THE ENDOCRINE SYSTEM IN SPORTS AND EXERCISE

VOLUME XI OF THE ENCYCLOPAEDIA OF SPORTS MEDICINE

AN IOC MEDICAL COMMISSION PUBLICATION

IN COLLABORATION WITH

THE INTERNATIONAL FEDERATION OF SPORTS MEDICINE

EDITED BY

WILLIAM J. KRAEMER AND ALAN D. ROGOL

THE ENDOCRINE SYSTEM IN SPORTS AND EXERCISE

THE ENDOCRINE SYSTEM IN SPORTS AND EXERCISE

VOLUME XI OF THE ENCYCLOPAEDIA OF SPORTS MEDICINE

AN IOC MEDICAL COMMISSION PUBLICATION

IN COLLABORATION WITH

THE INTERNATIONAL FEDERATION OF SPORTS MEDICINE

EDITED BY

WILLIAM J. KRAEMER AND ALAN D. ROGOL © 2005 International Olympic Committee Published by Blackwell Publishing Ltd Blackwell Publishing, Inc., 350 Main Street, Malden, Massachusetts 02148-5020, USA Blackwell Publishing Ltd, 9600 Garsington Road, Oxford OX4 2DQ, UK Blackwell Publishing Asia Pty Ltd, 550 Swanston Street, Carlton, Victoria 3053, Australia

The right of the Authors to be identified as the Authors of this Work has been asserted in accordance with the Copyright, Designs and Patents Act 1988.

All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, except as permitted by the UK Copyright, Designs and Patents Act 1988, without the prior permission of the publisher.

First published 2005

Library of Congress Cataloging-in-Publication Data The endocrine system in sports and exercise / edited by W.J. Kraemer and A.D. Rogol. p. ; cm. a (The encyclopaedia of sports medicine ; v. 11) Includes bibliographical references and index. ISBN-13: 978-1-4051-3017-2 (alk. paper) ISBN-10: 1-4051-3017-2 (alk. paper) 1. Endocrine glands. 2. ExerciseaPhysiological aspects. 3. SportsaPhysiological aspects. [DNLM: 1. Exerciseaphysiology. 2. Endocrine Systemaphysiology. 3. Sportsa physiology. QT 260 E556 2005] I. Kraemer, William J., 1953– II. Rogol, Alan David, 1941– III. Series. QP187.3.E93E53 2005 612.4—dc22 2004026395

ISBN-13: 978-1-4051-3017-2 ISBN-10: 1-4051-3017-2

A catalogue record for this title is available from the British Library

Set in 9/12pt Palatino by Graphicraft Limited, Hong Kong Printed and bound in India by Gopsons Papers Ltd, Noida

Commissioning Editor: Andy Robinson Project Manager: Alice Nelson Development Editor: Nick Morgan Production Controller: Kate Charman

For further information on Blackwell Publishing, visit our website: http://www.blackwellpublishing.com

The publisher’s policy is to use permanent paper from mills that operate a sustainable forestry policy, and which has been manufactured from pulp processed using acid-free and elementary chlorine-free practices. Furthermore, the publisher ensures that the text paper and cover board used have met acceptable environmental accreditation standards. Contents

List of Contributors, viii 7 Growth Hormone Variants and Human Exercise, 77 Foreword, xii wesley c. hymer, richard e. grindeland, bradley c. nindl and Preface, xiii william j. kraemer

8 Growth Hormone Binding Proteins, 94 1 Introduction, 1 gerhard baumann alan d. rogol and william j. kraemer 9 Resistance Exercise: Acute and Chronic Changes in Growth Hormone Concentrations, 2 Basic Principles and Mechanisms of 110 Endocrinology, 8 william j. kraemer, bradley c. nindl michael r. deschenes and and scott e. gordon keiichiro dohi

10 The Growth Hormone Response to Acute and 3 Exercise Testing: a Bridge Between the Chronic Aerobic Exercise, 122 High-Tech and the Humanathe Need for arthur l. weltman, laurie wideman, Innovative Technologies, 25 judy y. weltman and johannes d. dan m. cooper veldhuis

4 Measurement of Peptide Hormones, 36 11 Proopiomelanocortin and Exercise, 134 martin bidlingmaier, zida wu and heinz w. harbach and christian j. strasburger gunter hempelmann

5 Analysis of Low Molecular Weight Substances 12 Introduction to the Insulin-Like Growth Factor in Doping Control, 47 Signaling System, 156 mario thevis and wilhelm schänzer charles t. roberts, jr

6 The Reproductive Axis, 69 13 Exercise, Training and the GH–IGF-I Axis, 165 johannes d. veldhuis and alon eliakim, dan nemet and arthur l. weltman dan m. cooper

v vi contents

14 The Role of MGF and Other IGF-I Splice 23 Resistance Exercise and Testosterone, 319 Variants in Muscle Maintenance and atko viru and mehis viru Hypertrophy, 180 geoffrey goldspink, shi yu yang, 24 Exercise Response of β-Endorphin mahjabeen hameed, stephen and Cortisol: Implications on Immune harridge and pierre bouloux Function, 339 allan h. goldfarb 15 Adrenal Gland: Fight or Flight Implications for Exercise and Sports, 194 25 Neuroendocrine Modulation of the Immune michael kjær System with Exercise and Muscle Damage, 345 mary p. miles 16 The Adrenal Medulla: Proenkephalins and Exercise Stress, 200 26 The Impact of Exercise on Immunity: jill a. bush and n. travis triplett the Role of Neuroendocrine–Immune Communications, 368 andrea m. mastro and 17 Exercise and the Hypothalamic–Pituitary– robert h. bonneau Adrenal Axis, 217 warrick j. inder and gary a. wittert 27 Exercise Regulation of Insulin Action in Skeletal Muscle, 388 richard c. ho, oscar alcazar and 18 Influence of Energy Availability on laurie j. goodyear Luteinizing Hormone Pulsatility and Menstrual Cyclicity, 232 anne b. loucks 28 Hormone and Exercise-Induced Modulation of Bone Metabolism, 408 clifford j. rosen 19 Oral Contraceptive Use and Physical Performance, 250 29 Diet and Hormonal Responses: Potential jaci l. vanheest, carrie e. mahoney Impact on Body Composition, 426 and carol d. rodgers jeff s. volek and matthew j. sharman 20 Energy Balance and Exercise-Associated Menstrual Cycle Disturbances: Practical and 30 Neurohumoral Responses and Adaptations Clinical Considerations, 261 During Rest and Exercise at Altitude, 444 nancy i. williams and beth a. beidleman, janet e. staab mary jane de souza and ellen l. glickman

21 Regulation of Testicular Function: Changes 31 Neuroendocrine Influences on Temperature in Reproductive Hormones During Exercise, Regulation in Hot Environments, 466 Recovery, Nutritional Deprivation and lawrence e. armstrong and Illness, 279 jack a. boulant shalender bhasin 32 Alterations in Arginine Vasopressin with 22 Hormonal and Growth Factor-Related Exercise, Environmental Stress and Other Mechanisms Involved in the Adaptation of Modifying Factors, 487 Skeletal Muscle to Exercise, 306 carl m. maresh and fawzi kadi daniel a. judelson contents vii

33 Human Endocrine Responses to 36 Growth Hormone and Sport, 544 Exercise–Cold Stress, 499 jennifer d. wallace and john w. castellani and ross c. cuneo david w. degroot 37 Endocrinology of Overtraining, 578 34 Growth, Maturation and Hormonal andrew c. fry, juergen m. Changes During Puberty: Influence of steinacker and romain meeusen Sport Training, 512 james n. roemmich 38 Endocrinology of Sport Competition, 600 jay r. hoffman 35 Effects of Testosterone and Related Androgens on Athletic Performance in Men, 525 karl e. friedl Index, 613 List of Contributors

OSCAR ALCAZAR PhD, Research Division, Joslin PIERRE BOULOUX MD, Department of Medicine, Diabetes Center and Department of Medicine, Harvard Royal Free and University College Medical School, Medical School, Boston, MA 02215, USA University of London, Royal Free Campus, Rowland Hill Street, London, NW3 2PF, UK LAWRENCE E. ARMSTRONG PhD, Departments of Kinesiology and Physiology-Neurobiology, JILL A. BUSHPhD, Laboratory of Integrated University of Connecticut, Storrs, CT 06269, USA Physiology, Department of Health and Human Performance, University of Houston, Houston, TX 77204, USA GERHARD BAUMANN MD, Division of Endocrinology, Metabolism and Molecular Medicine, JOHN W. CASTELLANI PhD, Thermal and Northwestern University Feinberg School of Medicine and Mountain Medicine Division, US Army Research Institute Veterans Administration Chicago Health Care System, 303 for Environmental Medicine, 42 Kansas Street, Natick, MA East Chicago Avenue, Chicago, IL 60611, USA 01760-5007, USA

BETH A. BEIDLEMAN DSc, Biophysics and DAN M. COOPER MD, Center for the Study of Biomedical Modeling Division, US Army Research Institute Health Effects of Exercise in Children, Department for Environmental Medicine, Natick, MA 01760, USA of Pediatrics, Irvine College of Medicine, University of California, Irvine, CA 92868, USA SHALENDER BHASIN MD, UCLA School of Medicine, Drew-UCLA Reproductive Science Research ROSS C. CUNEO PhD, Department of Diabetes Center, Division of Endocinology, Metabolism and and Endocrinology, University of Queensland, Princess Molecular Medicine, Charles R. Drew University of Alexandra Hospital, 4102 Brisbane, Queensland, Australia Medicine and Science, Los Angeles, CA 90059, USA

MARTIN BIDLINGMAIER MD, DAVID W. DEGROOT MS, Thermal and Neuroendocrine Unit, Medizinische Klinik—Innenstadt, Mountain Medicine Division, US Army Research Institute Klinikum der LMU, Ziemssenstr. 1, 80336 Munich, for Environmental Medicine, 42 Kansas Street, Natick, MA Germany 01760-5007, USA

ROBERT H. BONNEAU PhD, Department of MICHAEL R. DESCHENES PhD, Department Microbiology and Immunology, Pennsylvania State of Kinesiology, The College of William and Mary, University College of Medicine, Milton S. Hershey Medical Williamsburg, VA 23187-8795, USA Center, Hershey, PA 17033, USA MARY JANE DE SOUZA PhD, Women’s JACK A. BOULANT PhD, Department of Exercise and Bone Health Laboratory, Faculty of Physical Physiology and Cell Biology, Ohio State University College Education and Health, 52 Harbord Street, University of of Medicine, Columbus, OH 43210, USA Toronto, Toronto, Ontario, M5S 2W6, Canada viii list of contributors ix

KEIICHIRO DOHI PhD, Osaka University of School, University of London, Royal Free Campus, Rowland Health and Sport Sciences, Asashirodai, Kumatori-cho, Hill Street, London, NW3 2PF, UK Sennan-gun, Osaka, 590-0496, Japan GUNTER HEMPELMANNMD, Department ALON ELIAKIM MD, Sackler School of Medicine, of Anesthesiology, Intensive Care Medicine, Pain Therapy, Tel-Aviv University and Child Health and Sports Center, University Hospital, Giessen, Rudolf-Buchheim-Str. 7, D- Pediatric Department, Meir General Hospital, Kfar-Saba 35385 Giessen, Germany 44281, Israel RICHARD C. HO PhD, Research Division, Joslin KARL E. FRIEDL PhD, US Army Research Institute Diabetes Center and Department of Medicine, Harvard of Environmental Medicine, Natick, MA 01760-7007, USA Medical School, Boston, MA 02215, USA

ANDREW C. FRY PhD, Exercise Biochemistry JAY R. HOFFMAN PhD, Department of Health and Laboratory, 135 Roane Field House, The University of Exercise Science, College of New Jersey, Ewing, NJ 08628, Memphis, Memphis, TN 38152, USA USA

School of Exercise, ELLEN L. GLICKMAN PhD, WESLEY C. HYMER PhD, Department of Leisure and Sport, Kent State University, Kent, OH 44513, Biochemistry and Molecular Biology, Pennsylvania USA State University, University Park, PA 16802, USA

ALLAN H. GOLDFARB PhD, Exercise and WARRICK J. INDER MD, Department of Sport Science Department, University of North Carolina- Medicine, St. Vincent’s Hospital, University of Melbourne, Greensboro, Greensboro, NC 27402-6170, USA Fitzroy, VIC 3065, Australia GEOFFREY GOLDSPINK PhD, ScD, DANIEL A. JUDELSON MA, Human Department of Surgery, Royal Free and University College Performance Laboratory, Department of Kinesiology, Medical School, University of London, Royal Free Campus, University of Connecticut, Storrs, CT 06269-1110, USA Rowland Hill Street, London, NW3 2PF, UK

FAWZI KADI PhD, Department of Physical Education LAURIE J. GOODYEAR PhD, Joslin Diabetes and Health, Örebro University, 70182 Örebro, Sweden Center, One Joslin Place, Boston, MA 02215, USA

SCOTT E. GORDON PhD, Human Performance MICHAEL KJÆR MD, PhD, University of Laboratory, East Carolina University, Greenville, NC 27858, Copenhagen, Sports Medicine Research Unit, Bispebjerg USA Hospital, Bispebjerg Bakke 23, DK-2400 Copenhagen NV, Denmark RICHARD E. GRINDELAND PhD, Life Science Division, NASA-Ames Research Center, Moffett WILLIAM J. KRAEMER PhD, Human Field, CA 94035, USA Performance Laboratory, Department of Kinesiology, University of Connecticut, Storrs, CT 06269-1110, USA MAHJABEEN HAMEED PhD, Department of Surgery, Royal Free and University College Medical School, ANNE B. LOUCKS PhD, Department of Biological University of London, Royal Free Campus, Rowland Hill Sciences, Ohio University, Irvine Hall 053, Athens, OH Street, London, NW3 2PF, UK 45701, USA

HEINZ W. HARBACHMD, Department of CARRIE E. MAHONEY BS, Department of Anesthesiology, Intensive Care Medicine, Pain Therapy, Kinesiology, University of Connecticut, Storrs, CT 06269- University Hospital, Giessen, Rudolf-Buchheim-Str. 7, D- 1110, USA 35385 Giessen, Germany CARL M. MARESH PhD, Human Performance STEPHEN HARRIDGE PhD, Department of Laboratory, Department of Kinesiology, University of Physiology, Royal Free and University College Medical Connecticut, Storrs, CT 06269-1110, USA x list of contributors

ANDREA M. MASTRO PhD, Department of Hillside Road, Unit 110, University of Connecticut, Storrs, Biochemistry and Molecular Biology, 431 South Frear CT 06269-1110, USA Building, Pennsylvania State University, University Park, PA 16802, USA JANET E. STAAB BS, Thermal and Mountain Medicine Division, US Army Research Institute for ROMAIN MEEUSEN PhD, Faculty of Physical Environmental Medicine, 42 Kansas Street, Natick, MA Education and Physiotherapy, Vrije Universiteit Brussel, 01760-5007, USA Brussels, 1050, Belguim CHRISTIAN J. STRASBURGER MD, MARY P. MILES PhD, Department of Health and Division of Endocrinology, Department of Internal Medicine, Human Development, Montana State University, Bozeman, Charité—Campus Mitte, Schumannstrasse 20/21, 10117 MT 59717, USA Berlin, Germany

DAN NEMET MD, Sackler School of Medicine, Tel- JUERGEN M. STEINACKER MD, PhD, Aviv University and Child Health and Sports Center, Section of Sports and Rehabilitation Medicine, University of Pediatric Department, Meir General Hospital, Kfar-Saba Ulm, 89070 Ulm, Germany 44281, Israel

BRADLEY C. NINDL PhD, Military Performance MARIO THEVIS PhD, Institute of Biochemistry, Division, US Army Research Institute of Environmental German Sport University Cologne, Carl-Diem Weg 6, 50933 Medicine, Natick, MA 01760, USA Cologne, Germany

CHARLES T. ROBERTS, JR PhD, Department N. TRAVIS TRIPLETT PhD, Department of of Pediatrics (NRC5), Oregon Health and Science Health, Leisure and Exercise Science, Appalachian State University, 3181 SW Sam Jackson Park Rd., Portland, OR University, Boone, NC 28608, USA 97239, USA JACI L. VANHEEST PhD, Department of CAROL D. RODGERS PhD, Department of Kinesiology, University of Connecticut, Storrs, CT 06269, Physical Education and Health, University of Toronto, USA, and Adjunct in Department of Physical Education and Toronto, Ontario, Canada, and Department of Physiology, Health, University of Toronto, Toronto, Ontario, M5S 2W6, Faculty of Medicine, University of Toronto, Ontario, M5S Canada 2W6, Canada JOHANNES D. VELDHUIS MD, Division of JAMES N. ROEMMICH PhD, Department of Endocrinology and Metabolism, Department of Internal Pediatrics, Division of Behavioral Medicine, State University Medicine, Mayo Medical and Graduate Schools of Medicine, of New York at Buffalo, 3435 Main Street, Buffalo, NY General Clinical Research Center, Mayo Clinic, Rochester, 14214-3000, USA MN 55905, USA

ALAN D. ROGOL MD, PhD, Clinical Pediatrics, ATKO VIRU DSc, PhD, Institute of Sports Biology, University of Virginia, ODR Consulting, 685 Explorers University of Tartu, 18 Ylikooli, Tartu 51014, Estonia Road, Charlottesville, VA 22911-8441, USA

Institute of Sports Biology, CLIFFORD J. ROSEN MD, Maine Center for MEHIS VIRUPhD, Osteoporosis Research and Education, St. Joseph Hospital, University of Tartu, 18 Ylikooli, Tartu 51014, Estonia 900 Broadway, Bangor, ME 04401, USA JEFF S. VOLEK PhD, Department of Kinesiology, WILHELM SCHÄNZER PhD, Institute of University of Connecticut, Storrs, CT 06269-1110, USA Biochemistry, German Sport University Cologne, Carl-Diem Weg 6, 50933 Cologne, Germany JENNIFER D. WALLACE PhD (Med) pending, Metabolic Research Unit, Department of Medicine, MATTHEW J. SHARMAN MA, Human University of Queensland, Princess Alexandra Hospital, Performance Laboratory, Department of Kinesiology, 2095 4102 Brisbane, Queensland, Australia list of contributors xi

ARTHUR L. WELTMAN PhD, Department of Laboratory, Penn State University, University Park, PA Human Services, Department of Medicine, General Clinical 16802, USA Research Center and Exercise Physiology Laboratory, Memorial Gymnasium, University of Virginia, GARY A. WITTERT MD, Department of Medicine, Charlottesville, VA 22908, USA Royal Adelaide Hospital, University of Adelaide, Adelaide, SA 5000, Australia JUDY Y. WELTMAN MS, General Clinical Research Center, University of Virginia, Charlottesville, ZIDA WU MD, Division of Endocrinology, Department VA 22908, USA of Internal Medicine, Charité—Campus Mitte, Schumannstrasse 20/21, 10117 Berlin, Germany LAURIE WIDEMAN PhD, Department of Exercise and Sport Science, University of North Carolina-Greensboro, SHI YU YANG PhD, MVet Sci, Department of Greensboro, NC 27402-6170, USA Surgery, Royal Free and University College Medical School, University of London, Royal Free Campus, Rowland Hill NANCY I. WILLIAMSScD, Noll Physiological Street, London, NW3 2PF, UK Research Center and Department of Kinesiology, 108 Noll Foreword

The Encyclopaedia of Sports Medicine is a highly chronic adaptations of the human organism to sport valuable series that has been produced during the conditioning result in major changes in endocrine last 17 years by the IOC Medical Commission. The function. The co-editors and contributing authors present edition becomes the eleventh volume in the of Encyclopaedia Vol. XI, The Endocrine System in series. Sports and Exercise, have provided a comprehensive A volume that is focused on the physiology of and authoritative coverage of this highly complex sport and exercise could be expected to concentrate system. This volume will serve as a leading reference on skeletal muscle metabolism and the supporting for physicians, scientists and graduate students for functions of the cardiovascular and respiratory sys- many years to come. tems. Indeed, these aspects of the biology of human I am pleased to congratulate the co-editors and performance have been featured in certain of the earl- the contributing authors on the quality of their work ier editions of The Encyclopaedia of Sports Medicine. and to welcome the newest volume into the presti- It is now entirely appropriate that attention be gious Encyclopaedia of Sports Medicine series. given to the ductless glands that produce the hor- mones that modify and control so many of the Dr Jacques Rogge important body functions. A great number of the IOC President

xii Preface

It is an honor for each of us to serve as Co-Editors physiology, lacking direct acknowledgment of its for this important contribution in the field of endo- own value as a discipline. The study of endocrino- crinology, more specifically sport and exercise logy as a specialty in medicine has been in existence endocrinology. We were fortunate to have an excep- for decades, while its applications in the field of tional group of scientists to participate in this sem- exercise and sport has been a more recent phe- inal work in our field. Thus, each chapter is written nomenon with typical focus on one or, at most, a few by one or more of the world’s leading scientists in hormones. This encyclopedia provides the reader their specific area of expertise. Their enthusiasm with a more complete view of the many targeted and excitement for the project and its importantance areas of study that have propelled this field for- is reflected in the very dynamic of each chapter. We ward over the past several decades. The number of acknowledge many of our other distinguished col- publications listed on PubMed with hormones and leagues who have also made influential contribu- exercise has increased by more than 30-fold over the tions to this field but were unable to participate in past 20 years. Advances in technology as well as this project, as this field of study has undergone greater interest in the endocrine system and its dramatic growth over the past 20 years. integration with the nervous and immune systems Each author was asked to develop a working have fueled the exponential rise in publications in paradigm that would not only provide a cutting endocrinology and exercise. edge overview of the field as it currently exists but The encyclopedia begins with a basic overview of would also provide a template for research over the the principles and concepts in the field of endo- coming years. This encyclopedia provides one of crinology. It then takes the reader through the the few resources for such a comprehensive view different endocrine glands allowing multiple per- of the many aspects of endocrinology in sport and spectives on function. The interactive effects of exercise. It is important to understand that each hormonal influences on the immune system, muscle chapter was not meant to be a comprehensive review and bone are then examined. The interactions with of the existing literature, but rather a state-of-the-art nutrition are then covered, providing a unique view conceptual framework for the area covered. Thus, it of one of the most important (and controversial) was not the purpose to provide exhaustive refer- aspects of sport and exercise performance. The text ences but rather a “cutting edge” perspective for use then provides a unique examination of the hor- by both clinical and basic scientists. It is our hope monal interfaces with environmental stresses. The that this encyclopedia serves to educate as well as study of the underlying hormonal aspects of sport inspire future research in the area of endocrinology itself remains a much less developed area of study in sport and exercise. due to the inability to access athletes under com- For so many years the field of exercise endocrino- petitive stress in game/match/meet conditions. In logy has been folded into the many other areas of this encyclopedia, chapters examining overtraining, xiii xiv preface the young athlete’s growth and development, and to the International Olympic Committee’s Medical the endocrinology of sport competition are covered, Commission and Sub-Commission on Publications providing some unique perspectives for future in the Sport Sciences for providing us this opportun- work. The problem of anabolic drug use is then ity to present to the world our field of study in its examined in detail. Ending with the study of full array. endocrine mechanisms involved with competitive stress, this volume of The Encyclopaedia of Sports William J. Kraemer Medicine series permits the student, clinician and Storrs, Connecticut practicing scientist the chance to view this field Alan D. Rogol more comprehensively. We are particularly indebted Charlottesville, Virginia Chapter 1

Introduction

ALAN. D. ROGOL AND WILLIAM J. KRAEMER

What is endocrinology? tion in conjunction with the neural and immune systems to transmit it to define ambient conditions Endocrinology is the science of intercellular and and to maintain metabolic homeostasis. These fun- intracellular communication. Classically, as defined damental concepts have also been greatly expanded by Bayliss and Starling more than 100 years ago, it by advances in cell biology, molecular biology and was the secretion of a chemical substance (hormone, genetics to help explain hormonal synthesis, action from the Greek word hormaein, to stimulate, to spur and perhaps integration, but the backbone of the into action) into the general circulation to cause endocrine system remains unchangedaintegration an effect at a distal site (target organ). Hormone of multiple systems and the maintenance of cellular receptors were unknown at that time. The hypo- and metabolic homeostasis. thesis as stated was used to describe the secretion Classically, hormones were defined by an and action of secretin, a chemical substance secreted ablation–replacement paradigm. Endocrine glands by the small intestine into the blood stream to stimu- were removed from normal animals, the gland late pancreatic exocrine secretion (Bayliss & Starling ground-up and extracted for the active substance. 1902). Restoration, by the extract of the function lost by Initially all hormones were considered produced removal of the gland, closed the loop and led to the in specialized glandsafor example, the thyroid, the discovery of many active hormonal compounds. adrenals or the pituitaryato act on single or mul- Over 150 years ago, Berthold showed the ‘proof-of- tiple organs as a means of homeostatic regulation concept’ of hormones by castrating roosters and of metabolism (biochemical control mechanisms) noting regrowth of the wattles and comb by replace- agene expression; biosynthetic pathways and their ment of the testis into a body cavity (Berthold 1849). enzymatic catalysis; the modification, transforma- The transplant was both ectopic and lacked innerva- tion and degradation of biological substances; the tion, permitting him to conclude that the testes biochemical mediation of the actions and interactions released a product that controlled the development of such substances; and the means for obtaining, (and maintenance) of the secondary sexual charac- storing and mobilizing stored energy (Table 1.1). teristics. The concept of homeostasis was stated by This fundamental concept has been subsequently Claude Bernard, who showed that the liver could (and greatly) expanded to account for many forms release glucose into the blood: of intercellular (and even intracellular) transfer of information in plants and animals (Table 1.2). Hor- The constancy of the internal environment mones represent many classes of biological mole- is the condition that life should be free and cules (Table 1.3). It now includes almost every tissue independent....So far from the higher animal and cell for they either produce or respond to hor- being indifferent to the external world, it is on mones. These hormonal systems integrate informa- the contrary in a precise and informed relation 1 2 chapter 1

Table 1.1 Selected hormones of the endocrine system and their actions.

Endocrine organ Hormone Major actions

Testes Testosterone Stimulates development and maintenance of male sex characteristics, growth and protein anabolism Ovaries Estrogens Develops female secondary sex characteristics; matures the epiphyses of the long bones Progesterone Develops female sex characteristics; maintains pregnancy; develops mammary glands Anterior pituitary Growth hormone Stimulates IGF-I and -II synthesis; stimulates protein synthesis, growth and intermediary metabolism Adrenocorticotropic hormone (ACTH) Stimulates glucocorticoid release in adrenal cortex Thyroid-stimulating hormone (TSH) Stimulates thyroid hormone synthesis and secretion Follicle-stimulating hormone (FSH) Stimulates growth of follicles in ovary, seminiferous tubules in testes and sperm production Luteinizing hormone (LH) Stimulates ovulation and production and secretion of sex hormones in ovaries and testes Prolactin (Prl) Stimulates milk production in mammary glands Posterior pituitary Antidiuretic hormone (ADH) Increases reabsorption of water by kidneys and stimulates contraction of smooth muscle Oxytocin Stimulates uterine contractions and release of milk by mammary glands Adrenal cortex Glucocorticoids Inhibits or retards amino acid incorporation into proteins (cortisol); stimulates conversion of proteins into carbohydrates (gluconeogenesis); maintains normal blood sugar level; conserves glucose; promotes metabolism of fat Mineralcorticoids Increases or decreases sodium–potassium metabolism; (aldosterone, increases body water deoxycorticosterone, etc.) Adrenal medulla Epinephrine Increases cardiac output; increases blood sugar, glycogen breakdown and fat mobilization Norepinephrine (some) Similar to epinephrine plus constriction of blood vessels Proenkephalins (e.g. peptide F, E) Analgesia; enhances immune function Thyroid Thyroxine Stimulates oxidative metabolism in mitochondria and cell growth Calcitonin Reduces blood calcium levels; inhibits osteoclast function Heart (cardiocytes) Atrial natriuretic hormone Facilitates the excretion of sodium and water; regulates blood pressure and volume homeostasis and opposes the actions of the renin–angiotensin system Pancreas Insulin Stimulates absorption of glucose and storage as glycogen Glucagon Increases blood glucose levels Parathyroids Parathyroid hormone Increases blood calcium; decreases blood phosphate Skin Vitamin D Produces vitamin D from 7-dehydrocholesterol and sunlight Adipose tissue Leptin Regulates appetite and energy expenditure

IGF-I, insulin-like growth factor I; IGF-II, insulin-like growth factor II. introduction 3

Table 1.2 Cellular communication.

Endocrine Chemical messenger (hormone) formed in a specialized tissue or organ (gland) and released into a circulation to affect another organ at a distance Paracrine Hormone synthesized in specialized cells to act on nearby cells Juxtacrine Hormone synthesized in one cell and acts on a contiguous cell Autocrine Hormone synthesized in one cell and acts on the surface of the same cell Intracrine Hormone synthesized in one cell and acts on the internal machinery of that cell without the necessity of cell surface receptors

Table 1.3 Chemical classes of hormonal molecules. Amino acid derivative Epinephrine, thyroid hormones Steroid Testosterone, cortisol, vitamin D Peptide Gonadotropin-releasing hormone, secretin Fatty acid derivatives Prostaglandins, leukotrienes

with it, in such a way that its equilibrium growth factor binding proteins (IGFBPs) that carry results from a continuous and delicate compen- IGF-I and II to their sites of action, prolong their sation, established as by the most sensitive of circulating half-life and dampen their effects as balances (Bernard 1957). hypoglycemic agents (LeRoith 2003). Most hormones are not secreted at a constant rate. Both of these results were known to Bayliss and Thus, the circulating concentrations undergo fluctu- Starling when they stated the secretin hypothesis ations that may be due to intermittent secretion or and coined the word hormone. Later, Walter Cannon alterations in metabolism (clearance). For example, emphasized that the constancy of the internal envir- the circulating concentrations of insulin are not uni- onment could only be achieved through the opera- form, but vary from minute-to-minute according to tion and integration of exquisitely co-ordinated the timing and composition of meals; from a base- physiologic processes which he named homeostatic line that is often determined by the relative mass of (Cannon 1939). lean and fat tissue and the regional distribution of Hormones may be fully active when secreted, the fat tissue (visceral versus subcutaneous). The for example cortisol; however, some may require concentrations of growth hormone vary in ‘pulses’ modification to become fully active, for example the throughout the day in response to meals and to conversion of tetraiodothyronine (thyroxine, T4) to sleep. Those of the gonadotropins, luteinizing hor- triiodothyronine (T3) by a specific deiodinase, the mone (LH) and follicle-stimulating hormone (FSH) conversion of testosterone to dihydrotestosterone and the sex steroids, estradiol and progesterone, (DHT) by 5-α reductase and the conversion of vit- vary over the approximately 28 days of the men- amin D3 to 1,25 dihydroxyvitamin D by two distinct strual cycle in addition to the 1- to 2-hourly pulses hydroxylases. Post-translational modification by of LH during the day. Adrenocorticotropic hor- phosphorylation, sulfation, or the addition of a lipid mone (ACTH) and cortisol usually maintain a daily chain may be required to permit solubility in the rhythm (high early in the morning and low at night) particular environment that the hormone finds itself. in those who sleep nocturnally. Disruption of these Because hormones may work at a distance, novel rhythms can lead to infertility or jet-lag. transport systems have evolved. Some steroid hormones have specific binding proteins to permit What do hormones do? solubility in aqueous environments (sex steroids bind to sex hormone binding globulin [SHBG] and For signal recognition the target cell produces a to albumin). There is a large series of insulin-like highly specific ‘receptor’ that recognizes the chemical 4 chapter 1

message (primary messenger) and begins a cascade CNS inputs of chemical reactions that lead to an alteration in the output of that cell, for example protein synthesis. Hypothalamus Tier I Receptor proteins may be expressed at the cell sur- Hypothalamic hormones face, cytoplasm and the nucleus. The receptors themselves may be part of the overall regulating Pituitary Tier II mechanism, for the signals from the hormones may Paracrine be modified by their number and affinity for their cytokines and Pituitary trophic growth factors ligand. Receptor endocytosis leads to the inter- hormone nalization of cell surface receptors; the hormone– Target gland Tier III receptor complex is subsequently dissociated and Peripheral leads to a diminution of hormonal effect. Receptor hormones desensitization (attenuated receptor signaling in the Fig. 1.1 Model for regulation of anterior pituitary presence of ligand) may alter the receptor’s affinity hormones secretion by three tiers of control. for the ligand, for example by phosphorylation (for Hypothalamic hormones impinge directly on their epinephrine). The major hormonal signaling path- respective target cells. Intrapituitary cytokines and ways use G-protein coupled receptors, tyrosine growth factors regulate tropic cell function by paracrine kinase receptors, serine/threonine kinase receptors, (and autocrine) control. Peripheral hormones exert negative feedback inhibition of respective pituitary ion channels, cytokine receptors and nuclear recep- trophic hormone synthesis and secretion. (Reproduced tors to activate the intracellular machinery. with permission from Ray & Melmed 1997.)

Control of hormone secretion ample visual and visceral inputs, blood borne and The endocrine system is organized in a hierarchal limbic signals. The concerted activation of this sys- manner. As an example, the hypothalamus releases tem leads to multiple outputs including physical thyrotropin-releasing hormone (TRH) which stimu- and behavioral changes. When these are adaptive, lates the anterior pituitary to produce thyroid- one usually remains within the homeostatic range stimulating hormone (TSH) which, in turn, acts on (see above). The hierarchy of this response is the thyroid to increase the synthesis and release of instructive for many of the other interactions of the the thyroid hormones, T4 and T3. The thyroid hor- neural, endocrine and immune systems. mones circulate bound to a specific binding protein, The central components include the parvocellular thyroxine binding protein (TBG), to affect virtually corticotropin-releasing hormone (CRH) and arginine all cells in the body through intracellular thyroid vasopressin (AVP) neurons within the paraventricu- hormone receptors and usually regulate the rate of lar nucleus of the hypothalamus, the CRH neurons metabolism of their targets (Fig. 1.1). of the paragigantocellular and parabranchial nuclei All of these systems are regulated by a series of of the medulla and the locus ceruleus (LC), and negative (and some positive) feedback loops to other mostly noradrenergic (norepinephrine [NE]) maintain the system within a narrow range of oper- cell groups of the medulla and pons (the LC/NE ation (homeostasis). However, this complex, hierar- sympathetic system, see Fig. 1.2). The peripheral chal system does not operate in a vacuum. There are limbs are the hypothalamic–pituitary–adrenal (HPA) multiple points of contact of the many endocrine axis, the efferent sympathetic–adrenomedullary sys- axes with the neural and immune systems, as ex- tem and some components of the parasympathetic emplified by the organism’s response to stress nervous system. AVP and CRH activate the pituit- (Fig. 1.2) (Selye 1950), which is served by a complex ary to produce proopiomelanocortin (POMC), which interaction of the central nervous system (CNS) and is eventually cleaved to produce ACTH (and many peripheral organs. This system must receive (and other centrally active peptides, such as α-melano- integrate) multiple neurosensory signals, for ex- cyte stimulating hormone, α-MSH). ACTH activates introduction 5

Circadian SS Stress GnRH rhythms Stress

LH, FSH

Sex Behavioral adaptation GH Glucocorticoids steroids GABA/BZD Serotonin Acetylcholine CRH

Arcuate N Visceral Bone LC/NE CRH POMC adiposity mass Symp. Syst.

NE Analgesia Fig. 1.3 Schematic representation of the detrimental AVP SP effects of chronic stress on adipose tissue metabolism NPY and bone mass. Stimulation is represented by solid lines ACTH and inhibition by light lines. FSH, follicle-stimulating hormone; GH, growth hormone; GnRH, gonadotropin- releasing hormone; LH, luteinizing hormone; SS, somatostatin. (Adapted with permission from Tsigos Epinephrine Glucocorticoids Norepinephrine & Chrousos 1995.)

Peripheral adaptation TSH and the peripheral thyroid hormones are less active). Further multiple metabolic axes are affected Fig. 1.2 Simplified representation of the central and by antagonizing the effects of growth hormone and peripheral components of the stress system, their functional interrelations and their relationships to other sex steroids on fat tissue catabolism and muscle and central nervous systems involved in the stress response. bone anabolism (Fig. 1.3). Activation is represented by solid lines and inhibition by dashed lines. ACTH, adrenocorticotropic hormone; AVP, arginine vasopressin; BZD, benzodiazepine; CRH, Exercise and endocrinology corticotropin-releasing hormone; GABA, γ-aminobutyric acid; LC/NE Sym. Syst., locus ceruleus/norepinephrine– Exercise produces dramatic challenges to the home- sympathetic system; NE, norepinephrine; NPY, ostatic mechanisms. The acute exercise responses neuropeptide Y; POMC, proopiomelanocortin; SP, can see metabolic increases of 10-fold or more. Force substance P. (Adapted with permission from Chrousos production can be repetitively high and maximal & Gold 1992. Copyright 1992, American Medical limits reached in typical training sessions. The Association.) challenges to the body under conditions of athletic competition are also dramaticaranging from per- the adrenal glands to produce cortisol and other formance of a marathon race in under 2 h and 10 min steroids that affect metabolic homeostasis and the to an Olympic weightlifter lifting a weight four times immune system. There are further interactions with his body mass. The mechanisms by which exercise the gonad axis (inhibition at multiple levels), the is tolerated and adapted to are intimately related to hypothalamic–pituitary–growth hormone/IGF-I axis hormonal regulation of physiological systems, but (inhibition at multiple levels) and the thyroid axis with both acute and chronic changes. (decreased production of TSH and inhibition of the Over the past 50 years or more, exercise and sport conversion of T4 to the more metabolically active physiology has continued to increase its study of T3ain fact it has been called the ‘euthyroid sick’ syn- hormonal mechanisms that mediate the exercise- drome in which there are diminished amounts of induced adaptations. For example, in resistance 6 chapter 1

training of primary importance to acute exercise such that the stimulus is highly specific in its nature. performance and subsequent tissue remodeling is Different from the general nature of stress proposed the role(s) played by many members of the endo- by Selye (1950) over 50 years ago, we now know that crine system (Kraemer & Ratamess 2003). Hormonal stress is very specific in its ‘fingerprint’ and mediat- elevations in response to resistance exercise take ing mechanisms. Thus, the magnitude of hormonal place in a unique physiological environment. Acute responses as well as the biocompartments in which elevations in circulating blood hormone concentra- they occur can differ dramatically. For example, tions (i.e. resulting from either increased secretion, with a resistance exercise that just stimulates an arm reduced hepatic clearance, plasma volume reduc- muscle, one notes only small or no increases in tions, reduced degradation rates) observed both circulating anabolic hormones, but growth factor during and immediately following a resistance concentrations (for example, IGF-I) may dramatic- exercise protocol present a greater likelihood of ally increase, specifically at the site of their action. interaction with receptors on either the target tissue Differences in hormonal responses also occur with cell membrane (e.g. peptides) or with nuclear/cyto- the magnitude of the exercise intensityalow in- plasmic receptors located within the target tissue tensity exercise produces a smaller magnitude of (e.g. steroid receptors) (Kraemer 2000). Coinciding hormonal response than higher intensity exercise. with blood hormonal concentrations is the number Thus, the influence of work, intensity, volume and of available receptors for binding and subsequent frequency all help create the exercise stimulus that cellular changes. Interaction with the receptor ini- occurs acutely with a single exercise session or tiates a myriad of events culminating in a specific repetitively as one continues with training. response, such as an increase in muscle protein Understanding the role of different hormones synthesis. Therefore, from the role of anabolic hor- within and among the various physiological com- mones (e.g. growth hormone, testosterone, IGFs) in munication systems is a challenge as few, if any, protein synthesis in response to resistance train- hormones act in isolation. Furthermore, with mul- ing to insulin’s role in glycogen metabolism with tiple-level communication being so important to endurance training, hormonal mechanisms have optimal homeostatic regulation, complex integra- become prominent in the study of exercise and tion of hormonal signals is required to respond to sport. Due to the ubiquitous nature of hormones, the multiple energy demands of exercise. no physiological system can adequately function Finally, the study of hormones and their roles in without one or more involved with its response and exercise and sport has permitted a better under- adaptation to one form or another of exercise. Such standing of the stresses of competition, overtraining dramatic hormonal influences have increased the and identified key factors in exercise prescription interest in endocrinology for those who investigate (e.g. intensities, frequencies and durations) that exercise and sport. can be optimized to create improved training pro- With exercise creating such a unique physiolo- grams and, ultimately, performance. Ultimately, the gical environment, one cannot merely extrapolate underlying physiological basis of any exercise or our understanding of resting homeostatic physiology sport stress has its underpinnings in endocrinolo- (endocrinology). The conditions with exercise are gical science.

References

Bayliss, V.M. & Starling, E.H. (1902) The hoden. Arch Anat Physiol Wiss Med 16, Kraemer, W.J. (2000) Neuroendocrine mechanism of pancreatic secretion. 42–46. responses to resistance exercise. In: Journal of Physiology 28, 325–334. Cannon, W.B. (1939) The Wisdom of the Essentials of Strength Training and Bernard, C. (1957) An Introduction to the Body. Norton, New York. Conditioning, 2nd edn. (Baechle, T., ed.). Study of Experimental Medicine. Dover, Chrousos, G.P. & Gold, P.W. (1992) The Human Kinetics, Champaign, IL: New York. concepts of stress and stress system 91–114. Berthold, A.A. (1849) Transplantation der disorders. JAMA 267, 1244–1252. Kraemer, W.J. & Ratamess, N.A. (2003) introduction 7

Endocrine responses and adaptations to Ray, D. & Melmed, S. (1997) Pituitary Tsigos, C. & Chrousos, G.H. (1995) Stress, strength and power training. In: Strength cytokine and growth factor expression endocrine manifestations and diseases. and Power in Sport, 2nd edn. (Komi, P.V., and action. Endocrine Reviews 18, In: Handbook of Stress Medicine (Cooper, ed.). Blackwell Science, Malden, MA: 206–228. G.L., ed.). CRC Press, Boca Raton, FL: 361–386. Selye, H. (1950) Stress and the general 61–65. Le Roith, D. (2003) The insulin-like growth adaptation syndrome. British Medical factor system. Experimental Diabesity Journal 4667, 1383–1392. Research 4, 205–212. Chapter 2

Basic Principles and Mechanisms of Endocrinology

MICHAEL R. DESCHENES AND KEIICHIRO DOHI

Introduction and general principles example of classical endocrine function, mediating agents can be released into the interstitial fluid to The overarching theme of physiology is the per- diffuse to and affect neighboring cells (paracrine sistent effort to maintain homeostasis within the effect), or even to interact with the same cell that organism. This drive to sustain a constant internal secreted them (autocrine effect). Indeed, some sub- milieu is challenged by regular perturbations of the stances, for instance insulin-like growth factor I organism’s internal and external environment. In (IGF-I), may exert their biological responses via the face of these changes, the capacity to achieve endocrine, paracrine and autocrine routes (Yakar homoeostasis is determined by effective commun- et al. 2002). More recently it has been proposed (Re ication among the cells of the organism. Two sys- 2003) that some growth factors and peptide hor- tems have evolved to provide such communication. mones may directly regulate activity within the cell The nervous system generally involves immediate of synthesis without ever exiting that cell (intracrine and brief responses to stimuli. In contrast, the effect). For the purpose of this chapter, only the fea- endocrine system typically is slower to respond to tures of the endocrine system will be addressed. alterations in the environment, but its responses Although scores of hormones have been identi- persist longer than those of the nervous system. The fied, and their biological activities regulate a plethora effects of the endocrine system are pervasive, and of physiological processes, several fundamental regulate the activity of virtually all cells within the principles of endocrine function have been estab- body. Each of these cells is perfused with blood; the lished. First, hormones are synthesized and secreted endocrine system capitalizes on this physiological by specialized ductless endocrine glands into the arrangement to communicate effectively through- bloodstream to be carried to ‘target’ cells that bind out the body. the hormone and respond to it by altering their The word ‘hormone’ derives from the Greek biological activity in a specific, preprogrammed word hormaein which means ‘to spur into action’ or fashion. Second, although some endocrine glands ‘to stimulate’. In 1902, Bayliss and Starling (1902) are the main components of organs specialized for described a substance secreted into the blood by one endocrine function (e.g. pituitary, thyroid), others organ (small intestine) that gave rise to a response in are located in organs that have distinct, primary another (pancreas). Thus, secretin became the first physiological functions (e.g. heart, gut, kidney). hormone identified. Today, a hormone is commonly Third, a single endocrine gland may produce more defined as ‘a chemical substance that is released into than one hormone. Fourth, with rare exception, the blood in small amounts and that, after delivery however, a single endocrine cell will synthesize and by the circulation, elicits a typical physiological release only one hormone. Fifth, a particular hor- response in other cells’ (Goodman 1994). It has mone may be secreted by more than one endocrine been discovered, however, that in addition to this gland. Sixth, a single hormone may elicit several 8 principles of endocrinology 9

different physiological responses by binding with the form of testosterone, yet trace amounts of corti- several different types of target cells. Seventh, any sol are also manufactured due to a minor presence one hormone, however, elicits only a single res- of the enzymes that comprise that steroid’s bio- ponse in any one type of target cell. Eighth, any one synthetic pathway. In viewing Fig. 2.1, it is easy to target cell may respond to numerous hormones, appreciate how this diversion among enzymatic each inducing its own biological response within pathways can occur. that cell. Ninth, a particular intracellular activity, The rate at which steroid hormones are produced glycolysis for example, can be regulated by more is dictatedaas in all enzymatic pathwaysaby the than one hormone. Tenth, a target cell’s responsive- activity of the rate limiting enzyme of the pathway, ness to a specific hormone may vary according to its i.e. the enzyme that catalyzes the slowest reaction own state of differentiation, presence of other hor- within the cascade. In all steroid hormone pro- mone, environmental conditions, etc. duction the rate limiting reaction is the conversion While the endocrine system regulates a host of of cholesterol to pregnenolone. Thus, factors that biological activities within the target cells of the increase the rate of steroid production do so primar- organism, the physiological effects of hormones can ily by accelerating the formation of pregnenolone, be broadly assigned to four areas. These are: (i) the and secondarily by amplifying the uptake of cho- digestion and metabolism (anabolic and catabolic lesterol from the blood into the endocrine gland mechanisms) of food nutrients; (ii) salt and water (Rhoades & Pflanzer 2003). balance; (iii) growth and development; and (iv) In the endocrine glands that manufacture reproductive function. steroids, there is no capacity for storage of the newly synthesized hormone. Accordingly, as the steroid Hormone categorization and synthesis is synthesized, it is secreted into the blood stream, and therefore the rate of the newly formed steroid’s All of the known hormones can be classified accord- release into the blood equals the rate of its produc- ing to their chemical composition and manner of tion within the endocrine cell. synthesis as either: (i) steroids; (ii) peptides/proteins; Peptide and protein hormones are comprised of or (iii) amines. Steroid hormones are derived from chains of amino acids. Should a small number of cholesterol and include the sex steroids (androgens, amino acids be involved (< 20), the hormone is typ- estrogens, progestins), which are produced in the ically referred to as a peptide hormone, but if 20 or gonads, as well as the glucocorticoids and mineralo- more amino acids comprise its structure, the chain is coids, which are synthesized by the adrenal glands. considered a protein hormone (Goodman 1994). In humans, the main androgen, or male sex steroid, Examples of peptide hormones include oxytocin, circulating in the blood is testosterone. Similarly, vasopressin, and somatostatin. Among the many estrogens are a family of female sex steroids, but (~ 100) protein hormones identified to date are in humans estradiol is the primary estrogen, and insulin, growth hormone, calcitonin and glucagon. among progestins, progesterone dominates. Cortisol Some of these proteins exist as relatively simple, is the principal glucocorticoid in humans, and single chains of amino acids, while others feature aldosterone is the major mineralocorticoid. disulfide bonds to connect different regions of the Since the same precursor moleculeacholesterol polypeptide sequence to convey complex tertiary ais used in the formation of all steroid hormones, it structure. Some protein hormones are even com- is the biosynthetic enzyme pathways present in the posed of multiple subunits that are bonded together endocrine gland that determines the specific steroid to form a single structure. that is mainly produced by that gland. Typically, Regardless of their final structure, all peptide/ however, small amounts of a secondary hormone protein hormones are synthesized within the endo- are concurrently synthesized as some diversion crine cell in a manner that is consistent with protein among the enzyme pathways occurs. For example, production in all cells. That is, peptide/protein the bulk of steroids synthesized by the testes is in hormone precursors are synthesized by ribosomes 10 chapter 2

Cholesterol

(side chain cleavage)

Pregnenolone (17α-Hydroxylase) 17-Hydroxypregnenolone (C17–20 Lyase) Dehydroepiandrosterone

A 3β-Hydroxysteroid D A 3β-Hydroxysteroid D A 3β-Hydroxysteroid D C Dehydrogenase-IsomeraseF CDehydrogenase-IsomeraseF CDehydrogenase-IsomeraseF

Progesterone (17α-Hydroxylase) 17-Hydroxyprogesterone (C17–20 Lyase)Androstenedione (Aromatase) Estrone

(21-Hydroxylase) (21-Hydroxylase) A17-HydroxysteroidD A17-HydroxysteroidD C Dehydrogenase F C Dehydrogenase F

11-Deoxycorticosterone 11-Deoxycortisol Testosterone (Aromatase) Estradiol

(11β-Hydroxylase) (11β-Hydroxylase) ANDROGEN (16α-Hydroxylase)

Corticosterone Cortisol Estriol GLUCOCORTICOID (18-Hydroxylase) ESTROGEN

A18-HydroxysteroidD C Dehydrogenase F

Aldosterone

MINERALOCORTICOID

Fig. 2.1 Biosynthetic pathways of steroid hormones. Involved enzymes are found in parentheses. ain conjunction with tRNA and mRNAaas much The increased intracellular concentration of calcium longer products than they ultimately appear in their causes the membrane of the secretory vesicle to fuse final form. These preprohormones contain signal with the cell’s plasma membrane, thus releasing the sequences indicating that the protein is intended for hormone via exocytosis. Typically the amount of release from the cell. Initial modifications of these peptide/protein hormone stored by the endocrine molecules occur in the endoplasmic reticulum, cell is limited and as a result the signal that stimu- where the ribosomes are attached, and include pro- lates the secretion of the cell’s hormone also triggers teolytic events that remove amino acid sequences the synthesis of additional quantities of that hor- aincluding the signal sequencearesulting in shorter mone (Rhoades & Pflanzer 2003). chains. The newly formed prohormones then journey The proteolytic cleavage of the prohormone to the Golgi complex where they undergo addi- during protein hormone synthesis conveys great tional modifications that include further proteolytic diversity in the number of hormones produced by cleavage and perhaps the addition of carbohydrate the endocrine system. The same precursor molecule (glycosylation) or phosphate (phosphorylation) can undergo differential processing resulting in the groups. Upon completion of these modifications, assembly of numerous end products. Perhaps the the Golgi complex pinches off a segment of its best example of this is the precursor proopiome- membrane to encapsulate the finished hormone in a lanocortin (POMC) which contains the amino acid vesicle. This secretory vesicle remains stored in the sequences of several peptide/protein hormones endocrine cell’s cytoplasm until the cell receives an including adrenocorticotropic hormone (ACTH), β- appropriate signal resulting in an influx of calcium. endorphin and β-lipotropic hormone, among others principles of endocrinology 11

(Krieger et al. 1980; Chretian & Seidah 1981). The thyroid hormones are part of the larger thyroglobu- specific cleavage enzymes expressed in the par- lin structure that is stored within the glandular cell. ticular endocrine gland synthesizing POMC deter- Upon stimulation to release thyroid hormone, pro- mines the principal hormone manufactured by that teolytic enzymes within the follicular cell break gland. For example, the anterior pituitary contains down the stored thyroglobulin liberating the T3 and a specific set of proteolytic enzymes that will yield T4 hormones intact, which are then released into the ACTH as its primary end product from the POMC circulation. prohormone. In contrast, neurons within the brain The catecholaminesaepinephrine and norepine- that produce POMC possess enzymes that cleave phrineaalso use tyrosine as their initial precursor, the molecule in such a way that β-endorphin is mainly but are produced in the medulla region of the secreted. This differential processing of POMC also adrenal glands. This adrenomedullary tissue is occurs in the placenta, reproductive organs, gas- actually a modified component of the sympathetic, trointestinal tract and lung (Liotta et al. 1982; or excitatory, branch of the autocrine nervous sys- Margioris et al. 1982). This gland specific expression tem. In fact, the adrenal medulla directly receives of different hormones from a common precursor input from nerve terminals of the sympathetic nerv- moleculeabased upon the enzyme profile present ous system serving as an example of neuroendo- ais not unlike the biosynthesis of steroids. crine function. Amine hormonesaalso referred to as amino acid Catecholamine synthesis occurs in the chroma- derivativesaare those that use the amino acid ffin cells of the adrenal medulla via a multistep tyrosine as their initial precursor. Included in this biosynthetic pathway. First, tyrosine is converted category are the thyroid hormones (thyroxine [T4] to 3,4-dihydroxyphenylalanine (dopa) by tyrosine and triiodothyronine [T3]) and the catecholamines hydroxylase which acts as the rate limiting step in (epinephrine and norepinephrine). Despite sharing the production of catecholamines. Dopa is then the same precursor molecule, the thyroid hormones converted to dopamine, which is then transformed and catecholamines differ in many respects includ- into norepinephrine, most of which is methylated ing their synthesis, transport through the blood by the enzyme phenylethanolamine–N–methyl stream and mechanism of action at target cells. Here transferase (PNMT) resulting in the formation of the amine hormones will be addressed separately in epinephrine. Both norepinephrine and epinephrine describing their synthesis. are considered catecholamines, but the stoichiome- Thyroid hormone formation is dependent upon try of their synthesis and release from the adrenal the uptake of both tyrosine and the mineral iodide gland is 1 : 4. Although circulating levels of nore- from the blood into the follicular cells of the thyroid pinephrine exceed those of epinephrine, the bulk gland. Tyrosine is used as a backbone for the manu- of this norepinephrine originates from the sym- facture of thyroglobulin which is a large glyco- pathetic nervous system where it functions as a protein that is stored in large amounts within the neurotransmitter and ‘spills over’ into the blood. follicular cells. With the uptake of iodide from Epinephrine, however, is the primary catecholam- the blood, the tyrosine residues of thyroglobulin ine hormone circulating in the bloodstream (Hedge become iodinated through a multistep reaction, et al. 1987). ultimately forming either T4 or T3, depending on the Upon their production, the same glandular cells number of iodide ions that bind with thyroglobulin. that synthesize the catecholamines store them as Initially, thyroglobulin reacts with either one or two chromaffin granules. Stimulation by the sympathetic iodide ions resulting in the production of either nervous system results both in the secretion of cate- monoiodotyrosine (MIT), or diiodotyrosine (DIT), cholamines through typical exocytotic actions, and respectively. In the next step of the enzymatic path- increased activation of tyrosine hydroxylase affect- way of thyroid hormone synthesis, two iodide ing a greater rate of catecholamine synthesis by the atoms are added to the thyroglobulin molecule chromaffin cells, so that intracellular catecholamine converting MIT to T3 and DIT to T4. At this point the depots can be replaced (Rhoades & Pflanzer 2003). 12 chapter 2

Regulation of hormone secretion hormone. In most cases, to assure a ready supply of hormone, the same stimulus that evokes the release The effect of a hormone on its target tissue is propor- of stored hormone also activates enzymatic path- tional to its concentration within the bloodstream. ways that synthesize that hormone. An influx of Several factors combine to determine the level of calcium into the endocrine cell’s cytosol usually pre- any biologically active hormone in the circulation. cedes the release of stored hormone and activation These variables include: (i) rate of secretion from of biosynthetic pathways. This regulated form of the endocrine gland into the circulatory system; (ii) release is evident in the secretion of peptide/protein for some hormones, i.e. thyroid hormones, rate of hormones, as well as catecholamines. activationathe conversion of T4 to T3awithin the With both constitutive and regulated hormonal blood; (iii) for lipophilic hormones, i.e. steroid and release, the stimuli that govern secretion are typic- thyroid hormones, degree of binding to plasma pro- ally: (i) changes in plasma nutrient or ion concentra- teins; and (iv) rate of inactivation and clearance tions; (ii) neurotransmitters released from neurons from the blood. Of these factors it is the first-rate of onto endocrine cells; or (iii) the binding of hormones release into the bloodathat is the principal deter- released by other endocrine glands. Generally, these minant of circulating hormone levels, particularly stimuli do not operate independently of each other. under non-exercising conditions (Sherwood 2004). To the contrary, alterations in endocrine secretion In general, there are two modes of hormonal usually are a function of receiving input from more release into the bloodstream (Kelly 1985). Constitut- than one type of stimulus. ive release refers to a continual, moderate discharge The responsiveness of endocrine glands to these of endocrine agent into the circulation. In this release stimuli is contingent upon a sensitive and effective mechanism, hormone exits as it is synthesized; there feedback system that conveys information from is no storage capacity within the endocrine gland. target tissues back to the hormone releasing organ. Consequently, upon receipt of stimulatory signal, The most prevalent form of communication regulat- synthetic pathways increase their activity and ing secretory rates of endocrine glands is negative newly formed hormone is directly released into the feedback. This type of feedback occurs when the circulation via passive diffusion through the cell’s activity of one system (endocrine gland) adjusts to plasma membrane. This release pattern governs negate, or offset, a change in another system (target circulating levels of steroid and thyroid hormones, tissue), thereby re-establishing homeostasis. For which like the cell’s plasma membrane, are lipo- example, an elevation in blood glucose triggers the philic in nature. Constitutive secretion is mediated secretion of insulin by the pancreas. The increased by alterations in the phosphorylation status of pro- blood-borne insulin level expedites the uptake of teins acting as enzymes in endocrine biosynthetic glucose into fat and muscle cellsatarget cells of pathways. insulinaresulting in a normalization of glucose con- Regulated release is the second mode of delivery of centration in the blood. hormone from the endocrine gland to the blood- There are even multiple forms of negative feed- stream. In this case, the rates of protein synthesis back regulation controlling hormone secretion. The and release are not directly coupled, as they are in release of several important hormones is directed constitutive release. Rather, endocrine glands that by the hypothalamic–pituitary axis. A small region employ regulated release possess the ability to store at the base of the brain, the hypothalamus produces newly synthesized hormone. It should be noted, several releasing hormones that are transported by however, that even in this type of endocrine gland a portal blood supply to the anterior segment of the storage potential is limited. Indeed, for any particu- pituitary gland. In this example of neuroendocrine lar hormone, it is rare for even a day’s supply to be function, the hypothalamus delivers releasing hor- stored and available for release (Baulieu 1990). mones to the pituitary which, in turn, alters its rate In regulated release, a stimulus causes the exo- of secretion of a number of hormones into the sys- cytotic release of prepackaged vesicles containing temic circulation where they can be transported principles of endocrinology 13

throughout the organism. These pituitary hormones Stimulus may directly bring about effects on target tissues, or they may react with a third gland in the axis, thereby First gland influencing this gland’s secretion of hormone into (hypothalamus) the bloodstream. releases hormone into . . . Regarding negative feedback, this neuroendo- crine axis can display both ‘short-loop’ and ‘long- Portal vasculature loop’ regulation (Vander et al. 2001). In an example Short-loop of short-loop negative feedback, high levels of pro- feedback lactin in the bloodstream are detected by the Second gland (anterior pituitary) hypothalamus causing it to increase its own secre- Long-loop releases hormone into . . . tion of dopamine to the pituitary, thus curtailing feedback that gland’s release of prolactin into the circula- tion. In the three organ neuroendocrine axis, long- Systemic circulation loop negative feedback is also evident. To illustrate this feedback mechanism, the regulation of cortisol Third gland secretion will be discussed. The axis involved (adrenals, gonads, etc.) features the hypothalamus, pituitary, and adrenal releases hormone into . . . cortex. The hypothalamus delivers corticotropin- releasing hormone (CRH) into the portal blood sys- Systemic circulation tem where it is carried to the anterior pituitary. Upon binding of CRH, the pituitary releases ACTH Fig. 2.2 Illustration of short-loop and long-loop negative feedback in the regulation of hormone release. into the systemic bloodstream so that it can be deliv- ered to the adrenal glands. In the cortex region of the adrenalarecall that the adrenal medulla produces crine gland secretes additional hormone until catecholaminesathe binding of ACTH triggers the the biological process it governs is adequate. An release of cortisol into the systemic circulation so example of positive feedback is the regulation of that it can elicit glucocorticoid effects on target endocrine function during childbirth. Oxytocin tissues such as the liver, skeletal muscle and adip- released by the posterior aspect of the pituitary ose. In this case, negative feedback occurs when the gland stimulates muscle contraction of the uterus. elevation in blood-borne cortisol blunts the release As the birthing process continues and stronger of ACTH from the pituitary and/or the secretion uterine contractions are required, the activity of the of CRH from the hypothalamus. As an example of uterus signals the pituitary to increase its secretion the exquisite integration of signals delivered and of oxytocin, thus amplifying the strength and fre- received throughout the endocrine system, the quency of uterine contractions enabling the com- hormone secretion of a multiorgan endocrine axis pletion of parturition. can be regulated both by short-loop and long-loop Although the primary function of the endocrine feedback (Vander et al. 2001). Illustrations of short- system is to maintain homeostasis, changes in the and long-loop negative feedback are displayed body’s internal and external environments are not in Fig. 2.2. the only regulators of hormone secretion. Indeed, Although negative feedback is much more preval- circulating concentrations of most hormones dis- ent, positive feedback is also used to adjust hormone play predictable fluctuations, or rhythms, over a secretion. In positive feedback, the hormone induced given period of time. The best studied endocrine change in the biological activity of the target tissue rhythm is the circadian, or diurnal, rhythm. ‘Circadian’ is monitored by the endocrine gland that initially refers to the pattern of peaks and troughs through- released the hormone. Should the target tissue’s out the roughly 24-h solar day, while ‘diurnal’ response be insufficient in magnitude, the endo- alludes to the day/night oscillations of hormone 14 chapter 2 secretion. Often these two terms are used inter- Hormone transport in the blood changeably. These natural, preprogrammed rhythms of endocrine release originate in the suprachias- Each hormone is released from the endocrine gland matic nucleus region within the hypothalamus. that produced it into the venous capillaries sur- This pacemaker regulates the secretion of hormones rounding that gland. After passing through the based upon its own internal clock, and elicits release lungs and heart, the hormone enters the general sys- patterns that are specific to each hormone. For temic circulation so that it may travel throughout example, cortisol levels in the blood are highest in the body. While traveling in the blood, a very small the morning, while growth hormone peaks during amount of hormone may be found adsorbed onto the night-time hours (Illnerova et al. 2000). the plasma membrane of red blood cells, but mostly In addition to variation over the 24-h day, hor- they are dissolved in the plasma (peptide/protein mone secretion often displays a regular pulsatile hormones), or bound to plasma proteins (steroids, pattern sometimes referred to as an ‘ultradian’ thyroid hormones). Due to their chemical structure rhythm that is superimposed on the underlying cir- the peptide/protein hormones are hydrophilic in cadian rhythm. These episodic bursts of hormone nature and readily dissolve in the blood’s plasma. release also appear to be directed by the hypotha- But once in the bloodstream these hormones are lamus and may have important physiological con- exposed to numerous proteolytic enzymes that sequences. For example, it has been demonstrated break them down, thus preventing them from inter- that even when total dosage is the same, glucose acting with their target tissues. Recall, however, that uptake into target tissue is enhanced when insulin endocrine glands that synthesize peptide/protein is delivered in pulses compared to a steady, non- hormones are able to store them and secrete them pulsatile pattern (Porksen 2002). upon receiving a stimulatory signal indicating a Although not as well studied in humans as in need for that particular hormone. animals, hormone secretion is also known to vary In contrast to the peptide/protein hormones, seasonally. These ‘circannual’ rhythms correspond steroid hormones are hydrophobic in structure, and to changes in the number of daylight hours, which as a result, are not soluble in the plasma. Con- are sensed by the pineal gland within the central sequently, the vast majority of these hormones nervous system (Short 1985). This gland, often (> 95%) travel through the bloodstream bound to alluded to as the ‘third eye’ due to its photosensit- various plasma proteins. The amine hormones also ivity, adjusts the amount of melatonin it secretes in are bound to proteins as they travel through the response to alterations in total daylight hours. In blood, albeit to various degrees. Only about 50% of animals demonstrating seasonal breeding behavior, blood-borne catecholamines are bound to carriers, the production of gonadotropins is modulated by while a much higher proportion (~ 99%) of thyroid these circannual variations in melatonin production hormones, which are hydrophobic in structure, are (Tamarkin et al. 1985). Other seasonal dependent in the bound state. behaviors, such as hibernation, migration and even Steroids may either bind to a plasma protein dis- changes in the color of fur, are guided by the pre- playing high specificity for a particular hormone dictable oscillations in circulating melatonin levels. (Table 2.1), or with lower specificity and affinity In humans, increased melatonin production, which to albumin and transthyretin, which are abundant occurs as the number of daylight hours decreases, circulating proteins. All carrier proteins, the high has been associated with altered mood states and specificity transport proteins and the ubiquitous even depression (Lewy et al. 1987). More recently it albumin and transthyretin, are synthesized by the has been established that in all mammalsaincluding liver before being released into the blood (Rhoades humansamelatonin plays a major role in ensuring & Plfanzer 2003). The low affinity albumin and proper circadian rhythmicity by interacting with transthyretin molecules contain several binding and influencing the suprachiasmatic nucleus (Pevet sites which can be bound not only by hormones but et al. 2002). also by other low molecular weight substances principles of endocrinology 15

Table 2.1 Examples of high specificity and affinity blood-borne hormone binding proteins.

Name Abbreviation/alternate name Hormone(s) bound

Sex hormone binding globulin SHBG/sex steroid binding protein Testosterone, estradiol Corticosteroid binding globulin CBG/transcortin Glucocorticoids, progesterone, aldosterone (minor amounts)

Thyroxine binding globulin TBG Triiodothyronine (T3), tetraiodothyronine (T4) Growth hormone binding protein GHBP Growth hormone Insulin-like growth factor IGFBPs IGF-I, IGF-II binding proteins

found in the blood. In contrast, the high specificity come the problem of insolubility in the plasma. It transport proteins typically possess a single binding has been demonstrated that due to their small size, site which is configured to interact with a specific most steroid and thyroid hormones readily dissolve hormone, and displays greater affinity at its ligand in the blood despite their hydrophobic chemical binding site than does albumin or transthyretin structure (Kronenberg et al. 2003). This suggests that (Baulieu 1990). The percentage of circulating hor- the binding proteins mainly act to provide a buffer- mone bound to albumin versus a specific transport ing or storage capacity, and to prevent the small, protein differs among hydrophobic hormones. For intact hormones from passing from the blood into example, equal amounts of bound aldosterone are the renal tubules of the kidneys where they can be complexed with specific binding proteins and albu- prematurely excreted from the body. In the case of min or transthyretin. But only 10% of circulating insulin-like growth factors, which are hydrophilic cortisol is bound to albumin or transthyretin, while in structure, specific binding proteins act to directly the vast majority (> 85%) is bound to corticosteroid mediate the biological effect imparted by the hor- binding globulin. mone/receptor complex at the target cell (Firth & Whether bound to albumin, transthyretin, or a Baxter 2002). specific transport protein, only the small amount of steroid or thyroid hormone that is unbound or ‘free’ Metabolic clearance of hormones is capable of binding with receptors sites expressed by target tissue. Thus, only the free portion of these After entering the circulation most hormones hormones is able to stimulate biological activity are rapidly degraded and feature half-lives of no within target cells. As this free steroid or thyroid more than 30 min, although this can vary greatly. hormone interacts with target tissue, some of the Catecholamines, for instance, have half-lives of only bound fraction dissociates from binding proteins, seconds, while thyroid hormones exhibit half-lives thereby forming a dynamic equilibrium between of several days. The length of any hormone’s half- the bound and unbound fractions that is influenced life in the bloodstream is dependent upon the by the need to maintain the organism’s homeostasis. metabolism and/or clearance from the circulation. In effect, the bound fraction serves as a buffer to As with transport, the chemical composition of the supply hormone to the target tissue as needed. This hormone will determine the specific method of its is presumed to compensate for the fact that unlike removal from the circulation, but in general, all peptide/protein hormones, the glands that synthes- hormones are primarily broken down within the ize steroids and thyroid hormones are unable to liver and eliminated via the renal system. Peptide/ store newly formed hormone to be released upon protein hormones and catecholamines are typic- demand. Indeed, it is no longer considered that the ally degraded by proteolytic enzymes in the blood primary function of transport proteins is to over- and the resultant amino acids are excreted through 16 chapter 2

the urinary tract. In contrast, steroid and thyroid sible, however, that two hormones with very similar hormones are resilient to circulating proteolytic chemical structures, for example insulin and insulin- enzymes, and they are mainly degraded in the liver like growth factor, will cross-react with each other’s through a series of reducing reactions which increase receptors. Under normal conditions and hormone the solubility of these hydrophobic molecules. concentrations, such cross-reactivity is minimal and Since steroids are small in structure, the improved the resultant cellular activity is negligible. solubility then allows the majority of them to be The strength of the bond formed between the hor- eliminated via the urinary tract, although a small mone and its receptor is described as the affinity portion is secreted directly from the liver into the of the ligand for its binding site. The affinity of the bile (Rhoades & Pflanzer 2003). Another, but less hormone–receptor complex determines how easily prevalent, mode of hormonal clearance is termed the bond can be broken. In high affinity reactions, ‘receptor-mediated endocytosis’. In this process much disturbance to environmental conditions, some larger protein hormones such as insulin are for example pH, temperature, etc., is necessary to internalized by the target cells after they have bound disrupt the hormone–receptor complex. In contrast, with receptors and elicited a response. After being low affinity bonds are easily broken and often engulfed by the target cell, the hormone is separ- require no change at all in the surrounding milieu. ated from its receptor and degraded by proteolytic The cross-reactivity described above displays this enzymes located within the cell’s cytoplasm. low affinity binding, which is why those hormone– receptor complexes typically result in very little Mechanism of action at target cells biological response in the cells where they occur. A dynamic equilibrium exists between unbound hormone in the blood perfusing the target tissue General principles and the hormone bound to the tissue’s receptors. The biological action(s) elicited by a hormone at its This equilibrium is essential to the endocrine sys- target cell is initiated by the binding of the hormone tem’s regulatory function. Consistent with the laws to its receptor produced by the cell. Indeed, this of mass action, the greater the amount of hormone reaction between hormone and receptor lies at the available to the targetacirculating concentrationa foundation of the endocrine system. In this manner the greater the probability that receptor sites will be an endocrine gland may release a hormone into the occupied by hormone, and thus the greater the bio- general circulation where it will perfuse all tissues logical response evident at the target. More directly within the body, but will exert its influence only stated, the target cell’s change in biological activity on those cells that express receptors specific for that is proportional to the number of hormone–receptor hormone. Typically, a target cell will express 2000– complexes formed, which is dictated by the hor- 100 000 receptors for any one hormone (Guyton & mone’s concentration in the bloodstream. Saturation Hall 1996). refers to the fraction of the target cell’s receptors that A receptor is a proteinasometimes a glycopro- is bound to hormone, and in physiological terms, teinathat possesses one or more binding sites for this is critical in determining the response of the cell ligand (hormone) and an effector, or active, site that to the hormone. Two factors primarily account for triggers a biological response in the cell expressing the degree of saturation of a cell’s binding sites: the that receptor. Each type of receptor displays a concentration of unbound circulating hormone and unique three-dimensional structure that is compli- the affinity of receptors for that hormone. mentary to the structure of a specific hormone, thus In considering the importance of saturation in conferring the property of specificity to the receptor. regulating a target cell’s response to a hormone, it is The complimentary chemical structures apparent in worth mentioning that, even at maximal biological the hormone and its receptor allow them to recog- response, not all of the cell’s receptors for that nize each other and interact in what is sometimes hormone are bound. This observation has lead to referred to as a ‘lock and key’ formation. It is pos- the ‘spare receptors’ concept of endocrine function. principles of endocrinology 17

That is, 100% occupancy of receptors is not neces- of available binding sites is obvious when one con- sary to elicit maximal biological response, suggest- siders that several types of cells, each with its own ing that the cell expresses more receptors than needs and homeostatic challenges, are sensitive to necessary. This ‘overexpression’ of receptors is, in the same hormone that is secreted into the systemic fact, appropriate when one recalls that the forma- circulation. For example, a disturbance in the envir- tion of hormone–receptor complexes is governed onment of one cell type may trigger increased by the rules of statistical probability. There are no release of a particular hormone to counter it. But forces of attraction that bring complimentary hor- another cell that is also responsive to that hormone mones and receptors together, although affinity may be at risk of homeostatic imbalance as a result sustains the bond once formed. There is simply a of the greater levels of that hormone. To prevent chance that circulating hormone and matching this, the cell may down-regulate, or temporarily binding site will come into contact with each other, decrease the number of receptors it makes available and then form a complex. Thus to enhance the prob- to bind the hormone. This would decrease the prob- ability that such random contacts will occur, target ability of hormone–receptor complex formation cells express more receptors than are needed to and dampen the cell’s sensitivity to the hormone, stimulate even maximal response. The expression thus maintaining its own homeostasis. Such adjust- of these ‘spare receptors’ is to be considered espe- ment in receptor number and/or affinity allows cially efficient and sensible given the very low con- each type of target tissue to respond in a manner centrations at which hormones travel in the blood appropriate to its own needs when exposed to a (10–8–10–12 mol·L–1). given amount of hormone traveling throughout Competition describes the process whereby dif- the entire organism. More long-term up- or down- ferent ligandsadisplaying similar chemical structure regulation of receptor expression is evident with amay compete for binding at the same receptor. dramatic alterations in the production of hormone Under normal physiological conditions, little com- through surgical removal of endocrine glands petition exists between hormones traveling through (Dahlberg et al. 1981), disease (Potier et al. 2002; the bloodstream. The potential for such competi- Pedersen & Vedickis 2003), or even as an adaptation tion, however, enables the pharmacological man- to chronic exercise training (Tchaikovsky et al. 1986; agement of some endocrine disorders, as drugs Deschenes et al. 1994). acting as antagonists may compete with endogen- Not only do target cells express receptors for a ous hormones that are over-secreted for binding host of different hormones so that various cellular sites, thereby preventing an inappropriately ampli- functions can be regulated by the endocrine sys- fied endocrine response. Also, the measurement tem, the same cell may respond to more than one of serum or plasma concentrations of hormones is hormone which affects the same biological pro- predicated upon competition as endogenously pro- cess within the cell. This phenomenon is known as duced hormones and labeled antigens vie for a redundancy, but should not be considered a super- given number of binding sitesaantibodiesapro- fluous or inefficient means of controlling cellular vided in the assay. physiology. For example, hepatocytes are sensit- Although, in general, the concentration of any ive to various hormonesaincluding glucagon and hormone present in the bloodstream dictates the epinephrineathat stimulate glycogenolysis and the degree of the biological response elicited by that release of glucose into the blood. However, different hormone, the target cell is capable of fine tuning its conditions may account for the need for additional response by adjusting the numberaand perhaps circulating glucose. Missing a meal may result in affinityaof the receptors it expresses for that hor- depressed glucose levels in the bloodstream, which mone. That is to ensure its proper function, target would be detected by the pancreas which would tissue may ‘up-regulate’ or ‘down-regulate’ its hor- respond by secreting glucagon to break down glyco- monal receptors to maintain its delicate homeostatic gen stores in the liver. Epinephrine, too, stimulates balance. The advantages of controlling the number glycogenolysis and the release of glucose from the 18 chapter 2

liver into the circulation. However, this is part of its attempt to achieve homeostasis, the availability an overall ‘fight or flight’ response elicited by the of hormone molecules must be sustained so that sympathetic nervous system and the adrenal secre- new molecules may bind with freshly unoccupied tion of epinephrine that includes increased heart receptors. rate, blood pressure and sweat rate. Obviously, these responses would be inappropriate to address Types of receptors and post-receptor actions a simple fasting induced state of hypoglycemia. Seen in this context, redundancy in the expression of At all target cells and with all hormones, the bio- receptors for the glycogenolytic hormones does not logical response evoked by a hormone must be appear wasteful, but rather a sophisticated approach preceded by the binding of the ligand (hormone to managing cellular homeostasis in the face of molecule) with its specific receptor expressed by many different challenges. the hormone sensitive cell. Despite the plethora of hormones produced by the endocrine system and the many cell types that respond to these hormones, Binding of hormone to receptor all receptors can be broadly categorized as either The biological response attributed to a hormone membrane bound or intracellular receptors. Mem- must be preceded by the binding of that hormone brane bound receptors are constituents of the tar- with its receptors located at its target cells. In this get cell’s plasma membrane and bind to peptide/ scenario, it is best to view the receptor as a mediator protein hormones, as well as catecholamines. As that transducts the extracellular message carried by indicated by their name, intracellular receptors the hormone to an intracellular signal that ulti- are located within the cell, and bind to steroid mately leads to a specific cellular response. and thyroid hormones which are small lipophilic At each hormonal receptor characterized to date, molecules that easily diffuse across the plasma only a single binding site exists so that at any time, membrane to enter the cell. only a single molecule of hormone can occupy a binding site to form a hormone–receptor complex. intracellular receptors Unlike enzymatic reactions where the substrate is altered by binding with enzyme, the hormone is In their unbound state there is some disparity in unaffected by the receptor to which it is bound. It the precise localization of intracellular receptors. is also noteworthy that the hormone–receptor com- Initially, it was believed that unbound intracellular plex is formed by a non-covalent bond that is receptors were found in the cell’s cytosol and that, reversible, and thus transient. upon hormone binding, the hormone–receptor Upon binding with its hormone, the receptor complex was translocated into the nucleus. How- undergoes a modification to its three-dimensional ever, it is now understood that even in the unbound structure. It is this conformational shift that activ- state, intracellular receptors for most steroid hor- ates the effector site of the receptor, setting forth mones are located within the nucleus. An exception a cascade of events which result in the cellular is the glucocorticoid receptor, which when unoccu- response that is ascribed to the involved hormone. pied is anchored to the cytoplasmic exterior of the It is important to recognize that while the binding nucleus (Lazar 2003). One feature that receptors of of hormone to receptor is reversible, the biological lipophilic hormones have in common is that they events initiated by the formation of hormone– are considered inactive when they are not bound receptor complex continue for some time after the with ligand. Characteristic of this inactive state is dissolution of that complex. On the other hand, this that these receptors are found in association with hormone induced cellular response is limited in heat shock proteins (HSPs), particularly HSP90 (Joab the absence of additional hormone–receptor com- et al. 1984; Catelli et al. 1985). Hormone binding to plex formation. Consequently, if the target cell is to the receptor results in a dissociation of the receptor remain activated for an extended period of time in from these chaperone proteins and dimerization of principles of endocrinology 19

newly liberated receptors. Upon this ‘activation’, As an illustration of the efficiency and integration of both cytoplasmic and nuclear receptors are trans- the endocrine system, it appears that the actions of a located to nuclear DNA, in order to interact with single lipohilic hormone bound to its intracellular hormone response elements (HREs). Once adhered receptors can regulate the transcriptional activity to these HREs, the hormone–receptor complex of several genes, altering the synthesis of several regulates the transcription of specific genes on the proteins. Yet this occurs in a co-ordinated manner DNA. Accordingly, intracellular receptors are con- in that each of the newly synthesized proteins par- sidered members of the larger family of transcrip- ticipates in a single, overall biological response tion regulatory factors. within the cell. For example, aldosterone functions Determining the amino acid sequence of these to maintain proper sodium levels within the body. It intracellular receptors has provided much insight does this by stimulating the reabsorption of sodium into the mechanisms of their interactions with hor- from the renal tubules when blood concentrations mones and DNA. At the C-terminal segment of of that mineral are low. To achieve this end, aldos- the protein structure is a sequence of about 250 terone increases the synthesis not only of sodium amino acids that are lipophilic in nature and form a channels and sodium pumps within the membranes structural pocket capable of binding steroid and of the tubules, but also the enzymes that manufac- thyroid hormones. The unique sequence of amino ture the adenosine triphosphate (ATP) used by acids comprising this pocket determines the recep- those pumps. This synchronized production of sev- tor’s specificity for a particular hormone. eral proteins subserves the reabsorption of sodium Binding of the hormone–receptor complex to from the renal filtrate in order to maintain proper specific HREs linked to hormone sensitive genes is osmolality and water balance within the body. directed by a DNA binding domain consisting of Since the mechanism(s) of action of intracellular approximately 70 amino acids that are also located endocrine receptors involves the management of at the C-terminus of the receptor protein. Common protein production, the biological activities stimu- to all DNA binding domains are two highly con- lated by steroid and thyroid hormones are typically served regions referred to as ‘zinc fingers’ which slow to occur and are long lasting. Recent evidence, directly interact with DNA (Scheidereit et al. 1986). however, suggests that in some cases lipophilic The ensuing genetic transcription is carried out by hormones induce fast acting responses of short a transcription complex that, in addition to the duration in target cells, without any change in pro- hormone–receptor complex, also includes ‘general tein expression (Oichinik et al. 1991). Although the transcription factors’ and ‘positively acting cofactors’ exact machinations of these rapid responses remain (McKenna et al. 1999). These coactivators enhance undefined, it appears that they are triggered by the the rate of transcription, at least in part, by unwind- interaction of steroid hormone with receptors ing the DNA double helix and stimulating RNA located on the plasma membrane of the target cell, polymerase activity (Kuo & Allis 1998). rather than with intracellular binding sites. Thus the cascade of events that result in the actions of steroid or thyroid hormones at target cells membrane bound receptors includes: 1 entrance of hormone via simple diffusion; As opposed to receptors that bind with lipophilic 2 formation of hormone–receptor complex; hormones which easily diffuse across the target 3 dissociation from HSPs and receptor dimerization; cell’s membrane to enter the cell, receptors for lipo- 4 translocation to DNA; phobic hormones (peptides/proteins) are located 5 binding to HREs; on the target’s plasmalemma since these hormones 6 assembly of transcription complexes; are not able to enter the cell. And because forma- 7 synthesis of gene specific mRNA; tion of the hormone–receptor complex occurs at 8 within cytosol, translation of proteins coded by the membrane’s extracellular surface, the resultant mRNA. intracellular activity is triggered by a pre-existing 20 chapter 2

signal transduction mechanism that provides a port when opened. The opening of the channel rapid response. In short, because membrane bound occurs upon binding of ligand, resulting in a con- receptors are linked to second messenger (the hor- formational shift within the membrane spanning mone acts as the first message) systems that are helices enabling specific ions to cross the plasmal- already in placeasimply needing to be activated by emma to enter, or exit, the cell. This type of receptor formation of hormone–receptor complexathe cellu- is found on excitable cells, i.e. neurons and myocytes, lar response elicited is quickly turned ‘on’ and can and the movement of ions electrically stimulates a be just as quickly turned ‘off’. response in the target cell. It is the chemical structure of membrane bound The best studied and most common receptor receptors that enables the forwarding of an extracel- that is directly linked with kinase activity involves lular message presented by a hormone molecule tyrosine kinase. Within the human genome, approx- to pre-existing signal transduction pathways within imately 100 receptor tyrosine kinases have been the cell. All membrane receptors are comprised of identified (Spiegel et al. 2003). Although character- three distinct sections, even though the receptor istics of the membrane spanning region are common itself is generally a single polypeptide chain (Spiegel to all of them, a large degree of variability exists at et al. 2003). The extracellular region is located at the extracellular region accounting for the diversity the N-terminal end of the chain, and consistently of ligands that interact with these receptors. Unlike expresses glycosylation sites. The carbohydrate most membrane bound receptors, members of the residues found at those sites appear to be involved tyrosine kinase family of receptors span the mem- in hormone binding, which occurs specifically in brane but once. Yet similar to many other ligand cysteine rich pockets. Each of the receptor’s trans- receptors, these proteins dimerize upon binding of membrane segmentsamembrane receptors typically hormone, a key step in activating the receptors. feature several membrane spanning regionsa Upon this activation, a specific region of the recep- consists of about 25 amino acids that are lipophilic tor’s intracellular domain phosphorylates tyrosine in nature and form a helical structure. The intracellu- residues of enzymes within the cell’s interior, thus lar region is at the C-terminal end of the chain and is eliciting the desired response of the cell. Moreover, responsible for the effector function of the receptor. the activated receptor is capable of phosphorylating Generally, the intracellular regionawhich is com- its own tyrosine residues located on its cytoplasmic posed of lipophobic amino acidsacontains regulat- region. This ‘autophosphorylation’ allows the ac- ory elements including phosphorylation sites. tivation state of the receptor to persist, therefore Although the actions of membrane bound recep- amplifying the signal carried by the hormone. A tors are carried out by numerous second messenger combination of the dissolution of the hormone– pathways, these post-receptor mechanisms can be receptor complex at the surface of the cell’s exterior categorized as: (i) ligand gated channels; (ii) recep- and the activity of intracellular phosphatases, which tor bound kinases; (iii) receptor bound guanylate de-phosphorylate enzymes, terminate the hormone cyclase; (iv) cytokine receptors; or (v) G protein induced biological response within the cell. coupled receptors. In the first three categories, when Other receptor bound kinases, less prevalent hormone binding occurs, the receptor itself directly than tyrosine kinase receptors, demonstrate similar stimulates cellular responses, but in cytokine recep- mechanisms of activity in that they phosphorylate tors and G protein coupled receptors, the recep- pre-existing enzymes found within the target’s tor activates another molecule within the target cell cytoplasm to alter cellular function. Both serine to carry out the final response attributed to the and threonine residues of enzymes residing within involved hormone. the cell can be phosphorylated by membrane In the case of ligand gated channels, the receptor receptors occupied by hormone. Again, cellular res- expresses not only an extracellular hormone bind- ponses cease when hormone–receptor complexes ing site, but also a channel within the membrane are no longer formed, and when phosphatase spanning regions that allows specific ions to trans- activity de-phosphorylates activated enzymes. This principles of endocrinology 21

demonstrates a common theme to the mechanism hormone’s message. As previously described, hor- of membrane receptor activity; cellular response is mone induced cellular responses are halted by turned ‘off’ both at the extracellular and intracellu- the unbinding of hormone from receptor, as well as lar levels. the activity of intracellular phosphatases which Receptors that mediate their responses via guany- inactivate enzymes within the target. late cyclase compose the final category of membrane The most prevalent category of membrane bound receptors where a component of the receptor pro- receptors is the family of G protein coupled receptors. tein itself stimulates cellular response. This category Over a thousand different ligands execute their is less ubiquitous than the other types of membrane biological activities via these receptors (Spiegel et al. receptors; it appears that only atrial natriuretic factor 2003). In all cases, however, the G protein and its utilizes this type of receptor to regulate the activity coupled receptor are two distinct proteins that are of its target cells. Here, again, the transduction of the functionally, but not structurally, linked together. first messenger (hormone) to intracellular activity Although these receptors contain seven membrane is restricted to the membrane molecule. After bind- spanning regions, they cannot directly alter intra- ing of ligand to the extracellular domain, a con- cellular activity upon binding of hormone. Instead, formational shift occurs throughout the structure located at the cytoplasmic lining of the plasmalemma resulting in the activation of guanylate cyclase; and neighboring the receptor is a G protein that an enzyme that is a constituent of the receptor’s stimulates cellular responses upon the formation of intracellular region. Upon activation, this enzyme hormone–receptor complexes. These G proteinsaso converts guanosine triphosphate (GTP) into cyclic termed because they require GTP to functionacan guanosine monophosphate (cGMP). The newly be either stimulatory (GS) or inhibitory (GI) in their formed cGMP is linked to a nucleotide dependent actions, with GS being far more commonplace. protein kinase which, in turn, phosphorylates All G proteins, whether GS or GI, consist of three enzymes within the cell’s cytoplasm. subunits termed α, β and γ. In the quiescent state, Cytokine receptors are quite similar to tyrosine the α subunit is bound to guanosine diphosphate, kinase receptors in their mechanism of action: these but upon binding of hormone to its receptor, this receptors also employ tyrosine kinases to evoke nucleotide is replaced by GTP, thus activating the G cellular responses. However, a major difference protein. Once activated, G proteins can stimulate exists in that cytokine receptors do not express the a host of different intracellular second messenger kinase enzyme itself in their intracellular tails. systems, therefore regulatingaeven simultaneously Rather, the enzyme is structurally uncoupled from anumerous intracellular activities. The major signal the cytokine receptor molecule. Members of the transduction mechanisms linked with G protein cytokine receptor familyawhich include growth activation will be addressed individually. hormone and prolactin receptorsaare comprised of 1 Cyclic adenosine monophosphate (cAMP). In this multiple subunits. Upon binding of ligand, these second messenger system, activated G proteins receptors form oligomers and activate the janus tyro- stimulate the enzyme adenylate cyclasealocated sine kinases, or JAKs, that are located in close prox- on the cytoplasmic lining of the plasmalemmaato imity to the receptor along the cytoplasmic lining convert ATP into cAMP. This reaction is similar to of the cell’s plasmalemma (Heim 1999). Cytokine the one stimulated by the interaction of hormone receptors themselves display no enzymatic activity with guanylate cyclase receptors where the binding (Argetsinger et al. 1993). It appears that the activa- of hormone results in the conversion of GTP to tion of JAKs occurs as a result of the drawing cGMP. In addition to the nucleotide substrate used, together of neighboring JAKs when oligomers of the systems differ in that, unlike guanosine cyclase, hormone–receptor complexes form, allowing the the enzyme adenylate cyclase is not a constituent JAKs to transphosphorylate each other’s tyrosine of the receptor molecule. residues. The activated kinases then phosphorylate Newly formed cAMP is capable of triggering num- cytosolic enzymes which, ultimately, carry out the erous intracellular processes, primarily by activating 22 chapter 2

various cAMP dependent protein kinases, also refer- receptor, phospholipase C cleaves phosphatidy- red to as PKAs. As with all kinases, PKAs stimulate linositol into diacylglycerol (DAG) and inositol cellular responses by phosphorylating enzymes of triphosphate (IP3), each of which triggers a cellular specific biochemical pathways. Because each type of response. Inositol triphosphate leaves the plasmal- target cell expresses its own set of PKAs and kinase emma to enter the cell’s cytoplasm where it re- activated pathways, the same second messenger acts with the endoplasmic reticulum to release its system may be used to control biochemical res- stored calcium into the cytosol. Increasing cytosolic ponses that are cell specific. And consistent with calcium levels is a common method of stimulating hormone action in general, at each step along the various cellular activities via calcium sensitive way the message of the hormone is amplified. enzymes.

The production of cAMP occurs in a matter of sec- Unlike IP3, the newly produced DAG remains onds, consequently the cellular response is quickly bound to the cytoplasmic lining of the plasmalemma evident; unless hormone–receptor complexes are where it can activate the membrane bound protein continually renewed the cellular action will be brief. kinase C (PKC) enzyme. As its name suggests, PKC Two enzymes terminate the actions induced by can only be turned ‘on’ by DAG in the presence of cAMP; phosphodiesterases cleave the bonds of cAMP elevated cytosolic calcium levels, thus the actions therefore inactivating it, and phosphatase activ- of IP3 are synergistic to those of DAG. As with all ity de-phosphorylates the enzymes stimulated by kinases, PKC functions by phosphorylating, and PKAs. As a result, responses typically evoked by thus activating, enzymes within the cell. cAMP are fast acting, and brief in duration. The activities stimulated by the phosphatidy- However, not all biological responses associated linositol signal transduction mechanism cease with cAMP are of short duration. It is now known when IP3 is eliminated by its conversion to inositol that cAMP can alter the transcription of certain pro- through a process of de-phosphorylation, and DAG teinsamimicking the actions of steroids and thyroid is inactivated by the addition of a phosphate group. hormonesaand therefore bring about long-term The activities of intracellular enzymes stimulated by modifications in cellular activity. Genes that are PKC are suppressed when cytosolic calcium levels sensitive to cAMP regulation contain sequences are returned to resting values. referred to as cAMP response elements that act as 3 In addition to the direct coupling of ion channel enhancers of transcriptional activity when stimu- opening with hormone binding as described earlier, lated. This stimulation occurs when a specific type receptor binding at the membrane may regulate of cAMP dependent protein kinase phosphorylates channel opening through a G protein intermediary. the cAMP responsive element binding protein (CREB). In this process, the membrane bound hormone Once phosphorylated, these CREBs act as trans- receptor and the channel that spans across the mem- cription factors and bind with the cAMP response brane are separate proteins. Upon formation of the elements located on the DNA. The genes responsive ligand-receptor complex, a nearby G protein located to these CREBs vary in accordance with target cell in the plasmalemma interacts with a neighboring type. This, again, is an illustration of how different channel protein. This interaction causes a conforma- target cells respond to the same signal transduction tional shift in the channel, resulting in its opening mechanism in a specific manner that is dictated by and movement of ions across the membrane. pre-existing ‘hard wiring’ of the cell. Typically, there is a large influx of ions into the cell 2 Phosphatidylinositolain another G protein related resulting in a substantial increase in the cytosolic second messenger system, a phospholipid constitu- concentration of that ion (Finn et al. 1996). Perhaps ent of the plasmalemmaaphosphatidylinositol—is the best example of this occurs in smooth muscle degraded by the activity of phospholipase C, a mem- cells where a sharp increase in cytosolic calcium brane bound enzyme that is stimulated by G protein levels occurs via G protein opening of ion speci- activity. Upon binding of hormone to its membrane fic channels in the plasmalemma. At rest, cytosolic principles of endocrinology 23

calcium concentrations of these cells are in the range strated when the cell responds to each hormone of 0.1–0.2 µmol·L–1, but upon hormone induced individually. To illustrate this, both growth hormone stimulation of calcium channel linked G proteins and cortisol elicit lipolysis in adipocytes but, when these levels quickly rise to more than 1 mmol·L–1. administered simultaneously, the rate of triglycer- This is a much greater increase than that observed ide break down is actually much greater than it following the opening of ligand gated ion channels. would be if those hormones were given separately In some cases, newly available calcium does not and their individual effects were added together. by itself stimulate a biological response within the In permissiveness the binding of one hormone at affected cell. Instead, the greatly increased con- the target cell must precede the binding of another centration of calcium dramatically augments the hormone, which only then stimulates a biological probability of its interaction with calcium specific response in the cell. In this case, it is said that the ini- intracellular binding proteins. Calmodulin is most tial hormone confers a permissive effect, allowing prominent among these binding proteins and it the target cell to respond to the second hormone. As exists in almost all cells. This high affinity protein an example of this, it has been found that in many possesses four binding sites. When these sites are target cells, the initial binding of thyroid hormone occupied by newly available calciumawhich can enables epinephrine to exert its biological effects also originate from intracellular storesathe calcium- on those cells. Lastly, antagonism occurs when the calmodulin complex activates enzymes, most com- influence of one hormone opposes, and effectively monly kinases. In turn, these activate enzymes that minimizes or even prevents, the action(s) of another directly catalyze the cellular activities attributed hormone at the target cell. To exemplify this, growth to the actions of the hormone bound to the extracel- hormone is known to antagonize the effects of lular surface of the target. This cellular response insulin at their shared target tissues. As a result, subsides with the dissolution of hormone–receptor the binding of growth hormone interferes with the complexes, therefore closing ion channels. Adenosine ability of insulin to promote glucose uptake and triphosphate-driven calcium pumps then deliver glycogen synthesis in the liver and skeletal muscle. cytosolic ions back to their intracellular (endoplas- mic reticulum) or extracellular sources of origin. Concluding comments

Integration of target cell responses Even in the brief description presented here, it is to hormones obvious that the mechanisms utilized by the endo- crine system to regulate target tissue responses are The integrative processes that characterize the characterized by a high degree of complexity and endocrine system are not only evident during the integration. Even within a single cell, physiological secretion of hormones but also in the responses of activities are governed both by steroid and protein target tissue to those hormones. More specifically, hormones, while various intracellular signal trans- the biological activity stimulated by one hormone duction mechanisms are employed in an effort to in a target tissue can be modified by the action of maintain homeostasis in the presence of continuous another hormone. Such integrated responsiveness and diverse environmental perturbations. How- of target cells is best exemplified by the phenomena ever, it is now understood that many patholo- of permissiveness, synergism, and antagonism. gical conditions, for example type 2 diabetes, can Synergism, also referred to as ‘potentiation’, occurs be directly attributed to dysfunction of these sig- when two different hormones stimulate the same nal transduction mechanisms. Accordingly, much biological activity in the target cell. However, rather research is currently being conducted that will than the effects of the two hormones being additive enhance our understanding of hormone regulated in magnitude, the response is greater than would be signal transduction in particular, as well as the func- evident by simply summing the responses demon- tion of the endocrine system in general. 24 chapter 2

References

Argetsinger, L.S., Campbell, G.S., Yang, X. subjective night and season of the year. Pevet, P., Botherel, B., Slotten, H. & et al. (1993) Identification of JAK2 as a Physiological Reviews 49 (suppl. 1), Saboureau, M. (2002) The chronobiotic growth hormone receptor-associated S1–S10. properties of melatonin. Cell and Tissue tyrosine kinase. Cell 74, 237–244. Joab, I., Radanyi, C., Renoir, J.M. et al. Research 309, 183–191. Baulieu, E.E. (1990) Hormones, a complex (1984) Immunological evidence for a Porksen, N. (2002) The in vivo regulation of communications network. In: Hormones: common non hormone-binding pulsatile insulin secretion. Diabetologia from Molecules to Disease (Baulieu, S.L. & component in ‘non-transformed’ chick 45, 3–20. Kelly, P.A., eds.). Chapman & Hall, New oviduct receptors for four steroids. Potier, M., Karl, M., Zhang, F. et al. (2002) York: 3–171. Nature 308, 850–853. Estrogen-related abnormalities in Bayliss, W.M. & Starling, E.H. (1902) The Kelly, R.B. (1985) Pathways of protein in glomerulosclerosis-prone mice: reduced mechanism of pancreatic secretion. eukaryotes. Science 230, 25–32. mesangial cell estrogen receptor Journal of Physiology 28, 325–353. Krieger, D.T., Liotta, A.S., Brownstein, M.J. expression and prosclerotic response to Catelli, M.G., Binart, N., Jung-Testas, I. & Zimmerman, E.A. (1980) ACTH, β- estrogens. American Journal of Pathology et al. (1985) The common 90-kD protein lipotropin and related peptides in brain, 160, 1877–1885. component of non-transformed ‘8S’ pituitary and blood. Recent Progress in Re, R. (2003) The intracrine hypothesis and steroid receptors is a heat-shock protein. Hormone Research 36, 277–344. intracellular peptide hormone action. EMBO Journal 4, 3131–3135. Kronenberg, H., Melmed, S., Larsen, P.R. Bioessays 25, 401–409. Chretian, M. & Seidah, N.G. (1981) & Polonsky, K. (2003) Principles of Rhoades, R. & Pflanzer, R. (2003) Human Chemistry and biosynthesis of endocrinology. In: Williams Textbook of Physiology, 4th edn. Thomson, propiomelanocortin. Molecular and Endocrinology (Larsen, P.R., Kronenberg, Brooks/Cole, Pacific Grove, CA. Cellular Biochemistry 34, 101–127. H., Melmed, S. & Polonsky, K., eds.). Scheidereit, C., Westphal, H.M., Carlson, Dahlberg, E., Snochowski, M. & W.B. Saunders, Co., Philadelphia: 1–9. C., Bosshard, H. & Beato, M. (1986) Gustafsson, J. (1981) Regulation of the Kuo, M.H. & Allis, C.D. (1998) Role of Molecular model of the interaction androgen and glucocorticoid receptors histone acetyltransferases and between the glucocorticoid receptor and in rat and mouse skeletal muscle cytosol. deacetylases in gene regulation. the regulatory elements of inducible Endocrinology 108, 1431–1440. Bioessays 20, 615–626. genes. DNA 5, 383–391. Deschenes, M.R., Maresh, C.M., Lazar, M.A. (2003) Mechanism of action of Sherwood, L. (2004) Human Physiology: Armstrong, L.E. et al. (1994) Endurance hormones that act on nuclear receptors. from Cells to Systems, 5th edn. Thomson, and resistance exercise induce muscle In: Williams Textbook of Endocrinology Brooks/Cole, Belmont, CA. fiber type specific responses in androgen (Larsen, P.R., Kronenberg, H., Melmed, Short, R.V. (1985) Photoperiodism, binding capacity. Journal of Steroid S. & Polonsky, K., eds.). W.B. Saunders, melatonin and the pineal: it’s only a Biochemical and Molecular Biology 50, Co., Philadelphia: 35–44. matter of time. In: Photoperiodism, 175–179. Lewy, A.J., Sack, R.L., Miller, S. & Hoban, Melatonin and the Pineal (Short, R.V., ed.). Finn, J.T., Greenwald, M.E. & Yau, K.W. T.M. (1987) Anti-depressant and Pitman Press, London: 1–8. (1996) Cyclic nucleotide-gated ion circadian phase-shifting effects of light. Spiegel, A., Carter-Su, C. & Taylor, S. channels: an extended family with Science 235, 352–354. (2003) Mechanism of hormones that act diverse functions. Annual Reviews in Liotta, A.S., Houghten, R. & Krieger, D.T. at the cell surface. In: Williams Textbook of Physiology 58, 395–426. (1982) Identification of a β-endorphin- Endocrinology (Larsen, P.R., Kronenberg, Firth, S.M. & Baxter, R.C. (2002) Cellular like peptide in cultured human placental H., Melmed, S. & Polonsky, K., eds.). actions of the insulin-like growth factor cells. Nature 295, 593–595. W.B. Saunders, Co., Philadelphia: 45–64. binding proteins. Endocrine Reviews 23, McKenna, N.J., Lanz, R.B. & O’Malley, Tamarkin, L., Baird, C.J. & Almeida, O.F.X. 824– 854. B.W. (1999) Nuclear receptor (1985) Melatonin: a coordinating signal Goodman, H.M. (1994) Basic Medical coregulators: cellular and molecular for mammalian reproduction. Science Endocrinology, 2nd edn. Raven Press, biology. Endocrine Reviews 20, 321–344. 227, 714–720. New York. Margioris, A.N., Liotta, A.S., Vaudry, N., Tchaikovsky, V.S., Astratenkova, J.V. & Guyton, A.C. & Hall, J.E. (1996) Textbook Boudin, G.W. & Krieger, D.T. (1982) Basharina, O.B. (1986) The effect of of Medical Physiology, 9th edn. W.B. Characterization of immunoreactive exercises on the content and reception Saunders, Co., Philadelphia. propiomelanocortin-related peptides in of the steroid hormones in rat skeletal Hedge, G.A., Colby, H.D. & Goodman, rat testes. Endocrinology 113, 463–471. muscle. Journal of Steroid Biochemistry 24, R.L. (1987) Clinical Endocrine Physiology. Oichinik, M., Murray, T.F. & Moore, F.L. 251–253. W.B. Saunders, Co., Philadelphia. (1991) A corticosteroid receptor in Vander, A., Sherman, J. & Luciano, D. Heim, M.H. (1999) The Jak–STAT neuronal membranes. Science 252, (2001) Human Physiology: the Mechanisms pathway: cytokine signaling from the 1848–1851. of Body Function, 8th edn. McGraw-Hill, receptor to the nucleus. Journal of Pedersen, K.B. & Vedickis, W.V. (2003) Boston. Receptor and Signal Transduction Research Quantification and glucocorticoid Yakar, S., Wu, Y., Setser, J. & Rosen, C.J. 19, 75–120. regulation of glucocorticoid receptor (2002) The role of circulating IGF-I: Illnerova, H., Sumova, A., Travnickova, Z., transcripts in two human leukemic cell lessons from human and animal models. Jac, M. & Jelinkova, D. (2000) Hormones, lines. Biochemistry 42, 10 978–10 990. Endocrine 19, 239–248. Chapter 3

Exercise Testing: a Bridge Between the High-Tech and the Humanathe Need for Innovative Technologies

DAN M. COOPER

Scope of exercise testing in rarely predict the impact of disease on exercise clinical medicine responses. Although tests like the 6-min walk (i.e. measuring An increasing number of investigators are using the distance walked by a subject in 6 min) continues measurements of human performance to probe to be a useful assessment of integrated responses, pathophysiology and disease mechanisms in a vari- such tools provide only crude insight into specific ety of conditions. Figure 3.1 shows the exponential physiological mechanisms of disease. How much increase in the number of published, peer-reviewed more information we could have from these simple randomized trials that utilized clinical exercise test- protocols if, for example, in addition to the distance ing. The widespread appeal of exercise testing is walked, we could measure body heat dynamics, that it allows clinicians to quantify physiological mechanical work performed and shifts in intramus- responses in a controlled setting that better reflects cular water with minimally invasive and intrusive the natural environment of their patients. Measure- devices. ments of almost any physiological variable, ranging from cardiovascular to hormonal, made only at rest, Advances in the technology of exercise testing 1600 Since the remarkable ‘growth spurt’ in human per- formance knowledge generated in the first half 1200 of the 20th century at such notable centers as the Harvard Fatigue Laboratory (Tipton 1998), the 800 development of new, clinically useful, technologies to assess human performance in response to physio- logical stresses like exercise has not kept pace with 400 the progress of many other areas of biomedical investigation. Treadmills and cycle ergometers have Number of PubMed citations 0 changed little since the first measurements of max- imal oxygen uptake were made in the 1920s. These devices have proved to be very useful for testing the 1966–701971–751976–801981–851986–901991–95 1996–2000 upper limits of gas exchange and metabolism, and are, therefore, suitable for studies of athletic perform- Fig. 3.1 Increasing number of clinical trials using exercise testing. Using the PubMed search engine, it is clear that ance in which near maximal efforts are critical. But the utilization of exercise testing in clinical trials is the emphasis on maximal efforts does not readily increasing rapidly. reflect the level and type of physical activity that 25 26 chapter 3

Fig. 3.2 ‘Found art’ in the General Clinical Research Center (GCRC) Human Performance Laboratory. A 7-year-old girl o creatively expresses her thoughts on the V 2max test. ‘I’m very sorry Dr. Cooper. I’m very tierd [sic] and I don’t want to do this again. I’m so sorry.’ determines quality of life in the clinical setting. science. Lagging behind this knowledge has been Moreover, typical exercise protocols are exhausting the development of tools for minimally invasive and uncomfortable, and often inappropriate for measures of neurological, intramuscular, cell sig- young children, the elderly and persons with most naling and vascular adjustments to exercise and diseases and disabilities (Cooper 1995; Metra et al. other stresses that could be used easily and safely in 1998) (Fig. 3.2). human beings. Most troubling is that this lack of Gas exchange during exercise can be measured progress has occurred despite an increasing use of precisely, but only with cumbersome and some- traditional exercise testing to probe mechanisms times painful mouthpieces which, in and of them- of human disease and to develop new therapies. selves, alter normal respiratory patterns (Lowhagen Moreover, there is growing recognition that health et al. 1999). Beyond the measurement of oxygen impairment directly related to physical inactivity is and carbon dioxide, little progress has been made increasing at an alarming rate (Booth et al. 2000; to exploit the potentially rich insight into disease Cooper et al. 2004). mechanisms that could be gained from online, con- The clinical exercise testing technology gap has tinuous measurement of nitric oxide (NO) and occurred despite a number of key technological and volatile organic compounds in the exhaled breath. conceptual breakthroughs where basic engineering Precise non-invasive quantification of physical activ- research is pointing the way toward novel clinical ity and energy expenditure under field conditions applications. In the 1970s and 1980s, the noted in free-living human beings remains an elusive Harvard biomedical engineer T. A. McMahon goal; such tools, like the use of stable-isotopically (McMahon 1984) reconfigured our understanding labeled bicarbonate (Zanconato et al. 1992; Coggan of the biomechanical principals that govern human et al. 1993), could revolutionize research focused on locomotion. His theories were tested on a running health outcomes in many areas of clinical and basic track built in his laboratory constructed so that science biomedical research. mechanical forces generated during human running Over the past 35 years, technological advances were matched by forces generated within the track. have facilitated an explosion of biomedical know- Equally remarkable was that McMahon developed ledge, particularly in molecular biology and neuro- a paradigm in which the biomechanical mechanisms exercise testing 27

of human locomotion translated into theoretical Remote biochemical imaging (e.g. 31P magnetic constructs that involved biochemical, cardiovas- resonance spectroscopy), pioneered by Britton cular and neurological control mechanisms. Such Chance and his colleagues (Chance 1994), also holds novel uses of integrative physiology toward solving the possibility of providing clinical investigators basic problems of human disease have proved to be with real time measurements of intramuscular high- a successful as a clinical research paradigm. energy phosphate during human performance (Zanconato et al. 1993). For example, Scheuermann- Imaging and spectroscopy approaches Freestone et al. (2003) recently examined both cardiac and skeletal muscle energetics during exer- A number of additional advances in non-invasive cise in patients with type 2 diabetes (Fig. 3.4). imaging approaches towards clinical human exer- Remarkably, although their cardiac morphology, cise testing also need to be mentioned in this con- mass and function appeared to be normal, the text. One involves attempts to link the tools of patients with diabetes had significantly lower phos- remote imaging with physiological function dur- phocreatine/adenosine triphosphate (PCr/ATP) ing exercise. Anatomic imaging, such as can occur ratios than the healthy volunteers. The cardiac with amazing precision using magnetic resonance PCr/ATP ratios correlated negatively with the fast- imaging, has been used to determine the profound ing plasma free fatty acid concentrations. Although changes in muscle water content and other physio- skeletal muscle energetics and pH were normal at logical mechanisms that occur even with brief rest, PCr loss and pH decrease were significantly exercise. But Patten and colleagues in a recent com- faster during exercise in the patients with diabetes, prehensive review of T2 mapping of muscle noted: who had lower exercise tolerance, and PCr recovery was slower in the patients. These investigators con- Despite demonstration of the capacity for cluded that type 2 diabetic patients with apparently imaging phenomena relevant to both exercise normal cardiac function have impaired myocardial physiology and clinical diagnosis, to date there and skeletal muscle energy metabolism related to have been surprisingly few clinical applications changes in circulating metabolic substrates. of T muscle mapping. Because MRI is nonin- 2 A major breakthrough in the understanding of vasive, it affords several advantages over tra- exercise responses and their alteration in disease ditional diagnostic modalities such as muscle states resulted from yet another interaction between biopsy or EMG for diagnosis of metabolic and engineers and physicians. The development of neuromuscular disorders in sports medicine, breath-by-breath measurement of gas exchange occupational medicine, and neurorehabilita- was pioneered in large measure in the Harbor– tion. MRI studies provide results rapidly for University of California, Los Angeles (UCLA) purposes of diagnostic decision-making and Laboratory of Drs. Brian Whipp and Karlman also offer outcome assessment for clinical or Wasserman (Wasserman et al. 1973). Wasserman, an exercise interventions without the patient suf- MD with a PhD in physiology, was a fellow in the fering the risks or discomforts associated with noted Cardiovascular Research Institute (CVRI) at repeated study (Patten et al. 2003). the University of California, San Francisco in the In their review, Patten and coworkers went on to 1960s. Whipp and Wasserman collaborated primar- point out that the alterations in T2 magnetic reson- ily with two engineers, Dr. William Beaver and, sub- ance imaging responses to exercise likely result sequently, Dr. Norman Lamarra. Lamarra received from two mechanisms: osmotically driven shifts of his PhD at UCLA in aerospace engineering for a muscle water that increase the volume of the intra- doctoral thesis devoted, oddly enough, entirely to cellular space; and from intracellular acidification analyzing the ontransient kinetics of oxygen uptake resulting from the end products of metabolism. An during exercise in human beings (Lamarra 1982)! example of the use of these techniques in a paretic Investigations into how disease influences these muscle and in a control subject are shown in Fig. 3.3. kinetics, and how these measurements prove to be 28 chapter 3

Fig. 3.3 T2-weighted axial magnetic resonance images of the arm obtained using a multiecho sequence (TE 20, 40, 60, 80 ms, TR = 2000, I NEX, 18 cm FOV) with in-plane resolution of 0.6–0.8 mm. The upper panel images illustrate the paretic (left) and non-paretic (right) arms at rest in an adult with post-stroke hemiparesis of 24 months duration. The lower panels illustrate the same arms following 40 dynamic elbow flexions performed at 80% of maximal effort. Following exercise, increased signal intensity (T2) is demonstrated in the flexor compartment bilaterally. The paretic side demonstrates markedly less activity-dependent increase in T2 and documents impaired muscle activation associated with post-stroke hemiplegia. (Data from Patten et al. 2003.) exercise testing 29

PCr Control Diabetic resting resting pH = 7.07 pH = 7.11

ATP

γ α PDE β Pi Fig. 3.4 Typical calf muscle 31P magnetic resonance spectra from a control subject and a patient with 5 0 –5 –10 –15 5 0 –5 –10 –15 type 2 diabetes at rest (upper panel, p.p.m. p.p.m. number of scans = 64), from the same patient at the end of exercise, and Control after Diabetic the same matched control at the 5.1 min end-exercise equivalent time (5.1 min) of exercise exercise at 5.1 min (lower panel, number of scans = 16). pH = 7.00 pH = 6.57 PCr, phosphocreatine; PDE, phosphodiesters; Pi, inorganic phosphate; α, β and γ indicate the three phosphate groups of adenosine triphosphate (ATP). Cytosolic pH was calculated from the chemical shift of Pi relative to PCr. Abscissa shows chemical shift in parts per million (p.p.m.). (Data from 10 5 0 –5 –10 –15 –20 10 5 0 –5 –10 –15 –20 Sheuermann-Freestone et al. 2003.) p.p.m. p.p.m. useful indicators of disease and its therapy, have ordinated function of the heart, the lungs, and proceeded at a relatively slow pace. the peripheral and the pulmonary circulation to The pioneers in the field of exercise physiology match the increased cellular respiration required clearly recognized that the biological responses to the to live and work (Wasserman et al. 1987). stress of exercise could ultimately be used to gain a We would add only that the ‘co-ordinated function’ greater understanding of fundamental processes at includes the neuromuscular and cellular signal the cellular and subcellular level. Wasserman’s 1975 transduction adaptive mechanisms as well. cartoon of ‘gears’ linking cellular function to gas exchange measured at the mouth aptly illustrates this concept (Fig. 3.5). But the primary use of exer- Development of new technologies— cise testing in the clinical setting has remained robots, strokes and maturation of almost invariably focused on cardiovascular meas- motor control urements. As noted by Wasserman and colleagues: In some cases, existing technologies are simply The authors would like to dispel a concept that not sufficient to test exercise and motor control in has developed in medicine, i.e., that there is a certain individuals and new devices need to be cardiac stress testing and pulmonary stress developed. For example, in dealing with stroke vic- testing. It is impossible to stress only the heart tims, traditional treadmills or cycle ergometers are or only the lungs. Exercise requires the co- often not feasible as either diagnostic or therapeutic 30 chapter 3

Muscle O2 and CO2 Ventilation activity transport (VA + VO = VE)

Peripheral Pulmonary Mitochondrion circulation circulation

PRO FLO PIRE O2 D O2 W EX D QCO C L VCO 2 V 2 Creat PO 4 Heart Muscleblood Lungs Pyr Lac V QO2 R O M CO IN VO 2 CONSU 2 FLO W SPIRED 2

Fig. 3.5 The metabolic ‘gears’ that Physiological response: QCO2 DilateSV Recruit VT link ventilation during exercise to the cellular and subcellular level. QO2 RecruitHR Dilate f (From Wasserman et al. 1987.)

tools because of the constraints and limitations to (a) 10 cm motor control caused by the brain injury. To get θ around these limitations, mechatronic devices and haptic robots have been developed by a number of investigators. Using these tools, new insight into the mechanisms that govern motor activity in stroke victims (Lum et al. 2002). For example, recent evidence suggests that brain injury can impair the ability to independently activ- ate shoulder and elbow muscles. Reinkensmeyer et al. (2002) hypothesized that if muscle activation patterns are constrained, then brain-injured sub- (b) jects should not be able to accurately grade initial hand movement direction during reaching toward a broad range of target directions. To test this hypo- thesis, Reinkensmeyer et al. used the mechatronic haptic robot to measure hand trajectories during reaching in three-space by 16 subjects with hemi- Fig. 3.6 Example hand trajectories for the ipsilesional (a) paretic stroke to an array of 75 targets distributed and contralesional arms (b) of a subject with severe stroke. throughout the workspace (Fig. 3.6). The contralesional (left) arm trajectories are flipped about Results of these innovative studies clearly suggest the sagittal plane. Reaching direction (θ) is defined that there are two identifiable classes of directional positive for movements to the right of straight ahead. Note that movement was constrained to essentially two control following stroke: largely preserved and directions (medial and lateral) for the severely impaired severely constrained. Since it has been hypothesized subject, even though substantial active range of motion that lower pathways substitute for corticospinal was preserved in these directions. (Data from ones following stroke, a possible explanation for Reinkensmeyer et al. 2002.) exercise testing 31

these two classes is that directional control is largely may arise from two sourcesaa relatively constant, preserved if some threshold fraction of corticospinal intrinsic source related to fundamental physiolo- pathways is spared. This hypothesis should be gical constraints of the developing motor system, and testable in the future via detailed functional imag- a more rapidly modifiable source that is modulated ing of key neural tracts. depending on the current motor context. The use of robots in measuring motor control is not limited to brain-injured patients. Interestingly, a Exhaled nitric oxide and exercise chance interaction between Reinkensmeyer’s group and the author of this chapter led to the idea that Clinicians are increasingly questioning the value mechatronic haptic robots could be used to measure of traditional measures of pulmonary function such motor control during development in normal chil- as the forced expiratory volume in 1 s (FEV1) so dren (Fig. 3.7). Children do not typically appear to commonly used to assess childhood asthma (Spahn move with the same skill and dexterity as adults, et al. 2004). A growing body of research has focused although they can still improve their motor perform- on arguably more direct measurements of lung ance in specific tasks with practice. inflammation, such as exhaled NO (Paredi et al. One possible explanation is that their motor per- 2002). NO performs many important functions in formance is limited by an inherently higher level of the lungs and can be detected in the exhaled breath movement variability, but that their motor adaptive of humans. Inflammatory diseases such as asthma ability is robust to this variability. To test this hypo- and cystic fibrosis alter exhaled NO levels. This has thesis, Takahashi et al. (2003) examined motor adap- generated interest in utilizing exhaled NO as a non- tation of 43 children (aged 6–17 years) and 12 adults invasive marker of lung inflammation. as they reached while holding the tip of a lightweight However, the exchange dynamics of NO are robot. The robot applied either a predictable, velocity- markedly different from the respiratory gases (oxy- dependent field (the ‘mean field’) or a similar field gen and carbon dioxide) whose exchange occurs that incorporated stochastic variation (the ‘noise predominantly in the alveolar region. In contrast, field’), thereby further enhancing the variability of NO exchange occurs in both alveolar and airway the subjects’ movements. Children exhibited greater compartments, and is thus highly dependent on the initial trial-to-trial variability in their unperturbed exhalation flow rate. This feature of NO exchange movements, but were still able to adapt comparably has confounded interpretation of exhaled NO in a to adults in both the mean and noise fields. Further- variety of clinical and physiological settings. Given more, the youngest children (aged 6–8 years) were the nature of NO exchange dynamics, and the multi- able to reduce their variability with practice to levels system physiological responses to exercise, it is not comparable to the remaining children groups, though surprising that there are inconsistencies in the not as low as adults. These results indicate that chil- reports of the impact of exercise on exhaled NO. dren as young as 6 years possess adult-like neural Recently, a number of investigators have devel- systems for motor adaptation and internal model oped paradigms to distinguish alveolar and air- formation that allow them to adapt to novel dynamic way contributions to exhaled NO (George 2004) environments as well as adults on average despite (Fig. 3.8). This approach provides greater specificity increased neuromotor or environmental noise. than exhaled concentration alone, and thus may Performance following adaptation is still more be able to address several unresolved questions variable than adults, however, indicating that move- regarding the impact of exercise on NO exchange. ment inconsistency, not motor adaptation inability, Given the nature of NO exchange dynamics, and the ultimately limits motor performance by children multisystem physiological responses to exercise, it and may thus account for their appearance of inco- is not surprising that there are inconsistencies in ordination and more frequent motor accidents (e.g. the reports of the impact of exercise on exhaled spilling, tripping). The results of this study also sug- NO. After exercise, exhaled NO concentration has gest that movement variability in young children been reported to be increased (Bauer et al. 1994), 32 chapter 3

Reach time (s) Spatial variability (cm)

1.5 1.5 **

1.0 1.0

0.5 0.5

0 0 6–8 9–12 13–17 >17 11–20 Last 10 Noise+ (a) Trials (b) Trials

Temporal variability (s) Scoring rate (%) 100 * * 6–8 1.0 9–12 80 13–17 0.8 >17 60 0.6 40 0.4

20 0.2

0 0 11–20 Last 10 Noise+ 11–20 Last 10 Noise+ (c) Trials (d) Trials

Fig. 3.7 Motor performance measures. (a) Reach time depended on age grouping (analysis of variance [ANOVA] linear contrast, p < 0.001). (b) Spatial variability depended on age grouping early in the experiment (mean over trials 11–20, ANOVA linear contrast over children groups, p < 0.001) but became independent of age grouping (p = 0.99) by the end of the experiment (i.e. last 10 trials). Children aged 6–8 years significantly reduced their spatial variability (paired one-sided t-test; ** p < 0.01) by the end of the experiment, but adults still maintained significantly lower levels compared to all other children groups (ANOVA planned comparison, p = 0.002). Spatial variability was significantly increased by the end of the noise field (‘noise+’ indicates last 10 trials of noise field) for all age groups (paired one-sided t-test, p < 0.001 all age groups) to levels that did not depend on age grouping (ANOVA linear contrast, p = 0.39). (c) Temporal variability also depended on age grouping over trials 11–20 (ANOVA linear contrast over children groups, p < 0.001) but became independent of age grouping by the last 10 trials of experiment (p = 0.38). Children aged 6–8 years significantly reduced their temporal variability (paired one-sided t-test, * p ≤ 0.05) by the end of the experiment. (d) Children aged 6–8 years improved their timing score rate (percent ‘just right’, t-test, *p ≤ 0.05) by the end of the experiment. Error bars show standard deviation across subjects. (Data from Takahashi et al. 2003.) exercise testing 33

= ′ JawNO JawNO – DawNOCair The latter suggests endogenously produced NO = may be useful to probe metabolic and structural fea- DawNO (CawNO – Cair) ′ tures of the airways during exercise. JawNO

Linking exercise testing to the new biology Calv, ss Cexh(t) The new technologies of genomic profiling, pro- teomics and flow cytometry have opened novel DawNO*Cair venues of research for exercise physiologists. The Alveolar region Airway region recent discoveries of the impact of physical activity (compartment #1) (compartment #2) on stress, inflammatory and immune function (Fleshner et al. 2003; Pedersen et al. 2003; Shephard Fig. 3.8 Schematic of two-compartment model used to describe nitric oxide (NO) exchange dynamics. Exhaled 2003) have created a paradigm shift in our ability no NO concentration (CE ) is the sum of two contributions, to understand the link between physical activity the alveolar region and the airway region, which depends and health. For example, Fehrenbach et al. (2000) on three flow-independent parameters: maximum total recently examined the role of exercise and fitness on volumetric flux of NO from the airway wall (JawNO, pls), 1 leukocyte expression of key immune modulators, diffusing capacity of NO in the airways (DawNO, pls , 1 no the heat shock proteins (HSPs) (Fig. 3.9). They found p.p.b. ), and steady-state alveolar concentration (CA , p.p.b.). JawNO, total flux (pls) of NO between the tissue and large increases in certain HSPs within the leuko- gas phase in the airway and is an inverse function of the cytes of subjects following exercise. HSPs inhibit no exhalation flow rate (VE); C , concentration of NO in the nuclear factor-κB, and this may explain the HSP car- gas phase within the airway compartment. (From George dioprotective effect that has been noted previously et al. 2004.) ( Joyeux et al. 1999; Powers et al. 2002). unchanged (Iwamoto et al. 1994), or decreased Summary (Maroun et al. 1995). By distinguishing alveolar and airway contributions to exhaled NO, Shin et al. Investigators are increasingly requesting human (2003) recently used a two-compartmental model performance/exercise studies on populations in of NO exchange during exercise in an attempt to whom the technology of ‘traditional’ exercise test- provide greater specificity than exhaled concentra- ing is inadequate; namely, the elderly, children and tion alone. They hoped to able to address several the disabled. The challenge we now face is to build unresolved questions regarding the impact of exer- on the past century of progress in exercise physi- cise on NO exchange. ology by investing in new technologies and approach

Significant changes were observed in J’awNO, that can help us understand how exercise is linked DawNO and CawNO 3-min post-exercise challenge, to fundamental disease processes. Moreover, the despite no significant changes in exhaled con- new approaches and technologies must be used for ± centration (CNOplat). DawNO (mean SD) increased integrating exercise testing with innovative, multi- ± (37.1 44.4%), whereas J’awNO and CawNO decreased disciplinary, research tools in biology that can pro- (−7.27 ± 11.1%, −26.1 ± 24.6%, respectively) 3-min vide new insights into mechanisms of disease at the post-exercise. Shin et al. (2003) concluded that the systemic and cellular level. flow-independent NO parameters provide greater specificity in characterizing NO exchange. It seems Acknowledgement that exercise acutely enhances elimination of NO from airway tissue stores. This effect may be due to This work was supported in part by National enhanced ventilation or an enhanced ability of NO Institutes of Health Grant HD26939 and the UCI to diffuse from the airway tissue to the gas phase. Satellite GCRC MO1 RR00827. 34 chapter 3

1.5 * * 1.5

1.0 1.0 mRNA mRNA — —

HSP 27 0.5 HSP 70 0.5

0.0 0.0 Rest +0h +3h +24 h Rest +0h +3h +24 h (a)Time (b) Time

Fig. 3.9 Descriptional presentation of mRNA expression of heat shock proteins HSP27 (a) and HSP70 (b) in leukocytes of athletes at rest and immediately, 3 h and 24 h after the half-marathon (n = 12). Semiquantitative reverse transcription- polymerase chain reaction (RT-PCR) was performed to assess mRNA expression. HSP27, HSP70 and β-actin were amplified under conditions to allow relative comparisons for a given mRNA. The specific mRNA values are described in relative units normalized to transcript levels of β-actin. Each curve represents a single subject. *Significant changes compared with resting values (P < 0.05). (Data from Fehrenbach et al. 2000.)

References

Bauer, J.A., Wald, J.A., Doran, S. & Soda, in the growing child: from the bench to protection. Fundamental and Clinical D. (1994) Endogenous nitric oxide in the playground. Current Opinions in Pharmacology 13, 1–10. expired air: effects of acute exercise in Pediatrics 16(3), 286–292. Lamarra, N. (1982) Ventilatory control, humans. Life Sciences 55, 1903–1909. Fehrenbach, E., Niess, A.M., Schlotz, E. cardiac output, and gas exchange dynamics Booth, F.W., Gordon, S.E., Carlson, C.J. & et al. (2000) Transcriptional and during exercise transients in man. PhD Hamilton, M.T. (2000) Waging war on translational regulation of heat shock thesis. UCLA, Los Angeles. modern chronic diseases: primary proteins in leukocytes of endurance Lowhagen, O., Arvidsson, M., Bjarneman, prevention through exercise biology. runners. Journal of Applied Physiology P. & Jorgensen, N. (1999) Exercise- Journal of Applied Physiology 88, 89(2), 704–710. induced respiratory symptoms are not 774–787. Fleshner, M., Campisi, J. & Johnson, J.D. always asthma. Respiratory Medicine 93, Chance, B. (1994) Non-invasive (2003) Can exercise stress facilitate 734–738. approaches to tissue bioenergetics. innate immunity? A functional role for Lum, P., Reinkensmeyer, D., Mahoney, R., Biochemical Society Transactions 22, stress-induced extracellular Hsp72. Rymer, W.Z. & Burgar, C. (2002) Robotic 983–987. Exercise Immunology Review 9, 6–24. devices for movement therapy after Coggan, A.R., Habash, D.L., Mendenhall, George, S.C., Hogman, M., Permutt, S. & stroke: current status and challenges L.A., Swanson, S.C. & Kien, C.L. (1993) Silkoff, P.E. (2004) Modeling pulmonary to clinical acceptance. Topics in Stroke

Isotopic estimation of CO2 production nitric oxide exchange. Journal of Applied Rehabilitation 8, 40–53. during exercise before and after Physiology 96, 831–839. McMahon, T.A. (1984) Muscles, Reflexes, endurance training. Journal of Applied Iwamoto, J., Pendergast, D.R., Suzuki, H. and Locomotion. Princeton University Physiology 75, 70–75. & Krasney, J.A. (1994) Effect of graded Press, Princeton, NJ. Cooper, D.M. (1995) Rethinking exercise exercise on nitric oxide in expired air Maroun, M.J., Mehta, S., Turcotte, R., testing in children: a challenge. American in humans. Respiration Physiology 97, Cosio, M.G. & Hussain, S.N. (1995) Journal of Respiratory and Critical Care 152, 333–345. Effects of physical conditioning on 1154–1157. Joyeux, M., Godin-Ribuot, D., Yellon, endogenous nitric oxide output during Cooper, D.M., Nemet, D. & Galassetti, P. D.M., Demenge, P. & Ribuot, C. (1999) exercise. Journal of Applied Physiology 79, (2004) Exercise, stress, and inflammation Heat stress response and myocardial 1219–1225. exercise testing 35

Metra, M., Nodari, S., Raccagni, D. et al. following severe stroke. Experimental adaptation, in children. Journal of (1998) Maximal and submaximal Brain Research 143, 525–530. Neurophysiology 90, 703–711. exercise testing in heart failure. Journal of Scheuermann-Freestone, M., Madsen, P.L., Tipton, C.M. (1998) Contemporary Cardiovascular Pharmacology 32 (suppl. 1), Manners, D. et al. (2003) Abnormal exercise physiology: fifty years after the S36–S45. cardiac and skeletal muscle energy closure of Harvard Fatigue Laboratory. Paredi, P., Kharitonov, S.A. & Barnes, metabolism in patients with type 2 Exercises and Sport Science Review 26, P.J. (2002) Analysis of expired air for diabetes. Circulation 107, 3040–3046. 315–339. oxidation products. American Journal of Shephard, R.J. (2003) Adhesion molecules, Wasserman, K., Whipp, B.J., Koyal, S.N. & Respiratory and Critical Care Medicine 166, catecholamines and leucocyte Beaver, W.L. (1973) Anaerobic threshold S31–S37. redistribution during and following and respiratory gas exchange during Patten, C., Meyer, R.A. & Fleckenstein, J.L. exercise. Reviews on Environmental Health exercise. Journal of Applied Physiology 35,

(2003) T2 mapping of muscle. Seminars in 33, 261–284. 236–243. Musculoskeletal Radiology 7, 297–305. Shin, H.W., Rose-Gottron, C.M., Cooper, Wasserman, K., Hansen, J.E., Sue, D.Y. & Pedersen, B.K., Steensberg, A., Keller, P. D.M., Hill, M. & George, S.C. (2003) Whipp, B.J. (1987) Principles of Exercise et al. (2003) Muscle-derived interleukin- Impact of high-intensity exercise on Testing and Interpretation. Lea & Febiger, 6: lipolytic, anti-inflammatory and nitric oxide exchange in healthy adults. Philadelphia, PA. immune regulatory effects. Pflugers Medicine and Science in Sports and Exercise Zanconato, S., Cooper, D.M., Barstow, T.J. 13 Archiv 446, 9–16. 35, 995–1003. & Landaw, E. (1992) CO2 washout Powers, S.K., Lennon, S.L., Quindry, J. Spahn, J.D., Cherniack, R., Paull, K. & dynamics during intermittent exercise & Mehta, J.L. (2002) Exercise and Gelfand, E.W. (2004) Is forced expiratory in children and adults. Journal of Applied cardioprotection. Current Opinion in volume in 1 second the best measure of Physiology 73, 2476–2482. Cardiology 17, 495–502. severity in childhood asthma? American Zanconato, S., Buchthal, S., Barstow, T.J. Reinkensmeyer, D.J., McKenna, C.A., Journal Respiratory and Critical Care & Cooper, D.M. (1993) 31P-magnetic Kahn, L.E. & Kamper, D.G. (2002) Medicine 169, 784–786. resonance spectroscopy of leg muscle Directional control of reaching is Takahashi, C.D., Nemet, D., Rose-Gottron, metabolism during exercise in children preserved following mild/moderate C.M. et al. (2003) Neuromotor noise and adults. Journal of Applied Physiology stroke and stochastically constrained limits motor performance, but not motor 74, 2214–2218. Chapter 4

Measurement of Peptide Hormones

MARTIN BIDLINGMAIER, ZIDA WU AND CHRISTIAN J. STRASBURGER

Introduction ysis of samples in an extremely short time period is often required during competitive events, enforcing Methods to quantify peptide hormone concentra- the need for rapid analytical methods. tions in biological fluids are a basic requirement for During the last 30 years, enormous progress has medical practice and biomedical research in the been made in the attempt to develop analytical field of endocrinology. Such methods ideally should methods for peptide hormone quantification fulfill- be very specific, identifying precisely and quantify- ing the above mentioned criteria. However, from ing accurately only the hormone of interest without the early days of radioimmunoassays to the latest interference from closely related peptides or un- developments in mass spectrometry (MS) tech- related matrix components. They must be extremely niques, it remains crucial to be aware of the pitfalls sensitive, allowing the detection of peptide hor- each method includes to avoid misinterpretation mones at the low concentrations observed in physio- of the data generated. In some cases, it might be logical and pathophysiological situations. Usually, impossible to combine different goals by one peptide hormone concentrations in circulation are method: For example, there is no doubt that MS is a in the nano- or picomolar range, making the sensiti- unique tool to identify a molecule with an extremely vity of the methods a crucial issue. Finally, measure- high degree of certainty (Binz et al. 2003; Gam et al. ment methods should be easy and rapid, allowing 2003; Kast et al. 2003). However, today’s MS methods application to large series of samples in an appropri- still are complex, time consuming and in many cases ate time. These methodological aspects of hormone require a very sophisticated sample preparation. measurement become even more important when The progress made in the applicability of MS tech- the analyses are done in the framework of the fight nologies to analyze purified peptide hormone pre- against doping in sports. Because of the potentially parations is not paralleled by the same progress severe consequences for the athlete, the commercial in analyzing hormone mixtures in more complex and political interests involved and the complicated biological matrices like serum (Liu & Bowers 1997; ethical and legal background, the level of precision, Black & Bowers 2000; Wu, S.L. et al. 2002), and the specificity and reliability for hormone measurements high degree of certainty achieved under optimal must be extremely high in this field. For example, conditions by no means indicates that MS would not the cross-reactivity from specific molecular isoforms be susceptible to interferences (Annesley 2003). of peptide hormones in immunoassays is rarely Finally, the costs of the equipment necessary are still known for commercial assays frequently used in much higher than those for immunological hor- clinical practice, whereas this is a crucial point for mone measurement methods. The rapid develop- several doping tests described below (chorinonic ment of new techniques in the area of MS together gonadotropin [hCG], erythropoietin [EPO], human with the development of instruments designed for growth hormone [hGH]). Finally, large-scale anal- high throughput analyses, the introduction of 36 measurement of peptide hormones 37

automated systems for sample preparation and the classical ‘competitive’ immunoassay and the ‘sand- increase in sensitivity recently achieved by research wich type’ immunoassay (Figs 4.1 and 4.2). In a groups might change the picture in the future (for competitive immunoassay, the hormone present in review see Binz et al. 2003). However, until today a sample ‘competes’ for antibody binding with a the vast majority of hormone analyses in clinical labeled form of the same hormone added to the practice and research is still done by the ‘traditional’ sample. The higher the endogenous hormone con- immunoassay techniques. Thus, this article is centration in a sample is, the lower the probability focused on the pitfalls and potentials of peptide hor- is that a labeled hormone molecule (called ‘tracer’) mone measurements by immunoassay techniques. binds to the antibody. Thus, the hormone concen- tration in the sample is inversely correlated to the General aspects of immunoassay signal obtained in the assay. Depending on the label methods used, competitive assays have been named radio- immunoassay (RIA), enzyme-immunoassay (EIA), Immunoassays employ the unique capability of luminescence-immunoassay (LIA) or fluorescence- specific immunoglobulins (antibodies) to recognize immunoassay (FIA). In contrast, ‘sandwich type’ and bind a certain three-dimensional structure on immunoassays consist of a ‘capture antibody’, which the surface of a molecule (the so-called ‘epitope’). binds the hormone from the sample, and a labeled This antigen–antibody reaction provides a high ‘detection antibody’, which is directed against degree of specificity. In combination with the use another epitope on the hormone surface and thus of powerful labeling substances for antibodies or allows ‘staining’ of the hormone molecules bound hormones, the dynamic range of immunoassays by the capture antibody. The more the hormone is could be greatly improved. Traditionally, radioactiv- present in a sample, the more the hormone is bound ity had been used as a detection system in hormone by the capture antibody, which in turn is trans- assays. Driven by concerns regarding environmen- lated into a signal by the ‘detection antibody’. Thus, tal, economical and health aspects of radiation, the signal obtained is directly proportional to the different non-isotopic methods have been devel- amount of hormone present in the sample. Similar oped for signal detection: These newer systems to the above mentioned nomenclature for the com- employ either an enzyme-catalyzed colorimetric petitive assays, the names of sandwich type immuno- or chemiluminescence reaction oraalternativelyaa assays are related to the detection system used: fluorescence dye for signal detection. Among the radioimmunometric assay (IRMA), enzyme linked advantages of these non-isotopic methods is the immunosorbent assay (ELISA), luminometric assay stability of the labelaradioactive labels undergo (ILMA) or fluorometric assay (IFMA). a decay, making lot-to-lot differences a common The difference in assay design between competit- problem. In some cases, the sensitivity of modern ive and sandwich type immunoassays has major non-isotopic detection systems exceeds that of implications on their applications and on the inter- radioactive labels, making their use even more pretation of the results: whereas competitive assays popular. However, in general the sensitivity of an require only one antibody andacorrespondinglya immunoassay method primarily depends on the one epitope, sandwich type immunoassays require affinity of the specific antibodies used, and often the two different antibodies and thus two different detection label is chosen simply because a certain epitopes. These epitopes must be in spatial distance, measurement device is present in a laboratory. because otherwise binding of the capture antibody would interfere with the binding of the detection Technical aspects of immunoassay antibody. Accordingly, sandwich type immuno- technology and its implications assays require a ‘larger molecule’ and therefore are used to measure larger peptide hormones like From a methodological point of view, immunoas- insulin, growth hormone (GH) or EPO, whereas says can be divided in two major subtypes: the the smaller peptides (adrenocorticotropic hormone 38 chapter 4

Signal 1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 0.0 1.0 10.0 100.0 (a) (b) Hormone (ng·mL–1)

Fig. 4.1 (a) Competitive assayaprinciple. Labeled hormone (‘tracer‘) and hormone from the sample compete for binding to antibody. (b) Competitive assayacalibration curve. The higher the hormone concentration, the lower the probability that a labeled tracer molecule is bound. Consequence: the more hormone, the lower the signal.

Signal 1 000 000

100 000

10 000

1000 0.1 1.0 10.0 100.0 (a) (b) Hormone (ng·mL–1)

Fig. 4.2 (a) Sandwich assayaprinciple. (i) Hormone from the sample binds to capture antibody. (ii) Bound hormone is detected by labeled antibody. (b) Sandwich assayacalibration curve. The higher the hormone concentration, the higher the probability that a labeled detection antibody is bound. Consequence: the more hormone, the more signal. measurement of peptide hormones 39

[ACTH], corticotropin-releasing hormone [CRH], [IGF-I] contains only about 40% IGF-I [Quarmby growth hormone releasing hormone [GHRH]) and et al. 1998]). To make the situation more complic- steroid hormones frequently are measured by com- ated, many peptide hormones naturally occurring petitive assays. The advantage of the sandwich type in the human body are a mixture of molecular assay format is that the recognition of a hormone isoforms rather than a homogenous substance molecule requires the presence of two independent (Nagy et al. 1994). This has been documented in very epitopes, making the methodaat least theoretically much detail, especially for hCG (Birken et al. 2003; amore specific and less prone to cross-reactivity of Lottersberger et al. 2003), but also for hGH partially related peptides. In addition, the analyte (Baumann 1999; Boguszewski 2003). Depending on (the hormone to be measured) and the calibrator the specific question an investigator or clinician (the standard preparation used) are chemically attempts to answer by measuring a hormone’s con- identical, whereas in the case of competitive assays, centration, the answer which of these isoforms or the calibrator is a modified, labeled hormone. These which mixture of isoforms would be the suitable modifications can lead to differences in the affinity ‘reference’ preparation varies significantly. In 1991, of the antibody to the naturally occurring hormone Roger Ekins, one of the inventors of immunoassay in a sample and to the modified hormone used as technology wrote that standardization of immuno- the calibrator of the assay. assays for heterogeneous antigens is impossible (Ekins 1991), and the ongoing discussions on assay Standardization of immunoassays standardization clearly demonstrate that we are far from having reached a consensus (Roger & Lalhou It is important to mention that all immunoassay 1994; Quarmby et al. 1998; Ranke et al. 2001). As to methods are ‘relative’ in nature. This means, that the second assumption, one has to be aware that the concentration of the substance of interest in a for many peptide hormones high- or low-affinity sample is determined in comparison to the concen- ‘binding proteins’ naturally occur in circulation tration contained in the calibrator. More precisely, (Baumann et al. 1988). Present in a sample, these the concentration is determined by comparing the binding proteins can interfere with the hormone ability of the antibodies used to bind the hormone measurement by either ‘capturing’ the tracer mole- in a biological sample (e.g. a patient’s serum) and in cule in a competitive assay or by preventing anti- the calibrator sample, respectively. Theoretically, body binding (Fisker & Orskov 1996). Depending this measurement technique relies on three funda- on assay type, the interference from binding pro- mental assumptions: (i) the calibrator is identical to teins in the sample will lead to under or overesti- the substance of interest in its physicochemical mation of the hormone concentration. A source of properties and in its three-dimensional structure; interference with no relation to the hormone of (ii) the epitopes on the surface of the molecule of interest is the presence of heterophilic antibodies in interest are freely accessible for the antibody in both the patient’s sample. These antibodies can ‘link’ the the biological sample and in the calibrator; (iii) the capture to the detection antibody directly, leading matrix of the calibrator is identical to the matrix of to falsely increased assay results (Kricka 1999). It the sample. Obviously it is very difficult to realize has been described that such antibodies can be these assumptions in an assay for peptide hormones: induced by pets housed by the patient (Park et al. In many cases, no international reference prepara- 2003), but in many cases the origin remains unclear. tion (IRP) is available, whereas in other cases more Methods have been developed to avoid such inter- than one reference preparation exists. In the case of ference, but the issue remains a problem (Emerson hGH, the IRP 80/505 is of pituitary extraction, et al. 2003; Preissner et al. 2003). Finally, the matrix whereas the IRP 88/624 and IRP 98/574 are of used to dissolve the calibrator rarely is identical to recombinant origin. In addition, some of the refer- human serumain many cases, animal sera or buffers ence preparations are of very poor quality and supplemented with albumin are used. The beha- impure (e.g. the IRP for insulin-like growth factor I vior of antibodies sometimes can be influenced by 40 chapter 4

matrix components, leading to differences in the some reports occurring in the laymen’s press, a high assay results depending on the calibrator’s matrix. concentration of hGH, for example, by no means is Because all these factors can have profound effects sufficient to prove that an athlete has been using on the hormone concentration measured, a careful recombinant hGH (Armanini et al. 2002). interpretation of immunoassay results is required. Another problem with the detection of doping Every single method must be evaluated in terms of with peptide hormones is their recombinant origin. cross-reacting substances, and ideally the exact Derived from the expression of the protein encoded nature of the epitope recognized by the antibodies by the human gene sequence transferred to an in involved is known. Obviously, ‘references ranges’ vitro system, the artificially produced hormones for hormone concentrations are method dependent are identical to the naturally occurring hormone in (Strasburger et al. 1996, 2001; Stenman et al. 1997; their amino acid sequence and thus in their phy- Quarmby et al. 1998; Wood 2001; Sharpe et al. 2002), sicochemical properties. Once a peptide hormone and the simple reference to ‘general’ normative data was injected, for many years it was impossible from textbooks should be avoided. Finally, there is to judge about the origin of a single hormone still a need for the development of IRPs for many molecule. peptide hormones. As demonstrated most recently Detecting peptide hormone doping is further com- for hCG (Birken et al. 2003), the development of plicated by the fact thatacompared to the rather suitable reference reagents is a sophisticated task, simple, small and stable steroid hormone molecules but helps to eliminate some factors of uncertainty in apeptide hormones are larger molecules exhibiting immunoassay measurements. a very sensitive three-dimensional structure. In many cases, peptide hormones are rapidly degraded, Peptide hormone measurements in metabolized and cleaved. The renal excretion in- doping tests volves complex processes, many of them being still poorly understood. Furthermore, the peptide hor- The Olympic Movement Anti-Doping Code as mone concentrations found in urine are often even of January 1st, 2003 contains the statement that much lower than those in blood. Therefore, the ‘the presence of an abnormal concentration of an material traditionally used in doping analyticsa endogenous hormone in class (E) or its diagnostic urineais of limited value for many peptide hor- marker(s) in the urine of a competitor constitutes mone tests. Blood sampling is required, bearing all an offence unless it has been proven to be due the ethical and legal problems discussed elsewhere to a physiological or pathological condition’ (see (Birkeland & Hemmersbach 1999). www.wada-ama.org/docs). Unfortunately, in the Finally, the problem to establish doping tests for case of peptide hormones the definition of an peptide hormones is related to the methodological ‘abnormal concentration’ is extremely difficult or difference in the measurement of peptide versus even impossible in many cases. Several peptide steroid hormones outlined above. For several years, hormones are secreted by the human body in a steroid hormones have been identified by gas chro- pulsatil rather than a continuous manner, or the matography/mass spectrometry (GC/MS), and secretion exhibits a circadian profile. Furthermore, cutting edge experience together with the appro- their concentration is usually influenced not only priate equipment is present in the International by age and gender but also by environmental fac- Olympic Committee (IOC) accredited laboratories. tors (temperature, altitude), stress (psychological In many cases, GC/MS determination is referred to or physiological), sleep, nutrition state or training as a ‘reference technology’ or ‘gold standard’ for the status. Many peptide hormones show an extremely quantitation of hormones. Unfortunately, this tech- short half-life time, leading to highly floatable nology has not been available for peptide hormone concentrations in circulation. Therefore, a simple analysis for many years, and there still is no estab- measurement of the hormone concentration in only lished procedure shown to be ready to use in doping a few cases (like hCG in men) is sufficient to demon- detection (Bowers 1997; Hilderbrand et al. 2003). All strate the misuse of the hormone. In contrast to methods currently available or at least expected to measurement of peptide hormones 41

be available in the near future rely on immunoas- recently that an international study group was able say techniques, in the case of EPO complemented by to establish six different IRPs for the most important an electrophoretic confirmation based on isoelectric hCG isoforms (Birken et al. 2003), and their intro- focusing (IEF). duction is expected to make assay results more comparable. Specific aspects in doping analytics Most recently, a tandem mass spectrometric ana- lysis technique has been described for confirmation of positive hCG tests (Gam et al. 2003). Following Chorinonic gonadotropin tryptic digestion, it was possible to identify a Chorinonic gonadotropin (hCG) is an important marker peptide (βT5) providing a specific finger- pregnancy-associated hormone that stimulates print for hCG. Combined with a specific immunoex- endogenous steroid production. The abuse of hCG traction procedure, the proposed method currently in male athletes has been described, aiming to is able to quantify hCG concentrations as low as enhance endogenous anabolic steroids without 5 IU·mL–1 Whether this MS approach in the field of changing the testosterone/epitestosterone ratio peptide hormone analysis will be reliable and (Stenman et al. 1997). Normally, hCG concentrations practicable in doping analytics remains to demon- in non-pregnant women and especially in men strated in large-scale studies. are extremely low. Only in a few pathological situations such as testicular cancer are know to be Erythropoietin accompanied by elevated concentrations in men (Lottersberger et al. 2003). Therefore, the case of The introduction of a doping test for recombinant hCG is a comparably simple situation in peptide human erythropoietin (rhEPO) at the 2000 Sydney hormone doping analytics: The presence of high Olympic games represents a major step forward in hCG levels in a male athlete are highly suspicious the fight against doping (Kazlauskas et al. 2002). For for doping. However, even frequently used in clin- the first time, a detection method for a peptide hor- ical practice, the quantification of hCG remains mone able to discriminate between the recombinant problematic. As a member of the glycoprotein hor- and the endogenous form of the hormone has been mone family, hCG is comprised of two distinct implemented into the official doping test program. subunits (α and β). The α subunit is shared with all To the present, detection of doping with rhEPO other members of this hormone family, whereas the relies on two different methods. The first is the urine- β subunit is specific and responsible for receptor based test as described by Lasne and de Ceaurriz binding and biological activity. The carboxyl termi- (Lasne & de Ceaurriz 2000; Lasne et al. 2002), which nal of the hCG molecule contains a highly glycosy- utilizes an IEF method in combination with a tech- lated region. Several molecular isoforms of hCG nique called ‘double-blotting’ enabling the visual- have been identified, and their heterogeneity con- ization of protein bands with a greatly reduced tributes substantially to the large between-method background staining (Lasne 2001, 2003). The ration- differences in existing hCG immunoassays (Cole & ale of the test is to detect the pattern of isoforms of Kardana 1992; Cole 1997; Cole et al. 2001). The situa- EPO in a sample, which is different between recom- tion is even worse in urine, where the spectrum of binant and endogenous EPO (Lasne et al. 2002; isoforms and degradation products is more com- De Frutos et al. 2003). The reason seems to be that plicated than in serum (Birken et al. 1996; O’Connor especially glycosylation are sensitive to the cellular et al. 1999). environment where a protein is produced. Because In addition to the heterogeneity of the analyte rhEPO is produced mainly in Chinese hamster in a sample, for many years the standard prepara- ovary cells, the pattern of glycosylation is different tions used to calibrate hCG assays contained sub- from that in the human kidney (Sasaki et al. 1987; stantial amounts of contaminating variants of hCG, Rice et al. 1992). Meanwhile, this test procedure has which reacted to a variable degree with the anti- been implemented in several IOC accredited labor- bodies in different immunoassays. It has been only atories and is used in doping analyses from many 42 chapter 4

sports associations (Catlin et al. 2002; Breidbach et al. been shown that at least two different approaches 2003). are able to discriminate whether or not an athlete A major problem with this technique is that it is has taken rhGH. The test methods developed are very expensive and time-consuming. In addition, still in the validation process and not yet imple- the likelihood of catching a cheating athlete is given mented as an official doping test, but this is only for the first 3–4 days after injection (Wide et al. expected to happen soon. 1995; Souillard et al. 1996). The attempts to circum- A major problem remains that in contrast to the vent these problems led to the development of situation with EPO, none of the proposed methods methods based on detecting changes in hematologic for detecting hGH doping today can be applied in parameters after injection of rhEPO. These changes urine samples. This is primarily due to the extremely have been demonstrated to be more pronounced low hGH concentrations found in urine making the than can be expected from the normally observed analysis not possible by currently available tech- variability, and some of them are detectable for niques. Furthermore, hGH secretion to urine is a as long as 4 weeks after rhEPO administration complex process which seems to be highly variable (Parisotto et al. 2001; Gore et al. 2003). The so-called and is still poorly understood (Saugy et al. 1996). ‘second generation blood test’ involves the measure- The IEF method used for EPO in urine samples is ment of EPO and of the soluble transferin receptor, not useful in the case of hGH, because hGH does both done by immunoassay techniques. Being much not contain any glycosylation sites. According to easier, cheaper and providing a much greater our knowledge today, the hGH molecules of recom- window of opportunity to detect rhEPO misuse, binant origin are almost identical to the main frac- this test procedure unfortunately relies on blood tion of hGH molecules secreted by the pituitary samples and gives a comparably high rate of false gland, and no distinct physico-chemical properties positives when the cut off is set to a level where of rhGH have been described. not too many abusers are missed. Thus, a current However, even no glycosylation sites are present; approach is to use the blood-based tests as a screen- hGH in circulation consists of a mixture of molecu- ing test to identify those samples which are more lar isoforms (Baumann 1999). The investigation of likely to contain rEPO, followed by the urine-based these isoforms is not as advanced as in case of hCG, IEF test as a confirmatory test applied only to suspi- but during the last years it was possible to identify cious samples. Gore et al. (2003) have clearly shown at least some of the major components. In addition that using the second-generation blood test for to the 22 kDa major isoform, consisting of 191 amino screening can drastically reduce the costs per posit- acids, a shorter 20 kDa hGH isoform lacking amino ive sample as judged by the urine test. In addition, acids 32–46 is the second most abundant form of a lot of work has been done to establish normative hGH in circulation (Hashimoto et al. 1998; Tsushima data and to identify factors potentially influencing et al. 1999; Leung et al. 2002). There are other even the test outcome (Sharpe et al. 2002; Ashenden et al. shorter isoforms described, but they are observed 2003; Parisotto et al. 2003). However, it is important less constantly and are not yet fully analyzed. Some to remember that the EPO molecules found in a of them have been shown to be cleaved or degraded urine sample can only be proven to be of recombin- hGH molecules. The isoforms of hGH seem to exist ant origin by the IEF urinary approach visualizing in monomeric, dimeric and multimeric complexes the typical glycosylation pattern. formed by either identical (homodimers) or differ- ent isoforms (heterodimers). Many of the hGH effects in the body are mediated Growth hormone and growth hormone- through a factor called insulin-like growth factor I dependent hormones (IGF-I). Produced mainly in the liver, but even The detection of doping with recombinant human locally in cartilage, bone and many other tissues, growth hormone (rhGH) has been thought to be IGF-I is secreted to the blood, where it is bound impossible for many years. However, recently it has by specific binding proteins (Le Roith et al. 2001). measurement of peptide hormones 43

The most important ones are IGF binding protein-3 clonal antibodies. Polyclonal antisera, however, are (IGFBP-3) and the acid labile subunit (ALS), both inherently limited in their quantity andaonce a produced under hGH control as well. At least for batch is finishedacannot be reproduced identically. IGFBP-3 it has been shown that this protein exerts Therefore, one can understand that the international effects independently from IGF-I binding and there- antidoping agencies put the development of suit- fore can be a seen as a ‘peptide hormone’ itself. able monoclonal antibodies onto the agenda. Once Together, IGF-I, IGFBP-3 and ALS form a 150 kDa the methodological details have been clarified, this ternary complex, possessing an increased half-life in test method could be very useful, comparably to the comparison to each molecule alone. blood test already in use for EPO doping detection. It was one strategy in the search for a suitable test Another approach more directly relies on the method to detect hGH doping to evaluate whether measurement of hGH: In contrast to the hGH or not the increase in the pharmacodynamic end- isoform spectrum secreted by the pituitary gland, points of hGH action, especially the increase in the rhGH as produced in bacteria usually consist of components of the ternary complex might exceed 22 kDa hGH only. The recombinant production of the normally observed variability (Dall et al. 2000). 20 kDa hGH has been described, but until today Such a test would include the advantage that the this preparation has been used only in a few clinical half-life of the pharmacodynamic endpoints of hGH trials. Recombinant hGH used to treat GH deficient action exceeds that of hGH, making possible a children, adolescents and adults is 22 kDa hGH, and longer window of opportunity for detecting hGH the same seems to apply for the hGH preparations abuse. An international consortium of researchers abused for doping purposes in sports. This ‘unifor- conducted a series of large-scale studies to invest- mity’ or ‘lack of heterogeneity’ in the recombinant igate the behavior of such pharmacodynamic hGH preparations in comparison to the many iso- endpoints of hGH action in relation to various con- forms secreted by the human pituitary gland and ditions like acute and chronic exercise, age, gender, normally present in circulation forms the basis for ethnic background or injury (Wallace et al. 1999, the so-called ‘differential immunoassay approach’ 2000; Longobardi et al. 2000; Ehrnborg et al. 2003). to detect doping with recombinant hGH (Wu, Z. The main outcome of these studies was that re- et al. 1999): the injection of recombinant monomeric peated administration of exogenous rhGH indeed 22 kDa hGH increases the relative abundance of this induces changes in pharmacodynamic endpoints, hGH isoform in circulation. This change in the iso- which can be discriminated from changes induced form composition is further increased by the reduc- by exercise or other stimulators. In more detail, the tion of pituitary hGH secretion due to the negative researchers developed a statistical model including feedback mechanism observed under chronic rhGH more than one of the pharmacodynamic endpoints treatment (Wallace et al. 2001a). Screening of mono- and tailored this model specifically for each gender. clonal antibodies raised against different hGH However, a crucial point is that the immunoassays preparations led to the development of two differ- used for determinations of the factors are suffi- ent immunoassays for hGH. The capture antibody ciently evaluated and that method specific reference of the first assay preferentially binds the 22 kDa iso- ranges are known. Not all commercially available form of human GH, whereas the capture antibody methods can fulfill these criteria, and therefore a of the second assay preferentially binds ‘pituit- careful selection of the immunoassays to be used is ary derived hGH’, consisting of multiple isoforms mandatory. Furthermore, because a highly sophist- (Bidlingmaier et al. 2000). Measuring on serum sam- icated statistical model is the backbone of this ple by both assays allows to calculate the relative test method, the variability of the assays must be abundance of 22 kDa hGH versus the other isoforms known, and a continuous supply with an identical (‘total hGH’), and thus identifying samples contain- batch of antibodies must be ensured. This could pro- ing an inappropriately high content of 22 kDa hGH. vide problems because many of the immunoassays It could be demonstrated that the change in the used until today rely on polyclonal rather than mono- molecular isoform composition is uniquely related 44 chapter 4

to rhGH application, whereas the increase in cir- variability must be determined exactly with respect culating hGH after exercise is accompanied by an to the potential impact of this variability on the increase in the non-22 kDa hGH isoforms as well ratio result. A reduction of the variability could be (Wallace et al. 2001b). Meanwhile, the original achieved by using the same capture antibody for approach has been dramatically improved in terms both, the 22 kDa preferential and the total hGH of sensitivity (Bidlingmaier et al. 2003). This was assay: on one microtiterplate, the upper half is made possible by the identification of new mono- covered by the 22 kDa assay capture monoclonal clonal antibodies. In addition, an independent antibody (mAb), whereas the lower half of the plate confirmatory set of assays was developed, based is covered by the total hGH capture mAb. After on new monoclonal capture antibodies, targeting adding calibrators, controls and samples to each independent epitopes. The latter is a prerequisite for half of the plate, the whole plate is covered by the the acceptability of immunoassays in a doping test same detection mAb. This reduces significantly the setting: each assay has to be confirmed by another variability always included in distributing samples assay, targeting independent structures of the mole- on two different plates (Bidlingmaier et al. 2000). cule of interest and thus providing further evidence for the identity of the molecule. Conclusions Inherently, the differential immunoassay ap- proach is limited to the detection of rhGH abusea The immunoassay techniques used for peptide hor- cadaveric hGH preparations, derived from extrac- mone measurements represent very sensitive tools tion of pituitary glands, have an isoform pattern to quantify the concentrations of the hormones in which cannot be discriminated from that naturally various biological fluids specifically. Appropriately occurring in the human body. Furthermore, due to selected antibodies, which must be characterized the very short half-life time of hGH in circulation regarding their epitopes, affinities and possibly (about 15 min), the window of opportunity for the cross-reacting peptides, confer a high degree of detection remains limited to 24–36 h. Obviously, specificity in hormone measurements, making these even the development of more sensitive assays will methods suitable even in the setting of doping test not overcome this limitation, because after dis- procedures. Being much easier, cheaper and less appearance of the recombinant substance and cessa- time consuming than other methods for peptide tion of the negative feedback, the normal pituitary hormone analysis, immunoassays have a high isoform secretion has been shown to start again. On potential, especially in large-scale screening. How- the other hand the fact that rhGH has to be taken by ever, for each assay used, method specific norm- daily subcutaneous injections enhances the prob- ative data have to be established. Furthermore, ability to catch a cheating athlete in unannounced potentially confounding factors must be identified out-of-competition tests. and appropriate methods for their elimination must The differential immunoassay approach also be developed. If these precautions are taken into makes it necessary to rigorously validate the consideration, the accuracy of immunoassay ana- immunoassays selected. Furthermore, because a lyses can be extremely high and very suitable for ratio is calculated, the within- and between-method use in a forensic environment.

References

Annesley, T.M. (2003) Ion suppression in British Journal of Sports Medicine 36(2), Baumann, G. (1999) Growth hormone mass spectrometry. Clinical Chemistry 148–149. heterogeneity in human pituitary and 49(7), 1041–1044. Ashenden, M.J., Gore, C.J., Parisotto, R. plasma. Hormone Research 51 (suppl. 1), Armanini, D., Faggian, D., Scaroni, C. et al. (2003) Effect of altitude on second- 2–6. & Plebani, M. (2002) Growth hormone generation blood tests to detect Baumann, G., Amburn, K. & Shaw, M.A. and insulin-like growth factor I in a erythropoietin abuse by athletes. (1988) The circulating growth hormone Sydney Olympic gold medallist. Haematologica 88(9), 1053–1062. (GH)-binding protein complex: a major measurement of peptide hormones 45

constituent of plasma GH in man. Cole, L.A. & Kardana, A. (1992) regulatory issues in antidoping Endocrinology 122(3), 976–984. Discordant results in human chorionic programs in sport. Current Sports Bidlingmaier, M., Wu, Z. & Strasburger, gonadotropin assays. Clinical Chemistry Medicine Reports 2(4), 226–232. C.J. (2000) Test method: GH. Bailliere’s 38(2), 263–270. Kast, J., Parker, C.E., van der Drift, K. Best Practice and Research. Clinical Cole, L.A., Shahabi, S., Butler, S.A. et al. et al. (2003) Matrix-assisted laser Endocrinology and Metabolism 14(1), (2001) Utility of commonly used desorption/ionization directed nano- 99–109. commercial human chorionic electrospray ionization tandem mass Bidlingmaier, M., Wu, Z. & Strasburger, gonadotropin immunoassays in the spectrometric analysis for protein C.J. (2003) Problems with GH doping diagnosis and management of identification. Rapid Communications in in sports. Journal of Endocrinological trophoblastic diseases. Clinical Chemistry Mass Spectrometry 17(16), 1825–1834. Investigation 26, 924–931. 47(2), 308–315. Kazlauskas, R., Howe, C. & Trout, G. Binz, P.A., Hochstrasser, D.F. & Appel, Dall, R., Longobardi, S., Ehrnborg, C. (2002) Strategies for rhEPO detection in R.D. (2003) Mass spectrometry-based et al. (2000) The effect of 4 weeks of sport. Clinical Journal of Sport Medicine proteomics: current status and potential supraphysiological growth hormone 12(4), 229–235. use in clinical chemistry. Clinical administration on the insulin-like Kricka, L.J. (1999) Human anti-animal Chemistry and Laboratory Medicine 41(12), growth factor axis in women and men. antibody interferences in immunological 1540–1551. GH-2000 Study Group. Journal of Clinical assays. Clinical Chemistry 45(7), 942–956. Birkeland, K.I. & Hemmersbach, P. (1999) Endocrinology and Metabolism 85(11), Lasne, F. (2001) Double-blotting: a The future of doping control in athletes. 4193–4200. solution to the problem of non-specific Issues related to blood sampling. Sports De Frutos, M., Cifuentes, A. & Diez-Masa, binding of secondary antibodies in Medicine (Auckland, NZ) 28(1), 25–33. J.C. (2003) Differences in capillary immunoblotting procedures. Journal Birken, S., Kovalevskaya, G. & O’Connor, electrophoresis profiles of urinary and of Immunological Methods 253(1–2), J. (1996) Metabolism of hCG and hLH to recombinant erythropoietin. 125–131. multiple urinary forms. Molecular and Electrophoresis 24(4), 678–680. Lasne, F. (2003) Double-blotting: a Cellular Endocrinology 125(1–2), Ehrnborg, C., Lange, K.H., Dall, R. et al. solution to the problem of nonspecific 121–131. (2003) The growth hormone/insulin-like binding of secondary antibodies in Birken, S., Berger, P., Bidart, J.M. et al. growth factor-I axis hormones and bone immunoblotting procedures. Journal (2003) Preparation and characterization markers in elite athletes in response to a of Immunological Methods 276(1–2), of new WHO reference reagents for maximum exercise test. Journal of Clinical 223–226. human chorionic gonadotropin and Endocrinology and Metabolism 88(1), Lasne, F. & de Ceaurriz, J. (2000) metabolites. Clinical Chemistry 49(1), 394–401. Recombinant erythropoietin in urine. 144–154. Ekins, R. (1991) Immunoassay Nature 405(6787), 635. Black, R.S. & Bowers, L.D. (2000) Mass standardization. Scandinavian Journal Lasne, F., Martin, L., Crepin, N. & de spectrometric characterization of of Clinical and Laboratory Investigation. Ceaurriz, J. (2002) Detection of the β-subunit of human chorionic Supplementum 205, 33–46. isoelectric profiles of erythropoietin in gonadotropin. Methods in Molecular Emerson, J.F., Ngo, G. & Emerson, S.S. urine: differentiation of natural and Biology 146, 337–354. (2003) Screening for interference in administered recombinant hormones. Boguszewski, C.L. (2003) Molecular immunoassays. Clinical Chemistry 49(7), Analytical Biochemistry 311(2), 119–126. heterogeneity of human GH: from basic 1163–1169. Le Roith, D., Bondy, C., Yakar, S., Liu, J.L. research to clinical implications. Journal Fisker, S. & Orskov, H. (1996) Factors & Butler, A. (2001) The somatomedin of Endocrinological Investigation 26(3), modifying growth hormone estimates hypothesis: 2001. Endocrine Reviews 274–288. in immunoassays. Hormone Research 22(1), 53–74. Bowers, L.D. (1997) Analytical advances 46(4–5), 183–187. Leung, K.C., Howe, C., Gui, L.Y. in detection of performance-enhancing Gam, L.H., Tham, S.Y. & Latiff, A. (2003) et al. (2002) Physiological and compounds. Clinical Chemistry 43(7), Immunoaffinity extraction and tandem pharmacological regulation of 20-kDa 1299–1304. mass spectrometric analysis of human growth hormone. American Journal of Breidbach, A., Catlin, D.H., Green, G.A. chorionic gonadotropin in doping Physiology. Endocrinology and Metabolism et al. (2003) Detection of recombinant analysis. Journal of Chromatography. B, 283(4), E836–843. human erythropoietin in urine by Analytical Technologies in the Biomedical Liu, C. & Bowers, L.D. (1997) Mass isoelectric focusing. Clinical Chemistry and Life Sciences 792(2), 187–196. spectrometric characterization of nicked 49(6 Pt. 1), 901–907. Gore, C.J., Parisotto, R. Ashenden, M.J. fragments of the β-subunit of human Catlin, D.H., Breidbach, A. Elliott, S. & et al. (2003) Second-generation blood chorionic gonadotropin. Clinical Glaspy, J. (2002) Comparison of the tests to detect erythropoietin abuse by Chemistry 43(7), 1172–1181. isoelectric focusing patterns of athletes. Haematologica 88(3), 333–344. Longobardi, S., Keay, N., Ehrnborg, C. et al. darbepoetin alfa, recombinant human Hashimoto, Y., Ikeda, I., Ikeda, M. et al. (2000) Growth hormone (GH) effects on erythropoietin, and endogenous (1998) Construction of a specific and bone and collagen turnover in healthy erythropoietin from human urine. sensitive sandwich enzyme adults and its potential as a marker of Clinical Chemistry 48(11), 2057–2059. immunoassay for 20 kDa human growth GH abuse in sports: a double blind, Cole, L.A. (1997) Immunoassay of human hormone. Journal of Immunological placebo-controlled study. The GH-2000 chorionic gonadotropin, its free Methods 221(1–2), 77–85. Study Group. Journal of Clinical subunits, and metabolites. Clinical Hilderbrand, R.L., Wanninger, R. & Endocrinology and Metabolism 85(4), Chemistry 43(12), 2233–2243. Bowers, L.D. (2003) An update on 1505–1512. 46 chapter 4

Lottersberger, C., Hoermann, R., Mann, K., combination of direct high-performance human growth hormone (20 K hGH) Schwarz, S. & Berger, P. (2003) Tumor- anion-exchange chromatography and 2- measured by a specific enzyme-linked and pregnancy-derived isoforms of aminopyridine derivatization. Analytical immunosorbent assay. Study Group human chorionic gonadotropin: Biochemistry 206(2), 278–287. of 20 K hGH. Journal of Clinical biological and diagnostic relevance. Roger, M. & Lalhou, N. (1994) Endocrinology and Metabolism 84(1), Hormone Research 59(3), 125–134. Heterogeneity of plasma gonadotropins. 317–322. Nagy, A.M., Meuris, S. & Robyn, C. (1994) Consequences on immunological Wallace, J.D., Cuneo, R.C., Baxter, R. Origin and significance of the properties of LH. Nuclear Medicine and et al. (1999) Responses of the growth heterogeneity of protein hormones. Biology 21(3), 349–357. hormone (GH) and insulin-like Nuclear Medicine and Biology 21(3), Sasaki, H., Bothner, B., Dell, A. & Fukuda, growth factor axis to exercise, GH 317–330. M. (1987) Carbohydrate structure of administration, and GH withdrawal in O’Connor, J.F., Kovalevskaya, G. & Birken, erythropoietin expressed in Chinese trained adult males: a potential test for S. (1999) Interference of luteinizing hamster ovary cells by a human GH abuse in sport. Journal of Clinical hormone β-core fragment in urinary erythropoietin cDNA. Journal of Endocrinology and Metabolism 84(10), gonadotropin assays. Clinical Chemistry Biological Chemistry 262(25), 12 059– 3591–3601. 45(12), 2290–2292. 12 076. Wallace, J.D., Cuneo, R.C., Lundberg, P.A. Parisotto, R., Wu, M., Ashenden, M.J. et al. Saugy, M., Cardis, C., Schweizer, C., et al. (2000). Responses of markers of (2001) Detection of recombinant human Veuthey, J.L. & Rivier, L. (1996) bone and collagen turnover to exercise, erythropoietin abuse in athletes utilizing Detection of human growth hormone growth hormone (GH) administration, markers of altered erythropoiesis. doping in urine: out of competition tests and GH withdrawal in trained adult Haematologica 86(2), 128–137. are necessary. Journal of Chromatography. males. Journal of Clinical Endocrinology Parisotto, R., Ashenden, M.J., Gore, C.J. B, Biomedical Applications 687(1), and Metabolism 85(1), 124–133. et al. (2003) The effect of common 201–211. Wallace, J.D., Cuneo, R.C., Bidlingmaier, hematologic abnormalities on the ability Sharpe, K., Hopkins, W., Emslie, K.R. et al. M. (2001a) Changes in non-22- of blood models to detect erythropoietin (2002) Development of reference ranges kilodalton (kDa) isoforms of growth abuse by athletes. Haematologica 88(8), in elite athletes for markers of altered hormone (GH) after administration of 931–940. erythropoiesis. Haematologica 87(12), 22-kDa recombinant human GH in Park, A., Edwards, M., Donaldson, M., 1248–1257. trained adult males. Journal of Clinical Ghatei, M. & Meeran, K. (2003) Lesson Souillard, A., Audran, M., Bressolle, F. Endocrinology and Metabolism 86(4), of the week: interfering antibodies et al. (1996) Pharmacokinetics and 1731–1737. affecting immunoassays in woman with pharmacodynamics of recombinant Wallace, J.D., Cuneo, R.C., Bidlingmaier, pet rabbits. BMJ (Clinical Research Ed.) human erythropoietin in athletes. Blood M. (2001b) The response of molecular 326(7388), 541–542. sampling and doping control. British isoforms of growth hormone to acute Preissner, C.M., O’Kane, D.J., Singh, R.J., Journal of Clinical Pharmacology 42(3), exercise in trained adult males. Journal Morris, J.C. & Grebe, S.K.G. (2003) 355–364. of Clinical Endocrinology and Metabolism Phantoms in the assay tube: heterophile Stenman, U.H., Unkila-Kallio, L., 86(1), 200–206. antibody interferences in serum Korhonen, J. & Alfthan, H. (1997) Wide, L., Bengtsson, C., Berglund, B. & thyroglobulin assays. Journal of Clinical Immunoprocedures for detecting Ekblom, B. (1995) Detection in blood and Endocrinology and Metabolism 88(7), human chorionic gonadotropin: clinical urine of recombinant erythropoietin 3069–3074. aspects and doping control. Clinical administered to healthy men. Medicine Quarmby, V., Quan, C., Ling, V., Compton, Chemistry 43(7), 1293–1298. and Science in Sports and Exercise 27(11), P. & Canova-Davis, E. (1998) How much Strasburger, C.J., Wu, Z., Pflaum, C.D. 1569–1576. insulin-like growth factor I (IGF-I) & Dressendorfer, R.A. (1996) Wood, P. (2001) Growth hormone: its circulates? Impact of standardization on Immunofunctional assay of human measurement and the need for assay IGF-I assay accuracy. Journal of Clinical growth hormone (hGH) in serum: a harmonization. Annals of Clinical Endocrinology and Metabolism 83(4), possible consensus for quantitative Biochemistry 38(Pt. 5), 471–482. 1211–1216. hGH measurement. Journal of Clinical Wu, S.L., Amato, H., Biringer, R. et al. Ranke, M.B., Feldt-Rasmussen, U., Bang, P. Endocrinology and Metabolism 81(7), (2002) Targeted proteomics of low-level et al. (2001) How should insulin-like 2613–2620. proteins in human plasma by LC/MSn: growth factor I be measured? A Strasburger, C.J., Bidlingmaier, M., Wu, Z. using human growth hormone as a consensus statement. Hormone Research & Morrison, K.M. (2001) Normal values model system. Journal of Proteome 55 (suppl. 2), 106–109. of insulin-like growth factor I and their Research 1(5), 459–465. Rice, K.G., Takahashi, N., Namiki, Y. et al. clinical utility in adults. Hormone Wu, Z., Bidlingmaier, M., Dall, R. & (1992) Quantitative mapping of the Research 55 (suppl. 2), 100–105. Strasburger, C.J. (1999) Detection of N-linked sialyloligosaccharides of Tsushima, T., Katoh, Y., Miyachi, Y. et al. doping with human growth hormone. recombinant erythropoietin, (1999) Serum concentration of 20 K Lancet 353(9156), 895. Chapter 5

Analysis of Low Molecular Weight Substances in Doping Control

MARIO THEVIS AND WILHELM SCHÄNZER

Historical aspects of doping control doping became obvious in 1886, when the cyclist analysis of low molecular weight drugs Linton died during a race Paris–Bordeaux caused by an overdose of caffeine (Prokop 2002). In the fol- Ever since sport competition has been known, the lowing 70 years, numerous cases of doping-related wish to artificially enhance power, skill and strength deaths were documented, primarily involving was also present. Philostratos and Galen already cyclists, such as the death of a cyclist caused by reported attempts of athletes of the ancient Olympic amphetamine poisoning (1949); the hospitalization Games to improve performance and endurance by of a cyclist for confusional toxical state from excess- means of concoctions of mushrooms and plant seeds ive use of amphetamines (1956); a cyclist in state as well as special diets, for example with bovine test- of shock from excessive use of sympathomimetics icles (Burstin 1963; Prokop 1970, 2002). Moreover, (1958); and hospitalization of a cyclist for intoxica- the athletes’ animals were targets of manipulation tion with amphetamines and analeptics (1962). All as described for the ancient Romans, who fed their these cases were uncovered in Italy (Venerando horses hydromel, a mixture of honey and water, in 1963). With the formation of doping commissions in order to increase their strength (Morgan 1957; France (1959), Austria (1962) and Italy (1963) followed Ariëns 1965). Major reasons for the phenomenon of by the Medical Sub-Commission of the International athletes trying to artificially obtain the cutting edge Olympic Committee (IOC) in 1967, a comprehens- advantage in competition are supposed to be fame, ive fight against doping was started. First official glory and honor as well as financial aspects, which doping controls were conducted during the Olympic became even more important with the introduction Games in Grenoble (1968) based on a first screening of professional sports and horse and greyhound assay for a selection of stimulants (Beckett et al. racing. At the beginning of the 20th century, first 1967), and the list of prohibited substances and scientific procedures for the determination of methods of doping has been modified and extended doping agents in horse saliva were developed by many-fold during the last decades. A huge variety Russian and Austrian scientists, and in 1910–1911 of methods have been established in order to deter- about 220 samples were investigated regarding mine doping-relevant compounds, their metabol- administration of alkaloids. Between 1938 and 1954, ites and their influence on endogenous parameters. principal procedures for the detection of stimulants such as amphetamines were presented, which Classification of drugs demonstrated limited sensitivity and susceptibility to interferences by biological matrices (Richter Prohibited compounds 1938; Keller & Ellenbogen 1952; Axelrod 1954), but in 1956, a commonly accepted assay based on liquid Owing to the wide variety of banned compounds, extraction, paper chromatography and visualiza- there is no list mentioning all restricted substances tion was introduced (Vidic 1956). The fatality of by name. Besides examples for each category, the 47 48 chapter 5

Table 5.1 Classification of prohibited compounds according to the International Olympic Committee (IOC) and World Anti-Doping Agency (WADA).

I. Prohibited classes of substances A. Stimulants B. Narcotics C. Anabolic agents D. Diuretics E. Peptide hormones, mimetics and analogs Scheme 5.2 Chemical structure of morphine, a F. Agents with anti-estrogenic activity representative of the class of opioid-like narcotics. G. Masking agents II. Prohibited methods A. Enhancement of oxygen transfer B. Pharmacological, chemical and physical manipulation pounds such as amphetamines, ephedrines, caffeine C. Gene doping and also cocaine (Scheme 5.1). The latter was III. Classes of prohibited substances in certain sports already used by Mexican Incas in the 16th century A. Alcohol when they utilized coca leaves to cope with dis- B. Cannabinoids C. Local anesthetics tances of more than 1000 miles (1609 km) between D. Glucocorticosteroids Cuzco and Quito within 5 days (www.g-o.de 2003), E. Beta-blockers and the very first identified doping substances found in athletes’ urine samples in the 20th century were related to amphetamines. Due to the presence of ephedrines in a series of cold medicines, and fur- extension ‘. . . and related substances . . .’ is added, thermore the caffeine content of several cold and resulting in prohibition of all compounds with hot beverages, urinary concentration limits for these correlated pharmacological or physicochemical selected compounds were established, requiring properties. In Table 5.1, the classification of drugs quantitation of these drugs in urine specimens, according to the IOC and World Anti-Doping whereas new regulations valid from January 2004 Agency (WADA) (International Olympic Com- do not prohibit caffeine anymore. mittee 2003) is shown. For several remedies and substances such as beta-blockers, corticosteroids, narcotics local anesthetics, marihuana and alcohol, restriction is limited to selected sport sections or their medical The class of narcotics includes compounds referred indication. to as opioid-like analgesics such as morphine (Scheme 5.2) and related substances with a few stimulants exemptions, for example ethylmorphine and co- deine, owing to their weak analgesic effect. Other Stimulants are most likely the oldest doping sub- non-opioid-like drugs, for instance acetylsalicylic stances known to human beings. They include com- acid or diclofenac, are allowed.

Scheme 5.1 Structure formulae of amphetamine (1), ephedrine (2), cocaine (3) and caffeine (4). doping analysis with gc-ms and lc-ms/ms 49

19-desmethylation 1-methylation 18 OH 1,2-oxidation 17-methylation 12 17 19 11 18 16 14 15 1 9 2-methylation 10 2 8 Testosterone 3 5 7 4 6 O Fig. 5.1 Chemical structure of 7-methylation testosterone and modifications 4,5-reduction resulting in various synthetic anabolic steroids. 4-chlorination

anabolic agents tazolamide), phenoxyacetic acid derivatives (e.g. ethacrynic acid), and potassium-retaining diuretics While stimulants and narcotics are of interest prim- (e.g. spironolactone), some of which are shown in arily in competition, anabolic agents are effective Scheme 5.3 (Wilhelm & deStevens 1976; Lang & during periods of exercise. Increasing muscle growth Hropot 1995; Möhrke & Ullrich 1995). and strength gaining, the drugs belonging to this category were added to the list of prohibited com- beta-blockers pounds of the IOC in 1976, and, since 1993, not only are steroids related to testosterone (Fig. 5.1) included Beta-receptor blocking agents (also referred to as in the class of anabolic agents but also drugs such as beta-blockers) are banned for sports such as shoot- β clenbuterol, as 2-agonists cause anabolic effects if ing or ski jumping owing to their sedative effects. In administered in doses significantly higher than for general, beta-blockers consist either of a phenyl ring therapeutic use (Reeds et al. 1988; Wagner 1989; structure bearing an oxypropanolamine side chain Stallion et al. 1991; Höher & Troidl 1995). that terminates in an isopropyl or tert. Butyl residue (e.g. propranolol, levobunolol), or a phenyleth- diuretics anolamine nucleus comprising substitutes such as nitrite functions (e.g. nifenalol). Modern β-receptor Diuretics are considered as doping agents based on blocking agents partially diverge from this principal two facts: (i) athletes competing in sports categor- structure, for instance the remedy nebivolol. Since ized by weight can decrease their body weight 1988 the class of beta-blockers belongs to the IOC list by artificially induced diuresis, i.e. increased urine of prohibited compounds, and some examples are flow; (ii) the increased loss of water dilutes the urine presented in Scheme 5.4. resulting in decreased concentration of excreted compounds. Regarding those drugs that are banned As shown in Table 5.1, additional classes of prohib- with respect to cutoff limits, diuretics can mask ited compounds, such as peptide hormones, belong the abuse of prohibited substances. The class of to the list of banned substances, but in this overview diuretics in particular demonstrates the possible we will focus only on a selection of low molecular physicochemical heterogeneity of one category of weight drugs. drugs. Diuretic agents can be divided into at least six groups owing to their structures, place and Analytical techniques mechanism of effect. There are so-called thiazides (e.g. hydrochlorothiazide), sulfamoylbenzoic acid The first commonly accepted screening and derivatives (e.g. furosemide), osmotic diuretics (e.g. confirmation methods for doping control analysis mannitol), carbonic anhydrase inhibitors (e.g. ace- included sample preparations based on liquid–liquid 50 chapter 5

Scheme 5.3 Structure formulae of selected diuretics: (1) hydrochlorothiazide, (2) furosemide, (3) mannitol, (4) acetazolamide, (5) ethacrynic acid and (6) spironolactone.

Scheme 5.4 Structure formulae of selected beta-blockers: (1) propranolol, (2) levobunolol, (3) nifenalol and (4) nebivolol.

extraction of urine, concentration of the extracts instruments and sophisticated sample preparation and separation of analytes by means of gas–liquid procedures, numerous assays have been developed chromatography (GLC) and thin-layer chromato- enabling a comprehensive analysis of elite athlete’s graphy (TLC). A flame ionization detector (FID) was urine samples for doping control purposes. Depend- interfaced to GLC indicating the presence of stimu- ing on the nature of compounds to be analyzed, lants of the amphetamine type by signals at com- different analytical tools have been employed, the parable retention times as observed with reference principal techniques of some of which will be compounds. Thin-layer chromatography was used described in the following. for the identification of strychnine and also ring- In general, chromatographic separation of extracts hydroxylated stimulants such as p-hydroxyam- obtained from biological material such as blood, phetamine and phenylephrine (Beckett et al. 1967). urine or hair, is more or less obligatorily preceding In case of ‘positive’ test results, additional informa- the analytical detection. Thus, more detailed infor- tion on the analytes was obtained by derivatization, mation on analytes is provided; for example, by micro infrared spectroscopy and also mass spectro- retention times that support for instance the distinc- metry (MS). With improvements and innovations tion between stereoisomers, or by separating lowly in analytical techniques, commercially available concentrated from highly abundant compounds. doping analysis with gc-ms and lc-ms/ms 51

Protective coating Fused silica tubing performed, for instance by trimethylsilylation, and the derivatized fused silica creates a suitable surface for being coated by the stationary phase. Stationary phase Stationary phase. The stationary phase is the most crucial component in terms of retention, separa- tion and peak shape of analytes in GC. This par- ticular part of capillary columns consists of either polysiloxanes/silicones, polyethylene glycols (PEGs) or so-called porous layer phases. The predomin- antly used stationary phases are the substituted Fig. 5.2 Principal composition of capillary gas polysiloxanes owing to their robustness and super- chromatography columns. ior lifetime. In Fig. 5.3, the general chemical struc- ture of polysiloxane-based stationary phases is presented, determined by the alternating silicon- oxygen backbone and two substitutes at each silicon Modern chromatographic systems are based either atom. The structure and amount of the substitutes on capillary gas chromatography or high perform- characterizes each stationary phase, and primarily ance liquid chromatography, both of which have four groups are utilized, mixed in different ratios: become extremely sophisticated and elaborated (i) methyl-; (ii) phenyl-; (iii) cyanopropyl-; and (iv) sciences. trifluoropropyl residues. With the choice of sub- stitutes and their relative presence in the stationary phase, the polarity of the GC column is defined as Gas chromatography listed in Fig. 5.4 for 14 common mixtures. Another Nowadays, capillary columns are the first choice stationary phase employed in GC is PEG that is for gas chromatography (GC) in doping control characterized mainly by the molecular weight or the analysis. Their excellent performance in terms of chain length of the polymer. A remarkable differ- compound separation, robustness, and the enorm- ence of this type of material to polysiloxane phases ous variety of different stationary phases provides is the possibility to designate a certain pH range, i.e. numerous options for method development and acidic or alkaline columns are available demon- possibilities to cope with tasks such as compre- strating an improved capability to separate acidic hensive screening assays or compound-specific or alkaline compounds, respectively. A main disad- analyses. The principal composition of a modern vantage is the high sensitivity to oxygen, in particu- capillary GC column is presented as cross section in lar at elevated temperatures, resulting in destruction Fig. 5.2, and the basic scaffold is represented by of the stationary phase and a correlated short life- three distinct parts: (i) outer protective coating; (ii) time. Porous layer stationary phases consist of small fused silica tubing; and (iii) stationary phase. porous particles (e.g. polymers, aluminium oxide, molecular sieve) that are bound to the fused silica Fused silica tubing. Fused silica is a synthetic, quartz- tubing by chemical linkers. Owing to their high like glass of very high purity regarding metal oxide affinity of gas adsorption, these phases are utilized contaminants. In general, fused silica bears a very primarily for the chromatography of very volatile active surface owing to its silanol functions, which compounds or gases, which usually demonstrate can interact with polar groups of analytes, such as poor retention on stationary phases such as poly- hydroxyl, carboxyl and thiol residues as well as prim- siloxane or PEG, and thus require temperatures ary and secondary amines, causing peak tailing and below 35°C. With so-called PLOT columns (porous a decreased peak intensity. Hence, a deactivation of layer open tubular), gas–solid adsorption is the silanol groups by appropriate chemical treatment is primary separation mechanism enabling efficient 52 chapter 5

CH3 CH3 CH3

Si OSiO Si O Si O

CH CH CH 3 3 n nm3

Methyl Methyl-phenyl

CH3 CH3 CH3 CH3

Si O Si O Si OSiOFig. 5.3 General chemical structure of polysiloxane-based stationary CH CH (CH2)3CN nm3 (CH2)2CF3 nm3 phases with selected substitutes (methyl-, phenyl-, cyanopropyl- Methyl-cyanopropyl Methyl-trifluoropropyl and trifluoropropyl residues).

resolution, film thickness primarily influences the

Increasing polarity 100% methyl 5% phenyl capacity of columns and the retention of analytes. 20% phenyl The length of a column is directly proportional to 35% phenyl 50% phenyl the number of theoretical plates, and as the resolu- 40–50% trifluoropropyl polyethylene glycol tion is proportional to the square root of the number 50% cyanopropyl of theoretical plates, resolution is also proportional 80% cyanopropyl to the square root of column length. In other words, 100% cyanopropyl doubling the column length (and thus the number Fig. 5.4 Stationary phase polarities depending on the of theoretical plates) does not improve resolution by choice of substitutes and their relative amounts. 100%, but in practise only by about 25–35%, and the reduction of column length by 50% causes a loss of resolution of approximately 15–25%. In addition, retention of very volatile substances, but the strong the number of theoretical plates is inversely propor- interaction between stationary phase and analyte tional to the column diameter. Hence, decreasing excludes their use for less volatile substrates. column diameter increases efficiency of GC. Taking again into account that resolution is proportional to Outer coating. As fused silica tubing is fairly fra- the square root of the number of theoretical plates, gile, it requires a protective outer coating. For that halving of the inner diameter also improves resolu- purpose, polyimide is commonly employed and tion by 25–35%; for instance, by exchanging a col- provides two advantages: first, polyimide-coated umn with an inner diameter of 0.25 mm by 0.11 mm. columns are robust, stable and do not require delic- The capacity of capillary columns is directly de- ate handling, and second, it covers and fills flaws pending on the film thickness of the stationary of the fused silica tubing, preventing the growth of phase. While phases of 0.1–0.2 µm are appropriate defects. for 20–50 ng of analyte, 0.5 µm are necessary for After considering the composition of capillary amounts greater than 125 ng. Thus, film thickness GC column, two additional parameters impact the must be adjusted to the expected load of com- chromatographic performance remarkably: column pounds injected onto the GC column. dimensions, i.e. length, diameter and film thickness, Also, carrier gas impacts the performance of GC, and carrier gases. as commonly described by the Van Deemter curve. While length and diameter mainly affect peak Frequently employed carrier gases are helium and doping analysis with gc-ms and lc-ms/ms 53

hydrogen, the latter one provides a superior For characterization of column performance, the optimal linear velocity resulting in an improved number of plates n (or the plate height h) is fre- performance regarding resolution. The disadvant- quently used as a measure. The Van Deemter equa- age of hydrogen is definitely its highly flammable tion in its simplest form describes the reciprocal character. relation of particle size to plate height, which means that a decrease in particle size entails an improve- ment in chromatography performance, namely re- Liquid chromatography solution (Yamashita & Fenn 1984a; Engelhardt et al. Comparable to GC, liquid chromatographic access- 1985). Hence, there is a bias to reduce particle sizes ories provide a wide variety of dimensions, pack- with the use of very short columns in high speed ing materials and, as compounds are carried chromatography in order to maintain required peak through the analytical columns by liquids, a series resolutions. of organic solvents and buffers. In addition, several The most commonly employed carrier material principles of liquid chromatography (LC) (normal for the preparation of stationary phases in RPC is phase, reversed phase, ion exchange, size exclusion) silica. Commercially available silicas differ in phys- can be applied, depending on the nature of target ical properties such as specific surface area, average analytes. In this brief overview, we will focus only pore diameter, specific pore volume and shape. on reversed phase chromatography (RPC) which is Assuming open cylindrical pores, the three first the most frequently used system in doping control mentioned variables are related by the equation analysis. φ= 3 × 10 4Vp/Osp φ= = Cartridges. In LC in particular, numerous diame- with average pore diameter (nm), Vp specific –1 = ters of column cartridges are available. With the pore volume (mL·g ) and Osp specific surface improvements in packing materials (i.e. stationary area (m2·g–1) (Scott 1982). Average surface areas are phases, see below), the most commonly employed approximately 300 m2·g–1 and pore diameters 10 nm, dimensions for small molecule analysis in doping giving rise to an approximate specific pore volume control laboratories have been reduced to column of 1 mL·g–1. Nowadays, almost exclusively spher- lengths of 30–120 mm and inner diameters of 1.0– ical silica is utilized as carrier for reversed phase LC 4.6 mm compared to earlier days of high perform- stationary phases, owing to the higher density com- ance liquid chromatography (HPLC) where longer pared to silica prepared by polymerization of silicic and broader columns were utilized. The shortening acid yielding irregular particle shapes. On the sur- has become an option with the highly selective and face, the silica carrier bears hydroxyl groups (silanol specific mass analyzers, which enable separation of functions), which are utilized to chemically bind compounds by means of their mass spectra, and phases and thus define the principal nature of the thus the need for optimal chromatographic distinc- HPLC column. An enormous variety of phases has tion between analytes has become minor important. been introduced to the market in order to provide In addition, performance of LC columns has been suitable material for each chromatographic chal- incredibly ameliorated, facilitating peak separation lenge, for instance the frequently used phenyl-, C18-, in reduced time frames. C8- and C4-phases, or more polar material such as cyano-, diol- or amino-phases as demonstrated in Stationary phase. As mentioned in the Gas chro- Fig. 5.5. In addition, even more sophisticated sys- matography section above, the stationary phase is tems enabling the separation of chiral molecules the utmost crucial point in LC as well. Here, differ- have been developed, and the most recent improve- ent parameters have to be considered such as par- ment in high-throughput analysis was achieved by ticle size (3–50 µm), carrier (spherical or irregular so-called monolithic columns consisting of a single silica), chemically bonded phases (e.g. C4, C8, C18), piece of an organic polymer or silica with flow- and pore size (50–4000 Å). through pores prepared by direct polymerization of 54 chapter 5

sulfate additives are not recommended but, for instance, ammonium acetate, ammonium formate or tetraethy-lammonium hydroxide (TEAH) are. Moreover, ESI is not compatible with ingredients favoring the generation of strong ion pairs resulting in neutralization of ions after desorption. Control of pH is of paramount importance in particular with ESI, because of its enhancing effect on analyte ion- ization. Compounds of basic character should be analyzed under acidic conditions utilizing additives such as acetic acid or formic acid in the 0.1– 1.0% range, while components of acidic nature (e.g. carboxylic acids) are preferably negatively ionized under alkaline conditions; for example, by means of addition of ammonium hydroxide. Commonly employed organic solvents are comparable to con- ventional HPLC without interfacing to MS, i.e. methanol, ethanol and acetonitrile.

Detectors

The detection and identification of chromatographic- Fig. 5.5 Commonly used stationary phases in reversed ally separated compounds is of paramount import- phase chromatography (RPC): (1) Phenyl-, (2) C -, (3) C -, 4 8 ance in doping control analysis. Many different (4) C -, (5) cyano- and (6) diol-phases. 18 types of detectors have been developed such as FID, nitrogen phosphorus detector (NPD), UV-detector appropriate monomers. These materials provide and mass selective detectors such as quadru- better stability and, more importantly, higher per- pole, time-of-flight, ion trap, Fourier-Transform ion formance than conventional columns packed with cyclotron resonance (FT-ICR), triple-stage quad- particles. rupole and sector mass analyzers. In addition, hybrid instruments composed by two or more of these Mobile phases. In case of RPC, the stationary phase units are commercially available, and numerous is the non-polar and the mobile phase the polar applications enabling the determination of adminis- component. Conventional RPC employs buffer sys- tration of prohibited compounds are based on a tems consisting of, for example, KH2PO4/H3PO4 combination of analyte-specific sample preparation, or NaH2PO4 in combination with organic solvents chromatography and sensitive as well as selective such as methanol or acetonitrile, which are very detectors. suitable for analyzers such as ultraviolet (UV) -detectors. But the coupling of HPLC to MS requires flame ionization detector different considerations regarding the eluents as there are restrictions on pH, solvent choice, solvent One of the oldest and universal detectors inter- additives and flow rates for LC in order to accom- faced to GC is the flame ionization detector (FID) plish optimal mass spectrometric results (Wheeler (Fig. 5.6). Here, the gas mixture eluting from the GC 1955). In general, atmospheric pressure ionization column is aerated, fortified with hydrogen and (e.g. electrospray ionization, ESI) demands volatile ignited electrically. The hydrogen–air flame itself solvent additives to prevent ion source contamina- creates only very few ions, but most organic com- tions; hence commonly used phosphate, borate or pounds generate an increased number of ions and doping analysis with gc-ms and lc-ms/ms 55

Vent the most popular analyzers for high performance LC since the late 1960s. The principle is based on Collector conventional spectrophotometry and the Beer– Lambert law: Hydrogen–air flame Air = ε Log I0/I cd = = with I0 intensity of the incident light, I intensity of the transmitted light, ε = extinction coefficient, = H2 c concentration of the absorbing analyte and d = path length of the cell. In practice, the eluent of a HPLC analysis is passed through a cell of fixed GC column length, a light source emitting in a distinct UV range (e.g. deuterium lamp, 190–600 nm) and, providing light of defined intensity is focused on the cell through a monochromator, the intensity of the Fig. 5.6 Principal design of flame ionization detectors transmitted light of selected wave length is detected (FIDs). by means of photo diodes. The detectors described so far enable the recogni- tion of compounds chromatographically separated electrons upon pyrolysis in the hydrogen–air flame, either by GC or HPLC. The main drawback of these and thus enable an improved conduction of electric- analyzers is the low specificity and correlated lack ity. To a so-called collector located near the flame, a of detailed information on the analytes. As a con- polarizing voltage is applied, which attracts gen- sequence, instruments based on MS have become erated ions producing a current that is proportional the first choice in doping control analysis being to the amount of sample being burned after elution employed in combination with both GC and HPLC, from the GC column. This particular current is complementary to the ‘traditional’ instrumentation sensed by an electrometer, converted to a digital sig- utilizing FID, NPD and UV detectors. In order to nal, and plotted as peak in a chromatogram. identify substances by MS, ionization is required, which can be accomplished by many different tech- nitrogen phosphorus detector niques in consideration of the respective chroma- tographic system. Nowadays, GC-MS frequently The nitrogen phosphorus detector (NPD) is a vari- employs techniques such as electron ionization and ant of the FID. The main difference is a glass chemical ionization, while HPLC is mainly inter- or ceramic bead that is located right above the jet of faced to mass spectrometers by ESI, atmospheric the hydrogen–air plasma. A low hydrogen/air ratio pressure chemical ionization (APCI) or also atmo- can not sustain a flame, resulting in decreased spheric pressure photo ionization (APPI). In the hydrocarbon ionization. But the alkali ions on the following, we briefly describe the principles of elec- bead surface facilitate the ionization of nitrogen- or tron ionization and ESI as well as quadrupole and phosphorus-containing compounds, thus favoring ion trap analyzers, the most commonly used tools in the detection of those compounds while suppress- doping control laboratories. ing the abundance of other, mainly hydrocarbon- based substances. Ionization techniques ultraviolet absorbance detector electron ionization

While FID and NPD can only be interfaced to GC, Electron ionization (EI) is widely used to generate the UV absorbance detector has proven to be one of positive ions of analytes separated by GC. Here, 56 chapter 5

Filament radical cations are generated inclining to dissociate into product ions, which are detected in the mass selective analyzer giving rise to a characteristic mass spectrum of the respective substance. These – e product ions can be radicals as well as even-electron ions, depending on rearrangement and elimination processes.

electrospray ionization To mass selective The development of the comparably soft ioniza- detector Anode tion technique utilizing electrospray was recently awarded with the Nobel Prize for chemistry in 2002. GC column John Fenn was honored for fundamental research regarding the ionization of macromolecules that was already introduced by Yamashita and Fenn in Fig. 5.7 Schematic assembly of an ion source with electron 1984 (Yamashita & Fenn 1984b; Voyksner 1997). ionization. With this technique, liquids containing protonated or deprotonated molecules are sprayed by means of a capillary tip at high voltages (1 kV and higher), electrons emitted from a cathode, the so-called forming charged droplets that shrink by solvent filament, are accelerated to an anode, and collide by evaporation and repeated droplet disintegration orthogonal orientation within the ion source with leading to very small and highly charged droplets. compounds eluting from the GC column as demon- Finally, gas-phase ions are produced from these strated in Fig. 5.7. The collision of electrons released very small droplets as discussed in two different with commonly utilized 70 eV (potential between theories, namely charged residue model (Dole et al. filament and anode is set to 70 V) with analytes in an 1968) and ion evaporation (Iribarne & Thomson 1976). ion source gives rise to a number of processes; for In Fig. 5.8, the main processes of ESI are presented. example: First, the penetration of imposed electric field into the liquid of the capillary leads to an enrichment of (1) AB + e– → AB+ + 2e– positive charges at the surface of the liquid, causing (2) AB + e– → AB2+ + 3e– destabilization of the meniscus, formation of a cone and a so-called jet-emission of droplets bearing an (3) AB + e– → AB– excess of positive ions. The charged droplets shrink For positive ionization, the most important mechan- by solvent evaporation while the charge remains ism is (1) with the generation of a positively charged constant. Hence, an increase of the electrostatic molecule (AB+) by liberation of an electron upon repulsion occurs, resulting in fission of the droplets impact of the accelerated electron released from the as they reach the Rayleigh stability limit. This phe- filament. In addition, also two electrons can be nomenon continues with ongoing evaporation of removed from the analyte AB giving rise to a doubly solvent until very small and highly charged droplets charged molecule AB2+ (2). Moreover, electrons can are created. Finally, as proposed by Dole et al. (1968), be incorporated into the molecule AB resulting in a only one ion remains in a singly charged droplet, negatively charged analyte AB– (3). Besides these and the evaporation of solvent gives rise to a examples of effects of EI on compounds introduced gas-phase ion (charged residue model). Alternatively, into an ion source by GC, the consequences of this the direct ion emission from droplets with a radius ionization technique also have to be taken into con- smaller than 10 nm was postulated, also generating sideration. As a major result of EI, highly energetic gas-phase ions (ion evaporation) (Kebarle & Ho 1997). doping analysis with gc-ms and lc-ms/ms 57

Capillary with solvents and analytes Connected segment

Taylor-cone Charged droplets U+Vcos2πft

Ions

Orifice plate Single rod

High-voltage power supply -U-Vcos2πft

Fig. 5.8 Processes in positive electrospray ionization. Fig. 5.9 General assembly of a quadrupole mass selective Application of high voltage to a liquid-filled capillary analyzer. Two opposing metal or gold-coated fused silica leads to enrichment of positive ions at the liquid surface, rods are electrically connected while adjacent rods are destabilization of the meniscus and generation of a cone isolated. Ions are accelerated from the ion source into the that emits droplets containing an excess of positive center of the quadrupole analyzer and their trajectories charges. are influenced by application of ac and dc voltage. Only ions of defined mass/charge ratios pass through the quadrupole at distinct voltages, which enables mass filtering. Mass analyzers

The ions generated from substances by various ion- ization techniques are analyzed by different mass positive or negative fields are established towards selective detectors such as quadrupole or ion trap the centerline of the quadrupole and, thus, positive instruments. Both analyzers operate under high ions passing through the rods are pushed away with vacuum and can be interfaced to GC (and thus EI) as positive and attracted with negative polarization. well as LC (including ESI) by respective coupling The extent of ion deflection is directly depending on systems. As EI is performed under high vacuum, the applied voltage, its frequency (i.e. the duration no special interface is necessary in order to control of exposure to alternating fields) and the mass different pressure levels, but ESI operates at atmo- of ions. In addition, to one segment positive direct spheric pressure. Hence sophisticated vacuum current (dc) is applied while the other segment is barriers are present in modern LC-ESI-MS(/MS) provided negative dc. As a result, ions of a distinct systems maintaining the required reduced pressure mass to charge ratio can travel through the two- in mass selective detectors. dimensional quadrupole field if the applied direct current and alternating current with a defined fre- quadrupole analyzers quency are appropriate for stable oscillating moving. These parameters can be modified for an optimized A quadrupole mass spectrometer is composed by mass selection, enabling the isolation of a single ion four rods consisting of various materials; for exam- or the recording of a full spectrum by scanning a dis- ple, fused silica coated with a gold layer. Opposing tinct range of mass/charge ratios over a given time rods are connected while adjacent segments are period (Budzikiewicz 1998; Lottspeich & Zorbas electrically isolated, as demonstrated in Fig. 5.9. 1998). Initially, ions that are generated in the ion source are Ions isolated by quadrupole instruments can accelerated into the center of the quadrupole. While be subjected to further experiments as frequ- applying alternating current (ac) to the segments, ently employed in triple stage quadrupole mass 58 chapter 5

End-cap electrode an ion inlet and one with an ion exit aperture) and a (ion entrance) Dipolar rf ring electrode, in the center of which ions are stored Ring by application of appropriate rf field established electrode between the ring electrode and the two end-cap electrodes. Ions that are generated by an external ion source (e.g. EI or ESI) are transferred into the ion He He trap, which contains a gas (e.g. helium, argon) at Ions approximately 1 mtorr. A commonly employed gas Aperture is helium that damps the trajectories of ions to the He He center of the ion trap, and moreover it reduces the kinetic energy of ions through collisions favoring Quadrupolar rf the storage of the ions within the trap. The storage of ions of a broad range of mass/charge ratios Ground End-cap electrode is directly depending on the ac voltage applied to (ion exit) the ring electrode. Hence, the consecutive ejection of ions enabling their mass selective detection is Fig. 5.10 Principal design of an ion trap mass accomplished by the so-called instability mode, spectrometer. The ring electrode is edged by two end- which is based on the successive increase of the ac cap electrodes establishing an rf field upon voltage application that enables the storage of ions in the center voltage applied to the ring electrode in combination of the ion trap. with an ac voltage applied to the end-cap electrodes causing resonant motion. Ions of defined mass/ charge ratios develop instable trajectories, are ejected spectrometers. Here, three quadrupoles are present, through the perforations of the end-cap electrode the first one of which is used to pick a single ion (ion exit) and detected with an electron multiplier. from the ion beam introduced into the mass spec- In addition to an efficient scan operation mode, trometer. In the second quadrupole, the so-called ion trap mass spectrometers offer possibilities of collision cell, the isolated ion hits molecules of a MSn experiments. With the selective removal of ions collision gas (e.g. argon, nitrogen) and dissociates from the trap, storage of a precursor ion of interest with respect to the applied accelerating voltage and and its resonant activation, collisionally activated dis- collision gas pressure. The resulting product ions sociation (CAD) is obtained, giving rise to product are finally measured in a third quadrupole, either ion spectra with a comparable amount of informa- by full scan or selected ion monitoring, providing tion as obtained with triple stage quadrupole MS/ information on fragmentation processes as well as MS experiments. But an important advantage of the structure of the analytes. ion trap technique is that trapping of ions allows the selection, storage and subsequent dissociation of quadrupole ion trap analyzers a product ion generated by MS/MS experiments, namely MS3, providing information on the fragmen- A frequently employed alternative to quadrupole tation pathway and composition of product ions, mass spectrometers are (quadrupole) ion trap which is of great interest, in particular in ESI-MS. analyzers. Already in 1953, Paul and co-workers described the applicability of a three-dimensional Sample preparation and purification quadrupole to ‘trap’ ions, and in the last decades, strategies sophisticated ion trap mass spectrometers were developed based on this invention (Louris et al. In doping controls, mainly urine specimens are 1987, 1989; Patterson et al. 2002). In Fig. 5.10, the obtained from athletes in order to be analyzed principal composition of an ion trap instrument is regarding prohibited compounds according to estab- shown, including two end-cap electrodes (one with lished anti-doping rules. In addition, also samples doping analysis with gc-ms and lc-ms/ms 59

of whole blood, plasma/serum and hair are obtained ives, organic solvents and pH values) (Geyer et al. for certain purposes of drug identification. In order 1997; Thevis et al. 2001; Deventer et al. 2002). to accomplish utmost specificity and required sensit- A very powerful tool turned out to be the so- ivity in analyses, sample pretreatment has become a called re-extraction, which was already employed very important part of doping controls, and various in 1952 (Keller & Ellenbogen 1952) and applied to strategies are present to purify specimens from salts doping control analysis in order to enhance purity and interfering compounds, and to isolate the target of samples to be analyzed. In particular, the extrac- analyte(s) in particular. tion of analytes under alkaline conditions into an organic solvent (e.g. diethyl ether) and subsequent retransfer of components of interest from the Liquid–liquid extraction organic into an acidified aqueous layer (e.g. 0.06 One of the oldest and most frequently utilized N HCl) proved to be very useful for confirmation β sample preparation steps is liquid–liquid extraction measurements, for instance for the 2-agonist clen- (LLE). The very first publications on the identifica- buterol, by highly efficient reduction of biological tion of benzedrine (a racemic mixture of amphet- background (Sigmund et al. 1998). amine sulfate) and metabolites in human urine employed LLE in order to determine excretion rates Solid-phase extraction (Richter 1938). Here, urine samples were alkalized by means of 2 N sodium hydroxide and extracted For a series of compounds, an alternative approach with petroleum ether. In the following decades, to purify biological samples and concentrate ana- numerous applications using LLE for the purifica- lytes is solid-phase extraction (SPE). Based on tion of biological fluids were presented, all of which various adsorbing materials such as polystyrene, were based on the same principle of adjusting pH C18 or ion exchange phases, several screening and and extraction of acidic, neutral or alkaline com- confirmation procedures in doping control analysis pounds (Keller & Ellenbogen 1952; Axelrod 1954; have been established. In 1968, Bradlow demon- Beckett & Rowland 1965; Beckett & Wilkinson 1965; strated the possibility to extract steroid conjugates Kolb & Patt 1965; Cartoni & Cavalli 1968). The with neutral resins (Bradlow 1968), and further knowledge about pI of analytes provides the option improvements in material and methods have pro- of multiple extraction of specimens and/or re- vided tools for the development of procedures for extraction of compounds at various pH values into doping control purposes, for example, for anabolic different solvents. With the desire for more detailed, steroids (Donike et al. 1984) and diuretics (Thieme specific and sensitive analytical assays, LLE pro- et al. 2001). A major advantage of SPE is the feas- cedures were improved and extended in order to ibility of full automation of sample extraction, in- remove coextracted interfering substances, and cluding conditioning of cartridges, sample loading, these principles have been utilized for up-to-date washing and subsequent elution. Moreover, derivat- screening and confirmation methods in doping ization of analytes can be accomplished during SPE control analysis (Donike et al. 1970; Spyridaki et al. (Lisi et al. 1991). 2001; Van Eenoo et al. 2001; Thevis et al. 2003b). For instance, stimulants such as amphetamines, Immunoaffinity chromatography ephedrines and related derivatives are commonly extracted from urinary matrices into ethers under A more recently introduced way of analyte isolation alkaline conditions, and the organic layer is con- from biological matrices is immunoaffinity chro- centrated and subsequently analyzed by GC-MS matography (IAC). In general, mono- or polyclonal and/or GC-NPD. Moreover, LLE is employed for antibodies able to bind non-covalently to distinct the analysis of corticosteroids (Fluri et al. 2001), parts of target molecules are linked to a carrier such diuretics, β-receptor blocking agents and anabolic as spherical agarose particles. The resulting ‘gel’ is steroids under various conditions (i.e. salt addit- placed in a column bearing a frit, urine or plasma is 60 chapter 5

Cyanate ester highly reactive

Agarose

Cyclic imidocarbonate slightly reactive

Isourea derivative Substituted imidocarbonate

Mixture of IAC column Non-covalently bond compounds antigen (target compound) Covalently bond Target analyte Sample antibody

Glass tube Gel Elution

Frit Target compound

Agarose-bead

Scheme 5.5 Activation of agarose and subsequent immobilization of antibodies. IAC, immunoaffinity chromatography. loaded onto the column and passed through the gel. fering signals. The principal production of an IAC The carrier-linked antibodies recognise target ana- column and its use is shown in Scheme 5.5. Com- lytes as they flow through the column and capture monly employed carriers are commercially avail- these compounds by generating non-covalent com- able agarose-beads. The surface of these beads is plexes, while other, non-antigenic substances miss- activated in order to chemically couple peptides ing the antibody–antigen reaction, are eluted. Under or proteins, which is accomplished by various various conditions, the non-covalent complex can techniques. One of the most frequently employed be broken without harming the antibody or analyte methods is cyanogen bromide (CNBr) activation of interest and the purified antigen is analyzed, for that was introduced in 1967 by Axen et al. (1967). example by GC-MS, basically without any inter- Under alkaline conditions, CNBr introduces cyanate doping analysis with gc-ms and lc-ms/ms 61

esters and imidocarbonates into the matrix of actions causing decreased volatility. One of the first agarose by reaction with its surface hydroxyl func- modifications of analytes was accomplished by tions to which antibodies are chemically connected acetylation utilizing acetic anhydride. The resulting as isourea or substituted imidocarbonate derivat- molecules have to be purified from remainders of ives (Hermanson et al. 1992). The beads bearing acetic acid and acetic anhydride, commonly achieved covalently fixed antibodies are placed in a column, by LLE. Because of this laborious and time-consum- which recognise and retain only specific analytes or ing way of preparing compounds for GC-MS ana- defined classes of compounds present in biological lysis, several derivatization agents were developed, fluid. With appropriate eluents, for example mix- preferably including trimethylsilyl (TMS) residues. tures of organic solvents and water, the captured There are for instance hexamethyldisilazane (HMDS), analytes are released from the antibodies, concent- trimethylsilyl-imidazole (TMSImi) and N-methyl- rated and analyzed by conventional techniques N-trimethylsilyltrifluoroacetamide (MSTFA) (Donike such as MS (Schänzer et al. 1996; Machnik et al. 1969), the latter one of which is nowadays the most 1999). The reduction of biological background frequently employed derivatization reagent, in par- results in improved signal/noise ratios, and thus in ticular in combination with ammonium iodide better detection limits of GC-MS and LC-MS/MS and ethanethiol, which gives rise to in situ gener- instruments. ated trimethyliodosilane (TMIS), a highly reactive trimethylsilylation reagent (Donike 1973; Donike & Zimmermann 1980). With TMIS, hydroxyl and Derivatization techniques amino groups are modified and also keto functions GC-MS is one of the primary tools for screening and are converted to their enol-TMS ethers, as demon- confirmational analyzes in doping control laborator- strated in Scheme 5.6. Several other powerful ies. These systems provide specificity and sensitiv- reagents have been used that introduce trifluo- ity for numerous compounds that are prohibited roacetyl (TFA)- or heptafluorobutyl (HFB) -residues according to the list of banned substances of IOC into molecules, for example N,N-bistrifluoroaceta- and WADA. But a principle requirement to identify mide, N-methyl-N-bistrifluoroacetamide (MBTFA) drugs by means of GC-MS is volatility. As many (Donike & Derenbach 1976) or N-methyl-N-bishep- β analytes are heavy volatile, for example 2-agonists, tafluoroburtyramide (MBHFB), respectively. The various diuretics and β-receptor blocking agents, trifluoroacetylation results in more stable derivat- they are derivatized to more volatile analogs by ives of amino functions than trimethylsilylation, means of different reagents. Here, hydroxyl and hence selective modification of analytes with TMS primary or secondary amino functions are target and TFA groups has also become possible, enabling groups because of their ability of hydrophilic inter- the gas chromatographic separation of stereoisomers

Scheme 5.6 Derivatization of doping-relevant analytes by means of trimethyliodosilane (TMIS). 62 chapter 5

of stimulants such as ephedrines (Opfermann & is methyltestosterone, a derivative of testosterone Schänzer 1996; Thevis et al. 2003b). obtained by introduction of an additional methyl group at C-17. Most anabolic steroids are subjects Qualitative analysis of doping-relevant of extensive metabolism, giving rise to a series of compounds metabolites, for instance by reduction of keto func- tions, oxidation of hydroxyl groups, introduction For most of the prohibited compounds in elite of hydroxyl residues and oxidation/reduction sports, qualitative evidence of their presence in of carbon-carbon bonds of the steroid nucleus urine samples of athletes is sufficient for a positive (Schänzer 1996). Following this phase-I metabol- test result. As many assays for screening and con- ism, conjugation to glucuronides and/or sulfates is firmation purposes are based on chromatography observed as common phase-II modification prior interfaced to MS, guidelines for the identification to renal excretion. In Scheme 5.7, the principal of substances by means of GC-MS or LC-MS(/MS) metabolism of methyltestosterone is demonstrated, systems have been established. The identification of generating 17α-methyl-5α-androstan-3α,17β-diol a compound is accomplished if the relative abund- (I), 17α-methyl-5β-androstan-3α,17β-diol (II) and ances of a specified amount of characteristic ions corresponding glucuronic acid conjugates. (depending on MS techniques) agrees, with respect Commonly employed strategies to identify meta- to allowed variations, with those obtained by ana- bolites of anabolic steroids are based on the lysis of the corresponding reference material. In enzymatic hydrolysis of phase-II metabolites to the addition, retention times of signals observed in corresponding phase-I metabolites, purification, con- urine samples of athletes and in appropriately centration, derivatization and subsequent GC-MS fortified reference urine specimens are compared analysis. For most anabolic steroid metabolites, no and may shift only within a defined time frame. endogenous production in humans is possible, Hence, knowledge about chromatographic and, in except for nandrolone, which will be discussed particular, mass spectrometric properties of respect- later. Hence, in case of qualitative determination of ive analytes is of utmost importance in order to these compounds in urine samples of athletes characterize and identify compounds in a complex elected for doping controls, a positive test result matrix such as urine. Numerous investigations will be reported. In Fig. 5.11a, the EI-mass spectrum regarding the mass spectrometric behavior of per- of 17α-methyl-5α-androstan-3α,17β-diol after bis- formance enhancing or masking agents and their trimethylsilylation is presented, and Fig. 5.11b detection for doping control purposes have been shows a typical GC-MS chromatogram with char- performed (Masse et al. 1989; de Boer et al. 1991; acteristic ion traces of the methyltestosterone metab- Donike et al. 1995; Ayotte et al. 1996; Shackleton et al. olites found in a urine sample. 1997; Bowers 1998; Aguilera et al. 1999; Thevis et al. 2000, 2001, 2002, 2003a; Ventura et al. 2000; Leinonen designer steroids et al. 2002), and the principles of some procedures will be described in the following. The problem of so-called designer steroids triggered an avalanche in the sport as well as the scientific world in October 2003 (Knight 2003) when the Anabolic steroids doping control laboratory of the University of Considering statistical data regarding conducted California, Los Angeles (UCLA) identified a com- doping controls and classes of abused drugs, pound related to gestrinone, a drug administered in anabolic steroids are the most frequently detected cases of endometriosis. Its hydrogenation at the prohibited compounds. In 2001, for example, more ethinyl residue at carbon 17 results in a steroid hor- than 40% of the banned substances identified by mone termed tetrahydrogestrinone (THG), which can 25 IOC-accredited laboratories were related to be considered as an analog to the highly efficient anabolic agents. One representative of this category anabolic steroid trenbolone, but the physiological doping analysis with gc-ms and lc-ms/ms 63

Scheme 5.7 Metabolism of methyltestosterone to 17α-methyl-5α-androstan-3α,17β-diol (I), 17α-methyl-5β-androstan- 3α,17β-diol (II) and their glucuronides.

143 CH 100 73 3 (II) m/z 143 OTMS 18 000 (I) 80 16 000 14 000 60 TMSO 12 000 H 10 000 m/z 435 40 8000 m/z 255 Abundance 20 6000 270 105 360 4000 Relative abundance (%) 213228255 318 435 0 2000 50 100 150 200 250 300 350 400 450 0 11.90 12.00 12.10 12.20 12.30 12.40 12.50 12.60 12.70 (a) m/z (b) Time (min)

Fig. 5.11 (a) Electron ionization-mass spectrum of the bis-trimethylsilylated 17α-methyl-5α-androstan-3α,17β-diol (mol. wt. = 450); (b) gas chromatography-mass spectrometry (GC-MS) chromatogram of characteristic ion traces of bis- trimethylsilylated 17α-methyl-5α-androstan-3α,17β-diol (I) and 17α-methyl-5β-androstan-3α,17β-diol (II). effects and side effects of THG have never been as THG, are invested with invisibility for these clinically investigated. The commonly employed conventional methods. Hence, more flexible assays strategy of doping control laboratories analyzing have been established enabling the detection of pharmaceutically produced and clinically tested known as well as unknown drugs by common struc- remedies is obviously not sufficient to cope with the tures such as a principal steroid nucleus, which is willingness of some athletes to cheat and to risk possible in particular by means of modern LC-MS/ their health in order to win by a short head against MS systems. competitors. Owing to the fact that many drug screening procedures are based on the comparison Endogenous steroids of reference compounds to urine samples by means of mass spectrometric techniques such as selected While the administration of anabolic steroids causes ion monitoring (SIM) or multiple reaction monitor- the presence of metabolites normally not occur- ing (MRM), unknown derivatives of remedies, such ring in human urine samples, doping by means of 64 chapter 5

testosterone is more difficult to uncover as it is also mation about analytes, specificity and selectivity produced endogenously. Here, different approaches of mass analyzers are provided by CAD of ionized have been established to determine applications of drugs and analysis of derived fragments. This testosterone, and the two most frequently utilized requires the knowledge of proton affinity and dis- tools are the testosterone/epitestosterone (T/E) sociation behavior after efficient activation of target ratio and the so-called isotope ratio mass spectro- molecules by CAD, which differs significantly from metry (IRMS) of carbons. The profile of endogenous fragmentation routes observed with EI. For diuret- β steroids varies under numerous influences (Geyer ics as well as for the majority of 2-agonists (except et al. 1996), but the ratio of T and E has proven to be for salbutamol, see Quantitative analysis of pro- a reliable parameter that indicates abuse of testo- hibited drugs below), qualitative analysis of these sterone as the production of epitestosterone is inde- drugs is sufficient, and typical ESI product ion pendent from testosterone. A threshold ratio of spectra of epithiazide and fenoterol are presented in six is utilized that does not immediately entail a Fig. 5.12. Commonly, extracted ion chromatograms positive test result but triggers follow-up studies to enable the sensitive detection of these compounds test for an abnormal but naturally elevated testo- in biological matrix, and confirmation of their pres- sterone concentration. ence is obtained by comparison of relative abund- With the availability of IRMS instruments, num- ances of product ions. In 2001, approximately 17.5% erous investigations have been published demon- of the banned substances detected in 25 IOC-accred- strating the possibility to distinguish between ited laboratories in doping control samples were β endogenously generated and chemically synthes- drugs related to 2-agonists, and 5% were classified ized testosterone by MS. Naturally produced tes- as diuretics. tosterone comprises different 13C/12C ratios than chemically synthesized analogs that are used for Quantitative analysis of medical supplementation. By means of GC, com- prohibited drugs bustion of eluting analytes and mass spectrometric measurement of resulting carbon dioxide provides For several compounds, including stimulants such information about the origin of testosterone by as ephedrines, metabolites of anabolic steroids such 13 12 β different contents of C and C (Horning et al. 1998; as nandrolone, and 2-agonists such as salbutamol, Aguilera et al. 1999). threshold levels have been established, which are the basis of the decision as to whether a sample is reported positive or negative. Various reasons are Diuretics and β -agonists 2 given for this regulation. Ephedrines are ingredients Two representatives of classes of compounds fre- of many remedies frequently used in cold therapy, quently analyzed by means of LC-MS(/MS) are hence their use is legal in the sense of anti-doping β diuretics and 2-agonists. In particular the category rules as long as their urinary concentration does not of diuretics demonstrates the chemical heterogene- exceed 5, 10, or 25 µg·mL–1 for the derivatives nore- ity of drugs administered for comparable or ident- phedrine, ephedrine and pseudoephedrine, respect- ical purposes. Here, primarily negative ionization ively. Salbutamol, a sympathomimetic agent, is β is performed (Thieme et al. 2001; Thevis et al. 2002, one of four permitted 2-agonists (in addition to 2003a) owing to the acidity of drugs belonging to salmeterol, terbutaline and formoterol), as long as the group of diuretics, but few compounds such as the application is conducted via inhalation. As the β triamterene require positive ionization. For 2- differentiation between orally administered pills agonists, protonation of analytes and detection and pressurized aerosol dosage is very complicated, of positively charged molecules is the method of the presence of salbutamol in competition samples choice. As described in Analytical techniques above, is reported to the corresponding federation if the ion sources interfacing LC to MS mainly generate urinary level exceeds 100 ng·mL–1, and in out-of- protonated or deprotonated molecules without any competition testing a threshold of 1 µg·mL–1 has considerable fragmentation. Thus, structure infor- been established owing to anabolic effects observed doping analysis with gc-ms and lc-ms/ms 65

424 100 135 m/z 135 m/z 107 100 O O H NO S S 2 2 NH 90 H3C 90 78 HO SCF3 80 80 Cl N N H 269 70 H 70 300 60 OH 60 205 50 HO OH 50 310 40 40 404 107 364 384 30 30 341 20 20 126 286 62 260 279 Relative abundance (%) Relative abundance (%) 152 304 95 10 10 150 228 164 170 169 0 0 40 60 80 100120140160180200220240260280300 50 100 150 200 250 300 350 400 (a)m/z (b) m/z

β = Fig. 5.12 (a) Electrospray ionization (ESI) product ion spectrum of the 2-agonist fenoterol (mol. wt. 303); (b) ESI product ion spectrum of the diuretic agent epithiazide (mol. wt. = 425).

with some sympathomimetics if administered in ionization techniques due to a faster sample prepara- doses much higher than therapeutically used. The tion as no derivatization of analytes is required. In presence of the nandrolone metabolite 5α-estran- addition, more flexible mass spectrometric experi- 3α-ol-17-one (norandrosterone) in urine samples of ments to determine and characterize known thera- elite athletes is observed to certain extents based on peutics and also unknown designer drugs have endogenous production. Thus, also here threshold become possible by frequently employed triple values have been introduced with 2 ng·mL–1 for quadrupole or ion trap analyzers supporting the male and 5 ng·mL–1 for female athletes. Numerous fight against doping and the illegal use of drugs. studies were performed in order to substantiate The range of substances important for doping con- these levels, and various ‘parameters’ influencing trol analysis has changed ever since lists of pro- the endogenous generation of this metabolite were hibited compounds and methods of doping have taken into consideration, for example, high physio- existed, and laboratories permanently expand or logical stress or pregnancy, which cause signific- modify analytical procedures within this dynamic antly elevated concentrations in urine specimens. process in order to improve performance regarding Quantitation of these substances is accomplished by sensitivity, specificity and flexibility to limit the mis- calibration curves utilizing regular sample prepara- use of drugs in sport as well as to protect athletes tion procedures and appropriate internal standards from false suspicion. Here, new developments con- with comparable or identical physicochemical prop- cerning high speed and high resolution chromato- erties (Schänzer & Donike 1995). graphy, high resolution and high sensitivity MS as well as modern ionization techniques provide valu- Summary able tools for analytical laboratories to obtain even more detailed information on the administered Doping control analysis of low molecular weight therapeutics, for example their chemical structures drugs is generally based on chromatographic and and metabolism, and enable extended time frames mass spectrometric techniques that enable the for the determination of drug abuse. As many detection and identification of remedies and their therapeutics, such as anabolic steroids, are applied metabolites in body fluids such as urine and blood. during out-of-competition periods but preserve While early procedures employed mainly GC, vari- performance-enhancing effects for several weeks, ous analyzers such as FIDs and NPDs, as well as doping control analysis requires in- as well as out- MS, recently published assays utilize primarily LC of-competition tests and the utmost specificity and interfaced to MS by means of atmospheric pressure sensitivity of analytical procedures. 66 chapter 5

References

Aguilera, R., Catlin, D.H., Becchi, M. et al. Burstin, S. (1963) La Lucha contra el Liquid Chromatography in Biochemistry (1999) Screening urine for exogenous Dopage. Revista de Derecho Sportiva 3, (Henschen, A., Hupe, K.P., Lottspeich, F. testosterone by isotope ratio mass 583–636. & Voelter, W., eds.). VCH spectrometric analysis of one Cartoni, G.P. & Cavalli, A. (1968) Detection Verlagsgesellschaft, Weinheim: 17–78. pregnanediol and two androstanediols. of doping by thin-layer and gas Fluri, K., Rivier, L., Dienes-Nagy, A. et al. Journal of Chromatography B 727, 95–105. chromatography. Journal of (2001) Method for confirmation of Ariëns, E.J. (1965) General and Chromatography 37, 158–161. synthetic corticosteroids in doping urine pharmacological aspects of doping. Deventer, K., Delbeke, F.T., Roels, K. & samples by liquid chromatography- In: Doping (de Schaepdryver, A. & Van Eenoo, P. (2002) Screening for 18 electrospray ionization mass Hebbelink, M., eds.). Pergamon Press, diuretics and probenecid in doping spectrometry. Journal of Chromatography Oxford: 27–50. analysis by liquid chromatography- A 926, 87–95. Axelrod, J. (1954) Studies on tandem mass spectrometry. Biomedical Geyer, H., Schänzer, W., Mareck-Engelke, sympathomimetic amines. II. The Chromatography 16, 529–535. U. & Donike, M. (1996) Factors biotransformation and physiological Dole, M., Mack, L.L., Hines, R.L. et al. influencing the steroid profile. In: disposition of d-amphetamine, d-p- (1968) Molecular beams of macroions. Proceedings of the 14th Cologne Workshop hydroxyamphetamine and d- Journal of Chemical Physics 49, 2240–2249. on Doping Analysis (Donike, M. Geyer, methamphetamine. Journal of Donike, M. (1969) N-Methyl-N- H., Gotzmann A. & Mareck-Engelke, U., Pharmacology and Experimental trimethylsilyl-trifluoracetamid, ein eds.). Sport & Buch Strauß, Cologne: Therapeutics 110, 315–326. neues Silylierungsmittel aus der Reihe 95–114. Axen, R., Porath, J. & Ernback, S. (1967) der silylierten Amide. Journal of Geyer, H., Schänzer, W., Mareck-Engelke, Chemical coupling of peptides and Chromatography 42, 103–104. U., Nolteernsting, E. & Opfermann, G. proteins to polysaccharides by means of Donike, M. (1973) Acylierung mit (1997) Screening procedure for anabolic cyanogen halides. Nature 24, 1302–1304. Bis(Acylamiden): N-Methyl- steroidsathe control of the hydrolysis Ayotte, C., Goudreault, D. & Charlebois, bis(trifluoracetamid) und with deuterated androsterone A. (1996) Testing for natural and Bis(trifluoracetamid), zwei neue glucuronide and studies with direct synthetic anabolic agents in human Reagenzien zur Trifluoracetylierung. hydrolysis. In: Proceedings of the 15th urine. Journal of Chromatography B 687, Journal of Chromatography 78, 273–279. Cologne Workshop on Doping Analysis 3–25. Donike, M. & Derenbach, J. (1976) Die (Schänzer, W., Geyer, H., Gotzmann, A. Beckett, A.H. & Rowland, M. (1965) selektive Derivatisierung unter & Mareck-Engelke, U., eds.). Sport & Determination and identification of kontrollierten Bedingungen: Ein Weg Buch Strauß, Cologne: 99–101. amphetamine in urine. Journal of zum Spuren-Nachweis von Aminen. Hermanson, G.T., Mallia, A.K. & Smith, Pharmacy and Pharmacology 17, 59–60 Zeitschrift Analytische Chemie 279, P.K. (1992) Immobilized Affinity Ligand Beckett, A.H. & Wilkinson, G.R. (1965) 128–129. Techniques. Academic Press, San Diego, Identification and determination of Donike, M. & Zimmermann, J. (1980) CA. ephedrine and its congeners in urine by Zur Darstellung von Trimethylsilyl-, Höher, J. & Troidl, H. (1995) Doping gas chromatography. Journal of Pharmacy Triethylsilyl- und tert. im Sport. Deutsche Zeitschrift für and Pharmacology 17 (Suppl.), S104–S106. -Butyldimethylsilyl-enoläthern Sportmedizin 46, 270–282. Beckett, A.H., Tucker, G.T. & Moffat, A.C. von Ketosteroiden für gas- Horning, S., Geyer, H., Gotzmann, A., (1967) Routine detection and chromatographische und Flenker, U. & Schänzer, W. (1998) identification in urine of stimulants and massenspektrometrische Detection of exogenous steroids by other drugs, some of which may be used Untersuchungen. Journal of 13C/12C analysis. In: Proceedings of the to modify performance in sport. Journal Chromatography 202, 483–486. 16th Cologne Workshop on Doping of Pharmacy and Pharmacology 19, Donike, M., Jaenicke L., Stratmann, D. Analysis (Schänzer, W., Geyer, H., 273–294. & Hollmann, W. (1970) Gas Gotzmann, A. & Mareck-Engelke, U., de Boer, D., Gainza Bernal, M.E., van chromatographic detection of nitrogen- eds.). Sport & Buch Strauß, Cologne: Ooyen, R.D. & Maes, R.A. (1991) The containing drugs in aqueous solutions 135–148. analysis of trenbolone and the human by means of the nitrogen detector. International Olympic Committee (IOC) urinary metabolites of trenbolone Journal of Chromatography 52, 237–250. (2003) List of Prohibited Substances and acetate by gas chromatography/mass Donike, M., Zimmermann, J., Bärwald, Methods of Doping. IOC, Lausanne. spectrometry and gas chromatography/ K.R. et al. (1984) Routinebestimmung Iribarne, J.V. & Thomson, B.A. (1976) On tandem mass spectrometry. Biological von Anabolika im Harn. Deutsche the evaporation of small ions from Mass Spectrometry 20, 459–466. Zeitschrift für Sportmedizin 35, 14–24. charged droplets. Journal of Chemical Bowers, L.D. (1998) Athletic drug testing. Donike, M., Ueki, M., Kuroda, Y. et al. Physics 64, 2287–2294. Clinics in Sports Medicine 17, 299–318. (1995) Detection of dihydrotestosterone Kebarle, P. & Ho, Y. (1997) On the Bradlow, H.L. (1968) Extraction of steroid (dht) doping: alterations in the steroid mechanism of electrospray mass conjugates with a neutral resin. Steroids profile and reference ranges for dht and spectrometry. In: Electrospray Ionization 11, 265–272. its 5 α-metabolites. Journal of Sports Mass Spectrometry—Fundamentals, Budzikiewicz, H. (1998) Medicine and Physical Fitness 35, 235–250. Instrumentation and Applications (Cole, Massenspektrometrie. Wiley–VCH, Engelhardt, H., Müller, H. & Schön, U. R.B., ed.). John Wiley & Sons, New York: New York. (1985) The column. In: High Performance 3–63. doping analysis with gc-ms and lc-ms/ms 67

Keller, R.E. & Ellenbogen, W.C. (1952) Experimental Pharmacology , vol. 117 Sigmund, G., Horning, S., Seinsch, I. & The determination of d-amphetamine (Greger, R.F., Knauf, H. & Mutschler, E., Schänzer, W. (1998) Confirmation of in body fluids. Journal of Pharmacology eds.). Springer-Verlag, Berlin: clenbuterol. Proceedings of the 16th and Experimental Therapeutics 106, 77–82. 173–200. Cologne Workshop on Doping Analysis Knight, J. (2003) Drugs bust reveals Morgan, C.E. (1957) Drug administration (Schänzer, W., Geyer, H., Gotzmann, A. athletes’ secret steroid. Nature 425, 752 to racing animals. Journal of the American & Mareck-Engelke, U., eds.). Sport & (News). Veterinary Medical Association 130, Buch Strauß, Cologne: 129–133. Kolb, H. & Patt, P.W. (1965) Beitrag 240–243. Spyridaki, M.H., Tsitsimpikou, C.J., Siskos, zum Arzneimittelnachweis in Opfermann, G. & Schänzer, W. (1996) P.A. & Georgakopoulos, C.G. (2001) Körperflüssigkeiten durch Trimethylsilylationaaspects for Determination of ephedrines in Gaschromatographie. Arzneimittel- derivatisation. In: Proceedings of the urine by gas chromatography-mass Forschung 15, 924–927. 14th Cologne Workshop on Doping spectrometry. Journal of Chromatography Lang, H.J. & Hropot, M. (1995) Discovery Analysis (Schänzer, W., Geyer, H., B 758, 311–314. and development of diuretic agents. In: Gotzmann, A. & Mareck-Engelke, U., Stallion, A., Zhang, F.S., Chance, W.T., Handbook of Experimental Pharmacology, eds.). Sport & Buch Strauß, Cologne: Foley-Nelson, T. & Fischer, J.E. (1991) vol. 117 (Greger, R.F., Knauf, H. & 247–252. Reversal of cancer cachexia in rats by Mutschler, E., eds.). Springer-Verlag, Patterson, G.E., Guymon, A.J., Riter, L.S. cimaterol and supplemental nutrition. Berlin: 141–172. et al. (2002) Miniature cylindrical ion Surgery 110, 678–684. Leinonen, A., Kuuranne, T. & Kostiainen, trap mass spectrometer. Analytical Thevis, M., Opfermann, G. & Schänzer, W. R. (2002) Liquid chromatography/ Chemistry 74, 6145–6153. (2000) Detection of the plasma volume mass spectrometry in anabolic steroid Prokop, L. (1970) The struggle against expander hydroxyethyl starch in human analysisaoptimization and comparison doping and its history. Journal of urine. Journal of Chromatography B 744, of three ionization techniques: Sports Medicine and Physical Fitness 10, 345–350. electrospray ionization, atmospheric 45–48. Thevis, M., Opfermann, G. & Schänzer, W. pressure chemical ionization and Prokop, L. (2002) Zur Geschichte des (2001) High speed determination of β- atmospheric pressure photoionization. Dopings. Sportärztewoche, December 1–7, receptor blocking agents in human urine Journal of Mass Spectrometry 37, Kaprun. by liquid chromatography/tandem 693–698. Reeds, P.J., Hay, S.M., Dorward, P.M. & mass spectrometry. Biomedical Lisi, A.M., Trout, G.J. & Kaslauskas, R. Palmer, R.M. (1988) The effect of β- Chromatography 15, 393–402. (1991) Screening for diuretics in human agonists and antagonists on muscle Thevis, M., Schmickler, H. & Schänzer, W. urine by gas chromatography-mass growth and body composition of young (2002). Mass spectrometric behavior spectrometry with derivatisation by rats (Rattus sp.). Comparative Biochemistry of thiazide-based diuretics after direct extractive alkylation. Journal of and Physiology C 89, 337–341. electrospray ionization and collision- Chromatography 563, 257–270. Richter, D. (1938) A colour reaction for induced dissociation. Analytical Lottspeich, F. & Zorbas, H. (1998) benzedrine. Lancet 238, 1275. Chemistry 74, 3802–3808. Bioanalytik. Spektrum Akademischer Schänzer, W. (1996) Metabolism of Thevis, M., Schmickler, H. & Schänzer, W. Verlag, Heidelberg. anabolic androgenic steroids. Clinical (2003a). Effect of the location of Louris, J.N., Cooks, R.G., Syka, J.E.P. et al. Chemistry 42, 1001–1020. hydrogen abstraction on the (1987) Instrumentation, applications, Schänzer, W. & Donike, M. (1995) fragmentation of diuretics in negative and energy disposition in quadrupole Synthesis of deuterated steroids for electrospray ionization mass ion-trap tandem mass spectrometry. GC/MS quantification of endogenous spectrometry. Journal of the American Analytical Chemistry 59, 1677–1685. steroids. In: Proceedings of the 12th Society for Mass Spectrometry 14, Louris, J.N., Amy, J.W., Ridley, T.Y. & Cologne Workshop on Dope Analysis 658–670. Cooks, R.G. (1989) Injection of ions (Donike, M., ed.). Sport & Buch Strauß, Thevis, M., Sigmund, G. & Schänzer, W. into a quadrupole ion trap mass Cologne: 93–112. (2003b) Confirmation of spectrometer. International Journal of Schänzer, W., Delahaut, P., Geyer, H., ephedrinesacomparison between GC- Mass Spectrometry and Ion Processes 88, Machnik, M. & Horning, S. (1996) Long- MS and LC-MS/MS. In: Proceedings of 97–111. term detection and identification of the 21st Cologne Workshop on Doping Machnik, M., Geyer, H., Horning, metandienone and stanozolol abuse in Analysis (Schänzer, W., Geyer, H., S. et al. (1999) Long-term detection of athletes by gas chromatography-high- Gotzmann, A. & Mareck-Engelke, U., clenbuterol in human scalp hair by gas resolution mass spectrometry. Journal of eds.). Sport & Buch Strauß, Cologne: chromatography-high-resolution mass Chromatography B 687, 93–108. 303–307. spectrometry. Journal of Chromatography Scott, R.P.W. (1982) Microbore columns in Thieme, D., Grosse, J., Lang, R., Mueller, B 723, 147–155. liquid chromatography. In: Techniques in R.K. & Wahl, A. (2001) Screening, Masse, R., Ayotte, C., Bi, H.G. & Dugal, R. Liquid Chromatography (Simpson, C.F., confirmation and quantification of (1989) Studies on anabolic steroids. III. ed.). John Wiley & Sons, Chichester: diuretics in urine for doping control Detection and characterization of 79–96. analysis by high-performance liquid stanozolol urinary metabolites in Shackleton, C.H., Phillips, A., Chang, T. chromatography-atmospheric pressure humans by gas chromatography-mass & Li, Y. (1997) Confirming testosterone ionization tandem mass spectrometry. spectrometry. Journal of Chromatography administration by isotope ratio mass Journal of Chromatography A 757, 497, 17–37. spectrometric analysis of urinary 49–57. Möhrke, W. & Ullrich, F. (1995) androstanediols. Steroids 62, Van Eenoo, P., Delbeke, F.T., Roels, K. & Metabolism of diuretics. In: Handbook of 379–387. De Backer, P. (2001) Simultaneous 68 chapter 5

quantitation of ephedrines in urine papierchromatographisch isolierter vol. 2 (Emmet, P.H., ed.). Reinhold, by gas chromatography-nitrogen- Arzneistoffe. Archives of Toxicology 16, New York: 105–165. phosphorus detection for doping control 63–73. Wilhelm, M. & deStevens, G. (1976) purposes. Journal of Chromatography B Voyksner, R.D. (1997) Combining liquid Antihypertensive agents. Progress in 760, 255–261. chromatography with electrospray mass Drug Research. 20, 197–259. Venerando, A. (1963) Pathologia del spectrometry. In: Electrospray Ionization www.g-o.de/kap4/40gboo91.thm doping e mezzi di controllo. Medicina Mass Spectrometry—Fundamentals, (06-2003) dello Sport 3, 945–959. Instrumentation and Applications (Cole, Yamashita, M. & Fenn, J.B. (1984a) Ventura, R., Segura, J., Berges, R. et al. R.B., ed.). John Wiley & Sons, New York: Electrospray ion source. Another (2000) Distinction of inhaled and oral 323–341. variation on the free-jet theme. salbutamol by urine analysis using Wagner, J.C. (1989) Abuse of drugs used to Journal of Chemical Physics 88, conventional screening procedures for enhance athletic performance. American 4451–4459. doping control. Therapeutic Drug Journal of Hospital Pharmacy 46, Yamashita, M. & Fenn, J.B. (1984b) Monitoring 22, 277–282. 2059–2067. Negative ion production with the Vidic, E. (1956) Eine Methode Wheeler, A. (1955) Reaction rates and electrospray ion source. Journal of zur Identifizierung selectivity in catalyst pores. In: Catalysis, Chemical Physics 88, 4671–4675. Chapter 6

The Reproductive Axis

JOHANNES D. VELDHUIS AND ARTHUR L. WELTMAN

Abstract Introduction The reproductive axis is largely stable in the face Critical assessment of the impact of vigorous exer- of intensive physical training in healthy men and cise training and strenuous sports engagements women. However, altered reproductive-hormone on the endocrine system is a recent medical accom- outflow is associated with any of numerous doping plishment. Indeed, the majority of earlier clinical protocols, exhaustive or non-escalating training reports cited in textbooks or forecast in contempor- schedules, concomitant weight loss, psychosocial ary advertisements of health-related supplements stress, inadequate caloric intake for workload and are flawed by significantly confounding issues, sustained strenuous exertion in adolescence. Expo- which render definitive interpretation impossible. sure to anabolic steroids induces loss of menstrual Scientific requirements for valid inference currently cyclicity in women, a decrease in spermatogenesis include (non-exclusively): (i) prospective evaluation in men, and a significant reduction in protective of demonstrably normal healthy individuals; (ii) (high-density lipoprotein) cholesterol concentra- supervised, graded, escalating training schedules of tions in both genders. More subtle endocrine and quantifiable intensity and duration; (iii) concurrent metabolic adaptations are detectable in the absence longitudinally monitored gender and age-matched of the foregoing risk factors, but their medical control cohorts without exercise intervention; (iv) significance is not established. For example, one verifiable adequacy of concomitant nutritional sup- longitudinal investigation showed that the only port to match individual energy demand; (v) docu- detectable effect of supervised long-distance run- mented absence of covert drug or hormone use ning training to complete a marathon in healthy (doping); (vi) baseline and anterograde assessment young women is slight abbreviation of the post- of perceived psychosocial stress; and (vii) compre- ovulatory (luteal) phase of the menstrual cycle hensive family history to identify genetic factors within the normal range. Cross-sectionally based known to predict reproductive disorders in the gen- reports describe an array of reproductive abnorm- eral population. At present, no single study fulfills alities in athletes, but such data are confounded the foregoing ensemble expectations. More subtle by one or more known comorbid factors. Accord- understanding of metabolic and endocrine implica- ingly, supervised, graded, non-exhaustive, voluntary tions of vigorous and competitive sports will thus strenuous exercise in healthy adults maintaining an require further clinical and physiological studies. adequate caloric intake has minimal potential for adverse reproductive consequences. Principles of reproductive physiology Contemporary knowledge of human reproduct- ive physiology embraces an integrative view of 69 70 chapter 6

Neuroanatomic locus Primary messenger (–) CNS Sex steroids (+) GnRH (–) Portal Hypothalamus microvasculature

Pituitary (–) gland LH LH, FSH, prolactin, sex steroids Circulation LH

Time

Ovary: estrogen, progesterone Gonad Testis: testosterone Gametogenesis

(+) Sex steroid-responsive tissues

Fig. 6.1 Schematized illustration of basic elements maintaining homeostasis of the male and female reproductive axes. Signals arising in the central nervous system and peripheral tissues converge on an ensemble of about 1200 hypothalamic gonadotropin-releasing hormone (GnRH) -secreting neurons. Synchrony of neuronal excitation releases pulses of GnRH into a microvascular portal-venous system (solid arrows [+]), which connects the mediobasal hypothalamus to the pituitary gland. GnRH drives transcription, translation, processing and secretion of luteinizing hormone (LH) and follicle- stimulating hormone (FSH) by gonadotrope cells. The general circulation delivers LH to gonadal (ovarian or testicular) compartments. LH stimulates the synthesis and secretion of testosterone (and thereby estradiol) in men and women, and progesterone in women after ovulation. In women, FSH promotes Graafian follicle development, oocyte fertilizability and ovarian responsiveness to LH. In men, FSH synergizes with LH-stimulated intratesticular testosterone to promote early stages of spermatogenesis in seminiferous tubules. Sex steroids act on peripheral target tissues, such as muscle, fat, breast, bone, skin, sexual organs, the hypothalamus and pituitary gland (open arrow [–]). In women, estrogen and progesterone exert sequential feedback (inhibition) and feedforward (enhancement), which cycle mediates the preovulatory LH surge. The same sex hormones stimulate uterine endometrial growth and maturation, which are necessary for implantation of a (fertilized) blastocyst in pregnancy.

hormonal signal exchange among the central nerv- adequate caloric repletion of expended energy is a ous system (hypothalamus), anterior pituitary gland minimal (necessary, but not sufficient) prerequisite (gonadotrope cells) and gonad (testis and ovary) for successful reproductive function during strenu- (Fig. 6.1). The foregoing set of neuroendocrine ous training (Loucks 2003). glands (the reproductive axis) is linked to signals Endocrine adaptations to physical exertion are associated with pubertal development, somatic transduced via multiple levels of non-exclusive growth, body composition, stress responses, appetite, control: (a) central-neural and blood-borne inputs energy expenditure and insulin action (Urban et al. to the hypothalamus; (b) hypothalamic integration 1988; Evans et al. 1992; Giustina & Veldhuis 1998). to yield convergent signal outflow to the anterior An integrative perspective of the reproductive axis pituitary gland; (c) feedback to the hypothalamus is necessary to reconstruct the pathogenesis of endo- via core body temperature, inflammatory mediators, crine disturbances and discern the mechanisms tissue-derived metabolites (e.g. lactate, free fatty of normal adaptations to exercise. For example, acids) and secreted products of target cells (insulin- the reproductive axis 71

like growth factor stimulated in the liver by growth of substrates (e.g. glucose), metabolites (e.g. lac- hormone); (d) exogenous substrates (glucose, fat tate) and hormones (sex steroids, cortisol, insulin, and amino acids); and (e) stress-adaptive thyroidal growth hormone, etc). Homeostasis requires de- and adrenal hormones (e.g. thyroxine, cortisol and terministic physiological mechanisms, low back- epinephrine). This minimal network of interactions ground randomness, a normal genetic endowment, governs hypothalamic stimulation of gonadotrope adequate psychosocial support and successful cells by intermittent (burst-like) release of the deca- adaptation to internal and external stress (Veldhuis peptide, gonadotropin-releasing hormone (GnRH). 1996). GnRH induces (feeds forward on) the pituitary syn- thesis and secretion of luteinizing hormone (LH) Hypothalamo–pituitary–gonadal and follicle-stimulating hormone (FSH). These two homeostasis in strenuous physical proteins are key effectors of the ovary and testis. The training size and time-pattern of GnRH pulses jointly deter- mine the amounts of LH and FSH secreted (Urban Longitudinal clinical studies et al. 1988; Evans et al. 1992; Veldhuis 1999). LH drives sex-steroid production (below). FSH controls Few clinical investigations have entailed prospect- sperm and egg development and the production of ive randomization of a demonstrably healthy cohort a feedback protein, inhibin. No clinically significant of volunteers into strata of unequal training intens- alterations in FSH or inhibin secretion have been ity, duration and/or type. One longitudinal study attributable to physical training (Veldhuis et al. assigned young menstruating women randomly to 1998). This statement is also true of prolactin, a lacta- either low (sublactate threshold) or high (supralac- tional hormone that increases transiently with a tate threshold) intensity physical training (Rogol et bout of exercise and in response to diverse psycho- al. 1992). The program comprised supervised long- logical and physical stressors. distance running for 18 months with the individual In men, LH acts on testicular Leydig cells to pro- goal of completing a marathon. Reproductive hor- mote the synthesis and secretion of 4–6 mg of mones were measured at baseline and after 1 year testosterone each day (Urban et al. 1988). In women, of gradually escalating training volume at the pre- LH stimulates ovarian theca and postovulatory assigned exercise intensity. LH, FSH, prolactin luteal cells to secrete about 0.15 mg of testosterone and estradiol did not change detectably in the low- (a precursor of estrogens) and 10–20 mg of proges- intensity exercise cohort or in a concurrent longit- terone per day (Evans et al. 1992). Testosterone, estro- udinal control group (recreationally active women). gen and progesterone reversibly suppress (feed Women who trained consistently above the indivi- back on) the hypothalamo–pituitary unit, thereby dually determined lactate threshold for an identical supervising intermittent secretion of GnRH, LH and total running distance of > 500 miles (> 800 km) FSH (Veldhuis 1999). Time-delimited inhibition by manifested a small (1.8-day) abbreviation of the sex-steroid hormones is a fundamental regulatory luteal phase of the menstrual cycle with no evident mechanism that preserves reproductive homeostasis. difference in mean progesterone concentrations In general, homeostasis is the aggregate outcome (Rogol et al. 1992). Shortening of luteal-phase length of reciprocal feedforward (stimulatory) and feed- did not exceed normal month-to-month differences back (repressive) interactions that enforce repeated, recognized in healthy premenopausal women. reversible incremental adjustments to maintain The importance of adequate nutritional intake hormone concentrations within the physiological to match total caloric expenditure is illustrated in a range for age, gender and species. Autoregulation is recent prospective randomized intervention. In this analogous to a cruise-control device, which holds 2-month-long study, women undertaking weight a vehicle’s speed within the desired preset range. reduction during training were more likely to Cascades of signal exchange in healthy indivi- develop disturbances in ovulatory or luteal-phase duals thereby ensure life-supporting concentrations timing than individuals training identically with 72 chapter 6

Baseline Fasting GnRH + Fasting 8 8 8 6 6 6 4 4 4 2 2 2

) 0 0 0

–1 0 400 800 1200 1600 0 400 800 1200 1600 0 400 800 1200 1600

16 16 16 12 12 12 8 8 8 4 4 4 0 0 0 0 400 800 1200 1600 0 400 800 1200 1600 0 400 800 1200 1600

Serum LH concentration (IU·L 20 20 20 16 16 16 12 12 12 8 8 8 4 4 4 0 0 0 0 400 800 1200 1600 0 400 800 1200 1600 0 400 800 1200 1600 Time (min)

Fig. 6.2 Short-term fasting suppresses, and gonadotropin-releasing hormone (GnRH) infusion restores, pulsatile luteinizing hormone (LH) release in healthy young men. Profiles are shown in three individuals (top, middle, bottom). LH concentrations were sampled every 10 min for 24 h. The triptych illustrates the fed session (left); day 3 of fasting with saline injections (middle); and day 3 of fasting accompanied by intravenous injection of synthetic GnRH (100 ng·kg–1 every 90 min) (right). Repeated GnRH pulses restore LH secretion. These data signify that caloric deprivation represses hypothalamic GnRH outflow to pituitary gonadotrope cells, which retain responsiveness to GnRH stimulation. (Reprinted with permission from Aloi et al. 1997.)

unrestricted caloric ingestion (Bullen et al. 1985). In induced by the injection of synthetic GnRH peptide each case, menstrual cycles became normal within (Fig. 6.3). Precisely why LH pulsatility is disrupted 6 months of completing the study. in some but not all long-distance runners is not known. Epidemiological correlations indicate that one or more apparent reproductive risk factors Observational studies often prevail (see below). Cross-sectional appraisal of hypothalamo–pituitary Table 6.1 highlights several factors that may –ovarian hormones in collegiate long-distance run- increase the risk of developing irregular menstrual ners has revealed a subset of subjects with irregular cycles in strenuously training young women. menstrual cyclicity due presumptively to reduced Among diverse considerations, deficient caloric hypothalamic GnRH release (Rogol et al. 1983; intake in relation to any given level of energy expend- Veldhuis et al. 1985; MacConnie et al. 1986) (Fig. 6.2). iture is an established risk for oligomenorrhea Blunted GnRH secretion in this setting is inferred in prospectively randomized analyses (vide infra) from decreased LH pulse frequency in the pre- (Bullen et al. 1985; Loucks 2003). Albeit not definit- sence of normal or accentuated pituitary LH release ive, several other inferred clinical associations are the reproductive axis 73

60 GnRH GnRH GnRH GnRH 2.5 µg 5.0 µg 10.0 µg 25.0 µg Controls 50 Runners ) –1 40

30

20 Serum LH (mIU·mL

10

051015 20 25 30 Sample number

Fig. 6.3 Accentuation of low-dose gonadotropin-releasing hormone (GnRH) -stimulated luteinizing hormone (LH) release in cross-sectionally recruited female collegiate long-distance runners with irregular menstrual cycles (see Fig. 6.2). Synthetic GnRH peptide was injected intravenously in escalating doses at the indicated times. LH concentrations were monitored by sampling blood from the forearm every 20 min. Greater LH secretion stimulated by low doses of GnRH in women athletes than controls indicates that the pituitary gland is capable of producing LH in response to an adequate (hypothalamic) GnRH stimulus. Clinically inferred risk factors associated with menstrual dysfunction in such individuals are highlighted in Table 6.1. (Adapted with permission from Veldhuis et al. 1985.)

Table 6.1 Factors associated with increased prevalence of menstrual irregularity in athletes studied cross-sectionally and/or by self-referral to a physician.

Apparent risk factor Putative mechanism

Self-imposed dieting, heavy laxative use, anorexia Metabolic and stress-adaptive reduction in hypothalamic or bulimia GnRH release Anabolic steroids, testosterone and progestins Direct feedback suppression of LH and GnRH secretion High doses of injected human growth hormone Prolactin-like effect to limit brain GnRH secretion Psychosocial stress Central nervous-system inhibition of GnRH outflow Personal or family history of reproductive disorder Unrelated primary disease; self-selection for study

GnRH, gonadotropin-releasing hormone; LH, luteinizing hormone. helpful in adumbrating possible avenues for further stimulation of LH secretion by the anterior pituitary investigation (Veldhuis et al. 1998). gland (Bergendahl et al. 1996, 2000; Loucks 2003). Insufficient total caloric intake, anorexia ner- For example, short-term fasting suppresses LH vosa, bulimia, binge eating, surreptitious vomiting pulses consistently in healthy adults. A study in and excessive use of purgatives all activate the young men showed that the 50% fall in LH and stress-adaptive hypothalamo–pituitary–adrenal axis. testosterone concentrations induced by fasting was Complex central-neural responses to metabolic prevented by injecting pulses of synthetic GnRH stress inhibit LH pulsatility by repressing hypotha- every 90 min intravenously (Aloi et al. 1997) (Fig. lamic GnRH release, thereby reducing feedforward 6.4). The outcome of this intervention demonstrates 74 chapter 6

15 12.5

10.0

10 7.5

5.0 5

2.5 ) –1 0 0.0 (a) 08001400 2000 0200 0800 (b) 08001400 2000 0200 0800

8 40 Serum LH (mIU·mL 6 30

4 20

2 10

0 0 (c) 08001400 2000 0200 0800 (d) 08001400 2000 0200 0800 Clock time

Fig. 6.4 Illustrative patterns of pulsatile luteinizing hormone (LH) release monitored every 20 min over 24 h in four young women identified cross-sectionally. Subjects were intercollegiate long-distance runners who reported irregular menstrual cycles. Some participants had a subnormal frequency (daily number) of LH pulses. The causal relationship, if any, to strenuous physical training is unknown. As discussed in the text, an independent longitudinal investigation demonstrated normal LH pulsatility before and after 1 year of endurance running in young women, each of whom ran at least 500 miles (800 km) (Rogol et al. 1992). (Reprinted with permission from Veldhuis et al. 1985.)

that caloric deficiency limits brain outflow of GnRH, A high degree of perceived psychosocial stress, but does block the pituitary gland. whether originating from or independently of the Abuse (non-therapeutic administration) of ana- training and competition environment, may also bolic steroids, testosterone, synthetic progestins and disrupt menstrual function (Evans et al. 1992; human growth hormone (which exerts a prolactin- Veldhuis et al. 1998). In addition, some athletes like inhibitory effect on GnRH secretion) can cause begin intensive training with pre-existing (unrecog- oligo- or amenorrhea (fewer than normal or no men- nized) hormonal abnormalities and/or a strong strual cycles, respectively) and oligo- or azoos- genetic potential for endocrine disease. Both factors permia (reduced or undetectable sperm counts). may be unmasked by strenuous athletic activity or Erythropoietin injections are not known to do so. detected independently of the training experience. the reproductive axis 75

Unresolved issues Relevant investigations would comprise longitud- inal, unbiased monitoring of FSH, inhibin and act- A significant and unresolved clinical issue is how ivin subunits, follistatin, spermatogenesis, Graafian two or more factors that individually heighten the follicle maturation, ovulation, luteinization, and risk of altered menstrual function interact adversely lifetable analyses of fecundity, fertility and live in the setting of vigorous physical training. This births inter alia. Anecdotal clinical observations pre- question is particularly difficult to address, because dict that changes in fecundity or fertility would be possible combinatorial factors are nearly illimitable. subtle rather than clinically significant. For example, what would be the overall effect The impact, if any, of physical exertion on the on menstrual function of strenuous training in a postpartum state is unknown. Illustrative unex- woman with a family but not a personal history of plored issues include the time required for recovery polycystic ovarian syndrome (present in 5.0–8.5% of menstrual cyclicity after pregnancy; lactational of otherwise healthy young women of diverse efficiency, milk volume and nutritional content for ethnicities), occasional prior menstrual irregularity the infant; and infant–maternal bonding. and minimal weight loss during the exercise pro- Greater clinical understanding is required of the gram? And how is the foregoing aggregate likeli- relatively neglected domain of sexual libido, poten- hood further influenced by the choice of a particular tia and satisfaction in athletes pursuing elite phys- sport; for example, competitive fencing, volleyball, ical training. And, medical assurance is needed that pole vaulting, the 500-m run or 100-m free-style all athletes have full access to instructional guides swimming. Observational inferences suggest that concerning risks of sexually transmissible disease menstrual dysfunction is less common among par- and communicable respiratory and gastrointestinal ticipants in cycling and swimming than other illness. Both necessities arise from the wide spec- athletic events (Evans et al. 1992). And under what trum of sociocultural, nutritional and geographic circumstances are risk factors that are inferable in venues engaged in international competition. women applicable to men? There are no explicit data at present to address such multifactorial Summary queries. In fact, bias of ascertainment, self-referral to physicians, exercise modality (e.g. greater or lesser Strenuous physical training in the healthy young cooling of core body temperature), perceived stress, adult (albeit not necessarily in children) does not probability of hormone abuse, weight loss or main- disrupt reproductive function significantly. This tenance, age of training onset and other unknown conclusion does not apply when training is accom- confounding factors vitiate definitive interpretation panied by deficient caloric intake, weight loss, aber- of cross-sectional data. rant eating habits, concomitant disease, high genetic In relation to the medical care of competitive ath- risk of a primary reproductive disorder, undue psy- letes, doctors should be cognizant that physical chosocial stress and/or abuse of alcohol, drugs or exertion alone is an unlikely proximate cause of hormonal agents. Additional clinical research is clinically significant endocrine changes. Thus, needed in behalf of the athletic and medical commun- physician-directed review of dietary habits, weight ities to clarify more subtle questions related to the loss, caloric intake, medical records, family history, impact of strenuous physical exertion on fecundity, systemic symptoms, physical signs and screening fertility, postpartum health and sexual function. laboratory tests can protect the athlete’s personal health. Thorough assessment is necessary to exclude Acknowledgments or identify remediable associated illness. Whether strenuous physical training affects This work was supported in part by the National sperm function or female fecundity (ability to con- Institutes of Health and the National Center for ceive) or fertility (carrying a healthy infant to full- Research Resources (Bethesda, Maryland, USA) via term delivery) has not been established rigorously. RR00585, AG23133 and DK60717. 76 chapter 6

References

Aloi, J.A., Bergendahl, M., Iranmanesh, A. women. New England Journal of Medicine year of endurance training. Journal of & Veldhuis, J.D. (1997) Pulsatile 312(21), 1349–1353. Applied Physiology 72(4), 1571–1580. intravenous gonadotropin-releasing Evans, W.S., Sollenberger, M.J., Booth, Jr., Urban, R.J., Evans, W.S., Rogol, A.D. et al. hormone administration averts fasting- R.A. et al. (1992) Contemporary aspects (1988) Contemporary aspects of discrete induced hypogonadotropism and of discrete peak detection algorithms. II. peak detection algorithms. I. The hypoandrogenemia in healthy, normal- The paradigm of the luteinizing paradigm of the luteinizing hormone weight men. Journal of Clinical hormone pulse signal in women. pulse signal in men. Endocrine Reviews 9, Endocrinology and Metabolism 82, Endocrine Reviews 13(1), 81–104. 3–37. 1543–1548. Giustina, A. & Veldhuis, J.D. (1998) Veldhuis, J.D. (1996) Neuroendocrine Bergendahl, M., Vance, M.L., Iranmanesh, Pathophysiology of the neuroregulation mechanisms mediating awakening of A., Thorner, M.O. & Veldhuis, J.D. (1996) of growth hormone secretion in the gonadotropic axis in puberty. Fasting as a metabolic stress paradigm experimental animals and the human. Pediatric Nephrology 10, 304–317. selectively amplifies cortisol secretory Endocrine Reviews 19(6), 717–797. Veldhuis, J.D. (1999) Male hypothalamic– burst mass and delays the time of Loucks, A.B. (2003) Energy availability, pituitary–gonadal axis. In: Reproductive maximal nyctohemeral cortisol not body fatness, regulates reproductive Endocrinology (Yen, S.S.C., Jaffe, R.B. & concentrations in healthy men. Journal of function in women. Exercise and Sports Barbieri, R.L., eds.). W.B. Saunders Co., Clinical Endocrinology and Metabolism Sciences Reviews 31(3), 144–148. Philadelphia: 622–631. 81(2), 692–699. MacConnie, S.E., Barkan, A.L., Lampman, Veldhuis, J.D., Evans, W.S., Demers, L.M. Bergendahl, M., Iranmanesh, A., Pastor, C., R.M., Schork, M.A. & Beitins, I.Z. (1986) et al. (1985) Altered neuroendocrine Evans, W.S. & Veldhuis, J.D. (2000) Decreased hypothalamic gonadotropin- regulation of gonadotropin secretion in Homeostatic joint amplification of releasing hormone secretion in male women distance runners. Journal of pulsatile and 24-hour rhythmic cortisol marathon runners. New England Journal Clinical Endocrinology and Metabolism 61, secretion by fasting stress in midluteal of Medicine 315, 411–417. 557–563. phase women: concurrent disruption Rogol, A.D., Veldhuis, J.D., Williams, Veldhuis, J.D., Yoshida, K. & Iranmanesh, of cortisol–growth hormone, cortisol– F.T. & Johnson, M.L. (1983) Pulsatile A. (1998) The effect of mental and luteinizing hormone, and cortisol– secretion of gonadotropins and prolactin metabolic stress on the female leptin synchrony. Journal of Clinical in male marathon runners: relation to reproductive system and female Endocrinology and Metabolism 85(11), the endogenous opiate system. Journal of reproductive hormones. In: Handbook of 4028–4035. Andrology 5, 21–27. Stress Medicine: an Organ System Bullen, B.A., Skrinar, G.S., Beitins, I.Z. et al. Rogol, A.D., Weltman, A., Weltman, J.Y. Approach (Hubbard, J. & Workman, E.A., (1985) Induction of menstrual disorders et al. (1992) Durability of the reproductive eds.). CRC Press, Boca Raton, FL: by strenuous exercise in untrained axis in eumenorrheic women during 1 115–140. Chapter 7

Growth Hormone Variants and Human Exercise

WESLEY C. HYMER, RICHARD E. GRINDELAND, BRADLEY C. NINDL AND WILLIAM J. KRAEMER

Introduction In this chapter we consider the biochemistry and physiological activities of the GH molecule and its Exercise stimulates release of human growth hor- variant forms. We examine the literature that shows mone (GH) from the anterior pituitary gland and elev- that GH variants in the circulation change after exer- ated concentrations of circulating GH ultimately cise. We also briefly review aspects of the cellular help contribute to the exercise-induced increases in biology of the pituitary with the hope that these muscle hypertrophy and fat breakdown as well as aspects help offer a fundamental basis for appreciat- other physiological responses. A review of the exer- ing how production of GH variants may relate to the cise/GH literature permits the generalization that it issue of what could happen after bouts of aerobic/ is largely the duration and intensity of the exercise resistance exercise. And finally we offer data that bout that regulates plasma concentrations of circu- strongly suggest the existence of a novel feedback lating GH. loop from the muscle to the pituitary. The authors It seems far less certain, however, that many are believe that this newly discovered feedback loop fully aware of the complexities underlying the GH might help explain one mechanism(s) underlying production ‘system’ in the pituitary gland itself. exercise-induced release of GH. Issues underlying these complexities often appear as studies published in specialty journals in which Measurement of human growth the major focus of the effort is not that of exercise- hormone induced release of GH. This situation has started to change. In today’s world, plasma concentrations of cir- For the reader primarily interested in exercise culating GH are almost always measured using physiology/medicine, the primary purpose of this immunoassay procedures. However, other detec- chapter is to bring the issue of GH system complex- tion systems exist and the method chosen to ity into sharper focus. What do we mean by ‘system measure GH in the blood is important. Before the complexity’? Elements of that complexity reside in introduction of GH immunoassays investigators signals delivered to the pituitary from other parts of usually relied on biological assays that often the body, for example muscles. Still other elements required the use of rats; certain of these bioassays reside in the GH producing cells themselves. The continue to be used in the authors’ laboratories final product of that complexity is the well-known (Roth et al. 1963; Hunter & Greenwood 1964). textbook form of GH molecule, as well as variants of Because they often yield interesting data that may at that molecule, that circulate in human blood. We times conflict with those obtained by immunoassay, contend that it is this ‘army of molecules’ that parti- we believe a review of such bioassay techniques is cipates in bringing about the physiological responses warranted. that have been ascribed to GH for many years. 77 78 chapter 7

Growth hormone bioassays: perspectives from the older literature Tests that were developed over 50 years ago re- flected the growing awareness at that time of the anabolic, lipolytic and diabetogenic actions of GH. A 1962 review (Papkoff & Li 1962) nicely sum- marized the essential details of some of these tests available at that time. In terms of body weight tests the following gener- alizations apply: large numbers of rats (10/dose level) are required; daily injections can be subcu- taneous or intraperitoneal; either intact or hypophy- sectomized rats are used; weight gain tests are relatively insensitive (50 µg·day–1 in the intact mat- ure female rat, 10 µg·day–1 in the immature hypox rat); indices of precision (calculated by dividing the standard deviation of the responses by the slope of the line) are > 0.2; and finally other hormones con- taminating a native GH preparation (e.g. thyroxine) may synergize with GH resulting in augmented weight gain. In spite of these disadvantages, the traditional rat weight gain bioassay, that measures the weight of hypox rats after subcutaneous admin- istration of GH for 10 days, is still required by regulators in the USA for individuals assessing the bioidentity and potency of GH preparations pro- duced by recombinant technologies. In terms of the rat tibial line GH bioassay pro- posed by Greenspan et al. (1949), a major advantage over the weight gain bioassay was the marked increase in assay sensitivity (a response can be detected with a total dose of 5 µg administered over a 4-day period). This assay quantifies the width of the uncalcified epiphyseal cartilage plate delineated from the silver nitrate stained calcified portions of the plate (Fig. 7.1). This test has been used by the authors of this chapter in many studies, including those investigating effects of exercise/bed rest on circulating GH. The list of biological effects of human GH increases steadily. As pointed out by Strasburger (1994), it has been known for a long time that GH is anabolic protein that promotes longitudinal bone growth. It is also lactogenic, has both insulin agon- istic and antagonistic properties, is lipolytic, stimu- lates ornithine decarboxylase in the liver, promotes growth hormone variants and human exercise 79

sodium and water retention, and modulates immune demand. A distinguishing structural feature is the system function among others. position of the cystine residues responsible for for- The IM-9 lymphocyte cell line and the 3T3-F422A mation of the large internal disulfide loop and adipocyte assays are more recent, noteworthy the smaller loop at the c-terminus. Also shown in examples of in vitro cell based biological assays. To Fig. 7.2 is an enzymatic cleavage site between our knowledge they have not been used to evaluate residues threonine-142 and tyrosine 143; a cleavage GH activities of plasma after exercise. Space limita- that results in a two-chain structure held together tions prohibit their consideration in this chapter. by these disulfide bridges. Generation of this two- chain form may result from the action of a mem- Growth hormone bioassays: brane bound protease during secretion from the new perspectives pituitary gland. On long-term storage of GH in solu- tion, deamidation of asparagine residues 149 and In an excellent study published by Roswall et al. 152, as well as loss of the first two residues at the (1996), careful comparisons are offered between two N-terminus, can occur. new GH bioassays developed in their laboratories Many structural variants of the GH molecule are and the hypox rat weight gain bioassay described in present in biological samples. In further studies section earlier. In order to fully appreciate the basis Roswall et al. (1996) generated two of these natur- for these new assays it is necessary to consider: (a) ally occurring variants using rhGH as their start- details of the structure of GH molecule produced ing preparation. One was the covalent dimer of by recombinant technology (recombinant human methionyl GH; the other was the 20 kDa trans- growth hormone, rhGH); (b) its structural variants criptional variant produced by deletion of residues and degraded forms; and (c) molecular interactions 32–46. These variants were used in studies later between these well-defined forms and the human described. growth hormone (hGH) receptor. How rGH molecules interact with membrane bound tissue GH receptors (GHRs) is obviously The primary structure of recombinant important for a more complete understanding of human growth hormone and its the importance and physiological consequences associated molecular landmarks of exercise-induced elevations in circulating GH. Studies by Cunningham and his colleagues ~ 15 The linear sequence of the 191 amino acid (22 kDa) years ago not only established the complete amino form of rhGH is shown in Fig. 7.2. This form, of acid sequence of membrane bound GHR, but also course, is identical to the native 22 kDa GH showed that the extracellular domain was essen- molecule that is synthesized in the pituitary gland tially identical to the glycosylated form of the recep- and released into the bloodstream on physiological tor isolated from human serum (Cunningham & Wells 1989; Cunningham et al. 1991). These invest- igators learned that one molecule of 22 kDa GH Fig. 7.1 (opposite) Rat epiphyseal cartilage plates prepared complexed with two molecules of the extracellular from hypophysectomized female rats after subcutaneous domain of GHR. At low GH concentrations receptor injection of saline or growth hormone (GH) standard for 4 binding occurs sequentially at two distinct sites on days prior to sacrifice. The total dose of GH administered is shown in the lower left corner of the panels. In this assay the hormone molecule. Figure 7.3 shows details cartilage plates are stained with silver nitrate and plate underlying receptor dimerization that will lead to widths measured with an ocular micrometer. Ten signal transduction in GH responsive tissues. measurements are taken across the plate and averaged. Understanding the basis of this detailed molecular µ Plates from control animals typically average 150 m physiology enabled Roswall et al. (1996) to develop and those from GH injected rats increase to > 200 µm in dose-dependant fashion. The assay is usually done in two different types of GH bioassays. One, termed double-blind fashion. Photographs kindly provided by high performance receptor binding chromatography Dr. Scott Gordon. (HPRBC), compares the ability of a test sample of 80 chapter 7

Denotes tryptic cleavage site 2-chain cleavage site

Denotes theoretical peptide number Oxidized methionine

des-Phe1 Pro2–hGH cleavage site Deamidated asparagine

Fig. 7.2 The linear sequence of recombinant human growth hormone (rhGH). The diagram indicates: (a) tryptic cleavage sites; (b) theoretical peptide number (T1–T21); (c) the two-chain enzymatic cleavage site; (d) methionine residues susceptible to oxidation; and (e) asparagines residues susceptible to deamidation. Residues 32–46, marked in black, are deleted in the 20 kDa splice variant of GH. (Adapted from Roswall et al. 1996.)

GH with that of a rhGH standard to form a stable rGH binding protein (BP) to assess functional activ- 2 : 1 receptor/rGH complex with soluble GHR. Non- ity of the GH-containing preparation. The IFA has denaturing size exclusion chromatography is used been used by to measure circulating levels of GH to analyze the resulting complex. Strasburger and after exercise (Nindl et al. 2000) colleagues have recently developed the immuno- In the second assay described by Roswall et al. functional enzyme linked immunoassay (IFA) that (1996), termed the cell proliferation assay (CP), cells is based on these earlier findings of Cunningham from a mouse myeloid leukemia cell line (FDC for and his colleagues (Strasburger et al. 1996). This assay ~ 20 h prior to -P1) are transfected with full length uses a GH monoclonal antibody and a biotinylated receptor and subsequently incubated with test growth hormone variants and human exercise 81

hGH

Receptor

Signal transduction

Sequential binding Sequential binding *Site 1 *Site 1 *Site 2

Fig. 7.3 Idealized model of a 22 kDa human growth hormone (hGH) molecule in plasma interacting with receptors on the plasma membrane of a target cell. Proper signal transduction requires receptor dimerization via two binding sites on the hormone molecule. samples measuring 3H-thymidine uptake into DNA parameters in human plasma after exercise. Data in as a measure of biological activity. A similar strategy Table 7.1, reproduced from the Roswall et al. (1996), was used by Wada et al. (1998) to measure GH show full activity and good correspondence between activity of certain variants using Ba/F3-hGHR cells. results in some samples (e.g. deamidated rGH and These cell based assays measure a response (i.e. oxidized rGH); poor activity with preparations of DNA synthesis) that is several steps distal to recep- dimer or trypsin treated rhGH; and ‘super-potency’ tor dimerization. Roswall et al. (1996) suggest that in the rat bioassay when the two-chain rhGH vari- this fact puts the assay: ‘several steps closer to the ant is tested. As indicated by Roswall et al. (1996) in vivo biological response’ (p. 36). and others, increased biological activity two-chain GH has been reported previously. Assay comparisons Even though the cell assays described by Roswall et al. (1996) and Wada et al. (1998) have yet to be used A comparison between the activities of genetic and for studies using blood samples collected before and chemical rGH variants assessed by the rat weight after exercise stress, it seems quite likely that they gain, HPRBC and CP assays are informative and will in the near term become important for under- important to consider before assessing similar standing GH activity.

Table 7.1 Biological activities of genetic and chemical variants of Sample description Rat bioassay HPRBC assay CP assay recombinant human growth = = hormone (rhGH) by the rat weight Deamidated rhGH 0.88 (n 1) 1.01 (n 3) 0.86 = = gain assay; the high performance Two-chain rhGH 1.78 (n 2) 0.75 (n 2) 0.77 = = receptor binding chromatography Aspartic acid mutant rhGH N149D 0.80 (n 1) 1.00 (n 4) 0.87 = = (HPRBC) assay and the cell Aspartic acid mutant rhGH N152D 1.22 (n 1) 0.95 (n 2) a = = proliferation (CP) assay. (Data Aspartic acid mutant rhGH 0.85 (n 1) 0.96 (n 4) 0.93 + taken from Roswall et al. 1996.) N149D N152D Des-Phe1Pro2-rhGH 0.46 (n = 1) 0.60 (n = 4) 0.94 Oxidized rhGH 0.79 (n = 4) 0.98 (n = 2) 0.77 Met-hGH covalent dimer 0.10 (n = 1) 0.07 (n = 5) 0.36 Aged rhGH 0.73 (n = 1) 0.96 (n = 3) a Trypsin-treated rhGH 0.47 (n = 1) 0.30 (n = 1) a

See Roswall et al. (1996) for a discussion of the principles underlying these data. Note that the two-chain variant, relative to the 22 kDa form, is highly potent in the rat weight gain assay and less so in the other in vitro biological assays. 82 chapter 7

Table 7.2 Estimated proportions of Proportion of total GH growth hormone (GH) variant forms GH form (%) in human plasma 15 min after secretion. (From Baumman 1991b.) Monomeric 22 kDa total 43.0 22 kDa free 21.0 22 kDa in high affinity complex 20.0 22 kDa in low affinity complex 2.0 20 kDa total 8.0 20 kDa free 5.5 20 kDa bound in high affinity complex 0.5 20 kDa bound in low affinity complex 2.0 Acidic GH (desamido- and acyl-GH) total 5.0 Acidic GH bound fractions Unknown Dimeric 22 kDa non-covalent dimer total 14.0 22 kDa disulfide dimer total 6.0 22 kDa dimers bound fraction Unknown 20 kDa non-covalent dimer total 5.0 20 kDa disulfide dimer total 2.0 20 kDa dimers bound fraction Unknown Acidic GH non-covalent dimer total 1.5 Acidic GH disulfide dimer total 0.5 Acidic dimers bound fractions Unknown Trimeric to pentameric 22 kDa non-covalent oligomers total 7.0 22 kDa disulfide oligomers total 3.0 22 kDa oligomers bound fraction Unknown 20 kDa non-covalent oligomers total 1.0 20 kDa disulfide oligomers total 0.5 20 kDa oligomers bound fraction Unknown Acidic GH oligomers (non-covalent and S–S) total 1.0 Acidic GH oligomers bound fraction Unknown Non-S–S-linked covalent oligomers total 1.0 Fragments 16 kDa, 12 kDa, and 30 kDa immunoreactive fragments Variable

The wide array of growth hormone Baumann paper (Baumann 1991b). It shows the forms in the blood estimated proportions of GH variant forms in the plasma 15 min after a secretory episode. Many of It has been suggested by Baumann (1991b) that these studies were done before techniques of recom- there may be as many as 100 different forms of GH binant technology became widely available. Not present in the human circulation. The concept that surprisingly, this type of information therefore was numerous molecular forms of the hormone arise obtained using traditional biochemical procedures. from post-transcriptional/translational modifica- The literature is rich with observations concern- tion of the single pituitary GH-N gene is certainly ing the biochemical nature of GH variants. A brief not new. Pioneering work from the laboratories of but non-comprehensive review of this literature is Lewis, Sinha, Kostyo and Baumann to name but important for the innate understanding of the het- a few, provide the foundation for the informa- erogeneity and complexity of the GH mechanisms tion offered in Table 7.2 that is taken from a 1991 potentially involved with exercise responses and growth hormone variants and human exercise 83

adaptations. The fact that plasma GH immunoreact- A recent review by the Lewis and Sinha group ivity consists of several molecular weight species summarizes the properties of five variant forms of that can be separated by size exclusion column chro- GH (Lewis et al. 2000). Two of these are the short and matography has been known for > 30 years. In the larger peptides generated from proteolytic cleavage past it was useful to categorize the three main between residues 43 and 44 of the GH molecule (see isomers (variants) of GH as little, big or big–big; a Fig. 7.2). Their evidence favors the concept that designation based upon their elution positions from the short peptide (GH [1–43]) potentiates the physi- the column. The physical nature of these size vari- ological effects of insulin; the larger peptide (GH ants is not nearly as well defined as those in studies [44–191]) has anti-insulin properties. In fact these using GH made by recombinant means. Neverthe- authors indicate that: ‘. . . we believe that this (larger less, the careful studies by the Baumann (Baumann peptide) is the long sought after diabetogenic sub- 1991a, 1991b, 1999; Baumann et al. 1994) and Lewis stance of the pituitary gland’ (p. 58). groups (Lewis et al. 2000), among others, that have A ~ 3 kDa peptide, isolated from both human attempted to characterize the big and big–big post-mortem pituitary tissue and human plasma, is variants, permit the conclusion that these variants active in the rat tibial line bioassay (Hymer et al. represent an oligomeric series. Similar oligomers in 2000). This peptide is not a fragment of the GH extracts of human pituitary tissue further support molecule. The relationship of this finding to variant this view. Most investigators therefore believe that forms described by Baumann is unclear (Baumann aggregation states up to at least pentameric hGH 1999). Limited amino acid sequence data, including exist and the distinction of big and big–big GH residues 9–25 in the middle of the peptide, indicate is arbitrary. The authors prefer to categorize the that this peptide is not a breakdown product of the oligomers in terms of their apparent molecular mass GH molecule. Most interesting is the finding that based on elution profiles following sephadex chro- many of these residues are non-polar and bear strik- matography. In addition to these oligomeric size ing similarity to the c peptide that is contained in the variants charge variants of the GH molecule are proinsulin molecule. Similar to the c peptide, this known. These are thought to be reflected in acety- human pituitary tibial peptide, obviously has bio- lated, deamidated or cleaved GH (see Fig. 7.2). logical activity. Unpublished data from one of our The careful study by Stolar et al. (1984) also laboratories (R.G.) shows that the rat pituitary also showed that a majority of the big and big–big GH contains a small peptide that is active in the tibial variants converts to little hGH (22 kDa) during bioassay. extraction and storage (e.g exposure to 4 mol potas- sium thiocynate [KSCN] and two freeze–thaw Variant forms of growth hormone: cycles result in ~ 70% conversion to little GH). The an emerging data base variants surviving this harsh treatment migrate as distinct species with apparent molecular weights of In spite of the fact that it has been known for many 45, 62, 80 and 110 kDa. These latter forms covert years that exercise is a potent stimulator of circulat- quantitatively (and mainly) to little GH after sulfhy- ing GH, it is only recently that the topic of possible dryl reduction. Acidic GH comprises a smaller com- exercise-induced generation of circulating GH vari- ponent. The 20 kDa variant may tend to dimerize ant forms been addressed at all (Nindl et al. 2003). In preferentially. this section we consider some of our prior informa- What is known about the biological activity of tion in the context of the background given previ- these oligomeric species? In general, the big forms ously. To summarize and analyze this information (dimer) are thought to have reduced activity in in a logical way requires an appreciation of the radioreceptor and rodent based assays. However, in following types of variables in the analysis: human the IFA, Strasburger et al. (1996) report that dimers, subject choice; exercise type (intensity/duration); on a molar basis, have slightly higher reactivity in GH assay type; methodology used to isolate the vari- the IFA (110%) relative to the 22 kDa monomer. ant forms; and special treatments of blood samples. 84 chapter 7

Table 7.3 Summary of recently published studies in which different types of growth hormone (GH) assays have been used to measure concentrations of GH in human plasma after aerobic or resistance exercise.

GH GH immunoassay bioassay Chemical Reference Exercise type/duration IRMA Poly Tibial IFA reduction

o Rubin et al. 2003 Treadmill: 60, 75, 90, 100% V 2max XXX for 10, 10, 5 & 2 min, respectively Hymer et al. 2001 Squats: 6 sets of 10-RM X X X X X o Wallace et al. 2001 Cycle: 80% V 2max for 20 min X X Nindl et al. 2001 Resistance: high volume, multiset X X X Bigbee et al. 2000 Treadmill: 27 m·min–1 for 15 min X X McCall et al. 2000 Vibration: stimulus-muscle X X afferents for 10 min Nindl et al. 2000 Squats: 6 sets of 10-RM X X McCall et al. 1997, 1999 Isometric plantar flexion: 30, 80, X X 100% MVC

Poly: assay using a polyclonal antiserum. Tibial: rat tibial line bioassay. Chemical reduction: those studies in which a reducing agent has been added to the plasma sample before assay. IFA, immunofunctional assay (as described by Strasburger et al. 1996); IRMA, immunoradiometric assay; RM, repetition maximum.

Table 7.3 provides a summary by subcategorizing GH increased and those of recombinant/pituitary studies (only human) according to our arbitrary GH decreased. Wallace et al. (2001) attributed the requirement that the blood sample has been studied increase in non-22 kDa isoforms to slower disap- by at least two procedures for the purposes of fur- pearance rates of 20 kDa and perhaps non-22 kDa ther understanding exercise-induced rises in circu- GH isoforms. Collectively, Wallace’s findings lating GH in terms of GH molecular heterogeneity. demonstrate that the proportion of GH isoform Only the Hymer study (Hymer et al. 2001) used frac- changed across acute exercise and into recovery. tionated plasma to measure GH variants; all the rest Although the 22 kDa was the predominant isoform in Table 7.3 used only neat plasma. detected at peak concentrations, isoforms of GH Data from studies to date thus far show that other than 22 kDa increased during the post- exercise can modify either the activity or molecular exercise period. These results suggest that the pro- character of GH in the circulation. Wallace et al. portion of 20 kDa, 17 kDa and possibly other (2001) utilized seven different assays to measure non-22 kDa isoforms (dimers, oligomers and GH GH in 17 aerobically trained men before and after bound to serum proteins) increase after exercise. o 20 min of cycle ergometry at 80% V 2max to further The authors postulated that the increase in the pro- characterize the response of GH molecular isoforms portion of isoforms other than 22 kDa after exercise to exercise. Serum was assayed with antibodies may be attributed to differential pituitary isoform specific for total, pituitary, 22 kDa, recombinant, secretion, the appearance of isoforms from non- non-22 kDa, 20 kDa and immunofunctional (IF) pituitary sources, generation of fragments, dimers GH. Salient findings from this study were: (a) all and oligomers in the circulation, and differences in forms of GH increased during and at the end of clearance rates of the different isoforms. The exercise; (b) 22 kDa GH was the predominant authors also speculated that the biological conse- isoform (73%) at the cessation of exercise; (c) the quences of their findings might potentially reside ratios of non-22 kDa/total GH and 20 kDa/total in enhanced diabetogenic effects of smaller GH growth hormone variants and human exercise 85

isoforms, which may serve to prevent post-exercise made between the IF versus IR GH concentrations hypoglycemia. in men and women before and after acute resistance Extending on Wallace’s work (Wallace et al. 2001), exercise (i.e. six sets of 10-RM squats separated by we (Hymer, Kraemer and Nindl) next conducted a 2-min rest periods). IF GH concentrations were study in which we fractionated human plasma in 35 determined by an enzyme linked immunosorbent women before and after acute resistance exercise assay (ELISA) purchased from Diagnostics Systems (six sets of 10-repetition maximum [RM] squats, Laboratories (DSL, Webster, TX), which was based separated by 2-min rest periods) using size exclu- on Strasburger’s work (Strasburger et al. 1996), and sion chromatography into three size classes (Hymer IR GH concentrations were determined by a mono- et al. 2001). Fraction A contained molecules > 60 kDa clonal IRMA purchased from Nichols (San Juan (presumably oligomers and/or monomeric GH Capistrano, CA). In this study, both men and bound to receptor); fraction B contained molecules women demonstrated similar increases for IR (men: 30–60 kDa (presumably homodimers and hetero- 1.47 vs. 25.0 ng·mL–1; women: 4.0 vs. 25.4 ng·mL–1) dimers); and fraction C contained GH molecules and IF (men: 0.55 vs. 11.7 ng·mL–1; women: 1.94 vs. < 30 kDa (presumably a mixture of 22, 20, 16, 12, and 10.4 ng·mL–1) GH following exercise. However, 5 kDa forms). All samples were then assayed using post-exercise IF GH was significantly less than IR the Diagnostic Systems Laboratory (DSL) IFA, the GH for both men and women. The ratio of IR/IF Nichols radioimmunometric assay (IRMA) and the after exercise was ~ 2 and similar for both men and National Institutes of Diabetes and Digestive and women. The correlation between post-exercise IR Kidney Diseases radioimmunoassay (NIDDK RIA). and IF GH was r = 0.83. This study initially indic- Additionally, we assayed all samples before and ated that about half of the GH isoforms measured after glutathione (GSH) treatment in order to deter- by the Nichols IRMA released after exercise did mine the effects of chemical reduction of disulfide not possess intact sites 1 and 2 required for receptor linked bonds. Recovered immunoreactivities were dimerization, thus suggesting biological inactivity. 4–11% in fraction A, 22–45% in fraction B and A following study considered the fact that GH is 44–72% in fraction C. Significant exercise-induced released in an episodic, pulsatile manner. IF GH was increases were observed for the lower molecular measured in 10 men who underwent two overnight weight GH moieties (30–60 kDa and < 30 kDa iso- blood draws with sampling every 10 min from 1700 forms), but not for the higher molecular weight GH to 0600 h. The overnight serial sampling was per- moieties (> 60 kDa). Another important finding was formed in both a control and an acute heavy resist- that chemical reduction of the post-exercise samples ance exercise condition (Nindl et al. 2001). For increased immunoassayable GH as measured by the the exercise condition, subjects performed a high- Nichols and NIDDK assays more than pre-exercise volume, multiset resistance exercise bout from 1500 samples, suggesting that exercise may specifically to 1700 h. IF GH was compared to the Nichols IRMA increase the release of disulfide-linked hormone and NIDDK’s polyclonal RIA. The Pulsar peak molecules and/or fragments. From these data, the detection system was used to evaluate the pulsatil- most important effect of acute resistance exercise ity profile characteristics of GH. Even though the appears to be on dimeric hormone. Because com- results from all three immunoassays were highly plexes of GH and BPs have longer half-lives than correlated (correlations ranged from 0.85 to 0.95), free GH, it is possible that dimeric GH might also the Nichols IRMA again yielded higher mean GH have a longer half-life. Therefore, the net effect of concentrations than did IF GH (3.98 vs. 1.83 ng·mL–1, the increase in GH isoforms within this molecular respectively). The results from the Nichols IRMA weight range would be to prolong the biological also yielded higher pulse amplitudes compared to activities of these forms after exercise. IF GH (8.0 vs. 4.63 ng·mL–1, respectively). In a study by Nindl et al. (2000) the first data com- The consistent finding in these studies (Nindl paring the effects of exercise on IF versus immuno- et al. 2000, 2001, 2003) was that IF GH measured reactive (IR) GH was presented. Comparisons were from the same sample was approximately one half 86 chapter 7

that of the Nichols IRMA, one of the most widely GH binding to receptors and in vitro bioactivity via used GH assays in clinical practice in the USA. Since competition for ligand (Strasburger et al. 1996). If it the IF assay purports only to measure biologically is true that GH complexed to BP is too large a mole- active forms of GH (i.e. only those forms of GH cule to traverse the capillary endothelium in order capable of inducing receptor dimerization are trans- to bind to cellular receptors, the lack of detection lated to an assay signal), the additional GH isoforms of the GH complexed in the IFA provides further detected by the Nichols IRMA are likely fragments support for the functional selectivity of the IFA. whose potential biological actions are not mediated Rubin et al. (2003) also compared the Nichols by the GHR. It has been reported that the GH frag- IR versus the DSL IFA in six endurance-trained men ment 44–191 is detectable in substantial levels in during intermittent treadmill running at progress- o human serum and may even antagonize GH action ively increasing intensities (60% V 2 max for 10 min, (Rowlinson et al. 1996). Since this fragment lacks 75% for 10 min, 90% for 10 min and 100% for 2 min). part of the N-terminus, it is unlikely to be detected Samples were assayed before and after the addition by the IF assay; however, the fragment could be of glutathione (GSH; 10 mmol for 18 h at room detected by the IRMA, depending on the targeted temperature) in order to break disulfide bonds epitopes. Alternatively, the additional GH isoforms between possible oligomeric GH complexes. For the detected by the Nichols IRMA could also be high IRMA, GH was elevated after the 75% exercise molecular weight variant forms of GH (Baumann intensity and remained elevated through 30 min of 1991a; Lewis et al. 2000). post-exercise. After adding GSH, the IRMA indic- Our findings conclusively show that at least some ated elevations in GH as early as 60% exercise intens- of the molecules released during secretory bursts ity and remained elevated 45 min into recovery. At are able to dimerize GHRs. In that sense, these 75%, the GSH assay run was higher than the non- molecules are biologically active. On the other hand, GSH assay run. With the IFA, GH was elevated at our data also demonstrated that GH isoforms are 60% in the non-GSH conditions, whereas the GSH released, both in exercise and non-exercise con- assay run indicated elevations at 75%. Both GSH and ditions, that are not capable of initiating signal non-GSH conditions remained elevated through transduction through the GHR. Based on the high 30 min of recovery. These data indicate that the correlations among the immunoassays and the addition of GSH to serum samples prior to assay via similar detection of the number of peaks and inter- an IRMA may break existing disulfide bonds aggreg- peak intervals, it appears that the immunoassays ated to GH molecules, thus altering the apparent report qualitatively comparable pictures of the GH assays signal to reveal total GH. response. The quantitative differences among the immunoassays have yet to be fully explained, but The complexity of the anterior pituitary are likely due to the existence of various molecular cell system isoforms. Other factors that contribute to the differ- ences in GH measurement could include the assay The GH-producing cell system of the anterior pitu- equilibrium conditions, buffer, tracer and standard itary gland is tremendously complex. How might used (Wood 2001). knowledge of that complexity help in unraveling It is important to consider the impact of BPs in the mysteries and mechanisms underlying the relation- IFA (Strasburger et al. 1996; Nindl et al. 2001). The ships between exercise and GH variants? Human IFA uses an rGHBP to bind site 1. One could infer pituitary tissue obtained after surgery or death is that a GH molecule already complexed to a GHBP difficult to obtain, so the rat pituitary gland has been would not be detected in this assay system, as site 1 the tissue of choice to study biology of the GH cell. would not be freely accessible. Also, a GH–BP com- Space limitations do not permit full review of this plex might be configured such that site 2 is not ex- topic. Instead, results from highly selected studies posed to the monoclonal antibody (mAb7B11). It has are summarized here, in the format of a few sen- been reported that the high-affinity GHBP inhibits tences for each point considered. growth hormone variants and human exercise 87

Table 7.4 The effect of chemical µ reduction on concentrations of rat GH ( g) growth hormone (GH) variants β +β contained in alkaline extracts Apparent mol. wt. (kDa) – ME ME Fold increase separated on non-reducing sodium < + + dodecylsulfate polyacrylamide gel 22 4.7 0.9 6.1 0.7 1.3 + + electrophoresis (SDS-PAGE) rat 22–57 1.8 0.2 3.2 0.5 1.8 + + pituitary prior to immunoassay. 57–77 2.1 0.2 8.1 0.4 3.9 + + (From Farrington & Hymer 1990.) 77–100 1.2 0.2 4.7 0.5 3.9 100–150 0.9 + 0.15 5.1 + 0.3 5.6 < 150 0.9 + 0.3 4.0 + 0.4 4.4

Hormone contained in different regions of the gel were eluted and reduced with mercaptoethanol prior to GH using a polyclonal antibody. Recovery of GH from the gel averaged 103% (n = 3 experiments).

1 Human pituitary glands weigh ~ 350 mg; those and GH containing granules isolated from the post- of rats 8–12 mg. Human pituitary glands contain mortem human pituitary also yield subpopulations 4–8 mg GH/gland; rats ~ 200 µg GH/gland. GH as well. Preliminary evidence indicates that the accounts for ~ 80% of the total hormone in the gland. more rapidly migrating particles are rich in tibial 2 Quantitative extraction of GH requires homogen- line bioactivity and relatively poor in IR GH. ization at alkaline pH (9–10). Non-reducing sodium 5 At least two types of rat pituitary GH cells can be dodecylsulfate polyacrylamide gel electrophoresis routinely separated by density gradient centrifuga- (SDS-PAGE) of pituitary extracts shows multiple tion. GH that is rich in tibial line activity is preferen- GH forms with apparent molecular weights of tially released from the more dense cell; and this 14–88 kDa. After chemical reduction there is a four hormone is oligomeric (Farrington & Hymer 1990). to sixfold increase in the GH immunoreactivity in 6 Implantation of GH cells into the cerebral vent- those forms with mass > 50 kDa (Table 7.4). High ricles of hypophysectomized rats show that receip- molecular weight forms of GH are aggregates that ents gain weight and have longer tibial, femoral are linked by disulfide bridges resulting from and pelvic bones (Weiss et al. 1978). Intraventricu- antiparallel alignment of cystine residues between lar implantation of the denser GH cells not only two GH monomers (Lewis et al. 1975). GH aggreg- increases body weight, it also increases width of the ates are released from the pituitary and are present epiphyseal plate and weight of the gastrocenimus in human plasma. muscle (Grindeland & Hymer unpublished). 3 Aggregates of GH are packaged in ~ 300 µ dia- Collectively these findings support the contention meter secretion granules. These granules contain that not only is there a molecular basis underlying biologically active GH (Hymer & McShan 1963). GH variants, but that there is an underlying cellular According to Dannies (1999) ‘not a great deal’ is biology basis as well. known about molecular packaging of the hor- mone into granules and, ‘...how cells concentrate Tibial line assay and exercise responses hormones is a major unanswered question in endocrinology’ (p. 3). In some hormone-packaging The issue of the dichotomy between measurements cell systems there is good evidence that different of GH concentrations made by tibial line bioassay secretory granule proteins are packaged in separate and those made by immunoassay is important granules in the same cell. It is probable that hetero- to consider in future experiments addressing the geneity in GH packaging is directly related to the relationship between GH variants and exercise. It is issue of GH variants and exercise. important to consider GH variants in terms of those 4 Subpopulations of rat GH containing granules molecular forms which react with high affinity anti- are separable by continuous flow electrophoresis; sera to the 22 kDa ‘native’, textbook form of GH 88 chapter 7

(immunoassayable GH; iGH) or to those form(s) Investigations by one of us (R.G.) revealed that the that do not react with such antisera, but which elicit rat responded to these stimuli by secretion of GH in growth as revealed by the tibial line bioassay (bioas- form(s) not recognized by antisera to the 22 kDa rat sayable GH; bGH). It is possible that the pituitary GH molecule (Ellis & Grindeland 1974). However, secretes variant forms of GH which have other func- these circulating immunologically unreactive forms tions (e.g. lipolytic activity), but literature on this promoted significant growth of the assay rat (Ellis & possibility is somewhat limited. Grindeland 1974). At the moment the tibial assay appears to be the These types of results permit the general, yet still assay of choice if one wishes to know the GH ‘status’ speculative, conclusion that while there is no obvious of the subject. In spite of the fact that the assay is relationship between iGH and bGH in the rat, vari- labor intensive, costly and lengthy it clearly offers ations in bioactive/immunoactive human GH information that is simply not obtainable any other concentrations in biological samples tend to change way. There is absolutely no question that today’s in the same direction. However, because titers of scientist or clinician is comfortable and familiar human iGH and bGH are not directly proportional with data that report plasma GH concentrations we believe that one should not use iGH measure- measured by immunoassay. However, apparent ments as an index of total circulating GH. concentrations of bGH reported in the literature Almost coincidental with these early studies on are often hundreds or even thousands of nanogram rat and human bGH was the finding that plasmin, a per milliliter! Why is this so? Because this bioassay protease, was active on subprimate GH (rat, bovine) measures the biological activity and not nanograms of and either reduced or abolished the immunological purified hormone. In this context it is important to activity of the 22 kDa hormone (Ellis et al. 1968). realize that purified GH from a number of different However, this treatment yielded peptides with species (human, bovine, murine), and their bGHs, normal or even enhanced biological activity. Other yield parallel dose response curves in the tibial laboratories have shown that human iGH, treated bioassay. Similar tibial growth curve responses with human plasmin, does not appear to lose elicited by these hormone preparations enable one immunological activity but to gain in biopotency to express biological potency in terms of a standard (Singh et al. 1974; Lewis et al. 1975; Nguyen et al. 22 kDa hormone preparation. 1981). Clearly the immunological/biological ratio of In any event it is important that, at the very least, the 22 kDa GH molecule can be significantly altered some appreciation of the difficulty and importance after enzyme treatment. of establishing the most meaningful quantification of plasma GH in a given physiological context of Human bed rest studies the subject is determined. Moreover any possible skepticism regarding measurement of bGH by tibial Humans performing a moderate exercise (i.e. plan- assay may be alleviated by the following two exam- tar flexion) lasting several minutes show a one to ples of the dichotomy issue between bGH and iGH. twofold increase in plasma bGH but no effect on iGH when done under the usual 1 G conditions. Dichotomy between bioassayable However, when subjected to absolute, head-down growth hormone and immunoassayable bed rest the identical exercise regimen does not growth hormone evoke an increase in bGH secretion and has no effect on iGH release (McCall et al. 1997). Interestingly, a The rat has played a major role in the evolution of few days after bed rest the increased bGH secretion our current understanding of the iGH–bGH dicho- in response to exercise returns (Table 7.5). tomy. For example, many investigators report that What bearing might these data from have on stimuli (e.g. cold exposure, hypoglycemia, exercise) GH variants and human exercise? In the exercise that elicit increased plasma iGH concentrations in physiology literature metabolic regulators are usu- humans have no effect on circulating iGH in the rat. ally invoked as a dominant mechanism by which growth hormone variants and human exercise 89

Table 7.5 Plasma immunoassayable and bioassayable stimulation has no effect on concentrations of either growth hormone concentrations. Pre- and post-exercise plasma or pituitary bGH or iGH. –1 during bed rest (ng·mL ). (Adapted from McCall et al. On the other hand, if the proximal end of the 1997.) severed nerve of a fast twitch muscle is stimulated, Day Hormone Pre Post there is a massive (one to twofold) increase in plasma bGH within 5 min after stimulation! Sig- Before bed rest nificantly, this increase in plasma bGH is mirrored –13/12 bGH 2146 0 3565* by a large decrease in pituitary gland concentrations iGH 5.3 4.7 –8/7 bGH 2162 0 4161* of bGH. However, there is no effect on iGH con- iGH 5.0 5.3 centrations in either plasma or the pituitary gland. Equally interesting is the finding that stimulation of During bed rest 2/3 bGH 2350 2203 the proximal end of the soleus nerve results in iGH 5.1 5.8 reduced plasma concentrations of bGH. This observa- 8/9 bGH 2433 2105 tion implies specificity of muscle groups in this iGH 4.8 5.2 afferent pathway. 13/14 bGH 2594 2085 These results are interesting from two points of iGH 4.7 5.2 view. First, they argue that there is a neural control After bed rest component to the pituitary GH system in addition +2/3 bGH 1807 2379 iGH 2.0 4.5 to a metabolic component. Secondly, we would +10/11 bGH 1881 0 4160* argue that these nerve stimulation studies provide iGH 4.9 5.3 insight into the physiological significance of bGH. If one assumes that one of GH’s most critical functions < *p 0.05 between pre- and post-exercise values. Growth is assuring a constant glucose supply to the heart hormone measured by immunoassay did not differ at any time or condition between pre- and post-exercise and brain, the results appear to make good sense. samples. bGH, bioassayable growth hormone; iGH, Massive GH release in response to high metabolic immunoassayable growth hormone. demand (e.g. fasting, hypoglycemia or cold expos- ure) and release of GH in response to activation of a quiescent muscle of locomotion would both offer increased muscle activity increases iGH secretion. It mechanisms for increased carbohydrate uptake by is interesting to find that none of the commonly tissues. These are protective mechanisms for the cited metabolic factors (e.g. lactate, plasma glucose) organism. appear to explain reduced plasma concentrations Of course it is well established that human iGH is of bGH. This obvious inconsistency led one of the secreted in response to exercise, but this response authors (R.G.) to ask the following question: ‘Is begins only after 15 or 20 min of exercise. We pro- there a neural mechanism that controls GH secre- pose that during quiescence the soleus and other tion?’ The answer seems to be yes; and it is relevant postural muscles, which are perhaps 80% active at to exercise physiologists. rest, signal the pituitary via muscle afferents to decrease bGH secretion and thereby permit non- Muscle afferents regulate release of neural and non-cardiac tissues to use glucose. As bioassayable growth hormone in rats the muscles of locomotion become active the mus- cles would signal the pituitary gland to secrete Initial studies used animals in which nerves serving bGH. The net effect would presumably be to inhibit hind-limb muscles were severed. When the distal utilization of glucose by active muscle and provide ends are stimulated electrically for 15 min, a treat- an alternate energy supply by mobilizing fatty acids ment that generates a pattern resembling that of a from the fat depots. rat walking at 1.5 miles per hour (2.4 km per hour) In Fig. 7.4 we offer a model, adapted from our (Gosselink et al. 1998, 2000; McCall et al. 2000), the published study (McCall et al. 2001), that closes a 90 chapter 7

to the pituitary. One of the more recent is that of (Circulation) Bone growth Paden et al. (1994) that describe ‘...a surprisingly BGH Muscle growth? extensive innervation of the anterior lobe of the Chronic Other functions? pituitary’ (p. 503). It is interesting that they are also unloading frequently associated with blood vessels and do not Anterior pituitary have the appearance of vasomotor fibers. Their dis- − + tribution is uneven and they appear to contact only a subset of glandular cells (GH/adrenocorticotropic SS (−) GRF (+) hormone; ACTH). Hypothalamus Acute versus chronic resistance exercise and bioassayable growth hormone In a recent study we (Kraemer, Hymer and Nindl) Spinal cord evaluated the effects of an acute resistance exercise bout (i.e. six sets of 10-RM squats separated by 2 min of rest) on bGH before and after 6 months Tibialis Soleus Gastrocnemius anterior of periodized resistance training in young, healthy women. Results from this study indicated that while Low threshold afferent fibers, activated by: acute resistance exercise did not alter circulating • Electrical stimulation bGH, 6 months of chronic resistance training clearly • Exercise • Vibration potentiated bGH concentrations (Fig. 7.5). These results suggest that one of the benefits of chronic Fig. 7.4 Model of proposed muscle afferent–pituitary resistance training is to increase the biological activ- feedback axis that is postulated to regulate release of ity of circulating GH. This new finding potentially bioactive growth hormone (BGH) from the anterior pituitary gland. BGH is defined as those form(s) of the represents one of the mechanisms by which muscle hormone molecule that stimulate widening of the and bone can benefit from resistance training. epiphyseal cartilage plate (see Fig. 7.1). This model is suggested by the studies of McCall et al. (2001) in which stimulation of afferents from ankle flexors of the rat (e.g. Summary tibialis anterior) or the entire posterior compartment of ankle extensors (e.g. soleus and gastrocnemius) stimulates The main points we have tried to emphasize in this release of BGH but inhibits BGH when only afferents from chapter can be summarized in the following way. the soleus muscle are stimulated. Chronic unloading 1 GH molecules are heterogeneous. In this chapter we inhibits the exercise-induced increase of plasma BGH. define and consider GH heterogeneity in different Afferents may activate pituitary GH cells directly. ways. These include: (a) size and charge variants of Possible physiological functions of BGH, other than stimulation of the tibial epiphyseal growth plate, have not the molecule(s) resulting from the single pituitary been identified. GRF, growth hormone releasing factor; GH gene; (b) activities of the molecule(s) in terms SS, somatostatin. (Adapted from McCall et al. 2001.) of the biological (in vivo) versus immunological (in vitro) signals they generate; and (c) cells in the pitu- itary that produce and release GH molecules. 2 Either aerobic or resistance exercise can result in differ- muscle/neural feedback loop to the pituitary GH ential release of GH variants into the circulation. To date system. In Fig. 7.4 the neural inputs are received by there are only a handful of studies that address the hypothalamic neurons. On the other hand, it is con- issue of GH variants after human exercise. Thus far ceivable that neural inputs might be received by the they support the idea that release of oligomeric anterior pituitary gland directly. There are a few forms of the 22 kDa GH monomer is increased after reports in the literature that describe nerve inputs exercise. The activities of circulating GH, measured growth hormone variants and human exercise 91 ) )

–1 6000 –1 6000 5000 5000 4000 4000 RT RT 3000 3000 Control Control 2000 2000 1000 1000 0 0 Growth hormone (ng·mL Pre-exercise Post-exercise Growth hormone (ng·mL Pre-exercise Post-exercise Pretraining treatment groups Post-training treatment groups (a)Tibial line growth hormone (b) Tibial growth hormone

Fig. 7.5 Bioassayable growth hormone in control versus exercise groups sampled pre and post an acute resistance exercise test consisting of six sets of 10-RM squats separated by a 2-min rest period before (a) and after (b) 6 months of periodized resistance training (RT) (unpublished data). by biological and immunological assays, are often appears to exist in humans as well as rats. It may be not parallel. The intensity and duration of the exer- an important factor in controlling the production cise appear to play a major role in this dichotomy and/or release of GH variants from the pituitary of activity. After training, resting concentrations of gland. GH active in the rat bone growth assay are increased. In conclusion, readers of this chapter know that 3 The GH-producing cell system in the rat pituitary the information explosion we are experiencing in gland is heterogeneous. GH cellular heterogeneity may today’s biological sciences results not only from also be present in the human pituitary gland, but these past studies but from a rapidly expanding techno- types of studies are difficult to do. Evidence supports logy and data base as well. This is obvious. In this the view that GH-containing secretion granules and chapter the authors have considered seminal works GH-producing cells are heterogeneous. This hetero- done some 50 years ago and tried to show that they geneity appears to have physiological relevance. are relevant today and help in a fuller appreciation Additional work is required to establish a link of the role that variant forms of the GH molecule between these components in exercising rats and might play in the beneficial effects of human exer- humans. cise. We have tried to show that a successful mar- 4 Regulatory mechanisms responsible for releasing GH riage between the experimental approaches used in variants from the pituitary may involve signals from endocrinology, endocrine biochemistry/cell biology neural pathways originating in muscles activated by and exercise physiology may lead to new insights exercise. In this chapter we offer evidence to support into the role that GH molecular heterogeneity plays the existence of a novel feedback loop from certain in human exercise. Obviously a start has been made, muscle groups to the pituitary gland. This loop but much more needs to be done.

References

Baumann, G. (1991a) Growth hormone Baumann, G. (1999) Growth hormone Grindeland, R.E. & Edgerton, V.R. heterogeneity: genes, isohormones, heterogeneity in human pituitary and (2000) Bioassayable growth hormone variants, and binding proteins. Endocrine plasma. Hormone Research 51 (suppl. 1), release in rats in response to a single Reviews 12(4), 424–449. 2–6. bout of treadmill exercise. Journal Baumann, G. (1991b) Metabolism of Baumann, G., Shaw, M., Amburn, K. et al. of Applied Physiology 89(6), 2174– growth hormone (GH) and different (1994) Heterogeneity of circulating 2178. molecular forms of GH in biological growth hormone. Nucear Medicine and Cunningham, B.C. & Wells, J.A. (1989) fluids. Hormone Research 36 (suppl. 1), Biology 21(3), 369–379. High-resolution epitope mapping 5–10. Bigbee, A.J., Gosselink, K.L., Roy, R.R., of hGH-receptor interactions by 92 chapter 7

alanine-scanning mutagenesis. Science Journal of Physiology. Endocrinology and GAP-43. Endocrinology 134(1), 244(4908), 1081–1085. Metabolism 281(4), E878–E887. 503–506. Cunningham, B.C., Ultsch, M., De Vos, Lewis, U.J., Pence, S.J., Singh, R.N. & Papkoff, H. & Li, C.H. (1962) Hypophyseal A.M. et al. (1991) Dimerization of the VanderLaan, W.P. (1975) Enhancement growth hormone. In: Methods in Hormone extracellular domain of the human of the growth promoting activity of Research II (Dorfman, R. ed.). Academic growth hormone receptor by a single human growth hormone. Biochemical and Press, New York: 671–704. hormone molecule. Science 254(5033), Biophysical Research Communications Roswall, E.C., Mukku, V.R., Chen, A.B. 821–825. 67(2), 617–624. et al. (1996) Novel assays based on Dannies P.S. (1999) Protein hormone Lewis, U.J., Sinha, Y.N. & Lewis, G.P. human growth hormone receptor as storage in secretory granules: (2000) Structure and properties of alternatives to the rat weight gain mechanisms for concentration and members of the hGH family: a review. bioassay for recombinant human growth sorting. Endocrine Reviews 20(1), 3–21. Endocrine Journal 47 (suppl.), S1–S8. hormone. Biologicals 24(1), 25–39. Ellis, S. & Grindeland, R.E. (1974) McCall, G.E., Goulet, C., Grindeland, Roth, J., Glick, S.M., Yalow, R.S. & Berson, Dichotomy between bioassayable and R.E. et al. (1997) Bed rest suppresses S.A. (1963) Secretion of human growth immunoassayable growth hormone. bioassayable growth hormone release hormone: physiologic and experimental In: Advances in Human Growth Hormone in response to muscle activity. Journal of modification. Metabolism 12, 577–579. Research (Raiti, S., ed.). DHEW Applied Physiology 83(6), 2086–2090. Rowlinson, S.W., Waters, M.J., Lewis, U.J. Publication No. (NIH) 74–612, US McCall, G.E., Goulet, C., Roy, R.R. et al. & Barnard, R. (1996) Human growth Government Printing Office, (1999) Spaceflight suppresses exercise- hormone fragments 1–43 and 44–191: in Washington, D.C.: 409–433. induced release of bioassayable growth vitro somatogenic activity and receptor Ellis, S., Nuenke, J.M. & Grindeland, R.E. hormone. Journal of Applied Physiology binding characteristics in human and (1968) Identity between the growth 87(3), 1207–1212. nonprimate systems. Endocrinology hormone degrading activity of the McCall, G.E., Grindeland, R.E., Roy, R.R. 137(1), 90–95. pituitary gland and plasmin. & Edgerton, V.R. (2000) Muscle afferent Rubin, M.R., Kraemer, W.J., Kraemer, R.R. Endocrinology 83(5), 1029–1042. activity modulates bioassayable growth et al. (2003) Responses of growth Farrington, M. & Hymer, W.C. (1990) hormone in human plasma. Journal of hormone aggregates to different Growth hormone aggregates in the rat Applied Physiology 89(3), 1137–1141. intermittent exercise intensities. adenohypophysis. Endocrinology 126, McCall, G.E., Gosselink, K.L., Bigbee, A.J. European Journal of Applied Physiology 1630–1638. et al. (2001) Muscle afferent-pituitary 89(2), 166–170. Gosselink, K.L., Grindeland, R.E., Roy, axis: a novel pathway for modulating Singh, R.N., Seavey, B.K., Rice, V.P., R.R. et al. (1998) Skeletal muscle afferent the secretion of a pituitary growth factor. Lindsey, T.T. & Lewis, U.J. (1974) regulation of bioassayable growth Exercise and Sport Sciences Reviews 29(4), Modified forms of human growth hormone in the rat pituitary. Journal of 164–169. hormone with increased biological Applied Physiology 84(4), 1425–1430. Nguyen, N.Y., Grindeland, R.E. & activities. Endocrinology 94(3), 883–891. Gosselink, K.L., Grindeland, R.E., Roy, Chrambach, A. (1981) Isolation of Stolar, M.W., Amburn, K. & Baumann, G. R.R. et al. (2000) Afferent input from rat human growth hormone isohormones (1984) Plasma ‘big’ and ‘big–big’ slow skeletal muscle inhibits D and E in milligram amounts (II), using growth hormone (GH) in man: an bioassayable growth hormone release. isoelectric focusing on polyacrylamide oligomeric series composed of Journal of Applied Physiology 88(1), gel. Preparative Biochemistry 11(2), structurally diverse GH monomers. 142–148. 173–189. Journal of Clinical Endocrinology and Greenspan, F.S., Li, C.H., Simpson, M.E. Nindl, B.C., Kraemer, W.J. & Hymer, Metabolism 59(2), 212–218. & Evans, H.M. (1949) Bioassay of W.C. (2000) Immunofunctional vs. Strasburger, C.J. (1994) Implications of hypophyseal growth hormone: the tibia immunoreactive growth hormone investigating the structure–function test. Endocrinology 45, 455–463. responses after resistance exercise in relationship of human growth hormone Hunter, W.M. & Greenwood, F.C. (1964) A men and women. Growth Hormone and in clinical diagnosis and therapy. radio-immunoelectrophoretic assay for IGF Research 10(2), 99–103. Hormone Research 41 (suppl. 2), 113–119. human growth hormone. Biochemistry Nindl, B.C., Hymer, W.C., Deaver, D.R. & Strasburger, C.J., Wu, Z., Pflaum, C.D. Journal 91(1), 43–56. Kraemer, W.J. (2001) Growth hormone & Dressendorfer, R.A. (1996) Hymer, W.C. & McShan, W.H. (1963) pulsatility profile characteristics Immunofunctional assay of human Isolation of rat pituitary granules and following acute heavy resistance growth hormone (hGH) in serum: a the study of their biochemical properties exercise. Journal of Applied Physiology possible consensus for quantitative and hormonal activities. Journal of Cell 91(1), 163–172. hGH measurement. Journal of Clinical Biology 117(1), 67–86. Nindl, B.C., Kraemer, W.J., Marx, J.O., Endocrinology and Metabolism 81(7), Hymer, W.C., Kirshnan, K., Kraemer, W., Tuckow, A.P. & Hymer, W.C. (2003) 2613–2620. Welsch, J. & Lanham, W. (2000) Growth hormone molecular Wada, M., Uchida, H., Ikeda, M. et al. Mammalian pituitary growth hormone: heterogeneity and exercise. Exercise and (1998) The 20-kilodalton (kDa) human applications of free flow electrophoresis. Sport Sciences Reviews 31(4), 161–166. growth hormone (hGH) differs from the Electrophoresis 21(2), 311–317. Paden, C., Moffett, C. & Benowitz, L. 22-kDa hGH in the complex formation Hymer, W.C., Kraemer, W.J., Nindl, B.C. (1994) Innervation of the rat anterior with cell surface hGH receptor and et al. (2001) Characteristics of circulating and neurointermediate pituitary hGH-binding protein circulating in growth hormone in women after acute visualized by immunocytochemistry human plasma. Molecular Endocrinology heavy resistance exercise. American for the growth-associated protein (Baltimore, Md) 12(1), 146–156. growth hormone variants and human exercise 93

Wallace, J.D., Cuneo, R.C., Bidlingmaier, Weiss, S., Bergland, R., Page, R., Turpen, C. Experimental Biology and Medicine 159, M. et al. (2001) The response of & Hymer, W.C. (1978) Pituitary cell 409–413. molecular isoforms of growth hormone transplants to the cerebral ventricles Wood, P. (2001) Growth hormone: its to acute exercise in trained adult males. promote growth of hypophysectomized measurement and the need for assay Journal of Clinical Endocrinology and rats. Proceedings of the Society for harmonization. Annals of Clinical Metabolism 86(1), 200–206. Biochemistry 38(5), 471–482. Chapter 8

Growth Hormone Binding Proteins

GERHARD BAUMANN

Growth hormone binding proteins (GHBPs) are affinity GHBPaa convention that will also be fol- soluble, circulating proteins that form complexes lowed in this chapter unless stated otherwise. with growth hormone (GH). They are integral parts of the somatotropic axis and have activities as mod- Nature and chemical properties ulators of GH action and GH transport in blood. An overview of the growth hormone–insulin-like The high affinity GHBP represents the extracellular growth factor (GH–IGF) axis is shown in Fig. 8.1. domain of the GHR (Leung, D.W. et al. 1987; Spencer et al. 1988). It is a single chain glycoprotein with a History molecular weight that varies widely depending on the species from 28 kDa (chicken) to 65 kDa The presence of GHBPs in blood was first postu- (humans), with most of the variation due to differ- lated in the 1960s (Irie & Barrett 1962; Touber & ences in glycosylation. The polypeptide backbone Maingay 1963; Collipp et al. 1964; Hadden & Prout accounts for approximately 28–30 kDa, with minor 1964), but these observations were not generally differences among species. The GHBP is evolution- accepted as physiological phenomena at that time arily conserved from teleost to humans; it has been (Berson & Yalow 1966a, 1966b). In 1977, Peeters & found in the blood of all vertebrates examined. In Friesen (1977) described a GH binding factor in some species it is derived directly from the GHR by pregnant mouse serum. However, this observation proteolysis, in others (rodents), it is synthesized as was also largely ignored. It was not until Baumann a separate gene product (see below). The precise and Herington independently described, character- structure of the GHBP is known in only a few spe- ized and partially purified GHBPs from human and cies, with the carboxy terminus unknown in many. rabbit serum (Ymer & Herington 1985; Baumann et al. Two subdomains, each composed of beta-pleated 1986; Herington et al. 1986b) that the GHBPs were sheets, have been identified: an amino-terminal generally accepted as authentic. Two GHBPs, one subdomain 1 containing the GH binding site; and a with high affinity and the other with low affinity for carboxy-terminal subdomain 2 that is involved in GH were described at that time (Baumann et al. dimerization of the GHR. An approximately 10 1986). While the high affinity GHBP was easy to amino acid linear stem region extends between sub- work with and soon became recognized as the GH domain 2 and the transmembrane helix of the GHR receptor (GHR) ectodomain (Leung, D.W. et al. 1987; (Baumann & Frank 2002). The precise cleavage site Baumann et al. 1988), it took some years to character- in the GHR giving rise to the GHBP has recently ize the low affinity GHBP (Baumann et al. 1990; Tar been mapped in the rabbit: cleavage occurs in the et al. 1990), ultimately leading to its recognition as extracellular stem region between subdomain 2 and α transformed 2-macroglobulin (Kratzsch et al. 1995b). the transmembrane domain, with the GHBP car- In general, the term ‘GHBP’ is used for the high boxy terminus at residue no. 238, i.e. eight residues 94 growth hormone binding proteins 95

Hypothalamus de novo. It contains a carboxyterminal ‘tail’ of 27 (mouse) or 17 (rat) amino acids in lieu of the trans- membrane domain in the GHR (Baumbach et al. 1989; Smith et al. 1989). The mouse and rat GHBP contain 273 and 255 amino acids, respectively. The + − GHRH ( ) Somatostatin ( ) extent of GHBP glycosylation differs among spe- (−) cies, but there is only limited information for rodent GHBP about the nature of the sugar moieties. (−) Mouse serum GHBP is glycosylated at three aspara- (−) Pituitary gine residues, wheras tissue-associated GHBP (see below) contains less carbohydrate and is glyco- GH GHBP sylated at two asparagine residues (Cerio et al. 2002). Rat serum GHBP contains sialic acid, whereas tissue-associated GHBP is rich in mannose (Frick Other tissues Liver et al. 1998). The details of the carbohydrate side chain structure are not known. In humans, two high affinity GHBPs exist (one containing and the other Circulating IGF-I Local IGF-I lacking the sequence encoded by exon 3 of the GHR gene) (Kratzsch et al. 1997b). This is the consequence of GHR polymorphism with respect to exon 3 Growth/Anabolism (Pantel et al. 2000; Seidel et al. 2003). The presence or absence of the exon 3-encoded sequence in the GHR Fig. 8.1 The hypothalamo–pituitary–somatotropic axis, or the GHBP has no significant functional con- also known as growth hormone–insulin-like growth factor (GH–IGF) axis. Pituitary GH secretion is under sequence with respect to GH binding. However, positive (stimulatory) control from the hypothalamus via small differences in the correlations between serum growth hormone releasing hormone (GHRH) and under GHBP level and anthropometric/metabolic para- negative (inhibitory) control via somatostatin. After meters have been reported for the two GHBP iso- secretion, GH binds to receptors in liver and virtually all forms (Seidel et al. 2003). other tissues, and to growth hormone binding proteins (GHBPs) in the circulation. There is interchange between The high affinity GHBP binds GH with dissoci- free GH, GHBP-bound GH and receptor-bound GH. ation constants ranging from 10–8 to 10–9 mol (Ymer Peripheral tissues produce IGF-I in response to GH. The & Herington 1985; Baumann et al. 1986; Herington liver is responsible for 60–70% of circulating IGF-I; other et al. 1986b; Smith et al. 1988; Massa et al. 1990). tissues are responsible for the rest. In tissues other than It exhibits somewhat lower affinity (K 10–6 to 10–7) the liver, local (paracrine/autocrine) IGF-I action is very d important. Direct (non-IGF-dependent) GH actions for the 20 000 kDa (20 K) variant of human GH are indicated by the dashed arrow. IGF-I feeds back (Baumann et al. 1986). Like the GHR, the GHBP has negatively on the hypothalamus and pituitary gland the capacity to form ternary (2 GHBP : 1 GH) com- to inhibit GH secretion. GH itself also inhibits its own plexes with GH, but due its low concentration in production via negative feedback at hypothalamic level biological fluids, the 1 : 1 GHBP–GH complex pre- (short-loop feedback). dominates under physiological conditions (Baumann et al. 1994). The association rate for the human outside the plasma membrane (Wang et al. 2002). GHBP is very rapid (~ 2 × 107·mol−1·min−1 at 37°C, Based on the sequence similarity of the rabbit and 80% maximum binding reached in 5 min), the dis- human GHRs in the extracellular stem region, it sociation rate is 3.7 × 10−2·min−1 at 37°C (disso- appears likely that the human GHBP has the same ciation half-time ~ 20 min) (Baumann et al. 1986; length, though this still needs direct confirmation. Veldhuis et al. 1993, Baumann 1995). In rodents, the GHBP is the product of an altern- The low affinity GHBP is a heterogeneous plasma atively spliced GHR mRNA and is synthesized component that binds GH with a Kd in the micromolar 96 chapter 8

range (Baumann et al. 1986, 1990; Massa et al. 1990; It is possible that other enzymes in the same class Tar et al. 1990; Leung, K.C. et al. 2000). It has high also contribute to GHBP shedding, but this remains binding capacity and in humans has been shown to to be investigated. The regulation of TACE activity α represent a modified form of 2-macroglobulin and the shedding process are still poorly under- α (‘transformed 2-macroglobulin’) (Kratzsch et al. stood. The conformational change induced in the 1995b). Relatively little is known about the mole- GHR by binding of GH (dimerization or change in cular nature of low affinity GHBPs in animals. predimerized GHR) renders it less prone to proteo- lysis than the monomeric, unliganded GHR (Zhang Generation and tissues source(s) et al. 2001). GHBP shedding is thought to occur prin- cipally if not exclusively at the cell surface, based on As indicated above, the high affinity GHBP is gener- the localization of active TACE and the fact that a ated by different mechanisms depending on spe- GHR variant with a long plasma membrane residence cies. In humans, rabbits and several other species, time (devoid of the cytoplasmic/internalization the GHBP is produced by proteolytic cleavage of domain) is a particularly good source of GHBP the juxtamembranous stem region of the GHR, a (Dastot et al. 1996). process named ‘shedding’ (Fig. 8.2). The enzyme Rodents generate GHBP by an entirely different involved in this process has been recently identified: mechanism. Both mouse and rat ghr genes contain a it is a zinc metalloproteinase of the ADAM family special exon (exon 8A) encoding the hydrophilic named TACE (tumor necrosis factor-α converting GHBP tail (see above) interposed between exons 7 enzyme), also known as ADAM-17 (Black et al. 1997; and 8 (Edens et al. 1994; Zhou et al. 1994, 1996). Exon Zhang et al. 2000). Mature, catalytically active TACE 8 encodes the transmembrane helix. Differential is a plasma membrane-resident enzyme that asso- mRNA splicing of exon 7 to either exon 8A or exon 8 ciates with the GHR, followed by cleavage and yields the GHBP or the GHR, respectively (Fig. 8.3). shedding of GHBP and a truncated GHR ‘remnant’, Both transcripts are expressed in the same tissues, which has its own intracellular fate. TACE is but it is unknown what regulates their relative responsible for cleavage of a number of transmem- expression. It should be noted that the mouse GHR brane proteins in their extracellular domain, with is not completely resistant to TACE proteolysis, at shedding of soluble ectodomains akin to the GHBP. least when induced by phorbol ester. However, the

Fig. 8.2 The generation of growth hormone binding protein (GHBP) Circulation via proteolytic cleavage of the growth hormone receptor (GHR). A membrane-bound zinc-dependent GHBP metalloproteinase (tumor necrosis Zn factor-α converting enzyme [TACE]) cleaves the GHR in its juxtamembranous stem region 8–9 amino acid residues outside the plasma membrane. The GHR ectodomain is shed as the GHBP, Plasma membrane which reaches the circulation. The GHR remnant protein, consisting of the transmembrane and intracellular domains, undergoes additional PKC TACE processing and may have its own MAPK Further processing biological role. TACE is activated by etc. GHR GHR Internalization remnant MAP kinase (MAPK) and protein Degradation kinase C (PKC) dependent pathways. growth hormone binding proteins 97

2 3 4 5 6 7 8 9 10 Human GHR gene

GHR mRNA 2 3 4 5 6 7 8 9 10

8a Mouse ghr gene

GHBP mRNA 2345 6 7 8a

Fig. 8.3 The human and mouse GHR genes and their products are shown. In humans and most other species, the only gene product is the growth hormone receptor (GHR). In rodents, an additional embedded exon (exon 8a) encodes the hydrophilic tail of the growth hormone binding protein (GHBP). Splicing exon 8a to exons 2–7 yields the GHBP mRNA (bottom). Splicing exon 8–10 to exons 2–7 yields the GHR mRNA (top). Exon 8a is not spliced to downstream exons because it contains a cleavage/polyadenylation site and lacks a canonical splice donor site at its 3’ end. Exon 1 of the GHR gene is not shown because it is not part of the coding region (it encodes the 5’ untranslated region[s]). Light tint denotes the extracellular, darker tint the intracellular domain. denotes the transmembrane domain, the hydrophilic GHBP tail.

cleavability of the mouse GHR is about two orders substantial portion of rodent GHBP remains associ- of magnitude lower than that of the rabbit GHR (G. ated with the plasma membrane (and intracellular Baumann, unpublished). In vivo, it appears that membranes) by an as yet unknown link (Frick et al. most if not all of the circulating rat GHBP is derived 1994, 1998); it has been suggested that an Arg- from the alternative mRNA splicing mechanism Gly-Asp sequence in the GHBP may interact with (Sadeghi et al. 1990). The two different mechanisms membrane integrins to provide a tether (Cerio et al. of GHBP generation are illustrated in Fig. 8.4. 2002). GHBP in the circulation differs from the Rhesus monkeys use both the proteolytic and tissue-associated form in its glycosylation moiety. an alternative mRNA splice mechanism for GHBP Membrane-associated GHBP forms have not been generation (Martini et al. 1997). In that case, the described in non-rodent species. The tissue source alternative mRNA encoding the GHBP derives from of the GHBP in species employing the proteolytic a readthrough into intron 7, which results in a 7 shedding mechanism is less clear as it is more amino acid ‘tail’ replacing the transmembrane difficult to differentiate the GHBP from the GHR. domain due to an intronic stop codon. It is not Since both the GHR and TACE are expressed ubi- known which mechanism predominates in generat- quitously, all tissues can theoretically contribute ing the GHBP in the monkey. to GHBP generation. However, the quantitative The tissue sources of the GHBP are well-defined aspects of GHBP generation by individual tissues in the rodent, where the GHBP can be specifically are not clearly established. Based on the relative recognized and differentiated from the GHR at abundance of GHRs in the liver, that organ is gen- both the mRNA and protein level by its unique erally thought to be a major source. However, it carboxyterminal tail. GHBP is expressed ubiquit- should be noted that this concept has not been ously, and generally is coexpressed with the GHR directly validated. Studies of venous gradients in (Carlsson, B. et al. 1990; Lobie et al. 1992). However, visceral organ effluents have not identified a major their expression is not necessarily regulated in a organ source (Segel et al. submitted for publication). parallel fashion (Walker et al. 1992). Of interest, a It appears likely that multiple tissues contribute 98 chapter 8

Humans, rabbit, Mouse, rat other species

NH2 NH2

Proteolytic cleavage Secretion

Receptor Binding protein

COOH COOH Receptor mRNA Receptor mRNA

(a) (b) BP mRNA

Fig. 8.4 The two mechanisms of growth hormone binding protein (GHBP) generation are shown. (a) The proteolytic shedding mechanism. (b) The splice variant/direct synthesis and secretion mechanism. (Adapted from Baumann, G. (1990) Growth hormone binding proteins. Trends in Endocrinology and Metabolism 1, 342–347. Copyright 1990, with permission from Elsevier.)

to the circulating GHBP pool, though their relative This concentration, together with its affinity, allows contributions remain to be determined. the GHBP to act as a buffer and dynamic modulator for circulating free GH. Under physiological and Growth hormone binding proteins in basal conditions, approximately 45% of circulating biological fluids GH in human blood is bound to the high affinity GHBP (Baumann et al. 1988, 1990). This proportion The high affinity GHBP is found in blood and most changes dynamically after a GH secretory spike other biological fluids, such as urine, lymph, milk, (Veldhuis et al. 1993). semen, follicular fluid and amniotic fluid (Hattori GHBP is also present within the cell (Herington et al. 1990; Postel-Vinay et al. 1991a; Amit et al. 1993; et al. 1986a; Lobie et al. 1991; Frick et al. 1994), but the Maheshwari et al. 1995; Harada et al. 1997). source, destination and function of this intracellular Cerebrospinal fluid contains no detectable GHBP GHBP is not clear. (Nixon & Jordan 1986). Unlike in rabbit milk, the The low affinity GHBP has only been described in GHBP found in human milk appears to related to the blood, where it circulates at micromolar concentra- prolactin receptor rather than the GHR (Mercado tions (Baumann et al. 1990; Leung, K.C. et al. 2000). In & Baumann 1994). The concentration of GHBP in humans, approximately 8% of circulating GH is blood varies over a 10-fold range; it is generally pre- bound to this GHBP; in rats it can be calculated that sent at nanomolar to subnanomolar concentrations. about 20% of GH is bound to the low affinity GHBP growth hormone binding proteins 99

(Barsano & Baumann 1989; Baumann et al. 1989a; decreased GH action is the formation of unproduct- Leung, K.C. et al. 2000). ive, non-signaling GHR/GHBP dimers (Fig. 8.5). Dimerization and proper GHR dimer conformation Functional aspects is necessary for signal transduction by the GHR. A GHR/GHBP dimer cannot perform that function. The principal known function of the GHBPs is to Formation of such heterodimers by GH binding form complexes with GH. Quantitatively, this is would inhibit GH action in a GHBP concentration- more important for the high affinity GHBP than the dependent fashion. Indeed, this effect has been low affinity GHBP. An indirect ‘function’ of GHBP demonstrated for naturally occurring and mutant generation is the inactivation of GHRs by cleaving forms of the GHR that lack the intracellular domain and shedding its ectodomain, a process that can be (Ayling et al. 1997; Ross et al. 1997; Iida et al. 1999). It viewed as receptor ‘decapitation’. GH binding has has not directly been proven that the same phe- multiple consequences. At the local (cellular/tissue) nomenon occurs with soluble GHBP, though this level, GHBP competes with the GHR for GH ligand, would be predicted. The short GHR, in contrast to the consequence of which is decreased GH action the GHBP, contains a transmembrane domain and (Fig. 8.5). This effect can be clearly shown in vitro, is membrane-anchored. If the GHR exists in the where GHBP inhibits GH binding to GHRs and GH membrane in a predimerized form even in the action in a dose-dependent manner (Lim et al. 1990; absence of GH binding (Ross et al. 2001), the mem- Mannor et al. 1991). An additional likely reason for brane-resident short GHR form is not necessarily

GH

GH GH GHBP

GHR GHR GHR GHR GHBP

Plasma membrane

Jak2 Jak2

(a) (b)Intracellular (c) signaling

Fig. 8.5 Inhibitory actions of the growth hormone binding protein (GHBP) on GH bioactivity. (a) GHBP competes with the GHR for GH. (b) Normal GH signaling through the growth hormone receptor (GHR). Binding of GH to the GHR induces GHR dimerization or activation of a preformed GHR dimer. The conformationally appropriate GHR-dimer binds Jak2 and initiates intracellular signaling. (c) A GHR–GHBP heterodimer is unable to initiate signaling, with the GHBP acting as a dominant negative inhibitor. 100 chapter 8

representative for the extracellular, soluble GHBP. prehensively address this issue. Because of these Therefore, the concept of GHR/GHBP heterodimers limitations, the following discussion is primarily remains to be formally validated. focused on the regulation of GHBP levels in serum. In contrast to its inhibitory action in vitro, the In humans, the principal factors affecting serum GHBP tends to enhance GH action in vivo. GHBP GHBP as part of normal physiology are develop- prolongs the plasma half-life of GH through forma- ment, gender, aging and nutrition. For reasons that tion of a complex that is too large for efficient glomer- are not known, serum GHBP concentrations in ular filtration and renal eliminationathe principal normal subjects vary over a 10-fold range (~ 0.3– route of GH clearance (Baumann et al. 1987a, 1989b). 3.0 nmol) (Rajkovic et al. 1994; Maheshwari et al. The complex also diminishes GH clearance through 1996); it is not clear whether this normal variablity GHR-mediated mechanisms, such as cellular inter- has biological significance. There is no significant nalization, and delays chemical degradation. In rats, diurnal variation in serum GHBP (Snow et al. 1990; the metabolic clearance of complexed GH is 10-fold Carmignac et al. 1992; Carlsson, L.M. et al. 1993), but lower than that of free GH (Baumann et al. 1989b). a minor seasonal variation with nadir in August In humans, the GH–GHBP complex has been estim- has been reported in children (Gelander et al. 1998). ated to have a plasma half-life of 25–29 min, versus GHBP levels are very low in the fetus, rise rapidly 4–9 min for free GH (Veldhuis et al. 1993). The during early childhood, stay constant through ado- GH–GHBP complex in blood serves as a circulating lescent and adult life, and decline in old age after GH reservoir, which dynamically dampens GH the age of 60 years (Daughaday et al. 1987; Holl et al. oscillations resulting from secretory pulses. GHBP, 1991; Martha et al. 1993; Maheshwari et al. 1996). A when given in large doses, has been shown to similar ontogenetic pattern is present in the rat enhance GH bioactivity in vivo, despite its inhibitory (Mulumba et al. 1991). GHBP levels are higher in actions in vitro (Clark et al. 1996). The net effect females than males, both in humans and more of the high affinity GHBP on GH action in the in- markedly in rodents (Massa et al. 1990; Hattori et al. tact organism is thus complex, concentration- and 1991; Rajkovic et al. 1994). This is probably in large compartment-dependent, and difficult to predict. part an estrogen effect. Maternal GHBP levels dur- Little is known about the effect of the low affinity ing pregnancy show a very marked species differ- GHBP on GH action. Based its low affinity, it is ence. In humans, there is only a small increase in likely to form a loose complex with GH that easily GHBP levels in early pregnancy (Blumenfeld et al. dissociates. Therefore, it probably has only a limited 1992), whereas in mice, GHBP levels in serum (and impact on GH dynamics and GH action. membrane-associated GHBP in the liver) show a very large increase (Cramer et al. 1992; Camarillo et al. 1998). It is the latter phenomenon that led to Regulation of growth hormone binding the first description of the GHBP (Peeters & Friesen protein production 1977). Rats also increase their GHBP during preg- In species that generate GHBP by the proteolytic nancy, though to a lesser degree than mice (Frick mechanism, GHBP production depends both on et al. 1998). An important factor regulating GHBP GHR expression and on the regulation of TACE levels is nutrition. Malnutrition decreases and over- activity. Both the GHR and TACE are expressed nutrition increases serum GHBP; and a highly sig- ubiquitously. GHR expression is regulated by nificant correlation exists between body mass index developmental stage, gender, species, metabolic and GHBP levels, and especially between visceral state, and differentially by tissue. Little is currently fat and GHBP levels (Hochberg et al. 1992; Martha known about the regulation of TACE activity. In et al. 1992; Roelen et al. 1997b). These changes paral- rodents, GHBP production is linked to expression lel those seen for IGF-I levels and probably reflect of the GHBP mRNA variant. This is also regulated the effect of insulin on GHR expression and conse- in a complex, tissue and metabolic state-dependent quently GHBP levels (Baxter & Turtle 1978; Mercado manner, and no systematic studies exist that com- et al. 1992; Kratzsch et al. 1996). growth hormone binding proteins 101

GH up-regulates GHBP in rodents (Sanchez- change in GHBP parallels altered GH sensitivity; Jimenez et al. 1990; Carmignac et al. 1992), but data the GHBP level is therefore thought to mirror GHR on this subject in humans are variable and incon- abundance in tissues. Foremost among the disor- sistent (see Baumann 2001 for review). It may be ders with abnormal GHBP is the genetic GH insens- concluded from this that GH does not have a major itivity syndrome due to inactivating mutations in effect on GHBP levels in humans. Of interest, the GHR gene (Laron syndrome), which results in acromegaly, a disease of chronic GH excess, is severe growth retardation and dwarfism (Rosenfeld associated with low to low normal GHBP in most et al. 1994). Absence or malfunction of the GHBP is studies (Amit et al. 1992; Roelen et al. 1992; Mercado generally a direct result of the mutant GHR, which et al. 1993; Kratzsch et al. 1995a; Fisker et al. 1996). is either not expressed (e.g. gene deletions, non- This is not necessarily a direct function of GH, but sense mutations), prematurely degraded or not may be a result of other adjustments occurring in properly directed to the plasma membrane (e.g. acromegaly. Thyroid hormone up-regulates GHBP some missense mutations), or unable to bind GH levels (Amit et al. 1991; Romero et al. 1996). (certain missense mutations) (see Baumann 2002 for Estrogens, especially when given by the oral route, an updated listing of known GHR mutations). increase GHBP levels in humans and rodents, but Absence of GHBP activity in the serum of patients lower serum GHBP in the rabbit (Weissberger et al. with Laron syndrome was the first strong indication 1991; Carmignac et al. 1993; Yu et al. 1996). Andro- that the GHBP is a GHR fragment (Baumann et al. gens lower serum GHBP levels (Postel-Vinay et al. 1987b; Daughaday & Trivedi 1987). About 80% of 1991b; Keenan et al. 1996; Yu et al. 1996). Glucocor- cases with Laron syndrome have low or undetect- ticoids lower GHBP in humans and rodents, but able GHBP in their blood (Woods et al. 1997). The increase GHBP in rabbit blood (Heinrichs et al. 1994; others have either normal serum GHBP or in rare Miell et al. 1994; Gabrielsson et al. 1995). Insulin up- cases even elevated GHBP. The underlying GHR regulates GHBP levels (Mercado et al. 1992; Massa mutations in GHBP-positive cases include those et al. 1993; Kratzsch et al. 1996), whereas IGF-I, in that prevent proper receptor dimerization or lack a single report, was shown to lower GHBP levels the intracellular signaling domain (Duquesnoy et al. (Silbergeld et al. 1994). 1994; Ayling et al. 1997; Kaji et al. 1997; Iida et al. Exercise and physical training have an effect on 1998; Gastier et al. 2000). A mutant receptor lacking plasma GHBP levels. Acute exercise, such as cycle the transmembrane helix results in massively elev- ergometry, induces a short-lived and mild increase ated serum GHBP activity, which represents the in GHBP (Wallace et al. 1999). Chronic endurance or mutant, soluble GHR rather than the normal GHBP fitness training has been shown to lower serum (Woods et al. 1996; Silbergeld et al. 1997). GHBP by 10–40% in most studies (Roemmich & Several disorders associated with acquired GH Sinning 1997; Eliakim et al. 1998b, 2001; Scheett et al. insensitivity are also characterized by abnormally 2002), but at least one study showed a small rise in low GHBP levels. Catabolic conditions, such as GHBP (Roelen et al. 1997a). GHBP is inversely related malnutrition, insulinopenic uncontrolled diabetes, to peak oxygen uptake and fitness (Eliakim et al. post-traumatic states and critical illness are examples 1998a). This is in part linked to the above-mentioned of acquired GH-resistance, characterized by low relatienship between adiposity and GHBP. The IGF-I levels despite normal or elevated GH secre- physiological significance of these exercise/training tion. Growth failure may occur in severe cases related GHBP changes is not fully understood. (Mauriac syndrome, a condition of poorly con- trolled diabetes, hepatomegaly and growth retarda- Growth hormone binding protein tion) (Mandell & Berenberg 1974; Mauras et al. and disease 1991). The fact that serum GHBP is decreased in GH-resistant states lends credence to the concept Several pathological conditions are associated with that serum GHBP concentration reflects tissue GHR abnormal serum GHBP levels. In most cases, the abundance. In animal models, catabolic states are 102 chapter 8

associated with decreased hepatic GHR levels and With regard to the low affinity GHBP, little is GH resistance (Baxter & Turtle 1978; Postel-Vinay known about changes in its serum level in either et al. 1982; Massa et al. 1993). Upon resolution of the normal physiology or in disease. underlying disease process, GH sensitivity, GHR expression and GHBP levels return to normal. The Assays for the growth hormone binding changes in GHR expression and consequent GHBP protein (Figs 8.6 and 8.7) release are thought to be largely mediated by insulin (Mercado et al. 1992; Hanaire-Broutin et al. 1996). The classical assay for both the high and the low The counterpart of GH resistance, GH hyper- affinity GHBPs measures their functional property sensitivity, is associated with elevated GHBP levels. of radiolabeled GH binding, with separation of The only well-recognized condition in this category bound from free GH by size exclusion chromato- is overnutrition/obesity, which is characterized by graphy (Baumann et al. 1986; Herington et al. 1986b). normal or elevated IGF-I levels despite suppressed This assay is quantitative under most circumstances GH secretion. It has long been known that over- because under physiological conditions the GHBPs weight children grow faster than lean children in serum are largely unoccupied. Correction for (Forbes 1977). Obesity is associated with elevated occupation of the high affinity GHBP by endogen- serum GHBP levels, probably reflecting increased ous GH must be made above a GH concentration tissue GHR expression (Hochberg et al. 1992; of 10 ng·mL–1 (Baumann et al. 1989a). Variations of Kratzsch et al. 1997a; Roelen et al. 1997b). Thus, the this fundamental GH binding assay employ other biochemical parameters as well as the functional methods for separation of bound from free GH, aspects of the GH–IGF axis in obesity are the exact such as charcoal or immunoprecipitation with anti- opposite from those in malutrition. GHR antibodies (Barnard et al. 1989; Amit et al. 1990;

Separate free from bound GH

8 Size exclusion chromatography III 6

/fraction) Vo Vt Fig. 8.6 Growth hormone binding -3 protein (GHBP) assays based on the 125 10 4 I-GH x principle of GH binding. To serum II containing GHBP and endogenous 2 GH radioiodinated GH is added, I (cpm I and GHBP–GH complex formation 125 IV 0 is allowed. Bound GH is then 50 100 150 separated from free GH by either Fraction number size exclusion chromatography (top), GH immunoprecipitation with anti-GH Incubate antibodies (center), or adsorption of GHBP free GH with activated charcoal Immunoprecipitation (bottom). The assay is quantitative of bound GH because under most conditions, GHBP is essentially unoccupied by endogenous ligand. At high endogenous GH levels (> 10 ng·mL–1), correction for GHBP saturation by unlabeled GH is Charcoal adsorption necessary. Different variations of this of free GH binding assay exist, using other modes of separating free from bound GH. growth hormone binding proteins 103

Coat plate with anti-GHR/GHBP antibody

GH

GHBP

Add serum sample and GH. Incubate

Enzyme

Anti-GH antibody

Add enzyme-coupled anti-GH antibody. Develop with substrate → Colorimetry

Fig. 8.7 Ligand-mediated immunofunctional assay (LIFA) for growth hormone binding protein (GHBP). This is a solid- phase, sandwich-type, two-site assay. A monoclonal antibody directed against the extracellular domain of the growth hormone receptor (GHR) (the GHBP), yet not interfering with its GH-binding site, is adsorbed to a microtiter plate. After washing, serum and an excess of exogenous GH are added, and formation of a ternary complex consisting of the antibody, GHBP and GH is allowed to form. After washing, an anti-GH-antibody coupled to an enzyme (e.g. horseradish peroxidase) is added. The antibody recognizes the solid-phase complex. Signal amplification is then achieved by the enzymatic reaction, using a color-yielding substrate.

Ho et al. 1993). Specific assays that differentiate the (Kratzsch et al. 1997b). Other assays with variations GHBP from the GHR have been devised by using on these methodological themes have been reported antibodies directed against the unique hydrophilic for several species, including human, rat and mouse tail of the rodent GHBP (Barnard et al. 1994); this GHBP. Unfortunately, relatively little information is strategy cannot be used for GHBP in species that use available about the correlation among these assays. the proteolytic shedding mechanism for GHBP gen- Commercial assays are available, but also suffer eration (e.g. humans, rabbits). A two-site sandwich- from a relative lack of crossvalidation with estab- type assay with enzyme linked immunosorbent lished assays. GHBP measurement is still primarily assay (ELISA) design (‘ligand-mediated immuno- a research tool; its main practical application in clin- functional assays [LIFAs]’) has been developed for ical medicine is in the diagnosis of GH insensitivity the human high affinity GHBP (Carlsson, L.M. et al. (Laron) syndrome. 1991). Its results correlate well with the conven- No standardized assays for the low affinity GHBP tional GH-binding assay, but for unknown reasons are available. This plasma component has been the absolute GHBP values are lower than those measured by GH binding assay followed by size obtained in other assays (Mercado et al. 1993). One exclusion chromatography (Baumann et al. 1989a; α assay with classical radioimmunoassay design Tar et al. 1990) or immunoprecipitation with anti- 2- (radiolabeled GHBP tracer and anti-GHBP anti- macroglobulin antibody (Kratzsch et al. 1995b). serum) has been reported for human GHBP (Kratzsch et al. 1995a). This type of assay is independ- Effect of the high affinity growth hormone ent of GH binding and able to measure dysfunc- binding protein on growth hormone tional GHBP with abnormal GH binding properties measurement in serum (such as mutant GHBP in certain cases of Laron syn- drome). A radioimmunoassay specific for the exon The high affinity GHBP can interfere with GH im- 3-containing human GHBP has also been developed munoassays in serum because of competition with 104 chapter 8

antibodies for GH. In general, antibodies, especi- rodents) as a soluble GHR mRNA splice variant. ally when polyclonal, have higher affinity for GH Cleavage by TACE occurs in the juxtamembranous, than the GHBP. Nevertheless, depending on the extracellular stem region of the GHR. The GHBP assay design, interference can be significant. Assays has complex effects on GH blood transport, GH particularly vulnerable are those employing relat- clearance and GH action, with both enhancing and ively low affinity monoclonal anti-GH antibodies, inhibitory modulation of GH bioactivity. The physio- quick turnaround assays that are run under brief logical significance of the GHBP for the somato- incubation, non-equilibrium conditions and insens- tropic axis is still incompletely understood. Serum itive assays employing a high serum volume. The GHBP levels appear to reflect GH responsivity of non-equilibrium assay design is problematic because the organism, presumably by reflecting GHR status time is required to transfer GHBP-bound GH from in tissues. Regulation of serum GHBP levels is com- the GHBP to the antibody. Reports of interference plex and in part variable among species; the principal in GH assays by the GHBP range from negligible regulators are ontogeny and development, nutri- to significant (Jan et al. 1991; Chapman et al. 1994; tion, gender/estrogen, and in rodents pregnancy. Jansson et al. 1997; Fisker et al. 1998). It is important Diagnostic use of the high affinity GHBP is currently to carefully examine each GH immunoassay for limited to genetic GH insensitivity (Laron) syn- GHBP interference as not all factors responsible for drome. GHBP can interfere in GH immunoassays, interference are known, and assay methodology and GH assays need to be optimized in this regard. α should be optimized to minimize GHBP effects. The low affinity GHBP is transformed 2- macroglobulin. It appears to have limited import- Conclusions ance for GH biology.

Two GHBPs exist in blood and other biological Acknowledgement fluids. The high affinity GHBP is the ectodomain of the GHR; it is either shed from cells by the action This chapter was supported in part by a merit re- of TACE, a member (ADAM-17) of the ADAM view grant from the Department of Veterans family of metalloproteinases, or directly secreted (in Affairs.

References

Amit, T., Barkey, R.J., Youdim, M.B. & Journal of Clinical Endocrinology and for the apportionment of hormone (and Hochberg, Z. (1990) A new and Metabolism 77, 33–39. receptor) into bound and free fractions convenient assay of growth hormone- Ayling, R.M., Ross, R., Towner, P. et al. in binding equilibria; or how to calculate binding protein activity in human (1997) A dominant-negative mutation bound and free hormone? Endocrinology serum. Journal of Clinical Endocrinology of the growth hormone receptor causes 124, 1101–1106. and Metabolism 71, 474–480. familial short stature. Nature Genetics 16, Baumann, G. (1995) Growth hormone Amit, T., Hertz, P., Ish-Shalom, S. et al. 13–14. binding to a circulating receptor (1991) Effects of hypo or hyper- Barnard, R., Quirk, P. & Waters, M.J. fragmentathe concept of receptor thyroidism on growth hormone-binding (1989) Characterization of the growth shedding and receptor splicing. protein. Clinical Endocrinology 35, hormone-binding protein of human Experimental and Clinical Endocrinology 159–162. serum using a panel of monoclonal and Diabetes 103(1), 2–6. Amit, T., Ish-Shalom, S., Glaser, B., antibodies. Journal of Endocrinology 123, Baumann, G. (2001) Growth hormone Youdim, M.B. & Hochberg, Z. (1992) 327–332. binding protein 2001. Journal of Pediatric Growth-hormone-binding protein in Barnard, R., Mulcahy, J., Garcia-Aragon, J. Endocrinology and Metabolism 14, patients with acromegaly. Hormone et al. (1994) Serum growth hormone 355–375. Research 37, 205–211. binding protein and hepatic GH binding Baumann, G. (2002) Genetic Amit, T., Dirnfeld, M., Barkey, R.J. et al. sites in the Lewis dwarf rat: effects of characterization of growth hormone (1993) Growth hormone-binding IGF-I and GH. Growth Regulation 4, deficiency and resistance: implications protein (GHBP) levels in follicular fluid 147–154. for treatment with recombinant from human preovulatory follicles: Barsano, C.P. & Baumann, G. (1989) growth hormone. American Journal of correlation with serum GHBP levels. Simple algebraic and graphic methods Pharmacogenomics 2, 93–111. growth hormone binding proteins 105

Baumann, G. & Frank, S.J. (2002) Baumbach, W.R., Horner, D.L. & Logan, protein-complex, and GH in healthy Metalloproteinases and the modulation J.S. (1989) The growth hormone-binding children. Journal of Clinical Endocrinology of growth hormone signalling. Journal of protein in rat serum is an alternatively and Metabolism 77, 356–361. Endocrinology 174, 361–368. spliced form of the rat growth hormone Carmignac, D.F., Wells, T., Carlsson, L.M., Baumann, G. & Shaw, M.A. (1988) receptor. Genes and Development 3, Clark, R.G. & Robinson, I.C. (1992) Immunochemical similarity of the 1199–1205. Growth hormone (GH) -binding protein human plasma growth hormone- Baxter, R.C. & Turtle, J.R. (1978) in normal and GH-deficient dwarf rats. binding protein and the rabbit liver Regulation of hepatic growth hormone Journal of Endocrinology 135, 447–457. growth hormone receptor. Biochemical receptors by insulin. Biochemical and Carmignac, D.F., Gabrielsson, B.G. & and Biophysical Research Communications Biophysical Research Communications 84, Robinson, I.C. (1993) Growth hormone 152, 573–578. 350–357. binding protein in the rat: effects of Baumann, G. & Shaw, M.A. (1990) A Berson, S.A. & Yalow, R.S. (1966a) Peptide gonadal steroids. Endocrinology 133, second, lower affinity growth hormone- hormones in plasma. Harvey Lectures 62, 2445–2452. binding protein in human plasma. 107–163. Cerio, R.J., Xing, F., Fatula, R.J. et al. Journal of Clinical Endocrinology and Berson, S.A. & Yalow, R.S. (1966b) State of (2002) Structurally distinct membrane- Metabolism 70, 680–686. human growth hormone in plasma and associated and soluble forms of GH- Baumann, G., Stolar, M.W., Amburn, K., changes in stored solutions of pituitary binding protein in the mouse. Journal of Barsano, C.P. & DeVries, B.C. (1986) growth hormone. Journal of Biological Endocrinology 172, 321–331. A specific growth hormone-binding Chemistry 241, 5745–5749. Chapman, I.M., Hartman, M.L., protein in human plasma: initial Black, R.A., Rauch, C.T., Kozlosky, C.J. Straume, M. et al. (1994) Enhanced characterization. Journal of Clinical et al. (1997) A metalloproteinase sensitivity growth hormone (GH) Endocrinology and Metabolism 62, disintegrin that releases tumour- chemiluminescence assay reveals lower 134–141. necrosis factor-α from cells. Nature 385, postglucose nadir GH concentrations Baumann, G., Amburn, K.D. & Buchanan, 729–733. in men than women. Journal of Clinical T.A. (1987a) The effect of circulating Blumenfeld, Z., Barkey, R.J., Youdim, Endocrinology and Metabolism 78, growth hormone-binding protein on M.B., Brandes, J.M. & Amit, T. (1992) 1312–1319. metabolic clearance, distribution, and Growth hormone (GH) -binding Clark, R.G., Mortensen, D.L., Carlsson, degradation of human growth hormone. protein regulation by estrogen, L.M. et al. (1996) Recombinant human Journal of Clinical Endocrinology and progesterone, and gonadotropins in growth hormone (GH)-binding protein Metabolism 64, 657–660. human: the effect of ovulation induction enhances the growth-promoting activity Baumann, G., Shaw, M.A. & Winter, R.J. with menopausal gonadotropins, GH, of human GH in the rat. Endocrinology (1987b) Absence of the plasma growth and gestation. Journal of Clinical 137, 4308–4315. hormone-binding protein in Laron- Endocrinology and Metabolism 75, Collipp, P.J., Kaplan, S.A., Boyle, D.C. & type dwarfism. Journal of Clinical 1242–1249. Shimizu, C.S.N. (1964) Protein-bound Endocrinology and Metabolism 65, Camarillo, I.G., Thordarson, G., Ilkbahar, human growth hormone. Metabolism 13, 814–816. Y.N. & Talamantes, F. (1998) 532–538. Baumann, G., Amburn, K. & Shaw, M.A. Development of a homologous Cramer, S.D., Barnard, R., Engbers, C., (1988) The circulating growth hormone radioimmunoassay for mouse growth Thordarson, G. & Talamantes, F. (1992) (GH)-binding protein complex: a major hormone receptor. Endocrinology 139, A mouse growth hormone-binding constituent of plasma GH in man. 3585–3589. protein RIA: concentrations in maternal Endocrinology 122, 976–984. Carlsson, B., Billig, H., Rymo, L. & serum during pregnancy. Endocrinology Baumann, G., Shaw, M.A. & Amburn, K. Isaksson, O.G. (1990) Expression of 130, 1074–1076. (1989a) Regulation of plasma growth the growth hormone-binding protein Dastot, F., Sobrier, M.L., Duquesnoy, P. hormone-binding proteins in health and messenger RNA in the liver and et al. (1996) Alternatively spliced forms disease. Metabolism 38, 683–689. extrahepatic tissues in the rat: co- in the cytoplasmic domain of the human Baumann, G., Shaw, M.A. & Buchanan, expression with the growth hormone growth hormone (GH) receptor regulate T.A. (1989b) In vivo kinetics of a covalent receptor. Molecular and Cellular its ability to generate a soluble GH- growth hormone-binding protein Endocrinology 73, R1–R6. binding protein. Proceedings of the complex. Metabolism 38, 330–333. Carlsson, L.M., Rowland, A.M., National Academy of Sciences of the United Baumann, G., Vance, M.L., Shaw, M.A. & Clark, R.G., Gesundheit, N. & Wong, States of America 93, 10 723–10 728. Thorner, M.O. (1990) Plasma transport W.L. (1991) Ligand-mediated Daughaday, W.H. & Trivedi, B. (1987) of human growth hormone in vivo. immunofunctional assay for Absence of serum growth hormone Journal of Clinical Endocrinology and quantitation of growth hormone- binding protein in patients with growth Metabolism 71, 470–473. binding protein in human blood. Journal hormone receptor deficiency (Laron Baumann, G., Lowman, H.B., Mercado, M. of Clinical Endocrinology and Metabolism dwarfism). Proceedings of the National & Wells, J.A. (1994) The stoichiometry 73, 1216–1223. Academy of Sciences of the United States of of growth hormone-binding protein Carlsson, L.M., Rosberg, S., Vitangcol, America 84, 4636–4640. complexes in human plasma: R.V., Wong, W.L. & Albertsson- Daughaday, W.H., Trivedi, B. & Andrews, comparison with cell surface receptors. Wikland, K. (1993) Analysis of 24-hour B.A. (1987) The ontogeny of serum GH Journal of Clinical Endocrinology and plasma profiles of growth hormone binding protein in man: a possible Metabolism 78, 1113–1118. (GH)-binding protein, GH/GH-binding indicator of hepatic GH receptor 106 chapter 8

development. Journal of Clinical distribution, turnover, and glycosylation Herington, A.C., Ymer, S. & Stevenson, J. Endocrinology and Metabolism 65, of the long and short growth hormone (1986b) Identification and 1072–1074. receptor isoforms in rat tissues. characterization of specific binding Duquesnoy, P., Sobrier, M.L., Duriez, B. Endocrinology 139, 2824–2830. proteins for growth hormone in normal et al. (1994) A single amino acid Gabrielsson, B.G., Carmignac, D.F., Flavell, human sera. Journal of Clinical substitution in the exoplasmic domain D.M. & Robinson, I.C. (1995) Steroid Investigation 77, 1817–1823. of the human growth hormone (GH) regulation of growth hormone (GH) Ho, K.K., Valiontis, E., Waters, M.J. & receptor confers familial GH resistance receptor and GH-binding protein Rajkovic, I.A. (1993) Regulation of (Laron syndrome) with positive GH- messenger ribonucleic acids in the rat. growth hormone binding protein in binding activity by abolishing receptor Endocrinology 136, 209–217. man: comparison of gel chromatography homodimerization. EMBO Journal 13, Gastier, J.M., Berg, M.A., Vesterhus, P., and immunoprecipitation methods. 1386–1395. Reiter, E.O. & Francke, U. (2000) Diverse Journal of Clinical Endocrinology and Edens, A., Southard, J.N. & Talamantes, F. deletions in the growth hormone Metabolism 76, 302–308. (1994) Mouse growth hormone-binding receptor gene cause growth hormone Hochberg, Z., Hertz, P., Colin, V. et al. protein and growth hormone receptor insensitivity syndrome. Human Mutation (1992) The distal axis of growth hormone transcripts are produced from a single 16, 323–333. (GH) in nutritional disorders: GH- gene by alternative splicing. Gelander, L., Bjarnason, R., Carlsson, binding protein, insulin-like growth Endocrinology 135, 2802–2805. L.M. & Albertsson-Wikland, K. (1998) factor-I (IGF-I), and IGF-I receptors in Eliakim, A., Brasel, J.A., Barstow, T.J., Growth hormone-binding protein levels obesity and anorexia nervosa. Mohan, S. & Cooper, D.M. (1998a) Peak over 1 year in healthy prepubertal Metabolism 41, 106–112. oxygen uptake, muscle volume, and the children: intraindividual variation and Holl, R.W., Snehotta, R., Siegler, B., growth hormone-insulin-like growth correlation with height velocity. Pediatric Scherbaum, W. & Heinze, E. (1991) factor-I axis in adolescent males. Research 43, 256–261. Binding protein for human growth Medicine and Science in Sports and Exercise Hadden, D.R. & Prout, T.E. (1964) A hormone: effects of age and weight. 30, 512–517. growth hormone binding protein in Hormone Research 35, 190–197. Eliakim, A., Brasel, J.A., Mohan, S., Wong, normal human serum. Nature 202, Iida, K., Takahashi, Y., Kaji, H. et al. W.L. & Cooper, D.M. (1998b) Increased 1342–1343. (1998) Growth hormone (GH) physical activity and the growth Hanaire-Broutin, H., Sallerin-Caute, B., insensitivity syndrome with high serum hormone-IGF-I axis in adolescent males. Poncet, M.F. et al. (1996) Effect of GH-binding protein levels caused by a American Journal of Physiology 275, intraperitoneal insulin delivery on heterozygous splice site mutation of the R308–R314. growth hormone binding protein, GH receptor gene producing a lack of Eliakim, A., Scheett, T.P., Newcomb, R., insulin-like growth factor (IGF) -I, and intracellular domain. Journal of Clinical Mohan, S. & Cooper, D.M. (2001) IGF-binding protein-3 in IDDM. Endocrinology and Metabolism 83, Fitness, training, and the growth Diabetologia 39, 1498–1504. 531–537. hormone-insulin-like growth factor I Harada, I., Tsutsumi, O., Momoeda, M. Iida, K., Takahashi, Y., Kaji, H. et al. (1999) axis in prepubertal girls. Journal of et al. (1997) Comparative concentrations Functional characterization of truncated Clinical Endocrinology and Metabolism 86, of growth hormone-binding protein in growth hormone (GH) receptor-(1-277) 2797–2802. maternal circulation, fetal circulation, causing partial GH insensitivity Fisker, S., Frystyk, J., Skriver, L. et al. (1996) and amniotic fluid. Endocrine Journal 44, syndrome with high GH-binding A simple, rapid immunometric assay for 111–116. protein. Journal of Clinical Endocrinology determination of functional and growth Hattori, N., Shimatsu, A., Kato, Y. & and Metabolism 84, 1011–1016. hormone-occupied growth hormone- Imura, H. (1990) Growth hormone and Irie, M. & Barrett, R.J. (1962) Immunologic binding protein in human serum. growth hormone binding protein in studies of human growth hormone. European Journal of Clinical Investigation human urine. Kidney International 37, Endocrinology 71, 277–287. 26, 779–785. 951–954. Jan, T., Shaw, M.A. & Baumann, G. (1991) Fisker, S., Ebdrup, L. & Orskov, H. (1998) Hattori, N., Kurahachi, H., Ikekubo, K. Effects of growth hormone-binding Influence of growth hormone binding et al. (1991) Effects of sex and age on proteins on serum growth hormone protein on growth hormone estimation serum GH binding protein levels in measurements. Journal of Clinical in different immunoassays. Scandinavian normal adults. Clinical Endocrinology 35, Endocrinology and Metabolism 72, Journal of Clinical and Laboratory 295–297. 387–391. Investigation 58, 373–381. Heinrichs, C., Yanovski, J.A., Roth, A.H. Jansson, C., Boguszewski, C., Rosberg, S., Forbes, G.B. (1977) Nutrition and growth. et al. (1994) Dexamethasone increases Carlsson, L.M. & Albertsson-Wikland, Journal of Pediatrics 91, 40–42. growth hormone receptor messenger K. (1997) Growth hormone (GH) assays: Frick, G.P., Tai, L.R. & Goodman, H.M. ribonucleic acid levels in liver and influence of standard preparations, GH (1994) Subcellular distribution of the growth plate. Endocrinology 135, isoforms, assay characteristics, and GH- long and short isoforms of the growth 1113–1118. binding protein. Clinical Chemistry 43, hormone (GH) receptor in rat Herington, A.C., Ymer, S., Roupas, P. & 950–956. adipocytes: both isoforms participate in Stevenson, J. (1986a) Growth hormone- Kaji, H., Nose, O., Tajiri, H. et al. (1997) specific binding of GH. Endocrinology binding proteins in high-speed cytosols Novel compound heterozygous 134, 307–314. of multiple tissues of the rabbit. mutations of growth hormone (GH) Frick, G.P., Tai, L.R., Baumbach, W.R. Biochimica et Biophysica Acta 881, receptor gene in a patient with GH & Goodman, H.M. (1998) Tissue 236–240. insensitivity syndrome. Journal of Clinical growth hormone binding proteins 107

Endocrinology and Metabolism 82, hormone binding protein in the rat. Mercado, M. & Baumann, G. (1994) A 3705–3709. Endocrinology 130, 3057–3065. growth hormone/prolactin-binding Keenan, B.S., Richards, G.E., Mercado, M. Maheshwari, H., Lillioja, S., Castillo, C.E., protein in human milk. Journal of Clinical et al. (1996) Androgen regulation of Mercado, M. & Baumann, G. (1995) Endocrinology and Metabolism 79, growth hormone binding protein. Growth hormone-binding protein in 1637–1641. Metabolism 45, 1521–1526. human lymph. Journal of Clinical Mercado, M., Molitch, M.E. & Baumann, Kratzsch, J., Blum, W.F., Ventz, M. et al. Endocrinology and Metabolism 80, G. (1992) Low plasma growth hormone (1995a) Growth hormone-binding 3582–3584. binding protein in IDDM. Diabetes 41, protein-related immunoreactivity in the Maheshwari, H., Sharma, L. & Baumann, 605–609. serum of patients with acromegaly is G. (1996) Decline of plasma growth Mercado, M., Carlsson, L.M., Vitangcol, regulated inversely by growth hormone hormone binding protein in old age. R. & Baumann, G. (1993) Growth concentration. European Journal of Journal of Clinical Endocrinology and hormone-binding protein determination Endocrinology 132, 306–312. Metabolism 81, 995–997. in plasma: a comparison of Kratzsch, J., Selisko, T. & Birkenmeier, G. Mandell, F. & Berenberg, W. (1974) The immunofunctional and growth (1995b) Identification of transformed Mauriac syndrome. American Journal of hormone-binding assays. Journal of α 2-macroglobulin as a growth hormone- Diseases of Children (1960) 127, 900–902. Clinical Endocrinology and Metabolism 76, binding protein in human blood. Journal Mannor, D.A., Winer, L.M., Shaw, M.A. & 1291–1294. of Clinical Endocrinology and Metabolism Baumann, G. (1991) Plasma growth Miell, J.P., Buchanan, C.R., Norman, 80, 585–590. hormone (GH)-binding proteins: effect M.R., Maheshwari, H.G. & Blum, W.F. Kratzsch, J., Keliner, K., Zilkens, T. et al. on GH binding to receptors and GH (1994) The evolution of changes in (1996) Growth hormone-binding protein action. Journal of Clinical Endocrinology immunoreactive serum insulin-like related immunoreactivity is regulated and Metabolism 73, 30–34. growth factors (IGFs), IGF-binding by the degree of insulinopenia in Martha, P.M., Jr., Reiter, E.O., Davila, N. proteins, circulating growth hormone diabetes mellitus. Clinical Endocrinology et al. (1992) The role of body mass in the (GH) and GH-binding protein as a 44, 673–678. response to growth hormone therapy. result of short-term dexamethasone Kratzsch, J., Dehmel, B., Pulzer, F. et al. Journal of Clinical Endocrinology and treatment. Journal of Endocrinology 142, (1997a) Increased serum GHBP levels in Metabolism 75, 1470–1473. 547–554. obese pubertal children and adolescents: Martha, P.M., Jr., Rogol, A.D., Carlsson, Mulumba, N., Massa, G., Ketelslegers, J.M. relationship to body composition, L.M., Gesundheit, N. & Blizzard, R.M. & Maes, M. (1991) Ontogeny and leptin and indicators of metabolic (1993) A longitudinal assessment of nutritional regulation of the serum disturbances. International Journal of hormonal and physical alterations growth hormone-binding protein in the Obsesity and Related Metabolic Disorders during normal puberty in boys. I. Serum rat. Acta Endocrinologica 125, 409–415. 21, 1130–1136. growth hormone-binding protein. Nixon, D.A. & Jordan, R.M. (1986) Kratzsch, J., Schreiber, G., Selisko, T. et al. Journal of Clinical Endocrinology and Conversion of CSF monomeric growth (1997b) Measurement of serum exon Metabolism 77, 452–457. hormone to large growth hormone with 3-retaining growth hormone-binding Martini, J.F., Pezet, A., Guezennec, C.Y. exposure to serum. Acta Endocrinologica protein in children and adolescents by et al. (1997) Monkey growth hormone 111, 289–295. radioimmunoassay. Hormone Research (GH) receptor gene expression. Pantel, J., Machinis, K., Sobrier, M.L. et al. 48, 252–257. Evidence for two mechanisms for the (2000) Species-specific alternative splice Leung, D.W., Spencer, S.A., Cachianes, G. generation of the GH binding protein. mimicry at the growth hormone et al. (1987) Growth hormone receptor Journal of Biological Chemistry 272, receptor locus revealed by the lineage of and serum binding protein: purification, 18 951–18 958. retroelements during primate evolution. cloning and expression. Nature 330, Massa, G., Mulumba, N., Ketelslegers, Journal of Biological Chemistry 275, 537–543. J.M. & Maes, M. (1990) Initial 18 664–18 669. Leung, K.C., Doyle, N. & Ho, K.K. (2000) characterization and sexual dimorphism Peeters, S. & Friesen, H.G. (1977) A growth Characterization of a low affinity of serum growth hormone-binding hormone binding factor in the serum binding protein for growth hormone in protein in adult rats. Endocrinology 126, of pregnant mice. Endocrinology 101, rat serum. Endocrinology 141, 138–145. 1976–1980. 1164–1183. Lim, L., Spencer, S.A., McKay, P. & Waters, Massa, G., Verhaeghe, J., Vanderschueren- Postel-Vinay, M.C., Cohen-Tanugi, E. & M.J. (1990) Regulation of growth Lodeweyckx, M. & Bouillon, R. (1993) Charrier, J. (1982) Growth hormone hormone (GH) bioactivity by a Normalization of decreased plasma receptors in rat liver membranes: effects recombinant human GH-binding concentrations of growth hormone- of fasting and refeeding, and correlation protein. Endocrinology 127, 1287–1291. binding protein by insulin treatment in with plasma somatomedin activity. Lobie, P.E., Barnard, R. & Waters, M.J. spontaneously diabetic BB rats. Hormone Molecular and Cellular Endocrinology 28, (1991) The nuclear growth hormone and Metabolic Research 25, 325–326. 657–669. receptor binding protein. Antigenic Mauras, N., Merimee, T. & Rogol, A.D. Postel-Vinay, M.C., Belair, L., Kayser, C., and physicochemical characterization. (1991) Function of the growth hormone- Kelly, P.A. & Djiane, J. (1991a) Journal of Biological Chemistry 266, insulin-like growth factor I axis in the Identification of prolactin and growth 22 645–22 652. profoundly growth-retarded diabetic hormone binding proteins in rabbit Lobie, P.E., Garcia-Aragon, J., Wang, B.S., child: evidence for defective target milk. Proceedings of the National Academy Baumbach, W.R. & Waters, M.J. (1992) organ responsiveness in the Mauriac of Sciences of the United States of America Cellular localization of the growth syndrome. Metabolism 40, 1106–1111. 88, 6687–6690. 108 chapter 8

Postel-Vinay, M.C., Tar, A., Hocquette, J.F. antagonist, B2036-PEG (pegvisomant), Endocrinology and Metabolism 70, et al. (1991b) Human plasma growth reveal effects of pegylation and evidence 417–420. hormone (GH)-binding proteins are that it binds to a receptor dimer. Journal Spencer, S.A., Hammonds, R.G., Henzel, regulated by GH and testosterone. of Clinical Endocrinology and Metabolism W.J. et al. (1988) Rabbit liver growth Journal of Clinical Endocrinology and 86, 1716–1723. hormone receptor and serum binding Metabolism 73, 197–202. Sadeghi, H., Wang, B.S., Lumanglas, A.L., protein. Purification, characterization, Rajkovic, I.A., Valiontis, E. & Ho, K.K. Logan, J.S. & Baumbach, W.R. (1990) and sequence. Journal of Biological (1994) Direct quantitation of growth Identification of the origin of the growth Chemistry 263, 7862–7867. hormone binding protein in human hormone-binding protein in rat serum. Tar, A., Hocquette, J.F., Souberbielle, J.C. serum by a ligand immunofunctional Molecular Endocrinology 4, 1799–1805. et al. (1990) Evaluation of the growth assay: comparison with Sanchez-Jimenez, F., Fielder, P.J., hormone-binding proteins in human immunoprecipitation and Martinez, R.R., Smith, W.C. & plasma using high pressure liquid chromatographic methods. Journal of Talamantes, F. (1990) Hypophysectomy chromatography gel filtration. Journal of Clinical Endocrinology and Metabolism 78, eliminates and growth hormone (GH) Clinical Endocrinology and Metabolism 71, 772–777. maintains the midpregnancy elevation 1202–1207. Roelen, C.A., Donker, G.H., Thijssen, J.H., in GH receptor and serum binding Touber, J.L. & Maingay, D. (1963) Koppeschaar, H.P. & Blankenstein, M.A. protein in the mouse. Endocrinology 126, Heterogeneity of human growth (1992) High affinity growth hormone 1270–1275. hormone. Its influence on a radio- binding protein in plasma of patients Scheett, T.P., Nemet, D., Stoppani, J. et al. immunoassay of the hormone in serum. with acromegaly and the effect of (2002) The effect of endurance-type Lancet 1, 403–405. octreotide treatment. Clinical exercise training on growth mediators Veldhuis, J.D., Johnson, M.L., Faunt, L.M., Endocrinology 37, 373–378. and inflammatory cytokines in pre- Mercado, M. & Baumann, G. (1993) Roelen, C.A., de Vries, W.R., Koppeschaar, pubertal and early pubertal males. Influence of the high-affinity growth H.P. et al. (1997a) Plasma insulin-like Pediatric Research 52, 491–497. hormone (GH) -binding protein on growth factor-I and high affinity growth Seidel, B., Glasow, A., Schutt, M. et al. plasma profiles of free and bound GH hormone-binding protein levels increase (2003) Association between the GH and on the apparent half-life of GH. after 2 weeks of strenuous physical receptor/exon 3 genotype and the level Modeling analysis and clinical training. International Journal of Sports of exon 3-positive GH-binding protein applications. Journal of Clinical Medicine 18, 238–241. in human serum. European Journal of Investigation 91, 629–641. Roelen, C.A., Koppeschaar, H.P., de Vries, Endocrinology 148, 317–324. Walker, J.L., Moats-Staats, B.M., Stiles, W.R. et al. (1997b) Visceral adipose tissue Silbergeld, A., Klinger, B., Keret, R. et al. A.D. & Underwood, L.E. (1992) Tissue- is associated with circulating high (1994) Serum growth hormone-binding specific developmental regulation of the affinity growth hormone-binding protein (GHBP) activity is decreased by messenger ribonucleic acids encoding protein. Journal of Clinical Endocrinology administration of insulin-like growth the growth hormone receptor and the and Metabolism 82, 760–764. factor I in three Laron syndrome siblings growth hormone binding protein in rat Roemmich, J.N. & Sinning, W.E. (1997) with normal GHBP. Proceeding of the fetal and postnatal tissues. Pediatric Weight loss and wrestling training: Society for Experimental Biology and Research 31, 335–339. effects on growth-related hormones. Medicine 206, 324–327. Wallace, J.D., Cuneo, R.C., Baxter, R. Journal of Applied Physiology 82, Silbergeld, A., Dastot, F., Klinger, B. et al. et al. (1999) Responses of the growth 1760–1764. (1997) Intronic mutation in the growth hormone (GH) and insulin-like Romero, G.S., Stephan, D.A., Sperling, hormone (GH) receptor gene from a girl growth factor axis to exercise, GH M.A. & Menon, R.K. (1996) Distinct with Laron syndrome and extremely administration, and GH withdrawal in sexual dimorphism in the effect of high serum GH binding protein: trained adult males: a potential test for hypothyroidism on the expression of the extended phenotypic study in a very GH abuse in sport. Journal of Clinical growth hormone receptor and growth large pedigree. Journal of Pediatric Endocrinology and Metabolism 84, hormone-binding protein gene in rat Endocrinology and Metabolism 10, 3591–3601. liver. Hormone Research 45, 273–278. 265–274. Wang, X., He, K., Gerhart, M. et al. Rosenfeld, R.G., Rosenbloom, A.L. & Smith, W.C. & Talamantes, F. (1988) (2002) Metalloprotease-mediated Guevara-Aguirre, J. (1994) Growth Gestational profile and affinity cross- GH receptor proteolysis and GHBP hormone (GH) insensitivity due to linking of the mouse serum growth shedding. Determination of extracellular primary GH receptor deficiency. hormone-binding protein. Endocrinology domain stem region cleavage site. Endocrine Reviews 15, 369–390. 123, 1489–1494. Journal of Biological Chemistry 277, Ross, R.J., Esposito, N., Shen, X.Y. et al. Smith, W.C., Kuniyoshi, J. & Talamantes, 50 510–50 519. (1997) A short isoform of the human F. (1989) Mouse serum growth hormone Weissberger, A.J., Ho, K.K. & Lazarus, L. growth hormone receptor functions as a (GH) binding protein has GH receptor (1991) Contrasting effects of oral and dominant negative inhibitor of the full- extracellular and substituted transdermal routes of estrogen length receptor and generates large transmembrane domains. Molecular replacement therapy on 24-hour growth amounts of binding protein. Molecular Endocrinology (Baltimore, Md) 3, 984–990. hormone (GH) secretion, insulin-like Endocrinology 11, 265–273. Snow, K.J., Shaw, M.A., Winer, L.M. & growth factor I, and GH-binding protein Ross, R.J., Leung, K.C., Maamra, M. et al. Baumann, G. (1990) Diurnal pattern of in postmenopausal women. Journal of (2001) Binding and functional studies plasma growth hormone-binding Clinical Endocrinology and Metabolism 72, with the growth hormone receptor protein in man. Journal of Clinical 374–381. growth hormone binding proteins 109

Woods, K.A., Fraser, N.C., Postel-Vinay, protein in rabbit serum. Molecular and Zhang, Y., Guan, R., Jiang, J. et al. (2001) M.C., Savage, M.O. & Clark, A.0J. (1996) Cellular Endocrinology 41, 153–161. Growth hormone (GH)-induced A homozygous splice site mutation Yu, Y.M., Domene, H.M., Sztein, J., dimerization inhibits phorbol ester- affecting the intracellular domain of Counts, D.R. & Cassorla, F. (1996) stimulated GH receptor proteolysis. the growth hormone (GH) receptor Developmental changes and differential Journal of Biological Chemistry 276, resulting in Laron syndrome with regulation by testosterone and estradiol 24 565–24 573. elevated GH-binding protein. Journal of of growth hormone receptor expression Zhou, Y., He, L. & Kopchick, J.J. (1994) Clinical Endocrinology and Metabolism 81, in the rabbit. European Journal of An exon encoding the mouse growth 1686–1690. Endocrinology 135, 583–590. hormone binding protein (mGHBP) Woods, K.A., Dastot, F., Preece, M.A. Zhang, Y., Jiang, J., Black, R.A., carboxy terminus is located between et al. (1997) Phenotype: genotype Baumann, G. & Frank, S.J. (2000) exon 7 and 8 of the mouse growth relationships in growth hormone TACE is a growth hormone binding hormone receptor gene. Receptor 4, insensitivity syndrome. Journal of Clinical protein (GHBP) sheddase: the 223–227. Endocrinology and Metabolism 82, metalloprotease TACE/ADAM-17 is Zhou, Y., He, L. & Kopchick, J.J. (1996) 3529–3535. critical for (PMA-induced) growth Structural comparison of a portion Ymer, S.I. & Herington, A.C. (1985) hormone receptor proteolysis and of the rat and mouse growth hormone Evidence for the specific binding of GHBP generation. Endocrinology 141, receptor/binding protein genes. Gene growth hormone to a receptor-like 4342–4348. 177, 257–259. Chapter 9

Resistance Exercise: Acute and Chronic Changes in Growth Hormone Concentrations

WILLIAM J. KRAEMER, BRADLEY C. NINDL AND SCOTT E. GORDON

Introduction cleaved from the extracellular domain of the GH receptor) adds further complexity to the nature and Resistance exercise is the most prolific form of spatial arrangement of circulating GH moieties. exercise producing anabolic effects in muscle and Thus, within the limitations for the scope of this connective tissues (Kraemer et al. 1996). Human chapter we will examine some of the basic responses growth hormone (GH), also called somatotropin, and adaptations of the 22 kDa GH to resistance a pleiotropic polypeptide, exerts a multitude of exercise and training. effects upon the metabolic state of the human body, in part mediating a myriad of metabolic and growth Growth hormone release and control processes. The complexity of pituitary release and production of GH, aggregates, and binding proteins Many pituitary hormones are under hypothalamic is just beginning to be elucidated with the inter- control. The hypothalamic hormones (e.g. releas- pretation of the literature on resistance exercise and ing factors) growth hormone releasing hormone training remaining a puzzle with parts of the picture (GHRH) and somatostatin (SS) serve to stimulate still requiring study (Nindl et al. 2003). Neverthe- and inhibit GH release, respectively. These releas- less, understanding the responses and adaptations ing factors are secreted from neurons originating in of the 22 kDa GH will continue to play an important the arcuate, periventricular and paraventricular part of our understanding of the adaptations to nuclei (i.e. hypothalamic nuclei) whose axon ter- resistance training. The hGH-N gene expresses the minals extend toward the median eminence. The main pituitary molecular weight variant in the GH releasing factors are delivered into a portal system family which is the 22 kDa form. This protein is 191 of veins which acts as a humoral pathway for direct amino acids in length (monomeric 22 kDa repres- delivery to the anterior pituitary. The portal blood ents ~ 21% of all circulating plasma GH). The next system originates from capillary loops in the median most prevalent form is the 20 kDa molecule (mono- eminence of the tuber cinereum and blood drains in meric 20 kDa represents ~ 6% of all circulating a parallel fashion down the pituitary stalk. In the plasma GH) formed through alternative mRNA anterior pituitary, the portal blood vessels break splicing during which amino acid residues 32–46 up into the sinusoids which provide the nutrient are cleaved out. Human GH can also undergo post- supply. This vascular supply is very small and the translational modification and peripheral tissue capillary permeability is quite high. In this manner, proteolytic cleavage at its site of action to form releasing factors from the hypothalamus may reach variants and aggregates (i.e. dimers, trimers, penta- secreting cells (i.e. somatotrophs) of the anterior mers) and fragments which exist in the circulation. pituitary (Fig. 9.1). Functionally distinct types (i.e. The existence of high- and low-affinity GH bind- type I and II or band 1 and band II cells) of somato- ing proteins (released from the pituitary and also trophs have been identified using density gradient 110 growth hormone and resistance exercise 111

Arcuate nuclei Paraventricular nuclei Hypothalamic neurons Supraoptic nucleus

Release of hypothalamic releasing factors Hypothalamus

Stalk Portal vessels Median eminence

Secreting cell of anterior pituitary (Somatotroph) Fig. 9.1 Hypothalamic–pituitary neural interface regulating growth Anterior pituitary hormone (GH) secretion. Posterior pituitary centrifugation. The somatotrophs in these two GH pulse amplitude is supported by studies that fractions can be distinguished morphologically by have reported that the GH response to GHRH is their staining characteristics and ultrastructure, and enhanced by administration of SS antagonists (viz. have been shown to vary in their responsiveness pyridostigmine [an acetylcholinesterase inhibitor] to GH secreting stimuli with the type I being more and hexapeptide GH-releasing peptides) (Cappa et al. responsive (Snyder et al. 1977). The physiological 1993). The cloning of a GH secretagogue receptor significance of this cellular heterogeneity and has also demonstrated that the GH secretagogue, the whether exercise can differentially influence these previously elusive endogenous ligand of the recep- cells remains unknown, but presents a provocative tor (recently identified as ghrelin), is part of a new area of future study as the complexities of exercise- physiological regulating system in GH secretion. induced release and adaptations interact with the It has been well established that GH is secreted heterogeneity of the different molecular weight from the anterior pituitary gland in an episodic, forms of GH and GH aggregates. pulsatile manner throughout the day with a dramatic The central mechanisms underlying pulsatile GH surge of release during slow-wave sleep (Fig. 9.2). release are classically thought to involve elevated This pulsatile release is under the regulatory control secretion of GHRH into the hypophysial portal of the two hypothalamic hormones that serve to blood during troughs or nadirs of SS secretion. The stimulate (i.e. growth factor releasing hormone) or precise molecular mechanisms responsible for this inhibit (i.e. SS) GH release. The balance between pulsatility remain speculative. Superimposed on these two hormones determines the relative magni- this basic mechanism are other control and modu- tude of initial release from the anterior pituitary. latory mechanisms that determine somatotroph It does appear that GH can exert acute negative responsiveness at any give point in time. There is feedback on its own release. When subjects are evidence that GHRH affects both GH biosynthesis administered a single dose of GH, subsequent GH and release, and SS inhibits release without affect- responses to GHRH are diminished or abolished ing biosynthesis. It is also thought that GHRH is (Scanlon et al. 1996). This ‘somatotroph desensitiza- required for the initiation of GH pulses, while SS tion’ mechanism can be reversed by prior activation dictates the amplitude of the pulse. SS control over of cholinergic pathways. Currently, it appears that 112 chapter 9

) 24 –1 22 20 18 16 14 12 10 8 6 4 Fig. 9.2 Example 13-h growth 2 hormone (GH) pulsatility profile

Growth hormone (ng·mL 0 for a young health male. Note that

100 300 500 the greatest pulse amplitude 1700 1900 2100 2300 occurred during sleeping hours Time of day (h) (approximately 0200 h). alterations in GH release are mediated through an issue. The interplay between exercise and many of inhibition of SS secreting cells via cholinergic path- these factors in the control of GH release is not fully ways (Giustina & Veldhuis 1998). As noted above SS understood. inhibits GH release, but not its synthesis. This is an important concept as it may explain ‘rebound’ GH Growth hormone release patterns secretion after SS priming and withdrawal (Giustina & Veldhuis 1998). Two unique characteristics of GH are its pulsatile There are many neuromodulators (i.e. neuropep- release and the degree of molecular heterogeneity. tides, neurotransmitters, physiological conditions) The GH-N gene expresses the main pituitary mole- that have putative roles in regulating GH secretion cular weight variant in the GH family, which is the via GHRH and SS have been an intense area of regu- 22 kDa form. This protein is 191 amino acids in latory mechanism research. Veldhuis et al. (2004) length with two disulfide cross-linkages (mono- have postulated ‘that sex-steroid-specific control meric 22 kDa represents ~ 21% of all circulating of SS and GHRH outflow may mediate the for- plasma GH). The next most prevalent form is the mer gender contrasts, whereas unknown (gender- 20 kDa molecule (monomeric 20 kDa represents independent) factors may determine the capability ~ 6% of all circulating plasma GH) formed through of exercise to significantly antagonize GH auto- alternative mRNA splicing during which amino inhibition’. Table 9.1 lists a summary of these acid residues 32–46 are cleaved out. Human GH can known modulators. These factors act to stimulate also undergo post-translational modification and α ( 1-adrenergic, amino acids, dopamine, muscarinic peripheral tissue proteolytic cleavage at its site of cholinergic, GABA(-B), galanin, growth hormone action to form variants and aggregates (i.e. dimers, releasing peptide (GHRP), histamine, hypoglycemia, trimers, pentamers) and fragments which exist in neuromedin C, opiates, serotonin, diabetes, acute the circulation. The existence of high- and low- and chronic exercise, starvation and stress) or inhi- affinity GH binding proteins (released from the α bit ( 1-adrenergic, GHRH immunization cortisol/ pituitary and also cleaved from the extracellular glucocorticoids, glucose, hypothyroidism, obesity, domain of the GH receptor) adds further complex- aging) GH release. Peripheral feedback regulation ity to the nature and spatial arrangement of circulat- of the somatotroph is mediated by the array of GH ing GH moieties. This molecular heterogeneity target actions. Namely, insulin-like growth factor I appears to have physiological significance as the (IGF-I), glucose and free fatty acids can each exert different forms have been shown to possess dif- feedback influences at hypothalamic/pituitary lev- ferent biological activities (e.g. relative potency in els. The array of multivariate neuroregulatory fac- bioassays) as well as their ability to be detected in tors shown in Table 9.1 that influence GH release immunoassays. The contents of the band I and band emphasize the concept that GH release is a complex II secretory cells in the pituitary can impact the growth hormone and resistance exercise 113

Table 9.1 Neuroregulatory modulators of growth hormone (GH) secretion in humans.

Effector Effect

α α β ↑ ↓ Adrenergic ( 1, 2 & 2) No effect, , Age ↓ Amino acids ↑ Autofeedback at hypothalamus by IGF-I ↓ Bombesin No effect basally, ↓ hypoglycemia effect Dopamine ↑ Muscarinic (chlolinergic & nicotinic) ↑ & ↓ Cortisol/glucocorticoids ↓ Diabetes mellitus (type 1 & type 2) ↑ & ↑ or ↓ Estrogen ↑ Amplitude Excitatory amino acids Unknown Exercise (acute & chronic) ↑ & ↑ Fatty acids ↓ GABA(-B) ↑ Basal ↓ (stim) Galanin ↑ GHRP ↑ Glucose ↓ Histamine ↑ Hypoglycemia ↓ Hypothryrodisim ↓ IGF-I (pituitary inhibition) Yes Immunization (or antagonist) (GHRH & SS) ↓ Amplitude & unknown Leptin Inversely correlated with GH Neuromedin C Unknown Neuropeptide Y ↓? Nitrc oxide No effect Obesity ↓ Opiates ↑ Senescence/aging ↓ Serotonin ↑ Starvation ↑ Stress (shock, restraint, endotoxin, psychological) ↑ Testosterone ↑ TRH No effect DHT No effect

DHT, dihydrotestosterone; GHRH, growth hormone releasing hormone; GHRP, growth hormone releasing peptide; GNRH, growth hormone releasing hormone; IGF-I, insulin-like growth factor I; SS, somatostatin; TRH, thyrotropin releasing hormone. profile of the composition of the circulatory concen- (e.g. gonadal steroids, thyroid hormones and IGF-I), trations of various GH isoforms and may eventually adiposity, fitness level and exercise have all been provide the insights into the vast array of physiolo- implicated as factors that influence GH concentra- gical mechanisms related to GH, yet not explained tions in the blood (see Fig. 9.1). Since the initial by the 22 kDa molecule alone (Hymer et al. 2001). observations of Hunter et al. (1965), it is well recog- nized that physical activity is a naturally occurring Growth hormone secretion stimulator of GH release into the circulation. GH has been linked to the promotion of anabolism in Many factors affecting GH secretion including age, both muscle and connective tissue. Specifically, it gender, diet and nutrients, stress, other hormones enhances cellular amino acid uptake and protein 114 chapter 9

synthesis in skeletal muscle, resulting in hypertrophy 2 The resistance loading used in the exercise. of both muscle fiber types (Noall et al. 1957; Ullman 3 The volume of exercise performed. & Oldfors 1989; Crist et al. 1991). Cartilage growth 4 The amount of rest between sets and exercises. and bone deposition can also be stimulated by GH, The activation of an adequate amount of muscle tis- increasing bone mineral density (BMD) and other sue is vital to increase plasma concentrations of GH. markers of bone formation (Isaksson et al. 1990; van The amount of tissue that is activated is influenced der Veen & Netelenbos 1990; Parfitt 1991; Bikle et al. by the resistance used, the total work and the type 1995; Orwoll & Klein 1995). While debatable with of exercise performed (e.g. small muscle group the re-emergence of the larger family of GH(s) and exercise vs. large muscle group exercise). The first GH binding proteins, classic dogma has proposed data supporting this paradigm were observed by that many of the anabolic effects of GH are mediated Vanhelder et al. (1984). In their study, GH signific- via IGF-I secreted by the liver and other tissues antly increased above rest after performing seven (Florini et al. 1996). While this may be true if GH is sets with a resistance of 85% of the 7-repetition only defined as the 22 kDa isoform, such a paradigm maximum (RM) in the squat exercise. However, may have to be re-examined if the GH superfamily when the resistance was reduced to 28% of the 7-RM of polypeptides and binding proteins are now con- while keeping rest periods the same and equating sidered. Nevertheless, exercise-induced GH release the total work, no changes in circulating concentra- is in part responsible (either directly or indirectly) tions of GH were observed after the exercise pro- for the anabolic effects of exercise. Furthermore, tocol. Based on ‘size principle’, the number of motor there is evidence in rats that GH may stimulate the units required to perform the 28% load was less than autocrine/paracrine production of IGF-I by skeletal needed for the 85% load. Thus, activation of enough muscle, cartilage, and bone cells themselves (Turner tissue appears to be a vital element of the exercise et al. 1988; Isaksson et al. 1990; Bikle et al. 1995). stimulus to elicit a significant GH response. Lastly, GH may interact both directly and indirectly The volume of exercise or total work has also been with androgens (Jørgensen et al. 1996), estrogens implicated in the magnitude of response. When (Holmes & Shalet 1996) and thyroid hormone each of the acute program variablesachoice of (Weiss & Refetoff 1996) with respect to secretion and exercise, order of exercise, rest periods, resistance target tissue actions. usedaare kept constant and only when the number of sets performed is increased thereby producing Resistance exercise-induced changes in more total work, is the response of GH increased. growth hormone concentrations This effect has been demonstrated in both men and women using whole body, multiexercise resistance It has become apparent over the past 15 years that protocols and 10-RM loadings (Mulligan et al. 1996; different exercise protocols will result in different Gotshalk et al. 1997). In addition, a higher level of concentrations of GH. To understand the composite strength fitness can also allow an individual to per- factors that may mediate such differential response form greater amounts of total work resulting in a to the myriad of different exercise protocols possible higher GH response as well (Ahtiainen et al. 2003). with resistance training, one must look at some of Little is known about the thresholds of total work the key factors that result from choices made in the to elicit an increase in GH in the circulation, but acute program variables in workout design (Fleck & likely would be interactive with the other elements Kraemer 2004). It is also apparent that interaction of of a workout protocol (e.g. length of rest periods various exercise factors is vital in determining the amount of tissue activated, and resistance used). magnitude of the GH response. The key external The interactive effects of different choices made factors that interact and produce an increased con- for the acute program variables (i.e. choice of exer- centration of GH in the circulatory biocompartment cise [e.g. large or small muscle groups], order of are: exercise [e.g. large muscle group first or small mus- 1 The amount of muscle mass recruited. cle groups first], rest period length [short duration growth hormone and resistance exercise 115

or long duration], resistance used [e.g. 5-RM or tion (Gordon et al. 1994). Reductions in the rest 10-RM] and number of sets or total work [low J or periods between sets of exercises with whole body high J]) in designing a resistance exercise session workouts have been shown to produce the most can result in a different GH response. While the dramatic increases in lactate responses with resist- combinations that could be used are many, prior ance exercise (Kraemer et al. 1990, 1993). However, research has shown that when comparing multi- reducing the rest period length will also impact the exercise, total body resistance protocols the three amount of resistance that can be lifted (Kraemer factors that appear to impact GH most dramatically et al. 1987). Thus, there appears to be a crucial modu- are the combination of high total work, short rest lation of the amount of resistance used and then period length (1 min between sets and exercise) and amount of tissue activated that drives the GH res- a moderately heavy 10-RM resistance in both men ponse. However, Takarada et al. (2000) has shown and women (Kraemer et al. 1990, 1993). that occlusion of the arm can have a dramatic effect Since any resistance training session uses a wide on GH resulting in significant increases with relat- array of different combinations of the acute exercise ively low intensity (20% of 1-RM) while no changes variables, GH responses are a function of these in GH were observed without occlusion. One may acute program variable choices. These appear to be conclude that for the 22 kDa molecule, hypoxia and the underlying governing principles for the crea- disruption of acid-base balance plays a major regu- tion of the exercise stimulus in resistance training. latory role in the stimulation of GH (Sutton 1983). The most dramatic finding related to these variable Resistance exercise workouts that have short rest structures and GH is the impact they have on periods (1 min rest between sets and exercises, mod- acid-base balance, which in turn appears to have a erate intensities [8–10-RM load ranges] and whole major role in stimulating the release of GH into the body workout protocols [8–10 exercises]) can pro- circulation. Each of the above four factors can be duce such physiological demands and results in the manipulated to produce an effect on metabolism most dramatic GH changes in the blood (Fig. 9.3). that either limits or promotes the accumulation The majority of these studies have examined the of hydrogen ions and decreases in blood pH, which short-term recovery responses (typically < 2 h) of in turn account for almost half of the shared vari- GH. The impact of GH pulsatile release at different ance with GH production. This makes acid-base times of day and GH roles over different phases of decreases (i.e. increase in ATP hydrolysis, decrease recovery from resistance exercise remains to be in pH, increase in hydrogen ion) prime determin- elucidated. McMurray et al. (1995) presented data ants of the amount of 22 kDa isoform in the circula- which utilized a resistance exercise protocol as the

25

20

15 5-RM/3 min 10-RM/1 min 10-RM/3 min 10 Fig. 9.3 GH responses to different rest period lengths with different resistance exercise protocols. The 10/1 post-exercise responses 5 p < 0.05 from the other protocols at each time point. All post-exercise concentrations are significantly 0 (P < 0.05) above resting values. Rest Mid 0 15 30 60 116 chapter 9

exercise perturbation on nocturnal GH responses. and readily released. These data also demonstrated The exercise protocol consisted of performing 3 × that enhanced IGF-I inhibitory feedback on pitu- 6–8-RM sets of six different exercises (18 sets total). itary GH release was unlikely because serum IGF-I Blood samples were obtained from 2100 to 0700 h. concentrations did not differ between the control The investigators did not observe any effect of and exercise conditions. It is well know that an array resistance exercise on nocturnal GH release. Con- of other metabolic and hormonal signals, such as versely, Nindl et al. (2001) examined GH pulsatile changes in GHRH, hexapeptides, or Ghrelin release, GH release over a longer period of time. The acute could also have mediated the observed GH response. heavy resistance exercise protocol began at 1500 h and was designed to be a high-volume workout that Influence of age included 50 total sets and recruited and activated a large amount of muscle tissue. Blood was sampled It has been shown that the acute GH response is every 10 min from 1700 to 0600 h under control and somewhat limited in older individuals (Craig et al. resistance exercise conditions. The results of the 1989; Pyka et al. 1992; Kraemer et al. 1999). A major study demonstrated that heavy resistance exercise factor contributing to this limited GH response in the late afternoon decreased overnight maximum maybe the magnitude of exertion displayed and the GH concentrations and GH pulse amplitude; how- inability to perform as much total work in an exer- ever, overall mean GH concentrations were not sig- cise protocol. It was reported in an investigation by nificantly reduced. Acute heavy resistance exercise Pyka et al. (1992) that lower blood lactate concentra- differentially influenced the temporal pattern of tions in the elderly during resistance exercise sup- the overnight release because GH was lower during ported the theory that the amount of effort exerted the first half of sleep, but greater during the second in a resistance exercise session may impact the half for the exercise versus control conditions. The resulting GH response. With the negative influence mean maximum GH concentration and mean pulse of aging on buffering capacities and toleration of amplitude were lower in the exercise vs. the control acidosis a combination of factors could help explain condition. the reductions in exercise-induced GH following Several possible explanations for this finding acute resistance exercise (Godfrey et al. 2003). Short- were speculated in that such results could be term training (10–12 weeks) does not appear to alter mediated by an increase in SS tone after exercise. this response (Craig et al. 1989; Kraemer et al. 1999). Although SS inhibits GH release, it does not negat- Ten-weeks periodized resistance training program ively affect GH biosynthesis. This may be an import- showed no significant changes in GH for resting ant concept because it may explain a ‘rebound’ in or exercise-induced concentrations for younger GH secretion after SS priming/withdrawal. The or older subjects (Kraemer et al. 1999). The lactate nocturnal peaks were lower for the exercise vs. response to the acute resistance exercise was lower control conditions during 2300–0300 h, but higher in the older than in the younger men with use of the during 0300–0600 h. Thus, even though the mean same relative resistance, while increases in lactic GH concentration was similar between the control acid do not change the acid-base balance (Robergs and exercise conditions, the temporal pattern of GH et al. 2004), it acts as an indication of reduced meta- release was clearly influenced by daytime exercise. bolic demands and a lower disruption of acid-base From a mechanistic perspective, the acute heavy status. This may partially explain the lower post- resistance exercise bout may have resulted in an exercise GH values in older individuals. Reductions elevated SS tone during 2300–0300 h. During this observed in resting and exercise-induced concentra- time, GH release was inhibited to some degree and tions of the 22 kDa GH concentrations do not appear concurrently; GH biosynthesis was not inhibited. to be changed with short-term training (Fig. 9.4). At ~ 0300 h, this SS tone was withdrawn, and GH There is only a rudimentary understanding of molecules biosynthesized and stored during the time the effects of exercise and aging on the physiolo- when GH release was inhibited were then available gical mechanisms underlying ‘somatopause’ (i.e. the growth hormone and resistance exercise 117

20 30 years 62 years 15 30 years 62 years * ) * * –1 * )

–1 * * 15 * * # * * * * # # 10 # * * * * 10 # # * * * # * * * 5 5 # Serum lactate (mmol·L Serum growth hormone ( µ g·L 0 0 Pre IP 5 15 30 Pre IP 5 15 30 Pre IP 5 15 30 Pre IP 5 15 30 (a) Pretraining Post-training (b) Pretraining Post-training

Fig. 9.4 Responses (means + SD) of lactate (a) and growth hormone (b) after acute heavy resistance exercise test (AHRET) before and after 10 weeks of periodized strength and power training for 30-year-old and 62-year-old men. *, Significantly different (P ≤ 0.05) from corresponding pre-exercise value. #, Statistically significant difference (P ≤ 0.05) between 30-year- old and 62-year-old men. s, Significantly different (P ≤ 0.05) from corresponding pretraining value. decrease within the GH–IGF-I system). Because this in competitive lifters and body builders have shown endocrine axis is considered to be of great import- the same responses of the 22 kDa GH molecule and ance in maintaining the integrity of the muscu- show no changes in the hypopituitary axis for rest- loskeletal system, it is conceptually pragmatic to ing concentrations (Häkkinen et al. 1988; Ahtiainen suggest that resistance exercise regimens should et al. 2003) when compared to untrained or lesser attempt to target it. Keeping in mind that GH also trained individuals. These findings are consistent exhibits a great deal of molecular heterogeneity, and with dynamic feedback mechanisms and pulsatility the standard radioimmunoassay (RIA) focuses on of the GH molecule and the many roles it may play the 22-kDa variant, this is an especially important in the homeostatic control of other metabolic and point for future studies, when it is considered that repair processes. McCall et al. (1999) did observe a the higher molecular weight species of GH could correlations between resting GH and the magnitude possess greater biological activity and that the lack of type I and type II muscle fiber hypertrophy (r = of changes or reductions in the immunoreactive GH 0.62 to 0.74, respectively). These relationships could may not present the complete picture of the adapta- be indicative of a role for repeated acute resistance tional responses of GH variants to resistance exer- exercise-induced GH elevations on cellular adapta- cise training. In other words, measurement of just tions in trained muscle. Changes in receptor sensit- the 22 kDa isoform may not tell the whole story of ivity, differences in feedback mechanisms, IGF-I the adaptive response of pituitary secretions of GH. potentiation, binding protein interactions, stimula- tion of other molecular weight variants in the pitu- Training adaptations itary somatotrophs as well as diurnal variations may all interact to mediate 22 kDa GH responses. To date, resistance training does not appear to affect resting concentrations of the 22 kDa GH isoform. Gender effects Surprisingly, no changes in the resting concentra- tions of GH have been observed in response to resist- The exercise-induced responses of women com- ance training in men and women of various ages pared to men appear to be similar in the absolute (Kraemer et al. 1999; McCall et al. 1999; Häkkinen magnitude of the value achieved but the relative et al. 2000; Marx et al. 2001). Even long-term training changes from rest are smaller (Kraemer et al. 1991). 118 chapter 9

This also impacts the acute exercise responses. tained in the type of repetitions they were perform- Some protocols (e.g. 5-RM, 3 min rest periods or low ing with training. With detraining the responses total work protocols) which may only marginally were muted to a similar degree across all groups. elevate GH concentrations in the circulation in men These data indicate that GH secretion may be sensit- result in no significant exercise-induced elevations ive to the specific muscle actions used during resist- in women due to the higher resting concentrations ance training. Such a response is supported by the (Kraemer et al. 1991, 1993). How such lower magni- relatively new construct as it has been shown that tudes of GH responses to acute exercise influence the anterior pituitary may be directly innervated by the adaptations of target tissues remains unknown. nerve fibers mostly with synapses on corticotroph It has been postulated that women’s higher rest- and somatotroph cells (Ju 1999). It has also been ing concentrations potentially compensate for lower suggested that ‘neural–humoral’ regulation of GH levels of other anabolic hormones which may min- secretion may take place such that a rapid neural imize the role of the acute elevation with exercise response is observed during the initial stress with stress. These have been observed in the early follicu- the humoral phase subsequently occurring (Ju 1999). lar phase of the menstrual cycle. Our recent data If this is the case, then it may be possible for higher indicate that estrogen in the form of oral contracept- brain centers (e.g. motor cortex) to play an active ives may have minimal effects on the GH response role in regulating GH secretion during resistance to resistance exercise response (unpublished data). exercise and this regulatory mechanism may be While overt influences of the menstrual cycle may sensitive to specific muscle actions used during mediate some of these responses, differential pat- resistance training. terns of pituitary GH release may also explain the gender differences as the GH mass and mode of GH Physiological impact secretion may be regulated differently in men and women (Pincus et al. 1996). Due to the many actions of GH and its complexity as a biological effector, its influence on skeletal muscle Specificity of exercise hypertrophy and other tissue anabolism has just started to be dissected beyond simplistic direct and One of the underlying principles in biological sci- indirect hypotheses. A multitude of important inter- ence as well as a principle of resistance training is acting factors are thought to contribute to resistance ‘specificity’. This may be seen when examining the training-induced skeletal muscle hypertrophy, and response of GH to resistance exercise as well. In a some investigators have raised the possibility that study by Kraemer et al. (2001) four different groups the pronounced rise in blood GH concentration of subjects trained for 19 weeks performing all elicited by acute heavy resistance exercise may be concentric repetitions, concentric repetitions with included with these mechanisms. This hypothesis is double the volume, or a typical program of con- supported by observations that optimal heavy resist- centric and eccentric repetitions or no training (con- ance training-induced skeletal muscle hypertrophy trol group). Subjects performed two exercise tests is compromised in hypophysectomized rats unless consisting of three sets of 30 isokinetic concentric synthetic or pituitary-extracted GH is administered contractions in one testing session and three sets (Goldberg & Goodman 1969; Grindeland et al. 1994). of 30 isokinetic eccentric contractions with knee However, directly contradicting this theory is the extensions. The tests were separated by 48 h. The fact that an acute bout of aerobic exercise also pro- acute response to a concentric stimulus was similar, duces an equally large increase in blood GH concen- but when an eccentric protocol was performed the trations as an acute bout of heavy resistance exercise group that had trained with the typical concentric while aerobic exercise training has little or no net and eccentric contraction style had the highest GH anabolic effect on skeletal muscle tissue (Kraemer response to the eccentric challenge, indicating sensit- et al. 1995). Therefore, assuming the GH response to ivity to the specific eccentric stimulus that was con- exercise is important to muscle tissue hypertrophy growth hormone and resistance exercise 119

and other tissue anabolism, it may be speculated However, there are some animal data to suggest that acute heavy resistance exercise and acute that exercise-induced GH is important for somatic aerobic exercise differ not in their effect on serum and muscle growth. Using the Hamster model, immunoreactive GH concentrations but in their effect Katarina Borer has pioneered work into the role of on serum bioactive GH concentrations, an hypo- GH in exercise modulated growth by demon- thesis that has recently received much attention strating that exercised golden hamsters (an animal (Hymer et al. 2001). Furthermore, the difference may continual growth model) exhibited increased basal be due to activation of different motor neuron pools pulsatile GH, skeletal elongation and permanent and subsequent sequence of receptor events that are increase in body mass and reductions in fat mass in part stimulated by the electromechanical activa- when compared to sedentary controls (Borer 1989, tion process of activated muscle fibers. Alternat- 1995). Since the pulsatile release of GH is an integral ively, immunoreactive GH may stimulate pituitary component of agonist function, it would appear that release of higher molecular weight binding proteins GH pulsatility in humans cannot be ignored as well and aggregates into the circulation; each of these as the heterogeneity of the GH super family of hypotheses remains to be substantiated with direct molecules and aggregates. However, these were evidence (Nindl et al. 2003). female hamsters and development takes place only In addition, the pulsatile release of GH is import- at certain windows of a developmental time period ant for mediating linear growth. The growth process leading to maturation. is episodic and discontinuous, whether measured during a period of a day or over longer periods of Summary time. For example, in rodents, linear growth is high- est when GH secretory bursts are separated by Acute resistance exercise can be a potent stimulus about 3 h of very low GH levels, as in the case in the for the 22 kDa GH isoforms with increases related male rat, and is reduced when GH levels show small to the type of exercise protocol utilized. Training deviations around a relatively baseline, as in the case appears to have no significant impact on resting of the female rat. Thus, sexually dimorphic mechan- concentrations of these isoforms in men or women isms regulating the pulsatile release of GH between of different ages. However, changes in pulsatile males and females (high amplitude GH pulses and release with acute exercise may be one mechanism low interpulse [GH]) characterize GH profiles in by which an altered secretory profile may be medi- males, whereas female rats have less regular pulses ated and such systems appear to differ between men and higher interpulse [GH] are partly responsible and women. Feedback loops and interactions with for differences in growth rate (Slob & Van der Werff other hormonal systems (e.g. IGF, GH binding pro- Ten Bosch 1975; Jaffe et al. 1998). Additionally, the teins, stimulation of molecular aggregates) remain temporal pattern of GH release is also coupled complex and an avenue for future research as it to key enzymes responsible for longitudinal bone relates to target tissue responses. It is evident that growth and it may be important to view the GH the responses and adaptations of GH will become release on a pulsatile basis rather than a single static a more integrated variable as the physiological measure and this is well illustrated by considering context for its function and roles becomes further the sexual dimorphism in GH release. delineated.

References

Ahtiainen, J.P., Pakarinen, A., Kraemer, Bikle, D.D., Harris, J., Halloran, B.P. et al. (GH) secretion and somatic growth. In: W.J. & Häkkinen, K. (2003) Muscle (1995) The molecular response of bone Hormones and Sport, (Laron, Z. & Rogol, hypertrophy, hormonal adaptations and to growth hormone during skeletal A., eds.). Serono Symposia Publications, strength development during strength unloading: regional differences. New York: 21–40. training in strength-trained and Endocrinology 136, 2099–2109. Borer, K.T. The effects of exercise on untrained men. European Journal of Borer, K.T. (1989) Exercise-induced growth. (1995) Sports Medicine (Auckland, Applied Physiology 89, 555–563. facilitation of pulsatile growth hormone NZ) 20(6), 375–397. 120 chapter 9

Cappa, M., Grossi, A., Benedetti, S. et al. adaptations in athletes to strength Kraemer, W.J., Fleck, S.J., Dziados, J.E. (1993) Effect of the enhancement of the training in 2 years. Journal of Applied et al. (1993) Changes in hormonal cholinergic tone by pyridostigmine on Physiology 65, 2406–2412. concentrations after different heavy- the exercise-induced growth hormone Häkkinen, K., Pakarinen, A., Kraemer, resistance exercise protocols in women. release in man. Journal of Endocrinological W.J., Newton, R.U. & Alen, M. (2000) Journal of Applied Physiology 75(2), Investigation 16, 421–424. Basal concentrations and acute 594–604. Craig, B.W., Brown, R. & Everhart, J. (1989) responses of serum hormones and Kraemer, W.J., Patton, J.F., Gordon, S.E. Effects of progressive resistance training strength development during heavy et al. (1995). Compatability of high- on growth hormone and testosterone resistance training in middle-aged and intensity strength and endurance levels in young and elderly subjects. elderly men and women. Journals of training on hormonal and skeletal Mechanisms of Ageing and Development Gerontology. Series A, Biological Sciences muscle adaptations. Journal of Applied 49, 159–169. and Medical Sciences 55, B95–B105. Physiology 78(3), 976–989. Crist, D.M., Peake, G.T., Loftfield, R.B., Holmes, S.J. & Shalet, S.M. (1996) Role of Kraemer W.J., Fleck, S.J. & Evans, W.J. Kraner, J.C. & Egan, P.A. (1991) growth hormone and sex steroids in (1996) Strength and power training: Supplemental growth hormone alters achieving and maintaining normal bone physiological mechanisms of body composition, muscle protein mass. Hormone Research 45, 86–93. adaptation. Exercise and Sport Sciences metabolism and serum lipids in fit Hunter, W.M., Fonseka, C.C. & Passmore, Reviews 24, 363–397. adults: characterization of dose- R. (1965) Growth hormone: important Kraemer, W.J., Häkkinen, K. Newton, dependent and response-recovery role in muscular exercise in adults. R.U. et al. (1999) Effects of heavy- effects. Mechanisms of Ageing and Science 150(699), 1051–1053. resistance training on hormonal Development 58, 191–205. Hymer, W.C., Kraemer, W.J., Nindl, B.C. response patterns in younger vs. older Fleck, S.J. & Kraemer, W.J. (2004) Designing et al. (2001) Characteristics of circulating men. Journal of Applied Physiology 87, Resistance Training Programs, 3rd edn. growth hormone in women after acute 982–992. Human Kinetics, Champaign, IL. heavy resistance exercise. American Kraemer, W.J., Dudley, G.A., Tesch, P.A. Florini, J.R., Ewton, D.Z. & Coolican, S.A. Journal of Physiology. Endocrinology and et al. (2001) The influence of muscle (1996) Growth hormone and the insulin- Metabolism 281(4), E878–887. action on the acute growth hormone like growth factor system in myogenesis. Isaksson, O.G.P., Nilsson, A., Isgaard, J. & response to resistance exercise and Endocrine Reviews 17(5), 481–517. Lindahl, A. (1990) Cartilage as a target short-term detraining. Growth Hormone Giustina, A. & Veldhuis, J.D. (1998) tissue for growth hormone and insulin- and IGF Research 11, 75–83. Pathophysiology of the neuroregulation like growth factor I. Acta Paediatrica McCall, G.E., Byrnes, W.C., Fleck, S.J. & of growth hormone secretion in Scandinavica. Supplement 367, 137–141. Kraemer, W.J. (1999) Acute and chronic experimental animals and the human. Jaffe, C.A., Ocampo-Lim, B., Krueger, K. hormonal responses to resistance Endocrine Reviews 19(6), 717–797. et al. (1998) Regulatory mechanisms training designed to promote muscle Godfrey, R.J., Madgwick, Z. & Whyte, G.P. of growth hormone secretion are hypertrophy. Canadian Journal of Applied (2003) The exercise-induced growth sexually dimorphic. Journal of Clinical Physiology 24, 96–107. hromeon response in athletes. Sports Investigation 102, 153–164. McMurray, R.G., Eubank, T.K. & Medicine (Auckland, NZ) 33(8), 599–613. Jørgensen, J.O.L., Vahl, N., Hansen, T.B. Hackney, A.C. (1995) Nocturnal Goldberg, A.L. & Goodman, H.M. (1969). et al. (1996) Influence of growth hormonal responses to resistance Relationship between growth hormone hormone and androgens on body exercise. European Journal of Applied and muscular work in determining composition in adults. Hormone Research Physiology 72, 121–126. muscle size. Journal of Physiology 200, 45, 94–98. Marx, J.O., Ratamess, N.A., Nindl, B.C. et al. 655–666. Ju, G. (1999) Evidence for direct neural (2001) Low-volume circuit versus high- Gordon, S.E., Kraemer, W.J., Vos, N.H., regulation of the mammalian anterior volume periodized resistance training Lynch, J.M. & Knuttgen, H.G. (1994) pituitary. Clinical Experimental in women. Medicine and Science in Sports Effect of acid-base balance on the growth Pharmacology and Physiology. Supplement and Exercise 33(4), 635–643. hormone resonse to acute high-intensity 26, 757–759. Mulligan, S.E., Fleck, S.J., Gordon, S.E. et al. cycle exercise. Journal of Applied Kraemer, W.J., Noble, B.J., Clark, M.J. (1996) Influence of resistance exercise Physiology 76(2), 821–829. & Culver, B.W. (1987) Physiologic volume on serum growth hormone and Gotshalk, L.A., Loebel, C.C., Nindl, B.C. responses to heavy-resistance exercise cortisol concentrations in women. et al. (1997) Hormonal responses of with very short rest periods. International Journal of Strength and Conditioning multiset versus single-set heavy- Journal of Sports Medicine 8(4), 247–252. Research 10(4), 256–262. resistance exercise protocols. Canadian Kraemer, W.J., Marchitelli, L., Gordon, S.E. Nindl, B.C., Hymer, W.C., Deaver, D.R. & Journal of Applied Physiology 22(3), et al. (1990) Hormonal and growth factor Kraemer, W.J. (2001) Growth hormone 244–255. responses to heavy resistance exercise pulsatility profile characteristics Grindeland, R.E., Roy, R.R., Edgerton, V.R. protocols. Journal of Applied Physiology following acute heavy resistance et al. (1994). Interactive effects of growth 69(4), 1442–1450. exercise. Journal of Applied Physiology hormone and exercise on muscle mass Kraemer, W.J., Gordon, S.E., Fleck, S.J. et al. 91(1), 163–172. in suspended rats. American Journal of (1991) Endogenous anabolic hormonal Nindl, B.C., Kraemer, W.J., Marx, J.O., Physiology 267, R316–R322. and growth factor responses to heavy Tuckow, A.P. & Hymer, W.C. (2003) Häkkinen, K., Pakarinen, A., Alen, M., resistance exercise in males and females. Growth hormone molecular Kauhanen, H. & Komi, P.V. (1988) International Journal of Sports Medicine heterogeneity and exercise. Exercise and Neuromuscular and hormonal 12(2), 228–235. Sport Sciences Reviews 31(4), 161–166. growth hormone and resistance exercise 121

Noall, M.W., Riggs, T.R., Walker, L.M. & Scanlon, M.F., Isa, B.G. & Dieguez, C. Ullman, M. & Oldfors, A. (1989) Effects of Christensen, H.N. (1957) Endocrine (1996) Regulation of growth hormone growth hormone on skeletal muscle. I. control of amino acid transfer: secretion. Hormone Research 46, Studies on normal adult rats. Acta distribution of an unmetabolizable 149–154. Physiologica Scandinavica 135, 531–536. amino acid. Science 126, 1002–1005. Slob, A.K. & Van der Werff Ten Bosch, J.J. Vanhelder, W.P., Radomski, M.W. & Orwoll, E.S. & Klein, R.F. (1995) (1975) Sex differences in body growth in Goode, R.C. (1984) Growth hormone Osteoporosis in men. Endocrine Reviews the rat. Physiology and Behavior 14, responses during intermittent weight 16(1), 87–116. 353–361. lifting exercise in men. European Journal Parfitt, A.M. (1991) Growth hormone and Snyder, G., Hymer, W.C. & Snyder, D.J. of Applied Physiology and Occupational adult bone remodelling. Clinical (1977) Functional heterogeneity Physiology 53(1), 31–34. Endocrinology 35, 467–470. somatotrophs isolated from the rat van der Veen, E.A. & Netelenbos, J.C. Pincus, S.M., Gevers, E.F., Robinson, C. anterior pituitary. Endocrinology 101(3), (1990) Growth hormone (replacement) et al. (1996) Females secrete growth 788–799. therapy in adults: bone and calcium hormone with more process irregularity Sutton, J.R. (1983) The hormonal responses metabolism. Hormone Research 33 (suppl. than males in both humans and rats. to exercise at sea level and at altitude. 4), 65–68. American Journal of Physiology 270(1 Pt. 1), Progress in Clinical and Biological Research Veldhuis, J.D., Patrie, J., Wideman, L. et al. E107–E115. 136, 325–341. (2004) Contrasting negative-feedback Pyka, G, Wiswell, R.A. & Marcus, R. (1992) Takarada, Y., Nakamura, Y., Aruga, S. et al. control of endogenously driven and Age-dependent effect of resistance (2000) Rapid increase in plasma growth exercise-stimulated pulsatile growth exercise on growth hormone secretion in hormone after low-intensity resistance hormone secretion in women and men. people. Journal of Clinical Endocrinology exercise with vascular occlusion. Journal Journal of Clinical Endocrinology and and Metabolism 75, 404–407. of Applied Physiology 88(1), 61–65. Metabolism 89(2), 840–846. Robergs, R.A., Ghiasvand, F. & Parker, D. Turner, J.D., Rotwein, P., Novakofski, J. Weiss, R.E. & Refetoff, S. (1996) Effect of (2004) Biochemistry of exercise-induced & Bechtel, P.J. (1988) Induction of thyroid hormone on growth: lessons metabolic acidosis. American Journal of mRNA for IGF-I and -II during growth from the syndrome of resistance to Physiology. Regulatory, Integrative and hormone-stimulated muscle thyroid hormone. Endocrinology and Comparative Physiology 287(3), hypertrophy. American Journal of Metabolism Clinics of North America 25(3), R502–516. Physiology 255, E513–E517. 719–730. Chapter 10

The Growth Hormone Response to Acute and Chronic Aerobic Exercise

ARTHUR L. WELTMAN, LAURIE WIDEMAN, JUDY Y. WELTMAN AND

Introduction GH secretion declines by approximately 14% per decade after the age of 40 years (Rudman et al. 1981; Growth hormone (GH) is secreted by the anterior Zadik et al. 1985; Iranmanesh et al. 1991) and is pituitary in a pulsatile pattern. Multiple GH iso- markedly reduced in obesity, even in younger indi- types and oligomers exist in plasma in addition to viduals (Veldhuis et al. 1991, 1995). Whereas GH the predominant 22 kDa protein (Baumann 1991; production falls by 50% every 7 years in men begin- Lewis et al. 2000; Nindl et al. 2003). Minor isoforms ning in young adulthood (Iranmanesh et al. 1991, do not change uniquely in response to exercise. 1998; Veldhuis et al. 1995), the decrease is nearly GH activates cells by dimerizing receptors and trig- twofold less rapid in premenopausal women gering a cascade of phosphorylation reactions that (Asplin et al. 1989; Winer et al. 1990; Weltman, A. et signal to the nucleus. al. 1994; van den Berg et al. 1996). Many age-related The amount of GH secreted in each pulse is under physical adaptations resemble those recognized in physiological control by peptidyl agonists and ant- GH-deficient adults, including reduced muscle agonists (Arvat et al. 1998; Mueller et al. 1999; Farhy mass and exercise capacity, increased body fat espe- et al. 2001; Bowers 2002). Brain (hypothalamic) cially abdominal visceral fat, unfavorable lipid and growth hormone releasing hormone (GHRH) stimu- lipoprotein profiles, reduction in bone mineral dens- lates GH synthesis and secretion and somatostatin ity and cerebro and cardiovascular disease. Which (SS) inhibits GH release without affecting its syn- is cause and which is effect is difficult to ascertain, in thesis (Giustina & Veldhuis 1998; Hartman 2000). A that intra-abdominal adiposity and limited exercise growth hormone releasing peptide (GHRP), ghre- also predict reduced GH production (Vahl et al. lin, expressed in the stomach, anterior pituitary 1997; Clasey et al. 2001). Below we highlight the gland and hypothalamus amplifies GH secretion effects of exercise on GH release. Several compre- via cognate receptor codistributed with the pep- hensive reviews discuss other physiological factors tide (Giustina & Veldhuis 1998; Kojima et al. 1999; (Veldhuis 1996a, 1996b; Veldhuis et al. 1997; Giustina Hartman 2000). Transgenic inactivation of cent- & Veldhuis 1998; Hartman 2000; Veldhuis & Bowers ral–neural GHRP receptors reduces GH secretion 2003a, 2003b). by one-third in the mouse (Shuto et al. 2002). These three effector molecules govern GH secretion by Acute aerobic exercise and growth convergent mechanisms (Veldhuis & Bowers 2003b). hormone secretion Many of the metabolic effects of GH are mediated by insulin-like growth factor I (IGF-I), which is Aerobic exercise stimulates GH release within synthesized in the liver and all nucleated cells under approximately 15 min and induces peak values at or the control of GH and tissue-specific hormones near the end of exertion (Lassarre et al. 1974; Sutton (Giustina & Veldhuis 1998). & Lazarus 1976; Kozlowski et al. 1983; Raynaud et al. 122 growth hormone and exercise 123

1983; Bunt et al. 1986; Chang et al. 1986; Felsing et al. (Pritzlaff et al. 1999; Pritzlaff-Roy et al. 2002). By 1992; Luger et al. 1992; Weltman, A. et al. 1992, 1994; simple linear regression analysis, women had a Cappon et al. 1994; Chwalbinska-Moneta et al. 1996; higher intercept value (greater baseline GH secre- Pritzlaff et al. 1999, 2000; Wideman et al. 1999, tion) and slope (accentuated sensitivity) to aerobic 2000a). The intensity and duration of the aerobic exercise (Fig. 10.2). stress, physical fitness, gender and age all influence The extent that these acute relationships apply to the GH response to exercise. Although the threshold graded chronic exercise training intensities is not of exercise intensity was envisioned earlier (Chang known. et al. 1986; Felsing et al. 1992; Chwalbinska-Moneta Exercise stimulates less GH secretion in healthy et al. 1996), randomly ordered separate day studies older individuals than in young individuals affirm that exercise intensity predicts GH secretion (Weltman, A. et al. 2000b, 2001). Analogous exercise in a linear dose–response fashion (Pritzlaff et al. intensity–GH response comparisons indicate that 2000; Pritzlaff-Roy et al. 2002). aging in man attenuates the graded effectiveness Women maintain higher GH concentrations than of exercise (slope term by 3.9-fold). An unresolved men at all ages, and manifest less orderly patterns of question raised by these outcomes is whether higher pulsatile GH release (Hartman et al. 1990; Veldhuis relative training intensities would be able to drive 1995; van den Berg et al. 1996; Pincus et al. 1996; young adult-like GH release in older men (Weltman, Engstrom et al. 1998; Giustina & Veldhuis 1998; A. et al. 2000b). Wideman et al. 1999). The time course of exercise- Exercise stimulates GH secretion 5.7–7.3-fold less induced GH release is similar by gender (Lassarre in post-menopausal women than in premenopausal et al. 1974; Bunt et al. 1986), but women differ in sev- women, whether or not post-menopausal individu- eral specific features; viz: anticipatory GH release als were receiving hormone replacement (Marcell before exercise and more rapid attainment of peak et al. 1999; Weltman, A. et al. 2001). Plausible mech- GH concentrations during exercise (Wideman et al. anistic bases for the consistent deficit in GH secre- 1999, 2000a; Pritzlaff-Roy et al. 2002). The effect tion in the older adult include excessive SS release of exercise is largely independent of circadian and diminished GHRH or possibly ghrelin drive. rhythmicity, since the time of day does not influence In fact, prolonged stimulation with GHRH or a the responses, at least in young men (Kanaley et al. synthetic GHRP (ghrelin surrogate) will elevate GH 2001). Young women and men achieve equival- secretion over 1–3 months in the elderly individual ent absolute GH concentrations during exercise (Evan et al. 2001; Richmond et al. 2001). (Wideman et al. 1999), but the fractional increase The GH response to exercise may also be blunted over baseline is higher in men (Bunt et al. 1986; in middle-aged men (Zaccaria et al. 1999). This infer- Wideman et al. 2000a). ence derives from study of a small group of young Figure 10.1 illustrates the impact of gender on (N = 5, age = 21 years) and middle-aged (N = 7, exercise intensity-dependent GH release in young age = 42 years) men who undertook incremental adults (Pritzlaff et al. 1999; Pritzlaff-Roy et al. 2002). exercise (50 watts every 3 min) until volitional In this study, men and women each undertook six exhaustion (Fig. 10.3). randomly ordered sessions (one control resting [C] Middle-aged men evidence lower and delayed and five exercise conditions [Ex]). Exercise com- peak GH release. If verified, diminutive GH prised 30 min of treadmill running at one of the responsiveness before later life would encourage following graded intensities (normalized to the investigations of earlier interventional strategies to individual lactate threshold [LT]): 25% and 75% of maintain favorable GH drive of anabolism. the difference between LT and rest (0.25 LT and 0.75 Relative adiposity and frank obesity suppress LT, respectively), LT, and 25% and 75% of the differ- baseline and stimulated GH production (Veldhuis o ence between LT and V 2peak (1.25 LT and 1.75 LT, et al. 1995; Weltman, A. et al. 2000b; Weltman, J.Y. respectively). GH responses to exercise increased et al. 2002). When compared to various pharmaco- progressively with increasing exercise intensity logic and physiologic stimuli, the magnitude of the 124 chapter 10

30 Males Control (N = 10) 0.25 LT 25 0.75 LT

) LT –1

L 1.25 LT . 20 1.75 LT µ g

15

10

Mean serum GH ( Onset of 5 exercise

0 7:00 7:40 8:20 9:00 9:40 10:20 11:00 11:40 12:20 13:00

30 Females Control (N = 8) 0.25 LT 25 0.75 LT

) LT –1

L 1.25 LT . 20 1.75 LT µ g Fig. 10.1 Mean serum growth 15 hormone (GH) concentrations monitored by sampling blood every Onset of 10-min for 6 h during rest (Control), exercise 10 and graded exercise defined as fractions (0.25, 0.75, 1.0, 1.25 and 1.75) Mean serum GH ( 5 of the individual lactate threshold (LT). Values are the mean ± SEM (N = 10 young men and eight young 0 women). (From Pritzlaff et al. 1999 7:00 7:40 8:20 9:00 9:40 10:20 11:00 11:40 12:20 13:00 and Pritzlaff-Roy 2002, used with Time permission.)

1600 ) − 1 Females (P = 0.02) = ·min X Slope 449 − 1 1200 X Intercept = 562

Fig. 10.2 Impact of gender on baseline (unstimulated) and exercise- 800 induced growth hormone (GH) secretion assessed over a graded range of intensity (see Fig. 10.1). The 400 Males (P = 0.08) higher slope of this relationship in X Slope = 277 women denotes a greater sensitivity X Intercept = 198 of the GH response to exercise, and Exercise & recovery (0900–1300 h) Integrated serum GH ( µ g·L the elevated intercept signifies higher 0 baseline (rest) GH secretion. (From 0 25 50 75 100 125 150 175 Pritzlaff-Roy 2002, used with Exercise intensity (% LT) permission.) growth hormone and exercise 125

80

70

60

50 ) − 1

40 GH ( µ g·L 30

20 Fig. 10.3 Growth hormone (GH) responses to maximal exercise at 10 exhaustion (E) and during recovery in young (diamonds) and middle- aged (circles) athletes before (open) 0 0 E and after (filled) a 4-month period of 60 180 300 600 900 training (Adapted from Zaccaria et al. 1200 1800 1999.) Seconds

GH response to exercise in obese subjects may exceed that induced by l-dopa or clonidine (Cordido et al. ) 15 − 1 1990; Tanaka et al. 1990), but not necessarily GHRH, pyridostigmine or l-arginine (Williams et al. 1984; ( µ g·L 5 Cordido et al. 1990, 1993; Maccario et al. 1997; Kelijman & Frohman 1998). Only GHRP and com- 0 (a) –1 0 1 3 5 bined secretagogues remain moderately (but not maximally) effective (Cordido et al. 1990, 1993; Maccario et al. 1997). Limited analyses of the inter-

) 15 action among gender, obesity and exercise suggest − 1 that abdominal visceral fat may be a key determin- ant of stimulated GH release (Clasey et al. 2001). ( µ g·L 5 Kanaley et al. (1999) examined exercise induced 0 GH secretion in cohorts comprising of non-obese, (b) –1 0 1 3 5 lower-body obese and upper-body obese (e.g. vis- cerally obese) women (Fig. 10.4). 15 ) − 1 Fig. 10.4 (right) Mean serum growth hormone (GH) concentrations for non-obese (a), lower body obese (b), ( µ g·L 5 and upper body obese (c) groups over the 6-h period of blood sampling. Exercise began at time zero and 0 o –1 0 60 180 300 continued for 30 min at an intensity of 70% V 2peak. (Adapted from Kanaley et al. 1999.) (c) Time (minutes) 126 chapter 10

400 4000

300 * 3000

* ) − 1

× min) 200 2000 − 1

(pmol·L * * ( µ g·L [Epinephrine] Integrated [GH] 100 1000

0 0 (a) Pre Week 3 Post (b) Pre Week 3 Post

30

20 *

) * − 1 Fig. 10.5 The effects of 6 weeks of training on integrated growth hormone (GH) concentration (a) and end exercise

(nmol·L 10 concentrations of epinephrine (b) and norepinephrine (c) in response to a constant load cycle ergometry exercise [Norepinephrine] bout (n = 6, Mean ± SD), Pre, Pretraining; Week 3, After 3 weeks of training; Post, After 6 weeks of training. 0 *P < 0.05 vs. pretraining. (From Weltman et al. 1997, used (c) Pre Week 3 Post with permission.)

The rank order of peak GH concentrations was µ –1 µ –1 Chronic endurance training and growth 13.7 g·L (non-obese), 6.8 g·L (lower body hormone release obese) and 3.5 µg·L–1 (upper body obese). Definit- ive radiological measurement of adipose-tissue Sustained endurance training limits acute exercise- topography will be needed to verify these anthro- induced GH release at the same absolute intensity pometric inferences. (Hartley et al. 1972; Weltman, A. et al. 1997). In

*

) ** − 1 10 000 *** ·min − 1 8000 Fig. 10.6 Effects of 1-year of run training on 24-h integrated C 6000 serum growth hormone (GH) @ LT > > concentrations. *1 year baseline LT > 4000 in lactate threshold (LT) group (P < 0.05); **at 1 year > LT group > Integrated (24 h) GH control (C) group (P < 0.05); ***at

concentration ( µ g·L 2000 1 year > LT group > @LT group Baseline 1 year (P < 0.05). (From Weltman et al. 1992, Time used with permission.) growth hormone and exercise 127

600 posity modulate long-term training effects is not yet established. In one study in older men (ages

) 50–78 years), serum IGF-I concentrations did not − 1

·L differ between marathon runners and age-matched − 1 sedentary subjects (Deuschle et al. 1998). In another 300 investigation, 1 year of exercise training (either supervised aerobic training [4 days a week] or

(6 h) (min· µ g supervised strength training [3 days a week]) in healthy older (aged 59–79 years) adults did not alter

Integrated GH concentrations 0 24-h integrated GH concentrations (Hartman et al. Pre- Post- 2000). Apparent unresponsiveness might be due exercise exercise to: (i) an insufficient training stimulus; (ii) lack of Fig. 10.7 The 6-h integrated serum growth hormone (GH) change in percentage body fat and abdominal vis- concentrations in 10 obese subjects before and after 16 ceral fat which correlate negatively with GH release weeks of aerobic exercise in response to an exercise bout and increase with age; and/or (iii) intrinsically o et al at 70% V 2peak for 30 min. (Adapted from Kanaley . diminished responsiveness of the aging GH–IGF-I 1999.) axis. Sixteen weeks of aerobic training in obese women young men, response down-regulation is evident yielded a significant training effect (e.g. increased o within the first 3 weeks of training (Fig. 10.5) V 2peak), but did not alter GH responses to acute (Weltman, A. et al. 1997). exercise at the same relative intensity (Fig. 10.7) On the other hand, total (24-h) GH secretion (Kanaley et al. 1999). However, training for 4 months appears to increase in trained individuals even on did not affect body weight, fat weight and fat free non-training days when some aerobic training was weight (estimated by skinfolds and bioelectrical targeted at an intensity above the lactate threshold impedance). Whether daily GH secretion rates in- (Fig. 10.6) (Weltman, A. et al. 1992). crease in this context is not known. Indeed, the pre- How healthy aging, gender and increased adi- cise relationship between acute exercise-stimulated

Human GH axis: neuromodulators

α 2-Adrenergic β 2-Adrenergic

? Exercise

? ? Dopamine Putative GHRPs Fig. 10.8 Schematic representation ? of the possible interactions or ? Cholinergic mechanisms that control exercise- induced growth hormone (GH) release. *, indicates the possibility GHRH Somatostatin Arginine that exercise modifies the normal autonegative feedback control of GH on growth hormone releasing GH hormone (GHRH) and somatostatin; +, denotes stimulation; –, denotes inhibition. (From Giustina & * * Veldhuis 1998. © 1998, The Endocrine IGF-I Society.) 128 chapter 10

160

) 140 − 1

120

100

80 S A 60 G AG 40

Serum GH concentration ( µ g·L 20

0 600 700 800 900 1000 1100 1200 (a) Clock time

160

) 140 − 1

120

100

80 S A 60 G Fig. 10.9 Mean serum growth AG hormone (GH) concentration (µg·L–1) 40 profiles basally and in response to saline (S), GH releasing peptide-2 (G),

Serum GH concentration ( µ g·L 20 and/or l-arginine (A) infusion at rest in men (a) and women (b). Data are 0 means ± SEM. Clock time (hours) is 600 700 800 900 1000 1100 1200 shown. (From Wideman et al. 2000b, (b) Clock time used with permission.) and training-enhanced (mean) GH production on tides, GABA and possibly excitatory amino acids non-training days remains unclear. (Thompson et al. 1993; Giustina & Veldhuis 1998; Weltman, A. et al. 2000a). Albeit correlational, plasma Neuroendocrine control of exercise- norepinephrine concentrations peak before and are induced growth hormone release proportionate to exercise-stimulated GH concentra- tions in young men with and without exercise The neuroendocrine basis underlying exercise- training (Weltman, A. et al. 1997, 2000a). Such data induced GH release is complex and remains enigm- are consistent with, but not proof of, central nora- α atic (Giustina & Veldhuis 1998; Wideman et al. 2000a). drenergic ( 2-adrenorecptor) drive. Serum ghrelin Presumptive mechanisms involve GHRH, SS and/ concentrations are higher in lean than obese adults or ghrelin (Fig. 10.8) (Veldhuis & Bowers 2003b). (Veldhuis & Bowers 2003b), but do not change Putative modulators include (non-exclusively) during 45 min of exercise and 3 h of recovery (Dall catecholamines, muscarinic agonists, opiatergic pep- et al. 2002). growth hormone and exercise 129

120 )

− 1 100

80

60 S A G 40 AG

20 Serum GH concentration ( µ g·L

0 600 700 800 900 1000 1100 1200 (a) Clock time

120 )

− 1 100

80

60 S A Fig. 10.10 Mean serum growth G 40 hormone (GH) concentration (µg·L–1) AG profiles basally and in response to saline (S), GH releasing peptide-2 (G), 20 and/or l-arginine (A) infusion with Serum GH concentration ( µ g·L exercise in men (a) and women (b). Data are means ± SEM. Clock time 0 (hours) is shown. (From Wideman 600 700 800 900 1000 1100 1200 et al. 2000a, used with permission.) (b) Clock time

Two probes of neuroendocrine pathways mediat- and l-arginine (but not GHRP-2) stimulated GH ing GH secretion are l-arginine and GHRP-2, which release is higher in women than men (Fig. 10.9) induce SS withdrawal and mimic ghrelin action (Wideman et al. 2000b). respectively. GHRP-2 also potentiates the effect of The synergy is equivalent in absolute terms in GHRH and mutes that of SS (Bowers et al. 1994; the female and the male before exertion. Exercise Pihoker et al. 1995; Popovic et al. 1995; Giustina potentiates maximal GH concentrations driven by & Veldhuis 1998; Bowers & Granda-Ayala 1999; l-arginine or GHRP-2 alone as well as together, Mueller et al. 1999; Veldhuis & Bowers 2003b). All and absolute responses are compatible by gender three agonists show sensitivity to gender (Merimee (Fig. 10.10) (Wideman et al. 2000a). et al. 1969; Benito et al. 1991; Bercu et al. 1991; Bowers Fractional responses to exercise (fold-increase 1993; Penelva et al. 1993; Veldhuis 1996a, 1998, 2003; above rest) are twofold higher in men than women Giustina & Veldhuis 1998; Jaffe et al. 1998; Veldhuis administered combined stimuli. Since GHRH ex- et al. 2001; Veldhuis & Bowers 2003a). At rest, basal pressly facilitates GHRP-2 and l-arginine actions, 130 chapter 10

synergy of the latter two secretagogues during exer- (supra-additive effect) (Weltman, A. et al. 2002). The cise strongly supports the inference that exertion later outcome could denote excessive somatosta- releases hypothalamic GHRH. This concept will need tinergic restraint, impaired outflow of endogenous to be tested by selective GHRH-receptor antagonists. GHRH (which synergizes with GHRP), and/or In complementation, estimates of stimulated GH down-regulation of the GHRP signaling pathway secretory-burst mass (Veldhuis et al. 1990) corrobor- as inferred independently in the aging human and ate the impact of l-arginine and GHRP-2 (Wideman experimental animal (Veldhuis et al. 2001, 2002; et al. 2000a, 2000b). However, GHRH, SS and put- Veldhuis 2003). A corollary question posed by cur- atively endogenous ghrelin are comodulated by rent data is how all three of exercise, age and gender β α inhibitory ( -adrenergic and 1-noradrenergic) and govern GH secretion (Brill et al. 2002). stimulatory (cholinergic and opiatergic) signal (Ghigo et al. 1993, 1994; Thompson et al. 1993; Giustina & Conclusions Veldhuis 1998; Mueller et al. 1999; Veldhuis & Yoshida 2000). Exercise appears to stimulate multiple Aerobic exercise is a potent stimulus for GH release, central–neural neurotransmitters (Sutton & Lazarus particularly in young men and women. Age and 1974; Uusitupa et al. 1982; Moretti et al. 1983; Bowers obesity blunt responsiveness markedly, putatively et al. 1984; Thompson et al. 1993; Giustina & by modifying metabolic signals to and effects of Veldhuis 1998). The integration of such responses is all three of GHR (stimulatory), SS (inhibitory) and a daunting challenge in exercise physiology. ghrlein (GHRP, amplifying). Gender further deter- An important emerging issue is the degree to mines GH secretion driven by exercise at any age. which specific GH secretagogues can amplify the In principle, impoverishment of physiological GH effects of exercise in aging individuals. In one pre- secretion in sedentary, obese and aged individuals liminary study in older men, combined stimulation would exacerbate visceral fat accumulation, dyslipi- with GHRP-2 and aerobic exercise elicited greater demia, relative insulin resistance, diminished bone GH secretion (summed amplitude) than either and muscle mass and (possibly) reduced quality of intervention alone, but did not act with true synergy life.

References

Arvat, E., Ceda, G.P., Di Vito, L. et al. Bercu, B.B., Weideman, C.A. & Walker, Bowers, C.Y. & Granda-Ayala, R. (1999) (1998) Age-related variations in the R.F. (1991) Sex differences in growth Stimulated release of GH in normal neuroendocrine control, more than hormone (GH) secretion by rats younger and older men and women. impaired receptor sensitivity, cause the administered GH-releasing In: Sex-Steroid Interactions with Growth reduction in the GH-releasing activity hexapeptide. Endocrinology 129, Hormone (Veldhuis, J.D. & Giustina, A., of GHRPs in human aging. Pituitary 1, 2592–2598. eds.). Serono Symposia, Norwell, MA: 51–58. van den Berg, G., Veldhuis, J.D., Frolich, 277–289. Asplin, C.M., Faria, A.C., Carlsen, E.C. M. et al. (1996) An amplitude-specific Bowers, C.Y., Maumenee, F.A., Reynolds, et al. (1989) Alterations in the pulsatile divergence in the pulsatile mode of GH G.A. & Hong, A. (1984) On the in vitro mode of growth hormone release in men secretion underlies the gender difference and in vivo activity of a new synthetic and women with insulin-dependent in mean GH concentrations in men and hexapeptide that acts on the pituitary diabetes mellitus. Journal of Clinical premenopausal women. Journal of to specifically release growth hormone. Endocrinology and Metabolism 69, Clinical Endocrinology and Metabolism 81, Endocrinology 114, 1537–1545. 239–245. 2460–2466. Bowers, C.Y., Veeraragavan, K. & Baumann, G. Growth hromone Bowers, C.Y. (1993) GH releasing Sethumadhavan, K. (1994) Atypical heterogeneity: genes, isohormones, peptidesastructure and kinetics. Journal growth hormone releasing peptides. variants and binding proteins. Endocrine of Pediatric Endocrinology 6, 21–31. In: Growth Hormone. II. Basic Clinical Reviews 12, 424–449. Bowers, C.Y. (2002) New insight into the Aspects (Bercu, B.B. & Walker, R.F., Benito, P., Avila, L., Corpas, M.S. et al. control of growth hormone secretion. eds.). Springer-Verlag, New York: (1991) Sex differences in growth In: Central and peripheral Mechanisms in 203–222. hormone response to growth hormone- Pituitary Disease (Kleinberg, D.L. & Brill, K.T., Weltman, J.Y., Anderson, S. releasing hormone. Journal of Clemmons, D.R., eds.). BioScientifica, et al. (2002) Relative rank order of Endocrinological Investigation 14, 265–268. Bristol, UK: 163–176. discrete secretagogue actions in the growth hormone and exercise 131

healthy aged male. Paper presented upon the activity of the GH-IGFs system does not increase 24-h GH secretion in at the 84th Annual Meeting of the in the elderly. International Journal of older adults. Paper presented at the Endocrine Society, San Francisco, CA. Sports Medicine 19, 250–254. 82nd Annual Meeting of the Endocrine Bunt, J.C., Boileau, R.A., Bahr, J.M., Engstrom, B.E., Karlsson, F.A. & Wide, L. Society, Toronto, Canada. Nelson, R.A. (1986) Sex and training (1998) Marked gender differences in Iranmanesh, A., Lizarralde, G. & differences in human growth hormone ambulatory morning growth hormone Veldhuis, J.D. (1991) Age and relative during prolonged exercise. Journal of values in young adults. Clinical adiposity are specific negative Applied Physiology 61, 1796–1801. Chemistry 44(6), 1289–1295. determinants of the frequency and Cappon, J.P., Brasel, J., Mohan, S. et al. Evan, W.S., Bowers, C.Y. & Veldhuis, J.D. amplitude of growth hormone (GH) (1994) Effect of brief exercise on (2001) Estradiol supplementation secretory bursts and the half-life of circulation insulin-like growth factor I. enhances pituitary sensitivity to endogenous GH in healthy men. Journal Journal of Applied Physiology 76, recombinant human (RH) GHRH-1, of Clinical Endocrinology and Metabolism 2490–2496. 44-amide in somatostatin (SS)- 73, 1081–1088. Chang, F.E., Dodds, W.G., Sullivan Kim, withdrawn postmenopausal women. Iranmanesh, A., South, S., Liem, A.Y. M.H. & Malarkey, W.B. (1986) The Paper presented at the Endocrine et al. (1998) Unequal impact of age, acute effects of exercise on prolactin and Society 83rd Meeting, Denver, CO. percentage body fat, and serum growth hormone secretion: comparison Farhy, L.S., Straume, M., Johnson, M.L. testosterone concentrations on the between sedentary women and women et al. (2001) A construct of interactive somatotrophic, IGF-I, and IGF-binding runners with normal and abnormal feedback control of the GH axis in the protein responses to a 3-day intravenous menstrual cycles. Journal of Clinical male. American Journal of Physiology 281, growth hormone-releasing hormone Endocrinology and Metabolism 62, R38–R51. pulsatile infusion in men. European 551–556. Felsing, N.E., Brasel, J.A. & Cooper, D.M. Journal of Endocrinology 139, 59–71. Chwalbinska-Moneta, J., Krysztofiak, H., (1992) Effect of low and high intensity Jaffe, C.A., Ocampo-Lim, B., Guo, W. Ziemba, A. et al. (1996) Threshold exercise on circulating growth hormone et al. (1998) Regulatory mechanisms of increases in plasma growth hormone in men. Journal of Clinical Endocrinology growth hormone secretion are sexually in relation to plasma catecholamine and and Metabolism 75, 157–162. dimorphic. Journal of Clinical blood lactate concentrations during Ghigo, E., Arvat, E., Bellone, J., Ramunni, J. Investigation 102, 153–164. progressive exercise in endurance- & Camanni, F. (1993) Neurotransmitter Kanaley, J.A., Weatherup-Dentes, M.M., trained athletes. European Journal of control of growth hormone secretion in Jaynes, E.B. et al. (1999) Obesity Applied Physiology 73, 117–120. humans. Journal of Pediatric attenuates the growth hormone Clasey, J.L., Weltman, A., Patrie, J. et al. Endocrinology 6, 263–266. response to exercise. Journal of Clinical (2001) Abdominal visceral fat and Ghigo, E., Arvat, E., Gianotti, L. et al. Endocrinology and Metabolism 84, fasting insulin are important predictors (1994) Interaction of salbutamol with 3156–3161. of 24-hour GH release independent of pyridostigmine and arginine on both Kanaley, J.A., Weltman, J.Y., Pieper, K.S. age, gender and other physiological basal and GHRH-stimulated GH et al. (2001) Cortisol and growth factors. Journal of Clinical Endocrinology secretion in humans. Clinical hormone responses to exercise at and Metabolism 86, 3845–3852. Endocrinology 40, 799–802. different times of day. Journal of Clinical Cordido, F., Dieguez, C. & Casanueva, Giustina, A. & Veldhuis, J.D. (1998) Endocrinology and Metabolism 86, F.F. (1990) Effect of central cholinergic Pathophysiology of the neuroregulation 2881–2889. neurotransmission enhancement by of growth hormone secretion in Kelijman, M. & Frohman, L.A. (1988) pyridostigmine on the growth hormone experimental animals and the human. Enhanced growth hormone (GH) secretion elicited by clonidine, arginine, Endocrine Reviews 19(6), 717–797. responsiveness to GH-releasing or hypoglycemia in normal and obese Hartley, L.H., Mason, J.W., Hogan, R.P. hormone after dietary manipulation in subjects. Journal of Clinical Endocrinology et al. (1972) Multiple hormonal obese and nonobese subjects. Journal of and Metabolism 70, 1361–1370. responses to graded exercise in relation Clinical Endocrinology and Metabolism 66, Cordido, F., Penalva, A., Dieguez, C. et al. to physical training. Journal of Applied 489–494. (1993) Massive growth hormone (GH) Physiology 33, 602–606. Kojima, M., Hiroshi, H., Date, Y. et al. discharge in obese subjects after the Hartman, M.L. (2000) Physiological (1999) Ghrelin is a growth-hormone combined administration of GH- regulators of growth hormone secretion. releasing acylated peptide from releasing hormone and GHRP-6: In: Growth Hormone in Adults, 2nd edn. stomach. Nature 402, 656–666. evidence for a marked somatotroph (Juul, A. & Jorgensen, J.O.L., eds.). Kozlowski, S., Chwalbinska-Moneta, J., secretory capability in obesity. Journal of Cambridge University Press, Vigas, M. et al. (1983) Greater serum GH Clinical Endocrinology and Metabolism 76, Cambridge, UK: 3–53. response to arm than to leg exercise 819–823. Hartman, M.L., Veldhuis, J.D., Vance, performed at equivalent oxygen uptake. Dall, R., Kanaley, J., Hansen, T.K. et al. M.L. et al. (1990) Somatotropin pulse European Journal of Applied Physiology 52, (2002) Plasma ghrelin levels during frequency and basal concentrations 11–135. exercise in healthy subjects and in are increased in acromegaly and are Lassarre, C., Girard, F., Durand, J. et al. growth hormone-deficient patients. reduced by successful therapy. Journal of (1974) Kinetics of human growth European Journal of Endocrinology 147, Clinical Endocrinology and Metabolism 70, hormone during submaximal exercise. 65–70. 1375–1384. Journal of Applied Physiology 37, 826–830. Deuschle, M., Blum, W.F., Frystyk, J. et al. Hartman, M.L., Weltman, J.Y., Patrie, J.T. Lewis, U.J., Sinha, Y.N., Lewis, G.P. (2000) (1998) Endurance training and its effect et al. (2000) Exercise training for 1 year Structure and properties of members of 132 chapter 10

the hGH family: a review. Endocrine hypothalamopituitary disconnection: Thompson, D.L., Weltman, J.Y., Rogol, Journal 47 (suppl.), S1–S8. evidence that GHRP-6 main action is A.D. et al. (1993) Cholinergic and opioid Luger, A., Watschinger, B., Duester, P. exerted at the hypothalamic level. involvement in release of growth et al. (1992) Plasma growth hormone and Journal of Clinical Endocrinology and hormone during exercise and recovery. prolactin responses to graded levels of Metabolism 80, 942–947. Journal of Applied Physiology 75, acute exercise and to a lactate infusion. Pritzlaff, C.J., Wideman, L., Weltman, J.Y. 870–878. Neuroendocrinology 56, 112–117. et al. (1999) Impact of acute exercise Uusitupa, M., Siitonen, O., Harkonen, M. Maccario, M., Valetto, M.R., Savio, P. et al. intensity on pulsatile growth hormone et al. (1982) Modification of the metabolic (1997) Maximal secretory capacity of release in men. Journal of Applied and hormonal response to physical somatotrope cells in obesity: comparison Physiology 87, 498–504. exercise by beta-blocking agents. with GH deficiency. International Journal Pritzlaff, C.J., Wideman, L., Blumer, J. Annals of Clinical Research 14 (suppl.), of Obesity 21, 27–32. et al. (2000) Catecholamine release, 165–167. Marcell, T.J., Wiswell, R.A., Hawkins, S.A. growth hormone secretion, and energy Vahl, N., Jorgensen, J.O., Skjaerback, C. et al. (1999) Age-related blunting of expenditure during exercise vs. recovery et al. (1997) Abdominal adiposity rather growth hormone secretion during in men. Journal of Applied Physiology 89, than age and sex predicts the mass and exercise may not be solely due to 937–946. patterned regularity of growth hormone increased somatostatin tone. Metabolism Pritzlaff-Roy, C.J., Wideman, L., Weltman, secretion in mid-life healthy adults. 48(6), 665–670. J.Y. et al. (2002) Gender governs the American Journal of Physiology 272, Merimee, T.J., Rabinowitz, D. & Fineberg, relationship between exercise intensity E1108–1116. S.E. (1969) Arginine-initiated release and growth hormone release in young Veldhuis, J.D. (1995) The neuroendocrine of human growth hormone: factors adults. Journal of Applied Physiology 92, regulation and implications of pulsatile modifying the response in normal man. 2053–2060. GH secretion: gender effects. New England Journal of Medicine 280, Raynaud, J., Capderou, A., Martineaud, Endocrinologist 5, 198–213. 1434–1438. J.-P. et al. (1983) Intersubject variability Veldhuis, J.D. (1996a) New modalities for Moretti, C., Favvri, A., Gnessi, L. et al. of growth hormone time course during understanding dynamic regulation of (1983) Naloxone inhibits exercise- different types of work. Journal of Applied the somatotropic (GH) axis: explication induced release of PRL and GH in Physiology 55, 1682–1687. of gender differences in GH athletes. Clinical Endocrinology 18, Richmond, E.J., Rogol, A.D., Clark, W., neuroregulation in the human. Journal 135–138. Dasdemir, D. & Veldhuis, J.D. (2001) of Pediatric Endocrinology and Metabolism Mueller, E.E., Locatelli, V. & Cocchi, D. Paradoxically heightened susceptibility 9, 237–253. (1999) Neuroendocrine control of to GH autonegative feedback in Veldhuis, J.D. (1996b) Physiological growth hormone secretion. Physiological midpuberty compared with prepuberty regulation of growth hormone (GH)- Reviews 79, 511–607. in healthy boys. Paper presented at the insulin-like growth factor type I (IGF-I) Nindl, B.C., Kraemer, W.J., Marx, J.O. et al. Endocrine Society 83rd Meeting, axis: predominant impact of age, (2003) Growth hormone molecular Denver, CO. obesity, gonadal function, and sleep. heterogeneity and exercise. Exercise and Rudman, D., Kutner, M.H., Rogers, C.M. Sleep 19, S221–S224. Sport Sciences Reviews 31, 161–166. et al. (1981) Impaired growth hormone Veldhuis, J.D. (1998) Neuroendocrine Penelva, A., Pombo, M., Carballo, A. secretion in the adult population. Journal control of pulsatile growth hormone et al. (1993) Influence of sex, age and Clinical Investigation 67, 1361–1369. release in man: relationship with gender. adrenergic pathways on the growth Shuto, Y., Shibasaki, T., Otagiri, A. et al. Growth Hormone and IGF Research 8, hormone response to GHRP-6. Clinical (2002) Hypothalamic growth hormone 49–59. Endocrinology 38, 87–91. secretagogue receptor regulates growth Veldhuis, J.D. (2003) A tripeptidyl Pihoker, C., Middleton, R., Reynolds, G.A., hormone secretion, feeding, and ensemble perspective of interactive Bowers, C.Y. & Badger, T.M. (1995) adiposity. Journal Clinical Investigation control of growth hormone secretion. Diagnostic studies with intravenous and 109, 1429–1436. Hormone Research 60, 86–101. intranasal growth hormone-releasing Sutton, J. & Lazarus, L. (1974) Effect of Veldhuis, J.D. & Bowers, C.Y. (2003a) peptide-2 in children of short stature. adrenergic blocking agents on growth Sex-steroid modulation of growth Journal of Clinical Endocrinology and hormone responses to physical exercise. hormone (GH) secretory control: three- Metabolism 80, 2987–2992. Hormone and Metabolic Research 6, peptide ensemble regulation under dual Pincus, S.M., Gevers, E., Robinson, I.C.A.F. 428–429. feedback restraint by GH and IGF-I. et al. (1996) Females secrete growth Sutton, J. & Lazarus, L. (1976) Growth Endocrine 22(1), 25–40. hormone with more process irregularity hormone in exercise: comparison of Veldhuis, J.D. & Bowers, C.Y. (2003b) than males in both human and rat. physiological and pharmacological Three-peptide control of pulsatile and American Journal of Physiology. stimuli. Journal of Applied Physiology 41, entropic feedback-sensitive modes of Endocrinology and Metabolism 270, 523–527. growth hormone secretion: modulation E107–E115. Tanaka, K., Inoue, S., Numata, K. et al. by estrogen and aromatizable androgen. Popovic, V., Damjanovic, S., Micic, D. (1990) Very-low-calorie diet-induced Journal of Pediatric Endocrinology 16 et al. (1995) Blocked growth hormone- weight reduction reverses impaired (suppl. 3), 587–605. releasing peptide (GHRP-6)-induced growth hormone secretion responses to Veldhuis, J.D. & Yoshida, K. (2000) GH secretion and absence of the growth hormone-releasing hormone, Impact of chronic training on pituitary synergic action of GHRP-6 plus GH- arginine and l-dopa in obesity. hormone secretion in the human. In: releasing hormone in patients with Metabolism 39, 892–896. Sports Endocrinology (Warren, M.P. & growth hormone and exercise 133

Constantini, N.W., eds.). Humana Press, age, percentage body fat, fitness, and Wideman, L., Weltman, J.Y., Shah, N. et al. Totowa, NJ: 57–76. 24-hour growth hormone release in (1999) The effects of gender on exercise- Veldhuis, J.D., Lassiter, A.B. & Johnson, healthy young adults: effects of gender. induced growth hormone release. M.L. (1990) Operating behavior of dual Journal of Clinical Endocrinology and Journal of Applied Physiology 87, or multiple endocrine pulse generators. Metabolism 78, 543–548. 1154–1162. American Journal of Physiology. Weltman, A., Weltman, J.Y., Womack, C.J. Wideman, L., Weltman, J.Y., Patrie, J.T. Endocrinology and Metabolism 259, et al. (1997) Exercise training decreases et al. (2000a) Synergy of l-arginine and E351–E361. the growth hormone (GH) response to growth hormone (GH)-releasing Veldhuis, J.D., Iranmenesh, A., Ho, K.K.Y. acute constant-load exercise. Medicine peptide-2 (GHRP-2) stimulation of GH et al. (1991) Dual defects in pulsatile and Science in Sports and Exercise 29, in men and women: Modulation by growth hormone secretion and clearance 669–676. exercise. American Journal of Physiology. subserve the hyposomatotropism of Weltman, A., Pritzlaff, C.J., Wideman, L. Regulatory, Integrative and Comparative obesity in man. Journal of Clinical et al. (2000a) Exercise-dependent growth Physiology 279, R1467–R1477. Endocrinology and Metabolism 72, hormone release is linked to markers Wideman, L., Weltman, J.Y., Patrie, J.T. 51–59. of heightened central adrenergic et al. (2000b) Synergy of l-arginine and Veldhuis, J.D., Liem, A.Y., South, S. et al. outflow. Journal of Applied Physiology growth hormone (GH)-releasing (1995) Differential impact of age, sex 89, 629–635. peptide-2 on GH release: influence of steroid hormones, and obesity on basal Weltman, A., Pritzlaff, C.J., Wideman, L. gender. American Journal of Physiology. versus pulsatile growth hormone et al. (2000b) The relationship between Regulatory, Integrative and Comparative secretion in men as assessed in an exercise intensity and growth hormone Physiology 279, R1455–R1466. ultrasensitive chemiluminescence assay. (GH) release is attenuated in older Williams, T., Berelowitz, M., Joffe, S.N. Journal of Clinical Endocrinology and men. Paper presented at the Fourth et al. (1984) Impaired growth hormone Metabolism 80, 3209–3222. International Conference of the Growth responses to growth hormone-releasing Veldhuis, J.D., Iranmanesh, A. & Hormone Research Society, Gothenberg, factor in obesity: a pituitary defect Weltman, A. (1997) Elements in the Sweden. reversed with weight reduction. pathophysiology of diminished growth Weltman, A., Anderson, S.M., Wideman, New England Journal of Medicine 311, hormone (GH) secretion in aging L. et al. (2001) Impact of short-term 1403–1407. humans. Endocrine 7, 41–48. estrogen supplementation in Winer, L.M., Shaw, M.A., Baumann, G. Veldhuis, J.D., Iranmanesh, A., Mulligan, postmenopausal women on (1990) Basal plasma growth hormone T. & Anderson, S. (2001) Interactive spontaneous and exercise stimulated levels in man: new evidence for regulation of postmenopausal growth pulsatile growth hormone (GH) rhythmicity of growth hormone hormone insulin-like growth factor axis secretion. Paper presented at the 83rd secretion. Journal of Clinical by estrogen and growth hormone- Annual Meeting of the Endocrine Endocrinology and Metabolism 70, releasing peptide-2. Endocrine 14, Society. Denver, CO. 1678–1686. 45–62. Weltman, A., Brill, K., Weltman, J.Y. et al. Zaccaria, M., Vanier, M., Piazza, P., Veldhuis, J.D., Iranmanesh, A., Mulligan, (2002) GHRP-2 partially rescues Noventa, D. & Ermolao, A. et al. (1999) T. & Bowers, C.T. (2002) Mechanisms impaired exercise stimulation of growth Blunted growth hormone response of conjoint failure of the somatotropic hormone (GH) release in older men. to maximal exercise in middle-aged and gonadal axes in ageing men. Paper presented at the 84th Annual versus young subjects and no effect of Novartis Foundation Symposium 242, Meeting of the Endocrine Society, San endurance training. Journal of Clinical 98–118. Fransisco, CA. Endocrinology and Metabolism 84, Weltman, A., Weltman, J.Y., Schurrer, R. Weltman, J.Y., Frick, K., Watson, D. et al. 2303–2307. et al. (1992) Endurance training amplifies (2002) Comparison of continuous and Zadik, Z., Chalew, S.A., McCarter, R.J. et al. the pulsatile release of growth hormone: intermittent exercise on 24-h growth (1985) The influence of age on the 24- effects of training intensity. Journal of hormone secretion in obese and non- hour integrated concentration of growth Applied Physiology 72, 2188–2196. obese young men. Paper presented hormone in normal individuals. Journal Weltman, A., Weltman, J.Y., Hartman, at the 84th Annual Meeting of the of Clinical Endocrinology and Metabolism M.L. et al. (1994) Relationship between Endocrine Society, San Fransisco, CA. 60, 513–516. Chapter 11

Proopiomelanocortin and Exercise

HEINZ W. HARBACH AND GUNTER HEMPELMANN

Introduction stressors induces overtraining syndrome. Changes in circulating hormonal concentrations have been The pituitary proopiomelanocortin (POMC) sys- proposed as indicators of overtraining, although tem is activated under physical, psychological and such hormonal changes are not always observed immunological stressors, which induce the release (Fry et al. 1998). of POMC fragments into the cardiovascular com- However, it remains to be elucidated, which partment. The response to physical stress includes purpose for POMC fragments are released into the the release of a variety of neuronal or humoral sig- cardiovascular compartment related to exercise nals from nervous, endocrine or immune systems conditions. This has to be emphasized despite of a ending up with morphological or functional altera- frequently offered answer suggesting the opposite tions, for example of muscular tissue or of the car- by referring to an ‘adaptation to stress’. The clari- diovascular system (Teschemacher 2003). fication of the functional significance of the POMC Adrenocorticotropic hormone (ACTH) and β- system, which is obviously activated by the human endorphin are the compounds mostly studied under organism every day and under multiple stress con- physical stress conditions, whereas further POMC ditions, still remains a target of interest (Teschemacher fragments, such as β-lipotropin (β-LPH), are also 2003). The exact mechanism(s) responsible for exer- released upon physical exercise. The secretion of cise-induced increases in β-endorphin, ACTH and POMC derivatives during exercise is an adaptive cortisol remain speculative. They may also differ in attempt of the athlete’s organism to cope with dif- dependence of the duration of exercise stress or in ferent stress situations. It is intimately linked to a dependence of the exercise stress intensity (i.e. sub- variety of psychological strategies that facilitate the maximal and maximal) exercise stress, when high- organism’s navigation through a stressful envir- intensity exercise at supramaximal oxygen (O2) β o onment (McCubbin 1993). ACTH and -endorphin consumption (V 2max) does not longer produce sig- appear to be released during exercise in a sufficient nificant increase in plasma β-endorphin (Kraemer, intensity and duration. The exercise response may W.J. et al. 1993). This review summarizes the current also be affected by the training status of the indivi- knowledge about POMC in exercise physiology. dual and the population being investigated (Goldfarb & Jamurtas 1997). Aerobic activities (endurance POMC and its occurrence in the exercise) are quite different from anaerobic activit- history of life ies, such as resistance exercise. In case of situations of overtraining challenge, a reduced ACTH response POMC—phylogenetic ontogeny and may reflect the athlete’s impaired ability to cope POMC-derived peptides with the stress situation. Whereas short-term over- training (overreaching) can be reversed by a more Proopiomelanocortin (POMC) is older than 500 mil- prolonged period of recovery, further exposure to lion years. This protein has not only been found in 134 proopiomelanocortin and exercise 135

ACTH β-LPH

γ-LPH β-END

Signal peptide N-term fragment γ-MSH JP α-MSH CLIP β-MSH β-END

acetyl-N- β-END (1–31)

acetyl-N- β-END (1–27)

acetyl-N- β-END (1–26)

acetyl-N- β-END (1–17)

acetyl-N- β-END (1–16)

Fig. 11.1 Schematic representation of known products of enzymatic processing of proopiomelanocortin (POMC). Signal peptide; N-terminal fragment; adrenocorticotropic hormone (ACTH); β-lipotropin (LPH); γ-LPH; α-melanotropin (MSH); β-MSH; γ-MSH; corticotropin-like intermediate lobe peptide (CLIP); joining peptide (JP); β-endorphin (1–31) (β-END (1–31)); β-END (1–27); β-END (1–26); β-END (1–17); β-END (1–16); acetyl-N-β-END (1–31); acetyl-N-β-END (1–27); acetyl-N-β-END (1–26); acetyl-N-β-END (1–17); acetyl-N-β-END (1–16). mammalians but already in invertebrates (Denef & of the 5’-untranslated sequence, the sequence of Van Bael 1998), and is also expressed in unicellular the signal peptide and the first 18 amino acids of organisms (Salzet et al. 1997). So far, POMC might be POMC. The third exon encodes the rest of the of significance both for the primitive living beings POMC precursor with the sequences of ACTH, β- and for the highest developed living systems. Much LPH and β-endorphin. Biological active peptides about the ontogeny of the human POMC system is arising from POMC are subjects to further process- not known, but it is known that POMC fragments ing, thus yielding smaller peptide products with are prepared in the human fetal organism already distinct biological activity. So far, POMC is pro- from the 5th week of gestation on (Facchinetti et al. cessed to pro-γ-melanocyte stimulating hormone 1987). POMC and two other major prohormones, (pro-γ-MSH) (also called N-POMC), corticotropin- proenkephalin and prodynorphin, give rise to the like-intermediate protein (CLIP), joining peptide three ‘families’ of opioid peptides: the endorphin, (JP), ACTH and β-LPH (Fig. 11.1). There are at least the enkephalin and the dynorphin families. 10 β-endorphin derivatives known to exist in the Peptide hormones and neuropeptides are synthes- mammalian organism, β-endorphin (1–27), (1–26), ized as part of a large protein precursor molecule (1–17) and (1–16), as well as their N-acetylated POMC from which they are post-translationally lib- forms, and in addition N-acetyl- β-endorphin (1–31) erated and modified through enzymatic processing (Teschemacher et al. 1990a; Höllt 1993; Young et al. (Fig. 11.1). In humans, there is one POMC gene per 1993). Pro-γ-MSH is cleaved to give an N-terminal haploid genome located on chromosome 2 (p23). It peptide (N-POMC (1–49)) ACTH to give α-MSH is 7665 base pairs long and consists of three exons and CLIP, and β-LPH is cleaved to β-endorphin and separated by two large introns (3–4 and 2–3 kb γ-LPH. The JP has no biological activity, although respectively). The first exon is about 100 nucleotides it contains a single cysteine residue, unique in in length and contains the 5’-untranslated region humans, which allows it to dimerize, and it is of the POMC mRNa. Exon 2 consists of about 150 thought that POMC may also dimerize within nucleotides and contains, besides a small portion the cell during biosynthesis (Bicknell et al. 1996). 136 chapter 11

N-POMC is involved in adrenal mitogenesis and cleaved into three big fragments, a so-called 16 K- α-MSH controls pigmentation in lower vertebrates. fragment, ACTH and β-LPH (Fig. 11.1); the latter The mRNA in the pituitary is 1072 bases long, POMC fragment consists of the sequences of γ- whereas the transcript in the brain is longer and LPH and β-endorphin, which are released in small most of the transcripts in the periphery are shorter, amounts also. In the melanotropic cells, the same i.e. about 800 bases long. POMC is produced in POMC fragments are released from POMC as in particularly high amounts in the pituitary gland. the corticotropic cells; however, they are further The POMC mRNA in the pituitary, in the brain and processed to release a series of smaller frag- in the periphery as far as comparable with pituitary ments, which are in part N-terminally acetylated or mRNA length is translated into the amino acid C-terminally amidated. α-Melanocyte stimulating sequence of ‘pre-proopiomelanocortin’, ‘pre-POMC’. hormone (α-MSH) or acetylated β-endorphin frag- After cleavage of the N-terminally located signal ments are typical examples for this type of POMC sequence from pre-POMC, POMC is left, which is derivatives (Loh 1992; Young et al. 1993; Castro & a protein of 241 amino acid residues, from which Morrison 1997). POMC fragments such as ACTH or β-endorphin are released (for review see Höllt 1993; Bertagna 1994). POMC—the precursor of ACTH and β-endorphin—and the endocrine POMC localization in the human organism regulation of the response to exercise stress The main site of expression of the POMC gene is the pituitary gland, in particular the corticotropic cells POMC release under physical stress conditions of the anterior lobe and the melanotropic cells of the intermediate lobe. The intermediate lobe as the genu- The ‘corticotropic part’ of the POMC system is ine location of the melanotropic cells in lower activated under the conditions of physical exercise, species is vestigial in the human pituitary gland. which results in the release of ACTH and β- So far the human organism lacks a well-defined endorphin immunoreactive material (IRM). ACTH intermediate lobe containing melanotropic cells. was studied as the main representative of the However, the POMC-expressing cells in the anterior POMC system for a long time under the aspect lobe are able to process POMC according to either of stress-induced activation of the ‘hypothalamic– type of enzymatic cleavage pattern, the melanotro- pituitary–adrenal axis’ (HPA axis) (Ganong et al. pic as well as the corticotropic one. These enzymatic 1987; Tache & Rivier 1993). ACTH is regarded as a systems, however, apparently are still regulated main valid parameter of the endocrine stress separately (Evans, V.R. et al. 1994). response besides epinephrine, norepinephrine and Outside the pituitary POMC transcripts have cortisol (Adams & Hempelmann 1991). Later β- been found as well, for example in the arcuate endorphin and β-LPH were recognized to be re- nucleus of the hypothalamus and other regions of leased from the pituitary under identical or similar the brain. Moreover, POMC-like transcripts have stress conditions (Owens & Smith 1987; McLoughlin been detected in a number of peripheral tissues. et al. 1993). These are the thyroid gland, thymus, adrenal med- ACTH and β-endorphin IRM concentrations in ulla, gonads, placenta, pancreas, kidneys, spleen, the plasma mostly were secreted in equimolar liver, gastrointestinal wall, skin, monocytes, macro- quantities in response to exercise (Akil et al. 1984; De phages and T-cells. Meirleir et al. 1986; Farrell et al. 1987; Rahkila et al. 1988; Strassman et al. 1989; Schwarz & Kindermann 1990; Heitkamp et al. 1993; Schulz et al. 2000). POMC expression in the human organism However, as shown in female marathon runners, In the corticotropic cells POMC is glycosylated and basal levels of ACTH and β-endorphin IRM can be phosphorylated and subsequently enzymatically different and the increase of ACTH levels under proopiomelanocortin and exercise 137

certain conditions, for example upon running to exercise proved to be elevated in marathon-trained exhaustion, can exceed the increase of β-endorphin athletes in comparison with untrained volunteers. IRM by a factor of five (Heitkamp et al. 1996). However, since cortisol levels did not reflect this Acute stress can be defined biochemically by difference, the authors of this study hypothesized a measuring an increase in catecholamine release. decreased HPA axis sensitivity to cortisol negative Chronic stress has proved more difficult to define in feedback. Results of a following study from Duclos biochemical terms, although in psychological terms et al. (1998) discarded this hypothesis of a decreased it has been considered as the lack of ability to cope adrenal sensitivity of the ACTH targets after pitui- with the environment or loss of control in a long- tary adrenal stimulation. term situation (McLoughlin et al. 1993). In the case of Marquet et al. (1999) administered low and high athlete’s feeling subjective symptoms of stress, com- doses of dexamethasone. After the last dose the bined with accelerated fatigability or diminished subjects performed a maximal cycling exercise, physical performance, a state of overtraining stress and blood was sampled just before and after each is given. exercise bout. Blood levels of ACTH, β-endorphin, cortisol and sex steroids except testosterone were lowered by dexamethasone at rest and after exer- POMC response to acute aerobic and cise. These effects were interpreted as an impair- endurance exercise ment of the adaptation to intense physical loads. Assuming that the stress response is a neuroen- Controversy exists regarding the effects of docrine mechanism that occurs in anticipation of duration of exercise and exercise dosage on β- physical exercise, De Vries et al. (2000) investigated endorphin release during exercise. Angelopoulos whether an incremental exercise protocol could be (2001) demonstrated an increased concentration of used as a model stressor. Subjects cycled at 40–100% β-endorphin maintained over time during intense o of the power output at the maximal O2 consumption exercise (trials at 80% V 2max for 10–30 minutes), o (V 2max) each up to exhaustion. Their results showed whereas Goldfarb et al. (1991) reported a gradual that increases in heart rate, lactate, epinephrine increase in β-endorphin concentration over time at o and norepinephrine reflect the relative workload, in 80% V 2max, and Taylor et al. (1994) observed a contrast to increases in ACTH hormone and β- linear increase in β-endorphin concentration with endorphin, which were observed only after exercise time. The disparity among the results may be due o reached an intensity of 80% V 2max. Their results to differences in the exercise dosage or a match demonstrated that activation of stress hormones between the rate of β-endorphin release and dis- o occur at different time points, the delayed response appearance at 50% V 2max conditions. The secretion of the HPA axis during incremental exercise con- of β-endorphin and ACTH were also described in trasted with the non-delayed HPA axis response dependence of the intensity of exercise (Rahkila et al. observed during psychological stress (De Vries et al. 1987, 1988). However, results from Petraglia et al. 2000). Farrell et al. (1983) compared ACTH levels (1990) at marathon race conditions and submaximal o before and after submaximal (80% of maximal O2 exercise at 50% V 2max indicate that the duration but o β consumption [V 2max]) and exhaustive isotonic not the workload were responsible for -endorphin o β exercise (100% V 2max) and also concluded that and -LPH release. exercise-induced increases in plasma ACTH and Goldfarb and Jamurtas (1997) reviewed signifi- their correlation with circulating cortisol depend on cant increases in β-endorphin at workloads of 60% o the intensity of isotonic exercise. At workloads of and 90% V 2max only at the higher workload and o β 40–60% V 2max -endorphin was not significantly dependant on the duration of exercise, respectively. elevated, but it was significantly elevated at 80% Graded or incremental exercise of an aerobic nature o β V 2max (Donevan & Andrew 1987; Langenfeld et al. appears to increase -endorphin. Exercise at 66% o β 1987; McMurray et al. 1987; Sforzo 1989). In a study and 57% V 2max was reported to increase plasma - by Duclos et al. (1997) ACTH levels after running endorphin in the untrained and trained individuals, 138 chapter 11

respectively. The authors indicated that the β- was measured at marathon race. All runners experi- endorphin levels did not differ significantly between enced a rise in β-endorphin activity from pre- the endurance-trained and untrained participants marathon to immediate post-marathon. However, either prior to or during exercise. hypercapnic ventilatory responsiveness showed no Lac et al. (1999) present reactions of cortisol to a significant change. Mahler et al. (1989) concluded dexamethasone treatment via ACTH blockade and that the natural increase in endogenous β-endorphin o to an exercise bout at V 2max. After exercise, plasma activity associated with marathon running does ACTH rose to 600% of basal value, whereas cortisol not modulate central chemosensitivity. Adult male o was unaffected, which was explained by differences students were given V 2max tests to confirm their in peripheral metabolism. aerobic fitness levels. The ‘fit group’ showed sig- Changes of plasma ACTH and β-endorphin were nificantly lower heart rate responses than the ‘non- observed under high-intensity cycle exercise and fit group’, and these fitness group differences were exposure to a simulated altitude of 4300 m (hypo- abolished by the opioid antagonist naltrexone. baric and hypoxic) and under sea level conditions: McCubbin (1993) suggested that regular aerobic Exercise at acute hypobaric hypoxia elicited no sig- exercise conditioning augments release of inhibit- nificantly different responses in plasma β-endorphin, ory opioids. ACTH or cortisol and no changes in β-endorphin/ ACTH molar ratio than those elicited under nor- POMC response to acute resistance exercise mobaric conditions (Kraemer, W.J. et al. 1991). A significant increase in serum ACTH was observed Aerobic activities (endurance exercise) are quite dif- in response to a 21 km non-competitive race at low ferent from anaerobic activities, such as resistance altitude (350 m below sea level) in comparison to exercise. A decrease in β-endorphin plasma concen- 620 m above sea level; it was proposed that ACTH tration following resistance exercise was observed may play a role in acclimatization to exercise at low as compared with pre-exercise (Pierce et al. 1994), altitudes (el Migdadi et al. 1996). and in addition, no relationship between affect and Repeated physical exercise including the follow- β-endorphin response to exercise was observed ing recovery period may be a useful model to study (McGowan et al. 1993). As Goldfarb and Jamurtas the effect of various rest intervals on subsequent (1997) also described, it appeared there is no definit- stress reactions. Therefore, Ronsen et al. (2002) ive response of β-endorphin to resistance exercise. designed a study that compared neuroendocrine In heavy-resistance exercise, it appeared that the and immune responses during days with two equal duration of exercise, length of the rest periods bouts of high-intensity endurance exercise, but dif- between exercise sets and blood lactate were key ferent periods of rest between the first and second exercise variables that influence increases in plasma bout. They demonstrated that when a second bout β-endorphin concentration (Kraemer, W.J. et al. of exercise was performed after 3 h compared with 1993). Their dataaonly one of six heavy-resistance 6 h of rest, increases in epinephrine, ACTH and exercise protocols examined resulted in a significant cortisol were augmented. They showed that recovery increase in plasma β-endorphin and serum cortisol time as such is a significant factor in achieving concentrationsademonstrated that the exercise sti- homeostasis between repeated sessions of endur- mulus for β-endorphin increase was characterized ance exercise. by longer-duration sets and shorter rest periods between sets and exercises. The number of repeti- β-Endorphin and ventilatory responsiveness. To invest- tions that the resistance (i.e. intensity) allows in the igate the hypothesis that endurance exercise may set (i.e. duration) and the rest period length are the lead to a decrease in ventilatory chemosensitivity as primary determinants of the physiological stress possibly mediated by an increase in endogenous β- (Kraemer, W.J. et al. 1989b). β-Endorphin increased endorphin, the hypercapnic ventilatory responsive- at the end of resistance exercise during energy bal- ness and circulating β-endorphin immunoreactivity ance, but significantly only during negative energy proopiomelanocortin and exercise 139

β o balance (Walberg-Rankin et al. 1992). -Endorphin increase in V 2max of volunteers but a blunted or whole blood lactate were not influenced by re- ACTH response to an absolute submaximal work sistance training experience in elite weightlifters rate in a 12-week training program. In addition, (Kraemer, W.J. et al. 1992). It appears that there is Luger et al. (1987) observed decreased ACTH levels no definitive response of β-endorphin to resist- in trained subjects under consideration of absolute ance exercise. Non-exhaustive performance such as workload. In contrast, Ronkainen et al. (1986), who heavy resistance exercise, although requiring high investigated the effects of endurance exercise on muscular activity for short time periods, did not lead the function of adrenal cortex of female runners, to an increase of β-endorphin levels at all (Pierce supposed no alteration of the function of the adrenal et al. 1993b), even decreased them (Pierce et al. 1994), cortex by chronic endurance exercise, when res- or only allowed an increase under approximately ponses of cortisol to intravenous ACTH injections exhaustive conditions (Kraemer, W.J. et al. 1993). were estimated. Schulz et al. (2000) demonstrated the increase of ACTH and β-endorphin IRM concentrations in the POMC response to chronic resistance exercise. Data plasma upon anaerobic exercise, correlated with the from Kraemer, W.J. et al. (1992) on effects of resist- increase of lactate levels observed upon anaerobic ance training experience demonstrated an increase exercise. Authentic β-endorphin (1–31) was only of β-endorphin, testosterone and lactate, whereas found in two plasma samples containing minor con- only testosterone was influenced by 2-years’ train- centrations of the peptide. They concluded that ing experience. the β-endorphin IRM released into blood under Deschenes et al. (2002) tested whether muscle anaerobic exercise was identical with authentic β- unloading were attributable to adaptations of the endorphin (1–31) only to a minor extent. The major neural system rather than modifications of myofiber part of the material in fact released into the blood size and confirmed that the loss of muscular strength upon anaerobic exercise was probably identical could primarily be attributed to a decreased capa- with β-LPH. city of the nervous system to excite the muscle. The results examining resistance exercise and In contrast to resistance training, which results in β-endorphin response are equivocal: partly due to strength gains and muscle hypertrophy, muscle the selection of participants; partly due to the inten- unloading evokes reductions in muscle perform- sity, duration and rest periods of exercise utilized ance. In addition, they observed that muscle un- (Kraemer, W.J. et al. 1989; Goldfarb & Jamurtas loading affects the hormonal milieu in a manner 1997); and partly due to the specificity of assays that promotes muscle atrophy: increase of the utilized (Harbach et al. 2000; Schulz et al. 2000). catabolic hormone cortisol in the absence of a con- Ethanol did not increase circulating cortisol con- comitant elevation of ACTH. centration above that caused by the resistance exer- Sleep deprivation did not alter the resting β- cise, but it appeared to have a more prolonged effect endorphin response, nor did it affect the β-endor- without a change in circulating ACTH (Koziris et al. phin response to high-intensity exercise (McMurray 2000). Unchanged resting concentrations of cortisol et al. 1988). during short periods of heavy strength training were also observed by Fry et al. (1998) and Raastad Influence of training. The influence of training on et al. (2001). basal plasma levels or on levels under exercise con- ditions has been reviewed (Goldfarb & Jamurtas 1997). A number of studies differing in the composi- Chronic training effects and POMC response: tion of volunteer groups as well as using different chronic aerobic and endurance exercise versus protocols obviously revealed controversial results: chronic resistance exercise Training has been reported to have increased, made POMC response to chronic aerobic and endurance no difference to or decreased β-endorphin levels exercise. Buono et al. (1987) observed a significant following exercise. Goldfarb and Jamurtas (1997) 140 chapter 11

interpreted the discrepancy in the literature as in the highly trained subjects. Like the levels of plasma part related to the type of training and its intensity, ACTH and cortisol, the level of plasma lactate was the methods used to measure β-endorphin and the coupled to exercise intensity, since all groups had mode of exercise utilized. A clear-cut experimental similar plasma lactate responses at matched exer- design was presented by Engfred et al. (1994), which cise intensities. Trained subjects required much did not just compare trained and untrained volun- higher absolute workloads to produce comparable teers but looked at the same subjects in a state before elevations in plasma lactate concentrations; they and after training as conducted and controlled in had much less activation of the HPA axis as com- the frame of their own study. Training-induced pared with untrained subjects at matched absolute changes in response to exercise of norepinephrine, workloads. Daily strenuous exercise was inter- epinephrine, growth hormone, β-endorphin and preted to lead to chronic ACTH hypersecretion insulin were similar in the groups. Incremental and adrenal hyperfunctionaphysical conditioning exhausting and prolonged exhausting endurance associated with adaptation mechanisms such as the exercise such as marathon running induced an ability to increase the capacity to handle a higher increase of similar magnitude in both β-endorphin workload with less pituitary–adrenal activation and ACTH concentration (Heitkamp et al. 1993). (Luger et al. 1987). Data of the same group showing ACTH and β-endorphin response to exercise work- elevated basal concentrations of ACTH and cortisol load proved to be dramatically reduced after a 5 and a blunted response to exogenous corticotropin- weeks’ training protocol. In a similarly designed releasing hormone (CRH) were compatible with study, untrained females were subjected to 8 weeks’ mild sustained hypercortisolism in highly trained endurance training and the effects of running runners (Luger et al. 1988). 30 minutes three times a week on ACTH and β- In studies comparing sedentary with trained endorphin levels were measured (Heitkamp et al. subjects, no training-induced adaptation of POMC 1998). Basal β-endorphin levels were not altered by derivative release was observed under different the training program, but basal ACTH levels exercise conditions (Kraemer, R.R. et al. 1989; increased. Both, ACTH as well as β-endorphin Goldfarb et al. 1991). Three training groups were levels, were dramatically increased immediately elected for maximal treadmill exercise. Sprint inter- after the exercise; this increase, however, was found vals, endurance training and combination training to be attenuated in case of β-endorphin. Thus, train- were observed over 10 weeks. No training-induced ing appeared to provoke a POMC fragment-specific hormonal changes were observed for the endu- response, which was difficult to recognize just upon rance group. While exercise-induced increases were acute exercise challenge of the POMC system. Results observed, the combination group exhibited signifi- obtained with untrained and trained volunteers cant post-training reductions in plasma responses of also showed differences between the two groups β-endorphin and ACTH (Kraemer, W.J. et al. 1989a). concerning different ACTH and β-endorphin res- In comparison with men, female marathon run- ponses, which can be interpreted in the same way ners showed higher baseline concentrations, lesser (Diego Acosta et al. 2001). Training also appears to increases in β-endorphin, lower baseline concentra- influence β-endorphin metabolism in the resting tions and larger increases in ACTH concentration state and during exercise in dependence of training after marathon running and treadmill (Heitkamp previously performed; mild endurance exercise et al. 1996). In addition, the same group observed o β 50% V 2max resulted in no change of -endorphin, a tendency to elevated basal ACTH in endurance whereas exercise at 66% was reported to increase β- training (Heitkamp et al. 1998). endorphin (Viru & Tendzegolskis 1995). Luger et al. (1987) examined plasma ACTH, POMC release under extreme physical stress conditions. cortisol and lactate responses in sedentary subjects, Ultramarathon foot race is extreme physical stress; moderately trained and highly trained runners. Basal resting serum ACTH and β-endorphin levels were concentrations of plasma ACTH were elevated in significantly elevated above normal range (altered proopiomelanocortin and exercise 141

baseline hormonal state), ACTH remained elevated, Kraemer 1997). Whereas short-term overtraining and β-endorphin plasma concentration was within (overreaching) can be reversed by a more prolonged the normal range without significant change and period of recovery, further exposure to stressors was interpreted as a hormonal adaptation to pro- induces overtraining syndrome. Substrates such as longed stress (Pestell et al. 1989). A more pronounced lactate, urea, enzymes (e.g. creatine kinase) or hor- response in trained subjects was only seen under mones (e.g. cortisol, testosterone, growth hormone) extreme exercise conditions (Farrell et al. 1987). are of interest in overtraining research, but we will Tharp already in 1975 made the following state- only report on ACTH and β-endorphin as represen- ments concerning the role of glucocorticoids in exer- tative POMC derivatives in the blood in overtrain- cise and training: Exhaustion produces a decrease ing analysis. in plasma glucocorticoid, which may represent a Urhausen et al. (1995) in parallel with Barron et al. defense mechanism to prevent depletion of body (1985) and Fry and Kraemer (1997) (for review see resources. Chronic exercise training produces Urhausen et al. 1995; Fry & Kraemer 1997) observed adrenal cortex hypertrophy and usually a smaller an impaired exercise-induced increase in ACTH rise in plasma glucocorticoids during an acute exer- after exhausting short-endurance test at 110% of cise bout than that obtained with non-trained sub- the individual threshold or in response to insulin- jects. The changes in glucocorticoid response during induced hypoglycemia in overtrained marathon training appear to be produced by decreased runners, which was ascribed to a negative feedback responsiveness to the adrenal cortex itself to ACTH through cortisol, depletion of the pituitary ACTH stimulation and possibly by adaptation of the HPA pool or change of the intracellular homeostasis axis which reduces the ACTH released in response (Urhausen et al. 1995). Kraemer et al. (1989b) also to stress (Tharp 1975). reported an impaired exercise-induced rise of ACTH Data from Viru and Tendzegolskis (1995) demon- in athletes during overtraining. Repeated acute or γ α strated that levels of 1–17-endorphin and 1–16- chronic exposure to a particular stress results in endorphin and the ratios to β-endorphin were adaptation, whereby the HPA axis becomes less significantly higher in untrained individuals, indi- responsive to subsequent or continued exposure cating an alteration of the metabolism of β-endorphin to that particular stress. Data from Wittert et al. in the resting state and during exercise as depen- (1996) show that intense physical training leads to dent on previous training. adaptive changes in basal HPA function, includ- ing a phase shift and increased pituitary in basal HPA function, a phase shift and increased pituit- Overtraining and addiction to exercise ary ACTH secretion, but also a blunted cortisol This chapter refers to exercise physiology in a state response. of imbalance between strain of exercise training During heavy endurance training or overreach- and the athlete’s tolerance of stress which leads to ing periods, the majority of findings give evidence overtraining. The literature on overtraining is con- of a reduced adrenal responsiveness to ACTH, com- fusing because of a lack of universal terminology pensated by an increased pituitary ACTH release (Fry & Kraemer 1997). Overtraining can be defined (Lehmann et al. 1998). A decreased β-endorphin as any training performed with incomplete recovery response was also reported in overtrained athletes between bouts of exercise, leading to physical per- (Urhausen et al. 1995). There is additionally evid- formance decrements (Fry & Kraemer 1997; Raastad ence for decreased intrinsic sympathetic activity et al. 2001). A further lack in the literature is that and sensitivity of target organs to catecholamines studies of overtraining do not consequently utilize (Lehmann et al. 1998). adequate changes in exercise volume or intensity, The parasympathetic, Addison type represents or do not carefully consider rest and recovery the dominant modern type of the overtraining syn- characteristics of the training programme, or do not drome. In an early stage, despite increased pituitary deal with actual decreases in performance (Fry & ACTH release, the decreased adrenal responsiveness 142 chapter 11

is no longer compensated and the cortisol response cortisol levels. Such responses are contrary to high decreases. In an advanced stage of overtraining syn- volume resistance exercise and overtraining with drome, the pituitary ACTH release also decreases. highly aerobic activities. It becomes apparent that However, this complete pattern is only observed some of the classical signs of overtraining, based subsequent to high-volume endurance overtraining on data from endurance athletes, cannot necessarily at high caloric demands. The functional alterations be applied to overtraining resulting from highly of pituitary–adrenal axis and sympathetic system anaerobic activities (Fry & Kraemer 1997). can explain persistent performance incompetence in With ACTH and resistance exercise overtraining, affected athletes (Lehmann et al. 1998). acute ACTH levels are attenuated with increased An exercise effect to be rather elicited in the exercise intensity. When strength performance has central nervous system than in the periphery is been decreased via maximal intensity resistance addiction to exercise. However, in a study targeting exercise overtraining, circulating ACTH levels did this question, scores on exercise-dependence survey not change either at rest or after exercise (Fry & were not correlated with β-endorphin plasma levels Kraemer 1997). (Pierce et al. 1993a). Previous research has attempted to identify Neuroendocrine responses to high volume resis- endocrine markers (‘overtraining markers’) of an tance exercise overtraining can be grouped with impending or concurrent overtraining syndrome highly aerobic activities, whereas excessive resis- for both aerobic and anaerobic activities. Data from tance training intensity (anaerobic activity) produces McGowan et al. (1993) support research showing a distinctly different neuroendocrine profile (Fry & resistance exercise does not produce the significant Kraemer 1997). Cortisol concentrations in response increase in β-endorphin immunoreactivity widely to resistance exercise at increased training volume reported after endurance exercise. In addition, or increased training intensity were compared: they showed no relationship between affect (mood) Cortisol increased at rest and acute increased train- and β-endorphin response to resistance exercise. ing volume, whereas no change or a slight decrease Despite speculations on an essential participation of was observed at increased training intensity. How- hypothalamic–pituitary hormonal changes in the ever, there was a lack of increased cortisol levels pathogenesis of overtraining, only few results with with high-intensity resistance exercise overtraining. respect to athletes who were actually in a state of This might be of interest for overtraining research, overtraining are available (Urhausen et al. 1995). since it has been shown that increased circulating Training intensity effect monitoring for resistance levels of cortisol may be associated with psycholo- exercise was only addressed by Fry and colleagues gical depression (Fry & Kraemer 1997). (Fry & Kraemer 1997; Fry et al. 1998). Because of It appears quite different when anaerobic activit- the lack of association between endocrine and per- ies, such as resistance exercise, are compared with formance alterations, they argued that the type of aerobic activities. Many of the overtraining symp- exercise does not readily permit use of hormonal toms identified for aerobic exercise have not been alterations and endocrine adaptations, respectively, reported for anaerobic overtraining protocols, and to monitor impending overtraining. In addition, altering some of the acute training variables for it appeared that short-term, high-relative-inten- resistance exercise results in a variety of different sity resistance exercise overtraining may not be physiological responses (Fry et al. 1998). Exercise- successfully monitored via circulating hormonal induced cortisol levels decreased when training concentrations. volume was doubled, arguing that the presence Implications of these neuroendocrine responses of increased resting levels of cortisol contributes to overtraining for adequate training should be an exhaustion of the HPA axis, thus preventing an managed by dosing the training stresses in a vari- adequate cortisol response to acute stress. In the able manner to avoid overtraining with long-term presence of maximal intensity resistance exercise disruption of homeostasis (Fry & Kraemer 1997). overtraining, no changes were observed for resting The magnitude of temporary fatigue and recovery proopiomelanocortin and exercise 143

rate (after heavy resistance exercise) may be an indica- was interpreted as a partial predominance of silent tion of effectiveness for long-term adaptations of ischemia during psychological stress (Miller et al. the neuromuscular system, which make it necessary 1993). In patients with CAD and exercise-induced to start the next training session after complete ischemia, public speaking produces psychological recovery (Ahtiainen et al. 2003). stress manifested by increased cardiovascular reac- tivity and causes an increase in plasma β-endorphin levels that was significantly correlated with pain Cardiovascular effects thresholds (Sheps et al. 1995). Sex differences in Exercise can be regarded as one of many stressors chest pain at exercise conditions were observed provoking cardiovascular reactions of the organism by the same authors. Women reported chest pain and therefore activating the pituitary POMC sys- more often than men during daily activities and tem. Shen et al. (1992) observed an exercise-induced during laboratory mental stressors but not during rise of β-endorphin plasma concentrations also in exercise. Men had lower scores than women on the presence of naloxone. Their results confirm a measures of depression or trait anxiety. Women had rise in intrinsic heart rate (heart rate following auto- significantly lower plasma β-endorphin levels at nomic blockade) and β-endorphin concentrations rest and at maximal mental stress. They concluded with acute exercise, but indicate that the changes in that their results reflect sex differences in the affect- intrinsic heart rate are not β-endorphin related. ive and discriminative aspects of pain perception, The β-endorphin response to exercise is of special which may help to explain sex-related differences in interest in patients with coronary artery disease clinical presentations (Sheps et al. 2001). (CAD). Letizia et al. (1996) observed patients with Plasma β-endorphin levels were studied due to suspected CAD and those with definite CAD under exercise-induced ischemia in patients with CAD. cycloergometric stress tests. At peak exercise There was no significant difference in plasma β- plasma levels of β-endorphin and ACTH increased endorphin levels during or after exercise between concomitantly in subjects that did not exhibit clinic symptomatic and asymptomatic patients; thus, the and electrocardiogram (ECG) signs of ischemia differences in circulating levels of β-endorphin and during stress test. β-Endorphin, ACTH and cortisol also ACTH were not associated with the presence or increased further during recovery. β-Endorphin absence of pain (Heller et al. 1987; Marchant et al. significantly increased at peak exercise in patients 1994). This observation is in accordance to the above with CAD and negative stress test, but ACTH and mentioned observation of Droste et al. (1991) that cortisol plasma levels were not significantly modi- exercise-induced elevation in pain threshold was fied. On the contrary in those patients with a posit- not related to plasma β-endorphin levels. ive stress test, the plasma levels of β-endorphin and POMC-correlated peptides were not modified. As Influence of age, race and gender an interpretation of this behavior they assumed the rise in β-endorphin concentration observed at peak Effects of chronic exertion on β-endorphin and the in patients with CAD and negative stress test as relationship to melatonin secretion were studied in associated with painless ischemia. well-trained athletes by Appenzeller and Wood Patients with asymptomatic ischemia on exercise (1992): β-endorphin and melatonin increased after had a significantly greater β-endorphin response exercise. Interestingly, the magnitude of this in- than those with angina. Public speaking elicited a crease was age-dependent. Chronic exertion was significantly larger β-endorphin increase than did associated with a decrease in exercise induced exercise. Patients with silent versus painful ischemia opioid release and in such individuals whose mela- experience had a greater β-endorphin response to tonin secretion was not β-endorphin related. The exercise. However, the β-endorphin response to first study demonstrating that older men can make a speech stressor was greater than to exercise. physiological adaptations in the endocrine system Increased β-endorphin response to a speech stressor with resistance training was made by Kraemer, W.J. 144 chapter 11

et al. (1999). They examined the adaptations of the tions, no major differences have yet been demon- endocrine system to heavy-resistance training in strated, with the exception of a slightly lower basal younger versus older men in a strength-power β-endorphin IRM concentration in women as com- training program. The amount of cortisol produced pared to men independent of the time of the wom- at resting levels was reduced and the response to the en’s menstrual cycle. resistance exercise stress was lower in the older ACTH, cortisol and β-endorphin were measured men. The decrease in resting concentrations of cor- in a treadmill run at 80% of a previously determined tisol throughout the training program in the older maximum heart rate of male and female runners men without significant changes in the ACTH con- and compared to the concentrations of sedentary centrations indicated that the ACTH receptors in the men and women by Kraemer, R.R. et al. (1989). The adrenal gland may have been ‘down-regulated’. run resulted in no rise in β-endorphin, ACTH and Changes in exercise-induced cortisol responses cortisol. β-Endorphin values were significantly observed after exercise were apparently mediated higher in men than in women. No sex or training by a reduction in ACTH responses to exercise stress. differences were seen with respect to change of hor- The inability to engage similar hormonal mech- mone concentrations over the course of the run. anisms in response to heavy-resistance exercise Their data indicated that gender and training do not training was interpreted as meaning that the plastic- affect ACTH and cortisol concentrations before, ity of the endocrine system in older men was altered during and after treadmill running at 80% of max- or impaired. imum heart rate, whereas β-endorphin concentra- Exhausting bicycle exercise in post-menopausal tions were higher in men under these conditions. women induced a strong release of ACTH, cortisol Plasma ACTH response to exercise was signific- and prolactin, and natural killer cell activity. Van- antly attenuated in lactating women performing o der-Pompe et al. (2001) assumed that low vigor graded treadmill exercise finally to elicit 90% V 2max in post-menopausal women interferes with the uptake in comparison to non-lactating women. endocrine and immune responses to exhausting These results were interpreted to mean that stress- exercise. responsive neurohormonal systems were restrained Yanovski et al. (2000) have shown that plasma in lactating women (Altemus et al. 1995). ACTH of African-American women was significantly HPA and hypothalamic–pituitary–gonadal (HPG) greater than that of Caucasian women, but that this axis modification, combined with cognitive impair- was not accompanied by greater plasma cortisol ments, have been reported in elderly subjects and concentrations. Plasma ACTH of African-American related to physical training status. Data from Struder women contained many ‘non-intact’ ACTH frag- et al. (1999) suggest that elderly endurance athletes ments which were not found in Caucasian women. reveal a prolonged secretion of glucocorticoids. Plasma ACTH, measured after intense exercise, was significantly greater in African-American women Physiological and pathophysiological aspects of than in Caucasian women, but plasma cortisol after exercise induced POMC release and consequences exercise was not different. The ACTH of African- to homeostasis of energy and nutrition American and Caucasian women did not appear to be equipotent at adrenal melanocortin-2 receptors Exercise can be regarded as one of many stressors because the greater ACTH of African-American provoking metabolic reactions of the organism and women did not lead to greater cortisol secretion. therefore activating the pituitary POMC system. Data from Goldfarb et al. (1998) suggested that However, in contrast to most of the other stressors, o β women cycling at 80% V 2max had a similar - there is no doubt that the organism tries to adapt to endorphin response to men independent of their this stressor not by reduction of morphology and menstrual cycle. Comparing ACTH or β-endorphin functions to the original state before exposition to plasma levels of female athletes with those of male exercise, but by conversion of morphology and athletes or volunteers under basal or exercise condi- functions to an altered, i.e. to a stressor-adapted proopiomelanocortin and exercise 145

state, thus counteracting future stress situations of nucleus or from the posterior pituitary, respect- the same type. ively, for enhancement of POMC fragment release Sustained hyperglycemia was observed to not (Inder et al. 1998). be a stimulus to enhanced β-endorphin secretion Thus, although the prerequisites or the conse- into plasma and this lack of response was not quences of anaerobic metabolism are apparently effected by prior exercise (Farrell et al. 1986). Lower closely related to the activation of the POMC β-endorphin levels were found in patients with system, lactate by itself does not seem to directly diabetes and silent myocardial ischemia at rest, stimulate the hypothalamic structures responsible following exercise, as compared with those with for pituitary POMC system activation (Petrides et al. silent myocardial ischemia who were not diabetic. 1999). Taylor et al. (1994) assumed base excess as the Hikita et al. (1993) assumed a less significant role best indicator of β-endorphin release or the primary of β-endorphin in diabetic patients than in non- stimulus for activation of the pituitary POMC sys- diabetic ones. tem. In contrast, endurance exercise (marathon and Angelopoulos (2001) observed comparatively steady-state cycling) reported that constant lactate higher β-endorphin concentrations during exercise levels were not related to β-endorphin increase under opioid antagonism (naloxone) as compared (Goldfarb et al. 1991; Heitkamp et al. 1996). Blood to placebo trial and assumed a positive feedback lactate levels do not appear to be related to HPA loop for β-endorphin on plasma glucose regula- hormone plasma concentrations at high exercise tion. Tabata et al. (1991) observed an abolished intensities (Kraemer, W.J. et al. 1989b). It might be corticotropin-releasing factor and ACTH increase associated with stimulation of testosterone secre- o after exercise at workloads of 50% V 2max until tion or plasma testosterone elevation during exer- exhaustion of healthy young men when their blood cise, but this requires further investigation (Raastad glucose concentrations were maintained at the pre- et al. 2000). However, lactate was significantly exercise level. The critical level for triggering the correlated with β-endorphin, ACTH and cortisol pituitary–adrenal–cortical axis was 3.3 mmol. Inder (Kraemer, W.J. et al. 1989a). Low volume resistive et al. (1998) demonstrated a significant rise in peri- exercise elevates lactate concentrations without pheral CRH at submaximal exercise, but this was altering endorphin (Kraemer, R.R. et al. 1996). not associated with hypoglycemia. Cortisol rise in Metabolic response to exercise stress should be physical exercise or during recovery may be initi- envisaged as a possible candidate for POMC frag- ated by the drop in blood glucose and may prevent ment function (Weissman 1990). In view of the fact the inflammatory and immune reactions from over- that the release of POMC fragments is obviously shooting and so may stabilize homeostasis of the dependent on anaerobic state with lactate levels organism (De Vries et al. 2000). beyond the anaerobic threshold, and that metabolic Numerous studies (for review see Steinberg & acidosis probably is a direct stimulus for POMC Sykes 1985; Cumming & Wheeler 1987; Sforzo 1989; fragment release, the most likely functional target Schwarz & Kindermann 1992; Hoffmann et al. 1996; of POMC fragment release would be counterbalanc- Goldfarb & Jamurtas 1997; Vassilakopoulos et al. ing the obvious metabolic derailment, exercise is 1999) have shown that activation of the pituitary responsible for, by increased energy supply of the POMC system is achieved by a certain degree of skeletal muscle (Knudtzon 1986; Evans, A.A. et al. metabolic demand, which is characterized by blood 1997). Since an anaerobic, acidotic state appears lactate levels beyond the anaerobic threshold, to be the stimulus for β-endorphin release, β- which again are reached upon incremental or short- endorphin targets might well be parts of a peri- term anaerobic exercise, or which also can be pheral system responsible for metabolic homeostasis reached after prolonged aerobic endurance exer- or acid/base equilibrium (Schulz et al. 2000). cise. Apparently the anaerobic state leads to the Acute amino acid supplementation enhanced release of CRH and arginin–vasopressin, known to the ACTH, luteinizing hormone (LH) and follicle- be released from the hypothalamic paraventricular stimulating hormone (FSH) response to CRH and 146 chapter 11

gonadotropin-releasing hormone (GnRH) (Di Luigi observations from Nagao et al. (2000) suggest that et al. 1999), whereas acetylsalicylic acid influences exercise-induced catecholamines modulate the ex- ACTH, β-endorphin and cortisol responses to exer- pression of adhesion molecules on natural killer cise-related stress in humans. Their data confirm cells, resulting in a mobilization of natural killer a role of prostaglandin in these responses. It might cells into the circulation. be of interest, because of the large use of anti- The mechanisms underlying exercise-associated inflammatory drugs in athletes, whether the inter- immune changes are multifactorial. Altered plasma action between acetylsalicylic acid and hormones glucose has also been implicated in decreasing stress might positively or negatively influence health hormone levels and thereby influencing immune status (Di Luigi et al. 2001). function (Nieman & Pedersen 1999). A further candidate related to stress, immunolo- gical function, insulin sensitivity and cardiovascular POMC and exercise and immunological influence disorders is dehydroepiandrosterone, secreted by Exercise can be regarded as one of many stressors the adrenal cortex in response to ACTH (for review provoking immunologic reactions of the organism see Kroboth et al. 1999). and therefore activating the pituitary POMC sys- Strenuous exercise recruits neutrophils and lym- tem. Effects on the immune system elicited by β- phocytes into the circulation and, during recovery, endorphin (Teschemacher et al. 1990b) are still a lymphocyte concentration rapidly decrease, result- matter of speculation. ing in lymphocytopenia if exercise intensity and An attractive candidate to be considered would duration are of sufficient magnitude. A novel be a POMC fragment influence on the immune finding of a study by Ronsen et al. (2002) was the system. Interactions of β-endorphin (Sibinga & more pronounced neutrophilia and lymphocyto- Goldstein 1988) or ACTH (via corticosteroid release penia during and after a second bout of exercise in from the adrenals) with cells of the immune system the short-rest trial compared to the long-rest trial. are well testified, such as effects of stress in general They showed that repeating exercise with only a (Fricchione & Stefano 1994) and of physical exercise few hours rest results in magnified neuroendocrine in particular (Jonsdottir et al. 1997). These effects stress responses and conceived that performing could contribute to both activation as well as daily repeated exercise sessions, thus imposing sub- suppression of separate functions of the immune stantial physical, psychological and metabolic stress, system responsible for defense against infections could lead to both adaptive and maladaptive altera- or necessary for morphological alterations in terms tions in the immune system (Ronsen et al. 2002). of elevated capability to cope with future exercise stress. Correlations between POMC derivatives and It is suggested that exercise can be employed as gonadal steroids or gonadotropins under a model of temporary immune suppression that exercise conditions occurs after severe physical stress. The increase in catecholamines and growth hormone mediate the To investigate the mechanisms of blunted adrenocor- acute effects of exercise on neutrophils, whereas tical responsiveness to exercise and mild hypercor- cortisol may be responsible for maintaining lym- tisolism in eumenorrheic and amenorrheic runners phopenia and neutrocytosis after exercise of long in comparison to eumenorrheic sedentary women duration. Lastly, the role of β-endorphin is less ACTH stimulation, tests were performed in the clear, whereas the cytokine response is closely presence and absence of dexamethasone suppres- related to muscle damage. However, POMC does sion. The cortisol response to stimulation in amen- not seem to be directly involved in the elevated orrheic runners was blunted in the presence of their cytokine level (Pedersen et al. 1997; Pedersen & mild hypercortisolism and appeared to be due to Hoffmann-Goetz 2000) or elevations of basal natural a normal limitation in maximal adrenal secretory killer cell activity (Dishman et al. 2000). However, capacity (De Souza et al. 1994). proopiomelanocortin and exercise 147

Ovulatory and eumenorrheic runners underwent elevated above the normal mean and β-endorphin exercise in the follicular and luteal phases and after was within the normal range, without significant GnRH agonist desensitization. Baseline peripheral change. A significant increase in cortisol was seen, β-endorphin and cortisol levels were not different with no change in cortisol-binding globulin. As a between the eumenorrheic and oligomenorrheic model of chronic physical stress, the results demon- groups. A significant increase in β-endorphin levels strated a significantly altered baseline hormonal in response to exercise occurred only in the eumen- state as reflected in the primary mediators of the orrheic group after GnRH agonist desensitization. stress response, the catecholamines and the HPA They concluded that alterations in menstrual cyclic- axis. The athletes’ response to severe exercise was ity and ovulation in conditioned runners were distinct from that of untrained individuals in who not due to an increase in opioid tone. The hypo- conjugated catecholamines decrease and ACTH thalamic–gonadotropic axis appeared to be intact in increase. This was interpreted as a possible hor- oligomenorrheic runners. Sex steroid administra- monal adaptation to prolonged stress (Pestell et al. tion had no effect on basal β-endorphin levels, but 1989). this probably was not due to pre-existing increased opioid tone (Meyer et al. 1999). New aspects of the functional Menstrual cycle alterations in athletes are accom- significance of POMC and interpretation panied by shortening of the luteal phase and sec- of stress definitions ondary amenorrhoea. So far it appears that the reduced LH secretion might be caused by an in- Functional significance of POMC derivatives creased CRH secretion inhibiting GnRH release. In addition, increased CRH tone leads to increased β- The functional significance of most of the POMC endorphin levels which also inhibits GnRH release systems scattered over the organism is unknown. (Keizer & Rogol 1990). So far this holds for the pituitary POMC system. It has been known for a very long time that the Effects of oral contraceptives on plasma β-endorphin. cleavage products of POMC are released into the Plasma immunoreactive β-endorphin levels at rest blood under ‘stress’ conditions. The pituitary gland were higher in non-pill users than in pill users. as an endocrine organ is destined for sending hor- Corticotropin levels at rest did not differ between monal messages to tissues of the whole organism o the pill and non-pill users. At 60% V 2max a slight via the cardiovascular compartment, receiving the increase was found in the concentrations of corti- orders for this kind of message transfer mainly from cotropin and β-endorphin in the non-pill but not in the central nervous system, but also, to a consider- o β the pill users. At 90% V 2max, plasma -endorphin ably smaller extent, over the cardiovascular com- and corticotropin levels increased significantly in partment itself. Apparently, the pituitary contains both groups. A threshold elevation of the intensity several message transfer systems thought for of exercise required to increase β-endorphin and specific transfer tasks such as the gonadotropic or corticotropin secretion by the use of oral contracept- the thyreotropic system. For each of these specific ives was concluded (Rahkila & Laatikainen 1992). message transfer systems, specific orders based on To examine individual hormonal responses to specific stimuli can be postulatedaas well as specific extreme physical stress, blood samples were taken targets in the periphery to be met by the hormonal from highly trained athletes before and within messengers released from the pituitary into the finishing a 1000-km ultramarathon foot race. ACTH blood (Mooren et al. 2005). levels were significantly elevated above the normal In fact, for the pituitary POMC system a big num- range. Immunoreactive β-endorphin, growth hor- ber of stimuli triggering its activation have been mone, prolactin, testosterone, cortisol and cortisol- demonstrated, which can all be comprised under binding globulin were within the normal range. the term ‘stress’. Physical as well as emotional stres- Catecholamines and ACTH remained significantly sors have been shown to induce the release of 148 chapter 11

POMC fragments into the cardiovascular compart- Methodological aspects of determination of ment. These stressors range from fear to pain, from POMC derivatives physical exercise and training to exhaustion exer- cise, from metabolic shifts to serious injury due to A large individual variation in the β-endorphin/β- parturition, surgery or even accidents. Peptides LPH response was noted by Farrell et al. (1982) and derived from POMC are, in function, as hormones Sheps et al. (1988). In addition, some methodological when they are secreted from pituitary cells into the difficulties concerning the determination of β- bloodstream. endorphin still exist. Whereas the determination However, it should be mentioned that phys- of ACTH always revealed clear-cut results, in ical and psychological stressors are processed in many studies β-endorphin IRM, in addition to β- the organism in a different way: the information endorphin, was determined to may be contain at launched upon peripheral tissue injury, in principle least 10 β-endorphin derivatives. In fact, recent carrying the character of life-threat, is immedi- findings showed that β-endorphin IRM was more ately signaled via spinal cord to the brain and is identical with β-LPH than with β-endorphin. In directly linked to the pituitary via hypothalamic studies, wherein β-LPH was determined in com- structures. In contrast, emotional stressors or the parison with ACTH or β-endorphin IRM using β- psychological components of physical stressors are LPH specific determination methods (Oleshansky et thought to be received and processed by supra- al. 1990, Petraglia et al. 1988), β-LPH reached about spinal structures such as the limbic system in a the same plasma concentrations as β-endorphin pretty complicated way before the resulting infor- IRM or ACTH. In contrast, further studies with mation is transferred via hypothalamic structures selective or even highly specific assays for authentic to the pituitary for activation of the POMC system β-endorphin showed that under exercise conditions (Herman & Cullinan 1997), i.e. the activation of the the plasma concentrations of authentic β-endorphin HPA axis. were low (Farrell et al. 1987; Engfred et al. 1994) to Since the isolation of the first endogenous minimal (Harbach et al. 2000; Schulz et al. 2000) opioids in 1975, interest has focused on deducing in comparison with ACTH or β-LPH. Fragments the analgesic properties of β-endorphin because from different regions of β-endorphin (1–31) might the amino acid sequence of its N-terminus is able have quite different functions, as demonstrated on to bind with opiod receptors. However, exercise- immune cells: β-endorphin interacts through its induced elevation in pain threshold was not related N-terminal fragment with opioid receptors whereas to plasma endorphin levels (Droste et al. 1991), its interactions with binding sites on complement or whereas elevated serum β-endorphin concentra- on thymocytes occur via its C-terminus. Therefore tions induced by exercise have been linked to the identity of ‘β-endorphin’ claimed to occur in the altered pain perception (Harber & Sutton 1984). plasma upon exercise is of considerable importance β-Endorphin has been correlated with several psy- for correct conclusions in terms of a possible func- chological and physiological changes, including tion (Schulz et al. 2000). ‘exercise-induced euphoria’, ‘runners’ high’, exer- Also, POMC fragment release from the adrenal cise dependence, negative mood state changes, medulla under physical exercise cannot be excluded food intake suppression, immune suppression and (Evans, C.J. et al. 1983). Besides ACTH and β-LPH, reproductive dysfunction. But the insulation func- further POMC derivatives, which apparently repres- tion provided by the blood brain barrier precludes ent small-sized β-endorphin fragments (Wiedemann any association between endorphin and central & Teschemacher 1983), may be released upon phys- nervous activity, and, in addition, an increase of ical exercise under certain conditions. The pro- β-endorphin associated with an intensive bout of portions of the plasma concentrations of defined aerobic exercise appears to be independent of a β-endorphin fragments may also vary as dependent tendency towards exercise dependence (Pierce et al. on the state of fitness of the volunteers (Viru & 1993a). Tendzegolskis 1995). proopiomelanocortin and exercise 149

In addition to catecholamine release, there is β-Endorphin and psychological effects evidence for an acute response of the pituitary during exercise POMC system to all kinds of stressors. However, there is no clear-cut information about the func- No biochemical link was observed for β-endorphin tional significance of the POMC fragments released that might explain the possible influence of physical under stress into the cardiovascular compartment. activity on depression (Williams & Getty 1986). Although there are well-known targets such as the However, lower β-endorphin resting plasma levels adrenal gland for ACTH, the functional significance of trained subjects were related to an adaptation to of the effects elicited by the POMC fragments at exercise training and a greater emotional stability peripheral targets is not clear at all; effects by ACTH and lower depression (Lobstein et al. 1989). Endur- via corticosteroids, for example in response to fear, ance training of 8 months duration appeared to are still a matter of speculation. Thus, the pituitary decrease the resting plasma β-endorphin concen- POMC system clearly plays the role of a ‘stress- trations and the depression scores (Lobstein & responder’ but as yet it cannot be classified as a Rasmussen 1991). ‘stress-adapter’, since it is unclear how it might con- Singh et al. (1999) observed different plasma tribute to the reduction of a disturbed function to a ACTH and cortisol responses to psychological and homeostatic state (Teschemacher 2003). Although a exercise stress tests after dexamethasone treatment. couple of well-known exercise effects have been Subjects were classified as responders based on linked to POMC fragments, in particular to opioid ACTH responses to exercise. A psychological stress peptides (for review see Cumming & Wheeler 1987), test raised heart rate, blood pressure, plasma ACTH the functional significance of the POMC fragments and cortisol levels in both high responders (HRs) released under exercise conditions is still a matter and low responders (LRs). HRs tended to have of speculation. Some of the functions discussed for higher heart rates and blood pressures in anticipa- POMC derivatives released into the cardiovascu- tion of the psychological stress test than LRs. ACTH lar compartment appear to be unlikely, since the responses of HRs were higher, although not sig- respective effects are not thought to be elicited in nificantly, throughout the psychological stress test peripheral tissues but in the central nervous sys- than LRs. HRs had a significantly greater cortisol tem; for example, reduction of depressive state, response to the psychological stress than LRs. They reduced anxiety, improved self-esteem, improved suggested that the adrenal cortex of the HRs were well beingaall in all, an improvement of mood hypersensitive to ACTH and concluded that men occasionally becoming manifest as ‘runner’s high’. who are highly responsive to exercise stress were Experimental data indicate that aerobic exercise can also highly responsive to psychological stress. activate central nervous system endogenous opioid Post-traumatic stress disorder may be associated systems, as shown by altered brain opioid levels with changes in endogenous opioid peptide func- and by increased levels of cerebrospinal fluid β- tion. To test this hypothesis, Vietnam combat veter- endorphin in running rats (Hoffmann et al. 1996). ans with post-traumatic stress disorder underwent The most likely hormonal role for circulating β- a standard exercise stress test. Resting plasma β- endorphin is the modulation of the adrenal response endorphin levels were comparable between veter- to stress, by controlling the release of cortisol in re- ans and controls. However, post-exercise plasma sponse to ACTH stimulation (McLoughlin et al. 1993). β-endorphin levels were significantly higher than Further candidates for POMC fragment functions in resting levels only in the post-traumatic stress dis- the periphery are certainly influences on food uptake order patients. These data suggested a differential or reproduction, since fat tissue or gonads represent alteration in plasma β-endorphin response to exer- peripheral targets. However, central POMC sys- cise in post-traumatic stress disorder (Hamner & tems in hypothalamic areas, in either case, appear to Hitri 1992). Healthy adolescent women received be more important candidates for influence on psychological and endocrine examinations. Anxiety reproductive dysfunction (Rivier & Rivest 1991). scores and frustration tests were used. On the basis 150 chapter 11

of the results of these tests, subjects were divided In case of situations of overtraining challenge, into two groups: normal subjects and subjects with a reduced ACTH response reflects the athlete’s evidence of anxiety and/or frustration. Plasma impaired ability to cope with the stress situation. levels of ACTH and β-endorphin were measured ACTH and cortisol responses of maximal intensity under basal conditions and after physical exer- resistance exercise overtraining are contrary to cise. Basal concentrations of ACTH and cortisol high-volume resistance exercise and overtraining were similar in the two groups, whereas basal β- with highly aerobic activities. Thus, it becomes endorphin levels were significantly higher in the apparent that signs of overtraining, based on data anxiety/frustration group than in the control group. from endurance athletes, cannot necessarily be A striking increase in plasma ACTH and a slight applied to overtraining resulting from highly ana- increase of β-endorphin levels were observed in the erobic activities. Future research must address the anxiety/frustration group after exercise. Absolute possible mechanism(s) of such a cortisol response, levels of ACTH and β-endorphin after physical whether it is due to adrenal cortex exhaustion or exercise were significantly higher in this group sympathetic or other control (Fry & Kraemer 1997). than in the control group. These findings indicate POMC fragments may have influence on the increased levels of adrenocorticotropic and opioid immune system, in particular at physical exercise activity in adolescent women with high scores on stress (Jonsdottir et al. 1997). It is suggested that psychological measures of anxiety and frustration exercise can be employed as a model of temporary (Gerra et al. 1992). In addition, significant elevations immune suppression that occurs after severe phy- of β-endorphin and CRH were observed after sical stress. The increase in catecholamines (epine- running and meditation and were associated as a phrine and norepinephrine) and growth hormone positive influence (Harte et al. 1995). mediate the acute effects of exercise on neutrophils, However, in one study by McGowan et al. (1993) whereas cortisol may be responsible for maintain- no significant relationship was observed between ing lymphopenia and neutrocytosis after exercise of pre- or post-exercise plasma β-endorphin and either long duration (Pedersen et al. 1997). However, the total mood disturbance. role of β-endorphin is less clear, but the cytokine response is closely related to muscle damage, and Conclusions stress hormones do not seem to be directly involved in the elevated cytokine level. Lastly, these pheno- The secretion of POMC or POMC derivatives dur- mena make it unlikely that β-endorphin plays a ing exercise is regarded as an adaptive attempt of major immune modulatory role in the immediate the athlete’s organism to cope with different stress recruitment of natural killer cells (for review see situations and is intimately linked to a variety of psy- Pedersen & Hoffman-Goetz 2000). chological strategies which facilitate its navigation The metabolic response to exercise stress might be through a stressful environment (McCubbin 1993). a candidate for POMC fragment function (Weissman The secretion of ACTH and β-endorphin and their 1990) in support mechanisms for energy and home- increase occurs in relation to the intensity and the ostasis (Knudtzon 1986). POMC is acting in a com- duration of the physical exercise. But the exercise plex interrelationship between the endocrine, response may also be affected by the training status metabolic, cardiac, neurologic and immune systems of the individual and the population being investig- under physical exercise conditions. What can be ated (Goldfarb & Jamurtas 1997). concluded from the diverse evidence about exercise Efforts to clarify the question of β-endorphin in- and the involvement of POMC? Given the inherent volvement in exercise effects by the blockade of opi- difficulties, the following ideas are summarized: oid receptors during exercise (Strassman et al. 1989; 1 Among POMC derivatives, the most information Angelopoulos 2001) revealed controversial effects arelated to exercise conditionsais available on on β-endorphin plasma levels but did not provide ACTH, which was studied as the main representa- further insight into the mechanisms under question. tive of POMC under the aspect of stress-induced proopiomelanocortin and exercise 151

activation of the HPA axis. Later β-endorphin and 6 Glucose and lactate are substrates, which were β-LPH were recognized to be coreleased from the intensively observed in exercise physiology because pituitary into the cardiovascular compartment of changes under physical and extreme stress. under exercise conditions. ACTH, cortisol and β-endorphin seem to have an 2 It is strictly the ‘corticotropic’ part of the pituitary influence. However, it is not yet clear in which way POMC system that is activated under physical they may contribute in stabilizing the homeostasis stress conditions. of the organism. 3 β-Endorphin plasma levels have universally been 7 Correlations between POMC derivatives and reported to increase with exercise. However, the gonadal steroids or gonadotropins are well testified. large individual variation in the β-endorphin/β- 8 Exercise is one of many stressors provoking LPH response and the methodological difficulties immunologic reactions and therefore activating the concerning the determination of authentic β-endor- pituitary POMC system. Effects of β-endorphin and phin are involved in the problem, and the biological ACTH can contribute to both activation as well as significance of β-endorphin is still not elucidated. suppression of the immune system. 4 Training and extreme physical stress lead to The response of POMC fragments, such as ACTH different ACTH and β-endorphin responses. The and β-endorphin, to exercise stress is complex and, stimulus responsible for β-endorphin release under in spite of multiple possible solutions, not yet eluci- anaerobic exhaustive exercise conditions is prob- dated. Despite evidence for an acute response of ably acidosis, whereas non-exhaustive performance, the pituitary POMC system to exercise stress (‘stress such as heavy resistance exercise, does not lead to sensor’), it cannot be classified as a ‘stress regulator’ an increase of endorphin levels. or ‘stress adapter’. Further investigation is neces- 5 Age, gender and race have an important influence saryain addition to the tremendous attempts of on POMC derivatives under exercise conditions, recent yearsato define the functional significance of but further investigation of this influence on adapta- the POMC fragments released at exercise stress. tion to acute and chronic exercise is necessary.

References

Adams, H.A. & Hempelmann, G. (1991) immunoreactivity during high-intensity Journal of Applied Physiology 63(6), The endocrine stress reaction in exercise with and without opiate 2499–2501. anesthesia and surgeryaorigin and blockade. European Journal of Applied Castro, M.G. & Morrison, E. (1997) significance. Anaesthesiologie Physiology 86(1), 92–96. Post-translational processing of Intensivmedizin Notfallmedizin Appenzeller, O. & Wood, S.C. (1992) proopiomelanocortin in the pituitary Schmerztherapie 26(6), 294–305. Peptides and exercise at high and and in the brain. Critical Reviews in Ahtiainen, J.P., Pakarinen, A., Kraemer, low altitudes. International Journal Neurobiology 11, 35–57. W.J. & Häkkinen, K. (2003) Acute of Sports Medicine 13 (suppl. 1), Cumming, D.C. & Wheeler, G.D. (1987) hormonal and neuromuscular responses S135–S140. Opioids in exercise physiology. Seminars and recovery to forced vs. maximum Barron, J.L., Noakes, T.D., Levy, W., Smith, in Reproductive Endocrinology 5, 171–179. repetitions multiple resistance exercises. C. & Millar, R.P. (1985) Hypothalamic De Meirleir, K., Naaktgeboren, N., Van International Journal of Sports Medicine dysfunction in overtrained athletes. Steirteghem, A. et al. (1986) β-Endorphin 24(6), 410–418. Journal of Clinical Endocrinology and and ACTH levels in peripheral blood Akil, H., Watson, G., Young, E. et al. (1984) Metabolism 60, 803–806. during and after aerobic and anaerobic Endogenous opioids: biology and Bertagna, X. (1994) Proopiomelanocortin- exercise. European Journal of Applied function. Annual Revisions of Neuroscience derived peptides. Endocrinology and Physiology 55(1), 5– 8. 7, 223–255. Metabolism Clinics of North America 23(3), De Souza, M.J., Luciano, A.A., Arce, J.C., Altemus, M., Deuster, P.A., Galliven, E., 467–485. Demers, L.M. & Loucks, A.B. (1994) Carter, C.S. & Gold, P.W. (1995) Bicknell, A.B., Savva, D. & Luciano, A.A. Clinical tests explain blunted cortisol Suppression of hypothalamic– (1996) Proopiomelanocortin and adrenal responsiveness but not mild pituitary–adrenal axis responses to function. Endocrine Research 22(4), hypercortisolism in amenorrheic stress in lactating women. Journal of 358–393. runners. Journal of Applied Physiology Clinical Endocrinology & Metabolism Buono, M.J., Yeager, J.E. & Sucec, A.A. 76(3), 1302–1309. 80(10), 2954–2959. (1987) Effect of aerobic training on the De Vries, W.R., Bernards, N.T., de Rooij, Angelopoulos, T.J. (2001) β-Endorphin plasma ACTH response to exercise. M.H. & Koppeschaar, H.P. (2000) 152 chapter 11

Dynamic exercise discloses different exercise in endurance-trained athletes. Fry, A.C., Kraemer, W.J. & Ramsey, L.T. time-related responses in stress Clinical Endocrinology 48(4), 493–501. (1998) Pituitary–adrenal–gonadal hormones. Psychosomatic Medicine 62(6), Engfred, K., Kjaer, M., Secher, N.H. et al. responses to high-intensity resistance 866–872. (1994) Hypoxia and training-induced exercise overtraining. Journal of Applied Denef, C. & Van Bael, A. (1998) A new adaptation of hormonal responses to Physiology 85(6), 2352–2359. family of growth and differentiation exercise in humans. European Journal of Ganong, W.F., Dallman, M.F. & Roberts, factors derived from the N-terminal Applied Physiology 68(4), 303–309. J.L. (eds.) (1987) Annals of the New York domain of proopiomelanocortin (N- Evans, A.A., Khan, S. & Smith, M.E. (1997) Academy of Sciences, vol. 512. The POMC). Comparative Biochemistry and Evidence for a hormonal action of β- Hypothalamic–Pituitary–Adrenal Axis Physiology. Part C, Pharmacology, endorphin to increase glucose uptake in Revisited. New York Academy of Toxicology & Endocrinology 119, 317–324. resting and contracting skeletal muscle. Sciences, New York. Deschenes, M.R., Giles, J.A., McCoy, R.W. Journal of Endocrinology 155(2), 387–392. Gerra, G., Volpi, R., Delsignore, R. et al. et al. (2002) Neural factors account for Evans, C.J., Erdelyi, E., Weber, E. & (1992) ACTH and β-endorphin strength decrements observed after Barchas, J.D. (1983) Identification of responses to physical exercise in short-term muscle unloading. American pro-opiomelanocortin-derived peptides adolescent women tested for anxiety Journal of Physiology. Regulatory, in the human adrenal medulla. Science and frustration. Psychiatry Research 41(2), Integrative and Comparative Physiology 221(4614), 957–960. 179–186. 282(2), R578–R583. Evans, V.R., Manning, A.B., Bernard, L.H., Goldfarb, A.H. & Jamurtas, A.Z. (1997) Di Luigi, L., Guidetti, L., Pigozzi F. Chronwall, B.M. & Millington, W.R. β-Endorphin response to exercise. An et al. (1999) Acute amino acids (1994) α-Melanocyte-stimulating update. Sports Medicine 24(1), 8–16. supplementation enhances pituitary hormone and N-acetyl-β-endorphin Goldfarb, A.H., Hatfield, B.D., Potts, J. & responsiveness in athletes. Medicine and immunoreactivities are localized in the Armstrong, D. (1991) β-Endorphin time Science in Sports and Exercise 31(12), human pituitary but are not restricted course response to intensity of exercise: 1748–1754. to the zona intermedia. Endocrinology effect of training status. International Di Luigi, L., Guidetti, L., Romanelli, F., 134(1), 97–106. Journal of Sports Medicine 12(3), 264–268. Baldari, C. & Conte, D. (2001) Facchinetti, F., Panerai, A.E., Petraglia, F. Goldfarb, A.H., Jamurtas, A.Z., Kamimori, Acetylsalicylic acid inhibits the pituitary et al. (1987) Proopiomelanocortin-related G.H. et al. (1998) Gender effect on β- response to exercise-related stress in peptides in feto-placental structures endorphin response to exercise. Medicine humans. Medicine and Science in Sports throughout pregnancy. Experimental and Science in Sports and Exercise 30(12), and Exercise 33(12), 2029–2035. Clinical Endocrinology 89(1), 48–54. 1672–1676. Diego Acosta, A.M., Garcia, J.C., Farrell, P.A., Gates, W.K., Maksud, M.G. & Hamner, M.B. & Hitri, A. (1992) Plasma Fernandez-Pastor, V.J. et al. (2001) Morgan, W.P. (1982) Increases in β-endorphin levels in post-traumatic Influence of fitness on the integrated plasma β-endorphin/β-lipotropin stress disorder: a preliminary report on neuroendocrine response to aerobic immunoreactivity after treadmill response to exercise-induced stress. exercise until exhaustion. Journal of running in humans. Journal of Applied Journal of Neuropsychiatry and Clinical Physiology and Biochemistry 57(4), Physiology 52(5), 1245–1249. Neuroscience 4(1), 59–63. 313–320. Farrell, P.A., Garthwaite, T.L. & Gustafson, Harbach, H., Hell, K., Gramsch, C. et al. Dishman, R.K., Warren, J.M., Hong, S. et al. A.B. (1983) Plasma adrenocorticotropin (2000) β-Endorphin (1–31) in the plasma (2000) Treadmill exercise training blunts and cortisol responses to submaximal of male volunteers undergoing physical suppression of splenic natural killer cell and exhaustive exercise. Journal of exercise. Psychoneuroendocrinology 25(6), cytolysis after footshock. Journal of Applied Physiology 55(5), 1441–1444. 551–562. Applied Physiology 88(6), 2176–2182. Farrell, P.A., Mikines, K.J., Bach, F.W., Harber, V.J. & Sutton, J.R. (1984) Donevan, R.H. & Andrew, G.M. (1987) Sonne, B. & Galbo, H. (1986) Plasma Endorphins and exercise. Sports Medicine Plasma β-endorphin immunoreactivity β-endorphin immunoreactivity: effects 1(2), 154–171. during graded cycle ergometry. Medicine of sustained hyperglycemia with and Harte, J.L., Eifert, G.H. & Smith, R. (1995) and Science in Sports and Exercise 19(3), without prior exercise. Life Sciences The effects of running and meditation 229–233. 39(11), 965–971. on β-endorphin, corticotropin-releasing Droste, C., Greenlee, M.W., Schreck, M. & Farrell, P.A., Kjær, M., Bach, F.W. & hormone and cortisol in plasma, and Roskamm, H. (1991) Experimental pain Galbo, H. (1987) β-Endorphin and on mood. Biological Psychology 40(3), thresholds and plasma β-endorphin adrenocorticotropin response to 251–265. levels during exercise. Medicine and supramaximal treadmill exercise in Heitkamp, H.C., Schmid, K. & Science in Sports and Exercise 23(3), trained and untrained males. Acta Scheib, K. (1993) β-Endorphin and 334–342. Physiologica Scandinavica 130(4), 619–625. adrenocorticotropic hormone Duclos, M., Corcuff, J.B., Rashedi, M., Fricchione, G.L. & Stefano, G.B. production during marathon and Fougere, V. & Manier, G. (1997) Trained (1994) The stress response and incremental exercise. European Journal versus untrained men: different autoimmunoregulation. Advanced of Applied Physiology 66(3), 269–274. immediate post-exercise responses of Neuroimmunology 4(1), 13–27. Heitkamp, H.C., Huber, W. & pituitary adrenal axis. A preliminary Fry, A.C. & Kraemer, W.J. (1997) Scheib, K. (1996) β-Endorphin and study. European Journal of Applied Resistance exercise overtraining and adrenocorticotrophin after incremental Physiology 75(4), 343–350. overreaching. Neuroendocrine exercise and marathon runningafemale Duclos, M., Corcuff, J.B., Arsac, L. et al. responses. Sports Medicine 23(2), responses. European Journal of Applied (1998) Corticotroph axis sensitivity after 106–129. Physiology 72(5–6), 417–424. proopiomelanocortin and exercise 153

Heitkamp, H.C., Schulz, H., Rocker, K. European Journal of Applied Physiology suspected and confirmed ischemic heart & Dickhuth, H.H. (1998) Endurance and Occupational Physiology 58(8), disease during exercise. Acta Cardiologica training in females: changes in 845–851. 51(1), 27–36. β-endorphin and ACTH. International Kraemer, R.R., Acevedo, E.O., Lobstein, D.D. & Rasmussen, C.L. Journal of Sports Medicine 19(4), 260–264. Dzewaltowski, D. et al. (1996) Effects (1991) Decreases in resting plasma β- Heller, G.V., Garber, C.E., Connolly, M.J. of low-volume resistive exercise on β- endorphin and depression scores after et al. (1987) Plasma β-endorphin levels in endorphin and cortisol concentrations. endurance training. Journal of Sports silent myocardial ischemia induced by International Journal of Sports Medicine Medicine and Physical Fitness 31(4), exercise. American Journal of Cardiology 17(1), 12–16. 543–551. 59(8), 735–739. Kraemer, W.J., Fleck, S.J., Callister, R. et al. Lobstein, D.D., Rasmussen, C.L., Dunphy, Herman, J.P. & Cullinan, W.E. (1997) (1989a) Training responses of plasma G.E. & Dunphy, M.J. (1989) β-endorphin Neurocircuitry of stress: central control β-endorphin, adrenocorticotropin, and and components of depression as of the hypothalamo–pituitary– cortisol. Medicine and Science in Sports powerful discriminators between adrenocortical axis. Trends in and Exercise 21(2), 146–153. joggers and sedentary middle-aged Neurosciences 20(2), 78–84. Kraemer, W.J., Patton, J.F., Knuttgen, H.G. men. Journal of Psychosomatic Research Hikita, H., Kurita, A., Takase, B. et al. et al. (1989b) Hypothalamic–pituitary– 33(3), 293–305. (1993) Usefulness of plasma β- adrenal responses to short-duration Loh, Y.P. (1992) Molecular mechanisms of endorphin level, pain threshold and high-intensity cycle exercise. Journal of β-endorphin biosynthesis. Biochemical autonomic function in assessing silent Applied Physiology 66(1), 161–166. Pharmacology 44, 843–849. myocardial ischemia in patients with Kraemer, W.J., Hamilton, A.J., Gordon, Luger, A., Deuster, P.A., Kyle, S.B. et al. and without diabetes mellitus. American S.E. et al. (1991) Plasma changes in β- (1987) Acute hypothalamic–pituitary– Journal of Cardiology 72(2), 140–143. endorphin to acute hypobaric hypoxia adrenal responses to the stress of Hoffmann, P., Jonsdottir, I.H. & Thoren, P. and high intensity exercise. Aviation, treadmill exercise. Physiologic (1996) Activation of different opioid Space, and Environmental Medicine 62(8), adaptations to physical training. systems by muscle activity and exercise. 754–758. New England Journal of Medicine News in Physiological Sciences 11, 223–228. Kraemer, W.J., Fry, A.C., Warren, B.J. et al. 316(21), 1309–1315. Höllt, V. (1993) Regulation of opioid (1992) Acute hormonal responses in elite Luger, A., Deuster, P.A., Gold, P.W., peptide gene expression. In: Handbook of junior weightlifters. International Journal Loriaux, D.L. & Chrousos, G.P. (1988) Experimental Pharmacology (Herz, A., of Sports Medicine 13(2), 103–109. Hormonal responses to the stress of Akil, H. & Simon, E.J., eds.). Springer, Kraemer, W.J., Dziados, J.E., Marchitelli, exercise. Advances in Experimental Berlin: 307–346. L.J. et al. (1993) Effects of different Medicine and Biology 245, 273–280. Inder, W.J., Hellemans, J., Swanney, M.P., heavy-resistance exercise protocols on McCubbin, J.A. (1993) Stress and Prickett, T.C. & Donald, R.A. (1998) plasma β-endorphin concentrations. endogenous opioids: behavioral and Prolonged exercise increases peripheral Journal of Applied Physiology 74(1), circulatory interactions. Biological plasma ACTH, CRH, and AVP in male 450–459. Psychology 35, 91–122. athletes. Journal of Applied Physiology Kraemer, W.J., Häkkinen, K., Newton, R.U. McGowan, R.W., Pierce, E.F., Eastman, N. 85(3), 835–841. et al. (1999) Effects of heavy-resistance et al. (1993) β-Endorphins and mood Jonsdottir, I.H., Hoffmann, P. & Thoren, P. training on hormonal response patterns states during resistance exercise. (1997) Physical exercise, endogenous in younger vs. older men. Journal of Perceptual and Motor Skills 76(2), 376–378. opioids and immune function. Acta Applied Physiology 87(3), 982–992. McLoughlin, L., Medback, S. & Grossman, Physiologica Scandinavica. Supplementum Kroboth, P.D., Salek, F.S., Pittenger, A.L., A.B. (1993) Circulating opioids in 640, 47–50. Fabian, T.J. & Frye, R.F. (1999) DHEA man. In: Handbook of Experimental Keizer, H.A. & Rogol, A.D. (1990) Physical and DHEA-S: a review. Journal of Clinical Pharmacology (Herz, A., Akil, H. & exercise and menstrual cycle alterations. Pharmacology 39(4), 327–348. Simon, E.J., eds.). Springer, Berlin: What are the mechanisms? Sports Lac, G., Marquet, P., Chassain, A.P. & 673–696. Medicine 10(4), 218–235. Galen, F.X. (1999) Dexamethasone in McMurray, R.G., Forsythe, W.A., Mar, Knudtzon, J. (1986) Effects of resting and exercising men. II. Effects M.H. & Hardy, C.J. (1987) Exercise proopiomelanocortin-derived peptides on adrenocortical hormones. Journal of intensity-related responses of β- on plasma levels of glucagon, insulin Applied Physiology 87(1), 183–188. endorphin and catecholamines. Medicine and glucose. Hormonal and Metabolic Langenfeld, M.E., Hart, L.S. & Kao, P.C. and Science in Sports and Exercise 19(6), Research 18, 579–583. (1987) Plasma β-endorphin responses to 570–574. Koziris, L.P., Kraemer, W.J., Gordon, S.E., 1-hour bicycling and running at 60% McMurray, R.G., Fafrowicz, J.F. & Berry, o Incledon, T. & Knuttgen, H.G. (2000) V 2max. Medicine and Science in Sports and M. (1988) The endorphin response of Effect of acute postexercise ethanol Exercise 19(2), 83–86. women to sleep loss and exercise. intoxication on the neuroendocrine Lehmann, M., Foster, C., Dickhuth, H.H. Aviation, Space, and Environmental response to resistance exercise. Journal of & Gastmann, U. (1998) Autonomic Medicine 59(2), 129–132. Applied Physiology 88(1), 165–172. imbalance hypothesis and overtraining Mahler, D.A., Cunningham, L.N., Skrinar, Kraemer, R.R., Blair, S., Kraemer, G.R. syndrome. Medicine and Science in Sports G.S., Kraemer, W.J. & Colice, G.L. (1989) & Castracane, V.D. (1989) Effects of and Exercise 30(7), 1140–1145. β-endorphin activity and hypercapnic treadmill running on plasma β- Letizia, C., Barilla, F., Cerci, S. et al. (1996) ventilatory responsiveness after endorphin, corticotropin, and cortisol β-Endorphin and propiomelanocortin- marathon running. Journal of Applied levels in male and female 10 K runners. correlates peptides response in Physiology 66(5), 2431–2436. 154 chapter 11

Marchant, B., Umachandran, V., regulation, integration, and adaptation. endorphin and corticotropin at rest Wilkinson, P. et al. (1994) Reexamination Physiological Reviews 80(3), 1055–1081. and during exercise. Gynecological of the role of endogenous opiates in Pedersen, B.K., Bruunsgaard, H., Klokker, Endocrinology 6(3), 163–166. silent myocardial ischemia. Journal of M. et al. (1997) Exercise-induced Rahkila, P., Hakala, E., Salminen, K. & the American College of Cardiology 23(3), immunomodulationapossible roles of Laatikainen, T. (1987) Response of 645–651. neuroendocrine and metabolic factors. plasma endorphins to running exercises Marquet, P., Lac, G., Chassain, A.P., International Journal of Sports Medicine 18 in male and female endurance athletes. Habrioux, G. & Galen, F.X. (1999) (suppl. 1), S2–S7. Medicine and Science in Sports and Exercise Dexamethasone in resting and Pestell, R.G., Hurley, D.M. & Vandongen, 19(5), 451–455. exercising men. I. Effects on R. (1989) Biochemical and hormonal Rahkila, P., Hakala, E., Alen, M., Salminen, bioenergetics, minerals, and related changes during a 1000 km K. & Laatikainen, T. (1988) β-Endorphin hormones. Journal of Applied Physiology ultramarathon. Clinical and Experimental and corticotropin release is dependent 87(1), 175–182. Pharmacology & Physiology 16(5), on a threshold intensity of running Meyer, W.R., Muoio, D. & Hackney, T.C. 353–361. exercise in male endurance athletes. Life (1999) Effect of sex steroids on β- Petraglia, F., Barletta, C., Facchinetti, F. Sciences 43(6), 551–558. endorphin levels at rest and during et al. (1988) Response of circulating Rivier, C. & Rivest, S. (1991) Effect of stress submaximal treadmill exercise in adrenocorticotropin, β-endorphin, on the activity of the hypothalamic– anovulatory and ovulatory runners. β-lipotropin and cortisol to athletic pituitary–gonadal axis: peripheral and Fertility and Sterility 71(6), 1085–1091. competition. Acta Endocrinologica 118(3), central mechanisms. Biology and el Migdadi, F., Bashir, N., Hasan, Z., al 332–336. Reproduction 45(4), 523–532. Hader, A.A. & Gharaibeh, M. (1996) Petraglia, F., Bacchi, M.A., Comitini, G. Ronkainen, H.R., Pakarinen, A.J. & Exercise at low altitude (Jordan valley) et al. (1990) Plasma β-endorphin and Kauppila, A.J. (1986) Adrenocortical causes changes in serum levels of β-lipotropin levels increase in well function of female endurance runners ACTH, insulin, cortisol and lactate. trained athletes after competition and and joggers. Medicine and Science in Endocrinological Research 22(4), non competitive exercise. Journal of Sports and Exercise 18(4), 385–389. 763–767. Endocrinological Investigation 13(1), Ronsen, O., Kjeldsen-Kragh, J., Haug, E., Miller, P.F., Light, K.C., Bragdon, E.E. et al. 19–23. Bahr, R. & Pedersen, B.K. (2002) (1993) β-Endorphin response to exercise Petrides, J.S., Deuster, P.A. & Mueller, G.P. Recovery time affects immunoendocrine and mental stress in patients with (1999) Lactic acid does not directly responses to a second bout of endurance ischemic heart disease, Journal of activate hypothalamic–pituitary exercise. American Journal of Physiology. Psychosomatic Research 37(5), 455–465. corticotroph function. Proceedings of Cell Physiology 283(6), C1612–C1620. Mooren, F., Völker, K., Pedersen, B.K., the Society for Experimental Biology and Salzet, M., Salzet-Raveillon, B., Schulz, A. & Teschemacher, H. (2005) Medicine 220(2), 100–105. Cocquerelle, C. et al. (1997) Inter- and intracellular signaling. In: Pierce, E.F., Eastman, N.W., Tripathi, H.L., Leech immunocytes contain Molecular and Cellular Exercise Physiology Olson, K.G. & Dewey, W.L. (1993a) proopiomelanocortin: nitric (Mooren, F.C. & Völker, K., eds). Human β-Endorphin response to endurance oxide mediates hemolymph Kinetics, Champaign, IL: 109. exercise: relationship to exercise proopiomelanocortin processing. Nagao, F., Suzui, M., Takeda, K., Yagita, H. dependence. Perceptual and Motor Skills Journal of Immunology (Baltimore, Md) & Okumura, K. (2000) Mobilization of 77(3) (Pt. 1), 767–770. 159(11), 5400–5411. NK cells by exercise: downmodulation Pierce, E.F., Eastman, N.W., Tripathi, H.T., Schulz, A., Harbach, H., Katz, N., Geiger, of adhesion molecules on NK cells by Olson, K.G. & Dewey, W.L. (1993b) L. & Teschemacher, H. (2000) β- catecholamines. American Journal of Plasma β-endorphin immunoreactivity: Endorphin immunoreactive material Physiology. Regulatory, Integrative and response to resistance exercise. Journal of and authentic β-endorphin in the Comparative Physiology 279(4), Sports Science 11(6), 499–52. plasma of males undergoing anaerobic R1251–R1256. Pierce, E.F., Eastman, N.W., McGowan, exercise on a rowing ergometer. Nieman, D.C. & Pedersen, B.K. (1999) R.W. et al. (1994) Resistance International Journal of Sports Medicine Exercise and immune function: recent exercise decreases β-endorphin 21(7), 513–517. development. Sports Medicine 27, 73–80. immunoreactivity. British Journal of Schwarz, L. & Kindermann, W. (1990) Oleshansky, M.A., Zoltick, J.M., Herman, Sports Medicine 28(3), 164–166. β-Endorphin, adrenocorticotropic R.H., Mougey, E.H. & Meyerhoff, J.L. Raastad, T., Bjoro, T. & Hallen, J. (2000) hormone, cortisol and catecholamines (1990) The influence of fitness on Hormonal responses to high- and during aerobic and anaerobic exercise. neuroendocrine responses to exhaustive moderate-intensity strength exercise. European Journal of Applied Physiology treadmill exercise. European Journal of European Journal of Applied Physiology 61(3–4), 165–171. Applied Physiology and Occupational 82(1–2), 121–128. Schwarz, L. & Kindermann, W. (1992) Physiology 59(6), 405–410. Raastad, T., Glomsheller, T., Bjoro, T. & Changes in β-endorphin levels in Owens, P.C. & Smith, R. (1987) Opioid Hallen, J. (2001) Changes in human response to aerobic and anaerobic peptides in blood and cerebrospinal skeletal muscle contractility and exercise. Sports Medicine 13(1), 25–36. fluid during acute stress. Bailliere’s hormone status during 2 weeks of heavy Sforzo, G.A. (1989). Opioids and exercise. Clinical Endocrinology and Metabolism 1, strength training. European Journal of An update. Sports Medicine 7(2), 415–437. Applied Physiology 84(1–2), 54–63. 109–124. Pedersen, B.K. & Hoffman-Goetz, L. (2000) Rahkila, P. & Laatikainen, T. (1992) Effect Shen, A., Chin, J., Fullerton, M., Jennings, Exercise and the immune system: of oral contraceptives on plasma β- G. & Dart, A. (1992) Increases in plasma proopiomelanocortin and exercise 155

β-endorphin concentrations during and mental function in sedentary and proliferation following exercise. exercise do not contribute to increases in endurance-trained elderly males. International Journal of Sports Medicine heart rate following autonomic blockade International Journal of Sports Medicine 22(6), 447–453. in man. British Journal of Clinical 20(3), 159–166. Vassilakopoulos, T., Zakynthinos, S. & Pharmacology 33(1), 89–92. Tabata, I., Ogita, F., Miyachi, M. & Roussos, C. (1999) Strenuous resistive Sheps, D.S., Koch, G., Bragdon, E.E. et al. Shibayama, H. (1991) Effect of low blood breathing induces proinflammatory (1988) The reproducibility of resting and glucose on plasma CRF, ACTH, and cytokines and stimulates the HPA axis in post exercise plasma β-endorphins. Life cortisol during prolonged physical humans. American Journal of Physiology Sciences 43(9), 787–791. exercise. Journal of Applied Physiology 277(4) (Pt. 2), R1013–R1019. Sheps, D.S., Ballenger, M.N., De Gent, G.E. 71(5), 1807–1812. Viru, A. & Tendzegolskis, Z. (1995) Plasma et al. (1995) Psychophysical responses to Tache, Y. & Rivier, C. (eds.) (1993) Annals endorphin species during dynamic a speech stressor: correlation of plasma of the New York Academy of Sciences, vol. exercise in humans. Clinical Physiology β-endorphin levels at rest and after 697. Corticotropin-Releasing Factor and 15(1), 73–79. psychological stress with thermally Cytokines: Role in the Stress Response. Walberg-Rankin, J., Franke, W.D. & measured pain threshold in patients New York Academy of Sciences, New Gwazdauskas, F.C. (1992) Response of with coronary artery disease. Journal of York. β-endorphin and estradiol to resistance the American College of Cardiology 25(7), Taylor, D.V., Boyajian, J.G., James, N. et al. exercise in females during energy 1499–1503. (1994) Acidosis stimulates β-endorphin balance and energy restriction. Sheps, D.S., Kaufmann, P.G., Sheffield, D. release during exercise. Journal of Applied International Journal of Sports Medicine et al. (2001) Sex differences in chest pain Physiology 77(4), 1913–1918. 13(7), 542–547. in patients with documented coronary Teschemacher, H. (2003) Weissman, C. (1990) The metabolic artery disease and exercise-induced Proopiomelanocortin. What role does response to stress: an overview and ischemia: Results from the PIMI study. this ACTH and endorphin precursor update. Anesthesiology 73(2), 308–327. American Heart Journal 142(5), 864–871. play in stress adaptation? Internistische Wiedemann, K. & Teschemacher, H. (1983) Sibinga, A. & Goldstein, A. (1988) Opioid Praxis 43, 605–615. β-Endorphin immunoreactive materials peptides and opioid receptors in cells of Teschemacher, H., Koch, G., Kramer, D. in human plasma determined by a the immune system. Annual Reviews of & Wiedemann, K. (1990a) β-Endorphin multiple radioimmunoassay system. Immunology 6, 219–249. immunoreactive material in the plasma: Life Sciences 33 (suppl. 1), 89–92. Singh, A., Petrides, J.S., Gold, P.W., what does it mean? In: Endorphins in Williams, J.M. & Getty, D. (1986) Effect of Chrousos, G.P. & Deuster, P.A. (1999) Reproduction and Stress (Distler, W. & levels of exercise on psychological mood Differential hypothalamic–pituitary– Beck, L., eds.). Springer, Berlin: 93–98. states, physical fitness, and plasma β- adrenal axis reactivity to psychological Teschemacher, H., Koch, G., Scheffler, H., endorphin. Perceptual and Motor Skills and physical stress. Journal of Clinical Hildebrand, A. & Brantl, V. (1990b) 63(3), 1099–1105. Endocrinology and Metabolism 84(6), Opioid peptides. Immunological Wittert, G.A., Livesey, J.H., Espiner, E.A. 1944–1948. significance? Annals of the New York & Donald, R.A. (1996) Adaptation of Steinberg, H. & Sykes, E.A. (1985) Academy of Sciences 594, 66–77. the hypothalamopituitary adrenal axis Introduction to symposium on Tharp, G.D. (1975) The role of to chronic exercise stress in humans. endorphins and behavioural processes; glucocorticoids in exercise. Medicine Medicine and Science in Sports and Exercise review of literature on endorphins and and Science in Sports and Exercise 7(1), 28(8), 1015–1019. exercise. Pharmacology, Biochemistry, and 6–11. Yanovski, J.A., Yanovski, S.Z., Boyle, A.J. Behavior 23(5), 857–862. Urhausen, A., Gabriel, H. & Kindermann, et al. (2000) Hypothalamic–pituitary– Strassman, R.J., Appenzeller, O., Lewy, W. (1995) Blood hormones as markers of adrenal axis activity during exercise A.J., Qualls, C.R. & Peake, G.T. (1989) training stress and overtraining. Sports in African American and Caucasian Increase in plasma melatonin, β- Medicine 20(4), 251–276. women. Journal of Clinical Endocrinology endorphin, and cortisol after a 28.5-mile Van-der-Pompe, G., Bernards, N.T., and Metabolism 85(8), 2660–2663. mountain race: relationship to Kavelaars, A. & Heijnen, C. (2001) An Young, E., Bronstein, D. & Akil, H. (1993) performance and lack of effect of exploratory study into the effect of Proopiomelanocortin biosynthesis, naltrexone. Journal of Clinical exhausting bicycle exercise on endocrine processing and secretion: functional Endocrinology and Metabolism 69(3), and immune responses in post- implications. In: Handbook of 540–545. menopausal women: relationships Experimental Pharmacology (Herz, A., Struder, H.K., Hollmann, W., Platen, P. between vigour and plasma cortisol Akil, H. & Simon, E.J., eds.). Springer, et al. (1999) Neuroendocrine system concentrations and lymphocyte Berlin: 393–421. Chapter 12

Introduction to the Insulin-Like Growth Factor Signaling System

CHARLES T. ROBERTS, JR

Introduction system, its various components and signaling mech- anisms, and its role in growth and development, The insulin-like growth factor (IGF) signaling sys- with an emphasis on human data. As other chapters tem plays a critical role in the growth and develop- in this volume will describe the effects of exercise ment of many tissues, and is an important mediator on specific components of the IGF system, such as of overall pre- and postnatal growth (reviewed in the IGF ligands and the IGFBPs, this chapter will Rosenfeld & Roberts 1999). The IGF system is also also discuss the potential effects of exercise on IGF implicated in pathophysiology, and plays a particu- signaling through the IGF receptors per se. larly important role in tumorigenesis. As described in more detail below, the IGF system is comprised Insulin-like growth factor system of the IGF ligands (IGF-I and IGF-II), a series of components cell-surface receptors that mediate the biological effects of the IGFs, and a family of insulin-like The insulin-like growth factor ligands growth factor binding protein (IGFBP) that modu- late the half-lives and bioavailability of the IGFs in igf-i and igf-ii structure the circulation and in extracellular fluids (Fig. 12.1). This review will provide an overview of the IGF The IGF-I and IGF-II ligands are encoded by large

6 IGF-I 1 2 4 IGF-II 5 A L 3 S Insulin A L S

Fig. 12.1 Schematic overview of the IGFBP-RsIGF-IR IR IRR IGF-II/ insulin-like growth factor (IGF) M6PR signaling system. 156 the igf signaling system 157

genes, which have been extensively characterized in The insulin-like growth factor receptors humans and rodents. The mature IGF-I peptide con- sists of B and A domains that are homologous to the The IGF-I and IGF-II ligands interact with an array B and A chains of insulin. Unlike the situation with of cell-surface receptors that may be present singly insulin, they are not proteolytically cleaved, but or in combination on target cells. It was initially remain linked in the mature peptide by a C domain thought that IGF-I primarily activated the type 1 analogous to the C peptide of insulin. Both IGF-I IGF or IGF-I receptor (IGF-IR), a transmembrane and IGF-II contain an additional short D domain tyrosine kinase structurally and functionally related that is not found in insulin. Additionally, the IGF-I to the insulin receptor (IR) (LeRoith et al. 1995; and IGF-II prohormones contain a C-terminal E Adams et al. 2000). IGF-II, on the other hand, was peptide that is cleaved in the Golgi apparatus known to interact with high affinity with the type during secretion. Alternative splicing of exons 5 and 2 IGF, or IGF-II receptor (IGF-IIR). Subsequent 6 of the human, rat and mouse IGF-I genes produces studies have shown that both IGF-I and IGF-II inter- alternative E peptides, but the physiological signific- act with the IGF-IR, albeit with a ~ threefold differ- ance of this prohormone diversity is unclear. The ence in affinity (IGF-I > IGF-II). The cloning of the presence of multiple leader exons in mammalian IGF-IIR cDNA revealed that it was identical to the IGF-I genes also results in variation in the signal previously characterized cation-independent man- peptide of the preprohormones, but the physiolo- nose-6-phosphate (M6P) receptor involved in endo- gical consequences of this are also unclear. The organ- cytosis and intracellular trafficking of M6P-tagged ization and splicing of the human and rodent IGF-II proteins. Although some early studies proposed genes is also complex, but involves non-coding an active role for the IGF-IIR in IGF-II signaling, exons and, therefore, does not affect the structure of based upon apparent sequence homology between the mature IGF-II molecule or its precursors. the cytoplasmic domain of the IGF-IIR and the intra- cellular loops of G-protein-coupled receptors, sub- igf-i and igf-ii expression sequent studies ruled out the ability of the short intracellular domain of the IGF-IIR to mediate sig- The expression of the IGF-II gene in rodents is nal transduction. The function of this molecule in widespread prenatally, but diminishes drastically IGF-II action is thought to reflect its ability to serve after birth, with the choroid plexus and the lepto- as a clearance receptor for the IGF-II, thereby influ- meninges being the persistent sites of synthesis in encing the extracellular concentration of IGF-II. adult animals. Murine expression of IGF-I, on the The biological effects of IGF-I result primarily other hand, is low prenatally and significantly from its activation of the IGF-IR. IGF-I does not increases during puberty, with hepatic production cross-react with the IR, except at pharmacological being a major contributor to overall IGF-I levels in doses, since the relative affinity of IGF-I for the IGF- the circulation. IGF-I is produced by numerous IR versus the IR differs by at least an order of magni- other adult organs, however, including kidney, tude. It was initially thought that IGF-II, like IGF-I, lung and bone, and exerts endocrine, paracrine and only bound appreciably to the IGF-IR as com- autocrine effects. This inverse pattern of IGF-I and pared to the IR. Studies in knockout mice lacking IGF-II expression in rats and mice initially led to the various combinations of the IGF system and the IR concept of IGF-II as a fetal growth factor and IGF-I suggested that IGF-II acted through the IR in early as an adult growth factor. This is not the situation in development, prior to detectable IGF-IR gene ex- humans, however, since both IGF-I and IGF-II are pression (Louvi et al. 1997). The molecular basis for produced throughout life by multiple tissues. In this phenomenon was revealed when it was dis- fact, the circulating levels of IGF-II are consistently covered that a splice variant of the IR displayed several-fold higher than those of IGF-I, which sup- high affinity for IGF-II. Specifically, the IR transcript ports that concept that there are potentially diver- is subject to alternative splicing of exon 11, which gent roles for the two IGFs in human physiology. encodes a 12-amino acid segment at the C terminus 158 chapter 12

of the extracellular β subunit. Previous studies had of intra-α subunit disulphide bonds in the Golgi shown that the IR isoform encoded by the mRNA apparatus of cells expressing both the IGF-IR and IR lacking the exon-11 sequence (IR-A) displayed a genes. While originally considered to represent a twofold higher affinity for insulin than the IR-B small proportion of the total number of IGF-IR and isoform specified by the exon 11-containing mRNA. IR in a given cell, some reports have suggested that It has been established that the IR-A isoform, in the formation of hybrids is preferred over the for- fact, functions as a high-affinity receptor for IGF-II mation of classical IGF-IR and IR heterotetramers. and produces predominantly proliferative effects as This could result from the preferential formation compared to the principally metabolic effects elicited of disulphide bonds between cysteine residues in by insulin stimulation of IR-B (Frasca et al. 1999). IGF-IR and IR α subunits themselves. Thus, in some Thus, IGF-I functions primarily by activating the circumstances, hybrid receptors may outnumber IGF-IR, while IGF-II can act through either the IGF- ‘pure’ IGF-IR or IR molecules at the cell surface. IR or through the A form of the IR. With respect to ligand binding, IGF-IR/IR hybrid receptors retain high affinity for IGF-I, but exhibit severely reduced affinity for insulin. It is thought Hybrid receptors and the insulin that this reflects the ability of IGF-I to efficiently receptor-related receptor bind to either IGF-IR α subunit, while tight insulin The scope of IGF signaling is made significantly binding requires its interaction with both of the β more complex by the existence of hybrid receptors subunits found in the IR. As a consequence, the ex- that result from the dimerization of IGF-IR and IR istence of significant number of hybrid receptors hemireceptors, each consisting of a single α and β may preferentially diminish cellular responsiveness subunit linked by disulphide bonds (Fig. 12.2). to insulin, but not IGF-I. This has been proposed as a These hybrid receptors are formed by the formation mechanism through which up-regulation of IGF-IR

IGF-IR IR-A IR-B IRR

Fig. 12.2 Insulin-like growth factor IGF-IR/IR-A IGF-IR/IR-B IR-A/IR-B IGF-IR/IRR IR-B/IRR (IGF) system receptors and hybrids. the igf signaling system 159

expression could result in insulin resistance in cells (ALS), in a 1:1:1 molar ratio. IGFBP-5 also forms expressing the IR. The situation with hybrid recep- ternary complexes with IGFs and ALS. While IGFBP- tors has been further complicated by the existence, 1 through -4 exhibit generally similar affinities for and the IGF-II-binding characteristics, of IR-A and IGF-I and IGF-II, IGFBP-5 and -6 bind IGF-II with IR-B. IR-A/IR-B hybrid receptors undoubtedly 10- and 100-fold greater affinities, respectively, than occur, since most cell express both splice variants. their binding to IGF-I. The IGFBPs do not bind The difficulty in distinguishing these variants experi- insulin. The IGFBPs control IGF action by increasing mentally has precluded an examination of the bind- the half-lives of circulating IGFs, by controlling their ing characteristics and signaling capabilities of this availability for receptor binding, and, in the case particular class of hybrid receptors. It has been of cell surface-associated IGFBPs, by potentially demonstrated, however, that IGF-IR/IR-A hybrids influencing their direct interaction with receptors. bind IGF-I, IGF-II and insulin, whereas IGF-IR/IR-B Each of the IGFBPs is subject to limited and poten- hybrids bind IGF-I with high affinity, IGF-II with tially regulated proteolysis by various proteases. low affinity, and do not bind to insulin (Pandini et al. Thus, ligand–receptor interactions in the IGF system 2002). Thus, the relative expression of the IGF-IR are subject to complex regulation as a result of IGFBP and IR genes and the degree of alternative splicing levels, expression profile, degree of cell–surface asso- of exon 11 of the IR gene governs the ability of a ciation and extent of proteolysis. given cell to respond to IGF-I, IGF-II and insulin. A series of studies performed over the last The insulin receptor-related receptor (IRR) is the several years has established the concept of IGF- third member of the IGF-IR/IR family and does not independent actions of some of the IGFBPs (Oh exhibit binding to IGFs or to insulin (Watt et al. 1998). IGFBP-3 and 5, in particular, have been 1993). Although still considered an orphan receptor, shown to exhibit effects on proliferation, migration it has been shown to form hybrids with the IR when and sensitivity to apoptosis that are independent of both entities are overexpressed in NIH-3T3 fibro- their effects on IGF signaling per se. Some of these blasts. The formal possibility exists, therefore, that ‘IGF-independent’ effects are still modulated by IGF-IR/IRR, IR-A/IRR, or IR-B/IRR hybrids may IGF binding to the responsible IGFBP, so that ‘IGF occur in the restricted set of tissues that express the receptor-independent action’ may be a more accur- IRR gene, and that the formation of such hybrids ate term for these novel functions. The cell-surface could, like the formation of IGF-IR/IR hybrids, or intracellular molecules that participate in those influence cellular IGF and insulin responsiveness. effects are poorly characterized, but IGFBP-3 and -5 In fact, a recent study that analyzed double and have been identified in the nuclei of cells following triple knockouts of the IR, IGF-IR and IRR genes exposure to exogenous recombinant protein. This demonstrated a role for the IRR in testis develop- aspect of IGFBP action, when clarified, will add an ment, presumably through its modulation of IR and important dimension to our understanding of the IGF-IR action through hybrid formation (Nef et al. IGF signaling system in general. 2003). IGF-I receptor and insulin receptor Insulin-like growth factor binding proteins signaling pathways The biological activities of the IGF ligands are The signaling pathways that mediate IGF action are, modulated by a family of high-affinity IGFBPs in large part, represented by those identified to date (IGFBP-1 through -6) that are found in the circula- for the IGF-IR (Fig. 12.3). Upon binding of IGF-I tion and in extracellular fluids (Jones & Clemmons or IGF-II to the extracellular α subunit (specifically, 1995). IGFBP-3 is the predominant IGFBP in serum, a binding region comprised of the internal cysteine- and most circulating IGF-I and IGF-II is not found in rich region and the adjacent C-terminal L2 domain a free form, but in a ternary complex with IGFBP-3 of the α subunit), a conformational change is induced and a third component, the acid-labile subunit in the trans-membrane β subunits, resulting in 160 chapter 12

Insulin/IGF-I Rs

Extracellular space

ras SHC Grb-2 sos p85 PDK2 Raf IRS-1,2(3,4) Gab1 p110 MEK Grb-2 PDK1 p85 sos ERK p110

PKB PKB

Ets AFX

Fig. 12.3 Insulin-like growth factor I receptor (IGF-IR) signal transduction pathways.

trans-autophosphorylation of the tyrosine kinase PI3 kinase pathway modulates differentiation and domain intrinsic to the cytoplasmic portion of the sensitivity to apoptosis. However, as is the case with β subunit. This process fully activates the receptor almost every aspect of the IGF system, nothing is tyrosine kinase, which autophosphorylates the addi- simple. In MCF-7 breast cancer cells, IGF-I-induced tional tyrosine residues in the juxtamembrane and mitogenesis requires the PI3 kinase pathway, but C-terminal domains that flank the tyrosine kinase not the MAP kinase pathway (Dufourny et al. 1997), domain. These residues, particularly tyrosine 950 in while in H19-7 neuronal cells, the PI3 kinase path- the juxtamembrane domain, then serve as docking way is also required for IGF-I-stimulated mitogen- sites for members of the insulin receptor substrate esis, and the MAP kinase cascade is necessary for (IRS) and the Shc adaptor protein families. Subse- IGF-I-induced differentiation (Morrione et al. 2000). quent phosphorylation of these proteins by the In terms of anti-apoptotic signaling by the IGF-IR, receptor tyrosine kinase allows IRS and Shc proteins protection of Rat-1 fibroblasts from ultraviolet-B- to engage factors such as Grb-2/SOS and the p85 induced apoptosis required PI3 kinase pathway regulatory subunit of PI3 kinase, leading to activa- activation, but not the MAP kinase pathway (Kulik tion of the MAP kinase and PI3 kinase cascades that et al. 1997), while IGF-IR-mediated protection of constitute the major signal transduction cascades PC12 cells from serum withdrawal-induced apop- emanating from the activated IGF-IR. Among the tosis involved both pathways acting in a synergistic ultimate targets of the MAP kinase and PI3 kinase fashion (Parrizas et al. 1997). These and other exam- cascades are members of the Ets and forkhead tran- ples suggest that the specific pathways involved in, scription factor families, which provides a mechan- and their relative contribution to, the effects of IGF- ism for IGF action at the cell surface to effect the IR signaling on growth, differentiation and apop- changes in gene expression that underlie the effects tosis are cell type-specific. of IGF signaling on cellular proliferation, differen- While IGF action can clearly be controlled by the tiation and apoptotic sensitivity. levels of extracellular ligand and the number (and There is a general notion that the MAP kinase types) of receptors at the cell surface, the relative pathway exerts effects on proliferation, and that the abundance of receptor targets may be an important the igf signaling system 161

factor in determining the effects of IGFs in a given expressed, and from human studies of populations target cell. For example, there are four members of such as Pygmies and rare individuals with muta- the IRS family, IRS-1 through 4, each of which has a tions affecting the IGF-IR and IGF-I genes. These similar, yet unique, structure. The presence of dif- findings are discussed below. ferent combinations of IRS proteins may result in differential responses to IGF-IR activation. In fact, Evidence from transgenic animals recent studies have suggested that IRS-3 and IRS-4 can inhibit processes activated through IRS-1 and prenatal growth IRS-2 (Tsuruzoe et al. 2001). The relative levels of Shc and IRS proteins may also be an important factor The role of IGF action in prenatal growth has been influencing IGF action, in that they have been inferred from the phenotypes of transgenic and shown to compete for binding to tyrosine 950 of the knockout mice in which the expression of the IGF-I, activated IGF-IR. IGF-II, IGF-IR and IGF-II/MGP receptor genes have The general characteristics of signaling and the been manipulated (DeChiara et al. 1991; Baker regulatory possibilities described above for the IGF- et al. 1993; Liu et al. 1993; Powell-Braxton et al. 1993). IR also apply to the IR (with IR-A and IGF-IR/IR IGF-I and IGF-II deficiency each results in a 40% hybrid receptors being relevant to our topic), but decrease in birthweight, with IGF-II knockout mice there are important differences between the IGF-IR also exhibiting placental growth retardation. Double and IR that may have important ramifications for knockouts exhibit an additive growth deficiency of differential actions of IGF-I and IGF-II. In the first 80%. IGF-I knockouts can exhibit general perinatal place, apart from the conserved tyrosine 950/960 in lethality, depending on the genetic background. the IGF-IR and the IR, the position and number IGF-IR knockout mice exhibit a 55% decrease in of tyrosine residues in the juxtamembrane and C- growth rate, which is less than that seen in the IGF- terminal domains that are subject to autophos- I/IGF-II knockouts, and invariably die of suffoca- phorylation differ between the IGF-IR and IR. In tion at birth due to inadequate development of the addition, the IGF-IR and the IR have been shown to musculature of the diaphragm. Additional deletion utilize different heterotrimeric G proteins as part of the IGF-I gene in IGF-IR knockout animals does of their signaling mechanisms (Dalle et al. 2001; not further decrease birth weight, suggesting that Kuemmerle & Murthy 2001), and other proteins IGF-I functions exclusively through the IGF-IR. In have been identified that interact specifically with contrast, IGF-II and IGF-IR double knockouts are the C terminus of the IGF-IR, but not with the IR. A more growth-retarded than single IGF-IR knock- final difference in signaling pathways engaged by outs, suggesting that IGF-II effects can be mediated the IGF-IR and the IR is the involvement of STAT-3 by another receptor during embryogenesis. Ana- (Zong et al. 1998, 2000; Prisco et al. 2001) and STAT-5 lysis of IGF-IR/IR knockouts (Louvi et al. 1997) (Okajima et al. 1998) in IGF-IR signaling. These dif- suggested that the IR was responsible for IGF-II ferences, in conjunction with the existence of differ- signaling not mediated by the IGF-IR. As mentioned ent classes of hybrid receptors, make IGF signaling above, it was subsequently found that alternative in cells that express the IR (or the IRR), in addition to splicing of exon 11 produces an IR isoform that the IGF-IR, extremely complicated. exhibits high affinity for IGF-II. An indirect role for the IGF-II/M6P receptor in prenatal growth (Lau The role of the insulin-like growth et al. 1994; Wang et al. 1994; Ludwig et al. 1997) factor system in growth and was inferred from the phenotype of IGF-II/M6P development receptor knockout mice, which exhibit modest fetal and placental overgrowth (25–40%). This pheno- The contributions of IGF action to growth and type has been interpreted as resulting from the development have been discerned from studies in excess IGF-II that is seen in the serum and tissues transgenic mice in which various components of the of these animals due to lack of the clearance function IGF signaling system have been ablated or over- of the IGF-II/M6P receptor. 162 chapter 12

postnatal growth 1995; Cortez et al. 1996). The molecular basis for this IGF-I resistance and, potentially, the growth IGF-I-deficient mice that do not die perinatally phenotype of the Efe population, appears to be a exhibit severely reduced postnatal growth, while defect in IGF-IR gene expression (Hattori et al. 1996). IGF-II-deficient mice, although smaller than normal Thus, decreased IGF action due to decreased IGF-IR at birth, have normal growth velocities postnatally. levels causes pre- and postnatal growth retardation These findings support the concept that IGF-I is the in humans. principal mediator of postnatal growth. The lack of a postnatal phenotype in IGF-II-deficient mice is human mutations affecting the igf-i not surprising in light of the normal shutoff of IGF-II and igf-ir genes expression in almost all mouse tissues postnatally. Global overexpression of IGF-I in transgenic mice One patient has been described who was homozy- produces generalized hyperplasia and organome- gous for a partial deletion of the IGF-I gene (Woods galy, resulting in adult animals that are 30% larger et al. 1996). This patient made no active IGF-I, exhib- than normal (Mathews et al. 1988). Conversely, post- ited severe pre- and postnatal growth retardation, natal overexpression of IGF-II does not result in and also presented with deafness, mental retarda- increased somatic growth (Rogler et al. 1994; Wolf tion and microcephaly, characteristics not found in et al. 1995). Again, the lack of a growth phenotype in patients with growth hormone deficiency or resist- IGF-II transgenics may reflect the lack of postnatal ance. This patient’s parents were heterozygous for IGF-II expression in postnatal rodents. the IGF-I gene mutation, had extremely low circu- To date, no convincing phenotype has been lating IGF-I levels and also exhibited short stature. observed in knockouts of any of the six IGFBPs, These findings provide additional support for a role including several double knockouts. This puzzling of IGF-I in both pre- and postnatal growth and lack of an effect may reflect the redundancy of func- development. tion among the six IGFBPs. A number of patients have been described that are hemizygous for the IGF-IR gene as the result of deletion of the distal arm of chromosome 15 Insulin-like growth factor system effects (Roback et al. 1991; Siebler et al. 1995) or ring chro- in humans mosome 15 syndrome (Tamura et al. 1993; Peoples The role of IGF action in human growth and devel- et al. 1995). All of these patients exhibited intrauter- opment has come from several lines of evidence, ine growth retardation and postnatal growth fail- including analysis of African Pygmies, a patient ure, as well as other developmental abnormalities. with mutational loss of the IGF-I gene and a series Although the growth-deficiency phenotype con- of patients with hemizygosity of the IGF-IR gene sistently manifested by these patients is consistent resulting from loss of the distal arm of chromo- with decreased IGF-IR levels, the fact that no direct some 15. demonstration of IGF resistance in cells derived from these patients has been reported makes this efe pygmies data supportive, but not definitive, evidence for a role of IGF-IR action in human pre- and postnatal Initial cross-sectional studies of Mbuti and Babinga growth and development. Pygmies concluded that the short stature of these populations was due to the lack of the pubertal Potential effects of exercise on insulin- growth spurt (Mann 1987). Subsequent longitudinal like growth factor signaling and action studies of Efe Pygmies demonstrated growth retar- dation at birth that increased in the first 5 years of The possible consequences of exercise on the function- life (Bailey 1990, 1991). More recently, it has been ing of the IGF signaling system and IGF-regulated shown that immortalized T- and B-cell lines from physiology can occur at many levels, including Efe Pygmies are IGF-I-resistant (Geffner et al. 1993, effects on local and circulating concentrations of IGF the igf signaling system 163

ligands and IGFBPs (as well as proteolysis of the study reported an effect of exercise on IR autophos- latter) or more subtle effects on IGF receptor expres- phorlyation (Youngren et al. 2001), but most studies sion and function at a cellular level. The relationship have described effects on more distal signaling com- between exercise and the IGFs and their binding ponents such as IRS-1 and IRS-2 (Chibalin et al. 2000; proteins are described in companion chapters, so Nagasaki et al. 2000) and the PI3 kinase pathway (Kim the following discussion will focus on exercise et al. 1998). These latter effects, however, would be and IGF signaling. There are, to date, no data on rather non-specific, as they affect signaling factors util- the relationship between exercise and expression or ized by a multitude of hormones and growth factors. the intrinsic activity of the IGF-IR that mediates the majority of IGF signaling. Conclusions An alternative possibility is that changes in IR expression or activity could modulate IGF signal- Despite decades of intensive investigation, there ing. This could be a direct effect if these changes remain basic aspects of the IGF system that are involved the IR-A isoform of the IR, since, as dis- poorly understood. Principal among these is the cussed above, this IR isoform is a functional recep- role of IGF-II in human physiology. Additionally, tor for IGF-II. Alteration of expression or activity of the effects of IGF-I and IGF-II in combination at a either IR isoform could also indirectly affect IGF-I or cellular level are entirely unknown, although most IGF-II signaling through the IGF-IR as a result of the human tissues are routinely exposed to a combina- formation of hybrid receptors. While the evidence tion of endocrine, paracrine and, often, autocrine to date supports an effect of exercise on general IGF-I and IGF-II. In the context of this volume, the insulin action and glucose metabolism in particular complexity of the IGF system provides a wide spec- (reviewed in Wojtaszewski et al. 2002; Zierath 2002), trum of targets through which the effects of exercise the extent of exercise-induced effects on IR levels on IGF-regulated physiology and pathophysiology or activity, particularly in humans, is unclear. One may be mediated.

References

Adams, T., Epa, V., Garrett, T. & Ward, C. African Efe Pygmies. Biochemical and like growth factor-I unresponsiveness in (2000) Structure and function of the type Molecular Medicine 58, 31–36. an Efe Pygmy. Biochemical and Biophysical 1 insulin-like growth factor receptor. Dalle, S., Ricketts, W., Imamura, T., Research Communications 193, 1216–1223. Cellular and Molecular Life Sciences: CMLS Vellonweider, P. & Olefsky, J.M. (2001) Geffner, M.E., Bersch, N., Bailey, R.C. & 57, 1050–1093. Insulin and insulin-like growth factor I Golde, D.W. (1995) Insulin-like growth Bailey, R.C. (1990) Growth of African receptors utilize different G protein factor-I resistance in immortalized T-cell Pygmies in early childhood. New signaling components. Journal of lines from African Efe Pygmies. Journal England Journal of Medicine 323, 1146. Biological Chemistry 276, 15 688–15 695. of Clinical Endocrinology and Metabolism Bailey, R.C. (1991) The comparative DeChiara, T.M., Robertson, E.J. & 80, 3732–3738. growth of Efe Pygmies and African Efstratiadia, A. (1991) Parental Hattori, Y., Vera, J.C., Rivas, C.I. et al. farmers from birth to age 5 years. Annals imprinting of the mouse insulin-like (1996) Decreased insulin-like growth of Human Biology 18, 113–120. growth factor II gene. Cell 64, 849–859. factor-I (IGF-I) receptor expression and Baker, J., Liu, J.P., Robertson, E.J. & Dufourny, B., Alblas, J., van Teeffelen, function in immortalized African pygmy Efstratiadia, A. (1993) Role of insulin- H.A.A.M. et al. (1997) Mitogenic T-cells. Journal of Clinical Endocrinology like growth factors in embryonic and signaling of insulin-like growth factor I and Metabolism 81, 2257–2263. postnatal growth. Cell 75, 73–82. in MCF-7 human breast cancer cells Jones, J.I. & Clemmons, D.R. (1995) Insulin- Chibalin, A.V., Yu, M., Ryder, J.W. et al. requires phosphatidylinositol 3-kinase like growth factors and their binding (2000) Exercise-induced changes in and is independent of mitogen-activated proteins: biological actions. Endocrine expression and activity of proteins protein kinase. Journal of Biological Reviews 16, 3–34. involved in insulin signal transduction Chemistry 272, 31 163–31 171. Kim, B., Cheng, H.-L., Margolis, B. & in skeletal muscle: differential effects on Frasca, F., Pandini, G., Scalia, P. et al. (1999) Feldman, E.F. (1998) Insulin receptor insulin-receptor substrates 1 and 2. Insulin receptor isoform A, a newly substrate 2 and Shc play different roles Proceedings of the National Academy of recognized, high-affinity insulin-like in insulin-like growth factor I signaling. Sciences of the United States of America 97, growth factor II receptor in fetal and Journal of Biological Chemistry 273, 38–43. cancer cells. Molecular and Cellular 34 543–34 550. Cortez, A.B., Van Dop, C., Bailey, R.C. Biology 19, 3278–3288. Kuemmerle, J.F. & Murthy, K.S. (2001) et al. (1996) IGF-I resistance in virus- Geffner, M.E., Bailey, R.C., Bersch, N., Coupling of the insulin-like growth

transformed B-lymphocytes from Vera, J.C. & Golde, D.W. (1993) Insulin- factor-I receptor tyrosine kinase to Gi2 in 164 chapter 12

human intestinal smooth muscle. Journal of breast cancer growth by IGF binding Tamura, T., Tohma, T., Ohta, T. et al. (1993) of Biological Chemistry 276, 7187–7194. proteins. Breast Cancer Research and Ring chromosome 15 involving deletion Kulik, G., Klippel, A. & Weber, M.J. Treatment 47, 283–293. of the insulin-like growth factor 1 (1997) Antiapoptotic signalling by the Okajima, Y., Matsumura, I., Nishiura, T. receptor gene in a patient with features insulin-like growth factor I receptor, et al. (1998) Insulin-like growth factor-I of Russell–Silver syndrome. Clinical phosphatidylinositol 3-kinase, and Akt. augments erythropoietin-induced Dysmorphology 2, 106–113. Molecular and Cellular Biology 17, proliferation through enhanced tyrosine Tsuruzoe, K., Emkey, R., Kriauciunas, 1595–1606. phosphorylation of STAT5. Journal of K.M., Ueki, K. & Kahn, C.R. (2001) Lau, M.M.H., Stewart, C.E.H., Liu, Z. et al. Biological Chemistry 273, 22 877–22 883. Insulin receptor substrate 3 (IRS-3) and (1994) Loss of the imprinted IGF2 cation- Pandini, G., Frasca, F., Mineo, R. et al. IRS-4 impair IRS-1- and IRS-2-mediated independent mannnose 6-phosphate (2002) Insulin/insulin-like growth signaling. Molecular and Cellular Biology receptor results in fetal overgrowth factor I hybrid receptors have different 21, 26–38. and perinatal lethality. Genes and biological characteristics depending on Wang, Z.-Q., Fung, M.R., Barlow, D.P. & Development 8, 2953–2963. the insulin receptor isoform involved. Wagner, E.F. (1994) Regulation of LeRoith, D., Werner, H., Beitner-Johnson, Journal of Biological Chemistry 277, embryonic growth and lysosomal D. & Roberts, C.T., Jr. (1995) Molecular 39 684–39 695. targeting by the imprinted Igf2/Mpr and cellular aspects of the insulin-like Parrizas, M., Saltiel, A.R. & LeRoith, D. gene. Nature 372, 464–467. growth factor I receptor. Endocrine (1997) Insulin-like growth factor-1 Watt, V.M., Shier, P., Chan, J., Petrisor, Reviews 16, 143–163. inhibits apoptosis using the B.A. & Mathi, S.K. (1993) IRR: a novel Liu, J., Baker, J., Perkins, A.S., Robertson, phosphatidylinositol 3’-kinase and member of the insulin receptor family. E.J. & Efstratiadias, A. (1993) Mice mitogen-activated protein kinase Advances in Experimental Medicine and carrying null mutations of the genes pathways. Journal of Biological Chemistry Biology 343, 125–132. encoding insulin-like growth factor I 272, 154–161. Wojtaszewski, J.F., Nielsen, J.N. & (IGF-1) and type 1 IGF receptor (IGF1R). Peoples, R., Milatorich, A. & Francke, U. Richter, E.A. (2002) Invited review: Cell 75, 59–72. (1995) Hemizygosity at the insulin- effect of acute exercise on insulin Louvi, A., Accili, D. & Efstratiadias, A. like growth factor I receptor (IGF1R) signaling and action in humans. Journal (1997) Growth-promoting interaction of locus and growth failure in the ring of Applied Physiology (Bethesda, Md) 93, IGF-II with the insulin receptor during chromosome 15 syndrome. Cytogenetics 384–392. mouse embryonic development. and Cell Genetics 70, 228–234. Wolf, E., Rapp, K., Blum, W.F., Kolb, H. Developmental Biology 189, 33–48. Powell-Braxton, L., Hollingshead, P., & Brem, G. (1995) Skeletal growth of Ludwig, T., Eggenschwiler, J., Fisher, P. Warburtin, C. et al. (1993) IGF-I is transgenic mice with elevated levels of et al. (1997) Mouse mutant embryos required for normal embryonic growth circulating insulin-like growth factor-II. overexpressing IGF-II exhibit in mice. Genes and Development 7, Growth Regulation 5, 177–183. phenotypic features of the Beckwith– 2609–2617. Woods, K.A., Camacho-Huebner, C., Wiedemann and Simpson–Golabi– Prisco, M., Peruzzi, F., Belletti, B. & Savage, P. & Clark, A.J.L. (1996) Behmel syndromes. Genes and Baserga, R. (2001) Regulation of Id gene Intrauterine growth retardation and Development 11, 3128–3142. expression by type I insulin-like growth postnatal growth failure associated with Mann, G.V. (1987) The riddle of Pygmy factor: roles of STAT3 and the tyrosine the insulin-like growth factor I gene. stature. New England Journal of Medicine 950 residue of the receptor. Molecular New England Journal of Medicine 335, 317, 709–710. and Cellular Biology 21, 5447–5458. 1363–1367. Mathews, L.S., Hammer, R.E., Behringer, Roback, E.W., Baraket, A.J., Dev, V.G. et al. Youngren, J.F., Keen, S., Kulp, J.L. et al. R.R. et al. (1988) Growth enhancement (1991) An infant with deletion of the (2001) Enhanced muscle insulin receptor of transgenic mice expressing human distal long arm of chromosome 15 autophosphorylation with short-term insulin-like growth factor I. (q26.1 -> qter) and loss of insulin-like aerobic exercise training. American Endocrinology 123, 2827–2833. growth factor I receptor gene. American Journal of Physiology 280, E528–E533. Morrione, A., Romano, G., Navarro, M. Journal of Medical Genetics 38, 74–79. Zierath, J.R. (2002) Invited review: exercise et al. (2000) Insulin-like growth factor I Rogler, C.E., Yang, D., Rosetti, L. et al. training-induced changes in insulin receptor signaling in differentiation of (1994) Altered body composition and signaling in skeletal muscle. Journal of neuronal H19-7 cells. Cancer Research 60, increased frequency of diverse Applied Physiology (Bethesda, Md) 93, 2263–2272. malignancies in insulin-like growth 773–781. Nagasaki, M., Nakai, N., Oshida, Y. et al. factor transgenic mice. Journal of Zong, C.S., Zeng, L., Jiang, Y., Sadowski, (2000) Exercise training prevents Biological Chemistry 269, 13 779–13 784. H. & Wang, L. (1998) Stat3 plays an maturation-induced decreases in Rosenfeld, R. & Roberts, C.T., Jr. (1999). The important role in oncogenic Ros- and insulin receptor substrate-1 and IGF System. Humana Press, Totowa, NJ. isulin-like growth factor I receptor- phosphatidylinositol 3-kinase in rat Siebler, T., Wlodzimierz, L., Terry, C.L. induced anchorage-independent skeletal muscle. Metabolism: Clinical and et al. (1995) Insulin-like growth factor I growth. Journal of Biological Chemistry Experimental 49, 954–959. receptor expression and function in 273, 28 065–28 072. Nef, S., Verma-Kurvari, S., Merenmies, J. fibroblasts from two patients with Zong, C.S., Chan, J., Levy, D.E. et al. (2000) et al. (2003) Testis determination requires deletion of the distal long arm of Mechanism of STAT3 activation by insulin receptor family function in mice. chromosome 15. Journal of Clinical insulin-like growth factor I receptor. Nature 426, 291–295. Endocrinology and Metabolism 80, Journal of Biological Chemistry 275, Oh, Y. (1998) IGF-independent regulation 3447–3457. 15 099–15 105. Chapter 13

Exercise, Training and the GH–IGF-I Axis

ALON ELIAKIM, DAN NEMET AND DAN M. COOPER

Introduction more thoroughly in recent years and, therefore, the GH–IGF-I response to exercise and training can Physical activity plays an important role in tissue assist competitive athletes and coaches in the evalu- anabolism, growth and development, yet little is ation of the training load. known about the mechanisms that link patterns of exercise with tissue anabolism. Considerable The GH–IGF-I axis anabolic stimulus arises even from the relatively modest physical activity of daily living (Leblanc The GH–IGF-I axis is composed of hormones, et al. 1990). Therefore, anabolic effects of exercise growth factors, binding proteins (BPs) and receptors training are not limited to individuals participating that regulate many essential life processes, including in competitive sports who are particularly focused growth and development, metabolic and reparative on improvement of muscle strength and endurance. processes and aging. Therefore, the understanding On the other hand, participation of young athletes of the axis must consider each individual com- in intense training, especially if associated with in- ponent and the interaction between them under adequate caloric intake, exposes the young athletes both physiologic and pathological conditions. The to several health risks and hazards, and may lead to axis starts in the central nervous system (CNS) a reduction in growth potential (Theintz et al. 1993). where several neurotransmitters (chatecholamines, The exercise-associated anabolic effects are age serotonin and cholinergic agents, etc.) stimulate and maturity dependent. It is remarkable that nat- the hypothalamus to synthesize growth hormone urally occurring levels of physical activity, energy releasing hormone (GHRH) and somatostatin (SS). expenditure and muscle strength exhibit some of GHRH stimulates the anterior pituitary to synthes- their most rapid increases during childhood and ize and secrete GH. In contrast, SS directly inhibits adolescence. The combination of rapid growth GH secretion. and development, high levels of physical activity GH is the major secretory product of the axis. One and spontaneous puberty-related increases in ana- of GH most important actions is the stimulation of bolic hormones (growth hormone [GH], insulin-like hepatic IGF-I synthesis. However, other effects of growth factor I [IGF-I] and sex steroids) suggest the GH on metabolism, body composition and tissue possibility of integrated mechanisms linking exer- differentiation are independent of IGF-I. GH exerts cise with a variety of anabolic responses. This chap- direct feedback effect on the two hypothalamic hor- ter will focus on the effect of physical activity and mones that control its secretion. Tissue GH effects exercise training on components of the GH–IGF-I result from the interaction between GH and the GH axis and on differences between systemic and local receptor. The GH receptor contains intracellular and (i.e. muscle) responses to exercise. Finally, the effect extracellular transmembrane domains. The extra- of exercise on the GH–IGF-I axis has been studied cellular domain is identical in structure to growth 165 166 chapter 13

hormone binding protein (GHBP) (Leung et al. tion. Both hormones stimulate increases in muscle 1987); therefore, a unique feature of this axis is mass and bone mineral density, and reduce fat that GH receptor number and activity can be deter- distribution. mined easily by measurements of circulating GHBP Several components of the axis are age de- levels. pendent. GHRH, GH, GHBP, IGF-I and IGFBP-3 IGF-I is part of the insulin-related peptides. IGF-I reach their peak circulating levels during puberty can act as a hormone, and these effects are GH (Mauras et al. 1987) and decrease with aging (Corpas dependent. However, the majority of IGF-I actions et al. 1993). These changes are partially sex-hormone occur primarily as a result of paracrine or autocrine dependent. Nutritional state has also a remarkable secretion and regulation, which are only partially influence on the GH–IGF-I axis. For example, fast- GH-dependent. IGF-I is responsible for most, but ing and malnutrition increase GH secretion, but not all, of the anabolic and growth promoting effects despite the elevated GH, IGF-I levels are reduced of GH. IGF-I stimulates SS secretion, and inhibits (Marimme et al. 1982), probably due to a decrease GH by a negative feedback mechanism (Berelowitz in GH receptors. In this chapter we will focus on et al. 1981), but it is not clear whether circulating or physical activity, another environmental regulator local central IGF-I is responsible to this feedback of the GH–IGF-I axis and its components. mechanism. The majority of circulating IGF-I is bound to The effect of a single exercise several IGFBPs. The most important circulating BP during adult life is IGFBP-3, which is synthesized growth hormone mainly in the liver, and is GH dependent. When bound to IGF-I, it complexes with an acid-labile sub- One should differentiate between the acute effects unit to form the circulating complex that carries of a single bout of exercise on the GH–IGF axis from most of the IGF-I in the serum. Some of the IGFBPs neuroendocrinological adaptations that occur in are GH dependent (e.g. IGFBP-3), but others, such physically active people or in response to long-term as IGFBP-1 and -2, are insulin dependent (being programs of endurance training. high when insulin levels are low). The interaction The GH response to exercise is dependent on between IGF-I and its BPs is even more complicated the duration and intensity of the exercise bout, the since some BPs stimulate (e.g. IGFBP-5), while fitness level of the exercising subject, the timing others inhibit (e.g. IGFBP-4) IGF-I anabolic effects of blood sampling, refractoriness of pituitary GH (Rajaram et al. 1997). secretion to exercise stimuli and other environmen- The effects of IGF-I result from its interaction tal factors. Therefore, standardized exercise proto- with two different receptors. Type I receptor has cols should be used to evaluate the GH response to tyrosine-kinase activity, and mediates most of the acute exercise. IGF-I effects. This receptor exhibit similarities to Several previous studies reported that the GH the insulin receptor and, therefore, may bind also response to exercise is greater in less fit subjects insulin which has known anabolic effects as well. (Buckler 1972). However, in those earlier studies, The type II receptor is identical to the mannose-6- subjects were asked to perform exercise tests at phosphate receptor, and binds IGF-II as well. the same absolute, rather than relative, power. As a Some hormones in the GH–IGF-I axis (i.e. GHRH, consequence, due to the great variability in fitness, SS and GH) have a pulsatile pattern of secretion, some subjects exercise below, while others exercise and it has been shown that the pulsatility of GH above, their lactic/anaerobic threshold (LAT). This secretion is significantly important for accelerated is important since several investigators (Felsing et al. growth rate (Clark et al. 1985). In contrast, IGF-I and 1992) have demonstrated that circulating GH levels IGFBPs level are relatively stable during the day. increased only in response to above, but not below In addition to the important effect on growth, GH LAT, and that exercise loads of 75–90% of maximal and IGF-I have a marked effect on body composi- aerobic power yielded a greater GH rise than milder exercise, training and the gh–igf-i axis 167 loads (Hartley et al. 1972; Sutton & Lazarus 1976). secretion (i.e. the subsequent GH response to exer- Therefore, results of studies in which the GH cise was attenuated), and suggested that exercise- response to exercise was tested at an absolute work induced elevation in free fatty acids or alterations in rate demonstrate simply that as individuals become parasympathetic–sympathetic tone might have been fitter, the stress associated with exercise at an abso- responsible (Casanueva et al. 1984, 1987; Klijman lute work rate diminishes. & Frohman 1991). Ronsen et al. (2001) showed a In contrast to the observation that the exercise recovery from pituitary refractoriness to GH secre- input should be sufficient to cause a sizeable meta- tion if a second bout of high intensity endurance bolic effect in order to stimulate GH secretion, we exercise was performed 3 h after the first session. previously demonstrated a small but significant GH Consistent with this report, integrated 1.5-h GH response to an exercise input that had no systemic concentrations were significantly greater if differ- effect on heart rate or circulating lactate levels (i.e. ences between the exercise bouts (30 min, 70% o unilateral wrist flexion) (Eliakim et al. 2000). These V 2max) were 3.5 h and not 1 h (Kanaley et al. 1997). data suggest that factors like perceived exertion and Environmental factors as well as some patholo- associated psychological stress may lead to activa- gical states may interfere with GH response to tion of the hypothalamic–pituitary axis and GH exercise. Administration of a high fat meal attenu- release even in exercises involving small muscle ated the magnitude of GH response to exercise groups. (Cappon et al. 1993), and this inhibition was corre- The duration of exercise for stimulation of GH lated with circulating levels of SS. High ambient secretion should be at least 10 min (Bar-Or 1983), temperature may in itself increase circulating GH since exercise of shorter duration (e.g. 5 min at above level, while low temperature attenuates GH release LAT [Felsing et al. 1992]) was not accompanied by (Buckler 1973). Obesity and polycystic ovarian syn- increases in circulating GH levels. Moreover, exer- drome (Wilkinson & Parkin 1974) are characterized cise-induced GH peak occurs 25–30 min after the by attenuated GH response to exercise. Exercise- start of the exercise, irrespective of the exercise associated GH release is reduced in amenorrheic duration (Schwarz et al. 1996; Eliakim et al. 1999), athletes, reflecting possibly decreased GHRH res- and occurs few minutes earlier in women (Wideman ponse to exercise compared to eumenorrheic athletes et al. 1999). Thus, when the exercise task is brief (e.g. (Waters et al. 2001). Finally, the age-related decline 10 min) a peak may be reached after its cessation, in GH secretion was associated also with reduced while when the exercise task is long (e.g. 45 min) the GH response to exercise (Zaccaria et al. 1999). peak may be reached while the individual is still The increase in circulating GH levels following exercising. Blood sampling, however, should be exercise may serve in the diagnosis of GH defici- timed to the exercise-induced GH peak. ency. Since GH is secreted in pulses, and during Pituitary refractoriness, a time in which the most of the day its levels are very low, a single ran- normal pituitary gland will not respond sufficiently dom blood sample can not differentiate between the to a stimulus for GH release could also influence the healthy and GH-deficient child. To overcome this, a GH response to exercise. We previously demon- number of provocative tests to stimulate pituitary strated that the GH response to exercise was inhib- GH release have been developed (Fraiser 1974). ited if a spontaneous, early morning, GH pulse Most of these provocative tests use pharmacological had occurred within 1 h prior to the exercise test agents (Cowell 1995) and present some risk (e.g. (Eliakim et al. 1999). This inhibition occurred prob- hypoglycemia) for the patient. Moreover, the inter- ably due to the phenomenon of GH autoinhibition pretation of a normal GH response to pharmacolo- (Pontiroli et al. 1991) in which the elevated circulat- gical stimuli can be questioned because it does not ing GH from the previous spontaneous pulse attenu- necessarily give information about physiological GH ated the pituitary’s response to exercise. Cappon secretion. These confounding factors have led a et al. (1994) previously demonstrated a refractory number of investigators to emphasize the role of period of at least 1 h following exercise induced GH physiological stimulation tests such as exercise, or 168 chapter 13

to focus on less variable circulating substances, GH (Marcus et al. 1990). In addition, earlier stud- like IGF-I and/or its BPs, in the diagnosis of GH- ies showed (Bang et al. 1990) that exercise led to deficiency in children (Rosenfeld et al. 1995). increases in IGF even in subjects with pituitary insufficiency. These studies suggest that the exer- insulin-like growth factors cise-associated increase in IGF-I is, in fact, not related to GH and must reflect rapid changes in It is now well established that acute exercise lead to IGF-I distribution in the circulation due to release an increase in circulating IGF-I levels. Interestingly, from marginal pools or changes in IGF-I removal. several studies demonstrated that the exercise- In addition, the transient nature of the increases induced IGF-I increase occurred following very suggests that hemodynamic or metabolic effects of short high intensity exercise (i.e. 90 s) (Kraemer et al. exercise might play a role. Exercise in humans 2000), occurred 10 min following the beginning of is accompanied by the rapid autotransfusion of endurance exercise (Bang et al. 1990; Schwarz et al. hemoconcentrated blood from the spleen into the 1996) and occurred in exercise of both below and cellular circulation (Flamm et al. 1990), by increased above the LAT (Schwarz et al. 1996). Exercise was blood flow to the exercising muscle and by loss of also associated with an increase in urinary IGF-I (De plasma water (Convertino et al. 1981). Each of these Palo et al. 2002). phenomenona might explain, in part, an increased The mechanism for the transient increase in cir- IGF concentration by changes in IGF flux and/or culating IGF-I in response to exercise is not read- volume of distribution. ily apparent. One possibility would be the classic Another possible source for the increase in circu- mechanism of increased hepatic IGF release due to lating IGF-I can be a release from the exercising exercise-induced secretion of GH. As noted earlier, muscle. To test this, we used a simple approach in GH increases significantly mainly in response to which subjects performed a unilateral repeated high intensity exercise, while IGF-I increases for flexion of the wrist against relatively high resist- both low and high intensity exercise. Moreover, ance, while during- and post-exercise blood samples circulating IGF-I reaches its peak before the GH peak were collected simultaneously from the basilic vein (i.e. 10 vs. 30 min) (Fig. 13.1), while increases in of both the exercising (representing local release) serum IGF-I, due to de novo IGF-I synthesis in the and resting arm (representing systemic response) liver and transport to the circulation, occurs sev- (Eliakim et al. 2000). We found a bilateral, simul- eral hours after the administration of endogenous taneous increase in IGF-I suggesting that the local

14 GH 340 12 IGF-I 320 10 )

) 300 –1

–1 8

6 280 GH (ng·mL

4 260 IGF-I (ng·mL Fig. 13.1 Typical growth hormone 2 240 (GH) and insulin-like growth factor I (IGF-I) response to intense 0 Exercise endurance-type exercise bout. Peak 220 IGF-I level occurs before the GH 0 peak emphasizing that the exercise- –10 0 10 20 30 40 50 60 70 induced increase in circulating IGF-I Time (min) is GH-independent. exercise, training and the gh–igf-i axis 169 exercising muscle was not the source for of the IGF-I have demonstrated that IGFBP-3 levels measured increase. by radioimmunoassay (RIA) increased with both Interestingly, Elias et al. (2000) showed a transient low and high intensity exercise, and were greater rise in circulating IGF-I immediately after exercise during the above-LAT protocols (Schwarz et al. 1996). test to exhaustion. IGF-I levels fell thereafter to In contrast to their RIA measurements, IGFBP-3 reach nadir level 60–90 min following the exercise, measured by Western ligand blotting (WLB) did and then returned gradually to baseline levels. not change. As a potential explanation for the dis- Consistent with this observation prolonged and crepancy between RIA and WLB data they sug- intense exercise sessions (1.5 h of intense soccer gested that the antibody used in the RIA recognizes practice in children [Scheet et al. 1999] or 1.5 h of both intact and fragment forms of IGFBP-3 while wrestling practice in adolescents [Nemet et al. 2002]) the WLB method measures only the intact form were associated with decreases in circulating IGF-I of IGFBP-3. They therefore measured the rate of levels. Interestingly, these sessions were associated IGFBP-3 proteolysis as a function of exercise and with increases in proinflammatory cytokines, and were intrigued to find that proteolysis did occur in the authors suggested that the increase in these the high intensity exercise, and that this increased proinflammatory cytokines mediated the decrease proteolysis was associated with the peak increase in in circulating IGF-I. IGF-I and IGF-II serum concentrations. It is import- The effect of resistance exercise on circulating ant, however, to note that increases in IGFBP-3 pro- IGF-I is inconsistent. Several studies found an teolytic activity occurred in this study only for high increase in circulating IGF-I and free IGF-I follow- intensity exercise, while increases in IGFBP-3, IGF-I ing strength exercise (Bermon et al. 1999). Moreover, and IGF-II occurred for both low and high intensity eccentric exercise was associated with increases in exercise. Moreover, other studies did not find sig- muscle IGF-I mRNA suggesting that IGF-I may nificant increase in IGF-I proteolytic activity follow- modulate tissue regeneration after mechanical dam- ing endurance exercise (Dall et al. 2001). age (Bamman et al. 2001). However, other studies The mechanism for the increased IGFBP-3 proteo- found no change in circulating IGF-I following lysis following high intensity exercise is not clearly heavy resistance exercise (Nindl et al. 2001), or even understood. It was suggested that changes in total reduced IGF-I levels the morning after high and and free-ionized serum calcium concentrations moderate intensity resistance workout (Raastad (Lamson et al. 1993) and marked serum and muscle et al. 2000). These conflicting results reflect probably acid-base changes (Martin & Baxter 1986) play a role differences in exercise protocols, fitness level of the in exercise-induced IGFBP-3 proteolysis. participants, timing of blood sampling, etc. Few studies demonstrated changes in other There have been far fewer investigations in IGFBPs following exercise, including increases in humans regarding the physiological responses of IGFBP-1 (Suikkary et al. 1989; Hopkins et al. 1994) IGF-II compared with IGF-I. Several studies (Bang and IGFBP-2 (Chadan et al. 1999; Nindl et al. 2001). et al. 1990; Schwartz et al. 1996) showed an acute, These BPs exist in the circulation in much smaller endurance exercise associated rise in IGF-II. Cir- quantities than IGFBP-3, and appear to play a lesser culating IGF-II may play an important role in bone role in circulating IGF bioavailability. However, it is growth and development (Mohan & Baylink 1991); clear that the exercise-induced effect on circulating however, the biological importance of the exercise- IGF-I is not only mediated by alteration of the induced increase in IGF-II has yet to be determined. amount of IGF-I but rather by the effect on its BPs. insulin-like growth factor The effect of exercise training on the binding proteins GH–IGF-I axis

Several studies found increases in IGFBP-3 follow- Several independent studies of healthy human beings ing endurance exercise (Schwarz et al. 1996; Chadan have demonstrated significant correlation between et al. 1999). Interestingly, Schwarz and coworkers physical fitness and circulating components of the 170 chapter 13

GH–IGF-I axis. Weltman et al. (1994) demonstrated 160 ) significant positive correlation between 24-h integ- –1 o 120 rated GH concentrations and peak V 2 in young healthy male, but not female, adults. They also 80 demonstrated inverse correlation between 24-h integrated GH concentrations and body fat in both 40 IGFBP-1 (ng·mL men and women. They suggested that hyper- 0 insulinemia associated with excess body fat and 0 20406080100 decreases levels of physical activity may reduce 800 GH release (Yamashita & Melmed 1986), and since ) men, in general, have more central fat than women –1 600 (Bouchard & Despres 1989), there is a greater influence of adiposity on GH secretion in men. They 400 further speculated that the higher estradiol levels in 200 female adults, which stimulate GH release (Ho et al. GHBP (ng·mL 1987), may oppose the inhibitory effect of insulin 0 0 20406080100 and therefore explain the gender-related differences ) of the correlation between fitness and GH con- –1

·kg 60 centration. Consistent with this hypothesis, we − 1 recently demonstrated in healthy prepubertal girls 40 a remarkable relationship between adiposity (deter- mined by the body mass index [BMI] percentile), (mL·min − 1 20

fitness and indirect indicators of GH responsiveness ·kg 2 O

(i.e. GHBP) and insulin sensitivity (i.e. IGFBP-1) · 0 V (Eliakim et al. 2001). It appears that above about 020406080 100 the 70 percentile of BMI for age, fitness, GH levels BMI (percentile) and insulin sensitivity all begin to decrease even Fig. 13.2 Relationship between age-adjusted body mass in healthy children (Fig. 13.2). Such information index (BMI) percentiles and insulin-like growth factor might prove to be clinically useful in identifying binding protein-1 (IGFBP-1) (top panel), growth hormone –1 children who would benefit from programs of phys- binding protein (GHBP) (middle panel), and Vo2peak·kg (bottom panel). The data suggest that there exists a ical activity. BMI percentile threshold (~ 70%) above which insulin The positive correlation between fitness and GH sensitivity, GH activity and fitness all begin to decrease in levels is consistent with animal experiments of otherwise healthy, prepubertal girls. Borer et al. (1986) who noted increased GH pulse amplitude in physically active, rapidly growing hamsters compared with sedentary controls. They late-pubertal girls (Eliakim et al. 1996, 2001). These suggested that the fit state was associated with cross-sectional data suggest that fitness in healthy, increased circulating endorphins and/or increased prepubertal and adolescent girls is associated with tissue sensitivity to endorphins. Endorphins inhibit anabolic adaptations of the GH–IGF-I system. SS and reduce the inhibition of pituitary GH secre- The significant correlation between fitness and tion, and as a consequence pituitary GH secretion is mean overnight GH levels resulted probably from increased. an increase in peak GH amplitude since only peak We previously described that both functional (i.e. amplitude (and not peak frequency or width) corre- o V 2max) and structural (i.e. thigh muscle volume lated with mean GH. determined by magnetic resonance images) indices The positive correlation between GHBP and fitness of fitness were correlated with mean overnight GH is unique. GHBP is the extracellular domain of the levels, GHBP and serum IGF-I levels in pre- and GH receptor (Rosenfeld 1994), and therefore reflect exercise, training and the gh–igf-i axis 171 tissue GH receptor capacity. Ligand-mediated suggesting that the BPs may play a role in the inter- receptor regulation appears to exist for GH and action between growth and exercise. With the GHBP in a number of situations. GHBP decreases in exception of IGFBP-3, the amount of IGFBPs in the acromegaly (Kratzsch et al. 1995) and during exogen- circulation is low, and the exact role of these IGFBPs ous recombinant human growth hormone (rhGH) in the circulation has yet to be determined. None- therapy (Legar et al. 1995), and is high in obesity theless, it is noteworthy that the cross-sectional despite low GH (Jorgensen et al. 1995). But ligand- relationships between fitness and IGFBPs were con- mediated receptor down-regulation does not ap- sistent with current understanding of the biological pear to operate during normal growth when both activity of IGF-I binding proteins based on tissue GH and GHBP increase in the prepubertal years studies. IGFBP-1 and -2, known to inhibit IGF-I (Baumman et al. 1989). Our data suggest that the function, were found to be inversely correlated with relatively trained or fit state in prepubertal and muscle mass (Eliakim et al. 1996, 2001). IGFBP-4, a adolescent girls is another example of simultaneous known inhibitor of the anabolic functions of IGF-I in increases in both GH and GHBP. The mechanism of bone tissue culture experiments (Mohan et al. 1995), these responses is not known, but suggests anabolic was found to be inversely correlated with muscle o adaptations of both the ligand and receptor. mass (Eliakim et al. 1996) and V 2max (Eliakim et al. Collectively, it seems that increasing levels of 1998a). In contrast, the IGF-I potentiating binding physical activity stimulate GH pulsatility and, as protein, IGFBP-5, was positively correlated with a consequence, circulating IGF-I. It is compelling muscle mass (Eliakim et al. 2001). Accordingly, these to speculate that the stimulation of the GH–IGF-I data suggest the possibility that the generally axis by exercise contributesaalong with genetic, increased IGF-I bioactivity in fitter subjects might nutritional and other environmental factorsato an not be related only to changes in circulating IGF-I increase in muscle mass and, ultimately, to improved but also to changes in IGFBPs. cardiorespiratory responses to exercise (such as Very few studies examined the effect of endur- o V 2peak). Our data suggest that this mechanism ance training on the GH–IGF-I axis longitudinally. operate in prepubertal and adolescent girls even as Smith, A.T. et al. (1987) studied a group of healthy spontaneous growth proceeds. young adult men for 10 days and found that o Significant positive correlation between V 2max increased physical activity exacerbated the well- and both circulating GH and IGF-I levels were also described reduction in IGF-I that accompanies found in healthy pre- and post-menopausal women caloric restriction. Reduced IGF-I associated with o (Kelley et al. 1990). Both V 2max and IGF-I concen- training has been observed in high school wrestlers trations decline with age. However, when the and in highly trained young female gymnasts influence of both age and fitness was analyzed using (Jahreis et al. 1991; Roemmich & Sinning 1997). In o multiple regression, V 2max remained the only inde- these studies the training program was accom- pendent predictor of circulating IGF-I. Therefore, it panied by loss of body mass providing clear evid- was concluded that the decrease in serum IGF-I was ence for a negative energy balance and catabolic probably the result of age related decline in physical state. However, a recent report demonstrated that activity and fitness, and was not related to aging while inadequate caloric intake and negative energy per se. Along with these observations, positive cor- balance is a major cause for the training-associated o relation between V 2max and circulating IGF-I were IGF-I decrease, IGF-I level may fall even when reported in young and old male adults (Poehlman & energy balance and weight stability are maintained Copeland 1990), and higher levels of IGF-I were (Nemet et al. 2004). found in trained middle-aged men (Manetta et al. We recently reported the effect of a brief (5 weeks) 2002). randomized, prospective endurance-type training Virtually all of the major IGFBPs (1–6), each of intervention on the GH–IGF-I axis in pre- and late- which is known to influence IGF-I bioactivity in dif- pubertal boys and girls. Based on the cross-sectional ferent ways, were also related to indexes of fitness, data, we hypothesized that training would lead 172 chapter 13

30 Trained expenditure in the training group, the exercise inter- vention was not accompanied by weight loss. 25 Control A potential mechanism for this seemingly un- 20 * expected systemic ‘catabolic-type’ adaptation was 15 the reduction in circulating GHBP in the trained 10 subjects. As noted earlier, lower GHBP circulating levels may reflect fewer tissue receptors and reduced 5 * * * tissue responsiveness to GH (Rosenfeld 1994). As 0 a consequence, since circulating IGF-I depend on –5 Prepubertal GH-induced hepatic production (Lowe 1991), IGF-I IGF-I (% change) girls –10 levels in the training group decreased. The exercise training effect on GHBP observed –15 Prepubertal Adolescent boys Adolescent boys here appears to be unique. In several pathological –20 girls states (malnutrition [Postel-Vinay et al. 1995], obes- –25 ity [Jorgensen et al. 1995]) the relationship between GH and GHBP suggest the well-described phe- Fig. 13.3 Effect of a brief endurance-type exercise training nomenon of ligand-induced receptor regulation. In on circulating insulin-like growth factor I (IGF-I) levels in studies (Eliakim et al. 1998b, 2001), training did lead prepubertal and adolescent boys and girls. Training was associated with a significant decrease or attenuated IGF-I to a reduction in GHBP, but without any change in response in all groups. mean GH or GH pulsatility. Thus, the possibility exists that lower GH receptor is the initiating mech- to increases in circulating GH and IGF-I levels. anism in the growth axis response to training. The Training was accompanied by about 15% higher mechanism for such a direct, training effect on GH total energy expenditure (by the doubly labeled receptors or GHBPs is not known. water technique), and resulted in significant increases Interestingly, training was associated with in- o in V 2max and thigh muscle volume in the trained creases in IGFBP-2, a known inhibitor of IGF-I ac- but not the control subjects (Eliakim et al. 1996, 1998, tion, and with reduction in IGFBP-5 in adolescent 2001; Scheet et al. 2002). In contrast to our hypo- women. IGFBP-5 is one of the binding proteins theses, there was a significant decrease in GHBP, that enhance some of the IGF-I mitogenic effects (i.e. IGF-I and IGFBP-3 in the trained prepubertal girls in bones [Mohan et al. 1995]). These observations (Eliakim et al. 2001); a significant decrease in IGF-I emphasize again that the exercise training induced and IGFBP-3, with a significant increase in IGFBP-2 decrease in IGF-I bioavailability is mediated not in the trained prepubertal boys (Scheet et al. 2002); only by acting on IGF-I itself but by altering its a significant decrease in IGF-I and IGFBP-5 in the binding proteins. trained late-pubertal girls (Eliakim et al. 1996); and Recently, Scheet et al. (2002) intriguingly suggested a significant decrease in GHBP and IGF-I, with a the hypothesis that proinflammatory cytokines significant increase in IGFBP-2 in the trained late- were involved in the training-induced decreases pubertal boys (Fig. 13.3) (Eliakim et al. 1998b). of IGF-I (Fig. 13.4) (Scheet et al. 2002). They demon- Interestingly, endurance training had no effect on strated that in brief 5-week endurance training in o GH pulsatility patterns in any of these groups. prepubertal boys increases in V 2max were pos- These effects are commonly observed in energy- itively correlated with changes in tumor necrosis deficient states like food deprivation or disease- factor-α (TNF-α). This suggested that children who associated malnutrition (Smith, W.J. et al. 1995; trained the hardest and had the biggest increase in Tonshoff et al. 1995), but ‘catabolic’ neuroendocrine fitness also had the largest increase in circulating adjustments occurred in the present studies even level of proinflammatory cytokines. In addition, though training increased thigh muscle volume. changes in IGFBP-3 were inversely correlated with Moreover, despite the significantly greater energy changes in TNF-α and interleukin-6 (IL-6), suggesting exercise, training and the gh–igf-i axis 173

80 ory response leads to lower level cytokines and to a lower steady state level of IL-1ra. Finally, and con- 60 * * sistent with the two phases hypothesis, longer peri- 40 ods of training (5 months [Koziris et al. 1999] and 1 year [Weltman et al. 1992]) were indeed associated 20 * with increases in circulating GH and IGF-I levels. 0 Very few studies have examined the effect of Change (%) resistance training on the GH–IGF-I axis. Raastad –20 Control et al. (2001) demonstrated that despite an increase IL-1 IL-6 TNF Trained –40 in muscle strength following 2 weeks of heavy strength training, IGF-I decreased in the 8th day and Fig. 13.4 The effect of 5-week aerobic-type exercise returned to baseline levels 4 days later. Kraemer intervention on proinflammatory cytokines (interleukin- et al. (1999) found no change in circulating IGF-I but 1β, interleukin-6 and tumor necrosis factor-α). The stress an increase in resting IGFBP-3 following 10 weeks of of training was associated with a significant increase of the resistance training program in young men. Twelve proinflammatory cytokines. IL, interleukin; TNF, tumor weeks of high-volume resistance training resulted necrosis factor. in type I and type II muscle fiber hypertrophy in college men, and there was a significant correla- that increase in the inflammatory response mediates tion between the muscle hypertrophy and training- the training-associated decrease of components of associated increase in GH level (McCall et al. 1999). the GH–IGF-I axis. Longer resistance training programs (25 weeks) These observations suggest the hypothesis that were associated with increase in serum IGF-I after a sudden imposition of a training program which the 13th week of training in young adults (Borst et al. is associated with substantial increase in energy 2001). Since increase in muscle strength occurred expenditure leads initially to an increase in pro- also mainly during the first 13 weeks, the authors inflammatory cytokines and, as a consequence, to suggested that IGF-I mediated, at least partially, the decreases in IGF-I levels. Further, if the training resistance training-associated improvement in mus- adaptation is successful, the proinflammatory cyto- cle strength. These results suggest that a biphasic kines fall, and with that decrease, the suppression change in circulating IGF-I occurs also in response of IGF-I diminishes, an anabolic ‘rebound’ in the to resistance training in young adults. In contrast, GH–IGF-I axis may occur, and IGF-I level exceed no changes in circulating IGF-I was found in old the pretraining level. Exactly how and when this adults (Kraemer et al. 1999; Häkkinen et al. 2001), switch takes place, and whether the initial catabo- suggesting age-dependent differences in the IGF-I lic-type stage is necessary for the ultimate anabolic axis response to strength training. adaptation, remains unknown. The training-induced increase in muscle mass, Adaptations to the training-induced increase in despite the circulating decrease in IGF-I level, sug- proinflammatory cytokines and decrease in circu- gest that the local tissue effect of exercise on growth lating IGF-I start early with a training-associated factors differ from systemic effects. Very few studies increase in IGF-I receptor binding capacity (Lee et al. have examined the effect of brief exercise or training 2000). This increase in IGF-I affinity probably reflects on skeletal muscle IGF-I levels. Yan et al. (1993) the phenomenon of ligand-mediated receptor up- demonstrated that an acute bout of eccentric exer- regulation. cise led to an increase in IGF-I immunoreactivity in Fitter subjects were found to have lower levels rat type II muscle 4 days post-exercise. Consistent of IL-1 receptor antagonist (IL-1ra). This agent is with Yan’s observation, we found that 5 days of stimulated by the inflammatory cytokines and act to treadmill training in young female rats resulted in block their biological activity at the receptor level. a significant increase in muscle size and muscle Therefore, training-associated reduced inflammat- IGF-I protein without changes in IGF-I mRNA and 174 chapter 13

circulating IGF-I (Eliakim et al. 1997). These results isoforms of IGF-I in the muscle (McKoy et al. 1999; suggest that the early mechanisms of the training Goldspink & Yang 2001). One of the isoforms is adaptation involve translational or post-translational detectable only after mechanical stimulation and increases in muscle tissue IGF-I, and that local therefore was named mechano growth factor (MGF). muscle IGF-I regulation may be dissociated from Its response to resistance training is higher in young central control mechanisms. Along these lines, adults, indicating age-related desensitivity to mech- Phillip et al. (1994) suggested that increases in IGF-I anical loading (Hameed et al. 2003). MGF is smaller in the kidney was not due to an increase in local and unglycosylated, and has a shorter half-life time synthesis but rather to an increase in IGF-I uptake and a different receptor binding affinity than the from the circulation by a non-membrane associated systemic liver IGF-I. The muscle has another iso- IGFBP-1. This observation emphasizes the possible form of IGF-I. This isoform is similar to the liver- role of local IGFBPs (which are strongly attached to type IGF-I, and is also up-regulated by exercise. the cell surface by a membrane integrin-binding Induction of MGF expression occurs mainly after domain) in the regulation of the early adaptation of stretch and electrical stimulation. MGF has an local IGF-I to exercise training. important role in local protein synthesis and in the Nevertheless, compensatory hypertrophy of the prevention of apoptosis and, therefore, plays a plantaris and soleus muscle after unilateral excision pivotal role in local tissue repair and remodeling. of the gastrocnemius tendon was accompanied by a This emphasizes again the significant importance of significant increase in IGF-I mRNA after only 2 days the local IGF-I response to exercise training. (levels of IGF-I protein were not measured in these What is the advantage to the organism of simultan- studies) (De Vol et al. 1990). This suggests that dif- eous central catabolism and local anabolism early in ferent types and amount of muscular work could the adaptation to increased physical activity? lead to a different time course and pattern of local We speculate that this adaptive mechanism might IGF-I increase. reduce global anabolic function, thereby conserving Longer periods of exercise training do, however, energy sources, but still allows for local tissue lead to stimulation of IGF-I gene expression. growth in response to environmental stresses like Zanconato et al. (1994) found an increase in hepatic exercise training. IGF-I gene expression following 4 weeks of endur- Consistent with this speculation is the phe- ance training in young rats. In addition, the authors nomenon of attenuated somatic growth and re- showed that the 4 weeks of endurance type training duced circulating IGF-I despite muscle adaptation led to increases in the exercising muscle IGF-I gene to intense exercise training in nutritionally self- expression and protein. Interestingly, inhibition of deprived young elite athletes (e.g. female gymnasts GH (by hypophysectomy [De Vol et al. 1990] or by [Theintz et al. 1993]). The dissociation between GH releasing hormone antibody [Zanconato et al. target tissue and central neuroendocrine GH–IGF-I 1994]) actually enhanced the local IGF-I response to responses was also demonstrated in other environ- increased muscular effort. It is clear from these mental conditions (Moromisato et al. 1996). Rats observations that inhibition of GH alone cannot exposed to hypoxia had a reduced growth rate block the autocrine and paracrine effects of IGF-I, and circulating IGF-I, but the relative size of their emphasizing the GH-independence of the ‘local’ heart and lung was increased along with local IGF-I IGF-I anabolic adaptations to physical activity. gene expression. This indicates that a local anabolic Singh et al. (1999) studied the effect of progressive adjustment to reduced oxygen carrying capacity resistance training in frail elderly humans. They had occurred but that the central response was found that training was associated with increased catabolic and overall growth was reduced. muscle strength and with muscle fiber remodeling, In summary, there are differences between the and that higher baseline muscle IGF-I predicted the local and systemic GH–IGF-I response to exercise increase in muscle strength. training (Fig. 13.5). The local and more important Recently, it was suggested that there are two skeletal muscle response is anabolic from very early exercise, training and the gh–igf-i axis 175

Kindermann 2000). For example, the change in the testosterone/cortisol ratio, as an indicator of Systemic response the anabolic–catabolic balance, has been used with Local response limited success to determine the physiological strain of training (Kuoppasalmi & Adlercreutz 1984; IGF-I Baseline levels Hoffman et al. 1997). The data presented here sug- gests that changes in circulating IGF-I may also serve as a possible marker of the training load, and that sustained reduced circulating IGF-I levels may indicate overtraining. This is particularly important Training duration in young competitive athletes, and particularly in Fig. 13.5 Differences between local and systemic sports that typically combine intensive training adaptations of insulin-like growth factor I (IGF-I) to with caloric restriction (aesthetic sports [e.g. gym- exercise training. While muscle IGF-I increases from very nastics] or weight-categories sports [e.g. wrestling]). early stages of training, there are at least two phases in Reduced circulating IGF-I level in these circum- the systemic IGF-I response. The first phase is an acute catabolic-type response characterized by decreases in stances indicates negative energy balance, which circulating IGF-I, but at some point anabolic rebound may lead to attenuated somatic growth (Jahreis et al. occurs and IGF-I increases. 1991; Theintz et al. 1993; Roemmich & Sinning 1997). Therefore circulating IGF-I levels can serve as a very stages, and result mainly from GH-independent important alerting sign, and the athletes should be autocrine and paracrine IGF-I release. The systemic aware that even when they reduce the training GH–IGF-I response to a training program has at intensity, they would not be always able to com- least two phases: the first is an acute catabolic-type pensate for the growth loss during long periods of response with a decrease in circulating IGF-I. At negative energy balance. some later point, depending probably also on the Measurements of IGF-I levels can also help the nutritional and energy balance of the individual, a athlete and coach in the preparation for competi- chronic anabolic adjustment of the GH–IGF-I axis tion. We recently determined the effect of 4 weeks occurs. Whether the initial catabolic phase is neces- of training on fitness, self-assessment physical con- sary for the later anabolic adaptation, and whether ditioning scores and circulating IGF-I in elite profes- this response is affected by the fitness level of the sional handball players during their preparation for individual is still unknown. the junior world championships (Eliakim et al. 2002). Training consisted of 2 weeks of intense train- ing followed by 2 weeks of relative tapering. Both Possible practical applications for the circulating IGF-I and physical conditioning scores coach and athlete decreased initially, and returned to baseline levels The effectiveness of physical training depends at the end of training. We found a significant posit- essentially on the intensity, volume, duration and ive correlation between the changes in circulating frequency of training, and on the individual ability IGF-I and the physical conditioning scores. to tolerate training. An imbalance between the train- Tapering down the training intensity prior to the ing load and the individual’s tolerance leads to competition is a well-known training methodology under or overtraining. Therefore, many efforts have to help the athlete to achieve his best performance. been made to find objective parameters to quantify The results of this study demonstrated that this the balance between training load and the athlete’s strategy is indeed associated with parallel changes tolerance, with limited success. The endocrine in both IGF-I (an objective measure) and in indi- system, by modulation of anabolic and catabolic vidual conditioning self-assessment (a subjective processes, plays a major role in the physiolo- measure). Therefore, both measures may assist gical adaptation to exercise training (Urhausen & coaches and athletes in their training preparations. 176 chapter 13

However, as noted earlier (e.g. Eliakim et al. 1996, 420 2001), despite the decrease in circulating IGF-I, 400 fitness improved. These results suggest that while 380 changes in circulating IGF-I are good markers of the 360 340 general condition and energy balance of the athlete, ) –1 320 they are not necessarily good predictors of the 300 athlete’s performance. 280 Recently, we measured IGF-I level throughout 260

the training season in elite judokas. IGF-I decreased IGF-I (ng·mL 240 significantly during periods of heavy training, but 220 returned to baseline levels during tapering down, 200 and reached a peak above baseline levels during the competition period (Fig. 13.6) (unpublished data). 0 Nemet et al. (2004) showed similar patterns for free IGF-I during the training season in competitive Heavy Taparingdown Preseason training training wrestlers. It is still unknown what the permitted Competition decrease of IGF-I during periods of heavy training Light–moderate is, or the optimal increase during periods of taper- Fig. 13.6 A typical example of insulin-like growth factor I ing down and reduced training intensity. However, (IGF-I) levels in elite judoka during the training season. we believe that the inability to increase circulating IGF-I level decreased significantly during heavy training, IGF-I levels before the target competition should and increased above baseline levels before the target be an alarming sign for both the athlete and his/her competition. coach that the athlete’s general condition is not optimal. should be paid by athletes and trainers to factors Finally, we demonstrated that local IGF-I changes that may inhibit the GH response to exercise and are more important than systemic changes in the attenuate the training effect (e.g. low exercise intens- muscle adaptation to exercise training. However, ity; short intervals between multiple daily exercise if indeed GH changes following exercise play a key sessions; exercise following high fat meal; continu- role in the anabolic effects of exercise, attention ous training despite menstrual irregularities, etc.).

References

Bamman, M.M., Shipp, J.R., Jiang, J. et al. hormone binding proteins in health and Borst, S.E., De Hoyos, D.V., Garzarella, L. (2001) Mechanical load increases muscle disease. Metabolism 38, 683–689. et al. (2001) Effects of resistance training IGF-I and androgen receptor mRNA Berelowitz, M., Szabo, M., Frohman, L.A. on insulin-like growth factor and IGF concentrations in humans. American et al. (1981) Somatomedin-C mediates binding proteins. Medicine and Science in Journal of Physiology 280, E383–E390. growth hormone negative feedback by Sports and Exercise 33, 648–653. Bang, P., Brandt, J., Degerblad, M. et al. effects on both the hypothalamus and Bouchard, C. & Despres, J.P. (1989) (1990) Exercise-induced changes in the pituitary. Science 212, 1281–1297. Variation in fat distribution with age insulin-like growth factors and their low Bermon, S., Ferrari, P., Bernard, P., Altare, and health implications. In: Physical molecular weight binding proteins in S. & Dolisi, C. (1999) Responses of total Activity and Aging (Spirduso, W.W. & healthy subjects and patients with and free insulin-like growth factor-I and Eckert, H.M., eds.). Human Kinetics, growth hormone deficiency. European insulin-like growth factor-I binding Champaign, IL: 78–106. Journal of Clinical Investigation 20, protein-3 after resistance exercise and Buckler, J.M.H. (1972) Exercise as a 285–292. training in elderly subjects. Acta screening test for growth hormone Bar-Or, O. (1983) Growth hormone Physiologica Scandinavica 165, 51–56. release. Acta Endocrinologica 69, 219–229. deficiencyausing exercise in the Borer, K.T., Nicolski, D.R. & Owens, V. Buckler, J.M.H. (1973) The relationship diagnosis. In: Pediatric Sports Medicine for (1986) Alteration of pulsatile growth between changes in plasma growth the Practitioner (Bar-Or, O., ed.). hormone secretion by growth inducing hormone levels and body temperature Springer-Verlag, New York: 182–191. exercise: involvement of endogenous occurring with exercise in man. Baumann, G., Shaw, M.A. & Amburn, K. opiates and somatostatin. Endocrinology Biomedicine 19, 193–197. (1989) Regulation of plasma growth 118, 844–850. Cappon, J.P., Ipp, E., Brasel, J.A. & Cooper, exercise, training and the gh–igf-i axis 177

D.M. (1993) Acute effects of high-fat and DeVol, D.L., Rotwein, P., Sadow, J.L., Felsing, N.E., Brasel, J.A. & Cooper, D.M. high-glucose meals on the growth Novakofski, J. & Bechtel, P.J. (1990) (1992) Effect of low and high intensity hormone response to exercise. Journal of Activation of insulin-like growth factor exercise on circulating growth hormone Clinical Endocrinology and Metabolism 76, gene expression during work-induced in men. Journal of Clinical Endocrinology 1418–1422. skeletal muscle growth. American Journal and Metabolism 75, 157–162. Cappon, J.P., Brasel, J.A., Mohan, S. & of Physiology 259, E89–E95. Flamm, S.D., Taki, J., Moore, R. et al. (1990) Cooper, D.M. (1994) Effect of brief Eliakim, A., Brasel, J.A., Mohan, S. et al. Redistribution of regional and organ exercise on circulating insulin-like (1996) Physical fitness, endurance blood volume and effect of cardiac growth factor-I. Journal of Applied training, and the GH–IGF-I system in function in relation to upright exercise Physiology 76, 2490–2496. adolescent females. Journal of Clinical intensity in healthy human subjects. Casanueva, F.F., Villanueva, L., Cabezas, J. Endocrinology and Metabolism 81, Circulation 81, 1550–1559. & Fernandez-Cruz, A. (1984) 3986–3992. Fraiser, S.D. (1974) A review of growth Cholinergic mediation of growth Eliakim, A., Moromisato, M.Y., hormone stimulation tests in children. hormone secretion elicited by arginine Moromisato, D.Y. et al. (1997) Increase Pediatrics 53, 929–937. clonidine and physical exercise in man. in muscle IGF-I protein but not IGF-I Goldspink, G. & Yang, S.Y. (2001) Effects Journal of Clinical Endocrinology and mRNA after 5 days of endurance of activity on growth factor expression. Metabolism 59, 526–532. training in young rats. American Journal International Journal of Sport and Exercise Casanueva, F.F., Villanueva, L., Dieguez, of Physiology 273, R1557–R1561. Metabolism 11 (suppl.), 21–27. C. et al. (1987) Free fatty acids block Eliakim, A., Brasel, J.A., Barstow, T.J., Häkkinen, K., Pakarinen, A., Kraemer, growth hormone (GH) releasing Mohan, S. & Cooper, D.M. (1998a) W.J. et al. (2001) Selective muscle hormone stimulated GH secretion in Fitness and the growth hormone– hypertrophy, changes in EMG and force, man directly at the pituitary. Journal of insulin-like growth factor-I axis in and serum hormones during strength Clinical Endocrinology and Metabolism 65, adolescent males. Medicine and Science training in older women. Journal of 634–642. in Sports and Exercise 30, 512–517. Applied Physiology 91, 569–580. Chadan, S.G., Dill, R.P., Vanderhoek, K. Eliakim, A., Brasel, J.A., Mohan, S., Wong, Hameed, M., Orrell, R.W., Cobbold, M., & Parkhouse, W.S. (1999) Influence of W.L.T. & Cooper, D.M. (1998b) Inceased Goldspink, G. & Harridge, S.D. (2003) physical activity on plasma insulin-like physical activity and the growth Expression of IGF-I splice variants in growth factor-I and insulin-like growth hormone–IGF-I axis in adolescent males. young and old human skeletal muscle factor-I binding proteins in healthy older American Journal of Physiology 275, after high resistance exercise. Journal of women. Mechanisms of Ageing and R308–R314. Physiology 547, 247–254. Development 109, 21–34. Eliakim, A., Brasel, J.A. & Cooper, D.M. Hartley, L.H., Mason, J.W., Hogan, R.P. Clark, R.G., Jansson, J.O., Isaksson, O. (1999) GH response to exercise: et al. (1972) Multiple hormonal & Robinson, I.C. (1985) Intravenous assessment of the pituitary refractory responses to graded exercise in relation growth hormone: growth responses period, and relationship with to physical training. Journal of Applied to patterned infusions in circulating components of the GH–IGF-I Physiology 33, 602–606. hypophysectomized rats. Journal of axis in adolescent females. Journal of Ho, K.Y., Evans, W.S. & Blizzard, R.M. Endocrinology 104, 53–61. Pediatric Endocrinology and Metabolism 12, (1987) Effects of sex and age on 24 h Convertino, V.A., Keil, L.C., Bernauer, 47–55. profile of growth hormone secretion E.M. & Greenleaf, J.E. (1981) Plasma Eliakim, A., Oh, Y. & Cooper, D.M. (2000) in man: importance of endogenous volume, osmolality, vasopressin and Effect of unilateral wrist flexion exercise edtradiol concentration. Journal of renin activity durin graded exercise in on circulating growth hormone, insulin- Clinical Endocrinology and Metabolism 64, man. Journal of Applied Physiology 50, like growth factor-I, and fibroblast 51–58. 123–128. growth factor-2. American Journal of Hoffman, J.R., Falk, B., Radom Isaac, S. Corpas, E., Harman, S.M. & Blackman, Physiology 279, R548–R553. et al. (1997) The effect of environmental M.R. (1993) Human growth hormone Eliakim, A., Scheet, T.P., Newcomb, R., temperature on testosterone and cortisol and human aging. Endocrine Reviews 14, Mohan, S. & Cooper, D.M. (2001) responses to high intensity, intermittent 20–39. Fitness, training and the growth exercise in humans. European Journal of Cowell, C.T. (1995) Short stature. In: hormone–insulin-like growth factor-I Applied Physiology 75, 83–87. Clinical Paediatric Endocrinology, 3rd edn. axis in prepubertal girls. Journal of Hopkins, N.J., Jakeman, P.M., Cwyfan (Brook C.G.D., ed.). Blackwell Science, Clinical Endocrinology and Metabolism 86, Hughes, S.C. & Holly, J.M.P. (1994) Oxford, UK: 136–172. 2797–2802. Changes in circulating insulin-like Dall, R., Lange, K.H., Kjaer, M. et al. (2001) Eliakim, A., Nemet, D., Bar-Sela, S., growth factor binding protein-1 No evidence of insulin-like growth Higer, Y. & Falk, B. (2002) Changes in during prolonged exercise: effect of factor binding protein-3 proteolysis circulating IGF-I and their correlation carbohydrate feeding. Journal of Clinical during a maximal exercise test in with self-assessment and fitness among Endocrinology and Metabolism 79, athletes. Journal of Clinical Endocrinology elite athletes. International Journal of 1887–1890. and Metabolism 86, 669–674. Sports Medicine 23, 600–603. Jahreis, G., Kauf, E., Frohner, G. & De Palo, E.F., Gatti, R., Lancerin, F. et al. Elias, A.N., Pandian, M.R., Wang, L. Schmidt, H.E. (1991) Influence of (2002) Urinary insulin-like growth et al. (2000) Leptin and IGF-I levels in intensive exercise on insulin-like growth factor-I in athletes before and after unconditioned male volunteers after factor-I, thyroid and steroid hormones in physical exercise, and in sedentary short term exercise. female gymnasts. Growth Regulation 1, subjects. Clinica Chimica Acta 322, 51–57. Psychoneuroendocrinology 25, 453–461. 95–99. 178 chapter 13

Jorgensen, J.O., Pedersen, S.B., Borglum, J. Lee, C., Eliakim, A., Brasel, J.A. & Cooper, Mohan, S., Nakao, Y., Honda, Y. et al. et al. (1995) Serum concentration of D.M. (2000) Effects of exercise training (1995) Studies on the mechanisms by insulin-like growth factors, IGF binding on red blood cell insulin-like growth which insulin-like growth factor (IGF) proteins 1 and 3, and growth hormone factor-I receptor binding in adolescent binding protein-4 (IGFBP-4) and IGFBP- binding proteins in obese women males. Journal of Pediatric Endocrinology 5 modulate IGF actions in bone cells. and the effect of growth hormone and Metabolism 13, 621–627. Journal of Biological Chemistry 270, administration: a double blind, placebo- Leung, D.W., Spencer, S.A., Cachaines, G. 20 424–20 431. controlled study. European Journal of et al. (1987) Growth hormone receptor Moromisato, D.Y., Moromisato, M.Y., Endocrinology 133, 65–70. and serum binding proteins: Zanconato, S. et al. (1996) Effect of Kanaley, J.A., Weltman, J.Y., Veldhuis, J.D. purification, cloning, and expression. hypoxia on lung, heart, and liver insulin- et al. (1997) Human growth hormone Nature 330, 537–540. like growth factor-I gene and receptor response to repeated bouts of aerobic Lowe, W.L., Jr. (1991) Biological actions expression in the newborn rat. Critical exercise. Journal of Applied Physiology 83, of the insulin-like growth factors. In: Care Medicine 24, 919–924. 1756–1761. Insulin-like Growth Factors: Molecular and Nemet, D., Oh, Y., Kim, H.S., Hill, M. & Kelijman, M. & Frohman, L.A. (1991) The Cellular Aspects (LeRoith, D., ed.). CRC, Cooper, D.M. (2002) Effect of intense role of the cholinergic pathway in Boca Raton, FL: 49–86. exercise on inflammatory cytokines and growth hormone feedback. Journal of Manetta, J., Brun, J.F., Maimoun, L. et al. growth mediators in adolescent boys. Clinical Endocrinology and Metabolism 72, (2002) Effect of training on the GH/IGF-I Pediatrics 110, 681–689. 1081–1087. axis during exercise in middle-aged Nemet, D., Connolly, P.H., Pontello- Kelley, P.J., Eisman, J.A., Stuart, M.C. et al. men: relationship to glucose Pescatello, A.M. et al. (2004) Negative (1990) Somatomedin-C, physical fitness, homeostasis. American Journal of energy balance plays a major role in and bone density. Journal of Clinical Physiology 283, E929–E936. the IGF-I response to exercise training. Endocrinology and Metabolism 70, Marcus, R.G., Buttterfield, G., Holloway, Journal of Applied Physiology 96, 276–282. 718–723. L. et al. (1990) Effects of short-term Nemet, D., Pontello, A.M., Rose-Gottron, Koziris, L.P., Hickson, R.C., Chatterton, administration of recombinant human C. & Cooper, D.M. (2004) Cytokines R.T. et al. (1999) Serum levels of total and growth hormone to elderly people. and growth factors during and after a free IGF-I and IGFBP-3 are increased Journal of Clinical Endocrinology and wrestling season in adolescent boys. and maintained in long-term training. Metabolism 70, 519–527. Medicine and Science in Sports and Exercise Journal of Applied Physiology 86, Marimme, T.J., Zapf, J. & Froesch, E.R. 36, 794–800. 1436–1442. (1982) Insulin-like growth factors in the Nindl, B.C., Kraemer, W.J., Marx, J.O. Kraemer, W.J., Häkkinen, K., Newton, R.U. fed and fasted states. Journal of Clinical et al. (2001) Overnight responses of the et al. (1999) Effects of heavy-resistance Endocrinology and Metabolism 55, circulating IGF-I system after acute training on hormonal response patterns 999–1002. heavy resistance exercise. Journal of in younger versus older men. Journal of Martin, J.L. & Baxter, R.C. (1986) Insulin- Applied Physiology 90, 1319–1326. Applied Physiology 87, 982–992. like growth factor binding protein from Phillip, M., Werner, H., Palese, T. et al. Kraemer, W.J., Harman, F.S., Vos, N.H. et human plasma. Journal of Biological (1994) Differential accumulation of al. (2000) Effects of exercise and alkalosis Chemistry 261, 8754–8760. insulin-like growth factor-I in kidneys on serum insulin-like growth factor-I Mauras, N., Blizzard, R.M., Link, K. et al. of pre- and postpubertal streptozotocin- and IGF binding protein-3. Canadian (1987) Augmentation of growth diabetic rats. Journal of Molecular Journal of Applied Physiology 25, 127–138. hormone secretion during puberty: Endocrinology 12, 215–224. Kratzsch, J., Blum, W.F., Ventz, M. et al. evidence for a pulse amplitude- Poehlman, E.T. & Copeland, K.C. (1990) (1995) Growth hormone binding protein modulated phenomenon. Journal of Influence of physical activity on insulin- related immunoreactivity in the serum Clinical Endocrinology and Metabolism 64, like growth factor-I in healthy younger of patients with acromegaly is regulated 596–601. and older men. Journal of Clinical inversely by growth hormone McCall, G.E., Byrnes, W.C., Fleck, S.J., Endocrinology and Metabolism 70, concentration. European Journal of Dickinson, A. & Kraemer, W.J. (1999) 1468–1473. Endocrinology 132, 306–312. Acute and chronic hormonal responses Pontiroli, A.E., Lanzi, R., Monti, L.D., Kuoppasalmi, K. & Adlercreutz, H. (1984) to resistance training designed to Sandoli, E. & Pozza, G. (1991) Growth Interaction between catabolic and promote muscle hypertrophy. Canadian hormone (GH) autofeedback on GH anabolic steroid hormones in muscular Journal of Applied Physiology 24, 96–107. response to GH-releasing hormone. exercise. In: Exercise Endocrinology McKoy, G., Ashley, W., Mander, J. et al. Role of free fatty acids and somatostatin. (Fotherby, K. & Pal, S.B., eds.). De (1999) Expression of insulin-like growth Journal of Clinical Endocrinology and Gruyter, Berlin: 65–98. factor-I splice variants and structural Metabolism 72, 492–495. Lamson, G., Giudice, L.C., Cohen, P. et al. genes in rabbit skeletal muscle induced Postel-Vinay, M.C., Saab, C. & Gourmelen, (1993) Proteolysis of IGFBP-3 may be a by stretch and stimulation. Journal of M. (1995) Nutritional status and growth common regulatory mechanism of IGF Physiology 516, 583–592. hormone binding protein. Hormone action in vivo. Growth Regulation 3, 91–95. Mohan, S. & Baylink, D.J. (1991) The Research 44, 177–181. Leblanc, A.D., Schneider, V.S., Evans, H.J., role of insulin-like growth factor-II Raastad, T., Bjoro, T. & Hallen, J. (2000) Engelbretson, D.A. & Krebs, J.M. (1990) in the coupling of bone formation to Hormonal responses to high and Bone mineral loss and recovery after 17 resorption. In: Modern Concepts of moderate intensity strength exercise. weeks of bed rest. Journal of Bone Mineral Insulin-like Growth Factors (Spencer, European Journal of Applied Physiology 82, Research 5, 843–850. E.M., ed.). Elsevier, New York: 169–184. 121–128. exercise, training and the gh–igf-i axis 179

Raastad, T., Glomsheller, T., Bjoro, T. & Singh, M.A., Ding, W., Manfredi, T.J. et al. Waters, D.L., Qualls, C.R., Dorin, R., Hallen, J. (2001) Changes in human (1999) Insulin-like growth factor-I in Veldhuis, J.D. & Baumgartner, R.N. skeletal muscle contractility and skeletal muscle after weight-lifting (2001) Increased pulsatility, process hormonal status during 2 weeks of exercise in frail elders. American Journal irregularity, and nocturnal trough heavy strength training. European Journal of Physiology 277, E135–E143. concentrations of growth hormone in of Applied Physiology 84, 54–63. Smith, A.T., Clemmons, D.R., Underwood, amenorrheic compared to eumenorrheic Rajaram, S., Baylink, D.J. & Mohan, S. L.E., Ben Ezra, V. & McMurray, R. athletes. Journal of Clinical Endocrinology (1997) Insulin-like growth factor binding (1987) The effect of exercise on plasma and Metabolism 86, 1013–1019. proteins in serum and other biological somatomedin-C/insulin-like growth Weltman, A., Weltman, J.Y., Schurrer, R. fluids: regulation and function. factor I concentrations. Metabolism 36, et al. (1992) Endurance training amplifies Endocrine Reviews 18, 801–831. 533–537. the pulsatile release of growth hormone: Roemmich, J.N. & Sinning, W.E. (1997) Smith, W.J., Underwood, L.E. & effects of training intensity. Journal of Weight loss and wrestling training: Clemmons, D.R. (1995) Effects of caloric Applied Physiology 72, 2188–2196. effects on growth-related hormones. and protein restriction on insulin-like Weltman, A., Weltman, J.Y., Hartman, Journal of Applied Physiology 82, growth factor-I (IGF-I) and IGF-binding M.L. et al. (1994) Relationship between 1760–1764. proteins in children and adults. Journal of age, percentage body fat, fitness, and 24- Ronsen, O., Haug, E., Pedersen, Clinical Endocrinology and Metabolism 80, hour growth hormone release in healthy B.K. & Bahr, R. (2001) Increased 443–449. young adults: effects of gender. Journal of neuroendocrine response to a repeated Suikkary, A.M., Sane, T., Seppala, M. et al. Clinical Endocrinology and Metabolism 78, bout of endurance exercise. Medicine (1989) Prolonged exercise increases 543–548. and Science in Sports and Exercise 33, serum insulin-like growth factor binding Wideman, L., Weltman, J.Y., Shah, N. et al. 568–575. protein concentrations. Journal of Clinical (1999) Effects of gender on exercise- Rosenfeld, R.G. (1994) Circulating growth Endocrinology and Metabolism 68, induced growth hormone release. Journal hormone binding proteins. Hormone 141–144. of Applied Physiology 87, 1154–1162. Research 42, 129–132. Sutton, J.R. & Lazarus, L. (1976) Growth Wilkinson, P.W. & Parkin, J.M. (1974) Rosenfeld, R.G., Albertsson-Wikland, K., hormone and exercise comparison of Growth hormone response to exercise Cassorla, F. et al. (1995) Diagnostic physiological and pharmacological in obese children. Lancet 2, 55. controversy: the diagnosis of childhood stimuli. Journal of Applied Physiology 41, Yamashita, S. & Melmed, S. (1986) Effects growth hormone deficiency revisited. 523–527. of insulin on rat anterior pituitary cells: Journal of Clinical Endocrinology and Theintz, G.E., Howald, H., Weiss, U. & inhibition of growth hormone secretion Metabolism 80, 1532–1540. Sizonenko, P.C. (1993) Evidence for a and mRNA levels. Diabetes 35, 440–447. Scheet, T.P., Mills, P.J., Ziegler, M.G., reduction of growth potential in Yan, Z., Biggs, R.B. & Booth, F.W. Stoppani, J. & Cooper, D.M. (1999) adolescent female gymnasts. Journal of (1993) Insulin-like growth factor Effect of exercise on cytokines and Pediatrics 122, 306–313. immunoreactivity increases in muscle growth mediators in prepubertal Tonshoff, B., Blum, W.F., Wingen, A.M. after acute eccentric contractions. Journal children. Pediatric Research 46, 429–434. & Mehls, O. (1995) Serum insulin-like of Applied Physiology 74, 410–414. Scheet, T.P., Nemet, D., Stoppani, J. et al. growth factors (IGFs) and IGF binding Zaccaria, M., Varnier, M., Piazza, P., (2002) The effect of endurance-type proteins 1, 2 and 3 in children with Noventa, D. & Ermolao, A. (1999) training on growth mediators and chronic renal failure: relationship to Blunted growth hormone response inflammatory cytokines in prepubertal height and glomerular filtration rate. to maximal exercise in middle-aged and early pubertal males. Pediatric The European Study Group for versus young subjects and no effect of Research 52, 491–497. Nutritional Treatment of Chronic Renal endurance training. Journal of Clinical Schwarz, A.J., Brasel, J.A., Hintz, R.L., Failure in Childhood. Journal of Clinical Endocrinology and Metabolism 84, Mohan, S. & Cooper, D.M. (1996) Acute Endocrinology and Metabolism 80, 2303–2307. effect of brief low and high intensity 2684–2691. Zanconato, S., Moromisato, D.Y., exercise on circulating IGF-I, II, and IGF- Urhausen, A. & Kindermann, W. (2000) Moromisato, M.Y. et al. (1994) Effects I binding protein-3 and its proteolysis in The endocrine system in overtraining. of training and growth hormone young healthy men. Journal of Clinical In: Sports Endocrinology (Warren, M.P. & suppression on insulin-like growth Endocrinology and Metabolism 81, Constantini, N.W., eds.). Humana Press, factor-I mRNA in young rats. Journal of 3492–3497. Totowa, NJ: 347–370. Applied Physiology 76, 2204–2209. Chapter 14

The Role of MGF and Other IGF-I Splice Variants in Muscle Maintenance and Hypertrophy

GEOFFREY GOLDSPINK, SHI YU YANG, MAHJABEEN HAMEED,

Introduction IGF-I, originally called ‘somatomedin’, was re- garded as a general growth factor produced by the Molecular biology methods have drastically changed liver under the influence of growth hormone (GH). the way we think about hormones. The classical Later, it became apparent that it is expressed by definition of a hormone was ‘a chemical substance most tissues and exists as different splice variants, produced in a specialized gland released into the each of which has a somewhat different action. blood stream and transported to different tissue to elicit a physiological purpose’. We are now aware Growth hormone–IGF-I axis that hormones, or factors, are produced by many different types of tissues, and that some have an The original somatomedin hypothesis originated in autocrine or a paracrine effect. This is exemplified the 1950s following early efforts to understand the by the insulin-like growth factor I (IGF-I) system, a regulation of somatic growth by pituitary-derived family of different forms of IGF-I, each of which has GH. It was suggested that this did not act directly a different biological effect. These are derived from on its target tissues to promote growth, but there the splicing of the IGF-I gene in response to different were intermediary substances involved (Daughaday signals. The human genome has now been sequenced & Reeder 1966). The term ‘somatomedin’ was later and comprises about 40 000 different genes. How- adopted (Daughaday et al. 1972) to reflect the growth ever, we know there are many more different pro- promoting actions of these substances, which were teins, so it is therefore evident that the same gene subsequently characterized and later called insulin- must be spliced differentially to generate this pheno- like growth factors (Rinderknecht & Humbel 1978; typic diversity. The IGF gene appears to have evolved Klapper et al. 1983). However, in 1985, Green et al. from a single insulin-like gene that is expressed in proposed the ‘dual effector hypothesis’ (Green et al. vertebrates. A similar insulin-like gene is present in 1985), which suggested that GH had direct effects the C. elegans, the nematode, and in the chordate on peripheral tissues that were not mediated by amphioxus, closely related to the common ancestor IGF-I and that GH stimulated local IGF-I produc- of the vertebrate. During vertebrate evolution how- tion. It is now clear that one of its main roles is stimu- ever, the gene has been duplicated to give rise to lating the release of IGF-I from the liver and that, in insulin, IGF-I and IGF-II genes. Experiments on C. addition, GH stimulates the formation of a ternary elegans have shown that the ancestral insulin-like IGF binding complex, including insulin-like growth gene prevents cell death and considerably extends factor binding protein 3 (IGFBP-3) and the acid- the life of the worm. The system is even more versatile labile subunit (ALS), which stabilizes IGF-I in the because the insulin-like growth genes can be spliced serum. GH is secreted from somatotroph cells located to produce different RNA transcripts and different within the anterior pituitary gland in a pulsatile polypeptides with different biological activity. manner, and also has itself other specific functional 180 igf-i splice variants in muscle growth 181

effects on local tissues. These include a protein still unclear. Systemic growth factors may be of rel- anabolic effect (increased DNA, RNA and protein atively minor importance in muscle hypertrophy. synthesis), stimulation of growth and calcification For example, in one study the overloaded muscles of cartilage, an increased mobilization of fats, and of hypophysectomized rats were still able to hyper- use of fats as an energy source. The exercise-induced trophy despite significantly reduced systemic IGF-I increase in GH is thought to mediate, either directly levels (Adams & Haddad 1996). These findings, or indirectly, many of the somatotrophic events sur- coupled with the simple observation that it is only rounding tissue remodeling (reviewed by Nindl challenged muscles that hypertrophy and not all the et al. 2003). muscles of the body, highlights the importance of a Unquestionably, the GH–IGF-I axis plays a role ‘local’ system of muscle adaptation. in postnatal growth and development, and these hormones reach their peak during adolescence. Expression of IGF-I splice variants However, with increasing age, a further decline in in muscle the circulating levels of GH and IGF-I occurs, such that older people can be regarded as partially GH Muscles can be stimulated to grow rapidly if deficient (Rudman et al. 1981). The relationship mechanically challenged; electrical stimulation of a between the age-related decline in the GH–IGF-I muscle held in a stretched position also promotes axis and loss of muscle mass and strength has been both muscle lengthening (by the addition of new extensively studied. In young GH-deficient adults, sarcomeres in series) and increases cross-sectional administration of recombinant human growth hor- area, by adding sarcomeres in parallel (Goldspink mone (rhGH) was shown to have positive effects on et al. 1992). Using this approach in combination with muscle mass and function (Cuneo et al. 1991). specific primers and RT-PCR (reverse transcription- Another study where GH-deficient adults were polymerase chain reaction) it was possible to detect treated with GH for an extended period of time con- two different RNA transcripts (Yang et al. 1996), cluded that there was not only increased muscle which subsequent cloning and sequencing identi- strength but also decreased body fat (Beshyah et al. fied as being derived from the IGF-I gene by altern- 1995). This led to the belief that older individuals ative splicing (Yang et al. 1996). The first, IGF-IEa with decreased levels of circulating GH and IGF-I was also detected in the resting muscle, and corres- would also benefit from rhGH therapy. However, ponded to the transcript commonly expressed in studies which have combined GH administration the liver. The second, not detected in resting muscle, and resistance training in both young (Yarasheski corresponded to IGF-IEb. et al. 1992) and older men (Yarasheski et al. 1995) The terminology of the IGF-I is a problem when have shown that the rates of protein synthesis are attempting to apply it to non-hepatic tissues and no greater when resistance training is combined therefore this splice variant in muscle was named with GH than when resistance training is performed mechano growth factor (MGF), due to its apparent alone. Furthermore, in older people the changes in regulation by mechanical signals (Yang et al. 1996); muscle mass and function have also been reported it has a different carboxy-peptide sequence to the to be similar between both groups (Lange et al. liver type of IGF-I. An additional problem is that 2002). It should be noted that studies involving the MGF would be classified as IGF-IEb in rat but IGF- use of rhGH, such as those listed above, are likely to IEc (as identified in hepatoma cells by Chew et al. have used the 22 kDa molecular isoform of GH. This 1995) in humans (Table 14.1). Also, in human muscle is the predominant form of GH found in the plasma an additional transcript has been detected (Hameed GH. The effects of the other molecular isoforms of et al. 2004), which confusingly has been termed GH, of which there are thought to be more than a IGF-IEb but which differs from the rat IGF-IEb. It is hundred (Baumann, 1991), are yet to be determined. therefore apparent that the muscle IGF-I isoforms The roles of circulating GH and IGF-I, particularly while related to the liver isoforms, need to be char- with regard to muscle adaptation in later life, are acterized separately (Table 14.1 and Fig. 14.1). 182 chapter 14

Table 14.1 The different names used to describe the short half-life unless bound to the binding protein insulin-like growth factor I (IGF-I) isoforms in the which may be intracellular. Therefore MGF can be literature. regarded as an autocine/paracrine or local growth IGF-I isoform Names used in the literature factor produced locally in response to mechanical stimuli and acts on those muscle fibers that produce IGF-IEa L.IGF-I*, m-IGF-I† it. Hence it is an important signaling molecule in the IGF-IEb local regulation of muscle growth. IGF-IEc IGF-IEb (rat)‡, MGF*

*Yang et al. (1996); †Musaro et al. (2001); ‡Rotwein et al. Systemic IGF-I is produced by (1986). active muscle MGF, mechano growth factor. As mentioned above, IGF-IEa is also expressed in Mechano growth factor skeletal muscle as well as in several other non- hepatic tissues. It has a similar sequence to the main As well as MGF being expressed in response to isoform produced by the liver and is therefore mechanical activity, it was noted that its E domain assumed to have a systemic action. However, had an insert that changes the open reading frame. muscle expresses at least two of the major binding Amino acids are encoded by nucleotide base triplets proteins of the systemic form of IGF-I and their and any insert that is not a multiple of three there- expression tends to be up-regulated as is that of IGF- fore changes the downstream sequence. There is a IEa, by exercise. As the IGF-IEa produced by muscle 52 base pair insert in the rat and a 49 base pair insert will bind to these binding proteins in the extracel- in humans. This has important functional conse- lular matrix as well as in the serum, it is expected quences, as the 3’RNA sequence codes for a differ- to have more effect on the muscles that produce it ent carboxy peptide sequence that is involved in than on other muscles; its action can be regarded as the recognition of the binding proteins. Also, in the autocrine and paracrine, as well as endocrine. case of MGF, the carboxy peptide (encoded in exons As well as having a different carboxy peptide 5 and 6) sequence acts as a different growth factor to sequence, the expression kinetics are different to the peptide that binds to the IGF-I receptor (encoded those of MGF. It was shown by Haddad and Adams in exons 3 and 4). The E-domain peptide alone has (2002), that in response to resistance-type exercise been shown to induce division of mononucleated in rats, the mRNA of MGF was expressed earlier myoblasts and thus activate the muscle satellite than IGF-IEa mRNA. In support of this, Hill and (stem) cells required for muscle hypertrophy and Goldspink (2003) showed that in rats that, following repair (Yang & Goldspink 2002). MGF is apparently muscle damage, MGF is produced as a pulse lasting not glycosylated and there is evidence that it has a a few days, whereas the expression of IGF-IEa

Fig. 14.1 Schematic representation of the human IGF-I gene comprising of EXONS 1 2 3 4 5 6 six exons. The three splice variants, IGF-IEa, IGF-IEb and IGF-IEc (also Human IGF-I gene known as mechano growth factor, MGF) expressed in muscle are shown IGF-IEc with the relevant exons. The black IGF-IEb box in exon 5 represents the first 49 base pairs (bp) (52 bp in the rat) IGF-IEa which gives rise to the alternatively spliced, mechano-sensitive MGF isoform. igf-i splice variants in muscle growth 183

increases as MGF declines but stays elevated for muscle, 48 h after an acute bout of eccentric, but not much longer. concentric contractions. Shortly (2.5 h) after a bout of eccentric cycling exercise, we again detected a significant increase in MGF, but not IGF-IEa. The IGF-I and its splice variants in human muscle eccentric exercise consisted of 60 min of reverse In a study of young military recruits, Hellsten et al. pedal cycling divided into six work intervals: 0– (1996) reported an increase in IGF-I immunoreact- 6 min at 50%, 6–12 min at 75%, 12–20 min at 100%, ivity in muscle after 7 days of strenuous exercise 20–25 min at 130%, 25–40 min at 100% and 40– (which included terrain marching and warfare exer- 60 min at 75% of the load corresponding to the pre- o cises). More, recently, in a study of elderly people, viously determined concentric V 2max (Hameed Singh et al. (1999) reported a 500% increase in IGF-I et al. 2003b). Thus, it is possible that Bamman et al. levels in the quadriceps muscles following a 12- (2001) were measuring increases in IGF-IEa and not week period of resistance training, as determined MGF at this latter time point. by immunohistochemistry. Resistance training con- Whilst it is clear that mechanical activity plays sisted of three sets of eight repetitions at 80% of the a pivotal role in regulating local IGF-I expression in most recently determined 1 repetition maximum (1- muscle, it was unclear as to whether there may be RM) for the hip and knee extensor muscles, 3 days further regulation, provided by other hormones, per week for 10 weeks. It is clear that this type of notably GH. Some insight into this possibility was exercise training, where the active muscles must gained from the results of a recent longitudinal overcome high loads, is the type of exercise that study where the relationship between exogenous results in muscle hypertrophy. However, these GH administration and strength training exercise studies failed to distinguish between the different was studied in older people. The subjects (age 74 ± IGF-I splice variants. Recently, the mRNA levels of 1 year) were assigned to either resistance training MGF and IGF-IEa were measured using real-time (consisting three different lower body exercises: quantitative PCR shortly (2.5 h) after a single bout leg press, seated knee extension and seated knee of high intensity knee extensor exercise. Subjects flexion, which were performed three times a week in this study performed 10 sets of six repetitions and consisted three to five sets of 8–12-RM per of the knee extensor muscles at 80% of their 1-RM. session) with placebo, resistance training combined (Hameed et al. 2003a). In young subjects it was with rGH administration alone. GH administration observed that MGF mRNA levels were significantly without training did not change MGF mRNA levels increased as a result of weightlifting exercise, but no when measured at 5 weeks (Fig. 14.2), but signific- such change was observed in older subjects. Fur- antly increased IGF-IEa levels (237%). In contrast, 5 thermore, at this short time point after exercise IGF- weeks of resistance training significantly increased IEa mRNA levels were unchanged in both groups. expression of MGF (163%) and to a lesser extent These observations were interesting in that they IGF-IEa (68%). However, when GH treatment was were in general agreement with animal experiments combined with exercise MGF levels were dramat- in which MGF levels were shown to increase before ically increased (456%). These data suggest that those of IGF-IEa, suggesting that the two isoforms exogenous GH administration causes an overall up- were differentially regulated. No relationship was regulation of transcription of the IGF-I gene prior to observed with muscle myosin heavy chain isoform splicing, which results in more of the primary tran- composition, but of note was the observation that script of IGF-I. In the absence of strength training the subject who showed the most dramatic increase exercise, splicing is towards the IGF-IEa isoform, in MGF was the subject whose muscles expressed but when combined with the mechanical loading the most MHC-IIX. Weightlifting exercise com- it splices towards the MGF isoform. Furthermore, prises both concentric and eccentric components in this study another splice variant, IGF-IEb was and in a recent study, Bamman et al. (2001) reported detected and cloned. The function of this third mus- a 62% increase in IGF-I mRNA concentration in the cle isoform is not yet known. 184 chapter 14

600 MGF * IGF-IEa 5 weeks 350 5 weeks 12 weeks 12 weeks † 500 * 300 * * 400 250 * 300 200 † 150 % change * % change 200 * * # 100 * * 100 50

0 0 (a)RT GH only RT + GH (b)RT GH only RT + GH

Fig. 14.2 Changes in the expression of mechano growth factor (MGF) (a) and IGF-IEa (b) mRNA after 5 (open bars) and 12 (lined bars) weeks in elderly men in the three intervention groups: Growth hormone only (GH), resistance training only (RT) and resistance training in combination with growth hormone (RT + GH). Values are expressed as a percentage change from baseline at 5 and 12 weeks. Bars and error bars represent mean values and SEM, respectively. *, Significant difference from baseline (#) from 5 weeks (P < 0.05). †, Significant difference in % change between isoforms (P < 0.05). (Data taken from Hameed et al. 2004.)

The structure of IGF-I Phe (insulin notation) is conserved as well. The most obvious differences between IGF-I and insulin are in IGF-I peptides exist as single chain polypeptides the C domain. The extra carboxy terminal regions consisting of about 70 amino acid residues. Not only endow the IGF-I splice variant with special proper- are these derived by alternative splicing of the IGF-I ties which determine their modus operandi. As well gene but, like insulin, the initial peptide undergoes as a hydrophobic core similar to that of proinsulin, a process of post-translational alteration. The prim- IGF-I has three S–S bridges which determine the ary structure of IGF-I is similar to that of proinsulin three dimensional conformation of this polypep- (43% sequence homology) as it comprises an amino tide. However the presence of the disulphide links terminal B region and an A region that is separated make it difficult to synthesise IGF-I chemically and by a short connecting C domain. Unlike proinsulin, native function structure and stability require the however, it also has a D region extension peptide presence of all three S–S bridges (Narhi et al. 1993). and an E peptide at its carboxyl terminus (Lowe, 1988), and is therefore longer than insulin. Receptors mediate the cellular The tertiary structure of IGF-I (Fig. 14.3) was effects of IGF-I initially predicted using computer graphics. This was based on the three-dimensional crystalline The biological activity of any hormone depends on structure of insulin as determined by X-ray diffrac- the ability of the target cells to respond to the signal tion. As mentioned, IGF-I is somewhat longer than in the extracellular milieu. This is a function of the insulin although its receptor domain is very similar cell receptors as well as post-receptor mechanisms. and acts as a marker for interactive molecular IGF-I and IGF-II molecules interact with an array graphics for visualizing the conformation of IGF-I of cell surface receptors that may be present indi- and IGF-II peptides. In predicting the tertiary struc- vidually or in various combinations on target cells. ture of IGF-I, the conservation of the cysteine and Both IGF-I and IGF-II are believed to interact with glycine residues between IGF-I and proinsulin are the IGF-I receptor (IGF-IR), that is structurally and particularly important. The hydrophobic core of functionally related to the insulin receptor (IR) with insulin: A2 Ile, A16 Leu, B12 Val, B15 Leu and B24 which it shares > 50% amino acid identity. Despite igf-i splice variants in muscle growth 185

6 7 5 8 9 4 10 8 9 A C 11 4 28 12 3 29 5 1 10 3 7 6 Fig. 14.3 Tertiary structure of 27 S-S S-S 4 insulin-like growth factor I (IGF-I) 2 1 2 11 3 6 2 B showing the A, B, C and D domains. 5 Amino acid residues are numbered 26 7 12 1 within each domain. The amino B 15 and A domains are separated by a 10 short interconnecting C domain. The 25 8 9 16 13 14 18 19 D domain and E domain (not shown) 11 form the carboxyl terminus of the 24 17 7 12 13 protein. As well as the hydrophobic 8 core similar to that of proinsulin the 23 20 14 17 molecule has three disulfide bridges 6 21 5 15 S-S (shown as S–S), which determine its 18 16 three-dimensional conformation. 22 D 2 1 Two of these are located in the core 4 3 21 19 of the molecule (B18–A20; A6–A11) B and the third on the surface (B6–A7). (Adapted from Blundell et al. 1983.) 20 this similarity, IGF-I does not tend to bind with the homoreceptor molecules at the cell surface (reviewed IR, except at pharmacological doses. This is largely by Le Roith & Roberts 2003). These IGF-IR/IR hybrid due to its affinity being approximately two orders of receptors bind IGF-I with high affinity, but have a magnitude higher for the IGF-IR compared to the reduced affinity for insulin. This can be attributed IR. As well as binding to the IGF-IR, IGF-II can also to the ability IGF-I to bind to either IGF-IR α sub- bind to a second receptor known as the IGF-IIR with unit, whereas for insulin to bind effectively requires high affinity. interaction with both the β subunits found in the IR. The IGF-IR is a tyrosine specific protein kinase It is interesting to note that among known muscle with an extracellular ligand-binding site. This growth factors, IGF-I is unique in its ability to stimu- receptor seems to mediate most actions of IGFs in late both proliferation and terminal differentiation skeletal muscle. For example, the IGF-IR appears to of post-mitotic myotubes. This may be due to dif- mediate amino acid and hexose uptake in rat soleus ferent IGF-I splice variants within muscle tissue. A muscles (Yu & Czech 1984), BC3H1 muscle cells recent study showed that different IGF-I splice vari- (De Vroede et al. 1984) and DNA synthesis in chick ants have different roles in myoblast proliferation muscle satellite cells (Duclos et al. 1991). The IGF-I and differentiation (Yang & Goldspink 2002). It also receptor is therefore believed to mediate several showed that different actions of these different actions of IGF-I, such as stimulation of amino acid splice variants are mediated through a different uptake, proliferation, differentiation and inhibition receptor. of protein degradation (Ewton et al. 1987). The complexity of IGF-I signaling is increased by IGF-I binding proteins the formation of hybrid receptors that result from the dimerization of IGF-IR and IR hemireceptors. The roles of the specific binding proteins in deter- Each hybrid receptor consists of a single α and β mining the autocrine or paracrine actions of the IGF subunit linked by disulphide bonds. In some cir- system are becoming increasingly apparent (Florini cumstances, these hybrid receptors may outnumber et al. 1996; Damon et al. 1997). Amongst the seven 186 chapter 14

IGFBPs described so far, four (IGFBP-2, 4, 5 and 6) numerous in vitro studies, where it has been shown are produced by different myoblast cell lines that IGF-I acts to increase the diameter of myotubes, whereas only IGFBP-4, 5 and 6 are expressed by suppress protein degradation, increase amino acid adult skeletal muscle (Florini et al. 1996; Putzer et al. uptake and stimulate protein synthesis (Ewton et al. 1998). IGFBPs have been associated with the sys- 1987; Vandenburgh et al. 1991; Florini et al. 1996; temic IGF system, but their expression in skeletal Semsarian et al. 1999; Bodine et al. 2001; Rommel muscle would have the effect of retaining IGF-I et al. 2001). Its expression during muscle hyper- within the muscle tissue. In the unbound state the trophy has been shown by using several animal IGF-I peptides have a short half-life and the IGFBPs models, including stretch-induced hypertrophy of were previously thought of as carrier proteins in the the muscle. For example, Schlechter et al. (1986) and serum. However, when expressed within the mus- Czerwinski et al. (1994) reported an increased cle they modulate the endocrine as well as the local expression of muscle IGF-I mRNA, and DeVol et al. influences of IGF-I (Mohan et al. 1996). The precise (1990) demonstrated that there was a threefold actions of IGFBPs at the tissue level are unknown, increase in IGF-I mRNA levels in the soleus and but it is apparent that they stabilize and augment plantaris muscles in 11–12-week-old female rats fol- local IGF-I bioavailability (Jones & Clemmons 1995; lowing tenotomy-induced hypertrophy. This par- Mohan et al. 1996; Clemmons et al. 1998). ticular study employed hypophysectomized rats, Both mRNA and protein levels of IGF-I and which further suggests that the observed increase in IGFBPs have been shown to be up-regulated during IGF-I mRNA expression was GH independent. regeneration after ischaemic injury (Jennische & Later studies utilizing a similar model of functional Hall 2000). In situ hybridization studies have shown overload in both normal and hypophysectomized IGFBP-5 to be restricted to regenerating muscle rats found that both mRNA and protein levels of cells, whereas connective tissue cells expressed IGF-I were increased in muscle, prior to the attain- IGFBP-4 (Boes et al. 1992). Mechanically loading, or ment of significant hypertrophy, and remained elev- unloading the muscle, has been shown to regulate ated for up to 28 days during the hypertrophy several of the binding proteins in muscle. For ex- process (Adams & Haddad 1996). In another study ample, Awede et al. (1999) reported that overload- which utilized treadmill training of GH-suppressed ing the muscles in mice increased the expression rats, levels of IGF-I mRNA and protein increased by of IGFBP-4 mRNA, but decreased that of IGFBP-5. 55% and 250% respectively (Zanconato et al. 1994). In contrast, unloading of mouse muscle resulted in Furthermore, overexpression (Coleman et al. 1995) reduced levels of IGFBP5 mRNA, but did not affect or direct infusion (Adams & McCue 1998) of IGF-I in levels of IGFBP-4. Both of these binding proteins muscle results in hypertrophy, whereas inhibition were assumed to mediate the effects of IGF-I via re- of intracellular signaling components associated gulation of the free IGF-I concentration in muscle with IGF-IR activation can prevent this response and possibly via competition with IGF receptors for (Bodine et al. 2001). Another study, examining the IGF-I (Awede et al. 1999). Experiments are under- association between local IGF-I overexpression and way to characterize the specific binding protein for atrophy induced by hind limb unloading, con- MGF, which differs from the other binding proteins cluded that overexpression of IGF-I in the muscles for the other splice variants. of transgenic mice was not shown to prevent unloading-induced atrophy (Criswell et al. 1998). Biological action of the IGF-I splice variants Genetic manipulation of IGF-I in muscle All the IGF-I splice variants have the same receptor- Transgenically modified mice that overexpress the binding domain encoded by exons 3 and 4. This is IGF-I gene have been produced in several laborator- apparently responsible for the anabolic effects of ies. Initial experiments of this kind by Mathews et al. IGF-I. These have been clearly demonstrated by (1988) in which the GH gene was overexpressed, igf-i splice variants in muscle growth 187

reported a 30% increase in muscle and an increase in Gene transfer of the different IGF-I bone density (Mathews et al. 1988). These experi- splice variants ments involved using a metallothionein promoter, which when activated increased the expression of The finding that muscle fibers can be transfected by IGF-I in most tissues. Later, experiments involved a simple intramuscular injection of a plasmid vector IGF-I transgenes that were under the control of containing the cDNA sequence of a gene of interest muscle regulatory elements, such as the α-actin pro- provides a method for treating medical conditions moter (Coleman et al. 1995). Although serum levels that are associated with marked muscle loss. More of IGF-I were not significantly elevated in these recently, experiments in which constructs con- mice, there was a considerable increase in muscle taining the IGF-I splice variants, including MGF mass. More recently, work by the same group has have been performed. For example, in one such shown that both muscle and motor neuron regen- experiment a plasmid construct containing MGF eration was enhanced in these mice when compared cDNA was injected into the muscles of normal mice to wild type (Rabinovsky et al. 2003). It is difficult to to determine the role of the MGF splice variant in determine whether the full length cDNA has been muscle maintenance. This study resulted in a 25% used in these gene transfer experiments and if this increase in the mean muscle fiber size in injected insert includes the sequence for a binding protein. muscle within 2 weeks, and this was shown to be This is important if we are to understand whether due to an increase in the size of the muscle fibers or not the expression of IGF-I is muscle specific; this (Goldspink & Yang 2001). Similar experiments by is usually determined by whether the introduced other groups have also been carried out using a viral DNA is under the control of a muscle specific regu- construct containing the liver type of IGF-I (IGF- latory sequence. IGF-I knock-out mice (lacking the IEa). This also resulted in a less than 20% increase in entire IGF-I gene sequence) have also been pro- muscle mass, but this took over 4 months to develop duced. Therefore, none of the IGF-I peptide can be (Barton-Davis et al. 1998). As well as treating certain transcribed (Baker et al. 1993). However, these mice medical conditions, gene transfer of this type may do not survive very long after birth, as their muscles well be open to abuse. The use of an adenoviral appear to be dystrophic. Interestingly, the results vector for delivery would make detection relatively of recent tissue specific gene deletion experiments simple using PCR based methods, as this virus using the Cre-loxP model of gene deletion have infects most cell types. Plasmid vectors are in con- questioned the role of GH and liver IGF-I in control- trast more difficult to detect, but new approaches ling postnatal growth and development (Sjogren that involve determining the different actions of the et al. 1999; Yakar et al. 1999). This homologous gene products from those of introduced gene are recombination system was used to create a liver underway. specific deletion of the IGF-I gene but allowed nor- mal expression of this gene in other non-hepatic IGF-I signaling pathways in muscle tissues, such as heart, muscle, fat, spleen and kid- hypertrophy ney. The effect of this liver specific gene deletion of IGF-I on growth and development in these mice The signaling pathways by which IGF-I promotes was to reduce circulating IGF-I levels at 6 weeks of skeletal muscle hypertrophy remain unclear, with age compared with wild type animals. Interestingly, roles suggested for both the calcineurin/NFAT measurements of body size and individual organ (nuclear factor of activated T-cells) pathway (Musaro weights at 6 weeks showed no difference between et al. 1999; Semsarian et al. 1999) and the P13-kinase/ the ‘knockout’ animals and their wild type litter- Akt pathway (Rommel et al. 2001). More recently, mates. Thus, postnatal growth and development studies investigating the hypertrophic response proceeded without the contribution of liver derived both in vitro and in vivo have reported that it is the IGF-I (Sjogren et al. 1999), emphasizing the role of Akt/mTOR pathway and not the calcineurin path- the local IGF-I system in the process. way which is involved in promoting hypertrophy, 188 chapter 14

by activating downstream targets such as P > 0.056 animals 20 months of age, whilst it was significantly kinase (Bodine et al. 2001; Rommel et al. 2001). In atrophied in the wild type animals. These genetic addition, it was also reported that IGF-I might in manipulation studies suggest that local IGF-I may fact act via Akt to inhibit the calcineurin/NFAT sig- be an important factor in maintaining muscle mass naling pathway during this process (Rommel et al. in old age. Recently, Owino et al. (2001) used tendon 2001). Unfortunately studies conducted to date have ablation to overload the soleus and plantaris mus- looked at the expression of total IGF-I in response to cles of rats to study any age-related differences in muscle overload. the response of the muscle IGF-I isoforms. Animals of different ages (4, 12, 24 months) were overloaded IGF-I and sarcopenia for 5 days. They reported that the MGF mRNA was up-regulated by ~ 1200% in the young animals Increasing age is associated with a loss of muscle when compared to the control leg, but was up-regu- and mass and function, but the mechanisms under- lated to a significantly less extent (~ 500%) in the old lying this process remain unclear. There is a reduc- animals. The Ea isoform was also up-regulated, but tion in circulating levels of GH and IGF-I (Rudman showed no clear age-related effect. In recent studies et al. 1981) and an actual loss of muscle fibers (Lexell on humans, and as mentioned previously, Hameed et al. 1988). There is some emerging evidence that et al. (2003a) reported that older men showed a IGF-I has a role to play the maintenance of muscle reduced MGF response when measured shortly in later life. Using a recombinant adeno-associated after a single bout of knee extensor weightlifting virus containing a myosin light chain (MLC1/3) exercise compared to young subjects. However, promoter, Barton-Davis et al. (1998) injected the Singh et al. (1999) showed that overall muscle IGF-I cDNA of an IGF-I construct into the extensor digi- levels could be increased in older people as a result torum longus (EDL) muscles of young (6 months) of a 10-week period (three times per week) of and old (27 months) mice. This resulted in overex- strength training exercise. As discussed previously, pression of IGF-I (Ea isoform) in this muscle, but did Hameed et al. (2004) showed that such a training not give rise to elevated circulating levels of IGF-I regimen could cause an increase in the level of tran- in plasma. Four months post-injection, the injected script expression all three of the IGF-I splice variants muscle of the younger animals was on average 15% (IGF-IEa, IGF-IEb and MGF) in the muscles of older larger and stronger than the non-injected muscle. people. Interestingly, it was MGF that showed the In the older animals, the injected muscle was 27% most marked increase (see Figure 14.2). The ability stronger when compared with the non-injected old to do this no doubt contributes to the retained capa- animals, resulting in similar values for mass and city, of even very elderly muscle, to hypertrophy in function to the 6-month-old animals. An age-related response to strength training exercise. improvement in muscle mass and function was also demonstrated in a very recent study using a mouse Mechano growth factor, satellite cells model in which transgenic animals were bred to and muscle damage overexpress the IGF-IEa gene (which they termed m.IGF-I) (Musaro et al. 2001). At 6 months of age, the Satellite cells in skeletal muscle were first described fibers of the transgenic animals were 32 µm in dia- by Mauro (1961). It is now realized that these cells meter as compared to 18 µm in the wild type animals. provide the extra nuclei for postnatal growth (Moss There was, however, preferential hypertrophy of & Leblond 1970; Schultz 1996) and that they are also the faster fiber types (46 µm in the transgenic animal involved in repair and regeneration following local vs. 32 µm in the wild type) with little effect on the injury of muscle fibers (Grounds 1998). In normal slow fibers (16 µm vs. 18 µm). This was attributed to adult undamaged tissue, the satellite cells are the low expression of the MLC regulatory cassette quiescent and usually detected just beneath the (1/3 locus) in slow muscle fibers. In the older trans- basal lamina. They express M-cadherin (M-cad) genic animals, muscle mass was maintained in (Bornemann & Schmalbruch 1994; Irintchev et al. igf-i splice variants in muscle growth 189

1994) when activated, and commence to coexpress much earlier transient response, which peaked at myogenic factors including c-met, MyoD and myf5 4 days post-bupivacaine injection and decreased and later myogenin (Cornelison & Wold 1997; thereafter. Following mechanical damage, the peak Beauchamp et al. 2000; Qu-Petersen et al. 2002). The in MGF expression occurred even earlier. It seems origin of satellite cells is still somewhat uncertain, that in both myotoxin- and mechanical activity- as they were thought to be residual myoblasts induced damage models, the temporal expression (reviewed by Seale & Rudnicki 2000), but there is pattern for each IGF-I splice variant showed sim- accumulating evidence that some may also origin- ilar differential gene splicing sequences, with MGF ate from pluripotent stem cells derived from pro- peaking before IGF-IEa. This temporal difference in genitor cells of the vasculature (Qu-Petersen et al. expression of the two muscle IGF-I mRNA tran- 2002). Pluripotent stem cells from bone marrow scripts has also been described in the rat following cells (Ferrari et al. 1998), as well as epidermal cells commencement of resistance exercise (Haddad & (Pye & Watt 2001), have also been shown to fuse and Adams 2002). As M-cad expression peaked well adopt the muscle phenotype when introduced into before IGF-IEa, whether it was measured as mRNA dystrophic muscle. or protein, it is unlikely that the systemic type It has been established that even in normal mus- of IGF-IEa is responsible for initial activation of cle, local injury does occur from time to time satellite cells. However, it is not possible to say from (Wernig et al. 1990), but in certain diseases such as these data whether this was due to an increase in the muscular dystrophies, the muscle fibers are number of satellite cells as it is known that quiescent markedly more susceptible to damage, in particular satellite cells do stain to some extent for M-cad pro- to the membrane (Cohn & Campbell 2000). The tein (Rosenblatt et al. 1994). Nevertheless, it repres- contractile system of muscle fibers also sustains ents a marked increase of M-cad whether it is in damage during eccentric contractions, that is to say existing satellite cells or an increase in the number of when the muscle is activated whilst being stretched. these cells or both. It is interesting to note that the forces generated by MGF and IGF-IEa splice variants apparently yield activation combined with stretch exceed even those the same mature peptide, which is derived from the of a maximal isometric contraction. In the muscle highly conserved exons 3 and 4 of the IGF-I gene. fibers involved, the sarcomeres may be pulled out These exons, present in all known IGF splice vari- to such a degree that there is no longer any overlap ants are known to encode the IGF-I receptor ligand of the actin and myosin filaments, thus causing domain. A mechanism of extracellular endoproteo- damage (Lieber & Friden 1999). lysis of the IGF-I prohormone results in the same During regeneration of skeletal muscle in young mature peptide (Gilmour 1994), even though the rats following ischaemia or myotoxin-induced dam- splice variants of IGF-I may have different 3’ age, elevated expression of IGF-I has been reported sequences including the E domain. It has been sug- (Jennische & Hansson 1987; Jennische et al. 1987; gested that IGF-I precursors could be pluripotent, in Edwall et al. 1989) which was diminished by the a form analogous to that of the prohormone pro- 15th day of recovery (Marsh et al. 1997). In a more piomelanocortin and proglucagon (Siegfried et al. recent study, the response of the different isoforms 1992). The observation that a synthetic peptide of IGF-I, IGF-IEa and MGF to such stimuli were derived from the rat Eb domain induces prolifera- studied for the first time and related to the activa- tion in epithelial cells is noteworthy (Siegfried et al. tion of muscle satellite (stem) cells in vivo (Hill 1992). The growth promoting properties of the MGF & Goldspink 2003). Results of the experiments, E peptide and its role as an independent growth fac- in which damage was induced by bupivacaine, tor is supported by recent cell culture experiments demonstrated a surge of IGF-IEa mRNA expression where stable transfection with MGF was shown to that was maximal at 11 days and diminished there- stimulate myoblast proliferation but differentiation after to similar levels as those in the non-injected was suppressed. The addition of a synthetic MGF animals. On the other hand, MGF mRNA showed a peptide or the medium from MGF-transfected cells 190 chapter 14

onto normal C2C12 cells also inhibited their differ- that MGF is not appropriately expressed in dys- entiation. Yet this inhibition was reversed when the trophic muscles (Goldspink 1996) and the decrease peptide or the medium were withdrawn. In con- in MGF mRNA levels in response to mechanical trast, cells of the liver type, systemic IGF-I (IGF-IEa) overload in older muscles (Owino et al. 2001). There positive clone did form myotubes and the normal is a deficiency of active satellite cells in both these cell lines showed less cellular proliferation as well situations in which local tissue repair becomes as forming myotubes. Of particular interest was the increasingly impaired. Future experiments invest- observation that when an IGF-I receptor antibody igating the expression of the two transcripts and was added to the muscle cell cultures, cell prolifera- activation of satellite cells in young and old muscles tion induced by MGF was not inhibited, whereas after the therapeutic application of MGF and IGF- their stimulation to increase in mass and to form IEa to ameliorate muscle loss are in progress. myotubes by IGF-I was reduced. These data strongly suggest that MGF is involved in another signaling Summary pathway in addition to that associated with the IGF- I receptor (Yang & Goldspink 2002). It is becoming increasingly clear that muscle adapta- The results of these studies provide additional tion to high resistance exercise as well as its res- insight into the complexity and implication of the ponse to contraction induced damage or otherwise, IGF-I system in conditions of damage and sub- is mediated at the local level by local growth and sequent regeneration. IGF-IEa and MGF are pro- repair factors. The IGF-I family comprises members duced by the active muscle in rodents and humans that play different roles in the growth and repair and have been shown to be positive regulators of processes. Alternative splicing is a subtle and muscle hypertrophy (McKoy et al. 1999; Goldspink sophisticated mechanism by which different IGF-I 2001; Owino et al. 2001; Hameed et al. 2003). How- isoforms can be generated from the same gene to ever, as reported here, the MGF isoform is acutely have different biological roles. Understanding these induced, whereas IGF-IEa has a delayed effect that roles and how these IGF-I isoforms may interact is sustained during the later phase of regeneration. with other control processes in muscle (e.g. testos- When comparing mechanical damage with myotoxin terone, myostatin, ubiquitin, etc.) will help in damage it is apparent that both involve a relatively optimizing training regimens for athletes. Further- rapid expression of the MGF splice variant, although more, an understanding of the signaling processes it may seem that this growth/repair factor has been triggered by exercise training will also provide us misnamed ‘mechano growth factor’. However, even with the means to develop more effective methods in the case of myotoxin-induced damage it is likely for testing exogenous enhancing substances used in that the damaged tissue mass is subjected to doping. increased mechanical strain. Also, it is known that the cells swell following damage, thus resulting in Acknowledgments the same cellular response. As the expression of the autocrine splice variant (MGF) precedes satellite cell This work was supported by grants from the activation it is likely that this form of IGF-I is asso- Wellcome Trust, the World Anti-Doping Agency ciated with satellite cell activation, not the systemic (WADA) and an EC (PENAM) grant for studying IGF-IEa type. This is in accord with the finding the effects of exercise including muscle damage.

References

Adams, G.R. (2002) Invited review: relationships among IGF-1, DNA Adams, G.R. & McCue, S.A. (1998) autocrine/paracrine IGF-I and skeletal content, and protein accumulation Localized infusion of IGF-I results in muscle adaptation. Journal of Applied during skeletal muscle hypertrophy. skeletal muscle hypertrophy in rats. Physiology 93, 1159–1167. Journal of Applied Physiology 81, Journal of Applied Physiology 84, Adams, G.R. & Haddad, F. (1996) The 2509–2516. 1716–1722. igf-i splice variants in muscle growth 191

Awede, B., Thissen, J., Gailly, P. & Lebacq, Chew, S.L., Lavender, P., Clark, A.J. & DeVol, D.L., Rotwein, P., Sadow, J.L., J. (1999) Regulation of IGF-I, IGFBP-4 Ross, R.J. (1995) An alternatively spliced Novakofski, J. & Bechtel, P.J. (1990) and IGFBP-5 gene expression by loading human insulin-like growth factor-I Activation of insulin-like growth factor in mouse skeletal muscle. FEBS Letters transcript with hepatic tissue expression gene expression during work-induced 461, 263–267. that diverts away from the mitogenic skeletal muscle growth. American Journal Baker, J., Liu, J.P., Robertson, E.J. & IBE1 peptide. Endocrinology 136, of Physiology 259, E89–E95. Efstratiadis, A. (1993) Role of insulin- 1939–1944. De Vroede, M.A., Romanus, J.A., like growth factors in embryonic and Clemmons, D.R., Busby, W., Clarke, J.B. Standaert, M.L. et al. (1984) Interaction postnatal growth. Cell 75, 73–82. et al. (1998) Modifications of insulin-like of insulin-like growth factors with a Bamman, M.M., Shipp, J.R., Jiang, J. et al. growth factor binding proteins and their nonfusing mouse muscle cell line: (2001) Mechanical load increases muscle role in controlling IGF actions. Endocrine binding, action, and receptor down- IGF-I and androgen receptor mRNA Journal 45 (suppl.), S1–S8. regulation. Endocrinology 114, 1917–1929. concentrations in humans. American Cohn, R.D. & Campbell, K.P. (2000) Duclos, M.J., Wilkie, R.S. & Goddard, C. Journal of Physiology. Endocrinology and Molecular basis of muscular (1991) Stimulation of DNA synthesis in Metabolism 280, E383–E390. dystrophies. Muscle and Nerve 23, chicken muscle satellite cells by insulin Barton-Davis, E.R., Shoturma, D.I., 1456–1471. and insulin-like growth factors: Musaro, A., Rosenthal, N. & Sweeney, Coleman, M.E., DeMayo, F., Yin, K.C. et al. evidence for exclusive mediation by H.L. (1998) Viral mediated expression (1995) Myogenic vector expression of a type-I insulin-like growth factor of insulin-like growth factor I blocks the insulin-like growth factor I stimulates receptor. Journal of Endocrinology 128, aging-related loss of skeletal muscle muscle cell differentiation and myofiber 35–42. function. Proceedings of the National hypertrophy in transgenic mice. Edwall, D., Schalling, M., Jennische, E. Academy of Sciences of the United States Journal of Biological Chemistry 270, 12 & Norstedt, G. (1989) Induction of of America 95, 15 603–15 607. 109–12 116. insulin-like growth factor I messenger Baumann, G. (1991) Growth hormone Cornelison, D.D. & Wold, B.J. (1997) ribonucleic acid during regeneration of heterogeneity: genes, isohormones, Single-cell analysis of regulatory gene rat skeletal muscle. Endocrinology 124, variants, and binding proteins. Endocrine expression in quiescent and activated 820–825. Reviews 12(4), 424–449. mouse skeletal muscle satellite cells. Ewton, D.Z., Falen, S.L. & Florini, J.R. Beauchamp, J.R., Heslop, L., Yu, D.S. et al. Developmental Biology 191, 270–283. (1987) The type II insulin-like growth (2000) Expression of CD34 and Myf5 Criswell, D.S., Booth, F.W., DeMayo, F. factor (IGF) receptor has low affinity for defines the majority of quiescent adult et al. (1998) Overexpression of IGF-I in IGF-I analogs: pleiotypic actions of IGFs skeletal muscle satellite cells. Journal of skeletal muscle of transgenic mice does on myoblasts are apparently mediated Cell Biology 151, 1221–1234. not prevent unloading-induced atrophy. by the type I receptor. Endocrinology 120, Beshyah, S.A., Freemantle, C., Thomas, E. American Journal of Physiology 275, 115–123. & Johnston, D.G. (1995) Comparison of E373–E379. Ferrari, G., Cusella-De Angelis, G., Coletta, measurements of body composition by Cuneo, R.C., Salomon, F., Wiles, C.M., M. et al. (1998) Muscle regeneration total body potassium, bioimpedance Hesp, R. & Sonksen, P.H. (1991) Growth by bone marrow-derived myogenic analysis, and dual-energy X-ray hormone treatment in growth hormone- progenitors. Science 279, 1528–1530. absorptiometry in hypopituitary adults deficient adults. I. Effects on muscle Florini, J.R., Ewton, D.Z. & Coolican, S.A. before and during growth hormone mass and strength. Journal of Applied (1996) Growth hormone and the insulin- treatment. American Journal of Clinical Physiology 70, 688–694. like growth factor system in myogenesis. Nutrition 61, 1186–1194. Czerwinski, S.M., Martin, J.M. & Bechtel, Endocrine Reviews 17, 481–517. Blundell, T.L., Bedarkar, S. & Humbel, R.E. P.J. (1994) Modulation of IGF mRNA Gilmour, R.S. (1994) The implications (1983) Tertiary structures, receptor abundance during stretch-induced of insulin-like growth factor mRNA binding and antigenicity of insulin like skeletal muscle hypertrophy and heterogeneity. Journal of Endocrinology growth factors. Federation Proceedings 42, regression. Journal of Applied Physiology 140, 1–3. 2592–2597. 76, 2026–2030. Goldspink, G. (1996) Muscle growth and Bodine, S.C., Stitt, T.N., Gonzalez, M. et al. Damon, S.E., Haugk, K.L., Swisshelm, K. muscle function: a molecular biological (2001) Akt/mTOR pathway is a crucial & Quinn, L.S. (1997) Developmental perspective. Research in Veterinary regulator of skeletal muscle hypertrophy regulation of Mac25/insulin-like growth Science 60, 193–204. and can prevent muscle atrophy in vivo. factor-binding protein-7 expression in Goldspink, G. (2001) Method of treating Nature Cell Biology 3, 1014–1019. skeletal myogenesis. Experimental Cell muscular disorders. United States Patent Boes, M., Booth, B.A., Sandra, A. et al. Research 237, 192–195. No: US 6 221 842 B1. (1992) Insulin-like growth factor binding Daughaday, W.H. & Reeder, C. (1966) Goldspink, G. & Yang, S.Y. (2001) Method protein (IGFBP)4 accounts for the Synchronous activation of DNA of treating muscular disorders. United connective tissue distribution of synthesis in hypophysectomized rat States Patent No: 6 221 842 B1. endothelial cell IGFBPs perfused cartilage by growth hormone. Journal Goldspink, G., Scutt, A., Loughna, P.T. through the isolated heart. Endocrinology of Laboratory and Clinical Medicine 68, et al. (1992) Gene expression in skeletal 131, 327–330. 357–368. muscle in response to stretch and force Bornemann, A. & Schmalbruch, H. (1994) Daughaday, W.H., Hall, K., Raben, M.S. generation. American Journal of Immunocytochemistry of M-cadherin in et al. (1972) Somatomedin: proposed Physiology 262, R356–R363. mature and regenerating rat muscle. designation for sulphation factor. Nature Green, H., Morikawa, M. & Nixon, T. Anatomical Records 239, 119–125. 235, 107. (1985) A dual effector theory of 192 chapter 14

growth-hormone action. Differentiation express insulin-like growth factor I. Mathews, L.S., Hammer, R.E., Behringer, 29, 195–198. Acta Physiologica Scandinavica 130, R.R. et al. (1988) Growth enhancement Grounds, M.D. (1998) Age-associated 327–332. of transgenic mice expressing human changes in the response of skeletal Jennische, E., Skottner, A. & Hansson, H.A. insulin-like growth factor I. muscle cells to exercise and (1987) Satellite cells express the trophic Endocrinology 123, 2827–2833. regeneration. Annals of the New York factor IGF-I in regenerating skeletal Mauro, A. (1961) Satellite cell of skeletal Academy of Sciences 20(854), 78–91. muscle. Acta Physiologica Scandinavica muscle fibers. Journal of Biophysical and Haddad, F. & Adams, G.R. (2002) Selected 129, 9–15. Biochemical Cytology 9, 493–495. contribution: acute cellular and Jones, J.I. & Clemmons, D.R. (1995) Insulin- Mohan, S., Baylink, D.J., Pettis, J.L. (1996) molecular responses to resistance like growth factors and their binding Insulin-like growth factor (IGF)-binding exercise. Journal of Applied Physiology 93, proteins: biological actions. Endocrine proteins in serum––do they have 394–403. Reviews 16, 3–34. additional roles besides modulating the Hameed, M., Orrell, R.W., Cobbold, M., Klapper, D.G., Svoboda, M.E. & Van Wyk, endocrine IGF actions? Journal of Clinical Goldspink, G. & Harridge, S.D. (2003a) J.J. (1983) Sequence analysis of Endocrinology and Metabolism 81(11), Expression of IGF-I splice variants in somatomedin-C: confirmation of 3817–20. young and old human skeletal muscle identity with insulin-like growth factor Moss, F.P. & Leblond, C.P. (1970) Nature after high resistance exercise. Journal of I. Endocrinology 112, 2215–2217. of dividing nuclei in skeletal muscle of Physiology 547, 247–254. Lange, K.H., Andersen, J.L., Beyer, N. growing rats. Journal of Cell Biology 44, Hameed, M., Toft, A.D., Pedersen, B.K., et al. (2002) GH administration changes 459–462. Harridge, S.D. & Goldspink, G. (2003b) myosin heavy chain isoforms in skeletal Musaro, A., McCullagh, K.J., Naya, F.J., The effects of eccentric exercise on IGF-I muscle but does not augment muscle Olson, E.N. & Rosenthal, N. (1999) IGF-1 isoform expression in the muscle of strength or hypertrophy, either alone induces skeletal myocyte hypertrophy young and elderly people. Journal of or combined with resistance exercise through calcineurin in association with Muscle Research and Cell Motility 24, training in healthy elderly men. Journal GATA-2 and NF-ATc1. Nature 400, 323–386. of Clinical Endocrinology and Metabolism 581–585. Hameed, M., Lange, K.H., Andersen, H., 87, 513–523. Musaro, A., McCullagh, K.J., Paul, A. Schjerling, P., Kjær, M., Harridge, S.D.R. Le Roith, D. & Roberts, C.T., Jr. (2003) The et al. (2001) Localized IGF-1 transgene & Goldspink, G. et al. (2004) The effect of insulin-like growth factor system and expression sustains hypertrophy and recombinant human growth hormone cancer. Cancer Letters 195, 127–137. regeneration in senescent skeletal and resistance training on IGF-I mRNA Lexell, J., Taylor, C.C. & Sjostrom, M. muscle. Nature Genetics 27, 195–200. expression in the muscles of elderly (1988) What is the cause of the ageing Narhi, L.O., Hua, Q.X., Arakawa, T. et al. men. Journal of Physiology 15(1), 231–240. atrophy? Total number, size and (1993) Role of native disulfide bonds in Hellsten, Y., Hansson, H.A., Johnson, L., proportion of different fiber types the structure and activity of insulin-like Frandsen, U. & Sjodin, B. (1996) studied in whole vastus lateralis muscle growth factor 1: genetic models of Increased expression of xanthine from 15- to 83-year-old men. Journal of protein-folding intermediates. oxidase and insulin-like growth factor I the Neurological Sciences 84, 275–294. Biochemistry 32, 5214–5221. (IGF-I) immunoreactivity in skeletal Lieber, R.L. & Friden, J. (1999) Mechanisms Nindl, B.C., Kraemer, W.J., Marx, J.O., muscle after strenuous exercise in of muscle injury after eccentric Tuckow, A.P. & Hymer, W.C. (2003) humans. Acta Physiologica Scandinavica contraction. Journal of Science and Growth hormone molecular 157, 191–197. Medicine in Sport/Sports Medicine heterogeneity and exercise. Exercise and Hill, M. & Goldspink, G. (2003) Expression Australia 2, 253–265. Sport Sciences Reviews 31(4), 161–166. and splicing of the insulin-like growth Lowe, W.L., Jr., Lasky, S.R., LeRoith, D. Owino, V., Yang, S.Y. & Goldspink, G. factor gene in rodent muscle is & Roberts, C.T., Jr. (1988) Distribution (2001) Age-related loss of skeletal associated with muscle satellite (stem) and regulation of rat insulin-like growth muscle function and the inability to cell activation following local tissue factor I messenger ribonucleic acids express the autocrine form of insulin- damage. Journal of Physiology 549, encoding alternative carboxyterminal like growth factor-1 (MGF) in response 409–418. E-peptides: evidence for differential to mechanical overload. FEBS Letters Irintchev, A., Zeschnigk, M., Starzinski- processing and regulation in liver. 505, 259–263. Powitz, A. & Wernig, A. (1994) Molecular Endocrinology (Baltimore, Md) Putzer, P., Breuer, P., Gotz, W. et al. Expression pattern of M-cadherin in 2, 528–535. (1998) Mouse insulin-like growth normal, denervated, and regenerating McKoy, G., Ashley, W., Mander, J. et al. factor binding protein-6: expression, mouse muscles. Developmental Dynamics (1999) Expression of insulin growth purification, characterization and 199, 326–337. factor-1 splice variants and structural histochemical localization. Molecular and Jennische, E. & Hall, C.M. (2000) genes in rabbit skeletal muscle induced Cellular Endocrinology 137, 69–78. Expression and localisation of IGF- by stretch and stimulation. Journal of Pye, D. & Watt, D.J. (2001) Dermal binding protein mRNAs in regenerating Physiology 516(2), 583–592. fibroblasts participate in the formation rat skeletal muscle. APMIS: Acta Marsh, D.R., Criswell, D.S., Hamilton, of new muscle fibres when implanted Pathologica, Microbiologica, et M.T. & Booth, F.W. (1997) Association into regenerating normal mouse muscle. Immunologica Scandinavica 108(11), of insulin-like growth factor mRNA Journal of Anatomy 198, 163–173. 747–755. expressions with muscle regeneration Qu-Petersen, Z., Deasy, B., Jankowski, R. Jennische, E. & Hansson, H.A. (1987) in young, adult, and old rats. American et al. (2002) Identification of a novel Regenerating skeletal muscle cells Journal of Physiology 273, R353–R358. population of muscle stem cells in igf-i splice variants in muscle growth 193

mice: potential for muscle regeneration. relation to age and adiposity. Journal of Wernig, A., Irintchev, A. & Weisshaupt, P. Journal of Cell Biology 157, 851–864. Clinical Investigation 67, 1361–1369. (1990) Muscle injury, cross-sectional Rabinovsky, E.D., Gelir, E., Gelir, S. et al. Schlechter, N.L., Russell, S.M., Spencer, area and fibre type distribution in (2003) Targeted expression of IGF-1 E.M. & Nicoll, C.S. (1986) Evidence mouse soleus after intermittent wheel- transgene to skeletal muscle accelerates suggesting that the direct growth- running. Journal of Physiology 428, muscle and motor neuron regeneration. promoting effect of growth hormone 639–652. FASEB Journal 17, 53–55. on cartilage in vivo is mediated by local Yakar, S., Liu, J.L., Stannard, B. et al. (1999) Rees, C. & Clemmons, D.R. (1998) production of somatomedin. Proceedings Normal growth and development in the Inhibition of IGFBP-5 binding to of the National Academy of Sciences of the absence of hepatic insulin-like growth extracellular matrix and IGF-I- United States of America 83, 7932–7934. factor I. Proceedings of the National stimulated DNA synthesis by a peptide Schultz, E. (1996) Satellite cell proliferative Academy of Sciences of the United States of fragment of IGFBP-5. Journal of Cellular compartments in growing skeletal America 96, 7324–7329. Biochemistry 71, 375–381. muscles. Developmental Biology 175, 84–94. Yang, S.Y. & Goldspink, G. (2002) Different Rinderknecht, E. & Humbel, R.E. (1978) Seale, P. & Rudnicki, M.A. (2000) A new roles of the IGF-I Ec peptide (MGF) and The amino acid sequence of human look at the origin, function, and ‘stem- mature IGF-I in myoblast proliferation insulin-like growth factor I and its cell’ status of muscle satellite cells. and differentiation. FEBS Letters 522, structural homology with proinsulin. Developmental Biology 218, 115–124. 156–160. Journal of Biological Chemistry 253, Semsarian, C., Wu, M.J., Ju, Y.K et al. Yang, S.Y., Alnaqeeb, M., Simpson, H. 2769–2776. (1999) Skeletal muscle hypertrophy & Goldspink, G. (1996) Cloning and Roberts, P., McGeachie, J.K. & Grounds, is mediated by a Ca2+-dependent characterization of an IGF-I isoform M.D. (1997) The host environment calcineurin signalling pathway. Nature expressed in skeletal muscle subjected determines strain-specific differences 400, 576–581. to stretch. Journal of Muscle Research and in the timing of skeletal muscle Siegfried, J.M., Kasprzyk, P.G., Treston, Cell Motility 17, 487–495. regeneration: cross-transplantation A.M. et al. (1992) A mitogenic peptide Yarasheski, K.E., Campbell, J.A., Smith, K. studies between SJL/J and BALB/c amide encoded within the E peptide et al. (1992) Effect of growth hormone mice. Journal of Anatomy 191(4), 585–594. domain of the insulin-like growth factor and resistance exercise on muscle Rommel, C., Bodine, S.C., Clarke, B.A. IB prohormone. Proceedings of the growth in young men. American Journal et al. (2001) Mediation of IGF-1-induced National Academy of Sciences of the United of Physiology 262, E261–E267. skeletal myotube hypertrophy States of America 89, 8107–8111. Yarasheski, K.E., Zachwieja, J.J., Campbell, by PI(3)K/Akt/mTOR and Singh, M.A., Ding, W., Manfredi, T.J. et al. J.A. & Bier, D.M. (1995) Effect of growth PI(3)K/Akt/GSK3 pathways. Nature (1999) Insulin-like growth factor I in hormone and resistance exercise on Cell Biology 3, 1009–1013. skeletal muscle after weight-lifting muscle growth and strength in older Rosenblatt, J.D., Yong, D. & Parry, D.J. exercise in frail elders. American Journal men. American Journal of Physiology 268, (1994) Satellite cell activity is required of Physiology 277, E135–E143. E268–E276. for hypertrophy of overloaded adult rat Sjogren, K., Liu, J.L., Blad, K. et al. (1999) Yu, K.T. & Czech, M.P. (1984) The type I muscle. Muscle and Nerve 17, 608–613. Liver-derived insulin-like growth factor insulin-like growth factor receptor Rotwein, P., Pollock, K.M., Didier, D.K. I (IGF-I) is the principal source of IGF-I mediates the rapid effects of & Krivi, G.G. (1986) Organization and in blood but is not required for postnatal multiplication-stimulating activity on sequence of the human insulin-like body growth in mice. Proceedings of the membrane transport systems in rat growth factor I gene. Alternative RNA National Academy of Sciences of the United soleus muscle. Journal of Biological processing produces two insulin-like States of America 96, 7088–7092. Chemistry 259, 3090–3095. growth factor I precursor peptides. Vandenburgh, H.H., Karlisch, P., Shansky, Zanconato, S., Moromisato, D.Y., Journal of Biological Chemistry 261, J. & Feldstein, R. (1991) Insulin and IGF-I Moromisato, M.Y. et al. (1994) Effect 4828–4832. induce pronounced hypertrophy of of training and growth hormone Rudman, D., Kutner, M.H., Rogers, C.M. skeletal myofibers in tissue culture. suppression on insulin-like growth et al. (1981) Impaired growth hormone American Journal of Physiology 260, factor I mRNA in young rats. Journal of secretion in the adult population: C475–C484. Applied Physiology 76, 2204–2209. Chapter 15

Adrenal Gland: Fight or Flight Implications for Exercise and Sports

MICHAEL KJÆR

Introduction thermoregulation; (iii) increase contractility of skel- etal muscle; and (iv) cause suppression/stimulation Fight and flight reaction has been important for of components involved in the human immune sys- humans throughout history. To be able to react tem during exercise. rapidly and mobilize energy for fast movements away from danger, or to be able to exert high Adrenal gland neuroendocrine changes amounts of muscle force in confrontations, requires with exercise a system by which a co-ordinated series of events can be stimulated by a fast-releasing hormonal sys- A major component in the autonomic control during tem. The adrenal gland, especially the adrenal exercise is the adrenergic activity, which can be medulla, is very well suited for this. This is because assessed in humans both by direct measurements of epinephrine has potent effects in this regards; electrical activity in superficial sympathetic nerves the half-life of epinephrine is very short whereby and by determination of circulating norepinephrine mobilization is quick. Finally, several of the effects and epinephrine in the blood. Direct recording of that epinephrine exerts can be initiated, even in the sympathetic activity in humans can, for practical resting state, when adequate stimulus (mental reasons, be performed to resting muscle only (Searls stress) can elevate circulating levels of epinephrine et al. 1988), whereas direct recording of sympathetic to a level that will bring forward, for example, activity to the adrenal medulla can be determined metabolic effects. in animal models. Epinephrine is released from the Energy turnover increases with onset of exercise, adrenal medulla in response to sympathetic neural and autonomic/endocrine mechanismsaespecially activity to the gland during exercise, whereas those involving hormones released from the adrenal adrenal cortical hormones (i.e. cortisol) are released glandaplay an important role in this regulation. due to pituitary hormonal stimulation by adreno- This results in increased mobilization of substrate: corticotropic hormone (ACTH). Levels of norepine- i.e. (a) glycogenolysis in skeletal muscle; (b) glucose phrine and epinephrine in arterial blood increase release from the liver to support muscle glycolysis; with exercise intensity, expressed by the percentage o (c) free fatty acid release from both adipose and of maximal individual performance (% V 2max), muscle tissue for beta-oxidation in skeletal muscle; and, as clearance of these hormones only changes and (d) increase protein synthesis. In addition to moderately with exercise, changes in plasma levels playing an important role in substrate mobilization can be attributed to changes in secretion and release and combustion, adrenal gland related endocrine (Kjær et al. 1985). It has been shown that whole-body mechanisms together with autonomic mechanisms clearance of epinephrine increases by 15% at low- can also: (i) influence blood distribution to various exercise intensities and decreases around 20% below organs; (ii) stimulate sweat glands and influence basal levels after more intense exercise (Kjær et al. 194 adrenal gland and exercise 195

1985). However, as the increase in plasma epine- physical training. This is supported by findings in phrine seen during dynamic exercise in humans is trained men (sprint) who showed a higher epine- five to 10-fold, these changes are caused by increases phrine response to short-term exercise compared in secretion from the adrenal medulla rather than with that of untrained counterparts (Jacob 2004). by changes in clearance. A major contributor to Interestingly, no significant difference could be epinephrine clearance is the hepatosplancnic area as obtained in trained versus untrained women using well as the kidneys. this short-term sprint test. In rats who underwent Epinephrine and cortisol responses are influenced 10 weeks of intense swim training, the adrenal by glucose levels during exercise, and falling glucose medullary volume and the adrenal content of epine- markedly enhance the responses of these two gluco- phrine was larger in trained rats (both male and counteregulatory hormones during prolonged exer- female) compared with controls who were either cise (Galbo et al. 1977). In addition to influencing weight matched, sham trained or cold stressed carbohydrate and fat metabolism, epinephrine per (Stallknecht et al. 1990). Interestingly, the training se has been shown to increase protein metabolism induced increase in adrenal medulla volume was in isolated electrically stimulated muscle. Although paralleled by the increase in adrenal gland weight, epinephrine responses can be seen to acute mental indicating that the major stimulus was on the stress, these changes are by far lower than the ones medulla rather than the adrenal cortex. Further- observed during physical exercise. more, no increase in total gland weight was seen after sham training or cold stress, indicating that Adrenomedullary activity after these ‘mental’ types of stress did not result in any physical training major gland adaptation. Although these findings indicate that the improved secretion capacity of Vigorous endurance training will reduce the cate- epinephrine is a result of training, this will most cholamine response to a given absolute workload, likely require several years of training. In well- whereas neither sympathetic nerve activity nor trained athletes who underwent hypoglycemia norepinephrine levels at maximal workloads differ before and 4–5 weeks after an injury that resulted in between individuals with different training status. inactivity, epinephrine responses did not change This supports the view that physical training does with this short-lasting alteration in activity level not alter the capacity of the sympathetic nervous (Kjær et al. 1992). It could be speculated that an early systemaand thus most likely neither the sympath- influence upon the adrenal medullary volume is etic activity directed towards the adrenal medullaa needed, for example during adolescence, and some but that responses to submaximal exercise are reports seem to indicate this (Jacob 2004). However, linked closely to the relative rather than to the even if this adaptation requires long-term stimula- absolute workload. Surprisingly, however, in a 24-h tion, it is interesting that endocrine glands appar- study it has been found that, if anything, highly ently can adapt to physical training and alter their trained individuals had a higher catecholamine secretion capacity, similar to other tissues like mus- release over the day compared with sedentary indi- cle and heart. viduals. Epinephrine response in trained versus sedentary individuals has been shown to be enlarged Motor control and neural reflex influence when stimulated by a variety of stimuli such as on adrenal gland responses hypoglycemia, caffeine, glucagon, hypoxia and hypercapnia (Kjær et al. 1986, 1988; Kjær & Galbo Trying to address the role of motor control for 1988). This indicates that the capacity to secrete adrenal endocrine responses during exercise can epinephrine from the adrenal medulla improves basically be studied either by enhancing the volun- with training, the development of a so-called ‘sports tary effect to carry out a certain workload and thus adrenal medulla’ reflecting a hypertrophy pheno- exagerating the central effort, or it can be studied by menon similar to that seen in a heart adapting to decreasing the contribution of motor centers by, for 196 chapter 15

example, performing involuntary exercise and, so to were absent. Interestingly, both ACTH and β- speak, ‘letting the work being done to you’. Attempts endorphin responses during submaximal exercise to increase motor center activity for a given force were abolished during epidural anesthesia (Kjær output has been adressed in humans using partial et al. 1989). In support of a role of afferent nerves neuromuscular blockade to weaken the maximal in adrenergic and pituitary responses, hormonal force by 30–70% (Kjær et al. 1987). It was found that levels in the blood increased in response to direct an exercise-induced increase in levels of circulating stimulation of these nerve fibers in cats (Vissing et al. catecholamines (and pituitary hormones [growth 1994). hormone and ACTH]) was augmented compared to control experiments with saline infusion. These Hepatosplanchnic glucose production findings are supported by experiments in paralyzed and adrenal gland responses to exercise cats, where direct stimulation of the subthalamic locomotor areas in the brain resulted in adrenergic During intense exercise the rise in hepatic glucose hormonal responses similar to the one seen during production parallels a rise in plasma epinephrine voluntary exercise (Vissing et al. 1989). Together, levels. In swimming rats, the removal of the adre- these experiments support the view that motor nal medulla reduced the hepatic glycogenolysis. center activity can directly stimulate sympatho- Furthermore, exercise-induced increase in hepatic adrenergic activity during exercise both directly glucose production was diminished by adreno- and independently of feedback from contracting demedullation in running rats (Sonne et al. 1985). muscle. That these central factors coupled to exer- However, most studies have been unable to demon- cise intensity are not sufficient to elicit a maximal strate any direct effect of epinephrine on liver glyco- adrenergic and pituitary hormonal responses can gen breakdown during exercise. In running dogs, be demonstrated in several ways. When exercising evidence has been provided that epinephrine may with a small muscle group, for example one-legged play a minor role in liver glucose output late during knee extension even at maximal intensity, only a exercise, probably due to an increased gluconeogenic small catecholamine response can be observed precursor level (Moates et al. 1988). Furthermore, (Savard et al. 1989). Furthermore, when maximal adrenalectomized individuals maintain a normal work output was reduced by more than 60% with rise in hepatic glucose production during exercise, neuromuscular blockade (tubocurarine), despite and only when epinephrine is infused in these pati- subjects working at the highest possible effort, ents was hepatic glucose production augmented adrenergic responses were far from maximal. during the early stages of exercise (Howlett et al. Mechanisms other than central motor control there- 1999). In humans the role of liver nerves and epine- fore need to be active during exercise, and one such phrine have been studied with application of local mechanism is neural feedback from contracting the anesthesia around the sympathetic coeliac ganglion skeletal muscle. This can be investigated by using innervating liver, pancreas and adrenal medulla lumbar epidural anesthesia in doses sufficiently (Kjær et al. 1993). Pancreatic hormones were stand- high enough to block impulses in thinner afferent ardized by the infusion of somatostatin, glucagon nerves (type III and IV or type C) but still preserving and insulin. During blockade, the exercise-induced motor nerves and the ability to perform exercise to epinephrine response was inhibited by up to 90%, the highest possible degree. Such an approach evid- and presumably liver nerves were also blocked, ently has weaknesses, and negative findings of a but this did not diminish the glucose production blockade does not definitely exclude a role of affer- response to exercise. This indicates that sympatho- ent nerves since a perfect distinction between affer- adrenergic activity is not responsible for the exercise- ent and efferent nerves cannot fully be obtained by induced rise in splanchnic glucose output. In further this approach. Nevertheless, during static exercise, support of this hypothesis, the exercise-induced but not during dynamic exercise, catecholamine increase in liver glucose production was identical in responses were inhibited when afferent responses liver-transplanted patients compared to healthy adrenal gland and exercise 197

control subjects, as well as kidney-transplanted Spinal-cord-injured individuals are characterized patients who received a similar hormonal and by a lack of voluntary control over their lower limbs, immunosuppressive drug treatment as the liver- and are absent of neural feedback from muscles transplanted patients (Kjær et al. 1996a). Finally, in to higher brain centers. Development of equipment experiments in exercising dogs that underwent a has allows for functionally electrical-stimulated selective blockade of hepatic α- and β-receptors, it exercise on an ergometer, causing oxygen uptake was demonstrated that circulating norepinephrine rates to rise to around 1.0–1.5 L·min–1. This allows for and epinephrine do not participate in the stimula- the study of carbohydrate and fat metabolism and tion of glucose production during intense exercise metabolic changes during exercise. Using involunt- (Coker et al. 1997). Taken together, sympathetic liver ary exercise in spinal-cord-injured individuals as an nerves or circulating norepinephrine play no role in intervention showed that, in the absence of motor glucose mobilization from the liver during exercise, control and neural feedback from muscle, mobiliza- and circulating epinephrine only plays a minor role tion of glucose from the liver was impaired, resulting during intense exercise and late, prolonged exercise. in a gradual drop in plasma glucose during exercise Cortisol has been shown only to play a minor role in (Kjær et al. 1996b). In line with this, healthy indivi- hepatic glucose production during exercise, and duals who were paralyzed by an epidural blockade only seems to be of importance if inadequate secre- also had impaired glucose mobilization response tion of other hormones is present. (Kjær et al. 1998). Furthermore, in spinal-cord- injured patients during voluntary arm cranking, Epinephrine effect on muscle euglycemia was maintained during exercise. These carbohydrate metabolism findings indicate that neural mechanisms are crucial for the matching of glucose mobilization to periph- Epinephrine has been demonstrated to stimulate eral glucose uptake during exercise, and that blood- glycogen breakdown in skeletal muscle during borne mechanisms are not sufficient to accomplish contraction both in exercising animals and humans this. During electrical exercise by spinal-cord-injured when supra-physiological doses were used (Richter individuals, the primary energy source is glyco- 1996). Later studies using more physiological epine- genolysis, and higher levels of lactate have been phrine doses have not been able to demonstrate found both in muscle and blood. Furthermore, the any increased glycogen breakdown despite higher glucose uptake is several-fold higher in spinal-cord- activation of phosphorylase compared with control injured patients compared with healthy controls studies). In line with this, in adrenalectomized indi- working at the same oxygen uptake rate. viduals no impairment in glycogen degradation was found during exercise, neither was muscle glyco- Sympathoadrenergic activity and genolysis increased by substituting epinephrine fat metabolism during exercise (Kjær et al. 2000). It was furthermore shown that activation of glycogen phosphorylase Intravenous infusion of epinephrine in resting and hormone-sensitive lipase is only present if these humans caused an increase in lipolytic activity individuals receive infusion with epinephrine to as determined by microdialysis of subcutaneous mimic changes in epineprine occuring in healthy adipose tissue, an effect that was desensitized by individuals during exercise. This indicates the im- repeated epinephrine infusions (Stallknecht 2003). portance of epinephrine in the activation of glyco- Spinal-cord-injured patients were investigated dur- genolytic and lipolytic pathways. It also indicates ing arm cranking, and lipolysis was determined by that such activation occurs in parallel for intramuscu- microdialysis in subcutaneous adipose tissue both lar triglyceride and glycogen by adrenergic activity, above and below the cutaneous border separat- and that the choice of substrate for energy produc- ing the sympathetic innervated region (proximal– tion takes place at another level in the muscle (Kjær clavicular region) from the desympathectomized et al. 2000). areas (distal–umbilical region) (Stallknecht et al. 198 chapter 15

2001). In both regions lipolysis increased with exer- that the choice of substrate for energy production cise, evidence that direct sympathetic innervation is takes place at another level. not of major importance for the lipolysis during muscular work. In contrast, circulating epinephrine Summary may be a likely candidate for lipolysis activation. Training causes a decreased adipose tissue mass and The adrenal gland plays an important role in releas- adipocyte size, but the sympathoadrenergic system ing hormones that are crucial for carbohydrate, does not seem to be crucial for this adaptation. fat and protein metabolism, as well as for regu- Not only adipose tissue but also intramuscular fat lating organ blood distribution, thermoregulation, can be stimulated by epinephrine, and both lipopro- immunmodulation and skeletal muscle contractility tein lipase (LPL) and hormone sensitive lipase during and after exercise. Epinephrine secretion (HSL) play important roles in this regulation. HSL increases with the relative work intensity, and might be under control by both contractions and adrenal medulla hypertrophy can occur in response epinephrine (Donsmark 2002), and it has recently to prolonged intense training (‘sports adrenal been shown that activation of HSL and glycogen medulla’). Both direct motor center activity and phosphorylase occurs in parallel in adrenalectom- afferent neural feedback signaling from contracting ized individuals who receive an infusion with muscle play a role in the regulation of hormonal epinephrine during exercise (Kjær et al. 2000). This release from the adrenal gland during exercise. could indicate that the mobilization of intramuscu- Adrenal gland hormones (epinephrine and cortisol) lar triglyceride and glycogen occurs when simul- participate in the redundant regulation of hepatic taneously stimulated by adrenergic activity, and glucose release during physical activity.

References

Coker, R.H., Krishna, M.G., Brooks Lacy, Kjær, M. & Galbo, H. (1988) The effect of humans: effect of sensory nervous D., Bracy, D.P. & Wasserman, D.H. physical training on the capacity to blockade. American Journal of Physiology (1997) Role of hepatic α- and β- secrete epinephrine. Journal of Applied 257, E95–E101. adrenergic receptor stimulation on Physiology 64, 11–16. Kjær, M., Mikines, K.J., Linstow, M., hepatic glucose production during Kjær, M., Christensen, N.J., Sonne, B., Galbo, H. & Nicolaisen, T. (1992) Effect heavy exercise. American Journal of Richter, E.A. & Galbo, H. (1985) Effect of 5 weeks detraining on epinephrine Physiology 273, E831–E838. of exercise on epinephrine turnover in response to insulin induced Donsmark, M., Langfort, J., Ploug, T. trained and untrained male subjects. hypogycemia in athletes. Journal of et al. (2002) Hormone-sensitive lipase Journal of Applied Physiology 59, Applied Physiology 72(3), 1201–1204. (HSL) expression and regulation by 1061–1067. Kjær, M., Engfred, K., Fernandes, A., epinephrine and exercise in skeletal Kjær, M., Farrell, P.A., Christensen, N.J. & Secher, N.H. & Galbo, H. (1993) muscle. European Journal of Sports Science Galbo, H. (1986) Increased epinephrine Regulation of hepatic glucose 2, 1–10. response and inaccurate glucoregulation production during exercise in man: Galbo, H., Christensen, N.J. & Holst, J.J. in exercising athletes. Journal of Applied role of sympathoadrenergic activity. (1977) Catecholamines and pancreatic Physiology 61, 1693–1700. American Journal of Physiology 265, hormones during autonomic blockade Kjær, M., Secher, N.H., Bach, F.W. & E275–E283. in exercising man. Acta Physiologica Galbo, H. (1987) Role of motor center Kjær, M., Keiding, S., Engfred, K. et al. Scandinavica 101, 428–437. activity for hormonal changes and (1996a) Glucose homeostasis during Howlett, K., Lorentsen, J., Bergeron, R. substrate mobilization in exercising exercise in humans with a liver or et al. (1999) Effect of adrenaline on man. American Journal of Physiology 253, kidney transplant. American Journal of glucose kinetics during exercise in R687–R695. Physiology 268, E636–E644. adrenalectomized subjects. Journal of Kjær, M., Bangsbo, J., Lortie, G. & Kjær, M., Pollack, S.F., Mohr, T. et al. Physiology 519, 911–919. Galbo, H. (1988) Hormonal response to (1996b) Regulation of glucose turnover Jacob, C., Zouhal, H., Prioux, J. et al. (2004) exercise in man: influence of hypoxia and hormonal responses during Effect of the intensity of training and physical training. American Journal exercise: electrical induced cycling in on catecholamine responses to of Physiology 254(23), R197–R203. tetraplegic humans. American Journal of supraphysiological exercise in Kjær, M., Secher, N.H., Bach, F.W., Sheikh, Physiology 271, R191–R199. endurance-trained men. European S. & Galbo, H. (1989) Hormonal and Kjær, M., Secher, N.H., Bangsbo, J. et al. Journal of Applied Physiology 91, 35–40. metabolic responses to exercise in (1998) Hormonal and metabolic adrenal gland and exercise 199

responses to electrically induced cycling exercise in humans: role of muscle mass. despite enlarged adrenal medulla in during epidural anesthesia in humans. American Journal of Physiology 257, trained rats. American Journal of Journal of Applied Physiology 80, H1812–1818. Physiology 259, R998–R1003. 2156–2162. Searls, D.R., Victor, R.G. & Mark, A.L. Stallknecht, B. Enevoldsen, L., Kjær, M., Howlett, J., Langfort, T. et al. (1988) Plasma norepinephrine and Lorentsen, J. et al. (2001) Role of the (2000) Adrenaline and glycogenolysis in muscle sympathetic discharge during symphathoadrenergic system in adipose skeletal muscle during exercise: a study rhythmic exercise in humans. Journal of tissue metabolism during exercise in in adrenalectomised humans. Journal of Applied Physiology 65, 940–944. humans. Journal of Physiology 536, Physiology 528, 371–378. Sonne, B., Mikines, K.J., Richter, E.A., 283–294. Moates, J.M., Lacy, D.B., Cherrington, A.D., Christensen, N.J. & Galbo, H. (1985) Role Vissing, J., Iwamoto, G.A., Rybicki, K., Goldstein, R.E. & Wasserman, D.H. of liver nerves and adrenal medulla in Galbo, H. & Mitchell, J.H. (1989) (1988) The metabolic role of the exercise- glucose turnover of running rats. Journal Mobilization of glucoregulatory induced increment in epinephrine. of Applied Physiology 59, 1640–1646. hormones and glucose by hypothalamic American Journal of Physiology 255, Stallknecht, B. (2003) Influence of Physical locomotor centers. American Journal of E428–436. Training on Adipose Tissue Metabolism— Physiology 257, E722–E728. Richter, E.A. (1996) Glucose utilization. with Special Focus on Effects of Insulin and Vissing, J., Iwamoto, G.A., Fuchs, I.E., In: Handbook of Physiology (Rowell, L.B. Epinephrine. PhD thesis. University of Galbo, H. & Mitchell, J.H. (1994) Reflex & Shepherd, J.T., eds.). Bermedica, Copenhagen, Denmark. control of glucoregulatory exercise Columbia, MD: 912–951. Stallknecht, B., Bülow, J., Frandsen, E. responses by group III and IV muscle Savard, G., Richter, E.A., Strange, S. et al. & Galbo, H. (1990) Diminished afferents. American Journal of Physiology (1989) Noradrenaline spillover during epinephrine response to hypoglycemia 266, R824–R830. Chapter 16

The Adrenal Medulla: Proenkephalins and Exercise Stress

JILL A. BUSH AND N. TRAVIS TRIPLETT

The adrenal gland by the anterior pituitary release of adrenocortico- tropic hormone. The two endocrine adrenal glands are relatively small glands that are situated superior to the kidney. Adrenal medullary chromaffin system These glands are essential for life; removal of them would result in death. Diseases of the adrenal The adrenal medulla contains cells known as chro- glands occur but are generally treated with syn- maffin cells which constitute the majority of space thetic derivatives of the hormones these glands pro- within the medulla. The chromaffin system serves duce. The main function of the adrenal glands is as the method of storage and secretion for cate- to respond to stressed of the body by producing cholamines and proenkephalin peptide fragments, hormones that regulate blood glucose, increase including peptide F, E, and B (Lewis et al. 1979; heart rate and ventilation, and maintain the fluid Viveros et al. 1979). They resemble storage and homeostasis in the body. The adrenal gland has the secretory granules of other endocrine or exocrine highest rate of blood flow in the body per gram of glands found within the body. These chromaffin tissue, and is comprised of two main types of tissue; granules are highly specialized, electron-dense and the cortex and medulla (Kaplan 1988). The adrenal osmiophilic organelles smaller than mitochondria. medulla is the innermost part of the adrenal gland Communication between the sympathetic chromaf- and comprises only about 10% of the total gland tis- fin system and the body is rapid, where the neural sue. The adrenal medulla is a sympathetic ganglion response is instantaneous and the hormonal res- in which postganglionic fibers have ended in secret- ponse occurs within minutes or hours. Stimulation ory vesicles. Upon nervous system stimulation, via of chromaffin cells by sympathetic preganglionic preganglionic nerves reaching the adrenal medulla neurons of the splanchnic nerve (Coupland 1965, via the splanchnic nerve, hormones and neurohor- 1972) occurs in response to the neurotransmitter mones (i.e. epinephrine and peptide F) are secreted acetylcholine secreted from preganglionic neurons from chromaffin cells. Adrenal medullary hormones arising from the thoracolumbar region of the spinal are not essential for life but are produced and cord (Fig. 16.1). secreted during stressful-like events. The adrenal Besides enkephalins and catecholamines, some cortex, on the other hand, is essential for life. This other constituents of the chromaffin granules exist outer cortex layer secretes glucocorticoids, involved including chromogranin A, nucleotides and ascor- in carbohydrate and protein metabolism (i.e. cor- bic acid. Extra-adrenal chromaffin cells are located tisol); mineralcorticoids, involved in maintenance adjacent to the aorta as paraganglia, in carotid of extracellular fluid volume (i.e. aldosterone); and bodies, in viscera, and in sympathetic ganglia. To minor-effecting sex hormones involved in repro- date, there is only indirect evidence for selective dis- duction. The adrenal cortex is stimulated primarily charge of granular contents upon neural stimulation 200 the adrenal medulla 201

Adrenal Cortex gland 80% Splanchnic nerve Medulla 20%

Kidney

Prevertebral Chromaffin cells Spinal chain ganglia Fig. 16.1 Neural innervation and cord Ach chromaffin granules of the adrenal Epi gland. Ach, acetylcholine; Epi, Epinephrine PF PF epinephrine; PF, peptide F.

of the adrenal medulla. However, exocytosis of sary constituent of the chromaffin granule (Fig. 16.2). the chromaffin granule is far from simple, and this Via hydrolysis of ATP on the exterior membrane process could provide a potential mechanism for surface, protons (H+) are injected into the granule, the differential release pattern with exercise stress. providing a more acidic, positively charged envir- Unfortunately, there is no data available to support onment versus the cytosol. An electrochemical H+ exocytosis and hence the differential release pattern gradient is created in the interior of the granule. Due of adrenal medullary neurohormone (i.e. peptide F to the biosynthetic pathways of epinephrine and and epinephrine) upon stimulation. Stimulation of dopamine, the energy supplied by the electrochem- nicotinic and muscarinic receptors is involved in ical proton gradient is necessary for the uptake acetylcholine-related exocytosis of the chromaffin of these two substances into the cell. The enzyme cells. Several endogenous substances (e.g. substance necessary to breakdown norepinephrine into epine- P, neurotensin, vasoactive intestinal peptide) within phrine is located in the cytosol of the cell. Thus nore- the adrenal medulla may competitively bind to pinephrine must be exported from the cell, exposed nicotinic receptors, thus inhibiting the release of to the enzyme, and epinephrine recaptured for epinephrine (Livett, B.G. 1984). High concentrations storage inside the granule. Other factors may also be of nicotine demonstrated a proportional release of involved in the exocytotic process of the granules. enkephalin-containing peptides and total catechola- A Ca2+ influx is a primary elicitor of exocytosis. mine concentration (Wilson et al. 1982; Livett, B.G. Stimulation via acetylcholine or high extracellular 1984). If exocytosis of chromaffin cells containing K+ increases the intracellular Ca2+ concentration of proenkephalin fragment peptide F and epinephrine the chromaffin cells. As the granules move toward can be altered by such endogenously produced sub- the plasma membrane, it may encounter the cyto- stances, then this would provide a plausible explana- skeletal structure actin. The increase in Ca2+ may tion for a differential release pattern of peptide F decrease the viscosity of this interaction permitting and epinephrine. easier granular movement. Calmodulin selectively Chromaffin granules contain membrane-bound phosphorylates membrane proteins, allowing for proteins and glycoproteins. Such soluble factors the fusion of plasma and granular membranes. include a proton pump adenosine triphosphatase Synexin, in the presence of Ca2+, can polymerize (ATPase), catecholamine and nucleotide carrier into large aggregates which bind to phospholipids proteins, cytochrome b561, actin, glycoproteins, and within the membrane. Also, recycling of the granu- dopamine β-hydroxylase pump ATPase is a neces- lar membrane to and from the plasma membrane is 202 chapter 16

Cell membrane

Ca2+ Ca2+ channel

CaM Ca2+/CaM Ca2+ + inactive active

CaM Ach Ach kinase II rec stimulation

Fig. 16.2 Signal transduction Catecholamine Phosphorylate several proteins pathway in acetylcholine-stimulated synthesis and Chromogranin A adrenal medullary cells. Ach, secretion Tyrosine hydroxylase acetylcholine; CaM, calmodulin. Chromaffin granule membrane (Modified from Yanagihara et al. Others 1996.)

continuously occurring. All of these aforementioned mental aspects of the proenkephalin polypeptide processes are important in the role of exocytosis precursor and the processing of this peptide precur- of the chromaffin granules containing epinephrine sor to peptide F in order to appreciate the potential and peptide F in the differential release of either physiological roles of peptide F. The preproen- neurohormone. kephalin polypeptide (30 kDa) is the precursor of The chromaffin cells also secrete enkephalins, the enkephalins ([Met]- and [Leu]-enkephalin) and a group of substances that possess morphine-like ECPs and is found in various regions of the body: analgesic activity and are also produced by the brain brain, adrenal medulla and activated T-lymphocytes, (Undenfriend & Kilpatrick 1984). The enkephalins and hemoglobin molecules (Hughes et al. 1975; and some larger peptide fragments which contain Kimura et al. 1980; Brantl et al. 1986; Zurawski et al. enkephalin sequences, known as the enkephalin- 1986; Martin et al. 1987; Ivanov et al. 1997; Zhao et al. containing peptides (ECPs), are derived from the 1997). large molecular weight preproenkephalin. These Peptide F, a 3.8 kDa ECP, is post-translationally ECPs include peptides in the 3–5 kDa range, which processed from preproenkephalin (Fig. 16.3) in a would include peptides E, F and B (Undenfriend & series of cleavages by trypsin (amino end) and car- Kilpatrick 1984; Hiddinga et al. 1990). Studies with boxypeptidase B (carboxyl end) which cleave at the these peptides have found parts of their amino acid enkephalin sequences (Undenfriend & Kilpatrick sequences to be highly conserved among different 1984). Peptide F, the more common peptide product species (Lewis & Stern 1983) and are found in the of proenkephalin precursor measured in circulation blood in physiologically relevant concentrations (Kilpatrick et al. 1980, 1981; Wasserman et al. 1986), with long half-lives (≥ 15 min), indicating their bio- consists of 34 amino acids (amino acid sequence 107– logical significance (Katzenstein et al. 1987). How- 140) and contains two [Met]-enkephalin sequences ever, many of the functions of these substances are (Lewis 1982; Lewis & Stern 1983), expressing struc- as yet unknown. tural similarities to classical opiates (Hansen & Morgan 1984). Opioid peptides typically play an Proenkephalin peptide F essential role in the brain as neurotransmitters where [Met]- and [Leu]-enkephalin are the most It is important to understand some of the funda- prominent end-products. the adrenal medulla 203

Preproenkephalin: 29.8 kDa

Signal peptide Peptide I: Peptide B: 18.2 kDa 4.9 kDa 2.6 kDa

Peptide E: 12.6 kDa 5.3 kDa 3.2 kDa

Peptide F: 8.6 kDa 3.8 kDa

Fig. 16.3 Post-translational processing of enkephalin-containing peptides from preproenkephalin. (Modified from Lewis & Stern 1983, Met-enkephalin Met-enk–Arg–Phe Undenfriend & Kilpatrick 1984 and Leu-enkephalin Met-enk–Arg–Gly–Leu Katzenstein et al. 1987.)

Interestingly, a breakthrough in the discovery of circulating period (i.e. 15–60 min) than the smaller the preproenkephalin precursor molecule became [Met]- and [Leu]-enkephalins (i.e. 1–2 min) (Kilpatrick apparent when Kimura et al. (1980) reported infor- et al. 1980, 1981; Kimura et al. 1980; Boarder & mation pertaining to possible precursors for the McArdle 1986). enkephalins, [Met]- and [Leu]-enkephalin, secreted Transportation of 3–8 kDa peptides through the from the adrenal medulla. With the discovery of circulation could be enhanced by the non-enkephalin few fragments in the brain, the determination of regions within the peptide structure (Boarder & the structure of the precursor became difficult. McArdle 1986), thus adding to the prolonged plasma ECPs derived from the biosynthetic processing of half-life of these peptide fragments. The roles of proenkephalin within the central nervous system the [Met]- and [Leu]-enkephalin in the periphery cannot fully account for the total amount of these became speculative at best due to a high potential peptides found in the circulatory pool (Lewis & for protease degradation, within 1–2 min (Kraemer Stern 1983). However, a higher concentration of et al. 1987). Thus, the larger structure of peptide F [Met]- and [Leu]-enkephalin is derived through with a longer prolonged circulating half-life (15– central nervous system processing (Lewis 1982; 60 min) provided the potential for physiological Lewis & Stern 1983). Discovery of more proenke- roles in the peripheral circulation (e.g. communica- phalin fragments in the adrenal medulla gave rise tion between different biocompartments). to better biochemical determination of sequences of the preproenkephalin precursor and to potential Concomittant release of peptide F and physiological stress responses. It was also thought epinephrine from adrenal medulla at the time that the more labile enkephalins were transported via the higher molecular weight frag- Peptide F is co-stored with epinephrine within chro- ments (e.g. to peptide F) since ECPs have a greater maffin cells of the adrenal medulla (Viveros et al. 204 chapter 16

1979; Lewis & Stern 1983). However, the biochem- epinephrine as well, as a close relationship exists ical and molar-equivalent relationship of epineph- between these neurohormones. The enzymatic pro- rine and peptide F within these chromaffin cells is cessing of newly synthesized preproenkephalin less understood. The relationship of the co-storage of may be affected by epinephrine and thus may affect peptide F and epinephrine may provide a potential the co-storage of peptide F (Wilson 1991). Wilson mechanism for differential release of these neuro- and colleagues (Wilson et al. 1982; Wilson 1991) hormones from the adrenal gland in response to found that pharmacological agents and neuro- stress. It is known that trypsin and carboxypepti- transmitters activated proenkephalin synthesis in dase β digestion of preproenkephalin within the the chromaffin cells. They found that inhibitors of adrenal medulla occurs within 2 h of proenkephalin vesicular catecholamine uptake (e.g. tetrabenazine precursor synthesis and that the peptide F fragment and resperine) enhanced the processing of proenke- is found in the chromaffin granules within 2 h of the phalin and thus the content of ECPs in the chro- precursor synthesis (Wilson 1991). Tetrabenazine, maffin cells. This data suggested that epinephrine which has been shown to inhibit catecholamine or possibly other constituents of the chromaffin cells uptake into the storage vesicles, did not inhibit (e.g. adenosine triphosphate [ATP]) may inhibit uptake of proenkephalin fragments into vesicles. In proenkephalin processing. Thus, this mechanism of fact, it enhanced the peptide’s assimilation (Wilson action can alter the molar concentration of peptide F 1991). This research would suggest that, not only is and epinephrine within different chromaffin cells. peptide F a final processing product assimilated It is important to understand the basic processing into the chromaffin vesicles, but also that termina- of epinephrine. Briefly, amines containing a 3,4- tion of the process may depend upon the co-storage dihyroxyphenyl nucleus are referred to as catecho- of epinephrine (Wilson 1991). lamines and are derived from the amino acid Studies have shown that epinephrine and enke- tyrosine. Tyrosine can also be derived from the phalin-containing fragments are co-released from conversion of phenylalanine in the liver via pheny- cultured adrenal chromaffin granules (Schultzberg lalanine hydroxylase. Tyrosine and phenylalanine et al. 1978; Livett, A.R. et al. 1981). This co-secretion, can be found in the diet of foods high in protein. under similar physiological stimulation, also under- Epinephrine is mainly derived from the adrenal scores the importance of a biological role for peptide medulla; however, the necessary enzymes involved F during stress. It is also possible that peptide F and in biosynthesis are also present within neurons of epinephrine can have a complementary action on the central nervous system in minute amounts. various biological target tissues. For example, both Norepinephrine is primarily found within the cent- epinephrine (McCarthy & Dale 1988) and peptide ral nervous system as a neurotransmitter for sym- F (Hiddinga et al. 1994) have modulating effects pathetic neurons, either inhibitory or excitatory. on the immune system. This may be similar to a Approximately 80% of total stimulated adrenal model used by other neurohormones and peptides medullary catecholamine secretion is epinephrine, which are co-stored and co-secreted and possess while the remaining 20% is norepinephrine. dissimilar/similar biological activities (e.g. insulin The initial step in epinephrine synthesis involves and serotonin within pancreatic β-cells in humans conversion of tyrosine to 3,4-dihydroxyphenylala- [Richter et al. 1986] and neuropeptide Y and nor- nine (dopa) via the enzyme tyrosine hydroxylase, epinephrine within bovine sympathetic nerves which is phosphorylated and activated when stimu- [Bastiaensen et al. 1988; De Potter et al. 1988]). lated by acetylcholine (Fig. 16.4). Dopa, in the pres- Epinephrine is the primary stress hormone in the ence of an aromatic l-amino acid decarboxylase, is body. Its biosynthesis and release may influence decarboxylated to form dopamine. The two afore- peptide F release and appearance in the blood mentioned enzymes are located in the cytosol of (Kraemer et al. 1985b, 1991). While it is important to the medulla where such enzymatic reactions occur. understand the processing and storage of peptide Dopamine enters the granule and is hydroxylated F, it is important to understand the processing of by dopamine β-hydroxylase from the granular mem- the adrenal medulla 205

Vascular β β β β Liver and - ( 1, 2, 3) adrenergic receptors. Epinephrine Diet compartment β conversion has a high affinity for 2-receptors which are located Phenylalanine extrajunctionally on non-innervated target cells. Phenylalanine Tyrosine hydroxylase Such receptors mediate lactate production and GI tract vasodilation in skeletal muscle. The role of epine- Tyrosine phrine release during acute exercise stress may Tyrosine hydroxylase obviate the role played by peptide F, as the pattern of response between the two neurohormones some- Dopa Aromatic L-amino acid times differs (Kraemer et al. 1985b, 1991). This is decarboxylase evident when data are expressed relative to the (cytoplasm) molar ratio of peptide F to epinephrine. A decrease Dopamine in the molar ratio (usually observed during exer- Dopamine β-hydroxylase (chromaffin granules) cise) would indicate the predominance of epine- Norepinephrine phrine secretion into the circulation, whereas as Phenylenthanolamine increase in the molar ratio (usually observed during N-methyl transferase (cytoplasm) recovery) would indicate the predominance of pep- Epinephrine tide F secretion. However, it cannot be discounted that different chromaffin cells contain varying con- Fig. 16.4 Biosynthetic pathway of adrenal medulla- centrations of each neurohormone (Wilson et al. derived epinephrine. (Modified from Genuth 1988.) 1982; Livett, B.G. 1984), or that chromaffin cells are selectively released upon stimulation. Epinephrine brane to form norepinephrine. Norepinephrine has been shown to play a greater role during exer- returns to the cytosol for methylation into epineph- cise (Kjær et al. 1985; Brooks et al. 1988; Kjær & rine via phenylethanolamine N-methyl transferase Galbo 1988), while peptide F may play a greater role (PNMT), using S-adenosyl-l-methionine as a donor. during recovery from exercise (Kraemer et al. 1985b, Epinephrine is reincorporated into the granule for 1991). The difference in release patterns may explain storage and secretory preparation. This process is important biological functions. The possibility ex- dependent on the energy from the proton pump ists that, even though epinephrine and peptide F are ATPase. Catecholamines within the granule are co-stored and co-secreted by similar stimuli, they bound to ATP and chromogranin A, preventing may have different physiological roles within the egress of the stored hormones. same biocompartment. In the periphery (i.e. adrenal gland), epinephrine, norepinephrine and dopamine serve as neurohor- Physiological role of peptide F mones, while in the central nervous system, they can serve as neurotransmitters. The plasma half-life Although peptide F contains two [Met]-enkephalin of epinephrine is about 1–2 min. The plasma concen- sequences, it responds only weakly to classical tration of epinephrine at rest is ~ 0.05 ng⋅mL–1 and opiate action tests (Lewis & Stern 1983). Therefore, can elevate from 0.27 ng⋅mL–1 to 4.1 ng⋅mL–1 during the [Met]-enkephalin sequence contained in peptide exercise. Epinephrine affects the state of arousal, the F is not necessarily indicative of enkephalin-like ‘fight or flight’ response, and the contractility of functions. The non-enkephalin segments of the pep- heart and skeletal muscle. Epinephrine elevates tide F molecule have an important role in their func- heart rate, increases blood flow to skeletal muscle, tion, as they may be the primary binding sequences increases the metabolic rate, and increases substrate for the molecule. Figure 16.5 outlines potential utilization during exercise via release of glucose and physiological roles for peptide F that will be dis- free fatty acids into the blood. Epinephrine binds cussed in the following section. to receptors on the cell surface of the plasma mem- To date, one of the most important evidences α α α brane of target cells, interacting with both - ( 1, 2) supporting the study of the response pattern of 206 chapter 16

Secretion of epinephrine Splanchnic nerve and peptide F into blood α β receptors stimulation following exocytosis of chromaffin cells

Epinephrine Metabolic effects on tissues in body

Adrenal medulla

HR Contractility Peptide F Glycogenolysis Gluc utilization Chromaffin cells containing various molar concentrations Circulate 15 min Potential interaction of epinephrine in plasma with erythrocyte and peptide F opiate-like receptors

Potential interaction Cleaved into peptide F? with immune system RBC Epinephrine only

Peptide F only Secrete proenkephalin Combination Enhance immune function RBC

Secrete Hb Immune opiate-like cells fragments Autocrine feedback to potential opiate-like receptors

Fig. 16.5 Potential interaction of peptide F within the three biocompartments of blood. HR, heart rate; RBC, red blood cell. (From Bush et al. under review.) peptide F is the interaction between peptide F and ever, the immune system may not be the only target the immune system (Hiddinga et al. 1994; Triplett- site for this proenkephalin peptide F fragment. McBride et al. 1998). In addition, it is possible that Other target tissues and sites may exist as this is an these interactions can take place in the blood bio- evolving form of study. compartments. Peptide F has been shown to have Hiddinga et al. (1994) reported a possible role for a relationship with T- and B-cells in vitro (Hiddinga peptide F as a regulatory neurohormone within the et al. 1994) and to be related to B-cell activation and immune system. Unlike the suppressive effects of individual fitness level (Triplett-McBride et al. 1998). [Met]-enkephalin on the immune system (Johnson The longer half-life of peptide F prolongs its bio- et al. 1982; Marotti et al. 1993), peptide F seems to logical viability to interact with the different blood enhance immune function (Hiddinga et al. 1994). In biocompartments and the immune system. How- vitro research using purified peptide F at physiolo- the adrenal medulla 207

gical concentrations demonstrated that, following sion of the antibody-forming response of B-cells, 15 min of peptide F incubation with T-cells, there and this action was blocked by the addition of was an increase in antibody-forming B-cells in cul- naloxone. This suggested that this low molecular ture, significantly enhancing the antigen-specific weight enkephalin was bound to opiate receptors antibody-forming cell response of lymphocytes to found on lymphocytes (Sibinga & Goldstein 1988). an antigenic challenge of trinitrophenyl-Ficoll (TNP- To test if concentration of either molecule was an Ficoll) (51 ng⋅mL–1). Studies in athymic nude mice important factor in modulating immune function, (e.g. mice expressing delayed and defective T- peptide F and [Met]-enkephalin were added simul- cell development) demonstrated that peptide F taneously to cultures in an equimolar concentration was directly involved in an alteration of the cell- (10 nmol) in the presence of naloxone. [Met]- mediated immune response (i.e. T-cell activation) enkephalin again induced suppression of the anti- rather than directly involved in the humoral-medi- body-forming response of B-cells, and this action ated immune response (i.e. antibody production was blocked by naloxone. Peptide F, on the other from B-cells). This data has shown that lymphocytes hand, induced an increase in antibody production, could be a possible target site for peptide F, and that and this increase was not blocked by naloxone. This subsequent interaction causes an enhancement in again suggested the possibility of different mech- immune function. anisms of receptor binding on the same immune cells. It may be that naloxone inhibited the immuno- suppressive effects of [Met]-enkephalin, thereby Methods of identifying the peptide F receptor potentiating the immunoenhancing effects of pep- Competitive binding with naloxone. Hiddinga et al. tide F. These data also implied that even though the (1994) also attempted to indirectly identify a pos- peptide F molecule contained [Met]-enkephalin sible receptor mechanism for peptide F through sequences, peptide F may not have interacted with naloxone competition. Naloxone functions as an opioid receptors found on lymphocytes. It may be antagonist for opiate receptor binding (e.g. [Met]- that the non-enkephalin (peptide) sequences of pep- enkephalin) (Simonds 1988). Murine splenocytes tide F play a greater role in receptor binding (Roth were either treated with peptide F or [Met]- et al. 1989). This line of research has proven to be enkephalin to determine which molecule had the very beneficial in determining and locating a pos- greatest effect on the antibody-forming cell response sible receptor and/or target site for peptide F, which of B-cells. Naloxone was added at final concentra- to date are not fully defined. Such data and the rela- tions of 0.1, 1.0 and 10.0 µmol (i.e. 10–1000 times tionship of immune cells and peptide F supported greater than the concentration of either plasma the need for research since peptide F was found in peptide F or plasma [Met]-enkephalin). At a 10 nmol quantifiable concentrations in the white blood cell concentration, purified peptide F and [Met]- biocompartment of the blood. enkephalin, respectively, yielded enhancement and suppression on the antibody-forming response of Immunochemical approach via flow cytometry. Since B-cells. To activate the lymphocytes, antigens (1% these studies in the 1980s, Bush et al. (under review) sheep red blood cells [SRBC] or TNP-Ficoll) were have performed other methodological techniques added to the culture at both an optimal (51 ng⋅mL–1) to determine the receptor for peptide F on immune and suboptimal (5 ng⋅mL–1) concentration follow- cells. It was reasoned that an immune cell respons- ing 15 min of incubation with naloxone (10 µmol) ive to peptide F would bear peptide F receptors. and/or peptide F (10 nmol). Naloxone was unable An immunochemical approach was used to concur- to block the antibody-forming response of B-cells rently identify cells displaying a peptide F receptor elicited by peptide F. This indicated that peptide F and to identify the leukocyte subtype. Monoclonal may interact with a yet-to-be-determined receptor antibodies to be used for identification of subclasses other than an opioid receptor (i.e. a peptide receptor) of human leukocytes were commercially available. on lymphocytes. [Met]-enkephalin-induced suppres- No antibodies were available to identify peptide F 208 chapter 16

Peptide F receptors Peptide F molecule

FITC Immune cell

Primary antibody Secondary antibody

Fig. 16.6 Indirect immunophenotyping technique for detection of a peptide F receptor. Primary antibody = rabbit anti-peptide F. Secondary antibody = goat anti-rabbit and fluorescein isothiocyanate (FITC) labeled. (From Bush et al. under review.)

receptors. In fact, peptide F receptors have not been with goat serum, human serum and calf serum as Fc- characterized. Therefore, an indirect approach and blocking agents. We were unable to reduce the bind- flow cytometry was employed (Fig. 16.6). ing of normal rabbit serum to levels acceptable to An antibody to the ligand (i.e. peptide F) was determine specific binding to peptide F (Bush et al. used to determine a receptor identified by ligand under review). Further research on determining the binding. A series of experiments including numer- receptor for peptide F needs to be done. ous controls to optimize the system and to verify specificity were performed. Such experiments in- Effect of exercise on adrenal medullary cluded variation of the dose of peptide F and anti- proenkephalins peptide F antibody; utilization of several blocking buffers to eliminate background staining; and test- Peptide F has been shown to elevate during exercise ing in the absence of exogenous peptide F, primary (Kraemer et al. 1985a, 1985b, 1990a, 1991; Bush et al. antibody, and secondary antibody. Normal rabbit 1998, under review; Triplett-McBride et al. 1998). serum (several different animals) was used as a neg- Three of the studies by Kraemer et al. (1985a, 1985b, ative control. Dual labeling techniques were used to 1992) compared peptide F responses in endurance- try to localize the labeling to a subclass of leukocytes. trained and untrained men. The pattern of release Using the indirect approach (Fig. 16.6), it was of peptide F in untrained men in response to found that between 20–30% of the white blood cycle ergometer exercise (8-min stages) at increasing cells stained positively for peptide F (Bush et al. intensities to Vo2peak (Kraemer et al. 1985a, 1985b) under review). The secondary antibody showed was marked by increases in peptide F concentra- essentially no binding in the absence of the primary tions with increasing exercise intensity, with a surge antibody, suggesting that antibody binding was approximately 5 min after the cessation of exercise. specific. Addition of peptide F to the cells did not Concentrations of peptide F were reduced, though greatly change the percentage of cells that bound not back to resting levels, by 15 min after exercise. the antibody, suggesting near endogenous satura- With trained men, peptide F elevated at approx- o tion of receptors. However, normal rabbit serum imately 50% V 2peak, dropped as exercise intensity o used in place of the primary antibody also produced approached 100% V 2peak, and then surged again at similar results. It was unlikely that binding to Fc- 5 min after the cessation of exercise. By 15 min after receptors occurred because cells were preincubated the cessation of exercise peptide F concentrations the adrenal medulla 209

had dropped, though not to resting concentrations urements. At acute altitude, the only significant dif- (Kraemer et al. 1985a, 1985b). The Triplett-McBride ference from pre-exercise peptide F concentrations et al. (1998) investigation was performed in women occurred at the mid-exercise measurement in the and the results were in contrast to prior studies in caffeine ingestion trial. For the chronic altitude con- men. For example, although the exercise mode was dition, the only significant increase in peptide F the same (cycle ergometer), only the trained group concentrations occurred with no caffeine ingestion showed any peptide F response to the exercise, with at the mid-exercise measurement. The authors attri- a peak at the highest exercise intensity of 80%. The buted most of these differences to the time of alti- untrained group showed no exercise response in tude exposure (17 days). peptide F. A study (Kraemer et al. 1991) examined peptide F Peptide F is also not under conscious control and catecholamine responses to exhaustive exercise as demonstrated by Kraemer et al. (1992) in an in- on a computerized cycle ergometer at various per- vestigation utilizing hypnosis and the responses centages (36%, 55%, 73% and 100%) of maximal of peptide F to exercise. This study also compared leg power. The focus of the study was to compare trained to untrained men performing cycle ergome- patterns of response of peptide F, epinephrine, o ter exercise at 25% and 50% V 2peak. The hypnosis lactate and norepinephrine. There was a significant condition involved the suggestion of performing increase in peptide F concentration immediately o cycle ergometer exercise at 50% and 75% V 2peak post-exercise at the 36% exercise intensity level, while subjects were actually cycling at 25% and 50% which was the longest in duration (3.5 min). There o V 2peak. The investigators found no significant dif- was also a significant increase in epinephrine imme- ferences in peptide F concentrations between condi- diate post-exercise at both the 36% and 55% exercise tions (control versus hypnosis) or between rest and intensity levels and at 15 min following the 100% exercise, suggesting that the actual intensity of exer- exercise intensity level. The results indicated that cise was still too low to elicit a peptide F response, these peptide F and epinephrine frequently have despite the suggestion of higher intensity. Other an inverse relationship at the higher intensity levels. studies by Kraemer et al. (1987, 1988, 1990a, 1991) There were also significant increases in norepine- examined peptide F responses to exercise in healthy phrine immediately after each exercise intensity, men, and demonstrated increases with exercise of 5 min after the low to moderate exercise intensit- long enough duration and high enough intensity. ies, and 15 min at the lowest exercise intensity level, Two studies (Kraemer et al. 1988, 1991) also util- suggesting a differential exercise response pattern ized cycle ergometer exercise, although varying than the encephalin. Similar exercise response exercise protocols were employed. The 1988 study patterns were observed with whole blood lactate examined peptide F responses to steady state following all the exercise intensity levels. Although o (80–85% V 2peak) exercise to exhaustion with and these relationships were not significant, the authors without caffeine and at varying altitudes and alti- point out that these relationships do not support the tude exposures (sea level, acute altitude and chronic theory of co-secretion of these substances from the altitude). The investigators found that peptide F chromaffin cells of the adrenal medulla as differen- concentrations were higher at sea level (mid- and tial response patterns to exercise may exist. post-exercise) and lower at chronic altitude (post- Another study (Kraemer et al. 1987) examined o exercise) after caffeine ingestion. The results also peptide F responses to steady state (70% V 2max) indicated that post-exercise peptide F concentra- treadmill exercise of long duration (100 min) before, tions were lower than sea level concentrations at during and after heat acclimation. The investigators both acute and chronic altitude after caffeine inges- found that higher concentrations of peptide F in the tion. The exercise responses of peptide F were heat were due to a reduction in degradation to [Met]- mixed. At sea level, only the caffeine ingestion trial enkephalin. The investigators also found no sig- produced significant differences from pre-exercise nificant differences pre- to post-exercise or between concentrations for the mid- and post-exercise meas- test days (before, during and after acclimation), and 210 chapter 16

proposed that these findings may be a result of protocol (control). A high intensity training protocol degradation processes in the longer duration exer- was performed for 2 weeks (i.e. 10 repetitions of cise. In another investigation, Kraemer et al. (1990a) 1-repetition maximum [1-RM] squat exercise every found that healthy men who performed treadmill day for 2 weeks) and a low intensity training proto- exercise of increasing intensity (7-min stages) exhib- col (i.e. 50% 1-RM squats 1 day per week for 2 weeks ited a peptide F response similar to that of the and 1-RM testing 1 day per week for 2 weeks). Acute untrained men in the 1985 investigations (Kraemer testing was administered at the beginning, middle et al. 1985a, 1985b). The main difference between and end of the 2-week training period. Subjects these two groups of men was that the peak in performed continuous repetitions at 70% 1-RM until peptide F concentration was at the maximal exer- exhaustion. No change in plasma peptide F concen- cise intensity for the healthy men, not 5 min into tration was observed in response to acute exercise recovery as was found in the 1985 investigations stress and/or the overtraining protocol. The over- using untrained men. This may be explained by training protocol may have produced adaptations the difference in fitness level and in exercise mode; in the adrenal medullary chromaffin system. The cycling (non-weight-bearing) versus treadmill lack of peptide F response to this type of training (weight-bearing). stimulus may have been due to inhibitory factors The modality of exercise (aerobic versus anaero- negatively affecting the proenkephalin biosynthesis bic) utilized can change the response of peptide F. into the chromaffin cells (Wilson et al. 1982; Livett, Kraemer et al. (1992) studied the response of plasma B.G. 1984; Wilson 1991) or exocytotic suppression of peptide F to a high intensity cycle ergometer test in existing peptide F from the chromaffin cells. 10 healthy, active men. Four different intensities Reductions in the immune function with over- were utilized: 100% maximal leg power (equivalent training have been observed (Mackinnon 1992). o to 318% V 2max for 6 s), 73% maximal leg power Such types of high intense training had negative o (equivalent to 230% V 2max for 15 s), 55% maximal effects on health and the immune system, since o leg power (equivalent to 175% V 2max for 45 s) and hormones such as cortisol and epinephrine, exhib- o 36% maximal leg power (equivalent to 115% V 2max iting immunosuppressive effects, were elevated for 180 s). This study showed the differential pattern during this type of training (Kuipers & Keizer 1988; of epinephrine and peptide F in response to the Fry et al. 1991, 1994). The lack of change in plasma exercise stress (i.e. epinephrine increased while peptide F concentration in response to overtraining peptide F decreased). With the longer duration of may have negatively impacted the immune system; o exercise, peptide F was elevated (i.e. 115% V 2max and under such conditions plasma peptide F would for 180 s) during exercise and returned to the resting not be available to reverse any negative effects on level by 5–15 min post-exercise. This study gave immune function produced by either cortisol or initial insight into the difference in peptide F con- epinephrine (Kuipers & Keizer 1988; Fry et al. 1991, centration in response to an anaerobic-type exer- 1994). cise, such as high-intense short-duration cycling or A second study (Bush et al. 1998) utilized a 16-set resistance exercise. Such data demonstrated that the resistance exercise protocol. Four sets each of 10-RM duration and/or volume of exercise may influence and 15-RM bench press, bent over row, military the concentration of peptide F in the plasma. press and squat exercise were performed by resist- Resistance exercise is anaerobic in nature. Two ance trained men. The 10- and 15-RM protocols both studies have examined the responses of peptide F produced a decrease in plasma peptide F concen- concentrations to resistance exercise (Bush et al. tration immediately following the acute resistance 1998; Fry et al. 1998). Fry et al. (1998) examined the exercise stress. At 15 min post-exercise, an increase peptide F response to an overtraining protocol. in plasma peptide F was observed following only Subjects were tested following a high intensity the 10-RM protocol. The higher forces involved in resistance exercise training protocol (overtrained) performing the 10-RM protocol may account for the versus a low intensity resistance exercise training differential response in plasma peptide F. An 80% the adrenal medulla 211

increase in peptide F concentration was observed volume and/or intensity resulting in a decrease in 4 h post-exercise following both protocols. This physical performance (Fry et al. 1991). An over- study suggested that circulating peptide F may play training study done by Fry et al. (1998) examined a role during recovery periods from moderate resist- resistance-trained adult men in which they per- ance exercise to potentially increase immune func- formed 100% 1-RM barbell squats for a 2-week tion during repair of disrupted muscle (Fridèn et al. period, resulting in a decrease in their 1-RM strength 1983; McCully & Faulkner 1985; Round et al. 1987; in the overtrained state. Plasma testosterone and Fielding et al. 1991; Nieman et al. 1995; McBride et al. cortisol were decreased while in the overtrained 1998). The increase in peptide F 4 h after resistance state. Growth hormone, however, was not influ- exercise was not observed in the control group. This enced by the high intensity resistance overtraining finding was quite surprising. It may be speculated protocol. Epinephrine is elevated in response to that there was an increase in production and/or high intensity resistance overtraining protocols (Fry secretion of peptide F from the adrenal medulla, or et al. 1994) with no effect of overtraining response to that peptide F bound to receptors on target tissues circulating levels of peptide F (Fry et al. 1998). This were released. indicates the high intensity resistance overtraining Studies during heat acclimation in adults have protocol appeared to overwhelm the capacity of shown a reduction in plasma [Met]-enkephalin the adrenal medullary chromaffin cells to secrete (Kraemer et al. 1987), potentially due to the increased peptide F. degradation of the peptide in the circulatory system of a decrease in the processing from the large pre- Effect of exercise on peptide F in other cursor unit. Circulating encephalin and opiates blood biocompartments released during stress (Viveros & Wilson 1983) and exercise (Howlett et al. 1984; Farrell et al. 1987) could It has been reported that upon activation of lym- mediate the effects of stress reduction. There appears phocytes, a secretagogue was the proenkephalin to be a significant difference in the enzyme concen- precursor (Roth et al. 1989). Lymphocytes differ in trations responsible for hydrolyzing enkephalins their origination and maturation. The lymphocytes such that the enkephalin concentrations were higher are a subclass of leukocytes, consisting of T- and B- in a trained individual (Jaskowski et al. 1989). cells, ranging in size from 6–10 µm and comprising Exercise studies performed to exhaustion indicate 20–25% of circulating leukocytes. Their functions an increase in plasma levels of [Met]-enkephalin are crucial during an immune response and include and that this is observed in an exercise-intensity recognition of antigens, production of antibodies dependent manner (Sommers et al. 1990). It is against invading antigens, production of lympho- known that [Met]-enkephalin is co-released with kines, action of cytotoxicity and memorization of epinephrine from the adrenal gland (Viveros et al. previous antigen encounters. T-lymphocytes are 1979; Wilson et al. 1982; Wilson 1991). In an exer- generated within the bone marrow and mature in o cise study of moderate (70% V 2max) to exhaustive the thymus, while B-lymphocytes are generated and o exercise (120% V 2max) focusing on the co-release mature within the bone marrow and fetal liver. patterns of epinephrine and [Met]-enkephalin, peak Proliferation of mature lymphocytes occurs while plasma epinephrine levels were observed 1 min they are surveying the system in the secondary post-exercise, whereas peak plasma [Met]-enkepha- lymphoid organs. lin levels were observed during the moderate exer- Approximately 1–2% of total lymphocytes recir- cise intensity, declined during the high exercise culate hourly, providing frequent communication intensity and returned to baseline within 1 min post- with any foreign cells present in the system. During exercise (Boone et al. 1992). Similar to the peptide F this time of circulation, proenkephalin fragments fragment, a differential exercise response pattern (e.g. peptide F) may interact with these immune exists for epinephrine and [Met]-enkephalin. cells. T-lymphocytes are categorized by their func- Overtraining is defined as an increase in training tional ability: T-helper and T-cytotoxic cells. T- and 212 chapter 16

0.12 Plasma WBC )

–1 0.10 A

0.08 A Fig. 16.7 Peptide F concentrations 0.06 A in plasma and white blood cell biocompartments at rest, mid- 0.04 exercise and for 60 min into recovery AA following 30 min of cycling at 80%

Peptide F (pmol·mL 0.02 o = ≤ V 2max. A p 0.05 difference 0 from corresponding pre-exercise, Pre- Mid- 051530 60 mean ± SE. WBC, white blood cell exercise exercise Post-exercise time (minutes) compartment.

B-lymphocytes are further differentiated by their pilot study was the presence of peptide F observed function during humoral-mediated or cellular- within the white blood cell biocompartment both mediated immunity. Cell-mediated immunity deals during rest and following exercise. Subsequently, with cell-to-cell interactions and involves primarily there was a decrease (p ≤ 0.05) in white blood cell the attacking of foreign substances or antigens. biocompartment peptide F concentration at 30 min Humoral-mediated immunity functions to supply post-exercise (Fig. 16.7). It might be speculated the system with antibodies (i.e. immunoglobulins) that during early recovery periods from exercise which have the ability to recognize the antigens (i.e. 5–15 min post-exercise), peptide F was bound (Sigal & Ron 1994). It is important to understand to immune cell receptors and then internalized, and the difference between the two immune responses thus a response of peptide F to exercise may not as Hiddinga et al. (1994) reported. Specifically, a have been readily apparent until 30 min post- significant cell-mediated response was produced exercise. The presence of peptide F 30 min post- between proenkephalin peptide fragments and T- exercise in the white blood cell biocompartment was cells, which then caused a humoral-mediated res- observed and may be attributed to a saturation of ponse (e.g. increased number of antibody-forming peptide F immune receptors. B-cells). The peptide F concentration observed in the Based on the known interaction between peptide plasma biocompartment following exercise was not F and immune cells (Hiddinga et al. 1994), a study similar to the concentration of peptide F observed in was designed to examine the presence of peptide the white blood cell biocompartment. It may be that F within the white blood cell biocompartment of there was a shift of peptide F concentration from blood. A progressive exercise stress protocol had the white blood cell biocompartment to the plasma been shown to cause an increase in plasma peptide F biocompartment, or that immune cell receptors for concentration (Kraemer et al. 1985b, 1991). In a pro- peptide F were fully saturated following exercise gressive endurance exercise protocol, eight healthy (i.e. immediate, 5 min and 15 min post-exercise). men (21.0 ± 1.0 years) performed a 30-min cycle These results indicated that a possible interaction o exercise at 80% V 2max and returned for a second between peptide F and cells of the immune system session under quiet control conditions. Blood sam- existed, and that exercise modulated the concen- ples were taken pre-exercise, mid-exercise, immedi- tration of peptide F within this biocompartment. ately post-exercise, 5 min, 15 min, 30 min and 60 This initial pilot data was studied within the con- min post-exercise. Similar to previous research, text of utilizing a progressive endurance exercise there was an exercise-induced response of peptide model. Examining the effect of acute resistance F in the plasma biocompartment at 5 min post- exercise on the peptide F concentration has been exercise (Kraemer et al. 1985a, 1985b). Unique to this a topic of interest. Such data in this pilot study the adrenal medulla 213

0.40 Plasma WBC RBC 0.35 )

–1 0.30

Fig. 16.8 Comparison of peptide 0.25 F concentration in three biocompartments following 0.20 = ≤ resistance exercise. A p 0.05 0.15 difference from corresponding plasma biocompartment; B = p ≤ 0.05 0.10 Peptide F (pmol·mL A difference from corresponding white AB 0.05 A A blood cell biocompartment, mean AB AB ± SE. RBC, red blood cell 0 biocompartment; WBC, white blood Pre-exercise 0 15 cell biocompartment. Post-exercise time (minutes) supported the examination of peptide F within the under resting conditions and in response to the white blood cell biocompartment, and the response resistance exercise. of peptide F in this biocompartment to resistance The increase in peptide F concentration in the exercise to secrete cytokines which produce not white blood cell biocompartment occurred with a only paracrine effects but also autocrine effects on significant increase in the total number of white the same activated immune cells to up-regulate blood cells measured in that biocompartment. receptor expression for further stimulation (Sigal During the recovery period, the number of peptide & Ron 1994). Although the receptor for peptide F F molecules per white blood cell was significantly has not been characterized, it may potentially bind increased by nearly 50%, indicating the potential to cells within the blood biocompartments (i.e. significance of the interaction of peptide F within immune cells). Erythrocytes possess peptide recep- the white blood cell biocompartment. The inter- tors for other hormones, such as insulin and insulin- action with the white blood cell biocompartment like growth factor I (Horuk et al. 1993; Hagino et al. can be supported by the fact that peptide F may be 1994). There is also the possibility of peptide F in binding to immune cell surface recpetors (Sibinga the red blood cell biocompartment interacting with & Goldstein 1988), subsequently becoming inter- similar peptide receptors or a yet to be determined nalized, and resulting in less measurable peptide opioid. F in either the plasma or white blood cell biocom- In another study (Bush et al. under review) men partment. The increase in the number of peptide F aged 22 ± 0.9 years performed an acute resistance molecules associated with immune cells during exercise protocol where they completed six sets of recovery can be potentially explained by: (i) a 10-RM squats on the Smith-squat machine. Blood down-regulation of immune cell surface receptors; samples for analysis of peptide F concentrations or (ii) an increase in production and secretion of were obtained at baseline, immediate post- and 15 peotide F from the adnreal medulla or immune min post-exercise. There was a hierarchy of peptide cells, as observed by an increase in peptide F con- F concentrations where the plasma biocompartment centrations with the white blood cell biocompart- showed the highest (p < 0.05) concentration of pep- ment. The measurable amount of peptide F in the tide F versus the white blood cell and red blood cell red blood cell biocompartment may be reflective of biocompartments at baseline and during recovery a shift in the peptide F among the three different (Fig. 16.8). Furthermore, the concentration in the blood biocompartments. However, the physiolo- white blood cell biocompartment was higher than gical significance of the observance of peptide F in that of the red blood cell biocompartment (Fig. 16.8). the white and red blood cell biocompartments This hierarchy of concentration was consistent both remains to be elucidated. 214 chapter 16

Summary antibody by B-cells (Hiddinga et al. 1994). However, this relationship has not yet been studied in vivo. Neural stimulation is the primary mechanism The complexity and degree of interactions between involved with the activation and release of com- the endocrine and immune systems are just begin- pounds in the adrenal medulla. Peptide F is an ECP ning to be understood. Some investigators have secreted from the chromaffin cells of the adrenal examined the possibility of bi-directional regulation medulla. Peptide F has been shown to be secreted between these systems (Stein et al. 1985; Bateman along with epinephrine during exercise (Kraemer et al. 1989). It has been shown that T-cells have et al. 1985a, 1985b, 1987, 1988, 1990a, 1991, 1992; receptors for enkephalins (Wybran et al. 1979) and Bush et al. 1998, under review; Triplett-McBride appear to have receptors specific for peptide F 1998). However, the pattern of release of these two (Hiddinga et al. 1994). It has also been demonstrated substances during and after exercise may differ that T-cells can synthesize and release the small (Kraemer et al. 1985b, 1990a, 1991). Interestingly, the ECPs for autoregulation of function (Blalock 1992, pattern of release of peptide F during and after an 1994). Therefore, it is possible that one of the main acute bout of exercise also differs between highly roles of peptide F is to enhance immune cell func- trained male endurance athletes and untrained men tion to counteract and balance the suppressive (Kraemer et al. 1985a, 1985b). Many of the biolo- effects of other endocrine hormones such as cortisol gical functions of peptide F are unknown, but it has and epinephrine. While the modulation of endogen- been demonstrated that peptide F may improve the ous peptide F may promote movement to various activation and function of the T-cells of the immune target tissues, it appears that the mechanisms are system (Hiddinga et al. 1994). Peptide F has been related to adrenal medullary function, primarily shown to enhance helper T-cell activation in vitro regulation of peptide F concentrations related to the and subsequently cause increased production of response of stress (i.e. exercise).

References

Bastiaensen, E., Miserez, B. & De Potter, W. Brantl, V., Gramsch, C., Lottspeich, F. De Potter, W.P., Dillen, L., Annaert, W. (1988) Subcellular fractionation of et al. (1986) Novel opioid peptides et al. (1988) Evidence for the co-storage bovine ganglion stellatum: co-storage derived from hemoglobin: hemorphins. and co-release of neuropeptide Y and of noradrenaline, Met-enkephalin and European Journal of Pharmacology 125, noradrenaline from large dense cored neuropeptide Y in large ‘dense-cored’ 309–310. vesicles in sympathetic nerves of the vesicles. Brain Research 442(1), 124–130. Brooks, S., Burrin, J., Cheetham, M.E. bovine vas deferens. Synapse 2(2), Bateman, A., Singh, A., Kral, T. & Solomon, et al. (1988) The responses of the 157–162. S. (1989) The immune–hypothalamic– catecholamines and β-endorphin to brief Farrell, P.A., Gustafson, A.B., Morgan, pituitary–adrenal axis. Endocrine Reviews maximal exercise in man. European W.P. & Pert, C.B. (1987) Enkephalin, 10, 92–112. Journal of Applied Physiology 57, 230–234. catecholamines, and psychological Blalock, J.E. (1992) Production of peptide Bush, J.A., Kraemer, W.J., Mastro, A.M. mode alterations: effect of prolonged hormones and neurotransmitters by the et al. (1998) Exercise and recovery exercise. Medicine and Science in Sports immune system. Chemical Immunology responses of adrenal medullary and Exercise 19, 347–353. 52, 1–24. neurohormones to heavy resistance Fielding, R.A., Meredith, C.A., O’Reilly, Blalock, J.E. (1994) The syntax of immune– exercise. Medicine and Science in Sports K.P. et al. (1991) Enhanced protein neuroendocrine communication. and Exercise 31(4), 554–559. breakdown after eccentric exercise in Immunology Today 15(11), 504–511. Bush, J.A., Kraemer, W.J. & Mastro, young and old men. Journal of Applied Boarder, M.R. & McArdle, W. (1986) A.M. (under review) Peptide F Physiology 71, 674–679. Breakdown of small enkephalin immunoreactivity in different Fridèn, J., Sjoström, M. & Ekblom, B. (1983) derivatives and adrenal peptide E by circulatory biocompartments after Myofibrillar damage following intense human plasma. Biochemical Pharmacology exercise stress (Peptides, in review). eccentric exercise in man. International 35, 1043–1047. Coupland, R.E. (1965) The Natural History Journal of Sports Medicine 4(3), 170–176. Boone, J.B., Sherraden, T., Pierzchala, K., of the Chromaffin Cell. Longman, London. Fry, A.C., Morton, A.R. & Keast, D. (1991) Berger, R. & Van Loon, G.R. (1992) Coupland, R.E. (1972) The chromaffin Overtraining in athletes, an update. Plasma Met-enkephalin and system. In: Catecholamines: Handbook of Sports Medicine 12, 32–65. catecholamine response to intense Experimental Pharmacology, vol. 33 Fry, A.C., Kraemer, W.J., Van Borselen, F. exercise in humans. Journal of Applied (Blaschko, H. & Muscholl, E. eds.). et al. (1994) Catecholamine responses to Physiology 73(1), 388–392. Springer-Verlag, Berlin: 16–39. short-term high-intensity resistance the adrenal medulla 215

exercise overtraining. Journal of Applied Medicine and Science in Sports and Exercise responses during heat acclimation in Physiology 77, 941–946. 21(2), 154–160. humans. Peptides 8, 715–719. Fry, A.C., Kraemer, W.J. & Ramsey, L.T. Johnson, H.M., Smith, E.M., Torres, B.A. Kraemer, W.J., Rock, P.B., Fulco, C.S. et al. (1998) Pituitary–adrenal–gonadal & Blalock, J.E. (1982) Regulation of (1988) Influence of altitude and caffeine responses to high-intensity resistance the in vitro antibody response by during rest and exercise on plasma exercise overtraining. Journal of Applied neuroendocrine hormones. Proceedings levels of proenkephalin peptide F. Physiology 85(6), 2352–2359. of the National Academy of Sciences 79, Peptides 9, 1115–1119. Genuth, S.M. (1988) The adrenal glands. 4171–4174. Kraemer, W.J., Diazdos, J.E., Gordon, S.E. In: Physiology (Berne, R.M. & Levy, M.N., Kaplan, N.R. (1988) The adrenal glands. et al. (1990a) The effects of graded eds). C.V. Mosby Co., St. Louis, MI: In: Textbook of Endocrine Physiology exercise on plasma proenkephalin 950–982. (Griffin, J.E. & Ojeda, S.R., eds.). Oxford peptide F and catecholamine responses Hagino, H., Shii, K., Yokono, K. et al. (1994) University Press, New York: 245–272. at sea level. European Journal of Physiology Enzyme-linked immunosorbent assay Katzenstein, G.E., Lund, D., Schultz, P. 61, 214–217. method for human autophosphorylated & Lewis, R.V. (1987) Target tissue Kraemer, W.J., Marchitelli, L., Gordon, insulin receptor: applicability to insulin- distribution of the proenkephalin S.E. et al. (1990b) Hormonal and growth resistant states. Diabetes 43(2), 274–280. peptides F, E, and B. Biochemical and factor responses to heavy resistance Hansen, P.E. & Morgan, B.A. (1984) Biophysical Research Communications 146, exercise protocols. Journal of Applied Structure-activity relationships in 1184–1190. Physiology 69(4), 1442–1450. enkephalin peptides. In: The Peptides: Kilpatrick, D.L., Lewis, R.V., Stein, S. Kraemer, W.J., Patton, J.F., Knuttgen, H.G. Analysis, Synthesis, Biology (Udenfied, S. & Udenfriend, S. (1980) Release of et al. (1991) Effects of high-intensity & Meienhofer, J. eds.). Academic Press, enkephalins and enkephalin-containing cycle exercise on sympathoadrenal– New York: 269–275. polypeptides from perfused beef medullary response patterns. Journal of Hiddinga, H.J., Katzenstein, G.E., adrenal glands. Proceedings of the Applied Physiology 70(1), 8–14. Middaugh, C.R. & Lewis, R.V. (1990) National Academy of Sciences USA 77(12), Kraemer, W.J., Lewis, R.V., Triplett, N.T. Secondary structure characteristics of 7473–7475. et al. (1992) Effects of hypnosis on proenkephalin peptides E, B, and F. Kilpatrick, D.L., Taniguchi, T., Jones, B.N., plasma proenkephalin peptide F Neurochemical Research 15(4), 393–399. Stern, A.S. & Shively, J.E. (1981) A highly and perceptual and cardiovascular Hiddinga, H.J., Isaak, D.D. & Lewis, R.V. potent 3200 dalton adrenal opioid responses during submaximal exercise. (1994) Enkephalin-containing peptides peptide that contans both a [Met]- and European Journal of Applied Physiology 65, processed from proenkephalin [Leu]-enkephalin sequence. Proceedings 573–578. significantly enhance the antibody- of the National Academy of Sciences of the Kuipers, H. & Keizer, H.A. (1988) forming cell responses to antigens. United States of America 78, 3265–3268. Overtraining in elite athletes: review Journal of Immunology 152, 3748–3759. Kimura S., Lewis, R.V., Stern, A.S. and directions for the future. Sports Horuk, R., Colby, T.J., Darbonne, W.C., et al. (1980) Probable precursors of Medicine 6, 79–92. Schall, T.J. & Neote, K. (1993) The [Leu]enkephalin and [Met]enkephalin Lewis, R.V. (1982) Enkephalin biosynthesis human erythrocyte inflammatory in adrenal medulla: peptides of 3–5 in the adrenal medulla. In: Regulatory peptide (chemokine) receptor. kilodaltons. Proceedings of the National Peptides: from Molecular Biology to Biochemical characterization, Academy of Sciences USA 77, 1681–1685. Function, vol. 33 (Costa, E. & Trabucchi, solubilization, and development of a Kjær, M. & Galbo, H. (1988) Effect of M., eds.). Raven Press, New York: binding assay for the soluble receptor. physical training on the capacity to 167–174. Biochemistry 32, 5733–5738. secrete epinephrine. Journal of Applied Lewis, R.V. & Stern, A.S. (1983) Howlett, T.A., Tomlin, S., Ngahfoong, L. Physiology 64, 11–16. Biosynthesis of the enkephalins and et al. (1984) Release of β-endorphin Kjær, M., Christensen, N.J., Sonne, B., enkephalin-containing polypeptides. and met-enkephalin during exercise Richter, E.A. & Galbo, H. (1985) Effect Annual Review of Pharmocology and in normal women: response to of exercise on epinephrine turnover in Toxicology 23, 353–372. training. British Medical Journal 288, trained and untrained male subjects. Lewis, R.V., Stern, A.S., Rossier, J., Stein, S. 1950–1952. Journal of Applied Physiology 59, & Udenfriend, S. (1979) Putative Hughes, J., Smith, T.W. Kosterlitz, H. et al. 1061–1067. enkephalin precursors in bovine (1975) Identification of two related Kraemer, W.J., Noble, B., Culver, B. & adrenal medulla. Biochemical and pentapeptides from the brain with Lewis, R.V. (1985a) Changes in plasma Biophysical Research Communications 89, potent opiate agonist activity. Nature proenkephalin peptide F and 822–829. 258, 577–579. catecholamine levels during graded Livett, A.R., Dean, D.M., Whelan, L.G., Ivanov, V.T., Karelin, A.A., Philippova, exercise in men. Proceedings of the Udenfriend, S. & Rossier, J. (1981) M.M., Nazimov, I.V. & Pletnev, V.Z. National Academy of Sciences of the United Co-release of enkephalin and (1997) Hemoglobin as a source of States of America 82, 6349–6351. catecholamines from cultured adrenal endogenous bioactive peptides: the Kraemer, W.J., Noble, B., Robertson, K. & chromaffin cells. Nature 289, 317–319. concept of tissue-specific peptide pool. Lewis, R.V. (1985b) Response of plasma Livett, B.G. (1984) Adrenal medullary Biopolymers 43, 171–188. proenkephalin peptide F to exercise. chromaffin cells in vitro. Physiological Jaskowski, M.A., Jackson, A.S., Raven, P.B. Peptides 6 (suppl. 2), 167–169. Review 64, 1103–1161. & Caffrey, J.L. (1989) Enkephalin Kraemer, W.J., Armstrong, L.E., McBride, J.M., Kraemer, W.J., Triplett- metabolism: effect of acute exercise Marchitelli, L.J., Hubbard, R.W. & McBride, N.T. & Sebastianelli, W. (1998) stress and cardiovascular fitness. Leva, H. (1987) Plasma opioid peptide The effect of resistance exercise on free 216 chapter 16

radical production. Medicine and Science Schultzberg, M., Lundberg, J.M., Hökfelt, catecholamins. Journal of Autonomic in Sports and Exercise 30(1), 67–72. T. et al. (1978) Enkephalin-like Nervous System 7, 41–58. McCarthy, D.A. & Dale, M.M. (1988) The immunoreactivity in gland cells and Viveros, O.H., Diliberto, E.J., Hazum, E. leukocytosis of exercise: a review and nerve terminals of the adrenal medulla. & Chang, E.J. (1979) Opiate-like model. Sports Medicine 6, 333–363. Neuroscience 3, 1169–1186. materials in the adrenal medulla: McCully, K.K. & Faulkner, J. (1985) Injury Sibinga, N.E.S. & Goldstein, A. (1988) evidence for storage and secretion with to skeletal muscle fibers of mice Opioid peptides and opioid receptors catecholamines. Molecular Pharmacology following lengthening contraction. in cells of the immune system. Annual 16, 1101–1108. Journal of Applied Physiology 59, 119–121. Review of Immunology 6, 219–249. Wasserman, K, Beaver, W. & Whipp, B. Mackinnon, L.T. (1992) Current Issues in Sigal, L.H. & Ron, Y. (1994) Cytokines. In: (1986) Mechanisms and pattern of blood Exercise Science Series: Exercise and Immunology and Inflammation, Basic lactate increase during exercise in man. Immunology. Human Kinetics, Mechanisms, and Clinical Consequences Medicine and Science in Sports and Exercise Champaign, IL. (Sigal, L.H. & Ron, Y., eds). McGraw- 18, 344–352. Marotti, T., Rabatic, S., Gabrilovac, J. Hill, New York: 482–490. Wilson, S.P. (1991) Processing of (1993) A characterization of the in vivo Simonds, W.F. (1988) The molecular basis proenkephalin in adrenal chromaffin immunomodulation by Met-enkephalin of opioid receptor function. Endocrine cells. Journal of Neurochemistry 57(3), in mice. International Journal of Review 9, 200–212. 876–881. Immunopharmacology 15(8), 919–926. Sommers, D.K., Loots, J.M., Simpson, S.F. Wilson, S.P., Chang, K.J. & Viveros, O.R. Martin, J., Prustowsky, M.B. & Angeletti, et al. (1990) Circulating met-enkephalin (1982) Proportional secretion of opioid R.H. (1987) Preproenkephalin mRNA in in trained athletes during rest, peptides and catecholamines from T-cells, macrophages, and mast cells. exhaustive treadmill exercise and adrenal chromaffin cells in culture. Journal of Neuroscience Research 18, 82–87. marathon running. European Journal of Journal of Neuroscience 2(8), 1150–1156. Nieman, D.C., Henson, D.A., Sampson, Clinical Pharmacology 38(4), 391–392. Wybran, J., Appelbroom, T., Famaey, J.P. C.S. et al. (1995) The acute immune Stein, M., Keller, S.E., Schleiter, S.J. (1985) & Govaert, A. (1979) Suggestive response to exhaustive resistance Stress and immunomodulation: the role evidence for receptors for morphine exercise. International Journal of Sports of depression and neuroendocrine and methionine-enkephalin on normal Medicine 16(5), 322–328. function. Journal of Immunology 135(2) human blood T lymphocytes. Journal of Richter, G., Ballmann, M. & Conlon, J.M. (suppl.), 827–833. Immunology 123, 1068–1070. (1986) Effect of streptozotocin Triplett-McBride, N.T., Mastro, A.M., Yanagihara, N., Oishi, Y., Yamamoto, H. administration upon the serotonin McBride, J.M. et al. (1998) Plasma et al. (1996) Phosphorylation of content of the pancreas and small proenkephalin peptide F and human B chromogranin A and catecholamine intestines of the rat. Hormone and cell responses to exercise stress in fit secretion stimulated by elevation of Metabolic Research 18(10), 663–665. and unfit women. Peptides 19(4), intracellular Ca2+ in cultured bovine Roth, K.A., Lorenz, R.G., Unanue, R.A. 731–734. adrenal medullary cells. Journal of & Weaver, C.T. (1989) Nonopiate active Udenfriend, S. & Kilpatrick, D.L. (1984) Biological Chemistry 271(29), 17 463– proenkephalin-derived peptides are Proenkephalin and the products of its 17 468. secreted by T helper cells. FASEB Journal processing: chemistry and biology. In: Zhao, Q., Garreau, I., Sannier, F. & Piot, 3(12), 2401–2406. The Peptides: Analysis, Synthesis, and J.M. (1997) Opioid peptides derived Round, J.M., Jones, D.A. & Cambridge, G. Biology (Udenfriend, S. & Meienhofer, J., from hemoglobin: hemorphins. (1987) Cellular infiltrates in human eds), vol. 6. Academic Press, Orlando, Biopolymers 43(2), 75–98. skeletal muscle: exercise induced FL: 25–67. Zurawski, G., Benedik, M., Kamb, B.J. et al. damage as a model for inflammatory Viveros, O.H. & Wilson, S.P. (1983) The (1986) Activation of mouse T-helper cells muscle disease? Journal of the adrenal chromaffin cell as a model to induces abundant preproenkephalin Neurological Sciences 82, 1–11. study the cosecretion of encephalin and mRNA synthesis. Science 232, 772–775. Chapter 17

Exercise and the Hypothalamic–Pituitary– Adrenal Axis

WARRICK J. INDER AND GARY A. WITTERT

The hypothalamic–pituitary–adrenal (HPA) axis and release (Catalano et al. 1986). Cortisol inhibits constitutes a hormonal network which is activated its own secretagogues, with a negative feedback during stress. ‘Stress’ may take many forms, includ- loop operating to inhibit ACTH from the pituitary ing acute episodes such as life-threatening illness, as well as acting at the hypothalamic level on both surgery or hemorrhage, or more chronic stressors corticotropin-releasing hormone (CRH) and argi- such as depression or eating disorders. All of these nine vasopressin (AVP) (Keller-Wood & Dallman forms of stress have been extensively studied, and 1984; Stewart 2003). This feedback tends to avoid the extent of the activation of the HPA axis quan- prolonged and inappropriate periods of hypercor- tified. Exercise, whether it is an acute episode of tisolism. CRH and AVP, present in the parvicellular strenuous exercise or chronic endurance training is region of the paraventricular nucleus in the hypo- also known to activate the HPA axis. This chapter thalamus (Pelletier et al. 1983), are the primary regu- will briefly outline the physiological regulation of lators of ACTH secretion (Orth 1992; Kjær, A. 1993). the HPA axis, followed by a comprehensive review CRH is a 41-amino-acid peptide first isolated in the of the specific effects of exercise and exercise train- sheep in 1981 (Vale et al. 1981). It is a potent stimulus ing on the axis in human subjects. of both ACTH production and secretion (Orth 1992). CRH binds to specific high affinity CRH receptors Overview of the regulation of cortisol on the corticotrope (Chen et al. 1993). Intracellular production by the adrenal cortex signaling is via the protein kinase A/cyclic AMP second messenger system (Aguilera et al. 1983). The target end-organ of the HPA axis, the adrenal Increased secretion of CRH is thought to play a cortex, secretes cortisol during basal conditions in major role to increase ACTH and cortisol levels a circadian pattern, with higher levels in the early during many forms of acute stress (Chrousos 1992; morning hours which fall over the course of the day, Orth 1992). AVP, a 9-amino-acid peptide, binds to reaching low levels by midnight (Krieger et al. 1971). a specific receptors on corticotropes known as V3- Cortisol has a number of physiological roles. These (sometimes designated V1b) receptors (Sugimoto include sodium and water balance and blood pres- et al. 1994). In turn, the protein kinase C second mes- sure control, maintenance of glucose homeostasis, senger system is activated, leading to ACTH release adipogenesis, inhibition of osteoblast function and (Liu et al. 1990). CRH and AVP are co-localized in anti-inflammatory actions including suppression of the median eminence of the hypothalamus (Whitnall the immune response (Stewart 2003). Cortisol is et al. 1987). AVP acts synergistically with CRH under the influence of adrenocorticotropic hormone to further amplify ACTH secretion during stress (ACTH) secreted by the anterior pituitary (Stewart (Gillies et al. 1982, Rivier & Vale 1983). AVP is also a 2003). ACTH binds to corticotropin receptors in the critical regulator of sodium and water balance and a adrenal cortex, which leads to cortisol production potent vasoconstrictor (Jard 1988). 217 218 chapter 17

The presence of additional stimulatory and in- are markedly increased (Alexander et al. 1991). In hibitory factors which may affect ACTH secretion humans, both acute high intensity exercise (Wittert has been proposed (Grossman & Tsagarakis 1989; et al. 1991) and prolonged submaximal exercise Alexander et al. 1996). A number of other hormones, (Inder et al. 1998a) are associated with increases in cytokines and neurotransmitters interact with the plasma AVP that parallel ACTH and cortisol levels. HPA axis mainly through effects on CRH and less The extent of the rise in AVP also appears to commonly AVP. There is evidence that leukemia determine the degree to which the exercise-induced inhibitory factor (LIF) is a stimulus to pituitary activation of the HPA axis is inhibited by exogenous ACTH secretion (Auernhammer & Melmed 2000), glucocorticoids (Petrides et al. 1994). In a group of but the physiological role of other factors interacting exercising men pretreated with 4 mg dexametha- directly at the level of the pituitary or adrenal sone, four out of 11 were noted to exhibit a significant glands remains unclear. The regulation of the HPA increment in ACTH and cortisol following exercise. axis is summarized in Fig. 17.1. In this group, the plasma AVP response was six times that of the group where dexamethasone abol- The effect of acute exercise stress ished the ACTH/cortisol rise (Petrides et al. 1997). on the HPA axis and factors modulating Subsequent studies comparing the group with- the response out dexamethasone suppressibility (designated high responders) have also shown a greater integrated cortisol response to psychological stress (Singh et al. Mechanism of the cortisol response to exercise 1999). This may identify a group of individuals Like other acute stressors, acute intense exercise is who have a greater HPA axis response to different a potent activator of the HPA axis (Luger et al. 1987). stressors mediated via AVP. Changes in plasma The increase in plasma cortisol occurs in spite of an AVP are also correlated with alterations in plasma increase in clearance from the circulation (Few osmolality, but the increase observed during intense 1974). Even anticipation of competition may result exercise is greater than expected due to osmolality in a cortisol increase (Suay et al. 1999) and psycho- changes alone (Wade & Claybaugh 1980). In par- logical stress occurring prior to exercise increases ticular, the increase in plasma AVP correlates with the cortisol response (Kaciuba-Uscilko et al. 1994). changes in plasma osmolality during prolonged sub- Studies in humans and large mammals have indic- maximal exercise, but this relationship is lost dur- ated an important role for both CRH and AVP in ing subsequent incremental exercise to exhaustion the exercise-induced stimulation of ACTH secretion (Inder et al. 1998a). The reduction in plasma volume which then causes the cortisol rise at the adrenal which occurs during exercise may also contribute to level. Noting the limitations of peripheral CRH the AVP response (Robertson & Athar 1976). measurement, some (Elias et al. 1991; Harte et al. β-Endorphin is an opioid peptide derived from 1995; Inder et al. 1998a) but not all (Luger et al. 1987; proopiomelanocortin (POMC) (Morley 1981). Pre- Wittert et al. 1991) studies have found an increase viously it was thought to be secreted in equimolar in plasma CRH after exercise (Fig. 17.2). Duration amounts to ACTH, and a number of studies have of the exercise may be important in explaining the examined the plasma β-endorphin response to exer- differences, along with the sensitivity of the CRH cise (Carr et al. 1981; Rahkila et al. 1987; Petraglia assay. When exercise is performed under condi- et al. 1988; Schwarz & Kindermann 1989). A major tions of a continuous CRH infusion, at a level where problem with much of this research is the find- the corticotrope CRH receptors would be fully ing of 100% cross-reactivity with β-lipotropin and saturated, a robust rise in ACTH and cortisol above acetylated forms of β-endorphin in most radioim- the elevated baseline is noted (Smoak et al. 1991). munoassays. Therefore, much of the β-endorphin- This implies a factor additional to CRH being cru- like immunoreactivity may not be opioid active. cial in the ACTH response. In the pituitary venous Using more specific immunoradiometric assays, it effluent of the exercising horse, plasma AVP levels is now clear that in the basal state, β-endorphin is exercise and the hpa axis 219

Hippocampus Glucocorticoid receptors Prefrontal cortex Processive Cytokine receptors (neurogenic) stress – (fear, restraint) Leptin receptors Raphe nucleus + BST Hypothalamic 5-HT Lateral GABAergic inputs septum Paraventricular GABA MeA nucleus Hypothalamic glutamatergic – + inputs Brain stem – catecholaminergic Hypothalamic + inputs glutamatergic inputs + + NTS CRH Subfornical org neuron OVLT + + Prostaglandins Systemic MePO (physiological) stress Brain stem (cytokines, hypoxia, Arcuate, vasculature hemorrhage) NPY and POMC neurons HYPOTHALAMUS Systemic (physiological) stress (osmotic challenge, macromolecules) to + CNS + – CRH AVP IL-1, IL-2, IL-6, TNFα

PITUITARY – Immune cells + ACTH – Cortisol

ADRENAL

Fig. 17.1 Regulation of the hypothalamic–pituitary–adrenal (HPA) axis. Neuronal and neurotransmitter input to hypothalamic corticotropin-releasing hormone (CRH) and arginine vasopressin (AVP) -containing neurons result in release of these hormones into the hypothalamic–hypophysial portal system. This causes an increase in secretion of adrenocorticotropic hormone (ACTH) from the pituitary, which in turn stimulates cortisol release from the adrenal cortex. Cortisol acts at both pituitary and hypothalamic level via negative feedback to inhibit ACTH, CRH and AVP. 220 chapter 17

1000 Exercise undetectable in most normal subjects (Gibson et al. β Control *** 1993). True -endorphin1–31 is released during exer- ) 800 cise into peripheral blood in only about 50% of sub- –1 *** jects, and represents only a small proportion of ** 600 β-endorphin immunoreactivity (Harbach et al. 2000). †† It appears, however, that exercise does increase cent- 400 †† †† † ral levels of endogenous opioid peptides (Thoren

Cortisol (nmol·L 200 et al. 1990). An infusion of naloxone, an opioid receptor antagonist, increases the perceived effort of (a) 0 exercise (Grossman et al. 1984; Sgherza et al. 2002). Highly trained athletes have evidence for increased 50 *** central opioid tone, probably induced by training (Inder et al. 1995). Basal plasma levels of β-endorphin 40 )

–1 do correlate with the subsequent ACTH response to 30 naloxone, a marker of central opioid tone (Inder et *** *** al. 1998b). Activation of endogenous opioid peptides †† 20 ** †† has been implicated in the positive mood effects †† of exercise. They have also been hypothesized in a ACTH (pmol·L 10 †† causative role in exercise induced hypothalamic amenorrhea (Laatikainen 1991). (b) 0

5 *** Effect of the intensity and duration of exercise †† *** *** > o †† Short duration exercise to 60% V 2max results in ) 4 ††

–1 ACTH and cortisol release proportional to the inten- sity of the exercise (Davies & Few 1973; Howlett 3 1987; Luger et al. 1987; Kjær, M. et al. 1988; Deuster et al. 1989; Wittert et al. 1991). Even high intens-

CRH (pmol·L 2 ity exercise for as short as 1 min activates ACTH and cortisol secretion (Buono et al. 1986). Short-term (c) 1 submaximal exercise does not result in activation of the HPA axis even in conditions of extreme 10 * o heat (Kenefick et al. 1998). Exercising at 50% V 2max for 20 min does not cause an increase in cortisol 8 o ) levels, while at 70% V 2max, ACTH and cortisol in- –1 crease (Luger et al. 1987). When subjects have been 6

4 Fig. 17.2 (left) Changes in plasma levels (geometric mean

AVP (pmol·L AVP † ± SEM) of (a) cortisol, (b) adrenocorticotropic hormone 2 (ACTH), (c) corticotropin-releasing hormone (CRH) and (d) arginine vasopressin (AVP) in six male athletes 0 undergoing a cycle ergometer-exercise protocol aiming to –30 0 30 60 90 o maintain each athlete at 70% V 2max for 1 h between 0 and (d) Time (min) 60 min, followed by an incremental increase in absolute intensity of 25 W every 2 min until exhaustion. Significant difference from baseline (–30 min) *P < 0.05, **P < 0.01, ***P < 0.001. Significant difference between exercise and control time-matched data †P < 0.01, ††P < 0.001. exercise and the hpa axis 221

subjected to a progressive increase in exercise intens- elicited an increase in ACTH and cortisol only in the o ity in 10-min blocks, commencing at 40% V 2max, in- latter part of the exercise when blood glucose o creased ACTH was only seen after 80% V 2max was decreased (Tabata et al. 1990). reached (de Vries et al. 2000). Following 1 h of cycle ergometer exercise at 70% Vo , cortisol levels 2max Timing of exercise increased from baseline; yet a further significant increase in AVP, CRH, ACTH and cortisol was The HPA axis response to some stimuli may be observed when the intensity was then progressively influenced by the ambient basal cortisol level. For increased over a 10-min period until exhaustion example, the cortisol increment may be less in the (Inder et al. 1998a). Using salivary as opposed to morning when basal levels are higher (DeCherney plasma cortisol, a cortisol rise following 1 h of exer- et al. 1985). This is thought to be due negative feed- o cise was observed only at 76% V 2peak, but not at back inhibition. Kanaley et al. (2001) demonstrated 45% or 62%, while 40 min of exercise failed to in- that although basal and peak cortisol levels in crease salivary cortisol at any of the three intensities response to exercise were highest at 0700 h, the (Jacks et al. 2002). incremental response was greater compared to con- These observations are best explained by the exer- trol day when the exercise occurred at 2400 h. In cise being performed at or exceeding the anaerobic contrast, when comparing area under the curve and threshold. Previous studies where the participants the circadian baseline, no difference was found in have exercised at the individualized anaerobic women exercising at different times of the day threshold have demonstrated that exercise below (Thuma et al. 1995). When bouts of equivalent exer- this point does not result in activation of the HPA cise are repeated later in the day, the ACTH and axis (Kindermann et al. 1982; Gabriel et al. 1992). cortisol response to the second bout are greater than During incremental exercise to exhaustion, plasma the first, or a single bout performed on another ACTH and β-endorphin rise only once the indi- day (Ronsen et al. 2001a). vidual anaerobic threshold is reached (Schwarz & Kindermann 1990). Type of exercise Although some studies have failed to demon- strate a cortisol rise in response to prolonged low Squats and intermittent higher intensity cycling intensity exercise (Hoffman et al. 1994), ultra-long result in a cortisol response in contrast to cycling distance running is associated with similar cortisol beneath the anaerobic threshold which does not rises to repeated bursts of shorter, higher intensity (Vanhelder et al. 1985). Estimation of the cortisol exercise (Nagel et al. 1992). Cortisol levels at the response to rowing has lead to conflicting results. completion of a 100-km ultramarathon were sig- Although a significant increases in plasma cortisol nificantly elevated over baseline (Pestell et al. 1989). in response to maximal (7 min) and submaximal A 75-km cross-country ski race significantly elev- (40 min) rowing on rowing apparatus and 8 × 2000 m ated plasma cortisol levels (Vasankari et al. 1993). It on the water (Snegovskaya & Viru 1993) has been has been suggested that the activation of the HPA shown to occur in one study, a subsequent study axis seen during prolonged low intensity exercise is was unable to demonstrate an increase in plasma dependent on the development of hypoglycemia cortisol in response to maximal rowing to exhaus- (Tabata et al. 1991). In six subjects who exercised for tion on a rowing ergometer ( Jurimae & Jurimae o 14 h at 50% V 2max, the cortisol, ACTH and CRH 2001). Similar results were found when the rowing responses were totally abolished when plasma glu- was performed at a lower intensity for 2 h, with no cose concentrations were maintained at pre-exercise change in plasma cortisol seen ( Jurimae et al. 2001). levels. The authors suggested a threshold plasma Cortisol levels were noted to increase following glucose concentration of < 3.3 mmol·L–1 (Tabata kayak races of 19 and 42 km, with the extent of et al. 1991). In an earlier study from the same group, the rise being more pronounced in the longer race o cycling at 50% V 2max for up to 3 h or exhaustion (Lutoslawska et al. 1991). Swimming for 30 min 222 chapter 17

caused an increase in plasma cortisol at higher women than men, possibly indicating a greater AVP ambient water temperatures but not at 20°C response in women or a reduced sensitivity to glu- (Deligiannis et al. 1993). Activation of the HPA axis cocorticoid negative feedback (Deuster et al. 1998). can be induced by resistance exercise as well as African-American women may have an increased through endurance exercise. During a high (100% of ACTH response to exercise, but cortisol responses are subject’s maximum) and moderate (70% of subject’s similar to Caucasian women (Yanovski et al. 2000). maximum) intensity strength workout, plasma cor- tisol increased more during the high intensity proto- Altitude col (Raastad et al. 2000). Three sets versus one set of resistance exercises resulted in a greater cortisol When comparative exercise has been done at low increase (Gotshalk et al. 1997). and moderate altitude, cortisol levels were demon- strated to increase during exercise under both con- ditions; however, ACTH did not show an increase Age above baseline at moderate altitude (el-Migdadi In middle-aged men, resistance exercise consist- et al. 1996). Comparison of athletes undergoing ing of sets of bench press, sit-ups and leg press, interval training at sea level and 1800 m did not increased plasma cortisol concentrations (Häkkinen reveal any statistically significant difference in cor- & Pakarinen 1995). This effect was not seen how- tisol response, although the sympathetic nervous ever in women or elderly men. When comparing system response was greater at higher altitude young, middle-aged and elderly men, Silverman (Niess et al. 2003). Marathon runners competing at and Mazzeo (1996) found lower basal cortisol in the high altitude develop higher basal cortisol levels elderly sedentary subjects, but elevated basal levels on acclimatization which increase further on com- in the trained subjects of all ages. Peak cortisol pletion of the race (Marinelli et al. 1994). Similar response to maximal exercise showed an age-related data were observed in a group of healthy volunteers decline, independent of training status. There was undertaking a rigorous trekking expedition in the no difference between men of different ages in the Himalayas, where cortisol levels increased after cortisol response to a 45-min submaximal exercise 2 weeks (Martignoni et al. 1997). While circadian (Silverman & Mazzeo 1996). rhythms were maintained, 30% of the subjects demonstrated failure of dexamethasone suppres- sion (Martignoni et al. 1997). Overall, it appears that Gender acclimatization to altitude results in an increase in Using 90 min of cycling at 80% of the anaerobic basal cortisol levels compared to sea level, while the o threshold (approximately 50% V 2max), there was cortisol response to exercise is preserved. no difference in cortisol response between men and women, matched for body mass index (BMI) and Nutrition physical fitness (Davis et al. 2000). Similar results have been obtained when 30 min of treadmill run- Several studies have examined the effects of dietary ning was employed (Kraemer, R.R. et al. 1989). changes and supplement use, before, during and Rahkila et al. (1987) found no difference between after exercise, on the plasma cortisol response. men and women in cortisol and ACTH response Ingestion of carbohydrates during prolonged (2.5 h) o either during a graded treadmill exercise to ex- cycling or running at approximately 70% V 2max haustion or anaerobic treadmill exercise. Therefore, resulted in a lower cortisol response to the exercise both sexes appear to have a similar response to aero- and a lower rating of perceived exertion (Utter et al. bic endurance exercise. Following dexamethasone, 1999). Similar results were obtained by Deuster et al. however, the plasma cortisol and AVP response (1992) who observed that ingestion of a 7% glucose o to high intensity (90–100% V 2max) was greater in polymer/fructose/electrolyte solution at a rate of exercise and the hpa axis 223

200 mL every 30 min abolished the exercise-induced 1991; Heitkamp et al. 1996). The recovery from o cortisol rise during a 2-h run at 60–65% V 2max com- intense, prolonged exercise in trained athletes is pared to exercise ingesting an equal volume of associated with greater plasma ACTH levels, but water. Murray et al. (1991) using a carbohydrate similar plasma cortisol levels as compared to con- solution compared to a water placebo, also noted an trol subjects (Duclos et al. 1997). Intensive training improvement in a 4.8-km performance test after 2 h for an ultramarathon (Tharp & Buuck 1974; Pestell o of prolonged cycling exercise at 65–75% V 2max, in et al. 1989; Wittert et al. 1996), or experimentally in addition to a reduction in the ACTH and cortisol recreational athletes (Lehmann et al. 1993), has been response. shown to increase pituitary ACTH secretion with- Three days on a ketogenic diet of equal energy out affecting plasma cortisol levels. A number of content resulted in higher cortisol levels both before other studies are consistent with this. In male and and after exercise in comparison to a control mixed female joggers, a season of training was without diet (Langfort et al. 1996). Glycerol, which has been affect on basal plasma cortisol levels (Ronkainen suggested as an aid to maintain hydration during et al. 1986). In middle- and long-distance runners, exercise, did not result in any difference in plasma neither an increase in training intensity nor 12 cortisol response to 1 h of cycle ergometer exercise weeks of intense training in competitive swimmers o at 70% V 2max followed by an incremental increase (Mujika et al. 1996) affected basal plasma cortisol in intensity to exhaustion (Inder et al. 1998c). levels (Lehmann et al. 1992; Flynn 1997). While this Creatine is a popular nutritional supplement suggests reduced adrenal responsiveness to ACTH, among athletes. Short-term creatine supplementa- Duclos et al. (1998) showed that that responsiveness tion for 5 days did not alter the cortisol response to a to ACTH did not appear to be decreased after 1 h session of heavy resistance training, although endurance training (Duclos et al. 1998) but the sens- levels tended to be higher during recovery follow- itivity of the HPA axis to glucocorticoid feedback, ing the creatine (Op’t Eijnde & Hespel 2001). Post- at least at the level of the pituitary, was decreased exercise feeding with whole food, supplemental (Duclos et al. 2001). In trained men, there is a drink, carbohydrate beverage or placebo made no decreased sensitivity of monocytes to cortisol 24 h difference to cortisol levels measured over 24 h fol- after the last bout of exercise. This may be related to lowing exercise (Bloomer et al. 2000). a process of desensitization that may act to protect the body from prolonged, exercise-induced cortisol The effect of exercise training on the secretion (Duclos et al. 2003). HPA axis In contrast to these studies, it has been shown that intensive training on a treadmill increased resting The objective of training is to optimize human per- plasma cortisol levels without affecting basal ACTH formance; training produces a number of neuro- (Kraemer, W.J. et al. 1989). In competitive swimmers, endocrine adaptations, which result in alterations a short-term increase in training distance may of the activity of the HPA axis. This response is increase basal plasma cortisol levels (without neces- determined by the training volume, intensity, type sarily affecting performance) (Kirwan et al. 1988), of exercise, optimum periods of rest (regeneration) and trained cyclists have been reported to have and poorly defined psychological factors. higher basal cortisol levels than untrained controls (Silverman & Mazzeo 1996). Age, gender, nutrition, mood, type and duration Effect of training on basal activity of the HPA axis of activity and training status may all contribute to The nature of the effect of training on basal activity the effects of physical training on the HPA axis. of the HPA axis remains unresolved. Normalization There is no difference in the response of male and of plasma cortisol levels after prolonged endurance female athletes to a sudden increase in training exercise may take up to 18–24 h (Lutoslawska et al. (O’Connor et al. 1991). In prepubescent gymnasts, 5 224 chapter 17

consecutive days of training (3 h each day at mod- decrease in intensity-dependent, exercise induced, erate intensity), had no significant effect on basal increases in cortisol levels (Fry & Kraemer 1997). plasma cortisol levels (Rich et al. 1992). In contrast, Although 2 years of intensive weight training was in a similar group of gymnasts 7–16 weeks of train- without affect on basal cortisol levels in adolescent ing resulted in increased basal cortisol levels, but males (Fry et al. 1994), 1 week of an overreaching energy intake was 31% below recommended levels stimulus induced an increase in early morning corti- (Filaire et al. 2003). Accordingly, the increased corti- sol levels in this group of subjects (Fry et al. 1994). In sol levels were likely the consequence of an energy young men, high-intensity weight training sufficient deficit rather than training per se. In well-nourished to induce an overtrained state resulted in a slightly young male gymnasts training had no significant increased exercise-induced testosterone/cortisol ratio effect in basal cortisol levels (Daly et al. 1998). but decreased exercise-induced cortisol. This hor- The response of plasma cortisol may be deter- monal profile is distinctly different from what has mined by both the duration and the nature of the been previously reported for other types of over- training program, because sprint interval (asso- training, indicating that high relative intensity resist- ciated with a substantial anaerobic component to ance exercise overtraining may not be successfully training) but not endurance training has been monitored via circulating testosterone and cortisol observed to increase basal plasma cortisol (Kraemer, (Fry et al. 1998). W.J. et al. 1989; Wittert 1996). An increase in training volume as opposed to intensity has been observed Effect of training on the response to subsequent to induce a decrease in both resting and exercise- acute exercise induced cortisol levels (Lehmann et al. 1992), which is characteristic of an overtraining syndrome. In The cortisol response to exercise of the same relative contrast a twofold increase in training volume has intensity (Rolandi et al. 1985; Deuster et al. 1989) is been shown not to affect plasma cortisol levels, and not affected by training, although the response to there was no difference in the endocrine responses the same absolute intensity may decrease (Buono to an increase in training volume with cross-training et al. 1987; Deuster et al. 1989; Botticelli et al. 1992; as compared to mode-specific training (Flynn et al. Hickson et al. 1994). In other words, some degree 1997). The hormonal response to 5 weeks of aerobic of adaptation occurs. In contrast, activation of the training is similar regardless of whether training HPA axis in response to supramaximal exercise was carried out under conditions to simulate in- is greater in endurance-trained subjects than in creased altitude (2500 m) or at sea level (Engred untrained subjects (Furell et al. 1987; Snegovskaya & et al. 1994). Similarly, season appears to have no Viru 1993). The nature of the exercise training may effect on the response to exercise training (Ronsen to some extent determine the response of the HPA et al. 2001b). In elderly men there is substantial indi- axis to subsequent acute exercise; when a substan- vidual variability in the effect of endurance training tial anaerobic component is present, the cortisol on the HPA axis, although in general the changes response to subsequent exercise may be increased appear to be similar to those that occur in younger (Fry et al. 1997). men (Struder et al. 1999). The effect of resistance exercise training on basal Exercise and abnormalities of the plasma cortisol levels is variable, and either no HPA axis change (Häkkinen et al. 1988a, 1988b; Fry et al. 1994) or a decrease has been reported (Alen et al. 1988); Overreaching and overtraining the difference probably related to lower training volumes. Increasing resistance, training volume, or If adaptation to the increased training regimen does intensity may result in an increase in resting cortisol not occur, or if insufficient recovery time is included levels (Fry & Kraemer 1997). A doubling of training in training regimens, overreaching and subsequently volume has also been observed to result in a overtraining may occur. Overreaching may be con- exercise and the hpa axis 225

Short-term Well Overreaching Overtraining intensified training adapted

ACTH Fig. 17.3 Effect of exercise training on plasma adrenocorticotropic Cortisol hormone (ACTH) and cortisol levels.

sidered as short-term overtraining and may be a cortisol or ACTH levels appear unaffected, but normal part of athletic training, or may be seen as a the responsiveness to exercise is decreased (Fry & short-term consequence of ultra-endurance events. Kraemer 1997). Some data suggests that hormone By contrast overtraining results in a state described responses to exercise load are superior in indicating as burnout or staleness, which is characterized by heavy training-induced stress when compared with increased fatigue, altered mood state, increased resting hormone levels. These responses indicated infections and suppressed reproductive function. decreased adrenocortical activity. However, since Overtraining may be, at least in part, the conse- marked individual differences were found in train- quence of inadequate periods of ‘regeneration’ ing- and overtraining-induced hormonal changes, (Lehmann et al. 1997). individual hormonal profiles are needed to follow- From the HPA axis standpoint, it has been sug- up training effects (Uusitalo et al. 1998). gested that the earliest stage of overreaching (or very early overtraining) may be reflected by a re- Athletic amenorrhea duced responsiveness to ACTH, which is initially compensated for by an increased pituitary ACTH Exercise-associated disorders of the reproductive response but a decreased adrenal cortisol response axis in women are associated with abnormalities of (Wittert et al. 1996; Lehmann et al. 1997). When fully the HPA axis; the response of plasma cortisol to evolved, the overtraining syndrome is charac- both maximal and submaximal acute exercise is terized by both increased basal and 24-h urinary reduced in amenorrheic compared to eumenorrheic cortisol levels and a reduced ACTH and cortisol athletes (Loucks & Horvath 1984; De Souza 1991). response to physical activity (Barron et al. 1985; Amenorrheic athletes have been found to have Lehmann et al. 1992). In highly trained distance run- higher mean basal (Ding et al. 1988; Loucks et al. ners who undertook a 38% increase in training 1989) early morning (Loucks et al. 1989) or mid- intensity over 3 weeks, six of the subjects developed afternoon- (De Souza et al. 1991) blood and 24-h sustained fatigue and the increase in serum cortisol, urine (Loucks et al. 1989) cortisol levels. There is also normally induced by 30 min of submaximal exer- evidence that basal CRH stimulation is increased cise, was lost (Verde et al. 1992). The severest and (Loucks et al. 1989; Hohtari et al. 1991) and that most fully evolved form of the overtraining syn- adrenal sensitivity to ACTH is reduced (De Souza drome is characterized by underactivity of both et al. 1991, 1994). the HPA axis and sympathoadrenal system. This complete pattern is only observed subsequent to Summary high-volume endurance overtraining at high caloric demands (Figs 17.3 and 17.4) (Lehmann et al. 1998). Trained and untrained individuals have a robust Whether overtraining induced by high volume increase in ACTH and cortisol in response to acute resistance exercise produces similar effects to highly high intensity exercise. This is mediated via the aerobic activities is not entirely clear. After maximal- hypothalamus by both CRH and AVP. The extent intensity-resistance exercise training, basal plasma of the cortisol rise is proportional to the relative 226 chapter 17

550 Controls be more prolonged to produce a significant increase 500 Athletes in cortisol levels. The effect of resistance exercise on

) 450

–1 the HPA axis appears to be most variable, and there 400 350 are both gender and age-specific affects. Variable 300 responses have been observed during other forms 250 of exercise, such as swimming and rowing. Carbo- 200 150 hydrate supplementation during more prolonged

Cortisol (nmol·L 100 exercise can attenuate or abolish the cortisol rise, 50 which may implicate relative hypoglycemia as 0 a potential factor inducing the activation of the 34567891011 12 13 14 15 16 HPA axis in this setting. At altitude, exercise still (a) Time (h) results in a significant rise in plasma cortisol, 90 Controls although basal cortisol levels are higher following 80 Athletes acclimatization. 70 ) Although a great deal has been learned about the –1 60 effects of physical activity on the HPA axis, there is 50 still much conflicting information and many areas 40 of confusion. The response of the HPA axis to 30 ACTH (ng·L any stress is dependent not only on the nature of 20 the stress, but also on the environment in which the 10 stress is imposed as well the inherent characteristics 0 34567891011 12 13 14 15 16 of the individual concerned (genetic factors, gender, (b) Time (h) personality, prior experience), concomitant stres- sors and nutritional state. In addition, the timing 250 and manner in which samples are collected will also ) –1 200 influence the results obtained. In general, it appears that the response to short- 150 term intensified training is an increased plasma 100 cortisol level, particularly if an anaerobic element is involved. Progressively adaptation occurs, with a 50 decreasing adrenal response to ACTH to exercise at

Cortisol (mmol·24 h the same relative intensity. With overreaching, the 0 (c) Controls Athletes absolute cortisol response is decreased, and when overtraining occurs, HPA axis activity is decreased Fig. 17.4 Adaptation of the hypothalamic–pituitary– adrenal (HPA) axis to chronic exercise stress in humans. as a whole (Barron et al. 1985). The precise factors Mean (± SEM) plasma concentrations of (a) cortisol, (b) that may modify an appropriate adaptive response adrenocorticotropic hormone (ACTH) and (c) a 24-h and lead to overreaching or overtraining remain to urinary free cortisol excretion in ultramarathon athletes be determined. and control subjects. The extent to which abnormalities of the HPA axis are simply consequences of, or are inherent to the pathophysiology of, a variety of conditions related to exercise training are not entirely clear. Whether intensity of the exercise as expressed as a percent- activity of the HPA axis can be used as a marker o age of V 2max. Older people may have a blunted of training stress and adaptation, by establishing response, while no difference exists between male the activity of and following longitudinally the and female subjects. At exercise levels below the HPA axis of individual athletes remains to be anaerobic threshold, the duration of exercise has to determined. exercise and the hpa axis 227

References

Aguilera, G., Harwood, J.P., Wilson, J.X. Carr, D.B., Bullen, B.A., Skrinar, G.S. et al. runners. Journal of Applied Physiology 76, et al. (1983) Mechanisms of action of (1981) Physical conditioning facilitates 1302–1309. corticotrophin-releasing factor and other the exercise-induced secretion of β- Deligiannis, A., Karamouzis, M., Kouidi, regulators of corticotropin release in endorphin and β-lipotropin in women. E., Mougios, V. & Kallaras, C. (1993) rat pituitary cells. Journal of Biological New England Journal of Medicine 305, Plasma TSH, T3, T4 and cortisol Chemistry 258, 8039–8045. 560–563. responses to swimming at varying water Alen, M., Pakarinen, A., Häkkinen, K. & Catalano, R.D., Stuve, L. & Ramachandran, temperatures. British Journal of Sports Komi, P.V. (1988) Responses of serum J. (1986) Characterization of Medicine 27, 247–250. androgenic-anabolic and catabolic corticotropin receptors in human Deuster, P.A., Chrousos, G.P., Luger, A. hormones to prolonged strength adrenocortical cells. Journal of Clinical et al. (1989) Hormonal and metabolic training. International Journal of Sports Endocrinology and Metabolism 62, responses of untrained, moderately Medicine 9, 229–233. 300–304. trained, and highly trained men to three Alexander, S.L., Irvine, C.H., Ellis, M.J. Chen, R., Lewis, K.A., Perrin, M.H. & Vale, exercise intensities. Metabolism 38, & Donald, R.A. (1991) The effect of W.W. (1993) Expression cloning of a 141–148. acute exercise on the secretion of human corticotropin-releasing-factor Deuster, P.A., Singh, A., Hofmann, A., corticotropin-releasing factor, arginine receptor. Proceedings of the National Moses, F.M. & Chrousos, G.C. (1992) vasopressin, and adrenocorticotropin Academy of Science of the United States of Hormonal responses to ingesting water as measured in pituitary venous blood America 90, 8967–8971. or a carbohydrate beverage during a 2 h from the horse. Endocrinology 128, 65–72. Chrousos, G.P. (1992) Regulation and run. Medicine and Science in Sports and Alexander, S.L., Irvine, C.H. & Donald, dysregulation of the hypothalamic– Exercise 24, 72–79. R.A. (1996) Dynamics of the regulation pituitary–adrenal axis. The Deuster, P.A., Petrides, J.S., Singh, A. of the hypothalamo–pituitary–adrenal corticotropin-releasing hormone et al. (1998) High intensity exercise (HPA) axis determined using a perspective. Endocrinology and promotes escape of adrenocorticotropin nonsurgical method for collecting Metabolism Clinics of North America 21, and cortisol from suppression by pituitary venous blood from horses. 833–858. dexamethasone: sexually dimorphic Frontiers in Neuroendocrinology 17, 1–50. Daly, R.M., Rich, P.A. & Klein, R. (1998) responses. Journal of Clinical Auernhammer, C.J. & Melmed, S. Hormonal responses to physical training Endocrinology and Metabolism 83, (2000) Leukemia-inhibitory factor- in high-level peripubertal male 3332–3338. neuroimmune modulator of endocrine gymnasts. European Journal of Applied Ding, J.H., Sheckter, C.B., Drinkwater, B.L., function. Endocrine Reviews 21, 313–345. Physiology and Occupational Physiology Soules, M.R. & Bremner, W.J. (1988) Barron, J.L., Noakes, T.I., Levy, W., Smith, 79, 74–81. High serum cortisol levels in exercise- C. & Millar, R.P. (1985) Hypothalamic Davies, C.T. & Few, J.D. (1973) Effects of associated amenorrhea. Annals of dysfunction in overtrained athletes. exercise on adrenocortical function. Internal Medicine 108, 530–534. Journal of Clinical Endocrinology and Journal of Applied Physiology 35, 887–891. Duclos, M., Corcuff, J.B., Rashedi, M., Metabolism 60, 803–806. Davis, S.N., Galassetti, P., Wasserman, Fougere, V. & Manier, G. (1997) Trained Bloomer, R.J., Sforzo, G.A. & Keller, B.A. D.H. & Tate, D. (2000) Effects of gender versus untrained men: different (2000) Effects of meal form and on neuroendocrine and metabolic immediate post-exercise responses of composition on plasma testosterone, counterregulatory responses to exercise pituitary adrenal axis. A preliminary cortisol, and insulin following resistance in normal man. Journal of Clinical study. European Journal of Applied exercise. International Journal of Sport Endocrinology and Metabolism 85, Physiology 75, 343–350. Nutrition Exercise and Metabolism 10, 224–230. Duclos, M., Corcuff, J.B., Arsac, L. et al. 415–424. DeCherney, G.S., DeBold, C.R., Jackson, (1998) Corticotroph axis sensitivity after Botticelli, G., Bacchi Modena, A., Bresciani, R.V. et al. (1985) Diurnal variation in the exercise in endurance-trained athletes. D. et al. (1992) Effect of naltrexone response of plasma adrenocorticotropin Clinical Endocrinology 48, 493–501. treatment on the treadmill exercise- and cortisol to intravenous ovine Duclos, M., Corcuff, J.B., Pehourcq, F. & induced hormone release in corticotropin-releasing hormone. Journal Tabarin, A. (2001) Decreased pituitary amenorrheic women. Journal of of Clinical Endocrinology and Metabolism sensitivity to glucocorticoids in Endocrinological Investigation 15, 61, 273–279. endurance-trained men. European 839–847. De Souza, M.J., Maguire, M.S., Maresh, Journal of Endocrinology 144, 363–368. Buono, M.J., Yeager, J.E. & Hodgdon, J.A. C.M. et al. (1991) Adrenal activation and Duclos, M., Gouarne, C. & Bonnemaison, (1986) Plasma adrenocorticotropin and the prolactin response to exercise in D. (2003) Acute and chronic effects cortisol responses to brief high-intensity eumenorrheic and amenorrheic runners. of exercise on tissue sensitivity to exercise in humans. Journal of Applied Journal of Applied Physiology 70, glucocorticoids. Journal of Applied Physiology 61, 1337–1339. 2378–2387. Physiology 94, 869–875. Buono, M.J., Yeager, J.E. & Sucec, A.A. De Souza, M.J., Luciano, A.-A., Arca, J.C., Elias, A.N., Wilson, A.F., Pandian, M.R. (1987) Effect of aerobic training on the Demers, L.M. & Loucks, A.B. (1994) et al. (1991) Corticotropin releasing plasma ACTH response to exercise. Clinical tests explain blunted cortisol hormone and gonadotropin secretion Journal of Applied Physiology 63, responsiveness but not mild in physically active males after acute 2499–2501. hypercortisolism in amenorrheic exercise. European Journal of Applied 228 chapter 17

Physiology and Occupational Physiology multiset versus single-set heavy- Hohtari, H., Salminen Lappalainen, K. 62, 171–174. resistance exercise protocols. Canadian & Laatikainen, T. (1991) Response of Engfred, K., Kjær, M., Secher, N.H. et al. Journal of Applied Physiology 22, 244–255. plasma endorphins, corticotropin, (1994) Hypoxia and training induced Grossman, A. & Tsagarakis, S. (1989) cortisol, and luteinizing hormone in adaptation of hormonal responses to The hunt for the CIA: factors which the corticotropin-releasing hormone exercise in humans. European Journal of demonstrate corticotrophin-inhibitory stimulation test in eumenorrheic and Applied Physiology 68, 303–309. activity. Journal of Endocrinology 123, amenorrheic athletes. Fertility and Few, J.D. (1974) Effect of exercise on the 169–172. Sterility 55, 276–280. secretion and metabolism of cortisol in Grossman, A., Bouloux, P., Price, P. et al. Howlett, T.A. (1987) Hormonal responses man. Journal of Endocrinology 62, 341–353. (1984) The role of opioid peptides in the to exercise and training: a short review. Filaire, E., Jouanel, P., Colombier, M., hormonal responses to acute exercise in Clinical Endocrinology 26, 723–742. Begue, R.J. & Lac G. (2003) Effects of man. Clinical Science 67, 483–491. Inder, W.J., Hellemans, J., Ellis, M.J. 16 weeks of training prior to a major Häkkinen, K. & Pakarinen, A. (1995) Acute et al. (1995) Elevated basal competition on hormonal and hormonal responses to heavy resistance adrenocorticotropin and evidence for biochemical parameters in young exercise in men and women at different increased central opioid tone in highly elite gymnasts. Journal of Pediatric ages. International Journal of Sports trained male athletes. Journal of Clinical Endocrinology and Metabolism 16, 741–750. Medicine 16, 507–513. Endocrinology and Metabolism 80, Flynn, M.G., Pizza, F.X. & Brolinson, P.G. Häkkinen, K., Pakarinen, A., Alen, M., 244–248. (1997) Hormonal responses to excessive Kauhanen, H. & Komi P.V. (1988a) Inder, W.J., Hellemans, J., Swanney, M.P., training: influence of cross training. Neuromuscular and hormonal Prickett, T.C. & Donald, R.A. (1998a) International Journal of Sports Medicine 18, adaptations in athletes to strength Prolonged exercise increases peripheral 191–196. training in 2 years. Journal of Applied plasma ACTH, CRH, and AVP in male Fry, A.C., & Kramer, W.J. (1997) Resistance Physiology 65, 2406–2412. athletes. Journal of Applied Physiology 85, exercise overtraining and overreaching. Häkkinen, K., Pakarinen, A., Alen, M., 835– 841. Neuroendocrine responses. Sports Kauhanen, H. & Komi, P.V. (1988b) Inder, W.J., Livesey, J.H. & Donald, R.A. Medicine (Auckland, NZ) 23, 106–129. Neuromuscular and hormonal (1998b) Peripheral plasma levels of β- Fry, A.C., Kraemer, W.J., Stone, M.H. responses in elite athletes to two endorphin in alcoholics and highly et al. (1994) Endocrine responses to successive strength training sessions in trained athletes and the relationship overreaching before and after 1 year of 1 day. European Journal of Applied to a measure of central opioid tone. weightlifting. Canadian Journal of Applied Physiology 57, 133–139. Hormone and Metabolic Research 30, Physiology 19, 400–410. Harbach, H., Hell, K., Gramsch, C. et al. 523–525. β Fry, A.C., Kraemer, W.J. & Ramsey, L.T. (2000) -endorphin1–31 in the plasma of Inder, W.J., Swanney, M.P., Donald, R.A., (1998) Pituitary–adrenal–gonadal male volunteers undergoing physical Prickett, T.C. & Hellemans, J. (1998c) responses to high-intensity resistance exercise. Psychoneuroendocrinology 25, The effect of glycerol and desmopressin exercise overtraining. Journal of Applied 551–562. on exercise performance and hydration Physiology 85, 2352–2359. Harte, J.L., Eifert, G.H. & Smith, R. (1995) in triathletes. Medicine and Science in Furell, P.A., Kjær, M., Bach, F.W. & The effects of running and meditation Sports and Exercise 30, 1263–1269. Galbo, H. (1987) Beta-endorphin and on β-endorphin, corticotropin-releasing Jacks, D.E., Sowash, J., Anning, J., adrenocorticotropin response to hormone and cortisol in plasma, and McGloughlin, T. & Andres, F. (2002) supramaximal treadmill exercise in on mood. Biological Psychology 40, Effect of exercise at three exercise trained and untrained males. Acta 251–265. intensities on salivary cortisol. Journal of Physiologica Scandinavica 130, 619–625. Heitkamp, H.C., Huber, W. & Strength and Conditioning Research 16, Gabriel, H., Schwarz, L., Steffens, Scheib, K. (1996) Beta-endorphin and 286–289. G. & Kindermann, W. (1992) adrenocorticotrophin after incremental Jard, S. (1988) Mechanisms of action Immunoregulatory hormones, exercise and marathon runningafemale of vasopressin and vasopressin circulating leucocyte and lymphocyte responses. European Journal of Applied antagonists. Kidney International. subpopulations before and after Physiology 72, 417–424. Supplement 26, S38–S42. endurance exercise of different Hickson, R.C., Hidaka, K., Foster C., Jurimae, J. & Jurimae, T. (2001) Responses intensities. International Journal of Sports Falduto, M.T. & Chatterton, R.T., Jr. of blood hormones to the maximal Medicine 13, 359–366. (1994) Successive time courses of rowing ergometer test in college rowers. Gibson, S., Crosby, S.R. & White, A. (1993) strength development and steroid Journal of Sports Medicine and Physical Discrimination between β-endorphin hormone responses to heavy-resistance Fitness 41, 73–77. and β-lipotrophin in human plasma training. Journal of Applied Physiology 76, Jurimae, J., Jurimae, T. & Purge, P. (2001) using two-site immunoradiometric 663–670. Plasma testosterone and cortisol assays. Clinical Endocrinology 39, Hoffman, J.R., Maresh, C.M., Armstrong, responses to prolonged sculling in male 445–453. L.E. et al. (1994) Effects of hydration state competitive rowers. Journal of Sports Gillies, G.E., Linton, E.A. & Lowry, P.J. on plasma testosterone, cortisol and Science 19, 893–898. (1982) Corticotropin releasing activity of catecholamine concentrations before Kaciuba-Uscilko, H., Porta, S., Nazar, K. the new CRF is potentiated several times and during mild exercise at elevated et al. (1994) Effect of mild psychological by vasopressin. Nature 299, 355–357. temperature. European Journal of Applied stress on physiological responses to Gotshalk, L.A., Loebel, C.C., Nindl, B.C. Physiology and Occupational Physiology exercise in men. Journal of Physiology and et al. (1997) Hormonal responses of 69, 294–300. Pharmacology 45, 429–439. exercise and the hpa axis 229

Kanaley, J.A., Weltman, J.Y., Pieper, K.S., graded exercise in men. Journal of 4000 m altitude. Hormone Research 41, Weltman, A. & Hartman, M.L. (2001) Physiology and Pharmacology 47, 361–371. 225–229. Cortisol and growth hormone responses Lehmann, M., Gastmann, U., Petersen, Martignoni, E., Appenzeller, O., to exercise at different times of day. K.G. et al. (1992) Training–overtraining: Nappi, R.E. et al. (1997) The effects of Journal of Clinical Endocrinology and performance, and hormone levels, after physical exercise at high altitude on Metabolism 86, 2881–2889. a defined increase in training volume adrenocortical function in humans. Keller-Wood, M.E. & Dallman, M.F. (1984) versus intensity in experienced middle- Functional Neurology 12, 339–344. Corticosteroid inhibition of ACTH and long-distance runners. British el-Migdadi, F., Bashir, N., Hasan, Z., secretion. Endocrine Reviews 5, 1–24. Journal of Sports Medicine 26, 233–242. al-Hader, A.A. & Gharaibeh, M. (1996) Kenefick, R.W., Maresh, C.M., Armstrong, Lehmann, M., Knizia, K., Gastmann, U. Exercise at low altitude (Jordan Valley) L.E. et al. (1998) Plasma testosterone and et al. (1993) Influence of 6-week, 6 days causes changes in serum levels of cortisol responses to training-intensity per week, training on pituitary function ACTH, insulin, cortisol and lactate. exercise in mild and hot environments. in recreational athletes. British Journal of Endocrine Research 22, 763–767. International Journal of Sports Medicine 19, Sports Medicine 27, 186–192. Morley, J.E. (1981) The endocrinology 177–181. Lehmann, M., Lormes, W., Opitzgress, A. of the opiates and opioid peptides. Kindermann, W., Schnabel, A., Schmitt, et al. (1997) Training and overtraininga Metabolism 30, 195–209. W.M. et al. (1982) Catecholamines, an overview and experimental results Mujika, I., Chatard, J.C., Padilla, S., growth hormone, cortisol, insulin, and in endurance sports. Journal of Sports Guezennec, C.Y. & Geyssant, A. (1996) sex hormones in anaerobic and aerobic Medicine and Physical Fitness 37, 7–17. Hormonal responses to training and its exercise. European Journal of Applied Lehmann, M., Foster, C., Dickhuth, H.H. tapering off in competitive swimmers: Physiology and Occupational Physiology & Gastmann, U. (1998) Autonomic relationships with performance. 49, 389–399. imbalance hypothesis and overtraining European Journal of Applied Physiology 74, Kirwan, J.P., Costill, D.L., Flynn, M.G. syndrome. Medicine and Science in Sports 361–366. et al. (1988) Physiological responses to and Exercise 30, 1140–1145. Murray, R., Paul, G.L., Seifert, J.G. & Eddy, successive days of intense training in Liu, J.P., Robinson, P.J., Funder, J.W. D.E. (1991) Responses to varying rates of competitive swimmers. Medicine and & Engler, D. (1990) The biosynthesis carbohydrate ingestion during exercise. Science in Sports and Exercise 20, 255–259. and secretion of adrenocorticotropin Medicine and Science in Sports and Exercise Kjær, A. (1993) Vasopressin as a by the ovine anterior pituitary is 23, 713–718. neuroendocrine regulator of anterior predominantly regulated by arginine Nagel, D., Seiler, D. & Franz, H. (1992) pituitary hormone secretion. Acta vasopressin (AVP). Evidence that Biochemical, hematological and Endocrinologica 129, 489–496. protein kinase C mediates the action of endocrinological parameters during Kjær, M., Bangsbo, J., Lortie, G. & Galbo, AVP. Journal of Biological Chemistry 265, repeated intense short-term running H. (1988) Hormonal response to exercise 14 136–14 142. in comparison to ultra-long-distance in humans: influence of hypoxia and Loucks, A.B. & Horvath, S.M. (1984) running. International Journal of Sports physical training. American Journal of Exercise-induced stress responses of Medicine 13, 337–343. Physiology 254, R197–R203. amenorrheic and eumenorrheic runners. Niess, A.M., Fehrenbach, E., Strobel, G. Kraemer, R.R., Blair, S., Kraemer, G.R. Journal of Clinical Endocrinology and et al. (2003) Evaluation of stress & Castracane, V.D. (1989) Effects of Metabolism 59, 1109–1120. responses to interval training at low and treadmill running on plasma β- Loucks, A.B., Mortola, J.F., Girton, L. & moderate altitudes. Medicine and Science endorphin, corticotropin, and cortisol Yen, S.S. (1989) Alterations in the in Sports and Exercise 35, 263–269. levels in male and female 10 K runners. hypothalamic–pituitary–ovarian and O’Connor, P.J., Morgan, W.P. & Raglin, J.S. European Journal of Applied Physiology and the hypothalamic–pituitary–adrenal (1991) Psychobiologic effects of 3 d of Occupational Physiology 58, 845–851. axes in athletic women. Journal of Clinical increased training in female and male Kraemer, W.J., Fleck, S.J., Callister, R. et al. Endocrinology and Metabolism 68, swimmers. Medicine and Science in Sports (1989) Training responses of plasma 402–411. and Exercise 23, 1055–1061. β-endorphin, adrenocorticotropin, and Luger, A., Deuster, P.A., Kyle, S.B. et al. Op’t Eijnde, B. & Hespel, P. (2001) Short- cortisol. Medicine and Science in Sports (1987) Acute hypothalamic–pituitary– term creatine supplementation does not and Exercise 21, 146–153. adrenal responses to the stress of alter the hormonal response to resistance Krieger, D.T., Allen, W., Rizzo, F. & treadmill exercise. Physiologic training. Medicine and Science in Sports Krieger, H.P. (1971) Characterization of adaptations to physical training. New and Exercise 33, 449–453. the normal temporal pattern of plasma England Journal of Medicine 316, Orth, D.N. (1992) Corticotropin-releasing corticosteroid levels. Journal of Clinical 1309–1315. hormone in humans. Endocrine Reviews Endocrinology and Metabolism 32, Lutoslawska, G., Obminski, Z., Krogulski, 13, 164–191. 266–284. A. & Sendecki, W. (1991) Plasma cortisol Pelletier, G., Desy, L., Cote, J. & Vaudry, H. Laatikainen, T.J. (1991) Corticotropin- and testosterone following 19-km and (1983) Immunocytochemical localization releasing hormone and opioid peptides 42-km kayak races. Journal of Sports of corticotropin-releasing factor-like in reproduction and stress. Annals of Medicine and Physical Fitness 31, immunoreactivity in the human Medicine 23, 489–496. 538–542. hypothalamus. Neuroscience Letters 41, Langfort, J., Pilis, W., Zarzeczny, R., Nazar, Marinelli, M., Roi, G.S., Giacometti, M., 259–263. K. & Kaciuba-Uscilko, H. (1996) Effect Bonini, P. & Banfi, G. (1994) Cortisol, Pestell, R.G., Hurley, D.M. & Vandongen, of low-carbohydrate-ketogenic diet on testosterone, and free testosterone in R. (1989) Biochemical and hormonal metabolic and hormonal responses to athletes performing a marathon at changes during a 1000 km 230 chapter 17

ultramarathon. Clinical and Experimental Ronsen, O., Haug, E., Pedersen, B.K. and mental function in sedentary and Pharmacology and Physiology 16, & Bahr, R. (2001a) Increased endurance-trained elderly males. 353–361. neuroendocrine response to a repeated International Journal of Sports Medicine 20, Petraglia, F., Barletta, C., Facchinetti, F. bout of endurance exercise. Medicine and 159–166. et al. (1988) Response of circulating Science in Sports and Exercise 33, 568–575. Suay, F., Salvador, A., Gonzalez-Bono, E. adrenocorticotropin, β-endorphin, Ronsen, O., Holm, K., Staff, H. et al. (2001b) et al. (1999) Effects of competition and its β-lipotropin and cortisol to athletic No effect of seasonal variation in outcome on serum testosterone, cortisol competition. Acta Endocrinologica 118, training load on immuno-endocrine and prolactin. Psychoneuroendocrinology 332–336. responses to acute exhaustive exercise. 24, 551–566. Petrides, J.S., Mueller, G.P., Kalogeras, K.T. Scandinavian Journal of Medicine and Sugimoto, T., Saito, M., Mochizuki, S. et al. et al. (1994) Exercise-induced activation Science in Sports 11, 141–148. (1994) Molecular cloning and functional of the hypothalamic–pituitary–adrenal Schwarz, L. & Kindermann, W. (1989) expression of a cDNA encoding the axis: marked differences in the Beta-endorphin, catecholamines, and human V1b vasopressin receptor. sensitivity to glucocorticoid cortisol during exhaustive endurance Journal of Biological Chemistry 269, suppression. Journal of Clinical exercise. International Journal of Sports 27 088–27 092. Endocrinology and Metabolism 79, Medicine 10, 324–328. Tabata, I., Atomi, Y., Mutoh, Y. & 377–383. Schwarz, L. & Kindermann, W. (1990) Miyashita, M. (1990) Effect of physical Petrides, J.S., Gold, P.W., Mueller, G.P. Beta-endorphin, adrenocorticotropic training on the responses of serum et al. (1997) Marked differences in hormone, cortisol and catecholamines adrenocorticotropic hormone during functioning of the hypothalamic– during aerobic and anaerobic exercise. prolonged exhausting exercise. European pituitary–adrenal axis between groups European Journal of Applied Physiology and Journal of Applied Physiology and of men. Journal of Applied Physiology 82, Occupational Physiology 61, 165–171. Occupational Physiology 61, 188–192. 1979–1988. Sgherza, A.L., Axen, K., Fain, R. et al. (2002) Tabata, I., Ogita, F., Miyachi, M. & Raastad, T., Bjoro, T. & Hallen, J. (2000) Effect of naloxone on perceived exertion Shibayama, H. (1991) Effect of low blood Hormonal responses to high- and and exercise capacity during maximal glucose on plasma CRF, ACTH, and moderate-intensity strength exercise. cycle ergometry. Journal of Applied cortisol during prolonged physical European Journal of Applied Physiology 82, Physiology 93, 2023–2028. exercise. Journal of Applied Physiology 71, 121–128. Silverman, H.G. & Mazzeo, R.S. (1996) 1807–1812. Rahkila, P., Hakala, E., Salminen, K. & Hormonal responses to maximal and Tharp, G.D. & Buuck, R.J. (1974) Adrenal Laatikainen, T. (1987) Response of submaximal exercise in trained and adaptation to chronic exercise. Journal of plasma endorphins to running exercises untrained men of various ages. Journals Applied Physiology 37, 720–722. in male and female endurance athletes. of Gerontology. Series A, Biological Thoren, P., Floras, J., Hoffman, P. & Seals, Medicine and Science in Sports and Exercise Sciences and Medical Sciences 51, D. (1990) Endorphins and exercise: 19, 451–455. B30–B37. physiological mechanisms and clinical Rich, P.A., Villani, R., Fulton, A. et al. Singh, A., Petrides, J.S., Gold, P.W., implications. Medicine and Science in (1992) Serum cortisol concentration Chrousos, G.P. & Deuster, P.A. (1999) Sports and Exercise 22, 417–428. and testosterone to cortisol ratio in elite Differential hypothalamic–pituitary– Thuma, J.R., Gilders, R., Verdun, M. & prepubescent male gymnasts during adrenal axis reactivity to psychological Loucks, A.B. (1995) Circadian rhythm training. European Journal of Applied and physical stress. Journal of Clinical of cortisol confounds cortisol responses Physiology and Occupational Physiology Endocrinology and Metabolism 84, to exercise: implications for future 65, 399–402. 1944–1948. research. Journal of Applied Physiology 78, Rivier, C. & Vale, W. (1983) Modulation Smoak, B., Deuster, P., Rabin, D. & 1657–1664. of stress-induced ACTH release by Chrousos, G. (1991) Corticotropin- Utter, A.C., Kang, J., Nieman, D.C. et al. corticotropin-releasing factor, releasing hormone is not the sole (1999) Effect of carbohydrate ingestion catecholamines and vasopressin. factor mediating exercise-induced and hormonal responses on ratings of Nature 305, 325–327. adrenocorticotropin release in humans. perceived exertion during prolonged Robertson, G.L. & Athar, S. (1976) The Journal of Clinical Endocrinology and cycling and running. European Journal of interaction of blood osmolality and Metabolism 73, 302–306. Applied Physiology and Occupational blood volume in regulating plasma Snegovskaya, V. & Viru, A. (1993) Steroid Physiology 80, 92–99. vasopressin in man. Journal of Clinical and pituitary hormone responses to Uusitalo, A.L., Huttunen, P., Hanin, Y., Endocrinology and Metabolism 42, rowing: relative significance of exercise Uusitalo, A.J. & Rusko, H.K. (1998) 613–620. intensity and duration and performance Hormonal responses to endurance Rolandi, E., Reggiani, E., Franceschini, R. level. European Journal of Applied training and overtraining in female et al. (1985) Comparison of pituitary Physiology and Occupational Physiology athletes. Clinical Journal of Sport Medicine responses to physical exercise in athletes 67, 59–65. 8, 178–186. and sedentary subjects. Hormone Stewart, P.M. (2003) The adrenal cortex. Vale, W., Spiess, J., Rivier, C. & Rivier, J. Research 21, 209–213. In: Williams Textbook of Endocrinology (1981) Characterization of a 41-residue Ronkainen, H.R., Parkarinen, A.J. & (Larsen, P., Kronenberg, H., Melmed, S. ovine hypothalamic peptide that Kauppila, A.J. (1986) Adrenocortical & Polonsky, K., eds.). Saunders, stimulates secretion of corticotropin function of female endurance runners Philadelphia: 491–551. and β-endorphin. Science 78, and joggers. Medicine and Science in Struder, H.K., Hollmann, W., Platen, P. 1394–1397. Sports and Exercise 18, 385–389. et al. (1999) Neuroendocrine system Vanhelder, W.P., Radomski, M.W., Goode, exercise and the hpa axis 231

R.C. & Casey, K. (1985) Hormonal and de Vries, W.R., Bernards, N.T., de Rooij, Wittert, G.A., Stewart, D.E., Graves, M.P. metabolic response to three types of M.H. & Koppeschaar, H.P. (2000) et al. (1991) Plasma corticotrophin exercise of equal duration and external Dynamic exercise discloses different releasing factor and vasopressin work output. European Journal of Applied time-related responses in stress responses to exercise in normal man. Physiology and Occupational Physiology hormones. Psychosomatic Medicine 62, Clinical Endocrinology 35, 311–317. 54, 337–342. 866–872. Wittert, G.A., Livesey, J.H., Espiner, E.A. Vasankari, T.J., Kujala, U.M., Heinonen, Wade, C.E. & Claybaugh, J.R. (1980) & Donald. R.A. (1996) Adaptation of O.J. & Huhtaniemi, I.T. (1993) Effects Plasma renin activity, vasopressin the hypothalamopituitary adrenal axis of endurance training on hormonal concentration, and urinary excretory to chronic exercise stress in humans. responses to prolonged physical exercise responses to exercise in men. Journal of Medicine and Science in Sports and Exercise in males. Acta Endocrinologica 129, Applied Physiology 49, 930–936. 28, 1015–1019. 109–113. Whitnall, M.H., Smyth, D. & Gainer, H. Yanovski, J.A., Yanovski, S.Z., Boyle, A.J. Verde, T., Thomas, S. & Shephard, R.J. (1987) Vasopressin coexists in half of the et al. (2000) Hypothalamic–pituitary– (1992) Potential markers of heavy corticotropin-releasing factor axons adrenal axis activity during exercise training in highly trained distance present in the external zone of the in African American and Caucasian runners. British Journal of Sports Medicine median eminence in normal rats. women. Journal of Clinical Endocrinology 26, 167–175. Neuroendocrinology 45, 420–424. and Metabolism 85, 2660–2663. Chapter 18

Influence of Energy Availability on Luteinizing Hormone Pulsatility and Menstrual Cyclicity

ANNE B. LOUCKS

Introduction most cases of menstrual disorders in anaerobic sports, too. Over the past 30 years three types of scientific Menstrual disorders caused by factors other studies have been conducted to investigate the high than athletic training, including pregnancy, lacta- prevalence of menstrual disorders in athletes. Sur- tion, eating disorders, pituitary tumors, hyperan- veys have found the highest prevalences in compet- drogenism, polycystic ovary syndrome, depression itive endurance, aesthetic and weight-class sports and various organic diseases, are beyond the scope and have led to several hypotheses about the cause of this review, even though some (such as polycystic of these disorders. Subsequent observational stud- ovary syndrome and anorexia nervosa) may be ies comparing hormone measurements in amenor- disproportionately represented amongst athletes rheic and eumenorrheic athletes contradicted most due to the self-selection of affected women into of those hypotheses. Then prospective experiments activities in which hyperandrogenism or low body controlling the administration of the two remain- weight, respectively, offer a competitive advantage. ing hypothetical causal factors, exercise stress and The subject of this review is the disruption of repro- energy availability, have shown that low energy ductive function by athletic training itself, the mech- availability (defined as dietary energy intake minus anism of which has been traced to the failure of exercise energy expenditure) disrupts reproduct- female athletes to sufficiently increase their dietary ive function in physically active women and that intake in compensation for their exercise energy exercise has no suppressive effect on reproductive expenditure, and which may be prevented by dietary function beyond the impact of its energy cost on reform without any moderation of their exercise energy availability. regimen. In this research, most observational studies of menstrual disorders in athletes have focused on Regulation of the reproductive system participants in aerobic sports, and most prospective experiments have employed aerobic exercise train- The length and regularity of menstrual cycles ing to investigate the mechanism of these disorders. vary considerably across the general population of Therefore, very little information is available con- women as well as during an individual woman’s cerning whatever differences there may be between reproductive years (Treloar et al. 1967). The median the influences of aerobic and anaerobic exercise on length of the menstrual cycle amongst North the prevalence and mechanism of menstrual dis- American adolescents is 33 days, and the median orders. However, since the prevalence of menstrual standard deviation of variations in the length from disorders in female bodybuilders is high, and since month to month is 7 days. By the age of 20 years, dietary restriction is common practice in bodybuild- the median length and standard deviation have ing, low energy availability probably accounts for decreased to 29 and 4 days, respectively. These 232 influence of energy availability 233

numbers continue to decline slowly until the age until the next menses is known as the luteal phase of of 40 years when both numbers begin to increase the cycle. greatly during the years preceding menopause, Estrogen and progesterone have profound influ- when menstruation ceases permanently. ences on the uterine endometrium as well as many Moreover, the average age of menarche (i.e. the other tissues in the body. Estrogen stimulates endo- first menstrual cycle) has declined dramatically metrial proliferation and progesterone causes it to over the past 150 years in all Western societiesa become highly vascularized. These are necessary, from an average of 17 years to 12.4 years (Styne & hospitable conditions for the successful implanta- Grumbach 1991; Anderson, S.E. et al. 2003). What tion of a fertilized egg. If no fertilization occurs reactivates the reproductive system at puberty is after several days, the ability of the corpus luteum to unknown, just as what deactivates it in infancy and secrete progesterone becomes exhausted, the struc- again at menopause are also unknown. tural integrity of the endometrium collapses, and The regulation of the menstrual cycle by the menstruation ensues. Under normal circumstances, hypothalamic–pituitary–ovarian (HPO) axis includes the secretory capacity of the corpus luteum is sus- both negative and positive feedback mechanisms, as tained long enough for the rapidly dividing cells well as inputs from the central nervous system and derived from a fertilized egg to become implanted other systems at various levels within the axis. The in the endometrium 6 or 7 days after fertilization. glands of the HPO axis secrete their hormones If the secretory capacity of the corpus luteum is rhythmically. Indeed, the secretion of gonadotropin- exhausted too soon, the endometrium sloughs off releasing hormone (GnRH) pulses into the pituitary before implantation can occur. The likelihood of portal blood by neurons within the hypothalamus this increases when the luteal phase is shorter than of the brain must occur at an optimal frequency if 10 days. Thus, infertility can result from either the the axis as a whole is to function normally. These failure of the ovary to release an egg for fertilization GnRH pulses stimulate the pulsatile secretion of or the failure of a fertilized egg to become properly luteinizing hormone (LH) from the pituitary gland implanted into the endometrium. into the blood. By sampling blood at 10 min inter- vals for 24 h, the frequency and amplitude of LH Characterization of female athletes pulses can be studied. In response to the proper pulsatile and monthly The ovarian axis rhythmic stimulation by LH and pituitary follicle- stimulating hormone (FSH), clusters of ovarian cells Amenorrheic athletes produce low levels of estrogen (‘ovarian follicles’) grow and secrete increasing and progesterone every day indicating a complete amounts of estrogen. Gradually, too, one of the fol- absence of follicular development, ovulation, and licles becomes dominant and eventually the rising luteal function (Fig. 18.1, AA) (Loucks et al. 1989). By concentration of estrogen in the blood exerts a contrast, even the most eumenorrheic competit- positive feedback on LH secretion. In response, the ive athletes display extended follicular phases and pituitary gland secretes a surge of LH into the blood, abbreviated luteal phases with blunted progesterone causing the dominant follicle to rupture, thereby concentrations (Fig. 18.1, CA) compared to eumen- releasing an egg cell for fertilization. The remaining orrheic sedentary women (Fig. 18.1, CS). Similar cells of the dominant follicle then undergo rapid observations have been made in eumenorrheic chemical and morphological changes to form the women running recreationally as little as 12 miles corpus luteum, a body that begins secreting both (20 km) per week (Ellison & Lager 1986; Broocks progesterone and estrogen into the blood. The inter- et al. 1990; Pirke et al. 1990; De Souza et al. 1998). val during which a dominant follicle develops, from The proximal cause of ovarian dysfunction in menses until ovulation, is known as the follicular amenorrheic and eumenorrheic athletes is a disrup- phase of the menstrual cycle. The interval during tion of the pulsatile rhythm of LH concentrations in which the corpus luteum is active from ovulation the blood, upon which ovarian function critically 234 chapter 18

Fig. 18.1 Urinary estrone- glucuronide (E G), an estradiol 45 1 CS metabolite, and pregnanediol- –1 CA glucuronide (PdG), a progesterone 30 metabolite, over an entire menstrual AA cycle in cyclic sedentary women (CS) 15 and cyclic athletes (CA), and over G ng·mg CR

1 an entire month in amenorrheic E athletes (AA). The mass of each 0 µ metabolite (ngE1 and PdG) excreted 6 in overnight urine samples was

–1 normalized to the mass (mg) of 4 creatinine (CR) excreted in the same samples. The black and open bars at the bottom of the figure indicate the 2 days of menses in the CS and and

PdG µ g·mg CR CA women, respectively, at the 0 beginning and the end of the cycle –16–80 8 16 of observation. (From Loucks et al. Day from significant increase in PdG 1989.)

depends (Veldhuis et al. 1985; Yahiro et al. 1987; ence of follicular and luteal suppression and ano- Loucks et al. 1989; Laughlin & Yen 1996). In a vulation appears to be extremely high. Repeated eumenorrheic sedentary young woman, the 24-h endocrine workups have found that 79% of eumen- LH profile in the early follicular phase is charac- orrheic female runners were luteally suppressed or terized by regular, high frequency pulses of low anovulatory in at least 1 month out of 3 (De Souza amplitude (Fig. 18.2, CS) (Loucks et al. 1989). During et al. 1998). sleep, the frequency slows and the amplitude increases. Eumenorrheic athletes display a slower, Other endocrine axes but still regular rhythm of larger pulses (Fig. 18.2, CA). Amenorrheic athletes display even fewer Mildly elevated resting serum cortisol levels in pulses, at irregular intervals (Fig. 18.2, AA). amenorrheic and eumenorrheic athletes (Ding et al. Experimental administration of GnRH has de- 1988; Loucks et al. 1989; De Souza et al. 1991, 1994; monstrated that the disruption of LH pulsatility in Laughlin & Yen 1996) encouraged the hypothesis amenorrheic and eumenorrheic athletes is caused that their reproductive disorders might be due to by a disruption of GnRH pulsatility and not by a the stress of exercise. Since cortisol is a glucoregulat- pituitary disorder (Veldhuis et al. 1985; Loucks et al. ory hormone activated by low blood glucose levels, 1989). Therefore, the neuroendocrine mechanisms however, these elevated cortisol levels might also by which the GnRH pulse generator can be dis- be explained as part of the physiological response to rupted are the focus of much research. chronic energy deficiency. The prevalence of amenorrhea in endurance, Indeed, more extensive endocrine observations aesthetic and weight-class sports can be as much as found that amenorrheic athletes also display low 10 times higher than in the general population (Otis levels of plasma glucose (Laughlin & Yen 1996), et al. 1997). The less severe disorders of ovarian insulin (Laughlin & Yen 1996), insulin-like growth function (follicular and luteal suppression and factor I (IGF-I) (Zanker & Swaine 1998a), insulin-like anovulation) may display no menstrual symptoms growth factor I/insulin-like growth factor binding at all so that affected women are entirely unaware protein-1 (IGF-I/IGFBP-1) (an index of IGF-I bio- of their condition until they undergo an endocrine availability) (Laughlin & Yen 1996), leptin (Laughlin workup. Amongst eumenorrheic athletes, the incid- & Yen 1997; Thong et al. 2000) and triiodothyronine influence of energy availability 235

Yen 1996) in addition to the mildly elevated cortisol levels (Loucks et al. 1989; De Souza et al. 1991;

) Laughlin & Yen 1996). All these abnormalities are

–1 20 * signs of chronic energy deficiency. * * * * CS * * * * * Compared to eumenorrheic sedentary women,

LH (IU·L 10 luteally suppressed eumenorrheic athletes also dis- play low levels of insulin (Laughlin & Yen 1996),

0 leptin (Laughlin & Yen 1997) and T3 (De Souza et al. 2003), and elevated levels of GH (Laughlin & Yen * 1996) and cortisol (Loucks et al. 1989; Laughlin & * ) * Yen 1996), but the magnitudes of these signs of –1 20 * * * * * CA energy deficiency are less extreme than in amenor- rheic athletes.

LH (IU·L 10

Energy intake and expenditure 0 Because of large variances between individuals, we were unable to distinguish between the dietary

) and exercise habits and histories of the amenorrheic

–1 20 * and eumenorrheic athletes that we studied (Loucks * AA * et al. 1989, 1992). At that time, it did not occur to us

LH (IU·L 10 * * to estimate each individual’s energy availability (dietary energy intake minus exercise energy expend- 0 iture). Both groups of athletes reported their similar body weights to be stable, despite dietary energy intakes indistinguishable from those of sedentary * ) women (Loucks et al. 1989). That is, their dietary

–1 20 energy intakes were much less than would be * AA * * expected for their level of physical activity, a fea-

LH (IU·L 10 * * ture commonly observed amongst athletic women (Drinkwater et al. 1984; Marcus et al. 1985; Nelson 0 et al. 1986; Kaiserauer et al. 1989; Myerson et al. 1991; 08001600 2400 0800 Laughlin & Yen 1996). Clock hours Recently, extensive observational data on the Fig. 18.2 The 24-h pulsatile rhythms of luteinizing energy and carbohydrate intakes of athletes in hormone (LH), expressed as international units per liter many sports have been compiled (Burke et al. 2001). (IU·L–1), for a cyclic (regularly menstruating) sedentary If these data are to be believed, one observation is woman (CS), a cyclic athlete (CA) and two amenorrheic particularly noteworthy: with the notable exception athletes (AA). Among amenorrheic athletes, the extensively quiescent, irregular pulsatile rhythms differed of cross-country skiers, female athletes consume ⎯ markedly. Pulsatile rhythms were considerably less ~ 30% less energy and carbohydrates normalized variable among cyclic sedentary women and athletes. for body weight⎯than do male athletes. (From Loucks et al. 1989.) The repeatedly reported combination of a stable body mass and an unexpectedly low dietary energy

(T3) (Myerson et al. 1991; Loucks et al. 1992; Zanker intake in female athletes is very controversial. Many & Swaine 1998a, 1998b), as well as low resting investigators have been skeptical of the dietary metabolic rates (Myerson et al. 1991). They also dis- records of female athletes, because studies com- play elevated growth hormone (GH) (Laughlin & paring such data to estimations or measurements 236 chapter 18

of their energy expenditure have repeatedly found chronic energy and carbohydrate deficiency result- apparently huge negative energy balances, some ing in the mobilization of fat stores, the slowing of exceeding 4 MJ·day–1 in athletes with stable body metabolic rate and a decline in glucose utilization, weights (Mulligan & Butterfield 1990; Wilmore et al. with more extreme abnormalities in amenorrheic 1992; Edwards et al. 1993; Beidleman et al. 1995; athletes and less extreme abnormalities in eumenor- Trappe et al. 1997; Hill & Davies 2002). Such large rheic athletes (Myerson et al. 1991; Loucks et al. 1992; discrepancies have been interpreted as indicating Jenkins, P.J. et al. 1993; Laughlin & Yen 1996, 1997; not that female athletes are undernourished but De Souza et al. 2003). So, while some might suggest rather that they grossly underreport their actual that lower energy and carbohydrate intakes would dietary intake. In support of this allegation, invest- be appropriate for women if their energy and carbo- igators have cited certain other special subpopula- hydrate expenditures were less than those of men, tions that have been found to underreport (Wilmore biochemical data demonstrate that female athletes et al. 1992; Edwards et al. 1993). Actually, underre- are, indeed, chronically energy deficient. porting of dietary intake is common in all popula- tions (Mertz et al. 1991) and a meta-analysis of Clinical concerns studies comparing dietary assessments to measure- ments of energy expenditure by doubly labeled The stability of body weight in athletes whose water found that women do not underreport more energy intake is estimated to be much less than than men (Trabulsi & Schoeller 2001). their energy expenditure has been attributed to Other investigators have questioned the methods an increase in metabolic ‘efficiency’ (Westerterp & used to measure energy intake and expenditure, Saris 1991; Westerterp et al. 1992), but ‘efficiency’ is and, indeed, the study that found virtually identical not the appropriate concept to apply to patholo- energy intakes in female and male cross-country gical adjustments to chronic energy deficiency. The skiers took extraordinary pains to achieve accurate oxidation of scarce metabolic fuels in the muscular measurements of energy intake (Sjodin et al. 1994). work of locomotion makes these fuels unavailable As a result of such concerns, quantitative criteria for immune function, growth, tissue turnover, re- have been developed to assess whether reported productive development and function, and other energy intakes in studies of various numbers of sub- important physiological functions. For example, jects over various lengths of time pass what might 50% of peak bone mass is deposited during adoles- be called the laugh test (Goldberg et al. 1991). cence, but the impairment of this process has led to On the other hand, a stable body weight is not some young amenorrheic athletes having the bone necessarily proof of energy sufficiency, because densities of 60-year-old women. Recently, a case behavior modification and endocrine-mediated study has been published documenting the 20-year alterations in resting metabolic rate can counteract history of clinical osteoporosis in an amenorrheic the potential influences of dietary energy excess or athlete (Zanker et al. 2004). deficiency on body mass (Leibel et al. 1995). Energy Bone is continuously remodeled at millions of intake and energy expenditure are also very local sites by coupled processes of bone resorption difficult to measure reliably, even with doubly by osteoclast cells followed by bone formation by labeled water. Considering the lack of confidence in osteoblast cells. Normally, an increase in osteoclast studies of energy balance in athletes, therefore, it is activity stimulates an increase in osteoblast activity surprising that investigations of energy intake and and bone resorption and formation are said to be expenditure in female athletes have not included coupled, but under pathological circumstances they biochemical measurements, because underreport- become uncoupled. When the activity of osteoclasts ing does not account for biochemical evidence of exceeds that of osteoblasts, the resulting net bone energy deficiency. As described above, metabolic loss is irreversible, because once they have left a substrates and hormones measured in amenorrheic bone remodeling site osteoblasts do not return to and eumenorrheic athletes tell a consistent story of finish the job of refilling it. influence of energy availability 237

At menopause, bone density declines because number of menses that they have missed (Drinkwater the normal suppression of osteoclast activity by et al. 1990), resulting in an increased rate of stress estrogen is released, thereby increasing the rate of fractures (Warren et al. 2003). Oral contraceptives bone resorption while bone formation is unaffected have been recommended to restore bone loss in (Marcus et al. 1996) or increased (Nielsen et al. 2004). amenorrheic athletes (Otis et al. 1997; Anderson, S.J. In younger women by contrast, energy deficiency et al. 2000), but this treatment has failed to increase reduces the rate of bone formation by suppress- bone density in young amenorrheic women by ing bone growth factors that stimulate osteoblast more than a few percent (Gulekli et al. 1994; Keen activity, and when this energy deficiency is severe & Drinkwater 1997; Zanker et al. 2004). Restoring enough to induce amenorrhea, the suppression of menses has been more successful for increasing osteoclast activity by estrogen is also released, bone density in some women, but recovered amen- thereby increasing the rate of bone resorption. Thus, orrheic athletes have not kept pace with the bone most studies of anorexia nervosa patients have growth of their eumenorrheic peers (Warren et al. found both a reduction in bone formation as well as 2003). In recovered anorexia nervosa patients who an increase in bone resorption (Stefanis et al. 1998; had been clinically well for 14–23 years, bone dens- Caillot-Augusseau et al. 2000; Hotta et al. 2000; Audi ity in the femur remained 14% lower than in con- et al. 2002; Gordon et al. 2002), although a few have trol subjects (Hartman et al. 2000). The inability found only reduced formation (Soyka et al. 1999) or of estrogen to restore lost bone in amenorrheic ath- increased resorption (Lennkh et al. 1999). Refeeding letes and anorexia nervosa patients may indicate a anorexia nervosa patients has been found either to potentially reversible suppression of bone forma- increase bone formation alone (Stefanis et al. 1998; tion caused by a continuing deficiency of nutritional

Hotta et al. 2000) or to increase bone formation while growth factors such as IGF-I, T3 and insulin, or it also reducing resorption (Caillot-Augusseau et al. may indicate that the bone loss is irreversible due to 2000; Heer et al. 2002; Soyka et al. 2002). the uncoupling of bone resorption and formation. Fewer studies of bone turnover have been con- IGF-I and IGF-I/IGFBP-3 (another index of IGF-I ducted in amenorrheic athletes. Unfortunately, the bioavailability) have been found to predict spinal results of these studies have been less consistent and femoral neck bone mass in gymnasts and runners than those of anorexia nervosa patients, with results (Maddalozzo & Snow 2000). When anorexia ner- showing no difference in either formation or resorp- vosa patients with osteopenia were treated with a tion (Hetland et al. 1993; Stacey et al. 1998), reduced combination of oral contraceptives and recombin- formation (Okano et al. 1995), and reduced forma- ant IGF-I their bone density did increase more than tion and resorption but with resorption outweigh- in a similar group who were administered oral con- ing formation (Zanker & Swaine 1998a). Therefore, traceptives alone (Grinspoon et al. 2000, 2002). The more studies of bone turnover in amenorrheic ath- benefit was only a few percent, however, and it letes need to be conducted before we can speak may have been illusory because the latter group confidently about whether the mechanism of bone inexplicably reduced their dietary intake by 40% loss in amenorrheic athletes differs from that in during the study resulting in a 15% decline in anorexia nervosa patients. Such studies will need to their endogenous IGF-1 production. Therefore, the be interpreted with care, however, since markers of superiority of the combined treatment has not been bone turnover are systemic in nature and may not established. This treatment has not yet been tested reveal local changes in bone turnover in the lumbar in amenorrheic athletes. vertebrae and other trabecular sites where bone loss Skeletal demineralization and osteoporotic frac- commonly occurs in amenorrheic athletes, espe- tures are the most alarming clinical consequence cially if bone density is simultaneously increasing of athletic amenorrhea. Osteoporotic spinal frac- elsewhere, such as in the heel. tures result in permanent disabilities and chronic In young athletic women, bone mineral density pain. Osteoporosis in a 20-year-old athlete is a dis- declines by as much as 20% in proportion to the aster. Osteopenia is a disaster waiting to happen. 238 chapter 18

Therefore, early detection and intervention are crit- Proposed mechanisms of reproductive ical for minimizing permanent skeletal damage. disturbances Physicians should not defer treatment until after osteopenia and osteoporosis are manifest. The As mentioned above, many competing hypotheses American College of Sports Medicine has published about the cause and mechanism of reproductive dis- a position stand on the female athlete triad as a orders in athletes have been offered over the years. syndrome requiring prompt intervention to prevent The three most prominent hypotheses are discussed chronic undernutrition from inducing reproductive below. disorders and skeletal demineralization (Otis et al. 1997). The American Academy of Pediatrics has Body composition published a similar warning (Anderson, S.J. et al. 2000). Since skeletal demineralization must pro- The body composition hypothesis held that the ceed for many months before a reduction in bone ovarian axis is disrupted when the amount of density is measurable, repeated measures of bio- energy stored in the body as fat declines below a chemical markers of bone turnover, which respond critical level (Frisch & McArthur 1974). Outside the immediately to undernutrition and hypoestrogen- research community this has been the most widely ism should be considered as early indicators of publicized explanation for menstrual disorders in skeletal demineralization. athletes, but it has been the least widely believed Of course, long before reductions in bone density explanation within the research community. Des- are measurable, the first clinical consequence of pite early associations of menarche and amenorrhea amenorrhea is infertility, since amenorrheic women with body composition (Frisch & McArthur 1974), are not developing egg cells that can be fertilized. later observations of athletes did not consistently Eumenorrheic physically active women with short verify this association (e.g. Laughlin & Yen 1996), luteal phases and low progesterone levels may also nor did they find the appropriate temporal relation- be at risk for infertility due to failures of implanta- ship between changes in body composition and tion. Paradoxically, irregularly cycling and oligo- menstrual function (for reviews see Scott & Johnston menorrheic athletes may be at increased risk for 1982; Loucks & Horvath 1985; Sinning & Little 1987; unintended pregnancies if they do not use con- Bronson & Manning 1991). Eumenorrheic and traceptives, because their day of ovulation is less amenorrheic athletes span a common range of body predictable than that of eumenorrheic women. composition (Loucks et al. 1984; Sanborn et al. 1987; Another consequence of the hypoestrogenism in Laughlin & Yen 1997) leaner than most eumenor- amenorrheic athletes is impaired endothelium- rheic sedentary women. Among women distance dependent arterial vasodilation (Hoch et al. 2003a, runners, energy balance is a better predictor of 2003b), which reduces perfusion of working muscle estradiol levels (r = 0.88) than are body mass index and increases the risk of developing cardiovascular (BMI) (r = 0.42) or percent body fat (r = 0.48) (Zanker disease, and which is restored by estrogen replace- & Swaine 1998c). Furthermore, if the growth and ment therapy and by the return of regular menstrual sexual development of young animals are blocked cycles (Hoch et al. 2003b). Impaired skeletal muscle by dietary restriction, normal LH pulsatility re- oxidative metabolism has also been reported in sumes only a few hours after ad libitum feeding is amenorrheic athletes, suggesting that they may be permitted, long before any change in body mass or at a physiological disadvantage in their ability to composition (Bronson 1986). Further evidence that perform repeated exercise bouts compared to their body composition does not play a causal role in the eumenorrheic competitors (Harber et al. 1998). In mechanism of menstrual disorders was reported in another report of estrogen-deficiency symptoms, an experiment on severely obese women (body 75% of amenorrheic athletes reported vaginal dry- weight ~ 130 kg; BMI ~ 47) (Di Carlo et al. 1999). ness compared to only 7% of eumenorrheic athletes Surgical reduction of their stomach volume reduced (Hammar et al. 2000). the amount of food that these patients could eat, re- influence of energy availability 239

sulting in rapid weight loss and amenorrhea while cortisol negative feedback (Rivier & Rivest 1991; the patients were still overweight (body weight Chrousos & Gold 1992). Furthermore, early experi- ~ 97 kg; BMI ~ 35). Thus, the body composition ments by Selye (1939) induced anestrus and ovarian hypothesis confused causation with correlation, atrophy in rats by abruptly forcing them to run which occurs in athletes because a lean body com- strenuously for prolonged periods. Others induced position and menstrual disorders are both conse- anestrus in rats by forced swimming (Asahina et al. quences of low energy availability. 1959; Axelson 1987), by forced running (Chatterton Nevertheless, interest in the body composition et al. 1990), and by requiring animals to run farther hypothesis was rejuvenated several years ago with and farther for smaller and smaller food rewards the discovery of the hormone leptin. Synthesized (Manning & Bronson 1989, 1991). Elevated cortisol and secreted by adipose tissue, leptin was origin- levels in such studies were interpreted as signs of ally thought to communicate information about fat stress, and the induced reproductive disorders were stores (Maffei et al. 1995). Later reports of leptin widely interpreted as evidence that ‘exercise stress’ varying profoundly before any changes in adiposity had a counter-regulatory influence on the female in response to fasting (Kolaczynski et al. 1996a; reproductive system. These experiments induced Weigle et al. 1997), dietary restriction (Weigle et al. extreme activations of the adrenal axis, however, 1997), refeeding after dietary restriction (Kolaczynski raising cortisol concentrations by several hundred et al. 1996a; Jenkins, A.B. et al. 1997) and overfeeding percent, in contrast to the mild 10–30% elevations (Kolaczynski et al. 1996b) led to the hypothesis that seen in amenorrheic athletes (Loucks et al. 1989; leptin signals information about dietary intake, Laughlin & Yen 1996), hypothalamic amenorrhea particularly carbohydrate intake (Jenkins, A.B. et al. patients (Berga et al. 1989) and anorexia nervosa 1997). Then leptin was found to be regulated by the patients (Gold et al. 1986). Whether such mild eleva- tiny flux of glucose through the hexosamine biosyn- tions in cortisol influence the GnRH pulse generator thesis pathway in muscle and adipose tissue (Wang was entirely speculative. et al. 1998; Rossetti 2000; Obici et al. 2002; Ravussin Only one experiment had successfully employed 2002). Since then, we have shown that the diurnal exercise to induce menstrual disorders in eumenor- rhythm of leptin depends actually on energy avail- rheic women. That experiment (Bullen et al. 1985) ability or more specifically on carbohydrate avail- imposed a high volume of aerobic exercise abruptly, ability (Hilton & Loucks 2000). in imitation of Selye (1939). It caused a large propor- tion of menstrual disorders in the first month, and an even larger proportion in the second. The dis- Exercise stress orders were more prevalent in a subgroup fed a con- The stress hypothesis held that the stress of exercise trolled weight-loss diet than in another subgroup activates the adrenal axis, which disrupts the GnRH fed for weight maintenance, but even the weight pulse generator by various mechanisms. To be maintenance subgroup may have been underfed, meaningfully distinct from the energy availability since body mass is an unreliable indicator of energy hypothesis, the stress hypothesis further implied balance (Leibel et al. 1995). that the adrenal axis is activated by some aspect of The first cracks in the stress hypothesis appeared exercise other than its energy cost. when glucose administration during exercise was Certainly, there are central and peripheral mech- found to prevent the cortisol response to exercise in anisms by which the adrenal axis can disrupt the both rats (Slentz et al. 1990) and in men (Tabata et al. ovarian axis. Considerable animal research has 1991) in the laboratory. This finding was later shown that GnRH neurons are disturbed by activa- confirmed in a field experiment when the cortisol tion of the hypothalamic–pituitary–adrenal axis response to a strenuous, prolonged hill walk was via pathways involving corticotropin-releasing hor- prevented by eating larger meals (Ainslie et al. mone (CRH) and endogenous opioid and proopi- 2003). Then LH pulsatility was disrupted in habitu- omelanocortin-derived peptides, or by increased ally sedentary, eumenorrheic women through a 240 chapter 18

combination of dietary restriction and exercise 90 90 (Williams et al. 1995), but this did not resolve the ambiguity about whether exercise stress or energy 45 45 availability disrupts the reproductive system in B 0 0 exercising women. –1 Since all previous animal and human investiga- –45 –45 tions of the influence of the ‘activity stress paradigm’ IEAIE A 90 90 on reproductive function had confounded the stress kcal·kg FFM of exercise with the stress of forcing animals to exer- 45 45 cise and/or with energy deficiency, there was, in D fact, no unconfounded experimental evidence that 0 0 the stress of exercise, independent of its energy cost, disrupts reproductive function in voluntarily exer- –45 –45 S X cising women. Therefore, we conducted an experi- ment to determine the independent effects of Fig. 18.3 Experimental design. Dietary energy intake (I) exercise stress and energy availability on LH pul- and exercise energy expenditure (E) were controlled to = –1 –1 satility (Loucks et al. 1998). We defined, measured achieve balanced (B 45 kcal·kg FFM ·day ) and deprived (D = 10 kcal·kg FFM–1·day–1) energy availability and controlled energy availability operationally as (A = I − E) treatments. Deprived energy availability was dietary energy intake minus exercise energy expend- achieved by dietary restriction alone in sedentary women iture. Needing an operational definition of exercise (S) and by exercise energy expenditure alone in exercising stress, however, we were confronted by a funda- women (X) (1 kcal = 4.18 kJ). (From Loucks et al. 1998.) mental problem. Despite 60 years of research on responses to exercise stress, we could find no object- ive definition of exercise stress itself. So, we defined increasing their dietary energy intake in compensa- exercise stress as everything associated with exercise, tion for their exercise energy expenditure. After except its energy cost. 4 days of these treatments, we drew blood samples Figure 18.3 shows the experimental design in from the subjects at 10-min intervals for 24 h to which we assigned habitually sedentary women of assess LH pulsatility. normal body composition to sedentary or exercising The results in Fig. 18.4 show that exercise stress groups and then administered balanced (45 kcal had no suppressive effect on LH pulse frequency, [188 kJ]·kg fat free mass [FFM]–1·day–1) and restricted whereas low energy availability suppressed LH 10 kcal [42 kJ]·kg FFM–1·day–1) energy availability pulse frequency, regardless of whether the low treatments to them in random order under con- energy availability was caused by dietary energy trolled conditions in the laboratory. The energy restriction alone or by exercise energy expenditure availability of the balanced sedentary group was alone. We also obtained similar results (not shown) achieved by feeding them 45 kcal [188 kJ]·kg when half of the reduction in energy availability FFM–1·day–1 of energy in the form of a clinical dietary was caused by dietary energy restriction and half by product. In the other trial, their energy availabil- exercise energy expenditure. Low energy availabil- ity was reduced by dietary restriction alone. The ity also suppressed T3, insulin, IGF-I and leptin exercising group expended 30 kcal [125 kJ]·kg (Hilton & Loucks 2000) while increasing GH and FFM–1·day–1 of energy in supervised exercise on a cortisol in a pattern very reminiscent of amenorrh- treadmill in the laboratory. Their 10 kcal [42 kJ]·kg eic and luteally suppressed eumenorrheic athletes. FFM–1·day–1 restricted energy availability was Unexpectedly, the effects of low energy availabil- achieved by feeding them a dietary energy intake ity on LH pulse frequency and on the metabolic hor- of 40 kcal [167 kJ]·kg FFM–1·day–1 similar to the mones were smaller in the exercising women than in sedentary women in their balanced treatment. Their the dietarily restricted women, even though their balanced energy availability was achieved by balanced and low energy availabilities were exactly influence of energy availability 241

20 0 divided into four 2-week phases in forest, desert, –1 mountain and swamp environments during which trainees undergo daily military skill training, 8– 15 –2 12 km patrols carrying 32 kg rucksacks, sleep depri- * –3 vation (~ 3.6 h of sleep per night) and dietary 10 intakes during alternate weeks of ~ 8.4 and ~ 21 MJ –4 per day. During the course, trainees lost ~ 12 kg of Pulses/24 h –5 5 body weight. Blood sampling at the end of each –6 * week revealed that T3, IGF-I and testosterone levels * fell ~ 20%, ~ 50% and ~ 70%, respectively, during 0 –7 –1 SX S X weeks on diets of 8.4 MJ·day and returned to normal initial levels during alternate weeks on diets Fig. 18.4 Left: luteinizing hormone (LH) pulse frequency of 21 MJ·day–1, despite continued exposure to all in sedentary (S) and exercising (X) women with the same other training stresses. Thus, exercise appears to energy availability. Right: reduction in LH pulse frequency caused by low energy availability in sedentary have no deleterious effect on reproductive function (S) and exercising (X) women. * = p < 0.01. (From Loucks in either men or women apart from the impact of its 2004; adapted from Loucks et al. 1998.) energy cost on energy availability. If the adrenal axis disrupts the GnRH pulse generator in athletes, it probably does so by mediating the influence of matched. This surprised us, because no-one had energy availability. ever hypothesized that exercise would be protective of reproductive function. Further investigation Energy availability revealed that the exercising women had a higher carbohydrate availability (defined observationally The energy availability hypothesis recognizes that as dietary carbohydrate intake minus carbohydrate mammals partition energy amongst six major meta- oxidation during exercise), due to a glucose-sparing bolic activities: cellular maintenance, thermoregula- alteration in skeletal muscle fuel selection during tion, immunity, locomotion, growth and reproduction energy deprivation. As a result, the carbohydrate (Wade & Schneider 1992). The expenditure of availabilities of the exercising and sedentary women energy in one function, such as locomotion, makes it were above and below the brain’s daily glucose unavailable for others, such as reproductive devel- requirement, respectively. opment and function. Specifically, this hypothesis Since this experiment, amenorrhea has been holds that failure to provide sufficient metabolic induced in monkeys by training them to run volun- fuels to meet the energy requirements of the brain tarily on a motorized treadmill for longer and longer causes an alteration in brain function that disrupts periods while their food intake remained constant the GnRH pulse generator. In this regard, it is (Williams et al. 2001a). The monkeys became amen- important to remember that because fatty acids do orrheic abruptly in 7–24 months, after one or two not cross the blood–brain barrier, the brain relies on cycles of luteal suppression. When the diet of half glucose for energy. Furthermore, the brain has no of the monkeys was then supplemented without glucose storage capacity, and in humans the brain is any moderation of their exercise regimen, their so large and so metabolically active that its daily menstrual cycles were restored (Williams et al. energy requirement exceeds the entire liver glyco- 2001b). The rapidity of recovery was directly related gen storage capacity (Bursztein et al. 1989). In addi- to the number of calories consumed. tion, because skeletal muscle lacks the enzyme to Further data undermining the stress hypothesis return glucose derived from muscle glycogen to the have been reported in a study of young male sol- bloodstream, muscle glycogen stores are not avail- diers participating in the 8-week US Army Ranger able to the brain. By contrast, skeletal muscle does training course (Friedl et al. 2000). This course is have access to liver glycogen stores. Therefore, 242 chapter 18

working skeletal muscle competes directly against the brain for all available carbohydrate stores in the 50 body. In a marathon race, working muscle oxidizes as much glucose in 2 h as the brain needs in a week. 25 Amplitude/3 Considerable data from biological field trials sup- port the hypothesis that mammalian reproductive 0 function depends on energy availability, particu- Effects (%) Effects Frequency larly in women (for reviews see Bronson & Manning –25 1989; Wade & Schneider 1992; Bronson & Heideman 1994; Wade et al. 1996; Schneider & Wade 2000). –50 Anestrus has been induced in Syrian hamsters by food restriction, the administration of pharmacolo- 0 10 20 30 40 50 –1 –1 gical blockers of carbohydrate and fat metabolism, Energy availability (kcal·kg FFM ·day ) insulin administration (which shunts metabolic Fig. 18.5 Dose-dependent effects of restricted energy fuels into storage) and cold exposure (which con- availability on luteinizing hormone (LH) pulse amplitude sumes metabolic fuels in thermogenesis) (Wade & ( top) and frequency ( bottom). Effects are expressed Schneider 1992). Disruptive effects on the reproduct- relative to values at 45 kcal·kg FFM–1·day–1. Effects on LH ive system were independent of body size and pulse amplitude are divided by three for graphical symmetry. As energy availability declines from 45 kcal·kg composition. FFM–1·day–1, effects occur below 30 kcal·kg FFM–1·day–1 Considerable laboratory research suggests that and become more extreme as energy availability is further GnRH neuron activity and LH pulsatility are regu- reduced (1 kcal = 4.18 kJ). (From Loucks & Thuma 2003.) lated by brain glucose availability via two separate mechanisms involving the area postrema (AP) in the caudal brain stem and the vagus nerve (Knobil energy. All subjects were administered the balanced 1990; Minami et al. 1995; Levin et al. 1999; Muroya energy availability treatment (45 kcal [188 kJ]·kg et al. 1999; Wade & Jones 2003). Glucose-sensing FFM–1·day–1) and one of the restricted energy avail- neurons in the AP of the hindbrain appear to trans- ability treatments in random order. mit information to the GnRH pulse generator in Figure 18.5 shows the dose-dependent effects the arcuate nucleus of the hypothalamus in the of energy availability on LH pulsatility. LH pulse forebrain via neurons containing catecholamines, frequency was suppressed and pulse amplitude neuropeptide Y and CRH. These glucose-sensing was increased below a threshold of energy avail- neurons are activated by fasting (Mizuno et al. 1999), ability at ~ 30 kcal [125 kJ]·kg FFM–1·day–1, suggest- due in part to reductions in the inhibitory influences ing that athletes may be able to prevent menstrual of insulin (Schwartz et al. 1992) and leptin (Mizuno disorders by maintaining energy availabilities above et al. 1998) as well as glucose. 30 kcal [125 kJ]·kg FFM–1·day–1. Having demonstrated that low energy avail- We also found that the disruption of LH pulsat- ability, not exercise stress, disrupts LH pulsatility ility is substantially more extreme in women with in exercising women, we investigated the dose– short luteal phases (Fig. 18.6). If the latter finding response relationship between energy availability is confirmed through further investigations, the and LH pulsatility and between energy availability screening of women for luteal length may be a and metabolic substrates and hormones in exercis- convenient way to identify those who need to take ing women (Loucks & Thuma 2003). Energy avail- extra care to avoid falling below the threshold of ability was set at 10, 20, 30 and 45 kcal [42, 84, 125 energy availability needed to maintain normal LH and 188 kJ]·kg FFM–1·day–1 by having all subjects pulsatility. perform 16 kcal [63 kJ]·kg FFM–1·day–1 of exercise at Statistical analysis showed that the dose- o 70% V 2max while consuming 25, 35, 45 or 60 kcal dependent effects on LH pulsatility were similar [104, 146, 188, or 251 kJ]·kg FFM–1·day–1 of dietary to those on the metabolic substrates glucose and influence of energy availability 243

75 cise energy expenditure in this experiment was A11/3 ~ 836 kcal [3.5 MJ]·day–1, these results suggest that 50 many women may be able to maintain normal LH pulsatility while running up to 8 miles (13 km) a day 25 A > 11/3 as long as they do not simultaneously reduce their 0 dietary energy intake below 45 kcal [188 kJ]·kg –1 –1 F>11 FFM ·day . If they do reduce their dietary energy Effects (%) Effects –25 intake, as many exercising women do, then they risk falling below the threshold of energy availability –50 F11 needed to maintain normal LH pulsatility. –75 We also determined the dose-dependent effects 0 10 20 30 40 50 of energy availability on biochemical markers of –1 –1 Energy availability (kcal·kg FFM ·day ) bone turnover in this experiment (Ihle & Loucks 2004). We found that bone resorption increased and Fig. 18.6 Dose-dependent effects of restricted energy availability on luteinizing hormone (LH) pulse amplitude became uncoupled from a suppression of bone for- (, top) and frequency (, bottom) in subgroups of mation only when energy availability was res- women with luteal phases of 11 days (, ) and > 11 days tricted severely enough to suppress estradiol (i.e. to < (, ). Women with shorter luteal phases ( 11 days) had 10 kcal [42 kJ]·kg FFM–1·day–1). If left to continue, been excluded from participation in the experiment. such uncoupling may cause irreversible reductions Effects are relative to values at 45 kcal·kg FFM–1·day–1. Effects on LH pulse amplitude are divided by three for in bone density. By contrast, bone formation was graphical symmetry. Women with luteal phases of 11 days impaired by much less severe restrictions of energy displayed substantially more extreme disruption of LH availability (i.e. as high as 30 kcal [125 kJ]·kg = pulsatility (1 kcal 4.18 kJ). (From Loucks & Thuma 2003.) FFM–1·day–1), a level not at all unusual in weight control programs, in close association with the β-hydroxybutyrate and to the metabolic hormones effects of low energy availability on insulin, IGF-I

GH and cortisol, and unlike those of the other and T3. Such reductions in the rate of bone forma- metabolic hormones including insulin, IGF-I, T3 tion may prevent even regularly menstruating, phys- and leptin. These results support the hypothesis ically active women from achieving their genetic that reproductive function reflects the availability potential for peak bone mass. of metabolic fuels, especially glucose, which may be Recently, a group of physically active university signaled in part by activation of the adrenal axis. women displayed high scores on a test of dietary The role of leptin in mediating the influence of restraint without any of them reporting irregular energy metabolism on the reproductive axis remains menstrual cycles (McLean et al. 2001). From the data controversial with both proponents (Hileman et al. provided, we estimate that their energy availabilit- 2000) and skeptics (Schneider & Wade 2000). The ies were reduced only ~ 22% compared to similar central melanocortin mechanism by which leptin women with low restraint scores. Similarly, monkeys inhibits food intake does not appear to mediate lep- maintained on a 30% restricted diet for 6 years tin’s stimulatory effect on the GnRH pulse generator showed no disruption of menstrual cycling or (Hohmann et al. 2000). reproductive hormones and no reduction in bone The maintenance of normal LH pulsatility in this mineral density, despite a fat mass 46% lower than experiment despite a 33% restriction of energy that in monkeys fed a control diet (Lane et al. 2001). availability to 30 kcal [125 kJ]·kg FFM–1·day–1 Since their body mass declined during the study, demonstrated for the first time that LH pulsatility is their dietary intake normalized to their body mass not simply proportional to energy availability. had, in fact, declined by only ~ 20%. Recent cross- Rather, the regulation of the reproductive system in sectional comparisons of estimated energy avail- women seems to be robust against reductions in ability in athletes also support the notion that energy availability as large as 33%. Since the exer- menstrual function is disrupted at the threshold of 244 chapter 18

energy availability needed to maintain normal LH macronutrient intake. Even a 20% increase in energy pulsatility (Thong et al. 2000). Amenorrheic athletes expenditure during 40 weeks of marathon training were estimated to habitually self-administer an induced no increase in energy intake (Westerterp energy availability of ~ 16 kcal [67 kJ]·kg FFM–1·day–1, et al. 1992). In our own laboratory, women say that while eumenorrheic athletes habitually self-admin- they have to force themselves to eat far beyond istered ~ 30 kcal [125 kJ]·kg FFM–1·day–1. their appetites to consume the amount of food that compensates their dietary energy intake for their Implications for athletic training exercise energy expenditure and to prevent the disruption of LH pulsatility. Other investigators Research done to date suggests that athletes may be have had to offer exercising amenorrheic monkeys able to prevent or reverse menstrual disorders by special treats to induce them to increase their energy increasing their dietary energy intake without any intake enough to restore their menstrual cycles modification of their exercise regimen. In short- (Williams et al. 2001b). Consequently, to improve term experiments, an energy availability of 30 kcal their performance while protecting their health, [125 kJ]·kg FFM–1·day–1 appears to be sufficient to athletes must learn to eat by discipline instead of preserve normal LH pulsatility. More short-term appetite. experiments are needed to determine the independ- Complicating that discipline, another part of the ent effects of carbohydrate and energy availability nutritional challenge for athletes is that energy on LH pulsatility, to identify the exact physiological balance is not the objective of athletic training. mechanism by which low energy or carbohydrate Athletic performance is maximized, in part, by a availability disrupts LH pulsatility, and to investig- sport-specific (and in team sports, position-specific) ate further whether low energy availability causes optimum body size, body composition and mix more extreme disruptions of reproductive function of stored metabolic fuels. Therefore, much of an in certain identifiable women. More prolonged athlete’s training aims to modify her body to experiments are needed to verify that short-term achieve these objectives. It is also important to know effects on LH pulsatility are predictive of chronic that macronutrients are metabolized differently effects on ovarian function, and to identify practical, and stored separately so that the conversion of one effective and acceptable interventions for prevent- macronutrient into another for storage does not rep- ing and reversing menstrual disorders in athletes. resent important metabolic pathways (Flatt 1988). Research is also needed to investigate how body Therefore, an athlete needs to manage fat, protein weight is maintained in chronically energy-deficient and carbohydrate balances separately to optimize athletes. their pursuit of sport-specific body size, body com- Part of the nutritional challenge for athletes is that position and energy store objectives. ‘there is no strong biological imperative to match As part of this discipline, athletes would benefit energy intake to activity-induced energy expend- from monitoring biomarkers of their progress and iture’ (Blundell & King 1999, p. 5581). Experimental pitfalls on the path toward their particular body food deprivation increases hunger, but the same size, body composition and energy store object- energy deficit produced by exercise energy expend- ives. Much applied research is needed to validate iture does not (Hubert et al. 1998). Studies have the utility of such biomarkers. Ideally, a single shown that hunger is suppressed briefly by a single measurement of a biomarker would provide the bout of intense exercise (Blundell & King 1998), and desired information accurately, unambiguously, two bouts of intense exercise in a single day induces inexpensively, conveniently, non-invasively, safely, no increase in ad libitum food intake on that or the privately and quickly without intentional or un- following 2 days (King et al. 1997). Furthermore, intentional confounding by other factors. Few, if large shifts in carbohydrate and fat oxidation any, biomarkers fulfill all these criteria. An obvious (Stubbs et al. 1995a, 1995b) and in glycogen stores candidate for monitoring progress in reducing fat (Snitker et al. 1997) produce no changes in ad libitum mass is skinfold thickness, which is simple, direct, influence of energy availability 245

immediate and inexpensive, and perhaps good institutional reforms like Rule 3 in National Col- enough. We are unaware of any inexpensive, non- legiate Athletic Association (NCAA) men’s wrestling invasive method for assessing muscle and liver (NCAA 2002) may be necessary to protect the health glycogen stores, which are essential for both athletic of female athletes. Rule 3 restricts weight loss object- performance and reproductive health, but a con- ives and sets individualized minimum weights for venient method for assessing carbohydrate defici- competition. In so doing, it prevents widespread ency is readily available. In our experience, urinary weight loss practices that had previously placed the ketones identify carbohydrate-deficient individuals health of participants at risk. almost as reliably at 30 kcal [125 kJ]·kg FFM–1·day–1 –1 –1 as they do at 10 kcal [42 kJ]·kg FFM ·day . A single Acknowledgments prolonged exercise bout on an energy-restricted diet is sufficient to elevate plasma ketones (Ainslie et al. This research was supported in part by the US Army 2003). Athletes can purchase ‘keto-sticks’ inexpens- Medical Research Acquisition Activity (USAM- ively in most pharmacies to monitor their urinary RAA) grant #DAMD 17-95-1-5053, and in part by ketones in the privacy of their own homes. Grant M01 RR00034 from the General Clinical The health of female athletes has always been Research Branch, Division of Research Resources, a high priority of women’s sports, but undernutri- NIH. The content of the information presented in tion has become standard practice, especially in this manuscript does not necessarily reflect the endurance, aesthetic and weight-class sports. If position or the policy of the government, and no equally pervasive voluntary reforms by athletes, official endorsement should be inferred. coaches and judges are unlikely, then mandatory

References

Ainslie, P.N., Campbell, I.T., Frayn, K.N. Axelson, J.F. (1987) Forced swimming pubertal development with et al. (2003) Physiological, metabolic, alters vaginal estrous cycles, body gonadotropin-releasing hormone. and performance implications of a composition, and steroid levels without Endocrinology 118(6), 2483–2487. prolonged hill walk: influence of energy disrupting lordosis behavior or fertility Bronson, F.H. & Heideman, P.D. (1994) intake. Journal of Applied Physiology 94(3), in rats. Physiology and Behavior 41(5), Seasonal regulation of reproduction 1075–1083. 471–479. in mammals. In: The Physiology of Anderson, S.E., Dallal, G.E. & Must, A. Beidleman, B.A., Puhl, J.L. & De Souza, Reproduction, Vol. 2 (Knobil, E. & Neill, (2003) Relative weight and race M.J. (1995) Energy balance in female J., eds.). Raven Press, New York: influence average age at menarche: distance runners. American Journal of 541–583. Results from two nationally Clinical Nutrition 61(2), 303–311. Bronson, F.H. & Manning, J.M. (1989) Food representative surveys of US girls Berga, S.L., Mortola, J.F., Girton, L. et al. consumption, prolonged exercise, and studied 25 years apart. Pediatrics 111, (1989) Neuroendocrine aberrations in LH secretion in the peripubertal female 844–850. women with functional hypothalamic rat. In: The Menstrual Cycle and Its Anderson, S.J., Griesemer, B.A., Johnson, amenorrhea. The Journal of Clinical Disorders (Pirke, K.M., Wuttle, W. & M.D. et al. (2000) American Academy Endocrinology and Metabolism 68, Schweiger, U., eds.). Springer-Verlag, of Pediatrics. Committee on Sports 301–309. Berlin: 42–29. Medicine and Fitness. Medical concerns Blundell, J.E. & King, N.A. (1998) Effects Bronson, F.H. & Manning, J.M. (1991) in the female athlete. Pediatrics 106(3), of exercise on appetite control: loose The energetic regulation of ovulation: 610–613. coupling between energy expenditure a realistic role for body fat. Biology of Asahina, K., Kitahara, F., Yamanaka, M. & and energy intake. International Journal Reproduction 44(6), 945–950. Akiba, T. (1959) Influence of excessive of Obesity and Related Metabolic Disorders Broocks, A., Pirke, K.M., Schweiger, U. exercise on the structure and function of 22 (suppl. 2), S22–S29. et al. (1990) Cyclic ovarian function in rat organs. Japanese Journal of Physiology Blundell, J.E. & King, N.A. (1999) Physical recreational athletes. Journal of Applied 9, 322–326. activity and regulation of food intake: Physiology 68(5), 2083–2086. Audi, L., Vargas, D.M., Gussinye, M. current evidence. Medicine and Science in Bullen, B.A., Skrinar, G.S., Beitins, I.Z. et al. (2002) Clinical and biochemical Sports and Exercise 31(11), S573–S583. et al. (1985) Induction of menstrual determinants of bone metabolism and Bronson, F.H. (1986) Food-restricted, disorders by strenuous exercise in bone mass in adolescent female patients prepubertal, female rats: rapid recovery untrained women. New England Journal with anorexia nervosa. Pediatric Research of luteinizing hormone pulsing with of Medicine 312(21), 1349–1353. 51(4), 497–504. excess food, and full recovery of Burke, L.M., Cox, G.R., Culmmings, N.K. 246 chapter 18

et al. (2001) Guidelines for daily content of amenorrheic and contraceptive administration on bone carbohydrate intake: do athletes achieve eumenorrheic athletes. New England density in anorexia nervosa. Journal of them? Sports Medicine 31(4), 267–299. Journal of Medicine 311(5), 277–281. Clinical Endocrinology and Metabolism Bursztein, S., Elwyn, D.H., Askanazi, J. Drinkwater, B.L., Bruemner, B. & Chesnut, 87(6), 2883–2891. & Kinney, J.M. (1989) Energy Metabolism, C.H., 3rd (1990) Menstrual history as a Gulekli, B., Davies, M.C. & Jacobs, H.S. Indirect Calorimetry, and Nutrition. determinant of current bone density in (1994) Effect of treatment on established Williams & Wilkins, Baltimore, MD: 146. young athletes. Journal of the American osteoporosis in young women with Caillot-Augusseau, A., Lafage-Proust, Medical Association 263(4), 545–548. amenorrhoea. Clinical Endocrinology M.H., Margaillan, P.H. et al. (2000) Edwards, J.E., Lindeman, A.K., Mikesky, 41(3), 275–281. Weight gain reverses bone turnover A.E. & Stager, J.M. (1993) Energy Hammar, M.L., Hammar-Henriksson, and restores circadian variation of bone balance in highly trained female M.B., Frisk, J., Rickenlund, A. & Wyon, resorption in anorexic patients. Clinical endurance runners. Medicine and Y.A. (2000) Few oligo-amenorrheic Endocrinology 52(1), 113–121. Science in Sports and Exercise 25(12), athletes have vasomotor symptoms. Chatterton, R.T., Jr., Hartman, A.L., Lynn, 1398–1404. Maturitas 34(3), 219–225. D.E. & Hickson, R.C. (1990) Exercise- Ellison, P.T. & Lager, C. (1986) Moderate Harber, V.J., Petersen, S.R. & Chilibeck, induced ovarian dysfunction in the rat. recreational running is associated with P.D. (1998) Thyroid hormone Proceedings of the Society for Experimental lowered salivary progesterone profiles concentrations and muscle metabolism Biology and Medicine 193(3), 220–224. in women. American Journal of Obstetrics in amenorrheic and eumenorrheic Chrousos, G. P. & Gold, P.W. (1992) The and Gynecology 154, 1000–1003. athletes. Canadian Journal of Applied concepts of stress and stress system Flatt, J.P. (1988). Importance of nutrient Physiology 23, 293–306. disorders. Journal of the American Medical balance in body weight regulation. Hartman, D., Crisp, A., Rooney, B. et al. Association 267, 1244–1252. Diabetes/Metabolism Reviews 4(6), (2000) Bone density of women who De Souza, M.J., Maguire, M.S., Maresh, 571–581. have recovered from anorexia nervosa. C.M. et al. (1991) Adrenal activation Friedl, K E., Moore, R.J., Hoyt, R.W. International Journal of Eating Disorders and the prolactin response to exercise in et al. (2000) Endocrine markers of 28(1), 107–112. eumenorrheic and amenorrheic runners. semistarvation in healthy lean men in a Heer, M., Mika, C., Grzella, I., Drummer, Journal of Applied Physiology 70(6), multistressor environment. Journal of C. & Herpertz-Dahlmann, B. (2002) 2378–2387. Applied Physiology 88(5), 1820–1830. Changes in bone turnover in patients De Souza, M.J., Luciano, A.A., Arce, J.C., Frisch, R.E. & McArthur, J.W. (1974) with anorexia nervosa during 11 weeks Demers, L.M. & Loucks, A.B. (1994) Menstrual cycles: fatness as determinant of inpatient dietary treatment. Clinical Clinical tests explain blunted cortisol of minimum weight for height necessary Chemistry 48(5), 754–760. responsiveness but not mild for their maintenance or onset. Science Hetland, M.L., Haarbo, J., Christiansen, C. hypercortisolism in amenorrheic 185, 949–951. & Larsen, T. (1993) Running induces runners. Journal of Applied Physiology Gold, P.W., Gwirtsman, H., Averinos, P.C. menstrual disturbances but bone mass 76(3), 1302–1309. et al. (1986) Abnormal hypothalamic– is unaffected, except in amenorrheic De Souza, M.J., Miller, B.E., Loucks, A.B. pituitary–adrenal function in anorexia women. American Journal of Medicine et al. (1998) High frequency of luteal nervosa: pathophysiologic mechanisms 95(1), 53–60. phase deficiency and anovulation in in underweight and weight-corrected Hileman, S.M., Pierroz, D.D. & Flier, J.S. recreational women runners: blunted patients. New England Journal of Medicine (2000) Leptin, nutrition, and elevation in follicle-stimulating 314(21), 1335–1342. reproduction: timing is everything. hormone observed during luteal- Goldberg, G.R., Black, A.E., Jebb, S.A. Journal of Clinical Endocrinology and follicular transition. Journal of Clinical et al. (1991) Critical evaluation of energy Metabolism 85(2), 804–806. Endocrinology and Metabolism 83(12), intake data using fundamental Hill, R.J. & Davies, P.S. (2002) Energy 4220–4232. principles of energy physiology: 1. intake and energy expenditure in elite De Souza, M.J., Van Heest, J., Demers, L.M. Derivation of cut-off limits to identify lightweight female rowers. Medicine and & Lasley, B.L. (2003) Luteal phase under-recording. European Journal of Science in Sports and Exercise 34(11), deficiency in recreational runners: Clinical Nutrition 45(12), 569–581. 1823–1829. evidence for a hypometabolic state. Gordon, C.M., Grace, E., Emans, S.J. Hilton, L.K. & Loucks, A.B. (2000) Low Journal of Clinical Endocrinology and et al. (2002) Effects of oral energy availability, not exercise stress, Metabolism 88(1), 337–346. dehydroepiandrosterone on bone suppresses the diurnal rhythm of leptin Di Carlo, C., Palomba, S., De Fazio, M. density in young women with anorexia in healthy young women. American et al. (1999) Hypogonadotropic nervosa: a randomized trial. Journal of Journal of Physiology. Endocrinology and hypogonadotropism in obese women Clinical Endocrinology and Metabolism Metabolism 278(1), E43–E49. after biliopancreatic diversion. Fertility 87(11), 4935–4941. Hoch, A.Z., Dempsey, R.L., Carrera, G.F. and Sterility 72(5), 905–909. Grinspoon, S., Thomas, E., Pitts, S. et al. et al. (2003a) Is there an association Ding, J.-H., Scheckter, C.B., Drinkwater, (2000) Prevalence and predictive factors between athletic amenorrhea and B.L., Soules, M.R. & Bremner, W.J. for regional osteopenia in women with endothelial cell dysfunction? Medicine (1988) High serum cortisol levels in anorexia nervosa. Annals of Internal and Science in Sports and Exercise 35(3), exercise-associated amenorrhea. Annals Medicine 133(10), 790–794. 377–383. of Internal Medicine 108, 530–534. Grinspoon, S., Thomas, L., Miller, K., Hoch, A.Z., Jurva, J., Staton, M. et al. Drinkwater, B.L., Nilson, K., Chesnut, Herzog, D. & Klibanski, A. (2002) Effects (2003b) Is endothelial dysfunction that C.H., 3rd et al. (1984) Bone mineral of recombinant human IGF-I and oral is associated with athletic amenorrhea influence of energy availability 247

reversible? Medicine and Science in Sports Kolaczynski, J.W., Ohannesian, J.P., et al. (1992) Hypothalamic–pituitary– and Exercise 35, S12. Considine, R.V., Marco, C.C. & Caro, J.F. thyroidal function in eumenorrheic and Hohmann, J.G., Teal, T.T., Clifton, D.K. (1996b) Response of leptin to short-term amenorrheic athletes. Journal of Clinical et al. (2000) Differential role of and prolonged overfeeding in humans. Endocrinology and Metabolism 75(2), melanocortins in mediating leptin’s Journal of Clinical Endocrinology and 514–518. effects on feeding and reproduction. Metabolism 81, 4162–4165. Loucks, A.B., Verdun, M. & Heath, E.M. American Journal of Physiology. Lane, M.A., Black, A., Handy, A.M. et al. (1998) Low energy availability, not stress Regulatory, Integrative and Comparitive (2001) Energy restriction does not of exercise, alters LH pulsatility in Physiology 278, R50–R59. alter bone mineral metabolism or exercising women. Journal of Applied Hotta, M., Fukuda, I., Sato, K. et al. (2000) reproductive cycling and hormones Physiology 84(1), 37–46. The relationship between bone turnover in female rhesus monkeys. Journal of McLean, J.A., Barr, S.I. & Prior, J.C. (2001) and body weight, serum insulin-like Nutrition 131(3), 820–827. Cognitive dietary restraint is associated growth factor (IGF) I, and serum IGF- Laughlin, G.A. & Yen, S.S.C. (1996) with higher urinary cortisol excretion binding protein levels in patients with Nutritional and endocrine-metabolic in healthy premenopausal women. anorexia nervosa. Journal of Clinical aberrations in amenorrheic athletes. American Journal of Clinical Nutrition 73, Endocrinology and Metabolism 85(1), Journal of Clinical Endocrinology and 7–12. 200–206. Metabolism 81(12), 4301–4309. Maddalozzo, G.F. & Snow, C.M. (2000) Hubert, P., King, N.A. & Blundell, J.E. Laughlin, G.A. & Yen, S.S.C. (1997) High intensity resistance training: (1998) Uncoupling the effects of energy Hypoleptinemia in women athletes: effects on bone in older men and expenditure and energy intake: appetite absence of a diurnal rhythm with women. Calcified Tissue International response to short-term energy deficit amenorrhea. Journal of Clinical 66(6), 399–404. induced by meal omission and physical Endocrinology and Metabolism 82(1), Maffei, M., Halass, J., Ravussin, E. et al. activity. Appetite 31(1), 9–19. 318–321. (1995) Leptin levels in human and Ihle, R. & Loucks, A.B. (2004) Dose– Leibel, R.L., Rosenbaum, M. & Hirsch, J. rodent: measurement of plasma leptin response relationships between energy (1995) Changes in energy expenditure and ob RNA in obese and weight- availability and bone turnover in young resulting from altered body weight. reduced subjects. Nature Medicine 1, exercising women. Journal of Bone and New England Journal of Medicine 332(10), 1155–1161. Mineral Research 19, 1231–1240. 621–628. Manning, J.M. & Bronson, F.H. (1989) Jenkins, A.B., Markovic, T.P., Fleury, A. Lennkh, C., de Zwaan, M., Bailer, U. et al. Effects of prolonged exercise on puberty & Campbell, L.V. (1997) Carbohydrate (1999) Osteopenia in anorexia nervosa: and luteinizing hormone secretion intake and short-term regulation of specific mechanisms of bone loss. Journal in female rats. American Journal of leptin in humans. Diabetologia 40(3), of Psychiatric Research 33(4), 349–356. Physiology 257(6 Pt. 2), R1359–R1364. 348–351. Levin, B.E., Dunn-Meynell, A.A. & Routh, Manning, J.M. & Bronson, F.H. (1991) Jenkins, P.J., Ibanez-Santos, X., Holly, J. V.H. (1999) Brain glucose sensing and Suppression of puberty in rats by et al. (1993) IGFBP-1: a metabolic signal body energy homeostasis: role in obesity exercise: effects on hormone levels and associated with exercise-induced and diabetes. American Journal of reversal with GnRH infusion. American amenorrhoea. Neuroendocrinology 57(4), Physiology 276(5 Pt. 2), R1223–R1231. Journal of Physiology 260(4 Pt. 2), 600–604. Loucks, A.B. (2004) Energy balance and R717–R723. Kaiserauer, S., Snyder, A.C., Sleeper, M. body composition in sports and exercise. Marcus, R., Cann, C., Madvig, P. et al. & Zierath, J. (1989) Nutritional, Journal of Sports Sciences 22, 1–14. (1985) Menstrual function and bone physiological, and menstrual status of Loucks, A.B. & Horvath, S.M. (1985) mass in elite women distance runners. distance runners. Medicine and Science in Athletic amenorrhea: a review. Medicine Endocrine and metabolic features. Sports and Exercise 21(2), 120–125. and Science in Sports and Exercise 17(1), Annals of Internal Medicine 102(2), Keen, A.D. & Drinkwater, B.L. (1997) 56–72. 158–163. Irreversible bone loss in former Loucks, A.B. & Thuma, J.R. (2003) Marcus, R., Feldman, D. & Kelsey, J. (1996) amenorrheic athletes. Osteoporosis Luteinizing hormone pulsatility is Osteoporosis. Academic Press, San Diego, International 7(4), 311–315. disrupted at a threshold of energy CA. King, N.A., Lluch, A., Stubbs, R.J. & availability in regularly menstruating Mertz, W., Tsui, J.C., Judd, J.T. et al. (1991) Blundell, J.E. (1997) High dose exercise women. Journal of Clinical Endocrinology What are people really eating? The does not increase hunger or energy and Metabolism 88(1), 297–311. relation between energy intake derived intake in free living males. European Loucks, A.B., Horvath, S.M. & Freedson, from estimated diet records and intake Journal of Clinical Nutrition 51(7), P.S. (1984) Menstrual status and determined to maintain body weight. 478–483. validation of body fat prediction in American Journal of Clinical Nutrition Knobil, E. (1990) The GnRH pulse athletes. Human Biology 56(2), 383–392. 54(2), 291–295. generator. American Journal of Obstetrics Loucks, A.B., Mortola, J.F., Girton, L. & Minami, S., Kamegai, J., Sugihara, H. et al. and Gynecology 163(5 Pt. 2), 1721–1727. Yen, S.S.C. (1989) Alterations in the (1995) Central glucoprivation evoked by Kolaczynski, J.W., Considine, R.V., hypothalamic–pituitary–ovarian and administration of 2-deoxy-D-glucose Ohannesian, J. et al. (1996a) Responses the hypothalamic–pituitary–adrenal induces expression of the c-fos gene in of leptin to short-term fasting and axes in athletic women. Journal of Clinical a subpopulation of neuropeptide Y refeeding in humans: a link with Endocrinology and Metabolism 68(2), neurons in the rat hypothalamus. Brain ketogenesis but not ketones themselves. 402–411. Research. Molecular Brain Research 33(2), Diabetes 45(11), 1511–1515. Loucks, A.B., Laughlin, G.A., Mortola, J.F. 305–310. 248 chapter 18

Mizuno, T.M., Kleopoulos, S.P., Bergen, Pirke, K.M., Schweiger, U., Broocks, A., intake in humans after manipulation H.T. et al. (1998) Hypothalamic pro- Tuschl, R.J. & Laessle, R.G. (1990) of glycogen stores. American Journal of opiomelanocortin mRNA is reduced by Luteinizing hormone and follicle Clinical Nutrition 65(4), 941–946. fasting and in ob/ob and db/db mice, stimulating hormone secretion patterns Soyka, L.A., Grinspoon, S., Levitsky, L.L., but is stimulated by leptin. Diabetes in female athletes with and without Herzog, D.B. & Klibanski, A. (1999) 47(2), 294–297. menstrual disturbances. Clinical The effects of anorexia nervosa on bone Mizuno, T.M., Makimura, H., Silverstein, J. Endocrinology 33, 345–353. metabolism in female adolescents. et al. (1999) Fasting regulates Ravussin, E. (2002) Cellular sensors of Journal of Clinical Endocrinology and hypothalamic neuropeptide Y, feast and famine. Journal of Clinical Metabolism 84(12), 4489–4496. agouti-related peptide, and Investigation 109(12), 1537–1540. Soyka, L.A., Misra, M., Frenchman, A. et al. proopiomelanocortin in diabetic mice Rivier, C. & Rivest, S. (1991) Effect of stress (2002) Abnormal bone mineral accrual in independent of changes in leptin or on the activity of the hypothalamic– adolescent girls with anorexia nervosa. insulin. Endocrinology 140(10), pituitary–gonadal axis: peripheral Journal of Clinical Endocrinology and 4551–4557. and central mechanisms. Biology of Metabolism 87(9), 4177–4185. Mulligan, K. & Butterfield, G.E. (1990) Reproduction 45, 523–532. Stacey, E., Korkia, P., Hukkanen, M.V.J., Discrepancies between energy intake Rossetti, L. (2000) Perspective: Polak, J.M. & Rutherford, O.M. (1998) and expenditure in physically active hexosamines and nutrient sensing. Decreased nitric oxide levels and bone women. British Journal of Nutrition 64(1), Endocrinology 141(6), 1922–1925. turnover in amenorrheic athletes with 23–36. Sanborn, C.F., Albrecht, B.H. & Wagner, spinal osteopenia. Journal of Clinical Muroya, S., Yada, Y., Shioda, S. & W.W., Jr. (1987) Athletic amenorrhea: Endocrinology and Metabolism 83(9), Takigawa, M. (1999) Glucose-sensitive lack of association with body fat. 3056–3061. neurons in the rat arcuate nucleus Medicine and Science in Sports and Exercise Stefanis, N., Mackintosh, C., Abraha, H.D., contain neuropeptide Y. Neuroscience 19(3), 207–212. Treasure, J. & Moniz, C. (1998) Letters 264(1–3), 113–116. Schneider, J.E. & Wade, G.N. (2000) Dissociation of bone turnover in Myerson, M., Gutin, B., Warren, M.P. Inhibition of reproduction in service anorexia nervosa. Annals of Clinical et al. (1991) Resting metabolic rate and of energy balance. In: Reproduction Biochemistry 35(Pt. 6), 709–716. energy balance in amenorrheic and in Context: Social and Environmental Stubbs, R.J., Harbron, C.G., Murgatroyd, eumenorrheic runners. Medicine and Influences on Reproductive Physiology and P.R. & Prentice, A.M. (1995a) Covert Science in Sports and Exercise 23, 15–22. Behavior (Wallen, K. & Schneider, J.E., manipulation of dietary fat and energy National Collegiate Athletic Association eds.). MIT Press, Cambridge, MA: density: effect on substrate flux and (2003) 2004 NCAA Wrestling Rules and 35–82. food intake in men eating ad libitum. Interpretations. National Collegiate Schwartz, M.W., Sipols, A.J., Marks, J.L. American Journal of Clinical Nutrition Athletic Association, Indianapolis, et al. (1992) Inhibition of hypothalamic 62(2), 316–329. IN: WR23–WR35, WI10–WI11, neuropeptide Y gene expression Stubbs, R.J., Ritz, P., Coward, W.A. WA9–WA10, WA27–WA32. by insulin. Endocrinology 130(6), & Prentice, A.M. (1995b) Covert Nelson, M.E., Fisher, N.C., Catsos, P.D. 3608–3616. manipulation of the ratio of dietary fat to et al. (1986) Diet and bone status in Scott, E.C. & Johnston, F.E. (1982) Critical carbohydrate and energy density: effect amenorrheic runners. American Journal fat, menarche, and the maintenance on food intake and energy balance in of Clinical Nutrition 43(6), 910–916. of menstrual cycles: a critical review. free-living men eating ad libitum. Nielsen, T.F., Ravn, P., Bagger, Y.Z., Journal of Adolescent Health Care 2(4), American Journal of Clinical Nutrition Warming, L. & Christiansen, C. (2004) 249–260. 62(2), 330–337. Pulsed estrogen therapy in prevention of Selye, H. (1939) The effect of adaptation to Styne, D.M. & Grumbach, M.M. (1991) postmenopausal osteoporosis. A 2-year various damaging agents on the female Disorders of puberty in the male and randomized, double blind, placebo- sex organs in the rat. Endocrinology 25, female. In: Reproductive Endocrinology controlled study. Osteoporosis 615–624. (Yen, S.S.C. & Jaffe, R.B., eds.). W.B. International 15(2), 168–174. Sinning, W.E. & Little, K.D. (1987) Body Saunders, Co., Philadelphia, PA: Obici, S., Wang, J., Chowdury, R. et al. composition and menstrual function 511–544. (2002) Identification of a biochemical in athletes. Sports Medicine 4(1), 34–45. Tabata, I., Ogita, F., Miyachi, M. & link between energy intake and energy Sjodin, A.M., Andersson, A.B., Hogberg, Shibayama, H. (1991) Effect of low blood expenditure. Journal of Clinical J.M. & Westerterp, K.R. (1994) Energy glucose on plasm CRF, ACTH, and Investigation 109(12), 1599–1605. balance in cross-country skiers: a study cortisol during prolonged physical Okano, H., Mizunuma, H., Soda, M. using doubly labeled water. Medicine exercise. Journal of Applied Physiology 71, et al. (1995) Effects of exercise and and Science in Sports and Exercise 26(6), 1807–1812. amenorrhea on bone mineral density in 720–724. Thong, F.S., McLean, C. & Graham, T.E. teenage runners. Endocrine Journal 42(2), Slentz, C.A., Davis, J.M., Settles, D.L., (2000) Plasma leptin in female athletes: 271–276. Pate, R.R. & Settles, S.J. (1990) Glucose relationship with body fat, reproductive, Otis, C.L., Drinkwater, B., Johnson, M., feedings and exercise in rats: glycogen nutritional, and endocrine factors. Loucks, A. & Wilmore, J. (1997) use, hormone responses, and Journal of Applied Physiology 88(6), American College of Sports Medicine performance. Journal of Applied 2037–2044. position stand. The female athlete triad. Physiology 69, 989–994. Trabulsi, J. & Schoeller, D.A. (2001) Medicine and Science in Sports and Exercise Snitker, S., Larson, D.E., Tataranni, P.A. Evaluation of dietary assessment 29(5), i–ix. & Ravussin, E. (1997) Ad libitum food instruments against doubly labeled influence of energy availability 249

water, a biomarker of habitual energy Warren, M.P., Brooks-Gunn, J., Fox, R.P. in the induction of menstrual cycle intake. American Journal of Physiology. et al. (2003) Persistent osteopenia in disturbances during strenuous exercise Endocrinology and Metabolism 281(5), ballet dancers with amenorrhea and training. Journal of Clinical Endocrinology E891–E899. delayed menarche despite hormone and Metabolism 86(11), 5184–5193. Trappe, T.A., Gastaldelli, A., Jozsi, A.C., therapy: a longitudinal study. Fertility Wilmore, J.H., Wambsgans, K.C., Brenner, Troup, J.P. & Wolfe, R.R. (1997) Energy and Sterility 80(2), 398–404. M. et al. (1992) Is there energy expenditure of swimmers during high Weigle, D.S., Duell, P.B., Connor, W.E. conservation in amenorrheic compared volume training. Medicine and Science in et al. (1997) Effect of fasting, refeeding, with eumenorrheic distance runners? Sports and Exercise 29(7), 950–954. and dietary fat restriction on plasma Journal of Applied Physiology 72(1), Treloar, A.E., Boynton, R.E., Behn, B.G. leptin levels. Journal of Clinical 15–22. & Brown, B.W. (1967) Variation of Endocrinology and Metabolism 82, Yahiro, J., Glass, A.R., Fears, W.B., the human menstrual cycle through 561–565. Ferguson, E.W. & Vigersky, R.A. (1987) reproductive life. International Journal of Westerterp, K.R. & Saris, W.H. (1991) Exaggerated gonadotropin response to Fertility 12(1 Pt. 2), 77–126. Limits of energy turnover in relation to luteinizing hormone-releasing hormone Veldhuis, J.D., Evans, W.S., Demers, L.M. physical performance, achievement of in amenorrheic runners. American et al. (1985) Altered neuroendocrine energy balance on a daily basis. Journal Journal of Obstetrics and Gynecology regulation of gonadotropin secretion in of Sports Sciences 9, 1–15. 156(3), 586–591. women distance runners. Journal of Westerterp, K.R., Meijer, G.A., Janssen, Zanker, C.L. & Swaine, I.L. (1998a) Clinical Endocrinology and Metabolism E.M., Saris, W.H. & Ten Hoor, F. (1992) Bone turnover in amenorrhoeic and 61(3), 557–563. Long-term effect of physical activity on eumenorrhoeic women distance Wade, G.N. & Jones, J.E. (2003) Lessons energy balance and body composition. runners. Scandinavian Journal of Medicine from experimental disruption of British Journal of Nutrition 68(1), 21–30. and Science in Sports 8(1), 20–26. estrous cycles and behaviors. Medicine Williams, N.I., Young, J.C., McArthur, J.W. Zanker, C.L. & Swaine, I.L. (1998b) and Science in Sports and Exercise 35, et al. (1995) Strenuous exercise with Relation between bone turnover, 1573–1580. caloric restriction: effect on luteinizing oestradiol, and energy balance in Wade, G.N. & Schneider, J.E. (1992) hormone secretion. Medicine and Science women distance runners. British Journal Metabolic fuels and reproduction in Sports and Exercise 27(10), 1390–1398. of Sports Medicine 32(2), 167–171. in female mammals. Neuroscience Williams, N.I., Caston-Balderrama, A.L., Zanker, C.L. & Swaine, I.L. (1998c) The and Biobehavioral Reviews 16(2), Helmreich, D.L. et al. (2001a) relationship between serum oestradiol 235–272. Longitudinal changes in reproductive concentration and energy balance in Wade, G.N., Schneider, J.E. & Li, H.Y. hormones and menstrual cyclicity in young women distance runners. (1996) Control of fertility by metabolic cynomolgus monkeys during strenuous International Journal of Sports Medicine cues. American Journal of Physiology exercise training: abrupt transition to 19(2), 104–108. 270(1 Pt. 1), E1–E19. exercise-induced amenorrhea. Zanker, C.L., Cooke, C.B., Truscott, J.G., Wang, J., Liu, R., Hawkins, M., Barzilai, N. Endocrinology 142(6), 2381–2389. Oldroyd, B. & Jacobs, H.S. (2004) Annual & Rossetti, L. (1998) A nutrient-sensing Williams, N.I., Helmreich, D.L., Parfitt, changes of bone density over 12 years in pathway regulates leptin gene D.B., Caston-Balderrama, A.L. & an amenorrheic athlete. Medicine and expression in muscle and fat. Nature Cameron, J.L. (2001b) Evidence for a Science in Sports and Exercise 36(1), 393(6686), 684–688. causal role of low energy availability 137–142. Chapter 19

Oral Contraceptive Use and Physical Performance

JACI L. VANHEEST, CARRIE E. MAHONEY AND CAROL D. RODGERS

Oral contraceptive use is common in both recre- classical endocrine axis known as the hypothalamic– ational and elite female athletes. Oral contraceptives pituitary–gonadal (HPG) axis (Fig. 19.1). Regulation are generally comprised of a combination of estro- of the HPG axis occurs through both short- and gen- and progesterone-like compounds. The com- long-loop feedback systems. position is dependant upon the varying levels of GnRH is secreted from the hypothalamus in a the two steroid derivatives produced by the various pulsatile manner throughout the menstrual cycle. manufacturers. In addition to their use for contra- Hypothalamic GnRH pulsatility is essential for regu- ception, women may also be prescribed oral contra- lar menstrual cyclicity. On average, the frequency of ceptives for cycle regulation and to manage various GnRH secretion is once per 90 min during the early cycle dysfunctions (i.e. amenorrhea or dysmenor- follicular phase and once per 60–70 min during the rhea). Side effects such as weight gain, fluid retention luteal phase. The release of both FSH and LH is and nausea initially caused many athletic women induced by GnRH with LH being the more sensit- to avoid the use of oral contraceptives. However, ive of the two hormones with respect to changes in with the introduction of lower dose formulations GnRH levels (Larsen et al. 2003). the negative side effects have been reduced and the FSH is secreted by the anterior pituitary gland numbers of women using these drugs has increased. and is essential for follicular growth. Its secretion The impact of the oral administration of ovarian is highest and most critical during the 1st week steroids on physical performance has been evalu- of the follicular stage. At the level of the ovary, FSH ated over the past several decades. Research, how- induces estrogen and progesterone secretion by ever, remains incomplete as to the potential positive activating aromatase and p450 enzymes. FSH also and/or negative effects of these compounds on sport induces the proliferation of granulosa cells and performance. The following chapter will address expression of LH receptors on granulosa cells. the endogenous hormonal control of the menstrual LH is secreted by the anterior pituitary gland and cycle, the various types and mechanisms of action is required for both growth of preovulatory follicles of exogenous hormones and the influence of these and luteinization and ovulation of the dominant drugs on physical performance. follicle. During the follicular phase of the menstrual cycle, LH induces androgen synthesis by theca cells; Overview of normal menstrual cycle stimulates proliferation, differentiation, and secre- tion of follicular thecal cells; and increases LH The menstrual cycle is regulated primarily by a receptors on granulosa cells. The preovulatory LH group of five hormones; gonadotropin-releasing surge drives the oocyte into the first meiotic division hormone (GnRH), follicle-stimulating hormone and initiates luteinization of thecal and granulosa (FSH), luteinizing hormone (LH), estrogen and pro- cells. The resulting corpus luteum produces high gesterone. These hormones are released as part of a levels of progesterone and some estrogen. 250 oral contraceptive use and physical performance 251

exert negative feedback on LH and FSH secretion; however, at very high levels estrogens exert positive feedback on LH and FSH secretion. Estrogen further induces proliferation of granulosa cells and syn- thesis of estrogen receptors, establishing a positive feedback loop on itself. In the uterine endometrial Hypothalamus cycle, estrogen induces proliferation of the endo- metrial glands (Knobil 1999). LHRH In the presence of a GnRH pulse, the pituitary and ovarian hormones exert mutual control over the circulating levels of each other. The complex inter- actions between pituitary and ovarian hormones involve forward control, positive feedback, and negative feedback mechanisms. They also serve to Pituitary sustain a self-perpetuating monthly endocrine cycle. Figure 19.2 illustrates the relationships between the LH FSH relative amounts of key hormones of the menstrual cycle. Day one of the menstrual cycle begins with menstruation, which occurs at the beginning of the follicular phase. The follicular phase of the menstrual cycle spans the first day of menstruation until ovulation. The Ovaries primary purpose of the follicular phase is to develop a viable follicle capable of undergoing ovulation. The early events of the follicular phase are initiated by a rise in FSH levels that occurs on the 1st day of the cycle. This rise in FSH levels can be attributed Progesterone Estrogen to a decrease in progesterone and estrogen levels at the end of the previous cycle and the subsequ- Fig. 19.1 The hypothalamic–pituitary–gonadal (HPG) axis. FSH, follicle-stimulating hormone; LH, luteinizing ent removal of their inhibitory effect on FSH. FSH hormone; LHRH, luteinizing hormone-releasing stimulates the development of 15–20 follicles each hormone. month and stimulates follicular secretion of estra- diol by up-regulating secretion of androgens by Estrogen is produced at the level of the ovary the theca externa, and by inducing the aromatase and is crucial for the development of the antrum enzyme receptor on granulosa cells (Yen 1999). FSH and maturation of the Graafian follicle. Estrogen is further induces expression of FSH receptors by fol- predominant at the end of the follicular phase, licles. As estradiol levels increase under the influ- directly preceding ovulation. Estradiol is primarily ence of FSH, estradiol inhibits the secretion of FSH derived from androgens produced by thecal cells. and FSH levels decrease. The androgens migrate from the thecal cells to the Under normal circumstances, one follicle evolves granulosa cells, where they are converted into estra- into the dominant follicle, destined for ovulation, diol by aromatase enzyme. The actions of estradiol while the remaining follicles undergo atresia. It is include induction of FSH receptors on granulosa currently not known how the dominant follicle is cells, proliferation and secretion of follicular thecal selected; yet it has been observed that the dominant cells, induction of LH receptors on granulosa cells, follicle always expresses an abundance of FSH and proliferation of endometrial stromal and epi- receptors. As FSH levels decrease towards the end thelial cells. At low circulating levels, estrogens of the follicular phase, the developing follicles must 252 chapter 19

1) Control by hypothalamus Inhibited by combination of Hypothalamus estrogen and progesterones Stimulated by high levels Releasing hormone of estrogen Anterior pituitary

FSH LH 2) Pituitary hormones in blood LH peak triggers ovulation and corpus luteum formation

LH

FSH

FSH LH 3) Ovarian cycle

Growing follicle Corpus Degenerating Maturing Ovulation luteum corpus luteum follicle Pre-ovulatory phase Post-ovulatory phase

Estrogen Progesterone and estrogen 4) Ovarian hormones in blood

Estrogen Progesterone Progesterone Estrogen and estrogen 5) Menstrual cycle

Endometrium

Fig. 19.2 The relationships between the relative amounts of key hormones 0 5101514 20 25 28 of the menstrual cycle. FSH, follicle- stimulating hormone; LH, luteinizing Menstruation Days hormone. compete for relatively small amounts of FSH. The FSH level they cease to develop and ultimately dominant follicle, with its high concentration of undergo atresia. The dominant follicle may also FSH receptors is able to acquire more FSH even release paracrine factors that stimulate apoptosis as FSH levels decrease. This enables the dominant in the other follicles. As the dominant follicle con- follicle to continue to synthesize estradiol which tinues to mature it secretes increasing amounts is essential for its complete maturation. Since the of estrogen. Estrogen levels peak towards the end of remaining follicles can no longer produce the the follicular phase of the menstrual cycle. At this needed amount of estradiol due to the decreasing critical point, estrogen exerts positive feedback on oral contraceptive use and physical performance 253

LH (positive feedback is usually associated with tions of the fallopian tube bring the oocyte into con- having estradiol present in excess of 1100 pmol·L–1 tact with the tubal epithelium to initiate migration for > 24 h), generating a dramatic preovulatory LH through the oviduct. surge. It is important to note that estrogen can only Generally, estrogen (estradiol) is produced con- exert a positive feedback on LH at this precise tinuously during the cycle. It is very low during stage in the menstrual cycle; if estrogen is artificially the early follicular phase, rising to a peak during provided earlier in the cycle, ovulation will not be the late follicular phase and triggering ovulation. induced (Yen 1999; Larsen et al. 2003). Progesterone and estrogen increase during the The luteal phase is characterized by the luteiniza- luteal phase, and the increased estrogen causes tion of those components of the follicle that were the endometrium, the inner lining of the uterus, to not ovulated. It is initiated by the LH surge. The thicken and mature. Progesterone helps to mature granulosa cells, theca cells and some surrounding and stabilize the endometrium. It also prevents connective tissue are all converted into the corpus further endometrial proliferation and mitosis and luteum, which eventually undergoes atresia. The changes the endometrium to a secretory structure major effects of the LH surge are the conversion that is ready for implantation of the fertilized ovum. of granulosa cells from predominantly androgen- At the end of the luteal phase, concentrations of converting cells to predominantly progesterone- estrogen and progesterone decrease, which triggers synthesizing cells by the expression of new LH menstruation and the subsequent breakdown of receptors. This fosters increased progesterone syn- the endometrium. Breakdown of the endometrium thesis, and a reduced affinity of granulose cells for occurs assuming no implantation of an ovum has estrogen and FSH in these cells. Combined, these occurred during this period. changes promote increased progesterone secretion Inhibin is a glycoprotein that has been isolated with some estrogen secretion. Progesterone secre- as a heterodimer with two subunits linked by dis- tion by the corpus luteum peaks between 5 and 7 ulphide bonds. The two isoforms of inhibin are days post-ovulation. High progesterone levels exert termed inhibin A and inhibin B. Inhibin is produced negative feedback on GnRH and subsequently by the granulosa cells and the corpus luteum in GnRH pulse frequency decreases. As GnRH pulse females (deKrester & Robertson 1989; Groome et al. frequency decreases, FSH and LH secretion also 1996). Control of inhibin synthesis and release is via decreases. The corpus luteum further loses its FSH the hormones FSH (granulosa cells) and LH (corpus and LH receptors. The lack of FSH and LH stimula- luteum). Level of inhibin (serum) peak mid-cycle tion precipitates atresia of the corpus luteum and and in the mid-luteal phase, with a decline in con- its subsequent evolution into the corpus albicans. centrations prior to menstruation (Sehested et al. With the decline of both estrogen and progesterone 2000). Inhibin has FSH-suppressing properties, as levels, an important negative feedback control on well as, more recently discovered immune, nervous FSH is removed and FSH levels rise once again to system and hemopoietic properties (deKrester & initiate the next menstrual cycle (Yen 1999). Robertson 1989; Sehested et al. 2000). In addition, The LH surge is required for ovulation. Under the inhibin may have paracrine actions related to influence of LH, the primary oocyte enters the final growth and maturation of the ovary and classical stage of the first meiotic division and divides into a endocrine actions in the maturation of the HPG axis secondary oocyte and the first Barr body. The LH (Groome et al. 1996; Sehsted et al. 2000). Additional surge induces release of proteolytic enzymes, which research is necessary to fully elucidate the role of degrade the cells at the surface of the follicle, and inhibin in human physiology. stimulates angiogenesis in the follicular wall and prostaglandin secretion. These effects of LH cause Overview of oral contraceptives the follicle to swell and rupture. At ovulation, the oocyte is expelled into the peritoneal cavity. The Oral contraceptive pills are available in three major oocyte adheres to the ovary and muscular contrac- formulationsafixed-dose, combination phasic and 254 chapter 19

daily progestin. Monophasic pill formulations con- cannot be overlooked when discussing elite sport tain a fixed dose of estrogen and progestin through- performance. out the cycle. Biphasic and triphasic (multiphasic) pills reduce the total amount of progestin through- Mechanism of action of oral out the cycle as to mimic normal physiologic levels. contraceptives Ethinyl estradiol and mestranol are typical synthetic estrogens used in oral contraceptive formulas. Older The mechanism by which estrogen and progesterone formulations contained mestranol in high doses prevent ovulation is through the suppression of the (50 µg) compared to the newer low dose formula- FSH and LH mid-cycle surge. Combination formu- tions (< 50 µg of estrogen compounds ethinyl estra- lations are highly effective at inhibiting the secretion diol or mestranol). First generation progestins in- of gonadotrophic hormone, and through this mech- clude norethindrone and norethindrone acetate or anism prevent ovulation. Conversely, progestin- ethynodiol diacetate and norethynodrel. Second only products are inconsistent in the suppression of generation progestins are typically norgesterel and ovulation and operate through the inhibition of the levonorgestrel (Larsen et al. 2003). Finally, desoges- release of GnRH from the hypothalamus. In this trel, norestimate and gestrodene are third genera- way, progestin only products attenuate the levels of tion progestins. The third generation progestins are FSH, LH, estradial and progesterone (Speroff et al. designed to be selective for progestin receptors, 1993). thus reducing the atherogenic properties of these These induced changes in circulating hormone compounds (Godsland et al. 1992). Table 19.1 lists levels precipitate changes in endometrial function- oral contraceptives that are currently available in ing. Changes in ovum transport and implantation, the USA. It is important to note that the new genera- coupled with alterations in the composition of the tion progestins are assumed to be less androgenic cervical mucus (increase viscosity and reduce vol- compared to the early formulations. However, all ume) are also evident (Larsen et al. 2003). Together, combination oral contraceptives are beneficial in the the hormonal and endometrial alterations result clinical treatment of various conditions including in a negative environment for success of ovulation, hirsuitism (Thorneycroft 1999). Low dose oral con- implantation and support of the ovum. traceptives are considered to have a lipid neutral Oral contraceptives also impact carbohydrate, affect resulting in a reduced atherosclerosis rates in protein and fat metabolism (Table 19.2). Carbohy- humans. drate metabolism is influenced by the dose, potency First generation oral contraceptives were associ- and chemical structure of the progestin in the ated with many negative side effects such as weight drug. The role of the synthetic estrogens on glucose gain, fatigue, headaches and nausea (Lebrun et al. metabolism is unclear, however they may act syn- 2003). These side effects have been minimized by ergistically with the progestins to cause impaired the second and third generation lower dose drugs. glucose tolerance. Epidemiological studies indicate These formulations have 30–40% lower levels of low risk or no risk for developing diabetes in oral hormones resulting in significantly reduced side contraceptive users (new generation). However, it is effects (Greenblatt 1985). Studies evaluating changes prudent to consider the glucose handling status of in body weight and body composition in both non- each woman when prescribing oral contraceptives. active and athletic women using oral contraceptives Low dose norethindrone-type progestins are prefer- support these findings. Female athletes (Lebrun able in these women. et al. 2003) and non-athletes (Rosenberg 1998) report Estrogens used in oral contraceptives increase the non-significant weight gains (~ 1 kg over 6 weeks) hepatic production of various globulins such as or no change in body weight and body composi- Factor V, VII, X and fibrinogen. These globulins tion, respectively. However, the impact of small enhance thrombosis. In addition, synthetic estro- body weight and body composition changes asso- gens also increase the synthesis of angiotensinogen ciated with oral contraceptive use in female athletes (which is converted to angiotensin) resulting in elev- oral contraceptive use and physical performance 255

Table 19.1 Oral contraceptive formulations. (Data from Mishell 1999 and Larsen et al. 2003.)

Product Brand name type Progestagen Estrogen

Levlen C 0.15 mg levonorgestrel 30 µg ethinylestradiol Tri-Levlen C, T 0.5 or 0.075 or 0.125 mg levonorgestrel 30 µg/40 µg /30 µg ethinylestradiol Ovcon-35 C 0.4 mg norethindrone 35 µg ethinylestradiol Ovcon-50 C 1.0 mg levonorgestrel 50 µg ethinylestradiol Desogen C 0.15 mg desogestrel 35 µg ethinylestradiol Mircette B, C 0.15 mg/0 mg desogestrel 30 µg/10 µg ethinylestradiol Micronor P 0.35 mg norethindrone Modicon C 0.5 mg norethindrone 35 µg ethinylestradiol Ortho-Cept C 0.15 mg desogestrel 30 µg ethinylestradiol Ortho-Cyclen C 0.25 mg norgestimate 35 µg ethinylestradiol Ortho-Novum 1/35 C 1.0 mg norethindrone 35 µg ethinylestradiol Ortho-Novum 1/50 C 1.0 mg norethindrone 50 µg mestranol Ortho-Novum C, T 0.5 mg/0.75 mg/L mg norethindrone 35 µg ethinylestradiol Ortho-Novum C, B 0.5 mg/1.0 mg norethindrone 35 µg ethinylestradiol Ortho-tricyclin C, T 0.18 mg/0.215 mg/0.25 mg norgestimate 35 µg ethinylestradiol Estrostep C, T 1.0 mg norethindrone 20 µg/30 µg/35 µg ethinylestradiol Loestrin 1/20 C 1.0 mg norethindrone 20 µg ethinylestradiol Loestrin 1.5/30 C 1.5 mg norethindrone 30 µg ethinylestradiol Norlestrin 1/50 C 1.0 mg norethindrone 50 µg ethinylestradiol Norlestrin 2.5/50 C 2.5 mg norethindrone 50 µg ethinylestradiol Brevicon C 0.5 mg norethindrone 35 µg ethinylestradiol Norinyl 1 + 35 C 1.0 mg norethindrone 35 µg ethinylestradiol Norinyl 1 + 50 C 1.0 mg norethindrone 50 µg ethinylestradiol Nor-Q.D. P 0.35 mg norethindrone Tri-Norinyl C, T 0.5 mg/1 mg/0.5 mg norethindrone 35 µg ethinylestradiol Demulen 1/35 C 1.0 mg ethynodiol diacetate 35 µg ethinylestradiol Demulen 1/50 C 1.0 mg ethynodiol diacetate 50 µg ethinylestradiol Alesse C 0.1 mg levonorgestrel 20 µg ethinylestradiol Lo/Ovral C 0.3 mg levonorgestrel 30 µg ethinylestradiol Nordette C 0.15 mg levonorgestrel 30 µg ethinylestradiol Ovral C 0.5 mg norgestrel 50 µg ethinylestradiol Ovrette P 75 mg norgestrel 30 µg ethinylestradiol Triphasil C, T 50 mg/75 mg/125 mg levonorgestrel 40 µg /40 µg /30 µg ethinylestradiol

B, biphasic; C, combination; P, progestagen only; T, triphasic. Androgenic activity (relative to 1 mg of norethindrone: norethindrone (1 mg) = 1; levonorgestrel (1 mg) = 8.3; drospirenone (1 mg) = 0; desogestrel (1 mg) = 3.4; norgestimate (1 mg) = 1.9; ethynodiol diacetate (1 mg) = 0.6). ated blood pressure in some women (Larsen 2003). total cholesterol and triglycerides, and a decrease The androgenic progestins increase the synthesis of in low-density lipoproteins (LDL) (Larsen 2003). sex hormone binding globulin (SHBG). SHBG binds Conversely, progestin elements cause a decrease in the 19-nortestosterone compounds used in oral con- HDL, total cholesterol and triglycerides, and an traceptives. Overall, the incidence of venous and increase in LDL. The negative impact of the pro- arterial thrombosis is directly related to the dose gestins in the early oral contraceptives was related of the estrogen, which is relatively low in the new to the high androgenic activity of these compounds. generation formulations. Today’s low dose oral contraceptives are less an- Synthetic estrogens used in oral contraceptives drogenic and appear to have a more lipid neutral cause an increase in high-density lipoproteins (HDL), response (Thorneycroft 1999). When evaluating the 256 chapter 19

Table 19.2 Metabolic effects of oral contraceptives. (Data from Dorflinger 1985 and Godsland et al. 1992.)

Steroid derivative Substrate pathway Metabolic effect

Ethinyl estradiol Protein Decrease amino acids Ethinyl estradiol Carbohydrate No change in plasma insulin or glucose tolerance Ethinyl estradiol Lipid Increase cholesterol, HDL and triglyceride Decrease LDL cholesterol 19-Nortestosterone derivatives Proteins None 19-Nortestosterone derivatives Carbohydrate Increase plasma insulin Decrease glucose tolerance 19-Nortestosterone derivatives Lipid Decrease cholesterol, HDL and triglyceride Increase LDL cholesterol

HDL, high-density lipoprotein; LDL, low-density lipoprotein.

metabolic response to various oral contraceptives, it type of oral contraceptives utilized. Research in is critical that the type, dose and combination of the early 1980s failed to demonstrate any significant each synthetic hormone be assessed. difference in performance between oral contracept- ive users and non-users (McNeill & Mozingo 1981; Impact of oral contraceptive use on Huisveld et al. 1983). A subsequent short duration physical performance (21 day) evaluation of 1 mg norethindrone also did not demonstrate a significant difference in o The literature regarding the impact of oral contra- V 2max between oral contraceptive users and non- ceptive use on performance has grown substantially users (Bryner et al. 1996). Interestingly, however, over the past several decades. Early studies explor- endurance trained women who used monophasic ing the effect of first generation, high dose drugs are oral contraceptives for a 6-month period experi- o often distinctly different from the more recent work enced a significant decrease in V 2peak (Notelovitz evaluating the low dose multiphasic compounds. et al. 1987). Further evaluation of a triphasic pre- While the focus of this review will be on the second paration performed by Casazza et al. (2002) with and third generation drugs that are currently being moderately active women also showed a significant o used by women participating in sport, earlier work reduction (–11%) in V 2peak following 4 months of will be discussed when appropriate. oral contraceptive use. Peak heart rate and minute ventilation, however, were not different These results were supported by the work of Lebrun Aerobic performance o et al. (2003) who showed a 4.7% decrease in V 2max o Maximal oxygen consumption (V 2max) is an often following a 2-month triphasic oral contraceptive used measure when assessing aerobic ability. intervention protocol. Normal ovarian hormone cyclicity does not appear Potential factors that could impact maximal o to affect V 2max (De Souza et al. 1990; Bemben et al. aerobic capacity include stroke volume, oxygen 1992, 1995; Lebrun et al. 1995; Lynch & Nimmo transport capacity, oxygen extraction capacity and o 1998); however, data examining the impact of hor- muscle blood flow. The reductions in V 2max may monal alterations as induced by changes in exogen- logically be associated with a reduction in stroke ous hormones, such as would occur with oral volume and/or oxygen transport capacity. Stroke contraceptive use, is less conclusive. volume has been shown to increase in perimeno- Studies evaluating the administration of exogen- pausal women following hormone replacement ous ovarian hormones have varied in the dose and therapy (Kamali et al. 2000). However, resting oral contraceptive use and physical performance 257

ferritin, hemoglobin concentration and serum iron tural muscle damage is a typical consequence of concentration remain unchanged in women using strenuous exercise such as resistance training. The oral contraceptives (Mooij et al. 1992) which would role of estradiol on membrane structure and per- suggest that these primary factors are not the mech- meability has been evaluated in both animal and anism by which aerobic potential is limited. human models. Blood flow regulation via mechanisms such as Women have responded to strenuous exercise sympathetic nervous system (SNS) activation has with lower serum creatine kinase (CK) concentra- also been cited as a factor that may be responsible tions compared to men, which has been associated for the observed reductions in maximal aerobic with lower total muscle mass in women (Shumate capacity. The evaluation of various conditions where et al. 1979; Rogers et al. 1985). However, animal studies ovarian hormones are elevated provides experi- support the role of estradiol in reducing membrane mental evidence for this mechanism. In pregnancy permeability in response to exercise (Amelink et al. where there is a period of high estrogen and proges- 1988, 1990; Bar & Amelink 1997). terone concentrations SNS activation is suppressed, Research evaluating the role of estradiol on in conjunction with a decreased level of circulating muscle damage following exercise in women using catecholamines (McMurray et al. 1993). In this way, oral contraceptives is both limited and equivocal. a reduction in SNS activation with oral contracept- Miles and Schneider (1993) reported no influence ive administration might also be responsible for the of estradiol on CK activity following exercise. The decreases in aerobic capacity shown in these studies. relationship between delayed onset muscle sore- ness and oral contraceptive use was reported fol- lowing a 50-min stepping routine in women using Anaerobic and strength performance and oral contraceptives compared to controls. Oral con- muscle damage traceptives were associated with a reduction in self- The current literature on the relationship of exo- reported soreness scores; however, the mechanism genous administration of ovarian hormones and for these findings remains unclear (Thompson et al. anaerobic metabolism and/or muscular strength is 1997). extremely limited. Early work by Petrofsky et al. Recent studies have utilized eccentric muscle (1976) identified a reduction in muscle strength damage protocols. Downhill running (30 min) was during the luteal phase of normally cycling females. used by Carter et al. (2001) to elicit damage in oral The potential for oral contraceptive use to blunt contraceptive users and non-users. Both groups this strength decline was subsequently evaluated in experienced significantly elevated CK and muscle a group of athletic women by Lebrun et al. (2003). soreness scores compared to baseline measures. Anaerobic performance on the anaerobic speed test Differences were evident between the groups at and isokinetic strength measured on the Cybex 72 h following damage with the oral contraceptive dynamometer was determined prior to and follow- users exhibiting a reduction in CK compared to the ing administration of a triphasic formulation for control group. These data support a protective role 2 months. Although this study failed to find a sig- of estrogen following muscle damage. In contrast, nificant difference in either anaerobic performance Savage and Clarkson (2002), using a 50 repetition or isokinetic strength, the relationship of high levels eccentric muscle contraction protocol of the elbow of ovarian hormones to anaerobic and/or strength flexors, showed no difference in measures of sore- performance capacity is significantly underexam- ness, serum CK or range of motion of the elbow ined and must be further researched before any joint. However, oral contraceptive users had delayed definitive conclusions can be established. force recovery (maximal isometric strength) at 2 The influence of endogenous and/or exogenous days following muscle damage. estrogens on exercise induced muscle damage has The role of exogenous estrogens on muscle been evaluated over the past decade. Ultrastruc- damage following exercise remains unclear. Animal 258 chapter 19

studies support the protective role (anti-oxidant illustrates the influence of estrogen ethinyl estradiol characteristics) of estradiol on membrane perme- and the 19-nortestosterone derivative progrestins ability. Human studies vary in methodology and on metabolism. The influence of the oral contra- outcome measures. Further research is necessary to ceptives is directly related to dosage and potency clarify the potential benefits of estrogen on muscle of the steroid derivative in the formulation. The stability following strenuous physical activity. metabolic effects at rest have been evaluated in relation to side effects of the drugs. Only recently have scientists begun to understand the role of oral Heat tolerance contraceptive use in female athletes during physical Menstrual cycle phase is associated with varied exercise. responses to thermal stressors. Exercise during the Current research on the metabolic alterations luteal phase is associated with an elevated core tem- associated with oral contraceptive use during perature (0.4°C) compared to the follicular phase exercise is limited. Cross-sectional studies have (Stachenfeld et al. 2000). In contrast, oral contracept- compared oral contraceptive users with non-users. ive users have less variability in core temperature Early work by Bonen et al. (1991) reported signi- between phases than non-users. The primary factor ficantly elevated free fatty acid concentrations for these phasic alterations in core temperature is in oral contraceptive users when compared with the oscillations in progesterone across the men- non-users during basal or resting conditions. This strual cycle (elevation during the luteal phase). Use tendency also appears to occur during exercise of oral contraceptives containing progestins would with reductions in respiratory exchange ratio (RER) cause elevations in core temperature during exer- and a potential enhancement in lipid oxidation also cise similar to women during the luteal phase. The evident in oral contraceptive users (Bemben et al. influence of progesterone on core temperature is 1995). further supported by the response of women using More recently, studies using stable isotope tech- the long-term contraceptive method of injectable nology to evaluate both glucose and lipid metabol- progesterone (i.e. Depot-Provera [Cheung et al. ism during exercise and at rest have demonstrated 2000]). These women experience an elevated core a decrease in glucose rate of appearance (Ra), dis- temperature during the 24–36 h following admin- appearance rate (Rd) and metabolic clearance rate istration of the drug. Martin and Buono (1997) in women using oral contraceptive during exercise o reported elevated core temperature (0.3°C) and on a cycle ergometer (45% or 65% V 2peak) (Suh et al. heart rate (8 b·min–1) in women using oral contra- 2003). Under similar exercise conditions, triglycer- ceptives containing synthetic progestins while exer- ide mobilization was measured using glycerol rate cising in the heat (30°C, 50% relative humidity). of appearance (infusion of [1,1,2,3,3-2H]glycerol). These changes in core temperature and heart rate Glycerol Ra increased significantly at both exercise are similar to those seen in women exercising in the intensities in the oral contraceptive condition. The heat during the luteal phase. It appears that second elevated Ra was associated with a significantly and third generation oral contraceptives mimic the increased plasma cortisol concentration but no dif- thermoregulatory responses of women exercising in ference in RER (Casazza et al. 2004). These data a hot environment during the luteal phase. Further would suggest that endogenous ovarian hormones research is necessary to more clearly understand the do play a role in affecting glucose and lipid meta- impact of low dose contraceptive use on perform- bolism during exercise. The exercise induced hor- ance in hot and humid environments. monal perturbations clearly dictate the metabolic flux in normal cycling women. Oral contraceptives (triphasic) provide a greater stimulus to lipid and Substrate utilization and metabolic flux glucose flux during exercise conditions. The level of The metabolic effects of contraceptive steroids circulating ovarian hormones appears to be critical varied depending on the formulation. Table 19.2 in these metabolic alterations. oral contraceptive use and physical performance 259

Summary Potential thermoregulatory issues exist for women using oral contraceptives; however, the responses Oral contraceptive use has grown in females parti- mimic women exercising during the luteal phase of cipating in sport and physical activity. The role that the menstrual cycle. The most prominent finding these exogenous ovarian hormones play in physical regarding oral contraceptive use during exercise performance remains to be fully elucidated. It is is on substrate metabolism. Studies using stable clear that third generation lower dose formulations isotope methodologies have provided significant result in reduced side effects for women. These insight into the interplay between lipid and glucose changes have made the use of oral contraceptives metabolism during exercise in women using exogen- more appealing to a broad group of female athletes ous hormones. including elite caliber performers. The literature examining oral contraceptive use The reductions seen in aerobic capacity appear during exercise is relatively limited. It is important to be associated with potential alterations in blood to consider the potential negative or ergogenic flow due to suppression of SNS activation coupled effects of exogenous ovarian steroid use in elite with a reduction in circulating catecholamines. These female athletes. Longitudinal studies of longer dura- findings are similar to those seen during pregnancy. tions are necessary to provide practical recommenda- Anaerobic and/or strength performance appear tions to female athletes. to be unchanged during oral contraceptive use.

References

Amelink, G.J., Kamp, H.H. & Bar, P.R. Bryner, R.W., Toffle, R.C., Ullrich, I.H. & Medicine and Science in Sports and Exercise (1988) Creatine kinase isoenzyme Yeater, R.A. (1996) Effect of low dose 22(5), 575–580. profiles after exercise in the rat: sex- oral contraceptives on exercise Dorflinger, L. (1985) Relative potency of linked differences in leakage of CK-MM. performance. British Journal of Sports progestins used in oral contraceptives. Pflugers Archiv 412, 417–421. Medicine 30, 36–40. Contraception 557, 31–37. Amelink, G.J., Koot, R.W., Erich, W.B.M., Casazza, G.A., Suh, S.H., Miller, B.F., Godsland, I.F., Crook, D. & Wynn, V. Van Gijn, J. & Bar, P.R. (1990) Sex-linked Navazio, F.M. & Brooks, G.A. (2002) (1992) Clinical and metabolic variation in creatine kinase release, and Effects of oral contraceptives on peak considerations of long-term oral its dependence on oestradiol, can be exercise capacity. Journal of Applied contraceptive use. American Journal of demonstrated in an in-vitro rat skeletal Physiology 93, 1698–1702. Obstetrics and Gynecology 166, 1955–1964. muscle preparation. Acta Physiologica Casazza, G.A., Jacobs, K.A., Suh, S.H. et al. Greenblaat, R.B. (1985) Oral Scandinavica 138, 115–124. (2004) Menstrual cycle phase and oral contraceptives: the state of the art. Bar, P.R. & Amelink, G.J. (1997) Protection contraceptive effects on triglyceride Clinical Therapeutics 8, 6–27. against muscle damage exerted by mobilization during exercise. Journal of Groome, N.P., Illingworth, P.J., O’Brien, oestrogen: hormonal or antioxidant Applied Physiology 97(1), 302–309. M. et al. (1996) Measurement of dimeric action? Biochemical Society Transactions Carter, A., Dobridge, J. & Hackney, A.C. inhibin B throughout the human 25, 50–54. (2001) Influence of estrogen on markers menstrual cycle. Journal of Clinical Bemben, D.A., Boileau, R.A., Bahr, J.M., of muscle tissue damage following Endocrinology and Metabolism 81, Nelson, R.A. & Misner, J.E. (1992) Effects eccentric exercise. Fiziologiia Cheloveka 1401–1405. of oral contraceptives on hormonal and 27, 133–137. Huisveld, I.A., Haspers, J.E.H. & metabolic responses during exercise. Cheung, S.S., McLellan, T.M. & Tenaglia, S. Bernink, M.J. (1983) The effect of oral Medicine and Science in Sports and Exercise (2000) The thermophysiology of contraceptives and exercise on 24, 434–441. uncompensatable heat stress. hemostatic and fibrinolytic mechanisms Bemben, D.A., Salm, P.C. & Salm, A.J. Physiological manipulations and in trained women. International Journal of (1995) Ventilatory and blood lactate individual characteristics. Sports Sports Medicine 4, 97–103. responses to maximal treadmill exercise Medicine (Auckland, N.Z.) 29, 329–359. Kamali, P., Muller, T., Lang, U. & during the menstrual cycle. Journal of deKretser, D.M. & Robertson, D.M. Clapp, J.F., 3rd. (2000) Cardiovascular Sports Medicine and Physical Fitness 35, (1989) The isolation and physiology responses of perimenopausal women to 257–262. of inhibin and related proteins. hormone replacement therapy. American Bonen, A., Haynes, F.J. & Graham, T.E. Biology of Reproduction 40, 33–47. Journal of Obstetrics and Gynecology 182, (1991) Substrate and hormonal De Souza, M.J., Maguire, M.S., Rubin, K.R. 17–22. responses to exercise in women using & Maresh, C.M. (1990) Effects of Knobil, E. (1999) The wisdom of the body oral contraceptives. Journal of Applied menstrual phase and amenorrhea on revisited. News in Physiological Sciences Physiology 70, 1917–1927. exercise performance in runners. 14, 1–11. 260 chapter 19

Larsen, P.R., Kronenberg, H.M., Melmed, of Laboratory Clinical Medicine 122, adolescent girls and adult women: S. & Polonsky, K.S. (2003) Williams 197–201. relation to age, stage of puberty, Textbook of Endocrinology. W.B. Saunders, Mishell, D.R. (1999) Contraception. In: menstrual cycle, follicle-stimulating Co., Philadelphia, PA. Reproductive Endocrinology: Physiology, hormone, luteinizing hormone, and Lebrun, C.M., McKenzie, D.C., Prior, J.C. & Pathophysiology and Clinical Management estradiol levels. Journal of Clinical Taunton, J.E. (1995) Effects of menstrual (Yen, S.S.C., Jaffe, R.B. & Barbieri, R.L., Endocrinology and Metabolism 85, cycle phase on athletic performance. eds.). W.B. Saunders, Co., Philadelphia, 1634–1640. Medicine and Science in Sports and Exercise PA: 676–708. Shumate, J.B., Brooke, M.H., Carroll, J.E. 27, 437–444. Mooij, P.N.M., Thomas, C.M.G., Doesburg, & Davis, J.E. (1979) Increased serum Lebrun, C.M., Petit, M.A., McKenzie, D.C., W.H. & Eskes, T.K.A.B. (1992) The creatine kinase after exercise: a sex- Taunton, J.E. & Prior, J.C. (2003) effects of oral contraceptives and linked phenomenon. Neurology 29, Decreased maximal aerobic capacity multivitamin supplementation on serum 902–904. with use of a triphasic oral contraceptive ferritin and hematological parameters. Speroff, L., DeCherney, A. & the Advisory in highly active women: a randomized International Journal of Clinical Board for the New Progestins (1993) controlled trial. British Journal of Sports Pharmacology and Therapeutics 30, 57–62. Evaluation of the new generation of oral Medicine 37, 315–320. Notelovitz, M., Zauner, C. & McKenzie, I. contraceptives. Obstetrics and Gynecology Lynch, N.J. & Nimmo, M.A. (1998) (1987) The effect of low-dose 81, 1034–1044. Effects of menstrual cycle and oral contraceptives on cardiorespiratory Stachenfeld, N.S., Silva, C. & Keefe, contraceptive use on intermittent function, coagulation, and lipids in D.L. (2000) Estrogen modifies the exercise. European Journal of Applied exercising young women: a preliminary temperature effects of progesterone. Physiology 78, 565–572. report. American Journal of Obstetrics and Journal of Applied Physiology 88, 1643–1649. McMurray, R.G., Mottola, M.F., Wolfe, Gynecology 156, 591–598. Suh, S.H., Casazza, G.A., Horning, M.A., L.A. et al. (1993) Recent advances in Petrofsky, J.S., LeDonne, D.M. & Rinehart, Miller, B.F. & Brooks, G.A. (2003) Effects understanding maternal and fetal J.S. (1976) Isometric strength and of oral contraceptives on glucose flux responses to exercise. Medicine and endurance during the menstrual cycle. and substrate oxidation rates during rest Science in Sports and Exercise 25, European Journal of Applied Physiology 35, and exercise. Journal of Applied Physiology 1305–1321. 1–10. 94, 285–294. McNeill, A.W. & Mozingo, E. (1981) Rogers, M.A., Stull, G.A. & Apple, F.S. Thompson, H.S., Hyatt, J.P., De Souza, Changes in the metabolic cost of (1985) Creatine kinase isoenzyme M.J. & Clarkson, P.M. (1997) The effects standardized work associated with activities in men and women following of oral contraceptives on delayed onset the use of an oral contraceptive. a marathon race. Medicine and Science muscle soreness following exercise. Journal of Sports Medicine 21, in Sports and Exercise 17, 679–682. Contraception 56, 59–65. 238–244. Rosenberg, M. (1998) Weight change with Thorneycroft, I.H. (1999) Update on Martin, J.G. & Buono, M.J. (1997) Oral oral contraceptive use and during the androgenicity. American Journal contraceptives elevate core temperature menstrual cycle. Contraception 58, 45–49. of Obstetrics and Gynecology 180, 288–294. and heart rate during exercise in the Savage, K.J. & Clarkson, P.M. (2002) Oral Yen, S.S.C. (1999) The human menstrual heat. Clinical Physiology (Oxford, England) contraceptive use and exercise-induced cycle: neuroendocrine regulation. In: 17, 401–409. muscle damage and recovery. Reproductive Endocrinology: Physiology, Miles, M.P. & Schneider, C.M. (1993) Contraception 66, 67–71. Pathophysiology and Clinical Management Creatine kinase isoenzyme MB may be Sehested, A., Juul, A., Andersson, A.M. (Yen, S.S.C., Jaffe, R.B. & Barbieri, R.L., elevated in healthy young women after et al. (2000) Serum inhibin A and inhibin eds.). W.B. Saunders, Co., Philadelphia, submaximal eccentric exercise. Journal B in healthy prepubertal, pubertal and PA: 191–217. Chapter 20

Energy Balance and Exercise-Associated Menstrual Cycle Disturbances: Practical and Clinical Considerations

NANCY I. WILLIAMS AND MARY JANE DE SOUZA

Introduction might be a contributing factor to alterations in menstrual cyclicity in female athletes. The first Historically, opportunities for women to engage in links between exercise-associated amenorrhea and physical activity increased dramatically with the compromised bone were made by Drinkwater et al. passing of Title IX in the USA. To date, over three (1984) and Cann et al. (1984) who observed signific- decades of subsequent research examining the antly lower bone mineral content in amenorrheic impact of exercise on women’s bodies and beha- runners. Drinkwater et al. (1990) then documented vioral aspects of the athletic lifestyle have brought a significant relationship between lumbar spine attention to gender-specific issues such as muscu- bone mineral density (BMD) and menstrual history loskeletal health, weight and diet concerns, and the in female athletes. An authoritative review by impact of exercise on the menstrual cycle. Looking Loucks and Horvath in 1985 (Loucks & Horvath back, it appears that research on the musculoskel- 1985) helped highlight the importance of exercise- etal effects of exercise and weight and diet concerns associated menstrual disturbances (EAMD) as a developed soon after observations of disturbances research area and synthesized the available informa- in reproductive function associated with exercise. tion concerning plausible mechanisms of the inter- After an early report documenting a higher pre- actions between exercise and reproductive function. valence of menstrual abnormalities in athletes Researchers’ attention was focused on the potential than non-athletes (Erdelyi 1962), numerous other contributions of body composition, training habits, studies confirmed this finding in the late 1970s and diet, physical stress and psychological stress in the early 1980s (Feicht et al. 1978; Dale et al. 1979; Baker etiology of EAMD. In 1987, a link between ‘eating et al. 1981), focusing mostly on amenorrhea and problems’ and amenorrhea in ballet dancers was delayed menarche in long-distance runners (Feicht reported by Brooks-Gunn et al. (1987), and in 1991, et al. 1978) and ballet dancers (Frisch et al. 1980; Wilmore (1991) put forth the idea that athletes in Warren 1980). Interestingly, while Frisch and Revelle both endurance or appearance sports are at an (1971) had put forth the hypothesis that menarche increased risk for disordered eating, secondary depended on the achievement of a critical body amenorrhea and bone mineral disorders. Shortly weight level of body fat, Warren (1980) observed after, the female athlete triad (Fig. 20.1), i.e. a con- that changes in body weight and body composition dition comprised of restrictive eating, menstrual did not correlate with the onset of menarche in disorders and skeletal demineralization was recog- dancers who acutely decreased their exercise due to nized (Yeager et al. 1993), and subsequently described injury. Warren thus popularized the concept that in a Position Stand published by the American the ‘energy drain’ or metabolic cost of exercise College of Sports Medicine (ACSM) in 1997 (Otis 261 262 chapter 20

Is energy deficiency the only factor contributing to EAMD?

Disordered eating Fig. 20.1 The relationship between components of the female athlete ? Psychological triad. Disordered eating and stress exercise combined to produce low energy availability, causing The Energy exercise-associated menstrual female deficiency disturbances (EAMD). Psychosocial athlete stress is theoretically depicted as a triad factor that may exacerbate the effects of low energy availability Bone loss Exercise-associated on EAMD. (Modified and reprinted Endothelial menstrual disorders with permission from Williams dysfunction? 2003.)

et al. 1997). Since that time, recognition of the import- Exercise-associated menstrual ance of the triad and the need to more specifically disturbances: definitions and prevalence define the components, establish the prevalence, physiological underpinnings and the clinical con- Overview sequences has driven the research in the area of exercise and the menstrual cycle, and stimulated a Important aspects of the relationship between exer- revision of the 1997 ASCM Position Stand. A current cise and the menstrual cycle have been revealed understanding of the practical and clinical consid- through descriptive, observational, cross-sectional erations of the effects of exercise on the menstrual and prospective studies utilizing both survey cycle in today’s active woman must include the methods and physiological approaches. Our current broader scope of the female athlete triad, and con- understanding is that EAMD exist on a spectrum of sequently recognize the complexity of interrelation- severity, i.e. luteal phase defects (LPD), anovulation ships among its components. A multidisciplinary and amenorrhea. Prior to discussing the definitions approach to this issue involving behavioral, physio- and prevalence of the different EAMD, attention logical, sociocultural and medical perspectives is is warranted toward the appropriate use of terms. warranted. This chapter will provide current infor- The term ‘exercise-induced menstrual disturbances’ mation on the practical issues relating to the impact implies that menstrual disturbances are directly of exercise on menstrual function, such as the related to participation in or the stress of physical identification, prevention and treatment of men- exercise per se. Contrary to this idea, a variety of strual disorders, an exploration of factors that pre- experimental approaches have yielded evidence dispose certain individuals to menstrual disorders that inadequate caloric intake is a causal factor in and considerations for the restoration of normal menstrual disturbances with exercise (Laughlin & menstrual function. Clinical issues arising from the Yen 1996; Loucks et al. 1998; Williams 1998; Williams association of physical activity and disturbances in et al. 2001b; De Souza et al. 2003) such that a state of reproductive function will also be examined. These low energy availability created by an imbalance of include the impact of endocrine changes associ- energy consumed as food and expended through ated with menstrual disturbances on fertility, bone exercise is the key factor in the onset of reproductive health and possibly the cardiovascular system, and suppression. The use of a more fitting term is thus the association of menstrual disturbances with warranted, i.e. ‘exercise-associated’ or ‘exercise- restrictive eating. related’ menstrual disturbances. Since estimates of energy balance and the menstrual cycle 263

Continuum of menstrual disturbances in athletes

Ovulatory Anovulation Amenorrhea Oligomenorrhea Luteal phase defect

Fig. 20.2 Continuum of reproductive disturbances, ranging from ovulatory cycles, subtle presentations of luteal phase defects (LPD) and anovulatory cycles to the most severe menstrual disturbance, amenorrhea. Physically active women and athletes fluctuate between ovulatory cycles, LPD and anovulatory disturbances frequently. It also seems probable that amenorrheic athletes may experience LPD during recovery from amenorrhea. (Reprinted with permission from De Souza 2003.)

the prevalence of secondary amenorrhea are higher as failure to achieve menarche by the age of 16 years in athletes than in sedentary populations (Drew (Loucks & Horvath 1985), has been repeatedly 1961), there is a correlation between athletic amenor- reported in athletes participating in many sports, rhea and physical exercise, but there is not a causal but again particularly in the aesthetic sports. The relationship. later ages at menarche in adolescent athletes are often attributed to regular exercise training without considering other factors known to influence pub- Amenorrhea ertal growth and maturation (Malina 1994). Malina Figure 20.2 depicts this continuum of menstrual (1994) states that in adequately nourished adoles- disturbances ranging from subtle perturbations like cents, the timing of menarche is very dependent on LPD and anovulatory cycles to the most severe pre- hereditary factors; but menarche is also influenced sentation, amenorrhea (De Souza 2003). In athletes, by a number of social or biocultural variables, the amenorrhea is hypothalamic in origin and is including the self-selective nature of participation manifested by suppressed levels of gonadotropins in some sports, like gymnastics, figure skating and and ovarian steroids (Loucks & Horvath 1985; ballet, where selection occurs for specific factors Velduis et al. 1985). The definition of amenorrhea in associated with a delayed or later maturation the literature has varied, but should be conservat- (Malina 1994). The mechanisms that link these fac- ively defined as no menses for a minimum of 3 tors to a later age of maturation remain undefined. months due to the clinical sequelae associated with Following the review of several studies that exam- chronic estrogen deficiency (Loucks & Horvath ined menstrual disturbances in athletes, the average 1985). The prevalence of amenorrhea in athletes age of menarche was about 1 year later (13 vs. ranges from 1% to 66% (Feicht et al. 1978; Dale et al. 12 years old) in the groups of amenorrheic athletes, 1979; Schwartz et al. 1981; Sanborn et al. 1982, 2000; but these differences were not statistically significant Loucks & Horvath 1985), is highest in those sports in the individual studies. In both human (Warren with an aesthetic component like gymnastics and 1980) and animal studies (Cheung et al. 1997), how- figure skating, and grossly exceeds estimates of this ever, the timing of puberty has been related to nutri- condition in sedentary women (2–5%) (Drew 1961; tional status (Wade et al. 1996), and specifically, to Petterson et al. 1973). increased leptin concentrations providing a per- missive effect on growth (Cheung et al. 1997). Since leptin concentrations can discriminate athletes of Delayed menarche differing menstrual status (Laughlin & Yen 1997; De Delayed menarche, or primary amenorrhea, defined Souza et al. 2003), and evidence exists for a causal 264 chapter 20

role of energy availability in the development of document ovulation are not always feasible, many menstrual disturbances (Williams et al. 2001a), it is researchers have relied on daily assessment of reasonable to speculate that energy status may play urinary ovarian steroid metabolites and urinary LH a role in the common observance of a later age of to confirm ovulation (or anovulation). Using this menarche in amenorrheic athletes, if participation approach, De Souza et al. (1998a) have reported a in exercise began prior to puberty. 16% prevalence of anovulatory cycles in women that exercise at recreational levels despite having characteristic regular menstrual intervals of 26–32 Oligomenorrhea days. Williams et al. (2000) have reported that in 32% Oligomenorrhea is defined by irregular and incon- of Division 1 athletes from a wide variety of sports sistent menstrual cycles lasting from 36–90 days in who self-reported regular menstrual bleeding of length (Loucks & Horvath 1985). The key parameter 26–32 days, ovulation could not be detected. In of interest in the definition of oligomenorrhea is most cases, a higher degree of estrogen exposure is the unpredictable nature of the intervals between present in anovulatory cycles compared to amenor- menstrual cycles. Given these inconsistent charac- rheic cycles; that is an anovulatory cycle is likely to teristics, it is a menstrual presentation that is diffi- have greater E2 production in a 30-day period than cult to study. As such, no definitive data exist on an equivalent period in an amenorrheic athlete. In the prevalence of oligomenorrhea in athletes, except both of the latter conditions, the absence of proges- to note that cycles of irregular length are often terone from luteinized cells renders estrogen actions reported in female athletes (Loucks & Horvath on some tissues unopposed. 1985) and oligomenorrhea as a menstrual category LPD have been reported in women engaged in all is frequently grouped together with amenorrhea in levels of physical activity, from strenuous to recre- many studies (Gremion et al. 2001; Csermely et al. ational exercise, (Shangold et al. 1979; Ellison & 2002; Cobb et al. 2003). When daily measurement Lager 1986; Broocks et al. 1990; Beitins et al. 1991; of hormones has not been feasible or affordable, Winters et al. 1996; De Souza et al. 1998a; De Souza investigators have also used definitions that have 2003). In women with LPD, the ovarian system func- included 3–4 or fewer menstrual cycles per year to tions at a level adequate for ovulation, but inade- define oligomenorrhea (Cobb et al. 2003). The ovar- quate to support successful implantation, since ian profile of an oligomenorrheic athlete displays the latter is dependent on adequate exposure of the erratic, unpredictable and presumably inadequate endometrium to P4 (Balash & Vanrell 1987). A estradiol (E2) production as a given follicle struggles reduction in luteal phase P4 production and abbre- to achieve dominance, but certainly may result in an viated luteal phases are the key characteristics of ovulatory cycle in an unpredictable manner. LPD. Reduced P4 production during the luteal phase is also referred to as luteal phase inadequacy or insufficiency to describe the poor quality of the Anovulation endometrium secondary to the low P4 levels. The Anovulation is defined as the absence of ovulation low P4 levels that occur are either low in volume or of an oocyte in the face of inadequate luteinizing low in duration of output, since they occur in the hormone (LH) secretion secondary to inadequate face of normal menstrual cycle lengths of 26–32 estrogen priming in the follicular phase and the days ( Jones 1976; Balash & Vanrell 1987). The other obvious absence of luteinization (Hamilton-Fairly presentation of LPD in athletes is the shortening of & Taylor 2003). Anovulatory cycles are character- the luteal phase, referring to luteal phases of 10 days ized by low E2 and progesterone (P4) levels through- or less (Sherman & Korenman 1974; Jones 1976; De out the cycle; however, much debate in the clinical Souza 2003). Clinically, LPD-associated P4 inade- forum continues regarding the specific criterion quacy causes asynchronous follicular growth in the for confirming anovulation (Malcolm & Cumming subsequent menstrual cycle, compromised oocyte 2003). Because serial ultrasound measurements to maturation and differentiated (out-of-phase) func- energy balance and the menstrual cycle 265

tion of the endometrium. All of the latter factors 10% of infertility and 25% of habitual abortion; how- are associated with low rates of cycle fecundity and ever, the infertility is related to poor P4 production high rates of embryonic loss, i.e. infertility and rather than ovulatory problems (Soules 1988). In spontaneous abortion (Jones 1976; Balash & Vanrell women desiring to become pregnant, decreases in

1987). It is important to understand that women the duration or amount of P4 secreted during the with exercise-associated LPD continue to ovulate, luteal phase has been correlated with the low cycle although some women do so as late as day 20 (De fecundity (Jones & Madigal-Castro 1970; Strott et al. Souza 2003), reflective of the short luteal phases that 1970). Blacker et al. (1997) found that only very small are apparent in association with some presentations differences in luteal phase P4 accounted for un- of LPD. The prevalence of LPD in non-active (i.e. explained infertility in a group of women compared sedentary) women is controversial, but estimates to aged-matched controls. No differences in existed vary from 2% to 5%, and from 3% to 20% in women in the number of preovulatory follicles, the rate of with infertility (Balash & Vanrell 1987; McNeely & follicular growth, or the mean diameter before fol- Soules 1988). LPD occur in athletes at a much licle rupture and timed endometrial biopsy (Blacker greater prevalence of approximately 79% than is et al. 1997). Alternatively, no differences were found reported in non-active women, representing the in salivary P4 obtained during the days prior to most common menstrual cycle abnormality associ- implantation in cycles that resulted in conception ated with exercise (De Souza 1998a). It is important and the levels on the same days that did not result in to note that EAMD occurs outside the competitive conception. The major determinant of cycle fecun- athletic arena as even recreationally active women dity was the robustness of the follicular phase, as the have been recognized to exhibit a high prevalence of probability of conception was significantly related

LPD (De Souza et al. 1998a). For a thorough review to the value of the mid-follicular E2 concentration of LPD in athletes, see a recent review by De Souza (Lipson & Ellison 1996). The previous studies sug- (2003). gest that subtle changes in luteal function and/

or follicular E2 levels may have implications for Exercise-associated menstrual fertility. To date, no studies have been performed in disturbances: clinical considerations female athletes with LPD to assess fertility but clearly the high prevalence of LPD in recreationally active women and athletes warrants more work in Infertility this area. A clinical consequence of exercise-associated amen- orrhea and anovulatory cycles is temporary infer- Effects on the cardiovascular system tility. This represents an obvious problem for women attempting to conceive. With amenorrhea, Due to an observed cardioprotective effect of the absence of menses is noticeable and thus the estrogens, persistently low E2 levels in amenorrheic individual is aware that conception is unlikely. athletes may also have adverse effects on cardio- However, if there is a spontaneous resumption of vascular health. Clinically, one of the earliest signs ovulation, pregnancy is possible because ovula- of cardiovascular disease is a decrease in endothe- tion may not be preceded by a return of menses. lial function, and is evident decades before overt Anovulatory cycles can be accompanied with regu- coronary artery disease is present (Celermajer 1997; lar and consistent menstrual bleeding intervals Luscher & Barton 1997; Schachinger et al. 2000). and thus the individual is not aware that conception Atherosclerotic disease progression and adverse is unlikely. This is in sharp contrast to irregular cardiovascular events have both been shown to cycle intervals indicative of oligomenorrheic cycles be associated with a decrease in endothelial func- where ovulation is unpredictable. Although diffi- tion (Celermajer 1997; Luscher & Barton 1997; cult to diagnose, LPD may be the most common Schachinger et al. 2000). In addition, physical abnormality observed in athletes and contribute to changes to the endothelium and the availability of 266 chapter 20

nitric oxideaan important endothelial-derived relax- these results suggest that chronic hypoestrogenism ing factoraare also decreased in association with may predispose young physically active and ath- atherosclerotic disease (Luscher & Barton 1997). letic women to early or premature cardiovascular dis- It is well-established that estrogen plays a sig- ease. To date, only two studies have examined the nificant role in endothelial-dependent blood flow impact of athletic amenorrhea on endothelial function. via nitric oxide and has effects on cholesterol and Zeni-Hoch et al. (2003) examined brachial artery lipoproteins (Steinberg 1987; Celermajer 1997; flow-mediated dilation (endothelium-dependent) in Luscher & Barton 1997; Mendelsohn & Karas 1999; amenorrheic athletes and compared them to women Schachinger et al. 2000). As such, hypoestrogenism with oligomenorrhea and age-matched controls. has been associated with endothelium-dependent Zeni-Hoch et al. (2003) reported that endothelial dysfunction (Celermajer 1997; Luscher & Barton function was 80% lower in athletes with amenor- 1997; Schachinger et al. 2000). The mechanism of rhea, compared to athletes with either normal action of estrogen on endothelial function is likely menstrual cycles or oligomenorrhea. Disturbingly, through the nitric oxide pathway, which is known the magnitude of impaired endothelium-dependent to be critical in vascular control and during reactive vasodilation in amenorrheic athletes was compar- hyperemia (Armour & Ralston 1998; Arora et al. able to data previously reported in otherwise healthy 1998; Ayres et al. 1998). Genomic and non-genomic post-menopausal women (Blumel et al. 2003) and effects of estrogen have been shown to play a older (60 + 2 years) coronary-artery-disease patients significant role in the up-regulation of endothelial (Celermajer et al. 1992) after a similar flow-mediated nitric oxide synthase (eNOS) and the subsequent stimulus. In contrast, endothelium-independent production and increased half-life of nitric oxide dilation to nitroglycerine was not different amongst (Armour & Ralston 1998; Simoncini et al. 2002). Some the groups. Recent data from our laboratory demon- studies (Armour & Ralston 1998; Simoncini et al. strate that amenorrheic athletes had lower resting 2002) have suggested that estrogen affects vascular (2.2 + 0.1 vs. 4.8 + 0.4; p < 0.001) and peak ischemic endothelial release of nitric oxide through actions (42.8 + 2.1 vs. 52.9 + 2.0; p = 0.004) blood flow res- that enhance the bioavailability of nitric oxide ponses (mL·100 mL–1·min–1) using lower limb strain- (production) by up-regulating the constitutive nitric gauge plethysmography (O’Donnell et al. 2004). The oxide synthase and, conversely, by inhibiting super- amenorrheic athletes also had lower resting supine oxide anion production. heart rate (50.5 + 4.8 vs. 58.8 + 1.9; p = 0.07) and Coincident with declining estrogen levels, a study supine resting systolic blood pressure (90.4 + 5.7 by Celermajer et al. (1994) has shown a reduction vs. 106.8 + 2.0 mmHg; p = 0.004), compared to their in endothelial-dependent vasodilation as soon as eumenorrheic counterparts. In light of the known 3 months after natural menopause. Impaired endo- link between estrogen and vascular function, the thelial function has also been demonstrated 1 week observed attenuated blood flow response is likely after surgical menopause (Ohmichi et al. 2003). Given a consequence of the chronic hypoestrogenism in these rapid reductions in endothelial-dependent the amenorrehic athletes. These findings confirm vasodilation following both surgical and natural altered flow-mediated endothelial-dependent vaso- menopause, it is logical to question the impact of dilation. Lower resting heart rate and systolic blood clinical and subclinical levels of hypoestrogenism in pressure may be indicative of altered autonomic physically active women. regulation, similar to that seen in anorexia nervosa Recent data from our laboratory has identified patients. These findings represent an additional reduced peripheral blood flow, and in another health paradigm associated with the female athlete laboratory impaired endothelial cell dysfunction in triad and suggests that hypoestrogenism in ame- amenorrheic athletes has been reported (Zeni-Hoch norrheic athletes may lead to deleterious cardio- et al. 2003). These cardiovascular findings are most vascular outcomes. Thus, the clinical sequelae of likely the result of chronic hypoestrogenism. Thus, hypoestrogenism in young amenorrheic women energy balance and the menstrual cycle 267

must be extended to include detrimental effects on greater risk of fracture is apparent (Kanis 2002). cardiovascular function that could have serious Similar data are available in women with E2 defici- consequences later in life. Since cardiovascular ency associated with anorexia nervosa. Since expos- health is of concern in this young cohort, more ure to E2, along with genetic and nutritional factors, evaluation is warranted. determines peak bone mass (Matkovic et al. 1994; Kanis 2002), a delay in menses related to exercise training may result in a lower than normal peak Bone health value. Peak bone mass appears to be a very good Chronic hypoestrogenism is a major cause of bone predictor of the rate of post-menopausal bone loss, loss in women, regardless of age (Ott 1990; Matkovic and higher levels will delay the risk of fracture (Ott et al. 1994). It is well-established that the chronic 1990; Matkovic et al. 1994; Kanis 2002). hypoestrogenism coincident with menopause is a It is well-documented that amenorrheic athletes major cause of osteoporosis in women (Matkovic unequivocally suffer from reductions in BMD, et al. 1994; Kanis 2002). Like the anorexic patient, particularly in the lumbar spine (Cann et al. 1984; female athletes who experience amenorrhea also ex- Drinkwater et al. 1984, 1990; Marcus et al. 1985; Keen perience chronic hypoestrogenism (Zipfel et al. 2001). & Drinkwater 1997; Tomten et al. 1998; Gremion Hypoestrogenism associated with secondary et al. 2001; Csermely et al. 2002; Cobb et al. 2003). amenorrhea in athletes contributes greatly to bone Even in oligomenorrheic athletes, very low BMD loss (Cann et al. 1984; Drinkwater et al. 1984, 1990; scores have been repeatedly reported (Tomten et al. Marcus et al. 1985; Matkovic et al. 1994). Since men- 1998; Gremion et al. 2001; Csermely et al. 2002; Cobb strual status is a reflection of E2 production, the et al. 2003). In one study of oligomenorrheic athletes, most severe menstrual perturbations exhibit the lumbar BMD was only 69% of that observed in greatest reductions in E2 production, and are asso- an aged-matched menstruating group of women ciated with the most severe consequences to bone. (Cobb et al. 2003). The low BMD experienced by The failure to achieve peak bone mass or premature amenorrheic athletes is both severe and likely irre- bone loss, and appropriate clinical diagnostic cri- versible, since resumption of menses offers minimal teria as measured by dual energy X-ray absorptio- improvements in BMD, permanently compromis- metry (DXA), i.e. the use of t-scores, should be ing the attainment of peak BMD (Drinkwater et al. encouraged (Ott 1990; Kanis 2002; Khan et al. 2002). 1986; Jonnavithula et al. 1993; Keen & Drinkwater Reduced BMD in amenorrheic athletes has been 1997). Even the administration of oral contracept- repeatedly reported by many investigators (Cann ives and other hormonal replacement strategies fail et al. 1984; Drinkwater et al. 1984, 1990; Marcus et al. to significantly improve BMD in amenorrheic and 1985; Keen & Drinkwater 1997; Tomten et al. 1998; oligomenorrheic athletes (Cumming 1996; Fagan Gremion et al. 2001; Csermely et al. 2002; Cobb et al. 1998). Dugowson et al. (1991) contend that the bone 2003). Most often a t-score of –1.0 to –2.5 is observed loss observed in amenorrheic athletes may be seri- in athletes with amenorrhea, and is referred to as ous enough to result in osteoporotic fractures well osteopenia (Kanis 2002; Khan et al. 2002). The clin- before menopause. ical criterion of osteopenia is associated with a 100% Based on available data, the prevalence of increase in the risk of fracture and is notably of long- osteopenia in amenorrheic athletes is estimated to term concern for the management of bone health in range from 1.4% to 50% in athletes and exercising these individuals. The clinical sequelae of a low women (Drinkwater et al. 1984, 1990; Dugowson peak bone mass or premature bone loss includes an et al. 1991; Rutherford 1993; Young et al. 1994; increased risk of stress fractures and fractures of the Lauder et al. 1999; Pettersson et al. 1999; Khan et al. hip and spine (Carbon et al. 1990; Otis et al. 1997; 2002; Cobb et al. 2003). It is important to recognize Korpelainen 2001). Moreover, if the clinical criterion that the clinical endpoint of osteopenia is not neces- of osteoporosis is achieved in an athlete, an even sarily accompanied by the other two components of 268 chapter 20

the triad (disordered eating and amenorrhea). That an independent nor an additive effect on bone is, the existence of even one of these problems has (Lindsay et al. 1978; Lindsay 1999). Simply put, the been associated with a decrease in BMD, presumably data available do not support any significant effect because both disorders are associated with varying of P4 or LPD on bone. If there is a decrease in degrees of hypoestrogenism (Cobb et al. 2003). Six bone mass in association with EAMD, such as LPD, percent of the oligomenorrheic and/or amenorrheic and anovulation, it is likely a product of decreased runners in the study by Cobb et al. (2003) had BMD exposure to E2. t-scores that were osteoporotic, whereas 48% had De Souza et al. (1997) have reported that regard-

BMD t-scores that were osteopenic. Cobb et al. less of significantly lower P4 levels in exercising (2003) also demonstrated that 26% of the women women with LPD (not combined with women with who were menstruating and had some evidence anovulation), BMD at the lumbar spine and femoral of disordered eating had BMD t-scores that were hip were comparable to that observed in ovulat- osteopenic. Sports medicine professionals should ory sedentary women, provided the E2 status was aggressively move toward the understanding that adequately maintained during the cycle. In the the existence of even one subclinical entity asso- Women’s Reproductive Health Study (Waller et al. ciated with the female athlete triad may have a 1998), sedentary women with LPD failed to present negative clinical impact on the long-term health of with reduced BMD as well. One concern is that in these athletes and physically active women. the study by De Souza et al. (1997), significantly

Although physical activity has generally been lower E2 levels were found in the follicular phase in considered to be protective of BMD, some research the LPD runners. Even in the ovulatory exercising suggests that a proportion of physically active women in that study, E2 levels were lower during women, even in the face of apparently normal men- the early follicular phase (days 2–5). Winters et al. strual cycles, have reduced BMD (Prior et al. 1990). (1996) have also reported reduced E2 levels in the Data reported by Prior et al. (1990) suggested that early follicular phase in trained runners. Import- LPD and anovulatory cycles resulted in progress- antly, they also found that the lumbar spine BMD ively more spinal bone loss over a 1-year period in was lower in their runners with the reductions in moderate- and long-distance runners, and they follicular phase E2, compared to the active controls. attributed this loss to lower P4 levels. Methodolo- Although this finding was not statistically signific- gical problems, however, are associated with the ant at p = 0.074, it was likely a product of inadequate Prior et al. (Prior et al. 1990; De Souza et al. 1997, statistical power, and the potential physiological 2003) data set that limit the usefulness of the infor- importance of the finding should not be dismissed. mation. These limitations include: (i) the inferior A limitation of the aforementioned studies is that methods used to assess LPD, which included the they are cross-sectional and limited by small sample use of basal body temperatures and the pooling of sizes. In the Michigan Bone Health Study, Sowers single blood samples from the follicular and luteal et al. (1998a, 1998b) found that 31 sedentary women phases as indicators of ovulatory status; and (ii) the who were slightly older than the subjects in the pre- confusing effect of combining E2-deficient anovu- vious studies (mean age of 37 years) and who had latory cycles with LPD cycles, since it is likely that reduced bone mass, also had the lowest E2 levels anovulatory cycles have very different E2 produc- during the luteal phase of two comprehensively tion patterns compared to ovulatory LPD cycles, monitored menstrual cycles. and thus have the potential to impact bone health. Thus, it may be necessary to follow exercising Since then, many researchers have been unable to women for several years while simultaneously reproduce the results of Prior et al. (1990) (Lindsay documenting menstrual status and E2 exposure to et al. 1978; Winters et al. 1996; De Souza et al. 1997; definitively answer the important question of long- Waller et al. 1998; Lindsay 1999). Moreover, in post- term effects of LPD and anovulation on BMD. It menopausal women, progestins, whether prescribed is likely that women with anovulatory cycles may alone or in combination with E2, provides neither certainly experience more serious consequences energy balance and the menstrual cycle 269

due to the greater severity of estrogen depletion physical activity) factors. Another factor pertinent association with this menstrual anomaly compared to the assessment of racial differences in EAMD is to ovulatory LPD cycles. that racial and ethnic differences in body image and dieting behavior exist (White et al. 2003). African- Energetics and exercise-associated American girls may be at a lessened risk for devel- menstrual disturbances: practical oping eating disorders because they do not display considerations the same degree of negative beliefs about body size and shape when compared to Caucasian girls (White et al. 2003). Demographics

With respect to the demographics of EAMD, the Low energy availability in practical terms highest prevalence of amenorrhea in athletes is found in sports that emphasize a low body weight caloric restriction such as figure skating, ballet, long-distance run- ning and gymnastics, but more recent studies have The evidence for modulation of reproductive func- documented menstrual abnormalities in a wide tion by energy availability in exercising women variety of sports (Erdelyi 1962; Feicht et al. 1978; comes from short- and long-term prospective stud- Dale et al. 1979; Schwartz et al. 1981; Sanborn et al. ies in humans and animals and from observational 1982; Loucks & Horvath 1985). Amenorrhea associ- and cross-sectional studies (Bullen et al. 1985; ated with exercise is a typically a diagnosis of exclu- Loucks 1989; Williams et al. 1995; Laughlin & Yen sion; other causes of reproductive disturbances, i.e. 1996; Wade et al. 1996; De Souza et al. 1998a; Loucks pregnancy, androgen excess, gonadal dysgenesis, et al. 1998; Loucks & Thuma 2003). The suppression uterine or premature ovarian failure, or pituitary of reproductive function during conditions of low dysfunction, must be ruled out (Warren 1996). In fuel availability is well-recognized in the animal physically active girls who have not experienced literature, and is thought to be an adaptive response menarche, primary amenorrhea may ensue despite to conserve fuel for more vital bodily processes, these individuals exhibiting normal growth (Warren such as cellular maintenance and locomotion, and 1996). While no studies have examined the pre- prevent the energetically costly investment of gesta- valence of EAMD with increasing age, increased tion and lactation (Wade et al. 1996). A key dif- gynecological maturity may afford some protection ference is that a state of low energy availability in from disruption of menstrual cycles (Rogol et al. today’s active women may represent a choice to 1992). The majority of studies examining EAMD consciously restrict food intake to a level below that have been performed in adolescents or women in required to match the calories expended through their early 20s. No studies have examined racial or exercise. Numerous studies have linked EAMD, ethnic differences in the prevalence or endocrine particularly amenorrhea, with significantly higher presentation of EAMD. Characterizing effects of scores on psychometric inventories of eating atti- race is important because studies in premenopausal tudes, indicating a conscious restriction of calor- and perimenopasual women show differences in ies associated with a drive for thinness and low some reproductive hormone concentrations when body fat (Brooks-Gunn et al. 1987; Myerson et al. Caucasian and African-American women are com- 1991; Wilmore et al. 1992; Laughlin & Yen 1996; pared (Manson et al. 2001; Reutman et al. 2002), and Lebenstedt et al. 1999; Warren et al. 1999). Dietary fat a high degree of variability when multiple races are is often restricted as well (Laughlin & Yen 1996; studied (Randolph et al. 2003). Additionally, because Loosli & Ruud 1998). When less severe forms of reproductive hormones change in varying directions menstrual disturbances are examined, i.e. anovula- with age and increasing body mass index (Randolph tion and LPD, scores on these eating inventories are et al. 2003), studies examining racial influences must still higher than those in subjects who menstruate adjust for these and other (smoking, alcohol use and normally (Lebenstedt et al. 1999). 270 chapter 20

physiological cues linking low energy availability to a suppression of menstrual cyclicity in female athletes (Furuta et al. While significant associations between weight and 2003). This is not surprising, as other gut peptides, diet concerns and restrictive eating have been such as CCK, galanin and peptide YY (PYY) (3–36), related to menstrual disturbances in female athletes, have been discovered to exist in the brain and have it should not be assumed that every athlete that has neuroendocrine functions (Lopez & Negro-Vilar EAMD is consciously restricting her food intake. 1990; Challis et al. 2003; Moran & Kinzig 2004). We Many studies describing the endocrine aspects recently reported that weight loss resulting from a of EAMD included only subjects that had no history diet and exercise intervention in young women of disordered eating (Loucks et al. 1989, 1992). If an leads to an increase in resting levels of ghrelin, and athlete is not consciously restricting her food but the change in ghrelin is significantly correlated with is still energy restricted, it is logical to question the change in body weight during the intervention whether the physiological cues to eat are not well- (Leidy et al. 2004). In another study we compared matched to energy needs. Perhaps strenuous exer- concentrations of ghrelin between women grouped cise in some individuals alters the levels of, and/or based on their exercise and menstrual status. Fast- sensitivity to, physiological cues for hunger or ing ghrelin levels were assessed in the following satiety in female athletes, or whether it influences groups of women: sedentary and ovulating norm- cognitive cues for, or the response to, food intake. ally; athletes who are ovulating normally; athletes In an effort to address a plausible mechanism for who are anovulatory or exhibit LPD; and athletes insufficient dietary intake in female athletes, one who are amenorrheic. Fasting ghrelin concentra- report by Hirschberg et al. (1994) found that the tions were found to be significantly elevated in the response of gastrointestinal hormone cholecys- amenorrheic athletes compared to all other groups tokinin (CCK), a satiety peptide produced in the (De Souza et al. 2004). Because the amenorrheic ath- gut, was reduced in female long-distance runners letes in the study exhibited other endocrine signs compared to non-athletes when a 500 kcal (2093 kJ) of energy deficiencyai.e. low levels of triiodothyro- meal was provided. Increased hunger was also nine, insulin, leptin and other alterations in meta- reported by the athletes (Hirschberg et al. 1994). In bolics hormones (Fig. 20.3)abut had been weight response to acute exercise, Hilton and Loucks (2000) stable, it appears that physiological cues to stimu- showed that leptin, a signal for satiety, decreases late food intake are responding appropriately to a with exercise only when energy availability is also state of low energy availability, whereas other cues, decreased. When the calories expended as exercise perhaps cognitive in nature, are preventing a return are compensated for with increased calorie intake, to energy homeostasis. This would agree with the leptin levels did not change significantly. In amen- findings of elevated concentrations of ghrelin in orrheic athletes, leptin levels are lower than would anorexics (Otto et al. 2001). With respect to how low be expected for their level of body fat (Laughlin energy availability develops in an active woman, & Yen 1997). Recently, the current authors have much still needs to be learned in order to identify investigated a link between EAMD and ghrelin, a the predisposing factors that predict which athletes recently discovered growth hormone secretagogue will consciously restrict their calories in addition to released from the stomach that stimulates hunger their training, and which athletes may simply not be and food intake in humans (Date et al. 2000). Ghrelin able to, or physiologically stimulated to, increase has been directly linked to the regulation of energy their food intake to meet their energy needs. homeostasis, and is one of the most powerful known orexigens (Nakazato 2001; Wren et al. 2001). prediction of exercise-associated Interestingly, one report thus far has linked in- menstrual disturbances based on tracerebroventricular ghrelin administration to a energy intake and expenditure suppression of LH pulsatility in ovariectomized rats, therefore establishing a plausible mechanism Despite advances in our understanding of the energy balance and the menstrual cycle 271

? Intermittent/transient Eumetabolic state Hypometabolic state hypometabolic state

Luteal phase deficient Ovulatory menstrual cycles Amenorrheic cycles menstrual cycles

Metabolic hormones and substrates relative to sedentary women

– Ghrelin – Ghrelin Ghrelin

Total T3 Total T3 Total T3

Leptin Leptin Leptin

– Insulin Insulin Insulin

– Growth hormone Growth hormone Growth hormone

– IGF-I/IGFBP-1 – IGF-I/IGFBP-1 IGF-I/IGFBP-1

Cortisol Cortisol Cortisol

– Glucose – Glucose Glucose

Reproductive hormones relative to sedentary women

LH pulsatility LH pulsatility LH pulsatility

– FSH FSH FSH

Estradiol Estradiol Estradiol

Progesterone Progesterone Progesterone

Fig. 20.3 The metabolic and reproductive hormone perturbations that have been identified to date and associated with exercise training and menstrual status, including eumenorrheic ovulatory cycles, luteal phase defect (LPD) cycles and amenorrhea. All values are depicted by arrows signifying the magnitude of the alteration reported. The proposed relationship to menstrual status is also shown. The repeated transitions from ovulatory cycles and LPD cycles are shown. Although not documented in the literature, it is likely that during recovery from amenorrhea, an individual is likely to experience LPD cycles. FSH, follicle-stimulating hormone; LH, luteinizing hormone; IGF-I, insulin-like growth factor I;

IGFBP-1, insulin-like growth factor binding protein 1; T3 triiodothyronine. (Modified and reprinted with permission from De Souza 2003.) 272 chapter 20

causal role that low energy availability plays in the wide range exists in the volume of exercise and or etiology of EAMD, we have not yet developed daily calorie intake that is associated with EAMD, practical guidelines regarding the optimal com- and prevents either from being used as a bench- bination(s) of dietary intake and exercise energy mark. Perhaps the most progress in actually expenditure to maintain ovulation in active women. quantifying changes in energy intake and energy Although it is believed that female athletes may not expenditure associated with changes in the repro- have to reduce their training to avoid EAMD, ductive axis in exercising women has been made by avoiding disruptions in menstrual function would Loucks et al. (Loucks & Thuma 2003). A threshold of conceivably necessitate the quantification of food energy availability between 20 and 30 kcal·kg–1 (84 intake and energy expenditure in order to determine and 126 kJ·kg−1) lean body mass was identified such how much food is required to match the level of that no effect on the reproductive axis was observed calories expended. Because efforts to optimize per- until this threshold was achieved. When energy formance may predispose an athlete to take in the availability of exercising subjects was held between minimum amount of calories required to prevent 20 and 30 kcal·kg–1 (84 and 126 kJ·kg−1) lean body fatigue but maintain ovulation, it would be helpful mass, LH pulsatility, a proxy indicator of the GnRH to know ‘how low you could go’ without comprom- pulse generator, was decreased. Importantly, because ising reproductive function. Many (Drinkwater et al. the initial energy availability of the subjects was 1984; Marcus 1985; Kaiserauer et al. 1989) but not 45 kcal·kg–1 (188 kJ·kg−1) lean body mass, these data all (Loucks 1989; Wilmore et al. 1992) studies of self- also show that a decrease of energy availability by reported calorie intake in amenorrheic athletes 33% had no effect on LH pulsatility, suggesting that report that calorie intake is lower than expected the female reproductive axis can withstand substan- for the level of estimated energy expenditure, and tial reductions in energy availability before the lower when compared to eumenorrheic athletes reproductive axis responds. Additionally, the data with similar training regimens. It is difficult to suggest that no change in LH pulsatility occurred translate the findings of these studies into advice for even though exercise calories were as high as practitioners, because there is a tendency for most 840 kcal (3516 kJ), which might represent an 8-mile subjects to under-report food intake. Regarding the (13-km) run. However, whether a similar threshold suppression of reproductive function associated to that identified by Loucks’ short-term studies with calorie restriction, neither the magnitude of holds true for the effects of prolonged training on the reduction in the calories consumed, nor the ovulation and menstrual cyclicity remains to be absolute amount of calories consumed appears to demonstrated. Further, whether the concept of a be what is sensed by gonadotropin-releasing hor- ‘threshold’ holds true for the development of less mone (GnRH) neurons. Support for this comes from severe menstrual disturbances, such as LPD, is Loucks’ studies where short-term energy deficits unknown. Ongoing studies aimed at inducing an are created by varying combinations of food intake quantifiable energy deficit with exercise combined and exercise and the resulting suppression of LH with calorie restriction and perturbing the men- pulsatility is dependent on the magnitude of the strual cycle are underway and may, in the future, energy deficit, regardless of if it is created through provide practical guidelines for athletes and active exercise or diet, or a combination (Loucks & Thuma women. 2003). Additionally, in studies in exercising mon- keys, food intake remained constant as amenorrhea prediction of exercise-associated developed due solely to increases in the energy menstrual disturbances based on cost of daily exercise (Williams et al. 2001a). This body weight and body fat dependence of EAMD on energy availability per se, and not a particular level of calorie intake or amount Predicting EAMD would be much easier if changes of exercise, presents a problem with regard to the in reproductive function occurred at a predictable detection and prevention of EAMD. It means that a level of body weight and or body fat, as originally energy balance and the menstrual cycle 273

proposed by Frisch and Revelle (1971). Bullen et al. the magnitude of change in energy availability (1985), in a prospective exercise training study over associated with the induction of EAMD may be two menstrual cycles, showed that the incidence more difficult to achieve than quantification of the of EAMD increases with weight loss that aver- magnitude of change in energy availability required aged –0.45 kcal·wk–1 (−1.88 kJ·wk−1) and –4.0 + 0.3 kg to restore menstrual cyclicity. One reason is that over the course of the study. Whether this rate and researchers have been more successful at document- magnitude of weight loss represent potential guide- ing the reversal of amenorrhea (Drinkwater et al. lines for women just beginning an exercise program 1986; Dueck et al. 1996; Kopp-Woodroffe et al. 1999; is questionable, since no individual correlations Perkins et al. 2001; Zeni-Hoch et al. 2003) in exercis- were reported between the magnitude of weight ing women than inducing amenorrhea with a diet loss and the severity of menstrual disturbances, and and/or exercise intervention (Williams et al. 2001a). menstrual cyclicity was disrupted in the group that In exercising monkeys who had developed amenor- maintained body weight. Other studies have shown rhea, Williams et al. (2001b) showed that, unlike the that EAMD exist in women who represent a range induction of amenorrhea, the reversal (as noted by of body weights and levels of percentage body fat ovulation) exhibited a linear dose–response relation- (Sanborn et al. 1987), and the induction of amenor- ship with increased energy availability, such that rhea can occur with no change in body weight the two monkeys that consumed 163% and 181% of (Williams et al. 2001a). Further, changes in repro- their baseline intake recovered in 16 and 12 days, ductive hormone secretion resulting from short- respectively, and two monkeys that consumed term changes in nutritional intake occur without 138% and 141% recovered in 57 and 50 days, respect- significant weight loss or gain (Bronson 1986; ively. The time in days required for recovery was Cameron & Nosbisch 1991). Lastly, the effect of inversely related to the amount of extra calories weight loss or the loss of body fat on reproductive consumed by each monkey (r = –0.97; P < 0.02). In function may or may not be dependent on initial contrast to the lack of change in body weight during body fat stores (Alvero et al. 1998), thereby imposing the induction of EAMD, body weight increased an additional interaction to consider. Therefore, significantly with the restoration of ovulation and advising female athletes not to lose a certain amount subsequent menses. There was a significant correla- of weight to prevent EAMD, or to gain a certain tion between body weight during the amenorrheic amount of weight to restore menstrual cyclicity, is period and the time to recovery, i.e. the monkeys not reliable. Guidelines regarding the prevention that weighed less recovered more quickly than the of EAMD based on body weight are further com- heavier monkeys, but there was no correlation plicated because of the large individual variability between the magnitude of change in weight and in the amount of weight lost in response to specific the rapidity of the restoration. Overall, body weight energy deficits (Ravussin et al. 2001). increased 3–11%. Notably, the time to recovery for these monkeys was much shorter than the time it reversing exercise-associated took to become amenorrheic. menstrual disturbances by increasing Although the reversal of exercise-associated ame- energy availability norrhea has been documented by several studies in humans, there have only been two small studies To date, the only non-medical advice provided for published on the reversal of amenorrhea attempted prevention and treatment of amenorrhea is that by a supervised intervention that modified diet and athletes reduce their training and increase their exercise (Dueck et al. 1996; Kopp-Woodroffe et al. caloric intake. No specific dietary guidelines are 1999). Dueck et al. (1996) were the first to publish the available, and the rationale for reducing training results of a reversal of amenorrhea intervention in may be an overly conservative approach, since an athlete consisting of a 15-week diet and exercise exercise per se does not appear to play a role in intervention. The intervention consisted of increas- exercise-associated amenorrhea. Quantification of ing daily energy intake by 360 kcal (1507 kJ) and 274 chapter 20

decreasing exercise training by 1 day per week. The to reverse amenorrhea with an ‘energy intervention’ increase in energy intake was accomplished with a is that the alteration in diet, exercise, or both, be liquid supplement. At the end of the 15-week inter- tailored to the individual’s own preference. For vention, the subject remained amenorrheic, despite some individuals, a reduction in training may not be a 2.7 kg increase in body weight and increased body as well-tolerated psychologically as an increase in fat from 8% to 14%, likely a product of improved calories, but for others a combination or only a and positive energy balance from –155 kcal·day–1 change in training may be preferred. Careful mon- (−648 kJ·day−1) at baseline to +683 kcal·day–1 (+2855 itoring of both exercise and diet is important since kJ·day−1) accomplished by the end of the 15-week there may be a tendency for an athlete to unknow- intervention. The intervention also resulted in a 148% ingly compensate for an increase in food intake increase in serum LH levels. Interestingly, menstru- with an increase in training frequency or duration ation did resume in this amenorrheic athlete after (Kopp-Woodroffe et al. 1999). In the latter study, an additional self-initiated 3 months of additional two out of four subjects could not completely compliance to the protocol. A second study (Kopp- adhere to the reduction in training called for by the Woodroffe et al. 1999) attempted the reversal of intervention. amenorrhea with a 20-week diet and intervention in four amenorrheic athletes. The women were again − Summary and conclusions provided with a 360 kcal·day–1 (1507 kJ·day 1) liquid supplement and asked to restrict exercise training for EAMD can be associated with significant clinical 1 day per week. Three of the four subjects resumed outcomes, and thus continued efforts aimed at normal menstrual function during or shortly follow- identification and prevention is necessary. Bone ing the intervention period, again likely a result of the loss and endothelial dysfunction has been linked improved energy balance of +164–292 kcal·day–1 to hypoestrogenism secondary to amenorrhea in (+686–1222 kJ·day−1). The fourth subject withdrew female athletes. The extent to which less severe from the intervention to begin hormone therapy for EAMD are associated with changes in bone and poor bone density. As reported in the study results, cardiovascular function remain unclear, but is a time to menstruation subsequent to ovulation in fertile area for future research. The cause of EAMD these three subjects ranged from 13 to 24 weeks. is not completely understood but a causal relation- Using data reported for calorie intake and expend- ship has been established between low energy iture, the estimated changes in energy balance availability and the induction of menstrual dis- ranged from +8% to +16%. turbances. Recovery of normal menstrual function The data from the previous studies provide the appears to depend directly on the magnitude of most useful information to date on how to achieve change in energy availability, accomplished through the restoration of menstrual cyclicity in amenorrheic an increase in food intake that may or may not be athletes. An important consideration in attempting combined with decreased exercise training.

References

Alvero, R., Kimzey, L., Sebring, N. Arora, S., Veves, A., Caballaro, A.E., Williamson, H.O. (1981) Female runners et al. (1998) Effects of fasting on Smakowski, P.A. & LoGerfo, F.W. (1998) and secondary amenorrhea: correlation neuroendocrine function and follicle Estrogen improves endothelial function. with age, parity, mileage, and plasma development in lean women. Journal of Journal of Vascular Surgery 27(6), hormonal and sex-hormone binding Clinical Endocrinology and Metabolism 83, 1141–1146. globulin concentrations. Fertility and 76–80. Ayres, S., Baer, J., Subbiah, M.T. & Ravi, H. Sterility 36, 183–187. Armour, K.E. & Ralston, S.H. (1998) (1998). Exercised-induced increase in Balash, J. & Vanrell, J.A. (1987) Corpus Estrogen upregulates endothelial lipid peroxidation parameters in luteum insufficiency and infertility: constitutive nitric oxide synthase amenorrheic female athletes. Fertility and a matter of controversy. Human expression in human osteoblast-like Sterility 69(1), 73–77. Reproduction 2, 557–567. cells. Endocrinology 139, 799–802. Baker, E.R., Mathur, R.S., Kirk, R.F. & Beitins, I.Z., McArthur, J.W., Turnbull, energy balance and the menstrual cycle 275

B.A., Skrinar, G.S. & Bullen, B.A. (1991) and adults at risk of atherosclerosis. Endocrinology and Metabolism 83, Exercise induces two types of human Lancet 340(8828), 1111–1115. 4220–4232. luteal dysfunction: conformation by Celermajer, D.S., Sorensen, K.E., De Souza, M.J., Miller B.E., Luciano, A.A. urinary free progesterone. Journal of Spiegelhalter, D.J. et al. (1994) Aging is et al. (1998b) Letter to the editor. Clinical Endocrinology and Metabolism 72, associated with endothelial dysfunction Ovulation and spine bone mineral 1350–1358. in healthy men years before the age- densityaauthor’s response. Journal of Blacker, C.M., Ginsburg, K.A., Leach, R.E., related decline in women. Journal of the Clinical Endocrinology and Metabolism Randolph, J. & Moghissi, K.S. (1997) American College of Cardiology 24(2), 83(10), 3757–3758. Unexplained infertility: evaluation of 471–476. De Souza, M.J., vanHeest, J.L., Demers, L. the luteal phase; results of the National Challis, B.G., Pinnock, S.B., Coll, A.P. et al. & Lasley, B.L. (2003) Luteal phase Center for Infertility Research at (2003) Acute effects of PYY3–36 on food deficiency in recreational runners: Michigan. Fertility and Sterility 67(3), intake and hypothalamic neuropeptide evidence for a hypometabolic state. 437–442. expression in the mouse. Biochemical and Journal of Clinical Endocrinology and Blumel, J.E., Castelo-Branco, C., Leal, T. Biophysical Research Communications Metabolism 88(1), 337–346. et al. (2003) Effects of transdermal 311(4), 915–919. De Souza, M.J., McConnell, H.J., estrogens on endothelial function Cheung, C.C., Thornton, J.E., Kuiper, J.L. O’Donnell, E., Lasley, B.L. & Williams, in postmenopausal women with et al. (1997) Leptin is a metabolic gate for N.I. (2004) Fasting ghrelin levels in coronary disease. Climacteric 6(1), the onset of puberty in the female rat. physically active women: relationship 38–44. Endocrinology 138, 855–858. with menstrual disturbances and Bronson, F.H. (1986) Food-restricted, Cobb, K.L., Bachrach, L.K., Greendale, G. metabolic status. Journal of Clinical preubertal, female rats: rapid recovery et al. (2003) Disordered eating, menstrual Endocrinology and Metabolism 10(5), of luteinizing hormone pulsing with irregularity, and bone mineral density in 433–448. excess food, and full of pubetal female runners. Medicine and Science in Drew, F.L. (1961) The epidemiology of development with gonadotropin- Sports and Exercise 35(5), 711–719. secondary amenorrhea. Journal of releasing hormone. Endocrinology 118, Csermely, T., Halvax, L., Schmidt, E. Chronic Disease 14, 396–401. 2483–2487. et al. (2002) Occurrence of osteopenia Drinkwater, B.L, Nilson, K., Chesnut, C.H. Broocks, A., Pirke, K.M., Schweiger, U. among adolescent girls with et al. (1984) Bone mineral content of et al. (1990) Cyclic ovarian function in oligo/amenorrhea. Gynecology and amenorrheic and eumenorrheic athletes. recreational athletes. Journal of Applied Endocrinology 2, 99–105. New England Journal of Medicine 311, Physiology 68(5), 2083–2086. Cumming, D.C. (1996) Exercise-associated 277–281. Brooks-Gunn, J., Warren, M.P. & amenorrhea, low bone density, and Drinkwater, B.L., Nilson, K., Ott, S. & Hamilton, L.H. (1987) The relation of estrogen replacement therapy. Archives Chestnut, C.H., 3rd. (1986) Bone mineral eating problems and amenorrhea in of Internal Medicine 156, 2193–2195. density after resumption of menses in ballet dancers. Medicine and Science in Dale, E., Gerlack, D.H. & Wilhite, A.L. amenorrheic athletes. Journal of the Sports and Exercise 19(1), 41–44. (1979) Menstrual dysfunction in distance American Medical Association 256, Bullen, B.A., Skrinar, G.S., Beitins, I.Z. et al. runners. Obstetrics and Gynecology 54, 380–382. (1985) Induction of menstrual disorders 47–53. Drinkwater, B.L., Bruemner, B. & by strenuous exercise in untrained Date, Y., Kojima, M., Hosoda, H. Chestnut, C.H., 3rd. (1990) Menstrual women. New England Journal of Medicine et al. (2000) Ghrelin, a novel growth history as a determinant of current bone 312, 1349–1353. hormone-releasing acylated peptide, is density in young athletes. Journal of the Cameron, J.L. & Nosbisch, C. (1991) synthesized in a distinct endocrine cell American Medical Association 263(4), Suppression of luteinizing hormone and type in the gastrointestinal tracts of 545–548. testosterone secretion during short term rats and humans. Endocrinology 141, Dueck, C.A., Matt, K.S., Manore, M.M. & food restriction in the adult male rhesus 4255–4261. Skinner, J.S. (1996) Treatment of athletic monkey (Macaca mulatta). Endocrinology De Souza, M.J. (2003) Menstrual amenorrhea with a diet and training 128, 1532–1540. disturbances in athletes: a focus on luteal intervention program. International Cann, C.E., Mortin, M.C., Genant, H.K. phase defects. Medicine and Science in Journal of Sports Nutrition 6, 24–40. & Jaffe, R.B. (1984) Decreased spine Sports and Exercise 35(9), 1553–1563. Dugowson, C.E., Drinkwater, B.L. & Clark, mineral content in amenorrheic women. De Souza, M.J., Miller, B.E., Sequenzia, J.M. (1991) Nontraumatic femur fracture Journal of the American Medical Association L.C. et al. (1997) Bone health is not in an oligomenorrheic athlete. Medicine 251, 626–629. affected by luteal phase abnormalities and Science in Sports and Exercise 23, Carbon, R., Sambrook, P.N., Deakin, V. and decreased ovarian progesterone 1323–1325. et al. (1990) Bone density of elite female production in female runners. Journal of Ellison, P.T. & Lager, C. (1986) Moderate athletes with stress fractures. Medical Clinical Endocrinology and Metabolism 82, recreational running is associated with Journal of Australia 153, 373–376. 2867–2876. lowered salivary progesterone profiles Celermajer, D.S. Endothelial dysfunction: De Souza, M.J., Miller, B.E., Loucks, A.B. in women. American Journal of Obstetrics does it matter? Is it reversible? (1997) et al. (1998a) High frequency of luteal and Gynecology 154, 1000–1003. Journal of the American College of phase deficiency and anovulation in Erdelyi, G.J. (1962) Gynecological survey Cardiology 30(2), 325–333. recreational women runners: blunted of female athletes. Journal of Sport Celermajer, D.S., Sorensen, K.E., Gooch, elevation in follicle-stimulating Medicine and Physical Fitness 2, V.M. et al. (1992) Non-invasive detection hormone observed during luteal- 174–197. of endothelial dysfunction in children follicular transition. Journal of Clinical Fagan, K.M. (1998) Pharmacologic 276 chapter 20

management of athletic amenorrhea. Keen, A.D. & Drinkwater, B.L. (1997) diets in female athletes: a red flag. Clinics in Sports Medicine 17, 327–331. Irreversible bone loss in former Physician and Sports Medicine 26(11), Feicht, C.B., Johnson, T.S., Martin, B.J. amenorrheic athletes. Osteoporosis 45–48. et al. (1978) Secondary amenorrhea International 7, 311–315. Lopez, F.J. & Negro-Vilar, A. (1990) in athletes. Lancet 2, 1145–1146. Khan, K.M, Liu-Ambrose, T., Sran, M.M. Galanin stimulates luteinizing hormone- Frisch, R.E. & Revelle, R. (1971) Height and et al. (2002) New criteria for the female releasing hormone secretion from weight at menarche and a hypothesis athlete triad syndrome? British Journal arcuate nucleus-median eminence of menarche. Archives of Disease in of Sports Medicine 36(1), 10–13. fragments in vitro: involvement of an Childhood 46, 695–701. Kopp-Woodroffe, S.A., Manore, M.M., α-adrenergic mechanism. Endocrinology Frisch, R.E., Wyshak, G. & Vincent, L. Dueck, C.A., Skinner, J.S. & Matt, K.S. 127(5), 2431–2436. (1980) Delayed menarche and (1999) Energy and nutrient status of Loucks, A.B. (1989) Alterations in the amenorrhea in ballet dancers. New amenorrheic athletes participating in a hypothalamic–pituitary–ovarian and England Journal of Medicine 303(1), 17–19. diet and exercise training intervention the hypothalamic–pituitary–adrenal Furuta, M., Funabashi, T., Kimura, F. program. International Journal of Sports axes in athletic women. Journal of Clinical (2003) Intro cerebroventricular Nutrition 9, 70–88. Endocrinology and Metabolism 68(2), administration of ghrelin rapidly Korpelainen, R., Orava, S., Karpakka, J., 402–411. suppresses pulsatile luteinizing Siira, P.& Hulkko, A. (2001) Risk factors Loucks, A.B. & Horvath, S.M. (1985) hormone secretion in ovariectomized for recurrent stress fractures in athletes. Athletic amenorrhea: a review. Medicine rats. Biochemical and Biophysical Research American Journal of Sports Medicine 29(3), and Science in Sports and Exercise 17(1), Communications 288(4), 780–785. 304–310. 56–72. Gremion, G., Rizzoli, R., Slosman, D., Lauder, T.O., Williams, M.V., Campbell, Loucks, A.B. & Thuma, J.R. (2003) Theintz, G. & Bonjour, J.P. (2001) Oligo- C.S. et al. (1999) The female athlete triad: Luteinizing hormone pulsatility is amenorrheic long-distance runners may prevalence in military women. Military disrupted at a threshold of energy lose more bone in spine than femur. Medicine 164, 630–635. availability in regularly menstruating Medicine and Science in Sports and Exercise Laughlin, G.A. & Yen, S.S.C. (1996) women. Journal of Clinical Endocrinology 33(1), 15–21. Nutritional and endocrine-metabolic and Metabolism 88, 297–311. Hamilton-Fairly, D. & Taylor, A. (2003) aberrations in amenorrheic athletes. Loucks, A.B., Laughlin, G.A., Mortola, J.F. Anovulation. BMJ (Clinical Research Ed.) Journal of Clinical Endocrinology and et al. (1992) Hypothalamic–pituitary– 327(7414), 546–549. Metabolism 81, 4301–4309. thyroidal function in eumenorrheic and Hilton, L.K. & Loucks, A.B. (2000) Low Laughlin, G.A. & Yen, S.S.C. (1997) amenorrheic athletes. Journal of Clinical energy availability, not the stress of Hypoleptinemia in women athletes: Endocrinology and Metabolism 75, exercise, suppresses the diurnal rhythm absence of a diurnal rhythm with 514–518. of leptin in healthy young women. amenorrhea. Journal of Clinical Loucks, A.B., Verdun, M. & Heath, E.M. American Journal of Physiology. Endocrinology and Metabolism 82, (1998) Low energy availability, not stress Endocrinology and Metabolism 278(1), 318–321. of exercise, alters LH pulsatility in E43–E49. Lebenstedt, M., Platte, P. & Pirke, K. exercising women. Journal of Applied Hirschberg, A.L., Lindholm, C., Carlstrom, (1999) Reduced resting metabolic rate Physiology 84, 37–46. K. & Von Schoultz, B. (1994) Reduced in athletes with menstrual disorders. Luscher, T.F. & Barton, M. (1997) Biology serum cholecystokinin response to food Medicine and Science in Sports and Exercise of the endothelium. Clinical Cardiology intake in female athletes. Metabolism 31(9), 1250–1256. 20(11) (suppl. 2), II-3–II-10. 43(2), 217–222. Leidy, H.J., Gardner, J.K., Frye, B.R. et al. McNeely, M.J. & Soules, M.R. (1988) The Jones, G.S. (1976) The luteal phase defect. (2004) Circulating ghrelin is sensitive to diagnosis of luteal phase deficiency: a Fertility and Sterility 27, 351–356. changes in body weight during a diet critical review. Fertility and Sterility 50, Jones, G.S. & Madigal-Castro, V. (1970) and exercise program in normal weight 1–15. Hormonal findings in association with young women. Journal of Clinical Malcolm, C.E. & Cumming, D.C. (2003) abnormal corpus luteum function in the Endocrinology and Metabolism 89(6), Does anovulation exist in eumenorrheic human: the lueal phase defect. Fertility 2659–2664. women? Obstetrics and Gynecology and Sterility 21, 1–13. Lindsay, R. (1999) The lack of effect of 102(2), 317–318. Jonnavithula, S., Warren, M.P., Fox, R.P. progesterone on bone. Journal of Sports Malina, R.M. (1994) Physical growth and & Lazaro, M. (1993) Bone density is Medicine 44(2), 215–220. biological maturation of young athletes. compromised in amenorrheic women Lindsay, R., Hart, D.M., Purdie, D. et al. Exercise and Sport Science Reviews 22, despite return of menses: a 2-year study. (1978) Comparative effects of oestrogen 389–433. Obstetrics and Gynecology 81(5), and a progesterone on bone loss in Manson, J.M., Sammel, M.D., Freeman, 669–674. postmenopausal women. Clinical E.W. & Grisso, J.A. (2001) Racial Kaiserauer, S., Snyder, A.C., Sleeper, M. Sciences and Molecular Medicine 54, differences in sex hormone levels in & Zierath, J. (1989) Nutritional, 193–195. women approaching the transition to physiological, and menstrual status of Lipson, S.F. & Ellison, P.T. (1996) menopause. Fertility and Sterility 75(2), distance runners. Medicine and Science in Comparison of salivary steroid profiles 297–304. Sports and Exercise 21(2), 120–125. in naturally occurring conception and Marcus, R., Cann, C., Madvig, P. et al. Kanis, J.A. (2002) Diagnosis of non-conception cycles. Human (1985) Menstrual function and bone osteoporosis and assessment of fracture Reproduction 11(10), 2090–2096. mass in elite women distance runners. risk. Lancet 359, 1929–1936. Loosli, A.R. & Ruud, J.S. (1998) Meatless Annals of Internal Medicine 102, 158–163. energy balance and the menstrual cycle 277

Matkovic, V., Jelic, T., Wardlow, G.M. et al. and patterns of neuro-endocrine & Gatz, M. (1979) The relationship (1994) Timing of peak bone mass in disturbance as prognostic indicators in between long distance running, plasma Caucasian females and its implication hypothalamic amenorrhea. Human progesterone and luteal phase length. for the prevention of osteoporosis. Reproduction 16(10), 2198–2205. Fertility and Sterility 31, 130–133. Inference from a cross-sectional model. Prior, J.C., Vigna, Y.M., Schechter, M.T. & Sherman, B.M. & Korenman, S.G. (1974) Journal of Clinical Investigation 93, Burgess, A.E. (1990) Spinal bone loss and Measurement of plasma LH, FSH,

799–808. ovulatory disturbances. New England estradiol, and P4 in disorders of the Mendelsohn, M.E. & Karas, R.H. (1999) Journal of Medicine 323, 1221–1227. human menstrual cycle: the short luteal The protective effects of estrogen on the Randolph, J.F., Sowers, M., Gold, E.B. et al. phase. Journal of Clinical Endocrinology cardiovascular system. New England (2003) Reproductive hormones in early and Metabolism 38, 89–93. Journal of Medicine 340(23), 1801–1811. menopausal transition: relationship to Simoncini, T., Varone, G., Fornari, L. Moran, T.H. & Kinzig, K.P. (2004) ethnicity, body size, and menopausal et al. (2002) Genomic and nongenomic Gastrointestinal satiety signals II. status. Journal of Clinical Endocrinology mechanisms of nitric oxide synthesis Cholecystokinin. American Journal of and Metabolism 88, 1516–1522. induction in human endothelial cells by Gastrointestinal Liver Physiology 286(2), Ravussin, E., Tschop, M., Morales, S., a fourth-generation selective estrogen G183–G188. Bouchard, C. & Heiman, ML (2001) receptor modulator. Endocrinology Myerson, M., Gutin, B., Warren, M.P. Plasma ghrelin concentration and 143(6), 2052–2061. et al. (1991) Resting metabolic rate and energy balance: overfeeding and Soules, M.R. (1988) Luteal-phase energy balance in amenorrheic and negative energy balance studies in deficiency: The most common eumenorrheic runners. Medicine and twins. Journal of Clinical Endocrinology abnormality of the menstrual cycle? Science in Sports and Exercise 23, 15–22. and Metabolism 86, 4547–4551. In: The Menstrual Cycle and Its Disorders Nakazato, M., Murakami, N., Date, Y. et al. Reutman, S.R., LeMasters, G.K., Kesner, (Pirke, K.M., Wuttke, W. & Schweiger, (2001) A role of ghrelin in the central J.S. et al. (2002) Urinary reproductive U., eds.). Springer-Verlag, New York: regulation of feeding. Nature 409, hormone level difference between 97–109. 194–198. African American and Caucasian Sowers, M., Crutchfield, M., Bandekar, R. O’Donnell, E., Goodman, J., Witt, J. et al. women of reproductive age. Fertility and et al. (1998a) Bone mineral density and (2004) Hypoestrogenism in amenorrheic Sterility 78(2), 383–391. its change in pre- and perimenopausal athletes modulates cardiovascular Rogol, A.D., Weltman, A., Weltman, J.Y. white women: the Michigan bone health function. Proceedings from the et al. (1992) Durability of the study. Journal of Bone and Mineral American College of Sports Medicine reproductive axis in eumenorrheic Research 13, 1134–1140. Annual Meeting. Medicine and Science in women during 1 year of endurance Sowers, M., Randolph, J.F., Crutchfield, M. Sports and Exercise 36(5), suppl., 533. training. Journal of Applied Physiology et al. (1998b) Urinary ovarian and Ohmichi, M., Kanda, Y., Hisamoto, K. 72(4), 1571–1580. gonadotropin hormone levels in et al. (2003) Rapid changes of flow- Rutherford, O.M. (1993) Spine and premenopausal women with low bone mediated dilatation after surgical total body bone mineral density in mass. Journal of Bone and Mineral Research menopause. Maturitas 44(2), 125–131. amenorrheic endurance athletes. Journal 13, 1191–1202. Otis, C.L., Drinkwater, B.L., Johnson, M., of Applied Physiology 74, 2904–8. Steinberg, D.S. (1987) Lipoproteins and Loucks, A.B. & Wilmore, J.H. (1997) Sanborn, C.F., Martin, B.J. & Wagner, the pathogenesis of atherosclerosis. ACSM Position Stand on the female W.W., Jr. (1982) Is athletic amenorrhea Circulation 76, 508–514. athlete triad. Medicine and Science in specific to runners? American Journal of Strott, C.A., Cargille, C.M., Ross, G.T. & Sports and Exercise 29(5), i–ix. Gynecology 143, 859–863. Lipsett, M.B. (1970) The short luteal Ott, S.M. (1990) Attainment of peak bone Sanborn, C.F, Albrecht, B.H. & Wagner, phase. Journal of Clinical Endocrinology mass. Journal of Clinical Endocrinology W.W. (1987) Athletic amenorrhea: lack and Metabolism 30, 246–251. and Metabolism 71, 1082A–1082C. of association with body fat. Medicine Tomten, S.E., Falch, J.A., Birkeland, K.I., Otto, B., Cuntz U., Fruehauf, E. et al. (2001) and Science in Sports and Exercise 19(3), Hemmersbach, P. & Hostmark, A.T. Weight gain decreases elevated plasma 207–212. (1998) Bone mineral density and ghrelin concentrations of patients with Sanborn, C.F., Horea, M., Siemers, B. & menstrual irregularities. A comparative anorexia nervosa. European Journal of Dieringer, K. (2000) Disordered eating study on cortical and trabecular bone Endocrinology 145, R5–R9. and the female athlete triad. Clinics in structures in runners with alleged Petterson, F., Fires, H. & Nillius, S.J. Sport Medicine 19(2), 199–213. normal eating behavior. International (1973) Epidemiology of secondary Schachinger, V., Britten, M.B. & Zeiher, Journal of Sports Medicine 19(2), 92–97. amenorrhea: prevalence and incidence A.M. (2000) Prognostic impact of Veldhuis, J.D., Evans, W.S., Demers, L.M. rates. American Journal of Obstetrics and coronary vasodilator dysfunction et al. (1985) Altered neuroendocrine Gynecology 117, 80–86. on adverse long-term outcome of regulation of gonadotropin secretion Pettersson, U., Stalnacke, B., Ahlenius, G. coronary heart disease. Circulation 101, in women distance runners. Journal of et al. (1999) Low bone mass density at 1899–1906. Clinical Endocrinology and Metabolism 61, multiple skeletal sites, including the Schwartz, B., Cumming, D.C., Riordan, E. 557–563. appendicular skeleton in amenorrheic et al. (1981) Exercise-associated Wade, G.N., Schneider, J.E. & Hui-Yun, L. runners. Calcified Tissue International 64, amenorrhea: a distinct entity? American (1996) Control of fertility by metabolic 117–25. Journal of Obstetrics and Gynecology 141, cues. American Journal of Physiology. Perkins, R.B., Hall, J.E., Martin, K.A. (2001) 662–670. Endocrinology and Metabolism 33(270), Aetiology, previous menstrual function Shangold, M.M., Freeman, R., Thysen, B. E1–E9. 278 chapter 20

Waller, K., Reim, J., Fenster, L. et al. (1996) Williams, N.I. (2003) Lessons from conservation in amenorrheic compared Bone mass and subtle abnormalities in experimental disruptions of the with eumenorrheic distance runners? ovulatory function in healthy women. menstrual cycle in humans and Journal of Applied Physiology 72(1), Journal of Clinical Endocrinology and monkeys. Medicine and Science in Sports 15–22. Metabolism 81, 663–668. and Exercise 35(9), 1564–1572. Winters, K.M., Adams, W.C., Meredith, Waller, K., Swan, S.H., Windham, G.C. Williams, N.I., Young, J.C., McArthur, J.W. C.N., Van Loan, M.D. & Lasley, B.L. et al. (1998) Use of urine biomarkers to et al. (1995) Strenuous exercise with (1996) Bone density and cyclic ovarian evaluate menstrual function in healthy caloric restriction: effect on luteinizing function in runners and active controls. premenopausal women. American hormone secretion. Medicine and Science Medicine and Science in Sports and Exercise Journal of Epidemiology 147(11), in Sports and Exercise 27(10), 1390–1398. 28(7), 776–785. 1071–1080. Williams, N.I., Clark, K.L., McConnell, Wren, A.M., Seal, L.J., Cohen, M.A. et al. Warren, M.P. (1980) The effects of H.J., Matuch, A. & O’Connor, K.A. (2001) Ghrelin enhances appetite and exercise on pubertal progression and (2000) Menstrual irregularities and increases food intake in humans. Journal reproductive function in girls. Journal of disordered eating in female athletes: of Clinical Endocrinology and Metabolism Clinical Endocrinology and Metabolism 51, survey vs. follow-up physiological 86, 5992–5995. 1150–1157. studies. Medicine and Science in Sports Yeager, K.K., Agostini, R., Nattiv, A. & Warren, M.P. (1996) Evaluation of and Exercise 32(5), S64. Drinkwater, B. (1993) The female athlete secondary amenorrhea. Journal of Clinical Williams, N.I., Caston-Balderrama, A., triad: disordered eating, amenorrhea, Endocrinology and Metabolism 81(2), Helmreich, D.L. et al. (2001a) osteoporosis. Medicine and Science in 437–442. Longitudinal changes in reproductive Sports and Exercise 25, 774–777. Warren, M.P. Voussoughian, F., Geer, E.B. hormones and menstrual cyclicity in Young, N., Formica, C., Szmukler, G. et al. et al. (1999) Functional hypothalamic cynomolgus monkeys during strenuous (1994) Bone density at weight-bearing amenorrhea: hypoleptinemia and exercise training: abrupt transition to and non weight-bearing sites in ballet disordered eating. Journal of Clinical exercise induced amenorrhea. dancers: the effects of exercise, Endocrinology and Metabolism 84, Endocrinology 142, 2381–2389. hypogonadism and body weight. Journal 873–877. Williams, N.I., Helmreich, D.L., Parfitt, of Clinical Endocrinology and Metabolism White, M.A., Kohlmaier, J.R., Varnado- D.B., Caston-Balderrama, A. & 78, 449–54. Sullivan, P. & Williamson, D.A. (2003) Cameron, J.L. (2001b) Evidence for a Zeni-Hoch, A., Dempsey, R.L., Carrera, Racial/ethnic differences in weight casual role of low energy availability G.F. et al. (2003) Is there an association concern: protective and risk factors for in the induction of menstrual cycle between athletic amenorrhea and the development of eating disorders disturbances during strenuous exercise endothelial cell dysfunction? Medicine and obesity among adolescent females. training. Journal of Clinical Endocrinology and Science in Sports and Exercise 35(3), Eating and Weight Disorders 8(1), and Metabolism 86, 5184–5193. 377–83. 20–25. Wilmore, J.H. (1991) Eating and weight Zipfel, S., Seibel, M.J., Lowe, B. et al. (2001) Williams, N.I. (1998) Low energy disorders in the female athlete. Osteoperosis in eating disorders: availability and reproductive International Journal of Sports Nutrition 2, a follow-up study of patients with disturbances: a review of clinical and 104–117. anorexia and bulimia nervosa Journal of hormonal effects. Journal of Dance Wilmore, J.H., Wambsgans, K.C., Clinical Endocrinology and Metabolism Medicine and Science 2(1), 19–31. Brenner, M. et al. (1992) Is there energy 86(11), 5227–5233. Chapter 21

Regulation of Testicular Function: Changes in Reproductive Hormones During Exercise, Recovery, Nutritional Deprivation and Illness

SHALENDER BHASIN

Introduction pituitary is stimulated by pulsatile gonadotropin- releasing hormone (GnRH) secretion, and regulated The mammalian testis is the site of germ cell develop- by negative feedbacks from gonadal hormones, ment and androgen production (Rommerts 2004). including gonadal steroids, and inhibin and activin. Testosterone, a 19-carbon steroid secreted by the testis, is the predominant androgen in most mam- Gonadotropin-releasing hormone mals. Testosterone plays a critical role in mam- secretion by hypothalamic neurons malian reproduction; it is essential for maintaining sexual function, germ cell development and access- Developmental migration of GNRH neurons. The ory sex organs. In the adult animal, testosterone neurons that secrete GnRH originate in the region has additional effects on the muscle, bone, hema- of the olfactory apparatus (Schwanzel-Fukuda & topoeisis, coagulation, plasma lipids, protein and Pfaff 1989) and migrate, along the olfactory and carbohydrate metabolism, psychosexual and cog- vomeronasal nerves, into the forebrain and then into nitive function. During sexual differentiation of their final location in the hypothalamus (Fig. 21.2). the mammalian fetus, testosterone masculinizes the This orderly migration of GnRH neurons requires Wolffian structures and causes the external genitalia the co-ordinated action of direction-finding mole- to form a scrotum and penis. In addition, increasing cules, adhesion proteins such as the KALIG-1 gene testosterone levels during pubertal development product and fibroblast growth receptor, and enzy- promote somatic growth and virilization of boys. mes that help the neuronal cells burrow their way Androgen production by the testes is regulated through intercellular matrix. Mutations of any of primarily by pituitary luteinizing hormone (LH), these proteins could arrest the migratory process while germ cell development requires the co- and result in GnRH deficiency. In a subset of patients, ordinated action of follicle-stimulating hormone failure of this developmental migration of GnRH (FSH) and high intratesticular testosterone con- neurons into their final location in the hypothala- centrations generated by the Leydig cells under the mus results in a clinical disorder called idiopathic influence of LH (Rommerts 2004). Paracrine interac- hypogonadotropic hypogonadism (IHH) that is tions between Sertoli and germ cells are also import- characterized by GnRH deficiency and impaired ant in the regulation of spermatogenesis, although gonadotropin secretion by the pituitary (Legouis the precise role of Sertoli cell in regulation of germ et al. 1991). cell development is not fully understood. Testicular function is regulated by a series of feed- Hypothalamus as the integrating center for the male back and feedforward mechanisms that operate at reproductive axis. The hypothalamus serves as the the level of the hypothalamus, the pituitary, and the integrating center for the reproductive system and testes (Fig. 21.1). Thus, LH and FSH secretion by the co-ordinates the regulatory signals from the higher 279 280 chapter 21

Light and other centers and the feedback from the gonads (Knobil environmental cues 1980; Crowley et al. 1991). The hypothalamus re- ceives neural input from the central nervous system n Olf motio actor E that reflects the effects of emotion, stress, light, olfact- E y ne b rg ss ala y ory stimuli, temperature and other environmental tre nc S e stimuli. The feedback signals from the gonads include steroid hormones (testosterone and estra- diol) and protein hormones (inhibins and actvins).

− Hypothalamus − Regulation of LH and FSH by pulsatile GnRH secretion. − GnRH GnRH is a major regulator of gonadotropin secre- tion, and increases LH and FSH secretion from pituit- + ary cells both in vitro and in vivo. Pulsatile secretion of GnRH is essential for maintaining normal LH and Pituitary gonadotrope FSH secretion from the pituitary (Belchetz et al. ++ 1978; Knobil 1980; Shupnik 1990; Crowley et al. 1991; − Weiss et al. 1992). Continuous GnRH infusion or − administration of a long-acting GnRH agonist leads − LH FSH to decreased LH and FSH secretion, a phenomenon referred to as down-regulation (Fig. 21.3) (Belchetz T E 2 + + et al. 1978; Knobil 1980). The pattern of GnRH signal (amplitude and frequency) is important in deter- Leydig Sertoli mining the quantity and quality of gonadotropins T → Inhibin cells T cells secreted (Belchetz et al. 1978; Haisenleder et al. 1988, 1991; Kim et al. 1988a, 1988b; Yuan et al. 1988; Germ cell compartment Shupnik 1990; Weiss et al. 1992). Marked increase in GnRH pulse frequency also desensitizes the gonadotrope, resulting in decreased LH and FSH secretion (Belchetz et al. 1978; Mercer et al. 1988; Sperm Shupnik 1990). The electrophysiologic activity of hypothalamic GnRH neurons is linked to episodic Fig. 21.1 A schematic diagram of the hypothalamic– pituitary–testicular axis. The hypothalamus serves as GnRH secretion. the integrating center for the male reproductive axis. The transcription of LH-β gene is induced by Emotion, olfaction, energy balance, light and stress exert pulsatile GnRH administration in vitro (Wierman their effects on human reproduction through circuits that impinge on hypothalamic gonadotropin-releasing et al. 1989; Shupnik 1990; Weiss et al. 1992). Con- hormone (GnRH) secreting neurons. Pulsatile GnRH tinuous infusion of GnRH up-regulates only the α- secretion by hypothalamic neurons stimulates luteinizing gene transcription, but not LH or FSH β-subunit hormone (LH) and follicle-stimulating hormone (FSH) secretion by the pituitary. LH binds to G-protein-coupled gene transcription (Haisenleder et al. 1988). Pulsatile receptors on the Leydig cells in the testis and stimulates GnRH administration also modifies polyadenyla- the production of testosterone; high intratesticular tion of LH subunit mRNA (Weiss et al. 1992). testosterone concentrations along with FSH are essential for initiating and maintaining spermatogenesis. The frequency of GnRH stimulus is important in Testosterone has a negative feedback effect on pituitary differential regulation of LH-β and FSH-β genes LH secretion and hypothalamic GnRH secretion directly (Haisenleder et al. 1988). Faster frequencies increase and indirectly through its conversion to estradiol. α β β Circulating glycoproteins secreted by the Sertoli cells, and LH- , and slower frequencies FSH- , lead- known collectively as inhibins, also regulate FSH ing to speculation that alterations in GnRH pulse secretion. As receptors for FSH and androgens are located frequency may be one mechanism by which two on the Sertoli cells, it is generally believed that the hormonal influences on germ cells are mediated via the functionally distinct gonadotropins can be regu- Sertoli cells. lated by a single hypothalamic-releasing hormone regulation of testicular function 281

poa vno ob ob gt poa gt 11E vno vno gt ob

13E vno 14E

16E

Fig. 21.2 Developmental origin and migration of the gonadotropin-releasing hormone (GnRH) secreting neurons. The neurons that secrete GnRH originate in the region of the olfactory apparatus. This figure shows the migratory route of GnRH secreting neurons in sagittal sections of the mouse embryo as determined by GnRH immunohistochemistry. On day 11, GnRH neurons are seen in proximity to the vomeronasal organ (vno) and the olfactory placode. By day 13, the migration of these neurons into the nasal septum and along the path of vomeronasal nerves and nervous terminalis is evident. By day 16, most of the GnRH neurons have entered the forebrain and some have entered into the preoptic area of the hypothalamus. gt, ganglion terminale; ob, olfactory bone; poa, preoptic area; vno, vomeronasal organ. (Reproduced from Schwanzel-Fukuda & Pfaff 1989.)

(Haisenleder et al. 1988). Continuous infusion of levels achieved after intravenous administration of GnRH or administration of a GnRH agonist leads to this dose (500–1000 pg·mL–1) are similar to those a decrease in LH-β mRNA levels but α mRNA levels obtained in the primates by direct sampling of the remain elevated (Haisenleder et al. 1988; Kim et al. hypophyseal-portal blood (Crowley et al. 1991). In 1988a, 1988b; Yuan et al. 1988). men with IHH, an interpulse interval of 2 h appears A great deal of information about the physiology optimum (Crowley et al. 1991). Increasing the fre- of GnRH secretion has emerged from examination quency of GnRH pulses leads to progressive de- of the LH and FSH pulse patterns in normal men crease in LH responsiveness to GnRH (Rebar et al. and women, and from GnRH replacement studies 1976). Decreasing the pulse frequency or increasing in patients with idiopathic hypogonadotropic hypo- the interpulse interval increases the amplitude of gonadism (IHH) (Urban et al. 1988; Crowley et al. the subsequent LH pulse. There is a linear relation- 1991). Studies in such patients with hypothalamic ship between the log of the dose of GnRH pulse and GnRH deficiency have indicated that GnRH pulses the amount of LH, FSH and free α subunit secreted at a dose of 25 ng·kg–1, given intravenously to men, (Spratt et al. 1986; Whitcomb et al. 1990). In the adult can replicate normal pulsatile LH secretion in all its man, the magnitude of the LH response to GnRH is characteristics (Crowley et al. 1991). The peak GnRH considerably greater than that of FSH. 282 chapter 21

40 1 Pulse/h 5 Pulse/h 1 Pulse/h 400

35 350

30 300 ) )

25 –1 –1 250

20 200

LH (ng·mL 15 150 FSH (ng·mL

10 100

5 50

0 0 1510550 10 1520 25 30 35 40 Days

Pulsatile Continuous Pulsatile 20 200

15 150 ) ) –1 –1

10 100 LH (ng·mL FSH (ng·mL

5 50

0 0 10550 10 1520 25 30 35 Days

Fig. 21.3 Knobil’s pioneering experiments demonstrated that the pulsatile nature of gonadotropin-releasing hormone (GnRH) signal is essential for maintaining normal luteinizing hormone (LH) and follicle-stimulating hormone (FSH) output. Knobil et al. (1980) ablated GnRH secreting neurons in the hypothalamus to create a monkey model that was deficient in GnRH. These monkeys also underwent gonadectomy in order to remove the feedback influences of the gonads. In this gonadectomized, GnRH deficient monkey model, pituitary LH and FSH secretion was normalized by pulsatile administration of GnRH at a frequency of one pulse per hour. Continuous infusion of GnRH decreased LH and FSH secretion, a phenomenon referred to as down-regulation. Similarly, administration of GnRH at an increased frequency of five pulses per hour resulted in decreased LH and FSH secretion. Restoration of pulse frequency to one pulse per hour restored normal LH and FSH output. (Reproduced with permission from Knobil et al. 1980 and Belchetz et al. 1978.) regulation of testicular function 283

Intensive sampling in normal adult men and & Rosenfield 2002). Cell specialization and pro- women reveals a wide spectrum of LH pulse char- liferation of differentiated cell types requires the acteristics (Urban et al. 1988). The median values for expression of transcription factors, Pit1 and Prop1. LH pulse parameters in men, reported in one such Pit-1 has a POU-specific and a POU-homeo DNA- recent study (Urban et al. 1988) were as follows: binding domain (Scully & Rosenfield 2002). The interpulse interval 55 min; LH peak duration 40 min; Prop1 gene encodes a transcription factor with a LH pulse amplitude 37% of basal (1.8 mLU·mL–1 single DNA-binding domain. While Pit-1 mutations incremental). Wide variations in LH pulse para- are associated with deficiencies of growth hormone meters in apparently healthy normal men and wo- (GH), thyroid-stimulating hormone (TSH) and pro- men dictate a need for caution in interpretation of lactin, mutations in Prop1 are associated with defi- mild deviations in LH pulse frequency and ampli- ciencies of LH and FSH in addition to deficiencies of tude parameters. The sampling frequency and the GH, prolactin and TSH. Expression of the HESX1 paradigm used to quantitate pulse characteristics gene precedes expression of Prop-1 and Pit-1; muta- can have significant impact on the false-negative tions in this gene are associated with septo-optic and false-positive rates and on the observed pulse dysplasia and panhypopituitarism (Parks et al. parameters (Urban et al. 1988). 1999). GnRH action on the gonadotrope is mediated via its binding to specific membrane receptors leading Biochemical structure and molecular biology of LH and to aggregation of receptors and calcium-dependent FSH. The family of pituitary glycoprotein hormones LH release (Conn et al. 1981, 1982). includes LH, FSH, TSH and chorionic gonadotropin (CG) (Sairam 1983; Ryan et al. 1987; Gharib et al. Gonadotropin secretion by pituitary 1990). Each of these hormones is heterodimeric, consisting of an α and a β subunit. The primary Functional anatomy and development of the pituitary structures of the α subunits of LH, FSH, TSH and gland. The bulk of immunocytologic evidence sug- CG are nearly identical within a species; the biologic gests that a single cell type within the pituitary specificity is conferred by the dissimilar β subunit. secretes both LH and FSH (Moriarty 1973; Kovacs Significant homology between the two subunits et al. 1985). Gonadotropes, the cells that secrete LH suggests that these subunits arose from a common and FSH, constitute about 10–15% of anterior pitu- ancestral gene. Individual subunits are not biolo- itary cells (Moriarty 1973; Kovacs et al. 1985) and are gically active; formation of the heterodimer is essen- dispersed throughout the anterior pituitary close tial for biologic activity. The subunits are linked to the capillaries. Gonadotropes are easily demon- internally by disulfide bonds; the location of the cys- strable in the fetal and prepubertal pituitary gland teines is important in determining the folding and (Childs et al. 1981); however, their number is low the three-dimensional structure of the glycoprotein before sexual maturation. Castration leads to an (Sairam 1983; Ryan et al. 1987; Gharib et al. 1990). increase in the size as well as the number of The α subunit of LH contains two asparagine-linked gonadotropes. Adenohypophyseal cells are derived carbohydrate chains, while the β-subunit chain con- from a common multipotential stem or progenitor tains one or two (Table 21.1) (Baenziger 1990). The cell. Genetic analyses of mutations associated with CG β subunit also contains O-linked oligosaccha- developmental disorders of the pituitary have re- rides not found on the LH dimer (Cole et al. 1984). vealed the molecular mechanisms of pituitary devel- Although free uncombined α subunit is secreted opment and cell lineage determination (Ingraham by the pituitary into the circulation, it is generally et al. 1988; Scully & Rosenfield 2002). Co-ordinated, believed that the free β subunit is not secreted to any temporal expression of a number of homeodomain significant degree via this route. transcription factors directs the embroyological Emergence of CG as a separate gonadotro- development of the pituitary and its differentiated pin occurred relatively recently during evolution cell types. Three homebox genes Lbx3, Lbx4 and (Kornfeld & Kornfeld 1976; Fiddes et al. 1984). Titf1 are essential for early organogenesis (Scully Unlike LH, which is found in the pituitaries of a 284 chapter 21

Table 21.1 General structure of the pituitary glycoprotein hormones. There is considerable homology among the glycoprotein α, LH-β, FSH-β, TSH-β and hCG-β genes and proteins. (Reproduced from Grossman et al. 1997.)

Gene No. of mRNA No. of No. of glycosylation sites Subunit Locus length (Kb) exons (introns) length (Kb) amino acids (location)*

Common α 6p21.1-23 9.4 4 (3) 0.8 92 2 (N: 52,78)† TSH-β 1p22 4.9 4 (3) 0.7 118‡ 1 (N: 23) LH-β 19q13.3 1.5 4 (3) 0.7 121 1(N: 30) CG-β 19q13.3 1.9§ 4 (3) 1.0 145 6 (N: 13,30; S: 121,127,132,138) FSH-β 11p13 3.9 4 (3) 1.8 111 2 (N: 7,24)

*Oligosaccharide chains are attached either to asparagine (N) (N-linked) or to serine (S) (O-linked). N or S residues are numbered according to their position in the respective sequence. †Free subunit may also contain an additional site of O-glycosylation at threonine (T) 39. ‡118-Amino acid coding region; six amino acids can be cleaved at the C-terminal end. §In contrast to all other glycoprotein hormone subunit genes which exist as a single copy, hCG is encoded by a cluster of six genes which vary in length. large number of species, CG is found only in the age enzyme (Simpson 1979; Mori & Marsh 1982), a placenta of certain mammalian species such as cytochrome P450-linked enzyme that converts cho- horses, baboons and humans (Fiddes et al. 1984). lesterol to pregnenolone, the rate-limiting step in The α and β subunits of LH and FSH are encoded testosterone biosynthesis. LH increases the delivery by separate genes (Fiddes et al. 1984). The LH–CG- of cholesterol to the side-chain cleavage enzyme, β gene cluster in the human incorporates seven thus increasing its capacity to convert cholesterol to CG-like genes, including one which codes for the pregnenolone (Simpson 1979; Mori & Marsh 1982). hLH-β gene (Fiddes et al. 1984). The general organ- A steroidogenesis acute regulatory protein (STAR) ization of the LH-β gene four exons and three introns makes cholesterol available to the cholesterol side- is similar to other glycoprotein hormone β genes chain complex and regulates the rate of testosterone (Table 21.1). biosynthesis (Clark & Stocco 1996). Peripheral ben- zodiazepine receptor, a mitochondrial cholesterol- Regulatory role of LH. Testosterone secretion by binding protein involved in cholesterol transport, is Leydig cells is under the control of LH. LH binds present in high concentration in outer mitochon- to G-protein-coupled receptors on Leydig cells drial membrane, and has also been proposed as and activates the cyclic adenosine monophos- an acute regulator of Leydig cell steroidogenesis. phate (cAMP) pathway. The luteinizing hormone– The long-term effects of LH include stimulation chorinonic gonadotropin (LH–hCG) receptor shares of gene expression and synthesis of a number of homology with other G-protein-coupled receptors key enzymes in the steroid biosynthetic pathway, such as rhodopsin, adrenergic, TSH and FSH recep- including the side-chain cleavage enzyme, 3-β tors (Fig. 21.4) (McFarland et al. 1989; Sprengel et al. hydroxysteroid dehydrogenase, 17-α hydroxylase 1990). These G-protein-coupled receptors are trans- and 17,20-lyase (Simpson 1979; Mori & Marsh 1982). membrane proteins that share a structural motif Although LH also activates phopholipase C path- of seven membrane-spanning domains. The amino way, it is unclear if this pathway is essential for LH- terminus of the receptor protein constitutes the mediated stimulation of testosterone production. In extracellular domain. The carboxy terminus con- addition, insulin-like growth factor-I, insulin-like sists of the seven membrane-spanning segments growth factor binding proteins, inhibins, activins, and a small cytoplasmic tail, and has several serines transforming growth factor-β, epidermal growth and threonines that may serve as phosphorylation factor, interleukin-1, basic fibroblast growth factor, sites (McFarland et al. 1989; Sprengel et al. 1990). gonadotropin-releasing hormone, vasopressin and LH increases the activity of the side-chain cleav- another group of poorly characterized mitochondrial regulation of testicular function 285

Fig. 21.4 General structure of the luteinizing hormone–chorinonic 133 gonadotropin (LH–hCG) receptor. NH R The normal amino acids are shown, 3 Extracellular but the activating mutations are indicated by black circles with light letters, whereas the inactivating mutations are indicated by gray M 398 S 616 I II III IV V VI VII circles with white letters. The corresponding codon numbers also are shown. Most activating mutations occur in the sixth transmembrane domain and third intracellular loop, COOH whereas inactivating mutations VVI Intracellular occur in the extracellular domain, A A transmembrane domains V–VII S I A and the third intracellular loop. F F S I P A Activating mutations include M 581 C T D F Asp578Gly, Ile542Leu, Asp564Gly, I 542 T 545 578 F 577 C I C A L I Asp578Tyr, Cys581Arg, Met571Ile, T 574 A I 571 M Thr577Ile, Ala568Val, Ala572Val and K K I K Met398Thr. Inactivating mutations Y A 568 F are Cys545X, Ala593Pro, Arg554X, I A N K K T D V A T Ser616Tyr and Arg133Cys. An X M R 564 554 N L indicates a stop codon, which is TAA, P K TAG or TGA. (Reproduced from Layman 1999.) proteins have been implicated in control of steroido- In the rat and non-human primate, testosterone genesis in Leydig cells. alone can maintain spermatogenesis when admin- istered shortly after hypophysectomy or stalk re- The regulatory role of FSH in the male mammal. section (Marshall et al. 1983; Sharpe et al. 1988; FSH binds to specific receptors on Sertoli cells and Stager et al. 2004). However, if testosterone is stimulates the production of a number of pro- given after a lapse of several weeks to months, it is teins including inhibin-related peptides, transferrin, much less effective in reinitiating spermatogenesis. androgen-binding protein, androgen receptor and Spermatogenesis maintained by treatment of hypo- 7-glutamyl transpeptidase. However, the role of physectomized male rodent or non-human primate FSH in the regulation of the spermatogenic process by testosterone is qualitatively, but not quantitat- remains poorly understood. The prevalent dogma is ively, normal (Marshall et al. 1983; Sharpe et al. 1988; that LH acts on Leydig cells to stimulate the pro- Stager et al. 2004). Combination of testosterone and duction of high intratesticular levels of testosterone FSH is more effective than testosterone alone in (Boccabella 1963; Steinberger 1971; Sharpe 1987). reinitiating spermatogenesis (Stager et al. 2004). Testosterone then acts on spermatogonia and prim- Thus, although LH alone can maintain or reinitiate ary spermatocytes leading the germ cells through spermatogenesis, FSH is required for quantitatively the meiotic division. FSH is felt to be essential for normal sperm counts. spermiogenesis, i.e. the maturation process by In men with prepubertal onset of both LH and which spermatids develop into mature spermato- FSH deficiency, LH or hCG alone is unable to initi- zoa. However, evidence from experimental animals ate spermatogenesis (Bardin et al. 1969; Matsumoto and patients with IHH treated with gonadotropins et al. 1983, 1984; Finkel et al. 1985). On the other suggest that FSH plays a more complex role in main- hand, if gonadotropin deficiency is acquired after taining quantitatively normal spermatogenesis. the patient has completed pubertal development, 286 chapter 21

LH or hCG alone can reinitiate qualitatively normal to its C-terminus containing seven transmembrane spermatogenesis (Finkel et al. 1985). Thus, FSH is segments (Sprengel et al. 1990). required for initiating the spermatogenic process but, once this has occurred, testosterone in high Feedback regulation of luteinizing doses can maintain spermatogenesis. These observa- hormone and follicle-stimulating tions suggest that FSH may be involved in some sort hormone secretion of ‘programming’ in the peripubertal period, after which LH alone can maintain germ cell develop- Feedback regulation by testosterone. Testosterone ment and maturation. plays an important role in feedback regulation of Androgen concentrations in the testis are much gonadotropins in the male. Serum LH levels rise higher than those in serum. However, considerable promptly and serum FSH gradually after castration confusion exists with regard to the significance of in a number of experimental animals (Yamamoto high intratesticular testosterone levels (Sharpe 1987; et al. 1970; Badger et al. 1978). The mRNAs for α, LH- Sharpe et al. 1988; Stager et al. 2004). For example, β (Gharib et al. 1986) and FSH-β (Gharib et al. 1987) stimulatory effects of exogenous testosterone on increase after castration, although the changes in spermatogenesis in the rat are not associated with FSH-α mRNA are more modest. The postcastra- proportionate elevations of intratesticular testo- tion rise in serum LH and LH-β mRNA levels is sterone levels. In the adult hypophysectorized or brought about both by an increase in the gona- GnRH antagonist-treated monkey that is supple- dotrope number and hypertrophy of individual mented with testosterone, a direct relationship gonadotropes (Childs et al. 1987). Testosterone between testicular testosterone levels and sper- replacement, started at the time of, or soon after, matogenesis has not been found (Stager et al. 2004). castration can attenuate the postcastration rise in α The method of post-mortem collection of testicular and LH-β mRNAs and serum LH levels, but has tissue affects estimation of intratesticular testo- little effect on FSH-β mRNA levels (Gharib et al. sterone concentrations (Stager et al. 2004). Thus, 1986, 1987). the relationship between intratesticular testosterone The effects of testosterone on FSH secretion and concentrations, FSH and spermatogenesis remains synthesis are complex. The net in vivo effect of poorly understood. Androgen receptors are present testosterone administration to normal men is inhibi- on Sertoli and peritubular cells, some Leydig cells tion of serum FSH levels (Swerdloff et al. 1979; and endothelial cells of the small arterioles. How- Winters et al. 1979). However, the direct effects ever, we do not know whether androgen receptors of testosterone on FSH output at the pituitary level are also present on germ cells. It is generally believed are stimulatory (Steinberger & Chowdhury 1977; that androgen effects on spermatogenesis are medi- Bhasin et al. 1987; Gharib et al. 1987). In isolated ated indirectly through Sertoli cells, although it is pituitary cell cultures, testosterone increases FSH possible that testosterone might also directly affect release into the media (Steinberger & Chowdhury germ cell development. Testosterone affects protein 1977). This is attended by a three to fourfold secretion by both round spermatids and Sertoli increase in FSH-β mRNA levels (Gharib et al. 1990). cells. The expression of androgen receptors is max- In intact male rats, in whom GnRH actions are imal in Stages VI–VII of the seminiferous epithe- blocked by administration of a GnRH antagonist, lium; testosterone regulates germ cell apoptosis in a testosterone increases serum FSH levels in a dose- stage-specific manner. dependent manner (Bhasin et al. 1987). Bhasin et al. The transduction of the FSH signals to the germ (1987) demonstrated that in castrated animals cells requires mediation of Sertoli cells, as FSH treated with a GnRH antagonist, graded doses of receptors are present on the Sertoli cells but are testosterone increase serum FSH levels. These data lacking on the germ cells. FSH receptor is also a G- indicate that the stimulatory effects of testosterone protein-linked polypeptide consisting of a glycosy- on serum FSH levels are not mediated through lated extramembranous domain which is connected effects on a gonadal inhibitor of FSH, but rather regulation of testicular function 287

directly at the pituitary level. Testosterone increases Feedback inhibition by estrogens. Estrogens can exert FSH-β but not LH-β mRNA levels. However in the both stimulatory and inhibitory effects on gonado- intact male animal, testosterone inhibits GnRH- tropin synthesis and secretion depending on the stimulated FSH secretion, accounting for the net dose, duration, the presence or absence of GnRH inhibition of serum FSH levels. and other physiologic factors. Data from experi- Testosterone inhibits LH secretion when given to mental animals suggest that the stimulatory effects normal men and rats (Santen 1975; Matsumoto et al. of estrogens on gonadotropin synthesis and secre- 1984; Veldhuis et al. 1984). These inhibitory effects tion in vivo are exerted directly at the pituitary level, are largely exerted at the hypothalamic level, a con- while the inhibitory effects are mediated at the clusion supported by observations that testosterone hypothalamic level (Neill et al. 1977; Clarke & decreases the frequency of LH pulses in eugonadal Cummins 1982; Zmeili et al. 1986; Saade et al. 1989; men (Matsumoto & Bremner 1984; Scheckter et al. Shupnik et al. 1989; Yamaji et al. 1992). Estrogen 1989; Finkelstein et al. 1991a). Androgens have no administration leads to a decrease in LH pulse fre- direct effects on LH-β mRNA levels in rat pituitary quency suggesting a hypothalamic site of action monolayer cultures. Similarly, in GnRH antagonist- (Shupnik et al. 1989). Estradiol treatment of hypo- treated male rats, graded doses of testosterone lead thalamic slices decreases GnRH mRNA expression only to an increase in FSH-β but not LH-β mRNA (Hall & Miller 1986). Finally, transcription of all levels (Bhasin et al. 1987). In contrast to rats, in men three gonadotropin subunits is negatively regulated with IHH, the amplitude of LH pulses, initiated and by estradiol in vivo, even though the direct in vitro maintained by pulsatile GnRH therapy, is reduced effects on pituitary are stimulatory (Neill et al. 1977; by testosterone administration, indicating that in Clarke & Cummins 1982; Saade et al. 1989). Estra- humans, testosterone has additional actions at the diol inhibits LH pulse amplitude in normal men pituitary level in attenuating pituitary LH response and in GnRH-deficient men maintained on GnRH to GnRH (Matsumoto et al. 1984; Scheckter et al. (Finkelstein et al. 1991b). These studies provide 1989; Finkelstein et al. 1991a). These studies demon- evidence that in men, estradiol inhibits LH by an strate that testosterone or one of its metabolites action predominantly at the pituitary site. inhibits gonadotropin secretion at both the pituitary and hypothalamic levels in men. Inhibins, activins and follistatins. The hypothesis The inhibitory effects of testosterone are medi- that a peptide of gonadal origin selectively regulates ated directly by testosterone itself and indirectly FSH secretion was postulated over 70 years ago through its metabolites, estradiol and dihydro- (McCullagh 1932); however, it wasn’t until 1985 that testosterone (DHT). Administration of inhibitors the structure of inhibin-related peptides was finally of aromatase or 5-α reductase is associated with characterized (Ling et al. 1985; Burger & Igarashi increases in LH concentrations consistent with 1988; Vale et al. 1988; Ying 1988). Inhibins are the role of estradiol and DHT in augmenting the dimeric proteins consisting of a common α subunit β β β inhibitory feedback effects of testosterone (Santen and one of two subunits, A or B (Fig. 21.5). The α β α β 1975; Finkelstein et al. 1991b; Gormley & Rittmaster heterodimers of : A form inhibin A and a: : B het- 1992). However, administration of a non-aromatiz- erodimers form inhibin B (Vale et al. 1988). Inhibins able androgen, DHT, also inhibits LH secretion A and B are equipotent in their FSH-inhibiting consistent with the proposal that aromatization potency, although inhibin B appears to be the of testosterone is not obligatory for mediating its dominant circulating form of inhibin in men. In β inhibitory effects on LH secretion (Santen 1975). addition, A subunits can form homodimers called α β Similarly, 5- reduction of testosterone is not essen- activin A or heterodimers with the B subunits tial for the LH-inhibitory effects of testosterone called activin AB (Mason et al. 1985). Both activins A (Gormley & Rittmaster 1992). The hypothalamic and B stimulate FSH secretion in vitro. effects of testosterone are mediated via opiatergic Inhibin-related peptides are widely distributed in pathways (Veldhuis et al. 1984). organ systems and have significant homology with 288 chapter 21

α β β subunit A subunit B subunit Although the original inhibin hypothesis postu- lated inhibin as a selective regulator of FSH, under 20 KDa 15 KDa 15 KDa some conditions inhibins also regulate LH output (Vale et al. 1988). Conversely, both FSH and LH Inhibin A Inhibin B regulate inhibin production by Sertoli cells in the rat and in the human male (McLachlan et al. 1988; Keman et al. 1989; Krummen et al. 1989). FSH and LH both increase inhibin-α mRNA (Krummen et al. Activin AActivin AB Activin B 1989, 1990). FSH effects on inhibin subunits are mediated via cAMP (Najmabadi et al. 1993). Follistatins are another class of FSH inhibitors (Ueno et al. 1987) that are made up of glycosylated Fig. 21.5 General structure of the inhibin/activin family single chain polypeptides with homology to pancre- of gonadal hormones. atic secretory inhibitory protein and human epider- mal growth factor. The mature follistatin protein members of a family of proteins that includes contains four repeating domains; three are highly Mullerian inhibiting substance, transforming growth similar among themselves and to hEGF and PSTI. factor-β, bone matrix proteins and the decapenta- The physiologic role of follistatins is not known; plegic gene complex of Drosophila (Vale et al. 1988; emerging data suggest that they may act primarily Ying 1988), which play an important role as regula- to suppress FSH release. Follistatins are potent tors of growth and differentiation in diverse tissues. inhibitors of estrogen production in granulosa cells Thus, activin interacts with erythropoietin to stimu- and can bind activin. Follistatins also act as binding late erythropoiesis. Activin is also an important proteins for other growth regulating proteins such regulator of homeobox genes. In the testis, activin as myostatin. has been shown to regulate spermatogonial multi- Activins regulate intragonadal function in both plication (Mather et al. 1990). the male and the female (Vale et al. 1988). In the The role of inhibins in the adult male animal testis, activins suppress LH-stimulated testosterone remains unclear. Immunoneutralization studies in production while inhibins suppress it. In granulosa rats reveal that infusion of anti-inhibin sera leads to cells, activins increase aromatase activity but inhibit an increase in serum FSH levels only in the female progesterone synthesis (Vale et al. 1988; Ying 1988). and prepubertal male animal, but not in the adult Activin B may also act as an autocrine/paracrine male animal (Rivier et al. 1986). These studies ques- mediator in the pituitary and modulate FSH-α gene tioned the in vivo role of inhibin as an FSH regulator expression. in the adult male. Subsequently, Culler & Negro- Vilar (1990) demonstrated that when Leydig cells in Ontogenesis of luteinizing hormone adult male rats are destroyed by a Leydig cell-specific and follicle-stimulating hormone toxin, ethane dimethane sulfonate (EDS), adminis- secretion during different phases of tration of anti-inhibin sera leads to an increase in human life serum FSH. These data suggest that under basal conditions in the adult male, testosterone plays a Fetal life. GnRH is demonstrable in the fetal hypo- more important role in FSH regulation and that thalamus at 6 weeks of gestation (Lee, P.A. 1988). inhibin effects are unmasked only when testosterone Fetal pituitary contains measurable amounts of levels are lowered. Indeed, using specific, two-site LH and FSH by 10 weeks, and by 11–12 weeks LH directed assays, an inverse relationship has been response to GnRH can be shown. Serum LH and demonstrated between circulating inhibin B levels FSH levels rise to a peak at about 20 weeks and FSH levels in healthy men and women, and in (Clements et al. 1980; Lee, P.A. 1988). In the second men with disorders of germ cell development. half of pregnancy, serum LH and FSH levels in the regulation of testicular function 289

fetus decline gradually. The mechanism of this the ability to respond to GnRH and to hCG, respect- decline is not known but several factors operative in ively. The response of prepubertal pituitary to the second half of pregnancy may be responsible. GnRH stimulus is relatively damped. In addition, The rise in sex steroid secretion by the fetal gonad, the GnRH-induced rise in serum FSH in prepubertal the rising maternal estrogen levels and the develop- humans is greater than that in LH. This is in contrast ment of negative feedback mechanisms may all to an adult individual in which a single dose of contribute to tonic inhibition of the hypothalamus GnRH causes a greater rise in LH. During pubertal and the pituitary gland by the ambient sex steroid maturation, serum LH and FSH levels rise (Sizonenko concentrations (Winter et al. 1975; Sizonenko 1979). 1979). The activation of FSH secretion precedes that Placental hCG plays a significant role in stimulat- of LH. Sleep-entrained pulsatile secretion of LH ing androgen production by the fetal testis in early is highly characteristic of early stages of puberty pregnancy (Lee, P.A. 1988). High androgen levels are ( Jakachi et al. 1982). required for differentiation of Wolffian structures in The use of highly sensitive two site-directed the male. In addition, FSH stimulates differentiation immunoradiometric, immunofluorometric and che- and development of seminiferous tubules. These miluminescent assays has revealed that between data are consistent with observations that patients 7 years of age and adulthood mean LH concentra- with IHH have normal differentiation of Wolffian tions increase over a 100-fold, mean FSH concentra- structures and external genitalia because the placen- tions sevenfold and estradiol levels 12-fold (Apter tal hCG drives the fetal testis to produce sufficient et al. 1989). The increase in FSH is gradual, but the androgen, even in the absence of pituitary LH and rise in LH is steep. The changes in FSH precede the FSH. However, because of FSH deficiency, these rise in LH concentrations. The mechanisms that patients have arrested or retarded development keep GnRH secretion in check during childhood of seminiferous tubules. Testicular descent is partly and trigger GnRH secretion at the onset of puberty dependent, during the later part of pregnancy, on remain unknown. androgen levels (Husmann 1991) which, during this period of fetal life, are maintained by pituitary LH. Changes in reproductive function during aging. There is agreement that serum testosterone levels decline Postnatal life and childhood years. After birth, serum progressively in men with advancing age (Fig. 21.6) LH and FSH levels rise again, albeit transiently (Pirke & Doerr 1973; Dai et al. 1981; Gray et al. 1991; (Winter et al. 1975; Sizonenko 1979). In the first Simon et al. 1992; Zmuda et al. 1997; Ferrini & 6 months of postnatal life, LH and FSH levels are Barrett-Connor 1998; Harman et al. 2001; Feldman measurable in blood (Winter et al. 1975). In fact, the et al. 2002; Matsumoto 2002); almost 25% of men pulsatile pattern of LH and FSH secretion is easily over the age of 70 have serum testosterone levels in discernible during this brief period of reactivation the hypogonadal range (Harman et al. 2001). As sex of the hypothalamic–pituitary axis (Winter et al. hormone binding globulin (SHBG) levels are higher 1976; Jakachi et al. 1982). Serum LH and FSH levels in older men than in younger men, the decrease in peak around 2–3 months of age and then decline to free and bioavailable testosterone with aging is undetectable levels by 9–12 months; serum testos- greater than the decline in total testosterone levels terone levels undergo similar changes. This brief (Pirke & Doerr 1973; Tenover et al. 1987; Ferrini period of postnatal life thus provides a window in & Barrett-Connor 1998; Harman et al. 2001). The which the normality of the hypothalamic–pituitary– diurnal rhythm of testosterone secretion, observed gonadal axis can be assessed before gonadotropin in younger men, is attenuated in older men. While and sex steroid levels fall back to the low range. mean serum levels of total, free and bioavailable During childhood years, the hypothalamic– testosterone fall in the later decades of life, many pituitary–gonadal unit is kept in abeyance until the elderly men retain serum testosterone in the normal onset of puberty (Lee, P.A. et al. 1976; Apter et al. male range. There is either no change or a modest 1978). However, the pituitary and the testis retain increase in circulating estradiol and estrone levels 290 chapter 21

20 0.5 ) (177)

(177) –1 ) 18 0.4

–1 (144) (151) 16 (151) (109) 0.3 (109) 14 (144) (43) 0.2 (158) (nmol·L (158) 12 Free T index Testosterone (nmol·nmol (43) 10 0.1 30 40 50 60 70 80 90 30 40 50 60 70 80 90 (a) Age (years) (b) Age (years)

Fig. 21.6 Age-related changes in serum total testosterone levels (a) and free testosterone (T) index (b) in men participating in the Baltimore longitudinal study of aging. (Reproduced from Harman et al. 2001.) with age due to the increased peripheral aromatiza- between LH and testosterone secretion than younger tion of androgen to estrogen (Pirke & Doerr 1973; men (Pincus et al. 1997). Therefore, aging is asso- Ferrini & Barrett-Connor 1998). ciated with abnormalities of the normal feedback The age-related changes in reproductive hor- control mechanisms that control the flow of informa- mones are compounded by the effects of conco- tion between components of the hypothalamic– mitant illness, changes in body composition and pituitary–testicular network, and a disruption of medications. Although the data on the relationship the orderly pattern of pulsatile hormonal secretion of age to serum androgen levels are mostly cross- (Pincus et al. 1997). sectional in nature, a few longitudinal studies (Zmuda et al. 1997; Harman et al. 2001; Feldman et al. Testosterone secretion, transport and 2002) have confirmed the aging-associated decline metabolism in serum testosterone levels. Some studies have been criticized for selecting elderly men who were Testosterone secretion. In males of most mammalian healthier than the general population. However, species, 95% of circulating testosterone is derived even after adjusting for illness, time of sampling, from testicular secretion. In man, 3–10 mg of testo- assay variability and medications, testosterone sterone is secreted daily by the testis (Horton 1978). levels are lower in older men than in younger men. Direct secretion of testosterone by adrenal, and peri- Aging-associated decline in testosterone levels pheral conversion of androstenedione, collectively occurs due to defects at all levels of the hypothalamic– account for another 500 µg of testosterone daily. pituitary–gonadal axis. Androgen secretion by the Only a small amount of dihydrotestosterone (approx- testis of elderly men is decreased due to primary imately 70 µg daily) is secreted by the human testis; abnormalities at the gonadal level. This is supported most of circulating DHT is derived from peripheral by their higher basal LH and FSH levels (Kaufman conversion of testosterone (Longcope & Fineberg & Vermeulen 1997; Morley et al. 1997; Feldman et al. 1985). 2002), decreased testosterone response to hCG Testosterone is produced in the testis by a het- and diminished Leydig cell mass in aging men erogeneous group of cells that includes the adult (Longcope 1973). In addition, secondary defects Leydig cells, Leydig cell precursors and immature may exist at the hypothalamic–pituitary level, as Leydig cells (Prince 2001). Studies in hypogona- indicated by the somewhat blunted LH and FSH dotropic (hpg) mice suggest that fetal development responses of older men to GnRH (Harman et al. of both Sertoli and Leydig cells is independent of 1982). The circulating levels of free α subunit are gonadotrophins; however, normal differentiation also higher in older men as compared to younger and proliferation of the adult Leydig cell population men (Harman et al. 1982). Older men secrete LH requires the presence of gonadotrophins. 46,XY more irregularly than younger men (Pincus et al. male humans with inactivating mutations of LH 1997). The older men also have less synchronicity receptor have varying degree of genital ambiguity regulation of testicular function 291

and Leydig cell agenesis, pointing to the important in SHBG and albumin are fertile and have normal role of LH in the regulation of Leydig cell develop- mating behavior. ment (Dufau 1988; Huhtaniemi & Toppari 1995). Sertoli cell number after birth is regulated by Testosterone metabolism. Testosterone is metabolized gonadotrophins. predominantly in the liver (50–70%) although some degradation also occurs in peripheral tissues, par- Androgen transport in the body. Ninety eight percent ticularly the prostate and the skin. Liver takes of circulating testosterone is bound to plasma pro- up testosterone from the blood and through a series teins: the SHBG and albumin (Vermeulen 1988; of chemical reactions, which involve 5-α- and 5-β- Rosner 1991). The SHBG binds testosterone with reductases, 3-α- and 3-β-hydroxysteroid dehydro- much greater affinity than albumin. Only 0.5–3.0% genases and 17-β-hydroxysteroid dehydrogenase, of testosterone is unbound. Although the preval- converts it into androsterone, etiocholanolone, both ent dogma assumes that only the unbound fraction inactive metabolites, and dihydrotestosterone and is biologically active, albumin-bound hormone 3-α-androstanediol. These compounds undergo glu- dissociates readily in the capillaries and may be coronidation or sulfation before their excretion bioavailable (Pardridge 1987). Pardridge (1987) has by the kidneys. Free and conjugated androsterone demonstrated that albumin- and SHBG-bound and etiocholanaolone are the predominant urinary androgens represent the major circulating pool of metabolites of testosterone. bioavailable hormone for testis or prostate. Further- more, these investigators have argued that SHBG- Testosterone as a prohormone: the role of DHT and estra- sex steroid complex may be nearly completely diol in mediating androgen action. Testosterone is con- available for influx through the blood–testis barrier verted in many peripheral tissues into its active or prostate plasma membrane; this view is not metabolites, estradiol 17-β and 5-α-DHT (Wilson universally shared (Pardridge 1987). et al. 1993; Grumbach & Auchus 1999). Aromatiza- SHBG is a glycoprotein, synthesized in the liver, tion of the A ring converts it into 17-β-estradiol. that possesses high affinity binding for testoster- In addition, reduction of the δ-4 double bond can one and estradiol (Vermeulen 1988; Rosner 1991). convert testosterone into 5-α-DHT. Testosterone Hepatic production of SHBG is regulated by insulin, actions in many tissues are mediated through these thyroid hormones, dietary factors and the balance metabolites. For instance, testosterone effects on between androgens and estrogens. SHBG is involved trabecular bone resorption, sexual differentiation in the transport of sex steroids in plasma, and its of the brain, plasma lipid concentrations, atheros- concentration is a major factor regulating their clerosis progression and some types of behaviors distribution between the protein-bound and free are mediated through its conversion to estrogen states. Plasma SHBG concentrations are decreased (Grumbach & Auchus 1999). The study of mice by androgen administration, obesity, hyperinsulin- that have mutantions in the estrogen receptor-α, ism and nephrotic syndrome (Rosner 1991). Con- estrogen receptor-β, or the aromatase gene has pro- versely, estrogen administration, hyperthyroidism, vided considerable insight into the role of estrogen many types of chronic inflammatory illnesses and in male mammals (Jones et al. 2000). These models of aging are associated with high SHBG concentra- estrogen deficiency exhibit significant disruption of tions. A locus that is associated with SHBG con- spermatogenesis and fertility, elevated testosterone centrations in African-Americans and Caucasians and LH levels, decreased bone mass and increased has been mapped to 1q44; several additional loci adiposity, indicating the important role of estrogens in African-Americans also exhibit linkage with in regulation of bone mass, gonadotropin regula- SHBG concentrations, suggesting that many genes tion, body composition and spermatogenesis (Smith regulate SHBG levels (Larrea et al. 1995). The et al. 1994; Carani et al. 1997). A small number of binding of testosterone to SHBG or albumin is not humans with inactivating mutations of the CYP19 essential for androgen action; rats that are deficient aromatase gene have been reported (Carani et al. 292 chapter 21

1997). Females with CYP19 gene mutations are mas- A large body of information about the role of culinized, fail to undergo pubertal development, DHT has emerged from the study of patients with have elevated levels of androgens and LH and FSH, autosomal recessive, steroid 5-α-reductase defi- polycystic ovaries and tall stature. Males with CYP19 ciency. 46,XY males with this syndrome contain aromatase mutations are characterized by osteo- normal male internal structures including testes, porosis, increased bone turnover, delayed epiphy- but exhibit ambiguous or female external genitalia seal fusion, tall stature and increased testosterone at birth (Cai et al. 1996; Mendonca et al. 1996). At but decreased estradiol levels (Carani et al. 1997). puberty, these individuals undergo partial viriliza- Two isoenzymes of steroid 5-α-reductase have tion and normal muscular development (Cai et al. been characterized (Wilson et al. 1993; Russell & 1993). Many, but not all, 46,XY individuals with this Wilson 1994). Type 1 steroid 5-α-reductase is disorder develop male gender identity, even if they expressed in many non-genital tissues, has been have been brought up as females. Their develop- mapped to chromosome region 5p15 and has a pH ment suggests that testosterone itself is able to optimum of 8. Type 2 steroid 5-α-reductase is stimulate psychosexual behavior, libido, develop- expressed in the prostate and other genital tissues, ment of the embryonic Wolffian duct, muscle devel- has been mapped to 2p23 and has a pH optimum opment, voice deepening, spermatogenesis and of 5.0 (Wilson et al. 1993). The biological role of axillary and pubic hair growth. In contrast, DHT is 5-α-reductase type 1 has not been ascertained fully. required for prostate development and growth, The gene-targeting experiments suggest that type the development of the external genitalia and male 1 enzyme plays a role in progesterone metabolism patterns of facial and body hair growth or male-pat- at the end of pregnancy (Wilson et al. 1993). The tern baldness. All 5-α-reductase-deficient kindreds, mice lacking 5-α-reductase enzyme type 1 have fail- studied to date, have been shown to have mutations ure of cervical ripening and fail to deliver (Wilson in steroid 5-α-reductase type 2, the predominant et al. 1993). form in the prostate (Cai et al. 1996; Mendonca et al. Testosterone effects on the prostate and seba- 1996). ceous glands of the skin require its 5-α reduction to DHT. DHT has been implicated in the pathophysio- Mechanism of androgen action. Most actions of testo- logy of benign prostatic hyperplasia and androgenic sterone and DHT are mediated through binding alopecia (Wilson et al. 1993). Type 2 isoenzyme is to an intracellular androgen receptor that acts as a the predominant form in the prostate, and has ligand-dependent transcription factor (Zhou et al. been implicated in the pathophysiology of benign 1994; Lee, D.K. & Chang 2003). Testosterone binds prostatic hypertrophy, hirsutism and possibly male- to the androgen receptor with half the affinity of pattern baldness. During embryonic life, testo- DHT, although the maximal binding capacity is sterone controls the differentiation of the Wolffian similar for both androgens. The DHT–androgen ducts into epididymis, vasa deferentia and seminal receptor complex has greater thermostability and vesicles. The development of structures from the a slower dissociation rate than the testosterone- urogenital sinus and the genital tubercle such as receptor complex. This may confer greater potency the scrotum, penis and penile urethra require the to DHT in mediating androgen effects in some action of DHT. Although testosterone and DHT androgen sensitive tissues, such as the prostate. both exercise anabolic effects on the muscle, steroid The androgen receptor has homology to other 5-α-reductase activity is very low or absent in the nuclear receptors, including the receptors for gluco- skeletal muscle, and we do not know whether 5-α corticoids, progesterone and mineralocorticoids reduction of testosterone to DHT is obligatory for (Zhou et al. 1994; Lee, D.K. & Chang 2003). The pre- mediating androgen effects on the muscle. Sim- dominant 919-amino acid, 110–114 kDa, androgen ilarly, the literature is unclear on whether androgen receptor protein has three conserved functional effects on sexual function in adult men are mediated domains: the steroid binding domain, the DNA through testosterone or DHT. binding domain and the transcriptional activational regulation of testicular function 293

domain; of these, the central, cysteine-rich DNA tein. An abnormal length of the polyglutamine tract binding domain is the most conserved. The single in exon 1 of the androgen receptor has been asso- copy androgen receptor gene spans a 90 kb region ciated with spinal and bulbar muscular atrophy, on the chromosomal region Xq11–12. In the absence also known as Kennedy’s disease. Although some of its ligand, the androgen-receptor protein is dis- studies have reported an association of polymor- tributed both in the nucleus and the cytoplasm. phisms in the lengths of polyglutamine and poly- However, androgen binding to the receptor causes glycine tracts with male infertility and risk of prostate it to translocate to the nucleus; amino acid sequences cancer, others have not confirmed these findings between 617–633 of the androgen receptor are (Casella et al. 2001). important for its nuclear migration and transac- tivation function. There is inconclusive evidence The biochemical pathways that link that some androgen effects may be mediated energy balance and reproductive axis through nongenomic mechanisms through mem- brane receptors. Humans have known since antiquity that energy The binding of androgens by the androgen recep- balance and nutritional status are intimately linked tor results in conformational changes in this protein; to the reproductive axis in both men and women. there is evidence that the binding of anti-androgens The onset of puberty, the length of the reproductive to the androgen receptor might induce a different period, the number of offspring and the age of set of conformational changes (Zhou et al. 1994; Lee, menopause have all been linked to body weight D.K. & Chang 2003). The androgen receptor can use and composition, particularly the amount of body two transactivation domains, AF1 and AF2, respect- fat (Frisch & McArthur 1974; Van Der Spruy 1985; ively. The transactivation domain AF1 (including Frisch 1989; Foster & Nagatani 1999). Normal repro- the so-called 1 and 5 regions) is located in the ductive function requires an optimal nutritional aminoterminal part of the receptor, whereas AF2 is intake; both caloric deprivation and consequent located in the carboxyterminal, ligand-dependent weight loss, and excessive food intake and obesity domain. In the intact receptor, both AF1 and AF2 are are associated with impairment of reproductive ligand-dependent and influenced by nuclear recep- function. The temporal aspects of sexual maturation tor coactivators. In contrast, in a truncated androgen are more closely associated with body growth than receptor, that is missing the ligand-binding domain, with chronological age (Penny et al. 1978; Frisch

AF1 becomes constitutively active. Hormone bind- 1989). In the animal kingdom, during periods of ing to the androgen receptor results in assembly of food scarcity, small animals with a short-life span tissue-specific co-activators and co-repressors that may not even achieve puberty before death (Foster determine the specificity of hormone action. & Nagatani 1999). In animals with longer life spans, The mutations of the androgen-receptor gene sexual maturation may be delayed during food have been associated with a wide spectrum of deprivation. Undernutrition, caused by famine, phenotypic abnormalities (Brinkmann 2001). Pati- eating disorders and exercise, results in weight loss ents with complete androgen insensitivity present and changes in body composition and endocrine with male pseudohermaphroditism, characterized milieu that can impair reproductive function by female external genitalia, a blind vaginal pouch (Penny et al. 1978; Bates et al. 1982; Rock et al. 1996). and well-developed breasts. Other patients with As a general rule, weight loss and body composition androgen-receptor mutations may have a male phe- changes resulting from undernutrition are asso- notype and milder abnormalities such as hypospa- ciated with reduced gonadotropin secretion; the dias, gynecomastia and infertility (McPhaul et al. decrease in FSH and LH levels correlates with the 1993). degree of weight loss (Fig. 21.7) (Penny et al. 1978). The lengths of the CAG and GCG repeats in exon However, both hypogonadotropic and hyper- 1 of the androgen-receptor gene have been linked to gonadotropic hypogonadism have been described transcriptional activity of androgen-receptor pro- in cachexia associated with chronic illnesses, such as 294 chapter 21

Starvation Exercise Illness Eating Exercise Stress disorders energy drain

Body weight and Fig. 21.7 Relationship between composition energy balance, exercise, illness and gonadotropin secretion. (a) Endocrine signals that emanate from adipocytes and are integrated in the Impaired hypothalamus and other central GnRH sensors regulate hypothalamic secretion gonadotropin-releasing hormone (GnRH) secretion. In addition, products and mediators of systemic inflammatory and stress responses during illness and exposure to Impaired reproductive function: physical, chemical or psychological decreased sex steroid secretion; stress affect the male reproductive impaired fertility; delayed puberty; axis at multiple levels, resulting in menstrual irregularities (a) impairment of sex steroid secretion, pubertal development and fertility. (b) Hypothetical depiction of the biochemical pathways that link Central control Central control energy balance and reproductive axis. The prevalent hypothesis is that of reproduction of energy balance Stress Illness metabolic signals that link energy − + − + + stores to hypothalamic GnRH NPY GALP NPY Leptin POMC CRH secretion originate in the adipocytes Leptin receptor − and are communicated to the + hypothalamic GnRH secreting GnRH +− + neurons through leptin, GALP and Food Energy other poorly understood chemical intake expenditure signals. Leptin, the product of the obesity gene (ob) inhibits neuropeptide Y (NPY) which has a tonic inhibitory effect on GnRH secretion. Leptin also has direct stimulatory effect on luteinizing hormone (LH) secretion by stimulating nitric oxide (NO) production in the gonadotropes. + Therefore, the net effect of leptin + Pituitary action is stimulation of LH secretion. Leptin NO Leptin decreases food intake while NOS other mediators of stress response, including products of the POMC Adipocytes gene and corticotropin-releasing energy hormone, inhibit food intake. FSH, stores LH & FSH follicle-stimulating hormone; NOS, nitric oxide synthase. (Reproduced (b) from Bross et al. 2000.) regulation of testicular function 295

human immunodeficiency virus (HIV) infection decrease in leptin secretion. While leptin is an (Arver et al. 1999). Collectively, these observations important metabolic signal that links energy bal- provide compelling evidence that energy balance is ance and the reproductive axis, it remains unclear an important determinant of reproductive function whether it is the primary trigger for the activation of in all mammals. the GnRH-pulse generator at the onset of puberty. We do not know the precise nature of the bio- Emerging evidence suggests that leptin is essential chemical pathways that connect energy metabolism but not sufficient for the initiation of puberty. and the reproductive axis, two biologic systems essential for the survival of all species. The preval- Historical and experimental illustrations ent hypothesis is that the metabolic signals, that re- of the link between nutritional status, gulate hypothalamic GnRH secretion, are mediated reproduction and fertility through leptin and neuropeptide Y (Aubert et al. 1998; Clarke & Henry 1999; Cunningham et al. 1999; Foster & Nagatani 1999). Leptin, the product of the The Dutch Hunger Winter. Between October 1944 and obesity (ob) gene, is a circulating hormone secreted May 1945, during the course of the Second World by the fat cells that acts centrally to regulate the War, German army restricted food supplies in activity of central nervous system effector systems certain Dutch cities (Fig. 21.8). This resulted in sub- that maintain energy balance (Schwartz et al. 1999). stantial reduction in average daily energy intake to Leptin stimulates LH secretion by activation of the less than 1000 kcal (Stein et al. 1973) in cities affected nitric oxide synthase in the gonadotropes (McCann by the German siege. In some adjacent cities, food et al. 1998), and inhibits neuropeptide Y secretion. supplies were not curtailed by the Germans (con- Neuropeptide Y has a tonic inhibitory effect on both trol cities). Studies by Stein et al. (1973) revealed leptin and GnRH secretion. Leptin also stimulates that 50% of women, affected by famine, developed nitric oxide (NO) production in the mediobasal amenorrhea, the conception rate dropped and there hypothalamus; NO stimulates GnRH secretion by was increased perinatal mortality, congenital mal- the hypothalamic GnRH-secreting neurons (McCann formations and schizophrenia. Thus, optimal caloric et al. 1998). More recent evidence suggests that intake is essential for normal fertility and prenatal additional pathways including those that involve a growth. galanin-like peptide (GALP) -link energy homeo- stasis, food intake and GnRH secretion in the hypo- The association of weight change and fertility in the thalamus (see Fig. 21.7) (Seth et al. 2004). The net !Kung San of Botswana. The !Kung San of Botswana effect of leptin action is stimulation of hypothalamic were a tribe of hunter–gatherers (Fig. 21.9) (Van Der GnRH secretion (Schwartz et al. 1999). Walt et al. 1978). The body weight of men and Caloric deprivation in mammals is associated women in the tribe varied throughout the year with reduced leptin levels and a concomitant reduc- depending upon food availability. In summer tion in LH levels (Schwartz et al. 1999). Leptin months, when food supply was abundant, body administration to calorically-deprived mice reverses weight increased, while the nadir of body weight the inhibition of gonadotropin secretion that attends was achieved in winter months. The number of food-restriction (Schwartz et al. 1999). Similarly, births in the tribe peaked about 9 months after the genetically ob/ob mice with leptin deficiency have peak of body weight (Van Der Walt et al. 1978). This hypogonadotropic hypogonadism and are infertile; is another example of how the availability of food treatment of these mice with leptin restores gona- regulates fertility patterns in nature. dotropin secretion and fertility (Mohamed-Ali et al. 1998; Schwartz et al. 1999). Thus, energy deficit and The Minnesota caloric deprivation experiment. In the weight loss are associated with impaired GnRH late 1940s, Ancel Keys and coworkers studied secretion; in part, because of changes in neuro- human starvation in an experiment in which 32 peptide Y and GALP activity and the consequent young men volunteered to live on the campus of 296 chapter 21

Famine city above 500 000 population Famine city 100 000–500 000 population Famine city 40 000–100 000 population Control city 100 000–500 000 population Control city 40 000–100 000 population

Groningen

Leeuwarden North Sea

Swolle Hearlem Enachade Amsterdam Mangolo The Hague Leiden Utrecht Rotterdam Mass R Mass R

Mass R Breda Tilburg

Eindhoven

Maastricht Hearlen 0km100

(a)

Born during Conceived famine during famine Oct '44-May '48 July '46-Feb '48 4000 8000 Total

births Births 3500 7000 Fig. 21.8 The Dutch Hunger Winter 3000 6000 during the German siege. During the German siege of the Netherlands 2500 5000 during the Second World War, residents of many Dutch cities 2000 4000 experienced severe curtailment Calorie of caloric intake (famine cities). Average daily caloric intake Average ration at 1500 3000 Adjacent cities that did not face conception curtailment of food supplies were 1000 2000 used as control cities (a). (b) The relationship between caloric 0 intake and the number of births. (b) J FMAM J J A S OND J FMAM J J A S OND J FMAM J J A S OND (Reproduced from Stein et al. 1973.) regulation of testicular function 297

production was reduced when men weighed ~ 25% 50 less than the normal weight for their height. Weight gain restored reproductive function.

45 106 Reproductive dysfunction in acute and chronic illnesses 40 104 There is a high prevalence of androgen deficiency Body weight in Lbs (…) 35 defined solely in terms of low testosterone levels 102 in men with chronic illness. Thus, even after the advent of potent anti-retroviral therapy, androgen 30 100 deficiency continues to be a common complication 98 of HIV infection in men. In an earlier survey of 25 150 HIV-infected men who attended our HIV clinic

Number of births ( —) Number of births ( 96 in 1997, approximately one third had serum total 20 and free testosterone levels in the hypogonadal 94 range (Arver et al. 1999). Other investigators have reported a similar prevalence of hypogonadism in 15 HIV-infected men (Dobs et al. 1996; Grinspoon et al. 1996). A recent survey of HIV-infected men found 10 the prevalence of low testosterone levels to be 20% ND J FMAM JJASON (Reitschel et al. 2000). Thus, androgen deficiency Month continues to be a common occurrence in HIV- infected men. Fig. 21.9 Relationship between feeding pattern and In our survey, 20% of HIV-infected men with low fertility in !Kung San of Botswana. The relationship between weight change and fertility in !Kung San of testosterone levels had elevated LH and FSH levels, Botswana. The !Kung San of Botswana were a tribe of and thus had hypergonadotropic hypogonadism hunter–gatherers until about 30 years ago. The body (Arver et al. 1999). These patients presumably had weight of the men and women in this tribe varied greatly primary testicular dysfunction. The remaining 80% throughout the year depending on the availability of had either normal or low LH and FSH levels. The food supplies. In the summer months, the food supply was more abundant and the body weight increased; men with hypogonadotropic hypogonadism either conversely, body weight decreased throughout the winter had a central defect at the hypothalamic or pituitary months. The number of births in the tribe peaked 9 site or a dual defect involving both the testis and the months after the achievement of peak body weight. These hypothalamic–pituitary centers. The pathophysio- anthropological data illustrate how availability of food logy of hypogonadism in HIV infection is com- can regulate fertility patterns. (Reproduced from Van Der Walt 1978.) plex and involves defects at multiple levels of the hypothalamic–pituitary–testicular axis. In a recent study, a majority of men with chronic the University of Minnesota and consume a diet obstructive lung disease had low total and free providing approximately 1600 kcal·day–1 (6.688 testosterone levels (Casaburi et al. 1996). Similarly, MJ·day−1), about two thirds of their normal energy there is a high frequency of hypogonadism in pati- requirement (Keys et al. 1950). The volunteers ents with cancer, end-stage renal disease on hemo- lost an average of 23% of their initial body weight; dialysis and liver disease (Handelsman & Dong more than 70% of body fat and 24% of lean tissue. 1993; Singh et al. 2001). Previous reports suggest Decreased caloric intake and subsequent weight that two thirds of men with end-stage renal disease loss first caused a loss of libido and a reduction in have low total and free testosterone concentrations prostate fluid, sperm motility and longevity. Sperm (Handelsman & Dong 1993). In a recent study, of the 298 chapter 21

39 men with end-stage renal disease on hemodia- Similarly, muscle mass, strength and perform- lysis who did not have diabetes mellitus, 24 (63%) ance and physical function are markedly impaired had serum total and free testosterone levels below in patients on hemodialysis (Johansen et al. 2001). the lower limit of normal male range (Singh et al. Exercise tolerance is attenuated (Kopple 1999; 2001). There is a high prevalence of sexual dysfunc- Johansen et al. 2001); peak oxygen uptake is typ- tion and spermatogenic abnormalities in men on ically reduced to roughly half the level predicted hemodialysis (Handelsman & Dong 1993). Muscle for healthy subjects. Although the etiology of sar- mass is decreased and fat mass increased, and copenia in end-stage renal disease is complex, the muscle performance and physical function are decrease in testosterone levels, a contributor to loss markedly impaired in men receiving hemodialysis of muscle mass and dysfunction, is potentially (Kopple 1999; Johansen et al. 2001). While androgen reversible. deficiency might contribute to the complex patho- physiology of sexual dysfunction and sarcopenia Changes in testosterone levels in men on hemodialysis, we do not know if any of during exercise these physiologic derangements can be reversed by androgen replacement. The literature on the effects of exercise on testicu- The pathophysiology of hypogonadism in chronic lar function is controversial in part because the illness is multifactorial; defects exist at all levels of published studies differed significantly in the type, the hypothalamic–pituitary–testicular axis (Bross mode, intensity and duration of exercise. Further- et al. 1998). Malnutrition, mediators and products more, few studies controlled for the confounding of the systemic inflammatory response, drugs such influence of nutritional intake and physical activity as ketoconazole and metabolic abnormalities pro- level. Therefore, it is not surprising that both duced by the systemic illness all contribute to a increases and decreases in circulating testosterone decline in testosterone production. levels have been reported in men undergoing exer- cise training. This is in contrast to women under- Low testosterone levels correlate with poor disease out- going heavy endurance exercise training in whom come. Low testosterone levels correlate with adverse consistent disruptions of menstrual cycle and disease outcome in HIV-infected men. Serum tes- ovarian estrogen production have been observed tosterone levels are lower in HIV-infected men (Frisch & Revelle 1970; Mahna et al. 1973; Warren who have lost weight than in those who have not 1980; Baker et al. 1981; Veldhuis et al. 1985; Loucks (Coodley et al. 1994). Longitudinal follow up of et al. 1989; Cumming 1996). The onset of menarche HIV-infected homosexual men reveals a progressive is often delayed in ballet dancers and in girls who decrease in serum testosterone levels; this decrease are engaged in heavy exercise (Frisch et al. 1980). is much greater in HIV-infected men who progress Similarly, a high prevalence of menstrual irregular- to acquired immunodeficiency syndrome (AIDS) ities, attenuated gonadotropin secretion and high than in those who do not. Serum testosterone levels cortisol levels has been documented in female run- in HIV-infected men decline early in the course of ners (Baker et al. 1981; Villanueva et al. 1986); how- events that culminate in wasting (Dobs et al. 1996). ever, the literature on the literature on the effects of Testosterone levels correlate with muscle mass and exercise on reproductive function in men is sparse exercise capacity in HIV-infected men (Grinspoon and less lucid (MacConnie et al. 1986; Skarda & et al. 1996). Although, patients with HIV infection Burge 1998). may lose both fat and lean tissue, the loss of lean Broadly, exercise-training regimens can be body mass is an important aspect of the weight loss classified into endurance and resistance exercise associated with wasting. There is a high prevalence regimens. Although serum testosterone levels may of sexual dysfunction in HIV-infected men. With increase in anticipation of agonistic events or an the increasing life expectancy of HIV-infected men, endurance exercise, most studies are in agreement frailty and sexual dysfunction have emerged as that mild endurance exercise has little or no clin- important quality of life issues. ically significant effect on circulating testosterone regulation of testicular function 299

levels (Skarda & Burge 1998; Smilios et al. 2003). In studies suggest that only nutritionally deleterious contrast, very severe endurance exercise training as eating behaviors are associated with lower bone exemplified by long distance running, especially mineral density and increased fracture risk. Fur- if accompanied by significant energy drain and thermore, some of the deleterious effects of lower weight loss, is associated with the lowering of serum sex hormone levels on bone mineral density may be testosterone levels (Remes et al. 1979; Häkkinen et al. offset by the direct beneficial effects of exercise and 1985; MacConnie et al. 1986; Dressendorfer & Wade physical activity on bone mineral density. 1991; Skarda & Burge 1998). For instance, male dis- Frost (1992) has hypothesized that the effect of tance runners who ran on average 92 km weekly exercise on bone mineral density depends crucially were found to have 10% lower bone mineral density on the force applied to the limbs during exercise. at the lumbar spine than a control group of men who He has speculated that only exercise regimens did not run (Frost 1992). In general, the distance that apply forces on the limbs in excess of a certain run correlates inversely with vertebral and femoral threshold level are capable of eliciting a bone bone mineral densities; the longer the distance run, remodeling response, whereas long distance run- the greater the energy drain, the lower the bone ning that exerts forces of a lower magnitude (5– mineral density. In one report of men participating 10 times the body weight) on the limbs does not in a 15-day 400-km race, serum testosterone levels increase bone mineral density. According to this declined significantly while cortisol levels increased hypothesis, the magnitude of bone loading is more (Dressendorfer & Wade 1991). Therefore, it is not important than the exercise mode or number of surprising that some of the lowest testosterone cycles. Thus, lean marathon runners or boot camp levels have been reported in army recruits who are trainees, with significant energy deficits, would be in intense boot camp training. expected to have lower testosterone levels, no coun- Remarkably, men who run 15–20 miles weekly terbalancing beneficial effects of bone loading, and have higher bone mineral density than age-matched consequently lower bone mineral density. controls who do not run (Frost 1992; Burrows et al. Previous studies have reported either no change 2003). Similarly, rowers have been reported to have or modest increments in serum total testosterone a higher bone mineral density than sedentary con- concentrations during regimens of resistance exer- trols. In another study, triathletes did not differ from cise training (Remes et al. 1979; Truls et al. 2000; sedentary controls in their bone mineral density. Ahtiainen et al. 2003). The small changes in testo- Thus, mild to moderate endurance exercise training sterone concentrations reported in some studies may have beneficial effects on bone mineral density were not sustained after completion of the exercise; with little or no effect on testosterone concentrations; in fact some studies have reported a significant very intense endurance exercise training lowers testo- decline in free testosterone levels during recovery sterone concentrations and decreases bone mineral from exercise. SHBG concentrations generally did density (Heinonen et al. 1995; Burrows et al. 2003). not change during the course of resistance exercise. There is agreement that long distance runners Some studies have reported increases in the testo- may have significant energy deficit and nutritional sterone to cortisol ratio during progressive training deficiencies that may affect bone mineral density for maximal strength gains. There is considerable independent of the effects of exercise on testo- interindividual variability in hormonal response to sterone levels (Burrows et al. 2003). Indeed, some resistance exercise training.

References

Ahtiainen, J.P., Pakarinen, A., Kraemer, Journal of Sports Medicine 24(6), Apter, D., Cacciatore, B., Alfthan, H. & W.J. & Häkkinen, K. (2003) Acute 410–418. Stenmain, U.-H. (1989) Serum LH hormonal and neuromuscular Apter, D., Pakarmen, A. & Vihko, R. (1978) concentrations increase 100-fold in responses and recovery to forced vs. Serum prolactin, FSH and LH during females from 7 years of age to maximum repetitions multiple puberty in girls and boys. Acta adulthood, as measured by time- resistance exercises. International Paediatrica Scandinavica 67, 417–423. resolved immunofluorometric assay. 300 chapter 21

Journal of Clinical Endocrinology and Brinkmann, A.O. (2001) Molecular basis Clarke, I.J. & Cummins, J.T. (1999) Metabolism 68, 53–57. of androgen insensitivity. Molecular and Direct pituitary effects of estrogen Arver, S.A., Sinha-Hikim, I., Beall, G. Cellular Endocrinology 179(1–2), 105–109. and progesterone on gonadotropin et al. (1999) Dihdrotestosterone and Bross, R., Casaburi, R., Storer, T.W. & secretion in the ovariectomized ewe. testosterone levels in HIV-infected men. Bhasin, S. (1998) Androgen effects on Neuroendocrinology 39, 267–274. Journal of Andrology 20, 611–618. body composition and muscle function: Clarke, I.J. & Henry, B.A. (1999) Leptin and Aubert, M.L., Pierroz, D.D., Gruaz, N.M. implications for the use of androgens as reproduction. Biology of Reproduction 4, et al. (1998) Metabolic control of anabolic agents in sarcopenic states. 48–55. sexual function and growth: role of Bailliere’s Clinical Endocrinology and Clements, J.A., Reves, F.L., Winter, J.S.D. neuropeptide Y and leptin. Molecular Metabolism 12(3), 365–378. & Faiman, C. (1980) Ontogenesis of and Cellular Endocrinology 140, 107–113. Bross, R., Singh, A.B. & Bhasin, S. (2000) GnRH in the human fetal hypothalamus. Badger, T.M., Edcox, C.E., Meyer, E.R., In: Atlas of Clinical Endocrinology: Human Proceedings of the Society for Experimental Bell, R.D. & Cicero, T.J. (1978) Nutrition and Obesity (Heber, D., ed.). Biology and Medicine 163, 437–444. Simultaneous changes in tissue and Blackwell Science, Philadelphia, PA. Cole, L.A., Perini, F., Birken, S. & Ruddon, serum levels of LH, FSH and LH/FSH Burger, H.G. & Igarashi, M. (1988) Inhibin: R. (1984) An oligosacchande of the O- releasing factor after castration in the definition and nomenclature, including linked type distinguishes the free from male rat. Endocrinology 102, 136–141. related substances. Journal of Clinical the combined form of hCG α subunit. Baenziger, J.U. (1990) The asparagine- Endocrinology and Metabolism 66, Biochemical and Biophysical Research linked ohgosaccharides of the 885–886. Communications 122, 1260–1267. glycoprotein hormones. In: Glycoprotein Burrows, M., Nevill, A.M., Bird, S. & Conn, P.M., Chafouleas, J.C., Rogers, D.C. Hormones (Chin, W.W. & Boime, L., Simpson, D. (2003) Physiological factors & Means, A.R. (1981) Gonadotropin eds.). Serono Symposia, Norwell, MA: associated with low bone mineral releasing hormone stimulates 11–18. density in female endurance runners. calmodulin redistribution in rat Baker, E.R., Mathur, R.S., Kirk, R.S. & British Journal of Sports Medicine 37(1), pituitary. Nature 292, 264–265. Williamson, H.O. (1981) Female runners 67–71. Conn, P.M., Rogers, D.C. & McNeil, R. and secondary amenorrhea: correlation Cai, L.Q., Zhu, Y.S., Katz, M.D. et al. (1996) (1982) Potency enhancement of with age, parity, mileage, and plasma 5α-reductase-2 gene mutations in the gonadotropin-releasing hormone hormonal and sex-hormone building Dominican Republic. Journal of Clinical agonist GnRH-mediated concentrations. Fertility and Sterility 36, Endocrinology and Metabolism 81(5), microaggregation stimulates 183–189. 1730–1735. gonadotropin release. Endocrinology 111, Bardin, C.W., Ross, G.T., Rifkind, A.B., Carani, C., Qin, K., Simoni, M. et al. (1997) 335–337. Cargille, C.M. & Lipsett, M.B. (1969) Effect of testosterone and estradiol in a Coodley, G.O., Loveless, M.O., Nelson, Studies of the pituitary Leydig cell axis man with aromatase deficiency. New H.D. & Coodley, M.K. (1994). Endocrine in young men with hypogonadotropic England Journal of Medicine 337(2), 91–95. function in HIV wasting syndrome. hypogonadism and hyposmia. Casaburi, R., Goren, S. & Bhasin, S. (1996) Journal of Acquired Immune Deficiency Comparison with normal men, Substantial prevalence of low anabolic Syndromes and Human Retrovirology 7, prepubertal boys and hypopituitary hormone levels in COPD patients 46–51. patients. Journal of Clinical Investigation undergoing rehabilitation. American Craig, B.W., Brown, R. & Everhart, J. (1989) 48, 2046–2056. Journal of Respiratory and Critical Care Effects of progressive resistance training Bates, G.W., Bates, S.R. & Whitworth, N.S. Medicine 153, A128. on growth hormone and testosterone (1982) Reproductive failure in women Casella, R., Maduro, M.R., Lipshultz, L.I. levels in young and elderly subjects. who practice weight control. Fertility & Lamb, D.J. (2001) Significance of the Mechanisms of Ageing and Development and Sterility 37, 373–378. polyglutamine tract polymorphism in 49(2), 159–169. Belchetz, P.E., Plant, T.M., Naka, Y., the androgen receptor. Urology 58(5), Crowley, W.F., Jr., Whitcomb, R.W., Keogh, E.J. & Knobil, E. (1978) 651–656. Jameson, J.L. et al. (1991) The Hypophyseal responses to continuous Childs, G.V., Ellison, D.G., Foster, L. neuroendocrine control of human and intermittent delivery of & Ramaley, J.A. (1981) Postnatal reproduction in the male. Recent Progress hypothalamic GnRH. Science 202, maturation of gonadotropes in the male in Hormone Research 47, 27–62. 631–643. rat pituitary. Endocrinology 109, Culler, M.D. & Negro-Vilar, A. (1990) Bhasin, S., Fielder, T.J. & Swerdloff, R.S. 1683–1693. Destruction of testicular Leydig cells (1987) Testosterone selectively increases Childs, G.V., Llovd, J.M., Unabia, G. et al. reveals a role of endogenous inhibin in serum FSH but not LH in GnRH (1987) Detection of LHβ mRNA in regulating follicle-stimulating hormone antagonist treated male rats: evidence individual gonadotropes after secretion in the adult male rat. Molecular for differential regulation of LH and castration: use of a new in situ and Cellular Endocrinology 70, 89–98. FSH secretion. Biology of Reproduction 37, hybridization method with a photo Cumming, D. (1996) Exercise-associated 55–59. biotinylated complementary RNA amenorrhea, low bone mineral density Boccabella, A.V. (1963) Reinitiation probe. Molecular Endocrinology and oestradiol replacement therapy. and restoration of spermatogenesis (Baltimore, Md) 1, 926–932. Archives of Internal Medicine 156, with testosterone propionate and Clark, B.J. & Stocco, D.M. (1982) STAR. 2193–2195. other hormones after a long term A tissue specific acute mediator of Cunningham, M.J., Clifton, D.K. & Steiner, posthypophysectomy regression period. steroidogenesis. Trends in Endocrinology R.A. (1999) Leptin’s actions on the Endocrinology 72, 787–798. and Metabolism 91, R1–R6. reproductive axis: perspectives and regulation of testicular function 301

mechanisms. Biology of Reproduction 60, human male I. Effects of testosterone in hypogondal men with acquired 216–222. administration in normal and GnRH immunodeficiency syndrome and Dai, W.S., Kuller, L.H., LaPorte, R.E. et al. deficient men. Journal of Clinical wasting. Journal of Clinical Endocrinology (1981) The epidemiology of plasma Endocrinology and Metabolism 73, and Metabolism 81, 4051–4058. testosterone levels in middle-aged men. 609–620. Grossman, M., Weintraub, B.D. & American Journal of Epidemiology 114, Foster, D.L. & Nagatani, S. (1999) Szkudlinski, M.W. (1997) Novel insights 804–816. Physiological perspectives on leptin as a into the molecular mechanisms of Dobs, A.S., Few, W.L., Blackman, M.R. regulator of reproduction: role in timing human thyrotropin action: structural, et al. (1996) Serum hormones in men puberty. Biology of Reproduction 60, physiological, and therapeutic with human immunodeficiency virus- 205–215. implications for the glycoprotein associated wasting. Journal of Clinical Frisch, R.E. (1989) Body weight and hormone family. Endocrine Reviews 18, Endocrinology and Metabolism 81, reproduction. Science 246, 432–436. 476–501. 4108–4112. Frisch, R.E. & McArthur, J.W. (1974) Grumbach, M.M. & Auchus, R.J. (1999) Dressendorfer, R.H. & Wade, C.E. (1991) Menstrual cycles: fatness as a Estrogen: consequences and Effects of a 15-d race on plasma steroid determinant of minimum weight for implications of human mutations in levels and leg muscle fitness in runners. height necessary for their maintenance synthesis and action. Journal of Clinical Medicine and Science in Sports and Exercise or onset. Science 185, 949–995. Endocrinology and Metabolism 84(12), 23(8), 954–958. Frisch, R.E. & Revelle, R.L. (1970) Height 4677–4694. Drinkwater, B.L., Nilson, K., Chestnut, and weight at menarche and a Häkkinen, K., Pakarinen, A., Alen, M. C.H. et al. (1984) Bone mineral content of hypothesis of critical body weights and & Komi, P.V. (1985) Serum hormones amenorrheic and eumenorrheic women. adolescent events. Science 169, 397–398. during prolonged training of New England Journal of Medicine 311, Frisch, R.E., Wushak, G. & Vincent, L. neuromuscular performance. European 2777–2782. (1980) Delayed menarche and Journal of Applied Physiology and Dufau, M.L. (1988) Endocrine regulation amenorrhea of ballet dancers. New Occupational Physiology 53(4), 287–293. and communicating functions of the England Journal of Medicine 303, 17–19. Haisenleder, D.J., Katt, J.A., Ortolano, G.A. Leydig cell. Annual Review of Physiology Frost, H.M. (1992) The role of changes in et al. (1988) Influence of GnRH pulse 50, 483–508. mechanical usage set points in the amplitude, frequency and treatment Feldman, H.A., Longcope, C., Derby, C.A. pathogenesis of osteoporosis. Journal of duration on the regulation of LH subunit et al. (2002) Age trends in the level of Bone and Mineral Research 7, 253–261. mRNAs and LH secretion. Molecular serum testosterone and other hormones Gharib, S.D., Bowers, S.M., Need, L.R. & Endocrinology (Baltimore, Md) 12, in middle-aged men: longitudinal Chin, W.W. (1986) Regulation of rat LH 338–343. results from the Massachusetts male subunit mRNAs by gonadal steroid Haisenleder, D.J., Dalkin, A.C., Ortolano, aging study. Journal of Clinical hormones. Journal of Clinical Investigation G.X., Marshall, J.C. & Shupnik, M.A. Endocrinology and Metabolism 87, 77, 582–589. (1991) A pulsatile GnRH stimulus is 589–598. Gharib, S.D., Wierman, M.E., Badger, T.M. required to increase transcription of Ferrini, R.L. & Barrett-Connor, E. (1998) & Chin, W.W. (1987) Sex steroid the gonadotropin subunit genes: Sex hormones and age: a cross-sectional hormone regulation of folicle- evidence for differential regulation of study of testosterone and estradiol and stimulating hormone subunit messenger transcription by pulse frequency in vivo. their bioavailable fractions in ribonucleic acid (mRNA) by gonadal Endocrinology 128, 509–517. community-dwelling men. American steroid hormones. Journal of Clinical Hall, S.H. & Miller, W.L. (1986) Regulation Journal of Epidemiology 147, 750–754. Investigation 80, 294–299. of ovine pituitary glycoprotein hormone Fiddes, J.C. & Talmadge, K. (1984) Gharib, S.D., Wierman, M.E., Shupnik, α subunit mRNA by 17-β-estradiol in Structure, expression and evolution of M.A. & Chin, W.W. (1990) Molecular cell culture. Biology of Reproduction 34, the genes for the human glycoprotein biology of the pituitary gonadotropins. 533–542. hormones. Recent Progress in Hormone Endocrine Reviews 11, 177–199. Handelsman, D.J. & Dong, Q. (1993) Research 40, 43–78. Glass, A.R. & Denster, P.A. (1987) Hypothalamic–pituitary–gonadal axis in Finkel, D.M., Phillips, J.L. & Snyder, P.J. Amenorrhea in Olympic marathon chronic renal failure. Endocrinology and (1985) Stimulation of spermatogenesis runners. Fertility and Sterility 4, 740–745. Metabolism Clinics of North America 22, by gonadotropins in men with Gormley, G. & Rittmaster, R. (1992) Effect 145–161. hypogonadotropic hypogonadism. New of finasteride, a 5-α reductase inhibitor, Harman, S.M., Tsitouras, P.D., Cost, P.T. England Journal of Medicine 313, 651–655. on serum gonadotropins in normal men. & Blackman, M.R. (1982) Reproductive Finkelstein, J.S., O’Dea, St. L, Whitcomb, Journal of Clinical Endocrinology and hormones in aging men II. Basal R.W. & Crowley, W.E., Jr. (1991a) Sex Metabolism 75, 484–488. pituitary gonadotropins and steroid control of gondotropin secretion Gray, A., Berlin, J.A., McKinlay, J.B. & gonadotropin responses to LHRH. in the human male II. Effects of estradiol Longcope, C. (1991) An examination of Journal of Clinical Endocrinology and administration in normal and GnRH research design effects on the association Metabolism 54, 547–551. deficient men. Journal of Clinical of testosterone and male aging: results Harman, S.M., Metter, E.J., Tobin, J.D., Endocrinology and Metabolism 73, of a meta-analysis. Journal of Clinical Pearson, J. & Blackman, M.R. (2001) 621–628. Epidemiology 44, 671–684. Longitudinal effects of aging on serum Finkelstein, J.S., Whitcomb, R.W., O’Dea, Grinspoon, S., Corcoran, C., Lee, K. et al. total and free testosterone levels in St. L. et al. (1991b) Sex steroid control (1996) Loss of lean body and muscle healthy men. Baltimore Longitudinal of gonadotropin secretion in the mass correlates with androgen levels Study of Aging. Journal of Clinical 302 chapter 21

Endocrinology and Metabolism 86, Knobil, E. (1980) The neuroendocrine Ling, N., Ying, S.-Y., Ueno, N. et al. (1985) 724–731. control of the menstrual cycle. Recent Isolation and partial characterization of Heinonen, A., Oja, P., Kannus, P. et al. Progress in Hormone Research 36, a Mr 32000 protein with inhibin activity (1995) Bone mineral density in female 53–88. from porcine follicular fluid. Proceedings athletes representing sports with Kopple, J.D. (1999) Pathophysiology of of the National Academy of Sciences of the different loading characteristics of the protein-energy wasting in chronic United States of America 182, 7217–7221. skeleton. Bone 17, 197–203. renal failure. Journal of Nutrition Longcope, C. (1973) The effect of human Horton, R. (1978) Sex steroid production 129(1S suppl.), 247S–251S. chorionic gonadotropin on plasma and secretion in the male. Andrologia Kornfeld, R. & Kornfeld, S. (1976) steroid levels in young and old men. 10(3), 183–194. Comparative aspects of glycoprotein Steroids 21, 583–592. Huhtaniemi, I. & Toppari, J. (1995) structure. Annual Review of Biochemisty Longcope, C. & Fineberg, S.E. (1985) Endocrine, paracrine and autocrine 45, 217–237. Production and metabolism of regulation of testicular steroidogenesis. Kovacs, K. & Horvath, E. (1985) dihydrotestosterone in peripheral Advances in Experimental Medicine and Morphology of adenohyphyseal cells tissues. Journal of Steroid Biochemistry Biology 377, 33–54. and, pituitary adenomas. In: The 23(4), 415–419. Husmann, D.A. (1991) Inhibition of Pituitary Gland (Martini, L., eds.). Loucks, A.B., Mortola, J.F., Girton, L. & testicular descent in the rat by anti Raven Press, New York: 25–56. Yen, S.S.C. (1989) Alterations in the androgens. Endocrinology 73, 1162A. Krummen, L.A., Toppari, J., Kim, W.H. hypothalamic–pituitary–ovarian and the Ingraham, H.A., Chen, R.P., Mangalam, et al. (1989) Regulation of testicular hypothalamic–pituitary–adrenal axis H.J. et al. (1988) A tissue-specific inhibin subunit messenger ribonucleic in athletic women. Journal of Clinical transcription factor containing a acid levels in vivo; effects of Endocrinology and Metabolism 68, homeodomain specifies a pituitary hypophysectomy and selective follicle- 402–411. phenotype. Cell 55, 519–529. stimulating hormone replacement. McCann, S.M., Kimura, M., Walczewska, Jakachi, R.I., Kelch, R.P., Sander, S.E. et al. Endocrinology 125, 1630–1637. A. et al. (1998) Hypothalamic control of (1982) Pulsatile release of LH in children. Krummen, L.A., Morelos, B.S. & Bhasin, S. FSH and LH by FSH-EF, LHRH, Journal of Clinical Endocrinology and (1990) The role of LH in regulation of cytokines, leptin and nitric oxide. α β Metabolism 55, 453–458. testicular inhibin and B subunit Neuroimmunomodulation 5, 193–202. Johansen, K.L., Chertow, G.M., da Silva, mRNAs in immature and adult animals. MacConnie, S.E., Barkan, A., Lampman, M., Carey, S. & Painter, P. (2001) Endocrinology 127, 1097–1104. R.M. et al. (1986) Decreased Determinants of physical performance Larrea, F., Carino, C., Hardy, D.O., Musto, hypothalamic gonadotropin-releasing in ambulatory patients on hemodialysis. N.A. & Catterall, J.F. (1995) Genetic horomone secretion in male marathon Kidney International 60(4), 1586–1591. variations in human testosterone– runners. New England Journal of Medicine Jones, M.E., Thorburn, A.W., Britt, K.L. estradiol binding globulin. Journal of 315, 411–417. et al. (2000) Aromatase-deficient (ArKO) Steroid Biochemistry and Molecular Biology McCullagh, D.R. (1932) Dual endocrine mice have a phenotype of increased 53(1–6), 553–559. activity of the testes. Science 76, 19–23. adiposity. Proceedings of the National Layman, L.C. (1999) Mutations in human McFarland, K.C., Sprengel, R., Academy of Sciences of the United States of gonadotropin genes and their Phillip, H. et al. (1989) Lutropin/ America 97(23), 12 735–12 740. physiologic significance in puberty and choriogonadotropin receptor: an Kaufman, J.M. & Vermeulen, A. (1997) reproduction. Fertility and Sterility 71(2), unusual member of the G-protein Declining gonadal function in elderly 201–218. coupled receptor family. Science 245, men. Bailliere’s Clinical Endocrinology Lee, D.K. & Chang, C. (2003) Endocrine 494–499. and Metabolism 11, 289–309. mechanisms of disease: expression and McLachlan, R.I., Matsumoto, A.M., Keman, D., Madigan, M.B., Bardin, W.C. degradation of androgen receptor: Burger, H.G., de Kretser, D.M. & & Chen, L.-C.C. (1989) Expression and mechanism and clinical implication. Bremmer, W.J. (1988) Relative roles regulation of testicular inhibin α subunit Journal of Clinical Endocrinology and of follicle-stimulating hormone and gene in vivo and in vitro. Molecular Metabolism 88(9), 4043–4054. luteinizing hormone in the control of Endocrinology (Baltimore, Md) 3, 29–35. Lee, P.A. (1988) Pubertal neuroendocrine inhibin secretion in normal men. Journal Keys, A., Brozek, J., Henschel, A. et al. maturation: early differentiation and of Clinical Investigation 82, 880–884. (1950) The Biology of Human Starvation. stages of development. Adolescent and McPhaul, M.J., Marcelli, M., Zoppi, S., University of Minnesota Press, Pediatric Gynecology 1, 3–14. Griffin, J.E. & Wilson, J.D. (1993) Genetic Minneapolis. Lee, P.A., Yenakis, T., Winer, J. & basis of endocrine disease. 4. The Kim, W.H., Swerdloff, R.S. & Bhasin, S. Matsenbaugh, S. (1976) Puberty in spectrum of mutations in the androgen (1988a) Regulation of α and rat LH β girls: correlation of serum levels of receptor gene that causes androgen mRNAs during GnRH agonist treatment gonadotropins, prolactin, androgens, resistance. Journal of Clinical in vivo in the male rat. Endocrinology 123, estrogens and progestins with physical Endocrinology and Metabolism 76(1), 2111–2116. changes. Journal of Clinical Endocrinology 17–23. Kim, W.H., Yuan, Q.X., Swerdloff, R.S. & and Metabolism 43, 775–784. Mahna, R.M., Harper, A.B., Avent, H.H. & Bhasin, S. (1988b) Regulation of α and Legouis, R., Hardelin, J.P., Levilliers, J. Campbell, D.E. (1973) Age at menarche LH β mRNAs during stimulatory and et al. (1991) The candidate gene for the in athletes and non-athletes. Medicine downregulatory phases of GnRH X-linked Kallmann syndrome encodes a and Science in Sports 5, 11–13. action. Biology of Reproduction 39, protein related to adhesion molecules. Marshall, G.R., Wickings, E.J., Ludecke, 847–853. Cell 67(2), 423–435. D.K. & Nieschlag, E. (1983) Stimulation regulation of testicular function 303

of spermatogenesis in stalk-sectional Moriarty, G.C. (1973) Adenohypophysis: Rietschel, P., Corcoran, C., Stanley, T. et al. rhesus monkeys by testosterone alone. ultrastructural cytochemistry. A review. (2000) Prevalence of hypogonadism Journal of Clinical Endocrinology and Journal of Histochemistry and among men with weight loss related Metabolism 57, 152–159. Cytochemistry 21, 855–892. to human immunodeficiency virus Mason, A.J., Hayflick, J.S., Ling, N. et al. Morley, J.E., Kaiser, F.E., Perry, H.M., 3rd infection who were receiving highly (1985) Complementary DNA sequences et al. (1997) Longitudinal changes in active antiretroviral therapy. Clinical of ovarian follicular fluid inhibin show testosterone, LH, and FSH in healthy, Infectious Diseases 31(5), 1240–1244. precursor structure and homology with older men. Metabolism 46, 410–413. Rivier, C., Rivier, J. & Vale, W. (1986) transforming growth factor-β. Nature Najmabadi, H., Rosenberg, L.A., Yuan, Inhibin mediated feedback control of 318, 659–663. Q.X. et al. (1993) Transcriptional and FSH secretion in the female rat. Science Mather, J.P., Attie, K.M., Woodruff, T.K., post-transcriptional regulation of 234, 205–208. Rice, G.C. & Phillips, D.M. (1990) intribin α subunit gene expression in rat Rock, C.L., Gorenflo, D.W., Drewnowski, Activin stimulates spermatogonial sertoli cell by 8-bromo cAMP. Molecular A. & Demitrack, M.A. (1996) Nutritional proliferation in germ–Sertoli cell co- Endocrinology (Baltimore, Md) 7, 469–476. characteristics, eating pathology, and cultures from immature rat testis. Neill, J.D., Patton, J.M., Daily, R.A., Tsour, hormonal status in young women. Endocrinology 127, 3206–3214. R.C. & Tindall, G.T. (1977) LHRH in American Journal of Clinical Nutrition 64, Matsumoto, A.M. (2002) Andropause: pituitary stalk brood of rhesus monkeys: 566–571. clinical implications of the decline in relationship to level of LH release. Rommerts, F.F.G. (2004) Testosterone: serum testosterone levels with aging in Endocrinology 101, 430–434. an overview of biosynthesis transport men. Journals of Gerontology. Series A, Pardridge, W.M. (1987) Plasma protein- metabolism and non-genomic actions. Biological Sciences and Medical Sciences mediated transport of steroid and In: Testosterone: Action, Deficiency and 57(2), M76–M99. thyroid hormones. American Journal of Substitution (Nieschlag, E. & Behr, H.M., Matsumoto, A.M. & Bremner, W.J. (1984) Physiology 252(2 Pt. 1), E157–E164. eds). Cambridge University Press, Modulation of pulsatile gonadotropin Parks, J.S., Brown, M.R., Hurley, D.L., New York: 1–37. secretion by testosterone in men. Journal Phelps, C.J. & Wajnrajch, M.P. (1999) Rosner, W. (1991) Plasma steroid-binding of Clinical Endocrinology and Metabolism Heritable disorders of pituitary proteins. Endocrinology and Metabolism 58, 609–614. development. Journal of Clinical Clinics of North America 20(4), 697–720. Matsumoto, A.M., Karpas, A.E., Endocrinology and Metabolism 84, Russell, D.W. & Wilson, J.D. (1994) Steroid Paulsen, C.A. & Bremner, W.J. (1983) 4362–4370. 5-α reductase: two gene/two enzymes. Reinitiation of sperm production in Penny, R., Goldstein, I.P. & Frasier, S.D. Annual Review of Biochemistry 63, 25–61. gonadotropin-suppressed normal men (1978) Gonadotropin excretion and Ryan, R.J., Keutmann, H.T., Charlesworth, by administration of FSH. Journal of body composition. Pediatrics 61, M.C. et al. (1987) Structure function Clinical Investigation 72, 1005–1015. 294–300. relationships of gonadotropins. Recent Matsumoto, A.M., Paulsen, C.A. & Pincus, S.M., Veldhuis, J.D., Mulligan, T., Progress in Hormone Research 43, Bremner, W.J. (1984) Stimulation of Iranmanesh, A. & Evans, W.S. (1997) 383–429. sperm production by human LH in Effects of age on the irregularity of LH Saade, G., London, D.R. & Clayton, R.N. gonadotropin suppressed normal men. and FSH serum concentrations in (1989) The interaction of GnRH and Journal of Clinical Endocrinology and women and men. American Journal of estradiol on LH and prolactin gene Metabolism 59, 882–887. Physiology 273(5 Pt 1), E989–E995. expression in hypogonadal (hpg) mice. Mendonca, B.B., Inacio, M., Costa, E.M. et Pirke, K.M. & Doerr, P. (1973) Age related Endocrinology 124, 1744–1753. al. (1996) Male pseudohermaphroditism changes and interrelationships between Sairam, M.R. (1983) Gonadotropic due to steroid 5 α-reductase 2 deficiency. plasma testosterone, oestradiol and hormones: relationship between Diagnosis, psychological evaluation, testosterone-binding globulin in normal structure and function with emphasis and management. Medicine; Analytical adult males. Acta Endocrinologica 74, on antagonists. In: Hormonal Proteins and Reviews of General Medicine, Neurology, 792–800. Peptides, vol. 11 (Li, C.H., ed.). Academic Psychiatry, Dermatology, and Pediatrics Prince, F.P. (2001) The triphasic nature of Press, New York: 1–79. 75(2), 64–76. Leydig cell development in humans, and Santen, R.J. (1975) Is aromatization of Mercer, J.E., Clements, J.E., Funder, J.W. comments on nomenclature. Journal of testosterone to estradiol required for & Clarke, I.J. (1988) LH β mRNA levels Endocrinology 168(2), 213–216. inhibition of LH secretion in men? are regulated primarily by GnRH and Rebar, R., Judd, H.L., Yen, S.S.C. et al. Journal of Clinical Investigation 56, not by negative estrogen feedback on the (1976) Characterization of the 1555–1563. pituitary. Neuroendocrinology 47, 63–66. inappropriate gonadotropin secretion in Scheckter, C.B., Matsumoto, A.M. & Mohamed-Ali, V., Pinkney, J.H. & polyvistic ovary syndrome. Journal of Bremner, W.J. (1989) Testosterone Coppack, S.W. (1998) Adipose tissue as Clinical Investigation 57, 1320–1325. administration inhibits gonadotropin an endocrine and paracrine organ. Remes, K., Kuoppasalmi, K. & secretion by an effect directly at the International Journal of Obesity and Related Adlercreutz, H. (1979) Effect of long- human pituitary. Journal of Clinical Metabolic Disorders 22, 1145–1158. term physical training on plasma Endocrinology and Metabolism 68, Mori, M. & Marsh, J.M. (1982) The site of testosterone, androstenedione, 397–401. LH stimulation of steroidogenesis in luteinizing hormone and sex-hormone- Schwanzel-Fukuda, M. & Pfaff, D.W. mitochondria of rat corpus luteum. binding globulin capacity. Scandinavian (1989) Origin of luteinizing hormone Journal of Biological Chemistry 257, Journal of Clinical and Laboratory releasing hormone neurons. Nature 338, 6178–6183. Investigation 39(8), 743–749. 161–163. 304 chapter 21

Schwartz, M.W., Baskin, D.G., Kaiyala, K.J. exercise-induced hypogonadism in male European Journal of Applied Physiology 82, & Woods, S.C. (1999) Model for the runners. Western Journal of Medicine 169, 121–128. regulation of energy balance and 9–12. Ueno, N., Ling, N., Yng, S.Y. et al. (1987) adiposity by the central nervous system. Smilios, I., Pilianidis, T., Karamouzis, M. Isolation and partial characterization of American Journal of Clinical Nutrition 69, & Tokmakidis, S.P. (2003) Hormonal follistatin, a novel Mr 35000 monomeric 584–596. responses after various resistance protein which inhibits the release of Scully, K.M. & Rosenfeld, M.G. (2002) exercise protocols. Medicine and Science FSH. Proceedings of the National Academy Pituitary development: regulatory codes in Sports and Exercise 35(4), 644–654. of Sciences of the United States of America in mammalian organogenesis. Science Smith, E.P., Boyd, J., Frank, G.R. et al. 84, 8282–8286. 295(5563), 2231–2235. (1994) Estrogen resistance caused by a Urban, R.J., Evans, W.S., Rogol, A.D. et al. Seth, A., Stanley, S., Jethwa, P. et al. (2004) mutation in the estrogen receptor gene (1988) Contemporary aspects of discrete Galanin-like peptide stimulates the in a man. New England Journal of Medicine peak detection algorithms 1. The release of gonadotropin-releasing 331, 1056–1061. paradigm of the LH pulse signal in men. hormone in vitro and may mediate the Spratt, D.I., Chin, W.W., Ridgeway, E.C. & Endocrine Reviews 9, 3–37. effects of leptin on the hypothalamo– Crowley, W.F., Jr. (1986) Administration Vale, W., Rivier, C., Hsueh, A.J.W. et al. pituitary–gonadal axis. Endocrinology of low dose pulsatile GnRH (1988) Chemical and biological 145(2), 743–750. administration to GnRH deficient men characterization of the inhibin family Sharpe, R.M. (1987) Testosterone regulates free α subunit secretion. of protein hormones. Recent Progress in and spermatogenesis. Journal of Journal of Clinical Endocrinology and Hormone Research 44, 1–34. Endocrinology 113, 1–2. Metabolism 62, 102–108. Van Der Spruy, Z.M. (1985) Nutrition and Sharpe, R.M., Donachie, K. & Cooper, L. Sprengel, R., Braun, T., Nikofics, K., reproduction. Clinics in Obstetrics and (1988) Reevaluation of intratesticular Segaloff, D.L. & Seeburg, P.H. (1990) Gynaecology 12, 579–604. level of testosterone required for The testicular receptor for FSH: structure Van Der Walt, L.A., Wilmsen, E.N. & quantitative maintenance of and functional expression of cloned Jenkins, T. (1978) Unusual sex hormone spermatogenesis in the rat. Journal of cDNA. Molecular Endocrinology patterns among desert-dwelling Endocrinology 117, 19–26. (Baltimore, Md) 4, 525–530. hunter gatherers. Journal of Clinical Shupnik, M.A. (1990) Effects of Stager, K., Behr, R., Kleiner, I., Weinbauer, Endocrinology and Metabolism 46, gonadotropin-releasing hormone on rat G.F., Bergmann, M. (2004) Expression 658–663. gonadotropin gene transcription in vitro: of activator of CREM in the testis Veldhuis, J.D., Rogol, A.D., Samojlik, E. requirement for pulsatile administration (ALP) during normal and impaired & Ertel, N.H. (1984) Role of endogenous for luteinizing hormone-β gene spermatosis genesis: correlation with opiates in the expression of negative stimulation. Molecular Endocrinology CREM expression. Molecular Human feedback actions of androgen and (Baltimore, Md) 4(10), 1444–1450. Reproduction 10, 129–135. estrogen on pulsatile properties of LH Shupnik, M.A., Gharib, S.D. & Chin, W.W. Stein, Z., Susser, M., Saenger, G. & secretion in man. Journal of Clinical (1989) Divergent effects of estradiol on Marolla, F. (1973) Famine and Human Investigation 74, 47–55. gonadotropin gene transcription in Development: The Dutch Hunger Winter Veldhuis, J.D., Evans, W.S., Demers, L.M. pituitary fragments. Molecular of 1944–1945. Oxford University Press, et al. (1985) Altered neuroendocrine Endocrinology (Baltimore, Md) 3, 474–478. New York. regulation of gonadotropin secretion Simon, D., Preziosi, P., Barrett-Connor, E. Steinberger, E. (1971) Hormonal control in women distance runners. Journal of et al. (1992) The influence of aging on of mammalian spermatogenesis. Clinical Endocrinology and Metabolism 61, plasma sex hormones in men: the Physiological Reviews 51, 1–22. 557–563. Telecom Study. American Journal of Steinberger, E. & Chowdhury, M. (1977) Vermeulen, A. (1988) Physiology of the Epidemiology 135, 783–791. The effects of testosterone propionate testosterone-binding globulin in man. Simpson, E.R. (1979) Cholesterol side chain and estradiol benzoate on the in vitro Annals of the New York Academy of cleavage, cytochrome P-450, and the synthesis of FSH. Biology of Reproduction Sciences 538, 103–111. control of steroidogenesis. Molecular 16, 403–408. Villanueva, A.L., Schlosser, C., Hopper, B. and Cellular Endocrinology 13, 213–227. Swerdloff, R.S., Campfield, L.A., Paiacios, et al. (1986) Increased cortisol production Singh, A.B., Norris, K., Modi, N. et al. A. & McClure, R.D. (1979) Suppression in women runners. Journal of Clinical (2001) Pharmacokinetics of a of human spermatogenesis by depot Endocrinology and Metabolism 63, 133–136. transdermal testosterone system in men androgen: potential for male Warren, M.P. (1980) The effects of with end stage renal disease receiving contraception. Journal of Steroid exercise on pubertal progression and maintenance hemodialysis and healthy Biochemistry 11, 663–670. reproductive functions in girls. Journal of hypogonadal men. Journal of Clinical Tenover, J.S., Matsumoto, A.M., Plymate, Clinical Endocrinology and Metabolism 51, Endocrinology and Metabolism 86(6), S.R. & Bremner, W.J. (1987) The effects 1150–1157. 2437–2445. of aging in normal men on bioavailable Weiss, J., Crowley, W.F., Jr. & Jameson, J.L. Sizonenko, P.C. (1979) Endocrinology in testosterone and luteinizing hormone (1992) Pulsatile gonadotropin-releasing preadolescents and adolescents 1. secretion: response to clomiphene hormone modifies polyadenylation of Hormonal changes during normal citrate. Journal of Clinical Endocrinology gonadotropin subunit messenger puberty. American Journal of Diseases of and Metabolism 65, 1118–1126. ribonucleic acids. Endocrinology 130, Children 132, 704–712. Truls, R., Trine, B. & Jostein, H. (2000) 415–420. Skarda, S.T. & Burge, M.R. (1998) Hormonal responses to high- and Whitcomb, R.W., O’Dea, St. L., Finkelstein, Prospective evaluation of risk factors for moderate-intensity strength exercise. J.S., Heavern, D.M. & Crowley, W.F., Jr. regulation of testicular function 305

(1990) Utility of free α subunit as an Winter, J.S.D., Hughes, I.A., Reyes, F.L. & Ying, S.-Y. (1988) Inhibins, activins alternative neuroendocrine marker of Faiman, C. (1976) Pituitary–gonadal and folbstatins: gonadal proteins GnRH stimulation of the gonadotroph in relations in infancy: patterns of serum modulating the secretion of FSH. the human; evidence from normal and gonadal steroid concentrations in man Endocrine Reviews 9, 267–293. GnRH-deficient men. Journal of Clinical from birth to 2 years of age. Journal of Yuan, Q.X., Swerdloff, R.S. & Bhasin, S. Endocrinology and Metabolism 70, Clinical Endocrinology and Metabolism 42, (1988) Differential regulation of rat LH 1654–1661. 679–686. α and β subunits during the stimulatory Wierman, M.E. & Wang, C. (1990) Winters, S.J., Sherins, R.J. & Loriaux, D.L. and downregulatory phases of GnRH Androgen selectively stimulates FSH β (1979) Studies on the role of sex steroids action. Endocrinology 122, 504–510. mRNA levels after GnRH antagonist in the feedback control of gonadotropin Zhou, Z.-X., Wong, C.-I., Sar, M. & Wilson, administration. Biology of Reproduction concentration in men. Journal of Clinical E.M. (1994) The androgen receptor: an 42, 563–571. Endocrinology and Metabolism 48, overview. Recent Progress in Hormone Wierman, M.E., Rivier, J.E. & Wang, C. 553–558. Research 49, 249–274. (1989) GnRH dependent regulation of Yamaji, T., Dierschke, D.J., Bhattacharya, Zmeili, S.M., Papavasiliou, S.S., Thorner, gonadotropin subunit mRNA levels in A.N. & Knobil, E. (1992) The negative M.O. et al. (1986) Alpha and LH β the rat. Endocrinology 124, 272–278. feedback control by estradiol and subunit mRNAs during the rat estrous Wilson, J.D., Griffin, J.E. & Russell, D.W. progesterone of LH secretion in the cycle. Endocrinology 119, 1867–1869. (1993) Steroid 5 α-reductase 2 deficiency. ovariectormized rhesus monkey. Zmuda, J.M., Cauley, J.A., Kriska, A. Endocrine Reviews 14(5), 577–593. Endocrinology 90, 771–778. et al. (1997) Longitudinal relation Winter, J.S.D., Faiman, C., Hobson, W. Yamamoto, M., Diebell, N.D. & between endogenous testosterone and et al. (1975) Pituitary–gonadal relations Bogdanove, E.M. (1970) Analysis cardiovascular disease risk factors in infancy 1. Patterns of serum of initial and delayed effects of in middle-aged men. A 13-year follow- gonadotropin levels from birth to 4 years orchidectomy and ovariectomy on up of former Multiple Risk Factor of age in man and chimpanzee. Journal of pituitary and serum LH levels in adult Intervention Trial participants. Clinical Endocrinology and Metabolism 40, and immature rats. Endocrinology 86, American Journal of Epidemiology 146, 545–557. 1102–1111. 609–617. Chapter 22

Hormonal and Growth Factor-Related Mechanisms Involved in the Adaptation of Skeletal Muscle to Exercise

FAWZI KADI

Introduction The enlargement of skeletal muscles: the role myonuclei and satellite cells Physical exercise can be assimilated to a very com- plex physiological stimulus that challenges diverse Enhanced contractile protein synthesis is an un- aspects of the cellular function. Skeletal muscle is equivocal condition for the increase in the size of one of the tissues that respond to exercise by under- muscle fibers in response to training. Both protein going a series of adjustments at the level of sev- synthesis and degradation rates are altered during eral of its components. The shortening velocity of the enlargement of skeletal muscles (Goldberg et al. skeletal muscles, the amount of force they are able to 1975). The increase in muscle protein synthesis generate and their capacity to resist fatigue are above resting levels occurs very rapidly, between important properties closely related to athletic per- 1 and 4 h after the completion of a single bout of formances. Thanks to the high degree of malleabil- exercise in humans (Wong & Booth 1990; Chesley ity of different muscular parameters such as fiber et al. 1992; Biolo et al. 1995; Phillips et al. 1997). At size, fiber type composition and capillarization, the onset of muscle hypertrophy, increased protein skeletal muscles adapt adequately to changes im- synthesis correlates with an increase in RNA activ- posed by training. However, skeletal muscles will ity (Laurent et al. 1978; Wong & Booth 1990). The adapt differently to endurance and strength exer- translation of mRNA is enhanced by factors whose cises, suggesting the existence of different sensing activity is known to be regulated by their phos- systems. Therefore the adaptive process of skeletal phorylation state (Frederickson & Sonenberg 1993; muscles to training can be viewed as orchestrated Wada et al. 1996). Paralleling these changes, amino local and peripheral events where hormonal, mech- acid transport into exercising muscles is also in- anical, metabolic and neural factors are key regu- creased following training. This would theoretically latory signals. Changes in the rate of synthesis of enhance the availability of amino acids for new hormones and growth factors, and the expression muscle protein synthesis (Biolo et al. 1997). of their receptors, are important signals involved in Following this initial step of muscle fiber hyper- the adaptive process allowing skeletal muscles to trophy, several lines of evidence indicate that meet the physiological demand of different types increased RNA levels (rather than RNA activity) of physical activities. A brief summary of the role appear to be essential for muscle fibers to continue played by some hormones and growth factors in to hypertrophy. In this regard, increased amount of muscle hypertrophy, in the regulation of muscle mRNA can be attributed to either increased gene fiber phenotype and in the remodeling of capillary transcription per myonucleus or to increased num- network form the substance of this chapter. ber of myonuclei. Adult muscle fibers contain hundreds of myonuclei, where each myonucleus sustains the protein synthesis over a finite volume 306 hormones, skeletal muscle and exercise 307

of cytoplasm, a concept called ‘nuclear domain’ source for the addition of new myonuclei into (Cheek 1985; Hall & Ralston 1989; Allen et al. 1999). hypertrophied fibers (Moss & Leblond 1971; It is important to note that although myonuclei are Schiaffino et al. 1976). Satellite cells are located post-mitotic, they are nonetheless able to sustain the between the basal lamina and the plasma mem- enlargement of fibers up to a certain limit after brane of muscle fibers (Mauro 1961); they are char- which the recruitment of new myonuclei becomes acterized by a high nuclear to cytoplasmic ratio, necessary. In agreement with this statement, results a well-developed Golgi apparatus, a prominent from animal and human studies showed that the rough endoplasmic reticulum and a heterochrom- hypertrophy of skeletal muscle fibers was accom- atic nucleus (Campion 1984). Several stimuli can panied by significant increases in the myonuclear activate satellite cells, which then undergo prolifera- number (Goldberg et al. 1975; Cabric & James 1983; tion. The daughter cells then fuse with the under- Winchester & Gonyea 1992; Allen et al. 1995; Kadi lying adult muscle fiber providing new myonuclei. 2000). In well-trained humans, the number of myo- The involvement of satellite cell-derived myonuclei nuclei in hypertrophied skeletal muscle fibers of in fiber hypertrophy is further supported by experi- power lifters is higher than that of sedentary sub- ments demonstrating that satellite cell activation jects, and a linear relationship between the number and proliferation is required to support the enlarge- of myonuclei and the cross-sectional area of muscle ment of muscles in animal models (Rosenblatt & fibers is found (Kadi et al. 1999a; Kadi 2000). The Parry 1992; Rosenblatt et al. 1994). addition of new myonuclei to the enlarged muscle In parallel with muscle fiber hypertrophy, it has fibers would play a role in the maintenance of a been shown that heavy resistance strength training constant myonuclei/cytoplasm ratio, i.e. nuclear induced a significant increase in the number of domain. The incorporation of additional myonuclei satellite cells in human skeletal muscles (Kadi 2000; into hypertrophying human muscle fibers was Roth et al. 2001). A 46% increase in satellite cell fre- reported both in young and elderly subjects (Hikida quency was reported in skeletal muscle of young et al. 1998; Kadi & Thornell 2000). women after 10 weeks of strength training (Kadi & While the number of myonuclei per fiber is Thornell 2000). More recently, an increase in the increased during muscle fiber hypertrophy, the con- number of satellite cells was found in skeletal mus- verse has also been documented in animal studies. cles of a group of men aged between 70 and 80 years A decrease in the number of myonuclei occurs as in response to endurance training (Charifi et al. a result of an ongoing atrophy of muscle fibers 2003a). Thus, satellite cells contribute to the acquisi- induced by spinal cord transection (Allen et al. tion of new myonuclei and to the renewal of their 1995), space flight (Allen et al. 1996) or hindlimb sus- own pool (Bischoff 1994; Schultz & McCormick pension (Hikida et al. 1997). Thus, the modulation 1994). Finally, the proliferation of satellite cells of the myonuclear population seems to be of great followed by the fusion of daughter cells together importance in the regulation of fiber size. However, would give rise to new muscle fibers (Kennedy et al. it is important to keep in mind that the enhancement 1988; Yamada et al. 1989; McCormick & Schultz 1992; of the myonuclear number would occur only when Antonio & Gonyea 1993; Kadi & Thornell 1999). the transcriptional capacity of existing myonuclei The newly formed muscle fibers would replace becomes unable to support the enlargement of the damaged fibers or contribute to fiber hyperplasia if fiber. Indeed, significant increases in myonuclear the number of newly generated fibers exceeds the number have been observed in muscle fibers that number of injured fibers upon training (Fig. 22.1). had hypertrophied by more than 26% (Cabric & James 1983; Allen et al. 1995; Roy et al. 1999; Kadi & The effects of androgenic-anabolic steroids Thornell 2000), but not by 6.8–15.5% (Giddings & Gonyea 1992). The use of androgenic-anabolic steroids is accom- As existing myonuclei in an adult muscle fiber panied by a remarkable increase in muscle size and are postmitotic, muscle satellite cells are the major strength in animal studies (Egginton 1987; Salmons 308 chapter 22

Training, hormones and growth factors hypertrophied muscle fibers (Kadi et al. 1999b). A main mechanism by which androgenic-anabolic steroids induce muscle hypertrophy is by activating and inducing the proliferation of satellite cells, S MN which subsequently incorporate into muscle fibers Muscle fiber MN MN or fuse together to form new muscle fibers. In agree- ment with this statement is the immunohistochem- ical localization of androgen receptors in cultured Activated satellite cell S satellite cells indicating that anabolic steroids can act directly on muscle satellite cells (Doumit et al. S 1996). S Dotter cells S S S S S S S S Androgen receptors

Blockade of androgen receptors by oxendolone, an androgen receptor antagonist, suppressed the hypertrophy caused by exercise (Inoue et al. 1994). 1. Provide additional myonuclei Although several factors are involved in the adapta- 2. Generate new muscle fibers or repair injured muscle fibers tion of muscle fibers to exercise, this experiment 3. Enhance satellite cell pool clearly shows that androgen receptors are import- Fig. 22.1 The effects of training, hormones and growth ant mediators of the exercise-induced muscle fiber factors on satellite cells. MN, myonucleus; S, satellite cell. hypertrophy. Androgen receptors belong to the family of ligand-responsive transcription regulators. When 1992). The administration of supra-physiological hormones bind to the receptor, it becomes activated doses of testosterone for 10 weeks in untrained and the hormone-receptor complex is translocated and trained men produced a significant increase in to the hormone responsive element within the muscle strength and in the cross-sectional area nucleus. The binding to selective genes increases the of the quadriceps (Bhasin et al. 1996). Androgenic- rates of transcription (Luke & Coffey 1994). Early anabolic steroids are known to increase the rate of reports indicated that androgen receptors were protein synthesis and to promote muscle growth located in the cytosol of muscle fibers (Krieg 1976; both in vivo and in vitro (Powers & Florini 1975; Max et al. 1981). Using immunohistochemistry with Rogozkin 1979). In humans, long-term anabolic specific polyclonal and monoclonal antibodies, the steroids usage accentuates the degree of fiber hyper- nuclear location of androgen receptors has been trophy of muscle fibers in well-trained power-lifters clearly demonstrated in nearly all tissues (Sar et al. (Kadi et al. 1999b). Skeletal muscles of power-lifters 1990; Takeda et al. 1990; Ruizeveld-De-Winter et al. taken anabolic steroids were characterized by 1991; Kimura et al. 1993; Janssen et al. 1994). In skel- both extremely large fibers and high myonuclear etal muscles, androgen receptors are expressed in numbers (Kadi et al. 1999b). Similarly, using animal myonuclei (Takeda et al. 1990; Kimura et al. 1993; models, it was also found that androgenic-anabolic Dorlochter et al. 1994; Kadi et al. 2000b) and in satel- steroids mediate their myotrophic effect by enhan- lite cells (Doumit et al. 1996). In normal resting cing the myonuclear content of muscles fibers and human muscle fibers, androgen receptors are ex- by increasing the number of muscle fibers (Galavazi pressed in some but not all myonuclei (Kadi et al. & Szirmai 1971; Sassoon & Kelley 1986; Joubert & 2000b), and differences in androgen receptor con- Tobin 1989; Joubert & Tobin 1995). Thus, androgenic- tent have been reported between human trapezius anabolic steroids would increase the myonuclear and vastus lateralis muscles (Kadi 2000). Similar number to sustain the protein synthesis of extremely intermuscular differences in androgen receptor hormones, skeletal muscle and exercise 309

content have also been demonstrated in frog skel- androgen receptors following training occur rapidly. etal muscles (Dorlochter et al. 1994). In agreement By 3 days of electrical stimulation, a 25% increase with the intermuscular differences in androgen in the number of androgen receptors was reported receptor content, it has been shown that the sensitiv- in the stimulated rat gastrocnemius muscle, and ity of guinea pig skeletal muscles to testosterone was followed by a progressive hypertrophy of the stimulation was greatest in the head and neck muscle (Inoue et al. 1993). region and gradually decreased from head to hind- The effects of androgenic-anabolic steroids on quarters (Kochakian & Tillotson 1957). Intermuscu- androgen receptor content have also been invest- lar differences in androgen receptor content might igated. In animal models and in cultured muscle well reflect differences in embryological origins, satellite cells, it has been shown that androgenic- nerve supply and functional requirements of differ- anabolic steroids may either up-regulate (Doumit ent muscles. et al. 1996) or down-regulate (Lin et al. 1993; Bricout Training influences the number of androgen et al. 1994) androgen receptor content. Doumit et al. binding sites in skeletal muscles. The increase in (1996) showed that the administration of testo- androgen receptors would lead to an enhancement sterone enhanced androgen receptor immunore- of the sensitivity of muscles to circulating andro- activity in porcine satellite cell nuclei. In contrast, gens. A significant increase in the number of andro- using radio-competition assay, it has been shown gen receptors has been shown to occur following that the concentration of androgen receptors was endurance and strength training, and electrical decreased following androgenic-anabolic steroids stimulation in animal studies (Inoue et al. 1993; treatment in rat soleus and extensor digitorum Deschenes et al. 1994). The amplitude of changes longus muscles (Bricout et al. 1994). In fact, the in androgen receptor content following training effects of androgenic-anabolic steroids on androgen is muscle dependent (Hickson & Kurowski 1986; receptors might also be muscle dependent. Frog Deschenes et al. 1994; Kadi et al. 2000b). Long-term skeletal muscle fibers in the shoulder region have strength training is associated with changes in been shown to be more sensitive to testosterone androgen receptor-containing myonuclei in human than fibers from other regions (Regnier & Herrera trapezius but not vastus lateralis muscle (Kadi et al. 1993a, 1993b). It has also been shown that two rabbit 2000b). Similar differences in the regulation of skeletal muscles characterized by similar fiber size, androgen receptors following exercise exist between fiber type composition, nerve and blood vessel sup- rat soleus, extensor digitorum longus, gastrocne- ply can differ greatly in their response to andro- mius and plantaris muscles (Hickson & Kurowski genic-anabolic steroid treatment (Salmons 1992). 1986; Salmons 1992; Inoue et al. 1993; Bricout et al. Finally, long-term self-administration of androgenic- 1994; Deschenes et al. 1994). anabolic steroids in humans was associated with Deschenes et al. (1994) examined the effects of changes in androgen receptor-containing myonu- endurance and resistance training on androgen clei in the trapezius but not in the vastus lateralis receptor content and receptor affinity to dihydro- (Kadi et al. 2000b). Clearly, further studies are testosterone in rat fast and slow skeletal muscles. warranted to better understand the modulation of Neither endurance nor resistance training induced androgen receptors in skeletal muscles in response alterations in androgen receptor affinity to dihy- to physiological and supra-physiological conditions. drotestosterone. Endurance training induced an enhancement in androgen binding capacity in the Fibroblast growth factors slow muscle whereas resistance training induced an enhancement in androgen binding capacity in The fibroblast growth factor (FGF) family comprises the fast muscle. Thus, alterations in androgen of ten members involved in different biological receptor content are not only muscle dependent, functions (Yamaguchi & Rossant 1995). Some FGF they also depend upon the type of physical activity isoforms have been shown to play an important role (endurance/strength exercises). The changes in in the enlargement of muscle fibers in response to a 310 chapter 22

physiological stimulus and others may exert their (Brahm et al. 1997), and IGF-I is considered as myotrophic role during the reparation of muscle an important factor mediating the enlargement of fibers following fiber damage. In this respect, FGF6 skeletal muscles in response to training. IGF-I is able is an important factor contributing to normal skel- to stimulate satellite cells proliferation, differenti- etal muscle regeneration following injury (Floss ation and fusion (Dodson et al. 1985; Florini et al. et al. 1997). FGF6 would be involved in important 1991; Quinn et al. 1994; Goldspink, D.F. et al. 1995). regenerative events such as the activation and pro- Immunohistochemical studies showed that IGF-I liferation of satellite cells and the expression of protein is expressed in satellite cells and in myo- important myogenic regulatory factors. FGF2 and tubes of regenerating rat skeletal muscles ( Jennische FGF4, which are localized in the myofiber peripheral et al. 1987; Jennische 1989; Jennische & Matejka matrix in adult skeletal muscles, are up-regulated 1992). during stretch mediated hypertrophy in an avian An increase in IGF-I immunoreactivity mostly wing-weighting model (Mitchell et al. 1999). The within the muscle fibers was observed following up-regulation and specific cellular location of FGF2 an acute bout of eccentric exercise in rat tibialis and FGF4 support their role in the activation and anterior muscle (Yan et al. 1993). In humans, 7 days proliferation of satellite cells (Mitchell et al. 1999). of strenuous exercise induced an elevation of IGF-I Located in the same matrix as satellite cells, it is immunoreactivity in the vastus lateralis muscle hypothesized that the release of these FGF isoforms (Hellsten et al. 1996). The immunostaining for IGF-I from heparin components might be involved in the was located on nuclei that might either represent generation of new muscle fibers following training myonuclei or satellite cells (Hellsten et al. 1996). (Yamada et al. 1989). Alterations in the interaction Enhanced IGF-I mRNA and protein in parallel with FGF/heparan-sulfate proteoglycans in exercising the hypertrophy of muscle fibers indicates that IGF- muscles would modulate the availability of FGF to I is a key factor involved in the process of muscle satellite cells. fiber enlargement via the recruitment of muscle Investigating the mechanisms governing the con- satellite cells (Adams & Haddad 1996). Bamman version of a mechanical load into a skeletal muscle et al. (2001) showed that IGF-I mRNA levels in- growth response, Clarke & Feeback (1996) found creased 48 h after the completion of a single bout that the release of FGF2 increases in parallel with of concentric and eccentric muscle loading in human increased mechanical load in a tissue culture model vastus lateralis muscle (the increase following ec- of differentiated human skeletal muscle cells. The centric loading being more important). In the same growth response was inhibited when the biologic study, a substantial increase in androgen receptor activity of FGF2 was neutralized (Clarke & Feeback mRNA was also found to occur following both con- 1996). This experiment strongly suggests that FGF2 centric and eccentric loading. release is an important autocrine mechanism for Advances in molecular biology allowed the transducing the stimulus of mechanical load into a discovery of new IGF-I isoforms (Yang et al. 1996; skeletal muscle growth response (Clarke & Feeback Goldspink, G. 1999; Hameed et al. 2002). Two main 1996). types of IGF-I expressed in skeletal muscles have been described. The first muscle isoform with a systemic mode of action is known as muscle liver- Insulin-like growth factors and receptor type L.IGF-I and is similar to the main liver IGF-I Insulin-like growth factor (IGF) isoforms are pro- (Yang et al. 1996). The second muscle isoform, called duced by many tissues and are important in both mechano growth factor (MGF) has been discovered embryonic and postnatal development. IGF-II is es- in skeletal muscles subjected to stretch and overload sential for normal embryonic development whereas and has an autocrine/paracrine action (Yang et al. IGF-I is important for both pre- and postnatal 1996). In response to stretch by immobilizing the growth (DeChiara et al. 1990; Baker et al. 1993). Exer- hindlimb in the extended position, a substantial cising skeletal muscles produce and utilize IGF-I increase in MGF was observed in rabbit extensor hormones, skeletal muscle and exercise 311

digitorum longus (Yang et al. 1996). Later, it has capacity and affinity as well as in IGF-I receptor been shown that both muscle L.IGF-I and MGF mRNA in rat skeletal muscle (Willis et al. 1997). mRNAs were significantly increased in rabbit Similarly, long-term training induced a significant extensor digitorum longus following stretch and increase in IGF-I and insulin receptor content (Willis stretch combined with electrical stimulation at 10 Hz et al. 1998). Altogether, these findings strongly sug- but not after electrical stimulation alone (McKoy gest that IGF-I and IGF-I receptors are essential et al. 1999). These experiments clearly showed that mediators of muscle fiber hypertrophy. IGF-I isoforms mediate the enlargement of skeletal muscles mainly in response to increased mechanical Capillarization of skeletal muscles load. The involvement of MGF in fiber hypertrophy has been investigated in young and elderly subjects. Blood vessels in the skeletal muscles form a rich net- A significant increase in MGF mRNA occured in the work of capillaries around muscle fibers. Capillaries young but not in the elderly subjects following 10 consist of a single layer of endothelial cells sur- sets of six repetitions of single-legged knee extensor rounded by a luminal glycocalyx and an ablum- exercise at 80% of 1-repetition maximum (1-RM) inal basement membrane. Located at the end of (Hameed et al. 2003). It was concluded that the the cardiorespiratory system, the capillary network reduced MGF response to high resistance exercise plays an important role in nutrients, and oxygen in elderly subjects might indicate an age-related and carbon dioxide exchange with muscle fibers. In desensitivity to mechanical loading. response to endurance training, a remodeling of the The administration of IGF-I has been considered capillary bed has been shown to occur in different for the treatment of various neuromuscular diseases human skeletal muscles (Andersen, P. & Henriksson characterized by muscular atrophy. In this respect, 1977; Hudlicka et al. 1992; Wang et al. 1993; Kadi et al. IGF-I has been successfully used to prevent the 2000a; Charifi et al. 2003b). development of steroid-induced muscle atrophy (Kanda et al. 1999). At the cellular level, it has been Vascular endothelial growth factor shown that daily growth hormone/IGF-I adminis- tration combined with functional overload of rat Vascular endothelial growth factor (VEGF) is a soleus muscle results in the enlargement of muscles heparin-binding endothelial cell-specific mitogen and a concomitant enhancement of the myonuclear mediating angiogenesis in different tissues. Both number of muscle fibers (McCall et al. 1998). exercise and hypoxia can cause an increase in VEGF IGF-I and IGF-II promote their growth effects mRNA in human skeletal muscle (Gustafsson et al. through IGF-I receptor. Mice with disturbed IGF-I 1999; Richardson et al. 1999). VEGF mRNA increased receptor die soon after birth (Baker et al. 1993). The in response to a single bout of exercise at a work lack of IGF-I receptor induces severe hypoplasia in load corresponding to 50% of peak work load in mice suggesting that IGF-I receptor is essential for normal healthy subjects and in patients with chronic the establishment of a mature skeletal muscle (Baker renal failure (Wagner et al. 2001). The receptors for et al. 1993). While functional inactivation of the VEGF (flt-1 and flk-1) are also up-regulated follow- IGF-I receptor induced a marked muscle hypoplasia ing both exercise and hypoxia (Takagi et al. 1996; and a decrease in MyoD and myogenin levels (two Gerber et al. 1997; Olfert et al. 2001). Thus, VEGF and important members of the myogenic regulatory its receptors are involved in the densification of the factors) (Fernandez et al. 2002), the opposite occurs capillary network in response to physical activity. when the IGF-I receptor is overexpressed (Quinn There is a strong increase in VEGF mRNA following et al. 1994). Therefore, IGF-I receptor is currently a single bout of exercise in untrained human skel- considered as a key regulator of muscle mass via its etal muscles. VEGF response is attenuated in skeletal action on muscle-specific genes. A single acute bout muscle of trained subjects (Richardson et al. 2000). of exercise has been shown to be accompanied by a This might reflect an initial intense proliferation of significant increase in the IGF-I receptor binding capillaries in untrained muscles occurring at the 312 chapter 22

beginning of a training program, followed by a of the myosin molecule exists in multiple isoforms period where the enhancement of the capillariza- and represents a major determinant of the force- tion occurs at a slower rate and requires high doses velocity properties of muscle fibers. The four most of training as the training status of muscles is important MyHC isoforms expressed in adult skel- improved. etal muscle fibers are: MyHC Iβ, MyHC IIA, MyHC Among factors controlling the expression of IIX/IID and MyHC IIB. Each isoform is charac- VEGF, hypoxia inducible factor 1 (HIF-1) subunit is terized by a specific shortening velocity and force currently considered as a major regulatory factor. production. Fibers containing MyHC I have a slow The effects of short-term exercise training on VEGF contracting velocity and produce less force than and HIF-1 have been studied in eight healthy males. fibers containing MyHC IIA, IIX and IIB. Within Although VEGF levels increased following seven muscle fibers containing fast MyHCs, the fastest training sessions, no changes in the levels of HIF-1 and strongest are those containing MyHC IIB fol- mRNA subunits were found (Gustafsson et al. 2002). lowed by fibers expressing MyHC IIX and MyHC This might indicate that an increase in HIF-1 mRNA IIA (Bottinelli et al. 1994a, 1994b). is not the only factor involved in the regulation The contractile properties of skeletal muscles are of training-induced VEGF enhancement. However, able to undergo significant changes in response to the intervention of different factors controlling the exercise. It is generally accepted that endurance remodeling of the capillary network probably oc- training results in a fast to slow transition of MyHC curs during serial transient time frames making it isoforms (Baumann et al. 1987; Schaub et al. 1989), difficult to assess their importance unless the whole whereas strength training causes an increase in adaptive process is monitored. MyHC IIA and a decrease in MyHC IIX (Staron et al. FGFs are also believed to play a role in skeletal 1991; Adams et al. 1993; Andersen, J.L. et al. 1994; Fry muscle angiogenesis. However, recent studies sug- et al. 1994; Kraemer et al. 1995; Kadi & Thornell 1999; gest that their contribution to angiogenesis is less Andersen, J.L. & Aagaard 2000). It is also suggested important than that played by VEGF (Richardson that muscle fibers containing MyHC IIX are seldom et al. 2000; Wagner et al. 2001). At present, although recruited in the normal daily activities of most people. it is suggested that VEGF is the most important If they become more recruited, as it is the case dur- angiogenic factor involved in the adaptation of ing training, they are converted into fibers contain- the capillary network in human skeletal muscles, ing MyHC IIA (type IIA fibers being more fatigue further studies are needed to advance knowledge resistant than type IIX fibers) (Goldspink, G. et al. about the importance of all angiogenic factors in 1991; Staron et al. 1991; Kraemer et al. 1995). During regulating skeletal muscle angiogenesis. endurance or strength training, the hormonal envir- onment of skeletal muscles is greatly disturbed and Contractile properties of muscle fibers these alterations are powerful signals able to trigger changes in myosin expression in exercising muscles. The existence of numerous types of fibers makes skeletal muscle a very heterogeneous tissue able to Effects of testosterone perform various functional capabilities. Immuno- histochemical and biochemical analysis of skeletal In some animal studies, a fast-to-slow shift of MyHC muscles revealed that this diversity in muscle fiber isoforms has been observed following androgenic- types reflects a broad spectrum of myosin isoforms. anabolic steroid treatment (Fritzsche et al. 1994; The myosin is the molecule primarily responsible, Czesla et al. 1997). An increase in fibers containing along with actin, for muscle contraction. The myosin MyHC IIA and a decrease in fibers containing molecule is made up of two heavy chains (MyHC) MyHC IIB has been reported in several skeletal and four light chains (MyLC) (Schiaffino & Reggiani muscles in rodents in response to androgenic- 1996; Pette & Staron 1997). The heavy chain portion anabolic steroid administration (Egginton 1987; hormones, skeletal muscle and exercise 313

Dimauro et al. 1992). In contrast, an androgenic Effects of growth hormone steroid-induced decrease in MyHC IIA-containing fibers in favor of MyHC IIB-containing fibers has It has been reported that growth hormone admin- also been reported (Kelly et al. 1985; Lyons et al. istration induced an increase in MyHC IIX in the 1986; Salmons 1992). These results suggest that the vastus lateralis of healthy elderly men (Lange effects of androgenic-anabolic steroids on the con- et al. 2002). The shift in MyHC isoforms towards tractile properties might be muscle-dependant and MyHC IIX has been interpreted as a change into a might also vary between species. In fact, there is more youthful MyHC composition as a decrease other evidence indicating that androgenic-anabolic in MyHC IIX usually accompanies aging in this steroids have no significant effects on muscle fiber muscle group (Lange et al. 2002). In contrast, it type composition. For example, muscle overload in has been shown that the amount of MyHC IIX in animal experiments induces an increase in the growth hormone-deficient patients was higher than expression of slow MyHC I, and the addition of in a normal control population (Daugaard et al. androgenic-anabolic steroids did not alter the pattern 1999). Furthermore, treatment of growth hormone- of MyHC expression (Boissonneault et al. 1987). Like- deficient patients with recombinant human growth wise, the administration of androgenic-anabolic hormone for 6 months had no effect on MyHC com- steroids did not modify the slow-to-fast shift in position (Daugaard et al. 1999). Similarly, it has been MyHC isoforms caused by hindlimb suspension ex- shown that growth hormone treatment significantly periments (Tsika et al. 1987). Finally, in well-trained increased the cross-sectional area of rat type II fibers power-lifters, there were no differences in trapezius in the soleus muscle with no significant effect on MyHC composition between androgenic-anabolic fiber type composition (Aroniadou-Anderjaska et al. steroid users and non-users (Kadi et al. 1999b). 1996). Whether increased concentration of growth hormone is associated with a slow-to-fast shift of myosin isoforms remains to be further investigated. Effects of estrogen

It is well-known that a reduction in force production Effects of thyroid hormones occurs at menopause (Greeves et al. 1999; Dionne et al. 2000; Meeuwsen et al. 2000). At the cellular Thyroid hormones exert a powerful action in the level, it has been shown that ovariectomy is asso- regulation of MyHC composition of skeletal mus- ciated with a fast-to-slow shift in MyHC isoforms cles (D’Albis & Butler-Browne 1993). Larsson & and a reduction in the level of spontaneous running Yu (1997) showed that the regulation of MyHC activity in rats (Kadi et al. 2002). MyHC changes isoforms in rat skeletal muscle by thyroid hormone tend to follow a general pathway of sequential is gender and muscle specific. Administration of ← ← ← transition in the order MyHC I IIA IIX IIB. 3.5.3’-triiodothyronine (T3) causes a down-regula- This result can be interpreted as an overall transi- tion of MyHC I and up-regulation of MyHC IIA in tion in the expression of fast isoforms towards male and female soleus, whereas the up-regulation slower isoforms, a specific up-regulation of the of MyHC IIX is observed only in male muscles slower isoforms of MyHC genes, or a specific (Larsson & Yu 1997). Treatment with T3 induces down-regulation of genes coding for fast MyHC no changes in MyHC isoforms in male extensor dig- isoforms following ovariectomy. When ovariec- itorum longus (EDL). The same treatment induces a tomized animals are allowed to run and are treated significant transition from MyHC IIA to MyHC IIB in with estrogen, there were no alterations in MyHC female EDL (Larsson & Yu 1997). Altogether, these composition (Kadi et al. 2002). Therefore, it can results show that the contractile properties of skeletal be suggested that physical activity together with muscles are under the control of several hormones estrogen treatment may help to maintain muscle and growth factors, and that changes in the hormonal characteristics of both slow and fast muscles. environment of skeletal muscles during exercise are 314 chapter 22

partly responsible for the adjustment of muscle of muscle physiology has just begun to develop phenotype to the physiological demand. It is now and there is still much to be discovered in order to becoming clear that changes in muscle structure and better understand the different signals involved in function in response to changes in the hormonal the various adaptive events occurring in skeletal environment can be muscle and gender dependent. muscles in response to different forms of physical activities. The delineation of the different steps of Conclusion the adaptation of muscles to exercise would provide the basis for the conception of individualized exer- This paper outlined only a few aspects related to cise prescriptions to optimize the quality of train- role of specific hormones and growth factors in the ing programs both in well-trained and sedentary regulation of some important muscular parameters healthy populations as well as in populations suf- responsible for the athletic performance. This area fering from specific diseases.

References

Adams, G.R. & Haddad, F. (1996) The Antonio, J. & Gonyea, W.J. (1993) Skeletal Bischoff, R., 1994. The satellite cell and relationships among IGF-1, DNA muscle fiber hyperplasia. Medicine and muscle regeneration. In: Myology (Engel, content, and protein accumulation Science in Sports and Exercise 25, A.G., Franzini-Armstrong, C., eds.). during skeletal muscle hypertrophy. 1333–1345. McGraw-Hill, New York: 97–117. Journal of Applied Physiology 81, Aroniadou-Anderjaska, V., Lemon, P.W., Boissonneault, G., Gagnon, J., Ho-Kim, 2509–2516. Gilloteaux, J. (1996) Effects of exogenous M.A. & Tremblay, R.R. (1987) Lack of Adams, G.R., Hather, B.M., Baldwin, K.M. growth hormone on skeletal muscle of effect of anabolic steroids on specific & Dudley, G.A. (1993) Skeletal muscle young female rats. Tissue and Cell 28, mRNAs of skeletal muscle undergoing myosin heavy chain composition and 719–724. compensatory hypertrophy. Molecular resistance training. Journal of Applied Baker, J., Liu, J.P., Robertson, E.J. & and Cellular Endocrinology 51, 19–24. Physiology 74, 911–915. Efstratiadis, A. (1993) Role of insulin- Bottinelli, R., Betto, R., Schiaffino, S. & Allen, D.L., Monke, S.R., Talmadge, R.J., like growth factors in embryonic and Reggiani, C. (1994a) Maximum Roy, R.R. & Edgerton, V.R. (1995) postnatal growth. Cell 75, 73–82. shortening velocity and coexistence of Plasticity of myonuclear number Bamman, M.M., Shipp, J.R., Jiang, J. myosin heavy chain isoforms in single in hypertrophied and atrophied et al. (2001) Mechanical load increases skinned fast fibres of rat skeletal muscle. mammalian skeletal muscle fibers. muscle IGF-I and androgen receptor Journal of Muscle Research and Cell Journal of Applied Physiology 78, mRNA concentrations in humans. Motility 15, 413–419. 1969–1976. American Journal of Physiology 280, Bottinelli, R., Betto, R., Schiaffino, S. Allen, D.L., Yasui, W., Tanaka, T. et al. 383–390. & Reggiani, C. (1994b) Unloaded (1996) Myonuclear number and myosin Baumann, H., Jaggi, M., Soland, F., shortening velocity and myosin heavy heavy chain expression in rat soleus Howald, H. & Schaub, M.C. (1987) chain and alkali light chain isoform single muscle fibers after spaceflight. Exercise training induces transitions composition in rat skeletal muscle fibres. Journal of Applied Physiology 81, of myosin isoform subunits within Journal of Physiology 478, 341–349. 145–151. histochemically typed human muscle Brahm, H., Piehl-Aulin, K., Saltin, B. & Allen, D.L., Roy, R.R. & Edgerton, V.R. fibres. Pflugers Archiv 409, 349–360. Ljunghall, S. (1997) Net fluxes over (1999) Myonuclear domains in muscle Bhasin, S., Storer, T.W., Berman, N. et al. working thigh of hormones, growth adaptation and disease. Muscle and Nerve (1996) The effects of supraphysiologic factors and biomarkers of bone 22, 1350–1360. doses of testosterone on muscle size and metabolism during short lasting Andersen, J.L. & Aagaard, P. (2000) strength in normal men. New England dynamic exercise. Calcified Tissue Myosin heavy chain IIX overshoot in Journal of Medicine 335, 1–7. International 60, 175–180. human skeletal muscle. Muscle and Nerve Biolo, G., Fleming, R.Y., Maggi, S.P. & Bricout, V., Germain, P., Serrurier, B. 23, 1095–1104. Wolfe, R.R. (1995) Transmembrane & Guezennec, C. (1994) Changes in Andersen, J.L., Klitgaard, H. & Saltin, B. transport and intracellular kinetics of testosterone muscle receptors: effects of (1994) Myosin heavy chain isoforms in amino acids in human skeletal muscle. an androgen treatment on physically single fibres from m. vastus lateralis of American Journal of Physiology 268, trained rats. Cellular and Molecular sprinters: influence of training. Acta E75–E84. Biology 40, 291–294. Physiologica Scandinavica 151, 135–142. Biolo, G., Tipton, K.D., Klein, S. & Wolfe, Cabric, M. & James, N.T. (1983) Andersen, P. & Henriksson, J. (1977) R.R. (1997) An abundant supply of Morphometric analyses on the muscles Capillary supply of the quadriceps amino acids enhances the metabolic of exercise trained and untrained dogs. femoris muscle of man: adaptive effect of exercise on muscle protein. American Journal of Anatomy 166, response to exercise. Journal of Physiology American Journal of Physiology 273, 359–368. 270, 677–690. E122–E129. Campion, D.R. (1984) The muscle satellite hormones, skeletal muscle and exercise 315

cell: a review. International Review of Dionne, I.J., Kinaman, K.A. & Poehlman, Galavazi, G. & Szirmai, J.A. (1971) Cytology 87, 225–251. E.T. (2000) Sarcopenia and muscle Cytomorphometry of skeletal muscle: Charifi, N., Kadi, F., Feasson, L. & Denis, C. function during menopause and the influence of age and testosterone on (2003a) Effects of endurance training on hormone-replacement therapy. Journal of the rat m. levator ani. Zeitschrift fur satellite cell frequency in skeletal muscle Nutrition, Health and Aging 4(3), 156–161. Zellforschung und Mikroskopische of old men. Muscle and Nerve 28, 87–92. Dodson, M.V., Allen, R.E. & Hossner, K.L. Anatomie 121, 507–530. Charifi, N., Kadi, F., Feasson, L. et al. (1985) Ovine somatomedin, Gerber, H.P., Condorelli, F., Park, J. (2003b) Enhancement of microvessel multiplication-stimulating activity, and & Ferrara, N. (1997) Differential tortuosity in the vastus lateralis muscle insulin promote skeletal muscle satellite transcriptional regulation of the two of old men in response to endurance cell proliferation in vitro. Endocrinology vascular endothelial growth factor training. Journal of Physiology 554, 559–569. 117, 2357–2363. receptor genes. Flt-1, but not Flk-1/ Cheek, D.B. (1985) The control of cell Dorlochter, M., Astrow, S.H. & Herrera, KDR, is up-regulated by hypoxia. Journal mass and replication. The DNA unit: A.A. (1994) Effects of testosterone on of Biological Chemistry 272, 23 659–23 667. a personal 20-year study. Early Human a sexually dimorphic frog muscle: Giddings, C.J. & Gonyea, W.J. (1992) Development 12, 211–239. repeated in vivo observations and Morphological observations supporting Chesley, A., MacDougall J.D., androgen receptor distribution. Journal muscle fiber hyperplasia following Tarnopolsky, M.A., Atkinson, S.A. & of Neurobiology 25, 897–916. weight-lifting exercise in cats. Smith, K. (1992) Changes in human Doumit, M.E., Cook, D.R. & Merkel, R.A. Anatomical Record 233, 178–195. muscle protein synthesis after resistance (1996) Testosterone up-regulates Goldberg, A.L., Etlinger, J.D., Goldspink, exercise. Journal of Applied Physiology 73, androgen receptors and decreases D.F. & Jablecki, C. (1975) Mechanism of 1383–1388. differentiation of porcine myogenic work-induced hypertrophy of skeletal Clarke, M.S. & Feeback, D.L. (1996) satellite cells in vitro. Endocrinology 137, muscle. Medicine and Science in Sports 7, Mechanical load induces sarcoplasmic 1385–1394. 185–198. wounding and FGF release in Egginton, S. (1987) Effects of an anabolic Goldspink, D.F., Cox, V.M., Smith, S.K. differentiated human skeletal muscle hormone on striated muscle growth et al. (1995) Muscle growth in response cultures. FASEB Journal 10, 502–509. and performance. Pflugers Archiv 410, to mechanical stimuli. American Journal Czesla, M., Mehlhorn, G., Fritzsche, D. & 349–355. of Physiology 268, 288–297. Asmussen, G. (1997) Cardiomyoplasty: Fernandez, A.M., Dupont, J., Farrar, R.P. Goldspink, G. (1999) Changes in muscle improvement of muscle fibre type et al. (2002) Muscle-specific inactivation mass and phenotype and the expression transformation by an anabolic steroid of the IGF-I receptor induces of autocrine and systemic growth factors (metenolone). Journal of Molecular and compensatory hyperplasia in skeletal by muscle in response to stretch and Cellular Cardiology 29, 2989–2996. muscle. Journal of Clinical Investigation overload. Journal of Anatomy 194, D’albis, A. & Butler-Browne, G. (1993) The 109, 347–355. 323–334. hormonal control of myosin isoform Florini, J.R., Ewton, D.Z. & Roof, S.L. Goldspink, G., Scutt, A., Martindale, J. expression in skeletal muscle of (1991) Insulin-like growth factor-I et al. (1991) Stretch and force generation mammals: a review. Basic and Applied stimulates terminal myogenic induce rapid hypertrophy and myosin Myology 3(1), 7–16. differentiation by induction of isoform gene switching in adult skeletal Daugaard, J.R., Bramnert, M., Manhem, P. myogenin gene expression. Molecular muscle. Biochemical Society Transactions et al. (1999) Effect of 6 months of GH Endocrinology (Baltimore, Md.) 5, 19, 368–373. treatment on myosin heavy chain 718–724. Greeves, J.P., Cable, N.T., Reilly, T. composition in GH-deficient patients. Floss, T., Arnold, H.H. & Braun, T. (1997) & Kingsland, C. (1999) Changes in European Journal of Endocrinology 141, A role for FGF-6 in skeletal muscle muscle strength in women following the 342–349. regeneration. Genes and Development 11, menopause: a longitudinal assessment DeChiara, T.M., Efstratiadis, A. & 2040–2051. of the efficacy of hormone replacement Robertson, E.J. (1990) A growth- Frederickson, R. & Sonenberg, N., 1993. therapy. Clinical Science (London, deficiency phenotype in heterozygous eIF-4E phosphorylation and the England) 97(1), 79–84. mice carrying an insulin-like growth regulation of protein synthesis. Gustafsson, T., Puntschart, A., Kaijser, L., factor II gene disrupted by targeting. In: Translational Regulation of Gene Jansson, E. & Sundberg, C.J. (1999) Nature 345, 78–80. Expression (Ilan, J., ed.). Plenum Press, Exercise-induced expression of Deschenes, M.R., Maresh, C.M., New York: 143–162. angiogenesis-related transcription and Armstrong, L.E. et al. (1994) Endurance Fritzsche, D., Krakor, R., Asmussen, G. growth factors in human skeletal and resistance exercise induce muscle et al. (1994) Effect of an anabolic steroid muscle. American Journal of Physiology fiber type specific responses in androgen (Metenolon) on contractile performance 276, 679–685. binding capacity. Journal of Steroid of the chronically stimulated latissimus Gustafsson, T., Knutsson, A., Puntschart, Biochemistry and Molecular Biology 50, dorsi in sheep. European Journal of Cardio- A. et al. (2002) Increased expression of 175–179. thoracic Surgery 8, 214–219. vascular endothelial growth factor in Dimauro, J., Balnave, R.J. & Shorey, C.D. Fry, A.C., Allemeier, C.A. & Staron, R.S. human skeletal muscle in response to (1992) Effects of anabolic steroids and (1994) Correlation between percentage short-term one-legged exercise training. high intensity exercise on rat skeletal fiber type area and myosin heavy chain Pflugers Archiv 444, 752–759. muscle fibres and capillarization. A content in human skeletal muscle. Hall, Z.W. & Ralston, E. (1989) Nuclear morphometric study. European Journal European Journal of Applied Physiology 68, domains in muscle cells. Cell 59, of Applied Physiology 64, 204–212. 246–251. 771–772. 316 chapter 22

Hameed, M., Harridge, S.D. & Goldspink, regenerating muscle of normal and ovariectomy. Journal of Muscle Research G. (2002) Sarcopenia and hypertrophy: hypophysectomized rats. Acta and Cell Motility 23, 335–339. a role for insulin-like growth factor-1 in Physiologica Scandinavica 146, 79–86. Kanda, F., Takatani, K., Okuda, S., aged muscle? Exercise and Sport Sciences Jennische, E., Skottner, A. & Hansson, H.A. Matsushita, T. & Chihara, K. (1999) Reviews 30, 15–19. (1987) Satellite cells express the trophic Preventive effects of insulin-like growth Hameed, M., Orrell, R.W., Cobbold, M., factor IGF-I in regenerating skeletal factor-I on steroid-induced muscle Goldspink, G. & Harridge, S.D. (2003) muscle. Acta Physiologica Scandinavica atrophy. Muscle and Nerve 22, 213–217. Expression of IGF-I splice variants in 129, 9–15. Kelly, A., Lyons, G., Gambki, B. & young and old human skeletal muscle Joubert, Y. & Tobin, C. (1989) Satellite cell Rubinstein, N. (1985) Influences of after high resistance exercise. Journal of proliferation and increase in the number testosterone on contractile proteins Physiology 547, 247–254. of myonuclei induced by testosterone of the guinea pig temporalis muscle. Hellsten, Y., Hansson, H.A., Johnson, L., in the levator ani muscle of the adult Advances in Experimental Medicine and Frandsen, U. & Sjodin, B. (1996) female rat. Developmental Biology 131, Biology 182, 155–168. Increased expression of xanthine 550–557. Kennedy, J.M., Eisenberg, B.R., Reid, S.K., oxidase and insulin-like growth factor I Joubert, Y. & Tobin, C. (1995) Testosterone Sweeney, L.J. & Zak, R. (1988) Nascent (IGF-I) immunoreactivity in skeletal treatment results in quiescent satellite muscle fiber appearance in overloaded muscle after strenuous exercise in cells being activated and recruited into chicken slow-tonic muscle. American humans. Acta Physiologica Scandinavica cell cycle in rat levator ani muscle. Journal of Anatomy 181, 203–215. 157, 191–197. Developmental Biology 169, 286–294. Kimura, N., Mizokami, A., Oonuma, T., Hickson, R.C. & Kurowski, T.G. (1986) Kadi, F. (2000) Adaptation of human Sasano, H. & Nagura, H. (1993) Anabolic steroids and training. Clinics skeletal muscle to training and anabolic Immunocytochemical localization of in Sports Medicine 5, 461–469. steroids. Acta Physiologica Scandinavica androgen receptor with polyclonal Hikida, R.S., Van-Nostran, S., Murray, J.D. 646, 1–52. antibody in paraffin-embedded human et al. (1997) Myonuclear loss in atrophied Kadi, F. & Thornell, L.E. (1999) Training tissues [see comments]. Journal of soleus muscle fibers. Anatomical Record affects myosin heavy chain phenotype Histochemistry and Cytochemistry. 41, 247, 350–354. in the trapezius muscle of women. 671–678. Hikida, R.S., Walsh, S., Barylski, N. et al. Histochemistry and Cell Biology 112, Kochakian, C.D. & Tillotson, C. (1957) (1998) Is hypertrophy limited in elderly 73–78. Influence of several C19-steroids on the muscle fibers? A comparison of elderly Kadi, F. & Thornell, L.E. (2000) growth of individual muscles of the and young strength-trained men. Basic Concomitant increases in myonuclear guinea pig. Endocrinology 60, 607–618. and Applied Myology 8(6), 419–427. and satellite cell content in female Kraemer, W.J., Patton, J.F., Gordon, S.E. Hudlicka, O., Brown, M. & Egginton, S. trapezius muscle following strength et al. (1995) Compatibility of high- (1992) Angiogenesis in skeletal and training. Histochemistry and Cell Biology intensity strength and endurance cardiac muscle. Physiological Reviews 72, 113, 99–103. training on hormonal and skeletal 369–417. Kadi, F., Eriksson, A., Holmner, S., Butler- muscle adaptations. Journal of Applied Inoue, K., Yamasaki, S., Fushiki, T. et al. Browne, G.S. & Thornell, L.E. (1999a) Physiology 78, 976–989. (1993) Rapid increase in the number of Cellular adaptation of the trapezius Krieg, M. (1976) Characterization of the androgen receptors following electrical muscle in strength-trained athletes. androgen receptor in the skeletal muscle stimulation of the rat muscle. European Histochemistry and Cell Biology 111, of the rat. Steroids 28, 261–274. Journal of Applied Physiology 66, 134–140. 189–195. Lange, K.H., Andersen, J.L., Beyer, N. Inoue, K., Yamasaki, S., Fushiki, T., Okada, Kadi, F., Eriksson, A., Holmner, S. & et al. (2002) GH administration changes Y. & Sugimoto, E. (1994) Androgen Thornell, L.-E. (1999b) Effects of anabolic myosin heavy chain isoforms in skeletal receptor antagonist suppresses exercise- steroids on the muscle cells of strength- muscle but does not augment muscle induced hypertrophy of skeletal muscle. trained athletes. Medicine and Science in strength or hypertrophy, either alone European Journal of Applied Physiology 69, Sports and Exercise 31, 1528–1534. or combined with resistance exercise 88–91. Kadi, F., Ahlgren, C., Waling, K., Sundelin, training in healthy elderly men. Journal Janssen, P.J., Brinkmann, A.O., Boersma, G. & Thornell, L.E. (2000a) The effects of Clinical Endocrinology and Metabolism W.J. & Van-Der-Kwast, T.H. (1994) of different training programs on the 87, 513–523. Immunohistochemical detection of the trapezius muscle of women with work- Larsson, L. & Yu, F. (1997) Gender-related androgen receptor with monoclonal related neck and shoulder myalgia. Acta differences in the regulatory influence antibody F39.4 in routinely processed, Neuropathologica 100, 253–258. of thyroid hormone on the expression of paraffin-embedded human tissues after Kadi, F., Bonnerud, P., Eriksson, A. & myosin isoforms in young and old rats. microwave pre-treatment. Journal of Thornell, L.-E. (2000b) The expression of Acta Physiologica Scandinavica 159, 81–89. Histochemistry and Cytochemistry 42, androgen receptors in human neck and Laurent, G., Sparrow, M. & Millward, D. 1169–1175. limb muscles: effects of training and self- (1978) Turnover of muscle protein in Jennische, E. (1989) Sequential administration of androgenic-anabolic the fowl. Changes in rates of protein immunohistochemical expression of steroids. Histochemistry and Cell Biology synthesis and beackdown during IGF-I and the transferrin receptor in 113, 25–29. hypertrophy of the anterior and regenerating rat muscle in vivo. Acta Kadi, F., Karlsson, C., Larsson, B. et al. posterior latissimus dorsi muscle. Endocrinologica 121, 733–738. (2002) The effects of physical activity Biochemical Journal 176, 407–417. Jennische, E. & Matejka, G.L. (1992) IGF-I and estrogen treatment on rat fast and Lin, M.C., Rajfer, J., Swerdloff, R.S. & binding and IGF-I expression in slow skeletal muscles following Gonzalez-Cadavid, N.F. (1993) hormones, skeletal muscle and exercise 317

Testosterone down-regulates the levels Pette, D. & Staron, R.S. (1997) Mammalian Series A, Biological Sciences and Medical of androgen receptor mRNA in smooth skeletal muscle fiber type transitions. Sciences 56, 240–247. muscle cells from the rat corpora International Review of Cytology 170, Roy, R., Monke, S., Allen, D. & Edgerton, cavernosa via aromatization to 143–223. V. (1999) Modulation of myonuclear estrogens. Journal of Steroid Biochemistry Phillips, S.M., Tipton, K.D., Aarsland, number in functionally overloaded and and Molecular Biology 45, 333–343. A., Wolf, S.E. & Wolfe, R.R. (1997) exercised rat plantaris fibers. Journal of Luke, M. & Coffey, D., 1994. The male Mixed muscle protein synthesis and Applied Physiology 87, 634–642. accessory tissues: structure, androgen, breakdown after resistance exercise in Ruizeveld-De-Winter, J.A., Trapman, J., and physiology. In: The Physiology of humans. American Journal of Physiology Vermey, M. et al. (1991) Androgen Reproduction (Knobil, E. & Neill, J., 273, E99–E107. receptor expression in human tissues: eds.). Raven Press, New York: Powers, M.L. & Florini, J.R. (1975) A direct an immunohistochemical study. Journal 1435–1488. effect of testosterone on muscle cells of Histochemistry and Cytochemistry 39, Lyons, G.E., Kelly, A.M. & Rubinstein, in tissue culture. Endocrinology 97, 927–936. N.A. (1986) Testosterone-induced 1043–1047. Salmons, S. (1992) Myotrophic effects of an changes in contractile protein isoforms Quinn, L.S., Steinmetz, B., Maas, A., anabolic steroid in rabbit limb muscles. in the sexually dimorphic temporalis Ong, L. & Kaleko, M. (1994) Type-1 Muscle & Nerve 15, 806–812. muscle of the guinea pig. Journal of insulin-like growth factor receptor Sar, M., Lubahn, D.B., French, Biological Chemistry 261, 13 278–13 284. overexpression produces dual effects F.S. & Wilson, E.M. (1990) McCall, G., Allen, D., Linderman, J. et al. on myoblast proliferation and Immunohistochemical localization of (1998) Maintenance of myonuclear differentiation. Journal of Cellular the androgen receptor in rat and human domain size in rat soleus after overload Physiology 159, 387–398. tissues. Endocrinology 127, 3180–3186. and growth hormone/IGF-I treatment. Regnier, M. & Herrera, A.A. (1993a) Sassoon, D. & Kelley, D.B. (1986) The Journal of Applied Physiology 84, Changes in contractile properties sexually dimorphic larynx of Xenopus 1407–1412. by androgen hormones in sexually laevis: development and androgen McCormick, K.M. & Schultz, E. (1992) dimorphic muscles of male frogs regulation. American Journal of Anatomy Mechanisms of nascent fiber formation (Xenopus laevis). Journal of Physiology 461, 177, 457–472. during avian skeletal muscle 565–581. Schaub, M.C., Brunner. U.T., Von- hypertrophy. Developmental Biology 150, Regnier, M. & Herrera, A.A. (1993b) Schulthess, C., Neidhart, M. & 319–334. Differential sensitivity to androgens Baumann, H. (1989) Adaptation of McKoy, G., Ashley, W., Mander, J. et al. within a sexually dimorphic muscle of contractile proteins in human heart and (1999) Expression of insulin growth male frogs (Xenopus laevis). Journal of skeletal muscles. Biomedica Biochimica factor-1 splice variants and structural Neurobiology 24, 1215–1228. Acta 48, S306–S312. genes in rabbit skeletal muscle induced Richardson, R.S., Wagner, H., Mudaliar, Schiaffino, S. & Reggiani, C. (1996) by stretch and stimulation. Journal of S.R. et al. (1999) Human VEGF gene Molecular diversity of myofibrillar Physiology 516, 583–592. expression in skeletal muscle: effect of proteins: gene regulation and functional Mauro, A. (1961) Satellite cell of skeletal acute normoxic and hypoxic exercise. significance. Physiological Reviews 76, muscle fibers. Journal of Biophysical and American Journal of Physiology 277, 371–423. Biochemical Cytology 9, 493–495. 2247–2252. Schiaffino, S., Bormioli, S.P. & Aloisi, M. Max, S.R., Mufti, S. & Carlson, B.M. Richardson, R.S., Wagner, H., Mudaliar, (1976) The fate of newly formed satellite (1981) Cytosolic androgen receptor in S.R. et al. (2000) Exercise adaptation cells during compensatory muscle regenerating rat levator ani muscle. attenuates VEGF gene expression in hypertrophy. Virchows Archiv. B: Cell Biochemical Journal 200, 77–82. human skeletal muscle. American Journal Pathology 21, 113–118. Meeuwsen, I.B., Samson, M.M. & of Physiology 279, 772–778. Schultz, E. & McCormick, K.M. (1994) Verhaa, H.J. (2000) Evaluation of the Rogozkin, V. (1979) Metabolic effects of Skeletal muscle satellite cells. Reviews of applicability of HRT as a preservative anabolic steroid on skeletal muscle. Physiology, Biochemistry and Pharmacology of muscle strength in women. Maturitas Medicine and Science in Sports 11, 123, 213–257. 36(1), 49–61. 160–163. Staron, R.S., Leonardi, M.J., Karapondo, Mitchell, P., Steenstrup, T. & Hannon, K. Rosenblatt, J.D. & Parry, D.J. (1992) D.L. et al. (1991) Strength and skeletal (1999) Expression of fibroblast growth Gamma irradiation prevents muscle adaptations in heavy-resistance- factor family during postnatal skeletal compensatory hypertrophy of trained women after detraining and muscle hypertrophy. Journal of Applied overloaded mouse extensor digitorum retraining. Journal of Applied Physiology Physiology 86, 313–319. longus muscle. Journal of Applied 70, 631–640. Moss, F.P. & Leblond, C.P. (1971) Satellite Physiology 73, 2538–2543. Takagi, H., King, G.L., Robinson, G.S., cells as the source of nuclei in muscles Rosenblatt, J.D., Yong, D. & Parry, D.J. Ferrara, N. & Aiello, L.P. (1996) of growing rats. Anatomical Record 170, (1994) Satellite cell activity is required Adenosine mediates hypoxic induction 421–435. for hypertrophy of overloaded adult of vascular endothelial growth factor in Olfert, I.M., Breen, E.C., Mathieu-Costello, rat muscle. Muscle and Nerve 17, 608–613. retinal pericytes and endothelial cells. O. & Wagner, P.D. (2001) Chronic Roth, S.M., Martel, G.F., Ivey, F.M. et al. Investigative Ophthalmology and Visual hypoxia attenuates resting and exercise- (2001) Skeletal muscle satellite cell Science 37, 2165–2176. induced VEGF, flt-1, and flk-1 mRNA characteristics in young and older men Takeda, H., Chodak, G., Mutchnik, S., levels in skeletal muscle. Journal of and women after heavy resistance Nakamoto, T. & Chang, C. (1990) Applied Physiology 90, 1532–1538. strength training. Journals of Gerontology. Immunohistochemical localization of 318 chapter 22

androgen receptors with mono- and of women after resistance training- Yamada, S., Buffinger, N., Dimario, J. & polyclonal antibodies to androgen quantitative ultrastructure and enzyme Strohman, R.C. (1989) Fibroblast growth receptor. Journal of Endocrinology 126, activity. Pflugers Archiv 424, 494–502. factor is stored in fiber extracellular 17–25. Willis, P.E., Chadan, S., Baracos, V. & matrix and plays a role in regulating Tsika, R.W., Herrick, R.E. & Baldwin, K.M. Parkhouse, W.S. (1997) Acute exercise muscle hypertrophy. Medicine and (1987) Effect of anabolic steroids on attenuates age-associated resistance to Science in Sports and Exercise 21, skeletal muscle mass during hindlimb insulin-like growth factor I. American S173–S180. suspension. Journal of Applied Physiology Journal of Physiology 272, 397–404. Yamaguchi, T.P. & Rossant, J. (1995) 63, 2122–2127. Willis, P.E., Chadan, S.G., Baracos, V. & Fibroblast growth factors in mammalian Wada, H., Ivester, C., Carabello, B., Parkhouse, W.S. (1998) Restoration of development. Current Opinion in Genetics Cooper, G. & Mcdermott, P. (1996) insulin-like growth factor I action in and Development 5, 485–491. Translational initiation factor eIF-4E, a skeletal muscle of old mice. American Yan, Z., Biggs, R.B. & Booth, F.W. link between cardiac load and protein Journal of Physiology 275, 525–530. (1993) Insulin-like growth factor synthesis. Journal of Biological Chemistry Winchester, P.K. & Gonyea, W.J. (1992) A immunoreactivity increases in muscle 271, 8359–8364. quantitative study of satellite cells and after acute eccentric contractions. Journal Wagner, P.D., Masanes, F., Wagner, H. myonuclei in stretched avian slow tonic of Applied Physiology 74, 410–414. et al. (2001) Muscle angiogenic growth muscle. Anatomical Record 232, 369–377. Yang, S., Alnaqeeb, M., Simpson, H. & factor gene responses to exercise in Wong, T.S. & Booth, F.W. (1990) Protein Goldspink, G. (1996) Cloning and chronic renal failure. American Journal metabolism in rat gastrocnemius muscle characterization of an IGF-1 isoform of Physiology 281, 539–546. after stimulated chronic concentric expressed in skeletal muscle subjected to Wang, N., Hikida, R.S., Staron, R.S. & exercise. Journal of Applied Physiology 69, stretch. Journal of Muscle Research and Cell Simoneau, J.A. (1993) Muscle fiber types 1709–1717. Motility 17, 487–495. Chapter 23

Resistance Exercise and Testosterone

ATKO VIRU AND MEHIS VIRU

The male sex hormone, testosterone, influences not testicular testosterone. Thus, the pituitary gonado- only sexual activity and emotional behavior (e.g. graphs secreting LH and the testes constitute the aggressiveness) but also contributes to metabolic pituitary–testicular system, and its activity is con- control. Testosterone is considered to be an anabolic trolled by nervous influences reaching the pituitary hormone; however, its role in metabolic control is gland from the hypothalamus via neurosecretion actually more extended. Testosterone influences by and by the feedback action of blood testosterone. its contra-action effects on several other hormones. Both LH-releasing hormone and LH are secreted In resistance training, the main role of testosterone in an episodic manner. Secretory bursts of GnRH is the induction of synthesis of contractile pro- last a few minutes at a time, once every 1–3 h. The teins in involved muscles. Beside that, during acute pulsative manner of testosterone is less pronounced. resistance exercises, as well as during competition, The blood testosterone level is highest early in the testosterone action seems to be essential for mobiliz- morning and decreases during the day. Most of ing performance capacity. the testosterone in blood is bound by sex hormone binding globulin (SHBG). The pituitary–testicular system Metabolic action of testosterone Testosterone is produced by the interstitial cells of Leydig, constituting of 20% of the mass of testes in Anabolic action. In 1935 Kochakian reported that, in adult men. Quantitatively, a much lesser amount of male castrated dogs, testosterone injection resulted testosterone is derivated from androgenic steroids in long-lasting nitrogen retention (Kochakian 1935). formed in the adrenal cortex. The Leydig cells Over the next few years, several papers confirmed secrete testosterone only when they are stimulated this result and showed that castration of male rats, by lutropin (LH; also called luteinizing hormone) guinea pigs or mice induces weakened skeletal from the pituitary gland. The quantity of secreted muscles due to a decreased protein synthesis rate. testosterone is in direct proportion to the amount Substitution therapy with testosterone abolished of LH available. The secretion of LH is stimulated these castration effects (Papanicolaou & Falk 1938; by a hypothalamic neurohormone, gonadotropin- Scow & Hayes 1955; Kochakian 1959; Kochakian releasing hormone (GnRH). In turn, blood testo- et al. 1964). sterone inhibits GnRH secretion (a pronounced According to present evidence, metabolic action negative feedback effect) as well as LH secretion by of testosterone is mediated through the androgen the pituitary (a weak negative feedback effect). Too receptor in the cytoplasm of cell. In the cell cyto- little testosterone allows the neurosecretory cells of plasm testosterone is bound by the receptor protein. thalamus to secrete large amounts of GnRH with The formed complex is activated and transferred followed increases in secretion of pituitary LH and into the cell nucleus. Binding of the complex by 319 320 chapter 23

chromosomal proteins triggers the production of of protein synthesis (anti-anabolic action) exerted mRNA specific for the synthesis of necessary pro- by them. Anti-catabolic action and the inhibition of tein(s) responsible for actualization of the testo- anti-anabolic action are founded on the competition sterone effect. In the skeletal muscle the main locus between testosterone and cortisol for the specific of testosterone action is the synthesis of myofibril- cellular receptors of glucocorticoids. Depending on lar proteins. This way testosterone contributes to the amount of glucocorticoid receptors occupied by the development of muscular hypertrophy. The testosterone, the cortisol catabolic and anti-anabolic receptor protein is common for testosterone and actions decrease (Mayer & Rosen 1977). several other androgens. In skeletal muscle fibers, as well as in bone cells, testosterone has a higher Stimulation of bone growth and calcium retention. affinity to receptor than other androgens. In other Testosterone increases the total bone matrix and cells the highest affinity belongs to 5α-dihydro- causes calcium retention. After puberty, bones grow testosterone. In these cases, after testosterone enters considerably in thickness and deposit additional the cell it has to be converted to 5α-dihydrotesto- amounts of calcium salts (Ritzen et al. 1981; Krabbe sterone. Conversion of testosterone to dihydrotesto- et al. 1982). The increase in bone matrix has been sterone is not found in muscle fibers and bone cells. related to the anabolic function of testosterone, and The information contained in the mRNA is trans- the deposition of calcium salts is secondarily caused lated for the synthesis of related cellular proteins in by the increase of bone matrix. ribosomes. The state of cellular receptors is regulated by Other effects. During adolescence and early adult- increasing/decreasing the number of binding sites hood, testosterone increases the basal metabolism and by changing the binding affinity. Thus, testoster- 5–10%. Testosterone is capable of increasing the rate one metabolism depends on testosterone produc- of erythropoiesis (Palacios et al. 1983). It also exerts a tion as well as the up- or down-regulation of cellular modest influence on sodium reabsorption in renal specific receptor. Therefore, the metabolic effect of tubules. the hormone augments not only by increased con- centration of the hormone in the intercellular fluid Testosterone in women but also as a result of the increased number of bind- ing sites and/or increased affinity of receptor to hor- Testosterone in female blood plasma originates mone. In conditions of receptor down-regulation, mainly from the adrenal cortex as a byproduct of glu- the high hormone level cannot produce a pronounced cocorticoid biosynthesis. The secretion of adrenal metabolic effect. Consequently, it is incorrect to put cortex contains androgenic steroids that can be the sign of equality between the blood concentration peripherally converted to testosterone. Adrenals of a hormone and its metabolic effect. begin to produce androgenic steroids at the begin- ning of the second decade of postnatal life in rela- Ontogenetic development of skeletal muscle is deeply tion to the adrenarche. The marker of this process is related to the metabolic effect of testosterone caus- the serum level of dehydroepiandrosterone sulfate ing the male body build to develop from the final (DHEAS) (Wierman & Crowley 1986). stages of puberty, and it is characterized by the The production of testosterone in women depends increase of musculature over that of the female. on the rate of biosynthesis of glucocorticoids stimu- Increased testosterone level after sexual maturation lated by adrenocorticotropic hormone (ACTH) from also warrants good faculties for strength, power and the anterior lobe of pituitary gland. Thus, in women speed training. LH has only a minor role, if at all, in the control of the blood level of testosterone. Therefore, when Anti-catabolic effects. Testosterone is capable of comparing various responses of blood testosterone inhibiting the catabolic effect of glucocortioids (anti- in men and women, the principal gender difference catabolic action) as well as reducing the suppression in the control mechanism has to borne in the mind. resistance exercise and testosterone 321

Although the blood level of testosterone in adult vascular volume, oxygen tension and glucose avail- women is about 10 times lower than in men, the ability. However, attention should have first been metabolic effects of testosterone are not so less given to the feedback by nervous impulses from pronounced. In women, testosterone metabolic receptors located in skeletal muscles (propriocep- effects increase in relation to estrogen production tors sensing the muscular tension and metaborecep- (Danhaive & Rousseau 1988). It has been suggested tors reacting to metabolite accumulation). The use that increased sensitivity in regard to testosterone of small doses of epidural anesthesia, to block the metabolic effect is related to up-regulation of andro- thin sensory afferents (mostly from metaborecep- genic receptor. Another possibility is that in women tors) and leave almost intact the thicker efferent testosterone has favored conditions for competition fibers and, subsequently, motor function, allowed for glucocorticoid receptors. the authors to demonstrate the essential role of nervous feedback from muscles for ACTH and β- Testosterone responses in resistance endorphin response (Kjær et al. 1989). exercise Significance of exercise intensity, duration and Exercise-induced hormonal responses depend on rest intervals for testosterone responses in four main determinants: (i) exercise intensity; (ii) resistance exercises exercise duration; (iii) level of adaptation to the concrete form of exercise; and (iv) homeostatic In cyclic exercises the level of intensity has been needs (Viru, A. 1992; Viru, A. et al. 1996). The action found to be significant for several hormone res- of these determinants is modified by several modu- ponses. Exercise intensities higher than the cor- lators, such as emotional strain, carbohydrate and responding threshold cause hormonal responses. oxygen availability, environmental temperature, Intensity thresholds for catecholamines (Lehmann biorhythms and fatigue (Viru, A. et al. 1996). et al. 1991), ACTH (Rahkila et al. 1988), cortisol (Port Three mechanisms are assumed to constitute the 1981), β-endorphin (Rahkila et al. 1988) and growth link between exercise and endocrine activities. One hormone (Chwalbinska-Moneta et al. 1996) are close of them is related to triggering nervous discharge to the anaerobic threshold. In men performing a from cerebral motor centers to spinal motoneurons cyclic incremental exercise test, the exercise intensity (central motor command). The importance of this has been noticed above that of the elevated blood for the activation of endocrine function has been level of testosterone (Jezˇova et al. 1985). The same shown in experiments that injected tubocurarine was indicated by the results of Galbo et al. (1977). in men. A bolus of 0.015 mg·kg–1 α-tubocurarine However, results of several other studies failed to caused a partial peripheral neuromuscular block- indicate a clear-cut dependence on exercise intensity. ade. As a result, it weakened the skeletal muscles. Taking into the consideration the significance Therefore, a stronger voluntary effort was necessary of the ‘strength’ of the central motor command, to produce a certain work output compared with most resistance and power exercises should be in- normal conditions. The increased voluntary effort tensive enough to evoke sufficiently strong central was confirmed by the higher rate of perceived exer- motor command for triggering hormone responses. tion in this experiment (Galbo et al. 1987). The Nevertheless, an instantaneous single application of ‘stronger’ central motor command was associated great force or muscle power does not mostly induce with exaggerated catecholamine, growth hormone hormonal responses. The duration of the effort or and ACTH responses during exercise performed at the number of repetitions seems to be important as a similar level of oxygen uptake without neuromus- well. Kraemer, W.J. et al. (1989, 1991b) compared cular blockade (Kjær et al. 1987). the action of cycling when 100%, 73%, 55% or 36% Galbo (1983) assumed that, during continuous of leg power was applied for pedaling. The max- exercises, hormone responses were influenced by imal duration of exercise was 6 s at 100%, 10 s at 73%, impulses from receptors sensing temperature, intra- 47 s at 55% and 3 min 31 s at 36%. In the case of the 322 chapter 23

highest exercise intensity, the 6 s of duration was suf- related, obviously, to the hemoconcentration. Galbo ficient to trigger the increase of norepinephrine in et al. (1977) interpreted the increase of testoster- blood serum but not to elevate levels of epinephrine one concentration by 13% in incremental exercise and cortisol. These results demonstrate that, in exer- as a manifestation of plasma volume reduction. cises of high power output, the triggering of hor- Kraemer, W.J. et al. (1992) also pointed to the pos- monal responses depends on the combined effect of sibility that in moderate resistance exercise sessions the rate of power output and duration of exercise. the elevated testosterone levels may be related to changes in plasma volume. Wilkerson et al. (1980) Kinetics of testosterone response. The rate of actual measured simultaneously changes of plasma vol- appearance of hormonal response is related to: (i) ume and testosterone concentration during cyclic the rate of reaching the triggering signal to the endo- exercise. Their conclusion was that in 20-min steady crine gland (the highest rate possessed to nervous state exercises the modest increase of testosterone activation of adrenal medulla; a comparatively high concentration was due to reduction of plasma vol- rate is signaling the activation of anterior lobe of ume but not due to elevated testosterone secretion. the pituitary gland with the aid of hypothalamus An additional reason for testosterone increase in neurohormones); (ii) the possibilities for immediate blood without elevated secretion is the reduced secretion of the already biosynthesized hormone metabolic clearance rate of testosterone during exer- (e.g. catecholamines bound in cytoplasmic granules cise (Sutton et al. 1978). of adrenomedullary cells); (iii) the rate of biosyn- According to these results, the actual increase of thesis of hormone in quantity necessary for increas- testosterone secretion in exercise should be a delayed ing the blood level of hormone and maintaining response. Nevertheless, Jezˇova et al. (1985) found a that level during the exercise. The kinetics of testo- significant increase of testosterone concentration sterone production has still not been investigated in (by 28%) after cycloergometric testing of 4.5 min the exercise situation. Padron et al. (1980) indicated duration performed at very high intensity (5 W·kg that a single injection of human chorinic gonadi- body weight–1). Obviously, the exercises cause a tropin in men was followed 2 h later by a prolonged rather complicated situation including the possibil- increase of testosterone level in the blood. These ity for altered kinetics of testosterone response. results suggest that the rate of secretory response of Results of Ahtiainen et al. (2003) showed that after testosterone depends on the rate of synthesis of this two sets of leg press at 12-repetition maximum (RM) hormone. Obviously, we have to assume that tes- (rest between sets 2 min) free and total testosterone tosterone secretion cannot be enhanced rapidly. concentrations were significantly higher than initial Variability of testosterone changes in exercise of values. The increased levels persisted up to the end short duration rather shows the normal fluctuations of session, whereas a trend for further elevation of of the basal level of testosterone than the actual free hormone level was observed (Fig. 23.1). Since secretory response. plasma volume decreased by 11.8% during the One-minute duration of high intensity exercise session in one protocol (maximum repetition) and (consecutive vertical jumps repeated at maximal by 14.4% in the other protocol (‘forced’ repetition), rate) evoked increases of concentration of anterior free testosterone response surpassed the hemo- pituitary and thyroid hormones 20–36%, but no concentration effect after first two sets and total change of growth hormone, prolactin and insulin- testosterone response after six to eight sets. Another like growth factor concentrations. Total and free study of this research team demonstrated that four testosterone and cortisol level elevated only 12–14% sets of squats of 10-RM (rest pauses 90 s) resulted (Bosco et al. 1996a). The magnitude of changes of in increases of concentrations of free and total testosterone and cortisol was similar to the reduc- testosterone as well as of ACTH, cortisol and lactate tion of plasma volume in short-term highly intense (Kraemer, W.J. et al. 1998a). However, again the con- cyclic exercises (Sejersted et al. 1986). Therefore, the tribution of the hemoconcentration effect remains. observed testosterone and cortisol changes were All in all it is necessary to bear in mind that the resistance exercise and testosterone 323

Free testosterone 40

35

30

25

20

Percent 15

10

5

0 Pre 2 sets 4 sets 2 sets 2 sets Post- Post- (a) 15 min 30 min

Testosterone 20 15 10 5 0 Pre 2 sets 4 sets 2 sets 2 sets Post- Post- –5

Percent 15 min 30 min Fig. 23.1 Dynamics of testosterone –10 during a resistance training session. –15 (a) Dynamics of free testosterone. (b) Dynamics of total testosterone. –20 Interrupted horizontal line, magnitude of hemoconcentration. –25 (Data from Ahtiainen et al. 2003.) (b) responses of the blood testosterone concentra- sets of 5-RM over 3-min rest pauses or three sets of tion do not linearly reflect the secretion responses. 10-RM over 1-min rest pauses (Kraemer, W.J. et al. Nevertheless, concentration changes are important 1990, 1991a). However, there is also another pos- because metabolic effects of hormones are related to sibility: six series of eight bench presses at 70% of hormone concentration in intercellular fluid. The maximal resistance (total 24 min) did not cause a total hormone amount in the body fluid compart- significant increase of testosterone level. Testosterone ments has significance for maintaining the hormone response was also not observed when the session effects. was followed by a maximal number of consecutive bench presses at 70% (Guezennec et al. 1986). Significance of intensity and duration of resistance ses- When two workload protocols (10-RM vs. 5-RM) sion. After sessions of resistance exercises a frequent were involved, testosterone increase in blood was finding is increased testosterone levels. This result greater after the four exercises using 10-RM (i.e. has been obtained in results of sessions with a dura- lower weight, greater total work) protocol as com- tion of approximately 30 min (Jürimäe et al. 1990), or pared with 5-RM (i.e. higher weight, lower total after four exercises consisting of either three to five work). After eight exercises no significant difference 324 chapter 23

in testosterone levels was found between the two Power sessions. A large number of repetitions protocols (Kraemer, W.J. et al. 1991a). Volvek et al. with low power output in bodybuilders decreased (1997a) observed that testosterone concentrations testosterone and increased growth hormone levels, increased modestly (by 7%) using a bench press whereas in weightlifters a high number of repeti- protocol of five sets to failure with 10-RM but tions increased testosterone concentrations without significantly more (by 15%) using a jump squat pro- changes in growth hormone level. When power out- tocol of 15 sets with 10 repetitions at 30% of 1-RM put remained close to maximum and the application squat. The significance of training session workload of force increased (and number of series decreased), was confirmed by Cotschalk et al. (1997): a three-set no significant hormone changes were found in heavy resistance protocol resulted in a greater in- weightlifters. Sprinters performed exercises at max- crease in testosterone concentration than a one-set imal power with a force of 60%. Although the num- protocol. Häkkinen and Pakarinen (1993) reported ber of series was modest, the total workload was that the increase of the total workload is responsible obviously high. In men, blood concentration of tes- for the blood testosterone response despite the neces- tosterone decreased together with the same change sary reduction of the exercise intensity. in blood levels of LH and cortisol. It is possible to Besides the characteristics of exercise, duration of suggest that the reversed hormonal responses were rest intervals has a significant effect on hormonal related to pronounced fatigue that developed dur- responses. The significance of rest intervals un- ing the session. This possibility was confirmed by a doubtably influenced the results of above-mentioned significant decrease of average power in full squats studies of Kraemer, W.J. et al. (1990, 1991a). After and by an increase of the total bioelectric activity of a 30-min intensive single circuit weight training ses- contracting muscles to power output in full squats sion (work/rest ratio 30 s; 30 s at 70% of 1-RM) male (Bosco et al. 2000). university students exhibited testosterone levels 24% higher than the initial values (Jürimäe et al. Training and modulators effects 1990). Summing up, Fleck and Kraemer (1997) affirmed Training effects. Few studies provide results on that the hormonal response in resistance exercise training effects in resistance training sessions. depends on muscle mass recruited, intensity of Guezennec et al. (1986) observed no significant workout, amount of rest between sets and exercises testosterone response after six series of eight bench and previous training. Both concentric and eccentric presses at 70% of 1-RM or after the maximal number exercises are able to increase the testosterone con- of repetitions at the same workload. The measure- centration in blood (Durand et al. 2003). ments were repeated each month for 4 months of Jensen et al. (1991) compared testosterone changes training, but the results were the same. Because the during and after endurance and strength training subjects were male weight-trained athletes, the lack sessions in the same men for both forms of training. of testosterone response might be related to previ- The mean increase of testosterone concentration ous training adaptation to test exercise. In studies on was 27% in the endurance session and 37% in the elite athletes, Häkkinen and Pakarinen (1993) found strength session. Differences between responses in that 20 sets of squats at 1-RM did not increase con- the two sessions were not significant but there were centrations of total and free testosterone, whereas large differences in the testosterone responses at the the testosterone level rose significantly when 10 sets individual level. A high correlation (r = 0.98) for were performed at 70% of 10-RM. If these results individuals was found between increases of testo- are related to previous adaptation to resistance sterone concentration after strength and endur- exercises then, interestingly, adaptation influences ance sessions. The authors suggested that the the response to high intensity workload but not to interindividual differences in testosterone responses voluminous workload. might be of importance for individual adaptation to In weightlifters, increased testosterone concentra- training. tion has been found after four series of six squats at resistance exercise and testosterone 325

90–95% of 6-RM as well as after nine to 10 squats at Subjects with a high trait anxiety showed signific- 60–65% of 6-RM (Schwab et al. 1993). In these ath- ant increases of testosterone in test exercise similar letes the exercises did not demonstrate adaptation to normal subjects, whereas concentrations of an by disappearance of testosterone responses. How- adrenal androgen, androstendione, increased less ever, since the test exercise was adjusted to the in subjects displaying a high trait anxiety (Diamond individual repetition maximum, the increased actual et al. 1989). workload avoided the disappearance of testoster- The above results presented that after a hard one response. power training session a pronounced decrease was Summing up, resistance training may remove the found in concentrations of testosterone and cortisol hormonal responses when the test exercise is similar (Bosco et al. 2000). It was suggested that this change to those frequently used in training. At the same was related to fatigue; more correctly, to a latent time, a possibility exists that long-term resistance fatigue appearing before the actual drop of working training promotes testosterone responses, at least in capacity. This suggestion has to be verified with the adolescent athletes. The same weightlifting protocol aid of further experiments. increased testosterone concentration in 17-year-old Häkkinen et al. (1988b) measured hormone res- juniors who had more than 2-years training experi- ponses in elite athletes performing two strength ence (Kraemer, W.J. et al. 1992). training sessions a day. Both testosterone and cor- The effects of resistance training differ from the tisol responses decreased after the first session, but outcome of endurance training, not only by changes the testosterone increased after the second session. at the level of muscle fibers and of aerobic capacity, The authors explained these results by the circadian but also in regard to influence on hormonal res- biorhythm: during the morning session the testo- ponses in exercises. The combination of strength sterone increase was masked by the decline in hor- and endurance training results in an attenuation of mone basal level, during the afternoon session the the performance improvements and adaptations lowered initial level of testosterone favored its typical for single-mode training (Kraemer, W.J. increase. et al. 1995a). Endurance training suppresses blood Results also became available showing that diet- testosterone level (Hackney 1996). The endogenous ary nutrients (Volek et al. 1997b), a protein–carbohy- stimulation of testosterone production gives less drate supplement consumed 2 h before (Kraemer, pronounced change in endurance-trained men com- W.J. et al. 1998b) or post-exercise nutrition (Bloomer pared to sedentary men (Hackney et al. 2003). et al. 2000) may influence the testosterone pattern in resistance training session or during the 24 h of Modulator influences on testosterone responses in resist- recovery. Amino acid supplementation (Fry et al. ance exercise. It has been believed that aggressiveness 1993) or the administration of either creatine (Volek favors the performance of resistance and power exer- et al. 1997a; Op’Teijnde & Hespel 2001) or ginseng cises. At the same time, aggressiveness is related to (Youl et al. 2002) failed to alter testosterone testosterone. Therefore a question arises as to whether responses. emotions of this type influence testosterone res- ponses in resistance exercises. In regard to this ques- Relationship of testosterone increase to LH response. tion, a new way of thinking arose from the results of Some studies indicated concomitant increases of Elias (1981). This study showed that judo fighting testosterone and LH levels during long-lasting increased the blood level of testosterone and cortisol training (Häkkinen et al. 1987, 1988a). However, a more in the winners than in the losers. In accord- possibility of uncorrelated changes of these two hor- ance, tennis players who exhibited a good mood and mones has also been shown (Häkkinen & Pakarinen self-assurance had higher testosterone levels before 1991). In resistance exercise sessions increases of competition. After the match, winners exhibited testosterone and LH are usually not parallel. In further elevation of the testosterone level, a decrease some studies concomitant declines of testosterone being observed in losers (Booth, A. et al. 1989). and LH have been observed (Häkkinen et al. 1988b; 326 chapter 23

Bosco et al. 2000). However, these data cannot be both gender groups without difference in the mag- used as evidence against the role of LH in testo- nitude of response. Resistance exercises during sterone response during exercises. The LH response 40 min increased concentration of another adrenal consists of a short-term burst of secretion, which is androgen, DHEA, more than endurance exercises in followed by a delayed increase in testosterone secre- women 19–69 years of age (Copeland et al. 2002). tion. Therefore, in experiments it is a complicated Kraemer, R.R. et al. (1995) compared hormonal task to ‘catch’ the LH response. More convenient is responses in healthy women in early follicular and to record the prolonged result of LH pulsation in luteal phases. A low-volume resistance exercise increased testosterone secretion and blood level. program caused greater estradiol response in the The exercise effect on pulsative secretion of LH has follicular phase. Growth hormone and andros- been studied in women and in relation to endurance tendione responses appeared only in luteal phase, exercises (Cumming et al. 1985; Weltman et al. 1990) whereas testosterone and progesterone did not but not in men and in regard to resistance exercise. response during this exercise. Endurance training may alter the biological activity A problem is whether the resistance exercises of endogenous LH (Di Luigi et al. 2002). Corres- influence the activity of enzymes in the adrenal ponding information regarding resistance training cortex responsible for forming androgen steroids. is not available. Neither confirming nor contradicting evidence is still available. However, individual differences in the production of androgens exist between women. Gender and age It has been shown that an acute resistance exercise Testosterone responses in women. Several studies test (ARET) increased concentrations of total testo- indicated that, differently from men, women do not sterone by 25%, free testosterone by 25% and SHBG react to resistance exercises by increased testos- by 4%. Multiple regression analysis indicated that in terone level in blood. When eight exercises for young healthy women the testosterone responses various muscles were performed with workloads to ARET is predicted by the subscapular to triceps of either 5-RM (rest intervals 3 min) or 10-RM ratio in skinfolds and by the ratio of upper-arm fat (rest intervals 1 min), testosterone concentration to mid-thigh fat assessed with magnetic resonance increased in men but not in women despite prior imaging. Waist-to-hip ratio in fat failed to have sig- recreational resistance training experience in both nificance as a discriminator for hormonal concen- contingents (Kraemer, W.J. et al. 1991a, 1993). In con- trations in women (Nindl et al. 2001). trast, Cumming et al. (1987) found a 20% increase of testosterone in women after a session consisting of Testosterone responses in elderly people. According to six isokinetic resistance exercises. Although the LH the results of Häkkinen and Pakarinenen (1995), level increased by 60%, it is unlikely that it was the a heavy resistance session caused testosterone in- physiological stimulus for testosterone production crease in both 30-year-old and 50-year-old men, but because women produce this hormone as a bypro- not in 70-year-old men. Women of any age failed to duct of steroid biosynthesis in the adrenal cortex. show testosterone response. In another study, Accordingly, those women who did not exhibit a Häkkinen et al. (1998) compared the acute effect cortisol response failed to show an increase in tes- of heavy resistance exercise on blood testosterone tosterone concentrations. The other female subjects in young (mean age 26.5 years) and old (mean age showed a parallel increase in testosterone and cor- 70 years) men. Exercises for legs, upper body and tisol concentrations (Cumming et al. 1987). their combination increased total testosterone level During strength exercises, Weiss et al. (1983) in young men but only leg exercise increased total observed rather similar relative increases in blood testosterone level in old men. Kraemer, W.J. et al. testosterone, in men from the high level by 21.6% (1998a) found that four sets of squats of 10-RM and in women from the low level by 16.7%. Andro- caused increased total and free testosterone levels in stendione concentration increased significantly in both 30- and 62-year-old men. The area under the resistance exercise and testosterone 327

response curve during and 30-min post-exercise In rats, seven times 1-min swimming with a high was greater in younger men. additional weight (of 12% body mass) simulated efforts in resistance training. The blood level of Testosterone responses in adolescents. Appearance of testosterone increased slightly. During the first 2 h testosterone in blood is related to puberty in boys of restitution the hormone level decreased. Four and to adrenarche in girls. During cyclic exercises hours after exercise a secondary rise in the hormone significant increase of testosterone level has been concentration was detected, exceeding the resting detected from the stage 4 of sexual maturation (by level by 1.5–2.5 times. Testosterone levels in cyto- Tanner’s scale) in boys (Fahey et al. 1979) and from plasm of skeletal and heart muscle fibers increased, the stage 5 in girls (Viru, A. et al. 1998). with the peak values 72 h after swimming (Fig. In resistance exercises, post-exercise as well as 23.2). At that time the specific binding of andro- pre-exercise testosterone levels were lower in 15- gens, as well as the number of binding sites, were year-old male athletes than in adult male athletes increased in the cytoplasm of skeletal muscle fibers (Pullinen et al. 1998). (Tchaikovski et al. 1986). It is possible to ask what is important for the post- exercise induction of adaptive protein synthesis Pattern of blood testosterone during the in exercised muscles, either testosterone response recovery period during the exercise or the secondary post-exercise In men, the testosterone level gradually returns to testosterone increase. The presented results of rat its initial level during the 1st hour following resist- experiments suggest that both increases of testo- ance sessions. In females, low blood levels of testo- sterone constitute an entire response. sterone remain without significant alterations both during the session and the 1st post-exercise hour Testosterone in monitoring of (Kraemer, W.J. et al. 1991a). This pattern is different resistance training from that found after cyclic exercises. After 2 h of bicycle ergometric exercise the testosterone level The fact that something is measured in athletes does dropped during the 6 h post-exercise. Twenty-four not mean training monitoring. It is possible to speak hours later it showed a normalizing trend, but the about training monitoring if the following five prin- mean testosterone concentration was still below the ciples are completely followed (Viru, A. & Viru 2001): initial values both in trained and untrained young 1 It is a process performed for the purpose of in- men (Viru, A. et al. 1992). Similar results have been creasing the effectiveness of training guidance. obtained by a Finnish research team (Kuoppasalmi 2 It is based on recording changes in an athlete 1980; Kuoppasalmi et al. 1980): after running for during various stages of the training or under the 13–14 km the testosterone level dropped during the influence of main elements of the sport activities. 1st hour and then remained low for 5 h. Ahtiainen 3 It is a highly specific process depending on the et al. (2003) also found that 30 min after a resistance sport event, performance level of the athlete and training session, free and total testosterone levels age/gender peculiarities. were below initial levels. When a 30-min resist- 4 Any method or measurement is in a sense train- ance session protocol prescribed a work-to-rest ratio ing monitoring if it provides reliable information of 30 s, 30-s testosterone concentration decreased related to the task being monitored. during the first 6 h of the post-exercise recovery 5 The information obtained from measurements (Jürimäe et al. 1990). Thus, the testosterone pattern has to be understandable so that necessary correct- after resistance exercises may be similar to that after ive changes in training can be made. endurance exercises. In the first and second morn- In the monitoring of the resistance training testo- ing after heavy resistance training, the level of total sterone basal level, responses to test exercise and the testosterone was normal but the level of free testo- ratio of testosterone/cortisol have been used. The sterone increased (Ahtiainen et al. 2003). idea is to get information about the contribution 328 chapter 23

Testosterone 2400 600

2100 500 1800 400 1500 –1 –1 g 1200 300 · g p pg·mL 900 Serum 200 600 Cytosol 100 300

0 0 (a)Exercise 0 h 2 h 4 h 12 h 24 h 48 h 72 h 144 h

1400 1.0 0.9 1200 0.8

1000 0.7 –1 0.6 800 Fig. 23.2 Post-exercise dynamics

0.5 protein of testosterone and its binding in g cpm skeletal muscle after swimming in 600 0.4 rats. (a) Closed triangles, testosterone Specific binding 0.3 in serum. Open circles, testosterone 400 fmol·m Number of binding sites 0.2 in cytosol of muscle fibers. (b) 200 Open squares, specific binding of 0.1 testosterone. Closed rhombs, number 0 0.0 of binding sites. (Data from (b)Exercise 0 h 2 h 4 h 12 h 24 h 48 h 72 h 144 h Tschaikovski et al. 1986.) of testosterone in training adaptations, and to be during the first 8 weeks the testosterone level informed about excessive training workloads. increased and the cortisol level decreased. After 16 weeks of training, levels of both hormones decreased. At the end of 24th week the mean concentrations Training effects of testosterone and cortisol did not differ from the Testosterone basal level during prolonged resistance initial values. The detraining did not accompany training. The related studies were initiated by hormonal changes. The effect of this training pro- Häkkinen from Finland. In 1985 he and his collabor- tocol on isometric force was significantly less pro- ators reported that heavy resistance training for 24 nounced than that in the high-resistance training weeks did not change the basal level of testosterone group (Häkkinen et al. 1985). in blood. Blood cortisol level decreased; the testo- Blood samples obtained every 4 months in elite sterone/cortisol ratio increased. This change was weightlifters showed testosterone increased from parallel to the increase in isometric force of the leg the month 8 to month 12 (Häkkinen et al. 1987). extensor muscles. During the followed detraining During the next year the increased testosterone for 12 weeks, both indices decreased to the pretrain- level persisted (Häkkinen et al. 1988a). The authors ing values. In combined training using jumping and suggested that increased testosterone level might strengthened exercises (weights of 60–80% 1-RM), create optimal conditions for intensive weight resistance exercise and testosterone 329

training to increase strength improvement in elite- gested in resistance training under the influence of strength athletes. In 1990 this team reported that, increased training volume. However, available data according to 51 weeks of follow-up of six elite suggests that in strength athletes most typical are weightlifters, the fitness level (evaluated by arbit- changes not in the cortisol level but in the testo- rary units) and blood testosterone level changed in sterone level. correlation (Busso et al. 1990). The meaning of ‘hard-training stage’ is a relative one depending on previous training. Therefore, the Effects of hard-training stages. Six weeks of pre- lack of testosterone changes in ‘intense’ training for paratory training of elite weightlifters was divided a couple of weeks (Häkkinen & Pakarinen 1991) or into a high voluminous stressful stage for 2 weeks 8 weeks of ‘heavy’ resistance training (Hickson et al. followed by 2 weeks of ‘normal’ training and then 2 1994; Potteinger et al. 1995) does not contradict the weeks of taper. During the first 2 weeks testoster- results of other studies. Also 12 weeks of heavy one concentration decreased significantly, cortisol resistance training did not change either the resting concentration elevated slightly and the ratio of level of testosterone or the responses in training testosterone/cortisol dropped. During next 2 weeks sessions (McCall et al. 1999). changes were modest, if at all. (Häkkinen et al. 1987). Kraemer, W.J. et al. (1996) suggests that in heavy The results of Busso et al. (1992) confirmed that in training hormonal changes are related to early elite weightlifters a 4-week stage of intensive train- adaptation, which later disappears. A question ing was accompanied with decreases in the blood arises in whether the decrease of hormonal changes testosterone level. Fry et al. (1994) investigated 1 indicates the necessity for further increase in the week of intensive training causing overreaching in training workout. junior weightlifters. During the 1st year the intense training week resulted in attenuated testosterone Testosterone in the taper stage. During the taper stage responses in test exercise. Over the 2nd year the after ‘stressful’ training (causing a decrease in tes- intensive training augmented the testosterone tosterone and a slight increase in cortisol levels), responses. testosterone remained low and cortisol decreased Kraemer, W.J. et al. (1995a) compared the effects (Häkkinen et al. 1987). However, different situations of hard endurance and strength training and their may exist. From the data of Busso et al. (1990) it is combination. During endurance training, testoster- possible that a sharp reduction of training workload one was constant but undulations occurred in the associates with the lowest testosterone level during cortisol response. In strength training, testosterone the training year. Another report of this team (Busso again stayed constant while cortisol response de- et al. 1992) also showed that the reduction of training creased. Combined training demonstrated changes workload for 2 weeks did not stop the testosterone of both testosterone and cortisol levels over the decline that appeared during the previous 4 weeks training period. of hard training. In power athletes a 2-week detrain- In endurance athletes hard-training stages result ing after heavy resistance training caused a decrease in two phases of hormonal changes (Viru, A. & Viru of cortisol and an increase of testosterone concentra- 2001). Typical for the 1st phase was increased corti- tion. Detraining does not impair the neuromuscular sol basal level and exaggerated cortisol increase in performance, as indicated by the results of most strenuous test exercise. Testosterone basal level was muscle strength and power tests and surface elec- constant or reduced and responses in test exercise tromyogram (EMG) activity (Hortobagyi et al. 1993). attenuated. The 2nd phase was indicated by sup- In recreationally strength-trained men, 6 weeks of pressed cortisol response in exercise while the detraining produced only a minimal change in per- basal level might be either increased or decreased. formance and hormonal levels (Kraemer, W.J. et al. Testosterone levels before and after test exercise 2002). were mostly low. Although the material is far from The taper problem is a rather complicated one sufficient for conclusions, two phases may be sug- because the taper stage may consist of various 330 chapter 23

degrees of reduction in the training load and in adaptive protein synthesis). Moreover, during acute various designs in regard to the choice of exercises exercise, cortisol makes an essential contribution and the pattern of their intensity. Further, the body to amino acid transamination, and thereby to the state when the reduction of training is considered formation of alanine, to glyconeogenesis, to the to be necessary may be also different. Moreover, synthesis of urea, to the synthesis of cateholamines the body responses both to overtraining and taper and supporting their effects on the post-receptor are individually variable. Therefore, summing-up level, and to the control of sodium–potassium fluxes results obtained in various athletes may actually through cellular membrane, etc. Therefore, insuffi- cause a loss of important information. cient or reverse responses of cortisol should make us Discussion of hormonal changes in the taper stage think about discrepancies in the metabolic control. led to an assumption that before competition the Adlercreutz et al. (1986) recommended using the taper stage has to be substituted by a short-term ratio between free testosterone and cortisol as an training stage for preconditioning the peak (at least indication of overstrain if the ratio decreases more high) performance. For power and strength events, than 30% or if the ratio is less than 0.35 × 10–3. This a high testosterone basal level may constitute an way, an extreme situation in the balance of anabolic important precondition for successful performance and catabolic stimulation is indicated. However, the (Viru, M. & Viru 2000). This is a problem still wait- proposed quantitative measure was later frequently ing systematic research. forgotten and any decrease in the ratio of testo- sterone/cortisol was considered a bad indication. Overtraining. The main hormonal indicators of With this background, conclusions on overtraining overtraining are suppressed production of cortisol, were made, although the authors’ data demon- testosterone, growth hormone, nocturnal excretion strated a good performance level. Even if the ratio of catecholamine and decreased sensitivity to hor- decreases more than 30% or is less than 0.35 × 10–3, mones (Kuipers & Keizer 1988; Urhausen et al. 1995; indications of overtraining may not exist (Kuipers & Lehmann et al. 1997, 1999; Viru, A. & Viru 2001). Keizer 1988; Urhausen et al. 1995). Several overtraining manifestations reflect a general In essence, the main shortcoming of the use of imbalance in the level of hypothalamic neuro- this ratio is that the hormone level in blood is not secretion and/or in the function of the anterior linearly related to its metabolic effects. This ratio has pituitary gland (Keizer 1998; Lehmann et al. 1998; been used despite the lack of information about the Hackney 1999). In accordance with the listed over- state of the glucocorticoid and androgen receptor. training manifestations, Fry et al. (1998) found Actually, the state of these receptors as well as the that in weight-trained men the overtrained state is competition between testosterone and glucocortioid indicated by strength decrements and a pronounced for glucocorticoid receptor are essential for deter- decrease in blood cortisol concentration. Overtrained minants of the outcome in anabolism/catabolism men exhibited a slightly increased testosterone level balance. Consequently, the ratio of testosterone/ after the test exercise. cortisol is too indirect to be used. In disagreement, several researchers believe that high cortisol levels in training are related to Testosterone, training effects and dysadaptation. The reason for this opinion is the performance catabolic effect of glucocorticoids that has been set in contrast to the anabolic action of testosterone. Induction of synthesis of myofibrillar proteins However, it should not be forgotten that in stress situations, including strenuous exercise, enhanced The main result of strength training is myofibrillar catabolism is not a misfortune but ultimately a neces- hypertrophy. This process is founded on the induc- sary adaptation response (creation of the pool of tion of synthesis of myosin and actin (Booth, F.W. & free amino acids to be used as additional substrate Thomason 1991). While various metabolic factors for oxidation as well as building blocks for the are able to induce this process, a powerful amplifier resistance exercise and testosterone 331

Type I fibers Down-regulation of androgen receptors Resistance exercises Fig. 23.3 Effect of resistance exercises on androgen receptor in muscle fibers Type II fibers Up-regulation of type I and type II. androgen receptors

of the induction of synthesis of contractile proteins (Urban et al. 1995; Kraemer, W.J. et al. 1996). When is testosterone (Viru, A. 1995). In fact, amplificat- training induced enhanced testosterone, cortisol ory effect of testosterone has been evidenced by the and growth hormone responses, the improvement prohibited use of anabolic steroids by athletes. of muscle strength was pronounced (Hansen et al. Frequently, although not always, the result was 2001). In disagreement, Hickson et al. (1994) reported enhanced training effects in strength and power increased cross-sectional area of fast-twitch fibers events (Rogozkin 1979; Wilson 1988; Lamb 1989). as a result of 8 weeks of heavy resistance training Rat experiments evidenced that anabolic steroids without increases of testosterone basal level or stimulate the synthesis of myofibrillar proteins, its responses in exercises. However, testosterone increase RNA polymerase activity and augment changes might take part during the recovery period; the corresponding training effects (Rogozkin 1979; also possible were alterations at the receptor level. Rogozkin & Feldkoren 1979). Accordingly, it has Thyroid hormones add their action on transcrip- been assumed that in normal conditions of training tion of oxidative enzymes and myosin (Freeberg & the synthesis of myofibrillar proteins is amplified by Hamolsky 1974; Konovalova et al. 1997). The action the endogenous androgens. of growth hormone, growth factors and insulin on It has been mentioned above (pp. 327–8) that, the translation process (Balon et al. 1990; Fryburg during the recovery period after exercises requiring et al. 1991) exerts a general support for the actualiza- strong muscle contractions, a number of binding tion of the adaptive protein synthesis (Fig. 23.4). sites increase the cytoplasm of working muscles in Training-induced muscle hypertrophy is con- association with increased production of proteins in trolled at transcription, translation and post- the muscle (Tchaikovski et al. 1986). The signific- translation level (Booth, F.W. & Thomason 1991), ance of androgen receptors in training hypertrophy whereas the used exercises may determine the was confirmed in rats: a pharmacological blockade relative significance of contribution of actions at of androgen receptors prevented training-induced each control level. Therefore, the hormonal effects muscle hypertrophy (Inoue et al. 1994). At the same on muscle hypertrophy also vary in dependence of time muscular activity itself increased the number the exercises used. of androgen binding sites in rat muscles (Inoue et al. After exercise, the catabolic influence of cortisol 1993). Control of testosterone action at the receptor is essential to provide free amino acids for protein level ensures a fiber-type specific stimulation of synthesis, to maintain the rate of protein degrada- protein synthesis in muscles during the recovery tion for renewal of protein structures and to adjust period. Resistance exercises induce down-regula- the number of protein molecules to the actual need tion of androgen receptors in slow-twitch fibers and (the post-translation control) (Viru, A. 1995). up-regulation of these receptors in fast-twitch fibers (Deschenes et al. 1994). Consequently, the tissue’s Preconditioning effect of testosterone susceptibility to testosterone’s effect increases selectively during resistance exercise in fast-twitch Data has suggested that performance in muscle muscle fibers (Fig. 23.3). power tests relates to the blood level of testosterone In humans, the testosterone effect on muscle in athletes. In professional soccer players, the results strength and protein synthesis has been confirmed of counter-movement jump positively correlated 332 chapter 23

Control at the level for 60 s and the change of blood testosterone con- of transcription centration during this test (Bosco et al. 1996a). Inductors Since the metabolic effect of testosterone is a time- Thyroid hormones consuming process due to the formation of the Testosterone testosterone-receptor complex, influence of this com- TRANSCRIPTION Genome plex on the genome and synthesis of new proteins (Liao 1977; Mainwaring 1977), it is not believed that, during short-term acute force efforts, jumping exer- mRNA cise or sprint distances, necessary time is provided to activate secretion of testosterone, to transport the Control at the level augmented amount of the hormone to working of translation muscle and to actualize related metabolic events. TRANSLATION Microsomes and polysomes Therefore, a hypothesis has been raised that testo- sterone has a preconditioning action on the applica- Insulin Growth hormone tion of muscle force and power (Viru, A. & Viru 2001). Competition performance (as well as the main part of a training sessions) is preceded by warming Cortisol New proteins up. Athletes are influenced by the anticipatory state. During competition each athlete can use three or six Control at the trials. Therefore, in competition power exercises are post-translation level performed on the level of altered hormone levels in Protein degradation blood. The significance of these hormonal alterations adjust the protein levels for performance depends on the time required for to the actual necessities the actualization of metabolic effects of hormones. Fig. 23.4 Hormonal control of synthesis of myofibrillar The hormones, which are bound on cellular mem- proteins. brane and acting through formation of cyclic adeno- sine monophosphate (cAMP) (e.g. catecholamines) need only a couple of seconds to evoke their meta- with the basal level of testosterone in blood (Bosco bolic effects. The hormones, which are bound by et al. 1996b). The specific relationship between specific receptors in cytoplasm and acting through testosterone level and the explosive strength of leg the induction of protein synthesis (e.g. testosterone muscles was supported by the fact that aerobic and other steroid hormones), in several cases require endurance, determined by Cooper’s 12-min runs more than 1 h to show metabolic effects. Conse- test, showed negative correlation with testosterone quently, performance in exercises of explosive type level (Bosco et al. 1996b). Comparison of testo- of power output may be influenced by hormonal sterone, determined by morning samples and the rise changes before the main performance. of the center of gravity in the counter-movement Two types of preconditioning of the perform- test in 97 high-level athletes, indicated the highest ance in power events, exerted by endogenous values of testosterone concentration and jumping testosterone, may be distinguished. The long-term performance in sprinters and the lowest values in preconditioning is related to the influence of tes- cross-country skiers, whereas soccer players exhib- tosterone on the development of fast twitch (FT) ited intermediate values (Bosco & Viru 1998). In muscles. Mainly, it is related to the pubertal period. accordance, Kraemer, W.J. et al. (1995b) reported a The short-term preconditioning is related to the positive correlation between testosterone level and influence of testosterone either on the central nerv- performance in double knee extension exercises. ous system or peripheral neuromuscular apparatus. Significant correlations were also found between The result is ‘tuning’ of the motor centers of the cent- average power output during continuous jumping ral nervous system for explosive performance. resistance exercise and testosterone 333

Long-term preconditioning. Explosive contractile activ- testosterone may enhance myofibrillar hypertrophy ity of muscles (jumping, sprinting performance, in resistance training. In power events, essential is etc.) is related to the percentage of FT fibers in leg the influence of high testosterone levels on central muscles (Costill et al. 1976; Bosco & Komi 1979). nervous structures. Various influences in early post- Thus, the formation of capacity for muscular activit- natal life play a role in the choice of neurons of ies of explosive type depends on the development the central nervous system, which become sensitive of FT fiber and fast motor units. Results of several to steroids. The spinal nucleus of the bulboca- studies indicate that testosterone is, plausibly, vernosus is highly androgen-sensitive. Testosterone responsible for improved anaerobic enzyme sys- regulates both the size of motoneurons of the spinal tems and structural development of FT fibers in nucleus of the bulbocavernosus and also related muscles. Bass et al. (1971) established that in tempor- muscle in adulthood (Kurz et al. 1986; Araki et al. alis muscle of guinea pig, sexual differentiation of 1991; Lubisher & Arnold 1995). This neuromus- enzyme pattern might be converted by testosterone. cular subsystem plays an important role in male Dux et al. (1982) demonstrated that pubertal castra- copulatory behavior. Still, there is no evidence that tion alters the structure of skeletal muscle. Most of testosterone influences the neural adaptations in them suffered in the development of FT muscles. strength training at the level of spinal motoneurons Krotiewski et al. (1980) confirmed the castration covering the function of major muscles of the body. effects in male rats. Testosterone substitution This may be suggested by the fact that androgens restored development of FT muscles in male cas- are able to influence the structure of neurons, trates. According to these results, it is possible to including dendritic branching and synapse forma- assume that during puberty inter-individual dif- tion in the adult brain (Arnold & Breedlove 1985; ferences in testosterone action determine the forma- Matsumoto 1992). The contribution of testosterone tion and development of FT fibers. Several studies in training-induced long-term neural adaptivity in male adolescents support this assumption. awaits investigation. Already at the onset of puberty (in 11–12-year-old boys) area of FT fibers as well as blood lactate level Short-term preconditioning. Another way to under- after 15 s all-out exercise correlated significantly stand the significance of testosterone in power exer- with testosterone level (Mero 1988). In circum- cises is the short-term preconditioning effect. In the pubertal boys testosterone levels in blood or saliva short-term preconditioning action, the testosterone correlated with maximal anaerobic power (Mero level in blood is, plausibly, highly significant. et al. 1990; Falgairette et al. 1991), maximal power Investigation of aggressive behavior showed weak output in incremental exercise (Fahey et al. 1979), positive correlations between the blood level of blood lactate level after Wingate test (Mero et al. testosterone and manifestations of aggressiveness 1990) and maximal voluntary strength (Mero 1988). in humans (Archer 1991; Book et al. 2001). The Bosco (1993) indicated that between 8.5 and 14.5 testosterone level is not the single determinant of years the rise of center gravity in counter movement aggressive behavior, which actually depends on the jump increases linearly in children of both gender. interaction of several factors, including previous From the age of 14.5 years a prevalence of boys experience, environment, danger of the situation, appeared. Typical for this age is a pronounced etc. (Mazur & Booth 1998). Thus, the testosterone increase of blood testosterone concentration. level is a preconditioning prerequisite for aggress- Thus, in the pubertal period, enhanced increase ive behavior but not the determinant of the expres- of testosterone concentration in blood obviously sion of aggressiveness. High levels of testosterone favors the development of FT fibers. Thereby a phe- made boys more impatient and irritable, which notype is formed which is characterized by high in turn increased their propensity to engage in testosterone level and effective performance in exer- aggressive–destructive behavior (Olweus et al. cise of explosive application of forces. 1988). By analogy, it is possible to assume that testo- Besides pubertal period, inherent high levels of sterone promotes changes in neurons, which are not 334 chapter 23

only related to increased aggressiveness but also to rons, synapses or muscle fibers is a matter of further favorable mobilization of neuromuscular capacity investigation. for explosive performance in power throwing, jumping and sprint events. In conclusion. The hypothesis on the preconditioning During sport competition, the short-term pre- of performance in power events by endogenous condition effect may appear due to the increased testosterone opens a wide spectrum of tasks for fur- testosterone concentration in blood as a result of ther research. Perspectives include testing various warming-up exercises, anticipatory state, emotional aspects of the hypothesis as well as detailed invest- strain aroused from performance of other compet- igations in order to establish the cellular–metabolic itors, as well as of prolonged concentration for the foundations for various actions of testosterone on forthcoming competition. It should be remembered nervous structures and muscle related to power that during sports competition in judo or tennis, performance. winners have higher levels of testosterone than losers (Elias 1981; Booth, A. et al. 1989). Therefore Conclusions Mazur and Booth (1998) suggested that testosterone prepared winners for effective performance. Testosterone level in blood increases in men during Some anecdotic examples support the suggestion resistance exercises. This response depends on sev- of the significance of testosterone in the precon- eral conditions, among those the primary significance ditioning of performance in power events. A couple seems to belong to the application of high muscle of days before an international competition, low forces or power output during a sufficiently long testosterone concentration was found in a discus time (> 10–15 min) over relatively short rest pauses. thrower. His performance was lower than expected. During short-term resistance exercises rapid testo- Within 3 weeks following the competition his tes- sterone response is mainly related to hemoconcen- tosterone concentration normalized. Further, after tration. After resistance training sessions a secondary return from winning an international competition a increase of blood testosterone concentration may decathlete showed an unusually high testosterone appear in the late recovery period concomitantly level (36 mmol·L–1). with increased binding of testosterone by specific On the background of extended experimental sites of androgen receptors in muscle fibers. Resist- material, Ingle (1952) affirmed that the permissive ance exercises, probably also power exercises, cause effect of hormones consists in enabling changes in the up-regulation of androgen receptors mainly in body function or metabolic processes (gives the per- fast-twitch fibers. The most important adaptive effect mission to change), although the hormone itself is of testosterone is the induction of the synthesis of not the direct cause of the change. The supposed contractile proteins, mainly in fast-twitch fibers. This permissive action of testosterone is obviously re- effect is founded on the formation of the testosterone- lated to the indirect effect of testosterone, which is receptor complex and its action on the genome. The actualized without participation of the androgen late recovery period after resistance training sessions receptor (Nieschlag & Behre 1998). The manifesta- is the time for actualization of the adaptive protein tions of the indirect effect of testosterone are: pro- synthesis induced by testosterone. The anabolic duction of insulin-like growth factor I; competition effect of testosterone-initiating transcription of syn- for the specific binding sites of glucocorticoids; thesis of myofibrillar proteins (release of related autocrine release of andromedins; transmembrane mRNA) is supported by action of growth hormone influx of extracellular calcium; and activation of and growth factors on the translation process. extracellular signal-related kinase cascade via bind- Action of testosterone on acute muscle per- ing to a yet unidentified extracellular receptor. formance is also possible without the contribution Specification of testosterone action on muscle force of androgen receptors. The resulting precondition and power generation in regard of these possibilit- effect of testosterone is still a suggestion and re- ies, as well as localization of related effect(s) in neu- quires systematic investigation. resistance exercise and testosterone 335

References Adlercreutz, H., Härkönen, M., perspectives of various models. track athletes. Journal of Applied Kuoppasalmi, K. et al. (1986) Effect Physiological Reviews 71, 541–585. Physiology 40, 149–154. of training on plasma anabolic and Bosco, C. (1993) Test di valutazione della Cotshalk, L.A., Loeb, C.C., Nindl, M. et al. catabolic steroid hormones and their donna nella practica del calcio. In: (1997) Hormonal responses of multiset response during physical exercise. Confegno Nazionale ‘Il Calcio Femminile’ versus singleset heavy-resistance International Journal of Sports Medicine 7 Aspetti Medici e Tecnici (Cambri, R. & exercise protocols. Canadian Journal of (suppl.), 27–28. Paterni, M., eds.). FIGC, Rome: Applied Physiology 22, 244–255. Ahtiainen, J.P., Pakarinen, A., Kraemer, 219–230. Cumming, D.C., Vickovic, M.M., Wall, W.J. & Häkkinen, K. (2003) Acute Bosco, C. & Komi, P. (1979) Mechanical S.R., Fluker, M.R. & Belcastro, A.N. hormonal and neuromuscular characteristics and fiber composition (1985) The effect of acute exercise on responses and recovery to forced vs. of human leg extensor muscles. pulsatile release of luteinizing hormone maximum repetitions multiple European Journal of Applied Physiology 41, in women runners. American Journal of resistance exercises. International 275–284. Obstetrics and Gynecology 153, 482–485. Journal of Sports Medicine 24, Bosco, C. & Viru, A. (1998) Testosterone Cumming, D.C., Wall, S.R., Galbraith, 410–418. and cortisol levels in blood of male M.A. & Delcastro, A.N. (1987) Araki, I., Harada, Y. & Kuno, M. (1991) sprinters, soccer players and cross- Reproductive hormone responses to Target-dependent hormonal control country skiers. Biology of Sport 15, 3–8. resistance exercise. Medicine and Science of neuron size in the rat spinal nucleus Bosco, C., Tihanyi, J., Rivalta, L. et al. in Sports and Exercise 19, 234–238. of the bulbocavernosus. Journal of (1996a) Hormonal responses in Danhaive, P.A. & Rousseau, G.G. (1988) Neurosciences 11, 3025–3033. strenuous jumping effort. Japanese Evidence for sex-dependent anabolic Archer, J. (1991) The influence of Journal of Physiology 46, 93–98. response to androgenic steroids testosterone on human aggression. Bosco, C., Tihanyi, J. & Viru, A. (1996b) mediated by glucocorticoid receptor in British Journal of Psychology 82, 1–28. Relationship between field fitness test rats. Journal of Steroid Biochemistry 29, Arnold, A.& Breedlove, M. (1985) and basal serum testosterone and 275–281. Organizational and activational effects cortisol levels in soccer players. Clinical Deschenes, M.R., Maresch, C.M., of sex steroids on brain and behavior: Physiology 16, 317–322. Armstrong, L.E., Kraemer, W.J. & a reanalysis. Hormones and Behavior 19, Bosco, C., Colli, R., Bonomi, R., von Covault, J.M. (1994) Endurance and 469–498. Duvillard, S.P. & Viru, A. (2000) resistance exercise induced muscle fiber Balon, T.W., Zorzano, A., Treadway, J.L., Monitoring of strength training: type specific responses in androgen Goodman, M.N. & Ruderman, N.B. neuromuscular and hormonal profile. binding capacity. Journal of Steroid (1990) Effects of insulin on protein Medicine and Science in Sports and Exercise Biochemistry 50, 175–179. synthesis and degradation in skeletal 32, 202–208. Diamond, P., Brisson, G.R., Candas, B. muscle after exercise. American Journal of Busso, T., Häkkinen, K., Pakarinen, A. & Perronet, F. (1989) Trait anxiety, Physiology 258, E92–E97. et al. (1990) A systems model of training submaximal physical exercise and blood Bass, A., Gutmann, E., Hanzlikova, V. & response and its relationship to androgens. European Journal of Applied Syrovy, I. (1971) Sexual differentiation hormonal responses in elite weight- Physiology 58, 699–704. of enzyme pattern and its conversion by lifters. European Journal of Applied Di Luigi, L., Guidetti, L., Baldari, C. et al. testosterone in temporalis muscle of the Physiology 61, 48–54. (2002) Physical stress and qualitative guinea pig. Physiologica Bohemoslovenica Busso, T., Häkkinen, K., Pakarinen, A. gonadotropin secretion: LH biological 20, 423–431. et al. (1992) Hormonal adaptations and activity at rest and after exercise in Bloomer, R.J., Sforzo, G.A. & Keller, B.A. modelled responses in elite weightlifters trained and untrained men. International (2000) Effects of meal form and during 6 weeks of training. European Journal of Sports Medicine 23, 307–312. composition on plasma testosterone, Journal of Applied Physiology 64, 381–386. Durand, R.J., Castrence, V.D., Hollander, cortisol, and insulin following resistance Chwalbinska-Moneta, J., Krysztofiak, H., D.B. et al. (2003) Hormonal responses exercise. International Journal of Sport Ziemba, A., Nazar, K. & Kaciuba- from concentric and eccentric muscle Nutrition, Exercise and Metabolism 10, Usc´ilko, H. (1996) Threshold increase in contraction. Medicine and Science in 415–424. plasma growth hormone in relation to Sports and Exercise 35, 937–944. Book, A.S., Starzyk, K.B. & Quinsley, V.L. plasma catecholamine and blood lactate Dux, L., Dux, E. & Guba, F. (1982) Further (2001) The relationship between concentration in endurance-trained data on the androgenic dependence testosterone and aggression: a meta- athletes. European Journal of Applied of the skeletal musculature: effect of analysis. Aggression and Violent Behavior Physiology 73, 117–120. pubertal castration on the structure of 6, 579–599. Copeland, J.L. Consitt, L.A., & Tremblay, the skeletal muscle. Hormonal and Booth, A., Shelley, G., Mazur, A., Thorp, G. M.S. (2002) Hormonal responses to Metabolic Research 14, 191–194. & Kittok, R. (1989) Testosterone, endurance and resistance exercise Elias, M. (1981) Serum cortisol, and winning and losing in human in females 19–69 years. Journals of testosterone, and testosterone-binding competition. Hormones and Behavior 23, Gerontology. Series A, Biological Sciences globulin response to competitive 556–571. and Medical Sciences 54, B158–B165. fighting in human males. Aggressive Booth, F.W. & Thomason, D.B. (1991) Costill, D.L., Daniels, J., Evans, W. et al. Behavior 7, 215–224. Molecular and cellular adaptation of (1976) Skeletal muscle enzymes and Fahey, D.T., Valle-Zuris, A.D., Oehlsen, G., muscle in response to exercise: fiber composition in male and female Trieb, M. & Seymour, J. (1979) Pubertal 336 chapter 23

stage differences in hormonal and training. Medicine and Science in Sports term strength training on human hematological responses to maximal and Exercise 28, 180–189. skeletal muscle: the importance of exercise in males. Journal of Applied Hackney, A.C. (1999) Neuroendocrine physiologically elevated hormone Physiology 46, 823–827. system. Exercise overload and levels. Scandinavian Journal of Medicine Falgairette, G., Bedu, M., Fellmann, N., regeneration. In: Overload, Performance and Science in Sports 11, 347–354. VanPraagh, E. & Coudert, J. (1991) Incompetency, and Regeneration in Sport Hickson, R.C., Hidaka, K., Foster, C., Bioenergetic profile in 144 boys aged 6 to (Lehmann, M., Foster, C., Gastmann, U., Falduto, M.T., Chatterton, R.J. (1994) 15 years with special reference to sexual Keizer, H. & Steinacker, J.M., eds.). Successive time courses of strength maturation. European Journal of Applied Kluwer Academic/Plenum, New York. development and steroid hormone Physiology 62, 151–156. Hackney, A.C., Szczepanowska, E. & Viru, responses to heavy-resistance training. Fleck, S.J. & Kraemer, W.J. (1997) Designing A. (2003) Basal testicular testosterone Journal of Applied Physiology 76, 663–670. Resistance Programs. Human Kinetics, production in endurance-trained men is Hortobagyi, T., Houmard, J.A., Stevenson, Champaign, IL. suppressed. European Journal of Applied J.R. et al. (1993) The effects of detraining Freeberg, A.S. & Hamolsky, M.W. (1974) Physiology 89, 198–201. on power athletes. Medicine and Science Effects of thyroid hormones on certain Häkkinen, K. & Pakarinen, A. (1991) in Sports and Exercise 25, 929–935. nonendocrine organ systems. In: Serum hormones in male strength Ingle, D.J. (1952) The role of the adrenal Handbook of Physiology, Endocrinology, athletes during short term strength cortex in homeostasis. Journal of Thyroid. American Physiological Society, training. European Journal of Applied Endocrinology 8, xxii–xxxvii. Washington, D.C.: 435–468. Physiology 63, 194–199. Inoue, K., Yamasaki, S, Fushika, T., Okada, Fry, A.C., Kraemer, W.J., Stone, M.H. Häkkinen, K. & Pakarinen, A. (1993) Acute Y. & Sugimoto, E. (1993) Rapid increase et al. (1993) Endocrine and performance hormonal responses to two different in the number of androgen receptors responses to high volume training and fatiguing heavy-resistance protocols following electrical stimulation of the amino acid supplementation in elite in male athletes. Journal of Applied rat muscle. European Journal of Applied junior weight athletes. International Physiology 74, 882–887. Physiology 66, 134–140. Journal of Sport Nutrition 3, 306–322. Häkkinen, K. & Pakarinen, A. (1995) Acute Inoue, K., Yamasaki, S., Fushiki, T., Okada, Fry, A.C., Kraemer, A.C., Stone, M.H. hormonal responses to heavy resistance Y., Sugimoto, E. (1994) Androgen et al. (1994) Endocrine responses to exercise in men and women at different receptor agonist suppresses exercise- overreaching before and after 1 year of ages. International Journal of Sports induced hypotrophy of skeletal muscle. weightlifting. Canadian Journal of Applied Medicine 16, 507–513. European Journal of Applied Physiology 69, Physiology 19, 400–410. Häkkinen, K., Pakarinen, A., Alen, M. 88–91. Fry, A.C., Kraemer, W.J. & Ramsey, L.T. & Komi, P.V. (1985) Serum hormones Jensen, J., Oftebro, H., Breigan, H. et al. (1998) Pituitary adrenal–gonadal during prolonged training of (1991) Comparison of changes in responses to high-intensity resistance neuromuscular performance. European testosterone concentration after strength exercise overtraining. Journal of Applied Journal of Applied Physiology 53, 287–293. and endurance exercise in well trained Physiology 85, 2352–2359. Häkkinen, K., Pakarinen, A., Alen, M., men. European Journal of Applied Fryburg, D.A., Gelfand, R.A. & Barrett, Kauhanen, H. & Komi, P.V. (1987) Physiology and Occupational Physiology E.J. (1991) Growth hormone acutely Relationship between training volume, 63, 467–471. stimulates forearm muscle protein physical performance capacity, and Jeˇzova, D., Vigaˇs, M., Tatár, P. et al. (1985) synthesis in normal humans. American serum hormone concentrations during Plasma testosterone and catecholamine Journal of Physiology 260, E499–E504. prolonged training in elite weight lifters. responses to physical exercise of Galbo, H. (1983) Hormonal and Metabolic International Journal of Sports Medicine 8 different intensities in men. European Adaptation to Exercise. Thieme, Stuttgart, (suppl.), 61–65. Journal of Applied Physiology and Germany. Häkkinen, K., Pakarinen, A., Alen, M., Occupational Physiology 54, 61–66. Galbo, H., Hummel, L. Petersen, I.B., Kauhanen, H. & Komi, P.V. (1988a) Jürimäe, T.K., Karelson, K., Smirnova, T. Christensen, N.J. & Bie, W. (1977) Neuromuscular and hormonal & Viru, A. (1990) The effect of a single- Thyroid and testicular responses to adaptations in athletes to strength circuit weight-training session on the graded and prolonged exercise in man. training in 2 years. Journal of Applied biochemistry of untrained university European Journal of Applied Physiology 36, Physiology 65, 2406–2412. students. European Journal of Applied 101–106. Häkkinen, K., Pakarinen, A., Alen, M., Physiology and Occupational Physiology Galbo, H., Kjær, M. & Secher, N.H. (1987) Kauhanen, H. & Komi, P.V. (1988b) 61, 344–348. Cardiovascular, ventilatory and Neuromuscular and hormonal Kjær, M., Secher, N.H., Bach, F.W. & catecholmine responses to maximal responses in elite athletes to two Galbo, H. (1987) Role of motor center exercise in partially curarized man. successive strength trainings. European activity for hormonal changes and Journal of Physiology 389, 557–568. Journal of Applied Physiology 57, 133–139. substrate mobilization in humans Guezennec, C.Y., Legar, L., Lhoste, F., Häkkinen, K., Pakarinen, A, Newton, R.U. American Journal of Physiology 253, Aymonod, M. & Pesquues, P.C. (1986) & Kraeer, W.J. (1998) Acute hormone R687–R697. Hormones and metabolite responses responses to heavy resistance lower and Kjær, M., Secher, N.H., Bach, F.W. & to weight-lifting training sessions. upper extremities exercise in young Galbo, H. (1989) Hormonal and International Journal of Sports Medicine 7, versus old men. European Journal of metabolic responses to exercise in 100–105. Applied Physiology 77, 312–319. humans: effect of sensory nervous Hackney, A.C. (1996) The male Hansen, S., Kvorning, T., Kjær, M. & blockade. American Journal of Physiology reproduction system and endurance Sjøgaard, G. (2001) The effect of short- 257, E85–E97. resistance exercise and testosterone 337

Kochakian, C.D. (1935) Effect of male intensity strength and endurance Lamb, R.D. (1989) Anabolic steroids and hormone on protein metabolism of training on hormonal and skeletal athletic performance. In: Hormones and castrate dogs. Proceedings of the Society muscle adaptations. Journal of Applied Sport (Lamb, R.D., ed.). Raven Press, for Experimental Biology and Medicine 32, Physiology 78, 976–989. New York: 257–273. 1064–1065. Kraemer, W.J., Häkkinen, K, Newton, R.U. Lehmann, M., Keul, L., Huber, G. & Prada, Kochakian, C.D. (1959) Mechanisms et al. (1995b) Factors in various strength M. (1981) Plasma catecholamines in of androgen action. Laboratory and power performance in man. In: trained and untrained volunteers during Investigations 8, 538–556. Proceedings of the XVth Congress of the graduated exercise. International Journal Kochakian, C.D., Costa, G. & Hill, J. (1964) International Society of Biomechanics. of Sports Medicine 2, 143–147. Effect of testosterone propionate on University of Jyväskylä, Jyväskylä, Lehmann, M., Foster, J.M., Steinacker, J.M. metabolism and incorporation of Finland: 508–509. et al. (1997) Training and overtraining: glysine-2-C 14 in guinea pig tissue Kraemer, W.J., Fleck, S.J. & Evans, W.J. overview and experimental results. preparations. Acta Endocrinologica 46 (1996) Strength and power training: Journal of Sports Medicine and Physical (suppl.), 92. physiological mechanisms of Fitness 37, 7–17. Konovalova, G., Masso, R., Ööpik, V. & adaptation. Exercise and Sports Science Lehmann, M., Foster, C., Dickhuth, H. Viru, A. (1997) Significance of thyroid Reviews 24, 363–397. & Gastmann, U. (1998) Autonomic hormones in post-exercise incorporation Kraemer, W.J., Häkkinen, K., Newton, R.U. imbalance hypotheses and overtraining of amino acids into muscle fibers: an et al. (1998a) Acute hormonal responses syndrome. Medicine and Science in Sports autoradiographic study. Endocrinology to heavy resistance exercise in younger and Exercise 30, 1140–1145. and Metabolism 4, 25–31. and older men. European Journal of Lehmann, M., Foster, C., Gastmann, U., Krabbe, S., Transbø´l, I. & Cristensen, C. Applied Physiology 77, 206–211. Keizer, H. & Steinacker, J.M. (1999) (1982) Bone mineral homeostasis, bone Kraemer, W.J., Volek, J.C., Bush, J.A., Definition, types, symptoms, findings, growth and mineralization during years Putukian, M. & Sebastianelli, W.J. underlying mechanisms, and frequency of pubertal growth: a unifying concept. (1998b) Hormonal responses to of overtraining and overtraining Archives for Diseases of Childhood 57, consecutive days of heavy-resistance syndrome. In: Overload, Performance 359–375. exercise with or without nutritional Incompetency, and Regeneration in Sport Kraemer, R.R., Heleniak, R.J., Tryniecki, supplementum. Journal of Applied (Lehmann, M., Foster, C., Gastmann, U., J.L. et al. (1995) Follicular and luteal Physiology 85, 1544–1555. Keizer, H. & Steinacker, J.M., eds.). phase hormonal responses to low- Kraemer, W.J., Konziris, L.P., Ratamess, Kluwer Academic/Plenum, New York: volume resistance exercise. Medicine and N.A. et al. (2002) Detraining produces 1–6. Science in Sports and Exercise 27, 809–817. minimal changes in physical Liao, S. (1977) Molecular action of Kraemer, W.J., Patton, J.F., Knuttgen, H.G. performance and hormonal variables in androgens. In: Biochemical Action of et al. (1989) Hypothalamic–pituitary– recreationally strength-trained men. Hormones (Litwack, G., ed.). Academic adrenal responses to short-duration Journal of Strength Conditioning Research Press, New York: 351–406. high-intensity cycle exercise. Journal of 16, 373–382. Lubisher, J.L. & Arnold, A.P. (1995) Applied Physiology 66, 161–166. Krotiewski, M., Kral, J.G. & Karlsson, J. Evidence for target regulation of the Kraemer, W.J., Marchitelli, L., Gordon, S.E. (1980) Effects of castration and development of androgen sensitivity in et al. (1990) Hormonal and growth factor testosterone substitution on body rat spinal motoneurons. Development of responses to heavy resistance protocols. composition and muscle metabolism in Neurosciences 17, 106–117. Journal of Applied Physiology 69, 1442–1250. rats. Acta Physiologica Scandinavica 109, McCall, G.E., Byrnes, W.C., Fleck, S.J., Kraemer, W.J., Gordon, S.E., Fleck, S.J. et al. 233–237. Dickinson, A. & Kraemer, W.J. (1999) (1991a) Endogenous anabolic hormonal Kuipers, H., & Keizer, H.A. (1988) Acute and chronic hormonal responses and growth factor responses to heavy Overtraining in elite athletes: review to resistance training designed to resistance exercise in males and females. and directions for the future. Sports promote muscle hypertrophy. Canadian International Journal of Sports Medicine 21, Medicine 6, 79–92. Journal of Applied Physiology 24, 96–107. 228–235. Kuoppasalmi, K. (1980) Plasma Mainwaring, W.I. (1977) The Mechanism of Kraemer, W.J., Patton, J.F., Knuttgen, H.G. testosterone and sex hormone binding Action of Androgens. Spinger-Verlag, et al. (1991b) Effects of high-intensity globulin in physical exercise. New York. cycle exercise on sympathoadrenal– Scandinavian Journal of Clinical and Matsumoto, A. (1992) Sex steroid medullary response pattern. Journal of Laboratory Investigations 40, 411–418. induction of synaptic reorganization in Applied Physiology 70, 8–14. Kuoppasalmi, K., Näveri, H., Härkönen, the adult neuroendocrine brain. Reviews Kraemer, W.J., Fry, A.S., Warren, B.J. et al. M. & Adlercreutz, H. (1980) Plasma of Neurosciences 3, 287–306. (1992) Acute hormonal responses in elite cortisol, androstendione, testosterone, Mayer, M. & Rosen, F. (1977) Interaction of junior weightlifters. International Journal and lutenizing hormones in running glucocortioids and androgens with of Sports Medicine 13, 103–109. exercise of different intensities. skeletal muscle. Metabolism 26, 937–962. Kraemer, W.J., Fleck, S.J., Dziados, J.E. Scandinavian Journal of Clinical and Mazur, A. & Booth, A. (1998) Testosterone et al. (1993) Changes in hormone Laboratory Investigations 40, 403–409. and dominance in human males. concentrations after different heavy- Kurtz, E.M., Sengelaub, D.R. & Arnold, Behavior and Brain Sciences 21, 353–397. resistance exercise protocol in women. A.P. (1986) Androgens regulate the Mero, A. (1988) Blood lactate production Journal of Applied Physiology 75, 594–604. dendritic length of mammalian and recovery from anaerobic exercise in Kraemer, W.J., Patton, J.F., Gordon, S.E. motoneurons in adulthood. Science 232, trained and untrained boys. European et al. (1995a) Compatibility of high- 395–398. Journal of Applied Physiology 57, 660–666. 338 chapter 23

Mero, A, Jaakola, L. & Komi, P.V. (1990) release is dependent on a threshold (1992) Stability and variability in Serum hormones and physical intensity of running exercise in male hormone responses to prolonged performance capacity in young boy endurance athletes. Life Sciences 43, exercise. International Journal of Sports athletes during 1-year training period. 551–558. Medicine 13, 230–235. European Journal of Applied Physiology 60, Ritzen, M., Aperia, A., Hall, K. et al. (eds) Viru, A., Smirnova, T., Karelson, K., 32–37. (1981) The Biology of Normal Human Snegovskaya, S. & Viru, M. (1996) Nieschlag, E. & Behre, H.M. (1998) Growth. Raven Press, New York. Determinants and modulators of Testosterone Action, Deficiency, and Rogozkin, V.A. (1979) Metabolic effects of hormonal responses in exercise. Biology Substitution. Springer-Verlag, New York. anabolic steroids. Medicine and Science in of Sport 13, 169–187. Nindl, B.C., Kraemer, W.J., Gotschalk, L.A. Sports 11, 160–163. Viru, A., Laaneots, L., Karelson, K., et al. (2001) Testosterone response after Rogozkin, V.A. & Feldkoren, B.I. (1979) Smirnova, T. & Viru, M. (1998) Exercise- resistance exercise in women: influence The effect of retabolil and training on induced hormone responses in girls at of regional fat distribution. International activity of RNA polymerase in skeletal different stages of sexual maturation. Journal of Sports Nutrition and Exercise muscle. Medicine and Science in Sports 11, European Journal of Applied Physiology 76, Metabolism 11, 451–465. 345–347. 801–805. Olweus, D., Mattson, Å., Schalling, D. & Schwab, R., Jounson, G.O., Housh, T.J., Viru, M. & Viru, A. (2000) Hormonelle Löw, H. (1988) Circulating testosterone Kinder, J.E. & Weir, J.P. (1993) Acute Veränderungen in der Taper. Phase levels and aggression in adolescent effects of different intensities of weight umittelbar vor einem Wettkampf. males: a causal analysis. Psychosomatic lifting on serum testosterone. Medicine Leisungssport 30(5), 4–7. Medicine 50, 261–272. and Science in Sports and Exercise 25, Volek, J.S., Boetes, M., Bush, J.A. et al. Op’Teijnde, B. & Hespel, P. (2001) Short- 1381–1385. (1997a) Response of testosterone and term creatine supplementation does not Scow, R.V. & Hagen, S.H. (1955) Effect cortisol concentration to high intensity alter the hormonal response to resistance of testosterone propionate on myosin, resistance exercise following creatine training. Medicine and Science in Sports collagen and other protein fractions in supplementation. Journal of Strength and Exercise 33, 449–453. striated muscle of gonadectomized Conditioning Research 11, 182–187. Padron, R.S., Wischussen, J. & Hudren, B. male guinea pigs. American Journal of Volek, J.S., Kraemer, W.J., Bush, J.A., (1980) Prolonged biphasic response Physiology 180, 31–36. Incedon, T. & Boete, M. (1997b) of plasma testosterone to single Sejersted, O.M., Vøllenstad, N.K. & Testosterone and cortisol in relation to intramuscular injection of human Medbø, J.I. (1986) Muscle fluid and dietary nutrients and resistance exercise. choronic gonadotropin. Journal of electrolyte balance during and following Journal of Applied Physiology 82, 49–54. Clinical Endocrinology 50, 1100–1104. exercise. Acta Physiologica Scandinavica Weiss, L.W., Cureton, K.J. & Thomson, Palacios, A., Campfield, L.A., McClure, 128 (suppl. 556), 119–127. F.N. (1983) Comparison of serum R.D., Stein, B. & Swerdoloff, R.S. (1983) Sutton, J.R., Coleman, M.J. & Casey, J.H. testosterone and androstendione Effects of testosterone enanthate on (1978) Testosterone production rate responses to weight lifting in men and hematopoiesis in normal men. Fertility during exercise. In: Third International women. European Journal of Applied and Sterility 40, 100–104. Symposium on Biochemistry of Exercise Physiology 50, 413–419. Papanicolaou, G.N. & Falk, E.A. (1938) (Landry, F. & Orban, W.A., eds.). Weltman, J.A., Seip, R.L., Weltman, A. General muscular hypertrophy induced Symposia Specialists, Miami: 227–234. et al. (1990) Release of lutenizing by androgenic hormones. Science 87, Tchaikovski, V.S., Astratenkova, I.V. & hormone and growth hormone after 238–239. Bashirina, O.B. (1986) The effect of recovery from maximal exercise. Journal Port, K. (1991) Serum and saliva exercise on the content of receptor of the of Applied Physiology 69, 196–200. cortisol responses and blood lactate steroid hormone in rat skeletal muscle. Wierman, M.E. & Crowley, W.F. (1986) accumulation during incremental Journal of Steroid Biochemistry 24, Neuroendocrine control of the onset exercise testing. International Journal of 251–253. of puberty. In: Human Growth: A Sport Medicine 12, 490–494. Urban, R.J., Bodenburg, Y.H., Gilkson, C. Comprehensive Treatise, vol. 2. Postnatal Potteinger, J.A., Judge, L.W., Cerny, J.A. & et al. (1995) Testosterone administration Growth: Neurobiology, 2nd edn. (Falkner, Pottinger, V.M. (1995) Effects of altering to elderly man increases skeletal muscle F. & Tanner, J.M., eds.). Plenum, New training volume and intensity on body strength and protein synthesis. American York: 225–241. mass, performance, and hormonal Journal of Physiology 269, E820–E826. Wilkerson, J.E., Horwath, S.M. & Gutin, G. concentrations in weight-event athletes. Urhausen, A., Gabriel, H. & Kindermann, (1980) Plasma testosterone during Journal of Strength Conditioning Research W. (1995) Blood hormones as markers of treadmill exercise. Journal of Applied 9, 55–58. training stress and overtraining. Sports Physiology 49, 249–253. Pullinen, T., Mero, A., MacDonald, E., Medicine 30, 351–376. Wilson, J.D. (1988) Androgen abuse by Pakarinen, A. & Komi, P.V. (1998) Viru, A. (1992) Plasma hormones and athletes. Endocrinological Review 9, Plasma catecholmine and serum physical exercise. International Journal of 181–199. testosterone responses to four units of Sports Medicine 13, 201–209. Youl, K.H., Hwak, K.S., JunLee, W. & resistance exercise in young and adult Viru, A. (1995) Adaptation in Sports Bryne, H.K. (2002) Effects of ginseng male athletes. European Journal of Applied Training. CRC Press, Boca Raton, FL. ingestion on growth hormone, Physiology 77, 413–420. Viru, A. & Viru, M. (2001) Biochemical testosterone, cortisol, and insulin-like Rahkila, P., Hakala, E., Alén, M., Monitoring of Sport Training. Human growth factor 1 responses to acute Salminen, K. & Laatikainen, T. (1988) Kinetics, Champaign, IL. resistance exercise. Journal of Strength Beta-endorphin and corticotropin Viru, A., Karelson, K. & Smirnova, K. Conditioning Research 16, 179–183. Chapter 24

Exercise Response of β-Endorphin and Cortisol: Implications on Immune Function

ALLAN H. GOLDFARB

β Introduction molecules can influence its synthesis. E is also synthesized within the brain and spinal cord, has β-Endorphin (βE) and cortisol are two important potent opioid actions within the central nervous neurohormones that influence immune response system and appears to modulate pain. and glucose levels that ultimately arise from a com- βE released into the circulation arises primarily mon molecule. This common molecule (preprohor- from the anterior pituitary gland. The large peptide mone) proopiomelanocortin (POMC) (Fig. 24.1) can POMC has a section towards the C-terminus known be cleaved to several peptide components. POMC as β-lipotropin that ultimately is cleaved to γ- is not only the precursor of adrenocorticotrophic lipotropin and βE. Both β- and γ-lipotropin molecules hormone (ACTH) that stimulates the production of help to mobilize lipid molecules from adipose tis- cortisol in the adrenals but also contains the peptide sue. βE within the circulation has been implicated in βE. The production of POMC and ultimately both a number of processes including modulation of cortisol and βE are regulated by factors that arise immune function, pain modulation and assisting in from the hypothalamus and the paraventricular glucose homeostasis. βE receptors have been identi- nucleus (PVN) in the brain. Corticotropin-releasing fied in numerous sites in the body including adipose hormone (CRH) arises from the hypothalamus and tissue, pancreas and skeletal muscle. However, the is the major stimulant to activate the release of exact role(s) βE may have on these tissues is still ACTH from the anterior pituitary gland. Argenine being elucidated. vasopression, which arises from the PVN, is also an Cortisol, the major glucocorticoid, has a negative activator of ACTH release into the circulation. There feedback impact on its own secretion at the level are numerous factors that can alter the release of of the anterior pituitary and the hypothalamus. POMC such as diurnal, emotional, physical and bio- Cortisol acts by binding to a cytosolic receptor, and chemical signals. Circulating cortisol will feedback this complex is then moved to the nucleus where it and inhibit the production of POMC although other binds to modulate gene expression. In this manner,

Signal peptide POMC β N N-terminal peptide J-peptide ACTH (1–39) -LPH (1–89) C Anterior lobe of N-terminal peptide (1–76) J-peptide ACTH (1–39) γ-LPH 1–56 βE 58–89 pituitary (1–30)

Fig. 24.1 Proopiomelanocortin (POMC) in the pituitary. Anterior pituitary lobe of pituitary. ACTH, adrenocorticotrophic hormone; βE, β-endorphin; β-LPH, β-lipotropin; γ-LPH, γ-lipotropin; J, joining peptide. The numbers coincide to the amino acid sequence for that section. 339 340 chapter 24

cortisol is the major inhibitor of the synthesis of ence the βE response to exercise (Goldfarb et al. CRH and the synthesis of POMC. Additionally, 1990; Heitkamp et al. 1996). cortisol has been shown to inhibit the release of syn- Incremental exercise and high intensity anaerobic thesized ACTH stored in vesicles within the anter- exercise has been reported to stimulate βE increases ior pituitary gland. There is a control of cortisol at within the circulation (Metzger & Stein 1984; Farrell the level of the hypothalamus. CRH control from et al. 1987; Goldfarb et al. 1987; Heitkamp et al. 1996). the hypothalamus demonstrates a circadian and Resistance exercise as a stimulus to circulating βE is pulsatile nature that manifests a pulsatile and varied limited. Conflicting reports exist and may be related response throughout the day for these factors. to differences in subjects, type of exercise intensity The greatest pulsatile secretion of ACTH typically and time of measurement. Kraemer, W.J. et al. (1993) occurs in the early morning. It should be noted that reported that βE levels in the circulation increased the suprachiasmatic nucleus of the hypothalamus in response to high total workload. These authors receives input from the optic nerve and this input noted that the total work, rest to work ratio and total has an impact on this circadian rhythm. Elimination force needed probably influenced the βE response. of the optic nerve can eliminate the circadian These authors also reported an increase in βE in 28 rhythm of ACTH and cortisol. elite male weightlifters after a moderate to high The formation of cortisol from cholesterol within intensity workload (Kraemer, W.J. et al. 1992). βE the adrenal cortex takes some time. Therefore, there levels were reported to be elevated in females in will be a delay in the cortisol peak compared to the response to three sets of resistance at 85% of their 1- increased pulsatile peaks of ACTH. The primary repetition maximum (1-RM) (Walberg-Rankin et al. function of cortisol is to help maintain glucose levels 1992). An increased βE/β-lipotrophin level was by enhancing mobilization of amino acids from pro- reported in response to weightlifting in five men teins to the liver to be precursors for glucose. The (Elliot et al. 1984). In contrast, low volume resistance stimulation of gluconeogenesis by cortisol as well as exercise did not result in any change in βE levels the stimulation of fat mobilization to enhance fat (Kraemer, R.R. et al. 1996). It appears that resistance metabolism helps to raise plasma glucose levels. exercise of sufficient intensity and volume (work- Cortisol also acts as an immunosuppressive agent load) can result in a transient increase in βE levels and has anti-inflammatory activity. within the circulation. Cortisol response to exercise has varied depend- Exercise influence on circulating ing on the type of exercise, the intensity of exercise β-endorphin and cortisol and the duration of the exercise. Generally, mild intensity moderate duration aerobic exercise does βE increases within the circulation have been docu- not appear to alter the cortisol level within the cir- mented in response to various aerobic and anaero- culation, although some reports have indicated a bic exercises (Goldfarb & Jamurtas 1997). Several decline in cortisol. In contrast, longer duration exer- studies have reported that circulating βE immuno- cise and more intense exercise have typically elev- reactivity can increase in response to exercise ated circulating cortisol levels. This may be related depending on the intensity of exercise (McMurray in part, to glucose homeostasis. When individuals et al. 1987; Goldfarb et al. 1990; Kraemer, W.J. et al. are given carbohydrate during long duration exer- 1993). It appears that a critical minimum intensity of cise the cortisol response is attenuated. Exercise > o β 60% of V 2max is needed to result in E elevation of sufficient duration generally will demonstrate with aerobic exercise (McMurray et al. 1987; Goldfarb an increase in circulating cortisol (Galbo 1983; et al. 1990, 1991; Rahkila & Laatikainen 1992). How- Petraglia et al. 1988). Cortisol levels increased in ever, this minimum may differ based on the indi- athletes who ran 1500 m and 10 000 m but not in vidual (Viru & Tendzegolskis 1995; Heitkamp et al. athletes whom competed in sprints (100 m) or per- 1996) and nutritional status of the subject. Addition- formed the discus throw (Petraglia et al. 1988). ally, the duration of the exercise appears to influ- Short-term exercise may have only minor changes exercise β-endorphin and cortisol immune function 341

in the cortisol level in plasma (Galbo 1983). It should was dose dependent and was not blocked by be noted that because of diurnal variations in cor- naloxone. Additionally, there was no βE effect on B- tisol levels, the increases are not always observed. lymphocytes. The proliferative response of spleno- Cortisol response to resistance exercise has equiv- cytes of adult male F344 rats to βE was enhanced ocal responses partly due to the intensity of the exer- 50–100% in a dose-dependent manner on T-cells cise and also related to total workload. In a recent (van den Bergh et al. 1991). It was noted that inter- study, it was reported that force repetition exercise leukin 2 (IL-2) was elevated as well as IL-2 receptor increased plasma cortisol levels to a greater extent expression prior to the βE proliferative effect on the that maximal isotonic exercise (Ahtiainen et al. T-cells. Additionally, naloxone was not effective in 2003). The results suggest that the workload has blocking the βE effect. Further evidence on the βE an influence on the cortisol response. In support proliferative effect on human T-lymphocytes was of this finding, resistance exercise using different shown using the mitogen concanavalin A (Owen intensities of 1-RM and varying sets was shown to et al. 1998). βE stimulated the mitogen response by influence the cortisol response (Smilios et al. 2003). 121–750% with a bell-shaped curve indicating that The results suggested that cortisol response was too high a dose would actually inhibit the response. similar in the high intensity exercise independent of It also appears that this response may change with number of sets whereas in lower intensity resistance time, dose or mitogen used (Millar et al. 1990). These exercise four sets increased cortisol more so than authors also noted that the inhibition of the immune two sets. High total work-resistance exercise was response to cortisol maybe partially reversed by βE. reported to increase cortisol as well as βE (Kraemer Therefore, the activation of βE may inhibit suppres- W.J. et al. 1993). It is interesting to note that the ele- sion of the immune response by cortisol in vivo. vation in cortisol was fairly rapid and occurred by The effect of βE on human natural killer cell func- the middle of the exercise as well as immediately tion was studied in vitro and was enhanced when following the activity and during the 15-min recov- βE was present (Kay et al. 1984). They also reported ery period. Not all studies confirm that cortisol that this was a dose dependent response and was levels will increase with high intensity exercise inhibited by naloxone. This suggests that the mode (Volek et al. 1997). These equivocal findings in corti- of action on natural killer cells appears to be differ- sol in response to resistance exercise may be related ent than the enhancement of T-lymphocyte func- in part to diurnal variations, nutritional factors and tion. The βE effect on natural killer cell activity training status of the subjects. (NKCA) and amount with exercise was also studied (Gannon et al. 1998). Naltrexone (an opioid-blocking β agent) was given 60 min prior to a 2-h moderate -Endorphin and immune function o intensity exercise (65% V 2max) and compared to a βE influence on immune function has been investig- placebo trial. The βE levels in the blood increased at ated in vitro but has not been adequately investig- 90 and 120 min with exercise and NKCA and counts ated in vivo. βE (both rat and human) was shown to were elevated. Furthermore, the naltrexone treat- stimulate T-lymphocyte proliferation (Hemmick & ment did not alter the exercise response in NKCA or Bidlack 1990). The data suggests that βE mode of counts. These authors suggested that βE probably action was not through an opioid receptor but might was not involved in the NKCA increase with exer- β be through the inhibition of the prostaglandin E1 cise. However, it is possible that E may work inde- effect on immune function. Synthetic βE was shown pendent of this receptor action ( Jonsdottir et al. to bind to non-opioid receptors on T-lymphocytes 2000). Chronic exercise (wheel running for 5 weeks) and was not blocked by naloxone or Met-enkephalin in spontaneously hypertensive rats enhanced NKCA. (Navolotskaya et al. 2001). The βE levels in cerebrospinal fluid increased after βE was shown in vitro to stimulate rat spleen lym- the running and also enhanced clearance of lym- phocytes by enhancing the proliferative response to phoma cells from the lungs. The δ-receptor antag- several mitogens (Gilman et al. 1982). The response onist naltrindole significantly but not completely 342 chapter 24

inhibited the enhanced NKCA after 5 weeks of exer- influence the immune response by interacting with cise. Neither κ- nor µ-receptor antagonists influ- interleukins (Zitnik et al. 1994). The inhibition may enced the natural killer cell response. These authors be at a number of levels including the reduction suggested a central δ-receptor mediated response of production of interleukin-1 and 6 in a dose- to the exercise training occurred. Peripheral βE dependent manner. βE stimulated the production given subcutaneously did not alter NKCA in vivo of interferon-γ (IFN-γ) from human mononuclear ( Jonsdottir et al. 1996). In contrast, NKCA after cells in response to concanavalin A in cultured central injection of a δ-opioid receptor agonist was cells (Brown & van Epps 1986). IFN-γ was enhanced enhanced (Band et al. 1992). In addition, a single in a dose-dependent manner using physiological βE injection into the intracerebral ventricle of a µ- (10–14–10–10 mol) and was not blocked by naloxone. agonist reduced NKCA. Furthermore, a single It appears that βE may act on a number of morphine injection into the periaquaduct area sup- immune factors both centrally and in the periphery pressed NKCA (Weber & Pert 1989). This suggests and may act through both opioid and non-opioid that central mediated βE may act to modulate receptors. Additionally, the action of βE may work NKCA via both µ-receptors and δ-receptors. Clearly through direct interaction with cortisol. more research with training programs is needed to substantiate the training response in other Cortisol and immune function populations. Additional modes of action of βE on the immune Cortisol has been generally thought of as an response include mononuclear cell chemotaxis (van immunosupressent and anti-inflammatory agent. Epps & Saland 1984; Pasnik et al. 1999), immuno- Corticosteroids given intravenously to humans globulin migration (van Epps & Saland 1984; Saland can induced lymphocytopenia, monocytopenia and et al. 1988) and lymphokine production (van Epps neutrophilia but may take up to 4 h to peak (Rabin & Saland 1984). Macrophages showed migration to et al. 1996). High doses of corticosteroids can result βE injected into the cerebral ventricles in rats (van in cell death of immature T- and B-lymphocytes Epps & Saland 1984). Human neutrophils demon- (Cohen & Duke 1984). Cortisol can regulate the strated enhanced migration to βE infusion and the immune system by induction of apoptosis of response was blocked by prior incubation with thymus and blood lymphocytes (Hirano et al. 2001). naloxone. Analogs of opioids appear to have dif- However, the metabolite of cortisol oxidation, cor- ferent responses when injected into the cerebral tisone can inhibit the apoptosis of these lympho- ventricles (Saland et al. 1988). Some may stimulate cytes. Human monocyte cell death was shown to be macrophages and others may influence neutrophils. enhanced by glucocorticoids (Schmidt et al. 1999). The chemotaxis response appears to be dose de- There was both a time and dose-dependent effect on pendent (Pasnik et al. 1999). High doses of βE monocyte apoptosis. IL-1 mediated activation was (10–3 mol) inhibited the chemotaxis response whereas partially responsible for the enhanced monocyte lower concentrations stimulated up-regulation of apoptosis. In addition, cortisol can inhibit tumor β α α neutrophils. Since, physiological E concentration necrosis factor- (TNF- ) and prostaglandin E2 is below the high dose level even when elevated (PGE2) by activated monocytes/macrophages (Hart by exercise or other stressors, it is probable βE et al. 1990). Interleukin-1 may also feedback to the would have a stimulatory influence on this immune hypothalamus to stimulate ACTH and cortisol syn- function. thesis (Besedovsky & del Rey 1987). Incubation It has been suggested that the opioid peptides of thymocytes and splenocytes with corticosterone such as βE and the enkephalins have a similar struc- in vitro resulted in apoptosis and necrosis of these tural component of interleukin-2 (Jiang et al. 2000). cells after 24 h (Hoffman-Goetz & Zajchowski 1999). Interleukin-2 and other interleukins are involved in The concentration of corticosterone in the medium the inflammatory response and are targets of βE and was similar to the level that near maximal exer- cortisol. It is highly likely that both βE and cortisol cise would produce. This suggests that cortisol may exercise β-endorphin and cortisol immune function 343

contribute to the apoptosis of lymphocytes and re- cortisol will have a time delay of several hours. The duced immunity after the intense exercise. influence of cortisol in response to moderate or low Corticosteroids have been shown to inhibit intensity exercise appears to be minimal on immune NKCA in vitro (Parillo & Fauchi 1978; Pedersen et al. function. However, intense exercise can induce the 1984). NKCA was shown to decrease in blood lym- cortisol response and this has an impact on immune phocytes after intravenous methylprednisolone function. In addition, chronic high intensity exercise pulse therapy in eight rheumatoid arthritic patients can reduce immune function (Pedersen & Hoffman- (Pedersen et al. 1984). Both in vitro and in vivo NKCA Goetz 2000). decreased in response to cortiocosteroids (Parillo & Fauchi 1978). However, the response in vivo was Summary different compared to in vitro. Adhesion of natural killer cell function to target cells was inhibited in Both βE and cortisol influence the immune function, response to pharmacological doses of corticosteroids with βE generally enhancing immune function and in vitro (Hoffman et al. 1981; Pedersen & Beyer 1986). cortisol acting as an immunosupressent. The inter- The natural killer cell activity of mononuclear cells play of βE and cortisol in regulating immune func- was inhibited by methylprednisolone and hydro- tion in response to both acute and chronic exercise cortisone in a dose-dependent manner and inhib- still needs clarification. The exercise effect on central ited adhesion to target cells (Pedersen & Beyer mediated βE and in vivo immune function also needs 1986). Inhibition of adhesion to target cells by corti- clarification. Adaptation effects to training also costeroids in mononuclear cells was dose depend- need further study. In addition, nutritional factors ent and was related to alteration in the methylation (i.e. carbohydrate level and antioxidants) have not of phospholipids (Hoffman et al. 1981). been adequately examined in relation to both βE The decline in immune function associated with and cortisol influence on the immune response.

References

Ahtiainen, J.P., Pakarinen, A., Kraemer, calcium-dependent endonulease in Journal of Physiology 275(6 Pt 2), W.J. & Häkkinen, K. (2003) Acute thymocyte nuclei leads to cell death. R1725–R1734. hormonal and neuromuscular responses Immunopharmacology 132, 38–42. Gilman, S.C., Schwartz, J.M., Milner, R.J., and recovery to forced vs. maximum Elliot, D.L., Goldberg, L., Watts, W.J. & Bloom, F.E. & Feldman, J.D. (1982) repetitions multiple resistance exercises. Orwoll, E. (1984) Resistance exercise Beta-endorphin enhances lymphocyte International Journal of Sports Medicine and plasma β-endorphin/β-lipotrophin proliferative responses. Proceedings of the 24(6), 410–418. immunoreactivity. Life Sciences 34(6), National Academy of Sciences of the United Band, L.C., Pert, A., Williams, W. et al. 515–518. States of America 79(13), 4226–4230. (1992) Central µ-opioid receptors van Epps, D.E. & Saland, L. (1984) Goldfarb, A.H. & Jamurtas, A.Z. (1997) mediate suppression of natural Beta-endorphin and met-enkephalin Beta-endorphin response to exercise: killer activity in vivo. Progress in stimulate human peripheral blood An update. Sports Medicine 24(1), 8–16. Neuroendocrinimmunology 5, 95–101. mononuclear cell chemotaxis. Journal of Goldfarb, A.H., Hatfield, B.D., Sforzo, G.A. van den Bergh, P., Rozing, J. & Immunology (Baltimore, Md.) 132(6), & Flynn, M.G. (1987) Serum β- Nagelkerken, L. (1991) Two opposing 3046–3053. endorphin levels during a graded modes of action of β-endorphin on Farrell, P.A., Kjaer, M., Bach, F.W. & exercise test to exhaustion. Medicine and lymphocyte function. Immunology 72(4), Galbo, H. (1987) Beta-endorphin and Science in Sports and Exercise 19(2), 78–83. 537–543. adrenocorticotropin response to Goldfarb, A.H., Hatfield, B.D., Armstrong, Besedovsky, H. & del Rey, A. (1987) supramaximal treadmill exercise in D. & Potts, J. (1990) Plasma β-endorphin Neuroendocrine and metabolic trained and untrained males. Acta concentration: response to intensity and responses induced by interleukin-1. Physiologica Scandinavica 30(4), duration of exercise. Medicine and Science Journal of Neuroscience Research 18(1), 619–625. in Sports and Exercise 22, 241–244. 172–178. Galbo, H. (1983) Hormonal and Metabolic Goldfarb, A.H., Hatfield, B.D., Potts, J. & Brown, S.L. & van Epps, D.E. (1986) Adaptation to Exercise. Thieme Verlag, Armstrong, D. (1991) Beta-endorphin Opioid peptides modulate production New York. time course response to intensity of of interferon γ by human mononuclear Gannon, G.A., Rhind, S.G., Suzui, M. et al. exercise: effect of training status. cells. Cell Immunology 103(1), 19–26. (1998) Beta-endorphin and natural killer International Journal of Sports Medicine Cohen, J.J. & Duke, R.C. (1984) cell cytolytic activity during prolonged 12(3), 264–268. Glucocorticoid activation of a exercise. Is there a connection? American Hart, P.H., Whitty, G.A., Burgess, D.R., 344 chapter 24

Croatto, M. & Hamilton, J.A. (1990) Kraemer, W.J., Dziados, J.E., Marchitelli, adrenocorticotropin, β-endorphin, Augmentation of glucocorticoid action L.J. et al. (1993) Effects of different β-lipotropin and cortisol to athletic on human monocytes by interleukin-4. heavy-resistance exercise protocols on competition. Acta Endocrinologica 118(3), Lymphokine Research 9(2), 147–153. plasma β-endorphin concentrations. 332–336. Heitkamp, H-Ch., Schmid, K. & Journal of Applied Physiology 74(1), Rabin, B.S., Moyna, M.N., Kusnecov, A., Scheib, K. (1996) Beta-endorphin and 450–459. Zhou, D. & Shurin, M.R. (1996) adrenocorticotrophic hormone McMurray, R.G., Forsythe, W.A., Mar, Neuroendocrine effect of immunity. In: production during marathon and M.H. & Hardy C.J. (1987) Exercise Exercise and Immune Function (Hoffman- incremental exercise. European Journal of intensity-related responses of β- Goetz, L., ed.). CRC Press, Boca Raton, Applied Physiology 66(3), 269–274. endorphin and catecholamines. Medicine FL: 21–38. Hemmick, L.M. & Bidlack, J.M. (1990) and Science in Sports and Exercise 19, Rahkila, P. & Laatikainen, T. (1992) Effect Beta-endorphin stimulates rat T 570–574. of oral contraceptives on plasma β- lymphocyte proliferation. Journal of Metzger, J.M. & Stein, E.A. (1984) Beta- endorphin and corticotropin at rest Neuroimmunology 29(1–3), 239–248. endorphin and sprint training. Life and during exercise. Gynecological Hirano, T., Horigome, A., Takatani, M. & Sciences 34, 1541–1547. Endocrinology 6(3), 163–166. Oka, K. (2001) Cortisone counteracts Millar, D.B., Hough, C.J., Mazorow, D.L. & Saland, L.C., van Epps, D.E., Maez, D., apoptosis-inducing effect of cortisol in Gootenberg, J.E. (1990) Beta-endorphin’s Gutierrez, L. & Samora, A. (1988) Acute human peripheral-blood mononuclear modulation of lymphocyte proliferation infusion of chemotactic or enkephalin- cells. International Immunopharmacology is dose, donor, and time dependent. analog peptides into rat cerebral 1(12), 2109–2115. Brain, Behavior, and Immunity 4(3), ventricles: scanning and transmission Hoffman-Goetz, L. & Zajchowski, S. (1999) 232–242. electron microscopy of leukocyte In vitro apoptosis of lymphocytes after Navolotskaya, E.V., Malkova, N.V., immigration in vivo. Journal of exposure to levels of corticosterone Zargarova, T.A. et al. (2001) Synthetic Neuroimmunology 18(3), 197–206. observed following submaximal β-endorphin-like peptide immunorphin Schmidt, M., Pauels, H.G., Lugering, N. exercise. Journal of Sports Medicine and binds to non-opioid receptors for β- et al. (1999) Glucocorticoids induce Physical Fitness 39, 269–274. endorphin on T lymphocytes. Peptides apoptosis in human monocytes: potential Hoffman, T., Hirata, F., Bougnoux, P. 22(12), 2009–2013. role of IL-1β. Journal of Immunology et al. (1981) Phospholipid methylation Owen, D.L., Morley, J.S., Ensor, D.M. (Baltimore, Md.) 163(6), 3484–3490.

and phospholipase A2 activation in & Miles, J.B. (1998) The C-terminal Smilios I., Pilianidis, T., Karamouzis, M. cytotoxicity by human natural killer tetrapeptide of β-endorphin (MPF) & Tokmakidis, S.P. (2003) Hormonal cells. Proceedings of the National Academy enhances lymphocyte proliferative responses after various resistance of Sciences of the United States of America responses. Neuropeptides 32(2), 131–139. exercise protocols. Medicine and Science 78(6), 3839–3843. Parillo, J.E. & Fauchi, A.S. (1978) in Sports and Exercise 35(4), 644–654. Jiang, C.L., Xu, D., Lu, C.L. et al. (2000) Comparison of the effector cells Viru, A. & Tendzegolskis, Z. (1995) Plasma Interleukin-2: structural and biological in human spontaneous cellular endorphin species during dynamic relatedness to opioid peptides. cytotoxicity: differential sensitivity exercise in humans. Clinical Physiology Neuroimmunomodulation 8(1), 20–24. of effector cells to in vivo and in vitro (Oxford, England) 15(1), 73–79. Jonsdottir, I.H., Johansson, C., Asea, A. et al. corticosteroids. Scandinavian Journal of Volek, J.S., Kraemer, W.J., Bush, J.A., (1996) Chronic intracerebroventricular Immunology 8, 99–107. Incledon, T. & Boetes, M. (1997) administration of β-endorphin Pasnik, J., Tchorzewski, H., Baj, Z., Luciak, Testosterone and cortisol in relationship augments natural killer cell cytotoxicity M. & Tchorzewski, M. (1999) Priming to dietary nutrients and resistance in rats. Regulatory Peptides 62, 113–118. effect of met-enkephalin and β- exercise. Journal of Applied Physiology Jonsdottir, I.H., Hellstrand, K., Thoren, P. endorphin on chemiluminescence, 82(1), 49–54. & Hoffmann, P. (2000) Enhancement of chemotaxis and CD11b molecule Walberg-Rankin, J., Franke, W.D. & natural immunity seen after voluntary expression on human neutrophils in Gwazdauska, F.C. (1992) Response of exercise in rats. Role of central opioid vitro. Immunology Letters 67(2), 77–83. β-endorphin and estradiol to resistance receptors. Life Sciences 66(13), 1231–1239. Pedersen, B.K. & Beyer, J.M. (1986) exercise in females during energy Kay, N., Allen, J. & Morley, J.E. (1984) Characterization of the in vitro effects of balance and energy restriction. Endorphins stimulate normal human glucocorticosteroids on NK cell activity. International Journal of Sports Medicine peripheral blood lymphocyte natural Allergy 41(3), 220–224. 13(7), 542–547. killer activity. Life Sciences 35(1), Pedersen, B.K. & Hoffman-Goetz, L. Weber, R.J. & Pert, A. (1989) The 53–59. (2000) Exercise and the immune system: periaqueductal gray matter mediates Kraemer, R.R., Acevedo, E.O., regulation, integration, and adaptation. opiate-induced immunosuppression. Dzewaltowski, D. et al. (1996) Effects of Physiology Review 80, 1055–1081. Science 14(245), 188–190. low-volume resistive exercise on β- Pedersen, B.K., Beyer, J.M., Rasmussen, A. Zitnik, R.J., Whiting, N.L. & Elias, J.A. endorphin and cortisol concentrations. et al. (1984) Methylprednisolone pulse (1994) Glucocorticoid inhibition of International Journal of Sports Medicine therapy induced fall in natural killer cell interleukin-1-induced interleukin-6 17(1), 12–16. activity in rheumatoid arthritis. Acta production by human lung fibroblasts: Kraemer, W.J., Fry, A.C., Warren, B.J. et al. Pathologica Microbiologica et Immunologica evidence for transcriptional and post- (1992) Acute hormonal responses in elite Scandinavica. Section C 92(5), 319–323. transcriptional regulatory mechanisms. junior weightlifters. International Journal Petraglia, F., Barletta, C., Facchinetti, F. American Journal of Respiratory Cell and of Sports Medicine 13(2), 103–109. et al. (1988) Response of circulating Molecular Biology 10(6), 643–650. Chapter 25

Neuroendocrine Modulation of the Immune System with Exercise and Muscle Damage

MARY P. MILES

The endocrine and nervous systems work together intensity and duration of exercise, the recovery to co-ordinate growth and development, to regulate state of the body and the availability of nutrients homeostasis and to co-ordinate the response of the during the exercise. In many respects, the responses body to stress. The study of the interrelationship and adaptations to exercise training are the cumu- between these systems is termed neuroendocrino- lative influence of repeated acute exercise bouts and logy. The immune system is controlled locally by resources provided to the body for recovery and chemical signals generated at the cellular level and adaptation. Exercise-induced muscle damage activ- systemically by the neuroendocrine system. How- ates the immune system both locally and systemic- ever, complex, bi-directional, anatomical and physio- ally and provides a model to study the inflammatory logical interactions exist among the three interrelated arm of the immune system. The neuroendocrine systems, nervous, endocrine and immune (Maˇsek immune response to muscle damage and other et al. 2003). All three systems have receptors for a stresses to muscle provides insight into the role shared set of ligands including cytokines, peptide that this complex system may play in eliciting hormones and neurotransmitters (Haddad et al. adaptations to exercise training, for example muscle 2002). Thus, the immune system is capable of exert- hypertrophy. Understanding the implications of ing influence on the neuroendocrine system as exercise-induced immune modulations and under- well as vice versa. Exercise is a form of stress to the standing that the immune system may have a role body, and as such, it elicits the stereotypic, neuro- in producing physiological adaptations to exercise endocrine stress response first described in 1936 by is important to the design of exercise programs for Hans Selye as ‘the general adaptation syndrome’ health and for athletic performance. (Selye 1936, p. 32). The complexity of the neuro- endocrine immune system is such that the immune Sympathetic nervous system response to exercise varies according to intensity and duration of exercise, environmental conditions, The autonomic nervous system includes the para- nutritional factors, level of recovery from previous sympathetic nervous system that controls resting training and tissue damage (Nieman 1997; Pedersen functions and the sympathetic nervous system that & Hoffman-Goetz 2000). enables the body to become physically active, as in The aim of this chapter is to examine the neuro- the ‘fight-or-flight’ response. Responses to stress, endocrine immune response to exercise stress from regardless of the nature of the stress, are co- several perspectives within the context of: (i) acute ordinated collectively by the sympathetic nervous exercise; (ii) exercise training; and (iii) exercise- system and the hypothalamic–pituitary–adrenal induced muscle damage. The key immune sys- (HPA) axis (discussed subsequently) (Tsigos & tem components are described in Table 25.1. The Chrousos 2002). Epinephrine and norepinephrine, immune response to acute exercise is influenced by also known as adrenaline and noradrenaline or 345 346 chapter 25 + plex (Th0) cells ediated, + et al . 2002; cells + et al . 2000; Rivest 2001; Suzuki α , IL-1 β IL-6, IL-10, IL-12). Following and are designated Th3 CD4 cells that regulate cellular immunity or the Th2 + cells will be activated for differentiation to either the cells that regulate humoral immunity and some + + Inflammation and natural immune defense against infection. Natural immunity protection against viral infection and tumors, Natural immunity protection against viral infection CD4 (MHC)-restricted cytotoxicity, e.g. killing virally infected cells (antibodies) Produce free radical oxygen species that destroy bacteria MHC-restricted cytotoxicity. Important for cell-m Natural immunity via non-major histocompatibility com Bind to molecular and cellular antigens (particularly bacterial) cytokines (TNF- phagocytosis of cellular debris, production inflammatory to form an antibody-antigen complex that induces phagocytosis by neutrophils and macrophages to eliminate the antigen Stimulation by antigen and cytokines from Th1 type CD4 and injure nearby cells. Remove small debris in the area of Th1 CD4 and some tumor cells. Important for early defense against viruses and some malignancies produce TGF- β phagocytosis and activation, macrophages are able to ‘present’ antigens to T-lymphocytes activate antigen- dependent/acquired immune defenses infection or inflammation by the process of phagocytosis inflammatory functions. A small minority of CD4 CD4 Key functions ‘acquired’ immune responses to kill infected cells T-lymphocytes induces production of immunoglobulins Co-ordination of immune responses. Undifferentiated Produced in bone marrow and released to the circulation. gland. Mature cells found in lymphatic tissue, spleen gland. Mature cells found in lymphatic tissue, spleen and blood in many extracellular fluids including blood and mucous, and stored in lymphatic tissues Produced by plasma cells (B-cells activated for antigen differentiate to become plasma cells. Mature B-cells found production) and released. Found in blood, saliva, mucosal Many adhere to vascular endothelial cells, particularly in the lungs the blood. After leaving circulation, monocytes differentiate to become macrophages secretions and throughout the body and blood Production/location ) Produced by bone marrow, upon activation antigen they + ) to vascular endothelial cells in lymphoid tissues. + 56 + ) ) CD16 + + – Key components of the immune system. (Data from Liles & Van Voorhis 1995; Shephard 1997; Elenkov et al . 2003.) CD8 CD4 + + Helper T-lymphocytes Production by bone marrow and maturation in the thymus (CD3 B-lymphocytes (CD19 Cytotoxic T-lymphocytes(CD3 Production by bone marrow and maturation in the thymus Natural killer (NK)cells (CD3 Production by bone marrow, found in blood and attached Steensberg Immunoglobulins (Ig) Table 25.1 Component Leukocytes Neutrophils Lymphocytes Monocytes/macrophages Monocytes are produced in bone marrow and found immune response to exercise and doms 347 T-cells, + and IL-1 β α production, stimulation of 2 T-cells + T-lymphocyte and NK cells, inhibits IgE + cells + Chemokine, recruits neutrophils to inflammatory sites, induction, recruitment of neutrophils and monocytes monocytes and macrophages, stimulates B-lymphocyte Stimulates the Th1 immune pathway, promotes cytotoxic activation of the HPA axis, inhibition TNF- Inhibition of NK cell activity, B- and T-cell proliferation IL-2 production, stimulation of acute phase protein synthesis, Stimulation of Th2 cells, stimulation immunoglobulin Th1 cytokine, potent stimulation of NK cell activity, Anti-tumor activity, initiation of inflammation via chemokine Induction of cerebral response to inflammation, e.g. fever, Stimulates macrophage activation, neutrophils and NK cells and production of antibodies by B-cells; inhibits Th2 cytokine production by CD4 induction of IL-6 synthesis IL-2 receptor expression, induces IL-6 synthesis of lymphocyte proliferation and antibody secretion by B-cells synthesis, stimulation of IL-10 and IL-1ra synthesis stimulates production of reactive oxygen species and degranulation by neutrophils production and proliferation by B cells, stimulation of allergic responses via IgE production, inhibits cytokine production by Th1 CD4 some macrophage functions; stimulation of IgA secretion by B-lymphocytes proliferation and antibody secretion secretion from B-lymphocytes activity of CD8 stimulation of prostaglandin E T-cells (Tc1 subset), T-cells (Tc1 subset) T-cells (Tc1 subset) and Th2 cytokine, stimulates B- and cytotoxic T-cells, induction of T-cells, monocytes and Th2 cytokine, inhibits cytokine production by Th1 CD4 + + + + T-cells, CD8 T-cells, CD8 T-cells, B-cells T-cells, CD8 + + + + T-cells, macrophages and other cells + Th0 and Th2 CD4 Produced by monocytes, macrophages and endothelial cells Th0 and Th1 CD4 Produced by Th0 and Th2 CD4 Th0 and Th1 CD4 Produced by monocytes monocytes/macrophages, but nearly all cells can be B-cells, as well pituitary and hypothalamic cells NK cells Produced by monocytes/macrophages and NK cells to a lesser extent by neutrophils, T- and B-lymphocytes other cells induced to produce IL-6, most notably muscle cells Th0 and Th1 CD4 α (IL-1 β ) Produced by monocytes/macrophages (TGF- β ) β (IL-10) Tumor necrosis factor- (TNF- α ) Interleukin-1 β Interleukin-2 (IL-2) Interleukin-4 (IL-6) (IFN- γ ) Interleukin-8 (IL-8) Interleukin-10 Cytokines Interferon- γ (IL-4) Interleukin-6 factor Interleukin-12 Transforming growth Th3 CD4 (IL-12) 348 chapter 25

simply catecholamines, are the neurotransmitters This ‘hard-wire’ connection has a major role in func- released by the sympathetic nervous system. Ac- tional modulation of immune cells, but it has little tivation of the sympathetic nervous system via the effect on exercise-induced leukocyte trafficking. locus ceruleus–norepinephrine system (LC/NE) stimulates release of epinephrine from the adrenal Leukocyte function medulla and norepinephrine from axon terminals of sympathetic neurons. Both of these neurotrans- The view that catecholamines have an immuno- mitters increase in the bloodstream during exercise, suppressive net effect is losing favor to a view that although the relative concentrations of norepine- recognizes a more complex variety of responses phrine exceed those of epinephrine by several (Elenkov et al. 2000). As illustrated in Fig. 25.1, the orders of magnitude (Weicker & Werle 1991; Kjær acute influence of catecholamines on the immune & Dela 1996). There is a linear increase in the con- system is complex, but generally leads to suppres- centration of catecholamines as the duration of exer- sion of the Th1 system (production of interleukin-2 cise increases (Kjær & Dela 1996). In contrast, the [IL-2], interferon-γ [IFN-γ] and regulation of cellu- increase in catecholamines in response to increas- lar immunity), no direct effect on the Th2 system ing exercise intensity is of a greater magnitude, (production of IL-4, IL-5, IL-6 and IL-10, and regula- more closely approximating an exponential increase tion of humoral immunity) and mixed effects on the (Kjær & Dela 1996). inflammatory system. A portion of the proinflam- matory response to catecholamines is the result of disinhibition of inflammatory cytokine synthesis Leukocyte trafficking that occurs when catecholamines inhibit IL-2 and The most dramatic effect of catecholamines on the IL-12 synthesis, i.e. stimulation via removal of immune system is to draw leukocytes from extra- inhibition. In contrast to acute exercise, chronic vascular storage depots to the circulation. Research exposure to catecholamines results in desensitiza- studies involving infusion of epinephrine or nore- tion and a cell type-specific down-regulation of β pinephrine to match exercise levels or involving 2-adrenergic receptors or other components of cell blockade of catecholamine receptors (adrenergic signaling (Elenkov et al. 2000). Thus, acute and receptors) during exercise clearly indicate that chronic stresses may have divergent effects on the β β epinephrine (ligand for 1- and 2-adrenergic recep- immune system. tors) recruits lymphocytes and neutrophils to the Catecholamines also are capable of modulating circulation during exercise (van Titts et al. 1990; natural killer (NK) cell function. Both epinephrine Kappel et al. 1991; Benschop et al. 1994; Schedlowski and norepinephrine, but particularly epinephrine, β et al. 1996). Norepinephrine (strong 1- and weak have induced increases in circulating NK cells and β 2-adrenergic receptor ligand) has a smaller effect decreases in the per cell cytotoxic activity of those on circulating leukocytes than epinephrine. Thus, it cells (Schedlowski et al. 1993; Klokker et al. 1997; generally is accepted that increased intracellular Kappel et al. 1998). The decrease in activity is likely cyclic adenosine monophosphate (cAMP) induced to be a function of catecholamine-induced decreases β by epinephrine binding to 2-adrenergic receptors in IL-2 and IL-12, cytokines that up-regulate NK-cell is a primary stimulus driving lymphocytes and cytotoxicity (see Fig. 25.1). neutrophils into the circulation (Boxer et al. 1980; While many cellular responses are induced dir- Weicker & Werle 1991; Schedlowski et al. 1996). ectly by rises in intracellular cAMP (Borger et al. Tissues where immune cells are produced or 1998), it has been suggested that the functional reside, including the thymus, spleen, lymph nodes, modulations of lymphocytes in response to cate- tonsils, bone marrow and gut-associated lymphoid cholamine increases are mediated by macrophages tissue (GALT), are innervated by the sympathetic and nitric oxide (Rabin et al. 1996). Evidence for this nervous system via noradrenergic and or neuropep- mechanism has also been presented in the rodent tide Y (NPY) nerve terminals (Elenkov et al. 2000). model (Blank et al. 1997). An example relevant to immune response to exercise and doms 349

STRESS

CNS (+)

Hypothalamus LC/NE (+)

Pituitary Spinal cord

(+) (+) Adrenal gland Sympathetic nerve terminals

Lymphatic tissue Other Cortex Medulla NE/NPY

Vascular tissue

E Circulation NE Bone Myocardium marrow

E/NE β-adrenergic stimulation

Immune cell responses: Th1 responses: Th2 responses: Inflammatory responses: α Leukocyte demargination TNF- IL-4 IL-8 IFN-γ GM-CSF Neutrophil phagocytosis and oxidative burst IL-2 MIP-1α Natural killer cell cytolytic activity IL-12 IL-10 B-cell Ig production/secretion IL-6 TGF-β Acute phase response

Fig. 25.1 (+) Indicates stimulation; dashed arrows indicate effect induced by removal of inhibition; boxes indicate neurotransmitters. CNS, central nervous system; E, epinephrine; GM-CSF, granulocyte-macrophage colony-stimulating factor; Ig, immunoglobulin; IL, interleukin; LC/NE, locus ceruleus–norepinephrine; NE, norepinephrine; NPY, neuropeptide Y; TGF, transforming growth factor; TNF, tumor necrosis factor. (Data from Sanders et al. 1997; Borger et al. 1998; Elenkov et al. 2000; Kohm & Sanders 2001.) 350 chapter 25

acute exercise is that Kappel et al. (1991) were able to presses a number of immune responses and is a key demonstrate that NK-cell activity per cell was lower regulatory element preventing the immune system 2 h following epinephrine infusion. There was a from over responding to immune challenges and twofold increase in monocytes at this time, and it becoming destructive. For example, inflammation was hypothesized that prostaglandins produced by left unchecked will lead to excessive tissue destruc- the monocytes inhibited the cytolytic activity of the tion and possibly death (Northoff et al. 1995; Suzuki NK cells. As hypothesized, when prostaglandin et al. 2002). production by monocytes was blocked using indo- The effects of cortisol on the immune system are methacin, the per cell decrease in NK-cell activity consistent with down-regulation of immune func- also was blocked. This suggests that prostaglandins tion, particularly down-regulation of inflammatory produced by monocytes, rather than a direct effect functions. For this reason, glucocorticoid analogs of epinephrine, inhibited NK-cell function. By con- are commonly used to treat inflammatory and auto- trast, it should be noted that Nieman et al. (1995a) immune pathologies (Ashwell et al. 2000). Binding measured decreased NK-cell activity despite indo- of cortisol to the intracellular glucocorticoid recep- methacin administration. Thus, this mechanism tor leads to activation of a glucocorticoid response may not be consistent across all situations. element (GRE) and an array of transcriptional responses (Pitzalis et al. 2002). Of particular interest Hypothalamic–pituitary–adrenal axis are the up-regulation of annexin I (formerly referred to lipocortin-1) and anti-inflammatory proteins, for The HPA axis consists of signals feeding forward example IL-1 receptor antagonist, and the down- from the hypothalamus to the anterior pituitary and regulation of cell adhesion molecules (CAM) and then to the adrenal cortex (Tsigos & Chrousos 2002; proinflammatory cytokines (Levine et al. 1996; Beishuizen & Thijs 2003). In response to many forms Pitzalis et al. 2002). of stress, the hypothalamus releases corticotropin- The level of HPA activity ideally falls within releasing hormone (CRH) and vasopressin (AVP). a range that allows for mounting an effective CRP, and to a lesser extent AVP, stimulate the anter- immune/inflammatory response when needed, but ior pituitary to produce adrenocorticotropic hor- does not allow excessive immune activity to become mone (ACTH). The main role of AVP is to promote destructive. Excessive stress, in a variety forms, fluid resorption by the kidneys. As an endocrine may result in chronically elevated cortisol levels hormone, ACTH travels through the circulation to and induce immunosuppression (Buckingham et al. stimulate the adrenal cortex to release glucocor- 1996). For example, immunosuppression associ- ticoid hormones, most notably cortisol. As such, ated with depression has been linked to elevations cortisol is the final product of the HPA axis. The in cortisol (Leonard & Song 1996). Reciprocally, effects of cortisol on the immune system and the adrenocortical insufficiency associated with low details of the relationships amongst the HPA axis glucocorticoid release has been associated with in- components and the inflammatory cytokines are creased susceptibility to autoimmune/inflammatory illustrated in Fig. 25.2. Cortisol acts in a negative conditions (Buckingham et al. 1996). Pathological feedback fashion to inhibit CRH and ACTH release. disorders that result in over and underproduction As a steroid hormone, cortisol is able to diffuse of glucocorticoids are identified as Cushing’s and through plasma membranes and bind to intracel- Addison’s diseases, respectively. Within the non- lular receptors (Riccardi et al. 2002). The cortisol/ pathological spectrum of adrenocortical reactivity, glucocorticoid receptor complex affects cellular high and low stress responders have been identified function primarily by up-regulating and down- with respect to the release of ACTH in response to regulating transcription of various proteins, but also stress (Petrides et al. 1997; Deuster et al. 1999). Thus, via more rapid means such as a Ca2+-dependent variability in the magnitude of the HPA response to mechanism (Buckingham et al. 1996). Cortisol sup- stress is expected and a number of factors modulate immune response to exercise and doms 351

Infection/ (+) STRESS tissue trauma

TNF-α CNS β IL-1 (+) IL-6 LIF Hypothalamus Hypothalamic Hypothalamic (–) (–) Pituitary Pituitary Thyroid Gonadal CRH Axis Axis

(+)

Pituitary (–)

TNF-α IL-1β ACTH IL-6 Circulation

Adrenal gland

(+)

Cortex Medulla

Circulation Cortisol Immune cell responses:

T Lymphocyte proliferation Neutrophil release from bone marrow Th1 responses: Th2 responses: Lymphocyte entry into tissues IL-2 IL-4 B cell IgE secretion IL-10 Monocyte/macrophage activation

Fig. 25.2 (+) Indicates stimulation; (–) indicates inhibition; boxes indicate neurotransmitters or hormones. ACTH, adrenocorticotropic hormone; CNS, central nervous system; CRH, corticotropin-releasing hormone; E, epinephrine; IL, interleukin; LC/NE, locus ceruleus-norepinephrine; LIF, leukemia inhibitory factor; NE, norepinephrine; NPY, neuropeptide Y; TGF, transforming growth factor; TNF, tumor necrosis factor. (Data from Weicker & Werle 1991; Chesnokova & Melmed 2002; Haddad et al. 2002; Riccardi et al. 2002; Tsigos et al. 2002.) 352 chapter 25

this variability. This variability also will manifest in Leukocyte trafficking variability in the immune response to stress, regard- less of the stressor. The effects of glucocorticoids on leukocyte migra- The two-way communication between the im- tion from tissue to tissue occur in a delayed time mune and neuroendocrine systems is such that course relative to catecholamines. Shifts in leuko- inflammatory cytokines, particularly tumor nec- cytes typically peak approximately 4 h after eleva- rosis factor-α (TNF-α), IL-1β, IL-6 and leukemia tions in cortisol are induced (Rabin et al. 1996). inhibitory factor (LIF) can stimulate the HPA axis to Broken down by leukocyte type in the circulation, induce the release of cortisol (Mastorakos et al. 1993; the response to glucocorticoid increases are concen- Chesnokova & Melmed 2002; Tsigos & Chrousos tration decreases in lymphocytes and monocytes 2002). LIF is necessary for inflammation-induced and increases in neutrophils (Rabin et al. 1996; release of ACTH and cortisol, and may be a poten- Nieman 1997). The net effect is a rise in leukocyte tiator of TNF-α and IL-1β induced activation of count exclusively owing to an influx of neutrophils. the HPA axis (Chesnokova & Melmed 2000, 2002; Pharmacological analogs also produce this profile. Chesnokova et al. 2002). That is, the chemical signals that promote the inflammatory process also initiate Cytokines a negative feedback loop to turn their own activity down. The reverse of this relationship has also Glucocorticoids are capable of inhibiting produc- been demonstrated. Specifically, most pituitary hor- tion of many cytokines, including IL-1, IL-2, IL-3, mones can be produced by lymphocytes (Carr & IL-4, IL-5, IL-6, IL-8, IL-10, IL-13, granulocyte- Blalock 1990). For example, lymphocytes stimulated macrophage colony-stimulating factor (GM-CSF), with recall antigen or with IL-12 will produce TNF-α and IFN-α (Ashwell et al. 2000; Riccardi et al. growth hormone (Malarkey et al. 2002). 2002). The influence on Th1 cytokine production In addition to modulating the immune system, is greater than that for Th2 cytokine production; glucocorticoids play an important role in enabling for example, IL-12 synthesis is profoundly inhibited the body to respond to stress in such a way that and IL-10 synthesis is only moderately inhibited increases the likelihood of survival in stressful (Ashwell et al. 2000). As such, glucocorticoids are situations. Glucocorticoids inhibit release of sex considered potent inhibitors of cellular immunity steroids and growth hormone (Tsigos & Chrousos and inflammation. 2002), thus removing the likelihood of the need to allocate resources of the body to non-essential Leukocyte function growth. Further, glucocorticoids inhibit the hypothalamic–pituitary–thyroid axis, an effect that The functional consequences of these glucocorticoid- results in a lowering of basal metabolic rate (Tsigos induced genomic and non-genomic effects are wide- & Chrousos 2002). To accommodate the need for spread through the immune system. Both NK- and energy in the tissues forced to respond to stress, T-cell functions are inhibited by cortisol (Ramirez & glucocorticoids promote gluconeogenesis, glyco- Silva 1997; Zhou et al. 1997; Ashwell et al. 2000). The genolysis, lipolysis and proteolysis (McMurray & ability of NK cells to lyse target cells is used to de- Hackney 2000; Steinacker et al. 2004). By promoting termine NK cell activity (NKCA). Cortisol down- gluconeogenesis and glycogenolysis by the liver, regulates NKCA by decreasing synthesis of effector cortisol helps to maintain blood glucose levels and proteins (Zhou et al. 1997). The cellular immune it is considered to be a ‘glucoregulatory’ hormone. response is dependent on the clonal proliferation Thus, growth processes are turned off, fertility is of antigen-specific T- and B-cells. Lymphocyte pro- decreased, metabolic rate is turned down to decrease liferation assays are a popular functional meas- the overall demand for energy on the body, and urement performed in vitro by exposing these cells energy substrates stored for a ‘rainy day’ are made to cytokines or polyclonal mitogens or antigens to available to the body. induce proliferation. T-cell proliferation is inhibited immune response to exercise and doms 353

β by cortisol (Ashwell et al. 2000; Cancedda et al. 2002). 1997 for review). NK cells have the greatest 2- The effect on B-cells is mixed, with inhibition of adrenergic receptor density, followed by CD8+ T- immunoglobulin G (IgG) and IgA and stimulation lymphocytes and B-lymphocytes and monocytes, of IgE synthesis (Ashwell et al. 2000). and then by T-helper type 1 CD4+ T-lymphocytes (Maisel et al. 1990; Schedlowski et al. 1996). T-helper + Acute exercise type 2 CD4 T-lymphocytes are the only lymphocyte subset that does not express β-adrenergic recep- While it is important to understand the effects of the tors (Sanders et al. 1997). Consistent with the cate- sympathetic nervous system and HPA axis hor- cholamine response, the NK response to exercise mones in isolation, exercise-induced immunomo- increases with intensity and duration of exercise dulation is the result of the collective influences of and the reported increases during exercise range all many physiological responses occurring simultan- the way from 50% to 900% (Shephard 1997). The eously or in sequence. Owing to the complexity of remaining lymphocyte subsets respond less dram- the systems involved and the number of potential atically as the density of β-adrenergic receptors factors that can enhance or attenuate the stress to the decreases. Additionally, the cells that enter the system as a whole, the array of potential influences circulation are predominately memory/activated on the immune response is virtually infinite. A cells with short telomeres in the case of lymphocytes single bout of exercise can induce the immune sys- (Bruunsgaard et al. 1999; Pedersen & Hoffman- tem to transiently redistribute leukocytes to the Goetz 2000), and segmented rather than banded bloodstream and amongst tissues, alter the func- nuclei in the case of neutrophils (Miles et al. 1998). tional capacity of leukocytes, induce the release of a This indicates that the influx of cells to the circula- host of molecules that regulate immune function, tion is made up of mature cells, rather than of newly produce transient or prolonged inflammation, and produced or released immature/naïve cells. in doing so, change the overall immune defense Cortisol gains increasingly more influence over level that the body has for protection against infec- that of the catecholamines as the duration of exer- tion and tumor cells. The extent of these changes cise continues. Approximately 1.5 h into endurance typically is greater as the magnitude of the exercise exercise, the influence of cortisol on leukocyte stress increases. The typical determinants of stress distribution and function overtakes that of the magnitude are intensity and duration of exercise catecholamines (Nieman 1997). If the duration (McMurray & Hackney 2000). However, the level of and intensity of exercise are sufficient to create an stress to which the neuroendocrine immune system increase in cortisol, then a cortisol-driven lympho- must respond is enhanced by additional factors cytosis occurs for 1–3 or 4 h post-exercise (Nieman such as insufficient recovery from previous exercise 1997). (Ronsen et al. 2002a, 2002b; McFarlin et al. 2003), low Neutrophils are similar to lymphocytes in that carbohydrate availability (Nieman et al. 1998; Green they increase in the circulation during exercise in a et al. 2003), hypoxic conditions (Klokker et al. 1995; dose–response fashion, and they differ from lym- Niess et al. 2003) and heat stress (Brenner et al. 1998; phocytes in that they have a second wave of eleva- Mitchell et al. 2002). tion that is delayed and sustained in the circulation a few hours following exercise that induces a sig- nificant cortisol increase (Peake 2002). While epine- Leukocyte trafficking phrine is credited with a portion of the induced Activation of the sympathetic nervous system elicits increase during exercise, a number of additional the exercise-induced increases among the different factors have been associated with the rise in neutro- leukocyte subsets. While there is tremendous inter- phils including IL-6, IL-8, granulocyte-colony stimu- individual variability in the magnitude of response, lating factor (G-CSF), growth hormone and muscle the overall pattern of leukocyte redistribution dur- damage (Hetherington & Quie 1985; Suzuki et al. ing exercise is robust and consistent (see Shephard 1999; Peake 2002; Yamada et al. 2002). The delayed 354 chapter 25

elevation of neutrophils appears to be the effect of The single most responsive cytokine to exercise cortisol and includes the entry of both immature is IL-6 which increases exponentially as the dura- band neutrophils from the bone marrow and mature tion of exercise progresses (Pedersen et al. 2001) and segmented cells from marginal pools (Suzuki et al. perhaps in correlation with rises in epinephrine 1996, 1999). (Papanicolaou et al. 1996). However, factors other Monocytes increase in the circulation as part of than epinephrine elicit the vast majority of the the transient response to acute exercise (Woods & exercise-induced rise in IL-6 (Jonsdottir et al. 2000; Davis 1994). This response is likely to be induced by Steensberg et al. 2001). With respect to magnitude enhanced hemodynamic forces and catecholamine- and time course, IL-6 concentrations increase from induced decreases in adhesion to vascular endo- two to 100-fold in the blood during endurance thelial cells (Woods et al. 1999). exercise and typically return to resting levels within a few hours post-exercise. When exercise is of shorter duration and biased toward the develop- Cytokines ment of high eccentric forces within active muscles, The pattern of cytokine production in response to the kinetics are delayed and a relatively small peak exercise is not entirely consistent with the catechola- of IL-6 has been measured a few hours post-exercise mine and glucocorticoid response. That is, stimula- (Bruunsgaard et al. 1997a). tion of the sympathetic nervous system should A key distinction to be made regarding IL-6 inhibit Th1 cytokine production (TNF-α, IL-2, IL-12 is that, while it has a role in the inflammatory and IFN-γ), have little effect on Th2 cytokine pro- response, it is produced more readily by skeletal duction (IL-4), and stimulate inflammatory cytokine muscle during exercise than by leukocytes, adipose, production (IL-6, IL-8, IL-10, transforming growth or the liver (Jonsdottir et al. 2000; Pedersen et al. factor-β [TGF-β]). Activation of the HPA axis should 2001; Febbraio et al. 2003b). A number of physiolo- inhibit production of some Th1 cytokines (IL-2) and gical roles for IL-6 in the response to exercise have some inflammatory cytokines (TNF-α, IL-1β, IL-6), been identified in recent years, and the regulatory and stimulate production of some Th2 cytokines role of IL-6 beyond the inflammatory response is (IL-4 and IL-10). These responses are what would be an emerging area of research (MacDonald et al. expected if catecholamines and glucocorticoids con- 2003; Steensberg et al. 2003). For example, as trolled the cytokine response to exercise. Consistent muscle glycogen is depleted, AMP-activated pro- with this model is the tendency for IL-2 and IFN-γ tein kinase (AMPK) increases, and a strong correla- production not to change or to decrease in response tion between AMPK and IL-6 release has been to exercise (Shephard 1997; Ibfelt et al. 2002; Suzuki measured (MacDonald et al. 2003). Thus, IL-6 may et al. 2002). Also consistent with this model is the be involved in systemic signaling related to energy tendency for IL-8 and IL-10 to increase in response needs of muscle cells during exercise. Plasma IL-6 to exercise (Shephard 1997; Nieman et al. 2001; increases occur in the absence of muscle damage Pedersen et al. 2001; Suzuki et al. 2002). However, and are independent of other markers of inflam- exercise elicits an inflammatory or ‘inflammatory- mation. Exercising, but not resting, skeletal muscle like’ response in a dose-dependent fashion that cells release IL-6 (Febbraio et al. 2003b). Further, includes elevations in TNF-α, IL-1β, IL-6, IL-10 and IL-6 gene expression (IL-6 mRNA) was detected IL-1 receptor antagonist (IL-1ra) (Ostrowski et al. in muscle, but not in mononuclear leukocytes 2000; Pedersen et al. 2001; Suzuki et al. 2002). Thus, (includes monocytes) collected from the blood fol- the exercise-induced increase in cortisol occurs in lowing strenuous exercise (Ostrowski et al. 1998b). part because the rise in inflammatory cytokines Thus, IL-6 will increase in the absence of tissue stimulates the HPA axis via the two-way com- damage and should be considered both as part munication network between the immune and of the inflammatory response and as part of an neuroendocrine systems (Smith, L.L. & Miles 2000; inflammatory-independent response to exercise Steensberg et al. 2003). (Pedersen et al. 2001). immune response to exercise and doms 355

is likely to be improved for a period of time after Leukocyte function exercise. Intensity and duration of exercise play an important One factor attenuating the functional effects of role in the subsequent response of immune cells, cortisol on immune function is that the sensitivity including neutrophils. The effects of exercise are of monocytes to cortisol is down-regulated by exer- somewhat consistent in that exercise that is moder- cise (DeRijk et al. 1996). Dexamethasone inhibition ate in duration and intensity transiently enhances of lipopolysaccharide (LPS) -stimulated IL-6 pro- functions such as phagocytosis, degranulation of duction by monocytes is attenuated following an proteolytic enzymes and oxidative burst (Pyne exercise bout (DeRijk et al. 1996; Smits et al. 1998). 1994). If exercise is of sufficient intensity and or This attenuation allows monocytes to increase IL-6 duration to elicit an increase in circulating gluco- production, despite the anti-inflammatory influence corticoids, then there are a number of functional of cortisol. However, the production of several consequences to cells of the immune system. The proinflammatory cytokines by monocytes, includ- oxidative burst of neutrophils tends to be enhanced ing IL-6, is decreased and the production of the by exercise (Pyne 1994; Suzuki et al. 1996, 1999). anti-inflammatory cytokines IL-10 and IFN-γ is not There is some evidence that exercise-induced in- affected by acute exercise (Smits et al. 1998). Thus, creases in IL-6, growth hormone and glutathione the influence of exercise on cytokine production is may be responsible for priming circulating neutro- complex and further research is needed to clarify phils (Atalay et al. 1996; Suzuki et al. 1999). If the this issue. It is important to remember that IL-6 exercise lasts for a prolonged period of time or is synthesis by monocytes is less significant follow- extremely strenuous, for example a marathon, then ing strenuous exercise than that by skeletal muscle the trend is for these functions to decrease (Müns (Steensberg et al. 2002). 1993; Pyne 1994). This decrease may be a function of In response to exercise of light to moderate the diminished capacity of neutrophils to respond intensity or short duration, i.e. that in which cortisol to stimuli after they have already been activated does not increase and the effects of catecholamines (Pyne 1994). Thus, moderate exercise may enhance prevail, redistribution of immune cells is accompan- and strenuous or intense exercise may reduce the ied by a few modest shifts in immune cell function. first line of defense against bacterial and viral B-cells are responsible for immunoglobulin (IgA, infections. IgE, IgG and IgM) production and increased serum Exercise of a variety of intensities and durations concentrations of both IgA and IgG have been meas- has been measured to enhance several macrophage ured after moderate exercise in some (Nehlsen- functions (Woods et al. 1994, 1999). Chemotaxis is Cannarella et al. 1991) but not all studies (Eliakim the process by which monocytes and other immune et al. 1997). Prolonged exercise decreases the levels cells follow a chemical gradient to move toward of salivary IgA, an indication of decreased mucosal sites of inflammation or infection. Phagocytosis and immune defenses (Walsh et al. 2002). In the long elements of the associated oxidative burst also are term, the process leading to development of anti- stimulated by intense exercise and mediated by body titers upon vaccination is not influenced by corticosteroids (Ortega et al. 1996). While the influ- a single stressful bout of exercise (Nieman 1997). ence of corticosteroids is anti-inflammatory, the net Thus, moderate exercise in which the catechola- effect of down-regulation of various inflammatory mines prevail do not appear to influence humoral stimuli may be an increase in macrophage activa- immunity to a significant extent, but prolonged tion and activity (Woods et al. 1994; Ortega et al. strenuous exercise in which the influence of cortisol 1997). However, Woods et al. (1994) have demon- prevails does decrease immune defense. strated in a murine model that the influence may be Cellular immunity may be lowered following indirect and related to reactive nitrogen molecules, strenuous exercise that includes the elevation of cor- such as nitric oxide. Thus, the ability of monocytes tisol. Antigen presentation by macrophages appears and macrophages to response to injury or infection to be attenuated owing to suppressed expression of 356 chapter 25

major histocompatibility complex II and this sup- more sensitive to cortisol and or strenuous exercise- pression correlates with cortisol elevations (Woods induced inhibition. et al. 1997). Consistent with a reduction in the anti- The NKCA response to exercise must be con- gen presentation process is the measurement of sidered both when NK-cell concentrations are smaller delayed-type hypersensitivity reactions increased, as during and immediately following following a half-marathon (Bruunsgaard et al. exercise, and when concentrations are decreased, 1997b). This indicates that stressful exercise results as in the hours following strenuous or prolonged in smaller cell-mediated immunity and antibody exercise. The mechanisms controlling the exercise- production responses to antigenic challenge. Addi- induced modulation of NKCA are likely to differ tionally, glucocorticoids are capable of inducing during each of these phases. For exercise of moder- apoptosis (programmed cell death) in lymphocytes ate intensity and duration, for example 45 min at o and there is some evidence that strenuous exercise 50% of V 2max, the increases in NK-cell concentra- promotes T-lymphocyte, lymphocyte, monocyte tions and NKCA are of similar magnitude, suggest- and neutrophil apoptosis in various compartments ing that the increase in NKCA is the result of in both humans and rodents (Hsu et al. 2002; Mooren increased NK cells (Nieman et al. 1993b). At higher et al. 2002; Hoffman-Goetz & Quadrilatero 2003). intensities, reports of the response of NKCA on a Recent research in which carbohydrate and placebo per NK cell basis are variable with some reporting supplementation elicited similar cortisol responses an increase in per cell activity (Nieman et al. 1993b; immediately post-exercise measured more apopto- Strasner et al. 1996), and others reporting a decrease sis in the placebo group (Green et al. 2003). This sug- in per cell activity (Nieman et al. 1995b; Nielsen et al. gests that cortisol is not likely to be a causative factor 1996). for apoptosis in response to exercise. During the delayed phase of decreased NK- The effect of exercise on lymphocyte proliferation cell concentrations, NKCA may be reduced to is unclear. The only conclusion that can be drawn approximately half of pre-exercise activity for a few regarding exercise-induced modulations in the hours. The decrease in NK-cell activity has been ability of T- and B-lymphocytes to proliferate is attributed to decrease in NK-cell numbers (Nieman that the data are mixed and inconsistent (Nielsen & 1997; Miles et al. 2002b), inhibition by prostaglandin

Pedersen 1997). Similarly, the effects of strenuous E2 (Pedersen & Ullum 1994; Rhind et al. 1999) and exercise on T- and B-cell functions are mixed. How- inhibition by cortisol (Berk et al. 1990). However, ever, there are many reports of decreased prolifera- the time frame for down-regulation of NKCA by tion responses of T- and or B-lymphocytes to mitogen cortisol does not match the time-course of exercise- stimulation following strenuous, prolonged exer- induced changes in NKCA, thus cortisol is not con- cise (Green et al. 2003). Examination of the propor- sidered a likely modulator of NKCA beyond the tion of activated CD4+ and CD8+ T-lymphocytes in effects of circulating numbers (Nieman et al. 1993b, the circulation expressing the CD69 surface marker, 1995a). revealed that the proportion of activated lympho- One means of distilling out the influence of corti- cytes was consistent across pre- and post-exercise sol is to examine the research comparing endurance measurements (Green et al. 2003). Based on this, exercise with and without carbohydrate supple- a decrease in T-cell proliferation would not be mentation. Carbohydrate supplementation during expected, even following exercise that elicits an prolonged endurance exercise decreases the stress increase in cortisol. Dohi et al. (2002) found that the hormone response (Nieman et al. 1998, 2001; Green proliferation response to a T- and B-cell mitogen et al. 2003). Thus, the difference in immune res- was lower in a group of women who had an increase ponses between conditions may reflect the types of in cortisol compared to those who did not have an functions influenced by cortisol, if the time course increase in cortisol, while there was no exercise- of differences in cortisol matches that of func- induced change in proliferation to T-cell mitogens. tional changes in immune cells. According to this This suggests that B-cell proliferation may be a bit paradigm, it appears that cortisol increases may be immune response to exercise and doms 357

associated with enhanced monocyte and neutrophil gen presentation occurs and has been hypothesized oxidative bursts (Nieman et al. 1998), phagocytosis to be an ‘open window’ of opportunity for weakened (Henson et al. 2000), enhanced production of some immune defenses to be overcome by a virus (Nieman cytokines, for example IL-8 and IL-10 (Nehlsen- 2000). The first line of defense against bacterial or Cannarella et al. 1997; Nieman et al. 2001), but not viral pathogens, including mucosal immunoglobu- with apoptosis (Green et al. 2003) or IL-6 (Nieman lins and the activity of neutrophils in the nasal et al. 2001). Reports on the effect carbohydrate sup- mucosa, is weakened following strenuous exercise plementation on lymphocyte proliferation are mixed (Müns 1993). Similarly, salivary production of IgA (Henson et al. 1998; Mitchell et al. 1998). Given the is reduced following strenuous exercise (Mackinnon two-way interaction between the immune system et al. 1989; Mackinnon & Jenkins 1993; Nieman and the HPA axis, it may be that the inflammatory et al. 2002), further adding to the impaired first responses drive the release of cortisol and not vice line of defense against URTI. Decreased salivary versa. Evidence to support this hypothesis is that IgA is the only immune measure that has been asso- infusion of recombinant IL-6 to match levels typ- ciated with the onset of URTI (Mackinnon 2000). ically measured during exercise elicits a substantial The tendency for both catecholamines and cortisol cortisol response (Steensburg et al. 2003). to inhibit type 1 immune responses and promote type 2 responses tilts the scale toward increase susceptibility to URTI. This pattern of T-cell and Susceptibility to illness cytokine responses has been definitively identified The down-regulation of various immune defenses following 2.5 h of treadmill running (Steensburg following strenuous and prolonged exercise may et al. 2001). Additionally, IL-6 produced during increase the likelihood that a person will succumb to exercise stimulates release of cortisol (Steensberg infection. Athletes are more susceptible to infectious et al. 2003). Thus, there are a number of mechanisms illness, particularly upper respiratory tract infection by which the neuroendocrine and immune systems (URTI), during periods of intense training and in are induced by strenuous and prolonged exercise to the few weeks following particularly stressful com- favor Th2 and inflammatory responses at the petitive events such as a marathon (Peters & expense of the Th1 system. This shift decreases viral Bateman 1983; Nieman 1998). Reports of incidence defenses and may increase the ability of viruses to of URTI in athletes under these conditions vary, but flourish, particularly those in the upper respiratory a rough estimate may be that the risk is approx- tract. imately double. That is, most athletes do not get URTI following strenuous competitions, but the risk Exercise training is significantly increased (Nieman 2000). The down- time in training or the potential affect on perform- The relationship between exercise training or phys- ance during URTI or other infections are issues of ical activity and immune status depends on the substantial concern for many competitive athletes. dose of exercise and likely also on the opportunity Additionally, data from athletes may apply to other for recovery from the stress of exercise. There seems physically demanding situations, including occupa- to be general agreement that moderate levels of tional or leisure time activities. training may stimulate overall immune defenses The mechanisms contributing to increased sus- while strenuous or severe training may suppress ceptibility to illness have not been definitively immune defenses (Nieman 1997; Shephard 1997). It identified; however, a number of known responses is likely that the immune changes induced by train- to strenuous exercise are suspected to contribute. ing are a component of a systemic response to stress, The period following strenuous exercise in which including the neuroendocrine response. Thus, the lymphocyte numbers and functions are decreased, ratio of stress to recovery may be a key component nasal neutrophil phagocytosis is decreased, there is dictating the shift from benefit to detriment as the a decrease in salivary IgA and attenuation of anti- stress of exercise training increases. 358 chapter 25

before and 75% of Vo after endurance training. Leukocyte trafficking 2max Thus, an equivalent absolute work rate requires a The number of leukocytes that reside in the circula- lower relative proportion of the maximal effort after tion at rest may be used as a gross indication of training. This is true for virtually all types of train- immune status; however, the significance of modest ing. With respect to the neuroendocrine response to fluctuations within or near the normal range has exercise, the rise in catecholamines and other hor- not been determined. Large deviations from the mones that occurred for this individual working at normal range can be used to identify pathologies. the same absolute rate will be attenuated before and For example, CD4+ T-cell concentrations can be after training (Kjær et al. 1988). measured to monitor the progression of human The influence of the neuroendocrine system on immunodeficiency virus/acquired immunodefici- leukocyte trafficking during exercise is a function ency syndrome (HIV/AIDS) (Hazenberg et al. 2003), of the relative intensity and duration of the exercise. and neutrophil concentrations can be measured It is the relative exercise intensity rather than the to determine whether inflammation is occurring absolute exercise intensity that dictates the neuro- (Smith, L.L. & Miles 2000). Both longitudinal and endocrine and consequent immune responses to cross-sectional studies have been employed to exercise. Research data consistent with this effect determine whether exercise training significantly of exercise training on the immune response to impacts concentrations of leukocyte subpopula- acute exercise have been presented for endurance tions at rest. A number of studies report increased exercise, resistance exercise and high-intensity numbers of neutrophils and NK cells in endurance interval exercise (Moyna et al. 1996; Shephard 1997; trained compared to non-trained individuals or in Miles et al. 2002a). response to moderate exercise training (Nieman 1997; Shephard 1997). Longitudinal studies of resist- Cytokines ance training have not measured higher NK-cell concentrations in response to training; however, Resting plasma cytokine concentrations typically one study measured an elevation after 3 months of do not differ between trained and untrained indi- training that did not persist through 6 months of viduals (Smith, J.A. et al. 1992; Shephard 1997). training (Flynn et al. 1999; Miles et al. 2002a). Exer- However, higher IL-1β and IL-6 concentrations cise training does not appear to have a consistent have been reported in endurance trained compared influence on resting concentrations of monocytes, to untrained individuals in some studies (Sprenger CD4+ and CD8+ T-lymphocytes, or B-lymphocytes et al. 1992; Mucci et al. 2000). This may be a function (Woods & Davis 1994; Shephard 1997; Miles et al. of decreased sensitivity to cortisol inhibition in the 2002a). With training that is considered strenuous, trained individuals (Duclos et al. 1999) or perhaps typically by athletes at the national or international the cumulative effects of exercise-induced inflam- levels, decreased concentrations of monocytes have mation (Shephard 1997). been reported (Shephard 1997). The absolute magnitude of increases and Leukocyte function decreases in circulating leukocyte populations in response to exercise varies considerably amongst While basal levels of the HPA axis hormones ACTH individuals. The influence of exercise training on and cortisol do not appear to be influenced by the catecholamine response to exercise has not been endurance training, decreased sensitivity of mono- definitively determined. Exercise training induces cytes to cortisol has been measured in endurance an increase in work capacity that decreases the relat- trained compared to untrained men (Duclos et al. ive proportion of maximal work intensity required 1999, 2001). Production of IL-6 by monocytes stimu- for any given absolute work effort. For example, the lated with LPS is greater for endurance trained oxygen consumption associated with running at a 6- compared to untrained men (Duclos et al. 1999). o min per mile (1.609 km) pace may be 90% of V 2max However, following an acute bout of endurance immune response to exercise and doms 359

exercise, the sensitivity of monocytes to glucocor- intensity are greatest, in many, but not all, investiga- ticoids for endurance trained men increased to tions of this phenomenon (Tharp & Barnes 1990; approximately that of the untrained men (Duclos Mackinnon & Hooper 1994; Gleeson et al. 1999; Pyne et al. 1999). Thus, the sensitivity of monocytes to cor- et al. 2000). The endocrine system may be partly tisol is enhanced by endurance training, but is not a responsible for this shift, that also may include static trait. The ramifications of this may reach a vast shifts in cell numbers and reactivity (B-, CD4+ and array of immune functions, as monocytes produce a CD8+ T-cells), the response to tissue damage and number of cytokines that potentially impact all overall stress (Shephard 1997). immune cells. In contrast, 12 weeks of progressive A number of investigations report that exercise resistance training did not influence the stimulated training increases the activity level of individual production of TNF-α, IL-1β, IL-2 or IL-6 by peri- NK cells such that greater NKCA per NK cell in the pheral blood mononuclear cells (PBMCs) collected circulation is consistently measured (Pedersen et al. from younger and older healthy subjects or subjects 1989; Nieman et al. 1990, 1993a; Woods et al. 1999). with rheumatoid arthritis (Rall et al. 1996). Thus, the Recent research using a rodent model (spontan- ability of leukocytes to produce cytokines does not eously hypertensive rats) suggests that: (i) there is a appear to be substantially impacted by exercise dose–response in which in vivo NKCA is enhanced training, with the possible exception of monocytes. as exercise volume increases, but this plateaus off A differential response to exercise has been iden- and no further gains are achieved beyond moderate tified between trained and untrained individuals volumes; and (ii) the gains in NKCA can be abol- with respect to neutrophil adherence and oxidative ished if sufficient rest and recovery do not occur burst (Pyne 1994; Shephard 1997). While this is con- ( Jonsdottir & Hoffmann 2000). This information, in sistent with increased sensitivity to cortisol, this conjunction with the research relating to the acute aspect of the neutrophil response to exercise train- effects of strenuous exercise, suggests that super- ing has not been investigated. There is no consist- imposing new exercise stresses before complete ency amongst reports of other aspects of neutrophil recovery has occurred may lead to an apparently function in response to exercise training or in cross- ‘chronic’ suppression. It is difficult to separate sectional comparisons of trained and untrained out adaptations to exercise training over time from individuals (Shephard 1997). lingering effects of the most recent exercise session Functional activity of T- and B-lymphocytes is not for athletes undergoing strenuous training (Shephard significantly altered in response to exercise training 1997). or in cross-sectional comparisons of trained and Overtraining is a situation in which training untrained subjects (Shephard 1997). Longitudinal volume or intensity exceeds the capacity of the body studies indicate that this is consistent for both to adapt and a feeling of fatigue and performance younger and older individuals and for endurance decrements are typical (Urhausen et al. 1995; Smith, and resistance training (Rall et al. 1996; Woods et al. L.L. 2000). Common characteristics associated with 1999; Miles et al. 2002a), although Woods et al. (1999) overtraining are sympathetic nervous system and measured a modest enhancement of T-lymphocyte HPA axis imbalances and an increased susceptib- proliferation following 6 months of aerobic training ility to illnesses (Fitzgerald 1991; Urhausen et al. in an elderly population. With respect to B-cell pro- 1995). This is not surprising given the interrelated- duction of immunoglobulins, levels in serum and ness of these systems. Smith, L.L. (2000) hypothes- saliva are comparable in trained versus untrained ized that tissue trauma, particularly muscle damage, individuals across the vast majority of comparisons, may result in increased inflammatory cytokines, fol- both cross-sectional and longitudinal (Shephard lowed by activation of the HPA axis, which results 1997; Potteiger et al. 2001). One exception to this is down-regulation of immune cell function. Further, that decreased salivary IgA has been measured in she suggests that this represents an advanced stage athletes undergoing particularly intensive portions of the general adaptation model to stress proposed of their training, i.e. when training volume and by Selye, and may be the point at which recovery 360 chapter 25

and survival are more important than adaptation factors, additional stressors and exposure to virus to stress. This model is consistent with the data pre- and/or the reactivation of dormant viruses. As noted sented in this chapter; however, more recent research previously, the ‘the general adaptation syndrome’ to test this hypothesis did not measure increased hypothesis of Selye (1936) proposes that all stressors levels of inflammatory cytokines in athletes during elicit some common responses, including activation a period of induced overtraining (Halson et al. 2003). of the HPA axis. Thus, exercise may best be viewed The complexities of the neuroendocrine and immune as one component of the total stress to which a per- systems make it very difficult to verify or reject sev- son must respond. eral components of this hypothesis. Cytokine control of muscle responses to Susceptibility to illness stress and damage The relationship between levels of exercise training Muscle cells release cytokines in response to exer- and physical activity to incidence of illness is con- cise, whether or not cellular damage occurs. Current sistent with the overall pattern of changes identified research suggests that many of these cytokines have in immune parameters. That is, moderate training regulatory functions and may be necessary signals may reduce susceptibility to illness somewhat, or for various cellular functions and as systemic reduce the duration and severity of symptoms signals to the brain, particularly to the HPA axis. (Nieman et al. 1990). Recent epidemiological studies That is, inflammation is just one of the functions in also confirmed a similar relationship between phys- which these molecules, particularly IL-6 and LIF, ical activity and URTI in a diverse, healthy, adult participate, and the production of these molecules population (Matthews et al. 2002) and school chil- within muscle is not dependent on tissue injury. For dren in Poland (Jedrychowski et al. 2001). Consist- example, LIF has a hypertrophic effect on skeletal ent with this trend toward lower URTI incidence muscle cells and induces satellite cell prolifera- with moderate levels of physical activity or exercise tion (Spangenburg & Booth 2002; Gregorevic et al. training is the measurement of increased salivary 2002). IL-6 may act as a systemic signal reflecting IgA in response to moderate exercise training low glycogen levels in exercising muscle cells (Klentrou et al. 2002). (MacDonald et al. 2003). However, when muscle When training intensifies and reaches high damage does occur, IL-6 and LIF participate in the volumes and or intensities, perhaps coupled with inflammatory response. As indicated previously, limited recovery, then there is an increase in suscept- TNF-α, IL-1β, IL-6 and LIF are capable of stimulat- ibility to URTI (Shephard 1997; Mackinnon 2000). ing the HPA axis in a negative feedback loop to Recent research with elite Australian swimmers has induce production of the anti-inflammatory hor- advanced IgA levels and stress-induced viral react- mone, cortisol (Mastorakos et al. 1993; Chesnokova ivation of dormant viruses, such as Epstein–Barr & Melmed 2002; Tsigos & Chrousos 2002). Unfortu- virus, as likely mechanisms contributing to the nately, the response of LIF to exercise scarcely has incidence of URTI during high level training been investigated and little is known at this point. (Gleeson et al. 2002). In a study of training volume The type of muscle damage most typically associ- and tennis players, salivary IgA levels were inversely ated with exercise is characterized by delayed onset related to training volume, but IgA levels were not muscle soreness (DOMS); for example, the soreness predictive of URTI incidence. Researchers continue felt in muscles the day following the performance to search for the link between training levels and of exercise to which one is not accustomed. Indir- immunosuppression. Illness occurrence is likely to ect indicators of this damage are DOMS, loss of be a multidimensional function of the temporal strength, swelling and elevations in serum creatine accumulation of acute, exercise-induced immuno- kinase activity (Miles & Clarkson 1994). The exercise- suppression from multiple training sessions, the induced inflammatory response involving tissue level of recovery from bout to bout, nutritional damage likely occurs in the following sequence: immune response to exercise and doms 361

(i) tissue injury/stress; (ii) resident macrophage Warhol et al. 1985). Following marathon-type races, activation; (iii) release of inflammatory ‘alarm’ increases in plasma or serum TNF-α, IL-1β, IL-6, cytokines TNF-α and IL-1β; (iv) stimulation of local IL-10 and IL-1ra have been measured (Northoff release of chemoattractants, for example IL-8; (v) et al. 1994; Drenth et al. 1995; Nehlsen-Cannarella local production of IL-6; (vi) stimulation of acute et al. 1997; Ostrowski et al. 1998a, 1998b; Henson et al. phase proteins by the liver; (vii) stimulation of the 2000; Nieman et al. 2001). High force eccentric exer- HPA axis; and (viii) leukocytosis and leukocyte cise also induces skeletal muscle damage (Fridén migration to site of injury (Smith, L.L. & Miles 2000). & Lieber 2001) and smaller increases in IL-1β, IL-6 IL-1β and TNF-α initiate the inflammatory cascade, and IL-1ra (Bruunsgaard et al. 1997a; Smith, L.L. et al. which includes production of IL-6, which in turn 2000; Chen & Hsieh 2001; Toft et al. 2002). Increases stimulates production of a number of counter- in TNF-α typically are not measured (Bruunsgaard inflammatory cytokines and other molecules, et al. 1997a; Toft et al. 2002), but this may reflect the including IL-10 and IL-1ra (Turnbull et al. 1994; difficulty of measuring a systemic response to a Smith, L.L. & Miles 2000). Local production of IL-1β local inflammatory response that is relatively small has been demonstrated in muscle biopsy samples in magnitude. The prolonged period of inflamma- following eccentric exercise associated with muscle tion within muscles is not accompanied by eleva- damage (Malm et al. 2000); however, the increases tions in cortisol (Pizza et al. 1995; Lenn et al. 2002). in blood are short-lived and typically small relative Thus, under reasonable circumstances, the inflam- to other cytokines such as IL-6 or IL-10 (Shephard matory response to muscle damage does not appear 1997; Suzuki et al. 2002). Both IL-1β and IL-6 act as substantial enough to stimulate the HPA axis for an growth factors to promote regeneration at the site of extended period of time. However, there is a neuro- tissue damage (Northoff et al. 1995). endocrine and immune basis for the cytokine Within the realm of inflammation, the process response to muscle damage to stimulate the HPA of regeneration is dependent on the removal of axis if IL-6, and possibly TNF-α, IL-1β or LIF, pro- damaged cellular debris by phagocytes. The cellu- duction was substantial. lar response to exercise-induced muscle damage The hypothalamus uses the neuroendocrine– includes infiltration by and activation of local immune network to integrate exercise-induced stress macrophages (Stupka et al. 2001; LaPointe et al. signals and central signals to induce appropriate 2002). Infiltration of injured tissue by neutrophils responses to those stresses (Steinacker et al. 2004). has been measured in some (Brickson et al. 2001; Recent muscle biopsy data suggest a link between MacIntyre et al. 2001), but not all investigations increases in circulating cortisol and activation of (LaPointe et al. 2002). Muscle biopsy data suggests proteolysis within the injured muscle (Willoughby that this process occurs over a period of weeks et al. 2003). Exercise-induced injury is mediated by (Lieber 1992). This certainly exceeds the duration of the ubiquitin-proteolytic pathway, and cortisol up- any detectable markers of inflammation in the sys- regulates gene expression of several components of temic circulation. Thus, much of the response to this pathway, for example ubiquitin itself, ubiquitin exercise-induced muscle injury occurs locally rather conjugating enzyme and glucocorticoid receptors. than systemically. Consistent with the protective effects known to occur If exercise is prolonged and strenuous or mechan- after a single exercise bout is the down-regulation ically stressful, for example containing a high-force of this system following a second bout of the same eccentric component, then some degree of tissue injurious exercise. Thus, the two-way network of damage may be expected and the components of the communication involves cytokine signals generated described response are measurable. Marathons and locally to signal the HPA axis and HPA axis signals comparable events induce skeletal muscle damage generated centrally to signal the muscle to repair evidenced by disruption of myofibrillar organiza- and regenerate in response to tissue injury. tion and an efflux of the intramuscular enzyme cre- The role of inflammatory mediators in the res- atine kinase from muscle to blood (Rogers et al. 1985; ponse to exercise-induced muscle damage has been 362 chapter 25

called into question in light of research in which immune modulation per se, for example LIF may be anti-inflammatory treatment modalities failed to acting as a growth factor and have nothing to do have a substantial impact on DOMS or the recovery with tissue injury. The two-way communication of function (Cheung et al. 2003). However, the network between the neuroendocrine and immune research cited above is a small sample of the evid- systems is such that cytokines produced by muscles ence of various components of the inflammatory are capable of activating the HPA axis. Reciprocal response measured in response to exercise-induced regulation of the neuroendocrine and immune sys- muscle damage. Thus, it may be prudent simply to tems is infinitely complex. Regardless, research stud- suggest that the influence of prostaglandin E2, the ies are consistent in measuring enhanced immune typical target of anti-inflammatory medications, is function in response to moderately stressful acute not a major player in the inflammatory response exercise or exercise training. The level of exercise (Semark et al. 1999). As a result, anti-inflammatory stress that the body can tolerate without immuno- treatments do not appear to interfere with or suppression and increased susceptibility to illness is enhance the healing process and have only a minor likely to vary according to the level of other stresses impact on soreness (Cheung et al. 2003). that the body must endure. A number of studies have measured suppressed function in some com- Summary ponents of immunity and increased URTI occur- rences in athletes when the physical demands of The immune and neuroendocrine responses to the training and competition are extreme. In light of stress of exercise certainly are linked. However, the interactions between the neuroendocrine and modulation of exercise-induced immune responses immune systems, individuals wishing to maximize also is influenced by the inflammatory response their training volumes and intensities may be well to tissue injury. Cytokines mediating the local and advised to reduce the stress response to exercise via systemic responses to exercise are often multifunc- adequate recovery and various nutritional strateg- tional, for example IL-6. Many of the functions of ies and to minimize other forms of stress to which these cytokines are unrelated to inflammation or the body must respond.

References

Ashwell, J.D., Lu, F.W.M. & Vacchio, M.S. Blank, S.E., Jones, T.B., Lee, E.G. et al. & Best, T.M. (2001) Oxidant production (2000) Glucocorticoids in T-cell (1997) Modulation of NK cell cytolytic and immune response after stretch development and function. Annual activity by macrophages in chronically injury in skeletal muscle. Medicine and Review of Immunology 18, 309–345. exercise-stressed mice. Journal of Applied Science in Sports and Exercise 33, Atalay, M., Marnila, P., Lilius, E.-M., Physiology 83, 845–850. 2010–2015. Hanninen, O. & Sen, C.K. (1996) Borger, P., Hoekstra, Y., Esselink, M.T. Bruunsgaard, H., Galbo, H., Halkjaer- Glutathione-dependent modulation of et al. (1998) Beta-adrenoceptor-mediated Kristensen, J. et al. (1997a). Exercise- exhausting exercise-induced changes inhibition of IFN-γ, IL-3, and GM-CSF induced increase in interleukin-6 is in neutrophil function in rats. European mRNA accumulation in activated related to muscle damage. Journal of Journal of Applied Physiology 74, 342–347. human T lymphocytes is solely Physiology 499, 833–841. β Beishuizen, A. & Thijs, L.G. (2003) mediated by the 2-adrenoceptor Bruunsgaard, H., Hartkopp, A., Mohr, T. Endotoxin and the hypothalamo– subtype. American Journal of Respiratory, et al. (1997b) In vivo cell-mediated pituitary–adrenal (HPA) axis. Journal of Cellular, and Molecular Biology 19, immunity and vaccination response Endotoxin Research 9, 3–24. 400–407. following prolonged, intense exercise. Benschop, R.J., Nieuwenhuis, E.E.S., Boxer, L.A., Allen, J.M. & Baehner, R.L. Medicine and Science in Sports and Exercise Tromp, E.A.M. et al. (1994) Effects of (1980) Diminished polymorphonuclear 29, 1176–1181. β-adrenergic blockade on immunologic leukocyte adherence. Journal of Clinical Bruunsgaard, H., Jensen, M.S., Schjerling, and cardiovascular changes induced by Investigation 66, 268–274. P. et al. (1999) Exercise induces mental stress. Circulation 89, 762–769. Brenner, I., Shek, P.N., Zamecnik, J. & recruitment of lymphocytes with an Berk, L.S., Nieman, D.C., Youngberg, W.S. Shephard, R.J. (1998) Stress hormones activated phenotype and short telomeres et al. (1990) The effect of long endurance and the immunological response to heat in young and elderly humans. Life running on natural killer cells in and exercise. International Journal of Sciences 65, 2623–2633. marathoners. Medicine and Science in Sports Medicine 19, 130–143. Buckingham, J.C., Loxley, H.D., Christian, Sports and Exercise 22, 207–212. Brickson, S. Hollander, J. Corr, D.T., Ji, L.L. H.C. & Philip, J.G. (1996) Activation of immune response to exercise and doms 363

the HPA axis by immune insults: roles strength and total work. European Journal Gleeson, M., Pyne, D.B., Austin, J.P. et al. and interactions of the cytokines, of Applied Physiology 85, 367–373. (2002) Epstein–Barr virus reactivation eicosanoids, and glucocorticoids. Drenth, J.P.H., Van Uum, S.H.M., and upper-respiratory illness in elite Pharmacology Biochemistry and Behavior Van Deuren, M. et al. J.W.M. (1995) swimmers. Medicine and Science in Sports 54, 285–298. Endurance run increases circulating IL-6 and Exercise 34, 411–417. Cancedda, C., Filaci, G., Puppo, F. et al. and IL-1ra but downregulates ex vivo Green, K.J., Croaker, S.J. & Rowbottom, (2002) Immune homeostasis requires TNF-α and IL-1β production. Journal D.G. (2003) Carbohydrate several biologic factors including of Applied Physiology 79, 1497–1503. supplementation and exercise-induced glucocorticoid hormones. Annals of Duclos, M., Minkhar, M., Sarrieau, A. changes in T-lymphocyte function. the New York Academy of Sciences 966, et al. (1999) Reversibility of endurance Journal of Applied Physiology 95, 49–63. training-induced changes on 1216–1223. Carr, D.J.J. & Blalock, J.E. (1990) glucocorticoid sensitivity of monocytes Gregorevic, P., Williams, D.A. & Lynch, Neuropeptide hormones and receptors by an acute exercise. Clinical G.S. (2002) Effects of leukemia inhibitory common to the immune and Endocrinology 51, 749–756. factor on rat skeletal muscles are neuroendocrine systems: bidirectional Duclos, M., Corcuff, J.-B., Pehourcq, F. & modulated by clenbuterol. Muscle and pathway of intersystem communication. Tabarin, A. (2001) Decreased pituitary Nerve 25, 194–201. In: Psychoneuroimmunology, 2nd edn. sensitivity to glucocorticoids in Haddad, J.J., Saade, N.E. & Safieh- Academic Press, New York: 573–588. endurance-trained men. European Garabedian, B. (2002) Cytokines and Chen, T.C. & Hsieh, S.S. (2001) Effects Journal of Endocrinology 144, 363–368. neuro-immune-endocrine interactions: of a 7-day eccentric training period Elenkov, I.J., Wilder, R.L., Chrousos, G.P. a role for the hypothalamic–pituitary– on muscle damage and inflammation. & Vizi, E.S. (2000) The sympathetic adrenal revolving axis. Journal of Medicine and Science in Sports and Exercise nerve: an integrative interface between Neuroimmunology 133, 1–19. 33, 1732–1738. two supersystems: the brain and the Halson, S.L., Lancaster, G.I., Jeukendrup, Chesnokova, V. & Melmed, S. (2000) immune system. Pharmacological Reviews A.E. & Gleeson, M. (2003) Leukemia inhibitory factor mediates 52, 595–638. Immunological responses to the hypothalamic pituitary adrenal axis Eliakim, A., Wolach, B. Kodesh, E. et al. overreaching in cyclists. Medicine and response to inflammation. Endocrinology (1997) Cellular and humoral immune Science in Sports and Exercise 35, 854–861. 141, 4032–4040. response to exercise among gymnasts Hazenberg, M.D., Otto, S.A., van Benthem, Chesnokova, V. & Melmed, S. (2002) and untrained girls. International Journal B.H. et al. (2003) Persistent immune Minireview: Neuro-immuno-endocrine of Sports Medicine 18, 208–212. activation inf HIV-1 infection is modulation of the hypothalamic– Febbraio, M.A., Ott, P., Nielsen, H.B. associated with progression of AIDS. pituitary–adrenal (HPA) axis by gp130 et al. (2003a) Hepatosplanchnic clearance AIDS (London, England) 17, 1881–1888. signaling molecules. Endocrinology 143, of interleukin-6 in humans during Henson, D.A., Nieman, D.C., 1571–1574. exercise. American Journal of Physiology, Parker, J.C.D. et al. (1998) Carbohydrate Chesnokova, V., Kariagina, A. & Melmed, Endocrinology and Metabolism 285, supplementation and the lymphocyte S. (2002) Opposing effects of pituitary E397–E402. proliferative response to long endurance leukemia inhibitory factor and SOCS-3 Febbraio, M.A., Steensburg, A., Starkie, running. International Journal of Sports on the ACTH axis response to R.L., McConell, G.K. & Kingwell, B.A. Medicine 19, 574–580. inflammation. American Journal of (2003b). Skeletal muscle interleukin-6 Henson, D.A., Nieman, D.C., Nehlsen- Physiology, Endocrinology and Metabolism and tumor necrosis factor-α release in Cannarella, S.L. et al. (2000) Influence 282, E111–E118. healthy subjects and patients with type 2 of carbohydrate on cytokine and Cheung, K., Hume, P.A. & Maxwell, L. diabetes at rest and during exercise. phagocytic response to 2 h of rowing. (2003) Delayed onset muscle soreness: Metabolism 52, 939–944. Medicine and Science in Sports and Exercise Treatment strategies and performance Fitzgerald, L. (1991) Overtraining 32, 1384–1389. factors. Sports Medicine 33, 145–164. increases the susceptibility to infection. Hetherington, S.V. & Quie, P.G. (1985) DeRijk, R.H., Petrides, J., Deuster, P., International Journal of Sports Medicine 12, Human polymorphonuclear leukocytes Gold, P.W. & Sternberg, E.M. (1996) S5–S8. of the bone marrow, circulation, and Changes in corticosteroid sensitivity Flynn, M.G., Fahlman, M., Braun, W.A. marginated pool: function and granule of peripheral blood lymphocytes after et al. (1999) Effects of resistance training protein content. American Journal of strenuous exercise in humans. Journal of on selected indexes of immune function Hematology 20, 235–246. Clinical Endocrinology and Metabolism 81, in elderly women. Journal of Applied Hoffman-Goetz, L. & Quadrilatero, J. 228–235. Physiology 86, 1905–1913. (2003) Treadmill exercise in mice Deuster, P.A., Zelazowska, E.B., Singh, A., Fridén, J. & Lieber, R.L. (2001) Eccentric increases intestinal lymphocyte loss via & Sternberg, E.M. (1999) Expression of exercise-induced injuries to contractile apoptosis. Acta Physiologica Scandanavica lymphocyte subsets after exercise and and cytoskeletal muscle fibre 179, 289–297. dexamethasone in high and low stress components. Acta Physiologica Hsu, T.-G., Hsu, K.-M., Kong, C.-W. responders. Medicine and Science in Scandanavica 171, 321–326. et al. (2002) Leukocyte mitochondria Sports and Exercise 31, 1799–1806. Gleeson, M., McDonald, W.A., Pyne, D.B. alterations after aerobic exercise in Dohi, K., Mastro, A.M., Miles, M.P. et al. et al. (1999) Salivary IgA levels and trained human subjects. Medicine and (2002) Lymphocyte proliferation in infection risk in elite swimmers. Science in Sports and Exercise 34, 438–442. response to acute heavy resistance Medicine and Science in Sports and Exercise Ibfelt, T., Petersen, E.W., Bruunsgaard, H., exercise in women: influence of muscle 31, 67–73. Sandmand, M. & Pedersen, B.K. (2002) 364 chapter 25

Exercise-induced change in type 1 in vivo. Pharmacological Reviews 53, Mackinnon, L.T. & Jenkins, D.G. (1993) cytokine-producing CD8+ T cells in 487–525. Decreased salivary immunoglobulins related to a decrease in memory T LaPointe, B.M., Frenette, J. & Côté, C. after intense interval exercise before and cells. Journal of Applied Physiology 93, (2002) Lengthening contraction-induced after training. Medicine and Science in 645–648. inflammation is linked to secondary Sports and Exercise 25, 678–683. Jedrychowski, W., Maugeri, U., Flak, E., damage but devoid of neutrophil Mackinnon, L.T., Chick, T.W., van As, A. & Mroz, E. & Bianchi, I. (2001) Cohort invasion. Journal of Applied Physiology Tomasi, T.B. (1989) Decreased secretory study on low physical activity level and 92, 1995–2004. immunoglobulins following intense recurrent acute respiratory infections in Lenn, J., Uhl, T., Mattacola, C. et al. (2002) endurance exercise. Sports Training, schoolchildren. Central European Journal The effects of fish oil and isoflavones on Medicine and Rehabilitation 1, 209–218. of Public Health 9, 126–129. delayed onset muscle soreness. Medicine McMurray, R.G. & Hackney, A.C. (2000) Jonsdottir, I.H. & Hoffmann, P. (2000) The and Science in Sports and Exercise 34, Endocrine responses to exercise and significance of intensity and duration 1605–1613. training. In: Exercise and Sport Science of exercise on natural immunity in rats. Leonard, B.E. & Song, C. (1996) Stress and (Garrett, W.E. & Kirkendall, D.T., eds.). Medicine and Science in Sports and Exercise the immune system in the etiology of Lippincott, Williams & Wilkins, 32, 1908–1912. anxiety and depression. Pharmacology Philadelphia, PA: 135–161. Jonsdottir, I.H. Schjerling, P., Ostrowski, K. Biochemistry and Behavior 54, 299–303. Maisel, A.S., Harris, C., Rearden, D.A. & et al. (2000) Muscle contractions induce Levine, S.J., Benfield, T. & Shelhamer, J.H. Michel, M.C. (1990) Beta-adrenergic interleukin-6 mRNA production in rat (1996) Corticosteroids induce receptors in lymphocyte subsets skeletal muscles. Journal of Physiology intracellular interleukin-1 receptor after exercise. Alterations in normal 528, 157–163. antagonist type I expression by a human individuals and patients with congestive Kappel, M., Tvede, N., Galbo, H. et al. airway epithelial cell line. American heart failure. Circulation 82, 2003–2010. (1991) Evidence the the effect of physical Journal of Respiratory Molecular and Malarkey, W.B., Wang, J., Cheney, C., exercise on NK cell activity is mediated Cellular Biology 15, 245–251. Glaser, R. & Nagaraja, H. (2002). Human by epinephrine. Journal of Applied Lieber, R.L. (1992) Skeletal Muscle Structure lymphocyte growth hormone stimulates Physiology 70, 2530–2534. and Function. Williams & Wilkins, interferon γ production and is inhibited Kappel, M., Poulsen, T.D., Galbo, H. & Baltimore, MD. by cortisol and epinephrine. Journal of Pedersen, B.K. (1998) Influence of minor Liles, W.C. & Van Voorhis, W.C. (1995) Neuroimmunology 123, 180–187. increases in plasma catecholamines on Review: nomenclature and biologic Malm, C., Nyberg, P., Engström, M. et al. natural killer cell activity. Hormone significance of cytokines involved (2000) Immunological changes in human Research 49, 22–26. in inflammation and host immune skeletal muscle and blood after eccentric Kjær, M. & Dela, F. (1996) Endocrine response. Journal of Infectious Diseases exercise and multiple biopsies. Journal of responses to exercise. In: Exercise and 172, 1573–1580. Physiology 529, 243–262. Immune Function (Hoffman-Goetz, L., MacDonald, C., Wojtaszewski, J.F.P., Masˇek, K., Slánsky`, J. Petrovicky`, P. & ed.). CRC Press, Boca Raton, FL: 1–20. Pedersen, B.D., Kiens, B. & Richter, E.A. Hadden, J.W. (2003) Neuroendocrine Kjær, M., Bangsbo, J., Lortie, G. & Galbo, (2003) Interleukin-6 release from human immune interactions in health H. (1988) Hormonal responses to skeletal muscle during exercise: relation and disease. International exercise in humans: influence of to AMPK activity. Journal of Applied Immunopharmacology 3, 1235–1246. hypoxia and physical training. American Physiology 95, 2273–2277. Mastorakos, G., Chrousos, G.P. & Weber, Journal of Physiology 254, R197–R203. McFarlin, B.K., Mitchell, J.B., McFarlin, J.S. (1993) Recombinant interleukin-6 Klentrou, P., Cieslak, T., MacNeil, M., M.A. & Steinhoff, G.M. (2003) Repeated activated the hypothalamic–pituitary– Vintinner, A. & Plyley, M. (2002) Effect endurance exercise affects leukocyte adrenal axis in humans. Journal of of moderate exercise on salivary number but not NK cell activity. Clinical Endocrinology and Metabolism 77, immunoglobulin A and infection risk Medicine and Science in Sports and 1690–1694. in humans. European Journal of Applied Exercise 35, 1130–1138. Matthews, C.E., Ockene, I.S., Freedson, Physiology 87, 153–158. MacIntyre, D.L., Sorichter, S., Mair, J., P.S. et al. (2002) Moderate to vigorous Klokker, M., Kjær, M., Secher, N.H. et al. Berg, A. & McKenzie, D.C. (2001) physical activity and risk of upper- (1995) Natural killer cell response to Markers of inflammation and respiratory tract infection. Medicine exercise in humans: effect of hypoxia myofibrillar proteins following and Science in Sports and Exercise 34, and epidural anesthesia. Journal of eccentric exercise in humans. European 1242–1248. Applied Physiology 78, 709–716. Journal of Applied Physiology 84, Miles, M.P. & Clarkson, P.M. (1994) Klokker, M., Secher, N.H., Madsen, P., 180–186. Exercise-induced muscle pain, soreness, Pedersen, M. & Perdersen, B.K. (1997) Mackinnon, L.T. (2000) Chronic exercise and cramps. Journal of Sports Medicine β β Adrenergic 1 and 1+2-receptor training effects on immune function. and Physical Fitness 34, 203–216. blockade suppress the natural killer Medicine and Science in Sports and Miles, M.P., Leach, S.K., Kraemer, W.J. cell response to head-up tilt in humans. Exercise 32, S369–S376. et al. (1998) Leukocyte adhesion Journal of Applied Physiology 83, Mackinnon, L.T. & Hooper, S.L. (1994) molecule expression during intense 1492–1498. Mucosal (secretory) immune system resistance exercise. Journal of Applied Kohm, A.P. & Sanders, V.M. (2001) responses to exercise of varying Physiology 84, 1604–1609. β Norepinephrine and 2-adrenergic intensity and during overtraining. Miles, M.P., Kraemer, W.J., Grove, D.S. receptor stimulation regulate CD4+ T International Journal of Sports Medicine 15, et al. (2002a) Effects of resistance training and B lymphocyte function in vitro and S179–S183. on resting immune parameters in immune response to exercise and doms 365

women. European Journal of Applied Lymphocyte, NK and LAK cell Northoff, H., Weinstock, C. & Berg, A. Physiology 87, 506–508. responses to maximal exercise. (1994) The cytokine response to Miles, M.P., Mackinnon, L.T., Grove, D.S. International Journal of Sports Medicine strenuous exercise. International Journal et al. (2002b) The relationship of natural 17, 60–65. of Sports Medicine 15, S167–S171. killer cell counts, perforin mRNA, and Nieman, D.C. (1997) Immune response Northoff, H., Enkel, S. & Weinstock, C. CD2 expression to post-exercise natural to heavy exertion. Journal of Applied (1995) Exercise, injury and immune killer cell activity in humans. Acta Physiology 82, 1385–1394. function. Exercise Immunology Review Physiologica Scandanavica 174, 317–325. Nieman, D.C. (1998) Exercise and 1, 1–25. Mitchell, J.B., Pizza, F.X., Paquet, A. et al. resistance to infection. Canadian Journal Ortega, E., Rodrigues, M.J., Barriga, C. (1998) Influence of carbohydrate status of Physiology and Pharmacology, 76, & Forner, M.A. (1996) Corticosterone, on immune responses before and after 573–580. prolactin and thyroid hormone as endurance exercise. Journal of Applied Nieman, D.C. (2000) Is infection risk hormonal mediators of the stimulated Physiology 84, 1917–1925. linked to exercise workload? Medicine phagocytic capacity of peritoneal Mitchell, J.B., Dugas, J.P., McFarlin, B.K. and Science in Sports and Exercise 32, macrophages after high-intensity & Nelson, M.J. (2002) Effect of exercise, S406–S411. exercise. International Journal of Sports heat stress, and hydration on immune Nieman, D.C., Nehlsen-Cannarella, S.L., Medicine 17, 149–155. cell number and function. Medicine Markoff, P.A. et al. (1990) The effects of Ortega, E., Forner, M.A. & Barriga, C. and Science in Sports and Exercise 34, moderate exercise training on natural (1997) Exercise-induced stimulation 1941–1950. killer cells and acute upper respiratory of murine macrophage chemotaxis: Mooren, F.C., Blöming, D., Lechtermann, tract infections. International Journal of role of corticosterone and prolactin as A., Lerch, M.M. & Völker, K. (2002) Sports Medicine 11, 467–473. mediators. Journal of Physiology 498, Lymphocyte apoptosis after exhaustive Nieman, D.C., Henson, D.A., Gusewitch, 729–734. and moderate exercise. Journal of Applied G. et al. (1993a) Physical activity and Ostrowski, K., Hermann, C., Bangash, A. Physiology 93, 147–153. immune function in elderly women. et al. (1998a) A trauma-like elevation Mosmann, T.R. & Sad, S. (1996) The Medicine and Science in Sports and of plasma cytokines in humans in expanding universe of T-cell subsets: Exercise 25, 823–831. response to treadmill running. Journal Th1, Th2 and more. Immunology Today Nieman, D.C., Miller, A.R., Hension, D.A. of Physiology 513, 889–894. 17, 138–146. et al. (1993b) Effects of high- vs. Ostrowski, K., Rohde, T., Zacho, M., Asp, Moyna, N.M., Acker, G.R., Weber, K.M. moderate-intensity exercise on natural S. & Pedersen, B.K. (1998b) Evidence et al. (1996). The effects of incremental killer cell activity. Medicine and Science in that interleukin-6 is produced in human submaximal exercise on circulating Sports and Exercise 25, 1126–1134. skeletal muscle during prolonged leukocytes in physically active and Nieman, D.C., Ahle, J.C., Henson, D.A. running. Journal of Physiology 508, sedentary males and females. European et al. (1995a) Indomethacin does not alter 949–953. Journal of Applied Physiology 74, 211–218. the natural killer cell response to 2.5 h of Ostrowski, K., Schjerling, P. & Pedersen, Mucci, P., Durand, F., Level, B., Bousquet, running. Journal of Applied Physiology 79, B.K. (2000) Physical activity and plasma J. & Préfaut, C. (2000). Interleukins 1-β, 748–755. interleukin-6 in humans: effect of -8, and histamine increases in highly Nieman, D.C., Henson, D.A., Sampson, intensity of exercise. European Journal trained, exercising athletes. Medicine C.S. et al. (1995b) The acute immune of Applied Physiology 83, 512–515. and Science in Sports and Exercise 32, response to exhaustive resistance Papanicolaou, D.A., Petrides, J.S., Tsigos, 1094–1100. exercise. International Journal of Sports C. et al. (1996) Exercise stimulates Müns, G. (1993) Effect of long-distance Medicine 16, 322–328. interleukin-6 secretion: inhibition by running on polymorphonuclear Nieman, D.C., Nehlsen-Cannarella, S.L., glucocorticoids and correlation with neutrophil phagocytic function of the Fagoaga, O.R. et al. (1998) Effects catecholamines. American Journal of upper airways. International Journal of of mode and carbohydrate on the Physiology, Endocrinology and Metabolism Sports Medicine 15, 96–99. granulocyte and monocytes response to 271, E601–E605. Nehlsen-Cannarella, S.L., Nieman, D.C., intensive, prolonged exercise. Journal of Peake, J.M. (2002) Exercise-induced Jessen, J. et al. (1991) The effects of acute Applied Physiology 84, 1252–1259. alterations in neutrophil degranulation moderate exercise on lymphocyte Nieman, D.C., Henson, D.A., Smith, L.L. and respiratory burst activity: possible function and serum immunoglobulin et al. (2001) Cytokine changes after a mechanism of action. Exercise levels. International Journal of Sports marathon race. Journal of Applied Immunology Review 8, 49–100. Medicine 12, 391–398. Physiology 91, 109–114. Pedersen, B.K. & Hoffman-Goetz, L. (2000) Nehlsen-Cannarella, S.L., Fagoaga, O.R., Nieman, D.C., Henson, D.A., Fagoaga, Exercise and the immune system: Nieman, D.C. et al. (1997) Carbohydrate O.R. et al. (2002) Change in salivary IgA regulation, integration, and adaptation. and the cytokine response to 2.5 h of following a competitive marathon race. Physiological Reviews 80, 1055–1081. running. Journal of Applied Physiology 82, International Journal of Sports Medicine 23, Pedersen, B.K. & Ullum, H. (1994) NK cell 1662–1667. 69–75. response to physical activity: possible Nielsen, H.B. & Pedersen, B.K. (1997) Niess, A.M., Fehrenbach, E., Strobel, G. mechanisms of action. Medicine and Lymphcyte proliferation in response et al. (2003) Evaluation of stress Science in Sports and Exercise 26, to exercise. European Journal of Applied responses to interval training at low 140–146. Physiology 75, 375–379. and moderate altitudes. Medicine and Pedersen, B.K., Tvede, N., Christensen, Nielsen, H.B., Secher, N.H., Kappel, M., Science in Sports and Exercise 35, L.D. et al. (1989) Natural killer cell Hanel, B. & Pedersen, B.K. (1996) 263–269. activity in peripheral blood of highly 366 chapter 25

trained and untrained persons. and reverses postexercise suppression to excessive stress? Medicine and Science International Journal of Sports Medicine of natural killer cell activity. American in Sports and Exercise 32, 317–331. 10, 129–131. Journal of Physiology 276, R1496–R1505. Smith, L.L. & Miles, M.P. (2000) Pedersen, B.K., Steensberg, A., Fischer, C. Riccardi, C., Bruscoli, S. & Micliorati, G. Exercise-induced muscle injury and et al. (2001) Exercise and cytokines with (2002) Molecular mechanisms of inflammation. In: Exercise and Sport particular focus on muscle-derived IL-6. immunomodulatory activity of Science (Garrett, W.E. & Kirkendall, D.T., Exercise Immunology Reviews 7, 18–31. glucocorticoids. Pharmacological Research eds.). Lippincott, Williams & Wilkins, Peters, E.M. & Bateman, E.D. (1983) 45, 361–368. Philadelphia, PA: 401–411. Ultramarathon running and upper Rivest, S. (2001) How circulating cytokines Smith, L.L., Anwar, A., Fragen, M. et al. respiratory tract infections. An trigger the neural circuits that control (2000) Cytokines and cell adhesion epidemiological survey. South African the hypothalamic–pituitary–adrenal molecules associated with high-intensity Medical Journal 64, 582–584. axis. Psychoneuroendocrinology 26, eccentric exercise. European Journal of Petrides, J.S., Gold, P.W., Mueller, G.P. 761–788. Applied Physiology 82, 61–67. et al. (1997) Marked differences in Rogers, M.A., Stull, G.A. & Apple, F.S. Smith, J.A., Telford, R.D., Baker, M.S., functioning of the hypothalamic– (1985) Creatine kinase isoenzyme Hapel, A.J. & Weidemann, M.J. (1992) pituitary–adrenal axis between groups activities in men and women following Cytokine immunoreactivity in plasma of men. Journal of Applied Physiology 82, a marathon race. Medicine and Science in does not change after moderate 1979–1988. Sports and Exercise 17, 679–682. endurance exercise. Journal of Applied Pitzalis, C., Pipitone, N. & Perretti, M. Ronsen, O., Kjeldsen-Kragh, J., Haug, E., Physiology 73, 1396–1401. (2002). Regulation of leukocyte- Bahr, R. & Pedersen, B.K. (2002a) Smits, H.H., Grunberg, K., DeRijk, R.H., endothelial interactions by Recovery time affects immunoendocrine Sterk, P.J. & Hiemstra, P.S. (1998) glucocorticoids. Annals of the New York responses to a second bout of endurance Cytokine release and its moduclation Acadamy of Science 966, 108–118. exercise. American Journal of Physiology, by dexamethasone in whole blood Pizza, F.X., Mitchell, J.B., Davis, B.H. Cellular Physiology 283, C1612–C1620. following exercise. Clinical and et al. (1995) Exercise-induced muscle Ronsen, O., Lea, T., Bahr, R. & Pedersen, Experimental Immunology 111, 463–468. damage: effect on circulating leukocyte B.K. (2002b) Enhanced plasma IL-6 and Spangenburg, E.E. & Booth, F.W. (2002) and lymphocyte subsets. Medicine and IL1ra responses to repeated vs. single Multiple signaling pathways mediate Science in Sports and Exercise 27, 363–370. bouts of prolonged cycling in elite LIF-induced skeletal muscle satellite Potteiger, J.A. Chan, M.A., Haff, G.G. et al. athletes. Journal of Applied Physiology 92, cell proliferation. American Journal of (2001) Training status influence T-cell 2547–2553. Physiology 283, C204–C211. responses in women following acute Sanders, V.M., Baker, R.A., Ramer-Quinn, Sprenger, H., Jacobs, C., Nain, M. et al. resistance exercise. Journal of Strength D.S. et al. (1997) Differential expression (1992) Enhanced release of cytokines, β and Conditioning Research 15, 185–191. of the 2-adrenergic receptor by Th1 and interleukin-2 receptors, and neopterin Pyne, D.B. (1994) Regulation of neutrophil Th2 clones: implications for cytokine after long-distance running. Clinical function during exercise. Sports Medicine production and B cell help. Journal of Immunology and Immunopathology 63, 17, 245–258. Immunology 158, 4200–4210. 188–195. Pyne, D.B., McDonald, W.A. Gleeson, M. Schedlowski, M., Falk, A., Rohne, A. Steensberg, A., Toft, A.D., Bruunsgaard, H. et al. (2000) Mucosal immunity, et al. (1993) Catecholamines induce et al. (2001) Strenuous exercise decreases respiratory illness, and competitive alterations of distribution and activity the percentage of type 1 T cells in the performance in elite swimmers. Medicine of human natural killer (NK) cells. circulation. Journal of Applied Physiology and Science in Sports and Exercise 33, Journal of Clinical Immunology 13, 91, 1708–1712. 348–353. 344–351. Steensberg, A., Keller, C., Starkie, R.L. et al. Rabin, B.S., Moyna, N.M., Kusnecov, A., Schedlowski, M., Hosch, W., Oberbeck, R. (2002) IL-6 and TNF-α expression in, and Zhou, D. & Shurin, M.R. (1996) et al. (1996) Catecholamines modulated release from, contracting human skeletal Neuroendocrine effects on immunity. human NK cell circulation and function muscle. American Journal of Physiology β In: Exercise and Immune Function via spleen-independent 2-adrenergic 283, E1272–E1278. (Hoffman-Goetz, L., ed.). CRC Press, mechanisms. Journal of Immunology 156, Steensberg, A., Fischer, C.P., Keller, C., Boca Raton, FL: 21–37. 93–99. Møller, K. & Pedersen, B.K. (2003) IL-6 Rall, L.C., Roubenoff, R., Cannon, J.G. et al. Selye, H. (1936) A syndrome produced by enhances plasma IL-1ra, IL-10, and (1996) Effects of progressive resistance diverse nocuous agents. Nature 138, cortisol in humans. American Journal of training on immue response in aging 32–36. Physiology, Endocrinology and Metabolism and chronic inflammation. Medicine and Semark, A., Noakes, T.D., St. Clair Gibson, 285, E433–E437. Science in Sports and Exercise 28, A. & Lambert, M.I. (1999) The effect Steinacker, J.M., Lormes, W., Reissnecker, 1356–1365. of a prophylactic dose of flubiprofen S. & Liu, Y. (2004) New aspects of the Ramirez, F. & Silva, A. (1997) on muscle soreness and sprinting hormone and cytokine response to Glucocorticoids enhance concanvalin performance in trained subjects. Journal training. European Journal of Applied A-induced mitogenic response through of Sport Science 17, 197–203. Physiology 91, 382–391. the inhibition of nitric oxide production. Shephard, R.J. (1997) Physical Activity, Strasner, A., Davis, J.M., Kohut, M.L. et al. Immunology 90, 66–73. Training, and the Immune Response. (1996) Effects of exercise intensity on Rhind, S.G., Gannon, G.A., Suzui, M., Cooper Publishing Group, Carmel, IN. natural killer cell activity in women. Shephard, R.J. & Shek, P.N. (1999) Smith, L.L. (2000) Cytokine hypothesis of International Journal of Sports Medicine 18,

Indomethacin inhibits circulating PGE2 overtraining: a physiological adaptation 56–62. immune response to exercise and doms 367

Stupka, N., Tarnopolsky, M.A., Yardley, eccentric exercise in young and elderly ubiquitin expression after repeated N.J. & Phillips, S.M. (2001) Cellular humans. American Journal of Physiology eccentric exercise. Medicine and Science adaptation to repeated eccentric 283, C289–C295. in Sports and Exercise 35, 2023–2031. exercise-induced muscle damage. Tsigos, C. & Chrousos, G.P. (2002) Woods, J.A. & Davis, J.M. (1994) Exercise, Journal of Applied Physiology 91, Hypothalamic–pituitary–adrenal axis, macrophages, and cancer defense. 1669–1678. neuroendocrine factors and stress. Medicine and Science in Sports and Exercise Suzuki, K., Sato, H., Kikuchi, T. et al. (1996) Journal of Psychosomatic Research 53, 26, 147–157. Capacity of circulating neutrophils to 865–871. Woods, J.A. Davis, J.M., Mayer, E.P., produce reactive oxygen species after Turnbull, A.V., Dow, R.C., Hopkins, S.J. Ghaffar, A. & Pate, R.R. (1994) Effects of exhaustive exercise. Journal of Applied et al. (1994) Mechanisms of activation exercise on macrophage activation for Physiology 81, 1213–1222. of the pituitary-adrenal axis by antitumor cytotoxicity. Journal of Applied Suzuki, K., Totsuka, M., Nakaji, S. et al. tissue injury in the rat. Physiology 76, 2177–2185. (1999). Endurance exercise causes Psychoneuroendocrinology 19, 165–178. Woods, J.A., Ceddia, M.A., Kozak, C. & interaction among stress hormones, Urhausen, A. Gabriel, H. & Kindermann, Wolters, B.W. (1997) Effects of exercise cytokines, neutrophil dynamics, and W. (1995) Blood hormones as markers of on the macrophage MHC II response to muscle damage. Journal of Applied training stress and overtraining. Sports inflammation. International Journal of Physiology 87, 1360–1367. Medicine 20, 251–276. Sports Medicine 18, 483–488. Suzuki, K., Nakaji, S., Yamada, M. et al. Walsh, N.P., Bishop, N.C., Blackwell, J., Woods, J.A., Davis, J.M., Smith, J.A. & (2002) Systemic inflammatory response Wierzbicki, S.G. & Montague, J.C. (2002) Nieman, D.C. (1999) Exercise and to exhaustive exercise. Cytokine Salivary IgA response to prolonged cellular innate immune function. kinetics. Exercise Immunology Review 8, exercise in a cold environment in trained Medicine and Science in Sports and Exercise 6–48. cyclists. Medicine and Science in Sports and 31, 57–66. Tharp, G.D. & Barnes, M.W. (1990) Exercise 34, 1632–1637. Yamada, M., Suzuki, K., Kudo, S. et al. Reduction of salivary immunoglobulin Warhol, M.J., Siegel, A.J., Evans, W.J. & (2002) Raised plasma G-CSF and IL-6 levels by swim training. European Journal Silverman, L.M. (1985) Skeletal muscle after exercise may play a role in of Applied Physiology 60, 61–64. injury and repair in marathon runners neutrophil mobilization into the van Tits, L.J.H., Michel, M.C., Grosse- after competition. American Journal of circulation. Journal of Applied Physiology Wilde, H. et al. (1990) Catecholamines Pathology 118, 331–339. 92, 1789–1794. β increase lymphocyte 2-adrenergic Weicker, H. & Werle, E. (1991) Interaction Zhou, J., Olsen, S., Moldovan, J. et al. (1997) β receptors via a 2-adrenergic, spleen- between hormones and the immune Glucocorticoid regulation of natural dependent process. American Journal of system. International Journal of Sports cytotoxicity: effects of cortisol on the Physiology 258, E191–E202. Medicine 12 (suppl. 1), S30–S37. phenotype and function of cloned Toft, A.D., Jensen, L.B., Bruunsgaard, H. Willoughby, D.S., Taylor, M. & Taylor, L. human nature killer cell line. Cellular et al. (2002) Cytokine response to (2003) Glucocorticoid receptor and Immunology 178, 108–116. Chapter 26

The Impact of Exercise on Immunity: the Role of Neuroendocrine–Immune Communications

ANDREA M. MASTRO AND ROBERT H. BONNEAU

Introduction to exercise– tions of exercise to physiology from a psychological neuroendocrine–immune relationships perspective are less well established. The exact mechanisms underlying these contributions have The many benefits of regular physical exercise are not been identified; but it is generally accepted that not only well-documented in the scientific literature the functioning of both the nervous and endocrine but have also received considerable attention in the systems are intimately associated with a variety of popular press. It is recognized that exercise has psychological states. The fact that exercise can affect wide-ranging effects on most organ systems of the components of both the nervous and endocrine body but, in general, emphasis has been placed on systems suggests that, in part, these effects may be its impact on cardiovascular and pulmonary func- mediated via a psychological pathway. In recent tion (reviewed in Booher & Smith 2003). However, years, substantial evidence has been collected to more recently, the cellular and molecular mechan- support a functional link between not only the isms underlying the effects of exercise on several nervous and endocrine systems but also among the other organs systems, especially the immune sys- nervous, endocrine and immune systems (Fig. 26.1) tem, have begun to be delineated (e.g. Pedersen (Conti et al. 2000; Ader et al. 2001). & Hoffman-Goetz 2000; Pedersen & Toft 2000; Given the many findings that both the nervous Suzuki et al. 2002; Woods et al. 2002; Lakier-Smith and endocrine systems are alone able to modu- 2003; Nieman 2003). The strong interest in the rela- late a variety of immune functions provides strong tionship between exercise and the immune function Nervous has lead to the recent publication of a number of system excellent review articles devoted to this subject (International Journal of Sports Medicine [volume 21, supplement 1, May 2000]; Immunology and Cell Bio- logy [volume 78, October 2002]). It has long been recognized that exercise can have a strong impact on a person’s overall mental condi- Exercise tion as well as their physical state (Glenister 1996; Fox 1999; Paluska & Schwenk 2000; Salmon 2001). Of particular recent interest is how one’s mental state can, in turn, influence a wide range of phy- siological parameters that together contribute to Endocrine Immune maintaining homeostasis and overall well-being. system system Although the physiological benefits of exercise have Fig. 26.1 Exercise and the nervous–endocrine–immune been recognized for quite some time, the contribu- network. 368 exercise and immunity 369

support for a psychological link between exercise clearly beyond the scope of this text and is discussed and immunity. As is outlined below, there is indeed in depth elsewhere (Robergs & Keteyian 2003). a neuroendocrine-mediated link between exercise and immune function. This link may play an import- the lines of communication between ant role in the development and progression of the nervous and endocrine systems diseases that are either immunologically resisted (e.g. infectious diseases, cancer) or that are caused It has been clear for quite some time that nervous by an undesirable activation of the immune system and endocrine systems function as a single inter- (e.g. allergy, autoimmunity). related system within the body. The functional rela- It is important to note that the information provided tionships that exist between these two systems and in this chapter is not intended to represent a compre- how they regulate a wide range of bodily processes hensive review of all literature dealing with exercise is the basis of the study of neuroendocrinology. This and immunity. For such information, the reader is relationship operates bi-directionally in that the directed to several recent reviews in this area endocrine system affects the nervous system and (Hoffman-Goetz 1996; Nieman & Pedersen 2000; the nervous system affects the endocrine system. Pedersen & Hoffman-Goetz 2000; Shephard & Shek Whereas hormones serve as the primary mediators 2000a; Hoffman-Goetz & Pedersen 2001). Rather, of the endocrine-to-nervous system communica- this chapter is specifically designed to enlighten the tion, the nervous system–endocrine communication reader on the neuroendocrine-mediated mechan- occurs within the site of direct interface, at the level isms underlying the relationship among exercise of the neuroendocrine cell itself. and specific aspects of immune function. It is also Hormones impact virtually every tissue in the intended to provide insight into the areas in which body as a result of their transport through the blood. an understanding of the relationship between exer- Functional interactions between hormones and cise and immune function can be further expanded their target tissues are highly dependent on specific as a result of recent advances in our knowledge of receptor-mediated binding of hormones to the cells immune function and state-of-the-art experimental which are able to respond. Depending on the hor- approaches that can quantify such function. Lastly, mone, these receptors are located on either the cell we will address the potential impact of exercise on membrane, within the cytoplasm, or in the nucleus. diseases whose prevention or cause is associated The mechanisms of hormone action are varied and with one or more aspects of immune function. include membrane transport, stimulation of gene transcription, and the activation of intracellular Neuroendocrine–immune second messengers such as cyclic adenosine mono- communications phosphate (cAMP). There are a variety of mechanisms that regulate the release of hormones into the neuroendocrine Neuroendocrine communications pathway. Endogenous circadian or diurnal rhythms Before launching into a discussion of the variety of release provide a tonic cyclic pattern of hormone of functional relationships that exist between the secretion that is independent of physiologic pertur- neuroendocrine and immune systems and their bation. Superimposed on these endogenous rhythms impact on one’s level of immunocompetence, it is are a number of highly complex positive and negat- first necessary to briefly review some basic prin- ive feedback loop mechanisms which help to main- ciples underlying the communications that exist tain endocrine-associated homeostasis. between the nervous and endocrine systems and how these communications are possibly influenced function and importance of by exercise. However, it should be noted that a com- neuroendocrine communications prehensive review of the both the nervous and endocrine systems, as well as exercise physiology, is Neuroendocrine connections are important in that 370 chapter 26

they allow our bodies to maintain a state of home- specific chemical and mechanical conditions that ostasis. For example, neuroendocrine relationships help to regulate, through the action of a variety of play a critical role in water conservation and main- hormones, a number of physiological functions tenance of body fluid osmolarity, blood volume and during exercise. Some of these functions include pressure, growth and development, metabolism, energy metabolism, fuel mobilization, fluid balance, electrolyte balance, ovulation and parturition and vascular hemodynamics and protein synthesis. behavior. Disorders in the generation, secretion, or These responses can vary among individuals and be response to hormones can lead to a wide range of influenced by both exercise intensity and gender. pathological conditions including diabetes insipidus, The exercise-induced, hormonally-based regula- osteoporosis and acromegaly, to name a few. tion of physiological processes involves a number The neuroendocrine system also provides adapt- of hormones including cortisol, growth hormone, ive physiological responses which allow one to vasopressin (antidiurectic hormone), renin, aldos- respond to and cope with changes in the environ- terone, thyroxin, insulin, glucagons, and the cate- ment. This ability to adapt to such changes is central cholamines, epinephrine and norepinephrine. The to maintaining a number of physiological parameters epinephrine and norepinephrine released from the within a ‘normal range’ that allows for survival. adrenal gland control changes in muscle metabol- For example, the hypothalamic–pituitary–adrenal ism, cardiac output and vascular resistance. There (HPA) axis is among the first physiological response may also be changes in the levels of other hormones systems to become activated in response to environ- (e.g. estrogen, follicle-stimulating hormone [FSH], mental ‘stress’. Although ‘stress’ can be thought of luteinizing hormone [LH] and testosterone, α- and from either a strictly psychological or physiological β-endorphins and enkephalins) that are not neces- standpoint, or a combination of both, the ultimate sarily associated with the maintenance of homeo- bodily response is the activation of the HPA axis stasis. Peptide hormones such as growth hormone, (Chrousos & Gold 1992; Dhabhar & McEwen 2001). insulin and prolactin increase with exercise as part A well-orchestrated communication between the of the metabolic response. Many, if not all, of these cells and tissues comprising the HPA axis is essen- hormones are able to bind to immune cells and elicit tial in maintaining homeostasis under conditions of a variety of cellular responses. stress. Signals from the limbic system trigger the The above mentioned neuroendocrine responses release of corticotrophin-releasing hormone (CRH) to exercise have been recognized for quite some from the paraventricular nucleus of the hypotha- time. In addition, many of the hormones that are lamus which, in turn, induces the anterior pituitary the products of these responses have been shown to release adrenocorticotropic hormone (ACTH). to affect various aspects of immune function both ACTH then enters the circulation where it interacts in vitro and in vivo in the context of studies totally with cells of the adrenal cortex to produce cortisol unrelated to exercise. However, as outlined in depth in humans (corticosterone in rats and mice). It is later this chapter, it is now recognized the exercise important to recognize that through the ability of itself may indeed influence immunity through the each of these compounds to feedback onto the actions of these neuroendocrine-derived hormones. organs in which they were made, the synthesis of these molecules is well-controlled. Nervous–endocrine–immune communications neuroendocrine adaptations introduction to neuroimmunology to exercise As is briefly outlined above, intricate levels of com- The ability of the body to tolerate exercise is regu- munication exist between nervous and endocrine lated by complex interactions between the auto- systems. In fact, most the other major organs systems nomic nervous system and the endocrine system. of the body (e.g. circulatory, respiratory, gastroin- The body responds to both neural stimulation and testinal, reproductive) are functionally linked with exercise and immunity 371

both the nervous and endocrine systems and/or At the cellular level, much more is known about the with each other in a number of ways. However, functioning of both the innate and adaptive arms from a historical perspective, the immune system of the immune system and how these two arms had generally been thought to function autonom- are intertwined via the synthesis of and response ously with little or no input from any of the other to cytokines. A better understanding of immune organ systems. However, during the past three events at the molecular levels have been advanced decades there have a plethora of both human by developments in molecular biology and through (Solomon & Moos 1964; Solomon et al. 1966; Solomon the use of transgenic animals. However, our under- 1981a, 1981b; reviewed in Glaser & Kiecolt-Glaser standing of how the neuroendocrine systems affect 1994) and animal (Solomon et al. 1968; Solomon these molecular events is just beginning to be 1969; Ader & Cohen 1975; reviewed in Bonneau et al. defined. 2001; Moynihan & Stevens 2001) studies that have provided substantial evidence that the immune sys- mediators and mechanisms of tem is functionally integrated with both the central neuroendocrine–immune interactions nervous system (CNS) and with the endocrine sys- tem. Given the fact that the identification of many of Recent experimental evidence has proven that the the complex interactions among these three systems CNS–endocrine–immune axis (neuroendocrine– had their roots in psychology, this area of study has immune axis) operates in a bi-directional fashion traditionally been known as ‘psychoneuroimmuno- (Chambers et al. 1993; Felten et al. 1993; Moynihan logy’ (Greer 2000) although a simpler term, ‘neuro- & Ader 1996; Stevens-Felten & Bellinger 1997). That immunology’ may be equally appropriate. is, the immune system receives signals from the Defining the relationship among the nervous, nervous system and the immune system provides endocrine, and immune systems is not easy for information to the nervous system. This intercel- several reasons. First and foremost, the immune lular communication can be mediated by products system itself is very complex. It is comprised of both of the immune system including cytokines, growth primary (bone marrow and thymus) and secondary factors and even neuropeptides made by the lym- (lymph nodes, spleen) immune organs distributed phocytes themselves. Therefore, the distinctions that throughout the body. Although for practical and had once been made among lymphokines, growth, ethical considerations blood-derived immune cells factors, hormones and neuropeptides with respect are studied most often in humans, one can not forget to the organ system in which they function are no the fact that there are immune cells in other parts of longer appropriate. the body that play important regulatory roles. For Studies of the relationship between neuro- example, a variety of immune-derived cells line peptides and the regulation of immune function both the respiratory and intestinal systems and con- have focused particularly on those neuropeptides tribute to what is known as mucosal immunity. derived from the polyprotein proopiomelanocortin Together with the intraepithelial lymphocytes which (POMC), particularly ACTH and β-endorphin. reside in the skin, these cells provide an important Other hormones such as cortisol, growth hormone, first line of defense against invading pathogenic prolactin and the catecholamines, epinephrine microorganisms. Thus, by limiting studies to only and norepinephrine are also central to our under- blood-borne immune cells we may compromise standing of the relationship between endocrine our ability to truly learn the importance of neuro- and immune function. Overall, the evidence for endocrine–immune interactions in overall immune- neuronal, endocrine and immune intersystem com- mediated defense. A second difficulty in defining munications have formed the basis of numerous the neuroendocrine–immune relationship is that studies that have investigated the relationship our understanding of immune function continues to among physical stressors such as exercise, immune grow at a rapid pace, independent of any additional function and the health status of an individual knowledge of the nervous and endocrine systems. (Fig. 26.2). 372 chapter 26

et al. 1985; Felten & Felten 1988; Bellinger et al. 1992; Homeostasis Madden et al. 1995) in much the same way that other organs of the body, such as the heart, are innervated by the fibers of nervous system. This innervation of the thymus, lymph nodes and spleen is important in Cytokines Neuroendocrine overall immune function given that these are sites Neuroendocrine Immune hormones of lymphocyte development and antigen-specific systems system lymphocyte activation. Such innervation had been referred to as a type of ‘hard wiring’ of the nervous system with the immune system. Disease function of neuroendocrine–immune Fig. 26.2 Health and disease. (Adapted from Chambers & connections Schauenstein 2000.) The bi-directional levels of communication among the cells, tissues and organs that comprise the nerv- The ability of the immune system to respond to ous, endocrine and immune systems suggests that neuroendocrine-derived peptides and hormones the overall control and function of these systems is depends on the presence of functional receptors on much more complex than had been once thought. the immune cells themselves. Indeed, there are a Thus, the physiological processes that are critical in variety of immune-derived cells (e.g. lymphocytes maintaining homeostasis and well-being under a and monocytes) that possess receptors for neuro- number of environmental insults can theoretically endocrine-derived peptides and hormones that are be controlled in many ways. identical to those receptors that are present on cells A series of well-orchestrated events at both the of the nervous and endocrine systems (Weigent et al. cellular and molecular level are necessary for the 1990; Blalock 1994; DeKloet et al. 1994; Garza & Carr proper functioning of the immune system. His- 1997). For example, catecholamines, opioids, sero- torically, the orchestration of these events has been tonin, vasopressin, ACTH, growth hormone and generally recognized to be mediated by the variety prolactin can all influence various aspects of the of cell types that comprise the immune system immune system though receptor-specific binding. itself. Cells such as B-lymphocytes, cytotoxic T- Likewise, there are receptors on cells of the neuro- lymphocytes, helper T-lymphocytes, macrophages, endocrine system for immune-derived products neutrophils, dendritic cells and natural killer (NK) such as cytokines. cells are the most common. In recent years, the In order for specific receptors to provide a link identification of the many molecules that each of between the neuroendocrine and immune systems, these cell types synthesize (e.g. cytokines, chemo- the neuroendocrine-derived products must first kines, growth factors, etc.) and their critical role come in contact with the cells of the immune system. in the regulation of overall immune function has Some products are released into the blood and cir- allowed for a better understanding of the immune culate to the immune cells that are located through- response at both the cellular and molecular levels. out the body. Alternatively, products of the nervous However, cells and molecules of the nervous and system may be released from nerve terminals in endocrine systems can also be key contributors to the direct proximity of immune cells located in the general immune function, thus making it more primary (thymus) and secondary (lymph nodes, difficult to fully understand what controls the func- spleen) lymphoid tissues. The latter mechanism is tioning of the immune system. supported by the finding that nerve fibers of the To date, it has been shown that nearly every cell autonomic nervous system directly innervate the type and function within the immune system can primary and secondary lymphoid tissues (Livnat be modulated by products of the nervous and exercise and immunity 373

endocrine systems. A comprehensive review of this Recently, there has been experimental data to sup- literature is beyond the scope of this chapter. port the notion that products of the neuroendocrine However, some broad examples include neuroen- system can contribute to the immune system’s role docrine effects on antigen processing and presenta- in the defense against cancer (Berczi et al. 1998; Ben- tion, antibody production, lymphocyte activation Eliyahu & Shakhar 2001; Turner-Cobbs et al. 2001; and proliferation, cytokine production and NK cell Sephton & Spiegel 2003). The successful defense lytic activity. These effects are mediated at the level against infection and cancer relies on an enhanced of the ligand–receptor interaction, second messenger level of immune function. In contrast, it may be systems and gene expression. Events such as exer- desirable to suppress immune function in order to cise and stress contribute to these effects by their reduce the severity of diseases that result as a con- ability to induce the synthesis of products of the sequence of an overactive or inappropriately tar- nervous and endocrine systems. geted immune response. A number of autoimmune The immune system is intimately involved in our disorders fall into this category and include well- well-being and survival. The ability to ward off the known diseases such as rheumatoid arthritis, juven- deleterious consequences of infectious pathogens ile diabetes, systemic lupus erythematosus (SLE) may depend on the generation of an immune res- and scleroderma. Likewise, it is desirable to dimin- ponse following vaccination either during child- ish the magnitude of the immune responses that hood (poliovirus, hepatitis virus) or as an adult are involved in mediating allergic reactions and that (influenza virus). For infectious pathogens for are responsible for the graft rejection in organ which no vaccinations currently exist, our ability transplant recipients. Exactly how products of the to mobilize an effective innate immune response neuroendocrine system contribute to each of these through the activation of immune cells such as responses is a subject of much interest. neutrophils, NK cells and macrophages is essential. It is important to note that our knowledge of For some pathogens, the further development of an the inner workings of the immune system has adaptive immune response, a response targeted rapidly involved during the past three decades. specifically toward the challenge pathogen, may be Advances in many areas of biological technology critical. Such a response is typically mediated by the have allowed for a substantially better understand- B-lymphocytes which produce antibodies and the ing of immune function at both the cellular and the T-lymphocytes which help destroy those cells that molecular levels. Concomitantly, there has been an the pathogen has invaded and relies on for its pro- increase in our knowledge of the role of the nerv- pagation and continued insult on the body. Lastly, ous and endocrine systems at each of these levels, our ability to generate and maintain immunological thus advancing the field of neuroimmunology and memory, as a consequence of vaccination and/or furthering our understanding of neuroendocrine– actual infection, is also important in our long-term immune interactions. defense against infection. The fact that products of the nervous and endocrine systems affect all of the The impact of exercise-induced effects above underscores the important role of both the on the immune system nervous and endocrine systems in the defense against infectious pathogens. Exercise and immunity There is also substantial interest in the role of the immune system in the development and progres- ‘Exercise helps relieve fatigue and other symptoms sion of cancer. Although the precise contributions of disease’, proclaimed a recent newspaper head- of the various components of the immune system in line. The article (Bertrand 2003) described how exer- the defense against cancer have not yet been fully cise helped individuals cope with diseases such defined, there is substantial evidence to believe that as multiple sclerosis and cancer. It included a immune components such as NK cells and cytotoxic ‘checklist’ of approaches to developing an exercise T-lymphocytes (CTL) may indeed play a key role. regimen. For example, there were suggestions of 374 chapter 26

how to monitor pulse and breathing as well as how gender and prior fitness level among other variables to pace activity or to modify the intensity and dura- in the humans being studied. Many studies have tion of exercise to meet one’s needs. Similar articles been carried out using animal models of exercise relating exercise to good health and immunity are and immunity. Although such studies have pro- common in popular journals, self-help books and vided valuable information regarding this relation- newspapers. While it is generally accepted that ship, the translation of this information to the some physical activity is better than no activity and human condition is not always straightforward. that exercise is ‘good for you’, exactly how exer- For the most part, the information discussed in this cise impacts on the immune system is not clear. chapter has been limited to that which has been Although one researcher in the field of exercise and obtained in human studies. immunity rather skeptically stated that, ‘. . . num- erous attempts to link exercise to meaningful Evaluating the impact of exercise on the immune alterations in immune function have been largely system in humans unconvincing (Moseley 2000, p. 128), the potential impact of exercise on immune function can not be Important in interpreting how exercise affects the ignored nor should studies of such an impact be immune system are the methodologies that are used terminated. This is especially true in light of our to assess immune function. Two commonly used increasing knowledge of immunity and the more assays in human subject studies are the determina- sophisticated means of quantifying immune function. tion of the subpopulation of the blood leukocytes The quest to understand the connections between (phenotyping) and ex vivo stimulation of lympho- exercise and health is driven by several perspect- cytes in culture (lymphocyte activation). Due to ives (Mackinnon 2000a). If exercise can be used to both convenience and ethical considerations, most treat individuals with disease, can exercise prevent human studies rely exclusively on peripheral blood disease? On the other hand, can a person exercise as source of lymphocytes for analysis. However, too much and actually increase susceptibility to blood contains only 1–2% of the immune cells in disease? For example, some athletes undergoing the body, and many of these cells are constantly intensive and long duration exercise training suffer trafficking throughout the body, entering and exit- from upper respiratory tract infections (URTIs). ing various sites where infectious pathogens are Does this observation imply that exercise can commonly encountered. Typically, flow cytometry induce immunosuppression and thereby increase is used to enumerate the populations of these cells in the risk of infection? Can exercise protect one from the blood. This flow cytrometric approach makes or exacerbate autoimmunity? What happens to the use of fluorescently tagged monoclonal antibodies immune system of either bedridden patients who that bind to specific cell surface proteins known are largely inactive or astronauts while in space? as CD, or cluster of differentiation, markers. These Does moderate exercise modulate the immune sys- surface proteins are used to differentiate and quant- tem? Questions regarding the relationship between itate the various cell types of the immune system. physical activity and good health have been asked For example, CD3+/CD4+ cells are T-helper cells for centuries, but the mechanisms by which exer- while CD3+/CD8+ cells are T-cytotoxic cells. How- cise relates to immunity have only seriously been ever, a change in the distribution and number of studied in a systematic approach for the past 30 these cells in blood does not necessarily indicate years or so. a change in cell function nor does such a change Coupled with the complexity of the immune sys- reflect an immune response at some other location tem is the complexity of the exercise protocol that in the body. is being studied. The potential impact of exercise A variety of approaches have been used to study has much to do with the type, mode, intensity and immune cell function. One commonly used tech- duration of the exercise. The effect of exercise on nique is ex vivo culture of lymphocytes in the pres- immunity may also be further confounded by age, ence of compounds such as concanavalin A or exercise and immunity 375

phytohemagglutinin, polyclonal T-cell mitogens suppression. However, if the stressor is perceived which stimulate the lymphocytes to produce a vari- as a ‘positive stressor’ (eustress), then the neuro- ety of cytokines, to express receptors and to divide. endocrine-mediated effects may actually result in Although this technique is fairly straightforward, immunoenhancement (Dhabhar & McEwen 2001). the results are subject to a variety of experimental Thus, in establishing associations between exercise manipulations and can be affected by the particu- and the immune function, one has to consider not lar combination of the immune-derived cells that only whether or not the exercise is even perceived as are located in the blood sample being tested. To a stressor but also if the stressor is one which results allow for changes in cell numbers after exercise, the in immunosuppression or immunoenhancement. resulting functional data are often mathematically In general, in studies of the relationship between normalized to the number of T-cells in the sample; exercise and immunity, researchers consider a bout however, the presence of other cell types in the of acute exercise as a model of induced stress or blood following exercise also can affect the magni- trauma while a long duration, high intensity exer- tude of T-lymphocyte activation. cise is used as a way to bring about immunosup- The information that can be gained from using pression (Hoffman-Goetz & Pedersen 1994, 2001). the above techniques are necessarily limited by the The response of the immune system to a bout of source of the lymphocyte sample and the fact that acute exercise has been compared to that of trauma measures of lymphocyte number and activation or surgery (an example of an acute bout of exercise potential in vitro are fairly general and may not is 60 min on a treadmill or bicycle). Following this accurately reflect what is occurring in vivo. Newer exercise, numerous changes have been observed technologies such as the ability to quantify markers including increased leukocyte mobilization, release of cellular activation and cytokine synthesis by cells of proinflammatory and anti-inflammatory cyto- directly ex vivo offer the promise of unraveling the kines, tissue damage, production of free radicals mechanisms underlying exercise-induced effects on and the activation of some of the pathways seen in immune responsiveness. inflammation such as acute phase response, coagu- lation and fibrinolysis. These observations had led some to conclude that muscle damage associated Acute exercise as a stressor with exercise initates an inflammatory response As both noted above and described in detail below, which is responsible for these immune changes exercise is known to have many beneficial effects on (Hoffman-Goetz 1996). However, it is now known the body, including the immune system. However, that muscle damage is not essential for many of depending on whether or not a given exercise is these changes in immune parameters (Pedersen & perceived as a stressor by the body will determine Hoffman-Goetz 2000). Nevertheless, when muscle its impact on immune function. The impact of damage does occur, the immune system, especially both physical and psychological stressors on the the innate system, can effectively bring about heal- immune system has been studied in great depth and ing. There are many changes that occur following a has been reviewed in detail elsewhere (Glaser & bout of acute exercise regardless of muscle damage. Kiecolt-Glaser 1994; Buckingham et al. 1997; Rabin For example, a general leukocytosis (i.e. a several- 1999; Marsland et al. 2002; Moynihan 2003; Padgett fold increase in circulating white blood cells) is a & Glaser 2003). Until recently, all forms of stress common finding. In exercise-induced leukocytosis, were thought to be generally immunosuppressive. all of the white blood cell populations usually However, it is now recognized that both the type increase to some extent. However, cells associated and degree of stress can determine its ultimate with the innate immune system such as neutrophils, impact on the immune system. For example, if a NK cells and monocytes increase more dramatic- stressor is perceived by the body as a ‘negative ally than do lymphocytes, the subclass of leukocytes stressor’ (distress), then both the type and degree of that is responsible for the acquired immune res- neuroendocrine activation may result in immuno- ponse. Within the lymphocyte population, the CD8+ 376 chapter 26

T-lymphocytes (cytotoxic T-lymphocytes) increase tors for all of these molecules; i.e. catecholamines, more than do the CD4+ T-lymphocytes (T-helper); growth hormone, β-endorphins, sex steroids and B-lymphocyte numbers typically change less. Leu- cortisol, and each of these hormones administered kocytosis is a rapid and transient event. It occurs individually has been reported to affect lymphocyte within about 30 min following the end of the exer- trafficking. These data generally support a model cise session but, after a few hours into the recovery in which catecholamines bind to the β-adrenergic phase, cell numbers usually return to baseline. receptors to bring about the immediate acute effects Sometimes there is a biphasic response in which while corticosteroids play a more important role there is a decline in leukocyte numbers followed in exercise of greater duration (Hoffman-Goetz & by a second increase. Because of the rapidity of the Pedersen 2001). increase in leukocytes, it is believed that hemody- Exercise may also bring about an increase in namics play the major role in recruiting cells from circulating levels of compounds that are agents of intravascular marginated pools and storage sites apoptosis (programmed cell death) for lympho- such as the spleen. Cortisol and catecholamines, cytes. For example, increased levels of cortisol and hormones that are both intimately associated with reactive oxygen species (ROS) could cause lympho- stress, are believed to be involved in this cell recruit- cyte apoptosis which would contribute to loss of ment (Mackinnon 2000a). Moreover, the recruited lymphocytes from the circulation. However, in a lymphocytes tend to be the memory cells (CD45RO+) study designed to test this hypothesis, it was noted in contrast to newly-differentiated cells (CD45RA+) that increases in cortisol, F2-isoprostanes (indic- (Gabriel et al. 1993). Memory cells are those which ative of ROS), epinephrine and norepinephrine have previously responded to foreign material increased with exercise but the total number of cir- (antigen) and are programmed to quickly respond culating apoptotic lymphocytes did not (Steensburg to another encounter with the same antigen. et al. 2002). This finding does not rule out the Whereas the newly differentiated cells tend to be importance of these molecules but suggests that predominantly located in lymphoid organs, the they may not necessarily influence circulating blood memory cells tend to home to tissues as well as to cell numbers. lymphoid organs. Thus, this somewhat selective leukocytosis of the two populations which reside lymphocyte subpopulations in different lymphoid compartments may provide clues to the hemodynamic and hormone driven T-lymphocytes are an essential component of cell- increases in these cells in the blood. How these exer- mediated acquired immunity and play an import- cise-induced, hormone-mediated changes in leuko- ant role in the defense against a number of viral cytosis influences immune responses to pathogens infections. Both of the T-cell subsets, T-helper (Th; is of significant interest. CD4+) and T-cytotoxic (Tc; CD8+), first increase and An intense bout of acute exercise also affects then decrease after severe, acute exercise (Steensburg the endocrine system by elevating the levels of et al. 2001). The Th cells provide a series of cytokines variety of hormones such as the catecholamines that are essential to regulate the immune response. (epinephrine, norepinephrine), growth hormone, Tc cells kill virally infected or tumor cells by direct β-endorphins, sex steroids and cortisol (Hoffman- contact. Each of these T-cell subsets are further Goetz & Pedersen 2001) in the blood. Thus, given classified by the cytokines they produce. A charac- their location it seems logical that these hormones teristic signature of a type 1 cytokine response, is could also have an influence on immune cell production of interferon-γ (IFN-γ) and interleukin-2 trafficking. In support of this hypothesis, the levels (IL-2), while type 2 cytokines include IL-4 and IL- of expression of β-adrenergic receptors on the white 6. It is the balance between these sets of T-cells blood cells roughly correlates with the increase in and their cytokines that influence whether the the cell numbers seen following exercise (Hoffman- immune response will be more directed to either Goetz & Pedersen 2001). Leukocytes have recep- a cell-mediated (type 1) or a humoral or antibody- exercise and immunity 377

mediated (type 2) response. It is this balance that is inflammatory mechanism for the removal of cells of essential for assuring an effective immune defense. all types within the body. For example, newly dif- It has been reported (Steensburg et al. 2001) that ferentiated lymphocytes undergo apoptosis if they the Th1 cells decrease while the Th2 are largely do not encounter their cognate antigen. In a recent unaffected after acute exercise stress. Further- publication, Green & Rowbottom (2003) suggested more, Ibfelt et al. (2002) found that the decrease that increased apoptosis may account for an appar- may be due largely to a decrease in memory cells ent decrease in cell proliferation. By combining vital (CD45RO+). Such a decrease in both memory cells staining with a fluorescent dye (CFSE) and flow and the type 1 cytokines associated with an exercise- cytometric analyses, this group found that exercise stressed immune system might result in the inabil- did not affect cell division but rather increased cell ity of one to mount an effective response against a death. Thus, in this study, the overall observed viral infection. decrease in lymphocyte expansion was interpreted to be the net result of cell division and apoptosis, lymphocyte activation and apoptosis with apoptosis being the greatest contributor. in culture This study was limited in the number of samples analyzed so it remains to be seen if the finding is Upon recognition of a specific antigen in vivo, both universal. T- and B-lymphocytes become activated, grow into In another case in which apoptosis was directly large blast cells and undergo several rounds of cell examined with the use of annexin V, a protein division. This clonal expansion of antigen-specific which binds to the membrane of apoptotic cells, lymphocytes assures that the cell numbers will be exhaustive but not moderate exercise increased adequate for mounting an effective defense. This lymphocyte apoptosis (Mooren et al. 2002). There clonal expansion is mimicked in vitro by the stimula- are various reports that exhaustive exercise brings tion of lymphocytes with polyclonal mitogens. about cellular changes which may participate in the The amount of DNA synthesis carried out by these apoptotic process such as DNA damage, increased stimulated cells has long been used as an index of in Ca2+ levels (Mooren et al. 2001), ROS, cortisol and vivo cell-mediated immune function. Based on this death–death ligand molecules. However, which one assay, it has often been observed that intense or pro- of these is cause of apoptosis and which is the effect longed exercise leads to a suppression of the DNA of apoptosis remain to be determined. Whether synthesis of lymphocytes sampled from the blood the observed apoptosis is indicative of a state of immediately after exercise (reviewed by Mackinnon immunosuppression or is simply a normal regulat- 2000a). This suppression is usually short-lived and ory mechanism associated with exercise-induced cells from blood samples taken a few hours after the stress, is not known. end of the exercise respond normally. This kinetic analysis suggests that the observed suppression natural killer cells may be, in part, explained by the redistribution of cells in the blood as monitored by the phenotype NK cells, which make up about 5–10% of the circu- analysis. Furthermore, the measured levels of 3H- lating white blood cell population, offer an immedi- thymidine incorporation by cells (a measure of ate and major response to the presence of virally DNA synthesis) is an average of all the cells in the infected cells and are believed to offer surveillance culture, yet some of the lymphocytes may be non- against cancer. Both NK-cell number and activity responders or, in fact, undergo cell death. have been examined in numerous acute and chronic Research based on the use of new techniques exercise studies. As is described below, some studies suggests that lymphocyte cell death via apoptosis have described major effects of exercise on NK cells may contribute, in part, to the apparent decrease while others do not. Faced with the plethora of in proliferation associated with exercise (Green & information regarding NK cells and their response Rowbottom 2003). Apoptosis is a normal and non- to exercise, Shephard & Shek (2000b) recently 378 chapter 26

carried out a meta-analysis of the existing literature. receptor, the B-lymphocyte becomes activated and After considering many parameters, including the develops into a plasma cell that, in turn, secretes type of exercise, they concluded that sustained, large quantities of antibodies into the blood, saliva, moderate exercise usually causes an increase in cir- and other mucosal tissues. These antibodies can culating NK cells during exercise, followed by a protect the host in several ways including neutraliz- decrease within about 1 h, and recovery to normal ing viruses, targeting bacteria for opsonization by levels within 2 h of the end of the exercise. Both phagocytes, activating complement, and binding to sustained vigorous exercise and a short bout of and clearing foreign particles. The immunoglobulin vigorous exercise brought about the same pattern G (IgG) class of antibody makes up the largest of changes in NK-cell numbers, but the magnitude proportion of serum immunoglobulins and plays of these changes were greater with the short bout of a critical role in each of the above mechanisms of moderate exercise, with no apparent effect of train- protection. ing on these numbers. These changes in circulating In general, the circulating levels of IgG or NK-cell numbers have been attributed to changes in total immunoglobulins have been found not to cardiac output and to increased synthesis of cate- change following acute exercise or chronic training cholamines associated with the exercise. (Mackinnon 2000a). Although there is a report of There also are reports of increases in NK-cell low serum immunoglobulin levels in elite swim- cytolytic activity as well as increases in their mers (Gleeson et al. 1995), the production of anti- numbers (Mackinnon 2000a). Does this change in bodies in response to vaccination is not suppressed activity reflect only the change in cell number or and there is no other obvious immunosuppression does activity on a per cell basis actually change? in these individuals (Gleeson et al. 1996; Bruunsgaard For most, but not all studies, it appears that, on a per et al. 1997b). Thus, exercise does not seem to affect cell basis, the activity is not affected (Miles et al. the overall B-cell-mediated component of the im- 2002). In summary, an acute bout of exercise leads mune response. to a transitory increase and then suppression of NK-cell counts and associated cytolytic activity. mucosal immunity However, there is no evidence that these fluctua- tions in number and function affects overall health. Immunoglobulins of the IgA and IgM subtypes are Authors of the meta-analysis study recommend enriched in mucosal tissues such as the linings of the sampling during exercise and beyond the 24-h oral and gastrointestinal cavities. These immuno- recovery period and to measurements of cate- globulins provide an important role in the protec- cholamines, cortisol and prostaglandins in order tion from pathogens that enter the body with both to more definitively examine how exercise affects air and food. In contrast to total serum immuno- NK cells. globulin, salivary IgA levels have been found to be reduced after an acute bout of exercise in high per- humoral immunity immunoglobulin formance athletes (reviewed by Gleeson & Pyne levels 2000). This observation was first published in 1982, by Tomasi et al. (1982) who reported decreased The humoral immune response is mediated by B- secretory IgA in cross-country skiers. The original lymphocytes, the immune cells of the body that pro- thought of these investigators was that part of this duce immunoglobulins more commonly referred decrease was due to cold temperatures. However, to as antibodies. These antibodies are glycoprotein reduced levels of salivary IgA have since been molecules that function as very specific cell mem- measured in other athletes with intensive training brane receptors for foreign molecules (antigens) schedules such as elite swimmers, tennis players, that are encoded for by all pathogens and, in some marathon runners, rowers, cyclists, hockey players cases, may be present on tumor cells. Upon bind- and kayakers (reviewed by Pedersen & Hoffman- ing of the antibody to the cell-associated antigenic Goetz 2000). Generally, the levels return to pre- exercise and immunity 379

exercise levels within about an hour, but in some pret these findings it is necessary to remember that cases, chronically depressed levels of salivary IgA the DTH response is a short-term response while were associated with long-term training. Moderate an immune response to vaccination involves a exercise does not appear to bring about these much longer period. Furthermore, another study changes. Salivary IgM levels mirrored the changes carried out in normal, healthy individuals, indicated in IgA. Salivary levels of IgA are of special interest that high intensity progressive resistance training because of the high incidence of URTIs in athletes did not affect a DTH response (Rall et al. 1996). involved in high intensity training. Yet, it has not The idea that training may be immunosuppress- been easy to directly link reduced salivary IgA with ive is based on the observation that overtraining increased risk of URTI (Pyne et al. 2001; Novas et al. or exhaustive training sometimes encountered by 2003). There are only a few studies that indicate a marathon runners or long-distance swimmers is general risk associated with exercise and suggest often associated with URTIs. These athletes also that salivary IgA and IgM can be used to predict may show decreased levels of salivary IgA and IgM. infection. The mechanisms by which reduced IgA is Thus, it is concluded that the infection is due to related to infection are speculative. The lower levels the reduction in the salivary immunoglobulins (see of salivary IgA may reflect secretion rates and fluid Mackinnon 2000b). In a recent review, Smith (2003) changes brought about by the exercise. These lower proposed that the URTIs associated with over- levels did not appear to be linked to increases in training or excessive exercise, might be associated cortisol caused by the exercise (McDowell et al. 1992; more with tissue trauma and the production of Dimitriou et al. 2002). Other stress-induced mole- stress molecules and hormones than with reduced cules such as prostaglandins may be involved immunoglobulin levels. Smith suggests that the (Tvede et al. 1989). training brings about type 2 (Th2) cytokine produc- tion and suppresses the production of type 1 (Th1) -associated cytokines. This balance of cytokines Athlete versus non-athlete would put the immune system in a compromised If exercise affects the immune system, is the position for fighting off viral infections. While this immune system of a well-trained athlete different hypothesis is supported by data in the literature, than a non-athlete? The simple answer is ‘no’. By the there is still the need to demonstrate that cytokine various criteria used to assess immunity, no major responses are affected by overtraining and that differences have been identified. The adaptive URTIs are the outcome. In summary, the majority of immune system remains unaffected for the most information suggests that the immune system of the part (Nieman & Pedersen 1999). Although a heavy trained athlete is not substantially different from bout of acute exercise even in the trained athlete that of a healthy sedentary person. However, excess- causes transient changes in immune cells in the ive or overtraining may, on a short-term basis, blood, there is no compelling evidence that these make the athlete more susceptible to URTIs but not changes are lasting or affect immune system func- necessarily due to immunosupression (Weidner tion. Furthermore, although it has been suggested et al. 1998). that this stress-like response leaves the athlete with a somewhat suppressed immune system, vaccina- Mechanisms for exercise–immune interactions— tion, which involves the co-ordinated function of the role of cytokines both B- and T-lymphocytes, has been shown not to be affected in male triathletes (Bruunsgaard et al. Although there is a variety of observations that 1997b). On the other hand, a skin test (delayed type point to exercise as a regulator of immune func- hypersensitivity [DTH]) response to recall antigen tion, the molecular mechanisms that underlie this was shown to be decreased when carried out imme- interaction of exercise and immunity remain to be diately after completing a half-ironman triathlon fully elucidated. On the other hand, there is evid- competition (Nieman & Pedersen 1999). To inter- ence that products of the neuroendocrine system 380 chapter 26

that are elicited by exercise can affect immune such as the ability to inhibit tumor necrosis factor-α function. (TNF-α) and IL-1 and the induction of IL-1 receptor It has long been recognized that leukocyte- antagonist. At one point, it was believed that exer- derived cytokines act as major communicators and cise-induced IL-6 resulted from muscle damage regulators of the immune system (reviewed by (Bruunsgaard et al. 1997a; Pedersen 2000). While Vilcek 2003). These low molecular weight proteins this may indeed be true, muscle damage is not or glycoproteins secreted by many cells in the body necessary to observe increases in IL-6. Although a in response to various stimuli are present in low correlation has been reported between the exercise- concentrations in the blood. In contrast to the classic induced increases in both IL-6 and epinephrine, endocrine-derived hormones which act on their the infusion of epinephrine did not mimic the extent target cells and tissues at distant sites, cytokines or the kinetics of changes in IL-6 found with exer- typically act locally in either an autocrine or para- cise (Steensberg et al. 2001). IL-6 is one of many crine manner. Many of these cytokines retain the cytokines (e.g. IL-1, TNF-α, IFN-γ, IL-12) that are name ‘interleukin’ (e.g. interleukin-2 or IL-2), thus known to stimulate the HPA axis. As is outlined implying their production by a leukocyte to their above, the products of the HPA axis, in particular ability to act on other leukocytes. Because they can the glucocorticoids, are able to have multiple effects act pleitropically and may be redundant in function, on immune function. IL-6 has a direct link to the it is often difficult to pinpoint their exact mechan- HPA axis given that pituitary cells express receptors ism of action. In fact, we now know that cytokines for IL-6 and respond to its presence (Besedovsky & produced by immune cells not only provide com- Del Rey 2001). Receptors for IL-6 and others cyto- munication among cells of the immune system kines have been identified in the brain as well. but also serve as a means of communication among Interestingly, Nybo et al. (2002), found that the the immune, nervous, and endocrine systems. The brain itself could release IL-6 following a bout of impact of exercise on cytokine production explains, prolonged exercise, albeit at a much lower level in part, how exercise can modulate both immune (approximately 100-fold less) than skeletal muscle. and neuroendocrine systems. In a recent review article, Suzuki et al. (2002) Initially, assays of biological activity (bioassays) summarized the results of a number of studies in were used to identify cytokines and growth factors which circulating cytokine levels were detected in in the plasma or serum following exercise. In 1983, the plasma following various exercise regimens. researchers (Cannon & Kluger 1983) showed that However, a major consideration presented in this the plasma obtained from subjects following exer- review was whether or not measuring the plasma cise, when injected into rats, caused an increase in levels of cytokines provides sufficient information body temperature. This bioassay for cytokines that since both the localization and utilization of cyto- produce fever, indicated the presence of IL-1 or kines in specific organs and tissues (both immune a related cytokine Both the development of more and non-immune) are also important in assessing sensitive assays for this and other cytokines and the cytokine contribution to overall immune function. production of recombinant molecules has suggested Regardless, in approximately one-half of these roles for many other cytokines following exercise studies, the proinflammatory molecule TNF-α was (Pedersen & Toft 2000). Furthermore the plethora of shown to increase. Plasma IL-6 concentrations also observed exercise-induced increases in cytokines increased with the greatest increases usually occur- led many to suggest that exercise causes an in- ring with exercise that caused potential muscle flammatory-like response. One such inflammation- damage. The anti-inflammatory cytokines inter- related cytokine that is produced in large amounts leukin-1 receptor antagonist (IL-1ra) and IL-10 were after exercise, and thus has been the subject of many seen to increase in many studies. In contrast, none of studies, is IL-6 (Pedersen & Toft 2000). Although IL- these studies reported a change in IFN-γ; some even 6 has been categorized as a proinflammatory mole- found a decrease in IL-2. cule it also has many anti-inflammatory properties In order to try to reconcile some of these conflict- exercise and immunity 381

ing reports, these same investigators (Suziki et al. viral infections 2002) selected 16 target cytokines, including pro- and anti-inflammatory, immunomodulatory, multi- The influence of the neuroendocrine system on functional, colony stimulating and chemokines. immunity to infectious pathogens has been better They measured both plasma and urine concentra- defined during the past decade. Many of the studies tions at 10 min and 2 and 24 h following a short- to define the mechanisms underlying this relation- duration maximal exercise. The findings were ship have relied on the use of stress-based models surprising. For example, although plasma TNF-α in mice (Sheridan et al. 1998; Bonneau et al. 2001). was not detected, the levels of TNF-α in the urine Although far fewer studies have explored the increased fivefold. An opposite finding was seen impact of stress on the development of infections in with IL-1β, where concentrations increased in the humans (Cohen & Herbert 1996; Cohen & Miller plasma but were undetectable in the urine. IL-2, IL- 2001), the significance of these studies is broad. 12, IFN-α and IFN-γ either were not detected or Stress and other psychosocial factors have also been did not change in either plasma or urine. The anti- shown to influence the ability of vaccinations for inflammatory cytokines changed more dramatically hepatitis B virus and influenza virus to elicit in response to exercise. For example, IL-1ra increased vaccine-specific immune responses (Yang & Glaser in plasma and urine as did IL-4. IL-6 increased in 2002). both fluids but only significantly in urine. While Although there is clearly a role for stress– the proinflammatory cytokines TNF-α and IL-1 neuroendocine–immune interactions in the defense increased, they showed a delayed response com- against infectious disease, there is little evidence in pared with anti-inflammatory cytokines, particu- humans that moderate exercise plays a signific- larly IL-1ra. Together, these results indicate that ant role in the defense against viral and bacterial because of the difference in kinetics of production, infections. For example, a recent study (Weidner half-life and clearance, sampling one cytokine at a & Schurr 2003) found that normally sedentary but time either during or after exercise is likely to pro- healthy individuals did not exhibit differences in vide misleading results. the type, severity, of duration of cold symptoms with the addition of exercise. Human immuno- deficiency virus (HIV) infection is directly related to The impact of exercise on diseases that are the immune system since HIV infects immune cells, immunologically resisted i.e. T-cells and macrophages. While exercise may The immune system has long been known to play a help boost the CD4 levels of HIV+ individuals it key role in the resistance to a variety of infectious does not reverse the disease. Likewise, exercise can pathogens. Although not as well-defined, there is increase the general well-being of individuals at all also evidence of a role for immune function in the ages but does not reverse or prevent age or disease successful defense against cancer. However, often influenced changes. forgotten are the undesirable contribution of the There have also been a number of recent murine- immune system in conditions such as allergy and based studies conducted in an attempt to define autoimmunity. Thus, the impact of exercise on the impact of exercise on the immune response to immune function can range from beneficial to and susceptibility to viral pathogens. Although the detrimental depending on the nature of the exercise findings of these studies vary depending on the and the role of the immune system in the disease of type of exercise and the nature of the pathogen, they interest. For many years, there was only anecdotal have been valuable in dissecting the many effects evidence that exercise can affect the development of exercise on immune-related processes includ- and/or progression of diseases that have an im- ing cytokine (Davis et al. 1998; Kohut et al. 1998, munological component. However, only recently has 2001a, 2001b) and antibody (Kohut et al. 2001a) pro- there been experimental proof that such a relation- duction, antigen presentation (Ceddia & Woods ship does indeed exist. 1999), macrophage function (Davis et al. 1997) and 382 chapter 26

macrophage chemotaxis (Ortega et al. 1997). Such variable for several types of cancer. A reduction in animal-based studies will continue to provide a circulating hormones and disruption of the men- wealth of information in helping to define the cel- strual cycle in some women undergoing strenuous lular and molecular mechanisms underlying the exercise has been invoked as a mechanism by which interactions among exercise, immunity and suscept- exercise can directly lessen the risk of breast cancer ibility to infectious pathogens. (reviewed by Hoffman-Goetz et al. 1998). However, few women exercise to this extent. Another recent exercise and cancer study with women who carry mutations in BRCA1 or BRCA2 genes, and thus putting them at a higher Does exercise reduce the risk of cancer? The simple risk for cancer, indicated that they may be protected answer is ‘yes’. Many, but not all, epidemiological from or delay the onset of breast cancer by remain- studies have indicated that there is indeed a correla- ing active and maintaining a healthful weight while tion between physical activity and a reduced risk young (King et al. 2003). of cancer (reviewed by Shephard & Shek 1995, Depending on the intensity, exercise may also 2000a; Gammon et al. 1998; McTiernan et al. 2003). bring about increased levels of catecholamines, Is this correlation due to a direct affect of exercise glucocorticoids, β-endorphins, growth hormone on the immune system and its ability to prevent the and prolactin in the circulation. These hormones all development and/or progression of cancer? While have a direct link to cells of the neuroendocrine and it is possible, there is no convincing evidence for immune systems but the link to cancer is much more such a direct connection. tenuous. Likewise, the changes in the levels of NK Several epidemiological studies suggest that cells or other leukocytes in the circulating blood physical activity correlates with the incidence of caused by intense exercise have never been directly various cancers including those of the colon, breast, linked to susceptibility to any disease, let alone prostate, testes and lung (reviewed by Gammon cancer. It should be noted that epidemiological et al. 1998). However, not all of these studies were studies are confounded by diet, body build, health carried out in the same way. For example, some history and other countless factors that can con- studies included activities related to a job or daily tribute to the development and progression of living (e.g. lifting boxes or housework) whereas cancer. Thus, there is no simple answer to explain other studies limited assessment to leisure time how an active life style may protect against cancera activities (e.g. weight training, aerobic exercise). if it even does! Other variables included the nature of the exercise (i.e. strenuous or moderate) and the time of life autoimmunity during which the activity was performed (reviewed by Gammon et al. 1998). A recent study (McTiernan If exercise can modulate immune responsiveness et al. 2003), to determine the impact of strenuous or can it affect autoimmunity? There are many dis- moderate recreational physical activity on the risk eases with an underlying autoimmune pathology; of breast cancer, indicated that risk decreased with for example, diabetes, multiple sclerosis, rheuma- exercise of longer duration and that the exercise did toid arthritis. Some of these can be reproduced in not have to be strenuous. This study was carried out animal models. There is anecdotal and experimental with post-menopausal women but other studies evidence to correlate some aspects of exercise stress have considered exercise at various ages of life with autoimmunity, but in general, exercise did (reviewed by Gammon et al. 1998). Whereas some not affect the development or cause of autoimmune of these studies found a correlation with age (e.g. diseases (Ferry 1996). The impact of stress and the Thune et al. 1997) others did not (e.g. Verloop et al. neuroendocrine system on aspects of autoimmunity 2000). Age is an especially important consideration has been well documented (Ligier & Sternberg 2001; for women because age inherently brings about a Prat & Antel 2001; Rogers & Brooks 2001; Wilder & change in reproductive hormones, a confounding Elenkov 2001). Therefore, it would be expected that exercise and immunity 383

exercise, though one or more components of the vascular compartments, the blood and plasma vol- neuroendocrine system, is able to regulate both the umes decrease. This hemoconcentration causes an development and progression of autoimmune dis- increase in hemoglobin and cells in the blood. ease. Exactly how this regulation takes place and Furthermore, blood may be directed away from how exercise can be used to control this regulation peripheral immune organs like the spleen with the remains to be determined. net effect that the numbers of leukocytes in the blood is increased (reviewed by Nielsen 2003). The impact of exercise on other systems Just as the heart rate rapidly increases, ventilation that may indirectly affect the immune also rapidly rises and is proportional to changes in system intensity in the exercise. The trachea enlarges to allow more, unrestricted air flow. This increased The body is clearly a well-integrated set of systems air flow may dry the mouth and have deleterious (Fig. 26.3). It would be unreasonable to assume that effects on the lining of the trachea, as well as limit all effects of exercise on immunity are mediated the amount of salivary immunoglobulins. These through neuroendocrine pathways. For one, the changes may partially explain why upper respirat- muscular system is obviously a prominent target of ory infections increase with intensive, long endur- exercise. During exercise, the muscles compete with ance exercise. other systems of the body for energy and also pro- Nielsen (2003) nicely discusses the lymphocyte duce waste. It has been suggested that the depletion response to maximal exercise from the point of view of glycogen, glucose and glutamine from muscle of the body’s cardiopulmonary responses. He sug- makes these compounds less available to the con- gests that the transient changes in cell numbers fol- tinually renewing cells of the immune system. lowing acute exercise can be explained by increased Moreover, waste products such as lactate may neg- blood flow, and that increases in ventilation and atively affect lymphocyte metabolism (Miles et al. blood pressure may affect the respiratory epithe- 2003). Muscle damage, likely to occur during certain lium making it more susceptible to infection. kinds of intensive exercise, also initiate the synthesis of inflammatory cytokines. Furthermore, the release Summary and conclusions of heat shock proteins by exercise stressed cells may act as ‘danger signals’ for the immune system There is no doubt that exercise impacts the neuro- (Moseley 2000) and activate antigen-presenting endocrine–immune axis. Exactly how, where and cells and lymphocytes. The cardiovascular system is to what extent exercise acts on this complex inter- responsible for transporting oxygen, nutrients and related system, remains to be further defined. wastes, and responds very rapidly to exercise. Heart Although there is a plethora of indicators of rate and blood pressure increase to assist the flow of immunomodulation, the ultimate question is how blood. As exercise causes fluids to leave the major these indicators are related to the functioning of the immune system and how this functioning is related to overall health. Correlations between exercise and Neuroendocrine-derived changes in immune parameters do not necessarily hormones translate to cause and effect. Even if they did, we Cytokines still might not understand when it is good to activ- Immune Exercise ate or to dampen an immune function, particularly Muscle system metabolism an inflammatory response. Cardiopulmonary/ It is difficult enough to evaluate the general status hemodynamics of the immune system in a healthy, homeostatic condition, let alone under stress from exercise. We Fig. 26.3 Mechanisms of exercise-mediated modulation of are able recognize a ‘broken’ immune system when the immune system. we see one, for example an immune system ravaged 384 chapter 26

by the HIV, inherited or acquired immunodeficien- provide a means by which to model the impact of cies and autoimmune disorders. In contrast, we do exercise on immunity. In addition, newer techniques not have a good handle on how to evaluate an and an expanding knowledge of the immune sys- apparently intact system. Illness is not necessarily tem have allowed scientists in the field to begin to an indicator of a failed immune system. Instead, it unravel some of the mysteries underlying immune may well indicate that the immune system is appro- function. It will take insightful collaborations be- priately responding to fight the acute phase of an tween scientists from the disciplines of kinesiology, infectious disease and is establishing immunolo- immunology and neuroendocrinology to move the gical memory in preparation for when the host is field forward. The goal is to reach the point where challenged with the infectious pathogen once again. the prescription is no longer a general ‘get plenty of For obvious ethical considerations, we can neither exercise’, but one in which a very specific recom- challenge humans with a pathogen nor extensively mendation is given of the intensity, mode, duration sample immune compartments where immune res- and kind of exercise needed to complement a stand- ponses to such pathogens generally occur. In these ard regimen of medical intervention in an attempt respects, studies in animals are valuable in that they enhance immune function.

References

Ader, R. & Cohen, N. (1975) Behaviorally receptors as a molecular mechanism for antigen presentation. Journal of Applied conditioned immunosuppression. communication between the immune Physiology 87, 2253–2258. Psychosomatic Medicine 37, 333–340. and neuroendocrine systems. Annals of Chambers, D.A. & Schauenstein, K. Ader, R., Felten, D.L. & Cohen, N. (eds.) the New York Academy of Sciences 741, (2000) Mindful immunology: (2001) Psychoneuroimmunology, 3rd edn. 292–298. neuroimmunomodulation. Trends in Academic Press, San Diego, CA. Bonneau, R.H., Padgett, D.A. & Sheridan, Immunology 21, 168–169. Bellinger, D.L., Lorton D., Felten S.Y. J.F. (2001) Psychoneuroimmune Chambers, D.A., Cohen, R.L. & Perlman, & Felten, D.L. (1992) Innervation of interactions in infectious disease: studies R.L. (1993) Neuroimmune modulation: lymphoid organs and implications in in animals. In: Psychoneuroimmunology, signal transduction and catecholamines. development, aging, and autoimmunity. 3rd edn. (Ader, R., Felten, D.L. & Cohen, Neurochemistry International 22, 95–110. International Journal of N., eds.). Academic Press, San Diego, Chrousos, G.P. & Gold, P.W. (1992) The Immunopharmacology 14, 329–344. CA: 483–497. concept of stress and stress systems Ben-Eliyahu, S. & Shakhar, G. (2001) The Booher, M.A. & Smith, B.W. (2003) disorders; overview of physical and impact of stress, catecholamines, and Physiological effects of exercise on the behavioral homeostasis. Journal of the the menstrual cycle on NK activity and cardiopulmonary system. Clinics in American Medical Association 267, tumor development: from in vitro Sports Medicine 22, 1–21. 1244–1252. studies to biological significance. In: Bruunsgaard, H., Galbo, H., Halljaer- Cohen, S. & Herbert, T.B. (1996) Health Psychoneuroimmunology, 3rd edn. Kristensen, J. et al. (1997a) Exercise- psychology: psychological factors and (Ader, R., Felten, D.L. & Cohen, N., induced increases in serum interleukin physical disease from the perspective eds.). Academic Press, San Diego, 6 in humans is related to muscle of human psychoneuroimmunology. CA: 545–563. damage. Journal of Physiology 499, Annual Review of Psychology 47, 113–142. Berczi, I., Chow, D.A., Baral, E. & Nagy E. 833–841. Cohen, S. & Miller, G.E. (2001) Stress, (1998) Neuroimmunoregulation and Bruunsgaard, H., Hartkopp, A., Mohr, T. immunity, and susceptibility to cancer. International Journal of Oncology et al. (1997b) In vivo cell mediated upper respiratory infection In: 13, 1049–1060. immunity and vaccination response Psychoneuroimmunology, 3rd edn. (Ader, Bertrand, A. (2003) ‘Healing power, following prolonged intense exercise. R., Felten, D.L. & Cohen, N., eds.). exercise helps relieve fatigue and other Medicine in Science and Sports and Exercise Academic Press, San Diego, CA: symptoms of disease’. St. Louis Post 29, 1176–1181. 499–509. Dispatch reported in Centre Daily Times. Buckingham, J.C., Gillies, G.E. & Conti, A., Maestroni, G.J.M., McCann, S.M. Monday, September 15, 2003. Cowell, A.-M. (eds.) (1997) Stress, et al. (eds.) (2000) Neuroimmunology: Besedovsky, H.O. & Del Rey, A. (2001) Stress Hormones, and the Immune System. Perspectives at the New Millennium, Cytokines as mediators of central and John Wiley & Sons, New York. Vol. 917. The New York Academy of peripheral immune-neuroendocrine Cannon, J.G. & Kluger, M.J. (1983) Sciences, New York. interactions. In: Psychoneuroimmunology, Endogenous pyrogen activity in human Davis, J.M., Kohut, M.L., Colbert, L.H. et al. 3rd edn. (Ader, R., Felten, D.L. & Cohen, plasma after exercise. Science 220, (1997) Exercise, alveolar macrophage N., eds.). Academic Press, San Diego, 617–619. function, and susceptibility to CA: 1–17. Ceddia, M.A. & Woods, J.A. (1999) respiratory infection. Journal of Applied Blalock, J.E. (1994). Shared ligands and Exercise suppresses macrophage Physiology 83, 1461–1466. exercise and immunity 385

Davis, J.M., Weaver, J.A., Kohut, M.L. et al. exercise effects on mucosal immunity. American Journal of Physiology 275, (1998) Immune system activation and Immunology and Cell Biology 78, 536–544. L1089–L1094. fatigue during treadmill running: role Gleeson, M., McDonald, W.A., Cripps, Kohut, M.L., Boehm, G.W. & Moynihan, of interferon. Medicine and Science in A.W. et al. (1995) Exercise, stress and J.A. (2001a) Moderate exercise is Sports and Exercise 30, 863–868. mucosal immunity in elite swimmers. associated with enhanced antigen- De Kloet, E.R., Oitzl, M.S. & Schobitz, B. Advances in Experimental Medical Biology specific cytokine, but not IgM antibody (1994) Cytokines and the brain 371(A), 571–574. production in aged mice. Mechanisms corticosteroid receptor balance: Gleeson, M., Pyne, D.B., McDonald, W.A. of Ageing and Development 122, relevance to pathophysiology et al. (1996) Pneumococcal antibody 1135–1150. of neuroendocrine–immune response in elite swimmers. Clinical Kohut, M.L., Boehm, G.W. & Moynihan, communication. Experimental Immunology 105, 238–244. J.A. (2001b) Prolonged exercise Psychoneuroendocrinology 19, 121–134. Glenister, D. (1996) Exercise and mental suppresses antigen-specific cytokine Dhabhar, F.S. & McEwen, B.S. (2001) health: a review. Journal of the Royal response to upper respiratory infection. Bidirectional effects of stress and Society of Health 116, 7–13. Journal of Applied Physiology 90, 678–684. glucocorticoid hormones on immune Green, K.J. & Rowbottom, D.G. (2003) Lakier-Smith, L. (2003) Overtraining, function. In: Psychoneuroimmunology, Exercise-induced changes to in vitro T- excessive exercise, and altered 3rd edn. (Ader, R., Felten, D.L. & Cohen, lymphocyte mitogen responses using immunity: is this a T helper-1 versus T N., eds.). Academic Press, San Diego, CFSE. Journal of Applied Physiology 95, helper-2 lymphocyte response? Sports CA: 301–338. 57–63. Medicine 33, 347–364. Dimitriou, L., Sharp, N.C. & Dougherty, Greer, S. (2000) What’s in a name? Ligier, S. & Sternberg, E.M. (2001) M. (2002) Circadian effects on the acute Neuroimmunomodulation or The neuroendocrine system and responses of salivary cortisol and IgA in psychoneuroimmunology. In: rheumatoid arthritis: focus on the well trained swimmers. British Journal of Neuroimmunology: Perspectives at the hypothalamo–pituitary–adrenal axis. Sports Medicine 36, 260–264. New Millennium, vol. 917 (Conti, A., In: Psychoneuroimmunology, 3rd edn. Felten, D.L. & Felten, S.Y. (1988) Maestroni, G.J.M., McCann, S.M. (Ader, R., Felten, D.L. & Cohen, N., Sympathetic noradrenergic innervation et al., eds.). The New York Academy eds.). Academic Press, San Diego, of immune organs. Brain, Behavior, and of Sciences, New York: 568–574. CA: 449–469. Immunity 2, 293–300. Hoffman-Goetz, L. (ed.) (1996) Exercise Livnat, S., Felten, S.Y., Carlson, S.L., Felten, D.L., Felten, S.Y., Bellinger, D.L. and Immune Function. CRC Press, Boca Bellinger, D.L. & Felten, D.L. (1985) & Madden, K.S. (1993) Fundamental Raton, FL. Involvement of peripheral and aspects of neural–immune signaling. Hoffman-Goetz, L. & Pedersen, B.K. central catecholamine systems in Psychotherapy and Psychosomatics 60, (1994) Exercise and the immune system: neural–immune interactions. Journal of 46–56. a model of the stress response? Neuroimmunology 10, 5–30. Ferry, A. (1996) Exercise and autoimmuine Immunology Today 15, 382–387. McDowell, S.L., Hughes, R.A., Hughes, diseases. In: Exercise and Immune Hoffman-Goetz, L. & Pedersen, B.K. (2001) R.J. et al. (1992) The effect of exhaustive Function (Hoffman-Goetz, L., ed.). Immune responses to acute exercise: exercise on salivary immunoglobulin A. CRC Press, Boca Raton, FL: 163–178. hemodynamic, hormonal and cytokine Journal of Sports Medicine and Physical Fox, K.R. (1999) The influence of physical influences In: Psychoneuroimmunology, Fitness 32, 412–415. activity on mental well-being. Public 3rd edn. (Ader, R., Felten, D.L. & Cohen, Mackinnon, L.T. (2000a) Exercise Health Nutrition 2, 411–418. N., eds.). Academic Press, San Diego, immunology: current issues. In: Gabriel, H., Schmitt, B., Urhausen, A. CA: 123–132. Nutrition and Exercise Immunology & Kindermann, W. (1993) Increased Hoffman-Goetz, L., Apter, D., Demark- (Nieman, D.C. & Pedersen, B.K., eds.). CD45RA+ CD45RO+ cells indicate Wahnefried, W. et al. (1998) Possible CRC Press, Boca Raton, FL: 3–24. activated T cells after endurance mechanisms mediating an association Mackinnon, L.T. (2000b) Overtraining exercise. Medicine and Science in Sports between physical activity and breast effects on immunity and performance in and Exercise 25, 1352–1357. cancer. Cancer 83, 621–628. athletes. Immunology and Cell Biology 78, Gammon, M.D., John, E.M. & Britton, J.A. Ibfelt, T., Petersen, E.W., Bruunsgaard, H., 502–509. (1998) Recreational and occupational Sandmand, M. & Pedersen, B.K. (2002) McTiernan, A., Kooperberg, C., White, E. physical activities and risk of breast Exercise-induced change in type 1 et al. (2003) Recreational physical cancer. Journal of the National Cancer cytokine-producing CD8+ T cells is activity and the risk of breast cancer in Institute 90, 100–117. related to a decrease in memory T cells. postmenopausal women. Journal of the Garza, H.H. & Carr, D.J. (1997) Journal of Applied Physiology 93, American Medical Association 290, Neuroendocrine peptide receptors on 645–648. 1331–1336. cells of the immune system. Chemical King, M.D., Marks, J.H., Mandell, J.B. & Madden, K.S., Sanders, V.M. & Felten, D.L. Immunology 69, 132–54. New York Breast Cancer Study Group. (1995) Catecholamine influences and Glaser, R. & Kiecolt-Glaser, J. (eds.) (2003) Breast and ovarian cancer risks sympathetic neural modulation of (1994) Handbook of Human Stress and due to inherited mutations in BRCA1 immune responsiveness. Annual Review Immunity, Academic Press, San Diego, and BRCA2. Science 302, 643–646. of Pharmacology and Toxicology 35, CA. Kohut, M.L., Davis, J.M., Jackson, D.A. 417–448. Gleeson, M. & Pyne, D.B. (2000) Special et al. (1998) Exercise effects on IFN-β Marsland, A.L., Bachen E.A., Cohen, S., feature for the Olympics: effects of expression and viral replication in lung Rabin, B. & Manuck, S.B. (2002) Stress, exercise on the immune system: macrophages after HSV-1 infection. immune reactivity and susceptibility 386 chapter 26

to infectious disease. Physiology and Interleukin-6 release from the human exercise on anxiety, depression, and Behavior 77, 711–716. brain during prolonged exercise. Journal sensitivity to stress: a unifying theory. Miles, M.P., Mackinnon, L.T., Grove, D.S. of Physiology 542, 991–995. Clinical Psychology Review 21, 33–61. et al. (2002). The relationship of natural Ortega, E., Forner, M.A. & Barriga, C. Sephton, S. & Spiegel, D. (2003) Circadian killer cell counts, perforin mRNA and (1997) Exercise-induced stimulation of disruption in cancer: a neuroendocrine– CD2 expression to post-exercise natural murine macrophage chemotaxis: role immune pathway from stress to disease? killer cell activity in humans. Acta of corticosterone and prolactin as Brain, Behavior, and Immunity 17, Physiologica Scandinavica 174, 317–325. mediators. Journal of Physiology 498, 321–328. Miles, M.P., Kraemer, W.J., Nindl, B.C. 729–734. Shephard, R.J. & Shek, P.N. (1995) Cancer, et al. (2003) Strength, workload, Padgett, D.A. & Glaser, R. (2003). How immune function and physical activity. anaerobic intensity and the immune stress influences the immune response. Canadian Journal of Applied Physiology 20, response to resistance exercise in Trends in Immunology 24, 444–448. 1–25. women. Acta Physiologica Scandinavica Paluska, S.A. & Schwenk, T.L. (2000) Shephard, R.J. & Shek, P.N. (2000a) 178, 153–163. Physical activity and mental health: Does regular physical activity reduce Mooren, F.C., Lechtermann, A., Fromme, current concepts. Sports Medicine 29, susceptibility to cancer? In: Exercise and A., Thorwestern, L. & Völker, K. (2001) 167–180. Immune Function (Shephard, R.J., ed.). Alterations in intracellular calcium Pedersen, B.K. (2000) Exercise and Adis International, Philadelphia, PA: signaling of lymphocytes after cytokines. Immunology and Cell Biology 131–154. exhaustive exercise. Medicine and Science 78, 532–535. Shephard, R.J. & Shek, P.N. (2000b) Effects in Sports and Exercise 33, 242–248. Pedersen, B.K. & Hoffman-Goetz, L. of exercise and training on natural killer Mooren, F.C., Blöming, D., Lechtermann, (2000) Exercise and the immune system: cell counts and cytolytic activity: a A., Lerch, M.M. & Völker, K. (2002) regulation, integration and adaptation. meta-analysis. In: Exercise and Immune Lymphocyte apoptosis after exhaustive Physiological Reviews 80, 1055–1081. Function (Shephard, R.J., ed.). Adis and moderate exercise. Journal of Applied Pedersen, B.K. & Toft, A.D. (2000) Effects International, Philadelphia, PA: 37–58. Physiology 93, 147–153. of exercise on lymphocytes and Sheridan, J.F., Dobbs, C., Jung, J. et al. Moseley, P.L. (2000) Exercise, stress, and cytokines. British Journal of Sports (1998) Stress-induced neuroendocrine the immune conversation. Exercise and Medicine 34, 246–251. modulation of viral pathogenesis and Sports Sciences Reviews 28, 128–132. Prat, A. & Antel, J.P. (2001) immunity. Annals of the New York Moynihan, J.A. (2003) Mechanisms of Neuroendocrine influences on Academy of Sciences 840, 803–808. stress-induced modulation of immunity. autoimmune disease: multiple sclerosis. Smith, L.L. (2003) Overtraining, excessive Brain, Behavior and Immunity 17 (suppl. In: Psychoneuroimmunology, 3rd edn. exercise, and altered immunity. Sports 1), S11–S16. (Ader, R., Felten, D.L. & Cohen, N., Medicine 33, 347–364. Moynihan, J.A. & Ader, R. (1996). eds.). Academic Press, San Diego, Solomon, G.F. (1969) Stress and antibody Psychoneuroimmunology: animal CA: 433–448. response in rats. International Archives of models of disease. Psychosomatic Pyne, D.B., McDonald, W.A., Gleeson, M. Allergy 35, 97–104. Medicine 58, 546–558. et al. (2001) Mucosal immunity, Solomon, G.F. (1981a) Emotion and Moynihan, J.A. & Stevens, S.Y. (2001) respiratory illness and competitive personality factors in the onset and Mechanisms of stress-induced performance in elite swimmers. Medicine course of autoimmune disease, modulation in animals. In: and Science in Sports and Exercise 33, particularly rheumatoid arthritis. In: Psychoneuroimmunology, 3rd edn. 348–353. Pyschoneuroimmunology (Ader, R., ed.). (Ader, R., Felten, D.L. & Cohen, N., Rabin, B.S. (1999) Stress, Immune Function, Academic Press, New York: 159–179. eds.). Academic Press, San Diego, CA: and Health: the Connection. Wiley-Liss, Solomon, G.F. (1981b) Immunological 227–250. New York. abnormalities in mental illness. In: Nielsen, H.B. (2003) Lymphocyte Rall, L.C., Roubenoff, R., Canon, J.G. et al. Pyschoneuroimmunology (Ader, R., ed.). responses to maximal exercise: a (1996) Effects of progressive resistance Academic Press, New York: 259–278. physiological perspective. Sports training on immune responses in aging Solomon, G.F. & Moos, R.H. (1964) Medicine 33, 853–867. and chronic inflammation. Medicine and Emotions, immunity, and disease. A Nieman, D.C. (2003) Current perspective Science in Sports and Exercise 28, speculative theoretical integration. on exercise immunology. Current Sports 1356–1365. Archives of General Psychiatry 11, Medicine Reports 2, 239–242. Robergs, R.A. & Keteyian, S.J. (eds.) (2003) 657–674. Nieman, D.C. & Pedersen, B.K. (1999) Fundamentals of Exercise Physiology: for Solomon, G.F., Moos, R.H., Fessel, W.F. Exercise and immune function. Sports Fitness, Performance, and Health, 2nd & Morgan, E.E. (1966) Globulins and Medicine 27, 73–80. edn. McGraw-Hill, Boston. behavior in schizophrenia. International Nieman, D.C. & Pedersen, B.K. (eds.) Rogers, M.P. & Brooks, E.B. (2001) Journal of Neuropsychiatry 2, 20–26. (2000) Nutrition and Exercise immunology. Psychosocial influence. immune Solomon, G.F., Levine, S. & Kraft, J.K. CRC Press, Boca Raton, FL. function, and the progression (1968) Early experiences and immunity. Novas, A.M., Rowbottom, D.G. & Jenkins, of autoimmune disease. In: Nature 220, 821–822. D.G. (2003) Tennis incidence of URTI Psychoneuroimmunology, 3rd edn. Steensberg, A., Toft, A., Bruunsgaard, H. and salivary IgA. International Journal of (Ader, R., Felten, D.L. & Cohen, N., et al. (2001) Strenuous exercise decreases Sports Medicine 24, 223–229. eds.). Academic Press, San Diego, the percentage of type 1 T cells in the Nybo, L., Nielsen, B., Pedersen, B.K, CA: 399–419. circulation. Journal of Applied Physiology Møller, K. & Secher, N.H. (2002) Salmon, P. (2001) Effects of physical 91, 1708–1712. exercise and immunity 387

Steensberg, A., Morrow, J., Toft, A.D., In: Psychoneuroimmunology, 3rd edn. duration of a viral upper respiratory Bruunsgaard, H. & Pedersen, B.K. (Ader, R., Felten, D.L. & Cohen, N., illness. Medicine and Science in Sports and (2002). Prolonged exercise, lymphocyte eds.). Academic Press, San Diego, CA: Exercise 30, 1578–1583. apoptosis and F2-isoprostanes. 565–582. Weigent, D.A., Carr, D.J. & Blalock, J.E. European Journal of Applied Physiology 87, Tvede, N., Heilmann, C., Halkjaer- (1990) Bidirectional communication 38–42. Kristensen, J. & Pedersen, B.K. (1989). between the neuroendocrine and Stevens-Felten, S.Y. & Bellinger, D.L. Mechanisms of B-lymphocyte immune systems. Common hormones (1997) Noradrenergic and peptidergic suppression induced by acute physical and hormone receptors. Annals of the innervation of lymphoid organs. exercise. Journal of Clinical Laboratory New York Academy of Sciences 579, Chemical Immunology 69, 99–131. Medicine 30, 169–173. 17–27. Suzuki, K., Nakaji, S., Yamoda, M. et al. Verloop, J., Rookus, M.A., van der Koou, Wilder, R.L. & Elenkov, I.J. (2001) Ovarian (2002) Systemic inflammatory response K. & van Leeuven, F.E. (2000) Physical and sympathoadrenal hormones, to exhaustive exercise. Cytokine kinetics. activity and breast cancer risk in women pregnancy, and autoimmune disease. Exercise Immunology Review 8, 6–48. aged 20–54 years. Journal of the National In: Psychoneuroimmunology, 3rd edn. Thune, I., Brenn, T., Lund, E. & Gaard, M. Cancer Institute 92, 128–135. (Ader, R., Felten, D.L. & Cohen, N., (1997) Physical activity and the risk of Vilcek, J. (2003) The cytokines: an eds.). Academic Press, San Diego, CA: breast cancer. New England Journal of overview. In: The Cytokine Handbook, 4th 421–431. Medicine 336(18), 1269–1275. edn. (Thomson, A.W. & Lotze, M.T., Woods, J.A., Lowder, T.W. & Keylock, K.T. Tomasi, T.B., Trudeau, F.B., Czerwinski, D. eds.). Academic Press, New York: (2002) Can exercise training improve & Erredg, S. (1982) Immune parameters 3–17. immune function in the aged? Annals of in athletes before and after strenuous Weidner, T. & Schurr, T. (2003) Effect of the New York Academy of Sciences 959, exercise. Journal of Clinical Immunology 2, exercise on upper respiratory tract 117–127. 173–178. infection in sedentary subjects. British Yang, E.V. & Glaser, R. (2002) Stress- Turner-Cobbs, J.M., Sephton, S.E. & Journal of Sports Medicine 37(4), 304–306. associated immunomodulation and Spiegel, D. (2001) Psychological effects Weidner, T.G., Cranston, T., Swchurr, T. its implications for responses to on immune function and disease & Kaminsky, L.A. (1998) The effect of vaccination. Expert Review of Vaccines 1, progression in cancer: Human studies. exercise training on the severity and 453–459. Chapter 27

Exercise Regulation of Insulin Action in Skeletal Muscle

RICHARD C. HO, OSCAR ALCAZAR AND LAURIE J. GOODYEAR

Introduction alternative signaling cascades like the PI3K-inde- pendent 5′-AMP-activated protein kinase (AMPK) Insulin stimulates pleiotropic pathways resulting pathway. While insulin and exercise have been in increases in both oxidative and non-oxidative shown to regulate independent pathways leading to glucose disposal, protein synthesis, gene transcrip- the regulation of certain metabolic processes (i.e. tion and cellular growth and hypertrophy. It has glucose transport), we will also discuss how exer- long been known that exercise is able to elicit similar cise and insulin are both able to regulate common effects in skeletal muscle. A single bout of exercise signaling pathways like the family of mitogen- can have profound effects on substrate metabolism, activated protein kinases (MAPKs). Finally, we will and chronic exercise is associated with adaptations discuss the clinical relevance of the various effects of that enhance mechanical and metabolic efficiency exercise on insulin action in skeletal muscle. in skeletal muscle. Among these adaptations are changes in glucose and glycogen metabolism, pro- Insulin action in skeletal muscle tein synthesis and hypertrophy, as well as changes in gene transcription. Glucose transport Efforts have been made to uncover the mechan- isms by which exercise is able to mimic and enhance Insulin stimulation has long been known to result in specific effects of insulin, many of which result in increases in GLUT4 translocation, glucose trans- clinically relevant adaptations. A fundamental issue port, glycogen synthesis and protein synthesis in in understanding the biological effects of exercise in skeletal muscle. Insulin elicits many of its metabolic skeletal muscle is elucidation of intracellular signal- effects in skeletal muscle by activation of the class- ing mechanisms that enable exercise to enhance ical PI3K pathway (Fig. 27.1). Under normal phy- insulin responsiveness. Interestingly, while it has siologic conditions, insulin binds to the α subunit been recognized that insulin and exercise utilize dis- of the insulin receptor (IR). This binding activates tinct signaling pathways leading to various cellular autophosphorylation activity in the β subunit and effects, more recent evidence has emerged showing subsequently causes tyrosine phosphorylation of that exercise and insulin utilize similar signaling numerous insulin receptor substrates. For example, intermediates as well. once activated, the pleckstrin homology domain of This chapter will highlight how insulin regulates the insulin receptor substrate-1 and -2 (IRS-1/-2) glucose transport, glycogen metabolism and protein can dock with the p85 regulatory subunit of PI3K metabolism utilizing the classical insulin-stimulated and activate its p110 catalytic subunit. Catalytic phosphatidylinositol 3-kinase (PI3K) pathway. We activity of PI3K results in the phosphorylation of will then review how exercise is able to mimic many phosphoinositide (PI) 4,5-bisphosphate to PI 3,4,5- of these metabolic effects in skeletal muscle utilizing trisphosphate (PIP3). PIP3 is necessary to activate 388 exercise regulation of insulin action 389

Insulin receptor α subunit

GLUT4 p110 β subunit p85 IRS PIP3

PDK1

Fig. 27.1 Insulin signaling through the classical phosphatidylinositol GLUT4 3-kinase (PI3K) pathway. Insulin translocation Akt mTOR binding to the extracellular insulin ? receptor α subunit activates catalytic activity in the corresponding AS160 transmembrane β subunit. The β subunit catalyzes the 4E-BP1 S6K phosphorylation of numerous insulin Glucose GSK3 receptor substrates, some of which transport (e.g. IRS-1/-2) associate with and activate the p85 and p110 subunits of PI3K. Activated PI3K results in the elF4E production of 3,4,5-trisphosphate (PIP ), and serial activation of 3- Glycogen 3 PP1 phosphoinositide-dependent protein synthase Protein kinase-1 (PDK1) and Akt. Akt synthesis signaling bifurcates resulting in the modification of proteins involved in the regulation of glucose transporter Glycogen translocation, glycogen synthesis and synthesis protein synthesis.

3-phosphoinositide-dependent protein kinase-1 Glycogen synthesis (PDK1), which phosphorylates Akt (also known as PKB) on threonine 308. Subsequent phosphoryla- The ability of insulin to stimulate glycogen syn- tion of Akt on serine 473 by a yet uncharacterized thesis was believed to involve the activation of pro- kinase further activates the enzyme. Glycogen syn- tein phosphatase-1 (PP1), and deactivation of GSK3 thase kinase-3 (GSK3), the mammalian target of (Cross et al. 1995, 1997; Brady et al. 1998; Liu & rapamycin (mTOR), and the 70-kDa S6 protein Brautigan 2000). Through its interaction with glyco- kinase (S6K) are among the established downstream gen-targeting subunits, insulin activates PP1, which substrates of Akt. Signaling through this classical catalyzes the dephosphorlyation and activation PI3K pathway leads to increases in glucose uptake of glycogen phosphorylase. However, mice lacking via the translocation of the insulin-sensitive GLUT4 the regulatory subunit of muscle-specific PP1 (PP1G) glucose transporter from subcellular vesicles to the exhibit normal insulin-stimulated glycogen synthase plasma membrane. AS160 has recently been pro- activation (Suzuki et al. 2001), suggesting the involve- posed to be an Akt-targeted signaling protein medi- ment of alternative pathways. Under basal condi- ating GLUT4 translocation (Sano et al. 2003). tions, active GSK3 exists in a non-phosphorylated 390 chapter 27

state. Active GSK3 phosphorylates and inhibits it was a logical hypothesis that exercise and insulin glycogen synthase activity. Upon stimulation by utilized similar signaling cascades in skeletal muscle. insulin, Akt catalyzes the phosphorylation of GSK3, However, several studies have clearly demonstrated converting it from its active to inactive form. This, in that proximal insulin-stimulated PI3K intermedi- turn, relieves the negative regulation that GSK3 ates are not involved in the mechanism by which exerts on glycogen synthase and thus promotes exercise elicits its metabolic effects. For example, glycogen accretion (Cross et al. 1995). Evidence for contractile activity does not lead to increases in insulin-induced deactivation of GSK3 has been autophosphorylation of isolated insulin receptors shown in both muscle cells (Cross 1995) and skeletal (Treadway et al. 1989) or IRS-1 and IRS-2 tyrosine muscle tissue (Markuns et al. 1999; Wojtaszewski phosphorylation (Goodyear et al. 1995; Sherwood et al. 2000). et al. 1999; Wojtaszewski et al. 1999a; Howlett et al. 2002). Consistent with this, muscle-specific insulin receptor knockout mice exhibit normal exercise- Protein synthesis mediated glucose transport (Wojtaszewski et al. The function of insulin and its mitogenic and 1999a). PI3K activity has been reported to be anabolic properties in various tissues has been well unchanged with skeletal muscle contraction (Good- characterized. Several lines of evidence have sug- year et al. 1995; Wojtaszewski et al. 1996, 1999a). gested that post-prandial increases in insulin are Additionally, wortmannin, a PI3K inhibitor, does associated with increased protein synthetic rates in not impair contraction-stimulated glucose transport skeletal muscle (Shah et al. 2000). Studies have sug- (Hayashi et al. 1998). The lack of activation of these gested that insulin-stimulated increases in rates of proximal PI3K signaling intermediates demon- protein synthesis are mediated by PI3K-dependent strates that the underlying molecular mechanisms mechanisms, since PI3K inhibitors can attenuate the leading to the insulin- and contraction-induced activation of key regulatory molecules involved in stimulation of glucose uptake and glycogen syn- protein synthesis. Presumably, this mechanism in- thesis in skeletal muscle are distinct. This is also volves the sequential activation of PI3K, Akt, mTOR supported by evidence showing that the effects of and S6K, and deactivation of eukaryotic initiation acute exercise and insulin on glucose transport are factor 4E binding protein 1 (4E-BP1) (reviewed in additive (Ruderman et al. 1971). Shah et al. 2000; Kimball et al. 2002). The ability Controversy exists regarding the potential role of of insulin to promote protein synthesis in skel- Akt in contraction-mediated signaling in skeletal etal muscle is also partially mediated by PI3K- muscle. While early studies reported that exercise dependent signaling to eukaryotic initiation factor and contraction did not result in the activation 2B (eIF2B), a guanine nucleotide exchange protein of Akt (Brozinick & Birnhaum 1998; Lund et al. involved in the regulation of mRNA translation ini- 1998; Widegren et al. 1998; Sherwood et al. 1999; tiation (Welsh et al. 1998). Existing data suggest that Wojtaszewski et al. 1999a), other studies have under non-stimulated conditions, GSK3 phosphory- shown significant activation or phosphorylation lates eIF2B on an inhibitory serine residue (Welsh of Akt in intact skeletal muscles in response to con- et al. 1998). Acute insulin stimulation leads to the traction (Turinsky & Damrau-Abney 1999; Nader & phosphorylation and inactivation of GSK3, result- Esser 2001). Recently, we and others have found ing in the activation of eIF2B in skeletal muscle. that both exercise in vivo and contraction in situ via sciatic nerve stimulation increases Akt phosphory- Exercise mimics insulin action in lation in multiple rat hindlimb muscles (Turinsky skeletal muscle & Damrau-Abrey 1999; Sakamoto et al. 2002, 2003). It has also been demonstrated that, like insulin, Like insulin, exercise is able to stimulate GLUT4 exercise alters the activity of GSK3 in rat skeletal translocation, glucose transport, glycogen synthesis muscles (Markuns et al. 1999). Insulin and exercise and protein synthesis in skeletal muscle. Therefore, increased glycogen synthase activity to a similar exercise regulation of insulin action 391

extent, however, the mechanisms involved have been Early evidence suggested a role for AMPK in the shown to be slightly different. Exercise deactivates regulation of fat oxidation (Vavvas et al. 1997; GSK3α and β activity, and increases phosphoryla- Hardie et al. 1998). Recently, studies from our labor- tion to a similar degree as does insulin (Sakamoto atory (Hayashi et al. 1998, 2000) and others (Hardie et al. 2003). Another study showed that bicycle et al. 1998; Russell et al. 1999) have also described a exercise increased glycogen synthase activity imme- role for AMPK in mediating contraction-induced diately following exercise without detectable deac- glucose transport. Both contraction and the AMPK tivation of either isoform of GSK3 in vastus lateralis activator 5-aminoimidazole-4-carboxamide-1-β-d- muscle (Wojtaszewski et al. 2001). There may be ribofuranoside (AICAR) stimulate glucose trans- alternative mechanisms in the regulation of GSK3 port by insulin-independent mechanisms (Hayashi activity in skeletal muscle. Interestingly, we have et al. 1998), and AMPK activity is associated with recently found that the muscle-specific regulatory glucose transport in contracting skeletal muscle subunit of PP1 is required for regulation of glycogen (Ihlemann et al. 1999). Mice overexpressing a metabolism under basal conditions and in response dominant inhibitory AMPK mutant were shown to to contractile, but not insulin, activity (Aschenbach exhibit complete inhibition of AICAR-mediated et al. 2001). Therefore, insulin-stimulated glycogen glucose uptake, but only partial inhibition of synthase activation seems to be regulated by GSK3, contraction-mediated glucose uptake into skeletal while exercise may regulate glycogen synthase via muscle (Mu et al. 2001). There are two isoforms of α α GSK3-independent mechanisms. the AMPK catalytic subunit ( 1 and 2), and another α study has recently demonstrated that AMPK 2 knockout mice exhibit significant impairments in Exercise-stimulated AMPK activation AICAR-mediated, yet completely normal skeletal Since PI3K appears not to be required for exercise- muscle glucose transport, in response to contrac- mediated glucose transport in skeletal muscle, an tion (Jorgensen et al. 2004). These data indicate that independent pathway activated by exercise has while AMPK is required for glucose transport in been the focus of much attention. The discovery response to AICAR, alternative exercise-responsive of the AMPK as an enzyme potently regulated by signaling pathways are involved in contracting exercise has been revealed as such a pathway. As skeletal muscle. a member of a metabolite-sensing protein kinase Recent studies suggest that AMPK is involved in family, AMPK acts as a fuel gauge monitoring cellu- the regulation of glycogen metabolism, although lar energy levels. Under conditions of decreased cel- existing data are equivocal, suggesting both an lular energy status (increase in the AMP/ATP and inhibitory and activating role of AMPK in the regu- creatine/phosphocreatine ratios), AMPK signaling lation of glycogen synthesis. For example, some functions to down-regulate adenosine triphosphate studies proposed that AMPK has the ability to phos- (ATP)-consuming pathways and up-regulate altern- phorylate key proteins of glycogen metabolism in ative pathways for ATP regeneration (Fig. 27.2). vitro, such as glycogen synthase (Carling & Hardie Skeletal muscle contraction is known to deplete 1989), which would be expected to inhibit glycogen these intracellular energy stores, and consequently, synthesis (Skurat et al. 1994) and phosphorylase AMPK has been shown to exhibit robust activa- kinase (Carling & Hardie 1989), the upstream effec- tion in response to both exercise and contraction tor of glycogen phosphorylase. It seems reasonable (Winder & Hardie 1996; Rasmussen & Winder 1997; to speculate that a major role of AMPK in contract- Vavvas et al. 1997). AMPK is activated by a complex ing muscle would be to promote glycogen degrada- mechanism that involves allosteric modification, tion and inhibit glycogen synthesis, as AMPK is decreases in phosphatase activities and phosphory- known to buffer intracellular ATP levels in various lation by an AMPK recently identified to be LKB1 cell systems. In accordance with this hypothesis, (Hawley et al. 2003; Ossipova et al. 2003; Lizcano there is a study showing an increased glycogen con- et al. 2004). tent in skeletal muscle of Hampshire pigs, which 392 chapter 27

GLUT4

EXERCISE

Glucose LKB1 transport ?

GLUT4 ? translocation Fig. 27.2 Exercise regulation of 5′-AMP-activated protein kinase (AMPK) in skeletal muscle. Exercise ? AMPK ACC activates AMPK through changes in cellular energy status (i.e. ? AMP : ATP), and potentially through the recently described AMPK, LKB1. ? Malonyl CoA AMPK is known to inhibit ATP- PGC-1 consuming pathways such as protein synthesis, and activate adenosine triphosphate (ATP) regenerating pathways such as fat oxidation. CPT-1 AMPK is also believed to regulate Gene Mitochondrial mitochondrial biogenesis via PGC- α transcription biogenesis 1 , as well as additional cellular processes such as glucose transport and gene transcription and through Protein Fat mechanisms that have not been synthesis oxidation fully elucidated. ACC, acetyl-CoA carboxylase.

γ harbor a point mutation in the 3 subunit of AMPK insulin-independent pathways (Wallberg-Henriksson that renders this enzyme less active (Milan et al. 2000). & Holloszy 1984, 1985). Interestingly, while oper- In contrast to this, reports have emerged showing ating through distinct signaling cascades during γ γ that similar mutations in the 1 and 2 subunits result exercise, the post-exercise period is characterized in constitutively active AMPK in conjunction with by enhancements in insulin sensitivity and respons- elevated muscle glycogen levels (Hamilton et al. iveness. These effects of exercise on insulin action 2001; Arad et al. 2002). Pharmacological activation and skeletal muscle metabolism persist well into the of AMPK by AICAR in hindlimb muscles has been post-exercise period. shown to result in the phosphorylation and deac- tivation of glycogen synthase in rat soleus, and red Insulin sensitivity: acute exercise increases and white gastrocnemius muscles (Wojtaszewski insulin action on glucose disposal et al. 2002). In contrast, it has also been reported that AMPK functions to increase glycogen synthesis In addition to the well-characterized increase in with chronic AICAR treatment in red, slow-twitch glucose transport during exercise, prior exercise and white, fast-twitch quadriceps and gastrocne- has also been reported to improve insulin action mius muscles (Winder et al. 2000; Buhl et al. 2001). in skeletal muscle. Richter et al. (1982) made the ini- Exercise is able to regulate glucose transport tial observation that previously exercised muscles and glycogen synthesis in skeletal muscle utilizing exhibit elevated insulin-stimulated glucose uptake, exercise regulation of insulin action 393

even when the effects of the exercise session per observed between insulin delivery and clearance se are no longer present. This finding has been (Bogardus et al. 1983; Devlin & Horton 1985; Richter verified in numerous animal and human studies et al. 1989). Furthermore, studies using isolated using a variety of experimental models. In humans, muscles (in vitro or perfused) where insulin levels an increase in insulin-stimulated whole body gluc- are controlled show that insulin sensitivity is ose utilization after exercise has been observed increased with exercise (Ruderman et al. 1971). (Bogardus et al. 1983; Devlin et al. 1987; Mikines et al. Additionally, the majority of studies show that 1988; Richter et al. 1989), and by using the arterioven- exercise does not increase the binding of insulin to ous balance technique this effect was shown to its receptor (Bonen et al. 1984; Zorzano et al. 1985; be primarily mediated by an increase in skeletal Treadway et al. 1989). Therefore, neither higher muscle glucose disposal (Ivy & Holloszy 1981; insulin delivery nor insulin binding to the receptor Richter et al. 1982). explains the increase in insulin action in previously An early study using a rat model first suggested exercised skeletal muscles. Taken together, these that the increase in insulin sensitivity with exer- data suggest that modulation of a post-insulin cise is restricted to the working muscle (Richter receptor event may be involved in the increased et al. 1984). One-legged exercise models in humans insulin action after exercise. have also demonstrated that the exercise-induced The cellular mechanisms leading to the increase increase in insulin-stimulated glucose uptake is in insulin sensitivity following exercise have been a local phenomenon restricted to the exercised mus- hypothesized to involve enhanced insulin signal- cles, and strongly suggested that changes in sys- ing. However, as discussed earlier, studies have temic factors are not the cause of the increase in shown that exercise does not utilize proximal PI3K insulin sensitivity to stimulate muscle glucose signaling molecules during exercise, yet specula- uptake (Richter et al. 1984, 1989). Interestingly, if tions were made that exercise resulted in enhance- isolated epithroclearis muscles from rats are con- ments in insulin-stimulated PI3K signaling in the tracted by applying electrical stimulation in vitro, post-exercise period. Prior exercise does not change no changes in insulin sensitivity are observed insulin binding to its receptor (Bonen et al. 1984; following the contraction period (Wardzala et al. Zorzano et al. 1985; Treadway et al. 1989) or increase 1985). This observation and a subsequent study insulin-stimulated receptor tyrosine kinase activity (Kolterman et al. 1980) led some to suggest that a fac- in skeletal muscles obtained from rats (Treadway tor released into the circulation during contractile et al. 1989; Goodyear et al. 1995) or humans activity is necessary for the post-exercise increase (Wojtaszewski et al. 2000). In fact, prior contraction in insulin sensitivity. Since there is an increase in of rat hindlimb skeletal muscles has been shown to insulin sensitivity to stimulate glucose transport in cause a paradoxical decrease in insulin-stimulated muscles of the perfused rat hindlimb after electrical tyrosine phosphorylation of IRS1 and IRS1-associated stimulation (Richter et al. 1984), it might be the para- PI3K activity (Wojtaszewski et al. 2000). crine action of a neurotrophic factor that initiates the In contrast to these reports, other evidence sug- events leading to the increase in insulin sensitivity gests that the increase in insulin responsiveness after exercise (Sasson et al. 1987). to stimulate muscle glucose transport immediately Exercise results in increases in blood flow to following exercise is associated with an increase working muscles, and it has been hypothesized that in insulin-stimulated PI3K activity (Houmard et al. differences in insulin delivery between exercised 1999; Chibalin et al. 2000; Kirwan et al. 2000). and rested muscles could explain the enhanced Furthermore, we have reported that prior exercise insulin action observed in the post-exercise period. resulted in enhanced insulin-stimulated IRS-2- However, using a euglycemic–hyperinsulinemic associated PI3K activity compared with insulin clamp, studies have shown that exercised legs stimulation alone (Howlett et al. 2002). The increase exhibit higher insulin sensitivity compared with in insulin-stimulated PI3K activity following exer- non-exercised legs, even when no differences are cise is short-lived, since it is not present when 394 chapter 27

insulin stimulation occurs after 30 min of recovery in insulin action. Three decades ago, Bjorntorp from exercise (J.F. Wojtaszewski & L.J. Goodyear, et al. (1972) suggested that exercise training might unpublished observation). Furthermore, if a physio- increase tissue sensitivity to insulin. This investiga- logical hyperinsulinemic clamp is applied 3 h after tion, along with several subsequent studies (Lohman one-legged exercise in humans, PI3K activity is et al. 1978; Johansen & Munch 1979; LeBlanc et al. not higher in the muscle from the exercised leg 1979, 1981; Seals et al. 1984), demonstrated that (Wojtaszewski et al. 1997). The lack of enhancement compared with sedentary individuals, trained indi- in insulin-stimulated PI3K activity in these later viduals tend to exhibit improvements in glucose time points after exercise is consistent with findings tolerance and insulin sensitivity. Additional studies in humans showing no change in insulin receptor using the hyperinsulinemic–euglycemic clamp have tyrosine kinase activity or IRS-1 tyrosine phospho- demonstrated that exercise-trained people have rylation showing no change in insulin receptor higher rates of insulin-stimulated glucose disposal tyrosine kinase activity or IRS-1 tyrosine phospho- than do their sedentary counterparts (Saltin et al. rylation (Wojtaszewski et al. 1997). Thus, although 1978; Rosenthal et al. 1983; Hollenbeck et al. 1984; there seems to be an up-regulation of some com- King et al. 1987; Mikines et al. 1989). Although these ponents of the insulin signaling pathway with results have been interpreted to indicate that exer- insulin immediately after exercise, these changes do cise training results in an increase in tissue sensitiv- not explain the long-lasting influence of exercise on ity to insulin, this concept has been complicated insulin action in skeletal muscle. These studies rule by the fact that, as discussed previously, an acute out a role for enhanced insulin signaling as a mech- bout of exercise produces major effects on both anism for increased glucose uptake after an acute whole body glucose disposal and metabolism exercise bout, and provide additional support to (Pruett & Oseid 1970; Bogardus et al. 1983; Devlin et the hypothesis that exercise and insulin act through al. 1985, 1987; Mikines et al. 1988, 1989) and skeletal distinct signaling mechanisms. muscle glucose uptake and metabolism (Holloszy Elevations in post-exercise insulin sensitivity are & Narahara 1965; Ivy & Hollozy 1981; Elbrink & also associated with the degree of muscle glycogen Phipps 1980; Fell et al. 1982; Richter et al. 1989) that depletion. It is generally known that increases in can persist well into the post-exercise period. Hence insulin sensitivity following exercise can persist if many of the effects of regular exercise may be due to muscle glycogen levels are kept low. Muscle gluc- the residual effects of the last individual exercise ose uptake following exercise remains elevated for session, rather than long-term adaptations to exer- up to 18 h if rats are fed a carbohydrate-free diet, cise training. Several studies have attempted to dis- while rates of uptake return to pre-exercise levels if criminate differences in insulin sensitivity between carbohydrate is provided following exercise (Young acute exercise effects and training adaptations et al. 1983). (Heath et al. 1983; Burstein et al. 1985; King et al. 1988; Mikines et al. 1989). Some of these studies suggest that the increases in insulin sensitivity asso- Exercise training increases insulin action on ciated with exercise training are transient (Burstein glucose disposal et al. 1985; King et al. 1988). One study has reported While the effects of an acute bout of exercise on that while 10 days of detraining resulted in insulin-stimulated glucose transport in the post- decreases in insulin sensitivity to a level comparable exercise period have been well established, these with sedentary individuals, performance of a single effects are quite transient with improvements in exercise bout by the trained subjects on the 11th day insulin responsiveness returning to pre-exercise of the protocol did not completely return the insulin levels, usually within 12 h of the preceding exercise response to the trained level (Heath et al. 1983). bout. It was reasonable to speculate that chronic These data suggest that effects of an acute bout exercise would result in adaptations that enable of exercise cannot completely account for increases skeletal muscle to exhibit more sustained increases in insulin action associated with exercise training. exercise regulation of insulin action 395

After comparing trained subjects with untrained pendent of short-term exercise effects are associated subjects who had undergone an acute bout of exer- with improvements of insulin action in skeletal cise and after comparing trained subjects before muscle. However, exercise-mediated improve- and after 5 days of detraining (Mikines et al. 1989), a ments in insulin sensitivity have been detected even different group of investigators concluded that an when studies have controlled for these potentially increase in maximal insulin action on whole-body confounding effects (Oshida et al. 1989; Hughes et al. glucose uptake (insulin responsiveness) but not 1993). Hence, exercise training appears to improve insulin sensitivity is a long-term adaptation caused insulin sensitivity independent of changes in body by endurance exercise training. Additional studies composition and there may be additive effects of have reported that maximal insulin-stimulated gluc- exercise and decreased adiposity. ose disposal was unaffected by 10 days of inactivity (King et al. 1988), suggesting that the reduction in Glycogen supercompensation: exercise increases insulin action following short-term inactivity is the insulin action on glycogen metabolism result of a decrease in insulin sensitivity and not a decrease in insulin responsiveness. The molecular The increase in glycogen synthesis in response to basis for this phenomenon has not been completely insulin stimulation has been well characterized. elucidated, but appears to be dependent on multiple Furthermore, exercise elicits a net breakdown of factors including humoral factors, muscle glyco- muscle glycogen despite increases in glycogen syn- gen concentrations and alterations in signaling thase activity. In 1966, Bergstrom and Hultman mechanisms. (1966) first described the concept of glycogen super- One study has recently reported that habitual compensation, whereby glycogen levels increase exercise was associated with decreased protein significantly in response to carbohydrate feeding expression of IR, IRS-1 and IRS-2 in trained versus following a glycogen depleting exercise bout. The untrained subjects (Yu et al. 2001). Another study preceding exercise bout leads to increases in skeletal found that 7 days of exercise training did not elicit muscle glycogen above that which is seen following changes in PI3K activity, despite increases in insulin a meal. This effect is mediated to a large extent by action in skeletal muscle (Tanner et al. 2002). The the post-exercise increases in glucose transport, enhanced activation by insulin was not associated glycogen synthase activity and the degree of glyco- with a greater insulin-stimulated insulin receptor gen depletion. or IRS-1 tyrosine phosphorylation. These data Increases in post-exercise glucose transport result suggest that the improvement in insulin action in subsequent elevations in the levels of glucose 6- associated with exercise training is not associated phosphate (G6P), an allosteric activator of glycogen with increases in PI3K signaling. Contrary to these synthase (Price et al. 2003). Therefore, the extent to data, trained rats have been shown to exhibit higher which an exercise bout increases glucose transport insulin-stimulated glucose transport, IRS-1/-2 activ- in the post-exercise period partially determines ity, PI3K association and Akt phosphorylation com- levels of glycogen synthesis. However, studies have pared with sedentary controls (Chibalin et al. 2000; shown that increases in glycogen synthase activity Luciano et al. 2002). Therefore, increases in insulin following exercise cannot fully account for the responsiveness resulting from chronic exercise observed elevations in glycogen accumulation may be mediated, in part by alterations in proximal (Bergstrom et al. 1972; Conlee et al. 1978; Nakatani PI3K signaling in skeletal muscle, but most likely et al. 1997; Greiwe et al. 1999). involves additional signaling molecules whose The degree of glycogen depletion resulting from characterization remains elusive. the preceding exercise bout is a critical factor in Efficacious exercise training programs can result determining the subsequent increases in muscle in changes in plasma lipids and body composi- glucose uptake and glycogen synthesis following tion, in particular, decreases in total body fat and exercise. In fact, the potentiating effect that exercise increases in lean body mass. These changes, inde- exerts on insulin action (glucose transport, glycogen 396 chapter 27

synthesis) is a function of the degree of glycogen bolic function (oxidative capacity) improve. In depletion achieved during the bout of exercise. particular, resistance exercise is associated with an Interestingly, post-exercise increases in glucose up-regulation of anabolic processes that allow for uptake and insulin sensitivity can be reversed with skeletal muscle growth and regeneration. One of the carbohydrate feeding and glycogen repletion (Host main underlying factors responsible for these adap- et al. 1998; Kawanaka et al. 2000). Similarly, main- tations is the ability of exercise to elicit transient taining low glycogen levels following exercise by increases in protein synthesis (Booth & Thomason carbohydrate restriction extends the effect of exer- 1991; Fluckey et al. 1996). A protocol involving 4 cise on insulin action in skeletal muscle (Garcia- days of resistance exercise resulted in elevations in Roves et al. 2003). However, a study performed in insulin-mediated protein synthesis in rat skeletal individual tissues from perfused hindquarter mus- muscle (Fluckey et al. 1996). Other studies have cle suggests that glycogen is not the only regulator reported no such effect following endurance exer- in the supercompensation phenomenon (Zorzano cise (Dohm et al. 1980; Balon et al. 1990). This may et al. 1986). In this report, glucose transport returned partially explain why chronic resistance exercise to near baseline values even though glycogen con- results in muscle hypertrophy, while chronic centrations were maintained at a low level. This endurance exercise does not. AMPK functions to result suggests the existence of alternate com- down-regulate energy consuming pathways, while pensatory mechanisms that may participate in the up-regulating ATP-regenerating pathways. Protein regulation of the supercompensation phenomenon synthesis is an expensive energy consuming pro- together with the glycogen levels. The precise cess, and subsequently, studies have shown that mechanism(s) involved in the supercompensation AMPK is involved in the inhibition of protein syn- effect remains to be elucidated. A recent study using thetic pathways (Bolster et al. 2002; Horman et al. AICAR treatment as a pharmacological model to 2002; Krause et al. 2002; Kimura et al. 2003). There- mimic exercise, in fast-twitch (epitrochlearis) and fore, increases in protein synthesis in skeletal muscle slow-twitch (flexor digitorum brevis) muscles, re- following exercise appear to occur via AMPK- vealed that AICAR has no effect on either glycogen independent mechanisms. synthase or glycogen phosphorylase in both muscle The capacity of exercise to mimic the anabolic types (Aschenbach et al. 2002). Therefore, this study effects of insulin has been attributed to the activa- suggests that AMPK does not directly regulate tion of similar signaling mechanisms. For example, glycogen synthase or glycogen phosphorylase in studies have demonstrated that exercise activates skeletal muscle. Consistent with this finding, a recent MAPK, Akt, S6K and eIF2B, all of which have investigation using four different phosphorylase been implicated in the regulation of skeletal muscle kinase substrates demonstrated that glycogen phos- protein synthesis by insulin (Baar & Esser 1999; phorylase is not a substrate of AMPK in vitro (Beyer Farrell et al. 2000; Sakamoto & Goodyear 2002). et al. 2000). Further efforts are required in the future Interestingly, exercise and insulin exhibit a synergy, to investigate the supercompensation mechanism since studies have reported that the ability of resist- and its regulation. ance exercise to increase protein synthesis (and eIF2B activity) in skeletal muscle is dependent on permissive levels of insulin (Fedele et al. 2000; Hypertrophy: exercise increases insulin action on Kostyak et al. 2001). Under moderately hypo- protein synthesis insulinemic conditions, resistance exercise has been While skeletal muscle is capable of responding reported to stimulate protein synthesis through acutely to changes in circulating humoral factors a mechanism believed to involve compensatory (insulin) and metabolic demands (exercise), skeletal increases in intramuscular IGF-1 signaling (Farrell muscle also exhibits the ability to adapt to chronic et al. 1999; Fedele et al. 2000). perturbations in such a way that size (hypertrophy), While resistance exercise is known to activate mechanical function (force generation) and meta- enzymes involved in protein synthesis, maximal exercise regulation of insulin action 397

protein accretion is highly dependent on additional tions in GLUT4 transcription, mRNA and protein factors. Exercise is able to improve the anabolic expression in skeletal muscle by up to 65% (Asp effects of insulin in skeletal muscle, and interactions et al. 1995; Kristiansen et al. 1996). Interestingly, with other hormones (e.g. insulin-like growth factor while studies have shown that eccentric contraction I [IGF-I], testosterone) and substrate availability can impair glucose metabolism, protein synthetic (amino acids in particular) determine to a large rates remain elevated. This is perhaps due to select- extent subsequent degrees of hypertrophy (reviewed ive impairments in PI3K signaling and GLUT4 in Tipton & Wolfe 2001). translocation (Fluckey et al. 1999) but normally activated MAPK signaling (Haddad & Addams 2002). In fact, MAPKs are not only activated by Caveat: exercise-induced muscle injury can muscle contraction, but have also been shown to be decrease insulin action on glucose disposal activated by stretch, muscle damage and injury and As described above, exercise is generally associated inflammatory cytokines (discussed later). with improvements in insulin-mediated glucose While acute moderate-intensity exercise and disposal in skeletal muscle. However, acute decreases endurance exercise training are associated with in insulin-stimulated glucose metabolism have increases in insulin-mediated glucose disposal, been documented following resistance exercise, a resistance exercise has been shown to elicit opposite phenomenon that is largely explained by the degree effects. An early study showed that high intensity to which the muscle is damaged. The association resistance exercise was associated with reductions between whole body injury, trauma and skeletal in glucose disposal (Kirwan et al. 1992). Acute resist- muscle insulin resistance has been well documented ance exercise has been shown to result in lower (Black et al. 1982). Exercise resulting in muscle insulin-stimulated glucose uptake compared with damage elicits an inflammatory response, whereby sedentary controls (Fluckey et al. 1999). The de- accumulation of inflammatory cells invades the gree of eccentric contraction involved in resistance affected tissues (Asp et al. 1997). These cells facilitate versus aerobic exercise explains, to a large degree, in the repair, remodeling and removal of damaged observed differences in insulin-stimulated glucose tissue. The release of inflammatory cytokines medi- metabolism. The anaerobic nature of resistance ate many of these repair mechanisms, however, the exercise results in significant decreases in muscle ability of cytokines (i.e. tumor necrosis factor-α) to glycogen. Interestingly, in addition to decreases in induce insulin resistance is also well known. insulin-stimulated glucose transport, eccentric exer- Due to force/area dynamics, muscle injury is cise has been shown to impair insulin-stimulated more likely to occur in response to eccentric (length- incorporation of glucose into glycogen in various ening of contracting muscle) compared with con- muscles (Asp et al. 1997). One study reported that centric (shortening of contracting muscle) muscle eccentric exercise resulted in a 16% decrease in contraction. While a single bout of concentric exer- maximal glycogen synthase activity compared with cise is associated with improvements in skeletal controls (Asp & Richter 1996). These impairments muscle insulin action (Richter et al. 1982, 1989; are often observed in the more glycolytic fast twitch Bogardus et al. 1983), a bout of eccentric exercise was muscles, which are highly recruited during resist- associated with impaired whole body insulin action ance exercise (Asp & Richter 1996). 48 h post-exercise (Kirwan et al. 1992). This effect While muscle damage resulting from a single bout was also observed in isolated muscle, suggesting of resistance exercise can result in decreases in that the insulin resistance is a local phenomenon skeletal muscle insulin action, other data suggest (Asp & Richter 1996). Consistent with differences in that resistance training is associated with improve- insulin-mediated glucose metabolism, concentric ments in insulin-stimulated glucose disposal (Miller exercise is commonly associated with increases in et al. 1994; Yaspelkis et al. 2002). Several factors help GLUT4 protein levels (Richter et al. 1989), while to explain the observed inconsistencies. First, resist- eccentric exercise has been associated with reduc- ance exercise includes both concentric and eccentric 398 chapter 27

components of muscle contraction. Resistance exer- time there has been intense interest in the regulation cise programs that include more eccentric contrac- of these pathways in skeletal muscle. tions would presumably lead to increased muscle injury and decreased insulin-action and vice versa. ERK signaling Furthermore, while an acute bout of eccentric- resistance exercise is associated with impairments Insulin has long been known to be a potent stimu- in glucose metabolism, chronic resistance exercise lator of ERK in skeletal muscle. Additionally, activa- elicits skeletal muscle adaptations that improve in- tion of ERK1/2 signaling has been reported in rat sulin action in skeletal muscle (e.g. GLUT4 expres- skeletal muscle with treadmill running (Goodyear sion). Finally, while untoward effects of an acute et al. 1996; Nader & Esser 2001), in vitro contraction bout of eccentric exercise are due to local effects (Hayashi et al. 1999; Wojtaszewski et al. 1999b, 2000; (inflammatory response), corresponding changes in Ryder et al. 2000; Wretman et al. 2000, 2001), in situ body composition (decreased fat mass, increased contraction (Sherwood et al. 1999; Martineau & muscle mass) associated with chronic resistance Gardiner 2001; Nader & Esser 2001), muscle over- exercise training are associated with improvements load (Carlson et al. 2001) and stretch (Boppart et al. in whole-body insulin action. 2001); in mouse muscle in response to treadmill run- ning (Dufresne et al. 2001); and in human skeletal Insulin and exercise stimulate muscle in response to cycle ergometer exercise MAPK signaling (Aronson et al. 1997a, 1997b; Widegren et al. 1998; Osman et al. 2000) and marathon running (Yu et al. While it is clear that insulin stimulates glucose 2001). Upstream of ERK1/2, both MEK1/2 and Raf1 disposal utilizing the PI3K pathway, and exercise (Aronson et al. 1997a, 1997b; Sherwood et al. 1999) appears to utilize PI3K-independent pathways (e.g. activities are increased by exercise and contraction. AMPK), recent evidence suggests that both exercise Molecules downstream of ERK1/2 that have been and insulin are able to activate common signaling shown to be activated by exercise include RSK2 cascades in skeletal muscle. While insulin has long (Goodyear et al. 1996; Aronson et al. 1997a, 1997b; been known to activate the family of MAPK, a grow- Sherwood et al. 1999; Krook et al. 2000; Osman et al. ing body of evidence suggests that MAPK signal 2000; Ryder et al. 2000; Yu et al. 2001) and the mito- transduction pathways play an important role in gen and stress-activated kinase 1/2 (MSK1/2) exercise signaling in skeletal muscle. The MAPKs (Ryder et al. 2000; Yu et al. 2001). MEK1/2 activa- represent an important family of signal transduc- tion is necessary for ERK1/2 activation since the tion proteins, which are expressed in all eukaryotic MEK1/2 inhibitor PD98059 inhibits muscle con- cells (Fig. 27.3). Four MAPK subgroups have been traction induced ERK1/2 phosphorylation (Hayashi described: (i) the extracellular-signal regulated kin- et al. 1999; Wojtaszewski et al. 1999b, 2000; Ryder ases (ERKs); (ii) c-Jun NH2-terminal kinases (JNKs); et al. 2000) and its downstream substrates RSK2 (iii) p38 MAPK; and (iv) ERK5/big MAP kinase 1 (Hayashi et al. 1999) and MSK1 (Ryder et al. 2000). (BMK1). ERKs are predominantly activated by With one-legged cycling exercise in human sub- growth factors, while JNKs and p38 MAPK are col- jects, activation of ERK1/2 (Aronson et al. 1997b; lectively known as stress-activated protein kinases, Widegren et al. 1998) and its downstream targets and have been implicated in a large number of cel- (Aronson et al. 1997b; Krook et al. 2000) is observed lular responses, including cell proliferation, dif- in the muscle from the exercised leg, but not in the ferentiation, hypertrophy, inflammation, apoptosis, resting leg. These data suggest that stimulation of carbohydrate metabolism and gene transcription ERK1/2 signaling in response to exercise in skeletal (Force & Bonventre 1998; Sweeney et al. 1999; Kyriakis muscle is due to primarily a local, tissue-specific & Avruch 2001). In 1996, it was first reported that phenomenon, rather than a systemic effect. The exercise activates ERK1/2, JNK and p38 signaling in molecules involved in this tissue specific stimula- skeletal muscle (Goodyear et al. 1996). Since that tion remain elusive. exercise regulation of insulin action 399

Exercise systemic factors

Exercise INSULIN autocrine and paracrine factors

GRB2 EXERCISE Signaling messengers

SOS

MAP3K Raf

Fig. 27.3 Exercise and insulin regulation of mitogen-activated MKK4/7 MKK3/6 MEK1/2 protein kinase (MAPK) signaling in skeletal muscle. Insulin potently stimulates ERK1/2 activity through a GRB2/SOS pathway. Exercise also regulates MAPK activity via JNK1/2 p38 ERK1/2 intracellular signaling molecules, as well as systemic and autocrine/paracrine factors. Downstream substrates of MAPKs Protein MNK1/2 MAPKAPK2/3 MSK1/2 RSK include various protein kinases, as kinases well as transcription factors. Transcription Mechanisms underlying the ability ATF-2 CREB CHOP C-Jun C-Fos Elk-1 MEF2 of exercise to induce changes in gene factors expression are relatively unknown; however, modulation via MAPK signaling represent likely candidates. Gene transcription

1998), knee extensions resulting in concentric and JNK signaling eccentric contractions of the quadricep muscles Unlike ERK, JNK is only modestly activated by (Boppart et al. 1999) and marathon running (Boppart insulin in skeletal muscle, and some reports show et al. 2000). In contrast to ERK1/2 signaling, activa- no activation (Aronson et al. 1996). Activation tion of the JNK cascade is sustained during in situ of JNK signaling has been reported in rat skeletal muscle contractions, whereas the activation of muscle in response to in vitro contractions (Boppart the ERK cascade is more rapid and transient. This et al. 2001), in situ contractions (Aronson et al. 1997a; suggests that the upstream proteins that regulate Martineau & Gardiner 2001), treadmill running the JNK signaling cascade are distinct from that of exercise (Goodyear et al. 1996), muscle overload ERK1/2 (Aronson et al. 1997a). Due to the lack of (Carlson et al. 2001) and mechanical stretch (Boppart selective cell permeable inhibitors of JNK signaling et al. 2001). In human subjects, JNK is activated in at this time, little is known about the physiological response to cycle ergometer exercise (Aronson et al. function of JNK in contracting skeletal muscle. 400 chapter 27

Downstream consequences of exercise-mediated p38 signaling MAPK signaling remain obscure; however, recent As is the case with JNK, skeletal muscle p38 data are emerging which implicate p38 in adaptive is weakly activated by insulin, if at all. However, responses to increase lipid metabolism in skeletal p38 is potently activated in rat skeletal muscle in muscle. Peroxisome proliferator activated receptors response to treadmill running (Goodyear et al. 1996; (PPARs) have received much attention as trans- Nader & Esser 2001), in vitro contractions (Ryder criptional regulators of lipid metabolism in various et al. 2000; Somwar et al. 2000; Wretman et al. 2000, tissues. PPARγ is predominantly expressed in adipo- 2001; Boppart et al. 2001), in situ contractions (Nader cytes and has been linked to adipocyte differenti- & Esser 2001), muscle overload (Carlson et al. 2001) ation. PPARα expression is high in skeletal muscle and mechanical stretch (Boppart et al. 2001). It has and has been shown to regulate the expression of also been shown that p38 signaling is increased in numerous genes involved in fat oxidation. Interest- humans during cycle ergometer exercise (Widegren ingly, p38 MAPK has been shown to regulate the et al. 1998) and marathon running (Boppart et al. activity of PPARα (Barger et al. 2001). Further- 2000; Yu et al. 2001). Phosphorylation of MAP- more, p38 has also been shown to regulate PPARγ KAPK-2, a downstream substrate for p38, is associ- coactivator-1 (PGC-1) (Knutti et al. 2001; Puigserver ated with changes in p38 activity during exercise/ et al. 2001). PGC-1 has been shown to regulate mito- contraction and is inhibited by the p38 antagonist chondrial biogenesis in skeletal muscle (Puigserver SB203580 (Krook et al. 2000; Ryder et al. 2000; Yu & Spiegelman 2003). Therefore, the adaptive effects et al. 2001). resulting from exercise training to improve fat util- ization could involve exercise-mediated MAPK activ- ity. Activation of these transcription factors leads Does exercise-mediated MAPK activation to the regulation of gene expression; however, there regulate gene transcription in skeletal muscle? is a paucity of data regarding the involvement of Chronic exercise can result in changes in skeletal skeletal muscle MAPK activation with exercise. muscle structure and function. Endurance training leads to increases in capillary density, increases Effect of MAPK activation on substrate in oxidative capacity and mitochondrial density, metabolism etc. The exact mechanisms regulating the adaptive response of skeletal muscle to repeated bouts of While the regulation of gene transcription is an exercise are unclear; however, MAPKs are known established function of MAPK, the involvement of to activate a number of transcription factors in MAPK in the regulation of cellular substrate meta- response to both insulin and exercise. Activation bolism is unclear. The ERK1/2 signaling cascade of these transcription factors may contribute to was previously proposed to be involved in the the adaptive effects of exercise in skeletal muscle regulation of both glucose transport and glycogen (Widegren et al. 2001). Using various cell systems, metabolism (Merrall et al. 1993). However, in subse- activated MAPKs have been shown to translocate quent studies, a MEK inhibitor that blocks activa- to the nucleus and phosphorylate numerous trans- tion of ERK1/2 had no effect on insulin-stimulated cription factors such as CREB, Elk, ATF2, c-Jun, c- glucose uptake in cultured adipocytes (Haruta et al. fos, c-Myc, AP-1, MEF2, NFAT and CHOP (Han et al. 1995; Tanti et al. 1996) and skeletal muscles (Hayashi 1997; Gomez et al. 2000; Kyriakis & Avruch 2001; et al. 1999; Wojtaszewski et al. 1999b). Thus, there is Hazzalin & Mahadevan 2002). Not only can MAPKs substantial evidence that ERK1/2 signaling is not regulate gene transcription by direct interaction with involved in the acute regulation of glucose uptake transcription factors, they can also activate other in response to insulin treatment or contraction in downstream substrates such as RSK2, MAPKAPK- skeletal muscle. 2/3 and MSK1/2 that can translocate to the nucleus The ERK1/2 signaling cascade has also been and phosphorylate numerous transcription factors. proposed to regulate insulin-stimulated activation exercise regulation of insulin action 401

of glycogen synthase. This hypothesis was based on robustly activate JNK activity, we hypothesized the finding that RSK2 could phosphorylate and that JNK could be involved in the regulation of inactivate GSK3 and phosphorylate and activate the contraction-stimulated glycogen synthase activ- glycogen-bound form of protein phosphatase-1 ity. Overexpression of wild type JNK1 in skel- (PP1-G) in vitro, two reported regulators of insulin- etal muscle in vivo dramatically increased basal and stimulated glycogen synthase activity (Dent et al. contraction-stimulated JNK activity. However, this 1990; Sutherland et al. 1993). However, inhibition of increase in JNK activity did not enhance basal and ERK signaling by a MEK inhibitor resulted in no contraction-stimulated glycogen synthase activ- inhibition of insulin-stimulated glycogen synthase ity in mouse skeletal muscle, suggesting that JNK activity (Lazar et al. 1995). More direct evidence came is not involved in the regulation of contraction- from a study using RSK2 knockout mice which stimulated glycogen synthase activity (Fujii et al. revealed that RSK2 is not necessary for insulin- 2001). Whether JNK is involved in glucose transport stimulated glycogen synthase activation, and in regulation in muscle is not known and will be an fact, the RSK2 null mice had greater increases in important area for future investigation. muscle glycogen synthase activity following insulin Recent studies have provided evidence that p38 is treatment (Dufresne et al. 2001). However, these involved in the regulation of contraction-stimulated studies do not rule out a role for RSK2 in the regula- glucose uptake in skeletal muscle. Somwar and tion of glycogen metabolism in the basal state, as colleagues demonstrated that p38 activity and the knockout mice had lower levels of muscle glucose uptake were increased in isolated extensor glycogen. digitorum longus (EDL) muscles with contraction A growing body of literature is emerging implic- in vitro (Somwar et al. 2000), and the p38 antagon- ating MAPK pathways in the regulation of lipid ist SB203580 abolished the activation of p38 and metabolism. Recently, a study has suggested that reduced contraction-stimulated glucose uptake by ERK signaling is associated with the plasma mem- 40–50%. However, p38 inhibitors are known to have brane fatty acid transporter FAT/CD36 (Todd & numerous non-specific effects and it is still unclear Turcotte 2003). Incubation of isolated muscle with whether the attenuation of glucose uptake was due PD98059 significantly attenuated the contraction- to inhibition of contraction-stimulated p38 activity induced increase in fatty acid uptake in skeletal or an indirect effect of the compound on other muscle. ERK has also been recently suggested signaling intermediates (Somwar et al. 2000). We to play a role in the activation of HSL in skeletal have recently showed that p38γ, the isoform highly muscle watt (Donsmark et al. 2003; Langfort et al. abundant in skeletal muscle, is a negative regulator 2003; Watt et al. 2003). These data suggest that ERK of GLUT4 expression and contraction-stimulated activation is involved in both the uptake of fatty glucose uptake (Ho et al. 2004). acids and hydrolysis of triglyceride in skeletal The impact of exercise on insulin-stimulated muscle. Whether chronic exercise improves insulin- MAPK activity remains to be uncovered. Interest- stimulated ERK activity in skeletal muscle remains ingly, cytokines that are released in response to to be elucidated. muscle damage and implicated in the negative re- There have been few studies examining the gulation of insulin-stimulated glucose metabolism effects of JNK activation in the regulation of carbo- (e.g. tumor necrosis factor-α) are also potent stimu- hydrate metabolism in skeletal muscle. One study lators of JNK and p38. Additionally, both JNK and demonstrated that activation of JNK by anisomycin, p38 activity have been shown to induce impair- a protein synthesis inhibitor, mimics insulin’s action ments in insulin-stimulated glucose transport in to stimulate glycogen synthesis in mouse skeletal skeletal muscle. However, because exercise results muscle in vivo (Moxham et al. 1996). Based on their in significant increases in glucose transport in para- findings, this group concluded that JNK stimulates llel with MAPK activation, the exact role of MAPK glycogen synthase activity through the regulation in contracting skeletal muscle warrants further of RSK3 and GSK3. Since exercise and contraction investigation. 402 chapter 27

Clinical implications and protein expression. Furthermore, epidemiolo- gical studies have determined that regular physical Throughout this chapter, we have discussed vari- exercise can reduce the risk of developing type 2 ous effects of exercise on insulin action in skeletal diabetes (Helmrich et al. 1991; Manson et al. 1991, muscle. Exercise clearly has the ability to improve 1992). glucose transport, glycogen synthesis and protein metabolism, as well as stimulate adaptive changes Summary through gene transcription. While the majority of studies reviewed in this chapter involve healthy Significant advances have been made in recent individuals, the impact of exercise on intermediary years in the elucidation of mechanisms by which metabolism also transcends to individuals exhibit- exercise increases insulin action in skeletal muscle. ing metabolic complications. It has long been recog- Discoveries have been made showing that both nized that physical exercise has important benefits exercise and insulin stimulate increases in glucose for people with obesity and diabetes (Trovati et al. transport, glycogen metabolism, protein synthesis 1984; Helmrich et al. 1991). However, the transient and long-term adaptations (e.g. hypertrophy). nature of post-exercise insulin sensitivity limits Interestingly, these effects are elicited through both the beneficial impact of physical activity. Chronic common and distinct signaling pathways. Further- exercise, on the other hand, results in multiple more, additive effects of exercise and insulin in physical and metabolic adaptations. Exercise train- the regulation of intermediary metabolism and ing improves glucose tolerance and insulin action in adaptive responses have a widespread impact in insulin-resistant humans (Hughes et al. 1993) and both health and disease. While exercise training can type 2 diabetic patients (Dela et al. 1994). The result in adaptations to improve performance, improvements in insulin sensitivity seem to be chronic physical activity can also prevent or reverse multifaceted, including alterations in body composi- metabolic defects observed in conditions such as tion, plasma lipid profiles, intracellular signaling type 2 diabetes.

References

Arad, M., Benson, D.W., Perez-Atayde, (1998) Exercise stimulates c-Jun NH2 Asp, S., Watkinson, A., Oakes, N.D. & A.R. et al. (2002) Constitutively active kinase activity and c-Jun transcriptional Kraegen, E.W. (1997) Prior eccentric AMP kinase mutations cause glycogen activity in human skeletal muscle. contractions impair maximal insulin storage disease mimicking hypertrophic Biochemical and Biophysical Research action on muscle glucose uptake in cardiomyopathy. Journal of Clinical Communications 251, 106–110. the conscious rat. Journal of Applied Investigation 109, 357–362. Aschenbach, W.G., Suzuki, Y., Breeden, K. Physiology 82, 1327–1332. Aronson, D., Fielding, R.A., Violan, A., et al. (2001) The muscle-specific protein Baar, K. & Esser, K. (1999) Phosphorylation Dufresne, S.D. & Goodyear, L.J. (1996) phosphatase PP1G/RGL(GM) is of p70(S6k) correlates with increased Exercise activates the MAP kinase essential for activation of glycogen skeletal muscle mass following signaling cascade in human skeletal synthase by exercise. Journal of Biological resistance exercise. American Journal of muscle. Diabetes 45, 103A. Chemistry 276, 39 959–39 967. Physiology 276, C120–C127. Aronson, D., Dufresne, S.D. & Goodyear, Aschenbach, W.G., Hirshman, M.F., Fujii, Balon, T.W., Zorzano, A., Treadway, J.L., L.J. (1997a) Contractile activity N. et al. (2002) Effect of AICAR treatment Goodman, M.N. & Ruderman, N.B. stimulates the c-Jun NH2-terminal on glycogen metabolism in skeletal (1990) Effect of insulin on protein kinase pathway in rat skeletal muscle. muscle. Diabetes 51, 567–573. synthesis and degradation in skeletal Journal of Biological Chemistry 272, Asp, S. & Richter, E.A. (1996) Decreased muscle after exercise. American Journal 25 636–25 640. insulin action on muscle glucose of Physiology 258, E92–E97. Aronson, D., Violan, M.A., Dufresne, S.D. transport after eccentric exercise in Barger, P.M., Browning, A.C., Garner, et al. (1997b) Exercise stimulates the rats. Journal of Applied Physiology 81, A.N. & Kelly, D.P. (2001) p38 mitogen- mitogen-activated protein kinase 1924–1928. activated protein kinase activates pathway in human skeletal muscle. Asp, S., Daugaard, J.R. & Richter, E.A. peroxisome proliferator-activated Journal of Clinical Investigation 99, (1995) Eccentric exercise decreases receptor α. A potential role in the 1251–1257. glucose transporter GLUT4 protein in cardiac metabolic stress response. Aronson, D., Boppart, M.D., Dufresne, human skeletal muscle. Journal of Journal of Biological Chemistry 276, S.D., Fielding, R.A. & Goodyear, L.J. Physiology 482, 705–712. 44 495–44 501. exercise regulation of insulin action 403

Bergstrom, J. & Hultman, E. (1966) Muscle (2001) Static stretch increases c-Jun inhibits glycogen synthase kinase-3 glycogen synthesis after exercise: an NH2-terminal kinase activity and p38 and activates glycogen synthase by enhancing factor localized to the muscle phosphorylation in rat skeletal muscle. rapamycin-insensitive pathways in cells in man. Nature 210, 309–310. American Journal of Physiology 280, skeletal muscle and adipose tissue. FEBS Bergstrom, J., Hultman, E. & Roch- C352–C358. Letters 406, 211–215. Norlund, A.E. (1972) Muscle glycogen Brady, M.J., Bourbonais, F.J. & Saltiel, Dela, F., Ploug, T., Handberg, A. et al. synthetase in normal subjects. Basal A.R. (1998) The activation of glycogen (1994) Physical training increases muscle values, effect of glycogen depletion by synthase by insulin switches from kinase GLUT4 protein and mRNA in patients exercise and of a carbohydrate-rich diet inhibition to phosphatase activation with NIDDM. Diabetes 43, 862–865. following exercise. Scandinavian Journal during adipogenesis in 3T3-L1 cells. Dent, P., Lavoinne, A., Nakielny, S. et al. of Clinical Laboratory Investigation 29, Journal of Biological Chemistry 273, (1990) The molecular mechanism by 231–236. 14 063–14 066. which insulin stimulates glycogen Beyer, A., Kitzerow, A., Crute, B. et al. Brozinick, J.T., Jr. & Birnbaum, M.J. (1998) synthesis in mammalian skeletal muscle. (2000) Muscle phosphorylase kinase is Insulin, but not contraction, activates Nature 348, 302–308. not a substrate of AMP-activated protein Akt/PKB in isolated rat skeletal muscle. Devlin, J.T. & Horton, E.S. (1985) Effects of kinase. Biological Chemistry 381, 457–461. Journal of Biological Chemistry 273, prior high-intensity exercise on glucose Bjorntorp, P., Fahlen, M., Grimby, G. 14679–14682. metabolism in normal and insulin- et al. (1972) Carbohydrate and lipid Buhl, E.S., Jessen, N., Schmitz, O. et al. resistant men. Diabetes 34, 973–979. metabolism in middle-aged physically (2001) Chronic treatment with Devlin, J.T., Hirshman, M.F., Horton, E.S. well trained men. Metabolism 21, 5-aminoimidazole-4-carboxamide-1- & Horton, E.D. (1987) Enhanced 1037–1044. β-d-ribofuranoside increases insulin- peripheral and splanchnic insulin Black, P.R., Brooks, D.C., Bessey, P.Q., stimulated glucose uptake and GLUT4 sensitivity in NIDDM men after single Wolfe, R.R. & Wilmore, D.W. (1982) translocation in rat skeletal muscles in a bout of exercise. Diabetes 36, 434–439. Mechanisms of insulin resistance fiber type-specific manner. Diabetes 50, Dohm, G.L., Kasperek, G.J., Tapscott, E.B. following injury. Annals of Surgery 196, 12–17. & Beecher, G.R. (1980) Effect of exercise 420–435. Burstein, R., Polychronakos, C., Toews, on synthesis and degradation of muscle Bogardus, C., Thuillex, P., Ravussin, E. C.J. et al. (1985) Acute reversal of the protein. Biochemical Journal 188, 255–262. et al. (1983) Effect of muscle glycogen enhanced insulin action in trained Donsmark, M., Langfort, J., Holm, C., depletion on in vivo insulin action in athletes. Diabetes 34, 756–760. Ploug, T. & Galbo, H. (2003) man. Journal of Clinical Investigation 72, Carling, D. & Hardie, D.G. (1989) The Contractions activate hormone-sensitive 1605–1610. substrate and sequence specificity of lipase in rat muscle by protein kinase C Bolster, D.R., Crozier, S.J., Kimball, S.R. the AMP-activated protein kinase. and mitogen-activated protein kinase. & Jefferson, L.S. (2002) AMP-activated Phosphorylation of glycogen synthase Journal of Physiology 550, 845–854. protein kinase suppresses protein and phosphorylase kinase. Biochimica et Dufresne, S.D., Bjorbaek, C., El Haschimi, synthesis in rat skeletal muscle through Biophysica Acta 1012, 81–86. K. et al. (2001) Altered extracellular down-regulated mammalian target of Carlson, C.J., Fan, Z., Gordon, S.E. & signal-regulated kinase signaling rapamycin (mTOR) signaling. Journal of Booth, F.W. (2001) Time course of the and glycogen metabolism in skeletal Biological Chemistry 277, 23 977–23 980. MAPK and PI3-kinase response within muscle from p90 ribosomal S6 kinase 2 Bonen, A., Tan, M.H. & Watson-Wright, 24 h of skeletal muscle overload. Journal knockout mice. Molecular and Cellular W.M. (1984) Effects of exercise on insulin of Applied Physiology 91, 2079–2087. Biology 21, 81–87. binding and glucose metabolism in Chibalin, A.V., Yu, M., Ryder, J.W. et al. Elbrink, J. & Phipps, B.A. (1980) Studies muscle. Canadian Journal of Physiology (2000) Exercise-induced changes in on the persistence of enhanced and Pharmacology 62, 1500–1504. expression and activity of proteins monosaccharide transport in rat skeletal Booth, F.W. & Thomason, R.B. (1991) involved in insulin signal transduction muscle following the cessation of the Molecular and cellular adaptation in in skeletal muscle: differential effects on initial stimulus. Cell Calcium 1, 349–358. response to exercise: perspectives of insulin-receptor substrates 1 and 2. Farrell, P.A., Fedele, M.J., Vary, T.C. et al. various models. Physiology Reviews 71, Proceedings of the National Academy of (1999) Regulation of protein synthesis 541–585. Sciences of the United States of America 97, after acute resistance exercise in diabetic Boppart, M.D., Aronson, D., Gibson, L. 38–43. rats. American Journal of Physiology 276, et al. (1999) Eccentric exercise markedly Conlee, R.K., Hickson, R.C., Winder, E721–E727. increases c-Jun NH(2)-terminal kinase W.W., Hagberg, J.M. & Holloszy, J.O. Farrell, P.A., Hernandez, J.M., Fedele, M.J. activity in human skeletal muscle. (1978) Regulation of glycogen et al. (2000) Eukaryotic initiation factors Journal of Applied Physiology 87, resynthesis in muscles of rats following and protein synthesis after resistance 1668–1673. exercise. American Journal of Physiology exercise in rats. Journal of Applied Boppart, M.D., Asp, S., Wojtaszewski, J.F. 235, R145–R150. Physiology 88, 1036–1042. et al. (2000) Marathon running Cross, D.A.E., Alessi, D.R., Cohen, P., Fedele, M.J., Hernandez, J.M., Lang, C.H. transiently increases c-Jun NH2- Andjelkovich, M. & Hemmings, B.A. et al. (2000) Severe diabetes prohibits terminal kinase and p38 activities in (1995) Inhibition of glycogen synthase elevations in muscle protein synthesis human skeletal muscle. Journal of kinase-3 by insulin mediated by protein after acute resistance exercise in rats. Physiology 526, 663–669. kinase B. Nature 378, 785–789. Journal of Applied Physiology 88, 102–108. Boppart, M.D., Hirshman, M.F., Sakamoto, Cross, D.A.E., Watt, P.W., Shaw, M. et al. Fell, R.D., Terblanche, S.E., Ivy, J.L., K., Fielding, R.A. & Goodyear, L.J. (1997) Insulin activates protein kinase B, Young, J.C. & Holloszy, J.O. (1982) Effect 404 chapter 27

of muscle glycogen content on glucose Han, J., Jiang, Y., Li, Z., Kravchenko, V.V. Hollenbeck, C.B., Haskell, W., Rosenthal, uptake following exercise. Journal of & Ulevitch, R.J. (1997) Activation of the M. & Reaven, G.M. (1984) Effect of Applied Physiology 52, 434–437. transcription factor MEF2C by the MAP habitual physical activity on regulation Fluckey, J.D., Vary, T.C., Jefferson, L.S. kinase p38 in inflammation. Nature 386, of insulin-stimulated glucose disposal in & Farrell, P.A. (1996) Augmented 296–299. older males. Journal of the American insulin action on rates of protein Hardie, D.G., Carling, D. & Carlson, M. Geriatric Society 33, 273–277. synthesis after resistance exercise in (1998) The AMP-activated/SNF1 Holloszy, J.O. & Narahara, H.T. rats. American Journal of Physiology 270, protein kinase subfamily: metabolic (1965) Changes in permeability to E313–E319. sensors of the eukaryotic cell. Annual 3-methylglucose associated with Fluckey, J.D., Ploug, T. & Galbo, H. (1999) Reviews of Biochemistry 67, 821–855. contraction of isolated frog muscle. Attenuated insulin action on glucose Haruta, T., Morris, A.J., Rose, D.W. et al. Journal of Biological Chemistry 240, uptake and transport in muscle (1995) Insulin-stimulated GLUT4 3493–3500. following resistance exercise in rats. translocation is mediated by a divergent Horman, S., Browne, G., Krause, U. Acta Physiologica Scandinavica 167, intracellular signaling pathway. et al. (2002) Activation of AMP- 77–82. Journal of Biological Chemistry 270, activated protein kinase leads to the Force, T. & Bonventre, J.V. (1998) Growth 27 991–27 994. phosphorylation of elongation factor 2 factors and mitogen-activated protein Hawley, S.A., Boudeau, J., Reid, J.L. et al. and an inhibition of protein synthesis. kinases. Hypertension 31, 152–161. (2003) Complexes between the LKB1 Current Biology 12, 1419–1423. Fujii, N., Boppart, M.D., Dufresne, S.D. tumor suppressor, STRADα/β and Host, H.H., Hansen, P.A., Nolte, L.A., et al. (2001) Overexpression of JNK in MO25α/β are upstream kinases in the Chen, M.M. & Holloszy, J.O. (1998) skeletal muscle does not alter glycogen AMP-activated protein kinase cascade. Glycogen supercompensation masks synthase activity. Diabetes 50, A276. Journal of Biology 2, 28.1–28.16. the effect of a traininginduced increase Garcia-Roves, P.M., Han, D.H., Song, Z. Hayashi, T., Hirshman, M.F., Kurth, E.J., in GLUT-4 on muscle glucose transport. et al. (2003) Prevention of glycogen Winder, W.W. & Goodyear, L.J. (1998) Journal of Applied Physiology 85, supercompensation prolongs the Evidence for 5′ AMP-activated protein 133–138. increase in muscle GLUT4 after exercise. kinase mediation of the effect of muscle Houmard, J.A., Shaw, C.D., Hickey, M.S. American Journal of Physiology 285, contraction on glucose transport. & Tanner, C.J. (1999) Effect of short-term E729–E736. Diabetes 47, 1369–1373. exercise training on insulin-stimulated Gomez, D.A., Martinez-Martinez, S., Hayashi, T., Hirshman, M.F., Dufresne, PI 3-kinase activity in human skeletal Maldonado, J.L., Ortega-Perez, I. & S.D. & Goodyear, L.J. (1999) Skeletal muscle. American Journal of Physiology Redondo, J.M. (2000) A role for the p38 muscle contractile activity in vitro 277, E1055–E1060. MAP kinase pathway in the nuclear stimulates mitogen-activated protein Howlett, K.F., Sakamoto, K., Hirshman, shuttling of NFATp. Journal of Biological kinase signaling. American Journal of M.F. et al. (2002) Insulin signaling after Chemistry 275, 13 872–13 878. Physiology 277, C701–C707. exercise in insulin receptor substrate-2- Goodyear, L.J., Giorgino, F., Balon, T.W., Hayashi, T., Hirshman, M.F., Fujii, N. deficient mice. Diabetes 51, 479–483. Condorelli, G. & Smith, R.J. (1995) et al. (2000) Metabolic stress and altered Hughes, V.A., Fiatarone, M.A., Fielding, Effects of contractile activity on tyrosine glucose transport: Activation of AMP- R.A. et al. (1993) Exercise increases phophoproteins and PI 3-kinase activity activated protein kinase as a unifying muscle GLUT-4 levels and insulin action in rat skeletal muscle. American Journal of coupling mechanism. Diabetes 49, in subjects with impaired glucose Physiology 268, E987–E995. 527–531. tolerance. American Journal of Physiology Goodyear, L.J., Chang, P.-Y., Sherwood, Hazzalin, C.A. & Mahadevan, L.C. (2002) 264, E855–E862. D.J., Dufresne, S.D. & Moller, D.E. MAPK-regulated transcription: A Ihlemann, J., Ploug, T., Hellsten, Y. & (1996) Effects of exercise and insulin continuously variable gene switch. Galbo, H. (1999) Effect of tension on on mitogen-activated protein kinase Molecular and Cellular Biology 3, 30–40. contraction-induced glucose transport signaling pathways in rat skeletal Heath, G.W., Gavin I.I., Hinderliter, J.M. in rat skeletal muscle. American Journal of muscle. American Journal of Physiology et al. (1983) Effects of exercise and lack of Physiology 277, E208–E214. 271, E403–E408. exercise on glucose tolerance and insulin Ivy, J.L. & Holloszy, J.O. (1981) Persistent Greiwe, J.S., Hickner, R.C., Hansen, P.A. sensitivity. Journal of Applied Physiology increase in glucose uptake by rat skeletal et al. (1999) Effects of endurance 55, 628–634. muscle following exercise. American exercise training on muscle glycogen Helmrich, S.P., Ragland, D.R., Leung, R.W. Journal of Physiology 241, C200–C203. accumulation in humans. Journal of & Paffenbarger, R.S. (1991) Physical Johansen, K. & Munch, O. (1979) The Applied Physiology 87, 222–226. activity and reduced occurrence of non- relationship between maximal oxygen Haddad, F. & Adams, G.R. (2002) Selected insulin-dependant diabetes mellitus. uptake and glucose tolerance/insulin contribution: acute cellular and New England Journal of Medicine 325, response ratio in normal young men. molecular responses to resistance 147–152. Hormone and Metabolic Research 11, exercise. Journal of Applied Physiology 93, Ho, R.C., Alcazar, O., Fujii, N., Hirshman, 424–427. 394–403. M.F. & Goodyear, L.J. (2004) p38 γ Jorgensen, S.B., Viollet, B., Andreelli, F. α Hamilton, S.R., Stapleton, D., O’Donnell, MAPK regulation of glucose transporter et al. (2004) Knockout of the 2 but not α ′ J.B. et al. (2001) An activating mutation expression and glucose uptake in L6 1 5 -AMP-activated protein kinase γ in the 1 subunit of the AMP-activated myotubes and mouse skeletal muscle. isoform abolishes 5-aminoimidazole-4- protein kinase. FEBS Letters 500, American Journal of Physiology 286, carboxamide-1-β-4-ribofuranosidebut 163–168. R342–R349. not contraction-induced glucose uptake exercise regulation of insulin action 405

in skeletal muscle. Journal of Biological kinase and acetyl-CoA carboxylase by kinase/Akt-1 pathway. European Journal Chemistry 279, 1070–1079. AMP-activated protein kinase and of Endocrinology 147, 149–157. Kawanaka, K., Nolte, L.A., Han, D.H., protein phosphatases in isolated Lund, S., Pryor, P.R., Ostergaard, S. et al. Hansen, P.A. & Holloszy, J.O. (2000) hepatocytes. European Journal of (1998) Evidence against protein kinase Mechanisms underlying impaired Biochemistry 269, 3751–3759. B as a mediator of contraction-induced GLUT-4 translocation in glycogen- Kristiansen, S., Asp, S. & Richter, E.A. glucose transport and GLUT4 supercompensated muscles of exercised (1996) Decreased muscle GLUT-4 and translocation in rat skeletal muscle. rats. American Journal of Physiology 279, contraction-induced glucose transport FEBS Letters 425, 472–474. E1311–E1318. after eccentric contractions. American Manson, J.E., Rimm, E.B., Stampfer, M.J. Kimball, S.R., Farrell, P.A. & Jefferson, L.S. Journal of Physiology 271, R477–R482. et al. (1991) Physical activity and (2002) Invited review: role of insulin in Krook, A., Widegren, U., Jiang, X.J. et al. incidence of non-insulin dependent translational control of protein synthesis (2000) Effects of exercise on mitogen diabetes mellitus in women. Lancet 338, in skeletal muscle by amino acids or activated protein kinase. American 774–778. exercise. Journal of Applied Physiology 93, Journal of Physiology 279, R1716–R1721. Manson, J.E., Nathan, D.M., Krolewski, 1168–1180. Kyriakis, J.M. & Avruch, J. (2001) A.S. et al. (1992) A prospective study Kimura, N., Tokunaga, C., Dalal, S. et al. Mammalian mitogen-activated protein of exercise and incidence of diabetes (2003) A possible linkage between kinase signal transduction pathways among US male physicians. Journal of the AMP-activated protein kinase (AMPK) activated by stress and inflammation. American Medical Association 268, 63–67. and mammalian target of rapamycin Physiology Reviews 81, 807–869. Markuns, J.F., Wojtaszewski, J.F.P. & (mTOR) signalling pathway. Genes to Langfort, J., Donsmark, M., Ploug, T., Goodyear, L.J. (1999) Insulin and Cells 8, 65–79. Holm, C. & Galbo, H. (2003) Hormone- exercise decrease glycogen synthase King, D.S., Dalsky, G.P., Staten, M.A. et al. sensitive lipase in skeletal muscle: kinase-3 activity by different (1987) Insulin action and secretion in regulatory mechanisms. Acta mechanisms in rat skeletal muscle. endurance trained and untrained Physiologica Scandinavica 178, 397–403. Journal of Biological Chemistry 274, humans. Journal of Applied Physiology 63, Lazar, D.F., Wiese, R.J., Brady, M.J. et al. 24 896–24 900. 2247–2252. (1995) Mitogen-activated protein kinase Martineau, L.C. & Gardiner, P.F. King, D.S., Dalsky, G.P., Cutler, W.E. kinase inhibition does not block the (2001) Insight into skeletal muscle et al. (1988) Effects of lack of exercise on stimulation of glucose utilization by mechanotransduction: MAPK activation insulin secretion and action in trained insulin. Journal of Biological Chemistry is quantitatively related to tension. subjects. American Journal of Physiology 270, 20 801–20 807. Journal of Applied Physiology 91, 254, E537–E542. LeBlanc, J., Nadeau, A., Boulay, M. & 693–702. Kirwan, J.P., Hickner, R.C., Yarasheski, Rousseau-Migneron, S. (1979) Effects Merrall, N.W., Plevin, R.J., Stokoe, D. et al. K.E. et al. (1992) Eccentric exercise of physical training and adiposity on (1993) Mitogen-activated protein kinase induces transient insulin resistance in glucose metabolism and 125I-insulin (MAP kinase), MAP kinase kinase and healthy individuals. Journal of Applied binding. Journal of Applied Physiology 46, c-Mos stimulate glucose transport in Physiology 72, 2197–2202. 235–239. Xenopus oocytes. Biochemical Journal 295, Kirwan, J.P., Del Aguila, L.F., Hernandez, LeBlanc, J., Nadeau, A., Richard, D. & 351–355. J.M. et al. (2000) Regular exercise Tremblay, A. (1981) Studies on the Mikines, K.J., Sonne, B., Farrell, P.A., enhances insulin activation of IRS-1- sparing effect of exercise on insulin Tronier, B. & Galbo, H. (1988) Effect of associated PI3-kinase in human skeletal requirements in human subjects. physical exercise on sensitivity and muscle. Journal of Applied Physiology 88, Metabolism 30, 1119–1124. responsiveness to insulin in humans. 797–803. Liu, J. & Brautigan, D.L. (2000) Insulin- American Journal of Physiology 254, Knutti, D., Kressler, D. & Kralli, A. (2001) stimulated phosphorylation of the E248–E259. Regulation of the transcriptional protein phosphatase-1 striated muscle Mikines, K.J., Sonne, B., Farrell, P.A., coactivator PGC-1 via MAPK-sensitive glycogen-targeting subunit and Tronier, B. & Galbo, H. (1989) Effect interaction with a repressor. Proceedings activation of glycogen synthase. Journal of training on the dose–response of the National Academy of Sciences of the of Biological Chemistry 275, 15 940–15 947. relationship for insulin action in men. United States of America 98, 9713–9718. Lizcano, J.M., Goransson, O., Toth, R. et al. Journal of Applied Physiology 66, Kolterman, O.G., Insel, J., Seakow, M. & (2004) LKB1 is a master kinase that 695–703. Olefsky, J.M. (1980) Mechanisms of activates 13 kinases of the AMPK Milan, D., Jeon, J.T., Looft, C. et al. (2000) insulin resistance in human obesity: subfamily, including MARK/PAR-1. A mutation in PRKAG3 associated with evidence for receptor and post-receptor EMBO Journal 23, 833–843. excess glycogen content in pig skeletal defects. Journal of Clinical Investigation 65, Lohman, D., Liebold, F., Heilmann, W., muscle. Science 288, 1248–1251. 1272–1284. Singer, J. & Pohl, A. (1978) Diminished Miller, J.P., Pratley, R.E., Goldberg, A.P. Kostyak, J.C., Kimball, S.R., Jefferson, L.S. insulin response in highly trained et al. (1994) Strength training increases & Farrell, P.A. (2001) Severe diabetes athletes. Metabolism Clinical and insulin action in healthy 50- to 65-yr-old inhibits resistance exercise-induced Experimental 27, 521–524. men. Journal of Applied Physiology 77, increase in eukaryotic initiation factor 2B Luciano, E., Carneiro, E.M., Carvalho, 1122–1127. activity. Journal of Applied Physiology 91, C.R. et al. (2002) Endurance training Moxham, C.M., Tabrizchi, A., Davis, R.J. 79–84. improves responsiveness to insulin and & Malbon, C.C. (1996) Jun N-terminal Krause, U., Bertrand, L. & Hue, L. (2002) modulates insulin signal transduction kinase mediates activation of skeletal Control of p70 ribosomal protein S6 through the phosphatidylinositol 3- muscle glycogen synthase by insulin in 406 chapter 27

vivo. Journal of Biological Chemistry 271, M.N. & Ruderman, N.B. (1982) Muscle GLUT4 translocation. Journal of Biological 30 765–30 773. glucose metabolism following exercise Chemistry 278, 14 599–14 602. Mu, J., Brozinick, J.T., Jr., Valladares, O., in the rat: increased sensitivity to Sasson, S., Edelson, D. & Cerasi, E. (1987) Bucan, M. & Birnbaum, M.J. (2001) A insulin. Journal of Clinical Investigation 69, In vitro autoregulation of glucose role for AMP-activated protein kinase 785–793. utilization in rat soleus muscle. Diabetes in contraction- and hypoxia-regulated Richter, E.A., Garreto, L.P., Goodman, 36, 1041–1046. glucose transport in skeletal muscle. M.N., Ruderman, N.B. & Garetto, L.P. Seals, D.R., Hagberg, J.M., Allen, W.K. Molecular Cell 7, 1085–1094. (1984) Enhanced muscle glucose et al. (1984) Glucose tolerance in young Nader, G.A. & Esser, K.A. (2001) metabolism after exercise: modulation and older athletes and sedentary men. Intracellular signaling specificity in by local factors. American Journal of Journal of Applied Physiology 56, skeletal muscle in response to different Physiology 246, E476–E482. 1521–1525. modes of exercise. Journal of Applied Richter, E.A., Mikines, K.J., Galbo, H. & Shah, O.J., Anthony, J.C., Kimball, S.R. & Physiology 90, 1936–1942. Kiens, B. (1989) Effect of exercise on Jefferson, L.S. (2000) 4E-BP1 and S6K1: Nakatani, A., Han, D.H., Hansen, P.A. insulin action in human skeletal muscle. translational integration sites for et al. (1997) Effect of endurance exercise Journal of Applied Physiology 66, nutritional and hormonal information training on muscle glycogen 876–885. in muscle. American Journal of Physiology supercompensation in rats. Journal of Rosenthal, M., Haskell, W.L., Solomon, R., 279, E715–E729. Applied Physiology 82, 711–715. Widstrom, A. & Reaven, G.M. (1983) Sherwood, D.J., Dufresne, S.D., Markuns, Oshida, Y., Yamanouchi, K., Hayamizu, S. Demonstration of a relationship J.F. et al. (1999) Differential regulation of & Sato, Y. (1989) Long-term mild jogging between level of physical training and MAP kinase, p70 S6 kinase and Akt by increases insulin action despite no insulin-stimulated glucose utilization contraction and insulin in rat skeletal o influence on body mass index or V 2max. in normal humans. Diabetes 32, muscle. American Journal of Physiology Journal of Applied Physiology 66, 408–411. 276, E870–E878. 2206–2210. Ruderman, N.B., Houghton, C.R.S. & Skurat, A.V., Wang, Y. & Roach, P.J. (1994) Osman, A.A., Pendergrass, M., Koval, J. Hems, R. (1971) Evaluation of the Rabbit skeletal muscle glycogen et al. (2000) Regulation of MAP kinase isolated perfused rat hindquarter for the synthase expressed in COS cells. pathway activity in vivo in human study of muscle metabolism. Biochemical Identification of regulatory skeletal muscle. American Journal of Journal 124, 639–651. phosphorylation sites. Journal of Physiology 278, E992–E999. Russell, R.R., III, Bergeron, R., Shulman, Biological Chemistry 269, 25 534–25 542. Ossipova, O., Bardeesy, N., DePinho, G.I. & Young, L.H. (1999) Translocation Somwar, R., Perreault, M., Kapur, S. et al. R.A. & Green, J.B. (2003) LKB1 (XEEK1) of myocardial GLUT-4 and increased (2000) Activation of p38 mitogen- regulates Wnt signalling in vertebrate glucose uptake through activation of activated protein kinase α and by insulin development. Nature Cell Biology 5, AMPK by AICAR. American Journal of and contraction in rat skeletal muscle: 889–894. Physiology 277, H643–H649. potential role in the stimulation of Price, T.B., Laurent, D. & Petersen, K.F. Ryder, J.W., Fahlman, R., Wallberg- glucose transport. Diabetes 49, (2003) 13C/31P NMR studies on the role Henriksson, H. et al. (2000) Effect of 1794–1800. of glucose transport/phosphorylation contraction on mitogen-activated Sutherland, C., Leighton, I.A. & Cohen, P. in human glycogen supercompensation. protein kinase signal transduction in (1993) Inactivation of glycogen synthase International Journal of Sports Medicine 24, skeletal muscle. Journal of Biological kinase-3β by phosphorylation: new 238–244. Chemistry 275, 1457–1462. kinase connections in insulin and Pruett, E.D.R. & Oseid, S. (1970) Effect of Sakamoto, K. & Goodyear, L.J. (2002) growth-factor signalling. Biochemical exercise on glucose and insulin response Invited review: intracellular signaling Journal 296, 15–19. to glucose infusion. Scandinavian Journal in contracting skeletal muscle. Journal of Suzuki, Y., Lanner, C., Kim, J.H. et al. of Clinical and Laboratory Investigation 26, Applied Physiology 93, 369–383. (2001) Insulin control of glycogen 277–285. Sakamoto, K., Hirshman, M.F., metabolism in knockout mice lacking Puigserver, P. & Spiegelman, B.M. (2003) Aschenbach, W.G. & Goodyear, L.J. the muscle-specific protein phosphatase

Peroxisome proliferator-activated (2002) Contraction regulation of Akt in PP1G/RGL. Molecular and Cellular Biology receptor-γ coactivator 1α (PGC-1α): rat skeletal muscle. Journal of Biological 21, 2683–2694. transcriptional coactivator and Chemistry 277, 11 910–11 917. Sweeney, G., Somwar, R., Ramlal, T. et al. metabolic regulator. Endocrinology Sakamoto, K., Aschenbach, W.G., (1999) An inhibitor of p38 mitogen- Reviews 24, 78–90. Hirshman, M.F. & Goodyear, L.J. (2003) activated protein kinase prevents Puigserver, P., Rhee, J., Lin, J. et al. (2001) Akt signaling in skeletal muscle: insulin-stimulated glucose transport but Cytokine stimulation of energy regulation by exercise and passive not glucose transporter translocation in expenditure through p38 MAP kinase stretch. American Journal of Physiology 3T3-L1 adipocytes and L6 myotubes. activation of PPARγ coactivator-1. 285, E1081–E1088. Journal of Biological Chemistry 274, Molecular Cell 8, 971–982. Saltin, G., Lingard, F., Houston, M. et al. 10 071–10 078. Rasmussen, B.B. & Winder, W.W. (1997) (1978) Physical training and glucose Tanner, C.J., Koves, T.R., Cortright, R.L. Effect of exercise intensity on skeletal tolerance in middle aged men with et al. (2002) Effect of short-term exercise muscle malonyl-CoA and acetyl-CoA chemical diabetes. Diabetes 28, 30–32. training on insulin-stimulated PI 3- carboxylase. Journal of Applied Physiology Sano, H., Kane, S., Sano, E. et al. (2003) kinase activity in middle-aged men. 83, 1104–1109. Insulin-stimulated phosphorylation of a American Journal of Physiology 282, Richter, E.A., Garetto, L.P., Goodman, Rab GTPase-activating protein regulates E147–E153. exercise regulation of insulin action 407

Tanti, J.F., Gremeaux, T., Grillo, S. et al. Welsh, G.I., Miller, C.M., Loughlin, A.J., Wojtaszewski, J.F.P., Nielsen, P., Kiens, B. (1996) Overexpression of a constitutively Price, N.T. & Proud, C.G. (1998) & Richter, E.A. (2001) Regulation of active form of phosphatidylinositol Regulation of eukaryotic initiation glycogen synthase kinase-3 in human 3-kinase is sufficient to promote Glut 4 factor eIF2B: glycogen synthase kinase-3 skeletal muscle: Effects of food intake translocation in adipocytes. Journal of phosphorylates a conserved serine and bicycle exercise. Diabetes 50, Biological Chemistry 271, 25 227–25 232. which undergoes dephosphorylation 265–269. Tipton, K.D. & Wolfe, R.R. (2001) Exercise, in response to insulin. FEBS Letters 421, Wojtaszewski, J.F.P., Jorgensen, S.B., protein metabolism, and muscle growth. 125–130. Hellsten, Y., Hardie, D.G. & Richter, E.A. International Journal of Sport Nutrition Widegren, U., Jiang, X.J., Krook, A. et al. (2002) Glycogen-dependent effects of Exercise and Metabolism 11, 109–132. (1998) Divergent effects of exercise on 5-aminoimidazole-4-carboxamide Todd, M.K. & Turcotte, L. (2003) ERK metabolic and mitogenic signaling (AICA) -riboside on AMP-activated inhibition abolishes the contraction- pathways in human skeletal muscle. protein kinase and glycogen synthase induced increase in muscle FA uptake. FASEB Journal 12, 1379–1389. activities in rat skeletal muscle. Diabetes FASEB Journal 17, 317.7 [abstract]. Widegren, U., Ryder, J.W. & Zierath, J.R. 51, 284–292. Treadway, J.L., James, D.E., Burcel, E. & (2001) Mitogen-activated protein kinase Wretman, C., Widegren, U., Lionikas, A., Ruderman, N.B. (1989) Effect of exercise signal transduction in skeletal muscle: Westerblad, H. & Henriksson, J. (2000) on insulin receptor binding and kinase effects of exercise and muscle Differential activation of mitogen- activity in skeletal muscle. American contraction. Acta Physiologica activated protein kinase signalling Journal of Physiology 256, E138–E144. Scandinavica 172, 227–238. pathways by isometric contractions in Trovati, M., Carta, Q., Cavalot, F. et al. Winder, W.W. & Hardie, D.G. (1996) isolated. Acta Physiologica Scandinavica (1984) Influence of physical training on Inactivation of acetyl-CoA carboxylase 170, 45–49. blood glucose control, glucose tolerance, and activation of AMP-activated protein Wretman, C., Lionikas, A., Widegren, U. insulin secretion, and insulin action in kinase in muscle during exercise. et al. (2001) Effects of concentric non-insulin-dependent diabetic American Journal of Physiology 270, and eccentric contractions on patients. Diabetes Care 7, 416–420. E299–E304. phosphorylation of MAPK(erk1/2) Turinsky, J. & Damrau-Abney, A. (1999) Winder, W.W., Holmes, B.F., Rubink, D.S. and MAPK(p38) in isolated rat skeletal Akt kinases and 2-deoxyglucose uptake et al. (2000) Activation of AMP-activated muscle. Journal of Physiology 535, in rat skeletal muscles in vivo: study with protein kinase increases mitochondrial 155–164. insulin and exercise. American Journal of enzymes in skeletal muscle. Journal of Yaspelkis, B.B., III, Singh, M.K., Trevino, Physiology 276, R277–R282. Applied Physiology 88, 2219–2226. B., Krisan, A.D. & Collins, D.E. (2002) Vavvas, D., Apazidis, A., Saha, A.K. et al. Wojtaszewski, J.F.P., Hansen, B.F., Urso, B. Resistance training increases glucose (1997) Contraction-induced changes in & Richter, E.A. (1996) Wortmannin uptake and transport in rat skeletal acetyl-CoA carboxylase and 5′-AMP- inhibits both insulin- and contraction- muscle. Acta Physiologica Scandinavica activated kinase in skeletal muscle. stimulated glucose uptake and transport 175, 315–323. Journal of Biological Chemistry 272, in rat skeletal muscle. Journal of Applied Young, J.C., Garthwaite, S.M., Bryan, J.E. 13 255–13 261. Physiology 81, 1501–1509. et al. (1983) Carbohydrate feeding speeds Wallberg-Henriksson, H. & Holloszy, J.O. Wojtaszewski, J.F.P., Hansen, B.F., reversal of enhanced glucose uptake in (1984) Contractile activity increases Kiens, B. & Richter, E.A. (1997) Insulin muscle after exercise. American Journal of glucose uptake by muscle in severely signaling in human skeletal muscle: time Physiology 245, R684–R688. diabetic rats. Journal of Applied Physiology course and effect of exercise. Diabetes 46, Yu, M., Blomstrand, E., Chibalin, A.V., 57(4), 1045–1049. 1775–1781. Krook, A. & Zierath, J.R. (2001) Wallberg-Henriksson, H. & Holloszy, J.O. Wojtaszewski, J.F.P., Higaki, Y., Hirshman, Marathon running increases ERK1/2 (1985) Activation of glucose transport M.F. et al. (1999a) Exercise modulates and p38 MAP kinase signalling to in diabetic muscle: responses to postreceptor insulin signaling and downstream targets in human skeletal contraction and insulin. American Journal glucose transport in muscle-specific muscle. Journal of Physiology 536, of Physiology 249, C233–C237. insulin receptor knockout mice. Journal 273–282. Wardzala, L.J., Hirshman, M.F., Pofcher, E. of Clinical Investigation 104, 1257–1264. Zorzano, A., Balon, T.W., Garetto, L.P., et al. (1985) Regulation of glucose Wojtaszewski, J.F.P., Lynge, J., Jakobsen, Goodman, M.N. & Ruderman, N.B. utilization in adipose cells and muscle A.B., Goodyear, L.J. & Richter, E.A. (1985) Muscle α-aminoisobutyric acid after long-term experimental (1999b) Differential regulation of MAP transport after exercise: enhanced hyperinsulinemia in rats. Journal of kinase by contraction and insulin in stimulation by insulin. American Journal Clinical Investigation 76, 460–469. skeletal muscle: metabolic implications. of Physiology 248, 546–552. Watt, M.J., Stellingwerff, T., Heigenhauser, American Journal of Physiology 277, Zorzano, A., Balon, T.W., Goodman, M.N. G.J. & Spriet, L.L. (2003) Effects of E724–E732. & Ruderman, N.B. (1986) Glycogen plasma adrenaline on hormone-sensitive Wojtaszewski, J.F.P., Hansen, B.F., Gade, J. depletion and increased insulin lipase at rest and during moderate et al. (2000) Insulin signaling and insulin sensitivity and responsiveness in muscle exercise in human skeletal muscle. sensitivity after exercise in human after exercise. American Journal of Journal of Physiology 550, 325–332. skeletal muscle. Diabetes 49, 325–331. Physiology 251, E664–E669. Chapter 28

Hormone and Exercise-Induced Modulation of Bone Metabolism

CLIFFORD J. ROSEN

Introduction organ infrequently thought of as a target tissue for systemic modulators. Yet the list of skeletal mediators Physical activity is an essential component of affected by chronic exercise is nearly endless and normal well-being. Within the last decade, more includes significant changes in circulating gonadal individuals have undertaken activity programs in steroids, adrenal steroids, cytokines, prostaglandins, order to ameliorate chronic medical conditions or growth hormone (GH), insulin, insulin-like growth to maintain their healthy status than ever before. factor I (IGF-I), leptin and others. All these endo- Some believe that long-term exercises will improve genous compounds influence other systems such as longevity and quality of life, while others contend metabolic fuel balance, cardiovascular fitness and that physical fitness enhances mental well-being muscle integrity to induce changes that may either as well as improving muscle function. Regardless be beneficial or harmful to the organism. of the physiology or the rationale, the long-term Weightlessness with manned space flight and the effects, either beneficial or harmful, associated with accompanying animal experiments provided the chronic exercise are an area of intense investigation. first in vivo proof of the importance of hormonal By the nature of its components, physical activ- effectors on skeletal growth, remodeling and mass. ity affects almost all hormonal systems, from β- Since those early days of space exploration, our endorphins in the brain, to local cytokines and understanding of the effects of physical training on chemokines in bone. Although the vast majority of hormonal balance, and its subsequent actions on the people believe the effects of long-term exercise are skeleton, has grown immensely. But there are still beneficial to virtually every organ system, drastic many unanswered questions. In this review, I will changes in the normal physiologic state of these focus first on the physiology of bone remodeling hormonal modulators can impact tissues in a negat- since that provides the background for understand- ive as well as positive way. Nowhere is this more ing skeletal homeostasis and the importance of cir- apparent than in the clinical syndrome of exercise- culating, as well as local growth factors in mediating induced amenorrhea and the female athlete’s adaptive processes in the skeleton. Next, I will touch triad. In predominantly younger women, chronic on the effects of acute exercise on bone cell function, and intense physical activity results in loss of the with a particular focus on the cellular aspects of gonadotropin pulse generator leading to estrogen strain within the basic multicellular unit (BMU) of deprivation. This in turn drastically affects the bone bone. In the Hormonal and skeletal responses to remodeling unit such that bone resorption exceeds long-term exercise section below, I will elaborate on formation resulting in significant bone loss. As such, the complex skeletal changes associated with sus- this interface between homeostatic processes related tained exercise programs, particularly as they relate to hormonal signaling and stress mechanisms in- to hormonal balances and imbalances. Finally, I will duced by exercise are most noticeable in bone, an discuss some potential clinical implications of our 408 hormone and exercise-induced modulation 409

current knowledge about exercise, hormonal modu- time when these three distinct activities are at their lation and skeletal homeostasis. highest level; i.e. adolescence. From about the age of 10–18 years, linear growth is very active and model- Physiology of normal and aberrant ing of the skeleton is in full force. As noted below, bone remodeling most studies that have examined the role of physical activity (i.e. loading the skeleton in one form or another, such as running or weightlifting) on bone Normal modeling and remodeling have noted a much more vigorous response in Mammalian skeletons grow and remodel during respect to changes in bone mineral density (BMD) life. Linear growth is accomplished at the growth during this time period than any other. plate and is regulated by GH and IGF-I. In rodents Remodeling is a constant process that dictates linear growth continues over a lifetime although it the metabolic needs of the skeleton and provides for is most pronounced during puberty. On the other the elasticity necessary for general physical activity. hand, the human skeleton which is more than just Every 10 years in humans the entire skeleton is calcium phosphate crystals melded together in a remodeled with the greatest turnover noted in the protein matrix, grows, models and then remodels. trabecular rich regions of the thoraco-lumbar spine Linear growth in humans occurs from the growth and several areas of the femur (Rosen 2003). Because plate, begins at birth and ceases after puberty. It is of this huge undertaking, it is not surprising that principally modulated by growth plate chondro- the mammalian skeleton is a highly organized and cytes. Modeling, which is essentially the process of physiologically active organ. Bone basically serves shaping a bone through resorption and formation, two purposes: (i) to maintain structure; and (ii) to occurs in response to several factors including preserve calcium homeostasis for all physiologic humoral substances, muscle activity and local factors. processes. As such, the mammalian skeleton is It too ceases after puberty in response to several uniquely designed for its protective and structural cues including changes in sex steroids and circulat- roles. There is an outer surface of cortical bone that ing IGF-I. Modeling is very dependent on the direc- surrounds the inner trabecular elements (Fig. 28.1). tion of the stress vectors that are modulated through Marrow bathes the trabecular skeleton while cor- muscle such that the actual cross-sectional shape of tical bone is nourished by periosteal vessels and a bone is not perfectly oval but rather slightly eccen- tric, depending on forces shaping it by loading. Remodeling is a very distinct homeostatic process Bone is a two component organ: cortical and trabecular bone analyzed by mCT compared to growth and modeling, even though the cellular players are similar. Remodeling allows the skeleton to reorganize itself without changing Trabecular bone it absolute mass and thus serves to enhance skel- etal integrity while maintaining metabolic balance, especially for essential ions such as calcium and phosphate (Rosen 2003). During remodeling, the rate of bone resorption or dissolution equals the Cortical bone rate of new bone formation. In contrast, new bone is added by growth and modeling due to linear expansion from the growth plate by chondrocytes, and expansion of lateral surfaces in the diaphysis Fig. 28.1 The two compartment model of the skeleton. by periosteal osteoblasts (OBs). Acquisition of peak The inner skeletal complex is made up of trabecular elements with a wide surface area that is bathed in bone bone mass requires optimization of all three distinct marrow. The outer shell is the cortex and on the far outer but overlapping processes. General physical activ- surface of the cortex is the perisoteal membrane. µCT, ity affects the growing skeleton, particularly at a micro computed tomography. 410 chapter 28

RANKL Activation

D-Pyr GH, IGF-I OPG RANK M-CSF

NTX CTX IL-1, IL-6, IL-11 IL-1, PTH, IL-6,

TNF, TGF-β – E2

IGF-I, II Calcium

Osteocalcin Osteocalcin OCHydroxyproline OB BSAP

PICP Matrix Proteases TGF-β H+ IGF-I LTBP IGF-II Collagen IGFBPs

Resorption - 20 days Formation - 100 days

120 days

Fig. 28.2 The bone remodeling cycle which is controlled by circulating and local growth factors and cytokines. Bone markers, including NTx, Ctx and D-Pyr, are fragments of collagen released during resorption. Formation markers including osteocalcin, BSAP and PICP are also noted by the osteoblast. BSAP, bone specific alkaline phosphatase; GH, growth hormone; IGF, insulin-like growth factor; IGFBPs, insulin-like growth factor binding proteins; IL, interleukin; M- CSF, macrophage colony-stimulating factor; OB, osteoblast; OC, osteoclast; PICP, procollagen peptide; RANK, receptor activator of nuclear factor kappa B; RANKL, RANK ligand; TGF, transforming growth factor; TNF, tumor necrosis factor. series of canaliculi connecting osteocytes to lining which contributes to coupling bone dissolution (i.e. cells and OBs (Fig. 28.2). The BMU defines the single resorption) to new bone formation, orchestrate bone functional component of bone remodeling and remodeling within a BMU (Figs 28.2 and 28.3). Pre- includes lining cells, OBs, osteoclasts (OCs) and osteoblasts (pre-OBs), derived from mesenchymal osteocytes (Figs 28.2 and 28.3). Gravitary forces stromal cells, and under the influence of a key tran- influence the BMU and stimulate cortical and tra- scription factor (Cbfa1, i.e. core binding factor I or becular remodeling. In respect to bone growth, RUNX2) represent target cells for initiation of the periosteal OBs and the underlying growth plate are remodeling cycle (Martin & Ng 1994; Thissen et al. principally responsible for longitudinal growth and 1994). Systemic and local factors, as well as signals lateral expansion. Both cortical and trabecular bone from osteocytes, enhance pre-OB differentiation, undergo remodeling, but the frequency of this pro- and this, in turn, leads to the synthesis and release cess is much less in the cortex than in the trabecular of macrophage colony-stimulating factor (M-CSF) components of the spine and distal femur. and receptor activator of nuclear factor kappa B Numerous growth factors and cytokines, each of ligand (RANKL) (Musey et al. 1993). These two hormone and exercise-induced modulation 411

and IGF-II, bound to high and low molecular weight insulin-like growth factor binding proteins (IGFBPs) (Ketelslegers et al. 1995). Similarly, the skeletal matrix also is highly enriched with these growth factors and other non-collagenous proteins, includ- ing all six IGFBPs and several IGFBP proteases. In addition, the type I IGF receptor is present on both OBs and OCs. It is now reasonably certain that skeletal IGFs originate from two sources: (i) de novo synthesis by bone forming cells (i.e. pre-OBs and fully differ- entiated OBs); and (ii) the circulation. In fact, some skeletal IGFs probably make their way into the matrix by way of specialized canaliculi and sinu- soids within the bone microcirculation (Rosen & Kessenich 1996; Rosen & Donahue 1998). IGFs, Fig. 28.3 The basic multicellular unit (BMU) consists bound to IGFBPs, can also be found within the of osteocytes connected by canaliculi to lining cells marrow milieu in close contact with the endosteal (cuboidal) and osteoblasts (ovoid). Osteoclasts are surface of bone. But, by most accounts, the vast multinucleated giant cells (see at the base of the lacuna) majority of IGF-I in bone is derived from local responsible for resorption of bone. The cellular response to loading begins with signals from the osteocytes to osteoblastic synthesis. Yet during active bone resorp- lining cells and osteoblasts, resulting in activation of the tion, as the matrix is dissolved, significant amounts remodeling sequence. of IGF-I and IGF-II are released from storage (i.e. binding to IGFBP-5 and hydroxyapatite) (Fig. 28.2). peptides are necessary and sufficient for the recruit- Subsequently, both IGFs recruit precursor OBs, ment of bone resorbing cells, i.e. the OCs. Once bone and possibly early OCs, to the bone surface where resorption occurs, calcium, collagen fragments and remodeling is occurring (Rosen & Kessenich 1996; growth factors such as the insulin-like growth Rosen & Donahue 1998; Heaney et al. 1999). factors (IGFs) and transforming growth factors Circulating and skeletal IGF-I are profoundly (TGFs), are released from the bony matrix. The latter influenced by nutritional determinants and physical factors enhance the recruitment of OBs to the bone activity. Growth retardation, a major feature of pro- surface, thereby setting the stage for collagen syn- tein calorie malnutrition in children, is associated thesis and matrix deposition/mineralization (Rosen with significant declines in circulating IGF-I despite & Donahue 1998). The entire remodeling cycle in enhanced GH secretion. Similarly elderly indivi- humans takes approximately 90 days, with the duals with poor protein intake have low serum IGF- majority of time consumed by the elaborate process I levels (Schurch et al. 1998). This almost certainly is of bone formation and subsequent mineralization due to reduced mRNA half-life of IGF-I in the liver. (Fig. 28.2) (Rosen & Donahue 1998). And, at each But, regardless of the mechanism, IGF-I is located in step, systemic hormones such as parathyroid hor- the final common homeostatic pathway affected by mone (PTH), estrogen, thyroxine and GH, influence changes in nutrient intake and energy balance. As the timing and direction of remodeling in a three- such, this peptide emerges as an important medi- dimensional space. ator of the skeletal response to stress. For example, a One of the most critical local and systemic recent study of elderly women who suffered a hip growth factor influencing bone remodeling is IGF-I fracture (i.e. the end-stage of osteoporosis) support (Fig. 28.3). IGF-I and IGF-II, the IGFs, are major this contention. After a hip fracture there is a pro- components of both the organic skeletal matrix and found decline in serum IGF-I levels, probably as the circulation. Indeed, the serum of most mammals result of poor nutrition and significant physical contains significant concentrations of both IGF-I inactivity, as well as a catabolic state (Schurch et al. 412 chapter 28

1998). Levels of IGF-I can be partially restored expression in bone, can trigger PTH release and through the use of recombinant IGF-I administered almost certainly is a principal factor in the secondary with IGFBP-3 (Boonen et al. 2002). This treatment hyperparathyroidism seen in elderly individuals. regimen in elderly patients after hip fractures, In addition, poor calcium diets and low vitamin D resulted in less bone loss and significant improve- likely contribute to dampening of the skeletal ment in functional outcomes (Boonen et al. 2002). response to loading (see Hormonal and skeletal This line of evidence supports the importance of responses to long-term exercise section below). Vit- a circulating mediator that affects skeletal respons- amin K is an essential co-factor for γ-carboxylation iveness to injury, particularly in relation to energy of osteocalcin, the most common non-collagenous status. protein in bone. Osteocalcin is produced by bone In children and young adults, exercise stimulates forming cells that are highly differentiated, and this GH secretion, which can result in higher levels of protein may be important in mineralization. The serum IGF-I. This effect is blunted in adults such expression and release of osteocalcin has also been that serum IGF-I concentrations are not statistically noted in loading studies. Other trace elements such different with sustained exercise regimens. How- as boron and strontium may affect bone cell func- ever, in respect to physical activity, anything that tion in vitro, although their role in the remodeling reduces dietary ingestion of essential nutrients (i.e. cycle is uncertain. Similarly, low levels of magne- prolonged physical activity with reduced intake, or sium may influence bone cell activity in vitro, but its compulsive undereating) in adults or children will role in remodeling and the response to exercise is override GH stimulation of IGF-I in the liver and still debated. significantly reduce circulating IGF-I concentrations. Exercise can also impact skeletal IGF-I expres- Aberrant remodeling sion. Several studies have shown that fluid flow in osteocytes and OBs can increase the expression of Aging and menopause are most frequently associ- IGF-I mRNA (Srinivasan & Gross 2000). Repetitive ated with modest uncoupling of the BMU resulting general physical activity not only increases IGF-I in accelerated bone resorption compared to bone expression in muscle but also in the periosteum and formation. Hence, over a lifespan, women can lose probably on the endosteal surface of bone. These approximately 42% of their spinal and 58% of their changes can have a profound effect on bone forma- femoral bone mass (Rosen 2003). Surprisingly, rates tion tipping the remodeling balance in a favorable of bone loss in the 8th and 9th decades of life may be direction, particularly during peak bone acquisi- comparable or exceed those found in the immediate tion. On the other hand, unloading removes the peri- and post-menopausal period of some women stimulus to bone formation, in part by making bone (Lacey et al. 1998; Robey & Bianco 1999). This is cells resistant to the actions of IGF-I (Sakata et al. due to uncoupling in the bone remodeling cycle of 2003) (see Hormonal and skeletal responses to long- older individuals resulting in a marked increase in term exercise section below). bone resorption but no change or a decrease in bone The remodeling cycle is sensitive to changes in formation (Martin & Ng 1994; Rosen & Donahue other nutrients which can profoundly affect growth 1998). However, the mechanisms that lead to an factors and cytokines in OBs. Phosphate balance is uncoupled bone remodeling unit, especially in the important for mineralization and low phosphate elderly, remain to be elucidated. Almost certainly, levels trigger activation of 1α-hydroxylase keying major changes in several hormonal factors (estro- the conversion of 25-hydroxyvitamin D to the active gen, testosterone, GH) as well as nutrient intake, compound 1,25-dihydroxyvitamin D. Conversely, dramatically affect the skeleton either singly or high levels of phosphorus stimulate PTH secretion in combination. The role of physical immobility resulting in marked activation of remodeling and on age-related changes in the skeleton remains enhanced bone resorption. Low calcium intake, uncertain, but inactivity is likely to accelerate bone coupled with vitamin D deficiency, inhibits IGF-I resorption in the elderly. hormone and exercise-induced modulation 413

Recent technological advances have made it easier to fractures. These determinants are particularly to monitor the remodeling process by defining important when considering the effects of both changes in bone specific turnover markers in patho- acute and chronic exercise programs on the aging logic states such as osteoporosis. Alterations in bone skeleton. turnover can be detected by several biochemical Weightlessness due to space flight has been markers including bone resorption indices (e.g. reported to induce the most rapid and highest rate urinary and serum N-telopeptide, C-telopeptide, of bone loss of any pathologic condition (Neuman and urinary free and total deoxypyridinoline) and 1970). This process is a function of accelerated bone bone formation markers (e.g. osteocalcin, procolla- resorption and simultaneous suppression of bone gen peptide, bone specific alkaline phosphatase) formation which begins immediately after reaching (see Fig. 28.2). In general, bone turnover markers zero gravity. Although hormonal replacement can are significantly higher in older than younger post- ameliorate part of this loss, restitution of normal menopausal women, and these indices are inversely bone mass only occurs with a normal gravity state. related to BMD (Beamer et al. 2000). For example in The role of physical inactivity that is not due to the EPIDOS trial of elderly European females, weightlessness or chronic bed rest, in the progres- the highest levels of osteocalcin, N-telopeptide, C- sion of osteoporosis is less well defined. Several telopeptide and bone specific alkaline phosphatase studies have shown that bed rest in healthy indi- were noted for those in the lowest tertile of femoral viduals can be associated with uncoupling of forma- bone density (Thissen et al. 1994). Also, increased tion from resorption, with a high resorptive state bone resorption indices were associated with a and suppressed formation (Chappard et al. 1995). A greater fracture risk independent of BMD (Thissen similar process likely occurs after a hip fracture et al. 1994). For those women in EPIDOS with low when immobility is significant and bone loss from bone density and a high bone resorption rate, there the contralateral hip can be significant even over a was a nearly fivefold greater risk of a hip fracture. In relatively short-time period (Sato et al. 2001). respect to exercise-induced changes in the skeleton, Less attention has been paid to the role that the bone turnover markers may provide some insight periosteal surface may play in aberrant remodeling into the role of systemic factors in modulating and the effects of inactivity on this region vis à vis the adaptive response to loading, particularly in hormonal and local factors. A recent 20-year pro- relation to bone resorption. spective study of post-menopausal Swedish women In contrast to a consistent pattern of high bone noted that bone loss from the radius averaged resorption indices with aging, bone formation nearly 2% per year. However in these same women, markers in osteoporotic patients are more variable. periosteal circumference increased (Fig. 28.4) sig- Serum osteocalcin levels are high in some individu- nificantly, such that the cross-sectional moment of als but this may be indicative of an increase in bone inertia, an index of bone strength, actually increased turnover rather than reflecting a true rise in bone (Ahlborg et al. 2003). This suggests that during the formation (Thissen et al. 1994). On the other hand, process of endosteal bone loss as a result of aging bone specific alkaline phosphatase, and procollagen or hormonal deficiency, the periosteum attempts peptide levels have been reported to be high, to compensate in order to maintain its inherent normal or low in elderly men and women (Musey strength (Duan et al. 2001). Currently it is not known et al. 1993). Bone histomorphometric indices in some how the periosteal envelope expands in response to patients are also quite variable. Thus, although loss of the inner aspect of bone, nor how it is sig- there is strong evidence for an age-associated rise in naled (Fig. 28.4) (Beck et al. 2000; Nelson et al. 2000). bone resorption, changes in bone formation are However, it should be noted that the two most inconsistent. Still, with aging and menopause, the important regulators of periosteal growth are mus- major skeletal abnormality is an uncoupling of cle activity and systemic IGF-I. Since the former can the remodeling unit that leads to bone loss, altered also induce skeletal IGF-I expression in the perios- skeletal architecture and an increased propensity teum and skeletal muscle, this peptide may be very 414 chapter 28

the critical mediator of this compartment. Further studies of this hypothesis are currently underway in several laboratories. Such studies will provide a strong rationale for studying the interface between physical activity, systemic hormonal modulators BMD 1.0 1.0 0.80 and bone remodeling.

X-sectional 1.77 1.77 1.77 area Hormonal and skeletal responses to Moment of 0.25 0.64 1.13 acute exercise inertia The skeleton is an extremely dynamic organ that Fig. 28.4 The skeletal effects of aging, including the loss responds not only to systemic hormonal factors of endosteal bone. In response to declining trabecular but to locally generated growth factors produced in bone, the periosteum increases in size. This enhances the response to applied forces (Table 28.1). As previ- biomechanical properties of the bone and prevents total ously noted, modeling, i.e. the process of growth failure of the skeleton. BMD, bone mineral density. and shaping, and remodeling, i.e. the process of bone renewal, are responsive to both hormonal critical for the compensatory skeletal response with mediators and local stresses. More is known about aging (Adams & Haddad 1996). Indeed, evidence the skeletal response to systemic mediators than it from transgenic and knockout mice tell us that cir- is to loading. However, there are some basic bio- culating IGF-I is important for modeling the skel- mechanical properties of the human skeleton that eton and providing it with a stimulus for optimal apply to any form of loading and are synergistic growth, particularly in the medial lateral dimension with hormonal factors that act on the BMU. (Bikle et al. 2002). Since the periosteum is highly Bone exhibits a remarkable capacity to adapt to vascular and pericytes may be the origin of OBs in changes in bone loading in order to optimize strength this compartment, it is not too radical to consider without unduly increasing weight. Surprisingly, the GH–IGF-I hormonal axis, which may be dir- the skeleton does this by altering its mass, its exter- ectly or indirectly modulated by physical activity, as nal geometry and its internal micro-architecture, a

Table 28.1 Systemic and local modulators of bone turnover: response to loading.

Hormone/ Systemic/ Local/ Loading growth factor CNS skeletal effects Reference(s)

PTH +++ – Synergistic Rosen 2003; Chow et al. 1998, Neer et al. 2001 GH +++ – Synergistic Forwood et al. 2001 IGF-I +++ ++++ +++++ Rosen & Donahue 1998, Rosen & Kessenich 1996, Bikle et al. 2002, Lean et al. 1996 +++ ++++ +++++ PGE2 (PGI) Klein-Nulend et al. 1997 Leptin +++ ++ ?? Takeda et al. 2002 Neuropeptides +++ ++? + Elefteriou et al. 2003, Takeda et al. 2002, Mason et al. 1997 TGF-β + +++++ + Srinivasan & Gross 2000 Estradiol ++++ ++ +/Synergistic Williams et al. 1995, Dueck et al. 1996, Otis et al. 1997, DiPietro & Stachenfeld 1997, Klibanski et al. 1995, Grinspoon et al. 2003, Smith & Rutherford 1993 Testosterone ++++ + Synergistic Rosen 2003 Cortisol ++++ ? Antagonistic Chrousos & Gold 1992

CNS, central nervous system; GH, growth hormone; IGF-I, insulin-like growth factor I; PGE2, prostaglandin E2; PGI, prostacyclin; PTH, parathyroid hormone; TGF-β, transforming growth factor-β. hormone and exercise-induced modulation 415

concept that has been around for more that 100 membrane receptor which is coupled to the frizzled years (Boyden et al. 2002). What has emerged more receptor on OBs and is activated by a family of recently, however, is a better understanding of the growth related peptides called Wnts. These growth factors that influence the skeletal response to load- factors can stimulate bone formation and mineral- ing, including sensitizing hormones such as PTH ization as well as cell proliferation, and are now and GH (Turner et al. 1997). To begin with, loads thought to work through a canonical pathway in applied to the skeleton are called stresses (i.e. force cells (Gong et al. 2001). LRP-5 interaction with the per unit area), while strain represents the measure Wnt frizzled ligand receptor complex results in of skeletal deformation in response to a stress (i.e. inhibition of β-catenin phosphorylation by glycogen change in length divided by its original length). It is synthetase kinase 3 (GSK-3) (Kato et al. 2002). GSK-3 the strain per se that generates an adaptive response facilitates the ubiquitin mediated breakdown of of bone to loading. Adaptation of hard tissues to β-catenin, while LRP-5 acts to prevent this break- stress is achieved by changes in bone resorption and down allowing for translocation of β-catenin into bone formation. Bone mass, geometry and trabecu- the nucleus where it interacts with a family of tran- lar orientation are altered as an adaptive response scription factors. An activating mutation of LRP-5 and these, in turn, are modulated by systemic factors. is responsible for the ‘high bone mass’ phenotype, Not all of the adaptive responses to stress are which has been reported in several families with equal because of the geometric properties of bone, healthy individuals but with BMD values 3–5 stand- but there is a direct relationship between the magni- ard deviations above normal, yet normal bone tude of strain, and the appropriate skeletal response remodeling (Boyden et al. 2002; Little et al. 2002). (Rubin & Lanyon 1985). The cellular mechanism Very recently investigators have noted that mutant responsible for that relationship is not known but LRP-5 expression was up-regulated by mechanical investigators have posited that there must be a stimulation, placing this system in the forefront of ‘mechnostat’ which regulates both periosteal forma- candidates for the elusive mechanostat (Bex et al. tion and endosteal resorption. If bone is loaded 2003). with more than 2500 microstrain, modeling occurs The type of load applied defines the particular at both sites, thereby allowing the bone to be more adaptive response of the skeleton in co-ordination resistant to deformation (Rubin & Lanyon 1985). with hormonal mediators. Three critical factors In contrast, when strain is reduced, modeling is related to physical activity modulate strain in the inhibited and endosteal remodeling with resorption human skeleton: frequency, duration and intensity occurs. The ‘putative’ mechanostat has never been (Rubin & Lanyon 1984). The latter is probably the identified, but theories abound about as to its loca- most important since intensity generates load. For tion and its mechanism of action. Most agree that example, gymnasts can experience ground forces osteocytes, buried within the cortex of bone but twelve times their body weight, while runners expe- connected to resting surface cells, sense fluid shifts rience only about three to five times their weight and gravitary forces (Noble & Reeve 2000). These (Grimston 1993; Grimston et al. 1993; Bassey et al. ‘strains’ trigger release of cellular factors which 1998). As such, the greatest increases in BMD among travel by way of canaliculi to the surface of bone active exercisers occurs at the point of impact (i.e. to activate resting OBs or pre-OBs (see Fig. 28.3) the hip) and is most pronounced in gymnasts rather (Turner 1999). This communication allows marrow than runners or walkers. stromal cells to be recruited and begin the differ- The frequency and duration of physical activity entiative process necessary for bone formation. are also important. Individuals who perform jumps Recently, evidence has emerged from genetic studies from short heights rapidly develop a peak force on of a family with high bone mass but normal bone impact; these subjects have greater increases in bone shape, that a previously unidentified pathway in density than do runners, and bone formation para- OBs may have ‘mechanostat’ like properties. Lipo- meters have been directly linked to strain rate, inde- protein receptor protein 5 (LRP-5) is a ubiquitous pendent of force (Morris et al. 1997; Fuchs et al. 2001). 416 chapter 28

In addition to the rate of change in force, evidence ing periosteal expansion and stimulating endosteal has emerged from both animal and human studies bone formation. Its potential synergy with physical that the frequency in which sound waves are gener- activity has not been tested in vivo, but studies are ated by a floor-based machine can affect the rate of currently being planned. bone formation and the acquisition of bone mass The acute response to physical activity or loading (Qin et al. 2003). Whether this represents a purely is systemic and local. The hormonal modulators skeletal effect, or one which is co-ordinated through include changes in classical stress hormones such skeletal muscle remains to be determined. However as cortisol and epinephrine. There are also bi- it does provide new insights into the bone modeling directional responses in the immune system related process. It is unknown whether there are additional to neuroendocrine function. GH levels rise, particu- hormonal factors which contribute to or are synger- larly in younger individuals, although changes in stic with this effect. IGF-I concentrations are not noticeable and are bal- Certain hormones can facilitate the ability of anced by energy needs and nutrient intake. Insulin mechanical loading to increase bone formation. The levels are generally suppressed while glucagon is two major ones are PTH and GH. Removal of the increased with physical activity. Accompanying the parathyroid glands in rats abolishes the skeletal rise in heart rate is the release of central nervous sensitivity to mechanical loading. Daily replace- system trophic factors such as neuropeptide Y, as ment with PTH restores that mechanical sensitivity well as systemic and local cytokines tied to the although the mechanism responsible for this effect immune response, including interleukin-1 (IL-1), is unknown (Chow et al. 1998). One candidate as a IL-4, IL-6, interferon-γ and IL-11 (Chrousos & Gold local mediator is IGF-I, which is induced by PTH 1992). There are also significant changes in circulat- directly through increases in skeletal gene expres- ing white blood cells as a response to these various sion. IGF-I null mice do not exhibit any anabolic immune modulators and sympathetic activity is response to PTH, nor do they show changes in also increased, leading to changes in blood pressure markers of bone formation (Bikle et al. 2002). The as well as heart rate. other important hormonal mediator is GH, another All of the above modulators have also been powerful inducer of local IGF-I. Lewis dwarf rats shown to impact the skeleton in one fashion or that do not have an intact GH–IGF-I axis do not another when exposure is chronic. The important respond to skeletal loading, but this mechano- role of neuropeptides in controlling OB function has sensitivity can be restored with GH replacement recently emerged, as has the possibility that leptin, (Forwood et al. 2001). Once again, it is not clear how an adipocyte-derived factor, plays a role in sup- GH mediates this effect, although replacement pressing bone formation and mediating hypothala- induces a significant rise in both local and systemic mic events related to obesity and starvation (Blum IGF-I. et al. 2003; Cock & Auwerx 2003; Elefteriou et al. The importance of a potent ‘modeling’ factor such 2003). Sympathetic overactivity is responsible for as GH cannot be understated in respect to the grow- the syndrome of reflex sympathetic dystrophy, and ing skeleton of pubertal children. Exercise-induced β-adrenergic blockers can prevent bone loss in changes in bone mass are most pronounced during rodents following ovariectomy (Takeda et al. 2002). the late prepubertal and pubertal growth phases However, none of these mediators play a major role (Bass 2000). Indeed, post-puberty, exercise-induced in the acute response of the skeleton to loading. That changes in bone mass are relatively minor, although adaptation is part of the mechanostat and is local- the effects on the periosteum require further stud- ized to the osteocyte and OB. Moreover, changes ies in older individuals. Recently, PTH has been that occur in the skeleton as a direct response to approved for the treatment of post-menopausal loading happen over relatively short-time intervals osteoporosis based on a large randomized placebo (e.g. minutes), well before hormonal modulators controlled trial (RPCT) of post-menopausal women are active. On the other hand, sustained changes in (Neer et al. 2001). It increases bone mass by enhanc- systemic factors are probably responsible for the hormone and exercise-induced modulation 417

more balanced skeletal response to physical activity osteocyte and OB and may be major signals for seen with long-term physical activity. recruitment of OB precursors from marrow stem Osteocytes, as noted previously, are buried deep cells. Selective inhibition of the inducible prostag- within the cortex of the skeleton (see Fig. 28.3). They landin synthase (COX-2) results in a greater sup- represent old OBs that have been entombed within pression of loading induced bone formation than a the matrix these cells have generated (Noble & Reeve non-selective blockade, although the clinical signific- 2000). Despite their relatively modest metabolic ance of this is currently unknown (Forwood 1996). rate, these terminally differentiated cells can sense As noted earlier, neuropeptides are important fluid sheer stress as mechanical loading. These cells systemic mediators of the stress response. For also communicate with bone lining cells and OBs example, neuropeptide Y is downstream of leptin through gap junctions by way of canaliculi which in the hypothalamus and can cause bone loss when can carry specific growth factors and cytokines (see administered into the cerebral ventricles of mice. Fig. 28.3). As a result, loading can transduce a relat- Also neuropeptide Y deficient mice have a marked ively rapid message to the master control cells of the increase in trabecular bone as do conditional knock- BMU in order to initiate remodeling. There is some out mice that have deletions of neuropeptide Y evidence to suggest that lining cells can even make limited to the hypothalamus. In addition to the sys- matrix in response to mechanical loading by de- temic and central nervous system effect, recent evid- differentiating into OBs. ence suggest these proteins may also be important The earliest changes in the osteocyte after load- in the acute local response to loading. A glutamate ing, is an increase in ocsteocytic glucose-6 phos- transported has been identified in bone tissue and phate dehydrogenase activity (Lean et al. 1996). This is down regulated immediately after mechanical occurs within minutes of the stimulus. At the loading (Laketic-Ljubojevic et al. 1999). Serotonin same time, OBs demonstrate an increase in intracel- receptors are present on osteocytes and OBs, particu- lular calcium, probably as a result of activation of larly in the periosteum suggesting that mechano- the phosphatidylinositol 3-kinase (PI3K) pathway transduction may in part be mediated through this which mediates intracellular calcium release. These pathway (Mason et al. 1997). changes result in stimulation of the mitogen activ- Two other major processes occur with loading ated protein (MAP) kinase signaling pathway, a key at the cellular level: prevention of apoptosis of OBs circuit for activating gene transcription. At 30– and osteocytes as a result of local growth factors 60 min after induction of stress on the bone, lining (IGF-I, leptin, IL-6, TGF-β) and the release and elab- cells and OBs begin to express c-fos, an important oration of nitric oxide, an important mediator of proto-oncogene necessary for both OB and OC OB activity and an inhibitor of osteoclastic function. function. IGF-I expression is up regulated in the Nitric oxide may be involved in mechanotransduc- OB at 1 h (Lean et al. 1996). Other factors clearly tion although the mechanism has not been defined contribute to the increase in OB activity, including (Turner et al. 1996). It has recently been shown to prostaglandins, TGF-β, neuropeptides and several down-regulate RANKL expression in stromal cells, matrix proteins that can activate integrins and a key factor in osteoclastogenesis (Rubin et al. 2003). enhance cell movement, particularly in marrow Nitric oxide also can bind to guanylyl cyclase stimu- stromal cells. lating cyclic guanosine monophosphate (cGMP), Prostaglandins likely play an important role in another intracellular mediator that can regulate the activation of bone remodeling with loading. gene transcription. Inhibition of prostaglandin synthesis by non-steroidal In sum, a host of local growth factors relatively anti-inflammatory drugs suppresses bone forma- rapidly enhance the number of OB progenitors and tion in vivo. The two most active prostaglandins terminally differentiated cells at the remodeling sur- released during this process are prostaglandin E2 face in response to loading. Interestingly, hormonal (PGE2) and prostacyclin (PGI2) (Klein-Nulend et al. mediators of the acute stress response that are active 1997). These compounds are synthesized both in the during the first minutes of physical activity do not 418 chapter 28

modulate the adaptive response in the skeleton to of resorption and increasing slightly bone forma- loading. But, ironically, their presence and regula- tion, a result of changes on the endosteal surface tion in the local bone milieu plays a major role in the of bone. On the other hand, changes in skeletal chronic adaptive response to loading. Cytokines, mass with loading in children can have a profound growth factors, neuropeptides, prostaglandins and effect on the growing skeleton, particularly on the nitric oxide contribute to an increase in bone forma- periosteal surface. For example, tennis and squash tion with physical activity. players who begin playing in their preadolescent years have a two to fourfold times greater differ- Hormonal and skeletal responses to ence in radial BMD between their playing arm and long-term exercise non-playing arm compared to those who start after puberty (Kannus et al. 1995). Similarly, young pre- As noted earlier, the skeleton is ideally suited to pubertal girls who undertook a regimen of regular respond to mechanical loading by signal trans- weight-bearing activity plus jumping for 10 months duction resulting in modeling of the periosteum and had a nearly 6% increase in femoral bone mineral trabecular reorientation. Hormonal changes as a content compared to an 8-month intervention in result of physical activity play a minor role at the adolescents (age 14.2 years) of a similar nature tissue level in accommodating the acute adaptive (Witzke & Snow 2000; Fuchs et al. 2001). In that lat- response. However, hormonal cues are particularly ter study, there was virtually no change in bone critical for regulating the process of bone remodel- mineral content or density at any site except the ing, an essential element of bone preservation in femoral trochanter. Thus site specificity and age adult life. As such, the skeletal adaptation to mech- are major determinants of the chronic response to anical loading which results from repetitive muscle loading. Finally, it should be noted that a major activity is balanced by the long-term changes in factor that is difficult to quantify, particularly in endocrine modulators such as the gonadal steroids, large-scale trials, is the amount of background phys- cortisol and neuropeptides. These changes will be ical activity occurring before or during the conduct discussed in this section. of the study. Although questionnaires are useful in Most investigators believe, although there are assessing the amount and frequency of background no long-term prospective data, that a lifetime of activity, these are certainly imprecise and subject physical activity benefits the skeleton (Bouxsein & to considerable variability, particularly in children Marcus 1994). This, indeed is the official position of where normal daily activities are quite pronounced. the American College of Sports Medicine (American However, a word of caution is necessary in rela- College of Sports 1998). However, skeletal adapta- tion to the assumption that loading the skeleton has tion to loading is very site specific, so that generalit- a more pronounced effect in growing children ies about the long-term benefits of physical activity rather than adults. Virtually all of the longitudinal on bone must be avoided. Moreover, as noted studies showing changes in bone mineral content earlier, the beneficial effects of loading the skeleton or BMD in adolescence or preadolescent children through exercise are much more pronounced before utilized areal measurements derived from DXA. and during puberty than later in life (Bass 2000). This technique is two-dimensional and very size Hence, changes in BMD, as measured by dual dependent; thus areal BMD would be expected to energy X-ray absorptiometry (DXA), are likely to be markedly increase during the peak skeletal growth. much greater in children than adults. In fact, most of This complicates interpretation of the longitudinal the trials related to physical activity in adults have data and bespeaks the need for studies looking at demonstrated only very modest changes in BMD of volumetric or three-dimensional BMD to determine the spine or hip, despite, in some cases, significant whether exercise-induced benefits to the skeleton musculoskeletal loading. This difference almost are purely growth mediated or in fact are related undoubtedly relates to the fact that loading of the directly to greater mineralization and matrix deposi- adult skeleton preserves bone mass by slowing rates tion, the critical determinants of BMD. hormone and exercise-induced modulation 419

The chronic hormonal and skeletal response to maintained BMD in the spine and sometimes in sustained physical exercise is complex and variable. the femur, but failed to prevent bone loss at other First it should be noted that there are numerous sites (Kohrt et al. 1997). On the other hand, women animal studies demonstrating a sustained skeletal receiving hormone replacement therapy (HRT) who response to chronic loading. Lanyon, Rubin and underwent a series of chronic exercise programs others have convincingly demonstrated in avian generally showed a marked increase in spine, total ulna that consecutive cycles of loading resulted in a body and femoral BMD (Kohrt et al. 1997, 1998). strain magnitude which enhanced endosteal bone The combination of exercise and calcium supple- formation (Rubin & Lanyon 1985). In fact, 36 cycles mentation in post-menopausal women may be per day (2050 microstrain at 0.5 Hz) increases bone greater than exercise or calcium supplementation formation and further increases in cycle number do alone, as reported in a meta-analysis performed by not increase bone mass (Rubin & Lanyon 1984). Specker (1996). As such there appears to be strong However, 1000 microstrain at 1 Hz led to a dose- evidence that the optimal skeletal response to exer- dependent increase in periosteal bone deposition cise in post-menopausal women requires adequate (Rubin & Lanyon 1985). Yet, exercise-induced stimu- calcium supplementation and HRT (Kanders et al. lation of bone cells in the rib following weight- 1988). bearing exercise does not induce bone formation. Exercise intervention studies in men are some- Furthermore, unloading of animal bones with what more conflicted than in women. Training of microgravity, hindlimb immobilization, spinal cord recreational male athletes between the ages of 25–52 injury, or sciatic neurectomy, result in significant years for 3 months with walking or running failed to bone loss. These studies confirm significant com- demonstrate an increase in bone mass at the spine, partmental differences (trabecular versus periosteal) humerus, femur or calcaneus. Similar but smaller as well as site specificity in regards to chronic load- studies in older men also failed to show any changes ing or unloading of the skeleton. with a 1-year-long exercise regimen. On the other Human adult studies have confirmed the results hand, in army recruits completing 14 months of in animals, although the size effect related to intensive physical training, leg bone mineral con- exercise intervention is much lower. For example tent increased 12.4% (Jones et al. 1989). This differ- exercise training in young premenopausal women ence may be a function of intensity, age or site of enhances areal BMD in a site-specific manner. maximal loading. However it does bespeak the Resistance training and weight-bearing exercises tremendous heterogeneity between studies, and the result in a very modest increase in lumbar spine, difficulty in performing systematic reviews using femoral and calcaneal BMD (Bassey et al. 1998). meta-analysis for exercise trials. And, there does not appear to be a significant dose One of the major limitations of loading trials in dependency in women for these changes, although humans relates to the endpoint being studied; i.e. those studies are not as neatly designed nor as con- areal BMD. DXA measurements of the spine or hip clusive as animal studies. Older premenopausal may not capture the true effect of skeletal loading women also show stabilization of areal BMD over a on bone mass. As noted previously, the skeleton is 12-month period in response to chronic loading; composed of both cortical and trabecular compon- however, these changes are not associated with a ents. Force induced changes in the periosteal and statistically significant increase in bone mass, sug- endosteal envelope that surround these compart- gesting that the effect size of the intervention may ments are not easily captured by an areal approx- decrease with increasing age (Pruitt et al. 1992; imation of bone mass. Indeed, changes in shape as Bassey et al. 1998). a result of skeletal modeling can not be assessed by The response of bone to chronic exercise in post- conventional DXA. Recently, evidence has emerged menopausal women is somewhat conflicted. In that changes in areal BMD with anti-osteoporosis early post-menopausal women without estrogen therapies, reflect only to a small degree the degree of replacement, resistance exercise either increased or risk reduction for subsequent fracture (Cummings 420 chapter 28

et al. 2002). In other words, the variance around frac- Decreased energy balance ture risk in relation to any intervention is much Low T3 Low IGF-I Low bone mass greater than that attributed to changes in BMD, and Low body fat/BMI Low E2 almost certainly reflects qualitative changes in both Low LH/FSH compartments of the skeleton. Hence, exercise inter- vention studies that employ BMD as a final end- Amenorrhea Anorexia point may underestimate the true effect on loading on bone strength and fracture risk. Since there are still no exercise trials with fracture as an endpoint, the degree to which changes in bone density reflect ultimate skeletal strength remains uncertain. Newer Increased risk of fracture, Increased morbidity, mortality imaging technology such as micro computed tomo- graphy (µCT), may allow for in vivo scanning of Fig. 28.5 The female athlete triad. This syndrome, trabecular bone, while magnetic resonance imaging common in young athletic women, represents a (MRI) can precisely delineate periosteal circum- compelling interface between hormones, exercise and bone adaptation. BMI, body mass index; E2, estradiol; ference, cortical thickness, trabecular number and FSH, follicle-stimulating hormone; IGF-I, insulin-like connectivity, muscle mass and marrow fat. These growth factor I; LH, luteinizing hormone; T3, endpoints may improve our capacity to define the triiodothyronine. effects of loading on the axial and appendicular skeleton, particularly in respect to aspects of bone geometry and quality. provide some basis for our understanding of the Regardless of the imaging tool, BMD represents balance between local and systemic factors. The the sum of many factors over the lifetime of an effects of a chronic exercise regimen on systemic individual. Particularly relevant to defining overall factors, such as gonadal steroids, remains a critical changes in bone quantity is the role of circulating aspect of the ultimate skeletal response to loading hormones, cytokines and growth factors. Indeed, in and must be considered within the context of the osteoporotic states, delineation of hormonal status entire organism. (estrogen in women, testosterone in men) repres- Much like the osteoporotic states of aging, the ents the first step in establishing the etiology of low pathologic syndrome now referred to as the female bone mass. Moreover, other endogenous hormones athelete’s triad, has provided investigators with such as PTH, 1,25-dihydroxyvitamin D, 25-OH significant insight into the interface between circu- vitamin D and thyroxine, all contribute to changes lating hormones, body mass and skeletal adaptation in bone remodeling and are important in deter- to exercise. This condition is defined by amenorrhea mining the length of the remodeling cycle and its (or hormonal disturbances), anorexia and low BMD subsequent balance. Similarly, circulating cytokines (Fig. 28.5) (Drinkwater et al. 1984; Marcus et al. 1985; such as IL-6 have been shown to be important in the Tomten et al. 1998). It is most commonly seen in pathogenesis of age-related bone loss and primary young adult women. To appreciate the adverse hyerparathryoidism (Nakchbandi et al. 2002). As effects of this syndrome on the skeleton and its discussed previously, IGF-I circulates in high con- relationship to circulating hormones, an overview centrations and plays a major role in skeletal model- of the changes that result from dietary restraint and ing and cortical bone mass (Rosen & Donahue 1998). excessive physical activity is necessary. Acromegaly, a condition of excess GH and IGF-I, is The hypothalamic–pituitary axis is the principle associated with large bone volumes and greater responder to stress related to dietary changes BMD (Rosen & Donahue 1998). By contrast, people and/or physical activity. Two major hypothalamic with conditions such as hypopituitarism have small factors, corticotrophin-releasing hormone (CRH), bones, increased skeletal fragility and reduced and gonadotropin-releasing hormone (GnRH) are BMD. These pathologic states are useful in defining significantly affected by energy balance associated the role of circulating factors on the skeleton and with activity patterns and nutrient availability. hormone and exercise-induced modulation 421

CRH controls the acute stress response; it in turn et al. (1995) showed that active eumenorrheic stimulates adrenocorticotropic hormone (ACTH) women experienced suppressed LH pulsatility after release from the pituitary gland, subsequently lead- only 3 days of physical training when dietary ing to cortisol production in the adrenals (Webster energy intake was reduced. This was completely et al. 2002). CRH is released into the median emin- reversed with appropriate dietary intake. Finally, ence of the hypothalamus through the portal circu- Dueck et al. (1996) showed in three amenorrheic and lation from neuronal cells of the parvaentricular three eumenorrheic women that increasing energy nucleus. It, combined with vasopressin, stimulates intake by 350 kcal·day–1 (1463 kJ·day−1) and reduc- ACTH. Subsequently the increase in cortisol pro- ing training by 1 day per week was associated with a duction enhances available energy stores by driving net increase in energy availability of 250 kcal·day–1 protein catabolism, stimulating fatty acid release (1045 kJ·day−1). At the end of 15 weeks, all six from adipose tissue and enhancing gluconeogen- athletes were retested and the amenorrheic athletes esis (Darmaun et al. 1988; Chrousos & Gold 1992). had resumption of LH pulsatility as well as a reduc- Cortisol also modulates the physiological effects of tion in cortisol secretion. various catecholamines and exerts negative feed- Chronic changes in the hypothalamic pituitary back control on the hypothalamic–pituitary axis. All system centered on GnRH and CRH can have major these processes are integral for an individual requir- effects on the skeleton and almost certainly are part ing major energy sources for the brain. But, long- of the pathogenetic mechanisms associated with term stimulation of this system can result in excess the female athelete’s triad (DiPietro & Stachenfeld cortisol release via the hypothalamic–pituitary axis, 1997; Otis et al. 1997). Weight loss, intentional or leading to increased serum cortisol concentrations unintentional due to an energy deficit, contributes and deleterious effects on muscle and bone. to a reduction in BMD by altering the mechanical The effects of long-term stress or overexer- properties of bone in relation to other body compon- cise on the GnRH, luteinizing hormone/follicle- ents and lowering systemic or local estradiol con- stimulating hormone (LH/FSH) system have also centrations. For example, reduced total body weight been studied in some detail. GnRH pulsatility is key means less gravitary forces placed on the skeleton, to the appropriate stimulation of both LH and FSH, thereby resulting in less strain, and a reduced adapt- the central regulators of ovarian steroidogenesis. A ive response. Moreover, fat is a major source of common factor in female athletes with amenorrhea peripheral estrogen production as a byproduct of or oligomenorrhea is a disruption in LH pulsatility, testosterone catabolism. Marrow and circulating almost certainly originating from the GnRH pulse estradiol levels drop in response to a decrease in generator (Loucks et al. 1992). The precise mechan- the total number of fat cells, or their cell density at ism for this early loss of pulsatility however remains peripheral sites such as the bone marrow. Also, a speculative. But a likely scenario is that a reduction reduction in body fat may reduce the number of in energy availability prevents the hypothalamic available marrow precursor cells that could even- pulse generator from providing the appropriate tually become preosteoblastic bone forming cells. signal for LH/FSH release. It seems likely that opti- Regardless of the mechanism, a drop in estradiol in mal menstrual function almost certainly requires a young women results in increased bone resorption, minimum energy requirement; when that threshold particularly on the endosteal border. If low estradiol is not attained, loss of gonadotropin pulsatility levels persist, resorption exceeds formation and results in a downstream shutdown in gonadotropin bone is lost. Moreover, the beneficial effects of phys- release. Several lines of evidence support this thesis. ical activity on the formation side may be dimin- Loucks & Callister (1993) demonstrated that, in ished or abolished by the lack of estrogen, as noted eumenorrheic sedentary women, reduced energy previously for post-menopausal women. intake over 5 days resulted in impaired LH pulsatil- Surprisingly, not all of the deleterious effects ity and a fall in trioiodothyronine. This drop occurred on BMD in amenorrheic athletes can be attributed at a threshold energy level of 20–25 kcal·kg–1 (84– to low circulating or skeletal estradiol. In fact, 105 kJ·kg−1) of lean body mass. Similarly, Williams restoration of BMD is unlikely to occur in anorexic 422 chapter 28

women even with HRT unless body weight is triad in men may not exist per se since there are no restored to at least 70% of ideal weight (Klibanski easy markers for hypothalamic dysfunction short et al. 1995). Thus energy restriction, low body weight of reduced libido. On the other hand, at least two and low body fat combine to drastically alter not groups have now demonstrated that male triath- only the set-point for the two major hypothalamic letes who are in the age range of 20 –40 years have factors noted earlier (GnRH and CRH) but also reduced total and free testosterone compared to for other factors that are nutritionally or energy non-exercising control men (Wheeler et al. 1984; dependent. These include triiodothyronine, dehy- Smith & Rutherford 1993). Moreover, although droepiandrosterone (DHEA), GH, IGF-I, IGFBP-1 exercise programs in elderly men have virtually no and several cytokines. In fact, parenteral IGF-I ther- effect on circulating testosterone, sex hormone apy can at least partially reverse the skeletal mani- binding globulin levels are higher in exercisers than festations of anorexia nervosa, particularly when controls, thereby affecting free testosterone con- combined with an anti-resorptive agent (Grinspoon centrations (Cooper et al. 1998). et al. 2003). In summary, the long-term effects of a vigorous Clinical correlates related to the exercise program on the skeleton are dependent interaction of exercise, hormones and on the energy threshold, nutrient intake, type of bone remodeling physical activity, as well as local and circulatory hormones. In general, weight-bearing exercises Exercise is considered by many to be essential promote bone formation and can, particularly in for overall well-being. The general benefits clearly younger individuals, enhance bone mass through outweigh the risks, and therefore prescriptions its adaptive response to strain. However, excessive for healthier bones should include a daily exercise exercise programs that can lead to negative energy regimen. Unfortunately, the overall impact of an balance and weight loss result in a significant exercise program on fracture risk in older men or decline in estradiol levels such that the skeleton is at women is not known. It is safe to assume that exer- significant risk for rapid bone loss. The end result of cise increases muscle performance and possibly these processes include a higher risk of fracture and muscle mass. This alone may be enough to reduce significant immobility and morbidity. Moreover, if falls in older individuals, and thereby indirectly the female athlete is young enough (i.e. 15–19 years reduce fractures. Loading of the skeleton with the of age), the loss of bone mass during this interval resultant adaptive response is optimized in situ- may not be restorable with return of menses and ations where there is adequate gonadal steroids, resumption of normal body weight. Since exer- although there are likely to be skeletal benefits for cise has become a more frequent habit among all older individuals that are gonadally insufficient. individuals, and thinness is certainly in vogue, par- Risks of a sustained exercise program are relat- ticularly in female college-aged students, the com- ively small and relate to problems loading areas of bination of several lifestyle factors in young females the skeleton that are not adaptable to stress, and may have a devastating impact on their long-term crossing a theoretic threshold where energy balance skeletal health, despite the short-term benefits becomes negative in the face of greater muscu- related to skeletal adaptation. loskeletal activity. The former issue has become It should be noted that there is likely to be an more important for the armed services, where the important gender effect on bone related to long- prevalence of stress fractures among new recruits term exercise such that women appear to be more range from 10–20% (Bennell et al. 1996a, 1996b). In susceptible to the adverse effects of sustained many cases these cortical fractures cannot be related negative energy balance than men. However, the to areal BMD, but rather are a function of training hypothalamic–pituitary–adrenal (HPA)–gonadal and readiness. In addition, for young female army axis in long-term male athletes has not been studied recruits, stress fractures of the pelvis and tibia are as extensively as the HPA axis of women athletes. In much more prevalent than in male recruits. This fact, the clinical counterpart of the female athlete’s may be a function of loading in regions of the skel- hormone and exercise-induced modulation 423

eton, particularly the periosteum, that are not used to female athlete’s triad; however, restoration of handling repetitive forces. These fractures cost the weight remains the best prognostic factor for re- armed services dearly in respect to combat readi- sumption of menses and improvement in skeletal ness, as well as the financial implications related health. Newer studies have been initiated with to caring for these individuals. Factors that predict DHEA, a weak adrenal androgen, to determine if stress fractures in military recruits or elite athletes this compound is more palatable and as effective as have not been well delineated, but clearly are some- estrogen in stopping bone resorption and enhancing what independent of bone density or the strength bone formation (Gordon et al. 2002). Other trials of the force applied (Bennell et al. 1996a, 1996b). with newer anabolic agents are being considered. Qualitative measures of bone, as determined by As noted previously, aging is associated with newer imaging technologies, may provide some significant bone loss, accelerated bone resorption clues for the future identification of those indivi- and a high risk of fractures. Defining the optimal duals susceptible to injury. Similarly, because there threshold for loading the skeleton and the resultant appears to be a heritable component to stress frac- energy balance will be mandatory in the future if tures, genetic studies may identify those high-risk exercise is to be used as a prescription for better individuals prior to the onset of training. bone health. In the meantime, a major push is Less clear cut are the clinical correlates related to needed to fully understand the fine balance between the deleterious effects of exercise on energy balance, local and systemic hormones and skeletal remodel- particularly in the young female athlete. Oral con- ing, particularly in respect to exercise programs and traceptives work in some subjects who have the their more global application.

References

Adams, G.R. & Haddad, F. (1996) The Beck, T.J., Looker, A.C., Ruff, C.B., mineral density in premenopausal relationships among IGF-1, DNA Sievanen, H. & Wahner, H.W. (2000) women. Calcified Tissue International content, and protein accumulation Structural trends in the aging femoral 73(1), 27–32. during skeletal muscle hypertrophy. neck and proximal shaft: analysis of the Boonen, S., Rosen, C., Bouillon, R. et al. Journal of Applied Physiology 81(6), Third National Health and Nutrition (2002) Musculoskeletal effects of the 2509–2516. Examination Survey dual-energy X-ray recombinant human IGF-I/IGF binding Ahlborg, H.G., Johnell, O., Turner, C.H., absorptiometry data. Journal of Bone and protein-3 complex in osteoporotic Rannevik, G. & Karlsson, M.K. (2003) Mineral Research 15(12), 2297–2304. patients with proximal femoral fracture: Bone loss and bone size after Bennell, K.L., Malcolm, S.A., Thomas, S.A. a double-blind, placebo-controlled pilot menopause. New England Journal of et al. (1996a) Risk factors for stress study. Journal of Clinical Endocrinology Medicine 349(4), 327–334. fractures in track and field athletes. A and Metabolism 87(4), 1593–1599. American College of Sports Medicine 12-month prospective study. American Bouxsein, M.L. & Marcus, R. (1994) Position Stand (1998) Exercise and Journal of Sports Medicine 24(6), 810–818. Overview of exercise and bone mass. physical activity for older adults. Bennell, K.L., Malcolm, S.A., Thomas, S.A., Rheumatic Diseases Clinics of North Medicine and Science in Sports and Wark, J.D. & Brukner, P.D. (1996b) The America 20(3), 787–802. Exercise 30(6), 992–1008. incidence and distribution of stress Boyden, L.M., Mao, J., Belsky, J. et al. (2002) Bass, S.L. (2000) The prepubertal years, a fractures in competitive track and field High bone density due to a mutation in uniquely opportune stage of growth athletes. A 12-month prospective study. LDL-receptor-related protein 5. New when the skeleton is most responsive to American Journal of Sports Medicine 24(2), England Journal of Medicine 346(20), exercise? Sports Medicine (Auckland, New 211–217. 1513–1521. Zealand) 30(2), 73–78. Bex, F., Green, P., Marzolf, J. et al. (2003) Chappard, D., Minaire, P., Privat, C. et al. Bassey, E.J., Rothwell, M.C., Littlewood, The human LRP5 G171V mutation in (1995) Effects of tiludronate on bone loss J.J. & Pye, D.W. (1998) Pre- and mice alters the skeleton response to limb in paraplegic patients. Journal of Bone and postmenopausal women have different unloading but not to ovariectomy. Mineral Research 10(1), 112–118. bone mineral density responses to the Journal of Bone and Mineral Research 18 Chow, J.W., Fox, S., Jagger, C.J. & same high-impact exercise. Journal of (suppl. 2), S463. Chambers, T.J. (1998) Role for Bone and Mineral Research 13(12), Bikle, D.D., Sakata, T., Leary, C. et al. (2002) parathyroid hormone in mechanical 1805–1813. Insulin-like growth factor I is required responsiveness of rat bone. American Beamer, W.G., Donahue, L.R. & Rosen, C.J. for the anabolic actions of parathyroid Journal of Physiology 274(1 Pt. 1), (2000) Insulin-like growth factor I and hormone on mouse bone. Journal of Bone E146–E154. bone: from mouse to man. Growth and Mineral Research 17(9), 1570–1578. Chrousos, G.P. & Gold, P.W. (1992) The Hormone & IGF Research 10 (suppl. B), Blum, M., Harris, S.S., Must, A. et al. (2003) concepts of stress and stress system S103–S105. Leptin, body composition and bone disorders. Overview of physical and 424 chapter 28

behavioral homeostasis. JAMA 267(9), Gong, Y., Slee, R.B., Fukai, N. et al. growth factor-I. Metabolism 1244–1252. (2001) LDL receptor-related protein 44(10 Suppl 4), 50–57. Cock, T.A. & Auwerx, J. (2003) Leptin: 5 (LRP5) affects bone accrual and eye Klein-Nulend, J., Burger, E.H., Semeins, cutting the fat off the bone. Lancet development. Cell 107(4), 513–523. C.M., Raisz, L.G. & Pilbeam, C.C. 362(9395), 1572–1574. Gordon, C.M., Grace, E., Emans, S.J. (1997) Pulsating fluid flow stimulates Cooper, C.S., Taaffe, D.R., Guido, D. et al. (2002) Effects of oral prostaglandin release and inducible et al. (1998) Relationship of chronic dehydroepiandrosterone on bone prostaglandin G/H synthase mRNA endurance exercise to the somatotropic density in young women with anorexia expression in primary mouse bone cells. and sex hormone status of older men. nervosa: a randomized trial. Journal of Journal of Bone and Mineral Research 12(1), European Journal of Endocrinology 138(5), Clinical Endocrinology and Metabolism 45–51. 517–523. 87(11), 4935–4941. Klibanski, A., Biller, B.M., Schoenfeld, Cummings, S.R., Karpf, D.B., Harris, F. Grimston, S.K. (1993) An application of D.A., Herzog, D.B. & Saxe, V.C. (1995) et al. (2002) Improvement in spine bone mechanostat theory to research design: The effects of estrogen administration density and reduction in risk of vertebral a theoretical model. Medicine and on trabecular bone loss in young women fractures during treatment with Science in Sports and Exercise 25(11), with anorexia nervosa. Journal of Clinical antiresorptive drugs. American Journal of 1293–1297. Endocrinology and Metabolism 80(3), Medicine 112(4), 281–289. Grimston, S.K., Willows, N.D. & Hanley, 898–904. Darmaun, D., Matthews, D.E. & Bier, D.M. D.A. (1993) Mechanical loading regime Kohrt, W.M., Ehsani, A.A. & Birge, S.J., Jr. (1988) Physiological hypercortisolemia and its relationship to bone mineral (1997) Effects of exercise involving increases proteolysis, glutamine, and density in children. Medicine and Science predominantly either joint-reaction or alanine production. American Journal of in Sports and Exercise 25(11), 1203–1210. ground-reaction forces on bone mineral Physiology 255(3 Pt. 1), E366–E373. Grinspoon, S., Miller, K., Herzog, D., density in older women. Journal of DiPietro, L. & Stachenfeld, N.S. (1997) The Clemmons, D. & Klibanski, A. (2003) Bone and Mineral Research 12(8), female athlete triad. Medicine and Science Effects of recombinant human insulin- 1253–1261. in Sports and Exercise 29(12), 1669–1671. like growth factor (IGF)-I and estrogen Kohrt, W.M., Ehsani, A.A. & Birge, S.J., Jr. Drinkwater, B.L., Nilson, K., Chesnut, administration on IGF-I, IGF binding (1998) HRT preserves increases in bone C.H., 3rd et al. (1984) Bone mineral protein (IGFBP)-2, and IGFBP-3 in mineral density and reductions in body content of amenorrheic and anorexia nervosa: a randomized- fat after a supervised exercise program. eumenorrheic athletes. New England controlled study. Journal of Clinical Journal of Applied Physiology 84(5), Journal of Medicine 311(5), 277–281. Endocrinology and Metabolism 88(3), 1506–1512. Duan, Y., Seeman, E. & Turner, C.H. (2001) 1142–1149. Lacey, D.L., Timms, E., Tan, H.L. et al. The biomechanical basis of vertebral Heaney, R.P., McCarron, D.A., Dawson- (1998) Osteoprotegerin ligand is a body fragility in men and women. Hughes, B. et al. (1999) Dietary changes cytokine that regulates osteoclast Journal of Bone and Mineral Research favorably affect bone remodeling in differentiation and activation. Cell 93(2), 16(12), 2276–2283. older adults. Journal of American Dietetic 165–176. Dueck, C.A., Matt, K.S., Manore, M.M. & Association 99(10), 1228–1233. Laketic-Ljubojevic, I., Suva, L.J., Maathuis, Skinner, J.S. (1996) Treatment of athletic Jones, B.H., Harris, J.M., Vinh, T.N. & F.J., Sanders, D. & Skerry, T.M. (1999) amenorrhea with a diet and training Rubin, C. (1989) Exercise-induced Functional characterization of N- intervention program. International stress fractures and stress reactions of methyl-d-aspartic acid-gated channels Journal of Sport Nutrition 6(1), 24–40. bone: epidemiology, etiology, and in bone cells. Bone 25(6), 631–637. Elefteriou, F., Takeda, S., Liu, X., classification. Exercise and Sport Sciences Lean, J.M., Mackay, A.G., Chow, J.W. Armstrong, D. & Karsenty, G. (2003) Review 17, 379–422. & Chambers, T.J. (1996) Osteocytic Monosodium glutamate-sensitive Kanders, B., Dempster, D.W. & Lindsay, R. expression of mRNA for c-fos and IGF-I: hypothalamic neurons contribute to the (1988) Interaction of calcium nutrition an immediate early gene response to an control of bone mass. Endocrinology and physical activity on bone mass in osteogenic stimulus. American Journal of 144(9), 3842–3847. young women. Journal of Bone and Physiology 270(6 Pt. 1), E937–E945. Forwood, M.R. (1996) Inducible cyclo- Mineral Research 3(2), 145–149. Little, R.D., Carulli, J.P., Del Mastro, R.G. oxygenase (COX-2) mediates the Kannus, P., Haapasalo, H., Sankelo, M. et al. (2002) A mutation in the LDL induction of bone formation by et al. (1995) Effect of starting age of receptor-related protein 5 gene results mechanical loading in vivo. Journal physical activity on bone mass in the in the autosomal dominant high-bone- of Bone and Mineral Research 11(11), dominant arm of tennis and squash mass trait. American Journal of Human 1688–1693. players. Annals of Internal Medicine Genetics 70(1), 11–19. Forwood, M.R., Li, L., Kelly, W.L. & 123(1), 27–31. Loucks, A.B. & Callister, R. (1993) Bennett, M.B. (2001) Growth hormone is Kato, M., Patel, M.S., Levasseur, R. et al. Induction and prevention of low-

permissive for skeletal adaptation to (2002) Cbfa1-independent decrease in T3 syndrome in exercising women. mechanical loading. Journal of Bone and osteoblast proliferation, osteopenia, American Journal of Physiology 264 Mineral Research 16(12), 2284–2290. and persistent embryonic eye (5 Pt. 2), R924–R930. Fuchs, R.K., Bauer, J.J. & Snow, C.M. (2001) vascularization in mice deficient in Loucks, A.B., Laughlin, G.A., Mortola, J.F. Jumping improves hip and lumbar spine Lrp5, a Wnt coreceptor. Journal of Cell et al. (1992) Hypothalamic–pituitary– bone mass in prepubescent children: Biology 157(2), 303–314. thyroidal function in eumenorrheic and a randomized controlled trial. Journal Ketelslegers, J.M., Maiter, D., Maes, M., amenorrheic athletes. Journal of Clinical of Bone and Mineral Research 16(1), Underwood, L.E. & Thissen, J.P. (1995) Endocrinology and Metabolism 75(2), 148–156. Nutritional regulation of insulin-like 514–518. hormone and exercise-induced modulation 425

Marcus, R., Cann, C., Madvig, P. et al. Pruitt, L.A., Jackson, R.D., Bartels, R.L. & Physiology and Occupational Physiology (1985) Menstrual function and bone Lehnhard, H.J. (1992) Weight-training 67(4), 330–334. mass in elite women distance runners. effects on bone mineral density in early Specker, B.L. (1996) Evidence for an Endocrine and metabolic features. postmenopausal women. Journal of Bone interaction between calcium intake and Annals of Internal Medicine 102(2), and Mineral Research 7(2), 179–185. physical activity on changes in bone 158–163. Qin, Y.X., Kaplan, T., Saldanha, A. & mineral density. Journal of Bone and Martin, T.J. & Ng, K.W. (1994) Mechanisms Rubin, C. (2003) Fluid pressure Mineral Research 11(10), 1539–1544. by which cells of the osteoblast lineage gradients, arising from oscillations in Srinivasan, S. & Gross, T. (2000) control osteoclast formation and intramedullary pressure, is correlated Intermittent rest enhances osteoblastic activity. Journal of Cellular Biochemistry with the formation of bone and activation induced by mechanical 56(3), 357–366. inhibition of intracortical porosity. loading. Transactions of the Orthopedic Mason, D.J., Suva, L.J., Genever, P.G. Journal of Biomechanics 36(10), Research Society 25, 628. et al. (1997) Mechanically regulated 1427–1437. Takeda, S., Elefteriou, F., Levasseur, R. expression of a neural glutamate Robey, P.B. & Bianco, P. (1999) Cellular et al. (2002) Leptin regulates bone transporter in bone: a role for excitatory mechanism of age related bone loss. In: formation via the sympathetic nervous amino acids as osteotropic agents? Bone Aging Skeleton (Rosen, C.J., Glowacki, J. system. Cell 111(3), 305–317. 20(3), 199–205. & Bilezikian, J., eds.). Academic Press, Thissen, J.P., Ketelslegers, J.M. & Morris, F.L., Naughton, G.A., Gibbs, J.L., San Diego: 145–157. Underwood, L.E. (1994) Nutritional Carlson, J.S. & Wark, J.D. (1997) Rosen, C.J. (2003) Restoring aging bones. regulation of the insulin-like growth Prospective 10-month exercise Scientific American 288(3), 70–77. factors. Endocrine Reviews 15(1), 80–101. intervention in premenarcheal girls: Rosen, C.J. & Donahue, L.R. (1998) Tomten, S.E., Falch, J.A., Birkeland, K.I., positive effects on bone and lean mass. Insulin-like growth factors and bone: Hemmersbach, P. & Hostmark, A.T. Journal of Bone and Mineral Research 12(9), the osteoporosis connection revisited. (1998) Bone mineral density and 1453–1462. Proceedings of the Society for Experimental menstrual irregularities. A comparative Musey, V.C., Goldstein, S., Farmer, P.K., Biology and Medicine 219(1), 1–7. study on cortical and trabecular bone Moore, P.B. & Phillips, L.S. (1993) Rosen, C.J. & Kessenich, C. (1996) The role structures in runners with alleged Differential regulation of IGF-1 and of IGF-I in senesence: implications for normal eating behavior. International IGF-binding protein-1 by dietary therapy. Endocrinologists 6, 102–106. Journal of Sports Medicine 19(2), 92–97. composition in humans. American Rubin, C.T. & Lanyon, L.E. (1984) Turner, C.H. (1999) Site-specific skeletal Journal of the Medical Sciences 305(3), Regulation of bone formation by applied effects of exercise: importance of 131–138. dynamic loads. Journal of Bone and Joint interstitial fluid pressure. Bone 24(3), Nakchbandi, I.A., Mitnick, M.A., Lang, R. Surgery. American Volume 66(3), 397–402. 161–162. et al. (2002) Circulating levels of Rubin, C.T. & Lanyon, L.E. (1985) Turner, C.H., Takano, Y., Owan, I. & interleukin-6 soluble receptor predict Regulation of bone mass by mechanical Murrell, G.A. (1996) Nitric oxide rates of bone loss in patients with strain magnitude. Calcified Tissue inhibitor L-NAME suppresses primary hyperparathyroidism. Journal International 37(4), 411–417. mechanically induced bone formation of Clinical Endocrinology and Metabolism Rubin, J., Murphy, T.C., Zhu, L. et al. in rats. American Journal of Physiology 87(11), 4946–4951. (2003) Mechanical strain differentially 270(4 Pt. 1), E634–E639. Neer, R.M., Arnaud, C.D., Zanchetta, J.R. regulates eNOS and RANKL expression Turner, C.H., Anne, V. & Pidaparti, R.M. et al. (2001) Effect of parathyroid via ERK1/2 in primary bone stromal (1997) A uniform strain criterion hormone (1–34) on fractures and bone cells. Journal of Bone and Mineral Research for trabecular bone adaptation: do mineral density in postmenopausal 18 (suppl. 2), S10. continuum-level strain gradients drive women with osteoporosis. New England Sakata, T., Halloran, B.P., Elalieh, H.Z. adaptation? Journal of Biomechanics 30(6), Journal of Medicine 344(19), 1434–1441. et al. (2003) Skeletal unloading induces 555–563. Nelson, D.A., Barondess, D.A., Hendrix, resistance to insulin-like growth factor I Webster, J.I., Tonelli, L. & Sternberg, E.M. S.L. & Beck, T.J. (2000) Cross-sectional on bone formation. Bone 32(6), 669–680. (2002) Neuroendocrine regulation of geometry, bone strength, and bone mass Sato, Y., Kaji, M., Higuchi, F. et al. immunity. Annual Review of Immunology in the proximal femur in black and white (2001) Changes in bone and calcium 20, 125–163. postmenopausal women. Journal of Bone metabolism following hip fracture Wheeler, G.D., Wall, S.R., Belcastro, A.N. and Mineral Research 15(10), 1992–1997. in elderly patients. Osteoporosis & Cumming, D.C. (1984) Reduced Neuman, W.F. (1970) Calcium metabolism International 12(6), 445–449. serum testosterone and prolactin levels in space flight. Life Sciences and Space Schurch, M.A., Rizzoli, R., Slosman, D. in male distance runners. JAMA 252(4), Research 8, 309–315. et al. (1998) Protein supplements 514–516. Noble, B.S. & Reeve, J. (2000) Osteocyte increase serum insulin-like growth Williams, N.I., Young, J.C., McArthur, J.W. function, osteocyte death and bone factor-I levels and attenuate proximal et al. (1995) Strenuous exercise with fracture resistance. Molecular and Cellular femur bone loss in patients with recent caloric restriction: effect on luteinizing Endocrinology 159(1–2), 7–13. hip fracture. A randomized, double- hormone secretion. Medicine and Science Otis, C.L., Drinkwater, B., Johnson, M., blind, placebo-controlled trial. Annals of in Sports and Exercise 27(10), 1390–1398. Loucks, A. & Wilmore, J. (1997) Internal Medicine 128(10), 801–809. Witzke, K.A. & Snow, C.M. (2000) Effects American College of Sports Medicine Smith, R. & Rutherford, O.M. (1993) Spine of plyometric jump training on bone position stand. The female athlete triad. and total body bone mineral density mass in adolescent girls. Medicine and Medicine and Science in Sports and Exercise and serum testosterone levels in male Science in Sports and Exercise 32(6), 29(5), i–ix. athletes. European Journal of Applied 1051–1057. Chapter 29

Diet and Hormonal Responses: Potential Impact on Body Composition

JEFF S. VOLEK AND MATTHEW J. SHARMAN

Introduction responses to feeding has not been undertaken. The studies reviewed in this chapter are from diverse In order to sustain life, there must be a continuous perspectives but share the common design com- supply of energy obtained from the diet. The prim- ponent of measuring the hormonal response to food ary energy-providing nutrients are protein, carbo- intake. We synthesized these postprandial hor- hydrate and fat. When food is ingested, there is an monal data in an attempt to shed light on the con- acute increase in plasma metabolites and hormones sistency or lack thereof related to the hormonal that is dependent, in part, on timing and macro- response to intake of meals with different com- nutrient distribution of the meal. Nutrients and position with and without exercise. We discuss hormones then interact at target tissues to regulate separately the effects of meals on growth hormone cellular processes. In the context of body composi- (GH), insulin-like growth factor I (IGF-I), testoster- tion, nutrients and hormones have important roles one, cortisol and insulin because these hormones in regulating skeletal muscle protein and adipose have a major role in regulating protein and lipid triacylglycerol balance, which over time impacts metabolism. For each hormone, we overview the body composition. Exercise also has independent postprandial response to meals of different macro- effects on nutrients and hormone availability and nutrient distribution and the exercise-induced hor- interacts with feeding to create a unique setting monal response to feeding before, during, or after that impacts metabolic processes including body exercise. Although theoretical, the implications of composition. This general scenario is depicted in diet-induced hormonal responses on body composi- Fig. 29.1. The postprandial phase of metabolism tion are discussed. We focus primarily on human includes the time when nutrients are being absorbed studies and only include animal work to provide from the gut and appear in the circulation. This time supporting information. period continues several hours after feeding and therefore most people are in a postprandial phase Effects of diet on hormones for the majority of their life. Similar to the situation linking the postprandial lipoprotein response to car- Growth hormone diovascular disease (Patsch et al. 1992; Ebenbichler et al. 1995), the postprandial hormonal response GH is a peptide hormone secreted from the anter- could be viewed as more physiologically important ior pituitary. In skeletal muscle, GH promotes a than the postabsorptive hormonal environment, positive protein balance by increasing protein syn- particularly for enhancing body composition. thesis and possibly inhibiting protein breakdown Given the important regulatory affects of hor- (Rooyackers & Nair 1997). These effects are contro- mones, it is surprising that a more systematic and versial in part because it is difficult to separate out comprehensive study of the effects of hormonal the effects mediated through stimulation of hepatic 426 impact of diet on hormones 427

Intake of Alteration Alteration Nutrients and hormones interact macronutrients in plasma in plasma at target tissues to regulate metabolites hormones protein and lipid balance

Protein Exercise stimulus Skeletal Carbohydrate muscle Fat

PRO AA

Glucose GH IGF-I Amino acids Testosterone Cortisol Chylomicrons Body Insulin composition

TAG FA

Adipocyte

Fig. 29.1 Intake of macronutrients results in appearance of glucose, amino acids and triacylglycerols in the form of chylomicrons in the plasma. These nutrients are also affected by exercise and together contribute to a release of hormones that then partition nutrients and interact with target tissues such as skeletal muscle and adipose tissue to regulate among other processes, protein synthesis/breakdown and adipose tissue lipolysis/lipogenesis, the balance of which over time impacts body composition. AA, amino acids; FA, fatty acids; GH, growth hormone; IGF-I, insulin-like growth factor-I; PRO, protein; TAG, triacylglycerols.

or skeletal muscle IGF-I. In adipose tissue, GH release pattern of GH. Studies have shown that pro- increases lipolysis and there has been specific inter- tein, fat and carbohydrate each have independent est in the C-terminal fragment of GH shown to have effects on regulation of GH secretion. An oral gluc- marked lipolytic and antilipogenic activity in vitro ose tolerance test results in significantly reduced and when administered to animals (Heffernan et al. GH levels (Hjalmarsen et al. 1996; Bernardi et al. 2000). GH dramatically decreases lipogenesis with a 1999; Nakagawa et al. 2002). Carbohydrate intake concomitant increase in muscle mass indicating a (0.7 g·kg–1·h–1) alone or in combination with protein powerful nutrient partitioning effect (Etherton 2000). (0.35 g·kg–1·h–1) resulted in declining GH levels over a 2-h period (van Loon et al. 2003). An oral glucose postprandial response to feeding tolerance test (75 g glucose) resulted in a slight and gradual decrease in GH for 2 h followed by a The GH response to meals has been shown to be significant increase that peaked at 240 min (Frystyk quite variable (Baker et al. 1972; van Loon et al. 2003), et al. 1997). Thus, hyperglycemia is associated with which may be explained in part by the pulsatile a decrease in GH, which may be followed by 428 chapter 29

rebound hypoglycemia and a subsequent increase significant effect on GH responses (Williams et al. in GH. This is consistent with work showing hypo- 2002). glycemia is a potent stimulator of GH (Roth et al. Meals also alter the GH response to submaximal 1963). Thus in response to carbohydrate, GH tends exercise. Whitley et al. (1998) compared the effects of to be decreased and increased in the early and late fasting to isoenergetic (956 kcal [40031 kJ]) high-fat postprandial periods, respectively. (74% fat) and high-carbohydrate (86% carbohy- Although infusion and, in some cases, oral inges- drate) meals consumed 4 h prior to cycling for 90 min tion of large doses of certain amino acids (e.g. on GH responses. The GH response during the fast- arginine, lysine and ornithine) can increase GH ing and carbohydrate trial were similar and signific- levels, the effect is quite variable and reduced by antly higher compared to the high-fat trial. Consistent physical training and high-protein diets (reviewed with the effects of glucose and fatty acids on GH by Chromiak & Antonio 2002). Ingestion of beef secretion, the lower GH levels during exercise after steak resulted in an increase in GH, presumably due the fat-rich meal was associated with higher blood to the amino acid content, but the increase was pre- glucose and fatty acid levels. Cappon et al. (1993) com- vented when heparin was injected to acutely elevate pared the effects of isoenergetic (520 kcal [2177 kJ]) circulating fatty acids (Fineberg et al. 1972). In line high-fat and high-carbohydrate liquid meals con- with an inhibitory effect of fatty acids on GH, other sumed 45 min before 10 min of intense exercise on work has confirmed that circulating fatty acids, but post-exercise GH responses. Compared to ingestion not triacylglycerols, inhibit GH secretion (Blackard of a noncaloric placebo, the post-exercise GH area et al. 1971). The time course of the GH response to under the curve was decreased by –54% after the food is not entirely clear. Acute ingestion of a liquid high-fat meal and by –25% after the high-carbohy- supplement (520 kcal [2177 kJ]) rich in fat or rich in drate meal. Post-exercise somatostatin was elevated carbohydrate did not affect GH measured 35 min after the high-fat meal and the authors postulate later (Cappon et al. 1993) nor did a mixed carbo- a potential link between dietary fat, somatostatin hydrate and protein meal (216 kcal [903 kJ]) affect and GH (Cappon et al. 1993). Somatostatin also GH levels measured 90 min later (Carli et al. 1992). decreases ghrelin (Schaller et al. 2003), a recently Alcohol ingestion (0.45 g ethanol·kg–1) consumed identified potent GH-releasing peptide primarily each hour for 3 h had no effect on GH responses produced in the stomach, providing an alternative compared to water (Rojdmark et al. 2000). fat-induced mechanism to decrease GH. The finding that a high-fat meal reduces post- exercise-induced responses to feeding exercise GH more than a high-carbohydrate meal is somewhat counterintuitive since carbohydrate Carbohydrate and protein intake alter the GH feeding would likely lead to higher glucose levels, response to resistance exercise. A protein and car- which should inhibit GH. In line with this role bohydrate supplement consumed immediately and of GH as a counter-regulatory hormone, several 2 h after resistance exercise increased GH during studies have shown that compared to placebo, car- late recovery when glucose levels were lower com- bohydrate beverages consumed before, during and pared to a placebo (Chandler et al. 1994), consistent after exercise result in higher blood glucose levels with the known effects of hypoglycemia on GH. and lower GH responses to exercise (Bonen et al. Our laboratory reported that a protein and carbohy- 1980; Tsintzas et al. 1996; Utter et al. 1999). Miller et al. drate supplement consumed before and immedi- (2002) showed that carbohydrate provided during ately after a bout of resistance exercise resulted in an 2 h of cycling blunted the post-exercise GH response enhanced acute GH response from 0–30 min post- compared to non-fat milk and a non-caloric placebo exercise compared to a non-caloric placebo despite despite similar glucose levels among the three trials similar glucose levels between trials (Kraemer et al. suggesting that GH response to exercise are medi- 1998). Another study reported that protein and car- ated by factors other than blood glucose. Studies bohydrate intake after resistance exercise had no have also shown that carbohydrate provided before impact of diet on hormones 429

and during 2 h of cycling (Murray et al. 1995) or total IGF-I levels measured 2 h later (van Loon et al. rowing (Henson et al. 2000) do not affect GH levels 2003). In another study, total IGF-I levels were not despite differences in glucose levels. In one of these affected when measured 4 h after a carbohydrate- studies (Murray et al. 1995), subjects also consumed rich breakfast (Bereket et al. 1996). A mixed meal the beverages with and without nicotinic acid, an had no effect on total IGF-I levels measured over the inhibitor of lipolysis. Nicotinic acid consumed with next 5 h (Svanberg et al. 2000) and another study either water or carbohydrate prevented the increase showed no changes in total IGF-I levels over a 24-h in fatty acids and resulted in increased post-exercise period during which subjects were fed mixed meals GH levels compared to water and carbohydrate (Frystyk et al. 2003). Consumption of ethanol did without nicotinic acid (Murray et al. 1995). Thus, the reduce IGF-I levels during the late postprandial GH response to exercise appears to be dependent in period (5–7 h after ingestion) (Rojdmark et al. 2000). part on glucose levels, but fatty acid levels also exert Thus, feeding does not appear to affect total IGF-I an independent effect. levels with the exception of a delayed decrease after Exercise-induced elevations in GH to a high-fat alcohol ingestion. However, feeding does appear to diet or fasting are accompanied by a more rapid alter IGF-I binding proteins. decline in plasma insulin and glucose concentra- tions during exercise suggesting that glucose sensit- Effects of macronutrients on other components of the ive receptors may modulate the GH response to IGF-I system. There are six high affinity insulin-like exercise (Galbo et al. 1979). Glucose infusion at the growth factor binding proteins (IGFBP-1 to 6) that end of exercise does not attenuate the greater GH circulate bound to the vast majority of IGF-I. IGFBP- response observed after a fat-rich diet (Galbo et al. 1 shows the most rapid dynamic regulation in 1979; Johannessen et al. 1981) again suggesting that plasma in response to meals and it has been sug- other substrates (e.g. glycogen, ketones, fatty acids) gested that it contributes to glucose regulation or hormones (e.g. insulin, catecholamines) may be by virtue of its role in countering the insulin-like involved in the regulation of GH secretion. bioactivity of IGF-I (i.e. hypoglycemic effect), most likely by controlling free levels of IGF-I (Lee et al. 1993, 1997). After an oral glucose tolerance test, Insulin-like growth factor I IGFBP-1 gradually decreased reaching a level –52% IGF-I is an anabolic hormone that stimulates growth below baseline after 180 min followed by a gradual in almost all tissues and is likely responsible for increase to values 74% above baseline at 5 h (Frystyk many of the effects of GH. IGF-I is primarily pro- et al. 1997). Opposite to IGFBP-1, free IGF-I was duced in the liver but also in other tissues including significantly decreased (–29% to –38%) during the skeletal muscle under stimulation by GH. In skeletal late postprandial period (270–330 min) and was muscle, IGF-I increases protein balance primarily by inversely related to IGFBP-1 at baseline and during increasing protein synthesis (Rooyackers & Nair the late postprandial period. Other studies have 1997). In adipose tissue, IGF-I has insulin-like effects observed decreases in IGFBP-1 in response to mixed stimulating glucose uptake and inhibiting lipolysis meals (Bereket et al. 1996; Frystyk et al. 2003), oral (Siddals et al. 2002). glucose (Bernardi et al. 1999) or intravenous glucose and insulin injections (Nyomba et al. 1997). Inverse postprandial response to feeding associations have been shown between IGFBP-1 and glucose, insulin and cortisol levels (Hopkins et al. Effects of macronutrients on total IGF-I levels. An oral 1994; Holden et al. 1995; Frystyk et al. 1997; Ricart & glucose tolerance test (75 g glucose) had no effect on Fernandez-Real 2001). Acute ingestion of ethanol total IGF-I levels measured every 30 min for 330 resulted in a rapid rise in IGFBP-1 that was independ- min in healthy subjects (Frystyk et al. 1997). Carbo- ent of changes in glucose, insulin and GH, suggest- hydrate intake (0.7 g·kg–1·h–1) alone or in combina- ing its regulation is partially mediated by other tion with protein (0.35 g·kg–1·h–1) had no effect on mechanisms. 430 chapter 29

Although there is a significant amount of evid- showed that a nutritional complete meal fed imme- ence supporting the role of the GH–IGF-I axis in diately after 2 h of treadmill running had no effect glucose metabolism (Holt et al. 2003), there has been on exercised induced total IGF-I levels compared to debate whether the IGF-I system is important in a food-deprived group of rats. Thus, acute feeding glucose regulation in response to metabolic stressors of any composition does not appear to affect sys- (e.g. feeding and/or exercise). This is because free temic levels of total IGF-I. IGF-I levels are generally not affected during the early postprandial period when IGFBP-1, glucose Effects of macronutrients on exercise-induced compon- and insulin are elevated but free IGF-I levels are ents of the IGF-I system. Similar to the situation after reduced during the late postprandial period and feeding, it has been proposed that IGFBP-1 may during the night when glucose and insulin are have an important role in glucose regulation during normalized (Frystyk et al. 1997, 2003). Thus, the and after exercise (Koistinen et al. 1996). Compared majority of evidence indicates that the IGF-I system to placebo, exercise-induced IGFBP-1 levels were does not partake in meal-related glucose regulation reduced in response to feeding of carbohydrate but does contribute to glucose disposal during the (Hopkins et al. 1994). In this study, IGFBP-1 and gluc- post-absorptive state. ose levels were correlated in the control condition Since IGF-I is produced in many tissues including but not during the carbohydrate feeding trial, skeletal muscle, it is possible that the IGF-I system suggesting that other factors other than glucose and is contributing to meal-related glucose regulation insulin regulate IGFBP-1 responses to prolonged through an autocrine and/or paracrine mechanism. exercise (Hopkins et al. 1994). Anthony et al. (2001) However, this hypothesis is unlikely based on showed that provision of a nutritionally complete findings from a recent study showing that feeding meal immediately after exercise did not alter the does not affect postprandial skeletal muscle IGF-I exercise-induced increase in hepatic IGFBP-1 mRNA mRNA expression in humans (Svanberg et al. 2000). expression or circulating IGFBP-1 levels despite higher glucose and insulin levels. Again, this is con- exercise-induced responses to feeding sistent with the hypothesis that exercise-induced increases in IGFBP-1 levels are not directly regu- Effects of macronutrients on total exercise-induced IGF-I lated by circulating glucose levels and do not responses. Consistent with the lack of a change of assist in preventing IGF-I induced hypoglycemia food on total IGF-I levels, studies have shown that by binding free IGF-I in plasma. Findings from a re- pre-exercise and post-exercise meals do not alter the cent study showed high correlations between liver total IGF-I response to exercise (Cappon et al. 1994; glycogen and IGFBP-1 responses to exercise, sug- Hopkins et al. 1994; Kraemer et al. 1998; Anthony gesting that the magnitude of liver glycogen deple- et al. 2001). A pre-exercise carbohydrate- and fat-rich tion may mediate IGFBP-1 responses to exercise meal had the same effect as a non-caloric placebo (Lavoie et al. 2002). Although speculative, it is pos- on IGF-I responses to 10 min of intense cycling sible that IGFBP-1 responses to exercise are modulat- exercise, suggesting that pre-exercise feeding and ing anabolic (growth), as opposed to the metabolic meal composition do not influence exercise-induced (glucose-lowering), effects of IGF-I. IGF-I levels (Cappon et al. 1994). There were no differences in total IGF-I levels in a group of men fed Testosterone a placebo or a glucose polymer solution during cycling exercise to fatigue (Hopkins et al. 1994). Our Testosterone is a steroid hormone secreted from laboratory showed that a protein and carbohydrate specialized cells in the testes in men and from the supplement consumed 2 h before and immediately ovaries in women. In skeletal muscle, testosterone after a bout of whole body resistance exercise increases protein balance primarily by increasing did not alter the exercise-induced IGF-I response protein synthesis whereas the effects on protein (Kraemer et al. 1998). Finally, Anthony et al. (2001) breakdown are unclear (Rooyackers & Nair 1997). impact of diet on hormones 431

In adipose tissue, testosterone inhibits lipid up- with only one to our knowledge performed in men take and LPL activity, and stimulates lipolysis by (Hjalmarsen et al. 1996). In this study, total testo- increasing the number of lipolytic β-adrenergic sterone and estimated free testosterone levels were receptors (De Pergola 2000). significantly decreased during a 2-h oral glucose to- lerance test. There were no changes in SHBG but LH postprandial response to feeding was significantly increased (Hjalmarsen et al. 1996). The above studies have all involved men. Studies The testosterone response to isocaloric (800 kcal generally indicate that testosterone is also decreased [3350 kJ]) meals that were high-fat (57% fat, 9% pro- modestly after an oral glucose tolerance test in tein, 34% carbohydrate) or low-fat (1% fat, 26% pro- healthy normal-weight women (Smith, S. et al. 1987; tein, 73% carbohydrate) was examined in healthy Falcone et al. 1990; Tiitinen et al. 1990; Aizawa & men (Meikle et al. 1990). Testosterone was not affected Niimura 1996; Ivandic et al. 1999), which may be after the low-fat meal, but postprandial total and attributed in part to the normal diurnal variation in free testosterone concentrations were approxim- the hormone. There have been a number of studies ately 30% lower for 4 h after the high-fat meal. The examining the androgen response to oral glucose fat-induced decrease was not related to changes tolerance tests in women with polycystic ovarian in other steroids (i.e. estrone, estradiol, dihydro- syndrome (PCOS) because they often exhibit both testosterone, lutenizing hormone [LH], percent elevated androgens and insulin, which are associ- free testosterone, or sex hormone binding globulin ated with insulin resistance. Insulin acts to stimulate [SHBG] binding capacity). This study indicates fat- testosterone in the ovaries (Cara & Rosenfield 1988), rich meals, but not carbohydrate-rich meals, decrease and thus hyperinsulinemia has been hypothesized postprandial testosterone levels. Our laboratory has to play a pathogenic role in women with PCOS also observed a significant reduction in total testo- (Bergh et al. 1993). In hyperandrogenic women or sterone (–22%) and free testosterone (–23%) after a those with PCOS, testosterone responses to glucose fat-rich meal in healthy men (Volek et al. 2001). tolerance tests also tend to decrease in a similar Another study compared testosterone responses manner as healthy controls (Falcone et al. 1990; to meals with different sources of protein (soy Tropeano et al. 1994) unless the women are also versus meat), amounts of fat (lean versus fatty meat) insulin resistant (Smith, S. et al. 1987) or obese and sources of fat (animal versus vegetable) (Habito (Tiitinen et al. 1990), and then there is a slight et al. 2001). The decrease in testosterone was greater increase in testosterone. after a low-fat meal consisting of lean meat (–22%) In summary, testosterone levels consistently drop compared to tofu (–15%); a meal consisting of lean after feeding and there is evidence this is dependent meat (–22%) compared to lean meat cooked with on the composition of the meal, particularly fat animal fat meal (–9%); and a meal consisting of lean content. Insulin may play a role in explaining some meat cooked with vegetable fat (–17%) compared to of the variability in testosterone responses to meals vegetable oil (–9%). Although these results do not because in men insulin and testosterone tend to necessarily agree with the findings mentioned exhibit an inverse relation whereas in women there above that only fat-rich meals reduce testosterone, is a positive association, especially in women with the study does provide further evidence that the PCOS (Haffner et al. 1994). However, findings from composition of meals, particularly the amount and studies using the euglycemic hyperinsulinemic type of fat, influences the circulating testoster- clamp procedure cast doubt on the hypothesis that one response to meals. All the meals resulted in insulin mediates the testosterone response to feed- decreases in testosterone, increases in LH and no ing because acute elevations in insulin were shown changes in SHBG. to have no effect on total or free testosterone levels A number of studies have examined the testo- in healthy normal-weight men (Ebeling et al. 1995; sterone response to an oral glucose tolerance test. Pasquali et al. 1997) or women (Diamond et al. 1991). The majority of these studies have been in women, The mechanism by which fat-rich meals decrease 432 chapter 29

postprandial testosterone concentrations may be compared the post-exercise testosterone responses related to the elevated chylomicrons or fatty acids to a mixed meal, an isocaloric beverage of similar after a fat-rich meal because they inhibit LH-stimu- nutrient content, and an isocaloric carbohydrate lated testosterone production in isolated Leydig beverage consumed immediately and 2, 4 and 7 h cells (Meikle et al. 1989, 1990). It is possible that after exercise. Compared to a placebo, post-resist- specific nutrients may interact with regions of the ance exercise testosterone levels were lower during testes and impact on the modulation of testosterone. all the meals at 0.5, 2.5, 4.5 and 8 h post-exercise. Alteration in the testicular plasma membrane and There is one study that did not show a difference in changes in the responsiveness of Leydig cells and free testosterone responses to resistance exercise subsequent testosterone synthesis as a result of between carbohydrate and a water placebo con- ingestion of different compositions of lipids has sumed during exercise (Tarpenning et al. 2001). been reported in rats (Sebokova et al. 1988, 1990). Collectively, these studies indicate that pre- and post-exercise meals decrease post-exercise testo- exercise-induced responses to feeding sterone levels compared to fasting. The decrease could be due in part to a decrease in the synthesis/ Resistance exercise has been shown to result in secretion of testosterone and/or an increase in acute elevations of testosterone that peak early metabolic clearance. Chandler et al. (1994) showed after exercise and return to baseline by about 60 min that the decrease in post-exercise testosterone was post-exercise; pre- and post-exercise meals alter not associated with a decrease in LH arguing this response (Kraemer et al. 1998). Our laboratory against a decrease in the rate of testosterone secre- showed that a protein and carbohydrate beverage tion, however there could still be a decrease in the consumed 2 h before and immediately after a bout testicular responsiveness to LH. Since post-exercise of whole body resistance exercise resulted in an meals increase muscle-specific protein synthesis increase in testosterone immediately after exercise during recovery (Tipton et al. 1999a, 2001; Rasmussen followed by a sharp decrease to values that were et al. 2000), the lower testosterone levels could be significantly below baseline (Kraemer et al. 1998). due in part to increased uptake in active skeletal In this study, the exact same exercise and supple- muscle. In support of this hypothesis, recent work in mentation protocol was performed on 3 consecutive our laboratory has shown that the meal-induced days. The greater reduction in post-exercise testo- decrease in testosterone after resistance exercise sterone with pre- and post-exercise protein and corresponds with an increase in skeletal muscle carbohydrate intake compared to placebo was androgen receptor content measured 60 min after observed on all 3 days, emphasizing the consistency exercise (unpublished observations). and reproducibility of the response (Kraemer et al. 1998). Chandler et al. (1994) examined the post- Cortisol exercise testosterone response to supplements con- taining either protein alone, carbohydrate alone, or Cortisol is an adrenal steroid hormone that is a combination of protein and carbohydrate con- regulated by pituitary adrenocorticotropin (ACTH), sumed immediately and 2 h after a bout of resist- which in turn is under the influence of hypothal- ance exercise in healthy men. In agreement with amic corticotropin-releasing hormone (CRH). This findings from our study, testosterone had decreased hypothalamic–pituitary–adrenal (HPA) axis is sens- to values below baseline by 30 min post-exercise itive to a variety of different stressors including during all the supplement treatments compared to feeding. At the whole body level, cortisol increases placebo. Testosterone values remained significantly protein breakdown but the effects in skeletal muscle below baseline 5–6 h whereas testosterone returned are unclear (Rooyackers & Nair 1997). Physiological to baseline shortly after exercise and stabilized concentrations of cortisol increase lipolysis in adip- throughout the recovery period (Chandler et al. ose tissue but the effects are less potent than GH 1994). In yet another study, Bloomer et al. (2000) (Divertie et al. 1991; Djurhuus et al. 2002). impact of diet on hormones 433

postprandial response to feeding after the glucose load compared to all other trials and the stress-induced cortisol response was posit- The effects of feeding on cortisol are complicated by ively correlated with glucose changes before the the fact that cortisol has a distinct diurnal variation stress test. Since restoration of the stress-induced and is sensitive to a number of internal and external HPA response after fasting was limited to the gluc- stressors. In general, feeding has been shown to ose load, which was the only trial to increase blood increase cortisol levels and the response is particu- glucose, the authors suggest that HPA responsive- larly evident at the noon meal (Follenius et al. 1982; ness is under control of centers sensing blood Knoll et al. 1984), which may serve to synchronize glucose levels, presumably in the hypothalamus the diurnal rhythm in HPA axis activity (Leal & (Gonzala-Bono et al. 2002). Moreira 1997). Although few studies have addressed Part of the discrepancy between the effects of the influence of meal composition on the HPA axis, meal composition on postprandial cortisol may be it appears that protein has the greatest stimulatory explained by differences in the pattern of body fat effect on cortisol (Slag et al. 1981; Ishizuka et al. 1983; distribution. Vicennati et al. (2002) recently showed Gibson et al. 1999) whereas carbohydrate feeding that women with abdominal fat distribution have inhibits cortisol (Jezova-Repcekova et al. 1980). a higher cortisol response to a high-carbohydrate In healthy men, salivary cortisol was shown to meal compared to a high-protein/high-fat meal and increase substantially after a mid-day protein-rich that women with a peripheral fat distribution meal (550 kcal [2299 kJ] with 39% protein) com- demonstrate the opposite effect, which was more in pared to fasting and this response although variable line with the response of a control normal-weight from subject to subject was highly reproducible group of women. within subjects (Gibson et al. 1999). In healthy women, Postprandial cortisol measurements must be salivary cortisol was also significantly increased interpreted in the context of the normal diurnal after a mid-day high-protein (630 kcal [2633 kJ] and variation of the hormone. For example, a morning 32% protein) compared to an isocaloric low-protein glucose tolerance test results in a decrease in cortisol (5% protein) meal, which resulted in a decrease for several hours after the meal (Walker et al. 2000; in cortisol (Gibson et al. 1999). Other studies have Reynolds et al. 2001, 2003). However, postprandial shown that meals containing 20–40% protein cortisol values are higher compared to the normal enhance postprandial cortisol levels compared to diurnal drop in cortisol in the morning hours meals with high carbohydrate or fat content and (Reynolds et al. 2001). that protein-free glucose or fat-rich meals do not The findings above showing a stimulatory effect stimulate cortisol release (Slag et al. 1981; Ishizuka of glucose on cortisol secretion are consistent with et al. 1983). The stimulation of cortisol by protein the findings of Gonzalez-Bono et al. (2002) and the may be due to specific essential amino acids such as notion that the HPA axis is responsive to increases tyrosine and tryptophan (Ishizuka et al. 1983). in circulating glucose, but the precise mechanisms The findings from the studies mentioned above regulating this effect remain unresolved. A role of α that protein has a stimulatory effect on cortisol but 1-adrenorecptors modulating ACTH secretion has carbohydrates have a negligible effect appear to be been proposed to mediate postprandial cortisol in conflict with other work that has shown glucose responses (Al-Damluji et al. 1987). In these experi- α to be important in the regulation of the HPA axis. ments, infusion of methoxamine (an 1-adrenore- Gonzalez-Bono et al. (2002) fed subjects either glu- ceptor agonist) enhanced and thymoxamine (an α cose, protein, fat or water and then 45 min later were 1-adrenoreceptor antagonist) attenuated the ACTH exposed to a psychosocial stress test. There were no and cortisol response to a standard meal. These differences in cortisol levels measured 45 min after authors also showed that patients with a recent pitu- the meals, which the authors suggest may have been itary ACTH deficiency and normal adrenal glands due to the timing of the meals. However, cortisol showed no ACTH or cortisol rise after a standard levels after the stress test were significantly higher meal (Al-Damluji et al. 1987). In further support of a 434 chapter 29

central mechanism, postprandial ACTH and cortisol exercise (Nehlsen-Cannarella et al. 1997; Henson levels have been shown to be positively correlated et al. 1998; Nieman 1998). However, the favorable after a meal (Vicennati et al. 2002). However, not all effects of carbohydrate feeding on immune function data point to a hypothalamic–pituitary mediated during and after exercise may not be mediated mechanism to explain postprandial cortisol secre- through a lower cortisol response (Green et al. 2003), tion (Fehm et al. 1983). In summary, cortisol tends to but rather some other mechanism such as better increase after feeding, especially protein, but the maintenance of plasma glucose or glutamine levels, response is dependent on the timing and composi- an important fuel source for immune cells (Bacurau tion of the meal, fat distribution pattern and must et al. 2002). A recent study showed that carbohy- be compared to the normal diurnal variation of the drate intake reduced the plasma interleukin-6 (IL-6) hormone. response to exercise, but IL-6 mRNA expression in skeletal muscle was not affected suggesting that exercise-induced responses to feeding carbohydrate feeding during exercise attenuates IL- 6 production by tissues other than skeletal muscle A large number of studies in the last 25 years have (Starkie et al. 2001). examined the effects of feeding before, during and after exercise on cortisol responses. The majority Insulin of these have examined carbohydrate feeding and a few have studied how different combinations of Insulin is a peptide hormone secreted by the pan- macronutrients influence cortisol. In terms of stud- creas. In skeletal muscle, insulin has anabolic effects ies that measured the cortisol response to resistance by increasing amino acid uptake and protein synthe- exercise, the majority indicate that intake of carbo- sis and inhibiting protein breakdown (Rooyackers hydrate or carbohydrate combined with protein & Nair 1997). Insulin is generally accepted as a stimu- before and after resistance exercise does not alter lator of protein synthesis only when adequate the cortisol response compared to placebo (Kraemer amino acids are available (Kimball & Jefferson 2002; et al. 1998; Bloomer et al. 2000; Koch et al. 2001; Kimball et al. 2002). Insulin has long been consid- Williams et al. 2002). One study showed that carbo- ered the most important regulator of adipose tissue hydrate intake during an acute bout of resistance metabolism. Adipose tissue lipolysis is exquisitely exercise significantly blunted the cortisol response sensitive to insulin at physiological concentrations (Tarpenning et al. 2001). This study further showed (Jensen et al. 1989). Small-to-moderate decreases that the reduction in post-resistance exercise cortisol in insulin can increase lipolysis several-fold, the was significantly related to increases in muscle fiber response being virtually immediate. Insulin also hypertrophy. stimulates lipogenesis by increasing glucose uptake Studies that measured the cortisol response to and activating lipogenic and glycolytic enzymes prolonged endurance or high-intensity exercise are (Kersten 2001). generally mixed with several showing that carbohy- drate intake before and during exercise results in postprandial response to feeding lower cortisol responses (Mitchell et al. 1990, 1998; Murray et al. 1991, 1995; Deuster et al. 1992; Nieman Carbohydrate ingestion leads to an increase in blood et al. 1998, 2003; Utter et al. 1999; Bacurau et al. 2002; glucose and a relatively similar increase in insulin Bishop et al. 2002; Green et al. 2003) whereas others concentrations (Nuttall et al. 1985). A meal rich in fat have shown no change in cortisol with carbohydrate results in lower insulin responses compared to feeding (Bonen et al. 1980; Tsintzas et al. 1996; Bishop meals rich in either carbohydrate (Nuttall et al. 1985) et al. 1999, 2001; Henson et al. 2000; Miller et al. 2002). or protein (Ullrich et al. 1985). Carbohydrate inges- The reduction in the exercise-induced cortisol tion results in elevated glucose and insulin levels response after carbohydrate supplementation may that depend primarily on the glycemic index or, contribute to a more favorable immune response to more precisely, the glycemic load of the food. The impact of diet on hormones 435

glycemic index was developed in 1981 and refers (1.2 g·kg–1·h–1), additional protein does not further to the effect of standard amounts of individual enhance the rate of glycogen resynthesis (Jentjens foods containing 50 g of available carbohydrate on et al. 2001). Enhanced insulin levels resulting from blood glucose compared with that of a control food carbohydrate combined with protein could be (usually white bread or glucose) (Jenkins et al. 1981). expected to have a favorable effect on net protein Comprehensive tables describing the glycemic balance because insulin is generally accepted as a index of over 1000 foods exist (Foster-Powell et al. stimulator of protein synthesis when adequate 2002). To account for differences in the amount of amino acids are available (Kimball & Jefferson 2002; carbohydrate in foods or meals, the concept of Kimball et al. 2002). glycemic load was introduced, which is the pro- duct of glycemic index and carbohydrate amount Implications of diet-induced hormonal (Salmeron et al. 1997). The glycemic load can predict changes on body composition glucose and insulin responses to individual foods across a wide range of portion sizes (Brand-Miller A goal of many athletes and individuals in general et al. 2003). Certain amino acids can increase insulin is to improve body composition, that is, to decrease and thus carbohydrate combined with protein can adipose tissue (fat mass) and/or increase skeletal enhance the insulin response (Nuttall et al. 1984, muscle tissue (lean body mass). Hormones are 1985). major regulators of protein turnover in skeletal muscle and triacylglycerol turnover in adipose exercise-induced responses to feeding tissue. As discussed above, quantity, quality and timing of dietary intake modulates many of the Given the synergistic effects of carbohydrate and hormones that regulate protein and triacylglycerol protein on insulin responses, there has been some turnover. Thus, from a theoretical perspective feed- interest in combining protein with carbohydrate to ing can be viewed as a strategy to alter the hormonal maximize insulin secretion in the hopes of enhanc- environment to alter protein and lipid balance, ing post-exercise glycogen resynthesis (Zawadzki which over time could lead to decreased fat mass et al. 1992; van Loon et al. 2000a) and protein and/or increased lean body mass. anabolism (Rasmussen et al. 2000; Tipton et al. 2001). Predicting the impact of a defined meal on protein In a study designed to determine the optimal and lipid balance and eventually body composition insulinotropic mixture of protein and carbohydrate, is however quite complex. Consider that most hor- it was determined that carbohydrate (0.7 g·kg–1·h–1) mones, at least those discussed in this article, affect consisting of 50% glucose and 50% maltodextrin both protein and lipid balance through multiple combined with protein (0.35 g·kg–1·h–1) consisting mechanisms (Fig. 29.2), and that unlike the situation of 50% wheat protein hydrolysate, 25% free leucine in vitro, hormones change in vivo simultaneously in and 25% free phenylalanine was most effective (van response to feeding. Thus, the interactions among Loon et al. 2000b). Although the effects of carbohy- various hormones determine the overall response in drate combined with protein on glycogen resyn- both skeletal muscle and adipose tissue. Because of thesis after resistance exercise are unknown, there these multiple interactions, translation of the infor- does appear to be a beneficial effect of ingesting mation known about postprandial and exercise- some protein and/or amino acids in combination induced hormonal responses into specific dietary with carbohydrate on glycogen resynthesis after recommendations aimed at improving body com- submaximal cycling exercise compared to the same position is a complicated task. Nutritional strategies amount of carbohydrate only (Zawadzki et al. 1992; during the time periods before and after exercise van Loon et al. 2000a). This effect is likely due to have been particularly studied and shown to have greater insulin secretion after combined carbo- relatively predictable affects on hormones and pro- hydrate and protein intake although it is pointed tein and lipid metabolism (Volek 2004), so this out that when carbohydrate intake is very high is an appropriate starting point for discussing the 436 chapter 29

lipolysis amino acid oxidation Adipose lipogenesis PRO synthesis Skeletal tissue LPL expression PRO breakdown muscle balance GH balance (–) (+)

lipolysis PRO synthesis glucose transport/oxidation PRO breakdown? IGF-I (+) (+)

Lipogenesis PRO synthesis PRO synthesis LPL expression PRO breakdown? TAG FA Testosterone PRO AA (–) (+)

Lipolysis PRO breakdown lipolysis LPL expression PRO breakdown Cortisol (–) (–) lipolysis lipogenesis amino acid uptake glucose uptake/oxidation PRO synthesis LPL expression PRO breakdown Insulin (+) (+)

Fig. 29.2 Hormonal regulation of body composition. Several hormones including growth hormone (GH), insulin-like growth factor I (IGF-I) and its binding proteins, testosterone, cortisol and insulin regulate both skeletal muscle protein (PRO) and adipose tissue balance through multiple mechanisms. Feeding affects the postprandial and exercise-induced concentrations of these hormones, which regulate PRO and amino acid (AA) cycling (i.e. protein synthesis/breakdown) in skeletal muscle and triacylglycerol (TAG) and fatty acid (FA) cycling (i.e. lipogenesis/lipolysis) in adipose tissue. These hormones do not change in isolation, rather the simultaneous interaction among hormones must be considered in terms of their ultimate impact on protein and lipid balance and lean body mass and fat mass.

implications of diet-induced hormonal responses. protein synthesis and protein breakdown, the bal- Theoretically, increases in GH, IGF-I and testoster- ance of which determines the anabolic response of one would all have favorable effects on skeletal muscle to resistance exercise. Quantity, quality and muscle and adipose tissue balance; that is, elevations timing of dietary intake after exercise influences in these hormones would contribute to increased nutrient and hormone availability at specific recep- lean body mass and decreased fat mass. Elevated tors at target tissues (i.e. skeletal muscle and adip- cortisol would be negative in terms of skeletal ose tissue). The contraction-induced mechanical muscle but might contribute to enhanced lipolysis. and chemical events in muscle interact with nutrient In contrast, elevated insulin would have positive and hormonal signals to regulate enzymes (e.g. effects on skeletal muscle and negative effects on glycogen synthase) and mediate gene level tran- lipid balance. scription and translation of proteins (Turner et al. 1988). Exercise also results in increased blood flow to the active skeletal muscles, which has important Nutrients, hormones, and skeletal muscle implications for pre-exercise meals that could protein balance enhance hormone interactions and the delivery of Exercise, particularly resistance training, stimulates nutrients to target receptors during and after exer- impact of diet on hormones 437

cise. The combined effect of muscular contraction lators of skeletal muscle protein synthesis (Kimball and the increased availability of hormones and & Jefferson 2002). Recent work indicates that it is nutrients have the potential to enhance the rate of the extracellular levels of essential amino acids in amino acid and glucose uptake, and promote the blood that regulate muscle protein synthesis an anabolic environment. Nutrient availability is as opposed to intramuscular amino acids (Bohe critical during this time as evidenced by studies et al. 2003). Whether these acute changes in skeletal showing little glycogen resynthesis (Pascoe et al. muscle protein metabolism reflect chronic changes 1993; Roy & Tarnopolsky 1998) and a negative pro- in lean body mass is unknown. tein balance (Biolo et al. 1995, 1997) in the absence of A recent study in elderly men investigated the nutritional intake after exercise. An anabolic nutri- effect of timing of protein and carbohydrate supple- tional and hormonal milieu favorably affects the mentation on muscle size and strength responses to balance of protein synthesis/degradation, which 12 week of resistance training (Esmarck et al. 2001). sets the stage for greater protein accretion and The supplement (10 g protein, 7 g carbohydrate) muscle fiber hypertrophy with chronic resistance was consumed immediately or 2 h after each train- training. The resulting increased force production ing session. The group who ingested the supple- capabilities can improve the intensity of subsequent ment immediately after exercise had significantly workouts and further enhance the resistance exer- greater increases in lean body mass, muscle fiber cise stimuli in a feedforward fashion. Eventually, area and quadriceps femoris area. These data indic- the repeated exposures to resistance exercise work- ate that altering the timing of calories, which affects outs will lead to measurable increases in muscle the nutrient and hormonal milieu, can impact strength and size. chronic adaptation to training. Specifically, early Feeding has been shown to be a simple and effect- intake of protein and carbohydrate after a workout ive method to alter rates of protein synthesis is more effective at increasing skeletal muscle (Svanberg et al. 2000). Infusion of amino acids or hypertrophy and lean body mass than a supplement exogenous administration of amino acids with consumed later. These findings are in conflict with or without carbohydrate stimulates protein syn- a study that showed no differences in acute meas- thesis after exercise (Bennet et al. 1989; Biolo et al. ures of protein balance when protein was ingested 1997; Tipton et al. 1999a, 1999b; Rasmussen et al. 1 or 3 h after exercise in healthy young subjects 2000). Carbohydrate intake after resistance exercise (Rasmussen et al. 2000). This apparent discrepancy decreases measures of protein breakdown and related to timing of protein ingestion highlights the slightly increases fractional muscle protein syn- importance of linking acute studies that measure thetic rate, which is likely due primarily to increases protein kinetics to long-term training studies that in insulin (Roy et al. 1997). Protein synthesis was assess outcome measures related to muscle size. stimulated ∼ 400% above pre-exercise values when The only study to date that has done this is work a protein and carbohydrate supplement (6 g essen- showing that carbohydrate intake during resistance tial amino acids and 35 g sucrose) was consumed exercise was an effective strategy to lower cortisol 1 or 3 h after resistance exercise (Rasmussen et al. levels and that the magnitude of cortisol reduction 2000). Consumption of this same protein and car- was directly related to measures of muscle fiber bohydrate supplement immediately before exercise hypertrophy (Tarpenning et al. 2001). resulted in increased amino acid delivery to muscle Practically no work has examined how acute and even greater net muscle protein synthesis diet-induced changes in testosterone, GH or IGF-I (Tipton et al. 2001). Essential amino acids have been might contribute to the acute changes in protein bal- shown to be primary regulators of muscle protein ance and chronic changes in lean body mass. Our synthesis with little contribution from non-essential laboratory has discovered that circulating testoster- amino acids (Smith, K. et al. 1998; Tipton et al. 1999a, one is consistently modulated by nutrient intake 1999b). The branched-chain amino acids, particu- and that a mixed meal alters post-exercise skeletal larly leucine, appear to be the most important stimu- muscle androgen content, indicating a possible link 438 chapter 29

between diet, testosterone and skeletal muscle pro- have increased dramatically in recent years in part tein metabolism. because they are advertised to be more effective in In summary, there appears to be an interaction promoting weight and fat loss. There were several between increased availability of amino acids, in- studies performed in the 1960s and 1970s that creased insulin and possibly other hormones after showed greater weight loss with a very low- exercise and the timing of supplement ingestion (i.e. carbohydrate compared to a low-fat diet, even when immediately before exercise) may be important to diets contained the same energy content (Rabast maximize the anabolic response (Esmarck et al. et al. 1979), suggesting a metabolic advantage (i.e. a 2001; Tipton et al. 2001). Consumption of a protein greater weight loss if carbohydrates are low com- and carbohydrate supplement at times around exer- pared to isoenergetic diets of different macronutri- cise (i.e. immediately before and immediately after ent composition). Very little follow-up work was exercise) may provide the ideal anabolic situation done until recently, as evidenced by several publica- for muscle growth. There does not appear to be any tions in 2003 again showing greater weight loss with value in including fat in the pre- or post-exercise very low-carbohydrate diets 3–6 months in dura- meals from a hormonal or metabolic perspective. tion (Brehm et al. 2003; Foster et al. 2003; Samaha et al. 2003; Sondike et al. 2003). Some early reports show that very low-carbohydrate diets result in Nutrients, hormones and adipose tissue preferential loss of fat and preservation of lean body triacylglycerol balance mass (Benoit et al. 1965; Young et al. 1971; Willi et al. Many of the studies designed to examine how feed- 1998; Meckling et al. 2002), suggestive of a nutrient ing affects hormonal responses to exercise have partitioning effect. In accordance with this notion, focused on enhancing protein balance or stimulat- we recently reported that a free-living 6-week ing glycogen formation with little attention directed very low-carbohydrate diet resulted in significant to the impact on adipose tissue lipid metabolism. decreases in fat mass and increases in lean body High-carbohydrate intakes are often encouraged to mass in normal-weight men (Volek et al. 2002). In a stimulate glycogen synthase and glycogen forma- follow-up study we showed that a very low-carbo- tion and even protein balance after exercise. The hydrate diet resulted in twofold greater whole body hormonal environment after carbohydrate feeding, fat loss and threefold greater fat loss in the trunk dominated by a surge in insulin, would inhibit lipo- region compared to a low-fat diet (Volek et al. 2004). lysis and potentially stimulate lipogenesis in adipose Carbohydrate restricted diets remain controversial tissue, a scenario not desirable if fat loss is a major (Blackburn et al. 2001; Freedman et al. 2001), yet goal. A better alternative would be to focus on high evidence indicates they are very effective in the quality protein with all the essential amino acids short-term and are not associated with any adverse and perhaps low-glycemic carbohydrates during effects (Volek et al. 2000, 2003; Sharman et al. 2002; the post-exercise period to stimulate protein synthe- Volek & Westman 2002). sis while keeping insulin low to prevent inhibition Very low-carbohydrate diets are low glycemic of lipolysis. Protein intake after exercise may have a load diets. Although the mechanisms by which very slight effect on enhancing GH, which would further low-carbohydrate diets benefit weight and fat loss support lipolysis and decrease lipogenesis. Sub- has not elucidated, a reduction in insulin is probably sequent meals throughout the day should focus on important in explaining a portion of the greater fat foods with a low-to-moderate glycemic load to keep loss (Volek et al. 2002). Inhibition of lipolysis occurs insulin levels low. Since the addition of protein and at relatively low concentrations of insulin with a fat to a meal slow the pace of digestion and lower half-maximal effect occurring at a concentration of the glycemic load, including quality sources of pro- 12 pmol·L–1 and a maximal effect at a concentration tein and healthy unsaturated fat should also be a of about 200–300 pmol·L–1 (Jensen et al. 1989). Thus, priority. even small reductions in insulin may be permissive The popularity of diets that restrict carbohydrate to mobilization of body fat on a low glycemic diet. A impact of diet on hormones 439

metabolic advantage, possibly driven by increased Given the important regulatory functions of hor- protein turnover to supply alanine for gluconeogen- mones on muscle and lipid balance, the influence of esis, is another plausible hypothesis to explain diet on endocrine function is arguably of consider- greater weight loss on very low-carbohydrate diets able importance. Optimizing the hormonal environ- (Feinman & Fine 2004). Although the benefits of ment in favor of an anabolic profile during the following a low glycemic load diet or a low- recovery period between exercise sessions would be carbohydrate, moderate protein and fat diet on fat advantageous for individuals attempting to max- loss look promising, the ideal diet for fat loss and imize gains in muscle size and strength. Specific- general health remains controversial. However, ally, by elevating the primary anabolic hormones overwhelming evidence indicates controlling insulin involved in muscle tissue growth (i.e. testosterone, levels through diet is important. GH, insulin and IGF-I) and/or decreasing major catabolic hormones (i.e. cortisol) a hormonal milieu Summary maximizing protein balance and muscle hyper- trophy would be created. If fat loss is a goal, then it It is clear that feeding and meal composition sig- would be advantageous to keep insulin levels low nificantly alter postprandial and post-exercise hor- and GH and testosterone levels elevated. From a monal responses. Yet the importance of acute and hormonal perspective, meals with a low glycemic chronic effects of these hormonal changes on skel- load with moderate amounts of quality protein and etal muscle and adipose tissue balance and, ulti- healthy unsaturated fat would be conducive for fat mately, body composition remain largely unknown. loss and general health.

References

Aizawa, H. & Niimura, M. (1996) Mild during mixed amino acid infusion: amino acids enhances the metabolic insulin resistance during oral glucose studies of incorporation of [1- effect of exercise on muscle protein. tolerance test (OGTT) in women with 13C]leucine. Clinical Science 76, 447–454. American Journal of Physiology. acne. Journal of Dermatology 23, 526–529. Benoit, F.L., Martin, R.L. & Watten, R.H. Endocrinology and Metabolism 273, Al-Damluji, S., Iveson, T., Thomas, J.M. (1965) Changes in body composition E122–E129. et al. (1987) Food-induced cortisol during weight reduction in obesity. Bishop, N.C., Blannin, A.K., Robson, P.J., α secretion is mediated by central 1 Annals of Internal Medicine 63, 604–612. Walsh, N.P. & Gleeson, M. (1999) The adrenoceptor modulation of pituitary Bereket, A., Wilson, T.A., Blethen, S.L. et al. effects of carbohydrate supplementation ACTH secretion. Clinical Endocrinology (1996) Effect of short-term fasting on on immune responses to a soccer- 26, 629–636. free/dissociable insulin-like growth specific exercise protocol. Journal of Anthony, T.G., Anthony, J.C., Lewitt, M.S., factor I concentrations in normal human Sports Sciences 17, 787–796. Donovan, S.M. & Layman, D.K. (2001) serum. Journal of Clinical Endocrinology Bishop, N.C., Blannin, A.K., Walsh, N.P. Time course changes in IGFBP-1 after and Metabolism 81, 4379–4384. & Gleeson, M. (2001) Carbohydrate treadmill exercise and postexercise Bergh, C., Carlsson, B., Olsson, J.H., beverage ingestion and neutrophil food intake in rats. American Journal of Selleskog, U. & Hillensjo, T. (1993) degranulation responses following o Physiology. Endocrinology and Metabolism Regulation of androgen production in cycling to fatigue at 75% V 2max. 280, E650–E656. cultured human thecal cells by insulin- International Journal of Sports Medicine Bacurau, R.F., Bassit, R.A., Sawada, L. et al. like growth factor I and insulin. Fertility 22, 226–231. (2002) Carbohydrate supplementation and Sterility 59, 323–331. Bishop, N.C., Gleeson, M., Nicholas, C.W. during intense exercise and the immune Bernardi, F., Petraglia, F., Seppala, M. & Ali, A. (2002) Influence of response of cyclists. Clinical Nutrition et al. (1999) GH, IGFBP-1, and IGFBP-3 carbohydrate supplementation on (Edinburgh, Lothian) 21, 423–429. response to oral glucose tolerance test in plasma cytokine and neutrophil Baker, H.W., Best, J.B., Burger, H.G. & perimenopausal women: no influence of degranulation responses to high Cameron, D.P. (1972) Plasma human body mass index. Maturitas 33, 163–169. intensity intermittent exercise. growth hormone levels in response to Biolo, G., Maggi, S.P., Williams, B.D., International Journal of Sport Nutrition and meals: a reappraisal. Australian Journal of Tipton, K.D. & Wolfe, R.R. (1995) Exercise Metabolism 12, 145–156. Experimental Biology and Medical Science Increased rates of muscle protein Blackard, W.G., Hull, E.W. & Lopez, A. 50, 715–724. turnover and amino acid transport after (1971) Effect of lipids on growth Bennet, W.M., Connacher, A.A., resistance exercise in humans. American hormone secretion in humans. Journal Scrimgeor, C.M., Smith, K. & Bennie, Journal of Physiology 268, E514–E520. of Clinical Investigation 50, 1439–1443. M.J. (1989) Increase in anterior tibialis Biolo, G., Tipton, K.D., Klein, S. & Wolfe, Blackburn, G.L., Phillips, J.C. & Morreale, muscle protein synthesis in healthy man R.R. (1997) An abundant supply of S. (2001) Physician’s guide to popular 440 chapter 29

low-carbohydrate weight-loss diets. Chromiak, J.A. & Antonio, J. (2002) Use K.H. (1983) The meal-related peak in Cleveland Clinic Journal of Medicine 68, of amino acids as growth hormone- plasma cortisol is not mediated by 761–774. releasing agents by athletes. Nutrition 18, radioimmunoassayable ACTH. Bloomer, R.J., Sforzo, G.A. & Keller, B.A. 657–661. Clinical Physiology and Biochemistry 1, (2000) Effects of meal form and De Pergola, G. (2000) The adipose tissue 329–333. composition on plasma testosterone, metabolism: role of testosterone and Feinman, R.D. & Fine, E.J. (2004) ‘A calorie cortisol, and insulin following resistance dehydroepiandrosterone. International is a calorie’ violates the second law of exercise. International Journal of Sport Journal of Obesity and Related Metabolic thermodynamics. Nutrition Journal 28, 9. Nutrition and Exercise Metabolism 10, Disorders 24, S59–S63. Fineberg, E.S., Horland, A.A. & Merimee, 415–424. Deuster, P.A., Singh, A., Hofmann, A., T.J. (1972) Free fatty acid concentrations Bohe, J., Low, A., Wolfe, R.R. & Rennie, Moses, F.M. & Chrousos, G.C. (1992) and growth hormone secretion in man. M.J. (2003) Human muscle protein Hormonal responses to ingesting water Metabolism 21, 491–498. synthesis is modulated by extracellular or a carbohydrate beverage during a 2 h Follenius, M., Brandenberger, G. & Hietter, not intracellular amino acid availability: run. Medicine and Science in Sports and B. (1982) Diurnal cortisol peaks and their a dose response study. Journal of Exercise 24, 72–79. relationships to meals. Journal of Clinical Physiology 552(Pt. 1), 315–324. Diamond, M.P., Grainger, D.A., Laudano, Endocrinology and Metabolism 55, Bonen, A., Belcastro, A.N., MacIntyre, K. & A.J., Starick-Zych, K. & DeFronzo, R.A. 757–761. Gardner, J. (1980) Hormonal responses (1991) Effect of acute physiological Foster, G.D., Wyatt, H.R., Hill, J.O. et al. during intense exercise preceded by elevations of insulin on circulating (2003) A randomized trial of a low- glucose ingestion. Canadian Journal of androgen levels in nonobese women. carbohydrate diet for obesity. New Applied Sport Sciences 5, 85–90. Journal of Clinical Endocrinology and England Journal of Medicine 348, Brand-Miller, J.C., Thomas, M., Swan, V. Metabolism 72, 883–887. 2082–2090. et al. (2003) Physiological validation of Divertie, G.D., Jensen, M.D. & Miles, J.M. Foster-Powell, K., Holt, S.H. & Brand- the concept of glycemic load in lean (1991) Stimulation of lipolysis Miller, J.C. (2002) International table young adults. Journal of Nutrition 133, in humans by physiological of glycemic index and glycemic load 2728–2732. hypercortisolemia. Diabetes 40, values, 2002. American Journal of Clinical Brehm, B.J., Seely, R.J., Daniels, S.R. & 1228–1232. Nutrition 76, 5–56. D’Alessio, D.A. (2003) A randomized Djurhuus, C.B., Gravholt, C.H., Nielsen, S. Freedman, M.R. (2001) Popular diets: a trial comparing a very low-carbohydrate et al. (2002) Effects of cortisol on lipolysis scientific review. Obesity Research 1, diet and a calorie-restricted low-fat diet and regional interstitial glycerol levels in 1S–40S. on body weight and cardiovascular risk humans. American Journal of Physiology. Frystyk, J., Grofte, T., Skjaerbaek, C. & factors in healthy women. Journal of Endocrinology and Metabolism 283, Orskov, H. (1997) The effect of oral Clinical Endocrinology and Metabolism 88, E172–E177. glucose on serum free insulin-like 1617–1623. Ebeling, P., Stenman, U.H., Seppala, M. growth factor-I and -II in health adults. Cappon, J.P., Ipp, E., Brasel, J.A. & Cooper, & Koivisto, V.A. (1995) Acute Journal of Clinical Endocrinology and D.M. (1993) Acute effects of high fat hyperinsulinemia, androgen Metabolism 82, 3124–3127. and high glucose meals on the growth homeostasis and insulin sensitivity in Frystyk, J., Nyholm, B., Skjaerbaek, C. hormone response to exercise. Journal of healthy man. Journal of Endocrinology et al. (2003) The circulating IGF system Clinical Endocrinology and Metabolism 76, 146, 63–69. and its relationship with 24-h glucose 1418–1422. Ebenbichler, C.F., Kirchmair, R., Egger, C. regulation and insulin sensitivity in Cappon, J.P., Brasel, J.A., Mohan, S. & & Patsch, J.R. (1995) Postprandial state healthy subjects. Clinical Endocrinology Cooper, D.M. (1994) Effect of brief and atherosclerosis. Current Opinion in 58, 777–784. exercise on circulating insulin-like Lipidology 6, 286–290. Galbo, H., Holst, J.J. & Christensen, N.J. growth factor I. Journal of Applied Esmarck, B., Andersen, J.L., Olsen, S. (1979) The effects of different diets and Physiology 76, 2490–2496. et al. (2001) Timing of post-exercise of insulin on the hormonal response to Cara, J.F. & Rosenfield, R.L. (1988) protein intake is important for muscle prolonged exercise. Acta Physiologica Insulin-like growth factor I and insulin hypertrophy with resistance training in Scandinavica 107, 19–32. potentiate luteinizing hormone-induced elderly humans. Journal of Physiology 535, Gibson, E.L., Checkley, S., Papadopoulos, androgen synthesis by rat ovarian 301–311. A. et al. (1999) Increased salivary cortisol thecal-interstitial cells. Endocrinology Etherton, T.D. (2000) The biology of reliably induced by a protein-rich 123, 733–739. somatotropin in adipose tissue growth midday meal. Psychosomatic Medicine 61, Carli, G., Bonifazi, M., Lodi, L. et al. (1992) and nutrient partitioning. Journal of 214–224. Changes in the exercise-induced Nutrition 130, 2623–2625. Gonzalez-Bono, E., Rohleder, N., hormone response to branched chain Falcone, T., Finegood, D.T., Fantus, I.G. & Hellhammer, D.H., Salvador, A. & amino acid administration. European Morris, D. (1990) Androgen response Kirschbaum, C. (2002) Glucose but not Journal of Applied Physiology and to endogenous insulin secretion during protein or fat load amplifies the cortisol Occupational Physiology 64, 272–277. the frequently sampled intravenous response to psychosocial stress. Chandler, R.M., Byrne, H.K., Patterson, glucose tolerance test in normal and Hormones and Behavior 41, 328–333. J.G. & Ivy, J.L. (1994) Dietary hyperandrogenic women. Journal of Green, K.J., Croaker, S.J. & Rowbottom, supplements affect the anabolic Clinical Endocrinology and Metabolism 71, D.G. (2003) Carbohydrate hormones after weight-training exercise. 1653–1657. supplementation and exercise-induced Journal of Applied Physiology 76, 839–845. Fehm, H.L., Holl, R., Klein, E. & Voigt, changes in T-lymphocyte function. impact of diet on hormones 441

Journal of Applied Physiology 95, response to food ingestion: evidence for International Journal of Sport Nutrition and 1216–1223. neuroendocrine signals from gut to Exercise Metabolism 11, 149–161. Habito, R.C. & Ball, M.J. (2001) brain. Journal of Clinical Endocrinology Koistinen, H., Koistinen, R., Selenius, L., Postprandial changes in sex hormones and Metabolism 57, 1111–1116. Ylikorkala, Q. & Seppala, M. (1996) after meals of different composition. Ivandic, A., Prpic-Krizevac, I., Jakic, M. Effect of marathon run on serum IGF-I Metabolism 50, 505–511. & Bacun, T. (1999) Changes in sex and IGF-binding protein 1 and 3 levels. Haffner, S.M., Valdez, R.A., Mykkanen, L., hormones during an oral glucose Journal of Applied Physiology 80, 760–764. Stern, M.P. & Katz, M.S. (1994) tolerance test in healthy premenopausal Kraemer, W.J., Volek, J.S., Bush, J.A., Decreased testosterone and women. Fertility and Sterility 71, 268–273. Putukian, M. & Sebastianelli, W.J. (1998) dehydroepiandrosterone sulfate Jenkins, D.J., Wolever, T.M., Taylor, R.H. Hormonal responses to consecutive concentrations are associated with et al. (1981) Glycemic index of foods: a days of heavy-resistance exercise with increased insulin and glucose physiological basis for carbohydrate or without nutritional supplementation. concentrations in nondiabetic men. exchange. American Journal of Clinical Journal of Applied Physiology 85, Metabolism 43, 599–603. Nutrition 34, 362–366. 1544–1555. Heffernan, M.A., Jiang, W.J., Thorburn, Jensen, M.D., Caruso, M., Heiling, V.J. & Lavoie, J.M., Fillion, Y., Couturier, K. & A.W. & Ng, F.M. (2000) Effects of oral Miles, J.M. (1989) Insulin regulation of Corriveau, P. (2002) Evidence that the administration of a synthetic fragment lipolysis in nondiabetic and IDDM decrease in liver glycogen is associated of human growth hormone on lipid subjects. Diabetes 38, 1595–1601. with the exercise-induced increase in metabolism. American Journal of Jentjens, R.L., van Loon, L.J., Mann, C.H., IGFBP-1. Journal of Applied Physiology 93, Physiology. Endocrinology and Metabolism Wagenmakers, A.J. & Jeukendrup, A.E. 798–804. 279, E501–E507. (2001) Addition of protein and amino Leal, A.M. & Moreira, A.C. (1997) Food Henson, D.A., Nieman, D.C., Parker, acids to carbohydrates does not enhance and the circadian activity of the J.C. et al. (1998) Carbohydrate postexercise muscle glycogen synthesis. hypothalamic–pituitary–adrenal axis. supplementation and the lymphocyte Journal of Applied Physiology 91, 839–846. Brazilian Journal of Medical and Biological proliferative response to long endurance Jezova-Repcekova, D., Vigas, M. & Klimes, Research 30, 1391–1405. running. International Journal of Sports I. (1980) Decreased plasma cortisol Lee, P.D., Conover, C.A. & Powell, D.R. Medicine 19, 574–580. response to pharmacological stimuli (1993) Regulation and function of Henson, D.A., Nieman, D.C., Nehlsen- after glucose load in man. Endocrinologia insulin-like growth factor-binding Cannarella, S.L. et al. (2000) Influence Experimentalis 14, 113–120. protein-1. Proceedings of Society for of carbohydrate on cytokine and Johannessen, A., Hagen, C. & Galbo, H. Experimental Biology and Medicine 204, phagocytic responses to 2 h of rowing. (1981) Prolactin, growth hormone, 4–29. Medicine and Science in Sports and Exercise thyrotropin, 3,5,3’-triiodothyronine, and Lee, P.D., Giudice, L.C., Conover, C.A. & 32, 1384–1389. thyroxine responses to exercise after fat- Powell, D.R. (1997) Insulin-like growth Hjalmarsen, A., Aasebo, U., Aakvaag, A. & and carbohydrate-enriched diet. Journal factor binding protein-1: recent findings Jorde, R. (1996) Sex hormone responses of Clinical Endocrinology and Metabolism and new directions. Proceedings of Society in healthy men and male patients with 52, 56–61. for Experimental Biology and Medicine 216, chronic obstructive pulmonary disease Kersten, S. (2001) Mechanisms of 319–357. during an oral glucose load. nutritional and hormonal regulation of van Loon, L.J., Saris, W.H., Kruijshoop, M. Scandinavian Journal of Clinical and lipogenesis. EMBO Reports 2, 282–286. & Wagenmakers, A.J. (2000a) Laboratory Investigation 56, 635–640. Kimball, S.R. & Jefferson, L.S. (2002) Maximizing postexercise muscle Holden, J.P., Butzow, T.L., Laughlin, G.A. Control of protein synthesis by amino glycogen synthesis: carbohydrate et al. (1995) Regulation of insulin-like acid availability. Current Opinion in supplementation and the application growth factor binding protein-1 during Clinical Nutrition and Metabolic Care 5, of amino acid or protein hydrolysate the 24-hour metabolic clock and in 63–67. mixtures. American Journal of Clinical response to hypoinsulinemia induced Kimball, S.R., Farrell, P.A. & Jefferson, L.S. Nutrition 72, 106–111. by fasting and Sandostatin in normal (2002) Invited review: role of insulin in van Loon, L.J., Saris, W.H., Verhagen, H. women. Journal of the Society for translational control of protein synthesis & Wagenmakers, A.J. (2000b) Plasma Gynecologic Investigation 2, 38–44. in skeletal muscle by amino acids or insulin responses after ingestion of Holt, R.I., Simpson, H.L. & Sonksen, P.H. exercise. Journal of Applied Physiology 93, different amino acid or protein mixtures (2003) The role of the growth hormone- 1168–1180. with carbohydrate. American Journal of insulin-like growth factor axis in glucose Knoll, E., Muller, F.W., Ratge, D., Clinical Nutrition 72, 96–105. homeostasis. Diabetic Medicine 20, 3–15. Bauersfeld, W. & Wisser, H. (1984) van Loon, L.J., Kruijshoop, M., Menheere, Hopkins, N.J., Jakeman, P.M., Hughes, Influence of food intake on P.P. et al. (2003) Amino acid ingestion S.C. & Holly, J.M. (1994) Changes in concentrations of plasma strongly enhances insulin secretion in circulating insulin-like growth factor- catecholamines and cortisol. Journal patients with long-term type 2 diabetes. binding protein-1 (IGFBP-1) during of Clinical Chemistry and Clinical Diabetes Care 26, 625–630. prolonged exercise: effect of Biochemistry 22, 597–602. Meckling, K.A., Gauthier, M., Grubb, R. carbohydrate feeding. Journal of Clinical Koch, A.J., Potteiger, J.A., Chan, M.A., & Sanford, J. (2002) Effects of a Endocrinology and Metabolism 79, Benedict, S.H. & Frey, B.B. (2001) hypocaloric, low-carbohydrate diet 1887–1890. Minimal influence of carbohydrate on weight loss, blood lipids, blood Ishizuka, B., Quigley, M.E. & Yen, S.S. ingestion on the immune response pressure, glucose tolerance, and body (1983) Pituitary hormone release in following acute resistance exercise. composition in free-living overweight 442 chapter 29

women. Canadian Journal of Physiology et al. (2003) Carbohydrate ingestion Ricart, W. & Fernandez-Real, J.M. (2001) and Pharmacology 80, 1095–1105. influences skeletal muscle cytokine No decrease in free IGF-I with increasing Meikle, A.W., Benson, S.J., Liu, X.H., mRNA and plasma cytokine levels after insulin in obesity-related insulin Boam, W.D. & Stringham, J.D. (1989) a 3-h run. Journal of Applied Physiology 94, resistance. Obesity Research 9, Nonesterified fatty acids modulate 1917–1925. 631–636. steroidogenesis in mouse Leydig cells. Nuttall, F.Q., Mooradian, A.D., Gannon, Rojdmark, S., Rydvald, Y., Aquilonius, A. American Journal of Physiology 257, M.C., Billington, C. & Krezowski, P. & Brismar, K. (2000) Insulin-like growth E937–E942. (1984) Effect of protein ingestion on factor (IGF)-1 and IGF-binding protein-1 Meikle, A.W., Stringham, J.D., Woodward, the glucose and insulin response to a concentrations in serum of normal M.G. & McMurry, M.P. (1990) Effects of standardized oral glucose load. Diabetes subjects after alcohol ingestion: evidence a fat-containing meal on sex hormones Care 7, 465–470. for decreased IGF-1 bioavailability. in men. Metabolism 39, 943–946. Nuttall, F.Q., Gannon, M.C., Wald, J.L. Clinical Endocrinology 52, 313–318. Miller, S.L., Maresh, C.M., Armstrong, & Ahmed, M. (1985) Plasma glucose Rooyackers, O.E. & Nair, K.S. (1997) L.E. et al. (2002) Metabolic response and insulin profiles in normal subjects Hormonal regulation of human muscle to provision of mixed protein– ingesting diets of varying carbohydrate, protein metabolism. Annual Review of carbohydrate supplementation during fat, and protein content. Journal of the Nutrition 17, 457–485. endurance exercise. International Journal American College of Nutrition 4, 437–450. Roth, J., Glick, S.M., Yalow, R.S. & Berson, of Sport Nutrition and Exercise Metabolism Nyomba, B.L., Berard, L. & Murphy, L.J. S.A. (1963) Hypoglycemia: a potent 12, 384–397. (1997) Free insulin-like growth factor I stimulus to secretion of growth Mitchell, J.B., Costill, D.L., Houmard, J.A. (IGF-I) in healthy subjects: relationship hormone. Science 140, 987–988. et al. (1990) Influence of carbohydrate with IGF-binding proteins and Roy, B.D. & Tarnopolsky, M.A. (1998) ingestion on counterregulatory insulin sensitivity. Journal of Clinical Influence of different macronutrient hormones during prolonged exercise. Endocrinology and Metabolism 82, intakes on muscle glycogen resynthesis International Journal of Sports Medicine 11, 2177–2181. after resistance exercise. Journal of 33–36. Pascoe, D.D., Costill, D.L., Fink, W.J., Applied Physiology 72, 1854–1859. Mitchell, J.B., Pizza, F.X., Paquet, A. et al. Robergs, R.A. & Zachwieja, J.J. (1993) Roy, B.D., Tarnopolsky, M.A., (1998) Influence of carbohydrate status Glycogen resynthesis in skeletal muscle MacDougall, J.D., Fowles, J. & on immune responses before and after following resistive exercise. Medicine Yarasheski, K.E. (1997) Effect of endurance exercise. Journal of Applied and Science in Sports and Exercise 25, glucose supplement timing on protein Physiology 84, 1917–1925. 349–354. metabolism after resistance training. Murray, R., Paul, G.L., Seifert, J.G. & Eddy, Pasquali, R., Macor, C., Vicennati, V. et al. Journal of Applied Physiology 82, D.E. (1991) Responses to varying rates of (1997) Effects of acute hyperinsulinemia 1882–1888. carbohydrate ingestion during exercise. on testosterone serum concentrations in Salmeron, J., Manson, J.E., Stampfer, M.J. Medicine and Science in Sports and Exercise adult obese and normal-weight men. et al. (1997) Dietary fiber, glycemic load, 23, 713–718. Metabolism 46, 526–529. and risk of non-insulin-dependent Murray, R., Bartoli, W.P., Eddy, D.E. & Patsch, J.R., Miesenbock, G., Hopferwieser, diabetes mellitus in women. JAMA 277, Horn, M.K. (1995) Physiological and T. et al. (1992) Relation of triglyceride 472–477. performance responses to nicotinic-acid metabolism and coronary artery disease: Samaha, F.F., Iqbal, N., Seshadri, P. et al. ingestion during exercise. Medicine studies in the postprandial state. (2003) A low-carbohydrate as compared and Science in Sports and Exercise 27, Thrombosis 12, 1336–1345. with a low-fat diet in severe obesity. 1057–1062. Rabast, U., Schonborn, J. & Kasper, H. New England Journal of Medicine 348, Nakagawa, E., Nagaya, N., Okumura, H. (1979) Dietetic treatment of obesity with 2074–2081. et al. (2002) Hyperglycaemia suppresses a low and high-carbohydrate diets: Schaller, G., Schmidt, A., Pleiner, J. et al. the secretion of ghrelin, a novel growth- comparative studies and clinical results. (2003) Plasma ghrelin concentrations hormone-releasing peptide: responses to International Journal of Obesity 3, 201–211. are not regulated by glucose or insulin: the intravenous and oral administration Rasmussen, B.B., Tipton, K.D., Miller, S.L., a double-blind, placebo-controlled of glucose. Clinical Science (London, Wolf, S.E. & Wolfe, R.R. (2000) An oral crossover clamp study. Diabetes 52, England) 103, 325–328. amino acid-carbohydrate supplement 16–20. Nehlsen-Cannarella, S.L., Fagoaga, O.R., enhances muscle protein anabolism after Sebokova, E., Garg, M.L. & Clandinin, Nieman, D.C. et al. (1997) Carbohydrate resistance exercise. Journal of Applied M.T. (1988) Modulation of receptor- and the cytokine response to 2.5 h of Physiology 88, 386–392. mediated gonadotropin action in rat running. Journal of Applied Physiology 82, Reynolds, R.M., Walker, B.R., Syddall, H.E. testes by dietary fat. American Journal of 1662–1667. et al. (2001) Elevated plasma cortisol in Physiology 254, E708–E712. Nieman, D.C. (1998) Influence of glucose-intolerant men: differences in Sebokova, E., Garg, M.L., Wierzbicki, A., carbohydrate on the immune response responses to glucose and habituation Thompson, A.B.R. & Clandinin, M.T. to intensive, prolonged exercise. Exercise to venepuncture. Journal of Clinical (1990) Alteration of the lipid Immunology Review 4, 64–76. Endocrinology and Metabolism 86, composition of rat testicular plasma Nieman, D.C., Nehlsen-Cannarella, S.L., 1149–1153. membranes by dietary (n-3) fatty acids Fagoaga, O.R. et al. (1998) Influence of Reynolds, R.M., Syddall, H.E., Walker, changes the responsiveness of Leydig mode and carbohydrate on the cytokine B.R., Wood, P.J. & Phillips, D.I. (2003) cells and testosterone synthesis. Journal response to heavy exertion. Medicine Predicting cardiovascular risk factors of Nutrition 120, 610–618. and Science in Sports and Exercise 30, from plasma cortisol measured during Sharman, M.J., Kraemer, W.J., Love, D.M. 671–678. oral glucose tolerance tests. Metabolism et al. (2002) A ketogenic diet favorably Nieman, D.C., Davis, J.M., Henson, D.A. 52, 524–527. affects serum biomarkers for impact of diet on hormones 443

cardiovascular disease in normal-weight volunteers. Journal of Nutritional et al. (2001) Effects of a high-fat diet men. Journal of Nutrition 132, 1879–1885. Biochemistry 10, 89–95. on postabsorptive and postprandial Siddals, K.W., Westwood, M., Gibson, J.M. Tipton, K.D., Rasmussen, B.B., Miller, S.L. testosterone responses to a fat-rich & White, A. (2002) IGF-binding protein- et al. (2001) Timing of amino acid– meal. Metabolism 50, 1351–1355. 1 inhibits IGF effects on adipocyte carbohydrate ingestion alters anabolic Volek, J.S., Sharman, M.J., Love, D.M. et al. function: implications for insulin-like response of muscle to resistance (2002) Body composition and hormonal actions at the adipocyte. Journal of exercise. American Journal of Physiology responses to a carbohydrate-restricted Endocrinology 174, 289–297. 281, E197–E206. diet. Metabolism 51, 864–870. Slag, M.F., Ahmad, M., Gannon, M.C. & Tropeano, G., Lucisano, A., Liberale, I. et al. Volek, J.S., Sharman, M.J., Gómez, A.L., Nuttall, F.Q. (1981) Meal stimulation (1994) Insulin, C-peptide, androgens, Scheett, T.P. & Kraemer, W.J. (2003) of cortisol secretion: a protein induced and β-endorphin response to oral An isoenergetic very low-carbohydrate effect. Metabolism 30, 1104–1108. glucose in patients with polycystic diet is associated with improved serum Smith, K., Reynolds, N., Downie, S., ovary syndrome. Journal of Clinical high-density lipoprotein cholesterol Patel, A. & Rennie, M.J. (1998) Effects of Endocrinology and Metabolism 78, (HDL-C), total cholesterol to HDL-C flooding amino acids on incorporation 305–309. ratio, triacylglycerols, and postprandial of labeled amino acids into human Tsintzas, O.K., Williams, C., Wilson, lipemic responses compared to a low-fat muscle protein. American Journal of W. & Burrin, J. (1996) Influence of diet in normal weight, normolipidemic Physiology. Endocrinology and Metabolism carbohydrate supplementation early in women. Journal of Nutrition 133, 275, E73–E78. exercise on endurance running capacity. 2756–2761. Smith, S., Ravnikar, V.A. & Barbieri, R.L. Medicine and Science in Sports and Exercise Volek, J.S., Sharman, M.J., Gomez, A.L. (1987) Androgen and insulin response 28, 1373–1379. et al. (2004) Comparison of very low- to an oral glucose challenge in Turner, J.D., Rotwein, P., Novakofski, J. carbohydrate and low-fat diet on fasting hyperandrogenic women. Fertility and & Bechtel, P.J. (1988) Induction of lipids, LDL subclasses, insulin Sterility 48, 72–77. messenger RNA for IGF-I and -II during resistance, and postprandial lipemic Sondike, S.B., Copperman, N. & Jacobson, growth hormone-stimulated muscle responses in overweight women. Journal M.S. (2003) Effects of a low-carbohydrate hypertrophy. American Journal of of the American College of Nutrition 23(2), diet on weight loss and cardiovascular Physiology 255, E513–E517. 177–184. risk factors in overweight adolescents. Ullrich, I.H., Peters, P.J. & Albrink, M.J. Walker, B.R., Soderberg, S., Lindahl, B. & Journal of Pediatrics 142, 253–258. (1985) Effect of low-carbohydrate diets Olsson, T. (2000) Independent effects Starkie, R.L., Rolland, J., Angus, D.J., high in either fat or protein on thyroid of obesity and cortisol in predicting Anderson, M.J. & Febbraio, M.A. (2001) function, plasma insulin, glucose, and cardiovascular risk factors in men and Circulating monocytes are not the triglycerides in healthy young adults. women. Journal of Internal Medicine 247, source of elevations in plasma IL-6 and Journal of American College of Nutrition 4, 198–204. TNF-α levels after prolonged running. 451–459. Whitley, H.A., Humphreys, S.M., American Journal of Physiology. Cell Utter, A.C., Kang, J., Nieman, D.C. et al. Campbell, I.T. et al. (1998) Metabolic Physiology 280, C769–C774. (1999) Effect of carbohydrate ingestion and performance responses during Svanberg, E., Ennion, S., Isgaard, J. & and hormonal responses on ratings of endurance exercise after high-fat Goldspink, G. (2000) Postprandial perceived exertion during prolonged and high-carbohydrate meals. resynthesis of myofibrillar proteins cycling and running. European Journal of Journal of Applied Physiology 85, is translationally rather than Applied Physiology and Occupational 418–424. transcriptionally regulated in human Physiology 80, 92–99. Willi, S.M., Oexmann, M.J., Wright, N.M., skeletal muscle. Nutrition 16, 42–46. Vicennati, V., Ceroni, L., Gagliardi, L., Collop, N.A. & Lyndon, L. (1998) The Tarpenning, K.M., Wiswell, R.A., Gambineri, A. & Pasquali, R. (2002) effects of a high-protein, low-fat, Hawkins, S.A. & Marcell, T.J. (2001) Comment: response of the ketogenic diet on adolescents with Influence of weight training exercise hypothalamic–pituitary–adrenocortical morbid obesity: body composition, and modification of hormonal response axis to high-protein/fat and high- blood chemistries, and sleep on skeletal muscle growth. Journal of carbohydrate meals in women with abnormalities. Pediatrics 101, 61–67. Science and Medicine in Sport 4, 431–446. different obesity phenotypes. Journal of Williams, A.G., Ismail, A.N., Sharma, A. Tiitinen, A., Pekonen, F., Stenman, U.H. & Clinical Endocrinology and Metabolism 87, & Jones, D.A. (2002) Effects of resistance Laatikainen, T. (1990) Plasma androgens 3984–3988. exercise volume and nutritional and oestradiol during oral glucose Volek, J.S. (2004) Nutritional practices for supplementation on anabolic and tolerance test in patients with polycystic resistance training. Medicine and Science catabolic hormones. European Journal of ovaries. Human Reproduction (Oxford, in Sports and Exercise 36, 689–696. Applied Physiology 86, 315–321. England) 5, 242–245. Volek, J.S. & Westman, E.C. (2002) Low Young, C.M., Scanlan, S.S., Im, H.S. & Tipton, K.D., Ferrando, A.A., Phillips, carbohydrate weight-loss diets revisited. Lutwak, L. (1971) Effect on body S.M., Doyle, D., Jr. & Wolfe, R.R. (1999a) Cleveland Clinic Journal of Medicine 69, composition and other parameters in Postexercise net protein synthesis in 849–862. obese young men of carbohydrate level human muscle from orally administered Volek, J.S., Gómez, A.L. & Kraemer, W.J. of reduction diet. American Journal of amino acids. American Journal of (2000) Fasting and postprandial Clinical Nutrition 24, 290–296. Physiology 276, E628–E634. lipoprotein responses to a low- Zawadzki, K.M., Yaspelkis, B.B., 3rd & Ivy, Tipton, K.D., Gurkin, B.E., Martin, S. & carbohydrate diet supplemented with J.L. (1992) Carbohydrate–protein Wolfe, R.R. (1999b) Nonessential amino n-3 fatty acids. Journal of the American complex increases the rate of muscle acids are not necessary to stimulate net College of Nutrition 19, 383–391. glycogen storage after exercise. Journal muscle protein synthesis in healthy Volek, J.S., Gómez, A.L., Love, D.M. of Applied Physiology 72, 1854–1859. Chapter 30

Neurohumoral Responses and Adaptations During Rest and Exercise at Altitude

BETH A. BEIDLEMAN, JANET E. STAAB AND ELLEN L. GLICKMAN

Introduction status (sick versus non-sick). This review therefore will focus on summarizing the results of numerous Upon arrival at high altitude, rapid physiologic studies that were conducted under a defined set of adjustments occur to compensate for the reduction conditions in order to develop a clearer understand- in ambient oxygen (O2). Immediate physiologic ing of the neurohumoral responses and adaptations adjustments, which include an increase in ventila- during rest and exercise at altitude. tion and heart rate, provide a rapid first line of The studies included in this review were as fol- defense against the reduced ambient O2 (Stenberg lows: (a) both chamber and field studies; (b) studies et al. 1966; Easton et al. 1986). Longer term physio- employing hypobaric hypoxia; (c) altitudes ranging logic adaptations, which include continued hyper- from 3500 to 5500 m; (d) both men and women; (e) ventilation, increased production of RBCs, changes both physically trained and untrained; (f) both in circulation, increased water and sodium excre- rapid and slow ascent; and (g) if the results of sick tion, and shifts in substrate utilization, provide a and non-sick subjects were separated, only results second line of defense against the sustained hypoxic from non-sick subjects. Acute altitude exposure was environment (Young, A.J. & Reeves 2002). The defined as <=3 days and chronic altitude exposure autonomic nervous system plays a critical role in was defined as 4–21 days of altitude exposure. This mediating many of the immediate physiologic review will also focus on human studies unless adjustments to altitude but hormonal alterations critical points can only be explained by including play a role in fine-tuning many of the body’s longer- animal work. term adaptations to altitude. Although several reviews have examined the Basic concepts neurohumoral response to altitude (Hoyt & Honig 1996; Ward et al. 2000; Richalet 2001; Swenson 2001; In order to understand the neurohumoral mech- Mazzeo & Reeves 2003), a clear consensus has not anisms responsible for adaptation to altitude, it is been reached regarding the role that hormones play imperative to first understand the basic concepts in successfully adapting to altitude. Conflicting associated with exposure to altitude. First, baromet- results in the literature are primarily due to differ- ric pressure (PB) decreases with increasing altitude ences in study design (field versus chamber studies), (Fig. 30.1). Although the percentage of O2 in a given time point of measurements during exposure (acute volume of air at altitude is the same as at sea level o versus chronic), altitude (low versus high elevation), (i.e. 20.93%), the ambient partial pressure of O2 (P 2) o = × environmental conditions (normobaric hypoxia ver- decreases (P 2 PB % O2). Second, the inspired o sus hypobaric hypoxia), gender, (male versus female), oxygen pressure (PI 2) is diminished in direct pro- physical fitness levels (trained versus untrained), portion to the reduction in barometric pressure o = − × o duration of ascent (rapid versus slow) and illness (PI 2 (PB 47) %O2), where the PH 2 is taken to 444 hormones and altitude 445

800 Physiologic adaptations to altitude 700 Ventilatory 600 Upon acute exposure to altitude, there is an imme- 500 diate increase in ventilation that is dependent on the 400 severity of hypoxia (Rahn & Otis 1949; Easton et al. 1986; Bisgard & Forster 1996). This increase in venti- 300 lation is primarily the result of hypoxic stimulation

Barometric pressure (mmHg) 200 of the peripheral chemoreceptors (Weil et al. 1970; 0 2000 4000 6000 8000 10 000 Lahiri & Delaney 1975; Fitzgerald & Dehghani Altitude (m) 1982). Immediately following this increase, ventila- tion declines over the next 10–30 min of altitude Fig. 30.1 The relationship between barometric pressure exposure. This response has been termed ‘hypoxic (PB) and altitude (m). ventilatory depression’ and is likely due to build up of inhibitory transmitters such as γ-aminobutyric acid or adenosine in the central nervous system (Smith et al. 2001). Within the next few hours to days 160 Sea level 4300 m of altitude exposure, there is a progressive increase 140 in ventilation that tends to level off by 7–10 days of altitude exposure (Forster et al. 1975). The mech- 120 anisms involved in ventilatory acclimatization to 100 altitude are highly debated, but thought to be due to a progressive increase in carotid body sensitivity

(mmHg) 80 2

o and central nervous system acid-base changes 60 (Severinghaus et al. 1963; Bisgard & Forster 1996; 40 Smith et al. 2001). The increase in ventilation with initial and continued altitude exposure mitigates 20 Inspired Alveolar Arterial Mixed the obligatory fall in partial pressure of alveolar O2 o o air gas blood venous (PA 2) that follows a reduction in PI 2. In fact, if blood compensatory hyperventilation did not occur, the o Fig. 30.2 The oxygen transport cascade at sea level and alveolar gas equation predicts that PA 2 would be o 4300 m. (Modified from Fulco & Cymerman 1988.) 10 Torr lower at 4000 m. This increase in PA 2 with hyperventilation facilitates O2 loading in the lungs and provides a rapid and continued first line of o defense against the reduced ambient P 2. be 47 Torr at a body temperature of 37°C. Thus, the O pressure gradient at each step in the O trans- 2 2 Hematologic port cascade is reduced upon exposure to altitude

(Fig. 30.2). As a result, less O2 is transferred from the Erythropoietin (described below) increases upon environment to the cells as the diffusion of O2 is acute exposure to altitude, which stimulates the o directly dependent on the P 2 gradient, according hematological cascade (Jelkmann 1992). Reticulo- to Fick’s law of diffusion. The result is an impaired cytes typically appear within 3–5 days (Grover et al. ability to deliver O2 to the tissues of the body upon 1998; Reeves et al. 2001) but red blood cell (RBC) exposure to altitude. Numerous short- and long- volume remains unchanged following 10 days of term physiologic adaptations occur to compensate altitude exposure (Sawka et al. 1996). Although for this hypoxemia. hemoglobin concentration [Hb] clearly increases in 446 chapter 30

lowlanders sojourning at altitudes for 5–12 days adaptations (Wolfel et al. 1991, 1998). Stroke volume (Maher et al. 1974; Horstman et al. 1980; Wolfel et al. decreases due to a decrease in plasma volume and/ 1991, 1998), this increase in [Hb] is most likely due to or increase in total peripheral resistance while heart hemoconcentration (i.e. decrease in plasma volume) rate remains elevated (Grover et al. 1986; Wolfel & rather than erythrocyte volume expansion (Sawka Levine 2001; Young, A.J. & Reeves 2002). Cardiac et al. 2000). If the altitude sojourn exceeds 1 month, output declines following chronic altitude exposure erythrocyte volume expansion occurs (Reynafarje because of the decrease in stroke volume (Grover 1957; Pugh 1964). The higher RBC volumes observed et al. 1986; Wolfel & Levine 2001; Young, A.J. & in high altitude natives compared to lowlanders Reeves 2002). The decline in cardiac output offsets o also supports increased RBC production follow- the increase in Ca 2 such that O2 delivery remains ing an extended period of time at altitude (Weil unchanged following chronic altitude exposure et al. 1968; Sanchez et al. 1970). The increase in RBC (Grover et al. 1976; Bender et al. 1988). Although a volume and thus Hb, which is the compound declining blood flow and cardiac output would not responsible for binding O2, ultimately improves O2 appear beneficial, the prolonged capillary transit transport capacity at altitude. time may minimize the alveolar–arterial diffusion o limitation at altitude such that Sa 2 is not comprom- ised at the expense of increased blood flow (Boushel Cardiovascular et al. 2001). Heart rate and cardiac output are increased imme- diately upon ascent to altitude (Stenberg et al. 1966; Fluid regulation Vogel & Harris, 1967; Wagner et al. 1986). The abrupt increase in cardiac output is due to an increase in Water and sodium excretion are increased in norm- heart rate, since stroke volume remains unchanged ally acclimatizing lowlanders with both acute and (Grover et al. 1986; Wolfel & Levine 2001). Sym- chronic exposure to altitude (Hoyt & Honig 1996). pathetic stimulation and parasympathetic with- This natriureis and diuresis elicits a depletion of drawal (described below) both contribute to this plasma volume and increase in [Hb] due to hemo- increase in heart rate. Normally, sympathetic stimu- concentration (Sawka et al. 2000), which counterbal- lation would also elicit peripheral vasoconstriction. ances the reduced supply of O2 to the tissues. The However, acute hypoxia overrides sympathetically- diuresis may also protect against altitude illness mediated vasoconstriction and induces vasodila- (Singh et al. 1969) due to a decrease in extracellular tion in all vascular beds except the lung such that fluid volume and brain edema (Hackett 1999). The total peripheral resistance is decreased (Wolfel & mechanisms underlying the natriuretic and diuretic Levine 2001). The specific local vasodilator respons- effect of hypoxia are not well understood, but suggest ible for overriding vasoconstriction is not known, that hypoxic stimulation of the peripheral chemore- but nitric oxide and epinephrine have been sug- ceptors results in decreased reabsorption of renal gested as potential candidates (Halliwill 2003). tubular sodium (Honig 1989). Tissue hypoxia may

Mixed-venous O2 content also decreases with acute also dilate renal vessels directly (Halliwill 2003), altitude exposure, which indicates increased peri- which would increase the glomerular filtration pheral O2 extraction (Wolfel et al. 1991, 1998). Thus, rate (Olsen et al. 1993) and thus sodium and water increases in both blood flow and peripheral O2 excretion. Several hormones may also play a key extraction upon acute exposure to altitude maintain role in the fluid regulation at high altitude, since the a tight coupling between metabolic supply and natriuretic and diuretic response is not abolished demand under limited O2 conditions. by renal nerve section (Schmidt et al. 1985; Karim With chronic altitude exposure, arterial O2 content et al. 1987). In fact, after adrenalectomy, the striking o (Ca 2), which is the product of arterial O2 saturation increase in urinary sodium excretion on exposure o –1 (Sa 2) and [Hb] multiplied by 1.34 mL O2·g [Hb] , is to hypobaric hypoxia is not observed (Lewis et al. increased due to ventilatory and hematological 1942). hormones and altitude 447

greater the hypoxemia, the more severe the illness Metabolic (Maggiorini et al. 1990; Honigman et al. 1993). Upon acute exposure to altitude, carbohydrates Individual AMS susceptibility can also be affected may be the preferred fuel since they have a high by the rate of ascent (Hansen et al. 1967; Honigman yield of adenosine triphosphate (ATP) per liter of O2 et al. 1993), prior acclimatization (Hansen et al. 1967; consumed (Brooks et al. 1991; Roberts et al. 1996b). Lyons et al. 1995), obesity (Honigman et al. 1993) and In support of this hypothesis, hypoxia inducible level of exertion to reach a given altitude (Roach factor-1α (HIF-1α), which is known to increase the et al. 2000). transcription of several genes involved in glucose The pathophysiology of AMS is highly debated uptake and glycolysis, is up-regulated upon acute but in general is thought to be a vasogenic-induced exposure to hypoxia (Semenza et al. 1994; Firth et al. cerebral edema, triggered by the initial hypoxic 1995). In addition, lactate accumulation is increased stimulus (Hackett et al. 1998). Vasogenic edema during both rest and exercise upon acute exposure could be the result of increased hydrostatic pressure to altitude, which has been attributed to accelerated across blood vessels due to an increase in cerebral glycolytic metabolism (Knuttgen & Saltin 1973). blood flow or an altered permeability of the blood– However, no differences in muscle glycogen utiliza- brain barrier. Jensen et al. (1990) reported no increase tion have been observed during exercise upon acute in cerebral blood flow in subjects suffering from exposure to exercise compared to sea level (Young, AMS. However, increased expression of vascular A.J. et al. 1982; Green et al. 1992). With chronic alti- endothelial growth factor has been reported in tude exposure, both increased (Brooks et al. 1991; humans (Walter et al. 2001) and animals (Xu & Roberts et al. 1996b) and decreased (Young, A.J. et al. Severinghaus 1998) exposed to hypoxia or hypobaric 1982, 1991; Beidleman et al. 1997; Braun et al. 2000) hypoxia, which would cause a generalized increased utilization of carbohydrates have been reported. capillary leakage of large molecules due to endo- However, lactate accumulation is consistently de- thelial activation in forming new blood vessels. pressed despite a similar O2 delivery following Resolution of AMS usually occurs following 1–2 chronic altitude exposure (Young, A.J. et al. 1982; days of high altitude residence if no further ascent is Brooks et al. 1991; Beidleman et al. 1997). This has attempted (Singh et al. 1969; Bärtsch et al. 1988). been termed the ‘lactate paradox’ and has been Physiological changes that may contribute to this attributed to decreased β-adrenergic stimulation of resolution of AMS include an increase in ventilation muscle glycogenolysis (Mazzeo & Reeves 2003). and diuresis, which can occur over the same few days of altitude residence (Singh et al. 1969; Hackett et al. 1982; Moore et al. 1986; Bärtsch et al. 1988). The Acute mountain sickness o increase in ventilation causes an increase in Pa 2, Acute mountain sickness (AMS) is a syndrome which reduces the initial hypoxic stimulus. An induced by hypoxia in unacclimatized individuals increased diuresis also helps to reduce symptoms who ascend rapidly to altitudes exceeding 2500 m of AMS (Singh et al. 1969) via a reduction in extracel- and remain there for more than several hours (Singh lular fluid volume, which lessens the magnitude et al. 1969; Hackett & Rennie 1976; Honigman et al. of cerebral edema, the causative factor for AMS 1993). Characteristic symptoms include headache, (Hackett 1999). nausea, vomiting, loss of appetite, fatigue, dizziness and sleep disturbances (Hackett & Rennie 1976; Maximal exercise performance Powles et al. 1978; Honigman et al. 1993; Sanchez del Rio & Moskowitz 1999; Bailey et al. 2000). The According to the Fick equation, O2 uptake is the onset of AMS usually occurs 4–12 h after ascent, and product of cardiac output (heart rate × stroke vol- o symptoms become most prominent after the 1st ume) and the difference in Ca 2 and venous O2 con- o o night spent at high altitude (Bärtsch et al. 1988). The tent (Cv 2). Maximal Ca 2 is decreased upon acute incidence of AMS increases with altitude; the exposure to altitude due to the reduction in ambient 448 chapter 30

Po and corresponding reduction in S o (Wolfel & 2 a 2 5 Steady-state O2 for cycle exercise Levine 2001). During maximal exercise at altitude, at power output of 100 W ) the body is unable to increase maximal ventilation, –1 4 heart rate and tissue O extraction beyond maximal 2 O2max values obtained at sea level to compensate for the 3 o reduction in Sa 2 (Stenberg et al. 1966; Grover et al. o 1986). Thus, maximal O2 uptake (V 2max) is reduced 2 o in direct proportion to the decrease in maximal Ca 2 upon acute exposure to altitude (Stenberg et al. 1966; 1 50% 70% Vogel et al. 1967). Oxygen uptake (L·min O2max O2max With chronic altitude exposure, maximal C O is a 2 0 increased compared to initial ascent due to increases Sea level 4300 m o in both Sa 2 and [Hb] (Wolfel & Levine 2001). Therefore, one would expect an increase in Vo Fig. 30.3 Effect of high altitude (4300 m) on the relationship 2max between absolute work load, oxygen uptake (Vo ), and However, research has repeatedly shown that 2 relative exercise intensity (% maximal oxygen uptake, Vo does not change following chronic altitude o 2max V 2max). (Modified from Young, A.J. & Young 1988.) exposure (Saltin et al. 1968; Young, A.J. et al. 1982; Bender et al. 1988; Wolfel et al. 1991, 1998; Beidleman o et al. 1997). The reason for an unchanged V 2max is extraction and submaximal exercise performance that maximal cardiac output decreases following will be similar at both sea level and during acute chronic altitude exposure due to a decrease in both exposure to altitude (Maher et al. 1974; Horstman maximal heart rate and stroke volume, which off- et al. 1980). o sets the increase in maximal Ca 2 (Grover et al. 1986; With chronic altitude exposure, both small and Wolfel & Levine 2001). Thus, maximal O2 delivery large muscular endurance performance is improved, remains the same upon acute and chronic altitude even when conducted at the same relative percent- o exposure (Grover et al. 1986). age of altitude-specific V 2max (Maher et al. 1974; Horstman et al. 1980; Fulco et al. 1994). Although not entirely understood, likely explanations for this Submaximal exercise performance improvement include a greater pulmonary diffu- Oxygen uptake at rest or during submaximal fixed sion time due to a decrease in blood flow (Boushel workloads at altitude does not change from sea et al. 2001), a greater O2 diffusion gradient from the level values (Reeves et al. 1967; Saltin et al. 1968; muscle capillary to the mitochondria (Fulco et al. o Raynaud et al. 1981). However, since V 2max is 1994; Beidleman et al. 2003) and sparing of muscle decreased at altitude, the relative exercise intensity glycogen due to an increase in blood glucose util- during a fixed submaximal workload can be greatly ization (Brooks et al. 1991; Roberts et al. 1996a). increased compared to that at sea level (Fig. 30.3). Furthermore, AMS symptomatology is decreased

In this example, the same submaximal O2 uptake following chronic altitude exposure which in and −1 o (2 L·min ) represents 50% of a subject’s V 2max at of itself may contribute to improvements in sub- o sea level and 70% of a subject’s V 2max at altitude. If maximal exercise performance. a fixed submaximal workload is maintained upon acute exposure to altitude, then ventilation, heart Neurohumoral responses to altitude rate and tissue O2 extraction will be increased and submaximal exercise performance will be decreased Sympathoadrenal response (Ekblom et al. 1975; Fulco et al. 2003). If the submaxi- mal workload is adjusted such that an individual is Plasma levels of norepinephrine (NE) are a function working at the same relative percentage of altitude- of the rate of spillover into the circulation from sym- o specific V 2max, then ventilation, heart rate, tissue O2 pathetic neuronal release (~ 70%), a small amount hormones and altitude 449

Rest 20 Exercise % % 2000 % % SL AA CA SL-AA AA-CA SL AA CA SL-AA AA-CA ) Rest 2.4 3.0 10.1 24 239 –1 Rest 470 647 487 38 –25 )

Exercise 8.8 10.5 18.5 20 34 –1 15 Exercise 1040 1840 1226 77 –33 % (R-Ex) 267 250 83 1500 % (R-Ex) 121 184 152

10 1000

5 500 Epinephrine (pmol·L Norepinephrine (nmol·L

0 0 (a)SL AA CA (b) SL AA CA

Fig. 30.4 Summarized results from 30 studies measuring (a) norepinephrine and (b) epinephrine levels during both rest and exercise during sea level (SL), acute altitude (AA) and chronic altitude (CA) exposure (Kotchen et al. 1973b; Fulco et al. 1985; Bärtsch et al. 1988, 1991a; Bouissou et al. 1988, 1989; Férézou et al. 1988, 1993; Kjær et al. 1988; Richalet et al. 1988; Knudtzon et al. 1989b; Mazzeo et al. 1991, 1995, 2000, 2001, 2003; Young, A.J. et al. 1991, Olsen et al. 1992; Ramirez et al. 1992; Young, P.M. et al. 1992; Loeppky et al. 1993; Antezana et al. 1995; Larsen et al. 1997; Rostrup 1998; Sevre et al. 2001; Basu et al. 2002; Beidleman et al. 2002; Bestle et al. 2002; Hasbak et al. 2002; Lundby & Steensberg 2004).

secreted by the adrenal medulla (~ 30%), and its altitude (Myhre et al. 1970; Jung et al. 1971; Jain et al. rate of removal from the plasma pool. Upon acute 1980). exposure to normobaric hypoxia or hypobaric Similar to exercise at sea level, NE levels increase hypoxia, numerous studies have reported an imme- one to threefold with exercise during acute and diate increase in sympathetic nervous system (SNS) chronic altitude exposure (Mazzeo et al. 1991, 1995, activity to skeletal muscle (Rowell et al. 1989; 2003; Young, A.J. et al. 1991), and the magnitude of Leuenberger et al. 1991; Morgan et al. 1995) that the response is dependent on the intensity of exer- remains elevated following chronic altitude expos- cise (Kjær et al. 1988; Braun et al. 2000; Mazzeo et al. ure (Hansen & Sander 2003). Although plasma NE 2000; Lundby & Steensberg 2004). Summarized levels do not always track SNS activity (Rowell et al. results suggest that the percentage of increase in 1989), summarized results from 30 studies also sug- NE during exercise from resting values is lower gest that there is a 24% increase in resting NE levels following chronic altitude exposure (~ 83%) com- upon acute altitude exposure that increases a fur- pared to acute altitude exposure (~ 250%) and sea ther 240% from acute to chronic altitude exposure level (~ 267%). However, NE levels reach higher (Fig. 30.4a). The mechanism for the initial and con- absolute levels during exercise with acute and tinued elevation in SNS activity in hypoxia is not chronic altitude exposure due to higher initial start- well known, but is unlikely due to the direct effect of ing levels (Fig. 30.4a). hypoxia since SNS activity increases with time at The immediate increase in SNS activity upon altitude, while hypoxemia decreases with time at acute exposure to altitude likely plays a role in altitude (Mazzeo & Reeves 2003). It appears that counterbalancing the effects of hypoxia-induced hypoxia may cause sympathetic activation directly vasodilation on most vascular beds so that blood via stimulation of the peripheral chemoreceptors pressure can be maintained (Wolfel & Levine 2001). (Marshall 1994) and brainstem neuronal pools (Reis The rising NE levels during both rest and exercise et al. 1994). In addition, unloading of cardiopul- following chronic altitude exposure results in monary baroreceptors could be involved due to a increased arterial and venous tone, increased sys- decrease in blood volume occurring over time at temic vascular resistance, decreased plasma volume 450 chapter 30

and decreased cardiac output (Seals et al. 2001; acute altitude exposure (Richardson et al. 1967; Koller Mazzeo & Reeves 2003). These physiologic adapta- et al. 1988). In humans acutely exposed to hypoxia, tions serve to maintain a tight coupling between pretreatment with propranolol (a β-adrenergic metabolic supply and demand by decreasing blood blocker) fails to prevent the acute increase in heart

flow as arterial O2 delivery is improved following rate (Koller et al. 1988). However, when propranolol chronic altitude exposure. and atropine (parasympathetic blocker) are com- Sympathetic stimulation of the adrenal gland bined, the increase in heart rate upon acute expos- upon acute altitude exposure also causes an increase ure to hypoxia is completely eliminated in both in plasma levels of epinephrine (EPI) that abates humans (Koller et al. 1988) and dogs (Hammill et al. following chronic altitude exposure (Mazzeo & 1979). With chronic altitude exposure, most (Hartley Reeves 2003). The summarized results from 30 stud- et al. 1974; Hughson et al. 1994; Boushel et al. 2001) ies suggest that resting EPI levels increase 38% with but not all (Grover et al. 1986) studies suggest a acute altitude exposure and return to sea level val- reversal of the initial decline in PNS activity such ues following chronic altitude exposure (Fig. 30.4b). that cardiac vagal activity is increased and maximal The primary mechanism for EPI release does appear heart rate decreased when compared to sea level to be hypoxia, since both the hypoxic stimulus values. The mechanism for the initial decrease and and EPI levels are increased upon acute altitude subsequent increase in PNS activity is not known, exposure and both lessen with chronic altitude resid- but may be induced via a baroreceptor-induced ence (Mazzeo & Reeves 2003). Wolfel et al. (1994) excitation of the vagal nerve in response to a slight o reported an inverse relationship between Sa 2 and drop and subsequent elevation in mean arterial plasma EPI levels with time at altitude. pressure with time at altitude (Hartley et al. 1974). Similar to exercise at sea level, EPI levels increase The immediate decrease in PNS activity serves to one to threefold with exercise from resting values increase blood flow under limited O2 conditions by during both acute and chronic altitude exposure increasing heart rate. The increase in PNS activity (Mazzeo et al. 1991, 1995, 2003; Young, A.J. et al. following chronic altitude exposure may maintain

1991), and the magnitude of the response is depend- myocardial O2 demand, which is roughly estimated ent on the exercise intensity (Kjær et al. 1988; Braun from the rate-pressure product (heart rate × mean et al. 2000; Mazzeo et al. 2000; Lundby & Steensberg arterial pressure), by decreasing exercise heart rate 2004). Summarized results suggest that the percent- as mean arterial pressure increases. age of increase in EPI during exercise from resting values is the same in all conditions (~ 120–180%), Glucocorticoid response but reaches higher absolute levels with exercise during acute altitude exposure due to higher initial Cortisol (COR) is the major glucocorticoid pro- starting levels (Fig. 30.4b). The increase and sub- duced by the adrenal cortex and is secreted in sequent decrease in EPI levels during both rest and response to increasing levels of adrenocorticotropic exercise likely contributes to the immediate increase hormone (ACTH) released from the anterior pitu- in heart rate, vasodilation, muscle glycogenolysis itary. Although individual study results may differ, and lactate accumulation upon acute altitude expos- the summarized results from 30 studies suggest ure that subsides following chronic exposure to that resting COR levels increase 33% with acute alti- altitude (Mazzeo & Reeves 2003). These physiologic tude exposure and return towards sea level values adaptations serve to increase O2 delivery and optim- following chronic altitude exposure (Fig. 30.5a). ize O2 utilization under limited O2 conditions and The trigger for the initial increase in ACTH and conserve cardiac energy as O2 supply improves. COR with altitude exposure may be partly related to increased peripheral chemoreceptor input as a result of the initial hypoxic stimulus, since deaf- Parasympathetic response ferentation of the arterial chemoreceptors results in Several studies suggest a reduction in resting para- diminished or abolished increases in ACTH and sympathetic nervous system (PNS) activity during COR (Raff et al. 1982; Honig et al. 1996). hormones and altitude 451

% % % % SL AA CA SL AA CA SL-AA AA-CA Rest SL-AA AA-CA Rest 354 470 388 33 –18 Exercise Rest 13 51 25 292 –51 600 Exercise 470 541 484 15 –10 60 % (R-Ex) 33 15 25

500 ) 50 –1 ) –1 400 40

300 30

200 20 Cortisol (nmol·L

100 Erythropoietin (IU·L 10

0 0 (a)SL AA CA (b)SL AA CA

Fig. 30.5 Summarized results from 30 studies measuring (a) cortisol during both rest and exercise (Moncloa et al. 1968; Singh et al. 1974; Frayser et al. 1975; Sutton 1977; Sutton et al. 1977; Claybaugh et al. 1982; Heyes et al. 1982; Mordes et al. 1983; Maresh et al. 1984; Bärtsch et al. 1988, 1991a; Bouissou et al. 1988; Friedl et al. 1988; Kjær et al. 1988; Richalet et al. 1989; Rock et al. 1989; Tunny et al. 1989; Kraemer et al. 1991; Sawhney et al. 1991; Ramirez et al. 1992; Vuolteenaho et al. 1992; Westendorp et al. 1993; Banfi et al. 1994, 1996; Beidleman et al. 1997, 2002; Larsen et al. 1997; Zaccaria et al. 1998; Braun et al. 2000; Basu et al. 2002) and 15 studies measuring (b) erythropoietin (EPO) at rest during sea level (SL), acute altitude (AA) and chronic altitude (CA) exposure (Abbrecht & Littell 1972; Milledge & Cotes 1985; Eckardt et al. 1989; Mairbaurl et al. 1990; Gunga et al. 1994; Klausen et al. 1996; Sawka et al. 1996; Gleiter et al. 1997; Chapman et al. 1998; Grover et al. 1998; Hudson et al. 1999; Bonfichi et al. 2000; Koistinen et al. 2000; Reeves et al. 2001; Ge et al. 2002).

Similar to exercise at sea level, COR levels from 15 studies suggest that initial ascent to altitude increase 30–100% with exercise during both acute induces a two to threefold increase in resting EPO and chronic altitude exposure (Bouissou et al. 1988; levels, with peak levels occurring within ~ 24–48 h, Kjær et al. 1988; Bärtsch et al. 1991a; Braun et al. that declines toward but does not reach sea level 2000), and the magnitude of the response is depend- values following chronic altitude exposure (Fig. ent on the intensity of exercise (Kjær et al. 1988; 30.5b). The EPO response to exercise with acute Braun et al. 2000). Summarized results suggest that and chronic altitude exposure has not been well the percentage of increase in COR during exercise studied. Thus, whether or not the EPO response is from resting values is the same in all conditions potentiated during hypoxic exercise cannot be (~ 15–33%) but reaches higher absolute values determined at this time. The initial stimulus for EPO with exercise during acute altitude exposure due production is related to the reduction in tissue O2 to higher initial starting values (Fig. 30.5a). The tension in the renal cortex upon acute altitude expos- increase in COR upon acute exposure to altitude ure (Jelkmann 1992). The actual renal O2 sensor during both rest and exercise most likely contributes is probably a heme protein (Goldberg et al. 1998), to increased cardiac contractility, cardiac output, or an enzyme termed hypoxia-inducible factor-α erythropoietin synthesis and energy mobilization prolyl-hydroxylase (HIF-PH) (Ivan et al. 2001; through gluconeogenesis (White et al. 1995). Jaakkola et al. 2001). In the presence of decreased O2 delivery, activation of this sensor appears to lead to the synthesis of HIF-1α, which then binds to the Erythropoietin active site on the enhancer region of the EPO gene Erythropoietin (EPO) is a glycoprotein hormone and activates transcription (Semenza et al. 1991). that is released primarily from the kidneys and to This immediate increase in EPO stimulates the a lesser degree by the liver. Summarized results hematological cascade which eventually leads to 452 chapter 30

an increase in RBC production if the altitude sojourn the kidney. Renin cleaves the circulating substrate is prolonged. angiotensinogen to angiotensin I, which is then The reason for the decrease in EPO to near sea further metabolized to its active form, angiotensin level values despite continuing hypoxemia is dif- II, by angiotensin-converting enzyme in the lung ficult to explain. Just like any other hormone, there vascular endothelium. Angiotensin II, a potent is a dynamic equilibrium between production and vasoconstrictor, stimulates secretion of aldosterone consumption. If increased erythroid activity follow- (ALDO) by the adrenal cortex. Renin is usually meas- ing chronic altitude exposure leads to increased EPO ured by incubating the sample to be assayed and consumption, plasma levels of EPO may decline in measuring the amount of angiotensin I generated. the face of elevated EPO turnover. Furthermore, This measures the plasma renin activity (PRA) of there may be other stimuli that change follow- the sample. Summarized results from 25 studies ing chronic altitude exposure such as increasing suggest a 9% decrease in resting PRA levels upon 2,3 diphosphoglycerate levels (Miller et al. 1973), acute altitude exposure that declines a further 10% decreasing sympathetic stimulation (Mazzeo & from acute to chronic altitude exposure (Fig. 30.6a). Reeves 2003) and down-regulation of HIF-1α The trigger for the slight and continued decrease in (D’Angelo et al. 2003), which could explain declin- resting PRA and plasma renin with altitude expos- ing EPO levels despite persistent systemic hypoxia. ure is also unknown, given that renin is typically released in response to β-adrenergic stimulation and decreasing blood pressure (Bouissou et al. 1989). Fluid regulatory hormones Since β-adrenergic stimulation is increased with Renin–angiotensin–aldosterone system. Renin is a gly- acute altitude exposure and decreased with chronic coprotein released from the juxtaglomerular cells of altitude exposure, one would expect a parallel

% % % % SL AA CA SL-AA AA-CA Rest SL AA CA SL-AA AA-CA 7 Rest 2.2 2.0 1.8 –9 –10 Exercise Rest 326 233 206 –29 –10 ) –2 –30 –1 Exercise 5.5 5.4 6.0 11 1000 Exercise 897 629 706 12 h % (R-Ex) 150 170 233 % (R-Ex) 175 169 244

–1 6 )

–1 800 5

4 600

3 400 2 200

1 Aldosterone (pmol·L

Plasma renin activity (ng·mL 0 0 (a)SL AA CA (b)SL AA CA

Fig. 30.6 Summarized results from 25 studies measuring (a) plasma renin activity during both rest and exercise (Kotchen et al. 1973a; Frayser et al. 1975; Maher et al. 1975; Sutton et al. 1977; Heyes et al. 1982; Milledge et al. 1983a; Bärtsch et al. 1988, 1991a, 1991b; Bouissou et al. 1988, 1989; Knudtzon et al. 1989b; Tunny et al. 1989; De Angelis et al. 1992, 1996; Olsen et al. 1992; Ramirez et al. 1992; Vuolteenaho et al. 1992; Rock et al. 1993; Westendorp et al. 1993; Antezana et al. 1995; Angelini et al. 1997; Zaccaria et al. 1998; Bestle et al. 2002; Robach et al. 2002) and 28 studies measuring (b) aldosterone during both rest and exercise (Frayser et al. 1975; Maher et al. 1975; Sutton et al. 1977; Claybaugh et al. 1982; Heyes et al. 1982; Milledge et al. 1983a, 1989; Maresh et al. 1984; Bärtsch et al. 1988, 1991a, 1991b; Bouissou et al. 1988, 1989; Knudtzon et al. 1989b; Tunny et al. 1989; De Angelis et al. 1992, 1996; Olsen et al. 1992; Ramirez et al. 1992; Vuolteenaho et al. 1992; Loeppky et al. 1993; Rock et al. 1993; Westendorp et al. 1993; Antezana et al. 1995; Angelini et al. 1997; Zaccaria et al. 1998; Bestle et al. 2002; Robach et al. 2002) during sea level (SL), acute altitude (AA) and chronic altitude (CA) exposure. hormones and altitude 453

increase and decrease in PRA and plasma renin. The natriuresis and diuresis upon acute and chronic hypoxic suppression of PRA with acute and chronic altitude exposure. These physiological adaptations altitude exposure may be related to the increase would be beneficial for increasing [Hb] and O2 in atrial natriuretic peptide with acute altitude transport as well as decreasing brain edema and exposure (Johnston et al. 1989), or stimulation of an thus AMS (Hackett 1999). intrarenal baroreceptor due to increased renal per- fusion following chronic altitude exposure (Hogan Atrial natriuretic peptide. Atrial natriuretic peptide et al. 1973; Olsen 1995). (ANP) is a polypeptide released by the cardiac atrial Similar to exercise at sea level, PRA or plasma muscle fibers in response to overstretch of the atria. renin levels increase one to threefold with exercise Although the literature is conflicting, summarized during both acute and chronic altitude exposure results of 21 studies suggest that plasma ANP (Maher et al. 1975; Bouissou et al. 1988; Bärtsch et al. increases 26% upon acute altitude exposure and 1991a; Rock et al. 1993). Summarized results suggest returns to sea level values following chronic altitude that the percentage of increase in PRA or renin exposure (Fig. 30.7a). The initial stimulus for ANP during exercise from resting values is the same in release may be the hypoxic pulmonary vasocon- all conditions (~ 150–233%), and absolute levels striction that occurs upon acute exposure to alti- reached during exercise are also similar in all con- tude, but does not explain the decrease in ANP ditions (Fig. 30.6a). given the sustained increase in pulmonary arter- Although suspected for some time, it is now ial pressure following chronic altitude exposure generally agreed that the production of angiotensin (Boussuges et al. 2000). Given the stimulatory effects II and activity of angiotensin-converting enzyme of HIF-1α on transactivation of ANP in hypoxic are not inhibited by hypoxia (Milledge & Catley atrial myocytes (Chen, Y.F. et al. 1997), up- and 1987; Raff et al. 1989). Thus, ALDO secretion down-regulation of HIF-1α with acute and chronic remains under predominant control of the renin– altitude exposure may explain increasing and de- angiotensin system even at altitude (Olsen 1995). creasing ANP levels (D’Angelo et al. 2003). Thus, it is not surprising that the summarized Similar to exercise at sea level, ANP levels results of 28 studies suggest that resting levels of increase 50–150% with exercise during both acute ALDO are also reduced 30–50% from sea level and chronic altitude exposure (Bärtsch et al. 1991a; values upon acute and chronic altitude exposure, Olsen et al. 1992; Rock et al. 1993; Braun et al. 2000). respectively (Fig. 30.6b). Although the most likely Summarized results suggest that the percentage of reason for the decreased ALDO levels with acute increase in ANP during exercise is lower following and chronic altitude exposure is the decreased renin chronic altitude exposure (~ 3%) compared to sea values (Olsen 1995), a decrease in plasma K+ and level (~ 88%) and acute altitude exposure (~ 72%) ACTH with altitude acclimatization may also con- (Fig. 30.7a). The immediate increase in ANP likely tribute (Okazaki et al. 1984). contributes to the increased diuresis and reduction Similar to exercise at sea level, ALDO levels in plasma volume observed upon acute altitude increase one to threefold with exercise during both exposure. acute and chronic altitude exposure (Sutton 1977; Bouissou et al. 1988; Bärtsch et al. 1991a; Braun et al. Antidiuretic hormone. Antidiuretic hormone (ADH) 2000; Beidleman et al. 2002). Summarized results is a peptide hormone stored in the posterior pitu- suggest that the percentage of increase in ALDO itary and released in response to increased osmotic during exercise from resting values is the same in all pressure. The summarized results of 16 studies conditions (~ 169–244%), but reaches lower abso- suggest that ADH increases 10% with acute altitude lute levels during acute and chronic altitude expos- exposure and returns to sea level values follow- ure due to lower initial starting levels (Fig. 30.6b). ing chronic altitude exposure (Fig. 30.7b). Due to The decline in ALDO levels during both rest and the limited number of studies, the ADH response exercise most likely contributes to the increased to exercise cannot be summarized. Bartsch and 454 chapter 30

% % % % SL AA CA SL AA CA SL-AA AA-CA Rest SL-AA AA-CA

) Rest 11.6 14.7 11.6 26 –21 Exercise Rest 2.3 2.6 2.3 10 –10 –1 3.0 25 Exercise 21.8 25.3 12.0 16 –53 ) % (R-Ex) 88 72 3 –1 2.5 20 2.0 15 1.5 10 1.0

5 0.5 Antidiuretic hormone (pmol·L Atrial natriuretic peptide (pmol·L 0 0.0 (a)SL AA CA (b)SL AA CA

Fig. 30.7 Summarized results from 21 studies measuring (a) atrial natriuretic peptide during both rest and exercise (Bärtsch et al. 1988, 1991a, 1991b; Bouissou et al. 1989; Kawashima et al. 1989; Milledge et al. 1989; Tunny et al. 1989; Koller et al. 1991; De Angelis et al. 1992, 1996; Olsen et al. 1992; Ramirez et al. 1992; Vuolteenaho et al. 1992; Loeppky et al. 1993; Rock et al. 1993; Westendorp et al. 1993; Antezana et al. 1995; Rostrup 1998; Zaccaria et al. 1998; Bestle et al. 2002; Robach et al. 2002) and 16 studies measuring (b) antidiuretic hormone at rest (Singh et al. 1974; Claybaugh et al. 1982; Heyes et al. 1982; Blume et al. 1984; Porchet et al. 1984; Bärtsch et al. 1988, 1991a, 1991b; Koller et al. 1991; De Angelis et al. 1992, 1996; Loeppky et al. 1993; Rostrup 1998; Sevre et al. 2001; Robach et al. 2002; Maresh et al. 2004) during sea level (SL), acute altitude (AA) and chronic altitude (CA) exposure. colleagues (Bärtsch et al. 1991a) reported the oppose the beneficial effects of diuresis, the reason same percentage of increase in ADH during exercise for the slight increase in ADH upon acute altitude both at sea level and upon acute altitude exposure, exposure is not clear. while Robach et al. (2002) found enhanced ADH release during exercise following chronic altitude Glucogregulatory hormones exposure. Clearly, further research needs to be con- ducted before definitive insight can be gained on Insulin. Insulin (INS) is a polypeptide hormone the ADH response to exercise at altitude. The slight secreted in response to high blood glucose levels by increase in ADH upon acute altitude exposure beta cells in the islets of Langerhans. Summarized could be the result of an increase in peripheral results from 13 studies suggest that plasma INS chemoreceptor activity (Share & Levy 1966) or increases 12% upon acute altitude exposure and plasma osmolality (Bestle et al. 2002; Maresh et al. returns toward but does not reach sea level values 2004), but doesn’t explain the decrease in ADH as following chronic altitude exposure (Fig. 30.8a). The chemoreceptor activity (Bisgard & Forster 1996) and initial increase and subsequent decrease in INS plasma osmolality (Blume et al. 1984; De Angelis secretion with altitude exposure is not entirely et al. 1992; Bestle et al. 2002; Maresh et al. 2004) known, since resting blood glucose levels are not increase following chronic altitude exposure. Heyes altered by acute altitude exposure (Young, P.M. et al. (1982) found that ADH levels were inversely et al. 1989b; Roberts et al. 1996b; Braun et al. 2001) correlated with blood pressure. Thus, the acute fall and tend to decrease following chronic altitude in blood pressure upon acute exposure to altitude exposure (Stock et al. 1978; Young, P.M. et al. 1989b; may stimulate an increase in ADH, while the Roberts et al. 1996b). The increase in INS with acute chronic increase in blood pressure following alti- altitude exposure may be due to β-adrenergic stimu- tude residence may cause ADH levels to return to lation. The subsequent decrease in INS following normal. Given that an increase in ADH would chronic altitude exposure may be due to rising NE hormones and altitude 455

% % % % SL AA CA SL-AA AA-CA Rest SL AA CA SL-AA AA-CA Rest 79 89 83 12 –7 Exercise Rest 3.0 2.7 2.1 –8 –22 Exercise 48 45 67 –6 48 Exercise 8.8 11.7 7.8 10 –33 100 14 % (R-Ex) 39 49 19 % (R-Ex) 193 333 271 )

–1 12 80 )

–1 10

60 8

6 40

Insulin (pmol·L 4 20 2 Growth hormone (ng·mL 0 0 (a)SL AA CA (b)SL AA CA

Fig. 30.8 Summarized results from 13 studies measuring (a) insulin during both rest and exercise (Sutton 1977; Stock et al. 1978; Sawhney et al. 1986; Young, P.M. et al. 1989a, 1992; Brooks et al. 1991; Sawhney et al. 1991; Young, A.J. et al. 1991; Roberts et al. 1996b; Larsen et al. 1997; Braun et al. 2000, 2001; Beidleman et al. 2002) and nine studies measuring (b) growth hormone during rest (Sutton et al. 1970; Sutton 1977; Raynaud et al. 1981; Kjær et al. 1988; Knudtzon et al. 1989a; Sawhney et al. 1991; Banfi et al. 1994; Larsen et al. 1997; Beidleman et al. 2002) during sea level (SL), acute altitude (AA) and chronic altitude (CA) exposure.

levels, which would suppress INS release via en- summarized results would be skewed by the low α hanced 2-adrenergic stimulation (Schmidt et al. number of GLG measurements at altitude, individ- 1997; Schäfers et al. 1999). ual study results will be examined. Given the anti- Similar to exercise at sea level, INS levels decrease insulin properties of GLG, one would expect a slight from 10% to 50% with exercise during both acute decrease upon acute altitude exposure that returns and chronic altitude exposure (Sutton 1977; Brooks toward normal sea level values following chronic et al. 1991; Young, P.M. et al. 1992; Braun et al. 2000; altitude exposure. Most (Kjær et al. 1988; Roberts Beidleman et al. 2002). Summarized results sug- et al. 1996b; Larsen et al. 1997) but not all (Beidleman gest that the percentage of decrease in INS during et al. 2002) studies have reported a slight 5% exercise from resting values is slightly less follow- decrease in GLG upon acute exposure to altitude. ing chronic altitude exposure (~ 19%) compared This slight decrease in GLG may facilitate enhanced to sea level (–39%) or acute altitude exposure glycolysis. However, only two studies have ex- (–49%) (Fig. 30.8a). However, if the studies were amined the resting GLG response following chronic conducted at the same relative intensity of exercise altitude exposure, and one reported a 55% increase at both sea level and altitude, the percentage of (Roberts et al. 1996b) while the other reported no decline in INS during exercise is similar (Braun change (Larsen et al. 1997). Thus, whether or not et al. 2000; Beidleman et al. 2002). The increased INS GLG changes following chronic altitude exposure levels upon acute altitude exposure may contribute remains unknown. to the preferential use of carbohydrates in order Similar to exercise at sea level, GLG increases to maximize O2 fuel efficiency under limited O2 10–30% during exercise at altitude (Roberts et al. conditions. 1996b; Beidleman et al. 2002), which is most likely related to the exercise-induced depletion of glucose. Glucagon. Glucagon (GLG) is also a polypeptide It does not appear that the GLG response to exercise hormone secreted in response to low blood glucose is potentiated by altitude exposure, but further levels by α cells in the Islets of Langerhans. Since research is needed in this area. 456 chapter 30

% % % % SL AA CA SL AA CA 14 SL-AA AA-CA Rest SL-AA AA-CA Rest 2.1 3.3 2.3 61 –30 160 Rest 87 148 117 70 –21 )

–1 12 140 ) 10 –1 120 100 8 80 6 60 4 40 Thyroxine (nmol·L 2 Triiodothyronine (nmol·L Triiodothyronine 20 0 0 (a)SL AA CA (b)SL AA CA

Fig. 30.9 Summarized results from 10 studies measuring (a) triiodothyronine (T3) and (b) thyroxine (T4) at rest (Surks et al. 1967; Kotchen et al. 1973b; Rastogi et al. 1977; Stock et al. 1978; Wright 1979; Mordes et al. 1983; Chakraborty et al. 1987; Férézou et al. 1988, 1993; Sawhney & Malhotra 1991) during sea level (SL), acute altitude (AA) and chronic altitude (CA) exposure.

secreted by the thyroid gland in response to Growth hormone thyroid-stimulating hormone (TSH) secreted by the Growth hormone (GH) is a polypeptide that is syn- anterior pituitary gland. Summarized results from thesized and secreted by cells in the anterior pitu- 10 studies suggest a 61% increase in resting T3 and itary gland. Although the literature is controversial, T4 upon acute altitude exposure that returns toward summarized results from nine studies suggest an sea level values following chronic altitude exposure 8% decrease in resting GH levels upon acute alti- (Fig. 30.9a,b). Given that most studies do not find an tude exposure that decline a further 22% from acute increase in TSH with altitude exposure, the reason to chronic altitude exposure (Fig. 30.8b). During for the acute increase and subsequent decrease in acute and chronic altitude exposure, the GH res- T3 and T4 with altitude exposure is unknown. ponse is elevated with exercise conducted at the However, since the thyroid gland contains β-adren- same absolute intensity (Sutton et al. 1970; Sutton, ergic receptors, the acute increase and subsequent 1977; Raynaud et al. 1981) but is actually decreased decrease in EPI levels may explain the thyroid hor- (Beidleman et al. 2002; Gutierrez et al. 2003) or mone response. unchanged (Kjær et al. 1988; Knudtzon et al. 1989a) Only one study has examined the thyroid hor- when exercise is conducted at the same relative mone response to exercise following chronic alti- intensity. The reason for a hypoxia-induced blunt- tude exposure, and they reported that the T3 and T4 ing of the GH response to exercise may be related to response to exercise was potentiated by altitude elevated cortisol levels directly inhibiting GH secre- exposure (Stock et al. 1978). However, the work load tion, or indirectly inhibiting GH secretion through was kept constant at both sea level and altitude in stimulation of somatostatin, an inhibitory stimulus this study, and the subsequent higher absolute to GH release (Chen, X.Q. & Du 2000). Given the intensity of exercise at altitude most likely con- lipolytic actions of GH, a decrease in GH hormone tributed to the greater increase in thyroid hormones. during acute and chronic altitude exposure may The transient elevation in basal metabolic rate enhance glycolysis and carbohydrate utilization. during the first few days at altitude is most likely mediated by increased thyroid hormone levels (Surks et al. 1967; Moore et al. 1987). Thyroid hor- Thyroid hormones mones may also play a permissive role in increasing

Triiodothyronine (T3) and thyroxine (T4) are both erythrocyte mass at altitude because an increase in hormones and altitude 457

metabolic rate increases O2 demand and thus ery- due to a direct effect of hypoxia on the testes or thropoiesis (Das et al. 1975). The physiologic adapta- indirect effect of lowered LH levels and elevated E2 tions stimulated by the thyroid hormones serve to levels (Sawhney et al. 1985; Friedl et al. 1988). increase O2 delivery under limited O2 supply. Integration of physiologic function and Reproductive hormones neurohumoral response at altitude

Estradiol and progesterone. Estradiol (E2) and proges- The integrated neurohumoral response to exercise terone (P4) are both secreted from the ovaries in and adaptations at altitude is complex (Fig. 30.10). response to changing levels of follicle-stimulating Some hormones exhibit an immediate increase hormone (FSH) and luteinizing hormone (LH) (i.e. EPO, EPI, COR, T3, T4) upon acute exposure to secreted from the anterior pituitary. Resting E2 and altitude followed by a decline to baseline sea level P4 are not changed with either acute (Beidleman values following chronic altitude exposure. Other et al. 1999; Reeves et al. 2001; Rock et al. 2001; Takase hormones exhibit a progressive decrease (ALDO, et al. 2002) or chronic altitude exposure (Reeves et al. PRA, GH) or no change (ADH, INS, GLG) with 2001; Rock et al. 2001) as long as women remain time at altitude while another hormone exhibits within the same phase of the menstrual cycle. a slow but continual increase (NE). Regardless

Similar to exercise at sea level, E2 and P4 increase of the complexity, it is important to understand 10–50% during exercise at altitude, and the response how neurohumoral responses influence physiologic is dependent on the intensity of exercise (Beidleman function at altitude because the neurohumoral et al. 1999). response often precedes physiologic adaptations. If the neurohumoral response is absent, it is unclear Testosterone. Testosterone is secreted by the testes in whether the physiologic adaptations described pre- response to changing levels of FSH and LH. Most viously would occur. (Vaernes et al. 1984; Sawhney et al. 1985; Friedl et al. The sympathoadrenal system plays a key role in 1988) but not all (Vander et al. 1978) studies report successful adaptation to altitude. Specifically, the that testosterone is decreased by acute and chronic immediate increase in EPI levels upon exposure to exposure to high altitude. Similarly, most report altitude improves O2 transport by increasing heart that LH and FSH decrease with acute and chronic rate, cardiac output and peripheral vasodilation via altitude exposure (Humpeler et al. 1980; Sawhney β-adrenergic stimulation (Mazzeo & Reeves 2003). et al. 1985; Friedl et al. 1988). The stimulus for this Increased EPI levels also augment glycolysis which decrease in testosterone is unknown, but may be increases the efficiency of O2 utilization. Increased

NE

300 EPI, COR, T3, T4 EPO PRA, ALDO, GH 200 ADH, INS, GLG

Fig. 30.10 Integrated neurohumoral response to altitude. ADH, 100 antidiuretic hormone; ALDO, aldosterone; COR, cortisol; EPI, % Change epinephrine; EPO, erythropoietin; 0 GH, growth hormone; GLG, glucagon; INS, insulin; NE, norepinephrine; PRA, plasma –100 SL 2 14 renin activity; T3, triiodothyronine; Days at altitude T4, thyroxine. 458 chapter 30

COR levels upon acute exposure to altitude also following chronic altitude exposure, the hormonal augment the actions of EPI by stimulating muscle response may be down-regulated in an appropriate and liver glycogenolysis. The cost of these changes fashion. is an increased metabolic rate and lactate produc- The fluid regulatory hormones are critical for tion (Mazzeo & Reeves 2003). As ventilatory acclim- successfully adapting to altitude. The progressive atization proceeds with time at altitude, arterial decline in PRA and ALDO contribute to the pro-

O2 content improves, the hypoxic stress subsides, gressive decline in plasma volume and extracellular EPI and COR levels fall, and heart rate, cardiac out- fluid volume with altitude exposure. These physio- put and lactate production return to baseline sea logic changes not only improve O2 transport but level values. The progressive increase in peripheral also decrease brain edema and thus AMS. In fact, chemoreceptor activity at altitude appears to con- many studies have reported that when PRA and tribute to the rising NE levels with time at altitude. ALDO are elevated, due to extensive time climbing The immediate increase in SNS activity upon acute to altitude, significant sodium and water retention exposure to altitude counterbalances the effects and increased incidence of AMS occur (Milledge of hypoxia-induced vasodilation on most vascular et al. 1983b, 1989; Bärtsch et al. 1988, 1991a). This beds so that blood pressure can be maintained suggests that if at all possible, exercise should be (Wolfel & Levine 2001). The progressive rise in SNS avoided within the first 24 h of altitude exposure activity with time at altitude contributes to increased (Roach et al. 2000). Given that ADH does not change arterial and venous tone, increased systemic vascu- significantly with altitude exposure, it does not lar resistance, decreased plasma volume and further appear to play a significant role in the altitude- decreases in cardiac output (Grover et al. 1986; induced alterations in fluid balance. Wolfel & Levine 2001). The cost of increased SNS The metabolic hormones likely play a role in sub- activity is a decrease in blood flow following strate utilization at altitude but they are less under- chronic altitude exposure. However, this allows for stood due to conflicting results in the experimental a more precise matching of metabolic supply and literature. Several studies have suggested increased demand by as arterial O2 content improves with utilization of carbohydrates upon acute and chronic acclimatization. exposure to altitude (Brooks et al. 1991; Roberts et al. The acute increase in EPO, which stimulates the 1996b) and other studies have reported decreased hematological cascade, also contributes to success- utilization of carbohydrates following chronic alti- ful altitude adaptation. Although there is a time lag tude exposure (Young, A.J. et al. 1982, 1991; Braun before RBC mass increases at altitude, the eventual et al. 2000). The acute increase in INS, small decrease α increase allows for improved O2 delivery. Animal in GLG, and up-regulation of HIF-1 upon expos- evidence suggests that the acute increase in EPO ure to altitude would support increased carbo- may also protect against ischemic brain injury hydrate utilization. Subsequent down-regulation of and, possibly, AMS (Brines et al. 2000). Increases in HIF-1α and decreased INS levels following chronic COR and the thyroid hormones upon acute expos- altitude exposure would support decreased carbo- ure to altitude may also increase EPO synthesis hydrate utilization, which has not been consistently (Golde et al. 1977; Kelly et al. 2000). Although the reported (Brooks et al. 1991; Roberts et al. 1996b). decrease in EPO following chronic altitude expos- Conflicting substrate results in the altitude liter- ure does not make intuitive sense, the benefits of ature may be due to differences in study design, expanded red cell volume may be detrimental to intensity of exercise, gender, diet control, level of blood flow due to an increase in blood viscosity hypoxemia and other unknown factors. Although it (Young, A.J. et al. 1996). Furthermore, HIF-1α, is clear that substrate utilization is altered with which is down-regulated with continued exposure exposure to altitude, the timing and mechanisms to hypoxia (D’Angelo et al. 2003), exerts a direct need further study. effect on transcription of EPO (Semenza et al. 1991). The reproductive hormones do not appear to play

Thus, as the body senses improved O2 transport a key role in successful adaptation to altitude. hormones and altitude 459

However, they may influence the low fertility rates summarizing the neurohumoral responses from a reported in several high altitude populations large number of studies under a defined set of con- (Clegg & Harrison 1971). One study suggested ditions, a clearer understanding concerning how lower luteal-phase progesterone values in sea level these responses affect physiologic adaptation at natives compared to high altitude natives (Escudero altitude has been gained. The hormonal trends pre- et al. 1996). Another study reported decreased testo- sented for the sympathoadrenal (EPI, NOR, COR), sterone levels and sperm count following a moun- hematologic (EPO) and fluid regulatory (PRA, taineering expedition to 5100 m that was reversed ALDO, ANP) response to altitude are consistent and only after a 3-month stay at sea level (Okumara clear due to the large number of studies examined. et al. 2003). Although the reproductive hormones However, further work examining the metabolic may exert effects on ventilation, erythropoiesis, hormones (INS, GH, GLG) and their response to and metabolism at sea level, their effects on these altitude is needed both during rest and exercise to physiologic adaptations at altitude are minimal clarify outstanding issues. (Muza et al. 2001; Beidleman et al. 2002). In conclusion, many neurohumoral responses Acknowledgments contribute to the short- and long-term physiologic adaptations that occur during both rest and exercise The authors would like to thank Dr. Charles Fulco at altitude that ultimately allows the healthy low- and Mr. Matthew Stueck for their editorial and lander to tolerate the hypoxemia of altitude. By research assistance in preparing this book chapter.

References

Abbrecht, P.H. & Littell, J.K. (1972) Plasma adrenal axis following glucocorticoid altitude exposure. Medicine and Science erythropoietin in men and mice during prophylaxis against acute mountain in Sports and Exercise 34, 430–437. acclimatization to different altitudes. sickness. Hormones and Metabolic Beidleman, B.A., Muza, S.R., Fulco, C.S. Journal of Applied Physiology 32, 54–58. Research 34, 318–324. et al. (2003) Intermittent altitude Angelini, C., Cogo, A., Madrigale, G., Bärtsch, P., Shaw, S., Franciolli, M., exposures improve muscular Pecchio, O. & Posani, L. (1997) Increased Gnadinger, M.P. & Weidmann, P. performance at 4300 m. Journal of aldosterone levels in acute mountain (1988) Atrial natriuretic peptide in acute Applied Physiology 95, 1824–1832. sickness at Capanna Regina Margherita. mountain sickness. Journal of Applied Bender, P.R., Groves, B.M., McCullough, Wilderness and Environmental Medicine 8, Physiology 65, 1929–1937. R.E. et al. (1988) Oxygen transport to 247–249. Bärtsch, P., Maggiorini, M., exercising leg in chronic hypoxia. Journal Antezana, A.M., Richalet, J.P., Noriega, I., Schobersberger, W. et al. (1991a) of Applied Physiology 65, 2592–2597. Galarza, M. & Antezana, G. (1995) Enhanced exercise-induced rise of Bestle, M.H., Olsen, N.V., Poulsen, T.D. Hormonal changes in normal and aldosterone and vasopressin preceeding et al. (2002) Prolonged hypobaric polycythemic high-altitude natives. mountian sickness. Journal of Applied hypoxemia attenuates vasopressin Journal of Applied Physiology 79, 795–800. Physiology 71, 136–143. secretion and renal response to Bailey, D.M., Davies, B., Milledge, J.S. et al. Bärtsch, P., Pfluger, N., Audetat, M. et al. osmostimulation in men. Journal of (2000) Elevated plasma cholecystokinin (1991b) Effects of slow ascent to 4559 m Applied Physiology 92, 1911–1922. at high altitude: metabolic implications on fluid homeostasis. Aviation Space and Bisgard, G.E. & Forster, H.V. (1996) for the anorexia of acute mountain Environmental Medicine 62, 105–110. Ventilatory responses to acute and sickness. High Altitude Medicine and Beidleman, B.A., Muza, S.R., Rock, P.B. chronic hypoxia. In: Handbook of Biology 1, 9–23. et al. (1997) Exercise responses after Physiology, Section 4: Environmental Banfi, G., Marinelli, M., Roi, G.S. et al. altitude acclimatization are retained Physiology (Fregley, M.J. & Blatteis, C.M., (1994) Growth hormone and insulin-like during reintroduction to altitude. eds.). Oxford University Press, New growth factor I in athletes performing a Medicine and Science in Sports and York: 1207–1257. marathon at 4000 m of altitude. Growth Exercise 29, 1588–1595. Blume, F.D., Boyer, S.J., Braverman, L.E. Regulation 4, 82–86. Beidleman, B.A., Rock, P.B., Muza, S.R. et al. (1984) Impaired osmoregulation at

Banfi, G., Marinelli, M., Bonini, P., Gritti, I. et al. (1999) Exercise Ve and physical high altitude. Studies on Mt. Everest. & Roi, G.S. (1996) Pepsinogens and performance are not affected by Journal of the American Medical Association gastrointestinal symptoms in mountain menstrual cycle phase at altitude. Journal 252, 524–526. marathon runners. International Journal of Applied Physiology 86, 1519–1526. Bonfichi, M., Baldini, A., Arcaini, L. et al. of Sports Medicine 17, 554–558. Beidleman, B.A., Rock, P.B., Muza, S.R. (2000) Haematological modifications Basu, M., Sawhney, R.C., Kumar, S. et al. et al. (2002) Substrate oxidation is altered after acute exposure to high altitude: (2002) Hypothalamic–pituitary– in women during exercise upon acute possible implications for detection of 460 chapter 30

recombinant erythropoietin misuse. cultured atrial cardiocytes. Hypertension lipemia after acute exposure to high British Journal of Haematology 109, 29, 75–82. altitude hypoxia. International Journal of 895–896. Claybaugh, J.R., Wade, C.E., Sato, A.K. Sports Medicine 14, 78–85. Bouissou, P., Guezennec, C.Y., Galen, F.X. et al. (1982) Antidiuretic hormone Firth, J.D., Ebert, B.L. & Ratcliffe, P.J. et al. (1988) Dissociated response of responses to eucapnic and hypocapnic (1995) Hypoxic regulation of lactate aldosterone from plasma renin activity hypoxia in humans. Journal of Applied dehydrogenase A. Interaction between during prolonged exercise under Physiology 53, 815–823. hypoxia-inducible factor 1 and cAMP hypoxia. Hormones and Metabolic Clegg, E.J. & Harrison, G.A. (1971) response elements. Journal of Biological Research 20, 517–521. Reproduction in human high altitude Chemistry 270, 21 021–21 027. Bouissou, P., Richalet, J.P., Galen, F.X. et al. populations. Hormones 2, 13–25. Fitzgerald, R.S. & Dehghani, G.A. (1982) (1989) Effect of β-adrenoceptor blockade D’Angelo, G., Duplan, E., Boyer, N., Neural responses of the cat carotid on renin–aldosterone and α-ANF during Vigne, P. & Frelin, C. (2003) Hypoxia and aortic bodies to hypercapnia and exercise at altitude. Journal of Applied up-regulated prolyl hydroxylase hypoxia. Journal of Applied Physiology 52, Physiology 67, 141–146. activity: a feedback mechanism that 596–601. Boushel, R., Calbet, J.L., Radegran, G. et al. limits HIF-1 responses during Forster, H.V., Dempsey, J.A. & Chosy, (2001) Parasympathetic neural activity reoxygenation. Journal of Biological L.W. (1975) Incomplete compensation of accounts for the lowering of exercise Chemistry 278, 38 183–38 187. CSF [H+] in man during acclimatization heart rate at high altitude. Circulation Das, K.C., Mukherjee, M., Sarkar, T.K., to high altitude (4300 m). Journal of 104, 1785–1791. Dash, R.J. & Rastogi, G.K. (1975) Applied Physiology 38, 1067–1072. Boussuges, A., Molenat, F., Burnet, H. et al. Erythrpoiesis and erythropoietin in Frayser, R., Rennie, I.D., Gray, G.W. & (2000) Operation Everest III (Comex ’97): hypo- and hyperthryroidism. Journal of Houston, C.S. (1975) Hormonal and Modifications of cardiac function Clinical Endocrinology and Metabolism 40, electrolyte response to exposure to secondary to altitude-induced hypoxia. 211–220. 17 500 ft. Journal of Applied Physiology An echocardiographic and doppler De Angelis, C., Farrace, S., Urbani, L. et al. 38, 636–642. study. American Journal of Respiratory and (1992) Effects of high altitude exposure Friedl, K.E., Plymate, S.R., Bernhard, W.N. Critical Care Medicine 161, 264–270. on plasma and urinary digoxin-like & Mohr, L.C. (1988) Elevation of plasma Braun, B., Mawson, J.T., Muza, S.R. et al. immunoreactive substance. American estradiol in healthy men during a (2000) Women at altitude: carbohydrate Journal of Hypertension 5, 600–607. mountaineering expedition. Hormones utilization during exercise at 4300 m. De Angelis, C., Ferri, C., Urbani, L. and Metabolic Research 20, 239–242. Journal of Applied Physiology 88, 246–256. & Farrace, S. (1996) Effect of acute Fulco, C.S., Cymerman, A., Rock, P.B. Braun, B., Rock, P.B., Zamudio, S. et al. exposure to hypoxia on electrolytes and & Farese, G. (1985) Hemodynamic (2001) Women at altitude: short-term water metabolism regulatory hormones. responses to upright tilt at sea level exposure to hypoxia and/or α- Aviation Space and Environmental and high altitude. Aviation Space and adrenergic blockade reduces insulin Medicine 67, 746–750. Environmental Medicine 56, 1172–1176. sensitivity. Journal of Applied Physiology Easton, P.A., Slykerman, L.J. & Fulco, C.S. & Cymerman, A. (1988) Human 91, 623–631. Anthonisen, N.R. (1986) Ventilatory performance and acute hypoxia. In: Brines, M.L., Ghezzi, P., Keenan, S. et al. response to sustained hypoxia in normal Human Performance Physiology and (2000) Erythropoietin crosses the adults. Journal of Applied Physiology 61, Environmental Medicine at Terrestrial blood–brain barrier to protect against 906–911. Extremes (Pandolf, K.B., Sawka, M.N. & experimental brain injury. Proceedings Eckardt, K.U., Boutellier, U., Kurtz, A. et al. Gonzalez, R.R., eds.). Benchmark Press, of the National Academy of Sciences (1989) Rate of erythropoietin formation Indianapolis: 467–495. of the United States of America 97, in humans in response to acute Fulco, C.S., Cymerman, A., Muza, S.R. et al. 10 526–10 531. hypobaric hypoxia. Journal of Applied (1994) Adductor pollicis muscle fatigue Brooks, G.A., Butterfield, G.E., Wolfe, R.R. Physiology 66, 1785–1788. during acute and chronic altitude et al. (1991) Increased dependence on Ekblom, B., Hout, R., Stein, E.M. & exposure and return to sea level. Journal blood glucose after acclimatization to Thorstensson, A.T. (1975) Effects of of Applied Physiology 77, 179–183. 4300 m. Journal of Applied Physiology 70, changes in arterial oxygen content on Fulco, C.S., Kambis, K.W., Friedlander, 919–927. circulation and physical performance. A.L. et al. (2003) Carbohydrate Chakraborty, S., Samaddar, J. & Batabyal, Journal of Applied Physiology 39, 71–75. supplementation improves time-trial S. (1987) Thyroid status of humans at Escudero, F., Gonzales, G.F. & Gonez, C. cycle performance at 4300 m altitude. high altitude. Clinica Chimica Acta 166, (1996) Hormone profile during the T03-07, USARIEM Technical Report. 111–113. menstrual cycle at high altitude. Natick, MA. Chapman, R.F., Stray-Gundersen, J. & International Journal of Gynaecology and Ge, R.L., Witkowski, S., Zhang, Y. et al. Levine, B.D. (1998) Individual variation Obstetrics 55, 49–58. (2002) Determinants of erythropoietin in response to altitude training. Journal Férézou, J., Richalet, J.P., Coste, T. & release in response to short-term of Applied Physiology 85, 1448–1456. Rathat, C. (1988) Changes in plasma hypobaric hypoxia. Journal of Applied Chen, X.Q. & Du, J.Z. (2000) Hypoxia lipids and lipoprotein cholesterol Physiology 92, 2361–2367. influences somatostatin release in rats. during a high altitude mountaineering Gleiter, C.H., Becker, T. & Wenzel, J. (1997) Neuroscience Letters 284, 151–154. expedition (4800 m) European Journal of Erythropoietin production in healthy Chen, Y.F., Durand, J. & Claycomb, W.C. Applied Physiology 57, 740–745. volunteers subjected to controlled (1997) Hypoxia stimulates atrial Férézou, J., Richalet, J.P., Serougne, C. hypobaric hypoxia: further evidence natriuretic peptide gene expression in et al. (1993) Reduction of postprandial against a role of adenosine. British hormones and altitude 461

Journal of Clinical Pharmacology 44, pathophysiology. Journal of the American Horstman, D., Weiskopf, R. & Jackson, R.E. 203–205. Medical Association 280, 1920–1925. (1980) Work capacity during 3-week Goldberg, M.A., Dunning, S.P. & Halliwill, J.R. (2003) Hypoxic regulation of sojourn at 4300 m: effects of relative Bunn, H.F. (1998) Regulation of the blood flow in humans. Skeletal muscle polycythemia. Journal of Applied erythropoietin gene: evidence that the circulation and the role of epinephrine. Physiology 49, 311–318. oxygen sensor is a heme protein. Science Advances in Experimental Medicine and Hoyt, R.W. & Honig, A. (1996) Body fluid 242, 1412–1415. Biology 543, 223–236. and energy metabolism at high altitude. Golde, D.W., Bersch, N., Chopra, I.J. & Hammill, S., Latham, W.W., Frost, W. In: Handbook of Physiology, Section 4: Cline, M.J. (1977) Thyroid hormones & Weil, J.V. (1979) Autonomic Environmental Physiology (Fregley, M.J. stimulate erythropoiesis in vitro. British cardiovascular control during hypoxia & Blatteis, C.M., eds.). Oxford Journal of Haematology 37, 173–177. in the dog. Circulation Research 44, University Press, New York: 1277–1289. Green, H.J., Sutton, J.R., Wolfel, E.E. et al. 569–575. Hudson, J.G., Bowen, A.L., Navia, P. (1992) Altitude acclimatization and Hansen, J.E. & Sander, M. (2003) et al. (1999) The effect of high altitude energy metabolic adaptations in skeletal Sympathetic neural overactivity in on platelet counts, thrombopoietin and muscle during exercise. Journal of Applied healthy humans after prolonged erythropoietin levels in young Bolivian Physiology 73, 2701–2708. exposure to hypobaric hypoxia. Journal airmen visiting the Andes. International Grover, R.F., Lufschanowski, R. & of Physiology 546, 921–929. Journal of Biometeorology 43, 85–90. Alexander, J.K. (1976) Alterations in the Hansen, J.E., Harris, C.W. & Evans, Hughson, R., Yamamoto, Y., McCullough, coronary circulation of man following W.O. (1967) Influence of elevation of R., Sutton, J. & Reeves, J. (1994) ascent to 3100 m altitude. Journal of origin, rate of ascent and a physical Sympathetic and parasympathetic Applied Physiology 41, 832–838. conditioning program on symptoms indicators of heart rate control at altitude Grover, R.F., Weil, J.V. & Reeves, J.T. of acute mountain sickness. Military by spectral analysis. Journal of Applied (1986) Cardiovascular adaptation to Medicine 132, 585–592. Physiology 77, 2537–2542. exercise at high altitude. In: Exercise Hartley, L.H., Vogel, J.A. & Cruz, J.C. Humpeler, E., Skrabal, F. & Bartsch, G. and Sports Science Reviews (Pandolf, K.B., (1974) Reduction of maximal exercise (1980) Influence of exposure to moderate ed.). Macmillan, New York: 269–302. heart rate at altitude and its reversal altitude on the plasma concentration of Grover, R.F., Selland, M.A., McCullough, with atropine. Journal of Applied cortisol, aldosterone, renin, testosterone, R.G. et al. (1998) β-Adrenergic blockade Physiology 36, 362–365. and gonadotropins. European Journal of does not prevent polycythemia or Hasbak, P., Lundby, C., Olsen, N.V., Applied Physiology 45, 167–176. decrease in plasma volume in men at Schifter, S. & Kanstrup, I.L. (2002) Ivan, M., Kondo, K., Yang, H. et al. (2001) 4300 m altitude. European Journal of Calcitonin gene-related peptide and HIFα targeted for VHL-mediated Applied Physiology 77, 264–270. adrenomedullin release in humans: destruction by proline hydroxylation:

Gunga, H.C., Krisch, K., Rocker, L. & effects of exercise and hypoxia. implications for O2 sensing. Science 20, Schobersberger, W. (1994) Time course Regulatory Peptides 108, 89–95. 464–468. of erythropoietin, triiodothronine, Heyes, M.P., Farber, H.W., Manfredi, F. Jaakkola, P., Mole, D.R., Tian, Y.M. et al. thyroxine, and thyroid-stimulating et al. (1982) Acute effects of hypoxia on (2001) Targeting of HIF-α to the von hormone at 2315 m. Journal of Applied renal and endocrine function in normal Hippel-Lindau ubiquitylation complex

Physiology 76, 1068–1072. humans. American Journal of Physiology by O2 regulated prolyl hydroxylation. Gutierrez, A., Gonzalez-Gross, M., Ruiz, 243, R265–R270. Science 20, 468–472. J.R., Mesa, J.L. & Castillo, M.J. (2003) Hogan, R.P., Kotchen, T.A., Boyd A.E., 3rd Jain, S.C., Bardhan, J., Swamy, Y.Y., Acute exposure to moderate high & Hartley, L.H. (1973) Effect of altitude Brishna, B. & Nayar, H.S. (1980) Body altitude decreases growth hormone on renin–aldosterone system and fluid compartments in humans during response to physical exercise in metabolism of water and electrolytes. acute high-altitude exposure. Aviation untrained subjects. Journal of Sports Journal of Applied Physiology 35, 385–390. Space and Environmental Medicine 51, Medicine and Physical Fitness 43, Honig, A. (1989) Peripheral arterial 234–236. 554–558. chemoreceptors and reflex control Jelkmann, W. (1992) Erythropoietin: Hackett, P.H. (1999) The cerebral etiology of sodium and water homeostasis. structure, control of production, and of high-altitude cerebral edema and American Journal of Physiology 257, function. Physiological Reviews 72, acute mountain sickness. Wilderness R1282–R1302. 449–489. and Environmental Medicine 10, 97–109. Honig, A., Wedler, B., Oppermann, H., Jensen, J.B., Wright, A.D., Lassen, N.A. Hackett, P.H. & Rennie, D. (1976) The Gruska, S. & Schmidt, M. (1996) Effect of et al. (1990) Cerebral blood flow in acute incidence, importance, and prophylaxis arterial chemoreceptor stimulation with mountain sickness. Journal of Applied of acute mountain sickness. Lancet 2, almitrine bismethylate on plasma renin Physiology 69, 433. 1149–1154. activity, aldosterone, ACTH and cortisol Johnston, C.I., Hodsman, P.G., Kohzuki, Hackett, P.H., Rennie, D., Hofmeister, S.E. in anaesthetized, artificially ventilated M. et al. (1989) Interaction between et al. (1982) Fluid retention and relative cats. Clinical and Experimental atrial natriuretic peptide and the renin hypoventilation in acute mountain Pharmacology and Physiology 23, 106–110. angiotensin aldosterone system. sickness. Respiration 43, 321–329. Honigman, B., Theis, M.K., Koziol-McLain, Endogenous antagonists. American Hackett, P.H., Yarnell, P.R., Hill, R. et al. J. et al. (1993) Acute mountain sickness in Journal of Medicine 87 (suppl. 6B), (1998) High-altitude cerebral edema a general tourist population at moderate 6B-24S–6B-28S. evaluated with magnetic resonance altitudes. Annals of Internal Medicine 118, Jung, R.C., Dill, D.B., Horton, R. & imaging: clinical correlation and 587–592. Horvath, S.M. (1971) Effects of age 462 chapter 30

on plasma aldosterone levels and adrenaline responses to simulated Mairbaurl, H., Schobersberger, W., Oelz,

hemoconcentration at altitude. Journal altitude. Clinical Science 44, 243–251. O. et al. (1990) Unchanged in vivo P50 of Applied Physiology 31, 593–597. Kotchen, T.A., Mongey, E.H., Hogan, R.P. at high altitude despite decreased Karim, F., Poucher, S.M. & Summerill, et al. (1973b) Thyroid responses to erythrocyte age and elevated 2,3 R.A. (1987) The effects of stimulating simulated altitude. Journal of Applied diphosphoglycerate. Journal of Applied carotid chemoreceptors on renal Physiology 34, 165–168. Physiology 68, 1186–1190. haemodynamics and function in dogs. Kraemer, W.J., Hamilton, A.J., Gordon, Maresh, C.M., Noble, B.J., Robertson, K.L. Journal of Physiology 355, 557–566. S.E. et al. (1991) Plasma changes in β- & Seip, R.L. (1984) Adrenocortical Kawashima, A., Kubo, K., Hirai, K. et al. endorphin to acute hypobaric hypoxia responses to maximal exercise in (1989) Plasma levels of atrial natriuretic and high intensity exercise. Aviation moderate-altitude natives at 447 Torr. peptide under acute hypoxia in normal Space and Environmental Medicine 62, Journal of Applied Physiology 56, 482–488. subjects. Respiration Physiology 76, 79–92. 754–758. Maresh, C.M., Kraemer, W.J., Judelson, Kelly, J.J., Martin, A. & Whitworth, Lahiri, S. & Delaney, R.G. (1975) D.A. et al. (2004) Effects of high altitude J.A. (2000) Role of erythropoietin in Relationship between carotid and water deprivation on arginine cortisol-induced hypertension. Journal of chemoreceptor activity and ventilation vasopressin release in men. American Human Hypertension 14, 195–198. in the cat. Respiration Physiology 24, Journal of Physiology 286, E20–E24. Kjær, M., Bangsbo, J., Lortie, G. & Galbo, 267–286. Marshall, J.M. (1994) Peripheral H. (1988) Hormonal response to exercise Larsen, J.J., Hansen, J.M., Olsen, N.V., chemoreceptors and cardiovascular in humans: influence of hypoxia and Galbo, H. & Dela, F. (1997) The effect of regulation. Physiological Reviews 74, physical training. American Journal of altitude hypoxia on glucose homeostasis 543–594. Physiology 254, R197–R203. in men. Journal of Physiology 504, Mazzeo, R.S. & Reeves, J.T. (2003) Klausen, T., Poulsen, T.D., Fogh-Andersen, 1241–1249. Adrenergic contribution during N. et al. (1996) Diurnal variations of Leuenberger, U., Gleeson, K. & acclimatization to high altitude: serum erythropoietin at sea level and Wroblewski, K. (1991) Norepinephrine perspectives from Pikes Peak. Exercise altitude. European Journal of Applied clearance is increased during acute and Sport Sciences Reviews 31, 13–18. Physiology 72, 297–302. hypoxemia in humans. American Journal Mazzeo, R.S., Bender, P.R., Brooks, G.A. Knudtzon, J., Bogsnes, A. & Norman, N. of Physiology 261, H1659–H1664. et al. (1991) Arterial catecholamine (1989a) Changes in prolactin and growth Lewis, R.A., Thorn, G.W., Koepf, G.F. responses during exercise with acute hormone levels during hypoxia and & Dorrance, S.S. (1942) The role of and chronic high-altitude exposure. exercise. Hormones and Metabolic Research the adrenal cortex in acute anoxia. American Journal of Physiology 261, 21, 453–454. Journal of Clinical Investigation 21, E419–E424. Knudtzon, J., Myhre, K., Rasch, W. et al. 33–46. Mazzeo, R.S., Brooks, G.A., Butterfield, (1989b) Reduced systolic blood pressure Loeppky, J.A., Roach, R.C., Selland, M.A. G.E. et al. (1995) Acclimatization to high elevations during maximum exercise at et al. (1993) Body fluid alteration during altitude increases muscle sympathetic simulated altitudes. Aviation Space and head-down bed rest in men at activity both at rest and during exercise. Environmental Medicine 60, 1072–1076. moderate altitude. Aviation Space and American Journal of Physiology 269, Knuttgen, H.G. & Saltin, B. (1973) Oxygen Environmental Medicine 64, 265–274. R201–R207. uptake, muscle high-energy phosphates, Lundby, C. & Steensberg, A. (2004) Mazzeo, R.S., Child, A., Butterfield, G.E. and lactate in exercise under acute Interleukin-6 response to exercise et al. (2000) Sympathoadrenal responses hypoxic conditions in man. Acta during acute and chronic hypoxia. to submaximal exercise in women after Physiologica Scandinavica 87, 368–376. European Journal of Applied Physiology 91, acclimatization to 4300 m. Metabolism 49, Koistinen, P.O., Rusko, H., Irjala, K. et al. 88–93. 1036–1042. (2000) EPO, red cells, and serum Lyons, T.P., Muza, S.R., Rock, P.B. & Mazzeo, R.S., Carroll, J.D., Butterfield, G.E. transferrin receptor in continuous and Cymerman, A. (1995) The effect of et al. (2001) Catecholamine responses to intermittent hypoxia. Medicine and altitude pre-acclimatization on acute α-adrenergic blockade during exercise Science in Sports and Exercise 32, 800–804. mountain sickness during reexposure. in women acutely exposed to altitude. Koller, E.A., Drechsel, S., Hess, T., Aviation Space and Environmental Journal of Applied Physiology 90, 121–126. Macherel, P. & Boutellier, U. (1988) Medicine 66, 957–962. Mazzeo, R.S., Dubay, A., Kirsch, J. Effects of atropine and propranolol on Maggiorini, M., Buhler, B., Walter, M. et al. (2003) Influence of α-adrenergic the respiratory, circulatory, and ECG & Oelz, O. (1990) Prevalence of acute blockade on the catecholamine response responses to high altitude in man. mountain sickness in the Swiss Alps. to exercise at 4300 m. Metabolism 52, European Journal of Applied Physiology 57, British Medical Journal 301, 853–855. 1471–1477. 163–172. Maher, J.T., Jones, L.G. & Hartley, L.H. Milledge, J.S. & Catley, D.M. (1987) Koller, E.A., Buhrer, A., Felder, L., (1974) Effects of high altitude exposure Angiotensin converting enzyme activity Schopen, M. & Vallotton, M.B. (1991) on submaximal endurance capacity of and hypoxia. Clinical Science 72, 149. Altitude diuresis: endocrine and man. Journal of Applied Physiology 37, Milledge, J.S. & Cotes, P.M. (1985) Serum renal responses to acute hypoxia of 895–898. erythropoietin in humans at high acclimatized and non-acclimatized Maher, J.T., Jones, L.G., Hartley, L.H., altitude and its relation to plasma subjects. European Journal of Applied Williams, G.H. & Rose, L.I. (1975) renin. Journal of Applied Physiology 59, Physiology 62, 228–234. Aldosterone dynamics during graded 360–364. Kotchen, T.A., Hogan, R.P., Boyd, A.E. exercise at sea level and high altitude. Milledge, J.S., Catley, D.M., Ward, M.P., et al. (1973a) Renin, noradrenaline and Journal of Applied Physiology 39, 18–22. Williams, E.S. & Clarke, R.A. (1983a) hormones and altitude 463

Renin–aldosterone and angiotensin- Olsen, N.V. (1995) Effect of hypoxaemia Raynaud, J., Drouet, L., Martineaud, J.P. converting enzyme during prolonged on water and sodium homeostatic et al. (1981) Time course of plasma altitude exposure. Journal of Applied hormones and renal function. Acta growth hormone during exercise in Physiology 55, 699–702. Anaesthesiology Scandanavia 39 humans at altitude. Journal of Applied Milledge, J.S., Catley, D.M., Williams, E.S., (Suppl. 107), 165–170. Physiology 50, 229–233. Withey, W.R. & Minty, B.D. (1983b) Olsen, N.V., Kanstrup, I.L., Richalet, J.P. Reeves, J.T., Grover, R.F. & Cohn, J.E. Effect of prolonged exercise at altitude et al. (1992) Effects of acute hypoxia on (1967) Regulation of ventilation during on the renin–aldosterone system. Journal renal and endocrine function at rest and exercise at 10 200 ft in athletes born at of Applied Physiology 55, 413–418. during graded exercise in hydrated low altitude. Journal of Applied Physiology Milledge, J.S., Beeley, J.M., McArthur, S. subjects. Journal of Applied Physiology 73, 22, 546–554. & Morice, A.H. (1989) Atrial natriuretic 2036–2043. Reeves, J.T., Zamudio, S., Dahms, T.E. peptide, altitude, and acute mountain Olsen, N.V., Hansen, J.M., Kanstrup, I.L., et al. (2001) Erythropoiesis in women sickness. Clinical Science 77, 509–514. Richalet, J.P. & Leyssac, P.P. (1993) Renal during 11 days at 4300 m is not affected Miller, M.E., Rorth, M., Parving, H.H. hemodynamics, tubular function, and by menstrual cycle phase. Journal of et al. (1973) pH effect on erythropoietin response to low-dose dopamine during Applied Physiology 91, 2579–2586. response to hypoxia. New England acute hypoxia in humans. Journal of Reis, D.J., Golanov, E.V., Ruggiero, D.A. Journal of Medicine 288, 706–710. Applied Physiology 74, 2166–2173. & Sun, M.K. (1994) Sympathoexcitatory Moncloa, F., Velasco, I. & Beteta, L. (1968) Porchet, M., Contat, H., Waeber, B., neurons of the rostral ventrolateral Plasma cortisol concentration and Nussberer, J. & Brunner, H.R. (1984) medulla are oxygen sensors and disappearance rate of 4-14C-cortisol in Responses of plasma arginine essential elements in the tonic and reflex newcomers to high altitude. Journal of vasopressin levels to rapid changes in control of the systemic and cerebral Clinical Endocrinology 28, 379–382. altitude. Clinical Physiology (Oxford, circulations. Journal of Hypertension. Moore, L.G., Harrison, G.L., McCullough, England) 4, 435–438. Supplement 12, S159–S180. R.E. et al. (1986) Low acute hypoxic Powles, A.P., Sutton, J.R., Gray, G.W. et al. Reynafarje, C. (1957) The influence of ventilatory response and hypoxic (1978) Sleep hypoxemia at altitude: high altitude on erythropoietic activity. depression in acute altitude sickness. its relationship to acute mountain Brookhaven Symposia in Biology 10, Journal of Applied Physiology 60, sickness and ventilatory responsiveness 132–146. 1407–1412. to hypoxia and hypercapnia. In: Richalet, J.P. (2001) The endocrine system. Moore, L.G., Cymerman, A., Huang, S. Environmental Stress: Individual Human In: High Altitude. An Exploration of et al. (1987) Propranolol blocks metabolic Adaptations (Folinsbee, L.J., Wagner, J.A., Human Adaptation (Hornbein, T.F. & rate increase but not ventilatory Borgia, J.F. et al., eds.). Academic Press, Schoene, R.B., eds.). Marcel Dekker, acclimatization to 4300 m. Respiration New York: 317–324. Inc., New York: 601–644. Physiology 70, 195–204. Pugh, L.G.C.E. (1964) Blood volume and Richalet, J.P., Mehdioui, H., Rathat, C. et al. Mordes, J.P., Blume, D., Boyer, S. et al. haemoglobin concentration at altitudes (1988) Acute hypoxia decreases cardiac (1983) High-altitude pituitary–thyroid above 18 000 ft (5500 m). Journal of response to catecholamines in exercising dysfunction on Mount Everest. New Physiology 170, 344–354. humans. International Journal of Sports England Journal of Medicine 308, Raff, H., Tzankoff, S.P. & Fitzgerald, R.S. Medicine 9, 157–162. 1135–1138. (1982) Chemoreceptor involvement in Richalet, J.P., Rutgers, V., Bouchet, P. Morgan, B., Crabtree, D., Palta, M. & cortisol responses to hypoxia in et al. (1989) Diurnal variations of acute Skatrud, J. (1995) Combined hypoxia ventilated dogs. Journal of Applied mountain sickness, colour vision, and hypercapnia evokes long-lasting Physiology 52, 1092–1096. and plasma cortisol and ACTH at sympathetic activation in humans. Raff, H., Ball, D.L. & Goodfriend, T.L. high altitude. Aviation Space and Journal of Applied Physiology 79, 205–213. (1989) Low oxygen selectively inhibits Environmental Medicine 60, 105–111. Muza, S.R., Rock, P.B., Fulco, C.S. et al. aldosterone secretion from bovine Richardson, D.W., Kontos, H.A., (2001) Women at altitude: ventilatory adrenocortical cells in vitro. American Raper, A.J. & Patterson, J.L., Jr. (1967) acclimatization at 4300 m. Journal of Journal of Physiology 256, E640–E644. Modification by β-adrenergic blockage Applied Physiology 91, 1791–1799. Rahn, H. & Otis, A.B. (1949) Man’s of the circulatory responses to acute Myhre, L.G., Dill, D.B., Hall, F.G. & respiratory response during and after hypoxia in man. Journal of Clinical Brown, D.K. (1970) Blood volume acclimatization to high altitude. Investigation 46, 77–85. changes during 3-week residence at American Journal of Physiology 157, Roach, R.C., Maes, D., Sandoval, D. high altitude. Clinical Chemistry 16, 7–14. 445–462. et al. (2000) Exercise exacerbates acute Okazaki, S., Tamura, Y., Hatano, T. Ramirez, G., Hammond, M., Agosti, S.J. mountain sickness at simulated high & Matsui, N. (1984) Hormonal et al. (1992) Effects of hypoxemia at sea altitude. Journal of Applied Physiology 88, disturbances of fluid–electrolyte level and high altitude on sodium 581–585. metabolism under altitude exposure in excretion and hormonal levels. Aviation Robach, P., Lafforgue, E., Olsen, N.V. et al. man. Aviation Space and Environmental Space and Environmental Medicine 63, (2002) Recovery of plasma volume after Medicine 55, 200–205. 891–898. 1 week of exposure at 4350 m. Pflugers Okumara, A., Fuse, H., Kawauchi, Y., Rastogi, G.K., Malhotra, M.S., Srivastava, Archiv 444, 821–828. Mizuno, I. & Akashi, T. (2003) Changes M.C. et al. (1977) Study of the pituitary– Roberts, A.C., Butterfield, G.E., in male reproductive function after high thyroid functions at high altitude in Cymerman, A. et al. (1996a) altitude mountaineering. High Altitude man. Journal of Clinical Endocrinology Acclimatization to 4300-m altitude Medicine and Biology 4, 349–353. and Metabolism 44, 447–452. decreases reliance on fat as a substrate. 464 chapter 30

Journal of Applied Physiology 81, man at high altitude. European Journal Acta Physiologica Scandinavia 173, 1762–1771. of Applied Physiology 62, 286–291. 409–417. Roberts, A.C., Reeves, J.T., Butterfield, G.E. Sawka, M.N., Young, A.J., Rock, P.B. et al. Share, L. & Levy, M.N. (1966) Effect of et al. (1996b) Altitude and β-blockade (1996) Altitude acclimatization and carotid chemoreceptor stimulation on augment glucose utilization during blood volume: effects of exogenous plasma antidiuretic hormone titer. submaximal exercise. Journal of Applied erythrocyte volume expansion. Journal American Journal of Physiology 210, Physiology 80, 605–615. of Applied Physiology 81, 636–642. 157–161. Rock, P.B., Johnson, T.S., Larsen, R.F. et al. Sawka, M.N., Convertino, V.A., Eichner, Singh, I., Khanna, P.K., Srivastava, M.C. (1989) Dexamethasone as prophylaxis E.R., Schneider, S.M. & Young, A.J. et al. (1969) Acute mountain sickness. for acute mountain sickness: effect of (2000) Blood volume: importance and New England Journal of Medicine 280, dose level. Chest 95, 568–573. adaptations to exercise training, 175–184. Rock, P.B., Kraemer, W.J., Fulco, C.S. et al. environmental stresses, and trauma/ Singh, I., Malhotra, M.S., Khanna, P.K. (1993) Effects of altitude acclimatization sickness. Medicine and Science in Sports et al. (1974) Changes in plasma cortisol, on fluid regulatory hormone response to and Exercise 32, 332–348. blood antidiuretic hormone and urinary submaximal exercise. Journal of Applied Schäfers, R.F., Nürnberger, J., Herrmann, catecholamines in high-altitude Physiology 75, 1208–1215. B. et al. (1999) Adrenoceptors mediating pulmonary oedema. International Journal Rock, P.B., Muza, S.R., Fulco, C.S. et al. the cardiovascular and metabolic effects of Biometeorology 18, 211–221. (2001) Women at altitude: effect of of α-methylnoradrenaline in humans. Smith, C.A., Dempsey, J.A. & Hornbein, menstrual-cycle phase on acute Journal of Pharmacology and Experimental T.F. (2001) Control of breathing at high mountain sickness during deployment Therapeutics 289, 918–925. altitude In: High Altitude. An Exploration to high altitude terrain. T01-18. Schmidt, M.E., Ledderhos, C. & Honig, A. of Human Adaptation (Hornbein, T.F. & USARIEM Technical Report. Natick, MA. (1985) Kidney function during arterial Schoene, R.B., eds.). Marcel Dekker, Inc., Rostrup, M. (1998) Catecholamines, chemoreceptor stimulation. I. Influence New York: 139–173. hypoxia and high altitude. Acta of unilateral renal nerve section, bilateral Stenberg, J., Ekblom, B. & Messin, R. (1966) Physiologica Scandinavia 162, 389–399. cervical vagotomy, constant artificial Hemodynamic response to work at Rowell, L.B., Johnson, D.G., Chase, P.B., ventilation, and carotid body simulated altitude, 4000 m. Journal of Comess, K.A. & Seals, D.R. (1989) chemoreceptor inactivation. Biomedica Applied Physiology 21, 1589–1594. Hypoxemia raises muscle sympathetic Biochimica Acta 44, 695–709. Stock, M.J., Chapman, C., Stirling, J.L. & activity but not norepinephrine in Schmidt, M.E., Goldstein, D.S., Schouten, Campbell, I.T. (1978) Effects of exercise, α resting humans. Journal of Applied J.L. et al. (1997) Acute 2 blockade altitude and food on blood hormone Physiology 66, 1736–1743. by idazoxan increases insulin and and metabolite levels. Journal of Applied Saltin, B., Grover, R.F., Blomqvist, C.G., lowers plasma glucose during Physiology 45, 350–354. Hartley, L.H. & Johnson, R.L. (1968) positron emission tomography. Surks, M.I., Beckwitt, H.J. & Chidsey, C.A. Maximal oxygen uptake and cardiac Psychopharmacology Bulletin 33, 253–259. (1967) Changes in plasma thyroxine output after 2 weeks at 4300 m. Journal Seals, D.R., Jone, P.P. & Davy, K.P. (2001) concentration and metabolism, of Applied Physiology 25, 400–409. Autonomic nervous system. In: High catecholamine excretion and basal Sanchez del Rio, M. & Moskowitz, M.A. Altitude. An Exploration of Human oxygen uptake during acute exposure (1999) High altitude headache: lessons Adaptation (Hornbein, T.F. & Schoene, to high altitude. Journal of Clinical from headaches at sea level. Advances in R.B., eds.). Marcel Dekker, Inc., New Endocrinology and Metabolism 27, 789–799. Experimental Medicine and Biology 474, York: 425–442. Sutton, J.R. (1977) Effect of acute hypoxia 145–153. Semenza, G.L., Nejfelt, M.K., Chi, S.M. on the hormonal response to exercise. Sanchez, C., Merino, C. & Figallo, M. (1970) & Antonarakis, S.E. (1991) Hypoxia- Journal of Applied Physiology 42, 587–592. Simultaneous measurement of plasma inducible nuclear factors bind to an Sutton, J.R., Young, J.D., Lazarus, L. et al. volume and cell mass in polycythemia enhancer element located 3′ to the (1970) Hormonal response to altitude. of high altitude. Journal of Applied human erythropoietin gene. Proceedings Lancet 2, 1194. Physiology 28, 775–778. of the National Academy of Sciences of the Sutton, J.R., Viol, G.W., Gray, G.W., Sawhney, R.C. & Malhotra, A.S. (1991) United States of America 88, 5680–5684. McFadden, M. & Keane, P.M. (1977) Thyroid function in sojourners and Semenza, G.L., Roth, P.H., Fang, H. & Renin, aldosterone, electrolyte, acclimatised lowlanders at high altitude Wang, G.L. (1994) Transcriptional and cortisol responses to hypoxic in man. Hormones and Metabolic Research regulation of genes encoding glycolytic decompression. Journal of Applied 23, 81–84. enzymes by hypoxia-inducible factor 1. Physiology 42, 421–424. Sawhney, R.C., Chhabra, P.C., Malhotra, Journal of Biological Chemistry 269, 23 Swenson, E.R. (2001) Renal function and A.S. et al. (1985) Hormone profiles at 757–23 763. fluid homeostasis. In: High Altitude. high altitude in man. Andrologia 17, Severinghaus, J.W., Mitchell, R.A., An Exploration of Human Adaptation 178–184. Richardson, B.W. & Singer, M.M. (1963) (Hornbein, T.F. & Schoene, R.B., eds.). Sawhney, R.C., Malhotra, A.S., Singh, T., Respiratory control at high altitude Marcel Dekker, Inc., New York: 525–568. Rai, R.M. & Sinha, K.C. (1986) Insulin suggesting active transport regulation Takase, K., Nishiyasu, T. & Asano, K. secretion at high altitude in man. of CSF pH. Journal of Applied Physiology (2002) Modulating effects of the International Journal of Biometeorology 30, 18, 1155–1166. menstrual cycle on cardiorespiratory 231–238. Sevre, K., Bendz, B., Hanko, E. et al. (2001) responses to exercise under acute Sawhney, R.C., Malhotra, A.S. & Singh, T. Reduced autonomic activity during hypobaric hypoxia. Japanese Journal of (1991) Glucoregulatory hormones in stepwise exposure to high altitude. Physiology 52, 553–560. hormones and altitude 465

Tunny, T.J., Van Gelder, J., Gordon, R.D. & Grover, R.F. (1968) The red cell Young, A.J. & Reeves, J.T. (2002) Human et al. (1989) Effects of altitude on atrial mass–arterial oxygen relationship adaptation to high terrestrial altitude. natriuretic peptide: the bicentennial in normal man. Journal of Clinical In: Medical Aspects of Harsh Environments Mount Everest expedition. Clinical and Investigation 47, 1627–1639. (Lounsbury, D.E., Bellamy, R.F. & Experimental Pharmacology and Physiology Weil, J.V., Byrne-Quinn, E., Sodal, I.E. Zajtchuk, R., eds.). Office of the Surgeon 16, 287–291. et al. (1970) Hypoxic ventilatory drive General, Borden Institute, Washington Vaernes, R.J., Owe, J.O. & Myking, O. in normal man. Journal of Clinical D.C.: 647–691. (1984) Central nervous reactions to a Investigation 49, 1061–1072. Young, A.J. & Young, P.M. (1988) Human 6.5-hour altitude exposure at 3048 m. Westendorp, R.G.J., Roos, A.N., Simons, acclimatization to high terrestrial Aviation Space and Environmental M. et al. (1993) Effects of hypoxia and altitude. In: Human Performance Medicine 55, 921–926. atrial natriuretic peptide on aldosterone Physiology and Environmental Medicine Vander, A.J., Moore, L.G., Brewer, G., secretion in healthy subjects. Journal of at Terrestrial Extremes (Pandolf, K.B., Menon, K.M. J. & England, B.G. (1978) Applied Physiology 75, 534–539. Sawka, M.N. & Gonzalez, R.R., eds.). Effects of high altitude on plasma White, P.C., Pescovitz, O.H. & Cutler Jr., Benchmark Press, Indianapolis: 497–543. concentrations of testosterone and G.B. (1995) Synthesis and metabolism Young, A.J., Evans, W.J., Cymerman, A. pituitary gonadotropins in man. of corticosteroids. In: Principles and et al. (1982) Sparing effect of chronic Aviation Space and Environmental Practice of Endocrinology and Metabolism high-altitude exposure on muscle Medicine 49, 356–357. (Becker, K.L., ed.). Williams & Wilkins, glycogen utilization. Journal of Applied Vogel, J.A. & Harris, C.W. (1967) Philadelphia: 647–661. Physiology 52, 857–862. Cardiopulmonary responses of resting Wolfel, E.E. & Levine, B.D. (2001) The Young, A.J., Young, P.M., McCullough, man during early exposure to high cardiovascular system at high altitude. R.E. et al. (1991) Effect of β-adrenergic altitude. Journal of Applied Physiology 22, Heart and systemic circulation. In: blockade on plasma lactate 1124–1128. High Altitude. An Exploration of Human concentration during exercise at high Vogel, J.A., Hansen, J.E. & Harris, C.W. Adaptation (Hornbein, T.F. & Schoene, altitude. European Journal of Applied (1967) Cardiovascular responses in man R.B., eds.). Marcel Dekker, Inc., New Physiology 63, 315–322. during exhaustive work at sea level York: 235–292. Young, A.J., Sawka, M.N., Muza, S.R. et al. and high altitude. Journal of Applied Wolfel, E.E., Groves, B.M., Brooks, G.A. (1996) Effects of erythrocyte infusion o Physiology 23, 531–539. et al. (1991) Oxygen transport during on V 2max at high altitude. Journal of Vuolteenaho, O., Koistinen, P., Martikkala, steady-state submaximal exercise in Applied Physiology 81, 252–259. V., Takala, T. & Leppaluoto, J. (1992) chronic hypoxia. Journal of Applied Young, P.M., Rock, P.B., Fulco, C.S. Effect of physical exercise in hypobaric Physiology 70, 1129–1136. et al. (1989a) Altitude acclimatization conditions on atrial natriuretic peptide Wolfel, E.E., Selland, M.A., Mazzeo, R.S. attenuates plasma ammonia secretion. American Journal of Physiology & Reeves, J.T. (1994) Systemic accumulation during submaximal 263, R647–R652. hypertension at 4300 m is related to exercise. Journal of Applied Physiology 63, Wagner, P.D., Gale, G.E., Moon, R.E. et al. sympathoadrenal activity. Journal of 758–764. (1986) Pulmonary gas exchange in Applied Physiology 76, 1643–1650. Young, P.M., Rose, M.S., Sutton, J.R. et al. humans exercising at sea level and Wolfel, E.E., Selland, M.E., Cymerman, A. (1989b) Operation Everest II: plasma

simulated altitude. Journal of Applied et al. (1998) O2 extraction maintains O2 lipid and hormonal responses during a Physiology 61, 260–270. uptake during submaximal exercise simulated ascent of Mt. Everest. Journal Walter, R., Maggiorini, M., Scherrer, U., with β-adrenergic blockade at 4300 m. of Applied Physiology 66, 1430–1435. Contesse, J. & Reinhart, W.H. (2001) Journal of Applied Physiology 85, 1092–1102. Young, P.M., Sutton, J.R., Green, H.J. et al. Effects of high-altitude exposure on Wright, A.D. (1979) Birmingham Medical (1992) Operation Everest II: metabolic vascular endothelial growth factor levels Research Expeditionary Society 1977 and hormonal responses to incremental in man. European Journal of Applied expedition: thyroid function and acute exercise to exhaustion. Journal of Applied Physiology 85, 113–117. mountain sickness. Postgraduate Medical Physiology 73, 2574–2579. Ward, M.P., Milledge, J.S. & West, J.B. Journal 55, 483–486. Zaccaria, M., Rocco, S., Noventa, D., (2000) Endocrine and renal systems at Xu, F. & Severinghaus, J.W. (1998) Rat Varnier, M. & Opocher, G. (1998) altitude. In: High Altitude Medicine and brain VEGF expression in alveolar Sodium regulating hormones at high Physiology. Chapman & Hall Medical, hypoxia: possible role in high-altitude altitude: basal and post-exercise levels. London: 178–201. cerebral edema. Journal of Applied Journal of Clinical Endocrinology and Weil, J.V., Jamieson, G., Brown, D.W. Physiology 85, 53–57. Metabolism 83, 570–574. Chapter 31

Neuroendocrine Influences on Temperature Regulation in Hot Environments

LAWRENCE E. ARMSTRONG AND JACK A. BOULANT

The nervous and endocrine systems act in unison acts rapidly (i.e. in 1–2 ms) over a small distance to to regulate homeostasis of body temperature and alter ion channel function and the electrical activity other physiological qualities that are essential to of post-synaptic cells. In some cases, these small- life. Because it is not possible to clearly divide the molecule transmitters cause acute nerve responses integrated efforts of the nervous and endocrine sys- or action potentials that relay afferent sensory tems (Galbo 1986), the term neuroendocrine refers to signals to the brain and efferent motor impulses the anatomical and functional interrelationships to muscles. In addition to norepinephrine (NE), sev- between nerves and endocrine organs. The majority eral other neurotransmitters serve to integrate the of neuroendocrine research has focused on blood- body’s responses to stress (DeSouza & Appel 1991; borne hormone and neuropeptide concentrations Toates 1995). Serotonin (5-hydroxytryptamine, in humans, and tissue levels of neurotransmitters 5-HT), for example, directly affects ACTH and in the central nervous system (CNS) of animals. prolactin secretion. The brain neurotransmitter Definitions of these and other relevant terms appear dopamine (i.e. a catecholamine precursor of NE) in the following four paragraphs. increases with exposure to various stressors, and A hormone is a signaling molecule that regulates also alters prolactin secretion. Acetylcholine (ACh) physiologic and metabolic functions by acting on excites the release of corticotropin-releasing hor- receptors located on or in target cells. Hormones mone from the hypothalamus (and ultimately can be stimulated by other hormones, nervous the release of the hormones ACTH and cortisol, in reflexes, or chemical transmitter substances. Some the hypothalamic–pituitary–adrenal [HPA] axis). are secreted by specific glands, transported by the Gamma-aminobutyric acid (GABA) inhibits the blood, and control responses in either specific target function of the HPA axis; glycine and glutamate, tissues (e.g. adrenocorticotrophic hormone, ACTH) two other amino acids, also act as neurotransmit- or in tissues located at multiple sites (i.e. epine- ters. Nitric oxide, which serves as a neurotransmit- phrine, growth hormone, thyroid hormone); these ter throughout the central and peripheral nervous are named endocrine secretions. Other signaling systems, is important in thermoregulation because molecules are released into the interstitial fluid and it results in blood vessel dilation. Interestingly, act on the receptors of adjacent cells; these are named nitric oxide regulates endocrine, autocrine and para- paracrine secretions. When the hormone is released crine functions. into the interstitial fluid and acts on the originating Some neuropeptides act slowly, over a relatively cell, it is named an autocrine secretion (Borer 2003). longer distance, to cause prolonged actions such as A neurotransmitter is a chemical mediator, syn- changes in the number of excitatory or inhibitory thesized in the cytosol of the presynaptic nerve receptors, long-term opening or closing of ion chan- terminal, that interacts with receptor molecules in nels, and activation-deactivation of specific genes in the post-synaptic membrane. A neurotransmitter the cell nucleus. The following neuropeptides have 466 neuroendocrine influences: thermoregulation 467

prompted considerable research interest among ganglionic sites. Postganglionic fibers arise from the neurophysiologists since the mid-1970s: ACTH, ganglion neurons, and these fibers then innervate prolactin, met-enkephalin, β-endorphin, human the effector organs. Postganglionic nerve endings growth hormone (hGH), peptide F, peptide Y, argi- of most sympathetic nerves secrete the neuro- nine vasopressin (AVP), luteinizing hormone, atrial transmitter NE (designated adrenergic), and most natriuretic peptide (ANP), vasoactive intestinal parasympathetic nerve endings secrete ACh (desig- peptide (VIP), somatostatin, oxytocin, galanin, α- nated cholinergic). An important exception exists: melanocyte-stimulating hormone, cholecystokinin innervation of cutaneous sweat glands, piloerector (CCK), insulin, glucagon, angiotensin II, brady- muscles and a few blood vessels is sympathetic kinin, calcitonin and thyrotropin-releasing hormone cholinergic. (Guyton & Hall 1996; Gibbins & Morris 2000). In a hot environment, activation of the sym- Many neuropeptides, released into the blood pathetic nervous system is evidenced by increased from the pituitary as hormones, are also released plasma NE. This represents an overflow of NE at within the CNS as neurotransmitters at much higher several peripheral sites, and often is used to repres- concentrations than in plasma. Thus, the same sub- ent the activity in the sympathetic–adrenomedullary stance may influence different target cells in differ- (SAM) axis (Zeisberger 1998). Sympathetic innerva- ent organs to a variable extent, depending on the tion of the adrenal medulla, as part of the SAM concentration of the neurotransmitter and the sensit- axis, provides an excellent example of the integrated ivity of the target cells. Receptor density, as altered effects of the nervous and endocrine systems. In by DNA-mediated protein synthesis, and receptor response to dysequilibria such as hyperthermia, subtype also may affect neuropeptide function reduced blood volume or blood pressure, and exer- (Zeisberger & Roth 1996). For example, at least three cise, relatively large quantities of epinephrine and 5-HT membrane receptors have been identified that NE are released by the adrenal medulla and circu- differ in binding properties and efferent signaling late to all tissues of the body. The effects of these pathways. hormones on organs are virtually the same as direct sympathetic nerve stimulation, but last 5–10 times Integrated neural and endocrine longer (i.e. 1–2 min after innervation) because these functions hormones are metabolized slowly (Guyton & Hall 1996). Therefore, organs may be stimulated simultan- The autonomic nervous system (ANS) regulates the eously in two ways: directly by sympathetic nerves visceral functions of the body such as body temper- and indirectly by adrenal medullary hormones. ature, sweating and arterial pressure. The ANS is Epinephrine and NE also stimulate organs that are activated by the hypothalamus, cerebral cortex, not innervated directly by sympathetic fibers; this is brain stem and spinal cord, without volitional con- significant because metabolic rate and heat produc- trol (Mosqueda-Garcia 1996). Efferent ANS signals tion can be increased in virtually all body cells. are transmitted through two subdivisions, named the sympathetic nervous system and parasym- Neurotransmitter roles in temperature pathetic nervous system, to glands, smooth muscle regulation: animal research or cardiac muscle. Sympathetic nerves originate in the spinal cord between the first thoracic and second Early models of thermoregulation that proposed lumbar vertebrae. About 75% of all parasympath- fixed, hard-wired neural circuitry have evolved etic fibers exit the brain via two vagus nerves, stimu- into complex paradigms that describe hypothalamic lating the entire thoracic and abdominal regions integration of different neurotransmitter systems (Hammel 1965). Both sympathetic and parasym- working concurrently (Zeisberger & Roth 1996). pathetic fibers make synaptic contact with neurons Despite numerous anatomical studies, relatively in autonomic ganglia located outside the spinal cord. little is known about the functional roles of these ACh is the most common neurotransmitter at these neurotransmitters because any given transmitter 468 chapter 31

may play either a stimulatory or an inhibitory role neurotransmitters (i.e. dopamine and NE) do not in temperature regulation, depending on its concen- cross the blood–brain barrier freely, histamine likely tration and the target organ involved. mediates these changes in core body temperature via peripheral nerves that affect heat loss (i.e. skin Peripheral effects. Body temperature is regulated via blood flow, sweating, or metabolic heat produc- the neurotransmitters ACh and NE, which act on tion). In contrast, peripherally administered sero- thermoregulatory effector organs (Zeisberger 1998) tonin (5-HT) evokes whole-body hyperthermia in the following ways. First, the tension of muscle (Feldberg & Myers 1964). Indeed, peripheral admin- contractions and exercise-induced heat production istration of several neuropeptides (i.e. angiotensin are controlled by the somatomotor system, via II, CCK, β-endorphin, α-melanocyte-stimulating hor- release of ACh at neuromuscular junctions. Sec- mone, neuropeptide Y, AVP, thyrotropin-releasing ond, sweat secretion is stimulated by sympathetic- hormone and substance P; see Clark & Fregly 1996) cholinergic nerve ganglia; ACh is involved at all increases or decreases heat storage. Until consen- such synapses. Third, skin blood flow changes sus is reached regarding the individual peripheral (i.e. changes in convective and radiative heat loss) effects of each, the pharmacological evidence will are primarily mediated by three neurotransmitters: remain unreconciled. NE at sympathetic nerve endings and the adrenal medulla, nitric oxide that is released locally and acts Central effects. The paramount role of the rostral as a potent vasodilator, or epinephrine that is hypothalamus, particularly the preoptic-anterior secreted by the adrenal medulla then transported hypothalamus (POAH), as the major temperature to cutaneous blood vessels. In addition to non- regulating center of the brain, has been established peptide neurotransmitters such as ACh and NE, in many animal species (Lomax & Green 1981; most vasoactive motor neurons contain coexist- Boulant 1996). The POAH receives and integrates ing neuropeptides. Immunohistochemical evidence nerve impulses from the skin and interprets the cir- indicates that the precise combination of these culating blood-borne compounds and endogenous neuropeptides, in peripheral neurons, varies with substances. Signals from the POAH are integrated the location of the vascular bed and species (Gibbins with impulses arising in the spinal cord, brain & Morris 2000). septum and CNS motor pathways. The following Although the complex interactions of neuro- sections describe several neurotransmitters and chemical transmitters in the control of body tem- endogenous substances that are thought to be perature have not yet been unequivocally defined, involved in both thermal afferent pathways and in experimental evidence indicates that peripheral sys- the POAH control of thermoregulation. temic injection of neurotransmitters and neuro- endocrines produce changes in body temperature. Glutamate. Glutamate is the primary excitatory The picture is complex, however, because these neurotransmitter in the brain and an important changes vary, not only with the animal species and neurotransmitter in thermoregulatory pathways. In environmental temperature, but also with the site, afferent pathways, glutamate is involved at spinal time and mode of injection (Clark & Fregly 1996). (Nishiyama et al. 2001) and supraspinal levels (Salt For example, histamine, administered peripher- & Turner 1998) in the integration and relay of ally into animals, affects core body temperature cutaneous thermal information. In the hypothala- (Brezenoff & Lomax 1970). The direction and mag- mus, microinjection studies indicate that glutamate nitude of these responses are a function of the ambi- injections have variable effects on body temper- ent temperature. A hypothermic response occurs ature (Clark 1979). Since glutamate excitation may in cool environments, whereas a hyperthermic res- play a role in both heat loss and heat produc- ponse occurs when environmental temperatures tion responses, these variable responses might be exceed skin temperature (Lomax & Green 1981). expected depending on the injection site, dose and Because circulating histamine and other central ambient temperature. An important microinjection neuroendocrine influences: thermoregulation 469

study was conducted by Kanosue et al. (1998), shivering thermogenesis by brown adipose tissue employing glutamate microinjections at various (Uno & Shibata 2001). POAH sites in anesthetized rats. These investigators showed that POAH glutamate had effects similar to Corticotropin-releasing factor (CRF). There is strong hypothalamic warming; i.e. eliciting tail vasodila- evidence that CRF release in the POAH is important tion in a thermoneutral environment and suppress- in fever and thermogenesis (Nakamori et al. 1993). ing shivering (Zhang et al. 1995) and non-shivering There are several distinct pathways for the induc- thermogenesis in a cold environment (Chen, X.-M. tion of fever. One well-documented fever pathway et al. 1998). involves interleukin-1α (IL-1α) and tumor necrosis Glutamate is also an important transmitter in factor-α (TNF-α) causing increases in a central efferent thermoregulatory pathways that descend mediator, prostaglandin E2, which alters the activity from the rostral hypothalamus. For rat tail blood of POAH neurons to induce fever. CRF appears flow, POAH projections descend to the midbrain to be a mediator in a separate fever pathway that where they release glutamate as a neurotransmitter. is associated with inflammation, stress and neural In the midbrain, glutamate excitation in one area trauma (Tache et al. 2001). Different cytokines (i.e. periaqueductal grey) elicits tail vasodilation, (IL-1β, IL-6, IL-8) and a different prostaglandin while glutamate in another area (i.e. vental tegmen- mediator (PGF2α) serve in this separate pathway, tal area) elicits tail vasoconstriction (Zhang et al. and it is proposed that these substances increase the 1997). Therefore, it is likely that there are two hypothalamic release of CRF to alter the activity of subpopulations of preoptic-anterior hypothalamic thermoregulatory neurons (Nakamori et al. 1993). neurons having descending axons controlling tail blood flow: one population controlling vasocon- Serotonin. Early studies found that (either induced striction and the other controlling vasodilation. or naturally-occurring) changes in hypothalamic Finally, there is also evidence that another popula- serotonin (5-HT) are associated with body temper- tion of POAH neurons send glutamate fibers to ature changes (Clark & Lipton 1986). While opposite the caudal midbrain and medulla to control non- responses occurred (depending on dose, site, spe- shivering thermogenesis by brown adipose tissue. cies, ambient temperature) many early rat studies indicated that POAH serotonin caused thermogen- GABA. Gamma-aminobutyric acid is the primary esis, heat retention, and increased body temperature. inhibitory neurotransmitter in the brain. Most In support of this, a recent microdialysis study of studies indicate that POAH microinjections of various neurotransmitters in the rat POAH found GABA-like drugs produce hypothermia (Minano a strong correlation between serotonin metabolites et al. 1989). Conversely, increases in body temper- and the rise in body temperature (Yasumatsu et al. ature occur with administration of antagonists of 1998). In some studies, hypothalamic 5-HT has two different types of GABA receptors; i.e. either been linked with either temperature increases or

GABAA (Horton, R.W. et al. 1988) or GABAB decreases, possibly depending on the type of 5-HT (Jackson & Nutt 1991). Studies also suggest that receptor stimulated. For example, pharmacological there are central interactions between GABAA studies link activation of 5-HT1A receptors with receptors and body temperature changes associated hypothermia (Malone & Taylor 2001), while 5-HT2A with prostaglandins and opioids (Eguchi et al. 1999). receptor activation has been linked with hyperther- In addition, GABA is an important transmitter mia (Sugimoto et al. 2000). in the efferent thermoregulatory pathways that descend from the rostral hypothalamus. Studies Neuroimmune system. The traditional view of immune indicate that GABA is the inhibitory transmitter in function has been replaced by a new paradigm that axons of some POAH neurons that project to the recognizes that immune function is co-ordinated midbrain to control tail blood flow (Kanosue 1998) with CNS function by the hypothalamus. This and to other brain stem areas controlling non- paradigm relies on the following facts: (a) immune 470 chapter 31

cells and neurons share membrane receptors for a Stressors variety of neuropeptides, corticosteroids and cyto- kines (i.e. immune system messenger proteins that amplify immune function and modify physiological Hypothalamus responses); (b) the organs of the immune system are innervated by sympathetic and parasympathetic CRH nerves, under hypothalamic control; (c) production ACTH of phagocytic cells and lymphocytes (i.e. B and T) Sympathetic Anterior pituitary are regulated by neuropeptides and corticosteroids spinal nerves emanating from the HPA axis; and (d) circulating hGH, PRL, βE cytokines bind to several brain regions, including Adrenal NE, E the neural pituitary gland, and stimulate these glands regions to produce releasing hormones. Such inter- NE, Y, P connections explain why infection and other immu- Immune cells nological stimuli influence thermoregulation (fever), Cortisol nutrition (anorexia) and psychological state, col- lectively termed ‘sickness behavior’ (Zeisberger & Roth 1996). Spleen, thymus and Cytokines other immune organs (IL-1, IL-6, TNF-α) The cytokines, a class of chemical messengers, are released into the blood in response to tissue injury, infection, and other hormones (e.g. epinephrine, Fig. 31.1 Brain–immune system interactions. Cytokines β are chemical messengers, produced by a variety of cells NE, hGH, prolactin, -endorphin). These messen- that complete a bi-directional communication network gers are critical to survival and include the sub- between the periphery and the brain. ACTH, corticotropin categories interleukins (IL-1α, IL-1β, IL-6), tumor (adrenocorticotropic hormone); βE, β-endorphin; CRH, necrosis growth factors (TNF-α), chemokines and corticotrophin-releasing hormone; E, epinephrine; hGH, interferons (IFN-α, IFN-β). Cytokines are released human growth hormone; IL-1, interleukin-1; IL-6, interleukin-6; NE, norepinephrine; P, substance P; by a variety of cells, including immune, endothelial PRL, prolactin; TNF-α, tumor necrosis factor-α; Y, and fat-storing cells (Smith 2000). The most com- neuropeptide Y. (Reprinted from Armstrong & VanHeest monly described actions of cytokines are: (a) their 2002 with permission.) release from immune cells; and (b) their suppres- sion of the early phase of inflammation, via feed- back to the brain (Armstrong & VanHeest 2002). As body temperature upward (Blatteis 1998). The illustrated in Fig. 31.1, they act in a way that re- resulting increase in body temperature can occur by sembles a diffuse sensory organ, providing the various means, including increasing heat produc- hypothalamus with information about a variety of tion responses (e.g. shivering and non-shivering processes occurring in the periphery of the body thermogenesis) and heat retention responses (e.g. (Maier & Watkins 1998; Smith 2000). In healthy hu- cutaneous vasoconstriction), as well as decreasing mans, cytokines have relatively little impact on body heat loss responses (e.g. sweating). It is likely that temperature regulation, but during illness they in- this is accomplished by inhibiting or enhancing the fluence heat balance in two ways. First, they increase activity of specific types of hypothalamic neurons, metabolic rate and heat production, as described in as described in the following section. the Human heat balance: production versus loss section below. Second, in response to microbial Influences on hypothalamic neurons. As mentioned (i.e. viruses, bacteria, mycobacteria, fungi) and non- above, the POAH plays an important role in all microbial agents, cytokines (i.e. IL-1β, IL-6 and physiological and behavioral responses that main- TNF-α) act as pyrogens, by altering the effector tain body temperature constant. Some of the neurons responses of the POAH and shifting the equilibrium in the POAH are thermosensitive and, therefore, neuroendocrine influences: thermoregulation 471

sense changes in body core temperature. These same retention. Neurons controlling these responses are neurons synaptically control other neurons that thought to be excited by temperature insensitive elicit heat loss responses (e.g. sweating), heat reten- neurons but synaptically inhibited by the warm tion responses (e.g. cutaneous vasoconstriction) and sensitive neurons. These neurons would appear to heat production responses (e.g. shivering and non- be ‘cold sensitive’ because hypothalamic cooling shivering thermogenesis). Exercise, for example, would decrease the firing rates of warm sensitive causes core temperature (and hypothalamic tem- neurons which, in turn, would decrease their inhibi- perature) to rise, and this is sensed by hypotha- tion on the innervated effector neuron. As a result, a lamic thermodetectors which initiate appropriate heat production effector neuron would increase its responses, such as sweating and an increase in skin firing rate during hypothalamic cooling. While only blood flow. a small proportion of POAH neurons are considered While most POAH neurons are relatively insens- to be cold sensitive, POAH warm sensitive neurons itive to temperature, about one-quarter of the spon- could synaptically inhibit other effector neurons taneously firing neurons are considered to be warm located in various brain stem locations. sensitive. Spontaneously firing neurons produce The neuroendocrines and neurotransmitters action potentials and have firing rates determined mentioned in the above sections generally have by the number of action potentials per second. predictable effects on the different types of POAH Warm sensitive neurons significantly increase their neurons. Neurochemicals that produce hypother- firing rates when the hypothalamus is warmed mia often have excitatory effects on POAH warm above 37°C, and these neurons also tend to decrease sensitive neurons. Excitation of warm sensitive neu- their firing rates when the hypothalamus is cooled rons increases heat loss responses and, thus, lowers below 37°C. Thus, warm sensitive neurons sense body temperature. In contrast, neurochemicals that both increases and decreases in body temperature. produce hyperthermia (such as fever-producing POAH warm sensitive neurons also receive synap- pyrogens) tend to inhibit warm sensitive neurons tic inputs from ascending somatosensory path- and excite cold sensitive neurons. This produces ways conveying information from thermoreceptors decreased sweating, decreased skin blood flow, in the skin and other body locations (e.g. spinal but increased shivering and non-shivering thermo- cord) (Boulant & Hardy 1974). For example, some genesis; and all of these responses increase body POAH warm sensitive neurons increase their firing temperature. rates during increases in either hypothalamic tem- In addition to regulating body temperature, the perature or skin temperature. In this way, POAH POAH also is important in reproduction. Some neurons can integrate central and peripheral POAH neurons contain and are sensitive to re- temperatures. productive neuroendocrines such as testosterone, POAH neurons form synaptic networks that con- estrogen or progesterone. Moreover, there are over- trol thermoregulatory responses. Various neuronal laps and strong interactions between temperature models have been postulated to explain these net- sensitive neurons and neurons that are sensitive works (Hammel 1965; Boulant 1996). For the neu- to reproductive endocrines. In a study of rat POAH rons controlling heat loss responses like sweating, tissue slices, about 30% of warm sensitive neurons most models suggest that these effector neurons were found to be excited by either testosterone or are synaptically excited by warm sensitive neurons estrogen. If warm sensitive neurons enhance heat and possibly inhibited by temperature insensit- loss responses, this may explain why animal studies ive neurons. Accordingly, hypothalamic warming sometimes link testosterone or estrogen administra- would increase the firing rate of warm sensitive tion to lowered body temperatures (Silva & Boulant neurons which, in turn, would synaptically increase 1986). The menstrual cycle of women may also illus- the firing rate of nearby neurons controlling sweat- trate the interactions between thermosensitive ing. A different synaptic arrangement is proposed and endocrine sensitive POAH neurons. Midway for neurons controlling heat production and heat through the menstrual cycle, just before ovulation, 472 chapter 31

some women show a brief drop in body temper- Therefore, the preoptic area of the anterior hypo- ature that is associated with peak estrogen levels thalamus senses rising blood and organ temperat- (Stephenson & Kolka 1999). Following this, there ures, and initiates sweat gland secretion concurrent often is a rapid rise in body temperature that stays with cutaneous blood vessel dilation. Elsewhere elevated for much of the later half of the cycle; in the CNS, the heated spinal cord also influences and this later period is associated with elevated pro- temperature regulation by inducing complimentary gesterone levels (Kolka & Stephenson 1997). As effector responses in skin sweat glands and blood mentioned for the rat tissue slice study above, vessels (Jessen 1996). Although prolonged strenu- some POAH warm sensitive neurons are excited by ous exercise in a hot environment typically results estrogen. If warm sensitive neurons facilitate heat in sweat losses of 0.8–2.0 L·h–1, the aforementioned loss and suppress heat production, this may explain marathon runner produced 3.71 L·h–1 of sweat the transient, mid-cycle drop in body temperature and incurred dehydration that was detrimental to during the estrogen peak. Moreover, similar neu- his running performance and health. The body ronal experiments (Boulant, unpublished) have dis- dissipates 2428 kJ (580 kcal) for each liter of sweat covered some POAH warm sensitive neurons that that is evaporated from the skin. Because 85–90% are inhibited by progesterone, which may explain of all heat dissipation occurs via sweat evaporation the sustained elevation in body temperature during in a hot-dry environment (Adams et al. 1975), and the last half of the cycle. because over 50% of heat loss in a hot-humid envir- onment may occur via radiation and convection Human heat balance: production from dilated skin blood vessels, the autonomic and versus loss neuroendocrine control of metabolism, sweating and blood flow are vital to thermoregulation, health To maintain internal body temperature at about and exercise performance. 37°C during waking and sleeping hours, humans During strenuous exercise, neuroendocrine constantly adapt to changes in air temperature, responses within the ANS support temperature humidity, air movement, solar radiation, baro- regulation, ventilation, cardiovascular responses, metric pressure and clothing insulation. Further, fluid-electrolyte balance and immune functions. as food is digested and metabolized, approxim- These widespread effects involve the SAM axis, ately 80% of all nutrient energy eventually is trans- the HPA axis, endogenous opioid messengers (i.e. formed to heat. If strenuous muscular exercise is endorphins, dynorphins, enkephalins) and fluid- undertaken, the rate of energy utilization can electrolyte regulating hormones (Borer 2003). During increase well above the resting metabolic rate. For exercise and rest, the ANS affects temperature regu- example, an elite marathon runner (body mass, lation primarily via sudomotor (i.e. in sweat glands) 66.9 kg; height 185 cm) required 5862 kJ (1400 kcal) and vasomotor (i.e. in blood vessels) responses. energy·h–1 (97.6 kJ·min−1 [23.3 kcal·min–1]; running The sudomotor response facilitates heat dissipation speed, 314 m·min–1), continuously for 2.2 h, to com- through sweat production and evaporation. An plete the 1984 Summer Olympics in Los Angeles increase in temperature is detected by central and (Armstrong et al. 1986); this was approximately peripheral thermoreceptors, and the eccrine sweat 8–12 times greater than his energy utilization at rest glands are activated via sympathetic cholinergic (i.e. 8.3–12.6 kJ·min–1 [2–3 kcal·min–1]). Such a heat neurons. These same thermoreceptors also affect load must be removed or serious hyperthermia (i.e. vasomotor responses (i.e. cutaneous vasoconstric- body temperature above 39–40°C), tissue damage tion or vasodilation) controlling skin blood flow, and death may result from exertional heat-stroke. which determines radiative and convective heat Indeed, this athlete experienced this malady twice loss to the environment. during his running career. Exclusive of exercise, the term thermogenesis refers Environmental–heat stress increases the need to adaptive or regulatory increases in metabolic rate for evaporative and radiative-convective cooling. that are associated with excess food consumption neuroendocrine influences: thermoregulation 473

(i.e. diet-induced thermogenesis); cold exposure, and 30 min (80°C), respectively (Hannuksela & cold adaptation and arousal from hibernation Samer 2001). After the sauna, metabolic rate re- (i.e. non-shivering thermogenesis); or responses to turns to baseline at about the same rate as body disease, injury and stress (often referred to as hyper- temperature. metabolism). While the preoptic anterior hypotha- In response to this rise in core body temperature, lamus exerts strong control over thermogenesis, hypothalamic efferent signals cause visible sweat several other neural structures can influence heat production within 8–12 min. The mean sweat rate production. These other structures may include the, ranges from 0.6–1.0 L·h–1 at 80–90°C, resulting in an ventromedial hypothalamus, posterior hypotha- average total loss of 300–500 mL during a typical lamus, paraventricular nucleus, cerebral cortex, sauna bath. Despite this sweat secretion, skin tem- hippocampus and brain stem structures such as the perature rises from 35–36°C (pre-exposure) to 40°C raphe nucleus and locus coeruleus (Rothwell 1994). within a few minutes (Hannuksela & Samer 2001). Furthermore, animal research has verified that Cardiovascular responses include a large increase several peptides (i.e. CRF, thyrotrophin-releasing in skin blood flow, from 0.5 to 7.0 L·min–1 (i.e. from factor [TRF], insulin, β-endorphin, angiotensin, soma- 5–10% to 50–70% of total resting cardiac out- tostatin, γ-melanocyte-stimulating hormone, CCK) put), which is accomplished by a decrease of peri- and cytokines (i.e. IL-1α, IL-1β, IL-6, IL-8, IFN-γ, pheral autonomic nerve activity (i.e. withdrawal of TNF-α) increase metabolic rate. sympathetic vasoconstriction), causing a decrease Acknowledging that the peripheral and central in total peripheral resistance (Zeisberger 1998). effects of specific neurotransmitters and neuro- This vasodilation in skin involves local production peptides remain largely unknown, the remainder of of the neurotransmitter nitric oxide (Kellogg et al. this chapter focuses on neuroendocrine responses 1999). But, because air temperature is much greater and adaptations to whole-body hyperthermia (i.e. than skin temperature, radiative and convective core body temperature > 38°C) in humans, at rest heat is gained through skin, not lost. Simultan- and during exercise in a hot environment. eously, the increase of sympathetic activity in thor- acic organs (i.e. increased vasoconstrictor tone) –1 Passive, intense heat exposure reduces splanchnic blood flow by 0.6 L·min . As in humans a result, most studies consistently report a decline in diastolic blood pressure (6–39 mmHg), but vari- The hot-dry Finnish sauna bath is taken in a paneled able pressure responses at systole. Because cardiac room that contains wooden benches and a rock-filled stroke volume remains constant, the increased heater. The air temperature ranges from 80–100°C cardiac output at rest is solely accomplished by at face level and 30°C at floor level, with a relative increased frequency of contraction. Mean resting humidity of 10–20%. A single exposure lasts 5– heart rate increases to 100 b·min–1 in experienced 20 min, but this exposure may be extended or bathers, and as high as 150 b·min–1 in some unac- repeated by experienced bathers. Human sauna customed adults who experience intense heat bathing provides an interesting model to study stress (Hannuksela & Samer 2001). Dehydration, the body’s neuroendocrine responses to extreme due to sweat losses, further reduces central blood hot-dry environments. volume. To maintain cardiac output, vasomotor reflexes decrease blood flow to splanchnic organs Heat production and dissipation. Mean resting meta- and increase heart rate. In addition to the release bolic rate increases 20%, ranging to 40% above the of epinephrine and NE, cardiovascular stability is pre-exposure level (Hasan et al. 1966). This increased enhanced by increased secretion of renin, angio- heat production, coupled with an environment that tensin II, aldosterone, AVP and cortisol (Hannuksela stifles radiative and convective heat loss, causes & Samer 2001). All of these hormones act to in- adult rectal temperature to rise 0.2, 0.4 and 1.0°C crease blood pressure and/or heart rate (Borer during exposures of 15 min (72°C), 20 min (92°C) 2003). 474 chapter 31

Table 31.1 Acute human thermoregulatory and stress hormone responses to severe heat exposure (80–100°C, 7–15%rh) in a sauna bath.

Exposure Human subjects Change of duration blood Hormone (min) Male Female concentration References

ACTH††† 30 7 Increased Jezova et al. 1985 20 11 Unchanged Laatikainen et al. 1988 30 8 Increased Vescovi et al. 1990 30 8 Increased Vescovi et al. 1992 20 8 Increased Jezova et al. 1994 20 8 Unchanged Jezova et al. 1994 β-Endorphin†‡ 30 7 Increased Jezova et al. 1985 20 11 Increased Laatikainen et al. 1988 30 8 Increased Vescovi et al. 1992 30 8 Increased Vescovi et al. 1990 Growth hormone§ 20 6 Increased Aldercreutz et al. 1976 20–40 37 18 Increased Lammintausta et al. 1976 30 10 7 Increased Leppaluoto et al. 1986 30 8 Increased Vescovi et al. 1992 Prolactin§¶ 20 6 Increased Aldercreutz et al. 1976 30 10 7 Increased Leppaluoto et al. 1986 20 11 Increased Laatikainen et al. 1988 30 8 Increased Vescovi et al. 1992 Cortisol†† 20 6 Increased Aldercreutz et al. 1976 30 7 Unchanged Jezova et al. 1985 20 11 Unchanged Laatikainen et al. 1988 Epinephrine*‡§** 10 5 Increased Hussi et al. 1977 20 11 Unchanged Laatikainen et al. 1988 20 8 Increased Jezova et al. 1994 20 8 Increased Jezova et al. 1994 Norepinephrine*‡§** 10 5 Increased Hussi et al. 1977 20 11 Increased Laatikainen et al. 1988 20 8 Increased Jezova et al. 1994 20 8 Increased Jezova et al. 1994

‡‡ Triiodothyronine (T3)** 30 10 7 Unchanged Leppaluoto et al. 1986 30 10 Increased & unchanged Strbak et al. 1987

ACTH, adrenocorticotropic hormone. *Responds to stressors as part of the hypothalamic–pituitary–adrenocortical (HPA) axis. †ACTH (adrenocorticotropic hormone) and β-endorphin are anterior pituitary products of a common precursor (proopiomelanocortin) (Vescovi et al. 1990). ‡Mediates skin blood flow (i.e. radiation and convective heat dissipation). §Stimulates secretion of sweat in eccrine glands (i.e. evaporative heat dissipation). ¶Prolactin may alter Na+–K+ and water transport in eccrine sweat gland membranes. **Increases metabolic rate and heat production. ††Responds to stressors as part of the sympathetic–adrenomedullary (SAM) axis. ‡‡The active form of thyroid hormone. neuroendocrine influences: thermoregulation 475

Peripheral hormone responses. Because present tech- 160 nology does not allow direct measurement of )

human brain neurotransmitters, and because these –1 neurotransmitters do not cross the blood–brain barrier freely, Table 31.1 describes changes of blood- 100 borne hormones (all of which are neurotransmitters or neuropeptides except cortisol and triiodothyro- ACTH (pg·mL nine, T3), following sauna exposure. These hor- mones are secreted by the anterior pituitary (ACTH, 50 β-endorphin, hGH, prolactin), adrenal cortex (corti- 21 )

sol), adrenal medulla (epinephrine, NE) and thy- –1 roid gland (T3) in response to a variety of stressors 17 (i.e. hyperthermia, intense exercise, dehydration, hypoxia, hypoglycemia, hypovolemia and psycho- 13 logical stress; Borer 2003). The stress hormones in

Table 31.1 indicate that a portion of the hormone Cortisol (ng·mL 9 response to sauna bathing (80–100°C) may result SAUNA from SAM axis or HPA axis responses to psycholo- 0153045 gical or environmental stresses. Time (min) Three factors complicate the interpretation of Fig. 31.2 Mean changes of plasma cortisol and Table 31.1. First, although all blood samples were adrenocorticotropic hormone (ACTH) (in seven healthy collected at rest and all of these hormones increase men) in response to intense heat exposure in a sauna for when animals are experimentally heated (Borer 30 min. X, p < 0.05; XX, P < 0.01. (Redrawn from Jezova 2003), it is impossible to attribute changes in human et al. 1985 with permission.) hormones to heat storage alone, unless central hyperthermia was verified via measurements of core body temperature. Unfortunately, only one of concentration of ACTH had occurred. Third, re- these hormone studies (Jezova et al. 1994) measured peated exposures to intense environmental heat and core body temperature validly (i.e. rectal, eso- repeated bouts of whole-body hyperthermia alter phageal or tympanic membrane temperature), and the secretion patterns of some hormones. Thus, the none included simultaneous measurements of sym- neuroendocrine responses in Table 31.1 may not be pathetic nervous system effector responses such as observed among experienced bathers who have sweat rate, cutaneous blood flow, or metabolic heat taken saunas throughout their lives (Kukkonen- production. Indeed, the variant findings of both Harjula et al. 1989). unchanged and increased hormone concentrations In light of these considerations, we believe that (column 5 in Table 31.1) may be due to different the following statements provide a reasonable sum- degrees of body temperature elevation. Second, mary of Table 31.1, regarding changes in circulating hormones (column 5, Table 31.1) may increase or hormones after resting sauna exposure: decrease at different rates, in response to hyperther- (a) hGH, β-endorphin, prolactin and NE were elev- mia, depending on the time that elapsed between ated in blood samples, following sauna exposure, in heat exposure and blood sample collections (Jezova all investigations. et al. 1985, 1994). Figure 31.2 illustrates this phe- Notes: nomenon, in that peak plasma concentrations of • Research evidence indicates that β-endorphin ACTH and cortisol were observed at different time may mediate blood flow (Navaratnam et al. 1992), points after sauna exposure. This likely was due to lowers the hypothalamic set point temperature continued release of cortisol by the adrenal glands, and reduces metabolic heat production (Kraemer, well after the peak pituitary secretion and plasma W.J. et al. 2003). 476 chapter 31

• Secretion from eccrine sweat glands is peri- on the plasma concentrations of some hormones pherally mediated by the neurotransmitter ACh (i.e. NE, epinephrine, cortisol) but not others (i.e. in sympathetic cholinergic nerve fibers, circulat- dopamine, growth hormone) (Brenner et al. 1997); ing NE, plus anterior pituitary hormones hGH (b) plasma prolactin concentration increases more (Boisvert et al. 1993) and prolactin (Follenius et al. during passive heating than during exercise 1979; Kaufman et al. 1988). (Kukkonen-Harjula 1989); and (c) plasma cortisol • An increased plasma NE concentration repres- may increase or decrease during low intensity exer- ents increased sympathetic and adrenal medul- cise, depending on its circadian rhythm (Thuma lary activity that maintains thermoregulatory et al. 1995). effector responses (Zeisberger 1998). During exercise–heat stress, numerous hormone, (b) Circulating ACTH, cortisol and epinephrine neuropeptide and neurotransmitter interactions may or may not increase following exposure to complicate a thorough description of human ther- heat, depending on the duration/intensity of expos- moregulation. Growth hormone is a prime example. ure and the degree of whole-body hyperthermia For example, the release of growth hormone is experienced. mediated by neuropeptides (i.e. galanin, thyroid- Notes: releasing hormone, neuropeptide Y, calcitonin, • ACTH and cortisol represent a neuroendocrine β-endorphin), neurotransmitters (i.e. ACh, NE, response to stress as part of the HPA axis. dopamine, 5-HT and GABA), metabolic substrates • Circulating epinephrine potentiates sweat secre- (i.e. blood glucose, l-arginine) and hormones (i.e. tion, but plays a secondary role to direct innerva- gonadal testosterone, estrogens, somatostatin and tion of eccrine sweat glands by sympathetic thyroid hormones) (Giustina & Veldhuis 1998). cholinergic nerve fibers (Zeisberger 1998). Several isoforms of growth hormone have been Summary statements (a) and (b) agree with previ- identified, but their distinct physiological roles ously published observations that NE (Laatikainen have not been clarified (Baumann 1999). Both tem- et al. 1988), prolactin (Aldercreutz et al. 1976; perature and exercise (i.e. exercise intensity, inter- Mills & Robertshaw 1981; Leppaluoto et al. 1986; mittency, duration and length of rest periods) are Laatikainen et al. 1988) and hGH (Leppaluoto et al. recognized as stimuli for growth hormone release. 1986) are sensitive markers of physiological strain in Thus, it is likely that multiple neurotransmitter the severe heat of a sauna (80–100°C, 7–15%rh). pathways, as well as a variety of peripheral feed- back signals, regulate growth hormone secretion Exercise–heat stress in humans either by acting directly on the anterior pituitary gland and/or modulating the hypothalamic release Neuroendocrine responses during exercise in a hot of growth-hormone releasing hormone or somato- environment (defined here as ≥ 35°C) are distinctly statin. This is relevant to body temperature regula- different from those that occur while resting in a tion because a deficiency of growth hormone reduces sauna bath. This is true because there are multiple sweat secretion from cutaneous eccrine glands stimuli for neurotransmitter or hormone release (Lobie et al. 1990; Borer 2003), decreases evaporative and, unlike rest, exercise alters the activity of several heat loss and increases the risk of serious hyper- neurotransmitter pathways concurrently or in suc- thermia during exercise (Hjortskov et al. 1995). cession (Giustina & Veldhuis 1998). Although most Figure 31.3 depicts the complex interaction of of these perturbations can be attributed to increased additional factors that may influence the neuro- sympathetic nervous tone (i.e. increased ventilation, endocrine regulation of human body temperature oxygen consumption, heart rate) or increased activ- during exercise–heat stress. The heart of this figure ity in the HPA and/or SAM axes, the endocrine (see shaded rectangle) involves neuroendocrine responses to concurrent heat and exercise stress are release, clearance and membrane receptors, trans- not always straightforward. For example: (a) exer- porter protein binding and release, as well as cise and thermal stressors have a synergistic effect sympathetic effector responses, that enhance heat neuroendocrine influences: thermoregulation 477

Exercise protocol Dietary and hydration state • Intensity (high versus low) • Diet (eucaloric versus hypocaloric) • Duration (long versus brief) • Hypoglycemia • Mode (cycling, running, resistance) • Hypohydration or dehydration • Design (continuous versus intermittent) • Osmolality of extracellular fluid

Neuroendocrine factors in body temperature regulation • Neurotransmitter/hormone release and interactions • Neurotransmitter/hormone metabolism/clearance • Transporter protein binding and release • Membrane receptor subtype, density and saturation • Sympathetic effector responses stimulated by hyperthermia and exercise: • eccrine sweat gland secretion • increased blood flow to cutaneous capillary beds and active skeletal muscles; reduced blood flow to splanchnic organs

Environmental factors Hereditary and acquired characteristics • Clothing microenvironment • Menstrual cycle, reproductive • Ambient conditions hormones • Temperature • Exercise-induced heat acclimatization • Relative humidity • Physical training • Wind speed • Exertional heat illness • Solar radiation

Fig. 31.3 Interactions of numerous factors that may influence the neuroendocrine regulation of human body temperature during exercise–heat stress. dissipation via sweating and cutaneous vasodila- Castellani et al. (1997) also identified NE as a more tion. These components of neuroendocrine function sensitive neuroendocrine marker, versus epine- are influenced by other stimuli (see unshaded phrine, of physiologic strain in the heat (33°C in rectangles) including features of the exercise pro- both; 150, 90-min cycling exercise; and 65, 50% o tocol, diet/metabolism/fluid balance, hereditary/ V 2max, respectively). Similarly, Nybo et al. (2002) acquired characteristics and environmental factors. reported that NE responded markedly, whereas Although today it is virtually impossible to explain epinephrine did not, when young men wore insu- this multifactorial problem simply, the following lated clothing (i.e. establishing uncompensable section describes our present understanding of heat stress) and cycled in a 20°C environment o these factors, as they influence neuroendocrine res- (60 min, 50% V 2max). However, during an intense o ponses to exercise–heat stress. (80% V 2max) 16.1-km outdoor run in the heat (30°C), both plasma epinephrine and NE concentra- Exercise protocol. The research of Hoffman et al. tions increased (Hartung et al. 1987), suggesting that (1994) advanced NE as the preferred neuroendo- either exercise intensity or posture (upright versus crine index of human exercise–heat strain. This seated) may alter catecholamine responses, thereby arose from the fact that neither epinephrine, testo- affecting thermoregulation and body heat storage. sterone nor cortisol increased in response to low- This raises the question, ‘What influences do the intensity treadmill exercise at 33°C, regardless of components of an exercise protocol have on human baseline hydration state or ad libitum water intake neuroendocrine and thermoregulatory responses?’ during exercise. Mora-Rodriguez et al. (1996) and Borer (2003) and other authorities state that 478 chapter 31

pituitary secretion of ACTH and β-endorphin, much additional study, especially with environ- adrenocortical release of cortisol and adrenome- mental temperatures above 35°C. dullay secretion of the stress hormones epinephrine and NE increase above exercise intensities of Diet and hydration state. Because the release of o 80–90% V 2max, in a mild environment. However, hormones and neurotransmitters may stimulate the reality may be more complex than this statement mobilization or storage of biochemical substrates, suggests. Borer’s concept evidently does not include dietary adequacy or insufficiency may alter neuro- investigations that reported: (a) no change in endocrine function. In this regard, research has ACTH, β-endorphin and cortisol at three supramax- focused on two independent variables: energy avail- o imal exercise intensities (175, 230, 318 V 2max; 6–46s ability and body hydration state. First, decreased duration; seated cycling; 23°C) but an increase in blood glucose concentrations (< 3.3 mmol·L–1), dur- o these hormones at 115 V 2max (Kraemer, W.J. et al. ing prolonged exercise, result in an increased 1989); (b) constant plasma β-endorphin levels dur- activation of the HPA axis in humans, as evidenced o ing the initial hour of cycling exercise (50% V 2max; by increased blood ACTH and cortisol concentra- 35°C), that increased during the final 75–120 min tions. Representative of increased physiological (Kelso et al. 1984); (c) a decrease in cortisol concen- strain, this phenomenon can be reversed by main- tration during five 15-s maximal anaerobic power taining blood glucose at a constant level (Tabata tests, separated by 30 s of active recovery (22 and et al. 1991). 35°C) (Hoffman et al. 1997); and (d) an increase in Second, a loss of body water also affects human circulating epinephrine, cortisol and NE concentra- thermoregulatory responses. For example, during o tions during 120 min of cycling at 40% V 2max (22°C), low-intensity walking (140 min total; four 25-min o with a plateau in NE from 60–120 min (Horton, T.J. exercise bouts; 28–30% V 2max) in a 49°C (20%rh) et al. 1998). Two of these research teams concluded environment, test subjects exhibited increased that hormonal responses were not duration- or plasma cortisol and growth hormone concentra- intensity-dependent (Kraemer, W.J. et al. 1989), tions (versus euhydration) when they were dehy- and were perhaps more duration-dependent than drated to –5% of initial body weight (Francesconi intensity-dependent (Hoffman et al. 1997). et al. 1984). Further, reduced plasma volume per Although the aforementioned findings are diffi- se (–14.6%, versus 0% in a control trial) affects heat cult to interpret, they likely are due to differences in dissipation during prolonged cycling, resulting in experimental designs. In support of this interpreta- increased rectal temperature during dehydration tion, the investigation conducted by Kraemer, W.J. (Roy et al. 2000) and increased whole-body sympath- et al. (1995) observed that: (a) different heavy- etic nervous drive (i.e. plasma NE increased but resistance training protocols resulted in different not epinephrine). Dehydration apparently affects blood β-endorphin and cortisol kinetics; and (b) the sweating, via sympathetic cholinergic innervation duration of force production and the length of rest and the neurotransmitters ACh and NE; it reduces periods between sets were the critical design fea- the sensitivity of eccrine sweat glands to hyperther- tures. Few studies have attempted to separate the mia and increases the temperature at which sweat- independent effects of exercise mode (i.e. cycling, ing begins (Armstrong & Maresh 1998). Thus, running, resistance exercise), design (i.e. continuous sweating begins later in exercise when an individual versus intermittent exercise), intensity, or duration on is dehydrated, versus euhydrated. neuroendocrine factors. One investigation (Brenner Senay (1979) concluded that these hypothalamic et al. 1997) offers a noteworthy exception, in that it responses are largely due to changes in the osmolal- distinguishes the physiological responses to air ity of brain tissue or extracellular fluid osmolality. temperatures of 23°C and 40°C from responses to Three publications from our laboratory support exercise. A thorough understanding of the neuro- this concept. When healthy males performed 90 min endocrine regulation of body temperature, as it is of continuous treadmill exercise (5% grade, 36% o influenced by various exercise protocols, requires V 2max) in a heated chamber (33°C, 56%rh), dew neuroendocrine influences: thermoregulation 479

point sensor measurements indicated that local hormone responses to exercise vary at different sweat rate decreased as plasma osmolality increased points in the menstrual cycle (Hansen & Weeke 1974) (from 281 to 324 mOsm·kg–1) and as body water loss and oral contraceptives potentiate the exercise- (from –0.2 and –6.7% of body mass) increased induced secretion of growth hormone (Bernardes & (Armstrong et al. 1997). During these experiments, Radomski 1998). These effects likely are due to Hoffman et al. (1994) observed that rehydration with the interaction of estrogen with the hypothalamic plain water during exercise significantly reduced regulation of growth hormone release. Estrogen the plasma NE response. This was true whether test also may affect radiative and convective heat loss participants began exercise in a euhydrated or by enhancing the bioavailability of nitric oxide, a hypohydrated state. In a separate study from our potent vasodilator (Chen, Z. et al. 1999). Further, laboratory, Castellani et al. (1997) observed that estradiol and progesterone are known modulators rehydration via either oral or intravenous fluids of sweat rate, cutaneous blood flow and metabolic (both 0.45% NaCl) reduced the plasma NE (i.e. sym- heat production, having the effect of altering core pathetic nerve activity), ACTH and cortisol (i.e. body temperature throughout the menstrual cycle HPA axis activity) increases that occurred in a sep- (Kolka 1999). Although not perfectly aligned with arate experiment that involved no fluid and exercise the question at hand, these studies provide direc- –heat stress. These findings are consistent with the tion and testable hypotheses for future research. concept that rehydration decreased the osmotic pressure (i.e. 307 mOsm·kg–1 without water versus Heat acclimatization adaptations. During 5–14 days 293 mOsm·kg–1 with water), this was sensed by of exposure to environmental stress (i.e. heat, cold, hypothalamic neurons, which in turn reduced both high altitude), a complex of systemic, ultrastruc- the sympathetic nervous drive for sweating and the tural and biochemical adaptations occur that are anterior pituitary release of ACTH. This reasoning named acclimatization in nature or acclimation in an links hydration state and osmolality to neuroendo- artificial environment (Armstrong & Maresh 1991). crine control of body temperature. Obviously, CNS and neuroendocrine adaptations Individual ions also may influence thermoregula- are involved in heat acclimatization, in the form tion. The thorough review of hypothalamic regula- of decreased rectal temperature (i.e. enhanced tion of body temperature by Myers (1980) concludes peripheral effector responses that increase heat dis- that whole-body hyperthermia can be reversed, sipation), heart rate, sweat/urine sodium concen- while laboratory animals exercise, by infusing cal- trations (i.e. aldosterone secretion from the adrenal cium ions into the hypothalamus. This suggests that cortex) and onset temperature for sweat production individual ions (i.e. calcium, sodium, potassium (Armstrong et al. 1987; Armstrong & Maresh 1998); and magnesium) play a role in thermoregulatory these occur concurrently with increased total sweat effector responses and heat dissipation. rate (i.e. sympathetic cholinergic innervation) and plasma volume (i.e. sympathetic vasodilatory con- α β Gender and reproductive hormones. After reviewing trol via - and 2-adrenergic receptors in precapil- numerous publications, we concluded that gender lary blood vessels; see Horowitz & Gival 1989). influences on human neuroendocrine regulation of An animal model of heat acclimation demon- body temperature are incompletely defined, espe- strated central neuroendocrine modifications, in cially because no studies involved exercise in the the form of increased NE and dopamine levels in the heat. However, a few gender-comparison studies POAH (Christman & Gisolfi 1985), in concert with involved either exercise in a mild environment enhanced peripheral heat-dissipating capacity. This (22°C; Horton, T.J. et al. 1998; Kraemer, R.R. et al. agrees with observations that heat-acclimated rats 1989; Davis et al. 2000) or rest in intense heat (90°C; consistently show enhanced thermoregulatory Jezova et al. 1994), and generated variant findings. sensitivity to POAH injections of various neuro- Reproductive hormone fluctuations also may set the modulating agents including NE and 5-HT (Gordon stage for gender differences. For example, growth 1993). 480 chapter 31

Ultrastructural and metabolic adaptations during development of heat acclimatization (Armstrong & heat acclimation have been identified by Horowitz Stoppani 2002). (1998). First, central adaptations have been observed Regarding thermoregulatory and stress hor- in the POAH, including altered membrane com- mones, Table 31.2 summarizes the adaptations ponents, that accompany peripheral changes in that have been reported in humans and animals, membrane receptor density, receptor affinity for subsequent to 7–10 days of exercise–heat acclima- specific molecules, and cytosolic Ca2+ concentra- tion in 35–49°C environments. Acknowledging that tion. Second, whole-body measurements of lower numerous factors influence chronic neuroendocrine o oxygen consumption (i.e. lower V 2 at a given adaptations during heat acclimation (see Fig. 31.3), workload) have been reported in humans who par- the primary benefit of Table 31.2 is that it emphas- ticipated in exercise–heat acclimation programs izes future research that is needed. (Sawka et al. 1983). Little was known about this phenomenon until recently. Animal studies have Physical training adaptations. Because exercise–heat shown that improved biochemical efficiency (i.e. acclimation involves consistent, repeated exercise, it lower metabolic rate) occurs in cardiac tissue, in the is logical that heat-acclimation adaptations would same time frame that increased glycogen content, overlap those that are derived from exercise train- enhanced cellular uptake of glucose and increased ing (Armstrong & Pandolf 1988). Indeed, an endur- glycolytic potential occur (Horowitz 2003). These ance training program in a cool environment results biochemical changes enhance the energy-generat- in a lowered threshold temperature for the onset of ing potential of the heart, allowing it to deal with sweating, increased plasma volume, increased car- the increased cardiovascular strain imposed by diac stroke volume, decreased heart rate, decreased environmental–exercise stress. skin temperature, decreased rectal temperature and o A recent review (Armstrong & Stoppani 2002) increased V 2max. These similarities support numer- describes the emerging paradigm of central nervous ous reports that highly-trained endurance athletes, system control of adaptations during repetitive training exclusively in cool environments, respond thermal stress, and emphasizes neural network to exercise in a hot environment as though they plasticity that is synchronized with altered bio- were heat acclimatized (Armstrong & Maresh 1991). chemical pathways. The authors describe several A few studies have evaluated hormonal responses plausible mechanisms for human physiological to multiweek physical training programs in cool adaptations, including: (a) CNS control of habitu- environments. In general, these studies have ation (i.e. diminished synaptic transmission), the shown an improved functional endocrine capacity, set point of core body temperature or neural circui- expressed as increased blood levels of ACTH, try; (b) biochemical and ultrastructural adaptations epinephrine, NE and β-endorphin (Viru 1992). (i.e. altered cell membrane characteristics; changes Unfortunately, the research regarding the influence in the intracellular concentration of organelles of physical training on neuroendocrine regulation and rate-limiting enzymes of energy metabolism); of human body temperature, in any environment, is and (c) effector organ responses (i.e. change of the limited and does not allow generalizations to be threshold temperature for increased blood flow or advanced at this time. sweat rate; morphological changes in sweat gland or neuromuscular junction size). Because many Exertional heatstroke and heat exhaustion. Exertional neurobiologists believe that information storage heat exhaustion involves an inability to continue is associated with the specific processing areas that exercise in the heat because of cardiovascular are engaged in homeostatic control, the anterior insufficiency. It is the most common heat-related hypothalamus (i.e. thermoregulatory integration), disorder in military, athletic and civilian settings medulla or pons (i.e. heart rate, blood pressure, ven- and recovery is complete within 24–48 h. It results tilation) and the reticular formation (i.e. integration from either water depletion, salt depletion or mixed of neural inputs) are theoretically implicated in the salt–water depletion (Armstrong & Anderson 2003). neuroendocrine influences: thermoregulation 481

Table 31.2 Summary of thermoregulatory and stress hormone adaptations during exercise–heat acclimation (35–49°C) in humans and animals.

Exposure Circulating duration hormone* Research findings (days) References

ACTH Undetermined β-Endorphin HA had no effect on β-endorphin levels in normal males† 8 Kraemer W.J. et al. 1987; Kraemer, W.J. et al. 2003 HA decreased β-endorphin levels in former heatstroke 7 Kraemer, W.J. et al. 2003 patients† hGH HA had no effect on hGH levels in 35°C and 49°C 10 Francesconi et al. 1984 environments when subjects were euhydrated Prolactin Undetermined Cortisol HA (intense intermittent exercise, 50 min) had no effect 8 Kraemer, W.J. et al. 1987 on cortisol levels HA (intense intermittent exercise, 56 min) decreased 8 Armstrong et al. 1989 cortisol levels HA (mild intensity 100 min) decreased cortisol levels 10 Francesconi et al. 1984 when subjects were hypohydrated‡ in a 35°C, but not in a 49°C, environment Epinephrine Undetermined Norepinephrine HA resulted in increased brain NE levels in rats 14 Christman & Gisolfi 1985 HA enhances thermal effector responsiveness to NE 21 Christman & Gisolfi 1980 injected into the POAH Dopamine HA resulted in increased brain dopamine levels§ in rats 14 Christman & Gisolfi 1985 Serotonin HA enhances thermal effector responsiveness to 5-HT d.n.p. Gordon 1993 injected into the POAH Thyroxine HA resulted in a sustained decrease of thyroxine in rats 14–30 Horowitz 1998; Horowitz 2003

ACTH, adrenocorticotropic hormone; d.n.p., data not provided; 5-HT, serotonin; hGH, human growth hormone; HA, exercise–heat acclimation in humans; NE, norepinephrine. *The physiological significance of these hormones appears in Table 31.1 footnotes. †β-Endorphin levels increased during each daily exposure to exercise–heat stress. ‡Hypohydration (–5% body weight loss) increased plasma cortisol levels (versus euhydration) on day 1 but not day 10 of HA. §As determined by analysis of a dopamine metabolite (3,4-dihydroxyphenylglycol).

Exertional heatstroke, a medical emergency, occurs heatstroke and exertional heat exhaustion. In addi- when the combined thermal stresses from the tion to the SAM-axis or HPA-axis stress responses to environment and muscular metabolism exceed the hyperthermia and exhaustive exercise, these invest- body’s heat dissipation capacity. A dangerous igations indicate that β-endorphin, dopamine and 5- hyperthermia may lead to profound CNS involve- HT participate in the pathophysiology of exertional ment, multisystem failure and death if the body is heat illnesses. not cooled promptly (Casa & Armstrong 2003). Ethical principles dictate that investigators shall Future research not induce exertional heat illness in test particip- ants. Therefore, virtually all investigations of the Many factors influence the neuroendocrine regula- neuroendocrine responses to exertional heat illness tion of body temperature during exercise in a hot arise from retrospective analyses. Table 31.3 sum- environment. In addition to the factors delineated in marizes the scientific literature regarding exertional Fig. 31.3, these include interactions among specific 482 chapter 31

Table 31.3 Thermoregulatory and stress hormone responses of exertional heat-illness patients and laboratory animals.

Circulating hormone* Illness† Experimental protocol Research findings References

ACTH Unknown β-Endorphin Exh 8-day laboratory HA‡; β-Endorphin level ↑ 20-fold Armstrong et al. 1988§ Exh in one male on day 8 (versus baseline) HS 7-day laboratory HA¶ β-Endorphin level ↑ within-day Kraemer, W.J. et al. 2003 (patients & controls); β-Endorphin response ↓ by day 8 of HA (patients only) HS Religious trek to Mecca, in desert β-Lipotropin/β-endorphin at Appenzeller et al. 1986 admission ↑ 42.7-fold (versus later recovery) HBI 24 rats; passive heat exposure for Blockade of opioid peptide** Sharma et al. 1997 4 h, 38°C receptors ↓ edema, cell damage and ischemia GH Exh Religious trek to Mecca, in desert No change during treatment Kashmeery 1995§ HS Religious trek to Mecca, in desert GH at admission ↑ 4.1-fold Appenzeller et al. 1986 (versus later recovery) Prolactin HS Religious trek to Mecca, in desert Prolactin at admission ↑ 3.9-fold Appenzeller et al. 1986 (versus later recovery) Cortisol Exh Religious trek to Mecca, in desert No change during treatment Kashmeery 1995§ Exh 8-day laboratory HA‡; Exh in one Post-Exh cortisol level ↑ Armstrong et al. 1988§ male on day 8 Twofold (versus day 1 and 4) HS Religious trek to Mecca, in desert Cortisol at admission ↑ 1.7-fold Appenzeller et al. 1986 (versus later recovery) Epinephrine Unknown Norepinephrine Unknown Dopamine CIND Rats; passive exposure, 42°C Depletion of brain dopamine†† Lin 1997 ↓ CIND and ↑ survival time Serotonin CIND Rats; passive exposure, 42°C Depletion of brain serotonin‡‡ Lin 1997 (5-HT) ↓ CIND and ↑ survival time

ACTH, adrenocorticotropic hormone; CIND, cerebral ischemia and neural damage in rats; Exh, human exertional heat exhaustion; GH, growth hormone; HA, exercise-induced heat acclimation in a laboratory; HBI, hyperthermic brain injury in rats; HS, human exertional heatstroke. *The physiological significance of these hormones appears in Table 31.1 footnotes. †To distinguish exertional heatstroke from exertional heat exhaustion, see text. ‡HA program involved 8 days of high-intensity treadmill running (50-min exercise, 100-min total exposure to 41°C). §Fluid-electrolyte regulating hormones also were measured (not shown). ¶HA consisted of 90-min walking at 40°C; nine former HS patients versus eight healthy, matched-control subjects. **Findings represent opioid peptides as a class, including β-endorphin; naloxone and naltrexone were used to block receptors. ††Dopamine depletion was accomplished by 6-hydroxydopamine injection. ‡‡Serotonin depletion was accomplished by injection of either methysergide or lysergic acid dithylamide. neurotransmitters and neuropeptides, multiple affer- Future research certainly is needed to interpret ent influences on hypothalamic and pituitary func- this complex matrix. Acknowledging that today’s tion, pulsatile release (i.e. growth hormone) or technology is inadequate, we propose the following circadian rhythms (i.e. ACTH, cortisol, hGH, pro- areas of investigation regarding neuroendocrine lactin) of some hormones, as well as feedback and influences on thermoregulation during exercise in a feedforward controls (Veldhuis & Yoshida 2000). hot environment: neuroendocrine influences: thermoregulation 483

• Determine the differences between central (i.e. • Identify the unique effects of exercise mode, brain) and peripheral (i.e. blood) neuroendocrine duration, intensity and intermittency on neuro- responses. endocrine responses. • Examine the specific effects of putative neuro- • Clarify the course of chronic neuroendocrine transmitters and neuropeptides on the different adaptations that occur during heat acclimatization. types of POAH neurons (e.g. warm sensitive, cold • Investigate the influence of nutritional influences, sensitive). including dehydration, on neuroendocrine responses. • Distinguish the individual effects of different • Evaluate the changes in neuroendocrine function stressors (i.e. exercise, intense environmental heat) following exertional heatstroke and heat exhaus- on neuroendocrine responses. tion, to determine if permanent disabilities result.

References

Adams, W.C., Fox, R.H., Fry, A.J. & Armstrong, L.E. & VanHeest, J.L. (2002) World Scientific, River Edge, NJ: MacDonald, I.C. (1975) The unknown mechanism of the 178–205. Thermoregulation during marathon overtraining syndrome. Sports Medicine Boisvert, P., Brisson, G.R. & Peronnet, F. running in cool, moderate and hot (New Zealand) 32, 185–209. (1993) Effect of plasma prolactin on environments. Journal of Applied Armstrong, L.E., Hubbard, R.W., Jones, sweat rate and sweat composition Physiology 38, 1030–1037. B.H. & Daniels, J.T. (1986) Preparing during exercise in man. American Aldercreutz, H., Kuoppasalmi, K., Alberto Salazar for the heat of the 1984 Journal of Physiology 264, F816–F820. Kosunen, K., Pakarinen, A. & Karonen, Olympic marathon. Physician and Sports Borer, K.T. (2003) Exercise as an S.L. (1976) Plasma cortisol, growth Medicine 14, 73–81. emergency and a stressor. In: Exercise hormone and prolactin levels during Armstrong, L.E., Hubbard, R.W., DeLuca, Endocrinology. Human Kinetics, exposure to intense heat. IRCS Medical J.P. & Christensen, E.L. (1987) Heat Champaign, IL: 77–95. Science: Endocrine System, Environmental acclimatization during summer running Boulant, J.A. (1996) Hypothalamic neurons Biology and Medicine 4, 546. in the northeastern United States. controlling body temperature. In: Appenzeller, O., Khogali, M., Carr, D.B. Medicine and Science in Sports and Exercise Handbook of Physiology: Section 4, et al. (1986) Makkah hajj: heat stroke and 19, 131–136. Environmental Physiology (Blatteis, C.M. endocrine responses. Annals of Sports Armstrong, L.E., Hubbard, R.W., Szlyk, & Fregly, M.J., eds.). Oxford University Medicine 3, 30–32. P.C., Sils, I.V. & Kraemer, W.J. (1988) Press, New York: 105–126. Armstrong, L.E. & Anderson, J.M. (2003) Heat intolerance, heat exhaustion Boulant, J.A. & Hardy, J.D. (1974) Heat exhaustion, exercise-associated monitored: a case report. Aviation Space The effect of spinal and skin collapse, and heat syncope. In: Exertional Environmental Medicine 59, 262–266. temperatures on the firing rate and local Heat Illnesses (Armstrong, L.E., ed.). Armstrong, L.E., Francesconi, R.P., thermosensitivity of preoptic neurons. Human Kinetics, Champaign, IL: 57–90. Kraemer, W.J. et al. (1989) Plasma Journal of Physiology 216, 1371–1374. Armstrong, L.E. & Maresh, C.M. (1991) cortisol, renin, and aldosterone during Brenner, I.K.M., Zamecnik, J., Shek, P.N. & The induction and decay of heat an intense heat acclimation program. Shephard, R.J. (1997) The impact of heat acclimatization in trained athletes. Sports International Journal of Sports Medicine exposure and repeated exercise on Medicine 12, 302–312. 10(1), 38–42. circulating stress hormones. European Armstrong, L.E. & Maresh, C.M. (1998) Armstrong, L.E., Maresh, C.M., Gabaree, Journal of Applied Physiology 76, 445–454. Effects of training, environment, and C.V. et al. (1997) Thermal and circulatory Brezenoff, H.E. & Lomax, P. (1970) host factors on the sweating response to responses during exercise: effects of Temperature changes following exercise. International Journal of Sports hypohydration, dehydration, and water microinjection of histamine into the Medicine 19, S103–S105. intake. Journal of Applied Physiology 82, thermoregulatory centers of the rat. Armstrong, L.E. & Pandolf, K.B. (1988) 2028–2035. Experientia 26, 51–52. Physical training, cardiorespiratory Baumann, G. (1999) Growth hormone Casa, D.J. & Armstrong, L.E. (2003) physical fitness and exercise–heat heterogeneity in human pituitary and Exertional heatstroke: a medical tolerance. In: Human Performance plasma. Hormone Research 51 (suppl. 1), emergency. In: Exertional Heat Illnesses Physiology and Environmental Medicine at 2–6. (Armstrong, L.E., ed.). Human Kinetics, Terrestrial Extremes (Pandolf, K.B., Bernardes, R.P. & Radomski, M.W. Champaign, IL: 29–56. Sawka, M.N. & Gonzalez, R.R., eds.). (1998) Growth hormone responses to Castellani, J.W., Maresh, C.M., Armstrong, Benchmark Press, Indianapolis, IN: continuous and intermittent exercise L.E. et al. (1997) Intravenous vs. oral 199–226. in females under oral contraceptive rehydration: effects on subsequent Armstrong, L.E. & Stoppani, J. (2002) therapy. European Journal of Applied exercise-heat stress. Journal of Applied Central nervous system control of heat Physiology 79(1), 24–29. Physiology 82(3), 799–806. acclimation adaptations: an emerging Blatteis, C.M. (1998) Fever. In: Physiology Chen, X.-M., Hosono, T., Yoda, T., Fukuda, paradigm. Reviews in the Neurosciences and Pathophysiology of Temperature Y. & Kanosue, K. (1998) Efferent 13, 271–285. Regulation (Blatteis, C.M., ed.). projection from the preoptic area for the 484 chapter 31

control of non-shivering thermogenesis the cerebral ventricles. A new concept Hjortskov, J.A., Jepsen, L.T., Nielsen, B. in rats. Journal of Physiology 512, 883–892. of temperature regulation. Journal of et al. (1995) Growth hormone deficiency Chen, Z., Yuhanna, I.S., Galcheva- Physiology (London) 173, 226–237. and hyperthermia during exercise: a Gargova, Z. et al. (1999) Estrogen Follenius, M., Brandenberger, G., Simeoni, controlled study of 16 GH-deficient receptor α mediates the nongenomic M. & Reinhardt, B. (1979) Plasma patients. Journal of Clinical Endocrinology activation of endothelial nitric oxide aldosterone, prolactin, ACTH: and Metabolism 80(11), 3335–3340. synthase by estrogen. Journal of Clinical relationships in man during heat Hoffman, J.R., Maresh, C.M., Armstrong, Investigation 103(3), 401–406. exposure. Hormone and Metabolic L.E. et al. (1994) Effects of hydration state Christman, J.V. & Gisolfi, C.V. (1980) Research 11, 180–181. on plasma testosterone, cortisol and Effects of repeated heat exposure Francesconi, R.P., Sawka, M.N. & Pandolf, catecholamine concentrations before on hypothalamic sensitivity to K.B. (1984) Hypohydration and and during mild exercise at elevated norepinephrine. Journal of Applied acclimation: effects on hormone temperature. European Journal of Applied Physiology 49(6), 942–945. responses to exercise/heat stress. Physiology 69, 294–300. Christman, J.V. & Gisolfi, C.V. (1985) Heat Aviation Space and Environmental Hoffman, J.R., Falk, B., Radom-Isaac, S. acclimation: role of norepinephrine in Medicine 55, 365–369. et al. (1997) The effect of environmental the anterior hypothalamus. Journal of Galbo, H. (1986) Autonomic temperature on testosterone and cortisol Applied Physiology 58(6), 1923–1928. neuroendocrine responses to exercise. responses to high intensity, intermittent Clark, W.G. (1979) Changes in body Scandinavian Journal of Sports Science 8, exercise in humans. European Journal of temperature after administration of 3–17. Applied Physiology 75, 83–87. amino acids, peptides, dopamine, Gibbins, I.L. & Morris, J.L. (2000) Pathway Horowitz, M. (1998) Do cellular heat neuroleptics and related agents. specific expression of neuropeptides and acclimation responses modulate central Neuroscience and Biobehavioral Reviews 3, autonomic control of the vasculature. thermoregulatory activity? News in 179–231. Regulatory Peptides 93(1–3), 93–107. Physiological Sciences 13, 218–225. Clark, W.G. & Fregly, M.J. (1996) Giustina, A. & Veldhuis, J.D. (1998) Horowitz, M. (2003) Matching the heart Evidence for roles of brain peptides Pathophysiology of the neuroregulation to heat-induced circulatory load: in thermoregulation. In: Handbook of of growth hormone secretion in heat-acclimatory responses. News in Physiology: Section 4, Environmental experimental animals and the human. Physiological Sciences 18, 215–221. Physiology, vol. 1 (Blatteis, C.M. & Endocrine Reviews 19(6), 717–797. Horowitz, M. & Gival, N. (1989) Heat Fregly, M.J., eds.). Oxford University Gordon, C.J. (1993) Temperature Regulation acclimation and heat stress: cardiac Press, New York: 139–153. in Laboratory Rodents. Cambridge output distribution, plasma volume Clark, W.G. & Lipton, J.M. (1986) University Press, New York: 180–190. expansion and the involvement Changes in body temperature after Guyton, A.C. & Hall, J.E. (1996) Textbook of of the adrenergic pathway. In: administration of adrenergic and Medical Physiology, 9th edn. W.B. Thermoregulatory Research and Clinical serotonergic agents and related Saunders Co., Philadelphia: 769–777. Applications (Lomax, P. & Schönbaum, drugs including antidepressants: II. Hammel, H.T. (1965) Neurons and E., eds.). Karger, Basel, Switzerland: Neuroscience and Biobehavioral Reviews 10, temperature regulation. In: Physiological 204–207. 153–220. Controls and Regulations (Yamamoto, Horton, R.W., LeFeuvre, R.A., Rothwell, Davis, S.N., Galassetti, P., Wasserman, W.S. & Brobeck, J.R., eds.). W.B. N.J. & Stock, M.J. (1988) Opposing

D.H. & Tate, D. (2000) Effects of gender Saunders Co., Philadelphia: 71–97. effects of activation of central GABAA on neuroendocrine and metabolic Hannuksela, M.L. & Samer, E. (2001) and GABAB receptors on brown counterregulatory responses to exercise Benefits and risks of sauna bathing. fat thermogenesis in the rat. in normal man. Journal of Clinical American Journal of Medicine 110(2), Neuropharmacology 27, 363–366. Endocrinology and Metabolism 85, 224–230. 118–126. Horton, T.J., Pagliasotti, M.J., Hobbs, K. DeSouza, E.B. & Appel, N.M. (1991) Hansen, A.P. & Weeke, J. (1974) Fasting & Hill, J.O. (1998) Fuel metabolism in Distribution of brain and pituitary serum growth hormone levels and men and women during and after long- receptors involved in mediating stress growth hormone responses to exercise duration exercise. Journal of Applied responses. In: Stress—Neurobiology and during normal menstrual cycles Physiology 85(5), 1823–1832. Neuroendocrinology (Brown, M.R., Koob, and cycles of oral contraceptives. Hussi, E., Sonck, T., Poso, H. et al. (1977) G.F. & Rivier, C., eds.). Marcel Dekker, Scandinavian Journal of Clinical Laboratory Plasma catecholamines in Finnish sauna. New York: 91–117. Investigations 34(3), 199–205. Annals of Clinical Research 9, 301–304. Eguchi, N., Minami, T., Shirafuji, N. et al. Hartung, G.H., Myhre, L.G., Tucker, D.M. Jackson, H.C. & Nutt, D.J. (1991) Inhibition (1999) Lack of tactile pain (allodynia) & Burns, J.W. (1987) Hormone and of baclofen-induced hypothermia in

in lipocalin-type prostaglandin D energy substrate changes during mice by the novel GABAB antagonist synthase-deficient mice. Proceedings of prolonged exercise in the heat. Aviation CGP 35348. Neuropharmacology 30, the National Academy of Sciences of the Space Environmental Medicine 58, 535–538. United States of America 96, 726–730. 24–28. Jessen, C. (1996) Interaction of body Feldberg, W. & Myers, R.D. (1963) A new Hasan, J., Karvonen, M.J. & Piironen, P. temperatures in control of concept of temperature regulation by (1966) Special review. Part I. thermoregulatory effector mechanisms. amines in the hypothalamus. Nature Physiological effects of extreme heat as In: Handbook of Physiology: Section 4, (London) 200, 1325. studied in the Finnish ‘sauna’ bath. Environmental Physiology, vol. 1 (Blatteis, Feldberg, W. & Myers, R.D. (1964) Effects American Journal of Physical Medicine C.M. & Fregly, M.J., eds.). Oxford on temperature of amines injected into 45(6), 296–314. University Press, New York: 127–138. neuroendocrine influences: thermoregulation 485

Jezova, D., Vigas, M., Tatar, P., Kraemer, W.J., Patton, J.F., Gordon, S.E. man. Journal of Clinical Endocrinology and Jurcovicova, J. & Palat, M. (1985) Rise et al. (1995) Compatibility of high- Metabolism 52(2), 279–283. in plasma β-endorphin and ACTH in intensity strength and endurance Minano, F.J., Serrano, J.S., Sancibrian, M. response to hyperthermia in sauna. training on hormonal and skeletal & Myers, R.D. (1989) Effects of Hormone and Metabolic Research 17, muscle adaptations. Journal of Applied central administration of GABA on 693–694. Physiology 78(3), 976–989. thermoregulation. In: Thermoregulation: Jezova, D., Kvetnansky, R. & Vigas, M. Kraemer, W.J., Armstrong, L.E. & Watson, Research and Clinical Applications (Lomax, (1994) Sex differences in endocrine G. (2003) The effects of exertional P. & Schönbaum, E., eds.). Karger, Basel, response to hyperthermia in sauna. Acta heatstroke and exercise–heat Switzerland: 184–186. Physiologica Scandinavica 150, 293–298. acclimation on plasma β-endorphin Mora-Rodriguez, R., Gonzalez-Alonso, J., Kanosue, K., Hosono, T., Zhang, Y.-H., concentrations. Aviation Space Belo, P.R. & Coyle, E.F. (1996) Plasma Chen, X.-M. (1998) Neuronal networks Environmental Medicine 74(7), 758–762. catecholamines and hyperglycemia controlling thermoregulatory effectors. Kukkonen-Harjula, K., Oja, P., Laustiola, influence thermoregulation in man Progress in Brain Research 115, 49–62. K. et al. (1989) Haemodynamic and during prolonged exercise in the heat. Kashmeery, A. (1995) Physiological studies hormonal responses to heat exposure in Journal of Physiology (London) 491(2), on heat exhaustion victims among a Finnish sauna bath. European Journal of 529–540. Mecca pilgrims. Acta Medica Austriaca Applied Physiology 58, 543–550. Mosqueda-Garcia, R. (1996) Central 22, 16–22. Laatikainen, T., Salminen, K., Hohvakka, autonomic regulation. In: Primer on the Kaufman, F.L., Mills, D.E., Hughson, A. & Pettersson, J. (1988) Response of Autonomic Nervous System (Robertson, D. R.L. & Peake, G.T. (1988) Effects of plasma endorphins, prolactin and & Polinsky, R.J., eds.). Academic Press, bromocriptine on sweat gland function catecholamines in women to intense San Diego, CA: 3–12. during heat acclimatization. Hormone heat in a sauna. European Journal of Myers, R.D. (1980) Hypothalamic control Research 29, 31–38. Applied Physiology 57, 98–102. of thermoregulation: neurochemical Kellogg, D.L., Liu, Y., Kosiba, I.F. & Lammintausta, R., Syvalahti, E. & mechanisms. In: Handbook of the O’Donnell, D. (1999) Role of nitric oxide Pekkarinen, A. (1976) Change in Hypothalamus, vol. 3. Behavioral Studies in the vascular effects of local warming hormones reflecting sympathetic of the Hypothalamus (Morgane, P.J. & of the skin in humans. Journal of Applied activity in the Finnish sauna. Annals of Panksepp, J., eds.). Marcel Dekker, Physiology 86, 1185–1190. Clinical Research 8, 266–271. New York: 83–210. Kelso, T.B., Herbert, W.G., Gwazdauskas, Leppaluoto, J., Huttunen, P. & Hirvonen, J. Nakamori, T., Morimoto, A. & Murakami, F.C., Goss, F.L. & Hess, J.L. (1984) (1986) Endocrine effects of repeated N. (1993) Effect of a central CRF Exercise–thermoregulatory stress sauna bathing. Acta Physiologica antagonist on cardiovascular and and increased plasma β-endorphin/ Scandinavica 128, 467–470. thermoregulatory responses induced β-lipotropin in humans. Journal of Lin, M.T. (1997) Heat stroke-induced by stress or IL-1β. American Journal of Applied Physiology 57(2), 444–449. cerebral ischemia and neuronal damage. Physiology 265, R834–R839. Kolka, M.A. (1999) Adaptations to Annals of the New York Academy of Navaratnam, N., Herndon, D.N., changing reproductive hormonal status. Sciences 873, 572–580. Woodson, L.C., Linares, H.A., Morris, S. In: Adaptation Biology and Medicine, vol. 2 Lobie, P.E., Breipohl, W., Lincoln, D.T., & Traber, D.L. (1992) Endorphin (Pandolf, K.B., Takeda, N. & Singal, P.K., Garcia-Aragon, J. & Waters, M.J. (1990) mediation of mesenteric blood flow eds.). Narosa Publishing House, New Localization of the growth hormone after endotoxemia in sheep. Critical Delhi, India: 360–364. receptor/binding protein in skin. Journal Care Medicine 20, 402–408. Kolka, M.A. & Stephenson, L.A. (1997) of Endocrinology 126(3), 467–471. Nishiyama, T., Gyermek, L., Lee, C. et al. Effect of luteal phase elevation in core Lomax, P. & Green, M.D. (1981) (2001) The analgesic interaction between temperature on forearm blood flow Histaminergic neurons in the intrathecal clonidine and glutamate during exercise. Journal of Applied hypothalamic thermoregulatory receptor antagonists on thermal and Physiology 82, 1070–1083. pathways. Federation Proceedings 40, formalin-induced pain in rats. Anesthesia Kraemer, R.R., Blair, S., Kraemer, G.R. 2741–2745. and Analgesia 92, 725–732. & Castracane, V.D. (1989) Effect of Maier, S.F. & Watkins, L.R. (1998) Nybo, L., Møller, K., Volianitis, S., Nielsen, treadmill running on plasma β- Cytokines for psychologists: B. & Secher, N.H. (2002) Effects of endorphin, corticotrophin, and cortisol implications of bi-directional hyperthermia on cerebral blood flow levels in male and female 10 K runners. immune-to-brain communication for and metabolism during prolonged European Journal of Applied Physiology 58, understanding behavior, mood and exercise in humans. Journal of Applied 845–851. cognition. Psychological Reviews 105, Physiology 93, 58–64. Kraemer, W.J., Armstrong, L.E., 83–207. Rothwell, N.J. (1994) CNS regulation of Marchitelli, L.J., Hubbard, R.W. & Malone, D.T. & Taylor, D.A. (2001) thermogenesis. Clinical Reviews in Leva, N. (1987) Plasma opioid peptide Involvement of somatodendritic Neurobiology 8(1/2), 1–10. responses during heat acclimation in 5-HT(1A) receptors in delta(9)- Roy, B.D., Green, H.J. & Burnett, M.E. humans. Peptides 8, 715–719. tetrahydrocannabinol-induced (2000) Prolonged exercise following Kraemer, W.J., Patton, J.F., Knutgen, H.G. hypothermia in the rat. Pharmacology diuretic-induced hypohydration: et al. (1989) Hypothalamic–pituitary– Biochemistry and Behavior 69, 595–601. effects on cardiovascular and thermal adrenal responses to short-duration Mills, D.E. & Robertshaw, D. (1981) strain. Canadian Journal of Physiology and high-intensity cycle exercise. Journal of Response of plasma prolactin to changes Pharmacology 78(7), 541–547. Applied Physiology 66(1), 161–166. in ambient temperature and humidity in Salt, T.E. & Turner, J.P. (1998) Reduction 486 chapter 31

of sensory and metabotropic glutamate Sugimoto, Y., Ohkura, M., Inoue, K. & during thermal stress. Hormone and receptor responses in the thalamus by Yamada, J. (2000) Involvement of the 5- Metabolic Research 22, 44–46. the novel metabotropic glutamate HT(2) receptor in hyperthermia induced Vescovi, P.P., Casti, A., Michelini, M. et al. receptor-1-selective antagonist S-2- by p-chloroamphetamine, a serotonin- (1992) Plasma ACTH, β-endorphin, methyl-4-carboxy-phenylglycine. releasing drug in mice. European Journal prolactin, growth hormone and Neuroscience 85, 655–658. of Pharmacology 403, 225–228. luteinizing hormone levels after thermal Sawka, M.N., Pandolf, K.B., Avellini, Tabata, I., Ogita, F., Miyachi, M. & stress, heat and cold. Stress Medicine 8, B.A. & Shapiro, Y. (1983) Does heat Shibayama, H. (1991) Effect of low blood 187–191. acclimation lower the rate of metabolism glucose on plasma CRF, ACTH, and Viru, A. (1992) Plasma hormones and elicited by muscular exercise? Aviation cortisol during prolonged physical physical exercise. International Journal of Space Environmental Medicine 54, 27–31. exercise. Journal of Applied Physiology 71, Sports Medicine 13(3), 201–209. Senay, L.C. (1979) Temperature regulation 1807–1812. Yasumatsu, M., Yazawa, T., Otokawa, M. and hypohydration: a singular view. Tache, Y., Martinez, V., Million, M. & et al. (1998) Monoamines, amino acids Journal of Applied Physiology 47(1), 1–7. Wang, L. (2001) Stress and the and acetylcholine in the preoptic area Sharma, H.S., Westman, J., Cervos- gastrointestinal tract. III. Stress-related and anterior hypothalamus of rats: Navarro, J., Dey, P.K. & Nyberg, F. alterations of gut function: role of brain measurements of tissue extracts and (1997) Opioid receptor antagonists corticotropin-releasing factor receptors. in vivo microdialysates. Comparative attenuate heat stress-induced reduction American Journal of Physiology 280, Biochemistry and Physiology 121, 13–23. in cerebral blood flow, increased G173–G177. Zeisberger, E. (1998) Biogenic amines and blood–brain barrier permeability, Thuma, J.R., Gilders, R., Verdun, M. & thermoregulatory changes. In: Progress vasogenic edema and cell changes in Loucks, A.B. (1995) Circadian rhythm in Brain Research, vol. 115 (Sharma, H.S. the rat. Annals of the New York Academy of cortisol confounds cortisol responses & Westman, J., eds.). Elsevier Science, of Sciences 873, 559–571. to exercise: implications for future New York: 159–176. Silva, N.L. & Boulant, J.A. (1986) Effects of research. Journal of Applied Physiology Zeisberger, E. & Roth, J. (1996) Central testosterone, estradiol, and temperature 78(5), 1657–1664. regulation of adaptive responses to heat on neurons in preoptic tissue slices. Toates, F. (1995) Stress. Conceptual and and cold. In: Handbook of Physiology: American Journal of Physiology 250, Biological Aspects. John Wiley and Sons, Section 4, Environmental Physiology, vol. 1 R625–R632. New York: 2–6. (Blatteis, C.M. & Fregly, M.J., eds.). Smith, L.L. (2000) Cytokine hypothesis of Uno, T. & Shibata, M. (2001) Role of Oxford University Press, New York: overtraining: a physiological adaptation inferior olive and thoracic IML neurons 579–595. to excessive stress? Medicine and Science in nonshivering thermogenesis in rats. Zhang, Y.-H., Motoko-Fujiwara, M., in Sports and Exercise 32, 317–331. American Journal of Physiology 280, Hosono, T. & Kanosue, K. (1995) Warm Stephenson, L.A. & Kolka, M.A. (1999) R536–R546. and cold signals from the preoptic area: Esophageal temperature threshold for Veldhuis, J.D. & Yoshida, M.D. (2000) which contribute more to the control of sweating decreases before ovulation in Impact of chronic training on pituitary shivering in rats? Journal of Physiology premenopausal women. Journal of hormone secretion in the human. In: 485, 195–202. Applied Physiology 86, 22–28. Contemporary Endocrinology: Sports Zhang, Y.-H., Hosono, T., Motoko- Strbak, V., Tatar, P., Angyal, R. et al. (1987) Endocrinology (Warren, M.P. & Fujiwara, M., Chen, X.-M. & Kanosue, K. Effects of sauna and glucose intake on Constantini, N.W., eds.). Humana (1997) Effect of midbrain stimulations TSH and thyroid hormone levels in Press, Totowa, NJ: 57–76. on thermoregulatory vasomotor plasma of euthyroid subjects. Metabolism Vescovi, P.P., Gerra, G., Pioli, M. et al. responses in rats. Journal of Physiology 36, 426–431. (1990) Circulating opioid peptides 503, 177–187. Chapter 32

Alterations in Arginine Vasopressin with Exercise, Environmental Stress and Other Modifying Factors

CARL M. MARESH AND DANIEL A. JUDELSON

Introduction tide, control of blood volume and blood pressure (Sampson 1992). In the healthy human body a complex set of endo- Synthesis of AVP mainly occurs in the cell bodies crine and behavioral mechanisms are present to of magnocellular neurosecretory neurons within control fluid and electrolyte balance. The inter- the supraoptic nuclei and to a lesser extent in the actions and co-ordinated effects of arginine vaso- paraventricular nuclei (Ludwig 1998). It is trans- pressin (AVP), the renin–angiotensin–aldosterone ported bound to a specific precursor molecule, system, thirst and drinking behavior and normal neurophysin II, as neurosecretory granules along kidney function are central to the preservation and nerve axons. When bound to its neurophysin, the maintenance of fluid and electrolyte homeostasis at half-life of AVP is increased fivefold, from about rest. Similarly, in the context of exercise and other 2 min to more than 10 min. There are at least four perturbations (e.g. environmental stress) AVP is a nerve tracts that deliver AVP to different parts of the major factor responsible for water conservation. In brain that are thought to be the major sites of its this chapter we focus our attention on the AVP observed central nervous system actions, but the response to various physical and pathological situ- major pathway terminates in the posterior pituitary. ations. We begin with a basic review of AVP physio- From here AVP and neurophysin II are released via logy followed by a review of the current knowledge calcium-dependent exocytosis into the systemic related to the AVP response to exercise, training, circulation in equimolar quantities. It also should be exercise and hydration status, modifying factors noted that the high levels of AVP in portal blood (including menstrual phase and status, pregnancy, provides the means by which it serves to modulate gender and aging) and environmental influences. anterior pituitary function (Sampson 1992). AVP circulates in the blood unbound to other proteins and is rapidly removed, mainly by the kidney, from Basic endocrinology of arginine vasopressin the circulation. In humans the major determinant of fluid balance is Three distinct AVP receptor subtypes have been the ability of the body to regulate the intake and identified. Each of these has seven transmembrane excretion of water. For the most part this is achieved spanning domains and each is G-protein coupled. by the stimulation and suppression of osmotically However, they are encoded by different genes, and sensitive cells in the hypothalamus that control the differ in tissue distribution, down-stream signal thirst mechanism and the release of AVP. AVP is a transduction and function (Hurbin et al. 1998). The neurohypophyseal hormone that plays an import- V2-receptor is most specific to the actions of AVP in ant regulatory role in water conservation and the the collecting duct of the distal nephron of the kidney. maintenance of body fluid osmolality, and in con- AVP has many actions but its primary physiolo- cert with aldosterone and atrial natriuretic pep- gical effect is regulation of water reabsorption in the 487 488 chapter 32

Table 32.1 Selected functions of arginine vasopressin (Maresh et al. 1985; McKenna & Thompson 1998). (AVP). Nevertheless, the normal relationship between Posm and AVP concentrations will breakdown due to: (i) Increases the passive water permeability of the cell membrane of nephron collecting ducts: rapid increases in Posm resulting in an exaggerated increases water retention AVP release; (ii) drinking rapidly, which suppresses reduces serum osmolality AVP release through afferent pathways originating A potent vasoconstrictor of splancnic, hepatic and renal in the oropharynx (Arnauld & du Pont 1982); and vessels (iii) aging, which increases sensitivity of AVP to an A neurotransmitter in CNS regulation of: osmotic stimulus (Helderman et al. 1978; Bevilacqua the secretion of ACTH et al. 1987). The influence of aging is presented in the cardiovascular system greater detail later in this review. temperature and other visceral function It is important to note for this discussion that Promotes hemostasis: similar to AVP, there also is a linear relationship release of endothelial coagulation factors between thirst and Posm in the physiological range increases platelet aggregation (Stevenson 1969). Mean osmotic threshold for thirst Increases blood volume and blood pressure perception approximates 281 mOsm·kg–1, such that

thirst occurs when Posm rises above this thres- ACTH, adrenocorticotropic hormone; CNS, central hold. Furthermore, as with AVP, there are specific nervous system. physiological conditions in which the relationship

between Posm and thirst will break down. These include: (i) drinking, which will reduce osmotic- distal nephron. Table 32.1 presents some of the ally stimulated thirst; (ii) an extracellular volume other functions attributed to AVP. The structure depletion, which will stimulate thirst through and transport processes of the distal nephron allow volume-sensitive cardiac afferents and the genera- the kidney to both concentrate and dilute urine in tion of circulating and intracerebral angiotensin II, proportion to the circulating AVP concentration. By a powerful dipsogen (Reid 1984); and (iii) aging, way of selective water channel proteins (aquapor- when both thirst and fluid intake can be blunted in ins) present in the wall of the distal nephron, water the elderly. is reabsorbed from the duct lumen along an osmotic Under specific conditions physiological factors gradient resulting in excretion of concentrated urine other than osmolality can increase AVP release. (Marples 2000). Some of these are presented in Table 32.2.

Regulation of arginine vasopressin release Exercise Under resting conditions the major stimulus of Endurance exercise AVP secretion is plasma osmolality (Posm) (Baylis & Robertson 1980). The osmoregulatory systems for By the early 1980s, research had confirmed that thirst and AVP secretion, and in turn the actions whole body endurance exercise increases the of AVP on renal water excretion, serve to maintain concentration of plasma AVP (Wade 1984). This –1 Posm within narrow limits: 284–295 mOsm·kg . In has been repeatedly shown with cycle ergometry general, the mean osmolar threshold for AVP (Convertino et al. 1980a, 1983; Melin et al. 1980; release approximates 284 mOsm·kg–1, above which Brandenberger et al. 1986, 1989; Grant et al. 1996; Roy plasma AVP increases in response to increases in et al. 2001a) and treadmill running (Beardwell et al.

Posm in a linear manner. Considerable variations in 1975; Wade & Claybaugh 1980; Maresh et al. 1985; both the threshold and sensitivity of AVP release De Souza et al. 1989; Melin et al. 1997; Mudambo have been noted between individuals, but these are et al. 1997). Physiologically, exercise stimulates AVP quite reproducible for a given person over time release through several mechanisms. Although arginine vasopressin 489

Table 32.2 Non-osmotic factors that can cause an increase baroreceptors, afferent sympathetic impulses and in arginine vasopressin (AVP) release. renal metabolism. Not all exercise, however, stimulates AVP secre- • A fall in blood pressure • Hemorrhage tion. Although exercise mode and duration are • Nausea/vomiting relatively unimportant determinants of AVP con- • Pain/anxiety centration, AVP release is highly intensity depend- • Increased body temperature ent (Wade 1984) and is only stimulated above • Hypoxia 40–65% of Vo (Convertino et al. 1981, 1983; • Hypercapnia 2max • Humoral factors: thyrotropin-releasing hormone, Montain et al. 1997). Exercise at intensities below o catecholamines, renin–angiotensin–aldosterone system, this % V 2max threshold may be incapable of stimu- endogenous opioids, prostaglandins, estrogens, lating the necessary changes in Posm and plasma progestins volume required to elicit an AVP response. Above • Drugs: morphine, narcotic analogs, nicotine, the threshold, a direct correlation between exer- anesthetics, tranqualizers, barbiturates cise intensity and AVP secretion has been fre- quently observed, although the linearity (Wade & Claybaugh 1980; Convertino et al. 1981) or curvilin- exercise-induced increases in Posm (Convertino et al. earity (Convertino et al. 1983) of the relationship is 1980b; Brandenberger et al. 1989; De Souza et al. still debated. Interestingly, exercise below the intens- 1989; Freund et al. 1991; Montain et al. 1997) and ity threshold may also affect AVP, but conversely. decreases in plasma volume (Wade & Claybaugh Freund et al. (1991) showed reduced AVP after o 1980; Melin et al. 1980; Freund et al. 1987; Shoemaker 20 min of cycling at 25% V 2max, and attributed this et al. 1998; Roy et al. 2001b) have each been exclus- decrease to exercise-induced increases in thoracic ively implicated as the sole cause of increased AVP, blood volume and resultant negative feedback by significant evidence exists against either as indi- atrial and aortic baroreceptors. vidual phenomena (Beardwell et al. 1975; Wade & While the physiological causes describing exer- Claybaugh 1980; Convertino et al. 1983; Maresh et al. cise-induced AVP release are reasonably clear, the 1985; Shoemaker et al. 1998; Melin et al. 2001; Roy ultimate effect and importance of increased AVP et al. 2001a, 2001b). Rather, it is the combination remains obscure. In his review, Wade (1984) sug- of increased Posm and decreased plasma volume gested AVP secretion during exercise could affect that dictate the AVP response to exercise (Beardwell several physiological factors, including initiation of et al. 1975; Convertino et al. 1981; De Souza et al. antidiuresis, regulation of plasma volume, regula- 1989; Grant et al. 1996; Melin et al. 1997). tion and secretion of sweat and training-induced Other than these primary influences, exercise hypervolemia. His conclusion, however, was that affects several other characteristics that may subtly the physiological importance of magnified AVP was adjust the AVP response. By enhancing sodium unclear. Unfortunately, much of this conclusion reabsorption (and therefore increasing Posm) plasma remains true requiring further research for com- renin activity and aldosterone may stimulate AVP plete clarification. (Convertino et al. 1980b; Wade & Claybaugh 1980), although conflicting reports exist (Convertino et al. Endurance training 1981; Shoemaker et al. 1998). Decreased blood pres- sure (Wade & Claybaugh 1980), increased sym- As a single bout of appropriate exercise effectively pathetic nervous system activity (Beardwell et al. stimulates AVP acutely, repeated exercise bouts 1975; Convertino et al. 1981) and decreased renal (i.e. endurance training) may chronically alter the metabolism of AVP (Wade & Claybaugh 1980) may control of AVP, both at rest and during exercise. also be influential. Thus, exercise stress globally Studies employing repeated measures and parallel stimulates AVP through changes in Posm and plasma designs to evaluate training consistently demon- volume and modifies this response with aortic strate similar resting AVP concentrations regardless 490 chapter 32

of training status (Geyssant et al. 1981; Freund et al. 3.5 Pre-training AVP 1987, 1988). Exercise responses, however, are more 3.0 Post-training AVP equivocal; AVP reaction to exercise is either un-

) * changed (Convertino et al. 1983; Freund et al. 1987), –1 2.5 decreased (Melin et al. 1980; Mudambo et al. 1997; 2.0 Shoemaker et al. 1998) or increased (Geyssant et al. 1981) after training. These contrasting results may 1.5 be explained by differences in exercise protocol.

AVP IR (fmol·L IR AVP 1.0 Specifically, Convertino et al. (1983) demonstrated that AVP responses were dependent on whether 0.5 exercise was assigned as absolute or relative work- 0.0 loads. Because exercise elicits a lower osmotic stimu- Lifters Runners lus at any given absolute workload after training, the magnitude of AVP stimulation at a specific abso- Fig. 32.1 Resting arginine vasopressin (AVP) responses before and after 12 weeks of high intensity resistance lute exercise intensity should decrease. At a given training (lifters) or 12 weeks of endurance training relative work rate, however, training should have (runners). * Indicates a significant difference pre- to little influence. Alternately, endurance training post-training within a group. may change exercise responses and regulation of plasma volume (Convertino et al. 1980b; Shoemaker et al. 1998), modify body weight and/or hematocrit Resistance exercise and training (Melin et al. 1980) and improve osmotic sensitivity (Freund et al. 1987), any of which may influence Because weightlifting exceeds the intensity thres- AVP regulation. hold required to stimulate AVP, resistance exercise, To eliminate the confounding effects of exercise similar to acute endurance exercise, may induce an intensity, several studies have assessed training AVP response (Kraemer et al. 1999). In the only using non-exercise protocols with fluid regulatory study to test this theory, Kraemer et al. (1999) exam- demands. Freund et al. (1988) examined the results ined the effect of leg pressing at 80% of maximum of a water load (1% of lean body mass drunk in force production to fatigue in trained power lifters 2 min) in trained and untrained subjects. Despite and untrained subjects. In response to the high similar Posm, trained subjects experienced smaller intensity exercise, AVP was unchanged in either decrements in AVP and less diuresis than untrained group, as were other potential endocrine regulators subjects. They concluded that exercise training: (i) of AVP (aldosterone, atrial natriuretic peptide and altered the osmotic regulation of AVP by blunt- angiotensin II). The authors attributed this complete ing oropharyngeal inhibition; and (ii) led to over- lack of a fluid regulatory hormonal response to the all fluid conservation. These findings contradict protocol’s insignificant impact on plasma volume. Convertino et al. (1993), who detailed the effects of Unpublished data from our laboratory, however, 5 h of water immersion on runners, swimmers and indicate that chronic resistance training may alter untrained subjects. In that study, runners exhibited AVP regulation. Although Kraemer et al. (1999) larger decreases in AVP and greater diuresis than showed power lifters and untrained subjects had untrained subjects, leading the authors to suggest similar resting AVP, we observed a significant dry-land endurance training enhanced barorecep- increase in baseline AVP after 12 weeks of heavy tor sensitivity and/or reduced renal tubule sensit- resistance training (Fig. 32.1). Resistance training ivity to AVP. Thus, while we are confident that had no effect on aldosterone, cortisol, renin activity endurance training has little effect on resting AVP, or lactic acid at rest or in response to the exercise further research is necessary to determine if and challenge. Based on this enhanced resting AVP, we how chronic exercise modifies responses to hydro- have speculated that the increased fat free mass dynamic stress. resulting from resistance training may be related to arginine vasopressin 491

35 exaggerated tissue fluid retention, not just enhanced EU + NW *† + myofibrillar protein content. Needless to say, the 30 EU W dearth of research in this area affords several excit- HY + NW ) 25 HY + W ing opportunities to further examine this topic. –1 20 † fmol·L

Exercise and hydration status –3 15

As total body water is closely tied to AVP control, ( × 10 10 pre-exercise hydration status has strong influences Arginine vasopressin 5 on the concentration of AVP elicited by exercise. Dehydration is a powerful stimulant to AVP release 0 and increases the quantity of AVP observed with exercise (Brandenberger et al. 1989; Melin et al. 1997, 10 2001; Montain et al. 1997; Roy et al. 2001a); likewise, *† ζ initiation of exercise in a hyperhydrated state (via 8 water supplementation) (Wade & Claybaugh 1980; ζ Brandenberger et al. 1989) or plasma volume ex- 6 pansion (Grant et al. 1996; Roy et al. 2001b) limits Thirst the AVP concentration. AVP response to exercise 4 appears to be preserved regardless of total body water, rather, hydration state alters baseline AVP 2 and this modification persists throughout exercise (Grant et al. 1996; Montain et al. 1997; Roy et al. 0 Pre IP 2001a). The effects of hydration on AVP are due to the changes in plasma volume, blood pressure Fig. 32.2 Changes in arginine vasopressin (top panel) and and/or osmolality (Wade & Claybaugh 1980; Grant thirst (bottom panel) before and after 90 min of walking in et al. 1996; Melin et al. 1997; Montain et al. 1997; Roy the heat. * Indicates significant difference compared to all other values at the same time point. † Indicates significant et al. 2001b). difference compared to corresponding Pre value. ζ Although pre-exercise hydration state modifies Indicates significant difference compared to both EU the AVP concentration during exercise, a prolonged values at the same time point. EU, euhydrated; H, bout affords the opportunity to significantly alter hydrated; IP, immediate post-exercise; NW, no water hydration state during exercise. Rehydration dur- intake; Pre, pre-exercise; W, ad libitum water intake. (From Maresh et al. 2004b.) ing prolonged (3–4 h) low intensity exercise blunts the AVP response (Mudambo et al. 1997), even if subjects are initially hypohydrated (Brandenberger exercise hydration status (euhydrated or hypohy- et al. 1986, 1989). Concomitant rehydration with drated) and water intake during exercise (ad libitum water or an isotonic carbohydrate–electrolyte solu- or no fluid intake). Subjects dehydrated before tion is more effective in limiting AVP than pre- the appropriate trials by intermittently exercising exercise hyperhydration (Brandenberger et al. 1986, the previous day and ~ 17 h of subsequent water 1989). Combinations of pre- and during-exercise deprivation. Immediately post-exercise, no signific- rehydration are also potent inhibitors of AVP dur- ant differences existed in AVP, plasma osmolality, ing exercise (Melin et al. 1997). or plasma volume among the euhydrated + water, Recent data from our laboratory (Maresh et al. euhydrated + no water and hypohydrated + water 2004a) also highlight the relevance of hydration trials, confirming that hydration prior to and/or state in modifying AVP during exercise. During that rehydration during exercise inhibited the release of study, subjects completed four 90-min walking AVP during low intensity exercise in the heat. Addi- trials in the heat; exercise bouts differed in pre- tionally, as displayed in Fig. 32.2, changes in thirst 492 chapter 32

(and resultant drinking) mirrored AVP, a rela- Table 32.3 Challenges stimulating an arginine tionship previously suggested by several authors vasopressin (AVP) response. (Verney 1947; Andersson et al. 1980; Nose et al. 1988). Plasma SNS

Challenge Posm volume activity Modifying factors Exercise ↑↓ ↑ Hypertonic saline infusion ↑↑ – Menstrual phase and status differences Water deprivation ↑↓ – Water load ↓↑ – The vast majority of research reports that baseline Posture change ↑↓ ↑↓ – AVP is similar across menstrual phases (Spruce et al. 1985; Vokes et al. 1988; De Souza et al. 1989; Trigoso SNS, sympathetic nervous system. et al. 1996; Stachenfeld et al. 1999a, 1999b; Claybaugh et al. 2000; Calzone et al. 2001). In one of the only studies to document serial measures throughout the Although the magnitude and osmotic regulation menstrual cycle, however, Forsling et al. (1981) of the AVP response elicited by exercise is similar noted significant cyclical variations in resting AVP in the luteal phase, the follicular phase (De Souza (AVP peaked at ovulation and minimized during et al. 1989; Stachenfeld et al. 1999a) and amenorrhea menstruation). Figure 32.3 shows that studies col- (De Souza et al. 1989), the osmolar threshold for lecting data only on 2 specific days may have missed AVP release decreases during the luteal phase in these peaks and nadirs because samples were gener- response to hypertonic saline infusion (Spruce ally obtained at the mid-point of each phase, when et al. 1985; Vokes et al. 1988; Trigoso et al. 1996; the concentration of AVP is moderate. This conclu- Stachenfeld et al. 1999b, 2001), potentially reducing sion and its physiological significance, however, free water clearance and maintaining hypotonic remain to be established. plasma. In the majority of cases, sensitivity of the

Significant research has also compared females’ AVP to a unit change in Posm was unchanged across AVP response to a variety of exercise, osmotic and phase (Vokes et al. 1988; Trigoso et al. 1996; dehydrating stimuli between menstrual phases. Stachenfeld et al. 1999b, 2001), although reductions in the AVP sensitivity during the luteal phase have been observed (Spruce et al. 1985). Different results 2.0 10 observed with exercise vs. infusion challenges are most likely due to the stimuli they present. Table 1.6 8

) 32.3 shows that although both challenges increase –1 P , exercise reduces plasma volume and stimu- 1.2 6 osm lates sympathetic nervous activity, while hyper- 0.8 4 tonic saline infusion increases plasma volume and

AVP ( µ u·mL has little effect on sympathetic nervous activity.

0.4 2 Frequency of study Further research has examined the effect of dosing estrogen, progesterone or combinations of 0.0 0 both on AVP response to a stimulus. These studies 0 4 8 1216202428 Day of cycle are challenging because subjects’ normal hormonal patterns must be non-existent (i.e. post-menopausal Fig. 32.3 Arginine vasopressin (AVP) measures collected women), artificial and controllable (i.e. oral or intra- throughout the menstrual cycle in eight healthy, muscular contraceptive users) or pharmacologic- eumenorrheic women. Shaded areas indicate the number ally eliminated. Similar to luteal phase alterations, of studies examining resting AVP on a given day of the menstrual cycle and, hence, why significant differences estrogen administration reduces the threshold Posm in baseline AVP may have not been found. (Portions of required to induce an AVP response (Forsling et al. the figure redrawn from Forsling et al. 1981.) 1982; Stachenfeld et al. 1998, 1999b; Calzone et al. arginine vasopressin 493

2001; Stachenfeld & Keefe 2002), increasing AVP release was significantly reduced during preg- release for any given Posm (Forsling et al. 1981; nancy. Considering the similar hormonal milieus Spruce et al. 1985; Claybaugh et al. 2000; Calzone et of pregnancy and the luteal phase, this finding is not al. 2001; Stachenfeld et al. 2001) and preserves the surprising. slope of the Posm–AVP curve (i.e. AVP sensitivity) (Stachenfeld et al. 1998, 1999b; Calzone et al. 2001; Gender differences Stachenfeld & Keefe 2002). As it readily crosses the blood–brain barrier, estrogen’s influence is most Based on limited research, few conclusions about likely mediated by a direct effect on the central nerv- gender-related control of AVP have been reached. ous system (Stachenfeld et al. 1998, 1999b, Calzone Baseline AVP has been reported as similar between et al. 2001; Stachenfeld & Keefe 2002). Progesterone, genders (Stachenfeld et al. 2001) and increased in on the other hand, typically has minor, inconsistent men (Stachenfeld et al. 1996). Although exercise effects on AVP (Forsling et al. 1982; Stachenfeld et al. appears to elicit a similar response regardless of 1999b; Calzone et al. 2001). gender (Maresh et al. 1985; De Souza et al. 1989; If, as suggested by the previous studies, estrogen Stachenfeld et al. 1996), females have demonstrated mediates menstrual control of AVP, why is the greater AVP during recovery than males, poten- osmolar threshold for AVP release reduced during tially due to higher concentrations of endogenous the luteal phase (when endogenous estrogen is estrogen (Stachenfeld et al. 1996). In response to a high), but not the late follicular phase (when 12 mL·kg–1 lean body mass water load, AVP was endogenous estrogen is also high)? Several plaus- similar between genders, but differences in Posm ible hypotheses exist. Use of synthetic, exogenous led authors to suggest that women experienced a steroids in many drug studies is potentially con- reduced osmotic threshold for AVP release and/or founding; natural and artificial hormones have an enhanced AVP release compared to men. The different bioavailabilities and binding character- importance of this shift was questioned, however, istics (Stachenfeld et al. 1998; Calzone et al. 2001). as women experienced reduced renal responsive- Alternately, estrogen likely interacts with other ness to AVP (Claybaugh et al. 2000). A combina- ovarian hormones: the increased luteinizing hor- tion salt infusion–water load elicited a greater slope mone or follicle-stimulating hormone characteristic in the Posm–AVP curve (i.e. greater AVP sensitiv- of the late follicular phase may inhibit, or the ity to changes in Posm) in men, potentially due to enhanced progesterone of the luteal phase may gender differences in arterial resistance, renal sens- facilitate estrogen’s full effect on AVP. The possib- itivity, androgen concentration, plasma volume ility that data collection has been mistimed (see and/or cardiopulmonary baroreceptor sensitivity Fig. 32.3) also cannot be eliminated. Finally, although (Stachenfeld et al. 2001). These varied, sometimes improbable, estrogen may not drive menstrual conflicting results preclude firm conclusion and will alterations in AVP regulation, as cyclic fluctuations hopefully be clarified with future research. in AVP clearance (Forsling et al. 1981, 1982) and/or volemic status (Trigoso et al. 1996) could modify Aging AVP control. Examination of subjects from 52 to 95 years of age has shown resting concentrations of AVP are similar Pregnancy regardless of age (Helderman et al. 1978; Rowe et al. In one of the only studies examining AVP during 1982; Phillips et al. 1984, 1991, 1993; Bevilacqua et al. pregnancy, Davison et al. (1984) compared gravidas’ 1987), although confounding reports exist (Kirkland responses to several experimental challenges dur- et al. 1984; Davies et al. 1995). The response to AVP ing their third trimester and again 8–10 weeks challenges, however, is more diverse. Overnight postpartum. AVP responded appropriately to all fasting (Duggan et al. 1993; Faull et al. 1993) and 24 h challenges, but the osmolar threshold for AVP of water deprivation (Phillips et al. 1984), have 494 chapter 32

produced contrasting results; elderly AVP response sometimes overriding stimulatory changes in plasma was either similar to (Phillips et al. 1984; Duggan volume, plasma sodium and Posm (Segar & Moore et al. 1993) or lower than (Faull et al. 1993) younger 1968; Wittert et al. 1992). Conversely, central blood subjects, sometimes independent of Posm. Water volume decreases during exposure to high ambient loads up to 20 mL·kg–1 demonstrate that elderly temperatures due to peripheral vascular vasodila- subjects fail to diminish their AVP as promptly as tion, resulting in decreased baroreceptor stimulation younger subjects (Phillips et al. 1993; Davies et al. and enhanced AVP output. This AVP stimulation 1995). Finally, although AVP response to hypertonic has been shown after exposures up to 2 h at temper- saline infusion is at least maintained (Phillips et al. atures of 42–50°C (107.6–122.0°F) and exceeds the 1991) and may be enhanced (Hong et al. 1977; Davies AVP stimulus resulting from hypovolemia and et al. 1995) in the aged, their responses to a change in hyperosmolality associated with sweating (Segar & body position (from recumbency to standing) are Moore 1968; Convertino et al. 1980b; Melin et al. 1991). significantly less than young subjects (Rowe et al. 1982; Bevilacqua et al. 1987). Altitude Physiologically, these apparently conflicting results may be explained by an age-related diver- Unfortunately, the release and control of AVP gence in the sensitivity of the primary receptors with high altitude exposure is still not completely controlling AVP release. Based on several studies, understood. Studies show that resting measures authors have suggested that aging increases the sens- of AVP increase (Hackett et al. 1978; Porchet et al. itivity of AVP to an osmotic stimulus (Helderman 1984; Rostrup 1998), decrease (Ramirez et al. 1992; et al. 1978; Phillips et al. 1984; Bevilacqua et al. 1987; Bestle et al. 2002; Robach et al. 2002) or do not change Davies et al. 1995). Research on changes in body (Rostrup 1998; Robach et al. 2002; Maresh et al. position indicates that volume–pressure input, 2004b) in response to significant elevation (2900– however, is decreased in the elderly (Rowe et al. 4559 m), frequently disconnected from Posm. Adding 1982; Bevilacqua et al. 1987). Thus, aging may sensit- further confusion are findings that high altitude ize the osmoreceptor and desensitize the barore- natives have higher resting AVP than lowlanders ceptor (Rowe et al. 1982; Bevilacqua et al. 1987). (Ramirez et al. 1992, 1993, 1998). At altitude, hyper-

Additionally, oropharyngeal control of AVP may be tonic saline infusion still increases Posm, but has altered with aging (Phillips et al. 1991). Exactly how inconsistent effects on AVP (Ramirez et al. 1992, these considerations explain the dehydration prob- 1993, 1995; Bestle et al. 2002). Furthermore, although lems of the elderly is currently unknown; however, 30 min of moderate intensity cycling elicited a a generalized reduction in AVP secretion is not to similar AVP response at sea level and after 7 days blame (Kirkland et al. 1984; Phillips et al. 1984, 1993). exposure to 4350 m, AVP responses were magnified when identical exercise was repeated 2 days after Environmental influences return to sea level (Robach et al. 2002). Besides differences in protocol, AVP is difficult to evaluate at altitude because acclimatization dynam- Temperature ically changes physiological responses, expanding The effects of temperature on AVP control are prim- the importance of measurement timing. Recent arily mediated via changes in intrathoracic blood work by our laboratory (Maresh et al. 2004b) has volume and resultant baroreceptor stimulation. clarified the role of acclimatization by examining Cold-induced cutaneous vasoconstriction increases the effects of a 24-h water deprivation trial at sea central blood volume and results in central pres- level, after 2 days exposure to 4300 m and after sure–volume receptor activation, typically dimin- 20-days exposure to 4300 m. The primary findings ishing AVP and increasing urine output. This AVP of that study are shown in Fig. 32.4: progressive inhibition has been shown after exposures of 30– altitude acclimatization modifies the relationship

60 min at temperatures of 4–13°C (39.2–55.4°F), between Posm and AVP in a time-dependent fashion. arginine vasopressin 495

14 ing this change include increases in AVP as a com- SL: y = .0633x -17.451 pensatory mechanism to defend blood pressure in 12 2 = = R .122, p .094 the face of decreased catecholamines (Rostrup 10 1998), renal resistance to AVP (Ramirez et al. 1998) or altered control of AVP by P , blood pressure 8 osm and plasma volume (Bestle et al. 2002; Robach et al. 6 2002; Maresh et al. 2004b). 4

2 Hyperbaria

0 Those studies finding a typical hyperbaria-induced diuresis show hyperbaric exposures decrease rest- 14 AA: y = (3.4e-49)(x19.7087) ing AVP (Hong et al. 1977; Claybaugh et al. 1984, 12 2 = = 1987, 1992, 1997; Matsui et al. 1987; Takeuchi et al. ) R .146, p .065 –1 1995; Torii et al. 1997; Park et al. 1998). Unlike the 10 time-dependent response to high altitude exposure, 8 this phenomenon is preserved whether the expos- ure is as minor as 1 h at 3 ATA (atmosphere’s abso- 6 lute pressure) (Torii et al. 1997) or as intense as 14 4 days at 31 ATA (Claybaugh et al. 1984). Limited data Plamsa AVP (pg·mL 2 on AVP responses to stimulating challenges dur- ing hyperbaria confirms resting results. Although 0 a 1 L water load appropriately (i.e. similarly to 1 14 ATA) reduced AVP at 31 ATA (Tekeuchi et al. 1995), PA: y = (1.8e-90)(x36.3366) head-up tilt (31 ATA) (Matsui et al. 1987), 20 min of 12 2 = = R .263, p .012 exercise at 80% of maximum heart rate (46 ATA) 10 (Claybaugh et al. 1997) and maximal exercise (37 ATA) (Claybaugh et al. 1997) elicited reduced, or no, 8 AVP response. Minimized AVP at hyperbaria may 6 result from pressure-induced magnification of neg- ative intrathoracic pressure and resultant increases 4 in thoracic blood volume. Through its actions on 2 central baroreceptors, increased thoracic blood vol- 0 ume decreases AVP and promotes diuresis (Hong et al. 1977; Claybaugh et al. 1984, 1987; Tao et al. 1992; 276 280 284 288 292 296 300 304 308 312 316 Torii et al. 1997). Alternately, reduced AVP may be Plasma osmolality (mOsm·L–1) caused by altered AVP receptor responsiveness (Raymond et al. 1980; Claybaugh et al. 1997), hypo- Fig. 32.4 Best-fit curve estimations of the relationship osmolality (Matsui et al. 1987; Torii et al. 1997) or as between Posm and PAVP for sea level (SL), acute altitude exposure (2 days at 4300 m) (AA) and prolonged altitude a direct effect of increased pressure on the central exposure (20 days at 4300 m) (PA). (From Maresh et al. nervous system (Claybaugh et al. 1987, 1992; Matsui 2004a.) et al. 1987).

Additionally, if the relationship between P and osm Summary AVP is forced into a linear model, the osmotic threshold of AVP release increases after prolonged AVP plays an important regulatory role in water altitude exposure. Physiological hypotheses explain- conservation and the maintenance of body fluid 496 chapter 32

osmolality. Exercise stimulates AVP release through conclusions about gender-related control of AVP several mechanisms, but principally through a com- have been reached. Aging increases the sensitivity bination of increased Posm and decreased plasma of AVP to an osmotic stimulus. High ambient volume. Endurance training may alter control of temperatures tend to enhance AVP secretion. The AVP at rest and during exercise, but recent findings AVP response to high altitude depends on several suggest that resistance training also may alter AVP factors, but most assuredly on the length of alti- regulation. Pre-exercise hydration status has a tude exposure. Typically, AVP concentrations are strong impact on the plasma AVP response to exer- reduced with hyperbaric-induced diuresis. Several cise. The majority of research reports that baseline of these results require further clarification. AVP is similar across menstrual phases and few

References

Andersson, B., Olsson, K. & Rundgren, M. Calzone, W.L., Silva, C., Keefe, D.L. & of hypervolemia. Journal of Applied (1980) ADH in regulation of blood Stachenfeld, N.S. (2001) Progesterone Physiology 48, 657–664. osmolality and extracellular fluid does not alter osmotic regulation of Convertino, V.A., Keil, L.C., Bernauer, volume. Journal of Parenteral and Enteral AVP. American Journal of Physiology. E.M. & Greenleaf, J.E. (1981) Plasma Nutrition 4, 88–96. Regulatory, Integrative and Comparative volume, osmolality, vasopressin, and Arnauld, E. & du Pont, J. (1982) Physiology 281, R2011–R2020. renin activity during graded exercise in Vasopressin release and firing of Claybaugh, J.R., Hong, S.K., Matsui, N. man. Journal of Applied Physiology 50, supraoptic neurosecretory neurons et al. (1984) Responses of salt- and water- 123–128 during drinking in dehydrated monkey. regulating hormones during a saturation Convertino, V.A., Keil, L. C. & Greenleaf, Pflugers Archiv 394, 195–201. dive to 31 ATA (SEADRAGON IV). J.E. (1983) Plasma volume, renin, and Baylis, P.H. & Robertson, G.L. (1980) Undersea Biomedical Research 11, vasopressin responses to graded Vasopressin response to hypertonic 65–80. exercise after training. Journal of Applied saline infusion to assess posterior Claybaugh, J.R., Matsui, N., Hong, S.K. Physiology. Respiratory, Environmental, pituitary function. Journal of the Royal et al. (1987) Seadragon VI: a 7-day and Exercise Physiology 54, 508–514. Society of Medicine 73, 255–260. dry saturation dive at 31 ATA. III. Convertino, V.A., Tatro, D.L. & Rogan, Beardwell, C.G., Geelen, G., Palmer, H.M., Alterations in basal and circadian R.B. (1993) Renal and cardiovascular Roberts, D. & Salamonson, L. (1975) endocrinology. Undersea Biomedical responses to water immersion in Radioimmunoassay of plasma Research 14, 401–411. trained runners and swimmers. vasopressin in physiological and Claybaugh, J.R., Goldinger, J.M., Moon, European Journal of Applied Physiology pathological states in man. Journal of R.E. et al. (1992) Urinary vasopressin and 67, 507–512. Endocrinology 67, 189–202. aldosterone and plasma volume during Davies, I., O’Neill, P.A., McLean, K.A., Bestle, M.H., Olsen, N.V., Poulsen, T.D. a saturation dive to 450 m. Undersea Catania, J. & Bennett, D. (1995) Age- et al. (2002) Prolonged hypobaric Biomedical Research 19, 295–304. associated alterations in thirst and hypoxia attenuates vasopressin Claybaugh, J.R., Freund, B.J., Luther, G., arginine vasopressin in response to a secretion and renal response to Müller, K. & Bennett, P.B. (1997) Renal water or sodium load. Age and Ageing osmostimulation in men. Journal of and hormonal responses to exercise 24, 151–159. Applied Physiology 92, 1911–1922. in man at 46 and 37 atmospheres Davison, J.M., Gilmore, E.A., Dürr, J., Bevilacqua, M., Norbiato, G., Chebat, E. absolute pressure. Aviation, Space, and Robertson, G.L. & Lindheimer, M.D. et al. (1987) Osmotic and nonosmotic Environmental Medicine 68, 1038–1045. (1984) Altered osmotic thresholds for control of vasopressin release in the Claybaugh, J.R., Sato, A.K., Crosswhite, vasopressin secretion and thirst in elderly: effect of metoclopramide. L.K. & Hassell, L.H. (2000) Effects of human pregnancy. American Journal of Journal of Clinical Endocrinology and time of day, gender, and menstrual cycle Physiology. Renal, Fluid, and Electrolyte Metabolism 65, 1243–1247. phase on the human response to a water Physiology 246, F105–F109. Brandenberger, G., Candas, V., Follenius, load. American Journal of Physiology. De Souza, M.J., Maresh, C.M., Maguire, M., Libert, J.P. & Kahn, J.M. (1986) Regulatory, Integrative and Comparative M.S. et al. (1989). Menstrual status and Vascular fluid shifts and endocrine Physiology 279, R966–R973. plasma vasopressin, renin activity, and responses to exercise in the heat. Convertino, V.A., Brock, P.J., Keil, L.C., aldosterone exercise responses. Journal European Journal of Applied Physiology 55, Bernauer, E.M. & Greenleaf, J.E. of Applied Physiology 67, 736–743. 123–129. (1980a) Exercise training-induced Duggan, J., Kilfeather, S., Lightman, S.L. Brandenberger, G., Candas, V., Follenius, hypervolemia: role of plasma albumin, & O’Malley, K. (1993) The association of M. & Kahn, J.M. (1989) The influence renin, and vasopressin. Journal of Applied age with plasma arginine vasopressin of the initial state of hydration on Physiology 48, 665–669. and plasma osmolality. Age and Ageing endocrine responses to exercise in Convertino, V.A., Greenleaf, J.E. & 22, 332–336. the heat. European Journal of Applied Bernauer, E.M. (1980b) Role of thermal Faull, C.M., Holmes, C. & Baylis, P.H. Physiology 58, 674–679. and exercise factors in the mechanism (1993) Water balance in elderly people: arginine vasopressin 497

is there a efficiency of vasopressin? Age Kirkland, J., Lye, M., Goddard, C., Journal of Applied Physiology 76, and Ageing 22, 114–120. Vargas, E. & Davies, I. (1984) Plasma 320–327. Forsling, M.L., Akerlund, M. & Strömberg, argininevasopressin in dehydrated Melin, B., Koulmann, N., Jimenez, C. et al. P. (1981) Variations in plasma elderly patients. Clinical Endocrinology (2001) Comparison of passive heat or concentrations of vasopressin during the 20, 451–456. exercise-induced dehydration on renal menstrual cycle. Journal of Endocrinology Kraemer, W.J., Fleck, S.J., Maresh, C.M. et water and electrolyte excretion: the 89, 263–266. al. (1999) Acute hormonal responses to a hormonal involvement. European Journal Forsling, M.L., Strömberg, P. & Åkerlund, single bout of heavy resistance exercise of Applied Physiology 85, 250–258. M. (1982) Effect of ovarian steroids in trained power lifters and untrained Montain, S.J., Laird, J.E., Latzka, W.A. & on vasopressin secretion. Journal of males. Canadian Journal of Applied Sawka, M.N. (1997) Aldosterone and Endocrinology 95, 147–151. Physiology 24, 524–537. vasopressin responses in the heat: Freund, B.J., Claybaugh, J.R., Dice, M.S. Ludwig, M. (1998) Dendritic release of hydration level and exercise intensity & Hashiro, G.M. (1987) Hormonal and vasopressin abd oxytocin. Journal of effects. Medicine and Science in Sports and vascular fluid responses to maximal Neuroendocrinology 10, 881–895. Exercise 29, 661–668. exercise in trained and untrained McKenna, K. & Thompson, C. (1998) Mudambo, K.S.M.T., Coutie, W. & Rennie, males. Journal of Applied Physiology 63, Osmoregulation in clinical disorders of M.J. (1997) Plasma arginine vasopressin, 669–675. thirst and thirst appreciation. Clinical atrial natruiretic peptide and brain Freund, B.J., Claybaugh, J.R., Hashiro, Endocrinology 49, 139–152. natruiretic peptide responses to long- G.M. & Dice, M.S. (1988) Hormonal and Maresh, C.M., Wang, B.C. & Goetz, K.L. term field training in the heat: effects renal responses to water drinking in (1985) Plasma vasopressin, renin of fluid ingestion and acclimatization. moderately trained and untrained activity, and aldosterone responses to European Journal of Applied Physiology 75, humans. American Journal of Physiology. maximal exercise in active college 219–225. Regulatory, Integrative and Comparative females. European Journal of Applied Nose, H., Mack, G.W, Shi, X. & Nadel, E.R. Physiology 254, R417–R423. Physiology 54, 398–403. (1988) Role of osmolality and plasma Freund, B.J., Shizuru, E.M., Hashiro, G.M. Maresh, C.M., Gabaree-Boulant, C.L., volume during rehydration in humans. & Claybaugh, J.R. (1991) Hormonal, Armstrong, L.E. et al. (2004a) Effect of Journal of Applied Physiology 65, electrolyte, and renal responses to hydration status on thirst, drinking, and 325–331. exercise are intensity dependent. Journal related hormonal responses during low- Park, Y.S., Claybaugh, J.R., Shiraki, K. & of Applied Physiology 70, 900–906. intensity exercise in the heat. Journal of Mohri, M. (1998) Renal function in Geyssant, A., Geelen, G., Denis, C. et al. Applied Physiology 93, 39–44. hyperbaric environment. Applied Human (1981) Plasma vasopressin, renin Maresh, C.M., Kraemer, W.J., Judelson, Science 17, 1–8. activity,and aldosterone: effect of D.A. et al. (2004b) Effects of high altitude Phillips, P.A., Rolls, B.J., Ledingham, exercise and training. European Journal and water deprivation on arginine J.G.G. et al. (1984) Reduced thirst after of Applied Physiology 46, 21–30. vasopressin release in men. American water deprivation in healthy elderly Grant, S.M., Green, H.J., Phillips, S.M., Journal of Physiology: Endocrinology and men. New England Journal of Medicine Enns, D.L. & Sutton, J.R. (1996) Fluid Metabolism 286, E20–E24. 311, 753–759. and electrolyte hormonal responses Marples, D. (2000) Water channels: Phillips, P.A., Bretherton, M., Johnston, to exercise and acute plasma volume who needs them anyway? Lancet 355, C.I. & Gray, L. (1991) Reduced osmotic expansion. Journal of Applied Physiology 1571–1572. thirst in healthy elderly men. American 81, 2386–2392. Matsui, N., Claybaugh, J.R., Tamura, Y. Journal of Physiology. Regulatory, Hackett, P.H., Forsling, M.L., Milledge, J. & et al. (1987) Seadragon VI: a 7-day Integrative and Comparative Physiology Rennie, D. (1978) Release of vasopressin dry saturation dive at 31 ATA. VI. 261, R166–R171. in man at altitude. Hormone and Hyperbaria enhances renin but Phillips, P.A., Bretherton, M., Risvanis, J. Metabolism Research 10, 571. eliminates ADH responses to head-up et al. (1993) Effects of drinking on thirst Helderman, J.H., Vestal, R.E., Rowe, J.W. tilt. Undersea Biomedical Research 14, and vasopressin in dehydrated elderly et al. (1978) The response of arginine 437–447. men. American Journal of Physiology. vasopressin to intravenous ethanol and Melin, B., Eclache, J.P., Geelen, G. et al. Regulatory, Integrative and Comparative hypertonic saline in man: the impact of (1980) Plasma AVP, neurophysin, Physiology 264, R877–R881. aging. Journal of Gerontology 33, 39–47. renin activity, and aldosterone during Porchet, M., Contat, H., Waeber, B., Hong, S.K., Claybaugh, J.R., Frattali, V. submaximal exercise performed until Nussberger, J. & Brunner, H.R. (1984) et al. (1977) Hana Kai II: a 17-day dry exhaustion in trained and untrained Response of plasma arginine saturation dive at 18.6 ATA. III. Body men. European Journal of Applied vasopressin levels to rapid changes in fluid balance. Undersea Biomedical Physiology 44, 141–151. altitude. Clinical Physiology 4, 435–438. Research 4, 247–265. Melin, B., Curé, M., Jimenez, C. et al. (1991) Ramirez, G., Hammond, M., Agosti, S.J. Hurbin, A., Boisin-Agasse, E., Orcel, H. & Plasma atrial natruiretic peptide and et al. (1992) Effects of hypoxemia at sea

Rabie A. (1998) The V1a and V1b, but not vasopressin during thermal dehydration level and high altitude on sodium the V2, vasopressin receptor genes are in supine position. Acta Physiologica excretion and hormonal levels. Aviation, expressed in the supraoptic nucleus Scandinavica 141, 227–230. Space, and Environmental Medicine 63, of the rat hypothalamus, and the Melin, B., Jimenez, C., Svourey, G. et al. 891–898. transcripts are essentially colocalized (1997) Effects of hydration state on Ramirez, G., Bittle, P.A., Agosti, S.J., Dietz, in the vasopressinergic magnocellular hormonal and renal responses during J. & Colice, G.L. (1993) Salt loading test neurons. Endocrinology 139, 4701–4707. moderate exercise in the heat. European in a population adapted to moderately 498 chapter 32

high altitude living (3000 m). Aviation, J.E. & Ojeda, S.R., eds.). Oxford food and water intake. In: The Space, and Environmental Medicine 64, University Press, New York: 106–117. Hypothalmus (Haymaker, W., Anderson, 831–838. Segar, W.E. & Moore, W.W. (1968) The E. & Nauta, W.J.H., eds.). Charles C. Ramirez, G., Pineda, D.O., Bittle, P.A. et al. regulation of antidiuretic hormone Thomas, Springfield, IL: 524–605. (1995) Salt excretory capacity in natives release in man. I. Effects of change in Takeuchi, H., Mohri, M., Shiraki, K. et al. adapted to moderate high altitude living position and ambient temperature on (1995) Diurnal renal responses in man after acute mobilization to sea level. blood ADH levels. Journal of Clinical to water loading at sea level and 31 ATM Aviation, Space, and Environmental Investigation 47, 2143–2151. abs. Undersea and Hyperbaric Medicine 22, Medicine 66, 1063–1070. Shoemaker, J.K., Green, H.J., Ball-Burnett, 61–71. Ramirez, G., Pineda, D., Bittle, P.A. et al. M. & Grant, S. (1998) Relationships Tao, H.Y., Chen, H.J., Zhang, H., Guo, J. (1998) Partial renal resistance to arginine between fluid and electrolyte hormones & Rong, F.K. (1992) Urinary ANP, vasopressin as an adaptation to high and plasma volume during exercise with ADH, and electrolyte excretion during altitude living. Aviation, Space, and training and detraining. Medicine and saturation-excursion diving to pressures Environmental Medicine 69, 58–65. Science in Sports and Exercise 30, 497–505. equivalent to 250 and 300 m. Undersea Raymond, L.W., Raymond, N.S., Frattali, Spruce, B.A., Baylis, P.H., Burd, J. & Biomedical Research 19, 159–169. V.P. et al. (1980) Is the weight loss of Watson, M.J. (1985) Variation in Torii, R., Sagawa, S., Wada, F. et al. (1997) hyperbaric habitation a disorder of osmoregulation of arginine vasopressin Mechanism for changes in vasopressin osmoregulation? Aviation, Space, and during the human menstrual cycle. during acute exposure at 3 ATM abs Environmental Medicine 51, 397–401. Clinical Endocrinology 22, 37–42. air. American Journal of Physiology. Reid, I.A. (1984) Actions of angiotensin Stachenfeld, N.S. & Keefe, D.L. (2002) Regulatory, Integrative and Comparative II on the brain: mechanisms and Estrogen effects on osmotic regulation of Physiology 273, R259–R264. physiological role. American Journal AVP and fluid balance. American Journal Trigoso, W.F., Wesly, J.M., Meranda, D.L. of Physiology: Renal Fluid Electolyte of Physiology: Endocrinology and & Shenker, Y. (1996) Vasopressin and Physiology 246, F533–543. Metabolism 283, E711–E721. atrial natruiretic hormone response to Robach, P., Lafforgue, E., Olsen, N.V. et al. Stachenfeld, N.S., Gleim, G.W., Zabetakis, hypertonic saline during the follicular (2002) Recovery of plasma volume after P.M. & Nicholas, J.A. (1996) Fluid and luteal phases of the menstrual cycle. 1 week of exposure at 4350 m. Pflugers balance and renal responses following Human Reproduction 11, 2392–2395. Archiv 444, 821–828. dehydrating exercise in well-trained Verney, E.B. (1947) The antidiuretic Rostrup, M. (1998) Catecholamines, men and women. European Journal of hormone and factors which determine hypoxia and high altitude. Acta Applied Physiology 72, 468–477. its release. Proceedings of the Royal Society Physiologica Scandinavica 162, 389–399. Stachenfeld, N.S., DiPietro, L., Palter, S.F. of London 135, 25–106. Rowe, J.W., Minaker, K.L., Sparrow, D. & Nadel, E.R. (1998) Estrogen influences Vokes, T.J., Weiss, N.M., Schreiber, J., & Robertson, G.L. (1982) Age-related osmotic secretion of AVP and body Gaskill, M.B. & Robertson, G.L. failure of volume–pressure-mediated water balance in postmenopausal (1988) Osmoregulation of thirst and vasopressin release. Journal of Clinical women. American Journal of Physiology. vasopressin during normal menstrual Endocrinology and Metabolism 54, Regulatory, Integrative and Comparative cycle. American Journal of Physiology. 661–664. Physiology 274, R187–R195. Regulatory, Integrative and Comparative Roy, B.D., Green, H.J. & Burnett, M. Stachenfeld, N.S., DiPietro, L., Kokoszka, Physiology 254, R641–R647. (2001a) Prolonged exercise following C.A. et al. (1999a) Physiological Wade, C.E. (1984) Response, regulation, diuretic-induced hypohydration effects variability of fluid-regulation hormones and actions of vasopressin during on fluid and electrolyte hormones. in young women. Journal of Applied exercise: a review. Medicine and Science Hormone and Metabolism Research 33, Physiology 86, 1092–1096. in Sports and Exercise 16, 506–511. 540–547. Stachenfeld, N.S., Silva, C., Keefe, D.L., Wade, C.E. & Claybaugh, J.R. (1980) Roy, B.D., Green, H.J., Grant, S.M. & Kokoszka, C.A. & Nadel, E.R. (1999b) Plasma renin activity, vasopressin Tarnopolsky, M.A. (2001b) Acute Effects of oral contraceptives on body concentration, and urinary excretory plasma volume expansion in the fluid regulation. Journal of Applied responses to exercise in men. Journal of untrained alters the hormonal response Physiology 87, 1016–1025. Applied Physiology 49, 930–936. to prolonged moderate-intensity Stachenfeld, N.S., Splenser, A.E., Calzone, Wittert, G.A., Or, H.K., Livesey, J.H. et al. exercise. Hormone and Metabolism W.L., Taylor, M.P. & Keefe, D.L. (2001) (1992) Vasopressin, corticotrophin- Research 33, 238–245. Sex differences in osmotic regulation of releasing factor, and pituitary adrenal Sampson, W.K. (1992) The posterior AVP and renal sodium handling. Journal responses to acute cold stress in normal pituitary and water metabolism. In: of Applied Physiology 91, 1893–1901. humans. Journal of Clinical Endocrinology Textbook of Endocrine Physiology (Griffin, Stevenson, J.A.F. (1969) Neural control of and Metabolism 75, 750–755. Chapter 33

Human Endocrine Responses to Exercise–Cold Stress

JOHN W. CASTELLANI AND DAVID W. DEGROOT

Introduction and cutaneous blood flow is decreased, limiting heat loss to the environment. These physiological The combination of cold exposure and exercise responses are mediated by the sympathetic nervous performed in this environment elicits profound system (SNS) in order to defend against a drop in physiological responses. These include increases body temperature. Plasma norepinephrine (NE) in metabolic heat production and vasoconstriction concentrations are used as an acute marker of SNS in order to maintain body temperature, changes in activity since NE is released from peripheral nerve fluid balance and changes in substrate mobilization endings during acute cold exposure. NE exerts its in order to fuel increased metabolic activity. Many thermoregulatory effects via β-adrenergic receptors of these physiological responses are associated with on skeletal muscle (metabolic heat production) and endocrine secretion and concentration changes. It is α-adrenergic receptors on smooth muscle (vasocon- acknowledged that differences in plasma levels as a striction). Plasma NE increases two to sixfold dur- result of cold exposure could be due to secretion, ing sedentary cold exposure (Wilkerson et al. 1974; clearance and volume of distribution that cannot be Johnson et al. 1977; O’Malley et al. 1984; Wang et al. differentiated in these studies. Most of the studies 1987; Weiß et al. 1988; Frank et al. 1997; Armstrong examining the role of cold exposure on hormone 1998; Castellani et al. 1998, 1999a, 1999b) and also responses have been completed during resting increases urinary NE excretion (Arnett & Watts exposure and these are still relatively few in num- 1960; Lamke et al. 1972), a 24-h marker of SNS activ- ber. The interaction of exercise and cold on the ity. The weighting/contribution factor of core and endocrine system is even less studied. The primary skin temperature for the NE response to resting cold purpose of this chapter is to survey the endocrine exposure is ~ 2 : 1, that is ~ 67% of the increase in NE responses to exercise–cold stress in humans. Animal is due to decreases in core temperature and 33% is studies will not be considered. Sedentary cold due to the fall in skin temperature (Frank et al. 1999). exposure responses are also reviewed to provide NE concentrations increase during exercise–cold background information and for use as a point of stress (Castellani et al. 2001), not surprisingly, but comparison to exercise studies. Table 33.1 presents the responses between cold and temperate environ- the hormones that will be discussed and presents a ments are likely dependent on whether core temper- general overview of the changes caused by resting ature declines. When high intensity exercise (~ 60% o and exercise–cold stress. V 2max) was performed for 2 h, plasma NE was not different between a dry, 15°C environment and a Catecholamines wet, windy, 5°C environment (Weller et al. 1997a). However, as soon as the intensity was lowered to o When people are exposed to the cold, metabolic heat less than 30% V 2max, and core temperature subse- production increases via shivering thermogenesis quently fell, NE concentrations were 240% higher in 499 500 chapter 33

Table 33.1 Summary of hormone responses during exercise–cold stress.

Response to resting Response to Hormone Physiological effects cold exposure exercise–cold stress Other observations

Catecholamines ↑ Cutaneous vasoconstriction ↑ 2–6-fold ↑ At exercise intensities ↑ o Heat production below 40% V 2max ↑ Glycogenolysis compared to temperate; ↑ Lipolysis ↑ Epinephrine during long duration exercise (> 6 h) ↑ ↓ ↑ T3 & T4 Metabolism vasodilator In 30 min to Levels after 6–9 h at o 3-h exposures 30% V 2max Plasma renin Converts angiotensinogen ↓ Or no change ↓ Following graded activity to angiotensin I exhaustive exercise Arginine ↑ Water conservation ↓ In water or air No change with graded vasopressin exercise Atrial natriuretic ↓ Sodium reabsorption No change ↑ After graded exercise peptide to exhaustion Cortisol Catabolic hormone; ↑ Or no change ↑ After 5-h cold–wet Responses may be o promotes lipolysis exposure at 30% V 2max; related to time of vasoconstriction ↓ After 1-h swimming at day & intensity- o 68% V 2max duration interaction Insulin Anabolic hormone; No change when ↓ With cold-water promotes glucose uptake core temperature exercise but not different does not fall; from temperate; no ↓ when core change during cold-air temperature exercise ↓ by 1°C Glucagon Catabolic hormone; No change No change ↑ glycogenolysis ↑ gluconeogenesis ↑ lipolysis β Growth Anabolic hormone; No change No change after 1-h 2-Adrenergic hormone ↑ free fatty acid mobilization swimming at 68% stimulation may ↑ o gluconeogenesis V 2max suppress GH levels; released in pulsatile manner Prolactin Lactation ↓ During and after Small increase with Possible marker of exposure exercise–cold stress cold acclimation but lower than exercise–heat stress Testosterone Anabolic hormone; No change Variable responses spermatogenesis depending on exercise stress Luteinizing Promotes testosterone estrogen Chronic exposure No effect from acute hormone & progesterone secretion to cold lowers exercise basal values

T3, triiodothyronine; T4, thyroxine. hormones and cold stress 501

the cold environment. However, if the exercise Plasma EPI concentrations during exercise–cold o intensity is relatively high (60% V 2max) but per- stress, compared to temperate conditions, are elev- formed over an extended duration (4 h) that causes ated if core temperature falls, similar to plasma NE. core temperature to fall, plasma NE concentrations EPI levels were elevated after 4–6 h of cold–wet are 45% higher in a cold–wet environment. Galbo exposure, compared to a 15°C dry condition when et al. (1979) found that swimming in 21°C water for the rectal temperature difference was 0.4°C (Weller 1 h (rectal temperature decreased by 0.8°C) increased et al. 1997b). Likewise, Galbo et al. (1979) observed NE concentrations 87% above that compared to 71% higher EPI levels after swimming in 21°C water swimming in 27°C. As well, when both core and versus 27°C. On the other hand, if core temperat- skin temperatures are low during submaximal and ures are not different between cold and temperate maximal exercise, absolute NE responses are the environments, EPI concentrations are the same greatest (Bergh et al. 1979). (Weller et al. 1997a, 1997b) or lower (Parkin et al. The interaction between skin temperature and 1999). exercise intensity may also influence plasma NE It is important for the reader to be aware that cold concentrations during exercise–cold stress when acclimatization might influence the NE response there are no differences in core temperature. Plasma to exercise–cold stress, although no studies have NE is doubled in 5°C air (versus 21°C) during cycle directly tested this. However, there have been ergometry at 50 W but not at 150 W (Stevens et al. several studies that have examined the effect of 1987). Furthermore, Weller et al. (1997b) found that cold acclimation on NE levels during resting cold when exercise initially begins at a low intensity (30% exposure. o V 2max) in a cold–wet environment, plasma NE There are three distinct types of cold acclimation: concentrations are three-times higher after 2 h even habituation, metabolic and insulative acclimation though rectal temperature is not different compared (Young 1996). Cold habituation, characterized by to dry–temperate conditions. Skin temperatures in blunted shivering and vasoconstrictor responses, these low intensity studies were 5–8°C lower in the typically occurs after cold exposures not severe cold. However, when cycle exercise is performed at enough to elicit falls in body core temperature. o 50–100% V 2max, there are no differences in NE con- Metabolic acclimatization is defined as a higher centrations between temperate and cold environ- metabolic activity. This has been observed in one ments (Quirion et al. 1989; Anderson & Hickey 1994). study and occurred after moderate cold exposure The effect of cold exposure on epinephrine (EPI) for 6 weeks. Insulative acclimatization is induced by is less clear. In contrast to NE, plasma EPI responses repeatedly lowering the body core temperature, do not appear to increase during resting cold-water which causes greater peripheral vasoconstriction, immersion (Weiß et al. 1988), cold-saline infusion lower skin temperatures and perhaps a shift in the (Frank et al. 1997) and cold-air exposure (Wang et al. onset of shivering thermogenesis so that shivering 1987; Armstrong 1998). However, Frank et al. (2002) does not begin until a lower mean body tempera- found that EPI levels increased when the core tem- ture is reached. When multiple cold-air exposures perature decreased by 1.0°C using cold-saline infu- were used to induce cold habituation (with little sion (skin temperature remained normal). That fall –0.5°C in core temperature), plasma NE values study did not collect peripheral venous (ante- declined during standardized cold air tests (Hesslink cubital) blood but instead sampled central venous et al. 1992; Leppäluoto et al. 2001). On the other blood and suggested this source is closer to the site hand, when 25 cold-water immersions (18°C water) of EPI release and is not affected by clearance. They were used for acclimating subjects (causing sig- suggest that the lack of an EPI response in other nificant falls in core temperature −1°C) that led to studies is due to the sampling methodology. If EPI is an insulative acclimation (lower core and skin tem- increased during cold exposure, it likely would peratures), plasma NE was significantly higher after increase shivering thermogenesis and fat mobiliza- acclimation, compared to preacclimation, during a tion via β-adrenergic receptors. standardized cold-air exposure (Young et al. 1986). 502 chapter 33

Thyroid hormones since these values were not reported. The elevated T3 and T4 following long duration exercise–cold The thyroid hormones, triiodothyronine (T3) and stress is more likely due to the exercise stimulus per thyroxine (T4), are very important for maintaining se, rather than the cold exposure, since acute cold basal metabolism and thermogenesis. The caloro- exposure has little effect on these hormones. genic effect of these hormones is believed to be Since cold habituation leads to blunted shivering caused by increasing energy consumption of the and higher skin temperatures, thyroid hormones sodium–potassium pump. Also, thyroid hormones could also change since they are involved in ther- have been shown to be vasodilators. Thus, these mogenic and vasodilatory responses. Hesslink et al. hormones may affect peripheral heat loss during (1992) exposed subjects to 4.4°C air for 30 min twice cold exposure. a day for 8 weeks (80 total exposures). They also

Sedentary cold-air exposure (4–10°C) from 30 min supplemented a group of subjects with T3 to arti- to 3 h has no effect on plasma thyroid-stimulating ficially suppress TSH and T4 levels in order to deter- hormone (TSH), T3 and T4 levels (Hershman et al. mine the relative importance of these hormones 1970; Wilson et al. 1970; Nagata et al. 1976; Tuomisto for inducing cold acclimation. The repeated cold et al. 1976; Leppäluoto et al. 1988). In contrast, exposure program did indeed induce habituation several authors (Golstein-Golaire et al. 1970; as demonstrated by reductions in oxygen uptake, O’Malley et al. 1984) have noted increases in these mean arterial pressure and plasma NE values sub- hormones after short-term resting cold exposure sequent to acute cold exposure. No differences whereas others (Solter & Misjak 1989) found that were observed between the supplemental T3 group TSH, total T3, reverse T3 and total T4 were all lower (low TSH and T4) and the non-supplemented group following an 8-h work day in a cold environment. and there were no changes in thyroid hormone con- Short-term cold stress should not bring about large centrations. Similarly, Leppäluoto et al. (2001) increases or decreases in plasma concentrations exposed subjects for 11 straight days for 2 h·day–1 in because most of the hormone is bound to plasma 10°C air and found no changes in T3, T4 and TSH, proteins that provides a substantial reservoir of even though skin temperatures were higher after thyroid hormone (Goodman 1994). acclimation. Savourey et al. (1994) studied eight men Only a few studies have examined thyroid hor- who underwent a 1°C cold-air test for 2 h before and mone responses during controlled exercise–cold after cold acclimation. Acclimation was induced by stress experiments. The response pattern could be 20 ice-water immersions to the thigh over a 1-month related to a number of factors including exercise period and resulted in lower rectal temperatures duration or intensity. T4 is elevated following 6–9 h during cold-air exposure after acclimation, com- of cold–wet exposure during low intensity exercise pared to preacclimation. T3, T4 and TSH were not (Dulac et al. 1987; Castellani et al. 2002) whereas different after the acclimation period. These studies

T3 responses did not change or increased. Thyroid demonstrate that thyroid hormones do not have a hormones likely increased during prolonged exer- role in cold acclimation induced over a short-time cise–cold stress to support lipid metabolism, since it period. amplifies β-adrenergic responses of the SNS. Core The pattern observed in fairly short acclima- temperature was not a likely mediator of this effect tion studies is somewhat different than that seen in since it reached only 36.9°C in one study (Castellani personnel over-wintering in Antarctica for 8–12 et al. 2002). TSH was not elevated during these low months. Reed et al. (1986, 1988, 1990) characterized intensity exercise studies, but increased by 90% dur- that Antarctica residence leads to an elevated TSH ing a 30-min moderate intensity performance swim response subsequent to thyroid-releasing hormone in 20°C water (Deligiannis et al. 1993). Likewise, stimulation, lower T3 levels and no changes in T4. free T4 levels increased by 46% during cold-water Sawhney et al. (1995) related lower T3 values to the swimming. Whether this effect was a result of cent- period of the Antarctic summer when physical ral (core) or peripheral (skin) input is not known activity levels are high and observed higher T3 in the hormones and cold stress 503

winter, when is was extremely cold and dark. lated by decreased pressure (volume) and decreased Interestingly, if personnel from Antarctica are given sodium and chloride flux in the kidney. Higher supplemental T4, declines in cognitive performance angiotensin and aldosterone levels increase sodium are attenuated (Reed et al. 2001). Likewise feelings of and water retention. ANP is released from right- fatigue and confusion are lessened in the supple- atrium cardiac myocytes in response to high central mental group. The reader should be aware that blood pressure/volume and causes greater sodium Antarctic sojourners are exposed to a multitude of and water excretion from the kidney. AVP is factors including low ambient temperature, extreme released from the posterior pituitary in response to light conditions, highly variable physical activity increased plasma osmolality as well as decreases in levels, high electromagnetic radiation and isolation. blood volume sensed by high-pressure (carotid) and Thus, it is difficult from these studies to discern if low-pressure (right atrium) receptors. AVP directly the cold per se has an effect on thyroid levels or if it increases water reabsorption in the collecting duct is due to the combination of all these environmental of the kidney. factors. Exposure to cold temperatures lowers total body Thyroid status is classically linked to cold sensit- water levels via several mechanisms including cold- ivity, i.e. people with thyroid insufficiency typic- induced diuresis (CID), sweating, respiratory water ally complain about feeling cold (Larsen et al. 1998). loss, blunted thirst, poor water availability and con-

Nagashima et al. (2002) recently showed that T4 scious under-drinking (O’Brien et al. 1998). These concentrations were 29% lower in women who com- body water losses are associated with changes in plained about feeling cold, compared to matched plasma volume and electrolyte concentrations that controls. Associated with this increased cold feel- can have a profound effect on fluid regulatory hor- ings were lower finger temperatures and metabolic mones. Cold exposure also increases central blood rates. However, the relative importance of thyroid volume and right atrial pressure. Exercise causes hormones during acute cold exposure in humans profound changes in plasma volume and electroly- is still not well understood. Surprisingly, there are tes. However, relatively few studies have examined little data (one subject, Thompson et al. 1971) on fluid regulatory hormones during rest in cold envir- acute thermoregulatory responses during cold ex- onments, and only one study has examined these posure in people who are hypothyroid. After T4 responses during exercise. administration, this subject exhibited a higher oxy- The most commonly studied effect of cold expos- gen consumption, but a blunted fall in skin temper- ure on body fluid balance is CID. Over 50 years ago, ature, so that the core temperature response was Bader et al. (1952) suggested that CID is caused by a about the same after 2 h in 10°C air. Thus, thyroid fall in AVP levels, subsequent to higher central hormones appear to have little effect on thermal blood volumes. Administering pitressin, an AVP balance as it simultaneously increases heat produc- analog, abolished the higher urine flows observed tion and heat loss. in the cold, without affecting the glomerular filtration rate, although this has not always been Fluid regulation observed (Lennquist 1972). If cold exposure causes an essential water diuresis then plasma tonicity Fluid balance is regulated by several hormones, should go up, but it does not. A series of studies including renin–angiotensin–aldosterone, atrial nat- (Lennquist et al. 1974; Wallenberg & Granberg 1974; riuretic peptide (ANP) and arginine vasopressin Atterhog et al. 1975; Knight & Horvath 1985; Deuster (AVP). These hormones monitor blood volume and et al. 1989) demonstrated that urine electrolyte plasma osmolality in order to precisely regulate excretion is also elevated with cold exposure sug- water and electrolyte losses to maintain normal gesting that CID is due to an isosmotic fluid loss. levels. Angiotensin and aldosterone are regulated What is not known is whether an elevated renal by the enzyme renin, which is released from the solute excretion (primarily sodium) leads to water juxtaglomerular cells in the kidney. Renin is regu- loss, or if cold increases both sodium and water 504 chapter 33

excretion via mechanisms independent of each concentrations are blunted when plasma levels are other. Some evidence supports that salt losses are highest in the morning. primary in CID. When NaCl supplementation was The effects of resting cold exposure on plasma used during a 48-h field exercise (Rogers et al. 1964) cortisol concentrations are equivocal as both in- in extreme cold, 30% less urine was secreted com- creases (Suzuki et al. 1967; Wilkerson et al. 1974; pared to a non-NaCl group even though both groups Kauppinen et al. 1989; Hennig et al. 1993; Tikuisis drank the same amount of water. et al. 1999) and no change (Golstein-Golaire et al. Resting studies that have examined fluid- 1970; Wilson et al. 1970; Ohno et al. 1987; Wittert et al. regulatory hormone responses during cold exposure 1992; Frank et al. 1997; Marino et al. 1998) have been (Segar & Moore 1968; Hiramatsu et al. 1984; Hassi observed. et al. 1991; Wittert et al. 1992; Hynynen et al. 1993; Exercise–cold stress studies, as with resting stud- Nakamitsu et al. 1994; Jansky et al. 1996; Arjamaa ies, show both increases and decreases. Potential et al. 1999, 2001; Sramek et al. 2000) commonly find reasons include the exercise duration and the time that plasma renin activity either does not change or of day. Galbo et al. (1979) studied swimming exer- falls, plasma aldosterone and ANP are not changed cise, whereas McMurray et al. (1994) and Castellani and AVP levels decrease. Urinary sodium excretion et al. (2002) used cycle ergometry and walking, was elevated in these studies. Cold-induced natriur- respectively. Only one study corrected for changes esis was not mediated via changes in aldosterone or in plasma volume and reported the use of standard- ANP, but possibly through the paracrine influence ized blood draws controlling for factors such as arm of urodilatin, an ANP-like compound secreted from position and posture (Castellani et al. 2002). Differ- the kidney (Nakamitsu et al. 1994; Bestle et al. 1999). ences in the change in core temperature between Lower AVP concentrations lead to increased water Galbo et al. (1979) (–0.8°C) and McMurray (1994) excretion, secondary to increases in central blood (0°C) could explain why cortisol declined in the volume and right atrial pressure. swimming study but not following cycling. How- Only one study has examined fluid-regulatory ever, Castellani et al. (2002) also observed a fall in hormones during exercise–cold stress. Therminarias core temperature (0.2–0.5°C) and found that exer- et al. (1992) found lower plasma renin activity and cise–cold stress increased cortisol concentrations. higher ANP levels during graded exercise to Perhaps the long duration cold exposure (~ 5 h) exhaustion in 10°C air, compared to 30°C air. No combined with moderate intensity exercise (50% o differences were observed for AVP. The higher V 2max) leads to a greater stress response compared ANP and lower plasma renin activity levels in 10°C to high intensity, short duration exercise bouts air were due to greater cardiac filling caused by a (Galbo et al. 1979). Pandolf et al. (1992) also did not central redistribution of blood volume. find any changes in cortisol after exercising in cold o water for 50 min at 50% V 2max. Interestingly, in Cortisol Castellani et al. (2002), cortisol increases occurred in the absence of an ACTH response suggesting that Plasma cortisol modulates many physiological pro- cortisol secretion during prolonged exercise–cold cesses that are important for responding to cold stress is ACTH-independent. exposure. These include increasing resting energy Time of day may be the most important factor in expenditure, inhibiting vasodilation, increasing free determining whether plasma cortisol changes as a fatty acid availability for substrate utilization, and result of exercise–cold stress. Castellani et al. (2002) functioning of the SNS. Cortisol secretion is regu- studied their subjects in the late afternoon/early lated by adrenocorticotrophic hormone (ACTH), evening when cortisol levels are typically lower and which is secreted by the anterior pituitary. Plasma stress responses are discriminated easier, whereas cortisol concentrations are lowest in the evening other studies exercised their subjects in the morn- hours and highest just before awakening. This point ing, when cortisol levels are high. Also, if exercise is important as stress-related changes in cortisol studies are initiated in the morning and are fairly hormones and cold stress 505

long and not intense (Ainslie et al. 2002), lower cor- insulin was significantly lower at baseline, but tisol concentrations post-cold exposure could just exercise–cold stress caused no further decline in reflect the normal diurnal rhythm. Utilization of serum insulin levels after fasting whereas insulin proper control groups will allow for correct inter- fell 68% in the non-fasting trial (Weller et al. 1998). pretation of exercise–cold stress studies. Similarly, 5 h of exhaustive exercise lowers baseline insulin levels and remains low during subsequent Insulin and glucagon cold–wet exposure (Tikuisis et al. 1999). Glucagon concentrations have been shown to Insulin and glucagon are the two primary fuel regu- both increase and decrease as a result of resting cold latory hormones. Insulin is an anabolic hormone exposure. In the two studies that found increased that promotes fuel storage throughout the body levels (Seitz et al. 1981; Tikuisis et al. 1999), plasma whereas glucagon acts on the liver to secrete glucose hormone values were not corrected for plasma vol- and β-hydroxybutyrate. Since moderate sedentary ume changes (decreased by 9–10%), which could cold exposure increases plasma glucose oxidation account for the higher values. In the other studies, by 138%, muscle glycogen oxidation by 109% and conducted either in cold water or cold air, glucagon lipid oxidation by 376% (Haman et al. 2002), it did not change as a result of cold exposure (Jacobs would appear that these hormones would have et al. 1984; Martineau & Jacobs 1989; Vallerand et al. a role in mobilizing substrates to fuel shivering 1995). thermogenesis. There are limited data on the response of Plasma insulin does not appear to change as glucagon to exercise–cold stress. During moderate a result of acute cold exposure where there are intensity swimming in cold water (21°C) that minimal changes in core temperature (Martineau & elicited a 0.8°C decline in core temperature (Galbo Jacobs 1989; Vallerand et al. 1995; Tipton et al. 1997; et al. 1979), glucagon levels did not change during or Haman et al. 2002; Koska et al. 2002), but during after exercise, while the same exercise elicited an a 10°C cold-water immersion that elicited a 1°C increase in glucagon in warm water. These findings decline in core temperature and higher metabolic suggest that either cold-water exercise suppresses rates, Jacobs et al. (1984) found a 32% decline in glucagon release or that a threshold core temperat- insulin concentrations. Cold-air exposure does ure needs to be reached before glucagon secretion appear to enhance insulin sensitivity or insulin occurs. Free fatty acids levels rose throughout the responsiveness in skeletal muscle. Following an exercise period and during recovery, but the same intravenous glucose tolerance test, plasma glucose response was observed during swimming in 27° falls more rapidly with lower plasma insulin con- and 33°C water, arguing against cold exposure centrations in 10°C air compared to a temperate increasing the need for fat metabolism. In a study of environment (Vallerand et al. 1988). diabetic patients (Ronnemaa et al. 1991), glucagon Exercise–cold stress studies are difficult to inter- levels did not change during exercise in either pret due to different research designs, exercise warm (30°C) or cold (10°C) conditions. Exercise was modes and intensities. It appears that, unlike rest, limited to three 15-min bouts of cycle ergometry at o lower insulin levels are not related to lower core 60% V 2max. Considering that the magnitude of the temperatures during exercise (Galbo et al. 1979) and glucagon response is dependent on the intensity that the response is not mediated via β-adrenergic and duration of exercise, the lack of a glucagon receptors (Lehtonen et al. 1984). Ten weeks of cold- response is not surprising. water exposure causes insulin to fall during light intensity cold-water swimming (Hermanussen et al. Growth hormone 1995). It is unclear if this is related to changes in body temperatures or to the higher basal insulin Growth hormone (GH) is secreted by the anterior concentrations. The catabolic state also may influ- pituitary and defends glucose concentrations by sus- ence the insulin response. Following a 36-h fast, taining lipolysis and inhibiting glucose metabolism. 506 chapter 33

Therefore, it may be important for regulating fuel GH release is affected by the interaction of exercise homeostasis during prolonged exercise–cold stress. and cold stress and what these changes mean for For example, by defending glucose concentrations overall fuel metabolism. and preventing hypoglycemia, GH may have a role in maintaining shivering thermogenesis. Reproductive hormones GH has been studied using various methodo- logies to induce cold stress, including cold-air Brain serotonin is linked to temperature regulation; exposure, ice ingestion and cool-water immersion increases in brain serotonin levels are associated combined with ice ingestion. The findings from with feelings of sleepiness and lethargy; and there is these studies are highly variable. Sitting in cold air evidence implicating serotonin in producing central for 2-h has been shown to either cause no change fatigue during exercise (Cheuvront & Sawka 2001). (4°C air, Golstein-Golaire et al. 1970) or lower GH Since central serotonin levels in humans cannot be concentrations (10°C, Leppäluoto et al. 1988), with measured, plasma prolactin (PRL) concentrations transient increases 30 min following exposure are an accepted marker for brain serotonergic activ- (Okada et al. 1970). Ice-water ingestion that lowered ity (Cheuvront & Sawka 2001) due to anatomical tympanic temperatures by 0.5–0.8°C demonstrated relationships within the brain as well as the finding either no change (Berg et al. 1966) or a significant that plasma PRL concentrations change in response decrease (Weeke & Gundersen 1983). The reader to serotonin agonists and antagonists. Thus PRL must be cognizant that many of these studies did changes during exercise–cold stress may be indicat- not employ a control group and also are measuring ive of changes in temperature regulation or useful GH at one point in time. This is important because as a marker of fatigue. GH is secreted in a pulsatile fashion and depending Resting cold exposure significantly decreases when the sample is taken, GH can be relatively high prolactin concentrations during and up to 90 min or be undetectable. Serial sampling over many post-exposure (Mills & Robertshaw 1981; O’Malley hours has not been done using repeated subjects et al. 1984; Leppäluoto et al. 1988), although the designs during or after cold exposure. studies varied slightly in exposure temperature and Only three studies have examined GH responses duration. during or after exercise–cold stress. Galbo et al. The literature is equivocal concerning exercise– (1979) found that exercising in 21°C water (decreased cold stress and acute changes in PRL, and there rectal temperature by 0.8°C) for 60-min elicited no appears to be influences of both exercise and cold- change in GH, whereas GH was elevated following stress intensity. No clear relationship exists be- exercise in 27°C and 33°C water. These data suggest tween PRL changes and temperature regulation. that lower body core temperatures suppress GH One study (Frewin et al. 1976) found no change in release or that a core temperature threshold is PRL levels during moderate intensity treadmill needed before plasma GH levels rise. Lehtonen et al. exercise in 10°C air, but an increase when the same (1984) exercised subjects for 1 h at –3°C at a heart exercise was performed in 40°C air. Another study rate of 140–150 b·min–1 on one of three treatments: (Ronnemaa et al. 1991) reported an increase in PRL β ° placebo, atenolol ( 1-adrenergic blocker) and pin- during exercise in 10 C air in patients with diabetes, β β dolol (blocks 1 and 2). They found that GH but, again, the PRL response at 10°C was lower was significantly higher during the pindolol trial compared to exercise at 30°C in these patients. The β compared to placebo, suggesting that 2-adrenergic increase in PRL during 10°C exercise in diabetics receptors modulate GH responses to exercise–cold may be due to the subject population, although PRL stress. Studies (Hermanussen et al. 1995) also show is lower in type I diabetics compared to healthy sub- that GH does not change in the post-exercise period jects (Ramires et al. 1993). A third study reported after swimming in 2–7°C water for 1.5–15.0 min. a decrease in PRL during an 8-h shift in meat- Since these studies only provide a ‘snapshot’ in processing plant workers exposed to –20°C to –40°C time, it is unknown how the normal pulsatility of temperatures (Solter & Misjak 1989). Other research hormones and cold stress 507

suggests that PRL levels only increase during exer- study (Johansen & Norman 1991), three of the four cise when core temperature increases above 38°C male subjects gained total body mass and lean mass (Cheuvront & Sawka 2001). Therefore, PRL as a sen- and lost body fat during a period of severe testo- sitive marker of temperature regulation or fatigue sterone deficiency (the authors remarked that testo- during exercise–cold stress is not supported by the sterone values during the last 22 days were in the literature. ‘female range’). This is in contrast to the remaining There is evidence, however, that PRL may be a subject, who lost a significant amount of total body marker of cold acclimation. Ten weeks of a winter mass, lean mass and body fat yet had a similar swimming program (2–7°C water) doubled the testosterone profile during the trek. The authors baseline PRL compared to preswimming concen- speculated that GH may have been elevated during trations while non-exercise control group PRL the trek and was responsible for the increased lean levels did not change, indicating the possibility of mass, but they present no data to support this seasonal variation (Hermanussen et al. 1995). hypothesis. Testosterone is an anabolic hormone that is res- Luteinizing hormone (LH) stimulates Leydig ponsible for growth and development of tissues cells in the testes to synthesize and secrete testo- that characterize males, including skeletal muscle sterone. Sedentary exposure for 2 h to 10°C air does growth. With respect to cold environments, this not change blood LH values (Leppäluoto et al. 1988). hormone has been studied in relation to sperm pro- The exercise studies are difficult to interpret due to duction. McConnell and Sinning (1984) found exer- environmental/field conditions and methodolo- cise–cold stress (running at 80% maximal heart rate gies and there is no clear linkage to testosterone. in 6.2°C air for 5 consecutive days) had no effect on Several minutes of swimming in ice water has no sperm count, semen sample volume, or total sperm effect on plasma LH (Hermanussen et al. 1995). per sample. They also found testosterone increased Solter and Misjak (1989) studied meat-processing on days 4 and 5 compared to baseline following workers who were exposed to ambient temper- daily cold-air exercise. However, the same response atures between –40°C and –20°C for 3.5 h·day–1 and was found whether the exercise was conducted in compared them to workers either exposed to less warm and hot conditions, indicating that the testo- severe cold (4–8°C) or normal room temperature. sterone results are caused by exercise and not cold No acute changes were observed in LH in any stress. On the other hand, resting cold exposure does group, although workers chronically exposed to not change testosterone concentrations (Leppäluoto extreme cold had lower basal LH values. LH et al. 1988) and other exercise–cold stress studies decreased over time during a 500-km ski-trek suggest testosterone decreases. During 10 days of (6 weeks) in Greenland, reaching a nadir during the outdoor military training in the winter (Hackney & 4th week, and then increasing during the final Hodgdon 1991), subjects who slept in tents at night 2 weeks (Johansen & Norman 1991). had a lower testosterone on day 5 compared to sub- Menstrual cycle status affects thermoregulatory jects who slept in warmer barracks at night. Testo- responses in cold environments. During the luteal sterone in the outdoor-sleeping group returned to phase, when endogenous levels of estrogen and normal on day 10. Testosterone decreased after an progesterone are high, shivering is lower at any 8-h shift in meat processing plant workers, but only given mean body temperature and there is also in the group exposed to the lowest ambient tem- higher heat flux from the body (Gonzalez & peratures (Solter & Misjak 1989). This group was the Blanchard 1998). Hessemer and Brück (1985) found only one who experienced intermittent exposure that the temperature onset for shivering was ~ 0.5°C and the effect of the alternating 1-h work and 1-h higher in the luteal phase, paralleling the rise in core rest (in normal room temperature conditions) is not temperature observed during this phase. Likewise, known. Core temperature was not reported during Charkoudian and Johnson (1999) also demonstrated this study, but it is likely that it did not change due that the onset of cutaneous vasoconstriction began to clothing worn by the workers. In a 42-day field at a higher internal temperature during the luteal 508 chapter 33

phase. These studies demonstrate that reproduct- exposure. Future studies need to systematically ive hormones influence thermoregulatory effector examine the interaction of exercise duration, mode responses in the cold. However, it is less clear what and intensity with cold exposure to determine effect cold exposure has on estrogen and proges- endocrine responses and, more importantly, the terone secretion. There are no studies describing role of the endocrine system in maintaining phy- how cold exposure impacts these reproductive siological homeostasis. Whether the changes in hormones. hormone levels as a result of cold exposure have important physiological or performance conse- Summary quences is unknown. An example is the role of thyroid hormones during acute cold exposure. Information on the effect of exercise–cold stress on What happens if a person with hypothyroidism endocrine responses is limited. Because there have exercises in a cold environment? Are they worse been relatively few studies completed and these off or does it not matter? This question cannot be have used a variety of experimental approaches, it answered at this point. The use of patients with is difficult to discern if the changes are due to the clinical endocrine disease may further our under- interaction of exercise–cold stress or if they are standing of the physiological role of specific due to the independent effects of exercise or cold hormones during exercise–cold stress.

References

Ainslie, P.N., Campbell, I.T., Frayn, K.N. Scandinavian Journal of Clinical Laboratory Castellani, J.W., Young, A.J., Kain, J.E. & et al. (2002) Physiological and metabolic Investigation 35, 311–317. Sawka, M.N. (1999b) Thermoregulatory responses to a hill walk. Journal of Bader, R.A., Eliot, J.W. & Bass, D.R. (1952) responses to cold water at different Applied Physiology 92, 179–187. Hormonal and renal mechanisms of cold times of day. Journal of Applied Physiology Anderson, D.E. & Hickey, M.S. (1994) diuresis. Journal of Applied Physiology 4, 87, 243–246. Effects of caffeine on the metabolic and 649–658. Castellani, J.W., Young, A.J., DeGroot, catecholamine responses to exercise in 5 Berg, G.R., Utiger, R.D., Schalch, D.S. D.W. et al. (2001) Thermoregulation and 28°C. Medicine and Science in Sports & Reichlin, S. (1966) Effect of central during cold exposure after several days and Exercise 26, 453–458. cooling in man on pituitary–thyroid of exhaustive exercise. Journal of Applied Arjamaa, O., Turunen, L., Makinen, T. et al. function and growth hormone secretion. Physiology 90, 939–946. (1999) Blood pressure and hormonal Journal of Applied Physiology 21, Castellani, J.W., Young, A.J., Stulz, D.A. responses to short whole body cold 1791–1794. et al. (2002) Pituitary–adrenal and exposure in subjects with high dietary Bergh, U., Hartley, H., Landsberg, L. pituitary–thyroid hormone responses salt intake. Applied Human Science 18, & Ekblom, B. (1979) Plasma during exercise–cold exposure after 7 203–209. norepinephrine concentration during days of exhaustive exercise. Aviation Arjamaa, O., Mäkinen, T., Turenen, L. submaximal and maximal exercise at Space and Environmental Medicine 73, et al. (2001) Are the blood pressure lowered skin and core temperatures. 544–550. and endocrine responses of healthy Acta Physiologica Scandinavia 106, Charkoudian, N. & Johnson, J.M. subjects exposed to cold stress altered 383–384. (1999) Reflex control of cutaneous by an acutely increased sodium intake? Bestle, M.H., Olsen, N.V., Christensen, P., vasoconstrictor system is reset by European Journal of Applied Physiology Jensen, B.V. & Bie, P. (1999) exogenous female reproductive and Occupational Physiology 84, Cardiovascular, endocrine, and renal hormones. Journal of Applied Physiology 48–53. effects of urodilatin in normal humans. 87, 381–385. Armstrong, D.W. (1998) Metabolic and American Journal of Physiology 276, Cheuvront, S.N. & Sawka, M.N. (2001) endocrine responses to cold air in R684–R695. Physical exercise and exhaustion from women differing in aerobic capacity. Castellani, J.W., Young, A.J., Sawka, M.N. heat strain. Journal of the Korean Society of Medicine and Science in Sports and Exercise & Pandolf, K.B. (1998) Human Living and Environmental Systems 8, 30, 880–884. thermoregulatory responses during 134–145. Arnett, E.L. & Watts, D.T. (1960) serial cold-water immersions. Journal Deligiannis, A., Karamouzis, M., Koukdi, Catecholamine excretion in men of Applied Physiology 85, 204–209. E., Mougios, V. & Kallaras, M.C. (1993)

exposed to cold. Journal of Applied Castellani, J.W., Young, A.J., Kain, J.E., Plasma TSH, T3, T4 and cortisol Physiology 15, 499–500. Rouse, A. & Sawka, M.N. (1999a) responses to swimming at varying water Atterhog, J., Carlens, P., Granberg, P. & Thermoregulation during cold temperatures. British Journal of Sports Wallenberg, L. (1975) Cardiovascular exposure: effects of prior exercise. Medicine 27, 247–250. and renal responses to acute cold Journal of Applied Physiology 87, Deuster, P.A., Smith, D.J., Smoak, B.L. exposure in water-loaded man. 247–252. et al. (1989) Prolonged whole-body cold- hormones and cold stress 509

water immersion: fluid and ion shifts. natriuretic petide in men during a Johnson, D., Hayward, J., Jacobs, T., Collis, Journal of Applied Physiology 66, 34–41. 2-hour stay in a cold room. Acta M., Eckerson, J. & Williams, R. (1977) Dulac, S., Quirion, A., DeCarufel, D. Physiologica Scandinavia 142, 481–485. Plasma norepinephrine responses of et al. (1987) Metabolic and hormonal Hennig, J., Laschefski, U., Becker, H., man in cold water. Journal of Applied responses to long-distance swimming in Rammsayer, T. & Netter, P. (1993) Physiology 43, 216–220. cold water. International Journal of Sports Immune cell and cortisol responses Kauppinen, K., Pajari-Backas, M., Volin, P. Medicine 8, 352–356. to physically and pharmacologically & Vakkuri, O. (1989) Some endocrine Frank, S.M., Higgins, M.S., Fleisher, L.A. induced lowering of body core responses to sauna, shower and ice et al. (1997) Adrenergic, respiratory, and temperature. Neuropsychobiology 28, water immersion. Arctic Medical Research cardiovascular effects of core cooling in 82–86. 48, 131–139. humans. American Journal of Physiology Hermanussen, M., Jensen, F., Hirsch, N. Knight, D. & Horvath, S. (1985) Urinary 272, R557–R562. et al. (1995) Acute and chronic effects of responses to cold temperature during Frank, S.M., Raja, S.N., Bulcao, C.F. winter swimming on LH, FSH, prolactin, water immersion. American Journal of & Goldstein, D.S. (1999) Relative growth hormone, TSH, cortisol, serum Physiology 248, R560–R566. contribution of core and cutaneous glucose and insulin. Arctic Medical Koska, J., Ksinantova, L., Sebokova, E. temperatures to thermal comfort and Research 54, 45–51. et al. (2002) Endocrine regulation of autonomic responses in humans. Journal Hershman, J.M., Read, D.G., Bailey, A.L., subcutaneous fat metabolism during of Applied Physiology 86, 1588–1593. Norman, V.D. & Gibson, T.B. (1970) cold exposure in humans. Annals of the Frank, S., Cattaneo, C., Wieneke- Effect of cold exposure on serum New York Academy of Sciences 967, Brady, M. et al. (2002) Threshold for thyrotropin. Journal of Clinical 500–505. adrenomedullary activation and Endocrinology 30, 430–434. Lamke, L.O., Lennquist, S., Liljedahl, S.O. increased cardiac work during mild Hessemer, V. & Brück, K. (1985) Influence & Wedin, B. (1972) The influence of cold core hypoxia. Clinical Science 102, of menstrual cycle on shivering, skin stress on catecholamine excretion and 119–125. blood flow, and sweating responses oxygen uptake of normal persons. Frewin, D., Frantz, A. & Downey, J. (1976) measured at night. Journal of Applied Scandinavian Journal of Clinical Laboratory The effect of ambient temperature on Physiology 59, 1902–1910. Investigation 30, 57–62. the growth hormone and prolactin Hesslink, R.L., D’Alesandro, M.M., Larsen, P.R., Davies, T.F. & Hay, I.D. (1998) response to exercise. Australian Journal Armstrong, D.W.I. & Reed, H.L. The thyroid gland. In: Williams Textbook of Experimental Biology and Medical (1992) Human cold air habituation of Endocrinology (Wilson, J.D., Foster, Science 54, 97–101. is independent of thyroxine and D.W., Kronenberg, H.M. & Larsen, P.R., Galbo, H., Houston, M.E., Christensen, thyrotropin. Journal of Applied Physiology eds.). W.B. Saunders, Co., Philadelphia: N.J. et al. (1979) The effect of water 72, 2134–2139. 389–515. temperature on the hormonal response Hiramatsu, K., Yamada, T. & Katakura, M. Lehtonen, A., Huupponen, R. & to prolonged swimming. Acta (1984) Acute effects of cold on blood Himanen, P. (1984) Effect of the Physiologica Scandinavia 105, 326–337. pressure, renin–angiotensin– short-term β-adrenoceptor bloackade Golstein-Golaire, J., VanHearlst, L., Bruno, aldosterone system, catecholamines and on exercise metabolism in cold O., Leclercq, R. & Copinschi, G. (1970) adrenal steroids in man. Clinical and weather. International Journal of Clinical Acute effects of cold on blood levels Experimental Pharmacology and Physiology Pharmacology, Therapy, and Toxicology 22, of growth hormone, cortisol, and 11, 171–179. 86–90. thyrotropin in man. Journal of Applied Hynynen, M., Ilmarinen, R., Tikkanen, I. Lennquist, S. (1972) Cold-induced diuresis. Physiology 29, 622–626. & Fyhrquist, F. (1993) Plasma atrial Scandinavian Journal of Urology and Gonzalez, R.R. & Blanchard, L.A. (1998) natriuretic factor during cold-induced Nephrology Supplement 9, 1–46. Thermoregulatory responses to cold diuresis. European Journal of Applied Lennquist, S., Granberg, P. & Wedin, B. transients: effects of menstrual cycle Physiology and Occupational Physiology (1974) Fluid balance and physical work in resting women. Journal of Applied 67, 286–289. capacity in humans exposed to cold. Physiology 85, 543–553. Jacobs, I., Romet, T.T., Frim, J. & Hynes, A. Archives of Environmental Health 29, Goodman, H.M. (1994) Basic Medical (1984) Effects of endurance fitness on 241–249. Endocrinology. Lippincott-Raven, responses to cold water immersion. Leppäluoto, J., Korhonen, I., Huttunen, P. Philadelphia, PA. Aviation Space and Environmental & Hassi, J. (1988) Serum levels of thyroid Hackney, A. & Hodgdon, J. (1991) Medicine 55, 715–720. and adrenal hormones, testosterone, Norwegian military field exercises in Jansky, L., Sramek, P., Savlikova, J. et al. TSH, LH, GH and prolactin in men the Arctic: endocrine and metabolic (1996) Change in sympathetic activity, after a 2-h stay in a cold room. Acta responses. Arctic Medical Research 50, cardiovascular functions and plasma Physiologica Scandinavia 132, 543–548. 137–141. hormone concentrations due to cold Leppäluoto, J., Korhonen, I. & Hassi, J. Haman, F., Peronnet, F., Kenny, G.P. et al. water immersion in men. European (2001) Habituation of thermal (2002) Effect of cold exposure on fuel Journal of Applied Physiology and sensations, skin temperatures, and utilization in humans: plasma glucose, Occupational Physiology 74, 148–152. norepinephrine in men exposed to cold muscle glycogen, and lipids. Journal of Johansen, A. & Norman, N. (1991) air. Journal of Applied Physiology 90, Applied Physiology 93, 77–84. Reproductive hormones during 42 days 1211–1218. Hassi, J., Rintamäki, H., Ruskoaho, H., of maximal physical effort, low McConnell, T. & Sinning, W. (1984) Leppäluoto, J. & Vuolteenaho, O. (1991) temperatures and general hardship. Exercise and temperature effects Plasma levels of endothelin-1 and atrial Arctic Medical Research 50, 142–147. on human sperm production and 510 chapter 33

testosterone levels. Medicine and Science acute cold stress. Clinical Endocrinology local cold acclimation. Journal of Applied in Sports and Exercise 16, 51–55. 21, 285–291. Physiology 76, 1963–1967. McMurray, R.G., Kocker, P.L. & Horvath, Pandolf, K.B., Gange, R.W., Latzka, W.A. Sawhney, R., Malhotra, A., Nair, C. S.M. (1994) Aerobic power and body et al. (1992) Human thermoregulatory et al. (1995) Thyroid function during a size affects the exercise-induced stress responses during cold-water immersion prolonged stay in Antarctica. European hormone responses to varying water after artificially induced sunburn. Journal of Applied Physiology and temperatures. Aviation Space and American Journal of Physiology. Regulatory Occupational Physiology 72, 127–133. Environmental Medicine 65, 809–814. Integrative and Comparative Physiology Segar, W.E. & Moore, W.W. (1968) The Marino, F., Sockler, J. & Fry, J. (1998) 262, R617–R623. regulation of antidiuretic hormone Thermoregulatory, metabolic and Parkin, J.M., Carey, M.F., Zhao, S., & release in man. I. Effects of change in sympathoadrenal responses to repeated Febbraio, M.A. (1999) Effect of ambient position and ambient temperature on brief exposure to cold. Scandinavian temperature on human skeletal blood ADH levels. Journal of Clinical Journal of Clinical Laboratory Investigation muscle metabolism during fatiguing Investigation 47, 2143–2151. 58, 537–546. submaximal exercise. Journal of Applied Seitz, H., Krone, W., Wilke, H. & Martineau, L. & Jacobs, I. (1989) Free fatty Physiology 86, 902–908. Tarnowski, W. (1981) Rapid rise in acid availability and temperature Quirion, A., Laurencelle, L., Paulin, L. plasma glucagon induced by acute cold regulation in cold water. Journal of et al. (1989) Metabolic and hormonal exposure in man and rat. Pflugers Archiv Applied Physiology 67, 2466–2472. responses during exercise at 20, 0, 389, 115–120. Mills, D. & Robertshaw, D. (1981) and –20°C. International Journal of Solter, M. & Misjak, M. (1989) Pituitary– Response of plasma prolactin to changes Biometerology 33, 227–232. gonadal response to extreme cold in ambient temperature and humidity in Ramires, P.R., Forjaz, C.L., Silva, M.E. et al. exposure in healthy men. Hormone and man. Journal of Clinical Endocrinology and (1993) Exercise tolerance is lower in type Metabolic Research 21, 343–344. Metabolism 52, 279–283. I diabetics compared with normal young Sramek, P., Simeckova, M., Jansky, L., Nagashima, K., Yoda, T., Yagashita, T. et al. men. Metabolism 42, 191–195. Savlikova, J. & Vybiral, S. (2000) Human (2002) Thermal regulation and comfort Reed, H.L., Burman, K.D., Shakir, K.M.M. physiological responses to immersion during a mild-cold exposure in young & O’Brian, J.T. (1986) Alterations in the into water of different temperatures. Japanese women complaining of hypothalamic–pituitary–thyroid axis European Journal of Applied Physiology and unusual coldness. Journal of Applied after prolonged residence in Antarctica. Occupational Physiology 81, 436–442. Physiology 92, 1029–1035. Clinical Endocrinology 25, 55–65. Stevens, G.H., Graham, T.E. & Wilson, Nagata, H., Izumiyama, T., Kamata, K. Reed, H.L., Ferreiro, J.A., Shakir, K.M.M., B.A. (1987) Gender differences in et al. (1976) An increase of plasma Burman, K.D. & O’Brian, J.T. (1988) cardiovascular and metabolic responses triiodothyronine concentration in man Pituitary and peripheral hormone to cold and exercise. Canadian Journal of

in a cold environment. Journal of Clinical responses to T3 administration during Physiology and Pharmacology 65, 165–171. Endocrinology and Metabolism 43, Anarctic residence. American Journal of Suzuki, M., Tonoue, T., Matsuzaki, S. & 1153–1156. Physiology 254, E733–E739. Yamamoto, K. (1967) Initial response Nakamitsu, S., Sagawa, S., Miki, K. et al. Reed, H.L., Brice, D., Shakir, K.M.M. et al. of human thyroid, adrenal cortex, and (1994) Effect of water temperature on (1990) Decreased free fraction of thyroid adrenal medulla to acute cold exposure. diuresis-natriuresis: AVP, ANP, and hormones after prolonged Antarctic Canadian Journal of Physiology and urodilatin during immersion in men. residence. Journal of Applied Physiology Pharmacology 45, 423–432. Journal of Applied Physiology 77, 69, 1467–1472. Therminarias, A., Flore, P., Oddul- 1919–1925. Reed, H.L., Reedy, K.R., Palinkas, L.A. Chirpaz, M.F., Gharib, C. & Gauquelin, O’Brien, C., Young, A.J. & Sawka, M.N. et al. (2001) Impairment in cognitive and G. (1992) Hormonal responses to (1998) Hypohydration and exercise performance during prolonged exercise during moderate cold exposure thermoregulation in cold air. Journal Antarctic residence: effect of thyroxine in young vs middle-age subjects. Journal of Applied Physiology 84, 185–189. supplementation in the polar of Applied Physiology 73, 1564–1571. Ohno, H., Yahata, T., Yamashita, K. & triiodothyronine syndrome. Journal of Thompson, R., Buskirk, E. & Whedon, G. Kuroshima, A. (1987) Effect of acute Clinical Endocrinology and Metabolism 86, (1971) Temperature regulation against cold exposure on ACTH and zinc 110–116. cold: effect of induced hyperthyroidism concentrations in human plasma. Rogers, T.A., Setliff, J.A. & Klopping, J.C. in men and women. Journal of Applied Japanese Journal of Physiology 37, (1964) Energy cost, fluid and electrolyte Physiology 31, 740–745. 749–755. balance in subarctic survival situations. Tikuisis, P., Ducharme, M.B., Moroz, D. & Okada, H., Miyai, K., Iwatsubo, H. & Journal of Applied Physiology 19, Jacobs, I. (1999) Physiological responses Kumahara, Y. (1970) Human growth 1–8. of exercised-fatigued individuals hormone secretion in normal adult Ronnemaa, T., Marniemi, J., Leino, A. et al. exposed to wet–cold conditions. Journal subjects during and after exposure to (1991) Hormone response of diabetic of Applied Physiology 86, 1319–1328. cold. Journal of Clinical Endocrinology 30, patients to exercise at cool and warm Tipton, M., Franks, G., Meneilly, G. & 393–395. temperatures. European Journal of Applied Mekjavic, I. (1997) Substrate utilization O’Malley, B., Cook, N., Richardson, A., Physiology and Occupational Physiology during exercise and shivering. European Barnett, D. & Rosenthal, F. (1984) 62, 109–115. Journal of Applied Physiology and Circulating catecholamines, Savourey, G., Caravel, J.P., Barnavol, B. & Occupational Physiology 76, 103–108. thyrotrophin, thyroid hormone and Bittel, J.H.M. (1994) Thyroid hormone Tuomisto, J., Mannisto, P., Lamberg, B.-A. prolactin responses of normal subjects to changes in a cold air environment after & Linnoila, M. (1976) Effect of cold- hormones and cold stress 511

exposure on serum thyrotrophin Weiß, M., Hack, F., Stehle, R., Pollert, R. & Horvath, S.M. (1974) Adaptation in levels in man. Acta Endocrinologica 83, & Weicker, H. (1988) Effects of man’s adrenal function in response to 522–527. temperature and water immersion on acute cold stress. Journal of Applied Vallerand, A.L., Firm, J. & Kavanaugh, plasma catecholamines and circulation. Physiology 36, 183–189. M.F. (1988) Plasma glucose and insulin International Journal of Sports Medicine 9, Wilson, O., Hedner, P., Laurell, S. et al. responses to oral and intravenous 113–117. (1970) Thyroid and adrenal response to glucose in cold-exposed humans. Journal Weller, A.S., Millard, C.E., Stroud, M.A., acute cold exposure in man. Journal of of Applied Physiology 65, 2395–2399. Greenhaff, P.L. & Macdonald, I.A. (1997a) Applied Physiology 28, 543–548. Vallerand, A.L., Zamecnik, J. & Jacobs, I. Physiological responses to a cold, wet, Wittert, G., Or, H., Livesey, J. et al. (1992) (1995) Plasma glucose turnover during and windy environment during Vasopressin, corticotropin-releasing cold stress in humans. Journal of Applied prolonged intermittent walking. American factor, and pituitary adrenal responses Physiology 78, 1296–1302. Journal of Physiology 272, R226–R233. to acute cold stress in normal humans. Wallenberg, L.R. & Granberg, P.O. (1974) Weller, A.S., Millard, C.E., Stroud, M.A., Journal of Clinical Endocrinology and The relationship between cold-induced Greenhaff, P.L. & Macdonald, I.A. Metabolism 75, 750–755. natriuresis and arterial blood pressure (1997b) Physiological responses to cold Young, A.J. (1996) Homeostatic responses in man. Scandinavian Journal of Clinical stress during prolonged intermittent to prolonged cold exposure: human Laboratory Investigation 34, 225–231. low- and high-intensity walking. cold acclimatization. In: Handbook of Wang L.C.H., Man, S.F.P. & Belcastro, A.N. American Journal of Physiology 272, Physiology: Environmental Physiology (1987) Metabolic and hormonal R2025–R2033. (Fregly, M.J. & Blatteis, C.M., eds.). responses in theophylline-increased cold Weller, A.S., Millard, C.E., Greenhaff, P.L. American Physiological Society, resistance in males. Journal of Applied & Macdonald, I.A. (1998) The influence Bethesda, MD: 419–438. Physiology 63, 589–596. of cold stress and 36-h fast on the Young, A.J., Muza, S.R., Sawka, M.N., Weeke, J. & Gundersen, H. (1983) The physiological responses to prolonged Gonzalez, R.R. & Pandolf, K.B. (1986) effect of heating and central cooling on intermittent walking in man. European Human thermoregulatory responses to serum TSH, GH, and norepinephrine in Journal of Applied Physiology and cold air are altered by repeated cold resting normal man. Acta Physiologica Occupational Physiology 77, 217–223. water immersion. Journal of Applied Scandinavia 117, 33–39. Wilkerson, J.E., Raven, P.B., Bolduan, N.W. Physiology 60, 1542–1548. Chapter 34

Growth, Maturation and Hormonal Changes During Puberty: Influence of Sport Training

JAMES N. ROEMMICH

Introduction trations of hormones throughout the day. This chap- ter reviews the processes of growth and maturation, The timing and tempo of pubertal maturation and the hormonal axes regulating growth and matura- somatic growth are under genetic control but modi- tion, the influence of acute exercise and of sport fied by nutritional status and a host of hormones. training on the hormonal axes controlling growth Growth and maturation are slowed during chronic and maturation, and the effects of exercise and sport undernutrition due to alterations in the hormonal training on the growth and maturation of youth. milieu including reduced sex steroid, insulin-like growth factor I (IGF) and thyroid hormone con- Growth and maturation centrations. Conversely, overnutrition results in early maturation and an increased growth velocity Linear growth and sexual maturation are mainly due to increased secretion of adrenal androgens, sex controlled by the growth hormone–insulin-like steroids and IGF-I. The influence of nutrition and growth factor I (GH–IGF-I) and the hypothalamic– endocrine function on growth and maturation have pituitary–gonadal (HPG) axes. Many of the growth led many to wonder if sport training influences the effects of GH, including long bone growth and rate of growth and maturation of young athletes. anabolic effects on muscle, are mediated through For instance, could the great exercise energy expend- IGF-I, which is synthesized locally by target tissues itures and limited nutritional intake of some ath- or the liver, the predominant source of circulating letes result in limited energy availability and lead to IGF-I. IGF-I also acts as a negative feedback signal to delayed puberty and slowed growth? This question reduce GH secretion. Serum IGF-I concentrations has received great public interest due to portrayals decline during periods of inadequate energy intake of ever younger gymnasts and figure skaters train- resulting in an opening of the feedback loop and ing year round, sometimes far from home, to pre- increased GH secretion, but low IGF-I production. pare their bodies and hone their skills with the The net result is relative GH resistance and slowed best coaches. Another common question is whether linear and somatic growth. sport training in the presence of adequate nutrition Prepubertal growth is almost exclusively depend- increases growth and accelerates the rate of pubertal ent on GH, IGF-I and thyroid, which has a permiss- maturation. An acute bout of exercise reliably ive effect on GH secretion. The marked acceleration increases the secretion of growth-related hormones in growth velocity at puberty results from the inter- and these increases in hormone secretion occur with action of the GH–IGF-I and HPG axes with a con- each bout of training. Chronic training alters the tinued permissive effect of thyroid hormone. In girls basal secretion of growth-related hormones so that and boys, rising estrogen concentrations stimulate the target tissues such as muscle and bone are twofold increases in GH secretion that are paralleled exposed to greater or smaller than normal concen- by increased serum IGF-I concentrations (Veldhuis 512 sport training and growth 513

et al. 1997, 2000; Roemmich et al. 1998; Mauras 2001). also play key roles in controlling macronutrient At low concentrations, sex steroids and GH syner- metabolism during the increased metabolic demands gistically stimulate skeletal growth, but at greater of exercise. Endocrine changes during exercise concentrations, estrogen induces epiphyseal fusion. promote a catabolic environment to enhance fuel Leptin is a recently discovered hormone that mobilization. Exercise stimulates GH, glucagon and assists in regulating long-term energy balance and cortisol secretion while secretion of the anabolic reproductive function by signaling the size of the hormone insulin is reduced to allow for the mob- fat stores and energy availability to the feeding and ilization of carbohydrates and fatty acids needed reproductive centers of hypothalamus (Cunningham to meet the increased energy demands of physical et al. 1999; Foster & Nagatani 1999). These dual roles activity. Exercise also stimulates testosterone and allow leptin to link nutritional status with proper estrogen secretion which may influence the pro- reproductive function in adults and to the onset of portion of carbohydrate and lipid utilization (Braun puberty in youth (Cunningham et al. 1999; Foster & Horton 2001) and perhaps stimulate tissue & Nagatani 1999). Anecdotal evidence for a role of repair post-exercise. The responses of key hormones leptin in human puberty includes leptin concentra- involved in growth and maturation to acute exercise tions of boys and girls rising just before puberty are summarized below. (Blum et al. 1997; Clayton et al. 1997; Garcia-Mayor et al. 1997; Mantzoros et al. 1997) and an inverse Acute exercise and the GH–IGF-I axis relationship between leptin concentrations and age at menarche (Matkovic et al. 1997). Experiments of Acute aerobic exercise so reliably increases the nature lend stronger evidence in that patients lack- serum GH concentration that it is used as a test ing leptin protein do not attain pubertal maturity to assess adequate GH secretion (Bouix et al. 1994; until leptin replacement therapy (Farooqi et al. 1999). Marin et al. 1994; Ghigo et al. 1996). Resistance exer- cise protocols that emphasize a great number of Nutrition modifies growth and repetitions and short rest periods also reliably pubertal timing increase GH secretion (Kraemer, W.J. et al. 1990, 1991; Häkkinen & Pakarinen 1995; Roemmich & The influence of nutritional status on pubertal Rogol 1997). Exercise influences on growth stimula- timing is clearly demonstrated by the steady decline tion could be hypothesized to be maximized during in age, but stable body weight at menarche from the puberty as exercise-induced GH secretion is great- mid-19th century to mid-20th century. The signals est during the adolescent growth spurt. Increased that reawaken the HPG axis at the onset of puberty concentrations of sex steroids, especially estrogen, and then keep it operative remain unclear. Presum- play a major role in enhancing exercise-induced GH ably, a non-gonadal mediated mechanism deter- release during puberty (Bouix et al. 1994; Marin et al. mines that the energy stores are large enough to 1994; Ghigo et al. 1996). sustain rapid growth, maturation and reproduction. As the downstream mediator of GH action, IGF-I Likely candidates for this signal are the availability secretion would also be expected to reliably increase of metabolic fuels and hormonal signals including with exercise, but it does not. Reviews of both leptin, insulin and components of the GH–IGF-I aerobic (Jenkins 1999) and resistance (Deschenes & axis. The influence of nutrition on the metabolic Kraemer 2002) exercise have highlighted the mixed signals controlling growth and pubertal maturation results regarding the IGF-I response to an exercise has been extensively reviewed (Foster & Nagatani bout. IGF-I and GH concentrations may not both 1999; Rogol & Roemmich 2000; Moschos et al. 2002). increase during exercise due to locally produced IGF-I in active target tissues, such as muscles acting Acute exercise and hormone release through paracrine or autocrine rather than endo- crine mechanisms (DeVol et al. 1990; Yan et al. 1993; Most hormones involved in growth and maturation Cappon et al. 1994). 514 chapter 34

investigated the effects of aerobic or resistance exer- Acute exercise stimulates the HPG axis cise on the reproductive axis of girls. Increased A short-term exercise bout in adult men increases serum concentrations of both estrogen and proges- serum testosterone concentrations in direct pro- terone have been noted after submaximal exercise portion to the exercise intensity (Hackney 1996). (Viru et al. 1998) and both submaximal (Viru et al. However, the evidence is contradictory regarding 1998) and maximal (Kraemer, R.R. et al. 2001) aero- increases in the secretion of testosterone’s upstream bic exercise increases serum testosterone concentra- signal, luteinizing hormone (LH) suggesting that tions of pubertal girls. increased testosterone secretion during short-term exercise is not LH-mediated. Exercise-induced Acute exercise and leptin increases in testosterone concentration are short- lived and return to baseline about 40 min after In adults, a single bout of exercise does not reliably exercise is stopped. In adults, testosterone concen- alter serum leptin concentrations immediately fol- trations begin to return to basal concentrations lowing exercise (Perusse et al. 1997; Racette et al. after about 60 min of exercise and may dip below 1997; Weltman et al. 2000; Kraemer, R.R. et al. 2002), basal concentrations before recovering over the next but may produce reductions 24–48 h after exercise 24–72 h (Hackney 1996). The impact of acute exer- (Essig et al. 2000; Olive & Miller 2001). Longer exer- cise on the HPG axis of youth has not yet been cise bouts of 2 or more hours are more likely to widely studied and there are not enough data to reduce serum leptin concentrations (Duclos et al. draw conclusions. Maximal aerobic exercise failed 1999; Karamouzis et al. 2002), but the reduction to increase serum testosterone concentrations of appears to be more due to negative energy balance boys regardless of stage of pubertal maturation than the exercise bout per se (van Aggel-Leijssen (Fahey et al. 1979), while weight training in adoles- et al. 1999; Hilton & Loucks 2000). There are not yet cent boys increases serum testosterone concentra- enough data to draw conclusions about the effects tions in some (Kraemer, W.J. et al. 1992; Fry et al. of an exercise bout on the leptin secretion of youth. 1993) but not all (Pullinen et al. 2002) studies. A single study, performed in adolescent female dis- Increased testosterone secretion in response to exer- tance runners, found that maximal exercise increased cise may be dependent on training history as youth serum leptin concentrations (Kraemer, R.R. et al. who trained for 2 or more years, regardless of age, 2001). have greater increases in serum testosterone con- centrations compared to youth who have trained for Exercise training and basal hormone shorter periods (Kraemer, W.J. et al. 1992; Fry et al. concentrations 1993). In women, a bout of endurance exercise also pro- The endocrine systems of the body adapt to the duces a reliable but short-lived increase in serum repeated stimulation of exercise training by: (a) testosterone (Keizer et al. 1987; Consitt et al. 2001) altering the strength of the exercise stimulus neces- and estradiol concentrations (Jurkowski et al. 1978; sary to increase or decrease hormone secretion; (b) Nicklas et al. 1989). The estradiol response is depend- altering the tissue responsiveness to the hormone by ent on the exercise intensity and the menstrual altering the amount of unbound ‘biologically active’ cycle phase with greater estradiol secretion during hormone by increasing or decreasing the number of the luteal phase than the follicular phase (Jurkowski circulating binding proteins that bind the hormone et al. 1978; Nicklas et al. 1989). In contrast, resistance and protect it from degradation but also render it exercise training by women does not produce a reli- biologically inactive, the number of cellular hormone able increase in serum testosterone concentrations receptors, or post-receptor function; and (c) work- (Kraemer, W.J. et al. 1993, 1998; Kraemer, R.R. et al. ing through neuroendocrine adaptations to alter 1995), but does increase serum estradiol concentra- basal and maximal hormone secretion. The follow- tions, especially during the luteal phase (Jurkowski ing sections summarize exercise training effects on et al. 1978; Nicklas et al. 1989). Very few studies have growth-related hormone axes. sport training and growth 515

increases in total and free-IGF-I (Koziris et al. 1999). Exercise training and the GH–IGF-I axis These catabolic alterations in the GH–IGF-I axis The most well-controlled study of the influence of may be reversed sometime after 5 weeks of training exercise training on resting GH secretion has shown as youth and adults who are more fit at the begin- that 1 year of endurance training 3 days per week ning of the training have the smallest increases in above the lactate threshold and 3 days per week at inflammatory cytokines (Scheett et al. 2002) and the the lactate threshold increased 24-h mean serum GH quickest normalization of serum IGF-I concentra- concentrations of women by 75%, but no change tions as training continues (Rosendal et al. 2002). was observed in women who trained 6 days per week at the lactate threshold (Weltman et al. 1992). Exercise training and the HPG axis Other studies have reported direct relationships between aerobic fitness and both resting GH secre- There has been a major increase in the number of tion (Weltman et al. 1992) and serum IGF-I concen- women participating in competitive sports over the trations (Poehlman & Copeland 1990) of adults. past 40 years. Enactment in the USA of Title IX of the With a couple of exceptions (Borst et al. 2001; Marx Educational Amendments Act of 1972, opportun- et al. 2001), resistance training has little effect on ities to participate in Olympic distance running resting GH secretion or serum IGF-I concentrations events, ice hockey and soccer, and the formation of of young and older adults (Kraemer, W.J. et al. 1998, highly publicized professional athletic teams have 1999; McCall et al. 1999; Häkkinen et al. 2001; all led to a greater number of women engaging in Deschenes & Kraemer 2002). strenuous exercise (Harber 2000; Loud & Micheli Aerobic fitness is directly related to GH secretion 2001). Many of these female athletes engage in great in adolescent girls and boys (Eliakim et al. 1996, volumes of exercise while being placed under self- 1998b) as are serum IGF-I concentrations (Eliakim imposed and external pressures to maintain a lean et al. 2001; Scheett et al. 2002) although not consist- physique to enhance performance and an esthetic ently (Eliakim et al. 1996, 1998a). While these results appearance. Unfortunately, the resulting low energy suggest that aerobic training upregulates the GH– availability can produce reproductive dysfunction IGF-I axis of youth, correlation does not prove in women ranging from luteal phase insufficiency to causality. Prospective studies do not support a direct secondary or ‘athletic’ amenorrhea (Williams et al. relationship between aerobic fitness and secre- 2001; De Souza et al. 2003; Loucks & Thuma 2003). tion of GH and IGF-I. Several short-term (5-weeks, The estimated prevalence of amenorrhea in athletes 90 min·day–1 of discontinuous exercise, 5 days per (5–66%) is much greater than in the general popula- week) training studies of youth that produced gains tion (2–5%) (Yeager et al. 1993), and the prevalence in aerobic fitness and anabolic effects on mid-thigh of less symptomatic conditions such as luteal phase muscle produced no changes in GH secretion insufficiency is probably even greater. (Eliakim et al. 1996, 1998a). These training protocols Reproductive dysfunction can occur within 8 also produced paradoxical reductions in serum IGF- weeks of strenuous training (Bullen et al. 1985) and I concentrations and alterations in growth hormone likely results from decreased activity of the hypo- binding protein (GHBP) and IGF-I binding pro- thalamic gonadotropin-releasing hormone (GnRH) teins reflecting a catabolic state (Eliakim et al. 1996, pulse generator resulting in suppressed LH pulse 1998a; Scheett et al. 2002). It is unclear how anabolic frequency and, ultimately, reduced estradiol secre- effects are seen in muscle tissue under a catabolic hor- tion (Williams et al. 2001; De Souza et al. 2003; monal environment, or why a catabolic environment Loucks & Thuma 2003). There are numerous health occurs, but this same exercise program increases issues resulting from reproductive dysfunction proinflammatory cytokines that may suppress including reduced bone density, scoliosis and car- the GH–IGF-I axis (Scheett et al. 2002). Aerobic train- diovascular risks (Constantini & Warren 1994), and ing also reduces IGF-I concentrations and alters the constant pressure to limit energy intake can IGF binding protein concentrations of some adults result in disordered eating (Sabatini 2001). Energy (Rosendal et al. 2002), while others have reported and nutrient deficiencies are very concerning in 516 chapter 34

young athletes as the immature reproductive axis 1998). The separate influence of body fat stores may be less resilient to nutritional stress resulting in and energy balance on serum leptin concentrations delayed menarche or primary amenorrhea (Warren and reproductive function is demonstrated in elite 1980; Frisch et al. 1981; Constantini & Warren 1994; amenorrheic athletes who have lower serum leptin Roemmich & Rogol 1995) and perhaps lifelong alter- concentrations than elite cyclic women despite sim- ations in reproductive function. Delayed menarche ilar amounts of body fat. The unusually low serum and nutritional deficiencies may also result in inade- leptin concentrations are explained by undernutri- quate bone mineral accrual, increasing the risk of tion in the amenorrheic athletes as their energy fractures and osteoporosis later in life (Sabatini 2001). intake was only 73% of the cyclic athletes (Thong The impact of sport training on the male repro- et al. 2000). The importance of undernutrition, inde- ductive system has been much less studied. The pendent of body fat, on serum leptin concentrations data are inconsistent but abnormalities include nor- and menstrual dysfunction has also been found in mal or elevated LH concentrations with testosterone recreationally active amenorrheic and cyclic women concentrations of 60–80% of control men, suggest- matched for body fat (Miller et al. 1998). These ing a modest central HPG impairment (Hackney data suggest that exercise training per se does not 1996, 2001; Hackney et al. 2003). Recent evidence alter serum leptin concentrations beyond training- suggests a peripheral mechanism may also reduce induced alterations in body fat as previously testosterone secretion as endurance-trained men reviewed (Kraemer, R.R. et al. 2002). Undernutrition have a lower testicular responsiveness in the form produces an independent negative effect on serum of reduced testosterone secretion to GnRH stimula- leptin concentrations. tion than sedentary control men (Hackney et al. 2003), with oligospermia in some men (Hackney Sport training, growth and maturation 1996, 2001). In contrast to women who can experi- ence reproductive axis dysfunction within several The data demonstrate that an acute bout of exercise months, reductions in serum testosterone are most and sport training alter the serum concentrations of prevalent in men who have been endurance training hormones involved in growth and maturation, but for several years (Hackney 1996, 2001). Women are most of the data have been collected in adults and more likely to limit energy availability by com- much more research is required in children and ado- bining endurance exercise with dietary restriction, lescents. In reproductively mature adults who are and the HPG axis of women may be more sensitive no longer growing, sport training: (a) can increase to energy availability due to the importance of ade- basal GH secretion, which increases lipolysis and quate energy stores for pregnancy and lactation. may help resist the accrual of body fat; and (b) can Few studies have examined the effect of endur- have detrimental effects on reproductive function ance exercise training on the hypothalamic–pituitary when combined with dietary restriction. Although –testicular (HPT) axis of pubertal male athletes. these data demonstrate the potential of sport train- During a short 2-month sport season there were no ing for altering serum concentrations of hormones changes in total or free testosterone concentrations involved in growth and maturation, they cannot be of adolescent male runners (Rowland et al. 1987). generalized to youth athletes. The following sections focus on the evidence, if any, supporting the stimu- latory or inhibitory effects of sport training on the Exercise training and leptin growth and pubertal maturation of youth athletes. There has been much interest in the influence of sport training on serum leptin concentrations due Non-weight control sports to the importance of energy availability for main- taining reproductive function and leptin’s role in There is some evidence that exercise may stimulate signaling both energy stores (i.e. body fat) and linear bone growth in well-nourished animal mod- energy availability (Dubuc et al. 1998; Wadden et al. els. Increased bone loading, as would occur with sport training and growth 517

physical activity, increases long bone growth of results in greater strains (Smith & Gilligan 1996). As adolescent rats (Simon 1978; Simon & Papierski such, female tennis players who start playing before 1982), and running accelerates growth in hamsters menarche have a twofold greater forearm bone via increased GH secretion (Borer & Kaplan 1977; mineral content (BMC) than those who start after Borer et al. 1979, 1986; Conn et al. 1993). Moreover, menarche (Kannus et al. 1995; Haapasalo et al. 1998). during recovery from protein–energy malnutrition, physically active children aged 2–4 years grow Weight control sports more in length and lean body mass than less active children over a 6-week period (Torun & Viteri 1994). Some athletes, including wrestlers, gymnasts and However, the animal studies may not model the some distance runners, combine great energy effects of exercise training on the growth of humans expenditures with dietary restraint to maintain a and the clinical study of children may not generalize low body weight. This practice affords the athlete well to healthy boys and girls engaging in intensive eligibility to compete in a weight class that may pro- sport training. Fortunately, there are large data sets vide them with a competitive advantage, to maintain of well-nourished youth involved in sport training a high strength to body weight ratio, or to maintain that make these generalizations unnecessary. a thin ‘esthetic’ body physique. These are the youth Studies of healthy youth show that for the vast at greatest risk of slowed growth and pubertal mat- majority of boys and girls participating in sport such uration. However, even within this select group of as track and field, baseball, basketball, volleyball, athletes only a small percentage experience limited football and soccer, training consists of exercise energy availability great enough to slow linear energy expenditures of 837–1674 kJ (200–400 kcal) growth and pubertal maturation. The remainder of 3–5 days per week with no concerns placed on low- this review will focus on the growth and maturation ering body weight. Many of these athletes are slightly of two of the most widely studied high-risk athletic taller and more mature than their non-athletic peers groups, female gymnasts and male wrestlers. (Malina 1994; Malina & Bielicki 1996; Malina et al. 1997). These results could be interpreted that exer- gymnasts cise increases the rate of growth and maturation, but it is more likely that these are self-selected inherited Gymnasts gain a competitive advantage by devel- traits because strength and power advantages caused oping muscular strength within a shorter, lighter by earlier maturation are major factors attracting and esthetic body. The performance of female gym- children to sport (Malina 1994). This issue high- nasts often decreases after the onset of puberty lights the difficulty in separating the effects of exer- due to increases in height and fat mass and develop- cise from genetically programmed growth as first ment of the secondary sexual characteristics. There preselection and later natural selection for a sport is much concern that elite gymnasts may use sport creates a non-random population that may skew the training and energy restriction to maintain a slender data about patterns of growth and maturation. physique and delay growth and puberty to remain The effects of sport training on hypertrophic competitive. The concern has become greater as growth such as an increase in size of a muscle, bone ever-younger and leaner girls participate at the elite or organ in response to a stimulus are clear and level of gymnastics. Gymnasts begin training hypertrophic growth for some tissues can be great- around age 5 years, and by the age of 10 years elite est during childhood and adolescence. For instance, gymnasts are training up to 36 h·wk–1, year round regular load-bearing exercise stimulates bone min- (Caine et al. 2001). From 1976 to 1992 the mean age of eral accrual in both laboratory animals and humans Olympic gymnasts decreased from 18 to 16 years, (Kannus et al. 1996; Lanyon 1996; Smith & Gilligan and, although the average height decreased 11% 1996). Immature bone increases bone formation during this time, the mean weight decreased by 22% more than mature bone to a given amount of stress (Sabatini 2001). because the greater compliance of immature bone Many studies have shown that gymnasts are 518 chapter 34

90 90

85 85

80 80

75 75

Segment length (cm) 70 Segment length (cm) 70 Fig. 34.1 Growth in sitting height (open circles) and leg length (closed circles) from bone age 10 years to 65 65 bone age 16 years. Gymnasts are (a) (b) shown in (a) and swimmers in (b). 10 11 12 13 14 15 16 10 11 12 13 14 15 16 Note the lack of increase in leg Bone age (year) Bone age (year) length after bone age 12 years in the Gymnasts Swimmers gymnasts. (From Theintz et al. 1993.) shorter than average and that height disparities However, there is some evidence that gymnastics increase with advancing age and the level of gym- training slows growth and pubertal maturation. nastics competition (Caine et al. 2001). All measures A controversial study reported a delayed age at of pubertal maturation, including bone age, devel- menarche, slowed linear growth, the lack of a nor- opment of the secondary sexual characteristics and mal growth spurt and reduced growth potential age at menarche, are also delayed in most competit- of gymnasts compare to swimmers (Theintz et al. ive gymnasts (Caine et al. 2001), which explains the 1993). Leg length velocity slowed at puberty with greater differences from normal in the height of gym- almost no growth after a bone age of 12 years nasts with advancing age. The growth and matura- (Fig. 34.1). The gymnasts also had significantly tion characteristics of gymnasts may be explained in lower serum concentrations of estradiol and IGF-I part by self-selection and genetics as elite gymnasts than swimmers even when analyzed by bone age are shorter than non-elite gymnasts before gymnas- (Theintz 1994), suggesting a depression of both of tics training and are children of short parents who the major axes involved in growth and maturation. also had later than average puberty (Peltenburg The adult height of gymnasts can be shorter than et al. 1984; Theintz et al. 1989; Baxter-Jones et al. expected based on mid-parental height (Lindholm 1995). Whether gymnastics training reduces growth et al. 1994), and growth spurts do not occur in some independent of self-selection and genetic influences girls until the training load is reduced (Lindholm on growth is unclear. Similar to menstrual dysfunc- et al. 1994; Caine et al. 2001). Other studies have tion in adult athletes, slowed growth and maturation reported accelerated or catch-up growth after retir- in girls may be more the result of undernutrition ing from gymnastics competition (Bass et al. 2000; or psychological stress than gymnastics training. Caine et al. 2001), suggesting that growth was Dietary restriction of 80–90% of recommended slowed when training. amounts is very prevalent in gymnasts and asso- ciated with reduced height and delayed bone age wrestlers (Caine et al. 2001). Olympic-level male gymnasts with adequate energy intake do not experience the Wrestling is made fairer by placing participants in same delayed puberty or hormonal alterations as weight classes so that a smaller wrestler does not their female counterparts (Weimann 2002), but this have to compete against a larger wrestler. Wrestlers may also be due to differences in sensitivity of the must weigh-in before competition to certify that sexes to nutritional insult. their body weight is within limits of the weight sport training and growth 519

class. To gain an advantage by competing in a lower 28 weight, wrestlers reduce their body weight through B

) A a combination of exercise, dietary restriction and –1 26 dehydration. Typical wrestling training lasts 2– C 2.5 h·day–1, 5–6 days per week. Adolescent wrestlers’ 24 nutrient intakes can consistently fall below 50% 22 of the recommended amounts during a 3–4 month season (Fig. 34.2) (Roemmich & Sinning 1997b) re- 20 A,B,C sulting in reduced protein nutritional status (Fig. 34.3) Prealbumin (mg·dL (Horswill et al. 1990). 18 This undernutrition is accompanied by low tes- Early Late Post tosterone and free-testosterone concentrations while Fig. 34.3 Serum prealbumin concentration (mg·dL–1) for serum LH concentrations are conserved (Roemmich the wrestler (open circles, n = 9) and control (filled circles, & Sinning 1997a) (Fig. 34.4). In a well-nourished n = 7) groups. Like letters are significantly different. individual, low testosterone concentrations result Values expressed as mean (± SE). For definitions see in reduced testosterone negative feedback and elev- Fig. 34.2. (Redrawn from Roemmich & Sinning 1997b.) ated LH concentrations. The lack of increase in LH secretion in adolescent wrestlers suggests a central disruption of the HPG axis, most likely a reduction receptor number. Down-regulation would result in GnRH secretion. Function of the GH–IGF-I axis in GH resistance and reduced growth, so that even was also altered with undernutrition in adolescent with increased GH concentrations, growth would wrestlers (Roemmich & Sinning 1997a) (Fig. 34.5). A be slowed to conserve energy for critical body partial GH resistance occurs as indicated by late- functions. season elevations in GH concentrations while IGF-I Although these nutritional and hormonal altera- concentrations are reduced. Decreases in serum tions have the potential of altering the growth and GHBP may also modulate serum GH concentra- maturation of wrestlers, over a sport season, the tions. GHBP enhances the growth promoting effects incremental growth in stature and rate of skeletal of GH. When GHBP concentrations are reduced, maturation of wrestlers is similar to that of controls more GH is released to continue growth at its gen- during both the sport season and post-season etically determined rate. Reductions in GHBP con- (Fig. 34.6) (Roemmich & Sinning 1996, 1997b). How- centrations also signify a down-regulation of GH ever, hypertrophic growth of fat-free mass and receptors, since GHBP is an index of GH tissue other soft tissues are slowed during the sport season

) 70 2.5 Fig. 34.2 Relative intake of –1 C D

60 C ) energy (kcal·kg–1·day–1), protein D 2.0 –1 –1 –1 –1 –1 50 (g·kg ·day ), fat (g·kg ·day ), and 1.5 carbohydrate (CHO, g·kg–1·day–1) 40 A,B B,D A,B 1.0 A,C B,D for wrestler (open circles, n = 9) and 30 A,C control (filled circles, n = 7) groups. 20 Fat (g·kg 0.5

All subjects were tested in mid- Energy (kcals·kg 10 0.0 November, 1–2 weeks prior to the first wrestling match (Early), 3.5–4.0 2.5

) C

months later (Late) depending on ) –1 D 8 C when the wrestler no longer qualified 2.0 –1 D for tournament competition, and 6 1.5 3.5–4.0 months after the wrestling A,B A,C B,D A,C B,D A,B season ended (Post). Like letters are 1.0 4 CHO (g·kg

significantly different. Values Protein (g·kg expressed as mean (± SE). (Redrawn 0.5 2 from Roemmich & Sinning 1997b.) Early Late Post Early Late Post 520 chapter 34

) ± –1 8 12 Fig. 34.4 Mean ( SE) concentrations

7 ) of testosterone, free-testosterone

–1 10 (Free-T), luteinizing hormone (LH), 6 A A 8 and sex hormone binding globulin 5 6 (SHBG) in the wrestler (open circles, 4 n = 9) and control (filled circles, A 4 = 3 LH (mIU·mL n 7) groups during the pre and late 2 2 portions of the wrestling season and Testosterone (ng·mL Testosterone during the post-season. Like letters 40 are significantly different. Serum ) ) 350 hormone concentrations are the mean –1 –1 35 325 of eight samples drawn every 20 min. 30 B B Like letters indicate p ≤ 0.05. Dashed A 200 A 25 A,B lines indicate upper and lower 20 150 normal range for each hormone 15 A,B concentration. For definitions see

Free T (pg·mL 100 DHEA-S ( µ g·dL 10 Fig. 34.2. (Redrawn from Roemmich Pre Late Post Pre Late Post & Sinning 1997a.)

Fig. 34.5 Mean (± SE) concentrations growth hormone (GH), insulin-like 10 500 growth factor I (IGF-I), growth ) )

8 –1 hormone binding protein (GHBP), –1 400 A,B A B 6 and insulin-like growth factor I 300 A B binding protein-3 (IGFBP-3) in the 4 = A,B wrestler (open circles, n 9) and 2 200 control (filled circles, n = 7) groups GH (ng·mL IGF-I (ng·mL 0 100 during the pre and late portions of the wrestling season and during the post-season. Serum hormone

400 ) 8 )

–1 concentrations are the mean of eight –1 7 samples drawn every 20 min. Like B 200 A letters are significantly different 6 and indicate p ≤ 0.05. Dashed lines indicate upper and lower 100 A,B 5 normal range for each hormone GHBP (pmol·L

0 IGFBP-3 (mg·mL 1 concentration. For definitions see Pre Late Post Pre Late Post Fig. 34.2. (Redrawn from Roemmich Test period Test period & Sinning 1997a.) followed by catch-up growth in the post-season will have an increased height percentile while late (Fig. 34.7) (Roemmich & Sinning 1996, 1997b). maturing athletes will have a decrease in height percentile. Although the average height for parti- Conclusion cipants in some sports may be less than the 50th percentile, this is probably due to self-selection rather Sport training produces acute and chronic altera- than sport training as their growth continues to tions in serum hormone concentrations but for track along the same relatively low percentile before the vast majority it does not alter the rate of genetic- and during training. Boys and girls with genetically ally programmed linear growth or pubertal devel- determined delayed puberty and/or short stature opment. Normal linear growth encompasses the would migrate towards gymnastics and wrestling third to 97th percentile for height and height velo- because their smaller size will be a disadvantage in city. Individual differences in pubertal timing may some sports, but an advantage in gymnastics and cause some athletes to temporarily track along a wrestling. Undernutrition caused by training and lower or higher percentile. Early maturing athletes dietary restriction appears to produce a disruption sport training and growth 521

175 of the HPG and a partial GH resistance in wrestlers and may also delay maturation of the reproductive axis of some elite gymnasts. The 3–4 month period 170 of undernutrition does slow hypertrophic growth of lean tissue but is followed by long periods of ade- quate nutrition so the period of undernourishment

Height (cm) 165 is not long enough to slow linear growth or matura- tion or reduce the adult height of wrestlers. Elite female gymnasts who are training and limiting their 160 dietary intake year round may be at greater risk for Pre Late Post limiting their growth. Gymnastics training may Fig. 34.6 Height for the wrestler (open circles, n = 9) slow growth, but more data are needed to confirm and control (filled circles, n = 7) groups. There were no this result. The separate effect of gymnastics train- differences in height between the groups. For definitions ing and dietary restriction on the growth, matura- see Fig. 34.2. (Redrawn from Roemmich & Sinning 1997b.) tion and neuroendocrine axes of gymnasts has not yet been adequately studied.

70 9 A A 65 8 A 60 7 A 55 B 6 B B B 50 Fat mass (kg) 5 A Body weight (kg) 45 4 14 60 Fig. 34.7 Body weight, percentage B A A body fat (% Fat), fat mass and fat free 12 55 B mass (FFM) for the wrestler (open circles, n = 9) and control (filled 10 50 circles, n = 7) groups. Like letters % Fat A C D 8 FFM (kg) 45 C,D are significantly different. Values A ± expressed as mean ( SE). For 6 40 definitions see Fig. 34.2. (Redrawn Pre Late Post Pre Late Post from Roemmich & Sinning 1997b.) Test period Test period

References van Aggel-Leijssen, D.P., van Baak, M.A., Growth and development of male Borer, K.T., Hallfrisch, J., Tsai, A.C., Tenenbaum, R., Campfield, L.A. & Saris, gymnasts, swimmers, soccer and tennis Hallfrisch, C. & Kuhns, L.R. (1979) The W.H. (1999) Regulation of average 24 h players: a longitudinal study. Annals of effect of exercise and dietary protein human plasma leptin level; the influence Human Biology 22(5), 381–394. levels on somatic growth, body of exercise and physiological changes in Blum, W.F., Englaro, P., Hanitsch, S. et al. composition, and serum lipid levels in energy balance. International Journal of (1997) Plasma leptin levels in healthy adult hamsters. Journal of Nutrition Obesity and Related Metabolic Disorders children and adolescents: dependence 109(2), 222–228. 23(2), 151–158. on body mass index, body fat mass, Borer, K.T., Nicoski, D.R. & Owens, V. Bass, S., Bradney, M., Pearce, G. et al. (2000) gender, pubertal stage, and testosterone. (1986) Alteration of pulsatile growth Short stature and delayed puberty in Journal of Clinical Endocrinology and hormone secretion by growth-inducing gymnasts: influence of selection bias Metabolism 82(9), 2904–2910. exercise: involvement of endogenous on leg length and the duration of Borer, K.T. & Kaplan, L.R. (1977) Exercise- opiates and somatostatin. Endocrinology training on trunk length. Journal of induced growth in golden hamsters: 118(2), 844–850. Pediatrics 136(2), 149–155. effects of age, weight, and activity Borst, S.E., De Hoyos, D.V., Garzarella, L. Baxter-Jones, A.D., Helms, P., Maffulli, N., level. Physiology and Behavior 18(1), et al. (2001) Effects of resistance training Baines-Preece, J.C. & Preece, M. (1995) 29–34. on insulin-like growth factor-I and IGF 522 chapter 34

binding proteins. Medicine and Science in Novakofski, J. & Bechtel, P.J. (1990) Frisch, R.E., Gotz-Welbergen, A.V., Sports and Exercise 33(4), 648–653. Activation of insulin-like growth factor McArthur, J.W. et al. (1981) Delayed Bouix, O.J., Brun, F., Fedou, C. et al. (1994) gene expression during work-induced menarche and amenorrhea of college Plasma β-endorphin, corticotrophin and skeletal muscle growth. American Journal athletes in relation to age of onset of growth hormone responses to exercise of Physiology 259(1 Pt. 1), E89–E95. training. JAMA 246(14), 1559–1563. in pubertal and prepubertal children. Dubuc, G.R., Phinney, S.D., Stern, J.S. & Fry, A.C., Kraemer, W.J., Stone, M.H. Hormone and Metabolic Research 26(4), Havel, P.J. (1998) Changes of serum et al. (1993) Endocrine and performance 195–199. leptin and endocrine and metabolic responses to high volume training and Braun, B. & Horton, T. (2001) Endocrine parameters after 7 days of energy amino acid supplementation in elite regulation of exercise substrate restriction in men and women. junior weightlifters. International Journal utilization in women compared to men. Metabolism 47(4), 429–434. of Sport Nutrition 3(3), 306–322. Exercise and Sport Sciences Reviews 29(4), Duclos, M., Corcuff, J.B., Ruffie, A., Roger, Garcia-Mayor, R.V., Andrade, M.A., 149–154. P. & Manier, G. (1999) Rapid leptin Rios, M. et al. (1997) Serum leptin levels Bullen, B.A., Skrinar, G.S., Beitins, I.Z. et al. decrease in immediate post-exercise in normal children: relationship to age, (1985) Induction of menstrual disorders recovery. Clinical Endocrinology 50(3), gender, body mass index, pituitary– by strenuous exercise in untrained 337–342. gonadal hormones, and pubertal stage. women. New England Journal of Medicine Eliakim, A., Brasel, J.A., Mohan, S. et al. Journal of Clinical Endocrinology and 312(21), 1349–1353. (1996) Physical fitness, endurance Metabolism 82(9), 2849–2855. Caine, D., Lewis, R., O’Connor, P., training, and the growth hormone– Ghigo, E., Bellone, J., Aimaretti, G. et al. Howe, W. & Bass, S. (2001) Does insulin-like growth factor I system in (1996) Reliability of provocative tests to gymnastics training inhibit growth of adolescent females. Journal of Clinical assess growth hormone secretory status. females? Clinical Journal of Sport Medicine Endocrinology and Metabolism 81(11), Study in 472 normally growing children. 11(4), 260–270. 3986–3992. Journal of Clinical Endocrinology and Cappon, J., Brasel, J.A., Mohan, S. & Eliakim, A., Brasel, J.A., Mohan, S., Wong, Metabolism 81(9), 3323–3327. Cooper, D.M. (1994) Effect of brief W.L. & Cooper, D.M. (1998a) Increased Haapasalo, H., Kannus, P., Sievanen, H. exercise on circulating insulin-like physical activity and the growth et al. (1998) Effect of long-term unilateral growth factor I. Journal of Applied hormone–IGF-I axis in adolescent males. activity on bone mineral density of Physiology 76(6), 2490–2496. American Journal of Physiology 275(1 Pt. female junior tennis players. Journal Clayton, P.E., Gill, M.S., Hall, C.M. et al. 2), R308–R314. of Bone and Mineral Research 13(2), (1997) Serum leptin through childhood Eliakim, A., Brasel, J.A., Barstow, T.J., 310–319. and adolescence. Clinical Endocrinology Mohan, S. & Cooper, D.M. (1998b) Peak Hackney, A.C. (1996) The male 46(6), 727–733. oxygen uptake, muscle volume, and the reproductive system and endurance Conn, C.A., Mial, S. & Borer, K.T. (1993) growth hormone–insulin-like growth exercise. Medicine and Science in Sports Food restriction postpones, not factor-I axis in adolescent males. and Exercise 28(2), 180–189. prevents, exercise-induced growth in Medicine and Science in Sports and Exercise Hackney, A.C. (2001) Endurance exercise hamsters. Growth, Development, and 30(4), 512–517. training and reproductive endocrine Aging 57(3), 193–204. Eliakim, A., Scheett, T.P., Newcomb, R., dysfunction in men: alterations in the Consitt, L.A., Copeland, J.L. & Tremblay, Mohan, S. & Cooper, D.M. (2001) hypothalamic–pituitary–testicular axis. M.S. (2001) Hormone responses to Fitness, training, and the growth Current Pharmaceutical Design 7(4), resistance vs. endurance exercise in hormone–insulin-like growth factor I 261–273. premenopausal females. Canadian axis in prepubertal girls. Journal of Hackney, A.C., Szczepanowska, E. & Viru, Journal of Applied Physiology 26(6), Clinical Endocrinology and Metabolism A.M. (2003) Basal testicular testosterone 574–587. 86(6), 2797–2802. production in endurance-trained men Constantini, N.W.& Warren, M.P. (1994) Essig, D.A., Alderson, N.L., Ferguson, is suppressed. European Journal of Applied Special problems of the female athlete. M.A., Bartoli, W.P. & Durstine, J.L. Physiology 89(2), 198–201. Bailliere’s Clinical Rheumatology 8(1), (2000) Delayed effects of exercise on the Häkkinen, K. & Pakarinen, A. (1995) Acute 199–219. plasma leptin concentration. Metabolism hormonal responses to heavy resistance Cunningham, M.J., Clifton, D.K. & Steiner, 49(3), 395–399. exercise in men and women at different R.A. (1999) Leptin’s actions on the Fahey, T.D., Del Valle-Zuris, A., Oehlsen, ages. International Journal of Sports reproductive axis: perspectives and G., Trieb, M. & Seymour, J. (1979) Medicine 16(8), 507–513. mechanisms. Biology of Reproduction Pubertal stage differences in hormonal Häkkinen, K., Pakarinen, A., Kraemer, 60(2), 216–222. and hematological responses to maximal W.J. et al. (2001) Selective muscle Deschenes, M.R. & Kraemer, W.J. (2002) exercise in males. Journal of Applied hypertrophy, changes in EMG and force, Performance and physiologic Physiology 46(4), 823–827. and serum hormones during strength adaptations to resistance training. Farooqi, I.S., Jebb, S.A., Langmack, G. et al. training in older women. Journal of American Journal of Physical Medicine and (1999) Effects of recombinant leptin Applied Physiology 91(2), 569–580. Rehabilitaion 81 (suppl. 11), S3–S16. therapy in a child with congenital leptin Harber, V.J. (2000) Menstrual dysfunction De Souza, M.J., Van Heest, J., Demers, L.M. deficiency. New England Journal of in athletes: an energetic challenge. & Lasley, B.L. (2003) Luteal phase Medicine 341(12), 879–884. Exercise and Sport Sciences Reviews 28(1), deficiency in recreational runners: Foster, D.L. & Nagatani, S. (1999) 19–23. evidence for a hypometabolic state. Physiological perspectives on leptin as a Hilton, L.K. & Loucks, A.B. (2000) Low Journal of Clinical Endocrinology and regulator of reproduction: role in timing energy availability, not exercise stress, Metabolism 88(1), 337–346. puberty. Biology of Reproduction 60(2), suppresses the diurnal rhythm of leptin DeVol, D.L., Rotwein, P., Sadow, J.L., 205–215. in healthy young women. American sport training and growth 523

Journal of Physiology. Endocrinology and and growth factor responses to heavy growth, maturation, and fitness of Polish Metabolism 278(1), E43–E49. resistance exercise in males and females. youth active in sport. International Horswill, C.A., Park, S.H. & Roemmich, International Journal of Sports Medicine Journal of Sports Medicine 18 (suppl. 3), J.N. (1990) Changes in the protein 12(2), 228–235. S179–S185. nutritional status of adolescent Kraemer, W.J., Fry, A.C., Warren, B.J. et al. Mantzoros, C.S., Flier, J.S. & Rogol, A.D. wrestlers. Medicine and Science in Sports (1992) Acute hormonal responses in elite (1997) A longitudinal assessment of and Exercise 22(5), 599–604. junior weightlifters. International Journal hormonal and physical alterations Jenkins, P.J. (1999) Growth hormone and of Sports Medicine 13(2), 103–109. during normal puberty in boys. V. exercise. Clinical Endocrinology 50(6), Kraemer, W.J., Fleck, S.J., Dziados, J.E. Rising leptin levels may signal the 683–689. et al. (1993) Changes in hormonal onset of puberty. Journal of Clinical Jurkowski, J.E., Jones, N.L., Walker, C., concentrations after different heavy- Endocrinology and Metabolism 82(4), Younglai, E.V. & Sutton, J.R. (1978) resistance exercise protocols in women. 1066–1070. Ovarian hormonal responses to exercise. Journal of Applied Physiology 75(2), Marin, G., Domene, H.M., Barnes, K.M. Journal of Applied Physiology 44(1), 594–604. et al. (1994) The effects of estrogen 109–114. Kraemer, W.J., Staron, R.S., Hagerman, priming and puberty on the growth Kannus, P., Haapasalo, H., Sankelo, M. F.C. et al. (1998) The effects of short-term hormone response to standardized et al. (1995) Effect of starting age of resistance training on endocrine treadmill exercise and arginine-insulin physical activity on bone mass in the function in men and women. European in normal girls and boys. Journal of dominant arm of tennis and squash Journal of Applied Physiology and Clinical Endocrinology and Metabolism players. Annals of Internal Medicine Occupational Physiology 78(1), 69–76. 79(2), 537–541. 123(1), 27–31. Kraemer, W.J., Häkkinen, K., Newton, R.U. Marx, J.O., Ratamess, N.A., Nindl, B.C. Kannus, P., Sievanen, H. & Vuori, I. (1996) et al. (1999) Effects of heavy-resistance et al. (2001) Low-volume circuit versus Physical loading, exercise, and bone. training on hormonal response patterns high-volume periodized resistance Bone 18 (suppl. 1), 1S–3S. in younger vs. older men. Journal of training in women. Medicine and Science Karamouzis, I., Karamouzis, M., Vrabas, Applied Physiology 87(3), 982–992. in Sports and Exercise 33(4), 635–643. I.S. et al. (2002) The effects of marathon Lanyon, L.E. (1996) Using functional Matkovic, V., Ilich, J.Z., Skugor, M. et al. swimming on serum leptin and plasma loading to influence bone mass and (1997) Leptin is inversely related to age neuropeptide Y levels. Clinical Chemistry architecture: objectives, mechanisms, at menarche in human females. Journal of and Laboratory Medicine 40(2), 132–136. and relationship with estrogen of the Clinical Endocrinology and Metabolism Keizer, H.A., Kuipers, H., de Haan, J., mechanically adaptive process in bone. 82(10), 3239–3245. Beckers, E. & Habets, L. (1987) Multiple Bone 18 (suppl. 1), 37S–43S. Mauras, N. (2001) Growth hormone and hormonal responses to physical exercise Lindholm, C., Hagenfeldt, K. & Ringertz, sex steroids. Interactions in puberty. in eumenorrheic trained and untrained B.M. (1994) Pubertal development in Endocrinology and Metabolism Clinics of women. International Journal of Sports elite juvenile gymnasts. Effects of North America 30(3), 529–544. Medicine 8 (suppl. 3), 139–150. physical training. Acta Obstetricia et Miller, K.K., Parulekar, M.S., Schoenfeld, Koziris, L.P., Hickson, R.C., Chatterton, Gynecologica Scandinavica 73(3), 269–273. E. et al. (1998) Decreased leptin levels R.T., Jr. et al. (1999) Serum levels of Loucks, A.B. & Thuma, J.R. (2003) in normal weight women with total and free IGF-I and IGFBP-3 are Luteinizing hormone pulsatility is hypothalamic amenorrhea: the effects increased and maintained in long-term disrupted at a threshold of energy of body composition and nutritional training. Journal of Applied Physiology availability in regularly menstruating intake. Journal of Clinical Endocrinology 86(4), 1436–1442. women. Journal of Clinical Endocrinology and Metabolism 83(7), 2309–2312. Kraemer, R.R., Heleniak, R.J., Tryniecki, and Metabolism 88(1), 297–311. Moschos, S., Chan, J.L. & Mantzoros, C.S. J.L. et al. (1995) Follicular and luteal Loud, K.J. & Micheli, L.J. (2001) Common (2002) Leptin and reproduction: a phase hormonal responses to low- athletic injuries in adolescent girls. review. Fertility and Sterility 77(3), volume resistive exercise. Medicine and Current Opinion in Pediatrics 13(4), 433–444. Science in Sports and Exercise 27(6), 317–322. Nicklas, B.J., Hackney, A.C. & Sharp, R.L. 809–817. McCall, G.E., Byrnes, W.C., Fleck, S.J., (1989) The menstrual cycle and exercise: Kraemer, R.R., Acevedo, E.O., Synovitz, Dickinson, A. & Kraemer, W.J. (1999) performance, muscle glycogen, and L.B. et al. (2001) Leptin and steroid Acute and chronic hormonal responses substrate responses. International Journal hormone responses to exercise in to resistance training designed to of Sports Medicine 10(4), 264–269. adolescent female runners over a 7-week promote muscle hypertrophy. Canadian Olive, J.L. &. Miller, G.D. (2001) season. European Journal of Applied Journal of Applied Physiology 24(1), 96–107. Differential effects of maximal- and Physiology 86(1), 85–91. Malina, R.M. (1994) Physical activity and moderate-intensity runs on plasma Kraemer, R.R., Chu, H. & Castracane, V.D. training: effects on stature and the leptin in healthy trained subjects. (2002) Leptin and exercise. Experimental adolescent growth spurt. Medicine and Nutrition 17(5), 365–369. Biology and Medicine (Maywood, NJ) Science in Sports and Exercise 26(6), Peltenburg, A.L., Erich, W.B., Zonderland, 227(9), 701–708. 759–766. M.L. et al. (1984) A retrospective growth Kraemer, W.J., Marchitelli, L., Gordon, S.E. Malina, R.M. & Bielicki, T. (1996) study of female gymnasts and girl et al. (1990) Hormonal and growth factor Retrospective longitudinal growth study swimmers. International Journal of responses to heavy resistance exercise of boys and girls active in sport. Acta Sports Medicine 5(5), 262–267. protocols. Journal of Applied Physiology Paediatrica 85(5), 570–576. Perusse, L., Collier, G., Gagnon, J. et al. 69(4), 1442–1450. Malina, R.M., Woynarowska, B., (1997) Acute and chronic effects of Kraemer, W.J., Gordon, S.E., Fleck, S.J. et al. Bielicki, T. et al. (1997) Prospective and exercise on leptin levels in humans. (1991) Endogenous anabolic hormonal retrospective longitudinal studies of the Journal of Applied Physiology 83(1), 5–10. 524 chapter 34

Poehlman, E.T. & Copeland, K.C. (1990) Rowland, T.W., Morris, A.H., Kelleher, Clinical Endocrinology and Metabolism Influence of physical activity on insulin- J.F., Haag, B.L. & Reiter, E.O. (1987) 82(10), 3414–3420. like growth factor-I in healthy younger Serum testosterone response to training Veldhuis, J.D., Roemmich, J.N. & and older men. Journal of Clinical in adolescent runners. American Journal Rogol, A.D. (2000) Gender and sexual Endocrinology and Metabolism 71(6), of Diseases of Children 141(8), 881–883. maturation-dependent contrasts in the 1468–1473. Sabatini, S. (2001) The female athlete triad. neuroregulation of growth hormone Pullinen, T., Mero, A., Huttunen, P., American Journal of the Medical Sciences secretion in prepubertal and late Pakarinen, A. & Komi, P.V. (2002) 322(4), 193–195. adolescent males and femalesaa general Resistance exercise-induced hormonal Scheett, T.P., Nemet, D., Stoppani, J. et al. clinical research center-based study. responses in men, women, and (2002) The effect of endurance-type Journal of Clinical Endocrinology and pubescent boys. Medicine and Science in exercise training on growth mediators Metabolism 85(7), 2385–2394. Sports and Exercise 34(5), 806–813. and inflammatory cytokines in pre- Viru, A., Laaneots, L., Karelson, K., Racette, S.B., Coppack, S.W., Landt, M. pubertal and early pubertal males. Smirnova, T. & Viru, M. (1998) Exercise- & Klein, S. (1997) Leptin production Pediatric Research 52(4), 491–497. induced hormone responses in girls at during moderate-intensity aerobic Simon, M.R. (1978) The effect of dynamic different stages of sexual maturation. exercise. Journal of Clinical Endocrinology loading on the growth of epiphyseal European Journal of Applied Physiology and and Metabolism 82(7), 2275–2277. cartilage in the rat. Acta Anatomica Occupational Physiology 77(5), 401–408. Roemmich, J.N. & Rogol, A.D. (1995) 102(2), 176–183. Wadden, T.A., Considine, R.V., Foster, Physiology of growth and development. Simon, M.R. & Papierski, P. (1982) Effects G.D. et al. (1998) Short- and long-term Its relationship to performance in the of experimental bipedalism on the changes in serum leptin dieting obese young athlete. Clinics in Sports Medicine growth of the femur and tibia in normal women: effects of caloric restriction 14(3), 483–502. and hypophysectomized rats. Acta and weight loss. Journal of Clinical Roemmich, J.N. & Rogol, A.D. (1997) Anatomica 114(4), 321–329. Endocrinology and Metabolism 83(1), Exercise and growth hormone: does one Smith, E.L. & Gilligan, C. (1996) 214–218. affect the other? Journal of Pediatrics Dose–response relationship between Warren, M.P. (1980) The effects of 131(1 Pt. 2), S75–S80. physical loading and mechanical exercise on pubertal progression and Roemmich, J.N. & Sinning, W.E. (1996) competence of bone. Bone 18 (suppl. 1), reproductive function in girls. Journal of Sport-seasonal changes in body 455–505. Clinical Endocrinology and Metabolism composition, growth, power and Theintz, G.E. (1994) Endocrine adaptation 51(5), 1150–1157. strength of adolescent wrestlers. to intensive physical training during Weimann, E. (2002) Gender-related International Journal of Sports Medicine growth. Clinical Endocrinology 41(3), differences in elite gymnasts: the female 17(2), 92–99. 267–272. athlete triad. Journal of Applied Physiology Roemmich, J.N. & Sinning, W.E. (1997a) Theintz, G.E., Howald, H., Allemann, Y. 92(5), 2146–2152. Weight loss and wrestling training: & Sizonenko, P.C. (1989) Growth and Weltman, A., Weltman, J.Y., Schurrer, R. effects on growth-related hormones. pubertal development of young female et al. (1992) Endurance training amplifies Journal of Applied Physiology 82(6), gymnasts and swimmers: a correlation the pulsatile release of growth hormone: 1760–1764. with parental data. International Journal effects of training intensity. Journal of Roemmich, J.N. & Sinning, W.E. (1997b) of Sports Medicine 10(2), 87–91. Applied Physiology 72(6), 2188–2196. Weight loss and wrestling training: Theintz, G.E., Howald, H., Weiss, U. & Weltman, A., Pritzlaff, C.J., Wideman, L. effects on nutrition, growth, maturation, Sizonenko, P.C. et al. (1993) Evidence et al. (2000) Intensity of acute exercise body composition, and strength. Journal for a reduction of growth potential in does not affect serum leptin of Applied Physiology 82(6), 1751–1759. adolescent female gymnasts. Journal of concentrations in young men. Medicine Roemmich, J.N., Clark, P.A., Mai, V. et al. Pediatrics 122(2), 306–313. and Science in Sports and Exercise 32(9), (1998) Alterations in growth and body Thong, F.S., McLean, C. & Graham, T.E. 1556–1561. composition during puberty: III. (2000) Plasma leptin in female athletes: Williams, N.I., Helmreich, D.L., Parfitt, Influence of maturation, gender, body relationship with body fat, reproductive, D.B., Caston-Balderrama, A. & composition, fat distribution, aerobic nutritional, and endocrine factors. Cameron, J.L. (2001) Evidence for a fitness, and energy expenditure on Journal of Applied Physiology 88(6), causal role of low energy availability nocturnal growth hormone release. 2037–2044. in the induction of menstrual cycle Journal of Clinical Endocrinology and Torun, B. & Viteri, F.E. (1994) Influence of disturbances during strenuous exercise Metabolism 83(5), 1440–1447. exercise on linear growth. European training. Journal of Clinical Endocrinology Rogol, A.D. & Roemmich, J.N. (2000) Journal of Clinical Nutrition 48 (suppl. 1), and Metabolism 86(11), 5184–5193. Puberty as a growth promoted event. S186–S189. Yan, Z., Biggs, R.B. & Booth, F.W. In: The Onset of Puberty (Bourguignon, Veldhuis, J.D., Metzger, D.L., Martha, (1993) Insulin-like growth factor J.P. & Plant, T.M., eds.). Elsevier Science, P.M., Jr. et al. (1997) Estrogen and immunoreactivity increases in muscle Amsterdam: 97–111. testosterone, but not a nonaromatizable after acute eccentric contractions. Journal Rosendal, L., Langberg, H., Flyvbjerg, A. androgen, direct network integration of of Applied Physiology 74(1), 410–414. et al. (2002) Physical capacity influences the hypothalamo-somatotrope (growth Yeager, K.K., Agostini, R., Nattiv, A. & the response of insulin-like growth hormone)–insulin-like growth factor I Drinkwater, B. (1993) The female athlete factor and its binding proteins to axis in the human: evidence from triad: disordered eating, amenorrhea, training. Journal of Applied Physiology pubertal pathophysiology and sex- osteoporosis. Medicine and Science in 93(5), 1669–1675. steroid hormone replacement. Journal of Sports and Exercise 25(7), 775–777. Chapter 35

Effects of Testosterone and Related Androgens on Athletic Performance in Men

KARL E. FRIEDL

In spite of quantitative differences in the gene that limiting mechanisms simply will not allow products of tissues from males and females, open-ended or cumulative effects to occur (e.g. there are few important sexual differences in changes in receptor density and binding activity, the genome, with the exception of the male- negative feedback loops and non-linear interac- specific Y chromosome. The major function of tions with other hormones) (Glass & Vigersky 1980). this chromosome is to determine sexual differ- In this case, observed effects from supraphysiolo- entiation; in its absence, ovaries develop. It is, gical doses of androgens may contrast those that then, the secretory products of the gonads that occur in nature, producing pharmacological effects are responsible for sexual dimorphism....As a that are very different than those associated with consequence of its widespread action, testo- male characteristics (e.g. stroke risk, gynecomastia, sterone is the major hormone responsible for the osmoregulatory changes). For practical purposes, sexual dimorphism of nonreproductive tissues. supraphysiological could be defined as a dose that Bardin & Catterall 1981, p. 1285 increases circulating levels of testosterone above the normal range, or other androgen formulations with Introduction rough bioequivalence to these above-normal testo- sterone levels (even though measured testosterone Although gender differences in body size and upper may be markedly suppressed); in some regards, any body strength may have been evident for centuries, administration of exogenous androgen is supra- the basis of these differences is still poorly under- physiological because it does not precisely mimic stood. This ignorance hampers our ability to predict circadian fluctuations and other properties of the the effects of supplemental androgens admin- body’s own hormonal secretions. In the absence of istered to normal adult males for performance data, speculative and anecdotal self-experiments enhancement. Androgens administered to andro- by athletes have led to widespread gym lore gen-deficient men, prepubertal boys and normal about androgen use, and non-users are left with women can produce marked effects on size and concerns about unfair performance advantages. A strength (Stanhope et al. 1988; Bhasin et al. 1997; greater understanding of androgen physiology can Elbers et al. 1999), but it does not necessarily follow improve the scientific basis of training regimens for that an excess of androgens to normal men will pro- athletes and help define the full range of modifiable vide more of the same effects along the same contin- human performance. This chapter considers the uum. There is no specific disease of androgen excess available data on testosterone and related andro- in males, perhaps suggesting that more androgen genic steroid effects on athletic performance and does indeed produce more of the same biological attempts to expand on earlier thoughtful reviews effects (Can a man be over-virilized?) (Bardin et al. (Wright 1980; Haupt & Rovere 1984; Wilson 1988; 1990). Alternatively, it could be reasonably expected Yesalis 2000). 525 526 chapter 35

Metabolism and action of testosterone as intra-abdominal fat metabolism, regulation of and related androgenic steroids key hepatic proteins, bone mineral metabolism and some brain effects (e.g. regulation of gonadotropin The basic physiology of testosterone production secretion). In muscle cells, testosterone appears to and secretion is relatively well known, while the far- act directly on androgen receptors, which are pre- reaching effects of testosterone on cells throughout sent in much lower density than in more androgen- the body, as well as interactions with other anabolic responsive tissues such as prostate, and then is hormones, are not as well understood (Bardin & excreted as a water-soluble conjugate, 3α-andro- Catterall 1981). Testosterone is a steroid hormone stanediol glucuronide; muscle lacks the 5α-reductase that is made from cholesterol through a series of to convert testosterone to DHT (Michel & Baulieu enzyme-regulated steps in the Leydig cells of the 1980; Hughes & Krieg 1988). Androgens, including testes of men, and in the adrenal cortex of men and testosterone and DHT, are bound to a variety of women. The key regulator of the testicular secretion carrier proteins in circulation, most specifically to is luteinizing hormone (LH), a protein hormone sex hormone binding globulin (SHBG), but also from the pituitary gland. LH secretion, in turn, non-specifically to albumin. This constitutes a large is regulated by gonadotropin-releasing hormone circulating ‘reserve’ pool of immediately available (GnRH), which is regulated in the hypothalamus. steroids that has some protection against metabol- LH and GnRH secretion are controlled, in part, by ism and excretion while bound, acting in a kinetic circulating levels of testosterone in a negative feed- balance with receptors and competitive ligands that back loop. If testosterone or a related androgenic may vary between tissues, with only 1–2% unbound compound is administered artificially, the system (or ‘free’) at any time. Testosterone activates the down-regulates, reducing the amount of testo- secretion of other potent anabolic hormones such as sterone synthesized by the testes. If higher than insulin-like growth factor I (IGF-I) and erythro- normal circulating levels are sustained by adminis- poeitin, and potential benefits to athletic perform- tration of androgen, the testes substantially reduce ance may accrue from these indirect effects of testosterone and sperm production, reducing the androgen. Localized effects of testosterone, includ- volume of the testes as the seminiferous tubules ing such interactions at the tissue levels, may be are depopulated of maturing germ cells (a readily very important and are just beginning to be appre- reversible phenomenon). This noticeable reduction ciated with the availability of modern research in testicular volume (Kiraly 1988; Friedl et al. 1991) methods such as transgenic animal studies. Testo- is one of the androgen effects that make blinded sterone can also bind to other receptors such as studies in androgen-experienced athletes difficult glucocorticoids, in this case providing competitive to conduct. inhibition of certain catabolic stress responses Testosterone’s effects on body tissues are far induced by cortisol; thus, part of the anabolic action more complicated than its production and secretion. of testosterone may actually be exerted through this Testosterone is directly active in most tissues, acting crossover binding to modify catabolic responses through a specific androgen receptor. Additionally, (Hickson et al. 1984; Janne 1990). in some tissues such as the prostate and hair fol- Pharmacological compounds that have been licles, testosterone is converted to 5α-dihydrotes- developed to extend or modify the effect of testo- tosterone (DHT). DHT has greater potency than sterone build on some of these understood effects testosterone on androgen receptors, but is also more of androgens (Liddle & Burke 1960; Kruskemper rapidly metabolized to biologically inactive 5α- 1968; Kochakian 1988). Thus, steroids that cannot be androstanediols. In fat cells, where high concentra- converted to estrogenic compounds or to DHT will tions of aromatase enzyme are present, some of the exhibit characteristic effects different from those testosterone is converted to estrogen, with estrogen of testosterone, with reduced effects on bone or actually playing important roles in the mediation of prostate, respectively (Sundaram et al. 1995). Some apparent testosterone actions in some tissues such steroid-using athletes include aromatase inhibitors androgens and performance 527

in the drug regimen to block conversion to estro- developed to try to reduce androgenic side effects gens, or use estrogen receptor blockers to prevent on prostate through its properties that prevent estrogenic actions, in attempts to prevent gyneco- aromatization to estrogens and its metabolism to mastia and other side effects (Friedl & Yesalis 1989). compounds that are weaker than DHT (e.g. 19- Most androgenic-anabolic steroids have been devel- nordihydrotestosterone) (Hershberger et al. 1953; oped to reduce the ‘androgenic’ effects on reproduct- Sundaram et al. 1995). Nandrolone esters (e.g. nan- ive tissues such as prostate, in favor of ‘anabolic’ drolone decanoate or ‘Deca-Durabolin’) and testo- effects on muscle and bone; these would focus on sterone esters (e.g. TE) are the most commonly used steroid structures that cannot be readily reduced androgenic preparations today, and these are the to DHT (Hershberger et al. 1953). Testosterone compounds that have been used in the majority of administered either orally or by injection promptly modern studies on performance effects. A wide reaches circulation and is rapidly removed by variety of androgens are active in muscle, while the liver and metabolized to products that can be requiring greater selectivity for some other tissues; readily excreted in urine or bile. Modifications with for example, protection of bone against glucocor- 17-alkylation protect testosterone against this rapid ticoid effects requires an aromatizable androgen, elimination, making methyl testosterone an orally while nandrolone compounds are effective in pre- active androgen, but the metabolites produced in serving muscle but not bone (Crawford et al. 2003). various tissues are 17-alkylated compounds that Any administration of exogenous steroid is non- will exhibit different binding properties and bio- physiological, even at doses that restore mean logical effects. Adverse effects on the liver are one circulating levels to ‘normal’, as there is a large of the properties inherent in the orally active 17- fluctuation in normal circulating levels of testo- alkylated androgens. Methandienone (Dianabol) is sterone during the day. the most important orally active 17-alkylated andro- Dietary supplements that are legally marketed as gen in the studies of athletic performance (Friedl ‘steroid replacers’ in the USA include a wide range 2000). This was the orally active drug of choice by of products (e.g. sterols, sapogenins, androgen steroid-using athletes in the 1960s and 1970s but metabolites, yohimbe bark, boron), none of which has been since taken off the market by major drug have any convincing scientific basis for perform- manufacturers (Yesalis et al. 1988). Modifications of ance enhancement in normal men. Some products testosterone with an ester side chain also protects have been tainted with banned substances that against rapid elimination, but because this side will give a positive drug test; others are simply chain can be hydrolyzed to provide graded release bogus and even harmful (Friedl et al. 1992b). There of testosterone into circulation from deep intra- is no over-the-counter drug that comes close to the muscular injection sites, the effects are different actions of testosterone, or significantly stimulates its than those of the 17-alkylated compounds. The production or release in the body (King et al. 1999). If duration of action of these esterified androgens is such an effective product surfaced, it would likely roughly dependent on the length of the side chain; a be added to the list of androgens and other sub- relatively short side chain in testosterone unde- stances banned by athletic federations. canoate makes this a relatively short acting testo- Most recently, new ‘designer’ steroids have been sterone (usually used orally) (Schurmeyer et al. 1983), detected in urinalyses from elite athletes (Catlin while high levels of testosterone can be maintained et al. 2002). Tetrahydrogestrinone is a compound with testosterone enanthate (TE) or testosterone related to gestrinone, a steroid used to treat en- cypionate in weekly injections, and the very long dometriosis in women, and is the latest to turn up in chained testosterone buciclate maintains circulating drug testing in a wide variety of sports including levels for months (Behre & Nieschlag 1992). A com- track and field, football and boxing (Knight 2003). pound that differs from testosterone only by the The novelty is not in some new marvelous mech- elimination of one methyl group at the 19 carbon anism of action, but rather in the fact that the position (’19-nortestosterone’ or nandrolone) was compound was new to human use and therefore not 528 chapter 35

specifically tested for in routine drug screens. larger than those observed in athletes in training, Other synthetic steroids have been administered in can be explained by inadequate energy intake and attempts to correct the abnormal profile of andro- not as an exercise-specific effect (Friedl 1997). As it gen metabolites that signal exogenous use of andro- turns out, even the concept of a female athletic gens (Aguilera et al. 2002). The use of new drugs amenorrhea produced by high volume exercise may with little or no evaluation of human safety should not be correct; earlier studies were confounded by be of great concern to normal healthy athletes. under-reporting food intake. As observed in men, For purposes of this chapter, anabolic steroids are energy deficit, not exercise, appears to be the key used to denote synthetic androgenic steroids other stressor affecting reproductive cycles in women than the testosterone compounds, but are used (Loucks & Thuma 2003; Loucks 2004). somewhat interchangeably, as anabolic and andro- Psychological factors also affect testosterone genic properties are notional and have never been levels in healthy men. The thrill of victory may pro- effectively separated (Kochakian 1988). duce short-term increases in circulating testosterone levels (Ursine et al. 1978; Elias 1981), while anxiety Is there a male athlete hypogonadal produces reliable declines (Bourne et al. 1968; Kreuz equivalent to athletic amenorrhea? et al. 1972; Francis 1981). Mood and motivation in human research is difficult to control or ethically manipulate. However, in one novel study of men in Environmental regulators of testosterone parachute training making a stressful jump out of a secretion high tower, the initial jumps resulted in a marked Female athletes have been commonly warned that stress effect with acute suppression of testosterone reproductive cycles may cease with too much exer- and elevation of cortisol (‘distress’), while later cise, as part of a ‘syndrome’ involving exercise, experienced jumps led to acute elevation of testo- amenorrhea and bone mineral loss. Extrapolations sterone and no change in cortisol (‘eustress’) (Ursine from the theory of an athletic amenorrhea in men et al. 1978). There is a large body of literature on the suggest that with high volume physical training, effects of defeat and chronic stress on testosterone testosterone will be suppressed in male athletes. suppression in animals that suggests this is a key Furthermore, it has been suggested that these regulator of normal testosterone in mammalian athletes might benefit from ‘replacement’ androgen species (Blanchard et al. 1993). In humans, chronic therapy. This interpretation may have started with anxiety appears to have a prominent suppressive military training studies in which marked reduc- effect on testosterone and may even provide a more tions in testosterone were noted in healthy young reliable indicator of a noxious stress than traditional men (Aakvaag et al. 1978). In normal men working measures of the hypothalamic–pituitary–adrenal hard with inadequate energy intake, testosterone axis such as cortisol (Kreuz et al. 1972; Francis 1981; can fall nearly to castrate levels. This has been shown Hellhammer et al. 1985). in high school wrestlers during their wrestling season while they are energy-restricted (Strauss Effects of resistance and endurance exercise et al. 1985; Roemmich & Sinning 1997) and in elite on testosterone soldiers during short- and long-term intensive train- ing (Opstad 1992a; Friedl et al. 2000). Guezennec Exercise itself does not appear to be an important et al. (1994) demonstrated a testosterone decline regulator of testosterone secretion through either with increasing energy deficit in French com- primary (actions at the testes) or higher levels mandos. However, testosterone levels are rapidly (actions at the pituitary or hypothalamus), although restored to normal with refeeding, even in the face little is known about the neurobiology of exercise of other persistent stressors such as workloads (Rogol et al. 1984), or any direct signals from exer- exceeding 25 MJ·day–1 (6000 kcal·day–1) (Friedl et al. cising muscle that may affect the gonadal axis. 2001). Thus, these large reductions, generally much Endurance athletes have been variously reported androgens and performance 529

% Ergometer exercise 20 Running events * 10 Skiing events min 50 100 150 200 250 300 350 400 450 500 Fig. 35.1 Apparent relationship –10 between duration of exercise and * serum testosterone from various –20 studies, suggesting a role for a –30 progressive metabolic stimulus such * as declining glucose availability on –40 regulation of testosterone secretion. * (From Viru 1985.) –50

to have modest reductions in resting testosterone hypogonadism that would normally justify treat- levels (Wheeler et al. 1984; De Souza et al. 1994), ment. Mechanisms for this change with prolonged or no differences (Bagatell & Bremner 1990; Lucia exercise have been variously postulated and invest- et al. 1996), compared to sedentary controls. Male igated. Catecholamine effects on testicular blood runners reducing their weekly mileage from flow (Collu et al. 1984) and cortisol suppression 81 km·wk–1 to 24 km·wk–1 demonstrated no change of testosterone production (Bambino & Hsueh in low, normal testosterone levels, even though 1981) do not seem to occur at physiological levels creatine kinase levels were reduced by half, reflect- (Sapolsky 1985, 1986). Increased clearance rates ing a reduction in muscular strain (Houmard et al. of testosterone by exercising muscle (e.g. increased 1994). The biological significance of this altered excretion of 3α-androstenedione-glucuronide) ap- status to low-normal circulating levels observed in pears to occur (Ponjee et al. 1994), but could be con- some studies is unknown. strued as an increased biological action reflected in Exercise can acutely increase or decrease cir- greater turnover at the muscle target site, rather culating testosterone, depending on the mode and than a deficiency state. It may be that this transient intensity of exercise (Schmid et al. 1982; Viru 1985). decline in testosterone that appears to increase with Increases occur during and after relatively short, duration of exercise (Fig. 35.1) is explained simply high intensity work such as resistance training or by a within-exercise energy deficit, highlighting sprint events, while declines are associated with the sensitivity of GnRH to low glucose levels in increasing duration endurance events and are espe- the hypothalamus (Opstad 1992b; Loucks 2004). A cially noted in marathon and ultra-running events. similar effect of a progressive ‘starvation’ profile Endurance exercise has been reported to decrease has been suggested for changes in thyroid hormone testosterone by as much as half, even after appro- levels during prolonged exercise (O’Connell et al. priate comparison to same time-of-day baseline 1979). Some extreme endurance athletes recognize values (Dessypris et al. 1976; Morville et al. 1979; the need for liquid carbohydrate replenishment Kuoppasalmi et al. 1980; Schurmeyer et al. 1984). during their exercise (Saris et al. 1989), although These effects observed during or shortly after the whether or not this better sustains testosterone exercise period, may persist into the next day or two levels and muscle mass has not been studied. after cessation of exercise. The physiological con- In contrast to prolonged endurance exercise, sequences of this short-term suppression are uncer- resistance and sprint exercise usually produces tain; this decline is comparable to the changes short-term elevation in testosterone concentration observed daily in normal men, and the observed (Sutton et al. 1973; Cumming et al. 1986; Kraemer reductions still do not fall into the range of a clinical et al. 1991), or no change observed (Guezennec et al. 530 chapter 35

1986). Several mechanisms would be reasonably ences in repetitions and duration of rest periods expected to temporarily increase serum concentra- between sets demonstrated large differences in tions. Liver blood flow markedly declines with growth hormone responses, but a minimal differ- increasing intensity of exercise (Rowell et al. 1964) ence in the magnitude of testosterone increases and would reduce hepatic clearance and temporar- in either routine. One can only speculate on the ily increase serum concentrations. Fluid shifts with significance of brief testosterone changes in either movement out of the vascular space during exercise direction to synergistic actions with other mus- (Wilkerson et al. 1980) would increase the con- culotrophic hormones such as local tissue activity centration of all protein-bound hormones such as of IGF-I, including in the final determination of testosterone, which cannot move with the fluid. muscle hypertrophy under the influence of strength Dopamine and prolactin may regulate testosterone demands for strength athletes and the slimming release, but in one study, testosterone changes were down of muscle mass that would be advantageous unaffected when the exercise-induced rise in pro- for distance runners. lactin was blocked by l-dopa (Jezova-Repcekova et al. 1982). Sympathetic activity may also play a role Effects on muscle mass: how big can in exercise-induced testosterone increase (Jezova & humans get? Vigas 1981). Increases in total body weight Effects of endogenous testosterone changes Androgen supplementation to normal men increases on performance body weight and amino acid incorporation into Temporary exercise-induced declines in testosterone muscle protein. The precise nature of the observed may not have much effect on body composition, as weight gain, how much can be added with con- much more profound reductions in testosterone tinuous androgen administration, and how much lasting weeks have relatively small effects. Artificial will remain after steroid cessation is far from clear. hypogonadism induced for 10 weeks in healthy In general, humans increase or lose lean and fat young men who were not exercising produced a mass in roughly the same proportions as they move 2 kg shift in lean-to-fat tissue balance (Mauras et al. up and down the scale of energy balance, with 1998). In healthy young soldiers who were under- approximately two-thirds of the changes in fat fed while working hard, individual differences in mass. Forbes has plotted these trends on the basis of the level of testosterone suppression was a relat- data from overfeeding and underfeeding studies ively small contributory factor in the proportion of (Fig. 35.2) (Forbes 1993). He suggests that supple- lean tissue lost during weight loss over 8 weeks, mentation with anabolic hormones moves indi- even though the association was statistically sig- viduals off the normal trajectory, with a substantial nificant (Friedl 1997). This is supported by find- increase in the lean component and even a reduction ings in animal studies, where fasting and exercise in the fat component. In other words, androgen caused smaller testosterone reductions in endur- supplementation has been suggested to provide a ance-trained animals and this difference was also nutrient partitioning effect similar to the class of β- associated with higher lipolysis and decreased adrenergic agonist drugs that have been developed glyogenolysis, possibly protecting muscle protein for animal food production (e.g. clenbuterol) that (Guezennec et al. 1984b). favor a specific increase in the lean mass com- Acute increases in testosterone observed with ponent. Forbes was one of the first to illustrate this resistance exercise may be enough to trigger tissue principle in a study with high-dose androgen receptor-level changes or other anabolic hormones, administration to normal men (see Fig. 35.3, below) but this remains unknown. In training studies con- (Forbes et al. 1992). However, all studies to-date, ducted by Kraemer et al. (1991), a comparison of including this one, leave nagging questions about bodybuilder to power lifter routines with differ- the role of dietary intakes, training effects, and the androgens and performance 531

Body composition response +8 hormones, food, exercise Testo

+6

+4 GH GH +Food +2 ∆ Lean (kg)

0

Exercise

−2 Fig. 35.2 Conceptual relationship –Food between fat and lean components during changes in body weight and the effect of superimposed anabolic −4 + + + + hormone administration. GH, growth –6 –4 –2 0 2 4 6 8 hormone. (From Forbes 1993.) ∆ Fat (kg) validity of the assumptions underlying each of the dosing regimen. There were no effects in normal methods used to estimate these changes. men at typical doses used in replacement therapy Weight gain is a well-known side effect of an- for hypogonadal men (e.g. 100 mg·wk–1 of testos- drogens used in clinical practice and in male con- terone cypionate or nandrolone decanoate, ND) traception trials using supraphysiological doses of (Crist et al. 1983), but doubling the dose (ND, injectable testosterone esters (typically 200 mg·wk–1) 200 mg·wk–1 for 8 weeks) produced significant (World Health Organization Task Force 1990; increases in weight (Hartgens et al. 2001). In another Young et al. 1993). A typical period of 6 weeks or study, androgen doses that produce ‘replacement’ longer of continuous testosterone ester administra- levels in normal men (TE, 100 mg·wk–1) did not tion to normal adult males provides about 3–5 kg of change body weight, while high levels of TE weight gain within the first few weeks (Mauss (300 mg·wk–1) and two doses of ND (100 mg·wk–1 et al. 1975; Friedl et al. 1990, 1991; Welle et al. 1992; and 300 mg·wk–1) produced significant weight gain Young et al. 1993; Pope et al. 2000). Orally active (Friedl et al. 1991). A study that was twice as long androgens have the same effects. A large series with comparable weekly dosing (TE 280 mg·wk–1) of studies in the early 1970s with methandione did not cause twice as much weight gain (Friedl (Dianabol) showed consistent increases of about et al. 1990). Bhasin et al. (1996) doubled the weekly 2 kg in men given at least 10 mg·day–1 for 3–4 week dose of testosterone (TE, 600 mg·wk–1 for 10 weeks) periods, regardless of physical training status and also observed a similar final increase of about (Friedl 2000). It is not clear that this weight gain 3.5 kg, while in the group of men who included effect is either dose- or time-dependent. In the male exercise with the same high dose, this weight contraceptive studies that have given androgen increase nearly doubled. Pope et al. (2000) found an (testosterone enanthate, TE) at a dose of 200 mg·wk–1 increase of 3.0 kg following a typical bodybuilder for more than a year, the individuals do not con- regimen of increasing doses of testosterone cypi- tinue to gain weight, despite the sustained doubling onate from 150 mg·wk–1 up to 600 mg·wk–1 over or tripling of circulating testosterone levels with this 6 weeks. (They reported all of this weight gain as 532 chapter 35

lean mass on the basis of skinfold thicknesses that technique dependent on certain assumptions that showed little change with androgen.) Although may not hold true with anabolic steroid administra- Forbes has suggested a dose–response relationship tion. Although the combined error of multiple to lifetime exposure to androgens (Forbes 1985), techniques in a multicompartment model of body these data suggest that supraphysiological doses of composition is generally smaller than the biological androgens provoke a specific weight increase as variation in body compartments measured (Friedl a threshold effect, after which mechanisms adjust et al. 1992a), multicompartment models may also and prevent further gains. If this were not true, one not resolve the question if steroid-induced changes would expect all normal men to continue to gain confound the assumptions of these methods. For weight from cumulative androgen exposure after example, if a portion of the androgen response the initial pubertal surge in testosterone, instead of includes an increase in total body water relative to reaching stable weights in the early 20s without the total lean mass, as has been suggested but never further dramatic gains. properly tested (Freed et al. 1975; Hervey et al. 1976; These increases in weight observed in controlled Wilson 1988), this could produce significant errors studies and clinical observations of high-dose in the estimated changes determined by each of the androgen administration to normal men reflect only principal body composition techniques employed, a portion of the apparent weight differences that by violating the assumptions about the composition have been estimated for anabolic steroid-using of the lean compartment for each method. Under- bodybuilders. In dedicated bodybuilders, Kouri water weighing assumes 1.1 g·cm–3 density of lean et al. (1995b) found a difference in weight between tissue, dual energy X-ray absorptiometry (DEA) steroids users and non-users of nearly 10 kg, with assumes an attenuation coefficient based on norm- comparable estimates of adiposity between the ally hydrated soft tissue, potassium-40 depends groups (~ 12.5% body fat). A comparison of the data on a fixed distribution in the lean compartment for 20 presteroid era Mr. America winners (1939– (66.6 mmol·kg–1), and deuterium dilution assumes a 1959) to a group of popular modern bodybuilders fixed hydration level (usually 72%) of the lean mass. indicated an estimated difference of 20 kg of mass, These might not be trivial errors. In semi-starvation presumably lean mass. Changes in nutrition and of normal men, where hydration of the lean mass health care over the past century have increased the increases in a subclinical edema, DEA measures lean mass of typical young US soldiers by nearly 3.5 kg more tissue than is actually present, based on 10 kg (Friedl 2004). Thus, it would be reasonable to simple scale weights (Friedl et al. 1994). Although ascribe some of the weight gain in bodybuilders speculative, a change in the opposite direction with since the 1940s and 1950s to a combination of this rapid increase in body mass produced by androgens improved nutrition, as well as exercise science could also be accompanied by increased water techniques; however, Kouri’s data provide at least a retention. If this is the case, each of the body com- rough estimate of the upper limit of the mean gain position methods would be expected to incorrectly that might be attributable to androgen use. The dif- estimate the changes in the lean mass component of ference of 3–5 kg gains in controlled trials and 10 kg weight. seen in androgen-using bodybuilders may reflect an Hervey attempted to resolve the question in his effect of repeated cycles of use, as well as training studies of high-dose methandienone (100 mg·day–1 and dietary interactions. for 6 weeks). Total body potassium measurements predicted that an average weight increase of 3.3 kg was composed of 6.3 kg lean mass gain and 2.5 kg Characterizing changes in body composition fat loss. However, body density measured by The nature of androgen-induced weight gain has underwater weighing predicted 2.4 kg lean mass never been adequately characterized. A key prob- gain and 0.9 kg fat gain, and skinfold thicknesses lem is that most studies have measured body did not change (Hervey 1976). A second study with composition using only one technique, with each the same dosing produced comparable average androgens and performance 533

(n = 7) (n = 7) ‡ 8 † (n = 7) = = LBM 6 (n 5) (n 5) † * † (n = 6) W † (n = 4) 4

† 2 † * * *

Change (kg) 0 Fig. 35.3 Changes in body weight and lean and fat mass components −2 Fat in healthy young men administered 3 mg·kg–1·wk–1 testosterone − enanthate for 12 weeks and following 4 cessation of drug administration. LBM, lean body mass; W, weight. 0 20406080100 10 30 50 70 90 110 130 150 170 (From Forbes et al. 1992.) Days on testosterone Days after testosterone weight gains of 3.6 kg and similar changes in the This was balanced at stable weight by an assumed total body potassium (Hervey et al. 1981). This time, regain of fat weight following an initial average loss total body nitrogen was also measured by neutron of 3.4 kg. The loss of at least some of the weight gain activation analysis, with an estimate that if the following steroid cessation is comparable to the increased body weight reflected new muscle, body observations in normal men in whom testosterone is nitrogen should increase by about 100 g. The meas- artificially suppressed. Normal men artificially ured increase was more than double this value, rais- reduced to prepubertal levels of circulating testo- ing new questions about the nature of the increase. sterone for 10 weeks had a 2 kg loss of lean mass, Forbes et al. (1992) reported on a typical regimen with a comparable increase in fat mass, as measured of approximately 200 mg·wk–1 of TE for 12 weeks, by DEA (Mauras et al. 1998). This suggests a labile showing an average weight gain of 4.1 kg (Fig. 35.3). component of lean mass that has some dependence Based on potassium-40 measurements, he estimated on threshold changes in androgen levels. an increase in the lean mass component of 7.5 kg by Several studies have used imaging or muscle the end of 12 weeks of drug administration. This biopsies to conduct morphological analyses on the was corroborated by urinary creatinine excretion cross-sectional area of muscle and muscle fibers. that increased by an average 358 mg·day–1 (on a Histological analyses of muscle biopsies from the meat-free diet), corresponding to an estimated lean vastus lateralis of normal subjects following admin- mass increase of 8.6 kg. istration of TE 200 mg·wk–1 for 6 weeks was not able There is an apparent rapid loss from the lean mass to demonstrate significant increases in muscle fiber compartment following androgen cessation. In this diameter (Griggs et al. 1989), while two other studies same study, Forbes et al. (1992) reported a prompt using athletes found significant increases in muscle decline in weight to about 2.5 kg within the first fiber area of this muscle (Alen et al. 1984; Kuipers 10 days following steroid cessation, and this new et al. 1993). A more recent study found that TE 300 weight was maintained over the next 6 months and 600 mg·wk–1 for 20 weeks in normal men with (Fig. 35.3). The estimated lean mass declined to endogenous androgen suppression significantly about half of the peak gain by 2 months following increased the cross-sectional area of muscle fibers steroid cessation, and further declined over 6 and myonuclear number in biopsies of the vastus months to just over 1 kg from the initial baseline. lateralis, indicating an effect of androgen on muscle 534 chapter 35

hypertrophy; there was no significant change in the as well as the essential role of estrogens in these proportion of type I and type II fibers (Sinha-Hikim effects. et al. 2002). This same study demonstrated dose- Studies of regional fat tissue metabolism suggest dependent increases in vastus muscle volume meas- that androgens reduce triglyceride accumulation in ured by magnetic resonance imaging (MRI), and an intra-abdominal fat, but do not have this effect on apparent dose-dependent increase in total lean subcutaneous fat (Marin 1995; Marin et al. 1996), mass at 300 and 600 mg·wk–1 TE. In their earlier although testosterone treatment produces an in- study, Bhasin et al. (1996) found significant increases crease in visceral fat in women (Elbers et al. 1997). in MRI-measured muscle cross-sections of both the An increase in lean mass might further influence triceps and quadriceps of the men receiving testo- body composition by changing energy require- sterone, whether or not they included exercise. ments, but there is no apparent androgen-specific However, no change was found for the exercise increase in energy metabolism. Welle et al. (1992) alone group, which supports the overall effect of found a 7% increase in basal metabolic rate with androgens on muscle mass increase. New state- TE (200 mg·wk–1 for 12 weeks), and this increase of-the-art techniques to quantify skeletal muscle could be completely accounted for by the increase in should be of great value in terms of ease and preci- skeletal muscle mass. An informative study of the sion for future studies (Gallagher et al. 1999). body composition changes was conducted using the Androgens increase muscle protein synthesis, but ‘Leydig clamp’ model of Bhasin and his colleagues, do not typically cause an increase in total body pro- with suppressed endogenous testosterone combined tein synthesis. Griggs et al. (1989) studied healthy with sub- and supraphysiological doses of testo- men administered approximately 200 mg·wk–1 TE sterone enanthate (25–600 mg·wk−1 for 20 weeks) for 12 wks and found typical increases in lean mass (Bhasin et al. 2001). Changes in lean men were based on potassium-40 and creatinine excretion, assessed by DEA, including regional lean and fat accompanied by a 27% mean increase in muscle estimates, and by multiple MRI slices of the thigh protein synthesis rates. The observed increases are and abdomen (Woodhouse et al. 2004). The DEA therefore presumed to be a shift in the balance data indicate increases in truncal and extremity between synthesis and degradation rates, with an lean mass with high doses, and increases in fat mass overall increase in turnover rates. Presumably, with subphysiological doses. MRI scans indicated the effect is mediated directly through androgen increases in adipose tissue volumes in all regions receptors in the muscle, but this mechanism has assessed (intra- and subcutaneous abdominal, and not been well characterized, and there appear to be intermuscular and subcutaneous thigh) at low major differences in the androgen responsiveness doses, with less consistent reductions in adipose of different muscles (Gustafsson et al. 1984; Hughes tissue volume at supraphysiological doses. These & Krieg 1988; Antonio et al. 1999). New studies data provide confirmation of androgen-induced of androgen-receptor signaling pathways in muscle increases in lean components of body composition are beginning to identify an inverse relationship and decreases in fat mass. between new muscle cell development and adipo- cyte development that appear to link androgen Mechanisms of performance effects on fat and muscle shifts (Lee 2002), as has enhancement been previously suggested for steroid hormone actions in bone in the balance of osteoblasts and Are observed strength gains only from increased adipocytes. Other transgenic studies that use tar- muscle mass? geted overexpression of androgen receptor (Wiren et al. 2003) and muscle-specific IGF-I (Paul & An increase in muscle size is associated with Rosenthal 2002) in muscularly overdeveloped increases in strength, since maximal force produc- ‘mighty mouse’ animal models are likely to produce tion in a muscle is directly proportional to the near-term breakthroughs in our understanding cross-sectional muscle area. It has been suggested of bone, muscle and fat regulation by androgens, that androgens produce their main effects simply androgens and performance 535

through the increase in muscle size, without other additive in the steroid and exercise group (Fig. 35.4). changes in neuromuscular performance (Schroeder In Bhasin’s more recent dose–response study, TE et al. 2003). There are other possible explanations (300 and 600 mg·wk–1) significantly increased 1 RM for androgen-induced strength gains in athletes, leg press (average increases of ~ 75 kg) and leg including neuromuscular adaptations and muscle power (~ 40–50 W increase), while no changes were biochemical and physiological alterations, but these observed at replacement or lower doses (Bhasin remain largely undefined. One of the first published et al. 2001). This corresponded to increases in thigh strength studies in normal men involved four med- volume and muscle fiber size in these two groups. ical students given 50 mg·day–1 methyl testoster- These strength changes did not appear to be dose- one along with creatine, and tested for changes in dependent. No changes were observed in leg press grip strength in a single-blind, crossover study repetitions to failure (‘fatigability’) (Storer et al. 2003). with 3 weeks of steroid use (Samuels et al. 1942). Separate from the effects of muscle cross-sectional No improvements in performance were observed. area, effects on psychomotor enhancement have Among the difficulties in objectively assessing the been suggested. Methandienone was reported to direct effects of androgens on muscular strength significantly shorten patellar reflex time in weight are the variability inherent in strength tests, estab- lifters (Ariel & Saville 1972), but an intensive study lishing appropriate controls for both individual of a small group of elite steroid-using Finnish experience and current training status, blinding the strength athletes showed no changes in psycho- subjects to their treatment (because of obvious motor or motor speed tests compared to a control changes such as testicular volume), and the poten- group of athletes (Era et al. 1988). Both groups tially important interactions with other factors improved physical performance during 24 weeks that may also be influenced by androgens, such as of training, without an apparent androgen effect motivation and training during the study. (Alen et al. 1984). Earlier studies with methandienone (10 mg·day–1, or more, for 3–4 weeks) produced significant im- Aggression and motivation provements of approximately 15% in one repetition maximum (1 RM) bench press performance, includ- Testosterone has become synonymous with aggress- ing some blinded tests and a mix of trained and ive power. One of the most difficult aspects of untrained subjects, as previously reviewed (Haupt studying androgen performance effects is ensuring & Rovere 1984; Friedl 2000). Methandienone was that the subjects are blinded to the treatment, in part quite consistent in increasing strength performance, because many subjects perceive a change in their although considerable discussion about the inter- mental outlook when they are receiving high-dose actions with exercise has never been resolved (e.g. androgens (Freed et al. 1975; Wright 1980). The Freed et al. 1975; Hervey et al. 1981). With careful change in mental status has been variously des- blinding, including a testosterone ‘replacement’ cribed as an increase in feelings of energy, a com- comparison dose group of TE (100 mg·day–1), a high pulsive focus on winning their event, and complete dose of TE (300 mg·day–1 for 6 weeks) produced disinhibition of aggressiveness, even to the point of significant increases in elbow flexion and knee murderous rage. Pope has reported extensively on extension tests (Friedl et al. 1991). Despite equival- the psychological effects of androgen supplementa- ent gains in body weight in ND treatment groups tion for over a decade (e.g. Pope & Katz 1988, 1994). 100 and 300 mg·day–1, there were no significant Pope’s research team found significant neuropsy- increases in strength. An increase in hip adduction chological effects in a thoughtful study designed to was the only muscle strength measure that changed mimic the increase in androgen dose employed by in untrained men receiving TE (200 mg·wk–1 for many bodybuilders by stepping the weekly dose up 24 weeks) (Young et al. 1993). Bhasin et al. (1996) from 150 mg·wk–1 to 300 mg·wk–1 to 600 mg·wk–1 measured an increase in 1 RM bench press with TE (Kouri et al. 1995a; Pope et al. 2000). In this, and other (600 mg·wk–1) and demonstrated that steroid and studies reviewed by Pope and our own researchers exercise each produced significant gains that were (e.g. Hannan et al. 1991), individual manic or 536 chapter 35

* P < 0.001 6

4 P = 0.02 2 Fat free mass (kg) 0 † P < 0.001 ) 2 600 P = 0.003 400 200 Triceps

area (mm 0 ‡ P < 0.001 ) 2 1200 † † P < 0.001 800 P < 0.001 400 area (mm

Quadriceps 0 Mean change * P < 0.001 † 20 † P < 0.005 10 0 Bench-press strength (kg) ‡ P < 0.001 40 † P < 0.001 Fig. 35.4 Changes in body 30 P = 0.004 composition and strength measures 20 in healthy young men administered –1 10 placebo or 600 mg·wk testosterone Squatting enanthate for 20 weeks with or strength (kg) 0 Placebo testosterone Placebo testosterone without an exercise program. No exercise Exercise (From Bhasin et al. 1996.)

hypomanic reactions have been observed, support- It is clear that almost irrational attitudes of invin- ing the anecdotal case reports on manic behavior in cibility provide a performance advantage, allowing androgen users. Pope et al. (2000) found significant athletes to withstand psychological fatigue and responses in a specially designed test, the Point discomfort to persist and win. However, neuro- Subtraction Aggression Paradigm (PSAP), as well physiologists have not completed the androgen as the Young Mania Rating Scale and daily diary connection between the behaviors, the neural cir- scores. The PSAP challenge test was particularly cuits, and the biochemistry. Hannan et al. (1991) novel and appropriate in this study, probing indi- found an increase in dopamine metabolites in viduals for a low threshold of aggressive response. peripheral circulation of men administered high- A better understanding of who is at risk for the dose androgens, suggesting the possibility of an idiosyncratic reactions is vitally important to future interaction with the dopaminergic system, which human studies. Conceivably, the action is based on has been implicated in a range of behaviors from crossover activity to corticosteroid receptors (Janne highly focused, goal-directed behavior to psychosis. 1990); psychotic episodes with high-dose cortico- Indeed, androgens have been since demonstrated steroids have been well documented (Lewis & to increase HVA/DA ratios in the striatum of rats Smith 1983). given one of four different androgens; methan- androgens and performance 537

dienone specifically increased dopamine synthesis poiesis, and stimulation of iron incorporation into (Thiblin et al. 1999). Hannan et al. (1990) had also red blood cells. While androgen treatment has pro- shown in animal studies that stereotypic cage- vided significant increases in hematocrit and exercising behavior was increased by androgen hemoglobin in anemic patients (Neff et al. 1981), administration. Exercise effects on brain dopamine studies of high-dose androgen administration to activity have been suggested to play a role in delay- normal men have not typically reported further ing exhaustion (Chaouloff et al. 1987) and might be increases (Mauss et al. 1975; Alen 1985; Kiraly 1988). interactive with androgens. However, Palacios et al. (1983) found a significant increase in red cell counts in normal men adminis- tered TE (200 mg·wk–1 for 16 weeks), with modest What do androgens do for endurance increases in hematocrit and hemoglobin. Although performance? rare, polycythemia has been reported in hypo- Androgens may provide endurance advantages gonadal patients treated with androgens, usually in through other non-psychological effects, including older patients. There may also be increases in blood both energy availability and the oxygen-carrying volume (Wilson 1988), although this has not been capacity of the blood. These could be beneficial to verified in athletes. Thus, it is unclear that andro- performance of endurance athletes such as distance gens boost hematological parameters in normal runners and cyclists. Earlier studies examined the men, but if they did, there could be clear benefits to effects of androgen supplementation on running and endurance performance. Erythrocyte infusions into swimming performance. In an article that received normal men can substantially increase maximal high visibility in Science, Johnson and O’Shea (1969) oxygen uptake (Muza et al. 1987; Sawka et al. 1987). reported that methandienone (10 mg·day–1 for The downside is an increased risk of thromo- 3 weeks) increased maximal oxygen uptake by 15%, embolytic injury such as stroke, especially with a and significantly more than the gain in non-blinded potentiating effect of dehydration during intensive control training partners. The same drug regimen performance (Sawka et al. 1996). This is comparable failed to improve swim performance over that of a to the risks that may be encountered by cyclists and control group (O’Shea 1970), as did administration other athletes using exogenous erythropoeitin for of oxandrolone (10 mg·day–1 for 6 weeks) (O’Shea performance enhancement (Adamson & Vapnek & Winkler 1970). Johnson et al. (1972) attempted 1991). to replicate the initial study reported in Science, Effects on the heart muscle and increases in stroke measuring aerobic capacity after the same drug regi- volume would theoretically also benefit endur- men of methandienone (10 mg·day–1 for 3 weeks) or ance athletes. While the effects of androgens on left stanazolol (6 mg·day–1 for 3 weeks), and a running ventricular hypertrophy have been observed in training program, but found no change in maximal strength athletes (e.g. McKillop et al. 1986), this has oxygen uptake or mile run time. Johnson speculated not been studied in endurance athletes, where ster- that the discrepancy in results could be explained by oid use appears to be less prevalent or far more covert. androgen-induced leg strength improvements that It remains to be determined if androgens enhance may have produced specific improvements in the lipolysis and energy availability for endurance per- bike test used in the original study, rather than a real formance. In one of the few studies to examine increase in aerobic capacity. No further tests of metabolic advantages, Guezennec et al. (1984a) androgens on aerobic performance in normal men tested endurance and metabolism of rats with elev- have been reported. ated androgen levels during treadmill running for There are plausible mechanisms to suggest 7 h, but found no significant effects of testosterone androgen benefits to endurance performance. Andro- other than a preservation of some of the glycogen gens boost red blood cell production through stores. Other metabolic effects are potentially medi- stimulation of erythropoeitin synthesis and secre- ated through hormones such as IGF-I (Hobbs tion, direct stimulation of bone marrow hemato- et al. 1993). Testosterone increases circulating IGF-I 538 chapter 35

levels, while nandrolone, which does not produce have been reported for bodybuilders. For example, the same metabolites including estrogens, had no in one review, 20 of the 28 case reports and most of effect on total IGF-I levels. This suggests that there the deaths involved bodybuilders (Friedl 2000). are differences in some of the biological actions of Health benefits appear to accrue in replacement various synthetic anabolic steroids, determined by therapy in aging men, similar to hormone replace- the receptor-binding affinities of the compounds ment therapy in women. These benefits include and their metabolites. Androgens act to modify maintaining muscle mass and strength, which may glucose disposal in normal men (300 mg·wk–1 ND be important in preventing injuries; maintenance or TE) through non-insulin mechanisms (Hobbs of bone mass and prevention of male osteoporosis; et al. 1996). antidepressant effects; and important effects on Other hormonal responses to exercise, such as the libido and motivation. Studies in the past decade acute rise in circulating growth hormone that occurs indicate that androgens can increase glucose dis- so dramatically in response to a serious strength posal separately from insulin-mediated mech- athlete routine (more repetitions and short rests) anisms (Hobbs et al. 1996), as well as reduce compared to a body builder routine (higher weight intra-abdominal fat (Marin et al. 1995), both of and longer rests) (Kraemer et al. 1991), may also which may be important in preventing type 2 dia- account for some of the effects currently ascribed betes. Reduction in fibrinolytic activity (Fearnley to anabolic steroids. Growth hormone responds to & Chakrabarti 1962) may reduce certain types of acute metabolic signals, and like testosterone, spe- cardiovascular risks, perhaps countering effects cifically stimulates muscle protein and bone mineral such as the reduction in HDL-cholesterol, but con- accretion, and stimulates the release of IGF-I ceivably increase hemorrhagic stroke risk. Although (Hindmarsh et al. 1997; Lee et al. 1997). It will be there is no defined disease of androgen excess in important to understand these interactions in the men, perhaps the effects of androgens are so ubiqui- hormonal responses to exercise in future training tous that a wide variety of adverse health outcomes studies in order to develop a systematic approach to ranging from prostate cancer to heart disease are all science-based nutrition and training guidance. diseases of androgen excess, just as being male is associated with shorter average lifespan. As a corol- Eunuchs may live longer, but do lary of this, athletes that use androgenic steroids androgen-using athletes die sooner? may be further trading lifespan for poorly defined performance advantages. Although eunuchs may The health risks associated with androgen supple- have a longer average lifespan than normally viril- mentation to normal men have been extensively ized men (Hamilton 1948), it is unknown if the reviewed and discussed elsewhere (e.g. Friedl 1990, reverse is true. Epidemiological studies of retired 2000). In summary, the primary risks are a revers- athletes have been proposed but are fraught with ible reduction in sperm count and an accompanying complications, particularly in accurately ascertain- testicular atrophy; hepatic adenoma that may be ing androgen administration histories and control- benign but may also result in hemorrhage and ling for differences such as risk-taking behaviors in death; an adverse serum lipid profile with marked steroid users compared to non-steroid users. suppression of high-density lipoprotein (HDL) Concern about the current increase in testoster- -cholesterol and unknown heart disease risk out- one supplementation of older men at least has a comes; changes in impulsivity and aggressiveness; tradeoff in disease prevention by reducing osteo- skin changes including hair loss, oily skin, and acne; porotic fractures and sarcopenia (that increases gynecomastia; and weight gain. Testosterone also musculoskeletal injury risks). Replacement doses plays a permissive role in prostate cancer, but a have been administered to many older men (> 45 causative role remains to be demonstrated. From years), and prescriptions for testosterone-replace- case reports of athletes using androgenic steroids, ment therapy have skyrocketed in the US in the past the majority of serious adverse health consequences 5 years, providing a new cohort of men for studies androgens and performance 539

that are likely to provide new insights into the diet and psychological stress are well-known factors health consequences of androgen supplementation. in peak athletic performance, and it appears that There is also a growing database of experience with endogenous testosterone levels can be markedly androgen administration to normal men from male affected by these factors, making it a useful indic- contraceptive studies around the world. Other data ator of training status for males. The current think- on high-dose supplementation is emerging from ing is that androgens may be useful in maintaining efforts to sustain lean mass in muscular dystrophy health and well-being in aging men as replacement and acquired immunodeficiency syndrome (AIDS) therapy to prevent sarcopenia and osteoporosis. patients. Supraphysiological doses that double or triple circulating levels have been tested in healthy young Conclusions men for male contraception without remarkable effects other than modest weight gain. Higher Androgenic steroids are banned at various levels of doses, particularly when combined with exercise, athletic competition, initially because of concerns may produce substantial gains in muscle mass and about an uneven playing field, but then also out of strength. Anecdotal reports of the use of extraordin- concern for the health risks to the athletes. The arily high levels of androgenic steroids that extend greatest advantage is in bodybuilding in which well beyond the observations in controlled studies, huge muscle development is the goal, and the involve primarily body builders and are associated majority of case reports of serious health outcomes with the majority of case reports of serious adverse have centered on these athletes who are more likely health effects. to use very high doses that exceed levels that can be ethically studied. Other athletes appear to derive Acknowledgements benefits from androgens for strength, endurance and overall psychological advantage. Prediction of The author gratefully acknowledges the editorial the specific and individual effects of androgens is assistance of Ms. Shari Hallas and technical assist- still uncertain. All of these compounds appear to ance provided by Sergeant Joseph Alemany. have some positive musculotrophic effect, but the The opinions and assertions in this chapter are nature of this effect is still not well understood and those of the author and do not necessarily represent almost certainly involves complex interaction with the official views or policies of the Army or the other hormones at the tissue level. Management of Department of Defense.

References

Aakvaag, A., Bentdal, O., Quigstad, K. et al. and muscle fiber characteristics of glucocorticoids upon testicular (1978) Testosterone and testosterone elite power athletes self-administering luteinizing hormone receptor and binding globulin (TeBG) in young men androgenic and anabolic steroids. steroidogenesis in vivo and in vitro. during prolonged stress. International Acta Physiologica Scandanavica 122, Endocrinology 108, 2142–2148. Journal of Andrology 1, 22–31. 535–544. Bardin, C.W. & Catterall, J.F. (1981) Adamson, J.W. & Vapnek, D. (1991) Antonio, J., Wilson, J.D. & George, F.W. Testosterone: a major determinant of Recombinant erythropoietin to improve (1999) Effects of castration and androgen extragenital sexual dimorphism. Science athletic performance. New England treatment on androgen-receptor levels 211, 1285–1294. Journal of Medicine 324, 698–699. in rat skeletal muscles. Journal of Applied Bardin, C.W., Caterall, J.F. & Janne, O.A. Aguilera, R., Hatton, C.K. & Catlin, D.H. Physiology 87, 2016–2019. (1990) The androgen-induced (2002) Detection of epitestosterone Ariel, G. & Saville, W. (1972) Effect of phenotype. Anabolic Steroid Abuse, doping by isotopic ratio mass anabolic steroids on reflex components. Research Monograph 102, 131–141. spectrometry. Clinical Chemistry 48, Journal of Applied Physiology 32, 795–797. Behre, H.M. & Nieschlag, E. (1992) 629–636. Bagatell, C.J. & Bremner, W.J. (1990) Sperm Testosterone buciclate (20 Aet-1) in Alen, M. (1985) Androgenic steroid effects counts and reproductive hormones in hypogonadal men: pharmacokinetics on liver and red cells. British Journal of male marathoners and lean controls. and pharmacodynamics of the new Sports Medicine 19, 15–20. Fertility and Sterility 53, 688–692. long-acting androgen ester. Journal of Alen, M., Häkkinen, K. & Komi, P.V. (1984) Bambino, T.H. & Hsueh, A.J.W. Clinical Endocrinology and Metabolism 75, Changes in neuromuscular performance (1981) Direct inhibitory effect of 1204–1210. 540 chapter 35

Bhasin, S., Storer, T.W., Berman, N. et al. De Souza, M.J., Arce, J.C., Pescatello, L.S., weightlifters. British Medical Journal 2, (1996) The effects of supraphysiologic Scherzer, H.S. & Luciano, A.A. (1994) 471–473. doses of testosterone on muscle size and Gonadal hormones and semen quality in Friedl, K.E. (1990) Reappraisal of the health strength in normal men. New England male runners. A volume threshold effect risks associated with the use of high Journal of Medicine 335, 1–7. of endurance training. International doses of oral and injectable androgenic Bhasin, S., Storer, T.W., Berman, N. Journal of Sports Medicine 15, 383–391. steroids. National Institute on Drug Abuse et al. (1997) Testosterone replacement Dessypris, A., Kuoppasalmi, K. & Research Monograph 102, 142–177. increases fat-free mass and muscle size Adlercreutz, H. (1976) Plasma cortisol, Friedl, K.E. (1997) Variability of fat and in hypogonadal men. Journal of Clinical testosterone, androstenedione, and lean tissue loss during physical exertion Endocrinology and Metabolism 82, luteinizing hormone in a non- with energy deficit. In: Physiology, Stress 407–413. competitive marathon run. Journal of and Malnutrition: Functional Correlates, Bhasin, S., Woodhouse, L., Casaburi, R. Steroid Biochemistry 7, 31–37. Nutritional Intervention (Kinney, J.M. & et al. (2001) Testosterone dose–response Elbers, J.M., Asscheman, H., Seidell, J.C., Tucker, H.N., eds.). Lippincott-Raven relationships in healthy young men. Megens, J.A. & Gooren, L.J. (1997) Publishers, New York: 431–450. American Journal of Physiology 281, Long-term testosterone administration Friedl, K.E. (2000) Effect of anabolic steroid E1172–E1181. increases visceral fat in female-to-male use on body composition and physical Blanchard, D.C., Sakai, R.R., McEwen, B., transsexuals. Journal of Clinical performance. In: Anabolic Steroids in Weiss, S.M. & Blanchard, R.J. (1993) Endocrinology and Metabolism 82, Sport and Exercise, 2nd edn (Yesalis, C.E., Subordination stress: behavioral, brain, 2044–2047. ed.). Human Kinetics Publishers, and neuroendocrine correlates. Behavior Elbers, J.M., Asscheman, H., Seidell, J.C. & Champaign, IL: 139–174. and Brain Research 58, 113–121. Gooren, L.J. (1999) Effects of sex steroid Friedl, K.E. (2004) Can you be large and Bourne, P.G., Rose, R.M. & Mason, J.W. hormones on regional fat depots as not obese? The distinction between body (1968) 17-OCHS levels in combat: assessed by magnetic resonance imaging weight, body fat, and abdominal fat in Special Forces ‘A’ team under threat of in transsexuals. American Journal of occupational standards. Diabetes attack. Archives of General Psychiatry 19, Physiology 276, E317–E325. Technology and Therapeutics 6, 732–749. 5–14. Elias, M. (1981) Serum cortisol, Friedl, K.E. & Yesalis, C.E. (1989) Catlin, D.H., Ahrens, B.D. & Kucherova, Y. testosterone, and testosterone-binding Self-treatment of gynecomastia in (2002) Detection of norbolethone, an globulin responses to competitive bodybuilders who use anabolic steroids. anabolic steroid never marketed, in fighting in human males. Aggressive Physician and Sportsmedicine 17, 67–79. athletes’ urine. Rapid Communications in Behavior 7, 215–224. Friedl, K.E., Hannan, C.J., Jr., Jones, R.E. Mass Spectometry 16, 1273–1275. Era, P., Alen, M. & Rahkila, P. (1988) & Plymate, S.R. (1990) HDL-cholesterol Chaouloff, F., Laude, D., Merino, D. et al. Psychomotor and motor speed in is not decreased if an aromatizable (1987) Amphetamine and α-methyl- power athletes self-administering androgen is administered. Metabolism p-tyrosine affect the exercise-induced testosterone and anabolic steroids. 39, 69–74. imbalance between the availability of Research Quarterly for Exercise and Sport Friedl, K.E., Dettori, J.R., Hannan, C.J., Jr., tryptophan and synthesis of serotonin in 59, 50–56. Patience, T.H. & Plymate, S.R. (1991) the brain of the rat. Neuropharmacology Fearnley, G.R. & Chakrabarti, R. (1962) Comparison of the effects of high dose 26, 1099–1106. Increase of blood fibrinolytic activity by testosterone and 19-nortestosterone Collu, R., Gibb, W. & Ducharme, J.R. (1984) testosterone. Lancet ii, 128–132. to a replacement dose of testosterone Role of catecholamines in the inhibitory Forbes, G.B. (1985) The effect of anabolic on strength and body composition effect of immobilization stress on steroids on lean body mass: the in normal men. Journal of Steroid testosterone secretion in rats. Biology dose–response curve. Metabolism 34, Biochemistry and Molecular Biology 40, of Reproduction 30, 416–422. 571–573. 607–612. Crawford, B.A., Liu, P.Y., Kean, M.T., Forbes, G.B. (1993) The companionship of Friedl, K.E., DeLuca, J.P., Marchitelli, L.J. & Bleasel, J.F. & Handelsman, D.J. (2003) lean and fat. In: Human Body Composition Vogel, J.A. (1992a) Reliability of body-fat Randomized placebo-controlled trial of (Ellis, K.J. & Eastman, J.D., eds.). Plenum estimations from a four-compartment androgen effects on muscle and bone Press, New York: 1–14. model using density, body water, and in men requiring long-term systemic Forbes, G.B., Porta, C.R., Herr, B.E. & bone mineral measurements. American glucocorticoid treatment. Journal of Griggs, R.C. (1992) Sequence of changes Journal of Clinical Nutrition 55, 764–770. Clinical Endocrinology and Metabolism 88, in body composition induced by Friedl, K.E., Moore, R.J. & Marchitelli, L.J. 3167–3176. testosterone and reversal of changes (1992b) Steroid replacers: let the Crist, D.M., Stackpole, P.J. & Peake, G.T. after drug is stopped. Journal of the athlete beware. National Strength and (1983) Effects of androgenic-anabolic American Medical Association 267, Conditioning Association Journal 14, 14–19. steroids on neuromuscular power and 397–399. Friedl, K.E., Moore, R.J., Martinez-Lopez, body composition. Journal of Applied Francis, K.T. (1981) The relationship L.E. et al. (1994) Lower limits of body fat Physiology 54, 366–370. between high and low trait in healthy active men. Journal of Applied Cumming, D.C., Brunsting, L.A., Strich, G., psychological stress, serum testosterone, Physiology 77, 933–940. Ries, A.L., & Rebar, R.W. (1986) and serum cortisol. Experientia 37, Friedl, K.E., Moore, R.J., Hoyt, R.W. Reproductive hormone increases in 1296–1297. et al. (2000) Endocrine markers of response to acute exercise in men. Freed, D.J.L., Banks, A.J., Longson, D. & semistarvation in healthy lean men in a Medicine and Science in Sports and Exercise Burley, D.M. (1975) Anabolic steroids in multistressor environment. Journal of 18, 369–373. athletics: crossover double-blind trial on Applied Physiology 88, 1820–30. androgens and performance 541

Friedl, K.E., Mays, M.Z., Kramer, T.R. & secretions as indicated by the effects of factor axis, body composition, and Shippee, R.L. (2001) Acute recovery of castration in man and by studies of metabolic parameters. Journal of Clinical physiological and cognitive function pathological conditions and the short Endocrinology and Metabolism 82, in US Army Ranger students in a lifespan associated with maleness. 2172–2176. multistressor field environment. In: The Recent Progress in Hormone Research 3, Hobbs, C.J., Plymate, S.R., Rosen, C.J. & Effect of Prolonged Military Activities in 257–322. Adler, R.A. (1993) Testosterone Man––Physiological and Biochemical Hannan, C.J., Jr., Friedl, K.E. & Plymate, administration increases insulin-like Changes––Possible Means of Rapid S.R. (1990) Effect of testosterone on growth factor I levels in normal men. Recuperation. Technical Report RTO- apomorphine induced movements Journal of Clinical Endocrinology and MP-042. Research and Technological in the rat. Workshop Conference on Metabolism 77, 776–779. Organization, North Atlantic Treaty Androgen Therapy: Biologic and Clinical Hobbs, C.J., Jones, R.E. & Plymate, S.R. Organization, Neuilly-sur-Seine Cedex, Consequences, 17–20 January 1990, Marco (1996) Nandrolone, a 19-nortestosterone, France: 6.1–6.10. Island, FL. enhances insulin-independent glucose Gallagher, D., Heymsfield, S.B. & Wang, Hannan, C.J., Jr., Friedl, K.E., Kettler, T.M. uptake in normal men. Journal of Clinical Z.M. (1999) Skeletal muscle markers. & Plymate, S.R. (1991) Psychological Endocrinology and Metabolism 81, In: Protein and Amino Acids. National and serum homovanillic acid changes 1582–1585. Academy Press, Washington, DC: in men administered androgenic Houmard, J.A., McCulley, C., Shinebarger, 255–277. steroids. Psychoneuroendocrinology 16, M.H. & Bruno, N.J. (1994) Effects of Glass, A.R. & Vigersky, R.A. (1980) 335–43. exercise training on plasma androgens Resensitization of testosterone Hartgens, F., Van Marken Lichtenbelt, in men. Hormone and Metabolism Research production in men after human W.D., Ebbing, S. et al. (2001) Body 26, 297–300. chorionic gonadotropin-induced composition and anthropometry in Hughes, B.J. & Krieg, M. (1988) Steroid desensitization. Journal of Clinical body builders: regional changes due to receptors and the muscular system. In: Endocrinology and Metabolism 51, nandrolone decanoate administration. Steroid Receptors and Disease: Cancer, 1395–1400. International Journal of Sports Medicine 22, Autoimmune, Bone, and Circulatory Griggs, R.C., Kingston, W., Jozefowicz, 235–241. Disorders (Sheridan, P.J., Blum, K. & R.F. et al. (1989) Effect of testosterone Haupt, H.A. & Rovere, G.D. (1984) Trachtenberg, M.C., eds.). Marcel on muscle mass and muscle protein Anabolic steroids: a review of the Dekker, New York: 415–433. synthesis. Journal of Applied Physiology literature. American Journal of Sports Janne, O.A. (1990) Androgen interaction 66, 498–503. Medicine 12, 469–484. through multiple steroid receptors. Guezennec, C.Y., Ferre, P., Serrurier, B. Hellhammer, D.H., Hubert, W. & National Institute on Drug Abuse Research et al. (1984a) Metabolic effects of Schurmeyer, T. (1985) Changes in Monograph 102, 178–186. testosterone during prolonged physical saliva testosterone after psychological Jezova, D. & Vigas, M. (1981) Testosterone exercise and fasting. European Journal of stimulation in men. response to exercise during blockade Applied Physiology 52, 300–304. Psychoneuroendocrinology 10, 77–81. and stimulation of adrenergic receptors Guezennec, C.Y., Serrurier, B., Aymonod, Hershberger, L.G., Shipley, E.G. & Meyer, in man. Hormone Research 15, M., Merino, D. & Pesquies, P.C. (1984b) R.K. (1953) Myotrophic activity of 141–147. Metabolic and hormonal response to 19-nortestosterone and other steroids Jezova-Repcekova, D., Vigas, M., Miklaj, L. short term fasting after endurance determined by modified levator ani & Jurcovicova, J. (1982) Plasma training in the rat. Hormone and Metabolic muscle method. Proceedings of the Society testosterone during bicycle ergometer Research 16, 572–575. for Experimental Biology and Medicine 83, exercise without and after l-dopa Guezennec, Y., Leger, L., Lhoste, F., 175–180. pretreatment. Endocrinologia Aymonod, M. & Pesquies, P.C. (1986) Hervey, G.R., Hutchinson, I., Knibbs, A.V. Experimentalis, 16, 3–8. Hormone and metabolite response et al. (1976) ‘Anabolic’ effects of Johnson, L.C. & O’Shea, J.P. (1969) to weight-lifting training sessions. methandienone in men undergoing Anabolic steroid: effects on strength International Journal of Sports Medicine 7, athletic training. Lancet 2, 699–702. development. Science 164, 957–959. 100–105. Hervey, G.R., Knibbs, A.V., Burkinshaw, L. Johnson, L.C., Fisher, G., Silvester, L.J. & Guezennec, C.Y., Satabin, P., Legrand, H. et al. (1981) Effects of methandienone on Hofheins, C.C. (1972) Anabolic steroid: & Bigard, A.X. (1994) Physical the performance and body composition effects on strength, body weight, oxygen performance and metabolic changes of men undergoing athletic training. uptake and spermatogenesis upon induced by combined prolonged Clinical Science 60, 457–461. mature males. Medicine and Science in exercise and different energy intakes Hickson, R.C., Kurowski, T.T., Capaccio, Sports 4, 43–45. in humans. European Journal of Applied J.A. & Chatterton, R.T., Jr. (1984) King, D.S., Sharp, R.L., Vukovich, Physiology 68, 525–530. Androgen cytosol binding in exercise- M.D. et al. (1999) Effect of oral Gustafsson, J.A., Saartok, T., Dahlberg, E. induced sparing of muscle atrophy. androstenedione on serum testosterone et al. (1984) Studies on steroid receptors American Journal of Physiology 247, and adaptations to resistance training in in human and rabbit skeletal muscle: E597–E603. young men: a randomized controlled clues to the understanding of the Hindmarsh, P.C., Fall, C.H.D., Pringle, trial. Journal of the American Medical mechanism of action of anabolic P.J., Osmond, C. & Brook, C.G.D. (1997) Association 281, 2020–2028. steroids. Progress in Clinical and Biological Peak and trough growth hormone Kiraly, C.L. (1988) Androgenic-anabolic Research 142, 261–290. concentrations have different steroid effects on serum and skin Hamilton, J.B. (1948) The role of testicular associations with the insulin-like growth surface lipids, on red cells, and on liver 542 chapter 35

enzymes. International Journal of Sports report of 14 cases and a review of the Neff, M.S., Goldberg, J., Slifkin, R.F. et al. Medicine 9, 249–252. literature. Journal of Affective Disorders 5, (1981) A comparison of androgens for Knight, J. (2003) Drugs bust reveals 319–332. anemia in patients on hemodialysis. athletes’ secret steroid. Nature 425, 752. Liddle, G.W. & Burke, H.A. (1960) New England Journal of Medicine 304, Kochakian, C.D. (1988) The evolution Anabolic steroids in clinical medicine. 871–875. from ‘the male hormone’ to anabolic- Helvetica Medica Acta 27, 504–513. O’Connell, M., Robbins, D.C., Horton, E.S., androgenic steroids. Alabama Journal of Loucks, A.B. (2004) Energy balance and Sims, E.A.H. & Danforth, E., Jr. (1979) Medical Sciences 25, 96–102. body composition in sports and exercise. Changes in serum concentrations of Kouri, E.M., Lukas, S.E., Pope, H.G. & Journal of Sports Sciences 22, 1–14. 3,3’,5’-triiodothyronine and 3,5,3’- Oliva, P. (1995a) Increased aggressive Loucks, A.B. & Thuma, J.R. (2003) triiodothyronine during prolonged responding in male volunteers following Luteinizing hormone pulsatility is moderate exercise. Journal of Clinical the administration of gradually disrupted at a threshold of energy Endocrinology and Metabolism 49, increasing doses of testosterone availability in regularly menstruating 242–246. cypionate. Drug and Alcohol Dependence women. Journal of Clinical Endocrinology Opstad, P.K. (1992a) Androgenic 40, 73–79. and Metabolism 88, 297–311. hormones during prolonged physical Kouri, E.M., Pope, H.G., Katz, D.L. & Lucia, A., Chicharro, J.L., Perez, M. et al. stress, sleep, and energy deficiency. Oliva, P. (1995b) Fat-free mass index (1996) Reproductive function in male Journal of Clinical Endocrinology and in users and nonusers of anabolic- endurance athletes: sperm analysis and Metabolism 74, 1176–1183. androgenic steroids. Clinical Journal of hormonal profile. Journal of Applied Opstad, P.K. (1992b) The Sports Medicine 5, 223–228. Physiology 81, 2627–2636. hypothalamo–pituitary regulation of Kraemer, W.J., Gordon, S.E., Fleck, S.J. et al. McKillop, G., Todd, I.C. & Ballantyne, D. androgen secretion in young men after (1991) Endogenous anabolic hormone (1986) Increased left ventricular mass in prolonged physical stress combined and growth factor responses to heavy a bodybuilder using steroids. British with energy and sleep deprivation. resistance exercise in males and females. Journal of Sports Medicine 20, 151–152. Acta Endocrinologica 127, 231–236. International Journal of Sports Medicine 12, Marin, P. (1995) Testosterone and regional O’Shea, J.P. (1970) Anabolic steroid: effects 228–35. fat distribution. Obesity Research 3, on competitive swimmers. Nutrition Kreuz, L.E., Rose, R.M. & Jennings, J.R. 609S–612S. Report International 1, 337–342. (1972) Suppression of plasma Marin, P., Lonn, L., Andersson, B. et al. O’Shea, J.P. & Winkler, W. (1970) testosterone levels and psychological (1996) Assimilation of triglycerides in Biochemical and physical effects of stress. A longitudinal study of young subcutaneous and intraabdominal an anabolic steroid in competitive men in Officer Candidate School. adipose tissue in vivo in men: effects swimmers and weightlifters. Nutrition Archives of General Psychiatry 26, of testosterone. Journal of Clinical Reports International 2, 351–362. 479–482. Endocrinology and Metabolism 81, Palacios, A., Campfield, L.A., McClure, Kruskemper, H.L. (1968) Anabolic Steroids. 1018–1022. R.D., Steiner, B. & Swerdloff, R.S. (1983) Academic Press, New York. Mauras, N., Hayes, V., Welch, S. et al. Effect of testosterone enanthate on Kuipers, H., Binkhorst, F.P., Hartgens, F., (1998) Testosterone deficiency in young hematopoiesis in normal men. Fertility Wijnen, J.A. & Keizer, H.A. (1993) men: marked alterations in whole body and Sterility 40, 100–104. Muscle ultrastructure after strength protein kinetics, strength, and adiposity. Paul, A.C. & Rosenthal, N. (2002) Different training with placebo or anabolic Journal of Clinical Endocrinology and modes of hypertrophy in skeletal muscle steroid. Canadian Journal of Applied Metabolism 83, 1886–1892. fibers. Journal of Cell Biology 156, 751–760. Physiology 18, 189–196. Mauss, J., Borsch, G., Bormacher, K. et al. Ponjee, G.A., DeRooy, H.A. & Vader, H.L. Kuoppasalmi, K., Naveri, H., Harkonen, (1975) Effect of a long-term testosterone (1994) Androgen turnover during M. & Adlerkreuz, H. (1980) Plasma oenanthate administration on male marathon running. Medicine and Science cortisol, androstenedione, testosterone, reproductive function: clinical in Sports and Exercise 26, 1274–1277. and luteinizing hormone in running evaluation, serum FSH, LH, testosterone, Pope, H.G., Jr. & Katz, D.L. (1988) exercise of different intensities. and seminal fluid analyses in normal Affective and psychotic symptoms Scandanavian Journal of Clinical and men. Acta Endocrinologica 78, 373–384. associated with anabolic steroid use. Laboratory Investigation 40, 403–409. Michel, G. & Baulieu, E.E. (1980) Androgen American Journal of Psychiatry 145, Lee, D.K.P. (2002) Androgen receptor receptor in rat skeletal muscle: 487–490. enhances myogenin expression and characterization and physiological Pope, H.G., Jr. & Katz, D.L. (1994) accelerates differentiation. Biochemical variations. Endocrinology 107, 2088–2097. Psychiatric and medical effects of and Biophysical Research Communications Morville, R., Pesquies, P.C., Guezennec, anabolic-androgenic steroid use. 294, 408–413. C.Y., Serrurier, B.D. & Guignard, M. Archives of General Psychiatry 51, Lee, D.K.P., Durham, S.K., Martinez, V. (1979) Plasma variations in testicular 375–382. et al. (1997) Kinetics of insulin-like and adrenal androgens during Pope, H.G., Jr., Kouri, E.M. & Hudson, J.I. growth factor (IGF) and IGF-binding prolonged physical exercise in man. (2000) Effects of supraphysiologic doses protein responses to a single dose of Annales d’Endocrinologie 40, 501–510. of testosterone on mood and aggression growth hormone. Journal of Clinical Muza, S.R., Sawka, M.N., Young, A.J. et al. in normal men. Archives of General Endocrinology and Metabolism 82, (1987) Elite special forces: physiological Psychiatry 57, 133–140. 2266–2274. description and ergogenic influence of Roemmich, J.M. & Sinning, W.E. (1997) Lewis, D.A. & Smith, R.E. (1983) Steroid- blood reinfusion. Aviation Space and Weight loss and wrestling training: induced psychiatric syndromes––a Environmental Medicine 58, 1001–1004. effects on growth-related hormones. androgens and performance 543

Journal of Applied Physiology 82, Schurmeyer, T., Wickings, E.J., Freischem, Viru, A.A. (1985) Hormones in Muscular 1760–1764. C.W. & Nieschlag, E. (1983) Saliva Activity. I. Hormonal Ensemble in Exercise. Rogol, A.D., Veldhuis, J.D., Williams, F.A. and serum testosterone following CRC Press, Boca Raton, FL: 87. & Johnson, M.L. (1984) Pulsatile oral testosterone undecanoate Welle, S., Jozefowicz, R., Forbes, G. & secretion of gonadotropins and prolactin administration in normal and Grigg, R.C. (1992) Effect of testosterone in male marathon runners––relation to hypogonadal men. Acta Endocrinologica on metabolic rate and body composition the endogenous opiate system. Journal of 102, 456–462. of normal men and men with muscular Andrology 5, 21–27. Schurmeyer, T., Jung, K. & Nieschlag, E. dystrophy. Journal of Clinical Rowell, L.B., Blackmon, J.R. & Bruce, R.A. (1984) The effect of an 1100 km run Endocrinology and Metabolism 74, (1964) Indocyanine green clearance and on testicular, adrenal and thyroid 332–335. estimated hepatic blood flow during hormones. International Journal of Wheeler, G.D., Wall, S.R., Belcastro, A.N. mild to maximal exercise in upright Andrology 7, 276–282. & Cumming, D.C. (1984) Reduced man. Journal of Clinical Investigation 43, Sinha-Hikim, I., Artaza, J., Woodhouse, L. serum testosterone and prolactin levels 1677–1690. et al. (2002) Testosterone-induced in male distance runners. Journal of the Samuels, L.T., Henschel, A.F. & Keys, A. increase in muscle size in healthy young American Medical Association 252, (1942) Influence of methyl testosterone men is associated with muscle fiber 514–516. on muscular work and creatine hypertrophy. American Journal of Wilkerson, J., Horvath, S. & Gutin, B. metabolism in normal young men. Physiology 283, E154–E164. (1980) Plasma testosterone during Journal of Clinical Endocrinology 2, 649–654. Stanhope, R., Buchanan, C.R., Fenn, G.C. treadmill exercise. Journal of Applied Sapolsky, R.M. (1985) Stress-induced & Preece, M.A. (1988) Double blind Physiology 49, 249–253. suppression of testicular function in the placebo controlled trial of low dose Wilson, J.D. (1988) Androgen abuse by wild baboon: role of glucocorticoids. oxandrolone in the treatment of boys athletes. Endocrine Reviews 9, 181–198. Endocrinology 116, 2273–2278. with constitutional delay of growth and Wiren, K., Zhang, X.W., Toombs, A., Sapolsky, R.M. (1986) Stress-induced puberty. Archives of Disease in Childhood Harada, S. & Jepsen, K. (2003) Targeted elevation of testosterone concentrations 63, 501–505. overexpression of androgen receptor in in high ranking baboons: role of Storer, T.W., Magliano, L., Woodhouse, L. osteoblasts results in complex skeletal catecholamines. Endocrinology 118, et al. (2003) Testosterone dose- phenotype in growing animals. Journal 1630–1635. dependently increases maximal of Bone and Mineral Research 18, S293 Saris, W.H.M., van Erp-Baart, M.A., voluntary strength and leg power, but [Abstract]. Brouns, F., Westerterp, K.R. & ten Hoor, does not affect fatigability or specific Woodhouse, L.J., Gupta, H., Bhasin, M. F. (1989) Study of food intake and tension. Journal of Clinical Endocrinology et al. (2004) Dose-dependent effects of energy expenditure during extreme and Metabolism 88, 1478–1485. testosterone on regional adipose tissue sustained exercise: the Tour de France. Strauss, R.H., Lanese, R.R. & Malarkey, distribution in healthy young men. International Journal of Sports Medicine 10, W.B. (1985) Weight loss in amateur Journal of Clinical Endocrinology and S26–S31. wrestlers and its effect on serum Metabolism 89, 718–726. Sawka, M.N., Young, A.J., Muza, S.R., testosterone levels. Journal of the World Health Organization Task Force Gonzalez, R.R. & Pandolf, K.B. (1987) American Medical Association 254, on Methods for the Regulation of Male Erythrocyte reinfusion and maximal 3337–3338. Fertility (1990) Contraceptive efficacy aerobic power. An examination of Sundaram, K., Kumar, N., Monder, C. & of testosterone-induced azoospermia in modifying factors. Journal of the American Bardin, C.W. (1995) Different patterns normal men. Lancet 336, 955–959. Medical Association 257, 1496–1499. of metabolism determine the relative Wright, J.E. (1980) Anabolic steroids and Sawka, M.N., Joyner, M.J., Miles, D.S. anabolic activity of 19-norandrogens. athletics. Exercise and Sport Sciences et al. (1996) American College of Sports Journal of Steroid Biochemistry and Reviews 8, 149–202. Medicine position stand. The use of Molecular Biology 53, 253–257. Yesalis, C.E., III, (ed.) (2000) Anabolic blood doping as an ergogenic aid. Sutton, J.R., Coleman, M.J., Casey, J. & Steroids in Sport and Exercise, 2nd edn. Medicine and Science in Sports and Exercise Lazarus, L. (1973) Androgen responses Human Kinetics, Champaign, IL. 28, i–viii. during physical exercise. British Medical Yesalis, C.E., III, Herrick, R.T., Buckley, Schmid, P., Pusch, H.H., Wolf, W. et al. Journal 1, 520–522. W.E. et al. (1988) Self-reported use (1982) Serum FSH, LH, and testosterone Thiblin, I., Finn, A., Ross, S.B. & Stenfors, of anabolic-androgenic steroids by in humans after physical exercise. C. (1999) Increased dopaminergic and elite power lifters. Physician and International Journal of Sports Medicine 3, 5-hydroxytryptaminergic activities in Sportsmedicine 16, 90–100. 84–89. male rat brain following long-term Young, N.R., Baker, H.W., Liu, G. & Schroeder, E.T., Terk, M. & Sattler, F.R. treatment with anabolic androgenic Seeman, E. (1993) Body composition (2003) Androgen therapy improves steroids. British Journal of Pharmacology and muscle strength in healthy men muscle mass and strength but not 126, 1301–1306. receiving testosterone enanthate for muscle quality: results from two studies. Ursine, H., Baade, E. & Levine, S. (eds.) contraception. Journal of Clinical American Journal of Physiology 285, (1978) Psychobiology of Stress. A Study of Endocrinology and Metabolism 77, E16–E24. Coping Men. Academic Press, New York. 1028–1032. Chapter 36

Growth Hormone and Sport

JENNIFER D. WALLACE AND ROSS C. CUNEO

Introduction promote linear growth in short and GH-deficient (GHD) children is well documented. Abuse of rhGH to augment height in normal children may theoret- Evidence of growth hormone doping in sport ically enhance performance in sports such as basket- Synthetic human growth hormone (hGH), pro- ball or swimming where height is advantageous. duced by recombinant DNA technology (recom- The physiological role and actions of GH in adult binant human growth hormone, rhGH), has been life has been established from studies of patients available for clinical use since the late 1980s. It is with GHD and in catabolic illnesses. GH is anabolic widely available on the black market, with the and can increase muscle mass, suggesting its use Underground Steroid Handbook promoting the per- to build muscle or as a reparative agent to sustain ceived benefits to athletes. In recent years in heavy training. The ability of GH treatment to Australia and the UK, rhGH has been stolen from reduce body fat has made it attractive to athletes hospital pharmacies and a pharmaceutical com- who wish to reduce body fat and improve their pany warehouse, much of it believed to be destined power to weight ratio. This lipolytic effect of GH is for use by sportspersons. The discovery in 1998 of also potentially attractive to the endurance athlete, the possession of vials of rhGH by Chinese swim- as an agent to release fatty acids into the blood- mers attending the World Swimming Champion- stream, as a source of energy, and hence as a means ships in Perth, Australia, and cycling teams in the of preserving glucose, in long duration events such Tour de France in Europe, has confirmed suspicions as the marathon. GH has also been shown to have that hGH is in use at the highest levels of sporting effects on the heart, thermoregulation and blood competition. volume which could be interpreted as potentially performance enhancing. The ability of GH to stimu- late endorphins, a natural analgesic, and to have Rationale for growth hormone abuse by athletes: positive effects on mood, may also allow athletes to a brief introduction train more effectively. GH treatment in GHD adults The abuse of rhGH to enhance performance pro- has been shown to increase aerobic exercise capacity o ceeds despite a lack of controlled studies to object- (V 2max) by up to 20%, even without training. How- ively confirm or refute potential benefit to athletes. ever it must be emphasized that these improve- What then might be the reasons for athletes choos- ments are the results of normalizing low GH levels ing to abuse growth hormone (GH)? In an attempt in deficient adults, and extrapolation of these to gain insight into the reasons athletes and coaches results to GH treatment in normal individuals is not might promote this activity the biological effects established. As an example, GH treatment appears of GH will be covered in detail below, but a brief to increase muscle strength in GHD adults, but overview is now provided. The use of rhGH to acromegalic patients who have chronically elevated 544 growth hormone and sport 545

GH levels have increased muscle size but with such a substance, in blood or urine, constitutes a reduced strength compared to normal adults. GH positive test. Unique problems exist for endogenous administration to weightlifters does not appear to proteins like GH and erythropoietin. enhance the ability to perform resistance work. GH as a marker of GH abuse. There are two main reasons why it is unlikely that the detection of GH Potential side effects of growth hormone abuse could be achieved by solely measuring GH doping in serum or urine. Firstly, GH is produced in a Physiological levels of GH, as used in adults treated pulsatile manner and is stimulated by a number of for GHD, merely replace endogenous production factors including exercise; an elevated level of in a doping athlete. It is therefore necessary for serum GH could possibly indicate doping, but it athletes to take supra-physiological doses of GH to could equally be the result of a naturally occurring achieve an effect. The main adverse effects of short- pulse or a response to stress or exercise. Secondly, term GH exposure are fluid retention and arthral- GH used in doping is produced by recombinant gias. However, if athletes wish to reproduce the DNA technology, and is identical to the major circu- effects of GH exposure they might be tempted to lating variant of GH produced by the human pitu- expose themselves to high levels of GH for an itary gland. It is therefore impossible to distinguish extended period of time, with the attendant risks of exogenous from endogenous GH. Manufacturers serious, often irreversible, adverse effects. Nature of rhGH are not keen to chemically label rhGH, as has provided us with a model of GH doping in a the process would involve considerable expense, condition of pathological GH excess. When this con- and necessitate clinical trials to demonstrate that dition presents in childhood, before bone fusion has labeling does not change the biological activity of occurred, it is known as gigantism, as it accelerates the product. Furthermore, labeling of GH would bone growth, producing abnormally tall indivi- also not result in detection of GH from unlicensed duals. If the condition occurs in adults it is called sources. From these considerations, strategies for a acromegaly. GH excess in adults does not influence test for GH abuse have evolved using naturally height, but has irreversible effects on bone growth, occurring markers of GH action in the human body. resulting in big hands and feet, a prominent fore- A detailed examination of the biological actions of head and jaw with splayed teeth, accelerated degen- GH, as they potentially apply to sporting perform- erative arthritis and enlargement and deformity of ance and to the selection of markers of GH abuse, is the vertebrae. Soft tissues are also affected, eventu- therefore now presented. ally resulting in the obvious features of a large nose and lips. Internal organs enlarge, leading to multi- Biological actions of growth hormone: system pathology, and a range of disorders such as rationale for the use of growth hormone high blood pressure, sleep apnoea, diabetes mellitus by athletes and a range of cancers (especially colon cancer) are common results of chronic pathological elevation The biological actions of GH have been clearly of GH. demonstrated in response to administration of rhGH to children with short stature, adults with GHD, adults and children suffering from a number of Development of a test for growth hormone catabolic states, and in disease states where circulat- abuse in sport ing GH levels are chronically high due to secretion While details of current progress will be presented from a pituitary tumor (gigantism or acromegaly). later in this chapter, some introductory comments From these clinical data, it is evident that GH has are required. Most of the agents on the list of banned growth promoting effects on long bones, is anabolic, drugs in sport are foreign or exogenous substances. lipolytic, has multiple effects on carbohydrate meta- In most cases, simple detection of the presence of bolism and is important for fluid homeostasis. In 546 chapter 36

addition it appears to have beneficial effects on Body composition physical strength and endurance, on thermoregula- tion, cardiovascular health, psychological well- The anabolic and lipolytic effects of GH are clearly being and on the immune system. We aim to present evident in the deranged body composition of chil- a detailed mechanistic examination of the effects dren and adults with GHD. In addition to a reduc- of GH as demonstrated by clinical conditions, tion in bone mass, childhood GHD is characterized and highlight the potential risks and benefits that by subnormal lean body mass and obesity, and sportspersons may, theoretically, experience should treatment with GH results in increases in bone and hGH be abused. muscle mass and a decrease in adipose tissue (Collip et al. 1973; De Boer et al. 1992). Discontinu- ation of GH treatment when children reach adult- Bone growth hood has been shown to result in a reduction in lean Reduced linear growth of long bones, short stature body mass and increase in fat mass within 12 months and reduced bone density are evident in children (Rutherford et al. 1989; Juul et al. 1998). with GHD (Shore et al. 1980). In contrast, GH The GHD syndrome in adults has only been re- hypersecretion in children (gigantism) results in cognized within the last decade. The first random- abnormally tall individuals as a result of linear over- ized controlled studies of GH replacement to adults growth of long bones with unfused epiphyses. The were published in 1989 (Rahkila et al. 1989; Salomon capacity of GH administration to stimulate long et al. 1989). Jørgensen and colleagues (Rahkila bone growth, in children with GHD, and in children et al. 1989) studied 22 GHD adults in a randomized, with short stature due to other causes, is well docu- double-blind placebo-controlled cross-over study mented (Root et al. 1998), and is a recognized treat- for 4 months with a daily GH dose of 2 IU·m2 (≅ 600 ment to augment adult height in children with GHD µg·m2 or 15 µg·kg–1·day–1). Treatment resulted in no world wide. GH abnormalities of adult onset would change in body weight, with an increase in the mean not be expected to affect height, as prior fusion of muscle volume of the thigh and a corresponding bone epiphyses would preclude changes in long decrease in mean thigh adipose tissue as assessed by bone length. Chronic GH hypersecretion in adults computerized tomography (CT). Subscapular skin- (acromegaly), results in several irreversible effects fold thickness was also significantly reduced with on bone, such as overgrowth of the jaw bone, and an GH treatment. Salomon et al. (1989) performed a ran- increase in the size of bones in the hands and feet. domized, double-blind placebo-controlled trial in GH treatment in GHD adults has been shown to 24 adults for 6 months using a GH dose of dose 0.07 increase bone turnover acutely. This results in a IU·kg–1 of body weight per day (≅ 20 µg·kg–1·day–1) biphasic effect on bone mineral content and bone and also showed no effect on body weight and a mineral density, with an initial fall followed by a mean increase in lean body mass of 5.5 kg and a return to baseline levels and increases above base- corresponding decrease in fat mass of 5.7 kg, as line levels (Vandeweghe et al. 1993; Ohlsson et al. assessed by measurement of total body potassium 1998). (TBK) counting. The criticisms of these studies have The theoretical benefits of GH use, in terms of been the small number of subjects studied and the bone metabolism in sportspersons, is twofold. GH high GH doses used. A review by Carroll et al. (1998) may result in increased height, if treatment were summarizes numerous studies which support the started prior to epiphyseal fusion, to benefit height ability of GH to increase lean body mass and related sporting performance. In adult athletes, GH decrease body fat, especially abdominal fat, using a may theoretically increase bone strength, thereby variety of GH doses and a large range of techniques minimizing stress fractures or ligamentous/ten- to analyse body composition. A large multicentered, dous bone-insertion fractures. The basis for such randomized, double-blind, placebo-controlled study expectations has not been subjected to scientific of GH administration to 166 GHD adult patients, study. using a much smaller dose than the early studies, growth hormone and sport 547

and employing a range of body composition tech- lack of effect on body cell mass with reduction in niques, has recently confirmed the ability of GH FFSTM being explained by the reduction in body treatment to increase lean and decrease fat tissue in water. this patient population (Cuneo et al. 1998). Limited data is available on the effect of GH Excessive GH secretion results in increased lean administration on body composition in normal body mass and reduced body fat. A study of 150 subjects and athletes. Crist et al. (1988) performed acromegalic subjects in Sweden revealed that they a double-blind placebo-controlled study of GH ad- had increased lean body mass and reduced fat mass ministration to eight subjects who acted as their when compared with 476 normal subjects (Bengtsson own controls. This study demonstrated body com- et al. 1989). Body composition was assessed by the position changes in highly conditioned adults given four-compartment model using assessment of TBK exogenous GH (30–50 µg·kg–1 three times per week), and body water. TBK was initially measured by a during 6 weeks of progressive resistance training. potassium radioisotope (42KCL) dilution technique Using hydrodensitometry an increase in fat free and later by whole body potassium counting, and mass (FFM) of 4% and a decrease in body fat of 4% total body water was measured using the tritiated was demonstrated. Hydrodensitometry was also water dilution technique. Body cell mass as estim- used by Yarasheski et al. (1992) in a placebo- ated by TBK showed a mean increase above normal controlled study of GH treatment in young men of 2.7 and 2.2 kg in acromegalic male and female undertaking heavy resistance exercise. Seven sub- subjects respectively, and fat mass was shown to be jects were treated with GH (40 µg·kg–1·day–1) dur- lower than predicted with men and women having ing the 12-week training program. GH treatment a mean fat mass 7.8 and 5.6 kg below normal. The resulted in a greater increase in FFM than placebo same group demonstrated that successful treatment treatment in subjects undertaking identical exercise. of acromegaly does not normalize body composi- Similar results were found when GH was adminis- tion, but results in persistent elevation of body cell tered to older men during resistance training. GH mass, a reduction in total body water and an administration with a training program resulted in increase in body fat (Bengtsson et al. 1989b). This greater increases in FFM than training alone: 23 may encourage athletes into believing that doping healthy male sedentary subjects, with a mean age with supra-physiological levels of GH may result of 67.1 years, took part in a 16-week program of in increases in lean body mass which can be main- progressive resistance training after being ran- tained after doping has ceased. It has been suggested domized to receive GH or placebo. Eight subjects however that ‘cured’ acromegalic subjects with low received GH at a dose of 12.5–24 µg·kg–1·day–1, serum GH may have abnormal 24-h secretory which resulted in an 8% increase (4.8 kg) in mean profiles which possibly contribute to maintenance FFM as assessed by hydrodensitometry (Yarasheski of lean body mass. A more recent study of body et al. 1995). An additional five subjects were unable composition in acromegalics (O’Sullivan et al. 1994) to complete the study due to side effects related to using dual X-ray absorptiometry and sodium dilu- fluid retention. Lack of change in vastus lateralis tion, has confirmed that acromegalics are less fat muscle protein synthesis and strength, along with than normal controls and increase their fat mass changes in body water that contributed to the following successful treatment. They demonstrated increased FFM with GH treatment suggest that that fat free soft tissue mass (FFSTM) was elevated the addition of GH to resistance training did not by acromegaly, but that this increase was due contribute to an increase in muscle tissue. Deyssig to increased body water and not increased body et al. (1993) was unable to show any effect of GH cell mass, perhaps suggesting a lack of anabolic on body composition in highly trained athletes effect of GH on acromegalic muscle. In contrast with low-fat mass in a placebo-controlled study to the Bengtsson study, successful treatment of of GH for 6 weeks at a dose of 0.09 U·kg–1·day–1 acromegaly resulted in a reduction in FFSTM in this (≅ 30 µg·kg–1·day–1). Measurements of fat and lean study, which furthermore was shown to represent a mass were however estimated only by skinfold 548 chapter 36

measurement which may have been a relatively trolled studies have shown a progressive increase in insensitive technique. quadriceps isometric force toward normal over Athletes may interpret the changes in body com- 12–36 months (Jørgensen, J.O.L. et al. 1990b; Rahkila position demonstrated by these studies to indicate et al. 1994; Johannsson et al. 1997; Janssen et al. 1999). that doping with GH could potentially result in sim- Proximal, limb-girdle muscle force has been shown ilar body composition changes resulting in increased to increase significantly in adults with GHD after lean body mass without training and decreased fat rhGH treatment (Cuneo et al. 1991a). A study of the mass. This would be an attractive proposition to effects of 5 years of GH therapy on 109 GHD adult those athletes striving to improve their power to subjects in a single center, prospective, open label weight ratio, or to body builders who aim to de- study has demonstrated normalization of isometric velop maximum muscle mass with minimal subcu- and isokinetic knee flexor and extensor strength and taneous body fat. an improvement but lack of normalization of hand grip strength (Svensson et al. 2003). Histology of vastus lateralis taken by needle Strength and endurance exercise biopsy from adults with GHD has shown normal Details of skeletal muscle function have been fiber type proportions and areas (Whitehead et al. assessed most carefully in the quadriceps group. 1989; Cuneo et al. 1992). No features of Cushingoid Before treatment, adults with GHD had signific- myopathy were evident. Differences in fiber type antly reduced isometric quadriceps force compared function have not been assessed in detail, but sig- to carefully matched, normal individuals (Cuneo nificant correlations between type I relative fiber et al. 1990). Most of this deficit could be explained by area and maximal aerobic performance and between reduced muscle cross-sectional area (Cuneo et al. type II relative fiber area and maximal force genera- 1990; Janssen et al. 1999). When force was expressed tion (Whitehead et al. 1989) suggest that individual per muscle cross-sectional area, an additional small muscle fiber-type function is qualitatively nor- deficit was still evident, particularly in males; pos- mal. Following treatment, vastus lateralis histology sible explanations include differences in training, either remained unchanged (Whitehead et al. 1989; muscle fiber or myofibrillar density, or neuromus- Cuneo et al. 1992), or showed an increase in type I cular excitation. Normal muscle histology and CT and type II mean fiber areas (Woodhouse et al. 1999). density of quadriceps muscle (Cuneo et al. 1992) The reason for the difference is unclear, but meth- suggest that muscle fiber density was not at fault. odological limitations of this technique are well Rutherford et al. (1991) performed an uncontrolled known. No features of an acromegalic myopathy study of GHD children during 1 year after the were seen following GH replacement. cessation of rhGH treatment when final height was In conclusion, GHD appears to cause a reduction attained. They found significant reductions in in skeletal muscle mass, not function, which results quadriceps cross-sectional area and force (with a in reduced muscle force generation which can be trend toward reduced muscle-fiber areas on biopsy) overcome in some but not all muscle groups with over the year. Isometric force in muscle groups GH treatment over a period of several years. other than the quadriceps have shown essentially In untreated adults with GHD, maximal oxygen o no difference from a normal population (Cuneo et al. uptake (V 2max) measured during cycle ergometry 1990). Respiratory muscle function may also be has been shown to be significantly reduced (mean of reduced in adults with GHD (Merola et al. 1996). 72–82% of that predicted from age, sex and height) Short-term controlled studies of rhGH treatment (Cuneo et al. 1991b; Whitehead et al. 1992). Power (4–6 months) (Rahkila et al. 1989; Cuneo et al. 1991a; output is also likely to be reduced (Rahkila et al. Woodhouse et al. 1999), have not been able to 1989; Cuneo et al. 1991b; Whitehead et al. 1992). demonstrate a statistically significant increase in Mean maximal heart rates of 90% and respiratory quadriceps force despite clear increases in thigh exchange ratio (RER) greater than 1.0 suggest that muscle cross-sectional area. Longer-term, uncon- maximal or near maximal subjective effort had been growth hormone and sport 549

obtained (Cuneo et al. 1991b). GHD may have a neg- proximal myopathy. The reason for this lack of ative chronotropic effect, demonstrated in some function despite apparent muscular hypertrophy is studies (Rahkila et al. 1989) but not others (Cuneo obscure but may relate to histological features char- et al. 1991b; Nass et al. 1995). Following rhGH treat- acterized by cellular infiltrates and type II fiber atro- ment, maximal exercise performance (maximal oxy- phy, type I fiber hypertrophy and connective tissue gen uptake and maximal power output) improved depositions (Nabarro 1987), compression or other markedly (mean changes 20–30%) (Rahkila et al. neuropathic processes, or other undefined factors. 1989; Cuneo et al. 1991b; Whitehead et al. 1992; Nass Athletes may consider rhGH abuse to augment et al. 1995). Submaximal exercise capacity, measured strength and/or endurance performance but con- as anaerobic threshold, also improved suggesting trolled data are lacking or discouraging. that activities performed during sedentary and strenuous daily life would be accomplished with Metabolic effects of growth hormone: less metabolic stress (Cuneo et al. 1991b; Nass protein anabolism et al. 1995). Some studies have controlled physical activity and still demonstrated beneficial effects In order to explore the mechanism responsible for (Salomon et al. 1989; Cuneo et al. 1991b). One other the increases in lean body mass in GHD adults study did not demonstrate a statistically significant treated with GH, studies of protein turnover have increase in maximal exercise capacity, but anaerobic been conducted in this patient population (Russell- threshold did increase, and this appeared to be a Jones et al. 1993). Whole body isotopic leucine major contributor to the subjective sensation of turnover studies were performed using L-1-13C- fatigue experienced by these patients (Woodhouse leucine to measure protein degradation (leucine et al. 1999). Explanations for improved aerobic per- Ra), and protein synthesis (non-oxidative leucine formance may relate to: (a) increased oxygen pulse Rd). Whole body protein turnover was measured (Cuneo et al. 1991b, Nass et al. 1995) and increased in 18 severely GHD adults, in a double-blind, cardiac output (Cuocolo et al. 1996); (b) increased placebo-controlled trial of GH/placebo at a dose lean body mass and skeletal muscle mass (Cuneo of 0.018 U·kg–1·day–1 for 1 month, followed by et al. 1991b); (c) altered fuel utilization (see below); 0.036 U·kg–1·day–1 (≅ 12 µg·kg–1·day–1) for 1 month. and (d) altered thermal homeostasis (see below). Results indicated that the increase in lean body Additionally, adults with childhood-onset GHD mass resulting from GH treatment in GHD adults have reduced lung volumes and respiratory muscle is due to an increase in protein synthesis with no force generation (Merola et al. 1996), whereas adult- change in protein degradation when results are onset patients only have a problem with the latter expressed in terms of lean body mass. The potent (Cuneo et al. 1991b). In acromegaly, maximal aero- anabolic effect of GH on whole body protein synthe- bic capacity appears to be impaired in the untreated sis has also been clearly demonstrated in normal state, and to improve only in patients who have an human subjects in response to GH administration improvement in cardiac output with normalization (Horber & Haymond 1990). Leucine turnover studies of serum GH and insulin-like growth factor I (IGF-I) in the normal population were mechanistically levels in response to treatment (Colao et al. 2000). consistent with those obtained from GHD subjects, Fatigue and impaired ventilatory threshold in acro- with GH administration resulting in an increase in megaly can be normalized with successful reduc- protein synthesis and no effect on proteolysis. tion of IGF-I (Thomas et al. 2002). A number of therapeutic uses of GH as an Therefore, substantial abnormalities of physical anabolic agent have been explored. Hypocaloric performance, likely to impinge on daily activities, feeding induces catabolism in normal subjects result from GHD, and quite dramatic improvements reflected by a loss of total body nitrogen. The capa- in clinically meaningful aerobic performance result city of GH to prevent nitrogen loss in normal subjects from rhGH replacement. In contrast to the situation with hypocaloric feeding was demonstrated (Manson in GHD subjects, patients with acromegaly have a & Wilmore 1986). Similarly GH administration has 550 chapter 36

been shown to conserve nitrogen in obese subjects shown to be unchanged by GH treatment, and a during short-term dietary restriction (Snyder et al. corresponding lack of change in muscle strength 1988), where 20 obese subjects were studied in a was evident. The authors have concluded that placebo-controlled trial of GH administration dur- increases in total body protein occur when training ing a 12-week restricted-energy intake of 24 kcal·kg is accompanied by GH treatment, but that extracel- (100 kJ·kg) ideal body weight (IBW)–1. Recombinant lular protein and not muscle mass accounts for this methionyl GH was injected intramuscularly every change. In contrast, experienced weightlifters were other day at a dose of 100 µg·kg IBW–1. The nitrogen- given GH for 14 days at a dose of 40 µg·kg–1·day–1, sparing effects of GH were only evident for the and 13C leucine turnover studies were performed first 5 weeks of the study. The authors attribute the before and after the treatment phase (Yarasheski lack of enduring GH effect on nitrogen retention to et al. 1993). The findings were that such short-term development of insulin-like growth factor I (IGF-I) GH treatment did not increase muscle protein resistance. synthesis in experienced weightlifters. Whole body The protein anabolic property of GH has been protein breakdown was also not affected by GH utilized in a number of catabolic states. Surgically treatment. This was, however, a very small study in produced acute stress can result in exaggerated pro- seven subjects and did not include a control group. tein catabolism, and GH treatment following sur- Exercise can therefore be considered as a model gery has been shown to preserve body cell mass and for catabolic stress in humans. The ability of GH to promote wound healing, and to accelerate wound overcome catabolism in many disease states, and to healing in severely burned children (Wilmore enhance wound healing, may lead some athletes & Catabolic 1991). Experimental wounds heal with to believe that the anabolic properties of GH may greater strength following GH treatment than with enable them to train harder with less protein break- placebo (Jørgensen, P.H. & Andreassen 1988; Belcher down and to enhance recovery from heavy training. & Ellis 1990). It has recently been shown that intra- The available experimental data do not however tendonous injection of IGF-I results in improved support such a concept. Empirical studies of wound healing of experimentally induced flexor tendonitis healing in athletes have not as yet been undertaken. in horses (Dahlgren et al. 2002). We have explored Nitrogen retention has been clearly documented the ability of GH to overcome protein catabolism following GH treatment in GHD (Henneman et al. in chronic liver disease (Wallace et al. 1996). GH 1960), in the long-term resulting in the previously treatment in this patient population was shown to mentioned changes in body composition. Protein result in a significant increase in lean tissue mass turnover studies show a direct effect of GH and IGF- as assessed by TBK and bioimpedance analysis. I on protein synthesis (Russell-Jones et al. 1993; The increase in TBK, the main intracellular ion, was Umpleby & Russell Jones 1996; Salomon et al. 1997), similar in magnitude to that observed in treatment while insulin’s anabolic action is mediated via of GHD adults (Cuneo et al. 1991a). a reduction in protein breakdown (Umpleby & Due to the ethical constraints of administering Sönksen 1985; Umpleby et al. 1986). Thus GH, IGF-I GH to athletes the effect of GH administration to and insulin appear to act as an ‘anabolic team’. athletes on protein turnover has not been extens- Fasting hyperinsulinaemia, generated by GH’s in- ively investigated. Young, untrained men were sulin antagonistic and insulinotropic (Davidson treated with GH in a 12-week placebo-controlled 1987; Press 1988; Fowelin et al. 1993; Jørgensen, study of resistance exercise (Yarasheski et al. 1992). J.O.L. et al. 1993; O’Neal et al. 1994) actions, may con- Turnover studies with 15N glycine and 13C leucine tribute to the anabolic effect of GH. showed that training supplemented with GH resulted in increased rates of whole body protein Metabolic effects of growth hormone: synthesis and greater whole body protein balance. lipolysis and lipid oxidation The in vivo fractional incorporation rate of leucine into protein was assessed with biopsy of the quad- Clinical evidence of the lipolytic action of GH is pro- riceps muscle. Muscle protein synthesis rate was vided by the ability of GH treatment to decrease the growth hormone and sport 551

excessive levels of adipose tissue in GHD adults, Hussain et al. (1994) administered either GH, IGF-I and for the reverse to result from successful treat- or a combination of GH and IGF-I to eight GHD ment of acromegaly. Direct evidence of the role of adult subjects. Circulating FFA levels were similarly GH in lipolysis has been demonstrated in many in elevated with GH or IGF-I and an increase in lipid vitro and in vivo studies. Harant et al. (1994) exposed oxidation occurred with both treatments. Com- fat cells from five GHD adults to GH and demon- bination therapy resulted in a synergistic effect. The strated a lipolytic effect as evidenced by glycerol direct effect of GH on adipose tissue is thought to appearance in medium. In an uncontrolled study, be achieved by its ability to enhance the reactivity they showed that 6 months of GH treatment of hormone-sensitive lipase to lipolytic hormones (0.125 IU·kg–1·wk–1 for 1 month and 0.25 IU·kg–1·wk–1 (Hussain et al. 1994). Hussain proposes that the lipo- [≅ 12 µg·kg–1·day–1] for the subsequent 5 months) lytic effect of IGF-I is indirect, and is a result of the further increased the lipolytic activity in fat cells ability of IGF-I to inhibit insulin secretion ‘therefore from these patients. releasing the brakes on lipolysis’ (Hussain et al. 1994). A number of studies have demonstrated the The demonstrated ability of GH to increase lipo- ability of GH administration to increase circulating lysis thereby increasing circulating FFA levels, and free fatty acid (FFA) levels in normal human sub- also to subsequently increase lipid oxidation may jects (Davidson 1987; Press 1988). Cheng and Kalant make it attractive to athletes performing extreme (1970) studied 10 adult human subjects and demon- endurance activities, such as marathon running, as strated that 5 mg of intravenous GH was able to it could potentially provide a source of circulating double serum FFA levels approximately 5.5 h later. fuel and at the same time protect carbohydrate stores. Subcutaneous adipose tissue, from 21 normal sub- jects, nine of which were undergoing abdominal Metabolic effects of growth hormone: surgery for non-malignant conditions, showed an carbohydrate metabolism increase in lipolysis, measured as glycerol release, in response to GH (Ottosson et al. 2000). Recently, GH has traditionally been seen as insulin- the effects of GH replacement in adults with GHD antagonistic, counteracting the actions of insulin to on exercise-induced changes in lipid metabolism restrain hepatic glucose production and promote have been examined. Gibney et al. (2003) performed peripheral glucose disposal (Davidson 1987; Press a randomized GH withdrawal study in replaced 1988). Indeed patients with acromegaly have a high GHD adults. Studies of both rest and exercise, after prevalence of carbohydrate intolerance and dia- 3 months of GH cessation, showed that GH in- betes mellitus. The situation with GHD and more creased the rate of appearance (Ra) of both glyc- subtle degrees of GH replacement is, however, erol and FFAs and the disappearance rate (Rd) of far more complex and the implications for athletic FFA. Glucose Ra is index of lipolysis from adipose performance difficult to assess. tissues since glycerol is not reincorporated into Insulin sensitivity is a feature of GHD in infancy triglycerides and FFA Ra is the parallel change to and childhood where fasting hypoglycemia causes glycerol Ra suggesting little FFA re-esterification. clinical concern. In adults with GHD, however, a The increased FFA Rd may result from either spectrum of insulin insensitivity has been described, increased FFA availability or lipid oxidation. This ranging from normal to insulin resistant, where the small study suggests GH is an important regulator differences have been attributed to the degree of of intermediary metabolism during exercise. GH global and especially visceral adiposity (Salomon excess is also associated with an increase in lipid et al. 1994; Hew et al. 1996). Alford et al. (1999) have oxidation (O’Sullivan et al. 1995). described severe insulin resistance in adults with Adipose tissue has been shown to express GH but untreated GHD, in association with visceral adipos- not IGF receptors and the lipolytic effect of GH ity and low intramuscular glycogen concentrations. is therefore thought to be a direct action of GH on Statistically, the severity of the defect correlated adipose GH receptors (GHRs). In order to elucidate with duration of GHD, degree of hypertriglyceride- the mechanism for lipolysis with GH administration mia and hyperinsulinemia. While the increases in 552 chapter 36

serum insulin and glucose tend to return toward Growth hormone, cholesterol metabolism and baseline after several months (Salomon et al. 1989, vascular health 1997; Bengtsson et al. 1993; Cuneo et al. 1998), more formal measurements of insulin action suggest GH appears to have a bi-modal pattern of effects on either normalization of (Fowelin et al. 1993) or per- vascular health. In early reports of patients with sistence of the resistant state (Alford et al. 1999; GHD, mostly involving children or young adults, Bramnert et al. 2003) following 6–24 months of GH GHD resulted in increased triglyceride and choles- replacement. The data favoring an amelioration of terol concentrations (predominantly low-density insulin resistance has been attributed to reductions lipoprotein [LDL] cholesterol) of a mild degree in in visceral adiposity and increases in lean mass 20–50% of cases (Merimee et al. 1972; Winter et al. (Fowelin et al. 1993; Christopher et al. 1998), and dis- 1979; Blackett et al. 1982; LaFranchi et al. 1985). In crepancies may relate to GH dose, duration, other adults with GHD, mild increases in LDL- and total pituitary hormone replacement therapies (notably cholesterol levels have been reported in 40–50% of glucocorticoids) and the persistence of hyperinsuli- patients compared to age, weight and sex-matched nemia (Alford et al. 1999). controls (Libber et al. 1990; Cuneo et al. 1993). These GH clearly affects fuel utilization. Acutely, GH same studies also reported a significant prevalence replacement in patients with GHD or short-term of hypertriglyceridemia, but differed in reporting treatment in normal individuals alters substrate increased or reduced HDL-cholesterol concentra- availability ( Jørgensen, J.O.L. et al. 1990a; Vahl et al. tions. Recent turnover studies have shown increased 1997), increases lipolysis and lipid oxidation (see Apo B-100 synthesis and decreased clearance in above) and reduces carbohydrate oxidation (Moller adults with GHD, providing further evidence for an et al. 1990), probably by means of substrate competi- atherogenic state in this syndrome (Cummings et al. tion. In contrast to these acute effects of essentially 1997). GH treatment improves many of these abnorm- transient pulses of GH, patients with acromegaly, alities (Cuneo et al. 1993, 1998; Rosén et al. 1994; who have chronically elevated GH serum concen- Beshyah et al. 1995; O’Neal et al. 1996). An 18-month trations, have been shown to have hyperglycemia open label study of 40 subjects showed that the most and hyperinsulinemia, resistance to the action notable changes in decreasing total/high-density of exogenous insulin and tendencies to increased lipoprotein (HDL) cholesterol ratio and increasing carbohydrate oxidation, increased non-oxidative HDL cholesterol were evident towards the end of glucose turnover and to have exaggerated amounts the observation period (Beshyah et al. 1995), sug- of intramuscular glycogen. Attempting to predict gesting that the cholesterol changes are likely to what this means for athletic performance is difficult, persist. Indeed, 10-year follow-up studies have con- but potentially long-duration aerobic exercise may firmed GH replacement in GHD adults maintains a benefit from the greater stores of glycogen with favorable lipid profile (Gibney et al. 1999). There are fat being used as a fuel initially, but this is entirely complex mechanisms thought to explain the GH speculative. Recently, the effects of brief, supra- effect on cholesterol metabolism, one of the main physiological doses of GH administered to competit- being an effect to up-regulate the hepatic LDL- ive cyclists have been reported (Lange, K.H. et al. receptor expression (Rudling et al. 1992), thereby 2002). In a double-blind, cross-over designed study, increasing LDL-cholesterol clearance and reducing an exaggerated lactate response to intense, aerobic hepatic cholesterol synthesis. exercise has been described, and some individuals The long-term follow-up of adults with hypo- appeared to be unable to complete the exercise pituitarism is characterized by increased total and following GH administration. This appears to be the cardiovascular mortality rates (Rosén & Bengtsson first controlled assessment of the effects of GH on 1990; Wüster et al. 1991; Bulow et al. 1997). These aerobic performance in athletes, and it does not sup- studies all assume that hypopituitary patients on port the theory that GH may be beneficial, but the conventional hormone replacement have GHD and data set is small and requires further assessment. optimal hormone replacement. More direct studies growth hormone and sport 553

however confirm an increased prevalence of ath- to examine the cardiac effects of chronically aug- erosclerotic plaques in both adults and children mented or decreased GH activity. Fazio et al. (1997) with GHD (Markussis et al. 1992; Capaldo et al. 1997). have characterized some cardiac structural and It is assumed that the dyslipidemia and visceral functional consequences of chronically altered GH adiposity account for much of the observed excess secretion. This group studied 30 adult acrome- mortality, but non-physiological replacement with galic subjects and 25 adults with childhood-onset other pituitary hormones (Al-Shoumer et al. 1995) GHD. Two groups of 30 normal controls were eva- and altered coagulation or fibrinolysis may also luated as control subjects. Echocardiography of contribute (Beshyah et al. 1993; Johansson et al. 1994, acromegalic and GHD hearts revealed an association 1996). Some data suggest the latter improves with with chronic circulating serum GH concentrations rhGH replacement (Cuneo 1998). A recent study of and left ventricular wall thickness (LVWT) and 10 years of rhGH treatment has shown a reduction associated total left ventricular mass (LVM), with in the prevalence of abnormal arterial compliance acromegalics having increased and GHD subjects was in GHD patients receiving long-term treatment decreased LVM. This suggests an anabolic effect of compared to those who did not (Gibney et al. 1999). GH on cardiac tissue. Evidence is available for both While this is a very small study with some selection direct anabolic effects of GH and IGF-I on cardiac biases, it represents the first direct clues that long- tissue (Saccá et al. 1994). Cardiac wall stress was term mortality may be reduced with rhGH replace- shown to be decreased in the high GH state and ment. In acromegaly, vascular mortality is one of the increased in the low GH state. An uncontrolled leading causes of death related to that condition, study of 12 months of GH treatment in 12 GHD sub- and appears predominantly related to hyperten- jects, with doses of rhGH ranging from 0.025 to sion, cardiomyopathy, obstructive sleep apnoea 0.050 IU·kg–1·day–1 (≅ 9–18 µg·kg–1·day–1) resulted and diabetes mellitus. in an increase in LVM secondary to an increase in LVWT, and a consequent reduction in cardiac wall stress (Fazio et al. 1997). By demonstrating that GH Cardiovascular effects of growth hormone is able to favorably affect cardiac structural and GH appears to be important for normal cardiovas- functional deficits evident in GHD, and by charac- cular health and may contribute to the increased car- terizing similar but exaggerated effects in acromegaly, diovascular morbidity and mortality in conditions the authors propose to have demonstrated that cir- of both GHD and GH excess. Hypopituitary patients, culating GH is implicated in a novel mechanism of replaced with other pituitary hormones but not GH, cardiac wall stress regulation in humans. In GHD have greater cardiovascular risk factors than normal adults, Cuneo et al. (1991c) and Caidahl et al. (1994) healthy controls as demonstrated by high body mass showed that GH treatment increased LVM (but index and plasma triglycerides, low plasma HDL not wall thickness), and increased stroke volume. cholesterol and a higher incidence of hypertension Caidahl et al. (1994) also demonstrated an increase (Rosen et al. 1993). Death from cardiovascular dis- in heart rate and cardiac output, and a decrease in ease in similar hypopituitary adults is more com- total peripheral resistance. The effect on peripheral mon than in the normal population, particularly as a vascular resistance (PVR) is consistent with known result of cerebrovascular disease. GH excess has renal/peripheral vascular effects of GH to be dis- also been shown to result in adverse cardiovascular cussed below. GH replacement has been shown to effects. Chronic supra-physiological levels of GH stimulate and normalize nitric oxide production in in acromegaly result in an increased prevalence of adults with GHD, which would be expected to cardiovascular disease and cardiovascular related result in marked arterial vasodilation and reduce deaths in these patients, particularly from a hyper- PVR (Böger et al. 1996). trophic cardiomyopathy (Carroll et al. 1998). The opportunity to study acromegalic and GHD GH administration to normal subjects. The first study of subjects provides us with natural models with which the short-term cardiac effects of GH administration 554 chapter 36

to normal men was published by Thuesen et al. crease exercise capacity, placing more emphasis on (1988). Eleven healthy male subjects with a mean the known detrimental effects of GH excess in acro- age of 31 ± 5 years were studied before and after megaly that causes left ventricle (LV) hypertrophy, 7 days of GH administration and 7 days after treat- and cardiac failure. The fact that GH increased LV ment cessation. Large doses of GH were adminis- mass in only 4 weeks suggested that detrimental tered subcutaneously (3.2 IU in the morning and effects of GH on the myocardium may occur 6.4 IU in the evening; total ≅ 3.2 mg). GH treatment rapidly. They speculated that concurrent use of resulted in a significant increase in resting heart anabolic steroids and GH may magnify the detri- rate but did not affect systolic or diastolic blood mental myocardial effects of each. pressure. Echocardiography showed no changes in heart wall thickness, but end-systolic diame- Renal effects of growth hormone ters decreased, circumferential shortening velocity increased, stroke volume was unchanged, but car- GH has a number of demonstrated effects on kidney diac output increased. The authors suggested that structure and function. In addition to the expected GH induced positive chronotropic and inotrophic anabolic effects, GH also influences renal hemody- effects. While PVR was not measured in this namics, body fluid homeostasis, renal acid-base study, the authors also suggested that the apparent homeostasis and possibly even contributes to gluc- increased myocardial contractility may be second- ose homeostasis. ary to an increase of peripheral blood flow. A recent study (Cittadini et al. 2002) has shown Anabolic effects. The anabolic effect of GH on kidney that supra-physiological doses of GH induce rapid tissue is evident in the renal hypertrophy resulting changes in cardiac morphology and function. Thirty from chronic exposure of to high levels of circulat- healthy male and 30 female volunteers were studied ing serum GH in the condition of acromegaly in a double-blind, randomized, placebo-controlled (Kopple & Hirschberg 1990). It is difficult to hypo- study of GH administration. Subjects were classi- thesize how this could impact positively on exercise fied as fit on the basis of two training sessions per performance, and the renal effects of long-term GH week for 1 year (nature, intensity and duration of doping would most likely be detrimental as it has exercise not defined). Subjects were assigned to been suggested that chronic exposure to high levels receive daily subcutaneous injections of GH at a of circulating GH might predispose to glomeru- dose of either 30 or 60 µg·kg–1·day–1 or placebo for losclerosis (Yang, C.-W. et al. 1993) and possibly 28 days with the first 7 days at a dose corresponding even renal failure (Kopple & Hirschberg 1990). to 50% of the target dose. Cardiovascular effects of GH were dose-dependent with the higher dose of Hemodynamic effects. The effects of GH on renal GH showing a 12% increase in LVM and a 10% hemodynamics are well documented. GH treatment increase in LVWT which was concentric in nature. increases renal plasma flow (RPF) and glomerular The lower, but still supra-physiological dose of GH filtration rate (GFR) in the GHD adult (Hirschberg & showed statistically non-significant changes similar Kopple 1988; Hirschberg et al. 1989; Caidahl et al. in direction but not in magnitude. No change in 1994). Chronic GH excess in acromegaly results in ejection phase indices were apparent with either elevated RPF and GFR which is reduced post dose of GH. Cardiac index, measured at rest but hypophysecotomy (see Kopple & Hirschberg 1990 still closely related to measures of exercise perform- for review). GH administration in normal subjects ance such as maximal oxygen consumption, was also results in an increase in RPF and GFR and increased by 11% with the higher GH dose, and was a reduction in renal vascular resistance (RVR) associated with a corresponding fall in systemic (Christiansen et al. 1981; Kopple & Hirschberg 1990). vascular resistance. In the absence of exercise per- For example, Christiansen et al. (1981) administered formance measures in this study, the authors were GH to seven normal male subjects twice a day for cautious in suggesting that such findings might in- 1 week, at doses of 2 and 4 IU morning and even- growth hormone and sport 555

ing, resulting in an increase in RPF from 554 to disease develop GH resistance as far as the ana- 601 mL·min–1 × 1.73 m2 and GFR from 114 to bolic effects of GH are concerned, but appear to 125 mL·min–1 × 1.73 m2. GH has no acute effect on remain sensitive to the antinatriuretic effects of GH the renal hemodynamics or filtration within 2 h of (Wallace & Cuneo 1995). administration, but RPF and GFR increase and renal vascular resistance (RVR) decreases within 24 h, Effects of GH administration on sodium and fluid home- paralleling changes in serum IGF-I (Hirschberg et al. ostasis in normal subjects. To demonstrate the anti- 1989; Kopple & Hirschberg 1990). Guler et al. (1989a, natriuretic effect of GH on normal subjects, and to 1989b) have demonstrated that IGF-I administra- explore potential mechanisms for this effect, Ho, tion also increases RPF and GFR and decreases K.Y. and Weissberger (1990) administered GH to RVR resistance in normal human subjects. There- six male subjects aged between 21 and 27 years at fore, the hemodynamic effects of GH are now a dose of 0.2 IU·kg–1·day–1 (≅ 70 µg·kg–1·day–1) for thought to be largely mediated by IGF-I, and that 5 days. This short-term supra-physiological treat- IGF-I acts as a potent vasodilator, an effect mediated ment resulted in a mean weight gain of 1.1 kg, and by nitric oxide (Schini-Kerth 1999). The potential reduced mean 24-h urinary sodium excretion from impact of GH induced changes in renal hemo- 197 to 38 mmol. Urine output was significantly dynamics on exercise performance are unknown, reduced from 1652 to 848 mL·day–1. These changes but the increase in RPF probably reflects the reduc- were associated with activation of the renin– tion in systemic vascular resistance and increased angiotensin system (RAS). Plasma-renin activity in- cardiac output. creased from 730 to 3608 fmol angiotensin-I·L–1·s–1 and aldosterone increased from 52 to 402 pg·mL–1. Renal gluconeogenesis. GH may contribute to en- GH has also been shown to increase plasma angio- hanced fuel homeostasis during endurance exercise. tensinogen concentrations and angiotensin II recep- GH-induced increases in renal gluconeogenesis, tor density in the both the kidney and adrenal of taking up lactic acid and converting it to glucose for dwarf rats (Wyse et al. 1993), providing additional export to the working muscle, may be significant points of influence of GH and the RAS. during prolonged work (Poortmans 1977), but This data strongly supports the role of the RAS in experimental demonstration of this hypothesis is GH induced effects on fluid homeostasis. It is how- yet to be shown. ever proposed that a number of other mechanisms are also possible (Ho, K.Y. & Kelly 1991), as the Fluid homeostasis. GH administration results in fluid retention associated with chronic excess in sodium retention with subsequent fluid retention. acromegaly is not associated with activation of the This antinatriuretic effect of GH is apparent from a RAS. A number of experimental studies in animals number of clinical observations. GHD adults have suggest that GH may have a direct effect on renal reduced extracellular water and plasma volume tubular absorption of sodium independent of other which is increased with GH treatment (Møller et al. effects on the RAS (Stein et al. 1952; Karlberg & 1996). The chronic GH excess of acromegaly results Ottosson 1982). in a corresponding increase in total body water and blood volume (Ikkos et al. 1954; Bengtsson et al. Atrial natriuretic peptide. Atrial natriuretic peptide 1989b). Fluid retention is a well-documented side (ANP) is secreted by the heart usually in response effect of GH treatment, and can cause oedema, to activation of cardiac stretch receptors in response weight gain, carpal tunnel syndrome and hyperten- to increased plasma volume. An increase in serum sion when given in supra-physiological doses to ANP stimulates renal tubular sodium excretion. GHD adults (Jørgensen, J.O.L. 1991). Fluid retention Short-term GH administration to GHD subjects with GH treatment is dose dependent and trans- (7 days) has no effect on blood levels of ANP itory when physiological replacement is achieved (Hoffman et al. 1996); however, long-term treatment in these GHD adults. Subjects with chronic liver (6 months), which normalizes circulating IGF-I 556 chapter 36

levels, has been shown to normalize ANP with no be used as an alternative means to alter blood pH change in angiotensin II (Ekman et al. 2002). High and enhance exercise time at maximal power out- levels of GH administered to normal healthy adults put without the adverse effects of oral bicarbonate for 2 weeks results in a marked reduction in circulat- ingestion. ing ANP levels, which is associated with an expan- sion in extracellular fluid (ECF) volume but no Thermoregulation change in plasma volume (Møller et al. 1991). The lack of change in plasma volume would perhaps Sweating is one of the most important thermoregu- explain the unexpected fall in ANP in the presence latory homeostatic mechanism for the exercising of expanded ECF as changes in ECF alone would athlete, and a number of factors suggest that human not be expected to activate cardiac stretch receptors. sweat gland function is influenced by the GH–IGF-I Møller et al. (1991) propose that the reduction in axis. GHD is associated with reduced sweating ANP in response to GH may be due to stimulation capacity (Sneppen et al. 2000). Patients born with of ANP release from the atrias and a subsequent a GHR defect resulting in severely stunted growth fall in atrial pressure. This would be consistent with (Laron syndrome) have severely impaired sweat reports of GH-induced increased myocardial con- secretion rates (Main et al. 1993). Hypohydrosis in tractility in normal humans (Thuesen et al. 1988). GHD poses thermoregulatory difficulties during exercise due to a predisposition to hyperthermia Renal acid-base metabolism. The role of GH in the (Juul et al. 1995). In contrast, acromegaly is charac- regulation of renal acid-base metabolism has been terized by hyperhydrosis, an excessive activation of demonstrated in both animals and man. Welbourne sweating mechanisms (Sneppen et al. 2000). Success- & Cronin (1991) studied isolated, perfused kidneys ful reduction of excessive sweating in acromegalics from hypophysectomized rats which exhibited a has been repeatedly achieved in studies using the marked reduction in net acid secretion in response long-acting somatostatin analog octreotide (Stewart to GH but not IGF-I or insulin. GH was able to et al. 1995; Davies et al. 1998). accelerate tubular H+ secretion in kidneys from GHRs have been localized in human skin and both hypophysectomized and intact rats, and to cultured skin fibroblasts (Oakes et al. 1992). In that enhanced urinary acidification (Welbourne & Cronin study GHR were demonstrated on eccrine sweat 1991). GH has also been shown to directly affect glands using a monoclonal antibody. Lange, M. ammoniagenesis in the isolated renal proximal et al. (2001) demonstrated that untreated GHD tubules of dogs (Chobanian et al. 1992). The authors adults have thin skin and a reduction in the area of suggest a role for hGH in the regulation of acid- eccrine sweat gland glomeruli compared to normal base metabolism under physiological conditions in subjects, whereas the area of the sebaceous glands which increased net acid excretion is important. appeared normal. Long-term GH treatment resulted Homeostasis of acid-base metabolism during acute in normalization of eccrine sweat gland glomeruli exercise is clearly such a scenario. GH has been area, improvment of the very low sweat secretory implicated in systemic and renal acid-base home- rate but did not appear to fully return skin thickness ostasis in adult humans. Sicuro et al. (1998) showed to normal in childhood onset GHD. The research that GH administration to normal subjects was able groups of both Lange and Hassan (Hasan et al. 2001; to partially correct experimentally induced chronic Lange, M. et al. 2001) have examined skin biopsies metabolic acidosis by raising serum bicarbonate from GHD adults and showed reductions in acety- levels via an increase in net acid excretion. Acidosis lcholinesterase (AchE) and vasoactive intestinal contributes to fatigue in many types of exercise, and polypeptide (VIP) in the nerves supplying sweat ingestion of sodium bicarbonate has been demon- glands, and showed that GH treatment increased strated to prolong exercise time at maximal power staining for AchE and VIP in the sudomotor nerves output (Jones et al. 1977). Vomiting and diarrhea are and restored sweat rates. They also report that potential side effects of ingesting large doses of acromegalic subjects have larger sweat gland acini sodium bicarbonate, so it is possible that GH could and greater density of innervation to sweat glands growth hormone and sport 557

and propose that this suggests a trophic effect of ial but not parasitic infection. The authors suggest GH on sweat gland epithelium and/or associated that this effect may relate to the inhibition of tumor nerves. necrosis factor by GH (Sartin et al. 1998). It is possible that GH doping could enhance ther- In humans, GH appears to have some clinically moregulation in competition conducted in extreme relevant effects on immunity. Both GHD children heat via a number of mechanisms: by reducing fat and adults have been shown to have abnormalities mass and therefore insulation, by mechanisms asso- in immune function variables. Both the concentra- ciated with increased blood flow via vasodilatory tion and proportion of natural killer (NK) cells and effects, by direct effects on sweat gland morphology, the stimulated NK-cell activities were reduced in or by influencing sudomotor synapse constituents. adult GHD patients compared to controls. GH treat- Excessive GH-induced sweating could theoretically ment did not, however, appear to alter this situ- compromize fluid homeostasis during prolonged ation, and these patients do not appear to be more exercise in the heat. susceptible to infection than the non-GHD popula- tion (Sneppen et al. 2002). We are unaware of literat- ure suggesting that patients with acromegaly have Growth hormone and the immune system altered immunity. In contrast, supra-physiological All elements of the GH–IGF axis are represented in doses of rhGH in severely ill humans in intensive the immune system. GH, GHRs, IGFs and their care has been shown to increase mortality as a result receptors and binding proteins are expressed in of infection (Takala, J. et al. 1999). the thymus, lymph nodes and peripheral blood While effects on the immune system within the lymphocytes of both children and adults (Yang, Y. physiological range of GH actions appears minimal, et al. 1999). This suggests that GH and IGF-I may this does not rule out the possibility that GH have both an endocrine and an autocrine/paracrine administration may protect against the immuno- effect on the immune system. The thymus gland has suppression of stress. GH is considered to have an been described as a target organ for GH and the important role in the immune system with reviewers effects of both pituitary-derived and extra-pituitary of the available literature suggesting that GH and GH are summarized by Savino et al. (2002). Known IGF-I, along with prolactin and thyroid hormones, actions of GH include enhancement of thymic are critical immunoregulatory factors. Mice knock- microenvironmental cell-derived secretory products out models for the primary hormones or their such as cytokines and thymic hormones, increased receptors suggest that, although these hormones thymic epithelial cell (TEC) proliferation in vitro, may not be required for lymphocyte development synergism with anti-CD3 to stimulate thymocyte or antigen responsiveness, they most likely play a proliferation, influence on thymocyte traffic, and mayor role in counteracting the effects of stress- enhancement of human T-cell progenitor engraft- mediated negative immunoregulatory factors, such ment into the thymus. as glucocorticoids ( Jeay et al. 2002). GH and/or IGF- A number of animal studies have shown that I may therefore ensure immune system homeostasis treatment with GH reduces susceptibility to infec- and reduce the susceptibility to stress-induced dis- tion. For example, a controlled study of rhGH ad- ease (Dorshkind & Horseman 2001). ministration to experimentally burned mice showed Prolonged periods of intense exercise training in the impaired immune response in these animals, human adults decrease neutrophil function, serum as reflected by a decreased interferon-γ response and salivary immunoglobulin concentrations, NK of splenic mononuclear cells, was overcome with cell numbers and possibly cytotoxic activity in peri- GH treatment. GH treatment has been shown to pheral blood, and an increase in the incidence of stimulate the appearance of cytostatic macrophages upper respiratory tract infections (URTIs) has been and to result in an increased survival rate in these documented in athletes (Nass et al. 1995; Dahlgren burned mice infected with herpes symplex virus et al. 2002). The immunoprotective value of GH type I (Takagi et al. 1997). GH administration to administration to animals discussed above may cattle has been shown to be protective against bacter- lead athletes to conclude that GH doping may 558 chapter 36

protect them from URTI and subsequent training and vasoactive intestinal peptide and increases in down-time. GH, IGF-I, insulin-like growth factor binding pro- tein-3 (IGFBP-3) and β-endorphin concentrations in cerebrospinal fluid. This was a significant study as it Hematological demonstrated that GH could cross the blood brain Chronic GH exposure in acromegaly results in barrier. High levels of HVA have been observed in increased total blood volume, plasma volume and psychiatric patients with clinical depression, and red blood cell volume, all of which are reduced with successful treatment of depression with antide- treatment (Henneman et al. 1960; Merola et al. 1996). pressant drugs is associated with a reduction in This mild hematinic effect may potentially benefit cerebrospinal fluid (CSF) levels of HVA. This aerobic-based sporting performance. strongly suggests that improvements in mood with GH treatment are related to effects of GH and/ or additional components of the GH–IGF axis such Growth hormone and central nervous as IGF-I and IGFBP-3 on the CNS. In a cross- system effects over, placebo-controlled, 9-month treatment study, GH and IGF receptors are distributed widely Burman et al. (1996) also showed that CSF concen- throughout the human central nervous system trations of monoamine metabolites and neuropep- (CNS) (Johansson et al. 1995; Nyberg & Burman tides involved in the control of attention and mood 1996), suggesting a role for GH and IGF-I in the were abnormal in GHD patients and normalized CNS. Adults suffering from GHD report a lower with GH treatment. The dopamine metabolite HVA than normal quality of life when responding to self- significantly decreased with GH treatment and the rating questionnaires even when compared with excitatory amino acid aspartate increased with 9 subjects suffering other chronic illnesses. Psycholo- months of GH treatment. They confirm the previous ical deficits such as depression, mental fatigue, low findings that GH can cross the blood–brain barrier. self esteem and reduced levels of life fulfilment are The CSF levels of HVA and aspartate were shown seen to significantly contribute to the reduced qual- to be similar between GHD adults and psychiatric ity of life (Wallymahmed et al. 1999). GH treatment patients suffering with severe depression, with the of GHD adults results in significant improvements change in these neurotransmitters in GHD subjects in perception of energy level and mood in these following GH treatment being equivalent in magni- patients (McGauley et al. 1990; Carroll et al. 1998; tude to the changes achieved with antidepressent Cuneo et al. 1998). GHD adults are also reported medication in psychiatric patients. Unlike Johansson to suffer from cognitive deficits such as impaired et al. (1995) they did not observe a change in CSF memory and lack of concentration which can be levels of β-endorphin with GH administration. This improved with GH treatment (Burman & Deijen lack of change in β-endorphin may be due to differ- 1998). It has been argued that the effects of GH ences in GH dose or assay used or may merely mask treatment on psychological parameters are merely a an initial increase in β-endorphin with short-term secondary to increases in lean body mass, decreases treatment and an equalibration with chronic treat- in body fat and enhanced exercise performance, but ment An elevation of β-endorphins and have been the changes in mood and cognition often appear shown to following exercise and are associated with well before any measurable changes in physical a post-exercise increase in positive mood (Harte status, which lead researchers to suspect that GH et al. 1995). may have direct effects on CNS function. Heavy training in athletes has been associated GH has central effects on major monoamine meta- with negative changes in psychological factors bolites and neuropeptides involved in the control of such as depression, anger, global mood disturbance attention and mood. Johansson et al. (1995) showed and a reduction in general well-being (Morgan et al. that 1 month of GH treatment in GHD adults 1988). It is possible that overtraining contributes to resulted in decreases in homovallinic acid (HVA) changes in GH–IGF axis as reductions in IGF-I have growth hormone and sport 559

been noted with exhaustive exercise (Jahreis et al. might be how might one go about constructing such 1991; Tigranian et al. 1992). We speculate that GH a test to detect GH abuse. doping may allow an athlete to undergo harder training without negative changes in mood by per- Potential markers of growth hormone haps normalizing levels of monoamine metabolites doping and neuropeptides and thereby preventing psycho- logical disruptions which could lead to negative Measuring GH in blood or urine is unlikely to repres- impacts on concentration and motivation. ent a useful test of GH abuse as manufactured and native hGH molecules are identical, and normal secretion of GH is pulsatile and occurrs naturally in Summary of theoretical beneficial (and response to a number of stimuli including exercise. detrimental) effects of growth hormone abuse Studies of rhGH treatment in clinical conditions by athletes (adult GHD, sepsis, burns, catabolic states and liver Table 36.1 outlines the main potential benefits that disease) and of patients with over-secretion of GH can be extrapolated from the literature regarding from a pituitary tumor has enabled us to define a GH effects in adult humans. It must be emphasized number of blood-borne substances that change in that most of these conclusions remain speculative response to GH administration. These ‘markers of and await appropriate scientific scrutiny. Table 36.2 GH action’ include: (a) components of the GH–IGF outlines potential adverse effects of GH abuse in axis; (b) markers of bone and collagen turnover; and athletes. (c) molecular isoforms of GH. GH-responsive ele- ments of the GH and IGF axis include GHBP, IGF-I , Development of a test for growth and a number of IGFBPs a number of which are hormone abuse in elite sport: the used clinically to diagnose acromegaly. A detailed physiological lessons review of the regulation of the GH–IGF axis is beyond the scope of this article, but the reader is Given that GH is a banned substance under interna- referred Cuneo et al. (2001). Markers of bone and tional doping control regulations, the next question collagen turnover were chosen given the prolonged

Table 36.1 Hypothetical reasons why athletes might abuse growth hormone: potential but unproven.

Effect Sporting advantage

Bone Height increase Increased bone turnover and trauma repair Anabolic Increased lean : fat ratio ? Increased muscle or connective tissue Increased injury recovery Cardiac Inotropic and chronotropic increases in cardiac output Vasodilatory enhancement of peripheral perfusion Lipolytic Increased lean : fat ratio Sparing carbohydrate oxidation in endurance event Carbohydrate metabolism ? Increased muscle glycogen content ? Increased gluconeogenesis Antinatriuretic/renal Preservation of sodium and fluid homeostasis Enhanced acid excretion Hematological Increased red cell production Sweat gland Increased sweat-mediated heat dissipation Vascular Increased cutaneous vasodilation-mediated heat dissipation Immunological Improved immunity in training-mediated immune stress Central nervous system Enhanced mood/motivation 560 chapter 36

Table 36.2 Potential adverse effects Acral enlargement Permanent facial and peripheral bony disfigurement of growth hormone (GH) abuse in Obstructive sleep apnoea athletes. Accelerated osteoarthritis Permanent articular cartilage damage Endocrine Diabetes mellitus and carbohydrate intolerance Goitre Antinatriuresis/vascular Hypertension Acromegalic cardiomyopathy Muscles Acromegalic myopathy Neurological Acromegalic peripheral neuropathy Malignancy Increased incidence of malignancies, particularly colonic Cutaneous Hyperhydrosis, papillomata

stimulation of both bone resorption and formation variants of GH are due to the ability of 22 kDa and that follow rhGH administration; for example, such 20 kDa GH to form both homo and hetero dimers markers stay elevated for months in adults with and polymers up to pentameric GH. Further size GHD after cessation of rhGH (Ohlsson et al. 1998). variants can be attributed to the binding of GH with The third area of investigation, molecular isoforms GHBPs. Two main binding proteins are known to of GH, probably warrants additional background circulate: a high affinity binding protein which is the discussion for the reader’s benefit. extracellular portion of the GHR (Leung et al. 1987), and a larger, low-affinity binding protein consisting of transformed α -macroglobulin. The high-affinity Molecular isoforms of growth hormone 2 binding protein strongly binds 22 kDa GH and hGH is produced in the anterior pituitary gland, shows minimal binding to 20 kDa GH. The low- and circulates in a variety of forms. Two genes affinity binding protein weakly binds 22 kDa GH for GH can be found on the long arm of chromo- and binds the 20 kDa variant with equal or slightly some 17: hGH-N or ‘normal GH’ and hGH-V, a higher affinity (Baumann 1991). Glycosylation is ‘variant’ which is found to circulate during preg- also thought to result in a number of size variants of nancy. The main expression product of GH-N is GH: 12 kDa, 27 kDa (Lewis et al. 1994), 24 kDa (Haro a 191-amino acid polypeptide of molecular weight et al. 1996). Proteolytic cleavage between residues 22 kDa. Different species of GH can be gener- 43 and 44 of 22 kDa GH is thought to be responsible ated by either: (a) alternate splicing of mRNA; (b) for the 5 kDa GH1–43 and 17 kDa GH44–191 forms of post-translational modification; (c) heterodimeric, circulating GH. Degradation of GH in peripheral homodimeric, or polymeric binding; and (d) pro- tissues has been suggested as the explanation for teolytic cleavage. circulating forms such as 12, 16 and 30 kDa (see A number of alternatively spliced transcripts of Bauman 1991 for detailed review). Other post- the hGH-N gene have been described, with an exci- translational modifications occur which result in sion within exon 3 results in a 20 kDa form of GH charge variants of GH, the most common of which missing residues 32–46. Messenger RNAs have also are acetylation and deamidation. Circulating GH is been found which predict a 17.5 kDa GH variant in thought to consist of predominantly 22 kDa GH which the whole of exon 3 is skipped (Lecomte et al. (72%) and to a lesser extent the 20 kDa variant (20%) 1987) and a 27 kDa GH variant which includes (Baumann et al. 1985), with other GH variants only intron 4 (Hampson & Rottman 1987). Although GH occurring in very low concentrations. However, variants of these sizes have been found in human there is some data to suggest that the main circulat- serum, it is not yet known if they are products of ing variant in some normal individuals may in fact alternative mRNA-splicing, or if they represent be the 17 kDa GH44–191 variant (Warner et al. 1993; post-translational modifications. A number of size Sinha & Jacobsen 1994). growth hormone and sport 561

bined with hGH administration); and (c) the dis- Biological actions of growth hormone isoforms appearance kinetics of these markers following the Twenty-two kilodalton GH has been shown to be cessation of hGH. Seventeen males, aged 18–40 responsible for a wide range of biological actions. It years and body mass index (BMI) 23.6 ± 0.6, with a has been suggested that these often contradictory high level of habitual aerobic activity and aerobic o ± –1 –1 actions of GH could possibly be explained by the fitness (V 2max 56.0 1.2 mL·kg ·min ) were studied action of individual GH variants. Purification of 5, on seven occasions. Following screening (visit 1), 17 and 20 kDa GH has allowed determination of subjects were randomized to either rest or exercise their biological effects. Five kilodalton GH has been studies in a cross-over design with the rest study shown to be insulin potentiating and to overcome controlling for postural effects. Subjects were then insulin resistance in yellow obese mice, and 17 kDa re-randomized to a double-blind, placebo-con- GH proves to be a highly diabetogenic fragment trolled, parallel study of rhGH or identical placebo when injected into dogs. Twenty kilodalton GH treatment at a dose of 0.15 IU·kg–1·day–1 for 7 days. binds to GHRs (Wada et al. 1997, 1998), and when Treatment was self-administered by subcutaneous injected into humans was shown to suppress serum abdominal injection at 20:00 h; the 7th dose was 22 kDa GH levels and to elevate serum FFA and given 3 h prior to the post-treatment visit (visit 4). IGF-I (Hashimoto et al. 2000). This suggests that it The exercise protocol was repeated on visit 4, and has lipolytic and growth promoting actions. The following rhGH withdrawal at visits 5, 6 and 7, authors suggest that this study demonstrates that which were performed 24, 48 and 96 h after the ‘regulation of 20 kDa-hGH secretion is physiologic- post-treatment visit (visit 4). All submaximal exer- ally the same as that of 22 kDa-hGH and that 20 K- cise tests used an identical protocol, consisting of hGH regulates hGH through GH-induced negative three consecutive stages: stage 1 was 5 min at feedback’ (Hashimoto et al. 2000, p. 601). 1 W·kg–1; stage 2 was 5 min at 2 W·kg–1; and stage 3 was 20 min at 65% of the workload achieved at the o predetermined V 2max (corresponding to approxim- History of involvement with the GH-2000 project o ately 80% V 2max). In 1993 a review of GH and exercise (Cuneo & Acute exercise induced transient equimolar in- Wallace 1994) was commissioned by Peter Sonksen, creases of approximately 20% for serum total IGF-I, in his capacity as a member of the Medical Sub- IGFBP-3 and acid-labile subunit (ALS), exceeding Committee of the International Olympic Committee the effect of hemoconcentration (see Mottram 1999 (IOC). Subsequently, the IOC and the European and Figs 36.1 & 36.2). Union funded research to develop a GH dope- This suggests to us a novel mechanism for the detection strategy, to be conducted by members of regulation of the IGF ternary complex. These results the GH-2000 group, centered mainly in St. Thomas’s confirm findings by Bang et al. (1990) where serum Hospital, London (Professor Peter Sonksen), Aarhus IGF-I increased by 26%, but the study was con- Hospital, Denmark (Professor Jens S. Christiansen), ducted in a very small number of subjects (three Sahlgrenska Hospital, Sweden (Professor Bengt-Åke men and three women) with no rest control, with o Bengtsson) and Naples Hospital, Italy (Professor subjects cycling for 30 min at 60% of V 2max. Studies Luigi Sacca), including scientists from Germany, conducted since the design of our study, using a Switzerland and Australia. comparable exercise intensity, have also confirmed the response of IGF-I to acute exercise. Ten minutes o the washout study of cycle ergometry at 72% of V 2max in 11 young active adult subjects (10 male and one female), Our main contribution to this program was ‘The resulted in a 14% increase in serum IGF-I, which Washout Study’, which aimed to define: (a) the peaked at the end of exercise, and was not statistic- markers most indicative of hGH administration; ally different to baseline levels 20 min following (b) their response to acute exercise (alone and com- exercise (Cappon et al. 1994). This study however 562 chapter 36

100 ) 150 –1 )

–1 80 60 100 40 50 20 GH (mU·L

0 IGFBP-1 ( µ g·L 0 –30 0 30 60 90 120 –30 0 30 60 90 120 Time (min) Time (min)

1.6 ) 600 –1 1.2 400 0.8 200 0.4 GHBP (nmol)

0 IGFBP-2 ( µ g·L 0 –30 0 30 60 90 120 –30 0 30 60 90 120 Time (min) Time (min) Fig. 36.1 The effect of acute exercise on the growth hormone–insulin-like 40 ) 6

–1 growth factor (GH–IGF) axis. = 30 Subjects (n 17) underwent random 4 sequence exercise study (closed 20 triangles; semi-recumbency from 2 10 t =−30 until 0 min; exercise from t = 0 min with 10 min two-stage warm-up IGFBP-3 (mg·L 0 Total IGF-I (nmol) Total 0 –30 0 30 60 90 120 –30 0 30 60 90 120 and 20 min at 65% of the workload o Time (min) Time (min) at V 2max [oxygen consumption o approximately 80% V 2max]; semi- recumbency from t = 30 to 120 min) )

–1 0.8 300 or rest study (open triangles; upright posture during equivalent ‘exercise’ 0.6 200 period indicated by box). ALS, acid- 0.4 labile subunit; GH, growth hormone; 100 0.2 GHBP, growth hormone binding ALS (nmol) protein; IGF, insulin-like growth 0 Free IGF-I ( µ g·L 0 factor; IGFBP, insulin-like growth –30 0 30 60 90 120 –30 0 30 60 90 120 Time (min) Time (min) factor binding protein. (Mottram 1999.) did not have a rest control as part of the design. A unique aspect of our study was the measure- Similar results were presented by Schwartz et al. ment of serum free IGF-I in response to acute (1996) using an identical exercise protocol, in 10 exercise in athletes, which was only assayed at active young male subjects, showing a mean the beginning and end of exercise, due to the increase in serum IGF-I of 13.3% at the end of exer- labor intensity and expense of performing the assay. cise, with IGF-I returning to baseline after 10-min The circulating concentration of free IGF-I did not rest. This paper also shows that the IGF-I response change with acute exercise, indicating that the to exercise is dependent upon the intensity of increase in total serum IGF-I was due to an increase exercise and is GH independent, as exercise at in IGF-I bound to IGF binding proteins and ALS. o approximately 46% of V 2max was unable to stimu- A detailed discussion of the IGF-I/IGFBP-3/ALS late GH release but still elicited an mean IGF-I ternary complex will be expanded below. A sudden response of 7.7%. major increase in free IGF-I during exercise would growth hormone and sport 563

) 200 ) 200 –1 –1 150 150 100 100 50 50

IGFBP-1 ( µ g·L 0 IGFBP-1 ( µ g·L 0 –30 0 30 60 90 120 –30 0 30 60 90 120 Time (min) Time (min)

) 600 ) 600 –1 –1

400 400

200 200

IGFBP-2 ( µ g·L 0 IGFBP-2 ( µ g·L 0 –30 0 30 60 90 120 –30 0 30 60 90 120 Time (min) Time (min)

) 8 ) 8 –1 –1 6 6 4 4 2 2

IGFBP-3 (mg·L 0 IGFBP-3 (mg·L 0 –30 0 30 60 90 120 –30 0 30 60 90 120 Time (min) Time (min) Fig. 36.2 The effect of recombinant human growth hormone (rhGH) treatment on insulin-like growth 500 500 factor binding protein (IGFBP) -1, -2 400 400 and -3 and acid-labile subunit (ALS). 300 300 Subjects underwent identical exercise 200 200 tests before (open symbols) or after

ALS (nmol) 100 ALS (nmol) 100 (closed symbols) randomization to treatment with placebo (circles; left 0 0 –1 –1 –30 0 30 60 90 120 –30 0 30 60 90 120 panel) or rhGH (0.15 IU.kg .day ; Time (min) Time (min) squares; right panel) for 7 days. be counterproductive as it has potent insulin-like free IGF-I. If this is the case, we could hypothesize effects and would be expected to negatively affect that the appearance of a bound pool of IGF-I, during both glucose and fat supply. If there is a require- acute exercise, could act as an acute reservoir of ment for additional IGF-I during acute exercise it IGF-I. IGF-I has been shown to be stable throughout would therefore make physiological sense for it to the day and is therefore probably tightly regulated. be delivered in a bound but releasable form. The appearance of an additional pool in the face of a There are a number of potential functions of an metabolic challenge could serve to maintain equilib- increase in bound IGF-I during acute exercise. It rium of both free and bound concentrations of is possible that free IGF-I is metabolically active serum IGF-I, while providing the additional IGF-I during acute exercise. The lack of change in concen- for short-term demands. Alternatively, exercise tration of free IGF-I during acute exercise may mask could possibly signal the release of IGF-I bound to an increase in the disappearance rate, or turnover, of binding proteins which could then be immediately 564 chapter 36

directed to target tissues, with the rising bound levels perfused rat pancreas is exposed to IGF-I, insulin in the blood merely reflecting the residue from the secretion is suppressed (Leahy & Vandekerkhove acute influx. It is also possible that the additional 1990), and acute infusion of IGF-I in humans is able IGF-I appearing in the circulation in response to to lower insulin even in the face of an euglycemic exercise only becomes metabolized at the end of clamp (Boulware et al. 1992). It is possible, therefore, exercise. It may remain bound to its constitutive that a small acute rise in IGF-I with exercise could binding proteins in circulation, distributed to target facilitate energy supply in the form of glucose tissues or released into the circulation as free IGF-I. and fat by reducing the exposure of hepatocytes We did not measure free IGF-I following exercise and adipocytes to circulating insulin, which would and are therefore unable to report on the post- favor uptake and not release by these tissues. Hav- exercise relationship between bound and free IGF-I. ing an influx of IGFBP-3 and ALS with the exercise- The origin of exercise induced IGF-I is unknown. induced IGF-I would be expected to avoid high Circulating IGF-I is thought to be largely hepatic in concentrations of free IGF-I which could then act via origin, secreted in response to GH, but may also hepatic and adipocyte insulin receptors to influence arise due to leakage or perhaps active secretion from uptake and net release of these exercise fuels. IGF-I non-hepatic tissues. The time course of GH stimu- is also thought to have a protein sparing effect by lated hepatic IGF-I secretion is not rapid, so if the means of inhibiting proteolysis. An acute increase in liver is the source of the rapid increase in serum IGF-I in response to exercise could therefore serve IGF-I in response to exercise it is not likely to be GH- the function of reducing potential protein catabolic related. Electrical nervous stimulation of an isolated effects. A number of studies have demonstrated perfused cat limb results in increased IGF-I in the that IGF-I administration leads to reduced protein perfusate, suggesting that exercise could possibly degradation. For example, leucine oxidation was result in secretion of IGF-I from tissues other than shown to decrease with IGF-I administration to the liver (Sara et al. 1982), and in human studies the male subjects (Nussey et al. 1994), and the appear- exercising limb has been shown to be a source of ance of branched chain amino acids was markedly IGF-I (Brahm et al. 1997b). Therefore local IGF-I decreased after administration of IGF-I to male and mediated effects seem feasible as a result of these female subjects (Boulware et al. 1992). exercise-induced changes in IGFs. The appearance of IGF-I in response to acute The biological actions of increased circulating exercise may have a physiological role in glucose IGF-I in response to acute exercise are currently transport into working muscles. Peripheral glucose unknown. If we consider the tissue distribution of disposal is increased during acute exercise. It is pos- IGF-I receptors, and results of acute infusion of sible that the enhanced uptake of glucose by muscle recombinant human IGF-I into human subjects, during exercise is facilitated by an IGF-I mediated we are able to speculate on a number of possible increase in muscle blood flow. The potent arterial metabolic effects which may be facilitated by vasodilatory properties of IGF-I were demonstrated exercise-induced IGF-I. The two metabolic fuel by infusing IGF-I into the brachial artery of human reservoirs of the body, namely the liver and adipose nine male subjects and producing a rapid increase in tissue lack functional receptors for IGF-I and are forearm blood flow within 5 min of the infusion. probably therefore not the main targets of exercise- This effect was attributed to IGF-I-induced nitric related serum IGF-I increases. The pancreas and oxide synthesis, as blockade of nitric oxide synthesis skeletal muscle on the other hand are rich in IGF-I with mononetyyl-l-arginine abolished the increase receptors and would therefore be logical targets for in blood flow to IGF-I (Kiowski et al. 1994). These acute changes in IGF-I. Acute exercise is character- results have been confirmed in a larger study by ized by a fall in circulating insulin. It is possible that Fryburg (1996). Napoli et al. (2003) have demon- the acute increase in IGF-I and subsequent action strated that GH infusion into the brachial artery via pancreatic IGF-I receptors may contribute to the acutely stimulates endothelium dependent vasodi- exercise-associated fall in insulin. When the isolated lation and the related nitric oxide pathway. They growth hormone and sport 565

8

6

4

2

0 Delta SDS (pre-exercise data)

–2

–4 IGF-I IGFBP-1 IGFBP-2 IGFBP-3 ALS

Fig. 36.3 The relative responses to growth hormone (GH) treatment. Data has been transformed into standard deviation scores (SDS) using mean and standard deviation from the pretreatment, pre-exercise data for the entire study group. Results report the differences in SDS (mean ± SEM) from pretreatment visit to visits 4, 5, 6 and 7 (3, 24, 51 and 99 h after the last dose of recombinant human growth hormone [rhGH], respectively) in the GH group (black, dark gray, mid-gray and light gray, respectively) and placebo group (black, dark gray, mid-gray and light-gray oblique stripes, respectively). show that forearm IGF-I levels did not change sumably as a result of IGF-mediated negative feed- during the infusion and suggest that the vasodilat- back. Therefore IGF-I, and perhaps several IGFBPs, ory effect of GH is independent of circulating IGF-I. may be suitable markers of hGH abuse, given Subcutaneous infusion of IGF-I, in humans, has their diurnal stability and limited response to acute been shown to affect capillary permeability in skin exercise. and retina (Franzeck et al. 1995a, 1995b). It is poss- ible that this effect could be activated post-exercise GH-IGF axis, relative responses (Fig. 36.3). Data for in vascualture other than that associated with the individual analytes were converted into standard skin and eyes, and could be a potential mechanism deviation scores (SDS), using the means and stand- involved in the facilitation of transport of IGF-I ard deviations of resting, pretreatment values as binary complexes to target tissues. reference data. The following changes in response to IGFBP-1 showed a prominent post-exercise incre- 7 days of rhGH administration in the GH group ment of 50%, perhaps minimizing post-exercise were observed: IGF-I +6.3 ± 0.9, IGFBP-1 –2.2 ± 0.9, hypoglycemia. Acute exercise did not change IGFBP-2 –0.3 ± 0.2, IGFBP-3 +1.6 ± 0.5 and ALS +2.7 IGFBP-2 and free IGF-I. Treatment with rhGH ± 0.5 SDS, compared to respective changes in the resulted in marked increases in free and total IGF-I, placebo group of +0.5 ± 0.2, −0.8 ± 0.4, +0.2 ± 0.3, IGFBP-3 and ALS, and minor reductions in IGFBP-1 +0.1 ± 0.3 and +0.1 ± 0.2 SDS (Fig. 36.3). Using the and -2, with all of these hGH-induced changes lack of overlap of individual data points to separate disappeared over 4 days. The total hGH response the groups, serum IGF-I allowed correct identifica- to acute exercise was abolished for several days in tion of rhGH-treated individuals at visit 4 in 87.5% many individuals following rhGH treatment, pre- and 100% of cases for pre- and post-exercise time 566 chapter 36

) 0.6 –1 14 )

–1 0.5 µ g·L 12

µ g·L 0.4 10 0.3 PIIIP ( 0.2

Osteocalcin ( 8 –30 0 30 60 90 120 –30 0 30 60 90 120 Time (min) Time (min)

Fig. 36.4 The effect of acute endurance-type exercise on markers ) 5 –1 of bone and soft-tissue collagen

14 )

–1 turnover. Subjects (n = 17) 12 4 underwent random sequence 10 µ g·L exercise study (semi-recumbency 3 =− 8 from t 30 until 0 min; exercise ICTP ( from t = 0 min with 10-min two- 6 2 stage warm-up and 20 min at 65% B-S Alk Phos (IU·L –30 0 30 60 90 120 –30 0 30 60 90 120 o of the workload at V 2max; semi- Time (min) Time (min) recumbency from t = 30 to 120 min; solid symbols and continuous line) or rest study 250 (upright posture during equivalent

) ‘exercise’ period indicated by box; –1 200 open symbols and dotted line). B-S

µ g·L Alk Phos, bone-specific alkaline 150 phosphatase; ICTP, carboxyterminal cross-linked telopeptide of type I PICP ( 100 collagen; PICP, carboxyterminal –30 0 30 60 90 120 propeptide of type I procollagen; Time (min) PIIIP, procollagen III N-terminal extension peptide. points, respectively and in 75% and 87.5% of cases at PIIIP and ICTP, most impressively in the latter two. visit 5 (27 h after the last dose of rhGH). Similarly, Disappearance half-times following cessation of serum ALS allowed correct identification in 100% rhGH for pre- and post-exercise markers ranged and 87.5% for pre- and post-exercise time points, from 248 to 770 h. Therefore, markers of bone and 100% and 50% at visit 5. Serum IGFBP-3 was and collagen turnover may be suitable markers of less sensitive at each visit. hGH use. Endurance and resistance type exercise, Acute exercise transiently increased serum bone- recovery post-exercise and differences in training specific alkaline phosphatase (BS-ALP; +16.1%), must be considered when interpreting PIIIP and carboxyterminal propeptide of type I procollagen bone markers (Takala, T.E. et al. 1986, 1989; Virtanen (PICP; +14.1%), procollagen III N-terminal exten- et al. 1993; Brahm et al. 1996, 1997a, 1997c). Given sion peptide (PIIIP; +5.0%), carboxyterminal cross- that PIIIP was perhaps the most promising ‘bone’ linked telopeptide of type I collagen (ICTP; +9.7%), marker to detect recent GH abuse, particular without change in osteocalcin (Fig. 36.4) (Karila concern must be expressed since this is a marker of et al. 1999). These changes clearly show a transient collagen turnover, and ligamentous injury may be increase in bone turnover following endurance considered a case for a ‘false positive test’. exercise with limited weight bearing. Exogenous To demonstrate the relative magnitude of res- rhGH treatment increased serum osteocalcin, PICP, ponse to acute exercise of all forms of GH measured, growth hormone and sport 567

100 ment groups was possible in most cases, but the Total GH effect may only be detectable for 24 h. Therefore, 22 kDa GH 90 Pit-GH molecular isoforms of hGH may be suitable markers rhGH of very recent rhGH administration. In our study, 80 GH–IFA 20 kDa GH urinary GH measurement was of limited use in non-22 kDa GH detecting rhGH administration (Wallace et al. un- 70 published data). ) –1 60 Progress after The Washout Study: 50 GH-2000 and others 40 One of the main sets of markers of GH action in ath-

Serum GH (mU·L letes revealed by The Washout Study was the IGF 30 system. Dall et al. (2000) extended these concepts in a study that recruited 50 male and 49 female subjects 20 aged 18–40 years. Subjects were healthy but not 10 athletic, given the entry requirement of two exer- cise sessions per week for 1 year. In a double-blind, 0 placebo-controlled, parallel dose finding study –30 0 30 60 90 120 rhGH (0.1 and 0.2 IU·kg–1·day–1) was administered Time (min) subcutaneously each evening. Unfortunately 33% and 20% of women in the high- and low-dose treat- Fig. 36.5 The relative magnitude, of responses of different molecular isoforms of growth hormone (GH), to acute ment groups, respectively, used oral contraceptives. endurance-type exercise. GH–IFA, immunofunctional This observation is important in interpreting the assay for growth hormone; Pit-GH, pituitary growth serum IGF-I responses, since oral estrogens inhibit hormone; rhGH, recombinant human growth hormone. hepatic IGF-I generation (Weissberger et al. 1991). Data presented as mean ± SEM. Serum IGF-I clearly increased with rhGH, with greater increases in men than women (Ho, K.K. serum concentrations have been converted into SI & Weissberger 1992; Burman et al. 1997), and in the units and presented together in Fig. 36.5. latter group a treatment effect was only seen in the The human pituitary produces several species of higher dose group. The mean concentrations are not hGH, the most abundant being 22 kDa, with smaller reported. Differentiation between the placebo and amounts of 20, 17, and 5 kDa monomeric isoforms other groups was reported, using 4 SD above the and oligomeric forms (Baumann 1991). Suppression whole group baseline data at 21 or 28 days of treat- of the non-22 kDa isoforms following exogenous ment as 50% and 33% in the high-dose female, and 22 kDa rhGH administration would be expected on 7% and 23% in the low-dose female group. In men, the basis of IGF-I mediated negative feedback. In corresponding figures were 86% and 64% in the collaboration with Swedish and German colleagues high-dose and 73% and 60% in the low-dose group. (Boguszewski et al. 1996; Wu et al. 1999), we have Differentiation persisted after cessation of rhGH for shown that acute endurance-type exercise increased 2 days in 13% and 6% in high- and low-dose female both 22 and non-22 kDa isoforms (Fig. 36.5). The groups and in 20% and 33% in high- and low-dose relative proportion of the latter increased from male groups. No differentiation was possible 5 days approximately 3% to 9% across the exercise proto- after the last dose. Serum IGFBP-3 and ALS were col. Treatment with rhGH suppressed the exercise- much less discriminatory between groups, but the induced responses, with an increase in the ratio of data was not presented. The IGF-I/IGFBP-3 ratio 22 kDa/total hGH and a decrease in non-22 kDa/ showed an interesting gender difference, with an total GH ratios (Fig. 36.6). Distinction between treat- apparent ceiling effect at 0.1 IU·kg–1·day–1 in the 568 chapter 36 ) 4 ) 4 –1 –1 3.5 3 3 2.5 2 2 1.5 1 1 0.5 20 kDa GH (ng·mL 0 20 kDa GH (ng·mL 0 0 10 20 30 40 0 50 100 150 200 GH-IFA (ng·mL–1) 22 kDa GH (mU·L–1) ) –1

) 4 15 –1 3.5 3 12 2.5 9 2 1.5 6 Fig. 36.6 Regression analysis of 1 3 non-22 kDa growth hormone 0.5 (GH) assays against total GH. 20 kDa GH (ng·mL 0 0 Data represent ‘normal’ or Non-22 kDa GH (mU·L 0 50 100 150 200 0 50 100 150 200 untreated individuals (i.e. data at –1 –1 Total GH (mU·L ) Total GH (mU·L ) all time points from the placebo group from rest and exercise 20 200 studies from visits 2–7 and recombinant human growth 15 150 hormone [rhGH] groups from before intervention for visits 2 10 100 and 3; open circles). The rhGH group at visit 4 (i.e. 3 h after the 5 50 last dose; solid squares), visit 5

rhGH : Pit-GH (%) (open triangles), visit 6 (open 0 0 diamonds) and visit 7 (double

Non-22 kDa : Total GH (%) 0 50 100 150 200 0 20 40 60 80 crosses) are shown separately. Pit- Total GH (mU·L–1) Pituitary GH (ng·mL–1) GH, pituitary growth hormone.

males but a more progressive dose response effect in 10% and 15% for IGF-I, IGFBP-3 and ALS over seven females. The authors considered this to be a good time points across 84 days. These data may con- biological marker of GH status in other conditions tribute to a detection system based on repeated (e.g. acromegaly), noted that it remained elevated measurements in elite athletes, where starting or for 2 weeks after ceasing rhGH in the current study, stopping hGH abuse may be detected as a change in but did not present this data in terms of being able excess of specific biological variability in an individ- to discriminate between treated and untreated ual athlete. To date, no data on the variation of such individuals. These data demonstrate that serum markers in truly elite athletes has been published to IGF-I (and perhaps IGF-I/IGFBP-3 ratio) may be a validate such an approach. Additional confounders suitable marker to detect rhGH abuse in a cross- that may reasonably be expected to contribute to sectional fashion, using a single time point sample day-to-day or month-to-month variability in IGF-I in an athlete compared to a population data set of may include acute exercise (Cuneo & Wallace 1994), comparable athletes. training phases (Hagberg et al. 1982; Jahreis et al. This same study examined the stability of the 1991; Tigranian et al. 1992), oral estrogens (Weiss- IGF system in the placebo groups. The mean intra- berger et al. 1991), pregnancy, anabolic steroid use individual coefficient of variation (CV) in the com- (Karila et al. 1999), illness or injury and season. bined male and female placebo groups was between Markers of bone turnover as potential agents to growth hormone and sport 569

detect hGH abuse were examined in the same football, rowing, etc.). Accordingly, the nature and cohort of healthy active adults. Markers of bone duration of the maximal tests were very different formation (osteocalcin, and PICP), bone resorption between sporting disciplines, and the majority of (ICTP) and extra-osseous collagen (PIIIP) were tests were not discipline specific (all skiers did examined. Stability of all markers in the placebo treadmill tests, and tennis players, footballers, groups was documented (CVs between 21% and swimmers, etc., performed cycle ergometry tests). 30%), but was somewhat higher than for the IGF The protocols employed would not appear to mimic axis markers. Examination of diurnal effects was competition. No resting control data were collected. not part of the study design, a potentially important Statistically significant increases in IGF-I, IGFBP-3, consideration as discussed by Neilsen et al. (1990). ALS, ICTP and PIIIP were demonstrated, peaking at Menstrual cycle was said to not influence these the end of exercise and declining to baseline values markers, but the statistical method to establish this over the subsequent 30–60 min. These results were conclusion was not presented. All four markers almost identical to The Washout Study. The limita- increased with rhGH treatment, with some import- tions of this study include the pooling of vastly dif- ant marker-specific differences. For example, men ferent sports, body types and stimuli applied, which had greater increments in PIIIP and ICTP, again appears to have resulted in very non-normally dis- reflecting the lower responses in women as noted tributed data given the non-parametric analysis for the IGF axis. There was a dose-dependent effect strategy used. While efforts to determine covariates with extra-osseous collagen marker (PIIIP) and the of some of these values has been undertaken, sport- bone resorption marker (ICTP) having a clearly ing discipline or nature of the physical stimulus greater response with the higher rhGH dose. Time- applied was not considered. The authors then specific responses were noted with more prolonged present 10–90 percentile concentrations for these effects after withdrawal for osteocalcin and PIIIP. In markers, but realistically many more observations general, bone and collagen markers remained elev- would be required to consider such data as ated for longer periods after withdrawal of rhGH normative and defining of an abnormal value in a treatment, some up to 56 days. The results were rhGH-abusing athlete. An important practical con- presented in a different fashion to the IGF marker sideration arising from this and earlier studies is publication, using a discriminant analysis to assess the recommendation to delay blood sampling from the utility of a single marker and combinations of athletes until at least 30 min after the completion markers to differentiate between the placebo and of exhaustive competition. This should be borne the treated group. False positive rates, an unaccept- in mind when considering the large scale cross- able situation where the test might falsely accuse sectional studies of markers in elite athletes, where a ‘clean’ athlete of abusing rhGH, were lowest after this criterion has not been uniformly applied. 4 weeks of rhGH (between 2.7% and 12.8%). Sens- Cross-sectional normal values for such markers itivity was 84–87% at the best time point, although have recently been reported in abstract form by the level of discrimination in terms of standard a GH-detection study group based in Australia deviations was not presented. (Howe et al. 2003). The aim was to provide reference GH-2000 also examined the effect of maximal data against which an athlete could be judged exercise in elite athletes on the above markers of the to either be normal or to be influenced by rhGH IGF axis and markers of bone and collagen turnover doping. As expected, widely varying results are (Ehrnborg et al. 2003). While the GH response to seen between differing sporting disciplines, differ- acute exercise has been well described, the effect of ent races and different body morphologies. This maximal exercise on most of these serum markers wide variability persists after statistical correction was unknown, certainly for the elite athlete. They for a number of morphological variables, suggest- studied 117 elite athletes spanning endurance sports ing that the use of a single serum sample of IGF-I (long-distance cyclists and cross-country skiers), (or related IGFBPs) will have low sensitivity. In power sports (weightlifters and sprint cyclists) and addition, the timing of venesection after competi- a mixture of other skill-based sports (alpine skiing, tion has not been considered in the study design, 570 chapter 36

contributing to a component of the variability, fur- nature of the tests; and (d) the acceptability of blood ther weakening the robustness of such a detection sampling as opposed to urine sampling. Blood strategy. sampling has been permitted at the Lillehammer Interestingly, the effects of GH administration and Nagano Winter Olympics, and it is believed on high intensity aerobic exercise and metabol- that many athletes may prefer blood sampling to the ism has recently been published (Lange, K.H. et al. invasion of privacy involved in urine collection. 2002). The authors studied seven cycle-trained males o ± –1 –1 (mean V 2max 65 1 mL·kg ·min ) in a randomized, Alternative testing methods. Mass spectrometry of double-blind, placebo-controlled, cross-over study proteins and stable isotope ratio assessment of the to assess the effect of a large, single, subcutaneous differences between endogenous and exogenous dose of rhGH (2.5 mg) 2 h prior to exercise. They proteins are being explored but as yet do not appear found that two subjects could not complete the 90- practical for widespread use as a test. min exercise protocol (staged at a workload that o elicited 65% and 75% of a predetermined V 2max) The social implications. Side effects of performance during the rhGH day but could do so on the placebo enhancing drugs appear not to influence elite day. While the significance of this finding is perhaps athletes. This apparent endorsement filters down to open to question, there were clear rhGH-induced drive the use of performance-enhancing drugs in increases in lipolytic responses (increased glycerol school-age athletes. We should endeavor to educate and FFA) and, surprisingly, in lactate. Unpublished our future athletes and put in place suitable tests to results from The Washout Study corroborate these deter widespread use of these agents. metabolic findings. The simple interpretation is that GH indices a lypolytic response and that this Acknowledgments may result in a competition for the type of fuel used during exercise, favoring lipid substrates as We would like to thank members of GH-2000, a opposed to carbohydrate sources. This would be multinational group investigating means of devel- consistent with the apparent decline in endurance oping a test to detect GH abuse in elite sport, who capacity, since lipid oxidation requires more oxygen have assisted in the generation of elements of this than does carbohydrate. work, and include Peter Sonksen (Team Leader, St. These studies confirmed the findings of The Thomas’s Hospital, London), Bengt-Åke Bengtsson Washout Study, allow the confident selection of (Sahlgrenska Hospital, Gothenberg, Sweden), Jens suitable markers and go a long way toward pro- Sandahl Christiansen (Aarhus Community Hospital, viding suitable reference data against which an Aarhus, Denmark), Luigi Sacca (Naples Hospital, athlete’s results could be compared. Questions Naples, Italy) and Christian Strasburger (University remain regarding: (a) the exact form of the testing Hospital, Munich, Germany). strategy (in- and out-of-competition testing); (b) We also gratefully acknowledge the support and which markers are suitable in non-Caucasian ath- collaboration of all members of GH-2000. Mr. Edward letes; (c) the statistical certainty required to label an Mulry has supported our bone research and we are athlete as having used hGH given the indirect eternally thankful to both he and his wife.

References

Alford, F.P., Hew, F.L., Christopher, M.C. D.G. (1995) Effect of glucocorticoid low molecular weight binding protein in & Rantzau, C. (1999) Insulin sensitivity replacement therapy on glucose healthy subjects and patients with in growth hormone (GH) -deficient tolerance and intermediary metabolites growth hormone deficiency. European adults and effect of GH replacement in hypopituitary adults. Clinical Journal of Clinical Investigation 20, 285–292. therapy. Journal of Endocrinological Endocrinology 42, 85–90. Baumann, G. (1991) Growth hormone Investigation 22, 28–32. Bang, P., Brandt, J., Degerblad, M. et al. heterogeneity: genes, isohormones, Al-Shoumer, K.A.S., Beshyah, S.A., (1990) Exercise-induced changes in variants, and binding proteins. Niththyananthan, R. & Johnston, insulin-like growth factors and their Endocrine Reviews 12, 424–449. growth hormone and sport 571

Baumann, G., Stolar, M.W. & Amburn, K. factor I on glucose, lipid, and amino Capaldo, B., Patti, L., Oliviero, U. et al. (1985) Molecular forms of circulating acid metabolism. American Journal of (1997) Increased arterial intima-media growth hormone during spontaneous Physiology. Endocrinology and Metabolism thickness in childhood-onset growth secretory episodes and in the basal state. 262, E130–E133. hormone deficiency. Journal of Clinical Journal of Clinical Endocrinology and Brahm, H., Piehl-Aulin, K. & Ljunghall, S. Endocrinology and Metabolism 82, Metabolism 60, 1216–1220. (1996) Biochemical markers of bone 1378–1381. Belcher, H.J.C.R. & Ellis, H. (1990) metabolism during distance running in Cappon, J., Brasel, J.A., Mohan, S. & Somatotropin and wound healing after healthy, regularly exercising men and Cooper, D.M. (1994) Effect of brief injury. Journal of Clinical Endocrinology women. Scandinavian Journal of Medicine exercise on circulating insulin-like and Metabolism 70, 939–943. and Science in Sports 6, 26–30. growth factor I. Journal of Clinical Bengtsson, B.-Å., Brummer, R.-J., Edén, S. Brahm, H., Piehl-Aulin, K., & Ljunghall, S. Endocrinology and Metabolism 76, & Bosaeus, I. (1989a) Body composition (1997a) Bone metabolism during 2490–2496. in acromegaly. Clinical Endocrinology 30, exercise and recovery: the influence of Carroll, P.V., Christ, E.R., Bengtsson, B.-Å. 121–130. plasma volume and physical fitness. et al. (1998) Growth hormone deficiency Bengtsson, B.-Å., Brummer, R.-J., Edén, S., Calcified Tissue International 61, 192–198. in adulthood and the effects of growth Bosaeus, I. & Lindstedt, G. (1989b) Body Brahm, H., Piehl-Aulin, K., Saltin, B. & hormone replacement: a review. Journal composition in acromegaly: the effect Ljunghall, S. (1997b) Net fluxes over of Clinical Endocrinology and Metabolism of treatment. Clinical Endocrinology 31, working thigh of hormones, growth 83, 382–395. 481–490. factors and biomarkers of bone Cheng, J.S. & Kalant, N. (1970) Effects of Bengtsson, B.-Å., Edén, S., Lönn, L. et al. metabolism during short lasting insulin and growth hormone on the flux (1993) Treatment of adults with growth dynamic exercise. Calcified Tissue rates of plasma glucose and plasma free hormone (GH) deficiency with International 60, 175–180. fatty acids in man. Journal of Clinical recombinant human GH. Journal of Brahm, H., Ström, H., Piehl-Aulin, K., Endocrinology 31, 647–653. Clinical Endocrinology and Metabolism Mallmin, H. & Ljunghall, S. (1997c) Chobanian, M.C., Julin, C.M., Molteni, 76, 309–317. Bone metabolism in endurance trained K.H. & Brazy, P.C. (1992) Growth Beshyah, S.A., Markussis, V., Harbourne, athletes: a comparison to population- hormone regulates ammoniagenesis in T., Nicolaides, A.N. & Johnston, D.G. based controls based on DXA, SXA, canine renal proximal tubule segments. (1993) Haemostatic mechanisms are quantitative ultrasound, and American Journal of Physiology 262, normal despite increased vascular biochemical markers. Calcified Tissue F878–F884. mortality in hypopituitary adults. International 61, 448–454. Christiansen, J.S., Gammelgaard, J., Hormone and Metabolic Research 25, Bramnert, M., Segerlantz, M., Laurila, E. Ørskov, H. et al. (1981) Kidney function 449–450. et al. (2003) Growth hormone and size in normal subjects before and Beshyah, S.A., Henderson, A., replacement therapy induces insulin during growth hormone administration Niththyananthan, R. et al. (1995) The resistance by activating the glucose– for 1 week. European Journal of Clinical effects of short and long term growth fatty acid cycle. Journal of Clinical Investigation 11, 487–490. hormone replacement therapy in Endocrinology and Metabolism 88, Christopher, M., Hew, F.L., Oakley, M., hypopituitary adults on lipid 1455–1463. Rantzau, C. & Alford, F. (1998) Defects metabolism and carbohydrate tolerance. Bulow, B., Hagmar, L., Mikoczy, Z., of insulin action and skeletal muscle Journal of Clinical Endocrinology and Nordstrom, C.H. & Erfurth, E.M. (1997) glucose metabolism in growth hormone- Metabolism 80, 356–363. Increased cerebrovascular mortality in deficient adults persist after 24 months Blackett, P.R., Weech, P.K., McConathy, patients with hypopituitarism. Clinical of recombinant human growth hormone W.J. & Fesmire, J.D. (1982) Growth Endocrinology 46, 75–81. therapy. Journal of Clinical Endocrinology hormone in the regulation of Burman, P. & Deijen, J.B. (1998) Quality and Metabolism 83, 1668–1681. hyperlipidemia. Metabolism 31, 117–120. of life and cognitive function in Cittadini, A., Berggren, A., Longobardi, S. Böger, R.H., Skamira, C., Bode-Böger, S.M. patients with pituitary insufficiency. et al. (2002) Supraphysiological doses et al. (1996) Nitric oxide may mediate the Psychotherapy and Psychosomatics 67, of GH induce rapid changes in cardiac hemodynamic effects of recombinant 154–167. morphology and function. Journal of growth hormone in patients with Burman, P., Hetta, J., Wide, L. et al. (1996) Clinical Endocrinology and Metabolism 87, acquired growth hormone deficiency. Growth hormone treatment affects brain 1654–1659. A double-blind, placebo-controlled neurotransmitters and thyroxine. Colao, A., Marzullo, P., Ferone, D. et al. study. Journal of Clinical Investigation 98, Clinical Endocrinology 44, 319–324. (2000) Cardiovascular effects of depot 2706–2713. Burman, P., Johansson, A.G., Siegbahn, A., long-acting somatostatin analog Boguszewski, C.L., Hynsjo, L., Johannsson, Vessby, B. & Karlsson, F.A. (1997) sandostatin LAR in acromegaly. Journal G., Bengtsson, B.-Å. & Carlsson, L.M. Growth hormone (GH) -deficient men of Clinical Endocrinology and Metabolism (1996) 22-kD growth hormone exclusion are more responsive to GH replacement 85, 3132–3140. assay: a new approach to measurement therapy than women. Journal of Clinical Collip, P.J., Curti, V., Thomas, J. et al. (1973) of non-22-kD growth hormone isoforms Endocrinology and Metabolism 82, Body composition changes in children in human blood [comment]. European 550–555. receiving human growth hormone. Journal of Endocrinology 135, 573–82. Caidahl, K., Edén, S. & Bengtsson, B.-Å. Metabolism: Clinical and Experimental 22, Boulware, S.D., Tamborlane, W.V., (1994) Cardiovascular and renal 589–595. Matthews, L.S. & Sherwin, R.S. (1992) effects of growth hormone. Clinical Crist, D.M., Peake, G.T., Egan, P.A. & Diverse effects of insulin-like growth Endocrinology 40, 393–400. Waters, D.L. (1988) Body composition 572 chapter 36

response to exogenous GH during (Karch, S.B., ed.). CRC Press, Boca renin–angiotensin–aldosterone system. training in highly conditioned adults. Raton, FL: 690–710. Clinical Endocrinology 57, 473–479. Journal of Applied Physiology 65, 579–584. Cuocolo, A., Nicolai, E., Colao, A. et al. Fazio, S., Cittadini, A., Sabatini, D. et al. Cummings, M.H., Christ, E.R., Umpleby, (1996) Improved left ventricular (1997) Growth hormone and heart A.M. et al. (1997) Abnormalities of very function after growth hormone performance. A novel mechanism of low density lipoprotein apolipoprotein replacement in patients with cardiac wall stress regulation in humans B-100 metabolism contribute to hypopituitarism: assessment with [see comments]. European Heart Journal the dyslipidemia of adult growth radionuclide angiography. European 18, 340–347. hormone deficiency. Journal of Clinical Journal of Nuclear Medicine 23, 390–394. Fowelin, J., Attvall, S., Lager, I. & Endocrinology and Metabolism 82, Dahlgren, L.A., van-der-Meulen, M.C., Bengtsson, B.-Å. (1993) Effects of 2010–2013. Bertram, J.E., Starrak, G.S. & Nixon, A.J. treatment with recombinant human Cuneo, R.C. (1998) Efficacy of growth (2002) Insulin-like growth factor-I growth hormone on insulin sensitivity hormone-replacement therapy: improves cellular and molecular aspects and glucose metabolism in adults with lipid and cardiovascular effects. of healing in a collagenase-induced growth hormone deficiency. Metabolism Endocrinologist 8, 22S–30S. model of flexor tendinitis. Journal of 42, 1443–1447. Cuneo, R.C. & Wallace, J.D. (1994) Growth Orthopaedic Research 20, 910–919. Franzeck, U.K., Dorffler-Melly, J., hormone, insulin-like growth factors Dall, R., Longobardi, S., Ehrnborg, C. et al. Hussain, M.A. et al. (1995a) Effects of and sport. Endocrinology and Metabolism (2000) The effect of 4 weeks of supra- subcutaneous insulin-like growth factor- Clinics of North America 1, 3–13. physiological GH administration on the I infusion on skin microcirculation. Cuneo, R.C., Salomon, F., Wiles, C.M. & IGF-axis in women and men. Journal of International Journal of Microcirculation, Sönksen, P.H. (1990) Skeletal muscle Clinical Endocrinology and Metabolism 85, Clinical and Experimental 15, 10–13. performance in adults with growth 4193–4200. Fryburg, D.A. (1996) NG-monomethyl- hormone deficiency. Hormone Research Davidson, M.B. (1987) Effect of growth l-arginine inhibits the blood flow but not 33 (suppl. 4), 55–60. hormone on carbohydrate and lipid the insulin-like response of forearm Cuneo, R.C., Salomon, F., Wiles, C.M., metabolism. Endocrine Reviews 8, muscle to IGF-I: possible role of nitric Hesp, R. & Sönksen, P.H. (1991a) 115–131. oxide in muscle protein synthesis. Growth hormone treatment in growth Davies, P.H., Stewart, S.E., Lancranjan, L., Journal of Clinical Investigation 97, hormone-deficient adults. I. Effects on Sheppard, M.C. & Stewart, P.M. (1998) 1319–1328. muscle mass and strength. Journal of Long-term therapy with long-acting Gibney, J., Wallace, J.D., Spinks, T. Applied Physiology 70, 688–694. octreotide (sandostatin-LAR) for the et al. (1999) The effects of 10 years of Cuneo, R.C., Salomon, F., Wiles, C.M., management of acromegaly. Clinical recombinant human growth hormone Hesp, R. & Sönksen, P.H. (1991b) Endocrinology 48, 311–316. (GH) in adult GH-deficient patients. Growth hormone treatment in growth De Boer, H., Blok, G.J., Voerman, H.J., Journal of Clinical Endocrinology and hormone-deficient adults. II. Effects on De Vries, P.M.J.M. & van der Veen, E.A. Metabolism 84, 2596–2602. exercise performance. Journal of Applied (1992) Body composition in adult Gibney, J., Healy, M.L., Stolinski, M. Physiology 70, 695–700. growth hormone-deficient men, et al. (2003) Effect of growth hormone Cuneo, R.C., Salomon, F., Wilmshurst, F. assessed by anthropometry and (GH) on glycerol and free fatty acid et al. (1991c) Cardiovascular effects of bioimpedance analysis. Journal of metabolism during exhaustive exercise growth hormone treatment in growth- Clinical Endocrinology and Metabolism 75, in GH-deficient adults. Journal of Clinical hormone-deficient adults: Stimulation 833–837. Endocrinology and Metabolism 88, of the renin–aldosterone system. Clinical Deyssig, R., Frisch, H., Blum, W.F. & 1792–1797. Science 81, 587–592. Waldhör, T. (1993) Effect of growth Guler, H.-P., Eckardt, K., Zapf, J., Bauer, C. Cuneo, R.C., Salomon, F., Wiles, C.M. et al. hormone treatment on hormonal & Froesch, E.R. (1989a) Insulin-like (1992) Histology of skeletal muscle in parameters, body composition and growth factor I increases glomerular adults with GH deficiency: comparison strength in athletes. Acta Endocrinologica filtration rate and renal plasma flow in with normal muscle and response to GH 128, 313–318. man. Acta Endocrinologica 121, 101–106. treatment. Hormone Research 37, 23–28. Dorshkind, K. & Horseman, N.D. (2001) Guler, H.-P., Schmid, C., Zapf, J. & Cuneo, R.C., Salomon, F., Watts, G.F., Anterior pituitary hormones, stress, and Froesch, E.R. (1989b) Effects of Hesp, R. & Sönksen, P.H. (1993) Growth immune system homeostasis. Bioessays recombinant insulin-like growth factor I hormone treatment improves serum 23, 288–294. on insulin secretion and renal function lipids and lipoproteins in adults with Ehrnborg, C., Lange, K.H., Dall, R. et al. in normal human subjects. Proceedings of growth hormone deficiency. Metabolism (2003) The growth hormone/insulin-like National Academy of Sciences of the United 42, 1519–1523. growth factor-I axis hormones and bone States of America 86, 2868–2872. Cuneo, R.C., Judd, S., Wallace, J.D. et al. markers in elite athletes in response to Hagberg, J.M., Seals, D.R., Yerg, J.E. et al. (1998) The Australian multicenter trial a maximum exercise test. Journal of (1982) Metabolic responses to exercise in of growth hormone (GH) treatment in Clinical Endocrinology and Metabolism young and older athletes and sedentary GH-deficient adults. Journal of Clinical 88, 394–401. men. Journal of Applied Physiology 65, Endocrinology and Metabolism 83, Ekman, B., Ohman, P.K., Arnqvist, H.J., 900–908. 107–116. Lindstrom, T. & Nystrom, F.H. (2002) Hampson, R.K. & Rottman, F.M. (1987) Cuneo, R.C., Wallace, J.D. & Sönksen, P. Individualized growth hormone Alternative processing of bovine growth (2001) Growth hormone abuse in elite substitution with normalized IGF-I hormone mRNA: nonsplicing of the final athletes. In: Drug Abuse Handbook levels does not stimulate the intron predicts a high molecular weight growth hormone and sport 573

variant of bovine growth hormone. Ho, K.Y. & Kelly, J.J. (1991) Role of growth increase isometric and isokinetic muscle Proceedings of National Academy of hormone in fluid homeostasis. Hormone strength in GH-deficient adults. Journal Sciences of the United States of America 84, Research 36 (suppl. 1), 44–48. of Clinical Endocrinology and Metabolism 2673–2677. Ho, K.Y. & Weissberger, A.J. (1990) The 82, 2877–2884. Harant, I., Beauville, M., Crampes, F. et al. antinatriuretic action of biosynthetic Johansson, J.O., Landin, K., Tengborn, L., (1994) Response of fat cells to growth human growth hormone in man Rosen, T. & Bengtsson, B.-Å (1994) High hormone (GH): effect of long term involves activation of the renin– fibrinogen and plasminogen activator treatment with recombinant human GH angiotensin system. Metabolism 39, inhibitor activity in growth hormone- in GH-deficient adults. Journal of Clinical 133–137. deficient adults. Arteriosclerosis and Endocrinology and Metabolism 78, Hoffman, D.M., Crampton, L., Sernia, C., Thrombosis 14, 434–437. 1392–1395. Nguyen, T.V. & Ho, K.K. (1996) Short- Johansson, J.O., Larson, G., Andersson, M. Haro, L.S., Lewis, U.J., Garcia, M. et al. term growth hormone (GH) treatment et al. (1995) Treatment of growth (1996) Glycosylated human growth of GH-deficient adults increases body hormone-deficient adults with hormone (hGH): a novel 24 kDa hGH-N sodium and extracellular water, but recombinant human growth hormone variant. Biochemical and Biophysical not blood pressure. Journal of Clinical increases the concentration of growth Research Communications 228, 549–556. Endocrinology and Metabolism 81, hormone in the cerebrospinal fluid Harte, J.L., Eifert, G.H. & Smith, R. (1995) 1123–1128. and affects neurotransmitters. The effects of running and meditation Horber, F.F. & Haymond, M.W. (1990) Neuroendocrinology 61, 57–66. on β-endorphin, corticotropin-releasing Human growth hormone prevents Johansson, J.O., Landin, K., Johannsson, hormone and cortisol in plasma, and on the protein catabolic side effects of G., Tengborn, L. & Bengtsson, B.-Å mood. Biological Psychology 40, 251–265. prednisone in humans. Journal of (1996) Long-term treatment with growth Hasan, W., Cowen, T., Barnett, P.S. Clinical Investigation 86, 265–272. hormone decreases plasminogen et al. (2001) The sweating apparatus Howe, C., Hardman, K., Sridevi, M. et al. activator inhibitor-1 and tissue in growth hormone deficiency, (2003) Ethnicity and the IGF system in plasminogen activator in growth following treatment with r-hGH and in elite athletes: implications for detecting hormone-deficient adults. Thrombosis acromegaly. Autonomic Neuroscience 89, growth hormone doping in sports. and Haemostasis 76, 422–428. 100–109. Proceedings of the Endocrine Society of Jones, N.L., Sutton, J.R., Taylor, R. & Hashimoto, Y., Kamioka, T., Hosaka, M. Australia 46, 92 (Abstract). Toews, C.J. (1977) Effect of pH on et al. (2000) Exogenous 20 K growth Hussain, M.A., Schmitz, O., Mengel, A. cardiorespiratory and metabolic hormone (GH) suppresses endogenous et al. (1994) Comparison of the effects responses to exercise. Journal of Applied 22 K GH secretion in normal men. of growth hormone and insulin-like Physiology 43, 959–964. Journal of Clinical Endocrinology and growth factor 1 on substrate oxidation Jørgensen, J.O.L. (1991) Human growth Metabolism 85, 601–606. and on insulin sensitivity in growth hormone replacement therapy: Henneman, P.H., Forbes, A.P., Moldawer, hormone-deficient humans. Journal of pharmacological and clinical aspects. M., Dempsey, E.F. & Carroll, E.L. (1960) Clinical Investigation 94, 1126–1133. Endocrine Reviews 12, 189–207. Effects of human growth hormone in Ikkos, D., Luft, R. & Sjögren, B. (1954) Jørgensen, J.O.L., Moller, N., Lauritzen, T. man. Journal of Clinical Investigation 39, Body water and sodium in patients et al. (1990a) Evening versus morning 1223–1238. with acromegaly. Journal of Clinical injections of growth hormone (GH) in Hew, F.L., Koschmann, M., Christopher, Investigation 33, 989–994. GH-deficient patients: effects on 24-hour M. et al. (1996) Insulin resistance in Jahreis, G., Kauf, E., Fröhner, G. & patterns of circulating hormones growth hormone-deficient adults: Schmidt, H.E. (1991) Influence of and metabolites. Journal of Clinical defects in glucose utilization and intensive exercise on insulin-like Endocrinology and Metabolism 70, glycogen synthase activity. Journal of growth factor l, thyroid and steroid 207–214. Clinical Endocrinology and Metabolism 81, hormones in female gymnasts. Growth Jørgensen, J.O.L., Pedersen, S.A., 555–564. Regulation 1, 95–99. Thuesen, L. et al. (1990b) Long-term GH Hirschberg, R. & Kopple, J.D. (1988) Janssen, Y.J.H., Doornbos, J. & Roelfsema, replacement therapy in adult GH Increase in renal plasma flow and F. (1999) Changes in muscle volume, deficient patients 1 year. Hormone glomerular filtration rate during growth strength, and bioenergetics during Research 33 (suppl. 3), 5 [Abstract]. hormone treatment may be mediated by recombinant human growth hormone Jørgensen, J.O.L., Møller, J., Alberti, K.G. insulin-like growth factor I. American (GH) therapy in adults with GH et al. (1993) Marked effects of sustained Journal of Nephrology 8, 249–253. deficiency. Journal of Clinical low growth hormone (GH) levels on Hirschberg, R., Rabb, H., Bergamo, R. & Endocrinology and Metabolism 84,279–284. day-to-day fuel metabolism: studies Kopple, J.D. (1989) The delayed effect of Jeay, S., Sonenshein, G.E., Postel-Vinay, in GH-deficient patients and healthy growth hormone on renal function in M.C., Kelly, P.A. & Baixeras, E. (2002) untreated subjects. Journal of Clinical humans. Kidney International 35, 865–870. Growth hormone can act as a cytokine Endocrinology and Metabolism 77(6), Ho, K.K. & Weissberger, A.J. (1992) Impact controlling survival and proliferation 1589–1596. of short-term estrogen administration on of immune cells: new insights into Jørgensen, P.H. & Andreassen, T.T. (1988) growth hormone secretion and action: signaling pathways. Molecular and The influence of biosynthetic human distinct route-dependent effects on Cellular Endocrinology 188, 1–7. growth hormone on biomechanical connective and bone tissue metabolism. Johannsson, G., Grimby, G., Sunnerhagen, properties and collagen formation in Journal of Bone and Mineral Research 7, K.S. & Bengtsson, B.-Å (1997) Two years granulation tissue. Hormone and 821–827. of growth hormone (GH) treatment Metabolic Research 20, 490–493. 574 chapter 36

Juul, A., Hjortskov, N., Jepsen, L.T. et al. branchpoint selection. Nucleic Acids deficient adults and the acute and (1995) Growth hormone deficiency Research 15, 6331–6349. prolonged effects of GH administration: and hyperthermia during exercise: a Leung, D.W., Spencer, S.A., Cachianes, G. a controlled experimental study. Clinical controlled study of 16 GH-deficient et al. (1987) Growth hormone receptor Endocrinology 44, 533–539. patients. Journal of Clinical Endocrinology and serum binding protein: purification, Moller, N., Rahkila, P.J., Alberti, K.G.M.M., and Metabolism 80, 3335–3340. cloning and expression. Nature 330, Flyvbjerg, A. & Schmitz, O. (1990) Juul, A., Vahl, N., Jørgensen, J.O. et al. 537–543. Short-term effects of growth hormone (1998) Consequences of stopping Lewis, U.J., Sinha, Y.N. & Haro, L.S. (1994) on fuel oxidation and regional substrate growth hormone (GH) therapy in Variant forms and fragments of human metabolism in normal man. Journal of young GH deficient patients with growth hormone in serum. Acta Clinical Endocrinology and Metabolism 70, childhood onset disease. Growth Paediatrica Supplementum 399, 29–31. 1179–1186. Hormone and IGF Research 8 (suppl. A), Libber, S.M., Plotnick, L.P., Johanson A.J. Morgan, W.P., Costill, D.L., Flynn, M.G., 15–19. et al. (1990) Long-term follow-up of Raglin, J.S. & O’Connor, P.J. (1988) Karila, T., Koistinen, H. & Seppala, T. hypopituitary patients treated with Mood disturbance following increased (1999) Growth hormone induced human growth hormone. Medicine 69, training in swimmers. Medicine and increase in serum IGFBP-3 level is 46–55. Science in Sports and Exercise 20, 408–414. reversed by anabolic steroids in McGauley, G.A., Cuneo, R.C., Salomon, F. Mottram, D.R. (1999) Banned drugs in substance abusing power athletes. & Sönksen, P.H. (1990) Psychological sportadoes the international olympic Clinical Endocrinology 49, 459–463. well-being before and after growth committee (IOC) list need updating? Karlberg, B.E. & Ottosson, A.-M. (1982) hormone treatment in adults with Sports Medicine (Auckland, NZ) 27, 1–10. Acromegaly and hypertension: role of growth hormone deficiency. Hormone Nabarro, J.D.N. (1987) Acromegaly. the renin–angiotensin–aldosterone Research 33 (suppl. 4), 52–54. Clinical Endocrinology 26, 481–512. system. Acta Endocrinologica 100, Main, K.M., Price, D.A., Savage, M.O. Napoli, R., Guardasole, V., Angelini V. 581–587. & Skakkebae, N.E. (1993) Decreased et al. (2003) Acute effects of growth Kiowski, W., Stoschitzky, K., Kim, J.H. sweating in seven patients with hormone on vascular function in human & Hussain, M.A. (1994) Insulin-like Laron syndrome. Journal of Clinical subjects. Journal of Clinical Endocrinology growth factor-I causes endothelium- Endocrinology and Metabolism 77, and Metabolism 88, 2817–2820. dependent arterial vasodilation. 821–823. Nass, R., Huber, R.M., Klauss, V. et al. Circulation 90(4) (Part 2), 242 [Abstract]. Manson, J.M. & Wilmore, D.W. (1986) (1995) Effect of growth hormone (hGH) Kopple, J.D. & Hirschberg, R. (1990) Positive nitrogen balance with human replacement therapy on physical work Physiological effects of growth hormone growth hormone and hypocaloric capacity and cardiac and pulmonary and insulin-like growth factor I on intravenous feeding. Surgery 100, function in patients with hGH deficiency the kidney. Mineral and Electrolyte 188–197. acquired in adulthood. Journal of Clinical Metabolism 16, 82–88. Markussis, V., Beshyah, S.A., Fisher C. Endocrinology and Metabolism 80, 552–557. LaFranchi, S., Hanna, C.E. & Illingworth, et al. (1992) Detection of premature Nielsen, H.K., Brixen, K. & Mosekilde, L. D.R. (1985) Effect of growth hormone atherosclerosis by high-resolution (1990) Diurnal rhythm and 24-hour replacement on development of ultrasonography in symptom-free integrated concentrations of serum hypothyroidism and hyperlipidemia. hypopituitary adults. Lancet 340, osteocalcin in normals: influence of age, Journal of Pediatrics 106, 588–593. 1188–1192. sex, season, and smoking habits. Calcified Lange, M., Thulesen, J., Feldt-Rasmussen Merimee, T.J., Hollander, W. & Fineberg, Tissue International 47, 284–290. U. et al. (2001) Skin morphological S.E. (1972) Studies of hyperlipidemia in Nussey, S.S., Hyer, S.L., Brada, M. & changes in growth hormone deficiency the HGH-deficient state. Metabolism 21, Leiper, A.D. (1994) Bone mineralization and acromegaly. European Journal of 1053–1061. after treatment of growth hormone Endocrinology 145, 147–153. Merola, B., Longobardi, S., Sofia, M. et al. deficiency in survivors of childhood Lange, K.H., Larsson, B., Flyvbjerg A. (1996) Lung volumes and respiratory malignancy. Acta Paediatrica et al. (2002) Acute growth hormone muscle strength in adult patients with Supplementum 399, 9–14. administration causes exaggerated childhood- or adult-onset growth Nyberg, F. & Burman, P. (1996) Growth increases in plasma lactate and glycerol hormone deficiency: effect of 12 months’ hormone and its receptors in the during moderate to high intensity growth hormone replacement therapy. central nervous system––location and bicycling in trained young men. Journal European Journal of Endocrinology 135, functional significance. Hormone of Clinical Endocrinology and Metabolism 553–558. Research 45, 1–2. 87, 4966–4975. Møller, J., Rahkila, P.J., Møller, N. et al. Oakes, S.R., Haynes, K.M., Waters, M.J., Leahy, J.L. & Vandekerkhove, K.M. (1991) Expansion of extracellular Herington, A.C. & Werther, G.A. (1992) (1990) Insulin-like growth factor-I volume and suppression of atrial Demonstration and localization of at physiological concentrations is a natriuretic peptide after growth growth hormone receptor in human skin potent inhibitor of insulin secretion. hormone administration in normal man. and skin fibroblasts. Journal of Clinical Endocrinology 126, 1593–1598. Journal of Clinical Endocrinology and Endocrinology and Metabolism 75, Lecomte, C.M., Renard, J.A. & Martial, J.A. Metabolism 72, 768–772. 1368–1373. (1987) A new natural hGH variant–– Møller, J., Frandsen, E., Fisker, S., Ohlsson, C., Bengtsson, B.-Å., Isaksson, 17.5. kd––produced by alternative Jørgensen, J.O.L. & Christiansen, J.S. O.G.P., Andreassen, T.T. & Slootweg, splicing. An additional consensus (1996) Decreased plasma and M.C. (1998) Growth hormone and bone. sequence which might play a role in extracellular volume in growth hormone Endocrine Reviews 19, 55–79. growth hormone and sport 575

O’Neal, D.N., Kalfas, A., Dunning, P.L. Rosén, T., Eden, S., Larson, G., long-term effects. Endocrinology and et al. (1994) The effect of 3 months of Wilhelmsen, L. & Bengtsson, B.-Å. Metabolism Clinics of North America 4, recombinant human growth hormone (1993) Cardiovascular risk factors in 121–128. (GH) therapy on insulin and glucose- adult patients with growth hormone Sara, V.R., Uvnas-Moberg, K., Uvnas, B. mediated glucose disposal and insulin deficiency. Acta Endocrinologica 129, et al. (1982) The distribution of secretion in GH-deficient adults: a 195–200. somatomedins in the nervous system of minimal model analysis. Journal of Rosén, T., Johannsson, G., Hallgren, P. the cat and their release following Clinical Endocrinology and Metabolism 79, et al. (1994) Beneficial effects of 12 neural stimulation. Acta Physiologica 975–983. months replacement therapy with Scandinavica 115, 467–470. O’Neal, D.N., Hew, F.L., Sikaris, K. et al. recombinant human growth hormone Sartin, J.L., Elsasser, T.H., Gunter, D.R. (1996) Low density lipoprotein particle to growth hormone deficient adults. & McMahon, C.D. (1998) Endocrine size in hypopituitary adults receiving Endocrinology and Metabolism Clinics of modulation of physiological responses conventional hormone replacement North America 1, 55–66. to catabolic disease. Domestic Animal therapy. Journal of Clinical Endocrinology Rudling, M., Norstedt, G., Olivecrona, H. Endocrinology 15, 423–429. and Metabolism 81, 2448–2454. et al. (1992) Importance of growth Savino, W., Postel-Vinay, M.C., Smaniotto, O’Sullivan, A.J., Kelly, J.J., Hoffman, D.M., hormone for the induction of hepatic S. & Dardenne, M. (2002) The thymus Freund, J. & Ho, K.K. (1994) Body low density lipoprotein receptors. gland: a target organ for growth composition and energy expenditure Proceedings of National Academy of hormone. Scandinavian Journal of in acromegaly. Journal of Clinical Sciences of the United States of America 89, Immunology 55, 442–452. Endocrinology and Metabolism 78, 6983–6987. Schini-Kerth, V.B. (1999) Dual effects of 381–386. Russell-Jones, D.L., Weissberger, A.J., insulin-like growth factor-I on the O’Sullivan, A.J., Kelly, J.J., Hoffman, D.M., Bowes, S.B. et al. (1993) The effects of constitutive and inducible nitric oxide Baxter, R.C. & Ho, K.K.Y. (1995) Energy growth hormone on protein metabolism (NO) synthase-dependent formation metabolism and substrate oxidation in adult growth hormone deficient of NO in vascular cells. Journal of in acromegaly. Journal of Clinical patients. Clinical Endocrinology 38, Endocrinological Investigation 22, 82–88. Endocrinology and Metabolism 80, 427–431. Schwarz, A.J., Brasel, J.A., Hintz, R.L., 486–491. Rutherford, O.M., Jones, D.A., Round, J.M. Mohan, S. & Cooper, D.M. (1996) Acute Ottosson, M., Lönnroth, P., Björntorp, P. & Preece, M.A. (1989) Changes in effect of brief low- and high-intensity & Edén, S. (2000) Effects of cortisol and skeletal muscle after discontinuation exercise on circulating insulin-like growth hormone on lipolysis in human of growth hormone treatment in young growth factor (IGF) I, II, and IGF- adipose tissue. Journal of Clinical adults with hypopituitarism. Acta binding protein-3 and its proteolysis in Endocrinology and Metabolism 85, Paediatrica Scandinavica. Supplement 356, young healthy men. Journal of Clinical 799–803. 61–63. Endocrinology and Metabolism 81, Poortmans, J.R. (1977) Exercise and renal Rutherford, O.M., Jones, D.A., Round, 3492–3497. function. Exercise and Sport Sciences J.M., Buchanan, C.R. & Preece, M.A. Shore, R.M., Chesney, R.W., Mazess, R.B., Reviews 5, 255–294. (1991) Changes in skeletal muscle and Rose, P.G. & Bargman, G.J. (1980) Bone Press, M. (1988) Growth hormone and body composition after discontinuation mineral status in growth hormone metabolism. Diabetes/Metabolism Reviews of growth hormone treatment in growth deficiency. Journal of Pediatrics 96, 4, 391–414. hormone deficient young adults. Clinical 393–396. Rahkila, P.J., Pedersen, S.A., Thuesen, L. Endocrinology 34, 469–475. Sicuro, A., Mahlbacher, K., Hulter, H.N. et al. (1989) Beneficial effects of growth Saccà, L., Cittadini, A. & Fazio, S. (1994) & Krapf, R. (1998) Effect of growth hormone treatment in GH-deficient Growth hormone and the heart. hormone on renal and systemic acid- adults. Lancet 1, 1221–1225. Endocrine Reviews 15, 555–573. base homeostasis in humans. American Rahkila, P.J., Thuesen, L. Müller, J. et al. Salomon, F., Cuneo, R.C., Hesp, R. & Journal of Physiology 274, F650–F657. (1994) Three years of growth hormone Sönksen, P.H. (1989) The effects of Sinha, Y.N. & Jacobsen, B.P. (1994) Human treatment in growth hormone-deficient treatment with recombinant human growth hormone (hGH)-(44-191), a adults: near normalization of body growth hormone on body composition reportedly diabetogenic fragment composition and physical performance. and metabolism in adults with growth of hGH, circulates in human blood: European Journal of Endocrinology 130, hormone deficiency. New England measurement by radioimmunoassay. 224–228. Journal of Medicine 321, 1797–1803. Journal of Clinical Endocrinology and Root, A.W., Kemp, S.F., Rundle, A.C., Salomon, F., Cuneo, R.C., Umpleby, A.M. Metabolism 78, 1411–1418. Dana, K. & Attie, K.M. (1998) Effect of & Sönksen, P.H. (1994) Interactions of Sneppen, S.B., Main, K.M., Juul, A. et al. long-term recombinant growth hormone body fat and muscle mass with substrate (2000) Sweat secretion rates in therapy in children––the National concentrations and fasting insulin growth hormone disorders. Clinical Cooperative Growth Study, USA, levels in adults with growth hormone Endocrinology 53, 601–608. 1985–1994. Journal of Pediatric deficiency. Clinical Science 87, 201–206. Sneppen, S.B., Mersebach, H., Ullum, H. Endocrinology and Metabolism 11, Salomon, F., Umpleby, A.M., Cuneo, R.C. & Feldt-Rasmussen, U. (2002) Immune 403–412. & Sönksen, P.H. (1997) Protein, fat, and function during GH treatment in Rosén, T. & Bengtsson, B.-Å. (1990) glucose metabolism during treatment GH-deficient adults: an 18-month Premature mortality due to with recombinant human growth randomized, placebo-controlled, cardiovascular disease in hormone in adults with growth double-blinded trial. Clinical hypopituitarism. Lancet 336, 285–288. hormone deficiency. Short- and Endocrinology 57, 787–792. 576 chapter 36

Snyder, D.K., Clemmons, D.R. & echocardiographic study. Danish Medical Wallace, J.D., Cuneo, R.C. & Abbott, W. Underwood, L.E. (1988) Treatment of Bulletin 35, 193–196. (1996) Growth hormone treatment in obese, diet-restricted subjects with Tigranian, R.A., Kalita, N.F., Davydova, adults with liver cirrhosis: effects on growth hormone for 11 weeks: effects N.A. et al. (1992) Observations on the body composition and metabolism. on anabolism, lipolysis, and body Soviet/Canadian transpolar ski trek: 10th International Congress of composition. Journal of Clinical status of selected hormones and Endocrinology. San Francisco, USA. Endocrinology and Metabolism 67, 54–61. biologically active compounds. Medicine Program and Abstracts, vol. 1, 116 Stein, J.D., Jr., Bennett, L.L., Batts, A.A. & and Science in Sports 33, 106–138. [Abstract]. Li, C.H. (1952) Sodium, potassium and Umpleby, A.M. & Russell Jones, D.L. Wallymahmed, M.E., Foy, P. & chloride retention produced by growth (1996) The hormonal control of protein MacFarlane, I.A. (1999) The quality of hormone in the absence of the adrenals. metabolism. Bailliere’s Clinical life of adults with growth hormone American Journal of Physiology 171, Endocrinology and Metabolism 10, 551–570. deficiency: comparison with diabetic 587–591. Umpleby, A.M. & Sönksen, P.H. (1985) patients and control subjects. Clinical Stewart, P.M., Kane, K.F., Stewart, S.E., The chalonic action of insulin in man. Endocrinology 51, 333–338. Lancranjan, I. & Sheppard, M.C. (1995) In: Substrate and Energy Metabolism Warner, M.D., Sinha, Y.N. & Peabody, Depot long-acting somatostatin analog (Garrow, J.S., Halliday, D. & Libbey, J., C.A. (1993) Growth hormone and (Sandostatin–LAR) is an effective eds.). Libbey, London: 169–178. prolactin variants in normal subjects. treatment for acromegaly. Journal of Umpleby, A.M., Boroujerdi, M.A., Brown, Relative proportions in morning Clinical Endocrinology and Metabolism 80, P.M., Carson, E.R. & Sonksen, P.H. and afternoon samples. Hormone and 3267–3272. (1986) The effect of metabolic control on Metabolic Research 25, 425–429. Svensson, J., Stibrant-Sunnerhagen, K. leucine metabolism in type 1 (insulin- Weissberger, A.J., Ho, K.K.Y. & Lazarus, L. & Johannsson, G. (2003) Five years of dependent) diabetic patients. (1991) Contrasting effects of oral and growth hormone replacement therapy in Diabetologia 29, 131–141. transdermal routes of estrogen adults: age- and gender-related changes Vahl, N., Moller, N., Lauritzen, T., replacement therapy on 24-hour growth in isometric and isokinetic muscle Christiansen, J.S. & Jørgensen, J.O.L. hormone (GH) secretion, insulin-like strength. Journal of Clinical Endocrinology (1997) Metabolic effects and growth factor l, and GH-binding protein and Metabolism 88, 2061–2069. pharmacokinetics of a growth hormone in postmenopausal women. Journal of Takagi, K., Suzuki, F., Barrow, R.E. et al. pulse in healthy adults: relation to age, Clinical Endocrinology and Metabolism 72, (1997) Growth hormone improves sex, and body composition. Journal of 374–381. immune function and survival in burned Clinical Endocrinology and Metabolism 82, Welbourne, T.C. & Cronin, M.J. (1991) mice infected with herpes simplex virus 3612–3618. Growth hormone accelerates tubular type 1. Journal of Surgical Research 69, Vandeweghe, M., Taelman, P. & Kaufman, acid secretion. American Journal of 166–170. J.-M. (1993) Short and long-term effects Physiology 260, R1036–R1042. Takala, J., Ruokonen, E., Webster, N.R. of growth hormone treatment on bone Whitehead, H.M., Gilliland, J.S., Allen, I.V. et al. (1999) Increased mortality turnover and mineral content in adults & Hadden, D.R. (1989) Growth hormone associated with growth hormone with growth hormone deficiency. treatment in adults with growth treatment in critically ill adults. New Clinical Endocrinology 39, 409–415. hormone deficiency: effect on muscle England Journal of Medicine 341, Virtanen, P., Viitasalo, J.T., Vuori, J., fibre size and proportions. Acta 785–792. Väänänen, K. & Takala, T.E.S. (1993) Paediatrica Scandinavica. Supplement 356, Takala, T.E., Vuori, J., Anttinen, H., Effect of concentric exercise on serum 65–67. Väänänen, K. & Myllyla, R. (1986) muscle and collagen markers. Journal of Whitehead, H.M., Boreham, C., McIlrath, Prolonged exercise causes an increase in Applied Physiology 75, 1272–1277. E.M. et al. (1992) Growth hormone the activity of galactosylhydroxylysyl Wada, M., Ikeda, M., Takahashi, Y. et al. treatment of adults with growth glucosyltransferase and in the (1997) The full agonistic effect of hormone deficiency: results of a 13- concentration of type III procollagen recombinant 20 kDa human growth month placebo controlled cross-over aminopropeptide in human serum. hormone (hGH) on CHO cells stably study. Clinical Endocrinology 36, 45–52. Pflugers Archiv 407, 500–503. transfected with hGH receptor cDNA.. Wilmore, D.W. (1991) Catabolic illness. Takala, T.E., Vuori, J.J., Rahkila, P.J. et al. Molecular and Cellular Endocrinology 133, Strategies for enhancing recovery. (1989) Carbonic anhydrase III and 99–107. New England Journal of Medicine 325, collagen markers in serum following Wada, M., Uchida, H., Ikeda, M. et al. 695–702. cross-country skiing. Medicine and (1998) The 20-kilodalton (kDa) human Winter, R.J., Thompson, R.G. & Green, Science in Sports and Exercise 21, 593–597. growth hormone (hGH) differs from the O.C. (1979) Serum cholesterol and Thomas, S.G., Woodhouse, L.J., Pagura, 22-kDa hGH in the complex formation triglycerides in children with growth S.M. & Ezzat, S. (2002) Ventilation with cell surface hGH receptor and hormone deficiency. Metabolism 28, threshold as a measure of impaired hGH-binding protein circulating in 1244–1249. physical performance in adults with human plasma. Molecular Endocrinology Woodhouse, L.J., Asa, S.L. & Thomas, S.G. growth hormone excess. Clinical (Baltimore, MD) 12, 146–156. (1999) Measures of submaximal aerobic Endocrinology 56, 351–358. Wallace, J.D. & Cuneo, R.C. (1995) Growth performance evaluate and predcit Thuesen, L., Christiansen, J.S., Sørensen, hormone treatment in adults with liver functional response to growth hormone K.E. et al. (1988) Increased myocardial cirrhosis. Proceedings of the Australian (GH) treatment in GH-deficient adults. contractility following growth hormone Society for Parenteral and Enteral Nutrition Journal of Clinical Endocrinology and administration in normal man. An Incorporated (AuSPEN) 40 [Abstract]. Metabolism 84, 4570–4577. growth hormone and sport 577

Wu, Z., Bidlingmaier, M., Dall, R. & dwarf rats. American Journal of Physiology growth in young men. American Strasburger, C.J. (1999) Detection of 265, E332–E339. Journal of Physiology. Endocrinology doping with human growth hormone. Yang, C.-W., Striker, L.J., Kopchick, J.J. and Metabolism 262, E261–E267. Lancet 353, 895. et al. (1993) Glomerulosclerosis in mice Yarasheski, K.E., Zachwieja, J.J., Wüster, C., Slenczka, E. & Ziegler, R. (1991) transgenic for native or mutated Angelopoulos, T.J. & Bier, D.M. (1993) Increased prevalence of osteoporosis bovine growth hormone gene. Kidney Short-term growth hormone treatment and arteriosclerosis in patients with International 43, S90–S94. does not increase muscle protein conventionally substituted pituitary Yang, Y., Guo, L., Ma, L. & Liu, X. (1999) synthesis in experienced weight lifters. insufficiency: is there a need for Expression of growth hormone and Journal of Applied Physiology 74, 3073–3076. additional growth hormone insulin-like growth factor in the immune Yarasheski, K.E., Zachwieja, J.J., Campbell, substitution? Klinische Wochenschrift 69, system of children. Hormone and J.A. & Bier, D.M. (1995) Effect of growth 769–773. Metabolic Research 31, 380–384. hormone and resistance exercise on Wyse, B., Waters, M.J. & Sernia, C. (1993) Yarasheski, K.E., Campbell, J.A., Smith, K. muscle growth and strength in older Stimulation of the renin–angiotensin et al. (1992) Effect of growth hormone men. American Journal of Physiology 31, system by growth hormone in Lewis and resistance exercise on muscle E268–E276. Chapter 37

Endocrinology of Overtraining

ANDREW C. FRY, JUERGEN M. STEINACKER AND ROMAIN MEEUSEN

Introduction Biological stress. The method in which biological systems deal with stress was eloquently described The pursuit of elite athletic performance often by the landmark work of Hans Selye (1956) as he requires athletes to ‘push themselves to the limit’. presented the general adaptation syndrome (GAS; Any coach or athlete understands that to attain Fig. 37.1). Although originally applied to typical the greatest performance possible, athletes must biological scenarios, the GAS nicely describes the routinely push themselves physically and mentally process that an athlete uses to respond and adapt to into new levels of performance. Such a tactic is often training stresses, both the short-term immediate called overload, and although it is necessary for stresses from a single training session, as well as the optimal athletic performance, it can also be quite long-term stresses from the chronic training pro- problematic. In particular, what happens when the gram. Upon initiating a vigorous training session, athlete goes ‘a little too far’ in their training and the athlete begins the alarm phase where the body either becomes stagnant or, of even greater concern, responds to the stresses applied. On completion begins to perform poorly. This training scenario is of the session, the athlete ideally enters the resist- often called overtraining, and is of great interest to ance stage where he/she counters the biological not only the athlete and coach, but also has implica- responses to the training insult. Eventually, the tions for occupational and military settings. athlete enters the adaptation stage where they adapt

Supercompensation Adaptation New level of stage performance Original level of performance Fig. 37.1 The general adaptation syndrome (GAS) as applied to athletic training and performance. Inadequate recovery before While the exhaustion stage as next training session Alarm illustrated may be analogous to Performance Resistance stage overtraining, it is more likely that stage the line representing inadequate recovery may be more indicative of Exhaustion the subtle onset of overreaching and stage overtraining often encountered in the sport setting. (Modified from Selye Time 1956.) 578 endocrinology of overtraining 579

to the original stress and move to a higher level Symptoms of overtraining. Perhaps the overarching of biological functioning. In sporting terms, the ath- question when examining the topic of overtraining lete performs better. The problem of overtraining is how can it be detected before it becomes a chronic arises when the individual is unable to properly problem? Unfortunately, experts on the topic ac- respond to the alarm phase, eventually reaching the knowledge that no specific, simple, and reliable exhaustion stage. In Selye’s original model, the parameters are known to diagnose overreaching exhaustion stage results in the death of the system. and overtraining in the earliest stage (Kuipers 1998). In the sporting world, this results in performance Regardless, overtraining research has suggested decrements, and the ‘death’ of any hope of optimal many possible markers of an impending or existing performance. overtrained state. Over 15 years ago, an excellent review of the topic suggested over 80 possible major Overtraining defined. Perhaps the most critical step symptoms in addition to performance decrements when discussing overtraining, is providing an opera- (physiological = 40 symptoms, psychological and tional definition of this and related terms. For the information processing = 12, immunological = 14 purposes of this chapter, we will define overtraining and biochemical = 18) (Fry, R.W. et al. 1991). Sub- as ‘the accumulation of training and non-training sequent years of study have only added to the list. stress resulting in long-term (i.e. several weeks or Of particular interest to the present chapter are the months or longer) decrement of performance capa- numerous endocrine and neuroendocrine symptoms city’ (Fry, A.C. 1998; Kreider et al. 1998). A less dele- and responses to overtraining. terious version of overtraining is often refered to as overreaching, which is ‘the accumulation of train- Physiological mechanisms. Obviously, any condition ing and non-training stress resulting in short-term as complex as overtraining, is influenced by numer- (i.e. several days) decrement of performance capa- ous physiological systems. Previous reviews have city’ (Fry, A.C. 1998; Kreider et al. 1998). Although addressed the contributions of neuromuscular con- many different terms have been applied to this phe- trol (Lehmann et al. 1999b), tissue trauma (Smith nomenon, this chapter will use these definitions 2004), the immune response (Robson 2003), psycho- when discussing both the short- and long-term con- logical (Morgan et al. 1987; O’Connor 1997) and clin- ditions. Perhaps the most important aspect of these ical depression (Armstrong & Van Heest 2002), to definitions is the requirement for performance name a few. It has also been suggested that women decrements. Although stress may be manifested in may be more susceptible to overtraining (Shephard may ways, this chapter will focus on the training 2000), and research efforts are beginning to address program-induced stresses (i.e. increased training the issue of gender (O’Connor et al. 1989; Uusitalo volume and/or intensity). It should also be noted et al. 1998). that ‘overreaching’ and ‘overtraining’ refer to the The central fatigue hypothesis has also been pre- actual stimulus, and the term ‘overtraining syn- sented as a possible contributor to overtraining drome’ refers to the resulting condition. As the (Newsholme et al. 1992). This theory is based on the overtraining literature is reviewed, it will become premise that the exercise-induced increase in circu- readily apparent that much of the information must lating concentrations of the amino acid tryptophan be extrapolated from research on overreaching and ultimately result in elevated levels of serotonin stressful training phases. This is certainly due to (5-hydroxytryptamine, 5-HT) in the the brain. The the difficulty in observing this phenomenon in a neurotransmitter serotonin has been associated controlled environment. Additionally, the authors with fatigue-like conditions, and could conceiv- would like to point out that actual long-term per- ably contribute to an overtraining syndrome. It formance decrements due to overtraining do not has been suggested that dietary supplementation occur as often as many might think. What many with the branched chain amino acids (BCAA; athletes and coaches call overtraining is actually leucine, isoleucine, valine) would minimize central overreaching, and recovery is relatively easy. uptake of tryptophan, resulting in an avoidance of 580 chapter 37

performance impairment. It should be noted, how- have been suggested (Meeusen 1999). Some of the ever, that thus far much of the research results recent advances in assessment of these types of are not supportive concerning the link between overtraining syndromes have been based on the BCAA supplementation, 5-HT and performance hypothalamic–pituitary–adrenal (HPA) and hypo- decrements due to overtraining (Fry, A.C. et al. 1993; thalamic–pituitary–gonadal (HPG) axes, as well as Meeusen et al. 1996). sympathetic activity (Meeusen et al. 2004; Steinacker et al. 2004). Detailed examination of these topics will Endocrine-related mechanisms. Perhaps the most be provided later in this chapter. studied overtraining-related physiological system is the endocrine and neuroendocrine system. In Not all sports are alike. Much of the overtraining- 1986, Adlercreutz and colleagues reported the rela- related research has focused on endurance types tionship between overtraining and the anabolic of sports and activities (Kuipers & Keizer 1988; hormone testosterone, the catabolic hormone cor- Fry, R.W. et al. 1991; Lehmann et al. 1993a, 1998b; tisol and their ratio (testosterone/cortisol, T/C, and Keizer 1998; Urhausen et al. 1998). However, it was free testosterone/cortisol, FT/C) (Adlercreutz et al. pointed out by van Borselen et al. (1992) that over- 1986). From an endocrinologist’s perspective, their training with heavy resistance exercise was likely suggestion that these ratios were indicative of the to present very different responses and symptoms. anabolic–catabolic status of the body was probably Later reviews of subsequent resistance exercise an oversimplification. However, these hormones overtraining research have reinforced this belief and their ratios have repeatedly been shown to alter (Fry & Kraemer 1997; Fry, A.C. 1998 & 1999; Stone in response to physical stresses such as overreach- & Fry 1998; Fry, A.C. et al. 2001). More recently, it is ing and overtraining (Häkkinen et al. 1987; Fry, A.C. becoming apparent is that overtraining for many et al. 1993). Closer examination of the hypothala- sports involve considerable contributions from a mic–pituitary regulation of these hormones during combination of aerobic training, resistance exercise, overtraining has suggested the potential role of this and sport-specific conditioning (Chicharro et al. axis on regulating peripheral concentrations of the 1998; Gorostiaga et al 1999; Murlasits et al. 1999; stress-related hormones (Barron et al. 1985; Wittert Kraemer et al. 2004; Maso et al. 2004). As such, the et al. 1996; Urhausen et al. 1998). characteristics of overtraining, and the resulting Eventually, an important role of the autonomic overtraining syndrome, are extremely dependent nervous system (ANS) in an overtraining syndrome on the physiological characteristics of the specific was suggested (Stone et al. 1991; Lehmann et al. sport. Therefore, this chapter will address the topic 1998a). Two types of ANS-based overtraining have of overtraining based on the type of activities being been termed the sympathetic and the parasym- studied. pathetic overtraining syndromes (Stone et al. 1991; Lehmann et al. 1998a). It has been suggested that the Overtraining in endurance sports sympathetic syndrome occurs prior to the parasym- pathetic syndrome. The sympathetic type is asso- Endurance sports and activities defined. A sport which ciated with a predominance of sympathetic nervous involves continuous activity over a longer time system activity, while the parasympathetic type is requires a physical capability defined as endurance. associated with a predominance of parasympathetic Endurance can be defined as the maximum dura- nervous system activity. As the individual attempts tion for which an exercise intensity can be sustained; to respond to the stressful training, sympathetic the relation between intensity or velocity and max- activity is elevated in an attempt to cope. As the imum time can be described by a typical logarithmic stress progresses, the sympathetic system becomes power–duration curve. Endurance training shifts exhausted, resulting in parasympathetic predom- this curve up so that an exercise intensity can be inance. Eventually, central mechanisms (i.e. brain sustained with longer duration and a given time neurotransmitters) contributing to this condition frame with higher intensity. Endurance activities endocrinology of overtraining 581

Fig. 37.2 Response of the hypothalamic–pituitary–adrenal (HPA) axis during normal training, Training Overreaching Overtraining overreaching and overtraining. ACTH When the body is able to adequately response react to to the metabolic stresses Lehmann et al. 1993 Barron et al. 1985 of training, adrenocorticotropic Wittert et al. 1996 hormone (ACTH) and cortisol are capable of responding (compensation). During Cortisol Lehmann et al. 1993 Barron et al. 1985 overreaching, the cortisol response is response Wittert et al. 1996 Lehmann et al. 1993 attenuated even though the ACTH Snyder et al. 1995 signal is augmented. During Urhausen et al. 1998 overtraining, both the ACTH and cortisol responses are blunted or missing (decompensation). (Modified Compensation Decompensation from Lehmann et al. 1999b; Steinacker et al. 2004.) can be classified according to the maximum dura- permit adjustment of training loads at high, sustain- tion in short-type (up to 4 min), middle-type (up able levels for a short time, and are relatively safe to 15 min) and long-type endurance (longer than (Steinacker et al. 1998). The process is called a super- 15 min). Many activities have endurance com- compensation cycle (Fig. 37.2) and is also referred to ponents such as in ball games in which endurance as reaching. The exhaustion and fatigue in the high exercise with lower intensity is combined with the load training phases should elicit corresponding sports specific exercise pattern. cellular stresses, and should consequently raise per- Endurance exercises can be also defined by formance in the recovery phases as an adaptation to metabolic activity which involves mainly aerobic the training overload (Lehmann et al. 1997b). In such pathwaysaKrebs cycle and respiratory chaina training cycles, fatigue is a common reaction in high and by the compartmentathe mitochondria. The load training phases and can be labeled as short- predominant substrate metabolism is fat and fat type overtraining (Lehmann et al. 1993a). This pro- oxidation, but at higher exercise intensities, or cess can be described as a prolonged or accumulated with lower fitness, carbohydrates are also oxidized; acute stress response, and can be compensated by for example, glucose and glycogen stores can be adequate recovery (Lehmann et al. 1997b). depleted (Baldwin et al. 1975; Richter et al. 2001). The first and classical training experiments were Training often consists of repetitive phases of nor- designed to clearly distinguish the effects of volume mal training, high training load phases, overload and intensity, keeping in mind that there is a close training phases, overreaching and recovery. During relationship between exercise duration, intensity and a training program, training load (defined by intens- substrate availability. The training group studied ity, duration and frequency of exercise) varies and by Lehmann and colleagues involved experienced should gradually increase in response to the train- middle- and long-distance runners who performed ing-induced adaptation of various physical systems 2 consecutive years of controlled training. Each year and in performance. This increase in training load is involved 4 weeks of standardized normal training, necessary to ensure further responses to a training followed by: (i) an unaccustomed ~ 35% increase in program. training volume (ITV) for 4 weeks; or (ii) by a ~ 51% Coaches often organize training in phases in increase in training intensity (ITI) for 4 weeks which training loads reach the athletes’ capacities, (Lehmann et al. 1991). In a further study, Lehmann followed by phases in which regeneration processes et al. (1993b) examined the effect of 6 weeks of en- should predominate training. Such training cycles durance training versus interval training and 3 582 chapter 37

weeks of recovery in recreational athletes on serum ponded with a lower ACTH and/or cortisol res- hormone levels and pituitary function (endurance- ponse in the state of overtraining. intensity training, EIT). The HPA axis is biphasically regulated. First, the acute stress and high metabolic load during an over- Increases in endurance training volume. In ITV, clas- reaching phase will lead to activation of the CRH sical symptoms of overtraining occurred such as and ACTH response and to a moderate adrenal fatigue, neuromuscular dysfunction and perform- desensitivation, meaning less cortisol response in ance incompetence. At rest and during submaximal relation to ACTH levels. Secondly, chronic stress exercise, plasma levels of lactate, glucose, glycerol, and overtraining will lead to central and peripheral free fatty acids and amino acids decreased. The down-regulation (Lehmann et al. 1997b; Steinacker nocturnal excretion of catecholamines decreased et al. 2004). Therefore, basal cortisol levels may (Lehmann et al. 1991, 1992a). In the endurance arm be used for analysis of the effects of training of EIT (Lehmann et al. 1993b), adrenocorticotropic (Steinacker et al. 2000). At an early stage during the hormone (ACTH) was increased, prolactin, thyroid- 1996 training camp, when training load was highest, stimulating hormone (TSH) and somatotropic hor- basal cortisol levels increased by 18% and decreased mone (STH) were unchanged. After pituitary slightly afterwards (Steinacker et al. 2000). Increases stimulation, cortisol, follicle-stimulating hormone in basal cortisol levels are indicative of a metabolic (FSH) and luteinizing hormone (LH) release were problem (e.g. glycogen deficiency) or of increased decreased. Endurance training in EIT had no effect training-dependent stress (Lehmann et al. 1997b). A on peripheral hormone levels (cortisol, aldosterone, common metabolic cause can be seen in glycogen insulin, prolactin) except for a small decrease in free depletion with increased counterregulatory activity testosterone (Lehmann et al. 1993b). of hormones such as cortisol, growth hormone (GH) and catecholamines (Gray et al. 1993). Elevated Increases in endurance training intensity. In ITI, only a basal cortisol levels are often seen as a normal small, but significant, decrease in plasma catechola- stress response to high intensity training, whereas mines at a given exercise intensity was observed, decreased basal cortisol levels and decreased pitu- which can be attributed to altered release and itary–adrenocortical responsiveness are a late sign of neuronal activity. The nocturnal excretion of cate- overreaching or overtraining (Lehmann et al. 1993a, cholamines decreased significantly, but to a much 1993b, 1997b). As discussed above, basal cortisol lower extent than in ITV (Lehmann et al. 1991). levels reflect an endpoint of the hypothalamic– pituitary–adrenocortical axis. Testing the axis by Mechanisms of endurance overtraining related to the functional tests will give much more information, endocrine system. During acute endurance exer- but they are time- and cost-consuming and are cise bouts, when muscle and liver glycogen are stressful for the athletes (Kuipers & Keizer 1988; diminished, the normal hormonal reaction involves Lehmann et al. 1998b; Steinacker et al. 2004). hypercortisolism and hypoinsulinism to maintain fuel homoestasis. With prolonged training, meta- Somatotropic hormonal reactions. The somatotrophic bolic imbalance is prolonged and dysfunction of the growth hormone–insulin-like growth factor I (GH– HPA axis and hypoinsulinism may occur (Lehmann IGF-I) axis is also involved in this hormonal regula- et al. 1999a). A similarity with the catabolic effects tion. The pulsatile pattern of GH is stimulated and of calorie restriction is striking (Jenkins et al. 1993; levels of GH rise by acute prolonged exercise Laughlin & Yen 1996; Argente et al. 1997). The (Kanaley et al. 1997). IGF-I levels increase, consecut- hypothalamic down-regulation in overreaching ively, after 1–2 weeks of training with moderate and overtraining was described by Barron et al. intensity and positive caloric balance (Snyder, D.K. (1985) and Lehmann et al. (1993b), who observed et al. 1989; Engfred et al. 1994; Roelen et al. 1997; that insulin-dependent hypoglycemia or attenu- Roemmich & Sinning 1997) and decrease more pro- ated corticotropin-releasing hormone (CRH) corres- foundly when training is exhausting and catabolic endocrinology of overtraining 583

mechanisms are predominant (Koistinen et al. 1996). expression of slow type fibers (Atalay et al. 1996; During prolonged exhausting training, the pulsatility Pette & Staron 1997). Glycogen deficiency is asso- of GH and IGF-I levels decrease simultaneously ciated with increased expression of local cytokines (Eliakim et al. 1998; Schmidt et al. 1995). There is evid- (interleukin-6) (Pedersen et al. 2001), decreased ence that depression of the GH–IGF-I axis is a com- expression of glucose transporters (Richter et al. mon finding during overreaching. The effects are 2001), and increased cortisol and decreased insulin related to down-regulation of the HPA axis (Barron secretion and β-adrenergic stimulation which lead et al. 1985; Steinacker et al. 1998, 2004). Furthermore, also to the loss of fast fibers (Pette & Staron 1997). down-regulation of the thyroid and gonadotropic The overtraining myopathy results from these meta- axes will be observed (Lehmann et al. 1997b). bolic and hormonal changes, and may be diagnosed Leptin, an adipocyte-derived hormone, is involved in muscle biopsies by a shift to slow type myosin the hypothalamic regulation of appetite, thermogen- and a decreased number and fiber area of fast type esis and metabolism (Friedman et al. 1997; Heiman fibers (Atalay et al. 1996; Steinacker et al. 2004). et al. 1997). Leptin depresses the excitatory transmit- Heat shock proteins are increased in muscle, pos- ter neuropeptide Y (NPY) expression in hypothala- sibly as hormonal molecular chaperones. A single mic neurons (Carro et al. 1998). NPY neurons are bout of exercise leads to induction of heat shock activated by fasting and express a high degree of protein 70 (HSP70) transcription but not of effective leptin receptors (Schwartz et al. 1996; Baskin et al. protein production; protein production rate is high 1999). Increased NPY levels increase hypothalamic after a week of training and peaks after approxim- neuronal activity and therefore depress the HPA ately 2 weeks (Liu et al. 1999). High intensive train- axes. Leptin is an example of a tissue hormone that ing leads to an induction of HSP70 while endurance has direct impact on the hypothalamic regulation training with low intensity does not. It can be con- and provides information about the metabolic state cluded that HSP70 response to training is depend- to other tissues like β cells or liver cells. Leptin ent on exercise intensity rather than exercise volume is a potent stimulator of pulsatile GH secretion (Liu et al. 2000). Whether accumulation of heat and the GH response to GH-releasing hormone shock proteins is protective, or indicates a training (Tannenbaum et al. 1998). Leptin plasma levels are response, or is related to trainability or stress low and apparently decrease with training load stability has to be determined in further studies (Leal-Cerro et al. 1998; Simsch et al. 2002). Whether (Steinacker et al. 2004). leptin has the potential as a marker for overreaching or overtraining has to be clarified in prospective Intrinsic sympathetic activity and catecholamines during studies (Steinacker et al. 2004). endurance overtraining. Increased intrinsic sympath- etic activity may be found in the early phase of heavy Muscle adaptations and heat shock proteins. Energy training (according to the above mentioned classical disturbances and glycogen depletion in skeletal model) which is an adaptive regulatory process muscle are common findings in early phases of (Fry, R.W. et al. 1991; Lehmann et al. 1997b, 1998a). overtraining, but it is now clear that restoring mus- However, if hypothalamic depression is more pro- cular glycogen will not decrease symptoms of over- found, intrinsic sympathetic activity decreases as training (Lehmann et al. 1997b; Snyder, A.C. 1998). indicated by decreased nocturnal catecholamine Muscular glycogen depletion is postulated as one of excretion and peripheral β-receptor down-regulation the metabolic causes that will initiate an overtrain- with compensatory increased plasma noradrenaline ing syndrome in which peripheral thyroidal hor- but ineffective catecholamine response (Lehmann et mones and GH levels decrease as well as levels of al. 1992b, 1998a; Wittert et al. 1996). The excretion of IGF-I (Lehmann et al. 1997b; Steinacker et al. 1999). free catecholamines during overnight rest can be All hormonal effects together will cause a shift in seen as the basal renal (urinary) catecholamine expression of myosin heavy chain expression in excretion reflecting the intrinsic activity or tone the direction of slow isoforms, and will lead to the of the sympathetic nervous system, as activating 584 chapter 37

mechanisms are clearly reduced during night rest. resistance exercise for supplemental training or Because noradrenaline concentrations in plasma develop high muscular force and power during and in cerebrospinal fluid are quite similar, circulat- their events. Needless to say, these sports/activities ing and excreted noradrenaline may also reflect the are difficult to study due to their complex nature. neuronal noradrenaline release in the brain. In over- Regardless, several attempts have been made to reaching and overtraining, resting and submaximal examine the manifestation of overtraining in these plasma noradrenaline concentrations may be in- activites. It is likely that each of these studies creased and β-adrenoreceptor density is decreased actually observed states of overreaching rather that (Jost et al. 1990; Lehmann et al. 1998a; Uusitalo et al. overtraining. 1998). Down-regulation of the intrinsic sympathetic When supplemental heavy resistance exercise has nervous system activity is a late finding in over- been added to the sport-specific training program trained athletes and may be related to symptoms of for American football (Murlasits et al. 1999) or team fatigue (Lehmann et al. 1992b). handball (Maso et al. 2004), decreases in resting levels of total testosterone and/or the T/C ratio Neuromuscular excitability. There is also indication have been reported. When interpreting these data, of desensitization of the motor end-plate (for an it must also be noted whether other aspects of the overview see Lehmann et al. 1998b). These findings training program were concurrently altered. For are related to the subjective feeling of some over- example, Murlasits et al. (1999) observed that the trained athletes that their muscles may not respond addition of resistance exercise alone was not prob- as normally and of peripheral weakness. The lemmatic, but when the volume of related condi- hypothesis was tested that a deterioration in neuro- tioning drills was increased (i.e. sprints, agility, muscular excitability (NME) also indicates an early metabolic conditioning, etc.), attenuated resting stage in the overtraining process during high intens- testosterone levels resulted. These results should ity training (Lehmann et al. 1997a). In the EIT trial, not be interpreted to mean resistance exercise NME was slightly improved after 3 weeks of train- should be avoided, but rather the cumulative train- ing, but deteriorated after 6 weeks, and was again ing stresses of the entire program must be con- normalized after 2 weeks of regeneration (Lehmann sidered when determining the proper resistance et al. 1995). The discrepancy between normalization exercise volume and intensity. of NME and still-deteriorated performance ability In the sport of rugby where high muscular forces after 2 weeks of regeneration reflects additional are common, overtraining (as determined by ques- significant, and probably central, mechanisms that tionnaire) was accompanied by decreased concen- explain persistent performance incompetence in trations of resting total testosterone (Maso et al. overtraining. Deterioration in NME may indicate an 2004). With the limited data available about sports early stage in the overtraining process during high- of this type, it appears that resting testosterone volume as well as high-intensity endurance over- and the T/C ratio may be most sensitive to the training, but normalization does not necessarily training stresses. Indeed, during a collegiate soccer indicate sufficient regeneration (Lehmann et al. season, Kraemer et al. (2004) observed that those 1995, 1999b). players beginning the season with the lowest T/C ratio also experienced the greatest performance Overtraining in strength sports and decrements as the season progressed. In a non-sport activities setting, Chicharro et al. (1998) reported that the FT/C ratio identified Special Forces soldiers who Resistance exercise and related sports. Many activities appeared to respond poorly to the specialized train- require considerable contributions from both anaer- ing for these elite troops. As previously mentioned, obic and aerobic energy systems (e.g. soccer, team it is likely that these situations represent overreach- handball, rugby, American football, military pre- ing and not true overtraining. It is readily apparent paration). As such, these activities either use heavy that much study needs to be done on these types endocrinology of overtraining 585

of activities, and that the T/C and FT/C ratios are for not only resting levels, but also for the exercise- sensitive enough to respond to the changing train- induced responses. It appears that prior weightlift- ing stresses. ing training experience of ≥ 2 years can significantly influence the hormonal response to a single bout Increased resistance exercise volume. A more thorough of exercise (Kraemer et al. 1992). Furthermore, it understanding of resistance exercise overtraining was subsequently demonstrated that weightlifters requires an understanding of the variables that who had previous experience undergoing an over- contribute to the resistance exercise training stress reaching phase of training were less impaired when (Kraemer & Nindl 1998; Fry, A.C. 1999). For a single exposed to a second overreaching phase 1 year later training session, these include the choice of exercise, (Fry, A.C. et al. 1994b). This second overreaching order of exercise, exercise volume, exercise intens- phase exhibited an actual increase in both total ity and inter-set rest intervals. For the long-term testosterone and the T/C ratio at rest and in resistance exercise program, this includes the vari- response to a standardized training session. The ation inherent with a periodized training program. cortisol response did not change. Additionally, the In order to more effectively determine the roles of most highly ranked lifters (but not more experi- each of these variables, it becomes necessary to enced) exhibited no relationship (r = 0.00) between attempt to control some or all of these variables for relative (%) changes in performance and relative research purposes. Of course, it is not possible to changes in the T/C ratio during the overreaching completely eliminate the role of any of these train- phase (Fry, A.C. et al. 2000). On the other hand, the ing variables because of their interdependence. relative change in T/C during this phase for lower The ability of chronic (≥ 2 yrs) training in elite ranked lifters was negatively related (r = –0.70) to competitive Olympic-style weightlifters to increase relative changes in performance. Both groups of resting testosterone levels has been reported by lifters appeared to benefit from the subsequent Häkkinen et al. (1987). Although the magnitude of training taper phase, but the highly ranked lifters this change was small it, nevertheless, illustrates responded especially well from both performance the plasticity of the endocrine system in highly and endocrine perspectives. As the T/C ratio resistance trained athletes. In a classic study from increased during the taper, the increase in perform- the same laboratory, changes in the training volume ance also increased (r = 0.92) for the highest ranked load (mass × repetitions; kilogram) for the Finnish lifters, and less so for the lower ranked lifters (r national weightlifting squad were shown to be = 0.51). Overall, the changes in the training volume related to testosterone and cortisol levels and the as well as the actual weightlifting performances T/C ratio (Häkkinen et al. 1987). As the volume load were strongly reflected by the testosterone and the increased, resting concentrations of testosterone T/C ratio. and the T/C ratio decreased, while resting cortisol Although study of the sport of weightlifting per- increased. This pattern was reversed as the volume mits examination of a closely monitored train- load was subsequently decreased. Although not ing environment, many athletes incorporate more an overtraining study, these findings demonstrate generic types of resistance exercise into their train- that changing training volume is reflected in the ing. Several studies have examined the endocrine hormonal profile. responses to increases in resistance exercise training These findings were supported when high vol- volume, although this increase was accomplished ume training of junior-age weightlifters resulted in differently for these studies. Raastad et al. (2001) an overreached state (Fry, A.C. et al. 1993). As was increased the training volume load approximately seen with the Finnish lifters, testosterone and the fourfold (from < 20 000 kg·wk–1 to > 76 000 kg·wk–1) T/C ratio decreased, and cortisol increased when while training everyday for 2 weeks. Volek et al. training volume was increased from 3–5 sessions (2004) and Ratamess et al. (2003) increased resistance per week to 2–4 sessions per day for a 1-week training volume for 4 weeks by training all body period. These decreases, however, were apparent parts five times per week. For their protocol, they 586 chapter 37

progressed from approximately 250 repetitions per Insulin concentrations have also been reported to day during week 1 to 75 repetitions per day for decrease due to high volume resistance exercise week 4. Concurrently, relative training intensity overreaching (Ratamess et al. 2003; Volek et al. 2004). increased from 10–12-repetition maximum (RM) Although the insulin response appeared to be re- loads during week 1 to 3-RM loads for week 4. lated to decreased glucose concentrations (Ratamess Strength perfomance was adversely affected in each et al. 2003), the impact of this anabolic hormone on of these studies, although an overreaching state skeletal muscle and its function during overtraining appeared to result rather than an advanced over- remains to be determined. Only one resistance exer- training syndrome. cise study has reported augmented resting (nocturnal In all three studies, resting concentrations of urinary) levels of epinephrine and norepinephrine total testosterone decreased when volume load was in response to overreaching (Raastad et al. 2001). increased (Raastad et al. 2001; Ratamess et al. 2003; Such a response would be indicative of a sympath- Volek et al. 2004). Although a similar decrease was etic overtraining syndrome. not apparent for direct measures of free testo- sterone, the ratio of testosterone to sex hormone Increased resistance exercise intensity and power. In an binding protein (T/SHBG) did decrease, most likely attempt to determine the effects of resistance exer- due to a training-induced increase in the absolute cise relative intensity (% 1-RM) on the etiology of concentrations of SHBG (Ratamess et al. 2003; Volek an overtraining syndrome, a series of studies were et al. 2004). As a result, not only did total testo- performed to isolate this training variable. When sterone decrease, but the bioavailable portion of maximal loads were used daily (10 × 100% 1-RM) for testosterone also decreased. When several days of 2 weeks on a squat simulating machine, 1-RM per- recovery were monitored, testosterone returned to formance was depressed and required 2–8 weeks baseline levels (Raastad et al. 2001). The decreases for recovery (Fry, A.C. et al. 1998). Contrary to what in testosterone occurred despite no alterations in has been observed with high volume resistance LH. No changes or slight transient increases were exercise overtraining, total testosterone was un- observed for resting cortisol concentrations (Raastad changed at rest, and actually increased slightly in et al. 2001; Ratamess et al. 2003; Volek et al. 2004). response to maximal exercise. This occurred simul- As with LH, no changes in ACTH were reported taneously with a slight decrease in acute LH (Raastad et al. 2001), suggesting a decreased sens- (preliminary data), suggesting that a mechanism itivity of the Leydig cells in the testes and of the other than the gonadotrophins were responsible for adrenal cortex to circulating levels of LH and the testosterone response. Bioavailable testosterone ACTH, respectively. It is important to note that, to was not altered as indicated by FT and %FT. No the authors’ knowledge, no overtraining study has change was observed for resting levels of cortisol, measured pulsatility characteristics of these trophic but acute responses of cortisol actually decreased hormones, and thus these LH and ACTH data rep- slightly. As with LH, no change was observed for resent preliminary results. Only Raastad et al. (2001) single time point ACTH (preliminary data), again examined measures related to the FT/C ratio, but suggesting that the trophic hormone for cortisol was reported that this was not altered. not responsible for the acute response. In actuality, In addition to the steroid hormones, increased the modest to non-existent hormonal responses to resistance training volume has sometimes resulted this high intensity protocol is not overly surprising, in slight increases in resting levels of immunoreact- since it has been previously shown that there is little ive GH (Ratamess et al. 2003), although circulating or no acute hormonal response to a single bout levels of IGF-I have reportedly been unaffected of high intensity, low volume resistance exercise (Raastad et al. 2001; Volek et al. 2004). Since IGF-I (20 × 100% 1-RM) (Häkkinen & Pakarinen 1993). functions not only in an endocrine fashion, but also Of particular interest for the high intensity resist- in paracrine and autocrine fashions, it is likely that ance exercise overtraining protocol was the increased these circulating values do not tell the entire story. acute response of circulating concentrations of both endocrinology of overtraining 587

epinephrine and norepinephrine (Fry, A.C. et al. as reported by Lehmann et al. (1998b) for endurance 1994a). When resting levels of catecholamines were athletes. monitored via nocturnal urine, only slight increases Due to the apparent sensitivity of high velocity were observed (effect size, ES = 0.42), suggesting and power performances, recent advances in this that the acute catecholamine responses are more line of study have examined resistance exercise sensitive to this high intensity overtraining protocol overtraining using excessive high power resistance (Fry, A.C. et al. 2004b). As with Raastad et al.’s (2001) exercise (Fry, A.C. et al. 2004a). An overtrained state report for high volume resistance exercise, these was induced by performing 10 × 5 speed squats at results are symptomatic of the sympathetic over- 70% 1-RM for 15 sessions over 7.5 days. As expected, training syndrome. Attempts to non-invasively no changes in resting levels of testosterone, cortisol monitor sympathetic activity in this overtraining or the T/C ratio were observed. Furthermore, model via heart rate variability has not been suc- nocturnal urine assessments exhibited no changes β cessful to date (Weir et al. 2002). It was interesting to in the catecholamines. Although 2-receptor density β note, however, that although the relative changes was not altered, the EPI/ 2 increased dramatically. (%) for the circulating catecholamines were strongly These results are similar to what was observed for correlated to various measures of muscular force the high intensity overtraining, and again indicate a (r = 0.79–0.96) for the control subjects, none of decreased catecholamine sensitivity (Lehmann et al. these relationships were significant for the over- 1998b). trained subjects. These results suggest that the high intensity resistance exercise overtraining protocol Endocrine mechanisms of resistance exercise overtrain- decreased sympathetic regulation of peripheral ing. When all the data available on resistance exer- muscle activity. It should be noted that several of cise overtraining are summarily examined, several the pre-exercise hormonal responses (i.e. T, FT, T/C, physiological patterns emerge. First, during high FT/C) to this protocol were also correlated to mus- intensity or high power overtraining, acute hor- cular force measures (r = 0.83–0.99) for only the monal responses appear to change before resting control subjects (Fry, A.C. et al. 1998). In the same levels do. Additionally, one of the ‘classic’ symp- study, significant correlations between the cate- toms of overtraining (i.e. decreased testosterone) cholamines and the hormonal responses were also does not occur, rather there may actually be a slight observed (unpublished data), so it is speculated increase. Peripheral maladaptation appears to occur β that, to some extent, the sympathetic system is via down-regulation of the 2-receptors in skeletal driving both the acute hormonal responses as well muscle, resulting in a decreased catecholamine sens- as muscular performance. Whether the hormonal itivity. All of these symptoms are consistent with environment is influencing acute muscle perform- the sympathetic overtraining syndrome. Secondly, ance is not clear. during high volume overtraining, the endocrine In an attempt to understand some of the under- responses appear somewhat like they do with lying physiological mechanisms for decreased per- endurance overtraining. This is, testosterone is formances with high intensity resistance exercise typically attenuated, while cortisol and the T/C β overtraining, skeletal muscle 2-receptor density ratio are either unaffected or attenuated. On the was examined (Fry, A.C. et al. 2004b). Not sur- other hand, increases in sympathetic activity are β prisingly, a 37% decrease in 2-receptor density beginning, at least at rest. While these are far from was observed after 2 weeks of the high intensity the classic parasympathetic overtraining syndrome, overtraining. When receptor densities were com- it is reflective of an overtrained state that is transi- bined with resting catecholamine data, the ratio tioning from a sympathetic type to one taking on a β β of epinephrine/ 2-receptor density (EPI/ 2) ex- few of the characteristics of the parasympathetic hibited an extremely large increase (ES = 1.16). type. To date, the data available on peripheral mal- β Although far from conclusive, these data are evid- adaptation of the 2-receptors leaves much to ques- ence of a decreased catecholamine sensitivity such tion. Are these receptors disappearing or are they 588 chapter 37

just being turned off via phosphorylation or inter- pituitary ACTH release. In the maladaptation stage, β nalization? Are the 2-receptor affinity or binding a decreased rise in pituitary hormones (ACTH, GH, characteristics being altered? How do these peri- LH, FSH) is reported (Barron et al. 1985; Lehmann pheral maladaptations ultimately influence central et al. 1993b, 1998b, 1999a, 1999b; Urhausen et al. regulation (i.e. central neurotransmitters, ANS, 1995, 1998; Wittert et al. 1996). But behind the seem- hypothalamic–pituitary axis) of the body? These ingly uniform acute hormonal response to exercise, and more questions remain to be studied. explaining the disturbance to the neuroendocrine system caused by overtraining is not that simple. Central regulation during overtraining The trigger that eventually leads to overtraining may be any of a number of mediators with separate Is the brain involved? Over the last few years, there regulatory mechanisms. has been significant interest in determining specific peripheral markers for the metabolic, physiological Which neurotransmitters? The observation that other and psychological responses to exercise that it is stressors in addition to exercise (job, social, travel, suggested are associated with overtraining. To date, inadequate nutrition, etc.) seem to predispose an relatively little attention has been placed on the role athlete to overtraining (Foster & Lehmann 1997), of the central nervous system in overtraining and links overtraining with the psychosocial signs of the fatigue during exercise and training (Meeusen maladaptive response to intense training. The work 1999). We will focus on the involvement of the cent- of Morgan and colleagues (Morgan et al. 1987, 1988; ral nervous system and the role of neurotransmit- O’Connor et al. 1989, 1991; Raglin & Morgan 1994; ters and neuromodulators in possible mechanisms O’Connor 1997) clearly indicates that psychological that underly overtraining. factors, especially mood state disturbances, are For several years it has been hypothesized that a effective in predicting the onset of overtraining. The hormonal mediated central dysregulation occurs symptoms associated with overtraining, such as during the pathogenesis of overtraining, and that changes in emotional behavior, prolonged feelings measurements of blood hormones could help to of fatigue, sleep disturbances and hormonal dys- detect the overtraining syndrome (Kuipers & Keizer functions are indicative of changes in the regulation 1988; Fry, R.W. et al. 1991; Urhausen et al. 1995, 1998; and co-ordinative function of the hypothalamus Fry & Kraemer 1997; Keizer 1998). The results of (Meeusen 1999; Armstrong & Van Heest 2002). the research devoted to this subject is far from The hypothalamus is under the control of several unanimous, mostly because of the differences in ‘higher’ brain centers and several neurotransmitters measuring methods, and/or detection limits of the (Meeusen & De Meirleir 1995). Amongst these analytical equipment used. transmitters, serotonin (5-HT) is known to play a For a long time, the T/C ratio was considered a major role in various neuroendocrine and beha- good indicator of the overtraining state. This ratio vioral functions, for example activation of the HPA decreases in relation to the intensity and duration axis, feeding and locomotion (Wilckens et al. 1992). of training and, as previously mentioned, it seems The possibility that impaired neurotransmission at likely that this ratio is an indicator of the actual the various central aminergic synapses is associated physiological strain of training, rather than over- with major disturbances of the central nervous sys- training (Urhausen et al. 1995; Meeusen 1999). Most tem such as depression (and possibly overtraining), of the literature agrees that the many types of over- has been receiving more attention over the past training and overreaching must be viewed on a con- several years. It has been suggested that exercise tinuum, with a disturbance, an adaptation and, exerts its putative psychological effects via the same finally, a maladaptation of the HPA axis (Lehmann neurochemical substrate (the monoamines) as the et al. 1993a, 1998b, 1999a, 1999b; Urhausen et al. 1995, antidepressant drugs known to increase the syn- 1998; Keizer 1998; Meeusen 1998, 1999). The HPA aptic availability of transmitters (Meeusen 1999; axis adaptation is often characterized by increased Armstrong & Van Heest 2002; Uusitalo et al. 2004). endocrinology of overtraining 589

The target brain regions of interest seem to be the stress is believed to increase the secretion of ACTH basal ganglia, the limbic system, the hippocampus by stimulating the secretion of CRH (Tuomisto & and the hypothalamus, which have been found to Mannisto 1985; Plotsky et al. 1989). play a role in motor behaviors and emotional and It has been shown previously that there exists motivational states (Dunn & Dishman 1991). Like- a controlling or modulating role of the serotonin wise, study of the neurobiological aspects of exer- system on the function of nigro-striatal and ventral cise may add to a more complete understanding of tegmental-limbic dopamine systems (Palfreyman the etiology and treatment of overtraining. et al. 1993). The fact that raphe cell bodies are densely There is evidence that central serotonergic sys- innervated with noradrenergic cell terminals em- tems act upon the sympathetic nervous system and phasize that the locus coeruleus norepinephrine the HPA axis, and that reciprocally glucocorticoids and the dorsal raphe serotonergic system recipro- and catecholamines (derived from sympathetic cally innervate each other (Tuomisto & Mannisto nerves and the adrenomedulla) affect central sero- 1985). We have to point out that interaction of sero- tonergic systems (Chaouloff 1993). In pathological tonergic, dopaminergic, noradrenergic and other situations such as in major depression (Dishman neurotransmitter and neuromodulator systems could 1997), post-traumatic stress disorders (PTSD) (directly or indirectly) influence central and peri- (Liberzon et al. 1999; Porter et al. 2004) and probably pheral parameters. also in overtraining, the glucocorticoids and the brain Traditionally, the main criterion for a stress monoaminergic systems apparently fail to restrain response is an increase in the secretion of stress the HPA response to stress (Meeusen et al. 2004). hormones. On this basis, a decline in hormonal secretion when stress is repeated or prolonged is Which mechanisms? As it has been shown that exer- commonly interpreted as indicating stress adapta- cise and training influence neurotransmitter release tion (Stanford 1993). A large body of evidence in various brain nuclei (Meeusen et al. 1994, 1995, indicates that stressful experiences also alter neuro- 1996, 1997, 2001), possible dysregulation at this transmitter metabolism and release in several brain level could play a key role in the maladaptation areas (Kuipers & Keizer 1988; Abercrombie et al. to the ‘stress’ of exercise, training and overtraining. 1989; Weizman et al. 1989; Keefe et al. 1990; Meeusen et al. (1997) showed that endurance train- Nisembaum et al. 1991; Imperato et al. 1992; Gresch ing decreases basal neurotransmitter outflow in the et al. 1994; Jordan et al. 1994; Finlay et al. 1995; Kirby striatum of rats, while maintaining the necessary et al. 1997). Repeated exposure to stress may lead to sensitivity for responses to acute exercise. These a different responsiveness to subsequent stressful observations raise the possibility that there could experiences depending on the stressor as well as on exist an exercise-induced change in receptor sensit- the stimuli paired with the stressor. This in turn may ivity (Meeusen et al. 1997). lead to either an unchanged, increased or decreased There is an extensive amount of anatomical, bio- neurotransmitter and receptor function. Behavioral chemical and physiological data lending support adaptation (release, receptor sensitivity, receptor to an excitatory influence of central serotonergic binding, etc.) in higher brain centers will certainly systems upon the HPA axis (Chaouloff 1993; Porter influence hypothalamic output (Lachuer et al. et al. 2004). Norepinephrine is also linked to the 1994). It has been shown that immobilization stress regulation of the HPA axis, which is dysregulated not only increases hypothalamic monoamine not only in overtraining (Keizer 1998), but also in release, but consequently CRH and ACTH secretion conditions such as in major depression and certain (Shintani et al. 1995). Chronic stress and the sub- anxiety disorders (Chaouloff 1993; Dishman 1997). sequent chronic peripheral glucocorticoid secretion In these conditions, glucocorticoids and brain could play an important role in the desensitiza- monoaminergic systems apparently fail to restrain tion of higher brain center responses during acute the HPA response to stress. Activation of the brain stressors, since it has been shown that in acute (and noradrenergic and serotonergic systems during also chronic) immobilization, the responsiveness of 590 chapter 37

hypothalamic CRH neurons rapidly falls (Cizza feature of the overtraining syndrome is the inability et al. 1993). These adaptation mechanisms could be to sustain intense exercise, and a decreased sports- the consequence of changes in neurotransmitter specific performance capacity when the training release, depletion of CRH and/or desensitization of load is maintained or even increased (Urhausen hypothalamic hormonal release to afferent neuro- et al. 1995; Meeusen et al. 2004). Athletes suffering transmitter input (Cizza et al. 1993). In overtraining from overtraining are able to start a normal training it is assumed that a ‘maladaptation’ to chronic exer- sequence or a race at their normal training pace, but cise (and other) stress occurs. There is evidence that are not able to complete the training load they are apart from other brain areas, the hippocampus given, or race as usual. Meeusen et al. (2004) recently down-regulates the HPA axis in both rats and prim- published a test protocol with two consecutive ates. Thorré et al. (1997) analysed the impact of maximal exercise tests separated by 4 h. With this streptozotocin (STZ) -elicited diabetes (a chronic protocol they found that in order to detect signs of stress for the body) on hippocampal extracellular 5- overtraining and distinguish them from normal HT levels both under basal conditions and during training responses or just overreaching, this method restraint stress. Restraint stress is a procedure known is a good indicator, not only of the recovery capacity to stimulate 5-HT synthesis/metabolism and release. of the athlete, but also of the ability to normally per- The chronic stress decreased hippocampal 5-HT form the second bout of exercise. The use of two metabolism, but did not alter baseline 5-HT release bouts of maximal exercise to study neuroendocrine (compared to controls), while restraint stress in- variations showed an adapted exercise-induced creased extracellular 5-HT levels in controls, but did increase of ACTH, prolactin and GH to two success- not increase 5-HT in the diabetic animals (Thorré ive exercise bouts. Particularly in overtrained ath- et al. 1997). This may indicate that in the chronically letes, the ‘overshoot’ of these hormones in the first stressed animals the central response to an acute exercise bout, followed by a complete suppression stressor (i.e. restraint) is impaired. in the second exercise bout is interesting (Fig. 37.3). The examples cited previously illustrate that This could indicate a hypersensitivity of the pituit- in several brain nuclei, chronic stress will create ary followed by an exhaustion afterwards. Previous an adaptation mechanism (autoreceptor mediated, reports that used a single exercise protocol found neurotransmitter interactions, or other mechanisms). similar effects (Meeusen et al. 2004). It appears that When an animal is confronted with a novel stress- the use of two exercise bouts is more useful in ful stimulus, sensitization will occur. However, in detecting overreaching for preventing overtraining. chronic, very intense stress situations (as in STZ Early detection of overreaching may be very import- diabetes where other peripheral hormonal mechan- ant in the prevention of overtraining. isms also play an important role), this sensitization of hippocampal 5-HT release does not occur. One Central mechanism conclusions. Several neurotrans- might speculate that in overtraining (the step mitters might play an important role in the onset of beyond coping with stress), a comparable mechan- the central dysfunctions that occur in the maladap- ism occurs. It remains to be elucidated if these tation to chronic exercise stress or overtraining. The results from a ‘non-exercise’ chronic stress situation multifactorial aspect of central nervous functioning can be translated to another chronic stress such as with different effects of increased neurotransmitter overtraining (Meeusen 1999). release in different brain regions, the complexity of several receptors and the mutual interaction of Assessment of overtraining. In overtrained athletes, a various neurotransmitters (having excitatory and number of signs and symptoms have been asso- inhibitory features) illustrates that this multifac- ciated with this imbalance between training and torial disturbance in the onset of overtraining could recovery. However, reliable diagnostic markers for give rise to more questions than answers. What distinguishing between well-trained, overreached remains to be seen is how individual elements of and overtrained athletes are lacking. A hallmark the several neurotransmitters and neuromodulators endocrinology of overtraining 591

ACTH Growth hormone 600 *# 12 000 Test 1 * * Test 1 500 Test 2 10 000 * Test 2

400 8000 * 300 6000

200 4000 % Change from rest 100 % Change from rest 2000

0 0 Pre-camp Post-camp OT (n = 1) Pre-camp Post-camp OT (n = 1) (a) (OR) (b) (OR)

PRL 350 * Test 1 300 * Test 2 250 * Fig. 37.3 Adrenocorticotropic hormone (ACTH) (a), * growth hormone (GH) (b) and prolactin (PRL) (c) results 200 for the two exercise tests to exhaustion respectively for the 150 trained group (Pre-camp), the overreached group (Post- camp, OR) and the overtrained athlete (OT). All data are 100

presented as percent increase from each baseline value, % Change from rest and presented as mean + SEM. *, Statistical significance at 50 p < 0.05 between exercise and rest; #, statistical significance 0 at p < 0.05 between the Pre-camp and the Post-camp Pre-camp Post-camp OT (n = 1) group. (Modified from Meeusen et al. 2004.) (c) (OR) contribute to this scheme: how modulation of pre- training are added to the story, it can become quite and post-synaptic receptors, coupled with rapid confusing. It is critical to remember that the actual or delayed changes in transmitter synthesis and definition of overtraining includes some type of release, accounts for the behavioral and perform- long-term performance decrement. Even though ance impact of a stress called overtraining. some of the classical symptoms of an overtraining Finally, we need not look at the present research syndrome may be apparent, overtraining has not data as being totally contradictory or inconsistent, occurred until performance is adversely affected. because it is clear that understanding brain function Regardless, several aspects of overtraining and takes more than just one series of experiments. At overreaching are becoming apparent. Much of the this moment it seems that it will probably take more maladaptations associated with these performance ‘brains’ than a human has available to understand decrements are due in part to regulation (or dysre- only a part of brain functioning. gulation) by the neuroendocrine system. How the various hormones and neurotransmitters respond is Summary dependent on the type of exercise stress encoun- tered. Although not all inclusive, Fig. 37.4 identifies Needless to say, the neuroendocrine system in a some of the key endocrine and related variables healthy human body is an extremely complex that respond to the various types of training. What scenario. When the additional complexities of over- is readily apparent from this figure is that the 592 chapter 37

Combined Resistance exercise activities Endurance exercise overtraining (team sports, overtraining military, occupational)

High High High High High volume intensity intensity volume volume (usually) NC—rest NC Testosterone —acute LH–NC or LH–NC LH–NC LH—

NC—rest NC—rest CRH & Cortisol —acute NC or NC —acute ACTH ACTH–NC ACTH–NC ACTH— acute sensitivity to ACTH

NC—rest Tes/cort ratio —acute or NC NC

NC or —rest NC or —rest —rest Catecholamines —acute —rest NC —acute —acute NC or β 2-Receptor density in skeletal muscle In lymphocytes

Catecholamine sensitivity

5-HT DA Central nervous GABA NE system GLU

Heat shock proteins (HSP70)

GH— GH–—NC or GH— acute GH— IGF-I—

Other IGF-I— IGF-I—NC or Insulin— acute TSH— T3— Insulin— Insulin— Leptin— Insulin— (rest)

Fig. 37.4 Overview of the endocrine and neuroendocrine responses and adaptations to various types of overtraining. ↑=increase, ↓=decrease; ACTH, adrenocorticotropic hormone; CRH, corticotrophin-releasing hormone; GABA, gamma- aminobutyric acid; GH, growth hormone; GLU, glucagon; IGF-I, insulin-like growth factor I; LH, luteinizing hormone;

NC, no change; T3, triiodothyronine; Tes/Cort, testosterone/cortisol; TSH, thyroid-stimulating hormone. endocrine responses can be quite different depend- the fundamental physiological and psychological ing on the physiological characteristics of the type differences inherent to each type of training and of training. As a result, it is very important that the sports that routinely utilize these training those studying overtraining have an appreciation of methods. endocrinology of overtraining 593

Overtraining Trophic hormones

Decreased Sympathetic Epi & NE sensitivity overtraining syndrome Acute hormones β 2-receptors

Probably not causal Catecholamine sensitivity

Relationship between Relationship between Epi or NE & muscle tes or tes/cort & performance muscle performance

? Performance

(a)

Overtraining CNS drive

Sympathetic exhaustion β 2-receptors Sympathetic Hypothalamic– drive pituitary drive

Fig. 37.5 Theoretical paradigm of the physiological progression of Catecholamine sympathetic (a) and parasympathetic sensitivity Epi & NE Trophic hormones (b) overtraining syndromes. The shaded areas indicate that the etiology of the sympathetic Energy availability Hormones overtraining syndrome focuses on peripheral mechanisms, while that of the parasympathetic overtraining Parasympathetic syndrome focuses on central overtraining mechanisms. CNS, central nervous syndrome Performance system; Epi, epinephrine; NE, norepinephrine; tes, testosterone; tes/cort, testosterone/cortisol. (b)

The role of the ANS in the onset of an over- eventually by a parasympathetic predominance. training syndrome is also becoming clearer. There Figure 37.5 illustrates theoretical progressions of appears to be a sequence in this response, with a each of these overtraining syndromes. As suggested sympathetic predominance occurring first, followed in this chapter, the limiting factors for a sympathetic 594 chapter 37

ANS overtraining paradigm

Normal training

Stressful training (no performance decrements)

Overreaching (short-term performance decrements)

Resistance Resistance Combined Endurance Endurance exercise exercise activities exercise exercise Low volume High volume High volume Mod. volume High volume High intensity Low intensity (usually) High intensity Mod. intensity · · (% 1-RM) (% 1-RM) (% VO2max) (% VO2max) Overtraining Sympathetic Mixed sympathetic & Parasympathetic (long-term overtraining parasympathetic overtraining performance syndrome syndrome decrements)

Acute Acute & chronic responses responses most sensitive sensitive

Limiting factors: Limiting factors: orthopedic & CNS & energy neuromuscular, availability, psychological psychological

Shortest Longest recovery? recovery?

Fig. 37.6 Theoretical paradigm of the progression from normal training to overtraining for various types of sports and physical activity as it relates to the autonomic nervous system (ANS). Note that the sympathetic and parasympathetic overtraining syndromes represent the extremes of overtraining, and that many sports/activities fall in between these extremes. Also note that many symptoms of the overtraining syndromes may be present during lesser stages of overtraining, such as overreaching and phases of stressful training, that do not result in performance decrements. overtraining syndrome appear to be mainly peri- such as orthopedic limitations with very intense pheral in nature, while those of the parasympathetic resistance exercise or related activities. Figure 37.6 overtraining syndrome are central. Not all types of summarizes the likely progression of a variety of overtraining appear to make it all the way to a training stresses as they progress from normal train- parasympathetic overtraining syndrome. It is likely ing to stressful training to overreaching to over- that other limiting variables prohibit this occurring, training. It should be apparent from Fig. 37.6 that endocrinology of overtraining 595

some of the symptoms sometimes associated with Notes overtraining may appear before overtraining is an actual problem. This figure may also help illustrate J.M. Steinacker was responsible for the section on why overtraining for so many sports is difficult to endurance sports, A.C. Fry was responsible for the study, since so many sports fall in the middle of this section on strength sports and activities, and R. continuum, and possess a combination of high-force Meeusen was responsible for the section on central and high-endurance characteristics. mechanisms.

References

Abercrombie, E., Keefe, K., DiFrischia, D. van Borselen, F., Vos, N.H., Fry, A.C. & prefrontal cortex induced by acute and & Zigmond, M. (1989) Differential effect Kraemer, W.J. (1992) The role of chronic stress effects of diazepam. of stress on in vivo dopamine release in anaerobic exercise in overtraining. Neuroscience 64(3), 619–628. striatum, nucleus accumbens, and National Strength and Conditioning Foster, C. & Lehmann, M. (1997) medial frontal cortex. Journal of Association Journal 14(3), 74–79. Overtraining. In: Running Injuries Neurochemistry 52, 1655–1658. Carro, E., Seoane, L.M., Senaris, R. et al. (Guten, G.N., ed.). W.B. Saunders, Co., Adlercreutz, H., Harkonen, M., (1998) Interaction between leptin and Philadelphia, PA: 173–187. Kuoppasalmi, K. et al. (1986) Effect neuropeptide Y on in vivo growth Friedman, J.E., Ferrara, C.M., Aulak, K.S. of training on plasma anabolic and hormone secretion. Neuroendocrinology. et al. (1997) Exercise training down- catabolic steroid hormones and their 68, 187–191. regulates ob gene expression in the response during physical exercise. Chaouloff, F. (1993) genetically obese SHHF/Mcc-fa(cp) International Journal of Sports Medicine 7, Physiolopharmacological interactions rat. Hormone and Metabolic Research 29, S27–S28. between stress hormones and central 214–219. Argente, J., Caballo, N., Barrios, V. et al. serotonergic systems. Brain Research. Fry, A.C. (1998) The role of training (1997) Multiple endocrine abnormalities Brain Research Reviews 18, 1–32. intensity in resistance exercise of the growth hormone and insulin-like Chicharro, J.L., Lopez-Mojares, L.M., overtraining and overreaching. In: growth factor axis in patients with Lucia, A. et al. (1998) Overtraining Overtraining and Overreaching in Sports anorexia nervosa: effect of short- and parameters in special military units. (Kreider, R.B., Fry, A.C. & O’Toole, M.L., long-term weight recuperation. Journal Aviation, Space, and Environmental eds.). Human Kinetics, Champaign, IL: of Clinical Endocrinology and Metabolism Medicine 69(6), 562–568. 107–130. 82, 2084–2092. Cizza, G., Kvetnansky, R., Tartaglia, M. Fry, A.C. (1999) Overload and Armstrong, L.E. & Van Heest, J.L. et al. (1993) Immobolisation stress regeneration during resistance exercise. (2002) The unknown mechanism rapidly decreases hypothalamic In: Overload, Performance Incompetence, of the overtraining syndrome: corticotropin-releasing hormone and Regeneration in Sport (Lehmann, M., clues from depression and secretion in vitro in the male 344/N Steinacker, J.M., Gastmann, U. & Keizer, psychoneuroimmunology. Sports fischer rat. Life Sciences 53(3), 233–240. H., eds.). Kluwer Academic/Plenum Medicine 32(3), 185–209. Dishman, R. (1997) The norepinephrine Publishers, New York: 149–162. Atalay, M., Seene, T., Hanninen, O. & hypothesis. In: Physical Activity and Fry, A.C. & Kraemer, W.J. (1997) Sen, C.K. (1996) Skeletal muscle and Mental Health (Morgan, W.P., ed.). Resistance exercise overtraining and heart antioxidant defences in response Taylor & Francis, Washington, DC: overreaching: neuroendocrine to sprint training. Acta Physiologica 199–212. responses, Sports Medicine 23(2), Scandinavica. 158, 129–134. Dunn, A. & Dishman, R. (1991) Exercise 106–129. Baldwin, K.M., Fitts, R.H., Booth, F.W., and the neurobiology of depression. Fry, A.C., Kraemer, W.J., Stone, M.H. et al. Winder, W.W. & Holloszy, J.O. (1975) Exercise and Sport Sciences Reviews 19, (1993) Endocrine and performance Depletion of muscle and liver glycogen 41–98. responses to high volume training and during exercise. Protective effect Eliakim, A., Brasel J.A., Mohan, S., Wong, amino acid supplementation in elite of training. Pflugers Archiv 354, W.L.T. & Cooper, D.M. (1998) Increased junior weightlifters. International Journal 203–212. physical activity and the growth of Sports Nutrition 3, 306–322. Barron, G., Noakes, T., Levy, W., Smidt, C. hormone–IGF-I axis in adolescent males. Fry, A.C., Kraemer, W.J., van Borselen, F. & Millar, R. (1985) Hypothalamic American Journal of Physiology 275, et al. (1994a) Catecholamine responses dysfunction in overtrained athletes. R308–R314. to short-term high intensity resistance Journal of Clinical Endocrinology and Engfred, K., Kjær, M., Secher, N.H. et al. exercise overtraining. Journal of Applied Metabolism 60, 803–806. (1994) Hypoxia and training-induced Physiology 77(2), 941–946. Baskin, D.G., Breininger, J.F. & Schwartz, adaptation of hormonal responses to Fry, A.C., Kraemer, W.J., Stone, M.H. et al. M.W. (1999) Leptin receptor mRNA exercise in humans. European Journal of (1994b) Exercise-induced endocrine identifies a subpopulation of Applied Physiology 68, 303–309. responses to over-reaching before and neuropeptide Y neurons activated by Finlay, J., Zigmond, M. & Abercrombie, E. after 1 year of weightlifting training. fasting in rat hypothalamus. Diabetes 48, (1995) Increased dopamine and Canadian Journal of Applied Physiology 828–833. norepinephrine release in medial 14(4), 400–410. 596 chapter 37

Fry, A.C., Kraemer, W.J. & Ramsey, L.T. Heiman, M.L., Ahima, R.S., Craft, L.S. Kraemer, W.J., French, D.N., Paxton, N.J. (1998) Pituitary–adrenal–gonadal et al. (1997) Leptin inhibition of the et al. (2004) Changes in exercise responses to high intensity resistance hypothalamic–pituitary–adrenal axis in performance and hormonal exercise overtraining. Journal of Applied response to stress. Endocrinology 13(8), concentrations over a Big Ten soccer Physiology 85(6), 2352–2359. 3859–3863. season in starters and nonstarters. Fry, A.C., Kraemer, W.J., Koziris, L.P. Imperato, A., Angelucci, L., Casolini, P., Journal of Strength and Conditioning et al. (2000) The relationship between Zocchi, A. & Puglisi-Allegra, S. (1992) Research 18(1), 121–128. testosterone, cortisol and weightlifting Repeated stressful experiences Kreider, R.B., Fry A.C. & O’Toole, M.L. performance. Journal of Strength and differently affect limbic dopamine (1998) Overtraining in sport: terms, Conditioning Research 14(3), 338–343. release during and following stress. definitions, and prevalence. In: Fry, A.C., Häkkinen, K. & Kraemer, W.J. Brain Research 577, 194–199. Overtraining and Overreaching in Sports (2001) Special considerations in strength Jenkins, P.J., Ibanez-Santos, X., Holly, J. (Kreider, R.B., Fry, A.C. & O’Toole, M.L., training. In: Handbook of Sports Medicine et al. (1993) IGFBP-1: a metabolic signal eds.). Human Kinetics, Champaign, IL: and Science––Strength Training for Sport associated with exercise-induced vii–ix. (Kraemer, W.J. & Häkkinen, K., eds.). amenorrhoea. Neuroendocrinology 57, Kuipers, H. (1998) Training and Blackwell Science, Oxford: 135–162. 600–604. overtraining: an introduction. Medicine Fry, A.C., Chiu, L.Z.F., Schilling, B.K. & Jordan, S., Kramer, G., Zukas, P. & Petty, F. and Sciences in Sports and Exercise 30(7), Weiss, L.W. (2004a) Effect of a dietary (1994) Previous stress increases in vivo 1137–1139. β supplement on skeletal muscle 2 biogenic amine response to swim Kuipers, H. & Keizer, H.A. (1988) receptor adaptations to high power stress. Neurochemical Research 19(12), Overtraining in elite athletes. Review resistance exercise overreaching 1521–1525. and directions for the future. Sports [abstract]. NSCA National Conference, Jost, J., Weiss, M. & Weicker, H. (1990) Medicine 6, 79–92. Minneapolis, MN. Sympathoadrenergic regulation and the Lachuer, J., Delton, I., Buda, M. & Tappaz, Fry, A.C., Schilling, B.K., Chiu, L.Z.F. adrenoceptor system. Journal of Applied M. (1994) The habituation of brainstem β et al. (2004b) Skeletal muscle 2 receptor Physiology 68, 897–904. catecholaminergic groups to chronic maladaptations to high intensity Kanaley, J.A., Weltman, J.Y., Veldhuis, J.D. daily restraint stress is stress specific like resistance exercise overtraining et al. (1997) Human growth hormone that of the hypothalamo–pituitary– [abstract]. NSCA National Conference, response to repeated bouts of aerobic adrenal axis. Brain Research 638, 196–202. Minneapolis, MN. exercise. Journal of Applied Physiology 83, Laughlin, G.A. & Yen, S.S. (1996) Fry, R.W., Morton, A.R. & Keast, D. (1991) 1756–1761. Nutritional and endocrine–metabolic Overtraining in athletes: an update. Keefe, K., Stricker, E., Zigmond, M. & aberrations in amenorrheic athletes. Sports Medicine 12, 32–65. Abercrombie, E. (1990) Environmental Journal of Clinical Endocrinology and Gorostiaga, E.M., Izquierdo, M., Iturralde, stress increases extracellular dopamine Metabolism 81, 4301–4309. P., Ruesta, M. & Ibanez, J. (1999) Effects in striatum of 6-hydroxydopamine- Leal-Cerro, A., Garcia-Luna, P.P., Astorga, of heavy resistance training on maximal treated rats: in vivo microdialysis R. et al. (1998) Serum leptin levels in and explosive force production, studies. Brain Research 527, 350–353. male marathon athletes before and after endurance and serum hormones in Keizer, H. (1998) Neuroendocrine aspects the marathon run. Journal of Clinical adolescent handball players. European of overtraining. In: Overtraining and Endocrinology and Metabolism 83, Journal of Applied Physiology 80(5), Overreaching in Sports (Kreider, R.B., Fry, 2376–2379. 485–493. A.C. & O’Toole, M.L., eds.). Human Lehmann, M., Dickhuth, H.H., Gendrisch, Gray, A.B., Telford, R.D. & Weidemann, Kinetics, Champaign, IL: 145–167. G. et al. (1991) Training–overtraining. M.J. (1993) Endocrine response to Kirby, L., Chou-Green, J., Davis, K. A prospective, experimental study with intense interval exercise. European & Lucki, I. (1997) The effects of experienced middle- and long-distance Journal of Applied Physiology 66, 366–371. different stressors on extracellular runners. International Journal of Sports Gresch, P., Sved A., Zigmond, M. & Finlay, 5-hydroxytryptamine and 5- Medicine 12, 444–452. J. (1994) Stress-induced sensitization of hydroxyindoleacetic acid. Brain Lehmann, M., Baumgartl, P., Wiesenack, dopamine and norepinephrine efflux Research 760(1–2), 218–230. C. et al. (1992a) Training–overtraining: in medial prefrontal cortex of the rat. Koistinen, H., Koistinen, R., Selenius, L., influence of a defined increase in Journal of Neurochemistry 63; 575–583. Ylikorkala, Q. & Seppala, M. (1996) training volume vs. training intensity on Häkkinen, K. & Pakarinen, A. (1993) Acute Effect of marathon run on serum IGF-I performance, catecholamines and some hormonal responses to two different and IGF-binding protein 1 and 3 levels. metabolic parameters in experienced fatiguing heavy-resistance protocols Journal of Applied Physiology 80, 760–764. middle- and long-distance runners. in male athletes. Journal of Applied Kraemer, W.J. & Nindl, B.C. (1998) Factors European Journal of Applied Physiology 64, Physiology 74, 882–887. involved with overtraining for strength 169–177. Häkkinen, K., Pakarinen, A., Alen, M., and power. In: Overtraining and Lehmann, M., Schnee, W., Scheu, R., Kauhanen, H. & Komi, P.V. (1987) Overreaching in Sports (Kreider, R.B., Fry, Stockhausen, W. & Bachl, N. (1992b) Relationships between training volume, A.C. & O’Toole, M.L., eds.). Human Decreased nocturnal catecholamine physical performance capacity and Kinetics, Champaign, IL: 69–86. excretion: parameter for an overtraining serum hormone concentrations during Kraemer, W.J., Fry, A.C., Stone, M.H. et al. syndrome in athletes? International prolonged training in elite weight lifters. (1992) Acute hormonal responses in elite Journal of Sports Medicine 13, 236–242. International Journal of Sports Medicine 8, junior weightlifters. International Journal Lehmann, M., Foster, C. & Keul, J. (1993a) 61–65. of Sports Medicine 13(20), 103–109. Overtraining in endurance athletes: endocrinology of overtraining 597

a brief review. Medicine and Science in Liu, Y., Mayr, S., Opitz-Gress, A. et al. protocol to detect subtle differences in Sports and Exercise 25, 854–862. (1999) Human skeletal muscle HSP70 (over)training status. European Journal Lehmann, M., Knizia, K., Gastmann, U. response to training in highly trained of Applied Physiology 91(2–3), 140–146. et al. (1993b) Influence of 6-week, 6 days rowers. Journal of Applied Physiology 86, Morgan, W., Brown, D., Raglin, J., per week, training on pituitary function 101–104. O’Connor, P. & Ellickson, K. (1987) in recreational athletes. British Journal of Liu, Y., Lormes, W., Baur, C. et al. (2000) Psychological monitoring of Sports Medicine 27, 186–192. Human skeletal muscle HSP70 response overtraining and staleness. British Lehmann, M., Jakob, E., Gastmann, U., to physical training depends on exercise Journal of Sports Medicine 21, 107–114. Steinacker, J.M. & Keul, J. (1995) intensity. International Journal of Sports Morgan, W., Costill, D., Flynn, M., Unaccustomed high mileage compared Medicine 21, 351–355. Raglin, J. & O’Connor, P. (1988) Mood to intensity training-related Maso, F., Lac, G., Filaire, E., Michauz, O. disturbance following increased training neuromuscular excitability in distance & Robert, A. (2004) Salivary testosterone in swimmers. Medicine and Science in runners. European Journal of Applied and cortisol in rugby players: correlation Sports and Exercise 20(4), 408–414. Physiology 70, 457–461. with psychological overtraining items. Murlasits, Z., Fry, A.C., Rubin, M.R. et al. Lehmann, M., Baur, S., Netzer, N. & British Journal of Sports Medicine 38(3), (1999) Hormonal responses to off-season Gastmann, U. (1997a) Monitoring high- 260–263. training program for American football intensity endurance training using Meeusen, R. (1998) Overtraining, indoor [abstract]. NSCA National Conference, neuromuscular excitability to recognize and outdoor. Vlaams tijdschrift voor Kansas City, MO. overtraining. European Journal of Applied Sportgeneeskunde & Sportwetenschappen Newsholme, E.A., Blomstrand, E., Physiology 76, 187–191. 19(74), 8–19. McAndrew, N. & Parry-Billings, M. Lehmann, M., Lormes, W., Opitz-Gress, A. Meeusen, R. (1999) Overtraining and the (1992) Biochemical causes of fatigue and et al. (1997b) Training and overtraining: central nervous system, the missing overtraining. In: Endurance in Sports an overview and experimental results link? In: Overload, Performance (Shaphard, R.J. Astrand, P.-O., eds.). in endurance sports. Journal of Sports Incompetence, and Regeneration in Blackwell Scientific, Oxford: 351–364. Medicine and Physical Fitness 37, 7–17. Sport (Lehmann, M., Steinacker, J.M., Nisembaum, L., Zigmond, M., Sved, A. & Lehmann, M., Foster, C., Dickhuth, H.H. Gastmann, U. & Keizer, H., eds.). Abercrombie, E. (1991) Prior exposure & Gastmann, U. (1998a) Autonomic Kluwer Academic/Plenum Publishers, to chronic stress results in enhanced imbalance hypothesis and overtraining New York: 187–202. synthesis and release of hippocampal syndrome. Medicine and Science in Sports Meeusen, R. & De Meirleir, K. (1995) norepinephrine in response to a novel and Exercise 30(7), 1140–1145. Exercise and brain neurotransmission. stressor. Journal of Neuroscience 11, Lehmann, M., Foster, C., Netzer, N. et al. Sports Medicine 20(3), 160–188. 1478–1484. (1998b) Physiological responses to Meeusen, R., Sarre, S., Michotte, Y., O’Connor, P. (1997) Overtraining and short- and long-term overtraining in Ebinger, G. & De Meirleir, K. (1994) The staleness. In: Physical Activity and Mental endurance athletes. In: Overtraining and effects of exercise on neurotransmission Health (Morgan, W.P., ed.). Taylor & Overreaching in Sports (Kreider, R.B., in rat striatum, a microdialysis study. Francis, Washington, DC: 145–160. Fry, A.C. & O’Toole, M.L., eds.). Human In: Monitoring Molecules in Neuroscience O’Connor, P., Morgan, W., Raglin, J., Kinetics, Champaign, IL: 19–46. (Louilot, A., Durkin, T., Spampinato, U. Barksdale, C. & Kalin, N. (1989) Lehmann, M., Foster, C., Gastmann, U., & Cador, M., eds.). Seigmosse: 181–182. Mood state and salivary cortisol levels Keizer, H. & Steinacker, J. (1999a) Meeusen, R., Smolders, I., Sarre, S. following overtraining in female Definitions, types, symptoms, findings, et al. (1995) The effects of exercise on swimmers. Psychoneuroendocrine 14, underlying mechanisms, and frequency extracellular glutamate (GLU) and 303–310. of overtraining and overtraining GABA in rat striatum, a microdialysis O’Connor, P., Morgan, W. & Raglin, J. syndrome. In: Overload, Performance study. Medicine and Science in Sports and (1991) Psychobiologic effects of 3 days Incompetence, and Regeneration in Sport Exercise 27(5), S215. of increased training in female and male (Lehmann, M., Steinacker, J.M., Meeusen, R., Chaouloff, F., Thorré, K. et al. swimmers. Medicine and Science in Sports Gastmann, U. & Keizer, H., eds.). (1996) Effects of tryptophan and/or and Exercise 23, 1055–1061. Kluwer Academic/Plenum Publishers, acute running on extracellular 5-HT and Palfreyman, M., Schmidt, C., Sorensen, S. New York: 1–6. 5-HIAA levels in the hippocampus of et al. (1993) Electrophysiological, Lehmann, M., Gastmann, U., Baur, S. food-deprived rats. Brain Research 740, biochemical and behavioural evidence

et al. (1999b) Selected parameters and 245–252. for 5-HT2 and 5-HT3 mediated control mechanisms of peripheral and central Meeusen, R., Smolders, I., Sarre, S. et al. of dopaminergic function. fatigue and regeneration in overtrained (1997) Endurance training effects on Psychopharmacology 112, S60–S67. athletes. In: Overload, Performance striatal neurotransmitter release, an Pedersen, B.K., Steensberg, A., Fischer, C. Incompetence, and Regeneration in Sport ‘in vivo’ microdialysis study. Acta et al. (2001) Exercise and cytokines with (Lehmann, M., Steinacker, J.M., Physiologica Scandinavica 159, 335–341. particular focus on muscle derived IL-6. Gastmann, U. & Keizer, H., eds.). Meeusen, R., Piacentini, M.F. & De Exercise Immunology Review 7, 18–31. Kluwer Academic/Plenum Publishers, Meirleir, K. (2001) Brain microdialysis in Pette, D. & Staron, R.S. (1997) Mammalian New York: 7–25. exercise research. Sports Medicine 31(14), skeletal muscle fiber type transitions. Liberzon, I., Taylor, S., Amdur, R. et al. 965–983. International Review of Cytology 170, (1999) Brain activation in PTSD in Meeusen, R., Piacentini, M.F., Busschaert, 143–223. response to trauma-related stimuli. B. et al. (2004) Hormonal responses in Plotsky, P., Cunninham, E. & Widmaier, E. Biological Psychiatry 45, 817–826. athletes: the use of a two bout exercise (1989) Catecholaminergic modulation 598 chapter 37

of corticotropin-releasing factor and Selye, H. (1956) The Stress of Life. McGraw- hormone and cytokine response to adrenocorticotropin secretion. Endocrine Hill, New York. training. European Journal of Applied Reviews 10, 437–458. Shephard, R. (2000) Exercise and training Physiology 91, 382–391. Porter, R., Gallagher, P., Watson, S. in women, part I: influence of gender Stone, M.H. & Fry, A.C. (1998) The role of & Young, A. (2004) Corticosteroid– on exercise and training responses. increased training volume in resistance serotonin interactions in depression: Canadian Journal of Applied Physiology exercise overtraining and overreaching. a review of the human evidence. 25(1), 19–34. In: Overtraining and Overreaching in Psychopharmacology 173, 1–17. Shintani, F., Nakaki, T., Kanba, S. et al. Sports (Kreider, R.B., Fry, A.C. & Raastad, T., Glomsheller, T., Bjoro, T. (1995) Involvement of interleukin-1 in O’Toole, M.L., eds.). Human Kinetics, & Hallen, J. (2001) Changes in human immobilization stress-induced increase Champaign, IL: 87–106. skeletal muscle contractility and in plasma adrenocorticotropic hormones Stone, M.H., Keith, R.E., Kearney, J.T. hormone status during 2 weeks of heavy and in release of hypothalamic et al. (1991) Overtraining: a review of strength training. European Journal of monoamines in rat. Journal of the signs and symptoms and possible Applied Physiology 84, 54–63. Neuroscience 15(3), 1961–1970. causes. Journal of Applied Sports Science Raglin, J. & Morgan, W. (1994) Simsch, C., Lormes, W., Petersen, K.G. Research 5, 35–50. Development of a scale for use in et al. (2002) Training intensity influences Tannenbaum, G.S., Gurd, W. & Lapointe, monitoring training-induced distress leptin and thyroid hormones in highly M. (1998) Leptin is a potent stimulator of in athletes. International Journal of Sports trained rowers. International Journal of spontaneous pulsatile growth hormone Medicine 15(2), 84–88. Sports Medicine 23, 422–427. (GH) secretion and the GH response to Ratamess, N.A., Kraemer, W.J., Volek, J.S. Smith, L.L. (2004) Tissue trauma: the GH-releasing hormone. Endocrinology et al. (2003) The effects of amino underlying cause of overtraining 139, 3871–3875. acid supplementation on muscular syndrome? Journal of Strength and Thorré, K., Chaouloff, F., Sarre, S. et al. performance during resistance training Conditioning Research 18(1), 185–193. (1997) Differential effects of restraint overreaching. Journal of Strength and Snyder, A.C. (1998) Overtraining and stress on hippocampal 5-HT metabolism Conditioning Research 17(2), 250–258. glycogen depletion hypothesis. Medicine and extracellular levels of 5-HT in Richter, E.A., Derave, W. & Wojtaszewski, and Science in Sports and Exercise 30, streptozotocin-diabetic rats. Brain J.F. (2001) Glucose, exercise and insulin: 1146–1150. Research 772, 209–216. emerging concepts. Journal of Physiology Snyder, D.K., Clemmons, D.R. & Tuomisto, J. & Mannisto, P. (1985) 535, 313–322. Underwood, L.E. (1989) Dietary Neurotransmitter regulation of anterior Robson, P. (2003) Elucidating the carbohydrate content determines pituitary hormones. Pharmacological unexplained underperformance responsiveness to growth hormone in Reviews 37, 249–332. syndrome in endurance athletes: the energy-resticted humans. Journal of Urhausen, A., Gabriel, H. & Kindermann, interleukin-6 hypothesis. Sports Medicine Clinical Endocrinology and Metabolism W. (1995) Blood hormones as markers of 33(10), 771–781. 69, 745–752. training stress and overtraining. Sports Roelen, C.A., de Vries, W.R., Koppeschaar, Stanford, C. (1993) Monoamines in Medicine 20(4), 251–276. H.P. et al. (1997) Plasma insulin-like response and adaptation to stress. Urhausen, A., Gabriel, H. & Kindermann, growth factor-I and high affinity growth In: Stress, From Synapse to Syndrome W. (1998) Impaired pituitary hormonal hormone-binding protein levels increase (Stanford, S. & Salmon, P., eds.). response to exhaustive exercise in after 2 weeks of strenuous physical Academic Press, London: 281–331. overtrained endurance athletes. Medicine training. International Journal of Sports Steinacker, J.M., Lormes, W., Lehmann, M. and Science in Sports and Exercise 30(3), Medicine 18, 238–241. & Altenburg, D. (1998) Training of 407–414. Roemmich, J.N. & Sinning, W.E. (1997) rowers before world championships. Uusitalo, A.L., Huttunen, P., Hanin, Y., Weight loss and wrestling training: Medicine and Science in Sports and Exercise Uusitalo, A.J. & Rusko, H.K. (1998) effects on growth-related hormones. 30, 1158–1163. Hormonal responses to endurance Journal of Applied Physiology 82, Steinacker, J.M., Kellmann, M., Böhm, training and overtraining in female 1760–1764. B.O. et al. (1999) Clinical findings and athletes. Clinical Journal of Sports Schilling, B.K., Fry, A.C., Chiu, L.Z.F. parameters of stress and regeneration in Medicine 8(3), 178–186. et al. (2002) Muscle and performance rowers before world championships. In: Uusitalo, A.L., Valkonen-Korhonen, M., adaptations to high-load resistance Overload, Performance Incompetence, and Helenius, P. et al. (2004) Abnormal exercise overtraining [abstract]. NSCA Regeneration in Sport (Lehmann, M., serotonin reuptake in an overtrained, National Conference, Las Vegas, NV. Steinacker, J.M., Gastmann, U. & Keizer, insomnic and depressed team athlete. Schmidt, W., Dore, S. Hilgendorf, A. H., eds.). Kluwer Academic/Plenum International Journal of Sports Medicine et al. (1995) Effects of exercise during Publishers, New York: 71–80. 25(2), 150–153. normoxia and hypoxia on the growth Steinacker, J.M., Lormes, W., Kellmann, M. Volek, J.S., Ratamass, N.A., Rubin, M.R. hormone–insulin-like growth factor I et al. (2000) Training of junior rowers et al. (2004) The effects of creatine axis. European Journal of Applied before World Championships. Effects on supplementation on muscular Physiology 71, 424–430. performance, mood state and selected performance and body composition Schwartz, M.W., Seeley, R.J., Campfield, hormonal and metabolic responses. responses to short-term resistance L.A., Burn, P. & Baskin, D.G. (1996) Journal of Sports Medicine and Physical training overreaching. European Identification of targets of leptin action Fitness 40, 327–335. Journal of Applied Physiology 91, 628–637. in rat hypothalamus. Journal of Clinical Steinacker, J.M., Lormes, W., Reissnecker, Weir, J.P., Fry, A.C., Schilling, B.K. et al. Investigation 98, 1101–1106. S. & Liu, Y. (2004) New aspects of the (2002) Heart rate variability is not endocrinology of overtraining 599

affected by high intensity resistance Experimental Therapeutics 249(3), Wittert, G., Livesey, J., Espiner, E. & exercise overtraining [abstract]. NSCA 701–707. Donald, R. (1996) Adaptation of the National Conference, Las Vegas, NV. Wilckens, T., Schweiger, U. & Pirke, K. hypothalamopituitary adrenal axis to

Weizman, R., Weizman, A., Kook, K. et al. (1992) Activation of 5-HT1C-receptors chronic exercise stress in humans. (1989) Repeated swim stress alters brain suppresses excessive wheel running Medicine and Science in Sports and Exercise benzodiazepine receptors measured in induced by semi-starvation in the rat. 28(8), 1015–1019. vivo. Journal of Pharmacology and Psychopharmacology 109, 77–84. Chapter 38

Endocrinology of Sport Competition

JAY R. HOFFMAN

The primary role of hormones is to maintain inter- ways controlling hormonal secretion patterns are nal equilibrium by interacting with various tissues similar for both types of stresses. The major differ- and organs in the body to combat a variety of ence is that the psychological stress is generally stresses. These stresses can either be physiological associated with the classic fight or flight response or psychological in nature, often stimulating a in which strong emotional components appear to change in the secretory pattern of various hormones have a key impact on the hypothalamic–pituitary– depending upon the specific stress encountered. adrenal axis. Assuming that an athlete is in a state of Exercise is a potent stimulus to the endocrine sys- homeostasis prior to competition, changes in hor- tem. The hormonal response to both acute exercise monal concentrations are likely a function of pre- and prolonged training emphasizes the important competition anxiety. Anxiety is a common emotional role that the endocrine system has in responding to expression of stress in sport and can influence the metabolic demands of exercise and the recovery athletic performance (Martens et al. 1990). Several mechanisms involved in tissue repair and remodel- studies have reported on the relationship between ing. Although a large body of literature exists con- the hormonal response and anxiety (Filaire et al. cerning the hormonal response to various exercise 2001b; Salvador et al. 2003), and the hormonal stresses, relatively little information is available response and behavior (Salvador et al. 1999). In concerning specific hormonal responses to acute addition, the athlete’s mood has also been shown sport competition and during a season of sport com- to play an important role in the endocrine response petition. Despite the ability to simulate physiolo- seen prior to exercise. gical stresses that may occur during competition Competition is thought to cause both anxiety and in a laboratory setting, differences in stress levels arousal due to the uncertainty associated with the between a laboratory measure of performance com- outcome and subsequent effects (Sapolsky 1994). As pared to actual competition are large, and may a result, this anxiety–arousal situation is perceived potentially have a much greater influence on the as a threat and causes an anticipatory rise in several hormonal response. Therefore, the purpose of this different hormones. Elevations in cortisol, testo- chapter is to discuss the endocrine response to acute sterone and catecholamines have been reported sport competition and during a season of athletic prior to the onset of the competitive activity. While competition. cortisol elevations prior to exercise have been con- sistently reported in the literature, the response of Precompetition hormonal response testosterone has been less reliable. Studies that have examined the anticipatory catecholamine response Changes to circulating concentrations of hormones to sporting events are less available, and are often are generally seen in response to either a physiolo- inferred by research examining different modes of gical or psychological stress. The mechanistic path- exercise. 600 endocrinology of sport competition 601

Anticipatory elevations in cortisol have been seen exercise intensities. This same pattern was not seen in a number of studies examining primarily com- in norepinephrine, suggesting that the mechanism bative sports (e.g. wrestling and judo) (Elias 1981; mediating the anticipatory catecholamine response Passelergue & Lac 1999; Salvador et al. 1999, 2003; is related to adrenal medulla activation where Filaire et al. 2001b). Precompetition cortisol concen- epinephrine is released in much greater quantities. trations have been reported to be associated with Although a similar exercise protocol failed to elicit winning (Elias 1981) and positively correlated to any anticipatory response from cortisol (Kraemer both self-confidence (r = 0.64) (Salvador et al. 2003) et al. 1989), the mechanistic pathway resulting in and anxiety (r ranging from 0.62 to 0.90) (Filaire et al. epinephrine release from the adrenal medulla may 2001b). An optimal level of anxiety is thought to also be responsible for the anticipatory rise in cor- be necessary for athletes to maximize performance tisol that was previously discussed. (Gould & Krame 1992). The anticipatory rise in The potential benefit of an elevated sympatho- cortisol likely reflects a psychological mechanism adrenal response prior to a short duration, max- used by athletes to increase precompetition arousal imal intensity sporting event is likely related to an and as part of a coping mechanism to combat the improved contractility of skeletal muscle. This may stress of impending competition. allow for greater power and force outputs during The role of an anticipatory rise in testosterone competition. In addition, the anticipatory catechol- is not well understood. Several investigators have amine response also appears to be a phenomenon suggested that elevations in testosterone are re- of training experience. Elevations in both epineph- lated to aggressive behavior (Hannon et al. 1991; rine and norepinephrine concentrations have been Bjorkqvist et al. 1994; Salvador et al. 1999). This reported prior to exercise in trained power lifters, would be beneficial for sports that have are physical but not in aged-matched untrained men (Kraemer in nature with a strong aggressive component. et al. 1999). It is likely that the elevation in cate- Salvador et al. (1999) reported a positive relation- cholamines is part of the psychological preparation ship between testosterone concentrations and the (i.e. ‘psyching-up’) that experienced athletes use number of attacks during judo competition. In addi- prior to competition. tion, precompetition testosterone concentrations have also been correlated to the mood state vigor = Endocrine response to a sports (Tanaka et al. 1989) and negatively correlated (r competition –0.67) to fatigue (Salvador et al. 2003). However, precompetition elevations in testosterone have not The endocrine response during competition likely been consistently seen as several studies examining plays an important role in substrate mobilization, athletes performing contact sports have failed to cardiovascular regulation, blood flow and fluid bal- support these findings (Gonzalez-Bono et al. 1999; ance adjustments, muscle (both cardiac and skel- Filaire et al. 2001b; Hoffman 2002; Salvador et al. etal) contractility and augmentation of the secretion 2003). rates of other hormones. Due to the logistical prob- Studies examining the catecholamine response lems associated with blood draws before, during to competition are limited. In a study examining and after competition, the volume of research that elite tennis players prior to Davis Cup competi- has examined the endocrine response to actual sport tion, significantly higher (three to fourfold) epine- competition is limited. Most of our understand- phrine but not norepinephrine concentrations were ing is inferred by laboratory measures that attempt seen prior to competition compared to practice con- to simulate actual performance, but as discussed ditions (Ferrauti et al. 2001). Kraemer et al. (1991) in earlier the psychological component to actual com- a laboratory-controlled examination clearly showed petition may cause a different response pattern. In that as subjects prepared to perform exercise at a addition, differences in both the physiological and maximal intensity an anticipatory elevation in epine- endocrine responses between trained competit- phrine was seen compared to all other submaximal ive athletes and non-competitive individuals may 602 chapter 38

be large, limiting our complete understanding of important in stimulating hepatic gluconeogenesis the endocrine response occurring during actual (Naveri et al. 1985). competition. Changes in catecholamine concentrations are also sensitive to blood glucose levels. Elevations in catecholamine concentrations have both a direct Endocrine changes during endurance and indirect role in increasing substrate mobiliza- competitions tion. During endurance exercise catecholamines Endurance events such as a marathon are of also appear to respond in a differential manner. sufficient duration to cause significant strain on Increases in norepinephrine concentrations occur metabolic systems. Often, both muscle and liver within 15 min of exercise without any change in glycogen levels are exhausted, limiting the ability epinephrine. Elevated norepinephrine concentra- to provide fuel to exercising muscle. In addition, tions are thought to stimulate hepatic gluconeogen- potential problems of exercising in the heat and esis by inhibiting insulin secretion (Glaster et al. dehydration may also cause additional strain on 1981). As exercise duration continues a greater need the athlete, causing significant changes to various for substrate mobilization is recognized by meta- hormones attempting to maintain thermoregulation boreceptors stimulating an increase in epinephrine and fluid balance. secretion. The elevation in epinephrine concentra- Endurance athletes generally compete at an exer- tions directly act to promote lipolysis (Smith 1980). cise intensity that does not exceed their anaerobic In addition, both growth hormone and cortisol also threshold. In well-conditioned endurance athletes have important roles in promoting lipolytic activity this occurs between 80–90% of their maximal aero- and increasing plasma free fatty acid concentrations bic capacity (Hoffman 2002). During prolonged during prolonged endurance exercise (Hoffman exercise at this intensity, such as in a marathon et al. 2002). or triathlon, glycogen stores within the muscle are Endurance exercise appears to be a potent stimu- not sufficient to provide the needed fuel; thus, in lus for both cortisol and growth hormone release. addition to the body’s reliance upon adipose tissue, Both of these hormones appear to be positively the rate of hepatic glucose production will become related to the duration of exercise (Bunt et al. 1986). markedly elevated (Wahren et al. 1971). Increased The influence of exercise intensity on the secretion sympathetic nervous system activity (elevations in patterns of these hormones though is less clear. It plasma catecholamine concentrations) and changes appears that exercise needs to be performed at a in glucagon and insulin concentrations are the prim- minimal intensity to elicit increases in the circulat- ary factors regulating hepatic glucose production ing concentrations of these hormones (Few 1974). and lipid metabolism. In addition, growth hormone However, most endurance events are performed at is also involved in lipolytic activity during pro- an intensity that appears to be sufficient to elicit longed endurance exercise. elevation in both cortisol and growth hormone con- Most of the knowledge attained of metabolic and centrations. Figure 38.1 provides a schematic figure endocrine responses during prolonged endurance outlining the proposed endocrine responses to an exercise has been achieved under laboratory set- endurance event such as a marathon. tings. In general, during prolonged exercise insulin Only a limited number of studies have examined concentrations will decline, while glucagon con- the endocrine response during endurance events centrations become elevated. Insulin depression such as marathons and triathlons. In these events appears to be a function of the duration of exercise. a 170% elevation in catecholamine concentrations As exercise duration increases a greater decrease in (Dearman & Francis 1983) and a 2–4.5-fold increase insulin concentrations is seen (Galbo 1981). Although in cortisol concentrations (Dearman & Francis 1983; elevated glucagon levels will increase hepatic gluc- Hale et al. 1983; Urhaussen & Kindermann 1987; ose production, the change in the molar ratio of Ponjee et al. 1994) have been reported. Although the glucagon to insulin has also been suggested to be as results of these studies clearly support the role of endocrinology of sport competition 603

Anterior pituitary

Liver Glycogen Growth hormone

Glucose Free fatty acids Lipid stores

Glucose Blood vessel

Glucagon Norepinephrine

Pancreas Cortisol Epinephrine Insulin

Adrenal Fig. 38.1 Endocrine response to an gland endurance event. these hormones in substrate mobilization during 1970) and creatine phosphate (Sutton et al. 1973). In prolonged competitive endurance events, other addition, the elevated cortisol levels seen in con- studies have also shown that these athletic en- junction with lower testosterone concentrations is deavors have a significant effect on the pituitary– indicative of greater catabolic activity and may limit testicular axis. During events that last less than 3 h the ability of skeletal muscle to repair damage no changes in circulating testosterone levels are associated with the endurance event. Interestingly, seen; however, increases in sex hormone binding during ultraendurance events such as an 1100-km globulin (SHBG) were seen, resulting in a decrease run testosterone levels will remain depressed for the in the testosterone/SHBG ratio and a decrease in the entire 20-day event (Schurmeyer et al. 1984). This is free, biologically active testosterone (Urhaussen & in contrast to other stress hormones measured (i.e. Kindermann 1987). As the endurance event exceeds thyroxine, cortisol and prolactin) that showed an 3–4 h, significant decreases in testosterone are seen ability to adapt to the prolonged exercise stress. (Tanaka et al. 1986; Lac & Berthon 2000), and may remain below baseline levels for up to 48 h post- Endocrine changes during strength/power competition (Tanaka et al. 1986). competitions The significance of this apparent anabolic deficit may be more important during the recovery period During competitions that emphasize strength and following competition than during the competition power (including sprint races) the demand on the itself. Testosterone has been reported to influence metabolic energy system to fuel exercise is generally the resynthesis of glycogen (Gillespie & Edgerton not a limiting factor. However, in competitions that 604 chapter 38

Epinephrine ACTH Cortisol Testosterone β-endorphin

Fig. 38.2 Proposed endocrine responses to a high intensity, Anticipatory effects Post-competition effects maximal effort. ACTH, (recovery period) adrenocorticotropic hormone.

rely on repetitive high intensity activity the hor- has been suggested in several other studies examin- monal responses and the stimuli that regulate their ing elite weightlifters (Kraemer 1992; Fry et al. 1993; secretion may differ compared to responses observed Kraemer et al. 1993). The role of testosterone in max- during a single event (i.e. 100-m sprint). During imal strength expression may be related to its inter- such competitions the endocrine response may action with the nervous system and enhancement of reflect both psychological preparation to the com- the secretion rates of growth hormone and growth petition and the physiological demands that occur factors (Kraemer 1992). However, these latter inter- during the competition. Figure 38.2 provides a actions may have more importance during the pre- schematic example of potential endocrine responses paration and recovery phases of competition, than to a high intensity competition involving a single having any acute effects during competition. maximal effort. During 100- and 400-m sprint competition, eleva- In athletes that participate in throwing events tions in both pituitary and adrenal hormones sug- such as the discus or hammer throw, anticipatory gest that the primary stimulus stimulating secretion elevations to pituitary and adrenal hormones (β- of these hormones is the psychological prepara- endorphin, adrenocorticotropic hormone [ACTH] tion preceding competition. During a 100-m sprint and cortisol) have been reported, with no other significant increases in β-endorphin, ACTH and hormonal changes seen post-competition (Petraglia cortisol concentrations have been reported prior to et al. 1988). This is not surprising considering that and following the race (Petraglia et al. 1988). While the duration of the event is comprised of only a increases in ACTH and cortisol are part of the anti- few seconds. During these maximal effort, short cipatory response to the event, the increase in β- duration competitions the anticipatory hormonal endorphin concentration could be associated with response may be critical for optimal performance. decreased pain sensitivity. This could explain how As discussed earlier, increases in catecholamine athletes despite suffering from muscle soreness or concentrations are important in eliciting maximal injury might still perform at optimal levels. strength performance, and changes in circulating During a longer duration sprint (400-m sprint) epinephrine have been correlated to short-term significant elevations in luteinizing hormone (LH) alterations in strength (Fry et al. 1994). In addition, and follicle-stimulating hormone (FSH), with sig- elevated sympathetic activity also appears to nificant decreases in total and free testosterone, have augment testosterone secretion (Jezova & Vigas been observed (Slowinska-Lisowka & Majda 2002). 1981). This may have certain significance consider- Following 24 h of rest, testosterone concentrations ing that recent research has shown that the testo- returned to baseline levels. Considering the short sterone/cortisol ratio (T/C ratio) is highly correlated duration of the event (45 s in these athletes) the (r = 0.92) to weightlifting performance in elite junior depressed testosterone concentrations seen follow- weightlifters (Fry et al. 2000). Although these invest- ing the race may reflect interactions associated with igators were careful not to imply a causal relation- the anticipatory hormonal response. As discussed ship based upon their correlational analyses, the earlier, both cortisol and testosterone concentra- importance of testosterone to strength performance tions may be elevated prior to a competitive event endocrinology of sport competition 605

involving maximal intensity exercise. However, the endocrine response prior to the first match. How- pre-event testosterone response has not been as ever, catecholamine levels were elevated at the end consistent as that seen in other hormones. Since of each match and corresponded to an improved cortisol has been shown to display inhibitory actions reaction time. The higher catecholamine response on testosterone secretion (Cumming et al. 1983), it is seen at the end of each match may reflect a greater possible that depressed testosterone concentrations arousal that contributed to the improved perform- observed immediately after the race reflect hor- ance. Interestingly, norepinephrine responded in monal interactions occurring prior to the start of a more sensitive manner than epinephrine during competition. Unfortunately, cortisol was not meas- the tournament. Norepinephrine was significantly ured in this study making this only a speculative elevated following each match, but epinephrine assumption. Regardless, the return of testosterone responses were not as consistent. The results were (total and free) to baseline levels was suggestive of a suggestive of a possible adrenal insufficiency that normal functioning recovery system. may have also contributed to a reduced force pro- Due to logistical difficulties there have been only duction observed during the study. a handful of investigations examining the hormonal The testosterone and cortisol responses reported response during prolonged high intensity sporting by Kraemer et al. (2001) contrasted from those events such as wrestling tournaments or American observed by Passelergue and Lac (1999). However, football games. During a wrestling match in which it should be acknowledged that the endocrine ana- wrestlers are engaged in combative competition for lysis in the former study was from plasma samples, up to 7 min, under conditions that may include fluid while in the latter study from saliva samples. and dietary restrictions, both psychological and Variability in hormonal measurements from differ- physiological stresses are apparent. Passelergue ent bio-compartments may have contributed to the and Lac (1999) observed an anticipatory rise in differences in the endocrine response seen between cortisol prior to each match during the 2-day study. these studies. Kraemer et al. (2001) observed a Cortisol concentrations continued to elevate (2.5- consistent reduction in testosterone concentrations fold from resting levels) throughout the competi- during the 2-day tournament with significant eleva- tion, but fell rapidly to baseline levels (within 1.5 h) tions in cortisol seen only during matches 1, 2 and 5. during the recovery period. Testosterone concen- Resting testosterone concentrations by the end of tration remained at basal levels during the com- the tournament fell below normal, and was similar petition, but began to significantly elevate during to levels seen in endurance runners performing the post-competitive period. During the recovery high volume training (Wheeler et al. 1984) and in period subjective feelings of fatigue, and an inability weightlifters participating in a week-long overtrain- to perform strenuous exercise was reported. How- ing protocol (Fry et al. 1993). The low testosterone ever, during this time a significantly higher T/C concentrations seen may have also contributed to ratio was observed, and remained elevated for 5 the reduced strength capability of the subjects by days following the tournament. This anabolic phase delaying recovery between matches. would likely enhance the recovery processes of the In high intensity, combative sports that are of athletes allowing for a faster return to both training longer duration (e.g. 60 min) such as American and competition. football the potential for greater performance de- Kraemer et al. (2001) examined wrestlers during crements, tissue damage and stress may exist. Only 5 matches over a 2-day tournament. They reported one study has examined the endocrine and per- that despite a 1-week weight loss period that formance responses to participation in a compet- resulted in a 6% loss of body mass by the athletes, no itive American football game (Hoffman et al. 2002). significant changes from baseline hormonal concen- Cortisol concentrations were higher prior to the trations (epinephrine, norepinephrine, testosterone game than baseline levels measured the day before. and cortisol) were observed prior to the first match. However, this may have been a function of normal The investigators did not observe any anticipatory diurnal changes. Regardless, the 52% increase (p > 606 chapter 38

0.05) in cortisol concentrations in starters between In the majority of studies examining high intensity baseline and pregame levels may also suggest a competitive sport activity the primary hormones potential anticipatory response. During the game examined have been cortisol and testosterone. The cortisol levels tended to return to baseline levels, but response of cortisol to a game has generally been were still significantly higher than red-shirt fresh- seen to increase, while the response of testosterone man players who did not play and served as a con- has been less consistent. The physiological roles of trol group. Since glucose levels were maintained cortisol (stimulation of gluconeogenesis, increase in during the game for both the players and the control blood glucose concentrations, facilitation of lipo- group, the significant difference seen in cortisol lysis and anti-inflammatory affects) do support likely represented a greater stress experienced by the findings in these studies. On the other hand, the the players participating in the game. Considering physiological roles of testosterone are primarily the significant elevations seen in muscle damage involved in mechanisms involved in recovery. markers seen as a result of game participation, the Thus, making the measurement of testosterone elevated cortisol concentrations appear to reflect the during the recovery period has greater importance anti-inflammatory properties that cortisol possesses in the hours and days following the competitive in response to tissue damage. performance. Unfortunately, often due to logistical During the American football game no changes in problems the ability to obtain samples from com- testosterone concentrations were seen from baseline petitive athletes for several days following a com- to the end of the game. However, testosterone petition is quite difficult. concentrations were at the low end of normal. Considering that the game was the last game of the Endocrine changes during moderate intensity season, it is possible that low testosterone levels may sporting events inhibit recovery mechanisms during the recovery period. However, the players were not examined During sport competitions of moderate intensity the past the post-game period and any inference to metabolic demands are quite different than those recovery can only be speculative. seen during higher intensity sports. Although glu- In high intensity sports that have limited phys- coneogenesis will still be of importance, a greater ical contact (i.e. basketball) changes in hormonal emphasis on lipid metabolism will likely be seen. In concentrations would likely reflect physiological addition, the anticipatory rise in many hormones stresses arising from psychological preparation or does not seem to be as prevalent in these sporting metabolic demands, and a reduced risk of physio- events compared to competitions involving high logical stresses caused by tissue damage. In the only intensity, maximal efforts. study known to have examined the hormonal Soccer is an interesting sport considering that it response to a basketball game, saliva samples were is 90 min in duration and involves sprints and runs collected from 16 members of two male professional of various intensities and durations. In a limited European basketball teams (eight from each team) number of studies, the pituitary–adrenal axis competing against one another 45 min before and appears to be quite active during the course of a 15 min after a game (Gonzalez-Bono et al. 1999). No game. Significant elevations in ACTH, cortisol, change in testosterone was seen during the contest, growth hormone and prolactin have been reported while significant cortisol elevations were noted. The (Lupo et al. 1985; Carli et al. 1986). Growth hormone investigators in this study were primarily compar- and prolactin were seen to increase only at halftime, ing the endocrine response to mood states in both while ACTH and cortisol continued to rise through- winners and losers from the game. The increased out the game. Cortisol levels returned to baseline cortisol concentrations seen at the end of the game levels within 45 min following the game. These does appear to be reflective of the metabolic stress hormonal responses do appear to be reflective of the that is associated with 40-min of high intensity metabolic demands of soccer; high intensity sprints activity. interspersed with periods of lower intensity activ- endocrinology of sport competition 607

ity. However, the varying intensities of activity centrations generally reflect the state of anxiety of during the course of a game do not appear to the athlete. As previously discussed, there does provide a sufficient stimulus to maintain growth appear to be an optimal level of anxiety for max- hormone secretions. Previous research has shown imizing athletic performance, and several investig- that growth hormone responses are greater during ators have suggested that the NE/EPI ratio may continuous exercise in comparison to intermittent provide an endocrine marker of optimal anxiety exercise, even when lactate levels are greater in the (Hoch et al. 1988; Ferrauti et al. 2001). With higher latter (Lugar et al. 1992). There apparently needs to concentrations of epinephrine there exists a greater be a minimum duration (10 min) of activity above chance for hormone–receptor interaction in skeletal the lactate threshold to stimulate consistent growth muscle potentially effecting contractile functioning hormone release (Hoffman 2002). of skeletal muscle. During a sport that requires max- Tennis is another sport that has an intermittent imal effort, augmentation of sympathetic activity style of play involving periods of high intensity act- appears to be beneficial. However, in a sport that ivity interspersed with numerous recovery periods. requires fine motor control, elevated sympathetic In contrast to soccer though, the high intensity activ- activity may result in a loss of accuracy and move- ity seen in tennis is played over a much smaller area. ment economy resulting in performance impair- As a result it appears that the metabolic demands ment. The loss of co-ordination may further increase of tennis players may be less than that seen in soccer anxiety levels by the added stress of poor perform- players. In a study on a simulated singles match ance. When a physiological stress stimulates cate- (85 min in duration) in NCAA Division I tennis cholamine release, the higher norepinephrine levels players, the results observed appeared to support appear to compensate for the mental anxiety by this (Bergeron et al. 1991). Blood glucose and plasma reducing the NE/EPI ratio. The lower NE/EPI ratios lactate concentrations were observed to remain at appear to correspond to a calmer athlete, and the basal levels, while cortisol concentrations decreased resulting catecholamine concentration reflects the and were significantly below baseline levels by the physiological stress of exercise and not the mental end of the match. At the end of the match testo- anxiety produced by competition (Ferrauti et al. 2001). sterone concentrations were significantly elevated from baseline measures. The physiological role of Endocrine response to a competitive testosterone during a tennis match however is not sport season clear, but it likely has an important role during recovery. There have been a number of studies that have In a separate study on tennis players, the import- examined the endocrine response to a competitive ance of catecholamine concentrations was exam- sports season. The primary purpose motivating ined in a match during the Davis Cup competition many of these studies was to provide a better (Ferrauti et al. 2001). A twofold increase was seen in understanding of hormonal changes to variations in both epinephrine and norepinephrine concentra- training intensity and volume that occur during the tions. Although data is limited, catecholamines may course of a competitive season, and to provide play an important role in fulfilling the metabolic insight to potential endocrine markers to overreach- demands of tennis. ing or overtraining. Overtraining is a general term The importance of catecholamines to perform- for describing an imbalance between training and ance may also be reflected by the ratio of nore- recovery (Kuipers & Keizer 1988). Most studies pinephrine to epinephrine (NE/EPI ratio). It has have used testosterone and cortisol, hormonal indic- been suggested that the NE/EPI ratio provides ators of anabolism and catabolism, respectively, as important information concerning the athlete’s abil- the primary endocrine measurement of training ity to perform, especially in sports that require fine stress. In this section, the hormonal response to a motor control for precise movements (Hoch et al. season of competition will be discussed as it relates 1988; Ferrauti et al. 2001). Elevated epinephrine con- to endurance and strength/power sports. 608 chapter 38

(range 22–36 years). It has been suggested that as Endocrine response to a season of competition in endurance athletes continue to train testosterone the endurance athlete concentrations are reduced as a result of hypothala- In most endurance sports, as the season progresses mic dysfunction (Barron et al. 1985; MacConnie et al. both the volume and intensity of training will elev- 1986), or as a result of elevated cortisol levels that ate (Hoffman 2002). This training progression may act to suppress testosterone secretions (Cumming be reflected by changes in circulating hormonal con- et al. 1983). In these experienced endurance-trained centrations. In some athletes the changes are drastic athletes, relatively short periods of reduced training and may reflect insufficient recovery and a possible (3–4 weeks) may not be sufficient to restore normal overtraining syndrome, while other athletes per- testicular function. forming the identical training program may have The studies reporting on endocrine changes little to no changes in endocrine markers of recovery, during a competitive season generally report on signifying adequate recovery from training. During changes occurring to the entire team. However, the course of a season the athlete will be competing individual athletes performing the same training in various competitions; however, the greatest program often respond differently. Hooper et al. emphasis is placed on the competitions that occur (1993) examining 14 male and female swimmers at the end of the season. Generally, the training reported no changes in any of the endocrine meas- program will be adjusted to help the athlete peak ures performed during the course of the competitive for these particular matches. This is referred to as season. When subjects were examined on an indi- a taper and involves a reduction in both training vidual basis three of the swimmers (all female) were volume and intensity, with a heavier emphasis on identified as being stale. These athletes were also volume reduction (Hoffman 2002). reported to have significantly higher norepine- Most studies examining endurance athletes have phrine concentrations during the taper period than reported no changes in resting testosterone, growth the athletes that were not suffering from any over- hormone, norepinephrine, ACTH and cortisol training syndrome. This highlights the importance during a season of competition (Vervoorn et al. 1991; of examining individual data. Further emphasizing Hooper et al. 1993; Lopez-Calbert et al. 1993; Flynn this point, other investigators have also reported on et al. 1994; Maresh et al. 1994; Mujicka et al. 1996; a tendency (p > 0.05) for depressed testosterone and Bonifazi et al. 1998). In contrast, other studies elevated cortisol levels (Lopez-Calbert et al. 1993). examining endurance athletes reported significant Though no significant changes were noted, five of decreases in testosterone concentrations following seven athletes studied had reduced testosterone periods of high volume training (Flynn et al. 1994; concentrations and a 29% reduction (p > 0.05) in the Tyndall et al. 1996), while studies primarily on T/C ratio was seen. marathon runners have often reported chronically A reduction in testosterone concentrations, eleva- low testosterone concentrations (Barron et al. 1985; tion in cortisol concentrations and reduction in the Houmard et al. 1990). Although the depressed tes- T/C ratio are all potential indicators of a catabolic tosterone concentrations in swimmers do appear state and provides information concerning the to return to baseline levels during a taper, this does recovery ability of the athlete, and the athlete’s abil- not appear to be the case for marathon runners ity to synthesize protein and to maintain muscle (Houmard et al. 1990). It is likely that prolonged mass. Changes in testosterone have been shown to high volume training and longer training experi- significantly correlate (r = 0.87) to changes in lean ence seen in these athletes contributes to a chronic body mass (Lopez-Calbert et al. 1993), and increases reduction in testosterone. In collegiate cross-country in the T/C ratio are also significantly correlated runners the training volume progressed incre- (r = 0.86) to improvements in swimming perform- mentally, and the athletes were relatively younger ance (Mujicka et al. 1996). (range 18–22 years) than the marathon runners A competitive endurance sport season is quite endocrinology of sport competition 609

stressing, both on the physiological mechanisms to that did not correspond to any negative changes in maintain homeostasis and on the psychological soccer performance (Filaire et al. 2001a). Studies on effects of training and competition. In an optimal swimmers involved in sprint events also reported situation the athlete is able to maintain normal no relationship between changes in cortisol, dehy- physiological function reflected by little to no droepiandrostenedione and performance (Chatard changes in the endocrine profile. Bonifazi et al. et al. 2002). In addition, unpublished data from this (2000) showed that following 18-weeks of training author’s laboratory on American college football in elite swimmers, precompetition cortisol levels players have also shown that testosterone and were reduced and corresponded to improvements cortisol levels are maintained during a season of in endurance swimming performance. In addition, competition. In contrast, Banfi et al. (1993) exam- 18-weeks of swim training has also been seen to ining speed skaters have suggested that the free result in elevated growth hormone concentrations testosterone/cortisol ratio may be a useful indicator (Bonifazi et al. 1998). This is a positive training adap- of temporary, incomplete recovery from training. tation considering that a greater growth hormone This is supported by other investigators that have concentration may reflect an improved ability to shown that the anabolic to catabolic hormonal ratio maintain muscle mass, enhance fatty acid utiliza- is a potential hormonal marker for incomplete tion and reduce utilization of glucose and amino recovery and overtraining syndrome (Adlercreutz acids (Rogol 1989). The psychological effect of a et al. 1986). The lack of consistency seen in the competitive sports season was also reflected by literature suggests that hormonal changes during a changes in epinephrine concentrations in competit- season may not be a reliable indicator of perform- ive swimmers (Hooper et al. 1993). A significant ance changes, but that such markers may have more depression in epinephrine concentrations was seen value when evaluating short duration training within 3 days of the conclusion of the competitive periods. season. This likely reflects the reduced emotional Wrestling has been shown to have a potent effect stress and anxiety from the end of the competitive on endocrine function during a competitive season season. (Strauss et al. 1985; Roemmich & Sinning 1997). However, in contrast to other strength/power athletes, wrestlers also participate in periods of Endocrine response to a competitive season in fasting and dehydration in order to make weight, the strength/power athlete which compounds the physiological stress of train- The number of studies that have examined ing. Decreases in body fat and body mass have been endocrine changes during a season in traditional shown to be significantly correlated (r = 0.72) to strength/power sports (i.e. basketball, American reduced testosterone concentrations (Strauss et al. football, sprinting, wrestling) are relatively few in 1985). Roemmich and Sinning (1997) have also comparison to endurance sports. Furthermore, no shown that dietary restriction and wrestling prac- published studies are available that have examined tice and competition produces a partial growth seasonal hormonal changes in popular American hormone resistance (elevated growth hormone team sports such as football or basketball. Similar to concentration with a decrease in growth hormone the research on endurance athletes, most studies binding protein) and a possible disruption of the examining the strength/power athlete have primar- pituitary–testicular axis (reflected by decreases in ily measured testosterone and cortisol concentra- testosterone and free testosterone concentrations). tions to provide information on the hormonal stress Interestingly, low testosterone concentrations were response and recovery capability in these athletes. not reported to be associated with any changes in Soccer players, although not a typical strength/ LH concentrations suggesting a possible impair- power athlete, have been reported to have a reduced ment of the hypothalamic–pituitary–gonadal axis. T/C ratio during the course of a competitive season However, the reduced testosterone concentrations 610 chapter 38 in these athletes were still within normal physiolo- Conclusion gical range. A change in the circulating concentration of The endocrine response to acute sport competition anabolic hormones is a positive adaptation for the was shown to reflect precompetition arousal and strength/power athlete. The primary reason being anxiety, the metabolic demands of the sport, the the role that testosterone and growth hormone has physical stress associated with competition and the in enhancing or facilitating the building of lean recovery capability of the athlete. The endocrine tissue (Hickson et al. 1994). Surprisingly, the ability response to competition appears to be dependent of these athletes to alter endocrine function and upon the nature of the sport. During competition the effects that hormonal changes have on strength requiring maximal effort over a short duration, and power performance is unclear. Experienced, the anticipatory hormonal response may have an competitive weightlifters have been shown to main- important role in strength and power expression, tain testosterone concentrations during a year of while during endurance competition the hormonal training, while strength levels were improved. response generally reflects the metabolic demands (Häkkinen et al. 1987). Only after 2 years of training of prolonged activity. In the post-competition re- did significant elevations in testosterone concentra- covery phase, the hormonal profile may provide tions become apparent in these experienced weight- an indication for the athlete’s ability to recover. lifters (Häkkinen et al. 1988). Alterations in the Analyses of the endocrine changes during a season endocrine profile of these athletes appears to be a of competition have primarily examined the ath- difficult to accomplish and may reflect an advanced lete’s ability to adapt to various training stresses, adaptive strategy to increase force production and to serve as an endocrine marker of the over- (Kraemer 1992). training syndrome.

References

Adlercreutz, H., Harkonen, M., exercise of adrenocorticotropin and Dearman, J. & Francis, K.T. (1983) Plasma Kuoppasalmi, K. et al. (1986) Effect growth hormone plasma concentrations levels of catecholamines, cortisol, and of training on plasma anabolic and in human swimmers. European Journal of β-endorphins in male athletes after catabolic steroid hormones and their Applied Physiology 78, 394–397. running 26.2, 6, and 2 miles. Journal of response during physical exercise. Bonifazi, M., Sardella, F. & Lupo, C. (2000) Sports Medicine 23, 30–38. International Journal of Sports Medicine 7 Preparatory versus main competitions: Elias, M. (1981) Serum cortisol, (suppl.), 227–228. differences in performances, lactate testosterone and testosterone-binding Banfi, G., Marinelli, M., Roi, G.S. & responses and pre-competition plasma globulin responses to competitive Agape, V. (1993) Usefulness of free cortisol concentrations in elite male fighting in human males. Aggressive testosterone/cortisol ratio during a swimmers. European Journal of Applied Behavior 7, 215–224. season of elite speed skating athletes. Physiology 82, 368–373. Ferrauti, A., Neumann, G., Weber, K. International Journal of Sports Medicine 14, Bunt, J.C., Boileau, R.A., Bahr, J.M. & & Keul, J. (2001) Urine catecholamine 373–379. Nelson, R.A. (1986) Sex and training concentrations and psychophysical Barron, J.L., Noakes, T.D., Levy, W., Smith, differences in human growth hormone stress in elite tennis under practice and C. & Millar, R.P. (1985) Hypothalamic during prolonged exercise. Journal of tournament conditions. Journal of Sports dysfunction in overtrained athletes. Applied Physiology 61, 1796–1801. Medicine and Physical Fitness 41, Journal of Endocrinology and Metabolism Carli, G., Bonifazi, M., Lodi, L. et al. (1986) 269–274. 60, 803–806. Hormonal and metabolic effects Few, J.D. (1974) Effect of exercise on the Bergeron, M.F., Maresh, C.M., Kraemer, following a football match. International secretion and metabolism of cortisol in W.J. et al. (1991) Tennis: a physiological Journal of Sports Medicine 7, 36–38. man. Journal of Endocrinology 62, 341–353. profile during match play. International Chatard, J.C., Atlaoui, D., Lac, G. et al. Filaire, E., Bernain, X., Sagnol, M. & Lac, G. Journal of Sports Medicine 12, 474–479. (2002) Cortisol, DHEA, performance and (2001a) Preliminary results on mood Bjorkqvist, K., Nygren, T., Bjorklund, A.C. training in elite swimmers. International state, salivary testosterone : cortisol ratio & Bjorkqvist, S.E. (1994) Testosterone Journal of Sports Medicine 23, 510–515. and team performance in a professional intake and aggressiveness: real effect or Cumming, D.C., Quigley, M.E. & Yen, soccer team. European Journal of Applied anticipation? Aggressive Behavior 20, S.S.C. (1983) Acute suppression of Physiology 86, 179–184. 17–26. circulating testosterone levels by cortisol Filaire, E., Sagnol, M., Ferrand, C., Maso, F. Bonifazi, M., Bela, E., Lupo, C. et al. (1998) in men. Journal of Clinical Endocrinology & Lac, G. (2001b) Psychophysiological Influence of training on the response to and Metabolism 57, 671–673. stress in judo athletes during endocrinology of sport competition 611

competition. Journal of Sports Medicine immediate effect on female runners’ Kraemer, W.J., Fry, A.C., Warren, B.J. et al. and Physical Fitness 41, 263–268. luteinizing hormone, follicle-stimulating (1993) Acute hormonal responses in elite Flynn, M.G., Pizza, R.X., Boone, B.J. et al. hormone, prolactin, testosterone and junior weightlifters. International Journal (1994) Indices of training stress during cortisol levels. American Journal of of Sports Medicine 13, 103–109. competitive running and swimming Obstetrics and Gynecology 146, 550–556. Kraemer, W.J., Fleck, S.J., Maresh, C.M. seasons. International Journal of Sports Hannon, C.J., Friedl, K.E., Zold, A., et al. (1999) Acute hormonal responses to Medicine 15, 21–26. Kettler, T.M. & Plymate, S.R. (1991) a single bout of heavy resistance exercise Fry, A.C., Kraemer, W.J., Stone, M.H. et al. Psychological and serum in trained power lifters and untrained (1993) Endocrine and performance homovanillic acid changes in men men. Canadian Journal of Applied responses to high-volume training and administered androgenic steroids. Physiology 24, 524–537. amino acid supplementation in elite Psychoneuroendocrinology 16, 335–343. Kraemer, W.J., Fry, A.C., Rubin, M.R. et al. junior weightlifters. International Journal Hickson, R.C., Hidaka, K., Foster, C., (2001) Physiological and performance of Sport Nutrition 3, 306–322. Falduto, M.T. & Chatterton, R.T., Jr. responses to tournament wrestling. Fry, A.C., Kraemer, W.J., van Borselen, F. (1994) Successive time course of strength Medicine and Science in Sport and Exercise et al. (1994) Catecholamine responses to development and steroid hormone 33, 1367–1378. short-term intensity specific resistance responses to heavy resistance training. Kuipers, H. & Keizer, H.A. (1988) exercise overtraining. Journal of Applied Journal of Applied Physiology 76, 663–670. Overtraining in elite athletes. Sports Physiology 77, 941–946. Hoch, F., Werle, E. & Weicker, H. (1988) Medicine 6, 79–92. Fry, A.C., Kraemer, W.J., Stone, M.H. et al. Sympathoadrenergic regulation in elite Lac, G. & Berthon, P. (2000) Changes in (2000) Relationships between serum fencers in training and competition. cortisol and testosterone levels and T/C testosterone, cortisol, and weightlifting International Journal of Sports Medicine 9 ratio during an endurance competition performance. Journal of Strength and (suppl. 2), 141–145. and recovery. Journal of Sports Medicine Conditioning Research 14, 338–343. Hoffman, J.R. (2002) Physiological Aspects of and Physical Fitness 40, 139–144. Galbo, H. (1981) Endocrinology and Sport Training and Performance. Human Lopez-Calbert, J.A., Navarro, M.A., metabolism in exercise. International Kinetics: Champaign, IL. Barbany, J.R. et al. (1993) Salivary steroid Journal of Sports Medicine 2, 2203–2211. Hoffman, J.R., Maresh, C.M., Newton, R.U. changes and physical performance in Gillespie, C.A. & Edgerton, V.R. (1970) The et al. (2002) Performance, biochemical, highly trained cyclists. International role of testosterone in exercise induced and endocrine changes during a Journal of Sports Medicine 14, 111–117. glycogen supercompensation. Hormones competitive football game. Medicine Lugar, A., Watschinger, B., Duester, P. and Metabolic Research 2, 364–366. and Science in Sports and Exercise 34, et al. (1992) Plasma growth hormone and Glaster, A.D., Clutter, W.E., Cryer, P.E., 1845–1853. prolactin responses to graded levels of Collins, J.A. & Bier, M.D. (1981) Hooper, S.L., Traeger-Mackinnon, L., acute exercise and to a lactate infusion. Epinephrine plasma thresholds for Gordon, R.D. & Bachmann, A.W. (1993) Neuroendocrinology 56, 112–117. lipolytic effects in man. Journal of Clinical Hormonal responses of elite swimmers Lupo, C., Baldi, L., Bonifazi, M. et al. (1985) Investigations 67, 1729–1738. to overtraining. Medicine and Science in Androgen levels following a football Gonzalez-Bono, E., Salvador, A., Serrano, Sports and Exercise 25, 741–747. match. European Journal of Applied M.A. & Ricarte, J. (1999) Testosterone, Houmard, J.A., Costill, D.L., Mitchell, J.B. Physiology and Occupational Physiology cortisol, and mood in a sports team et al. (1990) Testosterone, cortisol, and 54, 494–496. competition. Hormones and Behavior 35, creatine kinase levels in male distance MacConnie, S.E., Barkan, A., Lampman, 55–62. runners during reduced training. R.M., Schork, M.A. & Beitins, I.Z. (1986) Gould, D. & Krame, V. (1992) The arousal– International Journal of Sports Medicine Decreased hypothalamic gonadotropin- athletic performance relationship: 11, 41–45. releasing hormone secretion in male current status and future directions. In: Jezova, D. & Vigas, M. (1981) Testosterone marathon runners. New England Journal Advances in Sport Psychology (Horn, T., response to exercise during blockade of Medicine 315, 411–417. ed.). Human Kinetics: Champaign, IL: and stimulation of adrenergic receptors Maresh, C.M., Armstrong, L.E., Bergeron, 119–141. in man. Hormone Research 15, 141–147. M.F. et al. (1994) Plasma cortisol and Häkkinen, K., Pakarinen, A., Alen, M., Kraemer, W.J. (1992) Hormonal testosterone responses during a college Kauhanen, H. & Komi, P.V. (1987) mechanisms related to the expression swim season. Journal of Strength and Relationships between training volume, of muscular strength and power. In: Conditioning Research 8, 1–4. physical performance capacity, and Strength and Power in Sport (Komi, P.V., Martens, R., Vealay, R. & Burton, D. serum hormone concentration during ed.). Blackwell Scientific, London: (1990) Competitive Anxiety in Sport. prolonged training in elite weightlifters. 64–76. Human Kinetics: Champaign, IL: International Journal of Sports Medicine 8 Kraemer, W.J., Patton, J.F., Knuttgen, H.G. 142–149. (suppl.), 61–65. et al. (1989) Hypothalamic–pituitary– Mujicka, I., Claude-Chatard, J., Padilla, S., Häkkinen, K., Pakarinen, A., Alen, M., adrenal responses to short-term high- Yannick-Guezennec, C. & Geyssant, A. Kauhanen, H. & Komi, P.V. (1988) intensity cycle exercise. Journal of Applied (1996) Hormonal responses to training Neuromuscular and hormonal Physiology 66, 161–166. and its tapering off in competitive adaptations in athletes to strength Kraemer, W.J., Patton, J.F., Knuttgen, H.G. swimmers: relationship with training in 2 years. Journal of Applied et al. (1991) Effects of high-intensity performance. European Journal of Applied Physiology 65, 2406–2412. cycle exercise on sympathoadrenal– Physiology 74, 361–366. Hale, R.W., Kosasa, T., Krieger, J. & medullary responses patterns. Journal of Naveri, H., Kuoppasalmi, K. & Harkonen, Pepper, S. (1983) A marathon: the Applied Physiology 70, 8–14. M. (1985) Metabolic and hormonal 612 chapter 38

changes in moderate and intense long- Salvador, A., Suay, F., Gonzalez-Bono, E. the pituitary–testicular axis. Journal of term running exercises. International & Serrano, M.A. (2003) Anticipatory Endocrinology Investigations 9, 97–101. Journal of Sports Medicine 6, 276–281. cortisol, testosterone and psychological Tanaka, H., Yamamoto, K., Imamura, E. Passelergue, P. & Lac, G. (1999) Saliva responses to judo competition in young et al. (1989) Relationship between cortisol, testosterone and testosterone/ men. Psychoneuroendocrinology 28, plasma testosterone levels and cortisol ratio variations during a 364–375. psychological mood states or aptitudes wrestling competition and during the Sapolsky, R.M. (1994) Why Zebras Don’t Get in sportsmen. First IOC World Congress post-competitive recovery period. Ulcers. Freeman, New York. on Sport Sciences 3, 438–439. International Journal of Sports Medicine Schurmeyer, T., Jung, K. & Nieschlag, E. Tyndall, G.L., Kobe, R.W. & Houmard, J.A. 20, 109–113. (1984) The effect of an 1100 km run (1996) Cortisol, testosterone, and insulin Petraglia, F., Barletta, C., Facchinetti, F. on testicular, adrenal and thyroid action during intense swimming et al. (1988) Response of circulating hormones. International Journal of training in humans. European Journal of adrenocorticotropin, β-endorphin, Andrology 7, 276–282. Applied Physiology and Occupational β-lipotropin and cortisol to athletic Slowinska-Lisowska, M. & Majda, J. Physiology 73, 61–65. competition. Acta Endocrinologica 118, (2002) Hormone plasma levels from the Urhaussen, A. & Kindermann, W. (1987) 332–336. pituitary–gonadal axis in performance Behavior of testosterone, sex-hormone Ponjee, G.A.E., De Rooy, H.A.J. & Vader, athletes after the 400 m run. Journal of binding globulin (SHBG), and cortisol H.L. (1994) Androgen turnover during Sports Medicine and Physical Fitness 42, before and after a triathlon competition. marathon running. Medicine and Science 243–249. International Journal of Sports Medicine 8, in Sports and Exercise 26, 1274–1277. Smith, U. (1980) Adrenergic control of 305–308. Roemmich, J.N. & Sinning, W.E. (1997) human adipose tissue lipolysis. European Vervoorn, C., Quist, A.M., Vermulst, L.J.M. Weight loss and wrestling training Journal of Clinical Investigations 10, et al. (1991) The behaviour of the plasma effects on growth-related hormones. 343–344. free testosterone/cortisol ratio during Journal of Applied Physiology 82, Strauss, R.H., Lanese, R.R. & Malarkey, a season of elite rowing training. 1760–1764. W.B. (1985) Weight loss in amateur International Journal of Sports Medicine 12, Rogol, A.D. (1989) Growth hormone: wrestlers and its effect on serum 257–263. physiology, therapeutic use and testosterone levels. Journal of the Wahren, J., Felig, P., Ahlborg, G. & Jorfeldt, potential for abuse. In: Exercise and Sports American Medical Association 254, L. (1971) Glucose metabolism during Sciences Reviews (Pandolf, K.B., ed.). 3337–3338. leg exercise in man. Journal of Clinical Williams &Wilkins, Baltimore, MD: Sutton, J.R., Coleman, M.J., Casey, J.H. & Investigations 50, 2715–2725. 353–377. Lazarus, L. (1973) Androgen responses Wheeler, G.D., Wall, S.R., Belcastro, A.N. Salvador, A., Suay, F., Martinez-Sanchis, during physical exercise. British Medical & Cumming, D.C. (1984) Reduced S., Simon, V.M. & Brain, P.F. (1999) Journal 1, 520–522. serum testosterone and prolactin levels Correlating testosterone and fighting Tanaka, H., Cleroux, J., de Champlain, J., in male distance runners. Journal of in male participants in judo contests. Ducharme, J.R. & Collu, R. (1986) American Medical Association 252, Physiology and Behavior 68, 205–209. Persistent effects of a marathon run on 514–516. Index

Note: Page numbers in italic refer to figures response to exercise 218, 220 adrenal gland 194, 200 and/or tables separate from the text acute aerobic and endurance training cortex 200 137–8 cortisol secretion 217–18 abortion, spontaneous 265 acute resistance training 139 hypertrophy 141 acclimatization/acclimation chronic aerobic and endurance testosterone secretion 526 cold 501, 502–3, 507 training 139 functions 194 heat 211, 479–80, 481 chronic resistance training 139 medulla 11, 194, 200 acetazolamide 49, 50 in coronary artery disease 143 activity after physical training 195 N-acetyl-β-endorphin 135 effects of intensity and duration chromaffin system 200–2 acetylcholine (ACh) 466 220–1 hypertrophy/‘sports’ 195 action on adrenal medulla 200, 201 effects of training 140, 223–4, 581 motor control of responses 195–6 in thermoregulation 468 high/low responders 149 neuroendocrine changes with exercise acetylsalicylic acid 48 influence of timing 221 194–5 ACh see acetylcholine modulating factors 144, 222–3 response to altitude 448–50, 457–8 acid-labile subunit (ALS) 43, 159, 180 ultraendurance exercise 140–1 role of neural feedback in responses 196 Washout Study 561–7 role in cortisol secretion 13, 217–18, 339, adrenaline see epinephrine acromegaly 420, 544–5, 545, 546 340 adrenocorticotropin see ACTH and body composition 547 in strength/power competitions 604 aerobic exercise carbohydrate metabolism in 552 stress-induced release 4–5, 136–7, acute cardiovascular system in 553 140–1 effects on growth hormone secretion fluid balance in 555 and testosterone secretion in women and release 122–5, 513 GHBP levels 101, 171 320 POMC response to 137–8 and muscle strength 549 actin 201, 312 chronic, POMC response to 139 renal function in 554 activins 287–8 overtraining 142 thermoregulation in 556–7 acute mountain sickness 447, 458 aerobic metabolism, impact of oral ACTH 200, 370, 467 ADAM-17 (TACE) 96, 97, 100, 104 contraceptive use 256–7 in acclimatization to exercise at low addiction to exercise 141–3 age altitudes 138 Addison’s disease 350 effects on AVP release and responses actions 2 adenosine triphosphatase (ATPase) 488, 493–4 in amenorrhea 225 in chromaffin granules 201 effects on bone remodeling 412, 413, assay 37, 39, 148 in epinephrine synthesis 205 423 circulating concentrations 3 adenosine triphosphate, in chromaffin effects on GH–IGF axis 116–17, 166, 181 effects on immune system 146, 350, 351 granules 205 effects on HPA axis 144, 222 in exercise–heat acclimation 481 ADH see antidiuretic hormone effects on male reproductive axis in exertional heat illness 482 adipose tissue 289–90 in moderate intensity competitions 606 distribution and postprandial cortisol effects on muscle 188 in overreaching 225, 581 response 433 effects on POMC and derivatives in overtraining 141–2, 225, 226, 590, 591 effects of androgenic-anabolic steroids 143–4 endurance exercise 581, 582 534 and exercise-associated anabolic effects resistance exercise 586 effects of chronic stress 5 165 in passive intense heat exposure 474, triacylglycerol balance 438–9 and growth hormone secretion 122 475, 476 adolescents response to acute aerobic exercise postprandial response 433–4 effects of training on growth and 124–5 release from POMC 134, 135, 136, 339 maturation 516–20, 521 response to chronic endurance response to altitude 450 wrestlers 518–20, 521 training 126–7 613 614 index

age (cont.) amine hormones (amino acid derivatives) angiotensinogen 452 response to resistance exercise 116–17 11 annexin-I 350 and testosterone response to resistance transport 14 anorexia nervosa see eating disorders training 326–7 amino acids anovulation 234, 264–5 and thirst 488 and protein synthesis 437 and bone mineral density 268–9 aggression supplements 579–80 ANP see atrial natriuretic peptide and androgenic-anabolic steroids 535–7 5-aminoimidazole-4-carboxamide-1-β-d- antagonism 23 and testosterone 601 ribofuranoside (AICAR) 391–2, 396 antibodies AICAR 391–2, 396 5’AMP-activated protein kinase 354, capture 37 Akt 389, 390 391–2 detection 37 exercise-induced activation 390 amphetamine 48 heterophilic 39 and muscle hypertrophy 396 structure 48 antidiuretic hormone (ADH) albumin 3, 14–15 AMPK 354, 391–2 actions 2 testosterone binding 291, 526 AMS 447, 458 response to altitude 453–4, 458 aldosterone 9, 200, 503 anaerobic metabolism, impact of oral anxiety 528 actions 2, 19 contraceptive use 257–8 markers 607 response to altitude 452, 453, 458 anaerobic threshold 166, 167, 168, 603 and POMC derivative response to response to cold exposure 504 analytes exercise 149–50 response to exercise 370 immunoassay 39 precompetition 600–1 3-α-androstanediol 291 preparation and purification 58–62 aquaporins 488 5-α-androstanediol 526 androgen receptor 286, 292–3, 526 L-arginine 128–30 α-melanocyte stimulating hormone 135, exercise-associated changes 309 arginine vasopressin (AVP) 9, 350, 467, 136, 467 gene mutations 293 495–6 5α-reductase 291, 292 and muscle hypertrophy 308–9 and cold-induced diuresis 503 deficiency 292 satellite cells 308 effects of age on 488, 493–4 α 1–16-endorphin 141 and testosterone activity 319–20 endocrinology 487–8 α 2-macroglobulin, transformed 94, 96 androgenic-anabolic steroids 49, 331 environmental influences on 494–5 ALS see acid-labile subunit and aggression 535–7 in exercise–cold stress 500, 504 altitude 444–5 balance between androgenic and and fluid balance 487, 503 acute exposure 444 anabolic effects 527 functions 217, 488 and barometric pressure 444, 445 effects on androgen receptors 309 menstrual factors affecting 492–3 chronic exposure 444 effects on body composition 532–4 in pregnancy 493 effects on AVP responses 494–5 effects on endurance performance receptors 487 effects on HPA axis 222 537–8 regulation of release 488, 489 and fertility 459 effects on muscle 307–8, 312–13, 533–4 response to cold exposure 504 integration of physiologic function and effects on reproductive hormones 74 response to exercise 218, 220, 370 neurohumoral response 457–9 effects on total body weight 530–2 effects of intensity and duration 221 low 138 health risks and benefits associated endurance exercise 488–9 neurohumoral responses to 448–57 538–9 endurance training 489–90 oxygen transport cascade 444–5 metabolism 62 gender differences 493 physiologic adaptations to 445–8 and motivation 535–7 influence of hydration status 491–2 amenorrhea 74, 261, 408, 420–2, 528 performance enhancement mechanisms resistance exercise and training 490–1 and adrenocortical response to exercise 534–8 role in cortisol secretion 217, 339 146 qualitative analysis 62–3 in stress response 4, 5 body composition hypothesis 238–9 strength gains due to 534–5 aromatase 250, 251, 526 and bone mineral density 236–8, 261, terminology 528 aromatase inhibitors 526–7 267–9 use by bodybuilders 532, 538 AS160 389 and cardiovascular disease 238 androgens 9, 10 ascorbic acid 200 causes 233–4, 235 binding 526 ATP in chromaffin granules 205 definition 263 exogenous administration 525, 527 ATPase see adenosine triphosphatase diagnosis 269 supraphysiological doses 525 atrial natriuretic peptide (ANP) 21, 467, effects on muscle metabolism 238 see also androgenic-anabolic steroids 503 endocrine anomalies associated 225, insensitivity to 293 actions 2 234–5 regulation 279 effects of growth hormone on 555–6 and energy availability 235, 236, 241–4, transport 291 in exercise–cold stress 500, 504 269–74 see also testosterone response to altitude 453, 454 exercise stress hypothesis 239–41 androstenedione 290 response to cold exposure 504 and infertility 238 androsterone 291 autocrine effects 8, 466 prevalence 234, 263, 515 angiotensin I 452 autoimmune disorders 373, 382–3 primary see menarche, delayed angiotensin II 467, 503 autonomic nervous system 345, 467 reversal 273–4 response to altitude 452–3 in overtraining 580, 593–5 American football 605–6, 609 and thirst 488 in thermoregulation 472 index 615

β AVP see arginine vasopressin 2-agonists resorption indices 410, 413 azoospermia 74 permitted 64–5 responses to acute exercise 414–18 qualitative analysis 64, 65 responses to loading 414–18 B-lymphocytes 211, 346 bicarbonate, stable-isotopically labeled 26 responses to training 418–22 activation 377 bioassay of growth hormone 78–9, 79–81, skeletal structure 409–10 apoptosis 377 87–8, 88–9 stress fractures 422–3 influence of cortisol 353 N,N-bistrifluoroacetamide 61 as target tissue 408 interaction with peptide F 206, 207, 212 blood trabecular 409 response to exercise 355, 356, 376, 377 adaptation to altitude 445–6 remodeling 410 response to training 358, 359 biocompartments, peptide F turnover markers 410, 413, 568–9 barometric pressure and altitude 444, 445 concentrations 211–13 bone marrow 409 basal metabolic rate doping tests 40 bone matrix proteins 288 effects of testosterone on 320 samples 59 bone mineral density (BMD) response to altitude 456 second generation 42 during growth 409 basketball 606 effects of growth hormone on 558 and energy availability 243 bed rest 413 GHBPs in 98–9 female athletes 236–8, 261, 262, 267–9, Beer–Lambert law 55 hormone transport 14–15 420–2 beta-blockers 49, 50 blood pressure limitations of use as endpoint of studies β-catenin, in bone response to loading and oral contraceptive use 254–5 419–20 415 in passive intense heat exposure 473 male athletes 299 β-endorphin 467 BMD see bone mineral density measurement 418 assay 148 BMK1 398 response to training 418–22 derivatives 135 body composition bone specific alkaline phosphatase 413 effects of oral contraceptives on 147 effects of androgenic-anabolic steroids boron 412, 527 in exercise–heat acclimation 481 532–4 bradykinin 467 in exertional heat illness 482 and growth hormone 546–8 brain, glucose need and availability functional significance 148, 149–50, 339 hormonal regulation 436 241–2 and heart rate 143 measurement 532 breast cancer 382 and immune function 146, 341–2 multicompartment models 532

in overtraining 141–2 and nutrition 435–9 c-Jun NH2-terminal kinases 398, 399, 401 and pain perception 148 and reproductive disorders 238–9 c-met 189 in passive intense heat exposure 474, body mass index, relationship with C peptide 83, 157 475 components of GH–IGF axis 100, C-telopeptide 413 and psychological effects of exercise 170 CAD 143 149–50, 558 body temperature caffeine 48, 48 receptors 339 and menstrual cycle 471–2 calcitonin 2, 9, 467 release from POMC 134, 135, 136, 339 and oral contraceptive use 258 calcium response to exercise 218, 220, 340 in passive intense heat exposure 473 in bone remodeling 410–11, 412 acute aerobic and endurance training regulation see thermoregulation effects of testosterone on retention 320 137–8 body weight, effects of androgenic- ion channels 22–3 acute resistance training 138–9 anabolic steroids 530–2 post-menopausal supplementation chronic resistance training 139 bodybuilders, androgenic-anabolic steroid 419 in coronary artery disease 143 use 532, 538 calmodulin 23, 201 correlations with reproductive bone cAMP 21–2 hormones 147 basic multicellular unit (BMU) 408, 410, cAMP response elements 22 effects of intensity and duration 221 411, 412 cAMP responsive element binding modulating factors 143–4 cortical 409 proteins 22 physiological and pathophysiological remodeling 410 cancer aspects 145 effects of testosterone on growth 320 and exercise 382 in post-traumatic stress disorder 149 formation markers 410, 413 role of immune system 373, 382 ultraendurance exercise 140–1 functions 409 carbohydrates response to extreme physical stress growth plate 409, 410 dietary restriction 438–9 140–1 imaging technologies 418, 419, 420, 423 metabolism in strength/power competitions 604 linear growth 409 in acromegaly 552 stress-induced release 136–7, 140–1 mass, effects of chronic stress on 5 and altitude 447, 458 training effects 139–40 modeling 409 effects of growth hormone 551–2 and ventilatory responsiveness 138 in response to loading 415 effects of oral contraceptives 254 β-lipotropin (β-LPH) remodeling 409–12 regulation by JNK signaling 401 assay 148 aberrant 412–14 supplements 438 release from POMC 134, 135, 136, 339 monitoring 413 post-exercise 437 response to exercise 340 in response to loading 415 carboxypeptidase B 202, 204 stress-induced release 136 systemic and local modulators 414 cardiac ischemia, exercise-induced 143 616 index

cardiac output chromaffin granules 11, 200 effects on immune system 342–3, 350, and altitude 446 exocytosis 200–2 351, 352 in passive intense heat exposure 473 chromogranin A 200, 205 in acute exercise 353–4, 355–7, 376 cardiovascular disease CID 503–4 in training 358–9 and β-endorphin response to exercise circadian rhythms 13–14 effects of nutrition and feeding on 143 circannual rhythms 14 432–4 risk in female athletes 238, 265–7 clenbuterol 49 in endurance competitions 602, 603 cardiovascular system CLIP 135 changes through the season 608–9 in acromegaly 553 CNBr 60–1 in exercise–cold stress 500, 504–5 adaptation to altitude 446 cocaine 48, 48 in exercise–heat acclimation 481 effects of growth hormone 553–4 codeine 48 in exercise-induced hepatic glucose and POMC response to exercise 143 cognition, effects of growth hormone 558 production 197 castration 283, 286, 319 cold in exertional heat illness 482 catecholamines 200, 348 acclimatization/acclimation in female athletes 225, 234, 235, 421 clearance 15 and catecholamines 501 functions 217, 339–40 definition 204 and prolactin 507 in moderate intensity competitions 606, effects on natural killer cell function and thyroid hormones 502–3 607 348, 350 exposure in overreaching 225, 581 in endurance competitions 602, 608, 609 and catecholamines 499 in overtraining 141, 142, 225, 226, 580, in exercise–cold stress 499, 500, 501 and cortisol 504 581, 582, 592 interaction with immune system 348, and fluid regulation 504 in passive intense heat exposure 474, 349, 350 and glucagon 505 475, 476 in acute exercise 353–4, 376 and growth hormone 506 postprandial response 433–4 in moderate intensity competitions 607 and insulin 505 precompetition levels 600–1 in overtraining 583–4, 586–7, 592 and reproductive hormones 506, 507 in reproductive disorders 239 precompetition levels 600, 601 and thyroid hormones 502 response to altitude 450–1 response to exercise 370 habituation 501, 502 response to cold exposure 504 in strength/power competitions 604, sensitivity 503 response to exercise 195, 225–6, 340–1, 605 see also exercise–cold stress 370 synthesis 11 cold-induced diuresis 503–4 in coronary artery disease 143 see also epinephrine; norepinephrine collagen in bone remodeling 410–11 effects of feeding 434 Cbfa1 410 competition 600 effects of intensity and duration CCK 270, 467 endocrine responses during 601–7 220–1 cell proliferation assay 80–1 endocrine responses preceding 600–1 effects of training 140, 223–4, 581 cellular communication 1, 3, 8 endocrine responses through the season influence of timing 221 central nervous system, role of growth 607–10 influence of type of exercise 221–2 hormone 558–9 endurance 602–3, 608–9 mechanism 218, 220 CG see human chorionic gonadotropin moderate intensity 606–7 modulating factors 144, 222–3 childbirth, regulation of endocrine preparation for 175–6 secretion 217–18, 339–40 function during 13 recovery period 603 regulation 13 children strength/power 603–6, 609–10 in strength/power competitions 604–5, effects of training on bone mineral ultraendurance 603 605–6 density 418 core binding factor-1 410 changes through the season 609 effects of training on growth and coronary artery disease 143 see also testosterone/cortisol ratio maturation 516–20, 521 corpus albicans 253 cortisol/glucocorticoid receptor complex gymnasts 517–18 corpus luteum 233, 250, 253 350 male reproductive axis 289 corticosteroid binding globulin 15 CP 80–1 measurement of maturation of motor corticotropin-like intermediate protein creatine, and cortisol response to exercise control 31, 32 135 223 in non-weight control sports 516–17 corticotropin-releasing hormone (CRH) creatine phosphate 603 recombinant human growth hormone 350, 370 CREBs 22 abuse 544 assay 39 CRH see corticotropin-releasing hormone in weight control sports 517–20, 521 in female athletes 225, 420–1 cross-country running 608 cholecystokinin 270, 467 receptors 217 Cushing’s disease 350 cholesterol 9, 10, 340 response to exercise 218, 220 cyanogen bromide 60–1 effects of growth hormone 552–3 effects of intensity and duration 221 cyclic adenosine monophosphate 21–2 effects of oral contraceptives 255 role in cortisol secretion 13, 217, 339, cytokines 347 chondrocytes 409 340 actions 470 chromaffin cells 11, 200–1 in stress response 4, 5 control of muscle response to stress and concentrations of neurohormones 205 in thermoregulation 469 damage 360–2 extra-adrenal 200 cortisol 9, 200, 370 in fever 469, 470 storage of epinephrine and peptide F circulating concentrations 3, 14 interaction with HPA axis 352, 380 203–4 in delayed onset muscle soreness 361 receptors 4, 21 index 617

response to exercise 354, 355, 379–81 qualitative analysis of compounds training load 581 response to training 358–60 62–4, 65 and ventilatory responsiveness 138 in thermoregulation 470 quantitative analysis of compounds energy see also specific cytokines 64–5 availability 244 sample preparation and purification and EAMD 241–4, 269–74 DAG 22 58–62 optimal 270–2 dehydration Drosophila, decapentaplegic gene complex balance, and male reproductive axis and AVP response to exercise 491 288 293–5 in environmental heat stress 472 drugs, prohibited compounds 47–9, 50 deficiency in exercise–heat stress 478 DTH following exercise 379 in female athletes 235, 236, 269 dehydroepiandrosterone 146 dual energy X-ray absorptiometry 418, in male athletes 528 dehydroepiandrosterone sulfate 320 419, 532 markers of 244–5 delayed type hypersensitivity following Dutch Hunger Winter 295, 296 intake and expenditure, in female exercise 379 DXA 418, 419, 532 athletes 235–6 deoxycorticosterone 2 dynorphins 135 partition 241 deoxypyridinoline 413 status, and menarche 263–4 depression E2 see estradiol enkephalin-containing peptides 202–3 effects of exercise on 149–50 EAMD see exercise-associated menstrual post-translational processing from effects of growth hormone 558 disturbance preproenkephalin 203 derivatization techniques 61–2 eating disorders 232 enkephalins 135 designer steroids 62–3, 527–8 and bone mineral density 237, 267–8, secretion 202 desogestrel 254 420–1 enzyme-immunoassay 37 DHEAS 320 and estrogen deficiency 267 enzyme linked immunosorbent assay 37 DHT 3, 290, 291–2, 320, 526 and female athlete triad 261–2, 420–2 ephedrine 48 diabetes mellitus and race 269 quantitative analysis 64 muscle metabolism in 27 ECPs 202–3 structure 48 and POMC response to exercise 145 Efe Pygmies 162 epinephrine (EPI) 194, 345, 348 diacylglycerol 22 EI 55–6, 57 actions 2, 205 diclofenac 48 EIA 37 complementary action with peptide F 5α-dihydrotestosterone 3, 290, 291–2, 320, eIF2B 390, 396 204 526 electrolyte balance 487 effects on natural killer cell function binding 526 electron ionization 55–6, 57 348, 350 1,25 dihydroxyvitamin D 3 electrospray ionization 56, 57, 64, 65 in endurance competitions 602, 609 diiodotyrosine 11 ELISA 37 in exercise–cold stress 501 directional control following stroke 30–1 endocrinology, definition 1 in exercise–heat acclimation 481 discus throw 604 endocytosis, receptor-mediated 16 in exercise–heat stress 477–8 DIT 11 endometrium 233, 251, 253 in exertional heat illness 482 diuresis endorphins 135 interaction with immune system 348, and altitude 446 endurance, definition 580 349, 350 cold-induced 503–4 endurance competitions in moderate intensity competitions 607 diuretics 49 endocrine responses through the season in overtraining 586–7 osmotic 49, 50 608–9 in passive intense heat exposure 474, qualitative analysis 64, 65 endocrine responses to 602–3 476 structures 50 endurance exercise and training plasma concentration 205 thiazide 49, 50 AVP response to 488–90 plasma half-life 205 diurnal rhythms 13–14 definitions 580–2 precompetition levels 601 DOMS 360–2 effects on adrenomedullary activity ratio with peptide F 205 dopa 11, 204 195 release 201–2 dopamine 11, 204–5, 466 effects of androgenic-anabolic steroids and peptide F release 203–5 actions 205 537–8 response to altitude 449, 450, 457–8 in exercise–heat acclimation 481 effects on GH–IGF axis 126–8, 171–3, response to exercise 194–5, 209, 210, in exertional heat illness 482 174, 515 348, 370 in thermoregulation 469 effects on male reproductive axis 298–9, response to overtraining 211 dopamine β-hydroxylase 204–5 516, 528–30 response to stress 195 doping effects on satellite cells 307 response to training 195 analysis of low molecular weight increase in training intensity 581, 582 role in exercise-induced hepatic glucose substances 47–68 increase in training volume 581, 582 production 196–7 death related to 47 overtraining 141–2, 580–4, 608 role in leukocyte trafficking 348 history of 47 peptide F response to 208–10 role in lipolysis 197 need for specificity and reliability of POMC response to 137–8, 139 role in muscle carbohydrate metabolism measurements 36 short/middle/long type activities 581 197 peptide hormone assay 40–4 supercompensation cycle 581 in strength/power competitions 604, prohibited compounds 47–9, 50 training cycles 581 605 618 index

epinephrine (EPI) (cont.) ethylmorphine 48 menstrual cycles 233–6 and sympathetic–adrenal medullary ethynodiol diacetate 254 metabolic efficiency 236 axis 467 etiocholanolone 291 fenoterol, qualitative analysis 64, 65 synthesis 11, 204–5 Ets transcription factor family 160 fertility and altitude 459 in thermoregulation 468 eukaryotic initiation factor 2B 390, 396 fertilization 233 epithiazide 64, 65 euthyroid sick syndrome 5 fetus, hypothalamic–pituitary–gonadal epitopes in immunoassay 37, 39 exercise-associated menstrual disturbance axis 288–9

EPO see erythropoietin (EAMD) 71–2, 73, 233–8, 261–2, FEV1 33 ERK 398 274, 298, 515–16 fever 469, 470 signaling 398, 399, 400–1 and body weight/body fat 272–3 FIA 37 ERK5/big MAP kinase-1 398 clinical considerations 265–9 fibroblast growth factors and muscle erythropoiesis, effects of testosterone on continuum of 263 hypertrophy 309–10 320 definitions and prevalence 262–5 FID 50, 54–5 erythropoietin (EPO) demographics 269 fight or flight reaction 194, 205, 345, 600 assay 37, 41 and energy availability 235, 236, 241–4, flame ionization detector 50, 54–5 isoforms 41 269–74 fluid regulation 487 recombinant human (rhEPO) historical aspects 261 and altitude 446, 452–4 doping tests for 41–2 implications for athletic training 244–5 effects of growth hormone 555 risk of 537 metabolic and reproductive hormone in exercise–cold stress 500, 503–4 response to altitude 445, 451–2, 458 perturbations 271 fluid retention, in growth hormone secretion 526 and POMC derivatives 146–7 treatment 555 ESI 56, 57, 64, 65 prediction of 270–3 fluorescence-immunoassay 37 estradiol (E2)9 prevalence 515 fluorometric assay 37 actions 251 proposed mechanisms 238–44 follicle-stimulating hormone (FSH) 71 in anovulation 264 reversal 273–4 actions 2 and bone resorption 421 exercise–cold stress 499–511 biochemical structure and molecular and cardiovascular disease 265 exercise–heat stress 476–81, 482 biology 283–4 circulating concentrations 3 exercise performance at altitude in chronic illness 297 effects of aging in men 289–90 maximal 447–8 circulating concentrations 3 effects of strenuous physical training 71 submaximal 448 deficiency 285–6 impact on muscular damage 257–8 exercise testing effects of aging in men 290 and infertility 265 advances in technology 25–7 effects of strenuous physical training 71 in male reproductive axis 287, 291–2 following stroke 29–31 in female athletes 421 in menstrual cycle 251, 252–3 imaging in 27–9 fetal 288–9 in oligomenorrhea 264 and the new biology 33 in male reproductive axis 279, 285–6 predictors of level 238 number of clinical trials 25 feedback regulation 286–8 response to acute exercise 514 scope in clinical medicine 25 ontogenesis of secretion during response to altitude 457 use of robots 30–1 phases of life 288–90 and thermoregulation during exercise exertional heatstroke 480–1, 482 in menstrual cycle 233, 250, 251–2 479 exhaustion 141 receptors 251, 252, 285, 286 estrogen receptors 251 extracellular-signal regulated kinases see response to altitude 457 blocking agents 527 ERK secretion 250, 251 mutations 291 regulation 280–3 estrogens 9, 10, 71 fasting 73, 74 in strength/power competitions 604 actions 2 FAT/CD36 401 and weight loss 293 in amenorrhea 233, 234 feedback systems 4–5 follicular suppression 234 cardioprotective effects 238, 265, 266 negative 12–13 follistatins 288 effects on AVP responses 492–3 positive 13 forced expiratory volume in 1s 33 effects on endometrium 233 female athlete triad 408, 528 forkhead transcription factor family 160 effects on muscle fibers 313 hormonal changes in 420–2 4E binding protein-1 390 interaction with testosterone 526 need for intervention 238 FSH see follicle-stimulating hormone in male reproductive axis 287, 291–2 recognition of importance 261–2 furosemide 49, 50 in menstrual cycle 233, 250–1, 252–3 relationship between components 262 in oral contraceptives 254–5 treatment 423 G6P 395 response to acute exercise 514 female athletes G protein coupled receptors 4, 21, 284, response to cold exposure 508 bone mineral density 236–8, 261, 262, 487 synthetic 254–5 267–9 G proteins 21–3

and thermoregulation during exercise cardiovascular disease 238 GI 21 479 clinical concerns 236–8 GS 21 estrone, effects of aging in men 289–90 endocrine anomalies 234–5 GABA 466, 469 ethacrynic acid 49, 50 energy intake and expenditure 235–6 GABA receptors 469 ethinyl estradiol 254 infertility 3, 233, 238 galanin 270, 467 index 619

gamma-aminobutyric acid 466, 469 localization 18 gonadotropin-releasing hormone (GnRH) γ-lipotropin 135, 136, 339 glucocorticoid response element 350 71 γ 1–17-endorphin 141 glucocorticoids 9, 10, 200 deficiency 279 GAS 345, 360, 578–9 actions 2, 352 developmental origin and migration of gas chromatography (GC) 40, 51–3 effects on immune system 350–3 secreting neurons 279, 281 column composition 51 in exercise and training 141 effects of energy deficiency and weight detectors 54–5 response to altitude 450–1 loss on secretion 295 fused silica tubing 51 glucose effects of strenuous physical training 72, ionization techniques 55–6 competition for 241–2 73 mass analyzers 57–8 hepatosplanchnic production in in female athletes 234, 421 outer coating 52–3 response to exercise 196–7 fetal 288 sample preparation and purification metabolism in male reproductive axis 279–83, 319 58–62 effects of oral contraceptives 254, 258 in menstrual regulation 233, 250, 251, stationary phase 51–2 role of GH–IGF axis 430 253 gas chromatography–mass spectrometry plasma role in testosterone secretion 71, 526 55–6, 61 in female athletes 234 secretion 233, 250, 280–3, 319 gas exchange measurement 26, 27, 29, 30 in prolonged exercise 478 Graafian follicle 251 gas–liquid chromatography 50 postprandial response 434–5 graft rejection 373 GC see gas chromatography production granulocyte–macrophage colony- GC–MS 55–6, 61 in endurance competitions 602 stimulating factor, interaction with gender in moderate intensity competitions HPA axis 352 effects on AVP responses 493 606 granulosa cells 250, 251, 253 and growth hormone regulation at altitude 454–5 growth correlation with fitness 170 transport in muscle effects of training on 516–19, 520 release 119 effects of training 394–5 gymnasts 517–18 response to acute aerobic exercise exercise-related 197, 390–2 and male sexual maturation 293 123–4 insulin-regulated 388–9 prenatal 161 response to resistance exercise role of JNK signaling 401 prepubertal 512–13 117–18 role of p38 signaling 401 role of GH–IGF axis 119, 161–2, 409, influence on HPA axis 144, 222 glucose-6-phosphate 395 512–13, 546 influence on POMC and derivatives 144 GLUT4 glucose transporter 389, 397, 401 wrestlers 519–20, 521 and testosterone response to resistance glutamate 466, 468–9 growth hormone (GH) 9, 110, 165, 467 training 326 glycemic index 434–5 ability to cross blood–brain barrier 558 and thermoregulation during exercise glycemic load 434–5, 438 abuse see recombinant human growth 479 glycerol 223 hormone, abuse general adaptation syndrome 345, 360, glycine 466 acquired insensitivity/resistance to 578–9 glycogen 101–2 gestrinone 62, 527 resynthesis following endurance actions 2, 78–9, 166, 180–1 gestrodone 254 competitions 603 and anterior pituitary complexity 86–7 GH see growth hormone stores in endurance competitions 602 assay 77 GH-2000 Project 561–7, 569 synthesis in muscle bioassay 78–9, 79–81, 87–8 GH–IGF axis see growth hormone–insulin- exercise-related 390–2 dichotomy between bioassayable and like growth factor axis insulin-regulated 389–90 immunoassayable 88–9 GH-N gene 82, 110, 112 supercompensation 395–6 effects of GHBPs on 103–4 GHBPs see growth hormone binding glycogen phosphorylase 198 exercise-induced isoforms 83–6 proteins glycogen synthase immunoassay 37, 77, 80, 88, 103–4 ghrelin 111, 122, 127–30, 428 exercise-related regulation 391, 392 autoinhibition 167 and exercise-associated menstrual and glycogen supercompensation 395 bed rest studies 88–9 disturbance 270 insulin-related regulation 390 in bone response to loading 416 GHRH see growth hormone releasing regulation by ERK1/2 signaling 400–1 cadaveric preparations 44 hormone regulation by JNK signaling 401 circulating concentrations 3, 14 GHRs see growth hormone receptors glycogen synthase kinase-3 (GSK3) 389, deficiency 181, 544 gigantism 545, 546 389–90, 401 in adults 546–7 GLC 50 in bone response to loading 415 and biological actions of growth glucagon (GLG) 9, 467 effects of exercise 390–1 hormone 545–59 actions 2 glycogenolysis diagnosis 167–8 in endurance competitions 602 hepatic, in response to exercise 196–7 dual effector hypothesis 180 in exercise–cold stress 500, 505 muscular, in response to exercise 197 effects of age on 181 response to altitude 455 GM–CSF, interaction with HPA axis 352 effects on blood 558 response to cold exposure 505 GnRH see gonadotropin-releasing effects on body composition 546–8 response to exercise 370 hormone effects on carbohydrate metabolism glucocorticoid receptor 350 gonadotropes 283 551–2 620 index

growth hormone (GH) (cont.) resistance to 609 role in growth 119, 161–2, 409, 512–13, effects on cardiovascular system 553–4 response to altitude 455, 456 546 effects on cholesterol metabolism and response to cold exposure 506 role in thermoregulation 556–7 vascular health 552–3 role in central nervous system 558–9 growth hormone receptors 165–6 effects of GHBPs on activity 99–100 role in growth 119, 409, 512–13, 546 central nervous system 558 effects on lipolysis and lipid oxidation role in immune system 557–8 and GHBP 94–5, 96–7, 99–100 550–1 role in resistance exercise-induced GHR/ghr genes 97 effects on muscle fibers 313 muscle hypertrophy 118–19 mutations 101 effects on muscle strength 548–9 role in thermoregulation 556–7 products 97 effects of nutrition and feeding on secretion and release 111, 113–14 inactivation 99 426–9 control 110–12, 122 interaction with growth hormone 79, 81 effects on protein anabolism 549–50 neuroregulatory modulators 112, interaction with recombinant human effects on renal structure and function 113 growth hormone 79 554–6 patterns 112–13 polymorphism 95 in endurance competitions 602 pulsatile 111, 112, 119, 166 regulation of expression 100 changes through the season 609 rebound 112, 116 sweat gland 556 exercise-induced release 77, 181, 370 sexual dimorphism in 119 growth hormone releasing hormone acute aerobic exercise 122–6, 513 secretion granules 87 (GHRH) 110–12, 122, 128–30, 165 attenuating factors 167 system complexity 77 in exercise-induced GH release 127–30 chronic endurance training 126–8 Washout Study 561–7 growth hormone secretagogue see ghrelin effects of feeding on 428–9 wrestlers 519, 520 GSK3 see glycogen synthase kinase-3 neuroendocrine control 128–30 growth hormone binding proteins guanylate cyclase 21 resistance exercise and training 90, (GHBPs) 110, 112, 165–6 gymnasts, growth and maturation 517–18 91, 114–19, 513, 515 assay 102–4 single exercise bout 166–8 in biological fluids 98–9 hair samples 59 training effects 170–2, 173, 515 in disease 101–2 hammer throw 604 variant forms 83–6 effects on growth hormone assay 103–4 hand trajectories following stroke 30 Washout Study 561–7 function 99–101 Hb, and altitude 445–6 in exercise–cold stress 500, 505–6 generation and tissue sources 96–8 hCG see human chorionic gonadotropin in exercise–heat acclimation 481 historical aspects 94 heart rate in exercise–heat stress 476 low affinity 94, 95–6, 104 and altitude 446, 450 in exertional heat illness 482 assay 103 in passive intense heat exposure 473 in female athletes 235 in blood 98–9 response to exercise 143 genes 560 in disease 102 heat hypersecretion 101, 171, 420, 545 function 100 acclimation/acclimatization 211, and biological actions of growth nature and chemical properties 94–6 479–80, 481 hormone 545–59 in pregnancy 100 environmental heat stress 472 pathological 545 regulation of production 100–1 exercise–heat stress 476–81, 482 see also acromegaly; gigantism response to exercise 101 exertional heatstroke 480–1, 482 hypersensitivity to 102 effects of training 170–1, 172 passive intense exposure 473–6 interaction with growth hormone Washout Study 561–7 tolerance, impact of oral contraceptives receptors 79, 81 serum concentrations 100 258 international reference preparations 39 shedding 96, 98 heat exhaustion 480–1, 482 isoforms/variants 42, 43–4, 79, 82–3, wrestlers 519, 520 heat shock proteins 110, 112, 181, 560 growth hormone–growth hormone effects of exercise on expression by biological actions 561 binding protein complex 99–100 leukocytes 34 exercise-induced 83–6 growth hormone–insulin-like growth HSP90 19 as marker of rhGH abuse 545 factor (GH–IGF) axis 94, 95, 165–6, and immune system activation 383 markers of action 559 180–1 in overtraining 583, 592 in moderate intensity competitions 606, and bone remodeling 414 heatstroke, exertional 480–1, 482 607 in glucose metabolism 430 hemoconcentration and altitude 446 in overtraining 582–3, 586, 590, 591 influence of aging 116–17, 166, 181 hemodialysis 297–8 in passive intense heat exposure 474, influence of nutrition 166, 172 hemoglobin and altitude 445–6 475, 476 interaction with immune system 557–8 hexamethyldisilazane 61 pharmacodynamic endpoints of action in overtraining 582–3 HIF-1 312 43 relationship with body mass index 100, HIF-1α 447, 453, 458 postprandial response 427–8 170 high performance receptor binding post-transcriptional/translational response to exercise 116–17, 166–9, 513 chromatography 79–80 modification 82, 110, 112 endurance exercise and training hip fracture provocative tests 167–8 125–7, 171–3, 174, 515 IGF-I in 411–12 regulation of release by muscle afferents training effects 169–75, 515 risk factors 413 89–90 Washout Study 5561–7 histamine in thermoregulation 468 index 621

HIV infection 358 hyperglycemia 145, 427–8 preoptic-anterior (POAH) androgens in 297, 298 hyperhidrosis 556 neuron populations 469, 471 and exercise 381 hyperhydration 491 in reproduction 471–2 HMDS 61 hyperinsulinaemia 170 in thermoregulation 468, 469, 470–1 Holland, Dutch Hunger Winter 295, 296 hypermetabolism 473 role in male reproductive axis 279–80 homeostasis 1, 3, 8 hyperparathyroidism, secondary 412 hypoxia, at altitude 445, 446 and feedback 4 hyperthermia 472 hypoxia inducible factor-1 312 homovanillic acid 558 in exercise–heat stress 476–81, 482 hypoxia inducible factor-1α 447, 453, 458 hormone binding proteins 3, 14–15 and heatstroke/heat exhaustion 480–1, hypoxic ventilatory depression 445 effects on immunoassay 39 482 hormone–receptor complex 4 in passive intense heat exposure 473–6 IAC 59–61 affinity 16 hypoestrogenism idiopathic hypogonadotropic binding to HREs 19 and bone mineral density 267, 268 hypogonadism 279, 281, 289 cross-reactivity 16 and cardiovascular disease 266–7 IEF 41, 42 formation 18 hypoglycemia 145, 428 IFA 80 hormone replacement therapy and bone hypogonadism IFMA 37 mineral density 419 artificial induction 530 IGF-I see insulin-like growth factor-I hormone response elements 19 in chronic illness 297–8 IGF-IR see insulin-like growth factor-I hormone sensitive lipase 198 hypergonadotropic 293, 297 receptor hormones hypogonadotropic 279, 281, 289, 293, IGF-II see insulin-like growth factor-II actions 2 295 IGF-IIR see insulin-like growth factor-II activation 3 in male athletes 528–30 receptor binding to receptors 18 hypohidrosis 556 IGFBPs see insulin-like growth factor classification 9–11 hypopituitarism 552–3 binding proteins constitutive release 12 hypothalamic–pituitary axis, and growth IHH 279, 281, 289 definitions 1, 466 hormone secretion 110–11 IL-1 receptor antagonist (IL-1ra) functional principles 8–9 hypothalamic–pituitary–adrenal (HPA) in delayed onset muscle soreness 361 functions 3–4 axis 217, 225–6, 432 and exercise training 173 historical aspects 1, 3 delayed response 137 response to exercise 354, 380, 381 integration of target cell responses to 23 effects of nutrition and feeding on 432–4 illness, susceptibility to 357, 360, 374, 379, mechanisms of action 16–23 interaction with immune system 246, 381–3 metabolic clearance 15–16 342–3, 350–3 ILMA 37 post-translational modification 3 non-delayed response 137 IM-9 lymphocyte cell line assay 79 secretion in overreaching 224–5, 581 imaging, in exercise testing 27–9 rates 3 in overtraining 141, 224–5, 226, 581, 582 immune system 345 regulation 4–5, 12–14 regulation 219 in autoimmune disorders 373, 382–3 synthesis 9–11 response to exercise 218–22 in cancer 373, 382 terminology 1, 8 modulating factors 222–3 and cardiopulmonary response to transport in blood 14–15 training effects 140, 223–4, 581 exercise 383 HPA axis see response to stress 136, 350, 352, 370 cell-mediated response 212 hypothalamic–pituitary–adrenal role in regulation of hormone secretion components 371 axis 4–5, 12–13 function assessment 374–5 HPRBC 79–80 role in reproductive disorders 239 humoral response 212 HREs 19 hypothalamic–pituitary–gonadal axis in infection 357, 360, 373 HRT and bone mineral density 419 in childhood 289 influence of ACTH 146, 350, 351 HSL 198 effects of acute/chronic illness 297–8 influence of β-endorphin 146, 341–2 5-HT see serotonin and energy balance 293–7 interaction with GH–IGF axis 557–8 human chorionic gonadotropin (hCG) female athletes 233–4, 235, 420–2 interaction with HPA axis 246, 342–3, doping tests for 41 female reproductive cycle 232–3, 250–3 350–3 international reference preparations 41 fetal 288–9 interaction with peptide F 206–7, placental 289 homeostasis 69–71 211–12, 214 and spermatogenesis 285, 286 interaction of factors affecting 75 interaction with sympathetic nervous structure 283–4 male reproductive axis 279–93, 319 system 345, 348–50 human growth hormone see growth response to exercise key components 346–7 hormone acute exercise 514 mucosal immunity 355, 357, 371, 378–9 HVA 558 training effects 71–4, 515–16 and muscle damage 360–2 hydration status hypothalamic–pituitary–somatotropic axis neuroendocrine modulation see and exercise 491–2 see growth hormone–insulin-like neuroendocrine–immune axis in exercise–heat stress 478–9 growth factor axis response to exercise 146, 373–4 hydrochlorothiazide 49, 50 hypothalamus 4 acute exercise 353–7, 375–9 5-hydroxytryptamine see serotonin gonadotropin-releasing hormone athletes vs. non-athletes 379 hyperbaria, effects on AVP responses 495 secretion 279–83 evaluation 374–5 622 index

immune system (cont.) signaling through MAPK pathway central nervous system 558 response to overtraining 359–60 398–401 deficiency 162 response to training 357–60 signaling through PI3K pathway 388–9 effects of exercise on expression or in thermoregulation 469–70 insulin-like growth factor-I (IGF-I) 156, activity 163 immunoaffinity chromatography 59–61 165, 180 gene mutation 162 immunoassay 37–9 actions 166, 186 hybrids 158–9, 185 analyte 39 binding 42–3 knockout mice 161 calibrator 39 in bone remodeling 411–12 and muscle hypertrophy 311 competitive 37, 38, 39 in bone response to loading 416, 417 signaling pathways 159–61 differential approach 43–4 deficiency 161, 162 insulin-like growth factor-II (IGF-II) 156, in doping tests 40–4 during preparation for competitions 310 GHBPs 103 175–6 in bone remodeling 411 growth hormone 37, 77, 80, 88, 103–4 effects of age on 181 deficiency 161, 162 international reference preparations effects of nutrition and feeding on expression 157 (IRPs) 39, 40 411–12, 429–30 gene 180 matrix 39–40 in female athletes 234, 237 overexpression 162 sandwich 37, 38, 39s gene 180, 182 prohormone 157 standardization 39–40 expression 157 response to exercise 169 immunofunctional enzyme linked mutation 162 role in growth and development 161–2 immunoassay 80 genetic manipulation in muscle 186–7 structure 156–7 immunoglobulins 346 and growth 409 insulin-like growth factor-II receptor response to exercise 355, 357, 378–9 in hip fracture 411–12 (IGF-IIR) 157–8, 166, 185 response to training 359 as indicator of training load 175 role in prenatal growth 161 infection international reference preparation 39 insulin-like growth factor binding proteins immune system in 373 isoforms/variants 174, 310 (IGFBPs) 3, 156, 159, 185–6 susceptibility to 357, 360, 374, 379, actions and receptors 185 in bone remodeling 411 381–2, 557–8 expression in muscle 181–2, 183–4 and exercise training 171, 172, 173, 174 infertility 3, 233, 238, 265 gene transfer 187 IGF-independent/IGF receptor- and nutrition 295, 296 in muscle regeneration 189 independent effects 159 inflammation 358 terminology 182 IGFBP-1 159, 166, 169, 171 in delayed onset muscle soreness 360–2 knockout mice 187 effects of nutrition and feeding on inhibins 71 mediation of growth hormone effects 429–30 in female reproductive cycle 253 114, 122 in female athletes 234 isoforms 253, 287, 288 in muscle hypertrophy 186, 187–8, IGFBP-2 159, 166, 169 in male reproductive axis 287–8 310–11 and exercise training 171, 172 structure 287, 288 in muscle regeneration 189–90 expression 186 inositol triphosphate 22 overexpression 162, 186–7 IGFBP-3 43, 159, 166, 180 insulin 9, 467 in overtraining 175, 582–3, 586 and exercise training 172, 173 actions 2 and periosteal growth 413–14 exercise-induced increase 169 in skeletal muscle 388–98 and physical conditioning scores 175 in female athletes 237 assay 37 postprandial response 429–30 IGFBP-4 159, 166, 171, 186 circulating concentrations 3 prohormone 157, 189–90 IGFBP-5 159, 166 clearance 16 recombinant 412 and exercise training 171, 172 dietary maximization of secretion 435 regulation by growth hormone 180 expression 186 effects of nutrition and feeding on resistance 162 IGFBP-6 159 434–5 response to exercise 168–9, 412, 513 expression 186 in endurance competitions 602 effects of feeding 430 knockout mice 162 in exercise–cold stress 500, 505 effects of training 171–5, 515 Washout Study 561–7 in female athletes 235 resistance training 169, 173, 174, 183 insulin-like growth factor signaling system in growth hormone deficiency 551–2 and rhGH abuse 561–8 156–64 in overtraining 582, 586 Washout Study 561–7 see also individual components postprandial response 434–5 role in growth and development 161–2, insulin receptor (IR) 157, 185 resistance 158–9 512–13 effects of exercise on expression/activity muscle damage-related 397–8 role in immune system 557 163 response to altitude 454–5, 458 and sarcopenia 188 hybrids 158–9, 185 response to cold exposure 505 secretion 526 isoforms 157–8, 159, 163 response to exercise 370 by active muscle 182–4 knockout mice 161, 390 sensitivity to skeletal 411 in PI3K pathway 388, 389 and cold exposure 505 structure 156–7, 184, 185 signaling pathways 159–61 effects of exercise on 392–4 in wrestlers 519, 520 insulin receptor-related receptor 159, 161 effects of training 394–5 insulin-like growth factor-I receptor insulin receptor substrate family 160, 161, in growth hormone deficiency 551–2 (IGF-IR) 157–8, 166, 184–5 163 index 623

intercellular communication 1, 3, 8 isoelectric focusing 41, 42 Leydig cells 71, 279 interferon-α isotope ratio mass spectrometry, development 290–1 interaction with HPA axis 352 testosterone 64 testosterone secretion 284–5, 290, 319, response to exercise 381 526 interferon-γ 347 janus tyrosine kinases (JAKs) 21 LH see luteinizing hormone response to exercise 354, 380, 381 jet-lag 3 LH–hCG receptor 284, 285 interleukins 380 JNK 398 LIA 37 IL-1 signaling 399, 401 LIF see leukemia inhibitory factor interaction with HPA axis 352 joining peptide (JP) 135 ligand-mediated immunofunctional assay response to exercise 381 (LIFA) 103 response to training 173 Kennedy’s disease 293 lipid metabolism IL-1α, in fever 469 ketogenic diet 223 effects of epinephrine 197–8 IL-1β 347 ketones, urinary 245 effects of oral contraceptives 255, 258 in delayed onset muscle soreness kidney, effects of growth hormone 554–6 in endurance competitions 602 361 !Kung San 295, 297 in moderate intensity competitions 606 in fever 469 role of MAPK signaling 401 interaction with HPA axis 352 lactate lipocortin-1 350 response to exercise 354, 381 and POMC system activation 145 lipolysis response to training 358 response to exercise 209 effects of epinephrine 197–8 IL-2 347 lactate paradox 447 effects of growth hormone 550–1 interaction with HPA axis 352 lactation, influence on POMC and lipophilic hormones 18–19 response to exercise 354, 380, 381 derivatives 144 lipophobic hormones 19 IL-3 352 lactic/anaerobic threshold 166, 167, 168, lipoprotein lipase 198 IL-4 347, 352 603 lipoprotein receptor protein-5 415 IL-5 352 Laron syndrome 101, 103, 556 lipoproteins 9, 10, 340 IL-6 347 LAT 166, 167, 168, 603 effects of androgenic-anabolic steroids in delayed onset muscle soreness 361 LC see liquid chromatography 538 effects of feeding on response to LC–MS 55, 56, 57 effects of growth hormone 552–3 exercise 434 LC/NE system 348 effects of oral contraceptives 255 in fever 469 in stress response 4, 5 liquid chromatography (LC) 53–4 interaction with HPA axis 352, 380 leptin 239 detectors 54–5 in muscle response to damage 360 actions 2 ionization techniques 55, 56, 57 response to exercise 354, 355, 357, 380, deficiency 295 mass analyzers 57–8 381 effects on bone 416 sample preparation and purification response to training 172–3, 358 and exercise-associated menstrual 58–62 IL-8 347 disturbance 242, 243, 270 liquid chromatography–mass in delayed onset muscle soreness 361 in female athletes 234, 235 spectrometry 55, 56, 57 in fever 469 in male reproductive axis 295 liquid–liquid extraction 59 interaction with HPA axis 352 in overtraining 583 liver response to exercise 354 and puberty 263, 513 androgenic–anabolic steroid-associated IL-10 347 response to acute exercise 514 adenoma 538 in delayed onset muscle soreness response to training 516 glycogenolysis in response to exercise 361 leucine 437 196–7 interaction with HPA axis 352 [Leu]-enkephalin 202–3 LKB1 391 response to exercise 354, 380 leukemia inhibitory factor (LIF) LLE 59 IL-12 347 effects on HPA axis 352 locus ceruleus–norepinephrine system 4, interaction with HPA axis 352 in muscle response to damage 360 5, 348 response to exercise 381 as stimulus to ACTH secretion 218 low molecular weight substances, analysis IL-13 352 leukocytes 346 in doping control 47–68 interactions with β-endorphin 342 function LPD 264–5, 268–9, 515 interactions with cortisol 342 response to exercise 355–7 LPL 198 intracrine effects 8 response to training 358–60 LRP-5 415 iodide 11 role of catecholamines 348–50 luminescence-immunoassay 37 ion channels 4, 22–3 role of glucocorticoids 351, 352–3 luminometric assay 37 calcium 22–3 phenotyping 374 luteal phase defects 264–5, 515 ligand gated 20 trafficking and bone mineral density 268–9

IP3 22 response to exercise 353–4, 375–6 luteal suppression 234 IR see insulin receptor response to training 358 luteinizing hormone (LH) 71, 467 IRMA 37 role of catecholamines 348 actions 2 IRMS, testosterone 64 role of glucocorticoids 352 biochemical structure and molecular IRR 159, 161 levobunolol 49, 50 biology 283–4 IRS family 160, 161, 163 levonorgestrel 254 in chronic illness 297 624 index

luteinizing hormone (LH) (cont.) regulation in muscle by exercise and N-methyl-N- circulating concentrations 3 insulin 398–401 trimethylsilyltrifluoroacetamide 61 deficiency 285–6 marathon running 602–3, 608 methyltestosterone 62, 63 effects of aging in men 290 mass analyzers 57–8 MGF see mechano growth factor in exercise-associated menstrual mass spectrometry (MS) 36–7, 40, 50 mineralocorticoids 2, 9, 10, 200 disturbance 238, 239–45 hCG analysis 41 Minnesota caloric deprivation experiment in exercise–cold stress 500, 507 ionization techniques 55–6, 57 295, 297 in female athletes 233–4, 235, 421 mass analyzers 57–8 MIT 11 fetal 288–9 sample preparation and purification mitogen and stress-activated kinase-1/2 in male reproductive axis 279, 284–5, 58–62 398 286, 319, 325–6 Mauriac syndrome 101 monoclonal antibodies in immunoassay effects of training 516 MBHFB 61 43, 44 feedback regulation 286–8 MBTFA 61 monocytes/macrophages 346 ontogenesis of secretion during mechano growth factor (MGF) 174, 181, in delayed onset muscle soreness 361 phases of life 288–90 182 response to exercise 354, 355–7 in menstrual cycle 233, 250, 253 expression 182–3 response to training 358–9 in overtraining 386 gene transfer 187 monoiodotyrosine 11 response to altitude 457 in muscle 183–4 mood response to cold exposure 507 and muscle hypertrophy 310–11 alteration by exercise 149–50, 558 response to exercise 514 in muscle regeneration 189, 190 and overtraining 558–9 role in testosterone secretion 71, 284, signaling pathway involvement 190 role of growth hormone 558 290–1, 319, 526 MEK1/2 398 morphine 48, 48 secretion 233, 250, 251, 253 melatonin 13, 143 motivation and androgenic-anabolic regulation 280–3 menarche 233 steroids 535–7 in strength/power competitions 604 delayed 261, 263–4, 269, 298, 516 motor control, measurement of maturation training effects 71, 72, 73, 74 and nutritional status 263, 513 31, 32 and weight loss 293 see also puberty mountain sickness 447, 458 wrestlers 519, 520 menopause MRI, in exercise testing 27, 28 luteinizing hormone–chorionic calcium supplements following 419 MRS, in exercise testing 27, 29 gonadotropin receptor 284, 285 effects on bone remodeling 412 MS see mass spectrometry luteinizing hormone receptors 250, 251, and osteoporosis 267 MSK1/2 398 284, 285 menstrual cycle MSTFA 61 mutation 290–1 and body temperature 471–2 mTOR 389, 390 lutropin see luteinizing hormone effects of anabolic steroids 74 Mullerian inhibiting substance 288 lymphocytes 346 effects on AVP 492–3 muscle activation 374–5 effects of psychosocial stress 74 action of insulin in 388–90 response to exercise 146, 353, 355–7, female athletes 233–6 effects of exercise-induced damage 375–6 follicular phase 233, 251–3 397–8 trafficking 348 interaction of factors affecting 75 mimicry and enhancement by exercise length 232–3 390–8 M6P receptor 157, 161, 166 luteal phase 233 adaptation 306 M-cadherin 189 regulation 232–3, 250–3 angiogenesis 312 macrophage colony-stimulating factor and response to thermal stress 258 atrophy 307 β (M-CSF) 410 risk factors for irregularity in athletes 2-receptors, in overtraining 587–8, 592 magnesium 412 72–3 capillarization 311–12 magnetic resonance imaging, in exercise and thermoregulatory response to cold damage 189–90 testing 27, 28 507–8 immune system involvement 360–2, magnetic resonance spectroscopy, in see also exercise-associated menstrual 383 exercise testing 27, 29 disturbance impact of oral contraceptives 257–8 mammalian target of rapamycin 389, 390 menstruation 233, 251, 253 and insulin action 397–8 manic behavior in androgenic-anabolic mestranol 254 delayed onset soreness 360–2 steroid use 535–6 metabolic efficiency of female athletes 236 effects of aging 188 mannitol 49, 50 metabolic rate, resting, in female athletes effects of IGF-I overexpression 186–7 mannose-6-phosphate receptor 157, 161, 235 expression of IGF-I splice variants 166 [Met]-enkephalin 202–3, 467 181–2, 183–4 MAPK pathway 160 in heat acclimation 211 fibers in bone response to loading 417 in identification of peptide F receptor contractile properties 312–14 effects of activation on substrate 207 effects of androgenic–anabolic metabolism 400–1 response to exercise 211 steroids 312–13 and muscle hypertrophy 396 methandienone 527, 532–3, 535, 537 effects of estrogens 313 regulation of gene transcription in N-methyl-N-bisheptafluorobutyramide 61 effects of growth hormone 313 muscle 400 N-methyl-N-bistrifluoroacetamide 61 effects of thyroid hormones 313 index 625

glucose transport in 197, 388–9, 390–2, negative feedback 12–13 in passive intense heat exposure 474, 394–5 long-loop 13 475, 476 glycogen synthesis in 389–90, 390–2, short-loop 13 precompetition levels 601 395–6 neuroendocrine–immune axis 345, 368–9, response to altitude 448–50, 458 hypertrophy 306–11, 320 383–4, 470 response to exercise 194, 209, 348, 370 and androgen receptors 308–9 communications in 370–3 role in leukocyte trafficking 348 androgenic-anabolic steroid in delayed onset muscle soreness 360–2 in strength/power competitions 605 associated 307–8, 533–4 functions of connections 372–3 and sympathetic–adrenal medullary effects of exercise on insulin action mediators and mechanisms of axis 467 396–7 interactions 371–2 synthesis 11, 204–5 resistance exercise-induced 118–19, response to exercise 353–7, 373–81 in thermoregulation 468, 469 183 response to training 357–60 norepinephrine/epinephrine ratio 607 role of fibroblast growth factors role of HPA axis 350–3 norestimate 254 309–10 role of sympathetic nervous system 345, norethindrone 254 role of IGF-I 186, 187–8, 310–11 348–50 norethynodrel 254 and testosterone 307–8, 330–1 and susceptibility to illness 357, 360, norgesterel 254 IGF-I production 182–4 374, 379, 381–3 19-nortestosterone see nandrolone metabolic impairment in amenorrhea neuroendocrine systems 466, 467 NPD 55 238 adaptation to exercise 370 nuclear receptors 4 ontogenic development 320 communications in 369–70 nucleotides 200 in overtraining 583 neuroimmunology 345–67, 368–87 nutrition 426, 427 preconditioning effect of testosterone neuromuscular excitability in overtraining and adipose tissue triacylglycerol 331–4 584 balance 438–9 protein balance 436–8 neuropeptide Y 467, 583 and body composition 435–9 protein synthesis 390, 396–7 in bone response to loading 417 effects on cortisol 432–4 regeneration 189–90 in male reproductive axis 295 effects on GHBP 100 role in growth hormone secretion neuropeptides 466–7 effects on growth hormone 426–9 89–90 neurophysin II 487 effects on IGF-I 411–12, 429–30 strength neurotransmitters 466, 467 effects on insulin 434–5 gains due to androgenic-anabolic in overtraining 588–91 effects on leptin levels 516 steroids 534–5 in thermoregulation 467–72 effects on testosterone 430–2 impact of growth hormone neutrophils 346 in exercise–heat stress 478 administration 548–9 in delayed onset muscle soreness 361 influence on GH–IGF axis 166, 172 impact of oral contraceptives 257 response to exercise 146, 353–4, 355–7 influence on growth 513 testosterone action on 312–13, 320 response to training 358, 359 influence on HPA axis 222–3 myf5 189 trafficking 348 influence on hypothalamic– MyoD 189, 311 nifenalol 49, 50 pituitary–gonadal axis 293–7 myogenin 189, 311 nitric oxide (NO) 466 influence on puberty 513 myonuclei in muscle hypertrophy 306–7 in bone response to loading 417 low-carbohydrate diets 438–9 myosin isoforms 312–14 exhaled 31 and menstrual irregularity 72–3 in thermoregulation 468 ‘steroid replacer’ dietary supplements N-POMC 135, 136 nitrogen phosphorus detector 55 527 N-telopeptide 413 NK cells see natural killer cells wrestlers 519 naloxone, in identification of peptide F NKCA 352, 356, 359, 378 receptor 207 NO see nitric oxide obesity nandrolone 62, 527, 537–8 noradrenaline see norepinephrine and GHBP 102, 171 esters 527 norepinephrine 204, 345, 348, 466 and growth hormone secretion 122, 170 quantitative analysis 65 actions 2, 205 response to acute aerobic exercise 125 nandrolone decanoate 527 in chromaffin cells 201 response to chronic endurance weight gain due to 531 and cold acclimatization 501 training 127 narcotics 48 effects on natural killer cell function OBs see osteoblasts natriuresis and altitude 446 348 OCs see osteoclasts natural killer (NK) cells 346 in endurance competitions 602 oligomenorrhea 72, 74, 264 influence of β-endorphin 341–2 changes through the season 608 oligospermia 74 influence of catecholamines 348, 350 in exercise–cold stress 499, 501 Olympic Movement Anti-Doping Code 40 influence of cortisol 343, 352 in exercise–heat acclimation 481 oocyte 253 response to exercise 353, 356, 377–8 in exercise–heat stress 477–8 opioid peptide families 135 response to training 358, 359 in exertional heat illness 482 oral contraceptives 250, 253–4, 259 natural killer cytolytic activity 352, 356, interaction with immune system 348, and bone mineral density 237 359, 378 349, 350 effects on β-endorphin 147 NE/EPI ratio 607 in moderate intensity competitions 607 first generation 254 nebivolol 49, 50 in overtraining 583–4, 586–7 formulations 255 626 index

oral contraceptives (cont.) oxygen, and altitude 444–5, 446, 447–8 physical conditioning scores and IGF-I 175 impact on aerobic performance 256–7 oxytocin 9, 467 physical fitness, effects on GH–IGF axis impact on anaerobic performance actions 2 169–75 257–8 in childbirth 13 PI3K pathway 160 impact on heat tolerance 258 and effects of exercise on muscle 390 impact on muscle damage 257–8 p38 398, 400 effects of exercise on 163 impact on muscle strength 257 and glucose uptake by muscle 401 increased insulin-stimulated activity mechanism of action 254–6 signaling 400 following exercise 393–4 metabolic effects 256, 258 pain threshold, exercise-induced elevation insulin signaling through 388–9 side effects 250, 254 148 and protein synthesis in muscle 390 orexigens 270 paracrine effects 8, 466 and training effects on insulin action osteoblasts (OBs) 409, 410, 414 parasympathetic nervous system 345, 395 in bone remodeling 411 467 pineal gland 13

role in bone response to loading 415, in overtraining 141–2, 580, 593–4 PIP3 388–9 417 response to altitude 450 Pit-1 283 osteocalcin 412, 413, 569 parathyroid hormone pituitary osteoclasts (OCs) 410 actions 2 activation by stress 147–8 in bone remodeling 411 in bone response to loading 416 anterior osteocytes 409, 410 in treatment of post-menopausal growth hormone-producing cell role in bone response to loading 415, osteoporosis 416 system 86–7 417 paraventricular nucleus 339, 487 growth hormone secretion and release osteopenia in female athletes 237–8, 267–8 PCOS 232, 431 110–12 osteoporosis peptide B 200, 202, 203 interaction with muscle afferents in bone turnover markers in 413 peptide E 200, 202, 203 growth hormone secretion 89–90 female athletes 236–8 peptide F 200, 202–3, 214, 467 model for regulation of hormone male 538 complementary action with epinephrine secretion 4 and menopause 267 204 POMC localization to 136 treatment 416 interaction with immune system 206–7, functional anatomy and development ovarian follicles 233, 250, 251–3 211–12, 214 283 ovaries see physiological role 205–7 intermediate, POMC localization to 136 hypothalamic–pituitary–gonadal ratio with epinephrine 205 message transfer systems 147 axis receptor 207–8, 213 refractoriness 167 overload 578 release 201–2, 203 PKAs 22, 398 overreaching 134, 141 and epinephrine release 203–5 PKB see Akt definition 579 response to exercise 208–11, 214 PKC 22 effects on HPA axis 224–5, 581 in blood biocompartments 211–13 plasma indicators of 607 peptide I 203 osmolality overtraining 150, 578 peptide/protein hormones and AVP release 488 assessment 590, 591 biosynthesis 9–11 response to exercise 489 central fatigue hypothesis 579–80 clearance 15–16, 40 and thirst 488 central regulation 588–91 doping tests for 40–4 volume, response to exercise 489 definition 141, 211, 579 isoforms 39 plasma renin activity (PRA) effects on GH–IGF axis 582–3 measurement 36–46 and altitude 452–3, 458 effects on HPA axis 141, 224–5, 226, 581 pulsatile secretion 40 in exercise–cold stress 500, 504 effects on immune system 359–60 recombinant 40 plasmin 88 endocrine-related mechanisms 580 transport 14 POAH see hypothalamus, preoptic- endurance exercise and training 141–2, pericytes 414 anterior 580–4, 608 periosteum, in aberrant bone remodeling Point Subtraction Aggression Paradigm epinephrine response to 211 413–14 536 indicators of 134, 142, 175, 579, 607 peripheral benzodiazepine receptor 284 polyclonal antibodies in immunoassay 43 and mood 558–9 permissiveness 23 polycystic ovary syndrome 232, 431 overview of responses to 592 peroxisome proliferator activated polysiloxanes, in gas chromatography 51, peptide F response 210, 211 receptors 400 52 physiological mechanisms 579–80 PGC-1 400 POMC see proopiomelanocortin POMC response 141–3 phenylalanine 204 positive feedback 13 resistance exercise and training 142, phenylalanine hydroxylase 204 postpartum state, effects of physical 330, 580, 584–8 phosphate, in bone remodeling 412 exertion 75 short type 581 phosphatidylinositol 22 post-traumatic stress disorder 149 and stress 137 phosphatidylinositol-3-kinase pathway see potentiation 23 overtraining syndrome 134, 579 PI3K pathway power–duration curve 580 ovulation 233, 250, 251, 253 phosphodiesterases 22 PP1 389, 390 oxandrolone 537 phospholipase C 22 PP1-G 401 index 627

PPARγ coactivator-1 400 determination 148–9 Pygmies 162 PPARs 400 functional significance 147–8, 149, 339 pyridostigmine 111 PRA see plasma renin activity exercise-induced release 134 PYY 270 pregnancy acute aerobic and endurance exercise effects on AVP responses 493 137–8 quadrupole analyzers 57–8 GHBP levels 100 acute resistance exercise 138–9 quadrupole ion trap analyzers 58 pregnenolone 9 in cardiovascular disease 143 pre-osteoblasts (pre-OBs) 410 chronic aerobic and endurance race role in bone response to loading 415 exercise 139 and eating disorders 269 preproenkephalin 202–3 chronic resistance exercise 139 and exercise-associated menstrual post-translational processing of ECPs immunological influence 146 disturbance 269 203 influence of age, race and gender influence on POMC and derivatives preprohormones 10 143–4 144 pre-proopiomelanocortin 136 in overtraining/exercise addiction radioactive isotopes, in immunoassay 37 Prl see prolactin 141–3 radio-immunoassay 37, 103 procollagen peptide 413 physiological and pathophysiological radioimmunometric assay 37 prodynorphin 135 aspects 144–6 Raf1 398 proenkephalins 135, 200 training effects 139–40 RANKL 410 actions 2 ultraendurance exercise 140–1 rat tibial line assay 78, 83, 87–8 pro-γ-melanocyte stimulating hormone gene 135 rat weight gain assay 78 135, 136 expression 136 rate limiting enzymes 9 progesterone 9 localization 136 RBCs see red blood cells actions 2 phylogenetic ontogeny 134–6 reaching 581 in amenorrhea 233, 234 stress-induced release 4, 134, 136–7 receptor activator of nuclear factor kappa B in anovulation 264 extreme physical stress 140–1 410 circulating concentrations 3 structure 135, 339 receptors 3–4, 16 effects on AVP responses 493 Prop-1 283 binding to 18 effects on endometrium 233 propranolol 49, 50 chaperone proteins 18 effects of strenuous physical training 71 prostacyclin, in bone response to loading competition at 17 and infertility 265 417 cross-reactivity 16 influence on core temperature 258 prostaglandins desensitization 4

in luteal phase defects 264, 265 PGE2 down-regulation 17 in menstrual cycle 233, 250–1, 253 in bone response to loading 417 effector site activation 18 response to acute exercise 514 in fever 469 intracellular 18–19

response to altitude 457, 459 PGF2α, in fever 469 membrane-bound 18, 19–23 response to cold exposure 508 PGI2, in bone response to loading 417 extracellular region 20 secretion 71 protein intracellular region 20 and thermoregulation during exercise anabolism, role of growth hormone transmembrane regions 20 479 549–50 overexpression (spare) 16–17 progestins 9 balance in skeletal muscle 436–8 redundancy in expression 17–18 in oral contraceptives 254, 255 synthesis in muscle 390, 396–7 saturation 16–17 prohibited compounds 47–9, 50 effects of androgenic–anabolic specificity 16 prohormones 10 steroids 534 types 18–23 proinsulin 184 protein hormones see peptide/protein up-regulation 17 prolactin (Prl) 71, 467 hormones recombinant human growth hormone actions 2 protein kinase C 22 (rhGH) in cold acclimatization 507 protein kinases 22 abuse effects of strenuous physical training 71 stress-activated 398 in childhood 544 in exercise–cold stress 500, 506–7 protein phosphatase-1 389, 390 evidence 40, 544 in exercise–heat acclimation 481 glycogen-bound form 401 and mood 559 in exertional heat illness 482 protein supplements 438 potential side effects 545, 560 in moderate intensity competitions 606 post-exercise 437 rationale 544–5, 545–59 in overtraining 590, 591 PSAP 536 tests for 42–4, 545, 559–70 in passive intense heat exposure 474, psychological effects of exercise 149–50, and thermoregulation 557 475, 476 558 Washout Study 561–7 regulation of secretion 13 puberty associated molecular landmarks 79–81 response to cold exposure 506 delayed 516 chemical labeling 545 response to exercise 144, 370 effects of training on 516–19, 520 comparison between genetic and proopiomelanocortin (POMC) 10–11, 134 and growth 512–13 chemical variants 81 derivatives 134–6, 339 in gymnasts 518 effects on blood 558 correlations with reproductive influence of nutrition 513 effects on body composition 546–8 hormones 146–7 role of leptin 263, 513 effects on bone growth 546 628 index

recombinant human growth hormone gender effects 117–18 sex steroids 9, 70, 71 (rhGH) (cont.) specificity of exercise 118 SHBG see sex hormone binding globulin effects on carbohydrate metabolism training adaptations 117 Shc adaptor protein family 160, 161 551–2 increase in training intensity and power sickness behavior 470 effects on cardiovascular system 553–4 586–7 side-chain cleavage enzyme 284 effects on cholesterol metabolism and increase in training volume 585–6 skeleton vascular health 552–3 and insulin action in muscle 397–8 responses to acute exercise 414–18 effects on lipolysis and lipid oxidation intensity level 321–4 responses to training 418–22 550–1 monitoring 327–30 two compartment model 409–10 effects on muscle strength 548–9 overtraining 142, 330, 580, 584–8 see also bone effects on protein anabolism 549–50 peptide F response to 210–11, 212–13 soccer 606–7, 609 effects on renal structure and function physiological impact 118–19 sodium 554–6 POMC response to 138–9 effects of growth hormone 555 effects on reproductive hormones 74 power sessions 324 effects of testosterone on reabsorption immunoassay 39 taper stage 329–40 320 interaction with growth hormone training effects 324–5, 328–34 excretion at altitude 446 receptors 79 reticulocytes and altitude 445 solid–phase extraction 59 linear amino acid sequence 80 reversed phase chromatography 53–4 somatomedin see insulin-like growth primary structure 79–81 rhEPO see erythropoietin, recombinant factor-I therapy 181, 544, 546 human somatopause 116–17 adults 546–7 rhGH see recombinant human growth somatostatin (SS) 9, 165, 467 red blood cells (RBCs) hormone role in GH release 110–12, 116, 122 and altitude 445–6, 452, 458 RIA 37, 103 exercise-induced 127–30 effects of androgenic-anabolic steroids robots, use in exercise testing 30–1 somatotrophs 110–11 537 RPC 53–4 desensitization 111 re-extraction 59 RSK2 398, 401 peripheral feedback regulation 112 rehydration knockout mice 401 somatotropin see growth hormone and AVP response to exercise 491 RSK3 401 space flight 408, 413

during exercise 479 RUNX2 410 SPE 59 renal acid–base metabolism, effects of speed skating 609 growth hormone on 556 S6 protein kinase 389, 390 spermatogenesis 279, 285–6 renin 503 and muscle hypertrophy 396 spermiogenesis 285 in exercise–cold stress 500, 504 salbutamol, quantitative analysis 64–5 spironolactone 49, 50 response to altitude 452–3 SAM axis 467 splanchnic nerve 200 response to cold exposure 504 sample preparation and purification sprint competitions 604–5 response to exercise 370 58–62 SS see somatostatin reproductive axis see sapogenins 527 STAR 284 hypothalamic–pituitary–gonadal sarcopenia and IGF-I 188 STAT-3, in IGF-IR signaling 161 axis satellite cells 188–9, 190 STAT-5, in IGF-IR signaling 161 reproductive hormones 70, 71 androgen receptors 308 steroid hormones 9 correlations with POMC derivatives IGF-I expression 310 assay 39, 40 146–7 and muscle hypertrophy 307, 308 biosynthetic pathways 10 effects of strenuous physical training second messenger systems, G protein- clearance 16 71–4 associated 21–3 transport 14–15 in exercise–cold stress 500, 506–8 secretin 1, 3, 8 ‘steroid replacer’ dietary supplements 527 and menstrual regulation 233 serine/threonine kinase receptors 4 steroidogenesis acute regulatory protein response to altitude 457, 458–9 serotonin 466, 506 284 and thermoregulation during exercise in exercise–heat acclimation 481 sterols 527 479 in exertional heat illness 482 stimulants 48 see also specific hormones in overtraining 579–80, 588–90 strain, effects on bone 415 resistance exercise and training 5–6, 110 in thermoregulation 469 stress AVP response to 490–1 serotonin receptors 467 acute, biochemical definition 137 duration 323–4 in thermoregulation 469 adaptation to 345, 360, 578–9, 589 effects on IGF-I levels 169, 173, 174, 183, Sertoli cells 279, 285, 286, 291 chronic, biochemical definition 137 513 sex hormone binding globulin (SHBG) 3, effects on neurotransmitters 589–90 effects on male reproductive axis 299, 255 effects on testosterone secretion 528 321–34, 529–30 and aging 289 extreme physical, POMC response effects on satellite cells 307 concentrations 291 140–1 and growth hormone secretion 114–19, in endurance competitions 603 forms of 217 173, 513, 515 in overtraining 586 negative (distress) 375, 528 acute vs. chronic exercise 90, 91 testosterone binding 291, 319, 526 and overtraining 137 effects of age 116–17 in wrestlers 520 and pituitary activation 147–8 index 629

POMC release due to 4, 134, 136–7, anabolic 319–20, 526 in women 320–1 140–1 anti-catabolic 320 wrestlers 519, 520 positive (eustress) 375, 528 mechanism 292–3 testosterone buciclate 527 preceding competitions 600–1 metabolic 319–20 testosterone/cortisol (T/C) ratio psychosocial, effects on menstrual in acute/chronic illness 297–8 in endurance competition seasons 608 function 74 and aggression 601 in overtraining 580, 584–5, 587, 588, 592 and reproductive disorders 239–41 and aging 289–90 in strength/power competitions 604, response 4–5 binding and transport 291, 319, 526 609 HPA axis 136, 350, 352, 370 blood level 319 testosterone cypionate 527, 531 level 149 effects on bone growth and calcium testosterone enanthate 527 thermal 258 retention 320 effects on body composition 533–4 stress-activated protein kinases 398 effects on muscle fibers 312–13, 320 effects on body weight 531 stresses (physical), effects on bone 415 effects of nutrition and feeding on effects on endurance performance 537 stroke, exercise testing following 29–31 430–2 effects on strength 535 stroke volume and altitude 446 in endurance competitions 603 testosterone/epitestosterone ratio 64 strontium 412 changes through the season 608 3α-testosterone glucuronide 526 supercompensation cycle 581 esterification 527 testosterone undecanoate 527 suprachiasmatic nucleus 14 in exercise–cold stress 500, 507 tetrahydrogestrinone 62–3, 527–8 supraoptic nucleus 487 feedback regulation of LH and FSH TGF-β 288, 347 sweating 472 286–7 theca cells 250, 251, 253 in passive intense heat exposure 473 intratesticular levels 286 thermogenesis 472–3 regulation 472 metabolism 291, 526–8 diet-induced 472 role of growth hormone 556–7 in moderate intensity competitions non-shivering 471, 473 swimming 608, 609 607 shivering 471, 499, 505 sympathetic–adrenal medullary axis 467 modifications to produce anabolic and cold acclimatization 501 sympathetic nervous system 467 steroids 49 thermoregulation 466, 481–3 interaction with immune system 345, and muscle hypertrophy 307–8, 330–1 and exercise protocol 477–8 348–50 over-the counter alternatives to 527 in exercise–heat stress 476–81, 482 in overtraining 141, 580, 593–4 in overtraining 580, 584–6, 587, 592 and heat acclimatization/acclimation response to altitude 448–50, 457–8 postprandial response 431–2 479–80, 481 response to cold 499 precompetition levels 600, 601 in humans 472–3 synergism 23 preconditioning effect 331–4 influence of diet and hydration status synexin 201 as prohormone 291–2 478–9 qualitative analysis 63–4 influence of gender 479

T3 see triiodothyronine replacement therapy 538–9 and oral contraceptive use 258 T4 see thyroxine response to altitude 457, 459 in passive intense heat exposure 473–6 T/C ratio see testosterone/cortisol ratio response to cold exposure 507 role of GH–IGF axis 556–7 T/E ratio 64 response to exercise role of neurotransmitters, animal studies T-lymphocytes 211 acute exercise 514 467–72 activation 377 effects of feeding on 432 training effects 480 apoptosis 377 endurance exercise and training THG 62–3, 527–8 cytotoxic 346, 374, 375–6, 376–7 298–9, 516, 528–30 thin-layer chromatography 50 helper 346, 374, 376–7 modulating factors 325, 326–7 thirst 487, 488 influence of β-endorphin 341 monitoring of training 327–30 3T3-F422A adipocyte assay 79 influence of cortisol 352–3 recovery period 327 thrombosis and oral contraceptive use interaction with peptide F 206, 207, resistance exercise and training 299, 254–5 211–12 321–34, 529–30 thyroglobulin 11 response to exercise 356, 375–6, 376–7 significance of exercise intensity, thyroid hormones response to training 358, 359 duration and rest intervals 321–4 clearance 15, 16 TACE 96, 97, 100, 104 suppression 528–30 synthesis 11 target organ 1 training effects 324–5, 328–34 see also thyroxine; triiodothyronine TBG 4 in women 326, 514 thyroid-stimulating hormone (TSH) 4 temperature secretion 71, 290–1, 319, 526 actions 2 effects on AVP responses 494 environmental regulators 528 in exercise–cold stress 502 and thermoregulation 468 regulation 284 response to altitude 456 see also body temperature; 17-alkylation 527 thyrotropin-releasing hormone 4, 467

thermoregulation and spermatogenesis 285, 286 thyroxine (T4)3 tennis 607 in strength/power competitions 604–5, actions 2 testes 279 606 in cold acclimatization 502–3 atrophy 526, 538 changes through the season 609–10 and cold sensitivity 503 testosterone 3, 9, 49, 319 structure 49 in exercise–cold stress 500, 502 actions 2, 279, 526–8 and testicular function 279 in exercise–heat acclimation 481 630 index

thyroxine (T4) (cont.) transthyretin 14–15 urine regulation 4 trenbolone 62 analytical techniques 49–58 response to altitude 456–7 TRH 4, 467 in doping tests 40 response to exercise 370 triacylglycerol balance in adipose tissue sample preparation and purification synthesis 11 438–9 58–62 thyroxine binding protein 4 triathlon competition 602–3 urodilatin 504 Title IX 261, 515 triglycerides, effects of oral contraceptives URTI 357, 360, 374, 379, 557–8 TLC 50 255, 258

TMIS 61 triiodothyronine (T3)3 V3-receptors 217 TMSImi 61 in cold acclimatization 502–3 vaginal dryness 238 TNF-α see tumor necrosis factor-α effects on muscle fibers 313 vagus nerve 467 tracers 37 in exercise–cold stress 500, 502 Van Deemter equation 53 training 514 in female athletes 234–5 vascular endothelial growth factor (VEGF) avoidance of reproductive disorders in passive intense heat exposure 474 and acute mountain sickness 447 244–5 regulation 4 and muscle capillarization 311–12 AVP response to 489–90 response to altitude 456–7 vascular health, effects of growth hormone effects on adrenomedullary activity synthesis 11 552–3 195 trimethyliodosilane 61 vasoactive intestinal peptide 467 effects on GH–IGF axis 169–75, 515 trimethylsilyl-imidazole 61 VEGF see vascular endothelial growth effects on growth and puberty 516–19, trypsin 202, 204 factor 520 tryptophan in overtraining 579–80 ventilatory adaptation to altitude 445 effects on HPA axis 140, 223–4, 581 TSH see thyroid-stimulating hormone ventilatory responsiveness and β- effects on tumor necrosis factor-α (TNF-α) 347 endorphin 138 hypothalamic–pituitary–gonadal in delayed onset muscle soreness 361 VIP 467 axis 71–4, 515–16 in exercise training 172–3 vitamin D 2, 412

effects on immune system 357–60 in fever 469 vitamin D3 3 effects on insulin action in skeletal interaction with HPA axis 352 vitamin K 412 muscle 394–5 response to exercise 354, 381 effects on leptin 516 tumor necrosis factor-α converting Washout Study 561–7 effects on POMC and its derivatives enzyme (TACE) 96, 97, 100, 104 water excretion at altitude 446 139–40 tyrosine 11, 204 weightlessness 408, 413 indicators of load 175 tyrosine hydroxylase 11, 204 weightlifters 610 skeletal responses to 418–22 tyrosine kinase receptors 4, 20 Wnts 415 tapering intensity 175, 608 wortmannin 390 and thermoregulation during exercise ubiquitin 361 wrestlers 480 ultradian rhythms 13 during competition 605, 609–10 see also endurance exercise and training; ultraendurance exercise 140–1, 603 growth and maturation 518–20, 521 resistance exercise and training ultraviolet absorbance detector 55 transcription regulatory factors 19 upper respiratory tract infection 357, 360, yohimbe bark 527 transforming growth factor-β 288, 347 374, 379, 557–8 zinc fingers 19