INFORMATION TO USERS

This manuscript has been reproduced from the microfilm master. UMI films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may t>e from any type of computer printer.

The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistsict print, colored or poor quality illustrations and photographs, print Meedthrough, substandard margins, and improper alignment can adversely affect reproduction.

In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will b e noted. Also, if unauthorized copyright material had to t>e removed, a note will indicate the deletion.

Oversize materials (e.g., maps, drawings, charts) are reproduced t)y sectioning the original, beginning at the upper left-hand comer and continuing from left to right in equal sections with small overlaps.

Photographs included in the original manuscript have been reproduced xerographically in this copy. Higher quality 6" x 9" black arxf white photographic prints are available for any photographs or illustrations appearing in this copy tor an additional charge. Contact UMI directly to order.

Bell & Howell Information and Learning 300 North Zeeb Road, Ann Arbor, Ml 48106-1346 USA

UIVLI800-521-0600

THE RECOVERY OF FUNCTION AFTER SPINAL CORD CONTUSION INJURY IN RATS: SKELETAL MUSCLE, NEURAL, AND BEHAVIORAL ADAPTATIONS WITH AND WITHOUT EXERCISE

DISSERTATION

Presented in Partial Fulfillment of the Requirements for

the Degree Doctor of Philosophy in the Graduate

School of The Ohio State University

By

Karen J. Hutchinson, M.S., P.T.

*****

The Ohio State University 2000

Dissertation Committee:

Dr. D. Michele Basso, Adviser Approved by

Dr. James S. King, Co-Adviser

Dr. Jacqueline C. Bresnahan Adviser

Dr. Jon K. Linderman Neuroscience Graduate Program UMI Number 9962402

UMI*

UMI Microform9962402 Copyright 2000 by Bell & Howell Information and teaming Company. All rights reserved. This microform edition is protected against unauthorized copying under Title 17, United States Code.

Bell & Howell Information and Learning Company 300 North Zeeb Road P.O. Box 1346 Ann Arbor, Ml 48106-1346 ABSTRACT

Spinal cord injury (SCI) can result in significant and long lasting functional

motor deficits. In our first study, we evaluated one parameter that may influence motor

recovery following spinal cord injury; hindlimb (HL) skeletal muscle. Changes in

skeletal muscle properties appear to parallel the functional status of the animal. For

example, early after SCI, the animal presents with significant hindlimb paralysis/

paresis, as well as, significant muscle atrophy. Over a 10 week period, muscle atrophy

is attenuated corresponding to improved HL movement and locomotor activity. In a second study, we examined the effect on locomotion of three exercise paradigms designed to maximize rhythmicity, weight bearing (WB) or both in the hindlimbs after

SCI. Groups were trained to swim (rhythmicity, no WB), stand (WB, no rhythmicity), and treadmill step (rhythmicity and WB). Early significant locomotor improvements were identified for swim and treadmill training. Surprisingly, greater residual deficits occurred after stand training, suggesting that WB alone is insufficient to improve locomotion. Furthermore, SCI heightens reflex responses and lowers thresholds to somatosensory stimuli caudal to the injury. Exercise training (particularly treadmill) appeared to normalize sensation as shown during flexor withdrawal and von Frey hair testing. Engaging in early, moderate exercise after SCI also did not exacerbate lesion size as has been described for brain injury. Lastly, immunohistochemical and gel electrophoresis techniques determined that these exercise induced sensory and motor changes were not accompanied by changes in muscle fiber composition or size of a locomotor-related muscle of the hindlimb, the slow twitch soleus.

lU DEDICATION

I would like to dedicate this work to two very important people in my life. First,

to my soul mate and partner of several years, Edward T. Hutchinson. Your constant

support and encouragement have made the impossible seem probable; the stress seem

manageable; and the goal attainable. This document / achievement would not have

happened without you. Secondly, to my best fiiend 6om childhood, M.M.D., whose

daily struggle with paralysis is overshadowed by her grace and dignity. For 17 years I

have wimessed the strain that this injury places on the personal, professional and social

lives of its victims. The challenge of finding a “cine” should never be defeated by the tribulations associated with the effort.

IV ACKNOWLEDGMENTS

Many thanks go out to several people who helped me to make this publication possible.

To Lesley Fisher who assisted in virtually all aspects of this research, fix>m animal

training to spinal cord tissue processing, I am eternally grateful. Special thanks go to

Dr. B. Stokes and Dr. Lyn Jakeman for use and training on the MCID image analysis

system that allowed me to use sophisticated ‘state of the art’ computer software to

analyze spinal cord and skeletal muscle tissue. Thank you also to Tina Van Meter

whose patience and instruction was invaluable for cryostat sectioning of fresh unfixed

muscle samples and to Amy Tovar whose assistance in animal testing allowed me to

stay “masked” to the animal testing condition. Also, thank you to John Komon for

instruction on the use of the stereologer, to Pat Walters for her dedication to excellent

animal care and to Amy Lindsay for teaching me the von Frey Hair testing protocol.

Dr. Larry Sachs and Dr. John Buford were invaluable resources for the data analyses

included in these studies. I would also like to thank my committee; Dr. J. Linderman,

Dr. J. Bresnahan, and Dr. J. King for giving me faith and respect for the academic process. Finally, a special thanks goes out to my advisor. Dr. Michele Basso, for her vision, inspiration and constant dedication to the cause. It was through the course of many fruitful discussions that lead to the development of the ideas contained in this document. VITA

June 4 ,1 9 6 4 ...... Bom in Boston, MA

1982-1986 ...... B.S., Physical Therapy The University of Lowell, Lowell, MA

1986 —1990...... Staff Physical Ther^ist/Resource Clinician Inpatient and Outpatient NeurorehabiUtation Braintree Hospital

1990-1993...... M.S., Physical Therapy Division of Physical Therapy, School of Allied Medicine The Ohio State University, Columbus, OH

1993-199 4...... Graduate Research Associate Neuroscience Program, Department of Psychology The Ohio State University, Columbus, OH

1994-199 6...... Graduate Research Associate Neuroscience Program, Department of Psychobiology The Ohio State University, Columbus, OH

1996-Present...... Graduate Research Associate Neuroscience Program, Department Cell Biology, Neurobiology and Anatomy The Ohio State University, Columbus, OH

PUBLICATIONS

Research Publication

1. DS Nichols, TMGlenn, KJ Hutchinson. Changes in the mean center of balance during balance testing in young adults. Physical Therapy, 75(8): 699-706, (1995).

VI Published Abstracts

1. K Hutchinson, J Vensel, S Obermiller, L LeDufF, L Fisher, DM Basso. Early moderate exercise training after spinal cord injury does not exacerbate lesion size and attenuates changes in hindlimb sensation in the rat. Society for Neuroscience, 25:110, (1999).

2. KJ Hutchinson, LC Fisher, MB Johnson, J Linderman, DM Basso. Moderate exercise training improves locomotor recovery and does not exacerbate lesion size when implemented early after spinal cord injury in rat. Neurology Report, 22(5):181- 182, (1998).

3. CA Voll, Jr, LC Fisher, KJ Hutchinson, DM Basso. The severity of hypeireflexia is greater after moderate vs. mild experimental spinal cord injury in rats. Neurology Report, 22(5): 174, (1998).

4. MJ Crowe, KJ Hutchinson, DM Basso. Neurotrophin and insulin-like growth factor-1 mRNA expression in rat hindlimb muscles after spinal cord injury. Society for Neuroscience, 24(1):1149, (1998).

5. K Hutchinson, L Fisher, J Linderman, DM Basso. Exercise training improves recovery firom spinal cord contusion injury in the rat. Society for Neuroscience, 24(2): 1415, (1998).

6. K Hutchinson, J Linderman, DM Basso. Relationship between behavioral recovery, muscle structure and muscle ftmction after spinal cord injury in rat. Neurology Report, 21:175, (1998).

7. K Hutchinson, J Linderman, DM Basso. Effects of experimental spinal cord injury on behavior and hindlimb muscle ftmction in rat. Journal of Neurotrauma, 14:782, (1997).

8. K Hutchinson, J Linderman, DM Basso. Effects o f experimental spinal cord injury on hindlimb skeletal muscle in rat. Society for Neuroscience, 23:2190, (1997).

9. K Hutchinson, J Linderman, DM Basso. A comparison of Fischer 344 and FI Hybrid rats for appropriate animal models for neuromotor control and aging. Neurology Report, 20(4):15-16, (1996).

vu 10. K Hutchinson, M Sarter, B Givens. Single unit activity in medial prefirontal cortex of rats performing in a behavioral vigilance task. Society for Neuroscience, 22:1388, (1996).

11. DS Nichols, KJ Hutchinson, LA Colby. Reliability and validity of the CTSIB and Chattex Balance System in the evaluation of individuals with hemiparesis. Society for Neuroscience, 21(2):1203, (1995).

12. B Givens, K Hutchinson, K McMahon. The behavioral and neurophysiological effects of ethanol on sustained attention in rats. Society for Neuroscience, 21(2):1240, (1995).

13. D Nichols, T Glenn, K Hutchinson. The effects of balance training on gait components following stroke. Society for Neuroscience, 20(1):795, (1994).

14. D Nichols, T Glerm, K Hutchinson. Changes in the mean center of pressure related to testing conditions during balance testing. Physical Therapy, 74(5):S115, (1994).

15. D Nichols, K Hutchinson, L Colby. A correlational analysis of stroke patients’ performance on the CTSIB and Chattex dynamic balance unit. Physical Therapy, 74(5):S114, (1994).

FIELDS OF STUDY

Major Field: Neuroscience

VIU TABLE OF CONTENTS

Page

Abstract ...... ii

Dedication ...... iv

Acknowledgments ...... v

Vita...... vi

Table of Contents ...... ix

List of Tables ...... xiv

List of Figures ...... xv

List of Abbreviations ...... xvii

Chapters;

1 Skeletal Muscle Adaptations Following Spinal Cord Contusion Injury in Rat and the Relationship to Locomotor Function: A Time Course Study

1.1 Introduction ...... 1 1.2 Methods ...... 5 1.2.1 Subjects ...... 5 1.2.2 SCI Surgical Procedures ...... 6 1.2.3 Open Field Locomotion ...... 6 1.2.4 Muscle Dissection Procedures ...... 7 1.2.5 Muscle Wet Weight ...... 7 1.2.6 Muscle Physiology Testing ...... 8 1.2.7 Myosin Heavy Chain Expression ...... 9

ix 1.2.8 Spinal Cord Histology ...... 9 1.2.9 Statistical Analysis ...... 10 1.3 Results...... 11 1.3.1 Open Field Locomotion ...... 11 1.3.2 Spinal Cord Lesion Centers ...... 13 1.3.3 Body Weight and Muscle Wet Weight ...... 18 1.3.4 Muscle Physiology ...... 20 1.3.5 Myosin Heavy Chain Expression ...... 22 1.4 Discussion ...... 25 1.4.1 Comparison to Other Models ...... 25 1.4.2 Lesion Development ...... 26 1.4.3 Gross Muscle Atrophy ...... 28 1.4.4 Physiology ...... 30 1.4.5 Muscle Fiber Type ...... 33 1.4.6 Model Relevance to Human SCI ...... 34 1.4.7 Caveats ...... 36 1.5 Conclusions ...... 37 1.6 List o f References ...... 39

A Comparison of Exercise Interventions on Recovery from Spinal Cord Contusion Injury in Rats: I. Behavioral Perspective

2.1 Introduction ...... 47 2.1.1 Weight Bearing ...... 49 2.1.2 Rhythm Generation ...... 50 2.1.3 Rationale for Treatment Selection ...... 51 2.1.4 Onset of Training ...... 53 2.2 Methods ...... 54 2.2.1 Subjects ...... 54 2.2.2 SCI Surgical Procedures ...... 56 2.2.3 Exercise Training Paradigms ...... 56 2.2.3.1 Task Acquisition ...... 56 2.2.3.1.1 Treadmill...... 56 2.2.3.1.2 Swimming ...... 57 2.2.3.1.3 Standing ...... 57 2.2.3.2 Daily Exercise Training ...... 58 2.2.3.2.1 Treadmill...... 58 2.2.3.2.2 Swimming ...... 59 2.2.3.2.3 Standing ...... 60 2.2.4 Behavioral Testing ...... 61 2.2.4.1 Locomotion ...... 61 22.4.2 Walkway Analysis ...... 62 2.2.4.3 Tactile Sensitivity ...... 63 2 2.4.4 Segmental Reflex Testing/ Flexor Withdrawal 64 2.2.5 Spinal Cord Histology ...... 66 2.2.6 Lesion Volume ...... 67 2.2.7 Muscle Dissection Procedures ...... 67 2.2.8 Statistical Analysis ...... 67 2.3 Results ...... 68 2.3.1 Open Field Locomotion ...... 68 2.3.2 Walkway Locomotion ...... 70 2.3.3 Flexor Withdrawal ...... 74 2.3.4 von Frey Hair Testing ...... 76 2.3.5 Lesion Epicenter ...... 80 2.3.6 Lesion Volume ...... 84 2.4 Discussion ...... 84 2.4.1 Locomotion ...... 85 2.4.2 Neural Control of Coordinated Locomotion after Partial Lesions to the Spinal Cord ...... 86 2.4.3 Walkway Analysis ...... 88 2.4.4 Spinal Cord Learning ...... 89 2.4.5 Treadmill vs. Overground Locomotion ...... 90 2.4.6 Specificity of Training ...... 91 2.4.7 Reflex Testing and Somatosensory Examination ...... 92 2.4.8 Evidence of Hypersensitivity Caudal to SCI ...... 94 2.4.9 Relationship between Flexor Withdrawal and von Frey Hair Testing ...... 95 2.4.10 Outcome Measures for AUodynia ...... 96 2.4.11 Flexor Withdrawal Movement Response ...... 97 2.4.12 Mechanisms ...... 98 2.4.13 Peripheral vs. Central Effects of Training ...... 100 2.4.14 Lesion Volume ...... 102 2.4.15 Caveats...... 103 2.5 Conclusions ...... 104 2.6 List of References ...... 106

XI A Comparison of Exercise Interventions on Recovery from Spinal Cord Contusion Injury in Rats: H. Skeletal Muscle Effects

3.1 Introduction ...... 121 3.2 Methods ...... 126 3.2.1 Subjects ...... 126 3.2.2 Pre-Training ...... 126 3.2.3 Daily Training ...... 127 3 .2.4 SCI Surgical Procedures ...... 127 3.2.5 Muscle Dissection Procedures ...... 127 3.2.6 Muscle Wet Weight ...... 127 3.2.7 Myosin Heavy Chain Expression ...... 128 3.2.8 hmnunohistochemistry ...... 128 3.2.9 Statistical Analysis ...... 131 3.3 Results...... 131 3.3.1 Open Field Locomotion ...... 131 3.3.2 Muscle Wet Weight Values ...... 132 3.3.3 Cross-Sectional Area ...... 134 3.3.4 hmnunohistochemistry ...... 135 3.3.5 Gel Electrophoresis and MHC Expression ...... 143 3.4 Discussion ...... 144 3.4.1 Muscle Fiber Atrophy ...... 144 3.4.2 Application of Spinal Cord Contusion Injury for Analysis of Skeletal Muscle Change ...... 146 3.4.3 Wet Weight vs. Cross-Sectional Area ...... 148 3.4.4 Exercise and Muscle Atrophy ...... 148 3.4.5 Percent Fiber Type Expression ...... 151 3.4.6 The Gastrocnemii Muscles and Locomotion ...... 151 3.4.7 Caveats ...... 153 3.5 Conclusions ...... 153 3.6 List of References ...... 155

XU Appendices

4.1 Appendix A; BBB Subscoring Rules ...... 161 4.2 Appendix B: Task Training ...... 163 4.3 Appendix C: BBB Locomotor Scale ...... 165 4.4 Appendix D; Walkway Locomotion Evaluation Form ...... 168

Bibliography ...... 171

XIU LIST OF TABLES

Table Page

1.1 Mean area measurements of myelin sparing at the lesion epicenter 1, 3 and 10 weeks after moderate SCI or laminectomy ...... 14

1.2 Mean absolute muscle wet weight and body weight at 1, 3 and 10 weeks post SCI compared to CTL ...... 19

1.3 Contractile properties of the SOL muscle ftom SCI and CTL animals at I, 3 and 10 weeks ...... 21

1.4 Myosin heavy chain expression for the soleus and extensor digitorum longus muscles fi^om SCI and CTL groups at 1, 3 and 10 weeks ...... 23

2.1 Improvement in stepping time and belt speed for each rat during treadmill locomotion on training sessions 1, 5, and 29...... 59

2.2 Improvement in swimming performance as measured by total swimming time on training session 1, 5 and 29 for each animal ...... 60

2.3 Improvement in standing performance for each rat on training session 1, 5 and 29. Time spent standing on the hindlimbs and the maximal reward height reached ...... 61

3.1 Identification of muscle fiber type using monoclonal antibodies to I, Ua, Hb and IIx myosin heavy chain isoforms ...... 130

3.2 Mean muscle wet weight values for the extensor digitorum longus, plantaris, soleus, medial gastrocnemius, lateral gastrocnemius, tibiahs anterior and heart for each group ...... 133

3.3 Mean cross sectional area for type I, Ua, Db and IIx soleus muscle fiber for each group ...... 134

XIV LIST OF HGURES

Figure Page

1.1 Mean BBB locomotor scores for SCI and CTL groups over time ...... T2

1.2 Photomicrographs of myelin stained spinal cord sections through the lesion epicenter at 1, 3 and 10 weeks post SCI ...... 15

1.3 Myosin heavy chain protein assays for the soleus and extensor digitorum longus muscles of injured and control animals at 1,3 and 10 weeks ...... 24

2.1 Mean BBB locomotor scores over 49 days post SCI ...... 69

2.2 Ranking of walkway locomotion for all groups 7 weeks after moderate spinal cord contusion ...... 71

2.3 The percentage of animals in each group that demonstrated consistent stepping during walkway locomotion 7 weeks post SCI ...... 72

2.4 Stimulus intensity of pinch needed to elicit flexor withdrawal of the hindlimb ...... 75

2.5 Time to peak flexion for flexor withdrawal at 7 weeks post SCI ...... 77

2.6 Summed angular excursion of the hip, knee and ankle joints during the flexion phase of flexor withdrawal at 7 weeks post S C I ...... 78

2.7 Fifty percent response thresholds for von Frey Hair stimulus to the hind paw of rats ...... 79

XV LIST OF FIGURES (continued)

Figure Page

2.8 Photomicrographs of myelin stained spinal cord sections through the lesion epicenter ...... 81

2.9 Percent spared white matter at the lesion epicenter ...... 82

2.10 Volume of spared myelinated tissue through a 1 cm segment of the spinal cord containing the lesion epicenter ...... 83

3.1 Mean percentage of muscle fiber types in the soleus for each group as identified fiom immunohistochemistry for myosin heavy chain expression ...... 136

3.2 Myosin heavy chain antibody staining in the soleus ...... 138

3.3 Mean percentage of myosin heavy chain isofbrm expression firom soleus muscle homogenate using SDS-PAGE for all groups ...... 139

3.4 Photomicrograph of a type IIx muscle fiber fi’om the soleus of an SCI No-Ex animal as shown by myosin heavy chain antibody staining ...... 141

3.5 Myosin heavy chain expression in the soleus muscle firom representative an im als in each group using SDS-PAGE ...... 142

XVI LIST OF ABBREVIATIONS

Ab antibody

ANOVA analysis of variance

A4.74 mAb to Type Ha and IIx skeletal muscle fibers

A4.951 mAb to Type I skeletal muscle Eber

BBB Basso, Beattie, Bresnahan Locomotor Rating Scale, used to analyze gait in animals following spinal cord injury

C a^ calcium ion

CPG central pattern generator

CNS central nervous system

CSA cross-sectional area

CTL controls

DAB diaminobenzidene

DPO days post op

EDL extensor digitorum longus

EMG electromyography

EX exercise

FL forelimb

F W flexor withdrawal

xvu HL hindlimb

HLS hindlimb suspension

LAM laminectomy (control)

LFB luxol fast blue stain for myelin

LG lateral gastrocnemius (plantar flexor muscle)

Lo optimal length for physiological testing mAb monoclonal antibody

MG medial gastrocnemius (plantar flexor muscle)

MHC myosin heavy chain (muscle protein)

N2.261 mAh to type I and Ua skeletal muscle fibers

N3.36 mAb to type Ua, Ub, and Ux skeletal muscle fibers

OFL open field locomotion

PL plantaris (plantar flexor muscle)

Po peak tetanic tension (maximum force)

Pt peak twitch (force)

RT relaxation time

SCI spinal cord injury

SDS-PAGE sodium dodecyl sulfate-polyacrylamide gel electrophoresis

Sem standard error of the mean

SOL soleus (plantar flexor muscle)

xvm ST stand training abbreviation

SW swim training abbreviation

TA tibialis anterior (dorsiflexor muscle)

TCSA total cross-sectional area

TM treadmill training abbreviation

T8 thoracic vertebrae 8

VFH von Frey Hair

WW walkway

V ï RT one half reaction time, time for peak force output to decrease in half

I, Ha, myosin heavy chain proteins expressed in skeletal muscle Ex, nb

XIX CHAPTER 1

SKELETAL MUSCLE ADAPTATIONS FOLLOWING SPINAL CORD CONTUSION INJURY IN RAT AND THE RELATIONSHIP TO LOCOMOTOR FUNCTION: A TIME COURSE STUDY

Introduction

Animal models of spinal cord contusion injury, including the paradigm used at

The Ohio State University, reproduce the histopathologic sequellae of spinal cord injury (SCI) observed in approximately 40% of human trauma cases (Bunge et al.,

1993). Along with extensive tissue damage, the primary injury at the contusion site consists of a hemorrhagic necrosis occurring centrally in the cord (Balentine, Paris,

1978a; Balentine, Paris, 1978b). An exacerbation of the initial insult, referred to as the development of a secondary injury, involves further tissue and cell loss. Damage progresses radially and rostro-caudally within the cord and continues for several weeks following the initial trauma (reviewed in Schwab, Bartholdi, 1996; Balentine,

Paris, 1978b; Bresnahan, 1978; Bresnahan et al., 1976; Beattie et al., 1997; Crowe et al., 1997).

The locomotor outcome following experimental spinal cord contusion injury has been carefully documented using a standardized test for open field locomotion developed by Basso, Beattie and Bresnahan (BBB locomotor rating scale. Basso et al., 1995; Basso et al., 1996). For example, following a moderate thoracic SCI, an initial period of h in d lim b (HL) paralysis occurs. Within the first few weeks, animals show some movement of individual joints of the HLs and then subsequent weight support and stepping behaviors (Basso et al., 1995; Basso et al., 1996). The impact of this injury on HL skeletal muscle, however, has not been previously elucidated in this model system.

Two predominant animal models used to assess skeletal muscle adaptation following central nervous system (CNS) injury or changes in weight bearing conditions are the complete spinal cord transection (TX) model (Roy et al., 1991;

Talmadge et al., 1995a) and the hindlimb suspension (HLS) model (Thomason et al.,

1989; Thomason, Booth, 1990; Thomason et al., 1987a; Thomason et al., 1987b;

Wakatsuki et al., 1995; Jânkâlâ et al., 1997; Kandarian et al., 1992; Fauteck,

Kandarian, 1995; Diffee et al., 1991; Pierotti et al., 1990). The TX model mechanically disrupts all neuropil at the injury site and eliminates the transmission of descending and ascending information between the caudal cord and the brain. The

TX model differs firom the more clinically relevant contusion injury model in two ways. First, the contusion results in a peripheral rim of spared tissue (myelinated axons) permitting some communication between supraspinal centers and caudal sections of the spinal cord (Basso et al., 1995). Secondly, following TX anim als experience a complete loss of volitional locomotor capabilities (Berhmann et al.,

1992; Basso et al., 1996) whereas animals with moderate contusion injury regain some locomotor function without specific training (Basso et al., 1995; Basso et al.,

1996; Berhmann et al., 1992; Lovely et ai., 1986; Barbeau, Rossignol, 1987).

In the HLS model, the rat tail is used to elevate the HLs off the ground to

prevent weight bearing (Thomason, Booth, 1990; Roy et al., 1991). There is no

anatomical disruption within the peripheral or central nervous system, only an

absence of weight bearing behavior with the HLs. This system is relevant to

conditions with decreased HL weight bearing like bedrest, spaceflight, or paralysis.

Both HLS and spinal cord TX differ fiom peripheral nerve axotomy in which

the motor neuron is anatomically disconnected fiom the muscle fibers it innervates.

Following peripheral axotomy, pronounced muscle atrophy occurs (Buffelli et al.,

1997; Finol et al., 1981; Kean et al., 1974; Spector, 1985), but the contractile

properties of the soleus (SOL) muscle slow rather than speed up as is commonly seen

after unweighting (Herbert et al., 1988) or TX (Davey et al., 1981). In addition,

denervated muscle fibers show the presence of targetoid fibers (Stilwill, Sahgal,

1977), evidence of fiber type grouping (Hennig, Lomo, 1987; Stilwill, Sahgal, 1977),

and an increased iimervation ratio not commonly witnessed following a spinal cord

T X (C opeetal., 1986).

Following a moderate thoracic spinal cord contusion injury, animals present

with initial HL paralysis, i.e. a “non-weight bearing status”. In subsequent weeks,

HL plantar placement with weight bearing and limited locomotor capabilities are observed (Basso et al., 1996; Basso et al., 1995; Berhmann et al., 1992). We hypothesized that the adaptive responses of the HL skeletal muscle to contusion injury would change through out the recovery period, with early alterations being less

severe than TX but more severe than HLS. Given that some recovery of locomotor

behavior occurs following contusion injury, it was important to determine if changes

in âmction reflected changes in skeletal muscle parameters. This study was therefore

designed to characterize the time course for atrophy and the adaptation of biochemical

and physiological characteristics of skeletal muscle between normal and SCI rats that

occur over a 10 week time period following spinal cord contusion injury. We

evaluated key phases of behavioral function occurring at early (1 week), intermediate

(3 weeks) and late (10 weeks) time points after SCI. We selected the 1 week time

period because animals display significant paralytic and paretic behaviors during the

first week post contusion, similar to those seen after TX, which should be reflected in

changes in skeletal muscle properties. The 3 week time point was chosen because the

animals establish long term locomotor patterns during this period as evaluated on the

BBB scale. The 10 week time point was selected for chronic evaluation because

previous studies have shown that the plateau in locomotor behavior persists at least

through the 10 week time point after SCI (Basso et al., 2000).

Studies have shown that the greatest adaptive response to HLS and spinal cord

TX occurs in slow twitch more than fast twitch muscle fibers, and in extensor more than flexor muscle groups (Talmadge et al., 1995a). Therefore, the SOL muscle, being a slow twitch postural extensor, is firequently evaluated for being particularly vulnerable to changes in weight bearing status. The physiological studies presented here evaluated the SOL muscle for its contractile force and timing characteristics. In addition, wet weight measures of several locomotor muscles: SOL, plantaris (PL),

extensor digitorum longus (EDL), medial gastrocnemius (MG), lateral gastrocnemius

(LG), and tibialis anterior (TA); and, the myosin heavy chain (MHC) composition of

the SOL and EDL were completed. The MHC composition of the EDL was included

to see if the magnitude of change was similar for primarily slow twitch and fast twitch

muscles of similar size. We hypothesized that skeletal muscle atrophy would follow

spinal cord contusion injury and would be reflected in decreased force production

capabilities in the SOL. In addition, we expected that MHC expression would show a

conversion of muscle fiber toward a faster phenotype, particularly early after SCI

during periods of little or no weight bearing.

Methods

Subjects

Forty two female Sprague Dawley rats (weighing 220-300 grams at the start of

the experiment) were randomly assigned to either SCI or control (CTL) groups to

characterize the relationship between changes in muscle fiber phenotype and/or

physiologic function and locomotor ability after SCI. Rats were evaluated at early (1

week, n=14), intermediate (3 week, n=13 ) and late (10 week, n=l 1 ) stages of

functional recovery. Three rats fi*om the 10 week time point were discarded firom the study due to complications associated with the surgery. A subset of animals at each time point underwent physiological testing of the SOL muscle (n=4 rats per group).

Another subset of animals (n=18) was included for biochemical analyses of muscle fiber type (SCI=4, CTL=2 at each time point). In addition, animals served as normal, age- and weight-matched controls or were randomly assigned to sham-operated %e- matched controls.

SCI Surgical Procedures

Moderate SCI was produced using the OSU injury device described in detail elsewhere (Bresnahan et al., 1987; Stokes et al., 1992; Stokes et al., 1995). Briefly, animals received injections of antibiotic (gentocin 0.25 ml sc) and ketamine-xylazine

(80 mg/kg, 10 mg/kg ip) prior to surgery. Removal of the T8 lamina exposed the meningeal surface surrounding the spinal cord prior to placing the rat in a spinal firame for stabilization. The impact probe was lowered onto the dura to a pressure of

3 kdynes before the surface of the cord was displaced 1.1 mm over a 20 ms epoch

(Stokes et al., 1992; Stokes et al., 1995). This displacement produces a moderate SCI

(see histology below). Bleeding was stopped before suturing the incision in layers.

Subcutaneous lactated Ringers’ solution (5 ml) and antibiotic spray were administered afler completion of the surgery. Bladders were manually expressed two to three times daily until spontaneous voiding returned (~ 2 weeks). In addition, oral Vitamin C was given daily to all animals in order to prevent urinary tract infections.

Open Field Locomotion

Prior to surgery, animals were acclimated to the open field locomotor testing procedures. Rats were evaluated for locomotion in a plastic wading pool (90 cm diameter), for 4 minutes by two examiners using the Basso, Beattie, Bresnahan (BBB) Locomotor Rating Scale (Basso et al., 1995). Testing occurred every other day

during the first week and then weekly until sacrifice at 3 or 10 weeks.

Muscle Dissection Procedures

Several muscles associated with locomotion were dissected from the right HL

of each rat at the time of sacrifice. Under deep anesthesia (ketamine 80mg/kg,

xylazine lOmg/kg, ip), the following muscles were removed; the MG, LG, PL, SOL,

TA and EDL. After making a skin incision, the tendinous insertions of superficial

muscle groups were exposed using blunt dissection in order to preserve the blood and

nerve supply to deeper muscle groups. Muscles were removed by separating the

insertions firom the bone, reflecting the muscle back and detaching them fiom their origin. Dissection proceeded firom superficial to deep muscles. In order to separate the MG and LG, muscle attachments were cut along their common aponeurosis.

Muscle Wet Weight

Muscle wet weight served as a gross measure of muscle atrophy. Each muscle was cleaned of fat and excess connective tissue before being weighed. The PL, EDL and TA were then pinned to cork at their in situ length, immediately frozen in isopentane cooled by liquid nitrogen and stored at -80°C until further processing.

The SOL muscle was retained in a subset of animals for physiological testing (see below) before it was weighed and frozen. A subsequent -80°C freezer failure prevented further determination of muscle fiber cross-sectional area (see Chapter 3). Muscle Physiological Testing

Following dissection, the SOL muscle was immediately placed into a 2S°C

oxygenated (95%02,5%C02) Krebs-Ringer bicarbonate solution (137mM NaCl,

5mM KCl, 13 mM NaHCOs, 1.8 mM KHPO 4, 2 mM CaCh, ImM MgS 0 4 , Im M

Glucose, 10 mM caffeine, pH 7.4), supplemented with 25 uM turbocurarine chloride

(Segal, Faulkner, 1985). The distal tendon was attached in series with 2.0 silk to a

force transducer (Model 200B PCB Piezo-electronics, New York, NY) in the

recording chamber, with the proximal end attached to a rigid post. Using a

supramaximal 0.5 ms square wave pulse, (Grass Instrument S48), the muscle

length at which peak twitch (Pt) tension was greatest was considered the optimal

length (Lo) and was used to determine peak tetanic (Po) tension. Time to peak twitch

(tPt) and half relaxation time (14RT) were also recorded. Peak tetanic tension was

evoked by 0.5 ms supramaximal pulses delivered at 20,30,40, 60, 80, and 100 Hz

(400 ms train duration). The duration between stimulus trains was 2 minutes. Data

acquisition occurred by interfacing the Piezo-Electronic force transducer with a

Nicolet 2 channel digital oscilloscope (Model 720B, Oxford, England) along with a

CompuAdd 386 personal computer. On 4 of 23 records, a small portion o f the Pt

trace failed to be recorded. A second order polynomial regression equation was used to estimate the missing portion of each trace. The estimated value was then used to generate 14 RT values. Mvosin Heavy Chain Expression

A I m m frozen section of the midbelly was taken from the EDL and SOL of

SCI and CTL rats 1, 3 and 10 weeks post injury. The muscle tissue was minced with scissors in 9 vol of cold homogenization buffer [250mM sucrose, 100 mM KCl, 5 mM EDTA, and 20 mM tris (hydroxymethl) aminomethane pH 6.8] before homogenizing it with an electric tissue grinder. Total protein content of the homogenate was determined using a BCA kit (Pierce, Rockford, IL) according to manufacturers instructions. Samples were then diluted in Laemmli buffer (Laemmli,

1970) to a final total protein concentration of 0.25 mg/ml. For identification of MHC protein, SDS-PAGE was used according to the methods of Talmadge et al. (1993) and

15-22 pi of sample were loaded per lane. Coomassie stained gels were scanned using a 1-D scanning densitometer (Alpha Imager 2000, San Leandro, CA). Each band was represented as the percentage of the total MHC sampled.

Spinal Cord Histology

The subset of rats used for muscle physiology experiments underwent intracardiac perfusion with saline followed by 4% paraformaldehyde solution before a

1 cm block o f the spinal cord was removed at the lesion or laminectomy site. In all other animals, a 1 cm block of unfixed spinal cord tissue was carefully dissected at the lesion or lamincetomy site and postfixed in 10 % neutral buffered formalin for several days. Unfixed caudal spinal cord tissue was used for an umelated set of experiments. Given that no differences were noted in the percent of myelin sparing measured at the epicenter from perfused and postfixed tissue samples (ANOVA; group X time x fixation, p> .05), the histological data were pooled. All tissue blocks

were stored in fixative until they were embedded in paraffin (m inim um o f 3 days).

Tissue was then cut in the transverse plane (20 pm) with a rotary microtome, and

every fifth section was mounted on gelatin coated slides. Tissue was stained for

myelin using Luxol Fast Blue (LFB), and the lesion center was identified as the

section containing the largest central core lesion with the least myelin stained tissue.

The average location o f the tissue section representing the lesion epicenter was used

as a reference for identifying the “epicenter” in CTL animals. The spared tissue and

total cross-sectional area were measured using a Zeiss Axiophot Microscope attached

to MCID-M4 Image Analysis System (Imaging Research Inc., Ontario, Canada).

Im%es were captured using a Dage CCD72 8 bit camera at 3.125 x and projected in

black and white onto a 38 cm monitor. Digitized tissues sections were m anually

outlined onscreen using a P2 mouse while viewing the section under 20 x

magnification. White matter was considered “spared” if the myelin staining was

dense, contiguous and grossly normal in appearance, with little or no gliosis and few

swollen axons observed (Berhmann et al., 1992). Tissue sparing was expressed as a percent: area occupied by spared white matter per total cross sectional area of the cord measured at the lesion epicenter.

Statistical Analvsis

There were no group difTerences between the normal and sham CTL groups for behavioral performance (BBB scores. Repeated Measures ANOVA), spinal cord histology (% myelin sparing, ANOVA, CTL type x time) or muscle wet weight

10 values (ANOVA: CTL type x time;Xp > 05 for each), therefore CTL group data were

pooled for all analyses. Open field locomotion scores were analyzed using a repeated

measures ANOVA. Three cohorts of injured and CTL animals corresponding to 1,3

and 10 week survival periods were analyzed to preserve equal sampling sizes. Due to

small sample sizes, muscle physiology (Pt, Po, V i RT, tPt) and biochemistry (MHC)

values were analyzed using a Mann-Whitney U test. Spinal cord histology (total CSA

and % sparing) and muscle wet weight values, were analyzed using ANOVA (group x

time) comparing injiued to controls at each time point. Tukey post-hoc tests were

applied where appropriate. Significance was determined a priori at p < .05.

Results

Open Field Locomotion

Control animals showed normal locomotor behavior scoring 21 on the BBB

scale at all times tested (Figure 1.1). Rats with SCI demonstrated early locomotor

deficits with some spontaneous recovery occurring within the first 3 to 4 weeks after

injury. During the first week of recovery, BBB scores progressed fiom 0-7 which

corresponded to improvement firom flaccid paralysis to extensive HL joint movement.

BBB Scores for week two ranged firom 4-11, which represented a progression firom

movement at all HL joints to plantar placement of the hindpaws, weight support and early stepping behaviors. From 3 to 10 weeks after SCI, locomotor scores plateaued between 10-11 reflecting occasional to consistent weight-supported stepping but no

11 21

(I) 18 + c 15 CO o> E 12 lU 9 oon o (O 6 CD CD 3 CD CTL SCI 0 1 3 5 7 14 21 28 35 42 49 56 63 70 DAYS POST OP

Figure 1.1: Mean BBB locomotor scores for SCI and CTL groups over time. The SCI group had significantly lower scores than CTL throughout 70 days of recovery (*** p<.001). Note that recovery of weight bearing and stepping (BBB scores of 9 and above) occurred fiom 14 days post operative (Post Op).

1 2 coordination between the forelimbs (FLs) and HLs (Basso et ai., 1995). In addition,

deficits in trunk control, tail and paw position, as well as an inability to clear the toes

during swing persisted throughout the 10 week recovery period.

Spinal Cord Lesion Centers

Laminectomy surgeries did not produce any noticeable pathology in spinal

cord tissue as assessed by light microscopic examination of LFB stained sections

(data not shown). Since the total cross-sectional area (TCS A) of the spinal cord fiom

CTL rats increased slightly but significantly fiom 3 to 10 weeks after injury (Table

1.1), aged-matched CTLs were used for comparison to injured animals. In addition,

not all samples were included in the quantitative analyses due to damage fiom

histological processing (see Table 1.1 for the number of spinal cords analyzed).

Representative photographs of the lesion epicenters, fiom animals sacrificed at

1, 3 and 10 weeks post SCI, are shown in Figure 1.2. Contusion injury resulted in damage to cell bodies and axons located centrally in the cord, often sparing a peripheral rim of myelinated axons. The size of the lesion expanded significantly over time (p<.05) as evidenced by a striking decrease in the TCSA of the lesion epicenter fiom 1 to 10 weeks post injury (Table 1.1). By 10 weeks post SCI, the

TCSA of the cord at the lesion epicenter was 75% smaller than normal, age-matched

CTLs (1.0 vs 3.9 mm^, respectively; p< 05). Coincident with the decline in TCSA was a significant progressive decrease in spared white matter over time (Table 1.1, p<05). When tissue sparing was expressed as a percentage of the TCSA of the cord.

13 W h ite M a t t e r W hite M atter/ TCSA (mm^) S pa r in g (m m ^) TCSA(%)

SCICTL SCI CTLSCICTL 1 WEEK 3.2 (.2) 3.6 (.2) .46 (.1 )0 2.6 (.1) 1 5 (2 .3 )0 75 (.5) n=8 n=6 3 WEEK 1.8 ( . l ) t 3.5 (.1) .37 (.03)0 2.6 (.01) 2 0 (1 .1 )0 75 (.3) n=6 n=4 10 WEEK 1.0 ( . l ) t 3.9 (.1)* .14 (.0 2 ): 3.1 (.2) 1 4 (2 .4 )0 76 (.7) n=7 n=4

Table 1.1: Mean area measurements (± sem) of myelin sparing at the lesion epicenter 1, 3 and 10 wks after moderate SCI or laminectomy (CTL). Total cross sectional area (TCSA), total white matter sparing, and the percentage of white matter sparing were measured. The number of rats per group are listed under the TCSA values, t : p<.05 vs. 1 wk SCI and all CTL groups; O: p<.05 vs. all CTL groups; *: p<.05 vs. 1 and 3 wk CTL groups; p<.05 vs. 1 and 3 wk SCI and all CTL groups.

moderate SCI created approximately 15 ± 2 (sem)%, 20 ± 1% and 14 ± 2% sparing at

1, 3 and 10 weeks after injury, respectively (Table 1.1).

One week after a moderate SCI (1.1 mm displacement) there was no sparing

of gray matter apparent at the injury epicenter (Figure 1.2 A). The central lesion area

rarely demonstrated cystic cavities; rather it contained swollen, collapsed and

demyelinating axons, as well as cellular debris. In addition, a clear lattice type

structure with occasional light LFB staining was also noted throughout the lesion area

(Figure 1.2 A, small open arrows). This is in contrast to the darkly stained compacted

myelin noted in the peripherally spared tissue (Figure 1.2 A, arrowheads). Sparing was noted in all ftmicular regions with the least spared tissue observed in the dorsal funiculus. Small vacuoles were seen predominantly in the ventral and lateral

14 Figure 1.2. Photomicrographs of myelin stained spinal cord sections through the lesion epicenter at 1 (A), 3 (B) and 10 (C) weeks post SCI. Note the significant decrease in total cross sectional area of the spinal cord over time. Arrowheads (^) denote some areas of white matter sparing at the periphery of the cord. Lightly stained fibers occupy central regions of the lesion at 1 (A) and 3 (B) weeks (":>) but not 10 weeks (C) post injury. Cystic cavities develop at 3 weeks (B) and enlarge by 10 weeks (C) post SCI (<->). Scale bar: 300 pm.

15 % I

%

Figure 12 . 16 funicular LFB stained regions, as well as, occasionally in unstained central lesion areas. Reported analyses of 1 pm thick plastic sections through the lesion epicenter in rats 1 week post weight drop contusion injury (25 g-cm) showed a lesion area infiltrated with macrophages engulfing necrotic debris (Beattie et al., 1998).

Three weeks post SCI, the TCSA of the cord significantly decreased 45% fiom the 1 week value (1.8 vs 3.2 mm^; p<.05), while the area occupied by spared myelinated fibers decreased 20% (Table 1.1). There was a persistence of cellular debris noted centrally in the cord and moderate sized cavities had developed in the ventral and dorsolateral funiculi, as well as in central regions formerly occupied by white and gray matter (Figure 1.2 B, double ended arrows). Macrophages have previously been observed to infiltrate the central lesion at this time point although their appearance suggests less engulfinent of cellular debris (Beattie et al., 1998).

Scattered light LFB staining persisted throughout the lesion and an increased number of vacuoles was observed in both LFB stained and unstained regions. Spared myelin- stained tissue was seen in all funiculii throughout the peripheral rim (Figure 1.2 B, arrowheads), with the greatest loss of myelinated tissue noted in the dorsal funiculus regions.

At the 10 week time point, the TCSA of the cord was significantly reduced by 69% of the one week value (Table 1.1, 1.0 vs. 3.2 mm^; p<05) and by —75% of the age- matched CTLs (1.0 vs 3.9 mm^), producing an hourglass appearance upon gross inspection of the cord. Large cystic cavities were noted throughout central regions as

17 well as in ventral, lateral and occasionally dorsal funiculii (Figure 1.2 C, double ended arrows). Published electronmicrographs taken from tissue 8 weeks following a moderate weight drop contusion injury reveals the presence of “hypertrophic astrocytic processes” (Wrathall et al., 1998), which might be expected to occupy our lesion site at 10 weeks post injury. In addition, area measurements of dark LFB stained myelin in the periphery of the cord (Figure 1.2 C, arrowheads), most notable at this time point in the ventrolateral and dorsolateral regions, decreased by 70% from the 1 week values (Table 1.1). Occasionally, light LFB stained cellular debris was noted centrally.

Bodv Weight and Muscle Wet Weight

Injured animals presented with a transient decrease in mean body weight, 93% compared to CTLs animals, at 1 week, 96% at 3 weeks and 101% at 10 weeks post injury (Table 1.2). Although these values were not statistically different from CTLs at any time point, they do reflect a more rapid recovery of body weight than muscle wet weight over time following SCI (see below). All the HL muscles tested in this study (except for the EDL) showed an early, significant decrease in wet weight (Table

1.2). This resolved in three of the five muscles at intermediate (SOL, TA) or late time points (PL) (Table 1.2), but persisted for MG and LG muscles even at 10 weeks after injury. Since there is not a linear relationship between recovery of body weight and muscle wet weight after spinal cord contusion injury, normalizing muscle wet weight to body weight masks resolution of muscle atrophy. Therefore, absolute wet weight values are represented in Table 1.2.

18 MUSCLE WET WEIGHT MEASURES BW EDL PLSOL MGLGTA 1 WEEK CTL (n=6) 263 113 334 149 647 856 553 (1) (6) (12) (6) (43) (45) (17)

SCI (n=8) 244 106 260* 116* 482* 647* 445* (8) (5) (8) (7) (19) (39) (15) 3 WEEKS CTL (n=6) 269 125 352 143 725 951 619 (I) (4) (12) (4) (30) (23) (17)

SCI (n=8) 258 116 296* 139 610* 754* 560 (1) (4) (13) (10) (19) (23) (17) 10 WEEKS CTL (n=6) 288 130 383 142 770 1041 680 (1) (7) (5) (2) (12) (43) (13)

SCI (n=8) 289 129 336 133 651* 865* 609 (1) (4) (10) (6) (13) (31) (14)

Table 1.2: Mean absolute muscle wet weight (mg) and body weight (BW, in grams) at 1,3 and 10 wks post SCI compared to CTL. Note the significant decrease in muscle weight in most muscles at 1 wk post SCI which resolved by 10 wks for all but the MG/LG muscles. *: significantly different firom CTL at each time point, p<.05. Standard estimate of the mean shown in parentheses (sem). Number of rats per group listed (n= ). EDL: extensor digitorum longus; PL: plantaris; SOL: soleus; MG/LG: medial and lateral gastrocnemius; TA: tibialis anterior.

19 One week after SCI, the absolute wet weight o f all muscles, except the EDL, was significantly decreased (20-25%) compared to %e-matched CTLs (p<.05; Table

1.2). At the 3 week time point, loss of muscle wet weight was partially attenuated but remained significantly less than CTLs for the PL, MG and LG muscles (16-21%). In contrast, SOL mean wet weight returned to CTL levels at 3 weeks post op (p >.05), despite significant deficits in force production that occurred at this time point (see below). Ten weeks after injury, wet weights for PL, SOL, EDL and TA fiom injured animals were not different from CTL values, however, the MG and LG muscle continued to show significant decreases (15 % each).

Muscle Phvsiologv

At 1 week post injury, in vitro Pt and Po measures of the SOL showed decreases in absolute values for injured animals compared to CTLs which mirrored muscle wet weight loss (20%) (Table 1.3). However, neither absolute measures (g) nor values normalized to muscle wet weight (g/g) were different from CTLs 1 week after injury. At 3 weeks post SCI, absolute and normalized values for Pt and Po were decreased significantly approximately 51% and 46%, respectively vs. CTL (p<.05;

Table 1.3). This was surprising given that SOL wet weight was no longer different from CTLs at 3 weeks post injury. Interestingly, differences between groups in Pt and Po values (both absolute and normalized) were resolved by the 10 week time p o in t

20 There was no difference between SCI and CTL rats in their tPt responses at

any of the time points tested in this study. In addition, although overall RT values

were slower than those previously published (250 ms vs. 110 msXFitts, HoUoszy,

1977), the mean values were significantly faster (28%; p<.05) for injured animals 3

weeks after injury (Table 1.3).

IW eek 3 Week 10 W eek SCI CTL SCI CTL SCI CTL P t(g ) 32(2) 39(3) 23 (1)* 45(7) 43 (8) 48 (5)

P t(g /g ) 280 (39) 266 (17) 150(16)* 317(42) 330 (65) 340 (40)

P o (g ) 63 (13) 82(9) 36(5)* 78(14) 80(17) 90 (12)

Po (g/g) 536(101) 546 (49) 223 (10) * 550 (86) 603 (124) 709 (90)

tP t(m s) 107(3) 121 (5) 99 (13) 106 (4) 115(6) 116(9)

14 RT (ms) 222 (13) 326 (36) 209 (26) * 289 (9) 321 (17) 264(15)

Pt/Po .56 (.07) .50 (07) .68 (08) .59 (.05) .55 (.03) .49 (.01)

Table 1.3: Contractile properties of the SOL muscle fiom SCI and CTL animals at I, 3 and 10 weeks post op. Absolute (g) and normalized (g/g) values (mean ± sem) are reported for peak twitch (Pt) and peak tetanic tension OPo). Time to peak twitch (tPt), 14 relaxation tim e { V i RT) and specific tension (Pt/Po) are shown. (n=4 rats per group except 3 week SCI and 10 week CTL with n=3). * indicates significant differences than CTL p<.05.

21 Mvosin Heavy Chain Expression

In the SOL, there was no change noted in relative percent of MHC expression at I week after spinal cord contusion injury (Table 1.4). Three weeks following SCI, there was expression of the transitional llx heavy chain (5%, never observed in CTL animals) though this level of expression was not statistically significant. Ten weeks after SCI, no differences in MHC expression were observed between injured and

CTLs (Figure 1.3, Table 1.4).

In the EDL, MHC profiles showed a significant downregulation of lib (15%) with a corresponding increase in the llx isoform at I week post op. At 3 weeks after injury, the upregulation of nX was significantly different fiom controls with a corresponding (non-significant) mean decrease in Ilb MHC observed. No difference fiom CTL values was noted at the 10 week time point.

22 IW eek IW eek 3 W eek 3 W eek 10 W e e k 10 W eek SCI CTL SCI CTL SCI CTL SOL I 97.2 (5.7) 92.8 (10.2) 92.9 (8.4) 91.8(3.9) 96.1 (4.6) 96.2 (5.4) na 2.9 (5.7) 7.1 (10.0) 2.5 (3.0) 8J (3.9) 3.0 (4.0) 3.8 (5.4) fix 0 0 4.7 (5.5) 0 1.0 (2.0) 0 nb 0 0 0 0 0 0 EDL I 3.7 (6.5) 0 4.5 (3.1) 3.9 (5.5) 0 0 na 9.8 (4J) 7.3 (1.7) 7.2 (5.0) 9.4 (IJ) 3.0 (3.6) 6.4 (9.0) fix 40.4 (5.6) 31.1 (2.3) 54.6* (4.4) 40.9 (7.6) 47.7(12.2) 42.8 (5 2 ) nb 46.1* (8.0) 61.6 (4.0) 33.8 (5.3) 45.8(12.0) 49.3 (11.5) 50.9 (3.7)

Table 1.4: Myosin heavy chain (MHC) expression for the soleus (SOL) and extensor digitorum longus (EDL) muscles from SCI and CTL groups at 1, 3 and 10 weeks. The relative percentage (mean ± sem) of slow (I) and fast (Ua, Ux, lib) isoforms were determined with SDS-PAGE. (Columns do not add to 100% due to rounding). *: significantly different than CTL p<.05.

23 MHC SHS-PAGE

FIG U R E 1 3 Myosin heavy chain (MHC) protein assays for the soleus (SOL) (lanes 1-9) and extensor digitorum longus (EDL) (lanes 10-13) muscles of injured (I) and control (C) animals at 1, 3 and 10 wks. Lanes 4 and 5 demonstrate types Ila, IIx, and I MHC isoforms in the SOL 3 wks post injury while only types Ua and I isoforms are found in control animals (lane 6). After SCI, the EDL demonstrates a decrease in lib (lane 12) and an increase in IIx expression (lanes 10 and 12) relative to CTL at 1 wk (lane 11) and 3 wks (lane 13).

24 D iscussion

The major finding of this study was that a moderate spinal cord contusion injury produced transient changes in muscle phenotype and contractile properties which reflected the locomotor abilities of the animal. During the early period of paralysis and paresis pronounced muscle atrophy was observed in all but one of the muscles studied in this experiment This phase of non-weight bearing with the HLs appears to have induced in the SOL significantly decreased contractile force production and modest changes in phenotype expression towards fast muscle as detected at 3 weeks post SCI. In addition, decreased lib and increased IIx MHC expression in the EDL was identified early post injury time points only (1 and 3 weeks, respectively). The recovery of weight supported HL stepping was associated with a return to baseline levels for contractile properties, fiber phenotype and gross muscle atrophy by 10 weeks post injury. Thus, muscle plasticity appears to be extremely sensitive to locomotor deficits and their resolution induced by spinal cord contusion.

Comparison to Other Models

Altered neural activation patterns and changes in limb loading contribute to adaptations in skeletal muscle properties which influence motor performance under normal conditions. Therefore, it is important to determine the differences and similarities across the common models used to study muscle plasticity like spinal cord

TX, HLS and peripheral denervation as compared to the spinal cord contusion model.

An examination of locomotor behavior across different injury models / manipulations

25 clearly identifies differences in limb loading. Following spinal cord TX, the HLs are dependent and generally maintained in an extended position dragged behind the body during forward locomotion with the FLs (Basso et al., 1996; Berhmann et al., 1992;

Roy et al., 1991). Thus, animals with spinal cord transection made in adulthood rarely or never place weight across the HLs. After contusion injury, however, early

HL paralysis gives way to HL movements, plantar placing and ultimately stepping behaviors within the first 2-3 weeks post injury (Basso et al., 1995). Thus weight support by the affected HLs typically occurs even after moderate spinal cord contusion. Likewise, peripheral denervation injury (i.e. complete transection of the sciatic nerve) does not prevent an animal firom weightbearing on the affected HL, albeit clumsily so that the hindpaw of the flaccid limb flips between inversion, eversion and (subtalar) neutral with each step (Bennett, Xie, 1988). Lastly, HLS by definition does not allow weight bearing activity through the HLs but does not prevent full joint movement while the HLs are suspended in midair.

Lesion Development

The consequence to neural structures is also varied between models. As stated previously, spinal cord TX results in complete severance of ascending and descending axons across the lesion site, whereas contusion injiuy results in some sparing of fibers across the epicenter. The current data, using the contusion model, showed a progressive loss of myelinated axons at the lesion epicenter (Figures 1.2, Table 1.1) which is consistent with our previous reports (Basso et al., 1996), as well as several papers describing the evolving nature of spinal cord contusion injury (reviewed in

26 Schwab, Bartholdi, 1996; Balentine, Paris, 1978a; Balentine, Paris, 1978b; Basse et

al., 1995; Bresnahan, 1978; Bresnahan et al., 1976; Noble, Wrathall, 1989; Beattie et

al., 1997). Limited human data also suggest that SCI yields an expanding injury with

a progressive state of Wallerian Degeneration (of both ascending and descending

tracts) observed with increased time post injury (Becerra et al., 1995). The

progression of the lesion size is likely to result in altered descending neural “drive” to

motor neurons and intemeurons subserving the HLs leading to a decrease in

activation as witnessed by HL paresis. Peripheral denervation (i.e. axotomy of sciatic

nerve) on the other hand generally does not lead to sensory or motor neuron cell loss

(reviewed in Aldskogius, Kozlova, 1998) but includes structural adaptations between

neural-glial and glial-glial interactions. For example, astrocytic hypertrophy and

microglial proliferation have been described and are confined to the motor neuron and

central branches of sensory axons of the injured segment (Aldskogius, Kozlova,

1998). The influence of these changes caudal to the injury segment is unclear;

however, little or no effect on muscle activation at the injured segment likely results

since the axon fiom the motor neurons has been completely severed and action potentials fail to reach the corresponding muscle.

Lastly, HLS does not directly invoke damage to CNS structures, but few studies have assessed changes in CNS function following unweighting (Ishihara et al.,

1997). After 2 weeks of HLS, no change was noted in SOL motor neuron size distribution, mean soma cross-sectional area or mean succinate dehydrogenase activity (Ishihara et al., 1997), despite significant adaptations occurring in the SOL

27 muscle. In contrast, HLS is the animal model for spaceflight, and studies firom

animals following two to three weeks in space revealed a decrease in cytochrome c

expression in ventral motor neurons (Poliakov et al., 1995), and a decrease in protein

and mRNA content in the cytoplasm of large ventral motor neurons in the lumbar

cord (Gorbunova, Portugalov, 1976). These findings suggest a decrease in the

metabolic activity in these motor neurons due to lack of normal proprioceptive

afiTerent input (Gorbunova, Portugalov, 1976). Finally, upon returning to IG the MG

muscle was preferentially activated over the SOL during locomotor tasks, suggesting

altered central mechanisms vdiich favor recruitment of fast twitch motor units over

slow units (Hodgson et al., 1991; Recktenwald et al., 1999).

It is clear that each model, TX, contusion, HLS and peripheral denervation,

produce differential limb loading and motor neuron activation. Therefore, skeletal

muscle adaptations related to functional outcome after a contusion injury appear to be

distinct firom other models of muscle plasticity. From a neurorehabilitation perspective, it appears that the common models for muscle plasticity do not adequately represent the conditions inherent to human SCI. Hence it seems warranted and necessary to examine the extent and effectiveness of muscle plasticity in the contusion model which directly replicates human SCI.

Gross Muscle Atronhv

Data presented here are consistent with previous reports showing pronounced muscle atrophy following decreases in weight bearing status. We found that during periods of non-weight bearing (paralysis) gross muscle atrophy was pronounced.

28 However, once weight bearing had recovered the atrophy resolved for most muscles.

For example, a 20-25% loss of muscle mass occurred 1 week post SCI, a time when the HLs were non-weight bearing. These values for muscle atrophy are similar to those induced by HLS (30% Pierotti et al., 1990) but more extensive than after peripheral denervation (13%, Gundersen, 1985) at comparable time points.

Therefore a lack of a weight bearing, regardless of the lesion model, produces similar deficits in wet weight loss.

Three weeks following contusion injury animals had regained some locomotor function and were weight bearing through the HLs which was related to an attenuation of muscle wet weight loss in most muscle groups. This is in contrast to the decrease in wet weight values observed following two to four weeks of HLS (15-

37%, Brown, Hasser, 1995; 12-40%, Diffee et al., 1991), SC transection (30%,

Talmadge et al., 1995b), and peripheral denervation (64%, Bennett, Xie, 1988) in which muscle atrophy and non-weightbearing or altered-weightbearing conditions persist throughout the recording period.

Furthermore, muscle size may not only be determined by weight bearing but also by how the muscle is used following spinal cord contusion. We found atrophy in only the MG and LG muscles that persisted as long as 10 weeks post SCI. A differential atrophy response for the PL, SOL, EDL, TA versus the MG/LG during recovery suggests that either the animal cannot generate high forces ftom the MG/LG muscles or the activities in which the rat participates do not require sufficient MG/LG activity to overcome contusion-induced atrophy. Given that these muscles reach their

29 highest activi^ levels during movements requiring large force production like inclined running or swimming (de Leon et al., 1994; Gruner, Altman, 1980; Roy et al., 1985; Roy et al., 1991), two motor tasks in which none of our animals participated, it may not be surprising that atrophy persisted. Perhaps training in these tasks would alleviate the prolonged muscle atrophy observed for the MG/LG muscles

(see Chapter 3).

In addition, SOL mean wet weight values for injured animals were equivalent to CTLs three weeks after injury, despite the fact that force production capabilities

(Ft, Po) were significantly reduced at the same time point. This difference might suggest that deficits in excitation-contraction coupling mechanisms were responsible for the decrease force output instead of reduced fiber area. However, further evaluation o f fiber CSA is necessary to determine if attenuation of muscle wet weight loss following spinal cord contusion injury may be attributed to increases in non- contractile tissue (fat, connective tissue and/or interstitial fluid) or truly to changes in fiber size (Kandarian et al., 1991).

Phvsioloev

Changes in whole muscle contractile properties reveal significant decreases in

Pt and Po in most “unweighting” models. For example, one week after HLS

0-50% decreases in Pt and 30-50% decreases in Po have been described for the slow twitch SOL (Fitts et al., 1986; Pierotti et al., 1990), which is consistent with the 20% decrease in absolute force measurements observed following moderate contusion injury. Further loss of force output following two weeks of HLS (50-60%, Brown,

30 Hasser, 1995) or 4 weeks of denervation (60-80% decreases, Pt and Po respectively,

Spector, 1985) are also comparable to the 50% drop in force output noted at 3 weeks

post contusion injury. Unlike other models, however, SOL force measurements

returned to control values 10 weeks after injtuy.

Also, with the exception of the denervation model, contractile speed for the

SOL in most studies gets faster as the time spent “unweighted” increases. Decreases

in tPt occur following 1 week (22-25%, Fitts et al., 1986; Pierotti et al., 1990) or 2 weeks (40%, Fitts, Holloszy, 1977) of HL suspension, while 3 weeks after spinal cord TX a 23% decrease for tPt has been reported (Midrio et al., 1988). Consistent with the findings of Brown and Hasser (1995) which did not show significant changes in SOL tPt after 2 weeks of imweighting (HLS), the tPt values of the current study were not different from CTLs at any of the time points evaluated. Taken together, these data contrast with the peripheral denervation model which shows a 25-40% slower contraction time from 5 to 28 days after injury (Gimdersen, 1985; BuffeUi et al., 1997). The latter is thought to be due to impaired calcium (C a ^ resequestration mechanisms mediated by disconnection of the motor nerve from the muscle after denervation.

One-half RT measurements also did not change following 2 weeks of unweighting (Fitts et al., 1986; Pierotti et al., 1990; Brown, Hasser, 1995); however, 4 weeks of HLS produced a 26% faster time course for the SOL (Diffee et al., 1991).

Similarly, in the present study no differences were noted in relaxation times between injured and CTL animals 1 week after injury but by three weeks after SCI modest and

31 significantly faster Vi RT times (18%) were observed. At the 10 week time point,

injured animals were no different from controls. In addition, in the current study

absolute Vi RT measurements of Soleii muscle (for SCI and CTL rats) were

significantly slower than previously reported values for both in vivo and in vitro

studies (Fitts et al., 1986; Pierotti et al., 1990; Brown, Hasser, 1995). This may have

been due to the caffeine added to the incubation medium (Blough, 1997) intended to

facilitate maximum evoked twitch and tetanus responses fi;om an in vitro set up.

Caffeine causes the sarcoplasmic reticulum to release C a^ (Salviati et al., 1989;

Tamopolsky, 1994) and prevents C a^ re-uptake which may prolong the

resequestration of C a^ into the sarcoplasmic reticulum, an action necessary for

relaxation to occur. Both CTL and SCI muscle was subjected to this protocol m aking comparisons within our study possible, however, comparisons to studies using a non- caffeinated medium should be interpreted with caution. Further study is needed to contrast 14 RT values after moderate SCI with other models that manipulate neural input or HL loading.

Changes toward faster contractile times (14 RT) are often associated with a conversion of muscle fiber toward faster phenotype of myosin heavy chain. The modest increases (5%) in SOL IIx MHC expression only at the 3 week time point is in keeping with the fact that slight speeding up occurred for 14 RT values at this time point only.

32 Muscle Fiber Type

Histochemical and biochemical analyses of muscle fiber type following spinal

cord TX and HLS show corresponding changes toward faster profiles, with an

increase in darkly stained fibers using mATPase reactions (alkaline preincubation)

and a greater expression of fast MHC isoforms (Talmadge et al., 1995a; Jiang et al.,

1990; Lieber et al., 1986). Although most papers focus on chronic timepoints (4-12

months post injury), modest fiber type changes (3-16% increases in fast fibers) have

been described as early as 1 to 2 weeks following TX (Dupont-Versteegden et al.,

1998; Talmadge et al., 1995b) and HLS (Brown, Hasser, 1995). A 20% increase in

fast fibers was found 4 weeks after denervation (Spector, 1985). This early change in

fiber type is in contrast to the 90% conversion of slow to fast fiber noted in the SOL

one fill! year after spinal cord TX (Lieber et al., 1986). Thus, the longer that the limb

loading and/or neural activation is prevented the greater the magnitude of change in

muscle phenotype.

In the current study, changes toward faster MHC expression 3 weeks after

injury were statistically different in the EDL (9% increase in Ux with a corresponding

decrease in Ilb) and a slight though not significant upregulation (5% of Ux) occurred

in the SOL. These were only minor changes compared to the TX model, however, and most likely reflect the difference in activation patterns of the HL muscles following injury.

33 Model Relevance to Human SCT

It appears that spinal cord contusion injury represents a model of partial denervation of motomeuron and intemeuron pools below the level of injury due to loss of supraspinal and propriospinal axons (Basso et al., 2000). Skeletal muscle denervation likely occurs at segmental levels where the lesion has destroyed the gray matter of the cord which, in our model, is limited to the thoracic spinal cord. There appears to be little or no denervation of muscles innervated by motor neurons in the lumbar cord after midthoracic contusion. While both the contusion model and the peripheral denervation model show muscle fiber atrophy, peripheral nerve lesions result in immediate and significant slowing of contractile parameters and faster maximum shortening velocity (Spector, 1985; Kean et al., 1974). It is speculated that the slowed contractile properties are the result of impared excitation-contraction coupling mechanisms while altered myofibrillar proteins, changed toward faster isoforms, are responsible for the faster shortening velocities (reviewed in Talmadge et al., 1995a). In addition, denervated muscle fiber shows evidence of fiber type grouping, extra-junctional Acetyl-choline receptors, and increased innervation ratios

(McComas, 1996; Thesleff, 1974), none of which are commonly observed following experimental SCI (Cope et al., 1986; Talmadge et al., 1995a). It is important to note that the segmental connections of sensory and motor axons in the lumbar cord below the T8 lesion remain intact thereby preserving normal neurotrophic interactions between the muscle and motor neuron pools. The integrity of anatomical pathways from the spinal cord to the skeletal muscle distinguishes the contusion model from the

34 peripheral denervation model, A\iiich isolates the muscle from the motor neuron.

Thus, the response of the muscle to neural lesion appears to be influenced by both activation and neurotrophic support from the motor neurons.

The moderate contusion injury model generally produces changes in skeletal muscle similar to early changes seen following both spinal cord TX and HLS.

Weight bearing and HL movements / muscle contractions may account for the differences across experimental models observed at later time points. Spinal cord TX renders the HLs nearly motionless due to the loss of supraspinal input and volitional control. In contrast, HL movement continues during HLS and returns within the first week after moderate contusion injury with the primary difference between these models being a progression to weight bearing status after SCI but not after HLS. It is important to note that HL movement and weight bearing are likely to facilitate protein synthesis in the muscle thereby attenuating fiber atrophy and fiber type conversion

(Thomason et al., 1987b). The degree of preservation and/or recovery of normal skeletal muscle properties may play an instrumental role in the extent of motor recovery observed.

How well does experimental spinal cord contusion injury reproduce changes in rat HL skeletal muscle relative to their human counterparts? Due to the invasive nature of these analyses, few studies have addressed changes in skeletal muscle in humans following spinal cord contusion injury (Lotta et al., 1991; Grimby et al.,

1976; Stilwill, Sahgal, 1977; Castro et al., 1999). Analyses of muscle at chronic time

35 points (1-2 years post injury) found fiber conversion toward a faster phenotype, a pattern similar to that shown in the rat after spinal cord contusion in the present study

(Grimby et ai,, 1976; Shields, 1995; Round et ai., 1993). Following clinically complete SCI in humans, significant muscle atrophy is revealed (40% decrease in

CSA) as early as 6 weeks post injury, while only moderate changes in MHC expression occurred even as late as 24 weeks post injury (Castro et al., 1999). In the rat contusion model, we saw significant changes in gross muscle atrophy as early as 1 week post SCI and significantly greater expression of fast MHC expression at 1 and 3 weeks po. Therefore, there appears to be a high degree of conformity in muscle adaptations between human SCI and moderate contusion in the rat. Interestingly, it appears that the time course of these adaptations is faster in small mammals after SCI

(Talmadge et al., 1995a).

Caveats

Limited statistical significance between groups with respect to MHC expression and physiological ftmction following injury may be secondary to low sample size. A low sample size was used in this study in order to gather all the behavioral, neural and skeletal muscle data within the same animal.

Also, gender selection may show differences in the severity of change experienced following this manipulation. Male rats may have larger muscle wet weight values, higher control/baseline force measurements and therefore, theoretically could show greater decreases in wet weight and force production following injury than female rats. Female rats, however, were used in this study for two reasons.

36 Previous woric confirmed that spinal cord contusion lesions are not different with respect to lesion severity between weight-matched male and female rats (unpublished observations) and female rats are significantly easier to care for postoperatively.

Conclusions

The experimental spinal cord contusion model accurately reproduces both the

CNS histopathological scenario, as well as, the resultant changes in skeletal muscle properties that are seen following human spinal cord contusion injury. The major differences between the experimental model and SCI in humans are that most moderately lesioned animals recover the ability to locomote independent of treatment interventions whereas walking rarely occurs if no manipulations are used in clinical populations (Wickelgren, 1998). Therefore, this experimental model represents incomplete SCI in which voluntary activation of the HLs occurs with varying contributions to functional skills. In patients with complete lower limb paralysis chronic muscular changes will likely be more severe than noted here.

Further studies evaluating larger numbers of anim al s and including E M C analysis for changes in muscle activation / recruitment that occurs during locomotion after contusion injury, are clearly warranted. In addition, determination of fiber CSA and its relationship to muscle wet weight values is needed. An analysis of the factors which lead to locomotor recovery despite severe damage to the spinal cords of these animals, could bring insight into treatment interventions which may translate to the clinical population. These factors likely include: (a) requirements for trunk control in

37 quadrupedal versus bipedal gait, (b) fear of failing and its relationship to attempted locomotion, (c) early forced-use activiQr (water and food located at opposite ends of the cage), and finally (d) slumbering in bunches with roommate grooming activities, presumably resulting in changes in sensory feedback that may facilitate motor output.

38 LIST OF REFERENCES

1. Aldskogius H, Kozlova EN (1998) Central neuron-glial and glial-glial interactions following axon injury. Prog.Neurobiol. 55:1-26.

2. Balentine JD, Paris DU (1978a) Pathology of experimental spinal cord trauma I. The necrotic lesion as a function of vascular injury. Lab invest 39:236-253.

3. Balentine JD, Paris DU (1978b) Pathology of experimental spinal cord trauma U. Ultrastructure of axons and myelin. Lab lnvest 39:254-265.

4. Barbeau H, Rossignol S (1987) Recovery of locomotion after chronic spinalization in the adult cat. Brain Res. 412:84-95.

5. Basso DM, Beattie MS, Bresnahan JC (1995) A sensitive and reliable locomotor rating scale for open field testing in rats. J.Neurotrauma 12: 1- 21.

6. Basso DM, Beattie MS, Bresnahan JC (1996) Graded histological and locomotor outcomes after spinal cord contusion using the NYU weight drop device versus transection. Exp Neurol. 139:244-256.

7. Basso DM, Beattie MS, Bresnahan JC (2000) Descending systems contributing to locomotor recovery after mild or moderate spinal cord injury in rats. Manuscript in preparation.

8. Beattie MS, Bresnahan JC, Komon J, Tovar CA, Van Meter M, Anderson DK, Faden AI, Hsu CY, Noble LJ, Salzman S, Young W (1997) Endogenous repair after spinal cord contusion injuries in the rat. ExpNeurol. 148:453-463.

9. Beattie MS, Shuman SL, Bresnahan JC (1998) Apoptosis and spinal cord injury. Neuroscientist 4:163-171. 39 10. Becerra JL, Puckett WR, Heister ED, Quencer RM, Marcillo AH, Post MJ, Bunge RP (1995) MR-pathologic comparisons of wallerian degeneration. AJNR 16:125-133.

11. Bennett GJ, Xie YK (1988) A peripheral mononeuropathy in rat that produces disorders of pain sensation like those seen in man. Pain 33:87-107.

12. Berhmann DL, Bresnahan JC, Beattie MS, Shah BR (1992) Spinal cord injury produced by consistent mechanical displacement o f the cord in rats: Behavioral and histologic analysis. J.Neurotrauma 9:197-216.

13. Blough ER (1997) The effects of aging and overload on plantaris and extensor digitorum longus morphology and physiology in adult aged Fisher 344 and FI hybrid rats. Doctoral dissertation. The Ohio State University.

14. Bresnahan JC (1978) An electron-microscopic analysis of axonal alterations following blunt contusion of the spinal cord of the rhesus monkey (Macaca mulatta). J.Neurol.Sci. 37:59-82.

15. Bresnahan JC, Beattie MS, Todd FD, Noyes DH (1987) A behavioral and anatomical analysis of spinal cord injury produced by a feedback- controlled impaction device. Exp Neurol. 95:548-570.

16. Bresnahan JC, King JS, Martin OF, Yashon D (1976) A neuroanatomical analysis of spinal cord injury in the rhesus monkey (Macaca mulatta). J.NeuroI.Sci. 28:521-542.

17. Brown M, Hasser EM (1995) Weight-bearing effects on skeletal muscle during and after simulated bed rest. Arch.Phys.Med.Rehabil. 76:541- 546.

18. BuffeUi M, Pasino E, Cangiano A (1997) Paralysis of rat skeletal muscle equaUy affects contractile properties as does permanent denervation. JM uscle ResCeU MotU. 18:683-695.

40 19. Bunge RP, Puckett WR, Becerra JL, Marcillo A, Quencer RM (1993) Observations on the pathology of human spinal cord injury. A review and classification for 22 new cases with details from a case of chronic cord compression with extensive focal demyelination. Adv.Neurol. 59:75-89.

20. Castro MJ, Apple DF, Jr., Staron RS, Campos GER, Dudley GA (1999) Influence of complete spinal cord injury on skeletal muscle within 6 months of injury. J Appl Physiol. 86:350-358.

21- Cope TC, Bodine SC, Fournier M, Edgerton VR (1986) Soleus motor units in chronic spinal transected cats: Physiological and morphological alterations. J.Neurophysiol. 55:1202-1220.

22. Crowe MJ, Bresnahan JC, Shuman SL, Masters JN, Beattie MS (1997) Apoptosis and delayed regeneration after spinal cord injury in rats and monkeys. NatMed. 3:73-76.

23. Davey DF, Dunlop C, Hoh JFY, Wong SYP (1981) Contractile properties and ultrastructure of extensor digitorum longus and soleus muscles in spinal cord transected rats. Aust.J.Exp.Biol.Med.Sci. 59:393-404.

24. de Leon R, Hodgson JA, Roy RR, Edgerton VR (1994) Extensor- and flexor­ like modulation within motor pools of the rat hindlimb during treadmill locomotion and swimming. Brain Res. 654:241-250.

25. Diffee GM, Caiozzo VJ, Herrick RE, Baldwin KM (1991) Contractile and biochemical properties of rat soleus and plantaris after hindlimb suspension. Am J Physiol. 260:C528-C534

26. Dupont-Versteegden EE, Houle JD, Gurley CM, Peterson CA (1998) Early changes in muscle fiber size and gene expression in response to spinal cord transection and exercise. Am J Physiol. 275:C1124-Cl 133

27. Fauteck SP, Kandarian SC (1995) Sensitive detection of myosin heavy chain composition in skeletal muscle under different loading conditions. Am.J.Physiol. 268:C419-C424

41 28. Finol HJ, Lewis DM, Owens R (1981) The effects of denervation on contractile properties of rat skeletal muscle. Jfhysiol. 319:81-92.

29. Fitts RH, Holloszy JO (1977) Contractile properties of rat soleus muscle: Effects of training and fatigue. A m Jfhysiol. 233:C86-C91

30. Fitts RH, Metzger JM, Riley DA, Unsworth BR (1986) Models of disuse: A comparison of hindlimb suspension and immobilization. JAppLPhysiol. 60:1946-1953.

31. Gorbunova AV, Portugalov W (1976) Cytochemical investigations of proteins and RNA in spinal motoneurons and neurons of spinal ganglia of the rat after space flight. Aviat.Space Environ Med. 47:708-710.

32. Grimby G, Broberg C, Krotkeiwska I, Krotkiewski M (1976) Muscle fiber composition in patients with traumatic cord lesion. Scand.J Rehabil.Med. 8:37-42.

33. Gruner JA, Altman J (1980) Swimming in the rat: Analysis of locomotor performance in comparison to stepping. Exp.Brain Res. 40:374-382.

34. Gundersen K (1985) Early effects of denervation on isometric and isotonic contractile properties of rat skeletal muscle. Acta Physiol.Scand. 124:549-555.

35. Hennig R, Lomo T (1987) Effects of chronic stimulation on the size and speed of long-term denervated and innervated rat fast and slow skeletal muscles. Acta PhysioLScand. 130:115-131.

36. Herbert ME, Roy RR, Edgerton VR (1988) Influence of one-week hindlim b suspension and intermittent high load exercise on rat muscles. Exp.Neurol. 102:190-198.

37. Hodgson JA, Bodine-Fowler S, Roy RR, de Leon R, de Guzman CP, Koslovskaya I, Sirota M, Edgerton VR (1991) Changes in recruitment of rhesus soleus and gastrocnemius muscles following a 14 day spaceflight. Physiologist 34:8102-8103

42 38. Ishihara A, Oishi Y, Roy RR, Edgerton VR (1997) Influence of two weeks of non-weight bearing on rat soleus motoneurons and muscle fibers. Aviat.Space Environ.Med. 68:421-425.

39. Jânkâlâ H, Haqola VP, Petersen NE, Hârkônen M (1997) Myosin heavy chain mRNA transform to faster isoforms in immobili^sd skeletal muscle: A quantitative PCR study. J Appl Physiol. 82:977-982.

40. Jiang B, Roy RR, Edgerton VR (1990) Expression of a fast fiber enzyme profile in the cat soleus after spinalization. Muscle Nerve 13:1037- 1049.

41. Kandarian S, O'Brien S, Thomas K, Schulte L, Navarro J (1992) Regulation of skeletal muscle dihydropyridine receptor gene expression by biomechanical unloading. J.Appl.Physiol. 72:2510-2514.

42. Kandarian SC, Boushel RC, Schulte LM (1991) Elevated interstitial fluid volume in rat soleus muscles by hindlimb unweighting. J Appl Physiol. 71:910-914.

43. Kean CJ, Lewis DM, McGarrick JD (1974) Dynamic properties of denervated fast and slow twitch muscle of the cat. JPhysiol. 237:101-113.

44. Laemmli UK (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680-685.

45. Lieber RL, Frieden JO, Hargens AR, Ferringa ER (1986) Long-term effects of spinal cord transection on fast and slow rat skeletal muscle. II. Morphometric properties. ExpNeurol. 91:435-448.

46. Lotta S, Scelsi R, Alfonsi E, Saitta A, Nicolotti D, Epifani P, Carraro U (1991) Morphometric and neurophysiological analysis of skeletal muscle in paraplegic patients with traumatic cord lesion. Paraplegia 29:247-252.

47. Lovely RG, Gregor RJ, Roy RR, Edgerton VR (1986) Effects of training on the recovery of full-weight bearing stepping in the adult spinal cat. Exp Neurol. 92:421-435.

43 48. McComas AJ (1996) Skeletal Muscle Form and Function. ChampaignJL: Human Kinetics.

49. Midrio M, Betto DD, Betto R, Noventa D, Antico F (1988) Cordotomy- denervation interaction on contractile and myofibrillar properties of fast and slow muscles in the rat Exp.Neurol. 100:216-236.

50. Noble LJ, Wrathall JR (1989) Correlative analyses of lesion development and fimctional status after graded spinal cord contusion injuries in the rat Exp.Neurol. 103:34-40.

51. Pierotti DJ, Roy RR, Flores V, Edgerton VR (1990) Influence of 7 days of hindlimb suspension and intermittent weight support on rat muscle mechanical properties. AviatSpace Environ Med. 61:205-210.

52. Poliakov IV, Louri O, Edzherton VR, Krasnov IB (1995) Histochemistry and morphology of the anterior homs of spinal cords in rats after 9 day space flight. Aviakosm.Ekolog.Med. 29:30-32.

53. Recktenwald MR, Hodgson JA, Roy RR, Riazanski S, McCall GE, Kozlovskaya I, Washburn DA, Fanton JW, Edgerton VR (1999) Effects of spaceflight on rhesus quadrupedal locomotion after return to IG. J.Neurophysiol. 81:2451-2463.

54. Round JM, Barr FMD, Moffat B, Jones DA (1993) Fibre areas and histochemical fibre types in the quadriceps muscle of paraplegic subjects. J Neurol.Sci. 116:207-211.

55. Roy RR, Baldwin KM, Edgerton VR (1991) Plasticity of skeletal muscle: Effects of neuromuscular activity. Exerc.Sport Sci Rev. 19:269-312.

56. Roy RR, Hirota WK, Kuehl M, Edgerton VR (1985) Recruitment patterns in the rat hindlimb muscle during swimming. Brain Res. 337:175-178.

57. Roy RR, Hutchinson DL, Pierotti DJ, Hodgson JA, Edgerton VR (1991) EMG patterns of rat ankle extensors and flexors during treadmill locomotion and swimming. J Appl Physiol. 70:2522-2529.

44 58. Salviati G, Betto R, Ceoido S, Tegazzin V, Puppa AD (1989) Caffeine sensitivity of sarcoplasmic reticulum of fast and slow fibers from normal and malignant hyperthermia human muscle. Muscle Nerve 12:365-370.

59. Schwab ME, Bartholdi D (1996) Degeneration and regeneration of axons in the leisoned spinal cord. Physiol.Rev. 76:319-370.

60. Segal SS, Faulkner JA (1985) Temperature-dependent physiological stability of rat skeletal muscle in vitro. AmJPhysiol. 248:C265-C270

61. Shields RK (1995) Fatigability, relaxation properties, and electromyographic responses of the human paralyzed soleus muscle. JNeurophysiol. 73:2195-2206.

62. Spector SA (1985) Trophic effects on the contractile and histochemical properties of rat soleus muscle. J.Neurosci. 5:2189-2196.

63. Stilwill EW, Sahgal V (1977) Histochemical and morphologic changes in skeletal muscle following cervical cord injury: A study of upper and lower motor neuron lesions. Arch Phys Med Rehabil. 58:201-206.

64. Stokes BT, Homer PJ, Akino M (1995) Spinal cord injury modeling and functional assessment. In: Central Nervous System Trauma: Research Techniques (Ohnishi ST, Ohnishi T eds), pp 281-295. New York: CRC Press.

65. Stokes BT, Noyes DH, Behrmaim DL (1992) An electromechanical spinal injury technique with dynamic sensitivity. J.Neurotrauma 9:187-195.

66. Talmadge RJ, Roy RR (1993) Electrophoretic separation of rat skeletal muscle myosin heavy chain isoforms. J Appl Physiol. 75:2337-2340.

67. Talmadge RJ, Roy RR, Bodine-Fowler SC, Pierotti DJ, Edgerton VR (1995a) Adaptations in myosin heavy chain profile in chronically unloaded muscles. BAM 5:117-137.

45 68. Talmadge RJ, Roy RR, Edgerton VR (1995b) Prominence of myosin heavy chain hybrid fibers in soleus muscle of spinal cord-transected rats. J JApplPhysiol. 78:1256-1265.

69. Tamopolsky MA (1994) Caffeine and endurance performance. Sports Med. 18:109-125.

70. Thesleff S (1974) Physiological effects of denervation of muscle. Ann.N.YJAcad.Sci. 228:89-104.

71. Thomason DB, Biggs RB, Booth FW (1989) Protein metabolism and P- myosin heavy-chain mRNA in unweighted soleus muscle. AmJPhysiol. 257:R30O-R305

72. Thomason DB, Booth FW (1990) Atrophy of the soleus muscle by hindlimb unweighting. J.Appl.Physiol. 68:1-12.

73. Thomason DB, Herrick RE, Baldwin KM (1987a) Activity influences on soleus muscle myosin during rodent hindlimb suspension. JApplPhysiol. 63:138-144.

74. Thomason DB, Herrick RE, Surdyka D, Baldwin KM (1987b) Time course of soleus muscle myosin expression during hindlimb suspension and recovery. J.Appl.Physiol. 63:130-137.

75. Wakatsuki T, Ohira Y, Nakamura K, Asakura T, Ohno H, Yamamoto M (1995) Changes of contractile properties of extensor digitorum longus in response to creatine-analogue administration and/or hindlim b suspension in rats. Jpn J Physiol. 45:979-989.

76. Wickelgren I (1998) Teaching the spinal cord to walk. Science 279:319-321.

77. Wrathall JR, Li W, Hudson LD (1998) Myelin gene expression after experimental contusive spinal cord injury. J Neurosci. 18:8780-8793.

46 CHAPTER 2

A COMPARISON OF EXERCISE INTERVENTIONS ON RECOVERY FROM SPINAL CORD CONTUSION INJURY IN RATS: I. A BEHAVIORAL PERSPECTIVE

Introduction

Thousands of Americans suffer spinal cord injuries (SCI) each year due to blunt contusion of the spinal cord (Bunge et al., 1993). Subjects often suffer from the severe social, emotional, economic as well as the physical costs associated with a paralytic injury. Traditional rehabilitation utilizes compensatory strategies to treat the lasting paralysis and paresis which can result from SCI, but do little to address facilitation of recovery of motor ftmction in the impaired limbs (Wickelgren, 1998).

Recent pharmocological therapies, especially the administration of immunosuppressive and anti-lipid peroxidation drugs early after injury (i.e methylprednisolone. Bracken et al., 1990), have improved the movement capabilities of people with SCI, thereby necessitating the development of effective motor control interventions which facilitate recovery of ftmction.

Exercise is known to be beneficial and can improve motor outcome from neurological impairment (Nudo et al., 1996; Hesse et al., 1995; Taub et al., 1993; Wolf

47 et al-, 1989). However, there are at least two unresolved issues regarding exercise

training after SCI: (a) identification of the components of exercise or sensory-motor

experiences that best facilitate recovery; and, (b) a determination of the substrates (i.e.

muscle, nerve, motor neuron, primary afferents) which adapt to support improved

functional outcome.

The success of treadmill training in completely transected animals (Lovely et

al., 1986; Lovely et al., 1990; Barbeau, Rossignol, 1987; Hodgson et al., 1994; de

Léon et al., 1998b) and the recent reports of improved locomotor skills after treadmill training in humans (Wemig et al., 1995) highlights the importance of sensory-motor experience for motor recovery afier SCI. In addition, it has been shown that static or dynamic sensory-motor training provides a powerful stimulus for learning and induces adaptations in the central nervous system (CNS) that profoundly influence motor performance (Hodgson et al., 1994). We therefore decided to manipulate exercise in order to determine which of the parameters associated with treadmill locomotion produce the most extensive recovery: mechanical load, reciprocal HL movements or both. Understanding the neuromotor drive would then allow us to establish new and improved neurorehabilitation techniques.

The specifici^ of training has a marked impact on motor performance after

SCI which may potentially hamper carry over to other tasks. For example, following

T 12 spinal cord transection (TX) lesions cats were trained to either step on a moving treadmill belt, stand on a non-moving treadmill belt, or were untrained (Hodgson et al.,

1994). Comparison of stepping abilities across groups during treadmill locomotion

48 following 3 months of training in their respective tasks revealed that animals trained to step performed best on the moving treadmill (Hodgson et al., 1994). More

importantly, the stepping performance of animals trained to stand was maricedly worse than injured cats untrained on either task (Hodgson et al., 1994). Likewise, step trained cats were not as proficient at static standing as those trained to stand (Hodgson et al., 1994). Thus, skill learning occurred specific to the demands of the task but these improved fimctional capabilities did not transfer to novel tasks.

In the current study, we compared the therapeutic efficacy of three different training programs which manipulate two parameters of exercise: the application of minimal or extensive mechanical load, in combination with static or dynamic training.

The training tasks that we have chosen - standing (ST; maximal hindlimb load/weight bearing, no rhythmicity), swimming (SW; minimal hindlimb load/weight bearing; maximal rhythmicity) and treadmill (TM; maximal hindlimb load, maximal rhythmicity) - could easily be implemented in the clinic and represent varying degrees of the demand for weight bearing and rhythmic reciprocal activi^ fi-om the hindlimbs

(HL). Our hypothesis was that tasks which involve both weight bearing and rhythmic reciprocal activation of the HLs would best improve locomotor recovery following

SCI.

Weight Bearing

The process of bearing weight through the joints o f the HL activates cutaneous receptors, causes joint approximation and sends sensory afferent information into the spinal cord (Gordon, 1991). This sensory afferent information has been shown to play

49 an important role in the regulation of locomotor rhythms (Rossignol et al., 1988).

Weight bearing activities have also been shown to increase bone mineral density (Fiore et al., 1996; Collet et al., 1997), prevent muscle atrophy and block conversion of muscle fiber phenotype (Roy et al., 1991; Lovely et al., 1990; Roy et al., 1998). In

Chapter 3, we address the changes in HL skeletal muscle properties that result from a

1.1 mm spinal cord contusion injury and the impact that exercise training has on these changes. In this paper, the behavioral and neuroanatomical effects are examined.

R hythm Generation

Locomotion is characterized by smooth, coordinated and rhythmic activation of the forelimbs (FL) with the HLs (Cruse, Wamecke, 1992). Electromyographic (EMG) and ventral root recordings confirm the stereotypic pattern of activation that is observed during locomotor behavior (de Leon et al., 1994; Grillner, Shik, 1973;

Forsseberg et al., 1977). Studies in various animal species; lower vertebrates (Grillner et al., 1998), small mammals (Rossignol et al., 1988), dogs (Naito et al., 1990) and perhaps humans (Bussel et al., 1996; Calancie et al., 1994) suggest that the rhythm generating mechanisms for locomotion reside in the spinal cord. These mechanisms can be activated even in the isolated cord (i.e. transected) via electrical stimulation

(Orlovsky, 1972; Pearson, Rossignol, 1991), pharmacological activation (Harris-

Warrick, 1988) or exercise training (Rossignol et al., 1996). Given that sensory afferent information helps regulate locomotor rhythm (Rossignol et al., 1988), it is possible that rhythmic tasks may facilitate activation of central pattern generators

(CFG) leading to improved locomotor behavior.

50 Rationale for Treatment Selection

We were interested in determining which o f the exercise tasks chosen (TM,

SW, or ST) most influences motor recovery from SCI given that some spontaneous locomotor recovery occurs without any treatment intervention. For TM, animals performed quadrupedal stepping on the treadmill because we wanted to train for improved FL-HL coordination during open field locomotion (OFL) by requiring activation of both shoulder and pelvic girdle recruitment during training. Although treadmill locomotion is easier to elicit using bipedal stepping and by actively assisting limb advancement and placement during training (as is done during TM training after spinal injury in some labs, personal communication, R. de Leon, C. Flynn), the stepping skills would be less likely to carryover into improved FL-HL coordination in the open field. Therefore these techniques were not applied in the current study.

Based on our hypothesis, we proposed that rats who participated in TM training would show the greatest locomotor recovery compared to the other exercise groups and especially the injured unexercised rats.

Swim training was chosen as a comparison group because: (a) the recruitment order of HL muscles and limb segments is similar to a very fast step cycle on the treadmill (Gruner, Altman, 1980; Roy et al., 1985; Roy et al., 1991), and (b) the requirements for trunk control are diminished given that the body is suspended in the water which provides buoyancy. Thus swimming may place fewer demands for motor control on the lesioned animal than locomotion while providing an environment which encourages rhythmic HL movement (sink or swim). However, the increase in cycle

51 frequency during swimming compared to treadmill walking and the lack of a HL

weight bearing support phase such as occurs in overground locomotion led us to

hypothesize that S W training would improve outcomes compared to standing trained

and untrained but not TM trained rats.

Lastly, standing training in this study was performed differently than in previous studies where animals were required to stand in a normal quadrupedal posture on a non-moving treadmill belt (Hodgson et al., 1994; de Leon et al., 1998a) with load distributed through all four limbs. In addition, previous stand training was performed using a full-length trunk harness and periodic tail support making it unnecessary for the animal to develop and maintain trunk control. In the current study, we were interested in seeing if training static standing balance and maximizing the load distributed to the HLs using an upright, rearing posture would influence functional outcome. This rearing posture required the animal to support the majority of its body weight through extended HLs, using the FLs on the upright surface for postural adjustments. One question we have considered regarding recovery of coordinated locomotion revolves around the issue of trunk control: Do animals which lack coordinated locomotion in the open field following contusion injury lack the ability to temporally link motor neuron pools innervating appropriate FL and HL muscles which results in coordination or do they lack proper trunk control which disallows expression of coordinated behavior because they are posturally insecure (Ten

Cate, 1960; Ten Cate, 1962; Ten Cate, 1964). Testing the importance of trunk control for the execution of coordinated FL-HL movements on the treadmill confounds the

52 issue by changing the type and intensity of afferent input normally experienced by the spinal cord compared to overground locomotion.

Onset of Training

Early exercise intervention following neurological injury generally shows greater recovery of function compared to later or no intervention (Hodgson et al.,

1994; Wemig, Muller, 1992; de Leon et al., 1998b). However, early forced-use exercise in animal models of brain injury has shown negative consequences with respect to both the neural lesion size and behavioral outcome (Kozlowski et al., 1996).

Thus, any new and/or early treatment intervention has the potential to negatively impact on both the behavior (locomotion) and the neural lesion size (area and volume) regardless of its location in the brain or spinal cord. We therefore evaluated lesion parameters in the spinal cord, (location, area, and volume) to determine whether our training paradigms exacerbated the contusion injury.

To our knowledge, this is the first study to evaluate whether specific training regimens improve open field locomotor behavior in adult animals following SCI.

Most previous studies have addressed training-induced improvements only for the task on which the animals were trained (Smith et al., 1982; Rossignol et al., 1996; Barbeau,

Rossignol, 1987; Lovely et al., 1986; Hodgson et al., 1994; de Léon et al., 1998b; de

Léon et al., 1999; Edgerton et al., 1997; Belanger et al., 1996; Bregman, Goldberger,

1983; Robinson, Goldberger, 1985). The importance of transference of locomotor behaviors beyond the training protocol is the gold standard for new clinical interventions.

53 Lastly, changes in reflex and sensory processing have been identified caudal to

the spinal cord injury (Thompson et al., 1992; Advokat, Duke, 1999; Christensen,

Hulsebosch, 1997; Siddall et al., 1995; Kerasidis et al., 1987), with chronic pain

conditions affecting 11-94% of all SCI patients (Yezierski, 1996; Davidofif et al.,

1987). Segmental reflex activity has been shown to be modulated by exercise training

(Trimble et al., 1998; Skinner et al., 1996) and in our preliminary studies it appeared

that other sensory processing mechanisms may also be altered by exercise. Therefore,

flexor withdrawal (FW) responses and cutaneous sensation using von Frey Hair (vFH)

testing were recorded before and afier SCI to assess changes induced by injury and

exercise.

M ethods

Subjects

Forty-seven female Sprague Dawley rats weighing 250-300 grams at the start

of the study were randomly assigned to one o f five treatment groups; treadmill training

(SCI-TM, n=7), swim training (SCI-SW, n=6), stand (SCI-ST, n=9), untrained (SCI

No-Ex, n=6), and laminectomy controls (LAM CTL, n=7). A triple numbering system

was utilized in this study such that the evaluators were blinded to pre-operative training performance and exercise group when all outcome measures were collected.

Seven rats were removed from the study secondary to complications associated with the surgery. Data from one rat was not included in the analysis due to a lack of participation in the task: a swimmer who never swam.

54 In the first half of the experiment trained and untrained animals were randomly

housed 2-3 rats per c%e so that injured, non-exercising controls were housed with

animals in the exercise training groups and LAM CTLs. In the early stage of this

study, we believed that the performance of the injured non-exercising animals (SCI

No-Ex) showed greater motor behavior compared to our previous observations of

hundreds of animals with a similar severity of injury performing the same tasks used in

the current study. Although we were masked to the group assignment of the rats, we

came to this understanding when we failed to have any low performing animals, as

would be expected of animals with moderate SCI and no exercise. Given that the only

difference between the current animals and those of the past was the mixed housing

condition, we added a second group of injured rats (n=6) while the experiment was ongoing. These rats were housed together, 2-3 per cage, and served as our SCI No-Ex control group. Prior to breaking the code, we decided not to use the data fi’om the mixed-housed control group for statistical analysis.

Further comparison of our SCI No-Ex rats after mixed versus segregated housing confirmed our subjective beliefs. First, an increase in the extent of recovery was observed in the first week for randomly housed rats. Mean Basso, Beattie,

Bresnahan locomotor (BBB) scores for mixed and segregated rats were; 1.25 vs. 1.17 at

1 dpo; 1.0 vs. 0.17 at 3 dpo; 4.25 vs. 2.83 at 5 dpo; and, 7.0 vs. 5.92 at 7 dpo, respectively. Second, the mean rate of change in BBB scores from 1 to 7 dpo was

5.25 for the mixed-housed rats vs. 4.25 for the segregated rats. Finally, even in later stages of recovery, BBB subscoring (see Appendix A, BBB Subscore Scale) revealed a

55 slight benefit for mixed housing with an average score firom 5,6, and 7 weeks post injury of 1.5 points compared to i.0 fix>m segregrated animals. These results should be systematically studied with larger groups of animals and bring up an interesting question with regard to the influence on fimctional outcome that a trained animal has on its non-trained cage mate following neurological injury.

SCI Surgical Procedures

Moderate SCI was produced using the OSU injury device described in detail elsewhere (Bresnahan et al., 1987; Stokes et al., 1992). Briefly, animals received injections of antibiotic (gentocin 0.25 ml sc) and ketamine-xylazine (80 mg/kg, 10 mg/kg ip) prior to surgery. Removal of the T8 lamina exposed the meningeal surface surrounding the spinal cord prior to placing the rat in a spinal frame for stabilization.

The impact probe was lowered onto the dura to a pressure of 3 kdynes before the surface of the cord was displaced 1.1 mm over a 20 ms epoch (Stokes et al., 1992;

Berhmann et al., 1992). This produces a moderate spinal cord injury (see histology below). Bleeding was stopped before suturing the incision in layers. Subcutaneous lactated Ringers’ solution (5 ml) and antibiotic spray were administered after completion of the surgery. Bladders were manually expressed 2-3 times daily until spontaneous voiding returned (~ 2 weeks). In addition, oral Vitamin C was given daily to all animals in order to prevent urinary tract infections.

Exercise Training Paradigms

Task Acquisition. Prior to surgery animals were acclimated to their respective tasks (TM, SW, ST, see Appendix B) during daily sessions for 1 week. T readm ill:

56 Animals perfonned daily quadrupedal locomotion until they could maintain a forward

position on the treadmill belt moving at 11-13 m/min and continuously drank from a

liquid dispenser containing sugar water. Negative reinforcement (tail shock) was not

used during training in this study. Swimming: Rats learned to swim from one end to the other of a 75 cm long x 48 cm high x 30 cm wide glass tank filled with tap water maintained at 35°C. After each pass animals were removed from the end of the tank.

Following a short rest, the length of which depended on their past performance, i.e. 30s if no signs of stress to 2 min if sinking, rats were replaced in the starting location. This facilitated a straight swimming trajectory across the tank and prevented escape behaviors. In addition, rats received intermittent positive reinforcement for successful trials (finit loop). Standing: Rats were trained to stand on their HLs in a small

Plexiglas container for food reward (25 cm long x 14 cm wide x 30 cm high). During the first 10 minutes of a 20-minute training session, animals were continually enticed to search/explore on extended HLs for food and drink (apples, peanut butter, sugar water positioned at the upper edge of the container). In the second 10 minutes, animals continued to be enticed with positive rewards but in addition were replaced immediately upright onto their HLs each time they attempted to place their forelimbs on the floor of the tank.

Lastly, animals not engaged in daily training (SCI No-Ex, LAM CTL) received fruit loop rewards in their home cage 3 to 4 times per week. These animals were also handled for 10 min, 2 x week for the duration of the study in order to facilitate weekly

57 testing procedures. In our experience, animals that are not well handled can show significant variability in behavioral testing situations.

Dailv Exercise Training. We have observed a predictable stress response evoked by the injury itself which is marked by dark red potphyrrin expression around the animals’ eyes and nose early afier injury. This response tends to resolve within the first 4 days afier injury. In addition, supplemental fixxl and subcutaneous fluids to treat the dehydration that sometimes occurred following injury were no longer necessary at this time point. Therefore, training began 4 dpo for all exercise groups and lasted for 20-25 min per day, 5 days per week, for 7 weeks. Treadmill: Trunk support was provided as needed by a custom-made lycra vest that had holes cut out for the FLs and velcro closure on the back. The vest extended fiom the shoulder girdle down to the end of the rib cage. It was attached to a spring which was suspended on a cross bar located approximately 25.4 cm above the forward part of the treadmill belt. The spring served to prevent the animal fiom drifiting back on the treadmill belt but generally did not provide unweighting. On some vests, an extended piece of lycra could be unrolled down to the hip flexor region and used to support the lower trunk. Elastic supports were sewn laterally on the extended vest and were hand-held or attached to small hooks placed on the side walls of the treadmill to provide hindquarter support when necessary. Generally, animals required lower trunk support (unweighting) early afier injury, however, afier approximately 3-4 weeks of training animals were capable of stepping without lower trunk support. This extended piece of lycra was then only used occasionally and sometimes to prevent the rat fiom turning around on the

58 treadmill. Tail pinching which has been shown to improve gait patterns o f SCI

animals on the treadmill (Roy et al., 1991; Lovely et al., 1986; Edgerton et al., 1997),

was not employed in this study. The amount of training time actually spent stepping

was recorded for each animal (Table 2.1) in order to document a training effect.

RAT Session 1 (4 dpo) Se ssio n s (1 0 dpo) Session 29 (46 dpo) # Step Time Belt Speed Step Time Belt Speed Step Time Belt Speed R12 4:40 6 m /m in 16:36 6 m/min 20:00 18 m/min L12 4:00 5 8:00 7 20:00 12 B13 1:33 6 9:38 6 20:00 17 N15 3:22 6 18:00 6 20:00 13 B1 8:20 6 15:00 6 20:00 13 B4 7:23 5 14:40 6 19:45 13 L4 4:09 6 17:10 6 19:45 12

Table 2.1 : Improvement in stepping time (min) and belt speed (m/min) for each rat during treadmill locomotion on training sessions 1, 5, and 29. Stepping behavior included weight supported stepping as well as attempts made by forward limb advancement without weight support. Total training time was 20- 25 minutes per session. Belt speed was the highest speed tolerated on that session.

Swimming: Animals swam from one end to the other of the swimming tank and were

then removed. Rest times were gradually shortened from 2 min to 30 sec as their

performance improved (rats began eliciting HL cycling movements and were not

sinking). Generally, rest times were shortened approximately 15 seconds every week.

Time actually engaged in swimming was noted for each animal (Table 2.2) to record training effects. 59 RAT # Session 1 (4 dpo) Se ssio n s (1 0 dpo) Session 29 (46 dpo) Swim Time Swim Time Swim Time L I 2:22 2:42 11:55 L3 3:00 2:12 10:38 N4 4:02 3:36 10:50 N5 3:03 2:00 10:25 R4 3:18 3:12 13:17 R6 2:16 2:25 10:02

Table 2.2: Improvement in swimming performance as measured by total swimming time (min) on training session 1, 5 and 29 for each anim al. Tim ing values do not include the time spent placing and removing the animal &om the tank. Total training was 20-25 minutes per session.

Standing: Initially rats squatted on their haunches with their FLs placed on a 10 cm x

10 cm X 5 cm box in order to provide sufficient trunk support to facilitate weight bearing on the HLs. The box was removed by the fourth training session and as HL weight bearing performance improved the height o f food reward was raised for the training session. The highest height at which the animal successfully retrieved food rewards and the total time spent in HL weight bearing postures was recorded for each animal (Table 2.3) as a measure of training effects.

60 RAT Session 1 (4 dpo) SESSION 5 (10 dpo) Session 29 (46 dpo) # Stand Time Bottle H t Stand Time Bottle Ht Stand Time Bottle Ht N3 19:00 □ 12 cm 15:50 15 cm 14:44 21 cm R1 16:44 □ 12 11:10 15 14:15 20 R2 17:52 □ 12 15:42 15 15:10 21 R3 19:35 □ 12 15:09 14 14:54 21 N13 14:30 □ 15 13:16 16 12:01 20 N14 12:10 □ 15 7:00 16 12:40 20 R14 14:48 □ 11 12:19 16 12:55 19 N16 16:12 □ 11 14:35 16 13:54 20 BIS 15:34 □ 17 10:31 16 11:37 20

Table 2.3: Improvement in standing performance for each rat on training session 1,5 and 29. Time spent standing (min) on the hindlimbs and the maximal reward height reached (bottle h t, cm) increased across sessions for all animals in the group. Initially after SCI, rats were unable to independently support weight with the hindlimbs. Therefore, the animals were propped in an upright position using a small box under the forelimbs (□) until training session 4. Total training was 20-25 minutes per session.

Behavioral Testing

Locomotion. Prior to surgery, animals were acclimated to open field

locomotion (OFL) testing procedures which involved a 4 min observation period of

locomotion in a plastic wading pool (90 cm diameter). Testing by two examiners

using the BBB Locomotor Rating Scale (Basso et al., 1996; Appendix C) occurred preoperatively, every other day during the first week after injury and then weekly until

sacrifice at 7 weeks po.

6 1 Walkway Analysis. Rats were operantly conditioned to trayerse an eleyated wooden platform (14 cm w ide x 200 cm long) for fruit loop reward until the anim al crossed the platform ten times without stalls (daily sessions x 1 week). Analysis of locomotor behayior on the eleyated walkway (WW), following a 1.1 mm contusion injury to the spinal cord, reyealed that consistent plantar stepping did not occur in all animals and thereby precluded direct kinematic analyses. Howeyer, since motiyated, constrained locomotion often produced faster and more consistent gait patterns than when walking at self-selected speeds (i.e. as in OFL testing), locomotor characteristics were often easier to identify on the WW. In addition, we could videotape a clear unobstructed lateral view o f the animal during WW locomotion and compare anim als based on the quality of their walking performance. Three passes were eyaluated for quality of stepping patterns. As a conseryatiye measure of WW locomotion, we used the best three trials for each animal thereby giving the SCI No-Ex group the best opportunity to replicate the performance of exercised rats and minimize between group differences. (Eyaluation form and decision rules shown in Appendix D). Following videotape analyses, two obseryers (KH and DMB) masked to the experimental groups independently rank ordered from best to worst all animals in this study (Figure 2.2).

Animals were ranked based on the dominant HL moyement patterns obseryed.

Discrepancies in ranking between the 2 obseryers were resolyed by joint obseryation of the WW videotape.

62 Tactile Sensitivity. Withdrawal thresholds were determined on a subset of animals (n=20,4 to 6 rats per group) by applying a graded series of von Frey Hairs

(vFH, Stoelting, Wood Dale, XL; 8.5 to 125 g) to the plantar surface of the right hindpaw. The SW trained rats were not included in this analysis due to the fact that their training was completed prior to initiating the vFH analyses. Each vFH monofilament is calibrated to bend at a force that is approximately 10 fold greater than the previous filament (the kit ranges fiom a 5 mg force to 446 grams). Rats were acclimated to the testing procedures (4 x 20 min session) prior to the onset of behavioral testing. Animals were placed on an elevated Ya in.wire mesh with an inverted plastic cage (20 x 10 x 10 cm) confining their movement space. The 50% withdrawal threshold was determined using the up-down method of sensory testing

(Lindsey et al., 1999; Chaplan et al., 1994; Dixon, 1948). Pre-operative testing established baseline withdrawal thresholds as follows. Afier a 10 minute acclimation period in the apparatus, an 8.5 g vFH was applied to the plantar surface of the foot

(approximately 1 cm posterior to the fat pad of the middle phalange) fiom underneath the elevated wire mesh floor. The vFH was applied with a pressure that caused a slight bend in the hair after which the stimulus was removed. If the rat produced a brisk retraction of its hindpaw, the next lower vFH in the series was applied. If the rat did not retract its hindpaw in response to the vFH stimulus, the next higher vFH in the series was applied. Afier 20 stimulus presentations, the lowest gram force which produced a retraction at least 50% of the time determined the withdrawal threshold.

Thresholds were measured preoperatively, and 7,21,28,35,42, and 49 dpo SCI. The

63 1 week values were discarded, however, due to the limitations in our protocol requiring animals to be placed in a small space which limits their movement for testing. Given this constraint we were unable to accurately assess if animals confined to this small environment in the early paralytic phase possessed the trunk control to shift their body weight and produce a retraction response with their HL. It appeared that several animals flinched during testing but were unable to shift their weight off the HL thereby preventing the retraction response.

Segmental Reflex Testing / Flexor Withdrawal. Animals were videotaped for analysis of HL responses to pinch stimuli applied between the second and third metatarsal bones of the right hindpaw. A Panasonic wv-cl3S0 CCD video camera connected to a Panasonic VCR (image capture 60 fields/sec) was positioned at a 90 degree angle recording the lateral side of the animal. Images were down loaded onto a personal computer and analyzed using the Peak Motus Motion Analysis System.

Movement time and excursion of hip, knee and ankle movements were determined.

The greater trochanter, lateral femoral condyle and lateral malleolus were used to approximate the center of joint rotation for the hip, knee and ankle. Images were then hand digitized and data optimally smoothed with a Butterworth filter. Some rats rotated out of the 2-D plane of the camera and therefore their movement angles were inaccurate and were not included in analyses (Figure 2.5).

64 In addition, the pinch stimulus intensity necessary to evoke a flexor withdrawal

(FW) response of the right HL was recorded using the subjective rating scale that

follows:

1. Light compression of the thumb and forefinger (~ 25 g or less).

2. Moderate compression of the thumb and forefinger which results in some

deformation of the skin and underlying tissues (~30 to 55 g), (stimulus

intensity necessary to evoke a withdrawal response in a normal animal).

3. Heavy compression of the thumb and forefinger to the point which further

compression would break the skin (>60 g).

Since this test involves subjective evaluation of stimulus intensity we needed to

control as many variables as possible in order to yield reliable data. Therefore, a single

examiner, who was masked to the condition of the animal, randomly evaluated rats on

each testing day. Also, testing always occurred at approximately the same time o f the

light dark/cycle on the appropriate testing days.

Several attempts were made to quantify and calibrate FW stimulus intensity.

For example, a force transducer attached to the forefinger of the experimenter during pinch testing did not yield reproducible results. Also, several variations of forceps calibrated to deliver a set force when applied to the paw interfered with the movement responses. The previously described vFH testing protocol, therefore, was added to gain additional objective data for HL sensation.

65 Spinal Cord Histology

At the termination of the study, following an overdose of ketamine/xylazine anesthesia, a 1 cm block of unfixed spinal cord tissue was carefiUly dissected at the lesion or laminectomy site and postfixed in 10 % neutral buffered formalin. All tissue blocks were stored in fixative until they were embedded in paraffin (m in im u m o f 3 days). Tissue was then cut in the transverse plane (20 pm) with a rotary microtome and every fifth section was mounted on gelatin coated slides. Tissue was stained for myelin using Luxol Fast Blue (LFB) and the lesion epicenter was identified as the section containing the largest central core lesion with the least myelin stained tissue.

The average location of the tissue section representing the lesion epicenter was used as a reference for identifying the “epicenter” in LAM CTL animals. The spared tissue and total cross-sectional area were measured using a Zeiss Axiophot Microscope attached to MCID-M4 Image Analysis System (Imaging Research Inc., Ontario,

Canada). Images were captured using a Dage CCD72 8 bit camera at 3.125 x and projected in black and white onto a 38 cm monitor. Digitized tissue sections were manually outlined onscreen using a P2 mouse while viewing the section under 20 x magnification. White matter was considered “spared” if the myelin staining was dense, contiguous and grossly normal in appearance, with little or no gliosis and few swollen axons or vacuoles observed (Berhmaim et al., 1992). Tissue sparing was expressed as a percent: area occupied by spared white matter per total cross sectional area of the cord measured at the lesion epicenter.

6 6 Lesion V olum e

Lesion volume was determined from the 1 cm block o f harvested spinal cord tissue using the Cavalieri method (Michel, Cruz-Orive, 1988) and area assessment point counting procedure on the Stereologer vl.lb (Systems Planning and Analysis,

Inc., Alexandria, VA, area per point = 19,528 um^ (SPA, 1997). Systematic unbiased random sampling was used to select every sixth saved LFB stained section (12-14 sections per animal) which was projected via a CCD camera onto a computer screen at

6 X magnification. Sampling probes overlaying spared myelinated tissue were selected. Gray matter was not evaluated in this process. Following analysis of all appropriate sections for a given animal, the computer generated an estimated volume for spared myelinated fiber in mm^.

Muscle Dissection Procedures

At sacrifice several key HL muscles were dissected, wet weighed, pinned to cork at the in situ length and frozen in liquid nitrogen until further processing.

Analyses of the changes in skeletal muscle induced by SCI at 7 weeks post injury and the impact of exercise training upon them are described in Chapter 3.

Statistical Analvsis

BBB Scores were analyzed using repeated measures ANOVA. The Kruskal-

Wallis test was used to analyze group position on the WW ranking data. Also, Fisher’s

Exact Probability test was used to analyze consistent stepping behaviors among groups on the WW and the Spearman Rank Order Correlation Coefficient was used to

67 determine the degree of agreement between the two experimenters on WW ranking.

The gram-force value for the vFH threshold was transformed to the Log lo scale according to Chaplan (1994) and manufacturers instructions. Repeated measures

ANOVAs were used to analyze the transformed data for vFH thresholds, as well as, for

Flexor Withdrawal stimulus intensity data. Flexor withdrawal movement data for each time point were analyzed using a one-way ANOVA. Scheffe’s post hoc tests were employed as needed. A significance value of p< .05 was determined a priori for all comparisons. All measures of variance will be reported as means ± sem unless otherwise noted.

Results

Open Field Locomotion

The average BBB score fi-om the SCI No-Ex group during the first week post injury progressed fiom 0 to 6 which corresponded to an improvement fiom flaccid paralysis to extensive movement in two joints and slight movement of the third in each

HL (Figure 2.1). The only significant interactions that occurred for comparison of groups over time in the BBB test came during the first week after injury. At 5 dpo

SCI-TM rats scored significantly higher than SCI No-Ex rats (4.5 ± 0.8 sem vs 2.8 ±

0.5, p< .05). In addition, at 7 dpo SCI-TM and SCI-SW rats scored significantly higher than both SCI No-Ex and SCI-ST rats (SCI-TM 8 ± 0.4 sem, SCI-SW 7.4 ± 0.7,

SCI No-Ex 5.9 ± 0.8, SCI-ST 5.9 ± 0.6, p< 05). These scores reflect that SCI-TM and

SCI-SW rats showed greater HL movements and sweeping behaviors compared to the

6 8 21

18 o (0 15 c (0 12

LU 9 o€ C , o t / i 6 *— LAM CTL CD o - SCI NO-EX CD 3 €»— SCI-TM CD e - SCI-SW ^ SCI-ST 0

Pre 1 3 5 7 14 21 28 35 42 49 DAYS POST OP

Figure 2.1; Mean BBB locomotor scores (+ sem) over 49 days post SCI. Treadmill training (SCI-TM) and swim training (SCI-SW) significantly improved motor performance as compared to standing trained (SCI-ST) and untrained animals (SCI NO-EX) during the first week. *: SCI-TM significantly different than SCI NO-EX, p< .05. +: SCI-TM and SCI-SW significantly different than SCI NO-EX and SCI-ST, p< .05. Locomotor scores for all SCI groups were significantly lower than laminectomy controls (LAM CTL) at all time points (***: p< .001).

69 SCI No-Ex and SCI-ST trained groups. No significant differences were observed across all injured groups by 14 dpo and throughout the rest of the study. Locomotor patterns plateaued by 21 dpo (SCI-TM 10.3 ± 0.4 sem, SCI-SW 10.5 ± 0.3, SCI-ST

10.5 ± 0.2, and SCI No-Ex 10.8 ± 0.5). These values suggest that rats were capable of plantar stepping but deficits in FL/HL coordination, trunk control, paw position and toe clearance persisted. As expected, LAM CTL animals showed normal open field locomotor behavior scoring 21 on the BBB scale and were significantly different fiom

SCI animals at all times tested (Figure 2.1). Therefore, participating in TM and SW training following SCI facilitated recovery of early locomotor behaviors but did not significantly influence chronic locomotor patterns.

Walkwav Locomotion

A non-parametric statistical analysis (Kruskal-Wallis test) was used to determine if significantly more rats fiom one group ranked higher or lower than another group on WW locomotion. LAM CTL animals were significantly different in their group ranking compared to all other groups analyzed (best walkers with the highest ranking), but no other differences were observed between the other groups based on ranking alone (Figure 2.2). However, the ability to consistently step with both HLs on the WW following SCI is of important functional consequence (Figure

2.3) and did show significant group differences. Participation in TM and SW training did not improve consistent stepping abilities with 57% and 50% of the groups achieving consistent stepping, respectively, compared to 66 % of the SCI No- EX group. More importantly, participation in standing training activities appeared to

70 LAM LAM LAM LAM LAM LAM LAM ...TM... NoEx SW SW NoEx TM NoEx o TM z TM .. ST . SW TM NoEx ST ST ST NoEx TM -----SW— ST ST NoEx ST ST SW TM I SW

No Occa Freq/Cons Occa Consist Step Step Step Coord Coord

Figure 2.2; Ranking of walkway locomotion 7 weeks after moderate spinal cord contusion (LAM, laminectomy control; NoEx, untrained; TM, treadmill trained, SW, swim trained; ST, stand trained). Ranking was based on critical features of locomotion: frequency of stepping (none, occasional [Occa], or frequent to consistent [Freq/Cons]) and incidence o f coorcUnation (occasional or consistent). Note that six TM or SW animals achieved occasional coordination while no ST rats demonstrated any coordination. Quartile ranges are denoted by dotted lines.

71 2 100 O. Q. LU 80 & >- 60 LU & co z 40 o 0 1 20 *** I I LAMCTL SCINO-EX SCI-TM SCI-SW SCI-ST

Figure 2.3: The percentage of animals in each group that demonstrate consistent stepping during walkway locomotion 7 weeks post SCI. Note that none of the stand trained animals (SCI-ST) consistently stepped compared to 50-66% o f the untrained (SCI NO-EX), treadmill trained (SCI-TM), and swim trained (SCI-S\^ rats. 100% of laminectomy controls ^A M CTL) consistently stepped. ***: significantly different fi’om LAM CTL, p< .001.

72 considerably impair the ability to consistently step on the WW. None of the standing trained animals were able to show consistent stepping behavior on the WW (0%).

Analysis of this data set was hampered by low samples sizes, however, since Fisher’s exact probability test failed to demonstrate significant differences between a group of animals unable to consistently step and those that step at least half of the time (SCI-

TM and SCI-SW). The SCI-ST group was significantly lower than LAM CTLs. If the sample sizes were doubled for each group and the percentage of consistent steppers remained the same, all injured groups would be significantly different than LAM CTL animals, and SCI-ST rats would be significantly different fiom all other injured rats, as well as, LAM CTLs. Therefore, participating in ST training afier SCI impaired some aspects of locomotor recovery.

Videotaped observation of WW performance of injured animals also allowed for precise visual but not kinematic definitions of coordination. Our definition of coordination includes the stipulation that the FLs and HLs must alternate and one of the paired limbs (i.e. FL and ipsilateral or diagonal HL) must move into swing before the other paired limb was less than halfway through the swing phase. Three coordinated gait patterns were described with these general characteristics and are presented fiom higher to lower skill level with the percentage of animals per group demonstrating the pattern: (a) simultaneous diagonal FL-HL advancement (LAM

CTLs 100%, SCI trained rats 0%, SCI untrained rats 0%); (b) four point diagonal FL-

HL advancement is a slower version of the diagonal pattern (above) in which each limb moves sequentially and independently (SCI-TM 57%, SCI-SW 33%, SCI-ST 0%,

73 SCI No-Ex 50%); and finally, (c) simultaneous ipsilateral FL-HL advancement (SCI-

TM 14%, SCI-SW 33%, SCI-ST 0%, SCI No-Ex 50%). Lastly, as stated previously,

after proper conditioning on the WW only a diagonally coordinated pattern was

observed in our LAM CTLs and all such animals consistently traversed the WW in this

manner.

Flexor Withdrawal

Following SCI, a lower stimulus intensity elicited FW of the HL compared to

LAM CTLs (Figure 2.4). This test involves a subjective rating system with respect to

stimulus intensity, however, the experimenter (KH) was masked to the condition of the

animal thereby eliminating experimenter bias and inter-rater variability lending greater

confidence in any between group differences we detected. We found a significant

main effect for treatment group collapsed across time (Figure 2.4) such that SCI No-Ex

animals demonstrated a FW response to a significantly lower stimulus intensity than

most other groups (SCI No-Ex 1.5 ± 0.09, LAM CTL 1.9 ± 0.04, SCI-TM 1.8 ± 0.06,

SCI-ST 1.8 ± 0.06; p<.05). The stimulus intensity used for the SW group was not

significantly different than the SCI No-Ex group (p>.05). However, using compound contrasts which treat all exercise (EX) groups as one, we found that SCI-Ex rats (SCI-

TM, SCI-SW and SCI-ST) are not significantly different firom LAM CTLs (p> .05), but are different fiom SCI No-Ex animals with respect to the stimulus intensity necessary to evoke FW (p<.05). No interactions between group and time post injury were seen. Kinematic analysis of FW at pre-op, 1 and 7 weeks post SCI included time to peak flexion and the summative angular excursion across the hip, knee and ankle.

74 o 2 in + c S E co z liJ co « - LAM CTL -J O - SCI NO-EX • 3 ■m— SCI-TM S 1 # - SCI-SW h“ “ SCI—ST CO Pre-Op 7 21 35 49 DAYS POST-OR

Figure 2.4; Stimulus intensity of pinch needed to elicit flexor withdrawal of the hindlimb in laminectomy control (LAM CTL), untrained (SCI NO-EX), treadmill trained (SCI-TM), swim trained (SCI-SW), and stand trained (SCI-ST) groups. The SCI NO-EX group was significantly difiërent firom SCI-TM, SCI-ST, and LAM CTL across time (*).

75 Preoperative and 1 week post-op values showed no differences in movement time or angular excursion between groups (data not shown). However, 7 weeks post injury,

SCI No-Ex rats moved significantly faster in response to pinch stimuli compared to

LAM CTLs (Figure 2.5, LAM CTL 0.20 ± 0.01 sem vs. SCI No-Ex 0.17 ± 0.01 ms, p<.05). It appears that engaging in exercise training attenuated the faster movement response to pinch stimufi (SCI-TM 0.18 ± 0.02, SCI-SW 0.22 ± 0.02, SCI-ST 0.18 ±

0.01 ms) such that SCI-Ex animals were no longer different than LAM CTLs. These values were not significantly different firom the SCI No-Ex group (p> .05). At 7 weeks post injury, angular excursion of the HL tended to be less than LAM CTLs but this difference was only significant for the SCI-SW group (Figure 2.6; p<.05). von Frev Hair T esting

A main effect for group showed that the SCI No-Ex rats had a 50% threshold that was statistically different ft^om LAM CTLs (group means across time, SCI No-Ex

34 ± 5.4 sem vs. LAM CTL 75.8 ± 0 gram-force, p< 05). SCI-SW rats were not evaluated with the vFH protocol. Overall, SCI-TM (63 ± 7g) and SCI-ST (41 ± 7g) groups were not significantly different fi*om LAM CTLs or SCI No-Ex animals.

Interestingly, however, rats that participated in TM training initially showed decreased thresholds indicative of hypersensitivity (Figure 2.7) which, over time returned to pre­ op levels. Therefore, interactions of group by time showed that thresholds at 4 weeks po for both SCI-ST and SCI No-Ex were significantly lower compared to LAM CTLs

(SCI No-Ex 27.3 ±10 sem, SCI-ST 14.6 ± 4 vs. LAM CTL 75.8 ± 0 g-force, p<.05).

76 0.30

4 9 0.20

^

UJ W

T— ■ ' r LAMCTL SCINO-EX SCI-TM SCI-SW SCI-ST

Figure 2.5: Time to peak flexion for flexor withdrawal (mean + sem) at 7 weeks post SCI for laminectomy control (LAM CTL), untrained (SCI NO-EQ, treadmill trained (SCI-TM), swim trained (SCI-SW) and stand trained (SCI-ST) groups. **: SCI NO-EX sign^cantly faster than LAM CTL (p< .01). Number of rats per group in parentheses.

77 E «p 150

V < S 100

I I LAMCTL SCINO-EX SCI-TM SCI-SW SCI-ST

Figure 2.6: Summed angular excursion (mean + sem) of the hip, knee, and ankle joints during 6 e flexion phase of flexor withdrawal at 7 weeks post SCI for laminectomy control (LAM CTL), untrained (SCI NO-EX), treadmill trained (SCI-TM), swim trained (SCI-SW) and stand trained (SCI-ST) groups. •: SCI-SW sign^cantly less than LAM CTL (p< .05). Number of rats per group in parentheses.

78 (0 + 80 - C CO I 70-

3 60 - liJ g O u_

—I O * X 30 - CO UJ - m - LAM CTL " (K 20 • - o - SCI NO-EJi g 10 - -E h - SCi-TM SS O - A - SCI-ST u> Pre-Op 21 28 35 42 49 DAYS POST INJURY

Figure 2.7: Fifty percent response thresholds for von Frey hair stimulus (g) to the hindpaw of rats in laminectomy control (LAM CTL), untrained (SCI NO-EX), treadmill trained (SCI-TM),and stand trained (SCI-ST) groups across time. Note that the hypersensitivity for SCI-TM is ameliorated by 35 dpo. ♦: p< .05 vs. LAM CTL, +: p< .05 vs. LAM CTL and SCI-TM.

79 During weeks 6 and 7, SCI No-Ex rats (29.6 ± lOg and 24.8 ± 1 Ig, respectively) had lower thresholds compared to both SCI-TM and LAM CTL rats (SCI-TM 75.8 ± 0 sem, LAM CTL 75.8 ± 0 g, p<.05).

Lesion Epicenter

Laminectomy surgeries did not produce any noticeable pathology in spinal cord tissues as assessed by light microscopic examination of LFB stained sections

(data not shown). Contusion injury resulted in damage to cell bodies and axons located centrally in the cord, sparing a peripheral rim of myelinated axons (Figure 2.8).

There was no apparent sparing of gray matter at the injury center and the periphery contained swollen, collapsed and demyelinated axons. Moderate sized cystic cavities were noted centrally, as well as, in ventral and lateral funicular regions of the injured cord; however, a predominantly gliotic filled lesion epicenter was present. Electron micrographs taken from tissue 8 weeks following a moderate contusion injury

(Wrathall et al., 1998) revealed the presence of hypertrophic astrocytic processes which would be expected to occupy our lesion site as well. No qualitative differences in morphological presentation were apparent between LFB stained cords of SCI No-Ex and SCI-TM, SCI-SW and SCI-ST animals (Figure 2.8).

Quantitative analyses at the lesion epicenter showed a significant decrease in the percent of the cross-sectional area occupied by spared myelinated axons for SCI No-

Ex rats compared to LAM CTLs (Figure 2.9; SCI No-Ex 15.4 ± 4 sem vs. LAM CTL

80 Figure 2.8: Photomicrographs of myelin stain spinal cord section through the lesion epicenter of a representative animal from A: untrained (SCI No-Ex); B: treadmill trained (SCI-TM); C: Swim trained (SCI-SW) and D: stand trained (SCI-ST) groups. Scale bar: 150 um

81 80

? 70 (0 + 60 C o so E 40 O 2 30 ^ 20 O) 1 10

LAMCTL SCINO-EX SCI-TM SCi-SW SCi-ST

Figure 2.9: Percent spared white matter at the lesion epicenter (white matter per total cross-sectional area of the cord) for laminectomy control (LAM CTL), untrained (SCI NO-EX), treadmill trained (SCI-TM), swim trained (SCI-SW), and stand trained (SCI-ST) groups. Note that 75% of the cross-sectional area of the cord is occupied by white matter in LAM CTL rats. *: p< .05 vs. LAM CTL.

82 "U (0 s E ro "* E E Oz û : g c o

Ul > - LAM CTL SCI NO-EX SCi-TM SCI-SW SCI-ST

Figure 2.10: Volume of spared myelinated tissue (mm , mean ± sd) through a I cm segment of the spinal cord containing the lesion epicenter 7 weeks post SCI for laminectomy control (LAM CTL), untrained (SCI NO-EX), treadmill trained (SCI-TM), swim trained (SCI-SW), and stand trained (SCI-ST) groups. *: p< .05 vs. LAM CTL.

83 75.2 ± 0.5 p<.05). In addition, engaging in a moderate daily exercise program as

early as 4 dpo did not significantly exacerbate lesion size compared to SCI No-Ex rats

when evaluated 7 weeks post injury (SCI-TM 20.3 ± 2 sem, SCI-SW 24.4 ± 5, SCI-ST

14.2 ± 2 mm^ p>.05).

Lesion Volume

Lesion volume was analyzed on a set of serial sections spanning 1 cm rostral,

through and caudal to the lesion epicenter. Seven weeks alter a l l mm SCI, there was

a significant decrease in the volume of spared white matter (Figure 2.10) compared to

LAM CTL (SCI No-Ex 7.6 ± 0.7 sem vs. LAM CTL 25.9 ± 0.7 mm^, p< .05). More

importantly, participating in moderate exercise early afier injury and continuing for 7

weeks did not exacerbate lesion volume (SCI-TM 8.1 ± 0.7 sem, SCI-SW 7.9 ± 1.0,

SCI-ST 8.3 ± 0.7 mm^, p>.05).

Discussion

The major finding of this study was that treadmill and swim training facilitated early recovery of locomotor patterns as measured on the BBB scale. Importantly, however, videotaped WW analyses revealed that participating in stand training actually impaired some aspects of locomotor recovery after SCI (ability to consistently step).

Also, changes in sensory stimulus processing occurred secondary to SCI which were significantly attenuated by participation in an exercise training program (particularly

TM training). Furthermore, participating in exercise training at moderate intensity early after SCI did n o exacerbate t the lesion size.

84 Locom otion

Variations of the 5 point Tariov Scale have been used to analyze OFL following spinal cord contusion injury (Stokes, Homer, 1996; Berhmann et al., 1992; Kunkel-Bagden et al., 1992) with no or varied operational definitions making comparisons across laboratories difficult. The recent development of the BBB Scale (Basso et al., 1996),

Appendix B) has allowed for a more sensitive and reliable indicator o f OFL recovery and is now commonly used to evaluate locomotor behavior. Using the BBB scale, we identified small, yet significant behavioral improvements for treadmill and swimming trained animals during the early stage of locomotor recovery. However, even with use of the new sensitive BBB test, it was difficult to detect differences in locomotor patterns between groups during the late stages of recovery since most rats scored between a 10 and an 11 on the scale. Therefore, in addition to evaluating OFL with the

BBB scale, we also trained animals to locomote on the elevated WW for finit loop reward.

A critical element of locomotion and one that will greatly influence behavioral outcome is self-selected speeds for locomotion. The selection of locomotor speed can emphasize or mask locomotor deficits (Helgren, Goldberger, 1993). Furthermore, the motivated condition of the animal can significantly influence the speed o f locomotion

(Goldberger et al., 1990). Reproducible locomotor performance generally results from locomotion at faster speeds and therefore is preferred for analyses. In the current study, WW conditioning resulted in motivated locomotor passes that were consistent in level of control and generally faster in pacing versus self-selected walking speeds

85 observed in the open field (unpublished observations). Pertiaps of greater significance

is the constrained nature of the WW task. Although a relatively wide platform (15.24

cm) was used in the current study, greater precision for paw placement and better trunk

control are needed for WW versus OFL which may produce further separation of

animals according to locomotor skill. Also, the length of the pass is longer on the WW

than in the open field so that the stability of the locomotor pattern can be more

accurately assessed. Rats may be able to maintain locomotor patterns for a short

distance as in the open field test, but over longer distances on the WW we may see

their performance degrade. For instance, three animals judged to be consistently

stepping in OFL testing were unable to do so on the WW (data not shown).

Interestingly, one of the major findings of the WW analysis was the

identification of locomotor deficits observed in the standing trained animals. These

rats generally displayed poor trunk control, were more likely to dorsal step than SCI-

SW, SCI-TM, and SCI No-Ex rats and were consequently unable to consistently step

while walking on the WW (Figure 2.2).

Neural Control of Coordinated Locomotion after Partial Lesions to the

Spinal Cord

Several studies have described volitional open field and treadmill induced

walking capabilities following various partial lesions to the spinal cord (Basso et al.,

1994; Jiang, Drew, 1996; Brustein, Rossignol, 1998; Harris et al., 1994; Gôrska et al.,

1993; Bem et al., 1995; Little et al., 1988; Eidelberg et al., 1981; English, 1980).

Coordination is one parameter often assessed to compare locomotor capabilities across

8 6 different lesion models. While most studies show that animals regain some locomotor skill after partial lesions to the spinal cord, discrepancies exist as to which descending tracts support the coordinative interplay between the four limbs and maximize recovery from SCI.

It has been suggested that FL - HL coordination is the result of peripheral, propriospinal and supra-spinal inputs converging on CPG mechanisms located in the gray matter of the lumbar spinal cord (Rossignol et al., 1996; Brustein, Rossignol,

1998). Therefore, damage to any of these systems would be likely to result in impaired locomotor patterns. One would anticipate also that lesions to various areas within the spinal cord would result in different locomotor patterns as well. However, ventral spinal (Bem et al., 1995), dorsolateral funicular (Jiang, Drew, 1996), and dorsal column (English, 1980) lesions all result in a similar locomotor deficit including an impaired coupling between the limbs and a tendency for simultaneous advancement of the ipsilateral FL and HL. These data suggest that deficits in interlimb coordination patterns are related to the extent of spinal cord damage and not necessarily due to damage of a particular pathway (Jiang, Drew, 1996).

After a 1.1 mm spinal cord contusion injury, there is partial preservation of tracts which run in the periphery of the cord. Studies have shown at least partial damage to most major descending pathways in the cord afier a contusion lesion (Basso et al., 2000) with retrograde labeling revealing sparing in propriospinal, as well as brainstem regions, i.e. the lateral vestibular nucleus, reticular nuclei!, red nucleus, but generally not in the motor cortex whose role in locomotor recovery in rats following

87 SCI has been questioned (Muir, Wbishaw, 1999). However, similar to the results of specific lesions of the cord, recovery fiom contusion injury has been shown to be related to the overall amount of spared fibers at the epicenter (Noble, Wrathall, 1989;

Basso et al., 1995). The presence of FL-HL coordination is associated with greater over-all preservation of fibers at the epicenter. Therefore, it may be that the amount of sparing observed at the epicenter is the critical factor for the resumption of coordinated locomotion, perhaps allowing for a change in gain of segmental neurons which facilitate locomotor patterns. There may also be biomechanical constraints in need of consideration. For example, poor trunk control, irrespective of the particular tracts damaged, may lead to the ipsilateral limb advancement observed in previous experiments (Bem et al., 1995).

Walkwav Analvsis

We rank ordered rats on their WW performance based on the quality of their stepping patterns (i.e. plantar, dorsal or no stepping), the presence or absence of coordination, and their consistency across trials (Appendix D). Uninjured conditioned rats in our study produced only one locomotor pattern: a consistent diagonal coordinative pattern with the ipsilateral FL and contralateral HL. This pattern incorporated plantar stepping, normal trunk control with the nose and tail off the ground (i.e. no snifGng, and no tail dragging). This is in contrast to the 5 different locomotor patterns described for normal rats in non-motivated WW locomotion

(Cheng et al., 1996).

88 To accurately assess cocrdinaticii a minimum number of criteria should be met.

The animal should be motivated and performing in a continuous motion. In addition, the data analyzed should correspond with performance from the middle of the pass, not the beginning or end regions corresponding to the acceleration and deceleration phases. Adhering to these principles, we have found that a single locomotor pattern results from normal animals when adequate preoperative training and task acquisition are provided. This preoperative training likely accounts for the differences between the findings of Cheng et al. (1997) and the current results.

In the current study, using careful analyses of the videotaped WW performance, we did not observe the use of one consistent coordinative pattern across groups. The simultaneous advance of diagonal FL-HL pairs only occurred in LAM

CTL rats. The FL-HL coordination most frequently produced by the SCI-TM animals

(57% group) was the four point gait pattern, a slower version of the diagonal pattern mentioned above (see results). The SCI-SW (33%) and SCI No-Ex (50%) rats failed to demonstrate a predominant pattern but rather used the four point pattern or the ipsilateral FL-HL gait equally.

Spinal Cord Leaminp

As observed in previous studies (Barbeau, Rossignol, 1987; de Leon et al.,

1998a; de Léon et al., 1998b; Hodgson et al., 1994), animais in the current study showed significant improvement in performance with training on their respective tasks

(Table 2.1,2.2,2.3). Stand trained animals significantly increased the time spent and height achieved in static standing on the HLs. Swimmers increased the extent and

89 frequency of HL cycling (data not shown) as well as duration of their swimming

episodes. Treadmill steppers increased the total stepping time and maximum speed

achieved. Although early significant differences in BBB scores were identified for S W

and TM trained rats, the improvement in task performance listed above did not appear

to induce changes in chronic locomotor performance in the open field (Figure 3.1).

Treadmill versus Overground Locomotion

Treadmill locomotion yields a stereotypical HL stepping pattern (de Leon et al.,

1998b). During bipedal treadmill stepping, the FLs are placed on a platform and as the

treadmill belt moves the HLs are passively extended until the animal produces a step.

Afferent feedback from the moving limb plays a significant role in initiating and

modulating treadmill locomotion. The neural mechanisms necessary to elicit

locomotion on the moving treadmill belt are located in the spinal cord since animals

with complete spinal cord TX at low thoracic levels, with some trunk support, can

reproduce HL locomotor patterns (Roy et al., 1991; de Leon et al., 1998b). This is in

contrast to overground/OFL which is not dependent on segmental sensory afferents

(i.e. moving treadmill belt) but rather requires supraspinal (i.e hypothalamic,

mesencephalic locomotor region, reticulospinal) activation of spinal circuitry for the

initiation and perpetuation of gait (Grillner, Shik, 1973; Sinnamon, 1993; Armstrong,

1988). It may be that the execution of the propulsive phase (Ez-^Es) is where these two modes of locomotion differ the most (Smith et al., 1982).

90 Specificity of Traininy

To our knowledge, the relationship between exercise training in partially lesioned animals and carry over to OFL has not been previously investigated. Several studies have shown that animals subjected to severe SCI, i.e. complete TX of the thoracic cord and subsequent TM training, can leam to produce stepping patterns that closely resemble normal animals on the treadmill (Lovely et al., 1990; de Leon et al.,

1998b; Barbeau, Rossignol, 1987). As stated above, the locomotor capabilities of spinal cord transected animals is dependent on the afferent input provided by the moving treadmill belt and possibly the motion-dependent feedback elicited by the animals’ own movements (Guiliani, Smith, 1987). However, these animals with complete TX generally have no ability to produce volitional stepping movements in the open field and traditionally drag their HLs in extended positions during overground locomotion with the FLs (Roy et al., 1991; Smith et al., 1982; Basso et al., 1996).

In the current study, all groups significantly improved on their respective training tasks (Table 2.1, 2.2,2.3). However, despite the fact that most animals regain some locomotor ability following injury without specific training, no additional long term locomotor benefit was observed by engaging in the exercise tasks used in this study. In fact, stand training appeared to negatively effect some parameters of locomotion (see below). The lack o f a chronic training effect in OFL may have been due to training an insufScient length of time or intensity. Since not all groups trained on the TM, it is unlikely that poor propulsive responses (E2-*E3) are responsible for all deficits observed in OFL behavior. Perhaps a combined exercise group of TM and

91 ST training incorporating riiythmic HL activation, weight bearing and trunk control

training, might prove to be the most efBcacious treatment for this injury severity.

Data from the current study suggests that participating in ST training, but not

TM or S W training, may interfere with recovery of normal locomotor behavior. After

SCI, animals showed a significant increase in the extent and duration of HL weight

support during static standing with training (Table 2.3), however, this ability did not

carryover to better OFL. In fact, static standing training in this study, as in the

Hodgson et al. study (1994) using static standing training on the treadmill belt, resulted

in marked deficits in the resultant WW locomotor patterns (Figure 2.3, Appendix D).

Since this impairment introduced by standing training occurs both on and off the

treadmill (i.e. occurs in segmentally mediated locomotor behaviors like bipedal TM

training and supraspinally driven OFL) suggests that the mechanisms of “interference”

may be similar despite the difference in neural mechanisms necessary to elicit the

behavior. It is clear that training with a static “weight bearing-only” stimulus

condition is insufficient to improve voluntary locomotor behavior after SCI.

Reflex Testing and Somatosensorv Examination

Clinical investigations have revealed exaggerated segmental reflexes

(spasticity) (Robichaud, Agostinucci, 1996; Fung, Barbeau, 1994; Trimble et al., 1998)

and the presence of central dysesthetic pain (Davidoff et al., 1987; Yezierski, 1996)

caudal to SCI. We selected two tests to assess changes in hypersensitivity below the

injury: flexor withdrawal to pinch stimuli (Gale et al., 1985; Kerasidis et al., 1987) and vFH monofilament testing (Christensen, Hulsebosch, 1997).

92 Allodynia is a clinical condition, that can occur after neural injury, in Wiich a

normally non-noxious stimulus such as a touch, pinch or a thermal stimulus, becomes

noxious to the subject and provokes a withdrawal response. In FW testing performed

in the current study, a pinch stimulus was applied to the right hindpaw and an

allodynic response would be one in which a lighter than normal pinch intensity evokes

withdrawal. Similarly, with vFH testing a lighter than normal cutaneous stimulation to

the plantar surface of the paw using a lower gram-fbrce monofilament would evoke

paw retraction in an allodynic response.

We incorporated both FW and vFH testing for assessment of caudal cord

function following spinal cord contusion injury for two reasons. Flexor withdrawal was used as a global, semi-quantitative measure of sensory change while vFH testing provided more specific quantification of cutaneous sensation. Pinching the paw to evoke FW presumably stimulates deep muscle as well as cutaneous pressure and pain receptors and is graded based on subjective evaluation of stimulus intensity. The vFH protocol tests responses to cutaneous stimulation only and is based on the concept that the discrimination of stimulus intensity is based on a log-linear scale (a linear perception of stimulus intensity is correlated with a logio scale of actual gram force intensity) (Chaplan et al., 1994). It is likely that vFH testing is a more sensitive measure of cutaneous sensory changes than subjective ratings of stimulus intensity from pinch testing. However, both FW and vFH testing revealed the development of hypersensitivity after SCI.

93 Evidence of Hypersensitivity Caudal to SCI

A change in physiological properties of neuronal cells caudal to a neural injury

may play a role in the altered behavioral states observed after SCI. Decreases in the

intensity necessary to evoke flexor withdrawal (Helgren, Goldberger, 1993; Gale et al.,

1985; Siddall et al., 1995) and/or an ex%gerated movement response (Kerasidis et al.,

1987, current study) have previously been described during behavioral testing

following SCI. In addition, several labs have used electrophysiologic techniques to

evaluate changes in other segmental reflexes (H Reflex) and motor neuron properties caudal to a spinal cord TX lesion. Monosynaptic and polysynaptic reflexes following

SCI are easier to elicit and larger in amplitude in both human (Chang, Lien, 1991) and animal models (Hochman, McCrea, 1994a; Thompson et al., 1992; Hochman, McCrea,

1994b; Thompson et al., 1998). Loss of descending drive to motor neurons and intemeurons by complete transection of the spinal cord results in higher firing thresholds, transiently increased EPSP amplitudes and shorter duration after hyperpolarization potentials (Czeh et al., 1978; Cope et al., 1986; Hochman, McCrea,

1994a; Hochman, McCrea, 1994b). It is possible that decreased presynaptic inhibition and/or increased post-tetanic potentiation plays a role in the hypersensitivity of FW and vFH testing revealed in the current study (Thompson et al., 1998; Thompson et al.,

1992).

Interestingly, engaging in an exercise task (TM, S W, ST) attenuated the decreases in stimulus intensity for pinch testing observed after SCI. More importantly,

TM but not stand training ameliorated the decreased vFH thresholds observed

94 following SCI. Swim trained rats were not tested with the vFH protocol. Whether the

changes which result in a more normal response to sensory testing were due to changes

in peripheral receptor mechanisms or central stimulus processing is unclear (see

below). It is reasonable to suggest, however, that both phasic sensory-motor activity

(HL cycling) invoked by TM or SW training and tonic activation induced by stand training contributed to the attenuation of hypersensitivity to pinch stimuli identified in

FW testing. We observed an attenuation in response threshold to sensory stimulus testing (FW pinch) that appeared to be irrespective o f the type of training. The effect of differential training on vFH testing is remarkable but difficult to explain. Perhaps motor tasks (exercise) in general influence muscle receptors (evaluated in FW but not vFH testing) to a greater extent than cutaneous receptors and therefore all exercise tasks blunted the pinch stimulus response, whereas specific training paradigms may have differential impacts on cutaneous sensory processing. The latter would then lend itself to discrimination via vFH testing. Evaluation of SCI-SW rats with the vFH protocol would address whether phasic vs. tonic input into the spinal cord would prevent the development of hypersensitivity.

Relationship between Flexor Withdrawal and von Frev Hair Te.stiny

We have two separate measures which indicate lowered thresholds to sensory testing caudal to SCI (Figures 2.4 and 2.7) which lends further confidence that increased hypersensitivity in caudal circuits after SCI can be ameliorated or attenuated with exercise. The differential responses to the two tests may be due to the FW test being a less sensitive measure of sensory change. For example, in 86% of the cases in

95 which a decrease in threshold was identified via FW testing the vFH threshold was also decreased. However, in the reverse scenario, only 50% of the cases which showed decreased vFH thresholds also demonstrated a decrease in FW stimulus intensity (data not shown).

Outcome Measures for Allodvnia

In animal testing, consideration of a noxious stimulus as being painful requires evidence of supraspinally-mediated responses such as vocalizations, biting or escape behaviors coincident with stimulus application. Studies in animals with SCI

(hemisection lesions) have shown lowered vocalization thresholds upon vFH testing

(Christensen et al., 1996; Christensen, Hulsebosch, 1997). A reduction in the stimulus strength necessary to evoke flexor withdrawal or HL retraction (vFH), as determined in the current study, suggests greater sensitivity in the segmental circuitry of the cord

(mechanical allodynia) but does not definitively suggest a heightened response to painful stimuli (Christensen, Hulsebosch, 1997)

A recent study by Lindsey et al. (1999) compared graded contusion lesions of the spinal cord and the development of tactile sensitivity. This study found that rats subjected to a 25 gm-cm spinal cord contusion injury with the NYU device (a slightly more severe injury than the 1.1 mm displacement injury used in the current study) were more likely to develop mechanical allodynia (HL retraction, vocalizations and posturing with a normally non-noxious stimulus) than rats with a lighter injury (12.5 or a 6.25 gm-cm). Vocalization thresholds were recorded for vFH testing of the trunk only but did show a significantly decreased threshold caudal to the lesion (Lindsey et

96 al., 1999). Therefore in a model similar to the one used in the current study, hypersensitivity measured using traditional criteria for pain (vocalization) was observed rostral to the injury site. In the current study, vocalization was not used as a criteria for the development of allodynia because, once the animals (trained, injured or normal) were acclimated to the procedure for vFH testing, vocalizations did not occur.

Animals in the present study were also extensively handled for exercise training which may decrease the animals apprehension in testing conditions and change the nature in which the novel vFH stimulus is interpreted by the animal. In m ost cases when the rats produced a retraction they oriented toward the response-evoking stimulus and therefore we believe it was considered noxious by the animal.

Flexor Withdrawal Movement Response

In addition to lower stimulus intensities necessary to elicit FW, videotaped analyses of movement responses also show significantly faster movement times for

FW after SCI (Figure 2.5) and corresponds with exaggerated movements previously identified by Kerasidis et al (1987). Although all injured groups of animals had faster movement times for the production of FW, only the SCI No-Ex group was statistically faster than the LAM CTLs. Thus, in addition to attenuating the drop in stimulus intensity fi)r evoking FW, exercise appeared to normalize (slow) the movement response once FW was triggered. We are uncertain why SCI-SW rats moved through smaller ranges during FW testing. Perhaps increased resistive forces and sensory stimulation produced by the water encountered during swimming may influence motor neuron thresholds, ultimately resetting gain/depolarization thresholds and decreasing

97 movement responses. It is unlikely that lack of load (one difference between S W and

TM training) influenced this response since application of a load to the HLs would

require extensor activation and the phase of FW we examined was a flexor motor

neuron dominated event.

Lastly, the FW test we employed confounds stimulus intensity with the motor

response. For example not all animals were tested at each stimulus strength (i.e. 1,2,

and 3) so that the resultant movements could be compared. It is possible that faster

movement times were related to greater stimulus intensities required after injury,

however, none of our data show greater intensities, especially at the point where

significant motor responses were seen. In fact, we saw movement time changes at

seven weeks after SCI when intensity was reduced (Figures 2.4 and 2.5). Thus some

adaptation of the efferent limb of the response likely occurred.

Mechanisms

Several mechanisms have been posited to account for the lowered thresholds

observed following injury to the peripheral or central nervous system. Monofilament

(vFH) testing with varying grades of force have been used extensively for the detection

of sensory impairments following a variety of peripheral and central nervous system

injuries (Christensen et al., 1996; Christensen, Hulsebosch, 1997; Yezierski et al.,

1998). It has been shown that low (i.e. 1.4 gram) vFH strength will induce C and A

delta fiber firing long before actually resulting in a motor response (Leem et al., 1993).

Therefore, the behavioral response is presumably due to a summation of afferent input

into the spinal cord sufficient to evoke a movement response.

98 Primary afferent sprouting in the spinal cord may contribute to altered sensory thresholds and has been reported caudal to the injury following complete spinal cord transection (Krenz, Weaver, 1998), hemisection (Murray, Goldberger, 1974) and dorsal root sectioning lesions (Tessler et al., 1980; Tessler et al., 1981; Tessler et al.,

1984). Wide dynamic range neurons (WDR) in the dorsal horn through electrophysiological examination have shown increases in their peripheral receptive fields after spinal cord hemisection (Brenowitz, Pubols, 1981) and an increase in responsiveness after ischemic injury (Hao et al., 1991). The same stimulus, therefore, may activate a larger number of WDR neurons and/or to a greater extent after injury leading to an increased behavioral response. In addition, GAB A/glycine modulation of primary afferent inputs is altered after peripheral nerve injury and may contribute to a decrease in presynaptic inhibition that is suggested to also occur following the contusion lesion (Lin et al., 1996b; Lin et al., 1996a; Thompson et al., 1992). Excess excitatory amino acid excitotoxic injury (Yezierski, Park, 1993; Hide et al., 1995) and cholecystokinin content (ischemic injury, (Xu et al., 1994) have been correlated with the onset of mechanical allodynia after SCI and may additionally play a role following contusion injury. Lastly, one cannot rule out the possibility of peripheral sensitization of high threshold receptors (Na et al., 1993; Lôpez et al., 1998) or of altered central processing of low threshold A Beta fibers after SCI accoimting for allodynic responses

(Torebjork et al., 1992; Campbell et al., 1989). It is likely, however, that there are multiple factors involved in the onset and maintenance of mechanical allodynia following spinal cord contusion injury. Several papers have now established that

99 contusion injury serves as an excellent clinically relevant model from which to study the mechanism of altered sensory processing and pain after SCI (Siddall et al., 1995;

Lindsey et al., 1999, current study). Further systematic studies utilizing anatomic, physiologic and pharmacologic approaches are necessary to determine precisely the mechanisms involved and will ultimately facilitate the development of clinical intervention strategies.

Exercise may facilitate normal sensory stimulus processing by aiding in the remediation of any of the possible factors listed above. An additional provocative idea concerns the relationship between descending 5-HT pathways and motor activity.

TM activity has been correlated with increased 5 HT content in the extracellular fluid of the spinal cord as assessed by in vivo microdialysis (Gerin et al., 1995). In addition, the activity of raphe nucleii can be correlated with speed of TM training (Veasey et al.,

1995; Jacobs, Fomal, 1997). It is reasonable to suggest that engaging in these exercise tasks (especially TM training) may activate and facilitate spared 5-HT containing descending fibers in injured animals, which then influence dorsal horn neurons modulating responses to sensory testing.

Peripheral vs. Central Effects of Training

Physical therapists for decades have been aware of the importance of exercise activity in improving motor function after CNS injury and disease (Voss et al., 1985).

The prevailing wisdom revolved around the fact that paralytic muscle was presumed to be extremely weak and exercise was intended to strengthen weak muscles in order to impact on function. Research over the past 5-8 years, however, has clearly shown that

100 movement exercise has definite and distinguishable effects on CNS structures (Nudo et al., 1997). For example, use-dependent increases in dendritic branching have been observed in the contralateral homotypic cortex following sensorimotor cortex ablations

(Jones, Schallert, 1994). An increased number of synapses per Purkinje cell neuron with acrobatic activity and increased c^illary densi^ with aerobic activi^ have been reported in the cerebellum (Black et al., 1990; Kleim et al., 1997). Motor activities have also been associated with an increased number of synapses per neuron in the motor cortex (Kleim et al., 1996) and increased motomeuron size in the spinal cord

(Kanda, Hashizume, 1998). Increased trophic factor expression, BDNF,

(hippocampus. Keeper et al., 1995) and bFGF (hippocampus and cerebellum, Gomez-

Pinilla et al., 1998) may influence cell survival in these structures as well as motor performance, whereas increased catecholamine expression in the striatum (Meeusen et al., 1997) and cortex (Pagliari, Peyron, 1995) associated with exercise activities, may influence movement behavior and/or higher cognitive functioning. Lastly, endurance

TM exercise has been shown to increase axonal transport of periperal nerves (Jasmin et al., 1988) and increase CGRP expression in alpha motor neurons (Gharakhanlou et al.,

1999). Further evaluation is necessary to determine if neuroanatomical and/or neurophysiological changes occurred as a result of exercise following SCI which resulted in the normalization of sensory stimulus processing identified in the current study, i.e. changes in transmitter or receptor expression of the dorsal horn, blocking of detrimental sprouting responses, increased beneficial sprouting and/or changes in the neurotransmitter content/release in descending 5-HT fibers. Detrimental sprouting

101 may involve inappropriate contact of central processes from large myelinated fibers relaying touch information to traditional pain pathways, i.e. the anterolateral system.

Lesion Volume

Given that exercise has been shown to facilitate changes within the CNS, one must allow for the possibili^ that alterations may have both positive and negative consequences. Several investigators have described the evolving nature of a contusion injury to the spinal cord (Balentine, Paris, 1978a; Balentine, Paris, 1978b; Bresnahan,

1978). Not only is damage incurred at the particular site of injury, but over the period of several weeks after SCI there is a progressive increase in the lesion size with expansion in radial, as well as, rostro-caudal directions observed (Bresnahan, 1978); reviewed in Schwab, Bartholdi, 1996). Given that progressive lesion development has been attributed to some extent by irreparable ischemic damage to the cord (Balentine,

Paris, 1978a) and that our exercise training (presumably increasing metabolic demand) began early after injury during which time this progression occurs, the possibility exists that exercise training may influence lesion size (more likely exacerbate than attenuate). Kozlowski et al. (1996), have shown that animals engaged in extreme forced used paradigms (i.e. immediate casting of uninvolved forelimb leaving only the contralateral hemiparetic forelimb to negotiate daily activities) beginning the first day following sensorimotor cortex ablations showed an exacerbation of the cortical injury and an impairment of behavioral recovery (decreased hemiparetic FL use). If this extreme type of forced-use paradigm was initiated 7-14 days after the injury, no negative neuroanatomic or behavioral effects were observed (Kozlowski et al., 1996).

102 In addition, pretreatment with the NMDA receptor blocker, MK801, prevented the

exacerbation o f the lesion size brought about by forced use (Jones, Schallert, 1994;

Kozlowski et al., 1996). The fbrced-use effect was described as a situation in which

the cell bodies in the penumbral regions were “sensitive to behavioral pressure”,

meaning that activation of inputs into the sensorimotor cortex utilizing NMDA as their

neurotransmitter (i.e.thalamic inputs) caused the progression of the lesion. Clearly

engaging in motor activity had negative consequences in this case both on the lesion

size and the resultant motor behavior. Therefore, any motor activity/exercise

intervention introduced to the injured CNS needs to be considered for both the positive

and negative consequences to the nervous system and the resultant functional outcome.

Importantly, in the current study group means for lesion area and lesion volume for the

SCI-Ex groups were not statistically different from SCI No-Ex group. Therefore, participation in appropriate tasks at moderate exercise intensity 20-25 min/day early after injury should be considered a safe therapeutic treatment strategy for subjects with moderate SCI.

Caveats

In the current study a small sample size was evaluated using the vFH protocol

(n=20). Further studies utilizing larger numbers of animals including SCI-SW rats are needed. In addition, several attempts were made to quantify and calibrate FW stimulus intensity. For example, a force transducer attached to the experimenters forefinger during pinch testing did not yield reproducible results, and several variations of forceps calibrated to deliver a set force when applied to the paw interfered with the animals

103 movement responses. Pinch testing is useful when analyzing large groups of animals

and can clearly distinguish injured hum non-injured groups of animals in our

experience. However, this test is limited by the subjective nature of the stimulus

intensity necessary to elicit the response. Knowing precisely what force is necessary

to elicit FW would significantly increase the sensitivity and discriminative ability of

this test.

Conclusions

The field of exercise and its application to the recovery fi*om neurological

injury is really in its infancy. It appears that the old adage that doing something is better than nothing may not hold true with respect to exercise after neural injury.

Similar to the Hogdson et al study (1994), rats in the current study who were trained in static standing activities performed more poorly on locomotion on the elevated WW than the SCI No-Ex rats. Task specific training is becoming a theme in rehabilitation circles. What is practiced is as important as the length and intensity of that practice. It appears that engaging in moderate exercise as early as 4 days post injury is safe as long as appropriate tasks are chosen.

Significant improvements in locomotor outcomes are being identified for human SCI studies evaluating TM training (Wemig et al., 1995; Wemig, Muller, 1992;

Harkema et al., 1997). Appropriate protocols are currently in development (personal communication, S. Harkema). A provocative suggestion fi-om the current research is that even for subjects in whom motor recovery is not expected, engaging in

104 motor/exercise training may be beneficial to prevent the onset of chronic pain conditions given that 11-94 % of SCI subjects report the presence of debilitating dysesthetic pain syndromes (Davidoflf et al., 1987).

Lastly, it must be stressed that the results of the current study apply to animals subjected to a moderate 1.1 mm contusion injury to the midthoracic cord. It is possible that rats subjected to a greater or lesser severity of injury would respond differently both with respect to the exacerbation of the neural injury and the resultant behavioral outcome.

In the current study, it is possible that the duration and/or intensity of training was not great enough to invoke a long term behavioral change. Since it is now established that 20-25 minutes of moderately intense activity does not exacerbate the lesion site, more strenuous training protocols (i.e. faster TM speeds or longer training periods) should be evaluated.

105 LIST OF REFERENCES

1. Advokat C, Duke M (1999) Comparison of morphine-induced effects on thermal nociception, mechanoreception, and hind limb flexion in chronic spinal rats. Exp.Clin.Psychophatmacol. 7:219-225.

2. Armstrong DM (1988) The supraspinal control of mammalian locomotion. LPhysiol. 405:1-37.

3. Balentine JD, Paris DU (1978a) Pathology of experimental spinal cord trauma 1. The necrotic lesion as a function of vascular injury. Lablnvest. 39:236- 253.

4. Balentine JD, Paris DU (1978b) Pathology of experimental spinal cord trauma H. Ultrastructure of axons and myelin. Lab lnvest. 39:254-265.

5. Barbeau H, Rossignol S (1987) Recovery of locomotion after chronic spinalization in the adult cat. Brain Res. 412:84-95.

6. Basso DM, Beattie MS, Bresnahan JC (1995) A sensitive and reliable locomotor rating scale for open field testing in rats. J.Neurotrauma 12:1-21.

7. Basso DM, Beattie MS, Bresnahan JC (1996) Graded histological and locomotor outcomes after spinal cord contusion using the NYU weight drop device versus transection. Exp NeuroI. 139:244-256.

8. Basso DM, Beattie MS, Bresnahan JC (2000) Descending systems contributing to locomotor recovery after mild or moderate spinal cord injury in rats. Manuscript in preparation

9. Basso DM, Murray M, Goldberger ME (1994) Differential recovery of bipedal and overground locomotion following complete spinal cord hemisection in cats. Restor NeuroI Neurosci. 7:95-110.

10. Belanger M, Drew T, Provencher J, Rossignol S (1996) A comparison of treadmill locomotion in adult cats before and after spinalization. J.Neurophysiol. 76:471-491. 106 11. Bern T, Gôrska T, Majczynski H, Zmyslowski W (1995) Different patterns of fore-hindlimh coordination during overground locomotion in cats with ventral and lateral spinal lesions. Exp Brain Res. 104:70-80.

12. Berhmann DL, Bresnahan JC, Beattie MS, Shah BR (1992) Spinal cord injury produced by consistent mechanical displacement of the cord in rats: Behavioral and histologic analysis. J.Neurotrauma 9:197-216.

13. Black JE, Issacs KR, Anderson BJ, Alacantara AA, Greenough WT (1990) Learning causes synaptogenesis, whereas motor activity causes angiogenesis, in cerebellar cortex of adult rats. Proc.NatlAcad.Sci.U.S.A. 87:5568-5572.

14. Bracken MB, Shepard MJ, Collins WF, Holford TR, Young W, Baskin DS, Eisenberg HM, Flamm E, Leo-Summers L, Maroon J, Marshall LF, Perot PL, Jr., Peipmeier J, Sonntag VKH, W%ner FC, Wilberger JE, Winn HR (1990) A randomized, controlled trial of methylprednisolone or naloxone in the treatment of acute spinal-cord injury. N.Engl J.Med. 322:1405- 1411.

15. Bregman B, Goldberger ME (1983) Infant lesion effect. II. Sparing and recovery of function after spinal cord damage in newborn and adult cats. Brain Res. 285:119-135.

16. Brenowitz GL, Pubols LM (1981) Increased receptive field size of dorsal horn neurons following chronic spinal cord hemisections in cats. Brain Res. 216:45-59.

17. Bresnahan JC (1978) An electron-microscopic analysis of axonal alterations following blunt contusion of the spinal cord of the rhesus monkey (Macaca mulatta). J.Neurol.Sci. 37:59-82.

18. Bresnahan JC, Beattie MS, Todd FD, Noyes DH (1987) A behavioral and anatomical analysis of spinal cord injury produced by a feedback- controlled impaction device. Exp.Neurol. 95:548-570.

107 19. Brusteîn E, Rossignol S (1998) Recovery of locomotion after ventral and ventrolateral spinal lesions in the cat I. Deficits and adaptive mechanisms. J.Neurophysiol. 80:1245-1267.

20. Bunge RP, Puckett WR, Becerra JL, Marcillo A, Quencer RM (1993) Observations on the pathology of human spinal cord injury. A review and classification for 22 new cases with details from a case of chronic cord compression with extensive focal demyelination. Adv Neurol. 59:75-89.

21. Bussel B, Roby-Brami A, Yakovleff A (1996) Evidence for a spinal stepping generator in man. Electrophysiological study. Acta NeurobiolÆxp. 56:465-468.

22. Calancie B, Needham-Schropshire B, Jacobs P, Wilier K, Zych G, Green BA (1994) Involuntary stepping after chronic spinal cord injury. Evidence for central rhythm generator for locomotion. Brain 117:1143-1159.

23. Campbell JN, Raja SN, Meyer RA, Mackinnon SE (1989) Myelinated afferents signal the hyperalgesia associated with nerve injury. Pain 32:89-94.

24. Chang CW, Lien IN (1991) Estimate of motor conduction in human spinal cord: Slowed conduction in spinal cord injury. Muscle Nerve 14:990-996.

25. Chaplan SR, Bach FW, Pogrel JW, Chung JM, Yaksh TL (1994) Quantitative assessment of tactile allodynia in the rat paw. J.Neurosci Methods 53:55- 63.

26. Cheng H, Almstrom S, Gimenez-Lort L, Chang R, Ove Ogren S, Hoffer B, Olson L (1997) Gait analysis of adult paraplegic rats after spinal cord repair. Exp.Neurol. 148:544-557.

27. Cheng H, Cao Y, Olson L (1996) Spinal cord repair in adult paraplegic rats: Partial restoration of hind limb function. Science 273:510-513.

28. Christensen MD, Everhart AW, Pickelman JT, Hulsebosch CE (1996) Mechanical and thermal allodynia in chronic central pain following spinal cord injury. Pain 68:97-107.

108 29. Christensen MD, Hulsebosch CE (1997) Chronic central pain after spinal cord injury. J.Neurotrauma 14:517-537.

30. Collet P, Uebelhart D, Vico L, Moro L, Hartmann D, Roth M, Alexandre C (1997) Effects of 1- and 6- month spaceflight on bone mass and biochemistry in two humans. Bone 20:547-551.

31. Cope TC, Bodine SC, Fournier M, Edgerton VR (1986) Soleus motor units in chronic spinal transected cats: Physiological and morphological alterations. J.Neurophysiol. 55:1202-1220.

32. Cruse H, Wamecke H (1992) Coordination of the legs of a slow-walking cat. Exp Brain Res. 89:147-156.

33. Czeh G, Gallego R, Kudo N, Kuno M (1978) Evidence for the maintenance of motomeurone properties by muscle activity. J Physiol. 281:239-252.

34. Davidoff G, Roth E, Guarracini M, Sliwa J, Yarkony G (1987) Function-limiting dysesthetic pain syndrome among traumatic spinal cord injury patients: A cross-sectional study. Pain 29:39-48.

35. de Leon R, Hodgson JA, Roy RR, Edgerton VR (1994) Extensor- and flexor-like modulation within motor pools of the rat hindlimb during treadmill locomotion and swimming. Brain Res. 654:241-250.

36. de Leon R, Hodgson JA, Roy RR, Edgerton VR (1998a) Full weight-bearing hindlimb standing following stand training in the adult spinal cat. J.Neurophysiol. 80:83-91.

37. de Leon R, Hodgson JA, Roy RR, Edgerton VR (1998b) Locomotor capacity attributable to step training versus spontaneous recovery after spinalization in adult cats. J.Neurophysiol. 79:1329-1340.

38. de Leon R, Hodgson JA, Roy RR, Edgerton VR (1999) Retention of hindlimb stepping ability in adult spinal cats after the cessation o f step training. J.Neurophysiol. 81:85-94.

109 39. Dietz V, Colombo G, Jensen L, Baumgartner L (1995) Locomotor capacity of spinal cord in paraplegic patients. Ann.Neurol. 37:574-582.

40. Dixon WJ (1948) A method for obtaining and analyzing sensitivity data. J.Am.StatAssoc. 43:109-126.

41. Edgerton VR, de Leon R, Tillakaratne N, Recktenwald MR, Hodgson JA, Roy RR (1997) Use-dependent plasticity in spinal stepping and standing. Adv.Neurol. 72:233-247.

42. Bide PR, Stubhaug A, Stenehjem AE (1995) Central dysesthesia pain after traumatic spinal cord injury is dependent on V-Methyl-D-aspartate receptor activation. Neurosurgery 37:1080-1087.

43. Eidelberg E, Story JL, Walden JG, Meyer BL (1981) Anatomical correlates of return of locomotor function after partial spinal cord lesions in cats. Exp Brain Res. 42:81-88.

44. English AW (1980) Interlimb coordination during stepping in the cat: Effects of dorsal column section. J Neurophysiol. 44:270-279.

45. Fiore CE, Dieli M, Vintaloro G, Gibilaro M, Giacone G, Cottini E (1996) Body composition and bone mineral densi^ in competitive athletes in different sports. Int J.Tissue React. 18:121-124.

46. Forsseberg H, Grillner S, Rossignol S (1977) Phasic gain control of reflexes from the dorsum of the paw during spinal locomotion. Brain Res. 132:121- 139.

47. Fung J, Barbeau H (1994) Effects of conditioning cutaneous stimulation on the soleus H reflex in normal and spastic paretic subjects during walking and standing. J Neurophysiol. 72:2090-2104.

48. Gale K, Kerasidis H, Wrathall JR (1985) Spinal cord contusion in the rat: Behavioral analysis of functional neurologic impairment. Exp NeuroI. 88:123-134.

110 49. Gerin C, Becquet D, Privât A (1995) Direct evidence for the link between monoaminergic descending pathways and motor activity. 1. A study with microdialysis probes implanted in the ventral funiculus of the spinal cord. Brain Res. 704:191-201.

50. Gharakhanlou R, Chadan S, Gardiner P (1999) Increased activity in the form of endurance training increases calcitonin gene-related peptide content in lumbar motoneuron cell bodies and in sciatic nerve in the rat. Neuroscience 89:1229-1239.

51. Goldberger ME, Bregman BS, Vierck CJ, Brown M (1990) Criteria for assessing recovery of function after spinal cord injiuy:Behavioral methods. Exp.Neurol. 107:113-117.

52. Gordon J (1991) Spinal mechanisms of motor coordination. In: Principles of Neural Science. (Kandel ER, Schwartz JS, Jessell TM eds), pp 581-595. New York: Elsevier.

53. Gomez-Pinilla F, So V, Kesslak JP (1998) Spatial learning and physical activity contribute to the induction of fibroblast growth factor: Neural substrates for increased cognition associated with exercise. Neuroscience 85:53-61.

54. Gôrska T, Bem T, Majczynski H, Zmyslowski W (1993) Unrestrained walking in cats with partial spinal lesions. Brain Res.Bull. 32:241-249.

55. Grillner S, Buchanan JT, WaUén P, Brodin L (1998) Neural control of locomotion in lower vertebrates: From behavior to ionic mechanisms. In: Neural Control of Rhythmic Movement in Vertebrates (Cohen AH, Rossignol S, Grillner S eds), pp 1-40. New York: John Wiley and Sons.

56. Grillner S, Shik ML (1973) On the descending control of the lumbosacral spinal cord fi’om the "mesencephalic locomotor region". Acta Physiol.Scand. 87:320-333.

57. Gruner JA, Altman J (1980) Swimming in the rat: Analysis of locomotor performance in comparison to stepping. Exp Brain Res. 40:374-382.

I l l 58. Guiliani CA, Smith JL (1987) Stepping behaviors in chronic spinal cats with one hindlimb deafferented. J.Neurosci. 7:2537-2546.

59. Hao J-X, Xu X-J, Aldskogius H, Seiger A, Wiesenfeld-Hallin Z (1991 ) Allodynia-like effects in rat after ischaemic spinal cord injury photochemically induced by laser irradiation. Pain 45:175-185.

60. Harkema S, Hurley SL, Patel UK, Requejo PS, Dobkin BH, Edgerton VR (1997) Human lumbosacral spinal cord interprets loading during stepping. J.Neurophysiol. 77:797-811.

61. Harris-Warrick RM (1988) Chemical modulation of central pattern generators. In: Neural Control of Rhythmic Movements in Vertebrates (Cohen AH, Rossignol S, Grillner S eds), pp 285-331. New York: John Wiley and Sons.

62. Harris R, Little JW, Goldstein B (1994) Spared descending pathways mediate locomotor recovery after subtotal spinal cord injury. Neurosci Lett. 180:37-40.

63. Helgren ME, Goldberger ME (1993) The recovery of postural reflexes and locomotion following low thoracic hemisection in adult cats involves compensation by undamaged primary afferent pathways. Exp.Neurol. 123:17-34.

64. Hesse S, Bertelt C, Jahnke MT, Schaffrin A, Baake P, Malezic M, Mauritz KH (1995) Treadmill training with partial body weight support compared with physiotherapy in nonambulatory hemiparetic patients. Stroke 26:976-981.

65. Hochman S, McCrea DA (1994a) Effects of chronic spinalization on ankle extensor motoneurons. I. Composite monosynaptic la EPSPs in four motoneuron pools. J.Neurophysiol. 71:1452-1467.

66. Hochman S, McCrea DA (1994b) Effects of chronic spinalization on ankle extensor motoneurons. II. Motoneuron electrical properties. J.Neurophysiol. 71:1468-1479.

112 67. Hodgson JA, Roy RR, de Leon R, Dobkin B, Edgerton VR (1994) Can m am m alian spinal cord leam a motor task? Med.Sci.Sports Exercise 26:1491-1497.

68. Jacobs BL, Fomal CA (1997) Serotonin and motor activity. Curr.Opin.Neurobiol. 7:820-825.

69. Jasmin BJ, Lavoie P-A, Gardiner PF (1988) Fast axonal transport of labelled proteins in motoneurons of exercise-trained rats. Am J Physiol. 255:0731-C736

70. Jiang W, Drew T (1996) Effects of bilateral lesions of the dorsolateral funiculi and dorsal columns at the level o f the low thoracic spinal cord on the control of locomotion in the adult cat. I. Treadmill walking. J.Neurophysiol. 76:849-866.

71. Jones TA, Schallert T (1994) Use-dependent growth of pyramidal neurons after neocortical damage. J.Neurosci. 14:2140-2152.

72. Kanda K, Hashizume K (1998) Effects of long-term physical exercise on age- related changes of spinal motoneurons and peripheral nerves in rats. Neurosci Res. 31:69-75.

73. Kerasidis H, Wrathall JR, Gale K (1987) Behavioral assessment of functional deficit in rats with contusive spinal cord injury. J.Neurosci.Methods 20:167-189.

74. Kleim JA, Kapil VIJ, Ballard DH, Greenough WT (1997) Learning-dependent synaptic modifications in the cerebellar cortex of adult rat persist for at least 4 weeks. J.Neurosci. 17:717-721.

75. Kleim JA, Lussnig E, Schwarz ER, Comery TA, Greenough WT (1996) Synaptogenesis and FOS expression in the motor cortex of the adult rat afier learning motor skill. J.Neurosci. 16:4529-4535.

76. Kozlowski DA, James DC, Schallert T (1996) Use-dependent exaggeration of neuronal injury after unilateral sensorimotor cortex lesions. J.Neurosci. 16:4776-4786. 113 77. Krenz NR, Weaver LC (1998) Sprouting of primary afTerent fibers after spinal cord transection in the rat Neuroscience 85:443-458.

78. Kunkel-Bagden E, Dai HN, Bregman BS (1992) Recovery of ftmction after spinal cord hemisection in newborn and adult rats: Differential effects on reflex and locomotor ftmction. Exp NeuroI. 116:40-51.

79. Leem JW, Willis WD, Chung JM (1993) Cutaneous sensory receptors in the rat foot. J Neurophysiol. 69:1684-1699.

80. Lin Q, Peng YB, Willis WD (1996a) Inhibition of primate spinothalamic tract neurons by spinal glycine and GAB A is reduced during central sensitization. J Physiol. 76:1005-1014.

81. Lin Q, Peng YB, Willis WD (1996b) Role of GABA receptor subtypes in inhibition of primate spinothalamic tract neurons: Difference between spinal and periaqueductal gray matter. J.Neurophysiol. 75:109-123.

82. Lindsey AE, LoVerso RL, Tovar CA, Beattie MS, Bresnahan JC (1999) Rats with contusion spinal cord injury are hypersensitive to tactile and cold stimuli. Manuscript in preparation.

83. Little JW, Harris R, Sohlberg RC (1988) Locomotor recovery following subtotal spinal cord lesions in a rat model. Neurosci Lett. 87:189-194.

84. Lovely RG, Gregor RJ, Roy RR, Edgerton VR (1986) Effects of training on the recovery of fuU-weight bearing stepping in the adult spinal cat. Exp NeuroI. 92:421-435.

85. Lovely RG, Gregor RJ, Roy RR, Edgerton VR (1990) Weight-bearing hindlim b stepping in treadmill exercised adult spinal cats. Brain Res. 514:206-218.

86. Lopez SM, Pérez-Pérez M, Marquez JM, Naves FJ, Represa J, Vega JA (1998) p75 and TrkA Neurotrophin Receptors in human skin after spinal cord and peripheral nerve injury, with special references to sensory corpuscles. AnatRec. 251:371-383.

114 87. Meeusen R, Smolder J, Sarre S, DeMeiriier K, Keizer H, Semeeis M, Ebinger G, Michotte Y (1997) Endurance training effects on neurotransmitter release in rat striatum; An in vivo microdialysis study. Acta Physiol.Scand. 159:335-341.

88. Michel RP, Cruz-Orive LM (1988) Application of the Cavalieri principle and vertical sections method to lung: Estimation of volume and pleural surface area. J Microsc. 150:117-136.

89. Muir GD, Whishaw IQ (1999) Complete locomotor recovery following corticospinal tract lesions: Measurement of ground reaction forces during overground locomotion in rats. Behav.Brain Res. 103:45-53.

90. Murray M, Goldberger ME (1974) Restitution of function and collateral sprouting in the cat spinal cord: The partially hemisected animal. J.Comp.Neurol. 158:19-36.

91. Na HS, Leem JW, Chung JM (1993) Abnormalities of mechanoreceptors in a rat model of neuropathic pain: Possible involvement in mediating mechanical allodynia. J.Neurophysiol. 70:522-528.

92. Naito A, Shimizu Y, Handa Y (1990) Analyses of treadmill locomotion in adult spinal dogs. Neurosci.Res. 8:281-290.

93. Neeper SA, Gômez-Pinilla F, Choi J, Cotman C (1995) Exercise and brain neurotrophins. Nature 373:109-109.

94. Noble L J, Wrathall JR (1989) Correlative analyses of lesion development and Amctional status after graded spinal cord contusion injuries in the rat. Exp.Neurol. 103:34-40.

95. Nudo RJ, Milliken GW, Jenkins WM, Merzenich MM (1996) Use-dependent alterations of movement representations in primary motor cortex of adult squirrel monkeys. J.Neurosci. 16:785-807.

96. Nudo RJ, Plautz EJ, Milliken GW (1997) Adaptive plasticity in primate motor cortex as a consequence of behavioral experience and neuronal injury. Semin.Neurosci. 9:13-23. 115 97. Nudo RJ, W se BM, SiFuentes F, Milliken GW (1996) Neural substrates for the effects of rehabilitative training on motor recovery after ischemic infarct. Science 272:1791-1794.

98. Orlovsky ON (1972) Activity of vestibulospinal neurons during locomotion. Brain Res. 46:85-98.

99. P%liari R, Peyron L (1995) Norepinephrine release in the rat frontal cortex under treadmill exercise: A study with microdialysis. J Appl Physiol. 78:2121- 2130.

100. Pearson KG, Rossignol S (1991) Fictive motor patterns in chronic spinal cats. J Neurophysiol. 66:1874-1886.

101. Robichaud JA, Agostinucci J (1996) Airsplint pressure effects on soleus muscle alpha motomeuron reflex excitability in subjects with spinal cord injury. Arch Phys Med RehabU. 77:778-782.

102. Robinson GA, Goldberger ME (1985) Interfering with inhibition may improve motor function. Brain Res. 346:400-403.

103. Rossignol S, Chau C, Brustein E, Belanger M, Barbeau H, Drew T (1996) Locomotor capacities after complete and partial lesions of the spinal cord. Acta Neurobiol.Exp. 56:449-463.

104. Rossignol S, Lund JP, Drew T (1988) The role of sensory inputs in regulating patterns of rhythmic movement in higher vertebrates: A comparison between locomotion, respiration, and mastication. In: Neural Control of Rhythmic Movements in Vertebrates (Cohen AH, Rossignol S, Grillner S eds), pp 201-283. New York: John WUey and Sons.

105. Roy RR, Baldwin KM, Edgerton VR (1991) Plasticity of skeletal muscle: Effects of neuromuscular activity. Exerc.Sport SciRev. 19:269-312.

106. Roy RR, Hirota WK, Kuehl M, Edgerton VR (1985) Recruitment patterns in the rat hindlimb muscle during swimming. Brain Res. 337:175-178.

116 107. Roy RR, Hutchinson DL, Pierotti DJ, Hodgson JA, Edgerton VR (1991) EMG patterns of rat ankle extensors and flexors during treadmill locomotion and swimming. J.Appl.PhysioL 70:2522-2529.

108. Roy RR, Talmadge RJ, Hodgson JA, Zhong H, Baldwin KM, Edgerton VR (1998) Training effects on soleus of cats spinal cord transected (T12-13) as adults. Muscle Nerve 21:63-71.

109. Schwab ME, Bartholdi D (1996) Degeneration and regeneration of axons in the leisoned spinal cord. Physiol Rev. 76:319-370.

110. Siddall P, Xu CL, Cousins M (1995) Allodynia following traumatic spinal cord injury in the rat. NeuroReport 6:1241-1244.

111. Sinnamon H (1993) Preoptic and hypothalamic neurons and the initiation of locomotion in the anesthetized rat. Prog Neurobiol. 41:323-344.

112. Skiimer RS, Houle JD, Reese NB, Berry CL, Garcia-Rill E (1996) Effects of exercise and fetal spinal cord implants on the H-reflex in chronically spinalized adult rats. Brain Res. 729:127-131.

113. Smith JL, Smith LA, Zemicke RF, Hoy M (1982) Locomotion in exercised and nonexercised cats cordotomized at two or twelve weeks of age. Exp NeuroI. 76:393-413.

114. SPA I (1997) The stereologer handbook. Introduction to unbiased stereology and stereologer users guide. Alexandria, VA: Systems Planning and Analysis, Inc.

115. Stokes BT, Homer PJ (1996) Spinal cord injury modeling and outcome assesment. In: Neurotrauma (Narayan RK, Wilberger JE, Poviishock J eds), pp 1395-1403. New York: McGraw-Hill Co. Inc.

116. Stokes BT, Noyes DH, Behrmann DL (1992) An electromechanical spinal injury technique with dynamic sensitivity. J Neurotrauma 9:187-195.

117 117. Taub E, Miller NE, Novack TA, Cook EW, HI, Fleming WC, Nepomuceno CS, Connell JS, Crago JE (1993) Technique to improve chronic motor deficit after stroke. Arch.Phys.Med.Rehabil. 74:347-354.

118. ten Cate J (1960) Locomotor movements in the spinal pigeon. Exp.Biol. 37:609- 613.

119. ten Cate J (1962) Innervation of locomotor movements by the lumbrosacral cord in birds and mammals. J.Exp.Biol. 39:239-242.

120. ten Cate J (1964) Locomotor movements of the hind limbs in rabbits after isolation of the lumbosacral cord. J Exp Biol. 41:359-362.

121. Tessler A, Glazer E, Artymyshyn R, Murray M, Goldberger ME (1980) Recovery of Substance P in the cat spinal cord after unilateral lumbosacral deafiferentation. Brain Res. 191:459-470.

122. Tessler A, Himes BT, Artymyshyn R, Murray M, Goldberger ME (1981) Spinal neurons mediate return of Substance P following deafferentation of cat spinal cord. Brain Res. 230:263-281.

123. Tessler A, Himes BT, Soper K, Murray M, Goldberger ME, Reichlin S (1984) Recovery of Substance P but not Somatostatin in cat spinal cord after unilateral lumbosacral dorsal rhizotomy: A quantitative study. Brain Res. 305:95-102.

124. Thompson FJ, Parmer R, Reier PJ (1998) Alteration in rate modulation of reflexes to lumbar motoneurons after midthoracic spinal cord injury in the rat. I. Contusion injury. J.Neurotrauma 15:495-508.

125. Thompson FJ, Reier PJ, Lucas CC, Parmer R (1992) Altered patterns of reflex excitability subsequent to contusion injury o f the rat spinal cord. J Neurophysiol. 68:1473-1486.

126. Torebjork HE, Lundberg LER, LaMotte RH (1992) Central changes in processing of mechanoreceptive input in capsaicin-induced secondary hyperalgesia in humans. J Physiol. 448:765-780.

118 127. Trimble M, Kukulka CG, Behiman AL (1998) The effect of treadmill gait training on low Aequency depression of the soleus H reflex; Comparison of a spinal cord injured man to normal subjects. NeurosciXett. 246:186- 188.

128. Veasey SC, Fomal CA, Metzlar CW, Jacobs BL (1995) Response of serotonergic caudal raphe neurons in relation to specific motor activities in fieely moving cats. J.Neurosci. 15:5346-5359.

129. Voss DE, lonta MK, Myers BJ (1985) Proprioceptive neuromuscular facilitation patterns and techniques. Philadelphia: Harper and Row.

130. Wemig A, Muller S (1992) Laufband locomotion with body weight support improved walking in persons with severe spinal cord injuries. Paraplegia 30:229-238.

131. Wemig A, Muller S, Nanassy A, Cagol E (1995) Laufband therapy based on rules of spinal locomotion' is effective in spinal cord injured persons. Eur.J.Neurosci. 7:823-829.

132. Wickelgren I (1998) Teaching the spinal cord to walk. Science 279:319-321.

133. Wolf SL, Lecraw DE, Barton LA, Jaim BB (1989) Forced use of hemiplegic upper extremities to reverse the effect of teamed nonuse among chronic stroke and head-injured patients. Exp NeuroI. 104:125-132.

134. Wrathall JR, Li W, Hudson LD (1998) Myelin gene expression after experimental contusive spinal cord injury. J.Neurosci. 18:8780-8793.

135. Xu X-J, Hao J-X, Sieger A, Hughes J, Hokfelt T, Wiesenfeld-Hallin Z (1994) Chronic pain-related behaviors in spinally injured rats: Evidence for functional alterations of the endogenous cholecystokinin and opioid systems. Pain 56:271-277.

136. Yezierski RP (1996) Pain following spinal cord injury: The clinical problem and experimental studies. Pain 68:185-194.

119 137. Yezierski RP, Liu S, Ruenes GL, Kajander KJ, Brewer KL (1998) Excitotoxic spinal cord injury; Behavioral and morphological characteristics of a central pain model. Pain 75:141-155.

138. Yezierski RP, Park SH (1993) The mechanosensitivity of spinal sensory neurons following intraspinal injections of quisqualic acid in the rat. Neurosci.Lett. 157:115-119.

120 CHAPTERS

A COMPARISON OF EXERCISE INTERVENTIONS ON RECOVERY FROM SPINAL CORD CONTUSION INJURY IN RATS; n. SKELETAL MUSCLE EFFECTS.

Introduction

Behavioral motor function is related to the recruitment and the contractile properties of skeletal muscle. Studies have shown that the particular type of myosin heavy chain (MHC) protein expressed in skeletal muscle confers the properties of contractile speed and twitch force, i.e. rate of force production (Reiser et al., 1985a;

Reiser et al., 1985b; Sieck, 1994; SchiafBno et al., 1988). Four MHC proteins associated with rodent skeletal muscle fiber have been described. MHC type I is associated with slow speeds, produces small twitch forces and has low fatigability.

MHC types Ha, IIx and lib represent progressively faster speeds, larger twitch forces, and higher fatigability (Sieck et al., 1998; Goldspink, 1998).

Deficits in motor unit recruitment, secondary to injury or disease, result in impaired voluntary movement. For example, trauma to the spinal cord often results in paralysis or paresis caudal to the neural injury. Muscle atrophy and conversion of muscle fibers toward a faster phenotype following spinal cord injury (SCI) have been

121 described (Roy et al., 1991; Talmadge et al., 1995b; Lotta et al., 1991; Grimby et al.,

1976; Stilwill, Sahgal, 1977; Castro et al., 1999). In experimental models, these changes have been shown to occur to a greater extent in postural extensors versus flexor muscle groups and in slow twitch more than fast twitch fibers (Talmadge et al.,

1995b; Roy et al., 1991; Roy et al., 1999; Roy et al., 1998). The clinical relevance of the expected conversion o f slow twitch to fast twitch muscle fiber characteristics

(particularly for slow twitch muscle like the soleus, SOL), is a decrease in muscle contraction time and fatigue characteristics (Lieber, 1992). This renders the subject less able to generate low level, prolonged contractions (Lieber, 1992) and therefore less able to participate in tasks that require them, i.e. maintaining antigravity postures.

One animal model of SCI, the transection (TX) model, has been carefully described in terms of the rather significant and lasting muscle atrophy and fiber type conversion that results (Talmadge et al., 1995a; Roy et al., 1991; Roy et al., 1999; Roy et al., 1998; Dupont-Versteegden et al., 1998). This model completely eliminates descending and ascending connections between the caudal cord and locations rostral to the lesion site. In adult animals, TX of the low thoracic cord results in complete hindlimb (HL) paralysis with the dependent limbs usually extended and dragging during any attempt at open field locomotion (Roy et al., 1991; Basso et al., 1996). In comparison, spinal cord contusion injury represents a more clinically relevant model of SCI. Sparing of some ascending and descending fibers across the lesion site generally occurs along with varying amounts of volitional control of HL movements

(Stokes et al., 1995; Berhmann et al., 1992; Basso et al., 1996). The impaired motor

122 function induced by a moderate contusion injury results in changes in skeletal muscle properties (see Chapter I) and it is likely that these altered muscle properties further contribute to impaired locomotor function.

The recovery of voluntary movement appears to induce changes in skeletal muscle properties compared to animals that do not regain volitional control (see

Chapter 1 ; Lieber et al., 1986a). Therefore, effî>rts to improve motor recovery via training would be expected to further induce changes at the level of skeletal muscle.

In Chapter 2, we compared the therapeutic efficacy of three different exercise training regimens, initiated early after a contusion injury, on several behaviors including locomotion. The three tasks were standing (SCI-ST), swimming (SCI-SW) and treadmill walking (SCI-TM) which manipulate load (weight bearing) and rhythm generation in the HLs. Performance was compared to injured animals that did not engage in exercise (SCI No-Ex) and laminectomy control (LAM CTL) rats. Our hypothesis was that tasks which emphasize both load and rhythmic HL movements would facilitate greater recovery of locomotor behavior following SCI than tasks that do not.

In the current paper, we evaluate at 7 weeks post spinal cord contusion injury the extent of atrophy and fiber type conversion induced in HL skeletal muscle, as well as, the differential impact that exercise training has on these changes. Our hypotheses were, (a) animals engaged in exercise training would show improved locomotor recovery compared to animals that did not and, of the different exercise groups

123 treadmill trained animals would perform the best; (b) weight bearing exercise tasks

(stand and treadmill training) would target SOL muscle activation to prevent atrophy

(decreased CSA) and fiber type conversion, compared to non-weightbearing tasks

(swimming and non-exercising SCI animals); and, (c) due to the higher force requirements of HL muscles engaged in swimming training (Roy et al., 1985; Roy et al., 1991; de Leon et al., 1994), we predicted that animals engaged in this task would attenuate atrophy measures of larger, predominantly fast twitch muscle groups (medial gastrocnemius, MG; lateral gastrocnemius, LG) compared to animals not engaged in swim training.

Wet weight analyses of skeletal muscle is commonly used as an indicator of muscle atrophy implying that a decrease in fiber area will translate into a lighter weighing muscle. In addition, it has been shown that a decrease in fiber area is associated with a decrease in force production capabilities (Edgerton, Roy, 1996);

(Roy et al., 1991; Roy et al., 1998). In Chapter 1, we describe changes that occur over time in HL skeletal muscle as a result of moderate spinal cord contusion injury (1.1 mm, OSU device). Force production and wet weight values recorded at 1,3 and 10 weeks after injury suggest that atrophy of the SOL muscle occurs early following SCI

(wet weight deficits greatest at 1 week post injury and force production deficits greatest at 3 weeks post-op) and appear to be resolved by 10 weeks after injury. One of the goals of the current study, therefore, was to investigate the extent of skeletal muscle changes induced by contusion injury at an intermediate time point, i.e. between the previously recorded observations at 3 and 10 weeks post SCI.

124 As identified in our earlier study, SOL wet weight at 3 weeks afier injury, was not significantly different between SCI and control animals, yet the SCI muscle produced force values (peak twitch (Pt) and peak tetanic tension (Po)) at approximately 50% of the CTLs. This suggested that wet weight measures in some cases may not be sensitive to the changes in muscle fiber cross-sectional area that actually exist. A significant decrease in wet weight is generally a good indicator of muscle fiber atrophy/ decreased cross-sectional area (CSA) (Roy et al., 1998;

Grossman et al., 1998) whereas the attenuation of wet weight loss is more difBcult to interpret (Kandarian et al., 1991; Chapter 1). Therefore, a second goal of the current study was to use cross-sectional area measurements of individual fibers as a more precise anatomical assessment of muscle atrophy.

The SOL muscle was chosen for analyses for several reasons. First, the SOL is a slow twitch postural extensor and is vulnerable to SCI. Second, we can compare our results with physiological data collected previously in the contusion model. Lastly, since we did not expect to find differences in wet weights across groups for this muscle at the 7 week time point, any changes in fiber CSA would allow the two measures to be more easily compared.

Detailed analyses of MHC expression and immunohistochemical identification of fiber type were completed for the SOL muscle. In addition, as a screening tool for muscle atrophy, wet weight measures were determined for several muscle groups

(extensor digitorum longus, EDL, plantaris, PL, tibialis anterior, TA, MG, and LG) of the HL important for the execution of functional tasks.

125 Methods

Subjects

The same subjects from Chapter 2 were utilized in the current study. Briefly, female Sprague Dawley rats (n=47) weighing 250-300 grams at the start of the study were randomly assigned to one of five treatment groups; SCI-TM (n=7), SCI-SW

(n=6), SCI-ST (n=9), SCI No-Ex (n==6), and LAM CTLs (n=7). Seven rats were removed from the study secondary to complications associated with the surgery. Data from one rat was not included in analysis due to a lack of participation in the task (a swimmer who never swam).

In the first half of the experiment, rats were randomly housed two to three per cage. For the second phase of the study, rats were multiply housed based on their treatment group. Rats in the SCI No-Ex group showed greater motor skills in mixed housing versus group housing (see Chapter 2 Methods), therefore were treated separately and were not included in the SCI No-Ex statistical analyses. An additional group of rats (n=6) were group housed 2-3 per cage and were used for the SCI No-Ex.

No difference in motor performance for the trained groups occurred relative to housing situation, therefore all data was pooled for each o f the training groups.

Pre-Training

As described in Chapter 2, rats were acclimated to all tasks using the least amount of time on the task as possible to ensure accurate, compliant performance while preventing the acquisition of sophisticated motor skills.

126 Daily Training

As described in Chapter 2, exercise training for TM, SW and ST was initiated

4 dpo lasting 20-25 min/day, 5 days/week, for 7 weeks.

SCI Surgical Procedures

As described in Chapter 2, we performed a moderate contusion injury (1.1 mm

displacem ent) at Tg.

Muscle Dissection Procedures

Several muscles associated with locomotion were dissected from the right

hindlimb (HL) of each rat at the time of sacrifice. Under deep anesthesia (ketamine

80mg/kg, xylazine lOmg/kg, i.p.), the following muscles were removed; MG, LG, PL,

SOL, TA, and EDL. After making a skin incision, the tendinous insertions of

superficial muscle groups were exposed using blunt dissection in order to preserve the blood and nerve supply to deeper muscle groups. Muscles were removed by separating the insertions from the bone, reflecting the muscle back and detaching them from their origin. Dissection proceeded from superficial to deep muscles. In order to separate the MG and LG, muscle attachments were cut along their common aponeurosis.

Muscle Wet Weight

Each skeletal muscle was cleaned of fat and excess connective tissue before being wet weighed. The SOL was then pinned to cork at its in situ length, immediately frozen in isopentane cooled by liquid nitrogen and stored at -80°C until further

127 processing. The heart was carefully dissected, rinsed in saline and blotted dry before

the wet weight was determined.

Mvosin Heavy Chain Expression

A 3 mm frozen section from the midbelly of the SOL was taken after sacrifice

at 7 weeks post injury. The muscle tissue was minced with scissors in 9 vol of cold

homogenization buffer [250mM sucrose, 100 mM KCl, 5 mM EDTA, and 20 mM tris

(hydroxymethl) aminomethane pH 6.8] before homogenizing it with an electric tissue

grinder. Total protein content of the homogenate was determined using a BCA kit

(Pierce, Rockford, Illinois) according to manufacturers instructions and an EL-800

microplate reader (Bio-Tek Instruments, Inc. Winooski, Vt.). Samples were then

diluted in Laemmli buffer (Laemmli, 1970) to a final total protein concentration of .25

mg/ml. For identification of MHC protein, SDS-PAGE was used according to the

methods of Talmadge et al (1993) and 22 ul of sample were loaded per lane.

Coomassie stained gels were scanned using a 1 D scanning densitometer (Alpha

Imager 2000, San Leandro, CA). Each band was represented as the percentage of the

total MHC sampled. Using this procedure, any MHC isoform accounting for less than

4 % of total protein expression did not generate distinguishable boundaries for the ID

scanning densitometer. This occurred for Ila and IIx MHC. Therefore a category combining Ila, x was created to account for this expression.

Immimohistochemistrv

Serial sections firom the SOL midbelly were cut 10 um thick on a cryostat at

-22°C and stained with a series of monoclonal antibodies (mAb) specific to rat MHC

128 isoforms (Table 3.1). All antibodies were purchased at Developmental Studies

Hybridoma Bank (University of Iowa, Iowa City, lA). Antibodies A4.951, N2.26I, and A4.74 are IgG class antibodies which stain type I, I/Ha, and Ila/IIx MHCs, respectively (Table 3.1) (Webster et al., 1988; Hughes, Blau, 1992; Hughes et al.,

1993; Cho et al., 1993). N3.36 antibody stains all fast fibers and is an IgM class primary antibody (Table 3.1) (Hughes, Blau, 1992). Laminin (DI8) (Sanes et al.,

1990; Green et al., 1992), also an IgG class antibody, was used for the determination of fiber cross-sectional area (see below). Phosphate-buffered saline was used as a buffer for the immunoglobulin IgG class primary antibodies and tris (hydroxymethyl) aminomethane-buffered saline as a buffer for the IgM class primary antibody

(Talmadge et al., 1995b). Localization of primary antibody binding was determined using the avidin-biotin immunohistochemical procedure (Kits # PK-6102 and AK-

5010 Vector labs, Burlingame, CA) with diaminobenzidene (DAB) as the chromagen for IgG class primary antibodies and alkaline phosphotase for IgM.

Sections were maintained at room temperature in dessicant x 20 min., followed by a 20 min. re-hydration step with the appropriate buffer. After 45 min. in blocking solution at room temperature, the primary antibody was added. A no-primary-control slide was concurrently processed for each animal for comparison of background staining. Primary antibody incubation lasted overnight at 4°C. A washing step included 3x10 min with appropriate buffer in coplin jars. A one hour incubation at room temperature with the secondary antibody was again followed by a wash. Next, a

129 1 hour incubation with ABC reagent also at room temperature was followed by a wash

and development with DAB which lasted 3-5 minutes or alkaline phosphotase which

lasted 30 minutes.

One hundred fibers selected from a representative central section of each SOL

muscle (which could be followed across all stained serial sections) were analyzed for

MHC expression. In addition, SOL muscle fiber cross-sectional area was measured on

laminin stained sections using a Zeiss Axiophot Microscope attached to MCID-M4

Image Analysis System (Imaging Research Inc., Ontario, Canada). Images were

captured using a Dage CCD72 8 bit camera at 10 x and projected in black and white

onto a 38 cm monitor. We used laminin staining to demarcate the border of each fiber

by adjusting the optical density of the computer-microscope capture system which

then calculated the cross-sectional fiber area (um^).

MHC ISOFORM A ntibody I Ha IIx Hb

A4.951 + ———

N2.261 + ——

A4.74 — + —

N3.36 + + +

Table 3.1 : Identification of muscle fiber type using monoclonal antibodies to I, Ha, lib and IIx myosin heavy chain (MHC) isoforms. Antibody A4.951 labels type I MHC. Antibody N2.261 laW s types I and Ha MHC. Antibody A4.74 labels type Ha and Ox MHC. yùitibody N3.36 labels type Ha, Hb, and IIx MHC.

130 Statistical Analysis

Locomotor rating using the BBB scale was analyzed using repeated measures

ANOVA (see Chapter 2 methods). Muscle wet weight values, SOL fiber CSA, immunohistochemically determined percent fiber type, and MHC expression via gel electrophoresis were analyzed using a one way ANOVA. Scheffe^s post-hoc test was employed where appropriate. Data is presented as means ± sd unless otherwise noted.

A significance level of p<.OS was determined a priori.

Results

Open Field Locomotion

In Chapter 2 we describe the locomotor recovery of animals subjected to a 1.1 mm spinal cord contusion injury and subsequent participation in an exercise program beginning 4 dpo (Figure 2.1). Briefly, significant interactions that occurred for comparison of groups over time in the BBB test came only during the first week after injury. At 5 dpo, SCI-TM rats scored significantly higher than SCI No-Ex rats and at

7 dpo SCI-TM and SCI-S W rats scored significantly higher than both SCI No-Ex and

SCI-ST animals. No significant differences were observed across all injured groups by 14 dpo and throughout the rest of the study. Chronic locomotor patterns were

131 typically achieved by 21 dpo with mean BBB scores for SCI-TM =10.3 ± 0.4 sem,

SCI-SW =10.5 ± 0.3, SCI-ST =10.5 ± 0.2 and SCI No-Ex =10.8 ± 0.5 as compared to

LA M C TL = 21±0.

Muscle Wet Weight Values

At 7 weeks post injury, the wet weight values for the EDL, PL, SOL and TA of all injured animals were not statistically different from LAM CTLs (Table 3.2).

However, all SCI groups had significantly lighter LG wet weights than the LAM

CTLs. For example, the LG wet weight was 86% (SCI No-Ex and SCI-TM), 87%

(SCI-ST) and 90% (SCI-SW) of LAM CTL values (Table 3.2). In addition, the MG wet weight values for SCI-TM and SCI-ST were significantly lighter than LAM CTLs

(p<.05), but were not different from SCI-SW and SCI No-Ex.

Lastly, cardiac wet weight measures were determined at sacrifice and showed a trend that SCI-TM rats had heavier wet weights compared to SCI-ST animals (3.8 ±.14 sd and 3.40 ± 0.1 g, respectively; p=.05). In addition, the mean cardiac wet weight values for SCI No-Ex (3.46 ± 0.25g) and LAM CTLs (3.46± 0.21 g) were also lower than SCI-TM, although the differences were not significant. Therefore, it appeared that

TM training induced cardiovascular adaptation during recovery from SCI.

132 EDLPLSOL MGLG TAHEART LAM .892 2.57 .934 2.64 3.36 4.36 3.46 CTL (.05) (-13) (-09) (-21) (.14) (-23) (.21)

SCI .910 2 3 7 .946 2.41 2.96* 4.14 3.46 NO-EX (.05) (.17) (-07) (.13) (.15) (.16) (-25)

SCI- .921 2.43 .964 2 3 * 2.89* 4.11 3.80 TM (.05) (-12) (-07) (-11) (-14) (-17) (-14)

SCI- .866 2.45 .945 2.54 3.05* 4.32 3.60 SW (.02) (.16) (-07) (.20) (.19) (-21) (-33)

SCI- .903 2.39 .993 2.33* 2.92* 4.14 3.40+ ST (-06) (.13) (-08) (-19) (-17) (-18) (-10)

Table 3.2; Mean (± standard deviation) muscle wet weight values (g) for the extensor digitorum longus (EDL), plantaris (PL), soleus (SOL), medial gastrocnemius (MG), lateral gastrocnemius (LG), tibialis anterior (TA) and heart for each group. Note that exercise training did not resolve MG and LG atrophy (* p<.05 vs LAM CTL). There was a trend that standing training may have reduced the size of the heart as compared to treadmill trained animals (+ p=.05). LAM CTL: laminectomy control; SCI No-EX: untrained; SCI-TM: treadmill trained; SCI-SW: swim trained; SCI-ST: stand trained.

133 Cross-Sectional Area

Analysis of SOL muscle revealed that there was no significant difference in

CSA for any fiber type (I, Ila, IIx, or Co-expressing IJIaJIx) ftom any group evaluated in this study (Table 3.3). For each fiber type, the group means for SCI No-

Ex rats were always lower than all other groups. For example, mean Type I fiber CSA for SCI No-Ex rats was 12% below SCI-TM fiber area, 21% below SCI-SW, 3% below SCI-ST and 13% below LAM CTLs. Due to variability, however, in all instances the differences were not statistically significant.

LAM CTL SCI NO-EX SCI-TM SCI-SW SCI-ST (n=7) (n=6) (n=7) (n=6) (n=9) TYPE I 2346 (416) 2047 (325) 2301(520) 2591 (661) 2119(589) [660] [521] [596] [528] [782] TYPE HA 1478(351) 919(253) 1182(379) 1419(299) 1106(372) [40] [47] [79] [60] [83] TYPE n x NS NS 1012(283) NS 1729(0) [6] [8]

Co -E x press NS 900 (257) 1206(383) 1251 (373) 1061(359) I, HA, IDC [35] [19] [12] [27] GRAND 2291 (400) 1925(317) 2135(481) 2451 (445) 1997(577) MEAN [700] [600] [700] [600] [900]

Table 3.3: Mean (± standard deviation) cross sectional area (mm^) type I, Ha, Ilb and IIx soleus muscle fiber for each group. The number of animals analyzed fi^om each group are listed in parentheses under the group name. Approximately 100 fibers were analyzed for each animal. Of these fibers, the number measured for each type collapsed across animals is shown in brackets. NS: none occurred in the sample.

134 Immunohistnchemîstrv

The percent fiber type expressed (I, Ha, IIx, or co-expression of I, Ua, IIx) was measured fi’om the 100 sampled fibers per muscle and with one exception, an increased number of co-expressing fibers, no significant differences were observed across groups (Figure 3.1). Most fibers were stained with the monoclonal antibody that reacts with slow MHC (Figure 3.2). Positive staining of N2.261 along with a negative stain of A4.74 suggests slow type I fiber expression and occurred in approximately 88% of all the fibers tested. This finding was confirmed with a positive

A4.951 stain (Figure 3.2). However, a positive stain of bothN2.261 and A4.74, but negative A4.951 suggests Ila fiber expression and occurred in 6- 11% of all cases. It is possible that both Ila and IIx expression occurred in these instances but we represent them as Ha for the following reasons: (a) gel electrophoresis identified only minute quantities of IIx protein in the SOL muscle (Figure 3.3); (b) A4.74 appears to weakly label or fails to label IIx fibers (Hughes et al., 1993); and, (c) there is no commercially available monoclonal antibody to definitively identify a co-expressing MHC IIx fiber.

On the other hand, pure IIx fibers could be identified firom the negative stain of

N2.26I and positive staining of A4.74 and N3.36. The IIx fiber expression only occurred in < 0.5% of all muscle fibers analyzed (Figure 3.4). The number of Hb fibers in each sample was determined by comparing N3.36, which labels all fast fibers

(Ha, IIx, nb), to A4.74 which binds only Ha and IIx fibers (Hughes et al., 1993).

135 Figure 3.1. Mean percentage ( ± standard deviation) of muscle fiber types in the soleus for each group as identified firom immunohistochemical staining for myosin heavy chain (MHC) expression (type I, Ila, IIx, and Hb), Note the trend toward co- expression of I, Ha and IIx after SCI compared to normals (*; p=.07 vs. LAM CTL). LAM CTL: laminectomy control; SCI No-EX: untrained; SCI-TM: treadmill trained; SCI-SW: swim trained; SCI-ST: stand trained.

136 100 LAM CTL 80

60

40

20

0 100 SCI NO-EX 80

80 ■o 40 + C 20 (8 O 0 E 100 SCI-TM tr( 0 80 111 80 ffi LL 40 U. O 20 1 - 0 z 1 0 0 . UJ SCI-SW o 80 K LU 80 . Q. 40

20

100 SCI-ST 80

80

40

20

Ila Co-ex IIx "ilb" FIBER TY P E Figure 3.1

137 FIGURE 3.2. Myosin heavy chain antibody staining for A. type I isoforms (A 4.951); B. types I and Ila isoforms (N2.261); C. types Ila and IIx isoforms (A4.74); D. types Ila, lib, fix isoforms (N3 J6) for the soleus. Note that the soleus is primarily composed of type I fibers with occasional Ila evident. Scale bar: SOum

138 Figure 3.3; Mean percentage (± standard deviation) of myosin heavy chain (MHC) isoform expression from soleus muscle homogenate using (SDS-PAGE) for all groups. There was no lib expression and no significant differences noted between groups at 7 weeks after injury. LAM CTL: laminectomy control; SCI No-EX: untrained; SCI-TM: treadmill trained; SCI-SW: swim trained; SCI-ST: stand trained.

139 100 so LAM CTL

00

4 0 .

2 0 .

0 100 SCI NO-EX SO

60 ■ o 0» 40 + 20 c OCB 100 E SCI-TM CO SO K t u SO m U. 40 u _ 20 O 1 - z 1 0 0 . UJ SCI-SW oco so UJ so CL 40

20

100 SCI-ST SO

SO

40

20

* T Ila lla,x IIx Ilb FIBER TY P E

Figure 3.3 140 FIGURE 3.4: Photomicrograph of a type IIx muscle fiber from the soleus of an SCI No-Ex animal as shown by myosin heavy chain antibody staining for A. type I isoforms (A 4.951); B. types I and Ila isoforms (N2.261); C. types Ila and IIx isoforms (A4.74); D. types Ila, nb, Hx isoforms (N336). Scale bar: 30 um

141 MHC SDS-PAGE SOLEUS

K Ub I

Figure 3.5: Myosin heavy chain(MHC) expression in the soleus muscle from representative animals in each group using SDS-PAGE. Note that there was no lib MHC expression for the soleus. Lane 1: control MG; Lane 2: control diaghragm; Lane 3: MHC standard; Lane 4: stand trained; Lanes5,6,8 and 10: swim trained; Lanes 7 and 9: untrained; Lanes 11 and 13: treadmill trained; Lane 12: laminectomy control.

142 Since all muscle fibers showed identical staining patterns with these two antibodies, it

was determined that no Ilb fibers were present in this sample. This was confirmed

with gel electrophoresis results (Figure 3.5).

Lastly, co-expression of multiple proteins in a single fiber was determined

from positive serially stained sections for all mAbs enlisted (N2.261, A4.74, A4.951

and N3.36). There was a trend that SCI No-Ex SOL muscle had a higher percentage

of co-expressing fibers (type I, Ila a n d IIx) compared to LAM CTLs (5.8 ± 2.2 sem vs

0 %, respectively; p=.07). LAM CTL and SCI No-Ex values, however, were not

different from all other injured groups (2.7 ± 1.1 % SCI-TM, 2.0 ± 0.7% SCI-SW, and

3.0 ± 1.3 % SCI-ST, p>.10). It appears that engaging in any of the exercise programs

used in this study modestly attenuated co-expression of fiber types in the SOL muscle.

Gel Electrophoresis and MHC Expression

The relative content of skeletal muscle MHC proteins was determined from

horizontal sections from the SOL muscle midbelly and was expressed as a percent

(Figure 3.3). There was no significant difference in the percent of any of the MHC proteins expressed for any treatment group. In addition, in our experience clear

identification of bands using Coomassie staining was not possible for very low levels of protein (<4% of total expression, < 10 ng). This was a factor for Ila and Ox proteins which migrate closely on the separating gel and were low in amount in some

SOL muscles. We, therefore, designated a Ila, x category to accommodate this

143 diffîculty. A value of < 3%” was applied to the condition in which light staining

was identified for Ila and Ux (and confirmed via immunostaining) but clear boundaries

were not readily apparent

Discussion

The major finding of this study was that muscle atrophy persisted in the large

fast twitch LG muscle 7 weeks post spinal cord contusion, which was not attenuated

with any of the exercise tasks utilized in this study. However, no significant

differences in muscle wet weights were observed between groups for the EDL, PL,

SOL or TA; nor in CSA or fiber type for the slow twitch SOL muscle. There was a

trend toward an increase in the number of SOL fibers co-expressing multiple MHC

isoforms (p=.07) identified in contused animals that did not undergo exercise as

compared to LAM CTLs. The fact that none of the trained groups demonstrated this

co-expression above normal levels suggests that participation in exercise had a minor

normalizing effect on the transcriptional events that determine MHC isoform

expression in the SOL. Lastly, muscle atrophy most likely occurred early after injury

(see Chapter 1) and presumably was resolved by the 7 week time point. The

resolution of muscle atrophy but not locomotor deficits may merely indicate that other

muscles not evaluated in this study had persistent atrophy contributing to abnormal locom otor patterns.

Muscle Fiber Atronhv

Previous work has shown that SCI can lead to muscle atrophy and conversion of muscle fiber toward faster phenotype (Talmadge et al., 1995a; Roy et al., 1991 ; Roy

144 et al., 1998). Several studies have now carefully described the rapid and lasting impact of a thoracic spinal cord TX lesion on HL skeletal muscle (Lieber et al., 1986b;

Lieber et al., 1986a; Grossman et al., 1998; Dupont-Versteegden et al., 1998;

Talmadge et al., 1995b; Roy et al., 1999; Roy et al., 1998). As early as 5 days post

TX, the SOL showed a 32% decrease in CSA, an increase expression of Ila (56%) and nx (6%), as well as an increase from 0 to 11% in the number o f fibers co-expressing multiple MHC isoforms (Dupont-Versteegden et al., 1998). Ten days after TX, CSA dropped to 40% of CTL values with a 22% increase in IIx expression and a 22% increase in the number of co-expressing fibers (Dupont-Versteegden et al., 1998).

Reports from 15 to 30 days post-TX reveal approximately a 40% decrease in SOL wet weight with either 50% or 80% of the fibers, respectively, expressing more than one

MHC isoform (Talmadge et al., 1995b). Finally, one year post TX the SOL presented with 3% type I and 97% type n fibers using myosin ATPase histochemistry (Lieber et al., 1986a), compared to the 97% type I and 3% type H normally observed in uninjured

CTLs.

These changes noted for the slow twitch SOL are in contrast to more modest changes generally observed following TX for fast twitch muscle such as a 25% decrease in wet weight in the MG and 0% decrease in the TA evaluated 6 months post

TX (Roy et al., 1999; Roy et al., 1991). However, in the fast twitch EDL, a 35% decrease in CSA (IIx and lib fibers) an increase in Ux expression (42%), but no significant change in the numbers of co-expressing fibers were revealed as early as 10 days TX (Dupont-Versteegden et al., 1998).

145 Application of Spinal Cord Contusion Iniurv for Analysis of Skeletal Muscle Change

To our knowledge, skeletal muscle properties resulting from experimental

spinal cord contusion injury have not been previously reported (see Chapter 1).

Limited studies involving human subjects generally show a similar atrophic pattern and conversion of muscle fiber toward faster phenotype (Lotta et al., 1991; Grimby et al., 1976; Castro et al., 1999; Stilwill, Sahgal, 1977; Bumham et al., 1997; Shields,

1995), but appears to occur over a somewhat slower time course compared to adaptations in the TX models (Talmadge et al., 1995b; Dupont-Versteegden et al.,

1998). In addition, differences in injury severity, location, gender, body type, as well as the use of single biopsy specimens make the extent of skeletal muscle change in humans difficult to quantify.

Experimental models of spinal cord contusion are used extensively for analysis of pathological processes associated with SCI and often use locomotion as an outcome measure for evaluating various treatment conditions (Berhmann et al., 1992). Yet changes in skeletal muscle properties after spinal cord contusion have not been previously reported (see Chapter 1). A clear understanding of skeletal muscle adaptations to spinal cord contusion injury and the potential contribution to the resultant motor output warrants extensive examination. Skeletal muscle adaptation may influence motor outcome independently of changes in CNS pathological processes. Our earlier study (Chapter 1) showed that muscle force production appeared to follow the return to weight bearing and locomotion such that early

146 decreases in force had resolved at the late stage of recovery when a consistent locomotor pattern had been developed. The findings of the current study determined that at an intermediate time point (7 weeks po), muscle properties like fiber CSA and wet weight values had returned to near normal levels for the SOL. In Chapter 1, we describe that resolution of wet weight did not rectify decreases in force production. At

7 weeks after injury, no differences in SOL CSA between injured and CTL animals were observed suggesting resolution of force producing capabilities. It should be remembered that is it likely that a more severe injury would cause greater and longer lasting changes in skeletal muscle properties compared to the results presented here.

In contrast to the spinal cord TX model which results in severe and lasting paralysis, the more clinically relevant 1.1 mm spinal cord contusion produces an injury after which the animals regain volitional HL movements and eventually engage in some locomotor activity (Basso et al., 1995; Chapter 1). This appears to significantly attenuate changes in muscle properties identified in this injury model since no difference in EDL, PL, SOL, or TA wet weight occurred, nor were differences in SOL CSA or % fiber type observed even 7 weeks after injury.

However, wet weight values, which are used as an indication of muscle fiber atrophy, may in fact obscure actual changes in muscle fiber CSA. It appears that a significant decrease in muscle wet weight is linked to muscle fiber atrophy (Roy et al., 1998), however, it does not appear to predict the severity of atrophy (Edgerton, Roy, 1996;

Grossman et al., 1998; Thomason et al., 1987). It is not entirely clear if the lack of difference in wet weight values truly represents equality of fiber CSA. In Chapter 1,

147 we describe the early significant decreases observed in muscle wet weight measures after SCI. The resolution of wet weight, as identified at 3 weeks afier injury or at 7 weeks in the current study, could be due to an increase in fluid or non-contractile tissue (Kandarian et al., 1991). In this instance, evaluation of individual fiber CSA would confirm the relationship between wet weight and individual fiber atrophy.

Wet Weight vs. Cross-Sectional Area

It should be noted that while there was no statistical difference in group means between the wet weight values of SCI No-Ex and the injured exercised animals for the

SOL muscle, neither was there a statistical difference in the fiber CSA measurement.

However, a comparison of wet weight versus CSA revealed a discrepancy in the magnitude of change across groups. The wet weight of SCI No-Ex rats was 98% of the SCI-TM value, 100% of SCI-SW, 95% o f SCI-ST and 101% of LAM CTLs.

However, the CSA for the SCI No-Ex rats, collapsed across fiber types, was 90% of the size of SCI-TM SOL, 78% of SCI-SW, 96% of SCI-ST and 84% of LAM CTL.

Therefore, as seen in earlier studies (Chapter 1), wet weight values may mask true changes in fiber area. Cross-sectional area is clearly a more sensitive measure of muscle fiber change compared to wet weight values even at 7 weeks post SCI.

Exercise and Muscle Atronhv

Several studies have shown how exercise can attenuate muscle atrophy and conversion of muscle fiber phenotype following SCI (Roy et al., 1991; Roy et al.,

1998; Roy et al., 1999). The tasks used in this study were expected to have a differential effect on muscle atrophy and in fact mean wet weight values were in the

148 direction that was expected for each training group. For example, mean SOL wet weight values were greatest for TM and ST trained animals, tasks which preferentially recruit the SOL over other plantarflexor muscle groups (Walmsley et al., 1978; Roy et al., 1991). In addition, the MG, LG, PL and TA wet weights were greatest for SW

(Table 3.3) a task which is known to recruit increased activation firom predominantly fast twitch muscle groups (Roy et al., 1991; Jasmin, Gardiner, 1987) . However, except for the LG muscle, in most cases these values were not significantly different.

We also included the measurement of cardiac wet weight in this study as an indicator of general adaptation to exercise (cardiac hypertrophy). This measure has been used to evaluate pressure or volume overload of the heart (Nair et al., 1968). It is possible that the most significant behavioral effect of training was relative to overall cardiopulmonary fimess in our animals, a factor which was not directly assessed by our behavioral methods (see Chapter 2). We found a trend that TM training induced increased cardiac hypertrophy relative to other groups with the biggest difference being for the ST trained animals (p=.05).

One is limited in the number of behavioral tests that can be used within a single group of animals and still yield reliable data. One measurement we felt we could reliably make without jeopardizing other outcome measures was post-mortem analyses of cardiac wet weight. It is known that overwork due to pathology

(hypertension, stenosis), as well as, exercise training can induce adaptations in cardiac tissue; i.e. left ventricular hypertrophy (Shiverick et al., 1975; Ehasani et al., 1978;

Hickson et al., 1979; Blomqvist, Saltin, 1983). However, the precise nature of the

149 increase in wet weight induced by TM training compared to ST training in the current study is unknown and requires further study. Continuous locomotion on the treadmill belt would likely have a higher aerobic demand than static standing activities and would thus provide the stimulus for adaptation.

One last idea relative to training and the attenuation of muscle atrophy and fiber type conversion is in regards to the active or passive nature of the task. For example, Roy et al (1998) have shown that ST and TM training attenuate loss in fiber

CSA and wet weight after spinal cord TX injury. Passive cycling movements of the

HLs following TX also attenuated early loss in CSA (Dupont-Versteegden et al., 1998;

Roy et al., 1998). However, only TM training was found to attenuate the change toward faster MHC expression that accompanied muscle atrophy after SCI (Roy et al.,

1998). Neither static stand training nor passive rhythmic HL cycling were effective in attenuating conversion of muscle fiber toward faster phenotype (Dupont-Versteegden et al., 1998; Roy et al., 1998). One difference between these two studies however is the duration of the training phase. In the Roy study (Roy et al., 1998) animals were trained for 5 months, whereas Dupont’s study (Dupont-Versteegden et al., 1998) evaluated fiber type conversion within the first 10 days following injury. Since the half life of a MHC molecule has been described as 10 days (Thomason et al., 1987), perhaps extended passive cycling sessions for 5 months are necessary to see the amelioration MHC conversion after TX injury. If no such change occurs with

150 extended passive cycling training, it is then likely that volitional rhythmic weight bearing activity is a necessary requirement for normal transcriptional events related to

SOL MHC expression after SCI.

Percent Fiber Tvpe Expression

A comparison between the percent fiber type noted via immunohistochemical staining and gel electrophoresis showed some differences. For example, type I muscle fiber accounts for approximately 94 % of all LAM CTL SOL fiber types using gel electrophoresis measurements and approximately 88% using immunohistochemical sampling. It is possible that the discrepancy in fiber composition of the SOL related to error introduced by sampling. In addition, differences may be a reflection of fiber size. Type I fibers are larger than the fast fibers identified in this study (i.e. Type I fibers: 2346 ±416 sd vs. Type Ha: 1478 ± 351 um^for LAM CTLs). Therefore, absolute values of protein expression as determined by gel electrophoresis is dependent on both the proportion of fibers expressing the protein of interest and their relative size.

The Gastrocnemii Muscles and Locomotion

In the current study, LG atrophy persisted even 7 weeks post spinal cord contusion injury. This muscle is not thought to be a key player for locomotor behavior

(Smith et al., 1977), although it may play an important role in locomotor ftmction (see below). Electromyographic (EMG) records normally show low levels of activity for the LG compared to SOL muscle during TM locomotion at average speeds (Alaimo et al., 1984; Walmsley et al., 1978; Smith et al., 1977). Following spinal cord contusion

151 injury, rats can walk in the open field but with considerable residual deficits (Figures

1.1 and 2.1) such as toe drags, incoordination, abnormal paw and tail positioning, and poor trunk control. Perhaps the timing of LG activation during the locomotor cycle is more critical than the duration or amplitude of its EMG activation (Alaimo et al.,

1984; Smith et al., 1977). For example, LG bursts which peak slightly after SOL during the push off phase of gait (Smith et al., 1977) may lead to faster and improved clearance of the swing limb and subsequently improved limb placement, balance and coordination. It is clear that attenuation of LG atrophy did not occur when engaging in the tasks used in this study, and therefore these tasks probably did not recruit LG activation to a significant extent (Alaimo et al., 1984). Electromyographic analysis of the recruitment and timing of the LG relative to other lower limb muscle groups is necessary to determine the relative contribution of LG muscle activation to the locomotor pattern after contusion injury.

Muscle atrophy of the MG was also identified fi’om wet weight measures showing significant decreases in the SCI-TM and SCI-ST groups compared to the

LAM CTLs. However, no differences in wet weight were observed in the SCI No-Ex group or the SCI-SW groups. This seems to suggest that engaging in TM or ST training after contusion injury actually induced MG muscle atrophy. TM and ST tasks are known to target SOL muscle activation as much or more than the MG in normal animals (Smith et al., 1977; Walmsley et al., 1978). Theoretically, training could have resulted in hypertrophy of SOL muscle with compensatory MG atrophy (particularly since high force requirement tasks, like jumping or swimming, were not part of these

152 animals’ daily experience). However, SOL CSA was not significantly different

between groups (Table 3.3), which argues gainst SOL muscle hypertrophy. As in the

case of the LG muscle, EMG analysis of recruitment and timing of the MG relative to

other lower limb muscles groups during locomotion and training is necessary to

determine the impact of atrophy identified in this injury model.

Caveats

Due to the prohibitive cost of time and money, detailed analysis of muscle

fiber CSA and muscle fiber type were completed only for the SOL muscle group.

Although wet weight analyses were carried out for several HL muscles, further

evaluation is necessary to precisely determine the impact of spinal cord contusion on

fiber CSA and MHC expression, particularly for the MG and LG muscle groups. In

addition, EMG analyses of multiple HL muscles before and afier a contusive SCI,

would give a stronger indication of the relative contribution of each muscle to

recovered locomotor patterns.

Conclusion

It is possible that other muscles or muscle properties, which were negatively affected by this injury, were not evaluated in this study (i.e. stabilizing lower abdominal and/or hip extensor musculature or muscle fatigue characteristics). Also, following a moderate spinal cord contusion injury, locomotor deficits identified using the BBB assessment scale may be purely due to aberrant central processing mechanisms and impaired descending control independent of changes in peripheral

153 skeletal muscle. Lastly, one must keep in mind that the results of the current study are relative to the 1.1 mm spinal cord contusion injury. A more severe injury would have greater paralytic effects and subsequently may induce greater changes in skeletal muscle properties than those described in this communication.

154 LIST OF REFERENCES

1. Alaimo MA, Smith JL, Roy RR, Edgerton VR (1984) EMG activity of slow and fast ankle extensors following spinal cord transection. J.Appl.Physiol. 56:1608-1613.

2. Basso DM, Beattie MS, Bresnahan JC (1995) A sensitive and reliable locomotor rating scale for open field testing in rats. J Neurotrauma 12:1-21.

3. Basso DM, Beattie MS, Bresnahan JC (1996) Graded histological and locomotor outcomes after spinal cord contusion using the NYU weight drop device versus transection. Exp.Neurol. 139:244-256.

4. Berhmann DL, Bresnahan JC, Beattie MS, Shah BR (1992) Spinal cord injury produced by consistent mechanical displacement of the cord in rats: Behavioral and histologic analysis. J Neurotrauma 9:197-216.

5. Blomqvist CG, Saltin B (1983) Cardiovascular adaptations to physical training. Annu.Rev.Physiol. 45:169-189.

6. Burnham R, Martin T, Stein R, Bell G, Maclean 1, Steadward R (1997) Skeletal muscle fiber type transformation following spinal cord injury. Spinal Cord 35:86-91.

7. Castro MJ, Apple DF, Jr., Staron RS, Campos GER, Dudley GA (1999) Influence of complete spinal cord injury on skeletal muscle within 6 mo of injury. J.Appl.Physiol. 86:350-358.

8. Cho M, Webster SG, Blau H (1993) Evidence for myoblast-extrinsic regulation of slow myosin heavy chain expression during muscle fiber formation in embryonic development. J.Cell Biol. 121:795-810.

9. de Leon R, Hodgson JA, Roy RR, Edgerton VR (1994) Extensor- and flexor-like modulation within motor pools of the rat hindlimb during treadmill locomotion and swimming. Brain Res. 654:241-250.

155 10. EKipont-Versteegden EE, Houle JD, Gurley CM, Peterson CA (1998) Early changes in muscle fiber size and gene expression in response to spinal cord transection and exercise. Am.J.Physiol. 275:C1124-Cl 133

11. Edgerton VR, Roy RR (1996) Neuromuscular adaptations to actual and simulated spaceflight In: Handbook of Physiology (Fregly MJ, Blatteis CM eds), pp 721-763. New York: Oxford University Press.

12. Ehasani AA, Hagberg JM, Hickson RC (1978) Rapid changes in left ventricular dimensions and mass in response to physical conditioning and deconditioning. Am J.Cardiol. 42:52-56.

13. Goldspink G (1998) Selective gene expression during adaptation of muscle in response to different physiologic demands. Comp.Biochem.Physiol.B.Biochem.Mol.Biol. 120:5-15.

14. Green TL, Hunter DD, Chan W, Merlie JP, Sanes JR (1992) Synthesis and assembly of synaptic cleft protein s-laminin by cultured cells. J Biol.Chem. 267:2014-2022.

15. Grimby G, Broberg C, Krotkeiwska I, Krotkiewski M (1976) Muscle fiber composition in patients with traumatic cord lesion. Scand.J Rehabil.Med. 8:37-42.

16. Grossman EJ, Roy RR, Talmadge RJ, Zhong H, Edgerton VR (1998) Effects of inactivity on myosin heavy chain composition and size of rat soleus fibers. Muscle Nerve 21:375-389.

17. Hickson RC, Hammons GT, Holloszy JO (1979) Development and regression of exercise-induced cardiac hypertrophy. Am.J.Physiol. 236:H268-H272

18. Hughes SM, Blau HM (1992) Muscle fiber pattern is independent of cell lineage in postnatal rodent development. Cell 68:659-671.

19. Hughes SM, Cho M, Karsch-Mizrachi I, Travis M, Silberstein L, Leinwand LA, Blau H (1993) Three slow myosin heavy chains sequentially expressed in developing mammalian skeletal muscle. Dev Biol. 158:183-189.

156 20. Jasmin BJ, Gardiner PF (1987) Patterns of EMG activity of rat plantaris muscle during swimming and other locomotor activities. J.Appl.Physiol. 63:713- 718.

21. Kandarian SC, Boushel RC, Schulte LM (1991) Elevated interstitial fluid volume in rat soleus muscles by hindlimb unweighting. J.Appl.Physiol. 71:910- 914.

22. Laemmli UK (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680-685.

23. Lieber RL (1992) Skeletal muscle structure and function: Implications for rehabilitative and sports medicine. Baltimore: Williams and Wilkins Pub.

24. Lieber RL, Frieden JO, Hargens AR, Ferringa ER (1986a) Long-term effects of spinal cord transection on fast and slow rat skeletal muscle. II. Morphometric properties. Exp Neurol. 91:435-448.

25. Lieber RL, Johannson CB, Vahlsing HL, Hargens AR, Ferringa ER (1986b) Long-term effects of spinal cord transection on fast and slow rat skeletal muscle. I. Contractile properties. Exp.Neurol. 91:423-434.

26. Lotta S, Scelsi R, Alfbnsi E, Saitta A, Nicolotti D, Epifani P, Carraro U (1991) Morphometric and neurophysiological analysis of skeletal muscle in paraplegic patients with traumatic cord lesion. Paraplegia 29:247-252.

27. Nair KG, Cutilleta AF, Zak R, Koide T, Rabinowitz M (1968) Biochemical correlates of cardiac hypertrophy. I. Experimental model; changes in heart weight, RNA content, and nuclear RNA polymerase activity. Circules. 23:451-462.

28. Reiser PJ, Moss RL, Giulian GG, Greaser ML (1985a) Shortening velocity and myosin heavy chains of developing rabbit muscle fibers. J Biol.Chem. 260:14403-14405.

29. Reiser PJ, Moss RL, Giulian GG, Greaser ML (1985b) Shortening velocity in single fibers fiom adult rabbit soleus muscles is correlated with myosin heavy chain composition. J Biol.Chem. 260:9077-9080. 157 30. Roy RR, Baldwin KM, Edgerton VR (1991) Plasticity of skeletal muscle: Effects of neuromuscular activity. Exerc.Sport Sci.Rev. 19:269-312.

31. Roy RR, Hirota WK, Kuehl M, Edgerton VR (1985) Recruitment patterns in the rat hindlimb muscle during swimming. Brain Res. 337:175-178.

32. Roy RR, Hutchinson DL, Pierotti DJ, Hodgson JA, Edgerton VR (1991) EMG patterns of rat ankle extensors and flexors during treadmill locomotion and swimming. J Appl Physiol. 70:2522-2529.

33. Roy RR, Pierotti DJ, Baldwin KM, Zhong H, Hodgson JA, Edgerton VR (1998) Cyclical passive stretch influences the mechanical properties of the inactive cat soleus. Exp Physiol. 83:377-385.

34. Roy RR, Talmadge RJ, Hodgson JA, Oishi Y, Baldwin KM, Edgerton VR (1999) Differential response of fast hindlimb extensor and flexor muscles to exercise in adult spinalized cats. Muscle Nerve 22:230-241.

35. Roy RR, Talmadge RJ, Hodgson JA, Zhong H, Baldwin KM, Edgerton VR (1998) Training effects on soleus of cats spinal cord transected (T12-13) as adults. Muscle Nerve 21:63-71.

36. Sanes JR, Engvall E, Butkowski R, Hunter DD (1990) Molecular heterogeneity of basal laminae: Isoforms of laminin and collagen iv at the neuromuscular junction. J.Cell Biol. 111:1685-1699.

37. SchiafBno S, Ausoni S, Gorza L, Saggin I, Gundersen K, Lomo T (1988) Myosin heavy chain isoform and velocity of shortening of type 2 skeletal muscles fibers. Acta Physiol.Scand. 134:575-576.

38. Shields RK (1995) Fatigability, relaxation properties, and electromyographic responses of the human paralyzed soleus muscle. J Neurophysiol. 73:2195-2206.

39. Shiverick KT, Thomas LT, Alpert NR (1975) Purification of cardiac myosin: Application to hypertrophied myocardium. Biochim.Biophys.Acta 393:124-133.

158 40. Sieck G, Han YS, Prakash TS, Jones KA (1998) Crossbridge cycling kinetics, actomyosin ATPase activity and myosin heavy chain isofonns in skeletal and smooth respiratory muscles. Comp.Biochem.Physiol.B.Biochem.Mol.Biol. 119:435-450.

41. Sieck GC (1994) Physiological effects of diaphragm muscle denervation and disuse. Clin.Chest Med. 15:641-659.

42. Smith JL, Edgerton VR, Betts B, Collates TC (1977) EMG of slow and fast ankle extensors of cat during posture, locomotion, and jumping. J.Neurophysiol. 40:503-513.

43. Stilwill EW, Sahgal V (1977) Histochemical and morphologic changes in skeletal muscle following cervical cord injury: A study of upper and lower motor neuron lesions. Arch Phys Med Rehabil. 58:201-206.

44. Stokes BT, Homer PJ, Akino M (1995) Spinal cord injury modeling and functional assessment. In: Central Nervous System Trauma: Research Techniques (Ohnishi ST, Ohnishi T eds), pp 281-295. New York: CRC Press.

45. Talmadge RJ, Roy RR (1993) Electrophoretic separation of rat skeletal muscle myosin heavy chain isoforms. JjVppLPhysiol. 75:2337-2340.

46. Talmadge RJ, Roy RR, Bodine-Fowler SC, Pierotti DJ, Edgerton VR (1995a) Adaptations in myosin heavy chain profile in chronically unloaded muscles. BAM 5:117-137.

47. Talmadge RJ, Roy RR, Edgerton VR (1995b) Prominence of myosin heavy chain hybrid fibers in soleus muscle of spinal cord-transected rats. J Appl Physiol. 78:1256-1265.

48. Thomason DB, Herrick RE, Surdyka D, Baldwin KM (1987) Time course of soleus muscle myosin expression during hindlimb suspension and recovery. J.Appl.Physiol. 63:130-137.

159 49. Walmsley B, Hodgson JA, Burke RE (1978) Forces produced by medial gastrocnemius and soleus muscles during locomotion in fieely moving cats. J.NeurophysioI. 41:1203-1216.

50. Webster C, Silberstein L, Hays AP, Blau H (1988) Fast muscle fibers are preferentially affected in Duchenne muscular dystrophy. Cell 52:503- 513.

160 APPENDIX A

1 6 1 BBB Subscoring Scale

Rotated vs. Parallel Paw Toe Clearance Trunk Control Tail Position Placement Right Left Right Left Unstat)le 0 Down 0 IC/LO ICÆ.O None 0 0 Stable 1 Up/Down 1 R/R 0 0 Occ 1 1 Up 2 RJP. P/R 1 1 Freq 2 2 P/P 2 2 Cons 3 3 lO initial contact Occ= occasional LO= lift ôff Freq* frequent (>50%) R= rotated Cons- consistent (95-100%) P= parallel

162 APPENDIX B

163 TASK TRAINING: A) BIPEDAL TREADMILL LOCOMOTION B) QUADRUPEDAL TREADMILL LOCOMOTION Q STAND TRAINING D) SWIM TRAINING

164 APPENDIX C

165 BASSO, BEATTIE & BRESMAHAN LOCOMOTOR RATING SCALE

0 So observable hindlimb (HL) movement

1 Slight movement of one or two joints, usually the hip Sc/or knee

2 Extensive movement of one joint or Extensive movement of one joint and slight movement of one other joint

3 Extensive movement of two joints

4 Slight movement of all three joints of the HL

5 Slight movement of two joints and ezteiuive movement of the third

6 Extensive movement of two joints and slight movement of the third

7 Extensive movement ofaO three joints of the HL

S Sweeping with no weight support or Plantar placement of the paw srith no weight support

9 Plantar placement of the paw with weight support in stance only (Le. when stationary) or Occasional, Frequent or Consistent weight supported donal stepping and no plantar stepping

10 Occasional weight supported plantar steps, no FL-HL coordination

11 Frequent to consistent weight supported plantar steps and no FL-HL coordination

12 Frequent to consistent weight supported plantar steps and occasional FL-HL coonlination

13 Frequent to consistent weight supported plantar steps and frequent FL-HL coordination

14 Consistent weight supported plantar steps, consistent FL-HL coordination; and. Predominant paw position during locomotion is rotated (internal^ or externally) when it makes initial contact with the surface as well as Just before it is lifted off at the end of stance or

Frequent plantar stepping, consistent FL-HL coordination and occasional dorsal stepping

revised 2m /94

1 6 6 15 Consistent plantar stepping and Consistent FL-HL coordination; and. No toe clearance or occasional toe clearance during forward limb advancement Predominant paw position is parallel to the body at initial contact

16 Consistent plantar stepping and Consistent FL-HL coordination during gait; and Toe clearance occurs frequently during Torward limb advancement Predominant paw position is parallel at initial contact and rotated at lift off

17 Consistent plantar stepping and Consistent FL-HL coordination during gait; and Toe clearance occurs frequently during forward limb advancement Predominant paw position is parallel at initial contact and lift off

18 Consistent plantar stepping and Consistent FL-HL coordination during gait; and Toe clearance occurs consistently during forward limb advancement Predominant paw position is parallel at initial contact and rotated at lift off

19 Consistent plantar stepping and Comistent FL-HL coordination during gait; and Toe clearance occurs consfttently diiring forward limb advancement Predominant paw position is parallel at initial contact and lift off; and. Tail is down part or all of the time

20 Coiuistent plantar stepping and Coiuistent coordinated gait; consistent toe clearance; Predominant paw position is parallel at initial contact and lift off; and T runk instabüiQr Tail consistently up

21 Consistent plantar stepping and Coordinated gait, consistent toe clearance, predominant paw position is parallel throughout stance, consistent trunk stabOiQr; tad consistently up

DEFINITIONS Slight: partial joint movement through less than 1/2 the range o f joint motion Extensive: movement through more than half of the range of joint motion Sweeping: rhythmic movement of HL in which all three joints are extended, then fiilly flex and extend again; animal is usually sidelying and plantar surftce o f paw may or may not contact the ground; no weight support across the HL is evident No Weight Support: no contraction of the extensor muscles o f the HL during plantar placement of the paw; or no elevation of the hindquaner Weight Support: contraction o f the extensor muscles of the HL during plantar placement of the paw; or, elevation o f the hindquaner Plantar Stepping: The paw is in p la n ta contact r with weight suppon then the HL is advanced forward and p la n ta contact r with weight support is re-established Dorsal Stepping: Weight is supponed through the dorsal surface o f the paw at some point in the step (ycle. FL-HL Coordination: For every FL step a HL step is taken and the HLs alternate Occasional: less than or equal to half; s 50% Frequent: more than half but not always; 51-94% Consistent: nearly always or always; 95-100% T runk Instability: Lateral weight shifts which cause waddling ftom side to side or a partial collapse of the trunk. revised 2J2Sn*

167 APPENDIX D

168 WALKWAY LOCOMOTOR RANKING: DECISION RULES

ADDENDUM:

Stepping: Establish weight support, swing and reestablish weight supporL Evidence o f w e ^ t support includes visible muscle contraction and/or efevation of the hindquarters during weight bearing phases o f gait.

Frequent Stepping- if ^50% of the attenqpted steps are actual steps. However, consider the following. ( To be a Sequent stepper you must have more than 2 steps in one pass.) At terminal stance if the rat foils to lift off its’ hindlinnh (HL), ft is considered an e^qiected but not produced step. If more than half of the steps (^50% ) are expected steps the rat 6 dropped ftom Sequent to occasional stepping (unless ft is already occasional and ft will not drop any further). Le. if 2/4 steps, or 2/3 steps are eiqxcted but not produced -drop them down.

Swing vs. prolonged extension:Swing will result in a change in paw position relative to the position of the body / hindquarters, whereas prolonged extensfon will not. Dorsals:if ^50% o f the steps made are dorsals, drop them to occasional stepping.

Excess yeildcan be observed at either Initial Contact (IC) or midstance, whereas, dÿs refer specifically to IC.

Coordination:Consistent refers to, “for every forelimb (FL) step a HL step is taken and the HLs altemate”; rule ftom the BBB scale. Also, the pattern of coordftiatfon requires the presence o f at least 2 steps in one pass containing the elements : Diagonal—simuNnneous pattern. Contralateral (diagonal) FL and HL pairs advance - simultaneously ( seen in normal animals). - 4 point pattern; the amount of overlap between swing phase limbs will determine this pattern. If one pair is > halfway through swing before its’ pair initiates swing and the animal is consistently coordinated then that is scored as a 4 pomt pattern rather than a diagonal simultaneous pattern. Ipsilateral-simultaneous pattern. Advance of ipsiiateralFL/HL - 4 point pattern;Left FL, Left HL, Right FL , Right HL.

169 WALKWAY LOCOMOTION EVALUATION FORM

DATE: Rat#: Trial #1 Trial #1 Trial #1 Tape# Tima Coda > Evaluatof: Headed (L/R) LRL R L R Tiarl: ConaMantalappina (midlirM mmh) Conaiat. Coofd. Coca. Caod. Diagonal Panam 4ptaait inailMamOnlv Tiar 2: Fraq-> Conaiat. ataooina (not coofdJatap on 1HU HL Cdtapaa-* (ioa at inWai oortad ** exceaa vMd IC / danoa Hcpcino- 2alaoaarlaaa >2ataoa Plantar itaoa > 2 <2 DonalSlaH)ino->2 <2 TrunkCTL- (otf around)—laan Htah Normal Low Tai (UiXDown) 1 TItrJ: Otcadonal Sdnaiaa Fata lo aida HL CoHaoaa-unable to raaatablitti wL SuDoart Plantar Staooino-> 2 <2 DofaaiSlaooina->2 <2 HL Collapap-* dtoa at MHai contact ** axcaaa vield IC / alanoa HooDbiQ- 2ataoaorlaaa >2atapa SidaMno (L R . Mdda) Prokmoed aatanaorchaaa 1 1 1 Tam(UolDown) 1 Tier* Norttaaiaa PLPLnowtauppoit PLPLwtauDPOrt Draooina SkanQ (CompMalv llamd) Skana(agdandadHL'a> Talowaradopor WW ~ s > 1/2 ranaa for H. K. & A * = obviouawtsuooaftin atance. HQ raised COMMENTS:

170 BIBLIOGRAPHY

1. Advokat C, Duke M (1999) Comparison of morphine-induced effects on thermal nociception, mechanoreception, and hind limb flexion in chronic spinal rats. Exp.Clin.Psychopharmacol. 7:219-225.

2. Alaimo MA, Smith JL, Roy RR, Edgerton VR (1984) EMG activity of slow and fast ankle extensors following spinal cord transection. J Appl Physiol. 56:1608-1613.

3. Aldskogius H, Kozlova EN (1998) Central neuron-glial and glial-glial interactions following axon injiuy. Prog.Neurobiol. 55:1-26.

4. Armstrong DM (1988) The supraspinal control of mammalian locomotion. JPhysiol. 405:1-37.

5. Balentine JD, Paris DU (1978a) Pathology of experimental spinal cord trauma 1. The necrotic lesion as a function of vascular injury. Lab.lnvest. 39:236- 253.

6. Balentine JD, Paris DU (1978b) Pathology of experimental spinal cord trauma H. Ultrastructure of axons and myelin. Lab.lnvest. 39:254-265.

7. Barbeau H, Rossignol S (1987) Recovery of locomotion after chronic spinalization in the adiüt cat. Brain Res. 412:84-95.

8. Basso DM, Beattie MS, Bresnahan JC (1995) A sensitive and reliable locomotor rating scale for open field testing in rats. J.Neurotrauma 12:1-21.

9. Basso DM, Beattie MS, Bresnahan JC (1996) Graded histological and locomotor outcomes after spinal cord contusion using the NYU weight drop device versus transection. Exp Neurol. 139:244-256.

10. Basso DM, Beattie MS, Bresnahan JC (2000) Descending systems contributing to locomotor recovery after mild or moderate spinal cord injury in rats. Manuscript in preparation. 171 11. Basso DM, Murray M, Goldberger ME (1994) Differential recovery of bipedal and overground locomotion following complete spinal cord hemisection in cats. Restor.Neurol.Neurosci. 7:95-110.

12. Beattie MS, Bresnahan JC, Komon J, Tovar CA, Van Meter M, Anderson DK, Faden AI, Hsu CY, Noble LJ, Salzman S, Young W (1997) Endogenous repair after spinal cord contusion injuries in the rat. Exp Neurol. 148:453- 463.

13. Beattie MS, Shuman SL, Bresnahan JC (1998) Apoptosis and spinal cord injury. Neuroscientist 4:163-171.

14. Becerra JL, Puckett WR, Heister ED, Quencer RM, Marcillo AE, Post MJ, Bunge RP (1995) MR-pathologic comparisons of wallerian degeneration. AJNR 16:125-133.

15. Belanger M, Drew T, Provencher J, Rossignol S (1996) A comparison of treadmill locomotion in adult cats before and aAer spinalization. J.Neurophysiol. 76:471-491.

16. Bem T, Gorska T, Majczynski H, Zmyslowski W (1995) Different patterns of fore-hindlimb coordination during overground locomotion in cats with ventral and lateral spinal lesions. Exp .Brain Res. 104:70-80.

17. Beimett GJ, Xie YK (1988) A peripheral mononeuropathy in rat that produces disorders of pain sensation like those seen in man. Pain 33:87-107.

18. Berhmann DL, Bresnahan JC, Beattie MS, Shah BR (1992) Spinal cord injury produced by consistent mechanical displacement of the cord in rats: Behavioral and histologic analysis. J.Neurotrauma 9:197-216.

19. Black JE, Issacs KR, Anderson BJ, Alacantara AA, Greenough WT (1990) Learning causes synaptogenesis, whereas motor activity causes angiogenesis, in cerebellar cortex of adult rats. Proc.Natl.Acad.Sci.U.S.A. 87:5568-5572.

20. Blomqvist CG, Saltin B (1983) Cardiovascular adaptations to physical training. Annu.Rev.Physiol. 45:169-189.

172 21. Blough ER (1997) The effects of aging and overload on plantaris and extensor digitorum longus morphology and physiology in adult aged Fisher 344 and FI hybrid rats. Doctoral dissertation. The Ohio State University.

22. Bracken MB, Shepard MJ, Collins WF, Holford TR, Young W, Baskin DS, Eisenberg HM, Flamm E, Leo-Summers L, Maroon J, Marshall LF, Perot PL, Jr., Peipmeier J, Sonntag VKH, Wagner FC, Wilberger JE, Winn HR (1990) A randomized, controlled trial of methylprednisolone or naloxone in the treatment of acute spinal-cord injury. N.Engl.J.Med. 322:1405- 1411.

23. Bregman B, Goldberger ME (1983) Infant lesion effect. H. Sparing and recovery of function after spinal cord damage in newborn and adult cats. Brain Res. 285:119-135.

24. Brenowitz GL, Pubols LM (1981) Increased receptive field size of dorsal horn neurons following chronic spinal cord hemisections in cats. Brain Res. 216:45-59.

25. Bresnahan JC (1978) An electron-microscopic analysis of axonal alterations following blunt contusion of the spinal cord of the rhesus monkey (Macaca mulatta). J.Neurol.Sci. 37:59-82.

26. Bresnahan JC, Beattie MS, Todd FD, Noyes DH (1987) A behavioral and anatomical analysis of spinal cord injury produced by a feedback- controlled impaction device. Exp.Neurol. 95:548-570.

27. Bresnahan JC, King JS, Martin GF, Yashon D (1976) A neuroanatomical analysis of spinal cord injury in the rhesus monkey (Macaca mulatta). J.Neurol.Sci. 28:521-542.

28. Brown M, Hasser EM (1995) Weight-bearing effects on skeletal muscle during and after simulated bed rest. Arch.Phys.Med.Rehabil. 76:541-546.

29. Brustein E, Rossignol S (1998) Recovery of locomotion after ventral and ventrolateral spinal lesions in the cat I. Deficits and adaptive mechanisms. J.Neurophysiol. 80:1245-1267.

173 30. Buffelli M, Pasino E, Cangiano A (1997) Paralysis of rat skeletal muscle equally affects contractile properties as does permanent denervation. J.Muscle Res.CeU Motü. 18:683-695.

31. Bunge RP, Puckett WR, Becerra JL, Marcillo A, Quencer RM (1993) Observations on the pathology of human spinal cord injury. A review and classification for 22 new cases with details fi’om a case of chronic cord compression with extensive focal demyelination. Adv.Neurol. 59:75-89.

32. Burnham R, Martin T, Stein R, Bell G, Maclean I, Steadward R (1997) Skeletal muscle fiber type transformation following spinal cord injury. Spinal Cord 35:86-91.

33. Bussel B, Roby-Brami A, Yakovleff A (1996) Evidence for a spinal stepping generator in man. Electrophysiological study. Acta NeurobioLExp. 56:465-468.

34. Calancie B, Needham-Schropshire B, Jacobs P, Wilier K, Zych G, Green BA (1994) Involuntary stepping after chronic spinal cord injury. Evidence for central rhythm generator for locomotion. Brain 117:1143-1159.

35. Campbell JN, Raja SN, Meyer RA, Mackinnon SE (1989) Myelinated afferents signal the hyperalgesia associated with nerve injury. Pain 32:89-94.

36. Castro MJ, Apple DF, Jr., Staron RS, Campos GER, Dudley GA (1999) Influence of complete spinal cord injury on skeletal muscle within 6 months of injury. J.Appl.Physiol. 86:350-358.

37. Chang CW, Lien IN (1991) Estimate of motor conduction in human spinal cord: Slowed conduction in spinal cord injury. Muscle Nerve 14:990-996.

38. Chaplan SR, Bach FW, Pogrei JW, Chung JM, Yaksh TL (1994) Quantitative assessment of tactile allodynia in the rat paw. J.Neurosci.Methods 53:55- 63.

39. Cheng H, Almstrom S, Gimenez-Lort L, Chang R, Ove Ogren S, Hoffer B, Olson L (1997) Gait analysis of adult paraplegic rats after spinal cord repair. Exp.Neurol. 148:544-557.

174 40. Cheng H, Gao Y, Olson L (1996) Spinal cord repair in adult paraplegic rats: Partial restoration of hind limb function. Science 273:510-513.

41. Cho M, Webster SG, Blau H (1993) Evidence for myoblast-extrinsic regulation of slow myosin heavy chain expression during muscle fiber formation in embryonic development. J.Cell Biol. 121:795-810.

42. Christensen MD, Everhart AW, Pickelman JT, Hulsebosch CE (1996) Mechanical and thermal allodynia in chronic central pain following spinal cord injury. Pain 68:97-107.

43. Christensen MD, Hulsebosch CE (1997) Chronic central pain after spinal cord injury. J.Neurotrauma 14:517-537.

44. Collet P, Uebelhart D, Vico L, Moro L, Hartmann D, Roth M, Alexandre C (1997) Effects of 1- and 6- month spaceflight on bone mass and biochemistry in two humans. Bone 20:547-551.

45. Cope TC, Bodine SC, Fournier M, Edgerton VR (1986) Soleus motor units in chronic spinal transected cats: Physiological and morphological alterations. J.Neurophysiol. 55:1202-1220.

46. Crowe MJ, Bresnahan JC, Shuman SL, Masters JN, Beattie MS (1997) Apoptosis and delayed regeneration after spinal cord injury in rats and monkeys. Nat.Med. 3:73-76.

47. Cruse H, Wamecke H (1992) Coordination of the legs of a slow-walking cat. Exp Brain Res. 89:147-156.

48. Czeh G, Gallego R, Kudo N, Kuno M (1978) Evidence for the maintenance of motomeurone properties by muscle activity. J Physiol. 281:239-252.

49. Davey DF, Dunlop C, Hoh JFY, Wong SYP (1981) Contractile properties and ultrastructure of extensor digitorum longus and soleus muscles in spinal cord transected rats. Aust.J.Exp.Biol.Med.Sci. 59:393-404.

175 50. Davidoff G, Roth E, Guarracim M, Sliwa J, Yarkony G (1987) Function-iimiting dysesthetic pain syndrome among traumatic spinal cord injury patients: A cross-sectional study. Pain 29:39-48.

51. de Leon R, Hodgson JA, Roy RR, Edgerton VR (1994) Extensor- and flexor-like modulation within motor pools of the rat hindlhnb during treadmill locomotion and swimming. Brain Res. 654:241-250.

52. de Leon R, Hodgson JA, Roy RR, Edgerton VR (1998a) Full weight-bearing hindlimb standing following stand training in the adult spinal cat. J JNeurophysiol. 80:83-91.

53. de Leon R, Hodgson JA, Roy RR, Edgerton VR (1998b) Locomotor capacity attributable to step training versus spontaneous recovery after spinalization in adult cats. J.Neurophysiol. 79:1329-1340.

54. de Leon R, Hodgson JA, Roy RR, Edgerton VR (1999) Retention of hindlim b stepping ability in adult spinal cats after the cessation of step training. J.Neurophysiol. 81:85-94.

55. Dietz V, Colombo G, Jensen L, Baumgartner L (1995) Locomotor capacity of spinal cord in paraplegic patients. Ann.Neurol. 37:574-582.

56. Diffee GM, Caiozzo VJ, Herrick RE, Baldwin KM (1991) Contractile and biochemical properties of rat soleus and plantaris after hindlimb suspension. Am.J.Physiol. 260:C528-C534

57. Dixon WJ (1948) A method for obtaining and analyzing sensitivity data. JAm.StatAssoc. 43:109-126.

58. Dupont-Versteegden EE, Houle JD, Gurley CM, Peterson CA (1998) Early changes in muscle fiber size and gene expression in response to spinal cord transection and exercise. Am.J.Physiol. 275:C 1124-C1133

59. Edgerton VR, de Leon R, Tillakaratne N, Recktenwald MR, Hodgson JA, Roy RR (1997) Use-dependent plasticity in spinal stepping and standing. Adv Neurol. 72:233-247.

176 60. Edgerton VR, Roy RR (1996) Neuromuscular adaptations to actual and simulated spaceflight. In: Handbook of Physiology (Fregly MJ, Blatteis CM eds), pp 721-763. New Yoric: Oxford University Press.

61. Ehasani AA, Hagberg JM, Hickson RC (1978) Rapid changes in left ventricular dimensions and mass in response to physical conditioning and deconditioning. Am J.Cardiol. 42:52-56.

62. Eide PR, Stubhaug A, Stenehjem AE (1995) Central dysesthesia pain after traumatic spinal cord injury is dependent on JV-Methyl-D-aspartate receptor activation. Neurosurgery 37:1080-1087.

63. Eidelberg E, Story JL, Walden JG, Meyer BL (1981) Anatomical correlates of return of locomotor function after partial spinal cord lesions in cats. Exp.Brain Res. 42:81-88.

64. English AW (1980) Interlimb coordination during stepping in the cat: Effects of dorsal column section. J.Neurophysiol. 44:270-279.

65. Fauteck SP, Kandarian SC (1995) Sensitive detection of myosin heavy chain composition in skeletal muscle under different loading conditions. Am.J.Physiol. 268:C419-C424

66. Finol HJ, Lewis DM, Owens R (1981) The effects of denervation on contractile properties of rat skeletal muscle. J.Physiol. 319:81-92.

67. Fiore CE, Dieli M, Vintaloro G, Gibilaro M, Giacone G, Cottini E (1996) Body composition and bone mineral density in competitive athletes in different sports. Int J.Tissue React. 18:121-124.

68. Fitts RH, Hollos^ JO (1977) Contractile properties of rat soleus muscle: Effects of training and fatigue. Am.J.Physiol. 233:C86-C91

69. Fitts RH, Metzger JM, Riley DA, Unsworth BR (1986) Models of disuse: A comparison of hindlimb suspension and immobilization. J Appl Physiol. 60:1946-1953.

177 70. Forsseberg H, Grillner S, Rossignol S (1977) Phasic gain control of reflexes firom the dorsum of the paw during spinal locomotion. Brain Res. 132:121-139.

71. Fung J, Barbeau H (1994) Effects of conditioning cutaneous stimulation on the soleus H reflex in normal and spastic paretic subjects during walking and standing. J.Neurophysiol. 72:2090-2104.

72. Gale K, Kerasidis H, Wrathall JR (1985) Spinal cord contusion in the rat: Behavioral analysis of functional neurologic impairment. Exp.Neurol. 88:123-134.

73. Gerin C, Becquet D, Privât A (1995) Direct evidence for the link between monoaminergic descending pathways and motor activity. 1. A study with microdialysis probes implanted in the ventral funiculus of the spinal cord. Brain Res. 704:191-201.

74. Gharakhanlou R, Chadan S, Gardiner P (1999) Increased activity in the form of endurance training increases calcitonin gene-related peptide content in lumbar motoneuron cell bodies and in sciatic nerve in the rat. Neuroscience 89:1229-1239.

75. Goldberger ME, Bregman BS, Vierck CJ, Brown M (1990) Criteria for assessing recovery of function after spinal cord injury: Behavioral methods. Exp.Neurol. 107:113-117.

76. Goldspink G (1998) Selective gene expression during adaptation of muscle in response to different physiologic demands. Comp.Biochem.Physiol.B.Biochem.Mol.Biol. 120:5-15.

77. Gorbunova AV, Portugalov W (1976) Cytochemical investigations of proteins and RNA in spinal motoneurons and neurons of spinal ganglia of the rat after space flight. Aviat.Space Environ.Med. 47:708-710.

78. Gordon J (1991) Spinal mechanisms of motor coordination. In: Principles of Neural Science. (Kandel ER, Schwartz JS, Jessell TM eds), pp 581-595. New York: Elsevier.

178 79. Gômez-Pinilla F, So V, Kesslak JP (1998) Spatial learning and physical activity contribute to the induction of fibroblast growth factor; Neural substrates for increased cognition associated with exercise. Neuroscience 85:53-61.

80. Gorska T, Bern T, Majczynski H, Zmysiowski W (1993) Unrestrained walking in cats with partial spinal lesions. Brain Res.Bull. 32:241-249.

81. Green TL, Hunter DD, Chan W, Merlie JP, Sanes JR (1992) Synthesis and assembly of synaptic cleft protein s-laminin by cultured cells. JBiolChem. 267:2014-2022.

82. Grillner S, Buchanan JT, Wallen P, Brodin L (1998) Neural control of locomotion in lower vertebrates: From behavior to ionic mechanisms. In: Neural Control of Rhythmic Movement in Vertebrates (Cohen AH, Rossignol S, Grillner S eds), pp 1-40. New York: John Wiley and Sons.

83. Grillner S, Shik ML (1973) On the descending control of the lumbosacral spinal cord ftom the "mesencephalic locomotor region". Acta Physiol.Scand. 87:320-333.

84. Grimby G, Broberg C, Krotkeiwska I, Krotkiewski M (1976) Muscle fiber composition in patients with traumatic cord lesion. Scand.J.Rehabil.Med. 8:37-42.

85. Grossman EJ, Roy RR, Talmadge RJ, Zhong H, Edgerton VR (1998) Effects of inactivity on myosin heavy chain composition and size of rat soleus fibers. Muscle Nerve 21:375-389.

86. Gruner JA, Altman J (1980) Swimming in the rat: Analysis of locomotor performance in comparison to stepping. Exp.Brain Res. 40:374-382.

87. Guiliani CA, Smith JL (1987) Stepping behaviors in chronic spinal cats with one hindlimb deafferented. J.Neurosci. 7:2537-2546.

88. Gundersen K (1985) Early effects of denervation on isometric and isotonic contractile properties of rat skeletal muscle. Acta Physiol.Scand. 124:549- 555.

179 89. Hao J-X, Xu X-J, Aldskogius H, Seiger Â, Wiesenfeld-Hallin Z (1991) Allodynia-like effects in rat after ischaemic spinal cord injury photochemically induced by laser irradiation. Pain 45:175-185.

90. Harkema S, Hurley SL, Patel UK, Requejo PS, Dobkin BH, Edgerton VR (1997) Human lumbosacral spinal cord interprets loading during stepping. J.Neurophysiol. 77:797-811.

91. Harris-Warrick RM (1988) Chemical modulation of central pattern generators. In: Neural Control of Rhythmic Movements in Vertebrates (Cohen AH, Rossignol S, Grillner S eds), pp 285-331. New York: John Wiley and Sons.

92. Harris R, Little JW, Goldstein B (1994) Spared descending pathways mediate locomotor recovery after subtotal spinal cord injury. Neurosci.Lett. 180:37-40.

93. Helgren ME, Goldberger ME (1993) The recovery of postural reflexes and locomotion following low thoracic hemisection in adult cats involves compensation by undamaged primary afferent pathways. Exp Neurol. 123:17-34.

94. Hennig R, Lomo T (1987) Effects of chronic stimulation on the size and speed of long-term denervated and innervated rat fast and slow skeletal muscles. Acta Physiol.Scand. 130:115-131.

95. Herbert ME, Roy RR, Edgerton VR (1988) Influence of one-week hindlimb suspension and intermittent high load exercise on rat muscles. ExpNeurol. 102:190-198.

96. Hesse S, Bertelt C, Jahnke MT, Schafftin A, Baake P, Malezic M, Mauritz KH (1995) Treadmill training with partial body weight support compared with physiotherapy in nonambulatory hemiparetic patients. Stroke 26:976-981.

97. Hickson RC, Hammons GT, HoUoszy JO (1979) Development and regression of exercise-induced cardiac hypertrophy. Am J Physiol. 236:H268-H272

1 8 0 98. Hochman S, McCrea DA (1994a) Effects o f chronic spinaiization on ankle extensor motoneurons. I. Composite monosynaptic la EPSPs in four motoneuron pools. J.Neurophysiol. 71:1452-1467.

99. Hochman S, McCrea DA (1994b) Effects of chronic spinaiization on ankle extensor motoneurons. II. Motoneuron electrical properties. J.Neurophysiol. 71:1468-1479.

100. Hodgson JA Bodine-Fowler S, Roy RR, de Leon R, de Guzman CP, Koslovskaya I, Sirota M, Edgerton VR (1991) Changes in recruitment of rhesus soleus and gastrocnemius muscles following a 14 day spaceflight. Physiologist 34:8102-8103

101. Hodgson JA Roy RR, de Leon R, Dobkin B, Edgerton VR (1994) Can mammalian spinal cord learn a motor task? Med.Sci.Sports Exercise 26:1491-1497.

102. Hughes 8M, Blau HM (1992) Muscle fiber pattern is independent of cell lineage in postnatal rodent development. Cell 68:659-671.

103. Hughes 8M, Cho M, Karsch-Mizrachi I, Travis M, Silberstein L, Leinwand LA, Blau H (1993) Three slow myosin heavy chains sequentially expressed in developing mammalian skeletal muscle. Dev Biol. 158:183-189.

104. Ishihara A, Oishi Y, Roy RR, Edgerton VR (1997) Influence of two weeks of non-weight bearing on rat soleus motoneurons and muscle fibers. Aviat.8pace Environ.Med. 68:421-425.

105. Jacobs BL, Fomal CA (1997) Serotonin and motor activity. Curr.Opin Neurobiol. 7:820-825.

106. Jasmin BJ, Gardiner PF (1987) Patterns of EMG activity of rat plantaris muscle during swimming and other locomotor activities. J Appl Physiol. 63:713- 718.

107. Jasmin BJ, Lavoie P-A, Gardiner PF (1988) Fast axonal transport of labelled proteins in motoneurons of exercise-trained rats. Am J Physiol. 255:C731-C736

181 108. Jânkâlâ H, Haqola VP, Petersen NE, Hârkônen M (1997) Myosin heavy chain mRNA transform to faster isoforms in immobilized skeletal muscle; A quantitative PCR study. J.ApplJ*hysiol. 82:977-982.

109. Jiang B, Roy RR, Edgerton VR (1990) Expression of a fast fiber en^rme profile in the cat soleus after spinaiization. Muscle Nerve 13:1037-1049.

110. Jiang W, Drew T (1996) Effects of bilateral lesions of the dorsolateral funiculi and dorsal columns at the level of the low thoracic spinal cord on the control of locomotion in the adult cat. I. Treadmill walking. J Neurophysiol. 76:849-866.

111. Jones TA, Schallert T (1994) Use-dependent growth of pyramidal neurons after neocortical damage. J.Neurosci. 14:2140-2152.

112. Kanda K, Hashizume K (1998) Effects of long-term physical exercise on age- related changes of spinal motoneurons and peripheral nerves in rats. Neurosci.Res. 31:69-75.

113. Kandarian S, O'Brien S, Thomas K, Schulte L, Navarro J (1992) Regulation of skeletal muscle dihydropyridine receptor gene expression by biomechanical unloading. J.AppLPhysiol. 72:2510-2514.

114. Kandarian SC, Boushel RC, Schulte LM (1991) Elevated interstitial fluid volume in rat soleus muscles by hindlimb unweighting. J Appl Physiol. 71:910- 914.

115. Kean CJ, Lewis DM, McGarrick JD (1974) Dynamic properties of denervated fast and slow twitch muscle of the cat. J.Physiol. 237:101-113.

116. Kerasidis H, Wrathall JR, Gale K (1987) Behavioral assessment o f functional deficit in rats with contusive spinal cord injury. J.Neurosci.Methods 20:167-189.

117. Kleim JA, Kapil VIJ, Ballard DH, Greenough WT (1997) Learning-dependent synaptic modifications in the cerebellar cortex of adult rat persist for at least 4 weeks. J.Neurosci. 17:717-721.

182 118. Kleim JA, Lussnig E, Schwarz ER, Comery TA, Greenough WT (1996) Synaptogenesis and FOS expression in the motor cortex of the adult rat after learning motor skill. J.Neurosci. 16:4529-4535.

119. Kozlowski DA, James DC, Schallert T (1996) Use-dependent exaggeration of neuronal injury after unilateral sensorimotor cortex lesions. J.Neurosci. 16:4776-4786.

120. Krenz NR, Weaver LG (1998) Sprouting of primary afferent fibers after spinal cord transection in the rat. Neuroscience 85:443-458.

121. Kunkel-Bagden E, Dai HN, Bregman BS (1992) Recovery of fimction after spinal cord hemisection in newborn and adult rats: Differential effects on reflex and locomotor function. Exp Neurol. 116:40-51.

122. Laemmli UK (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680-685.

123. Leem JW, Willis WD, Chung JM (1993) Cutaneous sensory receptors in the rat foot. J.Neurophysiol. 69:1684-1699.

124. Lieber RL (1992) Skeletal muscle structure and function: Implications for rehabilitative and sports medicine. Baltimore: Williams and Wilkins Pub.

125. Lieber RL, Frieden JO, Hargens AR, Ferringa ER (1986a) Long-term effects of spinal cord transection on fast and slow rat skeletal muscle. H. Morphometric properties. Exp .Neurol. 91:435-448.

126. Lieber RL, Johannson CB, Vahlsing HL, Hargens AR, Ferringa ER (1986b) Long-term effects of spinal cord transection on fast and slow rat skeletal muscle. I. Contractile properties. Exp Neurol. 91:423-434.

127. Lin Q, Peng YB, Willis WD (1996a) Inhibition of primate spinothalamic tract neurons by spinal glycine and GABA is reduced during central sensitization. J.Physiol. 76:1005-1014.

183 128. Lin Q, Peng YB, Willis WD (1996b) Role of GABA receptor subtypes in inhibition o f primate spinothalamic tract neurons: Difference between spinal and periaqueductal gray matter. J.Neurophysiol. 75:109-123.

129. Lindsey AE, LoVerso RL, Tovar CA, Beattie MS, Bresnahan JC (1999) Rats with contusion spinal cord injury are hypersensitive to tactile and cold stimuli. Manuscript in preparation.

130. Little JW, Harris R, Sohlberg RC (1988) Locomotor recovery following subtotal spinal cord lesions in a rat model. Neurosci Lett. 87:189-194.

131. Lotta S, Scelsi R, Alfonsi E, Saitta A, Nicolotti D, Epifani P, Carraro U (1991) Morphometric and neiuophysiological analysis of skeletal muscle in paraplegic patients with traumatic cord lesion. Paraplegia 29:247-252.

132. Lovely RG, Gregor RJ, Roy RR, Edgerton VR (1986) Effects of training on the recovery of full-weight bearing stepping in the adult spinal cat. Exp Neurol. 92:421-435.

133. Lovely RG, Gregor RJ, Roy RR, Edgerton VR (1990) Weight-bearing hindlimb stepping in treadmill exercised adult spinal cats. Brain Res. 514:206-218.

134. Lopez SM, Pérez-Pérez M, Marquez JM, Naves FJ, Represa J, Vega JA (1998) p75 and TrkA neurotrophin receptors in human skin after spinal cord and peripheral nerve injury, with special references to sensory corpuscles. AnatRec. 251:371-383.

135. McComas AJ (1996) Skeletal Muscle Form and Function. Champaign,IL: Human Kinetics.

136. Meeusen R, Smolder J, Sarre S, DeMeirlier K, Keizer H, Semeels M, Ebinger G, Michotte Y (1997) Endurance training effects on neurotransmitter release in rat striatum: An in vivo microdialysis study. Acta Physiol.Scand. 159:335-341.

137. Michel RP, Cruz-Orive LM (1988) Application of the Cavalieri principle and vertical sections method to lung: Estimation of volume and pleural surface area. J.Microsc. 150:117-136.

184 138. Midrio M, Betto DD, Betto R, Noventa D, Antico F (1988) Cordotomy- denervation interaction on contractile and myofibrillar properties of fast and slow muscles in the rat. Exp .Neurol. 100:216-236.

139. Muir GD, Whishaw IQ (1999) Complete locomotor recovery following corticospinal tract lesions: Measurement of ground reaction forces during overground locomotion in rats. Behav.Brain Res. 103:45-53.

140. Murray M, Goldberger ME (1974) Restitution of function and collateral sprouting in the cat spinal cord: The partially hemisected animal. J.Comp.Neurol. 158:19-36.

141. Na HS, Leem JW, Chung JM (1993) Abnormalities of mechanoreceptors in a rat model of neuropathic pain: Possible involvement in mediating mechanical allodynia. J Neurophysiol. 70:522-528.

142. Nair KG, Cutilleta AF, Zak R, Koide T, Rabinowitz M (1968) Biochemical correlates of cardiac hypertrophy. 1. Experimental model; changes in heart weight, RNA content, and nuclear RNA polymerase activity. Circ Res. 23:451-462.

143. Naito A, Shimizu Y, Handa Y (1990) Analyses of treadmill locomotion in adult spinal dogs. NeuroscLRes. 8:281-290.

144. Neeper SA, Gômez-Pinilla F, Choi J, Cotman C (1995) Exercise and brain neurotrophins. Nature 373:109-109.

145. Noble LJ, Wrathall JR (1989) Correlative analyses of lesion development and functional status after graded spinal cord contusion injuries in the rat. ExpNeurol. 103:34-40.

146. Nudo RJ, Milliken GW, Jenkins WM, Merzenich MM (1996) Use-dependent alterations of movement representations in primary motor cortex of adult squirrel monkeys. J.Neurosci. 16:785-807.

147. Nudo RJ, Plautz EJ, Milliken GW (1997) Adaptive plasticity in primate motor cortex as a consequence of behavioral experience and neuronal injury. Semin.Neurosci. 9:13-23.

185 148. Nudo RJ, Wise BM, SiFuentes F, Milliken GW (1996) Neural substrates for the effects of rehabilitative training on motor recovery after ischemic infarct. Science 272:1791-1794.

149. Orlovsky ON (1972) Activity of vestibulospinal neurons during locomotion. Brain Res. 46:85-98.

150. Pagliari R, Peyron L (1995) Norepinephrine release in the rat frontal cortex under treadmill exercise: A study with microdialysis. J Appl Physiol. 78:2121- 2130.

151. Pearson KG, Rossignol S (1991) Fictive motor patterns in chronic spinal cats. J.Neurophysiol. 66:1874-1886.

152. Pierotti DJ, Roy RR, Flores V, Edgerton VR (1990) Influence of 7 days of hindlimb suspension and intermittent weight support on rat muscle mechanical properties. AviatSpace Environ.Med. 61:205-210.

153. Poliakov IV, Louri O, Edzherton VR, Krasnov IB (1995) Histochemistry and morphology of the anterior horns of spinal cords in rats after 9 day space flight Aviakosm.Ekolog.Med. 29:30-32.

154. Recktenwald MR, Hodgson JA, Roy RR, Riazanski S, McCall GE, Kozlovskaya I, Washburn DA, Fanton JW, Edgerton VR (1999) Effects of spaceflight on rhesus quadrupedal locomotion after return to IG. J.Neurophysiol. 81:2451-2463.

155. Reiser PJ, Moss RL, Giulian GG, Greaser ML (1985a) Shortening velocity and myosin heavy chains of developing rabbit muscle fibers. J.Biol.Chem. 260:14403-14405.

156. Reiser PJ, Moss RL, Giulian GG, Greaser ML (1985b) Shortening velocity in single fibers from adult rabbit soleus muscles is correlated with myosin heavy chain composition. J.Biol.Chem. 260:9077-9080.

157. Robichaud JA, Agostinucci J (1996) Airsplint pressure effects on soleus muscle alpha motomeuron reflex excitability in subjects with spinal cord injury. Arch.Phys.Med.RehabU. 77:778-782.

186 158. Robinson GA, Goldberger ME (1985) Interfering with inhibition may improve motor function. Brain Res. 346:400-403.

159. Rossignol S, Chau C, Brustein E, Belanger M, Barbeau H, Drew T (1996) Locomotor capacities after complete and partial lesions of the spinal cord. Acta NeurobioLExp. 56:449-463.

160. Rossignol S, Limd JP, Drew T (1988) The role of sensory inputs in regulating patterns of rhythmic movement in higher vertebrates: A comparison between locomotion, respiration, and mastication. In: Neural Control of Rhythmic Movements in Vertebrates (Cohen AH, Rossignol S, Grillner S eds), pp 201-283. New York: John Wiley and Sons.

161. Round JM, Barr FMD, Moffat B, Jones DA (1993) Fibre areas and histochemical fibre types in the quadriceps muscle of paraplegic subjects. J Neurol.Sci. 116:207-211.

162. Roy RR, Baldwin KM, Edgerton VR (1991 ) Plasticity of skeletal muscle: Effects of neuromuscular activity. Exerc.Sport Sci.Rev. 19:269-312.

163. Roy RR, Hirota WK, Kuehl M, Edgerton VR (1985) Recruitment patterns in the rat hindlimb muscle during swimming. Brain Res. 337:175-178.

164. Roy RR, Hutchinson DL, Pierotti DJ, Hodgson JA, Edgerton VR (1991) EMG patterns of rat ankle extensors and flexors during treadmill locomotion and swimming. J AppLPhysiol. 70:2522-2529.

165. Roy RR, Pierotti DJ, Baldwin KM, Zhong H, Hodgson JA, Edgerton VR (1998) Cyclical passive stretch influences the mechanical properties of the inactive cat soleus. Exp.Physiol. 83:377-385.

166. Roy RR, Talmadge RJ, Hodgson JA, Oishi Y, Baldwin KM, Edgerton VR (1999) Differential response of fast hindlimb extensor and flexor muscles to exercise in adult spinalized cats. Muscle Nerve 22:230-241.

167. Roy RR, Talmadge RJ, Hodgson JA, Zhong H, Baldwin KM, Edgerton VR (1998) Training effects on soleus of cats spinal cord transected (T12-13) as adults. Muscle Nerve 21:63-71.

187 168. Salviati G, Betto R, Ceoldo S, Tegazzin V. Puppa AD (1989) Caffeine sensitivity of sarcoplasmic reticulum of fast and slow fibers firom normal and malignant hyperthermia human muscle. Muscle Nerve 12:365-370.

169. Sanes JR, Engvall E, Butkowski R, Hunter DD (1990) Molecular heterogeneity of basal laminae: Isoforms of laminin and collagen iv at the neuromuscular junction. J.Cell Biol. 111:1685-1699.

170. Schiaffino S, Ausoni S, Gorza L, Saggin I. Gundersen K, Lomo T (1988) Myosin heavy chain isoform and velocity of shortening of type 2 skeletal muscles fibers. Acta PhysioLScand. 134:575-576.

171. Schwab ME, Bartholdi D (1996) Degeneration and regeneration of axons in the leisoned spinal cord. Physiol.Rev. 76:319-370.

172. Segal SS, Faulkner JA (1985) Temperature-dependent physiological stability of rat skeletal muscle in vitro. Am.J.Physiol. 248:C265-C270

173. Shields RK (1995) Fatigability, relaxation properties, and electromyographic responses of the human paralyzed soleus muscle. J.Neurophysiol. 73:2195-2206.

174. Shiverick KT, Thomas LT, Alpert NR (1975) Purification of cardiac myosin: Application to hypertrophied myocardium. Biochim.Biophys.Acta 393:124-133.

175. Siddall P, Xu CL, Cousins M (1995) Allodynia following traumatic spinal cord injury in the rat. NeuroReport 6:1241-1244.

176. Sieck G, Han YS, Prakash TS, Jones KA ( 1998) Crossbridge cycling kinetics, actomyosin ATPase activity and myosin heavy chain isoforms in skeletal and smooth respiratory muscles. Comp.Biochem.Physiol.B.Biochem.Mol.Biol. 119:435-450.

177. Sieck GC (1994) Physiological effects of diaphragm muscle denervation and disuse. Clin.Chest Med. 15:641-659.

188 178. Sinnamon H (1993) Preoptic and hypothalamic neurons and the initiation of locomotion in the anesthetized rat. Prog.Neurobiol. 41:323-344.

179. Skinner RS, Houle JD, Reese NB, Berry CL, Garcia-Rill E (1996) Effects of exercise and fetal spinal cord implants on the H-reflex in chronically spinalized adult rats. Brain Res. 729:127-131.

180. Smith JL, Edgerton VR, Betts B, Collates TC ( 1977) EMG of slow and fast ankle extensors of cat during posture, locomotion, and jumping. J.Neurophysiol. 40:503-513.

181. Smith JL, Smith LA, Zemicke RF, Hoy M ( 1982) Locomotion in exercised and nonexercised cats cordotomized at two or twelve weeks of age. Exp.Neurol. 76:393-413.

182. SPA I (1997) The stereologer handbook. Introduction to unbiased stereology and stereologer users guide. Alexandria, VA: Systems Planning and Analysis, Inc.

183. Spector SA (1985) Trophic effects on the contractile and histochemical properties of rat soleus muscle. J.Neurosci. 5:2189-2196.

184. Stüwill EW, Sahgal V (1977) Histochemical and morphologic changes in skeletal muscle following cervical cord injury: A study of upper and lower motor neuron lesions. Arch.Phys.Med.Rehabil. 58:201-206.

185. Stokes BT, Homer PJ (1996) Spinal cord injury modeling and outcome assesment. In: Neurotrauma (Narayan RK, Wilberger JE, Povlishock J eds), pp 1395-1403. New York: McGraw-Hill Co. Inc.

186. Stokes BT, Homer PJ, Akino M (1995) Spinal cord injury modeling and hmctional assessment. In: Central Nervous System Trauma: Research Techniques (Ohnishi ST, Ohnishi T eds), pp 281-295. New York: CRC Press.

187. Stokes BT, Noyes DH, Behrmarm DL (1992) An electromechanical spinal injury technique with dynamic sensitivity. J.Neurotrauma 9:187-195.

189 188. Talmadge RJ, Roy RR (1993) Electrophoretic separation of rat skeletal muscle myosin heavy chain isoforms. J.AppLPhysiol. 75:2337-2340.

189. Talmadge RJ, Roy RR, Bodine-Fowler SC, Pierotti DJ, Edgerton VR (1995a) Adaptations in myosin heavy chain profile in chronically unloaded muscles. BAM 5:117-137.

190. Talmadge RJ, Roy RR, Edgerton VR ( 1995b)Prominence of myosin heavy chain hybrid fibers in soleus muscle of spinal cord-transected rats. JAppLPhysiol. 78:1256-1265.

191. Tamopolsky MA (1994) Caffeine and endurance performance. Sports Med. 18:109-125.

192. Taub E, Miller NE, Novack TA, Cook EW, III, Fleming WC, Nepomuceno CS, Coimell JS, Crago JE (1993) Technique to improve chronic motor deficit after stroke. Arch.Phys.Med.RehabiI. 74:347-354.

193. ten Cate J (1960) Locomotor movements in the spinal pigeon. Exp.Biol. 37:609- 613.

194. ten Cate J (1962) Innervation of locomotor movements by the lumbrosacral cord in birds and mammals. J.Exp.BioI. 39:239-242.

195. ten Cate J (1964) Locomotor movements of the hind limbs in rabbits after isolation of the lumbosacral cord. J.Exp.BioI. 41:359-362.

196. Tessler A, Glazer E, Artymyshyn R, Murray M,Goldberger ME (1980) Recovery of Substance P in the cat spinal cord afterunilateral lumbosacral deafferentation. Brain Res. 191:459-470.

197. Tessler A, Himes BT, Artymyshyn R, Murray M, Goldberger ME (1981) Spinal neurons mediate return of Substance P following deafferentation of cat spinal cord. Brain Res. 230:263-281.

198. Tessler A, Himes BT, Soper K, Murray M, Goldberger ME, Reichlin S (1984) Recovery of Substance P but not Somatostatin in cat spinal cord after

190 unilateral lumbosacral dorsal rhizotomy: A quantitative study. Brain Res. 305:95-102.

199. Thesleff S (1974) Physiological effects of denervation of muscle. Ann.N.Y.Acad.Sci. 228:89-104.

200. Thomason DB, Biggs RB, Booth FW (1989) Protein metabolism and P-myosin heavy-chain mRNA in unweighted soleus muscle. Am.J.Physiol. 257:R300-R305

201. Thomason DB, Booth FW (1990) Atrophy of the soleus muscle by hindlimb unweighting. J.AppLPhysiol. 68:1-12.

202. Thomason DB, Herrick RE, Baldwin KM ( 1987a) Activity influences on soleus muscle myosin during rodent hindlimb suspension. J Appl Physiol. 63:138-144.

203. Thomason DB, Herrick RE, Surdyka D, Baldwin KM (1987b) Time course of soleus muscle myosin expression during hindlimb suspension and recovery. J.Appl.Physiol. 63:130-137.

204. Thompson FJ, Parmer R, Reier PJ (1998) Alteration in rate modulation of reflexes to lumbar motoneurons after midthoracic spinal cord injury in the rat. I. Contusion injury. J.Neurotrauma 15:495-508.

205. Thompson FJ, Reier PJ, Lucas CC, Parmer R (1992) Altered patterns of reflex excitability subsequent to contusion injury of the rat spinal cord. J.Neurophysiol. 68:1473-1486.

206. Torebjork HE, Lundberg LER, LaMotte RH (1992) Central changes in processing of mechanoreceptive input in capsaicin-induced secondary hyperalgesia in humans. J.Physiol. 448:765-780.

207. Trimble M, Kukulka CG, Behrman AL (1998) The effect of treadmill gait training on low frequency depression of the soleus H reflex: Comparison of a spinal cord injured man to normal subjects. Neurosci Lett. 246:186-188.

191 208. Veasey SC, Fomal CA, Metzlar CW, Jacobs BL (1995) Response of serotonergic caudal raphe neurons in relation to specific motor activities in fieeiy moving cats. J.Neurosci. 15:5346-5359.

209. Voss DE, lonta MK, Myers BJ (1985) Proprioceptive neuromuscular facilitation patterns and techniques. Philadelphia: Harper and Row.

210. Wakatsuki T, Ohira Y, Nakamura K, Asakura T, Ohno H, Yamamoto M (1995) Changes of contractile properties of extensor digitorum longus in response to creatine-analogue administration and/or hindlimb suspension in rats. Jpn. J.Physiol. 45:979-989.

211. Walmsley B, Hodgson JA, Burke RE (1978) Forces produced by medial gastrocnemius and soleus muscles during locomotion in fireely moving cats. J.Neurophysiol. 41:1203-1216.

212. Webster C, Silberstein L, Hays AP, Blau H (1988) Fast muscle fibers are preferentially affected in Duchenne muscular dystrophy. Cell 52:503-513.

213. Wemig A, Muller S (1992) Laufband locomotion with body weight support improved walking in persons with severe spinal cord injuries. Paraplegia 30:229-238.

214. Wemig A, Muller S, Nanassy A, Cagol E (1995) Laufband therapy based on rules of spinal locomotion' is effective in spinal cord injured persons. Eur J Neurosci. 7:823-829.

215. Wickelgren I (1998) Teaching the spinal cord to walk. Science 279:319-321.

216. Wolf SL, Lecraw DE, Barton LA, Jann BB ( 1989) Forced use o f hemiplegic upper extremities to reverse the effect of learned nonuse among chronic stroke and head-injured patients. Exp.Neiurol. 104:125-132.

217. Wrathall JR, Li W, Hudson LD (1998) Myelin gene expression after experimental contusive spinal cord injury. J.Neurosci. 18:8780-8793.

218. Xu X-J, Hao J-X, Sieger A, Hughes J, Hdkfelt T, Wiesenfeld-Hallin Z (1994) Chronic pain-related behaviors in spinally injured rats: Evidence for 192 functional alterations of the endogenous cholecystokinin and opioid systems. Pain 56:271-277.

219. Yezierski RP (1996) Pain following spinal cord injury: The clinical problem and experimental studies. Pain 68:185-194.

220. Yezierski RP, Liu S, Ruenes GL, Kajander KJ, Brewer KL (1998) Excitotoxic spinal cord injury: Behavioral and morphological characteristics of a central pain model. Pain 75:141-155.

221. Yezierski RP, Park SH (1993) The mechanosensitivity of spinal sensory neurons following intraspinal injections of quisqualic acid in the rat. NeurosciLett. 157:115-119.

193