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 be from any type of computer printer.
The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, 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 be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion.
Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand corner and continuing from left to right in equal sections with small overlaps. Each original is also photographed in one exposure and is included in reduced form at the back of the book.
Photographs included in the original manuscript have been reproduced xerographically in this copy. Higher quality 6" x 9" black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order.
A Bell & Howell Information Company 300 North Zeeb Road. Ann Arbor. Ml 48106-1346 USA 313/761-4700 800/521-0600
Order Number 9517043
Serotonergic control of gastrointestinal function
LePard, Kathy Jean, Ph.D.
The Ohio State University, 1994
UMI 300 N. ZeebRd. Ann Arbor, MI 48106
SEROTONERGIC CONTROL OF GASTROINTESTINAL FUNCTION
DISSERTATION
Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy in the Graduate School of The Ohio State University
By Kathy Jean LePard, B.S., M.S. * * * * * The Ohio State University 1994
Dissertation Committee: Richard C. Rogers, Ph.D. Jackie D. Wood, Ph.D. Robert L. Stephens Jr., Ph.D. Approved by
Advisor Department of Physiology To my Husband.
ii ACKNOWLEDGMENTS
I would like to express my sincere gratitude to the faculty of the Natural Science Division at Huntington College: my undergraduate advisor, Prof. J. Howald, along with Drs. W. Bordeaux, F. Jones, R. Brown, W. Weatherbee and R. Hale, for encouraging my pursuit of science, for academically preparing me for graduate school, but most importantly, for their neverending support and guidance throughout my undergraduate and graduate career. Thanks to my graduate advisor, Dr. Robert Stephens Jr, for his guidance, support and insight during my graduate research. Also, to the other members of my advisory committee, Drs. Helen Cooke, Richard Rogers, and Jackie Wood, for their helpful comments and suggestions. Thanks to the support staff of the Department of Physiology: C. Hagaar, D. Ranft, B. Spink, C. Hairston, R. Lynch and D. Vanek. Also to faculty which have graciously loaned equipment, drugs or provided knowledge: Drs. R. Rogers, J. Rail, P. Ward, C. Heesch, H. Cooke and L. Fisher. Special thanks to our laboratory technician J. Mohammed for her invaluable assistance and National Institutes of Health for funding (NIH DK 42880).
iii To my parents, Merle and Maxine Nice, I would like to express my indebtedness for their support, love and encouragement throughout my academic development. Finally, to my husband, Brian. Words cannot express my humble appreciation of your constant faith in me and your unwavering love to me despite the demands of my graduate career.
iv VITA
March 9, 1967 Born - Logansport, Indiana 1988-1989 Laboratory Assistant, Department of Chemistry Huntington College, Huntington, Indiana Summer 1988 Buckingham Research Fellow Department of Chemistry Miami University, Oxford, Ohio 1985-1989 B.S.S., Huntington College, Huntington, Indiana 1989-1992 M.S.S., College of Medicine, The Ohio State University Columbus, Ohio 1992-1994 Graduate Research Associate The Ohio State University Columbus, Ohio
PUBLICATIONS LePard, K.J. and R.L. Stephens Jr. Serotonin Inhibits Gastric Acid Secretion through a 5-Hydroxytryptamine^like Receptor in the Rat. Journal of Pharmacology and Experimental Therapeutics. 1994. 270:1-6. LePard, K.J., S. Gidener and R.L. Stephens Jr. Serotonin Inhibits, but 8-OH-DPAT Stimulates GAstric Acid Secretion through Vagal-Independent Mechanisms. Society for Neuroscience Abstracts. 1994. 20:1376. Stephens JR, R.L., K.J. LePard, J.R. Mohammed and P.E. Ward. Intracisternal Neutral Endopeptidase-24-11 Inhibitors Produce Inhibition in Gastric Acid Output: Independence of Opiate, Bombesin, or Neurotensin-Mediated Mechanisms. Regulatory Peptides. 1993. 46:549-555.
v LePard, K.J. and R.L. Stephens Jr. Characterization of Serotonergic Mechanisms Producing Inhibition of Gastric Acid Secretion. 1993. Gastroenterology. 104:A132. LePard, K.J. and R.L. Stephens Jr. Gastrin-Dependent Enterochromaffin-Like Cell Hyperplasia: Effects on Basal and Stimulated GAstric Acid Secretion. Gastroenterology. 1993. 104:A631. LePard, K.J. and R.L. Stephens Jr. Vagally-Mediated Serotonin (5HT) Release Inhibits Gastric Acid Secretion Via Receptors of the 5HT- Family. Society for Neuroscience Abstracts. 1993. 19:962. Shockley,R., K.J. LePard, and R.L. Stephens Jr. Fluoxetine Pretreatment Potentiates Intracisternal TRH Analogue- Stimulated Gastric Acid Secretion in Rats. 1992. 38:121-128. LePard, K.J. and R.L. Stephens Jr. Serotonin (5-HT) Inhibits Gastric Acid Secretion by Nonluminal, Vagal Independent Mechanisms. Society for Neuroscience Abstracts. 1992. 18:1529. McCann, M.J., K.J. Nice-LePard and R.C. Rogers. Dorsal Medullary Injection of Atrial Natriuretic Factor (ANF) Excites Vagal Efferents and Inhibits Gastric Motility. Brain Research. 1991. 549:247-252. FIELDS OF STUDY Major Field: Physiology Studies in: Cardiovascular Physiology Dr. G. Billman Dr. C. Heesch Endocrinology Dr. J. Grossie Gastrointestinal Physiology Dr. J. Rail Dr. R. Stephens Membrane Physiology Dr. J. Grossie Dr. B. Biagi Muscle Physiology Dr. J. Grossie Neurophysiology Dr. B. Stokes Dr. R. Rogers Physiological Laboratory Dr. J. Grossie Renal Physiology Dr. B. Biagi Respiratory Physiology Dr. H. Weiss Biochemistry Dr. K. Richardson Neuroscience Dr. G. Martin Electronic Instrumentation Dr. J.-P. Dujardin Drug Receptor Theory Dr. A. Burkman Statistics Dr. J. Powers
vi TABLE 07 CONTENTS
DEDICATION ii ACKNOWLEDGMENTS iii
VITA . V TABLE OF CONTENTS vii LIST OF FIGURES xiv LIST OF TABLES xvii
CHAPTER I. INTRODUCTION I. Historical Aspects 1 II. Regional Distribution of Serotonin in Mammalian Systems 2 A. Central Nervous System 2 B. Gastrointestinal Tract 2 III. Synthesis and Degradation 3 A. Synthesis 3 B. Degradation 4 IV. Localization of Serotonin Receptors 4 A. 5-HT1a Receptors 7 B. 5-HT1p Receptors 7 C. 5-HT2 Receptors 8 D. 5-HT3 Receptors 8 E. 5-HT4 Receptors 8 V. Overview 9 VI. Mechanism of Release of 5-HT from Gastrointestinal Stores: In Vivo Studies 10 A. Introduction 10 B. Release into the Lumen 10 1. Activation of Chemoreceptors 10 2. Central Vagal Control 11
vii 3. Mucosal Mast Cell Degranulation 12 4. Summary (TABLE 2) 13 C. Release into the Portal Circulation 15 1. Activation of Chemoreceptors 15 2. Central Vagal Control 15 3. Splanchnic Nervous Control 17 4. Endocrine Control 17 5. Summary (TABLE 3) 18 D. Release from Enterochromaffin Cells 21 1. Activation of Chemoreceptors 21 2. Central Vagal Control 21 3. Sympathetic Nervous System 22 4. Summary 22 VII. Mechanisms of Release of Serotonin from Gastrointestinal Stores: In Vitro Studies .... 22 A. Introduction 22 B. Release to Mucosal Side 23 1. Nutrients 23 2. Acid in the Duodenal Lumen 23 3. Cholinergic Receptors 24 4. Adrenergic Receptors 25 5. Enteric Neurons 26 6. Intraluminal Pressure 26 7. Summary 26 C. Release to Serosal Side 27 1. Nutrients 27 2. Cholinergic Receptors . 27 3. Adrenergic Receptors 29 4. Enteric Neurons 29 5. Second Messenger Systems 30 a. Cyclic 3 *,5 '-Adenosine Monophosphate (CAMP) 30 b. Cyclic 3',5'-Guanine Monophosphate (cGMP) 30 c. Calcium 30 6. Enzymatic Activity 31 7. Serotonin Receptors 31 8. Summary 32 D. Release from EC Cells or Tissue Segments ... 33 1. Cholinergic Receptors 33 2. Adrenergic Receptors 34 3. Enteric Neurons 34 4. Second Messengers: Calcium 35 5. Summary 35 VIII. Summary: Release of Serotonin 35 A. Release into the Lumen 35 1. Activation of Chemoreceptors 35 2. Central Vagal Control 36 B. Release into the Portal Circulation 36 1. Activation of Chemoreceptors 36
viii 2. Central Vagal Control 36 3. Enteric Neurons 37 IX. Effect of Serotonin on Gastrointestinal Function . 37 A. Exogenous Serotonin 37 1. Gastric Acid Secretion 37 a. In Vivo Studies 37 b. In Vitro Studies 39 2. Gastrointestinal Motility 41 a. In Vivo Studies 41 b. In Vitro Studies 43 3. Gastric Emptying 45 4. Gastrointestinal Blood Flow 45 5. Gastric and Duodenal Lesions 46 6. Vagal Activity 47 a. In Vivo Studies 47 In Vitro Studies 48 B. Endogenous Serotonin 49 1. Gastric Acid Secretion 50 a. Endogenous 5-HT in the Gastric Lumen 50 b. Endogenous 5-HT in the Portal Circulation 50 c. Depletion of Endogenous 5-HT .... 50 2. Gastrointestinal Motility 51 a. Endogenous 5-HT in Mucosal Mast Cells 51 b. Endogenous 5-HT and Enteric Neurons . 52 1) In Vivo Studies 52 2) In Vitro Studies 52 c. Depletion of Endogenous 5-HT .... 52 3. Gastric Emptying 53 a. Enteric Neurons 53 4. Vagal Activity 53 5. Absorption of Nutrients 53 a. Depletion of Endogenous 5-HT .... 53 X. Conclusion 54
CHAPTER II. Serotonin Receptor Subtype Mediating Inhibition of Gastric Acid Secretion 57 I. Overview 57 II. Hypothesis: Intravenous Serotonin Inhibits Acid Secretion by Acting through a Serotonin Receptor . 57 III. Methods 58 A. Animals 58 B. Measurement of Gastric Acid Secretion 58 C. Cannulation of the Splenic Artery 58 D. Protocol for Antagonist Studies 61
ix E. Protocol for Agonist Studies 61 F. Drugs 61 6. Drug Treatments for Antagonist Studies .... 62 H.. Drug Treatments for Agonist Studies ...... 62 I. Statistics 62 IV. Results 63 A. Effect of Intravenous Serotonin on Acid Secretory Responses to Pentagastrin 63 B. Ability of Local Gastric Infusion of Serotonin Antagonists to Reverse Serotonin-Induced Attenuation of Acid Secretion 65 C. Effect of Serotonin Agonists on Pentagastrin- Stimulated Gastric Acid Secretion 69 V. DISCUSSION 72
CHAPTER III. Mechanism of Action of Intravenous Serotonin to Inhibit Gastric Secretory Function . 78 I. Overview 78 II. Hypothesis: Serotonin Inhibits Acid Secretion by Acting Through Enteric Neurons 79 III. Method 79 A. Animals 79 B. Measurement of Gastric Acid Secretion 79 C. Cannulation of Vessels 79 1. Splenic Artery 79 2. Portal Vein 80 D. Methods for Assay of Serotonin Content .... 83 1. Gastric Perfusates 83 2. Whole Blood 83 3. Electrochemical Detection of Serotonin . . 84 E. Basic Experimental Protocol 85 F. Drugs 86 G. Statistics ..... 87 IV. Results 87 A. Source of Serotonin Mediating Inhibition of Acid Secretion: Gastric Lumen vs. Portal Circulation 87 1. Gastric Lumen 87 a. Physiologic Levels of Serotonin in the Gastric Lumen after Vagal Stimulation 87 1) Methods 87 2) Results 88
x b. Effect of Exogenous Serotonin in Gastric Lumen on Acid Secretion . . 91 1) Methods 91 2) Results 93 Portal Circulation 95 a. Physiologic Levels of Endogenous Serotonin in the Portal Circulation after Vagal Stimulation 95 1) Concentration of 5-HT 95 a) Methods 95 b) Results 95 2) Flow Through Portal Vein . . . . 97 a) Methods 97 b) Results 99 3) Net Release of Serotonin .... 101 a) Methods 101 b) Results 101 b. Levels of Serotonin in the Portal Circulation after Intravenous Serotonin 104 1) Methods 104 2) Results 104 c. Effect of Intravenous Serotonin on Acid Secretion 107 1) Methods 107 2) Results 107 of the Autonomic Nervous System 107 Enteric Nervous System 107 a. Positive Control: RX77368 108 1) Methods 108 2) Results 108 b. Role of the Enteric Nervous System 111 1) Methods 111 2) Results 111 Parasympathetic Nervous System: Vagus Nerve 113 a. Methods 113 b. Results 113 Sympathetic Nervous System: Splanchnic Nerves 115 a. Role of the Splanchnic Nerves .... 115 1) Methods 115 2) Results 115 b. Tissue Concentrations of Norepinephrine 118 1) Methods 118 2) Results 119 of Prostaglandins 121 Methods 121 Results 121
xi D. Gastric Mucosal Blood Flow 123 1. Positive Control: Mean Arterial Pressure . 123 a. Methods 123 b. Results 124 2. Role of Gastric Mucosal Blood Flow in Mediating Attenuation of Acid Secretion by 5-HT 128 a. Methods 128 b. Results 128 E. Role of Gastric Mucosal Mast Cells 130 1. Methods 130 2. Results 130 F. Role of the Adrenal Gland 132 1. Methods 132 2. Results 132 V. Discussion 135
CHAPTER IV. Role of Endogenous Serotonin in Modulating Gastric Acid Secretory Function 142 I. Overview 142 II. Hypothesis: Endogenous serotonin exerts an inhibitory tone on vagally-stimulated gastric acid secretion 143 III. Methods 143 A. Animals 143 B. Measurement of Gastric Acid Secretion 143 C. Chronic Close Intra-arterial Cannulation of Splenic Artery 143 D. Ability of Serotonin Antagonists to Prevent Inhibition of Acid Secretion by Intravenous Serotonin in Rats with Chronically Indwelling Splenic Artery Cannulas 145 1. Protocol . 145 2. Statistics 145 E. Effect of Serotonin Antagonists on RX77368- Stimulated Gastric Acid Secretion 146 1. Protocol 146 2. Statistics 147 F. Drugs 147 IV. Results 147 A. Effects of Serotonergic Antagonists on Inhibition of Acid Secretion by Intravenous Serotonin in Rats with Chronic Indwelling Splenic Cannula 147
xii B. Effect of Serotonin Antagonists on RX77368- Stimulated Gastric Acid Secretion in Chronically Cannulated Rat 150 V. Discussion 155
CHAPTER V. Serotonin Inhibits/ But 8-OH DPAT Stimulates Gastric Acid Secretion Through a Vagal-Independent, Adrenal Mediated Mechanism 161 I. Overview 161 II. Hypothesis: Intravenous Serotonin and 8-OH-DPAT Produce Opposite Effects on Gastric Function by Acting Through Different Receptors 162 III. Methods 162 A. Animals 162 B. Measurement of Gastric Acid Secretion 162 C. Cannulation of the Splenic Artery 162 D. Surgical Procedures 162 1. Vagotomy 162 2. Celiac Ganglionectomy 163 3. Adrenalectomy 163 E. Protocols 163 1. Basic Protocol 163 2. Dose-Response Study 164 3. Antagonist Studies ..... 164 F. Drugs 164 G. Statistics 164 IV. Results 165 A. Comparison of the Effect of Systemic 5-HT and 8- OH-DPAT 165 1. Time Course 165 2. Dose-Response 167 B. Mechanism of Action of 8-OH-DPAT to Enhance Acid 169 1. Effect of vagotomy or celiac ganglionectomy on 8-OH-DPAT-induced stimulation of gastric acid secretion 169 2. Effect of 8-OH-DPAT in adrenalectomized rats 169 C. Receptor Subtype Mediating Enhancement in Acid Secretion by 8-OH-DPAT: Spiperone vs. Idaxozan . 171 V. Discussion 173
LIST OF REFERENCES 176
xiii LIST OF FIGURES FIGURE EfiGE 1: Anatomy of the celiac axis of the rat 60 2: Time course of the effect of intravenous 5-HT on pentagastrin-stimulated acid secretion 64 3: A. Effect of local gastric infusion of various 5-HT antagonists on pentagastrin-stimulated (24 /xmol/kg/hr, i.v.) gastric acid secretion 66 B. Ability of local gastric infusion of various 5-HT antagonists to reverse the 5-HT-induced (3.5 /xmol/kg, i.v.) attenuation of pentagastrin- stimulated gastric acid secretion 66 4: Effect of close i.a. gastric infusion of either vehicle, 5-HT (0.88 /xmol/kg) or various 5-HT agonists (0.88 or 2.6 jimol/kg) on pentagastrin- stimulated (24 /Limol/kg/hr, i.v.) gastric acid secretion 71 5: A. 25 gauge spinal needle. B. Schematic of the portal cannula used for withdrawal of portal blood samples 81 6: Anatomy of the portal venous system of the rat. . . 82 7: Time course of the effect of RX77368 on acid and 5-HT secretion into the gastric lumen 90 8: Schematic of the triple lumen cannula 92 9: Effect of perfusion of various concentrations of 5-HT through the gastric lumen on pentagastrin- stimulated acid secretion 94 10: Time course of the effect of RX77368 on concentration of 5-HT in the portal circulation. . 96 11: Time course of the effect of vehicle (VEH) or RX77368 (RX) on flow of blood through the portal vein 100
xiv 12: Time course of the effect of RX77368 on net release of 5-HT into the portal circulation 103 13: Comparison of net release of 5-HT into the portal circulation after vehicle (control, 0.1% BSA, i.e.), RX77368 or 5-HT given via the femoral vein. 106 14: Schematic illustrating the theoretical basis for the enteric nervous system mediating vagally-stimulated acid secretion 109 15: Effect of TTX on RX77368-stimulated gastric acid secretion in chronic rats 110 16: A. Effect of local gastric administration of vehicle (V) or TTX (5 fxg, close i.a.) on pentagastrin- stimulated (24 /ig/kg/hr, i.v.) acid secretion. . . 112 B. Ability of local gastric administration of vehicle or TTX to reverse 5-HT-induced (3.5 jLtmol/kg, i.v.) attenuation of pentagastr in- stimulated acid secretion 112 17: Time course of the ability of 5-HT to inhibit carbachol-stimulated gastric acid secretion in rats after acute bilateral, cervical vagotomy 114 18: Time course of the ability of 5-HT to attenuate pentagastrin-stimulated acid secretion in sham vs. total celiac ganglionectomized (GX) rats 117 19: Effect of total celiac ganglionectomy on norepinephrine content in whole tissue sections of gastric corpus 120 20: A. Effect of vehicle (V) or piroxicam (PX, 15 jtimol/kg, i.pO on pentagastrin-stimulated (24 ng/kg/hr, i.v.) acid secretion 122 B. Ability of vehicle or piroxicam to reverse vehicle or 5-HT-induced (3.5 /xmol/kg, i.v.) attenuation of pentagastrin-stimulated acid secretion 122 21: Effect of SNP on MAP 125 22: Representative tracings of the effect of SNP, infused close to the gastric wall through the splenic artery, and intravenous vehicle (A.) or 5- HT (B.) on MAP 126 23: Time course of the effect of SNP on 5-HT-induced attenuation of acid secretion 129
xv 24: Effect of 5-HT to attenuate pentagastrin-stimulated acid secretion in vehicle vs. chronic dexamethasone-treated rats 131 25: Time course of the ability of 5-HT to attenuate pentagastrin-stimulated acid secretion in sham vs. acute, bilaterally adrenalectomized rats 134 26: A. Effect of local gastric infusion of various 5-HT antagonists on pentagastrin-stimulated (24 Mg/kg/hr, i.v.) gastric acid secretion in chronic rats 149 B. Ability of local gastric infusion of various 5-HT antagonists to reverse 5-HT-induced (3.5 jLtmol/kg, i.v.) attenuation of pentagastrin-stimulated acid secretion in chronic rats 149 27: Effect of metergoline on RX77368-stimulated acid secretion in chronic rats 151 28: Effect of methiothepin on RX77368-stimulated acid secretion in chronic rats 152 29: Effect of methysergide on RX77368-stimulated acid secretion in chronic rats 153 30: Effect of spiperone on RX77368-stimulated acid secretion in chronic rats 154 31: Time course of the effect of systemic vehicle, 5-HT or 8-OH-DPAT on pentagastrin-stimulated gastric acid secretion 166 32: Dose-response profile of the effects of intravenous (A.) 5-HT or (B.) 8-OH-DPAT on pentagastrin- stimulated gastric acid secretion 168 33: Comparison of the effect of systemic 8-OH-DPAT on pentagastrin-stimulated gastric acid secretion in acute adrenalectomized (ADX) versus sham-treated animals 170 34: A. Effect of vehicle, spiperone (2.5 jumol/kg) or idaxozan (2.5 /nmol/kg) on pentagastrin-stimulated (24 /xmol/kg/hr, i.v.) acid secretion. B. Ability of local gastric infusion of vehicle, spiperone (2.5 /imol/kg) or idaxozan (2.5 /zmol/kg) to reverse 8-OH-DPAT-induced (0.35 jttmol/kg, i.v.) enhancement of pentagastrin-stimulated gastric acid secretion 172
xvi LIST or TABLES TABLE PAGE 1: Pharmacologic basis for the classification of 5-HT receptor subtypes...... 6 2: Summary of the mechanisms mediating release of 5-HT into the gastric lumen in vivo 14 3: Summary of the mechanisms mediating release of 5-HT appearing in the portal circulation in vivo. ... 20 4: Summary of the serotonin antagonists and agonists used in these studies with their abbreviations and serotonin receptor subtype selectivity 68 5: Effect of intravenous infusion of either vehicle, 5- HT (3.5 jLtmol/kg) or 5-CT (3.5 /xmol/kg) on pentagastrin-stimulated (24 /nnol/kg/hr, i.v.) gastric acid secretion 70 6: Comparison of the affinities (Ki) of various 5-HT antagonists for 5-HT,, a, and D2 receptors 157
xvii CHAPTER I
INTRODUCTION
I. Historical Aspects As in many fields of science, observation and serendipity unite thoughts and ideas. This scenario describes the discovery of serotonin (5-HT). In 1934, Dr. Irvine Page suspected that the serum of blood contained a vasoconstrictor released from platelets upon clotting. The isolation and crystallization of this serum vasoconstrictor was initially performed to decontaminate experimental samples. In 1949, Dr. Page along with Drs. A. Green and M. Rapport at the Cleveland Clinic, were the first to identify 5-HT as the serum vasoconstrictor (Espamer and Asero, 1952? Twarog, 1988). Independently, in 1952, the Italian investigators Erspamer and Asero (Espamer and Asero, 1952; Twarog, 1988) first identified enteramine-containing endocrine cells in the gut [now identified as enterochromaffin (EC) cells]. They extracted the enteramine and identified it as 5-HT. They also ascribed the first biological activity to 5-HT, the contraction of smooth muscle. In 1951, thanks to Hamlin and Fischer (Espamer and Asero, 1952), 5-HT became commercially
1 2 available and many eager investigators began studies to assess its biologic significance. Two working hypotheses have been put forward to explain release of 5-HT from the EC cell in response to physiologic stimuli. Functional studies on.release of endogenous 5-HT suggest that the apical villi of the EC cell "taste" the luminal contents and, subsequent to appropriate stimuli, the basolateral surface responds by exocytosis of 5-HT (Fujita and Kobayashi, 1974; Kobayashi and Fujita, 1974). According to the second hypothesis, activation or deformation of mucosal mechanoreceptors induces release of 5-HT from the EC cell (Money et al. 1988; Kirchgessner et al. 1992). Both hypotheses implicated 5-HT as a 'chemical transducer' that communicates changes in the mucosal microenvironment to the gut.
II. Regional Distribution of Serotonin in Mammalian Systems. A. Central Nervous System. In 1953, 5-HT was discovered to be widely distributed in the brain. Neuropharmacologic studies implicated 5-HT in many aspects of personality and behavior, some of which continue to be actively pursued today. B. Gastrointestinal Tract. In 1965, Gershon, Drakontides, and Ross (Espamer and Asero, 1952) hypothesized that 5-HT was a neurotransmitter in the enteric nervous system. This provided new impetus to study the involvement 3 of serotonin in many aspects of gastrointestinal (GI) physiology, especially peristalsis and secretion. There are three main sources of 5-HT in the gut. These are EC cells, neurons and mucosal mast cells (MMC) of some species. Morphologically, the endocrine cell is pear shaped with apical villi and basolateral granules (Fujita and Kobayashi, 1974; Sasagawa et al. 1974). In the rat stomach, the EC cell is primarily located in the antrum (Gregg, 1966; Hakanson, 1970; Hasegawa et al. 1987) at the basal 1/3 of glands (Gregg, 1966). Serotonin is also found in the EC cells of the small intestine and colon. Only myenteric nerve cell bodies and their terminals contain 5-HT. Submucosal nerve cell bodies and terminals do not contain 5-HT, but can express 5-HT receptors. Serotonin is also found in the mast cells of the GI mucosa of the rat and mouse (Mate et al. 1980).
III. Synthesis and Degradation A. Synthesis. Serotonin is synthesized by the host cell, except for platelets which possess active uptake systems for
5-HT. The precursor, L-tryptophan, is hydrolyzed by tryptophan hydroxylase into 5-hydroxytryptophan. This is the rate-limiting step in the synthesis of 5-HT. The enzyme aromatic L-amino acid decarboxylase then uses 5-hydroxytryptophan as a substrate to produce 5-hydroxytryptamine (5-HT). 4 B. Degradation. The estimated half life of endogenous 5- HT in the circulation is 60-120 seconds (Gillis, 1985). Serotonin is rapidly inactivated. The principle pathways for inactivation of 5-HT are: 1) cellular reuptake mechanisms utilizing active transport systems [i.e. adrenergic neurons, platelets, mast cells (Verbeuren, 1989; Gillis, 1985)] and 2) oxidative deamination into 5-hydroxyindole acetic acid by the A isozyme of monoamine oxidase. Monoamine oxidase is primarily found in the liver and endothelial cells of blood vessels and lungs (Gillis, 1985). Approximately 90% of endogenous 5-HT released is taken up by platelets (Jaffe et al. 1977; Gillis, 1985). For comparison, studies estimate that exogenously administered 5-HT is deactivated by platelets (Jaffe et al. 1977; Gillis, 1985), the liver (Gillis, 1985) and the lung (Beck, 1985; Gillis, 1985) such that, after one passage through the circulation, over 90% of the 5-HT is rendered inactive.
IV. Localization of Serotonin Receptors. To date, five subtypes of 5-HT receptors have been
localized to the GI tract. These subtypes include 5-HT1A [rat
(Kirchgessner et al. 1993a)], 5-HT1p [rat (Branchek and Gershon, 1987; Kirchgessner et al. 1993a), guinea pig (Mawe et
al. 1986) and rabbit (Branchek et al. 1988)], 5-HT2 [guinea pig (Siriwardena et al. 1993)], 5-HT2B [rat (Cohen et al.
1992)], 5-HTj [rat (Champaneria et al. 1992)] and 5-HTA [guinea 5 pig (Pan and Galligan, 1994; Kilbinger and Wolf, 1992; Craig et al. 1990; Rizzi et al. 1992; Craig and Clarke, 1990)]. For information on the classification of 5-HT receptor subtypes based on their pharmacologic properties see TABLE 1. 6
TABLE 1: Pharmacologic basis for the classification of 5-HT receptor subtypes. Antagonists and agonists are listed in rank order of potency. Information taken from Hoyer et al. 1994. RECEPTOR SOBTYPE AGONISTS ANTAGONISTS
5-HTu _ DP-5-CT SDZ 216525 5-CT NAN-190 5-Methyl-urapidi1 SDZ 21009 8-OH-DPAT Methiothepin Metergoline Spiperone Methysergide
5-HT IP 5-OHIP N-hexanoyl-5-HTP-DP 6-OHIP N-acety1-5-HTP-DP Bufotenine Renzapride 5-HT, DOI Pirenperone a-Methyl-5HT Ketanserin RU 24969 Ritanserin 8-OH-DPAT Methiothepin 5-CT Spiperone Metergoline 5-HT, 2-Methyl-5HT Tropisetron Phenylbiguanide Zacopride Granisetron MDL 72222 Ondansetron Renzapride Quipazine Metoclopramide 5-HT, Cisparide SB 204070 5-MeOT GR 113808 Renzapride SDZ 205557 Zacopride Tropisetron 5-CT 7
A. 5-HT1a Receptors. In the rat, 5-HT1A receptors are found in the stomach, all levels of the intestine and in the colon, with very high density in the duodenum, corpus, antrum, pylorus and fundus (Kirchgessner et al. 1993a). Specifically, these receptors are found in the 1) myenteric plexus, 2) submucosal plexus, 3) mucosa at the base of glands, at the junction of the lamina propria and the muscularis mucosa, and 4) submucosa, close to mast cells and blood vessels
(Kirchgessner et al. 1993a). By contrast, no 5-HT1A receptors are found in the guinea pig GI tract (Kirchgessner et al.
1993a). The mRNA of the 5-HT1A receptor, as determined by in situ hybridization, is only found associated with neurons; no mRNA is found in the mucosa (Kirchgessner et al. 1993a). Kirchgessner et al. estimated that approximately 50% of myenteric neurons contain mRNA for 5~HT1A receptors (Kirchgessner et al. 1993a). To date, these receptors have been electrophysiologically characterized as mediating presynaptic inhibition in the enteric nervous system (Galligan et al. 1988; Galligan, 1992; Pan and Galligan, 1994). B. 5-HT.jp Receptors. Autoradiographic studies show that
5-HT1p receptors are localized to the: 1) myenteric plexus, 2) submucosal plexus, 3) lamina propria near glands, and 4) mucosa near 5-HT-containing EC cells (Kirchgessner et al. 1993a; Branchek et al. 1988). These receptors are electrophysiologically characterized as mediators of slow excitatory post-synaptic potentials (Mawe et al. 1986; Gershon 8 et al. 1989) and are hypothesized to be involved in peristalsis (Kirchgessner et al. 1992).
C. 5-HTz Receptors. Less information is available regarding receptors from the 5-HT2 family. 5-HT^ receptors are localized to small intestinal crypt cells of the guinea pig (Siriwardena et al. 1993). These receptors are hypothesized to be involved in secretion. 5-HT2B receptors of the smooth muscle of the rat stomach fundus have been cloned and are pharmacologically similar to the receptor mediating 5- HT-induced contractions of fundic strips (Cohen and Flundzinzki, 1987; Cohen et al. 1992). Both second messenger systems appear to involve hydrolysis of phosphotidyl inositol (Cohen et al. 1992; Siriwardena et al. 1993) through receptor coupling to G proteins (Wang et al. 1993).
D. 5—HTj Receptors. 5-HT3 receptors are localized to the terminal small intestine (Champaneria et al. 1992). Their presence is difficult to establish due to the inability in achieving saturated binding in many regions of the gut. These receptors are electrophysiologically characterized as directly gating ionic channels that produce rapid depolarizing responses (Wade et al. 1991; Kilbinger and Pfeuffer-Friederich, 1985).
E. 5-HT4 Receptors. Evidence exists for the distribution of 5-HTa receptors in the rat colon, ileum and oesophagus (Hoyer et al. 1994). These electrophysiological studies suggest 5-HT4 receptor activation enhances nerve-mediated 9 contractions of the guinea pig ileum (Rizzi et al. 1992; Craig and Clarke, 1990) through release of acetylcholine from enteric neurons (Kilbinger and Wolf, 1992). ,5-HT4 agonists increase peristaltic frequency of whole guinea pig ileal segments in vitro (Rizzi et al. 1992). The data suggest 5-HT4 receptors may be involved in peristaltic contractions.
V. Overview In vivo release of 5-HT into the gastric lumen and portal circulation will be described by reviewing mechanisms involved in luminal chemoreceptor-activated release along with central and enteric nervous control of release. In vitro studies will be reviewed to document the influence of cholinergic and adrenergic agents and stimulants of second messenger systems, along with neural and enzymatic activity on the vectorial release of 5-HT. A comparison of in vivo and in vitro studies regarding the mechanism of 5-HT release will conclude this section. Finally, studies describing the effects of the appearance of endogenous 5-HT in the lumen and portal circulation and actions of exogenous 5-HT on GI function will be reviewed. This will focus on the stomach and the small intestine and effects on acid secretion, motility, gastric emptying, blood flow, gastric lesions, and vagal activity. 10 VI. Mechanism of Release of 5-HT from Gastrointestinal stores: In Vivo Studies. A. Introduction. In vivo studies provide some advantages over in vitro studies. In these experiments, physiologic, pharmacologic or manual manipulations were performed to ascertain their individual influences on the release mechanism of 5-HT from the GI mucosa. More importantly, these in vivo studies provide opportunity for investigating the integration of individual systems to enhance release of 5-HT. Detail of the release mechanism(s) for 5-HT is obtained by analyzing its appearance in the lumen and/or portal circulation; release into these two compartments will be discussed individually. Serotonin appearing in the gastric lumen has the potential to interact with other mucosal cells (i.e. parietal cell) while 5-HT released basolaterally can interact not only with other mucosal cells but also with 1) afferent nerves of the vagus nerve, splanchnic nerves or enteric nervous system, 2) mucosal immune cells or 3) intramural blood vessels. B. Release into the Lumen 1. Activation of Chemoreceptors. The physiologic stimulus of a high-protein meal enhances the release of 5-HT into the dog jejunal lumen (Ferrara et al. 1987). Release of 5-HT is blocked by both atropine and propranolol. In this paradigm, both cholinergic and adrenergic systems exert 11 control over the release of 5-HT. Release of 5-HT occurs in response to protein in the intestinal lumen. 2. Central Vagal Control. The concentration of 5-HT in the rat gastric lumen increases in response to vagal stimulation (Cho et al. 1985; Raybould et al. 1990; Stephens and Tach6, 1989). Infusion of the muscarinic agonist, bethanechol, via the femoral vein, does not increase the concentration of 5-HT in the gastric lumen. The sole activation of muscarinic receptors by bethanechol does not provoke release of 5-HT into the gastric lumen (Stephens and Tache, 1989). The data also suggest that reduction in intragastric pH produced by bethanechol does not release 5-HT into the gastric lumen. The vagally-mediated release of 5-HT is prevented by cervical vagotomy (Stephens and Tache, 1989) [but not perivagal capsaicin (Raybould et al. 1990)], pretreatment with atropine (Cho et al. 1985; Stephens and Tach6, 1989) or adrenalectomy (Stephens and Tach£, 1989). Moreover, intracisternal injection of 0DT8-SS, a somatostatin analogue that stimulates acid secretion by vagal-dependent mechanisms and inhibits sympathetic outflow to the adrenals, decreases the concentration of 5-HT in the gastric lumen (Stephens, 1991). Therefore, in the rat, the participation of both muscarinic mechanisms and the adrenal component of the sympathetic nervous system appear essential for vagally- mediated release of 5-HT into the gastric lumen. 12 Basal release into the intestinal lumen is spontaneous (Gronstad et al. 1987; Ahlman et al. 1984; Ahlman et al. 1981a; Ahlman et al. 1981b; Cho et al. 1985; Gronstad et al. 1988a) and also under vagal control. In the rat jejunum, an increased concentration of 5-HT in the lumen is demonstrated after stimulation of the cervical vagus (Ahlman et al. 1984; Ahlman et al. 1981a). The increase in 5-HT concentration in the intestinal lumen has been studied most extensively in the cat. In both control cats (Gronstad et al. 1987; Ahlman et al. 1981b; Gronstad et al. 1988a; Zinner et al. 1982) and superior cervical ganglionectomized cats (Gronstad et al. 1987), stimulation of the thoracic or cervical vagus enhances release of 5-HT into the intestinal lumen. Pretreatment with propranolol (Gronstad et al. 1987; Gronstad et al. 1988a) or low doses of atropine (Gronstad et al. 1987) does not alter the pattern of 5-HT release. Administration of a high dose of atropine or hexamethonium abolishes thoracic vagal-stimulated release Of 5-HT into the intestinal lumen (Gronstad et al. 1987; Gronstad et al. 1988a). The studies suggest that vagally-mediated nicotinic mechanisms are involved in release of 5-HT into the intestinal lumen of the cat. 3. Mucosal Mast Cell Degranulation. Vagal stimulation significantly reduces both the mast cell count in rat gastric mucosa and submucosa (Cho et al. 1985) and the 5- HT fluorescence in MMC (Tobe et al. 1976). These reductions 13 are correlated with increased concentrations of 5-HT and histamine in the gastric lumen (Cho et al. 1985). Systemic atropine pretreatment prevents mast cell degranulation and decreases vagally-stimulated release of 5-HT and histamine into the gastric lumen (Cho et al. 1985). Vagally-induced degranulation of MMC may contribute to the concentration of 5- HT in the rat gastric lumen. 4. Summary (TABLE 2). In summary, release of 5-HT into the dog intestinal lumen in response to protein meals is both cholinergically and adrenergically controlled (Ferrara et al. 1987). Release evoked by vagal stimulation in the rat is mediated through release of acetylcholine from the vagus (Cho et al. 1985; Stephens and Tach6, 1989), with participation of circulating adrenal catecholamines (Stephens and Tache, 1989) , and in the cat through cholinergic vagus with emphasis on nicotinic pathways (Gronstad et al. 1987). Although the majority of the 5-HT released into the gastric lumen by vagal stimulation is thought to originate in EC cells, a portion may be from MMC degranulation (Cho et al. 1985; Tobe et al. 1976). The contribution of endogenous 5-HT in the lumen to the regulation of gut function is unknown. 14
TABLE 2: Summary of the mechanisms mediating release of 5-HT into the gastric lumen in vivo. Increased circulating catecholamines are involved in mediating the release of 5-HT into the gastric lumen of both the rat and cat whereas the participation of adrenergic nerves is species dependent. (+) involved in the mechanism of release. (-) not involved in the mechanism of release. NR: not reported.
adrenal cholinergic adrenergic catecholamines vacms nerves Rat + NR +
Cat + + 15 C. Release into the Portal Circulation 1. Activation of Chemoreceptors. Hypertonic glucose (Drapanas et al. 1962) or a high protein meal (Ferrara et al. 1987) in the dog intestinal lumen increases the concentration of 5-HT appearing in the hepatic portal (Drapanas et al. 1962) and systemic venous blood (Ferrara et al. 1987; Drapanas et al. 1962). Components of the meal, in contact with the duodenal mucosa, can generate release of 5-HT. As eluded to above (section VI.B.l.), 5-HT may act as an enterogastrone involved in the control of gastric acid secretion. In the dog, acid in the duodenum concomitantly enhances systemic, venous 5-HT concentration and decreases pentagastrin-stimulated acid secretion in control animals but not in denervated Heidenhain pouches (Jaffe et al. 1977). Although the response is vagally-mediated (Jaffe et al. 1977), no antagonists of 5-HT receptors were tested. Hence, an important hypothesis for the work reported in this dissertation was that acid, via enhancement in release of 5-HT and activation of 5-HT receptors on duodenal mucosal vagal afferents, evokes an inhibitory reflex that suppresses gastric acid secretion. The mechanism of action of 5-HT to inhibit acid secretion is discusses in CHAPTER III. 2. Central Vagal Control. Stimulation of the vagus releases 5-HT into the portal circulation of the rat (Tobe et al. 1976), cat (Pettersson, 1979; Gronstad et al. 1987; Siepler et al. 1980; Money et al. 1988; Gronstad et al. 16 1988a), rabbit (Horita and Carino, 1982; Horita et al.1985; Horita et al. 1988) and dog (Thompson, 1977). Due to apparent species differences, the mechanism of release, in this section on vagal control, will be discussed for each species individually. The concentration of 5-HT in portal whole blood is elevated in the rat after vagal stimulation (Tobe et al. 1976). Release into the portal circulation is partially inhibited by atropine pretreatment, abolished by atropine together with hexamethonium, and increased by 6-OH-dopamine (6-OH-DA) pretreatment (Tobe et al. 1976). This suggests that vagally-mediated release of 5-HT appearing in the portal circulation of the rat requires both muscarinic and nicotinic receptor activation. Adrenergic neurons may tonically inhibit the release of 5-HT directly or tonically inhibit the stimulation of release of 5-HT by acetylcholine. Vagal stimulation is reported to either increase (Pettersson, 1979; Gronstad et al. 1987; Siepler et al. 1980; Money et al. 1988; Gronstad et al. 1988a) or not alter (Ahlman et al. 1981b; Zinner et al. 1982) portal concentrations of 5- HT in the cat. Stimulation of the thoracic vagus in both control and hexamethonium treated animals increases the concentration of 5-HT appearing in the portal blood (Gronstad et al. 1987). After removal of the superior cervical ganglion and stimulation of the cervical vagus, no portal venous 5-HT increase is seen (Gronstad et al. 1987; Gronstad et al. 17 1988a). In addition, 5-HT concentration is unaltered in animals stimulated at the level of the cervical vagus when pretreated with propranolol or phenoxybenzamine; moreover, atropine or hexamethonium pretreatment is ineffective to attenuate release in the cat (Pettersson, 1979; Gronstad et al. 1988a). The studies suggest that intramural release of 5- HT, appearing in the portal circulation of the cat, is noncholinergically controlled and involves activation of adrenergic nerves. Stimulation of the rabbit vagus by intracisternal injection of thyrotropin releasing hormone increases the concentration of 5-HT in portal plasma (Horita and Carino, 1982; Horita et al. 1985; Horita et al. 1988). Release of 5- HT is blocked by pretreatment with hexamethonium or by bilateral vagotomy and is enhanced by fluoxetine, a 5-HT reuptake inhibitor (Horita and Carino, 1982). In the rabbit, release of 5-HT in response to intracisternal TRH is vagally dependent. 3. Splanchnic Nervous Control. In the cat, electrical stimulation of the sympathetic splanchnic nerves enhances concentration of 5-HT in the portal vein (Larsson et al. 1979). The mechanism of release was not investigated. 4. Endocrine Control. Recent work suggests that release of 5-HT into the portal venous circulation of the cat and dog is controlled by ^-adrenoceptors and that catecholamines, released by pentagastrin, may mediate release 18 of 5-HT (Gronstad et al. 1988b; Bech et al • 1992a). In contrast, another report states that pentagastrin does not release 5-HT into the portal circulation of the dog (Jaffe et al. 1977). Mechanistic studies in the cat show the concentration of 5-HT in the portal vein is decreased either by adrenalectomy in combination with propranolol (Gronstad et al. 1988b), by pretreatment with verapamil (calcium channel blocker) (Gronstad et al. 1988b), or by isoprenaline (Bech et al. 1992a). The involvement of the vagus is not reported. The mechanisms underlying the similar abilities of adrenalectomy plus propranolol (Gronstad et al. 1988b) or isoprenaline (Bech et al. 1992a) to decrease concentration of 5-HT in the portal vein, in spite of apparent opposing effects
on adrenergic receptors, is unclear. Interplay between f3- adrenoceptors and a-adrenoceptors, together with adrenal- derived catecholamines, may be important in mediating release of 5-HT into the portal circulation of the cat. 5. Summary (TABLE 3). Studies on the mechanism of in vivo release of 5-HT appearing in the portal circulation reveal marked species differences. Vagal stimulation increases portal 5-HT concentrations in the rat (Tobe et al. 1976), cat (Pettersson, 1979; Gronstad et al. 1988a) and rabbit (Horita and Carino, 1982; Horita et al. 1985; Horita et al. 1988). Release of 5-HT into the portal circulation appears to be mediated by cholinoceptor activation in the rat (Tobe et al. 1976) and adrenoceptor activation in the cat 19 (Pettersson, 1979; Gronstad et al. 1988a). The sympathetic nervous system may tonically inhibit 5-HT release in the rat (Tobe et al. 1976) but stimulate release in the cat (Pettersson, 1979; Larsson et al. 1979; Gronstad et al. 1988a). Purely endocrine mediated release of 5-HT appearing in the portal circulation may also occur (Gronstad et al. 1988b; Bech et al. 1992a). 20
TABLE 3: Summary of the mechanisms mediating release of 5-HT appearing in the portal circulation in vivo. Activation of cholinergic but not adrenergic vagal systems are implicated in the rat whereas in the cat, the latter is predominate. Species differences also exist regarding the involvement of adrenal catecholamines. (+) involved in the mechanism of release. (-) not involved in the mechanism of release.
adrenal cholinergic adrenergic catechol am* nog vacms vagus Rat Cat 21 D. Release from Enterochromaffin Cells 1. Activation of Chemoreceptors. The presence of acid in the duodenal lumen degranulates EC cells of the rat (Furman and Waton, 1989; Thompson, 1977) and guinea pig (Furman and Waton, 1989) duodenum. The degranulation of EC cells was confirmed histologically. The data supports the hypothesis that the EC cell 'tastes' the acid in the lumen and responds by degranulation. 2. Central Vagal Control. Electrical vagal stimulation releases 5-HT from EC cells of rat stomach and duodenum as measured by decreased fluorescence of 5-HT (Ahlman and Dahlstrom, 1983; Tobe et al. 1976) and of guinea pig as measured by a decrease in EC cell number staining for 5-HT (Thompson, 1977). In contrast, vagotomy slightly elevates 5- HT tissue concentrations in the rat (Heuser et al. 1979; Tobe et al. 1976) but decreases the number of 5-HT-staining EC cells in the dog (Thompson, 1977). Treatment of the neonatal rat with capsaicin does not alter GI 5-HT concentration (McGregor and Conlon, 1991). Vagal efferents modulate 5-HT tissue content in a species-dependent manner. Upon stimulation of the cervical vagus and ligation of the adrenals (to prevent compensatory mechanisms) in the cat, tissue concentrations of 5-HT are decreased (Ahlman and Dahlstrom, 1983). 5-HT release is not blocked by atropine or a-adrenoceptor blockers, but is blocked by propranolol and removal of the superior cervical ganglion suggesting 22 involvement of the adrenergic nerves (Ahlman and Dahlstrom, 1983). These studies lend supportive evidence to the involvement of the vagus and adrenergic nerves in the release mechanism of 5-HT. 3. Sympathetic Nervous System. The effect of the sympathetic system on the release of 5-HT from the EC cell has been studied by chemical sympathectomy using 6-OH-DA. In the rat, systemic 6-OH-DA increases the intestinal EC cell 5-HT content (Thompson, 1971) but in other studies, decreases the antral tissue 5-HT content (Tobe et al. 1976). The peripheral sympathetic nervous system may differentiate between gastric and intestinal stores of the amine. 4. Summary. These experiments, taken as a whole, suggest that 5-HT release from the EC cell is vagally mediated (Pettersson, 1979; Ahlman and Dahlstrom, 1983; Gronstad et al. 1988b; Heuser et al. 1979; Tobe et al. 1976; McGregor and Conlon, 1991; Thompson, 1977) and possibly through the adrenergic nerves in some species [i.e. cat (Ahlman and Dahlstrom, 1983)]. Peripheral sympathetic neurons may differentially influence gastric vs. intestinal stores of 5-HT (Thompson, 1971; Tobe et al. 1976).
VII. Mechanisms of Release of Serotonin from Gastrointestinal Stores: In Vitro Studies. A. Introduction. When investigating release mechanism of 5-HT, in vitro methods have some distinct advantages over in 23 vivo methods. Contributions by individual components of the GI system may be more easily examined by these approaches. Only release from distinct segments and cell types (i.e. neurons, EC cells) of the small intestine have been studied and is reviewed herein. B. Release to Mucosal Side 1. Nutrients. In addition to acid and protein in the lumen of the whole animal (section VI.B.l., C.l. and D.1.), hypertonic solutions in the lumen also stimulate release of 5-HT into the intestinal lumen. In the rabbit, application of dextrose or mannitol to the mucosal sheet (no muscle layers) releases 5-HT in a concentration-dependent manner (Martin et al. 1989). In contrast, dextrose in the lumen of an isolated perfused segment, increases the concentration of 5-HT appearing in the portal system, but not the intestinal lumen (Martin et al. 1989). This suggests 5-HT appearing in the lumen in the first study may simply be a consequence of diffusion. The dextrose-induced release of 5- HT is blocked by tetrodotoxin (TTX), but not by atropine, hexamethonium, phentolamine, or propranolol indicating a neural noncholinergic, nonadrenergic pathway regulating 5-HT release (Martin et al. 1989). 2. Acid in the Duodenal Lumen. Beginning at pH 4, a pH-dependent increase in the release of 5-HT from both mucosal and serosal sides of rabbit duodenal epithelium is observed (Kellum et al. 1983). Acid-stimulated release of 5- 24 HT into the lumen is inhibited by atropine, hexamethonium or propranolol but not by phentolamine. In addition, synergism of 5-HT release is observed in response to carbachol and isoproterenol. In vitro acid-stimulated 5-HT release from duodenal mucosa is mediated by activation of both muscarinic and /3 adrenergic receptors. 3. Cholinergic Receptors. Serosal application of various nicotinic agonists (Forsberg and Miller, 1983; Forsberg and Miller, 1982; Kellum et al. 1983) or antagonists (Kellum et al. 1983), acetylcholine alone (Forsberg and Miller, 1982) or in combination with neostigmine (Forsberg and Miller, 1983), carbachol (Forsberg and Miller, 1983) or atropine (Kellum et al. 1983) are ineffective in stimulating release of 5-HT from rabbit duodenal mucosa. In other studies, application of carbachol to mucosal plus serosal bathing solutions enhances release of 5-HT from the mucosal side (Kellum et al. 1983). Carbachol-induced release is independent of /9 adrenergic mechanisms. The effects of atropine, hexamethonium or a adrenergic antagonists on the carbachol response was not reported. Cholinergic receptor activation by the muscarinic agonist methacholine, applied to mucosal plus serosal surfaces of the preparation, increases net release of 5-HT predominantly in the mucosal direction (Kuemmerle et al. 1987). Methacholine- induced release of 5-HT is TTX-resistant. Exposure of the tissue to both TTX and methacholine increases release of 5-HT 25 above that seen for methacholine alone (Kuemmerle et al. 1987). In other studies (section VII.C.2.), pretreatment with an antibody to vasoactive intestinal peptide (VIP) reverses inhibition of 5-HT release by the muscarinic agonist oxotremorine (Rack6 et al. 1991) suggesting muscarinic receptor activation promotes release of VIP, which in turn, inhibits release of 5-HT. The data suggests that release of 5-HT into the intestinal lumen may be mediated by activation of muscarinic receptors present on enteroendocrine cells and enteric neurons may release inhibitory neurotransmitters when activated by muscarinic agonists. 4. Adrenergic Receptors. Serosal application of isoproterenol or epinephrine does not promote release of 5-HT from rabbit duodenal mucosa (Forsberg and Miller, 1982). In other studies with rabbit duodenum, the application of isoproterenol to mucosal plus serosal sides of the tissue increases release of 5-HT into the lumen (Kellum et al. 1983). Atropine does not alter isoproterenol-induced 5-HT release suggesting independence from cholinergic systems. Neither norepinephrine nor propranolol is effective in releasing 5-HT (Kellum et al. 1983). In similar experiments, simultaneous application of isoproterenol in combination with TTX potentiates spontaneous 5-HT release (Kellum et al. 1983).
The data suggests /3 adrenoceptors may enhance release of 5-HT while neuronal influences may inhibit spontaneous 5-HT release. 26 5. Enteric Neurons. In the rabbit, pretreatment with TTX alone elevates spontaneous 5-HT release into the lumen (Kuemmerle et al. 1987). This suggests that a tonic inhibitory influence by enteric neurons on spontaneous 5-HT release may exist. The inhibition is noncholinergic and nonadrenergic since exposure of the tissue to cholinergic and adrenergic agonists predominantly stimulates release of 5-HT (Kuemmerle et al. 1988; Forsberg and Miller, 1983; Forsberg and Miller, 1982). The release of 5-HT induced by TTX is enhanced by muscarinic agonists suggesting that muscarinic receptors are located on the EC cell, in close proximity to the EC cell, or activates a nearby cell which exerts a paracrine effect on the EC cell to enhance 5-HT release. 6. Intraluminal Pressure. In the guinea pig and rabbit isolated small intestine, a rise in intraluminal pressure, due to peristalsis, is highly correlated with enhanced release of 5-HT into the intestinal lumen (Bvilbring and Lin, 1958). In addition, activation of pressure receptors of the mucosa by puffs of nitrogen may evoke release of 5-HT from the gut (Kirchgessner et al. 1992). 7. Summary. Dextrose, applied to the duodenal lumen, increases the release of 5-HT into the lumen. The pathway involved in dextrose-induced release of 5-HT is mediated by noncholinergic and nonadrenergic enteric neurons (Martin et al. 1989). 27 Nicotinic agonists or antagonists have no effect on release of 5-HT into the lumen (Forsberg and Miller, 1983; Forsberg and Miller, 1982). In contrast, either muscarinic agonists or isoproterenol stimulates release of 5-HT into the lumen (Kellum et al. 1983). The data suggests that cholinergic receptors are preferentially located on the serosal side whereas adrenergic receptors are located on the mucosal side of the preparation. The enteric nervous system can enhance basal release of 5-HT from rabbit duodenal mucosa (Kuemmerle et al. 1987). However, when activated by an agonist, the enteric nervous system appears to provide a tonic inhibition of 5-HT release (Kuemmerle et al. 1987; Kellum et al. 1983). The data suggests that agonist activation of the enteric nervous system can promote release of neurotransmitters capable of inhibiting 5-HT release (Rack6 et al. 1991). C. Release to Serosal side 1. Nutrients. Mucosal application of dextrose to the isolated rabbit duodenum increases portal plasma 5-HT concentrations (Martin et al. 1989). Nutrients in contact with the mucosa of the duodenum can influence release of 5-HT into both the lumen and the serosa. 2. Cholinergic Receptors. Stimulation of muscarinic receptors by oxotremorine inhibits release of 5-HT into the venous effluent of the isolated small intestine of the guinea pig (Racke et al. 1991). Pretreatment with an antibody to 28 vasoactive intestinal peptide (VIP) reverses inhibition of 5- HT release by oxotremorine (Rack6 et al. 1991) suggesting muscarinic receptor activation promotes release of VIP, which in turn, inhibits release of 5-HT. Application of TTX enhances oxotremorine-induced release of 5-HT into the venous effluent as compared to control (Racke et al. 1991) suggesting that stimulatory muscarinic receptors are located on EC cells. In Ussing chamber experiments, serosal application of muscarinic agonists oxotremorine (Forsberg and Miller, 1983), acetylcholine along with neostigmine (Forsberg and Miller, 1983; Forsberg and Miller, 1982) or carbachol (Forsberg and Miller, 1982; Forsberg and Miller, 1983) increases release of 5-HT from the serosal side. Carbachol-induced release is inhibited by atropine (Forsberg and Miller, 1982) but is unaffected by hexamethonium (Forsberg and Miller, 1982) or TTX (Forsberg and Miller, 1983; Forsberg and Miller, 1982). Application of carbachol to the mucosal side (Forsberg and Miller, 1983), of neostigmine to the serosal side (Forsberg and Miller, 1982), or of carbachol to both sides of the mucosa (Kellum et al. 1983) has no effect on serosal 5-HT release. This suggests simultaneous activation of muscarinic receptors on the apical and basolateral side can prevent release of 5- HT. Application of the nicotinic agonist, DMPP, to the serosal side of the rabbit duodenal mucosa has no effect on release of 5-HT from the serosal side (Forsberg and Miller, 29 1983). Serotonin release appears to be mediated through activation of muscarinic receptors in the serosa or directly located on the EC cell. 3. Adrenergic Receptors. Ussing chamber experiments with rabbit duodenal mucosa demonstrate that mucosal plus serosal application of isoproterenol does not alter serosal 5- HT release (Kellum et al. 1983). In contrast, 5-HT release into the portal venous effluent is enhanced from isolated guinea pig ileum after treatment with isoprenaline (Rack6 et al. 1988). This enhanced release is prevented by propranolol but not TTX suggesting non-neuronal /3 adrenergic receptors are involved.
Treatment of the guinea pig ileum with clonidine, an a2 adrenergic agonist, decreases release of 5-HT into the portal venous effluent (Racke et al. 1988). Attenuation of 5-HT
release is prevented by pretreatment with tolazoline (a2 antagonist) but not prazosin (a, antagonist) or TTX suggesting
non-neuronal a2 receptors are involved (Racke et al. 1988). 4. Enteric Neurons. Upon application of TTX to the organ bath of the guinea pig ileum, the spontaneous release of 5-HT into the intact vascular system is attenuated 35% (Schworer et al. 1987; Racke et al. 1990; Rack6 et al. 1988). Imipramine, an inhibitor of norepinephrine and 5-HT reuptake into neurons and other cells, increases serosal 5-HT release 66% (Schworer et al. 1987). This suggests a component of 30 5-HT released is normally inactivated by reuptake into neurons and other cells. 5. Second Messenger Systems a. cyclic 3 ',5 '-Adenosine Monophosphate (cAMP). The application of 8-Br-cAMP (Forsberg and Miller, 1983), IBMX [phosphodiesterase inhibitor, (Forsberg and Miller, 1983)], AH 21-132 [phosphodiesterase inhibitor, (Racke et al.198811 or forskolin [adenylate cyclase agonist, (Forsberg and Miller, 1983; Racke et al. 1988)] to rabbit duodenal serosa or guinea pig ileum enhances the concentration of 5-HT appearing in the portal circulation. The cAMP-induced release of 5-HT is resistant to TTX suggesting enteric neurons are not involved (Rack6 et al. 1988). Adenylate cyclase systems may be one signal transduction mechanism involved in release of 5-HT from the gut. b. Cyclic 3',5'-Guanine Monophosphate (cGMP). Neither serosal nor mucosal application of 8-Br-cGMP to the rabbit duodenal mucosa has any effect on release of 5-HT (Forsberg and Miller, 1983). Guanylate cyclase signal transduction systems appear not to be involved in release of 5-HT. c. Calcium. Studies involving depletion of extracellular calcium (Forsberg and Miller, 1983; Schworer et al. 1987; Racke et al. 1990), chelation of extracellular calcium by EGTA (Forsberg and Miller, 1983), and application of calcium ionophores to increase intracellular concentrations 31 of calcium (Forsberg and Miller, 1983) demonstrate the necessity of elevated intracellular calcium levels for release of 5-HT from guinea pig isolated stomach (Rack6 et al. 1990), guinea pig ileal serosa (Schworer et al. 1987) and carbachol- stimulated rabbit duodenal serosa (Forsberg and Miller, 1983). The data suggest calcium may act as a second messenger since calcium channel blockers such as verapamil, nifedipine, and 8- (diethylamino)octyl 3,4,5-trimethoxybenzoate HCl have no effect on release of 5-HT (Forsberg and Miller, 1983). Taken together, the studies suggest that elevated intracellular calcium, either by acting as a second messenger or by mediating exocytosis, is important for agonist-stimulated 5-HT release. 6. Enzymatic Activity. Application of
DL-a-monofluoromethyldopa, an irreversible inhibitor of
L-tryptophan decarboxylase, to the perfusing medium of the guinea pig ileum organ bath has no effect on release of 5-HT from the serosal side of the tissue (Schworer et al. 1987). Preformed 5-HT comprises the bulk of released biogenic amine. 7. Serotonin Receptors. The participation of serotonergic receptors in the release profile of 5-HT has been investigated in the presence of TTX in the isolated guinea pig small intestine (Gebauer et al. 1993). Application of the 5-
HT3 agonist, 2-methyl 5-HT, enhances release whereas the 5-HT3 antagonists tropisetron, MDL 72222 or granisetron decrease concentration of 5-HT appearing in the portal venous effluent 32
(Gebauer et al. 1993). Receptors of the 5-HT3 family on non- neuronal elements enhance release of 5-HT. In contrast, in the presence of TTX, application of the
5-HT4 agonists 5-methoxytryptamine and BIMU8 inhibit release
of 5-HT whereas the 5-HT4 antagonist, tropisetron (/molar), enhances release of 5-HT into the portal venous effluent (Gebauer et al. 1993). Tropisetron may also suppress adenylate cyclase thereby decreasing cAMP and producing the opposite effect on 5-HT release (see section VII.C.5.a.). The
data suggest 5-HT4 receptors in association with non-neuronal elements inhibit release of 5-HT. Other evidence suggests
neither 5-HT, nor 5-HT2 receptors are located on EC cells (Gebauer et al. 1993). 8. Summary. Activation of serosal muscarinic receptors enhances the serosal release of 5-HT from rabbit duodenal mucosa (Forsberg and Miller, 1983; Forsberg and Miller, 1982). Activation of 0 adrenergic receptors enhance,
whereas activation of a2 receptors attenuates, 5-HT release (Racke et al. 1988). 5-HT is readily available for release as synthesis inhibitors are ineffective in altering release of 5- HT from rabbit duodenal mucosa (Schworer et al. 1987). Both cAMP (Forsberg and Miller, 1983) and intracellular calcium (Forsberg and Miller, 1983; Schworer et al. 1987) are second messenger candidates for mediating the release of 5-HT. 33 Further regulation of 5-HT release occurs at two levels. Serotonin receptors located on enteric neurons and EC cells may modulate release of 5-HT from guinea pig mucosa (Gebauer et al. 1993). Enteric TTX-sensitive neurons may stimulate spontaneous release of 5-HT from the guinea pig mucosa (Schworer et al. 1987; Forsberg and Miller, 1982). However, when stimulated by an agonist, enteric neurons of the guinea pig (Rack6 et al. 1991), but not the rabbit (Forsberg and Miller, 1983; Forsberg and Miller, 1982), inhibit release of 5-HT, probably through release of the inhibitory neurotransmitter VIP (Racke et al. 1991). D. Release from EC Cells or Tissue Segments 1. Cholinergic Receptors. Release of 5-HT from the dog and pig jejunum is enhanced by nicotine (Racke and Schworer, 1992). Hexamethonium, but not a bungarotoxin, blocks nicotine-enhanced release of 5-HT. In the pig, the nicotine stimulated release is TTX-insensitive and enhanced by scopolamine whereas in the dog, nicotine-stimulated release is TTX-sensitive and antagonized by scopolamine (Racke and Schworer, 1992). The data suggest that, in the pig, activation of cholinergic interneurons incites release of an inhibitory neurotransmitter [possibly VIP (Racke et al. 1991)] whereas in the dog, activation of cholinergic interneurons induces release of a stimulatory neurotransmitter, probably acetylcholine. Hence, oxotremorine inhibits 5-HT release in the pig by release of an inhibitory neurotransmitter but 34 enhances 5-HT release in the dog by directly stimulating the EC cell (Racke and Schworer, 1992). 2. Adrenergic Receptors. Incubation of rat duodenal tissue with norepinephrine, epinephrine, or isoprenaline depletes EC cells of 5-HT (Pettersson, 1979). The responses to norepinephrine and epinephrine are dose-dependent, with epinephrine-stimulated release being antagonized by propranolol. When a adrenergic receptors are blocked by phentolamine or phenoxybenzamine, the depleting effect of epinephrine is potentiated. Neither dopamine nor metoprolol given alone alters the EC cell 5-HT content [102]. Under adrenergic blockade, epinephrine solely interacts with adrenergic receptors. These results suggest catecholamines deplete EC cells of 5-HT through a adrenergic mechanism. Transmural field stimulation of the guinea pig intestine depletes EC cells of 5-HT (Pettersson, 1979). Depletion of 5- HT is attenuated by pretreatment with TTX or propranolol, given at a dose which does not have anesthetic properties (10"6 M). Therefore, enteric noradrenergic neurons are implicated in mediating the release of 5-HT. 3. Enteric Neurons. Treatment of either dog (Rack6 and Schworer, 1992) or pig (Racke and Schworer, 1992; Racke and Schworer, 1993) jejunal tissue with hexamethonium or TTX inhibits spontaneous release of 5-HT. Simultaneous application of these drugs is not additive (Racke and Schworer, 1992). 35 4. Second Messengers: Calcium. Calcium free media decreases spontaneous release of 5-HT from the dog and pig jejunum (Rack£ and Schworer, 1993; Rack£ and Schworer, 1992). Various calcium channel (L and N type) blockers also decrease the spontaneous release of 5-HT from the pig jejunum (Racke and Schworer, 1993). Calcium appears to play an important role in promoting release of 5-HT from the gut. 5. Summary. Evidence suggests that adrenergic receptors in the rat (Pettersson, 1979) and muscarinic receptors in the dog (Rack6 and Schworer, 1992; Racke and Schworer, 1993) are located at the level of the EC cell. When not activated by an agonist, the enteric nervous system, stimulates release of 5-HT (Rack6 and Schworer, 1992; Rack6 and Schworer, 1993; Anderson et al. 1991). In contrast, when activated by nicotinic agonists or electrical stimulation, the enteric nervous system can either enhance [i.e. dog (Rackd and Schworer, 1993)] or inhibit [i.e. pig (Racke and Schworer, 1992; Racke and Schworer, 1993)] release of 5-HT.
VIII. Summary: Release of Serotonin A. Release into the Lumen 1. Activation of Chemoreceptors. Constituents of a meal can effect the release of 5-HT from the duodenal mucosa. Whether this release mechanism involves exocytosis of granules at the apical surface of the EC cell (Ahlman and 36 Dahlstrom, 1983) or diffusion of basolateral release remains to be determined. Release of 5-HT in response to acid in the duodenum is vagally-dependent in the dog (Jaffe et al. 1977). In vitro studies in the rabbit implicate the involvement of both cholinergic and 0 adrenergic activation (Kellum et al. 1983). Hence, both parasympathetic and sympathetic mechanisms may be involved. 2. Central Vagal Control. The findings of in vitro studies correlate well with studies in the intact animal suggesting that other events, in addition to activation of cholinergic receptors, can participate in mediating the release of 5-HT into the lumen (Cho et al. 1985; Stephens and Tach6, 1989). B. Release into the Portal Circulation 1. Activation of Chemoreceptors. Constituents of a meal also promote release of 5-HT (Martin et al. 1989; Ferrara et al. 1987; Jaffe et al. 1977; Drapanas et al. 1962). Nonadrenergic and noncholinergic neurons of the enteric nervous system are implicated in mediating this release (Martin et al. 1989). 2. Central Vagal Control. Vagal stimulation increases the concentration of 5-HT appearing in the portal circulation (Pettersson, 1979; Gronstad et al. 1987; Ahlman and Dahlstrom, 1983; Siepler et al. 1980; Tobe et al. 1976). 37 In the rat, the role of the adrenal gland in mediating release of 5-HT has not been investigated. 3. Enteric Neurons. The relative state of activation of the enteric nervous system is significant in determining its involvement in mediating release of 5-HT from the gut. The enteric nervous system may release other neurotransmitters, in addition to VIP and acetylcholine, which modulate release of 5-HT.
IX. Effect of Serotonin on Gastrointestinal Function A. Exogenous Serotonin 1. Gastric Acid Secretion a. in Vivo Studies. Many reports confirm that 5-HT inhibits gastric acid secretion in rats (Furman and Waton, 1989; Cho and Ogle, 1986b; Thompson, 1977; Evans and Gidda, 1993; Tsukamoto et al. 1991), cats (Furman and Waton, 1989; Thompson, 1977), guinea pigs (Thompson, 1977), dogs (Beck, 1986b; Beck, 1985; Jaffe et al. 1977; Bech and Anderson, 1985; Furman and Waton, 1989; Johansen and Bech, 1991; Thompson, 1977) and man (Furman and Waton, 1989; Thompson, 1977). Specifically, in the rat, 5-HT has no effect on basal (Cho and Ogle, 1986b; Evans and Gidda, 1993) but inhibits pentagastrin- (Cho and Ogle, 1986b), histamine- (Cho and Ogle, 1986b; Thompson, 1977), methacholine- (Cho and Ogle, 1986b) and vagally- (Tsukamoto et al. 1991) stimulated gastric acid secretion. The mechanism of inhibition is independent of 38 the vagus (Cho and Ogle, 1986b). The role of prostaglandins (Cho and Ogle, 1986b; Canfield and Spencer, 1983; Bech, 1988) in mediating inhibition of acid secretion by 5-HT is unclear. Concerning the 5-HT receptor mediating the inhibition of
acid secretion, antagonists at 5-HT2 or 5-HT3 receptors do not prevent inhibition of acid secretion by systemic 5-HT (Evans
and Gidda, 1993). The nonselective 5-HT1A agonist, 8-OH-DPAT, inhibits acid secretion in a vagally-stimulated pylorus- ligated model (Evans and Gidda, 1993). Taken together, the studies suggest that inhibition of acid secretion by 5-HT may be mediated through a receptor of the 5-HT, family. The receptor mediating inhibition of acid secretion in the rat is discussed in detail in Chapter II. Early studies by Black et al. demonstrate the inhibitory effect of 5-HT on stimulated gastric acid secretion in dogs (Black et al. 1958). In other studies on dogs equipped with gastric fistula, systemic administration of 5-HT does not effect basal (Thompson, 1977), but inhibits acid secretion stimulated by pentagastrin (Beck, 1985; Jaffe et al. 1977; Bech and Anderson, 1985; Johansen and Bech, 1991; Bech, 1988), bethanechol (Beck, 1986b; Bech and Anderson, 1985; Bech, 1988) or urecholine (Thompson, 1977), with conflicting reports for histamine (Beck, 1986a; Thompson, 1977). Pretreatment with methysergide reverses 5-HT-induced inhibition of pentagastrin- (Beck, 1985; Bech and Anderson, 1985; Bech, 1988), histamine- (Beck, 1986a), or bethanechol- (Bech, 1988) with a single 39 report of methysergide-insensitive (equal doses) (Beck, 1986b) effects of 5-HT on bethanechol-stimulated output. Pretreatment with the nonspecific 5-HT, and p adrenergic antagonist, propranolol reverses 5-HT-induced inhibition of pentagastrin- (Beck, 1985; Bech and Anderson, 1985; Bech, 1988), histamine- (Beck, 1986a), or bethanechol- (Beck, 1986b; Bech, 1988) stimulated gastric acid secretion. Inhibition of pentagastrin-stimulated, but not bethanechol-or histamine- stimulated, gastric acid secretion is reversed by pretreatment
with renzapride [5-HT1p and 5-HT3 receptor antagonist and 5-HT4 receptor agonist, (Johansen and Bech, 1991)]. In this case, reversal of inhibition by 5-HT may be a consequence of
enhanced acetylcholine release by the 5-HT4 agonist activity of renzapride (Sanger, 1987; Clarke et al. 1989; Gullikson et al. 1991; Schiavone et al. 1990). Further studies with 5-HT
antagonists suggest that 5-HTz receptors are not involved (Bech, 1988). These studies taken together do not implicate
a role for 5-HT2, 5-HT3 or 5-HT4 receptors in mediating inhibition of acid secretion by 5-HT in the dog. b* In Vitro Studies. In rat in vitro studies, 5-HT inhibits gastric acid secretion stimulated by pentagastrin (Cho and Ogle, 1986b; Canfield and Spencer, 1983), histamine (Cho and Ogle, 1986b; Hsu et al. 1991; Canfield and Spencer, 1983), isoprenaline [(Canfield and Spencer, 1983), stimulates acid secretion by an 'atypical' 0 adrenoceptor], or methacholine (Cho and Ogle, 1986b) without 40 affecting acid secretion stimulated by bethanechol (Canfield and Spencer, 1983), cAMP (Canfield and Spencer, 1983) , or ICI 63197 [phosphodiesterase inhibitor, (Canfield and Spencer, 1983)]. Methysergide reverses 5-HT-induced inhibition of histamine- (Hsu et al. 1991), pentagastrin- (Canfield and Spencer, 1983) or isoprenaline- (Canfield and Spencer, 1983) stimulated gastric acid secretion. Propranolol is ineffective in reversing 5-HT-induced inhibition of acid secretion (Cho and Ogle, 1986b; Hsu et al. 1991). In addition, either indomethacin or ibuprofen, when given at a dose which significantly decreases PGE2 levels in the stomach mucosa, does not prevent inhibition of acid secretion by exogenous 5- HT (Cho and Ogle, 1986b). Another report suggests that prostaglandins may be involved (Canfield and Spencer, 1983). The role of endogenous prostaglandins in mediating the 5-HT- induced inhibition of acid secretion is unclear. In summary, the 5-HT-induced inhibition of stimulated gastric acid secretion in vitro is dependent on the secretogogue. Although the antagonists used exhibit poor selectivity, the earlier work suggests either a 5-HT, or 5-HT2 receptor mediates the inhibitory effect of 5-HT in the rat. In addition, in vitro application of 5-HT to the mucosal surface does not inhibit pentagastrin-stimulated gastric acid secretion (Canfield and Spencer, 1983). 5-HT in the lumen does not inhibit acid secretion. The mechanism of action of 41 5-HT to inhibit acid secretion is discussed in detail in Chapter III. 2. Gastrointestinal Motility Modulation of GI motility by 5-HT appears to be mediated, in part, by TTX- sensitive enteric neurons (Mizutani et al. 1992; Meulemans et al. 1993). As a result, the following pharmacologic studies primarily address 5-HT receptors on myenteric neurons. 7 a. In Vivo Studies. Systemically applied 5-HT dose-dependently increases intragastric pressure in spinal rats (Dhasmana et al. 1992). This change in intragastric
pressure is blocked by pretreatment with the general 5-HT1/2 receptor antagonists, methysergide and methiothepin, but not
by pretreatment with antagonists specific for 5-HT2 or 5-HT3 receptor subtypes. The 5-HT, agonists, RU 24969 or 5- carboxamidotryptamine, mimic the response to 5-HT. Based on drug efficacies, 5-HT,,-like receptors mediate the increase in intragastric pressure by 5-HT (Dhasmana et al. 1992). Serotonin also decreases pyloric pressure in the rat. This effect of 5-HT is tropisetron-sensitive indicating
involvement of 5-HT3 receptors (Yoshioka et al. 1990).
Receptors of the 5-HT, and 5-HT3 families appear to be involved in coordinating gastropyloric activity in the rat.
Administration of the selective 5-HT3 agonist 2-methyl 5- HT also enhances contractions of the dog jejunum in vivo (Mizutani et al. 1992). These contractions are sensitive to close intra-arterial (i.a.) administration of tropisetron, 42 hexamethonium, TTX or atropine (Hizutani et al. 1992). Enteric cholinergic interneurons are intimately involved in generating peristaltic motility patterns.
Motility of the GI tract is greatly enhanced by 5-HT4
receptor activation. The 5-HT4 agonists renzapride or SC- 49518 stimulate motility in the dog antrum (Yoshida et al. 1991; Hopkinson et al. 1989; Gullikson et al. 1991; Gullikson et al. 1993), duodenum (Yoshida et al. 1991; Gullikson et al. 1993), jejunum (Yoshida et al. 1991; Gullikson et al. 1991; Gullikson et al. 1993), and colon (Yoshida et al. 1991). This increase in motility is blocked by atropine (Yoshida et al.
1991). 5-HT, acting through 5-HT4 receptors, may enhance release of acetylcholine from interneurons and thereby stimulate GI motility. Serotonin decreases transit time in the mouse (Jacoby et al. 1991). This effect is not mimicked by agonists selective for 5-HT,, 5-HT2 or 5-HT3 receptor subtypes (Jacoby et al.
1991). The effect of 5-HT4 agonists was not tested; however, studies in dog (Yoshida et al. 1991) and guinea pig (Kilbinger and Wolf, 1992) suggest that 5-HT4 receptor activation increases colonic (Yoshida et al. 1991) and ileal (Kilbinger and Wolf, 1992) motility, respectively. 5-HT4 receptors may also mediate the increase in colonic motility by 5-HT in the mouse. In the ferret, close i.a. 5-HT increases contractile activity in the antrum and small intestine (Blackshaw and 43 Grundy, 1993b). The corpus either biphasically contracts or relaxes (Blackshaw and Grundy, 1993b). This illustrates that the response to 5-HT depends on the segment of gut investigated. Predominantly, the gut contracts in response to 5-HT. b. In Vitro Studies. Early work by Biilbring et al. utilized isolated guinea pig small intestinal loops to determine the effect of 5-HT on peristalsis (Biilbring and Lin, 1958). When the intestinal lumen was exposed to 5-HT, the peristaltic reflex was stimulated. When 5-HT was applied to the intestinal serosa, the peristaltic reflex was inhibited. The action of 5-HT was different depending on its access to the intestinal segment. Serotonin may be involved in the nonadrenergic- noncholinergic vagal relaxation of the stomach (Biilbring and Gershon, 1967). The guinea pig stomach in an organ bath relaxes in response to close i.a. 5-HT (Meulemans et al. 1993). These relaxations are prevented by pretreatment with the 5-HT1/2 antagonists xnethiothepin or metergoline, TTX or the nitric oxide synthesis inhibitor NG-nitro-L-arginine (Meulemans et al. 1993). The data suggest 5-HT induces relaxation of the guinea pig stomach through nitric oxide release from nonadrenergic-noncholinergic enteric neurons. Guinea pig ileum with the longitudinal muscle removed (to focus on receptors of the myenteric plexus) is used to determine the effect of 5-HT on contractility. Either 5-HT 44 (Butler et al. 1988; Butler et al. 1990; Gullikson et al.
1993), 2-methyl 5-HT [5-HT3 agonist, (Butler et al. 1988;
Butler et al.1990)1. renzapride [5-HT4 agonist and 5-HT3 antagonist, (Sanger, 1987)], zacopride [5-HT4agonist and 5-HT3 antagonist (Gidda et al. 1988)] or SC-49518 [5-HT4 agonist (Gullikson et al. 1993)] induces dose-dependent contractions of the ileum. The 5-HT3 antagonists (Butler et al. 1988; Butler et al. 1990; Gullikson et al. 1993) shift the dose response curve to the right whereas antagonists at the 5-HT1/2
(Butler et al. 1990; Gullikson et al. 1993), 5-HT2 (Butler et al. 1990) or 5-HT4 (Sanger, 1987) receptors are ineffective. Contractions are inhibited by atropine (Butler et al. 1990) or by desensitization of 5-HT receptors by high concentrations of 5-HT (Sanger, 1987), but not by hexamethonium, phentolamine, or propranolol (Sanger, 1987). The data suggests that, in the guinea pig, activation of a muscarinic receptor is the final mediator of the contraction elicited by 5-HT and that 5-HT3 and 5-HTA receptors are closely linked in modulating motility.
The individual roles of the 5-HT3 and 5-HTA receptors will be clarified as more selective 5-HT4 agonists and antagonists are generated. A recent report suggests that 5-HT stimulates contraction of both the stomach and the jejunum of the dog (de Ridder and Schuurkes, 1993; Hopkinson et al. 1989). Using longitudinal muscle/myenteric plexus preparations from stomach, the receptor mediated contraction by 5-HT does not conform to 5- 45
HTV 5-HT2, 5-HTJ, or 5-HT4 subtypes (de Ridder and Schuurkes, 1993). The only antagonist effective in this model was 1- naphthylpiperazine (de Ridder and Schuurkes, 1993), an antagonist of the 5-HT2B rat stomach fundus receptor (Cohen and Flundzinzki, 1987; Cohen et al. 1992). Since the effect of 5- HT (relaxation vs. contraction) and the antagonist profile of the rat and dog stomach are different, another 5-HT receptor subtype [mediates the effect of 5-HT in the dog (de Ridder and Schuurkes, 1993). 3. Gastric Emptying. Renzapride or SC-49518 enhances gastric emptying of both liquid or solid meals in the dog (Gullikson et al. 1991; Gullikson et al. 1993). In addition to stimulating GI motility (Yoshida et al. 1991;
Gullikson et al. 1991; Gullikson et al. 1993), 5-HT4 agonists may also enhance gastric emptying. 4. Gastrointestinal Blood Flow. Serotonin decreases gastric (Kitajima et al. 1991; Hashizume et al. 1978; Wong et al. 1990) and duodenal (Tsukamoto et al. 1991) mucosal blood flow in the rat as measured by the hydrogen gas clearance (Kitajima et al. 1991; Hashizume et al. 1978; Tsukamoto et al. 1991) or laser doppler (Wong et al. 1990) methods. Pretreatment by oral gavage with naftidrofuryl oxalate, a drug used clinically as a vasodilator, protects against 5-HT- induced ischemia (Salim, 1990). 5-HT perfused through the lumen of the cat intestine increases mucosal blood flow to the muscle layer of the 46 intestine as measured by the clearance of radiolabeled microspheres (Gronstad et al. 1986). Blood flow to the mucosal level also increases, but not significantly. The increase by 5-HT is prevented by either muscarinic blockade or local anesthesia but not by the 5-HT2 antagonist ketanserin (Gronstad et al. 1986). Cholinergic neurons may respond to 5- HT in the intestinal lumen by altering blood flow. 5-HT initially increases, then 15 minutes later decreases, gastric mucosal blood flow in the dog as measured by clearance of neutral red dye (Bech and Anderson, 1985). The long lasting effect of 5-HT is countered by either propranolol or methysergide. The results are confounded by the ability of propranolol or methysergide alone to enhance gastric mucosal blood (Bech and Anderson, 1985). Neither the receptor subtype nor mechanism mediating this effect by 5-HT has been characterized. 5. Gastric and Duodenal Lesions. 5-HT induces gastric (Tsukamoto et al. 1988; Takeuchi et al. 1986; Kitajima et al. 1991; Miyata et al. 1991; Hashizume et al. 1978; Ozdemir and Zimmermann, 1971; Wong et al. 1990) but not duodenal (Tsukamoto et al. 1991) erosion formation. The mechanism is unknown but 5-HT, at the doses used, also significantly reduces gastric mucosal blood flow (Salim, 1990; Tsukamoto et al. 1988; Kitajima et al. 1991; Wong et al. 1990) and glandular mucus content (Wong et al. 1990). Adrenalectomy exacerbates, but methysergide protects, the gastric mucosa 47 from 5-HT-induced lesions (Ozdemir and Zimmermann, 1971). Decreases in gastric mucosal blood flow and/or mucous formation may mediate gastric erosion formation by 5-HT. 6. Vagal Activity a. In Vivo Studies. In the rat, intravenous 5-
HT or 2-methyl 5-HT (5-HT3 agonist) increases gastric vagal efferent nerve activity recorded at the level of the esophago gastric ^junction (Yoshioka et al. 1990). The dose responses curves of 5-HT and 2-methyl-5-HT are shifted to the right by the 5-HTj antagonist, tropisetron (Yoshioka et al. 1990).
This suggests 5-HT3 receptors are located on vagal afferents. This hypothesis is confirmed by in vitro pharmacologic studies [section IX.A.6.b. (Ireland and Tyers, 1987c; Round and Wallis, 1986; Ireland et al. 1987b; Butler et al. 1990)]. 5-
HT may modulate vago-vagal reflexes through 5-HT3 receptors on vagal afferents. Intravenous 5-HT also increases cervical vagal afferent nerve activity in the rat (Yoshioka et al. 1992). The dose- response curve of 5-HT is shifted to the right by the 5-HT2 antagonist ketanserin or the 5-HT3 antagonist granisetron
(Yoshioka et al. 1992). Agonists at 5-HT2 (a-methyl 5-HT) and
5-HT3 (2-methyl 5-HT) receptors dose-dependently increase cervical vagal activity (Yoshioka et al. 1992). The agonists' effects are reversed by appropriate antagonist pretreatment (Yoshioka et al. 1992). The study suggests that, in addition to 5-HT3 receptors (Ireland and Tyers, 1987c; Round and 48 Wallis, 1986; Ireland et al. 1987b; Butler et al. 1990), vagal afferent nerves may also respond to 5-HT2 receptor activation. The location of the 5-HT responsive vagal afferents terminals in this study cannot be discerned as 5-HT is given intravenously and afferent nerve activity is recorded at the cervical vagus. In the ferret, vagal mucosal (Blackshaw and Grundy, 1993a), 7 but not vagal mechanosensitive (de Ridder and Schuurkes, 1993), afferents are responsive to 5-HT. The response of mucosal (Blackshaw and Grundy, 1993a) afferents to
5-HT is dose-dependent, blocked by treatment with the 5-HT3 antagonist granisetron, and insensitive to atropine and hexamethonium (Blackshaw and Grundy, 1993a). These fibers which respond to 5-HT also are activated by mucosal administration of acid or hypertonic saline (Blackshaw and Grundy, 1993a) suggesting that these receptor endings are in position to sample the luminal environment. In contrast, vagal mechanoafferents respond to 5-HT-induced changes in GI pressure, not 5-HT itself (de Ridder and Schuurkes, 1993). Therefore, activation of 5-HT receptors on peripheral endings of vagal afferents may modulate vago-vagal reflexes. b. In Vitro Studies. 5-HT enhances vagal nerve activity in vitro (Ireland and Tyers, 1987c; Round and Wallis, 1986; Ireland et al. 1987b; Butler et al. 1990). Segments of rat cervical vagus nerve dose-dependently depolarized after application of 5-HT (Ireland and Tyers, 1987c; Round and 49 Wallis, 1986; Ireland et al. 1987b; Butler et al. 1990). The
5-HT3 antagonists metoclopramide (Ireland and Tyers, 1987c; Ireland et al. 1987b; Ireland, 1987a; Butler et al. 1988), ondansetron (Butler et al. 1988), m-chlorophenylbiguanide (Ireland and Tyers, 1987c), MDL 72222 (Ireland and Tyers, 1987c), tropisetron (Ireland and Tyers, 1987c), tropacaine (Ireland and Tyers, 1987c), or quipazine (Ireland and Tyers, 1987c) produce a rightward shift of the dose response curve indicating that these drugs display antagonistic properties at the 5-HT3 receptor. Segments of the guinea pig vagus nerve also display similar behavior in response to 5-HT and 5-HT3 antagonists (Butler et al. 1990). Other experiments were performed using the sucrose gap technique on the rabbit nodose ganglia with a portion of the vagus nerve still intact (Round and Wallis, 1986; Round and Wallis, 1987). A dose response curve to 5-HT (Round and Wallis, 1986; Round and Wallis, 1987) with antagonism by MDL 72222 (Round and Wallis, 1987) , tropisetron (Round and Wallis, 1986), metoclopramide (Round and Wallis, 1987) , quipazine (Round and Wallis, 1987) or m-chlorophenylpiperazine (Round and Wallis, 1987) are obtained from this preparation. 5-HT3 receptors are present on the vagus nerve and nodose ganglia of the rat, guinea pig and rabbit. B. Endogenous Serotonin. Exogenous application of 5-HT indicates possible physiologic actions of the amine. A key question is the role of endogenous 5-HT in modulating 50 physiologic processes. Serotonin released into the gastric lumen and or appearing in the portal circulation will be discussed individually. 1. Gastric Acid Secretion a. Endogenous 5-HT in the Gastric Lumen. The elevated concentration of 5-HT in the lumen of rats treated with a chemical vagal stimulant is attenuated by adrenalectomy (Stephens, 1991). This reduction in 5-HT concentration in the gastric lumen is not correlated with a change in acid secretion (Stephens, 1991). Serotonin released into the gastric lumen does not inhibit stimulated acid secretion. b. Endogenous 5-HT in the Portal Circulation. Pretreatment with methysergide [cat (Izzat and Waton, 1987)], cyproheptadine [cat (Izzat and Waton, 1987)], or renzapride [dog (Johansen and Bech, 1991)] prevents 5-HT-induced decrease in pentagastrin-stimulated acid secretion. Due to the nonselective properties of the pharmacologic agents utilized, the endocrine role of 5-HT in mediating inhibition of pentagastrin-stimulated acid secretion should be further investigated. c. Depletion of Endogenous 5-HT. Vagal stimulation releases 5-HT into the gastric lumen (Cho et al. 1985; Stephens, 1991; Stephens and Tach6, 1989). Depletion of 5-HT stores by peripheral pretreatment with p-chlorophenylalanine (PCPA), an inhibitor of the rate- limiting enzyme in the synthesis of 5-HT, tryptophan 51 hydroxylase, reduces the concentration of 5-HT in whole stomach [65%] and brain [95%] tissue and decreases luminal concentrations of 5-HT [57%] (Stephens et al. 1989; Stephens et al. 1990). Moreover, PCPA pretreatment potentiates the gastric acid secretory response [43%] to vagal stimulation (Stephens et al. 1990). Decreased release of 5-HT is correlated with enhanced gastric acid secretion. The data suggest? that peripheral 5-HT exerts an inhibitory tone on vagally-stimulated gastric acid secretion. Recent work has suggested that 5-HT in the gastric lumen has no effect on stimulated gastric acid secretion (Stephens, 1991). However, since vagal stimulation can release 5-HT into the portal circulation (Pettersson, 1979; Gronstad et al. 1987; Siepler et al. 1980; Tobe et al. 1976; Gronstad et al. 1988a), it is plausible that this is the component modulating the vagally driven response. This is discussed in detail in Chapter III which investigates the source of 5-HT mediating inhibition of acid secretion. 2. Gastrointestinal Motility a. Endogenous 5-HT in Mucosal Mast Cells. Degranulation of MMC by BrX-537A inhibits duodenal and jejunal spiking activity but is followed by an abrupt return to normalcy (Bueno et al. 1991). Methysergide or zacopride decreases the time of inhibition in both the duodenum and jejunum. Ketanserin is effective in the duodenum whereas tropisetron is effective only in the jejunum. Serotonin 52 released from degranulating MMC disrupts duodenal and jejunal myoelectric activity through disparate 5-HT receptors (Bueno et al. 1991). b. Endogenous 5-HT and Enterio Neurons 1) in Vivo Studies. Various 5-HT antagonists have been utilized to determine the effect of 5-HT on GI motility. Administration of 5-HT3 antagonists to the guinea pig (Gidda et al. 1988) or dog (Gullikson et al. 1991; Yoshida et al. 1991) either has no effect (Yoshida et al. 1991; Gidda et al. 1988; Gullikson et al. 1991) or inhibits (Yoshida et al. 1991) motility of the GI tract. In humans, 5-
HT3 antagonists decrease spontaneous migrating myoelectric complex generation in the antrum but not in the duodenojejunal segment (Wilmer et al. 1993). Receptors of the 5-HT3 family may be involved in peristalsis. 2) In vitro studies. In the guinea pig,
5-HT3 antagonists enhance (Buchheit et al. 1985) , whereas 5-HT4 antagonists inhibit (Gidda et al. 1988), electrically generated contractions of the antrum and ileum, respectively.
The 5-HT3 and 5-HT4 receptors interplay to modulate contractile activity. c. Depletion of Endogenous 5-HT. In rats pretreated with PCPA, chemical vagal stimulation produces a 200% increase in gastric contractility as compared to vehicle treated rats (Stephens et al. 1990). In contrast, gastric motility stimulated by carbachol is not altered in PCPA 53 pretreated rats (Stephens et al. 1990). Selective depletion of 5-HT in the brain does not influence vagally-mediated motility changes. Serotonin in the gut produces an inhibitory influence on gastric contractility (Stephens et al. 1990). 3. Gastric Emptying a. Enteric Neurons. The involvement of various serotonin receptor subtypes in modulating gastric emptying is conflicting. Treatment of rats (Gidda et al. 1988; Schiavone et al. 1990), guinea pigs (Costall et al. 1987; Buchheit et al. 1985), or dogs (Gullikson et al. 1991) with granisetron (Costall et al. 1987), zacopride (Gidda et al. 1988; Schiavone et al. 1990) or tropisetron (Buchheit et al. 1985; Gidda et al. 1988; Gullikson et al. 1991) enhances gastric emptying.
Due to the affinity of these drugs for both 5-HT3 and 5-HT4 receptors, the subtype preferentially activated in each animal at the given dose is unclear. Nevertheless, 5-HT receptors on enteric neurons can influence gastric emptying. 4. Vagal Activity. Close i.a. administration of the
5-HT3 antagonist granisetron in the ferret greatly reduces or abolishes spontaneous discharge of mucosal vagal afferents, suggesting endogenous 5-HT activates vagal afferent discharge at rest (Blackshaw and Grundy, 1993a). 5. Absorption of Nutrients a. Depletion of Endogenous 5-HT. Gastrointestinal stores of 5-HT can also be depleted by 2,4- diamino-6-hydroxypyrimidine (DAHP), an inhibitor of 54 tetrahydrobiopterin (BHA), a cofactor for tryptophan hydroxylase, the rate-limiting enzyme in the synthesis of 5- HT. In the mouse, DAHP decreases EC cell 5-HT content in the duodenum and colon but not the stomach and ileum (Kobayashi et al. 1991). Reduction of 5-HT by DAHP treatment is correlated with impaired ability of the intestine to uptake glucose and diarrhea. For example, DAHP treatment also reduces norepinephrine and dopamine levels of the gut. Inhibition of
BH4 synthesis is a novel, but nonspecific, method to gain insight into the effect of endogenous 5-HT on GI function.
X. Conclusion This review examined the phenomena of release of 5-HT from the GI tract, covering the mechanisms regulating release into the luminal and hepatic portal compartments, and physiologic actions of 5-HT on the GI tract. The quantitation of concomitant release of 5-HT into the portal venous system and the lumen of the gut would be useful in clarifying in vivo release mechanisms. Neither the relative contribution of enteroendocrine cells, enteric neurons or MMC to the enhanced 5-HT levels has been determined. Nor is the relative contribution of the separate divisions of the GI tract (i.e. stomach vs. intestine) to release of 5-HT appearing in the lumen or portal circulation known. It is possible that separate release mechanisms exist for the different cellular and regional pools of 5-HT in the gut. These are a few of the 55 questions to be investigated concerning release of 5-HT into the portal circulation. Attempts to extrapolate physiologic actions of 5-HT from the effects of systemically administered 5-HT should be viewed with caution. 5-HT may not reach the desired site of action due to reuptake by the platelets, destruction by the liver and lungs, degradation by enzymes or barriers produced by unstirred layers in the gut. Uncertainties exist regarding the concentration of amine delivered to the receptor site and in location of the final effector site. Better methods of predicting the effect of endogenous 5-HT on GI function include close i.a. application of the amine selectively into the gastric circulation and utilization of the more specific and potent 5-HT antagonists that have become recently available. Many previous studies investigating the 5-HT receptor subtype mediating physiological responses to 5-HT in the gut were hampered by the nonselectivity of the agonists and antagonists. The studies described in this dissertation focus on gastric serotonin, a relatively unexplored pool of GI serotonin. The aims were: 1) to determine if vagal stimulation by the TRH analogue, RX77368, concomitantly enhanced release of endogenous 5-HT into the gastric lumen and portal circulation, 2) the ability of these pools to influence acid secretion, 3) the 5-HT receptor subtype mediating inhibition of gastric acid secretion and 4) the mechanism by 56 which 5-HT's physiologic action was mediated. Finally, the functional relevance of endogenous 5-HT in producing an inhibitory tone on vagally-stimulated gastric acid secretion was investigated. CHAPTER II
Serotonin Receptor subtype Mediating Inhibition of Gastrio Acid secretion
I. Overview Studies performed in the rat (Cho and Ogle, 1986b; Hsu et al. 1991; Canfield and Spencer, 1983) and in the dog (Beck, 1986b; Beck, 1986a; Black et al. 1958; Bech, 1988) demonstrate the ability of systemic 5-HT to inhibit stimulated gastric acid secretion. Depletion of endogenous gastrointestinal stores of 5-HT in the rat is correlated with potentiation of vagally stimulated gastric acid secretion and motility (Stephens et al. 1989). The receptor mediating this 5-HT- induced inhibitory response in the rat has not been clearly delineated. These studies described below utilize various 5- HT antagonists and agonists to test the hypothesis that intravenous serotonin inhibits acid secretion by acting through a serotonin receptor.
II. Hypothesis: Intravenous Serotonin Inhibits Acid Secretion by Acting through a Serotonin Receptor.
57 58 III. Methods A. Animals. Male Sprague-Dawley rats (180-300 g; Harlan Industries, Indianapolis, IN) were maintained ad libitum on Purina Laboratory Chow and tap water. They were housed under controlled conditions of temperature (22 + 1 "C) and illumination (6 a.m. to 6 p.m.). All experiments were performed in animals deprived of food but not water 18-24 hrs before the experiment. B. Measurement of Gastric Acid Secretion. Gastric secretion was collected from urethane anesthetized rats (1.5 g/kg, i.p.). The abdomen was opened by a midline incision, the pylorus exposed and ligated, and a double lumen cannula was placed through a small incision in the forestomach. Gastric secretion was collected by flushing the lumen with a 5-ml bolus of normal saline followed by a 5-ml bolus of air at 10-min intervals. Acid output was determined by titration of the flushed perfusate with 0.01 N NaOH to pH 7.0 on an automatic titrator (Radiometer, Copenhagen, Denmark). C. Cannulation of the splenic Artery. In these studies, either vehicle, 5-HT antagonists or 5-HT agonists were selectively infused into the gastric circulation through the splenic artery (FIGURE 1). The spleen was placed on moist gauze and the splenic artery was identified utilizing a stereomicroscope and was surgically isolated from the vein. The cannula (30-gauge needle attached to PE-10 tubing) assembly was inserted into the artery, tied and glued in place 59 with cyanoacrylate. The vehicle, antagonist or agonist was infused in a volume of 0.1 ml over the time course of 1 min. 60
^sssslsia*
FIGURE 1: Anatomy of the celiac axis of the rat. The common celiac artery branches off the aorta. Shortly thereafter, the celiac artery divides into three branches: hepatic, left gastric and splenic arteries. The splenic artery was cannulated at|. Drug was infused against the flow of blood in the direction of the arrows (•<-) . 61 D. Protocol for Antagonist Studies. After three 10-min basal periods, pentagastrin (24 jug/kg/hr, i.v.) was infused via the femoral vein. After six 10-min periods, the animal was pretreated with either vehicle or a 5-HT antagonist given close i.a. into the gastric circulation via the splenic artery. After three more periods, serotonin was infused into the femoral vein at a dose known to produce a 60% reduction in pentagastrin-stimulated gastric acid secretion. Acid secretions were collected for 6 additional 10-min periods. E. Protocol for Agonist Studies. After three 10-min basal periods, pentagastrin (24 /ig/kg/hr, i.v.) was infused via the femoral vein. After nine 10-min periods, either vehicle, 5-HT or a 5-HT agonist was injected close i.a. Acid secretions were collected for 6 additional 10-min periods. P. Drugs. 3-Tropanyl-indole-3-carboxylate (tropisetron; Research Biochemicals Inc., MA) was dissolved in 0.05% absolute ethanol and distilled water. Renzapride HC1 (Beecham Pharmaceutical Co., Harlow, UK), serotonin maleate (Sigma, St. Louis, MO), methiothepin mesylate, (+)-8-hydroxy-2-(n- dipropylamino)tetralin HBr [8-OH-DPAT], (+)-1-(4-iodo-2,5- dimethoxyphenyl)-2-aminopropane HC1 [(±)-DOI; Research Biochemicals Inc., MA] and 5-carboxamidotryptamine maleate (5- CT; generously donated by Glaxo, U.K.) were dissolved in distilled water. Ritanserin, metergoline, methysergide maleate, spiperone HC1 and 1-(2-methoxyphenyl)-4-[4-(2- phthalimido)butyl]piperazine HBr [NAN-190; Research 62 Biochemicals Inc., MA] were suspended in 0.1% Tween 80. Pentagastrin (Ayerst Laboratories Inc. New York, NY) was diluted using normal saline to the appropriate concentration. 6. Drug Treatments for Antagonist Studies. Groups of animals were pretreated with various serotonergic antagonists (TABLE 4) , at doses established to reverse 5-HT-induced effects. The following antagonists and doses (/wnol/kg/0.1 ml) were utilized: metergoline: 1.1 (Dhasmana et al. 1992); methiothepin: 1.1 (Dhasmana et al. 1992; Pettibone and Pflueger, 1984; Gartside and Cowen, 1990); methysergide: 3.5 (Takeuchi et al. 1986); NAN-190; 13 and 19 (Moser, 1991); renzapride: 0.17 (Schiavone et al. 1990); ritanserin: 1.3 (Mastai et al. 1990) ; spiperone: 1.3 (Murphy and Zemlan, 1990; Gartside and Cowen, 1990); and tropisetron: 0.04 (Turconi et al. 1991). Animals were then treated via the femoral vein with 5-HT (3.5 jumol/kg) to attenuate pentagastrin-stimulated gastric acid secretion. H. Drug Treatments for Agonist Studies. Groups of animals were treated close i.a. with either vehicle, an effective dose of 5-HT (0.88 /imol/kg) or a 5-HT agonist (0.88 or 2.6 /imol/kg). The following agonists were utilized (TABLE 4): 5-CT, (±)-8-DPAT and (+)-D0I. I. statistics. To determine pretreatment effects, data are expressed as a percentage of the difference between the average acid secretion 30 min before and after pretreatment divided by the average acid secretion before pretreatment. 63 In the cases of treatment with 5-HT or 5-HT agonists intravenously or 5-HT or 5-HT agonists close i. a., data are expressed as a percentage of the difference between the average of the three acid secretory measurements taken 30 min before treatment and the largest incremental change in 10-min acid output during the 30-min period after treatment. In the antagonist studies, the responses to the different close i.a. vehicle pretreatments were not significantly different; therefore, these data were pooled (see FIGURE 3). The data were analyzed utilizing the one-way analysis of variance (ANOVA) with post-hoc Student's t test. Differences between groups were considered significant if P<0.05.
IV. Results A. Effect of intravenous serotonin on Acid Secretory Responses to Fentagastrin. 5-HT administration (3.5 /umol/kg, i.v.) produced an 80% inhibition in pentagastrin- (24 /zmol/kg/hr, i.v.) stimulated gastric acid secretion in the 30- min period after injection (FIGURE 2). The nadir of acid secretion occurred 20 min after 5-HT. Acid secretion remained inhibited the remainder of the experiment. 64
VEHICLE 5-HT N - 4 N - 4
VEH/5-HT
TIME (MIN)
FIGURE 2: Time course of the effect of intravenous 5-HT on pentagastrin-stimulated acid secretion. Pentagastrin (24 umol/kg/hr, i.v.) infusion began after three 10-min basal periods. After nine 10-min periods, the animals were 9iven either intravenous vehicle or 5-HT (3.5 /mol/kg, l.v.). The acid secretory response was measured for six additional 1 min periods. M represents the number of animals m each group. Data are expressed as mean ± SEM. 65 B. Ability of Local Gastric Infusion of Serotonin Antagonists to Reverse Serotonin-Induced Attenuation of Acid Secretion. To control for actions of the 5-HT antagonists at nongastric sites, close i.a. infusion of the drug, at doses established to reverse 5-HT-induced effects (see section XII.G.), was performed. Close i.a. infusion of tropisetron (0.04 /nmol/kg), renzapride (0.17 jtxmol/kg) , ritanserin (1.3 fxmol/kg), methysergide (3.5 jxmol/kg), methiothepin (l.l /xmol/kg), metergoline (l.l /xmol/kg), spiperone (1.3 /nmol/kg) or NAN-190 (13 ^mol/kg) had no effect alone on pentagastrin- stimulated gastric acid secretion. In contrast, close i.a. infusion of a higher dose of NAN-190 (19 Mmol/kg) acted as a partial agonist to inhibit pentagastrin-stimulated gastric acid secretion (FIGURE 3A). Gastric close i.a. pretreatment with tropisetron or renzapride was ineffective in reversing the inhibition by 5-HT (3.5 /imol/kg, i.v.). Ritanserin showed only a trend in reversing the inhibition of acid secretion by exogenous 5-HT (FIGURE 3B) when given at a dose shown functionally effective in the rat (Leysen et al. 1985a). In contrast, close i.a. pretreatment with methysergide, methiothepin, metergoline or spiperone reversed the 5-HT- induced inhibition by 69 to 98% (FIGURE 3B). The high dose of NAN-190 (19 jumol/kg) which demonstrated partial agonist properties (FIGURE 3A), was unable to prevent 5-HT-induced inhibition of acid secretion. 66
FIGURE 3: A. Effect of local gastric infusion of various 5-HT antagonists on pentagastrin-stimulated (24 ymol/kg/hr, i.v.) gastric acid secretion. Data are expressed as mean ± SEM; *P<0.05. B. Ability of local gastric infusion of various 5-HT antagonists to reverse the 5-HT-induced (3.5 /xmol/kg, i.v.) attenuation of pentagastrin-stimulated gastric acid secretion. See TABLE 4 for abbreviations of antagonists. The numbers in section A. represent the number of animals in each group. Data are expressed as mean + SEM; *P<0.05; **P<0.01. 67 N= 22 4 4 4 5 4 3 5 4 6
i TS^zst tVjfl*
20 r 10 -
-80 - -90 — V TP RZ RT NN NN MS MT MQ SP .04 .17 1.3 13 19 3.5 1.1 1.1 1.3 5-HT ANTAQONIST (MMOL/KQ, CLOSE I.A.) FIGURE 3 68
TABLE 4: Summary of the serotonin antagonists and agonists_used in these studies with their abbreviations and serotonin receptor subtype selectivity.
SBR0T0M1N AMWHXSTS | DRUG SUBTYPE 1
Tropisetroo TP 5-HT3/4
Renzapride RZ 5-HTJ/1P
Ritanserin RT 5-HT2
Mathysergide MS 5-HT1/2
Methiothepin MT S-HTV2
Metergoline MG 5-HT1/2
Spiperone SP 5"HTU .
NAN-190 NN 5-HT1A/U
SEROTONIN ^QPNTSTS DROQ *BBiur SDBT XPB 5-CT 5-HT, 5-carboxamidotryptamine
(±)-8-hydroxy-2-(n- 8-OH-DPAT 5"HT1A dipropylamino)tetralin
(±)-l-(4-iodo-2,5-dimethoxy- DO I 5-HT2 phenyl)-2-aminopropane 69 C. Effect of Serotonin Agonists on Pentagastrin- Stimulated Gastric Acid Secretion. Intravenous infusion of equimolar doses (3.5 jumol/kg) of 5-HT or 5-CT inhibited acid secretion maintained by pentagastrin infusion (TABLE 5). Moreover, close i.a. administration of 5-HT (0.88 ^mol/kg) inhibited acid secretion by 48% but in contrast, equimolar close i.a. 5-CT only slightly inhibited acid secretion (29%), the latter response was not significantly different from vehicle. However, a 3-fold higher dose (2.6 /mol/kg) of close i.a. 5-CT was effective to inhibit acid secretion (62%) (FIGURE 4). In contrast, close i.a. administration of the selective 5-HT1A agonist, 8-OH-DPAT, did not inhibit acid secretion; indeed the high dose (2.6 jLtmol/kg) stimulated acid secretion by 11% (FIGURE 4). Moreover, close i.a. DOI (0.88 or 2.6 jumol/kg) was ineffective to inhibit gastric acid secretion (FIGURE 4). 70
TABLE 5: Effect of intravenous infusion of either vehicle, 5- HT (3.5 jLimol/kg) or 5-CT (3.5 /xmol/kg) on pentagastrin- stimulated (24 /mol/kg/hr, i.v.) gastric acid secretion. N represents the number of animals in each group. Data are expressed as mean + SEM; **P<0.01. N % change in acid secretion vehicle 3 13.7+1.8 5-HT 5 -56.6 + 11.9 ** 5-CT 5 -39.6 + 6.9 ** 71
20
-100 VEH 5HT 5CT DOI DPAT
FIGURE 4: Effect of close i.a. gastric infusion of either vehicle, 5-HT (0.88 jmol/kg) or various 5-HT agonists (0.88 or 2.6 |imol/kg) on pentagastrin-stimulated (24 jxinol/kg/hr, i.v.) gastric acid secretion. The numbers above the bars represent the number of animals in each group. Data are expressed as mean + SEM? *P<0.05? **P<0.01. 72 V. DISCUSSION The principle finding of this phase of the study was that 5-HT-induced inhibition of gastric acid secretion was mediated by a methysergide-, methiothepin-, metergoline- and spiperone- sensitive receptor. The ineffectiveness of both intravenous
(LePard and Stephens, 1993) and close i.a. tropisetron (5-HT3/4 antagonist) , renzapride (5-HT3/1p antagonist) , and ritanserin (S-HT^j,. antagonist) (FIGURE 3B) provide compelling evidence that 5-HT3, 5-HT1p and 5-HT2 receptors are not involved.
Although ritanserin [5-HT2 antagonist (Leysen, 1992)] pretreatment had a propensity to reverse the attenuation of acid secretion by intravenous 5-HT, a test of higher ritanserin doses was not justified due to the demonstrated ability of the administered dose to reverse 5-HT-induced effects (Mastai et al. 1990) and to antagonize binding to 5-
HT2 receptors (Leysen et al. 1985a).
A role of the recently characterized 5-HT4 receptor (Bockaert et al. 1992) in mediating inhibition of acid secretion is unlikely because of 1) the ineffectiveness of either intravenous (LePard and Stephens, 1993) or close i.a. tropisetron (resulting in probable micromolar concentrations in the gastric circulation) in reversing the 5-HT-induced effect (FIGURE 3B) and 2) the ineffectiveness of either intravenous (LePard and Stephens, 1993) or close i.a. renzapride [5-HT4 agonist, (Bockaert et al. 1992)] to alter gastric secretion in the 30-min pretreatment period before 5- 73 HT administration (FIGURE 3A). Hence, the antagonist data, taken together, suggest that the receptor mediating the inhibition of gastric acid secretion by 5-HT belongs to the 5-
HT1 family (Frazer et al. 1990). The data obtained from agonist studies also support the pharmacological profile of a receptor of the 5-HT, family (Frazer et al. 1990). Intravenous administration of equimolar 5-HT or 5-CT was effective in attenuating acid secretion (TABLE 5). When administered close i.a., however, a 3-fold higher concentration of 5-CT was needed to significantly inhibit acid secretion (FIGURE 4). The apparent reduced potency of 5-CT as compared to 5-HT may be a characteristic of the 5-HT receptor mediating the response, or may indicate that 5-HT produces inhibition of acid secretion by activating several subtypes of receptors. Moreover, the ineffectiveness of gastric close i.a. administration of DOI, a selective 5-HT2 agonist, is consistent with the conclusion that 5-HT2 receptors alone do not mediate the inhibitory effects of 5-HT on acid secretion. The gastric close i.a. administration of 5-HT agonists or antagonists in this study provides strong evidence for a gastric site of action. A recent study, using autoradiographic and in situ hybridization approaches, demonstrated the presence of high concentrations of 5-HT1A receptors in the rat stomach (Kirchgessner et al. 1993a). The ability of spiperone to reverse the 5-HT-induced inhibition of gastric acid secretion (FIGURE 3B) is supportive of a 5-HT1A 74 receptor-mediated response. However, several lines of evidence suggest that 5-HT1A receptors are not involved. NAN-
190, which possesses an affinity for 5-HT1A receptors one order of magnitude greater than spiperone (Glennon et al. 1988) , did not reverse 5-HT-induced attenuation of acid secretion (FIGURE 3B), but instead elicited a partial agonist response at a dose 10-fold greater than spiperone (FIGURE 3A). The partial agonist properties of NAN-190 have also been described in another study of acid secretion as well as other nonspecific effects (Evans and Gidda, 1993; Moser, 1991). Secondly, 5-CT displayed an apparent lower potency to inhibit acid secretion (FIGURE 4) as was demonstrated in a recent study of the effect of 5-HT on intraluminal pressure (Dhasmana et al. 1992).
Finally, spiperone, a 5-HT1A antagonist of nanomolar affinity (Leysen, 1992), was effective in antagonizing 5-HT's effect on acid secretion (FIGURE 3B) whereas 8-OH-DPAT, a high affinity
5-HT1A agonist (Miquel and Hamon, 1992), did not inhibit acid secretion (FIGURE 4). With respect to 5-HT,,-like receptors in the rat stomach, to date receptor binding studies have indicated only the presence of 5-HT1A (Kirchgessner et al. 1993a) and 5-HT1p (Branchek and Gershon, 1987; Kirchgessner et al. 1993a; Branchek et al. 1984; Branchek et al. 1988) receptors. The results of the present study preclude a role of 5-HT1A or 5-
HT1P receptors acting alone to mediate inhibition of acid output. The role of other 5-HT., subtypes in mediating 75 inhibition of acid secretion by 5-HT is suggested from this study. The effectiveness of spiperone, characterized as
relatively ineffective against the actions of 5-HT at 5-HT1B receptors, suggests that this subtype may not be involved. Novelty of the 5-HT receptor subtype involved in inhibiting gastric acid secretion can be inferred by the spiperone- sensitivity of this receptor, which is not apparently activated by 8-OH-DPAT. Although no evidence of gastrointestinal localization has been presented, criteria are being established for other receptors of the 5-HT, family that are activated by 5-CT (Humphrey, 1992). Taken together, results of the present study suggest that the receptor mediating inhibition of acid secretion is a S-H^-like receptor. Interestingly, the 5-HT receptor mediating attenuation of gastric acid secretion by exogenous 5-HT characterized herein is pharmacologically similar to the 5-HT receptor which mediates 5-HT-induced increases in intragastric pressure (Dhasmana et al. 1992). In both cases, 5-HT-induced events are sensitive to blockade by methysergide and methiothepin and
insensitive to blockade by 5-HT2 and 5-HT3 antagonists. Metergoline only prevented the action of the high dose of 5-HT on intragastric pressure (Dhasmana et al. 1992). The effectiveness of metergoline in our study, in contrast to the aforementioned study, might be explained by the more gastric- selective close i.a. administration of the antagonist in the 76 present work. In addition, in both studies, 5-CT was less potent than 5-HT. Hence, the 5-H^-like receptor in each case may comprise a similar subtype.
In previous work by others, the nonspecific 5-HT1/2 antagonist methysergide was characterized in the rat to reverse 5-HT-induced inhibition of pentagastrin- (Canfield and Spencer, 1983; Cho and Ogle, 1986a), histamine- (Hsu et al. 1991; Canfield and Spencer, 1983; Cho and Ogle, 1986a) and methacholine- (Kirchgessner et al. 1993b) stimulated gastric acid secretion. In vivo studies in the dog establish the ability of both methysergide (Beck, 1985; Bech, 1988) and renzapride (Johansen and Bech, 1991) to reverse the 5-HT- induced inhibition of pentagastrin-stimulated gastric acid secretion. The effectiveness of renzapride in the dog contrasts to the findings of the present study and suggests possible species differences.
The selective 5-HT1A agonist 8-OH-DPAT inhibits bethanechol- (Lyngso et al. 1992; Bech et al. 1992b) and histamine- (Lyngso et al. 1992; Bech et al. 1992b), but not pentagastrin- (Bech et al. 1992b), stimulated gastric acid secretion in the dog. In contrast to our findings, a recent report showed that systemic 8-OH-DPAT inhibited pentagastrin- stimulated gastric acid secretion in conscious rats with chronic gastric fistula (Evans and Gidda, 1993). Nevertheless, intravenous 8-OH-DPAT can act centrally to modulate autonomic outflow (Fozard et al. 1987; Gradin et al. 77 1985; Mccall et al. 1987). A central site of action may explain the disparity between intravenous vs. gastric close i.a. administration of 8-OH-DPAT. In our study, 8-OH-DPAT was given directly into the gastric circulation in a 100 til volume, markedly limiting its access to the central nervous system. In conclusion, evidence from our analysis of pharmacologic actions of several 5-HT antagonists and agonists suggests that 5-HT-induced inhibition of gastric acid secretion is mediated by receptors of the 5-HT, family. Aside from identification of the receptor subtype, important questions regarding serotonergic modulation of gastric function include the localization of receptors (gastric mucosa or enteric nervous system) and the role of released endogenous stores of 5-HT on integrated gastric function. CHAPTER III
Mechanism of Action of Intravenous Serotonin to Inhibit Gastric Secretory Function
I. Overview Systemic 5-HT administration produces changes in many gastrointestinal functional parameters including gastric acid secretion (Beck, 1985; Cho and Ogle, 1986b; Canfield and Spencer, 1983; Johansen and Bech, 1991). Studies performed in the rat (Cho and Ogle, 1986b; Hsu et al. 1991; Canfield and Spencer, 1983) and in the dog (Beck, 1986b; Beck, 1986a; Black et al. 1958; Bech, 1988) demonstrate the ability of intravenous 5-HT to inhibit stimulated gastric acid secretion. Application of exogenous 5-HT inhibits acid secretion in vitro (Hsu et al. 1991; Canfield and Spencer, 1983) and in vivo (Cho and Ogle, 1986b; LePard and Stephens, 1992) independently of the vagus nerve. Studies using the isolated rat stomach suggest prostaglandins, not the enteric nervous system, mediate inhibition of acid secretion by 5-HT (Canfield and Spencer, 1983). In contrast, studies using an ex vivo stomach preparation suggest prostaglandins are not involved
78 79 (Cho and Ogle, 1986b). The mechanism of 5-HT to inhibit acid secretion in the intact rat remains controversial. Due to autoradiographic localization of 5-HT, receptors to enteric neurons of the rat stomach (Kirchgessner et al. 1993a), their involvement in mediating inhibition of acid secretion by 5-HT was explored in this phase of the project.
II. Hypothesis: serotonin Inhibits Acid Secretion by Acting Through Enteric Neurons.
III. Method. A. Animals. Male Sprague-Dawley rats [180-220 g, Harlan Industries, Indianapolis, IN and 180-220 g, Zivic-Miller, Zelienople, PA (section IV.B.3.)] were maintained ad libitum on Purina Laboratory Chow and tap water. They were housed under controlled conditions of temperature (22 + 1 °C) and illumination (6 a.m. to 6 p.m.). All experiments were performed in animals deprived of food but not water 18-24 hrs before the experiment. B. Measurement of Gastric Acid Secretion. See Chapter II section III.B. c. Cannulation of Vessels. 1. Splenic Artery. See Chapter II section III.C. For the study investigating the role of gastric mucosal blood flow in mediating attenuation of acid secretion by 5-HT (section IV.D.2.a.), sodium nitroprusside was continuously 80 perfused at a rate of 13.6 /nl/min (Lopez-Belmonte et al. 1993) to dilate the gastric vasculature. 2. Portal Vein. In order to repetitively withdraw portal samples, a cannula was inserted into the portal vein at the level of the hillus above the entrance of the veins draining the stomach. The portal cannula consisted of two 1.5 cm segments of PE 50 tubing, with one end beveled at a 45° angle, connected to the ends of a 4.5 cm segment of heat shrink tubing (FIGURE 5). The portal cannula was then placed over a 25 gauge spinal needle. The bevels of the portal cannula and spinal needle were aligned such that the bevel of the portal cannula was 1-2 mm below the bevel of the spinal needle. The gut was deflected to the right with moist gauze and the portal vein was exposed (FIGURE 6). A 50 ml syringe was placed under the trunk of the rat to increase accessibility of the portal vein. Using a needle and suture (4.0), a ligature was tied into the fat adjacent to the portal vein. Providing tension on the distal gut, the tip of the needle, then PE tubing, was inserted into the vein according to the method of Kimura (Kimura et al. 1988a; Rich-Denson and Kimura, 1988; Kimura et al. 1988b). As the cannula was advanced, the needle was withdrawn and a syringe containing 10 U heparin was attached. The ligature was tied and glued in place using cyanoacrylate. Heparin (100 U) was flushed through the portal cannula to ensure patency. 81 K l\ PE SO J
Hecrt- Shrink Tubing
i I
Peso
• A FIGURE 5: A. 25 gauge spinal needle. B. Schematic of the portal cannula used for withdrawal of portal blood samples. The portal cannula consisted of two 1.5 cm segments of PE 50 tubing, with one end beveled at a 45° angle, connected to the ends of a 4.5 cm length of heat shrink tubing. The cannula was then placed over a 25 G (A) spinal needle. The bevels of the spinal needle and portal cannula were aligned with the cannula being 1-2 mm below the needle. ••
FIGURE 6: Anatomy of the portal venous system of the rat. For blood sampling, the portal vein was cannulated at position 1 at the hillus of the liver above the entrance of the veins draining the stomach. The cannula for indocyanine green dye infusion was inserted in the mesentery at position 2, approximately 2 cm below cannula 1. 83 D. Methods for Assay of Serotonin Content. 1. Gastric Perfusates. To a portion of the gastric perfusate an internal standard, 5-hydroxy-Na-methyltryptamine, was added, the mixture was centrifuged and a volume of 250 fil was injected into a high performance liquid chromatograph and analyzed for 5-HT (section iii.E.). Data were expressed either as ng 5-HT/10 min or as 2 hr integrated response after treatment (ng 5-HT/2 hr). 2. Whole Blood. Serotonin content in portal blood was determined according to the method described by Korpi (Korpi, 1984). In brief, immediately after collection of portal whole blood, samples were placed in microcentrifuge tubes and gassed with carbon monoxide (CO) for 30-60 seconds. A portion, 30 nl, was removed for analysis. Secondly, 410 /xl of 0.2 mM ice cold sodium metabisulfite, saturated with EDTA and gassed with CO, was added along with 10 jul of 54 nM solution of internal standard, 5-hydroxy-No-methyltryptamine oxalate, made with EDTA-saturated sodium metabisulfite solution. The sample was vortexed, then 50 jzl of 4 M perchloric acid was added to precipitate proteins. The sample was then treated in the following manner: 1) vortexed for 10 sec, 2) kept in an ice water bath for 5 min, 3) centrifuged for 1 min at 16,000 X g, 4) decanted into another microcentrifuge tube, and 5) recentrifuged for 30 sec. A volume of 10 /il was injected into the HPLC. The concentration 84 of 5-HT in the portal whole blood sample was expressed in fjLg/ml. 3. Electrochemical Detection of Serotonin.Serotonin content in gastric or whole blood samples was determined by high performance liquid chromatography (HPLC). The liquid chromatograph was equipped with a L-ECD-6A electrochemical detector, LC-6A pump, SIL-6A autoinjector and a SCL-6A system controller (Shimadzu, Japan). Data were displayed by the C- R5A chromatopac. In addition, chromatographs were stored in the computer by Chromatopac Data Archive Utility (Shimadzu, Japan). The system was connected to a SPHERI-5 ODS 5 micron column (220 X 4.6 nun, Applied Biosystems, San Jose, CA). The mobile phase consisted of the following components (g/L) dissolved in 8% absolute methanol and adjusted to pH 2.7:
NaH2P04, 24.6 and disodium EDTA, 0.04. The mobile phase was filtered through 0.45 fim filters (Rainin Instrument Co., Inc.; Woburn, MA) and degassed before use. The system was run at a flow rate of 1.0 ml/min yielding a pressure of 120 kgf/cm2, and the operating potential was +700 mV. For generation of the standard curve for HPLC analysis of 5-HT, stock solutions of both serotonin creatinine sulphate and the internal standard, 5-hydroxy-No-methyltryptamine oxalate were diluted to the appropriate concentrations utilizing saline adjusted to pH2 with hydrochloric acid. Standard curves were generated by injecting 0.5-2 ng of 5-HT and 1 ng of the internal standard. Linear regression of the 85 ratio of 5-HT to internal standard was used to determine concentration of 5-HT in the gastric samples (Hewlett Packard calculator). E. Basio Experimental Protocol. Animals were prepared according to section III.B. After three 10-min basal collections of gastric perfusates, acid secretion was stimulated by pentagastrin (24 /ug/kg/hr, i.v.). After nine additional 10-min periods, 5-HT (3.5 fimol/kg) was injected via the femoral vein. Acid secretion was collected for 6 additional 10-min periods. Data were expressed as percentage of the difference between the average of the three acid secretory measurements taken 30 min before 5-HT treatment and the largest incremental change in 10-min acid output during the 30 min period after 5- HT treatment (% change in acid secretion). To determine pretreatment effects by either tetrodotoxin (TTX) or piroxicam, data were expressed as a percentage of the difference between the average acid secretion 30 min before and after pretreatment divided by the average acid secretion before pretreatment (% change in acid secretion). Due to the varied individual acid secretory responses to pentagastrin obtained in sections IV.D.2. and IV.F. time course data were standardized with the average of the three acid secretory measurements taken 30 min before 5-HT treatment designated as 100%. 86 F. Drugs. The stable TRH analogue, RX77368 (Reckitt and Coleman, Kingston Upon Hull, UK), was obtained in powder form, was dissolved In 0.01% bovine serum albumin (BSA) and 0.9% saline at an Initial concentration of 1 mg/ml and was kept at -70°C. The peptide was injected intracisternally in a volume of 10 /xl. Serotonin maleate and baclofen for injection via the femoral vein (Sigma, St. Louis, MO) were dissolved in water. TTX with citrate buffer (Sigma, St. Louis, MO) and sodium nitroprusside (Elkins-Sinn Inc., Cherry Hill, NJ) were dissolved in saline. Sodium nitroprusside was protected from light with foil and was dissolved in chilled saline. Indocyanine green dye (Sigma, St. Louis, Mo) was dissolved in 0.01% BSA water. Piroxicam (a generous gift from Dr. Helen
Cooke) was suspended in 1.25% NaHC03. Carbachol (Alcon, Puerto Rico), pentagastrin (Ayerst Laboratories INC, New York, NY) and dexamethasone sodium phosphate (American Regent laboratories, Inc.; Shirley, NY) were diluted using normal saline to the appropriate concentration. RX77368 was diluted to the appropriate concentration using sterile 0.01% BSA saline. For HPLC determination of 5-HT, serotonin creatinine sulphate and the internal standard, 5-hydroxy-No- methyltryptamine oxalate (Sigma, . St. Louis, Mo) , were dissolved in 0.1 N perchloric acid with 1.5 mg sodium metabisulfite and stored at -70°C. For HPLC determination of 87 catecholamines, the internal standard 3,4-dihydroxybenzylamine HBr and (-)-norepinephrine bitartrate salt (Sigma, St. Louis, MO) were dissolved in 0.1 N perchloric acid. G. Statistics. In sections IV.A.l.a., IV.A.2.a., IV.B.2., IV.B.3. and IV.D. data were analyzed by Student £ test. In sections IV.A.l.b., lV.A.2.b., IV.B.l.a., IV.C., IV.E., and IV.F. data were analyzed utilizing the one way analysis of variance with post-hoc Student Neuraan Keulls £ test. To determine if enhancement of acid secretion by TTX had any subsequent effect on the 5-HT response (section IV.B.l.b.), the data were analyzed utilizing the two-way analysis of variance with post-hoc Student Neuman Keulls £ test. Differences between groups were considered significant if p < 0.05.
IV. Results. A. Source of Serotonin Mediating Inhibition of Acid Secretion: Gastric Lumen vs. Portal Circulation. 1. Gastric Lumen. a. Physiologic Levels of Serotonin in the Gastric Lumen after Vagal Stimulation. 1) Methods. To investigate the ability of 5-HT in the gastric lumen to modulate acid secretion, the physiologic levels of 5-HT in the gastric lumen achieved after vagal stimulation were determined. Animals were prepared as detailed in section III.B. except gastric secretions were 88 collected by flushing the lumen with a 3 cc bolus of normal saline followed by a 5 cc bolus of air. After three 10-min basal collections of gastric perfusates, animals were treated intracisternally with the TRH analogue RX77368 (100 ng/10 /xl) utilizing a stereotaxic instrument. Gastric perfusates were collected for twelve additional 10-min periods. Half of the gastric secretory sample (1.5 ml) was titrated to determine acid output with 0.01 N NaOH to pH 7.0 on an automatic titrator (Radiometer, Copenhagen, Denmark), correcting for the aliquot removed for 5-HT analysis. Data were expressed as /xmol acid/10 min. The remaining portion (1.5 ml) was analyzed for 5-HT content (sections III.D.l. and III.D.3.). Data were expressed as either ng/10 min or 2 hr integrated response. 2) Results. Serotonin was released spontaneously into the gastric lumen at an average rate of 1.73 + 0.33 ng/10 min (FIGURE 7). The concentration of 5-HT in the gastric lumen increased at a slower rate than acid secretion after vagal stimulation by RX77368. Thirty min after vagal stimulation, the concentration of 5-HT in the gastric lumen peaked. This peak in 5-HT concentration in the gastric lumen occurred 10 min after the peak in acid secretion. The concentration of 5-HT in the gastric lumen and acid secretion remained at a plateau for the remainder of the experiment (FIGURE 7). Overall, treatment with RX77368 increased the concentration of 5-HT in the gastric lumen by 88% as compared to vehicle treated animals [mean + SEM, 2 hr 89 integrated response (ng/2 hr): vehicle (N=5), 10.49 ± 2.87; RX77368 (N=4), 87.85 ± 9.67; PC0.0001]. 90
-©- ACID 5-HT N 6
o ao 0
H H Ul 111 £C AC 0 o Ul 111 CO <0 I- a X o 1 < K> TIME (MIN)
FIGURE 7: Time course of the effect of RX77368 on acid and 5-HT secretion into the gastric lumen. After three 10- min basal collections, RX77368 (RX, 100 ng, i.e.) was injected into the cisterna magna (10 fil). Gastric perfusates were collected for twelve additional 10-min periods. N represents the number of animals in each group. Data are expressed as mean + SEM. 91 b. Effect of Exogenous Serotonin in Gastric Lumen on Acid Secretion. 1) Methods. To determine if 5-HT in the gastric lumen inhibits acid secretion, exogenous 5-HT was perfused through the gastric lumen in varying concentrations in an attempt to mimic endogenous levels achieved after vagal stimulation by RX77368 (FIGURE 7). The aforementioned basic protocol for gastric collections (section III.F.) was followed except after nine 10-min periods, either vehicle (pH 2 saline) or various concentrations of 5-HT (10, 100 or 370 ng/10 min) was continuously perfused through the gastric lumen. To maximize exposure of the gastric mucosa to 5-HT, the 5-HT solution was slowly infused (1 ml/min, Harvard Pump) through a delivery tube (PE 50) extending approximately 2.5 cm beyond the gastric fistula (FIGURE 8). A portion of the gastric perfusate was analyzed for acid, with correction for the aliquot taken to determine 5-HT content (section III.D.l.). Acid secretions were collected for twelve additional 10-min periods. Data were expressed as either nmol acid/10 min or as % change after 5-HT treatment as described in section III.F. In addition, to confirm physiologic levels of 5-HT were indeed achieved in the gastric lumen, the concentration of 5-HT in the gastric perfusate was determined (section III.D.l. and III.D.3.) and expressed as ng 5-HT/10 min. 92
B.
ASPIRATE
FIGURE 8: Schematic of the triple lumen cannula. 5-HT was perfused through the discrete tubing (A) extending 2.5 cm beyond the inflow of the gastric fistula (B). denotes direction of flow. 93 2) Results. Perfusion of vehicle (pH 2 saline) slightly increased acid secretion (FIGURE 9A). 5-HT, perfused through the gastric lumen at levels encompassing the range seen experimentally after vagal stimulation with RX77368 (FIGURE 7), did not alter pentagastrin-stimulated gastric acid secretion [mean ± SEM, % change in acid secretion: vehicle (N=12), +53.8 ± 13.4; 10 ng/10 min (N=4), +58.3 ± 15.6; 100 ng/10 min (N=4), +51.5 ± 9.5; 370 ng/10 min (N=4), +80.5 ± 21.7; FIGURE 9]. The concentration of 5-HT in the gastric perfusate was always lower than the desired perfusing concentration (FIGURE 9). 94
FIGURE 9: Effect of perfusion of various concentrations of 5-HT through the gastric lumen on pentagastrin-stimulated acid secretion. Pentagastrin (24 ^mol/kg/hr, i.v.) infusion began (first arrow) after three 10-min basal periods. After nine additional 10-min periods (second arrow), the gastric lumen was perfused with either vehicle [A. pH 2 saline, (N=12)] or 5-HT [B. 10 ng/10 min, (N=4) ; C. 100 ng/10 min, (N=4); D. 370 ng/10 min, (N=4)]. Gastric perfusates were collected for six additional 10-min periods. The acid secretory response (open circles, /m°l/10 min) is displayed by the left y axis. The resultant concentration of 5-HT generated in the gastric perfusate (closed circles, ng/10 min) is displayed by the right y axis. Data are expressed as mean ± SEM. 95 2. Portal Circulation a. Physiologic Levels of Endogenous Serotonin in the Portal Circulation after Vagal Stimulation. 1) Concentration of 5-HT. a) Methods. To determine the concentration of 5-HT in the portal circulation after vagal stimulation, whole blood samples (100 fil) were taken from the portal vein at 10-min intervals. Collection syringes were equipped with 18 G needles and 10 fil of 0.5 jLtmol EDTA to prevent clotting. After three basal samples, the rats were injected intracisternally (10 nl) with either vehicle [0.1% bovine serum albumin (BSA) in saline] or the TRH analogue RX77368 (100 ng). Portal whole blood samples were collected for twelve additional 10-min periods. Whole blood samples were treated as detailed in section III.D.2. Serotonin content was determined as described in section III.D.3. Data were expressed as jug/ml. b) Results. Treatment of the rat with RX77368 (100 ng, i.e.) did not change the concentration of 5-HT in portal whole blood as compared to vehicle treated animals (FIGURE 10). Throughout the time course of the experiment, the concentration of 5-HT in either group did not change. 96
-O- VEHICLE RX77368 N - 5 N - 5
VEH/RX
I m 0.40
0.00 100 no no TIME (MIN)
FIGURE 10: Time course of the effect of RX77368 on concentration of 5-HT in the portal circulation. After three basal collections of portal blood, either vehicle (VEH, 0.1% BSA saline, i.e.) or RX77368 (RX, 100 ng, i.e.) was injected into the cisterna magna. Portal blood samples were collected for twelve additional 10-min periods. N represents the number of animals in each group. Data are expressed as mean + SEM. 97 2) Flow Through Portal Vein. a) Methods. The data of FIGURE 10, which demonstrated no change in the concentration of 5-HT in the portal blood, sharply contrasts with results of previous work describing the ability of vagal stimulation to enhance the concentration of 5-HT in the portal blood (Pettersson, 1979; Gronstad et al. 1987; Tobe et al. 1976; Gronstad et al. 1988a). Since vagal stimulation by RX77368 has been shown to increase both plasma catecholamine concentrations (Brown, 1981) and gastric mucosal blood flow (Tach6 et al. 1989), the effect of RX77368 on portal blood flow was determined. The flow of blood through the portal vein was determined by clearance of indocyanine green dye (3 mg/kg). This dye is completely metabolized by the liver (Stoeckel et al. 1980; Burns et al. 1989) and has been routinely used to determine blood flow (Stoeckel et al. 1980; Burczynski et al. 1987; Donald and Yipintsoi, 1973; Burns et al. 1989). In these studies, a second portal cannula (23 G needle attached to PE 50) for indocyanine green dye infusion (0.45 ml/min, Harvard Pump) was inserted 1.5-2.0 cm below the portal vein collection cannula at position 2, tied and glued with cyanoacrylate in the mesentery (FIGURE 6). Prior to time of flow analysis, portal blood samples were withdrawn to mimic the collection protocol detailed above [section IV.A.2.a.l)a)]. 98 In some animals, a second flow curve of a later time period was determined. In these animals, no portal samples were withdrawn between subsequent flow determinations. The high molecular weight sugar, ficoll (1 ml of a 4% solution, rate of 0.1 ml/min, i.v.), was administered for volume replacement. One min before injection of the dye, a blood sample (250
lil) was taken which served as the blank for spectrophotometry detection of dye content. At time 0:00, dye infusion began. Subsequent whole blood samples (250 /xl) were withdrawn in the following time sequence (min): 0:30, 0:45, 1:00, 1:15, 1:30, 1:45, 2:00, 3:00, 4:00, 5:00, 6:00, 7:00, 8:00 and 10:00. Every sample, including the blank, was immediately treated with 2.8 ml of chilled 0.1% BSA saline, vortexed well and chilled in the refrigerator. After one hr, 3 ml of chilled acetone was added while vortexing. Samples were then centrifuged at 4 "C at 100 X g for 20 min. If a sample was turbid, it was centrifuged for an additional 20 min. If a sample remained turbid, it was discarded from subsequent analysis. The clear supernatant was decanted into a disposable cuvette for determination of absorbance against the blank at 795 nm wavelength using a Perkin Elmer spectrophotometer. This wavelength was confirmed in our lab to produce the best linear regression of the indocyanine green standard curve (personal observation). The concentration of dye in the 99 sample was then calculated from the standard curve, plotted against time and the area under the curve was analyzed utilizing Sigma Scan (Jandel Scientific). Flow was calculated as the dose of the dye divided by the area under the curve and was expressed as ml/min. The dye aliquot used for determination of flow was also used for generation of the standard curve. At the end of the experiment, 5-8 additional portal samples (250 fil) were withdrawn. One sample was treated as the blank; whereas, to the other samples, dye was added resulting in concentrations encompassing the range seen experimentally. The samples were treated as detailed above. The absorbance vs. concentration curve was fitted to a straight line by regression analysis (Hewlett Packard calculator). b) Results. Treatment of the rat with RX77368 (100 ng) increased the flow of blood through the portal vein, especially in the first twenty min after intracisternal injection (FIGURE 11). Flow began to increase in the first time period following RX77368, but was not statistically significant. Twenty min after RX77368, the flow of blood was three times greater than control. Thirty min after RX77368, the flow of blood had stabilized and was not statistically different from vehicle treated rats. Blood flow did not deviate from control levels for the remainder of the experiment. 100
VEHICLE RX77368 N - 3-10 N - 3-10
VEH/RX
$ o U.
J L
TIME (MIN)
FIGURE 11: Time course of the effect of vehicle (VEH) or RX77368 (RX) on flow of blood through the portal vein. Basal portal samples were withdrawn until time of flow analysis. At this time, indocyanine green dye (3 mg/kg) was infused through cannula 2 (see FIGURE 6) and whole blood samples were withdrawn in the manner described in method section IV.A.2.a.2)a) above. N represents the range of animals for each group for each time. Data are expressed as mean + SEM; *P<0.05. 101 3) Nat Release of Serotonin. a) Methods. To determine the net release of 5-HT into the portal circulation, the concentration
of 5-HT (fig/ml) was multiplied by the flow (ml/min) to give net secretion (/*g/min) of 5-HT (Gerber and Payne, 1992). In the case of the two time periods before treatment daring which flow was not experimentally determined (-20 and -10), the flow at time 0 was considered to represent basal flow. For time periods after treatment where flow was not experimentally determined, the average of the flow on either side was taken as an estimation in calculation of net release of 5-HT. Data were expressed as net release of 5-HT (jug/min) or 30 min
integrated response (ng/30 min). b) Results. Although the concentration of 5-HT in the portal circulation did not change after vagal stimulation by RX77368 (FIGURE 10), the flow of blood through the portal vein was increased (FIGURE 13) suggesting the net secretion of 5-HT was also increased. RX77368 increased the net secretion of 5-HT into the portal circulation with the peak occurring twenty min after vagal stimulation by RX77368 (FIGURE 12). Net secretion of 5-HT at this time was about three times greater than control. Thirty min after RX77368, net secretion of 5-HT was not significantly different than control and remained at control levels for the remainder of the experiment. Consequently, treatment with RX77368 produced a net secretion of 120 nq of endogenous 5-HT 102 into the portal circulation in the 30 min after injection
[mean + SEM, ng/30 min: vehicle (N=5), 135 + 19; RX77368 (N=5), 256 + 35; p<0.05]. 103
-O- VEHICLE RX77368 N - 5 N - 5
VEH/RX • / *
I to
TIME (MIN)
FIGURE 12: Time course of the effect of RX77368 on net release of 5-HT into the portal circulation. Concentration of 5-HT in the whole blood sample was corrected for changes in portal blood flow to result in secretion of 5-HT into the portal circulation. N represents the number of animals in each group. Data are expressed as mean + SEM; *P<0.05. 104 b. Levels of Serotonin in the Portal Circulation after Intravenous Serotonin. X) Methods. To compare levels of 5-HT in the portal vein after intravenous administration of 5-HT with levels of 5-HT reached after vagal stimulation, the concentration of 5-HT in portal whole blood after intravenous administration of 5-HT was determined. To permit withdrawal of repetitive portal blood samples, a cannula was inserted into the portal vein (section III.C.2.) at position 1 (FIGURE 6) and the basic protocol to determine the effect of 5-HT on pentagastrin-stimulated gastric acid secretion (section III.F.) was followed. In addition, after six additional 10- min periods, portal collections began. After three collections of acid and portal samples, 5-HT (3.5 fimol/kg) was infused via the femoral vein. Both acid secretions and portal samples were collected for three additional 10-min periods. The portal samples were analyzed for 5-HT content (sections III.D.2. and III.D.3.). The concentration (/ig/ml) of 5-HT in the sample was multiplied by the basal flow (ml/min) of blood through the portal vein (FIGURE 11) to obtain an estimation of 5-HT content (fig/30 min). 2) Results. The concentration of 5-HT in portal whole blood increased 680% after intravenous administration of 5-HT [mean + SEM, concentration of 5-HT (jug/ml), N=4: before 5-HT, 0.48 + 0.10; after 5-HT, 3.30 + 0.31; P<0.001]. Basal flow was utilized to convert 105 concentration values to net release. The net 5-HT appearing in the portal circulation after intravenous 5-HT approximates that produced by vagal stimulation (FIGURE 13). 106
CONTROL RX77368 EX0QEN0U8 5-HT (100 NO, LC.) (3.5 /UMOL/KG, I.V.)
7IQUR2 13: Comparison of net release of 5-HT into the portal circulation after vehicle (control, 0.1% BSA, i.e.), RX77368 or 5-HT given via the femoral vein. The numbers in the bars represent the number of animals in each group. Data are expressed as mean + SEM; *p<0.05. 107 c. Effect of intravenous serotonin on Acid Secretion. 1) Methods. Gastric acid output was determined by following the basic protocol to determine the effect of 5-HT on pentagastrin-stimulated gastric acid secretion as described in section III.F. Data were described as jLtmol acid/10 min. 2) Results. Since 5-HT levels seen after vagal stimulation by RX77368 approximate those seen after intravenous 5-HT (FIGURE 13) at a dose that inhibits acid secretion by 80% (FIGURE 4), endogenous 5-HT released by vagal stimulation may modulate gastric acid output. This is addressed in detail in Chapter IV. B. Role of the Autonomic Nervous System. 1. Enteric Nervous System. To explore the involvement of the enteric nervous system in mediating the inhibition of acid secretion by 5-HT, TTX was administered close intra-arterially (i.a.) to the gastric circulation via the splenic artery. To determine effectiveness of the dose and route of administration of TTX, a control experiment was performed which theoretically involves the enteric nervous system. This experiment will be referred to as a positive control. In the subsequent experiment, the dose and route of administration of TTX, validated in the positive control, was used to investigate the involvement of the enteric nervous system in mediating inhibition of acid secretion by 5-HT. 108 a. Positive Control: RX77368. 1) Methods. Theory suggests the integrity of the enteric nervous system is required to mediate effects of the vagus on gastric function [FIGURE 14 (Wood, 1987)]. The ability of a central vagal stimulant, RX77368 (Tach6 et al. 1989), to effect gastric function is therefore dependent on enteric neuronal activity. As a positive control to determine the effectiveness of both the dose and route of
administration of TTX (5 ng, close i.a.), its effect on the RX77368-stimulated acid response in chronically cannulated rats was determined (see CHAPTER IV, section III.C.). After three 10-min basal collections of gastric perfusates, animals were pretreated with either vehicle or TTX close i.a. to the gastric circulation through the splenic artery. After three additional 10-min basal collections, animals were treated with the vagal stimulant RX77368 (100 ng, i.e.). Acid secretions were collected for twelve additional 10-min periods. Data were expressed as 2 hr integrated acid secretion after vehicle or RX77368 treatment (jumol/2 hr). 2) Results. Pretreatment with TTX (5 jug, close i.a.) abolished RX77368-stimulated (100 ng, i.c) gastric acid secretion (FIGURE 15). Basal acid secretion was not altered in animals pretreated with TTX as compared to vehicle pretreated groups. In animals pretreated with TTX then the vagal stimulant RX77368, acid secretion did not increase above basal levels. 109
RX77368 BRAIN
VAGUS
V ENTERIC mnNEURONS EFFECTOR
SYSTEM
FIGURE 14: Schematic illustrating the theoretical basis for the enteric nervous system mediating vagally-stimulated acid secretion. Stimulation of the vagus by intracisternal RX77368 increases acid secretion by acting through enteric neurons. Therefore, close gastric pretreatment of the stomach with TTX should prevent RX77368-stimulated acid secretion by preventing enteric neuronal transmission. 110
300
PRETX:
FIGURE 15: Effect of TTX on RX77368-stimulated gastric acid secretion in chronic rats. After three 10-min basal collections, animals were pretreated with either vehicle (V) or TTX (5 ng/, close i.a.). After three additional 10-min collections, animals were treated with either vehicle (V, 0.1% BSA saline, i.e.) or RX77368 (RX, 100 ng, i.e.)* The acid secretory response was measured for twelve additional 10-min periods. The numbers in the bars represent the number of animals in each group. Data are expressed as mean ± SEM; **P<0.01. Ill b. Role of the Enteric Nervous system. 1) Methods. Once data from the positive control experiment suggested both the dose and the route of administration of TTX effectively prevented neuronal transmission at the level of the enteric nervous system, TTX was used as a probe to assess the participation of the enteric nervous system in mediating inhibition of acid secretion by 5- HT. The basic protocol to determine the effect of 5-HT on pentagastrin-stimulated acid secretion (section III.F.) was followed except at +60 min the animal was pretreated with
either vehicle or TTX (5 nq) given close i.a. to the gastric circulation via the splenic artery. Data were expressed as described in section III.F. 2) Results. Pretreatment with TTX alone enhanced pentagastrin-stimulated acid secretion (FIGURE 16A). TTX was ineffective to reverse inhibition of acid secretion by 5-HT (FIGURE 16B). 112 A. Na 3 3
V TTX
FIGURE 16: A. Effect of local gastric administration of vehicle (V) or TTX (5 ng, close i.a.) on pentagastrin- stimulated (24 /xg/kg/hr, i.v.) acid secretion. Data are expressed as mean + SEM; **P<0.01. B. Ability of local gastric administration of vehicle or TTX to reverse 5-HT- induced (3.5 /iitiol/kg, i.v.) attenuation of pentagastr in- stimulated acid secretion. The numbers in section A. represent the number of animals in each group. Data are expressed as mean ± SEM. 113 2. Parasympathetic Nervous System: Vagus Nerve. a. Methods. To determine the involvement of the vagus nerve in mediating the inhibition of acid secretion by 5-HT, experiments were performed in rats after acute bilateral, cervical vagotomy. In order to obtain sufficient acid secretory levels for data analysis, acid secretion was stimulated by the cholinergic agonist carbachol. After three 10-min basal collections of gastric perfusates, the animal was pretreated with either vehicle or 5-HT (28 jtmol/kg, i.p.). After three additional basal collections, acid secretion was stimulated by subcutaneous injection of carbachol (1 mg/kg). Gastric perfusates were collected for 18 additional 10-min periods. Data were expressed as either /xmol acid/10 min or 3 hr integrated acid secretion (nmol/3 hr). b. Results. In animals pretreated with vehicle, acid secretion began to steadily increase 30 min after subcutaneous injection of carbachol. After 90 min, acid secretion reached a plateau and remained elevated the duration of the experiment. In animals pretreated with 5-HT, acid secretion only slightly increased 100 min after carbachol administration (FIGURE 17). Vagotomy was ineffective to reverse the inhibition in acid by systemic 5-HT [mean ± SEM, 3 hr integrated acid secretion (fimol/3 hr): vehicle (N=6), 207.2 ± 27.9; 5-HT (N=5), 43.0 ± 13.4; PC0.001; FIGURE 17]. 114
-©- VEHICLE 5-HT N - 6 N - 6
o -i o T TTT T t ZlL^e-e^IITT" VEH/5HT o I- CAIIBACHOL U1 DC o 111 CO a o < 10 MIN PERIODS
FIGURE 17: Time course of the ability of 5-HT to inhibit carbachol-stimulated gastric acid secretion in rats after acute bilateral, cervical vagotomy. After three 10-mm basal collections, animals were pretreated with either vehicle (VEH) or 5-HT (28 jumol/kg, i.p.). After three additional 10-min periods, animals were treated with carbachol (1 mg/kg, s.c.). The acid secretory response was measured for 18 additional 10- min periods. N represents the number of animals in each group. Data are expressed as mean + SEM. 115 3. Sympathetic Nervous System: Splanchnic Nerves, a. Role of the Splanchnic Nerves. 1) Methods. To determine the involvement of the splanchnic nerves in mediating the inhibition of acid secretion by 5-HT, experiments were performed two weeks after total celiac ganglionectomy (GX) (Zivic-Miller), at a time after peripheral sympathetic nerve terminals to the stomach have degenerated (Raybould et al. 1987). The basic protocol to determine the effect of 5-HT on pentagastrin-stimulated acid secretion (section III.F.) was followed. Data were expressed as /zmol acid/10 min or /xmol/90 min or as described in section III.F. 2) Results. Acid secretion reached a plateau 40 min after the start of pentagastrin infusion. The plateau reached by the GX group was greater than that achieved by the sham treated group [mean + SEM, 90 min integrated response (/xmol/90 min): sham (N=4) , 84.0 ± 7.1; GX (N=4), 155.8 + 20.8; P<0.05]. The difference in magnitude of acid secretory responses is probably due to the absence of inhibitory splanchnic nerves in the rats after celiac ganglionectomy. Acid secretion reached a nadir at 20 or 30 min after 5-HT and did not return to previous levels. Degeneration of splanchnic nerves in rats after celiac ganglionectomy was ineffective to reverse 5-HT-induced inhibition of acid secretion [mean + SEM, % change in acid 116 secretion: sham (N=>4), -51 ± 5; GX (N=4), 64 + 7; p=0.1687; FIQURE 18]. 117
SHAM GX N - 4 N - 4
• J
TIME (MIN)
FIGURE 18: Time course of the ability of 5-HT to attenuate pentagastrin-stimulated acid secretion in sham vs^. total celiac ganglionectomized (GX) rats. Pentagastrm (PG, 24 /xg/kg/hr, i.v.) infusion began after three 10-mm basal periods. After nine 10-min periods, the animal was treated with 5-HT (3.5 jumol/kg, i.v.). The acid secretory response was measured for six additional 10-min periods. N represents the number of animals in each group. Data are expressed as mean + SEM. 118 b. Tissue concentrations of Norepinephrine. 1) Methods. To confirm total celiac ganglionectomy indeed degenerated extrinsic splanchnic nerves to the stomach, the concentration of norepinephrine (NE) in full thickness tissue samples from the gastric corpus was determined. Immediately after the experiment, the animal was euthanized, the stomach was removed, and chilled in normal saline. The corpus was cut into four sections, internal standard, 3,4-dihydroxy-benzylamine HBr (1 ng/ml), was added and the tissue was homogenized using a polytron with 2 X 2 ml of 0.2 N perchloric acid with 0.8 mM sodium metabisulfite and 1 M EDTA then stored at -70 °C. The NE standards and tissue supernatants were extracted on 25 mg of acid washed alumina in a 1.5 ml plastic tube. Tris buffer (1.5 M, pH 8.6) was added and the sample was immediately shaken for 15-20 sec, then further shaken for 15- 20 min by table shaker. The alumina was allowed to fall to the bottom of the tube and the supernatant was aspirated. Each tube was washed 3 times with 1 ml distilled water, aspirating each time. Again, 1 ml of distilled water was added and the alumina slurry was transferred to microfilters loaded with RC 55 membranes. Samples were centrifuged for 4-5 min at 1000 X g. The receiving tube was replaced with a new
Eppendorph tube, then 200 fil of 0.1 N perchloric acid was added. The sample was vortexed (20 sec) allowed to stand for 10-15 min, then vortexed again (20 sec). Samples were 119 recentrlfuged at 1000 X g for 4-5 min. A volume of 10 /xl was injected into the HPLC for analysis. Data were expressed as ng NE/g corpus. The liquid chromatograph was equipped with a ESA Coulochem detector (Model 5100A), LC-6A pump (Shimadzu, Japan) and rheodyne (Cotati, CA). Data were displayed by the C-R5A chromatopac (Shimadzu, Japan). The system was connected to a phase II ODS 3 micron column (100 X 3.2 mm, Bioanalytical Systems, West Lafayette, IN). Mobile phase for elution of NE was composed of the following (g/L) dissolved in 8% absolute methanol and adjusted to pH 2.7: NaH2P04, 24.6; disodium EDTA, 0.04; and sodium octylsulphate, 100. The mobile phase was filtered through 0.45 fim filters (Rainin Instrument Co., Inc. ; Woburn, MA) and degassed before use. 2) Results. Norepinephrine content of the corpus was 84% lower in GX rats as compared to sham controls (FIGURE 19). The celiac ganglionectomy successfully degenerated extrinsic norepinephrine-containing splanchnic nerves to the stomach. 120
<0 3 a. oAC o 0 111z a
SHAM GX
FIGURE 19: Effect of total celiac ganglionectomy on norepinephrine content in whole tissue sections of gastric corpus. At the end of the experiment, the animal was sacrificed and the stomach was excised. For detailed methods regarding the analysis of norepinephrine content, see section IV.B.3.b.l). The numbers in the bars represent the number of animals in each group. Data are expressed as mean ± SEM; *P<0.05. 121 C. Role of Prostaglandins. 1. Methods. Conflicting reports exist regarding the involvement of gastric prostaglandins in mediating attenuation of acid secretion by 5-HT in vitro (Canfield and Spencer, 1983? Cho and Ogle, 1986b). To determine if prostaglandins mediate 5-HT-induced inhibition of acid secretion in vivo, the effect of piroxicam (15 pmol/kg, i.p.), a cyclo-oxygenase inhibitor, given at a dose shown to reduce prostaglandin Ez content of the stomach by 92% (Curtis et al. 1994), on 5-HT induced inhibition of acid output was determined. The basic protocol to determine the effect of 5-HT on pentagastrin- stimulated acid secretion (section III.F.) was followed except at +60 min animals were pretreated with piroxicam. Data were expressed as described in section III.F. 2. Results. Piroxicam showed a trend to enhance acid secretion which became more pronounced during the 30-60 min time period after intraperitoneal injection of piroxicam; however, this was not significant (FIGURE 20). Despite its tendency to enhance acid secretion, piroxicam pretreatment did not prevent attenuation of acid secretion by 5-HT (FIGURE 20B). Acid secretion reached a nadir 20-30 min after 5-HT and remained suppressed throughout the remainder of the experiment. 122 A. N 1M
-10* PRETX: 5HT 5HT
FIGURE 20: A. Effect of vehicle (V) or piroxicam (PX, 15 Mmol/kg, i.p.) on pentagastrin-stimulated (24 ng/kg/hr, i.v.) acid secretion. Data are expressed as mean + SEM. B. Ability of vehicle or piroxicam to reverse vehicle or 5-HT-induced (3.5 jmol/kg, i.v.) attenuation of pentagastrin-stimulated acid secretion. The numbers in section A. represent the number of animals in each group. Data are expressed as mean ± SEM; *P<0.05. 123 D. Gastric Mucosal Blood Flow. 1. Positive Control: Mean Arterial Pressure. a. Methods. Systemic 5-HT inhibits gastric mucosal blood flow (Bech and Anderson, 1985). Hence, gastric acid secretion may be limited by reduced blood flow to the mucosa (Leung et al. 1986). Nitric oxide, a gas synthesized by the vascular endothelium from i-arginine, causes local vasodilation of the gastric mucosal microvasculature by increasing cGMP formation (Stark and Szurszewski, 1992; Pique et al. 1989). To investigate the role of mucosal blood flow, the nitric oxide donor, sodium nitroprusside (SNP), was administered close to the gastric wall through the splenic artery at a dose shown to increase gastric mucosal blood flow and decrease systemic mean arterial pressure [MAP, 10 jLtg/kg/min (Lopez-Belmonte et al. 1993; Gardiner et al. 1993)]. To assess whether an adequate vasodilatory dose of SNP was used, the MAP of each animal was monitored by cannulating the femoral artery (Lopez-Belmonte et al. 1993). The changes in the parameter of MAP served as a positive control. After isolation of the femoral artery, the cannula (PE 50) was inserted, heparin (10 units/ml) was injected to ensure patency, and the cannula was attached to a pressure transducer (Gould P23 ID). MAP was continuously monitored on a chart recorder (Grass, Model 7D Polygraph). Before each experiment, the chart recorder was calibrated with a sphygmomanometer 124 (Omron, Model HEM-18). Data were expressed as change in MAP [mm mercury (Hg)]. b. Results. Close i.a. administration of SNP immediately decreased MAP in both vehicle (FIGURE 21 and 22A) or 5-HT challenged groups (FIGURE 21 and 22B). Intravenous vehicle had no effect on MAP (FIGURE 22A). In contrast, 5-HT incx'eased MAP during the time of intravenous administration via the femoral vein and continued for approximately 1 min, then decreased to pre-5-HT levels (FIGURE 22B). MAP remained suppressed during the SNP infusion. After SNP infusion halted, MAP returned to control levels without delay (FIGURE 22A and B). 125
0 X
a. <
uj zO < -100 X o VEH 5-HT
FIGURE 21: Effect of SNP on MAP. Close gastric infusion of SNP (10 jug/kg/hr, close i.a.) decreased MAP in both vehicle (VEH) and 5-HT treated groups. The numbers in the bars represent the number of animals in each group. 126
FIGURE 22: Representative tracings of the effect of SNP, infused close to the gastric wall through the splenic artery, and intravenous vehicle (A.) or 5-HT (B.) on MAP. First arrow: upon close i.a. infusion of SNP (SNP ON), MAP immediately decreased. Second arrow: effect of intravenous vehicle (A., VEH) or 5-HT (B.) infusion on MAP. Third arrow: after halting the SNP infusion (SNP OFF), MAP immediately returned to pretreatment levels. Horizontal bar denotes one minute; vertical bar represents 20 mm Hg. A* SNP ON SNP OFF • •
®* SNP ON SNP OFF • •
H M -J 128 2. Role of Gastric Mucosal Blood Flow in Mediating Attenuation of Acid Secretion by 5-HT. a. Methods. To determine if diminished blood supply to the mucosa mediates the inhibition of acid secretion by 5-HT, the vasodilator SNP was perfused close to the gastric circulation via the splenic artery. The basic protocol to determine the effect of 5-HT on pentagastrin-stimulated acid secretion (section III.F.) was followed except at +85 min SNP was infused via the splenic artery cannula [13.6 /il/min (Lopez-Belmonte et al. 1993)] and animals were given either vehicle or 5-HT through the femoral vein. Infusion of SNP was halted at +120 min. Data were expressed as described in section III.F. b. Results. Acid secretion reached a plateau 30 min after the start of pentagastrin infusion via the femoral vein. During close i.a. SNP infusion, acid secretion decreased steadily to reach a nadir 30 min after vehicle. After close i.a. infusion of SNP was halted, acid secretion slowly recovered to levels seen before SNP. Twenty min after intravenous 5-HT administration, acid secretion reached a nadir. Systemic 5-HT inhibited gastric acid secretion in the presence of SNP [mean + SEM, % change in acid secretion: vehicle (N=3), -31.0 + 5.9; 5-HT (N=3), -55.3 + 1.8; P<0.05; FIGURE 23]. 129
VEHICLE 5-HT N - 3 N - 3
V/5HT 1
TIME (MIN)
FIGURE 23: Time course of the effect of SNP °*J induced attenuation of acid secretion. Pentag^trin pig/kg/hr, i.v.) infusion began after three 10-mxn^ basal periods. After nine 10-min periods, the acid secretory T response plateaued in the range of 17-30 Mmol/10 an ve icl ( animal was treated at ninety mm with either ^. ^i before effective dose of 5-HT (3.5 fimol/kg, Five*call treatment, SNP (10 M9/Whr, close i.a.) was infus y to the gastric circulation through the splenic artery. Arter 35 min, close i.a. infusion of SNP was halted. Jhe acid secretory response was measured for six addition periods. N represents the number of animals in each gro p. Data are expressed as mean + SEM; **P<0.01. 130 E. Role of Gastric Mucosal Mast cells. 1. Methods. To investigate the role of gastric mucosal mast cells (MMC) in mediating inhibition of acid secretion by 5-HT, MMC were depleted by chronic pretreatment with dexamethasone [0.5 mg/kg, i.p. in a volume of 2.5 ml/kg; twice daily for 5 days; (Andersson et al. 1990? Heap and Kiernan, 1973)]. Two hours after the last dose, the experiment was performed following the basic protocol to determine the effect of 5-HT on pentagastrin-stimulated acid secretion as described in section III.F. Data are expressed as described as ninety min integrated acid response (/xmol/90 min) or as in section III.F. 2. Results. Acid secretion stimulated by pentagastrin was not significantly different in dexamethasone pretreated rats as compared to control rats [mean ± SEM, 90 min integrated response (jmol/90 min): vehicle (N=6), 239.3 + 49.7; dexamethasone (N=6), 356.8+65.2]. Serotonin inhibited acid secretion in rats depleted of MMC by chronic dexamethasone pretreatment. Although the response was significantly different from control, it was not as profound as the response of vehicle pretreated animals to 5-HT (FIGURE
24). 131
26
o O < z Ul O z -60- < X O -78-
PRETX: VEH DEX VEH DEX TX: VEH VEH 5-HT 5-HT
FIGURE 24s Effect of 5-HT to attenuate pentagastrin- stimulated acid secretion in vehicle 2£fLs. chronic dexamethasone-treated rats. To deplete MMC, animals were chronically pretreated with either vehicle (VEH) or dexamethasone (DEX) as described in section IV.E.l. Pentagastrin (24 jig/kg/hr, i.v.) infusion began after three 10-min basal periods. After nine 10-min periods, the animal was treated with either vehicle or an effective dose of 5-HT (3.5 Mmol/kg, i.v.). The acid secretory response was measured for six additional 10-min periods. The numbers above the bar® represent number of animals in each group. Data are expressed as mean + SEM; *P<0.05, **P<0.01. 132 F. Role of the Adrenal Gland. 1. Methods. Due to the necessity of the adrenal gland in mediating both release of 5-HT into the gastric lumen
! (Stephens, 1991) and the 5-HT1A agonist 8-0H-DPAT s effect on acid secretion (Chapter V), the role of the adrenal gland in mediating inhibition of acid secretion by 5-HT was determined in rats after bilateral adrenalectomy. In experimental animals, both adrenal glands were carefully excised from the animal. In sham controls, however, the adrenal glands were only exposed, not removed. Due to the inability of the adrenalectomized animal to withstand cardiovascular changes elicited by intravenous administration of 5-HT, 5-HT was administered close i.a. to the gastric circulation. At least 3 hr after adrenalectomy or sham treatment, collections of gastric perfusates began. The basic protocol to determine the effect of 5-HT (1.76 /umol/kg, close i.a.) on pentagastrin- stimulated acid secretion (section III.F.) was followed. Data were expressed as described in section III.F. 2. Results. Acid secretion stimulated by pentagastrin reached a plateau in 30 min. Twenty min after intravenous 5-HT, acid secretion reached a nadir in the sham group, then acid secretion slightly increased, but remained below 5-HT preinfusion levels throughout the experiment. In adrenalectomized rats, acid secretion decreased sharply 20 min after 5-HT, continued to fall throughout the experiment and reached basal levels one hour after 5-HT. Serotonin (1.76 133 jumol/kg, close l.a.) was effective to Inhibit acid secretion In both sham treated and adrenalectomized rats [mean ± SEM, % change In acid secretion: sham (N=5), -49.2 ± 5.2; adrenalectomy (N=4), -76.0 ± 8.2; P<0.05 FIGURE 25]. 134
SHAM ADX N - 5 N - 4
TIME (MIN)
FIGURE 25: Time course of the ability of 5-HT to attenuate pentagastrin-stimulated acute bilaterally adrenalectomized rats. Pentag;astnn ij ^mol/kg/hr, i.v.) infusion began after three "-«n basal neriods After nine 10-min periods, the acid secretory response plateaued in the range of 10-38 Minol/10 ' ^ animals were treated at ninety min with 5-HT (1.76 M^ol/kg, close i.a.). The acid secretory response was measured for 6 addiSioAai 10-min periods. N represents the number of animals in each group. Data are expressed as mean ± SEM. 135 V. Discussion. The main finding of these studies was that 5-HT, appearing in the portal circulation, inhibited gastric acid secretion through a mechanism that was independent of the autonomic nervous system, prostaglandins and mucosal mast cells. In addition, attenuation of acid output by 5-HT appeared not to be mediated by changes in gastric mucosal blood flow or adrenal-derived factors. Previous studies demonstrate vagal stimulation releases 5-HT into the gastric lumen (Hannun and Bell, 1987; Stephens, 1991; Cho et al. 1985; Stephens and Tach6, 1989) and portal circulation (Pettersson, 1979; Gronstad et al. 1987; Tobe et al. 1976; Gronstad et al. 1988a). Vagally-mediated release of 5-HT into the gastric lumen was confirmed (FIGURE 9). In contrast, preliminary studies suggested vagal stimulation did not alter levels of 5-HT appearing in the portal circulation (FIGURE 11). However, measuring the concentration of 5-HT in portal blood may not accurately reflect total secretion of 5- HT from the gut. In determining secretory rates of other substances from the gut, secretion is defined as the difference between the arterial and plasma concentrations of the hormone multiplied by plasma flow (Gerber and Payne, 1992). Blood flow is essential to determining secretory rates. Convergent hemodynamic studies demonstrate RX77368 markedly increases both plasma catecholamine levels (Brown, 1981) and gastric mucosal blood flow (Tache et al. 1989) 136 suggesting RX77368 may alter portal hemodynamics. Further studies demonstrated vagal stimulation by RX77368 increased portal blood flow (FIGURE IX) thereby enhancing net secretion
Of 5-HT (FIGURE 12).
Intravenous administration [FIGURE 4, (LePard and Stephens, 1994)], but not intragastric perfusion [FIGURE 9, (LePard and Stephens, 1992)], of 5-HT inhibited gastric acid secretion. In vitro studies confirm that gastric mucosal application of 5-HT does not alter pentagastrin-stimulated gastric acid secretion in the rat (Canfield and Spencer, 1983). The data suggest 5-HT appearing in the portal circulation mediates inhibition of acid secretion in vivo. To date, involvement of enteric neurons in mediating the inhibition of acid secretion by 5-HT has not been investigated in vivo. TTX given close to the gastric circulation did not prevent inhibition of acid secretion by 5-HT but did enhance pentagastrin-stimulated acid secretion (FIGURE 16). The increase in acid secretion after TTX administration may reflect removal of inhibition by splanchnic nerves as seen in GX rats (FIGURE 18). Studies utilizing the in vitro stomach preparation suggest the effect of 5-HT on acid secretion is independent of both enteric (Canfield and Spencer, 1983) and external (Cho and Ogle, 1986b; Canfield and Spencer, 1983) autonomic control. Both in vivo and in vitro data suggest that TTX-sensitive enteric neurons do not mediate 5-HT-induced attenuation of acid secretion. 137 There are conflicting reports regarding the role of gastric mucosal prostaglandins in mediating attenuation of acid secretion by 5-HT in vitro (Canfield and Spencer, 1983; Cho and Ogle, 1986b). The present study showed that pretreatment with the cyclo-oxygenase inhibitor piroxicam, at
a dose shown to reduce prostaglandin E2 content in the stomach by 92% (Curtis et al. 1994), was ineffective in reversing attenuation of acid secretion by intravenous 5-HT (FIGURE 20B). Prostaglandins are not involved in the inhibition of acid secretion by 5-HT in vivo. Serotonin may prevent degranulation of MMC, thereby possibly altering the functionally active pool of MMC-derived histamine or 5-HT. To investigate the role of MMC in mediating inhibition of acid secretion by 5-HT, animals were chronically pretreated with dexamethasone, an established protocol which depletes the rat stomach of MMC (Andersson et al. 1990? Heap and Kiernan, 1973; Rioux and Wallace, 1994). In vitro studies suggest dexamethasone 1) inhibits the production of various T cell-derived growth factors essential for differentiation and maturation of MMC and 2) acts directly on MMC to inhibit normal granule development (McMenamin et al. 1987). Dexamethasone also has effects on other immune cell populations, increasing and decreasing the number of neutrophils and macrophages, respectively (McMenamin et al. 1987). Hence, the treatment is nonspecific to MMC. Nonetheless, this established protocol depletes rat stomach 138 MMC (Andersson et al. 1990; Heap and Kiernan, 1973; Rioux and Wallace, 1994) and has been used to ascertain the role of MMC in modulating gastric function (Andersson et al. 1990; Rioux and Wallace, 1994). Previous studies suggest pentagastrin- stimulated acid secretion is not altered by chronic dexamethasone pretreatment (Andersson et al. 1990). This was confirmed in our model. Serotonin inhibited acid secretion in dexamethasone pretreated animals (FIGURE 24) suggesting that MMC do not mediate 5-HT-induced attenuation of gastric acid secretion. Systemic 5-HT suppresses gastric mucosal blood flow in the rat (Salim, 1990; Kitajima et al. 1991; Hashizume et al. 1978; Wong et al. 1990). Serotonin may attenuate acid secretion by impairing delivery of oxygen to mucosal parietal cells. In the present studies, 5-HT was capable of inhibiting acid secretion in the presence of a vasodilator given locally to the gastric circulation at a dose shown to significantly protect the mucosa from gastric erosions (Lopez-Belmonte et al. 1993; Gardiner et al. 1993) and to alter systemic MAP (Lopez-Belmonte et al. 1993). Changes in systemic MAP may reflect hypotensive effects of SNP after escape into the systemic circulation (Lopez-Belmonte et al. 1993). The data suggest 5-HT does not inhibit acid secretion by diminishing blood supply to the gastric mucosa. Since 5-HT attenuated gastric acid secretion in adrenalectomized rats (FIGURE 25), inhibition of acid 139 secretion by 5-HT is independent of adrenal-derived factors. Catecholamines from the adrenal are associated with inhibitory influences on acid secretion (Burks, 1994). Epinephrine administration at doses which mimic plasma concentrations reached after cerebro-ventricular corticotrophin releasing factor (CRF) administration does not alter pentagastrin- stimulated acid secretion (Druge et al. 1989). Adrenal- derived catecholamines do not mediate inhibition of acid secretion by 5-HT. The question remains as to the mechanism of 5-HT to attenuate acid secretion. Possible sites of action may include receptors associated with: 1) parietal cells or 2) gastric mucosal enteroendocrine cells. Serotonin may act directly on parietal cells. To date, demonstration of 5-HT receptor localization or the direct effect of 5-HT on gastric parietal cells is lacking. Secondly, 5-HT may modulate release of other secretogogues from gastric enteroendocrine cells. For example, 5-HT may exert paracrine effects to curtail histamine release from enterochromaffin-like (ECL) cells of the gastric mucosa. Again, no report demonstrates receptor localization or effect of 5-HT on isolated rat ECL cells. Serotonin has been shown to modulate release from D and G enteroendocrine cells of the gastric mucosa (Koop and Arnold, 1984). In the isolated rat stomach preparation, 5-HT inhibits somatostatin and enhances gastrin release, respectively (Koop and Arnold, 1984). There is precedent that 140 5-HT may Inhibit acid secretion by influencing release of secretogogues from other enteroendocrine cells of the rat gastric mucosa. Due to the multiplicity of secondary events in the whole animal model which may indirectly modulate gastric acid secretion, the site of the 5-HT, receptor mediating inhibition of acid secretion by 5-HT might be best studied utilizing an in vitro system. An isolated preparation permits experimentation with various drugs which may by fatal or less specific in the whole animal. Cultures of enriched, isolated parietal cells (Schepp et al. 1992; Berglindh, 1990) could be used to investigate the ability of 5-HT to modulate aminopyrine accumulation, second messenger activity, and ligand binding at the level of the parietal cell. Likewise, paracrine effects of 5-HT to modulate histamine release from ECL cells could be investigated utilizing an isolated oxyntic gland preparation (Chuang et al. 1992; Chuang et al. 1993; Berglindh, 1990). These are examples of further studies necessary to delineate the mechanism of action of 5-HT to inhibit gastric acid secretion. In conclusion, the mechanism of attenuation of acid secretion in vivo by 5-HT is independent of enteric neurons, vagal nerves, splanchnic nerves, prostaglandins, mucosal mast cells, gastric mucosal blood flow or adrenal-derived factors. Serotonin is capable of attenuating acid secretion in vitro; hence, this method may be best suited for further studies 141 determining the site of action which may include, but are not limited to, parietal cells or enteroendocrine cells of the gastric mucosa. CHAPTER IV
Role of Endogenous Serotonin in Modulating Gastric Acid Secretory Function
I. Overview Studies performed in the rat (Cho and Ogle, 1986b; Hsu et al. 1991; Canfield and Spencer, 1983) and in the dog (Beck, 1986b; Beck, 1986a; Black et al. 1958: Bech, 1988) demonstrate the ability of 5-HT to inhibit stimulated gastric acid secretion. Depletion of endogenous GI stores of 5-HT in the rat is correlated with potentiation of vagally-stimulated gastric acid secretion and motility at submaximal stimulation (Stephens et al. 1989; Stephens et al. 1990). Thus, 5-HT is proposed to exert an inhibitory tone on vagally-stimulated gastric acid secretion (Stephens et al. 1989; Stephens et al. 1990). The putative 5-HT receptor subtype mediating inhibition of acid secretion by intravenous 5-HT has recently been characterized as belonging to the 5-HT, family [CHAPTER II, (LePard and Stephens, 1994)]. The goal of studies in this phase of the project was to determine if endogenous 5-HT indeed exerts an inhibitory tone on vagally-stimulated gastric
142 143 acid secretion. Various 5-HT, antagonists were selectively administered to the gastric circulation, at doses shown to reverse inhibition by intravenous 5-HT, in an attempt to block the inhibitory action of endogenous 5-HT on gastric acid secretion. These studies investigated the role of endogenous 5-HT in modulating vagally-stimulated gastric acid secretion.
II. Hypothesis: Endogenous serotonin exerts an inhibitory tone on vagally-stimulated gastric acid secretion.
III. Methods A. Animals. See chapter II section III.A. Rats with chronically, indwelling cannula in the splenic artery were housed and fasted individually. B. Measurement of Gastric Acid Secretion. See chapter II section III.B. The patency of the chronically, indwelling cannula was confirmed during gastric surgery. C. Chronic Close Intra-arterial cannulation of Splenic Artery. To obtain selectivity, drugs were administered directly to the gastric circulation through a cannula inserted into the splenic artery. Due to variability in vagally- stimulated acid responses in animals with acutely placed splenic artery cannulas, it was necessary to develop a preparation with a chronically indwelling cannula in the splenic artery. In rats anesthetized with ketamine (80 mg/ml) and xylazine (2.5 mg/kg) anesthesia (i.p.), the ventral trunk 144 and back of the head were shaved. To help prevent infection, surgical instruments were sterilized in 100% ethanol, sterile saline was used and the abdomen was closed with sterile suture material (5-0 prolene, Ethicon Inc.). Each animal was given ampicillin (5 mg s.c., Wyeth Laboratories Inc., Philadelphia, PA) the day of experiment. The abdomen was shaved and swabbed with 100% ethanol, then opened by a midline incision. The stomach and spleen were placed on sterile gauze moistened with saline, the splenic artery was identified visually with a stereomicroscope and surgically isolated from the vein (FIGURE 1). The cannula (30 G needle attached to PE 10 tubing) assembly was inserted into the artery, tied and glued in place with cyanoacrylate. The stomach and spleen were replaced in the peritoneal cavity along with 15 cm of tubing. To help prevent adhesions, the peritoneal cavity was flooded with 10-15 ml of sterile saline. The abdominal wall was closed with sterile suture. The remaining tubing exited the abdominal cavity through the incision in the abdominal wall then was tunneled under the skin and secured at the back of the head. Finally, the abdominal skin was closed with wound clips. Heparin (250 nl, 10 U) was infused into the cannula, it was lightly clamped, and a 26 G stopper inserted to plug the cannula. Each rat was maintained under a heat lamp until recovery from the anesthesia, then individually housed. 145 The animals were allowed to recover for at least one week prior to the experiments. For three days after the surgery, the cannulas were flushed daily (250 Ml) with a mixture of heparin (10 U) and ampicillin (5 mg). Thereafter, the cannulas were flushed every other day with 250 pi of 10 U heparin. Each cannula was checked for patency on the day of the experiment. D. Ability of Serotonin Antagonists to Prevent Inhibition of Acid Secretion by Intravenous serotonin in Rats with Chronically Indwelling Splenic Artery Cannulas. 1. Protocol. In a portion of the studies, the ability of various 5-HT antagonists to prevent inhibition of acid secretion by intravenous 5-HT was determined using rats with chronically indwelling splenic artery cannulas. After three 10-min basal collections, acid secretion was stimulated by pentagastrin (24 /ig/kg/hr, i.v.). After six 10-min collections, the animal was pretreated by infusion of either vehicle or a 5-HT antagonist in a volume of 100 jul over a time course of one min into the splenic artery cannula. After three additional 10-min collections, the animal received a systemic dose of 5-HT (3.5 jLtmol/kg) via the femoral vein. Acid secretions were collected for six additional 10-min periods. 2. Statistics. To determine pretreatment effects, data are expressed as a percentage of the difference between the average acid secretion 30 min before and after 146 pretreatment divided by the average acid secretion before pretreatment. In the case of intravenous treatment with 5-HT, data are expressed as a percentage of the difference between the average of the three acid secretory measurements taken 30 min before treatment and the largest incremental change in 10-min acid output during the 30 min period after treatment. Those animals with cannula which were not patent responded in a manner not significantly different from those infused with vehicle, thus the data from these animals were integrated into the vehicle group. Data were analyzed by the one-way analysis of variance with post-hoc Student Neuman Keulls £ test. Differences between groups were considered significant if P<0.05. E. Effect of Serotonin Antagonists on RX77368-Stimulated Gastric Acid Secretion. 1. Protocol. After three 10-min collections of stomach perfusates, the animal was pretreated with either vehicle or a 5-HT antagonist given selectively to the gastric circulation via the splenic cannula. After three additional basal collections, the animal was treated with RX77368 (100 ng, i.e.). Gastric perfusates were collected for twelve additional 10-min periods. Data were expressed as 2 hr integrated acid response after RX77368 (/imol/2 hr). 147 2. Statistics. Data were analyzed utilizing the Student t test. Differences between groups were considered significant if P<0.05. F. Drugs. Methiothepin maleate (Research Biochemicals Inc., Natick, MA) was dissolved in water. Metergoline, methysergide and spiperone (Research Biochemicals Inc., Natick, MA) were suspended in 0.1% tween 80. Heparin (Elkins- Sinn INC, Cherry Hill, NJ), xylazine (Mobay Corp., Shawnee, KS) and ketamine (Parke-Davis, Morris Plains, NJ) were diluted using normal saline to the appropriate concentrations. See Chapter III section III.G. for information regarding RX77368.
IV. Results A. Effects of Serotonergic Antagonists on inhibition of Acid Secretion by Intravenous Serotonin in Rats with Chronic Indwelling Splenic Cannula. To establish effectiveness of the various 5-HT antagonists in the chronic model described earlier (see section III.C.), action to reverse inhibition of acid secretion by intravenous 5-HT was determined. Pretreatment with metergoline (1.1 jumol/kg) , methiothepin (1.1 or 0.55 /xmol/kg) , methysergide (3.5 /xmol/kg) or spiperone (2.5 /xmol/kg) alone had no significant effect on acid secretion at the concentrations tested (FIGURE 26A). Methiothepin at the 1.1 /xmol/kg dose had a tendency to inhibit acid secretion, but this was not significant. The lower dose of methiothepin (0.55 /xmol/kg) only slightly inhibited acid secretion (FIGURE 148 26A). These antagonists reversed the 5-HT-induced inhibition of acid secretion by 54-100% with metergoline and methysergide being the least and most effective, respectively (FIGURE 26B). These doses were then used to determine the effect of endogenous 5-HT on vagally-stimulated gastric acid secretion. 149
N 53 13 3 3 3 3 3 A. 40 8 20 < 11 0 Mi twjpa » z * ' -JO i 1 -40 § -eo * -80 -100
B. 40 a 20 o < 0 S -20
-40
-•0 1o * -80 -100 V MS MT MT MG SP 3.5 .55 1.1 1.1 2.5 yUMOL/KQ, CLOSE I.A.
PIQURE 26: A. Effect of local gastric Infusion of various 5-HT antagonists on pentagastrin-stimulated (24 jxg/kg/hr, i.v.) gastric acid secretion in chronic rats. Data are expressed as mean ± SEM. B. Ability of local gastric infusion of various 5-HT antagonists to reverse 5-HT-induced (3.5 /mol/kg, i.v.) attenuation of pentagastrin-stimulated acid secretion in chronic rats. Abbreviations: V, vehicle; MS, methysergide; MT, methiothepin; MG, metergoline; SP, spiperone. The number of animals for each group are given above section A. Data are expressed as mean + SEM; *P<0.05, **P<0.01. 150 B. Effect of Serotonin Antagonists on RX77368~Stimulated Gastric Acid secretion in Chronically Cannulated Rats. Serotonin antagonists were selectively administered to the gastric circulation via the splenic artery. Close i.a. administration of metergoline (1.1 jumol/kg) augmented RX77368- stimulated (100 ng, i.e.) gastric acid secretion by 40% (FIGURE 27). In contrast, neither methiothepin (0.55 /xmol/kg,
5-HT1/2 antagonist, FIGURE 28), methysergide (5-HT1/2 antagonist, FIGURE 29) nor spiperone (5-HT1A antagonist, FIGURE 30) enhanced RX77368-stimulated (100 ng, i.e.) gastric acid secretion. The high dose of methiothepin (1.1 jimol/kg), used in the study of 5-HT antagonism (FIGURE 26), could not be utilized in this phase of the study because of a high rate of mortality observed after intracisternal injection of RX77368. 151
N 300
FIGURE 27: Effect of metergoline on RX77368-stimulated acid secretion in chronic rats. After three 10-min basal periods, animals were pretreated with either vehicle (V) or metergoline (MG, 1.1 /tmol/kg, close i.a.) After three additional 10-min periods, the animal was treated with RX77368 (100 ng, i.e.). Acid secretions were collected for twelve additional 10-min periods. The numbers inside the bars represent the number of animals in each group. Data are expressed as mean + SEM; *P<0.05. 152
« 300
FIGURE 28: Effect of methiothepin on RX77368-stimulated acid secretion in chronic rats. After three 10-min basal periods, animals were pretreated with either vehicle (V) or methiothepin (MT, 0.55 /amol/kg, close i.a.) After three additional 10-min periods, the animal was treated with RX77368 (100 ng, i.e.). Acid secretions were collected for twelve additional 10-min periods. The numbers inside the bars represent the number of animals in each group. Data are expressed as mean + SEM. 153
600
FIGURE 29: Effect of methysergide on RX77368-stimulated acid secretion in chronic rats. After three 10-min basal periods, animals were pretreated with either vehicle (V) or methysergide (MS, 3.5 jtxmol/kg, close i.a.) After three additional 10-min periods, the animal was treated with RX77368 (100 ng, i.e.). Acid secretions were collected for twelve additional 10-min periods. The numbers inside the bars represent the number of animals in each group. Data are expressed as mean ± SEM. 154
IC X N -I O
H Ul K O Ul CO a o < o E i— CO < <9
FIGURE 30: Effect of spiperone on RX77368-stimulated acid secretion in chronic rats. After three 10-min basal periods, animals were pretreated with either vehicle (V) or spiperone (SP, 2.5 |umol/kg, close i.a.) After three additional 10-min periods, the animal was treated with RX77368 (100 ng, i.e.). Acid secretions were collected for twelve additional 10-min periods. The numbers inside the bars represent the number of animals in each group. Data are expressed as mean ± SEM. 155 V. Discussion The main finding of these studies was that metergoline but not methiothepin, methysergide or spiperone enhanced vagally-stimulated gastric acid secretion. Vagal stimulation releases 5-HT into the gastric lumen [(Stephens, 1991; Cho et al. 1985; Stephens and Tach6, 1989; LePard and Stephens, 1992); FIGURE 7] and portal circulation [(Pettersson, 1979; Gronstad et al. 1987; Tobe et al. 1976; Gronstad et al. 1988a; LePard and Stephens, 1993); FIGURE 12]. The dose of intravenous 5-HT effective to inhibit gastric acid secretion (CHAPTER II and FIGURE 26) results in portal 5-HT levels which approximate those seen after vagal stimulation by RX77368 (FIGURE 13). The 5-HT antagonists, if acting purely at the 5-
HT1 receptor mediating inhibition of acid secretion (CHAPTER II), should prevent the inhibitory action exerted by vagally- released endogenous 5-HT, resulting in an enhancement of acid secretion. Only pretreatment with metergoline enhanced RX77368-stimulated acid secretion. A possible explanation for the disparity in the data may reside in the activity of the 5-HT antagonists at other receptor systems. The 5-HT antagonists metergoline and methiothepin share two common features: 1] comparable affinities for 5-HT,, receptors (Miquel and Hamon, 1992; Leysen, 1992), and 2] equal doses required to reverse inhibition of acid secretion by intravenous 5-HT [CHAPTER II, (LePard and Stephens, 1994)]. However, these drugs are not 156 specific for 5-HT, receptors (TABLE 6). For example, compared to its affinity for 5-HT, receptors, methiothepin displays 10- fold higher affinity for a, adrenergic receptors and equal affinity for D2 dopaminergic receptors (Leysen, 1985b). In contrast, metergoline's affinity for a, and D2 receptors is 10- fold less than its affinity for 5-HT, receptors. This may speak to the ability of metergoline (FIGURE 27), but not methiothepin (FIGURE 28), to enhance RX77368-stimulated acid secretion. As to the ineffectiveness of the other antagonists utilized, they also display marked affinities for other receptor systems. Spiperone displays an affinity for D2 receptors which is 100-fold greater than its affinity for 5- HT, receptors. In contrast, methysergide1s affinity for 5-HT, receptors is 10-fold higher than D2 receptors and 100-fold higher than a, receptors [TABLE 6; (Miguel and Hamon, 1992; Leysen, 1992)]. Methysergide has been reported to display nonspecific effects on other neurotransmitter systems in the gut (Hirst and Silinsky, 1975). Effects of these 5-HT, antagonists on other receptor systems likely account for the disparity regarding the effect of inactivating gastric 5-HT receptors on vagally-stimulated gastric acid secretion (FIGURES 29, 30). 157
TABLE 6: Comparison of the affinities (Ki) of various 5-HT antagonists for 5-HT,, a, and D2 receptors. Each + represents a 10-fold change in affinity with (+) representing 100 nM and (+++++) 0.01 nM.
Ei
DRUG 5HU at Da
METERGOLINE <++++) <+++) (+++)
METHIOTHEPIN (++++) <++++•) <•+•+)
METHYSER6IDE (+++) <+) (++)
SPIPERONE (•++) (+*+) <++•+*) 158 The single dose of antagonists used in this study was chosen because each was effective in significantly reversing the inhibition of acid secretion by intravenous 5-HT in both an acute (FIGURE 2) and chronic (FIGURE 26) model. Experiments described in CHAPTER 3 show that the levels of 5- HT in the portal vein after vagal stimulation and intravenous 5-HT administration are very similar (FIGURE 13). As a result, a dose proven effective against intravenous 5-HT should also be effective against 5-HT released by vagal stimulation. In order to minimize the effects of the 5-HT, antagonists on other receptor systems (TABLE 6), the least effective dose of the agents were used in studies assessing the role of vagally-released 5-HT. As mentioned in the overview, depletion of peripheral 5- HT stores is correlated with an increase in vagally-stimulated gastric acid secretion (Stephens et al. 1989; Stephens et al. 1990). The tryptophan hydroxylase inhibitor, para- chlorophenylalanine (PCPA), used in the previous work is a well characterized, relatively selective tool routinely used to deplete endogenous 5-HT (Koe and Weissman, 1966). PCPA does have reported effects on other enzymes. For example, treatment with PCPA simulates phenylketonuria by inhibiting phenylalanine hydroxylase (Lipton et al. 1967). In addition, a preliminary report suggests PCPA may inhibit histidine decarboxylase as implied by a decrease in brain histamine levels (Manon et al. 1970). These effects appear secondary to 159 the profound inhibition of tryptophan hydroxylase by PCPA (Koe and Weissman, 1966; Welch and Welch, 1968; Eide et al. 1988). In contrast to the high affinities (nM) of the 5-HT antagonists utilized in this study for numerous heterologous receptor systems (TABLE 6), PCPA is quite selective. Data obtained from the previous PCPA studies might be weighed more heavily than that obtained from the present work utilizing 5- HT antagonists. To capitalize on the selectivity of PCPA, an attempt was made to solely deplete gastric 5-HT stores by chronic administration of PCPA close to the gastric circulation via the splenic artery cannula. Meaningful depletion of gastric 5-HT stores could not be obtained without significantly depleting brain stores of the amine (personal observation). For example, chronic close gastric treatment with PCPA (15 mg/250 nlf close i.a. for 3 days) decreased stomach 5-HT content only marginally (24%) but significantly reduced brain 5-HT content (49%) [mean + SEM, ng/g wet weight: 1) STOMACH: vehicle (N=3), 4867 + 116; PCPA (N=2), 3685; 2) BRAIN: vehicle (N=3), 1069 ± 94; PCPA (N=2), 543]. In conclusion, previous work suggests that endogenous 5- HT produces an inhibitory tone on vagally-mediated gastric acid secretion. Studies herein do not conclusively support this hypothesis; however, based on the nonselective properties of the antagonists utilized (TABLE 6), definitive progress in answering this question awaits the development of more 160 selective gastric 5-HT antagonists to confirm endogenous 5-HT Indeed produces an inhibitory tone on vagally-stimulated gastric acid secretion. CHAPTER V
Serotonin Inhibits, But 8-OH DPAT Stimulates Gastric Acid Secretion Through a Vagal-Independent, Adrenal Mediated Mechanism
I. Overview The gastrointestinal tract contains approximately 90% of the total body stores of 5-HT. This biogenic amine is present in entero-endocrine cells, myenteric neurons and in mast cells of the rodent gastrointestinal tract (Dhasmana et al. 1983; Gershon, 1991). 5-HT produces effects on many gastrointestinal functional parameters, including gastric acid secretion (Cho and Ogle, 1986b; LePard and Stephens, 1994), Evidence suggests that 5-HT inhibits gastric acid secretion (Canfield and Spencer, 1983; Ormsbee and Fondacaro, 1985; Evans and Gidda, 1993), and produces an inhibitory tone on vagally-stimulated acid output (Stephens et al. 1989; Stephens et al. 1990). Serotonin acts on heterologous receptors in the mammalian gastrointestinal tract, evidence exists for the presence of at least four receptor families [5-HT,, 5-HT2, 5-
HT3 and 5-HT4, (Dhasmana et al. 1983)]. In regard to receptors of the 5-HT, family in the rat stomach, autoradiographic
161 162 studies have indicated the presence of high concentration of
5-HT1a receptors in the lamina propria and myenteric ganglia (Kirchgessner et al. 1993b). However, the role of mucosal or
enteric 5-HT1A receptors in modulating gastric function has not been delineated. A recent report revealed that in contrast to
5-HT or 5-HT, agonists, infusion of the selective 5-HT1A receptor agonist 8-OH-DPAT close intra-arterially to the gastric circulation results in stimulation of gastric acid output (LePard and Stephens, 1994). The present study was
designed to further characterize this novel action of a 5-HT1A agonist to augment gastric acid secretion in the rat.
II. Hypothesis: Intravenous Serotonin and 8-OH-DPAT Produce Opposite Effects on Gastric Function by Acting Through Different Receptors.
III. Methods A. Animals. See Chapter II section III.A. B. Measurement of Gastric Acid Secretion. See Chapter II section III.B. C. Cannulation of the Splenic Artery. See chapter II section III.C. D. surgical Procedures. 1. Vagotomy. In some experiments the vagi were cut bilaterally at the cervical level 30 min before the start of experiments. After ligation and section of the nerve, 163 respiration was maintained through a tracheal cannula connected to a small animal ventilator (SAR-830? CWE, Inc. Ardmore, PA). The basic protocol as described below was followed (section III.E.l.). 2. Celiac Ganglionectomy. Celiac ganglionectomized rats were obtained commercially (Zivic-Miller Laboratories, Zellenopie, PA). The animals were utilized in experiments 2 weeks after the surgical procedure. The basic protocol as described below was followed (section III.E.l.). 3. Adrenalectomy. Bilateral adrenalectomy was performed via a midline incision in rats under urethane anesthesia. The adrenals were isolated and carefully removed with forceps. In the sham controls, the laparotomy was performed, the adrenals exposed but not excised, and the incision was closed. The animals in each group were treated according to the protocol of 8-OH-DPAT challenge (described below) three hours after surgery. The basic protocol as described below was followed (section III.E.l.). E. Protocols. 1. Basic Protocol. After three 10-min periods, pentagastrin (24 /xg/kg/hr, i.v.) was infused via the femoral vein. After nine 10-min periods, the animal was treated with an effective dose of 8-OH-DPAT (3.5 jumol/kg, personal observation) via the femoral vein. Acid secretion was collected for six additional 10-min periods. 164 2. Dose-Response Study. The basic protocol as described above (section III.E.l.) was followed except after nine 10-min periods, the animal was treated i.v. with either 5-HT (0.1, 1, 3, 3.5 and 10 /mol/kg) or 8-OH-DPAT (0.1, 0.2, 0.24, 0.26, 0.28, 0.35, 1, 3.5 and 4.3 /xmol/kg). 3. Antagonist Studies. The basic protocol as described above (section III.E.l.) was followed except after six 10-min periods, the animal was pretreated with either vehicle or the compounds of interest given close intra- arterially (close i.a.) to the gastric circulation via the splenic artery. After three more periods, the animals were challenged with 8-OH-DPAT (3.5 /amol/kg, i.v.). Acid secretion was collected for six additional 10-min periods. 7. Drugs. Serotonin creatine sulfate (Sigma, St. Louis, MO), (+)-8-hydroxy-2-(n-dipropylamino)tetralin HBr [8-OH- DPAT] and Idazoxan HC1 [Research Biochemicals Inc., MA] were dissolved in distilled water. Spiperone HC1 [Research Biochemicals Inc., Natick, MA] was suspended in 0.1% Tween 80. Pentagastrin (Ayerst Laboratories Inc. New York, NY) was diluted using normal saline to the appropriate concentration. G. statistics. In the time course data (FIGURES 31 and 33) each time point is expressed as a % change ± SEM of the average of the three acid secretory measurements taken just before systemic vehicle or drug administration. With respect to the dose response and the antagonist studies (FIGURES 32 and 34) data are expressed as a percent change between the 165 average of the three acid secretory measurements taken 30 min before treatment and the largest incremental change in 10-min acid output during the 30 min period after treatment. In some cases, the 60 min period after 8-OH-DPAT treatment was examined. The data were analyzed utilizing the one way analysis of variance (ANOVA) with post-hoc Student Neuman Keulls (dose-response data) or Dunnet's multiple range tests (time course data). Differences between groups were considered significant if p < 0.05.
IV. Results. A. Comparison of the Effect of Systemic 5-HT and 8-OH- DPAT on Pentagastrin-stimulated Acid Secretion. 1. Time Course. An equimolar dose (3.5 /xmol/kg, i.v.) of 5-HT or 8-OH-DPAT produced inhibition and stimulation, respectively, of gastric acid secretion stimulated by pentagastrin infusion (24 /^g/kg/hr, i.v.) [FIGURE 31]. The inhibitory effect of 5-HT (3.5 /imol/kg) peaked at the 30 min collection and the maximal effect was a 70% reduction in acid output. In contrast, the stimulatory effect of 8-OH-DPAT (3.5 /imol/kg) peaked at the 20 min perfusate analyzed for acid content and a 60% augmentation was produced. In both cases the response was prolonged for the 60 min period analyzed after challenge with these agents. 166
100 DPAT (N-3) 80 VEHICLE (N-3) 60 6HT (N-5) V/5HT/DPAT 40 20
-20 PQ
-60
-80.
-100 -20 0 20 40 60 80 100 120 140 TIME (MIN)
FIGURE 31: Time course of the effect of systemic vehicle, 5-HT or 8-OH-DPAT on pentagastrin-stimulated gastric acid secretion. Pentagastrin (24 jumol/kg, i.v.) infusion began after three 10-min basal periods. After nine 10-min periods, the acid secretory response plateaued in the range of 7-34 /nmol/10 min. At ninety min, the animals were given either vehicle, 5-HT or 8-OH-DPAT via the femoral vein at equimolar doses (3.5 /mol/kg). The acid secretory response was measured for 6 additional 10-min periods. N represents the number of animals in each group. Data are expressed as mean ± SEM. 167 2. Dose-Response. The effects of 5-HT administration were dose-dependent over the range of [0.1 - 10 jumol/kg, Figure 32A]. The peak effect observed (72% inhibition) occurred after the 10 jumol/kg dose. In contrast, 8-OH-DPAT administration produced a steep dose-response profile to stimulate acid secretion (Figure 32B). In the dose range of 0.1 - 0.28 /umol/kg, the full range of 8-OH-DPAT- stimulated acid secretion was observed, with the peak effect (90% stimulation) occurring at the 0.28 /mol/kg dose. Doses exceeding 0.28 /imol/kg produced stimulation ranging from 66 to 88% greater than pretreatment levels. 168
-iw e-HT (UMOL/KO, LVJ
I
-100 8-OH DPAT MKH7KQ, LVJ
FIGURE 32: Dose-response profile of the effects of intravenous (A.) 5-HT or (B.) 8-OH-DPAT on pentagastrin- stimulated gastric acid secretion. Pentagastrin (24 /xmol/kg, i.v.) infusion began after three 10-min basal periods. After nine 10-min periods, the animals were given either vehicle, 5- HT (0.1-10 /zmol/kg) or 8-OH-DPAT (0.1-4.3 /xmol/kg) via the femoral vein. The numbers in the columns represent the number of animals at each dose. Data are expressed as mean + SEM. 169
B. Mechanism of Action of 8-OH-DPAT to Enhance Acid. 1. Effect of vagotomy or celiac ganglionectomy on 8- OH-DPAT-induced stimulation of gastric acid secretion. Characterization of the novel stimulatory effect of 8-OH-DPAT was performed. Bilateral, cervical vagotomy or total celiac ganglionectomy did not attenuate the effect of 8-OH-DPAT (3.5 jimol/kg,i.v.) to augment gastric acid secretion stimulated by pentagastrin [mean ± SEM, % change in acid secretion: sham (n=3), +61 ± 16; vagotomy (n=3), +78 + 10; celiac ganglionectomy (n=3), +147 + 74], 2. Effect of 8-OH-DPAT in adrenalectomized rats. In contrast to its effect in sham treated controls, bilateral adrenalectomy abolished the stimulatory effect of 8-OH-DPAT (3.5 jimol/kg, i.v.). Indeed, a net inhibitory effect was produced as compared to the pretreatment control period [mean + SEM, % change in acid secretion (30 min period): sham (n=3), +38 + 11; adrenalectomy (n=4), -21 + 16; P<0.05; FIGURE 33]. Between 30-60 min after 8-OH-DPAT administration, the disparity between the sham and adrenalectomized animals increased [mean + SEM, % change in acid secretion (60 min period): sham (n=3), +55 ± 23; adrenalectomy (n=4), -58 + 6; PC0.01; FIGURE 33]. 170
SHAM -»-ADX N-4 N-5
200
T •. S 120 111 0z < 80 oX
TIME (MHO
FIGURE 33s Comparison of the effect of systemic 8-OH-DPAT on pentagastrin-stimulated gastric acid secretion in acute adrenalectomized (ADX) versus sham-treated animals. Pentagastrin (24 /xmol/kg» i.v.) infusion began after three 10- min basal periods. After nine 10-min periods, the acid secretory response plateaued in the range of 59-99 (imol/10 min. At ninety min, the animals were given 8-OH-DPAT (3.5 jmol/kg, i.v.). The acid secretory response was measured for 6 additional 10-min periods. N represents the number of animals in each group. Data are expressed as mean + SEM. 171 C. Receptor Subtype Mediating Enhancement in Acid Secretion by 8-OH-DPAT: Spiperone vs. Idaxozan. Since 8-0H-
DPAT can produce actions mediated by both 5-HT1A and a2- adrenergic receptors (Crist and Surprenant, 1987), its effect after close i.a. administration of the 5-HT1A antagonist, spiperone, or the selective a2-antagonist, idazoxan, was examined. The doses utilized were effective to reverse 5-HT- induced effects or activate a2-receptors in the rat after systemic administration (Murphy and Zemlan, 1990; Gartside and Cowen, 1990; Harland and Brown, 1988; LePard and Stephens, 1994). Close i.a. pretreatment with the two agents had no significant effect on pentagastrin-stimulated acid secretion (FIGURE 34A). Spiperone (2.5 /umol/kg) pretreatment reversed the 8-OH-DPAT-induced (0.35 jumol/kg) stimulation by 61% (FIGURE 34B). In contrast, idazoxan (2.5 nmol/kg) pretreatment resulted in a propensity for an enhanced acid response to 8-OH-DPAT [p=0.06, (FIGURE 34B)]. 172 A. 10#
r» A § •• z t§ hi a s • msn o -*• VEH 8P ID 2.5
B.
ID 2.5
FIGURE 34: A. Effect of vehicle, spiperone (2.5 /raol/kg) or idaxozan (2.5 /mol/kg) on pentagastrin-stimulated (24 /xmol/kg/hr, i.v.) acid secretion. B. Ability of local gastric infusion of vehicle, spiperone (2.5 fimol/kg) or idaxozan (2.5 jumol/kg) to reverse 8-OH-DPAT-induced (0.35 /mol/kg, i.v.) enhancement of pentagastrin-stimulated gastric acid secretion. The numbers in A. represent the number of animals in each group. Data are expressed as mean ± SEM; *P<0.05, **P<0.01. 173 V. Discussion The principle findings of this study were the disparate effects of systemic 5-HT and 8-OH-DPAT on gastric acid secretion and the adrenal-dependency of the novel stimulatory effect of 8-OH-DPAT. It is well documented that systemic 8- OH-DPAT stimulates adrenal catecholamine release (Bagdy et al. 1989; Durcan et al. 1991; Chaouloff and Jeanrenaud, 1987; Chaouloff, 1993). The site of action remains controversial, with evidence supporting both central and direct effects of this compound on the adrenal (Chaouloff, 1993). Accordingly, the site of the spiperone-sensitive receptor mediating the response in this report may be peripheral or central. The route of administration assures initial drug exposure to the stomach, however the appearance of spiperone to the central nervous system cannot be excluded. Given that the intact adrenal was essential to elicit the effects of systemic 8-OH-DPAT, it is likely that release of a chemical messenger from the adrenal mediates the gastric stimulatory effect of 8-OH-DPAT. Among the candidates are adrenal-derived steroids, peptides and/or biogenic amines. Adrenal-derived catecholamines or dopamine would be unlikely candidates because they are associated with inhibitory influences on gastric acid secretion. (Burks, 1994). Moreover, epinephrine administration at doses that mimic plasma concentrations reached after cerebro-ventricular CRF administration does not alter pentagastrin-stimulated gastric 174 acid output (Druge et al. 1989). By contrast, certain peptides found in the adrenal medulla can activate receptors which produce stimulatory effects on acid output, for example enkephalins [kappa agonists (Fox et al. 1988)] and atrial naturetic factor (Stapelfeldt et al. 1988). It has been reported that systemic 8-OH-DPAT produces hyperglycemia by activating peripheral (^-adrenoceptors (hyperglycemia presumably mediated by enhanced adrenal-derived plasma catecholamines) (Chaouloff and Jeanrenaud, 1987). In
the present report, the a2-adrenergic antagonist idaxozan augmented the effect of 8-OH-DPAT (FIGURE 34B) without changing the acid response in the 30 min before 8-OH-DPAT administration. The latter finding is strong evidence that in contrast to the case of adrenal catecholamine release,
activation of peripheral a2-adrenoceptors does not mediate the effect of 8-OH-DPAT on acid output. Thus, a spiperone-sensitive receptor mediates 8-OH-DPAT's
effect on acid output. In addition to being a potent 5-HT1A/2 receptor antagonist, spiperone can also interact with a,
adrenergic and D2 dopaminergic systems (Leysen, 1985b).
Antagonism of a, or D2 receptors alone results in inhibition (Pascaud et al. 1982) or is ineffective (Glavin, 1989) in modulating gastric acid output. Spiperone pretreatment in this study produced no significant change in acid output as compared to vehicle in the 30 min before 8-OH-DPAT challenge
(FIGURE 34A). A combined interaction between 5-HT1A receptors 175 and/or adrenergic and/or dopaminergic systems activated by 8- OH-DPAT in enhanced acid cannot be excluded.
Systemic administration of 5-HT1A agonists is associated with inhibition of gastric acid secretion in some models (Evans and Gidda, 1993). These studies utilized conscious pylorus-ligated rats, a model characterized by elevated acid output, indicative of a stomach under the influence of enhanced vagal tone (Brodie, 1966). The effects of systemic
5-HT1a agonists may be different on the stomach dependent on the magnitude of vagal tone. Disparities in pharmacological effects, dependent on the rat model utilized, have been demonstrated before in the study of adrenergic systems (Bernardini et al. 1986). The present investigation suggests that activation of 5-HT1A receptors results in stimulation of gastric acid secretion in the gastric fistula rat through an adrenal-mediated mechanism. LIST OF REFERENCES
Aerts, J.M.F.G. , Donker-Koopman, W.E., Koot, M., Barranger, J.A., Tager, J.M. and Schram, A.W.: Deficient activity of glucocerebrosidase in urine from patients with type 1 Gaucher disease. Clin Chem Acta 158: 155-164, 1986. Ahlman, H., Bhargava, H.N., Dahlstrom, A., Larsson, I., Newson, B. and Pettersson, G.: On the Presence of Serotonin in the Gut Lumen and Possible Release Mechanisms. Acta Physiol Scand 112: 263-269, 1981a. Ahlman, H., DeMagistris, L., Zinner, M. and Jaffe, B.M.: Release of Immunoreactive Serotonin into the Lumen of the Feline Gut in Response to Vagal Nerve Stimulation. Science 213: 1254-1255, 1981b. Ahlman, H. and Dahlstrom, A.: Vagal Mechanisms Controlling Serotonin Release from the Gastrointestinal Tract and Pyloric Motor Function. J Auton Nerv Syst 9: 119-140, 1983. Ahlman, H., Gronstad, K., Nilsson, O. and Dahlstrom, A.: Biochemical and Morphological Studies on the Secretion of 5-HT into the Gut Lumen of the Rat. Biogenic Amines 1: 63-73, 1984. Anderson, C.R., Campbell, G., O'Shea, F. and Payne, M.: The Release of Neuronal 5-HT from the Intestine of a Teleost Fish, Platycephalus bassensis. J Auton Nerv Syst 33: 239-246, 1991. Andersson, K., Mattsson, H. and Larsson, H.: The Role of Gastric Mucosal Histamine in Acid Secretion and Experimentally Induced Lesions in the Rat. Digestion 46: 1-9, 1990. Bagdy, G., Calogero, A.E., Murphy, D.L. and Szemeredi, K.: Serotonin Agonists Cause Parallel Activation of the Sympathoadrenomedu11 ary System and the Hypothalamo-Pituitary-Adrenocortical Axis in Conscious Rats. Endocrinology 125: 2664, 1989. Bech, K.: The Role of Somatostatin and Serotonin in the 6 Adrenoceptor Regulation of Gastric Function. Dan Med Bull 35: 122-140, 1988.
176 177 Bech, K. and Anderson, D.: Exogenous Serotonin and Gastric Mucosal Blood Flow in Conscious Dogs. Scand J Gastroenterol 20: 1285-1290, 1985. Bech, K., Johansen, B. and Lyngso, J.M.: Portal Vein Serotonin (5-HT) and 8 Adrenoceptor Influence on Gastric Secretion. Gastroenterol 102: A721, 1992a. (Abstract) Bech, K., Johansen, B. and Lyngso, J.M.: Portal Vein Serotonin (5-HT) and Inhibition of Gastric Secretion. 2nd International Symposium on Serotonin from Cell Biology to Pharmacology and Therapeutics 53, 1992b. (Abstract) Beck, K.: Effect of Serotonin on Pentagastrin-Stimulated Gastric Acid Secretion and Gastric Antral Motility in Dogs with Gastric Fistula. Scand J Gastroenterol 20: 1115-1123, 1985. Beck, K.: Exogenous Serotonin and Histamine-Stimulated Gastric Acid and Pepsin Secretion in Dogs. Scand J Gastroenterol 21: 1205-1210, 1986a. Beck, K.: Effect of Serotonin on Bethanechol-Stimulated Gastric Acid Secretion and Gastric Antral Motility in Dogs. Scand J Gastroenterol 21: 655-661, 1986b. Berglindh, T.: Gastric Glands and Cells: Preparation and In Vitro Methods. Methods in Enzymology 192: 93-107, 1990. Bernardini, C., Del Tacca, M., Soldani, G. and Martinotti, E.: The Effects of a2 Agonists and Antagonists on Gastric Acid Secretion. Arch Int Pharmacodyn Ther 281: 134, 1986. Black, J.W., Fisher, E.W. and Smith, A.N.: The Effects of 5-Hydroxytryptamine on Gastric Secretion in Anaesthetized Dogs. J Physiol 141: 27-34, 1958. Blackshaw, L.A. and Grundy, D.: Effects of 5-Hydroxytryptamine on Discharge of Vagal Mucosal Afferent Fibres from the Upper Gastrointestinal Tract of the Ferret. J Auton Nerv Syst 45: 41-50, 1993a. Blackshaw, L.A. and Grundy, D.: Effects of 5-Hydroxytryptamine (5-HT) on the Discharge of Vagal Mechanoreceptors and Motility in the Upper Gastrointestinal Tract of the Ferret. J Auton Nerv Sys't 45: 51-59, 1993b. Bockaert, J., Fozard, J.R., Dumuis, A. and Clarke, D.E.: The 5-HT4 receptor: a place in the sun. Trends in Pharmacol Sci 13: 141-145, 1992. 178 Branchek, T., Kates, M. and Gershon, M.D.: Enteric Receptors for 5-Hydroxytryptamine. Brain Res 324: 107-118, 1984. Branchek, T., Mawe, G.M. and Gershon, M.D.: Characterization and Localization of a Peripheral Neural 5-Hydroxytryptamine Receptor Subtype with a Selective Agonist, 3H-Hydroxyindalpine. J Neurosci 8: 2582-2595, 1988. Branchek, T.A. and Gershon, M.D.: Development of neural receptors for serotonin in the murine bowel. J Comp Neurol 258: 597-610, 1987. Brodie, D.A.: The mechanism of gastric hyperacidity produced by pylorus ligation in the rat. Am J Dig Dis 11: 231-241, 1966. Brown, M.R.: Thyrotropin releasing factor: A putative CNS regulator of the autonomic nervous system. Life Sci 28: 1789-1795, 1981. Buchheit, K.H., Costall, B., Engel, G., Gunning, S.J., Naylor, R.J. and Richardson, B.P.: 5-Hydroxytryptamine Receptor Antagonism by Metoclopramide and ICS 205-930 in the Guinea-Pig Leads to Enhancement of Contractions of the Stomach Muscle Strips Induced by Electrical Field Stimulation and Facilitation of Gastric Emptying in-vivo. J Pharm Pharmacol 37: 664-667, 1985. Bueno, L., Fargeas, M.J., Theodorou, V. and Fioramonti, J.: Involvement of 5-Hydroxytryptamine in the Intestinal Motor Disturbances Induced by Mast Cell Degranulation in Rats. Eur J Pharmacol 192(2): 263-269, 1991. Bulbring, E. and Gershon, M.D.: 5-Hydroxytryptamine Participation in the Vagal Inhibitory Innervation of the Stomach. J Physiol 192: 823-846, 1967. Bulbring, E. and Lin, R.C.Y.: The Effect of Intraluminal Application of 5-Hydroxytryptamine and 5-Hydroxytryptophan on Peristalsis; the Local Production of 5-HT and its Release in Relation to Intraluminal Pressure and Propulsive Activity. J Physiol 140: 381-407, 1958. Burczynski, F.J., Pushka, K.L., Sitar, D.S. and Greenway, C.V.: Hepatic Plasma Flow: Accuracy of Estimation from Bolus Injections of Indocyanine Green. Am J Physiol 252: H953-H962, 1987. Burks, T.F.: Neurotransmitters and Neurohormones. in: Physiology of the Gastrointestinal Tract, ed. L.R.Johnson (Raven Press, New York) p. 112, 1994. 179 Burns, E., Ball, C.E., Christie, J.P., Broadhead, G.D., Tucker, G.T. and Bax, N.D.S.: Direct and Indirect Measurement of the Hepatic Extraction Ratio of Indocyanine Green in the Rat. Clin Sci 76: 503-508, 1989. Butler, A., Hill, J.M., Ireland, S.J., Jordan, C.C. and Tyers, M.B.: Pharmacological Properties of GR38032F, a Novel Antagonist at 5-HT, Receptors. Br J Pharmacol 94: 397-412, 1988. Butler, A., Elswood, C.J., Burridge, J., Ireland, S.J., Bunce, K.T., Kilpatrick, G.J. and Tyers, M.B.: The Pharmacological Characterization of 5-HT3 Receptors in Three Isolated Preparations Derived from Guinea-Pig Tissues. Br J Pharmacol 101: 591-598, 1990. Canfield, S.P. and Spencer, J.E.: The Inhibitory Effects of 5-Hydroxytryptamine on Gastric Acid Secretion by the Rat Isolated Stomach. Br J Pharmacol 78: 123-129, 1983. Champaneria, S., Costall, B., Naylor, R.J. and Robertson, D.W.: Identification and Distribution of 5-HT, Recognition Sites in the Rat Gastrointestinal Tract. Br J Pharmacol 106: 693-696, 1992. Chaouloff, F.: Physiopharmacological Interactions Between Stress Hormones and Central Serotonergic Systems. Brain Res Rev 18: 1-32, 1993.
Chaouloff, F. and Jeanrenaud, B.: 5-HT1A and a2 Adrenergic Receptors Mediate the Hyperglycemic and Hypoinsulinemic Effects of 8-Hydroxy-2-(di-n-Propylamino)Tetralin in the Conscious Rat. J Pharmacol Exp Ther 243: 1159, 1987. Cho, C.C., Hung, K.M. and Ogle, C.W.: The Aetiology of Gastric Ulceration Induced by Electrical Vagal Stimulation in Rats. Eur J Pharmacol 110: 211-217, 1985. Cho, C.H., Hung, K.M. and Ogle, C.W.: The Aetiology of Gastric Ulceration Induced by Electrical Vagal Stimulation in Rats. Eur J Pharmacol 110: 211-217, 1985. Cho, C.H. and Ogle, C.W.: The inhibitory action of 5-hydroxytryptamine on gastric secretory function in rats. Br J Pharmacol. 87: 371-377, 1986a. Cho, C.H. and Ogle, C.W.: The Inhibitory Action of 5-Hydroxytryptamine on Gastric Secretory Function in Rats. Br J Pharmacol 87: 371-377, 1986b. Chuang, C.-N., Tanner, M., Chen, M.C.Y., Davidson, S. and Soil, A.H.: Gastrin Induction of Histamine Release from 180 Primary Cultures of Canine Oxyntic Mucosal Cells. Am J Physiol 263: G460-G465, 1992. Chuang, C.-N., Tanner, M., Lloyd, K.C.K., Wong, H. and Soil, A.H.: Endogenous Somatostatin Inhibits Histamine Release from Canine Gastric Mucosal Cells in Primary Culture. Am J Physiol 265: G521-G525, 1993.
Clarke, D.E., Craig, D.A. and Fozard, J.R.: The 5-HT4 Receptor : Naughty, but Nice. Trends in Pharmacological Sciences 10: 385-386, 1989. Cohen, M.L. and Flundzinzki, L.A.: Contractile Serotonergic Receptor in Rat Stomach Fundus. J Pharmacol Exp Ther 243: 264-269, 1987. Cohen, M.L., Nelson, D.L., Wainscott, D.B., Schenck, K., Kursar, D. and Baez, M.: Relationship of a Novel Serotonergic Receptor (5-HT2F) to the Contractile Serotonergic Receptor in the Rat Stomach Fundus. 2nd International Symposium on Serotonin from Cell Biology to Pharmacology and Therapeutics 50, 1992. (Abstract) Costall, B., Gunning, S.J., Naylor, R.J. and Tyers, M.B.: The Effect of GR38032F, a Novel 5-HT3 Receptor Antagonist on Gastric Emptying in the Guinea Pig. Br J Pharmacol 91: 263-264, 1987. Craig, D.A. and Clarke, D.E.: Pharmacological Characterization of a Neuronal Receptor for 5-Hydroxytryptamine in Guinea Pig Ileum with Properties Similar to the 5-Hydroxytryptamine4 Receptor. J Pharmacol Exp Ther 252: 1378-1386, 1990. Craig, D.A., Eglen, R.M., Walsh, L.K.M., Perkins, L.A., Whiting, R.L. and Clarke, D.E.: 5-Methoxytryptamine and 2-Methyl-5-Hydroxytryptamine-Induced Desensitization as a Discriminative Tool for the 5-HT3 and Putative 5-HT4 Receptors in the Guinea Pig Ileum. Naunyn-Schmied Arch Pharmacol 342: 9-16, 1990. Crist, J. and Surprenant, A.: Evidence that 8-Hydroxy-2-(N-dipropylamino)tetralin is a selective a2 adrenoceptor antagonist on guinea pig submucous plexus neurons. Br J Pharmacol 92: 341-347, 1987. Curtis, G.H., MacNaughton, W.K. and Gall, D.G.: Effect of Cyclooxygenase Inhibition on Macromolecular Transport in Rat Gastric Mucosa. Am J Physiol 265: G1135-G1140, 1994. de Ridder, W.J.E. and Schuurkes, J.A.J.: Cisapride and 5-Hydroxytryptamine Enhance Motility in the Canine Antrum via 181
Separate Pathways, not Involving 5-Hydroxytryptamine, 234 Receptors. J Pharmacol Exp Ther 264: 79-88, 1993. '' ' Dhasmana, K.M., Zhu, Y.N., Cruz, S.L. and Villalon, C.H.: Gastrointestinal Effects of 5-Hydroxytryptamine and Related Drugs. Life Sci 53: 1651-1661, 1983. Dhasmana, K.M., Villalon, C.M., Zhu, Y.N., Tadipdtri, S. and Saxena, P.R.: Role of 5-HTj-like Receptors in the Increase in Intragastric Pressure Induced by 5-Hydroxytryptamine in the Rat. Eur J Pharmacol 213: 293-299, 1992. Donald, D.E. and Yipintsoi, T.: Comparison of Measured and Indocyanine Green Blood Flows in Various Organs and Systems. Mayo Clin Proc 48: 492-500, 1973. Drapanas, T., McDonald, J.C. and Stewart, J.D.: Serotonin Release Following Instillation of Hypertonic Glucose into the Proximal Intestine. Ann Surg 156: 528-556, 1962. Druge, G., Raedler, A., Greten, H. and Lenz, H.J.: Pathways Mediating CRF-Induced Inhibition of Gastric Acid Secretion in Rats. Am J Physiol 256: G214, 1989. Durcan, M.J., Wozniak, K.M. and Linnoila, M.: Modulation of the Hypothermic and Hyperglycemic Effects of 8-OH-DPAT by a2 Adrenoceptor Antagonists. Br J Pharmacol 102: 222-226, 1991. Eide, P.K., Hole, K., Berge, O.-G. and Broch, O.J.: 5-HT Depletion with 5,7-DHT, PCA and PCPA in Mice: Differential Effects on the Sensitivity to 5-MeODMT, 8-OH-DPAT and 5-HTP as Measured by Two Nociceptive Tests. Brain Res 440: 42-52, 1988.
Erikson, A.: Gaucher disease. Norrbottnian Type (III). Neuropaediatric and neurobiological aspects of clinical patterns and treatment. Acta Paediatr.Scand. 326: 7-42, 1986. Espamer, V. and Asero, B.: Identification of Enteramine, the Specific Hormone of the Enterochromaffin Cell System as 5-Hydroxytryptamine. Nature 169: 800-801, 1952. Evans, D.C. and Gidda, J.S.: Serotonin (5-HT) Inhibits Gastric Acid Secretion Via 5-HT- Receptors in the Rat. Gastroenterol 104: A76, 1993. (Abstract) Ferrara, A., Zinner, M.J. and Jaffe, B.M.: Intraluminal Release of Serotonin, Substance P, and Gastrin in the Canine Small Intestine. Dig Dis Sci 32: 289-294, 1987. 182 Forsberg, E.J. and Miller, R.J.: Cholinergic Agonists Induce Vectorial Release of Serotonin from Duodenal Enterochromaffin Cells. Science 217: 355-356, 1982. Forsberg, E.J. and Miller, R.J.: Regulation of Serotonin Release from the Rabbit Intestinal Enterochromaffin Cells. J Pharmacol Exp Ther 227: 755-766, 1983. Fox, D.A. and Burks, T.F.: Roles of Central and Peripheral Mu, Delta and Kappa Opioid Receptors in the Mediation of Gastric Acid Secretory Effects in the Rat. J Pharmacol Exp Ther 244: 456, 1988. Fozard, J.R., Mir, A.K. and Middlemiss, D.N.: Cardiovascular Response to 8-Hydroxyl-2-(di-n-Propylamino)Tetralin (8-OH-DPAT) in the Rat: Site of Action and Pharmacological Analysis. J Cardiovasc Pharmacol 9: 328-347, 1987. Frazer, A., Maayani, S. and Wolfe, B.B.: Subtypes of Receptor for Serotonin. Ann Rev Pharmacol Toxicol 30: 307-348, 1990. Fujita, T. and Kobayashi, S.: The Cells and Hormones of the GEP Endocrine System. In: Gastro-Entero-Pancreatic Endocrine System, ed. by T. Fujita, pp. 1-16, The Williams and Wilkins Company, Baltimore, 1974. Furman, B.L. and Waton, N.G.: 5-Hydroxytryptamine and Peripheral Secretory Mechanisms. In: The Peripheral Actions of 5-Hydroxytryptamine, ed. by J.R. Fozard, pp. 274-300, Oxford University Press, New York, 1989. Galligan, J.J., Surprenant, A., Tonini, M. and North, R.A.: Differential Localization of 5-HT, Receptors on Myenteric and Submucosal Neurons. Am J Physiol 255: G603-G611, 1988. Galligan, J.J.: Differential Inhibition of cholinergic and Noncholinergic Neurogenic Contractions by 5-Hydroxytryptamine1A Receptor Agonists in Guinea Pig Ileum. J Pharmacol Exp Ther 260: 306-312, 1992. Gardiner, S.M., Kemp, P.A., Bennett, T., Palmer, R.M.J, and Moncada, S.: Regional and Cardiac Hemodynamic Effects of NG,NG, Dimethyl-L-Arginine and Their Reversibility by Vasodilators in Conscious Rats. Br J Pharmacol 110: 1457-1464, 1993. Gartside, S.E. and Cowen, P.J.: Mediation of ACTH and Prolactin Responses to 5-HTP by 5-HT2 Receptors. Eur J Pharmacol 179: 103-109, 1990.
•Gebauer, A., Merger, M. and Kilbinger, H.: Modulation by 5-HT3 and 5-HT4 Receptors of the Release of 5-Hydroxytryptamine from 183 the Guinea-Pig Small Intestine. Naunyn-Schmied Arch Pharmacol 347: 137-140, 1993. Gerber, J.G. and Payne, N.A.: The Role of Gastric Secretagogues in Regulating Gastric Histamine Release In Vivo. Gastroenterol 102: 403-408, -1992. Gershon, M.D., Mawe, G.M. and Branchek, T.A.: 5-Hydroxytryptamine and Enteric Neurones. In: The Peripheral Actions of 5-Hydroxytryptamine, ed. by J.R. Fozard, pp. 248-273, Oxford University Press, New York, 1989. Gershon, M.D.: Serotonin: Its Role and Receptors in Enteric Neurotransmission. In: Kynurenine and Serotonin Pathways, ed. by R. Schwarcz, S.N. Young and R.R. Brown, pp. 221-230, Plenum Press, New York, 1991. Gidda, J.S., Evans, D.C., Prime, P., Schenck, K. and Cohen, M.L.: Role of 5-HT-3 Receptor Antagonists in Gastrointestinal Motility. Gastroenterol 95: 867, 1988. (Abstract) Gillis, C.N.: Peripheral Metabolism of Serotonin. In: Serotonin and the Cardiovascular System, ed. by P.M. Vanhoutte, pp. 27-36, Raven Press, New York, 1985. Glavin, G.: Activity of Selective Dopamine DA-1 and DA-2 Agonists and Antagonists on Experimental Gastric Lesions and Gastric Acid Secretion. J Pharmacol Exp Ther 251: 726, 1989. Glennon, R.A., Naiman, N.A., Pierson, M.E., Titeler, M., Lyon, R.A. and Weisberg, E.: NAN-190: an Arylpiperazine Analog that Antagonizes the Stimulus Effects of the 5-HT1A Agonist 8-Hydroxy-2-(di-n-Proplamino)Tetralin (8-OH-DPAT). Eur J Pharmacol 154: 339-341, 1988. Gradin, K., Pettersson, A., Hedner, T. and Persson, B.: Acute Administration of 8-Hydroxy-2-(Di-n-Propylamino)Tetralin (8-OH-DPAT), a Selective 5-HT-Receptor Agonist, Causes a Biphasic Blood Pressure Response and a Bradycardia in the Normotensive Sprague-Dawley Rat and in the Spontaneously Hypertensive Rat. J Neural Transm 62: 305-319, 1985. Gregg, R.V.: The Gastric Argentaffin Cell Population of the Rat. J Morph 119: 81-88, 1966. Gronstad, K., Dahlstrom, A., Florence, L., Zinner, M.J., Ahlman, J. and Jaffe, B.M.: Regulatory Mechanisms in Endoluminal Release of Serotonin and Substance P from the Feline Jejunum. Dig Dis Sci 32: 393-400, 1987. 184 Gronstad, K.O., Dahlstrom, A., Jaffe, B.M. and Ahlman, H.: Regional and Selective Changes in Blood Flow of the Feline Small Intestine Induced by Endoluminal Serotonin. Acta Physiol Scand 127: 207-213, 1986. Gronstad, K.O., Zinner, M.J., Nilsson, O., Dahlstrom, A., Jaffe, B.M. and Ahlman, H.: Vagal Release of Serotonin into Gut Lumen and Portal Circulation via Separate Control Mechanisms. J Surg Res 44: 146-151, 1988a. Gronstad, K.O., Ahlund, L., Dahlstrom, A., Haggendal, J. and Ahlman, H.: A Possible Mechanism for the Release of Serotonin from the Gut Caused by Pentagastrin. J Surg Res 44: 473-478, 1988b. Gullikson, G.W., Loeffler, R.F. and Virina, M.A.: Relationship of Serotonin3 Receptor Antagonist Activity to Gastric Emptying and Motor-Stimulating Actions of Prokinetic Drugs in Dogs. J Pharmacol Exp Ther 258: 103-110, 1991. Gullikson, G.W., Virina, M.A., Loeffler, R.F., Yang, D-C., Goldstin, B., Wang, S.-X., Moummi, C., Flynn, D.L. and Zabrowski, D.L.: SC-49518 Enhances Gastric Emptying of Solid and Liquid Meals and Stimulates Gastrointestinal Motility in Dogs by a 5-Hydroxytryptamine4 Receptor Mechanism. J Pharmacol Exp Ther 264: 240-248, 1993. Hakanson, R.: New Aspects of the Formation and Function of Histamine, 5-Hydroxytryptamine and Dopamine in Gastric Mucosa. Acta Physiol. Scand. Suppl. 340: 1-34, 1970. Hannun, Y.A. and Bell, R.M.: Lysosphingolipids inhibit protein kinase C: Implications for the sphingolipidoses. Science 235: 670-674, 1987. Harland, D. and Brown, M.J.: Effects of Acute and Chronic Administration of Idazoxan on Blood Pressure and Plasma Catecholamine Concentrations of Rats. J Pharmacol Exp Ther 245: 265, 1988. Hasegawa, H., Yanagisawa, M., Inoue, F., Yanaihara, N. and Ichiyama, A.: Demonstration of Non-Neural Tryptophan 5-mono-oxygenase in Mouse Intestinal Mucosa. Biochem J 248: 501-509, 1987. Hashizume, T., Hirokawa, H., Aibara, S., Ogawa, H. and Kasahara, A.: Pharmacological and Histological Studies of Gastric Mucosal Lesions Induced by Serotonin in Rats. Arch Int Pharmacodyn Ther 236: 96-108, 1978. 185 Heap, B.J. and Kiernan, J.A.: Histological, Histochemical and Pharmacological Observation on Mast Cells in the Stomach of the Rat. J Anat 115: 315-325, 1973. Heuser, L., Gregg, R. and Max, M.H.: The Association of Pyloric Antral Tissue Levels of 5-Hydroxytryptamine and Restraint Stress Ulceration in the Rat. Am Surg 45: 771-773, 1979. Hirst, G.D.S. and Silinsky, E.M.: Some Effects of 5-Hydroxytryptamine, Dopamine and Noradrenaline on Neurones in the Submucous Plexus of the Guinea-Pig Small Intestine. J Physiol 251: 817-832, 1975. Hopkinson, G.B., Hinsdale, J. and Jaffe, B.M.: Contraction of Canine Stomach and Small Bowel by Intravenous Administration of Serotonin A Physiological Response?. Scand J Gastroenterol 24: 923-932, 1989. Horita, A. and Carino, M.A.: -Centrally administered thyrotropin-releasing hormone (TRH) Stimulates colonic transit and diarrhea production by a vagally mediated serotonergic mechanism in the rabbit. J Pharmacol Exp Ther 222: 367-371, 1982. Horita, A., Carino, M.A. and Pae, Y.-S.: Blockade by Naloxone and Naltrexone of the TRH-Induced Stimulation of Colonic Transit in the Rabbit. Eur J Pharmacol 108: 289-293, 1985. Horita, A., Carino, M.A. and Mori, T.: Morphine Antagonizes TRH-Induced Intestinal Transit and Hyperserotonemia in Rabbits. Eur J Pharmacol 145: 55-59, 1988. Hoyer, D., Clarke, D.E., Fozard, J.R., Hartig, P.R., Martin, G.R., Mylecharane, E.J., Saxena, P.R. and Humphrey, P.A.: International Union of Pharmacology Classification of Receptors for 5-Hydroxytryptamine (Serotonin). Pharmacol Rev 46: 157-203, 1994. Hsu, H.-C., Flancbaum, L.J., Kasziba, E., Merrill, G.F. and Fisher, H.: Roles of In Vitro- and In Vivo-Administered Histamine and Serotonin in Compound 48/80-Induced Gastric Acid Secretion in Isolated, Perfused Rat Stomach. Dig Dis Sci 36: 1708-1714, 1991. Humphrey, P.P.A.: 5-Hydroxytryptamine Receptors and Drug Discovery. In: Serotonin Receptor Subtypes: Pharmacological Significance and Clinical Implications. International Academy for Biomedical and Drug Research, ed. by S.Z. Langer, N. Brunello, G. Racagni and J. Mendlewicz, pp. 129-139, Karger, Basel, 1992. 186 Ireland, S.J.: Origin of 5-Hydroxytryptamine-induced Hyperpolarization of the Rat Superior Cervical Ganglion and Vagus Nerve. Br J Pharmacol 92: 407-416, 1987a. Ireland, S.J., Straughan, D.W. and Tyers, M.B.: Influence of 5-Hydroxytryptamine Uptake on the Apparent 5-Hydroxytryptamine Antagonist Potency of Metoclopramide in the Rat Isolated Superior Cervical Ganglion. Br J Pharmacol 90: 151-160, 1987b.
Ireland, S.J. and Tyers, M.B.: Pharmacological Characterization of 5-Hydroxytryptamine-Induced Depolarization of the Rat Isolated Vagus Nerve. Br J Pharmacol 90: 229-238, 1987c. Izzat, A. and Waton, N.G.: Release of 5-Hydroxytryptamine by Pentagastrin and Its Role in the 'Fade' of Stimulated Gastric Secretion in Cats. J Physiol 383: 499-507, 1987. Jacoby, H.I., Bonfilio, A.C. and Raffa, R.B.: Central and Peripheral Administration of Serotonin Produces Opposite Effects on Mouse Colonic Motility. Neuroscience Letters 122: 122-126, 1991. Jaffe, B.M., Kopen, D.F. and Lazan, D.W.: Endogenous Serotonin in the Control of Gastric Acid Secretion. Surgery 82: 156, 1977. Johansen, B. and Bech, K.: BRL 24924, a 5-Hydroxytryptamine Type 3 Antagonist, and Gastric Secretion of Acid and Pepsin in vivo. Digestion 48: 121-127, 1991. Kellum, J., McCabe, M., Schneier, J. and Donowitz, M.: Neural Control of Acid-Induced Serotonin Release from the Rabbit Duodenum. Am J Physiol 245: G824-G831, 1983. Kilbinger, H. and Pfeuffer-Friederich, I.: Two Types of Receptors for 5-Hydroxytryptamine on Cholinergic Nerves of the Guinea Pig Myenteric Plexus. Br J Pharmacol 85: 529-539, 1985.
Kilbinger, H. and Wolf, D.: Effects of 5-HT4 Receptor Stimulation on Basal and Electrically Evoked Release of Acetylcholine from Guinea-Pig Myenteric Plexus. Naunyn-Schmied Arch Pharmacol 345: 270-275, 1992. Kimura, R.E., LaPine, T.R. and Gooch, W.M.: Portal Venous and Aortic Glucose and Lactate Changes in a Chronically Catheterized Rat. Ped Res 23: 235-240, 1988a. Kimura, R.E., LaPine, T.R., Johnston, J. and Ilich, J.Z.: The Effect of Fasting on Rat Portal Venous and Aortic Blood 187 Glucose, Lactate, Alanine, and Glutamine. Ped Res 23: 241-244, 1988b. Kirchgessner, A.L., Tarair, H. and Gershon, M.D.: Identification and Stimulation by Serotonin of Intrinsic Sensory Neurons of the Submucosal Plexus of the Guinea Pig Gut: Activity-Induced Expression of Fos Immunoreactivity. J Neurosci 12: 235-248, 1992. Kirchgessner, A.L., Liu, M.-T. and Gershon, M.D.: Detection of the 5-HTlfl Receptor and 5-HT,. Receptor mRNA in the Rat Bowel and Pancreas: Comparison witn 5-HT1p Receptors. J Comp Neurol 327: 233-250, 1993a. Kirchgessner, A.L., Liu, M.-T., Howard, M.J. and Gershon, M.D.: Detection of the 5-HT1A Receptor and 5-HT1A Receptor mRNA in the Rat Bowel and Pancreas: Comparison with 5-HT1p Receptors, J Comp Neurol 327: 233-250, 1993b. (Abstract) Kitajima, M., Shimizu, A., Sakai, N., Otsuka, S., Mogi, M., Nakajima, M., Kiuchi, T., Ikeda, Y. and Oshima, A.: Gastric Microcirculation and Its Regulating Factors in Stress. J Clin Gastroenterol 13 (Suppl. 1): S9-S17, 1991. Kobayashi, S. and Fujita, T.: Emiocytotic Granule Release in the Basal-Granulated Cells of the Dog Induced by Intraluminal Application of Adequate Stimuli. In: Gastro-Entero-Pancreatic Endocrine System, ed. by T. Fujita, pp. 49-58, The Williams and Wilkins Company, Baltimore, 1974. Kobayashi, T., Hasegawa, H., Kaneko, E. and Ichiyama, A.: Gastrointestinal Serotonin: Depletion due to Tetrahydrobiopterin Deficiency Induced by 2,4-Diamino-6-hydroxypyrimidine Administration. J Pharmacol Exp Ther 1991.(In Press) Koe, B.K. and Weissman, A.: p-Chlorophenylalanine: A Specific Depletor of Brain Serotonin. J Pharmacol Exp Ther 154: 499-516, 1966. Koop, H. and Arnold, R.: Serotonergic Control of Somatostatin and Gastrin Release form the Isolated Rat Stomach. Regul Pept 9:101-108, 1984. Korpi, E.R.: Serotonin Determined in Whole Blood by Liquid Chromatography with Electrochemical Detection. Clin Chem 30: 487-488, 1984. Kuemmerle, J.F.,Kraus, H. and Kellum, J.M.: Serotonin Release is Mediated by Muscarinic Receptors on Duodenal Mucosal Cells. J Surg Res 43: 139-142, 1987. 188 Kuemmerle, J.F., Smith, E.H., Borum, E.H. and Kellum, J.M.: 6 Adrenoceptors on Duodenal Mucosal Mast Cells Mediate Venous Serotonin Release. J Surg Res 44: 740-744, 1988. Larsson, I., Ahlman, H., Bhargava, H.N., Dahlstrom, A., Pettersson, G. and Kewenter, J.: The Effects of Splanchnic Nerve Stimulation on the Plasma Levels of Serotonin and Substance P in the Portal Vein of the Cat. J Neural Transm 46: 105-112, 1979. LePard, K.J. and Stephens, R.L.: Serotonin (5-HT) Inhibits Gastric Acid Secretion by Nonluminal, Vagal Independent Mechanisms. Society for Neuroscience Abstracts 18: 1529, 1992. (Abstract) LePard, K.J. and Stephens, R.L.: Vagally-Mediated Serotonin (5HT) Release Inhibits Gastric Acid Secretion Via Receptors of the 5-HT. Family. Society for Neuroscience Abstracts 19: 962, 1993. (Abstract) LePard, K.J. and Stephens, R.L.: Serotonin Inhibits Gastric Acid Secretion Through a 5-Hydroxytryptamine1 -1 ike Receptor in the Rat. J Pharmacol Exp Ther 270: 1-6, 1994. Leung, F.W., Kauffman, G.L.Jr., Washington, J., Scremin, O.U. and Guth, P.H.: Blood Flow Limitation of Stimulated Gastric Acid Secretion in the Rat. Am J Physiol 250: G794-G799, 1986. Leysen, J.E., Gommeren, W., Van Gompel, J., Janssen, P.F.M. and Laduron, P.M.: Receptor-Binding Properties in Vitro and in Vivo of Ritanserin. Mol Pharmacol 27: 600-611, 1985a. Leysen, J.E.: Serotonergic Binding Sites. In: Serotonin and the Cardiovascular System, ed. by P.M. Vanhoutte, pp. 43-62, Raven Press, New York, 1985b.
Leysen, J.E.: 5-HTz Receptors: Location, Pharmacological, Pathological and Physiological Role. In: Serotonin Receptor Subtypes: Pharmacological Significance and Clinical Implication. International Academy for Biomedical and Drug Research, ed. by S.Z. Langer, N. Bruneilo, G. Racagni and J. Mendlewicz, pp. 31-43, Karger, Basel, 1992. Lipton, M.A., Gordon, R., Guroff, G. and Udenfriend, S.: p-Chlorophenylalanine-Induced Chemical Manifestations of Phenylketonuria in Rats. Science 156: 248-250, 1967. Lopez-Belmonte, J., Whittle, B.J.R. and Moncada, S.: The Actions of Nitric Oxide Donors in the Prevention or Induction of Injury to the Rat Gastric Mucosa. Br J Pharmacol 108: 73-78, 1993. 189
Lyngso, J.M., Johansen, B. and Bech, K.: 5-HTu Receptors and Gastric Acid Secretion In Vivo. 2nd International Symposium on Serotonin from Cell Biology to Pharmacology and Therapeutics 49, 1992. (Abstract) Manon, M.K., Aures, D. and Clark, W.G.: Lowering of Brain Histamine (Hm) by Parachlorophenylalanine (PCPA) and a New Histidine Decarboxylase Inhibitor. The Pharmacologist 12: 205, 1970. (Abstract) Mark, T., Dominguez, C. and Rywlin, A.M.: Gaucher's disease associated with chronic lymphocytic leukemia. South.Med.J. 75: 361-363, 1982. Martin, D.C., Magnant, A.D. and Kellum, J.M.: Luminal Hypertonic Solutions Stimulate Concentration-Dependent Duodenal Serotonin Release. Surgery 106: 325-331, 1989. Mastai, R., Giroux, L., Semret, M. and Huet, P.-M.: Ritanserin Decreases Portal Pressure in Conscious and Unrestrained Cirrhotic Rats. Gastroenterol 98: 141-145, 1990. Mate, L., Poston, G.J. and Thompson, J.C.: Serotonin. In: , , , 1980. Mawe, G.M., Branchek, T.A. and Gershon, M.D.: Peripheral Neural Serotonin Receptors : Identification and Characterization with Specific Antagonists and Agonists. Proc Natl Acad Sci USA 83: 9799-9803, 1986. Mccall, R.B., Patel, B.N. and Harris, L.T.: Effects of Serotonin., and Serotonin2 Receptor Agonists and Antagonists on Blood Pressure, Heart Rate and Sympathetic Nerve Activity. J Pharmacol Exp Ther 242: 1152-1159, 1987. McGregor, G.P. and Conlon, J.M.: Regulatory Peptide and Serotonin Content and Brush-Border Enzyme Activity in the Rat Gastrointestinal Tract Following Neonatal Treatment with Capsaicin; Lack of Effect on Epithelial Markers. Regul Pept 32: 109-119, 1991. McMenamin, C.C., Gault, E.A. and Haig, D.M.: The Effect of Dexamethasone on Growth and Differentiation of Bone-Marrow Derived Mucosal Mast Cells in vitro. Immunology 62: 29-34, 1987. Meulemans, A.L., Helsen, L.F. and Schuurkes, J.A.J.: The Role of Nitric Oxide (NO) in 5-HT-Induced Relaxations of the Guinea-Pig Stomach. Naunyn-Schmied Arch Pharmacol 348: 424-430, 1993. 190
Miquel, M.C. and Hamon, M.: 5-HT1 Receptor Subtypes: Pharmacological Heterogeneity. In: Serotonin Receptor Subtypes: Pharmacological Significance and Clinical Implications. International Academy for Biomedical and Drug Research, ed. by S.Z. Langer, N. Brunello, G. Racagni and J. Mendlewicz, pp. 13-30, Karger, Basel, 1992. Miyata, K., Kamato, T., Nishida, A. and Honda, K.: Studies on the Mechanism for the Gastric Mucosal Protection by Famotidine in Rats. Jpn J Pharmacol 55: 211-222, 1991. Mizutani, M., Neya, T. and Nakayama, S.: Ascending Contraction Mediated by 5-Hydroxytryptamine3 Receptors in Canine Small Intestine. Am J Physiol 263: G306-G311, 1992. Money, S.R., Kimura, K., Petroianu, A. and Jaffe, B.M.: Effects of Intravenous Calcium on Release of Serotonin into Jejunal Lumen and Portal Circulation. Dig Dis Sci 33: 977-981, 1988.
Moser, P.C.: The Effect of Putative 5-HT1A Receptor Antagonists on 8-OH-DPAT-lnduced Hypothermia in Rats. Eur J Pharmacol 193: 165-172, 1991.
Murphy, R.M. and Zemlan, F.P.: Selective Serotonin1A/1B Agonists Differentially Affect Spinal Nociceptive Reflexes. Neuropharmacology 29: 463-468, 1990. Ormsbee, H.S. and Fondacaro, J.D.: Action of Serotonin on the Gastrointestinal Tract. Proc Soc Exp Biol 178: 333-338, 1985. Ozdemir, I.A. and Zimmermann, B.: Serotonin and Restraint-Induced Gastrointestinal Ulceration. Surg Forum 22: 319-320, 1971.
Pan, H. and Galligan, J.J.: 5-HT1A and 5-HT4 Receptors Mediate Inhibition and Facilitation of Fast Synaptic Transmission in Enteric Neurons. Am J Physiol 266: G230-G238, 1994. Pascaud, X., Roger, A., Genton, M. and Roze, C.: Further Support for the Central Origin of the Gastric Antisecretory Properties of Clonidine in Conscious Rats. Eur J Pharmacol 86: 247, 1982. Pettersson, G.: The Neural Control of the Serotonin Content in Mammalian Enterochromaffin Cells. Acta Physiol Scand Suppl 470: 1-30, 1979. Pettibone, D.J. and Pflueger, A.B.: Effects of Methiothepin and Lysergic Acid Diethylamide on Serotonin Release in Vitro and Serotonin Synthesis in Vivo: Possible Relation to Serotonin Autoreceptor Function. J Neurochem 43: 83-90, 1984. 191 Pique, J.M., Whittle, B.J.R. and Esplugues, J.V.: The Vasodilator Role of Endogenous Nitric Oxide in the Rat Gastric Microcirculation. Eur J Pharmacol 174: 293-296, 1989. Rack6, K., Schworer, H. and Kilbinger, H.: Adrenergic Modulation of the Release of 5-Hydroxytryptamine from the Vascularly Perfused Ileum of the Guinea Pig. Br J Pharmacol 95: 923-931, 1988. Racke, K. , Schworer, H. and Kilbinger, H.: Effects of the Benzodiazepine Receptor Agonist Midazolam and Antagonist Flumazenil on 5-Hydroxytryptamine Release from Guinea-Pig Intestine in Vitro. Indirect Support for a "Natural" Benzodiazepine-like Substance in the Intestine. Naunyn-Schmied Arch Pharmacol 341: 1-7, 1990. Racke, K., Schworer, H., Agoston, D.V. and Kilbinger, H.: Evidence that Neuronally Released Vasoactive Intestinal Polypeptide Inhibits the Release of Serotonin from Enterochromaffin Cells of the Guinea Pig Small Intestine. Acta Endocrin 124: 203-207, 1991. Racke, K. and Schworer, H.: Nicotinic and Muscarinic Modulation of 5-Hydroxytryptamine (5-HT) Release from Porcine and Canine Small Intestine. Clin Investig 70: 190-200, 1992. Racke, K. and Schworer, H.: Characterization of the Role of Calcium and Sodium Channels in the Stimulus Secretion Coupling of 5-Hydroxytryptamine Release from Porcine Enterochromaffin Cells. Naunyn-Schmied Arch Pharmacol 347: 1-8, 1993. Raybould, H.E., Roberts, M.E. and Dockray, G.J.: Reflex Decreases in Intragastric Pressure in Response to Cholecystokinin in Rats. Am J Physiol 253: G165-G170, 1987. Raybould, H.E., Holzer, P., Reddy, S.N., Yang, H. and Tach6, Y.: Capsaicin-Sensitive Vagal Afferents Contribute to Gastric Acid and Vascular Responses to Intracisternal TRH Analog. Peptides 11: 789-795, 1990. Rich-Denson, C. and Kimura, R.E.: Evidence in vivo that most of the Intraluminally Absorbed Glucose is Absorbed Intact into the Portal Vein and not Metabolized to Lactate. Biochem J 254: 931-934, 1988. Rioux, K.P. and Wallace, J.L.: Mast Cell Activation Augments Gastric Mucosal Injury through a Leukotriene-Dependent Mechanism. Am J Physiol 266: G863-G869, 1994. Rizzi, C.A., Coccini, T., Onori, L., Manzo, L. and Tonini, M.: Benzimidazolone Derivatives: A New Class of 5-Hydroxytryptamine4 Receptor Agonists with Prokinetic and 192 Acetylcholine Releasing Properties in the Guinea Pig Ileum. J Pharmacol Exp Ther 261: 412-419, 1992. Round, A. and Wallis, D.I.: The Depolarizing Action of 5-Hydroxytryptamine on Rabbit Vagal Afferent and Sympathetic Neurones in vitro and its Selective Blockade by ICS 205-930. Br J Pharmacol 88: 485-494, 1986. Round, A. and Wallis, D.I.: Further Studies on the Blockade of 5-HT Depolarizations of Rabbit Vagal Afferent and Sympathetic Ganglion Cells by MDL 72222 and Other Antagonists. Neuropharmacology 26: 39-48, 1987. Salim, A.S.: Gastric Mucosal Cytoprotection in the Rat by Naftidrofuryl Oxalate. J Pharm Pharmacol 42: 125-127, 1990. Sanger,' G.J.: Increased Gut Cholinergic Activity and Antagonism of 5-Hydroxytryptamine M-Receptors by BRL 24924 : Potential Clinical Importance of BRL 24924. Br J Pharmacol 91: 77-87, 1987. Sasagawa, T., Kobayashi, S. and Fujita, T.: Electron Microscope Studies on the Endocrine Cells of the Human Gut and Pancreas. In: Gastro-Entero-Pancreatic Endocrine System, ed. by T. Fujita, pp. 17-38, The Williams and Wilkins Company, Baltimore, 1974. Schepp, W., Schmidtler, J., Dehne, K., Schusdziarra, V. and Classen, M.: Pertussis Toxin-Sensitive and Pertussis Toxin-Insensitive Inhibition of Parietal Cell Response to GLP-1 and Histamine. Am J Physiol 262: G660-G668, 1992. Schiavone, A., Volonte, M. and Micheletti, R.: The Gastrointestinal Motor Effect of Benzamide Derivatives is Unrelated to 5-HT, Receptor Blockade. Eur J Pharmacol 187: 323-329, 1990. Schworer, H., Racke, K. and Kilbinger, H.: Spontaneous Release of the Endogenous 5-Hydroxytryptamine and 5-Hydroxyindoleacetic Acid from the Isolated Vascularly Perfused Ileum of the Guinea Pig. Neuroscience 21: 297-303, 1987. Siepler, J.K., Ahlman, B.H.J., Bhargava, H.N., Donahue, P.E. and Nyhus, L.M.: A Pharmacokinetic Analysis of the Vagal Release of the 5-Hydroxytryptamine in the Cat. J Neural Transm 47: 99-105, 1980. Siriwardena, A.K., Smith, E.H., Borum, E.H. and Kellum, J.M.: Identification of a 5-Hydroxytryptamine (5-HTz) Receptor on Guinea Pig Small Intestinal Crypt Cells. Am J Physiol 265: G339-G346, 1993. 193 Stapelfeldt, W., Schusdziarra, V., Weigert, N. Allescher, H.D. and Classen, M.: Effect of Atrial Peptide on Gastric Acid Secretion in Rats. Clin Physiol Biochem 6: 262, 1988. Stark, M.E. and Szurszewski, J.H.: Role of Nitric Oxide in Gastrointestinal and Hepatic Function and Disease. Gastroenterol 103: 1928-1949, 1992. Stephens, R.L., Garrick, T., Weiner, H. and Tach6, Y.: Serotonin Depletion Potentiates Gastric Secretory and Motor Responses to Vagal but not Peripheral Gastric Stimulants. J Pharmacol Exp Ther 251: 524-530, 1989. Stephens, R.L., Garrick, T., Weiner, H. and Tach£, Y.: Endogenous Serotonin Produces an Inhibitory Tone on Vagally Stimulated Gastric Function. Ann NY Acad Sci 597: 114-127, 1990. Stephens, R.L. and Tach6, Y.: Intracisternal Injection of a TRH Analogue Stimulates Gastric Luminal Serotonin Release in Rats. Am J Physiol 256: G377-G383, 1989. Stephens, R.L.: Disparate Effects of the Intracisternal Rx77368 and 0DT8-SS on Gastric Acid and Serotonin Release : Role of Adrenal Catecholamines. Regul Pept 1991.(In Press) Stephens, R.L.Jr. and Tach6, Y.: Intracisternal injection of a TRH analogue stimulates gastric luminal serotonin release in rats. Am J Physiol 19: G377-G383, 1989. Stoeckel, K., McNamara, P.J., McLean, A.J., duSouich, P., Lalka, D. and Gibaldi, M.: Nonlinear Pharmacokinetics of Indocyanine Green in the Rabbit and Rat. J Pharm Biopharm 8: 483-496, 1980. Tach£, Y., Stephens, R.L. and Ishikawa, T.: Central Nervous System Action of TRH to Influence Gastrointestinal Function and Ulceration. Ann NY Acad Sci 553: 269-285, 1989. Takeuchi, K., Ohtsuki, H. and Okabe, S.: Pathogenesis of Compound 48/80-Induced Gastric Lesions in the Rat. Dig Dis Sci 31: 392-400, 1986. Thompson, J.H.: Serotonin and the Alimentary Tract. Res Comm Chem Path Pharm 2: 687-781, 1971. Thompson, J.H.: Serotonin (5-Hydroxytryptamine) and the Alimentary System. In: Serotonin in Health and Disease, ed. by W.B. Essman, pp. 201-392, Spectrum Publications, Inc., New York, 1977. 194 Tobe, T., Izumikawa, F., Sano, M. and Tanaka, C.: Release Mechanisms of 5-HT in Mammalian Gastrointestinal Tract - Especially, Vagal Release of 5-HT. In: Endocrine Gut and Pancreas, ed. by T. Fujita, pp. 371-380, Elsevier Scientific Publishing Company, Amsterdam, 1976. Tsukamoto, Y., Nakazawa, S., Segawa, K., Goto, H., Fukui, A., Takano, K., Hase, S., Arisawa, T., Osada, T. and Kawabe, Y.: The Effect of a New Proton Pump Inhibitor on Serotonin-Induced Gastric Mucosal Lesions in Rats. Gastroenterol Jap 23: 501-505, 1988. Tsukamoto, Y., Goto, H., Hase, S., Arisawa, T., Ohara, A., Suzuki, T. and Hoshino, H.: Serotonin-Induced Decrease of Duodenal Mucosal Blood Flow plus Acid Load Produces Duodenal Mucosal Lesion in Rats. Digestion 50: 99-103, 1991. Turconi, M., Donetti, A., Schiavone, A., Sagrada, A., Montagna, E., Nicola, M., Cesana, R., Rizzi, C. and Micheletti, R.: Pharmacological Properties of a Novel Class of 5-HT, Receptor Antagonists. Eur J Pharmacol 203: 203-211, 1991. Twarog, B.M.: Serotonin: History of Discovery. Comp Biochem Physiol 91C: 21-24, 1988. Verbeuren, T.J.: Synthesis, Storage, Release, and Metabolism of 5-Hydroxytryptamine in Peripheral Tissues. In: The Peripheral Actions of 5-Hydroxytryptamine, ed. by J.R. Fozard, pp. 1-25, Oxford University Press, New York, 1989. Wade, P.R., Mawe, G.M., Branchek, T.A. and Gershon, M.D.: Use of Sterioisomers of Zacopride to Analyze Action of 5-Hydroxytryptamine on Enteric Neurons. Am J Physiol 260: G80-G90, 1991. Wang, H.-Y., Eberle-Wang, K., Simansky, K.J. and Friedman, E.: Serotonin-Induced Muscle Contraction in Rat Stomach Fundus is Mediated by a G-a-z-Like Guanine Nucleotide Binding Protein. J Pharmacol Exp Ther 267: 1002, 1993. Welch, A.S. and Welch, B.L.: Effect of Stress and para-chlorophenylalanine upon Brain Serotonin, 5-hydroxyindole Acetic Acid and Catecholamines in Grouped and Isolated Mice. Biochem Pharmacol 17: 699-708, 1968. Wilmer, A., Tack, J., Coremans, G., Janssens, J., Peeters, T. and Vantrappen, G.: 5-Hydroxytryptamine3 Receptors are Involved in the Initiation of Gastric Phase-3 Motor Activity in Humans. Gastroenterol 105: 773-780, 1993. 195 Wong, S.H., Cho, C.H. and Ogle, C.W.: The Role of Serotonin In Ethanol-Induced Gastric Glandular Damage in Rats. Digestion 45: 52-60, 1990. Wood, J.D.: Physiology of the Enteric Nervous System. In: Physiology of the Gastrointestinal Tract, ed. by L.R. Johnson, pp. 67-109, Raven Press, New York, 1987. Yoshida, N., Mizumoto, A., Iwanaga, Y. and Itoh, Z.: Effects of 5-Hydroxytryptamine, Receptor Antagonists on Gastrointestinal Motor Activity in Conscious Dogs. J Pharmacol Exp Ther 256: 272-278, 1991. Yoshioka,' M., Ikeda, T., Togashi, H., Saito, Y. and Saito, H.: Effect of 5-Hydroxytryptamine on Gastric Motility and Efferent Gastric Vagus Nerve Activity in Rats. Res Comm Chem Path Pharm 70: 3-10, 1990. Yoshiok^, M., Ikeda, T., Abe, M., Togashi, H., Minami, M. and Saito, H.: Pharmacological Characterization of 5-Hydroxytryptamine-Induced Excitation of Afferent Cervical Vagus Nerve in Anaesthetized Rats. Br J Pharmacol 106: 544-549, 1992. Zinner, M.J., Jaffe, B.M., DeMagistris, L., Dahlstrom, A. and Ahlman, H.: Effect of Cervical and Thoracic Vagal Stimulation on Luminal Serotonin Release and Regional Blood Flow in Cats. Gastroenterol 82: 1403-1408, 1982.