Innervation of the Gastrointestinal Tract (Autonomic Nervous System)

INNERVATION OF THE GASTROINTESTINAL TRACT

The Autonomic Nervous System A series of books discussing all aspects of the autonomic nervous system. Edited by Geoffrey Burnstock, Autonomic Neuroscience Institute, Royal Free Hospital School of Medicine, London, UK. Volume 1 Autonomic Neuroeffector Mechanisms edited by G. Burnstock and C.H.V. Hoyle

Volume 8 Nervous Control of Blood Vessels edited by T. Bennett and S.M. Gardiner

Volume 2 Development, Regeneration and Plasticity of the Autonomic Nervous System edited by I.A. Hendry and C.E. Hill

Volume 9 Nervous Control of the Heart edited by J.T. Shepherd and S.F. Vatner

Volume 3 Nervous Control of the Urogenital Systsem edited by C.A. Maggi

Volume 10 Autonomic – Endocrine Interactions edited by K. Unsicker

Volume 4 Comparative Physiology and Evolution of the Autonomic Nervous System edited by S. Nilsson and S. Holmgren

Volume 11 Central Nervous Control of Autonomic Function edited by D. Jordan

Volume 5 Disorders of the Autonomic Nervous System edited by D. Robertson and I. Biaggioni

Volume 12 Autonomic Innervation of the skin edited by J.L. Morris and I.L. Gibbins

Volume 6 Autonomic Ganglia edited by E.M. McLachlan

Volume 13 Nervous Control of the Eye edited by G. Burnstock and A.M. Sillito

Volume 7 Autonomic Control of the Respiratory System edited by P.J. Barnes

Volume 14 Innervation of the Gastrointestinal Tract edited by S. Brookes and M. Costa

This book is part of a series. The publisher will accept continuation orders which may be cancelled at any time and which provide for automatic billing and shipping of each title in the series upon publication. Please write for details.

INNERVATION OF THE GASTROINTESTINAL TRACT

Edited by

Simon Brookes and Marcello Costa

London and New York

First published 2002 by Taylor & Francis 11 New Fetter Lane, London EC4P 4EE Simultaneously published in the USA and Canada by Taylor & Francis Inc, 29 West 35th Street, New York, NY 10001 Taylor & Francis is an imprint of the Taylor & Francis Group

This edition published in the Taylor & Francis e-Library, 2003. © 2002 Taylor & Francis All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. Every effort has been made to ensure that the advice and information in this book is true and accurate at the time of going to press. However, neither the publisher nor the authors can accept any legal responsibility or liability for any errors or omissions that may be made. In the case of drug administration, any medical procedure or the use of technical equipment mentioned within this book, you are strongly advised to consult the manufacturer’s guidelines.

British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging in Publication Data A catalog record for this book has been requested

ISBN 0-203-21701-2 Master e-book ISBN

ISBN 0-203-27305-2 (Adobe eReader Format) ISBN 0–415–28377–9 (Print Edition)

Contents Preface to the Series — Historical and Conceptual Perspective of the Autonomic Nervous System Book Series

vi

Preface

xii

Contributors

xviii

1 Enteric Reflexes that Influence Motility J.C. Bornstein, J.B. Furness, W.A.A. Kunze and P.P. Bertrand 2 Motor Control of the Stomach David Grundy and Michael Schemann

1 57

3 Control of Gastric Functions by Extrinsic Sensory Neurons Peter Holzer

103

4 Neural Control of the Large Intestine Kalina Venkova, Beverley Greenwood-Van Meerveld and Jacob Krier

171

5 Neurons of the Gallbladder and Sphincter of Oddi Gary M. Mawe, Erin K. Talmage, Lee J. Jennings, Kirk Hillsley and Audra L. Kennedy

189

6 Pharmacology of the Enteric Nervous System Marcello Tonini, Fabrizio De Ponti, Gianmario Frigo and Francesca Crema

213

7 Neuroeffector Transmission in the Intestine Charles H.V. Hoyle, Pam Milner and Geoffrey Burnstock

295

8 Neural Control of Intestinal Vessels Neela Kotecha

341

9 Enteric Neuro-Immunophysiology Jackie D. Wood

363

10 Cellular Organisation of the Mammalian Enteric Nervous System Simon J.H. Brookes and Marcello Costa

393

11 Development of the Enteric Nervous System Michael D. Gershon

469

Index

527

v

Preface to the Series — Historical and Conceptual Perspective of the Autonomic Nervous System Book Series The pioneering studies of Gaskell (1886), Bayliss and Starling (1899), and Langley and Anderson (see Langley, 1921) formed the basis of the earlier and, to a large extent, current concepts of the structure and function of the autonomic nervous system; the major division of the autonomic nervous system into sympathetic, parasympathetic and enteric subdivisions still holds. The pharmacology of autonomic neuroeffector transmission was dominated by the brilliant studies of Elliott (1905), Loewi (1921), von Euler and Gaddum (1931), and Dale (1935), and for over 50 years the idea of antagonistic parasympathetic cholinergic and sympathetic adrenergic control of most organs in visceral and cardiovascular systems formed the working basis of all studies. However, major advances have been made since the early 1960s that make it necessary to revise our thinking about the mechanisms of autonomic transmission, and that have significant implications for our understanding of diseases involving the autonomic nervous system and their treatment. These advances include: (1) Recognition that the autonomic neuromuscular junction is not a “synapse” in the usual sense of the term where there is a fixed junction with both pre- and post-junctional specialization, but rather the transmitter is released from mobile varicosities in extensive terminal branching fibres at variable distances from effector cells or bundles of smooth muscle cells which are in electrical contact with each other and which have a diffuse distribution of receptors (see Hillarp, 1959; Burnstock, 1986a). (2) The discovery of non-adrenergic, non-cholinergic nerves and the later recognition of a multiplicity of neurotransmitter substances in autonomic nerves, including monoamines, purines, amino acids, a variety of different peptides and nitric oxide (Burnstock et al., 1964, 1986b, 1997; Rand, 1992; Milner and Burnstock, 1995; Lincoln et al., 1995; Zhang and Snyder, 1995; Burnstock and Milner, 1999). (3) The concept of neuromodulation, where locally released agents can alter neurotransmission either by prejunctional modulation of the amount of transmitter released or by postjunctional modulation of the time-course or intensity of action of the transmitter (Marrazzi, 1939; Brown and Gillespie, 1957; Vizi, 1979; Fuder and Muscholl, 1995; MacDermott et al., 1999). vi

PREFACE TO THE SERIES

G. Burnstock — Editor of The Autonomic Nervous System Book Series

vii

viii INNERVATION OF THE GASTROINTESTINAL TRACT

(4) The concept of cotransmission that proposes that most, if not all, nerves release more than one transmitter (Burnstock, 1976; Hökfelt, Fuxe and Pernow, 1986; Burnstock, 1990a; Burnstock and Ralevic, 1996) and the important follow-up of this concept, termed “chemical coding”, in which the combinations of neurotransmitters contained in individual neurones are established, and whose projections and central connections are identified (Furness and Costa, 1987). (5) Recognition of the importance of “sensory-motor” nerve regulation of activity in many organs, including gut, lungs, heart and ganglia, as well as in many blood vessels (Maggi, 1991; Burnstock, 1993), although the concept of antidromic impulses in sensory nerve collaterals forming part of “axon reflex” vasodilatation of skin vessels was described many years ago (Lewis, 1927). (6) Recognition that many intrinsic ganglia (e.g., those in the heart, airways and bladder) contain interactive circuits that are capable of sustaining and modulating sophisticated local activities (Saffrey et al., 1992; Ardell, 1994). Although the ability of the enteric nervous system to sustain local reflex activity independent of the central nervous system has been recognized for many years (Kosterlitz, 1968), it has been generally assumed that the intrinsic ganglia in peripheral organs consist of parasympathetic neurones that provided simple nicotinic relay stations. (7) The major subclasses of receptors to acetylcholine and noradrenaline have been recognized for many years (Dale, 1914; Ahlquist, 1948), but in recent years it has become evident that there is an astonishing variety of receptor subtypes for autonomic transmitters (see Pharmacol. Rev., 46, 1994). Their molecular properties and transduction mechanisms have been characterised (see IUPHAR Compendium of Receptor Characterisation and Classification 2000). These advances offer the possibility of more selective drug therapy. (8) Recognition of the plasticity of the autonomic nervous system, not only in the changes that occur during development and ageing, but also in the changes in expression of transmitter and receptors that occur in fully mature adults under the influence of hormones and growth factors following trauma and surgery, and in a variety of disease situations (Burnstock, 1990b; Saffrey and Burnstock, 1994; Milner et al., 1999). (9) Advances in the understanding of “vasomotor” centres in the central nervous system. For example, the traditional concept of control being exerted by discrete centres such as the vasomotor centre (Bayliss, 1923) has been supplanted by the belief that control involves the action of longitudinally arranged parallel pathways involving the forebrain, brain stem and spinal cord (Loewy and Spyer, 1990; Jänig and Häbler, 1995). In addition to these major new concepts concerning autonomic function, the discovery by Furchgott that substances released from endothelial cells play an important role in addition to autonomic nerves, in local control of blood flow, has made a significant impact on our analysis and understanding of cardiovascular function (Furchgott and Zawadski, 1980; Burnstock and Ralevic, 1994). The later identification of nitric oxide as the major endothelium-derived relaxing factor (Palmer et al., 1988; see Moncada et al., 1991) (confirming the independent suggestion by Ignarro and by Furchgott) and endothelin as an endothelium-derived constricting factor (Yanagisawa et al., 1988; see Rubanyi and Polokoff, 1994) have also had a major impact in this area.

PREFACE TO THE SERIES

ix

In broad terms, these new concepts shift the earlier emphasis on central control mechanisms towards greater consideration of the sophisticated local peripheral control mechanisms. Although these new concepts should have a profound influence on our considerations of the autonomic control of cardiovascular, urogenital, gastrointestinal and reproductive systems and other organs like the skin and eye in both normal and disease situations, few of the current textbooks take them into account. This is largely because revision of our understanding of all these different specialised areas in one volume by one author is a near impossibility. Thus, this Book Series of 14 volumes is designed to try to overcome this dilemma by dealing in depth with each major area in separate volumes and by calling upon the knowledge and expertise of leading figures in the field. Volume I, deals with the basic mechanisms of Autonomic Neuroeffector Mechanisms which sets the stage for later volumes devoted to autonomic nervous control of particular organ systems, including Heart, Blood Vessels, Respiratory System, Urogenital Organs, Gastrointestinal Tract, Eye Function, Autonomic Ganglia, Autonomic-Endocrine Interactions, Development, Regeneration and Plasticity and Comparative Physiology and Evolution of the Autonomic Nervous System. Abnormal as well as normal mechanisms will be covered to a variable extent in all these volumes depending on the topic and the particular wishes of the Volume Editor, but one volume edited by Robertson and Biaggioni, 1995, has been specifically devoted to Disorders of the Autonomic Nervous System (see also Mathias and Bannister, 1999). A general philosophy followed in the design of this book series has been to encourage individual expression by Volume Editors and Chapter Contributors in the presentation of the separate topics within the general framework of the series. This was demanded by the different ways that the various fields have developed historically and the differing styles of the individuals who have made the most impact in each area. Hopefully, this deliberate lack of uniformity will add to, rather than detract from, the appeal of these books. G. Burnstock Series Editor

REFERENCES Ahlquist, R.P. (1948). A study of the adrenotropic receptors. Am. J. Physiol., 153, 586–600. Ardell, J.L. (1994). Structure and function of mammalian intrinsic cardiac neurons. In Neurocardiology, edited by J.A. Armour and J.L. Ardell, pp. 95–114. Oxford: Oxford University Press. Bayliss, W.B. (1923). The Vasomotor System. Longman: London. Bayliss, W.M. and Starling, E.H. (1899). The movements and innervation of the small intestine. J. Physiol. (Lond.), 24, 99–143. Brown, G.L. and Gillespie, J.S. (1957). The output of sympathetic transmitter from the spleen of a cat. J. Physiol. (Lond.), 138, 81–102. Burnstock, G. (1976). Do some nerve cells release more than one transmitter? Neuroscience, 1, 239–248. Burnstock, G. (1986a). Autonomic neuromuscular junctions: Current developments and future directions. J. Anat., 146, 1–30. Burnstock, G. (1986b). The non-adrenergic non-cholinergic nervous system. Arch. Int. Pharmacodyn. Ther., 280(suppl.), 1–15. Burnstock, G. (1990a). Co-Transmission. The Fifth Heymans Lecture – Ghent, February 17, 1990. Arch. Int. Pharmacodyn. Ther., 304, 7–33.

x

INNERVATION OF THE GASTROINTESTINAL TRACT

Burnstock, G. (1990b). Changes in expression of autonomic nerves in aging and disease. J. Auton. Nerv. Syst., 30, 525–534. Burnstock, G. (1993). Introduction: Changing face of autonomic and sensory nerves in the circulation. In Vascular Innervation and Receptor Mechanisms: New Perspectives, edited by L. Edvinsson and R. Uddman, pp. 1–22. San Diego: Academic Press Inc. Burnstock, G. (1997). The past present and future of purine nucleotides as signalling molecules. Neuropharmacology, 36, 1127–1139. Burnstock, G., Campbell, G., Bennett, M. and Holman, M.E. (1964). Innervation of the guinea-pig taenia coli: Are there intrinsic inhibitory nerves which are distinct from sympathetic nerves? Int. J. Neuropharmacol., 3, 163–166. Burnstock, G. and Milner, P. (1999). Structural and Chemical Organisation of the autonomic nervous system with special reference to non-adrenergic, non-cholinergic transmission. In Autonomic Failure: A Textbook of Clinical Disorders of the Autonomic Nervous System. 4th edn, edited by C.J. Mathias and R. Bannister, Oxford: Oxford University Press, pp. 63–71. Burnstock, G. and Ralevic, V. (1994). New insights into the local regulation of blood flow by perivascular nerves and endothelium. Br. J. Plast. Surg., 47, 527–543. Burnstock, G. and Ralevic, V. (1996). Cotransmission. In The Pharmacology of Smooth Muscle, edited by C.J. Garland and J. Angus, Oxford: Oxford University Press. Dale, H. (1914). The action of certain esters and ethers of choline and their reaction to muscarine. J. Pharmacol. Exp. Ther., 6, 147–190. Dale, H. (1935). Pharmacology and nerve endings. Proc. Roy. Soc. Med., 28, 319–332. Elliott, T.R. (1905). The action of adrenalin. J. Physiol. (Lond.), 32, 401–467. Fuder, H. and Muscholl, E. (1995). Heteroceptor-mediated modulation of noradrenaline and acetylcholine release from peripheral nerves. Rev. Physiol. Biochem. Physiol., 126, 265–412. Furchgott, R.F. and Zawadski, J.V. (1980). The obligatory role of endothelial cells in the relaxation of arterial smooth muscle by acetylcholine. Nature, 288, 373–376. Furness, J.B. and Costa, M. (1987). The Enteric Nervous System. Edinburgh: Churchill Livingstone. Gaskell, W.H. (1886). On the structure, distribution and function of the nerves which innervate the visceral and vascular systems. J. Physiol. (Lond.), 7, 1–80. Hillarp, N.-Å. (1959). The construction and functional organisation of the autonomic innervation apparatus. Acta Physiol. Scand., 46 (suppl. 157), 1–38. Hökfelt, T., Fuxe, K. and Pernow, B. (Eds.) (1986). Coexistence of neuronal messengers: A new principle in chemical transmission. In Progress in Brain Research, Vol. 68, Amsterdam: Elsevier. IUPHAR Compendium of Receptor Characterisation and Classification 2000, IUPHAR Media Ltd (London), UK. Jänig, W. and Häbler, H.-J. (1995). Visceral-Autonomic Integration. In Visceral Pain, Progress in Pain Research and Management, edited by G.F. Gebhart, Vol. 5, pp. 311–348. Seattle: IASP Press. Kosterlitz, H.W. (1968). The alimentary canal. In Handbook of Physiology, edited by C.F. Code, Vol. IV, pp. 2147–2172. Washington, DC: American Physiological Society. Langley, J.N. (1921). The Autonomic Nervous System, part 1. Cambridge: W. Heffer. Lewis, T. (1927). The Blood Vessels of the Human Skin and Their Responses. Shaw & Sons: London. Lincoln, J., Hoyle, C.H.V. and Burnstock, G. (1995). Transmission: Nitric oxide. In The Autonomic Nervous System, Vol. 1 (reprinted): Autonomic Neuroeffector Mechanisms, edited by G. Burnstock and C.H.V. Hoyle, pp. 509–539. The Netherlands: Harwood Academic Publishers. Loewi, O. (1921). Über humorale Übertrangbarkeit der Herznervenwirkung. XI. Mitteilung. Pflügers Arch. Gesamte Physiol., 189, 239–242. Loewy, A.D. and Spyer, K.M. (1990). Central Regulations of Autonomic Functions. New York: Oxford University Press. MacDermott, A.B., Role, L.W. and Siegelbaum, S.A. (1999). Presynaptic iootropic receptors and the control of transmitter release. Ann. Rev. Neurosci., 22, 443–485. Maggi, C.A. (1991). The pharmacology of the efferent function on sensory nerves. J. Auton. Pharmacol., 11, 173–208. Marrazzi, A.S. (1939). Electrical studies on the pharmacology of autonomic synapses. II. The action of a sympathomimetic drug (epinephrine) on sympathetic ganglia. J. Pharmacol. Exp. Ther., 65, 395–404. Mathias, C.J. and Bannister, R. (eds.) (1999). Autonomic Failure. 4th edn, Oxford: Oxford University Press. Milner, P. and Burnstock, G. (1995). Neurotransmitters in the autonomic nervous system. In Handbook of Autonomic Nervous Dysfunction, edited by A.D. Korczyn, pp. 5–32. New York: Marcel Dekker. Milner, P., Lincoln, J. and Burnstock, G. (1999). The neurochemical organisation of the autonomic nervous system. In Handbook of Clinical Neurology Vol 74(30): The autonomic nervous system – Part 1 – Normal Functions, edited by O. Appezeller, Amsterdam: Elsevier Science, pp. 87–134.

PREFACE TO THE SERIES

xi

Moncada, S., Palmer, R.M.J. and Higgs, E.A. (1991). Nitric oxide: physiology, pathophysiology, and pharmacology. Pharmacol. Rev., 43, 109–142. Palmer, R.M.J., Rees, D.D., Ashton, D.S. and Moncada, S. (1988). Arginine is the physiological precursor for the formation of nitric oxide in endothelium-dependent relaxation. Biochem. Biophys. Res. Commun., 153, 1251–1256. Rand, M.J. (1992). Nitrergic transmission: nitric oxide as a mediator of non-adrenergic, non-cholinergic neuroeffector transmission. Clin. Exp. Pharmacol. Physiol., 19, 147–169. Rubanyi, G.M. and Polokoff, M.A. (1994). Endothelins: molecular biology, biochemistry, pharmacology, physiology, and pathophysiology. Pharmacol. Rev., 46, 328–415. Saffrey, M.J. and Burnstock, G. (1994). Growth factors and the development and plasticity of the enteric nervous system. J. Auton. Nerv. Syst., 49, 183–196. Saffrey, M.J., Hassall, C.J.S., Allen, T.G.J. and Burnstock, G. (1992). Ganglia within the gut, heart, urinary bladder and airways: studies in tissue culture. Int. Rev. Cytol., 136, 93–144. Vizi, E.S. (1979). Prejunctional modulation of neurochemical transmission. Prog. Neurobiol., 12, 181–290. von Euler, U.S. and Gaddum, J.H. (1931). An unidentified depressor substance in certain tissue extracts. J. Physiol., 72, 74–87. Yanagisawa, M., Kurihara, H., Kimura, S., Tomobe, Y., Kobayashi, M., Mitsui, Y., Yazaki, Y., Goto, K. and Masaki, T. (1988). A novel potent vasoconstrictor peptide produced by vascular endothelial cells. Nature, 332, 411–415. Zhang, J. and Snyder, S.H. (1995). Nitric oxide in the nervous system. Annu. Rev. Pharmacol. Toxicol., 35, 213–233.

Preface Eating is an extremely hazardous way to make a living, biologically speaking. The strategy of ingesting foreign organisms, breaking them down and reusing the spare parts is fraught with risks. For one thing, the organism concerned is likely to resist being broken down for scrap and may, in fact, have designs on the body parts of the diner. For another, the chemicals required to digest food are equally capable of breaking down the structure of the gut wall and the rest of the consumer’s body as well. It seems that minimising these risks, while meeting the imperative to absorb nutrients, has influenced many of the design principles of the gastrointestinal tract. This long tube is composed of a variety of tissue types and is the largest internal organ of the body. Its main function is to digest food and absorb the released nutrients. It is subdivided into functionally distinct regions, so that ingested material is processed serially in a way that serves to minimise the risks of self-digestion and attack by ingested pathogens. After rapid propulsion through the thorax via the oesophagus, food is subjected to mechanical and chemical breakdown in the harshly acidic environment of the stomach. This tends to reduce the viability of (inadvertently) swallowed micro-organisms. If toxic material is detected, propulsion can be reversed and the ingested material can be rapidly expelled by vomiting. During its sojourn in the stomach, sphincteric muscles at either end of the organ prevent spillage of the corrosive acidic content into the susceptible neighbouring regions of the duodenum and oesophagus. Gastric contents are then gradually aliquoted into the duodenum at a carefully controlled rate, adapted to the calorific and nutrient content of the meal. In the duodenum, the acidity of the contents is neutralised and they are mixed with digestive enzymes, bile and watery secretions, to allow the complex chemistry of digestion to get to work. The resulting chyme is mixed, moved to and fro and gradually propelled along the small intestine, at a rate adapted to the requirements for absorption of nutrients. A few hours after the meal, the small intestine is basically empty, the residual matter having been expelled into the colon. Here it may reside for several days, as water and ions are reabsorbed, playing a potentially vital role in the water balance of the organism. Eventually, the near-solid material is expelled from the body. In mechanical terms, this is a decidedly non-trivial event. The prolonged retention of contents in the colon makes it inevitable that large numbers of bacteria will be present, posing an enormous potential danger to the organism. If large numbers of micro-organisms colonise the small intestine, the barrier between the vasculature and the outside world is rapidly compromised and xii

PREFACE

xiii

massive septicaemia results. Parasites in food may survive the digestive process and then attack the wall of the gut. The consuming organism defends itself against these perils with specialised motor patterns, which are triggered to remove such noxious threats with great rapidity from the body. The multiplicity of functions required of the digestive tract makes it a fascinating subject for biological research by a range of disciplines, including immunology, microbiology, biochemistry, oncology etc. However it provides an especially appealing subject for neuroscientists. Neuronal circuitry is intimately involved in controlling many of the functions of the gut. It controls the activity of the smooth muscle in the gut wall, which mixes and propels the contents. Between meals, specialised patterns of motor activity periodically empty sloughed-off epithelial cells and accumulated mucous, thus preventing the build-up of a culture medium for the ever-threatening bacteria. Other highly propulsive motor patterns can be triggered by toxic chemicals, bacterial infection, parasitic infestation or identified allergens. While we consider the symptoms associated with these motor patterns (diarrhoea and pain) to be unpleasant, there is little doubt that these specialised mechanisms are of profound adaptive significance to our survival. The feature of the innervation of the gastrointestinal tract that is most appealing to neuroscientists is the presence of extensive networks of neural circuits embedded in its walls – the enteric nervous system. Containing approximately the same number of nerve cells as the spinal cord, it is connected with the central nervous system via visceral afferents and sympathetic and parasympathetic efferents of the autonomic nervous system. The presence of an entire, specialised nervous system within this peripheral organ system provides an opportunity for detailed study, without having to contend with the central nervous system. Since the studies of Bayliss and Starling (1899), Langley and Magnus (1905) and Trendelenburg (1917), it has been apparent that the enteric nervous system is capable of displaying a great deal of autonomy in controlling gastrointestinal function. Thus the enteric nervous system is studied by gastroenterologists seeking to understand how gut function is controlled, and also by neuroscientists endeavouring to understand how a relatively simple, autonomous mammalian neural circuit is organised. Obviously these two motivations are not mutually exclusive and many students of the enteric nervous system have foot in both camps. The complexity of the enteric nervous system began to be recognised with the discovery of the two major neural plexuses, the submucous plexus by Meissner (1857) and the myenteric plexus by Auerbach (1862). Langley (1921) placed the intrinsic enteric innervation in a class of its own after he recognised that many of the neural functions depend on the circuits entirely intrinsic to the digestive tract, rather than being solely determined by extrinsic neural inputs. Despite this early insight, it took many more years before the enteric nervous system was recognised by gastroenterologists as a relatively independent component of the autonomic nervous system. Combined studies by neuroanatomists, pharmacologists and physiologists eventually demonstrated that the study of the enteric nervous system was not only relevant to gastroenterology, but essential for its progress. During much of the twentieth century, the rate of progress was relatively slow in this field, which seemed to have fallen into a gap between neuroscience and gastroenterology. During this period, much emphasis was placed on the myogenic control of gastrointestinal motility, with little direct investigation of the roles of enteric neurons. It is only recently

xiv INNERVATION OF THE GASTROINTESTINAL TRACT

that the cellular basis of the spontaneous electrical activity of gut smooth muscle has begun to be understood in detail. It has been discovered that the mysterious cells, first described by Cajal as “interstitial cells” are involved both in generating myogenic activity and in mediating the input of enteric motor neurons to the smooth muscle (Sander, 1996; Huizinga et al., 1997). This breakthrough promises eventually to bridge the historical divide between the two schools of thought (myogenic versus neurogenic) about the control of gut motility. In this volume of the series on the autonomic nervous system, the innervation of the gut by the enteric nervous system, and its interface with the extrinsic innervation, is examined from a number of different perspectives. It will become apparent that all of these different aspects of nervous control can be related directly to how gut function is adapted to minimise risk, while maximising digestive efficiency. For example, the detailed analysis of enteric neural circuits controlling reflex motor activity of the intestine is elegantly summarised in the chapter by Joel Bornstein and colleagues. They provide an insightful synthesis of data gathered over many years by their own laboratory and by others around the world, as to how simple circuits may propel the gut contents, preventing excessive distension and perhaps contribute to the expulsion of pathogens. As described above, the stomach plays a very different role in the process of dealing with food, as it is not involved in digestion or absorption but is rather adapted for storage, mechanical breakdown and aliquoting of contents. Not surprisingly, the neuronal control of gastric motility has many different features from those of the small intestine. This is summarised in the extensive review of David Grundy and Michael Schemann who have contributed widely to the study of both the extrinsic and intrinsic innervation of the stomach. A particularly important role for extrinsic sensory nerves in protecting the stomach from a range of damaging agents has been identified in the last decade. Peter Holzer has been at the centre of this field and has written an authoritative review, summarising the role and mechanisms of action of these nerves in gastric protection. At the other end of the gastrointestinal tract, the colon also plays a significant role as an organ of storage, but it has to deal with very different material. In particular, much of the activity of the colon is modulated by extrinsic neuronal inputs from both the sympathetic and parasympathetic divisions of the autonomic nervous system. For many years the contributions of the extrinsic innervation have been rather over-simplified. This is corrected in the chapter of Kalina Venkova, Beverely Greenwood-Van Meerveld and the late Jack Krier, who carried out so many foundation studies on the extrinsic innervation of the distal gut. The last specialised region to receive attention in this volume is the gallbladder and sphincter of Oddi. Gary Mawe’s group have established themselves as leading investigators of this often overlooked accessory to the gut. Given the amount of trouble that it causes clinically, the biliary system and its motility are medically important subjects for study. The major differences between the organisation of the enteric nervous system of the gallbladder and sphincter of Oddi and that of the extensively studied small intestine are highlighted in this chapter. Since the pioneering study of Paton (1955), the small intestine, particularly of the guinea-pig, has been a favoured preparation for pharmacologists around the world. A vast literature has developed about the effects of drugs acting on a huge range of receptors found in the enteric nervous system and smooth muscle of this preparation. Marcello

PREFACE

xv

Tonini and his colleagues have summarised an enormous body of work to provide an accessible, yet comprehensive summary of this field. It should be pointed out that the future ability of gastroenterologists to provide treatment for disorders of the gut will rely on the development of new drugs. Understanding the pharmacology of the enteric nervous system is a crucial step in this undertaking. This analysis is greatly extended by the review of Charles Hoyle, Pam Milner and Geoffrey Burnstock on the nature of neuroeffector transmission in the gut. A wide-ranging summary of the transmitters, receptors and pharmacology of neuromuscular transmission is provided in their contribution. Frequently, enteric neurobiologists forget about the secretory and absorptive functions of the gastrointestinal tract, which are so central to its role in digestion. Charles Hoyle and colleagues also give us a wonderful account of the major pathways, largely arising from the submucous plexus, which control epithelial function. Of course, the secretory and absorptive capability of the epithelium is closely coupled with maintenance of an adequate blood supply, as are the requirements for mucosal protection described in Peter Holzer’s review. Neela Kotecha’s chapter summarises what is known about the enteric and extrinsic control of the gut vasculature, providing a valuable foundation for appreciating how different aspects of neuronal control are integrated. Lay people and funding agencies are often particularly interested in what scientists can tell them about the nature of medical disorders. This is as true for the innervation of the gut as it is for any other organ system. Many of the disorders of the gastrointestinal tract are related to the interactions between the immune system and the nerve cells and fibres in the gut wall. Jackie Wood and his many colleagues, over a long period, have been pioneers in this field. In his chapter, Professor Wood provides a compelling account of the central role of mast cells and mediators from leukocytes in modifying and sometimes determining the motor patterns and secretion of the small and large intestine during pathogen challenge. In the chapter by Simon Brookes and Marcello Costa, the cellular organisation of the enteric nervous system and some of the changes associated with gut disorders have been summarised. This raises the question of how the enteric nervous system develops to the normal adult form. This field has made enormous progress in the last decade and at the forefront of the advance has been the New York-based group of Michael Gershon. In a scholarly review, Professor Gershon describes some of the major influences on the colonisation and phenotypic development of enteric neurones. He discusses evidence for the involvement of a number of powerful growth factors and their receptors. In addition, he examines how defects in them may contribute to the aganglionosis of the distal colon, which is one of the most common developmental disorders of the enteric nervous system. The investigations summarised in this volume are beginning to reveal a large repertoire of neural mechanisms present in the digestive tract. It is not possible for a single book to encompass all knowledge about such a large and rapidly growing field. However, the chapters presented here give a good indication of the state of current knowledge. By reading between the lines, the astute reader will also be able to identify some of the current gaps in our knowledge. For example, it should not be surprising that there have been only few attempts to link cellular mechanisms to the functions of the entire organ, either in vitro or in vivo. In only a few cases have investigators been able to establish the circumstances under which a particular neural mechanism actually operates in normal living

xvi INNERVATION OF THE GASTROINTESTINAL TRACT

conditions. The gap between cellular physiology and organ physiology is still very wide. It should be clear from the reviews presented here that we now know a lot about what the neural system in the gut can do, but we know much less about how it actually operates under normal conditions. We know even less about how it changes, at the cellular level, to deal with different diets, with pathogens and in disease states. This represents a major challenge for the next generation of investigators. It can be expected that addressing these questions will be invaluable in the rational design of new therapies for disorders of the gastrointestinal tract. Simon Brookes and Marcello Costa

REFERENCES Auerbach, L. (1862). Über einen plexus myentericus, einen bisher unbekannten ganglionervösen Apparat im Darmkanal der Wirbelthiere. Breslau, Morgenstern, E. p. 13. Bayliss, W.M. and Starling, E.H. (1899). The movements and innervation of the small intestine. Journal of Physiology (London), 24, 99–143. Huizinga, J.D., Thuneberg, L., Vanderwinden, J.M. and Rumessen, J.J. (1997). Interstitial cells of Cajal as targets for pharmacological intervention in gastrointestinal motor disorders. Trends in Pharmacological Sciences, 18, 393–403. Langley, J.N. and Magnus, R. (1905). Some observations of the movements of the intestine before and after degenerative section of the mesenteric nerves. Journal of Physiology (London), 33, 34–51. Meissner, G. (1857). Über die Nerven der Darmwand. Zeitschrift für Ration Medezin, 8, 364–366. Paton, W.D.M. (1955). The response of the guinea-pig ileum to electrical stimulation by coaxial electrodes. Journal of Physiology (London), 127, 40P–41P. Sanders, K.M. (1996). A case for interstitial cells of Cajal as pacemakers and mediators of neurotransmission in the gastrointestinal tract. Gastroenterology, 111, 492–515. Trendelenburg, P. (1917). Physiologische und Pharmakologische Versuche über die Dunndarmperistaltik. Naunyn Schmiedebergs Archiv für Experimentelle Pathologie und Pharmakologie, 81, 55–129.

Marcello Costa and Simon Brookes — Editors of this volume

Contributors P.P. Bertrand Department of Physiology University of Melbourne Victoria 3010 Australia J.C. Bornstein Department of Physiology University of Melbourne Victoria 3010 Australia Simon J.H. Brookes Department of Human Physiology and Centre for Neuroscience Flinders University GPO Box 2100, Adelaide South Australia 5001 Geoffrey Burnstock Autonomic Neuroscience Institute Royal Free and University College Medical School Royal Free Campus Rowland Hill Street London NW3 2PF UK Marcello Costa Department of Human Physiology and Centre for Neuroscience

Flinders University GPO Box 2100, Adelaide South Australia 5001 Francesca Crema Department of Internal Medicine and Therapeutics Section of Clinical and Experimental Pharmacology University of Pavia Piazza Botta 10 I-27100 Pavia Italy Fabrizio De Ponti Department of Internal Medicine and Therapeutics Section of Clinical and Experimental Pharmacology University of Pavia Piazza Botta 10 I-27100 Pavia Italy Gianmario Frigo Department of Internal Medicine and Therapeutics Section of Clinical and Experimental Pharmacology University of Pavia Piazza Botta 10 I-27100 Pavia Italy xviii

CONTRIBUTORS

J.B. Furness Department of Anatomy and Cell Biology University of Melbourne Victoria 3010 Australia Michael D. Gershon Department of Anatomy and Cell Biology Columbia University College of Physicians and Surgeons 630 W 168th Street New York NY 10032 USA Beverley Greenwood-Van Meerveld Oklahoma Foundation for Digestive Research Basic Science Labs V.A. Medical Center Oklahoma City OK 73104 USA David Grundy Department of Biomedical Science The University of Sheffield The Alfred Denny Building Western Bank Sheffield S10 2TN UK Kirk Hillsley Department of Anatomy and Neurobiology C423 Health Science Complex The University of Vermont Burlington VT 05405 USA Peter Holzer Department of Experimental and Clinical Pharmacology University of Graz, Universitätsplatz 4 A-8010 Graz Austria

xix

Charles H.V. Hoyle Department of Anatomy and Developmental Biology University College London Gower Street London WC1E 6BT UK Lee J. Jennings Department of Anatomy and Neurobiology C423 Health Science Complex The University of Vermont Burlington VT 05405 USA Audra L. Kennedy Department of Anatomy and Neurobiology C423 Health Science Complex The University of Vermont Burlington VT 05405 USA Neela Kotecha Department of Physiology Monash University Clayton Victoria-3800 Australia Jacob Krier Oklahoma Foundation for Digestive Research Basic Science Labs V.A. Medical Center Oklahoma City OK 73104 USA Wolfgang A.A. Kunze Department of Anatomy and Cell Biology University of Melbourne Victoria 3010 Australia

xx

INNERVATION OF THE GASTROINTESTINAL TRACT

Gary M. Mawe Department of Anatomy and Neurobiology C423 Health Science Complex The University of Vermont Burlington VT 05405 USA Pam Milner Autonomic Neuroscience Institute Royal Free and University College Medical School Royal Free Campus Rowland Hill Street London NW3 2PF UK Michael Schemann Department of Physiology School of Veterinary Medicine Bischofsholer Damm 15/102 D-30173 Hannover Germany Erin K. Talmage Department of Anatomy and Neurobiology C423 Health Science Complex The University of Vermont Burlington VT 05405 USA

Marcello Tonini Department of Internal Medicine and Therapeutics Section of Clinical and Experimental Pharmacology University of Pavia Piazza Botta 10 I-27100 Pavia Italy Kalina Venkova Oklahoma Foundation for Digestive Research Basic Science Labs V.A. Medical Center Oklahoma City OK 73104 USA Jackie D. Wood Department of Physiology College of Medicine The Ohio State University 300 Hamilton Hall 1645 Neil Avenue Columbus Ohio 43210 USA

1 Enteric Reflexes that Influence Motility J.C. Bornstein1, J.B. Furness2, W.A.A. Kunze2 and P.P. Bertrand1 1

Department of Physiology and Department of Anatomy and Cell Biology, University of Melbourne, Victoria 3010, Australia

2

The gut exhibits a variety of movements that depend on the region studied and the timing and composition of the last meal. Primary control of these movements is exerted by the enteric nervous system. Except in the oesophagus and the sphincters, external inputs modulate enteric neural activity, rather than directly controlling muscle movements. In isolated intestine, localised mechanical and chemical stimuli excite enteric neurons to produce stereotypic reflexes in the smooth muscle. The reflex circuits form the basis of the circuitry responsible for more complex motor patterns evoked by more generalised stimuli. The enteric neural circuitry responsible for stereotyped reflexes in the guinea-pig ileum has been studied in detail and the neurons involved have been identified. These include at least three types of intrinsic primary afferent neurons (IPANs), a population of ascending interneurons, three populations of descending interneurons, excitatory and inhibitory motor neurons innervating the circular muscle and longitudinal muscle motor neurons. Synaptic transmission in the ascending excitatory reflex pathway is primarily via fast excitatory synaptic potentials (EPSPs) mediated by acetylcholine acting at nicotinic receptors; slow EPSPs mediated via NK3 tachykinin receptors may also be involved. Acetylcholine is less important in descending pathways. Transmission from interneurons to inhibitory motor neurons depends on ATP acting at P2X receptors and tachykinins play a significant, but not exclusive, role in transmission from IPANs to descending interneurons. Transmitters at other synapses in this pathway remain to be identified. These properties of simple reflexes will be important for understanding more complex behaviour. KEY WORDS: myenteric plexus; intestinal reflexes; synaptic transmission; intrinsic primary afferent neurons; enteric neural circuits; neurotransmitters.

INTRODUCTION The gastrointestinal (GI) tract breaks down food into absorbable nutrients by mechanisms that are almost entirely without conscious control. While chewing, swallowing and defaecation are consciously initiated; processes such as digestion, secretion and absorption, and the motility of the GI tract are essentially automatic. Both motility and secretion are under the direct control of the nervous system. This is manifested at several different levels. The enteric nervous system (ENS) is contained 1

2

INNERVATION OF THE GASTROINTESTINAL TRACT

entirely within the wall of the GI tract and can mediate many gastrointestinal functions without input from the central nervous system. Nevertheless, under normal circumstances several levels of external control can be identified. Emotions such as fear have well known gastrointestinal effects without the involvement of a visceral stimulus. In addition, many visceral stimuli alter gastrointestinal behaviour via pathways that include neurons of the central nervous system (e.g. vago-vagal reflexes). However, some stimuli applied to the gut modify gastrointestinal functions via neural pathways that bypass the central nervous system, but include the prevertebral sympathetic ganglia. Furthermore, a variety of hormonal mechanisms can also regulate intestinal function. For example, the interdigestive motor patterns of the stomach and both acid secretion and gastric emptying after a meal are modified by the chemical composition of the chyme in the small intestine (Lin et al., 1990a,b; Spencer et al., 1990; Orloff et al., 1992). The latter two effects are depressed when all neural pathways into the stomach are interrupted, but they can still be observed. This implies that both the extrinsic nerves and hormonal influences are involved in the control of gastric function by the contents of the small intestine. Behaviours evoked by stimulation of the GI tract, that are independent of the endocrine system, fall into the general class of entero-enteric reflexes. Many, notably the vago-vagal reflexes referred to above (for review see Grundy and Scratcherd, 1989), depend on the integrity of connections between the GI tract and the central nervous system. Others depend only on intact connections between the GI tract and the prevertebral sympathetic ganglia (for review see Szurszewski and King, 1989). A characteristic feature of the extrinsic nervous supply to the GI tract is that efferent nerves provide terminals largely to the intrinsic ganglionated plexuses rather than directly to the muscle, except in the sphincters and in the striated muscle of the oesophagus (Furness et al., 1999). Thus, the extrinsic reflex pathways coordinate the activities of different regions of the GI tract largely by modifying the activity of enteric neurons, rather than via direct effects upon the muscle, the enteric neurons representing a final common level of control of gastrointestinal function. This chapter focuses on the control of motility. In particular, we discuss recent data about the neural circuitry responsible for motility reflexes and the way this circuitry is organised to produce the motor programs responsible for complex patterns of movement. MOTILITY REFLEXES AND MOTOR PROGRAMS The neural circuits that control real motor behaviours anywhere in the body are complex and are generally studied by the use of reduced preparations in which the number of factors affecting a motor response is kept to a minimum. In the extreme case, such reduced preparations give rise to simple reflexes in which each element of the circuit can be readily identified. At the next level, the neural circuits that produce more complex behaviours can be considered to act as motor pattern generators, producing coordinated and often complex interactions between different muscles to give rise to stereotyped behaviour. To take an example from another system, in the simple tendon jerk reflex, primary afferent neurons from muscle spindles directly excite motor neurons. However, when the spinal cord is isolated from central input, appropriate sensory stimuli trigger the entire suite of coordinated movements required for normal walking. The difference between the simple reflexes of reduced preparations and the more complex, but still stereotyped, motor

ENTERIC REFLEXES THAT INFLUENCE MOTILITY

3

patterns is that the simple reflexes always have the same form when initiated by a specific stimulus, while a single organ can generate several distinct motor patterns. In systems where pattern generators have been studied, including the GI tract, the neurons in simpler reflex circuits are an integral part of the pattern generator circuits. GI tract reflexes vary in complexity as they do in all other parts of the nervous system. Even if those reflexes involving the central nervous system or the prevertebral ganglia are excluded, enteric reflexes range from local constrictions resulting from pinches applied to the serosal surface to coordinated waves of activity sweeping along the intestine. The former may be analogous to the spinal monosynaptic reflex, while the latter are due to activation of complex motor patterns controlled by circuitry analogous to the motor pattern generators of the central nervous system. This chapter explores the circuitry underlying simple reflexes and the reflex motor patterns seen in the gut, with an emphasis on the ileum of the guinea-pig, about which we have the most detailed information.

PHYSIOLOGICAL STUDIES OF ENTERIC REFLEXES The first studies of intestinal reflexes were performed towards the end of the nineteenth century when it was shown that various stimuli applied to the intestine trigger movement of intestinal content or secretion across the intestinal wall (Nothnagel, 1882; Mall, 1896; Bayliss and Starling, 1899, 1900a,b). By 1899, it was clear that a substantial part of the mechanisms that control propulsion of intestinal contents must be confined to the intestinal wall. Bayliss and Starling (1899) showed that a bolus inserted into a region of intestine triggered a complex motor pattern which caused the bolus to pass anally along the intestine even when all connections to the central nervous system had been severed. This study identified two reflexes which appear to underlie much of the normal propulsive behaviour of the intestine; the ascending excitatory reflex, in which the circular muscle contracts oral to the point of stimulation, and the descending inhibitory reflex, in which the circular muscle relaxes anal (or aboral) to it. Bayliss and Starling believed that any stimulus that evoked one of these reflexes would evoke the other as well and they termed this the “Law of the Intestine”. Other workers, including Langley and Magnus (1905), quickly confirmed that propulsion of material along the intestine did not depend on the central nervous system. However, the motility patterns of the intestine in the presence, and absence, of food in vivo are more complex than the propulsive response to an inserted stimulus in vitro. STUDIES OF MOTOR PATTERNS IN VIVO Using radiometric methods, Cannon identified several distinct types of propulsive behaviour in the presence of food in vivo (Cannon, 1902, 1906, 1912). These can be broadly described as segmentation, peristalsis and peristaltic rushes. Segmentation seems to be a mixing behaviour and involves rhythmic stationary contractions that repeatedly divide and redivide a mass of intestinal content. In contrast, peristalsis and peristaltic rushes each propel the intestinal content in an anal direction, with the difference between them being a matter of degree, duration and velocity. He also identified propulsion of content in an oral direction, antiperistalsis.

4

INNERVATION OF THE GASTROINTESTINAL TRACT

Radiometric methods require the presence of a material opaque to X-rays in the lumen of the GI tract and have been extended by the use of radioactive labels within the food so that its movement can be monitored with scintigraphy. Such studies, in combination with the multisite recording methods discussed below, have provided valuable information about the effects of varying the composition of a meal on the motility of the intestine. An important observation is that the rate of movement of a meal through the stomach and along the intestine is critically dependent on its composition. Chemically inert meals move along the intestine more rapidly than meals with significant nutritive content, with the relative proportions of fat, carbohydrate and protein being significant determining factors (Eeckhout et al., 1984; Schemann and Ehrlein, 1986c; Buhner et al., 1988; Buhner and Ehrlein, 1989; Siegle, Schmid and Ehrlein, 1990; Schmid and Ehrlein, 1993). Even the pattern of contractions seen within the intestine varies according to the chemical composition of the meal. This is illustrated by a study by Schemann and Ehrlein (1986c) who found that meals containing nutrients induced a larger proportion of stationary contractions, attributed to segmentation, than did non-nutritive meals. The rate of propagation of propulsive contractions (peristalsis) was unchanged, but the length over which they propagated was substantially reduced by increased nutrient density. Thus, segmentation predominates when nutrient is present in lumen, while peristalsis predominates when the chyme is nutrient free (Schemann and Ehrlein, 1982, 1986b,c; Siegle and Ehrlein, 1988; Buhner and Ehrlein, 1989; Schmid and Ehrlein, 1993; Huge, Weber and Ehrlein, 1995). In many species, different patterns of activity occur in fed and fasted states. Studies using a variety of widely spaced recording devices – strain gauges, extracellular electrodes, multichannel pressure sensing catheters – have shown that when an animal is fasted a complex wave of activity propagates repeatedly, but slowly, along the intestine from the stomach to the caecum. The behaviour is variously known as the migrating myoelectric complex (MMC), the migrating motor complex or the interdigestive cycle (for a recent review of this, and other motility patterns in vivo, see Hasler (1999)). The MMC essentially consists of an anally propagating wave of strong excitation of the intestinal muscle that alternates with a longer period of very low activity or possibly inhibition of the muscle (Figure 1.1). These two phases (Phase III and Phase I, respectively) are separated by periods of irregular muscular activity (Phases II and IV). Interestingly, while MMCs in the small intestine have never been convincingly recorded in isolated tissue, they are seen in extrinsically denervated GI tract in vivo. Thus, the neural circuit generating MMCs must be contained within the intestinal wall, i.e. part of the enteric nervous system (Sarna et al., 1981). In many animals, including humans, MMCs disappear immediately after the animal eats and the activity of the intestine changes to the fed, or post-prandial state described above. MMCs are widely identified with the gastric antrum, the duodenum, jejunum and ileum. However, migrating motor patterns, which have been termed MMCs, have also been observed in the colon, but not in the caecum (Sarna, Prasad and Lang, 1988). The colonic MMCs are independent of those in the small bowel. Another motor pattern seen predominantly in the fasted state is the giant migrating contraction (GMC): a large contraction that is much briefer than phase III of the MMC, propagates rapidly and continues for a substantial distance (Sarna, Prasad and Lang, 1988; Sarna, 1991a,b; Otterson and Sarna, 1994; Sethi and Sarna, 1995). GMCs are seen in the ileum, the caecum and the

ENTERIC REFLEXES THAT INFLUENCE MOTILITY

J1 J2 J3 J4 J5 J6 J7 J8 J9 J10 J11 J12

III

I

II

5

III

24 min

80 mmHg

Figure 1.1 Multichannel manometric recordings taken from 12 sites (J1–J12) within the jejunum of a fasting human. Each recording site was separated from its neighbours by 2 cm. The different phases of the migrating motor complex are indicated for the upper trace as Phase I (I), Phase II (II) and Phase III (III), in this set of recordings Phase IV was not readily discernable. Recordings provided by Prof. Sushil Sarna.

colon and are clearly distinct from MMCs, but are similar to the peristaltic rushes seen in the fed state. In the fed state, the distal ileum has a characteristic pattern of motility not seen elsewhere in the GI tract (Fich et al., 1990). When a meal arrives in the terminal ileum of the dog, a series of high frequency contractions of the circular muscle is initiated. These contractions occur more frequently than the slow waves that are usually thought to control the frequency of contractions within the intestinal muscle. They are seen only when the material entering the terminal ileum contains nutrient; i.e. they are not seen during an isotonic saline infusion. As yet their function is unknown. Thus, in vivo studies have demonstrated the presence of several distinct motor patterns whose presence and persistence depends on the region being studied, the content of that region and of other regions. The relationship between some of these behaviours and the intestinal microcircuitry mediating simple reflexes will be discussed in the last section of this chapter. REFLEXES AND MOTOR PATTERNS IN ISOLATED SYSTEMS The work of Mall (1896) and then Bayliss and Starling (1899; 1900a,b) may fairly be described as the first substantial studies of motor patterns in isolated segments of intestine and hence in reduced systems. The results were essentially similar to those seen over the next 2 decades by other workers in studies performed on segments of intestine in vitro (Magnus, 1904; Langley and Magnus, 1905; Trendelenburg, 1917). Stimulation of the intestinal segment by irritating the mucosa or by distension with either a solid or liquid bolus evoked a stereotyped propulsive movement of the intestine that tended to move the stimulus in an oral to anal direction. The movement involved a shortening of the longitudinal muscle, followed by a contraction of the circular muscle oral to the stimulus and,

6

INNERVATION OF THE GASTROINTESTINAL TRACT

in some cases, a relaxation of the circular muscle anal to the stimulus. This last component of the response was not seen by all subsequent workers (e.g. Bolzer, 1949b), perhaps because the smooth muscle must first generate some active tension for a relaxation to be detected. Bolzer (1949a,b) recognised that peristalsis evoked by stroking the mucosa or by distension was a more complex phenomenon than a simple reflex. He found that contractions evoked oral to a solid bolus propagate anally along the intestine at a rate faster than the movement of the bolus itself. The contraction passed over the bolus implying that movement of the stimulus is not the sole mechanism leading to anal propagation of the contractions in the circular muscle. He speculated that the consistency of the bolus may be an important determinant of the final pattern of the propulsive behaviour. The idea that differing stimuli evoke different patterns of contractile behaviour has been supported by studies of intestinal segments isolated from the central nervous system in vivo and in vitro by Hukuhara and colleagues in the late 1950s and throughout the 1960s (Hukuhara, Yamagami and Nakayama, 1958; Hukuhara and Miyake, 1959; Hukuhara, Nakayama and Sumi, 1959; Hukuhara et al., 1959; Hukuhara, Nakayama and Nanba, 1960a,b, 1961; Hukuhara, Sumi and Kotani, 1961; Hukuhara and Fukuda, 1965; Hukuhara, Neya and Tsuchiya, 1969). In these studies, stretching the intestinal wall inhibited ongoing contractions oral to, and anal to, the point of stimulation in the dog and the rabbit small intestine, but mechanical or acid stimulation of the mucosa evoked contraction on the oral side and inhibition on the anal side. The response to a specific stimulus also depended on the species tested, with guinea-pigs showing only oral excitation as a result of chemical or mechanical stimulation of the mucosa. In this species, stretch evoked contractions on both the oral and the anal side, unlike the responses seen in dogs and rabbits. Anal excitation evoked by stretch has been seen only intermittently in subsequent studies of guinea-pig intestine (see below). Hukuhara’s studies demonstrated several important points about the motor behaviour of the intestine. First, removal of the myenteric plexus interrupted transmission of reflex activity along the intestine, but disruption of the submucous plexus did not. Thus, the myenteric plexus probably contains the interneurons and the cell bodies of motor neurons mediating these reflexes. Second, reflex responses in the circular muscle to chemical and mechanical stimulation of the mucosa were essentially identical. Third, mucosal stimulation does not always evoke responses similar to those evoked by stretch. Thus, the nature of the motor program activated by a stimulus may depend on the overall state of the tissue at the time the stimulus is applied. Finally, they showed that chemical stimuli (acetylcholine, histamine, pilocarpine or BaCl2) applied to the serosa could evoke stereotyped motor responses in the intestine. The variety of stimuli employed by Hukuhara and his colleagues, raises the issue of the modes of stimulation appropriate for studies of motor reflexes. Recently, three distinct sensory modalities have been shown to excite stereotyped intestinal reflexes consisting of excitation oral and inhibition anal to the point of stimulation in isolated segments of small intestine or distal colon in vitro. Distension, mechanical stimulation of the mucosa and acid applied to the mucosa, all evoke such reflexes (Smith and Furness, 1988; Smith, Bornstein and Furness, 1990, 1991, 1992; Smith et al., 1992; Yuan et al., 1991; Grider, 1994; Grider and Jin, 1994; Foxx-Orenstein, Kuemmerle and Grider, 1996; Smith and McCarron, 1998).

ENTERIC REFLEXES THAT INFLUENCE MOTILITY

7

Despite the dependence of propulsion and local motor patterns on the chemical composition of the chyme (see above), mechanical stimuli are nearly always used in studies of enteric motility reflexes, perhaps because the onset and offset of the stimulus can be well controlled by the investigator. Distension, or a localised stretch applied to the intestinal wall, has been the stimulus of choice in the vast majority of studies, either in vivo or in vitro. This does not affect chemoreceptors and so distension is a relatively pure stimulus. However, exclusive use of any stimulus modality to analyse propulsive behaviours carries the inherent assumption that the neural circuitry activated by that stimulus includes all the elements that might be excited by other stimuli. Detailed electrophysiological analysis of the functional classes of neurons excited by different sensory modalities in the guinea-pig small intestine indicates that these different stimuli excite common populations of circular muscle motor neurons (Smith, Bornstein and Furness, 1992). However, as discussed below, many of the primary afferent neurons excited by the different sensory modalities that trigger intestinal reflexes are functionally distinct, indicating that this assumption does not hold. The other underlying assumption of exclusive use of distension to initiate reflexes is that the motor program excited by this stimulus provides the basic pattern underlying all motility in the GI tract. Although the basic reflexes excited by the three distinct sensory modalities appear similar, the results outlined earlier in this chapter indicate that chemical stimuli may initiate motor patterns distinctly different from those evoked by distension. Thus, as “physiological stimuli” used in most studies of enteric reflexes are constrained to a single sensory modality or a mix of mechanical stimuli (see below), they may excite only a proportion of the behaviours of which the stimulated region is capable. When an isolated segment of intestine is distended by a static increase in the pressure of isotonic saline contained within the lumen, it exhibits rhythmic contractions of the circular muscle (e.g. Trendelenburg, 1917; Tonini et al., 1981; Buchheit and Buhl, 1991). These contractions propagate anally, propelling the saline in the same direction and can empty the entire segment. Prior to each circular contraction, the longitudinal muscle contracts in what is known as the “preparatory phase” of peristalsis. In a variant of this method, a length of intestine is steadily distended with increasing amounts of physiological saline until a propulsive reflex is generated (Bülbring and Lin, 1958; Kottegoda, 1969; Waterman, Costa and Tonini, 1992, 1994a). The behaviour evoked consists of a ring of contraction of the circular muscle, which typically starts at the oral end of the intestinal segment and propagates anally (Hennig et al., 1999; SchulzeDelrieu, 1999). The contraction usually occludes the lumen and is preceded by a relaxation of the circular muscle, so that the saline within the segment is propelled anally and expelled from its anal end. This method has the advantage of allowing a threshold for the initiation of the behaviour to be determined along with both the pressure changes and the transport of liquid through the intestinal segment. Such studies have shown how complex the propulsion of material along the intestine can be. For example, intraluminal acid reduces the threshold pressure needed to trigger propulsion, but leads to a reduced flow at any given stimulus pressure once threshold is exceeded (Hukuhara, Nakayama and Sumi, 1959). Similarly, blockade of inhibitory motor neurons reduces the threshold pressure for initiation of the contraction, but can lead to a complete cessation of propulsion (Waterman and Costa, 1994, 1994b). Moreover, while these methods produce responses that are similar to the behaviours seen in vivo, there are several possible differences. For example,

8

INNERVATION OF THE GASTROINTESTINAL TRACT

the normal content of the intestine is not as chemically inert as physiological saline and does not steadily, or evenly, distend long segments of intestine. A simpler behaviour can also be evoked by gradual filling of a segment of the GI tract. This is reflex accommodation which involves a relaxation of the circular muscle at the site of distension when the pressure inside the stomach (Azpiroz and Malagelada, 1990; Hennig et al., 1997), the small intestine (Waterman, Costa and Tonini, 1994a) or the large intestine (Davison and Pearson, 1979; Ciccocioppo et al., 1994) exceeds a threshold value. This threshold is typically below that required to evoke a propulsive reflex (Waterman, Costa and Tonini, 1994a). Thus, reflex accommodation may be part of the preparatory phase of the propulsive behaviour evoked in the intestine by a fluid distension, as well as a separate behaviour in its own right. Saline distensions have also been used to study motor behaviours in the stomach (Hennig et al., 1997) and colon (Ciccocioppo et al., 1994; Smith and Robertson, 1998). In the latter, however, artificial faecal pellets, which represent a more physiologically realistic stimulus, are often used (Costa and Furness, 1976; Foxx-Orenstein and Grider, 1996; Kadowaki, Wade and Gershon, 1996; Foxx-Orenstein, Jin and Grider, 1998). The pellet is inserted into the oral end of an isolated segment of colon and the time taken to expel it from the anal end is measured. This time is consistent between trials and provides a reliable measure of the propulsive behaviour of the colon. The propulsive mechanism is very similar to that in the small intestine with pellets being propelled anally by waves of contraction within the circular muscle that are preceded by relaxation of the same muscle layer. An elegant refinement of this method has been the use of a mobile balloon attached to a device for measuring rate of transit as the artificial pellet (Crema, Frigo and Lecchini, 1970; Frigo and Lecchini, 1970; Tonini et al., 1989; Ciccocioppo et al., 1994; Onori et al., 2000). This allows the size of the balloon to be varied between trials and the pressure generated to be determined along with the speed of propulsion. These studies have shown that, in the rabbit colon, the dependence of propulsion on inhibitory neuromuscular transmission is greater for larger pellets. Another type of experiment uses a confined stimulus, which is held stationary, to evoke reflexes within the intestine. Responses are measured as changes in the contractility of the smooth muscle layers (Holzer, 1989; Tonini and Costa, 1990; Holzer, Schluet and Maggi, 1993; Maggi et al., 1994; Spencer, Walsh and Smith, 1999, 2000), as electrical responses in the muscle (Hirst, Holman and McKirdy, 1975; Smith and Furness, 1988; Smith, Bornstein and Furness 1990, 1991; Smith et al., 1992; Yuan et al., 1991) or as changes in the activity of enteric neurons measured either extracellularly or intracellularly (Hirst and McKirdy, 1974; Hirst, Holman and McKirdy, 1975; Bornstein et al., 1991a; Smith, Bornstein and Furness, 1992). Local distension with a balloon evokes an excitatory reflex oral to the point of stimulation (Holzer, 1989; Smith, Bornstein and Furness, 1990, 1992; Tonini and Costa, 1990; Holzer, Schluet and Maggi, 1993; Maggi et al., 1994), but this reflex typically propagates a few centimetres orally rather than anally (Smith, Bornstein and Furness, 1990, 1992). Such stimuli (and also mechanical and chemical stimuli applied to the mucosa) may excite local reflexes which do not propagate at all. However, recent studies by Spencer, Walsh and Smith (1999, 2000) found that both radial distension and mechanical stimulation of the mucosa evoked anally propagating contractions of both longitudinal and circular muscle in the guinea-pig ileum.

ENTERIC REFLEXES THAT INFLUENCE MOTILITY

9

In contrast, electrophysiological studies have rarely shown any evidence that the excitatory reflex contractions evoked by stationary distensions propagate anally in the small intestine. Similarly, stimulation of the mucosa either mechanically or with acid evokes an orally propagating excitatory reflex (Smith and Furness, 1988; Smith, Bornstein and Furness, 1991), which has yet to be shown to propagate anally. Thus, the anal propagation of contraction seen in the propulsion experiments cannot be explained by a simple combination of reflexes evoked by stationary stimuli applied to a confined region of intestine. This will be discussed in the last section of this chapter.

NEURAL CIRCUITS MEDIATING REFLEXES EVOKED BY STATIONARY STIMULI IN VITRO The basic circuit excited by a stationary stimulus is now clearly established for the small intestine of the guinea-pig (Figure 1.2). This is based on the responses of enteric neurons and the intestinal muscle and correlated electrophysiological, immunohistochemical and morphological studies of individual enteric neurons. In combination, these experiments

Figure 1.2 Diagram showing the neural circuitry underlying simple reflexes within the guinea-pig small intestine. This circuit, which contains seven types of neurons, has been inferred from many experimental sources. At each point along the intestine there is a network of IPANs with outputs to all other neuronal subtypes other than SOM descending interneurons. The interneurons form chains along the intestine and also make specific connections to motor neurons (ascending interneurons to excitatory motor neurons, descending interneurons to inhibitory motor neurons), which also receive input from local IPANs. The identities of specific neuronal subtypes are shown below the main circuit and are used in subsequent figures.

10

INNERVATION OF THE GASTROINTESTINAL TRACT

have allowed virtually all component neurons of the reflex pathways to be identified both structurally and physiologically. A stationary stimulus excites one or more populations of intrinsic primary afferent neurons (IPANs), which project circumferentially to excite local motor neurons and separate populations of orally and anally directed interneurons. Orally directed (ascending) interneurons activate both excitatory motor neurons and other ascending interneurons, while the anally directed (descending) interneurons excite other descending interneurons and inhibitory motor neurons. A significant subset of the IPANs project anally to contact inhibitory motor neurons (Johnson et al., 1996; Johnson, Bornstein and Boucher, 1998; Bian, Bertrand and Bornstein, 2000). These studies are discussed in detail in the remainder of this section. CLASSIFICATION OF ENTERIC NEURONS There have been four primary methods of classifying enteric neurons prior to identification of their functions: morphological, neurochemical, pharmacological and electrophysiological. Morphological studies date back to the work of Dogiel (1899) in the nineteenth century. This work provided the basic description of the major morphological types of cell bodies seen in the many subsequent studies (e.g. Stach, 1979, 1989; Scheuermann and Stach, 1983; Brehmer and Stach, 1998; Brehmer et al., 1998). In the small intestine of the guinea-pig, two broad classes of neurons can be identified using morphological techniques: large multipolar neurons are usually called Dogiel type II neurons, and several poulations of monoaxonal neurons (Figure 1.3) (Furness, Bornstein and Trussell, 1988).

Figure 1.3 Representative silhouettes of myenteric neurons from the guinea-pig duodenum that had been injected with neurobiotin during electrophysiological recordings (Clerc et al., 1998a). These shapes are typical of the shapes of myenteric neurons in the duodenum, ileum, proximal colon and distal colon.

ENTERIC REFLEXES THAT INFLUENCE MOTILITY

11

A subset of the monaxonal neurons has short lamellar dendrites; these are Dogiel type I neurons. Other subsets of the monoaxonal neurons include a population with long filamentous dendrites and a population of small neurons (Furness, Bornstein and Trussell, 1988). Small neurons with short filamentous dendrites have been identified in the guineapig duodenum (Clerc et al., 1998a,b) and in the proximal colon (Messenger, Bornstein and Furness, 1994) and distal colon of this species (Lomax et al., 1999). These small filamentous neurons have similar projection patterns to the small neurons of the ileum and may represent the same functional class. The immunohistochemical studies that have led to classification of enteric neurons according to their neurochemistry and projections are described in detail elsewhere in this volume (see Chapter 11 of this volume). Electrophysiological identification of enteric neurons began in the early 1970s with studies using extracellular recording methods (for review see Wood, 1975). These were rapidly overtaken by classification schemes based on intracellular recordings that began a little later. The first substantial intracellular study of enteric neurons published was that of Nishi and North (1973) who divided neurons into type 1 and type 2 on the basis of their firing properties, specifically whether they adapted rapidly or slowly to an imposed depolarization. In work published a few months later, Hirst, Holman and Spence (1974) divided myenteric neurons of the duodenum into two groups on the basis of their synaptic inputs and the after-potentials following their action potentials. S-neurons exhibited prominent fast excitatory synaptic potentials (EPSPs) in response to electrical stimulation of synaptic inputs and had only a brief (<100 ms) hyperpolarization following an action potential. The second population, AH-neurons, appeared to lack fast EPSPs and their action potentials were followed by prolonged after-hyperpolarizing potentials (AHPs). The AH/S classification scheme has proved highly successful in the small intestine of the guinea-pig; in part, because intracellular dye injection has shown that all AH-neurons have Dogiel type II morphology, while all S-neurons are monoaxonal (for review see Bornstein, Furness and Kunze, 1994). An important extension has been the recognition that all AH-neurons have inflections on the falling phases of their action potentials (Iyer et al., 1988; Schutte, Kroese and Akkermans, 1995) and that these persist even when the AHP is suppressed by synaptic input (Kunze et al., 1998). In the duodenum and distal colon, some neurons classified as Dogiel type II neurons by shape lack a prolonged AHP, but still have an inflection on the falling phase of the action potential (Clerc et al., 1998a; Lomax et al., 1999). Although intracellular injection of dye allows prediction of whether a neuron belongs to the AH- or S-category from its morphology, the experimental conditions are crucial for determining some electrophysiological properties of enteric neurons. For example, the characteristic AHP of an AH-neuron is not prominent in Dogiel type II neurons in stretched preparations when the muscle is free to contract (Kunze et al., 1998), but is seen in all such neurons when muscle contraction is blocked with L-type calcium channel blockers (Iyer et al., 1988). Similarly, three recent reports differ in their description of the firing of ascending interneurons during prolonged depolarizations and in whether they exhibit slow EPSPs. One found that such neurons are rapidly accommodating under all conditions tested and exhibited slow EPSPs when synaptic inputs were stimulated electrically (Kunze et al., 1997), while another (Brookes et al., 1997) reported that they are slowly accommodating and lacked slow EPSPs. These studies differ in that the former was

12

INNERVATION OF THE GASTROINTESTINAL TRACT

carried out in freshly isolated tissue, while the latter was performed with neurons that had been in organ culture for about 3 days. In the third study, ascending interneurons were found to be slowly accommodating, but also to have slow EPSP input (Smith, Burke and Shuttleworth, 1999). Where possible in the discussion below, neurons are categorized on the basis of both their electrophysiology and their morphology determined by intracellular dye injection from the recording electrode. REFLEXES EVOKED BY LOCAL DISTENSIONS In the small intestine (Costa et al., 1985; Grider, 1989; Holzer, 1989; Tonini and Costa, 1990; Holzera, Schluet and Maggi, 1993; Tonini et al., 1996), and in the colon (Costa and Furness, 1976; Grider and Makhlouf, 1992; Grider, 1994; Grider and Jin, 1994; FoxxOrenstein, Kuemmerle and Grider, 1996), distension of the intestinal wall with a stationary balloon or a mechanical stretching device evokes a contraction that extends several centimetres orally from the stimulus. In the colon, such stimuli also evoke relaxations on the anal side of the stimulus site (Costa and Furness, 1976; Grider and Makhlouf, 1992; Grider, 1994; Grider and Jin, 1994; Foxx-Orenstein, Kuemmerle and Grider, 1996; Spencer, McCarron and Smith, 1999). The first electrophysiological studies of reflexes evoked by balloon distension focussed on anally directed reflex pathways and showed that inflation of a balloon inserted into the lumen of a segment of intestine evoked inhibitory junction potentials (IJPs) in the circular muscle 2–3 cm anal to the balloon (Hirst and McKirdy, 1974; Hirst, Holman and McKirdy, 1975). Essentially identical results have been obtained in subsequent electrophysiological studies (Figure 1.4B) (Smith, Bornstein and Furness, 1990; Yuan et al., 1991; Yuan, Furness and Bornstein, 1992). A second finding of the early electrophysiological studies of distension evoked anally directed reflexes was that IJPs evoked by distension were often followed by excitatory junction potentials (EJPs), although the EJPs were much more labile (Hirst, Holman and McKirdy, 1975). This descending excitation may be the electrophysiological equivalent of the anally propagating contractions that pass over a stimulating bolus as described by Bolzer (1949b) and others. A similar descending excitation has also been observed in both the ileum (Spencer, Walsh and Smith, 1999, 2000) and the distal colon (Smith et al., 1992). However, electrophysiological studies of the guinea-pig ileum have failed to consistently observe a descending excitation. The reason(s) for this remain unclear. Electrophysiological analysis of orally directed reflex pathways in the small intestine excited by stationary distensions has confirmed that the ascending reflex contractions of the circular muscle are accompanied by EJPs in individual muscle cells (Figure 1.4A) (Smith, Bornstein and Furness, 1990; Yuan et al., 1991; Yuan, Furness and Bornstein, 1992). They have also shown that an ascending inhibitory pathway is activated by distension (Smith et al., 1990), although this has not been seen in contractility studies, possibly because of the lack of background tension discussed above. Both ascending excitatory and ascending inhibitory reflexes have been seen in the distal colon (Smith et al., 1992). Stationary balloon distension also evokes electrophysiologically detectable reflexes in various regions of the stomach (Yuan, Brookes and Costa, 1997). However, the predominant

ENTERIC REFLEXES THAT INFLUENCE MOTILITY

13

Figure 1.4 Intracellular recordings (upper traces each panel) of (A) an excitatory junction potential (EJP) evoked in a circular muscle cell by a distension with a balloon located 15 mm anal to the impaled muscle cell and (B) an inhibitory junction potential (IJP) evoked in a circular muscle cell by a distension applied 15 mm oral. The lower traces in each panel show when the distension was applied and removed; the small fluctuations in these traces reflect a time marker operating every 2 s. In both cases, contractions were blocked by addition of nicardipine to the bathing solution. No EJP was recorded on the anal side of the distension. Note, prominent slow waves were present in each preparation.

response appears to be inhibition of the circular muscle both oral and anal to the distending stimulus. REFLEXES EVOKED BY MECHANICAL OR CHEMICAL STIMULATION OF THE MUCOSA Electrophysiological and contractility studies in the guinea-pig small intestine (Smith and Furness, 1988; Yuan et al., 1991; Yuan, Furness and Bornstein, 1992; Spencer, Walsh and Smith, 1999, 2000), contractility studies in the guinea-pig colon (Spencer, McCarron and Smith, 1999) and contractility studies in the rat colon and human jejunum in vitro indicate that mechanical stimulation of the mucosa evokes reflexes similar to those evoked by distension of the same preparations (Grider, 1994; Grider and Jin, 1994; Foxx-Orenstein, Kuemmerle and Grider, 1996; Smith and McCarron, 1998). In each preparation, disturbing the mucosa either by brushing the mucosal surface or, in the small intestine, by compressing the villi evokes excitation oral (Figure 1.5A) and inhibition anal to the region stimulated (Figure 1.5B). The properties of these reflexes are discussed below. Studies of chemical stimuli applied to the mucosa in vitro are rare. Dilute acid applied to the mucosa of the distal ileum of the guinea-pig evoked ascending excitatory and descending inhibitory reflexes similar to those evoked by mechanical stimulation of the mucosa (Smith, Bornstein and Furness, 1991). The responses were transient and when stimuli were repeated within a refractory period of about 2 min, no further response was

14

INNERVATION OF THE GASTROINTESTINAL TRACT

Figure 1.5 Intracellular recordings of (A) an EJP evoked by mucosal compression applied 15 mm anally and (B) an IJP evoked by mucosal compression applied 15 mm orally (arrows show duration of the compression). Contractions were blocked in each preparation with nicardipine. No EJP was recorded on the anal side of the compression. Slow waves were recorded in (A), but not in (B).

evoked. As yet other potential chemical stimulants have not been shown to excite reflexes in vitro, although certain fatty acids can trigger peristalsis in vivo (Thomas and Baldwin, 1971; Baldwin and Thomas, 1975; Stewart and Bass, 1976a,b) and fatty acids and other nutrients alter intestinal transit of a meal (Schemann and Ehrlein, 1986a,c; Ehrlein et al., 1987; Richardson et al., 1991; Huge, Weber and Ehrlein, 1995). INTERACTIONS BETWEEN REFLEXES EVOKED BY DIFFERENT STIMULI Interactions between reflexes evoked by different stimuli have been studied using a preparation in which distension was applied from the serosal surface and mechanical stimulation was applied to the mucosa and responses were recorded electrophysiologically from the circular muscle. Both distension evoked responses and responses to mucosal stimuli are transient even when the stimulus is maintained (Smith and Furness, 1988; Smith, Bornstein and Furness, 1990; Yuan et al., 1991). This is modality specific (Smith, Bornstein and Furness, 1991; Yuan et al., 1991); when the membrane potential change in the muscle evoked by distension has completely disappeared, mucosal stimulation still evokes a reflex, and vice versa. Indeed, maintained mucosal stimulation markedly enhances responses to distension even when any response to the former has vanished (Smith, Bornstein and Furness, 1991). These results indicate that the two types of stimulus activate reflex pathways that are at least partially separate. The relationship between the reflex pathways excited by acid pH and those excited by distension is similar. Prior treatment with dilute acid enhances distension evoked reflexes

ENTERIC REFLEXES THAT INFLUENCE MOTILITY

15

when the response to acid has disappeared (Smith, Bornstein and Furness, 1991). However, as the final concentration of acid at the level of the mucosa is unknown, it is difficult to determine the physiological significance of this observation. There are also interactions between reflexes evoked by similar stimuli at different sites. For example, if the intestine is distended at two distinct sites, then the response to the more distant distension is depressed if applied within 2 min of the nearer distension (Yuan, Furness and Bornstein, 1992). The depression is seen in both ascending and descending pathways and may be due to interference with transmission within the pathways, rather than an interaction at the level of the smooth muscle. REFLEX RESPONSES RECORDED FROM NEURONS Many results of studies of reflexes evoked in the circular muscle by either distension or mechanical stimulation of the mucosa are paralleled by observations of the activity of myenteric neurons following similar stimuli. Hirst and coworkers (Hirst and McKirdy, 1974; Hirst, Holman and McKirdy, 1975) found that luminal distension evoked bursts of fast EPSPs in virtually all myenteric S-neurons lying 2–5 cm anal to the stimulating balloon. AH-neurons did not respond to this stimulus. Many S-neurons responded only transiently to distension; the latencies of responses in these neurons were relatively short. Other S-neurons exhibited prolonged responses with much longer latencies even when distension was brief. Both mucosal deformation and distension from the serosal surface evoked bursts of fast EPSPs in S-neurons lying either oral (Figure 1.6A) or anal to the region stimulated (Figure 1.6B) (Bornstein et al., 1991a; Smith, Bornstein and Furness, 1992). The responses to distension were transient with maintained distension (Figure 1.6) and the number of fast EPSPs evoked declined markedly when distensions were repeated within 2 min of each other. When the mucosa was deformed, some neurons on the anal side exhibited very prolonged responses, but most responses in either orally or anally located neurons were transient (Figure 1.6) and repeated stimuli led to marked declines in the number of fast EPSPs evoked. There has been no similar analysis of the responses to chemical stimulation of the mucosa, although chemical stimulation of the mucosa evokes both fast and slow EPSPs in S-neurons lying circumferential to the site of stimulation (Kunze, Bornstein and Furness, 1995; Bertrand et al., 1997, 1998). Smith, Bornstein and Furness (1992) also showed that the reflex pathways excited by distension and deformation of the mucosa converge onto the same S-neurons, which were identified as motor neurons by their projections (see below). This study confirmed that stimulation via one modality evokes a normal reflex even when the response to the other stimulus at the same site had completely disappeared (see above). These results imply that separate populations of IPANs mediate the two reflexes, but that the motor neurons are the final common pathway for all the reflex circuits. While convergence of the responses to the different sensory modalities was not unexpected, a second type of convergence was more surprising. A significant subset of the S-neurons received synaptic input from both orally and anally directed reflex pathways (Smith, Bornstein and Furness, 1992). Most of these neurons were identified morphologically as longitudinal muscle motor neurons, others may have been circular muscle motor neurons or even interneurons.

16

INNERVATION OF THE GASTROINTESTINAL TRACT

Figure 1.6 Responses in two myenteric neurons to distension and brushing the mucosa at the same site. The neuron illustrated in (A) lay oral to the stimuli and thus was in the ascending reflex pathway. The neuron illustrated in (B) lay anal to the stimuli and was in the descending reflex pathway. In each case, the responses consisted of a transient burst of fast EPSPs. Both neurons were identified from their projections as circular muscle motor neurons, see Smith, Bornstein and Furness (1992). In both (A) and (B), the onset of the mucosal stimulation is indicated by the arrow.

IDENTITIES OF NEURONS IN THE REFLEX PATHWAYS INTRINSIC PRIMARY AFFERENT NEURONS It has recently been proved that some intestinal reflexes are mediated by primary afferent neurons with cell bodies in the enteric nervous system. Studies in rat colon and guinea-pig small intestine have shown that, after degeneration of all extrinsic nerve terminals in the gut wall, mechanical stimulation of the mucosa evokes ascending and descending reflexes indistinguishable from those seen when the extrinsic nerve supply is intact (Grider and Jin, 1994; Furness et al., 1995). While the cell type that detects this mechanical stimulus has not been identified, these results indicate unequivocally that some primary afferent neurons have their cell bodies within the intestinal wall. That is, there must be IPANs that are either directly sensitive to mechanical distortion of the mucosa or sensitive to some mediator released by non-neuronal cells within the mucosa.

ENTERIC REFLEXES THAT INFLUENCE MOTILITY

17

The same studies were less consistent when it came to responses to distension. While extrinsic denervation did not affect responses to distension in the guinea-pig small intestine (Furness et al., 1995), it abolished such responses in the rat colon (Grider and Jin, 1994). Thus, it appears that IPANs mediate distension evoked reflexes in the guinea-pig small intestine, while similar reflexes in rat colon may be mediated by extrinsic primary afferent neurons whose terminals remain functional when acutely isolated from their cell bodies. In this case, the outputs of the distension sensitive extrinsic primary afferent neurons may be via axon collaterals or synapses of passage in the myenteric plexus. Speculation about the identity of the IPANs precedes proof of their existence and dates back at least to Dogiel (1899), who suggested that they correspond to his type II neurons. He based this on the projections of neurons with this morphology, as they send processes both to the mucosa and to other myenteric ganglia, the ideal projection pattern for an IPAN mediating reflexes evoked by mucosal stimuli. The basic observation that type II neurons in the myenteric plexus project to the mucosa and make synapses with other myenteric ganglion cells has been abundantly confirmed (Furness, Bornstein and Trussell, 1988; Pompolo and Furness, 1988, 1995; Pompolo et al., 1989; Furness et al., 1990; Song, Brookes and Costa, 1991, 1994; Bornstein et al., 1991b; Brookes et al., 1995). A second population of Dogiel type II neurons has been identified in the submucous plexus (Bornstein, Furness and Costa, 1989) and these too project to both the mucosa (Costa et al., 1981; Keast, Furness and Costa, 1984) and the myenteric plexus (Kirchgessner and Gershon, 1988; Kirchgessner, Tamir and Gershon, 1992). Initial electrophysiological studies suggested AH-neurons might be IPANs, because fast synaptic inputs were not observed and so they were unlikely to be anything other than the first neuron in a reflex pathway (Hirst, Holman and Spence, 1974). This suggestion converged with that of Dogiel when intracellular injection of neuronal markers showed that myenteric AH-neurons had the morphology of Dogiel type II cells in the guinea-pig small intestine (Bornstein et al., 1984a, 1991b; Katayama, Lees and Pearson, 1986; Iyer et al., 1988). A similar observation was made for submucous AH-neurons (Bornstein, Furness and Costa, 1989). However, while fast synaptic input to AH-neurons is rare, these neurons exhibit prominent and powerful slow EPSPs in response to electrical stimulation of enteric nerve trunks (Katayama and North, 1978; Wood and Mayer, 1978b, 1979; Grafe, Mayer and Wood, 1979, 1980; Katayama, North and Williams, 1979; Johnson, Katayama and North, 1980; Johnson et al., 1981). Thus, the projections of AH/Dogiel type II neurons are consistent with them being IPANs, but their electrophysiological properties indicate that they may also act as interneurons. This is not unique to the IPANs of the intestine as primary afferent neurons with cell bodies in the mesencephalic trigeminal nucleus also have prominent synaptic inputs (Copray et al., 1990; Pedroarena et al., 1999). Confirmation that myenteric AH-neurons are IPANs has come from recent electrophysiological studies that directly measured the responses of these neurons to chemical and mechanical stimulation of the mucosa (Kunze et al., 1995; Bertrand et al., 1997, 1998). These studies show that application of acid (pH 1–5), base (pH 9–11) or the volatile fatty acid, acetate (pH 7.0), to the mucosa evokes bursts of action potentials in myenteric AH-neurons lying within 2 mm circumferential to the site of application (Figure 1.7). This occurs even when synaptic transmission is blocked with a low Ca2+, high Mg2+ solution, indicating that the action potentials are due to a sensory rather than a synaptic process.

18

INNERVATION OF THE GASTROINTESTINAL TRACT

Figure 1.7 Responses of three different AH/Dogiel type II neurons to application of (A) acetate (buffered to pH 7.2), (B) acid saline (pH 3) and (C) 5-HT to a patch of intact mucosa lying 1–3 mm circumferential to the impaled neurons. Each stimulus evoked a burst of action potentials in the impaled neuron. Figure modified from Bertrand et al. (1997).

Most AH/Dogiel type II neurons that respond to chemical stimulation of the mucosa do not respond to mild mechanical stimulation of the same region, but some do (Bertrand et al., 1998). Thus, many myenteric AH/Dogiel type II neurons appear to be chemosensitive IPANs. IPANs that are muscle mechanoreceptors The fact that distension evokes reflexes indicates that many primary afferent neurons are sensitive to mechanical changes in the intestinal smooth muscle. In the guinea-pig ileum, these mechanosensitive primary afferent neurons are intrinsic to the enteric nervous system (Furness et al., 1995). Furthermore, both ascending and descending reflexes can be evoked by distension when the mucosa and submucosa have been removed (Hirst, Holman and McKirdy, 1975; Smith, Bornstein and Furness, 1990; Yuan, Bornstein and Furness, 1994), indicating that IPANs sensitive to mechanical changes in the muscle have cell bodies and processes in the myenteric plexus. A variety of studies indicate that many IPANs are sensitive to an increase in the circumference of the intestinal tube, i.e. to the length of the circular muscle. For example, placing a sleeve around a segment of intestine to prevent an increase in intraluminal pressure stretching the intestinal wall, also prevents the initiation of the normal peristaltic motor pattern (Kosterlitz and Robinson, 1959). Similarly, slow stretch applied to an entire segment of intestine first excites reflex accommodation which relaxes circular muscle (Waterman, Costa and Tonini, 1994a) and then, as the length of the muscle continues to

ENTERIC REFLEXES THAT INFLUENCE MOTILITY

19

increase, the peristaltic motor pattern (Waterman, Costa and Tonini, 1992, 1994a,b; Waterman and Costa, 1994; Brookes et al., 1999a). However, while it is clear that the increase in length excites IPANs, whether this is due to an increase in tension within the muscle or to the actual change in length is unknown. Recent electrophysiological studies indicate that some myenteric IPANs are sensitive to tension in the outer smooth muscle layers (Kunze et al., 1998). Indeed, these neurons, which fire spontaneously in stretched preparations when the muscle is free to contract, appear to respond to the active generation of tension in the muscle as their firing disappears when contractions of the muscle are blocked (Figure 1.8). However, the change in firing between low and high levels of stretch may be slow, as only 3 of 10 neurons whose firing was followed for one second after the application of a rapid stretch increased their firing during this period. The effect of stretching the muscle can be mimicked by increasing spontaneous contractions of the muscle with the Ca2+ ionophore Bay-K 8864 and prevented by treatment with dispase, which breaks the connective tissue cross-bridges between the muscle and the myenteric plexus (Kunze et al., 1999). This indicates that these IPANs respond to the tension within the preparation rather than its macroscopic length. The spontaneous firing arises in the processes of the tension sensitive IPANs, because not all action potentials fully invade the soma and hyperpolarization of the soma often does not alter the rate of firing in the processes. However, there are discrepancies between the properties of the tension-sensitive IPANs directly identified by Kunze et al. (1998) and those that can be inferred for the IPANs that initiate reflexes evoked by distension. The compound EJPs evoked in circular muscle and bursts of fast EPSPs evoked in myenteric neurons oral to a stimulus by distension are transient and decay to baseline within 2–3 s even when the distension is maintained for up to 30 s (Smith, Bornstein and Furness, 1990, 1991, 1992; Smith et al., 1992; Yuan, Furness and Bornstein, 1992). The distensions used to evoke reflexes typically reach their peak in 150–500 ms and the latencies of the responses evoked in either neurons or the circular muscle are of the same order. This implies that

Figure 1.8 Intracellular recordings from a Dogiel type II neuron taken when the preparation was stretched to (A) 120% of its resting circumference, (B) 140% of its resting circumference and (C) 140% of its resting circumference in the presence of isoprenaline to relax the muscle. By contrast to Figures 1.4–1.7, nicardipine was not present in this experiment and the preparation was free to contract in (A) and (B). Modified from Kunze et al. (1998).

20

INNERVATION OF THE GASTROINTESTINAL TRACT

rapid increases in tension or muscle length excite some IPANs with very short delays for sensory transduction. Furthermore, these rapid distensions evoke reflex responses in neurons and circular muscle when the muscle is prevented from contracting by L-type Ca2+ channel blockers. By contrast, the directly identified tension-sensitive IPANs fire throughout a maintained stretch and depend on muscle contraction to produce their full response during maintained stretch (Kunze et al., 1998). These differences suggest that mechanosensitive myenteric IPANs may have two different responses to changes in the state of the intestinal muscle. Most, perhaps all, respond to a maintained increase in the active tension within the muscle with ongoing firing of action potentials in their sensory processes. These mechanoreceptors depend on contraction of the muscle and cease firing when the muscle relaxes. Some neurons, which may be a subset of the tension receptors, respond to rapid stretch with a short latency increase in firing that does not depend on muscle contraction. This type of response appears to adapt rapidly, declining to the static type response when the distension is maintained or disappearing entirely. Macroscopically, rapidly adapting mechanoreceptive IPANs would respond to the dynamic component of a tension increase (or perhaps to a length increase), while slowly adapting IPANs would signal the static component of the tension. Whether the mechanoreceptive IPANs are also chemosensitive remains to be determined. However, the slowly adapting IPANs have been unequivocally identified as AH/Dogiel type II neurons (Kunze et al., 1998), and neurons of this morphological type all project to the mucosa (Song, Brookes and Costa, 1994), so the projections of these neurons are consistent with this possibility. Furthermore, 50–80% of AH/Dogiel type II neurons respond to stretch, while more than 30% of these neurons have been found to respond to chemical stimuli applied to the mucosa. Thus, some overlap might be expected. However, although reflexes evoked in the circular muscle by acid stimulation of the mucosa are transient and depressed by prior stimulation with acid, such stimuli enhance subsequent responses to distension for several minutes even in the absence of a direct response to the acid (Smith, Bornstein and Furness, 1991). This suggests that reflexes evoked by acid stimuli and by rapid distension may be mediated by separate IPANs. Mucosal mechanoreceptors There is strong evidence that the IPANs that respond to mechanical stimulation of the mucosa are distinctly different from those that respond to distension. Studies using various activity dependent markers for neurons (Kirchgessner, Tamir and Gershon, 1992; Kirchgessner, Liu and Gershon, 1996) indicate that the cell bodies of the former are located in the submucous plexus. When nicotinic synaptic transmission was blocked, the only neurons that responded to mechanical stimulation of the mucosa, either by expression of Fos protein or by taking up FM1–43, had cell bodies in the submucous plexus. These neurons were immunoreactive for substance P, which is a characteristic feature of the submucous AH/Dogiel type II neurons in the guinea-pig small intestine (Bornstein, Furness and Costa, 1989). Retrograde and anterograde tracing studies indicate that these neurons project to the myenteric plexus (Kirchgessner and Gershon, 1988; Kirchgessner, Tamir and Gershon, 1992), as predicted for neurons mediating the reflexes evoked by mechanical stimulation of the mucosa.

ENTERIC REFLEXES THAT INFLUENCE MOTILITY

21

The projections of the IPANs responding to different sensory modalities provide further evidence that they represent distinct populations. Injection of intracellular markers into myenteric AH/Dogiel type II neurons has shown that most, if not all, have predominantly circumferential projections (Bornstein et al., 1991b). Some project more than half way around the intestinal wall (Stebbing and Bornstein, 1996). However, retrograde tracing studies indicate that some of these neurons also have long anal projections (Brookes et al., 1995). Analysis of the reflex pathways excited by stationary stimuli also indicates that some IPANs have long anal projections and that these are preferentially sensitive to distension rather than mechanical deformation of the mucosa (Johnson et al., 1996; Johnson, Bornstein and Boucher, 1998). This is consistent with the observation that the mechanosensitive IPANs in the submucosa have predominantly circumferential projections within the myenteric plexus (Kirchgessner and Gershon, 1988; Kirchgessner, Tamir and Gershon, 1992). A further distinction is that transmission from the IPANs sensitive to mucosal deformation appears to include a component mediated by NK3 tachykinin receptors, while transmission from the distension-sensitive IPANs does not (Johnson, Bornstein and Boucher, 1998). Thus, at least three distinct classes of IPANs can be identified in the guinea-pig ileum (Figure 1.9). These are a population of mechanoreceptive Dogiel type II neurons in the myenteric plexus, a population of myenteric Dogiel type II neurons that are sensitive to the chemical environment of the mucosa and a population of submucous Dogiel type II neurons that respond to mucosal deformation. Reflexes evoked by rapid stretch indicate that some mechanoreceptive myenteric IPANs respond preferentially to either rapid increases in tension or changes in length of the intestinal wall. Whether these functionally distinguishable IPANs are completely separate populations of neurons or whether they overlap in the sensory modalities they prefer remains to be determined.

Figure 1.9 Diagram showing the three different classes of intrinsic primary afferent neurons (IPANs) that have been unequivocally identified in physiological experiments on the guinea-pig myenteric plexus, all are AH/Dogiel type II neurons. The interconnections of these neurons are also shown. Chemosensitive IPANs (cross hatch) have outputs to, and receive input from, tension sensitive IPANs (solid) in the myenteric plexus. Mechanosensitive IPANs (checks) in the submucous plexus respond to mechanical deformation of the mucosa and have outputs to tension sensitive myenteric IPANs, which may in turn have outputs to the submucous IPANs (? indicates that this is less certain). MP, myenteric plexus; SMP, submucosal plexus.

22

INNERVATION OF THE GASTROINTESTINAL TRACT

INTERNEURONS Interneurons play a major role in the transmission of reflex activity along the intestine. Direct evidence for this has been obtained in a number of structural and functional studies, most notably in the guinea-pig ileum. Interneurons have been identified immunohistochemically and via direct injection of intracellular markers during electrophysiological experiments (see below). That they have a significant role in reflexes is illustrated by the observation that mucosal deformation evokes EJPs in the circular muscle 35 mm oral to the point of stimulation (Smith and Furness, 1988), but IPANs project no more than 1 mm orally and the excitatory motor neurons project less than 10 mm in the same direction. There is only one population of myenteric neurons that could bridge this gap, ascending interneurons that have been identified as S-neurons, have orally projecting axons and are immunoreactive for calretinin (Brookes et al., 1997). The axons of two neurochemically distinct subsets of anally projecting S-neurons, one immunoreactive for somatostatin and the other immunoreactive for nitric oxide synthase (NOS), provide collateral branches and synaptic terminals to other myenteric ganglia (Pompolo and Furness, 1995; Portbury et al., 1995; Stebbing and Bornstein, 1996). Another population of neurons, those immunoreactive for 5-hydroxytryptamine (5-HT), have similar projections within the myenteric plexus (Furness and Costa, 1982), but have not been identified electrophysiologically, although they have shapes characteristic of S-neurons. Each electrophysiologically identified class of descending interneurons makes ultrastructurally identifiable synaptic connections with inhibitory motor neurons (Pompolo and Furness, 1995; Portbury et al., 1995; Mann et al., 1997). The descending interneurons immunoreactive for 5-HT do not make significant numbers of synaptic connections with inhibitory motor neurons (Young and Furness, 1995). Descending interneurons are immunoreactive for choline acetyltransferase (ChAT) (Steele, Brookes and Costa, 1991; Costa et al., 1996; Li and Furness, 1998). Thus, there are at least three classes of cholinergic descending interneurons and one class of cholinergic ascending interneurons.

MOTOR NEURONS Electrophysiological studies of the projections of the excitatory and inhibitory motor neurons supplying the circular muscle indicate that each functional population can be subdivided on the basis of the lengths of their axons (Bornstein et al., 1986; Smith et al., 1988). Similar conclusions have been drawn from retrograde tracing studies, which also identified the neurochemistries of the different populations (for review see Costa et al., 1996, and Chapter 10 of this volume). In the guinea-pig small intestine, the inhibitory motor neurons are all immunoreactive for both NOS and vasoactive intestinal peptide (VIP), and the excitatory motor neurons are immunoreactive for ChAT and tachykinins. The short inhibitory motor neurons can be distinguished because they are immunoreactive for neuropeptide Y (NPY) (Uemura, Pompolo and Furness, 1995) and γ-aminobutyric acid (GABA) (Williamson, Pompolo and Furness, 1996), while the long neurons are immunoreactive for gastrin releasing peptide (GRP) (Uemura, Pompolo and Furness, 1995). Correlated electrophysiological, morphological and immunohistochemical studies indicate that all the

ENTERIC REFLEXES THAT INFLUENCE MOTILITY

23

circular muscle motor neurons are monoaxonal S-neurons (Bornstein et al., 1984a, 1991a; Katayama, Lees and Pearson, 1986; Stebbing and Bornstein, 1996; Kunze et al., 1997). The motor neurons that innervate the longitudinal muscle are also monoaxonal S-neurons (Bornstein et al., 1991a; Smith, Bornstein and Furness, 1992). Most are cholinergic and excitatory (Brookes, Steele and Costa, 1991; Brookes et al., 1992; Steele, Brookes and Costa, 1991). However, a small proportion contains GABA, NOS and/or VIP and may be inhibitory (Williamson, Pompolo and Furness, 1996). This is consistent with the results of some studies that indicate the presence of an inhibitory innervation (Kosterlitz and Lydon, 1968; Osthaus and Galligan, 1992), although electrophysiological studies of the longitudinal muscle usually only record EJPs (Hirst, Holman and McKirdy, 1975; Bauer and Kuriyama, 1982; Bywater and Taylor, 1983; Neil, Bywater and Taylor, 1983; Bornstein et al., 1986; Niel et al., 1986). FUNCTIONAL POPULATIONS OF NEURONS IN OTHER PREPARATIONS As yet there has been no complete study of the reflex pathways in other tissues comparable to those of the reflex pathways in the guinea-pig ileum, however, some features are common to all species and regions. For example, the circular muscle receives both excitatory and inhibitory innervation in every preparation studied. Nevertheless, immunohistochemical and electrophysiological studies of neurons elsewhere in the guinea-pig GI tract or in other species indicate that, despite the substantial similarities, there are differences between preparations. For example, in the guinea-pig, NOS is not found in the terminals of interneurons within the myenteric plexus of the duodenum (Clerc et al., 1998b) and somatostatin-immunoreactive interneurons project orally, rather than anally, in the distal colon (Messenger and Furness, 1990). Similarly, neurons electrophysiologically identical to the IPANs of the guinea-pig ileum have been identified in the guinea-pig duodenum (Hirst, Holman and Spence, 1974; Clerc et al., 1998a), proximal colon (Messenger, Bornstein and Furness, 1994), distal colon (Wade and Wood, 1988; Lomax et al., 1999) and rectum (Tamura and Wood, 1989; Tamura, 1992) as well as the mouse distal colon (Furukawa, Taylor and Bywater, 1986), rat duodenum (Brookes, Ewart and Wingate, 1988) and human colon (Brookes, Ewart and Wingate, 1987). However, such neurons are absent from the corpus of the guinea-pig stomach (Schemann and Wood, 1989), are rare in the antrum of the guinea-pig (Tack et al., 1992; Tack and Wood, 1992) and possibly in the pig ileum (Cornelissen et al., 1996). Thus, there are substantial differences between functionally equivalent neurons in the stomach and the rest of the GI tract, with less significant differences between similar neurons in other intestinal regions.

SYNAPTIC INTERACTIONS BETWEEN FUNCTIONAL CLASSES OF ENTERIC NEURONS TYPES OF SYNAPTIC POTENTIALS Electrical stimulation of enteric nerve trunks has revealed two forms of excitatory transmission in the myenteric plexus: fast (Figure 1.6) and slow EPSPs (Figure 1.10).

24

INNERVATION OF THE GASTROINTESTINAL TRACT

Figure 1.10 Slow excitatory synaptic potentials (EPSPs) in (A) an AH/Dogiel type II neuron and (B) and (C) two different S-neurons. In (A) and (B) the upper traces show hyperpolarizing current pulses injected through the recording electrodes to monitor resistance changes during the slow EPSPs. In both cases, there was an increase in input resistance of the neuron during the early part of the slow EPSP, although the relative magnitude of this fall was much greater in the AH/Dogiel type II neuron than in the S-neuron. (C) shows the effect of activation of a slow EPSP on the firing induced in an S-neuron by three successive 500 ms depolarizing current pulses of equal magnitude (1, 2, 3). A train of 10 stimuli at 20 Hz was delivered to an internodal strand between depolarizations 1 and 2 and evoked fast EPSPs in the S-neuron. These were insufficient to evoke an action potential during depolarization 2. Depolarization 1 evoked only a single action potential. However, during the rising component of the slow EPSP, which began at the end of depolarization 2, depolarization 3 triggered a train of action potentials that lasted throughout the current injection. i.e. the neuronal firing pattern was converted from rapidly adapting to slowly adapting. (C) was modified from Kunze et al. (1996).

Inhibitory synaptic potentials have also been seen in myenteric neurons (Wood and Mayer, 1978a; Johnson, Katayama and North, 1980; Johnson et al., 1981). However, the stimulus regime needed to evoke such inhibitory synaptic potentials is identical to that which evokes slow EPSPs, so that the excitatory responses usually occlude the inhibitory responses unless the slow EPSPs are blocked in some way (Christofi and Wood, 1993b). As a result, the inhibitory synaptic potentials have yet to be characterized. Similarly, while various different presynaptic inhibitory mechanisms have been described or inferred to exist in pharmacological studies (e.g. Morita, North and Tokimasa, 1982b), these have also not been well studied and this chapter will focus on the excitatory events. Three pharmacologically distinct types of fast EPSPs have been identified in the myenteric plexus. All involve similar conductance increases to Na+ and K+ and reverse at about 0 mV (Zhou and Galligan, 1996, 1998, 1999). It has long been known that many fast EPSPs are blocked by nicotinic antagonists (Nishi and North, 1973; Hirst, Holman and Spence, 1974). However, recent studies have shown that, although nicotinic blockade almost abolishes EPSPs in about a third of S-neurons, it is less effective in the remainder. P2X-receptor antagonists, suramin and PPADS, substantially depress the hexamethonium-insensitive fast EPSPs in most of these S-neurons (Galligan and Bertrand, 1994; LePard, Messori and Galligan, 1997). In around 10% of S-neurons, hexamethonium-insensitive fast EPSPs are blocked by the 5-HT3 receptor antagonist ondansetron (Zhou and Galligan, 1999).

ENTERIC REFLEXES THAT INFLUENCE MOTILITY

25

The exclusively nicotinic fast EPSPs appear to be in neurons of the ascending reflex pathways, while the fast EPSPs mediated via P2X receptors are in neurons of the descending pathways and may be confined to inhibitory motor neurons (LePard and Galligan, 1999; Johnson et al., 1999; Bian, Bertrand and Bornstein, 2000). The identity of the S-neurons exhibiting 5-HT3 receptor mediated fast EPSPs is unknown. Blockade of such receptors depresses reflexes in the ascending excitatory pathway (Yuan, Bornstein and Furness, 1994), but the only 5-HT immunoreactive neurons are descending interneurons (Costa et al., 1996). It has been suggested that glutamate may also mediate fast EPSPs in myenteric neurons (Liu et al., 1997). However, detailed pharmacological analysis of the effects of glutamate on enteric neurons indicates that these neurons express type 1 metabotropic glutamate receptors and not ionotropic glutamate receptors (Ren et al., 2000). Slow EPSPs last from 10 s to several tens of minutes depending on whether they are evoked in S- or AH-neurons and on the stimulus regime used to excite them (Katayama and North, 1978; Wood and Mayer, 1978b; Grafe, Mayer and Wood, 1980; Johnson, Katayama and North, 1980; Bornstein et al., 1984b; Clerc et al., 1999). In some S-neurons, a single focal stimulus applied to a circumferentially directed internodal strand evokes a slow EPSP (Bornstein et al., 1984b); but in AH-neurons, it is more common to employ a train of stimuli lasting 1–3 s at frequencies of 10–20 Hz. Prolonged trains of stimuli, up to 10 min, at a relatively low frequency (1 Hz) also evoke large slow EPSPs in AH-neurons (Clerc et al., 1999). The slow EPSPs in S-neurons are markedly shorter and have shorter latencies than those in AH-neurons for any given stimulus regime (Bornstein et al., 1984b). All slow EPSPs appear to involve a reduction in the K+ conductance of the postsynaptic membrane; in AH-neurons, this is often associated with an increase in the Cl − conductance (Bertrand and Galligan, 1994); whether a similar increase occurs in S-neurons is unknown. In both AH- and S-neurons, slow EPSPs apparently involve closure of a resting, voltage independent K+ conductance (Galligan, North and Tokimasa, 1989). In AH-neurons, however, closure of a calcium-dependent K+ channel also makes a major contribution (Grafe, Mayer and Wood, 1980; Galligan, North and Tokimasa, 1989; Bertrand and Galligan, 1994, 1995). Although calcium-dependent K+ channels have been identified in S-neurons (Tokimasa, Cherubini and North, 1983), these do not appear to be open at rest so they are not involved in the slow EPSPs of these neurons. The calciumdependent K+ channel in AH-neurons is blocked by BK channel specific toxins, iberiotoxin and charybdotoxin, but differs from a conventional BK channel in that it is insensitive to tetraethylammonium (TEA) and is not rapidly activated by calcium ions entering during an action potential (Kunze et al., 1994). The calcium-dependent K+ conductance underlies the slow AHPs of AH-neurons, so that slow EPSPs in these neurons depress their AHPs and enhance their firing during imposed depolarizations. The AHPs in AH-neurons normally limit their firing to the start of a prolonged depolarization, but during slow EPSPs the neurons fire throughout (see for example, Johnson, Katayama and North, 1980; Grafe, Mayer and Wood, 1980; Tokimasa, Cherubini and North, 1983; Kunze et al., 1994). That is, the AH-neuron is converted from a rapidly accommodating state to a slowly accommodating state. Slow EPSPs also cause a change in firing state from rapidly accommodating to slowly accommodating in most S-neurons (Figure 1.10C) (Kunze et al., 1997). In many S-neurons,

26

INNERVATION OF THE GASTROINTESTINAL TRACT

the change in state is due to blockade of an early onset voltage activated outward current that is blocked by TEA, but not by 4-aminopyridine (Kunze et al., 1997). A similar change in firing state is also seen under conditions in which IPANs are spontaneously active (Kunze et al., 1997, 1998). The identity of the neurotransmitters that mediate the slow EPSPs has been debated since they were first identified. It is now clear that the slow EPSPs are produced by a combination of different transmitters, and that this combination varies between different functional classes of neurons. Acetylcholine (ACh) acting at muscarinic receptors, 5-HT and tachykinins may all play a role in mediating these events. Muscarinic antagonists block an early component of the slow EPSPs in 25% of S-neurons and 10% of AH-neurons; this appears to be due to an action at M1 muscarinic receptors (North and Tokimasa, 1982; Morita, North and Tokimasa, 1982a; North and Tokimasa, 1983; North, Slack and Surprenant, 1985; Galligan, North and Tokimasa, 1989). This muscarinic component has a physiological role because pharmacological studies of reflexes indicate that muscarinic receptors play a role in transmission from IPANs to interneurons (Tonini and Costa, 1990; Johnson et al., 1996). However, most of the slow EPSPs seen in enteric neurons are not cholinergic. There is a body of pharmacological evidence indicating that many of the slow EPSPs in AH-neurons are mediated by 5-HT acting at a receptor subtype that has been identified as 5-HT1P (Branchek, Mawe and Gershon, 1988). The most powerful evidence comes from blockade of electrically evoked slow EPSPs by various antagonists and by anti-idiotypic antibodies, each of which also blocks the slow depolarizing effects of 5-HT on AH-neurons (Takaki et al., 1985; Mawe et al., 1989; Wade et al., 1994). However, slow EPSPs can still be evoked in both AH- and S-neurons after lesions that cause degeneration of all 5-HT containing nerve terminals (Bornstein et al., 1984b). Indeed, such lesions do not alter the slow EPSPs in S-neurons and those in AH-neurons show only very subtle changes. Furthermore, it has been shown using electron-microscopic immunocytochemistry that 5-HT neurons rarely make synapses with AH/Dogiel type II neurons (Young and Furness, 1995). Thus, 5-HT is unlikely to mediate slow EPSPs in S-neurons and may play only a minor role in AH-neurons. Recent data derived from experiments using specific antagonists against NK1 and NK3 tachykinin receptors confirm the conclusions of many less direct studies that tachykinins mediate many of the slow EPSPs in myenteric neurons. In particular, slow EPSPs evoked in AH-neurons by stimulating circumferentially directed internodal strands are blocked by either the specific NK3 receptor antagonist SR 142801 (Figure 1.11) or the specific NK1 receptor antagonist SR 140333 (Figure 1.12), or both (Johnson and Bornstein, 2002). Other pharmacological data indicating a role for tachykinins includes the fact that all the major types of mammalian tachykinins depolarize these neurons via a mechanism that mimics the slow EPSPs (Katayama and North, 1978; Katayama, North and Williams, 1979; Morita, North and Katayama, 1980; Johnson et al., 1981; Galligan, Tokimasa and North, 1987; Morita and Katayama, 1992; Bertrand and Galligan, 1994, 1995). Furthermore, slow EPSPs can be depressed by peptidases and by desensitization with substance P (Morita, North and Katayama, 1980). Tachykinin mediated slow EPSPs have also been identified in gall bladder neurons by blockade of NK3 receptors (Mawe, 1995) and AH-neurons of the guinea-pig proximal colon (Neunlist, Dobreva and Schemann, 1999).

ENTERIC REFLEXES THAT INFLUENCE MOTILITY

27

Figure 1.11 Effect of the NK3 tachykinin receptor antagonist SR 142801 on a slow EPSP evoked in an AH/Dogiel type II neuron by ten stimuli delivered at 10 Hz to a circumferentially directed internodal strand (star). (A) shows the control slow EPSP, while (B) shows the effect of the same stimulus after the preparation had been exposed to SR 142801 (100 nM) for 45 min. The slow EPSP is abolished. Upper traces show the membrane potential, while the lower traces show hyperpolarizing current pulses injected via the recording electrode to monitor resistance changes associated with the slow EPSP.

NK3 receptors have also been implicated in slow EPSPs in myenteric neurons of the guinea-pig stomach (Schemann and Kayser, 1991). Pharmacological evidence that other neurotransmitters mediate slow EPSPs of AH-neurons has been put forward recently. For example, specific antagonists of the actions of cholecystokinin (CCK) depress slow EPSPs in some myenteric AH-neurons (Schutte et al., 1997), although the relationship of these slow EPSPs to those mediated by tachykinins is unknown. While tachykinins depolarize both AH- and S-neurons in the guinea-pig ileum, direct pharmacological evidence for a role for these peptides in mediating slow EPSPs in the latter is very sketchy. There is indirect evidence that NK3 receptors play a role in transmission from IPANs to interneurons in both the ascending and descending reflex pathways (Johnson et al., 1996; Johnson, Bornstein and Boucher, 1998). Furthermore, studies using NK1 receptor internalization as a surrogate measure of receptor activation indicate that many S-neurons respond to tachykinins released when the mucosa is stimulated mechanically (Southwell et al., 1998). Nevertheless, in a study of myenteric neurons in the guinea-pig ileum, the slow EPSPs in 6 S-neurons were not blocked by an NK3 receptor antagonist and an NK1 antagonist blocked these responses in only 1 of 3 S-neurons (Johnson unpublished). As these experiments were performed in the presence of a muscarinic antagonist and 5-HT is unlikely to mediate slow EPSPs in S-neurons (see above), this suggests

28

INNERVATION OF THE GASTROINTESTINAL TRACT

Figure 1.12 Effect of the NK1 tachykinin receptor antagonist SR 140333 on a slow EPSP evoked in an AH/ Dogiel type II neuron by 20 focal pulses delivered at 10 Hz to a circumferentially directed internodal strand (arrow). (A) shows the control slow EPSP, (B) shows the effect of the same stimulus after the preparation had been exposed to SR 140333 (100 nM) for 40 min. Note, the slow EPSP in this experiment was faster in time course than that in Figure 1.11.

that a neurotransmitter other than a tachykinin, ACh or 5-HT mediates slow EPSPs in many S-neurons. Several other candidates are present in myenteric nerve terminals, all of which depolarize myenteric neurons. These include VIP (Williams and North, 1979), somatostatin (Katayama and North, 1980) and pituitary adenyl cyclase activating peptide (Christofi and Wood, 1993a). However, no direct evidence is available about these or other possible candidates as yet. TRANSMISSION BETWEEN IPANs AND MOTOR NEURONS (Figure 1.13) There is abundant structural and electrophysiological evidence for monosynaptic transmission between IPANs and excitatory and inhibitory motor neurons supplying each muscle layer. Pompolo and Furness (1988) used the fact that 80–90% of all IPANs in the myenteric plexus of the ileum are immunoreactive for calbindin, but no other neurons are, to study the synaptic connections of IPANs at an ultrastructural level. They found that calbindin-immunoreactive terminals contacted all myenteric neurons and those neurons

ENTERIC REFLEXES THAT INFLUENCE MOTILITY

29

Figure 1.13 Diagram showing the connections from intrinsic primary afferent neurons (IPANs) to motor neurons and the transmitters that can be inferred to act at the specific synapses. Acetylcholine (ACh) is likely to play a major role in transmission to local motor neurons, but is unlikely to be involved in transmission from the long anally projecting axons of some myenteric IPANs. Whether tachykinins (TK) have a role is unclear. In this and later Figures “?” indicates that the transmitter or receptor subtype has not been established unequivocally.

that were not IPANs had large numbers of such synapses. More recently, similar methods provided direct evidence for synapses between the IPANs and longitudinal muscle motor neurons and a subset of circular muscle motor neurons (Pompolo and Furness, 1990, 1993). Simultaneous recordings made from IPANs and structurally identified motor neurons showed that myenteric IPANs can cause slow EPSPs in many motor neurons (Kunze et al., 1993). Electrical stimulation of long circumferentially directed pathways within the myenteric plexus evokes fast EPSPs in motor neurons (Stebbing and Bornstein, 1996). As the only neurons in the myenteric plexus with long circumferential projections appear to be IPANs, this suggests that these neurons transmit to motor neurons via fast, as well as slow, EPSPs. Furthermore, chemical stimulation of the mucosa evokes volleys of fast EPSPs in nearby S-neurons with latencies similar to those of the action potentials evoked in IPANs by the same stimuli (Kunze et al., 1995; Bertrand et al., 1997, 1998). This

30

INNERVATION OF THE GASTROINTESTINAL TRACT

indicates that the fast EPSPs are not secondary to a slow EPSP in an interneuron as the rise time of a slow EPSP would add several hundred milliseconds to the latency of any secondary response. Analysis of transmission at these synapses indicates that both tachykinins and ACh play a role (Figure 1.13), but also implies that at least one other transmitter contributes to this process. The mechanoreceptive mucosal IPANs contain both substance P and ChAT (Bornstein and Furness, 1988; Kirchgessner, Tamir and Gershon, 1992). Nicotinic blockade substantially reduces the number of enteric neurons that express Fos or take up activity dependent markers in response to mechanical stimulation of the mucosa indicating an involvement of ACh at nicotinic receptors somewhere in this pathway (Kirchgessner, Tamir and Gershon, 1992; Kirchgessner, Liu and Gershon, 1996). Similar results with distension suggest that ACh may also have a role in transmission from distension-sensitive IPANs (Ritter, Costa and Brookes, 1997). This is consistent with the observations that all myenteric IPANs are immunoreactive for ChAT (Li and Furness, 1998) and that hexamethonium blocks fast EPSPs evoked in S-neurons by electrical stimulation of long circumferential pathways (Stebbing, 1996). Studies of descending inhibitory reflexes strongly suggest involvement of a transmitter other than ACh, ATP or a tachykinin in transmission from IPANs to inhibitory motor neurons. Divided organ bath studies indicate that synaptic transmission from the long anally projecting axons of distension-sensitive IPANs to inhibitory motor neurons was unaffected by blockade of nicotinic and muscarinic ACh receptors, P2X ATP receptors or NK1 and NK3 tachykinin receptors in the recording chamber (Johnson et al., 1996; Johnson, Bornstein and Burcher, 1998; Bian, Bertrand and Bornstein, 2000). Despite the failure of specific tachykinin antagonists to modify reflex transmission from IPANs to inhibitory motor neurons, slow EPSPs clearly have a physiological role in transmission between these two classes of neurons. When IPANs are spontaneously active, many morphologically and immunohistochemically identified inhibitory motor neurons change their firing state from rapidly adapting to slowly adapting (Kunze et al., 1997, 1998). TRANSMISSION FROM IPANs TO INTERNEURONS (Figure 1.14) Electrophysiological analyses indicate that ascending interneurons receive fast EPSPs (Stebbing and Bornstein, 1996; Brookes et al., 1997) and slow EPSPs (Kunze et al., 1993) from myenteric IPANs. However, slow EPSPs were not recorded in ascending interneurons that had been in organ culture for 3 days (Brookes et al., 1997). A direct analysis of the synaptic inputs from submucous IPANs to the ascending interneurons has not yet been undertaken. Nevertheless, substantial conclusions can be drawn from studies using divided organ baths and similar preparations. Blockade of nicotinic receptors in the stimulation chamber depresses ascending reflexes evoked in the circular muscle in a more oral recording chamber by either distension or mucosal distortion by 40–60% (Tonini and Costa, 1990; Johnson et al., 1996). These responses are further depressed to less than 20% of the control by desensitization of NK3 tachykinin receptors (Johnson et al., 1996). The response to mucosal distortion, but not that to distension, is also depressed by an NK3 antagonist in the presence of hexamethonium (Johnson, Bornstein and Burcher, 1998).

ENTERIC REFLEXES THAT INFLUENCE MOTILITY

31

Figure 1.14 Sites of transmission from intrinsic primary afferent neurons (IPANs) to ascending and descending interneurons and the transmitters/receptors that can be inferred to act at those sites. Transmission to ascending interneurons is likely to be primarily due to acetylcholine (ACh), but there is evidence for a lesser role for tachykinins acting at NK3 receptors. Tachykinins acting at NK3 receptors are involved in transmission to descending interneurons, presumably the nitric oxide synthase (NOS) interneurons, but the major transmitter at this site is unknown (indicated by + ?). Note, no connection between IPANs and SOM-immunoreactive descending interneurons has been identified. See Figure 1.2 for identities of specific neuronal subtypes.

Hyoscine also depresses the non-nicotinic component of each reflex when applied to the stimulation chamber and virtually abolishes reflexes evoked by mucosal distortion when added together with hexamethonium and an NK3 antagonist (Tonini and Costa, 1990; Johnson et al., 1996; Johnson, Bornstein and Burcher, 1998). Thus, these pharmacological studies indicate that distension-sensitive IPANs transmit to ascending interneurons via ACh acting on nicotinic and muscarinic receptors, which would produce fast and slow EPSPs, respectively. Submucous mechanosensitive IPANs employ both of these receptors and a tachykinin acting via NK3 receptors, which also implies transmission via fast and slow EPSPs. Ascending interneurons make their first synapses 3–4 mm oral to their cell bodies (Brookes et al., 1997), so some synapses affected by drugs in the stimulation chamber would be between interneurons and these probably involve exclusively cholinergic fast EPSPs (see next section). Thus, the possibility that all synaptic transmission between IPANs and ascending interneurons is via slow EPSPs cannot be excluded. Computer simulation of the enteric circuitry mediating ascending reflexes indicates that the time courses of the electrophysiologically recorded responses are more compatible with transmission at these synapses being exclusively via slow EPSPs than via fast EPSPs (Bornstein et al., 1997, 1998; Thomas, Bertrand and Bornstein, 1999). The nature of transmission from IPANs to descending interneurons is more difficult to infer. Pharmacological studies of the descending pathways indicate that nicotinic transmission plays only a minor role in transmission from either the distension-sensitive IPANs or the mucosal mechanoreceptors. Hexamethonium has virtually no effect on descending inhibitory reflexes over distances of up to 2 cm from the stimulus site (Smith, Bornstein and Furness, 1990; Johnson et al., 1996; Smith and Robertson, 1998; Bian, Bertrand and Bornstein, 2000). Furthermore, although descending interneurons that contain NOS receive fast EPSP input from what are presumably myenteric IPANs, the descending

32

INNERVATION OF THE GASTROINTESTINAL TRACT

interneurons that contain somatostain (SOM) rarely do (Stebbing and Bornstein, 1996). The SOM-immunoreactive descending interneurons do not appear to exhibit prominent slow EPSPs (Song et al., 1997), so it is possible that these neurons do not receive significant synaptic input from local myenteric or submucous IPANs. Indeed, quantitative immunocytochemical analysis of the neurochemistry of the inputs to the SOM-immunoreactive descending interneurons indicates that they receive very few structurally identifiable inputs from IPANs (Mann et al., 1997) (Pompolo and Furness, 1998). This suggests that the SOM-immunoreactive interneurons are not involved in the initial phase of transmission of descending reflexes. By contrast, slow EPSPs are common in NOS-immunoreactive descending interneurons, so it is possible that these slower non-cholinergic synaptic potentials mediate transmission from IPANs into the descending reflex pathways. This is consistent with studies showing that interference with NK3 tachykinin receptors in the stimulus region depresses descending inhibitory reflexes evoked by either mucosal distortion or distension, with the former being much more susceptible (Johnson et al., 1996; Johnson, Bornstein and Burcher, 1998). However, there has been no direct demonstration that NK3 receptors mediate slow EPSPs in the NOS-immunoreactive descending interneurons and other explanations for the effects of NK3 antagonists are possible (see below). There is no evidence available about transmission from IPANs to the 5-HT-immunoreactive class of descending interneurons. The results of divided organ bath experiments indicate that a neurotransmitter or receptor subtype, other than those already discussed, must be involved in transmission from IPANs to descending interneurons. Even when nicotinic receptors, suramin-sensitive receptors, NK1 and NK3 receptors are blocked in the region of the sensory stimulus transmission to the interneurons during reflexes still occurs. The nature of this extra transmitter remains unknown, as does the type of synaptic potential it produces. TRANSMISSION BETWEEN ASCENDING INTERNEURONS (Figure 1.15A) Only a single population of ascending interneurons, immunoreactive for ChAT, substance P, enkephalin and calretinin (Brookes, Steele and Costa, 1991; Costa et al., 1992; Brookes et al., 1997) has been identified immunohistochemically. These interneurons form chains with each receiving synapses from more anally located interneurons (Pompolo and Furness, 1993; Brookes et al., 1997). Most project 4–8 mm orally, although much longer neurons have been identified in retrograde tracing studies (Brookes et al., 1997). Both electrophysiological and pharmacological evidence indicates that transmission between ascending interneurons is via nicotinic fast EPSPs (Tonini and Costa, 1990; Bornstein et al., 1991a; Johnson et al., 1996; Brookes et al., 1997). For example, hexamethonium completely abolishes transmission of ascending reflexes through the central chamber of a three-chambered organ bath (Johnson et al., 1996). As the only synapses likely to be active in this chamber are those between ascending interneurons, this is strong evidence that transmission at such synapses is via nicotinic receptors (Figure 1.15A). Furthermore, local stimulation of ascending interneurons excites more oral ascending interneurons via fast EPSPs (Brookes et al., 1997). Despite the presence of tachykinin-immunoreactivity in ascending interneurons, three-chambered organ bath experiments indicate that tachykinins have no role in transmission between ascending interneurons (Johnson et al., 1996).

ENTERIC REFLEXES THAT INFLUENCE MOTILITY

33

Figure 1.15 Transmitters employed at synapses between (A) ascending interneurons and other ascending interneurons (exclusively ACh) and ascending interneurons and excitatory circular muscle motor neurons (predominantly ACh, with a minor role for tachykinins acting at NK3 receptors) and (B) descending interneurons and other descending interneurons and descending interneurons and inhibitory circular muscle motor neurons. The cell symbols are as shown in Figure 1.2. In B, synapses between NOS-immunoreactive descending interneurons and SOM-immunoreactive descending interneurons are shown as having no known transmitter, while synapses from the former to inhibitory motor neurons appear to use ATP acting at P2X receptors to produce fast EPSPs. The role of ACh in transmission between descending interneurons is uncertain, although these neurons are immunoreactive for ChAT and the vesicular acetylcholine transporter.

TRANSMISSION FROM ASCENDING INTERNEURONS TO MOTOR NEURONS (Figure 1.15A) Intracellular recordings from morphologically identified circular muscle motor neurons during ascending excitatory reflexes evoked by distension or mucosal distortion indicate that ascending interneurons transmit to motor neurons via fast EPSPs (Figure 1.6A) (Bornstein et al., 1991a; Smith, Bornstein and Furness, 1992). The divided organ bath studies of such reflexes indicate that this transmission is virtually abolished by nicotinic antagonists (Tonini and Costa, 1990; Johnson et al., 1996). Thus, it is probable that the fast EPSPs are mediated by ACh. However, the effects of blockade of NK3 receptors in the recording chamber of a divided organ bath suggest that these receptors have a minor role in transmission from ascending interneurons to the circular muscle motor neurons (Johnson, Bornstein and Boucher, 1998). Interestingly, when transmission along the ascending reflex pathway is blocked by hexamethonium for long periods, the reflex response recovers from 10% to about 30% of its control value and this may be due to increased activity of the tachykinins (Barthó, Holzer and Lembeck, 1989; Holzer, 1989). Transmission to the longitudinal muscle motor neurons from ascending interneurons is via fast EPSPs. Recordings made from such neurons identified by intracellular injection of

34

INNERVATION OF THE GASTROINTESTINAL TRACT

biocytin indicate that, like circular muscle motor neurons, they respond to activation of ascending reflex pathways with a burst of fast EPSPs (Smith, Bornstein and Furness, 1992). These are likely to be cholinergic as the ascending interneurons are immunoreactive for ChAT (Brookes, Steele and Costa, 1991; Steele, Brookes and Costa, 1991; Brookes et al., 1997). As yet no evidence for a role for slow transmission or a role for tachykinins has been identified. TRANSMISSION BETWEEN DESCENDING INTERNEURONS (Figure 1.15B) Morphologically identified descending interneurons (2–25 mm anal to the stimulus) respond to activation of descending reflex pathways by mucosal distortion with a burst of fast EPSPs (Bornstein et al., 1991a). The mucosal mechanoreceptive IPANs do not appear to have prominent anal projections (Johnson et al., 1996; Johnson, Bornstein and Boucher, 1998), so these EPSPs are probably from descending interneurons. However, these fast EPSPs may not be predominantly mediated via nicotinic receptors, as hexamethonium has little effect on these reflexes or on distension evoked reflexes over these distances (Johnson et al., 1996). Divided organ bath studies indicate that antagonists acting at 5-HT3 or 5-HT4 receptors do not modify either type of descending reflex (Yuan, Bornstein and Furness, 1994). Similar studies indicate that blockade of NOS activity actually enhances transmission from IPANs and has no effect on transmission between interneurons (Yuan, Bornstein and Furness, 1995). Another possible candidate is a purine acting through P2X receptors, but suramin does not depress descending reflexes in divided organ bath experiments, when added to a chamber lying between the stimulus chamber and the recording chamber (Johnson et al., 1999). Many of the synapses active in this chamber would have been expected to be between interneurons. Thus, no clear candidate has emerged as the likely neurotransmitter mediating fast transmission between the descending interneurons involved in descending inhibitory reflexes. Whether there is a role for slow excitatory transmission between descending interneurons is an open question. In two studies of the effects of mucosal distortion and/or distension on neurons in the descending reflex pathways, only one convincing slow EPSP has been recorded and this was in a neuron that was not identified morphologically (Bornstein et al., 1991a; Smith, Bornstein and Furness, 1992). By contrast, desensitization of NK3 tachykinin receptors in the intermediate chamber of a divided organ bath markedly depressed descending inhibitory reflexes evoked either by distension or mucosal deformation (Johnson et al., 1996). However, the NK3 antagonist SR 142801 in the same chamber had no effect on the reflexes (Johnson, Bornstein and Burcher, 1998). Furthermore, tachykinins are not found in descending interneurons, while they may be in the long anal projections of some IPANs. Thus, if tachykinins were involved in transmission within the intermediate chamber, it may have been at synapses between IPANs and one or more populations of descending interneurons. Muscarinic receptors are unlikely to be involved in transmission between the descending interneurons, because hyoscine in the intermediate chamber had little effect on descending reflexes (Johnson et al., 1996). The neurochemistries of the descending interneurons provide some clues as to the transmitters involved at interneuron to interneuron synapses. Interneurons immunoreactive for SOM and interneurons immunoreactive for NOS are each likely to be involved in motility

ENTERIC REFLEXES THAT INFLUENCE MOTILITY

35

reflexes because they contact inhibitory motor neurons (Mann et al., 1997) and each contains substances that excite myenteric neurons. Both classes of interneurons contain ACh, together with either SOM or VIP, pituitary adenylate cyclase-activating peptide and GRP (Costa et al., 1996), all of which excite myenteric neurons (Williams and North, 1979; Katayama and North, 1980; Zafirov et al., 1985; Christofi and Wood, 1993a). Furthermore, each makes chains of neurochemically identical interneurons (Portbury et al., 1995; Young, Furness and Povey, 1995; Mann et al., 1997) and has cross-connections to neurons of the other neurochemical class (Mann et al., 1997). Thus, it is unlikely that a single substance mediates transmission at all synapses between descending interneurons. TRANSMISSION BETWEEN DESCENDING INTERNEURONS AND MOTOR NEURONS (Figure 1.15B) Activation of descending reflex pathways by either distension or distortion of the mucosa evokes bursts of fast EPSPs in many morphologically identified inhibitory motor neurons supplying the circular muscle (Bornstein et al., 1991a; Smith, Bornstein and Furness 1992). The resistance of descending inhibition to hexamethonium discussed above indicates that these fast EPSPs are unlikely to be solely mediated by ACh acting at nicotinic receptors. They are also unlikely to be due to 5-HT acting at 5-HT3 receptors as granisetron (a 5-HT3 receptor antagonist) has no effect on descending reflexes when applied in the recording chamber of a divided organ bath (Johnson et al. unpublished observation). The transmitter mediating the major portion of these fast EPSPs is probably a purine acting at P2X receptors. Such receptors are prominent on inhibitory motor neurons and rare on excitatory motor neurons (Johnson et al., 1999) and purinergic fast EPSPs are preferentially evoked by electrical stimulation of descending pathways (LePard and Galligan, 1999). Furthermore, P2X antagonists in the recording chamber depress descending inhibitory reflexes recorded in the circular muscle to a substantially greater extent than they depress IJPs evoked by local electrical stimulation (Bian, Bertrand and Bornstein, 2000). There is no direct evidence for slow excitatory transmission between descending interneurons and inhibitory motor neurons. Blockade of tachykinin NK1 receptors in the recording chamber of a divided organ bath depresses descending reflexes and the effect appears to be selective for transmission from descending interneurons (Johnson, Bornstein and Boucher, 1998). However, these neurons do not contain tachykinins (Costa et al., 1996), so this effect is unlikely to be simply due to blockade of transmission between interneurons and the motor neurons. Rather it may be due to blockade of tachykinins released from spontaneously active IPANs in the vicinity of the inhibitory motor neurons thereby depressing the excitability of the latter (see below for a more complete discussion). Descending reflex input to longitudinal muscle motor neurons is also via fast EPSPs (Bornstein et al., 1991a; Smith, Bornstein and Furness, 1992). Whether these inputs arise from one or more populations of descending interneurons or from the anally directed axons of IPANs is unknown. There is structural evidence that the NOS/VIP/GRP interneurons make synaptic connections with longitudinal muscle motor neurons and with ascending interneurons, although the significance of the latter is unknown (Pompolo and Furness, 1995). The longitudinal muscle motor neurons also receive ultrastructurally defined inputs

36

INNERVATION OF THE GASTROINTESTINAL TRACT

from myenteric IPANs (Pompolo and Furness, 1995), although these may come from circumferentially directed pathways rather than descending pathways. While there is no definitive evidence as to the identity of the transmitter mediating the fast EPSPs evoked in these neurons by descending reflexes, there is no shortage of possible candidates. ACh, 5-HT, ATP and glutamate could all be involved. RETROGRADE TRANSMISSION FROM INTERNEURONS TO IPANs While nitric oxide (NO) has no role in transmission between descending interneurons, it does modulate the excitability of the descending reflex pathways. When NO synthesis is blocked in the stimulation chamber of a divided organ bath, the descending inhibitory response evoked either by distension or mucosal deformation is significantly enhanced (Yuan, Bornstein and Furness, 1995). This effect is not seen with ascending reflexes. This suggests that NO released from the cell bodies of descending interneurons depresses synaptic transmission between IPANs and descending interneurons (presumably those containing NOS). This is consistent with the suggestion that NOS-immunoreactive descending interneurons carry the excitatory signal from the IPANs into the descending reflex pathway. The effect of inhibition of NO synthesis on transmission between enteric neurons has not been directly studied. However, NO donors depress slow EPSPs evoked by electrical stimulation in IPANs (Tamura et al., 1993). Such slow EPSPs may result from activation of other IPANs (see next section). By contrast, NO donors did not modify fast EPSPs in the small number of neurons studied to date (Tamura et al., 1993). TRANSMISSION BETWEEN IPANs (Figure 1.16) IPANs are relatively unusual amongst primary afferent neurons in general because they make significant synaptic connections with each other. Ultrastructural studies indicate that calbindin neurons make synaptic connections with virtually all other myenteric neurons including Dogiel type II neurons (Pompolo and Furness, 1988). Furthermore, intracellular stimulation of an individual myenteric IPAN can evoke slow EPSPs in a second myenteric IPAN, if the stimulated neuron has an axon contacting the other neuron (Kunze et al., 1993). Repeated mucosal distortion enhances distension reflexes, even when the direct reflex response to mucosal stimulation has disappeared, which suggests an excitatory interaction between the submucosal and myenteric IPANs (Smith, Bornstein and Furness, 1991). This phenomenon persists for as long as the mucosal stimulation and for a considerable time afterwards suggesting that the decline in the direct response is not due to cessation of firing in the submucosal IPANs. The simplest explanation is that prolonged firing of mechanoreceptive submucosal IPANs evokes slow EPSPs in distension-sensitive myenteric IPANs thereby increasing their excitability. This implies differences in the transmission process between synapses from IPANs to interneurons and from IPANs to myenteric IPANs. As discussed above, there is strong evidence that the mechanosensitive submucosal IPANs transmit to ascending interneurons via fast EPSPs mediated by ACh acting on nicotinic receptors and slow EPSPs mediated via NK3 tachykinin receptors. The same

ENTERIC REFLEXES THAT INFLUENCE MOTILITY

37

Figure 1.16 Transmitters employed at synapses between IPANs. No distinction is made between different classes of IPANs in this diagram, because as yet pharmacological studies have not allowed such distinctions to be drawn. Both tachykinins acting at NK3 and NK1 receptors and ACh acting at muscarinic receptors appear to be involved. The predominant role of NK3 receptors is indicated by the heavier type.

neurons appear to transmit to descending interneurons via slow EPSPs mediated by a tachykinin acting on NK3 receptors and via another, unidentified, transmitter system. These conclusions are consistent with the immunohistochemical findings that submucosal IPANs are immunoreactive for both tachykinins and ChAT (Bornstein and Furness, 1988; Kirchgessner and Gershon, 1988; Kirchgessner, Tamir and Gershon, 1992; Kirchgessner, Liu and Gershon, 1996). Maintained distension does not enhance responses to mucosal distortion in the same region (Smith, Bornstein and Furness, 1991). This suggests that distension-sensitive myenteric IPANs do not provide synaptic input to submucosal IPANs. However, myenteric IPANs provide terminals to submucous ganglia (Furness et al., 1990) and there is indirect evidence that some slow EPSPs recorded in submucosal IPANs are due to activation of those terminals (Galligan et al., 1988). Other evidence that IPANs directly excite IPANs comes from analysis of the responses to chemical stimuli applied to the mucosa just circumferential to an impaled AH/Dogiel type II neuron. Although most neurons that responded did so by firing a burst of action potentials, some also exhibited a slow EPSP-like response and others only showed this slow EPSP-like response to the stimulus (Bertrand et al., 1997, 1998). Consideration of the known anatomy suggests that the slow EPSPs arose from connections between chemosensitive IPANs and other IPANs. This is consistent with the finding that chemical stimulation of the mucosa enhances reflexes evoked by distension (Smith, Bornstein and Furness, 1991). Thus, a variety of evidence indicates that mechanoreceptive IPANs with cell bodies in the submucosa and myenteric chemosensitive IPANs each modulate the activity of distension-sensitive IPANs via slow EPSPs. The question then arises whether mechanoreceptive myenteric IPANs have similar outputs to other myenteric IPANs, whether

38

INNERVATION OF THE GASTROINTESTINAL TRACT

mechano- or chemosensitive. Many myenteric IPANs in freely contracting stretched preparations of guinea-pig ileum are more excitable than in relaxed preparations (Kunze et al., 1998). The prolonged AHPs, characteristic of IPANs in relaxed preparations, are greatly diminished in stretched contracting preparations and the firing of IPANs accommodates more slowly in the latter condition. These are all features characteristic of the behaviour of IPANs during slow EPSPs, which typically reduce the amplitudes and durations of the AHPs in IPANs (Grafe, Mayer and Wood, 1980; Morita and North, 1985). This suggests that the firing of the myenteric IPANs sensitive to maintained tension leads to maintained slow EPSPs in virtually all other myenteric IPANs. These are not detected as individual synaptic events, but rather are seen as an overall increase in the excitability of the target neurons. Myenteric IPANs also exhibit slow EPSPs when pathways running longitudinally along the myenteric plexus are stimulated electrically, even when the stimulus is applied up to 1 cm away (Hodgkiss and Lees, 1984). The source of these slow EPSPs is unknown, but they may be due to activation of the long anally directed axons of distension-sensitive IPANs.

SENSORY TRANSDUCTION STRETCH As discussed above, the mechanosensitive myenteric IPANs can be inferred to have up to three distinct responses to stretch. Many of these neurons have been shown in electrophysiological recordings to respond to sustained stretch when the muscle is free to contract (Kunze et al., 1998). Others, or a subset of these, respond to the onset of stretch independently of muscle contraction (see above), while some appear to also respond to the passive length of the muscle (see above). In each case, the actual mechanisms that transduce changes in the muscle coat into neuronal firing have not been determined. They presumably all involve the presence of mechanosensitive ion channels in the neuronal membranes with the exact nature of the stimulus being determined by the nature of the coupling of these channels to the extracellular matrix or the intestinal muscle. Indeed, the IPANs that respond to sustained stretch in contracting preparations lose their ability to respond to such stretch when the collagen coupling them to the muscle is enzymatically removed (Kunze et al., 1999).

MUCOSAL MECHANORECEPTORS The cell bodies of the IPANs mediating responses to mechanical distortion of the mucosa have been localised to the submucosa (see above). There is a large body of evidence indicating that sensory stimuli act on the mucosa to release 5-HT from the enterochromaffin cells, and that this excites IPAN terminals in the mucosa. 5-HT applied to the lumen enhances peristaltic reflexes (Bülbring and Crema, 1958; Bülbring and Lin, 1958; Craig and Clarke, 1991; Costall, Naylor and Tuladhar, 1993; Tuladar, Kaisar and Naylor, 1997). Specific antagonists for 5-HT3 and 5-HT4 receptors either by themselves or in combination

ENTERIC REFLEXES THAT INFLUENCE MOTILITY

39

block reflexes evoked by mucosal distortion in the dog ileum (Neya, Mizutani and Yamasato, 1993), the human jejunum (Foxx-Orenstein, Kuemmerle and Grider, 1996), the rat colon (Grider, Kuemmerle and Jin, 1996) and the guinea-pig colon (Foxx-Orenstein, Kuemmerle and Grider, 1996). Studies in the guinea-pig small intestine have been less compelling with 5-HT3 and 5-HT4 antagonists having no apparent effect on the initiation of reflexes by mucosal deformation when applied to the region of the IPANs (Yuan, Bornstein and Furness, 1994). However, blockade of 5-HT1P receptors prevents the expression of Fos protein normally evoked in submucous AH/Dogiel type II neurons by mechanical deformation of the mucosa (Kirchgessner, Tamir and Gershon, 1992). Thus, 5-HT may act as a sensory intermediate between mechanical deformation of the mucosa and the IPANs that mediate the subsequent reflexes. Release of 5-HT from mucosal epithelial cells by mechanical deformation must be at least partially independent of extracellular Ca2+. Indeed, the mechanisms transducing chemical and distension stimuli must also be partially Ca2+-independent. Low Ca2+, high Mg2+ solutions in the region of the stimulated IPANs do not block descending reflexes evoked by mechanical deformation of the mucosa or by distension (Johnson et al., 1996; Johnson, Bornstein and Burcher, 1998) nor do they alter responses of chemosensitive IPANs to either acid or neutral acetate in the guinea-pig small intestine (Kunze et al., 1995; Bertrand et al., 1997). Calcitonin gene-related peptide (CGRP) has also been implicated in the sensory transduction process in studies of rat colon, human jejunum and guinea-pig colon (Grider, 1994; Foxx-Orenstein, Kuemmerle and Grider, 1996; Grider, Kuemmerle and Jin, 1996). However, immunohistochemical localization of CGRP in the intestinal mucosa to date do not provide a clear picture of where this peptide acts within the reflex pathways. In particular, most of the CGRP in mucosa of the guinea-pig ileum seems to be contained either within the terminals of extrinsic primary afferent neurons or in a population of secretomotor neurons (Furness, Costa and Keast, 1984; Gibbins et al., 1985). CHEMORECEPTORS The mechanisms of transduction of chemical stimuli applied to the mucosa are less clear. The mucosal epithelium represents a barrier between the stimulant and the sensory terminals of chemosensitive IPANs and whether the stimulants cross the mucosa or lead to release of a sensory mediator from epithelial cells is unknown. Electrical stimulation of the mucosa can evoke delayed bursts of action potentials in myenteric chemosensitive IPANs and these have been attributed to release of a chemical mediator from the mucosal epithelium (Bertrand et al., 1997, 1998). It is tempting to suggest that it is this unidentified mediator that transduces chemical stimuli applied to the mucosa into neural activity. The mediator is unlikely to be 5-HT, because 5-HT applied to the mucosa stimulates the chemosensitive myenteric IPANs of the guinea-pig ileum via 5-HT3 receptors and the delayed bursts of action potentials are unaffected by 5-HT3 receptor antagonists (Bertrand et al., 2000). Thus, other chemicals found within mucosal endocrine cells mediate responses to chemical stimuli. For example, CCK is found within these cells (Rehfeld, 1978; Noyer et al., 1980; Williams, 1982), enhances intestinal motility (Stewart and Bass, 1976a; Barthó et al., 1982; Scheurer et al., 1983; Chang, Lotti and Chen, 1984; Rodriguez-Membrilla et al.,

40

INNERVATION OF THE GASTROINTESTINAL TRACT

1995) and is a powerful excitant of enteric neurons. Indeed, myenteric IPANs are specifically excited by this peptide when it is applied to their cell bodies (Schutte et al., 1997). However, whether CCK or another peptide mediates chemical stimuli remains to be investigated.

RELATIONSHIPS BETWEEN SIMPLE REFLEXES, MOTOR PATTERNS AND MORE COMPLEX BEHAVIOUR Most of the discussion thus far has been about the simple reflexes evoked by stationary stimuli, usually of a single sensory modality, but the real situation is much more complex. When food is present in the intestine, the behaviour of any region will be determined by the interactions between simple reflexes evoked at many locations and via several different sensory modalities at each location, together with any neural mechanisms that may be activated solely by mixed stimuli. This might suggest that the overall outcome would be impossible to predict. However, some clear and consistent behaviours are seen in these circumstances and these appear to result from activation of motor patterns more complex than those responsible for the simple reflexes. Peristalsis, or the peristaltic reflex, is the most intensively studied of the motor patterns, largely because it can be repeatedly activated in a stereotyped fashion in vitro (see above). Other motor patterns seen in vivo such as the MMC and segmentation have been less easy to analyse. Once activated, the peristaltic pattern evoked by fluid distension of the small intestine involves a contraction of the circular muscle oral to a distended region of intestine containing much of the original fluid stimulus (a bolus) and a relaxation anal to the contracted region (Hennig et al., 1999). Both the contraction and the relaxation progress anally pushing the contents in an anal direction. This coordinated behaviour usually propagates for almost the full length of the intestinal segment in vitro, but in vivo the contraction typically dies away after progressing for varying distances along the gut. There have been many studies of peristaltic motor patterns and it has been assumed that they arise because the simple ascending excitatory and descending inhibitory reflexes are excited by the distension at the level of the fluid bolus. The anal propagation of the bolus is then due to movement of the bolus into a new segment of intestine, thereby stimulating a new set of reflexes. This idea is attractive in its simplicity, however, there are several observations, both recent and old, that cast doubt on the idea that the peristaltic motor pattern is a moving version of the stationary reflexes. This can be illustrated by three recent in vitro studies. In the first, it was found that the ascending excitatory reflex is evoked at a lower threshold distension than the peristaltic motor pattern (Tonini et al., 1996). Thus, activation of the ascending excitatory reflex is not sufficient to initiate propulsion of fluid along the intestine. In the second, it was found that circumferential stretch of an opened segment of intestine evoked a contraction of the circular muscle that progressed anally even though the stimulus could not move (Brookes et al., 1999a,b). In a third, an anally propagating peristaltic wave was initiated in an isolated segment of guinea-pig ileum by a localized mechanical stimulus even though the segment contained no content and so the stimulus could not move (Spencer, Walsh and Smith, 1999). These studies indicate that the peristaltic motor pattern involves activation of neural elements additional to those excited by

ENTERIC REFLEXES THAT INFLUENCE MOTILITY

41

simple reflexes and the second and third studies imply the existence of an anally directed excitatory pathway. Further evidence that the simple reflexes do not fully account for the peristaltic motor pattern comes from observations of the behaviour of the intestine or colon when a solid bolus is used to stimulate peristalsis. In both regions, it is common to observe a wave of circular muscle contraction passing over the bolus (Bolzer, 1949b), rather than remaining on the oral side of the bolus as would be expected if the contractions depended only on ascending excitation. Any model of the neural mechanisms underlying propulsive motor patterns needs to account for the basic properties and interconnections of the various functional classes of neurons identified above. Two features appear to be critical: the fact that virtually all neurons receive synaptic inputs from local IPANs and the fact that many of these inputs are via slow EPSPs lasting from several seconds to minutes. Figure 1.9 illustrates the basic circuit that can be deduced for the interconnections of the IPANs with each other, ignoring the possibility that there is a population of length sensitive neurons. Slowly adapting tensionsensitive IPANs contact most other myenteric IPANs including other neurons of the same class, chemosensitive IPANs and rapidly adapting distension-sensitive myenteric IPANs. The last population may also receive input from submucosal mechanosensitive IPANs. Chemosensitive neurons also contact the rapidly adapting distension-sensitive IPANs. All of these connections operate via slow EPSPs, so that changes in neuronal excitability can be expected to last much longer than the stimulus. POSSIBLE FEEDBACK EFFECTS WHEN THE MUSCLE TENSION CHANGES IPANs that are activated by muscle tension are in a peculiar situation, because the consequence of their activation, a change in muscle tension, will in turn change their excitability. We can speculate on the consequences of this feedback. Many of these IPANs provide inputs via slow EPSPs to neighbouring IPANs, interneurons and motor neurons (Kunze et al., 1993, 1997, 1998) (Figure 1.6B). Thus, changes in muscle tension would alter the firing of the slowly adapting tension-sensitive IPANs and, over time, modify the excitability of all the other myenteric IPANs and indeed all the other myenteric neurons in the region. This slow modulation of excitability would not play a role in the simple reflexes recorded electrophysiologically that were described above. Those reflexes were evoked under conditions where muscle contractions were prevented by blockade of L-type Ca2+ channels, thereby precluding ongoing activity in the slowly adapting tension-sensitive IPANs. By contrast, the descending excitatory reflex identified by Spencer, Walsh and Smith (1999, 2000) may depend on the ability of the muscle to contract thereby loading the slowly adapting receptors. This may account for the ability of P2 antagonists in the intermediate chamber of a three chambered organ bath to depress this reflex, even though they have no effect on descending inhibitory reflexes. However, the modulation of IPAN firing by activity in other IPANs can be expected to play a significantly greater role in responses to fluid distension in vitro or to the presence of food in the lumen in vivo. A slow fluid distension would increase firing in some slowly adapting tension-sensitive IPANs, which would then excite local motor neurons and also increase the excitability of other slowly adapting and rapidly adapting stretch-sensitive

42

INNERVATION OF THE GASTROINTESTINAL TRACT

IPANs. Because the local motor neurons include both excitatory and inhibitory motor neurons, the tension change in the muscle would depend on the relative strength of the input from local IPANs, on their local projections and the interactions of their outputs within the muscle. The excitatory motor neurons project either directly to the circular muscle or orally, but the inhibitory motor neurons project anally. Thus, it might be predicted that the effect of a slow stretch in a given region would be to increase the tension in the circular muscle at the oral edge of the region by comparison to that at the anal edge. This would tend to load the more oral slowly adapting tension-sensitive IPANs, while unloading the equivalent neurons lying more anally. Thus, firing would increase in the oral tension-sensitive IPANs and the rapidly adapting distension-sensitive IPANs would become more excitable, while more anal neurons would remain less excitable. The muscle at the oral end of the region originally distended would provide a positive feedback into the motor circuit. The effect of this in an isolated segment would be for a slow rise in excitation in the most oral part of the segment with only a small increase in tension until a threshold is passed and all the local neurons would fire together. Such a slow increase in tension leading to a threshold has recently been demonstrated in a slowly stretched opened segment of intestine which appears to mimic many of the normal properties of the peristaltic motor pattern (Brookes et al., 1999a). A key feature of this mechanism is that the peristaltic motor pattern is initiated at a site where descending inhibitory pathways are not active either because there is no more oral stimulus or because the reflex pathways have been interrupted. The latter has been observed in isolated intestinal segments in which the myenteric plexus has been lesioned (Waterman, Costa and Tonini, 1994b). In the intact intestine, antiperistalis resulting from the excitatory feedback would be prevented by the presence of food in more oral segments exciting descending inhibitory reflexes. Chemical stimulants within the lumen may modulate the positive feedback IPAN circuit in at least two ways. Firstly, chemical stimulants excite descending inhibitory pathways, so that slow distensions anal to a local increase in the concentration of a stimulant would be less likely to excite slowly adapting tension-sensitive IPANs and peristalsis would be more difficult to initiate. In essence, this would reduce the ability of a peristaltic motor pattern to enter the relaxed region until the chemical either dispersed or was absorbed. This would provide a mechanism for ensuring that digesta remains in a specific region to allow further digestion and absorption. Secondly, a raised concentration of a chemical at a particular location would tend to set the oral edge of a functional segment, because activity in chemosensitive IPANs would have a similar effect to that of slowly adapting tension-sensitive IPANs. This would feed into the same positive feedback loop excited by slow distension, thereby exaggerating any effects of distension and localising them to the region containing the chemicals. The chemical stimulant would be expected to reduce the threshold for initiation of the peristaltic motor pattern in this region. Widely distributed chemicals in the intestinal lumen enhance intestinal propulsion both in vivo (Thomas and Baldwin, 1971; Richardson et al., 1991) and in vitro (Bülbring and Crema, 1958; Bülbring and Lin, 1958; Craig and Clarke, 1991; Costall, Naylor and Tuladhar, 1993; Tuladar, Kaisar and Naylor, 1997). The pattern would, however, not propagate very far until the local concentration of the chemical stimulant falls to a level that no longer evokes descending inhibition to act as a brake on the effect of distension. Removal of the chemical stimulus might be expected to lead to

ENTERIC REFLEXES THAT INFLUENCE MOTILITY

43

a more rapid decline in the excitability of the inhibitory motor neurons than that of the local IPANs, because the slow EPSPs in S-neurons are much briefer than those in AH/Dogiel type II neurons (Bornstein et al., 1984b). Thus, the inhibitory input to the muscle may decline while the excitability of mechanoreceptive myenteric IPANs is still elevated. Accordingly, once a particular component of the digesta has been absorbed, the peristaltic motor pattern would be released and progress for some distance until descending inhibition again unloaded slowly adapting mechanoreceptive IPANs in front of it. The idea that a descending inhibition brakes the peristaltic motor pattern by unloading tension-sensitive neurons is also consistent with the long anal projections of the rapidly adapting stretch-sensitive IPANs (Brookes et al., 1995; Johnson et al., 1996; Johnson, Bornstein and Burcher, 1998). These neurons excite the inhibitory motor neurons (Johnson, Bornstein and Burcher, 1998) and would relax the circular muscle and unload slowly adapting tension-sensitive neurons. Thus, a component of the simple descending inhibitory reflexes would be expected to increase the overall threshold for the peristaltic motor program anal to the point of maximum activity. This would tend to slow the propagation of the peristaltic wave, if not prevent its propagation. Evidence for the existence of a nutrient driven brake on the peristaltic motor pattern has come from studies of intestinal motor patterns in vivo, where glucose infusion has been shown to produce a marked inhibition of jejunal contractions (Schmid and Ehrlein, 1993). These examples of possible interactions between IPANs, the excitatory and inhibitory motor neurons and the muscle show how several different motor patterns may be generated within a common, relatively simple, reflex circuit. The actual behaviour will depend on the temporal and spatial balance of activation of the different IPANs. This will depend on the chemical state of the lumen, the texture of the luminal content, the contractile state of the muscle, and the rate of filling and emptying of the segment. Because the muscle normally has its own rhythmic pacemakers, the slow waves, in any final output one would also need to consider the intrinsic rhythm that would be imposed by these upon the activity of slowly adapting mechanosensitive IPANs. These interactions may also affect the results of pharmacological experiments on simple reflexes evoked by stationary stimuli. For example, blockade of NK1 tachykinin receptors in the recording chamber of a divided organ bath depresses descending inhibitory reflexes recorded in the circular muscle (Johnson, Bornstein and Burcher, 1998). While this effect appears specific for transmission from descending interneurons to inhibitory motor neurons, it may not be due to direct interference with transmission between these two classes of neurons. Many IPANs are spontaneously active when the mucosa is intact (Kunze et al., 1997; Bertrand et al., 1997, 1998), so that the inhibitory motor neurons would be expected to be receiving a background slow EPSP input from local IPANs (Kunze et al., 1997). The NK1 antagonist may block the background activity, thereby reducing the excitability of the inhibitory motor neurons. If the safety factor for transmission from descending interneurons is low relative to the descending axons of IPANs, then the apparent effect would be a specific reduction in transmission from the former to the inhibitory motor neurons. The predominant motor pattern in an animal fed an inert meal appears to be propulsive, i.e. propagating waves of circular muscle contraction that are very similar to the peristaltic motor pattern seen in vitro (Schemann and Ehrlein, 1986c; Buhner and Ehrlein, 1989).

44

INNERVATION OF THE GASTROINTESTINAL TRACT

By constrast, when the meal contains nutrients, the predominant pattern consists of local contractions that spread only short distances along the intestine (Schemann and Ehrlein, 1986c; Buhner and Ehrlein, 1989). This appears to the analogue of the segmentation originally described by Cannon. Whether this is a neurally mediated motor programme is unclear. It has been argued that segmentation, which has not been recognised in vitro, is actually a myogenic phenomenon (Hasler, 1991). However, as can be seen from the discussion above stationary or slowly propagating rings of contraction may be produced by simple enteric reflex circuits by varying the level of excitability of the different neural elements. It is tempting to speculate that it is the release of the chemical (nutrient) brake postulated above which switches the motor activity of a segment of intestine from mixing to propulsion.

A FINAL COMMENT Over the last few years, methods for study of enteric reflexes and reflexly activated motor programs have grown in power and sophistication. However, many questions remain to be answered because the neural architecture is complex, there is a wide range of different neural interactions and several different stimuli may be considered to be physiological. The motor programmes and complex behaviour that govern intestinal function will probably prove to be elaborations of the neuronal mechanisms responsible for the simple reflexes described here. However, these represent only the base on which many other influences, like circulating hormones and the extrinsic nerves, impinge.

ACKNOWLEDGEMENTS This work was supported by a Program Grant (963213) of the NH&MRC (Australia).

REFERENCES Azpiroz, F. and Malagelada, J.-R. (1990). Perception and reflex relaxation of the stomach in response to gut distention. Gastroenterology, 98, 1193–1198. Baldwin, M.V. and Thomas, J.E. (1975). The intestinal intrinsic mucosal reflex; a possible mechanism of propulsive motility. In Functions of the stomach and intestine, edited by M.H.F. Friedman, pp. 75–91. Baltimore: University Park Press. Barthó, L., Holzer, P., Donnerer, J. and Lembeck, F. (1982). Effects of substance P, cholecystokinin octapeptide, bombesin, and neurotensin on the peristaltic reflex of the guinea-pig ileum in the absence and in the presence of atropine. Naunyn-Schmiedeberg’s Archives of Pharmacology, 321, 321–328. Barthó, L., Holzer, P. and Lembeck, F. (1989). Evidence for an involvement of substance P, but not cholecystokininlike peptides, in hexamethonium-resistant intestinal peristalsis. Neuroscience, 28, 211–217. Bauer, V. and Kuriyama, H. (1982). Evidence for non-cholinergic, non-adrenergic transmission in the guinea-pig ileum. Journal of Physiology, 330, 95–110. Bayliss, W.M. and Starling, E.H. (1899). The movements and innervation of the small intestine. Journal of Physiology, 24, 99–143. Bayliss, W.M. and Starling, E.H. (1900a). The movements and innervation of the small intestine. Journal of Physiology, 26, 125–138.

ENTERIC REFLEXES THAT INFLUENCE MOTILITY

45

Bayliss, W.M. and Starling, E.H. (1900b). The movements and the innervation of the large intestine. Journal of Physiology, 26, 107–118. Bertrand, P.P. and Galligan, J.J. (1994). Contribution of chloride conductance increase to slow EPSC and tachykinin current in guinea-pig myenteric neurones. Journal of Physiology, 481, 47–60. Bertrand, P.P. and Galligan, J.J. (1995). Signal-transduction pathways causing slow synaptic excitation in guinea pig myenteric AH neurons. American Journal of Physiology, 269, G710–G720. Bertrand, P.P., Kunze, W.A.A., Bornstein, J.C., Furness, J.B. and Smith, M.L. (1997). Analysis of the responses of myenteric neurons in the small intestine to chemical stimulation of the mucosa. American Journal of Physiology, 273, G422–G435. Bertrand, P.P., Kunze, W.A.A., Bornstein, J.C. and Furness, J.B. (1998). Electrical mapping of the projections of intrinsic primary afferent neurons to the mucosa of the guinea-pig small intestine. Neurogastroenterology and Motility, 10, 533–541. Bertrand, P.P., Kunze, W.A.A., Furness, J.B. and Bornstein, J.C. (2000). The terminals of myenteric intrinsic primary afferent neurons of the guinea-pig ileum are excited by 5-hydroxytryptamine acting at 5-hydroxytryptamine-3 receptors. Neuroscience, 101, 459–469. Bian, X.-C., Bertrand, P.P. and Bornstein, J.C. (2000). Descending inhibitory reflexes involve P2X mediated transmission from interneurons to motor neurons in guinea-pig ileum. Journal of Physiology, 528, 551–560. Bolzer, E. (1949a). Myenteric reflex. American Journal of Physiology, 157, 329–337. Bolzer, E. (1949b). Reflex peristalsis of the intestine. American Journal of Physiology, 157, 338–342. Bornstein, J.C. and Furness, J.B. (1988). Correlated electrophysiological and histochemical studies of submucous neurons and their contribution to understanding enteric neural circuits. Journal of the Autonomic Nervous System, 25, 1–13. Bornstein, J.C., Costa, M., Furness, J.B. and Lees, G.M. (1984a). Electrophysiology and enkephalin immunoreactivity of identified myenteric plexus neurones in the guinea-pig small intestine. Journal of Physiology, 351, 313–325. Bornstein, J.C., North, R.A., Costa, M. and Furness, J.B. (1984b). Excitatory synaptic potentials due to activation of neurons with short projections in the myenteric plexus. Neuroscience, 11, 723–731. Bornstein, J.C., Costa, M., Furness, J.B. and Lang, R.J. (1986). Electrophysiological analysis of projections of enteric inhibitory motor neurones in the guinea-pig small intestine. Journal of Physiology, 370, 61–74. Bornstein, J.C., Furness, J.B. and Costa, M. (1989). An electrophysiological comparison of substance P-immunoreactive neurons with other neurons in the guinea-pig submucous plexus. Journal of the Autonomic Nervous System, 26, 113–120. Bornstein, J.C., Furness, J.B., Smith, T.K. and Trussell, D.C. (1991a). Synaptic responses evoked by mechanical stimulation of the mucosa in morphologically characterized myenteric neurons of the guinea pig ileum. Journal of Neuroscience, 11, 505–518. Bornstein, J.C., Hendriks, R., Furness, J.B. and Trussell, D.C. (1991b). Ramifications of the axons of AH-neurons injected with the intracellular marker biocytin in the myenteric plexus of the guinea pig small intestine. Journal of Comparative Neurology, 314, 437–451. Bornstein, J.C., Furness, J.B. and Kunze, W.A.A. (1994). Electrophysiological characterization of myenteric neurons, how do classification schemes relate? Journal of the Autonomic Nervous System, 48, 1–15. Bornstein, J.C., Furness, J.B., Kelly, H.F., Bywater, R.A.R., Neild, T.O. and Bertrand, P.P. (1997). Computer simulation of the enteric neural circuits mediating an ascending reflex: roles of fast and slow excitatory outputs of sensory neurons. Journal of the Autonomic Nervous System, 64, 143–157. Bornstein, J.C., Furness, J.B., Kelly, H.F., Bertrand, P.P. and Kunze, W.A.A. (1998). Simulation of the activities of large-scale, anatomically realistic networks of intestinal (enteric) neurons during reflexes. In Humans to Proteins, edited by M.T. Michalewicz, pp. 77–95. Melbourne: CSIRO Publishing. Branchek, T.A., Mawe, G.M. and Gershon, M.D. (1988). Characterization and localization of a peripheral neural 5-hydroxytryptamine receptor subtype (5-HT1P) with a selective agonist, 3H-5-hydroxyindalpine. Journal of Neuroscience, 8, 2582–2595. Brehmer, A. and Stach, W. (1998). Regional structural differences in the neuronal composition of myenteric ganglia along the pig small intestine. Anatomical Record, 250, 109–116. Brehmer, A., Stach, W., Krammer, H.J. and Neuhuber, W. (1998). Distribution, morphology and projections of nitrergic and non-nitrergic submucosal neurons in the pig small intestine. Histochemistry and Cell Biology, 109, 87–94. Brookes, S.J.H., Ewart, W.R. and Wingate, D.L. (1987). Intracellular recordings from myenteric neurones in the human colon. Journal of Physiology, 390, 305–318. Brookes, S.J.H., Ewart, W.R. and Wingate, D.L. (1988). Intracellular recordings from cells in the myenteric plexus of the rat duodenum. Neuroscience, 24, 297–307. Brookes, S.J.H., Steele, P.A. and Costa, M. (1991). Calretinin immunoreactivity in cholinergic motor neurones, interneurones and vasomotor neurones in the guinea-pig small intestine. Cell and Tissue Research, 263, 471–481.

46

INNERVATION OF THE GASTROINTESTINAL TRACT

Brookes, S.J.H., Song, Z.-M., Steele, P.A. and Costa, M. (1992). Identification of motor neurons to the longitudinal muscle of the guinea-pig ileum. Gastroenterology, 103, 961–973. Brookes, S.J.H., Song, Z.-M., Ramsay, G.A. and Costa, M. (1995). Long aboral projections of Dogiel type II, AH neurons within the myenteric plexus of the guinea-pig small intestine. Journal of Neuroscience, 15, 4013–4022. Brookes, S.J.H., Meedeniya, A.C.B., Jobling, P. and Costa, M. (1997). Orally projecting interneurones in the guinea-pig small intestine. Journal of Physiology, 505, 473–491. Brookes, S.J.H., Chen, B.N., Costa, M. and Humphreys, C.M.S. (1999a). Initiation of peristalsis by circumferential stretch of flat sheets of guinea-pig ileum. Journal of Physiology, 516, 525–538. Brookes, S.J.H., D’Antona, G., Zagorodnyuk, V.P., Humphreys, C.M.S. and Costa, M. (1999b). Mechanisms of propagation of peristalsis in the guinea-pig ileum. Proceedings of the Australian Neuroscience Society, 10, 97. Buchheit, K.-H. and Buhl, T. (1991). Prokinetic benzamides stimulate peristaltic activity in the isolated guinea pig ileum by activation of 5-HT4 receptors. European Journal of Pharmacology, 205, 203–208. Buhner, S. and Ehrlein, H.J. (1989). Characteristics of postprandial duodenal motor patterns in dogs. Digestive Diseases and Science, 34, 1873–1881. Buhner, S., Ehrlein, H.J., Thoma, G. and Schumpelick, V. (1988). Effects of nutrients on gastrointestinal motility and gastric emptying after Billroth-I gastrectomy in dogs. Digestive Diseases and Science, 33, 784–794. Bülbring, E. and Crema, A. (1958). Observations concerning the action of 5-hydroxytryptamine on the peristaltic refllex. British Journal of Pharmacology, 13, 444–457. Bülbring, E. and Lin, R.C.Y. (1958). 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. Journal of Physiology, 140, 381–407. Bywater, R.A.R. and Taylor, G.S. (1983). Non-cholinergic fast and slow post-stimulus depolarization in the guinea-pig ileum. Journal of Physiology, 340, 47–56. Bywater, R.A.R. and Taylor, G.S. (1986). Non-cholinergic excitatory and inhibitory junction potentials in the circular smooth muscle of the guinea-pig ileum. Journal of Physiology, 374, 153–164. Cannon, W.B. (1902). The movements of the intestines studied by means of the Roentgen rays. American Journal of Physiology, 6, 251–277. Cannon, W.B. (1906). The motor activities of the stomach and small intestine after splanchnic and vagus section. American Journal of Physiology, 17, 429–442. Cannon, W.B. (1912). Peristalsis, segmentation and the myenteric reflex. American Journal of Physiology, 30, 114–128. Chang, R.S.L., Lotti, V.J. and Chen, T.-B. (1984). Further evidence that substance P partly mediates the action of cholecystokinin octapeptide (CCK-8) on the guinea pig ileum, but not gall bladder: studies with substance P antagonist. Neuroscience Letters, 46, 71–75. Christofi, F.L. and Wood, J.D. (1993a). Effects of PACAP on morphologically identified myenteric neurons in guinea pig small bowel. American Journal of Physiology, 264, G414–G421. Christofi, F.L. and Wood, J.D. (1993b). Presynaptic inhibition by adenosine A1 receptors on guinea pig small intestinal myenteric neurons. Gastroenterology, 104, 1420–1429. Ciccocioppo, R., Onori, L., Messori, E., Candura, S.M., Coccini, T. and Tonini, M. (1994). Role of nitric oxidedependent and -independent mechanisms in peristalsis and accomodation in the rabbit distal colon. Journal of Pharmacology and Experimental Therapeutics, 270, 929–937. Clerc, N., Furness, J.B., Bornstein, J.C. and Kunze, W.A.A. (1998a). Correlation of electrophysiological and morphological characteristics of myenteric neurons of the duodenum in the guinea-pig. Neuroscience, 82, 899–914. Clerc, N., Furness, J.B., Li, Z.S., Bornstein, J.C. and Kunze, W.A.A. (1998b). Morphological and immunohistochemical identification of neurons and their targets in the guinea-pig duodenum. Neuroscience, 86, 679–694. Clerc, N., Furness, J.B., Kunze, W.A., Thomas, E.A. and Bertrand, P.P. (1999). Long-term effects of synaptic activation at low frequency on excitability of myenteric AH neurons. Neuroscience, 90, 279–289. Copray, J.C.V.M., Liem, R.S.B., Ter Horst, G.J. and van Willigen, J.D. (1990). Origin, distribution and morphology of serotonergic afferents to the mesencephalic trigeminal nucleus of the rat. Neuroscience Letters, 121, 97–101. Cornelissen, W., Timmermans, J.-P., Van Bogaert, P.P. and Scheuermann, D.W. (1996). Electrophysiology of porcine myenteric neurons revealed after vital staining of their cell bodies. A preliminary report. Neurogastroenterology and Motility, 8, 101–109. Costa, M. and Furness, J.B. (1976). The peristalic reflex: An analysis of the nerve pathways and their pharmacology. Naunyn-Schmiedebergs Archives of Pharmacology, 294, 47–60. Costa, M., Furness, J.B., Llewellyn-Smith, I.J. and Cuello, A.C. (1981). Projections of substance P-containing neurons within the guinea-pig small intestine. Neuroscience, 6, 411–424.

ENTERIC REFLEXES THAT INFLUENCE MOTILITY

47

Costa, M., Furness, J.B., Pullin, C.O. and Bornstein, J. (1985). Substance P neurons mediate non-cholinergic transmission to the circular muscle of the guinea-pig intestine. Naunyn-Schmiedeberg’s Archives of Pharmacology, 328, 446–453. Costa, M., Brookes, S.J.H., Waterman, S.A. and Mayo, R. (1992). Enteric neuronal circuitry and transmitters controlling intestinal motor function. In Advances in the Innervation of the Gastrointestinal Tract, edited by G.E. Holle and J.D. Wood, pp. 115–121. Amsterdam, London, New York, Tokyo: Exerpta Medica, Elsevier Scientific Publishers B.V. Costa, M., Brookes, S.J.H., Steele, P.A., Gibbins, I.L., Burcher, E. and Kandiah, C.J. (1996). Neurochemical classification of myenteric neurons in the guinea-pig ileum. Neuroscience, 75, 949–967. Costall, B., Naylor, R.J. and Tuladhar, B.R. (1993). 5-HT4 receptor-mediated facilitation of the emptying phase of the peristaltic reflex in the guinea-pig isolated ileum. British Journal of Pharmacology, 110, 1572–1578. Craig, D.A. and Clarke, D.E. (1991). Peristalsis evoked by 5-HT and renzapride: evidence for putative 5-HT4 receptor activation. British Journal of Pharmacology, 102, 563–564. Crema, A., Frigo, G.M. and Lecchini, S. (1970). A pharmacological analysis of the peristaltic reflex in the isolated colon of the guinea-pig or cat. British Journal of Pharmacology, 39, 334–345. Davison, J.S. and Pearson, G.T. (1979). The role of intrinsic, non-adrenergic, non-cholinergic inhibitory nerves in the regulation of distensibility of the guinea-pig colon. Pflügers Archiv, 381, 75–77. Dogiel, A.S. (1899). Ueber den Bau der Ganglien in den Geflechten des Darmes und der Gallenblase des Menschen und der Säugetiere. Archiv für Anatomie und Physiologie, Anatomische Abteilung Jahrgang 1899, 130–158. Eeckhout, C., Vantrappen, G., Peeters, T.L., Janssens, J. and De Wever, I. (1984). Different meals produce different digestive motility patterns. Digestive Diseases and Science, 29, 219–224. Ehrlein, H.J., Thoma, G., Keinke, O., Tsiamitas, C. and Schumpelick, V. (1987). Effects of nutrients on gastrointestinal motility and gastric emptying after distal gastrectomy with Roux-Y gastrojejunostomy in dogs. Digestive Diseases and Science, 32, 538–546. Fich, A., Phillips, S.F., Neri, M., Hanson, R.B. and Zinsmeister, A.R. (1990). Regulation of postprandial motility in the canine ileum. American Journal of Physiology, 259, G767–G774. Foxx-Orenstein, A.E. and Grider, J.R. (1996). Regulation of colonic propulsion by enteric excitatory and inhibitory neurotransmitters. American Journal of Physiology, 271, G433–G437. Foxx-Orenstein, A.E., Jin, J.-G. and Grider, J.R. (1998). 5-HT4 receptor agonists and delta-opioid receptor antagonists act synergistically to stimulate colonic propulsion. American Journal of Physiology, 275, G979–G983. Foxx-Orenstein, A.E., Kuemmerle, J.F. and Grider, J.R. (1996). Distinct 5-HT receptors mediate the peristaltic reflex induced by mucosal stimuli in human and guinea pig intestine. Gastroenterology, 111, 1281–1290. Frigo, G.M. and Lecchini, S. (1970). An improved method for studying the peristaltic reflex in the isolated colon. British Journal of Pharmacology, 39, 346–356. Furness, J.B. and Costa, M. (1982). Neurons with 5-hydroxytryptamine-like immunoreactivity in the enteric nervous system: their projections in the guinea-pig small intestine. Neuroscience, 7, 341–349. Furness, J.B., Bornstein, J.C. and Trussell, D.C. (1988). Shapes of nerve cells in the myenteric plexus of the guinea-pig small intestine revealed by the intracellular injection of dye. Cell and Tissue Research, 254, 561–571. Furness, J.B., Costa, M. and Keast, J.R. (1984). Choline acetyltransferase and peptide immunoreactivity of submucous neurons in the small intestine of the guinea-pig. Cell and Tissue Research, 237, 329–336. Furness, J.B., Trussell, D.C., Pompolo, S., Bornstein, J.C. and Smith, T.K. (1990). Calbindin neurons of the guinea-pig small intestine: quantitative analysis of their numbers and projections. Cell and Tissue Research, 260, 261–272. Furness, J.B., Johnson, P.J., Pompolo, S. and Bornstein, J.C. (1995). Evidence that enteric motility reflexes can be initiated through entirely intrinsic mechanisms in the guinea-pig small intestine. Neurogastroenterology and Motility, 7, 89–96. Furness, J.B., Bornstein, J.C., Kunze, W.A.A. and Clerc, N. (1999). The enteric nervous system and its extrinsic connections. In Textbook of Gastroenterology, edited by T. Yamada, pp. 11–35. Philadelphia: Lippinscott Williams and Wilkins. Furukawa, K., Taylor, G.S. and Bywater, R.A. (1986). An intracellular study of myenteric neurons in the mouse colon. Journal of Neurophysiology, 55, 1395–1406. Galligan, J.J. and Bertrand, P.P. (1994). ATP mediates fast synaptic potentials in enteric neurons. Journal of Neuroscience, 14, 7563–7571. Galligan, J.J., North, R.A. and Tokimasa, T. (1989). Muscarinic agonists and potassium currents in guinea-pig myenteric neurones. British Journal of Pharmacology, 96, 193–203. Galligan, J.J., Tokimasa, T. and North, R.A. (1987). Effects of three mammalian tachykinins on single enteric neurons. Neuroscience Letters, 82, 167–171.

48

INNERVATION OF THE GASTROINTESTINAL TRACT

Galligan, J.J., Surprenant, A., Tonini, M. and North, R.A. (1988). Differential localization of 5-HT1 receptors on myenteric and submucosal neurons. American Journal of Physiology, 255, G603–G611. Gibbins, I.L., Furness, J.B., Costa, M., MacIntyre, I., Hillyard, C.J. and Girgis, S. (1985). Co-localization of calcitonin gene-related peptide-like immunoreactivity with substance P in cutaneous, vascular and visceral sensory neurons of guinea-pigs. Neuroscience Letters, 57, 125–130. Grafe, P., Mayer, C.J. and Wood, J.D. (1979). Evidence that substance P does not mediate slow synaptic excitation within the myenteric plexus. Nature, 279, 720–721. Grafe, P., Mayer, C.J. and Wood, J.D. (1980). Synaptic modulation of calcium-dependent potassium conductance in myenteric neurones in the guinea-pig. Journal of Physiology, 305, 235–248. Grider, J.R. (1989). Identification of neurotransmitters regulating intestinal peristaltic reflex in humans. Gastroenterology, 97, 1414–1419. Grider, J.R. (1994). CGRP as a transmitter in the sensory pathway mediating peristaltic reflex. American Journal of Physiology, 266, G1139–G1145. Grider, J.R. and Jin, J.-G. (1994). Distinct populations of sensory neurons mediate the peristaltic reflex elicited by muscle stretch and mucosal stimulation. Journal of Neuroscience, 14, 2854–2860. Grider, J.R. and Makhlouf, G.M. (1992). Enteric GABA: mode of action and role in the regulation of the peristaltic reflex. American Journal of Physiology, 262, G690–G694. Grider, J.R., Kuemmerle, J.F. and Jin, J.-G. (1996). 5-HT released by mucosal stimuli initiates peristalsis by activating 5-HT4/5-HT1P receptors on sensory CGRP neurons. American Journal of Physiology, 270, G778–G782. Grundy, D. and Scratcherd, T. (1989). Sensory afferents from the gastrointestinal tract. In Handbook of Physiology, 6: The Gastrointestinal System, Vol. I. Motility and Circulation, edited by J.D. Wood, pp. 593–620. Washington, D.C.: American Physiological Society. Hasler, W.L. (1991). Motility of the small intestine. In Textbook of Gastroenterology, edited by T. Yamada, pp. 158–180. Philadelphia: J.B. Lippincott Company. Hasler, W.L. (1999). Motility of the small intestine and colon. In Textbook of Gastroenterology, edited by T. Yamada, pp. 215–245. Philadelphia: Lippincott Williams & Wilkins. Hennig, G.W., Brookes, S.J.H. and Costa, M. (1997). Excitatory and inhibitory motor reflexes in the isolated guinea-pig stomach. Journal of Physiology, 501, 197–212. Hennig, G.W., Costa, M., Chen, B.G. and Brookes, S.J.H. (1999). Quantitative analysis of peristalsis in the guinea-pig small intestine using spatio-temporal maps. Journal of Physiology, 517, 575–590. Hirst, G.D.S. and McKirdy, H.C. (1974). A nervous mechanism for descending inhibition in guinea-pig small intestine. Journal of Physiology, 238, 129–143. Hirst, G.D.S., Holman, M.E. and McKirdy, H.C. (1975). Two descending nerve pathways activated by distension of guinea-pig small intestine. Journal of Physiology, 244, 113–127. Hirst, G.D.S., Holman, M.E. and Spence, I. (1974). Two types of neurones in the myenteric plexus of duodenum in the guinea-pig. Journal of Physiology, 236, 303–326. Hodgkiss, J.P. and Lees, G.M. (1984). Slow intracellular potentials in AH-neurons of the myenteric plexus evoked by repetitive activation of synaptic inputs. Neuroscience, 11, 255–261. Holzer, P. (1989). Ascending enteric reflex: Multiple neurotransmitter systems and interactions. American Journal of Physiology, 256, 540–545. Holzer, P., Schluet, W. and Maggi, C. (1993). Ascending enteric reflex contraction: roles of acetylcholine and tachykinins in relation to distension and propagation of excitation. Journal of Pharmacology and Experimental Therapeutics, 264, 391–396. Huge, A., Weber, E. and Ehrlein, H.J. (1995). Effects of enteral feedback inhibition on motility, luminal flow, and absorption of nutrients in proximal gut of minipigs. Digestive Diseases and Science, 40, 1024–1034. Hukuhara, T. and Fukuda, H. (1965). The motility of the isolated guinea-pig small intestine. Japanese Journal of Physiology, 15, 125–139. Hukuhara, T. and Miyake, T. (1959). The intrinsic reflexes in the colon. Japanese Journal of Physiology, 9, 49–55. Hukuhara, T., Nakayama, S. and Nanba, R. (1960a). Locality of receptors concerned with the intestino-intestinal extrinsic and intestinal muscular intrinsic reflexes. Japanese Journal of Physiology, 10, 414–419. Hukuhara, T., Nakayama, S. and Nanba, R. (1960b). The effects of 5-hydroxytryptamine upon the intestinal motility, especially with respect to the intestinal mucosal intrinsic reflex. Japanese Journal of Physiology, 10, 420–426. Hukuhara, T., Nakayama, S. and Nanba, R. (1961). The role of the intrinsic mucosal reflex in the fluid transport through the denervated colonic loop. Japanese Journal of Physiology, 11, 71–79. Hukuhara, T., Nakayama, S. and Sumi, T. (1959). The role of the intrinsic mucosal reflex in the fluid transport through the denervated intestinal loop. Japanese Journal of Physiology, 9, 406–418. Hukuhara, T., Neya, T. and Tsuchiya, K. (1969). The effect of the intrinsic mucosal reflex upon the propagation of intestinal contractions. Japanese Journal of Physiology, 19, 824–833.

ENTERIC REFLEXES THAT INFLUENCE MOTILITY

49

Hukuhara, T., Sumi, T. and Kotani, S. (1961). Comparative studies on the intestinal intrinsic reflexes in rabbits, guinea pigs and dogs. Japanese Journal of Physiology, 11, 205–211. Hukuhara, T., Yamagami, M. and Nakayama, S. (1958). On the intestinal intrinsic reflexes. Japanese Journal of Physiology, 8, 9–20. Hukuhara, T., Nakayama, S., Yamagami, M. and Miyake, T. (1959). On the intestinal extrinsic reflexes elicited from the small intestine. Acta Medica Okayama, 13, 113–121. Iyer, V., Bornstein, J.C., Costa, M., Furness, J.B., Takahashi, Y. and Iwanaga, T. (1988). Electrophysiology of guinea-pig myenteric neurons correlated with immunoreactivity for calcium binding proteins. Journal of the Autonomic Nervous System, 22, 141–150. Johnson, P.J. and Bornstein, J.C. (2002). Evidence that tachykinins mediate slow synaptic transmission to myenteric neurons in the guinea-pig ileum. Neuroscience, submitted. Johnson, S.M., Katayama, Y. and North, R.A. (1980). Slow synaptic potentials in neurones of the myenteric plexus. Journal of Physiology, 301, 505–516. Johnson, S.M., Katayama, Y., Morita, K. and North, R.A. (1981). Mediators of slow synaptic potentials in the myenteric plexus of the guinea-pig ileum. Journal of Physiology, 320, 175–186. Johnson, P.J., Yuan, S.Y., Bornstein, J.C. and Furness, J.B. (1996). Analysis of contributions of acetylcholine and tachykinins to neuro-neuronal transmission in motility reflexes in the guinea-pig ileum. British Journal of Pharmacology, 118, 973–983. Johnson, P.J., Bornstein, J.C. and Burcher, E. (1998). Roles of neuronal NK1 and NK3 receptors in synaptic transmission during motility reflexes in the guinea-pig ileum. British Journal of Pharmacology, 124, 1375–1384. Johnson, P.J., Shum, O.R., Thornton, P.D.J. and Bornstein, J.C. (1999). Evidence that inhibitory motor neurons of the guinea-pig small intestine exhibit fast excitatory synaptic potentials mediated via P2X receptors. Neuroscience Letters, 266, 169–172. Kadowaki, M., Wade, P.R. and Gershon, M.D. (1996). Participation of 5-HT3, 5-HT4, and nicotinic receptors in the peristaltic reflex of guinea pig distal colon. American Journal of Physiology, 271, G849–G857. Katayama, Y. and North, R.A. (1978). Does substance P mediate slow synaptic excitation within the myenteric plexus? Nature, 274, 387–388. Katayama, Y. and North, R.A. (1980). The action of somatostatin on neurones of the myenteric plexus of the guinea-pig ileum. Journal of Physiology, 303, 315–323. Katayama, Y., Lees, G.M. and Pearson, G.T. (1986). Electrophysiological and morphological characteristics of vasoactive intestinal peptide (VIP)-immunoreactive neurones in the guinea-pig myenteric plexus. Journal of Physiology, 378, 1–11. Katayama, Y., North, R.A. and Williams, J.T. (1979). The action of substance P on neurons of the myenteric plexus of the guinea-pig small intestine. Proceeding of the Royal Society of London, Series B, 206, 191–208. Keast, J.R., Furness, J.B. and Costa, M. (1984). The origins of peptide and norepinephrine nerve terminals in the mucosa of the guinea-pig small intestine. Gastroenterology, 86, 637–644. Kirchgessner, A.L. and Gershon, M.D. (1988). Projections of submucosal neurons to the myenteric plexus of the guinea pig intestine: in vitro tracing of microcircuits by retrograde and anterograde transport. Journal of Comparative Neurology, 277, 487–498. Kirchgessner, A.L., Liu, M.-T. and Gershon, M.D. (1996). In situ identification and visualization of neurons that mediate enteric and enteropancreatic reflexes. Journal of Comparative Neurology, 371, 270–286. Kirchgessner, A.L., Tamir, H. and Gershon, M.D. (1992). Identification and stimulation by serotonin of intrinsic sensory neurons of the submucosal plexus of the guinea-pig gut: activity-induced expression of Fos immunoreactivity. Journal of Neuroscience, 12, 235–248. Kosterlitz, H.W. and Lydon, R.J. (1968). Spontaneous electrical activity and nerve-mediated inhibition in the innervated longitudinal muscle strip of the guinea-pig ileum. Journal of Physiology, 200, 126P-128P. Kosterlitz, H.W. and Robinson, J.A. (1959). Reflex contractions of the longitudinal muscle coat of the isolated guinea-pig ileum. Journal of Physiology, 146, 369–379. Kottegoda, S.R. (1969). An analysis of possible nervous mechanisms involved in the peristaltic reflex. Journal of Physiology, 200, 687–712. Kunze, W.A.A., Bornstein, J.C. and Furness, J.B. (1995). Physiological identification of sensory nerve cells in a peripheral organ (the intestine) of a mammal. Neuroscience, 66, 1–4. Kunze, W.A.A., Furness, J.B. and Bornstein, J.C. (1993). Simultaneous intracellular recordings from enteric neurons reveals that myenteric AH neurons transmit via slow excitatory post-synaptic potentials. Neuroscience, 55, 685–694. Kunze, W.A.A., Bornstein, J.C., Furness, J.B., Hendriks, R. and Stephenson, D.S.H. (1994). Charybdotoxin and iberiotoxin, but not apamin, abolish the slow after-hyperpolarization in myenteric plexus neurons. Pflügers Archiv, 428, 300–306.

50

INNERVATION OF THE GASTROINTESTINAL TRACT

Kunze, W.A.A., Bertrand, P.P., Furness, J.B. and Bornstein, J.C. (1997). Investigation of a neural mechanism by which connection with the mucosa enhances the excitability of myenteric neurons. Neuroscience, 76, 619–634. Kunze, W.A.A., Furness, J.B., Bertrand, P.P. and Bornstein, J.C. (1998). Intracellular recording from myenteric neurons of the guinea-pig ileum that respond to stretch. Journal of Physiology, 506, 827–842. Kunze, W.A.A., Clerc, N., Bertrand, P.P. and Furness, J.B. (1999). Contractile activity in intestinal muscle evokes action potential discharge in guinea-pig myenteric neurons. Journal of Physiology, 517, 547–561. Langley, J.N. and Magnus, R. (1905). Some observations of the movements of the intestine before and after degenerative section of the mesenteric nerves. Journal of Physiology, 33, 34–51. LePard, K.J. and Galligan, J.J. (1999). Analysis of fast synaptic pathways in myenteric plexus of guinea-pig ileum. American Journal of Physiology, 276, G529–G538. LePard, K.J., Messori, E. and Galligan, J.J. (1997). Purinergic fast excitatory postsynaptic potentials in myenteric neurons of guinea pig: distribution and pharmacology. Gastroenterology, 113, 1522–1534. Li, Z.S. and Furness, J.B. (1998). Immunohistochemical localisation of cholinergic markers in putative intrinsic primary afferent neurons of the guinea-pig small intestine. Cell and Tissue Research, 294, 35–43. Lin, H.C., Doty, J.E., Reedy, T.J. and Meyer, J.H. (1990a). Inhibition of gastric emptying by acids depends on pH, titratable acidity, and length of intestine exposed to acid. American Journal of Physiology, 259, G1025–G1030. Lin, H.C., Doty, J.E., Reedy, T.J. and Meyer, J.H. (1990b). Inhibition of gastric emptying by sodium oleate depends on length of intestine exposed to nutrient. American Journal of Physiology, 259, G1031–G1036. Liu, M.-T., Rothstein, J.D., Gershon, M.D. and Kirchgessner, A.L. (1997). Glutamatergic enteric neurons. Journal of Neuroscience, 17, 4764–4784. Lomax, A.E.G., Sharkey, K.S., Bertrand, P.P., Low, A.W., Bornstein, J.C. and Furness, J.B. (1999). Correlation of morphology, electrophysiology and chemistry of neurons in the myenteric plexus of the guinea-pig distal colon. Journal of the Autonomic Nervous System, 76, 45–61. Maggi, C.A., Patacchini, R., Barthó, L., Holzer, P. and Santicioli, P. (1994). Tachykinin NK1 and NK2 receptor antagonists and atropine-resistant ascending excitatory reflex to the circular muscle of the guinea-pig ileum. British Journal of Pharmacology, 112, 161–168. Magnus, R. (1904). Versuche am uberlebenden Dunndarm von Saugeticren II. Mitteil: Die Beziehungen des Darmnervensystems zur automatischen Darmbewegung. Archiv für die gesamte Physiologie, 102, 349–363. Mall, F. (1896). A study of the intestinal contraction. Johns Hopkins Hospital Report, 1, 37–75. Mann, P.T., Southwell, B.R., Young, H.M. and Furness, J.B. (1997). Appositions made by axons of descending interneurons in the guinea-pig small intestine, investigated by confocal microscopy. Journal of Chemical Neuroanatomy, 12, 151–164. Mawe, G.M. (1995). Tachykinins as mediators of slow EPSPs in guinea-pig gall-bladder ganglia: involvement of neurokinin-3 receptors. Journal of Physiology, 485, 513–524. Mawe, G.M., Branchek, T.A. and Gershon, M.D. (1989). Blockade of 5-HT-mediated enteric slow EPSPs by BRL 24924: gastrokinetic effects. American Journal of Physiology, 257, G386–G396. Messenger, J.P. and Furness, J.B. (1990). Projections of chemically-specified neurons in the guinea-pig colon. Archives of Histology and Cytology, 53, 467–495. Messenger, J.P., Bornstein, J.C. and Furness, J.B. (1994). Electrophysiological and morphological classification of myenteric neurons in the proximal colon of the guinea-pig. Neuroscience, 60, 227–224. Morita, K. and Katayama, Y. (1992). Substance P inhibits activation of calcium-dependent potassium conductances in guinea-pig myenteric neurones. Journal of Physiology, 447, 293–308. Morita, K. and North, R.A. (1985). Significance of slow synaptic potentials for transmission of excitation in guinea-pig myenteric plexus. Neuroscience, 14, 661–672. Morita, K., North, R.A. and Katayama, Y. (1980). Evidence that substance P is a neurotransmitter in the myenteric plexus. Nature, 287, 151–152. Morita, K., North, R.A. and Tokimasa, T. (1982a). Muscarinic agonists inactivate potassium conductance of guinea-pig myenteric neurones. Journal of Physiology, 333, 125–139. Morita, K., North, R.A. and Tokimasa, T. (1982b). Muscarinic presynaptic inhibition of synaptic transmission in myenteric plexus of guinea-pig ileum. Journal of Physiology, 333, 141–149. Neunlist, M., Dobreva, G. and Schemann, M. (1999). Characteristics of mucosally projecting myenteric neurones in the guinea-pig proximal colon. Journal of Physiology, 517, 533–546. Neya, T., Mizutani, M. and Yamasato, T. (1993). Role of 5-HT3 receptors in peristaltic reflex elicited by stroking the mucosa in the canine jejunum. Journal of Physiology, 471, 159–173. Niel, J.P., Bywater, R.A.R. and Taylor, G.S. (1983). Apamin-resistant post-stimulus hyperpolarization in the circular muscle of the guinea-pig ileum. Journal of the Autonomic Nervous System, 9, 565–569. Nishi, S. and North, R.A. (1973). Intracellular recording from the myenteric plexus of the guinea-pig ileum. Journal of Physiology, 231, 471–491.

ENTERIC REFLEXES THAT INFLUENCE MOTILITY

51

North, R.A. and Tokimasa, T. (1982). Muscarinic synaptic potentials in guinea-pig myenteric plexus neurones. Journal of Physiology, 333, 151–156. North, R.A. and Tokimasa, T. (1983). Depression of calcium-dependent potassium conductance of guinea-pig myenteric neurones by muscarinic agonists. Journal of Physiology, 342, 253–266. North, R.A., Slack, B.E. and Surprenant, A. (1985). Muscarinic M1 and M2 receptors mediate depolarization and presynaptic inhibition in guinea-pig enteric nervous system. Journal of Physiology, 368, 435–452. Nothnagel, H. (1882). Zur chemischen Reizung der glatten Muskeln; zugleich als Beitrag zur Physiologie des Darmes. Archiv die Pathologie Anatomie und Physiologie, 88, 1–11. Noyer, M., Bui, N.D., Deschodt-Lanckman, M., Robberecht, P., Woussen, M.-C. and Christophe, J. (1980). Postnatal development of the cholecystokinin-gastrin family peptides in the brain and gut of rat. Life Sciences, 27, 2197–2203. Onori, L., Aggio, A., Taddei, G. and Tonini, M. (2000). Contribution of NK2 tachykinin receptors to propulsion in the rabbit distal colon. American Journal of Physiology, 278, G137–G147. Orloff, S.L., Bunnett, N.W., Walsh, J.H. and Debas, H.T. (1992). Intestinal acid inhibits gastric acid secretion by neural and hormonal mechanisms in rats. American Journal of Physiology, 262, G165–G170. Osthaus, L.E. and Galligan, J.J. (1992). Antagonists of nitric oxide synthesis inhibit nerve-mediated relaxations of longitudinal muscle in guinea pig ileum. Journal of Pharmacology and Experimental Therapeutics, 260, 140–145. Otterson, M.F. and Sarna, S.K. (1994). Neural control of small intestinal giant migrating contractions. American Journal of Physiology, 266, G576–G584. Pedroarena, C.M., Pose, I.E., Yamuy, J., Chase, M.H. and Morales, F.R. (1999). Oscillatory membrane potential activity in the soma of a primary afferent neuron. Journal of Neurophysiology, 82, 1465–1476. Pompolo, S. and Furness, J.B. (1988). Ultrastructure and synaptic relationships of calbindin-reactive, Dogiel type II neurons, in myenteric ganglia of guinea-pig small intestine. Journal of Neurocytology, 17, 771–782. Pompolo, S. and Furness, J.B. (1990). Ultrastructure and synaptology of neurons immunoreactive for gammaaminobutyric acid in the myenteric plexus of the guinea pig small intestine. Journal of Neurocytology, 19, 539–549. Pompolo, S. and Furness, J.B. (1993). Origins of synaptic inputs to calretinin immunoreactive neurons in the guinea-pig small intestine. Journal of Neurocytology, 22, 531–546. Pompolo, S. and Furness, J.B. (1995). Sources of inputs to longitudinal muscle motor neurons and ascending interneurons in the guinea-pig small intestine. Cell and Tissue Research, 280, 549–560. Pompolo, S. and Furness, J.B. (1998). Quantitative analysis of inputs to somatostatin-immunoreactive descending interneurons in the myenteric plexus of the guinea-pig small intestine. Cell and Tissue Research, 294, 219–226. Pompolo, S., Furness, J.B., Bornstein, J.C., Hendriks, R. and Trussell, D.C. (1989). Dogiel type II neurons in the guinea-pig small intestine: ultrastructure in relation to other characteristics. In Nerves in the Gastrointestinal Tract, edited by M.V. Singer and H. Goebell, pp. 57–67. Carnforth, U.K.: Martin Lister. Portbury, A.L., Pompolo, S., Furness, J.B., Stebbing, M.J., Kunze, W.A.A., Bornstein, J.C. et al. (1995). Cholinergic, somatostatin-immunoreactive, interneurons in the guinea-pig intestine: morphology, ultrastructure, connections and projections. Journal of Anatomy, 187, 303–321. Rehfeld, J.F. (1978). Immunochemical studies on cholecystokinin. II. Distribution and molecular heterogeneity in the central nervous system and small intestine of man and hog. Journal of Biological Chemistry, 253, 4022–4030. Ren, J., Hu, H.-Z., Liu, S., Xia, Y. and Wood, J.D. (2000). Glutamate receptors in the enteric nervous system: ionotropic or metabotropic? Neurogastroenterology and Motility, 12, 257–264. Richardson, A., Delbridge, A.T., Brown, N.J., Rumsey, R.D. and Read, N.W. (1991). Short chain fatty acids in the terminal ileum accelerate stomach to caecum transit time in the rat. Gut, 32, 266–269. Ritter, R.C., Costa, M. and Brookes, S.J.H. (1997). Nuclear Fos immunoreactivity in guinea pig myenteric neurons following activation of motor activity. American Journal of Physiology, 273, G498–G507. Rodriguez-Membrilla, A., Martinez, V. and Vergara, P. (1995). Peripheral and central cholecystokinin receptors regulate postprandial intestinal motility in the rat. Journal of Pharmacology and Experimental Therapeutics, 275, 486–493. Sarna, S.K. (1991a). Physiology and pathophysiology of colonic motor activity. (Part one of two) Digestive Diseases and Science, 36, 827–862. Sarna, S.K. (1991b). Physiology and pathophysiology of colonic motor activity. (Part two of two) Digestive Diseases and Science, 36, 998–1018. Sarna, S., Stoddard, C., Belbeck, L. and McWade, D. (1981). Intrinsic nervous control of migrating myoelectric complexes. American Journal of Physiology, 241, G16–G23. Sarna, S.K., Prasad, K.R. and Lang, I.M. (1988). Giant migrating contractions of the canine caecum. American Journal of Physiology, 254, 595–601.

52

INNERVATION OF THE GASTROINTESTINAL TRACT

Schemann, M. and Ehrlein, H.J. (1982). The utility of cellulose meals for studies on gastrointestinal motility in dogs. Digestion, 25, 194–196. Schemann, M. and Ehrlein, H.J. (1986a). 5-Hydroxytryptophan and cisapride stimulate propulsive jejunal motility and transit of chyme in dogs. Digestion, 34, 229–235. Schemann, M. and Ehrlein, H.J. (1986b). Effects of neurohormonal agents on jejunal contraction spread and transit in the fed dog. Gastroenterology, 90, 1950–1955. Schemann, M. and Ehrlein, H.J. (1986c). Postprandial patterns of canine jejunal motility and transit of luminal content. Gastroenterology, 90, 991–1000. Schemann, M. and Kayser, H. (1991). Effects of tachykinins on myenteric neurones of the guinea-pig gastric corpus: involvement of NK-3 receptors. Pflügers Archiv, 419, 566–571. Schemann, M. and Wood, J.D. (1989). Electrical behaviour of myenteric neurones in the gastric corpus of the guinea-pig. Journal of Physiology, 417, 501–518. Scheuermann, D.W. and Stach, W. (1983). External adrenergic innervation of the three neuron types of Dogiel in the plexus myentericus and the plexus submucosus externus of the porcine small intestine. Histochemistry, 77, 303–311. Scheurer, U., Varga, L., Drack, E., Burki, H.-R. and Halter, F. (1983). Measurement of cholecystokinin octapeptide-induced motility of rat antrum, pylorus, and duodenum in vitro. American Journal of Physiology, 244, 261–265. Schmid, H.R. and Ehrlein, H.J. (1993). Effects of enteral infusion of hypertonic saline and nutrients on canine jejunal motor patterns. Digestive Diseases and Science, 38, 1062–1072. Schulze-Delrieu, K. (1999). Visual parameters define the phase and the load of contractions in isolated guinea pig ileum. American Journal of Physiology, 276, G1417–G1424. Schutte, I.W.M., Kroese, A.B.A. and Akkermans, L.M.A. (1995). Somal size and location within the ganglia for electrophysiologically identified myenteric neurons of the guinea pig ileum. Journal of Comparative Neurology, 355, 563–572. Schutte, I.W.M., Hollestein, K.B.C.W., Akkermans, L.M.A. and Kroese, A.B.A. (1997). Evidence for a role of cholecystokinin as neurotransmitter in the guinea-pig enteric nervous system. Neuroscience Letters, 236, 155–158. Sethi, A.K. and Sarna, S.K. (1995). Contractile mechanisms of canine colonic propulsion. American Journal of Physiology, 268, G530–G538. Siegle, M.L. and Ehrlein, H.J. (1988). Digestive motor patterns and transit of luminal contents in canine ileum. American Journal of Physiology, 254, G552–G559. Siegle, M.-L., Schmid, H.-R. and Ehrlein, H.J. (1990). Effects of ileal infusions of nutrients on motor patterns of canine small intestine. American Journal of Physiology, 78–85. Smith, T.K. and Furness, J.B. (1988). Reflex changes in circular muscle activity elicited by stroking the mucosa: An electrophysiological analysis in the isolated guinea-pig. Journal of the Autonomic Nervous System, 25, 205–218. Smith, T.K. and McCarron, S.L. (1998). Nitric oxide modulates cholinergic reflex pathways to the longitudinal and circular muscle in the isolated guinea-pig distal colon. Journal of Physiology, 512, 893–906. Smith, T.K., Furness, J.B., Costa, M. and Bornstein, J.C. (1988). An electrophysiological study of the projections of motor neurones that mediate non-cholinergic excitation in the circular muscle of the guinea-pig small intestine. Journal of the Autonomic Nervous System, 22, 115–128. Smith, T.K., Bornstein, J.C. and Furness, J.B. (1990). Distension-evoked ascending and descending reflexes in the circular muscle of guinea-pig ileum: an intracellular study. Journal of the Autonomic Nervous System, 29, 203–213. Smith, T.K., Bornstein, J.C. and Furness, J.B. (1991). Interactions between reflexes evoked by distension and mucosal stimulation: electrophysiological studies of guinea-pig ileum. Journal of the Autonomic Nervous System, 34, 69–76. Smith, T.K., Bornstein, J.C. and Furness, J.B. (1992). Convergence of reflex pathways excited by distension and mechanically deforming the mucosa onto morphologically characterized myenteric neurons of the guineapig small intestine. Journal of Neuroscience, 12, 1502–1510. Smith, T.K., Bywater, R.A.R., Taylor, G.S. and Holman, M.E. (1992). Electrical responses of the muscularis externa to distension of the isolated guinea-pig distal colon. Journal of Gastrointestinal Motility, 4, 145–156. Smith, T.K., Burke, E.P. and Shuttleworth, C.W. (1999). Topographical and electrophysiological characteristics of highly excitable S neurones in the myenteric plexus of the guinea-pig ileum. Journal of Physiology, 517, 817–830. Smith, T.K. and Robertson, W.J. (1998). Synchronous movements of the longitudinal and circular muscle during peristalsis in the isolated guinea-pig distal colon. Journal of Physiology, 506, 563–577. Song, Z.-M., Brookes, S.J.H. and Costa, M. (1991). Identification of myenteric neurons which project to the mucosa of the guinea-pig small intestine. Neuroscience Letters, 129, 294–298.

ENTERIC REFLEXES THAT INFLUENCE MOTILITY

53

Song, Z.-M., Brookes, S.J.H. and Costa, M. (1994). All calbindin-immunoreactive myenteric neurons project to the mucosa of the guinea-pig small intestine. Neuroscience Letters, 180, 219–222. Song, Z.-M., Brookes, S.J.H., Ramsay, G.A. and Costa, M. (1997). Characterization of myenteric interneurons with somatostatin immunoreactivity in the guinea-pig small intestine. Neuroscience, 80, 907–923. Southwell, B.R., Woodman, H.L., Royal, S.J. and Furness, J.B. (1998). Movement of villi induces endocytosis of NK1 receptors in myenteric neurons from guinea-pig ileum. Cell and Tissue Research, 292, 37–45. Spencer, M.P., Sarr, M.G., Soper, N.J. and Hakim, N.S. (1990). Jejunal regulation of gastric motility patterns: Effect of extrinsic neural continuity to stomach. American Journal of Physiology, 258, G32–G37. Spencer, N., McCarron, S.L. and Smith, T.K. (1999). Sympathetic inhibition of ascending and descending interneurones during the peristaltic reflex in the isolated guinea-pig distal colon. Journal of Physiology, 519, 539–550. Spencer, N., Walsh, M. and Smith, T.K. (1999). Does the guinea-pig ileum obey the ‘law of the intestine’? Journal of Physiology, 517, 889–898. Spencer, N.J., Walsh, M. and Smith, T.K. (2000). Purinergic and cholinergic neuro-neuronal transmission underlying reflexes activated by mucosal stimulation in the isolated guinea-pig ileum. Journal of Physiology, 522, 321–331. Stach, W. (1979). Differentiated vascularization of Dogiel’s cell types and the preferred vascularization of type I/2 cells within plexus myentericus (Auerbach) ganglia of the pig. Anaatomischer Anzeiger, 145, 464–473. Stach, W. (1989). A revised morphological classification of neurons in the enteric nervous system. In Nerves and the Gastrointestinal Tract, edited by M.V. Singer and H. Goebell, pp. 29–45. Lancaster, UK: Kluwer Academic Publishers. Stebbing, M.J. (1996). Investigations on the neural control of the small intestine of the guinea-pig. PhD Thesis, University of Melbourne, Melbourne, Australia. Stebbing, M.J. and Bornstein, J.C. (1996). Electrophysiological mapping of fast synaptic inputs to morphologically and chemically characterised myenteric neurons of guinea-pig small intestine. Neuroscience, 73, 1017–1028. Steele, P.A., Brookes, S.J.H. and Costa, M. (1991). Immunohistochemical identification of cholinergic neurons in the myenteric plexus of guinea-pig small intestine. Neuroscience, 45, 227–239. Stewart, J.J. and Bass, P. (1976a). Effect of intravenous C-terminal octapeptide of cholecystokinin and intraduodenal ricinoleic acid on contractile activity of the dog intestine. Proceedings of the Society for Experimental Biology and Medicine, 152, 213–217. Stewart, J.J. and Bass, P. (1976b). Effects of ricinoleic and oleic acids on the digestive contractile activity of the canine small and large bowel. Gastroenterology, 70, 371–376. Szurszewski, J.H. and King, B.F. (1989). Physiology of prevertebral ganglia in mammals with special reference to inferior mesenteric ganglion. In Handbook of Physiology, 6: The Gastrointestinal System, volume I. Motility and Circulation, edited by J.D. Wood, pp. 519–592. Washington, DC: American Physiological Society. Tack, J.F. and Wood, J.D. (1992). Electrical behaviour of myenteric neurones in the gastric antrum of the guineapig. Journal of Physiology, 447, 49–66. Tack, J.F., Janssens, J., Vantrappen, G. and Wood, J.D. (1992). Actions of 5-hydroxytryptamine on myenteric neurons in guinea pig gastric antrum. American Journal of Physiology, 263, G838–G846. Takaki, M., Branchek, T., Tamir, H. and Gershon, M.D. (1985). Specific antagonism of enteric neural serotonin receptors by dipeptides of 5-hydroxytryptophan: Evidence that serotonin is a mediator of slow synaptic excitation in the myenteric plexus. Journal of Neuroscience, 5, 1769–1780. Tamura, K. (1992). Morphology of electrophysiologically identified myenteric neurons in the guinea pig rectum. American Journal of Physiology, 262, G545–G552. Tamura, K., Schemann, M. and Wood, J.D. (1993). Actions of nitric oxide-generating sodium nitroprusside in myenteric plexus of guinea pig small intestine. American Journal of Physiology, 265, G887–G893. Tamura, K. and Wood, J.D. (1989). Electrical and synaptic properties of myenteric plexus neurones in the terminal large intestine of the guinea-pig. Journal of Physiology, 415, 275–298. Thomas, E.A., Bertrand, P.P. and Bornstein, J.C. (1999). Genesis and role of coordinated firing in a feed forward network – a model study. Neuroscience, 93, 1525–1537. Thomas, J.E. and Baldwin, M.V. (1971). The intestinal mucosal reflex in the unanesthetized dog. American Journal of Digestive Diseases, 16, 642–647. Tokimasa, T., Cherubini, E. and North, R.A. (1983). Nicotinic depolarization activates calcium dependent gK in myenteric neurons. Brain Research, 263, 57–62. Tonini, M. and Costa, M. (1990). A pharmacological analysis of the neuronal circuitry involved in distensionevoked enteric excitatory reflex. Neuroscience, 38, 787–795. Tonini, M., Frigo, G., Lecchini, S., D’Angelo, L. and Crema, A. (1981). Hyoscine-resistant peristalsis in guineapig ileum. European Journal of Pharmacology, 71, 375–381.

54

INNERVATION OF THE GASTROINTESTINAL TRACT

Tonini, M., Crema, A., Frigo, G.M., Rizzi, C.A., Manzo, L., Candura, S.M. et al. (1989). An in vitro study of the relationship between GABA receptor function and propulsive motility in the distal colon of the rabbit. British Journal of Pharmacology, 98, 1109–1118. Tonini, M., Costa, M., Brookes, S.J.H. and Humphreys, C.M.S. (1996). Dissociation of the ascending excitatory reflex from peristalsis in the guinea-pig small intestine. Neuroscience, 73, 287–297. Trendelenburg, P. (1917). Physiologische und pharmakologische Versuche uber die Dunndarmperistaltik. Archiv für experimentelle Pathologie und Pharmakologie, 81, 55–129. Tuladar, B.R., Kaisar, M. and Naylor, R.J. (1997). Evidence for a 5-HT3 receptor involvement in the facilitation of peristalsis on mucosal application of 5-HT in the guinea-pig isolated ileum. British Journal of Pharmacology, 122, 1174–1178. Uemura, S., Pompolo, S. and Furness, J.B. (1995). Colocalization of neuropeptide Y with other neurochemical markers in the guinea-pig small intestine. Archives of Histology and Cytology, 58, 523–536. Wade, P.R. and Wood, J.D. (1988). Electrical behavior of myenteric neurones in guinea-pig distal colon. American Journal of Physiology, 254, G522–G530. Wade, P.R., Tamir, H., Kirchgessner, A.L. and Gershon, M.D. (1994). Analysis of the role of 5-HT in the enteric nervous system using anti-idiotopic antibodies to 5-HT receptors. American Journal of Physiology, 266, G403–G416. Waterman, S.A. and Costa, M. (1994). The role of enteric inhibitory motoneurons in peristalsis in the isolated guinea-pig small intestine. Journal of Physiology, 477, 459–468. Waterman, S.A., Costa, M. and Tonini, M. (1992). Modulation of peristalsis in the guinea-pig isolated small intestine by exogenous and endogenous opioids. British Journal of Pharmacology, 106, 1004–1010. Waterman, S.A., Costa, M. and Tonini, M. (1994a). Accomodation mediated by enteric inhibitory reflexes in the isolated guinea-pig small intestine. Journal of Physiology, 474, 539–546. Waterman, S.A., Tonini, M. and Costa, M. (1994b). The role of ascending excitatory and descending inhibitory pathways in peristalsis in the isolated guinea-pig small intestine. Journal of Physiology, 481, 223–232. Williams, J.A. (1982). Cholecystokinin: A hormone and a neurotransmitter. Biomedical Research, 3, 107–121. Williams, J.T. and North, R.A. (1979). Vasoactive intestinal polypeptide excites neurones of the myenteric plexus. Brain Research, 175, 174–177. Williamson, S., Pompolo, S. and Furness, J.B. (1996). GABA and nitric oxide synthase immunoreactivities are colocalized in a subset of inhibitory motor neurons of the guinea-pig small intestine. Cell and Tissue Research, 284, 29–37. Wood, J.D. (1975). Neurophysiology of Auerbach’s plexus and control of intestinal motility. Physiological Reviews, 55, 307–324. Wood, J.D. and Mayer, C.J. (1978a). Intracellular study of electrical activity of Auerbach’s plexus in guinea-pig small intestine. Pflügers Archiv, 374, 265–275. Wood, J.D. and Mayer, C.J. (1978b). Slow synaptic excitation mediated by serotonin in Auerbach’s plexus. Nature, 276, 836–837. Wood, J.D. and Mayer, C.J. (1979). Serotonergic activation of tonic-type enteric neurons in guinea pig small bowel. Journal of Neurophysiology, 42, 582–593. Young, H.M. and Furness, J.B. (1995). Ultrastructural examination of the targets of serotonin-immunoreactive descending interneurons in the guinea pig small intestine. Journal of Comparative Neurology, 356, 101–114. Young, H.M., Furness, J.B. and Povey, J.M. (1995). Analysis of connections between nitric oxide synthase neurons in the myenteric plexus of the guinea-pig small intestine. Journal of Neurocytology, 24, 257–263. Yuan, S.Y., Bornstein, J.C. and Furness, J.B. (1994). Investigation of the role of 5-HT3 and 5-HT4 receptors in ascending and descending reflexes to the circular muscle of guinea-pig small intestine. British Journal of Pharmacology, 112, 1095–1100. Yuan, S.Y., Bornstein, J.C. and Furness, J.B. (1995). Pharmacological evidence that nitric oxide may be a retrograde messenger in the enteric nervous system. British Journal of Pharmacology, 114, 428–432. Yuan, S.Y., Brookes, S.J.H. and Costa, M. (1997). Distension-evoked ascending and descending reflexes in the isolated guinea-pig stomach. Journal of the Autonomic Nervous System, 62, 94–102. Yuan, S.Y., Furness, J.B. and Bornstein, J.C. (1992). Post-stimulus depression of reflex changes in circular muscle activity in the guinea-pig small intestine. Journal of the Autonomic Nervous System, 40, 171–180. Yuan, S.Y., Furness, J.B., Bornstein, J.C. and Smith, T.K. (1991). Mucosal compression elicits polarized reflexes and enhances responses of the circular muscle to distension in the small intestine. Journal of the Autonomic Nervous System, 35, 219–226. Zafirov, D.H., Palmer, J.M., Nemeth, P.R. and Wood, J.D. (1985). Bombesin, gastrin releasing peptide and vasoactive intestinal peptide excite myenteric neurons. European Journal of Pharmacology, 115, 103–107.

ENTERIC REFLEXES THAT INFLUENCE MOTILITY

55

Zhou, X. and Galligan, J.J. (1996). P2X purinoceptors in cultured myenteric neurons of guinea-pig small intestine. Journal of Physiology, 496, 719–729. Zhou, X. and Galligan, J.J. (1998). Non-additive interaction between nicotinic cholinergic and P2X purine receptors in guinea-pig enteric neurons in culture. Journal of Physiology, 513, 685–697. Zhou, X. and Galligan, J.J. (1999). Synaptic activation and properties of 5-hydroxytryptamine3 receptors in myenteric neurons of guinea pig intestine. Journal of Pharmacology and Experimental Therapeutics, 290, 803–810.

2 Motor Control of the Stomach David Grundy1 and Michael Schemann2 1

Department of Biomedical Science, The University of Sheffield, The Alfred Denny Building, Western Bank, Sheffield S10 2TN, UK 2 Department of Physiology, School of Veterinary Medicine, Bischofsholer Damm 15/102, D-30173 Hannover, Germany The complex repertoire of gastric motor function is controlled and coordinated by reflexes that match gastric tone and peristaltic activity to the digestive needs of the individual. These reflexes operate both at the level of the enteric nervous system and through autonomic connections to the brain and spinal cord, with excitatory and inhibitory enteric motor neurones providing the final pathway for both intrinsic and extrinsic reflexes. The electrophysiological properties, neurochemical coding and projection patterns of these enteric neurones have been well characterised enabling the detailed circuit diagrams for contraction and relaxation of the corpus and antrum to be described. Extrinsic influences, from the vagus and splanchnic nerves, have a widespread action on enteric neural transmission, despite the relative paucity of extrinsic to intrinsic neurones. The stomach has a rich sensory innervation and the properties of vagal and spinal afferents have been well characterised. In contrast, the nature of the intrinsic primary afferent neurone has not been resolved despite abundant functional data for their existence. A concept for permissive control is presented, whereby autonomic inputs to enteric circuits set the “gain” for enteric reflexes. Plasticity in these circuits accounts for the adaptation of gastric function that occurs when the extrinsic innervations is compromised by surgery or disease. KEY WORDS: enteric nervous system; vagus; splanchnic nerves; stomach; neurochemistry; smooth muscle; gastric reflexes.

INTRODUCTION THE GASTRIC REFLEX REPERTOIRE Neural reflexes, together with endocrine and paracrine controls, serve to coordinate and control gastric motor and secretory function. Gastric motor function can be divided into two distinct activities which can be differentiated on a regional basis. The proximal stomach consisting of corpus and fundus serves as an important reservoir and has a remarkable capacity to accommodate large meals for subsequent delivery to the digestive and absorptive machinery of the gastrointestinal tract. The smooth muscle cells in this region generate myogenic tone and as such can be actively inhibited to bring about gastric 57

58

INNERVATION OF THE GASTROINTESTINAL TRACT

relaxation (see Szurszewski, 1987). This occurs during deglutition by a process known as receptive relaxation (Cannon and Lieb, 1911). Gastric relaxation also occurs during gastric filling and was recognised by Kelling (1903) as a means of adapting gastric capacity to increasing contents by maintaining the pressure within very narrow limits, hence the term adaptive relaxation. Relaxation also occurs as a prelude to vomiting when the stomach dilates to receive chyme being returned from the intestines in readiness for expulsion (see Grundy and Reid, 1994). In all three cases the relaxation is mediated by the parasympathetic innervation via the vagus nerve and, in particular for the adaptive relaxation, by the enteric nervous system (ENS). Gastric tone is also influenced by the sympathetic innervation especially during the widespread inhibition of gut motor function that occurs during “adynamic ileus” following surgical manipulation of the gut. However, the vagus nerve may also play a role in this response illustrating the importance of reflex interactions between vagal and splanchnic pathways to the gut (Abrahamsson, Glise and Glise, 1979). The antrum, in contrast, functions as a peristaltic pump to mix, grind and propel gastric contents through the pylorus into the duodenum. These phasic, rather than tonic, contractions originate in a pacemaker region in the middle or orad corpus with the dominant site of origin on the greater curve (Alvarez and Mahoney, 1922; Kelly and Code, 1971). The contractions propagate aborally towards the pylorus and in the fasting state periodically serve to clear the stomach of accumulated secretions, cell debris and undigested residue as part of the migrating motor complex (Code and Martlett, 1975). However, during gastric emptying of a meal, the corpus and antrum, together with the pylorus and duodenum, must act in a coordinated fashion to optimise delivery of chyme to the small intestine. Reflexes between corpus and antrum are responsible for the fine tuning of this coordinated activity and, when disrupted, the rate of gastric emptying can be excessively fast (post-vagotomy dumping syndrome) or slow (diabetic gastroparesis) leading to a range of symptoms including bloating, nausea and pain (see Meyer, 1994). However, under normal circumstances gastric reflexes serve to coordinate the corpus and antrum, while feedback from the small intestine concerning nutrient delivery, osmolarity and pH tend to inhibit the gastric drive to emptying and so can match the rate of gastric emptying to the capacity of the intestine to digest and assimilate nutrients. THE NEURAL BASIS OF REFLEX CONTROL The motor programmes for controlling gastric tone and peristaltic contractions rely on feedback from sensory receptors positioned both within and outside the gastric wall. These monitor the prevailing conditions and trigger the appropriate response in the motor supply to the gastric smooth muscle. This reflex activity is organised at a number of different levels, all of which interact in an integrative way (Figure 2.1). At the local level is the enteric nervous system which Langley (1921) recognised as containing all the elements of an independent nervous system. Reflexes are triggered by local sensory inputs with the integrative ability of the ENS allowing the assimilation of information from the muscle and mucosa to determine the appropriate motor response (Wood, 1994b). The ENS in turn is influenced by the extrinsic elements of the autonomic nervous system through the parasympathetic and sympathetic nerves, which modulate activity in the ENS according to the digestive needs of the individual as determined by both somatic and autonomic demands.

MOTOR CONTROL OF THE STOMACH

Figure 2.1

59

Schematic representation of the hierarchy of neural controls regulating gastric motor function.

Motor neurones in the ENS, be they excitatory or inhibitory, therefore provide the final pathway for both intrinsic and extrinsic reflexes. In this review the organisation of the neural control mechanisms which are responsible for this wide repertoire of gastric motor activity will be discussed.

THE ENTERIC NERVOUS SYSTEM The enteric nervous system in the stomach differs from other regions of the gut in that it consist of only one ganglionated plexus. The intrinsic innervation of the stomach is primarily located in the myenteric plexus since a ganglionated submucosal plexus is either very sparsely present or completely absent (Müller, 1920; Radke, Stach and Weiss, 1980; Christensen and Rick, 1985; Keast, Furness and Costa, 1985; Furness et al., 1991). Therefore, one must conclude that the myenteric plexus contains neurones involved in the regulation of both muscle and mucosal functions. Myenteric ganglia throughout the stomach do not run parallel to the circular muscle as they do in the distal small intestine, but rather they form a honeycombed-like meshwork similar to that seen in the proximal small intestine and the colon. The density and size of myenteric ganglia increases from the fundus to the antrum and there is also a decreasing gradient from lesser to greater curvature.

60

INNERVATION OF THE GASTROINTESTINAL TRACT

A subserosal plexus located along the lesser curvature has been described in the rat and opossum (Skad-Nielsen, Poulsen and Holst, 1987; King and Szurszewski, 1984). There is evidence that these parafascicular ganglion cells receive input from extrinsic vagal fibres and it has been suggested that they serve as relay stations to transmit vagal signals to the periphery (King and Szurszewski, 1984).

ELECTROPHYSIOLOGY AND NEUROPHARMACOLOGY OF GASTRIC MYENTERIC NEURONES Electrophysiological properties of gastric myenteric neurones Different classes of enteric neurones with specific electrical and synaptic properties have been described in different regions of the gut (for review see Wood, 1994a; Bornstein, Furness and Kunze, 1994; Chapters 1 and 10 of this volume). Their electrical properties range from inexcitable cells to highly excitable cells with spontaneous activity. In addition, the full repertoire of synaptic transmission, i.e. fast and slow excitatory and inhibitory potentials, has been described in the enteric nervous system and these may be modulated by presynaptic inhibitory or excitatory mechanisms. All neurones in the enteric nervous system appear to receive synaptic input although any one neurone does not usually receive the entire complement of synaptic input (see Wood, 1994b). Although the issue of terminology has not been finally resolved there is a general agreement on the existence of two major classes of cells which were first described in the small intestine (Nishi and North, 1973; Hirst, Holman and Spence, 1974, see also Chapters 1 and 10 in this volume). One class of cell has phasic or tonic spike discharge during the injection of depolarising current pulses and usually exhibits fast excitatory post-synaptic potentials (fEPSP) after stimulation of synaptic input. These cells have been termed S-cells or Type 1 neurones. In contrast AH-cells, also described as Type 2 neurones, usually fire fewer action potentials than S-cells and are characterised by a prominent Ca-hump during the repolarising phase of the action potential and by a prolonged after-hyperpolarisation following their action potentials, hence the term AH. While action potentials in S-cells are carried mainly by sodium currents, calcium currents contribute to action potentials in AH-cells. AH-cells very rarely receive fast synaptic input but slow EPSPs can be readily induced in these cells. S-cells may be activated by fast and slow excitatory input. In addition to S and AH cells, a number of other cell types with intermediate electrophysiological behaviour have been described throughout the gut (for review see Wood, 1994a; Bornstein, Furness and Kunze, 1994). AH-cells have received particular attention over the last years because a sub-population of these are likely to function as intrinsic sensory neurones (Kunze, Bornstein and Furness, 1995). The electrical and synaptic properties of gastric myenteric neurones differ from those in the small and large intestine in a number of ways. The most striking difference is the total lack of AH cells in the corpus region (Schemann and Wood, 1989a). A few neurones with properties similar to AH cells have been described in one study on antral myenteric neurones (Tack and Wood, 1992a). The majority of gastric myenteric neurones behave like phasic or tonic S cells and they have been termed Gastric I and Gastric II neurones, respectively (Schemann and Wood, 1989a). A small population of neurones (less than 5%)

MOTOR CONTROL OF THE STOMACH

61

which received fast synaptic input but were otherwise inexcitable were called Gastric III neurones (Schemann and Wood, 1989a). In the small intestine, initially inexcitable cells have been described which in time develop characteristics of AH cells (Wood, 1994b). However, it is unlikely that the gastric III neurones are AH cells since no spike generation occurs even after hours of impalement. In addition, unlike their counterparts in the intestine, inexcitable neurones in the gastric corpus can not be activated by forskolin. Unlike in the small intestine (see Chapters 1 and 10 of this volume) correlations between the neuronal types, i.e. Gastric I, II or III, with their morphological characteristics, their function as motor- or interneurones, and their pharmacology were not as obvious. The only relation found so far was that descending inhibitory neurones mostly exhibit Gastric I properties whereas Gastric I and II neurones were equally distributed in the population of ascending excitatory neurones. However, it still remains open at the moment whether the different neuronal types represent one cell class exhibiting different excitability levels, or whether their behaviour reflects functional specialisations. Synaptic properties of gastric myenteric neurones In addition to differences in electrical behaviour, there are also differences in the synaptic properties of gastric and intestinal enteric neurones. Synaptic transmission to gastric myenteric neurones, whether originating from intrinsic myenteric or extrinsic vagal fibres, appears to rely mainly on fEPSPs. In contrast to intestinal enteric neurones, all gastric myenteric neurones receive fEPSPs (Schemann and Wood, 1989b; Tack and Wood, 1991). The transmitter for these fEPSPs is acetylcholine which activates post-synaptic neurones via nicotinic receptors. The fEPSPs in Gastric I neurones usually reach threshold for spike discharge whereas the fEPSPs in Gastric II neurones are often subthreshold and of smaller amplitude. The small amplitude fEPSPs in Gastric III neurones never reach threshold for spike discharge. This may reflect differences in the number of cholinergic terminals synapsing onto the different cell types or may be simply related to different excitability level and passive properties of the membrane. Stimulation of either vagal or intrinsic fibres produced only occasionally slow EPSPs which are very likely mediated by acetylcholine or tachykinins (Schemann and Wood, 1989b; Tack and Wood, 1991; Schemann and Grundy, 1992). The relative paucity of slow excitatory and inhibitory synaptic events in gastric compared to intestinal enteric neurones is difficult to explain, in particular since the putative mediators of these two events are present in the gastric myenteric plexus or in extrinsic fibres supplying the myenteric plexus (Mawe et al., 1989; Schemann, Schaaf and Mäder, 1995) and moreover, exogenous application of mediators, including serotonin (Schemann, 1991a; Tack et al., 1992), substance P (Schemann and Kayser, 1991), noradrenaline (Schemann, 1991b; Tack and Wood, 1992b) and motilin (Tack, 1995) mimic slow EPSP-like responses in many neurones. Noradrenaline also has to be included in the list of putative mediators for slow EPSPs because it excites a small population of gastric myenteric neurones via α1 receptors (Schemann, 1991b; Tack et al., 1992). The functional relevance of the noradrenergic excitatory response remains unclear. It could be involved in excitatory and inhibitory responses observed after stimulation of intrinsic and vagal fibres because activation of α1 receptors decreases the amplitude of the rebound contractions while simultaneously

62

INNERVATION OF THE GASTROINTESTINAL TRACT

increasing the contractile on-response and the relaxation in guinea pig stomach (Hillsley, Schemann and Grundy, 1992). It is notable that many mediators which enhance or decrease excitability in intestinal myenteric neurones do not affect gastric myenteric neurones (Schemann, 1990, 1992). These include substances like vasoactive intestinal peptide (VIP), calcitonin gene related peptide (CGRP), gastrin releasing peptide (GRP) and galanin. Antidromic stimulation of extrinsic sensory nerves to cause the peripheral release of tachykinins might also contribute to the long lasting activation of myenteric neurones (Delbro et al., 1983). The concept of axon reflexes is supported by data showing that nearly all neurones in the gastric myenteric plexus are excited by the sensory neurotoxin capsaicin. The excitatory effect of capsaicin can be abolished by desensitising the neurones to substance P or to [MePhe7]NKB which would suggest that neuronal NK3 receptors are involved in the effects of capsaicin-sensitive extrinsic sensory neurones on the gastric myenteric plexus. Although slow EPSPs are only rarely induced, a long lasting excitation of gastric neurones may be achieved by prolonged stimulation of cholinergic synapses. This is possible because the fEPSPs in the stomach do not show the run-down phenomenon that suppresses fEPSP discharge in myenteric neurones of the small intestine. Recurrent activation and positive feedback regulation involving release of acetylcholine are also mechanisms to raise the excitation level over extended periods of time (Schemann and Wood, 1989b). Auto-inhibition of acetylcholine release via presynaptic muscarinic receptors does not appear to play a significant role in the stomach, which could explain the high discharge rate of stimulus evoked fEPSPs (Schemann and Wood, 1989b). Spontaneous fEPSPs in a minority of gastric myenteric neurones is evidence for ongoing release of acetylcholine which coincides with basal cholinergic tone present in isolated stomach preparations (Beani, Bianchi and Crema, 1971; Hennig, Brookes and Costa, 1997). Neither slow nor fast inhibitory post-synaptic potentials have been described in myenteric neurones of the stomach. Inhibitory mechanisms, however, do operate in the stomach and presynaptic inhibition is one mechanism by which fEPSP discharge is modulated. However, a substantial number of fEPSPs appear to be resistant to presynaptic modulation, which is not surprising in view of the dominant role of fEPSPs for synaptic transmission. The most effective substances which presynaptically inhibit acetylcholine release are neuropeptide Y (NPY), peptide YY (PYY) and pancreatic polypeptide (Schemann and Tamura, 1992). In addition, noradrenaline and serotonin act to presynaptically inhibit cholinergic fEPSPs and non-cholinergic slow EPSPs (Schemann, 1990, 1991a,b; Tack and Wood, 1992). Vago-enteric interactions The majority of gastric myenteric neurones receive vagal input as demonstrated directly by electrophysiological recordings (Schemann and Grundy, 1992). Electrical stimulation of the vagus evokes cholinergic fEPSPs mediated via nicotinic receptors. Thus both intrinsic and vagal extrinsic transmission is dependent on cholinergic nicotinic synapses. Cholinergic and nitrergic motor, as well as non-motorneurones (presumably interneurones), are synaptically activated by vagal fibres. Occasionally, high frequency stimulation of the

MOTOR CONTROL OF THE STOMACH

63

vagus was seen to produce a small amplitude slow EPSP, however, the mediator of the slow EPSPs has not been identified. Presynaptic inhibition is not restricted to fEPSPs evoked by intrinsic myenteric nerve fibres, but the same substances can also suppress vagal input to myenteric neurones. Indeed, some selectivity for vagal rather than intrinsic fEPSPs has been described for noradrenaline acting on α2 receptors to mediate presynaptic inhibition (Schemann, 1991b). This might represent a way in which vagal input to gastric myenteric neurones is selectively removed in order to favour, under certain conditions, the activation of intrinsic reflex circuits. Functional studies would support this view since a noradrenaline induced inhibition on gastric motility requires vagal cholinergic tone (Muren, 1957). The α2 receptor-mediated inhibition of transmitter release, mainly affecting acetylcholine release, is involved in the inhibition of electrically evoked contractile response in the stomach (Hillsley, Schemann and Grundy, 1992). Serotonin and NPY also have presynaptic inhibitory effects on some vagal synapses. NPY is released upon stimulation of nicotinic receptors and its release is blocked by hexamethonium (McIntosh, Dadgar and Kwok, 1992). PYY, which would presynaptically block both vagal and intrinsic fEPSPs has been shown to inhibit both basal gastric muscle contractions and those evoked by nerve stimulation by inhibiting cholinergic neurotransmission in the guinea-pig stomach (Wiley, Lu and Owyang, 1991). NPY-immunoreactivity in the myenteric plexus is not significantly affected by chemical sympathectomy (Mawe et al., 1989). Therefore, it seems possible that the presynaptic inhibitory effect on vagal inputs by NPY reflects the ability of myenteric NPY-positive neurones to actively shut down the vagal preganglionic input. This however is not a selective inhibition because in the same neurones intrinsic fEPSPs are also NPY-sensitive. Both electrophysiological and morphological studies indicated convergence as well as divergence of vagal inputs to myenteric ganglia (Schemann and Grundy, 1992; Berthoud, Jedrzejewska and Powley, 1990; Powley et al., 1994). These studies clearly demonstrated that the extent of vagal influences on gastric myenteric neurones had been previously underestimated. The concept that extrinsic fibres, in particular vagal fibres, innervate only a small number of myenteric neurones which would serve as command neurones has to be abandoned in favour of a more widespread influence of vagal fibres on the activity of gastric myenteric neurones (see below). NEUROCHEMICAL CODING OF ENTERIC NERVES IN THE STOMACH Plurichemical neurotransmission is one of the key features in the enteric nervous system and recognition of the way certain putative neurotransmitters or neuronal markers co-exist has been an important development in our ability to identify functionally distinct populations of neurones. The presence of a variety of transmitters in gastric myenteric neurones has been demonstrated in a number of different species (Gabriel et al., 1992; Kyosola, Veijola and Rechardt, 1975; Furness et al., 1991; Sternini, DeGiorgio and Furness, 1992; Van Aswegen et al., 1994; Jensen and Holmgren, 1991; Junquera et al., 1986; Schemann, Schaaf and Mäder, 1995; Schultzberg et al., 1980; Scheuermann et al., 1991; Wathuta, 1986; Groenewald, 1994; Yamamoto et al., 1994; Sann et al., 1998). Although the detailed chemical code of myenteric neurones varies between different regions of the gut and also between different species, it is notable that certain populations and their code appear to be

64

INNERVATION OF THE GASTROINTESTINAL TRACT

highly conserved. This applies especially for neurones which contain nitric oxide synthase (NOS) and choline acetyltransferase (ChAT) which, except for minor overlap, belong to separate populations. These two enzymes are markers for cholinergic and nitrergic neurons respectively (Schemann, Schaaf and Mäder, 1995) which appear to be the primary transmitters for enteric reflexes. The most detailed information on the neurochemical coding of myenteric neurones in the stomach has been obtained for the guinea-pig and the description below is derived mainly from this species (Table 2.1). The proportion of cholinergic versus nitrergic neurones is about 70–30% both in the fundus and corpus. However, the cholinergic and nitrergic populations may be further subdivided according to the additional presence of other antigens and in this respect more than ten neuronal subpopulations have been described in the guinea-pig stomach (Table 2.1). Cholinergic subpopulations (at least 8) outnumber the nitrergic subpopulations (at least 2). ChAT-positive cholinergic cell bodies are the predominant neuronal class in the gastric myenteric plexus. A substantial number of these do not appear to contain any other markers from a long list so far examined (Schemann, Schaaf and Mäder, 1995).

TABLE 2.1 Neurochemical coding of guinea-pig gastric myenteric neurones, their relative proportions and putative functions in the corpus and fundus. Chemical code

Population size (Corpus/Fundus)

Main projections

Putative functions

NOS/VIP/NPY/±ENK

28/16

NOS/VIP/NPY/GRP/ ±ENK NOS/(GRP)/(ENK) NOS/–

/4

descending to muscle (only/+ENK) descending to mucosa (only/–ENK) ?

inhibitory muscle motor neurones secreto and/or vasomotor neurones ?

/3 5/9

? descending to muscle

? inhibitory muscle motor neurones secreto-and/or vasomotor neurones interneurones

descending to mucosa

ChAT/VIP/NPY/GRP/ ±DBH ChAT/SP/ENK/±GRP ChAT/SP/ENK/CALRET/ ±GRP ChAT/SP/SOM/±GRP ChAT/±ENK/±GRP

11/12

descending intraganglionic ascending to mucosa

21/15

ascending to muscle

6/6

ascending to muscle

6/2 21/37

? ascending to muscle ascending to mucosa

5-HT/±SP/±VIP/±ChAT

2/2

intraganglionic intraganglionic

secreto-and/or vasomotor neurones excitatory muscle motor neurones excitatory muscle motor neurones ? excitatory muscle motor neurones secreto-and/or vasomotor neurones interneurones interneurones

MOTOR CONTROL OF THE STOMACH

65

However, substance-P is exclusively localised in one subpopulation of cholinergic neurones while some substance-P-positive as well as substance-P-negative cholinergic neurones also contain calretinin, enkephalins (ENK), GRP, 5-hydroxytryptamine (5-HT) or somatostatin (SOM). It is noteworthy that the number of SP-ergic myenteric neurones is twice as high in the corpus compared to the fundus region. Most nitrergic neurons in the guinea-pig are also VIP- and NPY-immunoreactive (Table 2.1). NO or a related compound, VIP and NPY are all candidates for inhibitory transmitters to gastric smooth muscle. A subpopulation of NOS, VIP and NPY containing neurones show also ENK immunoreactivity. Retrograde tracing studies would indicate that these neurones primarily innervate the muscle layers, in particular the circular muscle. Some nitrergic neurones did not show any neuropeptide-immunoreactivity (at least for the neuropeptides tested, see Table 2.1). They were NADPH-diaphorase positive and might be purely nitrergic. A proportion of these neurones were also found to be circular muscle motor neurones. ATP is also a prime candidate according to the purinergic hypothesis (Burnstock, 1972; Hoyle and Burnstock, 1989) and the coexistence of ATP and nitric oxide (NO) has been demonstrated in a subpopulation of myenteric neurones in the small and large intestine (Belai and Burnstock, 1994). However, it is unknown whether ATP is colocalised in these neurones in the stomach. NOS, VIP and NPY are also colocalised in ferret myenteric neurones but in this species one can visualise VIP-immunoreactive cell bodies without the need to pre-treat with colchicine to block axonal transport (Sann et al., 1998). This may mean that expression of VIP is higher in the ferret and could possibly relate to the eating habits of this carnivore which devour small prey whole and might therefore require a more pronounced and powerful relaxation of the stomach. In contrast NO appears to be the major mediator of relaxation in rodents. VIP and NPY are not exclusive to nitrergic neurones but are also present in a population of cholinergic neurones which in addition express DBH-immunoreactivity (Table 2.1). The DBH-cells do not show TH-immunoreactivity and it is unlikely that they synthesise noradrenaline (Mawe et al., 1989; Gabriel, Fekete and Halasy, 1989). Furthermore, this population is unlikely to be involved in the regulation of muscle function because they mainly project to the mucosa (Reiche, Pfannkuche and Schemann, 1997). A few TH-positive but DBH-negative, very likely dopaminergic cells, are however present in the ferret stomach (Sann et al., 1998). PNMT (phenylethanolamine-N-methyltransferase), the enzyme catalyzing the formation of adrenaline is not present in intrinsic neurones in the stomach (Back et al., 1995) but noradrenergic, TH/DBH-positive, and dopaminergic TH-positive and DBH negative nerve terminals can be visualised in sympathetic fibres surrounding myenteric ganglion cells (Mawe et al., 1989; Back et al., 1995). Some of the myenteric VIP neurones might project to the celiac ganglion, but their detailed coding is not known (Lee et al., 1986). There are significant differences in the chemical coding between the stomach and intestine which might be of functional significance. The relative high proportion of the putatively “pure” cholinergic neurones as well as the pronounced colocalisation of NPY in cholinergic neurones is striking in the stomach. CGRP is not present in intrinsic neurones in the stomach but is present in extrinsic sensory fibres where it is often colocalised with substance-P (see below). However, probably the most striking difference is the total lack of the calcium binding protein, calbindin, in myenteric neurones in the corpus. Very few

66

INNERVATION OF THE GASTROINTESTINAL TRACT

calbindin-positive neurones have been reported in the antrum region (Furness et al., 1988) and the fundus (less than 2%). This is of functional significance because one population of calbindin positive neurones are strong candidates for intrinsic sensory neurones (see below). Either the stomach contains only very few intrinsic sensory neurones, which is unlikely, or some other cell type serves this role. Moreover, the presence or absence of calbindin immunoreactivity should not be taken, per se, as evidence in favour of a particular function. The chemical coding as well as the morphology of calbindin-immunoreactive enteric neurones is different between species and various regions in the gut (Hodgkiss and Lees, 1983; Scheuermann et al., 1991; Furness et al., 1988) and hence they very likely subserve a variety of functions. Other putative transmitters classically involved in neural regulation within the central nervous system have been identified in the gastric myenteric plexus but their detailed chemical coding is unknown. GABAergic myenteric cell bodies are present in the stomach with fibres innervating the circular muscle (Krantis and Webb, 1989; Gilon Tappaz and Remacle, 1991; Tsai, Tsai and Wu, 1993). Glutamatergic nerve cell bodies and fibres in the circular muscle have also been found (Tsai, Tsai and Wu, 1994; Liu et al., 1997). Galanin and thyrotropin-releasing hormone are found in myenteric cell bodies and fibres supplying all layers of the stomach (Ekblad et al., 1985; Wang et al., 1995; Tsuruo et al., 1988). Neurochemical coding of vagal fibres Nerves within the vagus have either motor (efferent fibres) or sensory (afferent fibres) function. The vast majority of vagal motor fibres use acetylcholine as the transmitter and are therefore ChAT-positive. There is evidence for additional transmitters in vagal fibres but it is not known whether they are colocalised with acetylcholine or represent separate populations. Thus, galanin, CGRP, NOS, and substance P have been immunohistochemically localised in vagal fibres (Kirchgessner and Gershon, 1989; Forster and Southam, 1993). Only a minor proportion of vagal afferents innervating the stomach contained CGRP (Sternini and Anderson, 1992). Using different strategies and approaches there is evidence that cholinergic neurones exist in the vagal afferent system (Tsubomura et al., 1987; Ternaux et al., 1989; Falempin et al., 1989). However, their function has not been unequivocally demonstrated. Glutamate is the major transmitter of vagal afferent inputs to the brainstem (Sykes, Spyer and Izzo, 1997). Neurochemical coding related to motor activity in the isolated stomach It is noteworthy that both in the fundus and corpus the cholinergic neurones by far outnumber the nitrergic neurones despite the primary task of the proximal stomach to accommodate volume and hence to initiate inhibition of muscle activity. This reservoir function is therefore not paralleled by a higher number of inhibitory muscle motor neurones. This might point to an important role of disinhibition of cholinergic neurones in the mediation of relaxation (see below). Nevertheless, the chemical coding of gastric myenteric neurones agrees with the differential motor responses evoked by electrical stimulation of the isolated intact stomach or of gastric muscle strip preparations (Paton

MOTOR CONTROL OF THE STOMACH

67

and Vane, 1962; Armitage and Dean, 1966). The response to field stimulation is generally biphasic with an initial transient excitation followed by a longer lasting relaxation of the muscle. In some cases, an excitatory off (or rebound) response and a late excitation may occur while the relaxation may also have multiple components. The contraction/relaxation response is dependent on the stimulus frequencies with contractile responses dominating at lower frequencies (<1 Hz) and relaxation following the contractile response at higher frequencies (1–10 Hz) (Costall, Naylor and Tan, 1983). The initial tone of the preparation is also important in determining the balance of excitatory or inhibitory responses. Low tone favours an excitatory, high tone an inhibitory response (McSwiney and Wadge, 1928). The major mediator of the contractile response at the neuroeffector junction is acetylcholine acting at muscarinic receptors. The acetylcholine is probably released from more than one population of cholinergic myenteric neurones along with other transmitters including substance-P. The relaxation is mediated by non-adrenergic non-cholinergic (NANC) nerves. Several NANC transmitters including purines (Burnstock, 1972), VIP (Fahrenkrug et al., 1978), PACAP (Katsoulis et al., 1996) or NO (Bult et al., 1990) have been proposed and shown to be released upon field stimulation (Belai, Lefebvre and Burnstock, 1991). NPY is another candidate for inhibitory transmission to the muscle since immunoneutralisation to NPY caused a loss in vagal-induced inhibitory drive to the muscle (Grundy, Schemann and Hutson, 1993). At least in muscle strips of guinea-pig fundus, GRP which is present in a small population of nitrergic neurones evokes a relaxation. There has been an explosion of interest in the transmitter(s) released from these NANC inhibitory neurones because of the recent evidence suggesting NO may be involved. The novelty of a transmitter which is neither stored in pre-synaptic vesicles but synthesised de novo in response to an action potential, nor acts on post-synaptic receptor, but interacts intracellularly with guanylate cyclase to cause relaxation, has aroused considerable interest. Since NO is synthesised from L-arginine and analogues of L-arginine which act as inhibitors of NO synthesis are readily available, a role for NO in relaxations throughout the GI tract, including the stomach, has rapidly become established. It is not possible to determine which one of this list of substances is the primary transmitter because their contribution to the relaxation might vary with the species and/or the experimental conditions. Moreover, the coordinated release of two or more substances is likely to be involved in the relaxation response which can often reveal multiple components. A brief burst of low frequency stimulation evokes the short lasting relaxation which is followed by a sustained relaxation as stimulus frequencies and duration is increased. It is generally accepted that NO or a NO-releasing substance is involved in the initial, rapid relaxation, while the additional longer lasting relaxation is mediated by VIP (De Beurme and Lefebvre, 1988; D’Amato, Curro and Montuschi, 1992; Grundy, Gharib-Naseri and Hutson, 1993b; Meulemans, Eelen and Schuurkes, 1995). This conclusion is consistent with release studies demonstrating that low frequency stimulation is sufficient for maximal NO release whereas higher stimulation rates are required to release VIP (Boeckxstaens et al., 1992; Li and Rand, 1990; Takahashi and Owyang, 1995). Muscle cGMP levels increase before cAMP levels during intramural nerve stimulation supporting the view that NO and VIP act sequentially (Ito et al., 1990). Based on the finding that NO and VIP are colocalised in inhibitory myenteric neurones it seems reasonable to assume that these substances are frequency-dependently released from one specific population of inhibitory muscle motor

68

INNERVATION OF THE GASTROINTESTINAL TRACT

neurones. The effects of NO released from inhibitory motor neurones are considered to be more complex than that of VIP and not restricted to a direct effect on muscle activity but might involve an intermediary action on ICC (Daniels and Berezin, 1992; Ward et al., 1997). NO may also be able to amplify VIP effects on muscle by enhancing presynaptically the release of VIP (Grider et al., 1992). There is also evidence for VIP induced release of NO from smooth muscle (Murthy et al., 1993). Blockade of NO release also enhances neurally-induced cholinergic responses evoked by electrical field stimulation and ganglionic stimulation by nicotine or dimethylphenylpiperazinium (DMPP), but not muscular contractions evoked directly by acetylcholine or methacholine (Lefebvre, deVries and Smits, 1992; Baccari, Bertini and Calamai, 1993). This suggests that NO might exert an additional inhibitory influence on acetylcholine release at a prejunctional level. This would be in agreement with the putative role of a subset of NOS-positive myenteric gastric neurones as interneurones (Schemann, Schaaf and Mäder, 1995). However, we have been unable to demonstrate any presynaptic inhibitory effects of the NO-donor sodium nitroprusside on cholinergic fEPSP in the small intestine (Tamura, Schemann and Wood, 1993) or the stomach (own unpublished results). If this holds true, the inhibition of acetylcholine release by NO must occur at the neuromuscular junction rather than at the level of the interneuronal synapses. VIP and ATP also depress neurally induced cholinergic contractions but, in contrast to NO, both ATP and more significantly VIP, depress the muscular contractions evoked by exogenous acetylcholine which might indicate that ATP and VIP have postjunctional effects (Baccari, Calamai and Staderini, 1994). Alterations in bowel motility, such as ileus, might result from excessive concentrations of NO generated during endotoxicosis and inflammatory bowel disease (Salzmann, 1995). However, a significant role of nNOS is supported by a study in animals with gastric hypomotility. Dairy cows suffering from abomasal displacement show signs of upregulation of nitrergic pathways which has been suggested to be involved in the occurrence of a dilated abomasum (Geishauser, Reiche and Schemann, 1998). Electrical stimulation of the vagus nerve in vivo mediates essentially identical responses to those described above for intramural field stimulation, i.e. cholinergic excitation followed by NANC-mediated inhibition (Klee, 1912) with both NO and VIP contributing largely to the latter (Grundy, Gharib-Naseri and Hutson, 1993b). Both the vagal excitatory and inhibitory responses are hexamethonium sensitive, mimicked by the nicotinic agonist DMPP and therefore dependent upon a predominantly cholinergic nicotinic transmission (Greef and Holtz, 1956a,b; Greef, Kasperat and Osswald, 1962; Paton and Vane, 1962; Armitage and Dean, 1966; Yano et al., 1983). This conclusion is supported by the electrophysiological demonstration of vagally evoked nicotinic fEPSPs in gastric myenteric neurones (Schemann and Grundy, 1992) and the observation that vagal stimulation-induced NO and VIP release is mediated via nicotinic receptors (Ito et al., 1988; Takahashi and Owyang, 1995). Transmural stimulation is reported to evoke stronger contractions, more vigorous peristalsis and a more pronounced relaxation than electrical stimulation of the vagus alone (Paton and Vane, 1963; Armitage and Dean, 1966) while DMPP evokes a stronger relaxation than that evoked by stimulation of the vagus (Yano et al., 1983). It appears, therefore, that transmural stimulation activates nerves inaccessible to the vagus, which may be truly “autonomic” (Paton and Vane, 1963). This has been confirmed by electrophysiological studies demonstrating

MOTOR CONTROL OF THE STOMACH

69

that not all gastric myenteric neurones receive vagal input (Schemann and Grundy, 1992). PROJECTIONS AND CIRCUITS WITHIN THE MYENTERIC PLEXUS OF THE STOMACH Recent studies have demonstrated that the projection of cholinergic and nitrergic pathways within the myenteric plexus are polarised into ascending and descending projections in all regions including corpus, fundus (Brookes, Schemann and Hennig, 1994; Schemann and Schaaf, 1995; Pfannkuche et al., 1998; Brookes, Hennig and Schemann, 1998) and antrum (own unpublished results). The intracellular injection of myenteric neurones with neurobiotin allows the cell bodies of electrophysiologically characterised neurones to be identified and the terminals traced into the circular muscle, the longitudinal muscle or to other myenteric neurones (Schemann and Schaaf, 1995). From these morphological study it can be concluded that the gastric myenteric plexus contains motor neurones which project to the muscle. Other neurones project to myenteric cell bodies which we refer to as non-motor neurones, which because of their varicose endings in close proximity to other myenteric neurones, are likely to function as interneurones. The heterogeneous appearance of the terminals of these non-motor neurones, ranging from heavily branching basket-like endings to more simple varicose endings, might be taken as a preliminary indication of functional specialisation. The vast majority of myenteric neurones in the stomach were uniaxonal with paddle-shaped or filamentous dendrites but a small proportion of uniaxonal gastric neurones have axon collaterals supplying a number of different targets. One process always remains within the plexus, whereas the other branches project to the muscle, the mucosa or to both. Whether some specialised multitargeted neurones with intraganglionic endings or terminals in the muscle function as sensory neurones or have a specialised function as integrative neurones remains to be investigated. Motor neurones make up about two third of the gastric myenteric neurones; only a small population of less than 5% are multitargeted neurones; the remaining neurones belong to the class of non-motorneurones, which includes interneurones and the yet unidentified sensory neurones. Motor, non-motor and multitargeted neurones had no distinguishable electrophysiological characteristics. In the stomach, as in the small intestine, there is a projection preference of cholinergic and nitrergic myenteric neurones (Figure 2.2) (Schemann and Schaaf, 1995; Brookes, Hennig and Schemann, 1998; Pfannkuche et al., 1998). Intracellular labelling of myenteric neurones, sampled at random, indicates that the vast majority of cholinergic motor neurones project in an ascending direction. In contrast, the vast majority of nitrergic motor neurones are descending (Schemann and Schaaf, 1995). This projection pattern applies also for interneurones. Thus cholinergic interneurones have primarily ascending and nitrergric interneurones primarily descending projections. The fibres typically run for 0.5–20 mm in the longitudinal direction with 90% of the fibres projecting within 0.5–5 mm. Long projections over several centimetres are extremely rare in all regions of the stomach. An alternative labelling technique using the retrograde tracer DiI (1′-didodecyl3,3,3′,3′-tetramethyl-indocarbocyanine perchlorate) placed on a small area of circular

70

INNERVATION OF THE GASTROINTESTINAL TRACT

Figure 2.2 Ascending muscle motor neurones are cholinergic whereas descending muscle motor neurones are nitrergic. A small 50 µm bead of a retrograde tracer (DiI) has been placed onto the circular muscle. The tissue has been kept in organotypic culture and processed for immunohistochemistry for ChAT and NOS. On the left side (A–C) chemical code of ascending neurones are illustrated. The DiI filled cells are shown in B. The same field shows that the DiI cells are ChAT-positive but NOS-negative. On the right side (D–F) chemical coding of descending neurones are illustrated. The DiI filled cells (shown in E) are NOS-positive (D) but ChATnegative (F).

muscle (50 µm) has confirmed the projection polarity for myenteric neurones (Figure 2.3) (Brookes, Hennig and Schemann, 1998; Pfannkuche et al., 1998). In addition, using this technique, we have determined the detailed chemical coding of circular muscle motor neurones (Table 2.1). Ascending cholinergic neurones to this layer are both substance P-positive and substance P-negative and can be further divided by the additional presence or absence of GRP. Both populations of descending nitrergic neurones, those colocalised with VIP and NPY and those lacking the two peptides, project to the circular muscle. Functionally, this projection preference of cholinergic and nitrergic motor and interneurones would indicate the presence of a hardwired circuit which upon stimulation could initiate ascending excitation and descending inhibition of the circular muscle (Figure 2.4). It has to be stressed that this projection pattern of cholinergic and nitrergic myenteric neurones appears to be a general principle throughout the gut. The polarised circuits in the stomach may be differentially activated during gastric emptying in order to adjust excitatory and inhibitory tone to an appropriate level (Table 2.2 and Figure 2.4). This differential control has been recently demonstrated for the action of serotonin (Michel et al., 1997). Both muscle motor and interneurones are activated by serotonin. A fast transient excitatory response is mediated by 5-HT3 receptors and a slow long lasting activation is mediated by 5-HT1p receptors. It has been shown by intracellular dye injection in combination with the neurochemical coding that a proportion of the nitrergic neurones (30%) are activated by 5-HT1p receptors. In contrast, the 5-HT3 mediated activation is mainly restricted to cholinergic substance P-ergic neurones with 70% of the

MOTOR CONTROL OF THE STOMACH

71

Figure 2.3 Distribution of DiI filled myenteric circular muscle motorneurones. The retrograde tracer DiI was applied to the circular muscle. The spatial distribution of DiI filled cells is shown. In addition the figure demonstrates that both in the fundus and corpus NOS-positive neurones are mainly located oral to the application site (descending neurones) whereas ChAT-positive neurones are primarily located aboral to the application site (ascending neurones). The broken lines indicate the border between fundus, corpus and antrum. The dotted lines indicate the course of the circular muscle.

cholinergic neurones responding. The functional implication is apparent from the observation that the 5-HT1p receptor agonist 5-hydroxyindalpine induces muscle relaxation and inhibits the response to cholinergic transmural stimulation (Michel et al., 1997). In contrast the 5-HT3 receptor agonist 2-methyl-5-hydroxytryptamine induces an atropine sensitive increase in muscle tone and potentiate the cholinergic transmural stimulation response (Desai, Warner and Vane, 1994; Michel et al., 1997). In agreement with these observations, gastric emptying is increased by a 5-HT1p antagonist (Mawe, Branchek and Gershon, 1989; Coulie et al., 1997) whereas gastric emptying of a protein rich meal is delayed by a 5-HT3 antagonist (Forster and Dockray, 1990). In the isolated stomach preparations it has been shown that 5-HT evoked relaxation is linked to NO (Meulemans, Helsen and Schuurkes, 1993a). Comparable differential control may be postulated for other transmitters (Table 2.2), like noradrenaline and tachykinins which are known to affect only a subpopulation of myenteric neurones in the stomach (Schemann, 1991;

72

INNERVATION OF THE GASTROINTESTINAL TRACT

Cholinergic (nicotinic) enteric activation Cholinergic (nicotinic) vagal activation Presynaptic inhibition (NPYergic?) Presynaptic sympathetic inhibition (alpha-2)

Ascending excitatory cholinergic motorneurones

Descending non-cholinergic interneurone

Vagus

Sympathetics

5-HT1p mediated activation

NK-3 mediated activation

Descending inhibitory NANC motorneurones

5-HT3 mediated activation

Ascending cholinergic interneurone

Figure 2.4 This figure illustrates a wiring diagram based on results of neuropharmacological and neurochemical coding studies. The hardwired circuits may be activated through different receptors leading to graded and opposite changes in muscle activity. The sensory elements which might activate the circuits are unknown. Ascending cholinergic and descending nitrergic neurones might form the hardwired basis for a putative peristaltic-like reflex in the stomach.

Tack et al., 1991; Schemann and Kayser, 1991). In addition to the differential activation through post-synaptic receptors there also appears to be differential presynaptic inhibition. Thus, noradrenaline appears to predominantly inhibit cholinergic input to cholinergic motor neurones. There is no differential activation through cholinergic nicotinic receptors because all gastric myenteric neurones can be excited by exogenous acetylcholine and receive cholinergic fEPSPs independent of their function or neurochemical code (Schemann and Wood, 1989; Schemann and Schaaf, 1995). In the small intestine there is strong evidence that the projection polarity of cholinergic and nitrergic neurones in the myenteric plexus is the hardwired basis for the peristaltic reflex (see Chapters 1 and 10). This reflex is responsible for aboral transport of luminal

MOTOR CONTROL OF THE STOMACH

73

TABLE 2.2 Neuropharmacology of gastric myenteric neurones. Neurotransmitter/ Neuromodulator

Response

Receptor

Neuronal populations responding

Acetylcholine

post-synaptic excitatory: fast, transient post-synaptic excitatory: slow, long lasting post-synaptic inhibitory: slow, long lasting post-synaptic excitatory: fast, transient

~100% ~10% ~1% ~10%

Motilin Neuropeptide Y Noradrenaline

post-synaptic excitatory: slow, long lasting presynaptic inhibitory: block of fEPSP post-synaptic excitatory: slow, long lasting post-synaptic inhibitory: slow, long lasting resynaptic inhibitory: block of fEPSP

nicotinic muscarinic muscarinic strychnine sensitive ? ? α1 α2 α2

5-hydroxytryptamine

post-synaptic excitatory: fast, transient

5-HT3

post-synaptic excitatory: slow, long lasting

5-HT1p

post-synaptic inhibitory: long lasting presynaptic inhibitory: block of fEPSP post-synaptic excitatory: long lasting

5-HT1a 5-HT1a NK3

Glycine

Substance P

? ~80% ~17% ~7% ~30% (primarily cholinergic) ~70% of ascending cholinergic (primarily SPergic) neurones ~30% of descending nitrergic neurones ? ? ~80% (cholinergic and nitrergic)

content and mediates relaxation below and contraction above a stimulus (Bayliss and Starling, 1899; Langley and Magnus, 1905; Magnus, 1908). The coordinated behaviour of muscle activity might also be relevant for peristalsis in the stomach. The degree of relaxation and the level of contractile activity or tone at adjacent regions is a major determinant of the propulsive force of the peristaltic wave and thereby influences gastric emptying (Keinke, Schemann and Ehrlein, 1984). A functional descending inhibition is indicated by myotomy experiments designed to sever the intramural pathways between the corpus and antrum (Holle, 1992). As a result of this projection the contractile activity aboral to the myotomy increased and led to a constricted antrum. Increase in basal tone of isolated stomach preparations after blocking neural activity with tetrodotoxin would support an important role of an inhibitory tone (Meulemans, Helsen and Schuurkes, 1993b). Increased tone of the isolated stomach treated with the L-arginine analogue (L-NAME) would indicate that NO is responsible, although not solely, for maintenance of inhibitory tone (Hennig, Brookes and Costa, 1997). Irrespective of the possible existence of a local peristaltic reflex in the stomach, other nerve mediated reflexes have been demonstrated. In isolated guinea-pig stomach, gradual filling of the stomach results in reproducible adaptive responses consisting of a relaxation of the fundus together with a substantial increase in intra-gastric volume. This response is independent of extrinsic innervation and is NANC-mediated (Carlson, 1913; Paton and Vane, 1963; Desai, Sessa and Vane, 1991; Hennig, Brookes and Costa, 1997). Distension of the LES caused gastric relaxation in the isolated stomach preparation, a reflex which depends on intramural NANC mechanisms (Ohta, Nakazato and Ohga, 1985). Both

74

INNERVATION OF THE GASTROINTESTINAL TRACT

functional and morphological studies indicate that intrinsic gastric nerves may also modulate LES tone (Schulze-Delrieu et al., 1989; Brookes et al., 1996). Involvement of intrinsic neurones in distension evoked relaxation of the rat stomach is strongly suggested by a recent demonstration of c-fos expression in gastric myenteric neurones after refeeding fasting rats or distension of the rat stomach (Dimaline et al., 1995). While refeeding increased c-fos mRNA, physiological distension induced expression of c-fos, c-jun and c-myc in myenteric neurones. In this context it should be remembered that the initial increase in gastric pressure which occurs during distension is absent during feeding (when the stomach fills naturally) (Sick and Tedesco, 1908; Hennig, Brookes and Costa, 1997). INTRINSIC CONTROL OF GASTRIC MOTOR ACTIVITY Studies based on X-Ray examination of the stomach indicate the importance of muscle activity and tone just oral and aboral to a peristaltic wave (Keinke, Schemann and Ehrlein, 1984). The degree of relaxation and the level of contractile activity above and below a peristaltic contraction determines the propulsive force of the peristaltic wave and thereby the net transport of gastric content towards the pylorus. The spatial specificity and local modulation of neural activity determines the occurrence and timing of lumen occlusion relative to adjacent segments. Peristaltic activity in the stomach can be observed in the isolated stomach (Goltz, 1872; Paton and Vane, 1963; Armitage and Dean, 1966). Therefore, it must be concluded that intact extrinsic reflexes are not required for peristalsis, however, local axon reflexes might still be operative (see below). Moreover, even a muscle preparation where the myenteric plexus has been removed will respond to stretch with contractions indicating that myogenic mechanism may also be involved (Sick and Tedesco, 1908). Myogenic properties of gastric muscle are therefore involved in the intrinsic adaptations of gastric muscle tension to stretch and might be highly relevant for gastric compliance (Schulze-Delrieu, 1983; Hennig, Brookes and Costa, 1997). Around the time when Bayliss and Starling (1899) were describing the peristaltic reflex in the small intestine a number of investigators were attempting to investigate similar reflex responses in the stomach. However, it appears that the classical peristaltic reflex consisting of a contraction oral and a relaxation aboral to the stimulus (Bayliss and Starling, 1899) is not readily reproducible in the stomach. Yet there is extensive evidence for intrinsic reflex activity in the isolated stomach of a variety of species starting from the pioneering work of the end of the last century which demonstrated responses to mechanical (touching the serosal surface), chemical (crystals of sodium chloride) and electrical stimuli which sometimes induced peristaltic waves and produced variable motor responses in the stomach at the site of the stimulus as well as at oral and aboral sites (Lüderitz, 1891; Schultz, 1897a,b; Winkler, 1898; Auer, 1907; Sick and Tedesco, 1908, Schulze-Delrieu, 1990; Hennig, Brookes and Costa, 1997). In most cases, atropine sensitive local contractions and contractions oral to the stimulus were observed. Mechanical or chemical stimuli can also evoke an aboral contraction which sometimes obscured the small relaxations (Sick and Tedesco, 1911). These cholinergic influences are readily demonstrable in isolated muscle strips (Kuroda, 1924; Beani, Bianchi and Crema, 1971) where phasic activity will cease after removing the myenteric plexus (Sick and Tedesco, 1908).

MOTOR CONTROL OF THE STOMACH

75

The existence of ascending excitatory cholinergic and descending inhibitory nitrergic myenteric neurones in the stomach might represent the hardwired circuit for the stimulus evoked oral contraction and aboral relaxation (Figure 2.4), despite the inconsistency of this particular reflex pattern. The dilemma and a possible explanation for the high variability in reflex activity reported for the stomach is that in most studies reflexes occurring close to the site of stimulus have not been carefully distinguished from reflexes occurring at a distance. This is relevant as the projection of ascending and descending neurones in the gastric myenteric plexus is only a few millimetres (Schemann and Schaaf, 1995; Brookes, Hennig and Schemann, 1998; Pfannkuche et al., 1998). For initiation of phasic contractions and for the propagation of peristaltic contractions continuity within the myenteric plexus has to be preserved (Sick and Tedesco, 1908; Poos, 1923; Holle, Steinbach and Forth, 1994; Cannon, 1911; Chaplin et al., 1987; Chaplin and Duke, 1990). It is likely that a chain of cholinergic neurones is involved in the longitudinal spread of the peristaltic wave because the ganglionic blocker hexamethonium is able to prevent peristalsis in response to transmural stimulation in an isolated stomach preparation (Bennet Bucknell and Dean, 1966; Armitage and Dean, 1966). The neural control mechanisms in the stomach can obviously initiate a variety of different reflex patterns depending on whether adjustment more locally or at more distant sites are required. Recent studies in the isolated stomach indicate complex reflex responses coordinating motor activities in functionally distinct gastric regions like fundus, corpus and antrum (Yuan, Brookes and Costa, 1997). These reflex pathways have been deduced from intracellular muscle recordings in response to distension. Descending inhibitory reflex pathways existed between the fundus and corpus, ascending inhibitory and excitatory reflex pathways between the antrum and corpus and between corpus and fundus (Yuan, Brookes and Costa, 1997). It is not known whether there are specific innervation patterns for the circular and longitudinal muscle. In the stomach circular and longitudinal muscle layer contract at a given locus simultaneously suggesting that they are activated in a complementary fashion whereas in the intestine the two muscle layers are activated in a reciprocal fashion (Sarna, 1993). Gastric reflexes exist that coordinate antral, corporal and fundal activity. These reflexes are usually opposite to the peristaltic reflex and are significantly reduced but not totally abolished after extrinsic denervation (Andrews and Bingham, 1990). The remaining long descending reflexes are hexamethonium sensitive and hence depend on intramural polysynaptic pathways (Andrews, Grundy and Scratchard, 1980a). THE QUESTION OF INTRINSIC SENSORY NEURONS From the beginning of this century it has become apparent that the isolated stomach can orchestrate complex contractile responses to mechanical stimulation of the muscle or mucosa despite the absence of connections to the brain and spinal cord. Distension of the isolated stomach evokes or enhances peristaltic activity (Hofmeister and Schütz, 1886; Schütz, 1886). In an excised stomach it is possible to evoke peristaltic activity after mucosal stimulation (Bickel, 1925). Muscle strips with mucosa attached produce coordinated peristaltic activity which changed into a more irregular activity when the mucosa has been removed (Poos, 1923). In the isolated stomach preparation instillation of acidic solutions initiated a much more vigorous peristalsis than water or NaCl solutions (Sick

76

INNERVATION OF THE GASTROINTESTINAL TRACT

and Tedesco, 1908). Sensory nerves in the mucosa might transmit information to the myenteric plexus or directly to the muscle as indicated by the observation that lidocaine slows down propagation of peristaltic contractions (Ichikawa, 1989). A significant proportion of gastric myenteric cell bodies send their processes to the mucosa (Reiche, Pfannkuche and Schemann, 1997; Reiche and Schemann, 1998) and might be involved in mucosa/ muscle reflexes. However, neither in electrophysiological nor in neurochemical studies has the population of intrinsic sensory neurones in the stomach been identified to date. Such reflexes are organised entirely within the gut wall indicating that there must either be enteric sensory neurones or that collaterals of primary afferents, with cell bodies outside the gut wall, possess axon collaterals which project to elements of the enteric nervous system. While there is little information about the nature of the intrinsic sensory system in the stomach there is strong support for this concept as a basis for local reflexes in the small and large intestine. Dogiel (1899) recognised the heterogenous nature of enteric neurones and speculated that the Dogiel type-II neurons were sensory. Recent immunocytochemical studies demonstrating Dogiel type-II myenteric neurones with calbindin immunoreactivity and terminals extending to the mucosa provide support for this hypothesis (Furness et al., 1995). Intracellular recordings from these cells have been made during the application of mechanical and chemical stimuli to the mucosa and have confirmed that the peripheral terminals serve a sensory function and transmit information into the enteric circuits to modulate the excitability of S-neurones (Kunze, Bornstein and Furness, 1995). In contrast, Kirchgessner, Tamir and Gershon (1992), using the immuno-detection of fos as a marker for enteric neurones activated by mucosal stimulation, have suggested that the first-order sensory neurones have their cell bodies in the submucosal rather than myenteric plexus and that the extensive fos-immunoreactivity in the myenteric plexus arises from a cholinergic input from submucosal sensory neurones. However, the pattern of intrinsic afferent activation may be stimulus specific since Smith, Bornstein and Furness (1992) showed that distension and deformation of the mucosa activate different sensory neurones but a common final motor pathway. Numerous studies have demonstrated gastrointestinal reflexes in vitro which are abolished by tetrodotoxin and therefore are of intrinsic origin. Using elegant variations of the basic whole mount preparations for intracellular electrophysiology, several studies have characterised the synaptic responses in the ascending and descending enteric pathways that are an essential element of the peristaltic reflex (Hirst, Holman and McKirdy, 1975; Smith, Bornstein and Furness, 1990). The convergence of inputs following stimulation of mucosal endings and mechanoreceptors in the muscle points to a common motor programme for propulsion independent of the mechanism of activation. This programme could be triggered by intrinsic sensory neurones or axon collaterals of primary afferents and using capsaicin as a neurotoxin selective for the latter, there does indeed appear to be a deficit in reflex behaviour when primary afferents are eliminated in the colon (Grider and Jin, 1994). Distension localised to different regions of the isolated stomach also evokes a complex pattern of reflex excitation and inhibition revealed by intracellular recordings from the guinea-pig circular muscle (Yuan, Brookes and Costa, 1997). Both ascending excitation and descending inhibition were evident over relatively long distances between corpus and antrum. Thus, distension of the antrum evoked inhibition of the

MOTOR CONTROL OF THE STOMACH

77

corpus while distension of the corpus evoked inhibition followed by excitation of the fundus. When these inhibitory junction potentials were attenuated pharmacologically with L-NOARG and apamin, distension was shown to evoke ascending excitatory reflexes. No reflex activity was found in the antrum following stimulation of the corpus. The authors conclude that these reflexes could arise from intrinsic sensory neurones or from collaterals of extrinsic vagal or spinal afferents but offer no evidence either way. From experiments in which the movement of colonic segments were examined after degenerative section of the extrinsic innervation, Grider and Jin (1994), suggested that responses to muscle stretch were mediated by axon reflexes while reflexes from the mucosa were unaffected. In the small intestine, however, both muscle and mucosal reflex responses were maintained in chronically denervated segments (Furness et al., 1995). Thus there may be different patterns of sensory innervation to mucosa and muscle in colon and small intestine. Relating these observations to the stomach is difficult because in this region there is no ganglionated sub-mucosal plexus and neurones with Dogiel type II morphology, calbindin immunoreactivity or AH electrophysiology in the myenteric plexus are rare.

THE EXTRINSIC INNERVATION EXTRINSIC AFFERENTS The upper gastrointestinal tract has an extensive afferent innervation which conveys sensory information to the brainstem and spinal cord via vagal and splanchnic afferents respectively. Under normal circumstances the information conveyed in these afferents does not reach consciousness but is involved in reflexes which control gastric tone and motility. However, these afferents can mediate sensations which can range from vague feelings of fullness and satiety to discomfort, pain and nausea. According to the summation theory of nociception, pain is elicited when receptors are excessively stimulated rather than by specific nociceptors. In this respect splanchnic afferents can respond to subnoxious levels of stimulation which mediate reflex inhibition of gastrointestinal motor function and might mediate pain when the level of stimulation becomes noxious. Some afferents respond only at noxious levels and may be specific nociceptors. These may become sensitised during inflammation while other “silent” nociceptors may be recruited as a consequence of a range of inflammatory mediators (see Sengupta and Gebhart, 1994). In organs like the stomach (although the observations have been made in the bladder and colon) which can generate a range of sensations from mild fullness to intense pain, it is likely that the relative activation of low and high threshold afferent fibres is a determining factor (Cervero and Jänig, 1992). The splanchnic afferents have been likened to a trip-wire alarm system because the relatively small number of afferent fibres (<7% of DRG cells innervate the gastrointestinal viscera) have widespread connections in the spinal cord. There is little capacity for discrimination in the system but once it is activated it can have massive effector responses (Cervero and Tattersall, 1986). There is little evidence that vagal afferents mediate sensations but instead convey moment-to-moment information on gastrointestinal function to the brainstem. At the level of the diaphragm vagal afferent fibres outnumber efferent fibres by about 10 to 1

78

INNERVATION OF THE GASTROINTESTINAL TRACT

providing an enormous volume of afferent traffic that plays an important role in reflex control of gastrointestinal function and behavioural responses such as satiety (see Grundy, 1988). Vagal afferents have their cell bodies in the nodose ganglia and provide peripheral and central projections to the gut and the brain respectively. The terminals of primary afferents within the stomach wall have been visualised by a variety of fibre-tracing and immunocytochemical approaches. Vagal afferents have been shown to terminate in both mucosa and muscle where the endings may subserve a sensory function while other terminations in the myenteric plexus are more consistent with a role in mediating axon reflexes (Berthoud and Powley, 1992; Berthoud et al., 1997). However, the possibility that these intraganglionic endings may sense the neurochemical environment or electrical activity within the ganglia, as suggested by Kolossow and Milochin (1963), can not be discounted, especially in the absence of good evidence for functional synaptic inputs from vagal afferents to gastric myenteric neurones (Schemann and Grundy, 1992; Zheng et al., 1997). Using confocal microscopy, the terminal arborisations of individual vagal afferents have been identified in both the muscle and mucosal layers of the stomach and duodenum (Berthoud et al., 1997; Berthoud and Powley, 1992; Powley et al., 1994). The projections into the myenteric plexus have been termed intraganglionic laminar endings (IGLE) and have a high density in the stomach with about 30 to 50% of ganglia receiving at least one IGLE. A parent axon typically branches to form a highly arborised network on the surface of an individual myenteric ganglia immediately beneath the basal lamina and interdigitating with glial processes. This has been suggested to provide a mechanical link to the connective tissue matrix around the ganglia and is consistent with a mechanosensitive role detecting sheer forces generated by contraction of the overlying or underlying muscle. However, since the same fibres project into the muscle layer and endings here are better positioned to detect muscle tension, the role of IGLEs remains speculative. An “efferent” role for vagal afferent terminals in the myenteric plexus is inferred from the presence of a variety of peptides including substance P, cholecystokinin (CCK), SOM, VIP, GABA and CGRP in the cell bodies of these neurones in the nodose ganglia and which accumulate on the oral side of the ligated vagus indicating transport towards the periphery (Dockray and Sharkey, 1986; Varro et al., 1988; Szabat et al., 1992). Large numbers of nodose ganglion cells contain NADPH-diaphorase (Ichikawa and Helke, 1996). However, there is little information on the function of any of these agents derived from vagal afferents on gastric motor function although axon reflexes are inferred from responses to vagal stimulation that are hexamethonium resistant (Delbro and Lisander, 1982; Delbro et al., 1983). The terminations of splanchnic afferents are also extensively distributed within the gut wall and also project into the prevertebral ganglia where they can influence the activity of post-ganglionic sympathetic neurones. A major feature of the splanchnic afferent fibres in the stomach is that they are immunoreactive to CGRP and, to a variable extent in different species, also contain substance P (Green and Dockray, 1988). They supply the muscle layers, myenteric plexus and submucosal blood vessels and disappear after treatment with capsaicin or coeliac ganglionectomy. Retrogradely labelled gastric sensory neurones in the DRG were predominantly (85%) CGRP-positive and the majority of these neurones colocalised substance P (Green and Dockray, 1988). Other peptides and neurochemical

MOTOR CONTROL OF THE STOMACH

79

markers present in DRG neurones include ChAT, NOS, galanin, SOM, CCK, VIP, GRP, oxytocin but the functional significance of these is unknown (See Szurszewski and Miller, 1994). Peripheral release of CGRP from spinal afferents following noxious stimulation is implicated in hyperaemia that has a cytoprotective role following mucosal damage (Holzer 1994; Chapter 3 of this volume) and appears to play a role in the inhibition of gastric emptying seen after surgery (Plourde et al., 1993). The gastric emptying of acid and hypertonic saline is increased following immunoneutralisation of CGRP which implies that axon reflexes may also have a more physiological role (Forster and Dockray, 1991). CGRP nerve terminals are also distributed around gastric myenteric ganglia although an influence on synaptic transmission at this level has not been demonstrated. Substance P, in contrast, has a potent depolarising action on gastric myenteric neurones via the NK-3 receptor, but because substance P can also be of intrinsic origin it is not clear if this represents a potential action of afferent axon collaterals on myenteric reflex circuits (Schemann and Kayser, 1991). Splanchnic afferents also have collateral branches supplying the prevertebral ganglia where they may modulate transmission through the sympathetic pathway (see Szurszewski and Miller, 1994). Immunoreactive CGRP and substance P terminals form pericellular nests around principle ganglion cells in a number of species. Application of substance P mimics the noncholinergic sEPSP that follows electrical stimulation of the inputs to the prevertebral ganglia and both are reduced by desensitisation to substance P and after capsaicin treatment in vitro. Thus both morphological and electrophysiological data support the view that splanchnic afferents participate in peripheral reflexes and at the same time convey nociceptive information to the spinal cord. The role of CGRP in the pre-vertebral ganglia is not yet known. Another mechanism for reflexes through the prevertebral ganglia stems from a population of VIP containing myenteric neurones that project from the stomach and intestine to the coeliac and other prevertebral ganglia (Lee et al., 1986). These may also be ChAT positive (Mann et al., 1995). Postganglionic sympathetic neurones have been classified according to their transmitter phenotype into those that contain noradrenaline alone and others which co-localise either NPY or somatostatin (Macrae, Furness and Costa, 1986). These innervate different functional targets in the small intestine with those containing NPY supplying mainly blood vessels and, in the small intestine, those containing SOM supplying the submucosal plexus. Those containing only noradrenaline have the myenteric plexus as their target. All receive input from preganglionic sympathetic neurones but it is predominantly the latter and rarely the former neurones which receive inputs from spinal afferent and viscerofugal fibres suggesting that activation of these pathways will lead to inhibition of motor function rather than influencing gastric blood flow (Mann et al., 1995). This observation is consistent with in vitro studies that have demonstrated the loss of intestino-gastric inhibitory reflexes after the connection to the coeliac ganglion is severed (Kreulen, Muir and Szurszewski, 1983). An inhibitory reflex from stomach to intestine via the coeliac ganglia has also been described. This mechanism appears to utilise NO in ganglionic transmission, which is consistent with the presence of NADPH-diaphorase stained endings in the coeliac ganglia of rabbit and guinea-pig (Mazet et al., 1993; Mann et al., 1995) although in the latter study this may derive from fibres projecting from the

80

INNERVATION OF THE GASTROINTESTINAL TRACT

large intestine. Some preganglionic sympathetic neurones in the spinal cord may also be NADPH diaphorase positive (Anderson et al., 1993). The nature of the afferent inputs to the pre-vertebral ganglia have been assessed by recordings from principle ganglion cells during stimulation of the viscera. The synaptic input to the ganglion cell from these myenteric afferents appears to arise from mechanoreceptive elements that (in the colon) are “in-parallel” with the circular muscle. Thus, as the viscus fills there is tangential lengthening of the circular muscle which activates mechanosensitive afferents associated with it. Contractions which cause the viscus to empty are associated with off-loading of the mechanoreceptor despite the increase in wall tension and intraluminal pressure. Thus the viscerofugal myenteric neurone may be activated by stretch and not tension as is the case for primary vagal afferents (see below). A number of mediators are implicated in the distension evoked depolarisation of post-ganglionic sympathetic neurones including ACh, VIP, GABA and CCK with both pre-synaptic and post-synaptic mechanisms of action (see Szurszewski and Miller, 1994). There may also be inhibitory influences which allow the preganglionic inputs to gate the inputs from the periphery and as such the functional importance of these two inputs may vary under certain physiological conditions (Miolan and Neil, 1996). Afferent inputs to the CNS While sensory information is available to the local reflex circuits it is only in the CNS that a global picture of gastrointestinal events can be constructed from the convergent inputs from along the entire length of the gastrointestinal tract and from different receptor types signalling different information. These extrinsic afferents are therefore important to allow the stomach to function within the wider context of the entire gastrointestinal tract. There is a considerable literature on the discharge characteristics of gastrointestinal primary afferents, recorded directly either from the axon en route to the CNS or from the cell body in the nodose or dorsal root ganglia. A particular focus in these studies has been the functional characteristic of afferent fibres as determined by the stimulus-response characteristics of their receptive ending. In this way, the frequency of action potentials encodes the stimulus either as a consequence of a direct action on a generator region on the nerve terminal or following the stimulation of a secondary sense cell which encodes stimulus intensity by the release of a neuroactive substance which in turn causes afferent firing. The properties of afferents have also been inferred from recordings made from second order neurones in the pre-vertebral ganglia, spinal cord and brainstem (Raybould, Gayton and Dockray, 1988; Barber, Yuan and Burks, 1989; Anthony and Kreulen, 1990). However, these indirect recordings are influenced by the convergence of incoming information which might be a problem in the brainstem where information from different receptor types along almost the entire GI tract is channelled. Gastric mechanoreceptors Vagal mechanosensitive afferents in the stomach were first described by Paintal (1954). In 1957, Iggo coined the term “in-series tension receptors” because of the ability of these afferents to respond to both passive stretch and active contraction, although this may relate

MOTOR CONTROL OF THE STOMACH

81

to a functional arrangement rather than to an anatomical one (Berthoud and Powley, 1992; Grundy, 1988). Although the properties of muscle mechanoreceptors are homogeneous (all respond to stretch and contraction), they do show a range of spontaneous activity and threshold for activation (Cottrell and Iggo, 1984). Furthermore, receptors in different parts of the stomach behave quite differently because of the different functions of the proximal and distal stomach (Takeshima, 1971; Andrews, Grundy and Scratcherd, 1980b). The properties of muscle mechanoreceptors in the corpus reflect the reservoir function of this region. With the stomach empty they generate a low frequency irregular spontaneous activity which increases dramatically when the volume of the stomach is increased (Andrews, Grundy and Scratcherd, 1980b; Blackshaw, Grundy and Scratcherd, 1987). They therefore effectively monitor the degree of gastric filling. However, being tension receptors, they can also signal increases and decreases in muscle tone, the latter occurring during receptive relaxation. In contrast, the properties of receptors in the antrum reflect the function of this region as a muscular pump. Being less distensible than the corpus, with much of a meal being accommodated in the latter, antral mechanoreceptors do not readily respond to increases in gastric volume, but instead generate a burst of impulses as each wave of contraction passes over the receptive field. The amplitude and waveform of the contraction is reflected in the rate and duration of discharge in the mechanoreceptors, while the velocity and direction of propagation of the contraction will be encoded in the discharge from adjacent receptors. The response of gastric mechanoreceptors to stretch has recently been suggested to be influenced by CCK released when nutrients enter the duodenum. Gastric afferent activity is stimulated by CCK at high doses (Davison and Clarke, 1988) while at low doses CCK can sensitise the afferent’s response to stretch (Schwartz, McHugh and Moran, 1993). This sensitivity is suggested to underlie the satiety effect of CCK with gastric afferents signalling both the quantity of ingested material through distension and its quality in terms of nutrient density following its delivery to the duodenum. However, in the gastric corpus which undergoes pronounced relaxation in response to CCK, the reverse occurs and vagal mechanoreceptors are off-loaded as the stomach relaxes (Grundy, Bagaev and Hillsley, 1995). This observation is consistent with data which suggests that inhibition of food intake by gastric distension and intestinal nutrients is mediated by independent mechanisms (Cheng et al., 1993) and that the action of endogenous CCK on satiety is not via an endocrine action but through local paracrine mechanisms. A more likely target for CCK action is on the terminations of intestinal mucosal afferents which lie in close proximity to the enteroendocrine cells which release CCK in response to luminal nutrients, and which might be implicated in afferent signal transduction (Berthoud and Patterson, 1996). Splanchnic afferents are also mechanosensitive and will respond to stretch and during contraction (Morrison, 1977). However, these afferents are associated with the serosa and its mesenteric connections, where a single afferent terminates in multiple receptive fields which can cover a wide area and extend over more than one visceral target. As such, the relationship between wall tension and afferent activity is less precise than is the case with vagal afferents and these afferents are commonly referred to as “in parallel” receptors. The threshold for activation of splanchnic afferents is generally higher than for vagal afferents which is consistent with their role in visceral sensation. However, because they are active at physiological levels of stimulation these afferents may also play a role in reflex control

82

INNERVATION OF THE GASTROINTESTINAL TRACT

as well as the generalised suppression of gastrointestinal activity which accompanies pathophysiological levels of distension. In addition, splanchnic afferents, at least in more distal regions, can be sensitised by inflammatory agents such as bradykinin and other agents that can be released during ischaemia (Blumberg et al., 1983; Pan and Longhurst, 1996).

Mucosal afferents A diverse population of mucosal receptors have been described on the basis of experiments in which vagal afferent discharge is recorded in response to a range of stimuli applied to the mucosal epithelium (Grundy and Scratcherd, 1989). The endings are ideally positioned to signal information on the physical and chemical environment of the gut lumen and a complex picture of mucosal sensitivity has emerged (Mei, 1985). Much of the literature on mucosal afferents is concerned with those supplying the small intestine and especially the duodenum where chemosensitivity underlies the feedback mechanisms for gastric emptying. However, it is clear that mucosal sensitivity exists also in the stomach. These endings have a multi-modal pattern of activation (Iggo, 1957; Davison, 1972a; Clarke and Davison, 1978; Blackshaw and Grundy, 1990b) and respond whenever the luminal contents differ from iso-osmotic or neutral pH. The relevance of the latter for an organ that secretes acid is unclear but could be related to the normal cytoprotective role of gastric mucous and blood flow. Only when the normal cytoprotective mechanisms breakdown would acid back diffuse to the afferent endings in the lamina propria. Vagal mucosal afferents are also highly sensitive to deformation of the mucosa, generating a burst of impulses when the mucosa is stroked or touched. In this respect, these endings will detect the particulate nature of gastric contents and may play a role in trituration of solids (Davison, 1972b). Spinal afferents may also be sensitive to luminal acid but these have not been characterised electrophysiologically. However, it is clear that noxious agents applied to the mucosa (including acid, especially when the protective mucous has been eliminated with alcohol) activate capsaicin sensitive spinal afferents which mediate reflex vasodilatation via the local release of CGRP and protect the mucosa against pending damage (Holzer, 1994). These spinal afferents are suggested to be the “first line of defence” against mucosal damage while at the same time can activate autonomic reflexes to initiate more widespread reactions.

Intestinal afferents and feedback mechanism Intestinal and especially duodenal afferents play an important feedback role in the control of gastric emptying. Thus the rate of gastric emptying is proportional to the nutrient content of the meal to the extent that calories, whether from fats, carbohydrates or proteins, are delivered to the duodenum at a relatively constant rate (Hunt and Stubbs, 1975). However, while it might appear that the duodenum monitors calories, in reality the sensitivity to different nutrients is tuned to produce a comparable end-organ response. The concept of nutrient receptors has been readily accepted in the literature on reflex control of gut function but direct electrophysiological evidence is, in fact, quite sparse. The group of Mei and his colleagues are the main advocates of specific nutrient receptor sensitivity with separate

MOTOR CONTROL OF THE STOMACH

83

publications describing afferent responses to fats (Melone and Mei, 1991), amino acids (Jeanningros and Mei, 1980) and carbohydrates (Mei, 1978). The mechanisms underlying this nutrient sensitivity have not been elucidated but recent studies have focussed on the role of enteroendocrine cells in afferent signal transduction both for intrinsic and extrinsic afferents. The principle is that enteroendocrine cells are secondary sense cells in the gastrointestinal mucosa which are strategically positioned to monitor luminal composition (Fujita et al., 1979). These cells have a apical tuft of microvilli which is proposed to “taste” the luminal contents and in response to an appropriate stimulus release the contents of storage granules across the baso-lateral membrane to stimulate afferent terminals in close proximity. Good evidence is available that CCK acts in such a fashion. Thus, it has been shown that a subset of vagal mucosal afferents are exquisitely sensitive to CCK acting on the CCKa receptor (Richards et al., 1996). Functional data has demonstrated that many of the actions of CCK including the feedback regulation of gastric emptying (Holzer et al., 1994) and the inhibition of feeding behaviour (Smith et al., 1981) are mediated by this vagal action, and, moreover that endogenous CCK released by luminal exposure to protein acts in the same way (Reidelberger, 1992). Data suggesting a similar role for 5-HT are also available in the context of its role in nausea and vomiting and the gastrointestinal consequence of these behaviours (Andrews et al., 1990). Certainly a subpopulation of afferents are exquisitely sensitive to 5-HT (Blackshaw and Grundy 1990a,b; Hillsley, Kirkup and Grundy, 1998). For the latter it appears that the 5-HT3 receptor on vagal mucosal afferents is an important target for 5-HT released from EC cells and this has led to the development of a new class of potent anti-emetics (Andrews et al., 1990). For intrinsic afferents in the small intestine, however, there is good evidence that 5-HT1p receptors may be more important and that 5-HT release is an obligatory step in the transduction of both mechanical and chemical luminal stimuli (Kirchgessner, Tamir and Gershon, 1992). For extrinsic intestinal afferents, however, neither 5-HT- nor CCK-sensitive vagal afferents depend entirely on enteroendocrine cell mediators for their afferent sensitivity since both mechanosensitivity and sensitivity to luminal acid persist after the receptors for these agents have been blocked. It appears therefore that there are multiple transduction pathways for mucosal afferents (Richards et al., 1996).

CENTRAL REFLEX PATHWAYS SPINAL REFLEXES From the dorsal root ganglion the central processes of splanchnic afferents enter the spinal cord and project rostro-caudally in Lissauer’s tract, with collaterals passing into lamina I laterally and medially around the dorsal horn (de Groat, 1986). The main terminal field is in lamina V with some afferent projections to the dorsal grey commissure, to the region around the central canal and to laminae I and V on the contralateral side of the spinal cord. Medial projections of splanchnic afferents extend into lamina I and also into the ipsilateral dorsal column. The substantia gelatinosa (lamina II), which is the terminal site for somatic afferents, does not receive a strong visceral afferent projection, indicating a distinct pattern of processing of visceral and somatic afferent information (Cervero, 1985). This

84

INNERVATION OF THE GASTROINTESTINAL TRACT

pattern of innervation is reflected in the distribution of fos immunoreactivity in the thoracic segments of the spinal cord following gastric distension (Traub, Sengupta and Gebhart, 1996). Second order neurones ascend the spinal cord in the same region as the homologous somatic afferents through pathways such as the spinothalamic tract, spinoreticular tract and dorsal columns (Downman and Evans, 1957; Kuo and De Groat, 1985). Projections into lamina VII may make connections with dendrites of preganglionic sympathetic neurones located in the IML nucleus or near the central canal with ascending and descending pathways projecting to preganglionic neurones in adjacent segments of the spinal cord. Preganglionic fibres to the prevertebral ganglia emerge from the thoraco-lumbar spinal cord in the ventral roots. Projections to the coeliac ganglion arise between T5 and T12 in the guinea-pig and are probably organised in the same way as those to the pelvic organs with visceral vasoconstrictor neurones receiving different inputs from those involved in motility regulation (Jänig, 1988). In most species section of the splanchnics has minimal effects on normal gastric motor function whereas laparotomy and surgical manipulation of the viscera provokes widespread inhibition of gastric tone and motility that depends upon an intact sympathetic innervation (see Furness and Costa, 1987). However, in the dog, section of the splanchnic nerves leads to an increase in gastric motility and gastric emptying suggesting a physiological role (Carlson, 1913; McCrea, 1926). In the ferret, this sympathetic influence on gastric tone is only apparent after prior section of the vagus nerve or after pretreatment with atropine suggesting that there may be functional interactions between vagal and splanchnic influences in the periphery (Andrews, Grundy and Lawes, 1981). Furthermore, this interaction may be bidirectional (McSwiney and Robson, 1931; Mamber and Gershon, 1979). Somatic inputs converge onto the same spinal neurones that receive a splanchnic input. These somatic inputs are generally nociceptive, which provides the basis for the convergence-projection theory of referred pain. However, their role in the control of gastrointestinal function cannot be dismissed, especially since there is functional evidence for splanchnic projections to the dorsal motor vagal nucleus where they can modulate the vagal outflow to the gastrointestinal tract (Grundy, Salih and Scratcherd, 1981; Barber and Yuan, 1993; Renehan et al., 1995; Traub, Sengupta and Gebhart, 1996). BRAINSTEM REFLEXES Pertinent to this discussion is the fact that the full repertoire of vago-vagal reflexes to the entire gastrointestinal tract is achieved by just a few thousand pre-ganglionic efferent fibres, the vagus being a predominantly afferent nerve (Pretchl and Powley, 1990). An added complication arises because there are at least two distinct functional populations of pre-ganglionic vagal motor neurone that supply the stomach (ignoring the sub-classification into motor and secretomotor): those that when stimulated cause contraction and others that cause gastric relaxation. The contractile and relaxatory responses to vagal stimulation are both mediated by C-fibre inputs to gastric myenteric neurones via predominantly nicotonic synapses. However, the contractile response is largely cholinergic and blocked by atropine, and relaxation mediated by the so-called non-adrenergic, non-cholinergic (NANC) pathway that involves NO and VIP (see earlier discussion). One talks of excitatory and

MOTOR CONTROL OF THE STOMACH

85

inhibitory vagal pathways but this is in fact short-hand for two pathways to enteric circuits which influence the release of excitatory or inhibitory transmitters at the neuro-muscular junction. Vagal excitatory or inhibitory reflexes therefore arise from the selective activation or suppression of brainstem neurones projecting to different myenteric circuits for contraction or relaxation. These pathways through the brainstem are probably hardwired for specific peripheral responses in much the same way as those described above for the polarised enteric circuits. The processing of afferent information in the brainstem in relation to vago-vagal reflex control of gastrointestinal function has been investigated in neurophysiological studies in which the efferent discharge in pre-ganglionic vagal fibres was recorded during the application of stimuli designed to activate different afferent sub-populations in different regions of the gastrointestinal tract. Earlier work in the field utilised extracellular single fibre recording techniques to obtain efferent signals on route to the periphery (Davison and Grundy, 1978; Grundy, Salih and Scratcherd, 1981). This approach treats the brainstem as a “black-box” and while providing valuable information on overall central processing could not detail the mechanisms involved. To overcome this problem there has been a move towards direct recordings from the cell bodies of nucleus tractus solitarius (NTS) and vagal pre-ganglionic motor neurones within the brainstem using extracellular and intracellular electrode techniques (Appia et al., 1986; Barber and Burks, 1983; Ewart and Wingate, 1984; McCann and Rogers, 1992; Renehan et al., 1995; Zhang, Fogel and Renehan, 1992, 1995; Fogel, Zhang and Renehan, 1996; Chu et al., 1993). The intracellular approach offers an additional advantage in that the injection of markers such as neurobiotin through the recording electrode enables the morphology of the neurone to be correlated with its electrophysiology. However, in the absence of specific neurochemical or anatomical “markers” for the excitatory and inhibitory vagal motorneurones, investigators have been forced to interpret functions from the behaviour of individual neuronal responses to afferent stimulation which give rise to “specific” end-organ responses. Spontaneous activity If one focuses initially on the pre-ganglionic vagal outflow then a striking feature of these neurones is that more than 90% of them generate a spontaneous discharge in the absence of any intentional stimulation. In most cases this spontaneous firing is irregular and of low frequency, less than 2 imps–1, while in others the discharge is modulated in phase with on-going contractile activity in the stomach. That this discharge is generated centrally is evident from experiments in which the vagal afferent input is severed during maintained efferent recording. In some cases discharge increased after vagotomy demonstrating a tonic inhibitory vagal afferent input, while in other cases the reverse was observed (Davison and Grundy, 1978; Grundy, Salih and Scratcherd, 1981). The presence of spontaneous activity enables gastrointestinal reflex responses to be generated by both increases and decreases in efferent discharge (see below). This spontaneous activity may be generated by neurones of the dorsal motor nucleus of the vagus (DMNV) themselves or may result from synaptic influences elsewhere in the brainstem or higher centres which contribute the diversity of transmitters which exist in the dense synaptic neuropil around these neurones. Brain regions with a prominent input to this region of the brainstem

86

INNERVATION OF THE GASTROINTESTINAL TRACT

include the medullary raphe nuclei, the paraventricular nucleus of the hypothalamus and the central nucleus of the amygdala. These connections provide the basis for emotional and behavioural effects, and in particular the effect of stress on gastrointestinal function through a number of mediators including thyrotropin-releasing hormone, corticotropinreleasing factor and CCK (Tache, Maeda-Hagiwara and Turkelson, 1987; Diop et al., 1991; Bueno, 1993, 1994; Martinez et al., 1997). Central corticotropin-releasing factor may also be involved in the post-surgical inhibition of gastric emptying (Barquist et al., 1996). Reflex modulation of the vagal outflow The modulatory influence of vagal afferent inputs to brainstem circuits is dramatically illustrated when these afferents are electrically stimulated. Firing is suspended for several seconds in some efferent neurones after a single afferent shock, while in others there are reflex-evoked action potentials. Thus reflex circuits include both inhibitory and excitatory mechanisms and in some neurones a complex picture of both inhibition and excitation occurs. In the case of inhibition, it is not possible to determine if this arises as a direct consequence of inhibitory vagal afferent synapses onto vagal pre-ganglionic neurones or through a relay in the NTS. The latter is more likely since the majority of gastrointestinal vagal afferents project to the NTS (Kalia and Mesulam, 1980; Altschuler et al., 1989). Moreover, in a number of electrophysiological studies the predominant vagal input to the NTS is excitatory while DMNV neurones show inhibition in response to the same stimulus (Rogers, McTigue and Hermann, 1995). However, a direct vagal afferent projection to the DMNV has been described while in addition the dendrites of pre-ganglionic vagal motorneurones extend into the NTS where mono-synaptic contacts have been observed electron microscopically (Rinaman et al., 1989). Functional evidence for mono-synaptic vagal reflexes comes from the observation that some vagal efferent fibres generate a single action potential at fixed latency following a single electrical stimulus to vagal afferent fibres in both the contralateral and ipsilateral branches of the cervical vagus nerve. The limited latency “jitter” and the security of the response in terms of the probability of firing are in keeping with a mono-synaptic connection and are in marked contrast to the less secure responses seen in the majority of efferent neurones (Blackshaw and Grundy, 1988). This observation emphasises the variable extent to which the incoming afferent information is processed within the vago-vagal pathway. The function of the mono-synaptic pathway and the modality of afferent information processed in this way is not known but by analogy with the mono-synaptic somatic reflex may be the basis on which other modulating influences are exerted. Of the various subnuclei that make up the NTS, the subnucleus gelatinosus, medial and commissural nuclei are the principal targets for gastric afferents, although an afferent projection to the area postrema has also been noted (Sawchenko, 1983). The projection to area postrema may be associated with emesis triggered from gastrointestinal afferents (Leslie, 1986). From the NTS, neurones project to other nuclei within the brainstem which subserve the vago-vagal reflexes controlling gastric function. Other pathways ascend through the mid-brain and reticular nuclei to higher centres. While these pathways are well established anatomically, the functional implications for gastrointestinal control are unknown, although some undoubtedly have a role in the control of food intake.

MOTOR CONTROL OF THE STOMACH

87

Convergent vagal inputs Neurones in the NTS that receive a gastric vagal input possess a striking horizontal orientation of their dendrites extending over a 600 µm from medial NTS to solitary tract and as such pass through the terminal zones of all gastrointestinal afferent inputs providing a morphological basis for the dramatic convergence seen in the vago-vagal reflex pathway (Rogers and McCann, 1993; Rogers, McTigue and Hermann, 1995). An individual efferent neurone can be modulated by afferent fibres supplying different regions of the gastrointestinal tract, supplying different types of sensory modality and follow different pathways to the brainstem (Blackshaw and Grundy, 1989). The relative proportion of vagal efferent neurone that are excited or inhibited by a given stimulus varies with the region of gut being stimulated and the level of stimulation. In the rat and ferret the ratio of efferent fibres showing excitation and inhibition following gastric distension was about 50:50. However, when progressive more distal regions of the gastrointestinal tract were stimulated, the inhibitory influences became dominant (Grundy, Salih and Scratcherd, 1981). In some cases the inhibition was complete and all ongoing efferent firing was suspended for the duration of the distension. In other cases the inhibition was only partial. Similarly variations in the degree of excitation were also observed which again reflects the extent to which afferent information is processed. The observation that some efferent fibres were excited by the same stimulus, and at the same time as others were inhibited, viewed in the light of inhibitory and excitatory motor pathways, led to the hypothesis that these two motor pathways were reciprocally controlled (Davison and Grundy, 1978; Miolan and Roman, 1974). Thus, during gastric relaxation there would be activation of the vagal inhibitory pathway and concomitant inhibition of the excitatory pathway, an effect which would sharpen the response of either alone. In a number of more recent studies on brainstem mechanisms, reflex suppression of the vagal excitatory drive to the stomach is suggested to be the major means of gastric relaxation. For example, when the afferent stimulus was localised to the antrum by distending a balloon inserted through the pylorus, over 90% of DMNV responses were inhibitory (McCann and Rogers, 1992). A similar preference towards inhibitory responses in the DMNV was described by Zhang, Fogel and Renehan (1992) and Renehan et al. (1995) following intestinal distension. Since, both antral and intestinal distension gives rise to a vagally mediated gastric relaxation, the conclusion from both studies was that reflex events were brought about by a decrease in ongoing vagal cholinergic tone to the gastrointestinal tract. In contrast, functional studies of adaptive gastric relaxation favour a dominant role for enteric inhibitory pathways (Grundy, Gharib-Naseri and Hutson, 1993a; De Ponti, Azpiroz and Malagelada, 1989; Desai, Sessa and Vane, 1991). However, antrally mediated relaxation may have a disinhibition of cholinergic mechanisms as a contributing factor (Krowicki and Harby, 1996). An interpretational difficulty for all of these electrophysiological studies is that the target organ of the neurones under investigation is unknown. One assumes that they supply predominantly the stomach because this region receives the densest vagal input. Both morphological and electrophysiological data confirm that most myenteric ganglia in the stomach are innervated by the vagus while many less are innervated in the small intestine (Schemann and Grundy, 1992; Berthoud Jedrzejewska and Powley, 1990). The

88

INNERVATION OF THE GASTROINTESTINAL TRACT

discrepancy between data from vagal recording and brainstem studies may arise from the ability to target the gastric motor innervation in the brainstem since different gastrointestinal regions are organotopically organised within the DMNV with the gastric projecting neurones located medially. In contrast, recording from the vagus inevitably means a more random selection of fibres. The convergence of afferent inputs onto individual vagal motor fibres is striking. For example, single efferent fibres have been shown to respond to distension of the stomach, duodenum and colon in the ferret (Grundy, Salih and Scratcherd, 1981). Following truncal vagotomy, the gastric input was greatly attenuated while that from the duodenum and colon persisted, indicating that afferents from these more distal regions reached the brainstem via a non-vagal pathway. Convergent inputs from muscle mechanoreceptors and multimodal mucosal chemoreceptors have also been shown (Blackshaw and Grundy, 1989). In general, the vagal input is dominant and mechanoreceptive inputs more powerful than chemoreceptive ones. Nevertheless, the outcome from this considerable afferent convergence is that a particular population of preganglionic fibres provides the efferent limb of a variety of reflexes that require the same end organ response. Gastric corpus relaxation, for example, occurs under a variety of circumstances including receptive and adaptive relaxation to accommodate a meal, following gastric emptying into the duodenum, and during emetic episodes. A single efferent pathway may mediate all these responses and could account for vagal control through a relatively small number of efferent fibres. Vagal motor fibres may also contribute to pathological conditions such as paralytic ileus (Abrahamsson, Glise and Glise, 1979). There is strong evidence that splanchnic afferents project to the dorsal vagal complex and can therefore influence transmission through vago-vagal reflex pathways. Vagal efferent fibres received a non-vagal input following distension of the small and large intestine which was generally inhibitory (Grundy, Salih and Scratcherd, 1981). In contrast an excitatory input to the DMNV was shown following intestinal distension which appeared to bypass the predominantly inhibitory route via the NTS (Renehan et al., 1995). The authors concluded that spinal projections may be direct to neurones of the DMNV an interpretation that might explain the way gastric responses to low threshold stimulation is often reversed when the intensity of stimulation increases. These opposing actions may therefore have their basis in differential inputs from vagal and splanchnic afferents. COMMAND NEURONES Enteric neurones provide the final pathway for both intrinsic and extrinsic reflexes regulating gastric motor function. The command neurone hypothesis was proposed to explain the widespread influence on GI function of just a few thousand preganglionic vagal fibres. This hypothesis suggests that only a few myenteric neurones are synaptically driven by the vagus and these innervated neurones are suggested to play a key role by activating specific microcircuits for contraction or relaxation. Morphological data has produced conflicting data with respect to the extent of the preganglionic input to gastric myenteric neurones. On the one hand this innervation is suggested to be sparse with few enteric neurones receiving direct extrinsic input, an observation in keeping with the command

MOTOR CONTROL OF THE STOMACH

89

hypothesis (Kirchgessner and Gershon, 1989) while on the other hand, vagal efferent terminals were widely distributed within the gastric myenteric plexus (Berthoud et al., 1990). In an extension to this study, Berthoud estimated that in the rat fundus vagal terminals are in close apposition to between 25% and 50% of gastric myenteric neurones although synaptic contacts are impossible to verify with this approach. However, c-fos expression in large numbers of myenteric neurones following vagal stimulation is consistent with an extensive direct synaptic input (Berthoud, 1997) or via polysynaptic pathways within the ENS. Approximately one third of “innervated” neurones were NADPH-diaphorase positive and therefore likely to be inhibitory neurones to the muscle (Berthoud, 1995). The two third NADPH-diaphorase negative neurones that were also vagally innervated would include cholinergic motor neurones providing a morphological basis for the dual vagal motor pathways to the gastric smooth muscle. This proportion of innervated neurones matches the proportion of NOS and ChAT positive cells in the guinea-pig myenteric plexus and suggests that there is no preferential target for the vagal input to the enteric nervous system, a conclusion similar to that of an earlier study in which VIP, GRP and ENK neurones were examined (Berthoud, 1996). Electrophysiological studies in the guinea-pig corpus would tend to support the view that vagal terminals make both divergent and convergent contacts with myenteric neurones (Schemann and Grundy, 1992). Rather than there being few, specialised myenteric neurones receiving synaptic input from vagal preganglionic fibres it was found that the majority of neurones responded to electrical vagal stimulation with nicotinic fEPSPs of which most led to the generation of action potentials. This is suggestive of a highly divergent vagal input with widespread modulatory influence over gastric myenteric neurotransmission. These vagally innervated myenteric neurones also received nicotinic inputs from other myenteric neurones. The ENS is therefore capable of autonomous integrative activity in order to generate reflexes which are dependent upon intrinsic circuits but which are modulated by extrinsic inputs from the vagus. A variety of synaptic responses could be distinguished in response to vagal stimulation. In most cases increasing the stimulus strength above threshold produced a graded EPSP at a fixed latency (<1 ms) until finally firing an action potential. This observation suggests individual neurones receive multiple vagal inputs recruited as the stimulus strength increases and since most cells in a particular ganglion had similar latency this further suggests a monosynaptic input to many neurones. A minority of neurones generated a single fEPSP triggered action potential in an all-or-none fashion when stimulus strength was increased while a third type of response consisted of multiple EPSP at different latencies. A small number of neurones generated a slow EPSP following a train of stimuli delivered to the vagus but inhibitory post-synaptic potentials were never observed. The prominent influence of vagal fibres on myenteric neurones in the stomach might explain the quite dramatic effect of acute vagotomy on gastric functions. Although the circuits responsible for initiating the reflexes reside within the myenteric plexus, these circuits appear to depend on acetylcholine provided by the extrinsic nervous system. Support for this view is the finding that coordinated reflex activity in the stomach may be restored after acute vagotomy by raising the level of neural activity by a background infusion of acetylcholine or carbachol (Grundy, Hutson and Scratcherd, 1986; Grundy, 1989; Papasova and Atanassova, 1984). The intramural nervous system gradually adapts after

90

INNERVATION OF THE GASTROINTESTINAL TRACT

vagotomy so that it can perform as an independent organiser and modulator of motor function. Therefore, changes in gastric motility observed after acute vagotomy are usually fully reversible with time (Atanassova, Kortezova and Papasova, 1986). Based on these results we propose a permissive action of the vagus on stomach motility. PERMISSIVE CONTROL OF ENTERIC REFLEX CIRCUITS The electrophysiological data above suggests that rather than “command neurones” in the stomach which, once activated, produced a stereotyped excitatory or inhibitory response, there is a divergent vagal input to the majority of gastric myenteric neurones. This type of innervation suggests that the vagus may play a more permissive role in regulating enteric reflexes. Consistent with this view is the convergence of intrinsic and vagal inputs to individual neurones. The sympathetic input also fits easily into this concept since noradrenaline acts presynaptically to suppress cholinergic fEPSPs evoked by both the vagus and intrinsic fibres (Schemann, 1991b). Since the vagal input is tonically active, one might predict that after vagotomy one would lose a facilitatory input to the enteric circuits leading to a deficit in reflex control. In addition, sympathetic presynaptic inhibition primarily shuts down cholinergic tone because α2 receptor-mediated inhibition of fEPSPs mostly affects cholinergic neurones (Table 2.2). Certainly, acute vagotomy has dramatic effects on gastric motor function causing a suppression of fed motor patterns in the conscious dog (Hall, El-Sharkawy and Diamant, 1986), an increase in corpus tone (Grundy, GharibNaseri and Hutson, 1993) and disorganised antral peristaltic activity leading to delayed gastric emptying (Malbert, Mathis and Laplace, 1995). These deficits in motor function may not merely reflect the loss of specific reflex pathways but also this loss of a facilitatory input to the enteric reflex circuits. In this respect restoring this vagal tone by electrical stimulation has been shown to restore the corpus relaxatory response to local distension necessary for gastric accommodation (Grundy, 1993) and the antral contractile response to distension (Grundy, Hutson and Scratcherd, 1986). After chronic vagotomy, the deficit in gastric reservoir function appears to be temporary indicating that the enteric nervous system can adapt to the loss of the potentiating effect of the vagus, possibly by up-regulation of the nitrergic innervation (Andrews and Bingham, 1990; Takahashi and Owyang, 1997). It is not clear how changes in the intramural nerves compensates for the loss of vagal input. The degree of up-or-down regulation, reorganisation or plasticity after loss of central inputs is not known. Although the gross morphological integrity of the intramural elements is preserved after vagotomy, degeneration of vagal fibres is accompanied by regressive transneural changes that are rapid in onset (1–3 days after vagotomy) are short lasting (5–7 days) and presumably fully reversible (Vaithilingham, Wong and Ling, 1986). There has also been described a dramatic decrease in the number of synapses after vagotomy (Filogama and Gabella, 1970). Upregulation of muscarinic receptors on corpus muscle has also been reported after vagotomy (Keshavarzian et al., 1990) although the mechanisms contributing to restored reservoir function appear not to be muscarinic in the ferret (Grundy, Gharib-Nasari and Hutson, 1993a). What is lost after vagotomy or splanchnectomy are the reflex pathways that link distant regions of gastrointestinal tract and which normally ensure coordination of gut function and reinforce local reflex circuits. The corpus and antrum, for example, are

MOTOR CONTROL OF THE STOMACH

91

linked by reflex pathways that can be activated in vitro (Yuan, Brookes and Costa, 1997). However, there a bidirectional reflexes through the vagus which provides a facilitatory input to the antrum and may couple the amplitude of antral peristaltic contractions to the level of corpus distension and inhibitory mechanism from the antrum to the corpus which may play a role in retropulsion which helps triturate solid material prior to emptying (Rudge et al., 1990). The polarisation of these reflexes are such that excitatory influences occur in a descending direction while ascending reflexes are inhibitory, in other words in the opposite direction to the way the enteric circuits are wired. Inhibition of both antrum and corpus by chyme in the duodenum has both a vagal and splanchnic component (see Meyer, 1994). Nevertheless, the decentralised stomach is still capable of coordinated activity and non-neural mechanisms are undoubtedly necessary to complete the picture of gastric motor control (Spencer et al., 1990; Van Lier Ribbink, Sarr and Tanaka, 1989).

CONCLUSION The fundamental mechanisms for gastric relaxation and peristalsis exists within the isolated stomach with the basic myogenic properties of smooth muscle being regulated by enteric reflex circuits activated either by intrinsic sensory neurones, the identity of which has still to be determined, or by axon collaterals of primary sensory neurones. However, in the intact state there is the potential for parasympathetic and sympathetic modulation which may influence the extent to which these local circuits give rise to the effector responses. The important contribution of enteric circuits for gastric reflex activity is supported by studies demonstrating a permissive role for the vagus, a concept that recognises the complex interactions between extrinsic nerves and intrinsic circuits. Thus, a basal preganglionic parasympathetic tone, and possible tone in local axon reflexes, is required to keep intrinsic circuits that are responsible for the behaviour of gastric motor function operational.

REFERENCES Abrahamsson, H., Glise, H. and Glise, K. (1979). Reflex suppression of gastric motility during laparotomy and gastro-duodenal nociceptive stimulation. Scandinavian Journal of Gastroenterology, 14, 101–106. Altschuler, S.M., Bao, X.M., Bieger, D., Hopkins, D.A. and Miselis, R.R. (1989). Viscerotopic representation of the upper alimentary tract in the rat: sensory ganglia and nuclei of the solitary and spinal trigeminal tracts. Journal of Comparative Neurology, 283, 248–268. Alvarez, W.C. and Mahoney, L.J. (1922). Action currents in stomach and intestine. American Journal of Physiology, 58, 476–493. Anderson, C.R., Edwards, S.L., Furness, J.B., Bredt, D.S. and Snyder, S.H. (1993). The distribution of nitric oxide synthase-containing autonomic preganglionic terminals in the rat. Brain Research, 614, 78–85. Andrews, P.L.R. and Bingham, S. (1990). Adaptation of the mechanisms controlling gastric motility following chronic vagotomy in the ferret. Experimental Physiology, 75, 811–825. Andrews, P.L.R., Grundy, D. and Scratcherd, T. (1980a). Reflex excitation of antral motility induced by gastric distension in the ferret. Journal of Physiology (London), 298, 79–84. Andrews, P.L.R., Grundy, D. and Scratcherd, T. (1980b). Vagal afferent discharge from mechanoreceptors in different regions of the ferret stomach. Journal of Physiology (London), 298, 513–524.

92

INNERVATION OF THE GASTROINTESTINAL TRACT

Andrews, P.L.R., Grundy, D. and Lawes, I.N.C. (1981). The role of the vagus and splanchnic nerves in the regulation of intragastric pressure in the ferret. Journal of Physiology (London), 307, 401–411. Andrews, P.L.R., Davis, C., Bingham, S., Davidson, H.I.M., Hawthorn, J. and Maskell, L. (1990). The abdominal visceral innervation and the emetic reflex pathways, pharmacology, and plasticity. Canadian Journal of Physiology and Pharmacology, 68, 325–345. Anthony, T.L. and Kreulen, D.L. (1990). Volume-sensitive synaptic input to neurons in guinea pig inferior mesenteric ganglion. American Journal of Physiology, 259, G490–G497. Appia, F., Ewart, W.R., Pittam, B.S. and Wingate, D.L. (1986). Convergence of sensory information from abdominal viscera in the rat brain stem. American Journal of Physiology, 251, G169–G175. Armitage, A.K. and Dean, A.C.B. (1966). The effects of pressure and pharmacologically active substances on gastric peristalsis in a transmurally stimulated rat stomach-duodenum preparation. Journal of Physiology (London), 182, 42–56. Atanassova, E., Kortezova, N. and Papasova, M. (1986). Changes in the reactivity of the intrinsic nervous system of the stomach to hexamethonium after transthoracic vagotomy. Acta Physiologica et Pharmacologica Bulgarica, 12, 20–26. Auer, J. (1907). Gastric peristalsis in rabbits under normal and some experimental conditions. American Journal Physiology, 28, 347–360. Baccari, M.C., Bertini, M. and Calamai, F. (1993). Effects of L-NG-nitro arginine on cholinergic transmission in the gastric muscle of the rabbit. NeuroReport, 4, 1102–1104. Baccari, M.C., Calamai, F. and Staderini, G. (1994). Modulation of cholinergic transmission by nitric oxide, VIP and ATP in the gastric muscle. NeuroReport, 5, 905–908. Back, N., Ahonen, M., Soinila, S., Kivilaakso, E. and Kiviluoto, T. (1995). Catecholamine-synthesising enzymes in the rat stomach. Histochemistry and Cell Biology, 104, 63–67. Barber, W.D. and Burks, T.F. (1983). Brain stem response to phasic gastric distension. American Journal of Physiology, 245, G242–G248. Barber, W.D. and Yuan, C.S. (1993). Gastric vagal-splanchnic interactions in the brainstem of the cat. Brain Research, 487, 1–8. Barber, W.D., Yuan, C.S. and Burks, T.F. (1989). What does the proximal stomach tell the brain? In Nerves and the Gastrointestinal Tract, edited by M.V. Singer and H. Goebell, pp. 721–725. Lancaster: MTP Press Limited. Barquist, E., Bonaz, B., Martinez, V., Rivier, J., Zinner, M.J. and Tache, Y. (1996). Neuronal pathways involved in abdominal surgery-induced gastric ileus in rats. American Journal of Physiology, 270, R888–R894. Bayliss, W.M. and Starling, E.H. (1899). The movements and innervation of the small intestine. Journal of Physiology, 24, 99–143. Beani, L., Bianchi, C. and Crema, A. (1971). Vagal non-adrenergic inhibition of guinea-pig stomach. Journal of Physiology, 217, 259–279. Belai, A. and Burnstock, G. (1994). Evidence for coexistence of ATP and nitric oxide in non-adrenergic, noncholinergic (NANC) inhibitory neurones in the rat ileum, colon and anococcygeus muscle. Cell and Tissue Research, 278, 197–200. Belai, A., Lefebvre, R.A. and Burnstock, G. (1991). Motor activity and neurotransmitter release in the gastric fundus of streptozotocin-diabetic rats. European Journal of Pharmacology, 194, 225–234. Bennet, A., Bucknell, A. and Dean, A.C.B. (1966). The release of 5-hydroxytryptamine from the rat stomach in vitro. Journal of Physiology (London), 182, 57–65. Berthoud, H.R. (1995). Anatomical demonstration of vagal input to nicotinamide acetamide dinucleotide phosphate diaphorase-positive (nitrergic) neurons in rat fundic stomach. Journal of Comparative Neurology, 358, 428–439. Berthoud, H.R. (1996). Morphological analysis of vagal input to gastrin releasing peptide and vasoactive intestinal peptide containing neurons in the rat glandular stomach. Journal of Comparative Neurology, 370, 61–70. Berthoud, H.R., Jedrzejewska, A. and Powley, T.L. (1990). Simultaneous labelling of vagal innervation of the gut and afferent projection from the visceral forebrain with DiI injected into the dorsal vagal complex in the rat. Journal of Comparative Neurology, 301, 65–79. Berthoud, H.R. and Powley, T.L. (1992). Vagal afferent innervation of the rat fundic stomach: morphological characterization of the gastric tension receptor. Journal of Comparative Neurology, 319, 261–276. Berthoud, H.R. and Patterson, L.M. (1996). Anatomical relationship between vagal afferent fibres and CCKimmunoreactive entero-endocrine cells in the rat small intestinal mucosa. Acta Anatomica (Basel), 156, 123–131. Berthoud, H.R., Patterson, L.M., Neumann, F. and Neuhuber, W.L. (1997). Distribution and structure of vagal afferent intraganglionic laminar endings (IGLEs) in the rat gastrointestinal tract. Anatomy and Embryology, 195, 183–191. Bickel, A. (1925). Der nervöse Mechanismus der Sekretion der Magendrüsen und der Muskelbewegung am Magendarmkanal. Ergebnisse der Physiologie, 24, 230–280.

MOTOR CONTROL OF THE STOMACH

93

Blackshaw, L.A. and Grundy, D. (1988). Reflex responses of vagal efferent fibres influenced by gastrointestinal mechanoreceptors to electrical afferent stimulation in the anaesthetized ferret. Quarterly Journal of Experimental Physiology, 73, 1001–1004. Blackshaw, L.A. and Grundy, D. (1989). Responses of vagal efferent fibres to stimulation of gastric mechanoand chemoreceptors in the anaesthetized ferret. Journal of the Autonomic Nervous System, 27, 39–45. Blackshaw, L.A. and Grundy, D. (1990a). Vagal afferent responses and gastric motility changes following close systemic 5-hydroxytryptamine (5-HT) injection. Journal of Gastrointestinal Motility, 2, 133. Blackshaw, L.A. and Grundy, D. (1990b). The effect of CCK-8 on two classes of gastroduodenal vagal afferent fibre. Journal of the Autonomic Nervous System, 31, 191–202. Blackshaw, L.A., Grundy, D. and Scratcherd, T. (1987). Vagal afferent discharge from gastric mechanoreceptors during contraction and relaxation of the ferret corpus. Journal of the Autonomic Nervous System, 18, 19–24. Blumberg, H., Haupt, P., Jänig, W. and Kohler, W. (1983). Encoding of visceral noxious stimuli in the discharge patterns of visceral afferents from the colon. Pflügers Archiv, 398, 33–40. Boeckxstaens, G.E., Pelckmans, P.A., De Man, J.G., Bult, H., Merman, A.G. and Van Maercke, Y.M. (1992). Evidence for a differential release of nitric oxide and vasoactive intestinal polypeptide by nonadrenergic noncholinergic nerves in the rat gastric fundus. Archives Internationales de Pharmacodynamie et de Therapie, 318, 107–115. Bornstein, J.C., Furness, J.B. and Kunze, W.A.A. (1994). Electrophysiological characterization of myenteric neurones, how do classification schemes relate? Journal of the Autonomic Nervous System, 48, 1–15. Brookes, S.J.H., Schemann, M. and Hennig, G.W. (1994). Distribution and characterization of circular muscle motor neurons in the guinea-pig gastric corpus. Journal of Gastroenterology and Hepatology, 9, A75. Brookes, S.J.H., Chen, B.N., Hodgson, W.M. and Costa, M. (1996). Characterization of excitatory and inhibitory motor neurons to the guinea pig lower esophageal sphincter. Gastroenterology, 111, 108–117. Brookes, S.J.H., Hennig, G.W. and Schemann, M. (1998). Identification of motor neurons to the circular muscle of the guinea pig gastric corpus. Journal of Comparative Neurology, 397, 268–280. Bueno, L. (1993). Involvement of brain CCK in the adaptation of gut motility to digestive status and stress: a review. Journal of Physiology (Paris), 87, 301–306. Bult, H., Boeckxstans, G.E., Pelkmans, P.A., Jordaens, F.H., Van Maercke, Y.M. and Herman, A.G. (1990). Nitric oxide as an inhibitory nonadrenergic noncholinergic neurotransmitter. Nature, 304, 346–347. Burnstock, G. (1972). Purinergic nerves. Pharmacology Reviews, 24, 509–581. Cannon, W.B. (1911). The nature of gastric peristalsis. American Journal of Physiology, 24, 250–266. Cannon, W.B. and Lieb, C. (1911). The receptive relaxation of the stomach. American Journal of Physiology, 29, 267–273. Carlson, A.J. (1913). Contributions to the physiology of the stomach. VII. The inhibitory reflex from the gastric mucosa. American Journal of Physiology, 32, 389–397. Cervero, F. (1985). Visceral nociception: peripheral and central aspects of visceral nociceptive systems. Philosophical Transactions of the Royal Society of London, 308, 325–337. Cervero, F. and Jänig, W. (1992). Visceral nociceptor: a new word order? Trends in Neurosciences, 15, 374–378. Cervero, F. and Tattersall, J.E.H. (1986). Somatic and visceral sensory integration in the thoracic spinal cord. In Visceral Sensation, Progress in Brain Research, vol. 67, edited by F. Cervero and J.F.B. Morrison, pp. 189–205. Amsterdam: Elsevier. Chaplin, S.B. and Duke, G.E. (1990). Effect of denervation of the myenteric plexus on gastroduodenal motility in turkeys. American Journal of Physiology, 259, G481–G489. Chaplin, S.B., Duke, G.E., Hunt, H. and Degernes, L.A. (1987). Chemical denervation of the myenteric plexus of the muscular stomach of turkeys. Comparative Biochemistry and Physiology, 88C, 201–207. Cheng, C.A., Geoghegan, J.G., Lawson, D.C., Berlangieri, S.U., Akwari, O. and Pappas, T.N. (1993). Central and peripheral effects of CCK receptor antagonists on satiety in dogs. American Journal of Physiology, 265, G219–G223. Christensen, J. and Rick, G.A. (1985). Nerve cell density in submucous plexus throughout the gut of cat and opossum. Gastroenterology, 89, 1064–1069. Chu, X., Zhang, X., Renehan, W.E., Beierwaltes, W.H. and Fogel, S. (1993). Intestinal distension alters vagal efferent activity and small intestinal transport in vivo. American Journal of Physiology, 264, G341–G350. Clark, G.D. and Davison, J.S. (1978). Mucosal receptors in the gastric antrum and small intestine of the rat with afferent fibres in the cervical vagus. Journal of Physiology, 274, 55–67. Code, C.F. and Marlett, J.A. (1975). The interdigestive myo-electric complex of the stomach and small bowel of dogs. Journal of Physiology (London), 246, 289–309. Costall, B., Naylor, R.J. and Tan, C.C.W. (1983). Field stimulation-induced responses of circular smooth muscle from guinea-pig stomach, characterisation of the receptor mechanisms involved. Naunyn-Schmiedebergs Archives Pharmacology, 323, 155–161.

94

INNERVATION OF THE GASTROINTESTINAL TRACT

Cottrell, D.F. and Iggo, A. (1984). Tension receptors with vagal afferent fibres in the proximal duodenum and pyloric sphincter of sheep. Journal of Physiology (London), 354, 457–475. Coulie, B., Tack, J., Maes, B., Geypens, B., De Roo, M. and Janssens, J. (1997). Sumatriptan, a selective 5-HT1 receptor agonist, induces a lag phase for gastric emptying of liquids in humans. American Journal of Physiology, 272, G902–G908. D’Amato, M., Curro, D. and Montuschi, P. (1992). Evidence for dual components in the non-adrenergic noncholinergic relaxation in the rat gastric fundus: role of endogenous nitric oxide and vasoactive intestinal polypeptide. Journal of the Autonomic Nervous System, 37, 175–186. Daniels, E.E. and Berezin, I. (1992). Interstitial cells of Cajal: are they major players in control of gastrointestinal motility. Journal of Gastrointestinal Motility, 4, 1–24. Davison, J.S. (1972a). Response of single vagal afferent fibres to mechanical and chemical stimulation of the gastric and duodenal mucosa in cats. Quarterly Journal of Experimental Physiology, 57, 405–416. Davison, J.S. (1972b). The electrophysiology of gastrointestinal chemoreceptors. Digestion, 7, 312–317. Davison, J.S. and Clarke, G.D. (1988). Mechanical properties and sensitivity to CCK of vagal gastric slowly adapting mechanoreceptors. American Journal of Physiology, 255, G55–G61. Davison, J.S. and Grundy, D. (1978). Modulation of single vagal efferent fibre discharge by gastrointestinal afferents in the rat. Journal of Physiology (London), 284, 69–82. De Beurme, F.A. and Lefebvre, R.A. (1988). Vasoactive intestinal polypeptide as possible mediator of relaxation in the rat gastric fundus. Journal of Pharm and Pharmacology, 40, 711–715. de Groat, W.C. (1986). Spinal cord projections and neuropeptides in visceral afferent neurons. In Visceral Sensation. Progress in Brain Research. Vol. 67, edited by F. Cervero and J.F.B. Morrison, pp.165–187. Amsterdam: Elsevier. De Ponti, F., Azpiroz, F. and Malagelada, J.R. (1989). Relaxatory responses of canine proximal stomach to esophageal and duodenal dissension: importance of vagal pathways. Digestive Diseases and Sciences, 34, 873–881. Delbro, D. and Lisander, B. (1982). Hexamethonium-resistant gastric contractions by stimulation of the vagal nuclei. An antidromic activation of vagal afferents? Acta Physiologica Scandinavica, 116, 1–6. Delbro, D., Fandriks, L., Rossel, S. and Folkers, K. (1983). Inhibition of antidromically induced stimulation of gastric motility by substance P receptor blockade. Acta Physiologica Scandinavica, 118, 9–316. Desai, K.M., Sessa, W.C. and Vane, J.R. (1991). Involvement of nitric oxide in the reflex relaxation of the stomach to accommodate food or fluid. Nature, 351, 477–479. Desai, K.M., Warner, T.D. and Vane, J.R. (1994). 5-HT3 receptors do not mediate vagally-induced relaxation or contraction of the isolated stomach of the guinea-pig. British Journal of Pharmacology, 111, 346–350. Dimaline, R., Miller, S.M., Evans, D., Noble, P.J., Brown, P. and Poat, J.A. (1995). Expression of immediate early genes in rat gastric myenteric neurones: a physiological response to feeding. Journal of Physiology (London), 488, 493–499. Diop, L., Pascaud, X., Junien, J.L. and Bueno, L. (1991). CRF triggers the CNS release of TRH in stress-induced changes in gastric emptying. American Journal of Physiology, 260, G39–G44. Dockray, G.J. and Sharkey, K.A. (1986). Neurochemistry of visceral afferent neurones. Progress in Brain Research, 67, 133–148. Dogiel, A.S. (1899). Über den Bau der Ganglien in den Geflechten des Darmes und der Gallenblase des Menschen und der Säugetiere. Archiv für Anatomie und Physiologie, Anatomische Abteilung Jahrgang 1899, 130–158. Downman, C.B.B. and Evans, M.I. (1957). The distribution of splanchnic afferents in the spinal cord of cat. Journal of Physiology (London), 137, 66–79. Ekblad, E., Rökaeus, A., Håkanson, R. and Sundler, F. (1985). Galanin nerve fibres in the rat gut: distribution, origin and projections. Neuroscience, 16, 355–363. Ewart, W.R. and Wingate, D.L. (1984). Central representation of arrival of nutrient in the duodenum. American Journal of Physiology, 246, G750–G756. Fahrenkrug, J., Haglund, U., Jodal, M., Lundgren, O., Olbe, L. and Shaffalitzky de Muckadell, O.B. (1978). Nervous release of vasoactive intestinal polypeptide in the gastrointestinal tract of the cats; possible physiological implications. Journal of Physiology, (London), 284, 291–305. Falempin, M., Ternaux, J.P., Palouzier, B. and Chamoin, M.C. (1989). Presence of cholinergic neurons in the vagal afferent system: involvement in a heterogenous reinnervation. Journal of the Autonomic Nervous System, 28, 243–250. Filogamo, G. and Gabella, G. (1970). Effects of extrinsic denervation on the synapses of the myenteric plexus. Journal de Microscopie, 9, 281–284. Fogel, R., Zhang, X. and Renehan, W.E. (1996). Relationships between the morphology and function of gastric and intestinal distention-sensitive neurons in the dorsal motor nucleus of the vagus. Journal of Comparative Neurology, 364, 78–91.

MOTOR CONTROL OF THE STOMACH

95

Forster, E.R. and Dockray, G.J. (1990). The effect of ondansetron on gastric emptying in the conscious rat. European Journal of Pharmacology, 191, 235–238. Forster, E.R. and Dockray, G.J. (1991). The role of calcitonin gene-related peptide in gastric mucosal projection in the rat. Experimental Physiology, 76, 623–626. Forster, E.R. and Southam, E. (1993). The intrinsic and vagal extrinsic innervation of the rat stomach contains nitric oxide. NeuroReport, 4, 275–278. Fuijita, T., Kobayashi, S., Muraki, S., Saton K. and Shimoji, K. (1979). In Gut peptides: secretion, function and clinical aspects, edited by A. Miyoshi, pp. 47–52. Tokyo, Kodansha: Elsevier. Furness, J.B. and Costa, M. (1987). In The Enteric Nervous System. Edinburgh, London, Melbourne and New York: Churchill Livingstone. Furness, J.B., Keast, J.R., Pompolo, S., Bornstein, J.C., Costa, M., Emson, P.C. et al. (1988). Immunohistochemical evidence for the presence of calcium-binding proteins in enteric neurons. Cell and Tissue Research, 252, 79–87. Furness, J.B., Lloyd, K.C.K., Sternini, C. and Walsh, J.H. (1991). Evidence that myenteric neurons of the gastric corpus project to both the mucosa and the external muscle: myectomy operations on the canine stomach. Cell and Tissue Research, 266, 475–481. Furness, J.B., Johnson, P.J., Pompolo, S. and Bornstein, J.C. (1995). Evidence that enteric motility reflexes can be initiated through entirely intrinsic mechanisms in the guinea-pig small intestine. Neurogastroenterology and Motility, 7, 89–96. Gabriel, R., Fekete, E. and Halasy, K. (1989). Consecutive diaphorase-acetylcholinesterase histochemistry in the myenteric plexus of frog stomach. Acta Histochemica, 85, 135–141. Gabriel, R., Timmermans, J.-P., Adriaensen, D., De Groodt-Lasseel, M.H.A. and Scheuermann, D.W. (1992). Morphological features of the myenteric plexus of the stomach of the axolotl, Ambystoma mexicanum, revealed by immunohistochemistry. Journal of Histochemistry, 24, 181–189. Geishauser, T., Reiche, D. and Schemann, M. (1998). In vitro motility disorders associated with displaced abomasum in dairy cows. Neurogastroenterology and Motility, 10, 395–401. Gilon, P., Tappaz, M. and Remacle, C. (1991). Localisation of GAD-like immunoreactivity in the pancreas and stomach of the rat and mouse. Histochemistry, 96, 35–365. Goltz, F. (1872). Studien über die Bewegung der Speiseröhre und des Magens des Frosches. Archiv für die gesamte Physiologie, 6, 616–647. Greef, K. and Holtz, P. (1956a). Untersuchungen am isolierten Magenpräparat. Archiv für experimentelle Pathologie und Pharmacologie, 227, 427–435. Greef, K. and Holtz, P. (1956b). Über die Natur des Überträgerstoffes vagal-motorischer Magenerregungen. Archiv für experimentelle Pathologie und Pharmakologie, 227, 559–565. Greef, H., Kasperat, H. and Osswald, W. (1962). Paradoxe Wirkungen der elektrischen Vagusreizung am isolierten Magen- und Herzvorhofpräparat des Meerschweinchens sowie deren Beeinflussung durch Ganglienblocker, Sympathicolytica, Reserpin und Cocain. Naunyn-Schmiedebergs Archiv für experimentelle Pathologie und Pharmakologie, 243, 528–545. Green, T. and Dockray, G.J. (1988). Characterization of the peptidergic afferent innervation of the stomach in the rat, mouse and guinea-pig. Neuroscience, 25, 181–193. Grider, J.R., Murthy, K.S., Jin, J.G. and Makhlouf, G.M. (1992). Stimulation of nitric oxide from muscle cells by VIP: prejunctional enhancement of VIP release. American Journal of Physiology, 262, G774–G778. Grider, J.R. and Jin, J.G. (1994). Distinct populations of sensory neurones mediate the peristaltic reflex elicited by muscle stretch and mucosal stimulation. Journal of Neuroscience, 14, 2854–2860. Groenewald, H.B. (1994). Neuropeptides in the myenteric ganglia and nerve fibres of the forestomach and abomasum of grey, white and black Karakul lambs. Onderstepoort Journal of Veterinary Research, 61, 207–213. Grundy, D. (1988). Speculations on the structure/function relationship for vagal and splanchnic afferent endings supplying the gastrointestinal tract. Journal of the Autonomic Nervous System, 22, 175–180. Grundy, D. (1989). Permissive regulation of gastrointestinal functions by the vagus nerve. In Nerves and the Gastrointestinal Tract, edited by M.F. Singer and H. Boebell, p. 732. Lancaster: MTP Press. Grundy, D. (1993). Extrinsic reflex pathways in the upper gastrointestinal tract, and their central processing. In Advances in the Innervation of the Gastrointestinal Tract, edited by G.E. Holle and J.D. Wood, pp. 539–546. Amsterdam: Excerpta Medica. Grundy, D., Salih, A.A. and Scratcherd, T. (1981). Modulation of vagal efferent fibre discharge by mechanoreceptors in the stomach. Journal of Physiology (London), 319, 43–52. Grundy, D., Hutson, D. and Scratcherd, T. (1986). A permissive role for the vagus nerves in the genesis of antroantral reflexes in the anaesthetized ferret. Journal of Physiology (London), 381, 377–384. Grundy, D. and Scratcherd, T. (1989). Sensory afferents from the gastrointestinal tract. In Handbook of Physiology, Section 6, The Gastrointestinal System, Vol. 1, Motility and Circulation Part 1, edited by J.D. Wood, pp. 593–620. Bethesda: American Physiology Society.

96

INNERVATION OF THE GASTROINTESTINAL TRACT

Grundy, D., Gharib-Naseri, M.K. and Hutson, D. (1993a). Plasticity in the gastric inhibitory innervation after immunisation against VIP and vagotomy in the ferret. American Journal of Physiology, 265, G432–G439. Grundy, D., Gharib-Naseri, M.K. and Hutson, D. (1993b). Role of nitric oxide and vasoactive intestinal polypeptide in vagally mediated relaxation of the gastric corpus in the anaesthetized ferret. Journal of the Autonomic Nervous System, 43, 241–246. Grundy, D., Schemann, M. and Hutson, D. (1993). Correlation between morphology and function of neuropeptide-Y (NPY) neurones in the ferret gastric corpus. Journal of Gastrointestinal Motility, 5, 194. Grundy, D. and Reid, K. (1994). The physiology of nausea and vomiting. In Physiology of the gastrointestinal tract, third edition, edited by L.R. Johnson, pp. 879–901. New York: Raven Press. Grundy, D., Bagaev, V. and Hillsley, K. (1995). Inhibition of gastric mechanoreceptor discharge by cholecystokinin in the rat. American Journal of Physiology, 268, G355–G360. Hall, K.E., El-Sharkawy, T.Y. and Diamant, N.E. (1986). Vagal control of canine postprandial upper gastrointestinal motility. American Journal of Physiology, 250, G501–G510. Hennig, G.W., Brookes, S.J.H. and Costa, M. (1997). Excitatory and inhibitory motor reflexes in the isolated guinea-pig stomach. Journal of Physiology (London), 501, 197–212. Hillsley, K., Schemann, M. and Grundy, D. (1992). α-Adrenoceptor modulation of neurally evoked circular muscle responses of the guinea pig stomach. Journal of the Autonomic Nervous System, 40, 57–62. Hillsley, K., Kirkup, A.J. and Grundy, D. (1998). Direct and indirect actions of 5-hydroxytryptamine on the discharge of mesenteric afferent fibres innervating the rat jejunum. Journal of Physiology, 506.2, 551–561. Hirst, G.D.S., Holman, M.E. and McKirdy, H.C. (1975). Two descending nerve pathways activated by distension of guinea-pig small intestine. Journal of Physiology (London), 244, 113–127. Hirst, G.D.S., Holman, M.E. and Spence, I. (1974). Two types of neurones in the myenteric plexus of the duodenum in the guinea-pig. Journal of Physiology (London), 236, 303–326. Hodgkiss, J.P. and Lees, G.M. (1983). Morphological studies of electrophysiologically-identified myenteric plexus neurons of the guinea-pig ileum. Neuroscience, 8, 593–608. Hofmeister, F. and Schütz, E. (1886). Ueber die automatischen Bewegungen des Magens. Archiv für experimentelle Pathologie und Pharmakologie, 20, 1–33. Holle, G.E. (1992). Gastric reflex circuitry in the control of pyloric function and emptying: extrinsic and intrinsic pathways. In Advances in the Innervation of the Gastrointestinal Tract. In Excerpta Medica, edited by G.E. Holle and J.D. Wood, pp. 547–559. Amsterdam, London, New York, Tokyo: Elsevier Science. Holle, G.E., Steinbach, E. and Forth, W. (1994). Intrinsic corporoantropyloric coordination of motility and gastric emptying. American Journal of Physiology, 266, G255–G262. Holzer, P. (1994). Spinal afferent nerves: sensory, afferent and effector functions. In Innervation of the Gut: Pathophysiological Implications, edited by Y. Tache, D.L. Wingate and T.F. Burks, pp. 123–136. Boca Raton, FL: CRC Press. Holzer, H.H., Turkelson, C.M., Solomon, T.E. and Raybould, H.E. (1994). Intestinal lipid inhibits gastric emptying via CCK and a vagal capsaicin-sensitive afferent pathway in rats. American Journal of Physiology, 267, G625–G629. Hoyle, C.H.V. and Burnstock. G. (1989). Neuromuscular transmission in the gastrointestinal tract. In Handbook of Physiology, Section 6: The Gastrointestinal System, Vol. I: Motility and Circulation, edited by S.G. Schulz, J.D. Wood and B.B. Rauner, pp. 435–464, Bethesda, MD: American Physiological Society. Hunt, J.N. and Stubbs, D.F. (1975). The volume and energy content of meals as determinants of gastric emptying. Journal of Physiology (London), 245, 209–225. Ichikawa, H. (1989). Effect of enteric nervous system on antiperistaltic discharge. Nippon Heikatsukin Gakkai Zasshi, 25, 113–124. Ichikawa, H. and Helke, C.J. (1996). Coexistence of calbindin D-28k and NADPH-diaphorase in vagal and glossopharyngeal sensory neurons of the rat. Brain Research, 735, 325–329. Iggo, A. (1957). Tension receptors in the stomach and urinary bladder. Journal of Physiology, 128, 593–607. Ito, S., Kurokawa, A., Ohga, A., Ohga, T. and Sawabe, K. (1990). Mechanical, electrical and cyclic nucleotide responses to peptide VIP and inhibitory nerve stimulation in rat stomach. Journal of Physiology, 430, 337–353. Jänig, W. (1988). Integration of gut function by sympathetic reflexes. In Bailliere’s Clinical Gastroenterology, edited by D. Grundy and N.W. Read, pp. 45–62. London: Bailliere Tindall. Jeanningros, R. and Mei, N. (1980). Donnes preliminaire sur la response des chemorecepteurs intestinaux aux acides amines. Reproduction, Nutrition, Development, 20, 1615–1619. Jensen, J. and Holmgren, S. (1991). Tachykinins and intestinal motility in different fish groups. General Comparative Endocrinology, 83, 388–396. Junquera, C., Azanza, M.J., Parra, P., Peg, M.T. and Garin, P. (1986). The enteric nervous system of Rana ridibunda. General Pharmacology, 17, 597–605.

MOTOR CONTROL OF THE STOMACH

97

Kalia, M. and Mesulam, M.M. (1980). Brain stem projections of sensory and motor components of the vagus complex in the cat. II. Laryngeal, tracheobronchial, pulmonary, cardiac and gastrointestinal branches. Journal of Comparative Neurology, 193, 467–508. Katsoulis, S., Schmidt, W.E., Schwarzhoff, R., Folsch, U.R., Jin, J.G., Grider, J.R. and Makhlouf, G.M. (1996). Inhibitory transmission in guinea pig stomach mediated by distinct receptors for pituitary adenylate cyclase-activating peptide. Journal of Pharmacology and Experimental Therapeutics, 278, 199–204. Keast, J.R., Furness, J.B. and Costa, M. (1985). Distribution of certain peptide-containing nerve fibres and endocrine cells in the gastrointestinal mucosa of five mammalian species. Journal of Comparative Neurology, 236, 403–422. Keinke, O., Schemann, M. and Ehrlein, H.J. (1984). Mechanical factors regulating gastric emptying of viscous nutrient meals in dogs. Quarterly Journal of Experimental Physiology, 69, 781–795. Kelling, G. (1903). Untersuchungen über die Spannungszustände der Bauchwand, der Magen- und der Darmwand. Zeitschrift für Biologie, 26, 161–258. Kelly, K.A. and Code, C.F. (1971). Canine gastric pacemaker. American Journal of Physiology, 220, 112–118. Keshavarzian, A., Steck, T.B., Conway, D., Gordon, J.H. and Fields, J.Z. (1990). Canine gastric muscarinic receptors up-regulate after vagotomy. Digestive Diseases and Sciences, 35, 449–452. King, B.F. and Szurszewski, J.H. (1984). Intracellular recordings from vagally innervated intramural neurons in opossum stomach. American Journal of Physiology, 246, G209–G212. Kirchgessner, A.L. and Gershon, M.D. (1989). Identification of vagal efferent fibres and putative target neurons in the enteric nervous system of the rat. Journal of Comparative Neurology, 285, 38–53. Kirchgessner, A.L., Tamir, H. and Gershon, M.D. (1992). Identification and stimulation by serotonin of intrinsic sensory neurons of the submucosal plexus of the guinea pig gut: activity-induced expression of Fos immunoreactivity. Journal of Neuroscience, 12, 235–248. Klee, P. (1912). Der Einfluß der Vagusreizung auf den Ablauf der Verdauungsbewegungen. Pflügers Archives für die gesamte Physiologie, 145, 557. Kolossow, N.G. and Milochin, A.A. (1963). Morphologischer Nachweis der afferenten (sensiblen) Innervation der peripheren Neuronedes vegetativen Nervensystems. Zeitschrift für Mikroskopische Anatomie und Forschung, 70, 427–464. Krantis, A. and Webb, T. (1989). Autoradiographic localization of [3H] γ-aminobutyric acid in neuronal elements of the rat gastric antrum and intestine. Journal of the Autonomic Nervous System, 29, 41–48. Kreulen, D.L., Muir, T.C. and Szurszewski, J.H. (1983). Peripheral sympathetic pathways to gastroduodenal region in the guinea pig. American Journal of Physiology, 245, G369–G375. Krowicki, Z.K. and Hornby, P.J. (1996). Contribution of acetylcholine, vasoactive intestinal polypeptide and nitric oxide to CNS-evoked vagal gastric relaxation in the rat. Neurogastroenterology and Motility, 8, 307–317. Kunze, W.A., Bornstein, J.C. and Furness, J.B. (1995). Identification of sensory nerve cells in a peripheral organ (the intestine) of a mammal. Neuroscience, 66, 1–4. Kuo, D.C. and deGroat, W.C. (1985). Primary afferent projections of the major splanchnic nerve to the spinal cord and gracile nucleus of the cat. Journal of Comparative Neurology, 231, 421–434. Kuroda, S. (1924). Pharmakodynamische Studien zur Frage der Magenmotilität. Zeitschrift für die experimentelle Medizin, 39, 343–354. Kyosola, K., Veijola, L. and Rechardt, L. (1975). Cholinergic innervation of the gastric wall of the cat. Histochemistry, 44, 23–30. Langley, J.N. (1921). The Autonomic Nervous System, Part 1. Cambridge: W. Heffer and Sons. Langley, J.N. and Magnus, R. (1905). Some observations of movements of the intestine before and after degenerative section of the mesenteric nerves. Journal of Physiology (London), 33, 34–51. Lee, Y., Shiosaka, S., Hayashi, N. and Tohyama, M. (1986). The presence of vasoactive intestinal polypeptidelike immunoreactive structures projecting from the myenteric ganglion of the stomach to the celiac ganglion revealed by a double-labelling technique. Brain Research, 382, 392–394. Lefebvre, R.A., deVries, A. and Smits, G.J. (1992). Influence of vasoactive intestinal polypeptide and N G-nitroL-arginine methylester on cholinergic neurotransmission in the rat gastric fundus. European Journal of Pharmacology, 221, 235–242. Leslie, R.A. (1986). Comparative aspects of the area postrema: fine structural considerations help to determine function. Cellular and Molecular Neurobiology, 6, 95–119. Li, C.G. and Rand, M.J. (1990). Nitric oxide and vasoactive intestinal polypeptide mediate non-adrenergic, non-cholinergic inhibitory transmission to smooth muscle of the rat gastric fundus. European Journal of Pharmacology, 191, 303–309. Liu, M.T., Rothstein, J.D., Gershon, M.D. and Kirchgessner, A.L. (1997). Glutamatergic enteric neurons. Journal of Neuroscience, 17, 4764–4784.

98

INNERVATION OF THE GASTROINTESTINAL TRACT

Lüderitz, C. (1891). Das motorische Verhalten des Magens bei Reizung seiner äußeren Fläche. Archiv für die gesamte Physiologie, 49, 158–174. Macrae, I.M., Furness, J.B. and Costa, M. (1986). Distribution of subgroups of noradrenaline neurones in the coeliac ganglion of the guinea-pig. Cell and Tissue Research, 244, 173–180. Magnus, R. (1908). Die Bewegungen des Verdauungskanals. Ergebnisse. Physiologie, 7, 27–64. Malbert, C.H., Mathis, C. and Laplace, J.P. (1995). Vagal control of pyloric resistance. American Journal of Physiology, 269, G558–G569. Mamber, L. and Gershon, M.D. (1979). A reciprocal adrenergic-cholinergic axo-axonic synapse in the mammalian gut. American Journal of Physiology, 236, E738–E745. Mann, P.T., Furness, J.P., Pompolo, S. and Mader, M. (1995). Chemical coding of neurons that project from different regions of intestine to the coeliac ganglion of the guinea pig. Journal of the Autonomic Nervous System, 56, 15–25. Martinez, V., Rivier, J., Wang, L. and Tache, Y. (1997). Central injection of a new corticotropin-releasing factor (CRF) antagonist, astressin, blocks CRF- and stress-related alterations of gastric and colonic motor function. Journal of Pharmacology and Experimental Therapeutics, 280, 754–760. Mawe, G.M., Branchek, T.A. and Gershon, M.D. (1989). Blockade of 5-HT-mediated enteric slow EPSPs by BRL 24924: gastrokinetic effects. American Journal of Physiology, 257, G386–G396. Mawe, G.M., Schemann, M., Wood, J.D. and Gershon, M.D. (1989). Immunocytochemical analysis of potential neurotransmitters present in the myenteric plexus and muscular layers of the corpus of the guinea pig stomach. Anatomical Record, 224, 431–442. Mazet, B., Niolan, J.P., Niel, J.P. and Roman, C. (1994). New insights into the organisation of a gastroduodenal inhibitory reflex by the coeliac plexus. Journal of the Autonomic Nervous System, 46, 135–146. McCann, M.J. and Rogers, R.C. (1992). Impact of antral mechanoreceptor activation on the vago-vagal reflex in the rat: Functional zonation of responses. Journal of Physiology (London), 453, 401–411. McCrea, E.D.A. (1926). The nerves of the stomach and their relation to surgery. British Journal of Surgery, 13, 621–648. McIntosh, C.H.S., Dadgar, A. and Kwok, Y.N. (1992). Cholinergic stimulation of neuropeptide Y secretion from the isolated perfused rat stomach. Regulatory Peptides, 39, 83–94. McSwiney, B.A. and Wadge, W.J. (1928). Effects of variations in intensity and frequency on the contractions of the stomach obtained by stimulation of the vagus. Journal of Physiology (London), 65, 350. McSwiney, B.A. and Robson, J.M. (1931). The sympathetic innervation of the stomach. III. The interaction of the vagus and sympathetic nerves. Journal of Physiology (London), 69, 141–150. Mei, N. (1978). Vagal glucoreceptors in the small intestine of the cat. Journal of Physiology (London), 282, 485–506. Mei, N. (1985). Intestinal chemosensitivity. Physiological Reviews, 65, 211–237. Melone, J. and Mei, N. (1991). Intestinal effects of the products of lipid digestion on gastric electrical activity in the cat. Possible involvement of vagal intestinal receptors sensitive to lipids. Gastroenterology, 100, 380–387. Meulemans, A.L., Helsen, L.F. and Schuurkes, J.A.J. (1993a). The role of nitric oxide (NO) in 5-HT-induced relaxations of the guinea-pig stomach. Naunyn-Schmiedeberg’s Archives of Pharmacology, 348, 424–430. Meulemans, A.L., Helsen, L.F. and Schuurkes, J.A.J. (1993b). Role of NO in vagally-mediated relaxations of guinea-pig stomach. Naunyn Schmiedebergs Archives of Pharmacology, 347, 225–230. Meulemans, A.L., Eelen, J.G. and Schuurkes, J.A.J. (1995). NO mediates gastric relaxation after brief vagal stimulation in anesthetized dogs. American Journal of Physiology, 269, G255–G261. Meyer, E.A. (1994). The physiology of gastric storage and emptying. In Physiology of the Gastrointestinal Tract, 3rd edition, edited by L.R. Johnson, pp. 929–976. New York: Raven Press. Michel, K., Sann, H., Schaaf, C. and Schemann, M. (1997). Subpopulations of gastric myenteric neurons are differentially activated via distinct serotonin receptors: projection, neurochemical coding and functional implications. Journal of Neuroscience, 17, 8009–8017. Miolan, J.P. and Roman, C. (1974). Descharge unitaire des fibres vagales efferentes lors de la relaxation receptive de l’estomac du chien. Journal of Physiologie (Paris), 68, 693–704. Miolan, J.P. and Niel, J.P. (1996). The mammalian sympathetic prevertebral ganglia: Integrative properties and role in the nervous control of digestive tract motility. Journal of the Autonomic Nervous System, 58, 125–138. Morrison, J.F.B. (1977). The afferent innervation of the gastrointestinal tract. In Nerves and the Gut, edited by F.P. Brooks and P.W. Evers, pp. 297–326. New Jersey: Thorofare. Müller, L.R. (1920). Die Innervation des Magens: Anatomischer Teil. In Das Vegetative Nervensystem, edited by L.R. Müller, pp.127–134. Berlin: Springer. Muren, A. (1957). Influence of the vagal innervation on gastric motor responses to adrenaline and noradrenaline. Acta Physiologica Scandinavica, 39, 195–202.

MOTOR CONTROL OF THE STOMACH

99

Murthy, K.S., Zhang, K.M., Jin, J.G., Grider, J.R. and Makhlouf, G.M. (1993). VIP-mediated G protein-coupled Ca-influx activates a constitutive NOS in dispersed gastric muscle cells. American Journal of Physiology, 265, G660–G671. Nishi, S. and North, R.A. (1973). Intracellular recording from the myenteric plexus of the guinea pig ileum. Journal of Physiology (London), 231, 471–491. Ohta, T., Nakazato, H. and Ohga, A. (1985). Reflex control of the gastric motility by the vagus and splanchnic nerves in the guinea pig in vivo. Journal of the Autonomic Nervous System, 14, 137–149. Paintal, A.S. (1954). A study of gastric stretch receptors: Their role in the peripheral mechanism of satiation of hunger and thirst. Journal of Physiology, 121, 255–270. Pan, H.L. and Longhurst, J.C. (1996). Ischaemia-sensitive sympathetic afferents innervating the gastrointestinal tract function as nociceptors in cats. Journal of Physiology (London), 492, 841–850. Papasova, M. and Atanassova, E. (1984). Electrical activity of the stomach: Nervous control. Scandinavian Journal of Gastroenterology, 19, 43–53. Paton, W.D.M. and Vane, J.R. (1963). An analysis of the responses of the isolated stomach to electrical stimulation and to drugs. Journal of Physiology (London), 165, 10–46. Pfannkuche, H., Reiche, D., Sann, H. and Schemann, M. (1998). Different subpopulations of cholinergic and nitrergic myenteric neurones projecting to mucosa and circular muscle of the guinea-pig gastric fundus. Cell and Tissue Research, 292, 463–475. Plourde, V., Wong, H.C., Walsh, J.H., Raybould, H.E. and Tache, Y. (1993). CGRP antagonists and capsaicin on celiac ganglia partly prevent postoperative gastric ileus. Peptides, 14(6), 1225–1229. Poos, F. (1923). Zur Differenzierung der Magenfunktion hinsichtlich Reizbildung, Erregungsleitung und Tonus. Pflügers Archiv für die gesamte Physiologie, 201, 83–100. Powley, T.L., Holst, M.C., Boyd, D.B. and Kelly, J.B. (1994). Three-dimensional reconstructions of autonomic projections to the gastrointestinal tract. Microscopy Research and Technique, 29, 297–309. Pretchl, J.C. and Powley, T.L. (1990). The fiber composition of the abdominal vagus of the rat. Anatomy and Embryology, 181, 101–115. Radke, R., Stach, W. and Weiss, R. (1980). Innervation of the gastric wall related to acid secretion: a light and electron microscopy study on rats, rabbits and guinea-pigs. Acta Biologica Medica Germanica, 39, 687–696. Raybould, H.E., Gayton, R.J. and Dockray, G.J. (1988). Mechanisms of action of peripherally administered cholecystokinin octapeptide on brain stem neurons in the rat. Journal of Neuroscience, 8, 3018–3024. Reiche, D. and Schemann, M. (1998). Ascending choline acetyltransferase and descending nitric oxide synthase immunoreactive neurones of the myenteric plexus project to the mucosa of the guinea pig gastric corpus. Neuroscience Letters, 241, 61–64. Reiche, M., Pfannkuche, H. and Schemann, M. (1997). Identification of myenteric neurones innervating the mucosa of the guinea pig stomach. In Proceedings of the XXXIII International Congress of Physiological Sciences, P05014. Reidelberger, R.D. (1992). Abdominal vagal mediation of the satiety effects of exogenous and endogenous cholecystokinin in rats. American Journal of Physiology, 263, R1354–R1358. Renehan, W.E., Zhang, X., Beierwaltes, W.H. and Fogel, R. (1995). Neurons in the dorsal motor nucleus of the vagus may integrate vagal and spinal information from the GI tract. American Journal of Physiology, 268, G780–G790. Richards, W., Hillsley, K., Eastwood, C. and Grundy, D. (1996). Sensitivity of vagal mucosal afferents to cholecystokinin and its role in afferent signal transduction in the rat. Journal of Physiology (London), 497, 473–481. Rinaman, L., Card, J.P., Schwaber, J.S. and Miselis, R.R. (1989). Ultrastructural demonstration of a gastric monosynaptic vagal circuit in the nucleus of the solitary tract in the rat. Journal of Neuroscience, 9, 1985–1996. Rogers, R.C. and McCann, M.J. (1993). Intramedullary connections of the gastric region in the solitary nucleus: A biocytin histochemical tracing study in the rat. Journal of the Autonomic Nervous System, 42, 119–130. Rogers, R.C., McTigue, D.M. and Hermann, G.E. (1995). Vagovagal reflex control of digestion: Afferent modulation by neural and “endoneurocrine” factors. American Journal of Physiology, 268, G1–G10. Rudge, L., Grundy, D., Hutson, D., Scratcherd, T. and Kerrigan, D. (1990). Reflex co-ordination of corporal and antral contractions in the conscious dog. Experimental Physiology, 75, 801–809. Salzman, A.L. (1995). Nitric oxide in the gut. New Horizons, 3, 352–364. Sann, H., Hoppe, S., Baldwin, L., Grundy, D. and Schemann, M. (1998). Presence of putative neurotransmitters in the myenteric plexus of the gastrointestinal tract and in the musculature of the urinary bladder of the ferret. Neurogastroenterology and Motility, 10, 35–47. Sarna, S.K. (1993). Gastrointestinal longitudinal muscle contractions. American Journal of Physiology, 265, G156–G164.

100

INNERVATION OF THE GASTROINTESTINAL TRACT

Sawchenko, P.E. (1983). Central connections of the sensory and motor nuclei of the vagus nerve. Journal of the Autonomic Nervous System, 9, 13–26. Schemann, M. (1990). Neurophysiologie des enterischen Nervensystems. Zeitschrift für Verdauungskrankheiten, 6, 183–192. Schemann, M. (1991a). 5-Hydroxytryptamine-mediated responses in myenteric neurons of the guinea-pig gastric corpus: effect of ICS 205930 and cisapride. Journal of Gastrointestinal Motility, 3, 255–262. Schemann, M. (1991b). Excitatory and inhibitory effects of norepinephrine on myenteric neurons of the guineapig gastric corpus. Pflügers Archiv, 418, 575–580. Schemann, M. (1992). Characteristics of myenteric neurons in the gastric corpus. In Advances in the Innervation of the Gastrointestinal Tract. In Excerpta Medica, edited by G.E. Holle and J.D. Wood, pp. 147–156. Amsterdam, London, New York, Tokyo: Elsevier Science. Schemann, M. and Wood, J.D. (1989a). Electrical behaviour of myenteric neurons in the gastric corpus of the guinea-pig. Journal of Physiology (London), 417, 501–518. Schemann, M. and Wood, J.D. (1989b). Synaptic behaviour of myenteric neurons in the gastric corpus of the guinea pig. Journal of Physiology (London), 417, 519–535. Schemann, M. and Kayser, H. (1991). Effects of tachykinins on myenteric neurons of the guinea pig gastric corpus: involvement of NK-3 receptors. Pflügers Archiv, 419, 566–571. Schemann, M. and Grundy, D. (1992). Electrophysiological identification of vagally innervated enteric neurons in guinea pig stomach. American Journal of Physiology, 263, G709–G718. Schemann, M. and Tamura, K. (1992). Presynaptic inhibitory effects of the peptides NPY, PYY and PP on nicotinic EPSPs in guinea-pig gastric myenteric neurones. Journal of Physiology (London), 451, 79–89. Schemann, M. and Schaaf, C. (1995). Differential projection of cholinergic and nitroxidergic neurons in the myenteric plexus of the guinea pig stomach. American Journal of Physiology, 269, G186–G195. Schemann, M., Schaaf, C. and Mäder, M. (1995). Neurochemical coding of enteric neurons in the guinea pig stomach. Journal of Comparative Neurology, 353, 161–178. Scheuermann, D.W., Gabriel, R., Timmermans, J.-P., Adriaensen, D. and de Groodt-Lasseel, M.H.A. (1991). The innervation of the gastrointestinal tract of a chelonian reptile, Pseudemys scripta elegans. II. Distribution of neuropeptides in the myenteric plexus. Histochemistry, 95, 403–411. Schultz, P. (1897a). Die längsgestreifte (glatte) Musculatur der Wirbeltiere. II. Ihre Verrichtung. Archiv für Anatomie und Physiologie, 307–321. Schultz, P. (1897b). Zur Physiologie der längsgestreiften (glatten) Muskeln: Spontane Bewegungen, Tonus, Peristaltik. Archiv für Anatomie und Physiologie, 322–328. Schulzberg, M., Hökfelt, T., Nilsson, G., Terenius, L., Rehfeld, J.F., Brown, M. et al. (1980). Distribution of peptide- and catecholamine-containing neurons in the gastro-intestinal tract of rat and guinea-pig: immunohistochemical studies with antisera to substance P, vasoactive intestinal polypeptide, enkephalins, somatostatin, gastric/cholecystokinin, neurotensin and dopamine α-hydroxylase. Neuroscience, 5, 689–744. Schulze-Delrieu, K. (1983). Volume accommodation by distension of gastric fundus (rabbit) and gastric corpus (cat). Digestive Diseases and Sciences, 28, 625–632. Schulze-Delrieu, K. (1990). Exciting news on gastric neurons. Gastroenterology, 99, 562–564. Schulze-Delrieu, K., Percy, W.H., Ren, J., Shirazi, S.S. and von Derau, K. (1989). Evidence for inhibition of opossum LES through intrinsic gastric nerves. American Journal of Physiology, 256, G198–G205. Schütz, E. (1886). Über die Einwirkung von Arzneistoffen auf die Magenbewegungen. Archiv für experimentelle Pathologie und Pharmakologie, 27, 341–370. Schwartz, G.J., McHugh, P.R. and Moran, T.J. (1993). Gastric loads and CCK synergistically stimulate rat gastric vagal afferents. American Journal of Physiology, 265, R872–R876. Sengupta, J.M. and Gebhart, G.F. (1994). Gastrointestinal afferent fibres and sensation. In Physiology of the Gastrointestinal Tract, 3rd edition, edited by L.R. Johnson, pp. 483–519. New York: Raven Press. Sick, K. and Tedesko, F. (1908). Studien über die Magenbewegung mit besonderer Berücksichtigung der Ausdehnungsfähigkeit des Hauptmagens (Fundus). Deutsches Archiv für klinische Medizin, 92, 416–451. Skad-Nielsen, T., Poulsen, S.S. and Holst, J.J. (1987). Immunohistochemical detection of ganglia in the rat stomach serosa, containing neurons immunoreactive for gastrin-releasing peptide and vasoactive intestinal peptide. Histochemistry, 87, 47–52. Smith, G.P., Jerome, C., Cushin, B.J., Eterno, R. and Simansky, K.J. (1981). Abdominal vagotomy blocks the satiety effect of cholecystokinin in the rat. Science, 213, 1036–1037. Smith, T.K. and Furness, J.B. (1988). Reflex changes in circular muscle activity elicited by stroking the mucosa: an electrophysiological analysis in the isolated guinea-pig ileum. Journal of the Autonomic Nervous System, 25, 205–218. Smith, T.K., Bornstein, J.C. and Furness, J.B. (1990). Distension-evoked ascending and descending reflexes in the circular muscle of guinea-pig ileum: an intracellular study. Journal of the Autonomic Nervous System, 29, 203–217.

MOTOR CONTROL OF THE STOMACH

101

Smith, T.K., Bornstein, J.C. and Furness, J.B. (1992). Convergence of reflex pathways excited by distension and mechanical stimulation of the mucosa onto the same myenteric neurons of the guinea pig small intestine. Journal of Neuroscience, 12, 1502–1510. Spencer, M.P., Sarr, J.G., Soper, N.J. and Hakim, N.S. (1990). Jejunal regulation of gastric motility patterns: effect of extrinsic neural continuity to stomach. American Journal of Physiology, 258, G32–G37. Sternini, C. and Anderson, K. (1992). Calcitonin gene-related peptide containing neurones supplying the rat digestive system: differential distribution and expression pattern. Somatosensory and Motor Research, 9, 45–49. Sternini, C., DeGiorgio, R. and Furness, J.B. (1992). Calcitonin gene-related peptide neurons innervating the canine digestive system. Regulatory Peptides, 42, 15–26. Sykes, R.M., Spyer, K.M. and Izzo, P.M. (1997). Demonstration of glutamate immunoreactivity in vagal sensory afferents in the nucleus tractus solitarius of the rat. Brain Research, 762, 1–11. Szabat, E., Soinala, S., Happola, O., Linnala, A. and Virtanen, I. (1992). A new monoclonal antibody against the GABA-protein conjugate shows immunoreactivity in sensory neurons of the rat. Neuroscience, 47, 409–420. Szurszewski, J.H. (1987). Electrophysiological basis of gastrointestinal motility. In Physiology of the Gastrointestinal Tract, 2nd edition, edited by L.R. Johnson, pp. 383–422. New York: Raven Press. Szurszewski, J.H. and Miller, S.M. (1994). Physiology of prevertebral ganglia. In Physiology of the Gastrointestinal ract, 3rd edition, edited by L.R. Johnson, pp. 795–877. New York: Raven Press. Tache, Y., Maeda-Hagiwara, M. and Turkelson, C.M. (1987). Central nervous system action of corticotropinreleasing factor to inhibit gastric emptying in rats. American Journal of Physiology, 253, G241–G245. Tache, Y. and Yang, H. (1994). Role of medullary TRH in the vagal regulation of gastric function. In Innervation of the Gut: Pathophysiological Implications, edited by Y. Tache, D.L. Wingate and T.F. Burks, pp. 67–88. Boca Raton, FL: CRC Press. Tack, J.F. (1995). Georges Brohee Prize. Motilin and the enteric nervous system in the control of interdigestive and postprandial gastric motility. Acta Gastroenterologica Belgica, 58, 21–30. Tack, J.F. and Wood, J.D. (1991). Synaptic behaviour in the myenteric plexus of the guinea-pig gastric antrum. Journal of Physiology (London), 445, 389–406. Tack, J.F. and Wood, J.D. (1992a). Electrical behaviour of myenteric neurones in the gastric antrum of the guinea-pig. Journal of Physiology (London), 447, 49–66. Tack, J.F. and Wood, J.D. (1992b). Actions of noradrenaline on myenteric neurons in the guinea-pig gastric antrum. Journal of the Autonomic Nervous System, 41, 67–77. Tack, J.F., Janssens, J., Vantrappen, G. and Wood, J.D. (1992). Actions of 5-hydroxytryptamine on myenteric neurones in the guinea-pig gastric antrum. American Journal of Physiology, 263, G838–G846. Takahashi, T. and Owyang, C. (1995). Vagal control of nitric oxide and vasoactive intestinal polypeptide release in the regulation of gastric relaxation in rat. Journal of Physiology (London), 484, 481–492. Takahashi, T. and Owyang, C. (1997). Characterization of vagal pathways mediating gastric accommodation reflex in rats. Journal of Physiology (London), 504, 479–488. Takeshima, T. (1971). Functional classification of the vagal afferent discharges in the dog stomach. Japanese Journal of Smooth Muscle Research, 7, 19–27. Tamura, K., Schemann, M. and Wood, J.D. (1993). Actions of nitric oxide-generating sodium nitroprusside in myenteric plexus of guinea pig small intestine. American Journal of Physiology, 265, G887–G893. Ternaux, J.P., Falempin, M., Palouzier, B., Chamoin, M.C. and Portalier, P. (1989). Presence of cholinergic neurons in the vagal afferent system: Biochemical and immunohistochemical approaches. Journal of the Autonomic Nervous System, 28, 233–242. Traub, R.J., Sengupta, J.N. and Gebhart, G.F. (1996). Differential c-Fos expression in the nucleus of the solitary tract and spinal cord following noxious gastric distention in the rat. Neuroscience, 74, 873–884. Tsai, L.H., Tsai, W. and Wu, J.-Y. (1993). Action of myenteric gabaergic neurons in the guinea pig stomach. Neurochemistry International, 23, 187–193. Tsai, L.H., Tsai, W. and Wu, J.-Y. (1994). Effect of L-glutamic acid on acid secretion and immunohistochemical localization of glutamatergic neurons in the rat stomach. Journal of Neuroscience Research, 38, 188–195. Tsubomura, T., Okamoto, T., Kurahshi, K. and Fujiwara, M. (1987). Gastric excitation by stimulation of the vagal trunk after chronic supranodose vagotomy in cats. Journal of Pharmacology Experimental Therapeutics, 241, 650–654. Tsuruo, Y., Hökfelt, T., Visser, T.J., Kimmel, J.R., Brown, J.C., Verhofstadt, A. et al. (1988). TRH-like immunoreactivity in endocrine cells and neurons in the gastrointestinal tract of the rat and guinea-pig. Cell and Tissue Research, 253, 347–356. Vaithilingam, U.D., Wong, W. and Ling, E.A. (1986). Transneuronal changes in the myenteric ganglia of the monkey following vagotomy. Neuroscience, 17, 829–836. van Aswegen, G., Schoman, J.H., De Vos, V. and van Noorden, S. (1994). The oesophagus and stomach of the african elephant: a histological, immunocytochemical and immunofluorescence study. Onderstepoort Journal of Veterinary Research, 61, 223–229.

102

INNERVATION OF THE GASTROINTESTINAL TRACT

Van Lier Ribbink, J.A., Sarr, M.G. and Tanaka, M. (1989). Neural isolation of the entire canine stomach in vivoeffects on motility. American Journal of Physiology, 257, G30–G40. Varro, A., Green, T., Holmes, S. and Dockray, G.J. (1988). Calcitonin gene-related peptide in visceral afferent nerve fibres: quantification by radioimmunoassay and determination of axonal transport rates. Neuroscience, 26, 927–932. Wang, Y.F., Mao, Y.K., MacDonald, T.J. and Daniel, E.E. (1995). Distribution of galanin-immunoreactive nerves in the canine gastrointestinal tract. Peptides, 16, 237–247. Ward, S.M., Morris, G.A., Bayguinov, Y.R. and Sanders, K.M. (1997). Interstitial cells of Cajal mediate inhibitory neurotransmission in the lower esophageal and pyloric sphincters. Gastroenterology, 112, A850. Wathuta, E.M. (1986). The distribution of vasoactive intestinal polypeptide-like, substance P-like and bombesinlike immunoreactivity in the digestive system of the sheep. Quarterly Journal of Experimental Physiology, 71, 615–631. Wiley, J.W., Lu, Y. and Owyang, C. (1991). Mechanism of action of peptide YY to inhibit gastric motility. Gastroenterology, 100, 865–872. Winkler, H. (1898). Ein Beitrag zur Physiologie der glatten Muskeln. Archives für die gesamte Physiologie, 71, 357–398. Wood, J.D. (1994a). Application of classification schemes to the enteric nervous system. Journal of the Autonomic Nervous System, 48, 17–29. Wood, J.D. (1994b). Physiology of the enteric nervous system. In Physiology of the Gastrointestinal Tract, 3rd edition, edited by L.R. Johnson, pp. 423–482. New York: Raven Press. Yamamoto, Y., Kitamura, N., Yamada, J. and Yamashita, T. (1994). Immunohistochemical study of the distributions of peptide-and catecholamine-containing nerves in the omasum of the sheep. Acta Anatomica, 149, 104–110. Yano, S., Fujiwara, A., Ozaki, Y. and Harada, M. (1983). Gastric blood flow responses to autonomic nerve stimulation and related pharmacological studies in rats. Journal of Pharmacy and Pharmacology, 35, 641–646. Yuan, S.Y., Brookes, S.J.H. and Costa, M. (1997). Distension-evoked ascending and descending reflexes in the isolated guinea-pig stomach. Journal of the Autonomic Nervous System, 62, 94–102. Zhang, X., Fogel, R. and Renehan, W.E. (1992). Physiology and morphology of neurons in the dorsal motor nucleus of the vagus and the nucleus of the solitary tract that are sensitive to distention of the small intestine. Journal of Comparative Neurology, 323, 432–448. Zhang, X., Fogel, R. and Renehan, W.E. (1995). Relationships between the morphology and function of gastric and intestine-sensitive neurons in the nucleus of the solitary tract. Journal of Comparative Neurology, 363, 37–52. Zheng, H., Lauve, A., Patterson, L.M. and Berthoud, H.R. (1997). Limited excitatory local effector function of gastric vagal afferent intraganglionic terminals in rats. American Journal of Physiology, 273, G661–G669.

3 Control of Gastric Functions by Extrinsic Sensory Neurons* Peter Holzer Department of Experimental and Clinical Pharmacology, University of Graz, Universitätsplatz 4, A-8010 Graz, Austria The maintenance of gastric mucosal integrity depends on the rapid alarm of protective mechanisms in the face of pending injury. Afferent neurons of extrinsic origin constitute an emergency system that is called into operation when the gastric mucosa is endangered by noxious chemicals. The function of these chemoceptive afferents can selectively be manipulated and explored with the excitotoxin capsaicin. Since most of the homeostatic actions of capsaicin-sensitive afferents are brought about by peptides released from their peripheral endings in the gastric wall, the roles of these neurons are described as “sensory-efferent”, “sensory-motor”, “local effector” or “noceffector” to emphasise the functional similarity with the roles of autonomic efferents. When stimulated, chemoceptive afferents enhance gastric blood flow and activate hyperaemia-dependent and hyperaemia-independent mechanisms of protection and repair. Furthermore, they can influence the gastric secretion of acid, bicarbonate and mucus and modulate gastric motility and emptying. In the rodent stomach, the noceffector roles of sensory neurons are mediated by calcitonin gene-related peptide acting via CGRP1 receptors and neurokinin A acting via NK2 receptors, both peptides using nitric oxide as their second messenger. In addition, capsaicin-sensitive neurons form the afferent arc of autonomic reflexes that control secretory and motor functions of the stomach. The pathophysiological potential of the neural emergency system is best portrayed by the gastric hyperaemic response to acid backdiffusion, which is signalled by afferent nerve fibres. This mechanism limits damage to the surface of the mucosa and creates favourable conditions for rapid restitution and healing of the wounded mucosa. KEY WORDS: stomach; afferent neurons; capsaicin; calcitonin gene-related peptide; tachykinins; substance P; neurokinin A; nitric oxide; gastric blood flow; gastric secretion; gastric motor activity; peptic ulcer disease; ileus.

INTRODUCTION Nowhere else in the digestive system is the mucosa more endangered than in the gastroduodenal region, the threats being primarily of a chemical nature and of both endogenous and exogenous origin. Most tissues would rapidly disintegrate if exposed to the concentrations of hydrochloric acid that bathe but do not harm the gastric surface epithelium.

*Article completed on 3 February 1997. 103

104

INNERVATION OF THE GASTROINTESTINAL TRACT

This is because a multitude of physical, chemical and physiological factors, collectively forming the gastric mucosal barrier, prevent hydrogen ions from entering the tissue at quantities that would produce cell injury (Allen et al., 1993; Pabst, Wachter and Holzer, 1996). Although autonomic neurons have long been recognised to influence the capacity of the gastric mucosa to defend itself against damage, the possibility that neurons constitute an important element in gastric mucosal protection from injury was hardly thought of. This neglect of neurons may have been related to the understanding that the vagus nerve is a permissive factor in gastric lesion formation, given that subdiaphragmatic vagotomy can be beneficial in peptic ulcer disease. However, central stimulation of the vagus nerve can also protect the gastric mucosa from damage (Kato, Yang and Taché, 1995), which attests to the complexities in the autonomic regulation of gastric mucosal homeostasis. In his initial conception, Langley (1903) considered afferent neurons (“autonomic afferents”) to be a third subdivision of the autonomic nervous system. Although this concept was later abandoned, it remains convenient to label afferent neurons running in the vagus and pelvic nerves as “parasympathetic afferents” and those running in the splanchnic nerves as “sympathetic afferents” to denote their projections in the respective autonomic efferent nerves (Sengupta and Gebhart, 1994). It was only recently that “autonomic afferents” were discovered to constitute an important neural emergency system in the digestive tract. These neurons have been found to participate in the neural control of blood flow, secretory processes and motor activity and to regulate a variety of functions that can be viewed as increasing the resistance of the tissue to injury and facilitating the repair of damaged tissue. Since most of these actions are brought about by release of transmitter substances from the peripheral endings of afferent nerve fibres, the function of these neurons has been portrayed as “sensory-efferent” (Szolcsányi, 1984), “sensory-motor” (Burnstock, 1990), “local effector” (Holzer, 1988) or “noceffector” (Kruger, 1988) to emphasise the operational similarity with the physiological roles of autonomic efferent neurons. Many actions exerted by afferent neurons in the stomach, notably those on blood flow, resemble sensory nerve-mediated “neurogenic inflammatory reactions” seen in other tissues (Geppetti and Holzer, 1996). The discovery of these sensory neuron functions was

Figure 3.1 Use of capsaicin as a selective pharmacological tool to examine the functional implications of primary afferent neurons with fine (C- and Aδ-) nerve fibres.

CONTROL OF GASTRIC FUNCTIONS BY EXTRINSIC SENSORY NEURONS

105

made possible by the availability of capsaicin, a pharmacological tool with which the activity of primary afferent neurons with unmyelinated or thinly myelinated nerve fibres can selectively be manipulated (Holzer, 1991; Wood, 1993). Capsaicin is an excitotoxin which acutely stimulates a group of fine afferent nerve fibres and, depending on the dose and route of administration, may cause a long-lasting defunctionalisation of, and depletion of transmitter substances from, these neurons (Figure 3.1). In contrast, afferent neurons with thick myelin sheaths are not sensitive to capsaicin (Holzer, 1991). The selectivity of capsaicin’s excitotoxic action on afferent neurons derives from the fact that only these neurons express receptor binding sites (“vanilloid receptors”) for capsaicin and structurally related ligands (Szallasi, 1994). By taking use of the neurotoxic action of capsaicin it is possible to ablate afferent neurons without severing efferent autonomic neurons which unavoidably are cut by nerve transection. The exploitation of capsaicin’s neuropharmacological potential has brought unprecedented insights into the pathophysiological roles of primary afferent nerve fibres in the gastrointestinal tract, and it is the aim of this review to give an account of the experimental findings that have led to the elucidation of this neural emergency system in the stomach and to consider its physiological and pathophysiological implications.

EXTRINSIC AFFERENT NERVE FIBRES IN THE STOMACH The extrinsic primary afferent neurons supplying the stomach arise from two different sources (Figure 3.2). The spinal sensory neurons originate from cell bodies in the dorsal root ganglia and reach the stomach via the splanchnic and mesenteric nerves, while the sensory nerve fibres running in the vagus nerves have their cell bodies in the nodose and jugular ganglia. In small rodents (rat, mouse, guinea-pig) most of the spinal afferent neurons contain a variety of bioactive peptides including calcitonin gene-related peptide (CGRP) and the tachykinins substance P (SP) and neurokinin (NKA). CGRP and SP show a considerable degree of coexistence in the same afferent neurons but, while CGRP is

Figure 3.2 Schematic diagram showing the innervation of the mammalian stomach by primary afferent neurons.

106

INNERVATION OF THE GASTROINTESTINAL TRACT

contained in both small and large somata, SP occurs in small and medium-sized cell bodies only (Molander, Ygge and Dalsgaard, 1987; Su et al., 1987; Green and Dockray, 1988; McCarthy and Lawson, 1989, 1990; Kashiba et al., 1991). It is important to realise that the chemical coding of extrinsic afferents, i.e., the combination of peptides coexpressed in individual neurons, varies considerably across species. Thus, very little CGRP seems to be expressed in neurons of the human stomach (Tsutsumi and Hara, 1989; Sundler, Ekblad and Håkanson, 1991). The peptidergic innervation of the gastrointestinal tract is complicated by the fact that there are three separate populations of peptide-containing neurons, extrinsic primary afferent, extrinsic autonomic, and intrinsic enteric neurons, which have many peptide transmitters in common (Ekblad et al., 1985; Costa, Furness and Gibbins, 1986; Kirchgessner, Dodd and Gershon, 1988). However, extrinsic afferents differ from the other neuron populations in their chemical coding and sensitivity to capsaicin, let alone their physiological implications. CGRP and SP are coexpressed only in extrinsic afferents (Costa, Furness and Gibbins, 1986) and most of the CGRP present in afferent neurons appears to be CGRP-α, whereas the only form of CGRP in enteric neurons is CGRP-β (Mulderry et al., 1988; Sternini, 1992). Furthermore, only primary afferent neurons are sensitive to the excitotoxic action of capsaicin while autonomic (Holzer, 1991) and enteric (Barthó and Holzer, 1985; Green and Dockray, 1988) neurons are not. Table 3.1 summarises some of the neurochemical and biochemical changes that result from capsaicin-induced defunctionalisation of afferent neurons in the rat stomach and duodenum. These studies have revealed that the CGRP-containing nerve fibres in the rat stomach arise from extrinsic afferent neurons whereas the SP-immunoreactive fibres are of both extrinsic and intrinsic origin (Green and Dockray, 1988; Varro et al., 1988; Sternini, 1992). The extrinsic axons disappear from the wall of the gastrointestinal tract when the gut is surgically severed from its extrinsic nerve supply or the experimental animals are treated with a neurotoxic dose of capsaicin (Furness et al., 1982; Hayashi et al., 1982; Barja, Mathison and Huggel, 1983; Minagawa et al., 1984; Gibbins et al., 1985; Lee et al., 1987b; Sternini, Reeve and Brecha, 1987; Su et al., 1987; Green and Dockray, 1988; Kashiba et al., 1990; McGregor and Conlon, 1991). TABLE 3.1 Effect of capsaicin-induced defunctionalisation of afferent neurons on neurochemical and biochemical parameters in the rat stomach and duodenum. Capsaicin Parameter administration Systemic

Effect of nerve defunctionalisation

References

Alanine aminopeptidase activity No change (duodenum) McGregor and Conlon, 1991 Alkaline phosphatase activity No change (duodenum) McGregor and Conlon, 1991 Calcitonin gene-related peptide Loss Sternini, Reeve and Brecha, 1987; Su et al., 1987; Geppetti et al., 1988; Green and Dockray, 1988; Varro et al., 1988; Evangelista et al., 1991; Inui et al., 1991;

CONTROL OF GASTRIC FUNCTIONS BY EXTRINSIC SENSORY NEURONS

DNA synthesis Galanin Gastrin messenger RNA Glucagon HDCa mRNA Histamine content 5-Hydroxytryptamine content Leucine enkephalin content Methionine enkephalin content Neurokinin A Neuropeptide Y Nitric oxide synthase Prostaglandin formation Somatostatin Somatostatin messenger RNA

Substance P content

Substance P in muscle/ myenteric plexus Substance P in perivascular nerves Substance P in mucosa/ submucosa Vasoactive intestinal polypeptide

McGregor and Conlon, 1991; Renzi et al., 1991; Holzer, Lippe and Amann, 1992; Gray et al., 1994 No change Takeuchi et al., 1994 No change Su et al., 1987; Evangelista et al., 1992 No change (antrum) Sandvik et al., 1993 No change (duodenum) McGregor and Conlon, 1991 No change (corpus) Sandvik et al., 1993 No change Holzer et al., 1981 No change Holzer et al., 1981; McGregor and Conlon, 1991 No change Varro et al., 1988 No change Varro et al., 1988 No change

Renzi et al., 1991; Evangelista et al., 1992 No change Su et al., 1987; Evangelista et al., 1992 No change Forster and Southam, 1993 No change Holzer and Sametz, 1986 No change Inui et al., 1991 Reduction (corpus) Sandvik et al., 1993; Dimaline et al., 1994 No change (antrum) Sandvik et al., 1993; Dimaline et al., 1994 No change Holzer, Gamse and Lembeck, 1980; Geppetti et al., 1991; McGregor and Conlon, 1991; Renzi et al., 1991 Reduction (duodenum) Geppetti et al., 1988 No change Su et al., 1987; Green and Dockray, 1988 Loss Su et al., 1987; Green and Dockray, 1988 Reduction Su et al., 1987; Green and Dockray, 1988 No change Su et al., 1987; Evangelista et al., 1991, 1992; McGregor and Conlon, 1991; Tramontana et al., 1994a,b

Perivagal

Calcitonin gene-related peptide

No change

Raybould et al., 1992

Pericoeliacb

Calcitonin gene-related peptide

Depletion

Raybould et al., 1992

a

107

HDC, histidine decarboxylase; b administration of capsaicin locally to the coeliac ganglion.

108

INNERVATION OF THE GASTROINTESTINAL TRACT

Most of the extrinsic nerve fibres expressing CGRP and/or SP in the rodent stomach are spinal, rather than vagal, afferents which on their way from the dorsal root ganglia to the gut pass through the prevertebral ganglia (Figure 3.2) where they give off collaterals to form axodendritic and axosomatic synapses with the sympathetic ganglion cells (Hayashi et al., 1982; Matthews and Cuello, 1984; Lee et al., 1987a,b; Su et al., 1987; Green and Dockray, 1988; Kondo and Yamamoto, 1988; Lindh, Hökfelt and Elfvin, 1988). However, vagal afferent nerve fibres immunoreactive for SP and CGRP have been found to supply the oxyntic part of the feline and rat stomach (Hayashi et al., 1982; Bäck et al., 1994) while other studies hold that vagal afferents contribute little to the SP and CGRP content of the rat gastric corpus (Lee et al., 1987b; Green and Dockray, 1988). This controversy may be related to regional and species differences in the gastric projections of peptidergic vagal afferents. It seems as if the density of these vagal nerve fibres decreases from the proximal to the distal stomach as the gastric corpus of the cat is innervated by vagal SP-containing fibres (Hayashi et al., 1982) whereas the extrinsic supply of SP fibres in the antrum, pylorus and duodenum of the cat and guinea-pig is derived from spinal afferents (Hayashi et al., 1982; Lindh, Hökfelt and Elfvin, 1988). Within the rodent gastric wall it is particularly the arterial and arteriolar system that receives a dense supply by spinal afferent nerve fibres expressing CGRP, SP and NKA (Figure 3.2), whereas the venous system is rather sparsely innervated (Furness et al., 1982; Barja, Mathison and Huggel, 1983; Sharkey, Williams and Dockray, 1984; Uddman et al., 1986; Sternini, Reeve and Brecha, 1987; Su et al., 1987; Green and Dockray, 1988; Sternini, 1992). The peptide-containing axons run primarily in the connective tissue (tunica adventitia) surrounding the vessels and at the border between adventitia and muscle (media). In addition, some peptide-containing afferent nerve fibres also supply the myenteric plexus, the circular muscle layer and the mucosa of the gut (Furness et al., 1982; Minagawa et al., 1984; Sharkey, Williams and Dockray, 1984; Gibbins et al., 1985; Rodrigo et al., 1985; Uddman et al., 1986; Lee et al., 1987b; Sternini, Reeve and Brecha, 1987; Su et al., 1987; Green and Dockray, 1988; Kashiba et al., 1990; Sundler, Ekblad and Håkanson, 1991; Sternini, 1992; Jakab et al., 1993; Bäck et al., 1994; Schmidt et al., 1996).

PEPTIDERGIC AFFERENT NERVE FIBRES AND GASTRIC BLOOD FLOW VASODILATATION CAUSED BY AFFERENT NERVE STIMULATION The possibility that perivascular afferent nerve fibres control vascular functions in the stomach went unrecognised for a long time, as these axons were considered to subserve only sensory but not effector functions. However, with the advent of capsaicin it became feasible to prove that sensory neurons containing CGRP participate in the regulation of the rat gastric circulation. Stimulation of afferent nerve fibres by intragastric administration of capsaicin or its ultrapotent analogue resiniferatoxin leads to a marked increase in gastric mucosal blood flow (GMBF; Table 3.2) and blood flow through the left gastric artery (Figure 3.3). This effect is brought about by dilatation of submucosal arterioles but not venules (Chen et al., 1992; Chen and Guth, 1995). Other regions of the digestive tract,

CONTROL OF GASTRIC FUNCTIONS BY EXTRINSIC SENSORY NEURONS

109

Figure 3.3 Recording of the hyperaemic and dilator effect of intragastric (i.g.) capsaicin (160 µM) on the left gastric artery of an urethane-anaesthetised rat. The traces from the top to the bottom show heart rate (HR), mean arterial blood pressure (MAP), blood flow in the left gastric artery (BF/LGA) and vascular conductance in the left gastric artery (VC/LGA). Data from Wachter et al. (1995).

TABLE 3.2 Effect of capsaicin/resiniferatoxin-evoked stimulation of afferent neurons on motor, secretory and vascular functions of the oesophagus, stomach and duodenum. Species/region

Function

Rabbit oesophagus Blood flow Sphincter pressure Canine LOSa Canine stomach Acid backdiffusion Acid output 2-DODGb-evoked acid output Histamine-evoked acid output Mucosal blood flow TMPDc Plasma gastrin level Rat stomach

Acid backdiffusion Acid output

Effect of capsaicin

References

Increase Increase No change No change Reduction

Bass et al., 1991 Sandler et al., 1993 Sullivan et al., 1992 Soldani et al., 1992; Sullivan et al., 1992 Soldani et al., 1992

No change

Soldani et al., 1992

Increase No change No change

Sullivan et al., 1992 Sullivan et al., 1992 Soldani et al., 1992

No change No change

Lippe, Pabst and Holzer, 1989 Lippe, Pabst and Holzer, 1989; Matsumoto, Takeuchi and Okabe, 1991; Narita et al., 1995; Kang, Teng and Chen, 1996 Abdel Salam, Szolcsányi and Mózsik, 1994, 1995; Brzozowski et al., 1996b Brzozowski et al., 1996b

Decrease Pepsin output

Decrease

110

INNERVATION OF THE GASTROINTESTINAL TRACT

TABLE 3.2 Continued Species/region

Function

Effect of capsaicin

References

Rat stomach (Continued)

Bicarbonate output Emptying Motor activity in vivo

Increase Inhibition Increase (low dose) Reduction (high dose) Contraction/ relaxation Relaxation

Takeuchi et al., 1992b,c Kang et al., 1993; Hatakeyama et al., 1995 Holzer et al., 1991

Increase

Lippe, Pabst and Holzer, 1989; Holzer et al., 1990a, 1991; Li et al., 1991; Matsumoto, Takeuchi and Okabe, 1991; Takeuchi et al., 1991b; Chen et al., 1992; Leung, 1992a; Wallace, McKnight and Befus, 1992; Whittle, Lopez-Belmonte and Moncada, 1992; Brzozowski et al., 1993; Grønbech and Lacy, 1995; Merchant et al., 1994; Podolsky et al., 1994; Abdel Salam et al., 1996; Kang, Teng and Chen, 1996; Miyake et al., 1996 Holzer and Lippe, 1988; Lippe, Pabst and Holzer, 1989; Matsumoto, Takeuchi and Okabe, 1991; Takeuchi et al., 1991b Holzer, Pabst and Lippe, 1989 Holzer and Lippe, 1988

Motor activity in vitro (corpus) Motor activity in vitro (fundus) Mucosal blood flow

Takeuchi et al., 1991b Holzer-Petsche, Seitz and Lembeck, 1989 Lefebvre, De Beurme and Sas, 1991

TMPD

No change

TMPD fall due to ASAd TMPD fall due to ethanol Mucus output Vascular permeability

Attenuation No change Increase No change Increase

Kang et al., 1995a Holzer and Lippe, 1988; Hatakeyama et al., 1995; Sann et al., 1996 Lördal et al., 1996

Human stomach

Emptying Mucosal blood flow

Inhibitione Increasee

Horowitz et al., 1992 Graham et al., 1988

Rat duodenum

Bicarbonate output Mucosal blood flow Mucosal tissue integrity Vascular permeability

Increase Increase No change Slight increase

Takeuchi et al., 1991a, 1992b Leung, 1993b; Seno et al., 1996 Takeuchi et al., 1992a; Leung, 1993b Maggi et al., 1987a; Sann et al., 1996

a d

LOS, lower oesophageal sphincter; b 2-DODG, 2-deoxy-D-glucose; c TMPD, transmucosal potential difference; ASA, acetylsalicylic acid; e Effect of capsaicin contained in Jalapeno peppers or chilli powder.

in which capsaicin-evoked stimulation of sensory neurons augments blood flow, include the rabbit oesophagus (Bass et al., 1991), rat duodenum (Leung, 1993b; Seno et al., 1996), rat jejunum (Abdel Salam et al., 1996), rat colon (Leung, 1992b) and the mesenteric arteries of the rat and dog (Holzer, 1992; Jacobson et al., 1994).

CONTROL OF GASTRIC FUNCTIONS BY EXTRINSIC SENSORY NEURONS

111

The vasodilator effect of capsaicin and resiniferatoxin in the rat stomach, duodenum (Table 3.3), jejunum (Mathison and Davison, 1995) and colon (Leung, 1992b) depends on the integrity of the extrinsic afferent innervation of these organs, because defunctionalisation of afferent neurons by a neurotoxic dose of capsaicin abolishes the hyperaemic response to sensory nerve stimulation. Nerve-selective ablation of afferent nerve fibres has shown that only spinal afferent neurons passing through the coeliac ganglion participate in the capsaicin-evoked increment of GMBF (Li et al., 1991). However, the vagus nerves of the rat also contain afferent nerve fibres that give rise to gastric mucosal vasodilatation following electrical stimulation (Thiefin et al., 1990) but these vagal nerve fibres do not seem to contribute to the hyperaemia caused by intragastric capsaicin (Li et al., 1991) although they are ablated by perineural treatment with a neurotoxic dose of capsaicin (Thiefin et al., 1990).

TABLE 3.3 Neurons and mediators involved in the functional responses to capsaicin/resiniferatoxin-evoked stimulation of afferent neurons in the oesophagus, stomach and duodenum. Effect of capsaicin/ resiniferatoxin

Neurons/mediators

References

Increased acid elimination Increased bicarbonate output

Afferent neurons involved

Lippe, Pabst and Holzer, 1989

Afferent neurons involved

Takeuchi et al., 1991a, 1992b

Indomethacin-sensitive mechanisms involved Muscarinic and nicotinic ACha receptors involved

Takeuchi et al., 1991a, 1992b

Increased motor activity

Tachykinins involved

Holzer-Petsche, Seitz and Lembeck, 1989; Sandler et al., 1993

Decreased motor activity

Extrinsic afferent nerves involved Indomethacin-sensitive mechanisms involved CGRPb, SPc and VIPd ruled out

Lefebvre, De Beurme and Sas, 1991; Takeuchi et al., 1991b Takeuchi et al., 1991b

Afferent neurons involved

Holzer and Lippe, 1988; Holzer et al., 1991; Takeuchi et al., 1991b, 1994; Uchida, Yano and Watanabe, 1991b; Gray et al., 1994 Holzer and Lippe, 1988 Holzer and Lippe, 1988

Protection from injury

Sympathetic neurons ruled out Parasympathetic neurons ruled out Muscarinic ACh receptors involved Muscarinic ACh receptors ruled out

Takeuchi et al., 1991a

Holzer-Petsche, Seitz, and Lembeck, 1989; Lefebvre, De Beurme and Sas, 1991; Takeuchi et al., 1991b

Abdel Salam et al., 1995a Holzer and Lippe, 1988

112

INNERVATION OF THE GASTROINTESTINAL TRACT

TABLE 3.3 Continued Effect of capsaicin/ resiniferatoxin

Neurons/mediators

References

Protection from injury (Continued)

CGRP involved

Lambrecht et al., 1993a; Peskar et al., 1993; Kato, Yang and Taché, 1994; Merchant et al., 1994 Peskar et al., 1991; Brzozowski et al., 1993; Lambrecht et al., 1993a; Podolsky et al., 1994 Takeuchi et al., 1991b

Nitric oxide involved SP ruled out Neurokinin A involved Prostaglandins and leukotriene C4 ruled out Indomethacin-sensitive mechanisms ruled out Indomethacin-sensitive mechanisms involved

Vasodilatation

Afferent neurons involved

CGRP involved

Nitric oxide involved

Tachykinins ruled out

Indomethacin-sensitive mechanisms involved Muscarinic ACh receptors ruled out Nicotinic ACh receptors involved Noradrenaline α- and β-adrenoceptors ruled out Mast cell mediators ruled out

Stroff et al., 1996 Holzer et al., 1990a Holzer et al., 1990a Takeuchi et al., 1991b; Uchida, Yano and Watanabe, 1991b; Brzozowski et al., 1993; Mercer, Ritchie and Dempsey, 1994; Hatakeyama et al., 1995 Lippe, Pabst and Holzer, 1989; Holzer et al., 1991; Li et al., 1991; Matsumoto, Takeuchi and Okabe, 1991; Takeuchi et al., 1991b, 1994; Chen et al., 1992; Leung, 1992a; Wallace, McKnight and Befus, 1992; Abdel Salam et al., 1996 Li et al., 1991; Chen et al., 1992; McKie et al., 1994; Merchant et al., 1994; Chen and Guth, 1995 Whittle, Lopez-Belmonte and Moncada, 1992; Brzozowski et al., 1993; Podolsky et al., 1994; Chen and Guth, 1995; Merchant et al., 1995 Matsumoto, Takeuchi and Okabe, 1991; Grønbech and Lacy, 1994; McKie et al., 1994; Stroff et al., 1996 Matsumoto, Takeuchi and Okabe, 1991; Takeuchi et al., 1991b; Brzozowski et al., 1993 Lippe, Pabst and Holzer, 1989; Matsumoto, Takeuchi and Okabe, 1991; Abdel Salam et al., 1996 Leung, 1993a Lippe, Pabst and Holzer, 1989 Matsumoto, Takeuchi and Okabe, 1991; Wallace, McKnight and Befus, 1992; Grønbech and Lacy, 1994

a ACh, acetylcholine; b CGRP, calcitonin gene-related peptide; c SP, substance P; d VIP, vasoactive intestinal polypeptide.

CONTROL OF GASTRIC FUNCTIONS BY EXTRINSIC SENSORY NEURONS

113

PATHWAYS AND MEDIATORS OF NEUROGENIC VASODILATATION Calcitonin gene-related peptide (CGRP) and related peptides The neural pathways responsible for the capsaicin-evoked rise of GMBF in the rat stomach, which involve ganglionic transmission via nicotinic acetylcholine receptors (Leung, 1993a), have not conclusively been defined. Since noradrenergic neurons and cholinergic vasodilator neurons have been ruled out (Table 3.3) it would appear that the vasodilator response to capsaicin is due to a local mechanism within the gastric wall and brought about by release of vasodilator peptides from perivascular afferent nerve fibres (Holzer et al., 1990a, 1991). While the vasodilator mediators of vagal afferent nerve fibres have not yet been elucidated, there is conclusive evidence that CGRP plays an important mediator role in the rise of rat GMBF evoked by stimulation of spinal afferent nerve fibres, since the CGRP1 receptor antagonist CGRP8–37 prevents the gastric vasodilatation caused by intragastric and intravascular capsaicin (Table 3.3). It need be added, though, that the data obtained with CGRP8–37 prove the implication of CGRP1 receptors but not necessarily of CGRP itself. This is because peptides with some structural homology to CGRP, such as adrenomedullin formed in endothelial cells (Sugo et al., 1994) and amylin expressed in afferent neurons (Mulder et al., 1995) and endocrine cells of the gastric mucosa (Mulder et al., 1994a), are also capable of activating CGRP1 receptors, and vasodilator responses to both amylin (Gardiner et al., 1991) and adrenomedullin (Nuki

TABLE 3.4 Release of calcitonin gene-related peptide (CGRP), substance P (SP) and neurokinin A (NKA) from presumably afferent nerve fibres in the stomach and duodenum. Species/region

Peptide

Release stimulus

References

Rat stomach

CGRP

Acid (lowering of pH)

Ethanol Acid Capsaicin Capsaicin

Geppetti et al., 1991; Manela et al., 1995; Ren et al., 1995 Holzer et al., 1990b; Chiba et al., 1991; Geppetti et al., 1991; Inui et al., 1991; Renzi et al., 1991; Ren et al., 1993; Gray et al., 1994; Narita et al., 1995; Inaba et al., 1996 Inui et al., 1989 Manela et al., 1992 Inui et al., 1989 Kwok and McIntosh, 1990; Renzi et al., 1991; Katori, Ohno and Nishiyama, 1993; Hayashi et al., 1996 Katori, Ohno and Nishiyama, 1993; Hayashi et al., 1996 Smedfors, Theodorsson and Johansson, 1994 Fujimiya and Kwok, 1995 Renzi et al., 1988

Acidified 2 M NaCl Capsaicin Acidification

Gislason et al., 1995a Schmidt et al., 1996 Mueller et al., 1991

Capsaicin

SP/NKA

Rat duodenum

SP/NKA

Guinea-pig SP/NKA stomach Feline stomach CGRP Porcine stomach SP/NKA Human stomach SP a

Dibutyryl cAMPa Peptone Theophylline Capsaicin

cAMP, adenosine 3′,5′-cyclic monophosphate.

114

INNERVATION OF THE GASTROINTESTINAL TRACT

et al., 1993; Entzeroth et al., 1994) are antagonised by CGRP8–37. Whether amylin and adrenomedullin, which are less active than CGRP in causing vasodilatation (Gardiner et al., 1991; Nuki et al., 1993; Entzeroth et al., 1994), play a role in afferent nervemediated hyperaemia has not yet been tested. An implication that CGRP is involved is consistent with the ability of capsaicin and other stimuli to release this peptide from extrinsic afferent nerve fibres in the rat, guineapig and feline stomach (Table 3.4) and with the vascular pharmacology of the peptide. CGRP fails to affect gastric venules but potently dilates submucosal arterioles in the rat stomach (Chen et al., 1992; Chen and Guth, 1995) and enhances blood flow through the oesophagus and stomach of rat and rabbit (Bauerfeind et al., 1989; Lippe, Lorbach and Holzer, 1989; Holzer and Guth, 1991; Li et al., 1991; Holzer et al., 1993; Lopez-Belmonte and Whittle, 1993; Király et al., 1994a; McKie et al., 1994). The vasodilator action of the peptide is in keeping with the abundant presence of CGRP receptors on the muscle and endothelium of arteries and arterioles in the rat and canine stomach (Gates et al., 1989; Sternini, 1992), although one laboratory has failed to observe a hyperaemic response to CGRP in the canine stomach (Nakamura et al., 1996). The ability of CGRP8–37 to block the hyperaemia that in the rat stomach is elicited by both capsaicin and CGRP (Li et al., 1991; Chen et al., 1992; Holzer et al., 1994; Király et al., 1994a; Chen and Guth, 1995) indicates that the peptide dilates gastric blood vessels via an action on CGRP1 receptors. This is also true for the CGRP-evoked rise of mucosal blood flow in the rabbit oesophagus (McKie et al., 1994). Nitric oxide (NO) The vasodilator action of CGRP released by capsaicin in the rat stomach involves the formation of NO, since NO synthase inhibitors suppress the gastric hyperaemic reaction to both capsaicin (Table 3.3) and CGRP (Whittle, Lopez-Belmonte and Moncada, 1992; Holzer et al., 1993; Chen and Guth, 1995). NO is unlikely to be released from the CGRPcontaining axons, because ablation of capsaicin-sensitive afferent nerve fibres does not alter the activity of NO synthase in the rat stomach (Forster and Southam, 1993). The localisation of CGRP receptors on endothelial cells and the observation that the relaxant action of CGRP on precontracted arteries from the human stomach depends on the presence of the endothelium (Thom et al., 1987) suggest that CGRP increases the formation of NO in endothelial cells (Figure 3.4). However, the gastric vasodilator response to CGRP is only partly NO-dependent (Holzer et al., 1993) and less susceptible to inhibition by a NO synthase blocker than the vasodilator response to capsaicin (Chen and Guth, 1995). These observations indicate that part of the vasodilator action of CGRP in the stomach may be due to a direct action of the peptide on the vascular smooth muscle (Figure 3.4) and that the cellular sources of NO formed in response to capsaicin and CGRP are not fully identical. It need be considered in this context that, in addition to being the final vasodilator mediator of CGRP, NO can also facilitate the release of CGRP from afferent vasodilator axons (Wei et al., 1992; Holzer and Jocic, 1994; Hughes and Brain, 1994). While such a role of NO is conceivable in the in vivo stomach (Chen and Guth, 1995), there is little direct evidence that NO activates CGRP-releasing afferent nerve fibres in the vasculature of the rat stomach in vitro (Seyed Ebrahim, Stroff and Peskar, 1996).

CONTROL OF GASTRIC FUNCTIONS BY EXTRINSIC SENSORY NEURONS

115

Figure 3.4 Flow chart of the vasodilator and protective mechanisms initiated by stimulation of afferent neurons and subsequent release of peptide transmitters in the stomach. CGRP1-R, CGRP1 receptor; NK1-R, tachykinin NK1 receptor; NK2-R, tachykinin NK2 receptor.

Tachykinins SP and NKA are two other peptides that are released from afferent nerve fibres in the stomach and duodenum in response to capsaicin and other stimuli (Table 3.4). While SP/ NKA and CGRP comediate the capsaicin-evoked dilatation of submucosal arterioles in the guinea-pig ileum (Vanner, 1994), there is no evidence that SP and NKA acting via tachykinin NK1 or NK2 receptors play a role in the afferent nerve-mediated rise of GMBF in the oesophagus and stomach (Table 3.3). Thus, SP, NKA and NK2 receptor-selective agonists fail to dilate submucosal arterioles (Katori, Ohno and Nishiyama, 1993) and to augment blood flow in the rat stomach (Yokotani and Fujiwara, 1985; Holzer and Guth, 1991; Grønbech and Lacy, 1994; Holzer et al., 1994; Heinemann et al., 1996a; Stroff et al., 1996), and tachykinin NK1 or NK2 receptor antagonists do not alter the hyperaemic reaction to afferent nerve stimulation by intragastric capsaicin (Matsumoto, Takeuchi and Okabe, 1991; Stroff et al., 1996). To the contrary, SP has in fact been found to constrict rat gastric submucosal venules in a leukotriene C4-independent manner (Katori, Ohno and Nishiyama, 1993), which is consistent with the ability of NKA and NK2 receptor-selective analogues to decrease blood flow through the rat stomach (Heinemann et al., 1996a; Stroff et al., 1996; Lippe, Wachter and Holzer, 1997). In addition, SP reduces the mucosal hyperaemia which in the rat stomach is elicited by electrical stimulation of the vagus nerve (Yokotani and Fujiwara, 1985) or by capsaicin-evoked stimulation of afferent nerve fibres (Grønbech and Lacy, 1994). The inhibitory influence of SP on the capsaicin-induced hyperaemia involves liberation of mast cell proteases (Grønbech and Lacy, 1994) which may degrade CGRP, the vasodilator peptide released by capsaicin (Li et al., 1991). It must not go unnoticed here that the vascular effects of tachykinins in the stomach are subject to species differences. Unlike in the rat and rabbit, intravascular administration of SP in the dog increases blood flow through the coeliac artery (Prokopiw and McDonald,

116

INNERVATION OF THE GASTROINTESTINAL TRACT

1994) and in the wall of the stomach (Yeo, Jaffe and Zinner, 1984; Ito et al., 1993). The gastric hyperaemia in response to tachykinins is attenuated by atropine but remains unaltered by vagotomy or splanchnicotomy (Ito et al., 1993). Pharmacological analysis indicates that the vasodilator response to tachykinins in the canine stomach is brought about by activation of NK1 receptors (Ito et al., 1993), which is in keeping with the presence of some NK1 and NK2 binding sites on the endothelium and smooth muscle of submucosal arterioles and venules of the canine and human gut (Gates et al., 1988; Mantyh et al., 1988). A mediator role of tachykinins in neurogenic vasodilatation of the canine and human stomach awaits to be proved. Vasoactive intestinal polypeptide (VIP) and pituitary adenylate cyclase-activating peptide (PACAP) PACAP has been localised to primary afferent neurons (Mulder et al., 1994b) while VIP occurs primarily in intrinsic enteric neurons of the rat stomach (Ekblad et al., 1985). Whether PACAP or VIP, which potently increases blood flow through the rat (Holzer and Guth, 1991; Holzer et al., 1993) and canine (Nakamura et al., 1996) stomach, contributes to afferent nerve-mediated hyperaemia in the stomach has not yet been examined but is worth testing particularly in the human stomach. Prostaglandins and mast cell mediators The role of prostaglandins in the capsaicin-evoked rise of GMBF is not totally clear. Although capsaicin fails to alter the ex vivo formation of prostaglandin E2, 6-oxo-prostaglandin F1 and leukotriene C4 in the rat gastric mucosa (Holzer and Sametz, 1986; Holzer et al., 1990a), indomethacin has been reported to diminish the capsaicin-induced increase in GMBF in the rat (Table 3.3). The validity of the findings with indomethacin in terms of an implication of vasodilator prostaglandins remains ambiguous, therefore, and it is worth mentioning in this respect that the capsaicin-induced rise of blood flow in the rat jejunum (Mathison and Davison, 1995) is left unaltered by indomethacin. Vasoactive mediators derived from mast cells such as histamine do not seem to contribute to the capsaicinevoked rise of GMBF under normal circumstances (Table 3.3). How the inhibitory influence of cimetidine on the capsaicin-induced hyperaemia in the rat gastric mucosa is brought about (Kang, Teng and Chen, 1996) awaits clarification.

PHYSIOLOGICAL IMPLICATIONS OF AFFERENT VASODILATOR NERVE FIBRES IN THE STOMACH Local roles in the gastric circulation The ability of afferent nerve fibres to increase GMBF raises the question as to their physiological role in gastric circulatory control. There are two reports that basal blood flow through the rat gastric mucosa, particularly in the antrum, is lowered in rats pretreated with a neurotoxic dose of capsaicin (Uchida, Yano and Watanabe, 1993; Dembinski et al., 1995) but the majority of studies holds that defunctionalisation of

CONTROL OF GASTRIC FUNCTIONS BY EXTRINSIC SENSORY NEURONS

117

capsaicin-sensitive afferent neurons does not change basal blood flow in the rat gastric corpus (Table 3.5). It would seem, therefore, that peptidergic afferent nerve fibres do not play a role in regulating GMBF under basal conditions. The gastric vasodilator responses to CGRP, acetylcholine (ACh), adenosine, histamine and gastrin-17 do likewise not involve capsaicin-sensitive neurons (Table 3.5). While the gastric mucosal hyperaemia caused by pentagastrin is either reduced or unchanged in capsaicin-pretreated rats, there is consistent evidence that the rise of GMBF brought about by low doses of cholecystokinin (CCK) or CCK octapeptide depends on capsaicin-sensitive afferent neurons (Table 3.5). Further analysis indicates that the gastric vasodilator effect of CCK octapeptide is brought about by CCKB receptors (Heinemann et al., 1995) and involves CGRP and NO as vasodilator messengers (Table 3.6). The hyperaemia caused by epidermal growth factor is similarly dependent on capsaicin-sensitive afferent neurons, CGRP and NO (Konturek, Brzozowski and Sliwowski, 1996; Teng et al., 1996a). Although ablation of fine afferent neurons does not alter the gastric vasoconstrictor response to adrenaline, it does attenuate the autoregulatory escape from adrenalineinduced vasoconstriction in the rat gastric mucosa (Table 3.5), which indicates that vascular constriction is counteracted by afferent nerve-mediated vasodilatation. An analogous explanation may apply to the finding that gastric vasoconstriction caused by platelet-activating factor or blockade of NO synthesis is amplified after capsaicin-induced defunctionalisation of afferent neurons (Table 3.5). The interaction between the tonically active NO system (Piqué, Esplugues and Whittle, 1989; Walder, Thiemermann and Vane, 1990; Lippe and Holzer, 1992; Tepperman and Whittle, 1992; Holzer et al., 1993) and stimulusdriven afferent nerve fibres suggests that physiologically these two systems act in concert to provide an active dilator drive on the gastric microcirculation (Whittle, 1993). A similar relationship holds true for haemostasis which is attenuated by capsaicin-induced ablation of afferent neurons, an effect that is intensified by blockade of NO synthesis (Lippe et al., 1993). This interaction, which is due to enhancement of platelet aggregation, suggests that sensory nerve-induced facilitation of local blood flow is aided by a concomitant reduction of haemostasis (Lippe et al., 1993). Vagal efferent activation of gastric afferent vasodilator nerves Studies of the central and autonomic regulation of gastric functions indicate that efferent parasympathetic pathways in the vagus nerves interact with CGRP-containing afferent nerve fibres in the stomach. This interplay is instrumental for the ability of intracisternal injection of a stable analogue of thyrotropin-releasing hormone (TRH) to augment GMBF. The action of TRH not only requires activation of efferent pathways in the vagus nerves but also depends on an intact innervation of the stomach by capsaicin-sensitive afferents (Table 3.5) and involves ACh, CGRP and NO as mediators, whereas tachykinins acting via NK1 receptors, VIP and prostaglandins have been ruled out (Király et al., 1994a,b,c). Since the gastric hyperaemia evoked by muscarinic ACh receptor activation is also inhibited by CGRP8–37, it is evident that CGRP comes into play secondarily to ACh (Király et al., 1994a,b). This instance is reminiscent of the finding that the delayed component in the ACh-induced dilatation of the rat mesenteric vascular bed involves release of CGRP from capsaicin-sensitive nerve fibres (Takenaga et al., 1995). Analogously, the gastric vasodilator

118

INNERVATION OF THE GASTROINTESTINAL TRACT

TABLE 3.5 Effect of capsaicin-induced defunctionalisation of afferent neurons on haemodynamics in the rat stomach and duodenum. Capsaicin administration

Parameter

Effect of nerve defunctionalisation

References

Systemic

Arteriolar dilatation by acetylcholine Arteriolar dilatation by adenosine Arteriolar dilatation by capsaicin Arteriolar dilatation by CGRP Basal MBFa

No change

Chen, Li and Guth, 1992

No change

Chen, Li and Guth, 1992

Suppression

Chen, Li and Guth, 1992

No change

Chen, Li and Guth, 1992

No change

Lippe, Pabst and Holzer, 1989; Piqué, Esplugues and Whittle, 1990; Raybould et al., 1990, 1992; Thiefin et al., 1990; Holzer, Livingston and Guth,1991; Holzer et al., 1991; Li et al., 1991, 1992; Matsumoto, Takeuchi and Okabe, 1991; Matsumoto et al., 1992a,b; Takeuchi et al., 1991b, 1994; Chen, Li and Guth, 1992; Leung, 1992a; Uchida, Yano and Watanabe, 1993; Király et al., 1994a Uchida, Yano and Watanabe, 1993; Dembinski et al., 1995 Leung, 1992a

Reduction Escape from adrenaline vasoconstriction Increase in MBF by acid backdiffusion

Reduction

Inhibition

No change

Holzer, Livingston and Guth, 1991; Holzer et al., 1994; Leung, 1993b; Matsumoto et al., 1992a,b; Takeuchi et al., 1993, 1994; Grønbech and Lacy, 1996 Lippe, Pabst and Holzer, 1989; Holzer et al., 1991; Matsumoto, Takeuchi and Okabe, 1991; Matsumoto et al., 1992a; Takeuchi et al., 1991b, 1994; Leung, 1992b; Wallace, McKnight and Befus, 1992; Abdel Salam et al., 1996 Konturek et al., 1995b; Heinemann et al., 1996b Mercer et al., 1996

Reduction

Brzozowski et al., 1996a

Suppression

Konturek, Brzozowski and Sliwowski, 1996; Teng et al., 1996a

No change

Konturek et al., 1995a

Increase in MBF by capsaicin

Suppression

Increase in MBF by low dose CCKb Increase in MBF by high dose CCK Increase in MBF by E. coli endotoxin Increase in MBF by epidermal growth factor Increase in MBF by gastrin-17

Reduction

CONTROL OF GASTRIC FUNCTIONS BY EXTRINSIC SENSORY NEURONS

Increase in MBF by pentagastrin Increase in MBF by histamine Increase in MBF by 1 M NaCl

Perivagal

Pericoeliacf

No change

Livingston and Holzer, 1993

Reduction

Konturek et al., 1995b

No change

Leung, 1992a

Reduction

Matsumoto et al., 1992a

119

No change

Grønbech and Lacy, 1996

Increase in MBF by 2 M NaCl Increase in MBF by intracisternal TRHc Reduction of MBF by adrenaline Reduction of MBF by PAFd Reduction of MBF by NOSe inhibitors

No change

Endoh et al., 1993; Grønbech and Lacy, 1996

Inhibition

Király et al., 1994c

No change

Leung, 1992a

Exacerbation

Piqué, Esplugues and Whittle, 1990

Exacerbation

Tepperman and Whittle, 1992; Dembinski et al., 1995

Basal MBF

No change

Increase in MBF by capsaicin Increase in MBF by acid backdiffusion Increase in MBF by vagal nerve stimulation Increase in MBF by intracisternal TRH

No change

Raybould et al., 1990, 1992; Thiefin et al., 1990; Li et al., 1991 Li et al., 1991

No change

Raybould et al., 1992

Reduction

Thiefin et al., 1990

Reduction

Raybould et al., 1990

Basal MBF

No change

Increase in MBF by capsaicin Increase in MBF by acid backdiffusion

Reduction

Li et al., 1991; Raybould et al., 1992 Li et al., 1991

Reduction

Raybould et al., 1992

a

MBF, mucosal blood flow; b CCK, cholecystokinin; c TRH, thyrotropin-releasing hormone; d PAF, plateletactivating factor; e NOS, nitric oxide synthase; f administration of capsaicin locally to the coeliac ganglion.

response to low doses of intravenous CCK octapeptide is brought about by a vagovagal reflex and depends also on ACh, CGRP and NO as vasodilator messengers (Heinemann et al., 1996b). Although histamine acting via H1 receptors does not participate in the rise of GMBF caused by intracisternal TRH, it seems as if the interface between vagal efferents and CGRP-releasing afferents is constituted by mediators released from gastric mast cells (Király et al., 1994b,c).

120

INNERVATION OF THE GASTROINTESTINAL TRACT

TABLE 3.6 Evidence that physiological/pathophysiological functions of the stomach, that involve afferent neurons, are mediated by CGRP and NO. Mediator

Sensory nerve-mediated process

CGRP

Acid-induced increase in blood flow

NO

References

Li et al., 1992; Holzer et al., 1994; Gislason et al., 1995a Capsaicin-induced increase in blood flow Li et al., 1991; Chen, Li and Guth, 1992; Merchant et al., 1994: Chen and Guth, 1995 Increase in blood flow due to CCK-8 Heinemann et al., 1996b Increase in blood flow due to epidermal growth Konturek, Brzozowski and Sliwowski, 1996; factor Teng et al., 1996a Increase in blood flow due to intracisternal TRH Király et al., 1994a,b,c Inhibition of basal gastric acid output Lawson, Mantyh and Pappas, 1994; Kato et al., 1995 Capsaicin-induced inhibition of ACh and gastrin Ren et al., 1993 release Acid-induced inhibition of gastrin release Manela et al., 1995 Stimulation of basal somatostatin release Manela et al., 1995 Acid-induced release of somatostatin Manela et al., 1995 Capsaicin-induced release of somatostatin Inui et al., 1991; Ren et al., 1993 Endogenous gastroprotection Forster and Dockray, 1991a Protection by capsaicin Chiba et al., 1991; Lambrecht et al., 1993a; Peskar et al., 1993; Kato, Yang and Taché, 1994; Merchant et al., 1994 Protection by bombesin, CCK-8 and gastrin-17 Stroff et al., 1993, 1995; Stroff, Plate and Peskar, 1995a Protection by corticotropin-releasing factor Stroff, Plate and Peskar, 1995b Protection by epidermal growth factor Stroff and Peskar, 1995a Protection by NKA-related peptides Stroff et al., 1996 Protection by peptone Stroff and Peskar, 1996 Protection by intracisternal TRH Kato, Yang and Taché, 1994, 1996 Acid-induced inhibition of gastric emptying Forster and Dockray, 1991b Hyperosmolarity-induced inhibition of gastric Forster and Dockray, 1991b emptying Inhibition of gastric emptying after abdominal Plourde et al., 1993; Zittel et al., 1994a surgery Acid-induced increase in blood flow Lippe and Holzer, 1992; Holzer et al., 1994; Gislason et al., 1996 Capsaicin-induced increase in blood flow Whittle, Lopez-Belmonte and Moncada, 1992; Brzozowski et al., 1993; Podolsky et al., 1994; Chen and Guth, 1995; Merchant et al., 1995 Increase in blood flow due to CCK-8, Konturek et al., 1995b,c; Heinemann et al., 1996b gastrin-17 and pentagastrin Increase in blood flow due to E. coli endotoxin Brzozowski et al., 1996a Increase in blood flow due to epidermal growth Konturek, Brzozowski and Sliwowski, 1996 factor Increase in blood flow due to lansoprazole Murakami et al., 1996 Increase in blood flow due to intracisternal Király et al., 1994b TRH Protection by capsaicin Peskar et al., 1991; Brzozowski et al., 1993; Lambrecht et al., 1993a; Podolsky et al., 1994; Merchant et al., 1995

CONTROL OF GASTRIC FUNCTIONS BY EXTRINSIC SENSORY NEURONS

Protection by bombesin, CCK-8, gastrin-17 and pentagastrin

Protection by corticotropin-releasing factor Protection by ecabet Protection by E. coli endotoxin Protection by epidermal growth factor

Protection by hydrotalcit Protection by lansoprazole Protection by NKA-related peptides Protection by intracisternal TRH

121

Lambrecht et al., 1993b; Lambrecht, Stroff and Peskar 1993; Konturek et al., 1995b,c; Stroff et al., 1995; Stroff, Plate and Peskar, 1995a Stroff, Plate and Peskar, 1995b Kinoshita, Kume and Tamaki, 1995 Barrachina et al., 1995; Brzozowski et al., 1996a Stroff and Peskar, 1995a; Konturek, Brzozowski and Sliwowski, 1996 Lambrecht et al., 1993c Murakami et al., 1996 Stroff et al., 1996 Kato, Yang and Taché, 1994

PEPTIDERGIC AFFERENT NERVE FIBRES, VASCULAR PERMEABILITY AND LEUCOCYTE RECRUITMENT IN THE STOMACH PLASMA PROTEIN EXTRAVASATION A prominent feature of neurogenic inflammation in many tissues is an increase in venular permeability, a reaction that facilitates the extravasation of macromolecules, leucocytes and fluid (Holzer, 1992). Capsaicin-induced stimulation of afferent nerve fibres in the rat stomach does not consistently enhance leakage of Evans blue-labelled plasma albumin (Table 3.2), as studies reporting a lack of effect of capsaicin (Holzer and Lippe, 1988; Hatakeyama et al., 1995; Sann et al., 1996) are opposed by reports of moderate increases in vascular permeability caused by capsaicin (Lördal et al., 1996). At any rate, the plasma leakage which capsaicin or electrical stimulation of afferent fibres in the vagus and splanchnic nerves evokes in the stomach is small compared to the plasma protein extravasation seen in the oesophagus and distal rectum (Saria et al., 1983; Lundberg et al., 1984; Lördal et al., 1996). Afferent nerve-mediated increases in venular permeability outside the digestive system are mediated by SP and NKA acting via NK1 receptors (Holzer, 1992). Whether tachykinins mediate the erratic rise of plasma albumin leakage, that in the rat stomach is induced by capsaicin, is not known, and the effects of exogenous tachykinins on vascular permeability in the rat gut are inconsistent. Most studies hold that vascular permeability is not appreciably altered by SP (Saria et al., 1983; Lundberg et al., 1984; Lördal et al., 1996), while one study reports that high dose SP does increase plasma protein leakage in the rat stomach and duodenum (Nicolau et al., 1993). Pharmacological analysis indicates that all three tachykinin receptors control vascular permeability in the

122

INNERVATION OF THE GASTROINTESTINAL TRACT

rat gut in a regionally distinct manner (Nicolau et al., 1993; Lördal et al., 1996). The action of SP in the duodenum is mediated by NK1 receptors, while both NK2 and NK3 receptors might be responsible for the extravasation response in the stomach (Nicolau et al., 1993; Lördal et al., 1996). Since the plasma leakage which SP causes in the oesophagus and mesenteric vasculature of the rat remains unaltered by blockade of histamine H1/ H2 receptors (Saria et al., 1983; Lundberg et al., 1984), it would appear that histamine does not contribute to the response, although SP-positive nerve fibres lie in close proximity to mast cells in the rat mesentery and intestinal mucosa (Skofitsch, Savitt and Jacobowitz, 1985; Stead et al., 1987; Crivellato et al., 1991) and SP can act on rat intestinal mast cells to release histamine and other factors (Shanahan et al., 1985; Marshall et al., 1994). It is important to realise that the influence of afferent nerve fibres on vascular permeability is species-dependent, since capsaicin causes distinct plasma protein extravasation in the gastrointestinal tract of the mouse (Emanueli et al., 1996). Although not yet proven, tachykinins may definitely comediate the neurogenic leakage of plasma proteins in the murine gut, because SP is able to increase vascular permeability via stimulation of NK1 receptors (Emanueli et al., 1996; Figini et al., 1996). Further evidence that endogenous tachykinins participate in the physiological control of vascular permeability in the murine gut comes from the findings that basal plasma protein extravasation is elevated in mice made genetically deficient in neutral endopeptidase, a cell-surface enzyme known to degrade SP, and that a NK1 receptor antagonist corrects the increased vascular permeability in the knockout mice (Figini et al., 1996).

LEUCOCYTE RECRUITMENT Besides increasing vascular permeability and extravasation of macromolecules and fluid, afferent nerve stimulation and afferent nerve-derived sensory neuropeptides influence the adherence, emigration, proliferation and activity of leucocytes (McGillis, Mitsuhashi and Payan, 1990; Sung et al., 1992; Zimmerman, Anderson and Granger, 1992; Smith et al., 1993; Cabaner, Boudard and Bastide, 1995). Although SP enhances the adhesion of leucocytes to the endothelium of rat mesenteric venules (Zimmerman, Gaginella and Granger, 1991) it has not yet specifically been tested whether afferent nerve-derived peptides affect leucocyte recruitment in the stomach. In nonintestinal veins it is well established that SP, CGRP and VIP stimulate the adherence of neutrophils to the endothelium (Zimmerman, Anderson and Granger, 1992) and promote the emigration of neutrophils and eosinophils (Sung et al., 1992), actions which are accompanied by rapid translocation of P-selectin and upregulation of E-selectin expression in endothelial cells (Smith et al., 1993). SP has also been shown to contribute to the granulocyte infiltration which in the rat colon is induced by trinitrobenzene sulphonic acid (McCafferty, Sharkey and Wallace, 1994) and in the rat ileum is provoked by Clostridium difficile toxin A (Pothoulakis et al., 1994). Apart from granulocytes it is also lymphocytes (Stanisz, Befus and Bienenstock, 1986; Agro and Stanisz, 1993; Cabaner, Boudard and Bastide, 1995) and monocytes/macrophages (Bost, Breeding and Pascual, 1992) that may be influenced by afferent nerve-derived

CONTROL OF GASTRIC FUNCTIONS BY EXTRINSIC SENSORY NEURONS

123

neuropeptides, although this interaction has not yet been examined specifically in the stomach. The ability of SP immunoneutralisation to reduce lymphocyte proliferation in the inflamed small intestine of Trichinella spiralis-infected mice (Agro and Stanisz, 1993) and the upregulation of NK1 receptor expression in rat peritoneal macrophages exposed to endotoxin (Bost, Breeding and Pascual, 1992) underlines the pathophysiological potential of neuropeptide-immune system interactions.

PEPTIDERGIC AFFERENT NERVE FIBRES AND GASTRIC SECRETION IMPACT OF AFFERENT NEURONS ON GASTRIC SECRETION In view of their projections to submucosal blood vessels and the mucosa itself (Sternini, Reeve and Brecha, 1987; Green and Dockray, 1988; Sundler, Ekblad and Håkanson, 1991; Jakab et al., 1993; Bäck et al., 1994), extrinsic afferent neurons may, conceivably, influence the secretory activity of the stomach by their regulatory input on the gastric microcirculation and by a direct control of mucosal cell functions. Capsaicin-evoked stimulation of gastric afferent nerve fibres does not, however, significantly alter the basal secretion of acid in the rat and canine stomach (Table 3.2), whereas the elimination of acid from the gastric lumen is increased (Lippe, Pabst and Holzer, 1989; Matsumoto, Takeuchi and Okabe, 1991; Sullivan et al., 1992) as a result of enhanced GMBF (Holzer et al., 1990a, 1991) and bicarbonate secretion (Takeuchi et al., 1992b,c). Only under the pathological conditions of pylorus ligation in conscious rats have capsaicin and resiniferatoxin been found to reduce basal gastric acid output and to blunt the secretory responses to bethanechol, pentagastrin and histamine (Abdel Salam, Szolcsányi and Mózsik, 1994, 1995). The effects of sensory nerve stimulation on the gastric acid/ bicarbonate balance are related to capsaicin’s ability to enhance the release of somatostatin and to inhibit the release of ACh and gastrin in the rat isolated antrum, actions which are at least in part mediated by CGRP (Figure 3.5; Table 3.6). The stimulant effect of sensory nerve stimulation on gastric mucus secretion (Kang et al., 1995a) has not yet been examined systematically. Defunctionalisation of capsaicin-sensitive afferent neurons fails to change the basal secretion of gastric acid, pepsin and bicarbonate (Table 3.7). The basal secretion of mucus in the rat gastric corpus is likewise unaltered while that in the antrum is reduced (Uchida, Yano and Watanabe, 1993). The failure of capsaicin pretreatment to change basal gastric secretion is additionally reflected by its failure to change the permeability of the gastric mucosa to acid and ions (Holzer, Livingston and Guth, 1991; Matsumoto, Takeuchi and Okabe, 1991; Takeuchi et al., 1992c) and suggests that fine afferent nerve fibres do not take part in the regulation of basal gastric secretion. It cannot be ruled out, however, that chronic ablation of afferent neurons triggers compensatory mechanisms that balance the missing input from afferent neurons. This conjecture is supported by the observation that administration of the CGRP antagonist CGRP8–37 to dogs elevates basal acid secretion and amplifies the acid output induced by bombesin (Lawson, Mantyh and

124

INNERVATION OF THE GASTROINTESTINAL TRACT

Figure 3.5 Schematic diagram illustrating two different roles of capsaicin-sensitive afferent neurons in the regulation of acid secretion in the rat stomach. CGRP1-R, CGRP1 receptor; CNS, central nervous system; D, gastric D (somatostatin) cells; G, gastric G (gastrin) cells; NK2-R, tachykinin NK2 receptor.

Pappas, 1994). The acid output which in the rat stomach is caused by pylorus ligation or pentagastrin infusion is likewise facilitated by CGRP8–37 or a monoclonal CGRP antibody (Kato et al., 1995). There is a multitude of data showing that capsaicin-sensitive afferent neurons participate in stimulated gastric secretion (Table 3.7), but owing to discrepancies in the reports it is at present difficult to draw an integrated hypothesis as to how physiological stimulation of gastric secretion depends on input from afferent neurons. Thus, some laboratories hold that acid secretion induced by histamine, pentagastrin or 2-deoxy-D-glucose is reduced by pretreatment of rats with a neurotoxic dose of capsaicin, whereas other groups have failed to see any change (Table 3.7). Another unresolved issue concerns the relationship between vagal efferents and capsaicin-sensitive afferents in the control of gastric acid secretion. The rise of gastric acid output evoked by electrical stimulation of the vagus nerves remains unchanged after systemic pretreatment of rats with capsaicin (Table 3.7) but is attenuated after selective ablation of capsaicin-sensitive vagal afferents (Sharkey et al., 1991). The latter finding has been taken to speculate that vagal afferents may themselves be cholinergic, given that the secretory response to vagal nerve stimulation in untreated rats is abolished by hexamethonium and atropine (Sharkey et al., 1991). A further aspect of this question is mirrored by the observation that the inhibitory effect of a low dose of intracisternal TRH on gastric acid secretion involves ACh, prostaglandins and CGRP (Király et al., 1994c). No hypothesis is at present available that would incorporate all these findings in a coherent manner. There is little controversy over the observations that acid secretion elicited by distension of the gastric wall depends significantly on an intact sensory innervation of the stomach and involves both spinal and vagal afferents (Figure 3.5; Table 3.7). Capsaicininduced ablation of sensory neurons attenuates the acid secretory response to intragastric

CONTROL OF GASTRIC FUNCTIONS BY EXTRINSIC SENSORY NEURONS

125

peptone (Table 3.7), and the inhibition of peptone-stimulated gastric acid secretion, which is seen after intraduodenal administration of lipid, relies also in part on the integrity of vagal, but not spinal, afferent nerve fibres (Lloyd et al., 1993). In contrast, enterogastric inhibition of gastric acid secretion caused by intraduodenal acid is mediated by a reflex that involves both vagal and splanchnic afferents sensitive to the neurotoxin capsaicin (Saperas, Santos and Malagelada, 1995).

TABLE 3.7 Effect of capsaicin-induced defunctionalisation of sensory neurons on exocrine secretion in the rat stomach and duodenum. Capsaicin Parameter administration

Effect of nerve defunctionalisation

References

Systemic

Acid back diffusion

No change

Basal acid output

No change

Basal bicarbonate output

No change

Basal mucus output Basal mucus output Basal pepsin output Enterogastric reflexa (acid) Stimulated acid output (AChb, VNSc)

No change (corpus) Increase (antrum) No change Inhibition No change

Stimulated acid output (2-deoxy-D-glucose)

Reduction

Holzer, Livingston and Guth, 1991; Takeuchi et al., 1993, 1994 Szolcsányi and Barthó, 1981; Alföldi et al., 1986; Dugani and Glavin, 1986; Holzer and Sametz, 1986; Evangelista et al., 1989b; Lippe, Pabst and Holzer, 1989; Raybould and Taché, 1989; Esplugues et al., 1990 Takeuchi et al., 1991a, 1992c; Leung, 1993b Uchida, Yano and Watanabe, 1993 Uchida, Yano and Watanabe, 1993 Szolcsányi and Barthó, 1981 Saperas, Santos and Malagelada, 1995 Alföldi et al., 1986; Evangelista et al., 1989b; Esplugues et al., 1990; Barrachina et al., 1992 Evangelista et al., 1989b

Stimulated acid output (gastric distension) Stimulated acid output (histamine)

No change Suppression Reduction No change

Stimulated acid output (insulin) Stimulated acid output (pentagastrin)

Stimulated acid output (peptone)

No change No change

Reduction Reduction

Matsumoto et al., 1992b Esplugues et al., 1990; Barrachina et al., 1992 Alföldi et al., 1986 Esplugues et al., 1990; Ramos, Esplugues and Esplugues, 1992; Takeuchi et al., 1992a, 1994 Esplugues et al., 1990 Alföldi et al., 1986; Esplugues et al., 1990; Livingston and Holzer, 1993; Saperas et al., 1995 Dugani and Glavin, 1986 Ramos, Esplugues and Esplugues, 1992

126

INNERVATION OF THE GASTROINTESTINAL TRACT

TABLE 3.7 Continued Capsaicin Parameter administration

Perivagal

Effect of nerve defunctionalisation

References

Stimulated acid output (intracisternal TRH) Stimulated bicarbonate output (acid)

Reduction

Saperas et al., 1995

Inhibition

Takeuchi et al., 1991a, 1992c; Hamlet, Jonson and Fändriks, 1992; Inada and Satoh, 1996

Basal acid output

No change

Enterogastric reflex (acid) Enterogastric reflex (lipid) Stimulated acid output (bethanechol) Stimulated acid output (VNS)

No change

Raybould and Taché, 1989; Thiefin et al., 1990; Yoneda and Raybould, 1990; Sharkey et al., 1991; Lloyd et al., 1993; Saperas, Santos and Malagelada, 1995 Saperas, Santos and Malagelada, 1995

Reduction

Lloyd et al., 1993

No change

Raybould and Taché, 1989

No change

Thiefin et al., 1990

Reduction Reduction

Sharkey et al., 1991 Raybould and Taché, 1989

Reduction

Raybould and Taché, 1989

No change

Yoneda and Raybould, 1990

No change

Raybould and Taché, 1989

Reduction

Raybould et al., 1990; Saperas, Santos and Malagelada, 1995 Raybould et al., 1990

Stimulated acid output (gastric distension) Stimulated acid output (histamine) Stimulated acid output (H2Rd agonist) Stimulated acid output (pentagastrin) Stimulated acid output (intracisternal TRHe) Stimulated 5-HTf release (intracisternal TRH) Pericoeliacg

a

No change

Basal acid output

No change

Enterogastric reflex (acid) Enterogastric reflex (lipid) Stimulated acid output (gastric distension) Stimulated acid output (intracisternal TRH)

No change

Esplugues et al., 1990; Lloyd et al., 1993; Saperas, Santos and Malagelada, 1995 Saperas, Santos and Malagelada, 1995

No change

Lloyd et al., 1993

Suppression

Esplugues et al., 1990

No change

Saperas, Santos and Malagelada, 1995

Enterogastric reflex: inhibition of stimulated gastric acid output by intraduodenal acid; b ACh, acetylcholine; VNS, vagal nerve stimulation; d H2R, histamine H2 receptor; e TRH, thyrotropin-releasing hormone; f 5-HT, 5-hydroxytryptamine; g administration of capsaicin locally to the coeliac ganglion. c

CONTROL OF GASTRIC FUNCTIONS BY EXTRINSIC SENSORY NEURONS

127

A synopsis of the available data indicates that afferent neurons play two fundamentally different roles in the regulation of gastric acid secretion (Figure 3.5; Table 3.7). The inhibitory effect of afferent nerve stimulation on gastric acid secretion is likely to reflect the local release and action of CGRP in the stomach. In contrast, the implications of afferent neurons in the secretory responses to gastric distension, peptone and secretagogues point to a vagal reflex whose afferent reflex arc is constituted by capsaicin-sensitive sensory neurons (Figure 3.5). This also applies to the enterogastric inhibition of acid secretion, which is elicited by intraduodenal acid challenge. It is not yet clear whether the local antisecretory and the prosecretory reflex functions of afferent neurons in the stomach are activated simultaneously or separately and whether they are governed by the same or different nerve fibres. The acid-evoked output of bicarbonate from the rat stomach and duodenum is suppressed in capsaicin-pretreated rats (Table 3.7) as is the lansoprazole-induced facilitation of acid-evoked bicarbonate secretion in the rat duodenum (Inada and Satoh, 1996). One study holds that basal pepsin output is decreased by intragastric capsaicin (Brzozowski et al., 1996b) but it is not known whether afferent neurons play a role in stimulated pepsinogen secretion.

MEDIATORS BY WHICH AFFERENT NEURONS CONTROL GASTRIC SECRETION CGRP is most likely the transmitter by which afferent neurons in the rat suppress gastric acid output (Figure 3.5), an inference that is in keeping with the peptide’s high activity in inhibiting gastric acid secretion (Holzer, 1994). The antisecretory action of CGRP in the rat and canine stomach is mediated by CGRP1 receptors (Evangelista, Tramontana and Maggi, 1991; Lawson, Mantyh and Pappas, 1994; Kato et al., 1995) and related to the peptide’s ability to increase the release of somatostatin, while the release of gastrin and ACh is attenuated (Inui et al., 1991; Ren et al., 1991; Holzer, 1994; Manela et al., 1995) as is the release of histamine from enterochromaffin-like cells (Sandor et al., 1996). Tachykinins acting via NK2 receptors do not take part in the antisecretory response to CGRP (Plate et al., 1996). It is increasingly becoming clear that acid-evoked release of CGRP from the gastric mucosa (Table 3.4) represents a negative feedback mechanism (Figure 3.5) that limits excess secretion of acid when the acidity in the gastric juice rises (Kato et al., 1995; Manela et al., 1995; Ren et al., 1995). Thus, accumulation of gastric acid in the stomach seems to excite afferent nerve fibres which release CGRP, because the CGRP1 antagonist CGRP8–37 or a monoclonal CGRP antibody elevates acid output stimulated by pylorus ligation or pentagastrin infusion (Kato et al., 1995). It has moreover been demonstrated that CGRP acting via CGRP1 receptors contributes to the acid-evoked release of somatostatin and to the acid-evoked inhibition of gastrin release from the rat antral mucosa (Manela et al., 1995). An involvement of tachykinins in afferent nerve-mediated changes of gastric secretion has not yet been tested although such an implication is conceivable, given that the intraluminal

128

INNERVATION OF THE GASTROINTESTINAL TRACT

release of SP is related to the acidity of the gastric juice in humans (Mueller et al., 1991). Albeit tachykinins do not influence basal acid secretion in the rat and canine stomach, SP reduces the secretory response to vagal nerve stimulation, intravascular pentagastrin or intraluminal peptone (Konturek et al., 1981; Modlin, Lamers and Walsh, 1981; Martensson et al., 1984; Yokotani and Fujiwara, 1985; Schepp et al., 1990; Coruzzi, Adami and Bertaccini, 1991; Stroff et al., 1996). In contrast, pepsinogen secretion from the canine and guinea-pig stomach is stimulated by tachykinins, an effect that is brought about by NK1 receptors located directly on the chief cells (Vigna et al., 1989; Kitsukawa et al., 1996). The effects of tachykinins on gastric endocrine secretion depend on the species under study. Thus, the secretion of histamine from rat enterochromaffin-like cells (Sandor et al., 1996) and of gastrin from the canine and rat stomach remains unaltered by SP and NKA (Konturek et al., 1981; Modlin, Lamers and Walsh, 1981; Martensson et al., 1984; McIntosh et al., 1987), while gastrin release from the isolated porcine antrum is inhibited by SP and NKA through activation of NK1 and NK2 receptors (Schmidt et al., 1996). Somatostatin release from the isolated porcine antrum is stimulated by activation of NK1 and NK2 receptors (Schmidt et al., 1996), whereas somatostatin release from the rat isolated stomach (Figure 3.5) is attenuated by tachykinins acting via NK2 receptors on gastric D cells (Chiba et al., 1980; Kwok et al., 1985, 1988). An interaction with the somatostatin system is also indicated by the ability of SP to stimulate the expression of somatostatin messenger RNA in the gastric antrum of rats pretreated with a neurotoxic dose of capsaicin, but not of intact animals (Dimaline et al., 1994), although the functional significance of this finding is not yet understood. PACAP, another peptide that occurs in capsaicin-sensitive afferent neurons (Mulder et al., 1994b), is like CGRP able to suppress basal and pentagastrin/histamine-stimulated acid secretion in the rat stomach (Mungan et al., 1995). Whether the antisecretory effect of PACAP, which may be due to a direct action on parietal cells (Mungan et al., 1995), has any bearing on afferent nerve-mediated control of gastric acid output remains to be clarified.

PEPTIDERGIC AFFERENT NERVE FIBRES AND GASTRIC MOTOR ACTIVITY MOTOR EFFECTS AND MEDIATORS OF AFFERENT NERVE STIMULATION Sensory nerve stimulation with capsaicin has been found to exert both stimulant and inhibitory effects on the motor activity of the stomach (Table 3.2). While the longitudinal muscle of the rat isolated fundus is relaxed by capsaicin (Lefebvre, De Beurme and Sas, 1991), composite effects including contraction and relaxation have been observed in the circular muscle of the rat gastric corpus in vitro (Holzer-Petsche, Seitz and Lembeck, 1989). The motor activity of the rat gastric corpus in vivo is enhanced by low intragastric doses of capsaicin while high doses of the drug inhibit gastric motility (Table 3.2). This latter action may be related to the chilli- and capsaicin-induced

CONTROL OF GASTRIC FUNCTIONS BY EXTRINSIC SENSORY NEURONS

129

inhibition of gastric emptying, which has been observed in rats and humans (Table 3.2). The lower oesophageal sphincter pressure in the dog is enhanced by local administration of capsaicin which seems to activate an intrinsic neural reflex involving ganglionic transmission and muscle excitation via muscarinic ACh receptors (Sandler et al., 1993). The relaxant response of the rat stomach to capsaicin depends on extrinsic afferent neurons but does not involve SP, CGRP or VIP (Table 3.3). Tachykinins, though, participate in the capsaicin-induced increase of gastric motor activity (Table 3.3), which is consistent with the overall excitatory influence of tachykinins on gastrointestinal motility (HolzerPetsche, 1995). SP and/or NKA also mediate the capsaicin-evoked increase in lower oesophageal sphincter pressure in the dog (Table 3.3) and the contraction of the feline stomach caused by extrinsic nerve stimulation (Delbro et al., 1983). The contractions which vagal or splanchnic nerve stimulation elicits in the canine and feline stomach are believed to result from antidromic activation of extrinsic afferent neurons which through peripheral transmitter release stimulate cholinergic enteric motor neurons (Delbro et al., 1983; Lidberg et al., 1983; Nakazato et al., 1987; Ito et al., 1993). Analysis of the capsaicininduced contraction in the guinea-pig small intestine suggests that afferent nerve fibres stimulate enteric neurons through unidentified transmitters and that ACh and tachykinins are the transmitters by which enteric motor neurons cause muscle contraction (Barthó et al., 1994; Holzer and Barthó, 1996). Another mechanism by which tachykinins released from afferent nerve fibres may control gastrointestinal motility relates to their ability to depolarise sympathetic ganglion cells in the prevertebral ganglia (Otsuka and Yoshioka, 1993). Spinal afferent nerve fibres containing SP, NKA and CGRP send axon collaterals into the ganglia to make synaptic contacts with postganglionic sympathetic neurons (Hayashi et al., 1982; Matthews and Cuello, 1984; Lee et al., 1987a,b; Su et al., 1987; Green and Dockray, 1988; Kondo and Yamamoto, 1988; Lindh, Hökfelt and Elfvin, 1988), and there is evidence that slow synaptic transmission in the coeliac and mesenteric ganglia of the guinea-pig is at least in part mediated by SP and NKA (Otsuka and Yoshioka, 1993). It is hence likely that primary afferent neurons from the gut, which send tachykinin-releasing axon collaterals into prevertebral sympathetic ganglia, participate in short-loop reflexes that control gastrointestinal effector systems via stimulation of sympathetic output. Such a short-loop reflex is thought to underlie the inhibition of gastric motor activity elicited by intrajejunal acid (Cervero and McRitchie, 1982).

PHYSIOLOGICAL IMPLICATIONS OF AFFERENT NEURONS IN GASTRIC MOTOR CONTROL Capsaicin-sensitive afferents do not seem to play a role in the regulation of basal motor activity in the stomach but may contribute to physiological changes of motility (Table 3.8). Basal gastric motility, gastric emptying and gastrointestinal transit are left unaltered after systemic pretreatment of rats with capsaicin or after selective defunctionalisation of vagal or spinal afferents (Table 3.8). Furthermore, capsaicin-sensitive afferents do not

130

INNERVATION OF THE GASTROINTESTINAL TRACT

TABLE 3.8 Effect of capsaicin-induced defunctionalisation of afferent neurons on motor activity in the oesophagus, stomach and duodenum. Capsaicin Parameter administration

Effect of nerve References defunctionalisation

Systemic

Swallowing reflex Basal GEa and GMAb

Attenuation No change

Basal GITc

No change

Inhibition of GE by i.g.d acid, hyperosmolarity or peptone Inhibition of GMA by intrajejunal acid Inhibition of GMA by coeliac ganglion stimulation Inhibition of GIT by peritoneal irritation Inhibition of GMA by i.p.e acid, bradykinin or capsaicin Inhibition of GE, GMA and GIT by abdominal surgery

Attenuation

Jin et al., 1994 Cervero and McRitchie, 1982; Holzer, 1986; Forster et al., 1990; Holzer-Petsche, 1991; Takeuchi et al., 1991b; Barquist et al., 1992 Holzer, 1986; Holzer, Lippe and Holzer-Petsche, 1986 Forster et al., 1990

Attenuation

Cervero and McRitchie, 1982

No change

Cervero and McRitchie, 1982

Attenuation

Holzer, Lippe and Holzer-Petsche, 1986

Attenuation

Potentiation

Holzer, Lippe and Amann, 1992; Holzer-Petsche, 1992 Holzer, Lippe and Holzer-Petsche,1986; Holzer, Lippe and Amann, 1992; Barquist et al., 1992 Holzer-Petsche, 1991

No change

Matsumoto et al., 1992b

No change

Holzer-Petsche, 1991

Basal GE and GMA

No change

Inhibition of GE by atropine Inhibition of GE and GMA by cholecystokinin Inhibition of GE and GMA by secretin Inhibition of GE by i.d.f acid Inhibition of GMA by i.d. acid Inhibition of GE by i.d. lipid or maltose Inhibition of GMA by i.d. glucose or peptone Inhibition of GMA by i.d. soybean trypsin inhibitor Inhibition of GMA by duodenal distension Inhibition of GE by abdominal surgery Stimulation of GMA by vagal nerve stimulation

No change Attenuation

Attenuation

Raybould and Taché, 1988; Raybould and Hölzer, 1992; Barquist et al., 1996 Raybould and Taché, 1988 Raybould and Taché, 1988; Raybould and Hölzer, 1993b Raybould and Hölzer, 1993b; Lu and Owyang, 1995 Raybould and Hölzer, 1993a Raybould and Hölzer, 1993a Raybould and Hölzer, 1992; Hölzer et al., 1994 Raybould, 1991

Attenuation

Raybould, 1991

Attenuation

Hölzer and Raybould, 1992

No change

Plourde et al., 1993; Barquist et al., 1996 Raybould and Taché, 1988; Lu and Owyang, 1995

Stimulation of GMA by bethanechol Stimulation of GMA by 2-deoxy-D-glucose Stimulation of GMA by substance P and neurokinin A Perivagal

Attenuation

Attenuation No change Attenuation Attenuation

No change

CONTROL OF GASTRIC FUNCTIONS BY EXTRINSIC SENSORY NEURONS

Pericoeliacg

Stimulation of GMA by intracisternal TRH

No change

Raybould et al., 1990

Basal GE and GMA

No change

Inhibition of GE by cholecystokinin Inhibition of GE by secretin Inhibition of GE by i.d. acid Inhibition of GMA by i.d. acid Inhibition of GE by i.d. lipid Inhibition of GE by i.d. maltose Inhibition of GMA by coeliac ganglion stimulation Inhibition of GMA by duodenal distension Inhibition of GE by abdominal surgery

Attenuation

Esplugues et al., 1990; Raybould and Hölzer, 1992; Hölzer et al., 1994; Barquist et al., 1996 Raybould and Hölzer, 1993b

No change No change Attenuation No change Attenuation No change

Raybould and Hölzer, 1993b Raybould and Hölzer, 1993a Raybould and Hölzer, 1993a Hölzer et al., 1994 Raybould and Hölzer, 1992 Esplugues et al., 1990

Attenuation

Hölzer and Raybould, 1992

Attenuation

Plourde et al., 1993; Zittel et al., 1994a; Barquist et al., 1996

Intraduodenal Basal GE No change Inhibition of GE by i.d., Attenuation glucose or lipid Inhibition of GMA by secretin Attenuation

Zittel et al., 1994b Zittel et al., 1994b Lu and Owyang, 1995

a GE, gastric emptying; b GMA, gastric motor activity; c GIT, gastrointestinal transit; d i.g., intragastric; e i.p., intraperitoneal; f i.d., intraduodenal; g administration of capsaicin to the coeliac ganglion.

participate in the stimulation of gastric motility caused by electrical stimulation of the vagus nerves, systemic injection of 2-deoxy-D-glucose, systemic injection of SP/NKA, or intracisternal administration of a high dose of a stable TRH analogue (Table 3.8). In contrast, capsaicin-sensitive afferent neurons contribute to a number of inhibitory gastric motor reflexes including those evoked by intragastric acid, intraduodenal/intrajejunal acid and intraperitoneal administration of acid, capsaicin, bradykinin or CCK (Table 3.8). The inhibition of gastric motor activity and emptying elicited by intraduodenal administration of peptone, hyperosmolar saline, glucose, maltose or lipid or by duodenal distension depends likewise on capsaicin-sensitive afferent neurons (Table 3.8). Further analysis has revealed that the inhibitory motor effect of intraduodenal lipid is mediated by endogenous CCK acting via CCKA receptors on afferent neurons (Hölzer et al., 1994). The inhibition of gastric motility and emptying induced by intravenous secretin or intraduodenal lipid is in part mediated by vagal, but not spinal, afferents, whereas the reverse is true for the gastroparesis caused by abdominal surgery or irritation, which involves spinal, but not vagal, afferents (Table 3.8). Pharmacological studies indicate that the inhibition of gastric emptying, which is seen after intragastric administration of acid or hyperosmolar saline or after abdominal surgery relies on CGRP as a transmitter substance (Table 3.6). CGRP has, in addition, been found to contribute to other afferent nerve-mediated relaxant responses of gastrointestinal smooth muscle (Holzer and Barthó, 1996) but the sites at which CGRP participates in these reactions have not yet been elucidated.

131

132

INNERVATION OF THE GASTROINTESTINAL TRACT

PATHOPHYSIOLOGICAL IMPLICATIONS OF PEPTIDERGIC AFFERENT NERVE FIBRES IN THE STOMACH ABLATION OF SENSORY NEURONS COMPROMISES THE RESISTANCE OF THE GASTRIC MUCOSA TO INJURY Even before the implications of capsaicin-sensitive afferents in the neural regulation of gastric blood flow, secretion and motor activity were elucidated, it had become obvious that these neurons play a role in gastric mucosal defence against injury. The pertinent studies were fostered by the knowledge that capsaicin-sensitive afferent neurons are especially sensitive to noxious chemicals of endogenous and exogenous origin (Holzer, 1991; Wood, 1993) and can give rise to inflammatory processes which are designed to facilitate the recovery from tissue injury. The first evidence that sensory neurons strengthen the resistance of the gastric mucosa to damage was obtained by studying the effects of capsaicin-induced ablation of afferent neurons on experimentally imposed injury of the rat stomach (Table 3.9). Defunctionalisation of sensory neurons does not change the gastric mucosal potential difference (Matsumoto, Takeuchi and Okabe, 1991; Takeuchi et al., 1992c), does not cause damage by itself and fails to alter the gastric injury induced by cold and restraint stress (Table 3.9). Capsaicin-induced ablation of afferent neurons, however, leads to exacerbation of mucosal lesion formation in response to water immersion restraint stress, pylorus ligation, and a variety of injurious factors including hydrochloric acid, acetic acid, ammonia, acetylsalicylic acid (aspirin), indomethacin, platelet-activating factor, a thromboxane mimetic, endothelin-1, cysteamine, 2-deoxy-D-glucose, ethanol and taurocholate (Table 3.9). The involvement of neurons in gastric mucosal protection is further supported by the ability of the nerve conduction blocker tetrodotoxin (Esplugues, Whittle and Moncada, 1989) to enhance experimentally imposed damage to the gastric mucosa. Histological examination of the mucosa has revealed that not only the extent of macroscopically visible lesions but also the depth of damage is augmented by sensory nerve ablation (Esplugues, Whittle and Moncada, 1989, 1992; Piqué, Esplugues and Whittle, 1990; Whittle, Lopez-Belmonte and Moncada, 1990; Holzer, Livingston and Guth, 1991; Whittle and Lopez-Belmonte, 1991; Li et al., 1992; Pabst, Schöninkle and Holzer, 1993; Uchida, Yano and Watanabe, 1993). In addition to aggravating experimentally imposed damage, ablation of sensory neurons compromises the ability of certain factors to protect the gastric mucosa from injury (Table 3.10). The gastric mucosal adaptation to water immersion restraint stress is prevented by capsaicin pretreatment (Brzozowski et al., 1995) while “adaptive cytoprotection” caused by mild gastric irritants such as dilute ethanol, HCl or NaOH is either attenuated (Ko and Cho, 1995) or left unchanged (Evangelista, Maggi and Meli, 1988; Miller et al., 1989; Hatakeyama et al., 1996) by sensory nerve ablation. In contrast, the gastroprotection afforded by the antiulcer drug ecabet, the antacid hydrotalcit, the proton pump inhibitor lansoprazole, prostaglandin E2, NKA-related peptides, epidermal growth factor, corticotropin-releasing factor, intracisternal TRH, γ-aminobutyric acid, low dose endotoxin and laparotomy is reduced or abolished in capsaicin-pretreated rats (Table 3.10). Gastric mucosal protection induced by peptone, bombesin, CCK octapeptide and gastrin-17 is likewise abandoned by pretreatment of rats with capsaicin, although it need be added that

CONTROL OF GASTRIC FUNCTIONS BY EXTRINSIC SENSORY NEURONS

133

TABLE 3.9 Effect of capsaicin-induced defunctionalisation of afferent neurons on experimental lesion formation in the rat gastroduodenal mucosa. Region

Injurious factor

Effect of nerve defunctionalisation

References

Stomach

None

Failure to cause injury

Acetic acid Acetylsalicylic acid

Aggravation of injury Delay of healing Aggravation of injury

Ammonia Cold restraint stress Cysteamine 2-Deoxy-D-glucose

Aggravation of injury Injury unchanged Aggravation of injury Aggravation of injury

Endothelin-1 Ethanola

Aggravation of injury Aggravation of injury

Holzer and Sametz, 1986; Holzer and Lippe, 1988; Esplugues, Whittle and Moncada, 1989; Piqué, Esplugues and Whittle, 1990; Whittle, Lopez-Belmonte and Moncada, 1990; Uchida, Yano and Watanabe, 1993; Brzozowski et al., 1995; Grønbech and Lacy, 1996 Tramontana et al., 1994a Tramontana et al., 1994a Evangelista et al., 1988; Uchida, Yano and Watanabe, 1991a; Abdel Salam, Mózsik and Szolcsányi, 1995; Brzozowski et al., 1996b Uchida, Yano and Watanabe, 1991a Dugani and Glavin, 1986; Holzer et al., 1995 Holzer and Sametz, 1986 Uchida, Yano and Watanabe, 1991a; Matsumoto et al., 1992b Whittle and Lopez-Belmonte, 1991 Holzer and Sametz, 1986; Evangelista et al., 1988; Esplugues and Whittle, 1990; Szolcsányi, 1990; Yonei, Holzer and Guth, 1990; Evangelista and Maggi, 1991; Esplugues, Whittle and Moncada, 1992; Peskar, Nowak and Lambrecht, 1992; Pabst, Schöninkle and Holzer, 1993; Uchida, Yano and Watanabe, 1993; Kato, Yang and Taché, 1994; Abdel Salam, Mózsik and Szolcsányi, 1995; Karmeli et al., 1995

Hydrochloric acid

Restitution unchanged Delay of healing Aggravation of injury

Indomethacin

Injury unchanged Delay of healing Aggravation of injury Delay of restitution Aggravation of injury

Indomethacin plus acid

Aggravation of injury

Hypertonic saline/acid

Pabst, Schöninkle and Holzer, 1993 Peskar et al., 1995 Szolcsányi and Barthó, 1981; Holzer, Livingston and Guth, 1991; Matsumoto et al., 1992b; Raybould et al., 1992 Takeuchi et al., 1994 Takeuchi et al., 1994 Grønbech and Lacy, 1996 Grønbech and Lacy, 1996 Evangelista, Maggi and Meli, 1986; Holzer and Sametz, 1986; Whittle, Lopez-Belmonte and Moncada, 1990 Gray et al., 1994

134

INNERVATION OF THE GASTROINTESTINAL TRACT

TABLE 3.9 Continued Region

Injurious factor

Effect of nerve defunctionalisation

References

NG-monomethylL-arginine Platelet-activating factor

Induction of injury

Pylorus ligation Taurocholate plus acid

Aggravation of injury Aggravation of injury Restitutionb delayed Aggravation of injury

Whittle, Lopez-Belmonte and Moncada, 1990 Esplugues, Whittle and Moncada, 1989; Piqué, Esplugues and Whittle, 1990 Szolcsányi and Barthó, 1981 Takeuchi et al., 1993 Takeuchi et al., 1993 Brzozowski et al., 1995

Aggravation of injury Aggravation of injury Aggravation of injury

Maggi et al., 1987a Takeuchi et al., 1992a Leung, 1993b

Water immersion restraint stress Duodenum

Cysteamine Histamine Hydrochloric acid

Aggravation of injury

a

Aggravation of ethanol-induced damage depends on the concentration of ethanol and the region of the gastric mucosa – with 100% ethanol, damage in the rat gastric corpus may be so severe that sensory nerve defunctionalisation does not further exacerbate damage while aggravation of damage in the antrum region of the stomach is observed (Uchida, Yano and Watanabe, 1993; Peskar et al., 1995); b restitution measured by recovery of transmucosal potential difference.

only low doses of pentagastrin and CCK may in their gastroprotective effect depend on capsaicin-sensitive afferents (Table 3.10). The ability of CGRP, prostaglandin I2, atropine, cimetidine, terbutaline, sodium salicylate and SH-donating or SH-blocking compounds to prevent experimental damage in the rat stomach is left unaltered by defunctionalisation of capsaicin-sensitive afferent neurons (Table 3.10). There is evidence for additional relationships between afferent neurons and some systems that have a bearing on gastric mucosal integrity. Capsaicin-sensitive afferent neurons appear to synergise with sympathetic nerve fibres and substances released from the adrenal glands, since chronic adrenalectomy (Evangelista, Maggi and Meli, 1986; Evangelista et al., 1988) and chemical sympathectomy with guanethidine (Holzer and Sametz, 1986), which aggravate experimental damage on their own, prevent the ability of sensory nerve ablation to worsen experimental injury to the gastric mucosa. Mast cell mediators and free radicals may also play a role, because the capsaicin-evoked exacerbation of gastric damage is suppressed by ketotifen and tempol (Karmeli et al., 1995). Other factors that interact with afferent nerve-mediated gastroprotection include opioid peptides and opiates. The ability of morphine to exacerbate gastric mucosal damage is halted by sensory nerve ablation which aggravates experimental damage by itself (Esplugues, Whittle and Moncada, 1989, 1992; Esplugues and Whittle, 1990). These observations suggest that morphine weakens the resistance of the gastric mucosa to injury by inhibiting the gastroprotective role of capsaicin-sensitive afferent nerve fibres, a conclusion that is consistent with the ability of opioids to inhibit transmitter release from afferent neurons (Maggi, 1991; Holzer, 1992). A further important interaction takes place between capsaicin-

CONTROL OF GASTRIC FUNCTIONS BY EXTRINSIC SENSORY NEURONS

135

TABLE 3.10 Effect of capsaicin-induced defunctionalisation of afferent neurons on mucosal protection in the rat stomach. Protective agent

Effect of nerve ablation on gastroprotection

References

Atropine

No change

Bombesin Calcitonin gene-related peptide Capsaicin

Inhibition No change

Szolcsányi and Mózsik, 1984; Holzer and Sametz, 1986 Stroff, Plate and Peskar, 1995a Gray et al., 1994

Cholecystokinin Cholecystokinin octapeptide Cimetidine

No change Inhibition No change

Corticotropin-releasing factor Ecabet E. coli endotoxin Epidermal growth factor

Inhibition Inhibition Inhibition Inhibition

γ-Aminobutyric acid Gastrin-17

Inhibition Inhibition No change Inhibition Inhibition Inhibition Inhibition No change Inhibition Inhibition No change

Hydrotalcit (antacid) Lansoprazole Laparotomy Neurokinin A4–10 Pentagastrin Peptone Prostaglandin E2 Prostaglandin I2 SH-donating/-blocking compounds Sodium salicylate Terbutaline Intracisternal thyrotropin-releasing hormone analogue

Abolition

No change

Holzer and Lippe, 1988; Holzer et al., 1991; Takeuchi et al., 1991b; Uchida, Yano and Watanabe, 1991b; Gray et al., 1994 Konturek et al., 1995b; Mercer et al., 1996 Evangelista and Maggi, 1991; Stroff et al., 1993 Szolcsányi and Mózsik, 1984; Holzer and Sametz, 1986 Stroff, Plate and Peskar, 1995b Kinoshita, Kume and Tamaki, 1995 Barrachina et al., 1995; Brzozowski et al., 1996a Stroff and Peskar, 1995a; Konturek, Brzozowski and Sliwowski, 1996; Teng et al., 1996a,b Ren et al., 1996 Stroff et al., 1995 Konturek et al., 1995a Lambrecht et al., 1993c Murakami et al., 1996 Yonei, Holzer and Guth, 1990 Stroff et al., 1996 Konturek et al., 1995b Stroff and Peskar, 1996 Esplugues, Whittle and Moncada, 1992 Szolcsányi and Mózsik, 1984; Abdel Salam, Mózsik and Szolcsányi, 1995 Stroff and Peskar, 1995b

No change No change Inhibition

Stroff and Peskar, 1995b Holzer and Sametz, 1986 Kato, Yang and Taché, 1994

sensitive afferent nerve fibres and the NO system. NO synthase inhibitors, which do not cause damage in normal rats, give rise to extensive acid injury in capsaicin-pretreated rats (Whittle, Lopez-Belmonte and Moncada, 1990). This finding indicates that NO acts in concert with afferent neurons in maintaining the resistance of the gastric mucosa to injurious factors.

136

INNERVATION OF THE GASTROINTESTINAL TRACT

STIMULATION OF SENSORY NEURONS STRENGTHENS THE RESISTANCE OF THE RAT AND CANINE GASTRIC MUCOSA TO INJURY Szolcsányi and Barthó (1981) were the first to propose a protective function of capsaicinsensitive sensory neurons in the gastric mucosa, when they observed that intragastric administration of small quantities of capsaicin, to stimulate sensory nerve fibres, protected from injury induced by acid accumulation in the stomach. In subsequent studies it was found that acute intragastric administration of capsaicin prevented experimental injury in the rat stomach induced by pylorus ligation, water immersion restraint stress, haemorrhagic shock or injurious factors such as hydrochloric acid, aspirin, indomethacin, ethanol and acidified taurocholate (Table 3.11). The gastric damage caused by hydrochloric acid or taurocholate in the canine stomach is also suppressed by sensory nerve stimulation with capsaicin (Table 3.11). Numerous studies have shown that the protective effect of capsaicin results in a reduction of the depth of injury while surface damage is not prevented (Holzer, Pabst and Lippe, 1989; Holzer et al., 1990a, 1991; Lambrecht et al., 1993a; Kang et al., 1995a). In contradiction of traditional views, intragastric capsaicin does not irritate the gastric mucosa and does not in any way weaken the gastric mucosal barrier as has been assumed for some red pepper spices which contain many substances other than capsaicin. Capsaicin fails to enhance backdiffusion of gastric acid (Table 3.2), does not appreciably enhance vascular permeability in the gastric wall (Table 3.2), leaves the transmucosal potential difference unchanged (Table 3.2) and fails to cause any macroscopic or histological damage on its own (Table 3.11). It can be ruled out, therefore, that capsaicin strengthens mucosal resistance to injury by virtue of an irritant action on the mucosa. The gastroprotective effect of capsaicin is shared by chilli (Table 3.11). Acute or chronic administration of chilli powder (200 mg as single dose or 200 mg daily for 4 weeks) protects the rat gastric mucosa from ethanol injury (Kang et al., 1995a). Resiniferatoxin is likewise able to reduce the mucosal damage caused by ethanol (Szolcsányi, 1990; Tramontana et al., 1994b; Abdel Salam et al., 1995b), acidified aspirin (Abdel Salam et al., 1995b) and acidified salicylate (Abdel Salam et al., 1995a) in pylorusligated rats.

STIMULATION OF SENSORY NEURONS STRENGTHENS THE RESISTANCE OF THE HUMAN GASTRIC MUCOSA TO INJURY Importantly, the protective effects of capsaicin and chilli powder in the rat stomach are reproduced in the gastric mucosa of humans (Table 3.11). Thus, there is epidemiological evidence that dietary chilli ingestion has a protective action against peptic ulcer disease (Kang et al., 1995b). Careful examination of the multiracial population in Singapore has revealed that the Chinese who eat less chilli than the Malays and Indians have a higher frequency of peptic ulcer disease and that, within the Chinese population, the frequency of ulcer disease correlates inversely with the amount of chilli intake (Kang et al., 1995b). The gastroprotective effect of chilli was directly proven by the ability of 20 g chilli powder to reduce damage of the human gastroduodenal mucosa exposed to 600 mg aspirin

CONTROL OF GASTRIC FUNCTIONS BY EXTRINSIC SENSORY NEURONS

137

TABLE 3.11 Effect of afferent nerve stimulation by capsaicin, resiniferatoxin or chilli on experimental lesion formation in the mucosa of the oesophagus, stomach and duodenum. Species/region

Injurious factor

Effect of nerve stimulation

References

Rabbit oesophagus

Ethanol

Protection

Rat stomach

Acidified acetylsalicylic acid

Protection

Acidified indomethacin Acidified salicylate Acidified taurocholate

Protection

Holzer and Lippe, 1988; Holzer, Pabst and Lippe, 1989; Holzer et al., 1990a, 1991; Lippe, Pabst and Holzer, 1989; Takeuchi et al., 1992b Holzer, Pabst and Lippe, 1989; Abdel Salam et al., 1995b; Hatakeyama et al., 1995 Gray et al., 1994

Protection Protection

Ethanol

Protection

Haemorrhagic shock Hydrochloric acid

Protection Protection

Abdel Salam et al., 1995a Mercer, Ritchie and Dempsey, 1994; Merchant et al., 1994, 1995; Podolsky et al., 1994 Holzer and Lippe, 1988; Holzer et al., 1990a, 1991; Szolcsányi, 1990; Peskar et al., 1991, 1993; Peskar, Nowak and Lambrecht, 1992; Takeuchi et al., 1991b; Uchida, Yano and Watanabe, 1991b; Brzozowski et al., 1993; Lambrecht et al., 1993a; Tramontana et al., 1994b; Abdel Salam et al., 1995b; Kang et al., 1995a; Stroff et al., 1996 Teng et al., 1996b

Pylorus ligation Water immersion restraint stress

Protection Protection

Takeuchi et al., 1994; Abdel Salam, Mózsik and Szolcsányi, 1995; Hatakeyama et al., 1996 Szolcsányi and Barthó, 1981 Brzozowski et al., 1996b

Canine stomach

None Acid Acidified taurocholate

Failure to cause injury Protection Protection

Sullivan et al., 1992 Sullivan et al., 1992 Sullivan et al., 1992

Human stomach

None Acetylsalicylic acid

Failure to cause injury Protection

Graham et al., 1988; Yeoh et al., 1995 Yeoh et al., 1995

Rat duodenum

Hydrochloric acid

Protection

Leung, 1993b

(Table 3.11) and examined by endoscopy and biopsy (Yeoh et al., 1995). Ingestion of chilli alone fails to cause any macroscopic and microscopic injury of the human gastric mucosa (Yeoh et al., 1995), which is also true for ingestion of jalapeno-seasoned food (Graham et al., 1988).

138

INNERVATION OF THE GASTROINTESTINAL TRACT

PATHWAYS OF SENSORY NERVE-MEDIATED GASTRIC MUCOSAL PROTECTION The protective effect of intragastric capsaicin involves afferent nerve fibres because it is prevented by defunctionalisation of extrinsic sensory neurons (Table 3.3). Another indication that neurons are involved is the observation that capsaicin’s ability to reduce ethanol injury is suppressed by nerve conduction blockade with tetrodotoxin (Holzer et al., 1991) or lidocaine (Mercer, Ritchie and Dempsey, 1994). However, the pathways underlying sensory nerve-mediated gastric mucosal protection in the stomach have not yet been fully delineated. Atropine, acute (Abdel Salam et al., 1995a) and chronic (Miller et al., 1989) vagotomy have been reported to prevent capsaicin and resiniferatoxin from protecting the gastric mucosa from experimental injury. Another study, though, has shown that acute vagotomy, acute extirpation of the coeliac ganglion and acute ligation of the blood vessels to the adrenal glands are devoid of any influence on capsaicin’s gastroprotective effect (Holzer and Lippe, 1988). The adaptation to water immersion stress-induced gastric mucosal injury and vasoconstriction is likewise mediated by capsaicin-sensitive afferent neurons of nonvagal origin (Brzozowski et al., 1995). These discrepant data might be accommodated in the conjecture that the ability of capsaicin and resiniferatoxin to strengthen gastric mucosal resistance to injury depends on the local release of transmitters such as CGRP from sensory nerve fibres within the gastric wall, a process in which vagal efferent pathways play a permissive role. Evidence for such an interaction comes from studies with low doses of a stable TRH analogue injected into the cisterna magna (Kato, Yang and Taché, 1994, 1996) or dorsal nucleus of the vagus (Kato, Yang and Taché, 1995b). The gastroprotective action of centrally administered TRH involves activation of parasympathetic efferent neurons in the vagus nerves, ACh, CGRP, NO and prostaglandins (Kato, Yang and Taché, 1994, 1995b, 1996). It is not yet clear, though, whether ACh, CGRP, NO and prostaglandins interact with each other in parallel or in series, and in which order, but the possibility has to be considered that ACh increases the formation of NO, which in turn stimulates CGRP-releasing sensory nerve fibres via a prostaglandin link, as has been shown to occur in the rat skin (Holzer, Jocic and Peskar, 1995). The protective effect of intragastric peptone depends likewise on capsaicinsensitive afferent and cholinergic efferent fibres in the vagus nerve (Stroff and Peskar, 1996). Whether enteric neurons, which can be stimulated by transmitters released from extrinsic afferent nerve fibres (Barthó and Holzer, 1985; Holzer and Barthó, 1996), are implicated in sensory nerve-mediated gastroprotection awaits further investigation. MEDIATORS OF SENSORY NERVE-MEDIATED GASTRIC MUCOSAL PROTECTION CGRP There is compelling evidence that afferent nerve-mediated gastroprotection is mediated by nonadrenergic noncholinergic neurotransmitters (Holzer and Lippe, 1988) among which CGRP plays a central role (Figure 3.4; Tables 3.3 and 3.6). Close arterial administration of this peptide to the rat stomach prevents gastric damage induced by ethanol, acidified aspirin or endothelin-1 (Lippe, Lorbach and Holzer, 1989; Whittle and Lopez-Belmonte,

CONTROL OF GASTRIC FUNCTIONS BY EXTRINSIC SENSORY NEURONS

139

1991; Evangelista, Tramontana and Maggi, 1991; Lambrecht et al., 1993a; LopezBelmonte and Whittle, 1993). Intravenous (Clementi et al., 1993, 1994; Kato, Yang and Taché, 1994), subcutaneous (Maggi et al., 1987b) and intracisternal (Taché, 1992) administration of the peptide has also been reported to protect against gastric injury. The effect of CGRP to reduce ethanol-induced damage is mediated by CGRP1 receptors as it is prevented by the CGRP1 receptor antagonist CGRP8–37 (Lambrecht et al., 1993a; Kato, Yang and Taché, 1994), whereas the ability of CGRP to inhibit aspirin-induced injury is not antagonised by CGRP8–37 (Evangelista, Tramontana and Maggi, 1991). Conclusive pharmacological evidence for an implication of CGRP in afferent nervemediated gastric mucosal protection comes from the findings that the protective effects of intragastric capsaicin are blocked by the CGRP receptor antagonist CGRP8–37 (Lambrecht et al., 1993a; Kato, Yang and Taché, 1994; Merchant et al., 1994, 1995) and by immunoneutralisation of CGRP with polyclonal (Lambrecht et al., 1993a) and monoclonal (Peskar et al., 1993) antibodies to the peptide (Table 3.3). This important role of CGRP in gastric mucosal defence is corroborated by the observation that active immunisation of rats against CGRP exacerbates ethanol-induced injury in the stomach (Forster and Dockray, 1991a). CGRP has in addition been found to mediate the gastroprotective effects of NKArelated peptides, CCK octapeptide, gastrin-17, bombesin, corticotropin-releasing factor, epidermal growth factor and peptone (Table 3.6). The gastroprotective action of capsaicin is suppressed by NO synthase inhibitors (Table 3.3). A role of NO in the gastroprotection afforded by other factors that involve afferent nerve stimulation and CGRP release, such as CCK octapeptide, pentagastrin, gastrin-17, bombesin, corticotropin-releasing factor, epidermal growth factor, the antacid hydrotalcit, the proton pump inhibitor lansoprazole, and low dose endotoxin, has also been demonstrated (Table 3.6). Since the ability of CGRP to strengthen gastric mucosal defence against injurious factors is likewise abrogated by NO synthase inhibitors (Lambrecht et al., 1993a; Clementi et al., 1994) it would appear that stimulation of sensory nerve fibres by capsaicin results in the release of CGRP which, via formation of NO, augments the resistance of the gastric mucosa against experimental injury (Figure 3.4). Such a role of NO is consistent with its ability to prevent gastric mucosal injury and with the effect of NO synthase inhibitors to enhance gastric mucosal vulnerability (Whittle, 1993). Tachykinins The role of SP and NKA in afferent nerve-mediated gastroprotection has long escaped recognition, because SP is devoid of a protective action (Evangelista, Maggi and Meli, 1987b; Evangelista et al., 1989a) and has in fact been found to exaggerate gastric mucosal damage caused by acidified taurocholate or ethanol (Soper and Tepperman, 1986; Karmeli et al., 1991). The deleterious effect of SP involves activation of NK1 and NK3 receptors, mast cell degranulation and subsequent release of histamine, platelet-activating factor, leukotriene B4 and leukotriene C4 (Karmeli et al., 1991, 1993). The discharge of peptidedegrading proteases from mast cells is thought to explain why in the rat gastric mucosa exogenous SP inhibits the CGRP-mediated hyperaemic response to capsaicin (Grønbech and Lacy, 1994). Endogenous tachykinins share this adverse action of exogenous SP, since tachykinin antagonists reduce ethanol-induced injury (Karmeli et al., 1991, 1993;

140

INNERVATION OF THE GASTROINTESTINAL TRACT

Cho, Ko and Tang, 1994; Hayashi et al., 1996). That tachykinins contribute to ethanol damage is, in addition, suggested by the ability of ethanol to release SP into the gastric lumen and by the action of SP to constrict mucosal and submucosal venules (Karmeli et al., 1993; Katori, Ohno and Nishiyama, 1993; Hayashi et al., 1996). While NK1 receptor activation impairs mucosal homeostasis, NK2 receptor stimulation is beneficial for the gastric mucosa, at least under certain experimental conditions. Thus, NKA and related NK2 receptor agonists inhibit ethanol-induced injury to the rat gastric mucosa in spite of a decrease in gastric mucosal blood flow (Evangelista et al., 1989a, 1990; Heinemann et al., 1996a; Stroff et al., 1996). The protective effect of NK2 agonists depends on stimulation of afferent nerve fibres (Table 3.10), release of CGRP and formation of NO (Figure 3.4; Table 3.6). Since the NK2 receptor antagonist MEN-10,627 attenuates the protective action of intragastric capsaicin, it follows that endogenous tachykinins acting via NK2 receptors participate in the sensory nerve-mediated maintenance of gastric mucosal integrity (Stroff et al., 1996). The effect of NKA on gastric mucosal integrity, though, depends on the experimental model under study. While ethanol injury is reduced, damage caused by acid backdiffusion through a disrupted mucosal barrier is exacerbated by NKA, because this tachykinin suppresses the gastric hyperaemia that is evoked by acid backdiffusion and that is important to counteract the deleterious consequences of acid influx into the mucosa (Heinemann et al., 1996a). This antivasodilator action of NKA is mediated by NK2 receptors and thus is independent of the tachykinin’s vasoconstrictor action which in vivo is due to NK1 receptor activation (Heinemann et al., 1996a). Taken together, these data show that both CGRP acting via CGRP1 receptors and tachykinins acting via NK2 receptors are transmitters of afferent nerve-mediated gastroprotection, both transmitters using NO as second messenger (Figure 3.4; Tables 3.3 and 3.6). In addition, there are complex interactions between CGRP and tachykinins because, on the one hand, the gastroprotective effect of NK2 receptor agonists depends on the integrity of the afferent innervation of the stomach and involves release of CGRP (Stroff et al., 1996) and, on the other hand, mucosal protection due to CGRP is blunted by a NK2 receptor antagonist (Plate et al., 1996). Prostaglandins As the data are controversial, there is no conclusive evidence that eicosanoids play a messenger role in sensory nerve-mediated gastric mucosal protection. Intragastric capsaicin does not affect the ex vivo formation of prostaglandin E2, 6-oxo-prostaglandin F1α and leukotriene C4, and indomethacin fails to alter the gastroprotective effect of capsaicin (Holzer et al., 1990a). The ability of CGRP to enhance the resistance of the gastric mucosa to injury remains likewise unaltered by indomethacin (Lambrecht et al., 1993a). Some investigators, though, have found that the gastroprotective action of capsaicin is reduced by indomethacin (Table 3.3), but it has not been ascertained in these studies whether the effect of indomethacin is indeed due to inhibition of prostaglandin synthesis. It is worth considering in this context that prostaglandins may contribute to the activation of afferent nerve fibres under appropriate conditions (Dray and Perkins, 1993), as prostacyclin has been found to release CGRP and SP from rat sensory neurons (Hingtgen and Vasko, 1994).

CONTROL OF GASTRIC FUNCTIONS BY EXTRINSIC SENSORY NEURONS

141

MECHANISMS OF SENSORY NERVE-MEDIATED GASTRIC MUCOSAL PROTECTION Gastric mucosal hyperaemia From the available information it is obvious that CGRP and NO play an important mediator role both in the protective and in the hyperaemic response to capsaicin-evoked stimulation of afferent nerve fibres in the stomach (Figure 3.4). The identity of the messengers involved in the two processes points to a close relationship between the increase in GMBF and the increase in mucosal resistance to injury. The rise of GMBF is indeed also seen when capsaicin is administered together with an injurious concentration of ethanol (Holzer et al., 1991). Capsaicin prevents the ethanol-induced constriction of submucosal arterioles (Ohno et al., 1994), and there is a remarkable correlation between the hyperaemic response to capsaicin and the concomitant reduction of gross damage and deep haemorrhagic erosions (Holzer et al., 1991; Brzozowski et al., 1993). However, this parallelism does not prove that hyperaemia is the primary mechanism by which afferent nerve stimulation strengthens gastric mucosal defence against injury. Evidence is in fact accumulating that protective mechanisms other than vasodilatation are operated by afferent nerve fibres. Hyperaemia, nevertheless, can support a wide range of endogenous mechanisms of gastroprotection as seen with many vasodilator drugs (Guth, Leung and Kauffman, 1989; Holzer, Livingston and Guth, 1994). This interrelationship applies particularly to the gastric output of bicarbonate which is stimulated by sensory nerve stimulation (Takeuchi et al., 1992b,c) and which depends on an adequate blood flow to deliver bicarbonate to the surface mucus layer (Allen et al., 1993; Guttu et al., 1994). Other mechanisms of gastroprotection may be facilitated by afferent nerve-mediated hyperaemia or take place independently of blood flow alterations (Figure 3.6). Hyperaemia-independent mechanisms Several lines of evidence indicate that mechanisms other than a rise of GMBF play a significant part in the gastroprotective action of sensory nerve stimulation (Figures 3.4 and 3.6).

Figure 3.6 Summary of the mediators and presumed mechanisms by which afferent nerve stimulation enhances the resistance of the stomach to injury. CGRP1-R, CGRP1 receptor; NK2-R, tachykinin NK2 receptor.

142

INNERVATION OF THE GASTROINTESTINAL TRACT

The most conclusive hint has come from studies with the NK2 receptor antagonist MEN10,627 which attenuates the gastroprotective action of capsaicin but fails to inhibit the concomitant rise of GMBF (Stroff et al., 1996). This observation is consistent with the ability of NKA-related NK2 receptor agonists to enhance gastric mucosal resistance to experimental injury despite a marked reduction of GMBF (Stroff et al., 1996) and implies that the component of afferent nerve-mediated gastroprotection that involves endogenous tachykinins acting via NK2 receptors is independent of gastric mucosal hyperaemia. There are a number of other findings that indicate a dissociation of the gastroprotective and vasodilator activity of sensory nerve stimulation. For instance, the hyperaemic response to capsaicin in the mucosa of the canine gastric corpus is blocked by lidocaine whereas the protective effect of capsaicin is left unaltered by the local anaesthetic (Sullivan et al., 1992). In a rat gastric chamber model, capsaicin causes a prolonged increase in the resistance of the mucosa to injury that outlasts the increase in GMBF (Morris et al., 1993). Conversely, ablation of capsaicin-sensitive afferent neurons exacerbates the mucosal damage, that is caused by exposure of the rat stomach to hypertonic saline, but fails to abrogate the concurrent gastric vasodilatation (Grønbech and Lacy, 1996). Finally, CGRP counteracts the injurious influence of endothelin-1 and, at the same time, inhibits the hyperaemic response to endothelin-1 in the rat gastric mucosa (Lopez-Belmonte and Whittle, 1993). This observation suggests that CGRP enhances gastric mucosal resistance by protecting the vascular endothelium fom injury (Figure 3.6) rather than by causing vasodilatation (Lopez-Belmonte and Whittle, 1993). The nature of the hyperaemia-independent gastroprotective mechanisms which are operated by capsaicin-sensitive afferent neurons has not yet been identified with certainty although a number of possibilities have been envisaged (Figure 3.6). Inhibition of gastric emptying (Table 3.2) together with a possible increase in fluid secretion may result in dilution of the injurious factors in the gastric juice (Figure 3.6) and thus afford protection of the gastric mucosa (Hatakeyama et al., 1995). Whether the ability of sensory nerve stimulation to modulate gastric acid secretion (Tables 3.2 and 3.3) has any bearing on gastroprotection is not known. For the time being it seems more likely that the capsaicin-induced attenuation of acid output (Abdel Salam, Szolcsányi and Mózsik, 1994, 1995; Brzozowski et al., 1996b) reflects increased bicarbonate secretion (Takeuchi et al., 1992b,c) and acid elimination (Lippe, Pabst and Holzer, 1989) rather than reduced acid secretion. The resiniferatoxin-evoked decrease in acid output in pylorusligated rats (Abdel Salam et al., 1995a,b) may also reflect an increase in bicarbonate secretion rather than an antisecretory effect, given that in other studies capsaicin has failed to attenuate gastric acid secretion (Lippe, Pabst and Holzer, 1989; Matsumoto, Takeuchi and Okabe, 1991; Matsumoto et al., 1992a; Sullivan et al., 1992). Another protective mechanism is reflected by the ability of capsaicin to stimulate the secretion of gastric mucus (Kang et al., 1995a). This activity of capsaicin is mimicked by NO (Brown et al., 1993), whereas the actions of CGRP and tachykinins on gastric mucus production have not yet been studied. Finally, afferent nerve-derived peptide transmitters could modulate gastric mucosal integrity through their influence on the adherence, emigration, proliferation and activity of leucocytes (Figure 3.6) (Stead, Bienenstock and Stanisz, 1987; Sharkey, 1992), a possibility that has not yet been explored experimentally.

CONTROL OF GASTRIC FUNCTIONS BY EXTRINSIC SENSORY NEURONS

143

INFLUENCE OF SENSORY NEURONS ON RESTITUTION AND HEALING OF GASTRIC LESIONS There is mounting evidence that not only acute defence against pending injury but also repair of the injured gastric mucosa is facilitated by the sensory innervation of the stomach (Table 3.9). If the damage involves the superficial epithelium only, the mucosa is quickly repaired by migration of mucous cells over the denuded areas of the lamina propria (Allen et al., 1993). This process of rapid restitution is independent of afferent neurons, because restitution of the ethanol-injured gastric mucosa in the absence of luminal acid remains unchanged in rats treated with a neurotoxic dose of capsaicin (Pabst, Schöninkle and Holzer, 1993). Restitution, however, is promoted by the acid-induced hyperaemia that depends on afferent neurons, as has been observed in the rat stomach exposed to hypertonic saline plus acid (Grønbech and Lacy, 1996). The proper healing of gastric lesions is consistently facilitated by capsaicin-sensitive neurons, as the rate of repair of gastric ulcers induced by hydrochloric acid, acetic acid and ethanol is delayed after sensory nerve ablation (Table 3.9). It seems as if the afferent nerve-promoted acceleration of healing also depends on the sustained hyperaemia that afferent neurons trigger in response to luminal acid (Takeuchi et al., 1994). The retarded healing of gastric lesions seen in capsaicin-pretreated rats is not due to changes in prostaglandin formation, NO generation (Peskar et al., 1995) or gastric acid secretion (Takeuchi et al., 1994) but is associated with increased invasion of polymorphonuclear leucocytes into the gastric wall (Peskar et al., 1995). Particularly worth noting is the observation that capsaicin-pretreated, but not control, rats develop antral ulcers after gastric challenge with ethanol or excess hydrochloric acid (Uchida, Yano and Watanabe, 1993; Takeuchi et al., 1994; Peskar et al., 1995). The formation of antral ulcers has been discussed to arise from a decrease in GMBF and mucus secretion, which following afferent nerve ablation occurs in the gastric antrum but not corpus (Uchida, Yano and Watanabe, 1993). In keeping with a role of afferent neurons in ulcer healing is the observation that sensory nerve stimulation with low dose capsaicin promotes the healing of acetic acid-induced ulcers in the rat stomach (Kang, Teng and Chen, 1996). Interestingly enough, the ulcerhealing effect of capsaicin is counteracted by cimetidine which, unlike capsaicin, decreases GMBF (Kang, Teng and Chen, 1996). Whether the antagonistic influence of cimetidine on capsaicin-evoked promotion of ulcer healing is due to its antivasodilator effect has not directly been shown, and it may also be argued that it is the antisecretory action by which cimetidine removes acid as a factor that stimulates sensory nerve fibres in the stomach. The precise mechanisms by which afferent neurons promote healing of the injured gastric mucosa remain to be determined. It is relevant here to keep in mind that the sensory nerve-derived peptides CGRP, SP and NKA may exert trophic effects (Table 3.7) as they are capable of stimulating the proliferation of fibroblasts, vascular smooth muscle and endothelial cells (Nilsson, von Euler and Dalsgaard, 1985; Payan, 1985; Haegerstrand et al., 1990; Ziche et al., 1990a,b). Indirect evidence suggests that afferent nerve fibres may also exert a trophic influence on gastric mucosal cells. Thus, chronic intake of chilli at doses that are expected to stimulate sensory nerve fibres in the rat stomach enhances gastric mucosal mass but does not change crypt cell proliferation or experimental

144

INNERVATION OF THE GASTROINTESTINAL TRACT

carcinogenesis (Kang et al., 1992). Conversely, ablation of capsaicin-sensitive afferent neurons reduces the weight of the stomach, the gastric synthesis of DNA and the gastric content of DNA and RNA, effects which are exacerbated by concomitant inhibition of NO synthesis (Dembinski et al., 1995).

SENSORY NEURONS SIGNAL FOR AN INCREASE IN GASTRIC MUCOSAL BLOOD FLOW IN THE FACE OF PENDING ACID INJURY Neural pathways underlying the acid-evoked gastric hyperaemia The implications of peptidergic afferent neurons in the regulation of GMBF and gastric mucosal resistance to injury are put into pathophysiological perspective if their role in the gastric responses to acid challenge is considered. Acid backdiffusion through a disrupted gastric mucosal barrier has long been known to cause a prompt increase in GMBF (Whittle, 1977; Bruggeman, Wood and Davenport, 1979; Oates, 1990). Disruption of the rat gastric mucosal barrier with 15% ethanol in the presence of 0.15 M HCl in the gastric lumen causes a 2.5–4 fold rise of blood flow in the left gastric artery (Figure 3.7) and through the gastric mucosa (Holzer, Livingston and Guth, 1991; Holzer et al., 1994). The prompt vasodilator response requires rapid signalling between the acid-threatened surface of the mucosa and resistance vessels in the submucosa. Indeed, an involvement of neurons has been demonstrated by the observations that the acid-evoked rise of gastric blood flow is inhibited by tetrodotoxin (Holzer, Livingston and Guth, 1991) or pretreatment of rats with a neurotoxic dose of capsaicin (Table 3.5). This holds true both for the vasodilator response, which is seen after the gastric mucosal barrier has been disrupted by ethanol (Holzer, Livingston and Guth, 1991; Holzer et al., 1994) or taurocholate (Takeuchi et al., 1993) in the presence of a high luminal acid concentration, and for the hyperaemic

Figure 3.7 Recording of the acid-evoked hyperaemia in the left gastric artery of an urethane-anaesthetised rat. The response was elicited by disruption of the gastric mucosal barrier with ethanol (EtOH, 15%) in the presence of luminal acid (HCl, 0.15 M). The traces from the top to the bottom show heart rate (HR), mean arterial blood pressure (MAP) and blood flow in the left gastric artery (BF/LGA). Data from Holzer et al. (1994).

CONTROL OF GASTRIC FUNCTIONS BY EXTRINSIC SENSORY NEURONS

145

response to excess acid alone, which is seen in the chambered mucosa preparation only (Matsumoto et al., 1992a,b; Takeuchi et al., 1994). Inhibition of the hyperaemic reaction to gastric acid backdiffusion by tetrodotoxin, capsaicin pretreatment, morphine (Holzer, Livingston and Guth, 1991; Raybould et al., 1992), CGRP8–37 (Li et al., 1992), the bradykinin receptor antagonist HOE-140 (Pethö, Jocic and Holzer, 1994) and NKA or SP (Heinemann et al., 1996a) is associated with an aggravation of gross and histological damage to the mucosa. The nerve fibres involved in the acid-induced hyperaemia pass through the coeliac ganglion and hence originate from spinal ganglia (Raybould et al., 1992). The inhibitory effects of tetrodotoxin (Holzer, Livingston and Guth, 1991) and the autonomic ganglion blocking drug hexamethonium (Holzer and Lippe, 1992; Wachter et al., 1995) suggest that the increment of gastric blood flow in the face of pending acid injury results from a reflexlike mechanism (Figure 3.8). However, the individual arcs of this reflex and the possible autonomic or enteric pathways and relays which they use are not yet understood (Holzer, 1992). It is known that the acid-evoked rise of gastric blood flow relies on intact pathways in the splanchnic nerves and through the coeliac/superior mesenteric ganglion complex (Holzer and Lippe, 1992; Wachter et al., 1995), although a contribution from sympathetic noradrenergic neurons (Holzer and Lippe, 1992) and postganglionic parasympathetic or enteric cholinergic neurons (Bruggeman, Wood and Davenport, 1979; Holzer, Livingston and Guth, 1991) has been ruled out. The involvement of ganglionic transmission and pathways running in the splanchnic nerves is difficult to reconcile with the hypothesis that the gastric hyperaemic response to acid backdiffusion results from an axon reflex between mucosal and submucosal collaterals of afferent neurons (Holzer, Livingston and Guth, 1991; Holzer, 1992; Li et al., 1992). The main support for this conjecture comes from the evidence that the vasodilator transmitter of the acid-evoked gastric hyperaemia is CGRP (Table 3.6) which in the rat stomach occurs in extrinsic afferents only (Green and Dockray, 1988; Sternini, 1992). The axon

Figure 3.8 Effect of the CGRP1 receptor antagonist CGRP8–37 and the NO synthase inhibitor NG-nitro-Larginine methylester (L-NAME) on the gastric mucosal hyperaemia caused by gastric acid backdiffusion. D-NAME, inactive enantiomer of L-NAME. Data from Holzer et al. (1994).

146

INNERVATION OF THE GASTROINTESTINAL TRACT

reflex theory is also backed by the ability of acidification to cause a calcium-dependent release of CGRP from rat gastric tissue (Table 3.4), which is in line with other findings showing that hydrogen ions or factors generated in the tissue in response to acidification can stimulate capsaicin-sensitive afferent neurons (Cervero and McRitchie, 1982; Forster et al., 1990; Bevan and Yeats, 1991; Holzer, Lippe and Amann, 1992; Holzer-Petsche, 1992). The axon reflex concept does not explain, however, why hexamethonium, acute extirpation of the coeliac ganglion and acute splanchnic nerve transection inhibit the hyperaemic response to gastric acid backdiffusion (Holzer and Lippe, 1992; Holzer et al., 1994). Since gastric acid backdiffusion does not appear to activate afferent pathways in the spinal cord (Schuligoi et al., 1996), it is hypothesised that the acid-induced increase in GMBF is relayed by a peripheral neural circuitry (Figure 3.8) that depends on an excitatory or inhibitory input from splanchnic and/or enteric neurons (Holzer, 1992). Mediators of the acid-evoked gastric hyperaemia The acid-evoked vasodilatation in the rat stomach depends on CGRP or a related peptide acting via CGRP1 receptors (Figure 3.9), because the CGRP1 receptor antagonist CGRP8–37 depresses the acid-evoked rise of blood flow after disruption of the mucosal barrier with ethanol (Li et al., 1992; Holzer et al., 1994) or taurocholate (Merchant et al., 1995). The vasodilatation which in the rabbit oesophagus is seen after exposure to acidified bile salts is also inhibited by CGRP8–37 (McKie et al., 1994). Like the vasodilator response to exogenous CGRP, the acid-evoked rise of GMBF is blocked by NO synthase inhibitors (Figure 3.9) in an enantiomer-selective manner (Lippe and Holzer, 1992; Holzer et al., 1994). It would appear, therefore, that the gastric mucosal hyperaemia due to acid backdiffusion is mediated by release of CGRP from perivascular afferent nerve fibres (Table 3.6). CGRP in turn stimulates the formation of NO which acts as the final vasodilator messenger

Figure 3.9 Summary of the gastric vasodilator and somatic vasoconstrictor responses to acid backflux in the gastric mucosa. BRAD, bradykinin; 5-HT, 5-hydroxytryptamine.

CONTROL OF GASTRIC FUNCTIONS BY EXTRINSIC SENSORY NEURONS

147

(Figures 3.4 and 3.8). It is obvious that, as regards CGRP and NO, the vasodilator mediators of the hyperaemic response to acid backdiffusion in the rat stomach are identical with those of the hyperaemic response to sensory nerve stimulation with capsaicin. It is important to realise, though, that the vasorelaxant mediator role of NO is confined to the gastric microcirculation (Holzer et al., 1994), since NO synthase inhibitors block the acid-evoked rise of blood flow in the gastric mucosa but not that in the left gastric artery (Holzer et al., 1994). Thus, the effector mechanisms of neurogenic vasodilatation differ in the various levels of the gastric arterial tree. Tachykinin NK1 or NK2 receptor antagonists do not inhibit the hyperaemic response to gastric acid backdiffusion (Holzer et al., 1994; Heinemann et al., 1996a). SP and NKA, however, inhibit the acid-evoked rise of GMBF via activation of NK2 receptors (Heinemann et al., 1996a). The inhibitory effect of NKA on the gastric vasodilator response to acid backdiffusion remains unaltered by ketotifen and seems to be independent of the vasoconstrictor effect of the peptide which in vivo is brought about by stimulation of NK1 receptors (Heinemann et al., 1996a). The significance of the inhibitory effect of tachykinins on the acid-induced vasodilatation in the stomach is not yet clear but may reflect a negative feedback control mechanism. Histamine acting via histamine H1 receptors (Holzer, Livingston and Guth, 1991), vasodilator prostanoids (Lippe and Holzer, 1992), ACh acting via muscarinic receptors (Holzer, Livingston and Guth, 1991) and transmitters of noradrenergic neurons (Holzer and Lippe, 1992) have been ruled out as mediators of the hyperaemic reaction to acid challenge of the rat stomach. Bradykinin, to the contrary, plays some role, given that the bradykinin B2 antagonist HOE-140 (Pethö, Jocic and Holzer, 1994) attenuates the hyperaemia and enhances the mucosal damage elicited by barrier disruption in the presence of a high concentration of intraluminal acid (0.15 M HCl). It would hence seem that amounts of bradykinin sufficient to raise GMBF are formed only during severe acid challenge of the rat gastric mucosa (Pethö, Jocic and Holzer, 1994) but it remains to be examined whether bradykinin causes vasodilatation by its own or by stimulating the neural vasodilator system. Model and species differences in the acid-evoked gastric hyperaemia Model- and species-related differences in the mediators of the gastric hyperaemic response to acid backdiffusion become apparent when the data obtained with acidified ethanol or taurocholate in the rat are compared with those obtained with acidified hypertonic saline in the rat and cat. Sequential exposure of the rat gastric mucosa to hypertonic saline and acid has revealed that the initial hyperaemic response to hypertonic saline is independent of afferent neurons (Endoh et al., 1993; Grønbech and Lacy, 1996) whereas the sustained rise of GMBF in response to acid requires an intact sensory innervation of the stomach (Grønbech and Lacy, 1996). The participation of extrinsic afferent neurons in the hyperaemia, which in the feline stomach is elicited by exposure to hypertonic saline plus acid, has not yet been rigidly tested. Although some CGRP is released into the portal blood and makes a small contribution to the gastric vasodilatation caused by hypertonic saline plus acid (Gislason et al., 1995a), it appears as if histamine acting via histamine H1 and H2 receptors, prostaglandins (Gislason et al., 1995a,b), NO and possibly adenosine

148

INNERVATION OF THE GASTROINTESTINAL TRACT

(Gislason et al., 1996) play a more important role. It follows that, while the gastric vasodilator response to acid backdiffusion is conserved across different species, the transmitters mediating the rise of GMBF appear to be subject to considerable variation. It is not known which neuropeptides mediate gastric mucosal hyperaemia and protection in response to sensory nerve stimulation in the human stomach. There is reason to doubt that CGRP plays a significant role, given that CGRP-containing nerve fibres are scarce in the human stomach (Tsutsumi and Hara, 1989; Sundler, Ekblad and Håkanson, 1991). Since capsaicin releases VIP, but not CGRP, from the human isolated small intestine (Maggi et al., 1989) it is conceivable that VIP, rather than CGRP, is the major vasorelaxant transmitter of extrinsic afferent nerve fibres in the human gut. Somatic vasoconstriction associated with acid-evoked gastric hyperaemia Gastric hyperaemia in response to acid backdiffusion through a disrupted gastric mucosal barrier is aided by a reduction of blood flow to somatic vascular beds (Figure 3.8), e.g. those of the femoral artery (Wachter et al., 1995). Somatic vasoconstriction and gastric vasodilatation peak at different times and represent different entities with regard to pathways, mechanisms and mediators (Wachter et al., 1995, 1998). Further analysis has shown that it is the acid-induced gastric injury which gives rise to femoral vasoconstriction, because the gastric hyperaemia evoked by intragastric capsaicin, which does not injure the mucosa, is not accompanied by a reduction of femoral blood flow (Wachter et al., 1995). Although in part dependent on the extrinsic innervation of the stomach, the gastric acid-evoked constriction of the femoral artery does not involve capsaicin-sensitive afferent neurons and remains unaltered after blockade of ganglionic transmission with hexamethonium, blockade of noradrenergic sympathetic transmission with guanethidine, and ligation of the vascular supply to the adrenal glands (Wachter et al., 1995). Attempts to identify the factors by which gastric acid injury leads to somatic vasoconstriction have ruled out the implication of prostanoids, angiotensin II acting via AT1 receptors, vasopressin acting via V1 receptors, endothelin acting via ETA and ETB receptors (Wachter et al., 1995) and histamine acting via H1 and H2 receptors (Wachter et al., 1998). It has been observed, however, that acid injury of the rat gastric mucosa is associated with luminal release of 5-hydroxytryptamine (5-HT) and that the femoral vasoconstrictor response to gastric acid injury is significantly attenuated by blockade of 5-HT2 and 5-HT3 receptors (Wachter et al., 1998). Taken all available data together, it would seem as if acid challenge of the stomach releases 5-HT from the injured mucosa and that 5-HT in turn stimulates an unknown somatic vasoconstrictor mechanism via capsaicin-insensitive extrinsic neurons of the stomach (Figure 3.8). Physiological relevance of the acid-evoked gastric hyperaemia The observation that inhibition of the hyperaemic response to acid backdiffusion by a variety of drugs (Holzer, Livingston and Guth, 1991; Li et al., 1992; Raybould et al., 1992; Pethö, Jocic and Holzer, 1994; Heinemann et al., 1996a) goes hand in hand with the formation of deep mucosal erosions attests to the important defensive nature of this vascular reaction. By facilitating the disposal of acid, this mechanism prevents the build-up of

CONTROL OF GASTRIC FUNCTIONS BY EXTRINSIC SENSORY NEURONS

149

an injurious concentration of H+ ions in the tissue and thus limits acid damage to the surface of the mucosa (Whittle, 1977; Bruggeman, Wood and Davenport, 1979; Guth, Leung and Kauffman, 1989; Oates, 1990; Holzer, Livingston and Guth, 1994). The afferent nerve-mediated response to acid challenge thus reflects the existence of a neural emergency system which is called into operation when there is pending or actual acid injury to the stomach. In addition, this acid-evoked rise of gastric blood flow creates favourable conditions for rapid restitution (Grønbech and Lacy, 1996) and healing (Takeuchi et al., 1994) of the wounded mucosa. INVOLVEMENT OF AFFERENT NEURONS IN THE ACID-INDUCED SECRETION OF BICARBONATE Another pathophysiologically relevant implication of afferent neurons concerns the regulation of mucosal bicarbonate secretion in response to luminal acidification of the upper gastrointestinal tract (Table 3.2). Stimulation of afferent nerve fibres with capsaicin enhances the mucosal output of alkali from the rat stomach (Takeuchi et al., 1992b,c) and duodenum (Takeuchi et al., 1991a, 1992b), whereas defunctionalisation of capsaicinsensitive afferent neurons prevents these effects of capsaicin (Table 3.3) and blunts the gastric and duodenal secretion of bicarbonate in response to luminal acidification (Table 3.7). As is the case in the stomach, the ability of afferent nerve stimulation to enhance bicarbonate secretion from the duodenum is associated with hyperaemia and results in enhanced resistance to acid-induced deep mucosal damage (Leung, 1993b). Capsaicinsensitive afferent neurons are likewise involved in the effect of lansoprazole to facilitate acid-evoked bicarbonate secretion in the rat duodenum (Inada and Satoh, 1996). The pathological significance of the duodenal alkali response to luminal acidification is reflected by the observation that this process is severely compromised in duodenal ulcer patients infected with Helicobacter pylori (Hogan et al., 1996).

INVOLVEMENT OF AFFERENT NEURONS IN PATHOLOGICAL INHIBITION OF GASTRIC MOTOR ACTIVITY Abdominal surgery and peritoneal irritation or inflammation are known to inhibit gastrointestinal motor activity (Furness and Costa, 1974; Livingston and Passaro, 1990). The acute inhibition of gastric and intestinal motility after abdominal surgery or peritoneal irritation is in part relayed by a neural reflex that consists of capsaicin-sensitive afferent (Table 3.8) and sympathetic efferent (Holzer, Lippe and Holzer-Petsche, 1986; Holzer, Lippe and Amann, 1992; Holzer-Petsche, 1992) neurons. The pathways of the reflex run through the coeliac ganglion but not in the vagus nerves (Table 3.8). As CGRP8–37 and a monoclonal antibody to CGRP ameliorate surgery-induced gastroparesis, it follows that CGRP plays an important transmitter role in the process of acute postoperative motor inhibition (Table 3.6). Such a function is consistent with the ability of CGRP to inhibit gastric motility after local (Maton et al., 1988; Holzer-Petsche, Seitz and Lembeck, 1989; Katsoulis and Conlon, 1989; Lefebvre, De Beurme and Sas, 1991), systemic (Lenz, 1988; Chijiiwa et al., 1992) and intracisternal (Raybould, Kolve and Taché, 1988) administration.

150

INNERVATION OF THE GASTROINTESTINAL TRACT

The site at which CGRP mediates acute postoperative gastroparesis has been suggested to lie within the coeliac ganglion where it acts as a transmitter of an extraspinal intestinogastric inhibitory reflex (Zittel et al., 1994a). Tachykinins and VIP may also play a role, since preoperative administration of a tachykinin or VIP antagonist reduces acute postoperative motor inhibition in the rat and hastens the return of myoelectrical activity in the small intestine (Espat et al., 1995). How abdominal surgery activates afferent neurons to cause reflex inhibition of gastrointestinal motility is not yet fully understood. One study indicates that interleukin-1β is a contributory factor, since the inhibition of gastric emptying seen after abdominal surgery is attenuated by an interleukin-1β receptor antagonist and the ability of interleukin-1β to inhibit gastric emptying is prevented by CGRP8–37 (Coimbra and Plourde, 1996). A role of interleukin-1β in stimulating afferent neurons is consistent with the cytokine’s activity to facilitate the release of CGRP in the skin (Herbert and Holzer, 1994) and that of SP in the spinal cord (Malcangio et al., 1996). While acute shutdown of gastrointestinal motility by surgery or inflammation can be viewed as protective response which prevents movements that could be deleterious to the traumatised gut, continued blockade of motor activity gives rise to the clinical manifestations of ileus. The question as to whether peptidergic afferent neurons contribute to the prolonged, pathological form of postoperative and peritonitis-associated ileus has not yet been addressed. It may be speculated, though, that the conditions of inflammation, bacterial overgrowth and electrolyte imbalance, which are associated with ileus, may contribute to sensitisation of afferent pathways and to recruitment of otherwise inactive afferent neurons (Cervero, 1994; Mayer and Gebhart, 1994) and in this way perpetuate the reflex inhibition of gastrointestinal motility.

SUMMARY AND PERSPECTIVES CAPSAICIN-SENSITIVE AFFERENTS AS A NEURAL EMERGENCY SYSTEM The findings reviewed here prove the existence of a neural emergency system in the stomach, which is called into operation in the face of pending injury to the tissue but is not tonically active. As a result, blood flow to the stomach is greatly augmented, an effect that facilitates the delivery of bicarbonate to the surface epithelium and overlying mucus layer (Guttu et al., 1994), adds to the removal of injurious factors from the mucosa and promotes a wide range of processes that either reduce the vulnerability, or aid the repair, of the gastric mucosa (Guth, Leung and Kauffman, 1989; Oates, 1990; Allen et al., 1993; Holzer, Livingston and Guth, 1994). Apart from facilitation of gastric blood flow, augmentation of bicarbonate and mucus secretion and, under certain conditions, inhibition of acid output and motor activity have been shown to contribute to the overall protective role of peptide-containing afferent neurons. This neural emergency system, which acts in concert with other mechanisms of protection (Whittle, 1993), is operative not only in the stomach but also in other regions of the gastrointestinal system including the oesophagus (Bass et al., 1991; McKie et al., 1994), upper small intestine

CONTROL OF GASTRIC FUNCTIONS BY EXTRINSIC SENSORY NEURONS

151

(Rózsa et al., 1986; Evangelista, Maggi and Meli, 1987a; Maggi et al., 1987a; Remak, Hottenstein and Jacobson, 1990; Hottenstein et al., 1991; Takeuchi et al., 1991a, 1992a,b; Leung, 1993b; Jacobson et al., 1994; Vanner and Surprenant, 1996) and colon (Evangelista and Meli, 1989; Leung, 1992b; Eysselein et al., 1992; Goso et al., 1993; Reinshagen et al., 1994, 1996). Although most of the homeostatic actions are brought about by release of peptide transmitters from their peripheral endings, sensory nerve fibres also constitute the afferent arc of certain autonomic reflexes that regulate secretory and motor functions. Amongst the messenger molecules of afferent neurons, CGRP acting via CGRP1 receptors and tachykinins acting via NK2 receptors have proved to play a central role in enhancing mucosal resistance to pending injury, both classes of transmitter using NO as second messenger (Figures 3.4 and 3.6; Tables 3.3 and 3.6). CGRP and NO are potent vasodilators which, by increasing GMBF, support those protective mechanisms which depend on mucosal hyperaemia. NKA, to the contrary, strengthens mucosal resistance to injury by promoting hyperaemia-independent protective mechanisms. In addition, CGRP and other afferent nerve-derived peptides participate in the postoperative inhibition of gastric motility (Plourde et al., 1993; Zittel et al., 1994a) and in the pathophysiological regulation of gastric motor activity, secretion and immunity. In the long term, afferent neurons seem to be involved in the coordination of trophic and healing processes. This function is envisaged from the delayed restitution and healing of gastric damage in rats pretreated with a neurotoxic dose of capsaicin (Table 3.9) and the ability of SP, NKA and CGRP to stimulate the proliferation of endothelial cells (Haegerstrand et al., 1990; Ziche et al., 1990b), arterial smooth muscle cells (Nilsson, von Euler and Dalsgaard, 1985; Payan, 1985) and fibroblasts (Nilsson, von Euler and Dalsgaard, 1985; Ziche et al., 1990a) and from the ability of SP to stimulate angiogenesis (Ziche et al., 1990b). PATHOPHYSIOLOGICAL CONDITIONS TO ALARM THE NEURAL EMERGENCY SYSTEM To put the functional characteristics of capsaicin-sensitive afferents into proper pathophysiological perspective, the conditions under which these neurons are activated and the functional consequences which may arise from a dysfunction of these neurons need to be considered. A variety of chemicals including hydrochloric acid (Clarke and Davison, 1978; Cervero and McRitchie, 1982; Forster et al., 1990; Bevan and Yeats, 1991; Geppetti et al., 1991; Holzer, Livingston and Guth, 1991; Holzer-Petsche, 1992; Takeuchi et al., 1994; Manela et al., 1995), acetic acid (Leung, 1992b), the bacterial peptide N-formyl-methionylleucyl-phenylalanine (Giuliani et al., 1991) and oxygen radicals (Stahl, Pan and Longhurst, 1993) as well as ischaemia (Pan and Longhurst, 1996) and hypertonicity (Forster et al., 1990; Matsumoto et al., 1992a) are able to stimulate capsaicin-sensitive afferent neurons. These factors will activate the neurons either directly or indirectly via injury-induced formation or liberation of autacoids such as histamine, 5-HT, prostanoids, leukotrienes, bradykinin (Maggi, 1991; Stebbins, Theodossy and Longhurst, 1991; Dray and Perkins, 1993; Geppetti, 1993; Cervero, 1994; Mayer and Gebhart, 1994), interleukin-1β (Herbert and Holzer, 1994; Coimbra and Plourde, 1996; Malcangio et al., 1996), platelet-activating

152

INNERVATION OF THE GASTROINTESTINAL TRACT

factor (Piqué, Esplugues and Whittle, 1990), and endothelin-1 (Whittle and Lopez-Belmonte, 1991). The presence of some or, more likely, a combination of these factors will turn on and/or upregulate the neural emergency system which supports homeostasis by local release of protective transmitters. PRE- AND POSTJUNCTIONAL DEFECTS OF THE NEURAL EMERGENCY SYSTEM With the gastroprotective role of peptidergic afferent neurons in mind it is logical to assume that improper functioning of this neural emergency system is liable to weaken the resistance of the tissue to injurious stimuli and thus may be an aetiological factor in gastroduodenal ulcer disease. Evidence in favour of this conjecture is indeed accumulating (Figure 3.10), as there are reports that either a prejunctional dysfunction of the afferent nerve fibres (sensory neuropathy) or a postjunctional defect in the effector mechanisms disturbs gastric mucosal homeostasis. It need be emphasised, though, that sensory neuropathies resulting from excessive intake of capsaicin-containing peppers are not known. To the contrary, it rather seems as if dietary chilli intake is actually a factor that protects from peptic ulcer disease (Kang et al., 1995b). There are a number of reports that experimental or pathological injury of the stomach is associated with a change in the tissue concentration of CGRP, SP and other neuropeptides (Evangelista et al., 1992, 1993; Wattchow et al., 1992; Kaneko et al., 1993; Tramontana et al., 1994a), but it is not yet known whether these neurochemical alterations reflect changes in the functionality of the neural emergency system and whether they contribute to, or are secondary consequences of, the disease. Reflux of bile is a process that may have a bearing on gastritis and ulcer development, and it is pertinent to note that chronic treatment of rats with oral taurocholate causes a sensory neuropathy that manifests itself in a reduction of the capsaicin-evoked CGRP release and hyperaemia in the gastric mucosa (Narita et al., 1995). Analogously, chronic

Figure 3.10 Schematic diagram of the neural emergency system in the stomach and summary of the pathophysiological conditions that are associated with a malfunction of this system.

CONTROL OF GASTRIC FUNCTIONS BY EXTRINSIC SENSORY NEURONS

153

intake of nicotine by rats has been found to suppress the increase in mucosal blood flow due to gastric acid backdiffusion (Battistel et al., 1993). This observation has been related to a postjunctional depression of the vasodilator response to acid challenge, as chronic nicotine treatment is associated with enhanced formation of vasoconstrictor leukotrienes in the rat stomach (Battistel et al., 1993). Various insults ranging from smoking (Ainsworth et al., 1993) to infection with Helicobacter pylori (Hogan et al., 1996) inhibit the acid-stimulated secretion of bicarbonate from the mucosa of the human duodenum. Since the acid-evoked secretion of bicarbonate in the rat duodenum involves capsaicin-sensitive afferent neurons (Takeuchi et al., 1991a) it would seem as if some of the adverse effects of nicotine, smoking and Helicobacter pylori on the gastroduodenal mucosa may be related to a malfunction of afferent nervecontrolled effector systems, given that in H. pylori infection the acid-induced bicarbonate secretion is blunted due to a defect in the duodenal epithelium (Hogan et al., 1996). Other studies indicate that the gastropathy associated with experimental cirrhosis (Ferraz et al., 1995; Nishizaki et al., 1996) and uraemia (Quintero et al., 1992) is associated with an inadequate rise of GMBF in response to capsaicin or acid backdiffusion through a leaky gastric mucosal barrier. Since the supply of gastric submucosal arterioles by CGRPcontaining nerve fibres is unchanged in experimental cirrhosis (Ferraz et al., 1995; Nishizaki et al., 1996) and uraemia (Quintero et al., 1992) it can be inferred that the gastric circulatory dysregulation arises from a postjunctional defect, possibly in the vascular NO/ cyclic guanosine 3′,5′-cyclic monophosphate signalling pathway (Ferraz et al., 1995). It is worth noting here that capsaicin-sensitive afferent neurons have been suggested to contribute to the development of experimental portal hypertension (Lee and Sharkey, 1993), although such an implication has been refuted in another study (Fernández et al., 1994). Apart from pathological changes in the neural emergency system of the stomach it need be considered that the functionality of the system may change with the development and ageing of the afferent nervous system (Figure 3.10) and the effector control systems which afferent neurons interact with. The capsaicin- and acid-evoked hyperaemia in the rat gastric mucosa (Grønbech and Lacy, 1995; Miyake et al., 1996) and rat superior mesenteric artery (Seno et al., 1996) declines with age, a change that is associated with a significant reduction in the density of CGRP-containing nerve fibres around arterioles in the gastric submucosa (Grønbech and Lacy, 1995). It is hence likely that a sensory neuropathy is in part responsible for the compromised ability of the aged gastric mucosa to defend itself against injurious factors. From a pharmacological/toxicological perspective it need be borne in mind that certain drugs and medicines, that influence the release, action and metabolism of CGRP, NKA and NO, will interfere with the homeostatic function of peptidergic afferent neurons in the stomach. This is particularly true for CGRP and NKA receptor antagonists and NO synthase inhibitors. At the prejunctional level it is morphine and other opioid agonists that inhibit the release of peptide transmitters from afferent nerve fibres (Maggi, 1991; Holzer, 1992). Indeed, morphine worsens the vulnerability of the gastric mucosa by ethanol, platelet-activating factor and acid through an action that involves µ-opioid receptors and that depends on an intact afferent innervation of the stomach (Esplugues, Whittle and Moncada, 1989, 1992; Esplugues and Whittle, 1990; Piqué, Esplugues and Whittle, 1990; Holzer, Livingston and Guth, 1991; Whittle and Lopez-Belmonte, 1991). The release of

154

INNERVATION OF THE GASTROINTESTINAL TRACT

peptide transmitters from afferent nerve fibres may also be under the control of prejunctional adrenoceptors as it is the case in other tissues (Maggi, 1991; Holzer, 1992) but this possibility has not yet been explored in the gastric mucosa. A final question concerns the role that peptidergic afferent neurons play in chronic gastritis, nonulcer dyspepsia, ulcer disease and other functional disorders of the stomach. It is potentially important to note that the expression of SP is upregulated in the gastric mucosa of patients suffering from nonulcer dyspepsia with abdominal pain (Kaneko et al., 1993), but it remains to be elucidated whether this change has any bearing on gastric mucosal homeostasis. Further related to this issue is the question as to whether the local protective function of peptidergic afferent neurons in the stomach is associated with nociception. Although peptidergic afferent neurons supplying the stomach are connected to nociceptors (Cervero, 1994), gastric acid backdiffusion fails to activate spinal neurons as examined by the expression of messenger RNA for c-fos, a marker of neuronal excitation. It would seem, therefore, that the local emergency function of extrinsic afferents takes place without the concomitant transmission of afferent activity to the spinal cord (Schuligoi et al., 1996). It is in place to recall here that ulcer pain is likely to depend on the presence of inflammation in the gastric wall, which sensitises afferent pathways and recruits otherwise inactive nociceptors (Cervero, 1994; Mayer and Gebhart, 1994). The relationship between local emergency function and nociception needs hence to be examined under conditions of chronic injury and inflammation.

ACKNOWLEDGEMENTS Work performed in the author’s laboratory was supported by the Austrian Science Foundation (grants 7845, 9473, 9823 and 11834), the Austrian National Bank (grants 4207, 4905 and 6237) and the Franz Lanyar Foundation at the Medical Faculty of the University of Graz. The author is grateful to Drs. Ulrike Holzer-Petsche and Christof H. Wachter for drawing the computer graphs and to E. Fauland for secretarial help.

REFERENCES Abdel Salam, O.M.E., Szolcsányi, J. and Mózsik, G. (1994). Effect of resiniferatoxin on stimulated gastric acid secretory responses in the rat. Journal of Physiology (Paris), 88, 353–358. Abdel Salam, O.M.E., Mózsik, G. and Szolcsányi, J. (1995). Studies on the effect of intragastric capsaicin on gastric ulcer and on the prostacyclin-induced cytoprotection in rats. Pharmacological Research, 32, 209–215. Abdel Salam, O.M.E., Szolcsányi, J. and Mózsik, G. (1995). Capsaicin and its analogue resiniferatoxin inhibit gastric acid secretion in pylorus-ligated rats. Pharmacological Research, 31, 341–345. Abdel Salam, O.M.E., Bódis, B., Karádi, O., Nagy, L., Szolcsányi, J. and Mózsik, G. (1995a). Stimulation of capsaicin-sensitive sensory peripheral nerves with topically applied resiniferatoxin decreases salicylateinduced gastric H+ back-diffusion in the rat. Inflammopharmacology, 3, 121–133. Abdel Salam, O.M.E., Bódis, B., Karádi, O., Szolcsányi, J. and Mózsik, G. (1995b). Modification of aspirin and ethanol-induced mucosal damage in rats by intragastric application of resiniferatoxin. Inflammopharmacology, 3, 135–147. Abdel Salam, O.M.E., Szolcsányi, J., Pórszász, R. and Mózsik, G. (1996). Effect of capsaicin and resiniferatoxin on gastrointestinal blood flow in rats. European Journal of Pharmacology, 305, 127–136.

CONTROL OF GASTRIC FUNCTIONS BY EXTRINSIC SENSORY NEURONS

155

Agro, A. and Stanisz, A.M. (1993). Inhibition of murine intestinal inflammation by anti-substance P antibody. Regulatory Immunology, 5, 120–126. Ainsworth, M.A., Hogan, D.L., Koss, M.A. and Isenberg, J.I. (1993). Cigarette smoking inhibits acid-stimulated duodenal mucosal bicarbonate secretion. Annals of Internal Medicine, 119, 882–886. Alföldi, P., Obal, F., Toth, E. and Hideg, J. (1986). Capsaicin pretreatment reduces the gastric acid secretion elicited by histamine but does not affect the responses to carbachol and pentagastrin. European Journal of Pharmacology, 123, 321–327. Allen, A., Flemström, G., Garner, A. and Kivilaakso, E. (1993). Gastroduodenal mucosal protection. Physiological Reviews, 73, 823–857. Bäck, N., Ahonen, M., Häppölä, O., Kivilaakso, E. and Kiviluoto, T. (1994). Effect of vagotomy on expression of neuropeptides and histamine in rat oxyntic mucosa. Digestive Diseases and Sciences, 39, 353–361. Barja, F., Mathison, R. and Huggel, H. (1983). Substance P-containing nerve fibres in large peripheral blood vessels of the rat. Cell and Tissue Research, 229, 411–422. Barquist, E., Zinner, M., Rivier, J. and Taché, Y. (1992). Abdominal surgery-induced delayed gastric emptying in rats: role of CRF and sensory neurons. American Journal of Physiology, 262, G616–G620. Barquist, E., Bonaz, B., Martinez, V., Rivier, J., Zinner, M.J. and Taché, Y. (1996). Neuronal pathways involved in abdominal surgery-induced gastric ileus in rats. American Journal of Physiology, 270, R888–R894. Barrachina, M.D., Martinez-Cuesta, M.A., Canet, A., Esplugues, J.V. and Esplugues, J. (1992). Differential effects of locally applied capsaicin on distension-stimulated gastric acid secretion in the anaesthetised rat. Naunyn-Schmiedeberg’s Archives of Pharmacology, 346, 685–690. Barrachina, M.D., Calatayud, S., Moreno, L., Martinez-Cuesta, M.A., Whittle, B.J.R. and Esplugues, V. (1995). Nitric oxide and sensory afferent neurones modulate the protective effects of low-dose endotoxin on rat gastric mucosal damage. European Journal of Pharmacology, 280, 339–342. Barthó, L. and Holzer, P. (1985). Search for a physiological role of substance P in gastrointestinal motility. Neuroscience, 16, 1–32. Barthó, L., Maggi, C.A., Wilhelm, M. and Patacchini, R. (1994). Tachykinin NK1 and NK2 receptors mediate atropine-resistant ileal circular muscle contractions evoked by capsaicin. European Journal of Pharmacology, 259, 187–193. Bass, B.L., Trad, K.S., Harmon, J.W. and Hakki, F.Z. (1991). Capsaicin-sensitive nerves mediate esophageal mucosal protection. Surgery, 110, 419–426. Battistel, M., Plebani, M., Di Mario, F., Jocic, M., Lippe, I.T. and Holzer, P. (1993). Chronic nicotine intake causes vascular dysregulation in the rat gastric mucosa. Gut, 34, 1688–1692. Bauerfeind, P., Hof, R., Cucala, M., Siegrist, S., von Ritter, Ch., Fischer, J.A. and Blum, A.L. (1989). Effects of hCGRP I and II on gastric blood flow and acid secretion in anesthetized rabbits. American Journal of Physiology, 256, G145–G149. Bevan, S. and Yeats, J. (1991). Protons activate a cation conductance in a sub-population of rat dorsal root ganglion neurones. Journal of Physiology (London), 433, 145–161. Bost, K.L., Breeding, S.A. and Pascual, D.W. (1992). Modulation of the mRNAs encoding substance P and its receptor in rat macrophages by LPS. Regulatory Immunology, 4, 105–112. Brown, J.F., Keates, A.C., Hanson, P.J. and Whittle, B.J.R. (1993). Nitric oxide generators and cGMP stimulate mucus secretion by rat gastric mucosal cells. American Journal of Physiology, 265, G418–G422. Bruggeman, T.M., Wood, J.G. and Davenport, H.W. (1979). Local control of blood flow in the dog’s stomach: vasodilatation caused by acid back-diffusion following topical application of salicylic acid. Gastroenterology, 77, 736–744. Brzozowski, T., Drozdowicz, D., Szlachcic, A., Pytko-Polonczyk, J., Majka, J. and Konturek, S. (1993). Role of nitric oxide and prostaglandins in gastroprotection induced by capsaicin and papaverine. Digestion, 54, 24–31. Brzozowski, T., Konturek, S.J., Pytko-Polonczyk, J. and Warzecha, Z. (1995). Gastric adaptation to stress: role of sensory nerves, salivary glands, and adrenal glands. Scandinavian Journal of Gastroenterology, 30, 6–16. Brzozowski, T., Konturek, P.C., Sliwowski, Z. and Konturek, S.J. (1996a). Bacterial endotoxin lipopolysaccharide (LPS) protects gastric mucosa via endogenous prostaglandins (PG), nitric oxide (NO) and sensory nerves. Gut, 39(Suppl. 3), A155. Brzozowski, T., Konturek, P.C., Sliwowski, Z., Pytko-Polonczyk, J., Szlachcic, A. and Drozdowicz, D. (1996b). Role of capsaicin-sensitive sensory nerves in gastroprotection against acid-independent and acid-dependent ulcerogens. Digestion, 57, 424–432. Burnstock, G. (1990). Local mechanisms of blood flow control by perivascular nerves and endothelium. Journal of Hypertension, 8, S95–S106. Cabaner, C., Boudard, F. and Bastide, M. (1995). Modulation of the proliferative response of murine Peyer’s patch lymphocytes by different peptides: CCK-8, CGRP, CCK-PZ and VIP. Preliminary results. International Journal of Immunopathology and Pharmacology, 8, 123–133.

156

INNERVATION OF THE GASTROINTESTINAL TRACT

Cervero, F. (1994). Sensory innervation of the viscera: peripheral basis of visceral pain. Physiological Reviews, 74, 95–138. Cervero, F. and McRitchie, H.A. (1982). Neonatal capsaicin does not affect unmyelinated efferent fibers of the autonomic nervous system: functional evidence. Brain Research, 239, 283–288. Chen, R.Y.Z. and Guth, P.H. (1995). Interaction of endogenous nitric oxide and CGRP in sensory neuroninduced gastric vasodilation. American Journal of Physiology, 268, G791–G796. Chen, R.Y.Z., Li, D.-S. and Guth, P.H. (1992). Role of calcitonin gene-related peptide in capsaicin-induced gastric submucosal arteriolar dilation. American Journal of Physiology, 262, H1350–H1355. Chiba, T., Taminato, T., Kadowaki, S., Inoue, Y., Mori, K., Seino, Y., Abe, H., Chihara, K., Matsukura, S., Fujita, T. and Goto, Y. (1980). Effects of various gastrointestinal peptides on gastric somatostatin release. Endocrinology, 106, 145–14. Chiba, T., Kinoshita, Y., Nakamura, A. and Nakata, Y. (1991). Role of endogenous calcitonin gene-related peptide (CGRP) in gastric mucosal protection in the rat. Gastroenterology, 100, A634. Chijiiwa, Y., Kabemura, T., Misawa, T., Kawakami, O. and Nawata, H. (1992). Direct inhibitory effect of calcitonin gene-related peptide and atrial natriuretic peptide on gastric smooth muscle cells via different mechanisms. Life Sciences, 50, 1615–1623. Cho, C.H., Ko, J.K. and Tang, X.L. (1994). The differential mechanisms of mild irritants on adaptive cytoprotection. Journal of Gastroenterology and Hepatology, 9(Suppl. 1), S24–S28. Clarke, G.D. and Davison, J.S. (1978). Mucosal receptors in the gastric antrum and small intestine of the rat with afferent fibres in the cervical vagus. Journal of Physiology (London), 284, 55–67. Clementi, G., Amico-Roxas, M., Caruso, A., Cutuli, V.M.C., Maugeri, S. and Prato, A. (1993). Protective effects of calcitonin gene-related peptide in different experimental models of gastric ulcers. European Journal of Pharmacology, 238, 101–104. Clementi, G., Caruso, A., Prato, A., De Bernardis, E., Fiore, C.E. and Amico-Roxas, M. (1994). A role for nitric oxide in the anti-ulcer activity of calcitonin gene-related peptide. European Journal of Pharmacology, 256, R7–R8. Coimbra, C.R. and Plourde, V. (1996). Abdominal surgery-induced inhibition of gastric emptying is mediated in part by interleukin-1β. American Journal of Physiology, 270, R556–R560. Coruzzi, G., Adami, M. and Bertaccini, G. (1991). Effect of substance P and related neurokinins on gastric acid secretion. Advances in Experimental Medicine and Biology, 298, 157–165. Costa, M., Furness, J.B. and Gibbins, I.L. (1986). Chemical coding of enteric neurons. Progress in Brain Research, 68, 217–239. Crivellato, E., Damiani, D., Mallardi, F. and Travan, L. (1991). Suggestive evidence for a microanatomical relationship between mast cells and nerve fibres containing substance P, calcitonin gene related peptide, vasoactive intestinal polypeptide, and somatostatin in the rat mesentery. Acta Anatomica (Basel), 141, 127–131. Delbro, D., Fändriks, L., Rosell, S. and Folkers, K. (1983). Inhibition of antidromically induced stimulation of gastric motility by substance P receptor blockade. Acta Physiologica Scandinavica, 118, 309–316. Dembinski, A., Warzecha, Z., Ceranowicz, P. and Konturek, S.J. (1995). The role of capsaicin-sensitive sensory neurons and nitric oxide in regulation of gastric mucosal growth. Journal of Physiology and Pharmacology, 46, 351–362. Dimaline, R., Evans, D., Forster, E. and Dockray, G.J. (1994). Stimulation of gastric somatostatin mRNA abundance by substance P in capsaicin-treated rats. Neuroscience Letters, 172, 39–41. Dray, A. and Perkins, M. (1993). Bradykinin and inflammatory pain. Trends in the Neurosciences, 16, 99–104. Dugani, A.M. and Glavin, G.B. (1986). Capsaicin effects on stress pathology and gastric acid secretion in rats. Life Sciences, 39, 1531–1538. Ekblad, E., Ekelund, M., Graffner, H., Håkanson, R. and Sundler, F. (1985). Peptide-containing nerve fibers in the stomach wall of rat and mouse. Gastroenterology, 89, 73–85. Emanueli, C., Figini, M., Calaustro, E., Gamp, P.D., Gerard, C., Geppetti, P. and Bunnett, N.W. (1996). Bradykinin (BK) and substance P (SP) stimulate plasma extravasation in the gastrointestinal tract and pancreas of the mouse. Gastroenterology, 110, A1069. Endoh, K., Kao, J., Domek, M.J. and Leung, F.W. (1993). Mechanism of gastric hyperemia induced by intragastric hypertonic saline in rats. Gastroenterology, 104, 114–121. Entzeroth, M., Doods, H.N., Wieland, H.A. and Wienen, W. (1994). Adrenomedullin mediates vasodilation via CGRP1 receptors. Life Sciences, 56, PL19–PL25. Espat, N.J., Cheng, G., Kelley, M.C., Vogel, S.B., Sninsky, C.A. and Hocking, M.P. (1995). Vasoactive intestinal peptide and substance P receptor antagonists improve postoperative ileus. Journal of Surgical Research, 58, 719–723. Esplugues, J.V. and Whittle, B.J.R. (1990). Morphine potentiation of ethanol-induced gastric mucosal damage in the rat. Role of local sensory afferent neurons. Gastroenterology, 98, 82–89.

CONTROL OF GASTRIC FUNCTIONS BY EXTRINSIC SENSORY NEURONS

157

Esplugues, J.V., Whittle, B.J.R. and Moncada, S. (1989). Local opioid-sensitive afferent sensory neurones in the modulation of gastric damage induced by Paf. British Journal of Pharmacology, 97, 579–585. Esplugues, J.V., Ramos, E.G., Gil, L. and Esplugues, J. (1990). Influence of capsaicin-sensitive afferent neurones on the acid secretory responses of the rat stomach in vivo. British Journal of Pharmacology, 100, 491–496. Esplugues, J.V., Whittle, B.J.R. and Moncada, S. (1992). Modulation by opioids and by afferent sensory neurones of prostanoid protection of the rat gastric mucosa. British Journal of Pharmacology, 106, 846–852. Eysselein, V.E., Reinshagen, M., Patel, A., Davis, W., Nast, C. and Sternini, C. (1992). Calcitonin gene-related peptide in inflammatory bowel disease and experimentally induced colitis. Annals of the New York Academy of Sciences, 657, 319–327. Evangelista, S. and Maggi, C.A. (1991). Protection induced by cholecystokinin-8 (CCK-8) in ethanol-induced gastric lesions is mediated via vagal capsaicin-sensitive fibres and CCK-A receptors. British Journal of Pharmacology, 102, 119–122. Evangelista, S. and Meli, A. (1989). Influence of capsaicin-sensitive fibres on experimentally-induced colitis in rats. Journal of Pharmacy and Pharmacology, 41, 574–575. Evangelista, S., Maggi, C.A. and Meli, A. (1986). Evidence for a role of adrenals in the capsaicin-sensitive “gastric defence mechanism” in rats. Proceedings of the Society for Experimental Biology and Medicine, 182, 568–569. Evangelista, S., Maggi, C.A. and Meli, A. (1987a). Involvement of capsaicin-sensitive mechanism(s) in the antiulcer defence of intestinal mucosa in rats. Proceedings of the Society for Experimental Biology and Medicine, 184, 264–266. Evangelista, S., Maggi, C.A. and Meli, A. (1987b). Influence of peripherally administered peptides on ethanolinduced gastric ulcers in the rat. General Pharmacology, 18, 647–649. Evangelista, S., Maggi, C.A. and Meli, A. (1988). Lack of influence of capsaicin-sensitive sensory fibers on adaptive cytoprotection in rat stomach. Digestive Diseases and Sciences, 33, 1050–1051. Evangelista, S., Maggi, C.A., Giuliani, S. and Meli, A. (1988). Further studies on the role of the adrenals in the capsaicin-sensitive “gastric defence mechanism”. International Journal of Tissue Reactions, 10, 253–255. Evangelista, S., Lippe, I.T., Rovero, P., Maggi, C.A. and Meli, A. (1989a). Tachykinins protect against ethanolinduced gastric lesions in rats. Peptides, 10, 79–81. Evangelista, S., Santicioli, P., Maggi, C.A. and Meli, A. (1989b). Increase in gastric secretion induced by 2-deoxy-D-glucose is impaired in capsaicin pretreated rats. British Journal of Pharmacology, 98, 35–37. Evangelista, S., Maggi, C.A., Rovero, P., Patacchini, R., Giuliani, S. and Giachetti, A. (1990). Analogs of neurokinin A(4–10) afford protection against gastroduodenal ulcers in rats. Peptides, 11, 293–297. Evangelista, S., Renzi, D., Guzzi, P. and Maggi, C.A. (1991). Interactions between sialoadenectomy and capsaicinsensitive afferent fibers on gastric acid secretion in rats. Life Sciences, 48, PL37–PL41. Evangelista, S., Tramontana, M. and Maggi, C.A. (1991). Pharmacological evidence for the involvement of multiple calcitonin gene-related peptide (CGRP) receptors in the antisecretory and antiulcer effect of CGRP in rat stomach. Life Sciences, 50, PL13–PL18. Evangelista, S., Renzi, D., Tramontana, M., Surrenti, C., Theodorsson, E. and Maggi, C.A. (1992). Cysteamineinduced duodenal ulcers are associated with a selective depletion in gastric and duodenal calcitonin gene-related peptide-like immunoreactivity in rats. Regulatory Peptides, 39, 19–28. Evangelista, S., Tramontana, M., Panerai, C., Surrenti, C. and Renzi, D. (1993). Gastric lesions induced by concentrated ethanol are associated with a decrease in gastric calcitonin gene-related peptide-like immunoreactivity in rats. Scandinavian Journal of Gastroenterology, 28, 1112–1114. Fernández, M., Casadevall, M., Schuligoi, R., Pizcueta, P., Panés, J., Barrachina, M.D., Donnerer, J., Piqué, J.M., Esplugues, J.V., Bosch, J., Rodés, J. and Holzer, P. (1994). Neonatal capsaicin treatment does not prevent splanchnic vasodilatation in portal-hypertensive rats. Hepatology, 20, 1609–1614. Ferraz, J.G.P., McKnight, W., Sharkey, K.A. and Wallace, J.L. (1995). Impaired vasodilatory responses in the gastric microcirculation of anesthetized rats with secondary biliary cirrhosis. Gastroenterology, 108, 1183–1191. Figini, M., Emanueli, C., Geppetti, P., Lovett, M., Gamp, P.D., Kirkwood, K. et al. (1996). Spontaneous plasma extravasation in the gastrointestinal tract of neutral endopeptidase (NEP) knockout mice is mediated by bradykinin (BK) and substance P (SP). Gastroenterology, 110, A1071. Forster, E.R. and Dockray, G.J. (1991a). The role of calcitonin gene-related peptide in gastric mucosal protection in the rat. Experimental Physiology, 76, 623–626. Forster, E.R. and Dockray, G.J. (1991b). Immunoneutralization suggests that calcitonin gene-related peptide regulates gastric emptying in the rat. Neuroscience Letters, 131, 5–7. Forster, E.R. and Southam, E. (1993). The intrinsic and vagal extrinsic innervation of the rat stomach contains nitric oxide synthase. NeuroReport, 4, 275–278. Forster, E.R., Green, T., Elliot, M., Bremner, A. and Dockray, G.J. (1990). Gastric emptying in rats: role of afferent neurons and cholecystokinin. American Journal of Physiology, 258, G552–G556.

158

INNERVATION OF THE GASTROINTESTINAL TRACT

Fujimiya, M. and Kwok, Y.N. (1995). Effect of capsaicin on release of substance P-like immunoreactivity from vascularly perfused rat duodenum. Digestive Diseases and Sciences, 40, 96–99. Furness, J.B. and Costa, M. (1974). Adynamic ileus, its pathogenesis and treatment. Medical Biology, 52, 82–89. Furness, J.B., Papka, R.E., Della, N.G., Costa, M. and Eskay, R.L. (1982). Substance P-like immunoreactivity in nerves associated with the vascular system of guinea-pigs. Neuroscience, 7, 447–459. Gardiner, S.M., Compton, A.M., Kemp, P.A., Bennett, T., Bose, C., Foulkes, R. and Hughes, B. (1991). Antagonistic effect of human α-calcitonin gene-related peptide (8–37) on regional hemodynamic actions of rat islet amyloid polypeptide in conscious Long-Evans rats. Diabetes, 40, 948–951. Gates, T.S., Zimmerman, R.P., Mantyh, C.R., Vigna, R., Maggio, J.E., Welton, M.L., Passaro, E.P. and Mantyh, P.W. (1988). Substance P and substance K receptor binding sites in the human gastrointestinal tract: localization by autoradiography. Peptides, 9, 1207–1219. Gates, T.S., Zimmerman, R.P., Mantyh, C.R., Vigna, S.R. and Mantyh, P.W. (1989). Calcitonin gene-related peptide-receptor binding sites in the gastrointestinal tract. Neuroscience, 31, 757–770. Geppetti, P. (1993). Sensory neuropeptide release by bradykinin: mechanisms and pathophysiological implications. Regulatory Peptides, 47, 1–23. Geppetti, P. and Holzer, P. (1996). Neurogenic Inflammation. Boca Raton: CRC Press. Geppetti, P., Frilli, S., Renzi, D., Santicioli, P., Maggi, C.A., Theodorsson, E. and Fanciullacci, M. (1988). Distribution of calcitonin gene-related peptide-like immunoreactivity in various rat tissues: correlation with substance P and other tachykinins and sensitivity to capsaicin. Regulatory Peptides, 23, 289–298. Geppetti, P., Tramontana, M., Evangelista, S., Renzi, D., Maggi, C.A., Fusco, B.M. et al. (1991). Differential effect on neuropeptide release of different concentrations of hydrogen ions on afferent and intrinsic neurons of the rat stomach. Gastroenterology, 101, 1505–1511. Gibbins, I.L., Furness, J.B., Costa, M., MacIntyre, I., Hillyard, C.J. and Girgis, S. (1985). Co-localization of calcitonin gene-related peptide-like immunoreactivity with substance P in cutaneous, vascular and visceral sensory neurons of guinea pigs. Neuroscience Letters, 57, 125–130. Gislason, H., Guttu, K., Sørbye, H., Schifter, S., Waldum, H.L. and Svanes, K. (1995a). Role of histamine and calcitonin gene-related peptide in the hyperemic response to hypertonic saline and H+ back-diffusion in the gastric mucosa of cats. Scandinavian Journal of Gastroenterology, 30, 300–310. Gislason, H., Sørbye, H., Abdi-Dezfuli, F., Waldum, H.L. and Svanes, K. (1995b). Role of prostaglandins and histamine in hyperemic response to superficial and deep gastric mucosal injury and H+ back-diffusion in cats. Digestive Diseases and Sciences, 40, 1669–1678. Gislason, H., Varhaug, P., Sørbye, H., Waldum, H.L. and Svanes, K. (1996). Role of adenosine and nitric oxide in the hyperemic response to superficial and deep gastric mucosal injury and H+ back-diffusion in cats. Scandinavian Journal of Gastroenterology, 31, 14–23. Giuliani, S., Santicioli, P., Tramontana, M., Geppetti, P. and Maggi, C.A. (1991). Peptide N-formyl-methionylleucyl-phenylalanine (FMLP) activates capsaicin-sensitive afferent nerves in guinea-pig atria and urinary bladder. British Journal of Pharmacology, 102, 730–734. Goso, C., Evangelista, S., Tramontana, M., Manzini, S., Blumberg, P.M. and Szallasi, A. (1993). Topical capsaicin administration protects against trinitrobenzene sulfonic acid-induced colitis in the rat. European Journal of Pharmacology, 249, 185–190. Graham, D.Y., Smith, J.L. and Opekun, A.R. (1988). Spicy food and the stomach. Evaluation by videoendoscopy. Journal of the American Medical Association, 260, 3473–3475. Gray, J.L., Bunnett, N.W., Orloff, S.L., Mulvihill, S.J. and Debas, H.T. (1994). A role for calcitonin gene-related peptide in protection against gastric ulceration. Annals of Surgery, 219, 58–64. Green, T. and Dockray, G.J. (1988). Characterization of the peptidergic afferent innervation of the stomach in the rat, mouse, and guinea-pig. Neuroscience, 25, 181–193. Grønbech, J.E. and Lacy, E.R. (1994). Substance P attenuates gastric mucosal hyperemia after stimulation of sensory neurons in the rat stomach. Gastroenterology, 106, 440–449. Grønbech, J.E. and Lacy, E.R. (1995). Role of gastric blood flow in impaired defense and repair of aged rat stomachs. American Journal of Physiology, 269, G737–G744. Grønbech, J.E. and Lacy, E.R. (1996). Role of sensory afferent neurons in hypertonic damage and restitution of the rat gastric mucosa. Gastroenterology, 111, 1474–1483. Guth, P.H., Leung, F.W. and Kauffman, G.L. (1989). Physiology of the gastric circulation. In The Gastrointestinal System. Handbook of Physiology, Section 6, Vol. I, Part 2, edited by S.G. Schultz, pp. 1371–1404. Bethesda: American Physiological Society. Guttu, K., Sørbye, H., Gislason, H., Svanes, K. and Grønbech, J.E. (1994). Role of bicarbonate in blood flowmediated protection and repair of damaged gastric mucosa in the cat. Gastroenterology, 107, 149–159. Haegerstrand, A., Dalsgaard, C.J., Jonzon, B., Larsson, O. and Nilsson, J. (1990). Calcitonin gene-related peptide stimulates proliferation of human endothelial cells. Proceedings of the National Academy of Sciences USA, 87, 3299–3303.

CONTROL OF GASTRIC FUNCTIONS BY EXTRINSIC SENSORY NEURONS

159

Hamlet, A., Jonson, C. and Fändriks, L. (1992). The mediation of increased duodenal alkaline secretion in response to 10 mM HCl in the anaesthetized rat: support for the involvement of capsaicin-sensitive nerve elements. Acta Physiologica Scandinavica, 146, 519–525. Hatakeyama, Y., Matsuo, M., Tomoi, M., Ohtsuka, M. and Shimomura, K. (1995). Luminal dilution caused by certain mild irritants and capsaicin contributes to their gastric mucosal protection. American Journal of Physiology, 268, G200–G206. Hatakeyama, Y., Tomoi, M., Ohtsuka, M. and Shimomura, K. (1996). Implication of sensory neurons in the diverse mechanisms of adaptive cytoprotection in the rat stomach. Japanese Journal of Pharmacology, 70, 347–350. Hayashi, H., Ohsumi, K., Ueda, N., Fujiwara, M. and Mizuno, N. (1982). Effect of spinal ganglionectomy on substance P-like immunoreactivity in the gastroduodenal tract of cats. Brain Research, 232, 227–230. Hayashi, H., Nishiyama, K., Majima, M., Katori, M. and Saigenji, K. (1996). Role of endogenous substance P in ethanol-induced mucosal damage in the rat stomach. Journal of Gastroenterology, 31, 314–322. Heinemann, Á., Jocic, M., Holzer-Petsche, U., Pethö, G., Peskar, B.M., Horwell, D.C. and Holzer, P. (1995). Mediation by CCKB receptors of the CCK-evoked hyperaemia in rat gastric mucosa. British Journal of Pharmacology, 116, 2274–2278. Heinemann, Á., Jocic, M., Herzeg, G. and Holzer, P. (1996a). Tachykinin inhibition of acid-induced gastric hyperaemia in the rat. British Journal of Pharmacology, 119, 1525–1532. Heinemann, Á., Jocic, M., Peskar, B.M. and Holzer, P. (1996b). CCK-evoked hyperemia in rat gastric mucosa involves neural mechanisms and nitric oxide. American Journal of Physiology, 270, G253–G258. Herbert, M.K. and Holzer, P. (1994). Interleukin-1β enhances capsaicin-induced neurogenic vasodilatation in the rat skin. British Journal of Pharmacology, 111, 681–686. Hingtgen, C.M. and Vasko, M.R. (1994). Prostacyclin enhances the evoked release of substance P and calcitonin gene-related peptide from rat sensory neurons. Brain Research, 655, 51–60. Hogan, D.L., Rapier, R.C., Dreilinger, A., Koss, M.A., Basuk, P.M., Weinstein, W.M., Nyberg, L.M. and Isenberg, J.I. (1994). Duodenal bicarbonate secretion: eradication of Helicobacter pylori and duodenal structure and function in humans. Gastroenterology, 110, 705–716. Hölzer, H.H. and Raybould, H.E. (1992). Vagal and splanchnic sensory pathways mediate inhibition of gastric motility induced by duodenal distension. American Journal of Physiology, 262, G603–G608. Hölzer, H.H., Turkelson, C.M., Solomon, T.E. and Raybould, H.E. (1994). Intestinal lipid inhibits gastric emptying via CCK and a vagal capsaicin-sensitive afferent pathway in rats. American Journal of Physiology, 267, G625–G629. Holzer, P. (1986). Capsaicin-sensitive afferent neurones and gastrointestinal propulsion in the rat. NaunynSchmiedebergs Archives of Pharmacology, 332, 62–65. Holzer, P. (1988). Local effector functions of capsaicin-sensitive sensory nerve endings: involvement of tachykinins, calcitonin gene-related peptide and other neuropeptides. Neuroscience, 24, 739–768. Holzer, P. (1991). Capsaicin: cellular targets, mechanisms of action, and selectivity for thin sensory neurons. Pharmacological Reviews, 43, 143–201. Holzer, P. (1992). Peptidergic sensory neurons in the control of vascular functions: mechanisms and significance in the cutaneous and splanchnic vascular beds. Reviews of Physiology, Biochemistry and Pharmacology, 121, 49–146. Holzer, P. (1994). Calcitonin gene-related peptide. In Gut Peptides: Biochemistry and Physiology, edited by J.H. Walsh and G.J. Dockray, pp. 493–523. New York: Raven Press. Holzer, P. and Barthó, L. (1996). Sensory neurons in the intestine. In Neurogenic Inflammation, edited by P. Geppeti and P. Holzer, pp. 153–167. Boca Raton: CRC Press. Holzer, P. and Guth, P.H. (1991). Neuropeptide control of rat gastric mucosal blood flow. Increase by calcitonin gene-related peptide and vasoactive intestinal polypeptide, but not substance P and neurokinin A. Circulation Research, 68, 100–105. Holzer, P. and Jocic, M. (1994). Cutaneous vasodilatation induced by nitric oxide-evoked stimulation of afferent nerves in the rat. British Journal of Pharmacology, 112, 1181–1187. Holzer, P. and Lippe, I.T. (1988). Stimulation of afferent nerve endings by intragastric capsaicin protects against ethanol-induced damage of gastric mucosa. Neuroscience, 27, 981–987. Holzer, P. and Lippe, I.T. (1992). Gastric mucosal hyperemia due to acid back-diffusion depends on splanchnic nerve activity. American Journal of Physiology, 262, G505–G509. Holzer, P. and Sametz, W. (1986). Gastric mucosal protection against ulcerogenic factors in the rat mediated by capsaicin-sensitive afferent neurons. Gastroenterology, 91, 975–981. Holzer, P., Gamse, R. and Lembeck, F. (1980). Distribution of substance P in the rat gastrointestinal tract – lack of effect of capsaicin pretreatment. European Journal of Pharmacology, 61, 303–307. Holzer, P., Saria, A., Skofitsch, G. and Lembeck, F. (1981). Increase in tissue concentrations of histamine and 5-hydroxytryptamine following capsaicin treatment of newborn rats. Life Sciences, 29, 1099–1105.

160

INNERVATION OF THE GASTROINTESTINAL TRACT

Holzer, P., Lippe, I.T. and Holzer-Petsche, U. (1986). Inhibition of gastrointestinal transit due to surgical trauma or peritoneal irritation is reduced in capsaicin-treated rats. Gastroenterology, 91, 360–363. Holzer, P., Pabst, M.A. and Lippe, I.T. (1989). Intragastric capsaicin protects against aspirin-induced lesion formation and bleeding in the rat gastric mucosa. Gastroenterology, 96, 1425–1433. Holzer, P., Pabst, M.A., Lippe, I.T., Peskar, B.M., Peskar, B.A., Livingston, E.H. et al. (1990a). Afferent nervemediated protection against deep mucosal damage in the rat stomach. Gastroenterology, 98, 838–848. Holzer, P., Peskar, B.M., Peskar, B.A. and Amann, R. (1990b). Release of calcitonin gene-related peptide induced by capsaicin in the vascularly perfused rat stomach. Neuroscience Letters, 108, 195–200. Holzer, P., Livingston, E.H. and Guth, P.H. (1991). Sensory neurons signal for an increase in rat gastric mucosal blood flow in the face of pending acid injury. Gastroenterology, 101, 416–423. Holzer, P., Livingston, E.H., Saria, A. and Guth, P.H. (1991). Sensory neurons mediate protective vasodilatation in rat gastric mucosa. American Journal of Physiology, 260, G363–G370. Holzer, P., Lippe, I.T. and Amann, R. (1992). Participation of capsaicin-sensitive afferent neurons in gastric motor inhibition caused by laparotomy and intraperitoneal acid. Neuroscience, 48, 715–722. Holzer, P., Lippe, I.T., Jocic, M., Wachter, C., Erb, R. and Heinemann, Á. (1993). Nitric oxide-dependent and -independent hyperaemia due to calcitonin gene-related peptide in the rat stomach. British Journal of Pharmacology, 110, 404–410. Holzer, P., Livingston, E.H. and Guth, P.H. (1994). Neural, metabolic, physical, and endothelial factors in the regulation of the gastric circulation. In Physiology of the Gastrointestinal Tract, Third Edition, edited by L.R. Johnson, pp. 1311–1329. New York: Raven Press. Holzer, P., Wachter, C., Jocic, M. and Heinemann, Á. (1994). Vascular bed-dependent roles of the peptide CGRP and nitric oxide in acid-evoked hyperaemia of the rat stomach. Journal of Physiology (London), 480, 575–585. Holzer, P., Jocic, M. and Peskar, B.A. (1995). Mediation by prostaglandins of the nitric oxide-induced neurogenic vasodilatation in rat skin. British Journal of Pharmacology, 116, 2365–2370. Holzer, P., Lippe, I.T., Jocic, M., Wachter, C., Amann, R., Pabst, M.A. et al. (1995). Role of sensory neurons in the control of gastric mucosal blood flow and protection. In Neuroendocrinology of Gastrointestinal Ulceration, edited by S. Szabo and Y. Taché, pp. 1–14. New York: Plenum Press. Holzer-Petsche, U. (1991). Modulation of gastric contractions in response to tachykinins and bethanechol by extrinsic nerves. British Journal of Pharmacology, 103, 1958–1962. Holzer-Petsche, U. (1992). Blood pressure and gastric motor responses to bradykinin and hydrochloric acid injected into somatic or visceral tissues. Naunyn-Schmiedebergs Archives of Pharmacology, 346, 219–225. Holzer-Petsche, U. (1995). Tachykinin receptors in gastrointestinal motility. Regulatory Peptides, 57, 19–42. Holzer-Petsche, U., Seitz, H. and Lembeck, F. (1989). Effect of capsaicin on gastric corpus smooth muscle of the rat in vitro. European Journal of Pharmacology, 162, 29–36. Horowitz, M., Wishart, J., Maddox, A. and Russo, A. (1992). The effect of chilli on gastrointestinal transit. Journal of Gastroenterology and Hepatology, 7, 52–56. Hottenstein, O.D., Pawlik, W.W., Remak, G. and Jacobson, E.D. (1991). Capsaicin-sensitive nerves modulate resting blood flow and vascular tone in rat gut. Naunyn-Schmiedebergs-Archives of Pharmacology, 343, 179–184. Hughes, S.R. and Brain, S.D. (1994). Nitric oxide-dependent release of vasodilator quantities of calcitonin generelated peptide from capsaicin-sensitive nerves in rabbit skin. British Journal of Pharmacology, 111, 425–430. Inaba, N., Shibata, M., Onodera, S., Tanaka, M., Suzuki, T., Kase, N. et al. (1996). Capsaicin-induced calcitonin gene-related peptide release from isolated rat stomach measured with a new chemiluminiscent enzyme immunoassay. Japanese Journal of Pharmacology, 72, 223–229. Inada, I. and Satoh, H. (1996). Capsaicin-sensitive sensory neurons are involved in bicarbonate secretion induced by lansoprazole, a proton pump inhibitor, in rats. Digestive Diseases and Sciences, 41, 785–790. Inui, T., Chiba, T., Okimura, Y., Morishita, T., Nakamura, A., Yamaguchi, A. et al. (1989). Presence and release of calcitonin gene-related peptide in rat stomach. Life Sciences, 45, 1199–1206. Inui, T., Kinoshita, Y., Yamaguchi, A., Yamatani, T. and Chiba, T. (1991). Linkage between capsaicin-stimulated calcitonin gene-related peptide and somatostatin release in rat stomach. American Journal of Physiology, 261, G770–G774. Ito, S., Ohta, T., Honda, H., Nakazato, Y. and Ohga, A. (1993). Gastric vasodilator and motor responses to splanchnic stimulation and tachykinins in the dog. General Pharmacology, 24, 291–298. Jacobson, E.D., Berguer, R., Pawlik, W.W. and Hottenstein, O.D. (1994). Mesenteric purinergic and peptidergic vasodilators. In Physiology of the Gastrointestinal Tract, Third Edition, edited by L.R. Johnson, pp. 1647–1667. New York: Raven Press. Jakab, G., Webster, H.F., Salamon, I. and Mezey, E. (1993). Neural and non-neural origin of calcitonin generelated peptide (CGRP) in the gastric mucosa. Neuropeptides, 24, 117–122. Jin, Y.H., Sekizawa, K., Fukushima, T., Morikawa, M., Nakazawa, H. and Sasaki, H. (1994). Capsaicin desensitization inhibits swallowing reflex in guinea pigs. American Journal of Respiratory and Critical Care Medicine, 149, 261–263.

CONTROL OF GASTRIC FUNCTIONS BY EXTRINSIC SENSORY NEURONS

161

Kaneko, H., Mitsuma, T., Uchida, K., Furusawa, A. and Morise, K. (1993). Immunoreactive somatostatin, substance P, and calcitonin gene-related peptide concentrations of the human gastric mucosa in patients with nonulcer dyspepsia and peptic ulcer disease. American Journal of Gastroenterology, 88, 898–904. Kang, J.Y., Alexander, B., Barker, F., Man, W.K. and Williamson, R.C.N. (1992). The effect of chilli ingestion on gastrointestinal mucosal proliferation and azoxymethane-induced cancer in the rat. Journal of Gastroenterology and Hepatology, 7, 194–198. Kang, J.Y., Alexander, B., Math, M.V. and Williamson, R.C.N. (1993). The effect of chilli and its pungent ingredient capsaicin on gastrointestinal transit in the rat. Journal of Gastroenterology and Hepatology, 8, 513–516. Kang, J.Y., Teng, C.H., Wee, A. and Chen, F.C. (1995a). Effect of capsaicin and chilli on ethanol induced gastric mucosal injury in the rat. Gut, 36, 664–669. Kang, J.Y., Yeoh, K.G., Chia, H.P., Lee, H.P., Chia, Y.W., Guan, R. and Yap, I. (1995b). Chili – protective factor against peptic ulcer? Digestive Diseases and Sciences, 40, 576–579. Kang, J.Y., Teng, C.H. and Chen, F.C. (1996). Effect of capsaicin and cimetidine on the healing of acetic acid induced gastric ulceration in the rat. Gut, 38, 832–836. Karmeli, F., Eliakim, R., Okon, E. and Rachmilewitz, D. (1991). Gastric mucosal damage by ethanol is mediated by substance P and prevented by ketotifen, a mast cell stabilizer. Gastroenterology, 100, 1206–1216. Karmeli, F., Eliakim, R., Okon, E. and Rachmilewitz, D. (1993). Role of vasoactive intestinal peptide (VIP) in pathogenesis of ethanol-induced gastric mucosal damage in rats. Digestive Diseases and Sciences, 38, 1210–1219. Karmeli, F., Eliakim, R., Okon, E. and Rachmilewitz, D. (1995). Ketotifen and nitroxides decrease capsaicinaugmented ethanol-induced gastric damage in rats. Digestive Diseases and Sciences, 40, 1140–1146. Kashiba, H., Senba, E., Ueda, Y. and Tohyama, M. (1990). Relative sparing of calcitonin gene-related peptide-containing primary sensory neurons following neonatal capsaicin treatment in the rat. Peptides, 11, 491–496. Kashiba, H., Senba, E., Ueda, Y. and Tohyama, M. (1991). Cell size and cell type analysis of calcitonin generelated peptide-containing cutaneous and splanchnic sensory neurons in the rat. Peptides, 12, 101–106. Kato, K., Yang, H. and Taché, Y. (1994). Role of peripheral capsaicin-sensitive afferent neurons and CGRP in central vagally mediated gastroprotective effect of TRH. American Journal of Physiology, 266, R1610–R1614. Kato, K., Martinez, V., St. Pierre, S. and Taché, Y. (1995a). CGRP antagonists enhance gastric acid secretion in 2-hpylorus-ligated rats. Peptides, 16, 1257–1262. Kato, K., Yang, H. and Taché, Y. (1995b). Low doses of TRH analogue act in the dorsal motor nucleus to induce gastric protection in rats. American Journal of Physiology, 269, R1301–R1307. Kato, K., Yang, H. and Taché, Y. (1996). Role of prostaglandins and calcitonin gene-related peptide in central vagal cholinergic-dependent protection against gastric injury in urethane-anesthetized rats. Digestion, 57, 322–327. Katori, M., Ohno, T. and Nishiyama, K. (1993). Interaction of substance P and leukotriene C4 in ethanol-induced mucosal injury of rat stomach. Regulatory Peptides, 46, 241–243. Katsoulis, S. and Conlon, J.M. (1989). Calcitonin gene-related peptides relax guinea pig and rat gastric smooth muscle. European Journal of Pharmacology, 162, 129–134. Kinoshita, M., Kume, E. and Tamaki, H. (1995). Roles of prostaglandins, nitric oxide and the capsaicin-sensitive sensory nerves in gastroprotection produced by ecabet sodium. Journal of Pharmacology and Experimental Therapeutics, 275, 494–501. Király, A., Sütö, G., Livingston, E.H., Guth, P.H., St. Pierre, S. and Taché, Y. (1994a). Central vagal activation by TRH induces gastric hyperemia: role of CGRP in capsaicin-sensitive afferents in rats. American Journal of Physiology, 267, G1041–G1049. Király, A., Sütö, G., Livingston, E.H., Guth, P.H. and Taché, Y. (1994b). Peripheral mechanisms involved in gastric hyperemic response to central vagal stimulation in rats. Gastroenterology, 106, A819. Király, A., Sütö, G., Guth, P.H. and Taché, Y. (1994c). Vagally mediated gastric mucosal hyperemic response to a cytoprotective dose of intracisternal TRH analog: involvement of capsaicin sensitive afferents containing CGRP and mucosal mast cells in the rats. Gastroenterology, 106, A107. Kirchgessner, A.L., Dodd, J. and Gershon, M.D. (1988). Markers shared between dorsal root and enteric ganglia. Journal of Comparative Neurology, 276, 607–621. Kitsukawa, Y., Turner, R.J., Pradhan, T.K. and Jensen, R.T. (1996). Gastric chief cells possess NK1 receptors which mediate pepsinogen secretion and are regulated by agents that increase cAMP and phospholipase C. Biochimica et Biophysica Acta, 1312, 105–116. Ko, J.K.S. and Cho, C.H. (1995). The mechanistic pathway of gastric adaptive cytoprotection: a study on different components of the autonomic nervous system. Journal of Autonomic Pharmacology, 15, 205–214. Kondo, H. and Yamamoto, M. (1988). The ontogeny and fine structure of calcitonin gene-related peptide (CGRP)immunoreactive nerve fibers in the celiac ganglion of rats. Archives of Histology and Cytology, 51, 91–98.

162

INNERVATION OF THE GASTROINTESTINAL TRACT

Konturek, S.J., Jaworek, J., Tasler, J., Cieszkowski, M. and Pawlik, W. (1981). Effect of substance P and its C-terminal hexapeptide on gastric and pancreatic secretion in the dog. American Journal of Physiology, 241, G74–G81. Konturek, S.J., Brzozowski, T., Bielanski, W. and Schally, A.V. (1995a). Role of endogenous gastrin in gastroprotection. European Journal of Pharmacology, 278, 203–212. Konturek, S.J., Brzozowski, T., Pytko-Polonczyk, J. and Drozdowicz, D. (1995b). Comparison of cholecystokinin, pentagastrin, and duodenal oleate in gastroprotection in rats. Scandinavian Journal of Gastroenterology, 30, 620–630. Konturek, S.J., Brzozowski, T., Pytko-Polonczyk, J. and Drozdowicz, D. (1995c). Exogenous and endogenous cholecystokinin protects gastric mucosa against the damage caused by ethanol in rats. European Journal of Pharmacology, 273, 57–62. Konturek, S.J., Brzozowski, T. and Sliwowski, Z. (1996). Importance of sensory nerves and nitric oxide (NO) in gastric cytoprotection induced by epidermal growth factor (EGF). Gut, 39(Suppl. 3), A152. Kruger, L. (1988). Morphological features of thin sensory afferent fibers: a new interpretation of “nociceptor” function. Progress in Brain Research, 74, 253–257. Kwok, Y.N. and McIntosh, C.H.S. (1990). Release of substance P-like immunoreactivity from the vascularly perfused rat stomach. European Journal of Pharmacology, 180, 201–207. Kwok, Y.N., McIntosh, C.H., Pederson, R.A. and Brown, J.C. (1985). Effect of substance P on somatostatin release from the isolated perfused rat stomach. Gastroenterology, 88, 90–95. Kwok, Y.N., McIntosh, C.H., Sy, H. and Brown, J.C. (1988). Inhibitory actions of tachykinins and neurokinins on release of somatostatin-like immunoreactivity from the isolated perfused rat stomach. Journal of Pharmacology and Experimental Therapeutics, 246, 726–731. Lambrecht, N., Burchert, M., Respondek, M., Müller, K.M. and Peskar, B.M. (1993a). Role of calcitonin generelated peptide and nitric oxide in the gastroprotective effect of capsaicin in the rat. Gastroenterology, 104, 1371–1380. Lambrecht, N., Stroff, T., Korolkiewicz, R. and Peskar, B.M. (1993b). Role of nitric oxide (NO), prostaglandins and cholecystokinin (CCK)-8-A and -B receptors in the protective effect of pentagastrin in the rat stomach. Gastroenterology, 104, A128. Lambrecht, N., Trautmann, M., Korolkiewicz, R., Liszkay, M. and Peskar, B.M. (1993c). Role of eicosanoids, nitric oxide, and afferent neurons in antacid induced protection in the rat stomach. Gut, 34, 329–337. Lambrecht, N., Stroff, T. and Peskar, B.M. (1993). Endogenous nitric oxide (NO) as essential mediator of the protective effect of cholecystokinin (CCK)-8 in the rat stomach. Gastroenterology, 104, A128. Langley, J.N. (1903). The autonomic nervous system. Brain, 26, 1–26. Lawson, D.C., Mantyh, C.R. and Pappas, T.N. (1994). Effect of CGRP antagonist, α-CGRP8–37, on acid secretion in the dog. Digestive Diseases and Sciences, 39, 1405–1408. Lee, S.S. and Sharkey, K.A. (1993). Capsaicin treatment blocks development of hyperkinetic circulation in portal hypertensive and cirrhotic rats. American Journal of Physiology, 264, G868–G873. Lee, Y., Hayashi, N., Hillyard, C.J., Girgis, S.I., MacIntyre, I., Emson, P.C. et al. (1987a). Calcitonin generelated peptide-like immunoreactive sensory fibers form synaptic contact with sympathetic neurons in the rat celiac ganglion. Brain Research, 407, 149–151. Lee, Y., Shiotani, Y., Hayashi, N., Kamada, T., Hillyard, C.J., Girgis, S.I. et al. (1987b). Distribution and origin of calcitonin gene-related peptide in the rat stomach and duodenum: an immunohistochemical study. Journal of Neural Transmission, 68, 1–14. Lefebvre, R.A., De Beurme, F.A. and Sas, S. (1991). Relaxant effect of capsaicin in the rat gastric fundus. European Journal of Pharmacology, 195, 131–137. Lenz, J. (1988). Calcitonin and CGRP inhibit gastrointestinal transit via distinct neuronal pathways. American Journal of Physiology, 254, G920–G924. Leung, F.W. (1992a). Modulation of autoregulatory escape by capsaicin-sensitive afferent nerves in the rat stomach. American Journal of Physiology, 262, H562–H567. Leung, F.W. (1992b). Role of capsaicin-sensitive afferent nerves in mucosal injury and injury-induced hyperemia in rat colon. American Journal of Physiology, 262, G332–G337. Leung, F.W. (1993a). Inhibition of spinal afferent nerve-mediated gastric hyperemia by nicotine: role of ganglionic blockade. American Journal of Physiology, 264, H1087–H1092. Leung, F.W. (1993b). Primary sensory neurons mediate in part the protective mesenteric hyperemia after intraduodenal acidification in rats. Gastroenterology, 105, 1737–1745. Li, D.S., Raybould, H.E., Quintero, E. and Guth, P.H. (1991). Role of calcitonin gene-related peptide in gastric hyperemic response to intragastric capsaicin. American Journal of Physiology, 261, G657–G661. Li, D.S., Raybould, H.E., Quintero, E. and Guth, P.H. (1992). Calcitonin gene-related peptide mediates the gastric hyperemic response to acid back-diffusion. Gastroenterology, 102, 1124–1128.

CONTROL OF GASTRIC FUNCTIONS BY EXTRINSIC SENSORY NEURONS

163

Lidberg, P., Dahlström, A., Lundberg, J.M. and Ahlman, H. (1983). Different modes of action of substance P in the motor control of the feline stomach and pylorus. Regulatory Peptides, 7, 41–52. Lindh, B., Hökfelt, T. and Elfvin, L.G. (1988). Distribution and origin of peptide-containing nerve fibers in the celiac superior mesenteric ganglion of the guinea-pig. Neuroscience, 26, 1037–1071. Lippe, I.T. and Holzer, P. (1992). Participation of endothelium-derived nitric oxide but not prostacyclin in the gastric mucosal hyperaemia due to acid back-diffusion. British Journal of Pharmacology, 105, 708–714. Lippe, I.T., Lorbach, M. and Holzer, P. (1989). Close arterial infusion of calcitonin gene-related peptide into the rat stomach inhibits aspirin- and ethanol-induced hemorrhagic damage. Regulatory Peptides, 26, 35–46. Lippe, I.T., Pabst, M.A. and Holzer, P. (1989). Intragastric capsaicin enhances rat gastric acid elimination and mucosal blood flow by afferent nerve stimulation. British Journal of Pharmacology, 96, 91–100. Lippe, I.T., Sametz, W., Sabin, K. and Holzer, P. (1993). Inhibitory role of capsaicin-sensitive afferent neurons and nitric oxide in hemostasis. American Journal of Physiology, 265, H1864–H1868. Lippe, I.T., Wachter, C.H. and Holzer, P. (1997). Neurokinin A-induced vasoconstriction and muscular contraction in the rat isolated stomach: mediation by distinct and unusual NK2 receptors. Journal of Pharmacology and Experimental Therapeutics, 281, 1294–1302. Livingston, E.H. and Holzer, P. (1993). Gastric hyperemia accompanying acid secretion is not mediated by sensory nerves. Digestive Diseases and Sciences, 38, 1190–1194. Livingston, E.H. and Passaro, E.P. (1990). Postoperative ileus. Digestive Diseases and Sciences, 35, 121–132. Lloyd, K.C.K., Hölzer, H.H., Zittel, T.T. and Raybould, H.E. (1993). Duodenal lipid inhibits gastric acid secretion by vagal, capsaicin-sensitive afferent pathways in rats. American Journal of Physiology, 264, G659–G663. Lopez-Belmonte, J. and Whittle, B.J.R. (1993). The paradoxical vascular interactions between endothelin-1 and calcitonin gene-related peptide in the rat gastric mucosal microcirculation. British Journal of Pharmacology, 110, 496–500. Lördal, M., Hällgren, A., Nylander, O. and Hellström, P.M. (1996). Tachykinins increase vascular permeability in the gastrointestinal tract of the rat. Acta Physiologica Scandinavica, 156, 489–494. Lu, Y.X. and Owyang, C. (1995). Secretin at physiological doses inhibits gastric motility via a vagal afferent pathway. American Journal of Physiology, 268, G1012–G1016. Lundberg, J.M., Brodin, E., Hua, X.Y. and Saria, A. (1984). Vascular permeability changes and smooth muscle contraction in relation to capsaicin-sensitive substance P afferents in the guinea-pig. Acta Physiologica Scandinavica, 120, 217–227. Maggi, C.A. (1991). The pharmacology of the efferent function of sensory nerves. Journal of Autonomic Pharmacology, 11, 173–208. Maggi, C.A., Evangelista, S., Abelli, L., Somma, V. and Meli, A. (1987a). Capsaicin-sensitive mechanisms and experimentally induced duodenal ulcers in rats. Journal of Pharmacy and Pharmacology, 39, 559–561. Maggi, C.A., Evangelista, S., Giuliani, S. and Meli, A. (1987b). Anti-ulcer activity of calcitonin gene-related peptide in rats. General Pharmacology, 18, 33–34. Maggi, C.A., Santicioli, P., Del Bianco, E., Geppetti, P., Barbanti, G., Turini, D. et al. (1989). Release of VIPbut not CGRP-like immunoreactivity by capsaicin from the human isolated small intestine. Neuroscience Letters, 98, 317–320. Malcangio, M., Bowery, N.G., Flower, R.J. and Perretti, M. (1996). Effect of interleukin-β on the release of substance P from rat isolated spinal cord. European Journal of Pharmacology, 299, 113–118. Manela, F.D., Young, R.L., Ren, R. and Harty, R.F. (1992). Stimulation of rat antral CGRP release by intraluminal peptone. Annals of the New York Academy of Sciences, 657, 528–530. Manela, F.D., Ren, J., Gao, J., McGuigan, J.E. and Harty, R.F. (1995). Calcitonin gene-related peptide modulates acid-mediated regulation of somatostatin and gastrin release from rat antrum. Gastroenterology, 109, 701–706. Mantyh, P.W., Mantyh, C.R., Gates, T., Vigna, S.R. and Maggio, J.E. (1988). Receptor binding sites for substance P and substance K in the canine gastrointestinal tract and their possible role in inflammatory bowel disease. Neuroscience, 25, 817–837. Marshall, J.S., Kawabori, S., Nielsen, L. and Bienenstock, J. (1994). Morphological and functional characteristics of peritoneal mast cells from young rats. Cell and Tissue Research, 276, 565–570. Martensson, H.G., Akande, B., Yeo, C. and Jaffe, B.M. (1984). The role of substance P in the control of gastric acid secretion. Surgery, 95, 567–571. Mathison, R. and Davison, J.S. (1995). Regulation of jejunal arterioles by capsaicin-sensitive nerves in Nippostrongylus brasiliensis-sensitized rats. Journal of Pharmacology and Experimental Therapeutics, 273, 337–343. Maton, P.N., Sutliff, V.E., Zhou, Z.C., Collins, S.M., Gardner, J.D. and Jensen, R.T. (1988). Characterization of receptors for calcitonin gene-related peptide on gastric smooth muscle cells. American Journal of Physiology, 254, G789–G794.

164

INNERVATION OF THE GASTROINTESTINAL TRACT

Matsumoto, J., Takeuchi, K. and Okabe, S. (1991). Characterization of gastric mucosal blood flow response induced by intragastric capsaicin in rats. Japanese Journal of Pharmacology, 57, 205–213. Matsumoto, J., Takeuchi, K., Ueshima, K. and Okabe, S. (1992a). Role of capsaicin-sensitive afferent neurons in mucosal blood flow response of rat stomach induced by mild irritants. Digestive Diseases and Sciences, 37, 1336–1344. Matsumoto, J., Ueshima, K., Ohuchi, T., Takeuchi, K. and Okabe, S. (1992b). Induction of gastric lesions by 2-deoxy-D-glucose in rats following chemical ablation of capsaicin-sensitive sensory neurons. Japanese Journal of Pharmacology, 60, 43–49. Matthews, M.R. and Cuello, A.C. (1984). The origin and possible significance of substance P immunoreactive networks in the prevertebral ganglia and related structures in the guinea-pig. Philosophical Transactions of the Royal Society (London), Series B, 306, 247–276. Mayer, E.A. and Gebhart, G.F. (1994). Basic and clinical aspects of visceral hyperalgesia. Gastroenterology, 107, 271–293. McCafferty, D.M., Sharkey, K.A. and Wallace, J.L. (1994). Beneficial effects of local or systemic lidocaine in experimental colitis. American Journal of Physiology, 266, G560–G567. McCarthy, P.W. and Lawson, S.N. (1989). Cell type and conduction velocity of rat primary sensory neurons with substance P-like immunoreactivity. Neuroscience, 28, 745–753. McCarthy, P.W. and Lawson, S.N. (1990). Cell type and conduction velocity of rat primary sensory neurons with calcitonin gene-related peptide-like immunoreactivity. Neuroscience, 34, 623–632. McGillis, J.P., Mitsuhashi, M. and Payan, D.G. (1990). Immunomodulation by tachykinin neuropeptides. Annals of the New York Academy of Sciences, 594, 85–94. McGregor, G.P. and Conlon, J.M. (1991). 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. Regulatory Peptides, 32, 109–119. McIntosh, C.H., Bakich, V., Kwok, Y.N., Wong, J. and Brown, J.C. (1987). The effects of substance P, histamine and histamine antagonists on somatostatin and gastrin release from the isolated perfused rat stomach. Regulatory Peptides, 19, 253–263. McKie, L.D., Dunkin, B.J., Pennanen, M.F., Dunlap, K.W., Harmon, J.W. and Bass, B.L. (1994). Esophageal mucosal blood flow: a central role for calcitonin gene-related peptide. Surgery, 116, 409–418. Mercer, D.W., Ritchie, W.P. and Dempsey, D.T. (1994). Sensory neuron-mediated gastric mucosal protection is blocked by cyclooxygenase inhibition. Surgery, 115, 156–163. Mercer, D.W., Klemm, K., Cross, J.M., Smith, G.S., Cashman, M. and Miller, T.A. (1996). Cholecystokinininduced protection against gastric injury is independent of endogenous somatostatin. American Journal of Physiology, 271, G692–G700. Merchant, N.B., Dempsey, D.T., Grabowski, M.W., Rizzo, M. and Ritchie, W.P. (1994). Capsaicin-induced gastric mucosal hyperemia and protection: the role of calcitonin gene-related peptide. Surgery, 116, 419–425. Merchant, N.B., Goodman, J., Dempsey, D.T., Milner, R.E. and Ritchie, W.P. (1995). The role of calcitonin gene-related peptide and nitric oxide in gastric mucosal hyperemia and protection. Journal of Surgical Research, 58, 344–350. Miller, T.A., Smith, G.S., Stanislawska, M., Phan, T.M. and Henagan, J.M. (1989). Role of vagal innervation in adaptive cytoprotection. Digestive Diseases and Sciences, 34, 1318. Minagawa, H., Shiosaka, S., Inoue, H., Hayashi, N., Kasahara, A., Kamada, T. et al. (1984). Origins and threedimensional distribution of substance P-containing structures in the rat stomach using whole-mount tissue. Gastroenterology, 86, 51–59. Miyake, H., Inaba, N., Kato, S. and Takeuchi, K. (1996). Increased susceptibility of rat gastric mucosa to ulcerogenic stimulation with aging: role of capsaicin-sensitive sensory neurons. Digestive Diseases and Sciences, 41, 339–345. Modlin, I.M., Lamers, C.B. and Walsh, J.H. (1981). Stimulation of canine pancreatic polypeptide, gastrin, and gastric acid secretion by ranatensin, litorin, bombesin nonapeptide and substance P. Regulatory Peptides, 1, 279–288. Molander, C., Ygge, J. and Dalsgaard, C.J. (1987). Substance P-, somatostatin- and calcitonin gene-related peptide-like immunoreactivity and fluoride resistant acid phosphatase-activity in relation to retrogradely labeled cutaneous, muscular and visceral primary sensory neurons in the rat. Neuroscience Letters, 74, 37–42. Morris, G.P., Basu, B.K., Holitzner, C.A. and Williamson, T.E. (1993). Mucosal secretion and cytoprotection of the rat gastric mucosa. Journal of Clinical Gastroenterology, 17(Suppl. 1), S5–S10. Mueller, C.R., Ure, T., O’Dorisio, T.M., Barrie, R.J. and Woltering, E.A. (1991). Acute gastric pH changes alter intraluminal but not plasma peptide levels. Surgery, 110, 1116–1124. Mulder, H., Lindh, A.C., Ekblad, E., Westermark, P. and Sundler, F. (1994a). Islet amyloid polypeptide is expressed in endocrine cells of the gastric mucosa in the rat and mouse. Gastroenterology, 107, 712–719.

CONTROL OF GASTRIC FUNCTIONS BY EXTRINSIC SENSORY NEURONS

165

Mulder, H., Uddman, R., Moller, K., Zhang, Y.Z., Ekblad, E., Alumets, J. et al. (1994b). Pituitary adenylate cyclase activating polypeptide expression in sensory neurons. Neuroscience, 63, 307–312. Mulder, H., Leckström, A., Uddman, R., Ekblad, E., Westermark, P. and Sundler, F. (1995). Islet amyloid polypeptide (amylin) is expressed in sensory neurons. Journal of Neuroscience, 15, 7625–7632. Mulderry, P.K., Ghatei, M.A., Spokes, R.A., Jones, P.M., Pierson, A.M., Hamid, Q.A., Kanse, S., Amara, S.G., Burrin, J.M., Legon, S., Polak, J.M. and Bloom, S.R. (1988). Differential expression of α-CGRP and β-CGRP by primary sensory neurons and enteric autonomic neurons of the rat. Neuroscience, 25, 195–205. Mungan, Z., Hammer, R.A., Akarca, U.S., Komaki, G., Ertan, A. and Arimura, A. (1995). Effect of PACAP on gastric acid secretion in rats. Peptides, 16, 1051–1056. Murakami, I., Satoh, S., Asano, S. and Maeda, R. (1996). Role of capsaicin-sensitive sensory neurons and nitric oxide in the protective effect of lansoprazole, a proton pump inhibitor, on the gastric mucosa in rats. Japanese Journal of Pharmacology, 72, 137–147. Nakamura, T., Naruse, S., Ozaki, T. and Kumada, K. (1996). Calcitonin gene-related peptide is a potent intestinal, but not gastric, vasodilator in conscious dogs. Regulatory Peptides, 65, 211–217. Nakazato, Y., Sekine, H., Isogaya, M., Ito, S. and Ohga, A. (1987). Atropine- and hexamethonium-resistant motor response to greater splanchnic nerve stimulation in the dog stomach. Journal of the Autonomic Nervous System, 20, 35–42. Narita, M., Takahashi, S., Takeuchi, K. and Okabe, S. (1995). Desensitization of capsaicin-sensitive sensory neurons in rat stomachs on chronic treatment with sodium taurocholate. Japanese Journal of Pharmacology, 67, 321–328. Nicolau, M., Sirois, M.G., Bui, M., Plante, G.E., Sirois, P. and Regoli, D. (1993). Plasma extravasation induced by neurokinins in conscious rats: receptor characterization with agonists and antagonists. Canadian Journal of Physiology and Pharmacology, 71, 217–221. Nilsson, J., von Euler, A.M. and Dalsgaard, C.J. (1985). Stimulation of connective tissue cell growth by substance P and substance K. Nature, 315, 61–63. Nishizaki, Y., Guth, P.H., Sternini, C. and Kaunitz, J.D. (1996). Impairment of the gastric hyperemic response to luminal acid in cirrhotic rats. American Journal of Physiology, 270, G71–G78. Nuki, C., Kawasaki, H., Kitamura, K., Takenaga, M., Kangawa, K., Eto, T. and Wada, A. (1993). Vasodilator effect of adrenomedullin and calcitonin gene-related peptide receptors in rat mesenteric vascular beds. Biochemical and Biophysical Research Communications, 196, 245–251. Oates, P.J. (1990). Gastric blood flow and mucosal defense. In Gastric Cytoprotection, edited by D. Hollander and A.S. Tarnawski , pp. 125–165. New York: Plenum. Ohno, T., Hayashi, H., Hatanaka, K. and Katori, M. (1994). A preventive mechanism of capsaicin against ethanol-induced gastric mucosal injury in rats. Japanese Journal of Pharmacology, 64(Suppl. 1), 284P. Otsuka, M. and Yoshioka, K. (1993). Neurotransmitter functions of mammalian tachykinins. Physiological Reviews, 73, 229–308. Pabst, M.A., Schöninkle, E. and Holzer, P. (1993). Ablation of capsaicin-sensitive afferent nerves impairs defence but not rapid repair of rat gastric mucosa. Gut, 34, 897–903. Pabst, M.A., Wachter, C. and Holzer, P. (1996). Morphologic basis of the functional gastric acid barrier. Laboratory Investigation, 74, 78–85. Pan, H.L. and Longhurst, J.C. (1996). Ischaemia-sensitive sympathetic afferents innervating the gastrointestinal tract function as nociceptors in cats. Journal of Physiology (London), 492, 841–850. Payan, D.G. (1985). Receptor-mediated mitogenic effects of substance P on cultured smooth muscle cells. Biochemical and Biophysical Research Communications, 130, 104–109. Peskar, B.M., Respondek, M., Müller, K.M. and Peskar, B.A. (1991). A role for nitric oxide in capsaicin-induced gastroprotection. European Journal of Pharmacology, 198, 113–114. Peskar, B.M., Nowak, P. and Lambrecht, N. (1992). Effect of prostaglandins and capsaicin on gastric vascular flow and mucosal injury in endothelin-1-treated rats. Agents and Actions, 37(Suppl.), 85–91. Peskar, B.M., Wong, H.C., Walsh, J.H. and Holzer, P. (1993). A monoclonal antibody to calcitonin generelated peptide abolishes capsaicin-induced gastroprotection. European Journal of Pharmacology, 250, 201–203. Peskar, B.M., Lambrecht, N., Stroff, T., Respondek, M. and Müller, K.M. (1995). Functional ablation of sensory neurons impairs healing of acute gastric mucosal damage in rats. Digestive Diseases and Sciences, 40, 2460–2464. Pethö, G., Jocic, M. and Holzer, P. (1994). Role of bradykinin in the hyperaemia following acid challenge of the rat gastric mucosa. British Journal of Pharmacology, 113, 1036–1042. Piqué, J.M., Esplugues, J.V. and Whittle, B.J.R. (1989). The vasodilator role of endogenous nitric oxide in the rat gastric microcirculation. European Journal of Pharmacology, 174, 293–296. Piqué, J.M., Esplugues, J.V. and Whittle, B.J.R. (1990). Influence of morphine or capsaicin pretreatment on rat gastric microcirculatory response to PAF. American Journal of Physiology, 258, G352–G357.

166

INNERVATION OF THE GASTROINTESTINAL TRACT

Plate, S., Stroff, T., Patacchini, R., Maggi, C.A. and Peskar, B.M. (1996). Tachykinin NK2 receptor activation contributes to the protective but not acid secretion-inhibitory effect of calcitonin gene-related peptide in the rat stomach. Gastroenterology, 110, A1107. Plourde, V., Wong, H.C., Walsh, J.H., Raybould, H.E. and Taché, Y. (1993). CGRP antagonists and capsaicin on celiac ganglia partly prevent postoperative gastric ileus. Peptides, 14, 1225–1229. Podolsky, R.S., Grabowski, M., Milner, R., Ritchie, W.P. and Dempsey, D.T. (1994). Capsaicin-induced gastric hyperemia and protection are NO-dependent. Journal of Surgical Research, 57, 438–442. Pothoulakis, C., Castagliuolo, I., LaMont, J.T., Jaffer, A., Okeane, J.C., Snider, R.M. et al. (1994). CP-96,345, a substance P antagonist, inhibits rat intestinal responses to Clostridium difficile toxin A but not cholera toxin. Proceedings of the National Academy of Sciences USA, 91, 947–951. Prokopiw, I. and McDonald, T.J. (1994). Effects of tachykinins on the splanchnic circulation: a comparison of the dog and rat. Peptides, 15, 1189–1194. Quintero, E., Kaunitz, J., Nishizaki, Y., De Giorgio, R., Sternini, C. and Guth, P.H. (1992). Uremia increases gastric mucosal permeability and acid back-diffusion injury in the rat. Gastroenterology, 103, 1762–1768. Ramos, E.G., Esplugues, J. and Esplugues, J.V. (1992). Gastric acid secretory responses induced by peptone are mediated by capsaicin-sensitive sensory afferent neurons. American Journal of Physiology, 262, G835–G839. Raybould, H.E. (1991). Capsaicin-sensitive vagal afferents and CCK in inhibition of gastric motor function induced by intestinal nutrients. Peptides, 12, 1279–1283. Raybould, H.E. and Hölzer, H. (1992). Dual capsaicin-sensitive afferent pathways mediate inhibition of gastric emptying in rat induced by intestinal carbohydrate. Neuroscience Letters, 141, 236–238. Raybould, H.E. and Hölzer, H.H. (1993a). Duodenal acid-induced inhibition of gastric motility and emptying in rats. American Journal of Physiology, 265, G540–G546. Raybould, H.E. and Hölzer, H. (1993b). Secretin inhibits gastric emptying in rats via a capsaicin-sensitive vagal afferent pathway. European Journal of Pharmacology, 250, 165–167. Raybould, H.E. and Taché, Y. (1988). Cholecystokinin inhibits gastric motility and emptying via a capsaicinsensitive vagal pathway in rats. American Journal of Physiology, 255, G242–G246. Raybould, H.E. and Taché, Y. (1989). Capsaicin-sensitive vagal afferent fibers and stimulation of gastric acid secretion in anesthetized rats. European Journal of Pharmacology, 167, 237–243. Raybould, H.E., Kolve, E. and Taché, Y. (1988). Central nervous system action of calcitonin gene-related peptide to inhibit gastric emptying in the conscious rat. Peptides, 9, 735–738. Raybould, H.E., Holzer, P., Reddy, S.N., Yang, H. and Taché, Y. (1990). Capsaicin-sensitive vagal afferents contribute to gastric acid and vascular responses to intracisternal TRH analog. Peptides, 11, 789–795. Raybould, H.E., Sternini, C., Eysselein, V.E., Yoneda, M. and Holzer, P. (1992). Selective ablation of spinal afferent neurons containing CGRP attenuates gastric hyperemic response to acid. Peptides, 13, 249–254. Reinshagen, M., Patel, A., Sottili, M., Nast, C., Davis, W., Mueller, K. and Eysselein,V.E. (1994). Protective function of extrinsic sensory neurons in acute rabbit experimental colitis. Gastroenterology, 106, 1208–1214. Reinshagen, M., Patel, A., Sottili, M., French, S., Sternini, C. and Eysselein, V.E. (1996). Action of sensory neurons in an experimental rat colitis model of injury and repair. American Journal of Physiology, 270, G79–G86. Remak, G., Hottenstein, O.D. and Jacobson, E.D. (1990). Sensory nerves mediate neurogenic escape in rat gut. American Journal of Physiology, 258, H778–H786. Ren, J.Y., Young, R.L., Lassiter, D.C., Rings, M.C. and Harty, R.F. (1991). Calcitonin gene-related peptide: mechanisms of modulation of antral endocrine cells and cholinergic neurons. American Journal of Physiology, 262, G732–G739. Ren, J.Y., Young, R.L., Lassiter, D.C. and Harty, R.F. (1993). Calcitonin gene-related peptide mediates capsaicin-induced neuroendocrine responses in rat antrum. Gastroenterology, 104, 485–491. Ren, J.Y., Wang, Y., Liang, K., Gao, J., Brewer, K. and Harty, R.F. (1995). Mechanisms of proton-induced stimulation of CGRP release from rat antrum. Regulatory Peptides, 59, 103–109. Ren, J., Anderson, M., Jazzar, A., Xia, Y., Xiao, S. and Harty, R.F. (1996). GABA-induced gastroprotection is mediated by capsaicin-sensitive sensory nerves in the rat. Gastroenterology, 110, A239. Renzi, D., Santicioli, P., Maggi, C.A., Surrenti, C., Pradelles, P. and Meli, A. (1988). Capsaicin-induced release of substance P-like immunoreactivity from the guinea pig stomach in vitro and in vivo. Neuroscience Letters, 92, 254–258. Renzi, D., Evangelista, S., Mantellini, P., Santicioli, P., Maggi, C.A., Geppetti, P. and Surrenti, C. (1991). Capsaicin-induced release of neurokinin A from muscle and mucosa of gastric corpus: correlation with capsaicin-evoked release of calcitonin gene-related peptide. Neuropeptides, 19, 137–145. Rodrigo, J., Polak, J.M., Fernandez, L., Ghatei, M.A., Mulderry, P. and Bloom, S.R. (1985). Calcitonin generelated peptide immunoreactive sensory and motor nerves of the rat, cat, and monkey esophagus. Gastroenterology, 88, 444–451.

CONTROL OF GASTRIC FUNCTIONS BY EXTRINSIC SENSORY NEURONS

167

Rózsa, Z., Sharkey, K.A., Jancsó, G. and Varró, V. (1986). Evidence for a role of capsaicin-sensitive mucosal afferent nerves in the regulation of mesenteric blood flow in the dog. Gastroenterology, 90, 906–910. Sandler, A.D., Schlegel, J.F., DeSautel, M.G. and Maher, J.W. (1993). Neuroregulation of a chemosensitive afferent system in the canine distal esophagus. Journal of Surgical Research, 55, 364–371. Sandor, A., Kidd, M., Lawton, G.P., Miu, K., Tang, L.H. and Modlin, I.M. (1996). Neurohormonal modulation of rat enterochromaffin-like cell histamine secretion. Gastroenterology, 110, 1084–1092. Sandvik, A.K., Dimaline, R., Forster, E.R., Evans, D. and Dockray, G.J. (1993). Differential control of somatostatin messenger RNA in rat gastric corpus and antrum. Role of acid, food, and capsaicin-sensitive afferent neurons. Journal of Clinical Investigation, 91, 244–250. Sann, H., Dux, M., Schemann, M. and Jancsó, G. (1996). Neurogenic inflammation in the gastrointestinal tract of the rat. Neuroscience Letters, 219, 147–150. Saperas, E., Santos, J. and Malagelada, J.R. (1995). Role of vagal and splanchnic capsaicin-sensitive afferents in enterogastric inhibition of acid secretion in rats. American Journal of Physiology, 268, G286–G291. Saria, A., Lundberg, J.M., Skofitsch, G. and Lembeck, F. (1983). Vascular protein leakage in various tissues induced by substance P, capsaicin, bradykinin, serotonin, histamine and by antigen challenge. NaunynSchmiedeberg’s Archives of Pharmacology, 324, 212–218. Schepp, W., Schmidtler, J., Tatge, C., Schusdziarra, V. and Classen, M. (1990). Effect of substance P and neurokinin A on rat parietal cell function. American Journal of Physiology, 259, G646–G654. Schmidt, P., Poulsen, S.S., Hilsted, L., Rasmussen, T.N. and Holst, J.J. (1996). Tachykinins mediate vagal inhibition of gastrin secretion in pigs. Gastroenterology, 111, 925–935. Schuligoi, R., Herzeg, G., Wachter, C., Jocic, M. and Holzer, P. (1996). Differential expression of c-fos messenger RNA in the rat spinal cord after mucosal and serosal irritation of the stomach. Neuroscience, 72, 535–544. Sengupta, J.N. and Gebhart, G.F. (1994). Gastrointestinal afferent fibers and sensation. In Physiology of the Gastrointestinal Tract, Third Edition, edited by L.R. Johnson, pp. 483–519. New York: Raven Press. Seno, K., Lam, K., Leung, J.W.C. and Leung, F.W. (1996). Aging impairs afferent nerve function in rat intestine. Reduction of mesenteric hyperemia induced by intraduodenal capsaicin and acid. Digestive Diseases and Sciences, 41, 346–351. Seyed Ebrahim, J., Stroff, T. and Peskar, B.M. (1996). Nitric oxide mediates the vasodilator action but not the release of calcitonin gene-related peptide during afferent nerve stimulation in the rat stomach. Gastroenterology, 110, A1118. Shanahan, F., Denburg, J.A., Fox, J., Bienenstock, J. and Befus, D. (1985). Mast cell heterogeneity: effects of neuroenteric peptides on histamine release. Journal of Immunology, 135, 1331–1337. Sharkey, K.A. (1992). Substance P and calcitonin gene-related peptide (CGRP) in gastrointestinal inflammation. Annals of the New York Academy of Sciences, 664, 425–442. Sharkey, K.A., Williams, R.G. and Dockray, G.J. (1984). Sensory substance P innervation of the stomach and pancreas. Demonstration of capsaicin-sensitive sensory neurons in the rat by combined immunohistochemistry and retrograde tracing. Gastroenterology, 87, 914–921. Sharkey, K.A., Oland, L.D., Kirk, D.R. and Davison, J.S. (1991). Capsaicin-sensitive vagal stimulation-induced gastric acid secretion in the rat: evidence for cholinergic vagal afferents. British Journal of Pharmacology, 103, 1997–2003. Skofitsch, G., Savitt, J.M. and Jacobowitz, D.M. (1985). Suggestive evidence for a functional unit between mast cells and substance P fibers in the rat diaphragm and mesentery. Histochemistry, 82, 5–8. Smedfors, B., Theodorsson, E. and Johansson, C. (1994). HCl-stimulated duodenal HCO3-secretion in conscious rat. Interactions among VIP, nicotinic receptor mechanisms, and prostaglandins. Digestive Diseases and Sciences, 39, 2134–2142. Smith, C.H., Barker, J.N.W.N., Morris, R.W., MacDonald, D.M. and Lee, T.H. (1993). Neuropeptides induce rapid expression of endothelial cell adhesion molecules and elicit granulocyte infiltration in human skin. Journal of Immunology, 151, 3274–3282. Soldani, G., Mengozzi, G., Intorre, L., Pacini, F. and Evangelista, S. (1992). Acute intragastric application of capsaicin inhibits 2-deoxy-D-glucose- but not histamine-induced gastric acid secretion in the dog. Neuropeptides, 23, 221–225. Soper, B.D. and Tepperman, B.L. (1986). The effect of adrenergic, cholinergic and peptidergic salivary stimulants on gastric mucosal integrity in the rat. Journal of Physiology (London), 380, 329–340. Stahl, G.L., Pan, H.L. and Longhurst, J.C. (1993). Activation of ischemia- and reperfusion-sensitive abdominal visceral C fibre afferents. Role of hydrogen peroxide and hydroxyl radicals. Circulation Research, 72, 1266–1275. Stanisz, A.M., Befus, D. and Bienenstock, J. (1986). Differential effects of vasoactive intestinal peptide, substance P, and somatostatin on immunoglobulin synthesis and proliferations by lymphocytes from Peyer’s patches, mesenteric lymph nodes, and spleen. Journal of Immunology, 136, 152–156.

168

INNERVATION OF THE GASTROINTESTINAL TRACT

Stead, R.H., Bienenstock, J. and Stanisz, A.M. (1987). Neuropeptide regulation of mucosal immunity. Immunological Reviews, 100, 333–359. Stead, R.H., Tomioka, M., Quinonez, G., Simon, G.T., Felten, S.Y. and Bienenstock, J. (1987). Intestinal mucosal mast cells in normal and nematode-infected rat intestines are in intimate contact with peptidergic nerves. Proceedings of the National Academy of Sciences USA, 84, 2975–2979. Stebbins, C.L., Theodossy, S.J. and Longhurst, J.C. (1991). Cardiovascular reflexes evoked by histamine stimulation of the stomach. American Journal of Physiology, 260, H1098–H1105. Sternini, C. (1992). Enteric and visceral afferent CGRP neurons. Targets of innervation and differential expression patterns. Annals of the New York Academy of Sciences, 657, 170–186. Sternini, C., Reeve, J.R. and Brecha, N. (1987). Distribution and characterization of calcitonin gene-related peptide immunoreactivity in the digestive system of normal and capsaicin-treated rats. Gastroenterology, 93, 852–862. Stroff, T., Lambrecht, N., Burchert, M., Peskar, B.A. and Peskar, B.M. (1993). Calcitonin gene-related peptide (CGRP) is involved in the protective effect of cholecystokinin (CCK)-8 against ethanol in the rat stomach. Gastroenterology, 104, A199. Stroff, T. and Peskar, B.M. (1995a). Capsaicin-sensitive afferent nerves, calcitonin gene-related peptide and nitric oxide participate in the gastroprotective effect of epidermal growth factor (urogastrone) in the rat. Gastroenterology, 108, A227. Stroff, T. and Peskar, B.M. (1995b). Differential role of afferent neurons, calcitonin gene-related peptide and nitric oxide in various forms of gastroprotection. Gastroenterology, 108, A228. Stroff, T. and Peskar, B.M. (1996). The protective effect of intraluminal peptone in the rat stomach involves peptidergic and cholinergic neurons. Gastroenterology, 110, A265. Stroff, T., Plate, S., Respondek, M., Müller, K.M. and Peskar, B.M. (1995). Protection by gastrin in the rat stomach involves afferent neurons, calcitonin gene-related peptide and nitric oxide. Gastroenterology, 109, 89–97. Stroff, T., Plate, S. and Peskar, B.M. (1995a). Bombesin-induced protection in the rat stomach involves endogenous gastrin/cholecystokinin, afferent neurons, calcitonin gene-related peptide and nitric oxide. Gastroenterology, 108, A227. Stroff, T., Plate, S. and Peskar, B.M. (1995b). Corticotropin-releasing factor protects against ethanol-induced injury in the rat stomach: contribution of afferent neurons, calcitonin gene-related peptide and nitric oxide. Gastroenterology, 108, A227. Stroff, T., Plate, S., Seyed Ebrahim, J., Ehrlich, K.H., Respondek, M. and Peskar, B.M. (1996). Tachykinininduced increase in gastric mucosal resistance: role of primary afferent neurons, CGRP, and NO. American Journal of Physiology, 271, G1017–G1027. Su, H.C., Bishop, A.E., Power, R.F., Hamada, Y. and Polak, J.M. (1987). Dual intrinsic and extrinsic origins of CGRP- and NPY-immunoreactive nerves of rat gut and pancreas. Journal of Neuroscience, 7, 2674–2687. Sugo, S., Minamino, N., Kangawa, K., Miyamoto, K., Kitamura, K., Sakata, J. et al. (1994). Endothelial cells actively synthesize and secrete adrenomedullin. Biochemical and Biophysical Research Communications, 201, 1160–1166. Sullivan, T.R., Milner, R., Dempsey, D.T. and Ritchie, W.P. (1992). Effect of capsaicin on gastric mucosal injury and blood flow following bile acid exposure. Journal of Surgical Research, 52, 596–600. Sundler, F., Ekblad, E. and Håkanson, R. (1991). Occurrence and distribution of SP- and CGRP-containing nerve fibers in the gastric mucosa. Species differences. In Sensory Nerves and Neuropeptides in Gastroenterology: from Basic Science to Clinical Perspectives, edited by M. Costa, C. Surrenti, S. Gorini, C.A. Maggi and A. Meli, pp. 29–37. New York: Plenum. Sung, C.P., Arleth, A.J., Aiyar, N., Bhatnagar, P.K., Lysko, P.G. and Feuerstein, G. (1992). CGRP stimulates the adhesion of leukocytes to vascular endothelial cells. Peptides, 13, 429–434. Szallasi, A. (1994). The vanilloid (capsaicin) receptor: receptor types and species differences. General Pharmacology, 25, 223–243. Szolcsányi, J. (1984). Capsaicin-sensitive chemoceptive neural system with dual sensory-efferent function. In Antidromic Vasodilatation and Neurogenic Inflammation, edited by L.A. Chahl, J. Szolcsányi and F. Lembeck, pp. 27–52. Budapest: Akadémiai Kiadó. Szolcsányi, J. (1990). Effect of capsaicin, resiniferatoxin and piperine on ethanol-induced gastric ulcer of the rat. Acta Physiologica Hungarica, 75(Suppl.), 267–268. Szolcsányi, J. and Barthó, L. (1981). Impaired defense mechanism to peptic ulcer in the capsaicin-desensitized rat. In Gastrointestinal Defense Mechanisms, edited by G. Mózsik, O. Hänninen and T. Jávor, pp. 39–51. Oxford and Budapest: Pergamon Press and Akadémiai Kiadó. Szolcsányi, J. and Mozsik, G. (1984). Effects of capsaicin on the development of gastric mucosal damage by different necrotizing agents and of gastric cytoprotection by PGI2, atropine and cimetidine on rats. Acta Physiologica Hungarica, 64, 287–291.

CONTROL OF GASTRIC FUNCTIONS BY EXTRINSIC SENSORY NEURONS

169

Taché, Y. (1992). Inhibition of gastric acid secretion and ulcers by calcitonin gene-related peptide. Annals of the New York Academy of Sciences, 657, 240–247. Takenaga, M., Kawasaki, H., Wado, A. and Eto, T. (1995). Calcitonin gene-related peptide mediates acetylcholineinduced endothelium-independent vasodilation in mesenteric resistance blood vessels of the rat. Circulation Research, 76, 935–941. Takeuchi, K., Matsumoto, J., Ueshima, K. and Okabe, S. (1991a). Role of capsaicin-sensitive afferent neurons in alkaline secretory response to luminal acid in the rat duodenum. Gastroenterology, 101, 954–961. Takeuchi, K., Niida, H., Matsumoto, J., Ueshima, K. and Okabe, S. (1991b). Gastric motility changes in capsaicin-induced cytoprotection in the rat stomach. Japanese Journal of Pharmacology, 55, 147–155. Takeuchi, K., Matsumoto, J., Ueshima, K., Ohuchi, T. and Okabe, S. (1992a). Induction of duodenal ulcers in sensory deafferented rats following histamine infusion. Digestion, 51, 203–210. Takeuchi, K., Tachibana, K., Ueshima, K., Matsumoto, J. and Okabe, S. (1992b). Stimulation by capsaicin of gastric alkaline secretion in anesthetized rats. Japanese Journal of Pharmacology, 59, 151–157. Takeuchi, K., Ueshima, K., Matsumoto, J. and Okabe, S. (1992c). Role of capsaicin-sensitive sensory nerves in acid-induced bicarbonate secretion in rat stomach. Digestive Diseases and Sciences, 37, 737–743. Takeuchi, K., Ohuchi, T., Narita, M. and Okabe, S. (1993). Capsaicin-sensitive sensory nerves in recovery of gastric mucosal integrity after damage by sodium taurocholate in rats. Japanese Journal of Pharmacology, 63, 479–485. Takeuchi, K., Ueshima, K., Ohuchi, T. and Okabe, S. (1994). The role of capsaicin-sensitive sensory neurons in healing of HCl-induced gastric mucosal lesions in rats. Gastroenterology, 106, 1524–1532. Teng, C.H., Kang, J.Y., Chen, F.C. and Wee, A. (1996a). Effect of capsaicin and epidermal growth factor on gastroprotection in the rat: influence of sensory ablation, sialoadenectomy, and close arterial infusion with a calcitonin gene-related peptide antagonist. Gut, 39(Suppl. 3), A154. Teng, C.H., Kang, J.Y., Wee, A. and Lee, K.O. (1996b). Effect of epidermal growth factor, capsaicin and chilli ingestion on haemorrhagic shock-induced gastric mucosal injury. Gut, 39(Suppl. 3), A154. Tepperman, B.L. and Whittle, B.J.R. (1992). Endogenous nitric oxide and sensory neuropeptides interact in the modulation of the rat gastric microcirculation. British Journal of Pharmacology, 105, 171–175. Thiefin, G., Raybould, H.E., Leung, F.W., Taché, Y. and Guth, P.H. (1990). Capsaicin-sensitive afferent fibers contribute to gastric mucosal blood flow response to electrical vagal stimulation. American Journal of Physiology, 259, G1037–G1043. Thom, S.M., Hughes, A.D., Goldberg, P., Martin, G., Schachter, M. and Sever, P.S. (1987). The actions of calcitonin gene-related peptide and vasoactive intestinal peptide as vasodilators in man in vivo and in vitro. British Journal of Clinical Pharmacology, 24, 139–144. Tramontana, M., Renzi, D., Calabro, A., Panerai, C., Milani, S., Surrenti, C. et al. (1994a). Influence of capsaicinsensitive afferent fibers on acetic acid-induced chronic gastric ulcers in rats. Scandinavian Journal of Gastroenterology, 29, 406–413. Tramontana, M., Renzi, D., Panerai, C., Surrenti, C., Nappi, F., Abelli, L. et al. (1994b). Capsaicin-like effect of resiniferatoxin in the rat stomach. Neuropeptides, 26, 29–32. Tsutsumi, Y. and Hara, M. (1989). Peptidergic nerve fibers in human stomach and duodenal bulb: immunohistochemical demonstration of changes in fiber distribution in chronic gastritis. Biomedical Research, 10, 209–216. Uchida, M., Yano, S. and Watanabe, K. (1991a). Aggravation by the capsaicin treatment of gastric antral ulcer induced by the combination of 2-deoxy-D-glucose, aspirin and ammonia in rats. Japanese Journal of Pharmacology, 57, 377–385. Uchida, M., Yano, S. and Watanabe, K. (1991b). The role of capsaicin-sensitive afferent nerves in protective effect of capsaicin against absolute ethanol-induced gastric lesions in rats. Japanese Journal of Pharmacology, 55, 279–282. Uchida, M., Yano, S. and Watanabe, K. (1993). Involvement of CGRP, substance P and blood circulation in aggravating mechanism of absolute ethanol-induced antral lesions by capsaicin treatment in rats. Japanese Journal of Pharmacology, 62, 123–129. Uddman, R., Edvinsson, L., Ekblad, E., Håkanson, R. and Sundler, F. (1986). Calcitonin gene-related peptide (CGRP): perivascular distribution and vasodilatory effects. Regulatory Peptides, 15, 1–23. Vanner, S. (1994). Corelease of neuropeptides from capsaicin-sensitive afferents dilates submucosal arterioles in guinea pig ileum. American Journal of Physiology, 267, G650–G655. Vanner, S. and Surprenant, A. (1996). Neural reflexes controlling intestinal microcirculation. American Journal of Physiology, 271, G223–G230. Varro, A., Green, T., Holmes, S. and Dockray, G.J. (1988). Calcitonin gene-related peptide in visceral afferent nerve fibres: quantification by radioimmunoassay and determination of axonal transport rates. Neuroscience, 26, 927–932.

170

INNERVATION OF THE GASTROINTESTINAL TRACT

Vigna, S.R., Mantyh, C.R., Soll, A.H., Maggio, J.E. and Mantyh, P.W. (1989). Substance P receptors on canine chief cells: localization, characterization, and function. Journal of Neuroscience, 9, 2878–2886. Wachter, C., Heinemann, Á., Jocic, M. and Holzer, P. (1995). Visceral vasodilatation and somatic vasoconstriction evoked by acid challenge of the rat gastric mucosa: diversity of mechanisms. Journal of Physiology (London), 486, 505–516. Wachter, C.H., Heinemann, Á., Donnerer, J. and Holzer, P. (1998). Participation of 5-hydroxytryptamine in the femoral vasoconstriction evoked by acid challenge of the rat gastric mucosa. Journal of Physiology (London), 509, 541–550. Walder, C.E., Thiemermann, C. and Vane, J.R. (1990). Endothelium-derived relaxing factor participates in the increased blood flow in response to pentagastrin in the rat stomach mucosa. Proceedings of the Royal Society of London, Series B, 241, 195–200. Wallace, J.L., McKnight, G.W. and Befus, A.D. (1992). Capsaicin-induced hyperemia in the stomach: possible contribution of mast cells. American Journal of Physiology, 263, G209–G214. Wattchow, D.A., Jamieson, G.G., Maddern, G.J., Furness, J.B. and Costa, M. (1992). Distribution of peptidecontaining nerve fibers in the gastric musculature of patients undergoing surgery for gastroesophageal reflux. Annals of Surgery, 216, 153–160. Wei, E.P., Moskowitz, M.A., Boccalini, P. and Kontos, H.A. (1992). Calcitonin gene-related peptide mediates nitroglycerin and sodium nitroprusside-induced vasodilation in feline cerebral arterioles. Circulation Research, 70, 1313–1319. Whittle, B.J.R. (1977). Mechanisms underlying gastric mucosal damage induced by indomethacin and bile salts, and the actions of prostaglandins. British Journal of Pharmacology, 60, 455–460. Whittle, B.J.R. (1993). Neuronal and endothelium-derived mediators in the modulation of the gastric microcirculation: integrity in the balance. British Journal of Pharmacology, 110, 3–17. Whittle, B.J.R. and Lopez-Belmonte, J. (1991). Interactions between the vascular peptide endothelin-1 and sensory neuropeptides in gastric mucosal injury. British Journal of Pharmacology, 102, 950–954. Whittle, B.J.R., Lopez-Belmonte, J. and Moncada, S. (1990). Regulation of gastric mucosal integrity by endogenous nitric oxide: interactions with prostanoids and sensory neuropeptides in the rat. British Journal of Pharmacology, 99, 607–611. Whittle, B.J.R., Lopez-Belmonte, J. and Moncada, S. (1992). Nitric oxide mediates rat mucosal vasodilatation induced by intragastric capsaicin. European Journal of Pharmacology, 218, 339–341. Wood, J.W. (1993). Capsaicin in the Study of Pain. London: Academic Press. Yeo, C.J., Jaffe, B.M. and Zinner, M.J. (1984). The effects of intravenous substance P infusion on hemodynamics and regional blood flow in conscious dogs. Surgery, 95, 175–182. Yeoh, K.G., Kang, J.Y., Yap, I., Guan, R., Tan, C.C., Wee, A. and Teng, C.H. (1995). Chili protects against aspirin-induced gastroduodenal mucosal injury in humans. Digestive Diseases and Sciences, 40, 580–583. Yokotani, K. and Fujiwara, M. (1985). Effects of substance P on cholinergically stimulated gastric acid secretion and mucosal blood flow in rats. Journal of Pharmacology and Experimental Therapeutics, 232, 826–830. Yoneda, M. and Raybould, H.E. (1990). Capsaicin-sensitive vagal afferent fibers do not contribute to histamine H2 receptor agonist-induced gastric acid secretion in anesthetized rats. European Journal of Pharmacology, 186, 349–352. Yonei, Y., Holzer, P. and Guth, P.H. (1990). Laparotomy-induced gastric protection against ethanol injury is mediated by capsaicin-sensitive sensory neurons. Gastroenterology, 99, 3–9. Ziche, M., Morbidelli, L., Pacini, M., Dolara, P. and Maggi, C.A. (1990a). NK-1 receptors mediate the proliferative response of human fibroblasts to tachykinins. British Journal of Pharmacology, 100, 11–14. Ziche, M., Morbidelli, L., Pacini, M., Geppetti, P., Alessandri, G. and Maggi, C.A. (1990b). Substance P stimulates neovascularization in vivo and proliferation of cultured endothelial cells. Microvascular Research, 40, 264–278. Zimmerman, B.J., Gaginella, T.S. and Granger, D.N. (1991). Substance P: a modulator of leukocyte adhesion in rat mesenteric venules. Gastroenterology, 100, A265. Zimmerman, B.J., Anderson, D.C. and Granger, D.N. (1992). Neuropeptides promote neutrophil adherence to endothelial cell monolayers. American Journal of Physiology, 263, G678–G682. Zittel, T.T., Reddy, S.N., Plourde, V. and Raybould, H.E. (1994a). Role of spinal afferents and calcitonin generelated peptide in the postoperative gastric ileus in anesthetized rats. Annals of Surgery, 219, 79–87. Zittel, T.T., Rothenhöfer, I., Meyer, J.H. and Raybould, H.E. (1994b). Small intestinal capsaicin-sensitive afferents mediate feedback inhibition of gastric emptying in rats. American Journal of Physiology, 267, G1142–G1145.

4 Neural Control of the Large Intestine* Kalina Venkova, Beverley Greenwood-Van Meerveld and Jacob Krier Oklahoma Foundation for Digestive Research, Basic Science Labs, V.A. Medical Center, Oklahoma City, OK 73104, USA Efferent neurones in the extrinsic parasympathetic and sympathetic outflow from the central nervous system to the large intestine provide the reflex pathways which form and coordinate the motility patterns responsible for the storage, transport and evacuation of the colonic contents. Using experimental models of pelvic nerve and lumbar colonic nerve stimulation we have characterised the effects of selective activation of parasympathetic or sympathetic extrinsic pathways and have defined the neurotransmitters and receptors involved. In summary, we demonstrated that the extrinsic nervous control of the colon involves both prejunctional and direct smooth muscle effects. Efferent fibres to the colon provide cholinergic input to excitatory motoneurones in ganglia of the pelvic and the myenteric plexus, which induce smooth muscle contractions releasing acetylcholine and/or excitatory non-adrenergic, non-cholinergic (NANC) neurotransmitter(s). In addition, efferent parasympathetic fibres within the pelvic nerve may synapse with inhibitory motoneurones inducing muscle relaxation via the release of nitric oxide and/or inhibitory neurotransmitters such as vasoactive intestinal polypeptide and ATP. The relationship between nitric oxide-induced inhibition and a prostaglandin-dependent post-inhibitory excitation in colonic muscles is characteristic for NANC parasympathetic input to the colon. We believe that it plays a role in formation of parasympathetically controlled patterns of colonic activity. It has long been known that parasympathetic motor activity is under sympathetic influence by inhibitory α-adrenoceptors located on cholinergic nerve terminals. However, our observations have demonstrated direct motor responses in the circular muscle of non-sphincteric regions of the colon. Similar to vascular muscle innervation, postganglionic sympathetic fibres in the lumber colonic nerves co-release noradrenaline and ATP and express a presynaptic control mechanism operated by autoinhibitory α2-adrenoceptors. At the level of the muscle, the sympathetic motor response is formed by the activation of different adrenoceptor and purinoceptor subtypes i.e. α- and β-adrenoceptors and P2X receptors. KEY WORDS: colon; parasympathetic innervation; sympathetic innervation; adrenergic; cholinergic; NANC transmission.

INTRODUCTION The storage, transport and evacuation of colonic contents are accomplished by coordinated patterns of smooth muscle activity which in turn are regulated by the complex interaction between myogenic and neurogenic mechanisms. The neural regulation of colonic motility is provided by divisions of the autonomic nervous system organised in a three-rank

*Dedicated to the memory of our colleague and friend Jack Krier. 171

172

INNERVATION OF THE GASTROINTESTINAL TRACT

hierarchy. At the local level, a network of neurones located entirely within the gastrointestinal wall forms the enteric nervous system (ENS) which mediates simple stereotyped reflexes, such as ascending excitation and descending inhibition (Furness and Costa, 1987; Furness et al., 1995). At a higher level, peristaltic reflexes coordinate the activity of widely separated intestinal regions by neurones in the prevertebral ganglia (Szurszewski and King, 1989). Colonic motility is also under the influence of the central nervous system via neural connections contained within the extrinsic parasympathetic and sympathetic nerves supplying the gut. Until recently, the concept that parasympathetic nerves are excitatory to the colon and that sympathetic nerves cause inhibition of colonic motility was the generally accepted dogma. The most distal regions of the colon and rectum, receive prominent extrinsic parasympathetic innervation, with the pelvic nerves conveying the main parasympathetic innervation (McRorie, Krier and Adams, 1991) and providing phasic excitation during defecation and inhibition of the internal anal sphincter (Bouvier and Gonella, 1981b). The proximal part of the colon receives additional parasympathetic input through distal branches of the vagal nerves. The sympathetic nerves provide the major extrinsic supply to the caecum and proximal colon (Gonella, Bouvier and Blanquet, 1987; Furness and Bornstein, 1991). Postganglionic sympathetic fibres, which control colonic contractile activity, originate in the inferior mesenteric ganglion and in paravertebral ganglia, and travel to the colon in lumbar colonic nerves (de Groat and Krier, 1979). The most distal portion of the colon and the internal anal sphincter also receive sympathetic innervation through the hypogastric nerves (Baron, Janig and McLachlan, 1985a). Although the anatomical organisation of these pathways is well known, the physiological significance of the extrinsic innervation to the colon is less well understood. Except for the parasympathetic cholinergic excitation, the neuronal pathways and neurotransmitter mechanisms, involved in the extrinsic control of colonic contractile activity are still incompletely characterised. The lack of pharmacological data and the conflicting results obtained upon sympathetic nerve stimulation are due in part to the fact that, in most of the studies, colonic motility responses to extrinsic nerve stimulation have been investigated in vivo. These types of experiments provide accurate information about the functional significance of the extrinsic nervous control, however the technique does have limitations when studying receptors and pharmacological effects of specific neurotransmitters. A different approach was taken by Gillespie and his colleagues (Gillespie and McKenna, 1961; Gillespie and Khoyi, 1977) when they introduced an isolated preparation in which a colonic segment with the extrinsic lumbarcolonic and pelvic nerves attached was placed in an organ bath. In studies using this preparation the effects of extrinsic of sympathetic and parasympathetic nerve stimulation on the intraluminal pressure in the colon segment were investigated in the presence of selective adrenoceptor antagonists. The results supported the hypothesis that parasympathetic nerves induce motor responses which are controlled by sympathetic nerves and established that the main inhibitory effect of sympathetic nerves is due to activation of presynaptic inhibitory α-adrenoceptors located on cholinergic neurones. Subsequent data demonstrated the existence of postjunctional excitatory α-adrenoceptors in the circular muscle of non-sphincteric regions of the colon in different mammalian species (Gagnon and Belisle, 1970; Anuras and Christensen, 1981; Zhang et al., 1992) and in the human colon (Gagnon, Devroede and Belisle, 1972). While it is known that these excitatory

NEURAL CONTROL OF THE LARGE INTESTINE

173

postjunctional receptors respond to adrenoceptor agonists with contractile responses, it is unknown whether they are activated through the release of neurotransmitter(s) from sympathetic nerve fibres. The nature of parasympathetic motor responses has also proved to be more complicated since agonist-induced activation of nicotinic cholinoceptors in the ENS of the ileo-colonic region has been found to induce a prominent relaxatory response in the presence of atropine (Boeckxstaens et al., 1991). Studies using electrical field stimulation (EFS) of intramural nerve terminals, showed that nitric oxide (NO) and vasoactive intestinal polypeptide (VIP) are released and act as non-adrenergic, noncholinergic (NANC) inhibitory neurotransmitters in the ENS of the colon (Sanders and Ward, 1992). However, the physiological mechanisms of activation of NANC neurones in the colon and their functional significance still remains unresolved. In search of an answer, we concentrated our efforts on in vitro nerve-muscle preparations which are suitable for pharmacological studies of the mechanical responses of the colon circular or longitudinal muscle to stimulation of the extrinsic lumbar colonic (Venkova and Krier, 1993) or pelvic nerves (Kennedy and Krier, 1987a; Venkova and Krier, 1994). This chapter will review our findings related to the influence of lumbar and sacral autonomic neural pathways on the motor function of the colon with particular emphasis on the pharmacology of receptor mechanisms involved in the extrinsic neural control of colonic contractility.

PARASYMPATHETIC CONTROL OF THE COLON Preganglionic neurones in the sacral parasympathetic nucleus at intermediolateral regions of the sacral spinal cord are the source of efferent fibres that control the functions of the colon (Nadelhaft, de Groat and Morgan, 1980; de Groat et al., 1982; Nadelhaft et al., 1983). Parasympathetic efferent fibres emerge from the spinal cord in the second and third sacral ventral roots, than pass peripherally in a branch of the pelvic nerve to the pelvic plexus and to the colon, rectum and anal canal. Stimulation of sacral parasympathetic efferent fibres in animal models has led to conflicting results showing both excitation (de Groat and Krier, 1976, 1978) or inhibition (Fasth, Hulten and Nordgren, 1980) of contractility of the colon and rectum. Pelvic nerve stimulation in atropinised cats elicited a sustained contraction of the proximal colon and a relaxation of the rectum (Fasth et al., 1981). In our studies we investigated changes in colonic smooth muscle activity in response to stimulation of preganglionic pelvic nerve fibres using an isolated preparation of the cat colon. The findings revealed some new and exciting information about the extrinsic neural control of the colon (Kennedy and Krier, 1987a; Venkova and Krier, 1994). Our isolated preparation consists of a muscle strip, stripped of the mucosa and excised in the direction of the longitudinal muscle with one pelvic nerve attached to the muscle by connective tissue that carries branches of the nerve and a portion of the pelvicplexus ganglia. The preparation is mounted vertically in an organ bath to record isometric contractile activity of the longitudinal muscle. A bipolar platinum electrode is placed on the main branch of the pelvic nerve to stimulate preganglionic fibres and a pair of plate platinum electrodes, placed parallel to the muscle strip, supply transmural EFS. This preparation includes the efferent components of sacral parasympathetic pathways (preganglionic pelvic nerve fibres, pelvic plexus ganglia, colonic ganglia and colonic fibre bundles,

174

INNERVATION OF THE GASTROINTESTINAL TRACT

neurones and plexuses of the ENS and both muscle layers) and proved to be a powerful model to study the pharmacology of muscular motor responses in the colon. In the isolated colonic longitudinal muscle preparation with the pelvic nerve attached, we found that stimulation of preganglionic fibres induces cholinergic contractions, that are blocked by hexamethonium or atropine, and modulated by Met-enkephalin (Kennedy and Krier, 1987a,b). In the colon, sacral parasympathetic neurones provide excitatory input, synapsing with nicotinic cholinoceptors on postganglionic neurones in the colonic ganglia and/or in ganglia of the myenteric plexus, which then synapse with muscarinic cholinoceptors on smooth muscle cells. The data suggest that the activity of this pathway can be regulated by endogenous opioids acting on presynaptic δ-opioid receptors at the level of both ganglionic and nerve-muscle synapses. When adrenergic nerve terminals are depleted by guanethidine and cholinergic neuroeffector junctions are blocked by atropine, electrical stimulation of the pelvic nerve induces biphasic NANC responses of the longitudinal muscle of the colon. The NANC responses involve relaxation and/or inhibition of spontaneous phasic contractions, which occur during electrical stimulation and are followed by an excitatory component that develops upon cessation of the stimulus and which we referred to as an off-contraction (Venkova and Krier, 1994). A characteristic of the longitudinal muscle of the colon is that it maintains a low level of spontaneous tone (resting tension of about 0.1–0.2 g/mm2 cross sectional area) and generates spontaneous phasic contractions when isolated and suspended under standardised optimal tension in an organ bath. If the cholinergic and adrenergic input to the muscle is eliminated in the presence of atropine and guanethidine, most of the muscle strips lose phasic activity and remain mechanically silent. Under these conditions, stimulation of the pelvic nerve as well as EFS induce only an off-contractile response with its latency being considered a measure of NANC inhibition. The amplitude and the duration of the offcontractions are considered representative of a NANC excitation. Both phases of the response are neurally mediated because they are absent in preparations treated with tetrodotoxin, a neurotoxin known to abolish axonal neurotransmission. The responses to pelvic nerve stimulation are abolished also by hexamethonium, a ganglionic nicotinic cholinoceptor antagonist, that suppresses both inhibitory and excitatory components. In contrast, the responses to EFS are resistant to hexamethonium and are blocked only by tetrodotoxin. Therefore, it was assumed that in the colon NANC neurotransmission can be involved in both excitatory and inhibitory control of colonic motility and such control can be obtained through an extrinsic parasympathetic pathway that originates in the sacral spinal cord. ROLE OF NITRIC OXIDE We studied the involvement of nitric oxide (NO) as an inhibitory neurotransmitter of NANC responses to pelvic nerve stimulation in the cat colon using inhibitors of NO synthase (NOS): N ω-nitro-L-arginine (L-NNA), N ω-nitro-L-arginine methyl ester (L-NAME) and N ω-monomethyl-L-arginine (L-NMMA). NO synthesised from L-arginine by a calciumdependent constitutive NOS (Moncada, Palmer and Higgs, 1991) has been demonstrated to be a mediator of NANC relaxation following neuronal stimulation in a wide range of intestinal preparations (Christinck et al., 1991; Sanders and Ward, 1992) including the human colon (Boeckxstaens et al., 1993). We found that all NOS inhibitors reduced

NEURAL CONTROL OF THE LARGE INTESTINE

175

the latency of the off-contractions to both pelvic nerve and intramural stimulation, however neither of the NOS inhibitors was able to “shift” the contractile phase of the response to the beginning of the stimulus train i.e. bring the latency to zero (Figure 4.1), strongly suggesting that NO may not be the sole transmitter of this inhibitory response. Pretreatment with α-chymotrypsin at a concentration which abolished VIP-induced relaxation (Li and Rand, 1990), further decreased the duration of the inhibitory component of NANC responses. However, the effect of the NOS inhibitors was considered specific since it was reversed by the excess of L-arginine but not by D-arginine. Our results indicate that in the colon NO and an inhibitory peptide, most likely VIP, are released upon stimulation of the pelvic nerves and may act as co-transmitters to maintain the inhibitory phase of the smooth muscle response. To further establish the role of NO as an inhibitory neurotransmitter in the colon the distribution and abundance of NOS containing neurones have been examined using nicotinamide adenine dinucleotide phosphate (NADPH) diaphorase histochemistry and NOS immunohistochemistry (Barbiers et al., 1993; Furness et al., 1994; Timmermans et al., 1994). In muscle preparations of cat colon, with branches of the pelvic nerve attached, NADPH diaphorase histochemistry showed the existence of nitrergic neurones in the myenteric and submucosal plexuses in the distal part of the colon and rectum. Nitrergic neurones were found also in extramural parasympathetic ganglia connecting colonic branches of the pelvic nerve and in fibre bundles connecting colonic ganglia with the serosal surface of the distal colon (Venkova and Krier, unpublished observations).

Figure 4.1 Isolated cat pelvic nerve-colonic muscle preparation. Off-contractions induced by transmural stimulation (A) or stimulation of the pelvic nerve (B). Horizontal bars indicate the time of electrical stimulation. The upper traces show off-contractions elicited by EFS (2 Hz, 0.5 ms pulse duration, 30 s train duration) and pelvic nerve stimulation (8 Hz, 0.5 ms pulse duration, 45 s train duration) in the absence of NOS inhibitor. The middle traces illustrate the inhibitory effect of L-NNA (100 µM) on the latency, amplitude and duration of the off-contractile responses. A partial reversal of the inhibitory effects by L-arginine (120 µM) is shown in the lower traces. (From Venkova and Krier, 1994, with permission).

176

INNERVATION OF THE GASTROINTESTINAL TRACT

Although there is no direct evidence about the functional role of these nitrergic neurones, we postulate that they may be inhibitory motoneurones activated by preganglionic parasympathetic input through nicotinic cholinoceptors, because the NANC inhibitory responses to stimulation of the pelvic nerve were blocked by hexamethonium (Venkova and Krier, 1992, 1994). Other more complete studies in the guinea-pig found NOScontaining cell bodies in the submucosal plexus of the colon and rectum, and NOS containing terminals in the circular muscle, including the sphincters. Within the myenteric ganglia NOS-containing terminals were extremely dense in the proximal colon and rectum. NOS neurones have been classified as: (1) inhibitory motor neurones to the circular muscle throughout the gastrointestinal tract and (2) interneurones in the ileum and colon. The existence of two distinct subpopulations of NOS-containing enteric neurones acting as interneurones or inhibitory motor neurones was confirmed also for the ENS of the developing human digestive tract (Timmermans et al., 1994). In the cat colon we have demonstrated that motoneurones releasing NO are activated by extrinsic parasympathetic neurones running within the pelvic nerves and synapsing via nicotinic cholinoceptors. ROLE OF PROSTAGLANDINS The nature of the off-contractions, which follow inhibition of smooth muscle activity, has been a matter of speculation for more than two decades (Bennett, 1966). It is now apparent that the rebound excitation phenomenon is a complex event, which appears in response to intramural or extrinsic parasympathetic stimulation. In the longitudinal muscle of the colon, the NANC off-contraction is preceded by smooth muscle inhibition, requires the synthesis of prostaglandins and may also involve the effect of an excitatory NANC neurotransmitter. The pharmacology of the neurally mediated NANC off-contraction in the colon is still a matter of speculation, although it has been suggested that rebound contractions which follow inhibitory responses result from the effect of an excitatory NANC neurotransmitter, which is masked during stimulation (Wood and Marsh, 1973). An additional myogenic rebound phenomenon following smooth muscle membrane hyperpolarisation may contribute to the generation of off-contractions. Prostaglandins have been implicated in the development of rebound excitation in the guinea-pig taenia coli (Burnstock et al., 1975), mouse colon (Fontaine, Grivegnee and Reuse, 1984) and canine proximal colon (Ward et al., 1992) because cyclooxygenase inhibitors reduce the off-contractions or post-inhibitory smooth muscle membrane hyperpolarisation. In the cat colon longitudinal muscle we hypothesised that prostaglandins may play a role in mediating a component of the off-contraction. We tested our idea using indomethacin, an inhibitor of prostaglandin synthesis that blocks cyclooxygenase. Although preincubation with indomethacin did not change the tone or the spontaneous contractions, there was a concentration-dependent reduction in amplitude and duration of NANC off-contractions evoked by EFS or pelvic nerve stimulation. The maximum inhibition of amplitude and duration of off-contractions was an average of 45–70% of control responses measured in the absence of indomethacin (Figure 4.2). In contrast, indomethacin did not change the latency of the off-contractions evoked by either EFS or stimulation of the pelvic nerve, suggesting that the release of prostaglandins is a post-stimulus event involved only in the development of the off-contraction.

NEURAL CONTROL OF THE LARGE INTESTINE

177

Figure 4.2 Isolated cat pelvic nerve-colonic muscle preparation. Off-contractions induced by transmural stimulation (A) or stimulation of the pelvic nerve (B). Effects of indomethacin applied at concentrations of 10 µM and 30 µM on the latency (upper panels), amplitude (middle panels) and duration (lower panels) of the off-contractile responses. Filled symbols represent time-matched control responses and open symbols represent the effect of indomethacin. Note that indomethacin significantly (*P < 0.05; **P < 0.01) reduces the amplitude and duration of the off-contraction without changing its latency. (From Venkova and Krier, 1994, with permission).

Others have reported that off-excitation also occur in the canine colon, especially in the region where electrical slow waves are generated (Smith, Reed and Sanders, 1987; Sanders and Smith, 1989). Responses to activation of intrinsic inhibitory nerves were characterised by hyperpolarisation and reduction in slow wave amplitude and duration during the period of stimulation and a rebound enhancement of the slow waves immediately following the stimulus. In the canine colon, exogenous NO mimicked the inhibitory responses to nerve stimulation inducing: hyperpolarisation of the membrane potential, inhibitions of slow waves and a period of post-inhibitory rebound excitation following the inhibitory response to NO. Pretreatment with indomethacin reduced the rebound excitation following NO application without blocking the inhibitory components of the

178

INNERVATION OF THE GASTROINTESTINAL TRACT

responses (Ward et al., 1992). These results are compatible with the finding that in the cat colon NANC off-contractions induced by stimulation of the pelvic nerve are positively linked to the NO released during stimulation i.e. the magnitude of off-contractions decreased with the decrease of neurally released NO brought about by pretreatment with competitive NOS inhibitors (L-NAME, L-NNA and L-NMMA) or hydroquinone which enhances the formation of superoxide free radicals, shortening the life-time of NO (Venkova and Krier, 1994). In summary, motoneurones in the extrinsic parasympathetic outflow to the colon involve both cholinergic and NANC excitatory neurotransmission to induce colonic contraction. Inhibitory pathways are also activated by extrinsic parasympathetic neurones and are probably essential in coordinating colonic contractile activity to form complicated motility patterns involved in the storage and elimination of body waste. The chemistry of the inhibitory pathways remains to be elucidated though it is clear that NO plays an important role as an inhibitory motor transmitter.

SYMPATHETIC CONTROL OF THE COLON The lumbar sympathetic outflow to the large intestine consists of preganglionic fibres arising from neurones located in the autonomic nucleus in the spinal cord (L2–L4 segments), which in most species pass through the sympathetic chain of paravertebral ganglia and synapse with neurones in the prevertebral ganglia, e.g. inferior mesenteric and pelvic plexus ganglia. Postganglionic fibres arising from neurones in the inferior mesenteric ganglia divide into two principal groups: the lumbar colonic nerves and the hypogastric nerves. The lumbar colonic nerves form a diffuse network around the inferior mesenteric artery and accompany the artery to the distal colon and rectum (Costa and Gabella, 1971). Sympathetic postganglionic fibres that travel to the colon in lumbar colonic nerves innervate vascular and visceral smooth muscle (de Groat and Krier, 1979; Baron, Janig and McLachlan, 1985b). Sympathetic fibres also innervate the epithelial cells, providing adrenergic regulation of fluid and electrolyte transport in the colon (Hubel, 1985; Smith and McCabe, 1986). The most distal part of the colon and the internal anal sphincter receive sympathetic innervation through the hypogastric nerves. In most of the colon, adrenergic nerves densely innervate the circular muscle but sparsely innervate longitudinal muscle fibres. The longitudinal muscle receives dense adrenergic innervation in the most distal region of the colon and the rectum, as well as the internal anal sphincter (Howard and Garret, 1973). In general, the sympathetic supply to the large intestine is associated with a tonic inhibition of colonic contractility and the maintenance of high tone by the internal anal sphincter. However, the physiological significance of the sympathetic supply remains unclear, especially in the non-sphincteric regions of the colon. The influence of extrinsic sympathetic outflow on the contractile activity of the colon has been among the most studied but least understood mechanisms. Different models of sympathetic stimulation yielded controversial data about the sympathetic regulation of colonic motility. The classic experiments carried out in whole animals indicate relaxation of the colon and rectum and contraction of the internal anal sphincter in response to stimulation of extrinsic sympathetic pathways (Garry, 1933; Hulten, 1969). Subsequently, acute

NEURAL CONTROL OF THE LARGE INTESTINE

179

transection of the lumbar colonic nerves enhanced the basal intraluminal pressure and the amplitude of spontaneous phasic contractions in the non-sphincteric regions of the colon, indicating a tonic inhibitory control (de Groat and Krier, 1979). In the anaesthetised cat efferent electrical stimulation of the lumbar colonic and hypogastric nerves caused a sustained contraction of the distal colon and rectum which was inhibited by phentolamine and unaffected by hexamethonium suggesting direct α-adrenergic effect (Hedlung, Fasth and Hulten, 1984). In contrast, electrical stimulation of lumbar colonic nerves was found to decrease the basal intraluminal pressure and the amplitude of cholinergic pelvic nerveevoked contractions (Gillespie and McKenna, 1961; de Groat and Krier, 1976; Gillespie and Khoyi, 1977). Activation of lumbar sympathetic nerves caused a constriction in the sphincter region which was attributed to activation of smooth muscle α-adrenoceptors (Garret and Howard, 1975). The inhibitory effects observed in the non-sphincteric regions were related to presynaptic inhibition of cholinergic activity via α2-adrenoceptors (Gillespie and Khoyi, 1977; De Groat and Krier, 1976) and to a lesser extent to the direct activation of smooth muscle inhibitory β-adrenoceptors (Bouvier and Gonella, 1981a,b). In addition to inhibitory β-adrenoceptors on colonic muscle cells (Ek, 1985), excitatory α-adrenoceptors have been characterised in the circular muscle of non-sphincteric regions in cat and dog colon (Anuras and Christensen, 1981; Zhang et al., 1992; Venkova, Milne and Krier, 1994; Venkova and Krier, 1995) and in both circular and longitudinal muscle of the rat and human colon (Gagnon and Belisle, 1970; Gagnon, Devroede and Belisle, 1972). While it is known that activation of α-adrenoceptor is coupled to contraction of the circular muscle it is unknown whether the α-adrenoceptors are activated through the release of neurotransmitter from the lumbar sympathetic nerve fibres. To characterise the responses of the colon to stimulation of extrinsic sympathetic nerves we developed an isolated lumbar colonic nerve-muscle preparation (Venkova and Krier, 1993). The preparation includes a mucosa-free muscle strip, cut in the direction of the circular muscle, with attached branches of the lumbar colonic nerves. The spontaneous activity of the circular muscle is characterised by low resting tension (0.2–0.35 g/mm2 cross sectional area) and rhythmic low-amplitude contractions generated at a frequency of 4–5 cycles per min. Electrical stimulation of the lumbar colonic nerves causes a frequency-dependent increase in tone and in the amplitude of spontaneous contractions.

Figure 4.3 Isolated cat lumbar colonic nerve-colonic muscle strip. Inhibitory effects of guanethidine on contractile responses of the circular muscle induced by stimulation of colonic nerve fibres (2–20 Hz, 0.5 ms pulse duration, 45 s train duration). (From Venkova and Krier, 1993, with permission).

180

INNERVATION OF THE GASTROINTESTINAL TRACT

Lumbar colonic nerve-evoked contractions occur in the presence of hexamethonium or atropine but are abolished by either tetrodotoxin or guanethidine (Figure 4.3) and by sectioning the nerve between the electrode and the muscle. Thus, the isolated lumbar colonic nerve-colonic muscle preparation presents an in vivo model for studying sympathetic responses which avoids nonselective activation of intramural nerve terminals and provides the selective activation of neurotransmitter(s) release from sympathetic nerve fibres contained in branches of the lumbar colonic nerves. NORADRENALINE AND ATP AS CO-TRANSMITTERS OF SYMPATHETIC NERVES TO THE COLON In order to further characterise the pharmacology of the sympathetic innervation to the colon, experiments were designed using the lumbar colonic nerve-circular muscle preparation. Contractile responses to stimulation (2–20 Hz, 0.5 ms) of the lumbar colonic nerves were tested in the presence of the α-adrenoceptor antagonists phentolamine or prazosin. Contractions of colonic circular muscle induced by exogenous administration of noradrenaline or phenylephrine were completely inhibited by the α-adrenoceptor antagonists. In contrast, both agonists concentration-dependently reduced the lumbar colonic nerve-evoked contractile responses to approximately 50% of control value but did not abolish them. Since the remaining component of the responses was abolished by guanethidine, an agent used to deplete the presynaptic vesicles in sympathetic neurones, it was described as reflecting the co-release of a second neurotransmitter from the sympathetic terminals. The possibility of β-adrenoceptor inhibition being part of the overall response to lumbar colonic nerve stimulation was tested in preparations treated with propranolol, a selective β-adrenoceptor antagonist. Propranolol caused a concentration-dependent increase in the amplitude of lumbar colonic nerve-evoked contractions. Similarly, the inhibition of β-adrenoceptors caused a potentiation of the contractile responses to exogenous noradrenaline but not to phenylephrine. Taken together our findings suggest that lumbar sympathetic nerves regulate the contractile activity of the circular muscle in non-sphincteric regions of the colon releasing noradrenaline, which activates both excitatory α-adrenoceptors and inhibitory β-adrenoceptors. The co-release of one or more non-adrenergic neurotransmitters inducing a contractile response in colonic circular muscle, is also evident. Although the identity of the neurotransmitter co-released with noradrenaline in the colon is unknown, ATP is considered a putative candidate since in the guinea-pig colon vasoconstriction of submucosal arterioles following stimulation of sympathetic nerves is mediated by ATP (Evans and Surprenant, 1992). In the cat colon, administration of ATP and α,β-methylene ATP were found to contract the circular muscle and these contractile responses were abolished after desensitisation with α,β-methylene ATP. Moreover, we discovered that the contractions evoked by stimulation of lumbar colonic nerves decrease in amplitude after desensitisation of P2 purinoceptors with α,β-methylene ATP. Furthermore, ANAPP3 (arylazido-aminopropyl-ATP), known to act as specific P2X purinoceptor antagonist (Hogaboom, O’Donnell and Fedan, 1980; Fedan et al., 1982; Burnstock and Kennedy, 1985) caused an irreversible reduction in the amplitude of the lumbar colonic nerve-evoked responses. In both cases the remainder of the contractile response was

NEURAL CONTROL OF THE LARGE INTESTINE

181

Figure 4.4 Isolated cat lumbar colonic nerve-colonic muscle strip. (A) Inhibitory effects induced by the consecutive addition of prazosin and α,β-methylene ATP (α,β-mATP) on the contractions of the circular muscle elicited by stimulation of colonic nerve fibres. Horizontal bars indicate the time of electrical stimulation. (B) Summary graph of the inhibitory effects of prazosin (3 µM, open triangles) and α,β-mATP (5 µM, filled triangles) applied on the background of prazosin. The responses are expressed as percentage of controls (filled circles) in untreated preparations for each of the stimulus frequencies. (From Venkova and Krier, 1993, with permission).

further inhibited by prazosin. When α1-adrenoceptors were blocked and P2 purinoceptors were desensitised, the colonic contractile responses to lumber colonic nerve stimulation were virtually abolished, with the exception of a low-amplitude contraction developing in response to high frequency stimulation (Figure 4.4). The order in which prazosin and

182

INNERVATION OF THE GASTROINTESTINAL TRACT

α,β-methylene ATP were added to the bathing solution did not change the final effect suggesting that once released from the nerve terminal, noradrenaline and a nucleotide, probably ATP, interact with the smooth muscle receptors independently. The possible involvement of a third neurotransmitter responsible for a small portion of the excitatory response has not been studied. The effect of “nonadrenergic, noncholinegic, nonpurinergic” neurotransmitter is evident only for the high frequencies (10–20 Hz) of lumbar colonic nerve stimulation. The nature of such a neurotransmitter remains unknown, however its dependence to high stimulus frequencies, clearly suggests that a tachykinin or a prostaglandinlike modulatory substance may have been released. Presynaptic auto-inhibitory receptors, predominantly of the α2-adrenoceptor type, are a common feature of sympathetic junctions in vascular and visceral muscles (Starke, Gothert and Kilbinger, 1989). Therefore we aimed to establish whether the sympathetic neurotransmission to the colon is regulated by presynaptic autoinhibitory control. We examined the responses to stimulation of the lumbar colonic nerves in the presence of clonidine, a selective α2-adrenoceptor agonist or yohimbine, a selective α2-adrenoceptor antagonist. Clonidine caused a concentration-dependent reduction of lumbar colonic nerve evoked contractions, while yohimbine increased the responses. Both compounds had no effect on basal tone and spontaneous contractile activity. These results are consistent with the existence of prejunctional α2-adrenoceptors maintaining a negative feedback control of neurotransmitter release by the sympathetic terminals. However, whether α2-adrenoceptor activation is linked to inhibition of the co-release of a purinergic neurotransmitter or regulates only the release of noradrenaline remains to be determined. PURINE RECEPTORS Since 1970, when Geoffrey Burnstock postulated that ATP is an inhibitory NANC neurotransmitter in the mammalian intestine, the list of potential purinergic neurotransmitters has been expanded to include the adenine nucleotides (ATP, ADP and AMP) and the adenine nucleoside adenosine. In the past, problems of desensitisation and rapid interconversion of purine derivatives complicated investigation, however stable analogues resistant to metabolism are now available. According to the recent nomenclature the receptors for adenosine and adenine nucleotides are divided into two categories: P1 purine receptors activated by adenosine and designated as adenosine receptors (A1 and A2) and P2 purine receptors (P2X, P2Y, P2Z and P2T) activated by ATP and its analogues (Burnstock and Kennedy, 1985; Abbracchio and Burnstock, 1994). Activation of purine receptors in different smooth muscles shows a wide variety of effects including inhibition, relaxation or even biphasic responses. Furthermore, our experiments demonstrated that purine agonist induce different effects in one and the same muscle depending on the level of muscle tone. When the level of basal tone in isolated cat colon circular muscle is low ATP and the ATP analogues α,β-methylene-ATP (α,β-MeATP), β,γ-methylene-ATP (β,γ-MeATP) and 2-methylthio-ATP (2-MeSATP) induced concentration-dependent contractions with a potency order of α,β-MeAT > β,γ-MeATP > 2-MeSATP > ATP. This order of potency is considered characteristic for P2X purinoceptors as it has been defined in the urinary bladder (Hoyle, Knight and Burnstock, 1990) and in a number of blood vessels (see Burnstock, 1991). The existence of P2X purinoceptors causing contraction of the colon circular

NEURAL CONTROL OF THE LARGE INTESTINE

183

Figure 4.5 Schematic illustration of the extrinsic parasympathetic and sympathetic efferent pathways to colonic musculature. Preganglionic nerve fibres running with the pelvic nerves synapse with motoneurones in ganglia of the pelvic and myenteric plexus. In addition to the cholinergic contraction, the pelvic nerves induce NANC inhibitory responses. The putative inhibitory neurotransmitters released at the level of the neuromuscular junction, NO and VIP, contribute to the development of parasympathetic motor responses. Sympathetic postganglionic fibres emerging from the inferior mesenteric ganglion innervate the circular muscle of the colon and produce excitatory responses in non-sphincteric regions. There are pharmacological evidences that ATP is co-released with noradrenaline at the neuromuscular junction. The sympathetically mediated motor responses express both excitatory and inhibitory components induced by the activation of smooth muscle P2X purinoceptors and α1-adrenoceptors causing contractions and β-adrenoceptors causing relaxation. The activity of parasympathetic and sympathetic motor pathways to the large intestine is integrated by sympathetic nerve regulation via presynaptic α-adrenoceptors and α2-autoreceptors.

muscle was confirmed by the rightward shift of the concentration-effect curves in the presence of suramin, a selective and competitive antagonist of the contractile responses to α,β-MeATP and ATP in other smooth muscles (Leff, Wood and O’Connor, 1990; von Kügelgen, Bültmann and Starke, 1990). In the cat colon circular muscle suramin acted as a competitive antagonist of α,β-MeATP contractions with a pA2 value of 4.45 (Kb = 30.6 µM) and a slope of the Schild plot of 1.06. However, the contractile responses to ATP were less sensitive to suramin (Kb = 52.3 µM) suggesting that ATP may either activate a suramin-insensitive component of the P2X purinoceptor as was suggested in the mouse vas deferens (von Kügelgen, Bültmann and Starke, 1990) or may interact with another “nucleotide” receptor (O’Connor, Dainty and Leff, 1991). The latter is consistent with the development of inhibitory responses to ATP and 2-MeSATP in colonic circular muscles precontracted with noradrenaline. However, α,β-MeATP and β,γ-MeATP induced only contractile responses which were blocked by suramin, regardless of the

184

INNERVATION OF THE GASTROINTESTINAL TRACT

initial level of smooth muscle tone. A purine receptor which differs from the P2X purinoceptor, and is activated by ATP or by adenosine derived as the product of ATP degradation was suggested to be responsible for an inhibitory effect in the cat colon. Reactive blue 2, a selective antagonist of P2Y purinoceptors (Manzini, Hoyle and Burnstock, 1986), inhibited the relaxation induced by ATP or 2-MeSATP by 80–90%. The remaining relaxation was blocked by 8-(p-sulfophenyl)-theophylline, an antagonist of adenosine P1 purinoceptors (Kennedy, 1990). Furthermore, at low levels of muscle tone, reactive blue 2 increased the amplitude of contractile responses to ATP but did not change the contractions to α,β-MeATP. Data from other experiments indicate that inhibitory P2Y purinoceptors in the gastrointestinal tract are not restricted to the circular muscle of the colon, because relaxations to ATP which are blocked by reactive blue 2 were found in different regions including the gastric fundus (Lefebvre and Burnstock, 1991), duodenum (Manzini, Maggi and Meli, 1985) and caecum (Manzini, Hoyle and Burnstock, 1986). The pharmacological analysis of purine effects in the colon proved the existence of three distinct types of smooth muscle purinoceptors: P2X purinoceptors causing contraction, P2Y purinoceptors causing relaxation and inhibitory P1 purinoceptors activated by the hydrolytic breakdown of ATP to adenosine. The distribution of receptors amongst the extrinsic sympathetic and parasympathetic pathways to the colon that has been described in this review are schematically summarised in Figure 4.5.

REFERENCES Abbracchio, M.P. and Burnstock, G. (1994). Purinoceptors: are there families of P2X and P2Y purinoceptors? Pharmacology and Therapeutics, 64, 445–475. Anuras, S. and Christensen, J. (1981). Effects of autonomic drugs on cat colonic muscle. American Journal of Physiology, 240, G361–G364. Barbiers, M., Timmermans, J.-P., Scheuermann, D.W., Adriensen, D., Mayer, B. and De Groodt-Lasseel, M.H. (1993). Distribution and morphological features of nitrergic neurons in the porcine large intestine. Histochemistry, 100, 23–34. Baron, R., Janig, W. and McLachlan, E.M. (1985a). The afferent and sympathetic components of the lumbar spinal outflow to the colon and pelvic organs in the cat. I. The hypogastric nerve. Journal of Comparative Neurology, 238, 135–146. Baron, R., Janig, W. and McLachlan, E.M. (1985b). The afferent and sympathetic components of the lumbar spinal outflow to the colon and pelvic organs in the cat. III. The colonic nerves, incorporating an analysis of all components of the lumbar prevertebral outflow. Journal of Comparative Neurology, 238, 158–168. Bennett, M.R. (1966). Rebound excitation of the smooth muscle cells of the guinea-pig taenia coli after stimulation of intramural inhibitory nerves. Journal of Physiology (London), 185, 124–131. Boeckxstaens, G.E., Pelckmans, P.A., Ruytjens, I.F., Bult, H., De Man, J.G., Herman, A.G. and Van Maercke, Y.M. (1991). Bioassay of nitric oxide released upon stimulation of non-adrenergic, non-cholinergic nerves in the canine ileocolonic junction. British Journal of Pharmacology, 103, 1085–1091. Boeckxstaens, G.E., Pelckmans, P.A., Herman, A.G. and Van Maercke, Y.M. (1993). Involvement of nitric oxide in the inhibitory innervation of the human isolated colon. Gastroenterology, 104, 690–697. Bouvier, M. and Gonella, J. (1981a). Electrical activity from smooth muscle of the anal sphincteric area of the cat. Journal of Physiology (London), 310, 445–456. Bouvier, M. and Gonella, J. (1981b). Nervous control of the internal anal sphincter of the cat. Journal of Physiology (London), 310, 457–469. Burnstock, G. (1991). Distribution and roles of purinoceptors. Nucleosides Nucleotides, 10, 917–930. Burnstock, G. and Kennedy, C. (1985). Is there a basis for distinguishing two types of P2-purinoceptor? General Pharmacology, 16, 433–440.

NEURAL CONTROL OF THE LARGE INTESTINE

185

Burnstock, G., Cocks, T., Paddle, B. and Staszewska-Barczak, J. (1975). Evidence that prostaglandin is responsible for the “rebound contraction” following stimulation of non-adrenergic, non-cholinergic (“purinergic”) inhibitory nerves. European Journal of Pharmacology, 31, 360–362. Christinck, F., Jury, J., Cayabyab, F. and Daniel, E.E. (1991). Nitric oxide may be the final mediator of nonadrenergic non-cholinergic inhibitory junction potentials in the gut. Canadian Journal of Physiology and Pharmacology, 9, 1448–1458. Costa, M. and Gabella, G. (1971). Adrenergic innervation of the alimentary canal. Zeitschrift für Zellforschung Mikroskopische Anatomie, 122, 357–377. de Groat, W.C. and Krier, J. (1976). An electrophysiological study of the sacral parasympathetic pathway to the colon of the cat. Journal of Physiology (London), 260, 425–445. de Groat, W.C. and Krier, J. (1978). The sacral parasympathetic reflex pathway regulating colonic motility and defaecation in the cat. Journal of Physiology (London), 276, 481–500. de Groat, W.C. and Krier, J. (1979). The central control of the lumbar sympathetic pathway of the large intestine of the cat. Journal of Physiology (London), 289, 449–468. de Groat, W.C., Booth, A.M., Milne, R.J. and Roppolo, J.R. (1982). Parasympathetic preganglionic neurones in the sacral spinal cord. Journal of the Autonomic Nervous System, 5, 23–43. Ek, B. (1985). Studies on mechanisms for β-adrenoceptor mediated inhibition of colon motility. Acta Physiologica Scandinavica, 125 (Suppl. 543), 5–39. Evans, R.J. and Surprenant, A. (1992). Vasoconstriction of guinea-pig submucosal arterioles following sympathetic nerve stimulation is mediated by the release of ATP. British Journal of Pharmacology, 106, 242–249. Fasth, S., Hulten, L. and Nordgren, S. (1980). Evidence for a dual pelvic nerve influence on large bowel motility in the cat. Journal of Physiology (London), 298, 159–169. Fasth, S., Hulten, L., Nordgren, S. and Zeitlin, I.J. (1981). Studies on the atropine-resistant sacral parasympathetic vascular and motility responses in the cat colon. Journal of Physiology (London), 311, 421–429. Fedan, J.S., Hogaboom, G.K., Westfall, D.P. and O’Donnell, J.P. (1982). Comparison of the effects of arylazido aminoproprionyl ATP (ANAPP3), an ATP antagonist, on responses of the smooth muscle of guinea-pig vas deferens to ATP and related nucleotides. European Journal of Pharmacology, 85, 277–290. Fontaine, J., Grivegnee, A. and Reuse, J. (1984). Adrenoceptors and regulation of intestinal tone in the isolated colon of the mouse. British Journal of Pharmacology, 81, 231–243. Furness, J.B. and Bornstein, J.C. (1991). The enteric nervous system and its extrinsic connections. In Textbook of Gastroenterology, Vol. I, 2nd edition, edited by T. Yamada, pp. 2–24, Philadelphia: J.B. Lippincott. Furness, J.B. and Costa, M. (1987). The Enteric Nervous System. New York: Churchill Livingstone. Furness, J.B., Li, Z.S., Young, H.M. and Forstermann, U. (1994). Nitric oxide synthase in the enteric nervous system of the guinea-pig: a quantitative description. Cell and Tissue Research, 277, 139–149. Furness, J.B., Johnson, P.J., Pompolo, S. and Bornstein, J.C. (1995). Evidence that enteric motility reflexes can be initiated through entirely intrinsic mechanisms in the guinea-pig small intestine. Neurogastroenterology and Motility, 7, 89–96. Gagnon, D.J. and Belisle, S. (1970). Stimulatory effects of catecholamines on isolated rat colon after β-adrenergic blockade with oxprenolol and propranolol. European Journal of Pharmacology, 12, 303–309. Gagnon, D.J., Devroede, G. and Belisle, S. (1972). Excitatory effects of adrenaline upon isolated preparations of human colon. Gut,13, 654–657. Garrett, J.R. and Howard, E.R. (1975). Neural control of the internal anal sphincter of cats after chemical sympathectomy with 6-hydroxydopamine. Journal of Physiology (London), 247, 25P–27P. Garry, R.C. (1933). The nervous control of the caudal region of the large bowel. Journal of Physiology (London), 77, 422–431. Gillespie, J.S. and Khoyi, M.A. (1977). The site and receptors responsible for the inhibition by sympathetic nerves of intestinal smooth muscle and its parasympathetic motor nerves. Journal of Physiology (London), 267, 767–789. Gillespie, J.S. and McKenna, B.R. (1961). The inhibitory actions of the sympathetic nerves on the smooth muscle of the rabbit gut, its reversal by reserpine and restoration by catecholamines and DOPA. Journal of Physiology (London), 156, 17–34. Gonella, J., Bouvier, M. and Blanquet, F. (1987). Extrinsic nervous control of motility of small and large intestines and related sphincters. Physiological Reviews, 67, 902–961. Hedlund, H., Fasth, S. and Hulten, L. (1984). Efferent sympathetic nervous control of rectal motility in the cat. Acta Physiologica Scandinavica, 121, 317–324. Hogaboom, G.K., O’Donnell, J.P. and Fedan, J.S. (1980). Purinergic receptors: photoaffinity analog of adenosine triphosphate is a specific adenosine triphosphate antagonist. Science, 208, 1273–1276. Howard, E.R. and Garrett, J.R. (1973). The intrinsic myenteric innervation of the hind-gut and accessory muscles of defaecation in the cat. Zeitschrift für Zellforschung Mikroskopische Anatomie, 136, 31–44.

186

INNERVATION OF THE GASTROINTESTINAL TRACT

Hoyle, C.H., Knight, G.E. and Burnstock, G. (1990). Suramin antagonises responses to P2-purinoceptor agonists and purinergic nerve stimulation in the guinea-pig urinary bladder and taenia coli. British Journal of Pharmacology, 99, 617–621. Hubel, K.A. (1985). Intestinal nerves and ion transport: stimuli, reflexes and responses. American Journal of Physiology, 248, G261–G271. Hulten, L. (1969). Extrinsic nervous control of colonic motility and blood flow. An experimental study in the cat. Acta Physiologica Scandinavica Supplementum, 335, 1–116. Kennedy, C. (1990). P1- and P2-purinoceptor subtypes-an update. Archives Internationales de Pharmacodynamie et de Therapie, 303, 30–50. Kennedy, C. and Krier, J. (1987a). (Met5)enkephalin acts via δ-opioid receptors to inhibit pelvic nerve-evoked contractions of cat distal colon. British Journal of Pharmacology, 92, 291–298. Kennedy, C. and Krier, J. (1987b). δ-Opioid receptors mediate inhibition of fast excitatory postsynaptic potentials in cat parasympathetic colonic ganglia. British Journal of Pharmacology, 92, 437–443. Lefebvre, R.A. and Burnstock, G. (1990). Effects of adenosine triphosphate and related purines in the rat gastric fundus. Archives Internationales de Pharmacodynamie et de Therapie, 303, 199–215. Leff, P., Wood, B.E. and O’Connor, S.E. (1990). Suramin is a slowly equilibrating but competitive antagonist of P2X-receptors in the rabbit isolated ear artery. British Journal of Pharmacology, 101, 645–649. Li, C.G. and Rand, M.J. (1990). Nitric oxide and vasoactive intestinal polypeptide mediate non-adrenergic, noncholinergic inhibitory transmission to smooth muscle of the rat gastric fundus. European Journal of Pharmacology, 191, 303–309. Manzini, S., Maggi, C.A. and Meli, A. (1985). Further evidence for involvement of adenosine-5’-triphosphate in non-adrenergic, non-cholinergic relaxation of the isolated rat duodenum. European Journal of Pharmacology, 113, 399–408. Manzini, S., Hoyle, C.H.V. and Burnstock, G. (1986). An electrophysiological analysis of the effect of reactive blue 2, a putative P2-purinoceptor antagonist, on inhibitory junction potentials of rat caecum. European Journal of Pharmacology, 127, 197–204. McRorie, J., Krier, J. and Adams, T. (1991). Morphology and projections of myenteric neurones to colonic fibre bundles of the cat. Journal of the Autonomic Nervous System, 32, 205–215. Moncada, S., Palmer, R.M. and Higgs, E.A. (1991). Nitric oxide: physiology, pathophysiology and pharmacology. Pharmacological Reviews, 43, 109–142. Nadelhaft, I., de Groat, W.C. and Morgan, C. (1980). Location and morphology of parasympathetic preganglionic neurones in the sacral spinal cord of the cat revealed by retrograde axonal transport of horseradish peroxidase. Journal of Comparative Neurology, 193, 265–281. Nadelhaft, I., Roppolo, J., Morgan, C. and de Groat, W.C. (1983). Parasympathetic preganglionic neurons and visceral primary afferents in monkey sacral spinal cord revealed following application of horseradish peroxidase to pelvic nerve. Journal of Comparative Neurology, 216, 36–52. O’Connor, S.E., Dainty, I.A. and Leff, P. (1991). Further subclassification of ATP receptors based on agonist studies. Trends in the Pharmacological Sciences, 12, 137–141. Sanders, K.M. and Smith, T.K. (1989). Electrophysiology of colonic smooth muscle. In Handbook of Physiology, Section 6, Vol. I, edited by J.D. Wood, pp. 251–271. Bethesda, MA: American Physiological Society. Sanders, K.M. and Ward, S.M. (1992). Nitric oxide as a mediator of nonadrenergic noncholinergic neurotransmission. American Journal of Physiology, 262, G379–G392. Smith, P.L. and McCabe, R.D. (1986). Potassium secretion by rabbit descending colon: effects of adrenergic stimuli. American Journal of Physiology, 250, G432–G439. Smith, T.K., Reed, J.B. and Sanders, K.M. (1987). Origin and propagation of electrical slow waves in circular muscle of canine proximal colon. American Journal of Physiology, 252, C215–C224. Starke, K., Gothert, M. and Kilbinger, H. (1989). Modulation of neurotransmitter release by presynaptic autoreceptors. Physiological Reviews, 69, 864–989. Szurszewski, J.H. and King, B.F. (1989). Physiology of prevertebral ganglia in mammals with special reference to inferior mesenteric ganglion. In Handbook of Physiology, Section 6, Vol. I, edited by J.D. Wood, pp. 465–518. Bethesda, Maryland: American Physiological Society. Timmermans, J.-P., Barbiers, M., Scheuermann, D.W., Bogers, J.J., Adriensen, D., Fekete, E., Mayer, B., Van Merck, E.A. and De Groodt-Lasseel, M.H. (1994). Nitric oxide synthase immunoreactivity in the enteric nervous system of the developing human digestive tract. Cell and Tissue Research, 275, 325–345. Venkova, K. and Krier, J. (1992). A nitric oxide mediated component of nonadrenergic, noncholinergic responses of cat distal colon longitudinal muscle (Abstract). Gastroenterology, 103, 1376. Venkova, K. and Krier, J. (1993). Stimulation of lumbar sympathetic nerves evokes contractions of cat colon circular muscle mediated by ATP and noradrenaline. British Journal of Pharmacology, 110, 1260–1270. Venkova, K. and Krier, J. (1994). A nitric oxide and prostaglandin-dependent component of NANC off-contractions in cat colon. American Journal of Physiology, 266, G40–G47.

NEURAL CONTROL OF THE LARGE INTESTINE

187

Venkova, K. and Krier, J. (1995). Postjunctional α1- and β-adrenoceptor effects of noradrenaline on electrical slow waves and phasic contractions of cat colon circular muscle. British Journal of Pharmacology, 116, 3265–3273. Venkova, K., Milne, A. and Krier, J. (1994). Contractions mediated by α1-adrenoceptors and P2-purinoceptors in a cat colon circular muscle. British Journal of Pharmacology, 112, 1237–1243. von Kügelgen, I., Bültmann, R. and Starke, K. (1990). Interaction of adenine nucleotides, UTP and suramin in mouse vas deferens: suramin-sensitive and suramin-insensitive components in the contractile effect of ATP. Naunyn-Schmiedebergs Archives of Pharmacology, 342, 198–205. Ward, S.M., Dalziel, H.H., Thornbury, K.D., Westfall, D.P. and Sanders, K.M. (1992). Nonadrenergic, noncholinergic inhibition and rebound excitation in canine colon depend on nitric oxide. American Journal of Physiology, 262, G237–G243. Wood, J.D. and Marsh, D.R. (1973). Effects of atropine, tetrodotoxin and lidocaine on rebound excitation of guinea-pig small intestine. Journal of Pharmacology and Experimental Therapeutics, 184, 590–598. Zhang, L., Keef, K.D., Bradley, M.E. and Buxton, I.L. (1992). Action of α2A-adrenergic receptors in circular smooth muscle of canine proximal colon. American Journal of Physiology, 262, G517–G524.

5 Neurons of the Gallbladder and Sphincter of Oddi Gary M. Mawe, Erin K. Talmage, Lee J. Jennings, Kirk Hillsley and Audra L. Kennedy Department of Anatomy and Neurobiology, C423 Health Science Complex, The University of Vermont, Burlington, VT 05405 USA Following a meal, a cascade of events is initiated that results in the delivery of bile from the gallbladder to the lumen of the duodenum. An important messenger in the activation of this response, cholecystokinin, acts through neural mechanisms to increase pressure in the gallbladder and decrease resistance in the sphincter of Oddi, thus facilitating the flow of bile. Only recently have the neurons residing in the gallbladder and the sphincter of Oddi been directly studied to establish their morphological, electrical and neurochemical properties; the types of inputs that they receive; and how they respond to these signals. This chapter is organized to provide an overview of (1) the basic properties of gallbladder and sphincter of Oddi neurons, (2) the inputs to these cells that have been investigated, and (3) the reflex interactions that exist between the gut and the gallbladder, the gut and the sphincter of Oddi, and the gallbladder and the sphincter of Oddi. Although the neurons that migrate to, and ultimately innervate, the gallbladder and the sphincter of Oddi share a common ancestry with the neurons of the gut tube, the neurobiology of each of these regions is uniquely adapted to carry out their particular functions. KEY WORDS: autonomic ganglia; motility; innervation; enteric nervous system; biliary tract; immunohistochemistry; electrophysiology.

INTRODUCTION Developmentally, the sphincter of Oddi (SO) and the gallbladder arise from enteric precursors. The gallbladder forms as a ventral outgrowth of the primitive bile duct, and the SO forms as a specialization at the junction between the bile duct and the duodenum. The neural precursors that colonize these accessory regions of the gastrointestinal tract are likely to be derived from the vagal crest cells which also give rise to myenteric and submucous plexuses of the gut tube (Fontaine, Le Lievre and Le Douarin, 1977). Despite their common ancestry, the neurons that develop in the walls of the gallbladder and SO ultimately express electrical, morphological and neurochemical phenotypes that are different from one another, and different from those of the neurons in the gut tube. 189

190

INNERVATION OF THE GASTROINTESTINAL TRACT

Over the past several years, efforts in this and other laboratories have concentrated on elucidating the properties of the neurons that directly innervate the gallbladder and the SO. This chapter includes a summary of the morphological, neurochemical, and electrophysiological findings that have been reported within the past few years, as well as a discussion of how the neurons of these regions are functionally linked with the neurons of the enteric nervous system.

GALLBLADDER GANGLIA STRUCTURAL AND ELECTRICAL PROPERTIES OF GALLBLADDER NEURONS Morphological properties of gallbladder ganglia and neurons The general layout of the neural plexuses of the gallbladder have been studied in many species, including human, rhesus monkey, pig, dog, cat, marmoset, guinea-pig, North American opossums (Didelphis virginiana and Monodelphis domestica), Australian brush tailed possum (Trichosurus vultecula), and mouse (Alexander, 1940; Burnett, Gairns and Bacsich, 1964; Sutherland, 1966; Sutherland, 1967; Kyösola, 1977; Cai and Gabella, 1983a; Mawe and Gershon, 1989; Bauer et al., 1991; Talmage and Mawe, 1993; Talmage et al., 1996). Each layer of the gallbladder is innervated by a plexus of nerve fibres; the most conspicuous of which is found in the serosal layer where a prominent ganglionated network is located just outside of the muscularis. Here, a matrix of small, irregularly shaped ganglia, is interconnected by bundles of unmyelinated axons. The nerve bundles of the ganglionated plexus are contiguous with nerve bundles that follow the numerous blood vessels that exist in this layer. The neural plexus of the muscularis is largely composed of nerve fibres that travel parallel to the direction of muscle bundles. This plexus is somewhat sparse in some species such as the guinea-pig, but rather prominent in others such as the dog and human. A dense network of nerve fibres and occasional small ganglia exists in the lamina propria of the gallbladder. Nerve fibres in the mucosal plexus may pass singly or in bundles, and they often are seen in close association with the epithelium. The ultrastructural properties of the ganglionated plexus of the gallbladder have been studied in the guinea-pig (Cornbrooks, Pouliot and Mawe, 1992). Ultrastructurally, the ganglia and interganglionic connectives of the gallbladder appear to be hybrids of the enteric and non-enteric autonomic systems. Like enteric ganglia, gallbladder ganglia are surrounded by a connective tissue sheath and a layer of basal lamina that is confined to the exterior of the ganglion. The ganglia are comprised of neurons, glial cells, and a compact neuropil that is usually displaced to the periphery of the ganglia. Collagen, basal laminae, and intercellular spaces are noticeably absent from the interior of gallbladder ganglia. The neuropil of gallbladder ganglia consists of glial processes, unmyelinated axons, and nerve terminals that contain clear spherical and dense core vesicles. Unmyelinated axons in interganglionic connectives of the gallbladder are individually ensheathed by Schwann cell processes, as in non-enteric peripheral nerve bundles. Individual gallbladder neurons have been stained by intracellular injection of horseradish peroxidase or neurobiotin in guinea-pig and opossum ganglia (Mawe, 1990; Bauer et al., 1991; Cornbrooks, Pouliot and Mawe, 1992). The shapes of these neurons are

NEURONS OF THE GALLBLADDER AND SPHINCTER OF ODDI

191

reminiscent of parasympathetic neurons, and are simpler than those of enteric neurons. Gallbladder neurons consist of a soma and one or two long processes, with no appreciable dendritic arborization. The long processes of these cells appear to be mainly confined to the ganglionated plexus. The processes pass from their ganglion of origin into interganglionic nerve bundles where they often project for some distance and terminate. Frequently these processes exhibit large varicosities as they traverse one or more adjacent ganglia. This indicates that the circuitry for interganglionic communication exists in the ganglionated plexus of the gallbladder. It is doubtful that the intrinsic neurons of the gallbladder play a role in controlling vascular tone within the organ since processes from gallbladder ganglion cells do not project to the para- or perivascular plexuses that are associated with the rich vascular bed that exists in the serosa of the guinea-pig gallbladder. Electrical properties of gallbladder neurons Gallbladder neurons have been studied with intracellular recording techniques in the guinea-pig, North American opossum and human. In their resting state, gallbladder neurons are rather inexcitable, as spontaneous activity is only rarely observed, and they do not generate repetitive action potentials throughout the duration of a prolonged depolarizing current pulse (Table 5.1, Figure 5.1; Mawe, 1990; Bauer et al., 1991; Hillsley, Jennings and Mawe, 1998). Guinea-pig and human gallbladder neurons typically fire only one action potential at the onset of a prolonged depolarizing current pulse, though occasionally up to four spikes are recorded. In the guinea-pig, the upstroke of the action potential is largely due to a tetrodotoxin (TTX)-sensitive Na+ conductance, but a Ca2+ spike can be revealed in the presence of TTX and tetraethylammonium (TEA). The after-hyperpolarization (AHP) of guinea-pig gallbladder neurons, which lasts about 170 ms, involves two sequential Ca2+-dependent K+ events (Mawe, 1990). The AHP has a reversal potential at the equilibrium potential for potassium, and is completely abolished in a Ca2+-free Krebs solution. The early phase of the AHP is attenuated in the presence of TEA, and is likely to result from the activation of large-conductance Ca2+-activated potassium channels (BK channels), whereas the late phase of the AHP is diminished in the presence of apamin and is likely to involve small-conductance Ca2+-activated K+ channels (SK channels) (Figure 5.1B; Mawe, 1990). The late phase of the AHP apparently contributes to the rapid adaptation of TABLE 5.1 Electrical properties of guinea-pig gallbladder and sphincter of Oddi neurons. Data from Mawe, 1990; Wells and Mawe, 1993. Gallbladder

Excitability Membrane potential Input resistance AHP duration Morphology

Sphincter of Oddi

Gallbladder cell

Tonic cell

Phasic cell

AH cell

Low ≈−50 mV ≈80 MΩ ≈180 Ms Parasympathetic

High ≈−50 mV ≈70 MΩ ≈20 Ms Dogiel I

Low ≈−50 mV ≈70 MΩ ≈20 Ms Dogiel I

Low ≈−65 mV ≈45 MΩ >2 s Dogiel II

192

INNERVATION OF THE GASTROINTESTINAL TRACT

Figure 5.1 Active electrical properties of guinea-pig gallbladder neurons. A. Four consecutive overlapping traces showing the response of a neuron to prolonged depolarizing current pulses (250 ms; 0.2 nA). As is typical of these cells, this neuron generated only one action potential at the onset of the current pulse. B. Response of a neuron to a brief current pulse illustrating a typical AHP, which is composed of an early and a late phase that are sensitive to tetraethyl ammonium and apamin, respectively. C. Elimination of the late phase of the AHP by apamin (100 nM) converts the neuron from an adaptive to a non-adaptive state, thus allowing for the generation of repetitive action potentials. (From Mawe et al., 1997).

guinea-pig gallbladder neurons since suppression of this component of the AHP with apamin causes the cells to fire action potentials repetitively throughout the duration of a depolarizing current pulse (Figure 5.1C; Mawe, 1990). The neurons of the North American opossum gallbladder are classified into two groups (Bauer et al., 1991), adaptive neurons which respond to intracellular current pulses with a short burst of action potentials, and rapidly adaptive neurons that respond to current pulses with a single action potential. The adaptive cells are more numerous, comprising about 70% of the population. Action potentials of neurons in these ganglia are TTX-sensitive, and are followed by a brief AHP lasting about 30 ms. Synaptic inputs to guinea-pig gallbladder neurons Synaptic input to gallbladder neurons, in response to stimulation of interganglionic fibre bundles, has been studied in the guinea-pig and the opossum (Mawe, 1990; Bauer et al.,

NEURONS OF THE GALLBLADDER AND SPHINCTER OF ODDI

193

1991; Hillsley, Jennings and Mawe, 1998). Both fast and slow excitatory postsynaptic potentials (EPSPs) can be elicited, but inhibitory synaptic potentials have not been reported. Fast EPSPs, which can be activated by low frequency (0.5 Hz) stimulation of interganglionic fibre tracts, are sensitive to hexamethonium and are reversibly eliminated when the tissue is bathed in a Ca2+-free/high Mg2+ Krebs solution. Although interganglionic communication in the form of nicotinic EPSPs is likely to exist in these ganglia, the major source of nicotinic input to gallbladder neurons is from vagal preganglionic nerve fibres (see section on Vagal efferent fibres provide the driving force for gallbladder neurons; Mawe, Gokin and Wells, 1994). Slow EPSPs can be elicited by high frequency fibre tract stimulation (10–20 Hz) in about 30% of the neurons in guinea-pig (Mawe, 1990), and 20% of the neurons in opossum (Bauer et al., 1991). In the guinea-pig, the most thoroughly studied species, these slow synaptic events are Ca2+-dependent, and are associated with a decrease in input resistance. The slow EPSP is likely to involve an activation of a non-selective cation conductance since its amplitude decreases as cells are depolarized and increases when the membrane is hyperpolarized, with an estimated reversal potential near 0 mV. Recent evidence suggests that tachykinins and calcitonin gene-related peptide (CGRP) released from extrinsic afferent fibres mediate slow EPSPs in gallbladder ganglia (see section on Extrinsic sensory fibres; Mawe, 1995).

REGULATORY INPUTS TO GALLBLADDER GANGLIA Vagal efferent fibres provide the driving force for gallbladder neurons As described above, all gallbladder neurons exhibit nicotinic fast EPSPs in response to stimulation of interganglionic fibre tracts (Mawe, 1990; Bauer et al., 1991; Hillsley, Jennings and Mawe, 1998). Since gallbladder ganglia are developmentally associated with enteric ganglia, and since synaptic inputs to enteric ganglia are mainly intrinsic (Furness and Costa, 1987; Wood, 1989; Costa et al., 1992), it could not be presumed that nicotinic inputs to gallbladder neurons are from preganglionic neurons. However, it does appear that all gallbladder neurons do receive nicotinic input from the vagal preganglionic fibres. Several lines of evidence verify a significant vagal input to gallbladder ganglia. Stimulation of the vagus nerves causes gallbladder contraction, or, in the presence of atropine, relaxation (Ryan, 1987; Dodds, Hogan and Geenen, 1989; Dahlstrand, 1990). Neurons in the dorsal motor nucleus of the vagus have been retrogradely labelled from the gallbladder (Mawe, 1990). Also, nerve fibres that are immunoreactive for the biosynthetic enzyme of acetylcholine, choline acetyltransferase (ChAT), are present in ganglia, surrounding gallbladder neurons, and in the paravascular plexus, indicating that extrinsic cholinergic nerve fibres enter the gallbladder. Data from a study involving stimulation of extrinsic inputs to the gallbladder in normal and vagotomized guinea-pigs indicate that all gallbladder neurons receive vagal input, and that the vagal input is the major source of fast synaptic input to gallbladder neurons (Mawe, Gokin and Wells, 1994). Vagotomy results in the complete elimination of extrinsic nicotinic inputs, which are normally received by all gallbladder neurons. Therefore, unlike the ganglion cells in the bowel, all gallbladder neurons receive direct input

194

INNERVATION OF THE GASTROINTESTINAL TRACT

from the central nervous system. Following vagotomy, fast EPSPs could be elicited in some neurons by stimulation of interganglionic fibre tracts, indicating that interganglionic neural projections may exist amongst gallbladder ganglia. Sympathetic postganglionic fibres In addition to receiving input from vagal preganglionic fibres, intramural ganglia are the target of sympathetic postganglionic inputs to the gallbladder. The ganglionated plexus of the gallbladder is rich in nerve fibres that express catecholamine histofluorescence (Cai and Gabella, 1983a, 1984; Mawe and Gershon, 1989), as well as immunoreactivities for tyrosine hydroxylase and dopamine β-hydroxylase (Mawe and Gershon, 1989). It is likely that these nerve fibres arise in the celiac ganglia because celiac neurons are labelled following injection of retrograde axonal tracers in the wall of the gallbladder (Mawe and Gershon, 1989). We have established that noradrenaline can be released from catecholaminergic nerve fibres and that it acts to attenuate the release of acetylcholine from vagal terminals in gallbladder ganglia (Mawe, 1993; Mawe, Gokin and Wells, 1994). Exogenously applied noradrenaline decreases the amplitude of fast EPSPs in a concentration-dependent manner by acting on α2-adrenoceptors. The action of noradrenaline is mimicked by the α2-adrenoceptor agonist, clonidine, and is antagonized by the α2-adrenoceptor antagonist, yohimbine. When endogenous catecholamines are released from sympathetic nerves by tyramine application or by electrical stimulation of the vascular plexus, a yohimbine-sensitive decrease in fast synaptic activity is observed. Since it has been demonstrated that the major source of fast synaptic input to gallbladder neurons is vagal preganglionic fibres, and that fast EPSPs elicited by cystic nerve stimulation are sensitive to noradrenaline (Mawe, Gokin and Wells, 1994), it is likely that vagal input to the gallbladder can be attenuated by sympathetic activity. Therefore, the decrease in gallbladder tone, that can be elicited by stimulation of the splanchnic nerves (Pallin and Skoglund, 1964; Persson, 1971, 1972, 1973; Yamasato and Nakayama, 1990), may be the result of a presynaptic inhibitory effect of sympathetic nerves on the vagal terminals in gallbladder ganglia. Yamasato and Nakayama (1990) reported that subthreshold stimulation of the celiac nerve in the dog, which had no effect on gallbladder motility, markedly decreased the responsiveness of the gallbladder to vagal stimulation. It is possible that this inhibitory influence of sympathetic input on the vagal tone facilitates gallbladder filling. Furthermore, administration of sympathomimetics for treatment of low blood pressure could have the side effect of decreasing gallbladder motility by decreasing vagal output to gallbladder ganglia, thus leading to gallbladder stasis. Extrinsic sensory fibres The concept that sensory neurons can release neuroactive compounds from their peripheral processes, and initiate local reflex activity, has now been established in several organ systems, including the gallbladder. In the gallbladder, nerve fibres that are immunoreactive for substance P (SP) and CGRP are abundant in the ganglionated and vascular plexuses (Goehler, Sternini and Brecha, 1988; Maggi et al., 1989; Mawe and Gershon,

NEURONS OF THE GALLBLADDER AND SPHINCTER OF ODDI

195

1989). Furthermore, application of capsaicin to gallbladder strips causes contraction and the release of SP and CGRP (Maggi et al., 1989). Since gallbladder neurons do not express CGRP-immunoreactivity (Goehler, Sternini and Brecha, 1988; Mawe and Gershon, 1989; Talmage et al., 1992, 1996), the SP/CGRP-immunoreactive nerve fibres must be extrinsic and are likely to arise from thoracic sensory ganglion cells which project to the gallbladder (Mawe and Gershon, 1989). Application of tachykinins, CGRP or capsaicin to guinea-pig gallbladder neurons causes a depolarization that is similar to the slow EPSP recorded in these preparations (Mawe, 1995; Gokin, Jennings and Mawe, 1996). Both tachykinins and CGRP depolarize gallbladder neurons by activating a non-selective cation conductance, and increase the excitability of these cells. The tachykinin-mediated depolarization of gallbladder neurons has been studied with receptor-specific agonists and antagonists (Mawe, 1995). Naturally occurring tachykinins depolarize gallbladder neurons with a rank order potency that is characteristic of neurokinin-3 (NK3) receptors (neurokinin B > neurokinin A > SP). Consistent with this finding, the NK3 receptor agonist, senktide, was more potent than any of the naturally occurring tachykinins in eliciting a depolarization. Once the involvement of NK3 receptors was established, [Trp7,β-Ala8 ]-NKA(4–10), which acts as an antagonist at NK3 receptors, was used in the preparation. This compound shifted the concentration-effect curve for SP to the right, and depressed both capsaicin-induced depolarizations and stimulus-evoked slow EPSPs. Therefore, sensory fibres are likely to contribute to slow EPSPs in gallbladder ganglia by releasing tachykinins, and probably CGRP, in an axon reflex response. It is possible that such a response may contribute to gallbladder inflammation since release of neuroactive peptides from extrinsic sensory fibres is one of the initial steps in toxin A-induced inflammation in the bowel (Mantyh et al., 1996). Hormonal cholecystokinin Cholecystokinin (CCK) has long been recognized for its ability to cause gallbladder contractions (Ivy and Oldberg, 1928). However, the issue of whether CCK elicits its physiological effect through a direct action on gallbladder smooth muscle or by facilitating the neural output to the organ has been debated. Over the past 10 years, studies from several different laboratories provide strong support for the concept that, physiologically, CCK acts through neural mechanisms to cause gallbladder emptying. Studies of in vivo preparations have provided support for the view that the principle physiological effect of CCK in the gallbladder involves a neural mechanism (Grossman, 1975; Behar and Biancani, 1980, 1987; Takahashi et al., 1982; Gullo et al., 1984; Fisher, Rock and Malmud, 1985; Marzio et al., 1985; Strah et al., 1985, 1986; Pozo, Salida and Madrid, 1989; Hanyu et al., 1990b; Takahashi, May and Owyang, 1991). These studies demonstrate that gallbladder contractions elicited by feeding, or by intravascular injections of physiological post-feeding concentrations of CCK, were disrupted by atropine, hexamethonium, TTX, or vagal blockade. Similar results have been reported in an in vitro preparation by Brotschi, Pattavino and Williams, (1990). Furthermore, when applied to gallbladder muscle strips, CCK causes the release of ACh (Yau and Youther, 1984; Yamamura et al., 1986; Rakovska, Milenov and Bocheva, 1989; Galligan and Bertrand,

196

INNERVATION OF THE GASTROINTESTINAL TRACT

1994). Taken together, these data indicate that it is likely that feeding and CCK-induced physiological contraction of the gallbladder occurs through a neural mechanism rather than a direct action of CCK on smooth muscle. It is clear that CCK receptors do exist on gallbladder smooth muscle cells, but these receptors may not normally play a role in gallbladder emptying. It is plausible that the neural effect that was described was a result of actions of CCK on vagal afferent fibres and/or vagal terminals within gallbladder ganglia since hexamethonium and vagal blockade disrupt the meal- and CCK-induced gallbladder contraction (Hanyu et al., 1990b; Takahashi, May and Owyang, 1991). Electrophysiological studies of gallbladder whole mount preparations support this view. Direct studies of the actions of CCK in gallbladder ganglia have been conducted in the guinea-pig (Mawe, 1991; Mawe, Gokin and Wells, 1994) and in the opossum (Bauer et al., 1991) using intracellular electrophysiological recording techniques. In both of these species, CCK has a profound presynaptic facilitory effect on ganglionic transmission, but does not have a direct effect on the gallbladder neurons. Upon application of CCK, the amplitude of cholinergic fast EPSPs is increased, usually converting subthreshold EPSPs to suprathreshold EPSPs (Bauer et al., 1991; Mawe, 1991). CCK increases the quantal content three-fold without altering quantal size, but does not alter the sensitivity of these neurons to exogenously applied acetylcholine, indicating that CCK acts through a presynaptic mechanism (Mawe, 1991). Most importantly, it has been shown that CCK is quite potent in its ability to promote the release of acetylcholine. The concentration-effect relationship for CCK in gallbladder ganglia peaks at 1.0 nM, and has a half-maximal effective concentration (EC50) of 33 pM; the EC50 for the direct contractile effect of CCK on gallbladder muscle is 10 nM. In the presence of 10 pM CCK, which is within the range of postprandial serum levels of CCK (Takahashi, May and Owyang, 1991), the peptide increases synaptic currents by about 20%. The nerve terminals that are sensitive to CCK are from the vagus nerve since synaptic responses to cystic nerve stimulation are sensitive to CCK, and since these inputs are eliminated following vagotomy (Mawe, Gokin and Wells, 1994). CHEMICAL CODING OF GALLBLADDER NEURONS Despite the cumulative knowledge that has been gained from motility and ganglion electrophysiological studies, it is not entirely clear what compounds are actually released from gallbladder neurons to modulate smooth muscle function. It is recognized, from muscle strip studies involving electric field stimulation, that excitatory transmitters are released to promote gallbladder emptying, and it is possible that inhibitory transmitters are released to promote gallbladder filling. Two major theories have been proposed to explain how the gallbladder fills. One theory suggests that the gallbladder undergoes a “passive” filling between meals (Ryan, 1987; Shaffer, 1991), and the other theory suggests that the gallbladder actively expands to draw hepatic bile into its lumen, much as a bellows draws air in as it is expanded (Lanzini, Jazrawi and Northfield, 1987; Jazrawi et al., 1995). In order for the gallbladder to act as a bellows, a potent inhibitory output from the ganglia of the gallbladder would be necessary to induce an active relaxation. An important step in determining how gallbladder neurons could influence gallbladder function was to identify the neurochemical phenotypes of gallbladder neurons. We and

NEURONS OF THE GALLBLADDER AND SPHINCTER OF ODDI

197

others have studied the chemical coding of gallbladder neurons in several species, including guinea-pig, human, dog and opossum, and have identified sets of neuroactive compounds that are co-expressed. Characteristic trends, as well as interspecies differences, have emerged from these investigations. For example, all gallbladder neurons are likely to be cholinergic since all detectable neurons express immunoreactivity for the essential biosynthetic enzyme for acetylcholine, choline acetyltransferase (Talmage et al., 1996). Furthermore, gallbladder neurons can express immunoreactivities for tachykinins, vasoactive intestinal polypeptide (VIP), nitric oxide synthase (NOS) and either neuropeptide Y (NPY) (guinea-pig and human) or galanin (dog and opossum) (Keast, Furness and Costa, 1985; Talmage et al., 1992, 1996; De Giorgio et al., 1995). Species specific patterns of neuropeptide co-expression have been identified in neurons of the human, canine and guinea-pig gallbladder (Talmage et al., 1992, 1996; De Giorgio et al., 1995). These are schematically represented in Figure 5.2. In the human, the vast majority of neurons express VIP and NPY, and most of these neurons express SP as well (De Giorgio et al., 1995; Talmage et al., 1996). In the dog, where NPY appears to be replaced by galanin, most neurons express VIP and galanin, and most of these express SP as well (Talmage et al., 1996). In the guinea-pig, most neurons express SP and NPY, and a distinct subpopulation of neurons expresses VIP (Talmage et al., 1992). It is clear that gallbladder ganglia comprise neurons that express the enzymatic machinery to synthesize nitric oxide. In all species that have been studied, including the human (Talmage and Mawe, 1993; De Giorgio et al., 1995), monkey (De Giorgio et al., 1994), dog (Talmage and Mawe, 1993), opossum (Talmage and Mawe, 1993), guinea-pig (Talmage and Mawe, 1993; Grozdanovic et al., 1994; Siou, Belai and Burnstock, 1994), gerbil (Talmage and Mawe, 1993) and mouse (Grozdanovic, Baumgarten and Bruning, 1992), neurons in the gallbladder express NOS and/or NADPH-diaphorase (NADPH-DA) activity. In guinea-pig there is a direct correlation between NOS-immunoreactivity and NADPH-DA activity in gallbladder ganglia (Mawe et al., 1997). As mentioned above, all gallbladder neurons express ChAT; coexpression of ChAT and NOS is also observed in these ganglia (Figure 5.3). The physiological role of neurally released nitric oxide in the gallbladder, if there is one, is unclear at this time. However, nitric oxide can relax the gallbladder (Mourelle et al., 1993a; McKirdy, McKirdy and Johnson, 1994) and may modulate the CGRP-induced relaxation of the gallbladder (Kline and Pang, 1994).

Figure 5.2 Chemical coding in neurons in ganglia of the human, dog and guinea-pig gallbladder. Note the coexpression of neuroactive compounds that have excitatory (+) and inhibitory (–) effects on gallbladder smooth muscle. ACh, acetylcholine; VIP, vasoactive intestinal peptide; NPY, neuropeptide Y; SP, substance P; NOS, nitric oxide synthase. (Modified from Talmage et al., 1996).

198

INNERVATION OF THE GASTROINTESTINAL TRACT

Figure 5.3 Immunoreactivities for ChAT and NOS in ganglia of the guinea-pig gallbladder and sphincter of Oddi. Immunoreactivities in these double stained preparations illustrate the co-expression of ChAT and NOS in a subset of gallbladder neurons, and the exclusive expression of ChAT or NOS by sphincter of Oddi neurons. Note that in the micrographs of the SO ganglion, the NOS-immunoreactive neurons are located in the spaces that are devoid of ChAT immunoreactivity.

Unlike the ganglia of the bowel, ganglia of the gallbladder lack CGRP-immunoreactive neurons. Immunohistochemical studies in the human (De Giorgio et al., 1995; Talmage et al., 1996), dog (Talmage et al., 1996), pig (Sand, Tainio and Nordback, 1993), guineapig (Goehler, Sternini and Brecha, 1988; Mawe and Gershon, 1989; Talmage et al., 1992), Australian possum (Padbury, 1990), Monodelphis domesticus opossum (Talmage et al., 1996), and toad (Davies and Campbell, 1994) failed to identify neurons in the gallbladder wall that expressed immunoreactivity for CGRP. However, CGRP-immunoreactive nerve fibres have been described in the gallbladders of all of these species, where they are most abundant in the ganglionated plexus and in association with blood vessels, primarily in the paravascular plexus. In the human, dog, opossum, Australian possum, guinea-pig, and toad gallbladders these fibres have been shown to co-express SP-immunoreactivity, and probably originate in sensory ganglia (Goehler, Sternini and Brecha, 1988; Mawe and Gershon, 1989; Talmage et al., 1996). It is quite possible that CGRP can be released from these fibres as part of an axon reflex circuit (see section on Extrinsic sensory fibres of gallbladder ganglia), and exert influences on neurons, muscle and/or epithelial cells. CGRP causes a relaxation of gallbladder muscle (Kline and Pang, 1992) that is due to an activation of an ATP-dependent K+ conductance (Zhang et al., 1994). As described above, in guinea-pig gallbladder ganglia, CGRP causes a depolarization of neurons that is quite similar to the slow excitatory postsynaptic potential that occurs in these ganglia (Gokin, Jennings and Mawe, 1996).

NEURONS OF THE GALLBLADDER AND SPHINCTER OF ODDI

199

Together with the knowledge that all of the gallbladder neurons are cholinergic, these patterns of expression of neuroactive compounds are somewhat perplexing; as they suggest the coexpression of excitatory and inhibitory transmitters in single neurons of all of these species. In muscle strip assays, acetylcholine (Ryan, 1987), SP (Meldrum, Bojarski and Calam, 1987; Dahlstrand et al., 1988; Guo et al., 1989; Maggi et al., 1989) and NPY (Lillemoe, Webb and Pitt, 1988) have been shown to increase gallbladder tone, whereas VIP (Ryan and Cohen, 1977; Ryan and Ryave, 1978; Dahlstrand, Dahlström and Ahlman, 1989) can relax the precontracted gallbladder. The ostensibly opposing outputs from individual gallbladder neurons that express excitatory and inhibitory compounds could be explained by the following theoretical scenarios: (1) excitatory and inhibitory neuroactive compounds are separately released from a given neuron in response to distinct inputs; (2) compounds with opposing actions are released onto the same target sequentially, with one acting as a physiological antagonist of the other; or (3) both sets of compounds are co-released, but act on different targets. For example, if acetylcholine, SP and VIP were released together, acetylcholine and SP may act on the muscle to elicit a contraction, while VIP may act on epithelial cells. Neuroactive compounds may also act on adjacent nerve terminals to modulate further release. Further studies will be required to test these models.

SPHINCTER OF ODDI GANGLIA To date, many studies that have involved the direct study of SO ganglia have involved the guinea-pig model. Therefore, most of the following information is related to ganglia of the guinea-pig SO region. This is important to note because considerable interspecies variation exists with regard to the structure and function of the SO (see Mawe, 1992). In the guinea-pig, this area includes the terminal portion of the common bile duct (choledochal sphincter), an ampulla, and a terminal papilla (Dahlstrand et al., 1989; Hirose and Ito, 1991; Mawe, 1992). The studies that are reviewed here relate to neurons located throughout the anatomically defined regions of the ampulla and the choledochal sphincter. No regional differences have been observed in the electrical or morphological characteristics of neurons in these areas. STRUCTURAL AND ELECTRICAL PROPERTIES OF SPHINCTER OF ODDI NEURONS Morphological properties of sphincter of Oddi ganglia and neurons The ganglionated plexus of the guinea-pig SO includes a network of ganglia and interganglionic fibre bundles that is similar in appearance to the myenteric plexus of the small intestine. When stretched and pinned out flat, the SO region encompasses an area of approximately 20–25 mm2 containing an average of about 60 ganglia, and a total of about 2000 neurons (Wells and Mawe, 1993). The structure of individual SO neurons has been determined by injecting neurobiotin into individual neurons, from intracellular recording electrodes, and labelling the neurobiotin

200

INNERVATION OF THE GASTROINTESTINAL TRACT

with fluorophore- or horseradish peroxidase-conjugated avidin (Wells and Mawe, 1993). Guinea-pig SO neurons conform to the morphological classification scheme described by Dogiel for myenteric neurons (see Furness and Costa, 1987). The neurons of the SO that were classified as both tonic and phasic cells (see section on Electrophysiological properties of gallbladder neurons), which together represent the majority of SO neurons, have the shape of Dogiel type I cells. These neurons typically have a single long process and several very short processes emanating from the soma. In most cells, the short processes were flattened and “club-shaped”, and the remainder of the neurons had several thin processes that were approximately the length of the somal diameter. The long process of most cells, thought to be the axon, usually projects into adjacent ganglia. These processes frequently end as expansion bulbs, indicating that they may have been severed during the dissection. Processes of some cells were followed to strands of circular muscle that had not been removed during the tissue preparation. A small proportion of the neurobiotin-filled SO neurons, which are classified electrophysiologically as AH cells, are shaped like Dogiel type II cells. These neurons have several long branched processes that terminated both within the ganglionated plexus as well as within the muscularis. Calbindin-immunoreactive neurons in SO ganglia have a shape similar to that of the AH cells. Electrical properties of sphincter of Oddi neurons Unlike the guinea-pig gallbladder, where only one electrically classified type of neuron exists, three classes of neurons have been identified in the guinea-pig SO, on the basis of their active and passive electrical properties as determined by intracellular voltage recordings (Table 5.1, Figure 5.4; Wells and Mawe, 1993). The two most prevalent cell types, which together comprise about 95% of the total neuronal population, are called tonic and phasic cells. Tonic cells are very excitable; they generate spontaneous action potentials and elicit action potentials continuously throughout a prolonged suprathreshold depolarizing current pulse. The tonic cells of the SO are comparable to the S/type 1 cells of the small intestine (Wood, 1989) and rectum (Tamura and Wood, 1989), and the gastric Type I (Schemann and Wood, 1989) and colonic type 1 cells (Wade and Wood, 1988). Phasic cells have passive membrane characteristics that are similar to tonic cells, but differ in their active properties. Phasic cells do not exhibit spontaneous activity, and when directly

Figure 5.4 Typical responses of sphincter of Oddi tonic and phasic cells to a prolonged depolarizing current pulse. Tonic cells, which often exhibit spontaneous activity, generate action potentials throughout the duration of a depolarizing current pulse. Phasic cells, which are not spontaneously active, generate action potentials only at the onset of the current pulse, regardless of its amplitude or duration.

NEURONS OF THE GALLBLADDER AND SPHINCTER OF ODDI

201

stimulated, they adapt rapidly, typically firing only 1–3 action potentials at the onset of a prolonged depolarizing current pulse. This type of neuron has not been reported in the duodenum or small intestine; however neurons with similar properties have been described in the stomach (gastric type II; Schemann and Wood, 1989; Tack and Wood, 1992) and the distal colon (colonic type 4; Wade and Wood, 1988). The remainder of SO neurons, comprising about 5% of the population, are called AH cells. These cells have properties identical to those of the Type 2/AH cells of the bowel (Wood, 1989). Action potentials in these cells, which are TTX-insensitive, have a broadened shoulder during the repolarization, and the AHP of these cells lasts several seconds. Synaptic inputs to sphincter of Oddi neurons When interganglionic fibre bundles are stimulated, three different types of synaptic potentials can be elicited: fast EPSPs, slow EPSPs and inhibitory postsynaptic potentials (IPSPs). The fast excitatory input, which can be activated by low frequency stimulation, is present in most tonic and phasic cells, and some AH cells. These events are reversibly eliminated when the tissue is bathed in a Ca2+-free/high Mg2+ Krebs solution, and they are sensitive to hexamethonium. Potential sources of fast synaptic input to SO neurons include other SO neurons, neurons of the duodenum, and vagal preganglionic fibres. In some SO neurons, membrane depolarizations and hyperpolarizations can be activated by trains of high-frequency (10–20 Hz, 0.5 ms pulse) fibre tract stimulation (Wells and Mawe, 1993). These responses are eliminated by exposure to Ca2+-free Krebs, and therefore are presumed to be slow EPSPs and IPSPs. The sources and ionic mechanisms of slow excitatory inputs have not been evaluated, but SO neurons could receive modulatory input from other SO neurons, duodenal neurons, and/or extrinsic sensory fibres. Inhibitory synaptic inputs to SO neurons probably involves the activation of a K+ conductance, and these inputs are provided by sympathetic postganglionic input from the coeliac ganglion (see below). REGULATORY INPUTS TO SPHINCTER OF ODDI GANGLIA As indicated above, not as much is known about the neural control of the SO relative to other regions of the gut, including the gallbladder. Of the potential physiological modulators of SO ganglion function, two that have been studied directly in SO ganglia are sympathetic postganglionic inputs and CCK. Sympathetic inputs to sphincter of Oddi ganglia The action of sympathetic input to this region appears to increase SO tone. Activation of the sympathetic input to the feline SO, in vivo, by stimulating the splanchnic nerves or postganglionic nerve bundles, increases the tone of the sphincter musculature (Persson, 1972, 1973), whereas in non-sphincter regions of the gut, sympathetic input results in decreased motility (see Furness and Costa, 1987). Although the muscularis of the SO does contain catecholamine-histofluorescent nerve fibres, the area of most dense catecholaminergic innervation lies within the intrinsic

202

INNERVATION OF THE GASTROINTESTINAL TRACT

ganglionated neuronal plexus (Baumgarten and Lange, 1969; Mori, Azuma and Fujiwara, 1971; Cai and Gabella, 1983b). In the ganglionated plexus of the SO, immunoreactivities for tyrosine hydroxylase and dopamine β-hydroxylase are abundant, and coexistent in nerve fibres, but these synthetic enzymes are not expressed by SO neurons (Wells and Mawe, 1994). The actions of noradrenaline and endogenous catecholamine release have been studied in the guinea-pig SO (Wells and Mawe, 1994). Exogenous noradrenaline reversibly decreases the amplitude of fast EPSPs evoked by stimulation of interganglionic fibre tracts. Since noradrenaline does not alter the responsiveness of the neurons to acetylcholine, the action of noradrenaline in SO ganglia is thought to be presynaptic. The α2-adrenoceptor agonist, UK14304, mimics the noradrenaline-induced effect, and this effect is attenuated by the α2-adrenoceptor antagonist, Idazoxan. Agonists of α1- and β-adrenoceptor have no detectable effect in SO ganglia. Results of experiments involving application of exogenous noradrenaline are supported by the finding that release of endogenous catecholamines, by tyramine, causes an idazoxan-sensitive decrease in the fast EPSP. In SO neurons that exhibit IPSPs, noradrenaline causes hyperpolarization of the membrane potential. The IPSP and the noradrenaline-induced hyperpolarization have reversal potentials near the K+ equilibrium potential, and are inhibited by α2-adrenoceptor antagonists. Further evidence for the concept that sympathetic nerves mediate IPSPs is that desipramine, a catecholamine reuptake inhibitor, reversibly increases the amplitude of the IPSP. The findings from morphological and electrophysiological studies indicate that sympathetic nerves target SO ganglia, and that within these ganglia, noradrenaline can act both pre- and postsynaptically as an inhibitory neurotransmitter. Therefore, ganglia of the SO can be added to the growing list of enteric and parasympathetic ganglia that receive interganglionic input from sympathetic postganglionic nerves. Although noradrenaline has actions in SO ganglia that are similar to actions in other gastrointestinal ganglia, the net effect of sympathetic input to the sphincter is contraction rather than relaxation. The activation of divergent net effects through common mechanisms supports the view that precise circuitry exists within ganglia in various regions of the gut. Sympathetic pre- and postsynaptic activity is likely to have the net effect of decreasing the release of inhibitory substances and/or increasing the release of excitatory neuroactive compounds. Since the effects of noradrenaline are inhibitory it is likely that it decreases the release of inhibitory compounds. It is not clear whether the IPSPs occur selectively on inhibitory neurons of the SO, but this type of circuitry could contribute to the increased SO tone that results from sympathetic nerve stimulation. Actions of CCK on sphincter of Oddi neurons At the level of the SO, the stimulatory effect of CCK on bile flow appears to involve a neural mechanism. Several studies, in various species, have shown that the actions of CCK on the SO can be attenuated by muscarinic blockade with atropine and/or complete neural blockade with TTX (Behar and Biancani, 1987; Vogalis, Bywater and Taylor, 1989; Hanyu et al., 1990a). Furthermore, CCK-induced release of neuroactive compounds such as acetylcholine (Harada, Katsuragi and Furukawa, 1986), VIP (Wiley, O’Dorisio

NEURONS OF THE GALLBLADDER AND SPHINCTER OF ODDI

203

and Owyang, 1988; Dahlstrand et al., 1990) and nitric oxide (Pauletzki et al., 1993) have been reported. The actions of CCK on guinea-pig SO neurons have been investigated by application of CCK while recording from SO neurons with intracellular recording electrodes (Gokin, Hillsley and Mawe, 1997). CCK has a direct, excitatory effect on tonic and phasic cells in SO ganglia. The depolarization that is elicited by CCK likely involves the activation of a non-selective cation conductance since it reverses near 0 mV, and it is significantly reduced in a low Na+ solution. The CCK-induced depolarization of tonic and phasic cells involves the CCK-A receptor subtype; it is significantly reduced by the CCK-A antagonist, PD 140,548 N-methyl-D-glucamine, but not by the CCK-B receptor antagonist, PD 135,158 N-methyl-D-glucamine. Although SO neurons express CCK receptors, and CCK can act on these receptors to depolarize SO neurons, it is disputable whether hormonal CCK could act physiologically at this site to alter SO tone. This is because of the discrepancy between the sensitivity of SO neurons for CCK and the peak concentrations of CCK in the serum following a meal. The lowest concentration of CCK that results in a detectable response in SO neurons is 1 nM, and the EC50 in these neurons is 10 nM. Since peak serum concentrations of CCK following a meal are in the low picomolar range, hormonal CCK would be delivered at subthreshold concentrations. It is possible that CCK could have a paracrine effect on SO neurons due to the close proximity of the SO to the duodenum, where CCK is released. Alternatively, CCK may alter SO tone through vagal and/or duodenum-SO neural circuits (see section NEURAL INTERACTION BETWEEN THE GUT AND THE SPHINCTER OF ODDI). CHEMICAL CODING OF SPHINCTER OF ODDI NEURONS To identify potential chemical coding in sphincter of Oddi neurons, immunohistochemistry and histochemistry have been employed to test for the presence of putative neurotransmitters and related synthetic enzymes. In the guinea-pig SO, the pattern of chemical coding is less ambiguous than that found in the gallbladder; there is a distinction between the expression of excitatory and inhibitory compounds in SO neurons (Figure 5.3). In the guinea-pig SO, about 70% of the neurons are immunoreactive for ChAT, indicating that they are likely to synthesize acetylcholine and most of these neurons are also immunoreactive for SP and enkephalin (Wells, Talmage and Mawe, 1995; Talmage et al., 1997). Since acetylcholine, SP and opiates increase SO tone (Azuma and Fujiwara, 1973; Kudoh et al., 1981, Behar and Biancani, 1984; Dahlstrand et al., 1985, 1988), and since nerve fibres that are immunoreactive for SP and ChAT are abundant in the muscularis of the SO (Talmage et al., 1997), it is likely that the SP/enkephalin/ChAT-immunoreactive population includes excitatory motor neurons. A distinct population of guinea-pig SO neurons express NOS, and some of the NOS-positive neurons express VIP, whereas others express NPY. The NOS/VIP and NOS/NPY neurons are likely to be inhibitory neurons because both nitric oxide (Allescher et al., 1992; Kaufman et al., 1993; Mourelle et al., 1993b; Pauletzki et al., 1993) and VIP (Behar and Biancani, 1980; Wiley, O’Dorisio and Owyang, 1988; Dahlstrand et al., 1990) have been shown to decrease SO tone. To determine whether a correlation exists between the electrical and chemical coding properties of SO neurons, we conducted a study combining intracellular recording and dye

204

INNERVATION OF THE GASTROINTESTINAL TRACT

injection, along with chemical coding. Results of this study clearly demonstrated that there is no relationship between the electrical properties of SO neurons (Tonic versus Phasic) and their chemical coding patterns (cholinergic versus nitrergic) (Hillsley and Mawe, 1998).

REFLEX INTERACTIONS BETWEEN ENTERIC AND BILIARY NEURONS Since the beginning of the twentieth century, when Langman, Bayliss and Starling, and Trendelenberg were evaluating the neural activity of the bowel, it has been recognized that peristaltic reflex responses could be observed in the absence of the central nervous system. Evidence also exists for reflex interactions between the duodenum and the gallbladder and sphincter of Oddi, and between the gallbladder and sphincter of Oddi. Although these gutSO-gallbladder interconnections have not been evaluated as extensively as the intrinsic circuits of the gut, it is clear that the circuitry exists for reflex interactions between duodenal and biliary tract neurons.

NEURAL INTERACTIONS BETWEEN THE GUT AND THE GALLBLADDER In 1933, DuBois and Kistler found that stimulation of the duodenal ampulla causes the gallbladder to contract. This contractile response to ampulla stimulation was eliminated by transection of the bile duct, but gallbladder responses to vagal stimulation and stimulation of the cut end of the bile duct persisted. On the basis of these findings, DuBois and Kistler proposed that a direct neural connection exists between the gut and the gallbladder, and that these axons pass along the cystic duct. Morphological evidence from retrograde labelling experiments in the guinea-pig and Australian possum, show that the gallbladder receives extrinsic neural projections from several sources, and support the concept of a direct gut-gallbladder projection (Mawe and Gershon, 1989; Padbury et al., 1993). Sources of input to the gallbladder include the dorsal motor nucleus of the vagus nerve, nodose ganglia, coeliac ganglia, and thoracic dorsal root ganglia. In addition, when retrograde tracers were injected into the wall of the gallbladder, neurons of the duodenal myenteric plexus and the ganglia of the SO were labelled. These data indicate that, in addition to being regulated in conventional reflex responses that involve central nervous system processing, the circuitry exists for the gallbladder to receive direct inputs from the bowel. The evidence described above indicates that gut neurons project to the gallbladder where they exert an excitatory influence. However, the physiological relevance, if there is any, of such a pathway, and the precise origins and targets of the gut-gallbladder projection have not been resolved. According to the DuBois and Kistler study, activation of this circuit has an excitatory influence on gallbladder motor activity; however it is not known whether the target of this input is on gallbladder ganglia, on the smooth muscle cells, or at both of these locations. When retrograde tracers were injected into the gallbladder they could have been taken up by axon terminals in any layer of the organ.

NEURONS OF THE GALLBLADDER AND SPHINCTER OF ODDI

205

It is doubtful that the gut–gallbladder projection contributes to the fast EPSPs that have been studied in gallbladder ganglia. Following vagotomy, nicotinic input to gallbladder neurons, in response to stimulation of the nerve bundles that pass along the cystic duct, are completely eliminated (Mawe, Gokin and Wells, 1994). However, if this pathway is similar to the projection from enteric neurons to the pancreas (Kirchgessner and Pintar, 1991), there may be a serotonergic projection from the gut to the ganglia of the gallbladder. Nerve fibres that are immunoreactive for serotonin do exist in the ganglia of the gallbladder (Mawe and Gershon, 1989), and when serotonin is applied to gallbladder neurons, it causes a prolonged depolarization that is associated with an increase in the excitability of gallbladder neurons (Mawe, 1990). One potential physiological role of an excitatory neural pathway from the gut to the gallbladder would be to activate the gallbladder contractions that occur in coordination with the migrating myoelectric complex. During the interdigestive period, the gallbladder undergoes periods of increased intraluminal pressure, in phase with the migrating myoelectric complex, that are accompanied by a delivery of bile from the gallbladder to the lumen of the duodenum (Itoh et al., 1983; Ryan, 1987). It is thought that the migrating myoelectric complex serves a “housekeeping function”. According to this model, increased motor activity would advance undigested food from the proximal bowel toward the large intestines, and the associated delivery of bile into the intestinal lumen would facilitate the overall digestive process. NEURAL INTERACTIONS BETWEEN THE GUT AND THE SPHINCTER OF ODDI Duodenal distension and electrical field stimulation causes changes in SO tone in the dog, cat, and Australian possum (Wyatt, 1967; Thune, Jivegård and Svanvik, 1989; Saccone et al., 1994). In the possum, these responses are abolished both by crushing the duodenum between the site of stimulation and the SO and by TTX, but not by vagotomy. This indicates that neural projections from duodenal neurons to the SO are responsible for this response. Also in the Australian possum, Padbury and colleagues (1993) have demonstrated that injections of retrograde tracers into the SO resulted in retrograde labelling of duodenal myenteric neurons. In the guinea-pig, SO ganglia are richly innervated by calbindin-immunoreactive nerve fibres, but these ganglia contain few, if any, calbindin-immunoreactive neurons (Wells and Mawe, 1993). Since a high proportion of neurons in the nearby myenteric ganglia of the duodenum are calbindin-positive, the duodenum-myenteric projection neurons represent a likely source of calbindin-positive fibres in SO ganglia. This suggestion is supported by findings from studies demonstrating that the calbindin-immunoreactive neurons project from the myenteric plexus of the duodenum to SO ganglia (Kennedy and Mawe, 1998). Evidence from retrograde labelling studies validates the existence of a neural projection between the duodenum and the SO in the guinea-pig and Australian possum. Injection of DiI into the wall of the SO, in the Australian possum, in vivo, results in retrograde labelling of neurons in the duodenal myenteric plexus (Padbury et al., 1993). In vitro retrograde label studies in the guinea-pig have demonstrated that specific subpopulations of guineapig duodenal myenteric neurons project to the ganglionated plexus of the SO in guinea-pig (Kennedy and Mawe, 1998). An average of 110 neurons project from the duodenum to the

206

INNERVATION OF THE GASTROINTESTINAL TRACT

SO in the guinea-pig, and these neurons express ChAT-, but not NOS- or calretininimmunoreactivity. Importantly, many (~20%) of these projection neurons are calbindinpositive. This is a critical detail because calbindin is a marker for intrinsic primary afferent neurons in the guinea-pig enteric nervous system, and calbindin-positive neurons in the myenteric plexus send projections to the mucosa. The duodenum-SO projection neurons are depolarized by CCK, indicating that they are capable of responding to CCK released from the mucosa. Because calbindin-IR myenteric neurons send a projection to the mucosa, and postprandial release of CCK also occurs in the mucosa, it is possible that the duodenum-SO neural circuit is activated by postprandial release of mucosal CCK. To examine the nature of the synaptic signals received by SO neurons from neurons in the duodenum, membrane potential recordings were obtained from SO neurons with intracellular microelectrodes while duodenal nerve bundles and villi were stimulated (Kennedy and Mawe, 1999). Focal stimulation of interganglionic fibre bundles in the myenteric plexus of the duodenum, or duodenal mucosal villi, elicits nicotinic fast synaptic potentials in SO neurons. In these studies, antidromic action potentials were also detected in SO neurons indicating that SO neurons project to the duodenum. This reciprocal projection was verified morphologically by retrograde labelling studies demonstrating that SO neurons are labelled when DiI is applied to the duodenal myenteric plexus, in vitro. Thus, it is clear that duodenal neurons provide input to SO neurons and, at least in the guinea-pig, this input is in the form of nicotinic fast EPSPs through a reflex that can be activated at the duodenal mucosa. This circuit could play a role in any or all of the SOs functions. For example, it is possible that the neurons that project to the SO are activated, directly or through other neurons, by the release of CCK. According to this model, CCKinduced changes in SO tone would involve a local neural circuit rather than a direct action of hormonal CCK on SO neurons. This is consistent with the finding, described above, that SO neurons are not sensitive to the levels of CCK that exist in the serum. Another possible role of the duodenum-SO circuit could be to cause contraction of the SO in coordination with contractions of the duodenum. This would help prevent the reflux of lumenal contents into the biliary and pancreatic ducts. Finally, the gut-SO circuit could be involved in coordinating the relaxation of the SO in concert with the gallbladder contractions and delivery of bile that occur in phase with the migrating myoelectric complex. NEURAL INTERACTIONS BETWEEN THE GALLBLADDER AND THE SPHINCTER OF ODDI In addition to reflex interactions between the gut and the biliary tract, there exists a neural reflex circuit that links the gallbladder to the SO. In dogs, cats and humans, distension or electrical stimulation of the gallbladder results in a decreased motility, or flow resistance, in the SO (Wyatt, 1967; Thune, Thorness and Svanvik, 1986; Thune, Jivegård and Svanvik, 1989; Thune et al., 1991). In the cat, this response was eliminated by TTX or application of local anaesthetics to the bile ducts (Thune, Thorness and Svanvik, 1986). The mechanisms for the reflex between the gallbladder and the SO are unknown. Authors of these reports have suggested that the reflex may involve a direct neural link between the gallbladder and the SO; however, these studies have not ruled out the possibility of a vagal reflex mechanism.

NEURONS OF THE GALLBLADDER AND SPHINCTER OF ODDI

207

CONCLUDING REMARKS The neurons of the gallbladder have characteristics that are more similar to parasympathetic neurons than those of enteric neurons. To release their neuroactive compounds onto the effector tissues of the organ, gallbladder neurons must be stimulated, and the major source of excitatory input to these cells is vagal preganglionic fibres. Modulatory inputs that can up- or down-regulate the efficacy of this nicotinic ganglionic transmission include CCK and noradrenaline, which have presynaptic excitatory and inhibitory effects on vagal terminals, respectively. Furthermore, sensory fibres can release SP and CGRP in gallbladder ganglia to depolarize and increase the excitability of gallbladder neurons. The concept of how gallbladder neurons, which express acetylcholine plus an array of excitatory and inhibitory neuromodulators, conduct unambiguous signals to the smooth muscle and epithelial cells of the gallbladder needs to be resolved in future studies. The neurons of the SO represent a more heterogeneous population than those of the gallbladder. Electrically, SO ganglia comprise excitable neurons with spontaneous activity, less excitable neurons that do not exhibit spontaneous activity, and a small contingent of AH cells. Neurochemically, SO ganglia include neurons that appear to be either excitatory or inhibitory; the excitatory neurons express ChAT, SP and enkephalin, and the inhibitory neurons express NOS and VIP or NPY. Although the extrinsic inputs to SO ganglia have not been extensively studied as of yet, mounting evidence suggests that significant regulatory inputs to SO ganglia originate in the myenteric plexus of the duodenum. Future studies will focus on the neurochemical mediators of duodenal-SO interactions and the physiological activators of this pathway.

ACKNOWLEDGEMENTS The studies performed in the Mawe laboratory have been supported by NIH grants DK 45410 and NS 26995. We thank the alumni and active members of the Green Mountain Gallbag Company, including Wendy Pouliot, Ellen Cornbrooks, David Wells, Lei Zhang and Alex Gokin for their contributions to many of the studies that have been described here.

REFERENCES Alexander, W.F. (1940). The innervation of the biliary system. Journal of Comparative Neurology, 72, 357–370. Allescher, H.-D., Tougas, G., Vergara, P., Lu, S. and Daniel, E.E. (1992). Nitric oxide as a putative nonadrenergic noncholinergic inhibitory transmitter in the canine pylorus in vivo. American Journal of Physiology, 262, G695–G702. Azuma, H. and Fujiwara, M. (1973). Analysis of contractions induced by transmural stimulation in the isolated sphincter of Oddi. Japanese Journal of Pharmacology, 23, 723–732. Bauer, A.J., Hanani, M., Muir, T.C. and Szurszewski, J.H. (1991). Intracellular recordings from gallbladder ganglia of opossums. American Journal of Physiology, 260, G299–G306. Baumgarten, H.G. and Lange, W. (1969). Extrinsic adrenergic innervation of the extrahepatic biliary duct system in guinea-pigs, cats and rhesus monkeys. Zeitschrift für Zellforshung und Mikroskopische Anatomie, 100, 606–615. Behar, J. and Biancani, P. (1980). The effect of cholecystokinin and the octapeptide of cholecystokinin on the feline sphincter of Oddi and gallbladder. Journal of Clinical Investigation, 66, 1231–1239.

208

INNERVATION OF THE GASTROINTESTINAL TRACT

Behar, J. and Biancani, P. (1987). Pharmacologic characterization of excitatory and inhibitory cholecystokinin receptors of the cat gallbladder and sphincter of Oddi. Gastroenterology, 92, 764–770. Brotschi, E.A., Pattavino, J. and Williams, Jr. L.F. (1990). Intrinsic nerves affect gallbladder contraction in the guinea pig. Gastroenterology, 99, 826–830. Burnett, W., Gairns, F.W. and Bacsich, P. (1964). Some observations on the innervation of the extrahepatic biliary system in man. Annals of Surgery, 159, 8–26. Cai, W. and Gabella, G. (1983a). Innervation of the gallbladder and biliary pathways in the guinea pig. Journal of Anatomy, 136, 97–109. Cai, W. and Gabella, G. (1983b). The musculature of the gall bladder and biliary pathways in the guinea-pig. Journal of Anatomy, 136, 237–250. Cai, W.-Q. and Gabella, G. (1984). Catecholamine-containing cells in the nerve plexus of the guinea-pig gallbladder. Acta Anatomica, 119, 10–17. Cornbrooks, E.B., Pouliot, W.A. and Mawe, G.M. (1992). The structure of neurons and ganglia of the guinea pig gallbladder: light and electron microscopic studies. Journal of Comparative Neurology, 317, 31–44. Costa, M., Brooks, S., Waterman, S. and Mayo, R. (1992). Enteric neuronal circuitry and transmitters controlling intestinal motor function. In Advances in the Innervation of the Gastrointestinal Tract, edited by G.E. Holle and J.D. Wood, pp. 115–121. Amsterdam: Elsevier Science Publishers. Dahlstrand, C., Edin, E., Dahlstrom, A. and Ahlman, H. (1985). An in vivo model for the simultaneous study of motility of the gallbladder, sphincter of Oddi and duodenal wall in the cat. Acta Physiologica Scandinavica, 123, 355–362. Dahlstrand, C., Bjorck, S., Edin, R., Dahlstrom, A. and Ahlman, H. (1988). Substance P in the control of extrahepatic biliary motility in the cat. Regulatory Peptide, 20, 11–24. Dahlstrand, C., Dahlström, A. and Ahlman, H. (1989). Adrenergic and VIP-ergic relaxatory mechanisms of the feline extrahepatic biliary tree. Journal of the Autonomic Nervous System, 26, 97–106. Dahlstrand, C., Theordorson, E., Dahlström, A. and Ahlman, H. (1989). VIP-antisera inhibit the relaxatory motor responses of the feline sphincter of Oddi and gallbladder induced by VIP or vagal nerve stimulation. Acta Physiologica Scandinavica, 137, 375–378. Dahlstrand, C. (1990). The vagal nerves and peptides in the control of extrahepatic biliary motility. Acta Physiologica Scandinavia, 139, 1–52. Dahlstrand, C., Dahlström, A., Theodorsson, E., Rehfeld, J. and Ahlman, H. (1990). Is the CCK-8 induced relaxation of the feline sphincter of Oddi mediated by VIP neurons? Journal of the Autonomic Nervous System, 31, 75–84. Davies, P.J. and Campbell, G. (1994). The distribution and colocalization of neuropeptides and catecholamines in nerves supplying the gall bladder of the toad, Bufo marinus. Cell Tissue Research, 277, 169–175. De Giorgio, R.D., Parodi, J.E., Brecha, N.C., Brunicardi, F.C., Becker, J.M., Go, V.L.W. et al. (1994). Nitric Oxide Producing Neurons in the Monkey and Human Digestive System. The Journal of Comparative Neurology, 342, 619–627. De Giorgio, R.D., Zittel, T.T., Parodi, J.E., Becker, J.M., Brunicardi, F.C., Go, V.L.W. et al. (1995). Peptide immunoreactivities in the ganglionated plexuses and nerve fibres innervating the human gallbladder. Journal of the Autonomic Nervous System, 51, 37–47. Dodds, W.J., Hogan, W.J. and Geenen, J.E. (1989). Motility of the biliary system. In Handbook of Physiology. The Gastrointestinal System, Motility and Circulation, edited by S.G. Schultz, pp. 1055–1101. Bethesda, MD: American Physiological Society. DuBois, F.S. and Kistler, G.H. (1933). Concerning the mechanism of contraction of the gallbladder in the guinea pig. Proceedings of the Society for Experimental Biology and Medicine, 30, 1178–1180. Fisher, R.S., Rock, E. and Malmud, L.S. (1985). Cholinergic effects on gallbladder emptying in humans. Gastroenterology, 89, 716–722. Fontaine, J., Le Lievre, C. and Le Douarin, N.M. (1977). What is the developmental fate of the neural crest cells which migrate into the pancreas in the avian embryo. General and Comparative Endocrinology, 33, 394–404. Furness, J.B. and Costa, M. (1987). The Enteric Nervous System, New York: Churchill Livingston. Galligan, J.J. and Bertrand, P.P. (1994). ATP mediates fast synaptic potentials in enteric neurons. Journal of Neuroscience, 14, 7563–7571. Goehler, L.E., Sternini, C. and Brecha, N.C. (1988). Calcitonin gene-related peptide immunoreactivity in the biliary pathway and liver of the guinea-pig: distribution and colocalization with substance P. Cell and Tissue Research, 253, 145–150. Gokin, A.P., Hillsley, K. and Mawe, G.M. (1997). Cholecystokinin (CCK) depolarizes guinea pig sphincter of Oddi neurons by activating CCK-A receptors. American Journal of Physiology, 272, G1365–G1371. Gokin, A.P., Jennings, L.J. and Mawe, G.M. (1996). Actions of calcitonin gene-related peptide (CGRP) in guinea pig gallbladder ganglia. American Journal of Physiology, 271, G876–G883.

NEURONS OF THE GALLBLADDER AND SPHINCTER OF ODDI

209

Grossman, M.I. (1975). Gastrointestinal Hormones: Spectrums of Actions and Structure-Activity Relations. Thorofare: Charles Slack Inc. Grozdanovic, Z., Baumgarten, H.G. and Bruning, G. (1992). Histochemistry of NADPH-diaphorase, a marker for neuronal nitric oxide synthase, in the peripheral autonomic nervous system of the mouse. Neuroscience, 48, 225–235. Grozdanovic, Z., Mayer, B., Baumgarten, H.G. and Bruning, G. (1994). Nitric oxide synthase-containing nerve fibres and neurones in the gall bladder and biliary pathways of the guinea-pig. NeuroReport, 5, 837–840. Gullo, L., Bolondi, L., Priori, P., Casanova, P. and Labo, G. (1984). Inhibitory effect of atropine on cholyecystokinin-induced gallbladder contraction in man. Digestion, 29, 209–213. Guo, Y.-S., Singh, P., Gomez, G., Rajaraman, S. and Thompson, J.C. (1989). Contractile response of canine gallbladder and sphincter of Oddi to substance P and related peptides in vitro. Digestive Diseases and Sciences, 34, 812–817. Hanyu, N., Dodds, W.J., Layman, R.D. and Hogan, W.J. (1990a). Cholecystokinin-induced contraction of opossum sphincter of Oddi. Digestive Diseases and Sciences, 35, 567–576. Hanyu, N., Dodds, W.J., Layman, R.D. and Takahashi, I. (1990b). Mechanism of cholecystokinin-induced contraction of the opossum gallbladder. Gastroenterology, 98, 1299–1306. Harada, T., Katsuragi, T. and Furukawa, T. (1986). Release of acetylcholine mediated by cholecystokinin receptor from the guinea pig sphincter of Oddi. Journal of Pharmacology and Experimental Therapeutics, 239, 554–558. Hillsley, K. and Mawe, G.M. (1998). Correlation of electrophysiology, neurochemistry and axonal projections of guinea-pig sphincter of Oddi neurones. Neurogastroenterology and Motility, 10, 235–244. Hillsley, K., Jennings, L.J. and Mawe, G.M. (1998). Neural control of the gallbladder: an intracellular study of human gallbladder neurons. Digestion, 59, 125–129. Hirose, T. and Ito, Y. (1991). Excitatory and inhibitory responses of Oddi’s sphincter in guinea pigs. American Journal of Physiology, 260, G615–G624. Itoh, Z., Takahashi, I., Nakaya, M. and Suzuki, T. (1983). Interdigestive function of the gallbladder in the dog. In Functional Disorders of the Digestive Tract, edited by W.Y. Chey, pp. 259–265. New York: Raven Press. Ivy, A.C. and Oldberg, E. (1928). A hormone mechanism for gallbladder contraction and evacuation. American Journal of Physiology, 86, 599–613. Jazrawi, R.P., Pazzi, P., Petroni, M.L., Prandini, N., Paul, C., Adam, J.A., Gullini, S. and Northfield, T.C. (1995). Postprandial gallbladder motor function: refilling and turnover of bile in health and in cholelithiasis. Gastroenterology, 109, 582–591. Kaufman, H.S., Shermak, M.A., May, C.A., Pitt, H.A. and Lillemoe, K.D. (1993). Nitric oxide inhibits resting sphincter of Oddi activity. The American Journal of Surgery, 165, 74–80. Keast, J.R., Furness, J.B. and Costa, M. (1985). Distribution of certain peptide containing nerve fibres and endocrine cells in the gastrointestinal mucosa of five mammalian species. Journal of Comparative Neurology, 236, 403–422. Kennedy, A.L. and Mawe, G.M. (1998). Duodenal sensory neurons project to sphincter of Oddi ganglia in guinea pig. Journal of Neuroscience, 18, 8065–8073. Kennedy, A.L. and Mawe, G.M. (1999). Duodenal neurons provide nicotinic synaptic input to sphincter of Oddi neurons in guinea pig. American Journal of Physiology, 277, G226–G234. Kirchgessner, A.L. and Pintar, J.E. (1991). Guinea pig pancreatic ganglia: projections, transmitter content, and the type-specific localization of monoamine oxidase. Journal of Comparative Neurology, 305, 613–631. Kline, L.W. and Pang, P.K.T. (1992). Calcitonin gene related peptide relaxes cholecystokinin-induced contraction in guinea pig gallbladder strips in vitro. Canadian Journal of Physiology and Pharmacology, 70, 1571–1575. Kline, L.W. and Pang, P.K.T. (1994). Nitric oxide modulates the calcitonin gene-related peptide-induced relaxation in guinea pig gallbladder strips in vitro. Regulatory Peptides, 50, 207–212. Kyösola, K. (1977). Cholinesterases of the gall bladder. Histochemistry, 50, 337–346. Lanzini, A., Jazrawi, R.P. and Northfield, T.C. (1987). Simultaneous quantitative measurements of absolute gallbladder storage and emptying during fasting and eating in humans. Gastroenterology, 92, 852–861. Lillemoe, K.D., Webb, T.H. and Pitt, H.A. (1988). Neuropeptide Y: a candidate neurotransmitter for biliary motility. Journal of Surgical Research, 45, 254–260. Maggi, C.A., Sannticioli, P., Renzi, D., Patacchini, R., Surrenti, C. and Meli, A. (1989). Release of substance P- and calcitonin gene-related peptide-like immunoreactivity and motor response of the isolated guinea pig gallbladder to capsaicin. Gastroenterology, 96, 1093–1101. Mantyh, C.R., Pappas, T.N., Lapp, J.A., Washington, M.K., Neville, L.M., Ghilardi, J.R. et al. (1996). Substance P activation of enteric neurons in response to intraluminal Clostridium difficile toxin A in the rat ileum. Gastroenterology, 111, 1272–1280.

210

INNERVATION OF THE GASTROINTESTINAL TRACT

Marzio, L., DiGiammarco, A.M., Neri, M., Cuccurollo, F. and Malfertheiner, P. (1985). Atropine antagonizes cholecystokinin and cerulein induced gallbladder evacuation in man: a real time ultrasonographic study. American Journal of Gastroenterology, 80, 1–4. Mawe, G.M. (1990). Intracellular recording from neurones of the guinea-pig gall-bladder. Journal of Physiology (London), 429, 323–338. Mawe, G.M. (1991). The role of cholecystokinin in ganglionic transmission in the guinea-pig gall-bladder. Journal of Physiology (London), 439, 89–102. Mawe, G.M. (1992). Oddi: The Structure of Oddi’s Sphincter. In Sphincters: Normal Function – Changes in Diseases, edited by E.E. Daniel, S. Tsuchida, T. Tomita and M. Watanabe, pp. 175–187. Boca Raton: CRC Press. Mawe, G.M. (1993). Noradrenaline acts as a presynaptic inhibitory neurotransmitter in ganglia of the guinea-pig gall-bladder. Journal of Physiology (London), 461, 378–402. Mawe, G.M. (1995). Tachykinins as mediators of slow EPSPs in guinea-pig gall-bladder ganglia. Involvement of neurokinin-3 receptors. Journal of Physiology (London), 485, 513–524. Mawe, G.M. and Gershon, M.D. (1989). Structure, afferent innervation, and transmitter content of ganglia of the guinea pig gallbladder: relationship to the enteric nervous system. Journal of Comparative Neurology, 283, 374–390. Mawe, G.M., Gokin, A.P. and Wells, D.G. (1994). Actions of cholecystokinin and norepinephrine on vagal inputs to ganglionic cells in guinea pig gallbladder. American Journal of Physiology, 267, G1146–G1151. Mawe, G.M., Talmage, E.K., Cornbrooks, E.B., Gokin, A.P., Zhang, L. and Jennings, L.J. (1997). Innervation of the gallbladder: structure, neurochemical coding, and physiological properties of guinea pig gallbladder ganglia. Microscopy Research and Techniques, 39, 1–13. McKirdy, M.L., McKirdy, H.C. and Johnson, C.D. (1994). Non-adrenergic non-cholinergic inhibitory innervation shown by electrical field stimulation of isolated strips of human gall bladder muscle. Gut, 35, 412–416. Meldrum, L.A., Bojarski, J.C. and Calam, J. (1987). Effects of substance P and other neuropeptides on guineapig gallbladder muscle. Digestion, 37, 193–199. Mori, J., Azuma, H. and Fujiwara, M. (1971). Adrenergic innervation and receptors in the sphincter of Oddi. European Journal of Pharmacology, 14, 365–373. Mourelle, M., Guarner, F., Molero, X., Moncada, S. and Malagelada, J. (1993a). Regulation of gall bladder motility by the arginine-nitric oxide pathway in guinea pigs. Gut, 34, 911–915. Mourelle, M., Guarner, F., Moncada, S. and Malagelada, J.-R. (1993b). The arginine/nitric oxide pathway modulates sphincter of Oddi motor activity in guinea pigs and rabbits. Gastroenterology, 105, 1299–1305. Padbury, R.T.A. (1990). The innervation of the extrahepatic biliary tract with peptide- and catecholaminecontaining nerves. Doctor of Philosophy, Flinders University of South Australia. Padbury, R.T.A., Furness, J.B., Baker, R.A., Toouli, J. and Messenger, J.P. (1993). Projections of nerve cells from the duodenum to the sphincter of Oddi and gallbladder of the Australian possum. Gastroenterology, 104, 130–136. Pallin, B. and Skoglund, S. (1964). Neural and humeral control of gallbladder emptying mechanism in the cat. Acta Physiologica Scandinavica, 60, 358–362. Pauletzki, J.G., Sharkey, K.A., Davison, J.S., Bomzon, A. and Shaffer, E.A. (1993). Involvement of L-argininenitric oxide pathways in neural relaxation of the sphincter of Oddi. European Journal of Pharmacology, 232, 263–270. Persson, C.G.A. (1971). Adrenoceptor functions in the cat choledochoduodenal junction in vitro. British Journal of Pharmacology, 42, 447–461. Persson, C.G.A. (1972). Adrenergic, cholecystokinetic and morphine induced effects on extra-hepatic biliary motility. Acta Physiologica Scandinavica (Supplement), 383, 1–32. Persson, C.G.A. (1973). Dual effects of the sphincter of Oddi and gallbladder induced by stimulation of the right splanchnic nerves. Acta Physiologica Scandinavica, 87, 334–343. Pozo, M.J., Salida, G.M. and Madrid, J.A. (1989). Cholecystokinin-induced gallbladder contraction is influenced by nicotinic and muscarinic receptors. Archives Internationales de Physiologie, de Biochimie et de Biophysique, 97, 403–408. Rakovska, A., Milenov, K. and Bocheva, A. (1989). Effect of cholecystokinin octapeptide and somatostatin on the motility of guinea pig and canine gallbladder. Comparative Biochemistry and Physiology, 94C, 649–653. Ryan, J.P. (1987). Motility of the gallbladder and biliary tree. In Physiology of the Gastrointestinal Tract, edited by L.R. Johnson, pp. 695–721. New York: Raven Press. Ryan, J. and Cohen, S. (1977). Effect of vasoactive intestinal polypeptide on basal and cholecystokinin induced gallbladder pressure. Gastroenterology, 73, 870–872. Ryan, J.P. and Ryave, S. (1978). Effect of vasoactive intestinal polypeptide on gallbladder smooth muscle in vitro. American Journal of Physiology, 234, E44–E46.

NEURONS OF THE GALLBLADDER AND SPHINCTER OF ODDI

211

Saccone, G.T., Harvey, J.R., Baker, R.A. and Toouli, J. (1994). Intramural neural pathways between the duodenum and sphincter of Oddi in the Australian brush-tailed possum in vivo. Journal of Physiology (London), 481, 447–456. Sand, J., Tainio, H. and Nordback, I. (1993). Neuropeptides in pig sphincter of Oddi, bile duct, gallbladder, and duodenum. Digestive Diseases and Sciences, 38, 694–700. Schemann, M. and Wood, J.D. (1989). Electrical behavior of myenteric neurones in the gastric corpus of the guinea-pig. Journal of Physiology (London), 417, 501–518. Shaffer, E.A. (1991). Gallbladder Disease. In Paediatric Gastroenterology, edited by W.A. Walker and P.R. Durie, pp. 1152–1170. Philadelphia: BC Becker. Siou, G.P.S., Belai, A. and Burnstock, G. (1994). A developmental study of the localization of NADPHdiaphorase in the ganglionated plexus of the guinea-pig gallbladder. Cell and Tissue Research, 276, 61–68. Strah, K.M., Pappas, T.N., Melendez, R.L. and Debas, H.T. (1985). Anticholinegeric influence on exogenous and endogenous stimulation of gallbladder contraction. Gastroenterology, 88, 1601A. Strah, K.M., Pappras, T.N., Melendez, R.L. and Debas, H.T. (1986). Contrasting cholinergic dependence of pancreatic and gallbladder responses to cholecystokinin. American Journal of Physiology, 250, G665–G669. Sutherland, S.D. (1966). The intrinsic innervation of the gallbladder in Macaca rhesus and Cavia porcellus. Journal of Anatomy, 100, 261–268. Sutherland, S.D. (1967). The neurons of the gallbladder and gut. Journal of Anatomy, 101, 701–709. Tack, J.F. and Wood, J.D. (1992). Electrical behaviour of myenteric neurones in the gastric antrum of the guineapig. Journal of Physiology, 447, 49–66. Takahashi, I., Suzuki, T., Aizawa, I. and Itoh, Z. (1982). Comparison of gallbladder contraction induced by motilin and cholecystokinin in dogs. Gastroenterology, 82, 419–424. Takahashi, T., May, D. and Owyang, C. (1991). Cholinergic dependence of gallbladder response to cholecystokinin in the guinea pig in vitro. American Journal of Physiology, 261, G565–G569. Talmage, E.K., Pouliot, W.A., Cornbrooks, E.B. and Mawe, G.M. (1992). Transmitter diversity in ganglion cells of the guinea pig gallbladder: an immunohistochemical study. Journal of Comparative Neurology, 317, 45–56. Talmage, E.K. and Mawe, G.M. (1993). NADPH-diaphorase and VIP are co-localized in neurons of gallbladder ganglia. Journal of the Autonomic Nervous System, 43, 83–90. Talmage, E.K., Pouliot, W.A., Schemann, M. and Mawe, G.M. (1996). Structure and chemical coding of human, canine and opossum gallbladder ganglia. Cell and Tissue Research, 284, 289–302. Talmage, E.K., Hillsley, K., Kennedy, A.L. and Mawe, G.M. (1997). Identification of the cholinergic neurons in guinea-pig sphincter of Oddi ganglia. Journal of the Autonomic Nervous System, 64, 12–18. Tamura, K. and Wood, J.D. (1989). Electrical and synaptic properties of myenteric plexus neurones in the terminal large intestine of the guinea-pig. Journal of Physiology, 415, 275–298. Thune, A., Thorness, E. and Svanvik, J. (1986). Reflex regulation of flow resistance in the feline sphincter of Oddi by hydrostatic pressure in the billiary tract. Gastroenterology, 91, 1364–1369. Thune, A., Jivegård, L. and Svanvik, J. (1989). Flow resistance in the feline choledocho-duodenal sphincter as studied by constant-pressure and constant-perfusion techniques. Acta Physiologica Scandinavia, 135, 279–284. Thune, A., Saccone, G.T.P., Scicchitano, J.P. and Toouli, J. (1991). Distension of the gall bladder inhibits sphincter of Oddi motility in humans. Gut, 32, 690–693. Vogalis, F., Bywater, R.A.R. and Taylor, G.S. (1989). Propulsive activity of the isolated choledochoduodenal junction of the guinea pig. Journal of Gastrointestinal Motility, 1, 115–121. Wade, P.R. and Wood, J.D. (1988). Electrical behavior of myenteric neurons in guinea pig distal colon. American Journal of Physiology, 254, G522–G530. Wells, D.G. and Mawe, G.M. (1993). Physiological and morphological properties of neurons in the sphincter of Oddi region of the guinea pig. American Journal of Physiology, 265, G258–G269. Wells, D.G. and Mawe, G.M. (1994). Sympathetic input to ganglia of the guinea pig sphincter of Oddi. American Journal of Physiology, 266, G1162–G1169. Wells, D.G., Talmage, E.K. and Mawe, G.M. (1995). Immunohistochemical identification of neurons in ganglia of the guinea pig sphincter of Oddi. Journal of Comparative Neurology, 352, 106–116. Wiley, J.W., O’Dorisio, T.M. and Owyang, C. (1988). Vasoactive intestinal polypeptide mediates cholecystokinin-induced relaxation of the sphincter of Oddi. Journal of Clinical Investigation, 81, 1920–1924. Wood, J.D. (1989). Electrical and synaptic behavior of enteric neurons. In Handbook of Physiology, edited by S.G. Schultz, pp. 465–518. Bethesda, MA: American Physiological Society. Wyatt, A.P. (1967). The relationship of the sphincter of Oddi to the stomach, duodenum and gallbladder. Journal of Physiology (London), 193, 225–243. Yamamura, T., Takahashi, T., Kusunoki, M., Kantoh, M., Ishikawa, Y. and Utsunomiya, J. (1986). Cholecystokinin octapeptide-evoked [3H] acetylcholine release from guinea-pig gallbladder. Neuroscience Letters, 65, 167–170.

212

INNERVATION OF THE GASTROINTESTINAL TRACT

Yamasato, T. and Nakayama, S. (1990). Participation of the parasympathetic and sympathetic nerves in regulation of gallbladder motility in the dog. Acta Medica Okayama, 44, 79–86. Yau, W.M. and Youther, M.L. (1984). Modulation of gallbladder motility by intrinsic cholinergic neurons. American Journal of Physiology, 247, G662–G666. Zhang, L., Bonev, A.D., Mawe, G.M. and Nelson, M.T. (1994). Protein kinase A mediates activation of ATP-sensitive K+ currents by CGRP in gallbladder smooth muscle. American Journal of Physiology, 267, G494–G499.

6 Pharmacology of the Enteric Nervous System Marcello Tonini, Fabrizio De Ponti, Gianmario Frigo and Francesca Crema Department of Internal Medicine and Therapeutics, Section of Clinical and Experimental Pharmacology, University of Pavia, Piazza Botta 10, I-27100 Pavia, Italy The intestine is capable of complex behaviours which allow mixing, propulsion, digestion and absorption of food. The mechanism whereby the intrinsic neurons of the enteric nervous system and the circular smooth muscle coordinate their activities to produce propagated contractions have been the subject of intensive investigations for almost a century. Since the intestine contains a large number of neuroactive substances and an equally large number of receptors for a variety of chemicals, preparations of longitudinal muscle with attached myenteric plexus have been widely used by pharmacologists and therefore have provided detailed information about the effect of drugs on motor neurons of the gastrointestinal tract. The complexity of the chemical coding of enteric neurons mirrors the complexity of the receptors involved in neurogenic intestinal functions. This chapter is intended to review the findings that associate any pharmacologically induced change in enteric nerve activity with the appropriate receptor type or subtype recognized by means of selective agonists and/or antagonists. KEY WORDS: enteric nervous system; chemical coding; cholinergic transmission; adrenergic transmission; NANC transmission; serotonergic transmission; tachykinins.

INTRODUCTION An explosive growth has recently characterized the pharmacology of the enteric nervous system, thanks to the improved anatomical, electrophysiological and functional knowledge of the various classes of neurons intrinsic to the intestinal wall and of their chemical coding (Costa and Brookes, 1994; Furness et al., 1994; Costa et al., 1996). This gave impetus to the search for highly selective agonists and antagonists for the pharmacological characterization of the receptor subtypes involved in fast and slow transmission between neurons, in the excitatory and inhibitory transmission to smooth muscle cells, and in the modulation of transmitter release by presynaptic or prejunctional mechanisms (Starke, Göthert and Kilbinger, 1989; Fuder and Muscholl, 1995). 213

214

INNERVATION OF THE GASTROINTESTINAL TRACT

In spite of the fact that enteric neural pathways are involved in the regulation of several important functions, such as absorption/secretion, maintenance of vascular tone and motility, those involved in the regulation of gastrointestinal motility (i.e. the myenteric plexus) have been investigated much more extensively. This led to an outstanding development of the pharmacology of receptors mediating the effects of neurotransmitters on intestinal motility, which will be one of the main topics of the present chapter. Myenteric neurons can be divided into several functional classes that include sensory neurons, ascending and descending interneurons, ascending and descending (both short and long) motor neurons to the circular muscle, and excitatory motor neurons to the longitudinal muscle coat (Costa et al., 1996). All these neurons contain a wide array of neuroactive substances, which can be co-stored in and co-released from the same neuron upon appropriate stimulation. They include acetylcholine (ACh), tachykinins and opioid peptides (usually enkephalins) in the ascending pathways, and ACh, serotonin (5-hydroxytryptamine: 5-HT), somatostatin, vasoactive intestinal peptide (VIP), nitric oxide (NO) and dynorphin in descending interneurons, while descending motor neurons may contain VIP, NO, pituitary adenylate cyclase-activating peptide, gastrin-releasing peptide, dynorphin/ enkephalin and, as a result of pharmacological evidence, ATP. Some other substances are present in extrinsic pathways innervating the intestine (e.g. noradrenaline) and in mucosal endocrine cells (e.g. 5-HT) and mast cells. The complexity of the chemical coding of enteric neurons mirrors the complexity of the receptors (in terms of their subtypes, distribution, density, ionic and transduction mechanisms) involved in neurogenic intestinal functions. The presence of some of these receptors in the gut has been documented only from a pharmacological point of view and their functional significance is debated. In fact, it is a general rule that, for any given endogenous substance, a number of receptor subtypes exists even in the same tissue, leading either to similar or contrasting effects. Furthermore, it is now well documented that disease states can lead to alterations in the level of expression of a given receptor subtype. This concept of inducible receptors (Donaldson, Hanley and Villablanca, 1997) can of course profoundly affect drug action in vivo. In some cases, compounds behaving as antagonists on receptors activated by endogenous ligands offer a widely used pharmacological tool to assess the contribution of a given neurotransmitter in mediating a biological response. Indeed, if a given receptor is under tonic control by an endogenous transmitter, administration of an antagonist will elicit effects that are opposite to those of the endogenous ligand. This is based on the assumption that antagonists, according to the classical definition, lack efficacy. However, evidence has recently begun to accrue that, in the case of G protein-coupled receptors, some antagonists not only bind to the receptor, but also induce a conformational change that favours uncoupling of the receptor from its G protein (Schütz and Freissmuth, 1992; Kenakin, 1994). This implies that, in some in vitro systems with constitutive receptor activity, antagonists may show a property termed negative efficacy (or inverse agonism). Antagonists may thus be classified into compounds with no intrinsic activity and those with negative intrinsic activity (i.e. those reducing constitutive receptor activity, such as timolol or ICI 118551; Chidiac et al., 1994; Samama et al., 1994). These considerations cast some doubts on data obtained with receptor antagonists simply with the assumption that they prevent agonists binding. However, it should be noted that, while constitutive

PHARMACOLOGY OF THE ENTERIC NERVOUS SYSTEM

215

activity of the unliganded receptor is a well-established phenomenon in reconstituted systems with purified components, its physiological relevance is still unknown (Schütz and Freissmuth, 1992; Milligan, Bond and Lee, 1995) and, at the present state of knowledge, the effect observed with antagonists in vivo can be tentatively ascribed to antagonism of the endogenous ligand, unless evidence for inverse agonism is obtained. With these basic concepts in mind, this chapter is mainly intended to review those findings that associate any pharmacologically induced change in enteric nerve activity with the appropriate receptor type or subtype recognized by means of selective agonists and/or antagonists. Because of the complexity of the subject, this chapter should have had the size of a book. We hope the reader will forgive us if substantial information had to be neglected due to space limitations. Whenever possible, reference to extensive reviews on the subject has been provided.

PHARMACOLOGY OF CHOLINERGIC TRANSMISSION MUSCARINIC RECEPTORS Acetylcholine is the major transmitter in the enteric nervous system (Furness and Costa, 1987). Inhibition of muscarinic receptors by hyoscine was found to inhibit peristalsis (Tonini et al., 1981), although a weaker hyoscine- or atropine-resistant peristalsis may develop over time (Tonini et al., 1981; Barthó et al., 1982b; Schwörer and Kilbinger, 1988). One of the most popular intestinal preparations used to demonstrate ACh release from cholinergic nerve endings is the longitudinal muscle with attached myenteric plexus (LM-MP) from the guinea-pig small intestine (Paton and Zar, 1968). ACh is released from unstimulated LM-MP preparations and in greater amounts in response to electrical field stimulation. The postjunctional target of neuronally released ACh is a muscarinic receptor associated with membrane phosphoinositide breakdown, which causes the effector cell to contract (Eglen et al., 1994). Recently, this receptor has been recognized as the M3 receptor subtype using the M3-selective antagonist hexahydrosiladifenidol (HHSiD) in both the longitudinal (Giraldo et al., 1988) and circular muscle (Dietrich and Kilbinger, 1995). The prejunctional target of neuronally released ACh, as well as of applied muscarinic agonists, are inhibitory muscarinic autoreceptors, which depress, through a negative feed-back mechanism, the release of ACh itself. Indeed, muscarinic receptor antagonists were found to increase the electrically stimulated ACh release (Starke, Göthert and Kilbinger, 1989; Vizi et al., 1989). The evidence that muscarinic autoreceptors were antagonized with low affinity by the M1-selective antagonist pirenzepine (Kilbinger et al., 1984) and the M2-selective antagonist AF-DX 116 (Dammann et al., 1989) and with high affinity by the M3-selective antagonists HHSiD (Fuder, Kilbinger and Müller, 1985) and 4-DAMP (Kilbinger et al., 1984) suggests that these prejunctional sites belong to the M3 subtype. However, since HHSiD and 4-DAMP possess high affinity also for the M4 receptor, the participation of this site in negative feed-back mechanisms cannot be ruled out (Eglen and Watson, 1996). Furthermore, there is also evidence that cholinergic ganglionic transmission is regulated by presynaptic inhibitory muscarinic autoreceptors characterized so far as M2 subtypes (North, Slack and Surprenant, 1985).

216

INNERVATION OF THE GASTROINTESTINAL TRACT

In contrast with the longitudinal muscle, prejunctional autoreceptors that inhibit ACh release in the circular muscle of the guinea-pig ileum belong to the M1 subtype (Dietrich and Kilbinger, 1995). This is of particular relevance, since M1 receptors located on the cell bodies of myenteric S neurons (i.e. interneurons and motor neurons) cause a slow membrane depolarization (North and Tokimasa, 1982; Morita, North and Tokimasa, 1982) and stimulate resting ACh output (Kilbinger and Nafziger, 1985; Vizi et al., 1989) through a mechanism inhibited by pirenzepine. Functionally, M1 receptors on myenteric (inter) neurons are activated during the ascending excitatory reflex in response to localized gut wall distension. In fact, in a partitioned organ bath, administration of pirenzepine at the site of distension inhibited the reflex contraction recorded in the oral compartment. Conversely, administration of pirenzepine at the site of reflex recording was ineffective (Tonini and Costa, 1990). Previous experiments by Schwörer and Kilbinger (1988) had indicated that M1 receptors also play a role in peristalsis since nanomolar concentrations of pirenzepine were found to enhance peristalsis, an effect apparently in contrast with that observed in the ascending excitatory reflex (Tonini and Costa, 1990). Pirenzepine-induced facilitation of peristalsis was explained in terms of antagonism of M1 receptors activating an inhibitory pathway. Based on recent findings, however, this effect can be easily explained by the removal of a tonic M1 receptor mediated inhibition of ACh release from the circular muscle (Dietrich and Kilbinger, 1995). The existence of muscarinic heteroceptors, i.e. modulating release of neurotransmitters other than ACh (e.g. noradrenaline), is also documented (Alberts and Stjärne, 1982; Manber and Gershon, 1979; Yokotani and Osumi, 1993; Marino et al., 1997). NICOTINIC RECEPTORS Nicotinic receptors are ligand-gated ion channels which are present on certain subclasses of myenteric neurons, where they cause rapid membrane depolarization (fast excitatory post-synaptic potentials, fast EPSPs) leading to fast communication between neurons. Stimulation of these receptors evokes release of numerous transmitters including ACh (Yau, Dorsett and Parr, 1989), NO (Jin and Grider, 1992), somatostatin (Grider, 1989) and L-glutamate (Wiley, Lu and Owyang, 1991). A polysynaptic chain of cholinergic ascending interneurons are involved in the distension-evoked ascending excitatory reflex of the ileal circular muscle. The ACh released from these neurons acts on nicotinic receptors on interneurons and motor neurons. Inhibition of these receptors by hexamethonium added to the oral, intermediate or anal compartment of a partitioned organ bath, markedly reduced the amplitude of the reflex contraction (Tonini and Costa, 1990), indicating that fast ganglionic nicotinic transmission exerts a crucial role in such responses. In contrast, probably because ACh is not handled by all classes of descending interneurons, the descending inhibitory reflex is reportedly less sensitive to nicotinic receptor blockade (Smith and Furness, 1988; Smith, Bornstein and Furness, 1990; Smith and Robertson, 1998). Nicotinic ganglionic transmission is also crucial for normal peristalsis, since hexamethonium abolishes propulsion both in the ileum and colon (Fontaine, Van Nueten and Janssen, 1973; Barthó, Holzer and Lembeck, 1987; Barthó et al., 1989; Kadowaki, Wade and Gershon, 1996; Tonini et al., 1996). In the ileum, weaker peristalsis can be obtained

PHARMACOLOGY OF THE ENTERIC NERVOUS SYSTEM

217

using greater distending volumes, providing naloxone is present to block the inhibitory effects of endogenous opioids (Barthó, Holzer and Lembeck, 1987; Barthó et al., 1989). Thus, non-nicotinic (i.e. hexamethonium-resistant) ganglionic transmission also occurs during peristalsis. This form of transmission is inhibited by endogenous opioids. Typically, nicotinic receptors are localized to the somatodendritic region of excitatory interneurons and motor neurons. Recently, however, prejunctional and presynaptic nicotinic receptors have been described in myenteric plexus neurons of the guinea-pig small intestine, where they mediate the release of excitatory non-cholinergic (tachykinergic) transmitters (Galligan, 1999; Schneider and Galligan, 2000; Schneider, Perrone and Galligan, 2000).

PHARMACOLOGY OF NON-CHOLINERGIC EXCITATORY (TACHYKINERGIC) TRANSMISSION Tachykinins represent a family of peptides which have been isolated from mammals (substance P, neurokinin A, neurokinin B) and amphibians (physalaemin, phyllomedusin, uperolein, eledoisin, kassinin) and whose role in intestinal motility has been reviewed recently (Holzer and Holzer-Petsche, 1997a,b). In the gut, although tachykinins may be present in the enterochromaffin and immune cells of the mucosa, most of them are found in intrinsic enteric neurons and extrinsic primary afferent nerve fibres (Holzer and HolzerPetsche, 1997b). In the myenteric plexus of the guinea-pig small intestine, substance P (SP) immunoreactivity has been found in excitatory motor neurons projecting to the circular (CM) and longitudinal muscle (LM) layers, ascending interneurons and intrinsic primary afferent neurons (Llewellyn-Smith et al., 1988; Brookes and Costa, 1990; Brookes, Steele and Costa, 1991a,b, 1992; Brookes et al., 1992; Costa et al., 1996). SP and neurokinin A (NKA) have been found in synaptic vesicles from enteric neurons where they are co-released with ACh, whereas neurokinin B (NKB) is absent (Deacon et al., 1987; McDonald et al., 1988; Regoli, Bondon and Fouchère, 1994; Lippi et al., 1998).

TACHYKININ RECEPTORS Three distinct receptors mediate the biological effects of endogenous tachykinins in the gastrointestinal tract. They are known as tachykinin NK1, NK2 and NK3 receptors, which have some preferential (though sometimes negligible) affinity for SP, NKA and NKB, respectively (Maggi, 1995). Since NKB is not expressed in the gastrointestinal tract, NK3 receptors are agonized by SP and NKA, which may act as full agonists at these sites (Maggi, 1995). Nevertheless, the pharmacological characterization of tachykinin receptors was made possible by the development of potent and selective peptide-derived agonists, e.g. NK1 receptor: SP methyl ester, [Sar9]-SP sulphone; NK2 receptor: [β-Ala8]-NKA(4–10), [Nleu10]-NKA(4–10); NK3 receptor: senktide, [MePhe7]-NKB, and peptide (FK-888; MEN11420; PD-161182) or non-peptide antagonists (SR-140333; SR-48968; SB-142801) (Holzer and Holzer-Petsche, 1997a).

218

INNERVATION OF THE GASTROINTESTINAL TRACT

Effects of tachykinin receptor stimulation on neuromuscular preparations The NK1 and NK2 subtypes are located on smooth muscle cells, where they mediate contraction (Holzer and Lembeck, 1980; Costa et al., 1985; Maggi et al., 1990a, 1994a) and on neurons. In the guinea-pig ileum and colon LM, the direct contractile response to SP is mediated by the NK1 subtype (Lee et al., 1982; Briejer et al., 1993a), while in the CM both NK1 and NK2 receptors are involved (Maggi et al., 1990a, 1994a). In the colon, NK1 and NK2 receptors produce a non-adrenergic non-cholinergic (atropine-resistant) excitation of CM to nerve stimulation, which show a remarkable specialization. NK1 receptors mediate a “fast” excitatory transmission, which depend on the activation of voltage-sensitive calcium channels, whereas NK2 receptors mediate a “slow” excitatory transmission that is largely independent from calcium channels (Maggi, Zagorodnyuk and Giuliani, 1994b). Species-related pharmacological differences detected by variable affinities of competitive antagonists have emerged for both NK1 and NK2 receptors (Maggi, 1995). NK1 receptors are expressed on both ascending and descending neuronal pathways and on the interstitial cells of Cajal (Sternini et al., 1995; Legat et al., 1996; Portbury et al., 1996b; Johnson, Bornstein and Burcher, 1998; Lomax, Bertrand and Furness, 1998; Lecci et al., 1999; Bian et al., 2000). In CM strips from the guinea-pig small intestine, contractions mediated by NK1 receptors are atropine-resistant, but partially sensitive to tetrodotoxin (TTX) (Maggi et al., 1990a; Bian et al., 2000), suggesting that NK1 receptors are present, at least in part, on excitatory neurons, where they stimulate release of a non-cholinergic transmitter. In the same preparation, use of GR-73632, a selective NK1 receptor agonist, produced a TTX-sensitive decrease of spontaneous contractility, due to release of NO from inhibitory motor neurons (Lecci et al., 1999). NK1 receptor-mediated, TTX-sensitive relaxations due to release of NO have been also demonstrated in the colonic circular muscle (Bian et al., 2000). Lastly, prejunctional inhibitory NK1 receptors operating a feedback inhibition of ACh release have been described in LM-MP preparations (Kilbinger et al., 1986; Loeffler et al., 1994). Neuronal NK2 receptors have been described in descending pathways only (Portbury et al., 1996a; Zagorodnyuk and Maggi, 1995). In the guinea-pig colon CM strips, NKA and the NK2 receptor-selective synthetic agonist [β-Ala8]-NKA(4–10), in addition to causing a direct excitation, were found to activate inhibitory non-adrenergic non-cholinergic motor neurons with consequent release of NO and an apamin-sensitive inhibitory transmitter (Zagorodnyuk and Maggi, 1995). NK3 receptors are predominantly neuronal in origin. They are located on intrinsic primary afferent neurons (Mann et al., 1997a), and on ascending and descending pathways (Maggi et al., 1990a, 1993, 1994c; Legat et al., 1996; Johnson et al., 1996; Johnson, Bornstein and Burcher, 1998; Jenkinson et al., 1999). Intrinsic sensory neurons are connected to each other via synapses communicating with NK3 receptors, which therefore contribute to the reinforcement of perceptive signal (Furness et al., 1998). In ascending pathways, activation of NK3 receptors by senktide or NKB promotes release of tachykinins (SP and NKA) and ACh from the myenteric plexus of the guinea-pig ileum (Guard and Watson, 1987; Yau et al., 1992). NK3 receptors located on descending pathway induce release of NO and inhibition of circular muscle activity in both the ileum and colon (Maggi et al., 1993; Maggi, Zagorodnyuk and Giuliani, 1994c).

PHARMACOLOGY OF THE ENTERIC NERVOUS SYSTEM

219

Electrophysiological effects of tachykinins on neurons Substance P depolarizes 85% of AH neurons and 76% of S neurons in the myenteric plexus when applied electrophoretically. The depolarization is associated with an increase in input resistance due to inactivation of a resting potassium conductance (Katayama, North and Williams, 1979). Slow EPSPs in S and AH neurons evoked by repeated pulse stimuli are due to the release of a non-cholinergic neurotransmitter and can be abolished by desensitization to SP (Katayama and North, 1978; Johnson et al., 1981). Thus, tachykinins are likely to mediate some slow EPSPs in myenteric neurons. On pharmacological grounds, it might be predicted that NK3 (and NK1), but not NK2, receptors are involved (Hanani et al., 1988; Willard, 1990; Bertrand and Galligan, 1994, 1995). Role of endogenous tachykinins Electrically induced, nerve-mediated, non-cholinergic contractions of the CM are reduced in amplitude by antagonists at the NK1 and NK2 receptor subtypes (Costa et al., 1985; Barthó et al., 1992). This indicates the participation in the contractile response of both SP and NKA, which are co-released upon electrical stimulation (Lippi et al., 1998). The ascending reflex contraction of the CM produced by radial distension of the gut wall is abolished by hyoscine at low distensions, but partly reduced at higher distension volumes. Under the latter conditions, the reflex contraction is reduced by tachykinin NK1 and NK2 receptor antagonists (Costa et al., 1985; Holzer, 1989), and in particular, by those acting at the NK2 receptor subtype (Barthó et al., 1992; Maggi et al., 1994d). Exposure of guinea-pig ileum CM to SP produces depolarization and action potentials (Niel, Bywater and Taylor, 1983a). The depolarization is associated with decreased membrane resistance. Noncholinergic fast excitatory junction potentials in the intestinal CM are abolished by desensitization to SP and by SP receptor antagonists acting at NK1 and NK2 receptors (Niel, Bywater and Taylor, 1983a; Taylor and Bywater, 1986; Serio et al., 1998). Taken together, these results indicate that endogenous tachykinins mediate non-cholinergic excitatory transmission to the CM. This transmission is mainly mediated by the NK2 receptor subtype, although NK1 receptors participate in this event. There is now convincing evidence that tachykinins play a role in peristalsis. Figure 6.1 illustrates the distribution of tachykinin receptors on effector cells and intrinsic neuronal pathways (i.e. myenteric plexus), which mediate propulsive activity in the gastrointestinal tract. SP is released into venous effluent in response to distension of the small intestine during peristalsis. This release is blocked by TTX and enhanced by the nicotinic receptor agonist dimethylphenylpiperizine (DMPP) (Donnerer et al., 1984; Donnerer, Holzer and Lembeck, 1984). SP stimulates peristalsis when the intestine is distended by a subthreshold volume (Holzer and Lembeck, 1979) and increases the frequency of emptying (Barthó et al., 1982a). Furthermore, atropine-resistant peristalsis is greatly inhibited by tachykinin receptor antagonists (Barthó et al., 1982b), whereas hexamethonium-resistant peristalsis is abolished by spantide (Barthó et al., 1989). Analysis of the receptors by which tachykinins influence peristalsis has shown that activation of NK1 receptors inhibits, while activation of NK2 and NK3 facilitates propulsion (Holzer, Schluet and Maggi, 1995). The NK1 receptor-mediated depression of peristalsis involves descending motor pathways utilizing

220

INNERVATION OF THE GASTROINTESTINAL TRACT

Figure 6.1 Scheme illustrating the neuronal pathways regulating standing reflexes and peristalsis in the gastrointestinal tract. Radial distension of the gut wall by an intraluminal bolus excites ascending and descending pathways leading to excitatory and inhibitory reflexes. All three tachykinin receptor subtypes (NK1,2,3) are expressed on neurons where they mediate excitatory (+) or inhibitory effects (–) on propulsion. Tachykinin receptors located on smooth muscle cells mainly belong to the NK1 and NK2 subtypes and are excitatory in nature.

NO as a transmitter (Portbury et al., 1996b; Lecci et al., 1999), since the inhibitory effect was prevented by a nitric oxide synthase (NOS) inibitor (Holzer, 1997). On the other hand, blockade by SR-140333 of NK1 receptors markedly reduced the amplitude of inhibitory junction potentials evoked in the circular muscle by activation of descending pathways during motility reflexes. This clearly indicates that endogenous tachykinins acting via NK1 receptors partly mediate transmission to inhibitory motor neurons (Johnson, Bornstein and Burcher, 1998). Use of antagonists at NK1 (SR-140333), NK2 (SR-48968) and NK3 (SR-142801) receptors made it possible to establish that endogenous tachykinins acting via NK1 and NK2 receptors contribute to maintain intestinal peristalsis when cholinergic ganglionic and neuromuscular transmission via muscarinic receptors is suppressed. SR-142801 lacked a major influence on peristalsis, indicating that NK3 receptors play little role in peristalsis generation (Holzer et al., 1998). Conversely, NK3 receptors have been found to play a substantial role in transmission from intrinsic sensory neurons and from ascending (and descending) interneurons to excitatory motor neurons during motility reflexes (Johnson et al., 1996; Johnson, Bornstein and Burcher, 1998). It is interesting to point out that distension-evoked standing reflexes of the circular muscle subserve but cannot be identified with motor propulsive events (Tonini et al., 1996). In the guinea-pig isolated colon, NK1 and NK2 receptors contribute to peristalsis (FoxxOrenstein and Grider, 1996). In another study, blockade of NK1, NK2 and NK3 receptors inhibited propulsion. NK1 and NK2 receptors showed a synergistic interaction with muscarinic receptors, an effect observed also in the ileum (Holzer and Maggi, 1994), whereas inhibition produced by NK3 receptor blockade was additive with the decelerating effect

PHARMACOLOGY OF THE ENTERIC NERVOUS SYSTEM

221

caused by nicotinic receptor blockade. Simultaneous blockade of all the three receptors produced a 50% inhibition of the velocity of propulsion indicating that endogenous tachykinins acting at NK1, NK2 and NK3 receptors play an important role in maintaining propulsion (Tonini et al., 2001). In the rabbit isolated distal colon, NK2 and NK3 receptors located on descending pathways decelerate propulsion by activating a NO-dependent pathway. Both receptors, however, may also accelerate propulsion via distinct mechanisms. NK2 receptors mediate a facilitatory effect by a postjunctional synergistic interaction with muscarinic receptors, while NK3 receptors located on ascending pathways facilitate propulsion through a synergistic interaction with nicotinic receptors (Onori et al., 2000, 2001).

PHARMACOLOGY OF ADRENERGIC TRANSMISSION Noradrenaline is well established as the transmitter of postganglionic sympathetic neurons supplying the gastrointestinal tract, since it meets all the criteria for identification of a substance as a neurotransmitter (Furness and Costa, 1987). However, sympathetic neurotransmission may not be exclusively noradrenergic, as indicated, for instance, by the fact that adrenoceptor antagonists and reserpine pretreatment do not inhibit neurogenic vasoconstriction in submucosal arterioles, at least in the guinea-pig (Surprenant, 1994). In this species, noradrenergic axons supplying intestinal arterioles and submucous ganglia also contain neuropeptide Y and somatostatin, respectively, while sympathetic neurons reaching the myenteric plexus and the circular muscle coat contain neither peptide (Furness and Costa, 1987). The chemical coding of motility-inhibiting adrenergic neurons is different in the rat, a species where these neurons contain neuropeptide Y (McConalogue and Furness, 1994). Noradrenergic axons in the gut are most numerous in the myenteric and submucous ganglia and around arterioles, but there is also evidence of a sparse noradrenergic supply to the circular muscle and to the mucosa (Gabella, 1979). Concerning the sympathetic fibers reaching directly the muscle layer, in general the adrenergic innervation of the longitudinal muscle is less abundant than that of the circular layer (Gabella, 1979). Histochemical studies show that the noradrenergic innervation is particularly rich in the muscle coat of the so-called sphincteric regions such as the gastroesophageal junction, the pyloric region and the anal canal (Gabella, 1979; Furness and Costa, 1987). With few exceptions, stimulation of the sympathetic supply to sphincteric muscle is excitatory (Furness and Costa, 1974) and is thought to be due to a direct effect of noradrenaline on smooth muscle α-adrenoceptors. Electrophysiological studies have shown that noradrenergic fibers supplying the myenteric ganglia cause presynaptic inhibition of cholinergic transmission (Hirst and McKirdy, 1974a), while stimulation of noradrenergic nerves at frequencies of up to 50 Hz causes no changes in the membrane properties of myenteric neurons (Hirst and McKirdy, 1974b), although exogenous α2- and β1-adrenoceptor agonists can respectively hyperpolarize (Morita and North, 1981; Surprenant and North, 1985; Galligan and North, 1991; Wells and Mawe, 1994) or depolarize (Schemann, 1991; Tack and Wood, 1992) some neurons. Exogenous noradrenaline was shown to act presynaptically at different gut levels

222

INNERVATION OF THE GASTROINTESTINAL TRACT

inhibiting fast EPSPs (Nishi and North, 1973; Schemann, 1991; Tack and Wood, 1992; Mawe, 1993). The most direct evidence supporting the concept of a tonic sympathetic inhibition on gut motility would be provided by electrophysiological studies showing a tonic discharge of sympathetic efferent neurons supplying the myenteric plexus and/or smooth muscle cells. Unfortunately, while release of noradrenaline from sympathetic terminals accounts for the inhibitory post-synaptic potential (IPSP) through activation of postsynaptic α2-adrenoceptors in submucosal neurons, little information is available on IPSPs in myenteric neurons, where IPSPs are seldom recorded, and there are only a few data on the possible transmitter(s) involved (Surprenant, 1994). Although noradrenaline can hyperpolarize myenteric neurons, hyperpolarization of myenteric neurons in response to sympathetic activity has never been shown (McConalogue and Furness, 1994), while, in the case of secretomotor neurons, neuronal hyperpolarization seems to be the main mode of action of the sympathetic input (North and Surprenant, 1985). In a recent study, designed to test whether sphincter of Oddi ganglia are a target of sympathetic input (Wells and Mawe, 1994), noradrenaline, when applied to neurons exhibiting IPSPs (which were, however, a minority), induced membrane hyperpolarization. Interestingly, α2-adrenoceptor blockade inhibited, whereas application of desipramine increased the amplitude of the IPSPs, an observation consistent with the hypothesis put forward by the authors that noradrenaline may mediate IPSPs at least in this preparation. ADRENOCEPTOR SUBTYPES At present, adrenoceptors are classified into three major classes: α1, α2 and β, each further divided into several subtypes (Bylund et al., 1994; Alexander and Peters, 2000), but only fragmentary information is available on the distribution of adrenoceptor subtypes in the enteric nervous system (Table 6.1). On a pharmacological basis, α1-adrenoceptors are defined as receptors selectively stimulated by methoxamine, cirazoline or phenylephrine and competitively blocked by low concentrations of prazosin, WB-4101 or corynanthine, while α2-adrenoceptors are characterized by a high sensitivity to agonists such as clonidine, xylazine, UK14 304 and antagonists such as rauwolscine, yohimbine and idazoxan (Ruffolo and Hieble, 1994). Concerning β-adrenoceptors, although many experimental data can be explained in terms of the existence of β1- and β2-adrenoceptors, a β-adrenoceptor with “atypical” pharmacological characteristics was described in some tissues such as the adipose tissue, gastrointestinal tract and skeletal muscle (Arch et al., 1984; Bond and Clarke, 1987; Croci et al., 1988; Bianchetti and Manara, 1990; Molenaar et al., 1991) and is now classified as a β3-adrenoceptor (Alexander and Peters, 2000). This β-adrenoceptor displays atypically low pA2 values for conventional antagonists such as propranolol, ICI 118551 and CGP 20712A and high sensitivity to agonists such as BRL 37344 and SR 58611A (Arch and Kaumann, 1993). Among conventional β-adrenoceptor antagonists, alprenolol is one of the compounds displaying the highest affinity (Blue et al., 1990), although its pA2 value for β3-adrenoceptors is rather low (Arch and Kaumann, 1993). Molecular biological approaches have confirmed the existence of at least three human genes encoding different β-adrenoceptors corresponding to the β1-, β2- and β3-subtypes (Frielle et al., 1987;

Phenylephrine, methoxamine, cirazoline

Prazosin, corynanthine, WB4101

Gq/11

Selective agonists

Selective antagonists

Predominant effectors

CGP20712A, atenolol Gs

Gi/o

Xamoterol, prenalterol

Smooth muscle relaxation

Smooth muscle, enteric neurons (?)

β1-adrenoceptors

Yohimbine, idazoxan

Clonidine, UK14304

Presynaptic inhibition, smooth muscle contraction

Adrenergic neurons (autoreceptors), myenteric neurons (heteroceptors), smooth muscle

α2-adrenoceptors1

Gs

ICI118551

Terbutaline, salbutamol, ritodrine

Smooth muscle relaxation, facilitation of transmitter release2

Smooth muscle, peripheral noradrenergic neurons (?)

β2-adrenoceptors

Gs

SR59230A

BRL37344, SR58611A, CL316243

Smooth muscle relaxation

Smooth muscle

β3-adrenoceptors

α1- and α2-adrenoceptors are further divided into the following subtypes, respectively: α1A, α1B, α1D and α2A, α2B, α2C (Alexander and Peter, 2000), but only fragmentary information is available at present on their distribution in the gut; 2 This effect has been reported in peripheral organs other than the gastrointestinal tract (Langer, 1981; Brunn et al., 1994).

Smooth muscle contraction or relaxation, neuronal depolarization (gastric neurons)

Functional response

1

Smooth muscle, gastric neurons

Distribution in the gut

α1-adrenoceptors1

TABLE 6.1 Synopsis of adrenoceptors in the gut

PHARMACOLOGY OF THE ENTERIC NERVOUS SYSTEM 223

224

INNERVATION OF THE GASTROINTESTINAL TRACT

Kobilka et al., 1987; Emorine et al., 1989). More recently, the development of selective β3-antagonists has provided a new tool for β3-adrenoceptor identification in functional tests (Manara et al., 1995; De Ponti et al., 1996; De Ponti, 1997). ROLE OF PREJUNCTIONAL AND POSTJUNCTIONAL ADRENOCEPTORS Adrenoceptors of both α- and β-subtype have been recognized at different levels of the gastrointestinal tract where they are involved in the regulation of motility and secretion. In particular, α1-adrenoceptors are located postsynaptically on smooth muscle cells and, to a lesser extent, on intrinsic neurons, while α2-adrenoceptors may be present both pre- and postsynaptically (Table 6.1). At a presynaptic level, α2-adrenoceptors may act either as autoreceptors inhibiting noradrenaline release from adrenergic nerves (Langer, 1977; Starke, Göthert and Kilbinger, 1989), or as heteroceptors modulating release of other neurotransmitters (especially ACh) from nerve terminals (Paton and Vizi, 1969; Vizi, 1979). An important postsynaptic location of α2-adrenoceptors is that on enterocytes, where they control water and electrolyte absorption. β1- and β2-adrenoceptors are found mainly on smooth muscle cells, but the former may be present on enteric neurons (Bülbring and Tomita, 1987; Ek, Bjellin and Lundgren, 1986) (Table 6.1). Prejunctional adrenoceptors An electrophysiological study on guinea-pig antral myenteric neurons provided evidence for a concentration-dependent inhibition of cholinergic fast EPSPs and non-cholinergic slow EPSPs by application of noradrenaline or clonidine (Tack and Wood, 1992). Yohimbine and phentolamine, but not prazosin, reversed this effect, suggesting an involvement of α2-adrenoceptors. Interestingly, when noradrenaline was applied on cell bodies, no response was observed, in line with the notion that axo-axonal synaptic mechanisms are responsible for the adrenergic inhibitory effect (Tack and Wood, 1992). Another study carried out on myenteric neurons of the guinea-pig gastric corpus (Schemann, 1991) also found inhibition by noradrenaline of fast EPSPs, probably by activation of α2-adrenoceptors on extrinsic vagal fibers. Besides α2-adrenoceptors, it has been suggested that also α1- and β-adrenoceptors may be present on gastric neurons (Allescher et al., 1989; Schemann, 1991; Tack and Wood, 1992). Application of noradrenaline depolarized a small fraction of antral myenteric neurons and enhanced their excitability probably by activation of α1-adrenoceptors on somal membranes (Tack and Wood, 1992). The authors suggested that this α1-adrenoceptor may be located on non-cholinergic inhibitory neurons. In myenteric neurons of the guinea-pig gastric corpus, application of noradrenaline resulted in prolonged excitation in 40% of the neurons tested (Schemann, 1991). This excitatory effect, which is probably mediated by α1-adrenoceptors, seems peculiar to the stomach, since neuronal depolarization is not usually reported in other parts of the gut. Concerning the small bowel, many reports are available on presynaptic α2-adrenoceptors inhibiting neurotransmitter release at this level (Paton and Vizi, 1969; Wikberg, 1978; Drew, 1978; Vizi, 1979; Bauer and Kuriyama, 1982; Reese and Cooper, 1984; Wessler et al., 1987), while the presence of presynaptic β-adrenoceptors having a facilitatory effect

PHARMACOLOGY OF THE ENTERIC NERVOUS SYSTEM

225

on neurotransmitter release is less well documented (Alberts, Ögren and Sellstöm, 1985; Kotska et al., 1989). Evidence of a neuronal location of α2-adrenoceptors in the guinea-pig small intestine was provided by binding and functional studies (Drew, 1978; Wikberg and Lefkowitz, 1982): in the denervated longitudinal muscle, [3H]-clonidine binding was virtually abolished, indicating that denervation eliminated α2-adrenoceptors (Wikberg and Lefkowitz, 1982). Electrophysiological studies with intracellular recordings of guinea-pig myenteric neurons showed α2-adrenoceptor-mediated hyperpolarizations (resulting from an increase in potassium conductance) (Morita and North, 1981; Surprenant and North, 1985; Galligan and North, 1991) and inhibition of fast EPSPs (Galligan and North, 1991). Recent studies suggest that α2-autoreceptors and α2-heteroceptors modulating ACh release in the small bowel belong to the α2A/D subtype (Funk et al., 1995; Liu and Coupar, 1997). In the colon, sympathetic nerve stimulation reduces ACh release (Beani, Bianchi and Crema, 1969; Del Tacca et al., 1970) as well as the response to pelvic nerve stimulation (Gillespie and Khoyi, 1977) by activation of presynaptic α2-adrenoceptors (Marcoli et al., 1985), similarly to what has been described above for the small bowel. Indeed, clonidine is one of the most potent inhibitors of the colonic peristaltic reflex (Marcoli et al., 1985), an event requiring the integrity of neural circuits. Presynaptic α2-adrenoceptors can also modulate the release of the non-adrenergic, non-cholinergic transmitter, as indicated by inhibition of inhibitory junction potential amplitude in circular smooth muscle of guineapig caecum (Reilly, Hoyle and Burnstock, 1987) and by the contractile effect observed in colonic longitudinal muscle strips in mice (Fontaine, Grivegnée and Reuse, 1984). Postjunctional adrenoceptors Catecholamines can affect gastric motility through direct activation of smooth muscle adrenoceptors, which may belong to the α1-, α2-, β1-, β2- and β3-class (El-Sharkawy and Szurszewski, 1978; Verplanken, Lefebvre and Bogaert, 1984; Bülbring and Tomita, 1987; Coleman, Denyer and Sheldrick, 1987; McLaughlin and MacDonald, 1991; Cohen et al., 1995). β-mediated responses are usually inhibitory, α2-responses are excitatory, while the responses to α1-adrenoceptor stimulation are variable: both relaxations and contractions have been reported (Sahyoun, Costall and Naylor, 1982; Costall, Naylor and Tan, 1983; Chihara and Tomita, 1987; Bülbring and Tomita, 1987; MacDonald, Kelly and Dettmar, 1990; Mandrek and Kreis, 1992). In circular muscle strips of the porcine pyloric ring, the excitatory responses to catecholamines were unaffected by propranolol or yohimbine but were completely antagonized by prazosin and phentolamine, indicating the involvement of α1-adrenoceptors (Mandrek and Kreis, 1992). Binding studies with [3H]-prazosin and [3H]-yohimbine support the existence of both α1- and α2-adrenoceptors in the guinea-pig stomach, although they do not make it possible to distinguish between neural and smooth muscle receptors (Taniguchi et al., 1988). In some species, such as the rainbow trout, noradrenaline and adrenaline have been reported to induce smooth muscle contraction mainly by a direct action on muscular α2-adrenoceptors (Kitazawa, Kondo and Temma, 1986). Although functional responses can be obtained in the isolated stomach by stimulation of postjunctional α1- and α2-adrenoceptors, their physiological relevance in vivo is still unknown.

226

INNERVATION OF THE GASTROINTESTINAL TRACT

Both α- and β-adrenoceptors have been detected at the postjunctional level in the small bowel. Excitatory α1-adrenoceptors are found especially in the terminal ileum (Bauer, 1981, 1982; Bauer and Kuriyama, 1982), although inhibitory responses probably mediated by these receptors have also been described (Bülbring and Tomita, 1987). Postjunctional α2-adrenoceptors have been detected in a binding study (Ahmad et al., 1991), but their function is still disputed, although Bauer and Kuriyama (1982) suggested an inhibitory effect by observing reduction or suppression of the generation of excitatory junction potentials by clonidine. Postjunctional β-adrenoceptors mediating smooth muscle relaxation were initially characterized as belonging to the β1-subtype (Grassby and Broadley, 1984; Mian, Malta and Raper, 1984; Sim and Lim, 1983). In the rabbit ileum, motor inhibition by perivascular nerve stimulation was reduced by 75% by the β1-adrenoceptor antagonist atenolol, while the remainder of the response was blocked by the β2-adrenoceptor antagonist butoxamine (Greenwood, Davison and Dodds, 1990). Several reports showing low pA2 values of conventional β-blockers have now provided evidence in favour of the presence of β3-adrenoceptors (Salimi, 1975; Bond and Clarke, 1987; Bond and Clarke, 1988; Grassby and Broadley, 1987; Norman and Leathard, 1990; Taneja and Clarke, 1992; Growcott et al., 1993a; Growcott et al., 1993b; MacDonald et al., 1994), which, at least in some preparations, seem to contribute to a significant extent to isoprenaline-induced relaxation (Tesfamariam and Allen, 1994). Postjunctional α-adrenoceptors may belong to the α1- and α2-subtype: the former may be both inhibitory (Fontaine, Grivegnée and Reuse, 1984; Dettmar, Kelly and MacDonald, 1986; Reilly, Hoyle and Burnstock, 1987) and excitatory (Venkova, Milne and Krier, 1994), while activation of α2A-adrenoceptors, at least in the circular smooth muscle of the canine colon, results in a significant increase in contractile force, probably mediated by a decrease in cyclic AMP levels (Zhang et al., 1992). In line with the general rule of a postjunctional location of β-adrenoceptors, most colonic β-sites mediate relaxation by a direct action on smooth muscle cells (Ek, Jodal and Lundgren, 1987; Taniyama, Kuno and Tanaka, 1987; Kwon et al., 1993; Smith et al., 1993), except for a minor population of neuronal β1-adrenoceptors which have been proposed to inhibit the activity of cholinergic neurons (Ek and Lundgren, 1982; Ek, 1985). This hypothesis of a neuronal location was based on the observation that the effect of the partial β1-agonist prenalterol was markedly reduced by the neurotoxin tetrodotoxin and 6-hydroxydopamine pretreatment (Ek, Bjellin and Lundgren, 1986). Binding studies performed in rat colon with [125I]-(−)-pindolol and [3H]-dihydroalprenolol revealed a small population of β1- (20–25%) and a large β2-adrenoceptor population (Ek and Nahorski, 1986; Landi et al., 1992). Although binding studies have failed to detect β3-sites because of the lack of selective ligands (Landi et al., 1992), several functional studies have provided evidence in favour of the presence of β3-sites. Grivegnée, Fontaine and Reuse (1984), studying the relaxatory effect of isoprenaline on the canine colon, concluded that the inhibitory effect of conventional β-blockers on the agonist-induced relaxation was “weak even at high concentrations”. Since this is now an established hallmark for the presence of β3-adrenoceptors, several investigators have assessed their role in the modulation of colonic motility in vitro (Bianchetti and Manara, 1990; McLaughlin and MacDonald, 1990; Landi, Croci and

PHARMACOLOGY OF THE ENTERIC NERVOUS SYSTEM

227

Manara, 1993; MacDonald and Lamont, 1993; Koike et al., 1994; De Ponti et al., 1995, 1996) and found concentration-dependent relaxations by using selective β3-agonists such as SR 58611A or CGP 12177 (a partial agonist).

PHARMACOLOGY OF NON-ADRENERGIC, NON-CHOLINERGIC INHIBITORY TRANSMISSION Intrinsic inhibitory transmission in the gut has much attracted the attention of investigators, leading to the recognition that non-adrenergic, non-cholinergic (NANC) inhibitory neural responses can be demonstrated in virtually all sections of the gut. The nature of the transmitter(s) responsible for NANC inhibitory transmission has been a matter of debate: over the years, ATP, VIP-like peptides and NO have been proposed as the main mediators of NANC relaxatory responses. Recently, evidence has begun to accrue that the hypotheses about the identity of the transmitter are not mutually exclusive and that multiple inhibitory mechanisms may coexist (Maggi and Giuliani, 1993; Lefebvre, Smits and Timmermans, 1995; Smits and Lefebvre, 1996; Selemidis, Satchell and Cocks, 1997) and may operate at different gut levels in response to mechanical, chemical or electrical stimuli. NANC inhibitory neurons are involved in the relaxation of sphincter regions, in the accommodation processes in the gastric fundus, small and large intestine, as well as in the descending inhibition in response to localized gut wall distension and peristalsis (Ciccocioppo et al., 1994; Tonini et al., 2000). VIP-LIKE PEPTIDES The VIP family of peptides includes VIP, peptide histidine isoleucine (PHI), pituitary adenylate cyclase-activating peptide (PACAP), glucagon, secretin and gastrin-releasing factor. There is considerable evidence in favour of a transmitter role for VIP in the intestine (Furness and Costa, 1987; McConalogue and Furness, 1994). In the guinea-pig small intestine, VIP is found in short and long descending inhibitory CM motor neurons and cholinergic and non-cholinergic descending interneurons (Furness and Costa, 1979a; Costa et al., 1980a, 1996; Jessen et al., 1980; Schultzberg et al., 1980; Costa and Furness, 1983; Llewellyn-Smith et al., 1988; Brookes, Steele and Costa, 1991b). A small population of VIP-immunoreactive LM motor neurons may exist (Schultzberg et al., 1980; Costa and Furness, 1983). VIP-immunoreactive fibres form synapses on both VIP- and NOSimmunoreactive myenteric neurons (Llewellyn-Smith, Furness and Costa, 1985; Costa et al., 1996). PHI is derived from the same precursor protein as VIP, is colocalized with VIP in vesicles (Agoston et al., 1989), and is thus found in the same populations of neurons as VIP. PACAP was originally isolated from ovine hypothalamus. Two forms occur: a 39-amino acid peptide and a 27-amino acid peptide which show 68% homology with VIP (Miyata et al., 1989; Miyata et al., 1990). PACAP nerve fibres are found in the smooth muscle layers of all regions of the rat gastrointestinal tract (Köves and Arimura, 1990; Nagahama et al., 1998). PACAP-immunoreactive fibres are found in the myenteric plexus and muscularis

228

INNERVATION OF THE GASTROINTESTINAL TRACT

externa of the guinea-pig small and large intestine (Sundler et al., 1992; Portbury et al., 1995). While high affinity PACAP binding sites in the central nervous system do not bind VIP, in the periphery PACAP and VIP appear to share high affinity binding sites (Gottschall et al., 1990; Mao et al., 1998; Teng et al., 1998). Effects of VIP-like peptides on neuromuscular preparations VIP, at concentrations greater than 0.1 nM, enhances the cholinergic twitch and produces contraction of guinea-pig small intestine LM-MP preparations (Kusunoki et al., 1986). The contraction is partly blocked by hyoscine (Kusunoki et al., 1986), suggesting that VIP stimulates ACh and possibly tachykinin release (Katsoulis et al., 1992) from LM motor neurons. More direct evidence has shown that VIP causes increased release of [3H]-ACh from LM-MP preparations, isolated myenteric ganglia and from LM-MP synaptosomes (Yau, Youther and Verdun, 1985; Kusunoki et al., 1986; Yau, Dorsett and Youther, 1986b; Yau, Dorsett and Parr, 1989). Intra-arterial injection of VIP or PACAP1–27 or PACAP1–38 promotes ACh release from canine ileal circular muscle (Fox-Threlkeld et al., 1999). In the guinea-pig ileum CM, VIP produces concentration-dependent relaxations through a direct myogenic effect, which is resistant to apamin (Costa, Furness and Humphreys, 1986). PACAP causes relaxation of rat stomach, duodenum, jejunum, ileum and colon, through a direct action on the LM and CM (Mungan et al., 1992). In the human sigmoid colon, PACAP and VIP concentration-dependently inhibit phasic myogenic contractions of the LM. The effect of PACAP but not VIP is reduced by 80% by apamin. The inhibitory effect of VIP, but not PACAP, was attenuated by 70% by TEA (Schwörer et al., 1992). Similar findings were obtained in the rat colon (Ekblad, 1999), where the potassium channels involved in the VIP action are also sensitive to charybdotoxin (Kishi et al., 2000). These results suggest that PACAP and VIP act at different receptor subtypes which are coupled to different potassium channels, as also observed in the guinea-pig taenia coli (Jin et al., 1994). The receptors involved belong to the PAC1 subtype in the dog ileum (Fox-Threlkeld et al., 1999) and in the rat colon (Ekblad, 1999), whereas only the VIP2 /PACAP3 receptor (now VPAC2 receptor) is apparently expressed in gastric smooth muscle cells of rabbit and guinea-pig (Teng et al., 1998). Electrophysiological effects of VIP-like peptides on neurons and enteric smooth muscle cells VIP depolarizes myenteric neurons and mimics the slow EPSP in AH neurons (Williams and North, 1979; Zafirov et al., 1985). By combining electrophysiological and immunohistochemical methods, Katayama, Lees and Pearson (1986) demonstrated that all VIPimmunoreactive neurons belong to the S type. Non-cholinergic slow EPSPs could be evoked in 77% of VIP-immunoreactive neurons by stimulating orally or anally to the recording site. Thus, VIP neurons receive both ascending and descending inputs. PACAP1–27 or PACAP1–38 were found to depolarize 96% of AH neurons and 36% of S type neurons, respectively (Christofi and Wood, 1993).

PHARMACOLOGY OF THE ENTERIC NERVOUS SYSTEM

229

In the rat stomach, the apamin-resistant inhibitory junction potential (IJP) is mediated by VIP (Ito et al., 1990), whereas both VIP and PACAP are involved in the IJP generation in the longitudinal muscle of distal colon (Kishi et al., 1996). VIP produces a slow hyperpolarization of the CM of the guinea-pig ileum, antagonized by VIP10–28 (Crist, He and Goyal, 1992). In the murine gastric fundus circular muscle, the slow component of the IJP is also blocked by VIP10–28. This effect is apparently mediated by blockade of neuronal VIP receptors, which in turn cause release of NO as final transmitter (Mashimo et al., 1996). Role of endogenous VIP-like peptides VIP is a recognized transmitter of inhibitory motor neurons of the gut in a number of mammalian species, which is released upon a variety of physiological stimuli (Grider and Makhlouf, 1988b). In the rat and guinea-pig colon and human jeunum, distension-induced reflex relaxation is associated with increased VIP release and is blocked by VIP10–28 or VIP antibodies (Grider and Makhlouf, 1986, 1988b; Grider and Rivier, 1990; Foxx-Orenstein and Grider, 1996; Grider, Foxx-Orenstein and Jin, 1998). Release of PACAP from rat colon during reflex relaxation has also been documented (Grider et al., 1994). PACAP participates in the inhibitory transmission of guinea-pig teaenia coli, internal anal sphincter of opossum and in the canine ileal circular muscle (Jin et al., 1994; Chadker and Rattan, 1998; Fox-Threlkeld et al., 1999). Antibodies to somatostatin reduce both VIP release and the relaxation; exogenous somatostatin stimulates VIP release (Grider, Arimura and Makhlouf, 1987). Grider and Makhlouf (1986) have therefore proposed that somatostatin interneurons stimulate release of VIP from inhibitory motor neurons in response to distension. NITRIC OXIDE Nitric oxide is now known to act as a transmitter in the central and autonomic nervous system. The synthetic enzyme for NO, nitric oxide synthase (NOS) has been localized immunohistochemically in both central and peripheral neurons (Bredt, Hwang and Snyder, 1990). The previously characterized enzyme, NADPH diaphorase (Hope and Vincent, 1989), appears to be the same as NOS in some tissues and has therefore been used as a marker for NOS-containing neurons (Dawson et al., 1991; Hope et al., 1991). NOS catalyzes the synthesis of NO and L-citrulline from L-arginine via a Ca2+/calmodulindependent mechanism, which also requires the presence of tetrahydrobiopterin and NADPH as cofactors (Bredt and Snyder, 1989; Knowles et al., 1989; Mayer, John and Böhme, 1990). In some studies, concentrations of L-citrulline are used as an index of NOS activity (Currò, Volpe and Preziosi, 1996). NOS immunoreactivity has been demonstrated to be colocalized with VIP in discrete populations of myenteric neurons: descending inhibitory motor neurons and descending interneurons to other myenteric ganglia and submucous ganglia (Costa et al., 1991, 1996). Generally, the colocalization NOS/VIP has been recognized throughout the gastrointestinal tract of various species, from the stomach (Tonini et al., 2000) to the internal anal sphincter. Double immunostaining in the canine bowel revealed NOS and VIP in the same nerve varicosities, but never in the same organelles (Berezin et al., 1994).

230

INNERVATION OF THE GASTROINTESTINAL TRACT

The “receptor” for NO is the soluble form of guanylate cyclase. NO binds to the haeme group of this enzyme, producing a conformational change to the active state. Thus, NO stimulates synthesis of cGMP, which in turn is believed to regulate protein kinases, phosphodiesterases and ion channels (Goy, 1991; Vincent and Hope, 1992). A recent study (Franck et al., 1997) using a novel inhibitor of guanylate cyclase (ODQ), suggested that electrical and mechanical effects of endogenous and exogenous NO in the canine colon are largely due to cGMP synthesis, while no evidence was found in support of a cGMPindependent mechanism of NO action. NO is now believed to act as a NANC inhibitory transmitter in several regions of the gastrointestinal tract in different species. The following sections will concentrate largely on the effects of exogenous and endogenous NO in the guinea-pig small intestine. Effects of NO on neuromuscular preparations Electrically induced NANC relaxations of the CM in the guinea-pig ileum are reduced by approximately 50% by apamin, while the remaining relaxation is blocked by L-N Gnitroarginine methyl ester (L-NAME). Alone, L-NAME reduces the relaxation by 71% and this inhibition can be reversed by L-arginine (Humphreys, Costa and Brookes, 1991). These results, suggest that NO-induced relaxation depends in part on an apamin-resistant mechanism and in part on an apamin-sensitive one. Sodium nitroprusside (SNP), a nitric oxide donor, causes relaxation of the guinea-pig ileum (Osthaus and Galligan, 1992) which is partly apamin-sensitive (which means involvement of small conductance calcium-activated potassium channels). Furthermore, relaxations can be mimicked by 8-bromo-cGMP (Osthaus and Galligan, 1992). The exact mechanism whereby cGMP causes relaxation is not clear, although it is known to vary according to the tissue and species. In smooth muscle cells isolated from canine colon, cGMP increases the probability of opening of potassium channels (Thornbury et al., 1991). The resultant hyperpolarization reduces voltage-dependent calcium influx into the muscle cells, thereby reducing tone. However, not all cGMP-mediated relaxation is associated with hyperpolarization. In vascular smooth muscle, cGMP has been shown to stimulate Ca2+-ATPase activity, thus reducing cytoplasmic calcium concentrations (Lincoln, 1989). Grider and Makhlouf (1992) proposed a complex mechanism for relaxation of intestinal tissue: they reported that NO could be synthesized in isolated myenteric ganglia from the guinea-pig small intestine in response to DMPP and gastrin-releasing peptide (GRP) (Jin and Grider, 1992). Circular smooth muscle cells from rat colon and guinea-pig gastric fundus could also synthesize NO in response to VIP (Grider and Makhlouf, 1992; Grider et al., 1992). L-N G-nitroarginine (L-NNA) inhibited VIP-induced relaxation and suppressed NO production. In a further study, Jin et al. (1992) reported that relaxation induced by SNP could be abolished by a cGMP-dependent protein kinase (protein kinase G) inhibitor, whilst relaxation due to VIP was reduced by 42%. Furthermore, VIP-induced relaxation was reduced by 33% by an inhibitor of cAMP-dependent protein kinase (protein kinase A), whereas the response to SNP was unaffected. Grider et al. (1992) have therefore proposed that VIP is released from inhibitory motor neurons and acts on smooth muscle receptors leading to stimulation of adenylate cyclase and NO production. NO in turn stimulates guanylate cyclase and also feeds back to further stimulate VIP release from the

PHARMACOLOGY OF THE ENTERIC NERVOUS SYSTEM

231

inhibitory motor neuron. In the canine colon, both VIP and NO appear to be responsible for the relaxation. Huizinga, Tomlinson and Pintin-Quezada (1992) found that exogenous VIP produced relaxation of the CM which was partly TTX-sensitive. L-NNA also partly blocked the relaxation. The effect of L-NNA was not significantly different from that produced by TTX or TTX and L-NNA, indicating that NO is responsible for the nervemediated component of the relaxation to VIP. Similarly, relaxation produced by 5-HT in the canine terminal ileum and ileocolonic junction is inhibited by L-NNA (Bogers et al., 1991) and VIP- and electrically evoked relaxations of the opossum internal anal sphincter are blocked by L-NNA (Rattan and Chakder, 1992). However, the relaxation produced by VIP in the canine lower oesophageal sphincter (De Man et al., 1991), guinea-pig taenia coli (Grider et al., 1992), and pig, canine and human gastric fundus strips (Lefebvre, Smits and Timmermans, 1995; Bayguinov et al., 1999; Tonini et al., 2000) are not affected by L-NNA. Thus, the degree to which NO mediates relaxation produced by VIP and other transmitters varies according to species and region of intestine.

Electrophysiological effects of NO on neurons and enteric smooth muscle cells In the canine jejunum CM, a single brief pulse of NO produces transient membrane hyperpolarization and relaxation (Stark, Bauer and Szurszewski, 1991). When NO is applied repetitively, the initial hyperpolarization runs down, although spontaneous phasic contractions are still inhibited (Irons, Stark and Szurszewski, 1992). This suggests that NO relaxes the intestine by two mechanisms: one involving hyperpolarization and the other independent of hyperpolarization. Transmural electrical stimulation of guinea-pig ileum in the presence of atropine and SP desensitization produces a fast and slow IJP in the CM (Niel, Bywater and Taylor, 1983b; Bywater and Taylor, 1986). The fast IJP is apamin-sensitive, whereas the slow IJP is apamin-resistant (Niel, Bywater and Taylor, 1983b). L-NNA, a NOS inhibitor, has no effect on the fast IJP (Watson, Bywater and Taylor, 1992), but blocks the slow IJP and this effect is partly reversed by L-arginine (He and Goyal, 1992; Lyster, Bywater and Taylor, 1992), suggesting that NO is involved. The slow IJP is also blocked by VIP10–28 (Crist, He and Goyal, 1992). Furthermore, He and Goyal (1992) demonstrated that the slow hyperpolarizing response to VIP in CM is inhibited by L-NNA, suggesting that NO mediates the response to VIP. Thus, VIP released from inter- or motor neurons may stimulate NO synthesis in inhibitory motor neurons which in turn relaxes and hyperpolarizes the CM. This mechanism has also been proposed for the generation of slow IJPs in circular muscle strips of murine gastric fundus (Mashimo et al., 1996).

Role of endogenous NO The hypothesis that NO may serve as an inhibitory transmitter in the gastrointestinal tract is based on a great deal of evidence accumulated in the last decade. These findings generally indicate that neurogenic relaxations evoked by electrical field stimulation from gastrointestinal specimens of several animal species, including humans, depend on

232

INNERVATION OF THE GASTROINTESTINAL TRACT

a combination of NO and VIP release. Compared to VIP, NO is preferentially released at low frequency of stimulation (Tonini et al., 2000). NO seems to play an important role in maintaining tonic neural inhibition and compliance of the circular muscle. In fact, NO could be viewed as the primary inhibitory transmitter initiating accommodation in the guinea-pig ileum, rabbit colon and human gastric fundus (Waterman, Costa and Tonini, 1994; Ciccocioppo et al., 1994; Tonini et al., 2000). Apart from its action as an inhibitory transmitter at a postjunctional site, NO may also act as a modulator of excitatory transmission. Several reports now indicate that NO may inhibit cholinergic as well as noncholinergic transmitter release in several models (Knudsen and Tøttrup, 1992; Wiklund et al., 1993; Hryorenko, Woskowska and Fox-Threlkeld, 1994; Yunker and Galligan, 1996; Holzer-Petsche et al., 1996; Tonini et al., 2000). ADENOSINE TRIPHOSPHATE The presence of ATP in all cells makes it difficult to identify which neurons use ATP as a transmitter, although a recent study (McConalogue et al., 1996) provided evidence that ATP and its major metabolites are released from a neuronal source in the guinea-pig taenia coli. In other studies (Matsuo et al., 1997) ATP seems to derive mainly from smooth muscle cells. The best evidence that ATP may have a transmitter role in the enteric nervous system comes from studies using synaptosomes from LM-MP preparations. In these preparations, ATP is released in response to stimulation by high potassium, veratridine, ACh, or 5-HT (White, 1982; White and Leslie, 1982; White and Al-Humayyd, 1983; Al-Humayyd and White, 1985). Since the release is calcium-dependent, it is likely to represent a transmitter rather than a metabolic pool (White and Leslie, 1982). ATP is also released from guinea-pig taenia coli in response to electrical field stimulation. Some of this release is TTX-sensitive (Rutherford and Burnstock, 1978; McConalogue et al., 1996) and is therefore neuronal in origin. Relaxations in this preparation, like in many other gastrointestinal tissues, are mediated by P2Y purinoceptors modulating a subset of calcium-activated potassium channels (Kong, Koh and Sanders, 2000). Nevertheless, most of the measurable ATP is not neuronal in origin, since ATP can be released by smooth muscle relaxants such as papaverine and nitroglycerine (Kuchii, Miyahara and Shibata, 1973). The use of synaptosomes overcomes this problem. A further complication, however, is that ATP release from myenteric synaptosomes is reduced by 50–90% when the guinea-pigs are pretreated in vivo with 6-hydroxydopamine, suggesting that the majority of released ATP comes from noradrenergic nerve terminals of extrinsic nerves (White and Al-Humayyd, 1983; Al-Humayyd and White, 1985). ATP depolarizes S neurons, mimicking slow EPSPs, and hyperpolarizes AH neurons, mimicking slow IPSPs (Katayama and Morita, 1989). These effects are likely to be mediated by calcium-dependent potassium channel opening and closing, respectively (Katayama and Morita, 1989). Recent data also indicate that, in addition to ACh, ATP contributes to fast synaptic transmission through P2X receptors in myenteric neurons (Galligan and Bertand, 1994; LePard, Messori and Galligan, 1997). This transmission, which is antagonized by the P2X receptor antagonist PPADS (Lambrecht et al., 1992), predominantly occurs between descending interneurons and inhibitory motor neurons in descending inhibitory reflex pathways of guinea-pig ileum (Bian et al., 2000). Other

PHARMACOLOGY OF THE ENTERIC NERVOUS SYSTEM

233

electrophysiological evidence supports convincingly a role for endogenous ATP as an inhibitory transmitter. Single pulse transmural stimulation of the guinea-pig ileum produces a fast IJP which is TTX- and apamin-sensitive (Niel, Bywater and Taylor, 1983b; Crist, He and Goyal, 1992). This IJP, like the fast hyperpolarization induced by exogenous ATP, is antagonized by Reactive blue 2 and desensitization to α,β-methyleneATP, but is unaffected by VIP10–28 (Crist, He and Goyal, 1992). It is likely therefore, that ATP or a related nucleotide mediates the apamin-sensitive component of non-adrenergic, non-cholinergic inhibitory transmission. More recently, ATP was found to mediate fast relaxations or IJPs in response to electrical stimulation in the circular muscle of guinea-pig proximal colon and internal anal sphincter, the fast component of IJPs in the circular muscle of murine gastric fundus, and the IJPs in the circular muscle of porcine ileum via mechanisms sensitive to apamin, desensitization to α,β-methyleneATP, Reactive blue 2, and suramin, a non-selective P2 receptor antagonist (Maggi and Giuliani, 1996; Rae and Muir, 1996; Mashimo et al., 1996; Fernandez et al., 1998). Desensitization to ATP does not significantly alter peristalsis in the guinea-pig ileum (Weston, 1973) and rabbit colon (Tonini et al., 1982). It is therefore not clear whether endogenous ATP plays a role in peristalsis.

PHARMACOLOGY OF SEROTONERGIC TRANSMISSION Serotonin (5-HT) is present in enterochromaffin cells and neurons in the guinea-pig small intestine (Furness and Costa, 1982). Non-neuronally located 5-HT can be released in a calcium-dependent manner in response to increased intraluminal pressure (Bülbring and Lin, 1958; Bülbring and Crema, 1959; Schwörer, Racké and Kilbinger, 1987) or to vagal stimulation (Furness and Costa, 1987). 5-HT is localized in descending interneurons which project to other myenteric ganglia and/or submucous ganglia (Costa et al., 1982; Furness and Costa, 1982; Brookes, Steele and Costa, 1992). These neurons are also immunoreactive for cholineacetyl transferase. Ultrastructural studies indicate that 5-HT neurons form close contacts with other 5-HT neurons and with non 5-HT-immunoreactive Dogiel type I and type II neurons (Young and Furness, 1991). No obvious classification of enteric neurons has been found based on the number of 5-HT inputs they receive (Young and Furness, 1991). Release of 5-HT from these neurons can be demonstrated in response to depolarizing stimuli (Holzer and Skofitsch, 1984). 5-HT has a bewildering range of effects in the intestine, largely due to the presence of multiple receptor subtypes which appear to be present on several classes of myenteric neurons and smooth muscle cells (Table 6.2). The following section summarizes recent studies in which more selective agonists and antagonists have been used. Earlier work has been reviewed on numerous occasions (e.g. Costa and Furness, 1979; Furness and Costa, 1982, 1987). EFFECTS OF 5-HT ON NEUROMUSCULAR PREPARATIONS 5-HT contracts the ileal LM, producing a biphasic concentration-response curve (Buchheit et al., 1985; Eglen et al., 1990; Kilbinger and Wolf, 1992). The first phase has an EC50 for

8-OH-DPAT, buspirone

(±)WAY100635

Gi/o

Selective agonists

Selective antagonists

Effectors

Ketanserin, SB200646 MDL100907 SB204741

Gq/11

GR127935 (5-HT1B/D), SB216641 (5-HT1Bselective), BRL15572 (5-HT1D-selective)

Gi/o

Gq/11

α-Me-5-HT, BW723C86

α-Me-5-HT

sumatriptan

Longitudinal smooth muscle

5-HT2B

Contraction

Smooth muscle

5-HT2A

Contraction

Facilitation of Peristalsis, contraction

Myenteric neurons, Circular smooth muscle2

5-HT1B/D

Ligand-gated ion channel

Ondansetron granisetron

2-Me-5-HT, m-CPBG

Enhanced transmitter release

Myenteric neurons

5-HT3

Gs

Slow EPSP

Myenteric neurons

5-HT1P1

Gs

Go

5-HTP-DP

5-Carboxamido- Hydroxylated tryptamine indalpines

Relaxation

Smooth muscle

5-HT7

SB204070, Methiotepin, GR113808, GR125487 metergoline, SB258719

Renzapride, tegaserod (HTF919), ML10302, BIMU8

Enhanced transmitter release, relaxation

Myenteric neurons, smooth muscle

5-HT4

5-HT1P receptors are not included in the IUPHAR nomenclature of 5-HT receptors; 2 Sumatriptan is the only 5-HT1B/D receptor agonist so far tested and only fragmentary data exist on the possible involvement of these receptor subtypes in the gut.

Reduced transmitter release

Functional response

1

Myenteric neurons

Distribution in the gut

5-HT1A

TABLE 6.2 Synopsis of major 5-HT receptor subtypes in the gut.

234 INNERVATION OF THE GASTROINTESTINAL TRACT

PHARMACOLOGY OF THE ENTERIC NERVOUS SYSTEM

235

5-HT of 15 nM, indicating the involvement of a high affinity receptor (Eglen et al., 1990; Buchheit et al., 1985). This phase is selectively blocked by high concentrations of ICS 205–930 and mimicked by 5-MOT, BIMU 1, BIMU 8, cisapride and DAU 6236 (all 5-HT4 receptor agonists; Schuurkes et al., 1985; Eglen et al., 1990; Kilbinger and Wolf, 1992; Rizzi et al., 1992). These characteristics suggest that 5-HT is acting via the 5-HT4 receptor subtype. The first phase is also blocked by atropine, TTX, morphine and [D-Pro4, 7,9 D-Trp ]SP4–11 and augmented by physostigmine (Buchheit et al., 1985; Schuurkes et al., 1985). In the presence of antagonists to block all but the 5-HT4 receptor subtype, 5-HT and 5-MT enhance electrically evoked ACh release and associated twitch, and increase basal ACh release (Kilbinger and Pfeuffer-Friederich, 1985; Craig and Clarke, 1990; Fox and Morton, 1990; Kilbinger and Wolf, 1992). Cisapride also increases electrically evoked ACh release and the associated contraction (Taniyama et al., 1991). Cisapride has no effect on the contraction produced by exogenous ACh (Schuurkes et al., 1985). Consequently, the 5-HT4 receptor is likely to be located on LM motor neurons, where its activation stimulates the release of ACh and SP in a TTX-sensitive manner. The second phase of the concentration-response curve has an EC50 for 5-HT of 1.3 mM, indicating the involvement of a low affinity receptor (Buchheit et al., 1985; Eglen et al., 1990). This phase is blocked by low concentrations of ICS 205–930 (Buchheit et al., 1985; Eglen et al., 1990), suggesting that the 5-HT3 receptor subtype is involved. The contraction is inhibited by TTX, morphine, [D-Pro4, D-Trp7,9]SP4–11 and atropine and enhanced by physostigmine (Buchheit et al., 1985; Fox and Morton, 1989; Eglen et al., 1990). Activation of 5-HT3 receptors enhances basal ACh outflow (Kilbinger and Wolf, 1992). Thus, activation of 5-HT3 receptors may stimulate ACh and SP release from LM motor neurons. 5-HT also stimulates calcium-dependent, TTX-sensitive γ-aminobutyric acid (GABA) release through an action at 5-HT3 receptors (Shirakawa et al., 1989). At high concentrations (>10 µM), 5-HT acts at 5-HT2 receptors (formerly called D receptors) which are present on the smooth muscle and stimulate contraction (Engel et al., 1984; Buchheit et al., 1985; Richardson et al., 1985; Fox and Morton, 1989; Table 6.2). 5-HT produces a TTX-sensitive relaxation of precontracted LM in the presence of antagonists to 5-HT2, 5-HT3 and 5-HT4 receptors (Bill, Dover and Rhodes, 1990; Elswood and Bunce, 1992). This action is mimicked by carboxamidotryptamine and sumatriptan, suggesting that 5-HT1A and 5-HT1D receptors are involved (Bill, Dover and Rhodes, 1990; Elswood and Bunce, 1992). This relaxation may be due to inhibition of excitatory LM motor neurons, since activation of 5-HT1A receptors reduces the amplitude of the tachykinin-mediated twitch by a prejunctional mechanism (Galligan, 1992). It has also been reported that 5-HT1A agonists inhibit electrically evoked ACh outflow (Fozard and Kilbinger, 1985; Kilbinger and Wolf, 1992; Dietrich and Kilbinger, 1996) and reduce cholinergic twitch responses (Fozard and Kilbinger, 1985; Craig and Clarke, 1990). However, these effects may be due in part to an action of the agonists at histamine and muscarinic receptors (Galligan, 1992). 5-HT1A receptors also mediate inhibition of GABA release (Shirakawa et al., 1989). Some authors suggest caution in considering the involvement of 5-HT1A receptors in mediating the effects of 5-HT in the gut on the basis of data obtained with 8-OH-DPAT, a compound previously considered a selective 5-HT1A receptor agonist. After the discovery that 5-HT7 receptors mediate relaxation probably by a direct action on the smooth muscle

236

INNERVATION OF THE GASTROINTESTINAL TRACT

(Carter et al., 1995; Prins et al., 1999) and that 8-OH-DPAT is also a partial agonist at the 5-HT7 receptor, some functional data should probably be reconsidered (Vanhoenacker, Haegeman and Leysen, 2000). Agonists at 5-HT4 receptors (BIMU 8, cisapride and 5-HT) cause a concentrationdependent increase in the amplitude of submaximal cholinergic twitch responses. This effect is inhibited by micromolar concentration of ICS 205–930 and DAU 6285 (Tonini et al., 1992a), confirming that 5-HT4 receptors are involved. In addition to enhancing electrically evoked twitch responses, cisapride enhances the ascending reflex contraction produced by balloon distension of the ileum (Tonini, 1992). Since the drug was added to the anal compartment of a partitioned organ bath, which contains the sensory neurons and some interneurons involved in the reflex, the effect of cisapride may be due to enhancement of transmission between these neuronal classes. Electrophysiological studies (see below) indicate that cisapride enhances nicotinic transmission. The ability of 5-HT to enhance peristalsis has been known for many years (Bülbring and Crema, 1958; Bülbring and Lin, 1958). The receptor involved in this process has been the subject of several investigations. Renzapride (a 5-HT4 receptor agonist) and 5-HT initiate peristalsis in the guinea-pig ileum and this effect is blocked by high concentrations of ICS 205–930 (Craig and Clarke, 1991). Using a slightly different preparation, Tonini, Galligan and North (1989) and Rizzi et al. (1992) showed that the frequency of emptying was increased by cisapride, BIMU 1, BIMU 8 and DAU 6236; on close inspection of the recordings, the compliance of the intestinal wall also appears to have been decreased. These effects were inhibited by high concentrations of ICS 205–930, but not by ondansetron, indicating the involvement of 5-HT4 receptors. There was no effect of 5-HT4 receptor agonists on LM muscle contraction during the preparatory phase or on the maximal ejection pressure (Tonini, Galligan and North, 1989a; Rizzi et al., 1992). 5-HT4 receptor agonists could conceivably reduce the threshold volume required to trigger emptying of the intestine and the compliance of the intestinal wall, by stimulating excitatory motor neuron pathways or inhibiting inhibitory motor neuron pathways (Waterman, Costa and Tonini, 1992). The former hypothesis seems more likely, since 5-HT4 receptor agonists can enhance neuronal release of ACh and SP (Kojima and Shimo, 1996), while the inhibitory effects of 5-HT4 receptor agonists on an inhibitory pathway is less well documented. However, other mechanisms may be involved: first, enhancement of ACh and SP release from LM-MP preparations does not necessarily mean that these agonists enhance transmitter release from CM motor neurons. Second, enhancement of the cholinergic twitch response in CM by 5-HT4 agonists could equally well be due to removal of an inhibitory input to the muscle as to stimulation of excitatory motor neurons. Therefore, even though the net effect is prokinetic, it is unclear whether the underlying mechanism is one of excitation or inhibition. Interestingly, there are animal (Grider, Kuemmerle and Jin, 1996) and human (FoxxOrenstein, Kuemmerle and Grider, 1996) data suggesting that 5-HT released by mucosal stimulation initiates a peristaltic reflex by activating 5-HT4 receptors on sensory neurons containing calcitonin gene-related peptide (CGRP). These effects are mimicked by mucosal application of selective 5-HT4 receptor agonists (prucalopride and tegaserod: Grider, Foxx-Orenstein and Jin, 1998). Experimental evidence for this mechanism in humans is so far limited to the small bowel.

PHARMACOLOGY OF THE ENTERIC NERVOUS SYSTEM

237

Important differences may exist among different species/different gut levels. 5-HT4 receptor-mediated prokinesia may result from increased release of ACh (and tachykinins) from excitatory neurons and may operate in human small bowel and stomach (SakuraiYamashita et al., 1999; Schuurkes et al., 1991), whereas this pathway does not seem to operate in the human colonic circular muscle (Burke and Sanger, 1988; Burleigh and Trout, 1985). In addition, it should be noted that, in contrast with what observed in the one of the most widely used models [the guinea-pig colon, where neuronal 5-HT4 receptors mediate contractile responses that are mainly cholinergic in nature (Briejer et al., 1993b; Wardle and Sanger, 1993)], human colonic circular muscle strips are endowed with 5-HT4 receptors located on smooth muscle cells, where they mediate relaxation (McLean and Coupar, 1996; Prins et al., 2000b; Sakurai-Yamashita et al., 2000; Tam et al., 1995). A recent report (Prins et al., 2000a) suggests the presence of 5-HT4 receptors on cholinergic neurons supplying the longitudinal muscle in the human colon. Figure 6.2 illustrates the distribution of 5-HT4 receptors on effector cells and intrinsic neuronal pathways (i.e. myenteric plexus), which mediate propulsive activity in the gastrointestinal tract. 5-HT1B/D receptors are now emerging as possible targets of 5-HT action in the gut (Borman and Burleigh, 1997; Coulie et al., 1997; Coulie et al., 1999; Tack et al., 2000). However, sumatriptan is the only 5-HT1B/D receptor agonist so far tested and, to the best of our knowledge, no formal assessment of the involvement of this receptor subtype by the use of selective antagonists has been performed (Cipolla et al., 2001). A study carried out

Figure 6.2 Modulation of intestinal motility by serotonergic 5-HT4 receptors. These receptors, which are distributed at multiple neuronal sites, have an excitatory effect (+) on enteric neurons leading to transmitter release and facilitation of propulsion. 5-HT4 receptors are also located on smooth muscle cells where they produce relaxation (–).

238

INNERVATION OF THE GASTROINTESTINAL TRACT

in the human ileum suggested that, in the circular muscle, 5-HT-induced contraction is mediated via a receptor of the 5-HT1D subtype, whereas a receptor of the 5-HT2B subtype mediates the contractile response to 5-hydroxytryptamine of longitudinal muscle layer (Borman and Burleigh, 1997). Selective 5-HT1D receptor antagonists were not available at the time of this study to confirm the hypothesis. Finally, preliminary data suggest that 5-HT1B/D receptors are involved in mediating the facilitatory effect of sumatriptan on gastric accommodation in an in vivo canine model (De Ponti, 2000). ELECTROPHYSIOLOGICAL EFFECTS OF 5-HT ON NEURONS The electrophysiological effects of 5-HT on enteric neurons has been reviewed by Galligan (1995). 5-HT1A receptors are located on the cell bodies of AH neurons and on nerve terminals of cholinergic neurons (Galligan et al., 1988). Agonists at these receptors produce membrane hyperpolarization in AH neurons, but not S neurons, due to an increase in potassium conductance (Galligan et al., 1988; Galligan and North, 1991). These receptors also mediate presynaptic inhibition of ACh release which is recorded electrophysiologically as an inhibition of fast EPSPs (North et al., 1980; Galligan et al., 1988; Galligan and North, 1991). The amplitude of slow EPSPs is also reduced (Galligan et al., 1988; Galligan and North, 1991). 5-HT1P receptors are present on AH neurons where they mediate a slow depolarization associated with an increase in input resistance (Mawe, Branchek and Gershon, 1986). These receptors have been reported to be specifically antagonized by dipeptides of 5-hydroxytryptophan (5-HTP-DP) (Takaki et al., 1985a; Mawe, Branchek and Gershon, 1986). Activation of 5-HT3 receptors by 2-methyl-5-HT produces a fast depolarization in AH neurons which is associated with a decrease in input resistance, due to the opening of a cation channel (Mawe, Branchek and Gershon, 1986; Surprenant and Crist, 1988; Derkach, Surprenant and North, 1989; Mawe, Branchek and Gershon, 1989). This effect is blocked by low concentrations of ICS 205–930 (Mawe, Branchek and Gershon, 1986). Cisapride increases the amplitude of fast EPSPs through a presynaptic action, since it has no effect on the fast depolarization produced by exogenous ACh. This is likely to be due to stimulation of ACh release (Tonini, Galligan and North, 1989a). Cisapride had no effect on the resting membrane potential in several studies (Mawe, Branchek and Gershon, 1989; Tonini, Galligan and North, 1989a), but was reported to produce depolarization or hyperpolarization in a minority of cells (Nemeth et al., 1985; Nemeth and Gullikson, 1989). Cisapride does not alter non-cholinergic slow EPSPs in AH neurons (Tonini, Galligan and North, 1989a). ROLE OF ENDOGENOUS 5-HT Although 5-HT is present in and can be released from enterochromaffin cells and neurons in the guinea-pig ileum, it is difficult to prove its physiological role. 5-HT3 receptors are ligand-gated cation channels and mediate rapid depolarization of myenteric neurons (Surprenant and Crist, 1988; Derkach, Surprenant and North, 1989; Mawe, Branchek and Gershon, 1989). It is conceivable therefore, that at least some of the

PHARMACOLOGY OF THE ENTERIC NERVOUS SYSTEM

239

fast EPSPs recorded in the myenteric plexus are due to the action of an endogenous 5-HT-like substance on 5-HT3 receptors rather than the action of ACh at nicotinic receptors. This possibility has received little attention. However, Tonini, Galligan and North (1989a) found that the 5-HT3 receptor antagonists, ICS 205–930 and ondansetron, had no effect on the amplitude of fast EPSPs. It has been proposed that 5-HT mediates some or all slow EPSPs in the myenteric plexus. The latter hypothesis is unlikely, since lesioning 5-HT inputs to myenteric ganglia does not significantly alter the probability of recording slow EPSPs (Bornstein et al., 1984). However, 5-HT may be responsible for a proportion of slow EPSPs. Takaki et al. (1985b) and Mawe, Branchek and Gershon, (1986, 1989) reported that electrically evoked slow EPSPs could be blocked by 5-HTP-DP and renzapride but not by low concentrations of ICS 205–930. Some slow EPSPs may therefore be mediated by the action of 5-HT-like substances at 5-HT1P receptors. Since exogenous 5-HT can elicit fast and slow EPSPs, it is reasonable to propose that endogenous 5-HT might be responsible for non-cholinergic transmission. Such transmission has been shown to occur in the ascending excitatory reflex response to distension (Holzer, 1989; Tonini and Costa, 1990). However, endogenous 5-HT-like substances acting at 5-HT1, 5-HT1P, 5-HT3 and 5-HT4 receptors do not mediate the non-cholinergic component of ganglionic transmission in this reflex (Tonini et al., 1992b). Nevertheless, it is not known whether 5-HT may play a role in modulating cholinergic transmission in this reflex pathway through an action at presynaptic receptors. Endogenous 5-HT may play a role in the ascending excitatory reflex response to mucosal distortion, rather than distension. Bülbring and Crema (1958) and Bülbring and Lin (1958) originally proposed that 5-HT is involved in sensitizing mucosal processes of sensory neurons in the intestine. This was based on these pieces of evidence: first, 5-HT is present in enterochromaffin cells which are near the mucosal processes of presumed sensory neurons. Second, 5-HT could be released into the lumen of the intestine and this release is increased by increasing intraluminal pressure. Third, mucosal application of 5-HT stimulates peristalsis. More recently, Kirchgessner, Tamir and Gershon (1992) reported that stimulation of the mucosa with puffs of nitrogen gas results in c-fos expression in some myenteric and submucous neurons. Expression of c-fos is inhibited by dipeptides which are believed to inhibit 5-HT1P receptors. These investigators also demonstrated that antibodies to 5-HT1P receptors label submucosal calbindin-immunoreactive neurons, the proposed sensory neurons. The investigators therefore suggested that the response to nitrogen puffs may be due to release of 5-HT from enterochromaffin cells, which stimulates sensory neurons through an action at 5-HT1P receptors. However, these results await to be confirmed by other research groups. Furthermore, even if the conclusion drawn by Kirchgessner, Tamir and Gershon (1992) is correct, it does not necessarily mean that this mechanism operates during peristalsis. In fact, studies by other investigators, provide evidence that 5-HT is not necessarily involved. Schwörer, Racké and Kilbinger (1989) measured 5-HT release from enterochromaffin cells in the guineapig ileal mucosa in response to increased pressure in the lumen. If the intestine was infused with fluid at 22° C instead of 37° C, release of 5-HT was inhibited. However, there were no concomitant changes in peristalsis, thereby suggesting that 5-HT released from mucosal enterochromaffin does not play a significant role in peristalsis. Furthermore,

240

INNERVATION OF THE GASTROINTESTINAL TRACT

antagonists at 5-HT3 and 5-HT4 receptors do not appear to have any effect on peristalsis (Craig and Clarke, 1991), suggesting that endogenous 5-HT has little, if any, role in peristalsis through an action at these receptors. However, the facilitatory effect of mucosally applied 5-HT to reduce the peristaltic threshold in the guinea-pig ileum is probably mediated by 5-HT3 receptors located on the mucosal side (Tuladhar, Kaisar and Naylor, 1997), and in particular on the terminals of intrinsic primary afferent neurons (Bertrand et al., 2000).

PHARMACOLOGY OF AMINO ACIDS AS ENTERIC NEUROTRANSMITTERS γ-AMINOBUTYRIC ACID (GABA) The presence of GABA in the guinea-pig small intestine has been demonstrated by immunohistochemistry (Jessen, Hills and Saffrey, 1986; Saito and Tanaka, 1986; Pompolo and Furness, 1990) and autoradiography (Krantis and Kerr, 1981). GABA is contained in short inhibitory CM motor neurons and LM motor neurons (Krantis, Kerr and Dennis, 1986; Saito and Tanaka, 1986; Hills, Jessen and Mirsky, 1987; Furness et al., 1989b). Myenteric neurons also express the synthetic enzyme for GABA, glutamic acid decarboxylase (GAD) (Miki et al., 1983; Williamson et al., 1995) and a high affinity uptake mechanism for the neurotransmitter (Krantis and Kerr, 1981; Saffrey et al., 1983; Krantis, Kerr and Dennis, 1986). GABA is released from myenteric synaptosomes of the guinea-pig small intestine in response to depolarization by high potassium and veratridine in a calcium-dependent manner (Yau and Verdun, 1983). The release of GABA from LM-MP preparations of the guinea-pig small intestine has also been demonstrated during either high potassium or electrically-evoked stimulation (Taniyama et al., 1983a; Wiley, Lu and Owyang, 1991). Somatostatin, SP, 5-HT, neurotensin and cholecystokinin (CCK) can stimulate GABA release from LM-MP preparations (Kerr and Krantis, 1983; Taniyama et al., 1983a; Tanaka and Taniyama, 1985; Nakamoto, Tanaka and Taniyama, 1987; Sano, Taniyama and Tanaka, 1989; Shirakawa et al., 1989; Takeda et al., 1989). This release is blocked by TTX and low Ca2+ concentrations. GABA release can be inhibited presynaptically by agonists at M1 receptors and α2-adrenoceptors (Hashimoto, Tanaka and Taniyama, 1986). Immunohistochemical evidences have been provided for the presence of both GABAA and GABAB receptors in the enteric nervous system (Krantis et al., 1995; Nakajima et al., 1996). In particular, GABAA receptor immunoreactivity has been detected on the soma of myenteric ganglion cells in the rat distal colon (Krantis et al., 1995). In addition, molecular biological investigations have demonstrated an abundant and widespread distribution of different GABAA receptor subunit mRNAs in the rat’s small and large intestines (Zeiter, Li and Broussard, 1996; Poulter et al., 1999). GABAB immunoreactivity was found on the soma of neurons in both submucosal and myenteric ganglia at different levels in the rat gut (Nakajima et al., 1996).

PHARMACOLOGY OF THE ENTERIC NERVOUS SYSTEM

241

Effects of GABA on neuromuscular preparations GABA is not effective on nerve-free preparations. In the LM of the guinea-pig small intestine, GABA produces a transient contraction followed by a relaxation (Inouye et al., 1960; Krantis et al., 1980; Giotti et al., 1983; Taniyama et al., 1983b). Both effects are blocked by TTX (Krantis et al., 1980), indicating that GABA is not directly acting on the LM. GABA-induced contractions are mimicked by muscimol and associated with an increase in ACh release (Kleinrok and Kilbinger, 1983; Tonini et al., 1987). The enhancements both in tension and in ACh release are blocked by bicuculline, picrotoxin, hyoscine and atropine, suggesting the involvement of GABAA receptors. Hexamethonium reduced the phasic GABAA-mediated contraction by 20% in one study (Tonini et al., 1987), suggesting that GABA is capable of stimulating cholinergic interneurons. GABA has also been shown to enhance ACh release from synaptosomes isolated from LM-MP preparations of the guinea-pig small intestine, suggesting that the stimulatory effect of GABA is due, at least in part, to an action on nerve terminals (Yau and Verdun, 1983). Tonini et al. (1987) found that GABA-induced contractions were also reduced by [D-Pro4, D-Trp7,9]SP4–11, indicating that GABAA receptors are also involved in the stimulation of tachykinin release from motor neurons and interneurons in the LM pathways of the guinea-pig ileum. The presence of GABAA receptors is also reported on inhibitory neurons and their stimulation induces relaxant responses (Minocha and Galligan, 1993; Boeckxstaens et al., 1991). Several reports are available to suggest that GABA may also inhibit both electricallyand chemically-evoked contractions by acting prejunctionally to reduce ACh release, in a bicuculline- and picrotoxin-insensitive manner (Hobbiger, 1958a; Hobbiger, 1958b; Florey and McLennan, 1959; Inouye et al., 1960; Kleinrok and Kilbinger, 1983; Ong and Kerr, 1983). This effect is mimicked by homotaurine (3-APS) and baclofen, but not by muscimol (Kleinrok and Kilbinger, 1983; Hills et al., 1989). δ-Aminovaleric acid (DAVA), but not bicuculline, reversed the action of both GABA and baclofen (Ong and Kerr, 1983). On the whole, these observations suggest that inhibitory effects of GABA on contractile responses are due to inhibition of ACh release from motor neurons mediated via GABAB receptors. GABA does not have any significant effect on electrically stimulated contraction of the CM, however it reduces the frequency of spontaneous CM contractions (Ohkawa, 1987).

Electrophysiological effects of GABA on neurons Ionophoretic application of GABA to myenteric neurons of the guinea-pig ileum determines a depolarizing response of AH neurons in a bicuculline-sensitive manner. This effect is associated with an increase in chloride conductance and is prone to desensitization suggesting the involvement of GABAA receptors (Grafe, Galvan and Mayer, 1979; Cherubini and North, 1984a). Data from studies in acutely isolated myenteric plexus preparations and in myenteric neurons maintained in primary culture suggest that GABAA-mediated responses are potentiated by a number of drugs acting at benzodiazepine and barbiturate binding sites (Cherubini and North, 1985a; Bertrand and Galligan, 1992; Zhou and Galligan, 2000). Electrophysiological and immunohistochemical investigations

242

INNERVATION OF THE GASTROINTESTINAL TRACT

have recently demonstrated that myenteric neurons expressing calbindin immunoreactivity, a marker for AH neurons, have a high density of GABAA receptors (Zhou and Galligan, 2000). In addition to rapid depolarizing responses induced by ionophoretic application of GABA, superfusion of the agonist produces depolarizations in AH myenteric neurons which are mimicked by baclofen and are insensitive to bicuculline. This effect is associated with a smaller increase in conductance than that evoked by ionophoresis of GABA and does not desensitize. On the whole these observations suggest that this latter effect is mediated via activation of GABAB receptors probably located on nerve processes rather than on neuronal somata (Cherubini and North, 1984a). In isolated myenteric ganglia of the guinea pig ileum, GABA does not alter the resting membrane potential or conductance of S neurons (Cherubini and North, 1984b). However, GABA reduces the amplitude of evoked fast EPSPs without altering the response to ionophoretic application of ACh (Grafe, Galvan and Mayer, 1979; Cherubini and North, 1984b). This effect is mimicked by baclofen and resistant to bicuculline, suggesting the involvement of a GABAB receptor in the response (Cherubini and North, 1984b). In the same experimental model, GABAB receptors have also been shown to mediate the inhibition of slow EPSPs (Cherubini and North, 1984b). Thus, inhibition of both fast and slow EPSPs in S myenteric neurons is accomplished through GABAB receptors, which have been suggested to be located presynaptically to inhibit the release of both ACh and of a non-cholinergic excitatory transmitter.

Role of endogenous GABA Pharmacological studies have been described on the possible involvement of GABA receptors in the modulation of peristalsis with contrasting results. Indeed, early studies reported a GABA-mediated inhibition of peristalsis in a picrotoxin-sensitive manner (Hobbiger, 1958b; Inouye et al., 1960), whereas in a later study Schwörer and Kilbinger, (1988) could not demonstrate any significant effect of exogenous GABA on this functional parameter. In guinea-pig ileum, the amplitude of cholinergic twitches in the LM and the efficiency of peristalsis were enhanced by bicuculline (Schwörer and Kilbinger, 1988; Tonini et al., 1989b). However, these effects were not mimicked by picrotoxin or SR 95531, which also act on the GABAA receptor complex. These evidences might be explained on the basis of a non-specific effect of bicuculline (Tonini et al., 1989b), which might explain the discrepancy between an apparent effect of endogenous GABA on peristalsis (as deduced from the effect of bicuculline) and the lack of effect of exogenous GABA. However, in the guinea-pig colon, bicuculline has been shown to enhance the efficiency of peristalsis in normal but not in GABA-desensitized preparations, suggesting an action of bicuculline entirely dependent on the availability of functional GABA receptors (Frigo et al., 1987). In the same study bicuculline enhanced both cholinergic excitatory and non-adrenergic, non-cholinergic inhibitory responses to transmural stimulation and the release of ACh. These effects were completely abolished after desensitization with GABA (Frigo et al., 1987).

PHARMACOLOGY OF THE ENTERIC NERVOUS SYSTEM

243

SP- and neurotensin-induced stimulation of ACh release is partly reduced by bicuculline (Tanaka and Taniyama, 1985; Nakamoto, Tanaka and Taniyama, 1987). The authors concluded that endogenous GABA, which is released in response to SP and neurotensin, acts on GABAA receptors to stimulate ACh release. However, it seems likely that these effects are also due to non-specific actions of bicuculline, since the concentrations used in these studies (1–30 µM) were the same as those employed by Tonini et al. (1989b). Other evidence for non-specific effects of bicuculline has previously been reported in isolated rat atria, guinea-pig ileum and mouse vas deferens (Bartolini et al., 1985). Somatostatin-induced relaxation of the LM and the associated inhibition of ACh release, are blocked by phaclofen (Takeda et al., 1989). Somatostatin can stimulate GABA release (Takeda et al., 1989); therefore these results are likely to be due to a somatostatininduced GABA release which acts at GABAB receptors to inhibit ACh release and LM contraction. Since GABA is present only in motor neurons, and not in sensory neurons or interneurons, these results suggest that GABA may be released at the neuroeffector junction and act on local nerve terminals to inhibit (via GABAB receptors) ACh release. Exogenous GABA has a similar action on cholinergic nerve terminals in the myenteric plexus. Spontaneous neurogenic contractions of the guinea-pig ileum LM can be inhibited by atropine, bicuculline, picrotoxin, RU 5135, desensitization to GABA, 3-mercaptopropionic acid and cortisol (Ong and Kerr, 1984, 1989). Together with other results, these findings suggest that endogenous GABA is released from LM motor neurons and acts on local nerve terminals via GABAA receptors to stimulate ACh release. Immunohistochemical evidences have been provided for the presence also of inhibitory neurons co-expressing GABA and NOS immunoreactivity in the guinea-pig small intestine (Williamson, Pompolo and Furness, 1996). EXCITATORY AMINO ACIDS The concept that glutamate, the major excitatory neurotransmitter in the central nervous system (Ozawa, Kamiya and Tsuzuki, 1998), plays a role as a neurotransmitter also in the enteric nervous system is gaining increased acceptance (Liu et al., 1997; Sinsky and Donnerer, 1998). Immunoreactivity for glutamate has been detected in subsets of neurons in both the myenteric and submucosal plexus of the rat and guinea-pig ileum (Liu et al., 1997). In particular, glutamate immunoreactivity was found in enteric cholinergic neurons colocalized with substance P and/or calbindin (Liu et al., 1997). Immunohistochemical evidence has also been provided for glutamate storage into terminal axonal varicosities of myenteric neurons in the guinea-pig ileum (Liu et al., 1997). At this level, L-glutamate can be synthesized from the precursor, L-glutamine (Wiley, Lu and Owyang, 1991) and released during high potassium (Wiley, Lu and Owyang, 1991) or electrical-evoked (Sinsky and Donnerer, 1998) depolarization in a Ca2+-dependent manner. A recent study has demonstrated that both N-type and P-type calcium channels control most of stimulated glutamate release from LM-MP preparations of the guinea-pig ileum (Reis et al., 1999). Both ionotropic and metabotropic glutamate receptors have been found to be abundantly expressed in the enteric nervous system. Immunoreactivity for ionotropic receptors of the

244

INNERVATION OF THE GASTROINTESTINAL TRACT

N-methyl-D-aspartate (NMDA), α-amino-3-hydroxy-5methyl-4-isoxazole propionic acid (AMPA) and of the kainate type, and for group I metabotropic receptors of the mGluR5 type has been detected in both the myenteric and submucosal plexus of the guinea-pig ileum (Kirchgessner, Liu and Alcantara, 1997; Liu et al., 1997; Liu and Kirchgessner, 2000). In addition, expression of mRNAs encoding for NMDA and mGluR2/3 metabotropic receptors has been demonstrated within enteric ganglia at different levels in the guinea-pig and rat gut (Broussard et al., 1994, Burns, Stephens and Benson, 1994; Larzabal et al., 1999). Enteric neurons are also endowed with a glutamate-inactivating mechanism, represented by the high affinity neuronal transporter, EAAC1 whose expression along the gut is particularly extensive (Liu et al., 1997). Effects of excitatory amino acids on neuromuscular preparations Evidence from functional studies suggests that L-glutamate mediates contraction of the longitudinal muscle in the guinea-pig ileum through activation of NMDA receptors, probably located on excitatory cholinergic motor neurons (Luzzi et al., 1988; Shannon and Sawyer, 1989; Wiley, Lu and Owyang, 1991; Sinsky and Donnerer, 1998). Indeed, in LM-MP preparations of the guinea-pig ileum, L-glutamate, L-aspartate and NMDA determine contractile responses which are competitively antagonized by NMDA receptor antagonists and noncompetitively blocked by Mg2+ (Moroni et al., 1986; Luzzi et al., 1988; Shannon and Sawyer, 1989; Wiley, Lu and Owyang, 1991). An NMDA receptor has been suggested to be involved in the contraction of the guinea-pig ileum also since the action of glutamate is potentiated by glycine in a strychnine-insensitive manner (Luzzi et al., 1988). In addition, glutamate-induced contractions are inhibited by TTX and hyoscine, but not by hexamethonium (Moroni et al., 1986; Luzzi et al., 1988; Wiley, Lu and Owyang, 1991). On the whole, this evidence suggests that L-glutamate stimulates ACh release from LM motor neurons which in turn causes the contraction via activation of postjunctional muscarinic receptors. Indeed, pharmacological evidence has been provided for an NMDA-mediated enhancement of ACh release from myenteric neurons in the guinea-pig ileum and colon (Wiley, Lu and Owyang, 1991; Cosentino et al., 1995). At this latter level, application of NMDA enhances also the release of an inhibitory neurotransmitter such as noradrenaline. This effect has been correlated to the NMDA-mediated inhibition of peristalsis in the colon (Cosentino et al., 1995). The possible involvement of AMPA and kainate receptors in the modulation of motor function in the gastrointestinal tract has not yet been completely clarified. AMPA and kainate receptors do not seem to mediate a contractile response in the guinea-pig ileum, since neither quisqualate nor kainate are effective in producing contractions of LM-MP preparations at concentrations up to 1 mM (Moroni et al., 1986; Luzzi et al., 1988; Shannon and Sawyer, 1989). Recent pharmacological investigations have demonstrated that activation of AMPA receptors inhibits both electrically-induced contractions of the circular muscle and ACh release in the guinea-pig distal colon (Giaroni et al., 2000). In the same study, AMPA has been described to enhance the efficiency of peristalsis. On the basis of these apparently contrasting results, the authors suggested that AMPA receptors might have a more prominent role in the modulation of inhibitory inputs to the colon (Giaroni et al., 2000). Furthermore, in accordance with the inability of kainate to induce LM-MP

PHARMACOLOGY OF THE ENTERIC NERVOUS SYSTEM

245

contractions in the guinea-pig ileum (Luzzi et al., 1988), this compound was ineffective in modulating both ACh release and peristalsis in the colon of the same species (Giaroni et al., 2000). Electrophysiological effects of excitatory amino acids on neurons Electrophysiological studies of glutamate effects on enteric neurons have been described with contrasting results for the occurence of depolarizing responses and identification of glutamate receptors involved in the response. Grafe, Galvan and Mayer (1979) found no significant electrophysiological effect of L-glutamate in seven myenteric neurons studied, whereas Liu et al. (1997) found that application of glutamate evokes depolarizing responses in myenteric neurons that had fast and slow components. The fast component was mimicked by AMPA, whereas the slow component was mimicked by NMDA. In the same study, glutamate and ACh have been suggested to represent excitatory co-transmitters, both contributing to mediate fast ganglionic neurotransmission in AH/type 2 neurons. However, contrasting data have been recently reported suggesting that glutamate evokes only slowly activating depolarizing responses in submucous and myenteric neurons of the guinea-pig small intestine by activation of group I metabotropic receptors (Ren et al., 2000).

PHARMACOLOGY OF HISTAMINERGIC TRANSMISSION Histidine decarboxylase (HDC, a marker for histamine containing neurons) and its mRNA have been demonstrated in neurons in the rat central and peripheral nervous systems (Watanabe et al., 1984; Karhula et al., 1990). HDC immunoreactivity has been demonstrated in myenteric neurons in the rat small intestine (Ekblad et al., 1985). However, since the antibody used in these studies cross-reacts with guinea-pig L-dopa decarboxylase, the distribution of HDC in the myenteric neurons in the guinea-pig small intestine could not be determined (Ando-Yamamoto et al., 1986). In the intestine, histamine interacts with three types of receptor, namely H1, H2 and H3 receptors, for which there are currently available selective agonists and antagonists (Hill et al., 1997). H1 receptors are located postjunctionally where they mediate contraction (Zavecz and Yellin, 1982; Trzeciakowski, 1987; Barocelli et al., 1993). H2 receptors are present on LM motor neurons of guinea-pig small intestine where they facilitate the release of ACh and tachykinins (Zavecz and Yellin, 1982; Barker and Ebersole, 1982; Poli, Loruzzi and Bertaccini, 1990). These receptors excite myenteric AH neurons partly by enhancing chloride conductance (Starodub and Wood, 2000). H3 receptors are mainly prejunctional and inhibit the release of ACh and tachykinins from LM motor neurons (Trzeciakowski, 1987; Hew, Hodgkinson and Hill, 1990; Menkveld and Timmerman, 1990; Taylor and Kilpatrick, 1992). In addition, H3 receptors have been demonstrated to mediate inhibition of noradrenaline release from intestinal sympathetic nerves (Blandizzi et al., 2000; Liu et al., 2000). The effects mediated by histamine receptors have been recently assessed in distensionevoked ascending excitatory transmission in the guinea-pig ileum. Histamine acting at H1

246

INNERVATION OF THE GASTROINTESTINAL TRACT

receptors depressed synaptic transmission at low concentrations, whereas increased it at high concentrations. Activation of H2 receptors by the agonist dimaprit was associated with facilitation of transmission, whereas activation of H3 receptors by α-methylhistamine inhibited synaptic transmission (Izzo et al., 1998). Histamine is known to exist in mast cells and basophils in the intestinal mucosa, but the role of endogenous histamine in intestinal motility is unknown.

PHARMACOLOGY OF PEPTIDES AS ENTERIC NEUROTRANSMITTERS OPIOID PEPTIDES The guinea-pig small intestine contains peptides derived from proenkephalin and prodynorphin but not pro-opiomelanocortin (Steele et al., 1992). The presence of enkephalinlike and dynorphin-like immunoreactivity in the myenteric plexus has been reported by a number of researchers (Furness and Costa, 1987). Opioids are found in orally and anally directed interneurons, excitatory and inhibitory CM motor neurons and rarely in LM motor neurons (Brookes, Steele and Costa, 1991b, 1992; Steele et al., 1992; Costa et al., 1996). Recently, the endogenous opioid endomorphin-1 and endomorphin-2 have been detected in brain (Zadina et al., 1997) and spinal cord of various species, including man. However, endomorphins have not been found yet in the gut. Ligand binding and functional studies indicate that µ and κ receptors, and to a lesser extent δ receptors, are present in the myenteric plexus of mammals. The presence of δ receptors has been detected in the ileal submucosal plexus, at least in the guinea-pig. cDNA encoding an “orphan” receptor has been identified which has a high degree of homology to the classical opioid receptors. This receptor has been named ORL1 (opioid receptor like) (Mollereau et al., 1994; Nicholson et al., 1998). The endogenous peptide agonist for ORL1, nociceptin referred also as orphanin FQ, has been detected in porcine, rat and guinea-pig gastrointestinal tract (Osinski et al., 1999; Yazdani et al., 1999; O’Donnell et al., 2001). In the guinea-pig myenteric plexus, nociceptin immunoreactivity is expressed preferentially in excitatory motor neurons projecting to the longitudinal and circular muscle layers, as well as a small subgroup of descending interneurons (O’Donnell et al., 2001). Effects of opioids on neuromuscular preparations In the guinea-pig ileum, agonists at µ (morphine, normorphine, enkephalins, DAMGO, and endomorphin-1,2), κ receptors (dynorphin and U-50488H), and ORL1 receptors (nociceptin) inhibit electrically-induced longitudinal muscle cholinergic contractions by presynaptic inhibition of ACh release (Greenberg, Kosterlitz and Waterfield, 1970; Chavkin and Goldstein, 1981; Vizi et al., 1984; Kojima et al., 1994; Tonini et al., 1998; Sternini et al., 2000; O’Donnell et al., 2001). Opioids acting at µ and κ receptors have been shown directly to decrease ACh release evoked by substance P, neurotensin and caerulein (Yau, Verdun and Youther, 1983; Yau, Verdun and Youther, 1983; Yau, Dorsett and Youthe, 1986b; Huidobro-Toro et al., 1984). They can inhibit hyoscine-resistant,

PHARMACOLOGY OF THE ENTERIC NERVOUS SYSTEM

247

substance P-mediated contractions (Gintzler and Scalisi, 1982; Barthó et al., 1982b) and have been shown to reduce substance P release (Holzer, 1984). The action of agonists at ORL1 receptors is region specific as observed in the rat gastrointestinal tract. In this species, nociceptin inhibits cholinergic transmission in the stomach and small intestine, whereas it stimulates colonic contractions by inhibiting a neurogenic inhibitory pathway within the myenteric plexus (Yazdani et al., 1999). No specific effect of δ receptor agonists on excitatory LM motor neurons has been demonstrated. The agonists DPLPE and DPDPE were found to reduce LM contraction by 50% (Mosberg et al., 1983; Galligan et al., 1984; Ersparmer et al., 1989) or to be ineffective (Kojima et al., 1994). Whilst these effects were reversed by naloxone (Mosberg et al., 1983; Galligan et al., 1984) and the µ receptor antagonist CTP, (Shook et al., 1987) they were not antagonized by the specific δ receptor antagonist, ICI 174864 (Shook et al., 1987). Agonists at µ and κ receptors inhibit CM contraction in the whole wall, CM-MP and CM-axons preparations evoked by 5-HT (Harry, 1963; Johnson et al., 1987; Johnson, Costa and Humphreys, 1988). The δ agonist, DPLPE, partly inhibited circular muscle contraction at very high concentrations (3 mM); however this effect was not antagonized by the δ antagonist, ICI 174864. Morphine inhibits ascending excitatory reflex contractions when added to the oral, intermediate or anal compartments of a partitioned organ bath (Tonini et al., 1992c). This suggests that µ receptors are present on CM motor neurons and interneurons, where they inhibit transmitter release. Selective agonists at µ and κ receptors (PLO17 and U-10488, respectively) concentration-dependently inhibit the ascending excitatory reflex response in the rat ileum, whereas the δ agonist DPDPE is ineffective (Allescher et al., 2000). However, δ receptors seem to be involved together with µ receptors in the timing of the reflex response. The effects of opioid receptor stimulation on NANC inhibitory transmission have been investigated in rat, guinea-pig, canine and human gastrointestinal tract. Morphine was found to inhibit electrically evoked NANC inhibitory responses in rat gastric fundus strips (Dehpour et al., 1994), whereas loperamide (a non-selective µ receptor agonist) inhibited the 5-HT-induced NANC relaxation in isolated intact stomach from guinea-pig (Meulemans, Helsen and Schuurkes, 1993). Agonists acting at µ and κ receptors decreased the amplitude of NANC IJPs evoked by transmural nerve stimulation in the circular muscle of canine duodenum (Bauer and Szurszewski, 1991). In the circular muscle of guinea-pig and human colon, NANC inhibitory responses are reduced by activation of δ receptors (Hoyle et al., 1990; Zagorodnyuk and Maggi, 1994). Electrophysiological effects of opioids on neurons Early studies involving intracellular recordings of myenteric neurons showed that opioid agonists acting at µ receptors, such as met- and leu-enkephalin, morphine and normorphine hyperpolarize cell membranes (North and Tonini, 1977; Sakai, Hymson and Shapiro, 1978; North, Katayama and Williams, 1979; Cherubini and North, 1985b) via stereospecific and naloxone-reversible mechanisms, whereas opioids that are selective for κ receptors (dynorphin, U-50488H) do not (Cherubini and North, 1985b). Opioids acting at both receptor subtypes caused presynaptic inhibition of ACh release in the myenteric

248

INNERVATION OF THE GASTROINTESTINAL TRACT

plexus, depressing the amplitude of fast EPSPs (Cherubini and North, 1985b). However, µ and κ opioids inhibit transmitter release by different presynaptic inhibitory mechanisms; µ receptor activation results in hyperpolarization due to increased potassium conductance, whereas κ receptor activation may depress the release of ACh by directly reducing calcium entry into the nerve terminals (Cherubini and North, 1985b). A comprehensive study of the electrophysiological effects of DAMGO was undertaken by Pillai and Johnson (1991). This study demonstrated that 42% of S neurons are hyperpolarized by DAMGO. The magnitude of the hyperpolarization increased with increasing concentrations of DAMGO and was prevented by pre-application of naloxone. Fast EPSPs were significantly diminished in amplitude in 48% of S cells; this effect occurred in some cells which had not been hyperpolarized by DAMGO. This effect is probably due to a presynaptic action of opioids, since fast EPSPs evoked by exogenous nicotine are unaffected (Cherubini and North, 1985b; Cherubini, Morita and North, 1985). In AH cells, there was no effect of DAMGO on the resting membrane potential, input resistance, action potential duration or amplitude, nor on the amplitude or duration of the fast and slow after-hyperpolarizations. Propagation of action potentials to the soma of AH neurons was also unaffected by DAMGO. Ito and Tajima (1980) provided electrophysiological evidence that µ agonists do not act directly on the muscle. Morphine reduced the amplitude of the excitatory junction potential, but did not change the membrane potential, resistance or electrical threshold for activation of spikes in circular or longitudinal muscle. Thus the site of action of morphine was prejunctional. There is no evidence for inhibition of inhibitory junction potentials by opioids (Ito and Tajima, 1980). Role of endogenous opioids Trendelenburg (1917) was the first to demonstrate inhibition of peristalsis by morphine in the guinea-pig small intestine in vitro. Morphine, fentanyl and met-enkephalin inhibit both longitudinal and circular muscle contraction during the preparatory and emptying phases respectively (Schaumann, 1955; Kosterlitz and Robinson, 1957; Fontaine, Reuse and Van Nueten, 1973; Van Nueten, Van Ree and Vanhoutte, 1977). Fontaine, Reuse and Van Nueten (1973) reported a preferential effect of morphine and fentanyl on longitudinal muscle contraction. At submaximal doses, morphine, fentanyl and met-enkephalin increased the pressure at threshold, reduced longitudinal and circular muscle activity, increased filling of the intestine, reduced the volume expelled per wave and reduced the frequency of peristaltic waves (Fontaine, Reuse and Van Nueten, 1973; Van Nueten, Van Ree and Vanhoutte, 1977). Similarly, Kromer and co-workers (Kromer, Pretzlaff and Woinoff, 1981; Kromer, Steigemann and Shearman, 1982) have shown a dose-dependent inhibitory effect of normorphine in terms of the frequency of peristaltic waves. Inhibition was also produced by D-Ala2-Leu-enkephalinamide, β-casomorphin, DAMGO, ketocyclazocine and dynorphin (Kromer, Pretzlaff and Woinoff, 1980; Kromer, 1990). All of these results were obtained using the Trendelenburg preparation. However, peristalsis was also abolished by dermorphin, morphine, D-Ala2-D-Met5-enkephalin, FK 33–824 and dynorphin in the vascularly-perfused preparation (Holzer and Lembeck, 1979; Barthó et al., 1982a; Donnerer and Lembeck, 1985). At low concentrations, these opioids reduced

PHARMACOLOGY OF THE ENTERIC NERVOUS SYSTEM

249

both the frequency of peristaltic waves and the maximal ejection pressure. Barthó et al. (1982a) also demonstrated that FK 33–824 could abolish atropine-resistant peristalsis. A more recent study (Waterman, Costa and Tonini, 1992) has demonstrated that µ opioid receptor agonists reduce the compliance of the intestinal wall, the LM contraction during the preparatory phase and increase the threshold volume required to trigger emptying. Kappa receptor agonists reduce the maximal ejection pressure during the emptying phase and LM contraction and increase the threshold volume. Delta receptors also modulate peristalsis in the guinea-pig ileum. In agreement with studies by Kromer (1990), Waterman, Costa and Tonini (1992) demonstrated that δ agonists reduce the maximal ejection pressure, LM contraction during the preparatory phase and increase the threshold volume. In 1977, Schulz et al. reported that material with enkephalin-like activity was released from LM-MP strips. Furthermore, the enkephalin content of LM-MP strips was reduced by high frequency stimulation (1 and 10 Hz) – but not low frequency (0.1 Hz) – stimulation (Hughes, Kosterlitz and Sosa, 1978). Oka and Sawa (1979) showed that this release was calcium-dependent. The first direct measurement of met-enkephalin release from LM-MP strips was not achieved until 1986 (Glass, Chan and Gintzler, 1986). Enkephalin release was detectable following electrical stimulation at frequencies between 5 and 80 Hz. The release was tetrodotoxin-sensitive and calcium-dependent. Enkephalin release was enhanced in the presence of naloxone, but reduced by morphine pretreatment. This suggests that endogenous opioids can modulate their own release (Sublette and Gintzler, 1992). Preparations of synaptosomes from guinea-pig LM-MP strips have been shown to release met- and leu-enkephalin. Release increased following depolarization by potassium. Met-enkephalin release was detectable only in the presence of naloxone or atropine. Leu-enkephalin release was detectable only in the presence of naloxone, atropine or a purine receptor antagonist (1,3-dipropyl-8-p-sulphophenylxanthine) (Christofi, McDonald and Cook, 1990). Waterfield and Kosterlitz (1975) reported that naloxone stereospecifically increased the release of ACh in response to low frequency (0.017 Hz) electrical stimulation of the LM-MP preparation. The effect was less marked at a stimulation frequency of 10 Hz. Puig et al. (1977) combined different stimulation frequencies and showed that longitudinal muscle contraction evoked by stimulation at 0.1 Hz was reduced following stimulation at 10 Hz for 5 min. This inhibitory effect was antagonized in a dose-dependent manner by naloxone and naltrexone. Similar results have since been reported by Horàcek and Kadlec (1984). Gintzler and Scalisi (1982) reported that naloxone increased atropine-resistant contractions of the longitudinal muscle evoked by electrical stimulation at 20 Hz. This effect was unaltered by the addition of hexamethonium, but blocked by prior desensitization of the preparation to SP. Barthó et al. (1982b) similarly showed that naloxone enhanced atropineresistant contractions of the longitudinal muscle. These studies indicate that endogenous opioid peptides released by high frequency electrical stimulation can inhibit the release of SP from excitatory longitudinal muscle motor neurons. More direct evidence for this was provided by Holzer (1982). He directly measured SP release from LM-MP strips and showed that this was increased in the presence of naloxone. Research by Garzon et al. (1985, 1987) also indicated that endogenous opioid peptides inhibit the release of ACh and SP. They showed that contraction of LM-MP strips in response to a variety of peptides

250

INNERVATION OF THE GASTROINTESTINAL TRACT

(corticotropin-releasing factor, CCK, neurotensin, bombesin, angiotensin II, bradykinin and SP) was enhanced by naloxone in the presence and absence of atropine. In the guinea-pig colon, naloxone was found to increase resting and electrically evoked ACh release and electrically evoked noradrenaline release, and to enhance NANC relaxations in the circular muscle. The functional relevance of endogenous opioids appears to be enhanced after chronic sympathetic denervation (Marino et al., 1993). In the guinea- pig small intestine, naloxone enhances the amplitude of ascending excitatory reflex contractions when added to the oral compartment of a partitioned organ bath (Tonini et al., 1992c). Naloxone increases the rate of peristaltic contractions in the rat small intestine (Coupar, 1995). In the same preparation, CTOP and nor-binaltorphimine (nor-BNI), selective antagonists at µ and κ receptors respectively, increase the ascending excitatory reflex, whereas the selective δ antagonist ICI 174864 is ineffective (Allescher et al., 2000). In the rat colon, naloxone enhances distension evoked descending relaxation and VIP release (Grider and Makhlouf, 1987a). Together these findings suggest that endogenous opioids acting preferentially at µ and κ receptors modulate the release of excitatory and inhibitory transmitters in the gastrointestinal tract. Peristaltic activity rapidly “fatigues” when the Trendelenburg preparation is used. In this preparation, the oral end of the intestine is occluded and fluid enters the anal end of the intestine. The intestine empties fluid against the inflow head of pressure. Van Nueten, Janssen and Fontaine (1976) showed that the opioid receptor antagonist, naloxone, restored peristaltic activity in fatigued preparations. Furthermore, the organ bath solution surrounding a fatigued preparation inhibited peristalsis in non-fatigued segments. The latter effect was antagonized by naloxone. Van Nueten and colleagues therefore concluded that fatigue was due to the release of endogenous opioid peptides. In a similar study, Kadlec and Horàcek (1980) demonstrated that inhibition of peristalsis by a “stress stimulus” (12 cm H2O intraluminal pressure and 3 g longitudinal tension) was reversed by naloxone. Davison and Najafi-Farashah (1982) demonstrated that naloxone restored peristaltic activity in fatigued segments of intestine. At higher distending pressures, when intermittent myogenic activity was recorded, naloxone also restored continuous peristaltic activity. In a series of studies, Kromer and colleagues have shown that naloxone stereospecifically increases the frequency of peristaltic waves, reduces the number and duration of peristalsis-free intervals and transiently reduces the volume expelled (Kromer and Pretzlaff, 1979; Kromer and Woinoff, 1980; Kromer, Pretzlaff and Scheiblhuber, 1980; Kromer, Pretzlaff and Woinoff, 1980). Increasing the extracellular calcium concentration attenuated the effects of naloxone (Kromer, Scheiblhuber and Illes, 1980). These effects were observed in segments of ileum and to a lesser extent in the duodenum and jejunum from adult, fetal and pregnant guinea-pigs (Kromer and Pretzlaff, 1979; Kromer, Pretzlaff and Woinoff, 1980). A similar study (Kromer, Steigemann and Shearman, 1982) showed that peristaltic activity was also enhanced by the µ antagonist and proposed σ receptor agonist, SKF 10047. However more recently, Kromer (1990) reported that the µ antagonist, CTOP, did not significantly alter peristaltic activity. Instead, peristalsis was enhanced by the κ antagonist, nor-BNI. In more physiological preparations, in which both ends of the intestine are open and fluid is infused through the oral end, intra-arterial administration of naloxone has no effect

PHARMACOLOGY OF THE ENTERIC NERVOUS SYSTEM

251

on peristalsis (Holzer and Lembeck, 1979; Donnerer and Lembeck, 1985). Nevertheless, dynorphin is released into the venous effluent of the vascularly-perfused guinea-pig small intestine during peristalsis (Donnerer, Holzer and Lembeck, 1984). Dynorphin release increases during peristalsis. Donnerer, Holzer and Lembeck (1984) also showed that naloxone enhances the release of substance P and somatostatin during peristalsis, but not under resting conditions. Thus endogenous opioids may inhibit the release of SP and somatostatin during peristalsis. Another study, using more physiological conditions than the Trendelenburg technique, found that the power generated by the intestine when emptying was increased in the presence of nor-BNI, under certain conditions (Waterman, Costa and Tonini, 1992). Thus, endogenous opioids acting at κ receptors, inhibited peristalsis when the intestine was forced to empty against a high resistance. SOMATOSTATIN Somatostatin is a tetradecapeptide which is present in neuronal cell bodies in the myenteric and submucous plexuses of the guinea-pig small intestine (Costa et al., 1977, 1980b; Furness et al., 1980; Schultzberg et al., 1980; Endo, Uchida and Kobayashi, 1986; Portbury et al., 1995; Song et al., 1997). In the myenteric plexus, somatostatin is contained in 4% of neurons, which project up to 70 mm in an anal direction and form very long interneuronal chains (descending interneurons) that connect with both myenteric and submucous neurons (Song et al., 1997; Mann et al., 1997b). At the electron microscope level, the neurons can be seen to form axosomatic synapses with somatostatin-positive and -negative neurons (Endo, Uchida and Kobayashi, 1986). Somatostatin-containing neurons do not project to either the circular or longitudinal muscle layers (Costa et al., 1977, 1980b; Endo, Uchida and Kobayashi, 1986). Somatostatin is colocalized with ChAT, SP and CCK in descending interneurons (Brookes, Steele and Costa, 1992; Portbury et al., 1995; Costa et al., 1996; Mann et al., 1997b; Hens et al., 2000). Effects of somatostatin on neuromuscular preparations Somatostatin reduces the amplitude of the cholinergic twitch of the LM evoked by electrical field stimulation in LM-MP and in whole wall preparations of guinea-pig ileum (Guillemin, 1976; Cohen et al., 1978, 1979; Furness and Costa, 1979b; Jhamandas and Elliot, 1980). Contractions induced by neurotensin and caerulein, but not by SP, are also inhibited (Monier and Kitagbi, 1981; Teitelbaum et al., 1984; Yau, Lingle and Youther, 1983; Yau, Dorsett and Youther, 1986b). Somatostatin has no effect on LM contractions elicited by exogenous ACh (Guillemin, 1976; Cohen et al., 1978), carbachol (Furness and Costa, 1979b) or histamine (Cohen et al., 1979), indicating that somatostatin acts by inhibiting the release of ACh rather than by interfering with its action on the smooth muscle. Inhibition of resting and evoked ACh release by somatostatin has been measured directly (Yau, Lingle and Youther, 1983; Yau, Dorsett and Youther, 1986b; Teitelbaum et al., 1984; Takeda et al., 1989; Milenov and Atanassova, 1993). More recently, somatostatin was found to possess both excitatory and inhibitory effects on myenteric cholinergic transmission, which partly involve the release of endogenous GABA. This transmitter, in turn, may enhance or inhibit ACh release via activation of GABAA or GABAB receptors,

252

INNERVATION OF THE GASTROINTESTINAL TRACT

respectively (Roberts, Hasler and Owyang, 1993). The effect of somatostatin on the CM of the guinea-pig ileum has not been studied in detail. Kromer and Woinoff (1981) reported that somatostatin did not have any effects on the circular muscle in the presence of TTX, atropine, hexamethonium or 5-HT desensitization. Thus, somatostatin does not act directly on the CM. In the guinea-pig colon, somatostatin stimulates inhibitory motor neurons, since it produces a TTX-sensitive relaxation of the LM (Furness and Costa, 1979b). This relaxation is unaffected by guanethidine, and is therefore mediated by non-adrenergic neurons (Furness and Costa, 1979b). It is not clear whether somatostatin stimulates both the apamin-sensitive and -insensitive components of inhibitory transmission. In isolated smooth muscle cells from the stomach and colon of guinea-pig, somatostatin was found to possess a direct contractile activity (Corleto et al., 1997). In these cells, five somatostatin (SST) receptors have recognized (Corleto, Weber and Jensen, 1999), but the contractile effect was predominantly mediated by SST3 receptors in gastric cells and by SST5 receptors in colonic cells (Corleto et al., 1997). Electrophysiological effects of somatostatin on neurons Extracellular recordings of unit activity in the myenteric plexus indicate that somatostatin reduces the firing frequency of neurons (Williams and North, 1978). This effect is independent of extracellular calcium levels. Thus the action of somatostatin is likely to be a direct one on the neurons whose activity was being recorded. Intracellular recording demonstrated that somatostatin can depolarize and hyperpolarize both AH and S type neurons. The proportion of S neurons depolarized varied from 17–34% and those hyperpolarized varied from 20–31%, depending on whether the somatostatin was applied by perfusion or iontophoretically. Similarly, the proportion of AH neurons depolarized and hyperpolarized respectively was 9–29% and 8–20%. Depolarization was associated with an increase in input resistance and had a reversal potential near the equilibrium potential of potassium, suggesting that somatostatin acts by closing a potassium channel. The somatostatin-induced hyperpolarization was associated with an decrease in input resistance (Katayama and North, 1980; Liu et al., 2000). Role of endogenous somatostatin Endogenous somatostatin is released into the venous effluent during peristalsis in various species (Donnerer, Holzer and Lembeck, 1984; Schmidt, Rasmussen and Holst, 1993). This release was partly TTX-sensitive and therefore nerve mediated. However, a significant proportion is likely to represent release from mucosal endocrine cells. Somatostatin release could be increased by DMPP in a TTX-sensitive manner and by CCK-8 in TTXinsensitive manner. Somatostatin is released from isolated myenteric ganglia in response to nicotinic receptor stimulation by DMPP (Grider, 1989). The amount of somatostatin release is decreased by VIP and can be augmented by a VIP receptor antagonist (VIP10–28; Grider, 1989). This suggests that the release of somatostatin may be inhibited by endogenous VIP. Somatostatin has been reported to have a variety of effects on peristalsis. Holzer and Lembeck (1979) found that somatostatin had no immediate effect when infused intra-

PHARMACOLOGY OF THE ENTERIC NERVOUS SYSTEM

253

arterially. However, 12 minutes after the end of infusion, peristalsis became increasingly irregular and the volume of fluid expelled was decreased. This effect appeared to be irreversible (Holzer and Lembeck, 1979). Kromer and Woinoff (1981) studied peristalsis using the Trendelenburg method, in which the intestine “fatigues” and ceases to undergo normal peristalsis. They found that somatostatin stimulated peristalsis when it had been highly intermittent and inhibited it when it had been less intermittent. In the rat colon, somatostatin release increases along with VIP in response to the descending inhibitory reflex (Grider, Arimura and Makhlouf, 1987). Anti-somatostatin antibodies reduce VIP release and the reflex relaxation, whereas exogenous and endogenous somatostatin enhances both VIP/NO release and the reflex relaxation (Grider, 1994a). The latter effects are partly achieved through a decrease of met-enkephalin release, which exerts a tonic inhibitory influence on descending pathways. GASTRIN, CHOLECYSTOKININ AND CAERULEIN Cholecystokinins consist of a group of peptides which are C-terminal fragments of CCK-33 (Williams, 1982). Cholecystokinins, gastrin, caerulein (also known as ceruletide) and phyllocaerulein share a common pentapeptide sequence at their C-terminus and act with varying affinities at the same receptors, i.e. CCKA and CCKB receptors (Alexander and Peters, 2000). Immunohistochemical studies have demonstrated that CCK is localized in specific subpopulations of myenteric neurons in the guinea-pig small intestine: anally directed interneurons in the myenteric plexus, neurons projecting to the mucosa, intestinofugal neurons and neurons projecting from the myenteric to the submucous plexus (Furness and Costa, 1987; Brookes, Steele and Costa, 1992). In descending interneurons, CCK is colocalized with somatostatin, ChAT and SP. CCK is not present in motor neurons in either the CM or LM. Electrophysiological and immunohistochemical studies have demonstrated the presence of both CCKA and CCKB receptors in rat and guinea-pig myenteric neurons (Schutte, Akkermans and Kroese, 1997; Sternini et al., 1999a; Sayegh and Ritter, 2000). CCKA receptors are predominantly expressed in neurons containing VIP, but they are also expressed in neurons containing SP (Sternini et al., 1999a). Effects of CCK-like peptides on neuromuscular preparations Gastrin produces a contraction of the guinea-pig LM which is unaffected by hexamethonium but abolished or reduced by atropine, suggesting that the agonist acts on cholinergic LM motor neurons (Bennett, 1965; Yau et al., 1974; Yau, 1978). Caerulein and CCK-8 induce a pronounced and sustained TTX-sensitive contraction of the guinea-pig ileum LM. This contraction is partly inhibited by atropine and potentiated by eserine, suggesting that these peptides also stimulate release of ACh from LM motor neurons (Del Tacca et al., 1970; Hedner, 1970; Yau et al., 1974; Zetler, 1979; Hutchison and Dockray, 1981; Garzón et al., 1985). Increased ACh release in response to CCK-8, pentagastrin, gastrin and caerulein has been confirmed by direct measurement (Del Tacca, Soldani and Crema, 1970; Vizi et al., 1972, 1973; Yau, Lingle and Youther, 1983; Teitelbaum et al., 1984; Sano, Taniyama and Tanaka, 1989; Milenov and Atanassova, 1993). The

254

INNERVATION OF THE GASTROINTESTINAL TRACT

atropine-resistant contraction is abolished by SP desensitization (Hutchison and Dockray, 1981), suggesting that SP or a related tachykinin is involved. Circumstantial evidence in favour of this hypothesis was provided by Holzer (1984), who showed that CCK-8 stimulates the release of SP from LM-MP preparations in a TTX-sensitive manner. In some experiments, the hyoscine-resistant contraction is reduced by about 20–35% by hexamethonium (Del Tacca, Soldani and Crema, 1970; Hedner, 1970), indicating that caerulein and CCK-8 also stimulate cholinergic interneurons. Sano, Taniyama and Tanaka (1989) found that CCK-induced release of ACh from guinea-pig LM-MP preparations was inhibited by 42% by bicuculline, suggesting that part of the effect of CCK is indirect, via release of GABA which acts on GABAA receptors. Nevertheless, neurally-mediated effects may be contractile (via release of ACh/tachykinins) or relaxant (via release of VIP/NO) as demonstrated in the CCK-mediated relaxation of gastric fundus and emptying (Grider, 1994b). CCK also appears to act directly on the LM, since higher concentrations (0.1 mM) can produce contractions in the presence of TTX. A direct effect of CCK has been demonstrated in gastric fundus and in caecal smooth muscle cells where are present both receptor subtypes, which mediate a contractile response (Grider and Makhlouf, 1987b; de Weerth et al., 1997; Motomura et al., 1997). Caerulein induces TTX-sensitive, rhythmic contractions superimposed on increased tone of the guinea-pig ileum CM (Holzer, Lembeck and Donnerer, 1980; Barthó et al., 1987a). These contractions are blocked by atropine but not hexamethonium (Holzer, Lembeck and Donnerer, 1980), suggesting that caerulein acts by stimulating ACh release from CM motor neurons. The contractions are partly or completely inhibited by desensitization to SP, indicating that tachykinin release mediates part of the contractile response. Caerulein and pentagastrin lower the threshold distension required to trigger emptying of the gut and consequently increase the frequency of emptying when the intestine is infused with fluid at a constant rate (Frigo et al., 1971; Chijikwa and Davison, 1978; Tonini et al., 1989b). Similarly, CCK-8 produces a concentration-dependent increase in the frequency of peristaltic waves and in the volume of fluid expelled (Barthó et al., 1982a). Caerulein can initiate peristalsis in the undistended intestine (Holzer and Lembeck, 1979) and restore peristalsis which has been impaired by atropine or hexamethonium (Frigo et al., 1971; Barthó et al., 1989). This effect is likely to be due to the ability of CCK to stimulate release of ACh from CM motor neurons. Electrophysiological effects of CCK-like peptides on neurons The electrophysiological effects of CCK-like peptides in the guinea pig ileum have received little attention. CCK-8 mimics slow EPSPs in 68% of S neurons and 52% of AH neurons (Nemeth, Zafirov and Wood, 1985). CCK-8 also hyperpolarizes 17% of AH neurons and 5% of S neurons. These responses were unaffected by proglumide. Caerulein had similar effects to CCK-8, although hyperpolarizing responses were not recorded (Nemeth, Zafirov and Wood, 1985). Pentagastrin had no significant effect on AH neurons, and was not tested on S neurons (Nemeth, Zafirov and Wood, 1985). In another study, CCK-8 was found to depolarize 83% of S neurons. Depolarization was also obtained with the CCKB receptor agonist CCK-8NS. Use of CCKA and CCKB selective antagonists indicated that some neurons possess exclusively the CCKA and CCKB receptor subtype,

PHARMACOLOGY OF THE ENTERIC NERVOUS SYSTEM

255

but others possess both subtypes (Schutte, Akkermans and Kroese, 1997). Since selective antagonists for CCKA and CCKB (L-364,78 and L-365,260, respectively) were found to inhibit slow EPSPs in myenteric neurons, this provides evidence that neurally released CCK is involved in the mediation of slow EPSPs via CCKA and CCKB receptors (Schutte et al., 1997). Role of endogenous CCK-like peptides CCK-8 is released into the venous effluent in response to distension of the intestine during peristalsis (Donnerer et al., 1985). The nicotinic receptor agonist, DMPP, did not produce a significant increase in release above basal levels (Donnerer et al., 1985). It is therefore unlikely that CCK is released from neurons during peristalsis in response to excitation by cholinergic nicotinic transmission. The role of endogenous CCK in peristalsis is not known. Peristalsis was not significantly altered by the CCK receptor antagonist, CR 1409 (Barthó et al., 1987a). Similarly, lorglumide did not produce any consistent, significant effects on hexamethonium-resistant peristalsis other than an increase in the pressure at threshold (Barthó et al., 1989). The latter may reflect a decrease in the compliance of the intestinal wall, due to either a decrease in inhibitory neuronal tone or an increase in excitatory neuronal activity (Waterman, Costa and Tonini, 1992). Thus, endogenous CCK may stimulate inhibitory CM motor neurons or inhibit excitatory CM motor neurons. Both effects are possible, given the ability of CCK to depolarize and hyperpolarize myenteric neurons. In the duodenum of rat, use of selective CCKA and CCKB receptor antagonists (L-364,718 and L-365,260, respectively) had no effect on the ascending excitation to electrical field stimulation of the mucosa, but abolished the inhibitory effect of CCK-8 on electrical stimulation (Giralt and Vergara, 2000). The role of endogenous CCK in the descending inhibitory reflex has not been investigated. Since CCK is present in descending interneurons and not in motor neurons (Furness and Costa, 1987; Brookes, Steele and Costa, 1992), the most likely role for CCK would be in the descending inhibitory reflex, rather than the reflex pathways studied previously. NEUROPEPTIDE Y (NPY) AND PEPTIDE YY (PYY) NPY is a 36 amino acid peptide which has some sequence homology to PYY and pancreatic polypeptide (PP) (Tatemoto, Carlquist and Mutt, 1982). NPY and PYY are found in 5% of myenteric neurons, including anally projecting interneurons and short descending inhibitory motor neurons to the CM (Furness and Costa, 1987; Brookes, Steele and Costa, 1991b, 1992). NPY-immunoreactive nerve terminals synapse on non NPY-immunoreactive somata and fibres (Fehér and Burnstock, 1986). PYY-like immunoreactivity was not detectable by radioimmunoassay of LM-MP synaptosome samples (McDonald et al., 1988). In the guinea-pig gastric fundus and corpus, NPY has been localized on both ascending and descending pathways projecting to the mucosa (Pfannkuche et al., 1998; Reiche and Schemann, 1999). NPY is also present in secremotor neurons in the guineapig small intestine (Lomax, Bertrand and Furness, 1998). Five functional NPY receptors (Y receptors) have been recognized so far.

256

INNERVATION OF THE GASTROINTESTINAL TRACT

Effects of NPY on neuromuscular preparations NPY inhibits contractions of the LM induced by CCK-8 or electrical field stimulation (Garzón, Höllt and Sánchez-Blázquez, 1986; Holzer et al., 1987). This is not due to a direct inhibitory effect on the LM, since NPY does not alter concentration-response curves to ACh or carbachol in the presence of TTX (Garzón, Höllt and Sánchez-Blázquez, 1986; Holzer et al., 1987). Instead, NPY inhibits the release of ACh and a noncholinergic transmitter, probably a tachykinin, from LM motor neurons (Holzer et al., 1987; Takahashi, Yamamura and Utsunomiya, 1992). Similar results have been found in rabbit ileum, in which inhibition of ACh release is mediated by the Y5 receptor (Pheng et al., 1997). In the guinea-pig colon, PYY and NPY inhibit the twitch contractions mediated by the stimulation of cholinergic neurons and the resulting release of ACh. The inhibitory effect of PYY is mediated by a receptor located directly on cholinergic neurons, whereas the effect of NPY is mediated by noradrenaline release due to stimulation of a receptor located on adrenergic neurons (Sawa et al., 1995). NPY abolishes phasic TTX-sensitive contractions of the CM induced by DMPP or caerulein (Holzer et al., 1987). However, TTX-insensitive contractions evoked by carbachol are unaffected. Thus, NPY inhibits CM contractions through an action on excitatory CM motor neurons rather than by a direct effect on the CM. Such an action may also underlie the reduction in basal CM tone produced by NPY in some preparations (Holzer et al., 1987). NPY inhibits the ascending excitatory reflex contraction of the CM in response to balloon distension (Holzer et al., 1987). Both the atropine-sensitive and atropine-resistant components of the response are inhibited. Although these effects are likely to be due to inhibition of excitatory CM motor neurons, NPY needs to be tested in a partitioned organ bath to rule out the possibility that it also inhibits interneurons or sensory neurons. The inhibitory effect of NPY on the ascending excitatory reflex is significantly reduced by apamin (Holzer et al., 1987). NPY may therefore inhibit neurons by opening a calcium-dependent potassium channel in the neuronal membrane. Alternatively, NPY may stimulate inhibitory CM motor neurons which relax the muscle through an apamin-sensitive mechanism. At variance with the effects observed in the guinea-pig intestine, NPY enhances ACh release in the rat colon by acting on Y2 and Y4 receptors (Pheng et al., 1999). Y2 receptors have been also characterized in colonic smooth muscle cells (Feletou et al., 1998). Electrophysiological effects of NPY and related peptides on neurons NPY and PPY and bovine PP suppress nicotinic fast EPSPs (and ACh release) in guineapig gastric myenteric neurons via a mechanism insensitive to α-adrenoceptor blockade (Schemann and Tamura, 1992). These peptides have been found to inhibit fast EPSPs in myenteric neurons of descending colon through a mechanism involving activation of presynaptic Y2 receptors (Browning and Lees, 2000).

CALCITONIN GENE-RELATED PEPTIDE CGRP is a 37-aminoacid peptide localized in blood vessels and nerves of the gastrointestinal tract. Based upon the differential biological activity of various CGRP analogs, CGRP

PHARMACOLOGY OF THE ENTERIC NERVOUS SYSTEM

257

receptors have been classified into two major classes, namely the CGRP1 and CGRP2 subtypes (van Rossum, Hanisch and Quirion, 1997). In the guinea-pig ileum, CGRP is present in intrinsic and extrinsic fibres. The intrinsic fibres supply only the mucosa (Costa, Furness and Gibbons, 1986). Fibres innervating the myenteric plexus which are immunoreactive for CGRP are of extrinsic origin (Costa, Furness and Gibbons, 1986a). During peristalsis, CGRP may potentially be released from both sources. In fact, at least in the rat colon, extrinsic sensory pathways, which mediate the peristaltic response to muscle stretch, and intrinsic sensory pathways, which are involved in the peristaltic response to muscular stimulation, utilize CGRP as a sensory transmitter (Grider, 1994c). In the guinea-pig small intestine, however, extrinsic sensory neurons do not participate in the neural pathway mediating peristalsis, but stimulate peristaltic activity following capsaicin stimulation (Barthó and Holzer, 1995). CGRP produces a transient contraction followed by a longer relaxation of the guineapig ileum LM-MP preparation (Barthó et al., 1987b; Sun and Benishin, 1991). The contraction is blocked by TTX, atropine, 2-chloroadenosine, mepyramine and clonidine (Tippins et al., 1984; Sun and Benishin, 1991), suggesting that the contraction is mediated by ACh released from LM motor neurons and acting at muscarinic receptors. Histamine may also be involved in this response. The CGRP-induced relaxation is partly nerve-mediated, since it is partly blocked by TTX. However, CGRP also has a direct relaxing effect (Barthó et al., 1987b; Sun and Benishin, 1991) which is not antagonized by apamin (Holzer et al., 1989). CGRP has also been shown to inhibit electrically induced cholinergic contractions of the guinea-pig ileum LM-MP preparation (Barthó et al., 1987b) via CGRP1-like receptor activation (Tomlinson and Poyner, 1996). In smooth muscle cells isolated from the circular and longitudinal layers of guinea-pig ileum, CGRP triggers different intracellular pathways to induce relaxation: cAMP is involved in cells from both layers while NO is involved only in relaxation of circular smooth muscle cells (Rekik et al., 1997). CGRP induces phasic contractions of the guinea-pig ileum CM by stimulating ACh release from cholinergic motor neurons (Holzer et al., 1989). Part of the action of CGRP is likely to be on cholinergic interneurons, since the phasic contractions are reduced by hexamethonium (Holzer et al., 1989). CGRP modulates peristalsis by increasing the pressure which occurs at threshold (Holzer et al., 1989). This probably reflects a decrease in the compliance of the intestinal wall and increased CM tone due to either an increase in excitatory neuronal activity or a decrease in inhibitory neuronal activity. The former is more likely given the excitatory effect of CGRP on cholinergic motor neurons mentioned previously and that CGRP has not been reported to inhibit neurons. In agreement with this, CGRP was shown to enhance the amplitude of the ascending excitatory reflex (Holzer et al., 1989). In guinea-pig distal colon CGRP inhibits electrically-evoked NANC contractions through a postjunctional mechanism (Kojima, 1997). In electrophysiological studies, CGRP mimicked the slow EPSP in all myenteric AH neurons when applied by superfusion and pressure ejection. Depolarization was associated with an increase in input resistance, suppression of post-spike hyperpolarization and enhanced excitability (Palmer et al., 1986a). This effect may be mediated by a decrease in

258

INNERVATION OF THE GASTROINTESTINAL TRACT

the calcium-activated potassium conductance which is active in resting AH neurons (Palmer et al., 1986a). NEUROTENSIN Neurotensin immunoreactivity in the guinea-pig small intestine occurs mainly in the mucosa (Helmstaedter et al., 1977; Sundler et al., 1977; Schultzberg et al., 1980; Leander et al., 1984). In these studies, neurotensin-like immunoreactivity was not found in myenteric neurons of the guinea-pig ileum. However, other researchers have reported neurotensin-like immunoreactivity in these neurons (Reinecke et al., 1983; Tange, 1983). Neurotensin has been measured by radioimmunoassay in the external muscle layers of the guinea-pig ileum by one research group (Holzer et al., 1982) but was not found by another (Furness et al., 1982). It is possible that the discrepancies relate to the specificity of the antibodies used. Consequently, a peptide similar to neurotensin rather than authentic neurotensin may be present in the guinea-pig small intestine. Neurotensin binding sites have been demonstrated in the CM (Goedert, Hunter and Ninkovic, 1984). Recently, two neurotensin receptors have been identified in various tissues and selective antagonists for neurotensin-1 (NTS1) receptors have been developed (Vincent, Mazella and Kitabgi, 1999). Pharmacological experiments have demonstrated that neurotensin induces relaxation and then contraction of the guinea-pig LM-MP preparation (Kitagbi and Freychet, 1978, 1979a; Monier and Kitagbi, 1980; Goedert, Hunter and Ninkovic, 1984; Wagner and Wahl, 1986). The contraction is nerve-mediated, since it is TTX-sensitive (Kitagbi and Freychet, 1978, 1979a; Zetler, 1980; Garzón et al., 1985) and may be mediated by the release of ACh (Kitagbi and Freychet, 1978, 1979b; Zetler, 1980; Monier and Kitagbi, 1981; Huidobro-Toro and Way, 1982; Yau, Verdun and Youther, 1983; Teitelbaum et al., 1984; Garzón et al., 1985; Yau, Dorsett and Youther, 1986b; Nakamoto, Tanaka and Taniyama, 1987; Rakovska, 1993) and SP (Monier and Kitagbi, 1980). Neurotensin stimulates the release of ACh and SP from LM-MP preparations in a TTX-sensitive and calciumdependent manner (Yau, Verdun and Youther, 1983; Holzer, 1984; Yau, Dorsett and Youther, 1986b; Nakamoto, Tanaka and Taniyama, 1987). The neurotensin enhancement of ACh release is inhibited by stimulation of α2-adrenoceptor agonists (Rakowska, 1993). Stimulation of ACh release is unaffected by hexamethonium, hyoscine or SP receptor antagonists (Kitagbi and Freychet, 1979b; Nakamoto, Tanaka and Taniyama, 1987), suggesting that neurotensin may directly stimulate excitatory LM motor neurons. The action of neurotensin is likely to be at least partly at the level of the nerve terminals, because neurotensin is capable of stimulating ACh release from myenteric synaptosomes (Yau, Lingle and Youther, 1983). Neurotensin-induced contraction is partly inhibited by bicuculline. Since neurotensin also stimulates GABA release, it has been proposed that neurotensin may stimulate ACh release directly and also via release of GABA (Nakamoto, Tanaka and Taniyama, 1987). Neurotensin produces a direct relaxation of the guinea-pig ileum LM (Kitagbi and Freychet, 1978, 1979a; Goedert, Hunter and Ninkovic, 1984). This effect does not appear to involve cAMP or cGMP, since levels of these nucleotides are unaffected by neurotensin (Kitagbi and Freychet, 1979a). The relaxation is however, inhibited by apamin (Holzer

PHARMACOLOGY OF THE ENTERIC NERVOUS SYSTEM

259

et al., 1989), suggesting that neurotensin may open small conductance calcium-activated potassium channels. Neurotensin induces a direct relaxation of the guinea-pig ileum CM, which is apaminsensitive and blocked by antibodies to neurotensin. Furthermore, relaxation induced by nerve stimulation is partly blocked by neurotensin antibodies (Goedert, Hunter and Ninkovic, 1984). This led to the proposal that neurotensin is a NANC inhibitory transmitter in the guinea-pig ileum. In the rat stomach strips, neurotensin causes a contractile response, which results from direct activation of smooth muscle receptors (Nguyen et al., 1997). In canine ileal circular muscle devoid of myenteric plexus, neurotensin causes an initial hyperpolarization associated with inhibition of contractile activity, followed by an excitatory phase. The neurotensin-induced inhibitory effect is mediated by activation of apamin-sensitive, calcium-dependent potassium channels (Christinck, Daniel and Fox-Threlkeld, 1992). In rat proximal colon, neurotensin causes a summation of excitatory and inhibitory effects, which are dependent on the influx of Ca2+ via L-type channels (Mule and Serio, 1997). Neurotensin contracts human colonic circular smooth muscle strips. The insensitivity of this contractile effect to atropine, levocablastine (a NTS2 receptor agonist) or TTX and the blockade with SR 48692 and SR 142948 (two selective NTS1 receptor antagonists) is consistent with the involvement of non-neuronal NTS1 receptors (Croci et al., 1999). Neurotensin infused intravascularly increases the frequency of peristaltic waves and the volume of fluid emptied per unit time in the guinea-pig ileum (Barthó et al., 1982a). However, atropine-resistant peristalsis is initially stimulated and then inhibited by neurotensin (Barthó et al., 1982a). These results were interpreted by the authors to represent the stimulatory effect of neurotensin on ACh and SP release and its subsequent direct inhibitory effect on smooth muscle. The non peptide NTS1 receptor antagonist SR 48692 abolished both effects. The inhibitory action of neurotensin apparently involves an apamin-sensitive mechanism (Ohashi et al., 1996). Electrophysiological studies demonstrated that neurotensin mimics the slow EPSP in 48% of S neurons and 26% of AH neurons in the guinea-pig myenteric plexus. Calciumdependent hyperpolarization was recorded in 18% of AH neurons but not in S neurons (Williams, Katayama and North, 1979). MOTILIN Motilin is a 22-aminoacid residue polypeptide, which is released at about 100-min intervals during the interdigestive state, while the presence of nutrients in the duodenum strongly suppresses its endogenous release. Although, strictly speaking, motilin is a gastrointestinal hormone rather than a neurotransmitter, it will be briefly dealt with in this chapter in light of the interesting developments on motilin agonists as gastrointestinal prokinetics (Peeters, 1993; Clark et al., 1999; Siani et al., 2000). For the history of the discovery of motilin and erythromycin derivatives as prokinetics, the reader is referred to the reviews by Itoh (1997) and Peeters (1993). In this context, it will suffice to remember that erythromycin and its derivatives, which are devoid of antibacterial activity such as EM574 (motilin-like macrolides or motilides), are strong motilin

260

INNERVATION OF THE GASTROINTESTINAL TRACT

agonists and profoundly stimulate gut motility. This effect is brought about both through a direct action on smooth muscle motilin receptors (Depoortere, Peeters and Vantrappen, 1991; Hasler, Heldsinger and Owyang, 1992) and through neural receptors (Peeters, 1993; Van Assche, Depoortere and Peeters, 1995). Smooth muscle contraction is induced through a nifedipine-sensitive mechanism (which therefore involves entry of extracellular calcium), but there is evidence that the intracellular calcium pool may also have a role (Matthijs, Peeters and Vantrappen, 1989; Peeters, Matthijs and Vantrappen, 1991; Depoortere and Peeters, 1995; Farrugia et al., 1995). A neurally mediated effect of motilides in the gastrointestinal tract is indicated, at least in some models, by the observation that their effects are antagonized by atropine, hexamethonium or ondansetron (Peeters, 1993; Mizumoto et al., 1993; Fiorucci, Santucci and Morelli, 1993; Shiba et al., 1995; Van Assche, Depoortere and Peeters, 1995; Parkman, Pagano and Ryan, 1996; Boivin et al., 1997). Indeed, vagal and non-vagal cholinergic pathways as well as serotonergic pathways seem to play an important role in mediating the motor effect of motilides (Itoh, 1997; Inatomi et al., 1996). There are also data documenting the presence of motilin receptors in the central nervous system (Itoh, 1997). Inhibitory effects sometimes observed with motilides have been ascribed to stimulation of inhibitory neural pathways or Ca2+-channel blockade (Parkman, Pagano and Ryan, 1996; Minocha and Galligan, 1991; Furness et al., 1999). These effects, however, usually occur at high concentrations and are resistant to motilin receptor antagonists (Furness et al., 1999). GASTRIN-RELEASING PEPTIDE AND NEUROMEDIN B GRP is a 27-amino acid peptide which has been found in a variety of mammals along with the related peptide, neuromedin B (Spindel, 1986; Furness and Costa, 1987; Ohki-Hamazaki, 2000). Bombesin and alytesin are tetradecapeptides, isolated from amphibia, which share a similar sequence to the C-terminus of GRP (Ersparmer et al., 1972; Ersparmer and Melchiorri, 1980). Ranatensin is an endecapeptide which is also related to GRP (Ersparmer et al., 1972; Nakajima, 1981). GRP-immunoreactive neurons are found in the myenteric and submucous plexuses in the guinea-pig small and large intestine (Schultzberg et al., 1980; Hutchison, Dimaline and Dockray, 1981; Messenger, 1993). Myenteric neurons containing GRP include descending interneurons and long descending inhibitory motor neurons to the CM (Costa et al., 1984; Messenger, 1993). These neurons are also immunoreactive for VIP, NOS and dynorphin (Brookes, Steele and Costa, 1992). GRP has also been identified in LM-MP synaptosomes by radioimmunoassay (McDonald et al., 1988). In the guinea-pig gastric fundus, GRP is present in ascending excitatory muscle motor neurons and in neurons involved in the regulation of mucosal functions (Pfannkuche et al., 2000). Neuromedin B and GRP receptor mRNAs have been detected in human and rabbit colonic smooth muscle cells (Bitar and Zhu, 1993). In the cat, GRP and neuromedin B evoke concentration-dependent contractions in circular strips of esophagus and fundus (Milusheva et al., 1998). Similar responses have been detected in longitudinal strips of the duodenum and ileum longitudinal muscles (Kortezova et al., 1994; Milusheva et al., 1998).

PHARMACOLOGY OF THE ENTERIC NERVOUS SYSTEM

261

Bombesin, GRP and neuromedin B produce TTX-sensitive, tonic contractions of the guinea-pig ileum LM-MP preparation (Zetler, 1980; Pfannkuche, 1992). The contractions are inhibited by atropine, suggesting that these peptides stimulate the release of ACh from LM motor neurons (Ersparmer et al., 1972; Zetler, 1980; Garzón et al., 1985). Bombesin also stimulates SP release from LM-MP preparations in a TTX-sensitive manner (Holzer, 1984). Bombesin-, GRP- and neuromedin B-induced contractions are not altered by mepyramine, nicotine or methysergide (Ersparmer et al., 1972; Pfannkuche, 1992), suggesting that the peptides do not stimulate ACh release from cholinergic interneurons and that the contraction is independent of histamine and 5-HT. Bombesin increases the tone of guinea-pig ileum CM and induces rhythmic contractions (Holzer, Lembeck and Donnerer, 1980) in a TTX-sensitive way. However, this effect is not inhibited by tropicamide or hexamethonium (Holzer, Lembeck and Donnerer, 1980), suggesting that bombesin does not stimulate cholinergic CM motor neurons or interneurons. GRP modulates the mechanical activity of the human ileocaecal region in vitro. Longitudinal strips show a concentration-dependent increase in the rhythmic activity, whereas the circular strips react with a small decrease in tone. These effects are not affected by TTX, suggesting a direct action of GRP on smooth muscle probably mediated by distinct receptors (Vadokas et al., 1997). A study by Barthó et al. (1982a) demonstrated that intra-arterial infusion of bombesin in vitro stimulates peristalsis. This effect has not been investigated further. However, it is possible that the stimulation of peristalsis is a result of slow depolarization of excitatory CM motor neurons, causing increased excitability and increased release of a non-cholinergic transmitter. Bombesin and GRP mimic slow EPSPs in AH neurons. Depolarization is associated with an increase in input resistance which may be due to the closing of a calcium-dependent potassium channel (Zafirov et al., 1985). GALANIN Galanin is a 29-amino acid peptide originally isolated from porcine intestine (Tatemoto et al., 1983). Galanin immunoreactivity in guinea-pig ileal myenteric neurons is localized in specific subpopulations: neurons projecting to the mucosa, descending inhibitory motor neurons to the CM, descending interneurons and LM motor neurons (Melander et al., 1985; Furness et al., 1987). Galanin has also been identified by radioimmunoassay and high pressure liquid chromatography in LM-MP synaptosomes (McDonald et al., 1988). Galanin is present in anally projecting pathways in the guinea-pig colon (Messenger, 1993) and colocalizes with NOS and VIP in canine antrum, small bowel and colon (Wang et al., 1998). Galanin does not alter basal activity of the LM of guinea-pig ileum (Kuwahara, Ozaki and Tanaihara, 1989) and has no effect on neurally-evoked twitch responses (Kuwahara, Ozaki and Tanaihara, 1989, 1990). More recently, however, galanin was found to possess a marked inhibitory effect on twitch responses, probably via activation of GAL1 receptors (Sternini et al., 1999b), and on tritiated ACh release (Mulholland, Schoeneich and Flowe, 1992). Galanin does not alter the basal activity of the CM of guinea-pig ileum, but inhibits TTX-sensitive, electrically evoked contractions (Kuwahara, Ozaki and Tanaihara, 1989,

262

INNERVATION OF THE GASTROINTESTINAL TRACT

1990). Galanin inhibits both the hyoscine-sensitive and -resistant components of the CM twitch (Kuwahara, Ozaki and Tanaihara, 1989), suggesting that it may act by inhibiting release of ACh and a tachykinin from excitatory CM motor neurons. Yau, Dorsett and Youther (1986a) demonstrated that galanin inhibits the electrically stimulated release of [3H]ACh from the myenteric plexus. It has no effect on ACh-induced contractions in the presence of TTX (Kuwahara, Ozaki and Tanaihara, 1989). In agreement with this, Grider and Makhlouf (1988a) found that galanin alone had no contractile or relaxant effect on isolated circular smooth muscle cells. However, galanin potentiated the relaxant response to VIP, isoproteronol and dibutyryl cAMP. This effect was blocked by apamin and by tetraethylammonium at high concentrations although apamin and tetraethylammonium did not alter the response to VIP, isoprenaline or dibutyryl cyclic AMP alone (Grider and Makhlouf, 1988a). Thus galanin appears to act by opening potassium channels in the smooth muscle which in turn enhances the relaxation produced by agents increasing cyclicAMP levels. Galanin may contract ileal circular muscle of various species (pig, guineapig, rat and rabbit) and relaxes dog ileum by a direct myogenic effect (Botella et al., 1992). Recently, out of the three galanin receptors so far characterized, a single receptor subtype (i.e. the GAL2 receptor) was recognized in the rat intestine (Waters and Krause, 2000). Galanin hyperpolarizes myenteric neurons, decreases input resistance and suppresses excitability. These effects mimic slow IPSPs in myenteric neurons (Palmer et al., 1986b). Galanin also inhibits fast EPSPs in S neurons. However, this does not appear to be due to an action at a presynaptic inhibitory receptor, since galanin also inhibits the depolarization evoked by exogenous ACh (Tamura, Palmer and Wood, 1987). NEUROMEDIN U Neuromedin U was originally isolated from porcine spinal cord (Minamino, Kangawa and Matsuo, 1985), but it has also been extracted from the guinea-pig ileum (Murphy et al., 1990), and identified by radioimmunoassay and high pressure liquid chromatography (Augood, Keast and Emsom, 1988). Neuromedin U immunoreactivity is localized in Dogiel type II, calbindin-positive and -negative myenteric neurons, anally projecting myenteric neurons and myenteric neurons projecting to the submucous plexus (Furness et al., 1989a). It is not found in motor neurons. In the rat gastric fundus CM and in turtle small intestine LM, neuromedin U evokes a concentration-dependent contraction (Bockman et al., 1989; Maggi et al., 1990b; Benito-Orfila et al., 1991). Since these responses are unaffected by atropine or TTX (Bockman et al., 1989), they are likely to be due to a direct effect of neuromedin U on the muscle. However CM from frog stomach and LM from rat and frog small intestine are insensitive to neuromedin U. Similarly, neuromedin U does not have any significant effect on LM from the guinea-pig ileum (Minamino, Kangawa and Matsuo, 1985; Bockman et al., 1989). Human neuromedin U cDNA has been cloned and characterized (Austin et al., 1995). Neuromedin U mRNA is expressed in human small intestine and stomach (Hedrick et al., 2000). Recently, in the human and rat intestine, a receptor for neuromedin U has been identified, which probably corresponds to the orphan G protein-coupled receptor FM3 (Raddatz et al., 2000; Szerekes et al., 2000; Fujii et al., 2000).

PHARMACOLOGY OF THE ENTERIC NERVOUS SYSTEM

263

BRADYKININ Bradykinin receptors of the B2 but not B1 subtype are located on the LM and CM but not the myenteric plexus of guinea-pig ileum (Manning et al., 1982; Haasemann et al., 1991; Tousignant et al., 1991; Ransom et al., 1992), and on colonic CM (Hasler et al., 1995). The B2 receptor has been cloned and expressed in Xenopus oocytes; it is a G proteincoupled receptor that appears to stimulate a chloride current (McEachern et al., 1991). Bradykinin induces contraction of LM-MP preparations from the guinea-pig ileum (Garzón et al., 1985) which is largely unaffected by atropine (Khairallah and Page, 1961, 1963; Garzón et al., 1987). However, bradykinin reverses the inhibitory effects of opioids on the cholinergic twitch in a hyoscine-sensitive manner (Goldstein et al., 1983). Thus bradykinin appears to stimulate ACh release from cholinergic motor neurons to a variable degree. The non-cholinergic component of the contraction may be mediated by the release of a non-cholinergic transmitter or by a direct action of bradykinin on the smooth muscle. Bradykinin produces a biphasic response in the guinea-pig ileum CM, i.e. contraction followed by relaxation (Calixto and Medeiros, 1991). Neither response is mimicked by a B1 receptor agonist. Both the contraction and relaxation are unaffected by atropine, prazosin, yohimbine, propranolol, pyrilamine, [Leu8]BK1–8, TTX, glibenclamide or phorbol ester (Calixto and Medeiros, 1991). Thus neither response involves the action of endogenous transmitters at muscarinic, α1-adrenoceptor, α2-adrenoceptor, β-adrenoceptor, H1 or B1 receptors. Since the effect of bradykinin is TTX-insensitive, it is likely to be a direct effect on the smooth muscle. Furthermore, bradykinin does not open ATP-sensitive potassium channels or act via stimulation of protein kinase C activity. Bradykinin induces colonic smooth muscle contraction via B2 receptors, with phosphoinositide turnover activation and adenylate cyclase inhibition (Hasler et al., 1995). In this preparation, the bradykinin-induced contraction is mediated by influx of extracellular Ca2+ via non-selective cation channels and, in part, by the release of Ca2+ from loosely bound internal store (Zagorodnyuk, Santicioli and Maggi, 1998). In the taenia caeci, bradykinin produces a relaxation followed by a contraction. This response is inhibited by the B2 receptor antagonist Hoe 140. Since both phases of the response are TTXinsensitive, it is likely that they represent a direct effect on smooth muscle (Field et al., 1994). In the opossum lower oesophageal sphincter, bradykinin induces relaxation which is apamin-sensitive, followed by a contraction (Saha, Sengupta and Goyal, 1991). The contraction is inhibited by nifedipine, but not TTX, suggesting a direct effect of bradykinin on the muscle. In the rat isolated duodenum, bradykinin induces a biphasic response (relaxation followed by contraction), which is mediated by activation of B2 receptors (Rhaleb and Carretero, 1994). Previous evidence demonstrated that the B2 receptor-mediated relaxation was apamin sensitive (Hall and Morton, 1991; Greisbacher, 1992). Longitudinal strips of postmortem human ileum display a strong contractile response to bradykinin. This effect is mediated by constitutive B2 receptors and inducible B1 receptors (Zuzack et al., 1996).

ANGIOTENSIN Immunohistochemical studies suggest that the octapeptide angiotensin is present in enteric neurons of the guinea-pig ileum, although angiotensin has not been extracted from the external musculature (Furness and Costa, 1982).

264

INNERVATION OF THE GASTROINTESTINAL TRACT

Angiotensin contracts the LM of the guinea-pig ileum. This effect was originally believed to be due to, in part to, stimulation of ACh release from LM motor neurons for several reasons. The contraction was reduced by 60–85% by atropine and enhanced by anticholinesterase inhibitors (Khairallah and Page, 1961; Robertson and Rubin, 1962; Khairallah and Page, 1963; Blair-West and McKenzie, 1966). Hexamethonium does not inhibit the angiotensin-evoked contraction (Khairallah and Page, 1961; Robertson and Rubin, 1962; Khairallah and Page, 1963). However Paiva et al. (1976) demonstrated that the concentrations of atropine used in previous studies resulted in non-specific inhibition of contraction produced by a variety of agonists including histamine and bradykinin, both of which are believed to act directly on the muscle. The mechanism whereby angiotensin contracts the LM has therefore yet to be elucidated, but a direct effect on smooth muscle cells (as also observed in CM: Shimuta et al., 1999) is likely (Smith, Taylor and Whiting, 1994). In addition, angiotensin was found to produce a concentration-dependent inhibitory effect on twitch responses in LM-MP preparations (Smith, Taylor and Whiting, 1994). By contrast, in rat colonic preparations, angiotensin was found to facilitate ACh release at high concentrations (Voderholzer, Allescher and Muller-Lissner, 1995). The pharmacological and electrophysiological effects of angiotensin on myenteric neurons are unknown. ENDOTHELIN-LIKE PEPTIDES The endothelin (ET) family of peptides was originally isolated from cultured porcine aortic endothelial cells (Yanagisawa et al., 1988; Saida, Mitsui and Ishida, 1989). However, endothelins have also been found in human intestine and in human and porcine spinal cord neurons (Giaid et al., 1989; Shinmi et al., 1989; Egidy et al., 2000). ET-1, ET-2, ET-3 and vasoactive intestinal contractor, a related peptide, produce contraction of the isolated guinea-pig longitudinal smooth muscle cells through an action on the ETB receptor subtype. This contraction is dependent on intracellular and extracellular calcium (Yoshinaga et al., 1992). Apart from a direct effect on the smooth muscle, endothelins inhibit nerve-evoked LM contractions and ACh release (Maggi et al., 1989; Wiklund et al., 1989). This effect may be mediated by the ETA receptor subtype (Maggi et al., 1989). Endothelin receptors are present in enteric neurons and seem to play an important role in the development of the enteric nervous system (Gershon, 1997). ETA receptor activation stimulates, whereas ETB receptor activation inhibits peristalsis in guinea-pig isolated small intestine. The ability of BQ-788, a selective ETB receptor antagonist, to facilitate peristalsis per se, points to a physiological role of ETB receptors in peristaltic motor regulation (Shahbazian and Holzer, 2000).

CONCLUSIONS The intestine is capable of complex behaviours which allow mixing, propulsion, digestion and absorption of food. The mechanism whereby the intrinsic neurons of the enteric nervous system and the circular smooth muscle coordinate their activities to produce propagated contractions has been the subject of intensive investigations for almost a century. Peristalsis was originally described in 1899 and studied by a few investigators

PHARMACOLOGY OF THE ENTERIC NERVOUS SYSTEM

265

at the turn of the century. A burst of research activity occurred in the 1950s and 1960s. Since the 1980s interest has been focussed on the characterization of neuronal circuitries and mechanisms underlying peristalsis. An integrated approach to study a variety of parameters of intestinal function has shed some light on the factors which determine the threshold distension for triggering emptying, the strength of the contraction, the mechanism whereby the contraction propagates anally and the speed of propagation. The intestine contains a large number of neuroactive substances and an equally large number of receptors for a variety of chemicals. As a consequence, the LM-MP preparation of the guinea-pig ileum has been widely used by pharmacologists to test new drugs. We therefore have quite detailed information about the effect of drugs on LM motor neurons and the LM. Conversely, drugs are tested less frequently for their effects on CM motor neurons or CM activity. However, the irony is that the role of the LM in intestinal motor function is less understood than that of the CM. Furthermore, although we have detailed information on the effects of exogenous agents on a layer of muscle which appears to have a scarce role in peristalsis, we know little of the role of most endogenous chemicals. It is clear therefore, that future studies need to address this imbalance. The development of selective antagonists to receptors for neuropeptides, in particular, will greatly assist the ultimate goal of understanding the neuronal basis of peristalsis.

ACKNOWLEDGMENTS The authors wish to thank Dr. Amalia Di Nucci and Dr. Cristina Giaroni for their helpful suggestions during drawing up of the chapter, and Ms. Elena Giovanetti for assistance in preparing the reference list.

REFERENCES Agoston, D.V., Fahrenkrug, J., Mikkelson, J.D. and Whittaker, V.P. (1989). A peptide with N-terminal histidine and C-terminal isoleucine amide (PHI) and vasoactive intestinal peptide (VIP) are copackaged in myenteric neurones of the guinea pig ileum. Peptides, 10, 571–573. Ahmad, S., Allescher, H.D., Jang, E., Rausa, J. and Daniel, E.E. (1991). α2-Adrenergic receptors on the circular smooth muscle of canine small intestine. Canadian Journal of Physiology and Pharmacology, 69, 837–840. Alberts, P. and Stjärne, L. (1982). Facilitation, and muscarinic and α-adrenergic inhibition of the secretion of 3 H-acetylcholine and 3H-NA from guinea-pig ileum myenteric nerve terminals. Acta Physiologica Scandinavica, 116, 83–92. Alberts, P., Ögren, V.R. and Sellstöm, Å.I. (1985). Role of adenosine 3′,5′-cyclic monophosphate in adrenoceptormediated control of 3H-noradrenaline secretion in guinea-pig myenteric nerve terminals. Naunyn-Schmiedebergs Archives of Pharmacology, 330, 114–120. Alexander, S.P.H. and Peters, J.A. (2000). TiPS Receptor and Ion Channel Nomenclature (11th edition). Trends in Pharmacological Sciences Supplement 2000. London: Elsevier Science. Al-Humayyd, M. and White, T.D. (1985). 5-Hydroxytryptamine releases adenosine 5′-triphosphate from nerve varicosities isolated from the myenteric plexus of guinea-pig ileum. British Journal of Pharmacology, 84, 27–34. Allescher, H.D., Ahmad, S., Kostolanska, F., Kwan, C.Y. and Daniel, E.E. (1989). Modulation of pyloric motor activity via adrenergic receptors. Journal of Pharmacology and Experimental Therapeutics, 249, 652–659. Allescher, H.D., Storr, M., Brechmann, C., Hahn, A. and Schusdziarra, V. (2000). Modulatory effect of endogenous and exogenous opioids on the excitatory reflex pathway of the rat ileum. Neuropeptides, 34, 62–68.

266

INNERVATION OF THE GASTROINTESTINAL TRACT

Ando-Yamamoto, M., Hayashi, H., Taguchi, Y., Fukui, H., Watanabe, T. and Wada, H. (1986). Demonstration of immunohistochemical and immunochemical cross-reactivity of L-histidine and L-dopa decarboxylases using antibodies against the two enzymes. Biochemical and Biophysical Research Communications, 141, 306–312. Arch, J.R.S. and Kaumann, A.J. (1993). β3- and atypical β-adrenoceptors. Medicinal Research and Review, 13, 663–729. Arch, J.R.S., Ainsworth, A.T., Cawthorne, M.A., Piercy, V., Sennitt, M.V., Thody, V.E., Wilson, C. and Wilson, S. (1984). Atypical β-adrenoceptor on brown adipocytes as target for anti-obesity drugs. Nature, 309, 163–165. Augood, S.J., Keast, J.R. and Emsom, P.C. (1988). Distribution and characterization of neuromedin U-like immunoreactivity in rat brain and intestine and in guinea pig intestine. Regulatory Peptides, 20, 281–292. Austin, C., Lo, G., Nandha, K.A., Meleagros, L. and Bloom, S.R. (1995). Cloning and characterization of the cDNA encoding the human neuromedin U (NmU) precursor: NmU expression in the human gastrointestinal tract. Journal of Molecular Endocrinology, 14, 157–169. Barker, L.A. and Ebersole, B.J. (1982). Histamine H2-receptors on guinea-pig ileum myenteric plexus neurons mediate the release of contractile agents. Journal of Pharmacology and Experimental Therapeutics, 221, 69–75. Barocelli, E., Ballabeni, V., Caretta, A., Bordi, F., Silva, C., Morini, G. and Impicciatore, M. (1993). Pharmacological profile of new thioperamide derivatives at histamine peripheral H1-, H2-, and H3-receptors in guineapig. Agents and Actions, 38, 158–164. Barthó, L. and Holzer, P. (1995). The inhibitory modulation of guinea-pig intestinal peristalsis caused by capsaicin involves calcitonin gene-related peptide and nitric oxide. Naunyn-Schmiedebergs Archives of Pharmacology, 353, 102–109. Barthó, L., Holzer, P., Donnerer, J. and Lembeck, F. (1982a). Effects of substance P, cholecystokinin octapeptide, bombesin, and neurotensin on the peristaltic reflex of the guinea-pig ileum in the absence and in the presence of atropine. Naunyn-Schmiedebergs Archives of Pharmacology, 321, 321–328. Barthó, L., Holzer, P., Donnerer, J. and Lembeck, F. (1982b). Evidence for the involvement of substance P in the atropine-resistant peristalsis of the guinea-pig ileum. Neuroscience Letters, 32, 69–74. Barthó, L., Holzer, P. and Lembeck, F. (1987). Is ganglionic transmission through nicotinic receptors essential for the peristaltic reflex in the guinea-pig ileum? Neuropharmacology, 26, 1663–1666. Barthó, L., Holzer, P., Lembeck, F., Lippe, I.Th. and Setnikar, I. (1987a). Evaluation of a new and potent cholecystokinin antagonist on motor responses of the guinea-pig intestine. British Journal of Pharmacology, 90, 753–761. Barthó, L., Pethö, G., Antal, A., Holzer, P. and Szolcsányi, J. (1987b). Two types of relaxation due to capsaicin in the guinea pig isolated ileum. Neuroscience Letters, 81, 146–150. Barthó, L., Holzer, P., Leander, S. and Lembeck, F. (1989). Evidence for an involvement of substance P, but not cholecystokinin-like peptides, in hexamethonium-resistant intestinal peristalsis. Neuroscience, 28, 211–217. Barthó, L., Santicioli, P., Patacchini, R. and Maggi, C.A. (1992). Tachykininergic transmission to the circular muscle of the guinea-pig ileum: evidence for the involvement of NK2 receptors. British Journal of Pharmacology, 105, 805–810. Bartolini, A., Galli, A., Giotti, A., Giuliani, S., Malmberg-Aiello, P., Patacchini, R. and Renzi, G. (1985). Bicuculline: a GABAA antagonist with other pharmacological actions. British Journal of Pharmacology, 84, 547P. Bauer, V. (1981). Distribution and types of adrenoceptors in the guinea-pig ileum: action of α- and β-adrenoceptor agonists. British Journal of Pharmacology, 72, 201–210. Bauer, V. (1982). Distribution and types of adrenoceptors in the guinea-pig ileum: the action of α- and β-adrenoceptor blocking agents. British Journal of Pharmacology, 76, 569–578. Bauer, V. and Kuriyama, H. (1982). Homogeneous and non-homogeneous distribution of inhibitory and excitatory adrenoceptors in the longitudinal muscle of the guinea-pig ileum. British Journal of Pharmacology, 76, 603–611. Bauer, A.J. and Szurszewski, J.H. (1991). Effect of opioid peptides on circular muscle of canine duodenum. Journal of Physiology (London), 434, 409–422. Bayguinov, O., Keef, K.D., Hagen, B. and Sanders, K.M. (1999). Parallel pathways mediate inhibitory effects of vasoactive intestinal polypeptide and nitric oxide in canine fundus. British Journal of Pharmacology, 126, 1543–1552. Beani, L., Bianchi, C. and Crema, A. (1969). The effect of catecholamines and sympathetic stimulation on the release of acetylcholine from the guinea-pig colon. British Journal of Pharmacology, 36, 1–17. Benito-Orfila, M.A., Domin, J., Nandha, K.A. and Bloom, S.R. (1991). The motor effect of neuromedin U on rat stomach in vitro. European Journal of Pharmacology, 193, 329–333.

PHARMACOLOGY OF THE ENTERIC NERVOUS SYSTEM

267

Bennett, A. (1965). Effect of gastrin on isolated smooth muscle preparations. Nature, 208, 170–173. Berezin, I., Snyder, S.H., Bredt, D.S. and Daniel, E.E. (1994). Ultrastructural localization of nitric oxide synthase in canine small intestine and colon. American Journal of Physiology, 266, C981–C989. Bertrand, P.P. and Galligan, J.J. (1992). Alfaxalone, pentobarbital an diazepam potentiate γ-aminobutiric acidinduced depolarizations in single myenteric neurons of guinea pig intestine. Journal of Pharmacology and Experimental Therapeutics, 262, 677–682. Bertrand, P.P. and Galligan, J.J. (1994). Contribution of chloride conductance increase to slow EPSC and tachykinin current in guinea-pig myenteric neurones. Journal of Physiology, 481, 47–60. Bertrand, P.P. and Galligan, J.J. (1995). Signal-trasduction pathways causing slow synaptic excitation in guinea pig myenteric AH neurones. American Journal of Physiology, 269, G710–G720. Bertrand, P.P., Kunze, W.A., Furness, J.B. and Bornstein, J.C. (2000). The terminals of myenteric intrinsic primary afferent neurons of the guinea-pig ileum are excited by 5-hydroxytryptamine-3 receptors. Neuroscience, 101, 459–469. Bian, X.C., Bertrand, P.P., Furness, J.J. and Bornstein, J.C. (2000). Evidence for functional NK1-tachykinin receptors on motor neurones supplying the circular muscle of guinea-pig small and large intestine. Neurogastroenterology and Motility, 12, 307–315. Bianchetti, A. and Manara, L. (1990). In vitro inhibition of intestinal motility by phenylethanolaminotetralines: evidence of atypical β-adrenoceptors in rat colon. British Journal of Pharmacology, 100, 831–839. Bill, S.J., Dover, G.M. and Rhodes, K.F. (1990). Demonstration of 5-HT1A agonist actions of 5-carboxyamidotryptamine in the isolated transmurally stimulated ileum of the guinea-pig. British Journal of Pharmacology, 100, 483P. Bitar, K.N. and Zhu, X.X. (1993). Expression of bombesin-receptor subtypes and their differential regulation of colonic smooth muscle contraction. Gastroenterology, 105, 1672–1680. Blair-West, J.R. and McKenzie, J.S. (1966). Sodium concentration and the effect of angiotensin II on ileal smooth muscle. Experientia, 22, 291–292. Blandizzi, C., Tognetti, M., Colucci, R. and Del Tacca, M. (2000). Histamine H3 receptors mediate inhibition of noradrenaline release from intestinal sympathetic nerves. British Journal of Pharmacology, 129, 1387–1396. Blue, D.R., Bond, R.A., Adham, N., Delmendo, R., Michel, A.D., Eglen, R.M., Whiting, R.L. and Clarke, D.E. (1990). Antagonist characterization of atypical β adrenoceptors in guinea pig ileum: blockade by alprenolol and dihydroalprenolol. Journal of Pharmacology and Experimental Therapeutics, 252, 1034–1042. Bockman, C.S., Abel, P.W., Hicks, J.W. and Conlon, J.M. (1989). Evidence that neuromedin U may regulate gut motility in reptiles but not in mammals. European Journal of Pharmacology, 171, 255–257. Boeckxstaens, G.E., Pelckmans, P.A., Bult, H., De Man, J.G., Herman, A.G. and van Maercke, Y.M. (1991). Evidence for nitric oxide as mediator of non-adrenergic non-cholinergic relaxations induced by ATP and GABA in the canine gut. British Journal of Pharmacology, 102, 434–438. Bogers, J.J., Pelckmans, P.A., Boeckxstaens, G.E., De Man, J.G., Herman, A.G. and Van Maercke, Y.M. (1991). The role of nitric oxide in serotonin-induced relaxations in the canine terminal ileum and ileocolonic junction. Naunyn-Schmiedebergs Archives of Pharmacology, 344, 716–719. Boivin, M., Pinelo, L.R., St.-Pierre, S. and Poitras, P. (1997). Neural mediation of the motilin motor effect on the human antrum. American Journal of Physiology, 272, G71–G76. Bond, R.A. and Clarke, D.E. (1987). A response to isoprenaline unrelated to α- and β-adrenoceptor agonism. British Journal of Pharmacology, 91, 683–686. Bond, R.A. and Clarke, D.E. (1988). Agonist and antagonist characterization of a putative adrenoceptor with distinct pharmacological properties from the α- and β-subtypes. British Journal of Pharmacology, 95, 723–734. Borman, R.A. and Burleigh, D.E. (1997). 5-HT1D and 5-HT2B receptors mediate contraction of smooth muscle in human small intestine. Annals of New York Academy of Sciences, 812, 222–223. Bornstein, J.C., North, R.A., Costa, M. and Furness, J.B. (1984). Excitatory synaptic potentials due to activation of neurons with short projections in the myenteric plexus. Neuroscience, 11, 723–731. Botella, A., Delvaux, M., Frexinos, J. and Bueno, L. (1992). Comparative effects of galanin on isolated smooth muscle cells from ileum in five mammalian species. Life Sciences, 50, 1253–1261. Bredt, D.S. and Snyder, S.H. (1989). Nitric oxide mediates glutamate-linked enhancement of cGMP levels in the cerebellum. Proceedings of the National Academy of Sciences of the USA, 86, 9030–9033. Bredt, D.S., Hwang, P.M. and Snyder, S.H. (1990). Localization of nitric oxide synthase indicating a neural role for nitric oxide. Nature, 347, 768–770. Briejer, M.R., Akkermans, L.M.A., Meulemans, A.L., Lefebvre, R.A. and Schuurkes, J.A. (1993a). Substance P-induced contraction of the guinea-pig proximal colon through stimulation of post-junctional tachykinin NK1 receptors. European Journal of Pharmacology, 250, 181–183.

268

INNERVATION OF THE GASTROINTESTINAL TRACT

Briejer, M.R., Akkermans, L.M., Meulemans, A.L., Lefebvre, R.A. and Schuurkes, J.A. (1993b). Cisapride and a structural analogue, R 76,186, are 5-hydroxytryptamine4 (5-HT4) receptor agonists on the guinea-pig colon ascendens. Naunyn-Schmiedebergs Archives of Pharmacology, 347, 464–470. Brookes, S.J.H. and Costa, M. (1990). Identification of enteric motor neurones which innervate the circular muscle of the guinea-pig small intestine. Neuroscience Letters, 118, 227–230. Brookes, S.J.H., Steele, P.A. and Costa, M. (1991a). Calretinin immunoreactivity in cholinergic motor neurones, interneurones and vasomotor neurones in the guinea-pig small intestine. Cell Tissue Research, 263, 471–481. Brookes, S.J.H., Steele, P.A. and Costa, M. (1991b). Identification and immunohistochemistry of cholinergic and non-cholinergic circular muscle motor neurons in the guinea-pig small intestine. Neuroscience, 42, 863–878. Brookes, S.J.H., Steele, P.A. and Costa, M. (1992). The functional anatomy of the myenteric plexus. Proceedings of the Australian Physiology and Pharmacology Society, 23, 86–94. Brookes, S.J.H., Song, Z.-M., Steele, P.A. and Costa, M. (1992). Identification of motor neurones to the longitudinal muscle of the guinea-pig ileum. Gastroenterology, 103, 961–973. Broussard, D.L., Li, X., Pritchett, D.L., Lynch, D., Bannermann, P.G.C. and Pleasure, D. (1994). The expression of a NMDA receptor gene in the guinea-pig myenteric plexus. Neuroreport, 5, 973–976. Browing, K.N. and Lees, G.M. (2000). Inhibitory effects of NPY on ganglionic transmission in myenteric neurones of the guinea-pig descending colon. Neurogastroenterology and Motility, 12, 33–41. Brunn, G., Wessler, I., Anderson, G.P., Maclagan, J. and Racké, K. (1994). β-Adrenoceptor-mediated facilitation of endogenous noradrenaline release from rat isolated trachea. Naunyn-Schmiedebergs Archives of Pharmacology, 350, 459–463. Buchheit, K.-H., Engel, G., Mutschler, E. and Richardson, B. (1985). Study of the contractile effect of 5-hydroxytryptamine (5-HT) in the isolated longitudinal muscle strip from guinea-pig ileum. Naunyn-Schmiedebergs Archives of Pharmacology, 329, 36–41. Bülbring, E. and Crema, A. (1958). Observations concerning the action of 5-hydroxytryptamine on the peristaltic reflex. British Journal of Pharmacology, 13, 444–457. Bülbring, E. and Crema, A. (1959). The release of 5-hydroxytryptamine in relation to pressure exerted on the intestinal mucosa. Journal of Physiology, 146, 18–28. Bülbring, E. and Lin, R.C.Y. (1958). 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. Journal of Physiology, 140, 381–407. Bülbring, E. and Tomita, T. (1987). Catecholamine action on smooth muscle. Pharmacological Reviews, 39, 49–96. Burke, T.A. and Sanger, G.J. (1988). Regionally selective cholinergic stimulation by BRL 24924 in the human isolated gut. British Journal of Clinical Pharmacology, 26, 261–265. Burleigh, D.E. and Trout, S.J. (1985). Evidence against an acetylcholine releasing action of cisapride in the human colon. British Journal of Clinical Pharmacology, 20, 475–478. Burns, G.A., Stephens, K.E. and Benson, J.A. (1994). Expression of mRNA for the N-methyl-D-aspartate (NMDAR1) receptor by the enteric neurons of the rat. Neuroscience Letters, 170, 87–90. Bylund, D.B., Eikenberg, D.C., Hieble, J.P., Langer, S.Z., Lefkowitz, R.J., Minneman, K.P., Molinoff, P.B., Ruffolo, Jr., R.R. and Trendelenburg, U. (1994). IV. International Union of Pharmacology nomenclature of adrenoceptors. Pharmacological Reviews, 46, 121–136. Bywater, R.A.R. and Taylor, G.S. (1986). Non-cholinergic excitatory and inhibitory junction potentials in the circular smooth muscle of the guinea-pig ileum. Journal of Physiology, 374, 153–164. Calixto, J.B. and Medeiros, Y.S. (1991). Characterization of bradykinin mediating pertussis toxin-insensitive biphasic response in circular muscle of the isolated guinea pig ileum. Journal of Pharmacology and Experimental Therapeutics, 259, 659–665. Carter, D., Champney, M., Hwang, B. and Eglen, R.M. (1995). Characterization of a postjunctional 5-HT receptor mediating relaxation of guinea-pig isolated ileum. European Journal of Pharmacology, 280, 243–250. Chakder, S. and Rattan, S. (1998). Involvement of pituitary adenylate cyclase–activating peptide in opossum internal anal sphincter relaxation. American Journal of Physiology, 275, G769–G777. Chavkin, C. and Goldstein, A. (1981). Demonstration of a specific dynorphin receptor in guinea pig ileum myenteric plexus. Nature, 291, 591–593. Cherubini, E. and North, R.A. (1985a). Benzodiazepines both enhance γ-aminobutirate responses and decrease calcium action potentials in guinea-pig myenteric neurones. Neuroscience, 14, 309–315. Cherubini, E. and North, R.A. (1984a). Actions of γ-aminobutyric acid on neurons of guinea-pig myenteric plexus. British Journal of Pharmacology, 82, 93–100. Cherubini, E. and North, R.A. (1984b). Inhibition of calcium spikes and transmitter release by γ-aminobutyric acid in the guinea-pig myenteric plexus. British Journal of Pharmacology, 82, 101–105.

PHARMACOLOGY OF THE ENTERIC NERVOUS SYSTEM

269

Cherubini, E. and North, R.A. (1985b). µ and κ opioids inhibit transmitter release by different mechanisms. Proceedings of the National Academy of Sciences of the USA, 82, 1860–1863. Cherubini, E., Morita, K. and North, R.A. (1985). Opioid inhibition of synaptic transmission in the guinea-pig myenteric plexus. British Journal of Pharmacology, 85, 805–817. Chidiac, P., Hebert, T.E., Valiquette, M., Dennis, M. and Bouvier, M. (1994). Inverse agonist activity of β-adrenergic antagonists. Journal of Pharmacology and Experimental Therapeutics, 45, 490–499. Chihara, S. and Tomita, T. (1987). Mechanical and electrical responses to α-adrenoceptor activation in the circular muscle of guinea-pig stomach. British Journal of Pharmacology, 91, 789–798. Chijikwa, J.B. and Davison, J.S. (1978). The action of gastrin-like polypeptides on the peristaltic reflex in guinea-pig intestine. Journal of Physiology, 285, 68P–70P. Christinck, F., Daniel, E.E. and Fox-Threlkeld, J.E. (1992). Inhibitory and excitatory mechanisms of neurotensin action in canine intestinal circular muscle in vitro. Canadian Journal of Physiology and Pharmacology, 70, 1423–1431. Christofi, F.L. and Wood, J.D. (1993). Effects of PACAP on morphologically identified myenteric neurons in guinea pig small bowel. American Journal of Physiology, 264, G414–G421. Christofi, F.L., McDonald, T.J. and Cook, M.A. (1990). Neuropeptide release from isolated myenteric nerve endings derived from the guinea pig myenteric plexus. Regulatory Peptides, 30, 165–177. Ciccocioppo, R., Onori, L., Messori, E., Candura, S.M., Coccini, T. and Tonini, M. (1994). Role of nitric oxidedependent and -independent mechanisms in peristalsis and accommodation in the rabbit distal colon. Journal of Pharmacology and Experimental Therapeutics, 270, 929–937. Cipolla, G., Sacco, S., Crema, F., Moro, E., De Ponti, F. and Frigo, G.M. (2001). Gastric motor effects of triptans: open questions and future perspectives. Pharmacological Research, 43, 205–210. Clark, M.J., Wright, T., Bertrand, P.P., Bornstein, J.C., Jenkinson, K.M., Verlinden, M. and Furness, J.B. (1999). Erythromycin derivatives ABT 229 and GM 611 act on motilin receptors in the rabbit duodenum. Clinical and Experimental Pharmacology and Physiology, 26, 242–245. Cohen, M.L., Rosing, E., Wiley, K.S. and Slater, I.H. (1978). Somatostatin inhibits adrenergic and cholinergic neurotransmission in smooth muscle. Life Sciences, 23, 1659–1664. Cohen, M.L., Wiley, K.S., Yaden, E. and Slater, I.H. (1979). In vitro actions of somatostatin, D-Val1, D-Trp8somatostatin and glucagon in rabbit jejunum and guinea-pig ileum. Journal of Pharmacology and Experimental Therapeutics, 211, 423–429. Cohen, M.L., Granneman, J.G., Chaudhry, A., Schenck, K.W., Cushing, D.J. and Palkowitz, A.D. (1995). Is the atypical β-receptor in the rat stomach fundus the rat β3-receptor? Journal of Pharmacology and Experimental Therapeutics, 272, 446–451. Coleman, P.A., Denyer, L.H. and Sheldrick, K.E. (1987). β-adrenoceptors in guinea-pig gastric fundus – are they the same as the “atypical” β-adrenoceptors in rat adipocytes? British Journal of Pharmacology, 90, 40P. Corleto, V.D., Severi, C., Coy, D.H., Delle Fave, G., and Jensen, R.T. (1997). Colonic smooth muscle cells possess a different subtype of somatostatin receptor from gastric smooth muscle cells. American Journal of Physiology, 272, G689–G697. Corleto, V.D., Weber, H.C. and Jensen, R.T. (1999). Expression of somatostatin receptor subtypes on guinea-pig gastric and colonic smooth muscle cells. American Journal of Physiology, 277, G235–G244. Cosentino, M., De Ponti, F., Marino, F., Giaroni, C., Leoni, O., Lecchini, S. and Frigo, G.M. (1995). N-methylD-aspartate receptors modulate neurotransmitter release and peristalsis in the guinea-pig isolated colon. Neuroscience Letters, 183, 139–142. Costa, M. and Brookes, S.J.H. (1994). The enteric nervous system. The American Journal of Gastroenterology, 89, S129–S137. Costa, M. and Furness, J.B. (1979). On the possibility that an indoleamine is a neurotransmitter in the gastrointestinal tract. Biochemical Pharmacology, 28, 565–571. Costa, M. and Furness, J.B. (1983). The origins, pathways and terminations of neurons with VIP-like immunoreactivity in the guinea-pig small intestine. Neuroscience, 8, 665–676. Costa, M., Patel, Y., Furness, J.B. and Arimura, A. (1977). Evidence that some intrinsic neurons of the intestine contain somatostatin. Neuroscience Letters, 6, 215–222. Costa, M., Furness, J.B., Buffa, R. and Said, S.I. (1980a). Distribution of enteric neurons showing immunoreactivity for vasoactive intestinal polypeptide (VIP) in the guinea-pig intestine. Neuroscience, 5, 587–596. Costa, M., Furness, J.B., Llewellyn-Smith, I.J., Davies, B. and Oliver, J. (1980b). An immunohistochemical study of the projections of somatostatin-containing neurons in the guinea-pig intestine. Neuroscience, 5, 841–852. Costa, M., Furness, J.B., Cuello, A.C., Verhofstadt, A.A.J., Steinbusch, H.W.J. and Elde, R.P. (1982). Neurons with 5-hydroxytryptamine-like immunoreactivity in the enteric nervous system: their visualization and reactions to drug treatment. Neuroscience, 7, 351–363.

270

INNERVATION OF THE GASTROINTESTINAL TRACT

Costa, M., Furness, J.B., Yanaihara, N., Yanaihara, C. and Moody, T.W. (1984). Distribution and projections of neurons with immunoreactivity for both gastrin-releasing peptide and bombesin in the guinea-pig small intestine. Cell and Tissue Research, 235, 285–293. Costa, M., Furness, J.B., Pullin, C.O. and Bornstein, J. (1985). Substance P enteric neurons mediate noncholinergic transmission to the circular muscle of the guinea-pig intestine. Naunyn-Schmiedebergs Archives of Pharmacology, 328, 446–453. Costa, M., Furness, J.B. and Gibbins, I.L. (1986a). Chemical coding of enteric neurons. Progress in Brain Research, 68, 217–239. Costa, M., Furness, J.B. and Humphreys, C.M.S. (1986b). Apamin distinguishes two types of relaxation mediated by enteric nerves in the guinea-pig gastrointestinal tract. Naunyn-Schmiedebergs Archives of Pharmacology, 332, 79–88. Costa, M., Furness, J.B., Brookes, S.J.H., Bredt, D.S. and Snyder, S.H. (1991). Presence and chemical coding of neurons with nitric oxide synthase immunoreactivity in the guinea-pig small intestine. Proceedings of the Australian Physiology and Pharmacology Society, 22, 97P. Costa, M., Brookes, S.J., Steele, P.A., Gibbins, I., Burcher, E. and Kandiah, C.J. (1996). Neurochemical classification of myenteric neurons in the guinea-pig ileum. Neuroscience, 75, 949–967. Costall, B., Naylor, R.J. and Tan, C.C. (1983). Field stimulation-induced responses of circular smooth muscle from guinea-pig stomach. Naunyn-Schmiedebergs Archives of Pharmacology, 323, 155–161. Coulie, B., Tack, J., Maes, B., Geypens, B., De Roo, M. and Janssens, J. (1997). Sumatriptan, a selective 5-HT1 receptor agonist, induces a lag phase for gastric emptying of liquids in humans. American Journal of Physiology, 272, G902–G908. Coulie, B., Tack, J., Sifrim, D., Andrioli, A. and Janssens, J. (1999). Role of nitric oxide in fasting gastric fundus tone and in 5-HT1 receptor-mediated relaxation of gastric fundus. American Journal of Physiology, 276, G373–G377. Coupar, I.M. (1995). The peristaltic reflex in the rat ileum: evidence for functional µ- and δ-opiate receptors. Journal of Pharmacy and Pharmacology, 47, 643–646. Craig, D.A. and Clarke, D.E. (1990). Pharmacological characterization of a neuronal receptor for 5-hydroxytryptamine in guinea pig ileum with properties similar to the 5-hydroxytryptamine-4 receptor. Journal of Pharmacology and Experimental Therapeutics, 252, 1378–1386. Craig, D.A. and Clarke, D.E. (1991). Peristalsis evoked by 5-HT and renzapride: evidence for putative 5-HT4 receptor activation. British Journal of Pharmacology, 102, 563–564. Crist, J.R., He, X.D. and Goyal, R.K. (1992). Both ATP and the peptide VIP are inhibitory neurotransmitters in guinea-pig ileum circular muscle. Journal of Physiology (London), 447, 119–131. Croci, T., Aureggi, G., Guagnini, F., Manara, L., Gully, D., Fur, G.L. et al. (1999). In vitro functional evidence of different neurotensin-receptors modulating the motor response of human colonic muscle strips. British Journal of Pharmacology, 127, 1922–1928. Croci, T., Cecchi, R., Tarantino, A., Aureggu, G., Bianchetti, A., Boigegrain, R. et al. (1988). Inhibition of rat colon motility by stimulation of atypical β-adrenoceptors with new gut-specific agents. Pharmacological Research Communications, 20, 147–151. Currò D., Volpe, A.R. and Preziosi, P. (1996). Nitric oxide synthase activity and non-adrenergic non-cholinergic relaxation in the rat gastric fundus. British Journal of Pharmacology, 117, 717–723. Dammann, F., Fuder, H., Giachetti, A., Giraldo, E., Kilbinger, H. and Micheletti, R. (1989). AF-DX 116 differentiates between pre-junctional muscarine receptors located on noradrenergic and cholinergic nerves. Naunyn-Schmiedebergs Archives of Pharmacology, 339, 268–271. Davison, J.S. and Najafi-Farashah, A. (1982). Role of endogenous opiates in the peristaltic reflex. Journal of Physiology, 330, 87P–88P. Dawson, T.M., Bredt, D.S., Fotuhi, M., Hwang, P.M. and Snyder, S.H. (1991). Nitric oxide synthase and neuronal NADPH diaphorase are identical in brain and peripheral tissues. Proceedings of the National Academy of Sciences of the USA, 88, 7797–7801. De Man, J.G., Pelckmans, P.A., Boeckxstaens, G.E., Bult, H., Oosterbosch, L., Herman, A.G. et al. (1991). The role of nitric oxide in inhibitory non-adrenergic non-cholinergic neurotransmission in the canine lower oesophageal sphincter. British Journal of Pharmacology, 103, 1092–1096. De Ponti, F. (1997). Pharmacological criteria for the detection of β3-adrenoceptors. Is it time for a reappraisal? Trends in Pharmacological Sciences, 18, 52. De Ponti, F. (2000). Pharmacology of emesis and gastrointestinal motility: Implications for migraine. Functional Neurology, 15, 43–49. De Ponti, F., Gibelli, G., Crema, F. and Lecchini, S. (1995). Functional evidence for the presence of β3-adrenoceptors in the guinea-pig common bile duct and colon. Pharmacology, 51, 288–297. De Ponti, F., Gibelli, G., Croci, T., Arcidiaco, M., Crema, F. and Manara, L. (1996). Functional evidence of atypical β3-adrenoceptors in the human colon using the β3-selective adrenoceptor antagonist SR 59230A. British Journal of Pharmacology, 117,1374–1376.

PHARMACOLOGY OF THE ENTERIC NERVOUS SYSTEM

271

de Weert, A., von Schrenck, T., Gronewold, M., Freudenberg, F., Mirau, S., Schulz, M. et al. (1997). Characterization of CCK receptors in stomach smooth muscle: evidence for two subtypes. Biochimica et Biophysica Acta, 1327, 213–221. Deacon, C.F., Agoston, D.V., Nau, R. and Conlon, J.M. (1987). Conversion of neuropeptide K to neurokinin A and vesicular localization of neurokinin A and substance P in neurons of the guinea pig small intestine. Journal of Neurochemistry, 48, 141–146. Dehpour, A.R., Delfan, A., Mousavizadeh, K. and Mortazavi, S.R. (1994). Effects of atropine, pirenzepine, clonidine, and morphine on biphasic response of rat gastric fundus to field stimulation. General Pharmacology, 25, 951–955. Del Tacca, M., Soldani, G. and Crema, A. (1970). Experiments on the mechanism of action of caerulein at the level of the guinea-pig ileum and colon. Agents and Actions, 1, 176–182. Del Tacca, M., Soldani, G., Selli, M. and Crema, A. (1970). Action of catecholamines on release of acetylcholine from human taenia coli. European Journal of Pharmacology, 9, 80–84. Depoortere, I. and Peeters, T.L. (1995). Transduction mechanisms of motilin and motilides in rabbit duodenal smooth muscle. Regulatory Peptides, 55, 227–235. Depoortere, I., Peeters, T.L. and Vantrappen, G. (1991). Motilin receptors of the rabbit colon. Peptides, 12, 89–94. Derkach, V., Surprenant, A. and North, R.A. (1989). 5-HT3 receptors are membrane ion channels. Nature, 339, 706–709. Dettmar, P.W., Kelly, J. and MacDonald, A. (1986). α-Adrenoceptors in the rat colon. British Journal of Pharmacology, 88, 463P. Dietrich, C. and Kilbinger, H. (1995). Prejunctional M1 and postjunctional M3 muscarinic receptors in the circular muscle of the guinea-pig ileum. Naunyn-Schmiedebergs Archives of Pharmacology, 351, 237–243. Dietrich, C. and Kilbinger, H. (1996). 5-HT1A receptor-mediated inhibition of acetylcholine release from guinea pig myenteric plexus: potential mechanisms. Neuropharmacology, 35, 483–488. Donaldson, L.F., Hanley, M.R. and Villablanca, A.C. (1997). Inducible receptors. Trends in Pharmacological Sciences, 18, 171–181. Donnerer, J. and Lembeck, F. (1985). New models for the evaluation of opioid effects in the guinea-pig ileum. British Journal of Pharmacology, 85, 61–64. Donnerer, J., Barthó, L., Holzer, P. and Lembeck, F. (1984). Intestinal peristalsis associated with release of immunoreactive substance P. Neuroscience, 11, 913–918. Donnerer, J., Holzer, P. and Lembeck, F. (1984). Release of dynorphin, somatostatin and substance P from the vascularly perfused small intestine of the guinea-pig during peristalsis. British Journal of Pharmacology, 83, 919–925. Donnerer, J., Meyer, D.K., Holzer, P. and Lembeck, F. (1985). Release of cholecystokinin-immunoreactivity into the vascular bed of the guinea-pig small intestine during peristalsis. Naunyn Schmiedebergs Archives of Pharmacology, 328, 324–326. Drew, G.M. (1978). Pharmacological characterization of the presynaptic α-adrenoceptors regulating cholinergic activity in the guinea-pig ileum. British Journal of Pharmacology, 64, 293–300. Egidy, G., Juillerat-Jeanneret, L., Korth, P., Bosman, F.T. and Pinet, F. (2000). The endothelin system in normal human colon. American Journal of Physiology, 279, G211–G222. Eglen, R.M. and Watson, N. (1996). Selective muscarinic receptor agonists and antagonists. Pharmacology and Toxicology, 78, 59–68. Eglen, R.M., Reddy, H., Watson, N. and Challiss, A.J. (1994). Muscarinic receptor subtypes in smooth muscle. Trends in Pharmacological Sciences, 15, 114–119. Eglen, R.M., Swank, S.R., Walsh, L.K. and Whiting, R.L. (1990). Characterization of 5-HT3 and “atypical” 5-HT receptors mediating guinea-pig ileal contractions in vitro. British Journal of Pharmacology, 101, 513–520. Ek, B. (1985). Studies on mechanisms for beta-adrenoceptor mediated inhibition of colonic motility. Acta Physiologica Scandinavica, 125 (Supplement 546), 5–39. Ek, B. and Lundgren, B. (1982). Characterization of the β-adrenergic inhibition of motility in cat colon strips. European Journal of Pharmacology, 77, 25–31. Ek, B.A. and Nahorski, S.R. (1986). β-Adrenergic control of motility in the rat colon. II. Proportions of β1- and β2-adrenoceptors identified with 125I-(-)pindolol binding. Gastroenterology, 90, 408–413. Ek, B.A., Bjellin, L.A.C. and Lundgren, B.T. (1986). β-Adrenergic control of motility in the rat colon. I. Evidence for functional separation of the β1- and β2-adrenoceptor-mediated inhibition of colon activity. Gastroenterology, 90, 400–407. Ek, B., Jodal, M. and Lundgren, O. (1987). Autoradiographic location of β-adrenoceptor subtypes in cat colon smooth muscle. Acta Physiologica Scandinavica, 129, 353–360. Ekblad, E. (1999). Pharmacological evidence for both neuronal and smooth muscular PAC1 receptors and a VIPspecific receptor in the rat colon. Regulatory Peptides, 85, 87–92.

272

INNERVATION OF THE GASTROINTESTINAL TRACT

Ekblad, E., Wahlestedt, C., Håkanson, R., Sundler, F., Watanabe, T. and Wada, H. (1985). Is histamine a neurotransmitter in the gut? Evidence from histidine decarboxylase immunocytochemistry. Acta Physiologica Scandinavica, 123, 225–227. El-Sharkawy, T.Y. and Szurszewski, J.H. (1978). Modulation of canine antral circular smooth muscle by acetylcholine, noradrenaline and pentagastrin. Journal of Physiology (London), 279, 309–320. Elswood, C.J. and Bunce, K.T. (1992). Investigation of the 5-HT receptor mediating relaxation in guinea-pig proximal colon. Journal of Pharmacy and Pharmacology, 44, 264–266. Emorine, L.J., Marullo, S., Briend-Sutren, M.M., Patey, G., Tate, K., Delavier-Klutchko, C. and Strosberg, A.D. (1989). Molecular characterization of the human β3-adrenergic receptor. Science, 245, 1118–1121. Endo, Y., Uchida, T. and Kobayashi, S. (1986). Somatostatin neurons in the small intestine of the guinea-pig: a light and electron microscopic immunocytochemical study combined with nerve lesion experiments by laser irradiation. Journal of Neurocytology, 15, 725–731. Engel, G., Hoyer, D., Kalkman, H.O. and Wick, M.B. (1984). Identification of 5-HT2 receptors on longitudinal muscle of the guinea pig ileum. Journal of Receptor Research, 4, 113–126. Ersparmer, V. and Melchiorri, P. (1980). Active polypeptides: from amphibian skin to gastrointestinal tract and brain of mammals. Trends in Pharmacological Sciences, 1, 391–395. Ersparmer, V., Ersparmer, G.F., Inselvini, M. and Negri, L. (1972). Occurrence of bombesin and alytesin in extracts of the skin of three European discglossid frogs and pharmacological actions of bombesin on extracellular smooth muscle. British Journal of Pharmacology, 45, 333–348. Ersparmer, V., Melchiorri, P., Falconieri-Ersparmer, G., Negri, L., Corsi, R., Severini, C., Barra, D., Simmaco, M. and Kreil, G. (1989). Deltorphins: a family of naturally occurring peptides with high affinity and selectivity for δ opioid binding sites. Proceedings of the National Academy of Sciences of the USA, 86, 5188–5192. Farrugia, G., Rich, A., Rae, J.L., Sarr, M.G. and Szurszewski, J.H. (1995). Calcium currents in human and canine jejunal circular smooth muscle cells. Gastroenterology, 109, 707–717. Fehér, E. and Burnstock, G. (1986). Electron microscopic study of neuropeptide Y-containing nerve elements of the guinea pig small intestine. Gastroenterology, 91, 956–965. Feletou, M., Rodriguez, M., Beauverger, P., Germain, P., Imbert, J., Dromaint, S. et al. (1998). NPY receptor subtypes involved in the contraction of the proximal colon of the rat. Regulatory Peptides, 75–76, 221–229. Fernandez, E., Guo, X., Vergara, P. and Jimenez, M. (1998). Evidence supporting a role for ATP as non-adrenergic non-cholinergic inhibitory transmitter in the porcine ileum. Life Sciences, 62, 1303–1315. Field, J.L., Butt, S.K., Morton, I.K. and Hall, J.M. (1994). Bradykinin B2 receptors and coupling mechanisms in the smooth muscle of guinea-pig taenia caeci. British Journal of Pharmacology, 113, 607–613. Fiorucci, S., Santucci, S. and Morelli, A. (1993). 5-Hydroxytryptamine3-receptor antagonist modulates gallbladder emptying and motilin release induced by erythromycin. Digestive Diseases and Sciences, 38, 2236–2240. Florey, E. and McLennan, H. (1959). The effects of factor I and γ-aminobutyric acid on smooth muscle preparations. Journal of Physiology (London), 145, 66–76. Fontaine, J., Grivegnée, A. and Reuse, J.J. (1984). Adrenoceptors and regulation of intestinal tone in the isolated colon of the mouse. British Journal of Pharmacology, 81, 231–243. Fontaine, J., Reuse, J.J. and Van Nueten, J.M. (1973). Analysis of the peristaltic reflex in vitro: effects of some morphine-like drugs. Archives Internationales de Pharmacodynamie et de Therapie, 206, 376–380. Fontaine, J., Van Nueten, J.M. and Janssen, P.A.J. (1973). Analysis of the peristaltic reflex in vitro: effect of some antagonists. Archives Internationales de Pharmacodynamie et de Therapie, 203, 396–399. Fox, A.J. and Morton, I.K.M. (1989). Activities of some 5-HT analogues in 5-HT3 receptor mediated release of [3H]-ACh in the guinea-pig ileum. British Journal of Pharmacology, 96, 53P. Fox, A.J. and Morton, I.K.M. (1990). An examination of the 5-HT3 receptor mediating contraction and evoked [3H]-acetylcholine release in the guinea-pig ileum. British Journal of Pharmacology, 101, 553–558. Fox-Threlkeld, J.A., Daniel, E.E., Christinck, F., Woskowska, Z., Cipris, S. and McDonald, T.J. (1993). Peptide YY stimulates circular muscle contractions of the isolated perfused canine ileum by inhibiting nitric oxide release and enhancing acetylcholine release. Peptides, 14, 1171–1178. Fox-Threlkeld, J.A., McDonald, T.J., Woskowska, Z., Iesaki, K. and Daniel, E.E. (1999). Pituitary adenylate cyclase-activating peptide as a neurotransmitter in the canine ileal circular muscle. Journal of Pharmacology and Experimental Therapeutics, 290, 66–75. Foxx-Orenstein, A.E. and Grider, J.R. (1996). Regulation of colonic propulsion by enteric excitatory and inhibitory neurotransmitters. American Journal of Physiology, 271, G433–G437. Foxx-Orenstein, A.E., Kuemmerle, J.F. and Grider, J.R. (1996). Distinct 5-HT receptors mediate the peristaltic reflex induced by mucosal stimuli in human and guinea-pig intestine. Gastroenterology, 111, 1281–1290. Fozard, J.R. and Kilbinger, H. (1985). 8-OH-DPAT inhibits transmitter release from guinea pig enteric cholinergic neurons by activating 5-HT1A receptors. British Journal of Pharmacology, 86, 601P.

PHARMACOLOGY OF THE ENTERIC NERVOUS SYSTEM

273

Franck, H., Sweeney, K.M., Sanders, K.M. and Shuttleworth, C.W.R. (1997). Effects of a novel guanylate cyclase inhibitor on nitric oxide-dependent inhibitory neurotransmission in canine proximal colon. British Journal of Pharmacology, 122, 1223–1229. Frielle, T., Collins, S., Daniel, K.W., Caron, M.G., Lefkowitz, R.J. and Kobilka, B.K. (1987). Cloning of the cDNA for the human β1-adrenergic receptor. Proceedings of the National Academy of Sciences of the USA, 84, 7920–7924. Frigo, G.M., Lecchini, S., Falaschi, C., Del Tacca, M. and Crema, A. (1971). On the ability of caerulein to increase propulsive activity in the isolated small and large intestine. Naunyn-Schmiedebergs Archives of Pharmacology, 268, 44–58. Frigo, G.M., Galli, A., Lecchini, S. and Marcoli, M. (1987). A facilitatory effect of bicuculline on the enteric neurones in the guinea-pig isolated colon. British Journal of Pharmacology, 90, 31–41. Fuder, H. and Muscholl, E. (1995). Heteroreceptor-mediated modulation of noradrenaline and acetylcholine release from peripheral nerves. Reviews of Physiology, Biochemistry and Pharmacology, 126, 265– 412. Fuder, H., Kilbinger, H. and Müller, H. (1985). Organ selectivity of hexahydrosilanodifenidol in blocking pre-and postjunctional muscarinic receptors studied in guinea-pig ileum and rat heart. European Journal of Pharmacology, 113, 125–127. Fujii, R., Hosoya, M., Fukusumi, S., Kawamata, Y., Habata, Y., Hinuma, S. et al. (2000). Identification of neuromedin U as the cognate ligand of the orphan G protein-coupled receptor FM-3. Journal of Biological Chemistry, 275, 21068–21074. Funk, L., Trendelenburg, A.U., Limberger, N. and Starke, K. (1995). Subclassification of presynaptic α2-adrenoceptors: α2D-autoreceptors and α2D-adrenoceptors modulating release of acetylcholine in guinea-pig ileum. Naunyn-Schmiedebergs Archives of Pharmacology, 352, 58–66. Furness, J.B. and Costa, M. (1974). The adrenergic innervation of the gastrointestinal tract. Ergebnisse der Physiologie, 69, 1–51. Furness, J.B. and Costa, M. (1979a). Projections of intestinal neurons showing immunoreactivity for vasoactive intestinal polypeptide are consistent with these neurons being the enteric inhibitory neurons. Neuroscience Letters, 15, 199–204. Furness, J.B. and Costa, M. (1979b). Actions of somatostatin on excitatory and inhibitory nerves in the intestine. European Journal of Pharmacology, 56, 69–74. Furness, J.B. and Costa, M. (1982). Identification of gastrointestinal neurotransmitters. In Handbook of experimental pharmacology, volume 59/I. Mediators and drugs in gastrointestinal motility I. Morphological basis and neurophysiological control, edited by G. Bertaccini, pp. 383–460. Berlin: Springer-Verlag. Furness, J.B. and Costa, M. (1987). The enteric nervous system. Edinburgh: Churchill Livingstone. Furness, J.B., Bornstein, J.C., Pompolo, S., Young, H.M., Kunze, W.A.A. and Kelly, H. (1994). The circuitry of the enteric nervous system. Neurogastroenterology and Motility, 6, 241–253. Furness, J.B., Clark, M.J., Wright, T., Bertrand, P.P., Bornstein, J.C. and Verlinden, M. (1999). An action of erythromycin in the intestine that is not mediated via motilin receptors. Clinical and Experimental Pharmacology and Physiology, 26, 100–104. Furness, J.B., Costa, M., Murphy, R., Beardsley, A.M., Oliver, J.R., Llewellyn-Smith, I.J. et al. (1982). Detection and characterization of neurotransmitters, particularly peptides, in the gastrointestinal tract. Scandinavian Journal of Gastroenterology, 17, 61–70. Furness, J.B., Costa, M., Rökaeus, Å., McDonald, T.J. and Brooks, B. (1987). Galanin-immunoreactive neurons in the guinea-pig small intestine: their projections and relationships to other enteric neurons. Cell and Tissue Research, 250, 607–615. Furness, J.B., Eskay, R.L., Brownstein, M.J. and Costa, M. (1980). Characterization of somatostatin-like immunoreactivity in intestinal nerves by high pressure liquid chromatography and radioimmunoassay. Neuropeptides, 1, 97–103. Furness, J.B., Kunze, W.A.A., Bertrand, P.P., Clerc, N. and Bornstein, J.C. (1998). Intrinsic primary afferent neurons of the intestine. Progress in Neurobiology, 54, 1–18. Furness, J.B., Pompolo, S., Murphy, R. and Giraud, A. (1989a). Projections of neurons with neuromedin U-like immunoreactivity in the small intestine of the guinea-pig. Cell and Tissue Research, 257, 415–422. Furness, J.B., Trussell, D.C., Pompolo, S., Bornstein, J.C., Maley, B.E. and Storm-Mathisen, J. (1989b). Shapes and projections of neurons with immunoreactivity for γ-aminobutyric acid in the guinea-pig small intestine. Cell and Tissue Research, 256, 293–301. Gabella, G. (1979). Innervation of the gastrointestinal tract. International Review of Cytology, 59, 129–193. Galligan, J.J. (1992). Differential inhibition of cholinergic and noncholinergic neurogenic contractions by 5-hydroxytryptamine1A receptor agonists in guinea pig ileum. Journal of Pharmacology and Experimental Therapeutics, 260, 306–312.

274

INNERVATION OF THE GASTROINTESTINAL TRACT

Galligan, J.J. (1995). Electrophysiological studies of 5-hydroxytryptamine receptors on enteric neurons. In Serotonin and Gastrointestinal Function, edited by T.S. Gaginella and J.J. Galligan, pp. 109–126. Boca Raton, FL: CRC Press. Galligan, J.J. (1999). Nerve terminal nicotinic cholinergic receptors on excitatory motoneurons in the myenteric plexus of guinea pig intestine. Journal of Pharmacology and Experimental Therapeutics, 291, 92–98. Galligan, J.J. and Bertrand, P.P. (1994). ATP mediates fast synaptic potentials in enteric neurons. Journal of Neuroscience, 14, 7563–7571. Galligan, J.J. and North, R.A. (1991). Opioid, 5-HT1A and α2 receptors localized to subsets of guinea-pig myenteric neurons. Journal of Autonomic Nervous System, 32, 1–12. Galligan, J.J., Mosberg, H.I., Hurst, R., Hruby, V.J. and Burks, T.F. (1984). Cerebral δ opioid receptors mediate analgesia but not the intestinal motility effects of intracerebroventricularly administered opioids. Journal of Pharmacology and Experimental Therapeutics, 229, 641–648. Galligan, J.J., Surprenant, A., Tonini, M. and North, R.A. (1988). Differential localization of 5-HT1 receptors on myenteric and submucosal neurons. American Journal of Physiology, 255, G603–G611. Garzón, J., Höllt, V., Schulz, R. and Herz, A. (1985). Excitatory neuropeptides activate opioid mechanisms in the guinea pig ileum. Neuropeptides, 5, 583–586. Garzón, J., Höllt, V. and Sánchez-Blázquez, P. (1986). Neuropeptide Y is an inhibitor of neural function in the myenteric plexus of the guinea-pig ileum. Peptides, 7, 623–629. Garzón, J., Höllt, V., Sánchez-Blázquez, P. and Herz, A. (1987). Neural activation of opioid mechanisms in guinea pig ileum by excitatory peptides. Journal of Pharmacology and Experimental Therapeutics, 240, 642–649. Gershon, M.D. (1997). Genes and lineages in the formation of the enteric nervous system. Current Opinion in Neurobiology, 7, 101–109. Giaid, A., Gibson, S.J., Ibrahim, N.B.N., Legon, S., Bloom, S.R., Yanagisawa, M. et al. (1989). Endothelin-1, an endothelium-derived peptide, is expressed in neurons of the human spinal cord and dorsal root ganglia. Proceedings of the National Academy of Sciences of the USA, 86, 7634–7638. Giaroni, C., Zanetti, E., Marino, F., Cosentino, M., Senaldi, A., Somaini, L. et al. (2000). Glutamate receptors of the AMPA type modulate neurotransmitter release and peristalsis in the guinea-pig isolated colon. Life Sciences, 67, 1747–1757. Gillespie, J.S. and Khoyi, M.A. (1977). The site and receptors responsible for the inhibition by sympathetic nerves of intestinal smooth muscle and its parasympathetic motor nerves. Journal of Physiology (London), 267, 767–789. Gintzler, A.R. and Scalisi, J.A. (1982). Effects of opioids on noncholinergic excitatory responses of the guinea-pig isolated ileum: inhibition of release of enteric substance P. British Journal of Pharmacology, 75, 199–206. Giotti, A., Luzzi, S., Spagnesi, S. and Zilletti, L. (1983). GABAA and GABAB receptor-mediated effects in guinea-pig ileum. British Journal of Pharmacology, 78, 469–478. Giraldo, E., Viganò, M.A., Hammer, R. and Ladinsky, H. (1988). Characterization of muscarinic receptors in guinea pig ileum longitudinal smooth muscle. Journal of Pharmacology and Experimental Therapeutics, 33, 617–625. Giralt, M. and Vergara, P. (2000). Inhibition of ascending contraction elicited by mucosal stimulation in the duodenum of the rat. Neurogastroenterology and Motility, 12, 173–180. Glass, J., Chan, W.C. and Gintzler, A.R. (1986). Direct analysis of the release of methionine-enkephalin from guinea pig myenteric plexus: modulation by endogenous opioids and exogenous morphine. Journal of Pharmacology and Experimental Therapeutics, 239, 742–747. Goedert, M., Hunter, J.C. and Ninkovic, M. (1984). Evidence for neurotensin as a non-adrenergic, noncholinergic neurotransmitter in guinea-pig ileum. Nature, 311, 59–62. Goldstein, D.J., Ropchak, T.G., Keiser, H.R., Atta, G.J., Argiolas, A. and Pisano, J.J. (1983). Bradykinin reverses the effect of opiates in the gut by enhancing acetylcholine release. Journal of Biological Chemistry, 258, 12122–12124. Gottschall, P.E., Tatsuno, I., Miyata, A. and Arimura, A. (1990). Characterization and distribution of binding sites for the hypothalamic peptide, pituitary adenylate cyclase activating polypeptide. Endocrinology, 127, 272–277. Goy, M.F. (1991). cGMP: the wayward child of the cyclic nucleotide family. Trends in the Neurosciences, 14, 293–299. Grafe, P., Galvan, M. and Mayer, C.J. (1979). Actions of GABA on guinea-pig myenteric neurons. Pflügers Archives, 382, R44. Grassby, P.F. and Broadley, K.J. (1984). Characterization of beta-adrenoceptors mediating relaxation in the guinea-pig ileum. Journal of Pharmacy and Pharmacology, 36, 602–607.

PHARMACOLOGY OF THE ENTERIC NERVOUS SYSTEM

275

Grassby, P.F. and Broadley, K.J. (1987). Partial agonists at guinea-pig atrial β-adrenoceptors display relaxations responses in the guinea-pig ileum independent of β-adrenoceptor stimulation. General Pharmacology, 18, 25–31. Greenberg, R., Kosterlitz, H.W. and Waterfield, A.A. (1970). The effects of hexamethonium, morphine and adrenaline on the output of acetylcholine from the myenteric plexus-longitudinal muscle preparation of the ileum. British Journal of Pharmacology, 40, 553P–554P. Greenwood, B., Davison, J.S. and Dodds, W.J. (1990). Influence of selective α- and β-adrenoceptor antagonists on the control of motor activity and transmural potential differences in the rabbit ileum in vitro. Journal of Gastrointestinal Motility, 2, 47–52. Grider, J.R. (1989). Somatostatin release from isolated ganglia of the myenteric plexus. American Journal of Physiology, 257, G313–G315. Grider, J.R. (1994a). Interplay of somatostatin, opioid, and GABA neurons in the regulation of peristaltic reflex. American Journal of Physiology, 267, G696–G701. Grider, J.R. (1994b). Role of cholecystokinin in the regulation of gastrointestinal motility. Journal of Nutrition, 124, 1334S–1339S. Grider, J.R. (1994c). CGRP as a transmitter in the sensory pathway mediating peristaltic reflex. American Journal of Physiology, 266, G1139–G1145. Grider, J.R. and Makhlouf, G.M. (1986). Colonic peristaltic reflex: identification of vasoactive intestinal peptide as mediator of descending relaxation. American Journal of Physiology, 251, G40–G45. Grider, J.R. and Makhlouf, G.M. (1987a). Role of opioid neurons in the regulation of intestinal peristalsis. American Journal of Physiology, 253, G226–G231. Grider, J.R. and Makhlouf, G.M. (1987b). Regional and cellular heterogeneity of cholecystokinin receptors mediating muscle contraction in the gut. Gastroenterology, 92, 175–180. Grider, J.R. and Makhlouf, G.M. (1988a). The modulatory function of galanin: potentiation of VIP-induced relaxation in isolated smooth muscle cells. Gastroenterology, 94, A157. Grider, J.R. and Makhlouf, G.M. (1988b). Vasoactive intestinal peptide. Transmitter of inhibitory motor neurons of the gut. Annals of the New York Academy of Sciences, 527, 369–377. Grider, J.R. and Makhlouf, G.M. (1992). Nitric oxide (NO) is generated from muscle cells by the action of relaxant neurotransmitter, VIP. Gastroenterology, 102, A735. Grider, J.R. and Rivier, J.R. (1990). Vasoactive intestinal peptide (VIP) as a transmitter of inhibitory motor neurons of the gut: evidence from the use of selective VIP antagonists and VIP antiserum. Journal of Pharmacology and Experimental Therapeutics, 253, 738–742. Grider, J.R., Arimura, A. and Makhlouf, G.M. (1987). Role of somatostatin neurons in intestinal peristalsis: facilitatory interneurons in descending pathways. American Journal of Physiology, 253, G434– G438. Grider, J.R., Murthy, K.S., Jin, J.G. and Makhlouf, G.M. (1992). Stimulation of nitric oxide from muscle cells by VIP: prejunctional enhancement of VIP release. American Journal of Physiology, 262, G774–G778. Grider, J.R., Katsoulis, S., Schmidt, W.E. and Jin, J.G. (1994). Regulation of descending relaxation phase of intestinal peristalsis by PACAP. Journal of Autonomic Nervous System, 50, 151–159. Grider, J.R., Kuemmerle, J.F. and Jin, J.G. (1996). 5-HT released by mucosal stimuli initiate peristalsis by activating 5-HT4 /5-HT1P receptors on sensory CGRP neurons. American Journal of Physiology, 270, 778–782. Grider, J.R., Foxx-Orenstein, A.E. and Jin, J.G. (1998). 5-Hydroxytryptamine-4 recepor agonists initiate the peristaltic reflex in human, rat, and guinea pig intestine. Gastroenterology, 115, 370–380. Grivegnée, A.R., Fontaine, J. and Reuse, J. (1984). Effect of dopamine on dog distal colon in-vitro. Journal of Pharmacy and Pharmacology, 36, 454–457. Growcott, J.W., Holloway, B., Green, M. and Wilson, C. (1993a). Zeneca ZD7114 acts as an antagonist at β3-adrenoceptors in rat isolated ileum. British Journal of Pharmacology, 110, 1375–1380. Growcott, J.W., Wilson, C., Holloway, B. and Mainwaring, S. (1993b). Evaluation of ICI D7114, a putative stimulant of brown adipocytes, on histamine-contracted guinea-pig ileum. British Journal of Pharmacology, 109, 1212–1218. Guard, S. and Watson, S.P. (1987). Evidence for neurokinin-3 receptor-mediated tachykinin release in the guinea-pig ileum. European Journal of Pharmacology, 144, 409–412. Guillemin, R. (1976). Somatostatin inhibits the release of acetylcholine induced electrically in the myenteric plexus. Endocrinology, 99, 1653–1654. Haasemann, M., Buschko, J., Faussner, A., Roscher, A.A., Hoebeke, J., Burch, R.M. et al. (1991). Anti-idiotypic antibodies bearing the internal image of a bradykinin epitope. Production, characterization, and interaction with the kinin receptor. Journal of Immunology, 147, 3882–3892. Hall, J.M. and Morton, I.K. (1991). Bradykinin B2 receptor evoked K+ permeability increase mediates relaxation in the rat duodenum. European Journal of Pharmacology, 193, 231–238.

276

INNERVATION OF THE GASTROINTESTINAL TRACT

Hanani, M., Chorev, M., Gilon, C. and Selinger, Z. (1988). The actions of receptor-selective substance P analogs on myenteric neurons: an electrophysiological investigation. European Journal of Pharmacology, 153, 247–253. Harry, J. (1963). The action of drugs on the circular muscle strip from the guinea-pig isolated ileum. British Journal of Pharmacology, 20, 399–417. Hashimoto, S., Tanaka, C. and Taniyama, K. (1986). Presynaptic muscarinic and alpha-adrenoceptor-mediated regulation of GABA release from myenteric neurones of the guinea-pig small intestine. British Journal of Pharmacology, 89, 787–792. Hasler, W.L., Heldsinger, A. and Owyang, C. (1992). Erythromycin contracts rabbit colon myocytes via occupation of motilin receptors. American Journal of Physiology, 262, G50–G55. Hasler, W.L., Kurosawa, S., Takahashi, T., Feng, H., Gaginella, T.S. and Owyang, C. (1995). Bradykinin acting on B2 receptors contracts colon circular muscle cells by IP3 generation and adenylate cyclase inhibition. Journal of Pharmacology and Experimental Therapeutics, 273, 344–350. He, X.D. and Goyal, R.K. (1992). VIP associated inhibitory junction potential in the guinea-pig ileum involves nitric oxide. Gastroenterology, 102, A457. Hedner, P. (1970). Effect of the C-terminal octapeptide of cholecystokinin on guinea pig ileum and gall-bladder in vitro. Acta Physiologica Scandinavica, 78, 232–235. Hedrick, J.A., Morse, K., Shan, L., Qiao, X., Pang, L., Wang, S. et al. (2000). Identification of a human gastrointestinal tract and immune system receptor for the peptide neuromedin U. Molecular Pharmacology, 58, 870–875. Helmstaedter, V., Taugner, C.H., Feurle, G.E. and Fossman, W.G. (1977). Localization of neurotensinimmunoreactive cells in the small intestine of man and various mammals. Histochemistry, 53, 35–41. Hens, J., Schrodl, F., Brehmer, A., Adriaensen, D., Neuhuber, W., Scheuermann, D.W. et al. (2000). Mucosal projections of enteric neurons in the porcine small intestine. Journal of Comparative Neurology, 421, 429–436. Hew, R.W., Hodgkinson, C.R. and Hill, S.J. (1990). Characterization of histamine H3-receptors in guinea-pig ileum with H3-selective ligands. British Journal of Pharmacology, 101, 621–624. Hill, S.J., Ganellin, C.R., Timmerman, H., Schwartz, J.C., Shankley, N.P., Young, J.M. et al. (1997). International Union of Pharmacology, XIII. Classification of histamine receptors. Pharmacological Reviews, 49, 253–278. Hills, J.M., Jessen, K.R. and Mirsky, R. (1987). An immunohistochemical study of the distribution of enteric GABA-containing neurons in the rat and guinea-pig intestine. Neuroscience, 22, 301–312. Hills, J.M., Dingsdale, R.A., Parsons, M.E., Dolle, R.E. and Howson, W. (1989). 3-Aminopropylphosphinic acid – a potent, selective GABAB receptor agonist in the guinea-pig ileum and rat anococcygeus muscle. British Journal of Pharmacology, 97, 1292–1296. Hirst, G.D.S. and McKirdy, H.C. (1974a). Presynaptic inhibition at mammalian peripheral synapse? Nature, 250, 430–431. Hirst, G.D.S. and McKirdy, H.C. (1974b). A nervous mechanism for descending inhibition in guinea-pig small intestine. Journal of Physiology, 238, 129–143. Hobbiger, F. (1958a). Antagonism by γ-aminobutyric acid to the actions of 5-hydroxytryptamine and nicotine on isolated organs. Journal of Physiology (London), 144, 349–360. Hobbiger, F.J. (1958b). Effects of γ-aminobutyric acid on the isolated mammalian ileum. Journal of Physiology (London), 142, 147–164. Holzer, P. (1982). An enquiry into the mechanism by which substance P facilitates the phasic longitudinal contractions of the rabbit ileum. Journal of Physiology (London), 325, 377–392. Holzer, P. (1984). Characterization of the stimulus-induced release of immunoreactive substance P from the myenteric plexus of the guinea-pig small intestine. Brain Research, 297, 127–136. Holzer, P. (1989). Ascending enteric reflex: multiple neurotransmitter systems and interactions. American Journal of Physiology, 256, G540–G545. Holzer, P. (1997). Involvement of nitric oxide in the substance P-induced inhibition of intestinal peristalsis. Neuroreport, 8, 2857–2860. Holzer, P. and Holzer-Petsche, U. (1997a). Tachykinins in the gut. Part I. Expression, release and motor function. Pharmacology and Therapeutics, 73, 173–217. Holzer, P. and Holzer-Petsche, U. (1997b). Tachykinins in the gut. Part II. Roles in neural excitation, secretion and inflammation. Pharmacology and Therapeutics, 73, 219–263. Holzer, P. and Lembeck, F. (1979). Effect of neuropeptides on the efficiency of the peristaltic reflex. NaunynSchmiedebergs Archives of Pharmacology, 307, 257–264. Holzer, P. and Lembeck, F. (1980). Neurally mediated contraction of ileal longitudinal muscle by substance P. Neuroscience Letters, 17, 101–105.

PHARMACOLOGY OF THE ENTERIC NERVOUS SYSTEM

277

Holzer, P. and Maggi, C.A. (1994). Synergistic role of muscarinic acetylcholine and tachykinin NK2 receptors in intestinal peristalsis. Naunyn-Schmiedebergs Archives of Pharmacology, 349, 194–201. Holzer, P. and Skofitsch, G. (1984). Release of endogenous 5-hydroxytryptamine from the myenteric plexus of the guinea-pig isolated small intestine. British Journal of Pharmacology, 81, 381–386. Holzer, P., Lembeck, F. and Donnerer, J. (1980). Caerulein, substance P, serotonin and cholinomimetics induce rhythmic contractions of the intestinal circular muscle. Naunyn-Schmiedebergs Archives of Pharmacology, 312, 131–137. Holzer, P., Bucsics, A., Saria, A. and Lembeck, F. (1982). A study of the concentrations of substance P and neurotensin in the gastrointestinal tract of various mammals. Neuroscience, 7, 2919–2924. Holzer, P., Lippe, I.Th., Barthó, L. and Saria, A. (1987). Neuropeptide Y inhibits excitatory enteric neurons supplying the circular muscle of the guinea pig small intestine. Gastroenterology, 92, 1944–1950. Holzer, P., Barthó, L., Matusák, O. and Bauer, V. (1989). Calcitonin gene-related peptide action on intestinal circular muscle. American Journal of Physiology, 256, G546–G552. Holzer, P., Schluet, W. and Maggi, C.A. (1995). Substance P stimulates and inhibits intestinal peristalsis via distinct receptors. Journal of Pharmacology and Experimental Therapeutics, 274, 322–328. Holzer, P., Lippe, I.Th., Heinemann, A. and Barthó, L. (1998). Tachykinin NK1 and NK2 receptor-mediated control of peristaltic propulsion in the guinea-pig small intestine in vitro. Neuropharmacology, 37, 131–138. Holzer-Petsche, U. and Moser, R.L. (1996). Participation of nitric oxide in the relaxation of the rat gastric corpus. Naunyn-Schmiedebergs Archives of Pharmacology, 354, 348–354. Hope, B.T. and Vincent, S.R. (1989). Histochemical characterization of neuronal NADPH-diaphorase. Journal of Histochemistry and Cytochemistry, 37, 653–661. Hope, B.T., Michael, G.J., Knigge, K.M. and Vincent, S.R. (1991). Neuronal NADPH diaphorase is a nitric oxide synthase. Proceedings of the National Academy of Sciences of the USA, 88, 2811–2814. Horácek, J. and Kadlec, O. (1984). An interaction of endogenous and exogenous opiates in the guinea-pig isolated ileum. Archives Internationales de Pharmacodynamie et de Therapie, 267, 13–22. Hoyle, C.H., Kamm, M.A., Burnstock, G. and Lennard-Jones, J.E. (1990). Enkephalins modulate inhibitory neuromuscular transmission in circular muscle of human colon via delta-opioid receptors. Journal of Physiology, 431, 465–78. Hryorenko, L.M., Woskowska, Z. and Fox-Threlkeld, J.E.T. (1994). Nitric oxide (NO) inhibits release of acetylcholine from nerves of isolated circular muscle of canine ileum: relationship to motility and release of nitric oxide. Journal of Pharmacology and Experimental Therapeutics, 271, 918–926. Hughes, J., Kosterlitz, H.W. and Sosa, R.P. (1978). Enkephalin release from the myenteric plexus of the guineapig small intestine in the presence of cyclohexamide. British Journal of Pharmacology, 63, 397P. Huidobro-Toro, J.P. and Way, E.L. (1982). Possible modulatory role of dynorphin on the excitation by neurotensin on the guinea pig myenteric plexus. Neuroscience Letters, 30, 309–314. Huidobro-Toro, J.P., Zhu, Y.-X., Lee, N.M., Loh, H.H. and Way, E.L. (1984). Dynorphin inhibition of the neurotensin contractile activity on the myenteric plexus. Journal of Pharmacology and Experimental Therapeutics, 228, 293–303. Huizinga, J.D., Tomlinson, J. and Pintin-Quezada, J. (1992). Involvement of nitric oxide in nerve-mediated inhibition and action of vasoactive intestinal peptide in colonic smooth muscle. Journal of Pharmacology and Experimental Therapeutics, 260, 803–808. Humphreys, C.M.S., Costa, M. and Brookes, S.J.H. (1991). Nitric oxide mediates the apamin-insensitive component of transmission from enteric inhibitory motor neurons to the circular muscle of the guinea-pig small intestine and colon. Proceedings of the Australian Physiology and Pharmacology Society, 22, 144P. Hutchison, J.B. and Dockray, G.J. (1981). Evidence that the action of cholecystokinin octapeptide on the guinea pig ileum longitudinal muscle is mediated in part by substance P release from the myenteric plexus. European Journal of Pharmacology, 69, 87–93. Hutchison, J.B., Dimaline, R. and Dockray, G.J. (1981). Neuropeptides in the gut: quantification and characterization of cholecystokinin octapeptide-, bombesin- and vasoactive intestinal polypeptide-like immunoreactivities in the myenteric plexus of the guinea-pig small intestine. Peptides, 2, 23–30. Inatomi, N., Sato, F., Marui, S., Itoh, Z. and Omura, S. (1996). Vagus-dependent and vagus-independent mechanism of action of the erythromycin derivative EM574 and motilin in dogs. Japanese Journal of Pharmacology, 71, 29–38. Inouye, A., Fukuya, M., Tsuchiya, K. and Tsujiyoka, T. (1960). Studies on the effects of aminobutyric acid on the isolated guinea-pig ileum. Japanese Journal of Physiology, 10, 167–182. Irons, W.A., Stark, M.E. and Szurszewski, J.H. (1992). Effect of repetitive application of nitric oxide on circular muscle of canine jejunum. Gastroenterology, 102, A461. Ito, Y. and Tajima, K. (1980). Action of morphine on the neuroeffector transmission in the guinea-pig ileum and in the mouse vas deferens. Journal of Physiology (London), 307, 367–383.

278

INNERVATION OF THE GASTROINTESTINAL TRACT

Ito, S., Kurokawa, A., Ohga, A., Ohta, T. and Sawabe, K. (1990). Mechanical, electrical and cyclic nucleotide responses to peptide VIP and inhibitory nerve stimulation in rat stomach. Journal of Physiology, 430, 337–353. Itoh, Z. (1997). Motilin and clinical application. Peptides, 18, 593–608. Izzo, A.A., Costa, M., Mascolo, N. and Capasso, F. (1998). The role of histamine H1, H2 and H3 receptors on enteric ascending synaptic transmission in the guinea pig ileum. Journal of Pharmacology and Experimental Therapeutics, 287, 952–957. Jenkinson, K.M., Morgan, J.M., Furness, J.B. and Southwell, B.R. (1999). Neurons bearing NK3 tachykinin receptors in the guinea-pig ileum revealed by specific binding of fluorescently labelled agonists. Histochemistry and Cell Biology, 112, 233–246. Jessen, K.R., Saffrey, M.J., Van Noorden, S., Bloom, S.R., Polak, J.M. and Burnstock, G. (1980). Immunohistochemical studies of the enteric nervous system in tissue culture and in situ: localization of vasoactive intestinal polypeptide (VIP), substance-P and enkephalin immunoreactive nerves in the guinea-pig gut. Neuroscience, 5, 1717–1735. Jessen, K.R., Hills, J.M. and Saffrey, M.J. (1986). Immunohistochemical demonstration of GABAergic neurons in the enteric nervous system. Journal of Neuroscience, 6, 1628–1634. Jhamandas, K. and Elliot, J. (1980). Comparative effects of somatostatin and enkephalins on the guinea-pig ileum and rat vas deferens. Canadian Journal of Physiology and Pharmacology, 58, 1389–1392. Jin, J.G. and Grider, J.R. (1992). Nitric oxide is produced in ganglia isolated from the myenteric plexus of guineapig intestine. Gastroenterology, 102, A463. Jin, J.G., Katsoulis, S., Schmidt, W.E. and Grider, J.R. (1994). Inhibitory transmission in tenia coli mediated by distinct vasoactive intestinal peptide and apamin-sensitive pituitary adenylate cyclase activating peptide receptors. Journal of Pharmacology and Experimental Therapeutics, 270, 433–439. Jin, J.G., Kuemmerle, J.F., Grider, J.R. and Makhlouf, G.M. (1992). Relaxation induced by VIP in isolated gastric smooth muscle cells is mediated by activation of both cAMP- and cGMP-dependent kinase. Gastroenterology, 102, A463. Johnson, S.M., Katayama, Y., Morita, K. and North, R.A. (1981). Mediators of slow synaptic potentials in the myenteric plexus of the guinea-pig ileum. Journal of Physiology (London), 320, 175–186. Johnson, S.M., Costa, M., Humphreys, C.M.S. and Shearman, R. (1987). Inhibitory effects of opioids in a circular muscle-myenteric plexus preparation of guinea-pig ileum. Naunyn-Schmiedebergs Archives of Pharmacology, 336, 419–424. Johnson, S.M., Costa, M. and Humphreys, C.M.S. (1988). Opioid µ and κ receptors on axons of cholinergic excitatory motor neurons supplying the circular muscle of guinea-pig ileum. Naunyn-Schmiedebergs Archives of Pharmacology, 338, 397–400. Johnson, P.J., Bornstein, J.C., Yuan, S.Y. and Furness, J.B. (1996). Analysis of contributions of acetylcholine and tachykinins to neuro-neuronal transmission in motility reflexes in the guinea-pig ileum. British Journal of Pharmacology, 118, 973–983. Johnson, P.J., Bornstein, J.C. and Burcher, E. (1998). Roles of neuronal NK1 and NK3 receptors in synaptic transmission during motility reflexes in the guinea-pig ileum. British Journal of Pharmacology, 124, 1375–1384. Kadlec, O. and Horácek, J. (1980). Inhibition of peristaltic activity in the guinea-pig ileum by specific stress stimulus: its reversal by naloxone and indomethacin. Life Sciences, 27, 1557–1562. Kadowaki, M., Wade, P.R. and Gershon, M.D. (1996). Participation of 5-HT3, 5-HT4, and nicotinic receptors in the peristaltic reflex of the guinea pig distal colon. American Journal of Physiology, 271, G849–G857. Karhula, T., Panula, P., Steinbusch, H. and Happola, O. (1990). Immunohistochemical localization of 5-hydroxytryptamine, histamine and histidine decarboxylase in the rat major pelvic and coeliac-superior mesenteric ganglion. Journal of the Autonomic Nervous System, 31, 91–99. Katayama, Y. and Morita, K. (1989). Adenosine 5′-triphosphate modulates membrane potassium conductance in guinea-pig myenteric neurones. Journal of Physiology, 408, 373–390. Katayama, Y. and North, R.A. (1978). Does substance P mediate slow synaptic excitation within the myenteric plexus? Nature, 274, 387–388. Katayama, Y. and North, R.A. (1980). The action of somatostatin on neurones of the myenteric plexus of the guinea-pig ileum. Journal of Physiology, 303, 315–323. Katayama, Y., North, R.A. and Williams, J.T. (1979). The action of substance P on neurons of the myenteric plexus of the guinea-pig small intestine. Proceedings of the Royal Society London (Biological Sciences), B206, 191–208. Katayama, Y., Lees, G.M. and Pearson, G.T. (1986). Electrophysiology and morphology of vasoactiveintestinal-peptide-immunoreactive neurones of the guinea-pig ileum. Journal of Physiology (London), 378, 1–11.

PHARMACOLOGY OF THE ENTERIC NERVOUS SYSTEM

279

Katsoulis, S., Schmidt, W.E., Clemens, A., Schworer, H. and Creutzfeldt, W. (1992). Vasoactive intestinal polypeptide induces neurogenic contraction of guinea-pig ileum. Involvement of acetylcholine and subtance P. Regulatory Peptides, 38, 155–164. Kenakin, T. (1994). On the definition of efficacy. Antagonists may possess negative efficacy. Trends in Pharmacological Sciences, 15, 408–409. Kerr, D.I.B. and Krantis, A. (1983). Uptake and stimulus-evoked release of [3H]γ-aminobutyric acid by myenteric nerves of guinea-pig intestine. British Journal of Pharmacology, 78, 271–276. Khairallah, P.A. and Page, I.H. (1961). Mechanism of action of angiotensin and bradykinin on smooth muscle in situ. American Journal of Physiology, 200, 51–54. Khairallah, P.A. and Page, I.H. (1963). Effects of bradykinin and angiotensin on smooth muscle. Annals of the New York Academy of Sciences, 104, 212–221. Kilbinger, H. and Nafziger, M. (1985). Two types of neuronal receptors modulating acetylcholine release from guinea-pig myenteric plexus. Naunyn-Schmiedebergs Archives of Pharmacology, 328, 304–309. Kilbinger, H. and Pfeuffer-Freiderich, I. (1985). Two types of receptors for 5-hydroxytryptamine on the cholinergic nerves of the guinea-pig myenteric plexus. British Journal of Pharmacology, 81, 529–539. Kilbinger, H. and Wolf, D. (1992). Effects of 5-HT4 receptor stimulation on basal and electrically evoked release of acetylcholine from guinea-pig myenteric plexus. Naunyn-Schmiedebergs Archives of Pharmacology, 345, 270–275. Kilbinger, H., Halim, S., Lambrecht, G., Weiler, W. and Wessler, I. (1984). Comparison of affinities of muscarinic antagonists to pre- and postjunctional receptors in the guinea-pig ileum. European Journal of Pharmacology, 103, 313–320. Kilbinger, H., Stauss, P., Erlhof, I. and Holzer, P. (1986). Antagonist discrimination between subtypes of tachykinin receptors in the guinea-pig ileum. Naunyn-Schmiedebergs Archives of Pharmacology, 334, 181–187. Kirchgessner, A.L., Tamir, H. and Gershon, M.D. (1992). Identification and stimulation by serotonin of intrinsic sensory neurons of the submucosal plexus of the guinea pig gut: activity-induced expression of Fos immunoreactivity. Journal of Neuroscience, 12, 235–248. Kirchgessner, A.L., Liu, M.T. and Alcantara, F. (1997). Excitotoxicity in the enteric nervous system. Journal of Neuroscience, 17, 8804–8816. Kishi, M., Takeuchi, T., Suthamnatpong, N., Ishii, T., Nishio, H., Hata, F. et al. (1996). VIP- and PACAPmediated nonadrenergic, noncholinergic inhibition in longitudinal muscle of rat distal colon: involvement of activation of charybdotoxin- and apamin-sensitive K+ channels. British Journal of Pharmacology, 119, 623–630. Kishi, M., Takeuchi, T., Katayama, H., Yamazaki, Y., Nishio, H., Hata, F. et al. (2000). Involvement of cyclic AMP-PKA pathway in VIP-induced, charibdotoxin-sensitive relaxation of longitudinal muscle of the distal colon of Wistar-ST rats. British Journal of Pharmacology, 129, 140–146. Kitagbi, P. and Freychet, P. (1978). Effects of neurotensin on isolated intestinal smooth muscles. European Journal of Pharmacology, 50, 349–357. Kitagbi, P. and Freychet, P. (1979a). Neurotensin: contractile activity, specific binding, and lack of effect on cyclic nucleotides in intestinal smooth muscle. European Journal of Pharmacology, 55, 35–42. Kitagbi, P. and Freychet, P. (1979b). Neurotensin contracts the guinea-pig longitudinal ileal smooth muscle by inducing acetylcholine release. European Journal of Pharmacology, 56, 403–406. Kitazawa, T., Kondo, H. and Temma, K. (1986). α2-Adrenoceptor-mediated contractile response to catecholamines in smooth muscle strips isolated from rainbow trout stomach (Salmo gairdneri). British Journal of Pharmacology, 89, 259–266. Kleinrok, A. and Kilbinger, A. (1983). γ-Aminobutyric acid and cholinergic transmission in the guinea-pig ileum. Naunyn-Schmiedebergs Archives of Pharmacology, 322, 216–220. Knowles, R.G., Palacios, M., Palmer, R.M.J. and Moncada, S. (1989). Formation of nitric oxide from L-arginine in the central nervous system: a transduction mechanism for stimulation of the soluble guanylate cyclase. Proceedings of the National Academy of Sciences of the USA, 86, 5159–5162. Knudsen, M.A. and Tøttrup, A. (1992). A possible role of L-arginine-nitric oxide pathway in the modulation of cholinergic transmission in the guinea-pig taenia coli. British Journal of Pharmacology, 107, 837–841. Kobilka, B.K., Dixon, R.A.F., Frielle, T., Dolhman, H.G., Bolanowski, M.A., Sigal, I.S. et al. (1987). cDNA for the human β2-adrenergic receptor: a protein with multiple membrane-spanning domains and encoded by a gene whose chromosomal location is shared with that of the receptor for platelet-derived growth factor. Proceedings of the National Academy of Sciences of the USA, 84, 46–50. Koike, K., Takayanagi, I., Muramutsu, M., Ohki, S. and Horinouchi, T. (1994). Involvement of β3-adrenoceptor in the relaxation response in guinea pig taenia caecum. Japanese Journal of Pharmacology, 66, 213–220. Kojima, S. (1997). Calcitonin gene related peptide inhibits nonadrenergic noncholinergic contraction of guineapig distal colon. Japanese Journal of Pharmacology, 73, 255–257.

280

INNERVATION OF THE GASTROINTESTINAL TRACT

Kojima, S. and Shimo, Y. (1996). Investigation into the 5-hydroxytryptamine-induced atropine-resistant neurogenic contraction of guinea-pig proximal colon. British Journal of Pharmacology, 117, 1613–1618. Kojima, Y., Takahashi, T., Fujina, M. and Owyang, C. (1994). Inhibition of cholinergic transmission by opiates in ileal myenteric plexus is mediated by κ receptor. Involvement of regulatory inhibitory G protein and calcium N-channels. Journal of Pharmacology and Experimental Thererapeutics, 268, 965–970. Kong, I.D., Koh, S.D. and Sanders, K.M. (2000). Purinergic activation of spontaneous transient outward currents in guinea pig taenia colonic myocytes. American Journal of Physiology, 278, C352–C362. Kortezova, N., Mizhorkova, Z., Milusheva, E., Coy, D.H., Vizi, E.S. and Varga, G. (1994). GRP-preferring bombesin receptor subtype mediates contractile activity in cat terminal ileum. Peptides, 15, 1331– 1333. Kosterlitz, H.W. and Robinson, J.A. (1957). Inhibition of the peristaltic reflex of the isolated guinea-pig ileum. Journal of Physiology, 136, 249–262. Kotska, P., Sipos, S.N., Kwan, C.Y., Niles, L.P. and Daniel, E.E. (1989). Identification and characterization of presynaptic and postsynaptic beta adrenoceptors in the longitudinal smooth muscle/myenteric plexus of dog ileum. Journal of Pharmacology and Experimental Therapeutics, 251, 305–310. Köves, K. and Arimura, A. (1990). Immunohistochemical demonstration of pituitary adenylate cyclase activating polypeptide in ovine gut. Clinical Research, 38, 969A. Krantis, A. and Kerr, D.I.B. (1981). Autoradiographic localization of [3H]-γ-aminobutyric acid in the myenteric plexus of the guinea-pig small intestine. Neuroscience Letters, 23, 263–268. Krantis, A., Costa, M., Furness, J.B. and Orbach, J. (1980). γ-Aminobutyric acid stimulates intrinsic inhibitory and excitatory nerves in the guinea-pig intestine. European Journal of Pharmacology, 67, 461–468. Krantis, A., Kerr, D.I.B. and Dennis, B.J. (1986). Autoradiographic study of the distribution of [3H]γ-aminobutyrate-accumulating neural elements in guinea-pig intestine: evidence for a transmitter function of γ-aminobutyrate. Neuroscience, 17, 1243–1255. Krantis, A., Shabnavard, L., Nichols, K., deBlas, A.L. and Staines, W. (1995). Localization of GABAA receptor immunoreactivity in NO synthase positive myenteric neurons. Journal of Autonomic Nervous System, 53, 157–165. Kromer, W. (1990). Reflex peristalsis in the guinea pig isolated ileum is endogenously controlled by kappa opioid receptors. Naunyn-Schmiedebergs Archives of Pharmacology, 341, 450–454. Kromer, W. and Pretzlaff, W. (1979). In vitro evidence for the participation of intestinal opioids in the control of peristalsis in the guinea pig small intestine. Naunyn-Schmiedebergs Archives of Pharmacology, 309, 153–157. Kromer, W. and Woinoff, R. (1980). Peristalsis in the isolated guinea-pig ileum during opiate withdrawal. Naunyn-Schmiedebergs Archives of Pharmacology, 314, 191–193. Kromer, W. and Woinoff, R. (1981). Dual action of somatostatin upon peristalsis in the guinea pig isolated ileum. Neuroendocrinology, 33, 136–139. Kromer, W., Pretzlaff, W. and Scheiblhuber, E. (1980). In vitro evidence for an involvement of intestinal endorphins in the control of peristalsis in the guinea pig ileum. Comparison to rabbit, rat, cat and dog small intestine. In Endogenous and Exogenous Opiate Agonists and Antagonists, edited by E.L. Way, pp. 337–340. New York: Pergamon Press. Kromer, W., Pretzlaff, W. and Woinoff, R. (1980). Opioids modulate periodicity rather than efficacy of peristaltic waves in the guinea pig ileum in vitro. Life Sciences, 26, 1857–1865. Kromer, W., Scheiblhuber, E. and Illes, P. (1980). Functional antagonism by calcium of an intrinsic opioid mechanism in the guinea-pig ileum. Neuropharmacology, 19, 839–843. Kromer, W., Pretzlaff, W. and Woinoff, R. (1981). Regional distribution of an opioid mechanism in the guineapig isolated intestine. Journal of Pharmacy and Pharmacology, 33, 98–101. Kromer, W., Steigemann, N. and Shearman, G.T. (1982). Differential effects of SKF 10 047 (N-allyl-normetazocine) on peristalsis and longitudinal muscle contractions of the isolated guinea-pig ileum. NaunynSchmiedebergs Archives of Pharmacology, 321, 218–222. Kuchii, M., Miyahara, J.T. and Shibata, S. (1973). [3H]-adenosine nucleotide and [3H]-noradrenaline uptake by cold stored guinea-pig taenia caecum; mechanical effects and release of [3H]-adenosine nucleotide by noradrenaline, papaverine and nitroglycerine. British Journal of Pharmacology, 49, 642–650. Kusunoki, M., Tsai, L.H., Taniyama, K. and Tanaka, C. (1986). Vasoactive intestinal polypeptide provokes acetylcholine release from the myenteric plexus. American Journal of Physiology, 251, G51–G55. Kuwahara, A., Ozaki, T. and Tanaihara, N. (1989). Galanin suppresses neurally evoked contractions of the circular muscle in the guinea-pig ileum. European Journal of Pharmacology, 164, 175–178. Kuwahara, A., Ozaki, T. and Yanaihara, N. (1990). Structural requirements for galanin action in the guinea-pig ileum. Regulatory Peptides, 29, 23–29. Kwon, S.C., Ozaki, H., Hori, M. and Karaki, H. (1993). Isoproterenol changes the relationship between cytosolic Ca2+ and contraction in guinea-pig taenia caecum. Japanese Journal of Pharmacology, 61, 57–64.

PHARMACOLOGY OF THE ENTERIC NERVOUS SYSTEM

281

Lambrecht, G., Friebe, T., Grimm, U., Windscheif, U., Bungardt, E., Hildebrandt, C. et al. (1992). PPADS, a novel functionally selective antagonist of P2 purinoceptor-mediated responses. European Journal of Pharmacology, 217, 217–219. Landi, M., Bianchetti, A., Croci, T. and Manara, L. (1992). Phenylethanolaminotetralines compete with [3H]dihydroalprenolol binding to rat colon membranes without evidencing atypical β-adrenergic sites. Biochemical Pharmacology, 44, 665–672. Landi, M., Croci, T. and Manara, L. (1993). Similar atypical β-adrenergic receptors mediate in vitro rat adipocyte lipolysis and colonic motility inhibition. Life Sciences, 53, PL297–PL302. Langer, S.Z. (1977). Presynaptic receptors and their role in the regulation of transmitter release. British Journal of Pharmacology, 60, 481–497. Langer, S.Z. (1981). Presynaptic regulation of the release of catecholamines. Pharmacological Reviews, 32, 337–362. Larzabal, A., Losada, J., Mateos, J.M., Benitez, R., Garmilla, I.J., Kuhn, R. et al. (1999). Distribution of the group II metabotropic glutamate receptors (mGluR2/3) in the enteric nervous system of the rat. Neuroscience Letters, 276, 91–94. Leander, S., Ekman, R., Uddman, R., Sundler, F. and Håkanson, R. (1984). Neuronal cholecystokinin, gastrinreleasing peptide, neurotensin, and β-endorphins in the intestine of the guinea-pig: distribution and possible motor functions. Cell and Tissue Research, 235, 521–531. Lecci, A., De Giorgio, R., Barthó, L., Sternini, C., Tramontana, M., Corinaldesi, R., Giuliani, S. and Maggi, C.A. (1999). Tachykinin NK1 receptor-mediated inhibitory responses in the guinea-pig small intestine. Neuropeptides, 33, 91–97. Lee, C.M., Iversen, L.L., Hanley, M.R. and Sandberg, B.E.B. (1982). The possible existence of multiple receptors for substance P. Naunyn-Schmiedebergs Archives of Pharmacology, 318, 281–287. Lefebvre, R.A., Smits, G.J.M. and Timmermans, J.P. (1995). Study of NO and VIP as non-adrenergic non-cholinergic neurotrasnmitters in the pig gastric fundus. British Journal of Pharmacology, 116, 2017–2026. Legat, F.J., Althuber, P., Maier, R., Griesbacher, T. and Lembeck, F. (1996). Evidence for the presence of NK1 and NK3 receptors on cholinergic neurones in the guinea-pig ileum. Neuroscience Letters, 207, 125–128. LePard, K.J., Messori, E. and Galligan, J.J. (1997). Purinergic fast excitatory postsynaptic potentials in myenteric neurons of guinea-pig: distribution and pharmacology. Gastroenterology, 113, 1522–1534. Lincoln, T.M. (1989). Cyclic GMP and mechanisms of vasodilation. Pharmacology and Therapeutics, 41, 479–502. Lippi, A., Santicioli, P., Criscuoli, M. and Maggi, C.A. (1998). Depolarization evoked co-release of tachykinins from enteric nerves in the guinea-pig proximal colon. Naunyn-Schmiedebergs Archives of Pharmacology, 357, 245–251. Liu, L. and Coupar, I.M. (1997). Characterization of pre- and post-synaptic α-adrenoceptors in modulation of the rat ileum longitudinal and circular muscle activities. Naunyn-Schmiedebergs Archives of Pharmacology, 356, 248–256. Liu, M.T. and Kirchgessner, A.L. (2000). Agonist- and reflex-evoked internalization of metabotropic glutamate receptor 5 in enteric neurons. Journal of Neuroscience, 20, 3200–3205. Liu, M.T., Rothstein, J.D., Gershon, M.D. and Kirchgessner, A.L. (1997). Glutamatergic enteric neurons. Journal of Neurosciences, 17, 4764–4784. Liu, S., Xia, Y., Hu, H., Ren, J., Gao, C. and Wood, J.D. (2000). Histamine H3 receptor-mediated suppression of inhibitory synaptic transmission in the submucous plexus of guinea-pig small intestine. European Journal of Pharmacology, 397, 49–54. Llewellyn-Smith, I.J., Furness, J.B. and Costa, M. (1985). Light and electron microscopic immunocytochemistry of the same nerves from whole mount preparations. Journal of Histochemistry and Cytochemistry, 33, 857–866. Llewellyn-Smith, I.J., Furness, J.B., Gibbins, I.L. and Costa, M. (1988). Quantitative ultrastructural analysis of enkephalin-, substance P- and VIP- immunoreactive nerve fibres in the circular muscle of the guinea pig small intestine. Journal of Comparative Neurology, 272, 139–148. Loeffler, S., Holzer, P., Maggi, C.A. and Kilbinger, H. (1994). Inhibition by NK1 receptors of the electrically evoked release of acetylcholine from guinea pig myenteric neurones. Naunyn-Schmiedebergs Archives of Pharmacology, 349, 352. Lomax, A.E., Bertrand, P.P. and Furness, J.B. (1998). Identification of the populations of enteric neurons that have NK1 tachykinin receptor immunoreactivity in the guinea-pig small intestine. Cell and Tissue Research, 294, 27–33. Luzzi, S., Zilletti, L., Franchi-Micheli, S., Gori, A.M. and Moroni, F. (1988). Agonists, antagonists and modulators of excitatory amino acid receptors in the guinea-pig myenteric plexus. British Journal of Pharmacology, 95, 1271–1277.

282

INNERVATION OF THE GASTROINTESTINAL TRACT

Lyster, D.J.K., Bywater, R.A.R. and Taylor, G.S. (1992). Nitric oxide mediates complex changes in membrane potential in the circular muscle of isolated guinea-pig ileum. Proceedings of the Australian Physiology and Pharmacology Society, 23, 6P. MacDonald, A. and Lamont, M. (1993). Effects of selective antagonism of β-adrenoceptor sub-types on responses to isoprenaline in rat distal colon in vitro. British Journal of Pharmacology, 110, 1551–1555. MacDonald, A., Kelly, J. and Dettmar, P.W. (1990). Pre- and post-junctional α-adrenoceptor-mediated responses in the rat gastric fundus in vitro. Journal of Pharmacy and Pharmacology, 42, 752–757. MacDonald, A., Forbes, I.J., Gallacher, D., Heeps, G. and McLaughlin, D.P. (1994). Adrenoceptors mediating relaxation to catecholamines in rat isolated jejunum. British Journal of Pharmacology, 112, 576–578. Maggi, C.A. (1995). The mammalian tachykinin receptors. General Pharmacology, 26, 911–944. Maggi, C.A. and Giuliani, S. (1993). Multiple inhibitory mechanisms mediate non-adrenergic non-cholinergic relaxation in the circular muscle of the guinea-pig colon. Naunyn-Schmiedebergs Archives of Pharmacology, 347, 630–634. Maggi, C.A. and Giuliani, S. (1996). Characterization of the apamin- and L-nitroarginine-resistant NANC inhibitory transmission to the circular muscle of guinea-pig colon. Journal of Autonomic Pharmacology, 16, 131–145. Maggi, C.A., Giuliani, S., Patacchini, R., Rovero, P., Giachetti, A. and Meli, A. (1989). The activity of peptides of the endothelin family in various mammalian smooth muscle preparations. European Journal of Pharmacology, 174, 23–31. Maggi, C.A., Patacchini, R., Giachetti, A. and Meli, A. (1990a). Tachykinin receptors in the circular muscle of the guinea-pig ileum. British Journal of Pharmacology, 101, 996–1000. Maggi, C.A., Patacchini, R., Giuliani, S., Turini, D., Barbanti, G., Rovero, P. et al. (1990b). Motor response of the human isolated small intestine and urinary bladder to porcine neuromedin U-8. British Journal of Pharmacology, 99, 186–188. Maggi, C.A., Patacchini, R., Meini, S. and Giuliani, S. (1993). Nitric oxide is the mediator of tachykinin NK3 receptor-induced relaxation in the circular muscle of the guinea-pig ileum. European Journal of Pharmacology, 240, 45–50. Maggi, C.A., Patacchini, R., Meini, S., Quartara, L., Sisto, A., Potier, E., Giuliani, S. and Giachetti, A. (1994a). Comparison of tachykinin NK1 and NK2 receptors in the circular muscle of the guinea-pig ileum and proximal colon. British Journal of Pharmacology, 112, 150–160. Maggi, C.A., Zagorodnyuk, V. and Giuliani, S. (1994b). Specialization of tachykinin NK1 and NK2 receptors in producing fast and slow atropine-resistant neurotransmission to the circular muscle of the guinea-pig colon. Neuroscience, 63, 1137–1152. Maggi, C.A., Zagorodnyuk, V. and Giuliani, S. (1994c). Tachykinin NK3 receptor mediates NANC hyperpolarization and relaxation via nitric oxide release in the circular muscle of the guinea-pig colon. Regulatory Peptides, 53, 259–274. Maggi, C.A., Patacchini, R., Barthó, L., Holzer, P. and Santicioli, P. (1994d). Tachykinin NK1 and NK2 receptor antagonists and atropine-resistant ascending excitatory reflex to the circular muscle of the guinea-pig ileum. British Journal of Pharmacology, 112, 161–168. Manara, L., Badone, D., Baroni, M., Boccardi, G., Cecchi, R., Croci, T. et al. (1995). Aryloxopropanolaminotetralins are the first selective antagonists for atypical (β3) β-adrenoceptors. Pharmacology Communications, 6, 240–245. Manber, L. and Gershon, M.D. (1979). A reciprocal adrenergic-cholinergic axoaxonic synapse in the mammalian gut. American Journal of Physiology, 236, E738–E745. Mandrek, K. and Kreis, S. (1992). Regional differentiation of gastric and of pyloric smooth muscle in the pig: mechanical responses to acetylcholine, histamine, substance P, noradrenaline and adrenaline. Journal of Autonomic Pharmacology, 12, 37–49. Mann, P.T., Southwell, B.R., Ding, Y.Q., Shigemoto, R., Mizuno, N. and Furness, J.B. (1997a). Localization of neurokinin 3 (NK3) receptor immunoreactivity in the rat gastrointestinal tract. Cell and Tissue Research, 289, 1–9. Mann, P.T., Southwell, B.R., Young, H.M. and Furness, J.B. (1997b). Appositions made by axons of descending interneurons in the guinea-pig small intestine, investigated by confocal microscopy. Journal of Chemical Neuroanatomy, 12, 151–164. Manning, D.C., Snyder, S.H., Kachur, J.F., Miller, R.J. and Field, M. (1982). Bradykinin receptor-mediated chloride secretion in intestinal function. Nature, 299, 256–261. Mao, Y.K., Wang, Y.F., Moogk, C., Fox-Threlkeld, J.E., Xiao, Q., McDonald, T.J., et al. (1998). Locations and molecular forms of PACAP and sites and characteristics of PACAP receptors in canine ileum. American Journal of Physiology, 274, G217–G225. Marcoli, M., Lecchini, S., De Ponti, F., D’Angelo, L., Crema, A. and Frigo, G.M. (1985). Subsensitivity of enteric cholinergic neurones to α2-adrenoceptor agonists after chronic sympathetic denervation. NaunynSchmiedebergs Archives of Pharmacology, 329, 271–277.

PHARMACOLOGY OF THE ENTERIC NERVOUS SYSTEM

283

Marino, F., Creta, F., De Ponti, F., Giaroni, C., Lecchini, S. and Frigo, G.M. (1993). Opioid pathways exert a tonic restraint in the guinea-pig isolated colon: changes after chronic sympathetic denervation. Journal of Pharmacy and Pharmacology, 45, 668–670. Marino, F., Cosentino, M., De Ponti, F., Giaroni, C., Somaini, L., Bombelli, R. et al. (1997). Muscarinic modulation of endogenous noradrenaline release from adrenergic terminals in the guinea-pig colon. Journal of Autonomic Pharmacology, 17, 365–372. Mashimo, H., He, X.D., Huang, P.L., Fishman, M.C. and Goyal, R.K. (1996). Neuronal constitutive nitric oxide synthase is involved in murine enteric inhibitory neurotransmission. Journal of Clinical Investigations, 98, 8–13. Matsuo, K., Katsuragi, T., Fujiki, S., Sato, C. and Furukawa, T. (1997). ATP release and contraction mediated by different P2-receptor subtypes in guinea-pig ileal smooth muscle. British Journal of Pharmacology, 121, 1744–1748. Matthijs, G., Peeters, T.L. and Vantrappen, G. (1989). The role of intracellular calcium stores in motilin induced contractions of the longitudinal muscle of rabbit duodenum. Naunyn-Schmiedebergs Archives of Pharmacology, 339, 332–339. Mawe, G.M. (1993). Noradrenaline as a presynaptic inhibitory neurotransmitter in ganglia of the guinea-pig gallbladder. Journal of Physiology (London), 461, 387–402. Mawe, G.M., Branchek, T. and Gershon, M.D. (1986). Peripheral neural serotonin receptors: identification and characterization with specific agonists and antagonists. Proceedings of the National Academy of Sciences of the USA, 83, 9799–9803. Mawe, G.M., Branchek, T. and Gershon, M.D. (1989). Blockade of 5-HT-mediated enteric slow EPSPs by BRL 24924: gastrokinetic effects. American Journal of Physiology, 257, G386–G396. Mayer, B., John, M. and Böhme, E. (1990). Purification of a Ca2+/calmodulin-dependent nitric oxide synthase from porcine cerebellum. Cofactor-role of tetrahydrobiopterin. FEBS Letters, 277, 215–219. McConalogue, K. and Furness, J.B. (1994). Gastrointestinal neurotransmitters. Baillière’s Clinical Endocrinology and Metabolism, 8, 51–76. McConalogue, K., Todorov, L., Furness, J.B. and Westfall, D.P. (1996). Direct measurement of the release of ATP and its major metabolites from the nerve fibres of the guinea-pig taenia coli. Clinical and Experimental Pharmacology and Physiology, 23, 807–812. McDonald, T.J., Christofi, F.L., Brooks, B.D., Barnett, W. and Cook, M.A. (1988). Characterization of content and chromatographic forms of neuropeptides in purified nerve varicosities prepared from guinea pig myenteric plexus. Regulatory Peptides, 21, 69–83. McEachern, A.E., Shelton, E.R., Bhakta, S., Obernolte, R., Bach, C., Zuppan, P. et al. (1991). Expression cloning of a rat B2 bradykinin receptor. Proceedings of the National Academy of Sciences of the USA, 88, 7724–7728. McLaughlin, D.P. and MacDonald, A. (1990). Evidence for the existence of “atypical” β-adrenoceptors (β3-adrenoceptors) mediating relaxation in the rat distal colon in vitro. British Journal of Pharmacology, 101, 569–574. McLaughlin, D.P. and MacDonald, A. (1991). Characterization of catecholamine-mediated relaxation in rat isolated gastric fundus. Evidence for an atypical β-adrenoceptor. British Journal of Pharmacology, 103, 1351–1356. McLean, P.G. and Coupar, I.M. (1996). Stimulation of cyclic AMP formation in the circular smooth muscle of human colon by activation of 5-HT4-like receptors. British Journal of Pharmacology, 117, 238–239. Melander, T., Hökfelt, T., Rökaeus, Å., Fahrenkrug, J., Tatemoto, K. and Mutt, V. (1985). Distribution of galaninlike immunoreactivity in the gastro-intestinal tract of several mammalian species. Cell and Tissue Research, 239, 253–270. Menkveld, G.J. and Timmerman, H. (1990). Inhibition of electrically evoked contractions of guinea-pig ileum preparations mediated by the histamine H3 receptor. European Journal of Pharmacology, 186, 343–347. Messenger, J.P. (1993). Immunohistochemical analysis of neurons and their projections in the proximal colon of the guinea-pig. Archives of Histology and Cytology, 56, 459–474. Meulemans, A.L., Helsen, L.F. and Schuurkes, J.A. (1993). The role of nitric oxide (NO) in 5-HT-induced relaxations of the guinea-pig stomach. Naunyn-Schmiedebergs Archives of Pharmacology, 348, 424–430. Mian, M.A., Malta, E. and Raper, C. (1984). A homogeneous population of α1-adrenoceptors subserves inhibitory responses in guinea-pig ileal preparations. Journal of Pharmacy and Pharmacology, 36, 698–699. Miki, Y., Taniyama, K., Tanaka, C. and Tobe, T. (1983). GABA, glutamic acid decarboxylase, and GABA transaminase levels in the myenteric plexus of humans and other mammals. Journal of Neurochemistry, 40, 861–865. Milenov, K. and Atanassova, E. (1993). Effects of cholecystokinin octapeptide and somatostatin on the motility and release of [3H]acetylcholine in canine colon. Comparative Biochemistry and Physiology, 106, 337–342.

284

INNERVATION OF THE GASTROINTESTINAL TRACT

Milligan, G., Bond, R.A. and Lee, M. (1995). Inverse agonism: pharmacological curiosity or potential therapeutic strategy? Trends in Pharmacological Sciences, 16, 10–13. Milusheva, E.A., Kortezova, N.I., Mizhorkova, Z.N., Papasova, M., Coy, D.H., Balint, A. et al. (1998). Role of different bombesin receptor subtypes mediating contractile activity in cat upper gastrointestinal tract. Peptides, 19, 549–556. Minamino, N., Kangawa, K. and Matsuo, H. (1985). Neuromedin U-8 and U-25: novel uterus stimulating and hypertensive peptides identified in porcine spinal cord. Biochemical and Biophysical Research Communications, 130, 1078–1085. Minocha, A. and Galligan, J.J. (1991). Erythromycin inhibits contractions of nerve-muscle preparations of the guinea-pig small intestine. Journal of Pharmacology and Experimental Therapeutics, 257, 1248–1252. Minocha, A. and Galligan, J.J. (1993). Excitatory and inhibitory responses mediated by GABAA and GABAB receptors in guinea pig distal colon. European Journal of Pharmacology, 230, 187–193. Miyata, A., Arimura, A., Dahl, R.R., Minamino, N., Uehara, A., Jiang, L. et al. (1989). Isolation of a novel 38 residue-hypothalamic polypeptide which stimulates adenylate cyclase in pituitary cells. Biochemical and Biophysical Research Communications, 164, 567–574. Miyata, A., Jiang, L., Dahl, R.R., Kiada, C., Kubo, K., Fujino, M. et al. (1990). Isolation of a neuropeptide corresponding to the N-terminal 27 residues of the pituitary adenylate cyclase activating polypeptide with 38 residues (PACAP38). Biochemical and Biophysical Research Communications, 170, 643–648. Mizumoto, A., Sano, I., Matsunaga, Y., Yamamoto, O., Itoh, Z. and Ohshima, K. (1993). Mechanism of motilininduced contractions in isolated perfused canine stomach. Gastroenterology, 105, 425–432. Molenaar, P., Roberts, S.J., Kim, Y.S., Pak, H.S., Sainz, R.D. and Summers, R.J. (1991). Localization and characterization of two propranolol resistant (-)[125I]cyanopindolol binding sites in rat skeletal muscle. European Journal of Pharmacology, 209, 257–262. Mollereau, C., Parmentier, M., Mailleux, P., Butour, J.L., Moisand, C., Chalon, P. et al. (1994). ORL1, a novel member of the opioid receptor family. Cloning, functional expression and localization. FEBS Letters, 341, 33–38. Monier, S. and Kitagbi, P. (1980). Substance P-induced autodesensitization inhibits atropine resistant neurotensin-stimulated contractions in guinea pig ileum. European Journal of Pharmacology, 65, 461–462. Monier, S. and Kitagbi, P. (1981). Effects of β-endorphin, met-enkephalin and somatostatin on the neurotensininduced neurogenic contraction in the guinea-pig ileum. Regulatory Peptides, 2, 31–42. Morita, K. and North, R.A. (1981). Clonidine activates membrane potassium conductance in myenteric neurones. British Journal of Pharmacology, 74, 419–428. Morita, K., North, R.A. and Tokimasa, T. (1982). Muscarinic presynaptic inhibition in myenteric plexus of guinea-pig ileum. Journal of Physiology, 333, 141–149. Moroni, F., Luzzi, S., Franchi-Micheli, S. and Zilletti, L. (1986). The presence of N-methyl-D-aspartate-type receptors for glutamic acid in the guinea pig myenteric plexus. Neuroscience Letters, 68, 57–62. Mosberg, H.I., Hurst, R., Hruby, V.J., Gee, K., Yamamura, H.I., Galligan, J.J. et al. (1983). Bis-penicillamine enkephalins possess highly improved specificity toward delta opioid receptors. Proceedings of the National Academy of Sciences of the USA, 80, 5871–5874. Motomura, Y., Chijiiwa, Y., Iwakiri, Y. and Nawata, H. (1997). Direct contractile effect of cholecystokinin octapeptide on caecal circular smooth muscle cells of guinea pig via both CCK-A and CCK-B/gastrin receptors. Life Sciences, 60, 499–504. Mule, F. and Serio, R. (1997). Mode and mechanism of neurotensin action in rat proximal colon. European Journal of Pharmacology, 319, 269–272. Mulholland, M.W., Schoeneich, S. and Flowe, K. (1992). Galanin inhibition of enteric cholinergic transmission: guanosine triphosphate-binding protein interactions with adenylate cyclase. Surgery, 112, 195–201. Mungan, Z., Arimura, A., Ertan, A., Rossowski, W.J. and Coy, D.H. (1992). Pituitary adenylate cyclase-activating polypeptide relaxes rat gastrointestinal smooth muscle. Scandinavian Journal of Gastroenterology, 27, 375–380. Murphy, R., Turner, C.A., Furness, J.B., Parker, L. and Giraud, A. (1990). Isolation and microsequence analysis of a novel form of neuromedin U from guinea pig small intestine. Peptides, 11, 613–617. Nagahama, M., Tsuzuki, M., Mochizuchi, T., Iguchi, K. and Kuwahara, A. (1998). Light and electron microscopic studies of pituitary adenylate cyclase-activating peptide (PACAP)-immunoreactive neurons in the enteric nervous system of rat small and large intestine. Anatomy and Embriology, 198, 341–352. Nakajima, T. (1981). Active peptides in amphibian skin. Trends in Pharmacological Sciences, 2, 202–206. Nakajima, K., Tooyama, I., Kuriyama, K. and Kimura, H. (1996). Immunohistochemical demonstration of GABAB receptors in the rat gastrointestinal tract. Neurochemistry Research, 21, 211–215. Nakamoto, M., Tanaka, C. and Taniyama, K. (1987). Release of γ-aminobutyric acid and acetylcholine by neurotensin in guinea-pig ileum. British Journal of Pharmacology, 90, 545–551.

PHARMACOLOGY OF THE ENTERIC NERVOUS SYSTEM

285

Nemeth, P.R. and Gullikson, G.W. (1989). Gastrointestinal stimulating drugs and 5-HT receptors on myenteric neurons. European Journal of Pharmacology, 166, 387–391. Nemeth, P.R., Ort, C.A., Zafirov, D.H. and Wood, J.D. (1985). Interactions between serotonin and cisapride on myenteric neurons. European Journal of Pharmacology, 108, 77–83. Nemeth, P.R., Zafirov, D.H. and Wood, J.D. (1985). Effects of cholecystokinin, caerulein and pentagastrin on electrical behaviour of myenteric neurons. European Journal of Pharmacology, 116, 263–269. Nguyen-Le, X.K., Neugebauer, W., Gobeil, F., Pheng, L.H., Nsa Allogho, S. and Regoli, D. (1997). Pharmacological heterogeneity of neurotensin receptors: an in vitro study. Canadian Journal of Physiology and Pharmacology, 75, 547–551. Nicholson, J.R., Paterson, S.J., Menzies, J.R., Corbett, A.D. and McKnight, A.T. (1998). Pharmacological studies on the “orphan” opioid receptor in central and peripheral sites. Canadian Journal of Physiology and Pharmacology, 76, 304–313. Niel, J.P., Bywater, R.A.R. and Taylor, G.S. (1983a). Effect of substance P on non-cholinergic fast and slow post-stimulus depolarization in the guinea-pig ileum. Journal of Autonomic Nervous System, 9, 573–584. Niel, J.P., Bywater, R.A.R. and Taylor, G.S. (1983b). Apamin-resistant post-stimulus hyperpolarization in the circular muscle of the guinea-pig ileum. Journal of Autonomic Nervous System, 9, 565–569. Nishi, S. and North, R.A. (1973). Intracellular recording from the myenteric plexus of the guinea-pig ileum. Journal of Physiology (London), 231, 471–491. Norman, B.J. and Leathard, H.L. (1990). Evidence that an atypical β-adrenoceptor mediates the inhibition of spontaneous rhythmical contractions of rabbit isolated jejunum induced by ritodrine and salbutamol. British Journal of Pharmacology, 101, 27–30. North, R.A. and Surprenant, A. (1985). Inhibitory synaptic potentials resulting from α2-adrenoceptor activation in guinea-pig submucous plexus neurones. Journal of Physiology (London), 358, 17–33. North, R.A. and Tokimasa, T. (1982). Muscarinic synaptic potentials in guinea-pig myenteric plexus neurones. Journal of Physiology (London), 333, 151–156. North, R.A. and Tonini, M. (1977). The mechanism of action of narcotic analgesics in the guinea-pig ileum. British Journal of Pharmacology, 61, 541–549. North, R.A., Katayama, Y. and Williams, J.T. (1979). On the mechanism and site of action of enkephalin on single myenteric neurons. Brain Research, 165, 67–77. North, R.A., Henderson, G., Katayama, Y. and Johnson, S.M. (1980). Electrophysiological evidence for presynaptic inhibition of acetylcholine release by 5-hydroxytryptamine in the enteric nervous system. Neuroscience, 5, 581–586. North, R.A., Slack, B.E. and Surprenant, A. (1985). Muscarinic M1 and M2 receptors mediate depolarization and presynaptic inhibition in guinea-pig enteric nervous system. Journal of Physiology (London), 368, 435–452. O’Donnell, A.M., Ellis, L.M., Riedl, M.S., Elde, R.P. and Mawe, G.M. (2001). Distribution and chemical coding of orphanin FQ/nociceptin-immunoreactive neurons in the myenteric plexus of guinea pig intestines and sphincter of Oddi. Journal of Comparative Neurology, 430, 1–11. Ohashi, H., Tanaka, K., Kiuchi, N., Unno, T. and Komori, S. (1996). Modulation of peristalsis by neurotensin in isolated guinea-pig intestinal segments. European Journal of Pharmacology, 301, 129–136. Ohkawa, H. (1987). Effects of γ-aminobutyric acid on the non-adrenergic inhibitory potentials in circular smooth muscle cells of the guinea-pig duodenum. Biomedical Research, 8, 185–194. Ohki-Hamazaki, H. (2000). Neuromedin B. Progress in Neurobiology, 62, 297–312. Oka, T. and Sawa, A. (1979). Calcium requirements for electrically-induced release of an endogenous opiate receptor ligand from the guinea-pig ileum. British Journal of Pharmacology, 65, 3–5. Ong, J. and Kerr, D.I.B. (1983). GABAA- and GABAB-receptor-mediated modification of intestinal motility. European Journal of Pharmacology, 86, 9–17. Ong, J. and Kerr, D.I.B. (1984). Evidence for a physiological role of GABA in the control of guinea-pig intestinal motility. Neuroscience Letters, 50, 339–343. Ong, J. and Kerr, D.I.B. (1989). Modulation of spontaneous motility by GABAA receptor antagonism in the guinea-pig isolated ileum. Neuroscience Letters, 101, 203–208. Onori, L., Aggio, A., Taddei, G. and Tonini, M. (2000). Contribution of NK2 tachykinin receptors to propulsion in the rabbit distal colon. American Journal of Physiology, 278, G137–G147. Onori, L., Aggio, A., Taddei, G., Ciccocioppo, R., Severi, C., Carnicelli, V. et al. (2001). Contribution of NK3 tachykinin receptors to propulsion in the rabbit isolated distal colon. Neurogastroenterology and Motility, 13, 211–219. Osinski, M.A., Pampusch, M.S., Murtaugh, M.P. and Brown, D.R. (1999). Cloning, expression and functional role of a nociceptin/orphanin FQ receptor in the porcine gastrointestinal tract. European Journal of Pharmacology, 365, 281–289. Osthaus, L.E. and Galligan, J.J. (1992). Antagonists of nitric oxide synthesis inhibit nerve-mediated relaxations of longitudinal muscle in guinea pig ileum. Journal of Pharmacology and Experimental Therapeutics, 260, 140–145.

286

INNERVATION OF THE GASTROINTESTINAL TRACT

Ozawa, S., Kamiya, H. and Tsuzuki, K. (1998). Glutamate receptors in the mammalian central nervous system. Progress in Neurobiology, 54, 581–618. Paiva, T.B., Mendes, G.B., Aboulafia, J. and Paiva, A. (1976). Evidence against mediation of the effect of angiotensin II on the guinea pig ileum. Pflügers Archives, 365, 129–133. Palmer, J.M., Schemann, M., Tamura, K. and Wood, J.D. (1986a). Calcitonin gene-related peptide excites myenteric neurons. European Journal of Pharmacology, 132, 163–170. Palmer, J.M., Schemann, M., Tamura, K. and Wood, J.D. (1986b). Galanin mimics slow synaptic inhibition in myenteric neurons. European Journal of Pharmacology, 124, 379–380. Parkman, H.P., Pagano, A.P. and Ryan, J.P. (1996). Erythromycin inhibits rabbit pyloric smooth muscle through neuronal motilin receptors. Gastroenterology, 111, 682–690. Paton, W.D.M. and Vizi, E.S. (1969). The inhibitory action of noradrenaline and adrenaline on acetylcholine output by guinea-pig ileum longitudinal muscle strips. British Journal of Pharmacology, 35, 10–28. Paton, W.D.M. and Zar, A. (1968). The origin of acetylcholine released from guinea-pig intestine and longitudinal muscle strip. Journal of Physiology (London), 194, 13–33. Peeters, T.L. (1993). Erythromycin and other macrolides as prokinetic agents. Gastroenterology, 105, 1886–1899. Peeters, T.L., Matthijs, G. and Vantrappen, G. (1991). Ca2+ dependence of motilide-induced contractions in rabbit duodenal muscle strips in vitro. Naunyn-Schmiedebergs Archives of Pharmacology, 343, 202–208. Pfannkuche, H.J. (1992). Characterization of the bombesin receptor subtype in guinea-pig ileum longitudinal muscle preparations. Naunyn-Schmiedebergs Archives of Pharmacology, 345, R116. Pfannkuche, H., Reiche, D., Sann, H. and Schemann, M. (1998). Different subpopulations of cholinergic and nitrergic myenteric neurones project to mucosa and circular muscle of the guinea-pig gastric fundus. Cellular and Tissue Research, 292, 463–475. Pfannkuche, H., Firzlaff, U., Sann, H., Reiche, D. and Schemann, M. (2000). Neurochemical coding and projection patterns of gastrin-releasing peptide-immunoreactive myenteric neurone subpopulations in the guineapig gastric fundus. Journal of Chemical Neuroanatomy, 19, 93–104. Pheng, L.H., Quirion, R., Iyengar, S., Fournier, A. and Regoli, D. (1997). The rabbit ileum: a sensitive and selective preparation for the neuropeptide Y Y5 receptor. European Journal of Pharmacology, 333, R3–R5. Pheng, L.H., Perron, A., Quirion, R., Cadieux, A., Fauchere, J.L., Dumont, Y. et al. (1999). Neuropeptide Y-induced contraction is mediated by neuropeptide Y Y2 and Y4 receptors in the rat colon. European Journal of Pharmacology, 374, 85–91. Pillai, N.P. and Johnson, S.M. (1991). The electrophysiological effects of [D-Ala2, N-Me-Phe4, Gly5-ol]-enkephalin on guinea-pig myenteric neurons. European Journal of Pharmacology, 192, 227–233. Poli, E., Coruzzi, G. and Bertaccini, G. (1990). Ranitidine but not famotidine releases acetylcholine from the guinea pig myenteric plexus. Agents Actions, 30, 191–194. Pompolo, S. and Furness, J.B. (1990). Ultrastructure and synaptology of neurons immunoreactive for gammaaminobutyric acid in the myenteric plexus of the guinea pig small intestine. Journal of Neurocytology, 19, 539–549. Portbury, A.L., Furness, J.B., Southwell, B.R., Wong, H., Walsh, J.H. and Bunnett, N.W. (1996a). Distribution of neurokinin-2 receptors in the guinea-pig gastrointestinal tract. Cell and Tissue Research, 286, 281–289. Portbury, A.L., Furness, J.B., Young, H.M., Southwell, B.R. and Vigna, S.R. (1996b). Localization of NK1 receptor immunoreactivity to neurons and interstitial cells of the guinea-pig gastrointestinal tract. Journal of Comparative Neurology, 367, 342–351. Portbury, A.L., Pompolo, S., Furness, J.B., Stebbing, M.J., Kunze, W.A., Bornstein, J.C. et al. (1995). Cholinergic, somatostatin-immunoreactive interneurons in the guinea-pig intestine: morphology, ultrastructure, connections and projections. Journal of Anatomy, 187, 303–321. Poulter, M.O., Singhal, R., Brown, L.A. and Krantis, A. (1999). GABAA receptor subunit messenger RNA expression in the enteric nervous system of the rat: implications for functional diversity of enteric GABAA receptors. Neuroscience, 93, 1159–1165. Prins, N.H., Briejer, M.R., Van Bergen, P.J., Akkermans, L.M. and Schuurkes, J.A.J. (1999). Evidence for 5-HT7 receptors mediating relaxation of human colonic circular smooth muscle. British Journal of Pharmacology, 128, 849–852. Prins, N.H., Akkermans, L.M.A., Lefebvre, R.A., Cheyns, P. and Schuurkes, J.A.J. (2000a). Cholinergic 5-HT4 receptor stimulation enhances canine and human colon longitudinal muscle contractility. Neurogastroenterology and Motility, 12, 267. Prins, N.H., Shankley, N.P., Welsh, N., Briejer, M.R., Lefebvre, R.A., Akkermans, L.M. et al. (2000b). An improved in vitro bioassay for the study of 5-HT4 receptors in the human isolated large intestinal circular muscle. British Journal of Pharmacology, 129, 1601–1608. Puig, M.M., Gascon, P., Craviso, G.L. and Musacchio, J.M. (1997). Endogenous opiate receptor ligand: electrically induced release in the guinea pig ileum. Science, 195, 419–420.

PHARMACOLOGY OF THE ENTERIC NERVOUS SYSTEM

287

Raddatz, R., Wilson, A.E., Artymyshyn, R., Bonini, J.A., Borowsky, B., Boteju, L.W. et al. (2000). Identification and characterization of two neuromedin U receptors differentially expressed in peripheral tissues and the central nervous system. Journal of Biological Chemistry, 275, 32452–32459. Rae, M.G. and Muir, T.C. (1996). Neuronal mediators of inhibitory junction potentials and relaxation in the guinea-pig internal anal sphincter. Journal of Physiology (London), 493, 517–527. Rakovska, A.D. (1993). Functional and neurochemical evidence that neurotensin-induced release of acetylcholine from Auerbach’s plexus of guinea-pig ileum is presynaptically controlled via α2-adrenoceptors. Neurochemical Research, 18, 737–741. Ransom, R.W., Young, G.S., Schneck, K. and Goodman, C.B. (1992). Characterization of solubilized bradykinin B2 receptors from smooth muscle and mucosa of guinea-pig ileum. Biochemical Pharmacology, 43, 1823–1827. Rattan, S. and Chakder, S. (1992). Role of nitric oxide as a mediator of internal anal sphincter relaxation. American Journal of Physiology, 262, G107–G112. Reese, J.H. and Cooper, J.R. (1984). Noradrenergic inhibition of the nicotinically-stimulated release of acetylcholine from guinea-pig ileal synaptosomes. Biochemical Pharmacology, 33, 1145–1147. Regoli, A., Bondon, A. and Fouchère, J.L. (1994). Receptors and antagonists for substance P and related peptides. Pharmacological Review, 46, 551–599. Reiche, D. and Schemann, M. (1999). Mucosa of the guinea pig gastric corpus is innervated by myenteric neurones with specific neurochemical coding and projection preferences. Journal of Comparative Neurology, 410, 489–502. Reilly, W.M., Hoyle, C.H.V. and Burnstock, G. (1987). Pre- and postjunctional adrenoceptor types in the circular muscle of the guinea-pig caecum. Naunyn-Schmidebergs Archives of Pharmacology, 336, 37–47. Reinicke, M., Forssmann, W.G., Thiekotter, G. and Troepel, J. (1983). Localization of neurotensin-immunoreactivity in the spinal cord and peripheral nervous system of the guinea pig. Neuroscience Letters, 37, 37–42. Reis, H.J., Massensini, A.R., Prado, M.A.M., Gomez, R.S., Gomez, M.V. and Romano-Silva, M.A. (2000). Calcium channels coupled to depolarization-evoked glutamate release in the myenteric plexus of the guinea-pig ileum. Neuroscience, 101, 237–242. Rekik, M., Delvaux, M., Frexinos, J. and Bueno, L. (1997). The calcitonin gene related peptide activates both cAMP and NO pathways to induce relaxation of circular smooth muscle cells of guinea-pig ileum. Peptides, 18, 1517–1522. Ren, J., Hu, H.Z., Liu, S., Xia, Y. and Wood, J.D. (2000). Glutamate receptors in the enteric nervous system: ionotropic or metabotropic? Neurogastroenterology and Motility, 12, 257–264. Rhaleb, N.E. and Carretero, O.A. (1994). Role of B1 and B2 receptors and of nitric oxide in bradykinin-induced relaxation and contraction of isolated rat duodenum. Life Sciences, 55, 1351–1363. Richardson, B.P., Engel, G., Donatsch, P. and Stadler, P.A. (1985). Identification of serotonin M-receptor subtypes and their specific blockade by a new class of drugs. Nature, 316, 126–131. Rizzi, C.A., Coccini, T., Onori, L., Manzo, L. and Tonini, M. (1992). Benzimidazolone derivatives: a new class of 5-hydroxytryptamine-4 receptor agonists with prokinetic and acetylcholine releasing properties in the guinea pig ileum. Journal of Pharmacology and Experimental Therapeutics, 261, 412–419. Roberts, D.J., Hasler, W.L. and Owyang, C. (1993). GABA mediation of the dual effects of somatostatin on guinea-pig ileal myenteric cholinergic transmission. American Journal of Physiology, 264, G953–G960. Robertson, P.A. and Rubin, D. (1962). Stimulation of intestinal nervous elements by angiotensin. British Journal of Pharmacology, 19, 5–12. Ruffolo, Jr. R.R. and Hieble, J.P. (1994). α-Adrenoceptors. Pharmacology and Therapeutics, 61, 1–64. Rutherford, A. and Burnstock, G. (1978). Neuronal and non-neuronal components in the overflow of labelled adenyl compounds from guinea-pig taenia coli. European Journal of Pharmacology, 48, 195–202. Saffrey, M.J., Marcus, N., Jessen, K.R. and Burnstock, G. (1983). Distribution of neurons with high affinity uptake sites for GABA in the myenteric plexus of the guinea-pig, rat and chicken. Cell and Tissue Research, 234, 231–235. Saha, J.K., Sengupta, J.N. and Goyal, R.K. (1991). Effects of bradykinin and bradykinin analogs on the lower esophageal sphincter: characterization of an inhibitory bradykinin receptor. Journal of Pharmacology and Experimental Therapeutics, 259, 265–273. Sahyoun, H.A., Costall, B. and Naylor, R.J. (1982). Catecholamines act at α2-adrenoceptors to cause contraction of circular smooth muscle of guinea-pig stomach. Journal of Pharmacy and Pharmacology, 34, 381–385. Saida, K., Mitsui, Y. and Ishida, N. (1989). A novel peptide, vasoactive intestinal contractor, of a new (endothelin) peptide family. Journal of Biological Chemistry, 264, 14613–14616. Saito, N. and Tanaka, C. (1986). Immunohistochemical demonstration of GABA containing neurons in the guinea-pig ileum using purified GABA antiserum. Brain Research, 376, 78–84. Sakai, K.K., Hymson, D.L. and Shapiro, R. (1978). The mode of action of enkephalins in the guinea-pig myenteric plexus. Neuroscience Letters, 10, 317–322.

288

INNERVATION OF THE GASTROINTESTINAL TRACT

Sakurai-Yamashita, Y., Takada, K., Takemura, K., Yamashita, K., Enjoji, A., Kanematsu, T. et al. (1999). Ability of mosapride to bind to 5-HT4 receptor in the human stomach. Japanese Journal of Pharmacology, 79, 493–496. Sakurai-Yamashita, Y., Yamashita, K., Yoshimura, M. and Taniyama, K. (2000). Differential localization of 5-hydroxytryptamine-3 and 5-hydroxytryptamine-4 receptors in the human rectum. Life Sciences, 66, 31–34. Salimi, M. (1975). Comparison of β-adrenoceptor blocking properties of sotalol, oxprenolol, propranolol and pindolol on rabbit intestinal smooth muscle. Pharmacology, 13, 441–447. Samama, P., Pei, G., Costa, T., Cotecchia, S. and Lefkowitz, R.J. (1994). Negative antagonists promote an inactive conformation of the β2-adrenergic receptor. Journal of Pharmacology and Experimental Therapeutics, 45, 390–394. Sano, I., Taniyama, K. and Tanaka, C. (1989). Cholecystokinin, but not gastrin, induces gamma-aminobutyric acid release from myenteric neurons of the guinea pig ileum. Journal of Pharmacology and Experimental Therapeutics, 248, 378–383. Sawa, T., Mameya, S., Yoshimura, M., Itsuno, M., Makiyama, K., Niwa, M. et al. (1995). Differential mechanism of peptide YY and neuropeptide Y in inhibiting motility of guinea-pig colon. European Journal of Pharmacology, 276, 223–230. Sayeg, A.I. and Ritter, R.C. (2000). CCK-A receptor activation induces fos expression in myenteric neurons of rat small intestine. Regulatory Peptides, 88, 75–81. Schaumann, W. (1955). The paralysing action of morphine on the guinea-pig ileum. British Journal of Pharmacology, 10, 456–461. Schemann, M. (1991). Excitatory and inhibitory effects of norepinephrine on myenteric neurons of the guineapig gastric corpus. Pflügers Archives, 418, 575–580. Schemann, M. and Tamura, K. (1992). Presynaptic inhibitory effects of the peptides NPY, PYY and PP on nicotinic EPSPs in guinea-pig gastric myenteric neurones. Journal of Physiology (London), 451, 79–89. Schmidt, P., Rasmussen, T.N. and Holst, J.J. (1993). Release of immunoreactive somatostatin, vasoactive intestinal polypeptide (VIP), and galanin during propulsive complexes in isolated pig ileum. Peptides, 14, 215–220. Schneider, D.A. and Galligan, J.J. (2000). Presynaptic nicotinic acetylcholine receptors in the myenteric plexus of guinea pig intestine. American Journal of Physiology, 279, G528–G535. Schneider, D.A., Perrone, M. and Galligan, J.J. (2000). Nicotinic acetylcholine receptors at sites of neurotransmitter release to the guinea pig intestinal circular muscle. Journal of Pharmacology and Experimental Therapeutics, 294, 363–369. Schultzberg, M., Hokfelt, T., Nilsson, G., Terenius, L., Rehfeld, J.F., Brown, M. et al. (1980). Distribution of peptide- and catecholamine-containing neurons in the gastro-intestinal tract of rat and guinea-pig: immunohistochemical studies with antisera to substance P, vasoactive intestinal polypeptide, enkephalins, somatostatin, gastrin/cholecystokinin, neurotensin and dopamine β-hydroxylase. Neuroscience, 5, 689–744. Schulz, R., Wüster, M., Simantov, R., Snyder, S. and Herz, A. (1977). Electrically stimulated release of opiatelike material from the myenteric plexus of the guinea pig ileum. European Journal of Pharmacology, 41, 347–348. Schutte, I.W., Akkermans, L.M. and Kroese, A.B. (1997). CCKA and CCKB receptor subtypes both mediate the effects of CCK-8 on myenteric neurons in the guinea-pig ileum. Journal of the Autonomic Nervous System, 67, 51–59. Schutte, I.W., Hollestein, K.B., Akkermans, L.M. and Kroese, A.B. (1997). Evidence for a role of cholecystokinin as neurotransmitter in the guinea-pig enteric nervous system. Neuroscience Letters, 236, 155–158. Schütz, W. and Freissmuth, M. (1992). Reverse intrinsic activity of antagonists on G protein-coupled receptors. Trends in Pharmacological Sciences, 13, 376–380. Schuurkes, J.A.J., Van Nueten, J.M., Van Daele, A.J., Reyntjens, A.J. and Janssen, P.A.J. (1985). Motorstimulating properties of cisapride on isolated gastrointestinal preparations of the guinea pig. Journal of Pharmacology and Experimental Therapeutics, 234, 775–783. Schuurkes, J.A.J., Meulemans, A.L., Obertop, H. and Akkermans, L.M.A. (1991). 5-HT4 receptors on the human stomach. Journal of Gastrointestinal Motility, 3, 199. Schwörer, H. and Kilbinger, H. (1988). Enhancement of guinea-pig intestinal peristalsis by blockade of muscarinic M1-receptors. British Journal of Pharmacology, 93, 715–720. Schwörer, H., Racké, K. and Kilbinger, H. (1987). Spontaneous release of endogenous 5-hydroxytryptamine and 5-hydroxyindoleacetic acid from the isolated vascularly perfused ileum of the guinea-pig. Neuroscience, 21, 297–303. Schwörer, H., Racké, K. and Kilbinger, H. (1989). Temperature-dependent effects of increased intraluminal pressure on serotonin release from the vascularly perfused guinea pig ileum. Naunyn-Schmiedebergs Archives of Pharmacology, 336, 483–486. Schwörer, H., Clemens, A., Katsoulis, S., Köhler, H., Creutzfeldt, W. and Schmidt, W.E. (1992). Effects of pituitary adenylate cyclase activating peptide (PACAP) and vasoactive intestinal polypeptide (VIP) on human colonic motility. Naunyn-Schmiedebergs Archives of Pharmacology, 345, R116.

PHARMACOLOGY OF THE ENTERIC NERVOUS SYSTEM

289

Selemidis, S., Satchell, D.G. and Cocks, T.M. (1997). Evidence that NO acts as a redundant NANC inhibitory neurotransmitter in the guinea-pig isolated taenia coli. British Journal of Pharmacology, 121, 604–611. Serio, R., Mulé, F., Bonvissuto, F. and Pastorino, A. (1998). Tachykinins mediate noncholinergic excitatory neuronal responses in the circular muscle of rat proximal colon. Canadian Journal of Physiology and Pharmacology, 76, 684–689. Shahabazian, A. and Holzer, P. (2000). Regulation of guinea pig intestinal peristalsis by endogenous endothelin acting at ETB receptors. Gastroenterology, 119, 80–88. Shannon, H.E. and Sawyer, B.D. (1989). Glutamate receptors of the N-methyl-D-aspartate subtype in the myenteric plexus of the guinea pig ileum. Journal of Pharmacology and Experimental Therapeutics, 251, 518–523. Shiba, Y., Mizumoto, A., Nobuhiro, I., Haga, N., Yamamoto, O. and Itoh, Z. (1995). Stimulatory mechanism of EM523-induced contractions in postprandial stomach of conscious dogs. Gastroenterology, 109, 1513–1521. Shimuta, S.I., Borges, A.C., Prioste, R.N. and Paiva, T.B. (1999). Different pathways for Ca2+ mobilization by angiotensin II and carbachol in the circular muscle of the guinea-pig ileum. European Journal of Pharmacology, 367, 59–66. Shinmi, O., Kimura, S., Yoshizawa, T., Sawamura, T., Uchiyama, Y., Sugita, Y. et al. (1989). Presence of endothelin-1 in porcine spinal cord: isolation and sequence determination. Biochemical and Biophysical Research Communications, 162, 340–346. Shirakawa, J., Takeda, K., Taniyama, K. and Tanaka, C. (1989). Dual effects of 5-hydroxytryptamine on the release of γ-aminobutyric acid from myenteric neurones of the guinea-pig ileum. British Journal of Pharmacology, 98, 339–341. Shook, J.E., Pelton, J.T., Wire, W.S., Hirning, L.D., Hruby, V.J. and Burks, T.F. (1987). Pharmacologic evaluation of a cyclic somatostatin analog with antagonist activity at mu opioid receptors in vitro. Journal of Pharmacology and Experimental Therapeutics, 240, 772–777. Siani, M.A., Skillman, A.G., Carreras, C.W., Ashley, G., Kuntz, I.D. and Santi, D.V. (2000). Development and screening of a polyketide virtual library for drug leads against a motilide pharmacophore. Journal of Molecular Graphics and Modelling, 18, 497–511. Sim, M.K. and Lim, J.M.E. (1983). Adrenergic receptor-mediated response of the rabbit small and large intestine. Japanese Journal of Pharmacology, 33, 409–413. Sinsky, M. and Donnerer, J. (1998). Evidence for a neurotransmitter role of glutamate in guinea pig myenteric neurons. Neuroscience Letters, 258, 109–112. Smith, C.F., Taylor, K.J. and Whiting, E.M. (1994). Discovery and characterization of an inhibitory angiotensin receptor in guinea-pig ileum. Archives Internationales de Pharmacodynamie et de Therapie, 327, 87–95. Smith, T.K. and Furness, J.B. (1988). Reflex changes in circular muscle activity elicited by stroking the mucosa: an electrophysiological analysis in the isolated guinea-pig ileum. Journal of Autonomic Nervous System, 25, 205–218. Smith, T.K. and Robertson, W.J. (1998). Synchronous movements of the longitudinal and circular muscle during peristalsis in the isolated guinea-pig distal colon. Journal of Physiology, 506, 563–577. Smith, T.K., Bornstein, J.C. and Furness, J.B. (1990). Distension-evoked ascending and descending reflexes in the circular muscle of the guinea-pig ileum: an intracellular study. Journal of Autonomic Nervous System, 29, 203–217. Smith, T.K., Ward, S.M., Zhang, L., Buxton, I.L.O., Gerthoffer, W.T., Sanders, K.M. et al. (1993). β-Adrenergic inhibition of electrical and mechanical activity in canine colon: role of cAMP. American Journal of Physiology, 264, G708–G717. Smits, G.J.M. and Lefebvre, R.A. (1996). ATP and nitric oxide: inhibitory NANC neurotransmitters in the longitudinal muscle-myenteric plexus preparation of the rat ileum. British Journal of Pharmacology, 118, 695–703. Song, Z.M., Brookes, S.J., Ramsay, G.A. and Costa, M. (1997). Characterization of myenteric interneurons with somatostatin immunoreactivity in the guinea-pig small intestine. Neuroscience, 80, 907–923. Spindel, E. (1986). Mammalian bombesin-like peptides. Trends in the Neurosciences, 9, 130–133. Stark, M.E., Bauer, A.J. and Szurszewski, J.H. (1991). Effect of nitric oxide on circular muscle of canine small intestine. Journal of Physiology (London), 444, 743–761. Starke, K., Göthert, M. and Kilbinger, H. (1989). Modulation of neurotransmitter release by presynaptic autoreceptors. Physiological Reviews, 69, 864–989. Starodub, A.M. and Wood, J.D. (2000). Histamine H2 receptor activated chloride conductance in myenteric neurons from guinea pig small intestine. Journal of Neurophysiology, 83, 1809–1816. Steele, P.A., Costa, M., Murphy, R. and Waterman, S.A. (1992). Endogenous opioids in the enteric nervous system. Proceedings of the Australian Physiology and Pharmacology Society, 23, 95–110. Sternini, C., Su, D., Gamp, P.D. and Bunnett, N.W. (1995). Cellular sites of expression of the neurokinin-1 receptor in the rat gastrointestinal tract. Journal of Comparative Neurology, 358, 531–540.

290

INNERVATION OF THE GASTROINTESTINAL TRACT

Sternini, C., Wong, H., Pham, T., De Giorgio, R., Miller, L.J., Kuntz, S.M. et al. (1999a). Expression of cholecystokinin A receptors in neurons innervating the rat stomach and intestine. Gastroenterology, 117, 1136–1146. Sternini, C., Pham, T., Minnis, J., Wong, H., Spelta, V., Fiori, E. et al. (1999b). The inhibitory action of galanin on myenteric cholinergic neurons is partly mediated by galanin-1 receptors. Gastroenterology, 116, A648. Sternini, C., Brecha, N.C., Minnis, J., D’Agostino, G., Balestra, B., Fiori, E. et al. (2000). Role of agonistdependent receptor internalization in the regulation of µ opioid receptors. Neuroscience, 98, 233–241. Sublette, E. and Gintzler, A.R. (1992). Stimulus frequency and intensity: critical determinants of opioid enhancement or inhibition of evoked methionine-enkephalin release. Brain Research, 599, 165–170. Sun, Y.D. and Benishin, C.G. (1991). Effects of calcitonin gene-related peptide on longitudinal muscle and myenteric plexus of guinea pig ileum. Journal of Pharmacology and Experimental Therapeutics, 259, 947–952. Sundler, F., Håkanson, R., Hammer, R., Alumets, J., Carraway, R., Leeman, S. and Zimmerman, E. (1977). Immunohistochemical localization of neurotensin in endocrine cells of the gut. Cell and Tissue Research, 178, 313–321. Sundler, F., Ekblad, E., Absood, A., Håkanson, R., Köves, K. and Arimura, A. (1992). Pituitary adenylate cyclase activating peptide: a novel vasoactive intestinal peptide-like neuropeptide in the gut. Neuroscience, 46, 439–454. Surprenant, A. (1994). Control of the gastrointestinal tract by enteric neurons. Annual Reviews of Physiology, 56, 117–140. Surprenant, A. and Crist, J. (1988). Electrophysiological characterization of functionally distinct 5-HT receptors on guinea-pig submucous plexus. Neuroscience, 24, 283–295. Surprenant, A. and North, R.A. (1985). µ-Opioid receptors and α2-adrenoceptors coexist on myenteric but not on submucous neurones. Neuroscience, 16, 425–430. Szekeres, P.G., Muir, A.I., Spinage, L.D., Miller, J.E., Butler, S.I., Smith, A. et al. (2000). Neuromedin U is a potent agonist at the orphan G protein-coupled receptor FM3. Journal of Biological Chemistry, 275, 20247–20250. Tack, J.F. and Wood, J.D. (1992). Actions of noradrenaline on myenteric neurones in the guinea pig antrum. Journal of Autonomic Nervous System, 41, 67–78. Tack, J., Coulie, B., Wilmer, A., Andrioli, A. and Janssens, J. (2000). Influence of sumatriptan on gastric fundus tone and on the perception of gastric distension in man. Gut, 46, 468–473. Takahashi, T., Yamamura, T. and Utsunomiya, J. (1992). Human pancreatic polypeptide, neuropeptide Y and peptide YY reduce the contractile motility by depressing the release of acetylcholine from the myenteric plexus of the guinea pig ileum. Gastroenterologia Japonica, 27, 327–333. Takaki, M., Branchek, T., Tamir, H. and Gershon, M.D. (1985a). Specific antagonism of enteric neuronal serotonin receptors by dipeptides of 5-hydroxytryptophan: evidence that serotonin is a mediator of slow synaptic excitation in the myenteric plexus. Journal of Neuroscience, 5, 1769–1780. Takaki, M., Mawe, G.M., Barasch, J. and Gershon, M.D. (1985b). Physiological responses of guinea pig myenteric neurons secondary to the release of endogenous serotonin by tryptamine. Neuroscience, 16, 223–240. Takeda, T., Taniyama, K., Baba, S. and Tanaka, C. (1989). Putative mechanisms involved in excitatory and inhibitory effects of somatostatin on intestinal motility. American Journal of Physiology, 257, G532–G538. Tam, F.S., Hillier, K., Bunce, K.T. and Grossman, C. (1995). Differences in response to 5-HT4 receptor agonists and antagonists of the 5-HT4-like receptor in human colon circular smooth muscle. British Journal of Pharmacology, 115, 172–176. Tamura, K., Palmer, J.M. and Wood, J.D. (1987). Galanin suppresses nicotinic synaptic transmission in the myenteric plexus of guinea-pig small intestine. European Journal of Pharmacology, 136, 445–446. Tanaka, C. and Taniyama, K. (1985). Substance P provoked γ-aminobutyric acid release from the myenteric plexus of the guinea-pig small intestine. Journal of Physiology, 362, 319–329. Taneja, D.T. and Clarke, D.E. (1992). Evidence for a noradrenergic innervation to “atypical” β adrenoceptors (or putative β-3 adrenoceptors) in the ileum of guinea pig. Journal of Pharmacology and Experimental Therapeutics, 260, 192–200. Tange, A. (1983). Distribution of peptide-containing endocrine cells and neurons in the gastrointestinal tract of the dog: immunohistochemical studies using antisera to somatostatin, substance P, vasoactive intestinal polypeptide, met-enkephalin, and neurotensin. Biomedical Research, 4, 9–25. Taniguchi, T., Nishikawa, H., Yokotani, K. and Fujiwara, M. (1988). The bindings of 3H-prazosin and 3 H-yohimbine to α adrenoceptors in the guinea-pig stomach. Life Sciences, 42, 2341–2347. Taniyama, K., Miki, Y., Kusunoki, M., Saito, N. and Tanaka, C. (1983a). Release of endogenous and labeled GABA from isolated guinea pig ileum. American Journal of Physiology, 245, G717–G721. Taniyama, K., Kusunoki, M., Saito, N. and Tanaka, C. (1983b). GABA evoked ACh release from isolated guinea pig ileum. Life Sciences, 32, 2349–2353.

PHARMACOLOGY OF THE ENTERIC NERVOUS SYSTEM

291

Taniyama, K., Kuno, T. and Tanaka, C. (1987). Distribution of β-adrenoceptors associated with cAMP-generating system in cat colon. American Journal of Physiology, 253, G378–G382. Taniyama, K., Nakayama, S., Takeda, K., Matsuyama, S., Shirakawa, J., Sano, I. et al. (1991). Cisapride stimulates motility of the intestine via the 5-hydroxytryptamine receptors. Journal of Pharmacology and Experimental Therapeutics, 258, 1098–1104. Tatemoto, K., Carlquist, M. and Mutt, V. (1982). Neuropeptide Y – a novel brain peptide with structural similarities to peptide YY and pancreatic polypeptide. Nature, 296, 659–660. Tatemoto, K., Rökaeus, Å., Jornvall, H., McDonald, T.J. and Mutt, V. (1983). Galanin – a novel biologically active peptide from porcine ileum. FEBS Letters, 164, 124–128. Taylor, G.S. and Bywater, R.A.R. (1986). Antagonism of non-cholinergic excitatory junction potentials in the guinea-pig ileum by a substance P analogue antagonist. Neuroscience Letters, 63, 23–26. Taylor, S.J. and Kilpatrick, G.J. (1992). Characterization of histamine-H3 receptors controlling non-adrenergic non-cholinergic contractions of the guinea-pig isolated ileum. British Journal of Pharmacology, 105, 667–674. Teitelbaum, D.H., O’Dorisio, T.M., Perkins, W.E. and Gaginella, T.S. (1984). Somatostatin modulation of peptide-induced acetylcholine release in guinea-pig ileum. American Journal of Physiology, 246, G509–G514. Teng, B., Murthy, K.S., Kuemmerle, J.F., Grider, J.R. and Makhlouf, G.M. (1998). Selective expression of vasoactive intestinal peptide (VIP) 2/pituitary adenylate cyclase-activating polypeptide (PACAP) 3 receptors in rabbit and guinea pig gastric and taenia coli smooth muscle cells. Regulatory Peptides, 77, 127–134. Tesfamariam, B. and Allen, G.T. (1994). Beta1- and beta2-adrenoceptor antagonist activities of ICI-215001, a putative beta3-adrenoceptor agonist. British Journal of Pharmacology, 112, 55–58. Thornbury, K.D., Ward, S.M., Dalziel, H.H., Carl, A., Westfall, D.P. and Sanders, K.M. (1991). Nitric oxide and nitrocysteine mimic nonadrenergic, noncholinergic hyperpolarization in canine proximal colon. American Journal of Physiology, 261, G553–G557. Tippins, J.R., Morris, H.R., Panico, M., Etienne, T., Bevis, P., Girgis, S. et al. (1984). The myotropic and plasma-calcium modulating effects of calcitonin gene-related peptide (CGRP). Neuropeptides, 4, 425–434. Tomlinson, A.E. and Poyner, D.R. (1996). Multiple receptors for calcitonin gene-related peptide and amylin on guinea-pig ileum and vas deferens. British Journal of Pharmacology, 117, 1362–1368. Tonini, M. (1992). An appraisal of the action of cisapride and other substituted benzamide prokinetics on myenteric neurons. In Gastrointestinal Dysmotility: Focus on Cisapride, edited by R.C. Heading and J.D. Wood, pp. 85–91. New York: Raven Press. Tonini, M. and Costa, M. (1990). A pharmacological analysis of the neuronal circuitry involved in distensionevoked enteric excitatory reflex. Neuroscience, 38, 787–795. Tonini, M., Frigo, G., Lecchini, S., D’Angelo, L. and Crema, A. (1981). Hyoscine-resistant peristalsis in guineapig ileum. European Journal of Pharmacology, 71, 375–381. Tonini, M., Onori, L., Lecchini, S., Frigo, G.M., Perucca, E. and Crema, A. (1982). Mode of action of ATP on propulsive activity in rabbit colon. European Journal of Pharmacology, 82, 21–28. Tonini, M., Onori, L., Rizzi, C.A., Perucca, E., Manzo, L. and Crema, A. (1987). Involvement of substance P in the excitatory action of GABAA agonists on cholinergic neurons in the guinea-pig ileum. NaunynSchmiedebergs Archives of Pharmacology, 335, 629–635. Tonini, M., Galligan, J.J. and North, R.A. (1989). Effects of cisapride on cholinergic neurotransmission and propulsive motility in the guinea pig ileum. Gastroenterology, 96, 1257–1264. Tonini, M., De Petris, G., Onori, L., Manzo, L., Rizzi, C.A. and Crema, A. (1989). The role of GABAA receptor function in peristaltic activity of the guinea-pig ileum: a comparative study with bicuculline, SR 95531 and picrotoxinin. British Journal of Pharmacology, 97, 556–562. Tonini, M., Candura, S., Onori, L., Coccini, T., Manzo, L. and Rizzi, C.A. (1992a). 5-Hydroxytryptamine4 receptor agonists facilitate cholinergic transmission in the circular muscle of guinea pig ileum: antagonism by tropisetron and DAU 6285. Life Sciences, 50, PL173–PL178. Tonini, M., Coccini, T., Onori, L., Candura, S., Rizzi, C.A. and Manzo, L. (1992b). The influence of neuronal 5-hydroxytryptamine receptor antagonists on non-cholinergic ganglionic transmission in the guinea pig enteric excitatory reflex. British Journal of Pharmacology, 107, 5–7. Tonini, M., Costa, M., Brookes, S.J. and Humphreys, C.M. (1996). Dissociation of the ascending excitatory reflex from peristalsis in the guinea-pig small intestine. Neuroscience, 73, 287–297. Tonini, M., Fiori, E., Balestra, B., Spelta, V., D’Agostino, G., Di Nucci, A. et al. (1998). Endomorphin-1 and endomorphin-2 activate µ-opioid receptors in myenteric neurons of the guinea-pig small intestine. NaunynSchmiedebergs Archives of Pharmacology, 358, 686–689. Tonini, M., De Giorgio, R., De Ponti, F., Sternini, C., Spelta, V., Dionigi, P. et al. (2000). Role of nitric oxideand vasoactive intestinal polypeptide-containing neurones in human gastric fundus strip relaxations. British Journal of Pharmacology, 129, 12–20. Tonini, M., Waterman, S.A., Candura, S.M., Coccini, T. and Costa, M. (1992c). Sites of action of morphine on the ascending excitatory reflex in the guinea-pig small intestine. Neuroscience Letters, 144, 195–198.

292

INNERVATION OF THE GASTROINTESTINAL TRACT

Tonini, M., Spelta, V., De Ponti, F., De Giorgio, R., D’Agostino, G., Stanghellini, V. et al. (2001). Tachykinindependent and -independent components of peristalsis in the guinea pig isolated distal colon. Gastroenterology, 120, 938–945. Tousignant, C., Guillemette, G., Barabe, J., Rhaleb, N.E. and Regoli, D. (1991). Characterization of kinin binding sites: identity of B2 receptors in the epithelium and the smooth muscle of the guinea pig ileum. Canadian Journal of Physiology and Pharmacology, 69, 818–825. Trendelenburg, P. (1917). Physiologische und Pharmakologische Versuche über die Dünndarmperistaltik. Naunyn-Schmiederbergs Archives of Experimental Pathology and Pharmacology, 81, 51–129. Trzeciakowski, J.P. (1987). Inhibition of guinea pig ileum contractions mediated by a class of histamine receptor resembling the H3 subtype. Journal of Pharmacology and Experimental Therapeutics, 243, 874–880. Tuladhar, B.R., Kaisar, M. and Naylor, R.J. (1997). Evidence for a 5-HT3 receptor involvement in the facilitation of peristalsis on mucosal application of 5-HT in the guinea-pig isolated ileum. British Journal of Pharmacology, 122, 1174–1178. Vadokas, B., Ludtke, F.E., Lepsien, G., Golenhofen, K. and Mandrek, K. (1997). Effects of gastrin-releasing peptide (GRP) on the mechanical activity of the human ileocaecal region in vitro. Neurogastroenterology and Motility, 9, 265–270. Van Assche, G., Depoortere, I. and Peeters, T.L. (1995). Motilin binding to microsomal and synaptosomal membranes from rabbit antrum and distal colon. Gastroenterology, 108, A703. Van Nueten, J.M., Janssen, P.A.J. and Fontaine, J. (1976). Unexpected reversal effects of naloxone on the guinea-pig ileum. Life Sciences, 18, 803–809. Van Nueten, J.M., Van Ree, J.M. and Vanhoutte, P.M. (1977). Inhibition by met-enkephalin of peristaltic activity in the guinea-pig ileum, and its reversal by naloxone. European Journal of Pharmacology, 41, 341–342. van Rossum, D., Hanisch, U.K. and Quirion, R. (1997). Neuroanatomical localization, pharmacological characterization and functions of CGRP, related peptides and their receptors. Neuroscience Biobehavioral Reviews, 21, 649–678. Vanhoenacker, P., Haegeman, G. and Leysen, J.E. (2000). 5-HT7 receptors: current knowledge and future prospects. Trends in Pharmacological Sciences, 21, 70–77. Venkova, K., Milne, A. and Krier, J. (1994). Contractions mediated by α1-adrenoceptors and P2-purinoceptors in a cat colon circular muscle. British Journal of Pharmacology, 112, 1237–1243. Verplanken, P.A., Lefebvre, R.A. and Bogaert, M.G. (1984). Pharmacological characterization of alpha adrenoceptors in the rat gastric fundus. Journal of Pharmacology and Experimental Therapeutics, 231, 404–410. Vincent, S.R. and Hope, B.T. (1992). Neurons that say NO. Trends in Neurosciences, 15, 108–113. Vincent, J.P., Mazella, J. and Kitabgi, P. (1999). Neurotensin and neurotensin receptors. Trends in Pharmacological Sciences, 20, 302–309. Vizi, S.E. (1979). Presynaptic modulation of neurochemical transmission. Progress in Neurobiology, 12, 181–290. Vizi, S.E., Bertaccini, G., Impicciatore, M. and Knoll, J. (1972). Acetylcholine-releasing effect of gastrin and related polypeptides. European Journal of Pharmacology, 17, 175–178. Vizi, S.E., Bertaccini, G., Impicciatore, M. and Knoll, J. (1973). Evidence that acetylcholine released by gastrin and related polypeptides contributes to their effect on gastrointestinal motility. Gastroenterology, 64, 268–277. Vizi, S.E., Ono, K., Adam-Vizi, V., Duncalf, D. and Foldes, F.F. (1984). Presynaptic inhibitory effect of met-enkephalin on [14C]acetylcholine release from the myenteric plexus and its interaction with muscarinic negative feedback inhibition. Journal of Pharmacology and Experimental Therapeutics, 230, 493–499. Vizi, S.E., Kobayashi, O., Törocsik, A., Kinjo, M., Nagashima, H., Manabe, N., Goldiner, P.L., Potter, P.E. and Foldes, F.F. (1989). Heterogeneity of presynaptic muscarinic receptors involved in modulation of transmitter release. Neuroscience, 31, 259–267. Voderholzer, W.A., Allescher, H.D. and Muller-Lissner, S.A. (1995). The effect of hormones and peptides involved in water balance on rat colonic motility in vitro. Neurogastroenterology and Motility, 7, 15–21. Wagner, F. and Wahl, M. (1986). Effects of neurotensin in feline pial arteries, guinea-pig ileum and portal vein of rat and guinea-pig. Archives Internationales de Pharmacodynamie et de Therapie, 282, 240–251. Wang, Y.F., Mao, Y.K., Fox-Threlkeld, J.E., McDonald, T.J. and Daniel, E.E. (1998). Colocalization of inhibitory mediators, NO, VIP and galanin, in canine enteric nerves. Peptides, 19, 99–112. Wardle, K.A. and Sanger, G.J. (1993). The guinea-pig distal colon – a sensitive preparation for the investigation of 5-HT4 receptor-mediated contractions. British Journal of Pharmacology, 110, 1593–1599. Watanabe, T., Taguchi, Y., Shiosaka, S., Tanaka, J., Kubota, H., Terano, Y. et al. (1984). Distribution of the histaminergic neuron system in the central nervous system of rats: a fluorescent immunohistochemical analysis with histidine decarboxylase as a marker. Brain Research, 295, 13–26. Waterfield, A.A. and Kosterlitz, H.W. (1975). Stereospecific increase by narcotic antagonists of evoked acetylcholine output in guinea-pig ileum. Life Sciences, 16, 1787–1792.

PHARMACOLOGY OF THE ENTERIC NERVOUS SYSTEM

293

Waterman, S.A., Costa, M. and Tonini, M. (1992). Modulation of peristalsis in the guinea-pig isolated small intestine by exogenous and endogenous opioids. British Journal of Pharmacology, 106, 1004–1010. Waterman, S.A., Costa, M. and Tonini, M. (1994). Accommodation mediated by enteric inhibitory reflexes in the isolated guinea-pig small intestine. Journal of Physiology, 474, 539–546. Waters, S.M. and Krause, J.E. (2000). Distribution of galanin-1, -2 and -3 receptor messenger RNAs in central and peripheral rat tissues. Neuroscience, 95, 265–271. Watson, M., Bywater, R.A.R. and Taylor, G.S. (1992). Nitric oxide mediates the slow apamin-insensitive inhibitory junction potential in the proximal and distal colon of the guinea-pig. Proceedings of the Australian Physiology and Pharmacology Society, 23, 5P. Wells, D.G. and Mawe, G.M. (1994). Sympathetic input to ganglia of the guinea-pig sphincter of Oddi. American Journal of Physiology, 266, G1162–G1169. Wessler, I., Eshenbruch, V., Halim, S. and Kilbinger, H. (1987). Presynaptic effects of scopolamine, oxotremorine, noradrenaline and morphine on [3H]acetylcholine release from the myenteric plexus at different stimulation frequencies and calcium concentrations. Naunyn-Schmiedebergs Archives of Pharmacology, 335, 597–604. Weston, A.H. (1973). The effect of desensitization to adenosine triphosphate on the peristaltic reflex in guineapig ileum. British Journal of Pharmacology, 47, 606–608. White, T.D. (1982). Release of ATP from isolated myenteric varicosities by nicotinic agonists. European Journal of Pharmacology, 79, 333–334. White, T.D. and Al-Humayyd, M. (1983). Acetylcholine releases ATP from varicosities isolated from guinea-pig myenteric plexus. Journal of Neurochemistry, 40, 1069–1075. White, T.D. and Leslie, R.A. (1982). Depolarization-induced release of adenosine 5′-triphosphate from isolated varicosities derived from the myenteric plexus of the guinea pig small intestine. Journal of Neuroscience, 2, 206–215. Wikberg, J. (1978). Differentiation between pre- and postjunctional-receptors in guinea pig ileum and rabbit aorta. Acta Physiologica Scandinavica, 103, 225–239. Wikberg, J. and Lefkowitz, R.J. (1982). Alpha2 adrenergic receptors are located prejunctionally in the Auerbach’s plexus of the guinea pig small intestine: direct demonstration by radioligand binding. Life Sciences, 31, 2899–2905. Wiklund, N.P., Wiklund, C.U., Ohlen, A. and Gustafsson, L.E. (1989). Cholinergic neuromodulation by endothelin in guinea pig ileum. Neuroscience Letters, 101, 342–346. Wiklund, C.U., Olgart, C., Wiklund, N.P. and Gustafson, L.E. (1993). Modulation of cholinergic and substance P-like neurotransmission by nitric oxide in the guinea-pig ileum. British Journal of Pharmacology, 110, 833–839. Wiley, J.W., Lu, Y. and Owyang, C. (1991). Evidence for a glutamatergic neural pathway in the myenteric plexus. American Journal of Physiology, 261, G693–G700. Willard, A.L. (1990). Substance P mediates synaptic transmission between rat myenteric neurones in cell culture. Journal of Physiology (London), 426, 453–471. Williams, J.A. (1982). Cholecystokinin: a hormone and a neurotransmitter. Biomedical Research, 3, 107–121. Williams, J.T. and North, R.A. (1978). Inhibition of firing of myenteric neurones by somatostatin. Brain Research, 155, 165–168. Williams, J.T. and North, R.A. (1979). Vasoactive intestinal polypeptide excites neurones of the myenteric plexus. Brain Research, 175, 174–177. Williams, J.T., Katayama, Y. and North, R.A. (1979). The action of neurotensin on single myenteric neurones. European Journal of Pharmacology, 59, 181–186. Williamson, S., Faulkner-Jones, B.E., Cram, D.S., Furness, J.B. and Harrison, L.C. (1995). Transcription and translation of two glutamate decarboxylase genes in the ileum of rat, mouse and guinea pig. Journal of Autonomic Nervous System, 55, 18–28. Williamson, S., Pompolo, S. and Furness, J.B. (1996). GABA and nitric oxide synthase immunoreactivities are colocalized in a subset of inhibitory motor neurons of the guinea-pig small intestine. Cell and Tissue Research, 284, 29–37. Yanagisawa, M., Kurihara, H., Kimura, S., Tomobe, Y., Kobayashi, M., Mitsui, Y. et al. (1988). A novel potent peptide produced by vascular endothelial cells. Nature, 332, 411–415. Yau, W.M. (1978). Effect of substance P on intestinal muscle. Gastroenterology, 74, 228–231. Yau, W.M. and Verdun, P.R. (1983). Release of γ-aminobutyric acid from guinea pig myenteric plexus synaptosomes. Brain Research, 278, 271–273. Yau, W.M., Makhlouf, G.M., Edwards, L.E. and Farrar, J.T. (1974). The action of cholecystokinin and related peptides on guinea pig small intestine. Canadian Journal of Physiology and Pharmacology, 52, 298–303. Yau, W.M., Lingle, P.F. and Youther, M.L. (1983). Modulation of cholinergic neurotransmitter release from myenteric plexus by somatostatin. Peptides, 4, 49–53.

294

INNERVATION OF THE GASTROINTESTINAL TRACT

Yau, W.M., Verdun, P.R. and Youther, M.L. (1983). Neurotensin: a modulator of enteric cholinergic neurons in the guinea pig small intestine. European Journal of Pharmacology, 95, 253–258. Yau, W.M., Youther, M.L. and Verdun, P.R. (1985). A presynaptic site of action of substance P and vasoactive intestinal polypeptide on myenteric neurons. Brain Research, 330, 382–385. Yau, W.M., Dorsett, J.A. and Youther, M.L. (1986a). Evidence for galanin as an inhibitory neuropeptide on myenteric neurons in the guinea pig small intestine. Neuroscience Letters, 72, 305–308. Yau, W.M., Dorsett, J.A. and Youther, M.L. (1986b). Inhibitory peptidergic neurons: functional difference between somatostatin and enkephalin in myenteric plexus. American Journal of Physiology, 250, G60–G63. Yau, W.M., Dorsett, J.A. and Parr, E.L. (1989). Characterization of acetylcholine release from enzymedissociated myenteric ganglia. American Journal of Physiology, 256, G233–G239. Yau, W.M., Mandel, K.G., Dorsett, J.A. and Youther, M.L. (1992). Neurokinin3 receptor regulation of acetylcholine release from myenteric plexus. American Journal of Physiology, 263, G659–G664. Yazdani, A., Takahashi, T., Bagnol, D., Watson, S.J. and Owyang, C. (1999). Functional significance of a newly discovered neuropeptide, orphanin FQ, in rat gastrointestinal motility. Gastroenterology, 116, 108–117. Yokotani, K. and Osumi, Y. (1993). Cholinergic M2 muscarinic receptor-mediated inhibition of endogenous noradrenaline release from the isolated vascularly perfused rat stomach. Journal of Pharmacology and Experimental Therapeutics, 264, 54–60. Yoshinaga, M., Chijiiwa, Y., Misawa, T., Harada, N. and Nawata, H. (1992). EndothelinB receptor on guinea pig small intestinal smooth muscle cells. American Journal of Physiology, 262, G308–G311. Young, H.M. and Furness, J.B. (1991). An ultrastructural study of the 5-HT immunoreactive inputs to neurons of the myenteric plexus of the guinea-pig ileum. Proceedings of the Australian Physiology and Pharmacology Society, 22, 106P. Yunker, A.M.R. and Galligan, J.J. (1996). Endogenous NO inhibits NANC but not cholinergic neurotransmission to circular muscle of guinea-pig ileum. American Journal of Physiology, 271, G904–G912. Zadina, J.E., Hackler, L., Ge, L.J. and Kastin, A.J. (1997). A potent and selective endogenous agonist for the mu-opiate receptor. Nature, 386, 499–502. Zafirov, D.H., Palmer, J.M., Nemeth, P.R. and Wood, J.D. (1985). Bombesin, gastrin releasing peptide and vasoactive intestinal peptide excite myenteric neurons. European Journal of Pharmacology, 115, 103–107. Zagorodnyuk, V. and Maggi, C.A. (1994). Electrophysiological evidence for different release mechanism of ATP and NO as inhibitory NANC transmitters in guinea-pig colon. British Journal of Pharmacology, 112, 1077–1082. Zagorodnyuk, V. and Maggi, C.A. (1995). Neuronal tachykinin NK2 receptors mediate release of non-adrenergic non-cholinergic inhibitory transmitters in the circular muscle of the guinea-pig colon. Neuroscience, 69, 643–650. Zagorodnyuk, V., Santicioli, P. and Maggi, C.A. (1998). Evidence for the involvement of multiple mechanisms in the excitatory action of bradykinin in the circular muscle of guinea-pig colon. Naunyn-Schmiedebergs Archives of Pharmacology, 357, 197–204. Zavecz, J.H. and Yellin, T.O. (1982). Histamine receptors in the myenteric plexus-longitudinal muscle of the guinea-pig ileum: H1- and H2-receptor-mediated potentiation of the contractile response to electrical stimulation. Journal of Pharmacology and Experimental Therapeutics, 223, 177–182. Zeiter, D.K., Li, X. and Broussard, D.L. (1996). Identification of the GABAA receptor alpha-subunit mRNA in rat intestine. Brain Research Molecular Brain Research, 39, 241–244. Zetler, G. (1979). Antagonism of cholecystokinin-like peptides by opioid peptides, morphine or tetrodotoxin. European Journal of Pharmacology, 60, 67–77. Zetler, G. (1980). Antagonism of the gut-contracting effects of bombesin and neurotensin by opioid peptides, morphine, atropine or tetrodotoxin. Pharmacology, 21, 348–354. Zhang, L., Keef, K.D., Bradley, M.E. and Buxton, I.L.O. (1992). Action of α2A-adrenergic receptors in circular smooth muscle of canine proximal colon. American Journal of Physiology, 262, G517–G524. Zhou, X. and Galligan, J.J. (2000). GABAA receptors on calbindin-immunoreactive myenteric neurons of guinea pig intestine. Journal of Autonomic Nervous System, 78, 122–135. Zuzack, J.S., Burkard, M.R., Cuadrado, D.K., Greer, R.A., Selig, W.M. and Whalley, E.T. (1996). Evidence of a bradykinin B1 receptor in human ileum: pharmacological comparison to the rabbit aorta B1 receptor. Journal of Pharmacology and Experimental Therapeutics, 277, 1337–1343.

7 Neuroeffector Transmission in the Intestine Charles H.V. Hoyle1, Pam Milner2 and Geoffrey Burnstock2 1

Department of Anatomy and Developmental Biology, University College London, Gower Street, London WC1E 6BT, UK 2 Autonomic Neuroscience Institute, Royal Free and University College Medical School, Royal Free Campus, Rowland Hill Street, London NW3 2PF, UK This chapter focuses on neuromuscular and neuroepithelial transmission in the intestine. There seems to be an ever increasing number of putative neurotransmitters in the enteric nervous system, but relatively few have a defined role in either neuromuscular or neuroepithelial transmission even though their chemical coding may have been established. In addition to covering the “old” classic transmitters, acetylcholine and noradrenaline, some more novel candidate transmitters such as endothelin and pituitary adenylate cyclase-activating polypeptide are discussed, as well as the “modern” classical transmitters ATP and vasoactive intestinal polypeptide (VIP), and the notso-classical nitric oxide. Co-transmission is a widespread phenomenon in the enteric nervous system and particularly with regard to inhibitory transmission, it is difficult to discuss one transmitter, such as ATP, in isolation from others such as nitric oxide and VIP. An area of current research is the plasticity of the enteric nervous system, and how expression of neurotransmitters changes in response to ageing, trauma, surgery or chronic drug treatment, and what the mechanisms underlying these changes are. Some pathological conditions are considered, namely diabetes mellitus, idiopathic chronic constipation, Hirschsprung’s disease and ulcerative colitis, in all of which intestinal neuromuscular transmission may be affected. KEY WORDS: intestinal neuromuscular transmission; intestinal neuroepithelial transmission; enteric nervous system; co-transmission; neuroplasticity.

INTRODUCTION HISTORICAL CONCEPTS Studies of neurotransmission in the enteric nervous system have played an important role in revealing the diversity of the autonomic nervous system as a whole and in developing new concepts in neurotransmission that are now being fully realised. The first clear evidence for non-adrenergic, non-cholinergic (NANC) neurotransmission came from two studies in 1963, one an electrophysiological study on guinea-pig intestine and 295

296

INNERVATION OF THE GASTROINTESTINAL TRACT

the other a pharmacological study on cat stomach. These studies showed recordings of inhibitory junction potentials (IJPs) in intestinal smooth muscle during stimulation of enteric nerves in the presence of adrenergic and cholinergic blocking agents (Burnstock et al., 1963) and relaxation of stomach following vagal stimulation with adrenergic and cholinergic blockade (Martinson and Muran, 1963). They challenged the classical view of antagonistic actions of noradrenaline and acetylcholine causing either constriction or relaxation. In 1970, the purine nucleotide, adenosine 5′-triphosphate (ATP), was proposed as a principal transmitter in NANC neurotransmission (Burnstock et al., 1970; Burnstock, 1972). The recent molecular biological approaches to identify the molecular structures for purine receptor subtypes has greatly reinforced the concept of purinergic neurotransmission (Burnstock, 1997). MULTIPLICITY OF NEUROTRANSMITTERS IN THE ENTERIC NERVOUS SYSTEM Hints that there are in fact several different neurotransmitters in autonomic nerves came from ultrastructural studies of the enteric nervous system. At least 9 distinguishable types of axon profile were described (Cook and Burnstock, 1976). Subsequently, using newly available immunohistochemical techniques, several biologically active peptides were localised in neural elements of the gut (Furness and Costa, 1987). In addition to many polypeptides, 5-hydroxytryptamine (5-HT), dopamine and γ-aminobutyric acid (GABA) were proposed as autonomic neurotransmitters (see Gershon, Mawe and Branchek, 1989; Hills and Jessen, 1992). More recently, nitric oxide (NO) has been added to the list of neurotransmitters in the gastrointestinal tract (for reviews see Rand, 1992; Saffery et al., 1992; Sanders and Ward, 1992; Lefebvre, 1995; Lincoln, Hoyle and Burnstock, 1997). The rapid expansion of the number of proposed enteric neurotransmitters in recent years, including endothelin (Inagaki et al., 1991; Lin and Lee, 1992), vasoactive intestinal contractor (Saida, Mitsui and Ishida, 1989), secretoneurin (Schmid et al., 1995; Schürman et al., 1995; Dun et al., 1997), glutamate (Burns and Stephens, 1995) and carbon monoxide (Rattan and Chakder, 1993; Verma et al., 1993; Werkström et al., 1997), makes it likely that the list is still incomplete (Table 7.1). CHEMICAL CODING, CO-TRANSMISSION AND NEUROMODULATION The concept of co-transmission, that some nerves release more than one neurotransmitter (Burnstock, 1976) is well illustrated in the enteric nervous system where several different neuropeptides have been localised in a single neurone. Many of these substances act as neuromodulators, enhancing or diminishing the release or actions of primary transmitters. The precise combinations of substances contained in individual enteric neurones and their projections and central connections, termed “chemical coding”, have been defined in an elaborate series of surgical manipulations of the guinea-pig intestine (for reviews see Furness and Costa, 1987; Furness et al., 1992). Fourteen separate classes of neurones, accounting for more than 90% of the myenteric neurones of the guinea-pig ileum, have now been identified (Costa et al., 1996). Whilst there are some inter-species and interregional variabilities, some aspects appear to be constant: the primary transmitters are

NEUROEFFECTOR TRANSMISSION IN THE INTESTINE

297

TABLE 7.1 Proposed neurotransmitters/neuromodulators in the enteric nervous system. Acetylcholine Adenosine 5′-triphosphate Angiotensin Calcitonin gene-related peptide Carbon monoxide Cholecystokinin Dopamine Dynorphins Endorphins Endothelin Enkephalins Galanin γ-Aminobutyric acid Gastrin Gastrin releasing peptide/bombesin Helodermin Helospectin 5-Hydroxytryptamine MERGL

Neo-endorphin Neurokinin A, B Neuromedin B, C, U Neuropeptide Y Neurophysin Neurotensin Nitric oxide Noradrenaline Oxytocin Peptide histidine isoleucine Peptide histidine methionine Pituitary adenylate cyclase-activating peptide Secretoneurin Somatostatin Substance P Thyrotropin-releasing hormone Vasoactive intestinal peptide Vasoactive intestinal contractor

generally well preserved in functionally equivalent neurones but their co-transmitters and neuromodulators may differ between species and regions of the gastrointestinal tract (Furness et al., 1994). Use of high affinity antagonists for many peptide receptors and recent advances in cloning genes encoding peptide receptor subtypes have enabled a detailed description of the molecular events involved in enteric peptidergic transmission (Dockray, 1994). Although there are many different transmitter substances in the gut, most are involved in neurotransmission or neuromodulation at the ganglion level and/or may have a trophic role. The number involved in neuromuscular transmission is more limited.

AUTONOMY OF THE ENTERIC NERVOUS SYSTEM The ability of the enteric nervous system to sustain local reflex activity independent of the central nervous system has been recognised for many years since the original studies of Bayliss and Starling (1899). This functional autonomy is attributed to the integrated intrinsic ganglionated plexuses which innervate the gastrointestinal tract. Scanning and electronmicroscopic studies have shown that the organisation of these ganglia is closer to that of the central nervous system than that of the sympathetic or parasympathetic ganglia (Gabella, 1972; Jessen and Burnstock, 1982; Gabella, 1990). Similarities with the central nervous system include the compact organisation of neural and glial cells with a paucity of extracellular space, the lack of penetration of either connective tissue or blood vessels into the ganglia and the limited access of intravascular molecules i.e. a blood-ganglion barrier analogous to the blood-brain barrier. For these reasons, enteric tissues have been proposed

298

INNERVATION OF THE GASTROINTESTINAL TRACT

as grafts into the central nervous system as prospective treatment of neurodegenerative diseases. Enteric neurones and associated cells implanted together with the surrounding smooth muscle are able to survive in the corpus striatum for up to a year and stimulate axonal sprouting in the central nervous system (Tew, Anderson and Burnstock, 1992; Tew et al., 1996).

INTRINSIC GANGLIONATED PLEXUSES Of the two main interconnecting ganglionated plexuses of the enteric nervous system, the myenteric plexus, which lies between the external longitudinal and circular muscle coats, and the submucous plexus, which lies between the circular muscle and muscularis mucosae, the myenteric plexus contains most of the intrinsic neurones of the gut and is the main source of neuromuscular innervation of the gut wall (Furness and Costa, 1987). Using guinea-pig small intestine, it has been shown that the motor neurones for the circular muscle are located entirely in the myenteric plexus whereas most of those for the mucosa are located in the submucosal ganglia. Intrinsic myenteric neurones also synapse with cells in the same or other ganglia, running both orally and anally, and project to submucosal ganglia. They also connect with autonomic ganglia outside the walls of the gastrointestinal tract and send afferents to the central nervous system. Some enteric neurones project from the intestine to innervate the mesenteric arteries and arterioles of the colon (Holzer, Gamse and Lembeck, 1980). Videomicroscopic analyses have revealed that intrinsic submucosal neurones, by regulating neurogenic vasodilatation of submucosal arterioles, are involved in local physiological control of mucosal blood flow (Neild, Shen and Surprenant, 1990). Intestinal secretion is largely under reflex neural control by the enteric nervous system. Endocrine cells of the intestinal epithelium are receptive to stimulants that evoke secretion of peptides and amines from their basal aspect, and activate the neurones that run close to the epithelium. These neurones run to the myenteric plexus and from there, via interneurones, to the submucous plexus (Jodal, 1990).

EXTRINSIC INNERVATION Post-ganglionic sympathetic nerve fibres from the coeliac and superior and inferior mesenteric ganglia and pre-ganglionic parasympathetic fibres running in the vagus and pelvic nerves innervate all parts of the gut, including the nerve plexuses, muscle layers, blood vessels and epithelium. Many sympathetic nerves are associated with blood vessels and are related to vasomotor action. The vasoconstrictor innervation to the arteriolar network in the submucous plexus is mediated solely by extrinsic sympathetic nerves that release primarily ATP to act on P2X receptors on the arteriolar smooth muscle, with noradrenaline acting as a prejunctional modulator via α2-adrenoceptors causing depression of neurotransmission (Evans and Surprenant, 1992). Sensory-motor nerves with their cell bodies in dorsal root ganglia also project to the gut and blood vessels in the submucosa (Szurszewski and King, 1989). Submucosal arteriolar vasodilatation is mediated by extrinsic sensory reflex pathways which are likely to be activated during inflammatory conditions (Vanner and Surprenant, 1996).

NEUROEFFECTOR TRANSMISSION IN THE INTESTINE

299

NEUROMUSCULAR TRANSMISSION CONTROL OF INTESTINAL SMOOTH MUSCLE CONTRACTILITY The extrinsic and intrinsic control of gut smooth muscle operates in concert with the hormonal and mechanical control mechanisms stimulated by gut contents. The dominant role of the intrinsic innervation is apparent after sympathetic and parasympathetic denervation. Intrinsic neurones are either excitatory or inhibitory. Localised distension of the intestine, mechanical deformation of the mucosa or application of acid to the mucosa elicits both ascending excitatory and descending inhibitory reflexes to the circular muscle (Smith, Bornstein and Furness, 1990). A descending excitatory reflex to the circular muscle has also been reported (Kow, Brookes and Costa, 1993). These polarised events involve a chain of neurones, sensory interneurones and motor neurones. In the guinea-pig small intestine, enteric motility reflexes can be initiated through entirely intrinsic mechanisms (Furness et al., 1995, 1998). Definitive identification of intrinsic primary afferent neurones has recently been established by intracellular recordings in response to chemical and mechanical stimuli of the gut (Furness et al., 1998).

AUTONOMIC NEUROMUSCULAR JUNCTION The junction between autonomic nerve terminals and smooth muscle is not a well-defined structure and lacks both the pre- and postjunctional specialisations found at a skeletal muscle motor end plate (Gabella, 1972, 1995; Burnstock, 1986). Autonomic nerves do not release transmitter solely from terminals per se, but rather from varicosities that occur at intervals of 5–15 µm along axons. In autonomically innervated organs the distance of the cleft between the varicosity and the smooth muscle, a neuromuscular junction, is between 20 nm (as in the vas deferens or sphincter pupillae) and 1–2 µm (in large elastic arteries). Short, thickened regions of the membranes of varicosities are associated with aggregation of vesicles, and these membranes may represent sites of transmitter release, but post-junctional thickenings are not seen on the smooth muscle membrane. One of the essential features of autonomic neuromuscular transmission is that transmitter is released in passage from varicosities during conduction of an impulse along an autonomic axon. Furthermore, it is possible that a given impulse will evoke release from only some of the varicosities that it encounters (Blakeley and Cunnane, 1979; Blakeley, Cunnane and Petersen, 1982; Brock and Cunnane, 1995). The effector is a muscle bundle rather than a single cell; individual smooth muscle cells are connected by low-resistance bridges that allow electrotonic spread of activity within the effector bundle. Morphologically the sites of electrotonic coupling are represented by gap junctions (or nexuses). The size of these gap junctions varies from punctate junctions to junctional areas of 1 µm2 in diameter. Little is known about the quantity and arrangement of gap junctions in effector bundles relative to the density of autonomic innervation. Circular muscle layers tend to be more densely innervated than longitudinal layers. In small animals such as rats, mice and guinea-pigs, innervation of the longitudinal layer of the intestine is very sparse, with no nerve bundles penetrating the muscle coat. In the guinea-pig taenia coli, nerve bundles containing only 3–5 axons are found in muscle bundles. In the circular

300

INNERVATION OF THE GASTROINTESTINAL TRACT

muscle there may be 25–78 close neuromuscular junctions per 1,000 muscle cells in the toad intestine, and in the guinea-pig small intestine there are claimed to be 10–50 axons of inhibitory junction potential-producing neurones per functional unit of circular muscle (for reviews see Hoyle and Burnstock, 1989; Gabella, 1995). Autonomic neuromuscular junctions are probably labile structures with varicosities that are able to move along axons; the lack of postjunctional specialisation is consistent with this view. Another aspect of this type of junction is that it is particularly accessible at both pre- and postjunctional sites for neuromodulatory influences, where local agents may enhance or exacerbate release of neurotransmitter or alter the extent or time course of neurotransmitter action. The possible role of the interstitial cells of Cajal in neuromuscular transmission in the gut has been debated for many years (Rogers and Burnstock, 1966; Bortoff, 1976; Komuro, 1982). More recently, evidence has been presented that interstitial cells may play a role as pacemaker cells, having intimate contact with neurones and muscle cells (Torihashi et al., 1993). They are themselves contractile and they influence the rate of contraction of the muscle and may modulate neural transmission to the muscle (Thuneberg, 1982; Sanders, 1996; Shuttleworth and Sanders, 1996). The importance of these cells in normal and abnormal gastrointestinal motility is gaining recognition.

EXCITATORY NEUROMUSCULAR TRANSMISSION Acetylcholine Excitatory enteric neurones that project to intestinal smooth muscle utilise tachykinins and acetylcholine as neurotransmitters. Acetylcholine, either applied or released from neurones, produces a rapid depolarisation, or excitatory junction potential (EJP), via muscarinic receptors. In response to low-strength stimulation, perhaps a single electrical pulse, the EJP is transient, usually lasting less than 1 s, with a latency of the order of 100 ms. At higher strengths of stimulation the EJP may be accompanied by an action potential (Brock and Cunnane, 1995). Cholinergic EJPs have been recorded from several gut muscles (Hoyle and Burnstock, 1989; Brock and Cunnane, 1995). During repetitive stimulation of cholinergic nerves, EJPs may facilitate and develop faster rise times and larger amplitudes or depression may occur, in which case these parameters reduce. Although increasing either sodium or calcium conductance in a resting membrane will result in a depolarisation, the excitatory effects of acetylcholine acting on muscarinic receptors involve increases in permeabilities to these ions as well as to potassium and perhaps chloride ions. The reversal potential for muscarinic action of around –5 mV is indicative of simultaneous depolarising and hyperpolarising ion fluxes. In the guinea-pig ileum longitudinal muscle and rabbit jejunum and rectum longitudinal muscles, the cholinergically induced EJP reverses between –5 and +5 mV, while the non-adrenergic, non-cholinergic EJP in the guinea-pig ileum reverses at –27 mV. Although this might suggest an involvement of chloride ions, it could also reflect a simultaneous increase in potassium conductance along with sodium or calcium conductance (Hoyle and Burnstock, 1989; Brock and Cunnane, 1995). Single-cell voltage-clamp studies of rabbit ileum longitudinal muscle show that muscarinic receptors open channels that are permeable to sodium

NEUROEFFECTOR TRANSMISSION IN THE INTESTINE

301

and potassium, slightly permeable to calcium, and impermeable to chloride (Bolton and Lim, 1991). Muscarinic receptors are not ligand-gated ion channels: all muscarinic receptors that have been cloned are G protein-coupled receptors. In smooth muscle, excitatory effects are predominantly mediated via G proteins coupled to phosphatidylinositol metabolism and subsequent calcium mobilisation (Gq/11), or to cation channels (Bolton and Lim, 1991; Zholos and Bolton, 1994; Prestwich and Bolton, 1995). Additionally, muscarinic receptors in intestinal smooth muscle may be coupled to a pertussis toxin-sensitive G protein (Gi/o) that is involved in inhibition of calcium-dependent potassium channels (Cole and Sanders, 1989; Prestwich and Bolton, 1995). Pharmacologically, the predominant muscarinic receptor on intestinal smooth muscle is selectively antagonised by 4-diphenylacetoxy-N-methylpiperidine methiodide (4-DAMP), and is of the M3 subtype (Grider, Bitar and Makhlouf, 1987; Lucchesi et al., 1989; De Vos, 1993). M2 receptors are also present, although they form a smaller population (Lucchesi et al., 1989; De Vos, 1993). Substance P Within the gut, substance P-containing neurones do not appear to project for long distances; in transected and re-anastomosed ileum no change in substance P immunoreactivity could be found 2–4 mm away from the injury sites (Keast, Furness and Costa, 1982), implying that the substance P neurones are confined to segments or arcades of the gut. Many excitatory motor neurones contain both acetylcholine and tachykinins whilst fewer contain either acetylcholine or tachykinins (Brookes, Steele and Costa, 1991). Excitatory enteric neurones innervating circular muscle project either locally or orally (Bornstein, Furness and Kunze, 1994) whilst those innervating longitudinal muscle normally have short local projections. Opioid peptides (dynorphin, enkephalin and opioid-related peptides) co-localised in many excitatory enteric neurones do not appear to be primary neurotransmitters but rather have a neuromodulatory role, generally inhibiting motor neurotransmission (Tonini et al., 1992). Calretinin localised in some excitatory neurones projecting to longitudinal muscle is not found in excitatory neurones projecting to circular muscle. Substance P, produces slower responses than does acetylcholine, and its actions are mediated via neurokinin (NK) receptors (Maggi et al., 1990, 1997; Hoyle, 1996a). Tachykinins are preferentially released during high frequency neuronal firing but are probably also released with acetylcholine at lower frequencies (Holzer, Schulet and Maggi, 1993). Non-cholinergic contractile response to high frequency stimulation is effectively blocked by tachyphylaxis to substance P, and by selective NK receptor antagonists. In the superfused ileum in vitro, substance P is released during field stimulation in a manner that is calcium-, temperature-, and frequency-dependent, and tetrodotoxin- (TTX) sensitive; the amount released is enough to produce a physiological response (Baron, Jaffé and Gintzler, 1983). Similarly, high potassium-stimulated substance P release from isolated longitudinal muscle-myenteric plexus preparations is calcium-dependent, whereas field stimulationevoked release of substance P is TTX-sensitive. Raised intraluminal pressure in a segment of guinea-pig ileum, which will initiate peristalsis, is accompanied by substance P release. The substance P antagonist (D-Pro2,D-Trp7,9 )-substance P (DPDTSP) also inhibits atropineresistant peristalsis, as does substance P tachyphylaxis and hexamethonium indicating that

302

INNERVATION OF THE GASTROINTESTINAL TRACT

Figure 7.1 NK1 receptor immunoreactivity in nerve cells in various regions of the guinea-pig gastrointestinal tract. (A) Wholemount preparation of a myenteric ganglion in the corpus of the stomach showing NK1 receptor immunoreactive nerve cell bodies. Cells show Dogiel type 1 morphology and immunoreactivity is largely confined to the cell surface membrane. (B, C) Wholemount preparations of myenteric ganglia in the antrum of the stomach. Staining appears particularly punctate in this region with large aggregations of immunoreactivity dotting the cell surface membrane. (D) Wholemount preparation of a myenteric ganglion in the duodenum illustrating immunoreactive nerve cells (arrow heads) and interstitial cells of Cajal (small arrows) at the level of the myenteric ganglia. The nerve cell bodies exhibit Dogiel type 1 morphology with flattened lamellar dendrites and a prominent axon. (E) Wholemount preparation of the myenteric ganglia of the distal colon showing three NK1 receptor immunoreactive neurones. Scale bars A, B, C, E = 20 µm; D = 50 µm. (Reproduced from Portbury et al., 1996, with permission from Wiley-Liss, Inc., a subsidiary of John Wiley and Sons Inc.).

substance P in intrinsic neurones has an active role in coordination of peristalsis (Barthó et al., 1982). Reflex contractions of circular muscle oral to sites of distension stimuli normally have a low threshold but in the presence of muscarinic blockade they have a high threshold. This high-threshold stimulation is sensitive to the substance P antagonists. Hence substance P neurones are final in a noncholinergic reflex pathway to the circular muscle and perhaps interneuronal, within the myenteric plexus, in a cholinergic reflex pathway (Costa et al., 1985). A recent study of the distribution of NK1 receptors in guinea-pig intestine showing lack of immunoreactivity to these receptors in the smooth muscle but strong immunoreactivity on the interstitial cells of Cajal at the inner surface of the circular muscle supports pharmacological evidence that tachykinin receptors on longitudinal muscle are a subtype of NK1 receptors (Chassaing et al., 1992) and suggests that circular muscle may be indirectly excited via interstitial cells of Cajal (Figure 7.1) (Portbury et al., 1996): these cells are linked to the muscle by gap junctions and are considered to be responsible for pacemaker activity in the gut (Vogalis, Ward and Sanders, 1991; Sanders, 1996). NK2 receptors are localised on circular and longitudinal muscle cells (Grady et al., 1996).

NEUROEFFECTOR TRANSMISSION IN THE INTESTINE

303

In the guinea-pig ileum, substance P stimulates only those smooth muscle cells that have an excitatory neural input (Bauer and Kuriyama, 1982). During continuous application of substance P (0.5 µM), the primary phase of neurogenic depolarisation is abolished, and this depolarisation is inhibited by substance P antagonists, indicating that substance P is the transmitter. There appears to be a functional interaction between acetylcholine and substance P, perhaps released from the same neurones, in that there is a synergistic pro-kinetic action that involves muscarinic and NK2 receptors (Holzer and Maggi, 1994). Substance P and neurokinin A (NKA, a selective NK2 receptor agonist) evoke depolarisation of colonic smooth muscle cells, by opening non-selective cation channels, similar to those opened by acetylcholine, with a reversal potential close to 0 mV (Lee, Shuttleworth and Sanders, 1995). Noradrenaline Many adrenergic nerves in the muscle layers of the gastrointestinal tract are associated with blood vessels and are probably related to vasomotor action. In the peripheral nervous system, nearly all adrenergic cell bodies are confined to sympathetic ganglia that have their terminals in the myenteric plexus, submucous plexus, mucosa, and associated blood vessels. Adrenergic cell bodies are rarely seen, if at all, in the enteric plexuses (Furness and Costa, 1987). This distribution appears to be similar in many species, such as in the rat intestine, guinea-pig ileum, pig ileum, human ileum and cat ileum (Hoyle and Burnstock, 1989). However, intrinsic adrenergic cell bodies are found in the myenteric plexus of the guinea-pig proximal colon, and these account for 50–60% of the adrenergic fibres in the myenteric plexus; the remaining 40–50% and all the adrenergic fibres in the submucous plexus are extrinsic in origin (Furness and Costa, 1971a,b). Microsurgical interruption of the myenteric plexus shows that there are no ascending or descending adrenergic pathways because the adrenergic fibres are confined to arcades of the gut supplied by blood vessels and neurones entering via the mesentery (Furness, Costa and Llewellyn-Smith, 1981). Noradrenergic transmission is often excitatory in sphincteric muscles. For example, in the cat ileocolonic sphincter, peripheral splanchnic nerve stimulation is excitatory, is mimicked by phenylephrine, and is blocked by phentolamine (Rubin et al., 1980). The cat internal anal sphincter behaves similarly, and electromyograph records show that hypogastric nerve stimulation evokes EJPs that are mediated via α-adrenoceptors (Bouvier and Gonella, 1981). Endothelin Endothelin was originally identified as a 21 amino acid peptide with pressor activity, synthesised by endothelial cells. During identification of genes producing the family of endothelins in mouse intestine (endothelin-1, - 2 and - 3), a gene encoding a peptide closely related to endothelin-2 was identified. This peptide which differs from endothelin-2 by one amino acid residue was originally found only in the intestine, and not in endothelial cells, and although it possessed pressor activity it was a potent spamsogen in guinea-pig ileum, hence its name: vasoactive intestinal constrictor (VIC) (Ishida et al., 1989; Saida, Mitsui and Ishida, 1989; Saida et al., 1996). Immunoreactivity for endothelin-1 is found in

304

INNERVATION OF THE GASTROINTESTINAL TRACT

human colon myenteric and submucous plexuses, where it is co-localised with vasoactive intestinal polypeptide (VIP), and endothelin receptors are mostly neuronal, with smaller populations on smooth muscle cells or epithelial cells (Inagaki et al., 1991; Escrig et al., 1992). Rat myenteric neurones in culture display endothelin-1 immunoreactivty and high levels of the appropriate mRNA (Eaker et al., 1995). Interestingly, myenteric neurones do not contain mRNA for endothelin-2 (de la Monte et al., 1995). Several molecular forms of endothelin exist (Hoyle, 1996a); endothelin-1, endothelin-2, endothelin-3, VIC and the related peptides, sarafotoxin-a, sarafotoxin-b and sarafotoxin-c all evoke concentration-dependent contraction of the ileum (Wiklund et al., 1991; Wollberg et al., 1991; Yoshinaga et al., 1992; Kan, Niwa and Taniyama, 1994). In addition to contracting ileal muscle, endothelin-1 inhibits release of acetylcholine while potentiating contractions evoked by exogenous acetylcholine (Wiklund et al., 1991). In contrast, VIC evokes release of acetylcholine, via non-endothelin receptors, but it also acts on endothelin receptors to evoke contractions (Kan, Niwa and Taniyama, 1994). Endothelins may also cause relaxation of guinea-pig ileal longitudinal muscle, mediated via apamin-sensitive Ca2+-dependent K+-channels (Lin and Lee, 1992). Whether endothelin-1 (or any other endothelin-related neuropeptide) is a neuromuscular transmitter, or is involved in ganglionic transmission or neuromodulation, remains to be established. INHIBITORY NEUROMUSCULAR TRANSMISSION Most examples of inhibitory neuromuscular transmission in the intestine can be accounted for by the troika of ATP, nitric oxide (NO) and VIP. Although there is no direct evidence that these neurochemicals are in fact co-transmitters, it is very likely that they are (see Hoyle, 1996b). In recent years, the discovery of NO as an inhibitory neurotransmitter has commanded a lot of attention, and related evidence has been reviewed many times (for example Rand, 1992; Sanders and Ward, 1992; Lefebvre, 1995; Lincoln, Hoyle and Burnstock, 1995, 1997). In some regions of the gastrointestinal tract, interstitial cells of Cajal are involved in nitrergic inhibitory transmission, acting as intermediary amplifiers between the inhibitory nerves and the smooth muscle (for reviews see Sanders, 1996; Shuttleworth and Sanders, 1996). The current that causes the hyperpolarisation of the IJP in response to nerve stimulation seems to be solely due to potassium efflux. The reversal potential of the non-adrenergic, non-cholinergic IJP has been determined; it lies very close to the potassium equilibrium potential (Hoyle and Burnstock, 1989). Also, in several studies, the dependency of IJP amplitude on membrane potential has been shown to be linear, again suggesting the involvement of only the one ion. Pharmacological investigations have also shown that potassium ion efflux is responsible for IJPs. The polypeptide toxin apamin, which is extracted from bee venom, blocks calcium-dependent potassium ion channels and abolishes IJPs by blocking this ion channel in the guinea-pig taenia coli (Maas et al., 1980). ATP, nitric oxide and vasoactive intestinal peptide The evidence that ATP acts as a neuromuscular transmitter in the enteric nervous system is now quite convincing, and has been reviewed extensively (Hoyle, 1992, 1996b; Bauer,

NEUROEFFECTOR TRANSMISSION IN THE INTESTINE

305

1993; Itoh et al., 1995; Burnstock, 1996, 2001). Although there is little evidence from histochemical studies for the co-localisation of ATP with VIP or nitric oxide synthase (NOS) in inhibitory motoneurones, there are many examples of regions of the gastrointestinal tract in which neuromuscular transmission is mediated by more than one transmitter, and generally inhibitory transmission is carried out by members of this trio (Hoyle, 1996b). Although desensitisation of P2X receptors by α,β-methylene ATP or antagonism by arylazidoaminopropionyl ATP (ANAPP3) has effectively demonstrated that ATP is a neurotransmitter in nerves supplying autonomically innervated structures such as the vas deferens, seminal vesicles, urinary bladder, heart, and some blood vessels (Hoyle, 1992, 1994, 1996b; Sneddon, McLaren and Kennedy, 1996; Burnstock, 1997), neither of these agents successfully antagonises inhibitory P2 receptors in the gut, which are typically subclass P2Y. In the guinea-pig taenia coli, ANAPP3 does antagonise low concentrations of ATP (<100 µM), but not specifically, and has no significant action against inhibitory field stimulation (Westfall et al., 1982). In the myenteric plexus of the rat colon, nearly all the quinacrine-positive nerves contain NADPH-diaphorase activity, but in the ileum and anococcygeus muscle only a subpopulation of quinacrine-positive nerves also contain NADPH-diaphorase activity (Belai and Burnstock, 1994). If the assumption that the neurones that take up quinacrine are purinergic is correct, then it would appear that some nerves probably co-transmit with ATP and NO, since NADPH-diaphorase activity can sometimes be utilised as a marker for NOS (see Lincoln, Hoyle and Burnstock, 1997). Some of the quinacrine-positive neurones may be involved in ganglionic rather than neuromuscular transmission. Although ATP can be released from myenteric neurones during field stimulation or from synaptosomal preparations from the guinea-pig myenteric plexus (see White, 1988; Hoyle and Burnstock, 1989; Hoyle, 1992) it has not been clarified whether the purinergic nerves are motoneurones or interneurones, or both. Fast excitatory postsynaptic potentials mediated via ATP acting on P2X receptors have been recorded in the guinea-pig ileum myenteric plexus (LePard, Messori and Galligan, 1997) (Figure 7.2). The pharmacological profile of ATP receptors in cultured myenteric neurones suggests that they are formed from P2X2, but in situ the pharmacological profile is less clear, and has been described as being more similar to that of P2X4 or P2X6 (Barajas-López et al., 1996). The postsynaptic neurone is possibly cholinergic, since ATP stimulates release of acetylcholine from cholinergic motoneurones via P2X receptors (Matsuo et al., 1997). The muscular receptor in the intestine is P2Y, being G protein-coupled rather than a ligandgated ion channel. In the rat small intestine, according to homology with cloned receptors, it is P2Y1, and is coupled to phosphatidylinositol metabolism (Blottiere, Loirand and Pacaud, 1996; Pacaud et al., 1996). In the guinea-pig small intestine it is P2Y (Matsuo et al., 1997), although the subtype is as yet unidentified. Human small intestinal muscle contains appreciable levels of mRNA for P2Y6, which, when expressed in a cell line, couples to phosphatidyl inositol metabolism (Communi, Parmentier and Boeynaems, 1996). In the circular muscle of the guinea-pig ileum, a single pulse of electrical field stimulation evokes a fast IJP transient hyperpolarisation that has a latency of onset of approximately 100 ms and a rise time of less than 500 ms; this is a fast inhibitory junction potential (IJP). This event is mimicked by local application of ATP and in parallel with the response to ATP it is abolished by apamin (a K+ channel blocker) or Reactive blue 2 or following

306

INNERVATION OF THE GASTROINTESTINAL TRACT

Figure 7.2 Effect of antagonism of P2X receptors on non-cholinergic fast excitatory synaptic potentials in guinea-pig ileum myenteric plexus. Under control conditions, stimulation of the presynaptic nerve evokes a fast EPSP. The nicotinic antagonist, hexamethonium (100 µM), reduces the amplitude of the EPSP by approximately 50%, and addition of the P2 receptor antagonist, PPADS (10 µM) causes its abolition. After recovery from PPADS, in the presence of hexamethonium, desensitisation of P2X receptors by α,β-methylene ATP (α,β-me ATP, 1 µM) also abolishes the EPSP. (Reproduced from LePard, Messori and Galligan, 1997, with permission).

desensitisation to α,β-methylene ATP (Figure 7.3). In the presence of apamin, a short train of pulses of electrical field stimulation evokes a hyperpolarisation that has a relatively long latency of onset, and a rise time approaching 1.5 s. This slow IJP is mimicked by VIP and like the response to applied VIP, it is antagonised by the VIP receptor antagonist VIP(10–28) (Figure 7.4). Furthermore, when NOS is inhibited, the slow IJP is attenuated (Figure 7.5), sometimes unmasking a depolarisation, and not affecting the fast IJP. Thus in this tissue, ATP is the transmitter responsible for the fast IJP and both VIP and NO contribute to the slow IJP (Crist, He and Goyal, 1992; He and Goyal, 1993). It cannot be stated with certainty that these three substances are co-transmitters, but VIP is extensively co-localised with NOS in the nerves supplying the circular muscle (Costa et al., 1992, 1996), so these two are likely to be co-transmitters, but there is no evidence one way or the other as to whether or not ATP is released from this same population. Physiologically, the inhibitory transmission to the circular muscle is brought about via reflexes involved in descending inhibition. These reflexes prepare the bowel to receive the contents, and facilitate transit along the intestinal tract. It is not immediately clear what the advantage is of having a biphasic IJP or of having phases of fast and slow inhibitory transmission.

NEUROEFFECTOR TRANSMISSION IN THE INTESTINE

307

Figure 7.3 Effects of desensitisation to α,β-methylene ATP (100 µM for 30 min), the P2 receptor antagonist, Reactive blue 2 (200 µM), and the vasoactive intestinal polypeptide (VIP) receptor antagonist, VIP(10–28) (1 µM) on non-adrenergic, non-cholinergic inhibitory junction potentials (IJPs) in guinea-pig ileum. Intracellular recordings of smooth muscle membrane potential. IJPs were evoked by electrical field stimulation (single pulses). The upward transient preceding the hyperpolarisation is the stimulus artefact. Resting membrane potential –57 to –47 mV. Note that antagonism of P2 receptors causes inhibition of IJP amplitude, while antagonism of VIP receptors has no significant effect. (Modified from Crist, He and Goyal, 1992, with permission from the Physiological Society.)

The guinea-pig internal anal sphincter is similar to the ileal circular muscle in that it too supports a biphasic IJP. The initial faster phase is mimicked by ATP, blocked by apamin and suramin, as are responses to applied ATP (Lim and Muir, 1986; Baird and Muir, 1990; Rae and Muir, 1996). The slower secondary phase is blocked by NOS antagonists (Rae and Muir, 1996). There is no evidence that VIP is also involved in neurotransmission in the guinea-pig internal anal sphincter (Lim and Muir, 1986). Nitric oxide Neurones that can synthesise NO are well represented in the enteric nervous system, throughout the gastrointestinal tract, and the techniques used to visualise them have been reviewed recently (Lincoln, Hoyle and Burnstock, 1997). In dog duodenum and ileum, neurogenic relaxations, evoked by electrical stimulation or agents such as ATP, GABA or 5-HT, are blocked by oxyhaemoglobin or NOS inhibitors (Toda, Baba and Okamura, 1990; Boeckxstaens et al., 1991; Bogers et al., 1991; Toda, Tanobe and Baba, 1991; Toda et al., 1992). Similarly, in the rat duodenum longitudinal muscle, NANC inhibition evoked by nicotine is blocked by L-NNA (Irie et al., 1991). In the human ileum circular muscle, relaxant responses of NANC inhibitory transmission can be inhibited by up to 65% by L-NNA (Maggi et al., 1991), implying that another transmitter is responsible for the residual 35%. Nitrergic pathways may provide a tonic inhibitory influence over excitation in the small intestine. Inhibition of NOS in cat, rat and guinea-pig

308

INNERVATION OF THE GASTROINTESTINAL TRACT

Figure 7.4 Effects of desensitisation to α,β-methylene ATP (100 µM for 30 min), the P2 receptor antagonist, Reactive blue 2 (200 µM), and the vasoactive intestinal polypeptide (VIP) receptor antagonist, VIP(10–28) (1 µM) on non-adrenergic, non-cholinergic inhibitory junction potentials (IJPs) in guinea-pig ileum in the presence of apamin (1 µM). Intracellular recordings of smooth muscle membrane potential. IJPs were evoked by electrical field stimulation (four pulses, 20 Hz). Apamin blocks the fast IJP and reveals a relatively low-amplitude, slow IJP. The upward transients preceding the hyperpolarisations are stimulus artefacts. Resting membrane potential –47 to –36 mV. Note that antagonism of VIP receptors causes inhibition of IJP amplitude, while antagonism of P2 receptors has no significant effect. Note also that in the presence of VIP(10–28) there is a long-latency, slow depolarisation that is due to substance P. (Modified from Crist, He and Goyal, 1992, with permission).

Figure 7.5 Examples of the effect of the nitric oxide synthase inhibitor, NG-nitro-L-arginine (L-NNA, 200 µM) on the slow inhibitory junction potential produced in the presence of apamin and substance P desensitisation, in the guinea-pig ileum. (A) A single pulse of electrical field stimulation (EFS) evokes a slow IJP followed by a slow depolarisation that are both abolished by L-NNA. (B) A short train of EFS (four pulses, 20 Hz) evokes a slow IJP and slow depolarisation that are both attenuated, but not abolished, by L-NNA. (C) In a different cell, the IJP evoked by EFS (four pulses, 20 Hz) is abolished by L-NNA, and a short latency depolarisation is unmasked. Intracellular recordings of smooth muscle membrane potential. The upward transients preceding the hyperpolarisations are stimulus artefacts. Resting membrane potential –46 to –45 mV. (Modified from He and Goyal, 1993, with permission from the Physiological Society).

NEUROEFFECTOR TRANSMISSION IN THE INTESTINE

309

increases intestinal compliance, or causes increases in intraluminal pressure and contractile activity (Maggi et al., 1991; Calignano et al., 1992; Gustafsson and Delbro, 1993; Waterman and Costa, 1994; Waterman, Costa and Tonini, 1994), potentiating ascending excitation (Allescher et al., 1992). Neuromodulation by NO is a pronounced phenomenon in the small intestine. For example, in guinea-pig small intestine, exogenous NO decreases the nerve-stimulation-evoked release of both substance P and acetylcholine (Gustafsson et al., 1990; Wiklund et al., 1993a). The effect of inhibition of NOS on substance P-mediated neuromuscular transmission in the guinea-pig ileum is quite dramatic (Figure 7.6). In dog duodenal longitudinal muscle, inhibition of NOS results in potentiation of cholinergic transmission (Toda, Tanobe and Baba, 1991). Almost paradoxically, in the guinea-pig ileum NO can also have excitatory actions, stimulating the release of both acetylcholine and substance P from intramural neurones (Gustafsson et al., 1990; Wiklund et al., 1993a). Excitatory effects of NO have been found to occur in rat small intestine longitudinal muscle as well. NANC nerve stimulation evokes a contractile response, as does sodium nitroprusside, while L-NNA abolishes the NANC contraction, and L-arginine reverses this effect (Barthó et al., 1992). Thus, it would appear that in this organ, under the prevalent conditions, nitrergic transmission is excitatory. Sequential labelling with the NADPH diaphorase technique for labelling NOS in the canine proximal colon showed that 94% of neurones that responded to exogenous NO with an increase in cGMP-like immunoreactivity were NADPH diaphorase-negative. None of the myenteric neurones that responded to electrical field stimulation with an increase in cGMP-like immunoreactivity was NADPH diaphorase-positive, and only one submucosal neurone with cGMP-like immunoreactivity was also NADPH diaphorase-positive (Shuttleworth et al., 1993). The electrical field-stimulated increase in cGMP-like immunoreactivity was blocked by nitroarginine (100 µM). Here NANC-inhibitory neuromuscular transmission has all the characteristics of being mediated predominantly via NO, being antagonised by oxyhaemoglobin, L-NAME, L-NNA, mimicked by exogenous NO or S-nitrosocysteine, and being potentiated by the cGMP phosphodiesterase inhibitor M&B22948 (also known as zaprinast) (Dalziel et al., 1991; Thornbury et al., 1991; Ward et al., 1992a,b; Shuttleworth, Sanders and Keefe, 1993; Shuttleworth et al., 1993). The fast IJP in the canine colon is abolished by inhibition of NOS (Ward, McKeen and Sanders, 1992c). In addition, rebound excitation seen after NANC inhibition in the colon is also attenuated by inhibition of NOS, and by oxyhaemoglobin, and is mimicked by NO (Ward et al., 1992b). The same is true in the cat colon (Venkova and Krier, 1994). In the human colon, although L-NNA can reduce the amplitude of responses to NANC inhibitory transmission (Burleigh, 1992; Tam and Hillier, 1992; Boeckxstaens et al., 1993), the fast IJP is unaffected by L-NNA or L-NAME nor is it affected by oxyhaemoglobin. However, inhibition of NOS causes a reduction of the sustained hyperpolarisation that follows the rapid IJP when trains of pulses are applied (Keef et al., 1993). VIP is colocalised extensively with NOS in the human colon (Keranen et al., 1995), but its role as an inhibitory transmitter remains to be substantiated. Certainly it does not mediate the IJP (Hoyle et al., 1990). In the guinea-pig large intestine, where VIP may be co-localised with NOS (Costa et al., 1992; Berezin et al., 1994), little or no interaction between VIP and NO has been found.

310

INNERVATION OF THE GASTROINTESTINAL TRACT

Figure 7.6 Effects of inhibition of nitric oxide synthase on excitatory neuromuscular transmission in guineapig ileum longitudinal muscle. Contractions were evoked by electrical field stimulation (3 Hz, 0.2 ms, 180 pulses at 5 min intervals). (A) NG-nitro-L-arginine (L-NOARG, 20 µM) enhances the tonic “hump” phase of the nerve-mediated contraction with little or no effect on the initial twitch. (B) Atropine (1 µM) abolishes the initial twitch, and the residual component is enhanced by L-NOARG (20 µM). (C) The enhancement of the non-cholinergic contraction by L-NOARG (20 µM) is reversed by L-arginine (1 mM), and the substance P receptor antagonist, spantide (30 µM) abolishes the non-cholinergic contractions. (Reproduced from Wiklund et al., 1993a, with permission from the Nature Publishing Group).

NEUROEFFECTOR TRANSMISSION IN THE INTESTINE

311

Unlike gastric smooth muscle, the taenia coli does not generate NO in response to VIP (Grider et al., 1992; Jin et al., 1993), although NO is produced by EFS both here and in intertaenial longitudinal muscle (Shuttleworth, Murphy and Furness, 1991; Grider et al., 1992; Jin et al., 1993; Wiklund et al., 1993b). Several effects of NO are only manifest in the guinea-pig taenia coli when the cholinergic system is intact, i.e. in the absence of atropine (Shuttleworth, Murphy and Furness, 1991; Grider et al., 1992; Jin et al., 1993; Wiklund et al., 1993b; Ward et al., 1996). Under these conditions a neurogenic relaxation and hyperpolarisation can be recorded that is sensitive to haemoglobin and L-NNA. This situation is curious because the actions of an inhibitory transmitter, NO, seem to be dependent upon an excitatory transmitter, acetylcholine. The internal anal sphincter has been extensively investigated, particularly in the opossum. Like other gastrointestinal sphincters (lower oesophageal sphincter, pylorus, choledochal duodenal sphincter or sphincter of Oddi, ileocaecal sphincter or ileocolonic junction) it receives an inhibitory nitrergic innervation. In isolated preparations, relaxant responses to electrical field stimulation (EFS) are abolished by L-NNA, the effects of which may be reversed by L-arginine but not D-arginine (Chakder and Rattan, 1992, 1993a). Release of NO as a result of EFS can be determined using a chemiluminescence assay (Chakder and Rattan, 1993b), and responses to EFS can be inhibited by hydroquinone (Chakder and Rattan, 1992) and recombinant human haemoglobin (Rattan, Rosenthal and Chakder, 1995). In vivo, rectal distension evokes relaxation of the internal anal sphincter via a descending inhibition, or rectoanal reflex. This reflex is blocked by L-NNA, which also blocks relaxations evoked by sacral nerve stimulation (Rattan, Sarkar and Chakder, 1992). When the hypogastric nerve is stimulated, the opossum internal anal sphincter responds with an initial relaxation followed by a phase of contraction; both elements are attenuated by L-NNA (Rattan and Thatikunta, 1993). It is also likely that inhibitory transmission involves VIP since it co-exists with NOS (Lynn et al., 1995). Both NO and VIP, as well as EFS, induce elevation of cGMP and cAMP levels in the smooth muscle (Chakder and Rattan, 1993a). Relaxant responses to VIP may be severely inhibited by L-NNA or hydroquinone (Chakder and Rattan, 1992), implying that its responses involve NO. However, others have found no significant effect of L-NNA on responses to VIP (Tøttrup, Glavind and Svane, 1992). The human internal anal sphincter is innervated by NOS-containing nerves, and neurogenic relaxations in isolated preparations are abolished by L-NNA or oxyhaemoglobin (Burleigh, 1992; O’Kelly, Brading and Mortensen, 1993). In vivo, topical application of glyceryl trinitrate (GTN, nitroglycerin) to the anus causes a reduction in anal pressure (Loder et al., 1994). This observation led to the suggestion that GTN can be used to treat conditions such as anal fissures, haemorrhoids or proctalgia, which benefit from a reduced anal pressure (Loder et al., 1994; Gorfine, 1995). Nitric oxide and interstitial cells of Cajal In canine proximal colon, interstitial cells of Cajal are located on the submucosal surface of the circular smooth layer, and on the surface of smooth fasciculi lying adjacent to the connective tissue septa. Ablation of the interstitial cells prevents generation of slow waves within the fasciculus in regions close to the area of ablation (Ward and Sanders, 1990).

312

INNERVATION OF THE GASTROINTESTINAL TRACT

Electrically, interstitial cells of Cajal are spontaneously active, with phasic depolarisations similar to the slow-wave of colonic smooth muscle, driven in part by cyclic activity of Ca2+ ion channels and Ca2+-activated K+ channels (Langton et al., 1989). The electrical activity of smooth muscle cells is governed by interstitial cells. When isolated preparations of canine proximal colon are incubated with rhodamine123, the activity of the smooth muscle cells alters. Rhodamine123 is selectively taken up by the interstitial cells, and is cytotoxic, thus the activity of the smooth muscle cells alters as the rhodamine123 poisons the influential interstitial cells (Ward, Burke and Sanders, 1990). In canine proximal colon in vitro, electrical field stimulation and application of NO evoke increases in cGMP-like immunoreactivity in interstitial cells located at the submucosal surface of the circular muscle layer, which indicates that the interstitial cells are targets of enteric neurones that utilise NO as a transmitter (Shuttleworth et al., 1993). In culture, application of nitric oxide causes elevation of intracellular Ca2+ in interstitial cells of Cajal (due to release from intracellular stores), and a decrease in smooth muscle cells (Publicover, Hammond and Sanders, 1993). The interstitial cells release NO that then acts on the smooth muscle cells, presumably the nitric oxide-induced elevation in intracellular calcium is the stimulus for NOS activation within the interstitial cell. Thus it appears that the interstitial cells, interposed between some neurones and some smooth muscle cells, amplify the inhibitory nitrergic transmission in the enteric nervous system (Publicover, Hammond and Sanders, 1993). The spontaneous pattern of electrical and mechanical rhythmic activity of colonic smooth muscle is tonically suppressed by the release of NO from interstitial cells of Cajal (Keef et al., 1997): the potential for interstitial cells to synthesise NO is demonstrated by their expression of constitutive NOS-immunoreactivity (Xue et al., 1994). Mice with mutations in the receptor tyrosine kinase c-kit, W/W(V) mutants, have markedly fewer interstitial cells of Cajal in the region of the myenteric plexus than do the wild-type mice. They also lack electrical slow wave activity in their smooth muscle cells, indicating that interstitial cells play a dominant role in the generation of rhythmic activity in intestinal muscles (Ward et al., 1994). Similarly, mutant mice lacking steel factor, which is the endogenous ligand for c-kit, also lack myenteric interstitial cells of Cajal and smooth muscle slow wave activity (Ward et al., 1995). Furthermore, in W/W(V) mutants, nitrergic inhibition of gastric smooth muscle is impaired (Burns et al., 1996). Pituitary adenylate cyclase-activating peptide Pituitary adenylate cyclase-activating peptide (PACAP) exists in two forms, PACAP-38 and PACAP-27 which is PACAP-38(1–27). This neuropeptide is astonishingly well conserved amongst diverse animal species and a gene that encodes PACAP-27 has been identified in the urochordate Chelyosoma productum. This molecular form has only one amino acid residue that differs from mammalian PACAP-27, representing approximately 96% homology. Furthermore, PACAP is closely related to VIP with PACAP-27 having approximately 68% sequence homology with VIP (Hoyle, 1998). Within the intestine, the majority of PACAP belongs to the intrinsic innervation, and is found in nerve fibres in all layers of the human intestine, and in cell bodies of neurones that are more numerous in the submucous plexus than in the myenteric plexus (Sundler et al., 1992).

NEUROEFFECTOR TRANSMISSION IN THE INTESTINE

313

Functionally, PACAP relaxes intestinal smooth muscle, inhibiting spontaneous activity and causing concentration-dependent relaxation of carbachol-contracted preparations of sigmoid colon (Schworer et al., 1993). In many cases, VIP and PACAP appear to act on the same receptor, but in the muscle coat of the intestine, PACAP acts independently of VIP receptors, probably via PACAP type I receptors that are selective for PACAP and are activated by relatively high concentrations of VIP (Mao et al., 1998). For example, in the guinea-pig taenia coli, relaxations evoked by PACAP are blocked by apamin, while those evoked by VIP are not. Further, responses to PACAP but not VIP, are inhibited by the PACAP antagonist PACAP(6–38), or by desensitisation to PACAP (Jin et al., 1994). It has been suggested that PACAP may be responsible for part of the non-adrenergic, noncholinergic inhibitory transmission in this tissue, since evoked release of PACAP is frequency-dependent and the treatments that inhibit PACAP activity also inhibit responses to nerve stimulation (Jin et al., 1994). At an electrophysiological level, PACAP hyperpolarises the smooth muscle membrane of the guinea-pig taenia coli. This hyperpolarisation is abolished by apamin, and reduced by the P2 receptor antagonist, suramin; following desensitisation to PACAP, the IJP is reduced (McConalogue, Lyster and Furness, 1995). However, in colonic circular muscle, apamin and suramin inhibit the IJP and simultaneously inhibit hyperpolarisation evoked by PACAP or α,β-methylene ATP, but pyridoxalphosphate-6azophenyl-2′,4′-disulphonic acid (PPADS) inhibits the IJP and responses to α,β-methylene ATP without affecting responses to PACAP (Zagorodnyuk et al., 1996). Thus, the role of PACAP in the fast IJP is not entirely clear. Descending inhibition, evoked in the large intestine, is accompanied by release of PACAP. Application of neutralising antibodies raised against PACAP-27 and PACAP-38, or application of PACAP(6–38) attenuate the inhibition. VIP(10–28) and VIP-antibodies also inhibit the relaxation, but the descending inhibition is abolished when they are applied in combination with the anti-PACAP treatments (Grider et al., 1994). Nitric oxide appears to be involved in the release of PACAP, since inhibition of NOS by NG-nitro-Larginine reduces PACAP-release by approximately 50% (Grider et al., 1994). Noradrenaline In most regions of the gut, adrenergic nerve stimulation or exogenously applied noradrenaline results in an inhibitory action, with decreases in mechanical tone or spontaneous activity, hyperpolarisation of the smooth muscle membrane, and inhibition of firing of action potentials. In many isolated smooth muscle preparations such as guinea-pig stomach, guinea-pig taenia coli (Bennett, Burnstock and Holman, 1966; Ito and Kuriyama, 1975), or human taenia coli (Stockley and Bennett, 1977), the adrenergic contribution in response to field stimulation is apparent only at higher frequencies of stimulation, usually >5 Hz, whereas non-adrenergic inhibitory responses are evokable by single pulses of electrical stimulation. In non-sphincteric muscles, where adrenergic nerve fibres are concentrated in the myenteric and submucous plexuses, it has been suggested that the relaxation of the smooth muscle may not be due to a direct action of noradrenaline released from sympathetic nerves but instead is due to an inhibition of a tonic excitatory pathway, most likely cholinergic, and as such represents a neuromodulation rather than neuromuscular transmission. Noradrenaline may act directly on the smooth muscle, especially when the adrenergic

314

INNERVATION OF THE GASTROINTESTINAL TRACT

neurones are stimulated at high frequency and the neurotransmitter can overspill to act on muscle receptors (Gillespie and Maxwell, 1971). Exogenous noradrenaline is a potent inhibitor of acetylcholine release and neurotransmission in the myenteric plexus, and because neither exogenous noradrenaline nor sympathetic nerve stimulation affects the membrane potential of ganglion cell bodies in the myenteric plexus, the site of adrenergic action is probably remote from the cell body, i.e. on the axon terminals. Reciprocal axoaxonic synapses between adrenergic and cholinergic nerves in the enteric plexus have been observed, and acetylcholine and noradrenaline are able to inhibit their own and each other’s release (see Hoyle and Burnstock, 1989). In addition to modulating release of acetylcholine, noradrenaline can inhibit release of other neurotransmitters. In the guinea-pig taenia coli, noradrenaline reduces the relaxant effect of non-adrenergic inhibitory nerve stimulation (Sakato, Shimo and Bando, 1972); in the guinea-pig duodenum (Ohkawa, 1983) and caecum circular muscle (Reilly, Hoyle and Burnstock, 1987), and rat caecum circular muscle (Hoyle et al., 1988) the non-adrenergic IJP in response to transmural stimulation is reduced in amplitude by adrenoceptor agonists and is enhanced by guanethidine (Ohkawa, 1983), implying the existence of a tonic neuronal adrenergic modulation of the IJP. In the rabbit colon, where preganglionic vagal stimulation can evoke EJPs or IJPs, simultaneous splanchnic nerve stimulation blocks the vagal excitatory, but not inhibitory, effect; pelvic nerve stimulation excitation is very effectively blocked by lumbar colonic nerve stimulation (Gillespie and Khoyi, 1977). In this case the presynaptic inhibition is mediated via α-adrenoceptors, whereas direct inhibition observed at high stimulation frequencies is mediated via β-adrenoceptors.

NEUROEPITHELIAL TRANSMISSION IN THE INTESTINE Ion transport across the intestinal epithelium plays a role of paramount importance in body fluid homeostasis. The flux of ions, principally chloride, bicarbonate, sodium and potassium, through intestinal epithelial cells generates osmotic gradients across the epithelium which govern the rate and direction of passage of water to and from gut lumen and vascular system. They are innervated on their basal side, principally by nerves that have their cell bodies in the submucous plexus, but also by nerves originating in the myenteric plexus or in sympathetic ganglia, and by extrinsic sensory neurones. The epithelial cells have tonic activity but are under humoral and neural control. Secretomotor neurones from the submucous plexus transmit with acetylcholine and VIP, and NPY appears to be the main antisecretory transmitter. In general, the enterocytes on villi in the small intestine, and lining the lumen of the large intestine are absorptive, one of their primary features being that they transport chloride ions from the gut lumen across their apical membrane via a chloride-bicarbonate exchange. The epithelial cells that line the crypts of the small and large intestine are secretory, and a major feature is that these cells lack a chloridebicarbonate exchanger but have chloride ion channels in their apical membrane, which provide a route for chloride ions to flow from inside the cell, across the apical membrane, into the gut lumen.

NEUROEFFECTOR TRANSMISSION IN THE INTESTINE

315

SECRETORY TRANSMITTERS Acetylcholine Choline acetyltransferase, the enzyme that catalyses production of acetylcholine from acetyl-CoA and choline, is found in myenteric and submucosal neurones projecting to the epithelium. These neurones may also contain cholecystokinin (CCK), calcitonin gene-related peptide (CGRP), neuropeptide Y (NPY) and somatostatin, and possibly galanin (Keast, Furness and Costa, 1982; Bornstein, Furness and Costa, 1987; Furness et al., 1987; Bornstein and Furness, 1988; Cooke, 1989; Porter et al., 1996). Thus it is likely that acetylcholine is a co-transmitter with these other substances. Choline acetyltransferase is also found in submucosal interneurones, in a population that does not appear to contain any neuropeptide. These nerves synapse in submucosal ganglia with secretomotor neurones (Bornstein, Furness and Costa, 1987). A third population of nerve fibres, which contains choline acetyltransferase, is sensory, projecting to the epithelium, and additionally contain substance P (Bornstein, Furness and Costa, 1987; Cooke, 1989). Stimulation of intramural neurones in the intestine evokes secretion that is partially mediated by acetylcholine (Wu, Kisslinger and Gaginella, 1982; Carey et al., 1987; Kuwahara et al., 1987a; Carey and Cooke, 1989; Chandan et al., 1991a; Javed and Cooke, 1992; Hogan et al., 1993). Acetylcholine promotes secretion from epithelial cells by acting on muscarinic receptors, however it has been argued that it would be incorrect to regard the cholinergic nerves as “secretomotor” because some of the neuropeptides that they contain, particularly NPY and somatostatin, would evoke absorption. In intestinal mucosa, the primary site for secretion appears to be in the crypts (Browning et al., 1978). There is probably not a homogeneous population of muscarinic receptors on the epithelial cell because the pharmacological profile of muscarinic receptor antagonists is not consistent with a single known subtype. The subtypes that are present appear to be M1 and M3, with M3 being preponderant (Carey et al., 1987; Kuwahara et al., 1987b; Chandan et al., 1991a,b; O’Malley et al., 1995). The main effect of acetylcholine is to inhibit sodium and chloride absorption from the lumen, across the apical membrane, to stimulate chloride secretion across the basolateral membrane and, in some tissues, to stimulate bicarbonate transport across the apical membrane (Cooke, 1989; Jodal, 1990; Chandan, O’Grady and Brown, 1991; Hogan et al., 1993). In porcine small intestine, muscarinic receptors involved in mucosal transport are found not only on epithelial cells but also on neurones. Activation of the epithelial receptor promotes chloride secretion, while activation of the neuronal receptor inhibits chloride secretion, presumably by stimulating release of an anti-secretory transmitter or by inhibiting tonic release of a pro-secretory transmitter (Chandan et al., 1991b). Vasoactive intestinal peptide and related peptides Vasoactive intestinal polypeptide (VIP) and peptide histidine isoleucine (PHI) are colocalised with dynorphin and galanin in guinea-pig ileum submucosal neurones (Furness et al., 1987), and in the rat VIP is co-localised in submucosal neurones with NPY (Cox, Rudolph and Gschmeissner, 1994). These neurones project to the mucosa, forming an aganglionic villar plexus. They receive excitatory and inhibitory input from myenteric neurones, excitatory input from submucosal neurones, and inhibitory input from sympathetic

316

INNERVATION OF THE GASTROINTESTINAL TRACT

Figure 7.7 Effects of vasoactive intestinal polypeptide (38 nM) on baseline and neurally evoked secretory responses in guinea-pig ileum. (A) Short-circuit current response to stimulation of enteric neurones (a, fast cholinergic phase; b, sustained phase). (B) Effect of VIP on baseline short-circuit current and on neurally evoked response. Stimulus parameters were: 0.5 ms, 10 Hz, 10 V (3.3 mA), applied as indicated by bar. Note that VIP increases the short-circuit current and enhances the response to nerve stimulation, especially the initial cholinergic phase. (Reproduced from Cooke et al., 1987, with permission from Harcourt).

neurones (Bornstein, Furness and Costa, 1987; Bornstein and Furness, 1988). Release of VIP from guinea-pig ileal and colonic mucosa has been measured, and is evoked by nicotinic stimulation or depolarisation by potassium chloride (Okuno et al., 1988; Reddix et al., 1994). VIP is a powerful secretagogue, with greater potency than its evolutionary homologues (see Hoyle, 1998), PHI, helodermin and PACAP-28 (Cox and Cuthbert, 1989b; Reddix et al., 1994). The mechanism of action of VIP involves an increase in intraepithelial cell production of cAMP (Laburthe et al., 1979; Beubler, 1980; Broyart et al., 1981; Prieto et al., 1981; Reimer et al., 1996), and subsequent activation of a cAMP-dependent protein kinase (Cohn, 1987). The relationship between intracellular levels of cAMP and stimulation of secretion by VIP is not entirely clear, and secretion may be evoked by concentrations of VIP that do not induce measurable changes in cAMP (Beubler, 1980), and increasing concentrations of VIP can evoke increasing rates of secretion even though intracellular levels of cAMP have been increased maximally (Laburthe et al., 1979; Broyart et al., 1981). Further, the increase in cAMP by VIP is thought to contribute to potentiation of cholinergicallyevoked secretion, as shown in Figure 7.7. The agonist potency of VIP and its evolutionary homologues, at evoking cAMP production, correlate well with the ability of the homologues to displace epithelial binding of radiolabelled VIP (Salomon et al., 1993) as well as their ability to evoke secretion (Cox and Cuthbert, 1989b; Reddix et al., 1994). The predominant site of activity of VIP is likely to be in crypt cells rather than villar cells, because, in response to VIP, enterocytes lining the crypt produce more cAMP than enterocytes on the villi , although the potency of VIP is the same in the two regions (Amelsberg et al., 1996). Homologues of VIP, for example PHI and the reptilian helodermin, also promote secretion in the small intestine. In the rat ileum, these substances cross-desensitise one another, which implies that they act on the same types of receptor (Cox and Cuthbert, 1989b). The actual receptor subtype remains to be elucidated, but it appears to be resistant to

NEUROEFFECTOR TRANSMISSION IN THE INTESTINE

317

antagonists formed from VIP fragments such as VIP(10–28), [Lys1,Pro2,5,Arg3,4,Tyr6]VIP or [4Cl-D-Phe6,Leu17]VIP, or those derived from growth hormone releasing factor (GRF), which is an evolutionary relative of VIP (Hoyle, 1998), such as [AcTyr1,D-Phe2] GRF(1–29)-NH2 and [AcTyr1]hGRF-(1–40)-OH (Cox and Cuthbert, 1989b; Burleigh and Kirkham, 1993). In Ussing chamber experiments, application of PACAP to the basolateral surface of human jejunal and colonic mucosa stimulates chloride secretion (Fuchs et al., 1996). Both PACAP-27 and PACAP-38 are potent secretogogues in rat ileal mucosa (Cox, 1992). PACAP-27, but not PACAP-38 or VIP, at low, subnanomolar, concentrations causes secretion that is sensitive to TTX, desensitisation to substance P, and also capsaicinisation, and is likely to be mediated via stimulation of sensory nerve terminals. At higher, nanomolar, concentrations both PACAP-27 and PACAP-38 induce secretion, which like VIP, is insensitive to TTX (Cox, 1992). In guinea-pig colon, the direct effects of VIP and responses to non-cholinergic nerve stimulation are attenuated by the nonadecapeptide VIP fragment VIP(10–28) (Reddix et al., 1994), providing further evidence that VIP is a neurotransmitter in this tissue. VIP(10–28) may not always be a useful tool, because in rat colon, high concentrations of VIP(10–28) (1 and 3 µM) evoke low amplitude short circuit currents, but do not affect responses to applied VIP (Burleigh and Kirkham, 1993). Chymotrypsin, an enzyme that degrades VIP, and anti-VIP antisera have been used with extremely limited success to demonstrate a transmitter role of VIP in neuroepithelial transmission (Hubel, 1984; Cooke et al., 1987). Binding studies have shown that in the human small intestine epithelium, PACAP and VIP bind the same receptor, which has a higher concentration in villar cells than in crypt cells (Salomon et al., 1993), and is probably a PACAP-VIP type II receptor. However, in rodents it is likely that there is more than one type of receptor present in the mucosa. For example, in rat small intestine, low concentrations of PACAP-27 evoke transient secretory responses that are sensitive to TTX (Figure 7.8), desensitisation by substance P or pre-treatment with capsaicin, indicating an action mediated via stimulation of sensory nerve terminals (Cox, 1992). This action of PACAP-27 is not shared by VIP or PACAP-38, and is indicative of a separate receptor. In guinea-pig large intestine VIP, PACAP-27, PACAP-38, PHI and helodermin evoke responses that are inhibited, but not abolished by atropine. Additionally, TTX abolishes responses to PACAP-27 and PACAP-38, but only attenuates responses to VIP, PHI and helodermin, and then only responses evoked by high concentrations (Kuwahara et al., 1993). These results also indicate the presence of at least two receptors, one on nerve terminals and one on epithelial cells, with PACAP recognising only the neural receptor. Substance P Substance P is localised in neurones that have their cell bodies in the myenteric plexus in guinea-pig small intestine, and which do not receive a synaptic input (Bornstein, Furness and Costa, 1987). These neurones also contain choline acetyltransferase (Bornstein and Furness, 1988; Cooke, 1989). It is unclear whether there is a true physiological role for the action of substance P released from these nerves acting on the epithelial cells. In small and large intestine, across a range of species that includes rat, mouse, guinea-pig and dog,

318

INNERVATION OF THE GASTROINTESTINAL TRACT

Figure 7.8 Effects of pituitary adenylate cyclase-activating polypeptide (PACAP-27) on short-circuit current in rat small intestinal mucosa. PACAP-27 was applied cumulatively (↑) in the absence (upper trace) and presence (lower trace) of tetrodotoxin (TTX, 100 nM, ▲). Vasoactive intestinal polypeptide (VIP, 100 nM, ∆) was added at the end of the traces in order to define the maximum secretory response. The inset shows the response to a single, 1 nM, concentration of PACAP-27. Baseline currents are indicated at the start of the recordings. All doses were applied to the basolateral side of the epithelium. (Modified from Cox, 1992, with permission from the Nature Publishing Group).

responses to substance P are qualitatively similar in that substance P acts partly directly on the epithelial cells and partly indirectly on mucosal nerve terminals or submucosal neurones (Keast, Furness and Costa, 1985a; Fox et al., 1986; Perdue, Galbraith and Davison, 1987; Rangachari, Prior and McWade, 1990; Burleigh and Yull, 1992; Parsons et al., 1992; Wang et al., 1995). When cholinergic transmission is inhibited by atropine in guinea-pig small intestine, the residual non-cholinergic secretory response is markedly attenuated by either desensitisation to substance P or by anti-substance P antisera (Perdue, Galbraith and Davison, 1987). However, if the substance P is present only in sensory nerves, then under these experimental conditions, the substance P would be released from peripheral terminals due to antidromic stimulation of these nerves. This situation could occur under physiological conditions because it has been shown that lightly stroking guinea-pig colonic mucosa induces chloride secretion that is inhibited by the selective NK1 receptor antagonist GR-82334 (Cooke et al., 1997). In rat colon, it is unlikely that substance P is involved in neuroepithelial transmission because it has been shown that application of a neurokinin receptor antagonist, while blocking secretory responses to applied substance P, does not block responses to non-cholinergic nerve stimulation (Burleigh and Yull, 1992). Likewise, in guinea-pig ileum,

NEUROEFFECTOR TRANSMISSION IN THE INTESTINE

319

NK1 receptor antagonists that block responses to applied substance P do not significantly affect neurogenic secretion (Reddix and Cooke, 1992). The receptor activated by substance P may be species-dependent. In guinea-pig small intestine two populations of neurokinin receptors have been identified, one on cholinergic and non-cholinergic nerve terminals in the mucosa, and one on the epithelial cell (Keast, Furness and Costa, 1985a; Perdue, Galbraith and Davison, 1987). In contrast, another group has found that the response to substance P is mediated via NK1 receptors, with no evidence for NK2 or NK3 receptors, in guinea-pig small intestine (Reddix and Cooke, 1992). In these experiments, responses to substance P were markedly attenuated by TTX, but not atropine, implying that the receptor is on non-cholinergic neurones, an action that had been observed previously (Kachur et al., 1982). However, the mRNA for NK1 receptors has been located in isolated colonic epithelial cells and in colonic crypts (Cooke et al., 1997), indicating that epithelial cells of the large intestine probably express NK1 receptors. In rat colon, it is likely that NK1, NK2 and NK3 receptors are present. When the activity of a series of tachykinins was examined (Cox et al., 1993), including substance P and its evolutionary homologues, neurokinin A, neurokinin B and neuropeptide Y (Hoyle, 1998), all were found to evoke secretion. The responses to some tachykinins were abolished by TTX, while some were reduced and some were unaffected. The use of selective agonists and antagonists led to the conclusion that NK1 receptors are both neuronal and epithelial, NK2 receptors are predominantly epithelial with a comparatively small functional neuronal population (on non-cholinergic nerves), and NK3 receptors are neuronal (Cox et al., 1993). Responses to substance P may involve an interaction with mast cells. In mouse small intestine, substance P evokes increases in short circuit current (chloride secretion) via NK1 receptors on enteric nerves and epithelial cells (Wang et al., 1995). H1 and H2 receptor antagonists reduce the secretory responses to non-cholinergic nerve stimulation and applied substance P. In the genetically mast cell-deficient WBB6F1 W/W(V) mouse, responses to substance P are weaker than in control mice, and are unaffected by H1 and H2 antagonists. Thus, in control mice, responses to substance P are augmented by histamine released from mucosal mast cells (Wang et al., 1995). In some pathological conditions, substance P pathways may become involved. For example, in rats infected with an intestinal parasite, the helminth Nippostrongylus brasiliensis, there is a marked increase in the density of innervation of the mucosa by substance P-containing nerves, in rat small intestine (Masson et al., 1996). Furthermore, responses to substance P become reduced to approximately 25% of their control value, at a time when cholinergic neuroepithelial transmission is abolished (Masson et al., 1996). ATP Exogenous ATP stimulates active ion transport across intestinal epithelial mucosa (Kohn, Newey and Smyth, 1970; Korman et al., 1982; Cuthbert and Hickman, 1985; Richards et al., 1987). The ability of ATP to act from the serosal surface of the intestinal epithelium to increase short circuit current, raises the possibility that release of ATP from enteric nerves may be a mechanism for causing intestinal secretion (Korman et al., 1982). In isolated crypts of rat distal colon, basolaterally applied ATP increases [Ca2+]i and acts as a secretagogue via a P2Y receptor, probably P2Y1. Furthermore, P2Y receptor agonist-induced

320

INNERVATION OF THE GASTROINTESTINAL TRACT

[Ca2+]i elevations are most marked at the crypt base, which is the secretory part of the colonic crypt. These responses are not mediated by adenosine (Leipziger et al., 1997). There is at present no conclusive evidence that ATP is involved in neuroepithelial transmission and in addition to a potential neural source, ATP released from neighbouring cells such as epithelial cells and fibroblasts may be available to mucosal sites. Purine receptors, notably UTP-sensitive receptors including P2Y2 (P2U), are also localised to the apical domain of secretory cells (Inoue et al., 1997; Leipziger et al., 1997). ATP may act via degradation to adenosine and subsequently elicit chloride secretion by occupation of adenosine receptors localised on epithelial cells (Dho, Stewart and Foskett, 1992; Stutts et al., 1995). An indirect action of ATP has been proposed following observations of partial TTX-inhibition of exogenous ATP-induced chloride secretion in rat colonic epithelium: this suggests that ATP acts predominantly on neuronal elements in the lamina propria (Cuthbert and Hickman, 1985).

ANTI-SECRETORY TRANSMITTERS Noradrenaline In vivo, intra-arterial injection of noradrenaline promotes water absorption from the small intestine. This effect is mediated via α2-adrenoceptors, since it is mimicked by UK 14,304 but not phenylephrine, and is blocked by yohimbine (Liu and Coupar, 1997). More specifically, the receptor in rat intestine is likely to be the α2A or α2D subtype, as evidenced by the antagonistic effect of BRL 44408 (Liu and Coupar, 1997). There is tonic sympathetic activity, shown by α2-adrenoceptor antagonists decreasing basal rates of absorption, and chemical sympathectomy with 6-hydroxydopamine resulting in inhibited absorption. In vitro, xylazine, which is another α2-agonist, induces a yohimbine-sensitive reduction in short-circuit current in voltage-clamped small intestinal mucosa (Cox and Cuthbert, 1989a). However, the antisecretory activity of xylazine is markedly attenuated by piroxicam, indicating that endogenous eicosanoid formation is involved (Cox and Cuthbert, 1989a). The human colonic epithelial cell line, Colony-1, also possesses α2-adrenoceptors that are functionally antisecretory (Holliday, Tough and Cox, 1997).

Neuropeptide Y and related peptides Nerve fibres containing NPY are distributed throughout the wall of the intestine, belonging to myenteric, submucosal and extrinsic neurones (Keast, Furness and Costa, 1985b). At an ultrastructural level, NPY-containing nerves lie in close apposition to mucosal epithelial cells (Feher and Burnstock, 1986). Furthermore, NPY is co-stored with VIP in a population of neurones that have their cell bodies in the submucous plexus and innervate mucosal epithelial cells (Cox, Rudolph and Gschmeissner, 1994). Neuropeptide Y is an antisecretory neuropeptide, which in vitro inhibits chloride secretion from intestinal mucosa (Friel, Miller and Walker, 1986; Hubel and Renquist, 1986; Cox and Cuthbert, 1988; Cox et al., 1988). Piroxicam and indomethacin, both cyclooxygenase inhibitors, block responses to applied NPY, indicating that production of

NEUROEFFECTOR TRANSMISSION IN THE INTESTINE

321

prostaglandins is involved (Cox and Cuthbert, 1988; Cox et al., 1988). The NPY receptor is located preponderantly on the basolateral rather than apical aspect of the epithelial cells. The subtype of receptor may be species-dependent: for example, in the human epithelial cell line, Colony-6, Y1 receptors are present, in rat and rabbit intestinal mucosa there are Y2 receptors, and in dog intestinal epithelium there are Y4; at all these receptors PYY is more potent than or equipotent with NPY (Playford and Cox, 1996). However, in Colony-6, the pharmacological profile indicates an heterogeneous population of receptors (Cox and Tough, 1995). In rat jejunum, transport and ligand-binding studies also indicate an heterogeneous population of Y1 and Y2 receptors, together with a third, pancreatic polypeptideselective receptor (Souli et al., 1997). In humans, intravenous infusion of NPY increases jejunal net absorption of water, sodium, potassium and chloride ions under basal conditions, and markedly reduces the secretion stimulated by intraluminal instillation of prostaglandin E2 (PGE2) (Holzer-Petsche et al., 1991). In guinea-pig large intestinal mucosa, NPY has little or no effect on basal secretion, but was able to inhibit secretion evoked by either nerve stimulation, VIP or the cholinomimetic, bethanechol, (McCulloch et al., 1987). Similarly, in pig distal colon, NPY inhibits secretion induced by stimulating nerves electrically or with leukotriene C4 (Traynor, Brown and O’Grady, 1995). In rat small intestine, close arterial injection of NPY has no effect on basal secretion, but inhibits that evoked by PGE2 (Saria and Beubler, 1985). Peptide YY is an evolutionary analogue of NPY (Hoyle, 1998), but is found in enterochromaffin cells rather than enteric nerves. It also has antisecretory effects, mediated via Y receptors (Playford and Cox, 1996). Y receptors are G protein-coupled receptors that mediate inhibition of adenylate cyclase (Cox and Krstenansky, 1991; Mannon, Mervin and Sheriff-Carter, 1994; Playford and Cox, 1996). The human colonic epithelial cell line, Colony-1, is unresponsive to NPY and PYY, but in these cells, pancreatic polypeptide (PP) is a potent inhibitor of VIP-stimulated secretion. Responses to PP are unaffected by preincubation with PYY, implying that this cell line bears Y4 receptors (Holliday and Cox, 1996).

PLASTICITY OF EXPRESSION OF NEUROTRANSMITTERS PLASTICITY IN THE AUTONOMIC NERVOUS SYSTEM The particular combination and quantity of neuroactive substances expressed by neurones is partly pre-programmed and partly determined by the influence of trophic factors that trigger the expression or suppression of the appropriate genetic machinery. The plasticity of the autonomic nervous system during ageing, following trauma, surgery, after chronic exposure to drugs and in disease is well documented (Burnstock, 1990; Milner and Burnstock, 1994). The enteric nervous system is no exception to this. For example, whilst the total number of myenteric neurones in the intestine declines with age (Gabella, 1990), an increasing proportion remaining contain NOS (Santer, 1994; Belai, Cooper and Burnstock, 1995; Belai and Burnstock, 1999). During mammalian hibernation, when there are extended periods of gastrointestinal inactivity, there is a selective increase in the number of substance P and CGRP-containing enteric neurones (Shochina et al., 1997). There are

322

INNERVATION OF THE GASTROINTESTINAL TRACT

also many examples of plasticity of enteric neurones during chronic denervation and in colonic disease which are given below. Effects of chronic extrinsic denervation Surgical extrinsic denervation of the gut results in an altered innervation profile of submucosal arteries in guinea-pig ileum. Normally, submucosal arteries are innervated by extrinsic sensory nerves, which contain both substance P and CGRP, and sympathetic nerves, which contain noradrenaline, ATP and NPY. Six weeks after denervation, substance P- (without CGRP) and VIP-containing fibres are abundant in these vessels, possibly as a result of sprouting of axons of intrinsic origin (Galligan, Costa and Furness, 1988). Myenteric neurones appear to be essential for normal, sympathetic reinnervation of gastrointestinal circular smooth muscle. After surgical extrinsic denervation of the jejunum combined with chemical myenteric denervation with benzyldimethyltetradecylammonium chloride, the pattern of sympathetic nerves reinnervating the gut is altered and there is evidence for innervation of the circular muscle by the submucous plexus, probably induced by the increased production of neurotrophic factors from the hyperplastic smooth muscle (Luck et al., 1993). These changes may represent a part of the adaptive response of the enteric nervous system which permits normal intestinal function in the absence of extrinsic neuronal inputs. In some species, VIP levels in the small intestine are increased following long-term extrinsic denervation (Nelson, Sarr and Go, 1991), probably as a result of enhanced transcriptional regulation in the intrinsic ganglia, as occurs following extrinsic denervation and transection of the intestine (Stadelmann et al., 1996). Surgical coeliac ganglionectomy does not affect innervation of the intestine by substance P-containing nerve fibres or neurokinin receptor-binding density (Ouyang et al., 1996). Effects of chronic sympathectomy Long-term chemical sympathectomy leads to altered peptide expression in the gut wall that differs from the responses to surgical sympathectomy. Five months after guanethidine sympathectomy of neonatal rats, levels of CGRP and substance P increase in the myenteric plexus and surrounding smooth muscle of the ileum while NPY levels are unchanged. More surprisingly, noradrenaline levels, which are depleted six weeks after treatment, return to normal (Figure 7.9). This suggests that enteric neurones may turn on synthesis of noradrenaline and specific neuropeptides when the extrinsic sympathetic innervation is irreversibly destroyed (Milner et al., 1995). The functional implications of these changes are awaited. Guanethidine sympathectomy of adult rats leads to transitory increases in VIP and neurotensin levels in the colon one week after cessation of treatment, which normalise five weeks later (Nelson et al., 1988). Immunosympathectomy by neonatal administration of antiserum to nerve growth factor (NGF) leads to an increase in VIP, galanin, and substance P in nerves of the myenteric plexus of the ileum of the rat at four and eight weeks of age, but has no effect on noradrenaline, CGRP- and NPY-containing nerves (Belai, Aberdeen and Burnstock, 1992). Thus NGF availability may be an important regulatory agent in the post-natal expression of at least some enteric neuropeptides.

NEUROEFFECTOR TRANSMISSION IN THE INTESTINE

323

Figure 7.9 Noradrenaline levels in the external muscle layers containing the myenteric plexus of ileum from neonatal guanethidine-treated rats at 6, 12 and 20 weeks of age (dotted bars) compared with age-matched controls (clear bars). ** P < 0.001. Results are expressed as mean ± S.E.M (µg/g tissue). The number of animals per group is given below the horizontal axis. (Reproduced from Milner et al., 1995, with permission from Elsevier Science).

NEUROEFFECTOR DYSFUNCTION IN PATHOLOGICAL CONDITIONS Diabetes mellitus Gastrointestinal dysfunction is frequently reported in diabetes mellitus, with either diarrhoea or constipation. No causal relationship between altered motility and changes in innervation in the gut has been established, but autonomic neuropathy has been implicated in reduced peristalsis, dilation of the oesophagus, gastric retention and disordered small intestinal movement and colonic atony or megacolon. Studies on the streptozotocin-treated diabetic rat, which has been used as a model of diabetes mellitus, have shown that during the course of the progression of diabetes, there are differential changes in the expression of neurotransmitters/neuromodulators in nerves supplying the bowel. Whilst there are degenerative changes in VIP- and noradrenalinecontaining nerves early on in the development of the disease, the expression of 5-HT, substance P and CGRP in nerve fibres is altered later on. The lack of release of VIP and CGRP from enteric nerves of early-diabetic rats during transmural electrical stimulation was reversed by acute application of insulin (Burnstock, Mirsky and Belai, 1988). Rigorous control of glycaemia in streptozotocin-induced diabetes prevents the increased expression of VIP and galanin in the myenteric plexus (Belai et al., 1996). NPY-immunoreactive

324

INNERVATION OF THE GASTROINTESTINAL TRACT

nerve fibres appear comparatively resistant to change in diabetes although increased NPY levels in ileal myenteric neurones have been reported after a long diabetic period . In comparison with the ileum and proximal colon, the distal colon, appears to be relatively resistant to neurodegenerative changes due to diabetes (Figure 7.10). This may be explained by functional differences along the intestine or by the difference in origin of the sympathetic nerves to the ileum and proximal colon and to the distal colon (Belai et al., 1991). A reduction of the expression of NOS in the myenteric plexus of the antrum of the diabetic rat is not apparent in the ileum and colon (Takahashi et al., 1997; Wrzos et al., 1997). Idiopathic chronic constipation Patients with idiopathic constipation with normal bowel diameter have an increased whole-gut transit time. This may be related to an imbalance of enteric transmitter release, since VIP levels are reduced in the myenteric plexus and muscle layers of patients with this colonic motility disorder while levels of substance P and NPY are normal (Koch et al., 1988; Milner et al., 1990). In view of its role as an inhibitory neurotransmitter in peristalsis, an abnormally low concentration of VIP would possibly correlate with excess segmenting and reduced propulsion in constipation. Disturbances in the function of cholinergic innervation of the taenia coli of the colon have also been reported (Burleigh, 1988). 5-HT levels are elevated in circular smooth muscle and mucosa in this disorder, but unchanged in preparations containing the plexuses (Lincoln et al., 1990). Reduced numbers of substance P and VIP-immunoreactive nerve fibres in colonic circular muscle have been reported in biopsy samples taken from children with severe intractable constipation (Hutson, Chow and Borg, 1996). Reduced levels of substance P have also been reported in the mucosa (Goldin et al., 1989) and in submucosa isolated from rectal biopsies from patients with slow transit constipation showing normal levels of VIP and somatostatin (Tzavella et al., 1996). It appears then that disturbances or imbalances of both excitatory and inhibitory elements of intestinal innervation may contribute to bowel motility disorders associated with chronic constipation. In constipation associated with idiopathic megacolon or megarectum, abnormalities in inhibitory systems may contribute to abnormal gut function and subsequent bowel dilatation; in the rectum of these patients the density of innervation by nerves containing VIP and NADPH diaphorase (a marker for NO) is increased in the muscularis mucosae and lamina propria but decreased in the longitudinal muscle layer (Gattuso et al., 1996). Hirschsprung’s disease Hirschsprung’s disease is a congenital disorder of the gut characterised by the absence of enteric ganglia in variable lengths of the terminal colon. The aganglionic gut, in the absence of inhibitory neurones, is constricted and there is formation of megacolon on the oral side. There is a striking hyperinnervation of the colonic smooth muscle by extrinsic nerves while intrinsic neurones do not appear to enter the aganglionic segment (Hamada et al., 1987). In the absence of postganglionic cells, cholinergic fibres appear to seek alternative targets, in particular the smooth muscle; because of this, cholinergic nerve stimulation can cause colonic spasticity. In adjacent segments of ganglionic bowel, the extrinsic innervation appears

NEUROEFFECTOR TRANSMISSION IN THE INTESTINE

325

Figure 7.10 Immunofluorescence micrographs showing VIP-immunoreactive nerve fibres in the myenteric plexus of (Column A) ileum and (Column B) distal colon from control (C) and 8-week (D8), 16-week (D16) and 25-week (D25) streptozotocin-diabetic rats. Note that in the ileum, after 8-weeks of diabetes there was an increase in fluorescence intensity and density of VIP-immunoreactive nerve fibres compared to controls, followed by a progressive decrease by 16 and 25 weeks. In the distal colon, the pattern of change was different, with no difference in VIP-immunoreactivity after 8 weeks of diabetes but increased fluorescence intensity of VIP-immunoreactive nerve fibres mainly around the unstained ganglion cells of the plexus which was restored to control levels by 25 weeks of diabetes. Calibration bar = 30 µm. (Reproduced from Belai et al., 1991, with permission from Harcourt).

326

INNERVATION OF THE GASTROINTESTINAL TRACT

normal and comparatively sparse (Gannon, Noblet and Burnstock, 1969). Nerve fibres containing NPY appear to proliferate in the aganglionic region (Larsson, 1994), probably because they have their cell body oustide the gut wall. In the aganglionic sections there is a depletion of neuronal substance P and VIP (Bishop et al., 1981; Larsson, 1994), enkephalin, galanin, PACAP and CGRP (Larsson, Malmfors and Sundler, 1983; Shen et al., 1992; Larsson, 1994). Several authors have reported an absence of neuronal NOS from aganglionic segments, although it may be found within extrinsic nerve fasciculi (Larsson, 1994; O’Kelly et al., 1994; Larsson et al., 1995; Guo et al., 1997). Neuronal NOS mRNA is expressed in aganglionic segments of Hirschsprung’s disease only at very low levels (Kusafuka and Puri, 1997). There is no presence of inhibitory transmission in the aganglionic segment, either at a mechanical or electrophysiological level (Kubota, Ito and Ikeda, 1983; Tomita et al., 1995). In the lethal spotted mouse (or piebald lethal mouse), a mutant that develops foetal megacolon subsequent to colonic aganglionosis, although inhibitory transmission in affected large intestine is lost, the smooth muscle still relaxes in response to VIP or the NO generator, sodium nitroprusside (Chakder, McHugh and Rattan, 1997). In tissue from patients with long aganglionic regions, there may be small relaxant neurogenic responses in smooth muscle lying adjacent to the ganglionic region (Kamimura, Kubota and Suita, 1997); these are probably due to nerve fibres penetrating from the ganglionic region, or could be due to NOS contained in extrinsic fasciculi or rami. Intracellular recordings have shown that there are no differences in resting membrane potentials in the smooth muscle cells from ganglionic and aganglionic segments, but whereas ganglionic segments support regular slow-wave activity, the activity in aganglionic segments is very irregular (Kubota, Ito and Ikeda, 1983). Single pulses of electrical field stimulation evoke non-adrenergic relaxation in ganglionic but not aganglionic segments, whereas EJPs and IJPs can be recorded in ganglionic but not aganglionic segments (Kubota, Ito and Ikeda, 1983). The lethal spotted mouse also demonstrates a lack of junction potentials in the constricted aganglionic segment (Okasora and Okamoto, 1986). The lack of IJPs in either the human condition or the animal model may indicate that the spasticity in the colon is due to the lack of postganglionic, non-adrenergic inhibitory neurones, in addition to the proliferation of cholinergic neurones. The lack of these neurones is responsible for the non-propagation of migrating contractile complexes from ganglionic to aganglionic segments of the colon. Three susceptible genes have been identified in this disorder, the RET proto-oncogene, the endothelin B receptor gene, and the endothelin 3 gene (Edery et al., 1994; Amiel et al., 1996; Hirata, 1996). Ulcerative colitis In ulcerative colitis, in which the mucosa becomes inflamed, there are changes in the gut musculature, and probably its innervation, although the inflammation does not usually extend beyond the submucosa. The haustration of the colon is destroyed due to shortening of the longitudinal muscles (taenia coli) and relaxation of the circular muscles. This results in a decrease in the distal resistance to faecal flow, decreased transit time, diarrhoea, and compromised water absorption.

NEUROEFFECTOR TRANSMISSION IN THE INTESTINE

327

There are changes in kallikrein and kininogen activity in the muscle layers (Zeitlin and Smith, 1973) in ulcerative colitis. The increased activity of kinins in the muscle layers may be responsible, at least in part, for the loss of haustration in the colon, because in the human taenia coli, bradykinin has been shown to cause an increase in tone, and in the circular muscle it causes a loss of tone (Fishlock, 1966). It is also likely that the kinins modulate neuronal activity directly, although they could also modulate it indirectly by affecting the postjunctional muscle membrane. Decreased colonic tissue levels of VIP in the mucosal/submucosal layers in ulcerative colitis together with increased VIP mRNA expression supports the suggestion of axonal degeneration and increased expression of neuronal VIP (Schülte-Bockholt et al., 1995). Greater responses of the colonic smooth muscle to electrical field stimulation in the presence of adrenergic and cholinergic blockade in tissues from patients with ulcerative colitis suggest that NANC inhibitory nerves play an important role in the observed impaired motility (Tomita, Munakata and Tanjoh, 1998).

SUMMARY AND FUTURE DIRECTIONS Most information on neurotransmission in the enteric nervous system is on animal gut, with still comparatively little information available in man. The recent advances in identifying and localising an increasing number of putative neurotransmitters and neuromodulators in the enteric nervous system and molecular biological approaches to identify their receptor subtypes will expand the search for selective agonists and antagonists as tools to unravel further the mechanisms of intestinal neurotransmission. These will have potential use to alleviate disorders of motility or secretion. There is growing recognition of the plasticity of enteric nerves, some of which may be adaptive to allow normal functioning of the gut. This opens up the possibility of manipulation of these events to facilitate beneficial compensatory changes and offers new therapeutic strategies to improve gastrointestinal function.

REFERENCES Allescher, H.D., Sattler, D., Piller, C., Schusdziarra, V. and Classen, M. (1992). Ascending neural pathways in the rat ileum in vitro – effect of capsaicin and involvement of nitric oxide. European Journal of Pharmacology, 217, 153–162. Amelsberg, M., Amelsberg, A., Ainsworth, M.A., Hogan, D.L. and Isenberg, J.I. (1996). Cyclic adenosine3′,5′-monophosphate production is greater in rabbit duodenal crypt than in villus cells. Scandinavian Journal of Gastroenterology, 31, 233–239. Amiel, J., Attie, T., Jan, D., Pelet, A., Edery, P., Bidaud, C. et al. (1996). Heterozygous endothelin receptor B (EDNRB) mutations in isolated Hirschsprung disease. Human Molecular Genetics, 5, 355–357. Baird, A.A. and Muir, T.C. (1990). Membrane hyperpolarization, cyclic nucleotide levels and relaxation in the guinea-pig internal anal sphincter. British Journal of Pharmacology, 100, 329–335. Barajas-López, C., Huizinga, J.D., Collins, S.M., Gerzanich, V., Espinosa-Luna, R. and Peres, A.L. (1996). P2xpurinoceptors of myenteric neurones from the guinea-pig ileum and their unusual pharmacological properties. British Journal of Pharmacology, 119, 1541–1548. Baron, S.A., Jaffé, B.M. and Gintzler, A.R. (1983). Release of substance P from the enteric nervous system: direct quantitation and characterization. Journal of Pharmacology and Experimental Therapeutics, 227, 365–368.

328

INNERVATION OF THE GASTROINTESTINAL TRACT

Barthó, L., Holzer, P., Donnerer, J. and Lembeck, F. (1982). Evidence for the involvement of substance P in the atropine-resistant peristalsis of the guinea-pig ileum. Neuroscience Letters, 32, 69–74. Barthó, L., Kóczán, G., Pethö, G. and Maggi, C.A. (1992). Blockade of nitric oxide synthase inhibits nerve-mediated contraction in the rat small intestine. Neuroscience Letters, 145, 43–46. Bauer, V. (1993). NANC transmission in intestines and its pharmacological modulation. Acta Neurobiologica Experientia (Warsaw), 53, 65–77. Bauer, V. and Kuriyama, H. (1982). The nature of non-cholinergic, non-adrenergic transmission in longitudinal and circular muscle of the guinea-pig ileum. Journal of Physiology, 332, 375–391. Bayliss, W.M. and Starling, E.H. (1899). The movements and innervation of the small intestine. Journal of Physiology, 24, 99–143. Belai, A. and Burnstock, G. (1994). Evidence for coexistence of ATP and nitric oxide in non-adrenergic, noncholinergic (NANC) inhibitory neurones in the rat ileum, colon and anococcygeus muscle. Cell and Tissue Research, 278, 197–200. Belai, A. and Burnstock, G. (1999). Distribution and colocalisation of nitric oxide synthase and calretinin in the myenteric neurons of developing, aging and Crohn’s disease human small intestine. Digestive Diseases Science, 44, 1579–1587. Belai, A., Lincoln, J., Milner, P. and Burnstock, G. (1991). Differential effect of streptozotocin-induced diabetes on the innervation of the ileum and distal colon. Gastroenterology, 100, 1024–1032. Belai, A., Aberdeen, J. and Burnstock, G. (1992). Differential effects of immunosympathectomy on the expression of rat enteric neurotransmitters. Neuroscience Letters, 139, 157–160. Belai, A., Cooper, S. and Burnstock, G. (1995). Effect of age on NADPH-diaphorase-containing myenteric neurones of rat ileum and proximal colon. Cell and Tissue Research, 279, 379–383. Belai, A., Calcutt, N.A., Carrington, A.L., Diemel, L.T., Tomlinson, D.R. and Burnstock, G. (1996). Enteric neuropeptides in streptozotocin-diabetic rats; effects of insulin and aldose reductase inhibition. Journal of the Autonomic Nervous System, 58, 163–169. Bennett, M.R., Burnstock, G. and Holman, M.E. (1966). Transmission from intramural inhibitory nerves to the smooth muscle of the guinea-pig taenia coli. Journal of Physiology, 182, 541–558. Berezin, I., Snyder, S.H., Bredt, D.S. and Daniel, E.E. (1994). Ultrastructural localization of nitric oxide synthase in canine small intestine and colon. American Journal of Physiology, 266, C981–C989. Beubler, E. (1980). Influence of vasoactive intestinal polypeptide on net water flux and cyclic adenosine 3′,5′-monophosphate formation in the rat jejunum. Naunyn Schmiedebergs Archives of Pharmacology, 313, 243–247. Bishop, A.E., Polak, J.M., Lake, B.D., Bryant, M.G. and Bloom, S.R. (1981). Abnormalities of the colonic regulatory peptides in Hirschsprung’s disease. Histopathology, 5, 679–688. Blakeley, A.G.H. and Cunnane, T.C. (1979). The packeted release of transmitter from the sympathetic nerves of the guinea-pig vas deferens: an electrophysiological study. Journal of Physiology, 296, 85–96. Blakeley, A.G.H., Cunnane, T.C. and Petersen, S.A. (1982). Local regulation of transmitter release from rodent sympathetic nerve terminals. Journal of Physiology, 325, 93–109. Blottiere, H.M., Loirand, G. and Pacaud, P. (1996). Rise in cytosolic Ca2+ concentration induced by P2-purinoceptor activation in isolated myocytes from the rat gastrointestinal tract. British Journal of Pharmacology, 117, 775–780. Boeckxstaens, G.E., Pelckmans, P.A., Bult, H., De Man, J.G., Herman, A.G. and van Maercke, Y.M. (1991). Evidence for nitric oxide as mediator of non-adrenergic non-cholinergic relaxations induced by ATP and GABA in the canine gut. British Journal of Pharmacology, 102, 434–438. Boeckxstaens, G.E., Pelckmans, P.A., Herman, A.G. and Van Maercke, Y.M. (1993). Involvement of nitric oxide in the inhibitory innervation of the human isolated colon. Gastroenterology, 104, 690–697. Bogers, J.J., Pelckmans, P.A., Boeckxstaens, G.E., De Man, J.G., Herman, A.G. and van Maercke, Y.M. (1991). The role of nitric oxide in serotonin-induced relaxations in the canine terminal ileum and ileocolonic junction. Naunyn Schmiedeberg’s Archives of Pharmacology, 344, 716–719. Bolton, T.B. and Lim, S.P. (1991). Action of acetylcholine on smooth muscle. Zeitschrift fur Kardiologie, 80(Suppl. 7), 73–77. Bornstein, J.C. and Furness, J.B. (1988). Correlated electrophysiological and histochemical studies of submucous neurons and their contribution to understanding enteric neural circuits. Journal of the Autonomic Nervous System, 25, 1–13. Bornstein, J.C., Furness, J.B. and Costa, M. (1987). Sources of excitatory synaptic inputs to neurochemically identified submucous neurons of guinea-pig small intestine. Journal of the Autonomic Nervous System, 18, 83–89. Bornstein, J.C., Furness, J.B. and Kunze, W.A.A. (1994). Electrophysiological characterisation of myenteric neurons : how do classification schemes relate? Journal of the Autonomic Nervous System, 48, 1–15. Bortoff, A. (1976). Myogenic control of intestinal motility. Physiological Reviews, 56, 418–434.

NEUROEFFECTOR TRANSMISSION IN THE INTESTINE

329

Bouvier, M. and Gonella, J. (1981). Nervous control of the internal anal sphincter of the cat. Journal of Physiology, 310, 457–469. Brock, J.A. and Cunnane, T.C. (1995). Electrophysiology of neuroeffector transmission in smooth muscle. In Autonomic Neuroeffector Mechanisms, edited by G. Burnstock and C.H.V. Hoyle, pp. 121–213. Chur, Switzerland: Harwood Academic Press. Brookes, S.J.H., Steele, P.A. and Costa, M. (1991). Identification and immunohistochemistry of cholinergic and non-cholinergic circular muscle motor neurones in the guinea-pig small intestine. Neuroscience, 42, 863–878. Browning, J.G., Hardcastle, J., Hardcastle, P.T. and Redfern, J.S. (1978). Localization of the effect of acetylcholine in regulating intestinal ion transport. Journal of Physiology, 281, 15–27. Broyart, J.P., Dupont, C., Laburthe, M. and Rosselin, G. (1981). Characterization of vasoactive intestinal peptide receptors in human colonic epithelial cells. Journal of Clinical Endocrinology and Metabolism, 52, 715–721. Burleigh, D.E. (1988). Evidence for a functional cholinergic deficit in human colonic tissue resected for constipation. Journal of Pharmacy and Pharmacology, 40, 55–57. Burleigh, D.E. (1992). NG-nitro-L-arginine reduces nonadrenergic, noncholinergic relaxations of human gut. Gastroenterology, 102, 679–683. Burleigh, D.E. and Kirkham, S.E. (1993). Lack of effect of three putative vasoactive intestinal peptide receptor antagonists on vasoactive intestinal peptide-induced secretory responses in rat colon. European Journal of Pharmacology, 249, 239–242. Burleigh, D.E. and Yull, D.N. (1992). Evidence against substance P as a neurotransmitter at the neuroepithelial junction in rat colonic mucosa. Journal of Pharmacy and Pharmacology, 44, 257–258. Burns, G.A. and Stephens, K.E. (1995). Expression of mRNA for the N-methyl-D-aspartate (NMDAR1) receptor and vasoactive intestinal polypeptide (VIP) co-exist in enteric neurons of the rat. Journal of the Autonomic Nervous System, 55, 207–210. Burns, A.J., Lomax, A.E., Torihashi, S., Sanders, K.M. and Ward, S.M. (1996). Interstitial cells of Cajal mediate inhibitory neurotransmission in the stomach. Proceedings of the National Academy of Sciences USA, 93, 12008–12013. Burnstock, G. (1972). Purinergic nerves. Pharmacological Reviews, 24, 509–581. Burnstock, G. (1976). Do some nerve cells release more than one transmitter? Neuroscience, 1, 239–248. Burnstock, G. (1986). Autonomic neuromuscular junctions: current developments and future directions. Journal of Anatomy, 146, 1–30. Burnstock, G. (1990). Changes in expression of autonomic nerves in aging and disease. Journal of the Autonomic Nervous System, 30, 525–534. Burnstock, G. (1996). Purinergic neurotransmission. Seminars in the Neurosciences, 8, 171–257. Burnstock, G. (1997). The past, present and future of purine nucleotides as signalling molecules. Neuropharmacology, 36, 1127–1139. Burnstock, G. (2001). Purinergic signalling in the gut. In Handbook of Experimental Pharmacology, volume 151, Purinergic and Pyrimidinergic Signalling II, edited by M.P. Abbracchio and M. Williams, pp. 141–238. Berlin: Springer-Verlag. Burnstock, G., Campbell, G., Bennett, M. and Holman, M.E. (1963). Inhibition of the smooth muscle of the taenia coli. Nature, 200, 581–582. Burnstock, G., Campbell, G., Satchell, D. and Smythe, A. (1970). Evidence that adenosine triphosphate or a related nucleotide is the transmitter substance released by non-adrenergic inhibitory nerves in the gut. British Journal of Pharmacology, 40, 668–688. Burnstock, G., Mirsky, R. and Belai, A. (1988). Reversal of nerve damage in streptozotocin-diabetic rats by acute application of insulin in vitro. Clinical Science, 75, 629–635. Calignano, A., Whittle, B. J. R., Di Rosa, M. and Moncada, S. (1992). Involvement of endogenous nitric oxide in the regulation of rat intestinal motility in vivo. European Journal of Pharmacology, 229, 273–276. Carey, H.V. and Cooke, H.J. (1989). Neuromodulation of intestinal transport in the suckling mouse. American Journal of Physiology, 256, R481–R486. Carey, H.V., Tien, X.Y., Wallace, L.J. and Cooke, H.J. (1987). Muscarinic receptor subtypes mediating the mucosal response to neural stimulation of guinea-pig ileum. American Journal of Physiology, 253, G323–G329. Chakder, S. and Rattan, S. (1992). Neurally mediated relaxation of opossum internal anal sphincter: influence of superoxide anion generator and the scavenger. Journal of Pharmacology and Experimental Therapeutics, 260, 1113–1118. Chakder, S. and Rattan, S. (1993a). Involvement of cAMP and cGMP in relaxation of internal anal sphincter by neural stimulation, VIP, and NO. American Journal of Physiology, 264, G702–G707. Chakder, S. and Rattan, S. (1993b). Release of nitric oxide by activation of nonadrenergic noncholinergic neurons of internal anal sphincter. American Journal of Physiology, 264, G7–G12. Chakder, S., McHugh, K.M. and Rattan, S. (1997). Inhibitory neurotransmission in lethal spotted mutant mice: a model for Hirschsprung’s disease. Gastroenterology, 112, 1575–1585.

330

INNERVATION OF THE GASTROINTESTINAL TRACT

Chandan, R., Hildebrand, K.R., Seybold, V.S., Soldani, G. and Brown, D.R. (1991a). Cholinergic neurons and muscarinic receptors regulate anion secretion in pig distal jejunum. European Journal of Pharmacology, 193, 265–273. Chandan, R., Megarry, B.H., O’Grady, S.M., Seybold, V.S. and Brown, D.R. (1991b). Muscarinic cholinergic regulation of electrogenic chloride secretion in porcine proximal jejunum. Journal of Pharmacology and Experimental Therapeutics, 257, 908–917. Chandan, R., O’Grady, S.M. and Brown, D.R. (1991). Modulation of Na+, Cl– and HCO3– transport by carbachol in pig distal jejunum. European Journal of Pharmacology, 193, 257–264. Chassaing, G., Lavielle, S., Brunissen, A., Carruette, A., Garret, C., Petitet, F. et al. (1992). [Pro9]SP and [pGlu6, Pro9]SP (6–11) interact with two different receptors in the guinea-pig ileum as demonstrated with new SP antagonists. Neuropeptides, 23, 73–79. Cohn, J.A. (1987). Vasoactive intestinal peptide stimulates protein phosphorylation in a colonic epithelial cell line. American Journal of Physiology, 253, G420–G424. Cole, W.C. and Sanders, K.M. (1989). G proteins mediate suppression of Ca2+-activated K+ current by acetylcholine in smooth muscle cells. American Journal of Physiology, 257, C596–C600. Communi, D., Parmentier, M. and Boeynaems, J.M. (1996). Cloning, functional expression and tissue distribution of the human P2Y6 receptor. Biochemical and Biophysical Research Communications, 222, 303–308. Cook, R.D. and Burnstock, G. (1976). The ultrastructure of Auerbach’s plexus in the guinea-pig. I. Neuronal elements. Journal of Neurocytology, 5, 171–194. Cooke, H.J. (1989). Role of the “little brain” in the gut in water and electrolyte homeostasis. FASEB Journal, 3, 127–138. Cooke, H.J., Zafirova, M., Carey, H.V., Walsh, J.H. and Grider, J. (1987). Vasoactive intestinal polypeptide actions on the guinea-pig intestinal mucosa during neural stimulation. Gastroenterology, 92, 361–370. Costa, M., Brookes, S.J.H., Steele, P.A., Gibbins, I., Burcher, E. and Kandiah, J. (1996). Neurochemical classification of myenteric neurons in the guinea-pig ileum. Neuroscience, 75, 949–967. Cooke, H.J., Sidhu, M., Fox, P., Wang, Y.Z. and Zimmermann, E.M. (1997). Substance P as a mediator of colonic secretory reflexes. American Journal of Physiology, 272, G238–G245. Costa, M., Furness, J.B., Pullin, C.O. and Bornstein, J. (1985). Substance P enteric neurons mediate non-cholinergic transmission to the circular muscle of the guinea-pig intestine. Naunyn Schmiedeberg’s Archives of Pharmacology, 328, 446–453. Costa, M., Furness, J.B., Pompolo, S., Brookes, S.J.H., Bornstein, J.C., Bredt, D.S. and Snyder, S.H. (1992). Projections and chemical coding of neurons with immunoreactivity for nitric oxide synthase in the guineapig small intestine. Neuroscience Letters, 148, 121–125. Cox, H.M. (1992). Pituitary adenylate cyclase activating polypeptides, PACAP-27 and PACAP-38: stimulators of electrogenic ion secretion in the rat small intestine. British Journal of Pharmacology, 106, 498–502. Cox, H.M. and Cuthbert, A.W. (1988). Neuropeptide Y antagonises secretagogue evoked chloride transport in rat jejunal epithelium. Pflügers Archives, 413, 38–42. Cox, H.M. and Cuthbert, A.W. (1989a). Antisecretory activity of the α2-adrenoceptor agonist, xylazine in rat jejunal epithelium. Naunyn Schmiedeberg’s Archives of Pharmacology, 339, 669–674. Cox, H.M. and Cuthbert, A.W. (1989b). Secretory actions of vasoactive intestinal polypeptide, peptide histidine isoleucine and helodermin in rat small intestine: the effects of putative VIP antagonists upon VIP-induced ion secretion. Regulatory Peptides, 26, 127–135. Cox, H.M., Cuthbert, A.W., Hakanson, R. and Wahlestedt, C. (1988). The effect of neuropeptide Y and peptide YY on electrogenic ion transport in rat intestinal epithelia. Journal of Physiology, 398, 65–80. Cox, H.M. and Krstenansky, J.L. (1991). The effects of selective amino acid substitution upon neuropeptide Y antisecretory potency in rat jejunum mucosa. Peptides, 12, 323–327. Cox, H.M. and Tough, I.R. (1995). Functional characterization of receptors with affinity for PYY, NPY, [Leu31,Pro34]NPY and PP in a human colonic epithelial cell line. British Journal of Pharmacology, 116, 2673–2678. Cox, H.M., Tough, I.R., Grayson, K. and Yarrow, S. (1993). Pharmacological characterisation of neurokinin receptors mediating anion secretion in rat descending colon mucosa. Naunyn Schmiedeberg’s Archives of Pharmacology, 348, 172–177. Cox, H.M., Rudolph, A. and Gschmeissner, S. (1994). Ultrastructural co-localization of neuropeptide Y and vasoactive intestinal polypeptide in neurosecretory vesicles of submucous neurons in the rat jejunum. Neuroscience, 59, 469–476. Crist, J.R., He, X.D. and Goyal, R.K. (1992). Both ATP and the peptide VIP are inhibitory transmitters in guinea-pig ileum circular muscle. Journal of Physiology, 447, 119–131. Cuthbert, A.W. and Hickman, M.E. (1985). Indirect effects of adenosine triphosphate on chloride secretion in mammalian colon. Journal of Membrane Biology, 86, 157–166.

NEUROEFFECTOR TRANSMISSION IN THE INTESTINE

331

Dalziel, H.H., Thornbury, K.D., Ward, S.M. and Sanders, K.M. (1991). Involvement of nitric oxide synthetic pathway in inhibitory junction potentials in canine proximal colon. American Journal of Physiology, 260, G789–G792. de la Monte, S.M., Quertermous, T., Hong, C.C. and Bloch, K.D. (1995). Regional and maturation-associated expression of endothelin 2 in rat gastrointestinal tract. Journal of Histochemistry and Cytochemistry, 43, 203–209. De Vos, W.C. (1993). Role of the enteric nervous system in the control of migrating spike complexes in the feline small intestine. American Journal of Physiology, 265, G628–G637. Dho, S., Stewart, K. and Foskett, J.K. (1992). Purinergic receptor activation of Cl- secretion in T84 cells. American Journal of Physiology, 262, C67–C74. Dockray, G.J. (1994). Physiology of enteric neuropeptides. In Physiology of the Gastrointestinal Tract, 3rd edition, edited by L.R. Johnson, pp. 169–209. New York: Raven press. Dun, N.J., Dun, S.L., Lin, H.H., Hwang, L.L., Saria, A. and Fischer-Colbrie, R. (1997). Secretoneurin-like immunoreactivity in rat sympathetic, enteric and sensory ganglia. Brain Research, 760, 8–16. Eaker, E., Sallustio, J., Kohler, J. and Visner, G. (1995). Endothelin-1 expression in myenteric neurons cultured from rat small intestine. Regulatory Peptides, 55, 167–177. Edery, P., Lyonnet, S., Mulligan, L.M., Pelet, A., Dow, E., Abel, L. et al. (1994). Mutations of the RET protooncogene in Hirschsprung’s disease. Nature, 367, 378–380. Escrig, C., Bishop, A.E., Inagaki, H., Moscoso, G., Takahashi, K., Varndell, I.M. et al. (1992). Localisation of endothelin like immunoreactivity in adult and developing human gut. Gut, 33, 212–217. Evans, R.J. and Surprenant, A. (1992). Vasoconstriction of guinea-pig submucosal arterioles following sympathetic nerve stimulation is mediated by the release of ATP. British Journal of Pharmacology, 106, 242–249. Feher, E. and Burnstock, G. (1986). Electron microscopic study of neuropeptide Y-containing nerve elements of the guinea-pig small intestine. Gastroenterology, 91, 956–965. Fishlock, D. J. (1966). Effect of bradykinin on human isolated small and large intestine. Nature, 212, 1533–1535. Fox, J.E., McDonald, T.J., Alford, L. and Kostolanska, F. (1986). Tachykinin activation of muscarinic inhibition in canine small intestine is SPP in nature. Life Sciences, 39, 1123–1128. Friel, D.D., Miller, R.J. and Walker, M.W. (1986). Neuropeptide Y: a powerful modulator of epithelial ion transport. British Journal of Pharmacology, 88, 425–431. Fuchs, M., Adermann, K., Raab, H.R., Forssmann, W.G. and Kuhn, M. (1996). Pituitary adenylate cyclase-activating polypeptide: a potent activator of human intestinal ion transport. Annals of the New York Academy of Science, 805, 640–647. Furness, J.B. and Costa, M. (1971a). Monoamine oxidase histochemistry of enteric neurones in the guinea-pig. Histochimie, 28, 324–336. Furness, J.B. and Costa, M. (1971b). Morphology and distribution of intrinsic adrenergic neurones in the proximal colon of the guinea-pig. Zeitschrift für Zelforschungen den Mikroskopische Anatomie, 120, 346–363. Furness, J.B. and Costa, M. (1987). The Enteric Nervous System. Edinburgh: Churchill Livingstone. Furness, J.B., Costa, M. and Llewellyn-Smith, I.J. (1981). Branching patterns and projections of enteric neurons containing different putative transmitters. Peptides, 2(Suppl. 2), 119–121. Furness, J.B., Costa, M., Rokaeus, A., McDonald, T.J. and Brooks, B. (1987). Galanin-immunoreactive neurons in the guinea-pig small intestine: their projections and relationships to other enteric neurons. Cell and Tissue Research, 250, 607–615. Furness, J.B., Bornstein, J.C., Murphy, R. and Pompolo, S. (1992). Roles of peptides in transmission in the enteric nervous system. Trends in Neuroscience, 15, 66–71. Furness, J.B., Bornstein, J.C., Pompolo, S., Young, H.M., Kunze, W.A.A. and Kelly, H. (1994). The circuitry of the enteric nervous system. Neurogastroenterology and Motility, 6, 241–253. Furness, J.B., Johnson, P.J., Pompolo, S. and Bornstein, J.C. (1995). Evidence that enteric motility reflexes can be initiated through entirely intrinsic mechanisms in the guinea-pig small intestine. Neurogastroenterology and Motility, 7, 89–96. Furness, J.B., Kunze, W.A.A., Bertrand, P.P., Clerc, N. and Bornstein, J.C. (1998). Intrinsic primary afferent neurons of the intestine. Progress in Neurobiology, 54, 1–18. Gabella, G. (1972). Fine structure of the myenteric plexus in the guinea-pig ileum. Journal of Anatomy, 111, 69–97. Gabella, G. (1990). On the plasticity of form and structure of enteric ganglia. Journal of the Autonomic Nervous System, 30, 559–566. Gabella, G. (1995). Fine structure of post-ganglionic nerve fibres and autonomic neuroeffector junctions. In Autonomic Neuroeffector Mechanisms, edited by G. Burnstock and C.H.V. Hoyle, pp. 1–31. Chur, Switzerland: Harwood Academic Press. Galligan, J.J., Costa, M. and Furness, J.B. (1988). Changes in surviving nerve fibers associated with submucosal arteries following extrinsic denervation of the small intestine. Cell and Tissue Research, 253, 647–656.

332

INNERVATION OF THE GASTROINTESTINAL TRACT

Gannon, B.J., Noblet, H.R. and Burnstock, G. (1969). Adrenergic innervation of bowel in Hirschsprung’s disease. British Medical Journal, 3, 338–340. Gattuso, J.M., Hoyle, C.H., Milner, P., Kamm, M.A. and Burnstock, G. (1996). Enteric innervation in idiopathic megarectum and megacolon. International Journal of Colorectal Diseases, 11, 264–271. Gershon, M.D., Mawe, G.M. and Branchek, T.A. (1989). 5-HT and enteric neurones. In The Peripheral Actions of 5-HT, edited by J.R. Fozard, pp. 247–273. Oxford: Oxford University Press. Gillespie, J.S. and Khoyi, M.A. (1977). The site and receptors responsible for the inhibition by sympathetic nerves of intestinal smooth muscle and its parasympathetic motor nerves. Journal of Physiology, 267, 767–789. Gillespie, J.S. and Maxwell, J.D. (1971). Adrenergic innervation of sphincteric and non-sphincteric smooth muscle in the rat intestine. Journal of Histochemistry and Cytochemistry, 19, 676–681. Goldin, E., Karmeli, F., Selinger, Z. and Rachmilewitz, D. (1989). Colonic substance P levels are increased in ulcerative colitis and decreased in chronic severe constipation. Digestive Diseases Science, 34, 754–757. Gorfine, S.R. (1995). Treatment of benign anal disease with topical nitroglycerin. Diseases of the Colon and Rectum, 38, 453–456. Grady, E.F., Baluk, P., Bohm, S., Gamp, P.D., Wong, H., Payan, D.G. et al. (1996). Characterization of antisera specific to NK1, NK2 and NK3 neurokinin receptors and their utilization to localize receptors in the rat gastrointestinal tract. Journal of Neuroscience, 16, 6975–6986. Grider, J.R., Bitar, K.N. and Makhlouf, G.M. (1987). Identification of muscarinic M2 receptors on single muscle cells of the human and guinea-pig intestine. Gastroenterology, 93, 951–957. Grider, J.R., Murthy, K.S., Jin, J.G. and Makhlouf, G.M. (1992). Stimulation of nitric oxide from muscle cells by VIP: prejunctional enhancement of VIP release. American Journal of Physiology, 262, G774–G778. Grider, J.R., Katsoulis, S., Schmidt, W. and Jin, J.G. (1994). Regulation of the descending relaxation phase of intestinal peristalsis by PACAP. Journal of the Autonomic Nervous System, 50, 151–159. Guo, R., Nada, O., Suita, S., Taguchi, T. and Masumoto, K. (1997). The distribution and co-localization of nitric oxide synthase and vasoactive intestinal polypeptide in nerves of the colons with Hirschsprung’s disease. Virchows Archives, 430, 53–61. Gustafsson, B.I. and Delbro, D. (1993). Tonic inhibition of small intestinal motility by nitric oxide. Journal of the Autonomic Nervous System, 44, 179–187. Gustafsson, L.E., Wiklund, C.U., Wiklund, N.P., Persson, M.G. and Moncada, S. (1990). Modulation of autonomic neuroeffector transmission by nitric oxide in guinea-pig ileum. Biochemical and Biophysical Research Communications, 173, 106–110. Hamada, Y., Bishop, A.E., Federici, G., Rivosecchi, M., Talbot, I.C. and Polak, J.M. (1987). Increased neuropeptide Y-immunoreactive innervation of aganglionic bowel in Hirschsprung’s disease. Virchows Archives, 411, 369–377. He, X.D. and Goyal, R.K. (1993). Nitric oxide involvement in the peptide VIP-associated inhibitory junction potential in the guinea-pig ileum. Journal of Physiology, 461, 485–499. Hills, J.M. and Jessen, K.R. (1992). Transmission: γ-amino butyric acid (GABA), 5-hydroxytryptamine (5-HT) and dopamine. In Autonomic Neuroeffector Mechanisms, edited by G. Burnstock and C.H.V. Hoyle, pp. 465–507. Chur, Switzerland: Harwood Academic Publishers. Hirata, Y. (1996). Endothelin peptides. Current Opinions in Nephrology and Hypertension, 5, 12–15. Hogan, D.L., Yao, B., Steinbach, J.H. and Isenberg, J.I. (1993). The enteric nervous system modulates mammalian duodenal mucosal bicarbonate secretion. Gastroenterology, 105, 410–417. Holliday, N.D. and Cox, H.M. (1996). The functional investigation of a human adenocarcinoma cell line, stably transfected with the neuropeptide Y Y1 receptor. British Journal of Pharmacology, 119, 321–329. Holliday, N.D., Tough, I.R. and Cox, H.M. (1997). Inhibition of cyclic AMP-dependent chloride secretion by PP receptors and α2-adrenoceptors in a human colonic epithelial cell line. Naunyn Schmiedeberg’s Archives of Pharmacology, 355, 183–189. Holzer, P., Gamse, R. and Lembeck, F. (1980). Distribution of substance P in rat gastrointestinal tract: lack of effect of capsaicin treatment. European Journal of Pharmacology, 61, 303–307. Holzer, P. and Maggi, C.A. (1994). Synergistic role of muscarinic acetylcholine and tachykinin NK-2 receptors in intestinal peristalsis. Naunyn Schmiedeberg’s Archives of Pharmacology, 349, 194–201. Holzer, P., Schulet, W. and Maggi, C.A. (1993). Ascending enteric reflex contraction: roles of acetylcholine and tachykinins in relation to distension and propagation of excitation. Journal of Pharmacology and Experimental Therapeutics, 264, 391–396. Holzer-Petsche, U., Petritsch, W., Hinterleitner, T., Eherer, A., Sperk, G. and Krejs, G.J. (1991). Effect of neuropeptide Y on jejunal water and ion transport in humans. Gastroenterology, 101, 325–330. Hoyle, C.H.V. (1992). Transmission: purines. In Autonomic Neuroeffector Mechanisms, edited by G. Burnstock and C.H.V. Hoyle, pp. 367–407. Chur, Switzerland: Harwood Academic Publishers. Hoyle, C.H. (1994). Non-adrenergic, non-cholinergic control of the urinary bladder. World Journal of Urology, 12, 233–244.

NEUROEFFECTOR TRANSMISSION IN THE INTESTINE

333

Hoyle, C.H.V. (1996a). Neuropeptides: Essential Data. Oxford: Bios Scientific Publishers. Hoyle, C.H.V. (1996b). Purinergic cotransmission: parasympathetic and enteric nerves. Seminars in the Neurosciences, 8, 207–215. Hoyle, C.H.V. (1998). Neuropeptide families: evolutionary perspectives. Regulatory Peptides, 73, 1–3. Hoyle, C.H.V. and Burnstock, G. (1989). Neuromuscular transmission in the gastrointestinal tract. In Handbook of Physiology – The Gastrointestinal System, Volume I, Section 6, edited by J.D. Wood, pp. 435–464. Bethesda, MA: American Physiological Society. Hoyle, C.H.V., Reilly, M.W., Lincoln, J. and Burnstock, G. (1988). Adrenergic, but not cholinergic or purinergic, responses are potentiated in the caecum of the diabetic rat. Gastroenterology, 94, 1357–1367. Hoyle, C.H., Kamm, M.A., Burnstock, G. and Lennard-Jones, J.E. (1990). Enkephalins modulate inhibitory neuromuscular transmission in circular muscle of human colon via delta-opioid receptors. Journal of Physiology, 431, 465–478. Hubel, K.A. (1984). Chymotrypsin, ileal chloride transport, and neurotransmitters. American Journal of Physiology, 247, G253–G260. Hubel, K.A. and Renquist, K.S. (1986). Effect of neuropeptide Y on ion transport by the rabbit ileum. Journal of Pharmacology and Experimental Therapeutics, 238, 167–169. Hutson, J.M., Chow, C.W. and Borg, J. (1996). Intractable constipation with a decrease in substance P immunoreactive fibres: is it a variant of neuronal intestinal dysplasia. Journal of Pediatric Surgery, 31, 580–583. Inagaki, H., Bishop, A.E., Escrig, C., Wharton, J., Allen-Mersh, T.G. and Polak, J.M. (1991). Localization of endothelin-like immunoreactivity and endothelin binding sites in human colon. Gastroenterology, 101, 47–54. Inoue, C.N., Woo, J.S., Schwiebert, E.M., Morita, T., Hanaoka, K., Guggino, S.E. and Guggino, W.B. (1997). Role of purinergic receptors in chloride secretion in Caco-2 cells. American Journal of Physiology, 272, C1862–C1870. Irie, K., Muraki, T., Furukawa, K. and Nomoto, T. (1991). L-NG-Nitro-arginine inhibits nicotine-induced relaxation of isolated rat duodenum. European Journal of Pharmacology, 202, 285–288. Ishida, N., Tsujioka, K., Tomoi, M., Saida, K. and Mitsui, Y. (1989). Differential activities of two distinct endothelin family peptides on ileum and coronary artery. FEBS Letters, 247, 337–340. Ito, Y. and Kuriyama, H. (1975). Responses to field stimulation of the smooth muscle membrane of the guineapig stomach. Japanese Journal of Physiology, 25, 333–344. Itoh, H., Sakai, J., Imoto, A. and Creed, K.E. (1995). The control of smooth muscle tissues by nonadrenergic non-cholinergic (NANC) nerve fibres in the autonomic nervous system. Journal of Smooth Muscle Research, 31, 67–78. Javed, N.H. and Cooke, H.J. (1992). Acetylcholine release from colonic submucous neurons associated with chloride secretion in the guinea-pig. American Journal of Physiology, 262, G131–G136. Jessen, K.R. and Burnstock, G. (1982). The enteric nervous system in tissue culture: a new mammalian model for the study of complex nervous networks. In Trends in Autonomic Pharmacology, Vol. II, edited by S. Kalsner, pp. 95–115. Baltimore: Urban and Schwartzenberg. Jin, J.G., Murthy, J.R., Grider, J.R. and Makhlouf, G.M. (1993). Activation of distinct cAMP- and cGMPdependent pathways by relaxant agents in isolated gastric muscle cells. American Journal of Physiology, 264, G470–G477. Jin, J.G., Katsoulis, S., Schmidt, W.E. and Grider, J.R. (1994). Inhibitory transmission in tenia coli mediated by distinct vasoactive intestinal peptide and apamin-sensitive pituitary adenylate cyclase-activating peptide. Journal of Pharmacology and Experimental Therapeutics, 270, 433–439. Jodal, M. (1990). Neuronal influence on intestinal transport. Journal of Internal Medicine Supplement, 732, 125–132. Kachur, J.F., Miller, R.J., Field, M. and Rivier, J. (1982). Neurohumoral control of ileal electrolyte transport. I. Bombesin and related peptides. Journal of Pharmacology and Experimental Therapeutics, 220, 449–455. Kamimura, T., Kubota, M. and Suita, S. (1997). Functional innervation of the aganglionic segment in Hirschsprung’s disease – a comparison of the short- and long-segment type. Journal of Pediatric Surgery, 32, 673–677. Kan, S., Niwa, M. and Taniyama, K. (1994). Specific receptor for vasoactive intestinal contractor in myenteric cholinergic neurones. European Journal of Pharmacology, 258, 139–143. Keast, J.R., Furness, J.B. and Costa, M. (1982). Sources of peptide-containing nerves in the mucosa of the guinea-pig small intestine. British Medical Bulletin, 38, 247–252. Keast, J.R., Furness, J.B. and Costa, M. (1985a). Different substance P receptors are found on mucosal epithelial cells and submucous neurons of the guinea-pig small intestine. Naunyn Schmiedebergs Archives of Pharmacology, 329, 382–387. Keast, J.R., Furness, J.B. and Costa, M. (1985b). Distribution of certain peptide-containing nerve fibres and endocrine cells in the gastrointestinal mucosa in five mammalian species. Journal of Comparative Neurology, 236, 403–422.

334

INNERVATION OF THE GASTROINTESTINAL TRACT

Keef, K.D., Du, C., Ward, S.M., McGregor, B. and Sanders, K.M. (1993). Enteric inhibitory neural regulation of human colonic circular muscle: role of nitric oxide. Gastroenterology, 105, 1009–1016. Keef, K.D., Murray, D.C., Sanders, K.M. and Smith, T.K. (1997). Basal release of nitric oxide induces an oscillatory motor pattern in canine colon. Journal of Physiology, 499, 773–786. Keranen, U., Vanhatalo, S., Kiviluoto, T., Kivilaakso, E. and Soinila, S. (1995). Colocalization of NADPH diaphorase reactivity and vasoactive intestinal polypeptide in human colon. Journal of the Autonomic Nervous System, 54, 177–183. Koch, T.R., Carney, J.A., Go, L. and Go, V.L. (1988). Idiopathic chronic constipation is associated with decreased colonic vasoactive intestinal peptide. Gastroenterology, 94, 300–310. Kohn, P.G., Newey, H. and Smyth, D.H. (1970). The effect of adenosine triphosphate on the transmural potential in rat small intestine. Journal of Physiology, 208, 203–220. Komuro, T. (1982). The interstitial cells in the colon of the rabbit. Cell and Tissue Research, 222, 41–51. Korman, L.Y., Lemp, G.F., Jackson, M.J. and Gardner, J.D. (1982). Mechanism of action of ATP on intestinal epithelial cells: cyclic AMP-mediated stimulation of active ion transport. Biochimica et Biophysica Acta, 721, 47–54. Kow, L., Brookes, S.J.H. and Costa, M. (1993). Enteric reflexes in organotypic culture of guinea-pig small intestine. Proceedings of the Australian Neuroscience Society, 4, 79. Kubota, M., Ito, Y. and Ikeda, K. (1983). Membrane properties and innervation of smooth muscle cells in Hirschsprung’s disease. American Journal of Physiology, 244, G406–G415. Kusafuka, T. and Puri, P. (1997). Altered mRNA expression of the neuronal nitric oxide synthase gene in Hirschsprung’s disease. Journal of Pediatric Surgery, 32, 1054–1058. Kuwahara, A., Bowen, S., Wang, J., Condon, C. and Cooke, H.J. (1987a). Epithelial responses evoked by stimulation of submucosal neurons in guinea-pig distal colon. American Journal of Physiology, 252, G667–G674. Kuwahara, A., Kuwahara, Y., Mochizuki, T. and Yanaihara, N. (1993). Action of pituitary adenylate cyclaseactivating polypeptide on ion transport in guinea-pig distal colon. American Journal of Physiology, 264, G433–G441. Kuwahara, A., Tien, X.Y., Wallace, L.J. and Cooke, H.J. (1987b). Cholinergic receptors mediating secretion in guinea-pig colon. Journal of Pharmacology and Experimental Therapeutics, 242, 600–606. Laburthe, M., Prieto, J.C., Amiranoff, B., Dupont, C., Hui Bon Hoa, D. and Rosselin, G. (1979). Interaction of vasoactive intestinal peptide with isolated intestinal epithelial cells from rat. 2. Characterization and structural requirements of the stimulatory effect of vasoactive intestinal peptide on production of adenosine 3′: 5′-monophosphate. European Journal of Biochemistry, 96, 239–248. Langton, P., Ward, S.M., Carl, A., Norell, M.A. and Sanders, K.M. (1989). Spontaneous electrical activity of interstitial cells of Cajal isolated from canine proximal colon. Proceedings of the National Academy of Sciences USA, 86, 7280–7284. Larsson, L.T. (1994). Hirschsprung’s disease – immunohistochemical findings. Histology and Histopathology, 9, 615–629. Larsson, L.T., Malmfors, G. and Sundler, F. (1983). Peptidergic innervation in Hirschsprung’s disease. Zeitschrift fur Kinderchirugie, 38, 301–304. Larsson, L.T., Shen, Z., Ekblad, E., Sundler, F., Alm, P. and Andersson, K.E. (1995). Lack of neuronal nitric oxide synthase in nerve fibers of aganglionic intestine: a clue to Hirschsprung’s disease. Journal of Pediatric Gastroenterology and Nutrition, 20, 49–53. Lee, H.K., Shuttleworth, C.W. and Sanders, K.M. (1995). Tachykinins activate nonselective cation currents in canine colonic myocytes. American Journal of Physiology, 269, C1394–C1401. Lefebvre, R.A. (1995). Nitric oxide in the peripheral nervous system. Annals of Medicine, 27, 379–388. Leipziger, J., Kerstan, D., Nitschke, R. and Greger, R. (1997). ATP increases [Ca2+]i and ion secretion via a basolateral P2Y-receptor in rat distal colonic mucosa. Pflügers Archives, 434, 77–83. LePard, K.J., Messori, E. and Galligan, J.J. (1997). Purinergic fast excitatory postsynaptic potentials in myenteric neurons of guinea-pig: distribution and pharmacology. Gastroenterology, 113, 1522–1534. Lim, S.P. and Muir, T.C. (1986). Neuroeffector transmission in the guinea-pig internal anal sphincter: an electrical and mechanical study. European Journal of Pharmacology, 128, 17–24. Lin, W.W. and Lee, C.Y. (1992). Intestinal relaxation by endothelin isopeptides: involvement of Ca2+-activated K+ channels. European Journal of Pharmacology, 219, 355–360. Lincoln, J., Crowe, R., Kamm, M.A., Burnstock, G. and Lennard-Jones, J.E. (1990). Serotonin and 5-hydroxyindoleacetic acid are increased in the sigmoid colon in severe idiopathic constipation. Gastroenterology, 98, 1219–1225. Lincoln, J., Hoyle, C.H.V. and Burnstock, G. (1995). Transmission: nitric oxide. In Autonomic Neuroeffector Mechanisms, edited by G. Burnstock and C.H.V. Hoyle, pp. 509–539. Chur, Switzerland: Harwood Academic Publishers.

NEUROEFFECTOR TRANSMISSION IN THE INTESTINE

335

Lincoln, J., Hoyle, C.H.V. and Burnstock, G. (1997). Nitric Oxide in Health and Disease. Cambridge: Cambridge University Press. Liu, L. and Coupar, I.M. (1997). Role of α2-adrenoceptors in the regulation of intestinal water transport. British Journal of Pharmacology, 120, 892–898. Loder, P.B., Kamm, M.A., Nicholls, R.J. and Phillips, R.K. (1994). “Reversible chemical sphicterotomy” by local application of glyceryl trinitrate. British Journal of Surgery, 81, 1386–1389. Lucchesi, P.A., Romano, F.D., Scheid, C.R., Yamaguchi, H. and Honeyman, T.W. (1989). Interaction of agonists and selective antagonists with gastric smooth muscle muscarinic receptors. Naunyn Schmiedeberg’s Archives of Pharmacology, 339, 145–151. Luck, M.S., Dahl, J.L., Boyeson, M.G. and Bass, P. (1993). Neuroplasticity in the smooth muscle of the myenterically and extrinsically denervated rat jejunum. Cell and Tissue Research, 271, 363–374. Lynn, R.B., Sankey, S.L., Chakder, S. and Rattan, S. (1995). Colocalization of NADPH-diaphorase staining and VIP immunoreactivity in neurons in opossum internal anal sphincter. Digestive Diseases Science, 40, 781–791. Maas, A.J.J., Den Hertog, A., Ras, R. and Van den Akker, J. (1980). The action of apamin on guinea-pig taenia caeci. European Journal of Pharmacology, 67, 265–274. Maggi, C.A., Patacchini, R., Giachetti and Meli, A. (1990). Tachykinin receptors in the circular muscle of the guinea-pig ileum. British Journal of Pharmacology, 101, 996–1000. Maggi, C.A., Barbanti, G., Turini, D. and Giuliani, S. (1991). Effect of NG-monomethyl-L-arginine (L-NMMA) and NG-nitro-L-arginine (L-NOARG) on non-adrenergic, non-cholinergic relaxation in the circular muscle of the human ileum. British Journal of Pharmacology, 103, 1970–1972. Maggi, C.A., Catalioto, R.M., Criscuoli, M., Cucchi, P., Giuliani, S., Lecci, A. et al. (1997). Tachykinin receptors and intestinal motility. Canadian Journal of Physiology and Pharmacology, 75, 696–703. Mannon, P.J., Mervin, S.J. and Sheriff-Carter, K.D. (1994). Characterization of a Y1-preferring NPY/PYY receptor in HT-29 cells. American Journal of Physiology, 267, G901–G907. Mao, Y.K., Wang, Y.F., Moogk, C., Fox-Threlkeld, J.E.T., Xiao, Q., McDonald, T.J. and Daniel, E.E. (1998). Locations and molecular forms of PACAP and sites and characteristics of PACAP receptors in canine ileum. American Journal of Physiology, 274, G217–G225. Martinson, J. and Muran, A. (1963). Excitatory and inhibitory effects of vagus stimulation on gastric motility in the cat. Acta Physiologica Scandinavica, 57, 309–316. Masson, S.D., McKay, D.M., Stead, R.H., Agro, A., Stanisz, A. and Perdue, M.H. (1996). Nippostrongylus brasiliensis infection evokes neuronal abnormalities and alterations in neurally regulated electrolyte transport in rat jejunum. Parasitology, 113, 173–182. Matsuo, K., Katsuragi, T., Fujiki, S., Sato, C. and Furukawa, T. (1997). ATP release and contraction mediated by different P2-receptor subtypes in guinea-pig ileal smooth muscle. British Journal of Pharmacology, 121, 1744–1748. McConalogue, K., Lyster, D.J. and Furness, J.B. (1995). Electrophysiological analysis of the actions of pituitary adenylate cyclase-activating peptide in the taenia of the guinea-pig caecum. Naunyn Schmiedeberg’s Archives of Pharmacology, 352, 538–544. McCulloch, C.R., Kuwahara, A., Condon, C.D. and Cooke, H.J. (1987). Neuropeptide modification of chloride secretion in guinea-pig distal colon. Regulatory Peptides, 19, 35–43. Milner, P. and Burnstock, G. (1994). Trophic factors and the control of smooth muscle development and innervation. In Airways Smooth Muscle: Development and Regulation of Contractility, edited by D. Raeburn and M.A. Giembycz, pp. 1–39. Basel: Birkhauser Verlag. Milner, P., Crowe, R., Kamm, M.A., Lennard-Jones, J.E. and Burnstock, G. (1990). Vasoactive intestinal polypeptide levels in sigmoid colon in idiopathic constipation and diverticular disease. Gastroenterology, 99, 666–675. Milner, P., Lincoln, J., Belai, A. and Burnstock, G. (1995). Plasticity in the myenteric plexus of the rat ileum after long-term sympathectomy. International Journal of Developmental Neuroscience, 13, 385–392. Neild, T.O., Shen, K.Z. and Surprenant, A. (1990). Vasodilation of arterioles by acetylcholine released from single neurones in the guinea-pig submucosal plexus. Journal of Physiology, 420, 247–265. Nelson, D.K., Sarr, M.G. and Go, V.L.W. (1991). In vivo neural isolation of the canine jejunoileum: temporal adaptation of enteric neuropeptides. Gut, 32, 1336–1341. Nelson, D.K., Service, J.E., Studelska, D.R., Brimijoin, S. and Go, V.L.W. (1988). Gastrointestinal neuropeptide concentrations following guanethidine sympathectomy. Journal of the Autonomic Nervous System, 22, 203–210. O’Kelly, T., Brading, A. and Mortensen, N. (1993). Nerve mediated relaxation of the human internal anal sphincter: the role of nitric oxide. Gut, 34, 689–693. O’Kelly, T.J., Davies, J.R., Tam, P.K., Brading, A.F. and Mortensen, N.J. (1994). Abnormalities of nitric-oxideproducing neurons in Hirschsprung’s disease: morphology and implications. Journal of Pediatric Surgery, 29, 294–299.

336

INNERVATION OF THE GASTROINTESTINAL TRACT

O’ Malley, K.E., Farrell, C.B., Boyle, K.M. and Baird, A.W. (1995). Cholinergic activation of Cl– secretion in rat colonic epithelia. European Journal of Pharmacology, 275, 83–89. Ohkawa, H. (1983). Inhibitory junction potentials of the guinea-pig duodenum in the treatment with catecholamines. Tohoku Journal of Experimental Medicine, 140, 209–219. Okasora, T. and Okamoto, E. (1986). Electrophysiological and pharmacological study on innervation of the aganglionic colon in Hirschsprung’s disease of human and murine model. Zeitschrift fur Kinderchirugie, 41, 93–96. Okuno, M., Shinomura, Y., Himeno, S., Kashimura, M. and Tauri, S. (1988). Effects of cholinergic agonists on VIP release from Meissner’s plexus of rabbit ileum. Biomedical Research, 9, 119–125. Ouyang, A., Zimmermann, K., Wong, K.L., Sharp, D. and Reynolds, J.C. (1996). Effect of celiac ganglionectomy on tachykinin innervation, receptor distribution and intestinal responses in the rat. Journal of the Autonomic Nervous System, 61, 292–300. Pacaud, P., Feolde, E., Frelin, C. and Loirand, G. (1996). Characterization of the P2Y-purinoceptor involved in the ATP-induced rise in cytosolic Ca2+ concentration in rat ileal myocytes. British Journal of Pharmacology, 118, 2213–2219. Parsons, A.M., Seybold, V., Chandan, R., Vogt, J., Larson, A.A., Murray, C.R. et al. (1992). Neurokinin receptors and mucosal ion transport in porcine jejunum. Journal of Pharmacology and Experimental Therapeutics, 261, 1213–1221. Perdue, M.H., Galbraith, R. and Davison, J.S. (1987). Evidence for substance P as a functional neurotransmitter in guinea-pig small intestinal mucosa. Regulatory Peptides, 18, 63–74. Playford, R.J. and Cox, H.M. (1996). Peptide YY and neuropeptide Y: two peptides intimately involved in electrolyte homeostasis. Trends in Pharmacological Sciences, 17, 436–438. Portbury, A.L., Furness, J.B., Young, H.M., Southwell, B.R. and Vigna, S.R. (1996). Localisation of NK1 receptor immunoreactivity to neurons and interstitial cells of the guinea-pig gastrointestinal tract. Journal of Comparative Neurology, 367, 342–351. Porter, A.J., Wattchow, D.A., Brookes, S.J., Schemann, M. and Costa, M. (1996). Choline acetyltransferase immunoreactivity in the human small and large intestine. Gastroenterology, 111, 401–408. Prestwich, S.A. and Bolton, T.B. (1995). G-protein involvement in muscarinic receptor-stimulation of inositol phosphates in longitudinal smooth muscle from the small intestine of the guinea-pig. British Journal of Pharmacology, 114, 119–126. Prieto, J.C., Laburthe, M., Hoa, D.H. and Rosselin, G. (1981). Quantitative studies of vasoactive intestinal peptide (VIP) binding sites and VIP-induced adenosine 3′: 5′-monophosphate production in epithelial cells from duodenum, jejunum, ileum, cecum, colon and rectum in the rat. Acta Endocrinologica Copenhagen, 96, 100–106. Publicover, N.G., Hammond, E.M. and Sanders, K.M. (1993). Amplification of nitric oxide signaling by interstitial cells isolated from canine colon. Proceedings of the National Academy of Science USA, 90, 2087–2091. Rae, M.G. and Muir, T.C. (1996). Neuronal mediators of inhibitory junction potentials and relaxation in the guinea-pig internal anal sphincter. Journal of Physiology, 493, 517–527. Rand, M.J. (1992). Nitrergic transmission: nitric oxide as a mediator of non-adrenergic, non-cholinergic neuroeffector transmission. Clinical and Experimental Pharmacology and Physiology, 19, 147–169. Rangachari, P.K., Prior, T. and McWade, D. (1990). Epithelial and mucosal preparations from canine colon: responses to substance P. Journal of Pharmacology and Experimental Therapeutics, 254, 1076–1083. Rattan, S. and Chakder, S. (1993). Inhibitory effect of CO on internal anal sphincter: heme oxygenase inhibitor inhibits NANC relaxation. American Journal of Physiology, 265, G799–G804. Rattan, S. and Thatikunta, P. (1993). Role of nitric oxide in sympathetic neurotransmission in opossum internal anal sphincter. Gastroenterology, 105, 827–836. Rattan, S., Sarkar, A. and Chakder, S. (1992). Nitric oxide pathway in rectoanal inhibitory reflex of opossum internal anal sphincter. Gastroenterology, 103, 43–50. Rattan, S., Rosenthal, G.J. and Chakder, S. (1995). Human recombinant hemoglobin (rHb1.1) inhibits nonadrenergic noncholinergic (NANC) nerve-mediated relaxation of internal anal sphincter. Journal of Pharmacology and Experimental Therapeutics, 272, 1211–1216. Reddix, R.A. and Cooke, H.J. (1992). Neurokinin 1 receptors mediate substance P-induced changes in ion transport in guinea-pig ileum. Regulatory Peptides, 39, 215–225. Reddix, R., Kuhawara, A., Wallace, L. and Cooke, H.J. (1994). Vasoactive intestinal polypeptide: a transmitter in submucous neurons mediating secretion in guinea-pig distal colon. Journal of Pharmacology and Experimental Therapeutics, 269, 1124–1129. Reilly, W.M., Hoyle, C.H.V. and Burnstock, G. (1987). Pre- and post-junctional adrenoceptor subtypes in the circular muscle of the guinea-pig caecum. Naunyn Schmiedeberg’s Archives of Pharmacology, 336, 37–47.

NEUROEFFECTOR TRANSMISSION IN THE INTESTINE

337

Reimer, R., Odes, H.S., Beil, W., Schwenk, M., Muallem, R. and Sewing, K.F. (1996). Bicarbonate secretion in the guinea-pig duodenum: functional characterization of peptide hormone receptors in duodenal enterocytes. Pharmacology, 52, 339–346. Richards, N.W., Allbee, W.E., Giginella, T.S. and Wallace, L.J. (1987). Exogenous ATP-stimulated calcium uptake in isolated rat intestinal epithelial cells. Life Sciences, 40, 1665–1672. Rogers, D.C. and Burnstock, G. (1966). The interstitial cell and its place in the concept of the autonomic ground plexus. Journal of Comparative Neurology, 126, 255–285. Rubin, M., Fournet, J., Snape, W.J. and Cohen, S. (1980). Adrenergic regulation of ileocaecal sphincter function in the cat. Gastroenterology, 78, 15–21. Saffrey, M.J., Hassall, C.J.S., Hoyle, C.H.V., Belai, A., Moss, J., Schmidt, H.H.H.W. et al. (1992). Colocalisation of nitric oxide synthase and NADPH-diaphorase activity in cultured myenteric neurones. NeuroReport, 3, 333–336. Saida, K., Mitsui, Y. and Ishida, N. (1989). A novel peptide, vasoactive intestinal contractor, of a new (endothelin) peptide family. Molecular cloning, expression, and biological activity. Journal of Biological Chemistry, 264, 14613–14616. Saida, K., Gao, X., Matsui, M. and Mitsui, Y. (1996). Vasoactive intestinal contractor (VIC)/ mouse ET-2 and VIC receptor: biological activity, gene expression, and specific receptor. Nippon Rinsho, 54, 1115–1121. Sakato, M., Shimo, Y. and Bando, T. (1972). Effect of noradrenaline on non-adrenergic inhibitory response of guinea-pig taenia coli. Japanese Journal of Physiology, 22, 137–138. Salomon, R., Couvineau, A., Rouyer-Fessard, C., Voisin, T., Lavallee, D., Blais, A. et al. (1993). Characterization of a common VIP-PACAP receptor in human small intestinal epithelium. American Journal of Physiology, 264, E294–E300. Sanders, K.M. (1996). A case for interstitial cells of Cajal as pacemakers and mediators of neurotransmission in the gastrointestinal tract. Gastroenterology, 111, 492–515. Sanders, K.M. and Ward, S.M. (1992). Nitric oxide as a mediator of nonadrenergic noncholinergic neurotransmission. American Journal of Physiology, 262, G379–G392. Santer, R.M. (1994). Survival of the population of NADPH-diaphorase stained myenteric neurons in the small intestine of aged rats. Journal of the Autonomic Nervous System, 49, 115–121. Saria, A. and Beubler, E. (1985). Neuropeptide Y (NPY) and peptide YY (PYY) inhibit prostaglandin E2-induced intestinal fluid and electrolyte secretion in the rat jejunum in vivo. European Journal of Pharmacology, 119, 47–52. Schmid, K.W., Kunk, B., Kirchmair, R., Totsch, M., Bocker, W. and Fischer Colbrie, R. (1995). Immunohistochemical detection of secretoneurin, a novel neuropeptide endoproteolytically processed from secretogranin II, in normal human endocrine and neuronal tissues. Histochemistry Journal, 27, 473–481. Schülte-Bockholt, A., Fink, J.G., Meier, D.A., Otterson, M.F., Telford, G.L., Hopp, K. et al. (1995). Expression of mRNA for vasoactive intestinal polypeptide in normal human colon during inflammation. Molecular and Cellular Biochemistry, 142, 1–7. Schürman, G., Bishop, A.E., Facer, P., Eder, U., Fischer-Colbrie, R., Winkler, H. et al. (1995). Secretoneurin: a new peptide in the human enteric nervous system. Histochemistry and Cell Biology, 104, 11–19. Schworer, H., Clemens, A., Katsoulis, S., Kohler, H., Creutfeldt, W. and Schmidt, W.E. (1993). Pituitary adenylate cyclase-activating peptide is a potent modulator of human colonic motility. Scandinavian Journal of Gastroenterology, 28, 625–632. Shen, Z., Larsson, L.T., Malmfors, G., Absood, A., Hakanson, R. and Sundler, F. (1992). A novel neuropeptide, pituitary adenylate cyclase-activating polypeptide (PACAP), in human intestine: evidence for reduced content in Hirschsprung’s disease. Cell and Tissue Research, 269, 369–374. Shochina, M., Belai, A., Toole, L., Knight, G. and Burnstock, G. (1997). Neurochemical coding in the myenteric plexus of the upper gastrointestinal tract of hibernating hamsters. International Journal of Developmental Neuroscience, 15, 353–362. Shuttleworth, C.W., Murphy, R. and Furness, J.B. (1991). Evidence that nitric oxide participates in non-adrenergic inhibitory transmission to intestinal muscle in the guinea-pig. Neuroscience Letters, 130, 77–80. Shuttleworth, C.W., Sanders, K.M. and Keef, K.D. (1993). Inhibition of nitric oxide synthesis reveals noncholinergic excitatory neurotransmission in the canine proximal colon. British Journal of Pharmacology, 109, 739–747. Shuttleworth, C.W. and Sanders, K.M. (1996). Involvement of nitric oxide in neuromuscular transmission in canine proximal colon. Proceedings of the Society for Experimental Biology and Medicine, 211, 16–23. Shuttleworth, C.W., Xue, C., Ward, S.M., de Vente, J. and Sanders, K.M. (1993). Immunohistochemical localization of 3′,5′-cyclic guanosine monophosphate in the canine proximal colon: responses to nitric oxide and electrical stimulation of enteric inhibitory neurons. Neuroscience, 56, 513–522. Smith, T.K., Bornstein, J.C. and Furness, J.B. (1990). Distension-evoked ascending and descending reflexes in the circular muscle of guinea-pig ileum: an intracellular study. Journal of the Autonomic Nervous System, 29, 203–218.

338

INNERVATION OF THE GASTROINTESTINAL TRACT

Sneddon, P., McLaren, G.J. and Kennedy, C. (1996). Purinergic cotransmission: sympathetic nerves. Seminars in the Neurosciences, 8, 201–205. Souli, A., Chariot, J., Voisin, T., Presset, O., Tsocas, A., Balasubramaniam, A. et al. (1997). Several receptors mediate the antisecretory effect of peptide YY, neuropeptide Y, and pancreatic polypeptide on VIP-induced fluid secretion in the rat jejunum in vivo. Peptides, 18, 551–557. Stadelmann, A.M., Telford, G.L., Appel, D.A., Walgenbach-Telford, S., Hopp, K., Meier, D. A. and Koch, T.R. (1996). Expression of mRNA for vasoactive intestinal peptide in rat small intestine. Molecular and Cellular Endocrinology, 116, 31–67. Stockley, H. and Bennett, A. (1977). Relaxations mediated by adrenergic and non-adrenergic nerves in human isolated taenia coli. Journal of Pharmacy and Pharmacology, 29, 533–537. Stutts, M.J., Lazarowski, E.R., Paradiso, A.M. and Boucher, R.C. (1995). Activation of CFTR Cl- conductance in polarized T84 cells by luminal extracellular ATP. American Journal of Physiology, 268, C425–C433. Sundler, F., Ekblad, E., Absood, A., Hakanson, R., Koves, K. and Arimura, A. (1992). Pituitary adenylate cyclase activating peptide: a novel vasoactive intestinal peptide-like neuropeptide in the gut. Neuroscience, 46, 439–454. Szurszewski, J.H. and King, B.F. (1989). Physiology of prevertebral ganglia in mammals with special reference to inferior mesenteric ganglion. In Handbook of Physiology – The Gastrointestinal System, Volume I, Section 6, edited by J.D. Wood, pp. 519–592. Bethesda, MA: American Physiological Society. Takahashi, T., Nakamura, K., Itoh, H., Sima, A.A.F. and Owyang, C. (1997). Impaired expression of nitric oxide synthase in the gastric myenteric plexus of spontaneously diabetic rats. Gastroenterology, 113, 1535–1544. Tam, F.S. and Hillier, K. (1992). The role of nitric oxide in mediating non-adrenergic non-cholinergic relaxation in longitudinal muscle of human taenia coli. Life Sciences, 51, 1277–1284. Tew, E.M.M., Anderson, P.N. and Burnstock, G. (1992). Implantation of the myenteric plexus into the corpus striatum of adult rats: survival of the neurones and glia and interactions with host brain. Restorative Neurology and Neuroscience, 4, 311–321. Tew, E.M., Saffrey, M.J., Anderson, P.N. and Burnstock, G. (1996). NADPH-diaphorase-containing enteric neurones survive for a year in the adult rat striatum. NeuroReport, 7, 958–960. Thornbury, K.D., Ward, S.M., Dalziel, H.H., Carl, A., Westfall, D.P. and Sanders, K.M. (1991). Nitric oxide and nitrosocysteine mimic nonadrenergic, noncholinergic hyperpolarization in canine proximal colon. American Journal of Physiology, 261, G553–G557. Thuneberg, L. (1982). Interstitial cells of Cajal: intestinal pacemakers? Advances in Anatomy Embryology and Cell Biology, 71, 1–130. Toda, N., Baba, H. and Okamura, T. (1990). Role of nitric oxide in non-adrenergic, non-cholinergic nervemediated relaxation in dog duodenal longitudinal muscle strips. Japanese Journal of Pharmacology, 53, 281–284. Toda, N., Tanobe, Y. and Baba, H. (1991). Suppression by NG-nitro-L-arginine of relaxations induced by nonadrenergic, non-cholinergic nerve stimulation in dog duodenal longitudinal muscle. Japanese Journal of Pharmacology, 57, 527–534. Toda, N., Baba, H., Tanobe, Y. and Okamura, T. (1992). Mechanism of relaxation induced by K+ and nicotine in dog duodenal longitudinal muscle. Journal of Pharmacology and Experimental Therapeutics, 260, 697–701. Tomita, R., Munakata, K., Kurosu, Y. and Tanjoh, K. (1995). A role of nitric oxide in Hirschsprung’s disease. Journal of Pediatric Surgery, 30, 437–440. Tomita, R., Munakata, K. and Tanjoh, K. (1998). Role of non-adrenergic, non-cholinergic inhibitory nerves in the colon of patients with ulcerative colitis. Journal of Gastroenterology, 33, 48–52. Tonini, M., Waterman, S.A., Candura, S.M., Coccini, T. and Costa, M. (1992). Sites of action of morphine on the ascending excitatory reflex in the guinea-pig small intestine. Neuroscience Letters, 144, 195–198. Torihashi, S., Kobayashi, S., Gerthoffer, W.T. and Sanders, K.M. (1993). Interstitial cells in deep muscular plexus of canine small intestine may be specialized smooth muscle cells. American Journal of Physiology, 265, G638–G645. Tøttrup, A., Glavind, E.B. and Svane, D. (1992). Involvement of the L-arginine-nitric oxide pathway in internal anal sphincter relaxation. Gastroenterology, 102, 409–415. Traynor, T.R., Brown, D.R. and O’Grady, S.M. (1995). Neuromodulation of ion transport in porcine distal colon: NPY reduces secretory actions of leukotrienes. American Journal of Physiology, 269, R426–R431. Tzavella, K., Riepl, R.L., Klauser, A.G., Voderholzer, W.A., Schindlbeck, N.E. and Muller-Lissner, S.A. (1996). Decreased substance P levels in rectal biopsies from patients with slow transit constipation. European Journal of Gastroenterology and Hepatology, 8, 1207–1211. Vanner, S. and Surprenant, A. (1996). Neural reflexes controlling intestinal microcirculation. American Journal of Physiology, 271, G223–G230. Venkova, K. and Krier, J. (1994). A nitric oxide and prostaglandin-dependent component of NANC off-contractions in cat colon. American Journal of Physiology, 266, G40–G47.

NEUROEFFECTOR TRANSMISSION IN THE INTESTINE

339

Verma, A., Hirsch, D.J., Glatt, C.E., Ronnett, G.V. and Snyder, S.H. (1993). Carbon monoxide: a putative neural messenger. Science, 259, 381–384. Vogalis, F., Ward, S.M. and Sanders, K.M. (1991). Correlation between electrical and morphological properties of canine pyloric circular muscle. American Journal of Physiology, 260, G390–G398. Wang, L., Stanisz, A.M., Wershil, B.K., Galli, S.J. and Perdue, M.H. (1995). Substance P induces ion secretion in mouse small intestine through effects on enteric nerves and mast cells. American Journal of Physiology, 269, G85–G92. Ward, S.M. and Sanders, K.M. (1990). Pacemaker activity in septal structures of canine colonic circular muscle. American Journal of Physiology, 259, G264–G273. Ward, S.M., Burke, E.P. and Sanders, K.M. (1990). Use of rhodamine123 to label and lesion interstitial cells of Cajal in canine colonic circular muscle. Anatomica Embryologica Berlin, 182, 215–224. Ward, S.M., Dalziel, H.H., Bradley, M.E., Buxton, I.L., Keef, K., Westfall, D.P. et al. (1992a). Involvement of cyclic GMP in non-adrenergic, non-cholinergic inhibitory neurotransmission in dog proximal colon. British Journal of Pharmacology, 107, 1075–1082. Ward, S.M., Dalziel, H.H., Thornbury, K.D., Westfall, D.P. and Sanders, K.M. (1992b). Nonadrenergic, noncholinergic inhibition and rebound excitation in canine colon depend on nitric oxide. American Journal of Physiology, 262, G237–G243. Ward, S.M., McKeen, E.S. and Sanders, K.M. (1992c). Role of nitric oxide in non-adrenergic, non-cholinergic inhibitory junction potentials in canine ileocolonic sphincter. British Journal of Pharmacology, 105, 776–782. Ward, S.M., Burns, A.J., Torihashi, S. and Sanders, K.M. (1994). Mutation of the proto-oncogene c-kit blocks development of interstitial cells and electrical rhythmicity in murine intestine. Journal of Physiology, 480, 91–97. Ward, S.M., Burns, A.J., Torihashi, S., Harney, S.C. and Sanders, K.M. (1995). Impaired development of interstitial cells and intestinal electrical rhythmicity in steel mutants. American Journal of Physiology, 269, C1577–C1585. Ward, S.M., Dalziel, H.H., Khoyi, M.A., Westfall, A.S., Sanders, K.M. and Westfall, D.P. (1996). Hyperpolarization and inhibition of contraction mediated by nitric oxide released from enteric inhibitory neurones in guinea-pig taenia coli. British Journal of Pharmacology, 118, 49–56. Waterman, S.A. and Costa, M. (1994). The role of enteric inhibitory motoneurons in peristalsis in the isolated guinea-pig small intestine. Journal of Physiology, 477, 459–468. Waterman, S.A., Costa, M. and Tonini, M. (1994). Accommodation mediated by enteric inhibitory reflexes in the isolated guinea-pig small intestine. Journal of Physiology, 474, 539–546. Werkström, V., Ny, L., Persson, K. and Andersson, K.-E. (1997). Carbon monoxide-induced relaxation and distribution of haem oxygenase isoenzymes in the pig urethra and lower oesophagogastric junction. British Journal of Pharmacology, 120, 312–318. Westfall, D.P., Hogaboom, G.K., Colby, J., O’Donnell, J.P. and Fedan, J.S. (1982). Direct evidence against a role of ATP as the nonadrenergic noncholinergic inhibitory neurotransmitter in guinea-pig taenia coli. Proceedings of the National Academy of Science USA, 79, 7041–7045. White, T.D. (1988). Role of adenine compounds in autonomic neurotransmission. In Pharmacology and Therapeutics 38th edition, Vol. 38, edited by C. Bell, pp. 129–168. London: Pergamon Press. Wiklund, N.P., Wiklund, C.U., Cederqvist, B., Ohlen, A., Hedqvist, P. and Gustafsson, L.E. (1991). Endothelin modulation of neuroeffector transmission in smooth muscle. Journal of Cardiovascular Pharmacology, 17(Suppl. 7), S335–S339. Wiklund, C.U., Olgart, C., Wiklund, N.P. and Gustafsson, L.E. (1993a). Modulation of cholinergic and substance P-like neurotransmission by nitric oxide in the guinea-pig ileum. British Journal of Pharmacology, 110, 833–839. Wiklund, N.P., Leone, A.M., Gustafsson, L.E. and Moncada, S. (1993b). Release of nitric oxide evoked by nerve stimulation in guinea-pig intestine. Neuroscience, 53, 607–611. Wollberg, Z., Bdolah, A., Galron, R., Sokolovsky, M. and Kochva, E. (1991). Contractile effects and binding properties of endothelins/sarafotoxins in the guinea-pig ileum. European Journal of Pharmacology, 198, 31–36. Wrzos, H.F., Cruz, A., Polavarapu, R., Shearer, D. and Ouyang, A. (1997). Nitric oxide synthase (NOS) expression in the myenteric plexus of streptozotocin-diabetic rats. Digestive Diseases Science, 42, 2106–2110. Wu, Z.C., Kisslinger, S.D. and Gaginella, T.S. (1982). Functional evidence for the presence of cholinergic nerve endings in the colonic mucosa of the rat. Journal of Pharmacology and Experimental Therapeutics, 221, 664–669. Xue, C., Pollock, J., Schmidt, H.H., Ward, S.M. and Sanders, K.M. (1994). Expression of nitric oxide synthase immunoreactivity by interstitial cells of the canine proximal colon. Journal of the Autonomic Nervous System, 49, 1–14.

340

INNERVATION OF THE GASTROINTESTINAL TRACT

Yoshinaga, M., Chijiiwa, Y., Misawa, T., Harada, N. and Nawata, H. (1992). Endothelin-B receptor on guineapig small intestinal smooth muscle cells. American Journal of Physiology, 262, G308–G311. Zagorodnyuk, V., Santicioli, P., Maggi, C.A. and Giachetti, A. (1996). The possible role of ATP and PACAP as mediators of apamin-sensitive NANC inhibitory junction potentials in circular muscle of guinea-pig colon. British Journal of Pharmacology, 119, 779–786. Zeitlin, I.J. and Smith, A.N. (1973). Mobilization of tissue kallikrein in inflammatory disease of the colon. Gut, 14, 133–139. Zholos, A.V. and Bolton, T.B. (1994). G-protein control of voltage dependence as well as gating of muscarinic metabotropic channels in guinea-pig ileum. Journal of Physiology, 478, 195–202.

8 Neural Control of Intestinal Vessels Neela Kotecha Department of Physiology, Monash University, Clayton, Victoria-3800, Australia Intestinal circulation is important in its own right due to its immense capacity for demand-related regulation of blood flow. As such, intestinal blood flow is influenced by well-developed neural regulatory mechanisms involving extrinsic nerves (sympathetic and sensory) as well as the intrinsic submucosal plexus of the enteric nervous system. The submucosal nerves subserve a vasodilator role as illustrated by studies on postprandial hyperaemia of the small intestine. However, the activity of the intrinsic nerves can be modified by extrinsic sympathetic nerves which act as a “brake” and serve to divert blood away from the intestine, when required. Although the mechanisms underlying sympathetic nerve transmission in the intestinal vessels have been a focus of study for the past two decades, it is only in the last decade that our understanding of mechanisms underlying responses to stimulation of sensory and intrinsic nerves has increased. It is clear from recent work that endothelial paracrine factors, the electrical connectivity between endothelium and vascular smooth muscle, a direct action on the vascular smooth muscle plus the interaction between vasomotor pathways are all involved in the neural regulation of vascular tone of the intestinal vessels. Despite substantial progress, we still do not know which particular subclasses of nerves are involved in normal physiological responses, nor do we have much understanding of changes in neural control resulting from diseased states such as diabetes, which is associated with widespread neuropathy and dysfunction of the enteric nervous system. Evaluating the roles of intrinsic and extrinsic nerves is an essential step towards understanding the neural regulation of intestinal arterioles in normal and pathophysiological states. The next decade promises to be both exciting and challenging as we piece together the jigsaw of neural control of intestinal vessels. KEY WORDS: autonomic nervous system; intestinal microcirculation; enteric nervous system; sensory nerves.

INTRODUCTION The splanchnic circulation is the largest strictly regional systemic circulation in the body as it receives at least one-fourth of the cardiac output. Its total capacity is as great as the entire blood volume. About a third of the splanchnic circulation passes through the hepatic circulation to the liver and the bulk of the remainder is distributed to the stomach and the small and large intestines through the mesenteric circulation. The intestinal circulation is interesting not only because it receives a large proportion of the cardiac output but also because of its impressive potential for demand related up- or downregulation of the blood 341

342

INNERVATION OF THE GASTROINTESTINAL TRACT

flow (Mitchell and Blomqvist, 1971). In the resting mammal, 10–15% of the cardiac output is distributed to the small intestine and 2–3% to the colon (Lundgren, 1984). The two together total nearly 20% of the cardiac output which is equal to blood flow in either the renal or the cerebral circulation. The vascular beds of the small and large intestine are the major circuits supplied by the coeliac artery and the superior and inferior mesenteric vessels. The superior mesenteric artery conveys the lion’s share of the splanchnic blood flow and perfuses the small intestine and the proximal bowel whereas the distal bowel is supplied mainly by the inferior mesenteric artery.

ANATOMICAL ARRANGEMENT OF BLOOD VESSELS IN THE INTESTINE The superior mesenteric artery arises from the anterior surface of the aorta at the level of the first lumbar vertebral body. It enters the mesentery and ends there by forming an arch with one of its own branches, the ileal branch of the ileocoelic artery, forming the superior mesenteric loop. The superior mesenteric veins lie to the right of the superior mesenteric artery in its course through the mesentery. The mesenteric arteries divide and anastomose forming a series of arterial arcades that divide into two branches as they reach the intestine and run longitudinally along the mesenteric border in the aboral and oral directions. Small vessels are derived from the first series of arcades and these contribute to a second series. As the vessels progress further into the mesentery, more complex patterns are formed and three or even four tiers of arcades are present. The terminal series of arcades give off a pair of small vessels, which pass around the small intestine on either side of the mesenteric border, penetrating into the intestinal layers and forming the arteriolar tree. The arteriolar tree lies in a sheet of connective tissue, along with the submucous plexus, between the overlying circular muscle layer and the underlying mucosa. The submucosal arterioles have been implicated as the major resistance vessels within the intestine and therefore the main determinant of blood flow to a particular region and the site of the local and remote control systems that continually regulate the rate of blood flow by changing vascular smooth muscle tone (Lundgren, 1984). However recent work suggests that the arcade small arteries and the submucosal arterioles contribute roughly equally to the mesenteric resistance and point to a major role for small arteries in resistance regulation in this vascular bed (Fenger-Gron, Mulvany and Christensen, 1995). The capillaries arising from the arterioles feed the mucosa. In the villi of the mucosa, the inflow capillary and the outflow capillary are close together, separated by perhaps just 10 µm. There are no arterio-venous shunts or if they exist, these are of minor functional importance in the gastrointestinal vascular bed, (Lundgren, 1984) and the extensive network in the mucosa cannot be bypassed. The short distance between the inflow and outflow capillaries, i.e. the countercurrent flow arrangement, permits diffusion of lipid soluble substances to cross the capillary endothelium thereby short-circuiting their passage through the capillary bed in the tip of the villus (Jodal and Lundgren, 1986). Oxygen meets this requirement and at normal or high blood pressure only a small portion of the oxygen carried by the feeder vessels escapes the long trip to the villus tip. In low flow states, however, the proportion of oxygen diffusing directly from inflow to outflow

NEURAL CONTROL OF INTESTINAL VESSELS

343

vessels is much greater. In other words, from base to tip of the villus there is an oxygen gradient; this may be a factor in cell death at the tip – ordinarily a normal process of mucosal cell turnover. The collecting venules converge to form individual mesenteric veins leaving the intestine near the entry points of the arteries of the mesenteric border eventually leading the blood flow into the hepatic circulation. Overall, the intestinal microcirculation consists of three major vessel types each subserving a different primary function. Arterioles offer resistance to the flow of blood, thus allowing these vessels to control blood pressure and flow to the intestine. Capillaries permit exchange of substances from blood to cells and vice versa. Lastly, venules store and release blood as needed by the rest of the body.

FACTORS THAT INFLUENCE VASCULAR TONE The blood flow to the intestine is normally highly variable, responding to the need for fluid secretion and absorption and the metabolic demands of the mucosa and muscle. Vascular tone is a balance between vasoconstrictor and vasodilator influences that are prevalent. These factors include circulating vasoactive agents, such as secretin, gastrin, and cholecystokinin (CCK), substances released from the endothelium, such as endotheliumderived relaxing factor (EDRF), endothelium-derived hyperpolarizing factor (EDHF) and endothelin, local metabolic products and neurogenic mediators released from both vasoconstrictor and vasodilator nerves that innervate the blood vessels. (See Table 8.1 for a list of abbreviations used in this review). These factors are not mutually exclusive; one can modulate the influence or release of another. Overall, these mechanisms work in order to maintain an adequate blood flow to meet the requirements of the target organ. TABLE 8.1 List of abbreviations. ACh ATP cAMP CCK CGRP ChAT 4-DAMP DYN EJP GAL EDHF EDRF ENS 5-HT NA NADPH NO NPY

Acetylcholine Adenosine 5′-triphosphate Cyclic adenosine monophosphate Cholecystokinin Calcitonin gene-related peptide Choline acetyl transferase 4-diphenylacetoxy methylpiperidine methiodide Dynorphin Excitatory junction potential Galanin Endothelium-derived hyperpolarizing factor Endothelium-derived relaxing factor Enteric nervous system 5-Hydroxytryptamine Noradrenaline Nicotinamide adenine dinucleotide phosphate Nitric oxide Neuropeptide Y

344

INNERVATION OF THE GASTROINTESTINAL TRACT

In this review, the focus will be on the neurogenic influences that control vessel calibre of the intestinal blood vessels. It has long been known that intestinal blood flow is influenced by well developed neural regulatory mechanisms within the enteric nervous system (ENS) of the intestine itself, as well as by extrinsic nerves. Relaxation of vascular smooth muscle induced by changes in neurogenic activity is achieved by two mechanisms acting either independently or simultaneously. The first of these mechanisms involves withdrawal of vasoconstrictor activity of sympathetic nerves. The second mechanism involves the release of substance/s which promotes vascular smooth muscle relaxation. The function of large arteries is to conduct blood to the distributing vessels. The smaller arteries and arterioles form the resistance vessels that are primarily involved in controlling blood flow to the organ. It is here that neural control, receptors for circulating hormones, and local factors have the greatest influence on organ blood flow because small changes in diameter will alter blood flow resistance to the greatest extent (resistance ∝ 1/radius4). Hence, I will focus my discussion mainly to neural mechanisms involved in the control of tone in arterioles of the small intestine as this is the most thoroughly investigated area of the gastrointestinal tract.

INNERVATION OF INTESTINAL BLOOD VESSELS The intestinal mucosa is richly supplied with nerve endings that are involved with enhancement of blood flow in the intestine. The control of intestinal blood vessels is complex, involving cholinergic inhibitory, adrenergic excitatory, and non-adrenergic, non-cholinergic inhibitory nerves of the autonomic nervous system. Intestinal blood vessels are specialised in that they are controlled by both extrinsic and intrinsic nerves. Extrinsic nerves include the postganglionic vasoconstrictor sympathetics arising from the prevertebral sympathetic ganglia and the sensory nerves arising from the dorsal root ganglia. No parasympathetic vasodilator innervation of the arterioles of the small intestine has been demonstrated, however the parasympathetic innervation of the colonic circulation arises from two main sources; the vagal and the pelvic nerves. Electrical stimulation of the vagal fibres to the colon does not elicit any flow changes apart from those caused by increased motility, whereas on pelvic nerve stimulation, blood flow increases three-fold (Hultén, Jodal and Lundgren, 1969). The mucosal lining of the gastrointestinal tract is provided with an extensive nervous supply, the ENS. Although known for more than a century, its function has been studied in detail only in the past two decades. Apart from controlling motility of the gut, the intrinsic nerves (all unmyelinated) of the ENS appear to form a dilator nerve supply to the arterioles in the intestine. There appears to be a paucity of innervation of the collecting venules of the guinea-pig small intestine and the rat mesentery (Furness, 1973; Furness and Costa, 1987). EXTRINSIC VASOCONSTRICTOR NERVES These arise from the sympathetic ganglia to supply the intestinal blood vessels where they form a meshwork of perivascular nerves. The nerves that supply the arterioles are restricted to the adventitial surface; there are many varicosities from which transmitter

NEURAL CONTROL OF INTESTINAL VESSELS

345

release may occur along the surface of the arteriole and ultrastructural studies indicate that virtually all of the varicosities form close appositions with arteriolar smooth muscle (Luff, McLachlan and Hirst, 1987). Sympathetic nerves are also prominent around myenteric and submucous ganglia (Furness and Costa, 1987). The morphological simplicity of the submucous arteriolar preparation makes it ideal for the study of vasomotor neurotransmission (Hirst, 1977; Hirst and Neild, 1978). The function of the extrinsic sympathetic nerves is primarily to reduce blood flow to the intestine. The sympathetic nerves supplying the gastrointestinal blood vessels are tonically active and they participate in cardiovascular homeostasis by altering intestinal vascular resistance to meet regional and whole body demands. Stimulation of the sympathetic nerves to the submucous arterioles causes a transient depolarisation of the arteriolar smooth muscle membrane potential, which is referred to as an excitatory junction potential (EJP). That these EJPs result from nerve stimulation has been demonstrated by their abolition by tetrodotoxin (TTX), which prevents propagation of action potentials by blocking Na+ channels, and by their absence when the arterioles have been extrinsically denervated (Hirst, 1977). During trains of stimuli, the EJPs summate to exceed a threshold, triggering an action potential which leads to constriction of the arteriole (Hirst, 1977). This increases the vascular resistance and decreases blood flow to the intestine. There is now plenty of evidence to suggest that noradrenaline (NA) and ATP act as co-transmitters, being released from sympathetic nerves in variable proportions depending on the tissue, the species and on the parameters of stimulation (for a review see Burnstock and Ralevic, 1994). Considerable variation exists in the ratio of NA:ATP released in different vessels. ATP appears to be the major component of sympathetic co-transmission (via P2X purinoceptors) in the intestinal arterioles of the guinea-pig and the function of NA is to act prejunctionally on α2 receptors to decrease the amount of transmitter released (Evans and Surprenant, 1992) or to cause a slow depolarisation accompanied by a slow contraction, via its action on extrajunctional α1 receptors. Additionally, ATP has a direct action on P2Y purinoceptors on the endothelium to cause the release of NO, thus effecting vasodilatation (Ralevic and Burnstock, 1988). Neuropeptide Y (NPY) is also found in most sympathetic nerves, including all sympathetic nerves supplying vascular smooth muscle (Uddman et al., 1985; McLachlan and Llewellyn, 1986; Morris et al., 1986). NPY has no vasoconstrictor effect on the submucous arterioles unlike other vascular beds (Neild and Kotecha, 1990). Normally NPY potentiates the actions of neurogenic or exogenously applied vasoconstrictors to the arterioles via its actions on the Y1 receptors (Xia, Neild and Kotecha, 1992). Alternatively, NPY can cause inhibition via the Y2 receptors localised on the smooth muscle membrane (Neild and Lewis, 1995). NPY can also modulate vasodilator neurotransmission by decreasing release from intrinsic vasodilator nerves via prejunctional Y2 receptors (Kotecha, 1998). Also both NA (via α2 receptors) and NPY can modulate release from sensory nerves (Kawasaki et al., 1990b, 1991). Hence, there is great scope for intercommunication between the sympathetic, sensory and intrinsic vasodilator nerves, in that NPY released during sympathetic nerve stimulation will not only potentiate the effects of constrictors that are co-released but also attenuate vasodilator neurotransmission, thus enhancing the effects the constrictors. On the other hand, direct effect of NPY on the muscle Y2 receptors would cause inhibition of the vascular

346

INNERVATION OF THE GASTROINTESTINAL TRACT

Figure 8.1 Schematic representation of putative interactions between sympathetic, sensory and intrinsic nerves and their possible effects on the vascular smooth muscle and endothelium of submucosal arterioles.

smooth muscle. This latter role could be fulfilled by release of NPY from the NPY neurones in the submucosal vasodilator ganglia. These neurones project to the arterioles and NPY released by them could also act prejunctionally on the sympathetic nerve terminals to decrease the size of the excitatory junction potentials evoked by sympathetic stimulation (Kotecha, 1998; Neild and Kotecha, 1990). Figure 8.1 shows an overview of putative interactions between the extrinsic sympathetic and sensory nerves with the intrinsic nerves, and their effect on submucosal arterioles. EXTRINSIC SENSORY NERVES Primary afferent sensory nerves originating from the dorsal root ganglia project along the mesenteric arteries to enter the intestine and provides a perivascular network of fibres around the arterial system. The neuropeptides substance P (SP) and calcitonin generelated peptide (CGRP), potent vasodilators in many systems, are the principal transmitters in primary afferent nerves and have been shown to co-exist in the same perivascular terminals (Gibbins et al., 1985). Apart from the control exhibited by intrinsic nerves and extrinsic sympathetic nerves, extrinsic sensory nerves can also influence arteriolar tone. The majority of these fibres continue to run along the arterioles of the submucous plexus, ultimately sending branches which terminate in the mucosa and around some neurones in

NEURAL CONTROL OF INTESTINAL VESSELS

347

the ganglia of the submucous plexus. Several groups (Lundgren, Svanvik and Jivegard, 1989; Rozsa and Jacobson, 1989; Hottenstein et al., 1991) have proposed a role for extrinsic sensory nerves in the regulation of intestinal blood flow in vivo. For example, Kawasaki et al. (1991) have shown that, in the rat, stimulation of these nerves produces a substantial vasodilatation in the mesenteric arteries which can be antagonised by a fragment of CGRP, implying that CGRP is a vasodilator neurotransmitter. Further evidence for the participation of sensory nerves in the control of arteriolar tone in the submucosa comes from the study of Vanner (1993) who used capsaicin to excite the sensory nerves. Vanner (1993) found that preconstricted guinea-pig submucosal arterioles relaxed when exposed to capsaicin, suggesting that extrinsic sensory nerves provide an additional neural pathway within the submucosal plexus for modulation of blood flow. The exact mechanism via which capsaicin causes its effects is unclear but integrity of sensory nerves is a prerequisite for its actions as arterioles that have been extrinsically denervated do not respond to capsaicin (Vanner, 1993). In the submucous arteriolar preparation, small constrictions elicited by perivascular nerve stimulation are augmented in the presence of human CGRP(8–37) (unpublished personal observations), suggesting that CGRP is released from the sensory nerves running along the arteriole to inhibit vasoconstriction. Deactivation of the sensory nerves using capsaicin also results in augmentation of neurogenic constrictions; this potentiation is accompanied by an increase in the amplitude of the evoked EJPs suggesting that there is an increase in the amount of transmitter released from the sympathetic nerves (Coffa and Kotecha, 1999; Kotecha and Neild, 1995b). The corollary of this is that in in vitro experiments, sensory nerves are normally stimulated along with sympathetic nerves in the perivascular nerve bundle, to inhibit transmitter release from the sympathetic nerves. Although both the sensory neuropeptides (i.e. SP and CGRP) have been implicated in the modulation of arteriole tone in the intestinal microcirculation (Vanner, 1994), their mode of action is different. It is unlikely that neurally released SP, which is known to be a totally endotheliumdependent vasodilator, can reach the endothelium in sufficient quantity to effect inhibition of the vascular smooth muscle. The most likely role of neurogenic SP is to inhibit transmitter release from the sympathetic nerves, whilst the postjunctional actions of sensory nerves are mediated by CGRP (Coffa and Kotecha, 1999). Several mechanisms underlying the control of blood flow by sensory nerves have been implicated and may differ between species and different vascular beds of the same species. These mechanisms include postsynaptic action on the smooth muscle membrane to cause inhibition (Han, Naes and Westfall, 1990; Kawasaki et al., 1990a; Remak, Hottenstein and Jacobson, 1990), excitatory effect on intrinsic vasodilator nerves (Vanner and MacNaughton, 1995), inhibitory effect on the sympathetic nerves (Coffa and Kotecha, 1999), effect on the endothelium to release NO (Hill and Gould, 1995) or antidromic excitation of motor collaterals of sensory nerves (axon reflex; Lundgren, Svanvik and Jivegard, 1984; Rozsa and Jacobson, 1989). All of these mechanisms may be involved in the control of intestinal vessels and the particular neural circuits involved may depend on the stimulus. For example, Meehan, Hottenstein and Kreulen (1991) have shown that electrical stimulation of sensory nerves produces an inhibitory junction potential, which was sensitive to capsaicin and TTX, in guanethidine-treated mesenteric arterial smooth muscle, suggesting a postjunctional site of action. Sensory nerve mediated recovery of

348

INNERVATION OF THE GASTROINTESTINAL TRACT

blood flow after sustained sympathetic nerve stimulation has been demonstrated in the rat superior mesenteric artery (Remak, Hottenstein and Jacobson, 1990) and may involve an endothelium-dependent mechanism similar to that seen in the rat irideal arterioles (Hill and Gould, 1995). Immunological studies have shown that extrinsic sensory nerve fibres project not only to the arterioles but also to the submucosal and myenteric plexuses (Gibbins, Furness and Costa, 1987). Thus, there is scope for extrinsic sensory nerves to excite submucosal vasodilator neurones that project to the arterioles. Vanner (1993) ruled out the possibility that sensory neurones could excite intrinsic cholinergic neurones on the basis that capsaicin-induced vasodilatations could not be prevented by the muscarinic antagonist, 4-diphenylacetoxy methylpiperidine (4-DAMP). However, Kotecha and Neild (1995a) have shown that although intrinsic cholinergic neurones are the main candidates for vasodilator effects of submucosal ganglion stimulation, there is clearly a role for other neurones. In fact, Rozsa and Jacobson (1989) have demonstrated that bile-oleate-induced intestinal vasodilatation involves primary afferent nerve fibres of the gut that release vasoactive intestinal peptide (VIP), and this release is not via nicotinic cholinergic synapses. Also sensory nerves are subject to modulation by NA and NPY released from the sympathetic nerve terminal (Kawasaki et al., 1990b; 1991). In turn, sensory nerves can modulate transmitter release from sympathetic nerves (Coffa and Kotecha, 1999). Whatever the mechanism by which sensory nerves have their effect, this neural pathway may contribute to responses evoked by different stimuli that transit through the intestinal lumen. INTRINSIC NERVES The arterioles of the small intestine are supplied with nerves that originate in the submucosal plexus of the ENS. In 1975, Hirst and McKirdy recorded the first synaptic potentials from the neurones of the guinea-pig submucous plexus. The presence of putative vasodilator transmitters in these neurones of the submucous ganglia (Furness and Costa, 1987) prompted Neild, Shen and Surprenant (1990) to investigate the effects of stimulating these ganglia on the arterioles of the submucosa. The anatomy of the ENS has been studied extensively but only in the guinea-pig. The submucous ganglia of the guinea-pig intestine contain at least three types of neurones which project to the submucous arterioles (Galligan, Costa and Furness, 1988; Brookes, Steele and Costa, 1991). They can be distinguished by their characteristic pattern of immunoreactivity for a variety of peptides. One type shows immunoreactivity for VIP, galanin (GAL) and dynorphin (DYN), and makes up 45% of the neurones in the plexus. Another shows immunoreactivity for SP and choline acetyl transferease (ChAT) and makes up 11% of the total neurone count, and the third shows immunoreactivity for calretinin and ChAT and makes up 12% of the total (Brookes, Steele and Costa, 1991). Recent work done in our laboratory (unpublished) has found that the neurones containing NPY also appear to project to the submucous arterioles. In addition to NPY, these neurones, which are known to project to the mucosa, contain ChAT, CCK, CGRP, DYN(1–8), somatostatin and possibly GAL (Song et al., 1992). All neurones are distributed throughout the submucous ganglia and will be referred to as VIP, SP, NPY and calretinin neurones. The vasodilator effects of these neurones can be demonstrated in isolated preparations of the submucosal plexus of the guinea-pig (Neild, Shen and Surprenant, 1990; Kotecha and Neild, 1995a). Many of the substances

NEURAL CONTROL OF INTESTINAL VESSELS

349

found in the neurones that project to the arterioles can cause vasodilatation in these and other arteries. However, the presence of multiple messengers in neurones does not imply that all of them subserve a transmitter role and it is not possible to decide from histochemical investigation alone if the innervation is efferent or afferent. Acetylcholine (ACh) appears to be the main mediator of this vasodilatation but others are clearly involved (Kotecha and Neild, 1995a); vasodilatation to 48% of the ganglia tested were completely abolished by the muscarinic antagonist, pirenzepine, 31% were partly abolished and 21% were not affected; as the individual ganglia contains from one to ten neurones (Furness and Costa, 1987), it was not surprising that some ganglia did not contain a cholinergic neurone. Vanner and Surprenant (1991) have also found cholinergic and non-cholinergic components, attributed to SP and VIP, of vasodilatation (in response to submucosal ganglionic stimulation) in the submucous arterioles of the guinea-pig colon. The neurones responsible for the cholinergic component of vasodilatation in the submucous arterioles of the ileum, could be the calretinin-containing neurones (Brookes, Steele and Costa, 1991) or the putative NPY neurones mentioned above. The only other neurone type in the ganglia known to contain ChAT are the SP neurones but these are unlikely to be involved as they cannot be stimulated at a rate greater than 1 Hz (Bornstein, Furness and Costa, 1989) but the vasodilator effect produced by ganglion stimulation was graded with stimulus frequency up to 10 Hz (Neild, Shen and Surprenant, 1990). Additionally, there appears to be no evidence for the involvement of SP in the dilator response in normal preparations, although it could be demonstrated in arterioles that had been extrinsically denervated for more than 30 days (Galligan et al., 1990) and the source of this neurally released SP appears to be the myenteric neurones (Jiang and Surprenant, 1992). Brunsson et al. (1995) have shown that the effects of SP in feline small intestine were via release of VIP. CHOLINERGIC TRANSMISSION Cholinergic neurones have several important roles. Cholinergic secretomotor neurones augment water and electrolyte secretion. They supply excitatory input to gastric acid secreting cells and to the external muscle and some function as interneurons. As mentioned before, ACh is the main mediator of arteriolar vasodilatation to submucosal ganglionic stimulation (Neild, Shen and Surprenant, 1990; Kotecha and Neild, 1995a). In the rat skin, ACh activates sensory nerves to cause its effect (Ralevic et al., 1992). However, Neild, Shen and Surprenant (1990) did not see a block of response to ACh when the submucosal arterioles were extrinsically denervated to get rid of the sensory nerves, suggesting that the cholinergic response in the submucous arterioles is not mediated via the sensory nerves. The endothelium plays a crucial role in mediating the vasodilator response to ACh and it has been shown that neurogenic ACh released from submucosal ganglia reaches the endothelium to mediate its effect via the release of nitric oxide (NO) (Andriantsitohaina and Surprenant, 1992; Kotecha and Coffa, 1999). Ach-induced vasodilatation is accompanied by a rapidly developing hyperpolarization of the smooth muscle (Kotecha and Neild, 1995a) similar to the hyperpolarization caused by NO in other arteries (Tare et al., 1990). Whilst a NO-independent hyperpolarization (Hashitani and Suzuki, 1997) and a direct inhibitory effect on the intestinal arteriole smooth muscle (Kotecha, 1999) can be

350

INNERVATION OF THE GASTROINTESTINAL TRACT

demonstrated with exogenous ACh, it is clear that all of the inhibitory effects of neurogenically-released ACh on the intestinal arterioles are mediated via the release of NO from the vascular endothelium (Kotecha and Coffa, 1999). Additionally, the role of AChinduced hyperpolarization in modulating tone is also not clear since NO appears to be almost exclusively implicated in the dilator response of these arterioles to ACh when constricted using exogenously applied U46619 (Andriantsitohaina and Surprenant, 1992) or phenylephrine (Coffa and Kotecha, 1999). Recently Plane and Garland (1996) have demonstrated that vasodilator responses can be changed dramatically from NO to EDHF just by changing the mode of induction of tone. Contribution of Ca2+ from internal stores in response to different constrictor agents may vary in the different vascular beds (Low et al., 1996). As such, it may be envisaged that EDHF might be more effective against a contractile agonist that relies on voltagedependent Ca2+ influx, such as ATP (Reilly and Hirst, 1996). Although the ACh induced hyperpolarization (in conjunction with release of endothelial NO) may underlie the inhibition of applied ATP constrictions, these factors fail to account for the inhibition of neurogenic ATP constrictions (Kotecha, 1999). There is clearly another powerful inhibitory mechanism involved. Therefore, it seems that although endothelial factors can account for inhibition of constrictions mediated via exogenously applied constrictors, there appears to be a limited role for the physiological effectiveness of the endothelium in situations where the contractile state of the arteriolar muscle is determined by the level of sympathetic nerve activity. Hence, apart from a direct postjunctional effect of vasodilator nerves, an additional mechanism that needs to be considered is the neural cross talk between the vasodilator nerves and the extrinsic sympathetic nerves. Vasodilator nerves can regulate sympathetic activity by influencing the release of transmitter from the sympathetic nerves, as shown by a decrease in the amplitude of EJP (Kotecha and Neild, 1995b) probably by an action of ACh on the prejunctional M2 receptors on the sympathetic nerve terminals (Kotecha and Neild, 1993). The structure of the sympathetic nerves around the guinea-pig submucosal arterioles has been examined in detail (Luff, McLachlan and Hirst, 1987). No specialised junctions between nerves have been seen that would account for the interaction between vasodilator and sympathetic nerves, although close approaches between nerve axons in paravascular nerve bundles are common (Luff, McLachlan and Hirst, 1987). The putative vasodilator nerves that run in these bundles are immunoreactive for ChAT (Brookes, Steele and Costa, 1991). Recent work (Kotecha, 1999) has confirmed that exogenously applied ACh, at low concentration, can significantly reduce the amplitude of the EJPs evoked by stimulation of the perivascular nerves, without significantly altering the time constant of decay of EJPs. This confirms ACh to be at least one of the mediators involved in reduction of transmitter release from sympathetic nerves. Decrease in sympathetic outflow mediated via prejunctional muscarinic receptors has been known for some time (Eglen and Whiting, 1990; Fernandes et al., 1991; Komori and Suzuki, 1987; Kotecha and Neild, 1993). ACh-mediated inhibition of neurogenic constrictions have been compared with exogenous ATP constrictions, in the presence of agents that prevent the actions of NO and prostanoids. ACh causes much more potent inhibiton of neurogenic constrictions than ATP-constrictions, suggesting that ACh-induced hyperpolarization is not accountable. This implicates a prejunctionally mediated action of ACh on the sympathetic

NEURAL CONTROL OF INTESTINAL VESSELS

351

transmission (Kotecha, 1999). Hence, there appears to be a very powerful role of prejunctional inhibitory mechanism and the one that appears to be physiologically important. In conclusion, although release of endothelial factors in response to ACh can be demonstrated, we must be cautious about its importance and question its physiological role in the vascular bed that is being investigated. It is also clear that prejunctional mechanisms can be very powerful and it is possible that in this vascular bed only prejunctional effects modulate neurogenic constriction whereas endothelial factors may be important when looking at circulating constrictor agents. NON-CHOLINERGIC TRANSMITTERS Vasodilatation of the arteriole that was not sensitive to the muscarinic antagonist pirenzepine could be divided into two types; one that was not accompanied by hyperpolarization (attributable to VIP) and the other that was associated with a hyperpolarization (Kotecha and Neild, 1995a). A few experiments have been conducted in which dilatation has been obtained by intracellular stimulation of a single neurone, rather than a whole ganglion (Bornstein and Neild, unpublished; Surprenant, unpublished). The neurone has been marked with dye and later processed for immunohistochemical identification of peptides. The one clear result that has emerged from these few experiments is that VIP cells can cause vasodilatation. Also, stimulation of the whole ganglion can release VIP to cause vasodilatation as deduced from comparison of the smooth muscle membrane potential profile obtained during vasodilatation obtained by exogenous application of VIP and that obtained when the ganglion is stimulated. These dilatations are not associated with hyperpolarization of the smooth muscle membrane and are insensitive to muscarinic receptor antagonist, pirenzepine (Kotecha and Neild, 1995a). Itoh et al. (1985) also found that VIP caused no change in membrane potential in the rabbit mesenteric artery, although Standen et al. (1989) found that VIP opened potassium channels in isolated cells from the rat and rabbit mesenteric arteries. However, in both rabbit mesenteric arteries (Itoh et al., 1985) and cat cerebral arteries (Edvinsson et al., 1985) VIP caused an increase in cyclic adenosine monophosphate (cAMP) production which could cause vasodilatation without a change in membrane potential. VIP axons are associated with small blood vessels, mostly in the submucosa and the mucosa, throughout the gastrointestinal tract. VIP is a potent dilator of intestinal blood vessels (Furness and Costa, 1987). VIP is also involved in the mediation of vasodilator response of submucous arterioles in the guinea-pig colon to ganglion stimulation (Vanner and Surprenant, 1991). Vasodilator responses associated with pirenzepine-insensitive hyperpolarization were presumably mediated by GAL, DYN or some other unidentified substance whose vasodilator effects and effects on membrane potential have not been characterised (Kotecha and Neild, 1995a). Both GAL and DYN cause hyperpolarization of a similar magnitude and time course and cannot be readily distinguished from each other (Kotecha and Neild, 1995a). GAL has been reported to cause a small increase in K+ conductance in mudpuppy parasympathetic (Parsons and Konopka, 1991) and cultured RINm5f cells (an insulin-secreting pancreatic cell line; Homaidan, Sharp and Nowak, 1991), but in both these tissues its major action was to reduce the current flowing through voltage-sensitive

352

INNERVATION OF THE GASTROINTESTINAL TRACT

Ca++ channels. This may also be the basis of its vasodilator action in the submucous arterioles. Among other substances found in the putative NPY neurones that project to the arteriole, CCK is capable of producing vasodilatation. Somatostatin increases absorptive fluxes in small and large intestine and reduces secretion that has been enhanced by theophylline, prostaglandin, 5-hydroxytryptamine (5-HT) or VIP (Furness and Costa, 1987). The roles of CCK and somatostatin in mediating responses to the submucous neurone stimulation have not been studied. There is also a possibility that NO nerves, originating in the enteric plexuses, mediate control of intestinal vessels. Yu, Li and Deng (1993) suggested that the effect of bradykinin in perfused rat mesentery may be mediated via NO released from nerves. More recently, Gyoda et al. (1995) have shown the presence of NADPH-diaphorase positive nerves and demonstrated that electrical field stimulation of the guinea-pig superior mesenteric artery induced a vasodilator response that was mediated by NO and CGRP. It is not clear at this stage whether such nerves are present on the small arteries and arterioles of the intestine although several workers have reported the presence of NO nerves in the myenteric and submucosal plexus and around submucosal arterioles of the monkey and human digestive system. NO-nerves have also been described in the ganglia of the two plexuses and within the blood vessels throughout the guinea-pig small and large intestines but, as yet, there is no indication of perivascular NO nerve fibres (Nichols, Krantis and Staines, 1992; McConalogue and Furness, 1993; De Giorgio et al., 1994). Release of NO from autonomic nerves and myenteric ganglia of the guinea-pig has also been demonstrated (Grider and Jin, 1993; Wiklund et al., 1993). It is not clear at this stage what role neurogenic NO plays in the regulation of intestinal blood flow, although NO is an important mediator for other endogenous vasodilator substances which act on the endothelium. Table 8.2 summarises the actions of the various transmitters involved in the control of intestinal blood vessels and the receptor types, where known.

TABLE 8.2 Summary of putative transmitters of intestinal nerves and their actions. Transmitter Sympathetic neurones ATP NA

Receptor

Putative actions

P2X P2Y α1

Mediator of arteriolar constriction Release of NO from endothelium Mediator of slow arteriolar constriction May be involved in autoregulatory escape Prejunctional inhibition of transmitter from sympathetic and sensory nerves Potentiator of vasoconstrictors Inhibition of arteriolar smooth muscle Prejunctional inhibition of submucosal and sensory nerves

α2 NPY

Sensory neurones SP CGRP and SP

Y1 Y2

NK1

Inhibition of transmitter release from sympathetics Mediators of dilatation in arterioles

NEURAL CONTROL OF INTESTINAL VESSELS

353

TABLE 8.2 Continued Transmitter

Receptor

Putative Actions Mediators of axon reflex Prejunctional excitation of submucous neurones

Submucous neurones ACh

Dynorphin Galanin NPY VIP

M3 M2

Y2

Release of NO from endothelium Inhibition of transmitter release from sympathetics Mediator of dilatation in arterioles Mediator of dilatation in arterioles Mediator of dilatation in arterioles Inhibition of transmitter release from sympathetics Mediator of dilatation in arterioles

NORMAL CONTROL OF INTESTINAL VESSELS The neural control of blood flow to the intestine is targeted towards maintaining mucosal integrity and meeting the needs for secretion and absorption. AUTOREGULATORY ESCAPE Increase in the resistance of a vascular bed, in the presence of a continuous vasoconstrictor stimulus, is often not maintained. Spontaneous relaxation (i.e. “autoregulatory escape”) appears to result from redistribution of blood flow within the intramural vessels. The decreased resistance is due to a partial dilatation of the small arteries and arterioles (Ross, 1971; Marshall, 1982). The vasoconstrictor stimulus may be neural (Folkow, Öberg and Rubinstein, 1964; Greenway, 1984) due to exposure to a vasoconstrictor substance (Ross, 1971). The presence of an autoregulatory escape from the influence of vasoconstrictor stimuli has been demonstrated in several species and appears to be a property of small arteries and arterioles (Fara, 1971; Duling et al., 1981). Several mechanisms (metabolic, myogenic, tissue-pressure) probably underlie the autoregulatory escape mechanisms (Lundgren, 1984) and it is not possible to quantify the contribution of these mechanisms. However, it is of interest to survey the evidence for a neural contribution. The relaxation of the submucous arteriole during an escape is preceded by repolarisation of the membrane potential (Neild and Kotecha, 1989). Neild and Kotecha (1989) proposed that NA (via α receptors) activated adenylate cyclase, leading to a rise in cAMP, favouring Ca2+ uptake into internal stores. Intracellular Ca2+ levels would then fall, thus closing calcium-dependent chloride channels, leading to hyperpolarization and vasodilatation. Additionally, ATP could act on P2Y purinoceptors on the endothelium to release NO, thus mediating vasodilatation (Ralevic and Burnstock, 1988). It is also possible that sensory nerves may be involved in this autoregulation since they appear to have a role in recovery of blood flow after sustained constriction by perivascular nerve stimulation in rat superior mesenteric artery (Remak, Hottenstein and Jacobson, 1990). Demonstration of Y2 receptors

354

INNERVATION OF THE GASTROINTESTINAL TRACT

on the arteriolar muscle of the submucosa (Neild and Lewis, 1995) opens up another possibility for autoregulatory escape mechanism, since NPY, released from sympathetic nerves, could act on these Y2 receptors to cause inhibition. Escape may be physiologically important mechanism for avoiding tissue ischaemia during extreme vasoconstrictor stimuli and it would not be surprising if various complex neural control mechanisms are in place to fulfil this need. POSTPRANDIAL REFLEX VASODILATATION IN THE SMALL INTESTINE Digestion is an energy-consuming process that increases the oxygen demands of the small intestine. The entry of partly digested food and bile into the small intestine provokes dilatation of mucosal arterioles that can double blood flow to the mucosa (see Lundgren, 1984). This effect has been ascribed to local reflexes, hormones, and vasoactive metabolic products as it can be demonstrated in segments of intestine that have been acutely extrinsically denervated. The ENS aims at adjusting blood flow according to the metabolic demands of the tissue and therefore it seems logical to presume that the ENS would, at least in part, mediate change in the intestinal haemodynamics postprandially. There is plenty of evidence now to suggest that local neural reflex mechanisms may be responsible for the postprandial increase in blood flow in the ileum and the proximal colon, whereas the reflex vasodilatation evoked in the distal colon is mediated via the pelvic nerves (Biber, Lundgren and Svanvik, 1971; Fasth et al., 1977). Local neural reflex mechanisms would require all the neural elements, i.e. sensory, inter and motor neurones, to be present in an isolated piece of intestine. These reflexes can be divided into two types depending on the sensory neurones that mediate the signal transduction and the type of stimulus that evokes the reflex (i.e. chemical or mechanical). Both vasodilator reflexes are probably involved in normal digestive processes and although distinct populations of sensory neurones may be activated, these probably converge on the same subset of enteric motor neurones effecting vasodilatation presumably via release of VIP. LOCAL INTRINSIC VASODILATOR REFLEX-ELICITED BY MECHANICAL STIMULATION All the elements necessary for a reflex are wholly intrinsic to the intestine and removal or destruction of extrinsic nerves has no bearing on the reflex response to the right stimulus. The intrinsic sensory neurone is likely to be the proposed SP, calbindin immunoreactive neurone which is located in the submucosal plexus (Kirchgessner, Tamir and Gershon, 1992). Mechanical stimulation of the mucosa of an extrinsically denervated intestinal segment can increase blood flow in the small bowel, often more than two-fold; this increase is sensitive to TTX, lidocaine (a local anaesthetic), and 5-HT receptor blocking agents suggesting that the vascular response was elicited via an intramural nervous reflex arch involving 5-HT receptors (Biber, Lundgren and Svanvik, 1971; Biber, Fara, and Lundgren, 1974). In addition, Biber, Fara and Lundgren (1973) showed that transmural electrical field stimulation gave a TTX-sensitive increase in intestinal blood flow which again was indicative of

NEURAL CONTROL OF INTESTINAL VESSELS

355

an intramural vasodilator reflex. Sensitivity of this reflex to 5-HT receptor antagonists has led many workers to suggest that 5-HT released from enterochromaffin cell is involved in mediating it (Lundgren, Svanvik and Jivegard, 1989; Vanner, Jiang and Surprenant, 1993). Both VIP and ACh have been implicated as neurotransmitters at the vascular smooth muscle effecting vasodilatation (Lundgren, Svanvik and Jivegard, 1989; Vanner, Jiang and Surprenant, 1993). AXON REFLEX-ELICITED BY CHEMICAL STIMULI In this type of reflex, the activated sensory neurones are extrinsic, with cell bodies in the dorsal root ganglia and axonal projections to the motor limb of the reflex. These reflexes still occur in a freshly isolated piece of intestine where the sensory terminals and the collaterals that these nerves send out to the motor limb of the reflex are functional. This reflex is sensitive to capsaicin, a sensory neurotoxin (Holzer, 1991), or to prior denervation of the sensory afferents if time is allowed for the sensory afferents to degenerate before testing vasodilator reflexes in an isolated preparation of ileum. Rozsa and Jacobson (1989) have showed that the vasodilator response in rats, to a bile/oleate mixture in the intestinal lumen, was almost abolished by capsaicin treatment, indicating a role for the dorsal roots sensory neurones. It has long been known that the stimulation of the same nerves results in release of significant quantities of VIP from the intestine, and this could be responsible for some vasodilatation. This prompted Rozsa and Jacobson (1989) to apply VIP antiserum to the small intestine and this caused a dose-dependent reduction in the vasodilator response to the bile/oleate mixture. This response was insensitive to hexamethonium, implying that the connection was not by a cholinergic synapse. As immunoreactivity for VIP has not been demonstrated in primary sensory fibres in most species, it appears that there must be a connection from sensory fibres to the VIP-containing neurones of the ENS via a non-nicotinic interneurone. However, there is a possibility that the motor collaterals of sensory afferents may release VIP, in view of reports by Maggi et al. (1989, 1990) that VIP might be present in sensory nerves of human isolated ileum. The action of cholera toxin on the small intestine also involves intestinal vasodilator neurones (Cassuto et al., 1983; Jiang et al., 1993) and direct involvement of VIP neurones is supported by the work of Jiang et al. (1993) who have proposed that VIP neurones are activated by the B subunit of cholera toxin to effect vasodilatation.

INTESTINAL DISORDERS Intestinal ischaemia results in hypoxic injury and death of cells or tissue layers and often kills the patient. Usually death is a result of irreversible shock and multisystem organ failure but severe intestinal ischaemia is the central pathophysiological event. A study by Jrvinen et al. (1994) has revealed that 82% of the 214 patients with acute intestinal ischaemia died over a period of 30 days after diagnosis; 86% of these patients had a prior history of cardiovascular disease. Since many of the intestinal ischaemic conditions are secondary to other life-threatening disorders, usually cardiovascular, there is often a delay in the diagnosis of this condition, which may be fatal. As most of the small intestine is

356

INNERVATION OF THE GASTROINTESTINAL TRACT

supplied by a single artery (i.e. the superior mesenteric artery), intestinal ischaemia results if blood flow to this artery is impaired, making the mucosa of the intestine susceptible to injury. As we have already seen, the intestinal vessels are supplied with a complex system of extrinsic and intrinsic neural pathways that work towards controlling the blood flow to meet the need for fluid secretion and absorption and metabolic demands of the muscle and mucosa. An aberration in any of these pathways may result in vascular insufficiency leading to undesirable clinical problems. For example, over-activity of the sympathetic component seen in some forms of hypertension will culminate in constriction of mesenteric blood vessels and sharply reduced blood flow. An early victim of ischaemic injury is the endothelial cell; this results in impaired control of vascular tone by neurogenic mechanisms that act via release of vasorelaxants from the endothelium, thus adding to the ischaemic insult. Decreased sensory nerve function in ageing animals may contribute to age-related alterations in circulatory haemodynamics in the rat mesentery (Li and Duckles, 1993) and this may be a contributory factor in age related intestinal disorders. Alterations in neurally mediated responses may also be the underlying cause of intestinal problems associated with some diseases. In diabetes, gastrointestinal neuropathy can involve virtually the entire length of the gastrointestinal tract. Clinical symptoms such as constipation and diarrhoea (Ross, 1993), which may be a result of a motility dysfunction may also be aggravated by abnormal fluid secretion or absorption and may indicate widespread abnormalities in extrinsic and intrinsic neural function. For example, Ralevic, Belai and Burnstock (1993) have demonstrated that vasodilator function of sensory-motor nerves in the mesenteric arterial bed of streptozotocin-induced diabetic rats is severely impaired. Since sensory nerves appear to form an integral part of the neural control of intestinal vessels, a deficit in these nerves could be expected to lead to many problems related to vascular insufficiency in the intestine. Thus, deficits in sensory nerve-mediated responses, including autoregulation, may culminate in ischaemia and tissue hypoxia. Changes in reflex vasodilatation in response to food, may underlie some of the gastrointestinal disorders associated with diabetes. It is interesting to draw parallels between disorders associated with neuropathy with results obtained in experiments where denervation has been part of the procedure. Neild, Shen and Surprenant (1990) showed that vasodilator response, in guinea-pig submucous arterioles, to submucosal ganglion stimulation was bigger after the arterioles had been extrinsically denervated. Galligan et al. (1990) showed that this neurogenically-mediated vasodilator response (after extrinsic denervation) was mediated by SP although SP was not involved in the neurogenic vasodilator response normally. Jiang and Surprenant (1992) later showed that this neurogenic SP response was mediated by myenteric neurones that had re-innervated the arterioles. In essence, this extrinsic denervation mimics diabetic neuropathy in which the sensory nerves are substantially affected. This speculation is supported by the work of Belai and Burnstock (1990) who showed that there was an increase in SP-like immunoreactive nerve fibres in the ileal myenteric plexus of 16 week streptozotocin-diabetic rats. Changes in the content of other neuropeptides have also been shown in both the submucosal and myenteric plexuses of diabetic rat ileum (Belai and Burnstock, 1990; Belai et al., 1993).

NEURAL CONTROL OF INTESTINAL VESSELS

357

It is tempting to speculate that in face of natural denervation, extrinsic and intrinsic, the tissue responds by developing pathways to compensate for the deficiency and this may be a normal remedial mechanism in diseased states, like diabetes, at least in the early stages of the disease. Pronounced changes in gut neuropeptide content and innervation patterns have also been observed in inflamed intestine of patients with inflammatory bowel disease but it is not clear whether these changes are due to altered synthesis and release from intrinsic and/ or extrinsic neurones and nerve fibres (Eysselein and Nast, 1991). Mayer, Raybould and Koelbel (1988) have suggested that neurogenic inflammation, elicited by activation of and release of mediators from primary afferent nerves, may be an important factor in inflammatory bowel disease. There is an ever-growing number of reports devoted to the study of neural impairment in various intestinal disorders (Hirschsprung’s disease, Crohn’s disease, ulcerative colitis) but how this relates to neural dysfunction in intestinal vascular tone has not been the focus of these studies. It is not within the scope of this review to assess how this might affect the control of tone in the intestinal vessels; however it stands to reason that altered haemodynamics in the intestinal circulation, related to neural dysfunction, may contribute to the pathogenesis of these disorders.

SUMMARY The precise coordination of mucosal function and smooth muscle activity necessary to maintain normal gut function is the prime function of the ENS. For optimum functioning, the blood flow to the intestine needs to be controlled so as to cope with the ever-changing requirement of the intestinal muscle, mucosa and digestive processes. The normal smooth interactions between the overlapping complex neural pathways – both extrinsic and intrinsic – that are involved in the control of intestinal blood vessels are responsible for adequate intestinal perfusion. Dysfunction of the bowel often involves both secretion/ absorption and motility. For example, bacterial toxins may stimulate sensory neurones to the mucosa which excite both secretomotor neurones and distension-sensitive neurones to promote motility. Stimulation of the secretomotor reflexes would lead to secretion of water and electrolytes, which in turn may lead to further activation of stretch reflexes. Although our knowledge of the pathways from sensory elements to physiological responses is quite extensive for peristaltic reflexes, it is somewhat fragmented when it comes to vasodilator reflexes. This is despite a dramatic development in our understanding of some the neural elements involved in controlling intestinal blood flow in the past couple of decades. Also the invaluable neurophysiological understanding of the normal control of intestinal vascular tone falls somewhat short of providing information about the neural basis of vascular tone under pathophysiological conditions. Substances present in the nerves may function as neurotransmitters leading to vasoconstriction or vasodilatation. They may also modulate the effect or release of other transmitters thus forming a means of communication between nerves. Additionally they may be involved in forming a “backup” system to ensure optimal blood flow or to compensate for selective neural deficiencies seen in diseased states.

358

INNERVATION OF THE GASTROINTESTINAL TRACT

Questions that need to be addressed include; Which substances are acting as neurotransmitters? Where are the sites of action of each of the transmitters involved? What are the roles of the various nerves in normal physiological and pathophysiological conditions? The next decade promises to be both exciting and challenging, as we piece together the jigsaw of neural control of intestinal blood vessels.

ACKNOWLEDGEMENTS I am deeply grateful to Dr Tim Neild and Professor Mollie Holman for providing useful criticism of the manuscript.

REFERENCES Andriantsitohaina, R. and Surprenant, A. (1992). Acetylcholine released from the guinea-pig submucosal neurones dilates arterioles by releasing nitric oxide from the endothelium. Journal of Physiology, 453, 493–502. Belai, A. and Burnstock, G. (1990). Changes in adrenergic and peptidergic nerves in the submucous plexus of streptozotocin-diabetic rat ileum. Gastroenterology, 98, 1427–1436. Belai, A., Facer, P., Bishop, A., Polak, J.M. and Burnstock, G. (1993). Effect of streptozotocin-diabetes on the level of VIP mRNA in myenteric neurones. Neuroreport, 4, 291–294. Biber, B., Fara, J. and Lundgren, O. (1973). Intestinal vasodilatation in response to transmural electrical field stimulation. Acta Physiologica Scandinavica, 87, 277–282. Biber, B., Fara, J. and Lundgren, O. (1974). A pharmacological analysis of intestinal vasodilatation mechanisms in the cat. Acta Physiologica Scandinavica, 90, 673–683. Biber, B., Lundgren, O. and Svanvik, J. (1971). Studies on the intestinal vasodilatation observed after mechanical stimulation of the mucosa of the gut. Acta Physiologica Scandinavica, 82, 177–190. Bornstein, J.C., Furness, J.B. and Costa, M. (1989). An electrophysiological comparison of substance P immuno reactive neurons with other neurons in the guinea-pig submucous plexus. Journal of the Autonomic Nervous System, 26, 113–120. Brookes, S.J.H., Steele, P.A. and Costa, M. (1991). Calretinin immunoreactivity in cholinergic motorneurones, interneurones and vasomotor neurones of the guinea-pig small intestine. Cell and Tissue Research, 263, 471–481. Brunsson, I., Fahrenkrug, J., Jodal, M., Sjoqvist, A. and Lundgren, O. (1995). Substance P effects on blood flow, fluid transport and vasoactive intestinal polypeptide release in the feline small intestine. Journal of Physiology, 483, 727–734. Burnstock, G. and Ralevic, V. (1994). New insights into local regulation of blood flow by perivascular nerves and endothelium. British Journal of Plastic Surgery, 47, 527–543. Cassuto, J., Siewert, A., Jodal, M. and Lundgren, O. (1983). The involvement of intramural nerves in cholera toxin induced intestinal secretion. Acta Physiologica Scandinavica, 117, 195–202. Coffa, F.P. and Kotecha, N. (1999). Modulation of sympathetic nerve activity by perivascular sensory nerves in the arterioles of the guinea-pig small intestine. Journal of the Autonomic Nervous System, 77, 125–132. De Giorgio, R., Parodi, J.E., Brecha, N.C., Brunicardi, F.C., Becker, J.M., Go, V.L. et al. (1994). Nitric oxide producing neurons in the monkey and human digestive system. Journal of Comparative Neurology, 342, 619–627. Duling, B.R., Gore, R.W., Dacey, R.G. and Dawson, D.N. (1981). Methods for isolation, cannulation and in vitro study of single microvessels. American Journal of Physiology, 241, H108–H116. Edvinsson, L., Fredholm, B.B., Hamel, E., Jansen, I. and Verrcchia, C. (1985). Perivascular peptides relax cerebral arteries concomitant with stimulation of cyclic adenosine monophosphate accumulation or release of an endothelium derived relaxing factor in the cat. Neuroscience Letters, 58, 213–217.

NEURAL CONTROL OF INTESTINAL VESSELS

359

Eglen, R.M. and Whiting, R.L. (1990). Heterogeneity of vascular muscarinic receptors. Journal of Autonomic Pharmacology, 19, 233–245. Evans, R.J. and Surprenant, A. (1992). Vasoconstriction of guinea-pig submucosal arterioles following sympathetic nerve stimulation is mediated by the release of ATP. British Journal of Pharmacology, 106, 242–249. Eysselein, V.E. and Nast, C.C. (1991). Neuropeptide and inflammatory bowel disease. Zeitschrift für Gastroenterologie-Verhandlungsband, 26, 253–257. Fara, J.W. (1971). Escape from tension induced by noradrenaline or electrical stimulation in isolated mesenteric arteries. British Journal of Pharmacology, 43, 865–867. Fasth, S., Hultén, L., Lundgren, O. and Nordgren, S. (1977). Vascular responses to mechanical stimulation of the mucosa of the cat colon. Acta Physiologica Scandinavica, 101, 98–104. Fenger-Gron, J., Mulvany, M.J. and Christensen, K.L. (1995). Mesenteric blood pressure profile of conscious, freely moving rats. Journal of Physiology, 488, 753–760. Fernandes, F.A., Alonso, M.J., Marin, J. and Salaices, M. (1991). M3 muscarinic receptor mediates prejunctional inhibition of noradrenaline release and the relaxation in cat femoral artery. Journal of Pharmacy and Pharmacology, 43, 644–649. Folkow, B., Öberg, B. and Rubinstein, E.H. (1964). A proposed differentiated neuro-effector organisation in muscle resistance vessels. Angiologica, 1, 197–208. Furness, J.B. (1973). Arrangement of blood vessels and their relation with adrenergic nerves in the rat mesentery. Journal of Anatomy, 115, 347–364. Furness, J.B. and Costa, M. (1987). The Enteric Nervous System. London: Churchill Livingstone. Galligan, J.J., Costa, M. and Furness, J.B. (1988). Changes in surviving nerve fibres associated with submucosal arteries following extrinsic denervation of the small intestine. Cell and Tissue Research, 253, 647–656. Galligan, J.J., Jiang, M.-M., Shen, K.-Z. and Surprenant, A. (1990). Substance P mediates neurogenic vasodilatation in extrinsically denervated guinea-pig submucosal arterioles. Journal of Physiology, 420, 267–280. Gibbins, I.L., Furness, J.B., Costa, M., Macintyre, I., Hillyard, C.J. and Girgis, S. (1985). Co-localisation of calcitonin gene-related peptide-like immunoreactivity with substance P in cutaneous, vascular and visceral sensory neurons in guinea-pigs. Neuroscience Letters, 57, 125–130. Gibbins, I.L., Furness, J.B. and Costa, M. (1987). Pathway-specific patterns of the co-existence of substance P, calcitonin gene-related peptide, cholecystokinin and dynorphin in neurons of the dorsal root ganglia of the guinea-pig. Cell Tissue Research, 248, 417–437. Greenway, C.V. (1984). Neural control and autoregulatory escape. In Physiology of the Intestinal Circulation, edited by A.P. Shepherd and D.N. Granger, pp. 61–71. New York: Raven Press. Grider, J.R. and Jin, J.G. (1993). Vasoactive intestinal peptide release and L-citrulline production from isolated ganglia of the myenteric plexus: evidence for regulation of vasoactive intestinal peptide release by nitric oxide. Neuroscience, 54, 521–526. Gyoda, Y., Tsukada, Y., Saito, A. and Goto, K. (1995). Role of nitric oxide and neuropeptides in neurogenic vasodilatation of the guinea-pig mesenteric artery. European Journal of Pharmacology, 279, 83–92. Han, S.P., Naes, L. and Westfall, T.C. (1990). Calcitonin gene-related peptide is the endogenous mediator of nonadrenergic-noncholinergic vasodilatation in rat mesentery. Journal of Pharmacology and Experimental Therapeutics, 255, 423–428. Hashitani, H. and Suzuki, H. (1997). K+ channels which contribute to the acetylcholine-induced hyperpolarization in smooth muscle of the guinea-pig submucosal arteriole. Journal of Physiology, 501, 319–329. Hill, C.E. and Gould, D.J. (1995). Modulation of sympathetic vasoconstriction by sensory nerves and nitric oxide in rat irideal arterioles. Journal of Pharmacology and Experimental Therapeutics, 273, 918–926. Hirst, G.D.S. (1977). Neuromuscular transmission in arterioles of the guinea-pig submucosa. Journal of Physiology, 273, 263–275. Hirst, G.D.S. and McKirdy, H.C. (1975). Synaptic potentials recorded from neurons of the submucous plexus of guinea-pig small intestine. Journal of Physiology, 249, 369–385. Hirst, G.D.S. and Neild, T.O. (1978). An analysis of excitatory junction potentials recorded from arterioles. Journal of Physiology, 280, 84–104. Holzer, P. (1991). Capsaicin: cellular targets, mechanism of action, and selectivity for thin sensory neurons. Pharmacological Reviews, 43, 143–201. Homaidan, F.R., Sharp, G.W. and Nowak, L.M. (1991). Galanin inhibits a dihydropyridine – sensitive Ca2+ current in the RINm5f cell line. Proceedings of the National Academy of Sciences USA, 88, 744–748. Hottenstein, O.D., Pawlik, W.W., Remak, G. and Jacobson, E.D. (1991). Capsaicin-sensitive nerves modulate resting blood flow and vascular tone in rat gut. Naunyn Schmiedebergs Archives of Pharmacology, 343, 179–184. Hultén, L., Jodal, M. and Lundgren, O. (1969). Extrinsic nervous control of colonic blood flow. Acta Physiologica Scandinavica Supplement, 335, 39–49.

360

INNERVATION OF THE GASTROINTESTINAL TRACT

Itoh, T., Sasafuri, T., Makita, Y., Kanmura, Y. and Kuriyama, H. (1985). Mechanisms of vasodilatation induced by vasoactive intestinal polypeptide in rabbit mesenteric artery. American Journal of Physiology, 249, H231–H240. Jiang, M.M. and Surprenant, A. (1992). Re-innervation of submucosal arterioles by myenteric neurons following extrinsic denervation. Journal of The Autonomic Nervous System, 37, 145–154. Jiang, M., Kirchgessner, A., Gershon, M.D. and Surprenant, A. (1993). Cholera toxin-sensitive neurons in guinea-pig submucosal plexus. American Journal of Physiology, 264, G86–G94. Jodal, M. and Lundgren, O. (1986). Countercurrent mechanisms in the mammalian gastrointestinal tract. Gastroenterology, 91, 225–241. Jrvinen, O., Laurikka, J., Salenius, J.P. and Tarkka, M. (1994). Acute intestinal ischaemia. A review of 214 cases. Annals Chirurgiae et Gynaecologiae, 83, 22–25. Kawasaki, H., Nuki, C., Saito, A. and Takasaki, K. (1990a). Role of calcitonin gene-related peptide-containing nerves in the vascular adrenergic neurotransmission. Journal of Pharmacology and Experimental Therapeutics, 252, 403–409. Kawasaki, H., Nuki, C., Saito, A. and Takasaki, K. (1990b). Adrenergic modulation of calcitonin gene-related peptide (CGRP)-containing nerve-mediated vasodilatation in the rat mesenteric resistance vessels. Brain Research, 506, 287–290. Kawasaki, H., Nuki, C., Saito, A. and Takasaki, K. (1991). NPY modulates neurotransmission of CGRP-containing vasodilator nerves in rat mesenteric arteries. American Journal of Physiology, 261, H683–H690. Kirchgessner, A.L., Tamir, H. and Gershon, M.D. (1992). Identification and stimulation by serotonin of intrinsic sensory neurons of the submucosal plexus of the guinea-pig gut: Activity-induced expression of Fos immunoreactivity. The Journal of Neuroscience, 12, 235–248. Komori, K. and Suzuki, H. (1987). Heterogenous distribution of muscarinic receptors in the rabbit saphenous artery. British Journal of Pharmacology, 92, 657–664. Kotecha, N. (1998). Modulation of submucosal arteriolar tone by neuropeptide Y Y2 receptors in the guinea-pig small intestine. Journal of the Autonomic Nervous System, 70, 157–163. Kotecha, N. (1999). Mechanisms underlying ACh induced modulation of neurogenic and applied ATP constrictions in the submucosal arterioles of the guinea-pig small intestine. British Journal of Pharmacology, 126, 1625–1633. Kotecha, N. and Coffa, F.P. (1999). Modulation of sympathetic nerve activity by perivascular sensory nerves in the arterioles of the guinea-pig small intestine. American Journal of Physiology, 277, H1441–H1446. Kotecha, N. and Neild, T.O. (1993). Muscarinic M2 and M3 receptors may mediate vasodilator action of submucous neurons on the submucous arterioles of the guinea-pig small intestine. Proceedings of the Australian Physiological and Pharmacological Society, 24, 193P. Kotecha, N. and Neild, T.O. (1995a). Vasodilatation and smooth muscle membrane potential changes in arterioles from the guinea-pig small intestine. Journal of Physiology, 482, 661–667. Kotecha, N. and Neild, T.O. (1995b). Actions of vasodilator nerves on arteriolar smooth muscle and neurotransmitter release from sympathetic nerves in the guinea-pig small intestine. Journal of Physiology, 489, 849–855. Li, Y. and Duckles, S.P. (1993). Effect of age on vascular content of calcitonin gene-related peptide and mesenteric vasodilator nerve activity in the rat. European Journal of Pharmacology, 236, 373–378. Low, A.M., Kotecha, N., Neild, T.O., Kwan, C.Y. and Daniel, E.E. (1996). Relative contributions of extracellular Ca2+ and Ca2+ stores to smooth muscle contraction in arteries and arterioles of rat, guinea-pig, dog and rabbit. Clinical and Experimental Pharmacology and Physiology, 23, 310–316. Luff, S.E., McLachlan, E.M. and Hirst, G.D.S. (1987). An ultrastructural analysis of the sympathetic neuromuscular junctions on arterioles of the submucosa of the guinea-pig ileum. Journal of Comparative Neurology, 257, 578–594. Lundgren, O. (1984). Microcirculation of the gastrointestinal tract and pancreas. In The Handbook of Physiology; section 2: The Cardiovascular System: volume IV. Microcirculation, part 2, pp. 799–863. Bethesda, MA: American Physiological Society. Lundgren, O., Svanvik, J. and Jivegard L. (1989). Enteric Nervous System. Physiology and pathophysiology of the intestinal tract. Digestive Diseases and Sciences, 34, 264–283. Maggi, C.A., Santicioli, P., Del Bianco, E., Geppetti, P., Barbanti, G., Turini, D. et al. (1989). Release of VIPbut not CGRP-like immunoreactivity by capsaicin from the human isolated small intestine. Neuroscience Letters, 98, 317–320. Maggi, C.A., Giuliani, S., Santicioli, P., Patacchini, R., Said, S.I., Theodorsson, E. et al. (1990). Direct evidence for the involvement of vasoactive intestinal polypeptide in the motor response of the human isolated ileum to capsaicin. European Journal of Pharmacology, 185, 169–178. Marshall, J.M. (1982). The influence of sympathetic nervous system on individual vessels of the microcirculation of skeletal muscle of the rat. Journal of Physiology, 332, 169–186.

NEURAL CONTROL OF INTESTINAL VESSELS

361

Mayer, E.A., Raybould, H. and Koelbel, C. (1988). Neuropeptides, inflammation, and motility. Digestive Diseases and Sciences, 33, 71S–77S. McConalogue, K. and Furness, J.B. (1993). Projections of nitric oxide synthesizing neurons in the guinea-pig colon. Cell Tissue Research, 271, 545–553. McLachlan, E.M. and Llewellyn, I.J. (1986). The immunohistochemical distribution of neuropeptide Y in lumbar pre- and paravertebral sympathetic ganglia of the guinea-pig. Journal of The Autonomic Nervous System, 17, 313–324. Meehan, A.G., Hottenstein, O.D. and Kreulen, D.L. (1991). Capsaicin-sensitive nerves modulate inhibitory junction potentials and dilatation in guinea-pig mesenteric artery. Journal of Physiology, 443, 161–174. Mitchell, J.H. and Blomqvist, G. (1971). Maximal oxygen uptake. New England Journal of Medicine, 284, 1018–1022. Morris, J.L., Murphy, R., Furness, J.B. and Costa, M. (1986). Partial depletion of neuropeptide Y from noradrenergic perivascular and cardiac axons by 6-hydroxydopamine and reserpine. Regulatory Peptides, 13, 147–162. Neild, T.O. and Kotecha, N. (1989). A study of the phasic response of arterioles of the guinea pig small intestine to prolonged exposure to norepinephrine. Microvascular Research, 38, 186–199. Neild, T.O. and Kotecha, N. (1990). Actions of neuropeptide Y on arterioles of the guinea-pig small intestine are not mediated by smooth muscle membrane depolarization. Journal of the Autonomic Nervous System, 30, 29–36. Neild, T.O. and Lewis, C.J. (1995). Reduction of vasoconstriction mediated by neuropeptide Y Y2 receptors in arterioles of the guinea-pig small intestine. British Journal of Pharmacology, 115, 220–221. Neild, T.O., Shen, K.-Z. and Surprenant, A. (1990). Vasodilatation of arterioles by acetylcholine released from single neurons in the guinea-pig submucosal plexus. Journal of Physiology, 420, 247–265. Nichols, K., Krantis, A. and Staines, W. (1992). Histochemical localization of nitric oxide-synthesizing neurons and vascular sites in the guinea-pig intestine. Neuroscience, 51, 791–799. Parson, R. L. and Konopka, L.M. (1991). Analysis of the galanin-induced decrease in membrane excitability in mudpuppy sympathetic neurons. Neuroscience, 43, 647–660. Plane, F. and Garland, C.J. (1996). Influence of contractile agonists on the mechanism of endothelium-dependent relaxation in rat isolated mesenteric artery. British Journal of Pharmacology, 119, 191–193. Ralevic, V., Belai, A. and Burnstock, G. (1993). Impaired sensory-motor nerve function in the isolated mesenteric arterial bed of streptozotocin-diabetic and ganglioside-treated streptozotocin-diabetic rats. British Journal of Pharmacology, 110, 1105–1111. Ralevic, V. and Burnstock, G. (1988). Actions mediated by P2-purinoceptor subtypes in the isolated perfused mesenteric bed of the rat. British Journal of Pharmacology, 95, 637–645. Ralevic, V., Khalil, Z., Dusting, G.J. and Helme, R.D. (1992). Nitric oxide and sensory nerves are involved in the vasodilator response to acetylcholine but not calcitonin gene-related peptide in rat skin microvasculature. British Journal of Pharmacology, 106, 650–655. Reilly, M.J. and Hirst, G.D.S. (1996). Differences in the response to purinergic nerve stimulation and applied ATP in the guinea-pig vas deferens. Journal of the Autonomic Nervous System, 57, 93–100. Remak, G., Hottenstein, O.D. and Jacobson, E.D. (1990). Sensory nerves mediate neurogenic escape in rat gut. American Journal of Physiology, 258, H778–H786. Ross, G. (1971). Escape of mesenteric vessels from adrenergic and noradrenergic vasoconstriction. American Journal of Physiology, 221, 1217–1222. Ross, M.A. (1993). Neuropathies associated with diabetes. Medical Clinics of North America, 77, 111–124. Rozsa, Z. and Jacobson, E.D. (1989). Capsaicin-sensitive nerves are involved in bile-oleate-induced hyperemia. American Journal of Physiology, 256, G476–G481. Song, Z.M., Brookes, S.J.H., Steele, P.A. and Costa, M. (1992). Projections and pathways of submucous neurons to the mucosa of the guinea-pig small intestine. Cell and Tissue Research, 269, 87–98. Standen, N.B., Quayle, J.M., Davies, N.W., Brayden, J.E., Huang, Y. and Nelson, M.T. (1989). Hyperpolarizing vasodilators activate ATP-sensitive K+ channels in arterial smooth muscle. Science, 245, 177–180. Tare, M., Parkington, H.C., Coleman, H.A., Neild, T.O. and Dusting, G.H. (1990). Nitric oxide derived from the endothelium causes hyperpolarization and relaxation of arterial smooth muscle. Nature, 346, 69–71. Uddman, R., Ekblad, E., Edvinsson, L., Hakånson, R. and Sundler, F. (1985). Neuropeptide Y- like immunoreactivity in perivascular nerve fibres of the guinea-pig. Regulatory Peptides, 10, 243–257. Vanner, S. (1993). Mechanism of action of capsaicin on submucosal arterioles in the guinea pig ileum. American Journal of Physiology, 265, G51–G55. Vanner, S. and Surprenant, A. (1991). Cholinergic and noncholinergic submucosal neurons dilate arterioles in guinea pig colon. American Journal of Physiology, 261, G136–G144. Vanner, S., Jiang, M. and Surprenant, A. (1993). Mucosal stimulation evokes vasodilatation in submucosal arterioles by neuronal and nonneuronal mechanisms. American Journal of Physiology, 264, G202–G212. Vanner, S. (1994). Co-release of neuropeptides from capsaicin-sensitive afferents dilates submucosal arterioles in guinea pig ileum. American Journal of Physiology, 267, G650–G655.

362

INNERVATION OF THE GASTROINTESTINAL TRACT

Vanner, S. and MacNaughton, W.K. (1995). Capsaicin-sensitive afferent nerves activate submucosal secretomotor neurons in guinea pig ileum. American Journal of Physiology, 269, G203–G209. Wiklund, N.P., Leone, A.M., Gustafsson, L.E. and Moncada, S. (1993). Release of nitric oxide evoked by nerve stimulation in guinea-pig intestine. Neuroscience, 53, 607–611. Xia, J., Neild, T. and Kotecha, N. (1992). Effects of neuropeptide Y and agonists selective for neuropeptide Y receptor subtypes on arterioles of the guinea-pig small intestine and the rat brain. British Journal of Pharmacology, 107, 771–776. Yu, X.J., Li, Y.J. and Deng, H.W. (1993). The regulatory effect of bradykinin on the actions of sensory nerves in the perfused rat mesentery is mediated by nitric oxide. European Journal of Pharmacology, 241, 35–40.

9 Enteric Neuro-Immunophysiology Jackie D. Wood Department of Physiology, College of Medicine, The Ohio State University, 300 Hamilton Hall, 1645 Neil Avenue, Columbus, Ohio 43210, USA The intestine has an effective defence mechanism that purges the bowel of foreign threats through an integrated system of mast cells, the enteric nervous system (ENS) and effectors made-up of the musculature, secretory epithelium and blood vasculature. Enteric mast cells acquire and retain memory of antigenic threats as they appear and reappear in the digestive tract throughout a lifetime. Antibody-based memory enables mast cells to detect and respond to forbidden antigens whenever they appear. Insofar as mast cells send signals to the enteric nervous system, their function is analogous to sensory neurons. Like sensory neurons, mast cells detect and signal. They release chemical signals (e.g. histamine) that alert the ENS to the presence within the lumen of a potential threat. The ENS responds by calling-up from its library of programs a coordinated cascade of events that quickly purges the threat. The program starts by stimulating the secretion of H2O, electrolytes and mucous to flush the antigen from the cryptic depths of the mucosa and suspend it in solution in the lumen. This is followed by initiation of powerful propulsive motility that rapidly moves the secretions and suspended antigen toward the anal outlet. These defensive events are accompanied by the side effects of abdominal distress, urgency and diarrhoea. KEY WORDS: enteric nervous system; mast cells; histamine; neural programs; antigens.

INTRODUCTION The gastrointestinal tract exhibits patterns of behaviour that are characteristic for specific digestive and pathological states. Activity of the musculature, mucosal epithelium and blood vasculature are coordinated to produce these specific patterns. A neural program for each pattern is stored in the memory of a program library in the enteric nervous system (brain-in-the-gut). The programs are identified by their motility component. They are: (1) segmentation motility in the digestive state of the small intestine; (2) migrating motor complex in the interdigestive state of the small intestine; (3) orthograde power propulsion in the small and large intestine; (4) retrograde power propulsion in the small intestine during emesis; (5) haustral formation in the large intestine; (6) physiological ileus (Wood, 1995a,b). Selective secretory behaviour is organised by the program in concert with the motility pattern. 363

364

INNERVATION OF THE GASTROINTESTINAL TRACT

Activation of the program for orthograde power propulsion is a significant outcome of immuno-neural communication. This can occur after sensitisation of the enteric immune system to foreign antigens. Sensitisation may be caused by foreign antigens in the form of foodstuffs, toxins or invading organisms. After sensitisation, a second exposure to the same antigen triggers predictable integrated behaviour of the intestinal effector systems (e.g. Harari, Russell and Castro, 1987; Baird and Cuthbert, 1987). Neurally coordinated activity of the musculature, mucosa and blood vasculature results in organised behaviour of the whole intestine that rapidly expels the antigenic threat from the lumen. Recognition of an antigen by the sensitised immuno-neural apparatus activates the power propulsion program, the motor component of which is coordinated with copious secretion of water, electrolytes and mucous into the intestinal lumen, (Castro, 1989; Wang, Palmer and Cooke, 1991; Cowles and Sarna, 1990, 1991; Sarna et al., 1991). Detection by the enteric immune system and signalling to the enteric nervous system initiates the adaptive behaviour. Power propulsion is a specialised form of intestinal motility that forcefully and rapidly propels any material in the lumen over long distances to effectively empty the lumen. Its occurrence is accompanied by abdominal distress and diarrhoea in animal models and humans (Sethi and Sarna, 1991; Phillips, 1995). Output of the power propulsion program reproduces the same stereotyped motor behaviour in response to radiation exposure, mucosal contact with noxious stimulants or antigenic detection by the sensitised enteric immune system (Sarna et al., 1991). The neural program for power propulsion incorporates connections between submucous and myenteric divisions of the enteric nervous system that coordinate mucosal secretion with motor behaviour (Cooke, Wang and Rogers, 1993). The program is organised to stimulate copious secretion that flushes the mucosa and holds intraluminal contents in liquid suspension in the receiving segment ahead of the powerful propulsive contractions, which in turn, empty the lumen. The overall benefit is rapid excretion of material recognised by the immune system as threatening. The side effects are symptoms of abdominal distress and diarrhoea. Aside from containing a second brain with as many neurons as the spinal cord, the digestive tract is recognised as the largest lymphoid organ in the body together with a unique compliment of mast cells. In its position at one of the dirtiest of interfaces between the body and outside world, the intestinal mucosal immune system continuously encounters dietary antigens, bacteria, viruses and toxins. Physical and chemical barriers at the epithelial interface are insufficient to exclude fully the large antigen load, thereby allowing chronic challenges to the mucosal immune system. Evidence suggestive of direct communication between the mucosal immune system and the intrinsic neural networks in the intestine is derived from electrophysiological recording from enteric neurons in antigen sensitised animal models. The communication is meaningful and results in adaptive behaviour of the bowel in response to circumstances within the lumen that are threatening to the functional integrity of the whole animal. Communication is chemical in nature (paracrine) and incorporates specialised sensing functions of intestinal mast cells for specific antigens, together with the capacity of the enteric nervous system for intelligent interpretation of the signals. Immuno-neural integration progresses sequentially, beginning with immune detection followed by signal transfer to enteric microcircuits followed by neural interpretation and then selection of a specific neural program of coordinated mucosal secretion and motor propulsion (power propulsion) that effectively clears

ENTERIC NEURO-IMMUNOPHYSIOLOGY

365

the antigenic threat from the intestinal lumen. Investigation of immuno-neural interactions merges the disciplines of mucosal immunology and enteric neurophysiology and will be the focus of this chapter.

BRAIN-IN-THE-GUT CONCEPT The conceptual model for the enteric nervous system is the same as the brain and spinal cord (Figure 9.1). Like the vertebrate brain and spinal cord, the enteric nervous system is organised with the neural elements and integrated circuitry necessary for independent processing of sensory information and the programming of organised behaviour of effector systems in the control of the intraluminal environment of the bowel. The neural elements are sensory neurons, interneurons and motor neurons. Each kind of neuron has specialised regions (e.g. cell bodies, axons and dendrites) that are differentiated in expression of specific receptors, ionic channels and other functions that underlie basic mechanisms of neural signalling and information processing. The three kinds of neurons are synaptically connected

Figure 9.1 The conceptual model for enteric neuro-immunophysiology encompasses the brain, the enteric nervous system, neuro-immune communication and behaviour of the digestive effector systems. The enteric nervous system is an independent integrative nervous system. It processes information derived from sensory neurons central nervous inputs and immune/inflammatory cells (e.g. mast cells). The enteric neural networks contain reflex microcircuits and a library of gut behavioural programs. Interneuronal circuits determine the outflow of information in motor neurons to the intestinal effector systems. Coordinated activity of the effectors determines momentto-moment behaviour of the gut. Mast cells detect threatening antigens and alert the enteric nervous system to their presence. Mast cells signal the enteric nervous system by releasing paracrine mediators. The central nervous system signals the enteric nervous system through a brain to mast cell connection as well as direct neural pathways. (Reproduced with permission from Wood, 1995c).

366

INNERVATION OF THE GASTROINTESTINAL TRACT

into integrated circuits that process sensory information and program the variety of digestive functions found in specialised regions of the digestive tract during ever changing demands of the ingestive/digestive cycles of the functioning gut. Enteric integrated circuits determine the distinctive patterns of motility found along the gastrointestinal tract. Microcircuits in the intestine incorporate polysynaptic reflex circuits analogous to those in the spinal cord and basic to all recognised forms of motility. The integrated circuits store the programs for organisation of the commonly occurring digestive and interdigestive motility patterns emphasised above, as well as a library with additional programs that determine less frequent behaviours such as retropulsion during emesis, in the small bowel. Microcircuits in the myenteric division of the enteric nervous system contain the cell bodies of the motor neurons to the gastrointestinal musculature. Like spinal motor neurons, enteric motor neurons represent the final common pathway to the effector systems. Enteric motor neurons are uniaxonal neurons with Dogiel Type I morphology. There are both excitatory and inhibitory motor neurons to the musculature with specific axonal projections for each (reviewed by Brookes and Costa, 1994). Acetylcholine and substance P are recognised as important neurotransmitters released by excitatory motor neurons at neuromuscular junctions; whereas, nitric oxide and vasoactive intestinal peptide are implicated as inhibitory neurotransmitters (reviewed in Murray et al., 1991; Sanders and Ward, 1992; Makhlouf and Grider, 1993; Maggi et al., 1994; Wood, 1994; Maggi, 1995). Microcircuits of the submucous division of the enteric nervous system contain the cell bodies of the secretomotor neurons to the secretory epithelium. Secretomotor neurons stimulate secretion of water, electrolytes and mucous by releasing acetylcholine and vasoactive intestinal peptide at neuroepithelial junctions (reviewed by Cooke, 1989). Axonal collaterals of submucous secretomotor neurons project to the submucous vasculature. When the neuron fires to stimulate secretion from the crypts, the axon collaterals simultaneously release acetylcholine at their junctions with the vasculature (Adriantsitohaina and Surprenant, 1992). Acetylcholine activates the vascular endothelium to release nitric oxide, relax the arteriolar muscle and thereby increase mucosal blood flow in support of the demands of stimulated secretion in the intestinal crypts (reviewed by Vanner and Surprenant, 1996). Integrative microcircuits formed by synaptic connections of interneurons determine the timing and strength of neural outflow in the motor neuronal pathways to the musculature and secretory epithelium. In addition to individual control of each of these effector systems, the internuncial synaptic circuits coordinate the activity of each of the systems for homeostatic behaviour at the level of the integrated organ system. The enteric nervous system can be perceived, justifiably, as a mini-brain with specialised programs in juxtaposition to the effector systems it controls (Figure 9.1). Like the central nervous system, the enteric brain continuously processes sensory code on the momentto-moment state of each segment of gut and uses mechanisms of set-point determination and negative feedback for automatic control of the moment-to-moment intraluminal and intramural states. This model assumes that the central nervous system monitors the activity in the peripheral networks and transmits commands to the distant minibrain as appropriate for adjustment of gastrointestinal function in maintenance of whole body homeostasis.

ENTERIC NEURO-IMMUNOPHYSIOLOGY

367

ENTERIC IMMUNOLOGY The intestinal tract is colonised from birth with a complement of lymphoid and myeloid cells that fluctuates with luminal conditions and pathophysiological states (reviewed by McDonald, 1993). A variety of cell types including polymorphonuclear leukocytes, lymphocytes, macrophages, dendrocytes and mast cells are present in varying numbers in the intestinal mucosa and/or lamina propria and may be present in the muscle coats and connective tissue lamina between circular and longitudinal muscle coats. These are often found in close histo-anatomical association with the neuronal elements of the enteric nervous system and vagal nerve projections (Stead et al., 1987; Befus, 1994; Gottwald et al., 1995; Williams, Berthoud and Stead, 1995). In the normal bowel, both histo-anatomical and immunophysiological evidence suggest that elements of the enteric immune system are strategically positioned to cooperate with the enteric nervous system in establishing a first line of defense against foreign invasion at a vulnerable interface between the body and the outside environment. In pathophysiological states of inflammation (e.g. Crohn’s disease, ulcerative colitis and parasitic infection) close histo-anatomical proximity of lymphocytes and polymorphonuclear leukocytes to enteric nerve elements suggest the possibility for released cytokines and chemical mediators to reach and influence enteric nervous functions. Electrophysiological studies in enteric neurons described later in the chapter confirm that mediators of this nature do in fact alter electrical and synaptic behaviour of enteric neurons. ENTERIC MAST CELLS All of the cell types related to immunity and inflammation are putative sources of paracrine signals to the enteric nervous system. Nevertheless, most is known about signalling between mast cells and the neural elements of the local microcircuits of the enteric nervous system. The cytoplasm of mast cells is characterised by large numbers of electron-dense granules that are sites of storage for a variety of pre-formed chemical mediators. In addition to mediators that are preformed and stored, are substances that are newly synthesised in response to stimulation. Mast cells can be stimulated by antigens or secretagogues to secrete the mediators. Antigen stimulation involves receptors for antibodies on mast cells all of which have highaffinity receptors for IgE or other immunoglobulins (depending on animal species). When the receptors are occupied by antibodies to a sensitising antigen and cross-linking occurs by interaction of the sensitising antigen with the bound antibody, the mast cells release a melange of mediators. The list of mast cell mediators is long and only a partial list of those implicated in enteric immuno-neural communication by direct electrophysiological neural recording are listed in Table 9.1. Intestinal mast cells proliferate during infection of the intestine with nematode parasites such as Trichinella spiralis and Nippostrongylus brasiliensis. Animal models infected with these parasites as well as food allergy models utilising hyper-sensitivity to milk protein have proved informative in studies on mast cell involvement in enteric immuno-neural communication. In the sensitisation models, a second exposure to antigen isolated from the nematode, or to the milk protein, β-lactoglobulin results in predictable integrated behaviour of the intestinal

368

INNERVATION OF THE GASTROINTESTINAL TRACT

TABLE 9.1 Mast cell mediators: a truncated list. Preformed and stored

Newly synthesised

Biogenic amines Histamine 5-Hydroxytryptamine

Platelet activating factor

Enzymes Proteases Peroxidase Superoxide dismutase Chemotactic factors for inflammatory cells

Nitric oxide

Cytokines Tissue necrosis factor-α Interleukins IL-4

Arachidonic acid metabolites Leukotrienes, e.g. LTB4, LTC3 Prostaglandins, e.g. PGD2 Cytokines Tissue necrosis factor-α Interleukins IL-1β, IL-6

effector systems (e.g. Harari, Russell and Castro, 1987; Baird and Cuthbert, 1987; McKay and Perdue, 1993a,b). Recognition of the antigen by antibodies bound to sensitised mast cells triggers degranulation and release of mediators. The mediators then become messengers to the brain-in-the-gut which responds by suppressing other programs in its library and running the program designed to eliminate the antigen from the lumen (i.e. power propulsion coupled with copious secretion). In this respect, intestinal mast cells are uniquely equipped and situated to recognise agents that threaten whole body integrity and then signal to the enteric nervous system to program an appropriate response. Mast cell function in immuno-neural communication is an immune analogue of sensory detection and information coding in the classic sense of the nervous system (Figure 9.2). Sensory neurons, are genetically programmed to express a detection mechanism for a specific stimulus energy that remains unchanged throughout the life of the individual. Mast cells, on the other hand, acquire specific detection capabilities through the flexibility of recognition functions inherent in antibody expression by the immune system. Detection specificity functions can be acquired throughout life due to formation of new antibodies that become associated with mast cells. The output signals from mast cells are analogous to those from sensory neurons. Both mast cells and sensory neurons ultimately code information on the sensed parameter as a chemical message that is decoded by internuncial information-processing circuits in the nervous system. Apart from local signalling to the enteric nervous system, central messages from the cephalic brain to the enteric nervous system may be transmitted to the enteric brain through the mast cells. This is a brain-gut interaction by which central psychological status may be linked to irritable states of the digestive tract. Functional evidence for a brain-to-mast cell connection is found in reports of Pavlovian conditioning of mast cell degranulation in the gastrointestinal tract (MacQueen et al., 1989). Release of mast cell protease into the systemic circulation is a marker for degranulation of enteric mucosal mast cells. This can be demonstrated as a conditioned response

ENTERIC NEURO-IMMUNOPHYSIOLOGY

369

Figure 9.2 Detection and signalling functions of mast cells are analogous to sensory neurons. Receptor regions of sensory neurons (e.g. Pacinian corpuscles) detect changes in stimulus energy and transduce the changes into action potential code. The action potential code is transformed to a chemical code at the synapse of the sensory fibre with the second order neuron in the central neural processing circuitry for interpretation. Sensory neurons detect and code the same stimulus throughout its life. Intestinal mast cells detect sensitising antigens and use a chemical code to signal the presence of the antigen to the processing circuits of the enteric nervous system. The enteric nervous system interprets and responds to the mast cell signals with outputs that are adaptive for the animal. Mast cells detection is based on immune functions that enable the cells to learn to detect and remember antigenic substances and signal the presence of the allergen to the enteric nervous system whenever it reappears. Mast cell detection and signalling is updated for new allergens as they are encountered throughout life.

in laboratory animals to either light or auditory stimuli and in humans as a conditioned response to stress (Santos et al., 1998), suggestive of a brain-to-enteric-mast cell connection. Central nervous influence on mast cells in the upper gastrointestinal tract is suggested as well by work showing association between vagal afferents and enteric mast cells (Williams, Berthoud and Stead, 1995) and increased levels of histamine in intestinal mast cells in response to vagal nerve stimulation (Gottwald et al., 1995). Findings that stimulation of neurons in the brain stem by thyrotropin-releasing hormone evokes degranulation of mast cells in the rat small intestine are added evidence for brainmast cell interactions (Santos et al., 1996). Overall, the brain-mast cell connection is significant because the gastrointestinal symptoms associated with mast cell degranulation are expected to be the same whether the mast cells are stimulated by antigen-antibody crosslinking or neurotransmitters. Intracerebroventricular injection of thyrotropin-releasing hormone in the rat evokes the same kinds of inflammation and erosions in the stomach as cold-restraint stress. In the large intestine, restraint stress exacerbates nociceptive responses and these effects are associated with increased release of histamine from mast cells (Gue et al., 1997). Intracerebroventricular injection of corticotropin-releasing factor mimics the responses to stress. Intracerebroventricular injection of a corticotropin- releasing factor antagonist or pretreatment with mast cell stabilising drugs suppresses stress-induced lower gastrointestinal responses.

370

INNERVATION OF THE GASTROINTESTINAL TRACT

Mast cell degranulation may release mediators that sensitise silent nociceptors in the large intestine. In animals, degranulation of enteric mast cells results in a reduced threshold for pain responses to intestinal distension that is prevented by treatment with mast cell stabilising drugs (Coelho, Fioramonti and Bueno, 1998). In addition to signalling the enteric minibrain, mast cells release messenger substances that attract leukocytes into the lamina propria from the vascular system. Introduction of purified subunits of Clostridium difficile toxin into intestinal loops stimulates influx of polymorphonuclear leukocytes as well as activating the neural program for coupled secretion and power propulsion. Blockade of enteric nerves by tetrodotoxin, treatment with tachykinin NK1 antagonists or mast cell stabilisation by the drug ketotifen all prevent the transmigration of polymorphonuclear cells into the lamina propria and the acute inflammatory response to the toxin (Rachmilewitz et al., 1992; Castagliuolo et al., 1994; Pothoulakis, Castagliuolo and LaMont, 1998). This suggests that enteric nervous input utilising substance P, which is a putative transmitter for enteric slow excitatory postsynaptic potentials and axonal reflexes mediated by primary afferent neurons (see below), participates in toxin-induced mast cell degranulation and the release of chemoattractant factors for inflammatory cells. Application of toxin-A to neurons in the enteric nervous system alters both the electrical behaviour of the neuronal cell bodies and inhibitory noradrenergic neurotransmission to secretomotor neurons in the submucous plexus (Xia et al., 1999). Altered electrical behaviour includes slow EPSP-like depolarisation and elevated excitability. Tetrodotoxin or a histamine H2 receptor antagonist does not affect the depolarisation evoked by toxin-A. Failure of the histamine antagonist to suppress the actions of toxin-A indicates that its neuronal actions are direct and not mediated by degranulation of intramural mast cells. Among the actions of toxin-A on neurotransmission is suppression of inhibitory postsynaptic potentials evoked in uniaxonal neurons in the submucous plexus by stimulation of sympathetic nerve fibres. The toxin acts presynaptically to suppress the release of noradrenaline from the sympathetic innervation of the enteric nervous system. This results in removal of sympathetic braking action from secretomotor neurons. Together with toxinevoked neuronal excitation, this may be an underlying factor in the diarrhoea associated with C. difficile overgrowth in the large intestine. Substance P is known to be a secretagogue for histamine release from mast cells (Cocchiara et al., 1999) but often in concentrations so high that the physiological relevance has been questioned (Shanahan et al., 1985). It was generally assumed that interactions of the hydrophobic N-terminal moieties of substance P with the lipid bilayer, not the basic amino acids believed to bind the receptor site, was responsible for degranulation of mast cells by the neuropeptide (Repke et al., 1987). The findings of Castagliuolo et al. (1994) with the C. difficile suggest that this may not be the case and that NK-1 receptors may mediate substance P action on the mast cells. Relative to these uncertainties, Befus (1994) points out the fact that mast cell populations are heterogenous, that there may be selective release of mast cell mediators by different physiological messengers and histamine release may not be the optimum end point for assaying effects of putative secretagogues. For example, patch clamp studies show that picomolar concentrations of substance P evokes changes in ion channel behaviour in peritoneal mast cells (Janiszewski, Bienenstock and Blennerhassett, 1994) and the IC50 for mast cell-mediated intestinal mucosal secretion is 100 nM for substance P and 1 nM for substance P(4–11) (Wang et al., 1995).

ENTERIC NEURO-IMMUNOPHYSIOLOGY

371

SIGNALLING FROM MAST CELLS TO THE BRAIN-IN-THE-GUT Several mast cell-derived mediators share common neuropharmacological actions on electrical and synaptic behaviour of neurons in the enteric nervous system. These include histamine, 5-hydroxytryptamine, adenosine, interleukin-1β (IL-1β), interleukin-6 (IL-6), tumour necrosis factor α (TNFα), and platelet activating factor. HISTAMINE Histamine is not localised to any extent in enteric neurons and is not considered as a putative neurotransmitter in enteric microcircuits (Panula et al., 1985). The principal source of histamine in the intestine is the population of mast cells. Understanding of histaminergic actions on intestinal neurons is derived from results obtained in electrophysiological studies on single neurons of the small and large intestine of the guinea-pig. These studies were done with either extracellular electrodes that detect action potential discharge of individual neurons or intracellular microelectrodes that record changes in neuronal membrane potential, input resistance, action potential discharge and synaptic potentials. Detailed accounts of the methods, as well as a review of the cellular neurophysiology of enteric neurons can be found in Wood (1989, 1994). Mayer and Wood (1975) reported that application of histamine excited neurons in the myenteric plexus of cat small intestine. Subsequent work in the guinea-pig revealed two important actions of histamine on neuronal elements of the enteric microcircuits. One of the actions occurs at the neuronal cell body and consists of long-lasting excitation. The second is at nicotinic synapses, where histamine acts to suppress synaptic transmission. These actions are found in the circuitry of both the myenteric and submucous plexuses of the small and large intestine. Histamine mimics slow synaptic excitation The excitatory effects of histamine mimic slow synaptic excitation (Nemeth, Ort and Wood, 1984; Tamura and Wood, 1992; Frieling, Cooke and Wood, 1993). Slow synaptic excitation (slow EPSP) is reviewed in detail in Wood (1994). It is a response detected by intracellular microelectrodes when specific neurotransmitters are released from enteric axons at synapses on the cell bodies of the recorded neurons (Figure 9.3). Binding of slow EPSP mediators to their receptors activates transduction mechanisms that change conductance states of a variety of ionic channels to produce the characteristic alterations in electrical behaviour. The changes in electrical behaviour during slow EPSPs include depolarisation of the membrane potential, increase in the electrical resistance of the membrane and enhanced excitability reflected by spike discharge. In addition, hyperpolarising after-potentials in AH/Type 2 (after-hyperpolarisation/Dogiel Type II neurons) neurons are suppressed to permit repetitive spike discharge. AH/Type 2 neurons are a distinct class of enteric neurons characterised by low excitability and the presence of long-lasting hyperpolarising after-potentials. In the resting state

372

INNERVATION OF THE GASTROINTESTINAL TRACT

Figure 9.3 Histamine and 5-hydroxytryptamine mimic slow synaptic excitation in AH/Type 2 enteric neurons. Slow synaptic excitation is characterised by depolarisation and a prolonged train of action potentials. The slow EPSP in the top trace was evoked by focal electrical stimulation of the axon responsible for synaptic input. Application of histamine (middle trace) or 5-hydroxytryptamine (bottom trace) by pressure microejection from a fine-tipped pipette mimics the depolarisation and spike train of the slow synaptic response.

and in the absence of slow synaptic mediators, these neurons either do not fire or fire only one or a few action potentials to a depolarising stimulus. The action potentials are followed by hyperpolarising after-potentials that last for several seconds. After-hyperpolarisation is a mechanism that prevents repetitive discharge when the neuron is not under the influence of a slow excitatory signal. Suppression of the after-hyperpolarisation enables the repetitive discharge of spikes that is seen to occur in prolonged trains during slow synaptic activation (Figure 9.3). Repetitive discharge is an important functional event because each action potential, after starting in the cell body of the multipolar neuron, will propagate to the terminals of each neurite where the neuron’s neurotransmitters will be released at synapses with other elements of the microcircuits. Signal transduction for slow EPSPs Transduction of slow synaptic signals involves activation of adenylate cyclase and second messenger function of adenosine 3′,5′-cyclic monophosphate (cAMP) (Palmer, Wood and Zafirov, 1986, 1987a). Excitatory receptors for slow synaptic excitation are coupled to adenylate cyclase through G proteins in enteric neuronal membranes (Tamura, Itoh and Wood, 1995).

ENTERIC NEURO-IMMUNOPHYSIOLOGY

373

Histamine receptors for slow excitatory responses Histamine H2 receptors are the mediators of the slow excitatory response to histamine in cell bodies of enteric neurons in the guinea-pig (Nemeth, Ort and Wood, 1984; Tamura and Wood, 1992; Freiling, Cooke and Wood, 1993). The selective H2 agonist, dimaprit, mimics the excitatory actions of histamine and the H2 antagonist, cimetidine, blocks them. Exposure to histamine elevates levels of cAMP in myenteric ganglia and this action is also blocked by selective histamine H2 receptor antagonists and mimicked by selective agonists (Xia, Fertel and Wood, 1996). The hyperexcitability evoked by interaction of histamine with the H2 histamine receptor does not desensitise during prolonged exposure. All of the slow EPSP-like actions of histamine persist unabated for as long as histamine is present in the bathing media (Tamura and Wood, 1992). This is expected to continuously energise the microcircuits that control the behaviour of the intestinal effector systems and provide a long-lasting drive to maintain the immune “alarm” program in effect. Chronic release of histamine from mast cells is likely to do this in the living animal. Occurrence of this activity in the microcircuits that control the secretomotor neurons to the intestinal crypts is predicted to enhance intestinal secretion leading to diarrhoeal symptoms like those associated with infective agents or food allergies. Watery diarrhoea symptoms associated with mastocytosis and microscopic colitis in humans have been treated effectively with histaminergic blocking drugs (Baum, Paramjit and Miner, 1989).

Pattern generation evoked by histamine Pattern generators are networks of neurons that generate the timing and phasing cues for the rhythmic behaviours of effector systems. They are often called central pattern generators because the network is capable of generating the same basic activity pattern in experimental isolation from other regions of the nervous system. Behaviours driven by neural pattern generators occur in fixed action patterns that may be triggered by an external stimulus or by intrinsic command inputs. These fixed action patterns are similar to reflexes, but differ in important respects. Unlike reflexes, the duration, latency and intensity of effector activity of fixed action patterns are not determined by the stimulus alone. Behaviours driven by pattern generators are also unlike reflexes in that the fixed action pattern can occur in the absence of sensory input, whereas reflexes cannot. Although, sensory feedback often modifies ongoing motor activity, it is not essential for the generation or timing of a motor pattern. This is programmed by the pattern generator and communicated to the motor neurons for execution of the pattern. A common characteristic of pattern generating networks is the occurrence of rhythmic discharge of spike bursts in individual neurons of the network. This is the case for networks that determine fixed action patterns of movements in both vertebrate and invertebrate central nervous systems. Exposure of the neural networks in the intestinal submucous plexus to histamine evokes patterned bursts of spikes suggestive of a central pattern generator in a subpopulation of neurons in the network (Figure 9.4). At periodic intervals in the presence of histamine, the membrane potential depolarises spontaneously and this is accompanied by a crescendo of

374

INNERVATION OF THE GASTROINTESTINAL TRACT

Figure 9.4 Pattern generation evoked by histamine in an AH/Type 2 submucous neuron of the guinea-pig colon. (A) Recurrent trains of action potentials evoked by exposure of the neuron to histamine. (B) Continuous traces of a single train of action potentials from A recorded on an expanded time base. Action potential discharge was preceded by spontaneous depolarisation of the membrane potential. Intracellular injection of constant current depolarising pulses did not evoke spikes prior to the train. The current pulses triggered spikes as the membrane potential depolarised and excitability increased, leading up to the spike train. Spikes continued to be evoked by the depolarising current pulses until the membrane potential repolarised and excitability was reduced to a silent period before the next spike train. (Reproduced with permission from Frieling, Cooke and Wood, 1993).

spike discharge that subsides to silence prior to the start of the next cycle. In most neurons, the recurrent spike bursts occur at intervals of 2–3 min. The patterned behaviour can be evoked by histamine H2 receptor agonists, but not by histamine H1 receptor agonists (Frieling, Cooke and Wood, 1993). The effects of histamine are blocked by histamine H2 but not by histamine H1 receptor antagonists. Patterned discharge evoked by histamine occurs only in the networks of the submucous division of the enteric nervous system; it is not found in the myenteric division (Tamura and Wood, 1992). The electrophysiological observations suggest that histamine acts like a neuromodulator to activate a central pattern generator in the enteric neural networks. Discussion below will show how the central generator drives rhythmic patterns of effector behaviour when the neural networks are overlaid with histamine, either from experimental application or from antigen-stimulated release from mucosal mast cells. As in other nervous systems, it may be that the enteric neural networks are not hard-wired groups of neurons that exist only in an active or inactive mode and generate only one pattern of behaviour when in the active mode. The enteric networks may be like other ensembles of neurons known to be capable of producing a variety of behaviours depending on the kind of neuromodulatory overlay. This is sometimes referred to as multiple task processing where different neuromodulators reconfigure different rhythmic outputs from a single neural network.

ENTERIC NEURO-IMMUNOPHYSIOLOGY

375

Presynaptic inhibition by histamine Presynaptic inhibition is the suppression of release of neurotransmitters from release sites on axons. In the enteric nervous system, this occurs at both fast and slow excitatory synapses and at neuroeffector junctions. Presynaptic inhibition may involve axo-axonal transmission, whereby release of a neurotransmitter from one axon acts at receptors on another to suppress release of transmitter from the second axon or it can be mediated by substances released from mast cells or other non-neuronal cells into the milieu surrounding the synaptic circuits. Ten or more kinds of receptors for neurotransmitters/neuromodulators are known to be involved in presynaptic inhibition in the enteric nervous system (reviewed in Wood, 1994). Histamine is one of these mediators. It acts at presynaptic receptors on cholinergic axons to suppress fast EPSPs at nicotinic synapses in the enteric microcircuits (Tamura, Palmer and Wood, 1987). Histamine also acts on sympathetic nerve terminals in the submucous plexus to suppress the release of noradrenaline and the occurrence of slow inhibitory postsynaptic potentials that are mediated by noradrenaline in secretomotor neurons (Figure 9.5). This action of histamine is blocked by histamine H3 receptor antagonists and

Figure 9.5 Histamine suppresses slow inhibitory postsynaptic potentials (IPSPs) mediated by release of noradrenaline from sympathetic postganglionic neurons in the submucous plexus of guinea-pig small intestine. (A) Control IPSP evoked by focal electrical stimulation of sympathetic postganglionic fibres. (B) Suppression of the IPSP during exposure to histamine. (C) Return of the IPSP after washout of histamine.

376

INNERVATION OF THE GASTROINTESTINAL TRACT

is mimicked by histamine H3 receptor agonists, suggesting that histamine H3 receptors are responsible for the presynaptic inhibitory action of histamine. Discussion later in the chapter will describe how histamine released during antigenic degranulation of intestinal mast cells acts in a like manner. 5-HYDROXYTRYPTAMINE 5-Hydroxytryptamine is another preformed mediator expected to be released during degranulation of intestinal mast cells in most mammals except the guinea-pig. In the guineapig and some other species, 5-hydroxytryptamine localisation is restricted to neurons and enterochromaffin cells. Two actions of 5-hydroxytryptamine on guinea-pig enteric neurons are essentially the same as those of histamine, whereas in other respects the actions differ from histamine (Wood and Mayer, 1979; Mawe, Branchek and Gershon, 1986; Wade and Wood, 1988; Frieling, Cooke and Wood, 1991; Tack et al., 1992). 5-Hydroxytryptamine acts at 5-HT1p receptors on the neuronal soma (Mawe, Branchek and Gershon, 1986) to produce long-lasting excitatory responses that mimic all aspects of slow synaptic excitation and the slowly activating excitatory response to histamine (Figure 9.3). The 5-HT1p receptor is a metabotropic receptor linked through G proteins to activate adenylate cyclase and elevate levels of cAMP as part of the transduction process in myenteric ganglia from the guinea-pig (Fiorica-Howells, Wade and Gershon, 1993; Xia, Fertel and Wood, 1994). 5-Hydroxytryptamine acts at presynaptic receptors on cholinergic axons to suppress fast EPSPs at nicotinic synapses in the enteric networks (North et al., 1980). This action is also the same as the presynaptic inhibitory action of histamine at nicotinic synapses. The presynaptic inhibitory receptors behave like the 5-HT4 receptor subtype (Frieling, Cooke and Wood, 1991). Unlike histamine, and contrary to an earlier report (Surprenant and Crist, 1988), we have not found presynaptic inhibition of slow inhibitory postsynaptic potentials in the guinea-pig submucous plexus. 5-Hydroxytryptamine also produces fast excitatory depolarising responses that are similar to fast nicotinic excitatory postsynaptic potentials in enteric neurons. These responses are mediated by 5-HT3 ionotropic receptors that function as part of ligandgated channel complexes (Derkach, Surprenant and North, 1989). Responses similar to the 5-HT3 receptor mediated responses have not been found for histamine in the enteric nervous system. 5-Hydroxytryptamine also differs from histamine in not evoking the rhythmic discharge seen for histamine in submucous neurons. This carries over to the motor output from the submucosal network which does not show the rhythmic oscillations of effector behaviour that will be discussed later in the chapter for histamine and antigen exposure in the sensitised intestine. The actions of 5-hydroxytryptamine, when compared with histamine, suggest a neuromodulatory function that may be associated with network reconfiguration that differs from the “alarm” program that is called into play by the neuromodulatory action of histamine. Whereas, 5-hydroxytryptamine is involved in both neural and paracrine transmission, histamine seems to act solely as an “alarm” message to the enteric nervous system signalling the presence of a threat in the intestinal lumen.

ENTERIC NEURO-IMMUNOPHYSIOLOGY

377

ADENOSINE The extent to which adenosine is released from mast cells during antigen exposure or in response to secretagogues is unclear. If adenosine were released as a neuromodulatory overlay on the enteric networks, it is clear that the effect would be a braking action on excitatory processes within the circuits. All of the actions of adenosine that are mediated by the adenosine A1 receptor subtype on neural elements in the enteric microcircuits of the guinea-pig are inhibitory. Adenosine acts through A1 receptors to suppress both fast nicotinic excitatory postsynaptic potentials and slow excitatory transmission. It inhibits excitation in the cell bodies of AH/Type 2 neurons by opening K+ channels and hyperpolarising the membrane (Palmer, Wood and Zafirov, 1987a,b; Christofi and Wood, 1993, 1994). Suppression of slow excitatory transmission by adenosine seems to involve A1 receptors positioned both pre- and postsynaptically. Application of adenosine after the onset of stimulus-evoked slow EPSPs aborts the response, indicating a post-synaptic action. In release studies of putative neurotransmitters for slow synaptic excitation (e.g. tachykinins), adenosine A1 stimulation suppresses the release of putative neurotransmitters for slow synaptic excitation (e.g. tachykinins) in studies on release from myenteric plexus preparations (Christofi, McDonald and Cook, 1990; Broad et al., 1992). Part of the mechanism of adenosinergic inhibition at the cell somas of enteric neurons involves inhibition of adenylate cyclase and suppression of cAMP formation (Xia, Fertel and Wood, 1997). Adenosine acts through G protein-coupled adenosine A1 receptors to inhibit stimulation of adenylate cyclase by forskolin (Zafirov, Palmer and Wood, 1985) and by histamine (Palmer, Wood and Zafirov, 1987b; Xia, Fertel and Wood, 1997). In contrast to its inhibitory action on the slow excitatory actions of histamine and on histamine stimulation of adenylate cyclase, adenosine does not inhibit the slow excitatory actions of 5-hydroxytryptamine nor the stimulation of adenylate cyclase by 5-hydroxytryptamine (Palmer, Wood and Zafirov, 1987b; Xia, Fertel and Wood, 1997). This may be a reflection of subtlety of neuromodulatory events that reconfigure the neural networks for changes in program output. Presynaptic inhibition by adenosine at fast nicotinic synapses in the enteric microcircuits occurs at the A1 type of P1 purinoreceptor (Barajas-Lopez, Surprenant and North, 1991; Christofi, Tack and Wood, 1992; Christofi and Wood, 1994). This occurs also at noradrenergic inhibitory synapses on neurons in the submucous plexus.

CYTOKINES Investigation of actions of cytokines on the nervous elements of the enteric neural networks is preliminary. The cytokines investigated are TNFα, IL-1β and IL-6. These are of potential importance because they may be released in paracrine fashion from a variety of immune and inflammatory cells as well as mast cells. Actions of the three cytokines are essentially the same. Like histamine and 5-hydroxytryptamine, they produce excitatory responses that mimic slow synaptic excitation (Xia et al., 1995, 1999). The natural IL-1 receptor antagonist (nIL-1ra) prevents the excitatory action of IL-1β. All three cytokines also suppress the release of noradrenaline from

378

INNERVATION OF THE GASTROINTESTINAL TRACT

sympathetic nerve terminals in the enteric nervous system and thereby block inhibitory neurotransmission to secretomotor neurons in the submucous plexus. This can be observed as suppression of noradrenergic inhibitory postsynaptic potentials in microelectrode recording from the neurons (Xia et al., 1995; 1999). Ruhl, Hurst and Collins (1994) reported that IL-1β suppressed the release of noradrenaline from intestinal preparations in vitro and that IL-6 acted synergistically with IL-1β in the suppression of noradrenaline release. Microelectrode recording from submucous neurons with Dogiel I morphology (presumably secretomotor neurons) revealed that application of either IL-1β or IL-6 dose-dependently suppressed the amplitudes of stimulusevoked noradrenergic inhibitory postsynaptic potentials (Xia et al., 1999). Application of IL-1β and IL-6 combined produced greater suppression of the inhibitory synaptic potentials than predicted from the additive effects of the two cytokines applied separately. Neither of the cytokines applied separately or in combination suppressed the amplitude of inhibitory responses to micropressure pulses of noradrenaline. Absence of reduction in responses to noradrenaline while noradrenergic neurotransmission was suppressed meets criteria for a presynaptic inhibitory action of the cytokines on release of noradrenaline. The results with the cytokines, like those with histamine, suggest that accumulation of these mediators, derived either from degranulation of mast cells or other cell types associated with inflammatory states, will lead to suppression of the physiological effects of activation of the sympathetic innervation to the bowel. The physiological actions of sympathetic activation are presynaptic inhibition at nicotinic synapses resulting in shut-down of enteric microcircuits and inhibition of excitatory secretomotor neurons to the intestinal crypts. This predicts neural network behaviour that reflects removal of the braking action of the sympathetic nervous input to the intestine resulting in hyper-activity of the circuits with further enhancement produced by the excitatory action of the cytokines on the neurons in the network. Inhibition of the sympathetic brake on secretomotor neurons in parallel with neuronal stimulation would result in enhanced secretion from the crypts and development of a potential for secretory diarrhoea. PLATELET ACTIVATING FACTOR, LEUKOTRIENES AND PROSTAGLANDIN D2 Similarities and differences are found for the actions of lipid membrane derived mediators. Platelet activating factor (PAF) has excitatory actions on enteric neurons that are essentially the same as for the other above mentioned immuno-physiological mediators that mimic slow synaptic excitation (Xia et al., 1996). It acts in concentration-dependent manner on the membranes of the neuronal cell bodies to depolarise the membrane potential, increase input resistance, increase action potential discharge during intraneuronal injection of depolarising current and to induce spontaneous spike discharge. Like the other mediators, it acts presynaptically to inhibit release of noradrenaline from sympathetic nerves in the submucous plexus and thereby remove the sympathetic brake from the secretomotor innervation of the intestinal crypts. Leukotrienes B4, E4, D4 and C4 (LTB4, LTE4, LTD4 and LTC4) also have excitatory actions that are essentially the same as described for other putative mediators in immunoneural communication (Frieling et al., 1995). Like histamine, but unlike the other mediators

ENTERIC NEURO-IMMUNOPHYSIOLOGY

379

described above, LTE4, LTD4 and LTC4 evoke recurrent periodic discharge of trains of action potentials in submucous neurons. Prostaglandin D2 (PGD2) has excitatory actions on enteric neurons that are essentially the same as described above for PAF (Frieling et al., 1994b). Unlike the actions of histamine and 5-hydroxytryptamine, PGD2 does not suppress fast nicotinic neurotransmission in the submucous microcircuits, but does inhibit some forms of slow synaptic excitation. NITRIC OXIDE Although nitric oxide has important consequences in inflammatory responses and is formed and released by intestinal mast cells, it appears to have minimal effects on the neuronal elements of the enteric microcircuits. Release of nitric oxide from sodium nitroprusside does not affect resting membrane potentials in myenteric neurons of the guinea-pig intestine, except for an occasional decrease in input resistance attributable to suppression of excitatory neurotransmitter release (Tamura, Schemann and Wood, 1993). Release of nitric oxide does not alter fast nicotinic neurotransmission, but does suppress noncholinergic slow excitatory postsynaptic potentials. It does not affect the slow depolarising responses to 5-hydroxytryptamine or substance P, suggestive of a presynaptic site of action for suppression of stimulus-evoked slow synaptic excitation. The nitric oxide synthase inhibitor, N-nitro-L-arginine methyl ester, does not affect resting membrane excitability or excitatory synaptic events in myenteric neurons of the guinea-pig.

NEURAL NETWORKS AND EFFECTOR SYSTEMS BEHAVIOUR The mucosa, musculature and blood vasculature are the primary effector systems involved in overt intestinal responses to mast cell degranulation. Mast cell degranulation leads to stimulated secretion from the intestinal crypts in association with increased mucosal blood flow. Degranulation of mast cells is associated also with a specialised form of motility that was referred to as power propulsion earlier in the chapter. The migrating spike bursts recorded with serosal electrodes by Mathias et al. (1976) in response to enterotoxin and by Schanbacher et al. (1978) in response to infection by T. spiralis are reflections of the muscle behaviour during power propulsion. The giant migrating contractions recorded with serosal strain gauges in vivo in antigen-sensitised and other animal models also reflect the occurrence of power propulsion (Cowles and Sarna, 1990, 1991; Sarna et al., 1991). RESPONSES TO HISTAMINE AND LEUKOTRIENES In preparations of guinea-pig colonic mucosa in vitro, application of histamine evokes transient increases in secretion resulting from the direct action of histamine with histamine H1 receptors on the enterocytes and evoked release of prostaglandins and other unidentified mediators (Wang and Cooke, 1990). After the transient response to histamine subsides, cyclical bursts of electrolyte secretion appear and are seen as changes in short-circuit current

380

INNERVATION OF THE GASTROINTESTINAL TRACT

Figure 9.6 Histamine evokes periodic cycles of secretion in the mucosa of the guinea-pig large intestine. The record shows short-circuit current indicative of electrogenic chloride secretion as investigated in an Ussing flux chamber during continuous exposure to histamine. (Modified with permission from Wood, 1991).

in Ussing chamber studies. They persist unchanged for periods of hours (Figure 9.6; Wang and Cooke, 1990). Application of the nerve-blocking agent tetrodotoxin abolishes the secretory cycles, indicating that they are neurally driven. The secretory cycles are prevented by drugs that block the histamine H2 receptor subtype. Selective histamine H2 receptor agonists reproduce the cyclical behaviour induced by histamine, whereas histamine H1 receptor agonists do not. Exposure to atropine or mecamylamine suppresses the recurrent secretory cycles. The action of atropine, taken together with what is known about cholinergic stimulation in mucosal preparations, suggests that the cycles are driven mainly by the release of acetylcholine from secretomotor neurons at neuroepithelial junctions. The blocking action of mecamylamine suggests that nicotinic synapses are active in the microcircuits responsible for generating the patterned output to the mucosa. Extended exposure to LTE4, LTD4 or LTC4 was also reported to evoke recurrent cycles of electrolyte secretion in the guinea-pig colon reminiscent of the cyclical behaviour induced by histamine (Frieling et al., 1995). The response to the leukotrienes, like that of histamine, was blocked by tetrodotoxin, atropine or a nicotinic blocker. Cooke, Wang and Rogers (1993) recorded short-circuit current and muscle contraction simultaneously during exposure to histamine in Ussing chamber studies. They found that phasic contractions recorded by strain gauges oriented along the circular muscle axis of the flat sheet preparations were synchronised with the secretory cycles. A contraction appeared to be coupled to each secretory cycle with each contraction lagging behind the secretory event. Coupling of the secretory cycles with phasic contractions required neural connections between the myenteric and submucous plexuses. Severing the connections prevented the contractile responses without affecting the secretory cycles. NEURAL PATTERN GENERATION The cyclic trains of action potentials described earlier in the chapter for submucous neurons during exposure to histamine are apparently the neurophysiological correlates of the cyclic secretory and contractile behaviour evoked by histamine in full-thickness preparations of intestine in Ussing chamber studies. Results of the neurophysiological and integrated system studies suggest that the cyclic behaviour is driven by a neural pattern generator that switches secretomotor neurons between active and inactive states to drive

ENTERIC NEURO-IMMUNOPHYSIOLOGY

381

periodic bursts of secretion of water and electrolytes. Activity of the pattern generator is part of the specialised program stored in the program library of the enteric nervous system. An overlay of histamine on the neural network commands the program into action. The program includes mechanisms for activating the population of secretomotor neurons to the intestinal crypts and motor neurons to the musculature. It also incorporates timing functions that coordinate secretion with motility. The power propulsion pattern of muscular behaviour and associated hypersecretion triggered by noxious stimulation of the mucosa in vivo are postulated to be the integrated output of the immuno-neural “alarm” program. Patterned neural discharge and cyclic secretory behaviour and coupled contractions seen in neurophysiological and Ussing chamber studies in vitro represent fragmented operations of the program. The integrated output of the neural network in response to histamine can be attributed to two actions on single elements of the circuits. Excitation of the neuronal cell body by histamine H2 receptors and presynaptic inhibition mediated by histamine H3 receptors at nicotinic synapses in the circuits are the only definable actions at the single neuron level of organisation. The patterned output emerges from integration at higher order levels of circuit organisation. Like integrative neural networks in general, understanding of the integrated circuit level of organisation in the enteric nervous system is less advanced than understanding of actions at the level of the individual elements of the network. Two aspects of enteric integrated circuit function are especially interesting, but not well understood mechanistically. One is the emergent mechanism that accounts for the programming of the central pattern generator when exposed to histamine. The second is the mechanism by which a population of submucosal secretomotor neurons is recruited to fire simultaneously in supplying synchronous excitatory drive to the secretory epithelium. Slow synaptic excitation is thought to underlie the latter phenomenon, whereas there are few clues to the mechanism underlying pattern generation. SLOW SYNAPTIC EXCITATION Slow synaptic excitation is postulated to be the basic function in a mechanism for gating spread of excitation among the neurons in an intestinal microcircuit. Slow synaptic gating and feed forward excitation have been proposed as a synaptic mechanism in circuits that recruit synchronous discharge in neuronal pools consisting of hundreds of enteric neurons. This is necessary in neuronal populations responsible for generation of the same motor event simultaneously in the musculature around the circumference and for extended distances along the longitudinal axis of an intestinal segment. A similar mechanism can explain synchronous recruitment of secretomotor neurons to the intestinal epithelium to generate the cyclical secretory behaviour resulting from histamine action and immuno-neural communication in the sensitised intestine described later in the chapter. Slow synaptic excitation and slow paracrine excitation underlie a mechanism for longlasting activation or inhibition of gastrointestinal effector systems. The prolonged discharge of spikes during the slow excitatory event drives the release of neurotransmitter from the axon for the duration of the event measured in seconds or minutes (Figure 9.3). It results in prolonged inhibition or excitation at neuronal synapses and neuroeffector junctions.

382

INNERVATION OF THE GASTROINTESTINAL TRACT

Contractile responses of the muscles and secretory responses of the epithelium are sluggish events that last for several seconds from start to completion. The train-like discharge of spikes during slow EPSPs is the candidate correlate of long-lasting responses of the effector systems in the functioning gastrointestinal tract. COORDINATED RECRUITMENT Slow excitatory events in cell somas of AH/Type 2 neurons emerge as an explanation for coordinated recruitment functions at the microcircuit level of organisation. Coordinated recruitment refers to the simultaneous conversion of a pool of several hundred AH/Type 2 neurons from the hypo- to hyper-excitable state with repetitive spike discharge being initiated in the cell somas of each of the individual members of the population. Coordinated firing within the AH/Type 2 interneuronal population ultimately drives the behaviour of motor neurons to the effector systems (Figure 9.7). Importance of coordinated recruitment is the requirement that populations of motor neurons be stimulated to activity in synchrony in order to evoke simultaneous responses of the musculature or secretory epithelium in a segment of digestive tract with several square centimetres of surface area. This is believed to be a function of “driver circuits” comprised of interconnected AH/Type 2 neurons (Wood, 1995a; Figure 9.7). Driver circuits synchronise motor and secretory events around the circumference of a defined length of bowel.

Figure 9.7 Driver circuits are formed by populations of AH/Type 2 neurons that are synaptically connected for feed forward synaptic excitation. Synchronous discharge in the driver circuit is the mechanism responsible for simultaneous activation of populations of motor neurons to gastrointestinal effectors. The cell somas of AH/ Dogiel II neurons in driver circuits behave like a somal gate that is closed when the soma is in its hypoexcitable state and opened when the somal membrane is made hyperexcitable by slow synaptic input. Open gates permit feed forward (positive feedback) excitation in the circuit; closed gates inactivate the feed forward function of the circuit.

ENTERIC NEURO-IMMUNOPHYSIOLOGY

383

These circuits are thought to provide simultaneous synaptic drive to subpopulations of inhibitory or excitatory neurons to the muscle layers during operation of propulsive motor programs. On-off behaviour of the driver circuits is the most plausible function to account for the on-off cyclical behaviour of neurally driven electrolyte and water secretion described earlier in the chapter. The off-state of the driver circuit corresponds to the low excitability condition of the AH/Type 2 neurons that make-up the circuit; whereas, the on-state is the hyperexcitable condition of slow synaptic excitation or paracrine actions of histamine or other neuromodulators. Sixteen or more neuro/paracrine mediators have receptors on enteric neurons that mimic slow synaptic excitation when activated and an equivalent number of neuro/paracrine modulators inhibit slow synaptic excitation (reviewed by Wood, 1994). The AH/Type 2 neurons in the driver networks are assumed to be the multipolar Dogiel Type II neurons that have extensive ramifications of axonal projections in the circumferential axis of the intestine (Bornstein et al., 1991). They are thought to be interneurons that are synaptically interconnected for feed forward excitation. This is a kind of synaptic connectivity in which the neurons of the circuit make recurrent slow excitatory synaptic connections one with another (Kunze, Furness and Bornstein, 1993). It is a positive feedback system that leads to rapid amplification of excitation in the population of driver interneurons. It underlies a mechanism that ensures simultaneous activation of the entire network around the circumference of the segment of bowel (Thomas, Bertrand and Bornstein, 1999). Simultaneous activation is obviously important for effective application of the forces necessary for propulsion of the luminal contents and coordination of mucosal secretory events with behaviour of the musculature. GATING FUNCTIONS Slow excitatory events in cell somas of AH/Type 2 neurons emerge also as an explanation for a form of gating function at the microcircuit level of organisation. Control of feed forward excitation and coordinated recruitment of neuronal activity develops from a gating function of the cell body of AH/Type 2 neurons. Gating in this case refers to the control of propagation of action potentials between the neurites arising from opposite poles of the multipolar cell somas of AH/Type Dogiel Type II neurons. Slow excitatory events are an integral part of the gating mechanism. Figure 9.7 illustrates how the gating mechanism is thought to work in this kind of neuron. The membrane of the cell body of AH/Type 2 neurons can exist in an inexcitable state, hyper-excitable state or intermediate state of excitability. Inexcitability reflects the absence of slow excitatory mediators, hyper-excitability reflects exposure to elevated levels of one or more mediator and intermediate states reflect intermediate levels of one or the other of the sixteen or so putative excitatory mediators. When the cell body is in the inexcitable state, action potentials propagating toward the cell body in one of its neurites cannot fire the membrane of the cell soma and the gate is closed. In intermediate states of excitability where the somal membrane happens to be fired by an inbound spike, the action potential in the cell body will be followed by the characteristic after-hyperpolarisation and the prolonged refractory period will prevent firing of the somal membrane by any additional incoming spikes. In this state, the partially open somal gate fractionates the transfer of information

384

INNERVATION OF THE GASTROINTESTINAL TRACT

across the cell body. In the hyper-excitable state, the probability that the cell body will be fired repetitively by inbound spikes is dramatically increased. This occurs because excitability of the membrane is enhanced, membrane resistance and space constant are increased, and after-spike hyperpolarising potentials are suppressed. The gate is wide-open in this state and each arriving action potential in any given neurite is conducted across the cell body to activate spike discharge in all neurites around the periphery of the multipolar neuron. When this occurs, output of the discharge in the cell body is distributed to multiple synaptic sites in the circuitry by the extensive ramifications of the Dogiel Type II neurites. PRESYNAPTIC INHIBITION A significant aspect of the importance of presynaptic inhibition emerged from work on electrolyte secretion in Ussing chamber studies. The amplitude of recurrent cycles of electrolyte secretion, seen in Ussing chamber studies on guinea-pig large intestinal mucosa, was found to be smaller when the cycles were evoked by histamine than when the selective histamine H2 receptor agonist, dimaprit, was applied to evoke the cycles (Wang and Cooke, 1990). This suggests that activation of a histamine receptor different from the histamine H2 receptor subtype is involved in suppression of the neurogenic secretory activity. Phamacological and electrophysiological evaluation indicates that the receptor is the histamine H3 receptor located at presynaptic terminals or motor nerve terminals at neuroeffector junctions. Use of the drug burimamide, which was mentioned earlier in the chapter as acting as an antagonist at presynaptic histamine receptors at nicotinic synapses, helps to clarify why dimaprit evokes larger amplitude secretory cycles than histamine. The amplitude of the histamine cycles grows to the size of the cycles evoked by dimaprit when burimamide is applied in the presence of histamine. Burimamide does not change dimaprit-evoked cycles. On the other hand, application of the selective histamine H3 receptor agonist, sodium methylhistamine, reduces the amplitude of the cycles evoked by dimaprit. These observations together with neurophysiological evidence for presynaptic inhibitory histamine H3 receptors (Tamura, Palmer and Wood, 1987) point to activation of the presynaptic inhibitory receptors as the factor responsible for the blunted amplitude of the recurrent secretory cycles when evoked by histamine. The cycles are larger when evoked by dimaprit because this selective histamine H2 receptor agonist spares the presynaptic inhibitory receptors while stimulating the slow excitatory receptors on the cell bodies of the AH/ Type 2 neurons in the network. A presynaptic braking action appears to control excitability in the circuitry and thereby, the strength of outflow to the effector system. The positive feedback configuration of driver circuits, as described above, would be expected to produce runaway amplification if left unchecked after activation by histamine H2 receptors on the cell bodies. Simultaneous activation of histamine H3 receptors leading to suppression of the strength of transmission at synapses in the network applies a brake on excitability in the whole circuit and in this way maintains a check on the strength of outflow from the circuit to the effector system. The presence of two selective receptor subtypes for histamine at two key locations in the circuit seems to be basic for neuromodulatory

ENTERIC NEURO-IMMUNOPHYSIOLOGY

385

reconfiguration of the circuit into a different output mode when the circuit is exposed to a paracrine overlay of histamine.

IMMUNO-NEUROPHYSIOLOGY IN THE SENSITISED INTESTINE NEURAL BEHAVIOUR IN THE ANTIGEN-SENSITISED INTESTINE The effects of histamine and other mast cell mediators on neurophysiological functions predict similar effects when mast cells are degranulated by sensitising antigens. Guinea-pig models have proved useful for investigation of the neurophysiology of antigen-sensitised bowel. The general protocol is to induce gastrointestinal immunity and consequently hyper-sensitivity to milk protein or the intestinal parasite T. spiralis (Frieling, Cooke and Wood, 1994; Frieling et al., 1994a). Milk hypersensitivity is produced in guinea-pigs by ingestion of cow’s milk in place of drinking water for a 3-week period. Hypersensitivity to T. spiralis is produced by oral inoculation of guinea-pigs with live infective-stage larvae. After 3 weeks of milk ingestion, or 6–8 weeks after T. spiralis infection, submucous or myenteric plexus preparations are set up for intracellular recording from the neurons. Electrophysiological and synaptic behaviour of the sensitised tissue and non-sensitised controls is then compared during application of β-lactoglobulin for the milk sensitised animals or an extract of T. spiralis larvae for the nematode infection. Submucous neurons in guinea-pig colon sensitised to either milk protein or T. spiralis antigens are found to have a higher incidence of spontaneously occurring action potentials and fast nicotinic excitatory post-synaptic potentials than neurons from non-sensitised animals (Frieling, Cooke and Wood, 1994; Frieling et al., 1994a). This activity is reminiscent of the spontaneous spike discharge evoked by application of histamine in submucous plexus preparations. Application of antigens from T. spiralis larvae or β-lactoglobulin produce no changes in neuronal behaviour in preparations from nonsensitised guinea-pigs. In neurons from sensitised animals, the same application of antigen evokes electrical changes similar to slow excitatory synaptic transmission mentioned earlier in the chapter. The changes consist of slowly activating membrane depolarisation, increased input resistance, suppression of postspike hyperpolarising potentials and enhanced excitability reflected by increased frequency of repetitive spike discharge during injection of depolarising current (Figure 9.8). Application of a selective histamine H2 receptor antagonist prevents or reverses the excitatory action of antigen exposure in the sensitised tissues (Figure 9.8). This is reminiscent of the action of histamine and reversal of its action by histamine H2 antagonists. The effects of antigen exposure on fast excitatory neurotransmission at nicotinic synapses in the sensitised preparations are also reminiscent of the effects of histamine on fast synaptic behaviour. Introduction of T. spiralis antigen to the parasite-sensitised submucous plexus preparations, or β-lactoglobulin to the milk-sensitised preparation, results in suppression of fast nicotinic transmission in both (Figure 9.8). Like the presynaptic inhibitory action of histamine on fast nicotinic transmission, the inhibitory action

386

INNERVATION OF THE GASTROINTESTINAL TRACT

Figure 9.8 Effects of exposure to β-lactoglobulin or Trichinella spiralis antigen in submucous neurons from the colon of guinea-pigs previously infected with T. spiralis or cow’s milk. (A) Excitation of an AH/Type 2 submucous neuron in a milk-sensitised preparation. Each action potential evoked by injection of depolarising current pulses was followed by after-hyperpolarisation (AH) characteristic of AH/Type 2 neurons. Application of β-lactoglobulin resulted in suppression of the AH and enhanced excitability reflected by spike discharge evoked by each depolarising current pulse. (B) Enhanced neuronal excitability evoked by application of T. spiralis was reversed by application of histamine antagonists. Intrasomal injection of depolarising current pulse (lower traces) evoked 2 spikes (upper traces) at the onset of the pulse, in the control, prior to application of the antigen in the superfusion solution. In the presence of T. spiralis somatic antigen, enhanced excitability in the same neuron is seen as repetitive spike discharge at increased frequency throughout the depolarising current pulse. The excitatory action of the antigen was suppressed by cimetidine with the antigen still present.

of antigen exposure can be reversed by treatment with the histamine H3 antagonist burimamide (Figure 9.9).

EFFECTOR BEHAVIOUR IN THE ANTIGEN-SENSITISED INTESTINE Effector behaviour in the antigen-sensitised intestine generally fits with predictions derived from the actions of histamine on neural elements of the enteric nervous system. Mastocytosis occurs in the T. spiralis sensitised guinea-pig model and levels of histamine in the intestinal wall are elevated (Russell and Castro, 1989). Basal secretion from the mucosa is enhanced and secretory responses to substance P, carbachol and electrical field stimulation are facilitated in Ussing chamber studies (Wang, Palmer and Cooke, 1991).

ENTERIC NEURO-IMMUNOPHYSIOLOGY

387

Figure 9.9 Suppression of fast excitatory postsynaptic potentials (EPSPs) occurs during exposure to β-lactoglobulin in a neuron from a milk-sensitised guinea-pig. The histamine H3 receptor antagonist, burimamide, reversed suppression of the EPSP. In the control prior exposure to β-lactoglobulin, focal electrical stimulation of an interganglionic connective evoked a fast EPSP. Exposure to β-lactoglobulin suppressed the fast EPSP. Application of burimamide with the antigen present reversed the antigen effect. The amplitude of the EPSP returned to near pre-challenge level after washout of the antigen and burimamide.

Introduction of the sensitising antigen evokes recurrent cyclical secretory behaviour. Each of these forms of behaviour in the sensitised intestine is suppressed by histamine H2 antagonists. Exposure of isolated segments of intestine from sensitised animals to T. spiralis antigen evokes powerful propulsive motility (Alizadeh, Castro and Weems, 1987; Alizadeh, Weems and Castro, 1989). Histamine application mimics the propulsive motor responses and neural blockade by tetrodotoxin eliminates both the response to antigen and histamine. The highly propulsive motility evoked by antigen in the sensitised intestine in vitro is reproduced in vivo in a dog model. Instillation of T. spiralis into the large intestine of conscious dogs, after an infection with the parasite, triggered motor behaviour characteristic of power propulsion (“giant migrating contractions”) associated with watery diarrhoea (Cowles and Sarna, 1990, 1991; Sethi and Sarna, 1991).

CONCLUSIONS The immunoneurophysiological evidence leads to the conclusion that the minute-tominute behaviour of the bowel, whether it be normal or pathologic, is determined primarily

388

INNERVATION OF THE GASTROINTESTINAL TRACT

by integrative functions of the enteric nervous system. Information input processed by the brain-in-the-gut is derived from local sensory receptors, the central nervous system and a compliment of immunological cells including mast cells. Intestinal mucosal mast cells utilise the capacity of the immune system for detection of new antigens and long-term memory, for recognition of the antigen if it ever reappears in the gut lumen. Should the antigen reappear, the mast cells signal its presence to the enteric nervous system. The enteric nervous system interprets the mast cell signal as a threat and calls up from its program library secretory and propulsive motor behaviour organised for rapid and effective eradication of the threat. Operation of the “alarm” program protects the integrity of the bowel, but at the expense of side effects that include abdominal distress and diarrhoea. The signalling function of intestinal mast cells is analogous to that of sensory neurons. Sensory neurons detect changes in stimulus energy and transfer the information to integrated neural circuits for processing. Mast cells, like sensory neurons are detectors. However, unlike sensory neurons that are endowed with mechanisms for detection and transduction of a single form of stimulus, mast cells utilise immunological memory functions to detect foreign antigens as they appear throughout the life of the individual. Unlike sensory neurons, mast cells use paracrine signalling mechanisms for direct transfer of chemical information to the integrative neural circuits of the brain-in-the-gut. The integrative circuits can receive and interpret the chemical signals from the mast cells. The enteric nervous system responds to the signals as if they were a labelled code for the presence of a threat in the intestinal lumen. The histoanatomical proximity of intestinal mast cells and enteric neural elements and an expanding collection of immunophysiological observations described in this chapter indicate that mast cells in the mucosa and elsewhere in the intestine are strategically positioned to establish a first line of defence against foreign intrusion at a vulnerable interface between the body and the outside world. Evidence in this chapter implicates histamine as one of the significant paracrine signals from mast cells to the enteric nervous system. When the enteric microcircuits detect histamine, the message is processed as a labelled signal indicating mast cell degranulation, which in turn is a specific label for the presence of a sensitising antigen. Microelectrode recording from neurons in the sensitised bowel has clarified the neurophysiological events associated with antigen detection at the interface for information transfer between mucosal mast cells and enteric neurons. This approach confirmed specificity for detection and signalling of the sensitising antigen and added to the evidence for significant involvement of histamine in immuno-neural signalling. Work at the whole organ level reviewed in this chapter has established that an overlay of histamine in the enteric nervous system results in an organised pattern of coordinated behaviour of mucosal secretion and motility. The pattern of effector behaviour generated by the neural circuitry emerges from two actions of histamine at the level of the individual neural elements that make-up the circuits. One action is dramatically enhanced excitability of the membrane of the neuronal cell bodies mediated by histamine H2 receptors. The second is presynaptic inhibition mediated by histamine H3 at nicotinic synapses in the circuitry. How the actions of histamine at the cell bodies and nicotinic synapses are transformed by the integrated circuitry into the neurally programmed behaviour seen at the level of the whole organ is not yet fully understood.

ENTERIC NEURO-IMMUNOPHYSIOLOGY

389

ACKNOWLEDGEMENTS The concepts of enteric neuro-immunophysiology reviewed in this chapter emerged, in part, from the collaborative work of my former colleague, Prof. Helen J. Cooke and several postdoctoral visitors to our laboratories. These include: Fedias Christofi, Thomas Frieling, Jeffrey M. Palmer, Kenji Tamura, Yu-Z. Wang, Yun Xia and Dimiter Zafirov. The work in my laboratory on neuro-immune signalling in the enteric nervous system was supported by National Institutes of Health Grants RO1 NS17363, RO1 AM26742, and RO1 DK37238.

REFERENCES Alizadeh, H., Castro, G.A. and Weems, W.A. (1987). Intrinsic jejunal propulsion in the guinea pig during parasitism with Trichinella spiralis. Gastroenterology, 93, 784–790. Alizadeh, H., Weems, W.A. and Castro, G.A. (1989). Long-term influence of enteric infection on jejunal propulsion in guinea pigs. Gastroenterology, 97, 1461–1468. Andriantsitohaina, R. and Surprenant, A. (1992). Acetylcholine released from guinea-pig submucosal neurones dilates arterioles by releasing nitric oxide from endothelium. Journal of Physiology (London), 453, 493–502. Baird, A.W. and Cuthbert, A.W. (1987). Neuronal involvement in Type 1 hypersensitivity reactions in gut epithelia. British Journal of Pharmacology, 92, 647–655. Barajas-Lopez, C., Surprenant, A. and North, R.A. (1991). Adenosine A1 and A2 receptors mediate presynaptic inhibition and postsynaptic excitation in guinea pig submucosal neurons. Journal of Pharmacology and Experimental Therapeutics, 258, 490–495. Baum, C.A., Paramjit, B. and Miner, P.B. (1989). Increased colonic mucosal mast cells associated with severe watery diarrhea and microscopic colitis. Digestive Diseases and Sciences, 34, 1462–1465. Befus, D. (1994). Reciprocity of mast cell-nervous system interactions. In Innervation of the Gut: Pathophysiological Implications, edited by Y. Tache, D.L. Wingate and T.F. Burks, pp. 315–329. Boca Raton, FL: CRC Press. Bornstein, J.C., Hendriks, R., Furness, J.B. and Trussell, D.C. (1991). Ramifications of the axons of AH-neurons injected with the intracellular marker biocytin in the myenteric plexus of the guinea pig small intestine. Journal of Comparative Neurology, 314, 437–451. Broad, R.M., McDonald, T.J., Brodin, E. and Cook, M. (1992). Adenosine A1 receptors mediate inhibition of tachykinin release from perfused enteric nerve endings. American Journal of Physiology, 262, G525–G531. Brookes, S.J.H. and Costa, M. (1994). Enteric motor neurons. In Innervation of the Gut: Pathophysiological Implications, edited by Y. Tache, D.L. Wingate and T.F. Burks, pp. 237–248. Boca Raton, FL: CRC Press. Castagliuolo, I., LaMont, J.T., Letourneau, R., Kelly, C., O’Keane, J.C., Jaffer, A. et al. (1994). Neuronal involvement in the intestinal effects of Clostridium difficile toxin A and Vibrio cholerae enterotoxin in rat ileum. Gastroenterology, 107, 657–665. Castro, G.A. (1989). Gut immunophysiology: regulatory pathways within a common mucosal immune system. News in Physiological Sciences, 4, 59–64. Christofi, F.L. and Wood, J.D. (1993). Presynaptic inhibition by adenosine A1 receptors on guinea-pig small intestinal myenteric neurons. Gastroenterology, 104, 1420–1429. Christofi, F.L. and Wood, J.D. (1994). Electrophysiological subtypes of inhibitory P1 purinoceptors on myenteric neurones of guinea-pig small bowel. British Journal of Pharmacology, 113, 703–710. Christofi, F.L. McDonald, T.J. and Cook, M.A. (1990). Adenosine receptors are coupled negatively to release of tachykinin(s) from enteric nerve endings. Journal of Pharmacology and Experimental Therapeutics, 253, 290–295. Christofi, F.L., Tack, J. and Wood, J.D. (1992). Suppression of nicotinic synaptic transmission by adenosine in myenteric ganglia of the guinea-pig gastric antrum. European Journal of Pharmacology, 216, 17–22. Cocchiara, R., Albeggiani, G., Lampiasi, N., Bongiovanni, A., Azzolina, A. and Geraci, D. (1999). Histamine and tumor necrosis factor-α production from purified rat brain mast cells mediated by substance P. NeuroReport, 10, 575–578. Coelho, A.M., Fioramonti, J. and Bueno, L. (1998). Mast cell degranulation induces delayed rectal allodynia in rats: role of histamine and 5-HT. Digestive Diseases and Sciences, 43, 727–737.

390

INNERVATION OF THE GASTROINTESTINAL TRACT

Cooke, H.J. (1989). Role of the “little brain” in the gut in water and electrolyte homeostasis. FASEB Journal, 3, 127–138. Cooke, H.J., Wang, Y.Z. and Rogers, R. (1993). Coordination of Cl– secretion and contraction by a histamine H2-receptor agonist in guinea pig distal colon. American Journal of Physiology, 265, G973–G978. Cowles, V.E. and Sarna, S.K. (1990). Effects of T. spiralis infection on intestinal motor activity in the fasted state. American Journal of Physiology, 259, G693–G701. Cowles, V.E. and Sarna, S.K. (1991). Trichinella spiralis infection alters small bowel motor activity in the fed state. Gastroenterology, 101, 664–669. Derkach, V., Surprenant, A. and North, R.A. (1989). 5-HT3 receptors are membrane ion channels. Nature, 339, 706–709. Fiorica-Howells, E., Wade, P.R. and Gershon, M.D. (1993). Serotonin-induced increase in camp in ganglia isolated from the myenteric plexus of the guinea pig small intestine – mediation by a novel 5-HT receptor. Synapse, 13, 333–349. Frieling, T., Cooke, H. J. and Wood, J.D. (1991). Serotonin receptors on submucous neurons in the guinea-pig colon. American Journal of Physiology, 261, G1017–G1023. Frieling, T., Cooke, H.J. and Wood, J. D. (1993). Histamine receptors on submucous neurons in the guinea-pig colon. American Journal of Physiology, 264, G74–G80. Frieling, T., Cooke, H.J. and Wood, J.D. (1994). Neuroimmune communication in the submucous plexus of guinea pig colon after sensitization to milk antigen. American Journal of Physiology, 267, G1087–G1093. Frieling, T., Palmer, J.M., Cooke, H.J. and Wood, J.D. (1994a). Neuroimmune communication in the submucous plexus of guinea pig colon after infection with Trichinella spiralis. Gastroenterology, 107, 1602–1609. Frieling, T., Rupprecht, C., Kroese, A.B.A. and Schemann, M. (1994b). Effects of the inflammatory mediator prostaglandin D2 on submucosal neurons and secretion in guinea pig colon. American Journal of Physiology, 266, G132–G139. Frieling, T., Dobreva, G., Rupprecht, C., Musial, A., Becker, C. and Schemann, M. (1995). Leukotrienes activate submucosal neurons to evoke cyclical chloride secretion in guinea-pig colon. Gastroenterology, 108, A819. Gottwald, T.P., Hewlett, B.R., Lhotak, S. and Stead, R.H. (1995). Electrical stimulation of the vagus nerve modulates the histamine content of mast cells in the rat jejunal mucosa. NeuroReport, 7, 313–317. Gue, M., Del Rio-Lacheze, C., Eutamene, H., Theodorou, V., Fioramonti, J. and Bueno, L. (1997). Stressinduced visceral hypersensitivity to rectal distension in rats: role of CRF and mast cells. Neurogastroenterology and Motility, 9, 271–279. Harari, Y., Russell, D.A. and Castro, G.A. (1987). Anaphylaxis mediated epithelial Cl secretion and parasite rejection in rat intestine. Journal of Immunology, 128, 1250–1255. Janiszewski, J., Bienenstock, J. and Blennerhassett, M.D. (1994). Picomolar doses of substance P trigger electrical responses in mast cells without degranulation. American Journal of Physiology, 267, C138–C145. Kunze, W.A.A., Furness, J.B. and Bornstein, J.C. (1993). Simultaneous intracellular recordings from enteric neurons reveal that myenteric AH neurons transmit via slow excitatory postsynaptic potentials. Neuroscience, 55, 685–694. MacQueen, G., Marshall, J., Perdue, M., Siegel, S. and Bienenstock, J. (1989). Pavolvian conditioning of rat mucosal mast cells to secrete rat mast cell protease II. Science, 243, 83–84. Maggi, C.A. (1995). The mammalian tachykinin receptors. General Pharmacology, 26, 911–944. Maggi, C. A., Patacchini, R., Meini, S., Quartara, L., Sisto, A., Potier, E. et al. (1994). Comparison of tachykinin NK-1 and NK-2 receptors in the circular muscle of the guinea-pig ileum and proximal colon. British Journal of Pharmacology, 112, 150–160. Makhlouf, G.M. and Grider, J.R. (1993). Nonadrenergic noncholinergic inhibitory transmitters of the gut. News in Physiological Sciences, 8, 195–199. Mathias, J.R., Carlson, G.M., DiMarino, A.J., Bertiger, G., Morton, H.E. and Cohen, S. (1976). Intestinal myoelectric activity in response to live Vibrio cholerae and cholera enterotoxin. Journal of Clinical Investigation, 58, 91–96. Mawe, G.M., Branchek, T.A. and Gershon, M.D. (1986). Peripheral neural serotonin receptors: Identification and characterization with specific antagonists and agonists. Proceedings of the National Academy of Sciences USA, 83, 9799–9803. Mayer, C.J. and Wood, J.D. (1975). Properties of mechanosensitive neurons within Auerbach’s plexus of the small intestine of the cat. Pflugers Archiv, 357, 35–49. McDonald, T. (1993). Antigen-independent and antigen-dependent development of mucosal immunity and hypersensitivity. In Immunophysiology of the Gut, edited by W.A. Walker, P.R. Harmatz and B.K. Wershil, pp. 99–109. San Diego: Academic Press. McKay, D.M. and Perdue, M.H. (1993a). Intestinal epithelial function: the case for immunophysiological regulation. Cells and mediators (1). Digestive Diseases and Sciences, 38, 1377–1387.

ENTERIC NEURO-IMMUNOPHYSIOLOGY

391

McKay, D.M. and Perdue, M.H. (1993b). Intestinal epithelial function: the case for immunophysiological regulation. Implications for disease (2). Digestive Diseases and Sciences, 38, 1735–1745. Murray, J., Du, C., Ledlow, A., Bates, J.N. and Conklin, J.L. (1991). Nitric oxide – mediator of nonadrenergic noncholinergic responses of opossum esophageal muscle. American Journal of Physiology, 261, G401–G406. Nemeth, P.R., Ort, C.A. and Wood, J.D. (1984). Intracellular study of effects of histamine on electrical behaviour of myenteric neurons in guinea-pig small intestine. Journal of Physiology (London), 355, 411–425. North, R.A., Henderson, G., Katayama, Y. and Johnson, S.M. (1980). Electrophysiological evidence for presynaptic inhibition of acetylcholine release by 5-hydroxytryptamine in the enteric nervous system. Neuroscience, 5, 581–586. Palmer, J.M., Wood, J.D. and Zafirov, D.H. (1986). Elevation of cyclic adenosine monophosphate mimics slow synaptic excitation in myenteric neurones of the guinea-pig. Journal of Physiology (London), 376, 451–460. Palmer, J.M., Wood, J.D. and Zafirov, D.H. (1987a). Transduction of aminergic and peptidergic signals in enteric neurones of the guinea-pig. Journal of Physiology (London), 387, 371–383. Palmer, J.M., Wood, J.D. and Zafirov, D.H. (1987b). Purinergic inhibition in the small intestinal myenteric plexus of the guinea-pig. Journal of Physiology (London), 387, 357–369. Panula, P., Kaartinen, M., Macklin, M. and Costa, E. (1985). Histamine-containing peripheral neuronal and endocrine systems. Journal of Histochemistry and Cytochemistry, 33, 933–941. Phillips, S.F. (1995). Motility disorders of the colon. In Textbook of Gastroenterology, edited by T. Yamada, D.H. Alpers, C. Owyang, D.W. Powell and F.E. Silverstein, pp. 1856–1875. Philadelphia: J.B. Lippincott. Pothoulakis, C., Castagliuolo, I. and LaMont, J.T. (1998). Nerves and intestinal mast cells modulate responses to enterotoxins. News in Physiological Sciences, 13, 58–63. Rachmilewitz, D., Karmeli, F., Eliakim, R., Pothoulakis, C., Joshi, M.A. and Kelly, C.P. (1992). Ketotifen decreases intestinal secretion and inflammation induced by Clostridium difficile toxin A (TxA) in rat ileum. Gastroenterology, 102, A235. Repke, H., Piotrowski, W., Bienert, M., and Foreman, J.C. (1987). Histamine release induced by Arg-Pro-LysPro(CH2)11CH3 from rat peritoneal mast cells. Journal of Pharmacology and Experimental Therapeutics, 243, 317–321. Ruhl, A., Hurst, S. and Collins, S.M. (1994). Synergism between interleukins 1 beta and 6 on noradrenergic nerves in rat myenteric plexus. Gastroenterology, 107, 993–1001. Sanders, K.M. and Ward, S.M. (1992). Nitric oxide as a mediator of nonadrenergic noncholinergic neurotransmission. American Journal of Physiology, 25, G379–G392. Santos, J., Saperas, E., Mourelle, M., Antolin, M. and Malagelada, J.-R. (1996). Regulation of intestinal mast cells and luminal protein release by cerebral thyrotropin-releasing hormone in rats. Gastroenterology, 111, 1465–1473. Santos, J., Saperas, E., Nogueiras, C., Mourelle, M., Antolin, M., Cadahia, A. et al. (1998). Release of mast cell mediators into the jejunum by cold pain stress in humans. Gastroenterology, 114, 640–648. Sarna, S.K., Otterson, M.F., Cowles, V.E., Sethi, A.K. and Telford, G.L. (1991). In vivo motor response to gut inflammation. In Effects of immune cells and inflammation on smooth muscle and enteric nerves, edited by S.M. Collins and W.J. Snape, pp. 181–195. Boca Raton, FL: CRC Press. Schanbacher, L.M., Nations, J.K., Weisbrodt, N.W. and Castro, G.A. (1978). Intestinal myoelectric activity in parasitized dogs. American Journal of Physiology, 234, R188–R195. Sethi, A.K. and Sarna, S.K. (1991). Colonic motor activity in acute colitis in conscious dogs. Gastroenterology, 100, 954–963. Shanahan, F.L., Denburg, J.A., Fox, J., Bienenstock, J. and Befus, A.D. (1985). Mast cell heterogeneity. Effects of neuroenteric peptides on histamine release. Journal of Immunology, 135, 1331–1337. Stead, R.H., Tomioka, M., Quinonez, G., Simon, G.T., Felten, S.Y. and Bienenstock, J. (1987). Intestinal mucosal mast cells in normal and nematode-infected rat intestines are in intimate contact with peptidergic nerves. Proceedings of the National Academy of Sciences USA, 84, 2975–2979. Surprenant, A. and Crist, J. (1988). Electrophysiological characterization of functionally distinct 5-hydroxytryptamine receptors on guinea-pig submucous plexus. Neuroscience, 24, 283–295. Tack, J., Janssens, J., Vantrappen, G. and Wood, J.D. (1992). Actions of 5-hydroxytryptamine on myenteric neurons in the guinea-pig gastric antrum. American Journal of Physiology, 263, G838–G846. Tamura, K. and Wood, J.D. (1992). Effects of prolonged exposure to histamine. Digestive Diseases and Sciences, 37, 1084–1088. Tamura, K., Palmer, J.M. and Wood, J.D. (1987). Presynaptic inhibition produced by histamine at nicotinic synapses in enteric ganglia. Neuroscience, 25, 171–179. Tamura, K., Schemann, M. and Wood, J.D. (1993). Actions of nitric oxide-generating sodium nitroprusside in myenteric plexus of guinea-pig small intestine. American Journal of Physiology, 159, 107–110. Tamura, K., Itoh, H. and Wood, J.D. (1995). G-proteins couple excitatory and inhibitory receptors to adenylate cyclase in AH/type 2 myenteric neurons of guinea-pig small intestine. Gastroenterology, 108, A1010.

392

INNERVATION OF THE GASTROINTESTINAL TRACT

Thomas, E.A., Bertrand, P.P. and Bornstein, J.C. (1999). Genesis and role of coordinated firing in a feed forward network: A model study of the enteric nervous system. Neuroscience, 93, 1525–1537. Vanner, S. and Surprenant, A. (1996). Neural reflexes controlling the intestinal microcirculation. American Journal of Physiology, 271, G223–G230. Wade, P.R. and Wood, J.D. (1988). Actions of serotonin and substance P on myenteric neurons of guinea-pig distal colon. European Journal of Pharmacology, 148, 1–8. Wang, Y.Z. and Cooke, H.J. (1990). H2 receptors mediate cyclical chloride secretion in guinea pig distal colon. American Journal of Physiology, 258, G887–G893. Wang, Y., Palmer, J. and Cooke, H. (1991). Neuro-immune regulation of colonic secretion in guinea pigs. American Journal of Physiology, 260, G307–G314. Wang, L., Stanisz, A.M., Wershil, B.K., Galli, S.J. and Perdue, M. (1995). Substance P induces ion secretion in mouse small intestine through effects on enteric nerves and mast cells. American Journal of Physiology, 269, G85–G92. Williams, R.M., Berthoud, H.R. and Stead, R. (1995). Association between vagal afferent nerve fibers and mast cells in rat jejunal mucosa. Gastroenterology, 108, A941. Wood, J.D. (1989). Electrical and synaptic behavior of enteric neurons. In Handbook of Physiology, The Gastrointestinal System, Motility and Circulation, edited by J.D. Wood, pp. 465–517. Bethesda, MD: American Physiological Society. Wood, J. D. (1991). Communication between minibrain in gut and enteric immune system. News in the Physiological Sciences, 6, 64–69. Wood, J.D. (1994). Physiology of the enteric nervous system. In Physiology of the Gastrointestinal Track, 3rd edition, edited by L.R. Johnson, J. Christensen, M.J. Jackson, E.D. Jacobson and J.H. Walsh, pp. 423–482. New York: Raven Press. Wood, J.D. (1995a). Gastrointestinal neurophysiology. In Medical Physiology, edited by R.A. Rhoades and G.A. Tanner, pp. 487–504. Boston: Little, Brown Co. Wood, J. D. (1995b). Gastrointestinal motility. In Medical Physiology, edited by R.A. Rhoades and G.A. Tanner, pp. 505–529. Boston: Little, Brown Co. Wood, J.D. (1995c). Physiological and pathophysiological paracrine functions of intestinal mast cells in enteric neuro-immune signalling. In Gastrointestinal Tract and Endocrine System, edited by M.F. Singer, R. Ziegler and G. Rohr, pp. 254–263. Dordrecht: Kluwer Academic Publishers. Wood, J.D. and Mayer, C.J. (1979). Serotonergic activation of tonic-type enteric neurons in guinea-pig small bowel. Journal of Neurophysiology, 42, 582–593. Xia, Y., Fertel, R.H. and Wood, J.D. (1994). Stimulation of formation of cAMP by 5-hydroxytryptamine in myenteric ganglia isolated from guinea pig small intestine. Life Sciences, 55, 685–692. Xia, Y., Zafirov, D.H., Cooke, H.J. and Wood, J.D. (1995). Actions of tumor necrosis factor (TNF-α) on electrical and synaptic behavior in the submucous plexus of guinea-pig small intestine. Gastroenterology, 108, A945. Xia, Y., Fertel, R. and Wood, J. (1996). Stimulation of formation of adenosine 3’,5’-phosphate by histamine in myenteric ganglia isolated from guinea-pig small intestine. European Journal of Pharmacology, 316, 81–85. Xia, Y., Lu, G., Zafirov, D.H., Sarna, S.K. and Wood, J.D. (1996). Effects of platelet-activating factor on electrical and synaptic behavior of neurons in the submucous plexus of guinea-pig small intestine. Gastroenterology, 110, A782. Xia, Y., Fertel, R.H. and Wood, J.D. (1997). Suppression of cAMP formation by adenosine in myenteric ganglia of guinea-pig small intestine. European Journal of Pharmacology, 320, 95–101. Xia, Y., Hu, H.-Z., Liu, S., Ren, J., Zafirov, D. and Wood, J.D. (1999). IL-1β and IL-6 excite neurons and suppress nicotinic and noradrenergic neurotransmission in guinea-pig enteric nervous system. Journal of Clinical Investigation, 103, 1309–1316. Zafirov, D.H., Palmer, J.M. and Wood, J.D. (1985). Adenosine inhibits forskolin-induced excitation in myenteric neurons. European Journal of Pharmacology, 113, 143–144.

10 Cellular Organisation of the Mammalian Enteric Nervous System Simon J.H. Brookes and Marcello Costa Department of Human Physiology and Centre for Neuroscience, Flinders University, GPO Box 2100, Adelaide, South Australia 5001 The enteric nervous system plays a central role in the control of gastrointestinal function. To understand how its circuits operate, it is necessary to identify and characterise the different functional classes of enteric neurones and determine their cellular properties and connectivity. The most powerful way to distinguish the different classes of neurones has been to study the combinations of neurochemicals (“chemical coding”) that they contain in their cell bodies and axons. Some of the neurochemicals in the chemical codes, such as neurotransmitters, may be functionally important in their own right. The role of other molecules, such as calcium binding proteins, are currently unclear, however they are still useful as markers to distinguish various classes of neurones. In combination with axonal tracing techniques, chemical coding has provided a quantitative account of all of the major classes of neurones in one preparation, the guinea-pig ileum. Extensive electrophysiological and morphological studies have been readily incorporated into this account. This has revealed an exquisite degree of organisation, comparable to many parts of the central nervous system. Each class of neurone has characteristic combinations of neurotransmitters, neuromodulators, synaptic inputs, projections and soma-dendritic morphology. The methodologies developed in the guinea-pig small intestine are increasingly being applied to other regions of gastrointestinal tract and to preparations from other species. While many characteristics appear to be shared by different preparations, substantial differences are beginning to emerge, even between neurones with apparently identical functions. It is clear that the guinea-pig ileum is not representative of the enteric nervous system in general, nor of the human enteric nervous system in particular. However, neither is any other animal model. Fortunately, with the techniques now available, it is possible to study directly specimens of human gut, reducing the reliance on animal models to answer fundamental questions of the cellular organisation of the human enteric nervous system. Changes in the chemical coding during development and in diseases will cast light onto both the mechanisms underlying plasticity of the enteric nervous system and the processes underlying physiopathological changes. KEY WORDS: enteric nervous system; afferent neurones; motor neurones; interneurones; neurotransmission; chemical coding.

INTRODUCTION REASONS FOR STUDYING THE ENTERIC NERVOUS SYSTEM The enteric nervous system plays an important role in the day to day survival of the organism. The control of motility is essential for the digestion and absorption of nutrients. 393

394

INNERVATION OF THE GASTROINTESTINAL TRACT

The coordinated movements of the gut allow mixing and sufficient exposure of the contents to the absorptive surface of the wall while avoiding stasis, which would allow the gut flora to overwhelm the body’s defences. In addition, by optimising the rate of propulsion, together with mucosal secretory activity and blood flow, the enteric nervous system plays a profoundly important role in setting the Na+, K+ and water balance of the whole body. When these control mechanisms are compromised, for example by infection, severe diarrhoea can result. Many of the readers of this volume live in relatively wealthy countries, with reliable water supplies, good infrastructure and adequate provision of medical services. In this case, such diarrhoea is unpleasant and inconvenient, as are many of the other disorders involving the enteric nervous system. In less wealthy parts of the world, where water supplies are limited and sometimes contaminated and where medical services are unavailable or unaffordable, dehydration caused by diarrhoea is a major cause of mortality. Children and the elderly people are especially vulnerable. Even in western countries, chronic dysfunction of the enteric nervous system has an enormous impact on quality of life. In the most severe cases, such as pseudo-obstruction or Hirschsprung’s disease, radical surgical treatment is urgently required. The vital role played by the enteric nervous system in the functioning of the entire body, provides a powerful justification for basic research aimed at understanding how this circuitry operates in health and disease. For many of us working in this field, there is another rationale for our study. Understanding how spindly little nerve cells can interact to give rise to adaptive, coordinated behaviour holds a deep intellectual fascination. For neurobiologists of a faint-hearted disposition, who are intimidated by the complexity of the central nervous system, the enteric nervous system is a wonderful alternative. The gut is the only organ in mammals (with the debatable exception of the heart) where entire neuronal circuits including sensory neurones, interneurones and motor neurones can be isolated from the central nervous system and still show reflex responses of physiological significance. Thus it is possible to study the cellular basis of entire behaviours which merely involve tens of thousands of neurones, rather than the uncountable millions potentially influencing central pathways. Understanding enteric neuronal circuits still represents a considerable intellectual challenge, which hopefully will identify some of the principles of organisation which may be shared by other neuronal circuits including those in the brain. AIMS OF THIS REVIEW The enteric nervous system and its control of gastrointestinal function have been extensively studied using morphological, electrophysiological, biochemical and pharmacological approaches. Each approach provides different types of information about the cells of the ENS. It is clear that to understand the cellular basis of neuronal control, it is necessary to have an accurate characterisation of the types of nerve cells that make up the circuitry. An adequate description of a nerve cell must incorporate morphological, electrophysiological, biochemical and pharmacological aspects. Fortunately, a substantial number of “multi-disciplinary” studies of the ENS have been carried out, in which electrophysiological and pharmacological approaches have been combined. This combinatorial approach

CELLULAR ORGANISATION OF THE MAMMALIAN ENTERIC NERVOUS SYSTEM

395

has been most extensively pursued in the guinea-pig small intestine, by a number of groups around the world. The aim of this review is to attempt to synthesise some of the more significant findings into an account of the different functional classes of nerve cells in this preparation. We will then discuss some of the similarities and differences between functional classes of cells in other regions of the guinea-pig gut and in the gastrointestinal tract of other species.

A BRIEF DESCRIPTION OF THE ENTERIC NERVOUS SYSTEM The enteric nervous system, classified as the third division of the autonomic nervous system (Langley, 1921), is made up of the enormous number of nerve cells which lie within the wall of the gut. These nerve cells have cell bodies located in two ganglionated plexuses. The myenteric plexus (Auerbach’s plexus) lies between the outer longitudinal smooth muscle layer and the circular smooth muscle (see Figure 10.1). The submucous plexus is closely associated with the thick layer of connective tissue lying between the

Figure 10.1 Dimensions of the enteric nervous system of the guinea-pig ileum. A tube of intestine, maximally distended has a diameter of approximately 6 mm. From this a 1 mm ring is shown at higher magnification, with the myenteric plexus schematically shown. At the right is a higher magnification of the myenteric plexus, with ganglia arranged in rows. At maximal stretch there are approximately 2–2.5 rows of ganglia per millimetre.

396

INNERVATION OF THE GASTROINTESTINAL TRACT

circular muscle and the innermost mucosal tissue. In small animals such as the guinea-pig, the submucous plexus forms basically a single network, however in larger animals, including humans; it can be anatomically subdivided. There is an inner submucous plexus (Meissner’s plexus) and an outer submucous plexus (Schabadasch’s or Henle’s plexus) with an intermediate plexus, at least in humans (Hoyle and Burnstock, 1989). It has been estimated that the guinea-pig small intestine contains 2,750,000 myenteric neurones and 950,000 submucous neurones (Gabella, 1987). The mouse small intestine has a higher density of neurones, but lower overall numbers (about 400,000 myenteric and 330,000 submucous neurones) due to its shorter length, whereas the sheep, with its extended small intestine, was estimated to have over 30,000,000 myenteric and over 50,000,000 submucous neurones. It appears that, on average, the individual nerve cells have a greater size in the larger species (Gabella and Trigg, 1984). These estimates of numbers of enteric neurones may in fact be underestimates, since they are based on a histochemical method that does not stain every nerve cell body (Young et al., 1993a; Karaosmanoglu et al., 1996).

CLASSIFICATION OF ENTERIC NEURONES As in the rest of the nervous system, the millions of enteric nerve cells are not entirely different from one another; they appear to be organised into a relatively small number of different classes. Neurones of each class share a combination of characteristics, which distinguishes them from neurones in all other classes. These characteristics include where they project to (i.e. their targets), their soma-dendritic morphology, the combinations of transmitters and modulators that they contain (“chemical coding”), their electrophysiological properties, their synaptic inputs and their synaptic outputs. The expression of these features is under genetic control, played out as a cascade of interacting mechanisms during the development of the enteric nervous system (see chapter by Michael Gershon in this volume). It is likely that a relatively small number of key developmental events determine the phenotype of each and every enteric neurone. Thus, the ultimate way to distinguish classes of enteric nerve cells would be to identify the crucial decision points in their ontogeny. While enormous advances have been made in the last decade in understanding the colonisation of the gut by enteric nerve cell precursors (Gershon, 1997; Gershon, 1998; Taraviras and Pachnis, 1999), the factors which determine most of the differences between classes of cells have yet to be identified. For this reason, to identify the classes of enteric nerve cells, it is necessary to examine their characteristics in the mature ENS, at the single cell level, and determine how these vary systematically.

CHEMICAL CODING Since the development of indirect immunohistochemical staining methods (Coons, 1958), the localisation of neurochemicals has been one of the most successful means to distinguish different classes of nerve cells from one another. An enormous number of studies have now been carried out analysing the distribution of one substance at a time in nerve

CELLULAR ORGANISATION OF THE MAMMALIAN ENTERIC NERVOUS SYSTEM

397

cells and other cells of the gastrointestinal tract. Immunohistochemical localisation is now a standard tool for determining where a particular protein or peptide is expressed. A recent search has shown that over 1000 studies have used immunohistochemical methods to localise substances in enteric nerve cells. Many of these studies have been carried out with the aim of characterising a substance and its distribution. They give invaluable information about the range of neurochemicals present in enteric ganglia. The ability of such data to distinguish between different functional classes of nerve cells however, is usually rather limited, since this has not been the primary aim of most of the studies. We will not attempt to summarise all of the data in these many studies. Rather we will attempt to review those studies in which systematic attempts have been made to distinguish between different types of enteric neurones. The finding that cells can contain more than one releasable, bioactive compound was an important discovery. First suggested in endocrine cells (Pearse, 1969), co-existence of neurochemical markers was demonstrated for enteric neurones in the early 1980s (Schultzberg, 1980). Subsequent use of immunohistochemical double-labelling, and later multiple-labelling, led to the concept of “chemical coding” which suggests that each class of neurone, with its characteristic projections, morphology and other features, can be distinguished by its particular combination of neurochemicals (Costa, Furness and Gibbins, 1986). This has turned out to be a very powerful way to distinguish different classes of neurones, which has been widely applied to enteric neurones and to other parts of the nervous system. In combination with selective lesions (“myectomies” and “myotomies”, as well as transmural crushes), chemical coding has been used to identify the gross projections of neurochemically defined classes of enteric neurones. It should be mentioned that this approach, however, has consistently under-estimated the lengths of projections of enteric neurones which can be measured more accurately and quantitatively using retrograde tracing techniques (Brookes, 2001). It has become clear that to distinguish different classes of enteric neurones from one another requires systematic and quantitative analysis of patterns of co-existence of multiple markers (see Figure 10.2). This has been most thoroughly achieved in the guinea-pig small intestine (Costa et al., 1996). For this reason, we will describe in detail the classes of nerve cells in the guinea-pig small intestine, integrating data on the chemical coding, projections, morphology, electrophysiological characteristics and connectivity of these neurones. We will then compare this data with the rather more sparse information about classes of enteric nerve cells in other parts of the guinea-pig gastrointestinal tract. Lastly, we will briefly discuss some of the major similarities and differences that have been identified in the gastrointestinal tract of other species that have been systematically studied with multiple labelling immunohistochemistry, lesions or retrograde tracing techniques. TERMINOLOGY Throughout this review the word “type” will be used to describe a subset of neurones which can be distinguished by the presence of a particular characteristic. For example, neurones which are immunoreactive for vasoactive intestinal polypeptide (VIP) are defined as a VIP-immunoreactive “type”. It is likely that types can be further subdivided

398

INNERVATION OF THE GASTROINTESTINAL TRACT

A

B

C

D

Figure 10.2 Distinguishing nerve cell bodies in myenteric ganglia by chemical coding. Double labelling for the calcium binding proteins calbindin and calretinin (A) is shown combined in a single micrograph. Although colour information is lost, calretinin cells appear paler. The outlines of the cells are shown in (B). Note the smooth outlines of calbindin-immunoreactive Dogiel type II cells (oblique crosshatching) and the more irregular, Dogiel type I outlines of the calretinin immunoreactive cells (vertical hatching). In (C) a preparation was triple labelled for NOS, calretinin and NFP triplet. In (D) the chemical coding of the labelled cell bodies is shown as horizontal hatching (Neurofilament protein triplet), vertical hatching (calretinin) and as grey tint (NOS). Note that NOS and calretinin do not coexist, but NOS and NFP are found in many nerve cell bodies.

on the basis of other characteristics. The term “class” will be applied to groups of neurones which share a number of characteristics and which cannot be subdivided on the basis of current knowledge. A “functional class” is one in which a role can be deduced from the projections of the cells, their pattern of activation and their transmitter content. It should be noted that within any class, there may be variability in some characteristics. It is important to avoid subdividing classes unnecessarily, since this undermines the aim of classification, which is to reduce number of cells to manageable limits. A useful criterion for identifying classes is that the cells in each class differ from those of other classes by at least two co-varying characteristics. This largely overcomes the problem of mistaking variability within a single characteristic (e.g. intensity of immunoreactivity or size of the cell soma) for differences between functional classes of cells. For example, some Dogiel type II cells (which are characterised by multiple long processes arising from a smooth cell body) are immunoreactive for calbindin, while others are not (Iyer et al., 1988; Brookes et al., 1995). In this review they will be grouped in the same class because, according to current knowledge, they do not differ consistently on any other characteristic such as projection, morphology or neurochemical coding. In contrast, some Dogiel type II cells have a “dendritic” appearance with numerous short filamentous dendrites (Stach, 1989; Bornstein et al., 1991b; Brookes et al., 1995). Many of these “dendritic” cells have long descending projections within the myenteric plexus which “non-dendritic” Dogiel type II cells lack (Brookes et al., 1995). They also make rather fewer varicose endings in their ganglion of origin, compared to other Dogiel type II cells (Bornstein et al., 1991b). This suggests that “dendritic” Dogiel type II cells may comprise a separate class, although their function has not yet been unequivocally identified.

CELLULAR ORGANISATION OF THE MAMMALIAN ENTERIC NERVOUS SYSTEM

399

PRIMARY AFFERENT NEURONES IDENTIFICATION OF ENTERIC PRIMARY AFFERENT NEURONES Early studies on extrinsically denervated intestine by Bayliss and Starling indicated that the enteric nervous system was capable of mediating entire reflexes, including peristalsis, in the absence of connections with the central nervous system (Bayliss and Starling, 1899). In their landmark paper, they wrote: “Local stimulation of the gut produces excitation above and inhibition below the excited spot. These effects are dependent on the activity of the local nervous mechanism.” Their observations were supported by later studies on isolated preparations of intestine, which confirmed that the enteric nervous system was entirely capable of producing integrated, adaptive responses to physiological stimuli (Langley and Magnus, 1905; Trendelenburg, 1917). An inescapable conclusion of those studies was that there must be primary afferent neurones within the enteric nervous system, which are capable of transducing physiological stimuli into neuronal activity, in order to activate enteric neuronal circuitry. In the last decade, this view has been challenged (Wood, 1994a). It has been reported that motor reflexes evoked by distension in the isolated rat colon were abolished by prior extrinsic denervation, suggesting that intramural collaterals of extrinsic afferent nerve fibres (either spinal or vagal in origin) were required for the reflex activity (Grider and Jin, 1994). It is clear that spinal afferent neurones have extensive collateral branching within enteric ganglia (Gibbins et al., 1985) and the same applies to vagal afferent nerve fibres (Berthoud and Powley, 1992). However, extrinsic denervation of the guinea-pig ileum has been demonstrated to have little effect on distension-activated motor reflexes (Furness et al., 1995). In this study, the effectiveness of extrinsic denervation was tested by confirming the disappearance of immunoreactivity for tyrosine hydroxylase, a marker of extrinsic sympathetic nerve fibres. In our laboratory, we have demonstrated that slow circumferential stretch of isolated segments of guinea-pig ileum evokes, at a sharp threshold, an abrupt contraction of the circular muscle corresponding to the initiation of peristalsis (Brookes et al., 1999). This motor activity, which is somewhat more complex in nature than the simple ascending reflex, is preserved after 4 days in organ culture, when extrinsic nerve fibres have degenerated (unpublished observations). These studies strongly suggest that intrinsic sensory neurones are present, and trigger motor reflexes, in the absence of extrinsic afferent collaterals within the gut wall. For a considerable period, the identity of the primary afferent neurones in the enteric nervous system remained uncertain. It was speculated, on the basis of morphology, that the multipolar enteric neurones, known as Dogiel type II neurones, were likely to be sensory in function (Dogiel, 1899). Early intracellular recordings from enteric neurones revealed a type of cell with a long after-hyperpolarisation following their action potentials, which have come to be known as AH cells (Hirst, 1974) or AH type II cells (Wood, 1994a). These cells were reported to lack fast synaptic inputs to drive their firing, unlike the majority of cells recorded. It was speculated that they might be primary afferent neurones, with no need for a synaptic drive (Hirst, 1974). Later, it was shown that AH cells have Dogiel type II morphology (Bornstein et al., 1984a; Erde, Sherman and Gershon, 1985; Katayama, Lees and Pearson, 1986; Iyer et al., 1988; Hendriks, Bornstein

400

INNERVATION OF THE GASTROINTESTINAL TRACT

and Furness, 1990; Brookes et al., 1995). Interestingly, a few years after their initial characterisation, AH cells were shown to receive excitatory synaptic inputs, although these had a much slower time course than the fast, nicotinic synaptic inputs recorded in other types of cells (Wood and Mayer, 1978).

FUNCTIONAL IDENTIFICATION OF ENTERIC PRIMARY AFFERENT NEURONES Mucosal stimuli Identification of the enteric primary afferent neurones required functional studies. The first of these used expression of the immediate early gene product, Fos, to identify nerve cell bodies in the submucous ganglia of the guinea-pig small intestine which had been activated by cholera toxin or mechanical stimulation (using bubbles of nitrogen, blown against the villi) (Kirchgessner, Tamir and Gershon, 1992). Hexamethonium, a blocker of nicotinic neurotransmission, was used to prevent indirect activation of pathways, which might otherwise have confounded the results. One particular class of cells was identified in the submucous ganglia that had developed Fos expression in responses to the stimuli. These cells were immunoreactive for the calcium binding protein, calbindin, and/or the neuropeptide, substance P. Later studies have shown that these neurones comprise approximately 13% of all submucous neurones in the guinea-pig small intestine (Song et al., 1992) and that they can be retrogradely labelled from both the mucosa and from the myenteric plexus. Intracellular dye fills have shown that these cells have morphology similar to the Dogiel type II cells in the myenteric plexus, with several long axonal processes which appear to contact neurones in nearby ganglia (Bornstein et al., 1989; Evans, Jiang and Surprenant, 1994). In addition they have electrical properties similar to the AH cells of the myenteric plexus, with rare fast synaptic inputs, broad action potentials with a characteristic inflection on the falling phase and a long after-hyperpolarisation (Bornstein et al., 1989). Later it was shown that similar submucous primary afferent neurones were involved in entero-pancreatic reflexes evoked by stimulating the mucosa with either nitrogen bubbles or glucose (Kirchgessner, Liu and Gershon, 1996). Interestingly, these submucous primary afferent neurones appear to be activated indirectly by mechanical stimuli applied to the mucosa. Superfusion with the 5-HT1P receptor antagonist N-acetyl-5-hydroxytryptophyl-5-hydroxytryptophan amide, blocked the activation of these neurones, suggesting that non-neuronal enterochromaffin cells in the mucosa may have activated the nerve terminals of the primary afferent neurones via the release of 5-hydroxytryptamine (5-HT) (Kirchgessner, Tamir and Gershon, 1992). A recent study has suggested that an additional class of submucous neurones may also function as primary afferent neurones. Using both expression of Fos and uptake of the styryl dye FM2–10, it has been suggested that a subset of calcitonin gene-related peptide (CGRP)immunoreactive neurones in the submucous plexus may also function as primary afferents, probably being activated by 5-HT (Kirchgessner, Tamir and Gershon, 1992) released from enterochromaffin cells in the mucosa (Pan and Gershon, 2000). This is interesting because CGRP has been reported in a separate class of unipolar submucous neurones which are also immunoreactive for neuropeptide Y (NPY), choline acetyltransferase

CELLULAR ORGANISATION OF THE MAMMALIAN ENTERIC NERVOUS SYSTEM

401

(ChAT) and a number of other markers (see section “Submucosal neurones and other mucosally-projecting neurones”). Numerous myenteric neurones expressed Fos when the mucosa was stimulated by nitrogen bubbles or by choleragenoid, however this was largely blocked when hexamethonium was included in the bathing medium. It was suggested that myenteric neurones were indirectly activated by the stimuli and that there were no primary afferent neurones in the myenteric plexus which responded to these stimuli (Kirchgessner, Tamir and Gershon, 1992). Later it was shown that myenteric Dogiel type II neurones in the small intestine are quite resistant to the expression of Fos and only expressed low levels, even in response to prolonged electrical stimulation (Ritter, Costa and Brookes, 1997). This leaves open the possibility that myenteric Dogiel type II neurones might function as primary afferents to the mucosa. Earlier studies, using retrograde labelling in vitro with the carbocyanine dye, DiI, had shown that probably all Dogiel type II cells in the myenteric plexus have a projection to the mucosa, in the guinea-pig ileum (Song, Brookes and Costa, 1994a). Direct evidence supporting a role for Dogiel type II cells in transducing mucosal stimuli came from the elegant studies of Kunze and colleagues (Kunze, Bornstein and Furness, 1995; Bertrand et al., 1997) who made intracellular recordings from AH cells while applying substances to the mucosa. AH cells fired action potentials in response to solutions of high and low pH and to neutral solutions of short chain fatty acids. Many of the responses were not blocked in AH cells by bathing solutions containing a low concentration of calcium ions, indicating that the responses were not indirectly mediated by other cells synapsing onto the afferent terminal. In contrast, many S cells responded to mucosally applied chemicals with bursts of fast excitatory post synaptic potentials: these were blocked in low calcium solution, indicating that they reflected indirect activation, presumably via primary afferent neurones. Thus it can be concluded that some myenteric Dogiel type II neurones respond to chemical stimulation of the mucosa and then activate other enteric neurones.

Stretch stimuli It has been known since the nineteenth century that stretch is also a powerful stimulus for activating enteric neuronal circuitry. Bayliss and Starling demonstrated that distension of the dog small intestine triggers characteristic polarised reflexes in the smooth muscle layers (Bayliss and Starling, 1899). Later it was shown that distension of isolated intestine by fluid could trigger more complex propagating, peristaltic activity (Trendelenburg, 1917). It is likely that the transduction sites of the primary afferent neurones involved in these responses lie outside the mucosa. Removal of the mucosa and submucosa does not prevent the expression of stretch-induced polarised reflexes in the guinea-pig small intestine (Smith, Bornstein and Furness, 1990), although it does block reflex responses to mucosal distortion (Smith and Furness, 1988), nor does it prevent peristalsis in this preparation (Yokoyama et al., 1990; Tsuji et al., 1992). Identification of the stretch-sensitive myenteric neurones was hampered for a long time by movements of the preparation dislodging the microelectrode. This was solved by a modification of the floating microelectrode technique (Woodbury and Brady, 1956) which allowed impalements to be maintained while the level of stretch of the preparation was changed (Kunze, 1998).

402

INNERVATION OF THE GASTROINTESTINAL TRACT

Using this approach, it was shown that a subset of Dogiel type II neurones fired spontaneously when the preparation was stretched to 140% of its resting length. The preparations consisted of longitudinal muscle and myenteric plexus, with most of the circular muscle removed. This is important because it indicates that there is a transduction site for mechanical stimulation located in the primary or secondary branches of the myenteric plexus. Hyperpolarising the nerve cell body during stretch affected the amplitude but not the frequency of action potentials, indicating that transduction did not occur in the cell body. This makes it highly likely that there are specific mechanically-sensitive sites located in the axons of these Dogiel type II cells. In a later study, it was shown that these neurones appear to be primarily sensitive to intramural tension, rather than length. Drugs which decreased smooth muscle activity also reduced stretch-induced firing in these primary afferents (Kunze et al., 1999) whereas Bay K8644, which opens L-type calcium channels, and hence activates smooth muscle contractility, increased firing. Circular muscle was largely absent from the preparations, but this did not alter their responses to circular stretch. This suggests that the neurones are probably responsive to any distortion of the ganglia in the region of their transduction sites. Further support for this comes from the observation that applying fine probes to the top of the ganglion activated stretch-sensitive Dogiel type II neurones even though force was applied vertically to the ganglion surface (Kunze et al., 2000). It is interesting to note that similar mechanisms seem to operate in the vagal mechanoreceptors to the upper gastrointestinal tract which also have transduction sites in myenteric ganglia (Zagorodnyuk and Brookes, 2000). Vagal mechanoreceptors have also been established to function as intramural tension receptors (Iggo, 1955). In a recent study, Kunze et al. (2000) succeeded in recording whole cell currents in Dogiel type II neurones in situ, using the methods developed by Gola and colleagues to enzymatically remove the surface of sympathetic ganglia (Gola et al., 1992). This confirmed that the transduction sites were located on the axons, since small depolarising generator potentials were evoked by mechanical stimulation of axons, whereas cell bodies responded with a hyperpolarisation, apparently mediated by stretch-sensitive BK channels (Kunze et al., 2000). ELECTROPHYSIOLOGICAL CHARACTERISTICS OF ENTERIC PRIMARY AFFERENT NEURONES There has been a lot of discussion over the relationship between electrophysiological and morphological properties of enteric neurones. It has become clear that many of the characteristics used to classify enteric nerve cells are actually shared by neurones belonging to several different functional classes. In addition, many of the characteristics show variability, both between cells, and within the same cell under different physiological conditions. For example, the long after-hyperpolarisation which follows action potentials in some neurones, was originally used to define AH cells (Hirst, 1974). In another influential paper, published a year earlier, enteric nerve cells were classified as type I and type II, on the basis of the number of action potentials evoked by depolarising current pulses (Nishi and North, 1973). Since these classifications largely overlap, the two schemes were quite reasonably combined and an electrophysiological type of AH/type II cells was identified (Wood, 1994). However, both the long after hyperpolarisation and the relative inexcitability

CELLULAR ORGANISATION OF THE MAMMALIAN ENTERIC NERVOUS SYSTEM

403

of these cells are substantially modified by slow excitatory synaptic inputs (Wood and Mayer, 1978). In addition, both are affected by the level of activation of calcium dependent potassium currents (Hirst, Johnson and van Helden, 1985a,b). Since these currents may be activated by a leak conductance of calcium ions, caused by the intracellular microelectrode, the criteria of excitability, and amplitude of the long after-hyperpolarisation cannot be used by itself to identify the functional class to which an impaled neurone belongs. This uncertainty should be borne in mind in the following sections where the electrophysiological characteristics of enteric primary afferent neurones, identified as AH cells, are reviewed. However, it should also be noted that enteric primary afferent neurones are amongst the largest nerve cell bodies in enteric ganglia, making them easy to impale with microelectrodes. It is likely therefore that the great majority of studies on AH cells have probably been on enteric primary afferent neurones.

Currents in AH cells AH cells, have been demonstrated to have a substantial calcium component to their action potentials, leading to a tetrodotoxin-resistance of spikes evoked by depolarising current pulses (North, 1973). This calcium influx gives rise to a pronounced inflection on the falling phase of the action potential, and is associated with a longer duration, measured at half peak amplitude, of soma action potentials. The inflection has been proposed to be a useful criterion for identifying AH cells, since it is largely resistant to the effects of membrane potential, synaptic inputs etc. (Schutte, Kroese and Akkermans, 1995). However, a similar inflection is seen on the action potentials of some somatostatin (SOM)-immunoreactive interneurones in the guinea-pig ileum (Song et al., 1997a), thus it is not unique to AH cells. The long after-hyperpolarisation following action potentials is largely due to the activation of a calcium-dependent potassium current (Hirst, Johnson and van Helden, 1985b) which has subsequently been shown to be due to opening of BK channels, which are sensitive to iberiotoxin and charybdotoxin (Kunze et al., 1994). A persistent calcium-dependent potassium conductance appears to contribute to the resting membrane potential of AH cells (North and Tokimasa, 1987), explaining the observation that calcium-free bathing solution causes depolarisation of AH cells (Wood et al., 1979; Grafe, Wood and Mayer, 1980). This is likely to be substantially due to BK channels too, since charybdotoxin depolarised AH cells (Kunze et al., 1994). A number of studies have characterised other ion channels in AH cells in the guinea-pig small intestine. It has been shown that in addition to inward calcium currents and outward potassium currents mediated via calcium-activated potassium channels, there are also inward sodium currents, which are likely to contribute to the depolarisation during action potentials. There is also a delayed rectifier (potassium current) and a transient outward current (Hirst, Johnson and Helden, 1985a) and a background potassium conductance (Galligan, North and Tokimasa, 1989). Another current present in AH cells is the inwardly rectifying cation current IH, which may contribute to stabilising the membrane potential (Galligan et al., 1990b) of these neurones during the after-hyperpolarisation. This current causes a characteristic “sag” in the response of AH cells to hyperpolarising current pulses, but is also seen in other classes of enteric neurones.

404

INNERVATION OF THE GASTROINTESTINAL TRACT

MORPHOLOGY OF ENTERIC PRIMARY AFFERENT NEURONES The intrinsic primary afferent neurones of the gut have AH cell electrophysiological characteristics and Dogiel type II soma-dendritic morphology. The details of their projections have been characterised using immunohistochemical labelling combined with lesions, intracellular dye filling, and retrograde tracing techniques. The first firm indication of the projections and morphology of these cells came with the recognition that a large proportion of them, at least in the guinea-pig ileum, contain immunoreactivity for the calcium binding protein, calbindin (Furness et al., 1988). The distinctive soma-dendritic morphology indicated that a substantial proportion of Dogiel type II cells contained this marker. This was confirmed when immunoreactivity for calbindin was detected in over 80% of electrophysiologically identified myenteric AH cells (Iyer et al., 1988). Importantly, calbindin appeared to be confined to this group of cell bodies, in the myenteric plexus, and was not present in S cells, which do not have Dogiel type II morphology. Large number of calbindin-immunoreactive varicosities were present in myenteric and submucous ganglia and in the mucosa, but not in the circular or longitudinal muscle layers. This is consistent with a role for these neurones as primary afferent neurones or as interneurones, but not as motor neurones. It was suggested at the time of these studies that the presence of calbindin might be related to the calcium-dependence of action potentials and the long after-hyperpolarisation in these neurones. However, many Dogiel type II neurones with apparently identical electrophysiological characteristics lack detectable calbindin immunoreactivity (Iyer et al., 1988; Brookes et al., 1995). In addition, many calbindin-immunoreactive neurones in the guinea-pig proximal colon have Dogiel type I morphology and have S cell electrophysiological characteristics (Messenger, Bornstein and Furness, 1994), thus the functional significance of this calcium binding protein remains to be determined. Studies of the effects of lesions on calbindin-immunoreactive nerve fibres in the ileum suggested that Dogiel type II neurones projected for short distances circumferentially as well as to the submucous ganglia and mucosa (Furness et al., 1990b). Early studies on the soma-dendritic morphology of enteric neurones used dyes or tracers such as Procion Yellow (Hodgkiss and Lees, 1983) or Lucifer Yellow (Bornstein et al., 1984a), which appear to have limited ability to diffuse through the cytoplasm of nerve cells. The use of more rapidly diffusing tracers such as biocytin (Horikawa and Armstrong, 1988) allowed for considerably better filling of enteric nerve cells, revealing their finer processes. It became apparent that Dogiel type II neurones formed two major morphological classes. Most of the cells had several long processes which ran circumferentially, giving rise to many varicose endings in the neuropile of their ganglion of origin and other ganglia further circumferentially (Bornstein et al., 1991b). In fact, retrograde labelling studies in our laboratory have shown that when tracers are applied selectively to myenteric ganglia (after careful removal of all overlying circular muscle) Dogiel type II neurones make up all of the circumferentially projecting neurones more than 3 mm from the application site. About 10% of the cells had numerous short dendritic processes and were classified as “dendritic Dogiel type II” neurones according to the morphological classification scheme of Stach (Stach, 1989). These neurones also projected circumferentially, but appeared to give rise to relatively few endings within their ganglion of origin (Bornstein et al., 1991b).

CELLULAR ORGANISATION OF THE MAMMALIAN ENTERIC NERVOUS SYSTEM

405

No polarised projections up or down the gut were apparent from this study. Using retrograde tracing techniques, it was later demonstrated that the “dendritic Dogiel type II neurones” had long aboral projections within the myenteric plexus, up to 100 mm in length (Brookes et al., 1995). In fact, this class of cells actually comprised the major single source of aborally directed axons in the guinea-pig small intestine (Brookes et al., 1995; Song, Brookes and Costa, 1996). During all of the intracellular dye filling experiments it was reported that the axons of Dogiel type II neurones often ended abruptly as a retraction bulb, in the circular muscle. This was due to the removal of the bulk of the circular muscle prior to impalement of myenteric nerve cell bodies. This observation suggested that many Dogiel type II cells projected to overlying layers of the gut wall. Using retrograde tracing it was later shown that all calbindin immunoreactive myenteric neurones in the guinea-pig small intestine, and in fact probably all Dogiel type II neurones, project to the overlying mucosal layers (Song, Brookes and Costa, 1994a). The projections of enteric primary afferent neurones in the submucous plexus have also been largely established. Early studies on their immunoreactivity gave differing estimates of the proportion of submucous neurones which contained calbindin immunoreactivity (Furness et al., 1990b; Kirchgessner, Tamir and Gershon, 1992). It has recently become clear that some antisera to calbindin reveal many more neurones than other antisera (Reiche et al., 1999) and this may explain some of the discrepancies between studies. Nevertheless, a population of tachykinin-immunoreactive neurones in the submucous ganglia, with and without calbindin immunoreactivity, were identified in a number of studies (Bornstein and Furness, 1988). Intracellular recording and dye filling revealed that, like their myenteric counterparts, these cells have AH-cell-like properties with multipolar morphology (Bornstein et al., 1989; Evans, Jiang and Surprenant, 1994). They make extensive projections within the submucous ganglia, often appearing to contact other nerve cell bodies (Bornstein et al., 1989; Evans, Jiang and Surprenant, 1994). They can be retrogradely labelled by tracers applied to both the mucosa and to the myenteric plexus (Kirchgessner, Tamir and Gershon, 1992; Song et al., 1992) and thus project to both targets.

CHEMICAL CODING OF ENTERIC PRIMARY AFFERENT NEURONES The submucous primary afferent neurones were reported to be immunoreactive for the calcium binding protein calbindin and to contain immunoreactivity for tachykinins (Kirchgessner, Tamir and Gershon, 1992). In addition to these neurochemicals, this class of submucous neurones is known to be immunoreactive for ChAT (Furness, Costa and Keast, 1984) and was later shown to contain neuromedin U (NMU) (Furness et al., 1989a). The myenteric primary afferent neurones have very similar chemical coding. The majority of them contain immunoreactivity for calbindin (Iyer et al., 1988) and for tachykinins (Song, Brookes and Costa, 1991) and for ChAT (Steele, Brookes and Costa, 1991) and many also contain NMU (Furness et al., 1989a). It is very clear from this that primary afferent neurones in the submucous plexus and in the myenteric plexus are strikingly similar to one another, showing nearly identical patterns of projection, electrophysiological properties, chemical coding and morphology.

406

INNERVATION OF THE GASTROINTESTINAL TRACT

SYNAPTIC INPUTS TO ENTERIC PRIMARY AFFERENT NEURONES In early studies, myenteric neurones were classified as AH cells because of the presence of the long after-hyperpolarisation. In contrast, other cells which usually lacked the after-hyperpolarisation, had prominent fast excitatory synaptic inputs following focal electrical stimulation of the preparation. This led to their being named “S cells” (Hirst, 1974). Subsequently, it was suggested that AH cells do in fact receive fast synaptic inputs, but that these are of small amplitude and require signal averageing to be visualised clearly (Grafe et al., 1979). This is consistent with recent reports that calbindin-immunoreactive myenteric neurones (which include many of the enteric primary afferent neurones of the guinea-pig ileum), maintained in culture, generate smaller currents to nicotine than do non-calbindin neurones (Zhou and Galligan, 2000). Nevertheless, there have been reports of some AH cells receiving prominent fast excitatory synaptic inputs, in the small intestine (Bornstein, Furness and Kunze, 1994; Wood, 1994a), large intestine (Wade and Wood, 1988) and rectum (Tamura and Wood, 1989). It should also be remembered that in many of these studies, recorded cells were not routinely dye filled, so it cannot be assumed that they had Dogiel type II morphology. In the small intestine, very few Dogiel type II neurones receive fast synaptic inputs (Brookes et al., 1995; Kunze et al., 1999). It seems likely that much of the confusion over this matter may result from the fact that AH cells and Dogiel type II cells are not identical. It is likely that some AH cells, classified on the basis of the long afterhyperpolarisation or inflection on their action potentials, do not have Dogiel type II morphology. Some of these AH/non-Dogiel type II cells almost certainly receive powerful fast synaptic inputs. Prominent among these are the SOM-immunoreactive descending interneurones which have been electrophysiologically characterised (Song et al., 1997a). Many of these cells had a long after-hyperpolarisation and they often had an inflection on their action potentials, which had a longer duration than most S cells. These interneurones also appear to have the IH current which contributes greatly to the “look and feel” of AH cells. Intracellular dye fills of these neurones reveal a smooth ovoid cell body with numerous long processes, similar to AH cells. However, when biocytin or biotinamide are used as labels, it becomes clear that these neurones have a single long axon, which consistently runs aborally (Song et al., 1997a; Meedeniya et al., 1998), and numerous shorter filamentous dendrites (Portbury et al., 1995b; Song et al., 1997a; Meedeniya et al., 1998). It seems very likely that these neurones may have been classified as AH cells by some investigators but not by others. This matter is important because the presence of fast synaptic inputs would profoundly alter the understanding of how enteric primary afferent neurones are likely to function in enteric neuronal circuitry. On the current balance of evidence, it seems reasonable to assume, in the guinea-pig ileum, that Dogiel type II neurones function as enteric primary afferent neurones, transducing both mucosal and/or mechanical stimuli, without being synaptically driven via fast synaptic inputs from other classes of neurones. Other types of AH cells probably play very different roles and it is likely that most of these lack Dogiel type II morphology. However, it is not possible to exclude the possibility that a small subset of AH cells, with Dogiel type II morphology do receive fast synaptic inputs.

CELLULAR ORGANISATION OF THE MAMMALIAN ENTERIC NERVOUS SYSTEM

407

Slow synaptic inputs to enteric primary afferent neurones Although AH cells show few fast synaptic inputs, it became apparent that following trains of electrical stimulation, many AH cells showed long-lasting depolarisation which was due to slow synaptic excitation (Wood and Mayer, 1978). These slow excitatory postsynaptic potentials (slow EPSPs) were mimicked by 5-HT and it was proposed that this might be the neurotransmitter that is responsible for them (Wood and Mayer, 1978). This was supported, a few years later, by the observation that large number of enteric nerve fibres in both myenteric and submucous ganglia, contained immunoreactivity for 5-HT (Costa et al., 1982; Furness and Costa, 1982). Slow EPSPs were associated with a depolarisation, which lasted from 30 s up to several minutes. Typically there was a decrease in input impedance of the cells at the peak of the potential, suggesting that there was a net closure of ion channels. Under current clamp conditions it was clear that slow EPSPs had a reversal potential near –90 mV, the equilibrium potential for potassium ions. During the slow EPSP, the long after-hyperpolarisation which follows action potentials in AH cells was suppressed and there was a profound increase in excitability, with cells firing repeatedly during depolarising current pulses and often showing spontaneous firing. All of these effects were mimicked by 5-HT and, furthermore, desensitisation with 5-HT reduced electrically stimulated slow EPSPs (Wood and Mayer, 1979). A number of studies were then carried out to determine whether or not 5-HT was the only transmitter that might mediate slow EPSPs. An early study demonstrated that substance P (SP) also mimicked the slow EPSP in AH cells (Katayama and North, 1978) and that desensitisation with SP, or chymotrypsin treatment, reduced or blocked the synaptic potential (Morita, North and Katayama, 1980). This led to some debate as to whether SP or 5-HT was “the transmitter”. An elegant study used lesions to the myenteric plexus to distinguish between these possibilities. It was known that 5-HT immunoreactive neurones had descending projections within the myenteric plexus (Furness and Costa, 1982) whereas SP immunoreactive neurones gave rise to many varicosities close to their cell bodies (Costa et al., 1980a). Lesions were used to remove long descending inputs from 5-HT containing neurones and intracellular recording were made in these partially denervated islands. It was found that both fast and slow EPSPs persisted in these areas, indicating that local pathways were sufficient, although there was a reduction in amplitude of synaptic events, which was attributed to the loss of long descending pathways (Bornstein et al., 1984b). Later studies have shown that much of the SP (or tachykininlike) immunoreactivity in the myenteric plexus is likely to arise locally from Dogiel type II neurones (Song, Brookes and Costa, 1991) which have profuse circumferential projections (Hendriks, Bornstein and Furness, 1990; Bornstein et al., 1991b; Song, Brookes and Costa, 1991, 1996; Brookes et al., 1995). This evidence does not rule out a role for 5-HT in slow synaptic inputs to AH cells, but suggests that SP, or a related tachykinin is likely to be involved, under some circumstances. An early ultrastructural study suggested that 5-HT immunoreactive nerve varicosities are associated with the cell bodies of dye filled AH cells (Erde, Sherman and Gershon, 1985) although a subsequent study failed to find any special association (Young and Furness, 1995). The story is also complicated by the fact that a number of other endogenous substances have been demonstrated to mimic slow EPSPs on AH cells. For example pituitary adenyl cyclase activating peptide (Christofi and

408

INNERVATION OF THE GASTROINTESTINAL TRACT

Wood, 1993a), VIP, bombesin and gastrin-releasing peptide (GRP) (Zafirov et al., 1985a), cholecystokinin (CCK) and the gastrin analogue, pentagastrin (Nemeth, Zafirov and Wood, 1985) and CGRP (Palmer et al., 1986) all mimic slow EPSPs. Using selective antagonists, evidence has been provided that endogenous CCK may contribute to slow EPSPs via both CCKA and CCKB receptors (Schutte et al., 1997). In addition, histamine also mimics slow EPSPs, although it is unlikely to be an endogenous mediator of these synaptic events (Nemeth, Ort and Wood, 1984). The intracellular mechanisms underlying slow EPSPs are of considerable interest, given the long time course of these events. All of the candidate receptors belong to the superfamily of receptors with 7 transmembrane regions and are likely to be coupled via G-proteins to second messenger pathways within cells. The observation that slow EPSPs were mimicked by intracellular injection of cyclic AMP (Zafirov et al., 1985b) and by membrane permeant cAMP analogues (Palmer, Wood and Zafirov, 1986) suggested a role for adenylcyclase in generating slow EPSPs. This effect was also mimicked by forskolin which raises intracellular cAMP levels (Nemeth et al., 1986). Adenosine, hyperpolarises AH cells, increase the amplitude of long after-hyperpolarisations, reduces their input impedance and reduces excitability (Palmer, Wood and Zafirov, 1987). It has been proposed that this is due to reduction in intracellular cAMP concentrations. Direct measurement suggests that this is one of the effects of adenosine (Xia, Fertel and Wood, 1997). Adenosine also blocked the effects of forskolin (Palmer, Wood and Zafirov, 1987). Using adenosine as a tool, it was demonstrated that the effects of several of the potential transmitters of the slow EPSP were also blocked by adenosine, however the effects of exogenous SP, CGRP and 5-HT were not reduced. Unfortunately, it was not possible to use adenosine as a tool to investigate the nature of second messenger pathways activated during the slow EPSP because adenosine also causes potent presynaptic inhibition of both fast and slow synapses onto enteric neurones (Christofi and Wood, 1993b). Early studies had suggested that during the slow EPSP, a calcium-dependent potassium current was inhibited (Grafe, Wood and Mayer, 1980). Later it was shown that SP reduced such a current which contributed to both resting membrane potential and the long afterhyperpolarisation (Morita and Katayama, 1992). Such a reduction in a resting potassium conductance would explain the observation that there was a marked increase in input impedance in most cells during the slow EPSP. However, in many recordings it had been noticed, particularly during the early phase of the slow EPSP that there was sometimes an initial decrease in input impedance. The currents underlying this event were systematically studied using single electrode voltage clamp (Bertrand and Galligan, 1994). It was shown that slow EPSPs are frequently associated with an increase in chloride currents which sum with the decrease in potassium conductance. While forskolin and cAMP analogues mimic the decrease in potassium conductances, the second messenger pathways associated with the increase in chloride conductance have not been identified (Bertrand and Galligan, 1995). Sources and significance of slow EPSPs Experimentally, slow EPSPs are typically evoked by repetitive focal electrical stimulation relatively close to the recorded nerve cell body. Sometimes the stimulating electrode is

CELLULAR ORGANISATION OF THE MAMMALIAN ENTERIC NERVOUS SYSTEM

409

placed on an internodal strand or on a nearby part of the ganglion. From our experience with retrograde labelling techniques, we estimate that such stimuli probably activate 200–2000 axons synchronously, which makes them rather “un-physiological”. It is important therefore to determine under what conditions slow EPSPs are likely to occur, in vivo, which cells are likely to give rise to them and what their role is likely to be. Several studies have examined the presence of slow EPSPs without using electrical stimuli. Using paired intracellular recording, it was shown that single AH cells, with Dogiel type II morphology, could give rise to measurable slow EPSPs in other myenteric neurones (Kunze et al., 1993) including AH cells. This was significant because it demonstrated that it was not necessary to stimulate large number of axons simultaneously. However, repetitive stimulation was required, indicating that multiple action potentials were required to evoke measurable potentials. When chemical stimuli were applied to the mucosa, some AH cells were activated directly by the stimulus, whereas others were activated indirectly, via slow excitatory synaptic inputs. In contrast, all S cells were activated indirectly, via synaptic inputs (Bertrand et al., 1997). This suggests that firing in chemosensitive primary afferent neurones, presumably myenteric AH cells, can evoke slow EPSPs in other primary afferent neurones. The same mechanism appears to operate when primary afferent neurones are activated by stretch (Furness et al., 1998). Thus it appears that primary afferent neurones in the guinea-pig small intestine are organised into self-exciting assemblies, which are then capable of giving synchronised responses to stimuli. This idea was proposed by Jackie Wood, long before type II/AH neurones had been identified as enteric primary afferents. He suggested that the slow EPSP may modulate the spread of excitability across the soma of Dogiel type II neurones (Wood, 1989) (see also Chapter 9 of this volume). ENTERIC PRIMARY AFFERENT NEURONES IN OTHER REGIONS AND SPECIES It is clear that a great deal of data has been gathered about the projections, chemical coding, electrophysiological characteristics and morphology of the intrinsic primary afferent neurones of the guinea-pig small intestine. The great bulk of this data was obtained before the function of these cells had been identified. This particular type of enteric neurone has been the focus of a great deal of study, for several rather arbitrary reasons. First of all, they are large and easy to impale with intracellular microelectrodes and are frequently encountered, because they make up a substantial proportion of all enteric neurones. Their electrophysiological characteristics are intriguingly unusual and they have a distinctive morphology that allows them to be unequivocally distinguished from other classes of cells in this preparation. This raises the question then, as to whether primary afferent neurones are similar in other regions of the gut and in other species. At present, this question cannot be answered directly, since it is only in the guinea-pig ileum that functional studies of sensory transduction by enteric neurones have been carried out. However, it is possible to determine whether or not there are electrophysiological AH cells, and cells with Dogiel type II morphology present in other preparations. It should be stated, at the outset, that the only absolute requirement for an enteric neurone to function as a primary afferent is the ability to produce a generator potential in response to a mechanical or chemical stimulus. This must reflect the presence of appropriate stretch-activated channels or ligand-gated mechanisms in the processes of these cells. Since none of the

410

INNERVATION OF THE GASTROINTESTINAL TRACT

molecules responsible for transduction have yet been identified in enteric primary afferent neurones, they cannot be used to identify them. It should be pointed out that all of the other electrical characteristics found in AH/type II cells, such as their action potential shape, after-hyperpolarisation, channels and currents, probably function only to modulate the responses of these neurones to physiological stimuli, or to influence transmitter release from their terminals. There is no reason, per se, to believe that any of these characteristics is absolutely essential for a neurone to function as a primary afferent. It seems intuitively possible that a number of different combinations of electrical characteristics could meet the integrative requirements of enteric primary afferent neurones. Thus the presence or absence of after-hyperpolarisations, spike inflections etc. may not be a good indicator of whether a particular neurone functions as a primary afferent in any gut preparation other than the guinea-pig ileum. For example, the absence of a calcium component to the soma action potential, or the lack of an after-hyperpolarisation could not be taken to rule out the possibility that a particular neurone functions as a primary afferent in the mouse, rat, pig or human gut, for example. Perhaps the best correlation with a role as a primary afferent is the soma-dendritic morphology of the neurone. Throughout the body, primary afferent neurones are generally bipolar, multipolar or pseudo-unipolar – this applies to those with cell bodies in dorsal root ganglia, nodose or petrosal ganglia, trigeminal and other cranial ganglia. It also applies to myenteric and submucous Dogiel type II neurones in the guinea-pig small intestine, which have been shown to function as primary afferent neurones. The multipolar morphology may be functionally important since it ensures spatial separation between transduction sites and synaptic release sites. It would be reasonable to expect that enteric primary afferent neurones in other species and regions of gut would also be multipolar. In his original work, Dogiel (1899) described multipolar (type II cells) in several species and regions of gut, including small and large intestine of guinea-pigs, in the gall bladder and in human intestine. In the guinea-pig, Dogiel type II cells have been described in the ileum, duodenum (Clerc et al., 1998), and proximal colon (Messenger, Bornstein and Furness, 1994; Neunlist and Schemann, 1997; Neunlist, Dobreva and Schemann, 1999), distal colon (Lomax et al., 1999) and rectum (Tamura, 1992; Tamura, 1997). In each of these regions some correlation has been noted between Dogiel type II morphology and AH cell electrophysiological characteristics. However, Dogiel type II neurones do not appear to be present in the proximal guinea-pig stomach, based on retrograde labelling of large number of neurones (Brookes et al., 1998). Neither is the AH cell electrophysiological type present in this area (Schemann and Wood, 1989). Dogiel type II cells also appear to be absent from the oesophageal myenteric plexus (Brookes et al., 1996) although intracellular recordings have not been made from these neurones. Interestingly, neurones with AH cell electrophysiological characteristics have been recorded from the distal stomach of the guinea-pig (Tack and Wood, 1992) but were not morphologically identified. Immunohistochemical and retrograde labelling studies have shown a small population of neurones in the myenteric plexus of the guinea-pig antrum which have large smooth cell bodies and several long processes. This suggests that a small population of Dogiel type II neurones may be present in this region (SJH Brookes and GW Hennig, unpublished observations). Thus it appears that multipolar neurones, with AH cell electrophysiological characteristics (a calcium-dependent action potential, with an inflection on

CELLULAR ORGANISATION OF THE MAMMALIAN ENTERIC NERVOUS SYSTEM

411

the falling phase, usually followed by a long after-hyperpolarisation) are present throughout the enteric nervous system of the guinea-pig, at least distal to the mid stomach. It should be noted, however that the long after-hyperpolarisation, in particular, shows considerable variability in amplitude, being powerfully suppressed when slow excitatory synaptic inputs have been activated (Wood, 1994b). Dogiel type II neurones, with AH cell electrophysiological characteristics have also been described in the myenteric plexus of other species, including the rat small intestine (Brookes, Ewart and Wingate, 1988), rat large intestine (Browning and Lees, 1996) and in the pig small intestine (Cornelissen et al., 2000). It should be noted however that the correlation in the pig small intestine was not as straightforward as reported elsewhere. In a systematic study of the long after hyperpolarisation, in porcine myenteric cells with Dogiel type II morphology, it was reported that only a small proportion (17%) showed a slow after hyperpolarisation comparable to that seen in the guinea-pig (Cornelissen et al., 2000). Cells classified as AH cells have been recorded with intracellular microelectrodes in the mouse colon (Furukawa, Taylor and Bywater, 1986) and occasionally in the human colon (Brookes, Ewart and Wingate, 1987), although their morphology was not determined, so it is not clear whether or not they were multipolar neurones. Thus it is clear that neurones with Dogiel type II morphology and neurones with AH cell electrophysiology are found in the enteric nervous system of most of the preparations of small or large intestine in a range of species. Where tested, there has usually been some correlation between the characteristics, although they do not always match one-for-one. The functional significance of this variability remains to be determined. It may reflect minor differences in the way that stimuli are frequency coded by afferents or it may indicate that some Dogiel type II cells play roles other than as primary afferents. Concerning the chemical coding of Dogiel type II neurones, differences between the guinea-pig and other species have been reported. In the guinea-pig, Dogiel type II enteric primary afferent neurones are immunoreactive for ChAT (Steele, Brookes and Costa, 1991; Costa et al., 1996) with or without immunoreactivity for the calcium binding protein calbindin (Iyer et al., 1988; Brookes et al., 1995). Similar coding is seen in Dogiel type II/AH neurones in the proximal colon of the guinea-pig (Neunlist and Schemann, 1997; Neunlist, Dobreva and Schemann, 1999). In the small intestine, the majority of these primary afferent neurones are also immunoreactive for tachykinins (Song, Brookes and Costa, 1991) and for NMU (Furness et al., 1989a). In other species, Dogiel type II neurones have been reported to contain immunoreactivity for CGRP, notably in the small intestine of the pig (Scheuermann et al., 1991) and the human (Timmermans et al., 1992). While care has to be taken to identify soma-dendritic morphology purely on the basis of immunoreactivity, a recent study has confirmed that porcine Dogiel type II neurones, retrogradely labelled from the mucosa, frequently contain immunoreactivity for CGRP, in addition to ChAT and tachykinins (Hens et al., 2000). Myenteric Dogiel type II cells in the rat ileum have also been shown to contain ChAT (Mann, Furness and Southwell, 1999), at least those that were immunoreactive for calbindin. These differences in chemical coding may have profound importance for interpreting functional studies. CGRP immunoreactive nerve cell bodies are found throughout the rat enteric ganglia (Su et al., 1987; Sternini and Anderson, 1992). If some of these are the Dogiel type II cells, this could explain the observations of Grider and colleagues which suggest that enteric

412

INNERVATION OF THE GASTROINTESTINAL TRACT

primary afferent neurones may release CGRP to activate reflex circuits in the rat colon (Grider, 1994). ENTERIC PRIMARY AFFERENT NEURONES: SUMMARY From functional studies it is clear that enteric neuronal circuits can be specifically activated by physiological stimuli such as stretch, distortion of the villi and chemical stimulation, indicating the existence of enteric primary afferent neurones. To date, neurones which respond to these stimuli have been directly demonstrated only in the myenteric and submucous ganglia of the guinea-pig small intestine. These neurones all appear to have a large smooth cell soma, sometimes with short filamentous dendrites, but always characteristically have several long axonal processes (and hence have “Dogiel type II” morphological features). In the guinea-pig ileum these neurones also have characteristic electrophysiological characteristics, with a significant inward calcium current during their action potentials which are followed by a long after-hyperpolarisation. These cells receive powerful slow excitatory synaptic inputs, substantially arising from other primary afferent neurones, but receive small and sparse fast synaptic inputs. Intuitively, this is compatible with the neurones being activated by physiological stimuli (rather than being synaptically driven by powerful fast synaptic inputs) but with their excitability being modulated by slow synaptic contacts from other neurones of the same class, which are thus organised into coordinated assemblies. Exactly how many classes of intrinsic primary afferent neurones exist in the guinea-pig ileum is not currently clear. At least three classes can be distinguished on morphological grounds. These include the cells located in submucous ganglia, myenteric “dendritic” Dogiel type II neurones, with long descending projections (Brookes et al., 1995) and “non-dendritic” Dogiel type II neurones. Three physiological types have also been identified, including the submucous neurones responding to mechanical stimulation of the mucosa, myenteric neurones responding to chemical stimuli and myenteric neurones that are stretch sensitive. Whether or not any of the cells respond to two or more modes of stimulation remains to be determined, but it seems likely given estimates of the proportions of cells that are activated under different circumstances (Furness et al., 1998; Kunze and Furness, 1999). The calcium influx during the action potential, and the related long after-hyperpolarisation may not be directly involved in sensory transduction. It seems likely that these characteristics may have little to do with the ability of these neurones to detect physiological stimuli, but may rather be involved in modulating the output from enteric primary afferent neurones onto other neurones in the enteric ganglia. As such, the long after-hyperpolarisation and calcium spike may well be epi-phenomena, associated with a sensory role, but not directly, causally involved in it. If this is the case, primary afferent neurones in other regions of gut and in other species may show different combinations of such electrophysiological characteristics. It would be premature to expect all enteric primary afferent neurones to appear morphologically, neurochemically and electrophysiologically identical to those of the guinea-pig small intestine. Lastly, it is worth pointing out that although recent evidence has demonstrated convincingly that many Dogiel type II cells function as primary afferents, this does not exclude other roles for them. Slow synaptic input to Dogiel type II neurones may, under some

CELLULAR ORGANISATION OF THE MAMMALIAN ENTERIC NERVOUS SYSTEM

413

circumstances, drive them to produce action potentials in the absence of physical or chemical stimuli (Wood, 1994b). This suggests that Dogiel type II neurones may, under some circumstances, function as interneurones in the enteric nervous system. All of the classes of Dogiel type II neurones discussed above have axonal projections to the mucosa (Song, Brookes and Costa, 1994a). It is possible that the axons within the mucosa may not be exclusively afferent in function. It is possible that they might influence epithelial cells in the mucosa, since they give rise to abundant calbindin-immunoreactive endings in close apposition to the mucosal epithelium. In addition, these cells are likely to be cholinergic (from the presence of ChAT immunoreactivity) and cholinergic mechanisms play such an important role in regulating epithelial secretion (Cooke and Reddix, 1994) (see also Chapter 7 of this volume).

ENTERIC MOTOR NEURONES Since the earliest studies on isolated or denervated preparations of intestine (Bayliss and Starling, 1899; Langley and Magnus, 1905; Cannon, 1912; Trendelenburg, 1917) it has been clear that there are intrinsic motor neurones within the enteric nervous system. It is technically very straightforward to record mechanical activity of the smooth muscle layers of preparations of gastrointestinal tract. Since the first studies using electrical field stimulation (Paton, 1955), the guinea-pig ileum, in particular, has become a standard pharmacological preparation, in which to test an enormous range of pharmacological agents on neuromuscular transmission and smooth muscle contractility (see review by Marcello Tonini and colleagues in this volume). Despite the widespread use of this preparation, the motor neurones responsible for this transmission have only been identified in the last decade. For many years it was unclear which major morphological type enteric motor neurones belonged to, with Dogiel suggesting that type I cells were probably motor (Dogiel, 1899), whereas some later authors considered that type II cells were likely to function as motor neurones (Hill, 1927). The first real insights came with the use of immunohistochemical methods to analyse the axons of enteric motor neurones within the muscle layers. In particular, extensive branching by VIP-immunoreactive nerve axons was seen in the circular muscle of the guinea-pig ileum (Costa et al., 1980b; Costa and Furness, 1983). A later study demonstrated that the nerve cell bodies that gave rise to VIP-containing axons were largely of the Dogiel type I morphological class with S cell electrophysiological characteristics (Katayama, Lees and Pearson, 1986). A similar finding was made for enkephalin (ENK)-immunoreactive axons (Furness, Costa and Miller, 1983) and cell bodies (Bornstein et al., 1984a). Since either ENK or VIP is present in the majority of motor axons in the circular muscle layer of the guinea-pig ileum (Llewellyn Smith et al., 1988), it could be concluded that a substantial proportion of motor neurones must have Dogiel type I morphology. This was later supported by observations that immunoreactivity for the calcium binding protein, calbindin, was present in a substantial proportion of Dogiel type II neurones (Iyer et al., 1988) but was not present in many varicose motor axons within the circular muscle layer (Furness et al., 1988, 1990b).

414

INNERVATION OF THE GASTROINTESTINAL TRACT

LONGITUDINAL MUSCLE MOTOR NEURONES The longitudinal muscle layer of the guinea-pig ileum is a thin layer of tissue lying just below the serosa, from 15–40 µm thick, depending on the degree of stretch of the tissue. It is innervated by the ramifying network of varicose nerve fibres making up the tertiary plexus, which lies on its inner face, but which does not penetrate deeply in the muscle layer (Llewellyn-Smith et al., 1993; Gabella, 1994). Retrograde labelling, with the carbocyanine dye 1,1′-didodecyl-3,3,3′,3′-tetramethyl-indocarbocyanine perchlorate (DiI) applied in solid form to the tertiary plexus, labelled longitudinal muscle motor neurones in the myenteric plexus (Brookes et al., 1992). All of the nerve cell bodies were located within 3 mm of the DiI application site, indicating that the motor neurones have relatively short projections to the region of muscle that they innervate. The cell bodies were typically very small (12–15 µm in the short axis) with a single axon. Some had clear Dogiel type I morphology, with short lamellar dendrites, but in others, the dendrites were so restricted that they were better classified as “simple cells" (Furness, Bornstein and Trussell, 1988). Longitudinal muscle motor neurones were often located in small clusters, located near the points at which internodal strands entered the myenteric ganglia. Intracellular dye fills of these neurones have revealed that they branch extensively within the tertiary plexus, close to their cell bodies (Bornstein et al., 1991a; Furness et al., 2000). Intracellular recordings have shown that they are highly excitable S cells which fire tonically to depolarising current pulses (Smith, Burke and Shuttleworth, 1999). Analysis of the chemical coding of retrogradely labelled longitudinal muscle motor neurones revealed that the great majority in the guinea-pig ileum (>97%) were immunoreactive for ChAT and hence likely to be cholinergic (Brookes et al., 1992). These motor neurones are likely to release the acetylcholine that mediates the excitatory junction potential evoked by electrical stimulation (Cousins et al., 1993) and which underlies the widely-studied cholinergic twitch of the guinea-pig ileum (Paton, 1955). It is well established that inhibitory junction potentials are relatively small in the longitudinal muscle of the guinea-pig ileum (Bywater and Taylor, 1986), at least in comparison to those recorded in the circular muscle layer. This suggests that the longitudinal muscle layer probably receives relatively little direct inhibitory innervation. Consistent with this is the relatively sparseness of VIP-immunoreactive (Costa et al., 1980b; Costa and Furness, 1983) and nitric oxide synthase (NOS)-immunoreactive (Costa et al., 1992) varicose nerve fibres in the tertiary plexus of this region. This was matched by the observation that only 3% of retrogradely labelled motor neurones to the longitudinal muscle were immunoreactive for VIP (Brookes et al., 1992). However it is still possible to record relaxations of the longitudinal muscle in responses to electrical stimulation (Osthaus and Galligan, 1992; Yunker and Galligan, 1998). This suggests either that the 3% of motorneurones can still be functionally important, or that overflow of inhibitory transmitters from the circular muscle can occur following repetitive, synchronous activation of large number of axons. Of the retrogradely labelled cell bodies of motor neurones to the longitudinal muscle of the guinea-pig ileum, approximately 87% were immunoreactive for the calcium binding protein, calretinin (Brookes et al., 1992). Calretinin immunoreactivity is present in many varicose nerve fibres in the tertiary plexus (Brookes, Steele and Costa, 1991a), where it coexists with ChAT, but not with neurofilament protein (NFP) triplet. Calretinin is found

CELLULAR ORGANISATION OF THE MAMMALIAN ENTERIC NERVOUS SYSTEM

415

in two classes of neurones within the myenteric plexus of the guinea-pig ileum, which together comprise 26% of all cells. Approximately 21% of all myenteric neurones are calretinin-immunoreactive longitudinal muscle motor neurones, indicating that about 24% of all myenteric neurones innervate the longitudinal muscle in this preparation. About half of all longitudinal muscle motor neurones were immunoreactive for tachykinins (Brookes et al., 1992), all of which were also immunoreactive for ChAT. It is well established that the longitudinal muscle of the guinea-pig ileum receives a non-cholinergic excitatory innervation (Ambache and Freeman, 1968) which is substantially mediated by SP (Franco, Costa and Furness, 1979; Bjorkroth, 1983). The cholinergic twitch can be effectively stimulated by electrical stimuli delivered at low frequencies (typically 0.1 Hz) whereas non-cholinergic excitation, recorded in the presence of muscarinic antagonists, requires trains of stimuli. This led to speculation that tachykinins, or other mediators of non-cholinergic excitation, may be released from different classes of nerves from those releasing acetylcholine. It appears from the anatomical data that tachykinins are probably released by a subset of cholinergic nerve fibres and that the apparent frequency-dependence may reflect differential release of transmitters stored in different vesicles in the same nerve endings in response to repetitive action potentials (Maggi, Holzer and Giuliani, 1994). Many other neurochemicals found in enteric neurones are strikingly absent from the tertiary plexus. There are very few axons immunoreactive for SOM (Costa et al., 1980c), 5-HT (Costa et al., 1982), calbindin (Furness et al., 1988; Furness et al., 1990b) which each mark other substantial populations of enteric neurones. However, immunoreactivity for γ-amino butyric acid (GABA) is present, particularly after pre-loading the preparation with exogenous GABA (Jessen, Hills and Saffrey, 1986; Furness et al., 1989b). It appears that GABA is present within a subset of motor neurones to both the circular and longitudinal muscle layers, largely (but not exclusively) in those immunoreactive for NOS (Williamson, Pompolo and Furness, 1996). Longitudinal muscle innervation: other regions The relative paucity of direct, inhibitory innervation of the longitudinal muscle of the guinea-pig ileum is not the case for most other regions of gut. The longitudinal muscle layer of the stomach, proximal colon and distal colon of the guinea-pig all show more pronounced nerve mediated relaxations (Costa, Furness and Humphreys, 1986) in response to electrical stimulation. Indeed, the taenia coli (or more properly taenia caeci) was one of the first preparations at which non-cholinergic, non-adrenergic inhibitory innervation of the gastrointestinal tract was described (Burnstock et al., 1963; Burnstock, Campbell and Rand, 1966). It is tempting to speculate why the ileum may be rather differently organised. It is clear from functional studies that during peristaltic emptying, there is a marked lengthening of the longitudinal muscle, while the circular muscle is contracting to expel the fluid contents (Trendelenburg, 1917). This led to the suggestion that the two muscle layers may be reciprocally innervated, such that one muscle layer contracts while the other relaxes (Kottegoda, 1970). A good deal of evidence now suggests that this is not the case. In particular, in the guinea-pig ileum, the longitudinal muscle remains contracted during circular muscle contraction, if the preparation is first opened up into a flat sheet, preventing passive interactions due to the incompressibility of fluid contents (Brookes

416

INNERVATION OF THE GASTROINTESTINAL TRACT

et al., 1999). If contractions of circular and longitudinal muscle are carefully monitored during peristaltic emptying in tubular preparations, the longitudinal muscle starts to lengthen only after circular muscle contractions have occluded the lumen of the intestine (Hennig et al., 1999). When contractions of the two layers are monitored independently, using force transducers attached to regions where the layers have been separated, the two are seen to contract simultaneously (Spencer, Walsh and Smith, 1999). Thus it appears that during peristaltic emptying, contraction of the circular muscle is sufficiently powerful to overcome the active contraction of the longitudinal muscle. It is possible, then, that active relaxation of the longitudinal muscle may not be functionally important in this region. Under similar conditions, in the colon, there is evidence for a direct, functional inhibitory innervation of the longitudinal muscle causing relaxation in response to physiological stimuli (Smith and Robertson, 1998). Longitudinal muscle motor neurones: other regions, other species As described above, the longitudinal muscle of the guinea-pig ileum is innervated mostly by excitatory motor neurones with short, local projections. The excitatory motor neurones did not have a preferential direction of projection and the inhibitory motor neurones were too few in number to determine whether they showed any polarity. The only other preparation in which longitudinal muscle motor neurones have been studied in such detail is in the upper stomach of the guinea-pig (Michel, Reiche and Schemann, 2000). Here it was found, not surprisingly, that there was a substantial inhibitory innervation of the muscle layer, with approximately 44% of all motor neurones being immunoreactive for NOS and hence probably inhibitory in function. In addition, the neurones had a marked polarity, with inhibitory neurones being located orally and excitatory motor neurones being located anal to the muscle that they innervated. There were also significant differences in the chemical coding of excitatory motor neurones to the longitudinal muscle of the gastric corpus compared to those of the ileum. For example, about half of the gastric excitatory motor neurones contained ENK immunoreactivity, whereas this was very sparsely distributed in the tertiary plexus of the ileum (Furness, Costa and Miller, 1983). Fewer than 4% of gastric motor neurones to the longitudinal muscle contained calretinin immunoreactivity. This points to substantial functional differences between regions within the same species, presumably reflecting the different physiological requirements related to specialised motor activity. It seems likely that the polarised distributions of inhibitory and excitatory motor neurones seen in the stomach may be more typical than the specialised innervation of the ileum. Since inhibitory reflexes have been evoked in the longitudinal muscle of the guinea-pig colon (Smith and Robertson, 1998), it seems likely that polarised inhibitory innervation of the longitudinal muscle layer may exist here too. Intracellular dye fills of longitudinal muscle motor neurones in the distal colon reveal them to be small unipolar cells with short dendrites, often with a filamentous appearance (Lomax et al., 1999). Their chemical coding has not been investigated in detail and it is not currently clear whether inhibitory motor neurones have a different polarity to excitatory motor neurones. In the human intestine, longitudinal muscle motor neurones have been retrogradely labelled and seen to be numerous, small cells, with relatively short projections to the muscle that they

CELLULAR ORGANISATION OF THE MAMMALIAN ENTERIC NERVOUS SYSTEM

417

innervate (Wattchow et al., 1995). Their chemical coding has not, as yet, been investigated so it is not clear whether excitatory and inhibitory motor neurones show a polarised distribution. CIRCULAR MUSCLE MOTOR NEURONES In most regions of the gastrointestinal tract, the circular muscle layer is the major component of the muscularis externa. It is densely innervated by enteric excitatory and inhibitory motor neurones. In the guinea-pig ileum, it had been shown from myectomy operations that the circular muscle is exclusively innervated by motor neurones with cell bodies in the myenteric plexus (Wilson et al., 1987). Four types of junction potential have been recorded in the circular muscle of the guinea-pig ileum. These include cholinergic excitatory junction potentials (Hirst, Holman and McKirdy, 1975), non-cholinergic excitatory junction potentials (Bywater, Holman and Taylor, 1981; Bauer and Kuriyama, 1982a,b) and non-cholinergic, non-adrenergic (NANC) fast inhibitory junction potentials (Hidaka and Kuriyama, 1969; Hirst, Holman and McKirdy, 1975; Holman and Weinrich, 1975). The addition of apamin, a toxin which blocks channels that mediate the fast inhibitory junction potential (Shuba and Vladimirova, 1980), reveals an additional, slow hyperpolarising junction potential (Niel, Bywater and Taylor, 1983). To identify and characterise the motor neurones that give rise to junction potentials, retrograde tracing studies were carried out, in vitro. DiI was applied in solid form to the microdissected surface of the circular muscle where it was taken up by the axons of motor neurones and transported back to their cell bodies of origin over the following 3–4 days in organotypic culture. Using this methodology, circular muscle motor neurones could be distinguished from all intermingled classes of nerve cell bodies in the myenteric ganglia (Brookes and Costa, 1990). It was confirmed that the great majority of nerve cell bodies labelled from the circular muscle had Dogiel type I morphological characteristics and that they were located both oral and anal to the region of the muscle that they innervated. Combining immunohistochemical labelling with retrograde tracing allowed results from this approach to be integrated with the enormous volume of immunohistochemical data already available. It quickly became apparent that excitatory motor neurones, identified by their immunoreactivity for ChAT, accounted for nearly three quarters of all circular muscle motor neurones. They were located either very close to the circular muscle that they innervated (local projections <2 mm in length) or were located aborally, up to 8 mm from their field of innervation (Brookes, Steele and Costa, 1991b). Inhibitory motor neurones, immunoreactive for VIP, accounted for the remaining quarter of retrogradely labelled nerve cell bodies and were all located oral to their field of innervation, from 0.5–25 mm oral to the DiI application site. Both excitatory and inhibitory motor neurones had a single axon and short lamellar, or occasionally filamentous dendrites, but the excitatory motor neurones were consistently smaller than the inhibitory motor neurones. It also became clear that long, descending inhibitory motor neurones had larger nerve cell bodies than those with shorter projections. This size difference correlated with the presence of immunoreactivity for NFP triplet in the longer, but not shorter motor neurones. Subsequent studies of coexistence of markers (Costa et al., 1996) confirmed that long inhibitory motor neurones were immunoreactive for VIP, NOS and GRP whereas shorter

418

INNERVATION OF THE GASTROINTESTINAL TRACT

inhibitory motor neurones were immunoreactive for VIP, NOS, ENK and NPY and thus form a distinct functional class (Brookes, Steele and Costa, 1991b). Some of these short inhibitory motor neurones are immunoreactive for GABA (Williamson, Pompolo and Furness, 1996) and, from previous studies examining the effects of lesions, some must be immunoreactive for galanin (GAL) (Furness et al., 1987). There were also differences between local and long excitatory motor neurones, although it was not possible to separate these cells into two functional classes. Long excitatory motor neurones were immunoreactive for NFP triplet and were more often immunoreactive for ENK than were shorter motor neurones. The great majority of the cholinergic motor neurones were also immunoreactive for tachykinins, suggesting that non-cholinergic excitation is mediated by the same neurones as cholinergic excitation (Brookes, Steele and Costa, 1991b). The distinct polarities of inhibitory and excitatory motor neurones to the circular muscle of the guinea-pig ileum extended previous functional studies examining the effects of lesions on the amplitude of junction potentials. Inhibitory motor neurones were demonstrated to have projections up to 30 mm long, in the anal direction, although the majority of cells had much shorter aboral projections (Bornstein et al., 1986). Another conclusion of this study was that single inhibitory motor neurones probably innervated very wide bands of circular muscle, extending up to 12 mm along the length of the ileum. This was based on the amplitudes of inhibitory junction potentials on the oral side of lesions to the myenteric plexus. Direct tracing of the fields of innervation of inhibitory motor neurones, using electrical stimulation of antidromic action potentials suggest that single motor neurones innervate bands of circular muscle up to about 1 mm in width (Brookes, unpublished observations). The discrepancy is probably due to the incorrect assumption (Bornstein et al., 1986) that indirect, fast excitatory synaptic activation of motor neurones could be entirely blocked by hexamethonium (Galligan and Bertrand, 1994). In a similar study on non-cholinergic, excitatory junction potentials, it was concluded that excitatory motor neurones (at least those responsible for non-cholinergic junction potentials) projected orally within the myenteric plexus for distances up to 10 mm (Smith et al., 1988) although the majority had projections less than 1 mm long. The same characteristic polarisation of inhibitory and excitatory motor neurones has been seen in several other preparations in other regions of gut and in species other than the guinea-pig. For example, in the guinea-pig gastric corpus, inhibitory motor neurones to the circular muscle were larger, less abundant and located oral to the muscle that they innervated. Excitatory motor neurones were smaller, more numerous and located locally or anally to their field of innervation (Brookes et al., 1998). Similar finding has been reported in the guinea-pig gastric fundus (Pfannkuche et al., 1998a,b). Likewise, the smooth muscle of the guinea-pig lower oesophageal sphincter is innervated by inhibitory motor neurones located orally (usually within the body of the oesophagus) and by excitatory motor neurones lying directly under the sphincter or just anal to it (Brookes et al., 1996). Here however, unlike all other regions studied to date, inhibitory motor neurones outnumbered excitatory motor neurones, accounting for approximately 70% of all filled cells. This may reflect the high intrinsic tone of the sphincter, where excitatory motor neurone input may be functionally less important than in other, phasically active regions of the gut. The gastric sling muscle (also sometimes referred to as the oblique muscle) is known to contribute to the tone of the gastroesophageal junction

CELLULAR ORGANISATION OF THE MAMMALIAN ENTERIC NERVOUS SYSTEM

419

(Preiksaitis and Diamant, 1997), but shows a similar pattern of innervation to the corpus and fundus circular muscle, with excitatory motor neurones in the majority (Yuan and Brookes, 1999). From quantitative multiple labelling immunohistochemical studies, it has been estimated that inhibitory motor neurones to the circular muscle layer comprise up to 17% of all myenteric neurones (Costa et al., 1996). Excitatory circular muscle motor neurones make up at least 12% of all myenteric neurones. Thus, circular muscle motor neurones in total make up between a quarter and a third of all myenteric neurones. Not surprisingly, enteric motor neurones are often recorded when making random impalements with an intracellular microelectrode into myenteric ganglia. It has also been possible to record from motor neurones that have previously been retrogradely labelled with DiI. Not surprisingly, all recordings to date indicate that motor neurones to the circular muscle of the guinea-pig ileum have the physiological characteristics of S cells, receiving prominent fast excitatory synaptic input. This could be predicted from the observation that the majority are immunoreactive for ENK and or VIP (Bornstein et al., 1984a; Katayama, Lees and Pearson, 1986). It is clear that S cells can be subdivided into several subtypes, which correlate with their projections and functional type. For example, many S cells recorded randomly in myenteric ganglia fire phasically to depolarising current pulses. This rapid accommodation has been attributed to an outwardly rectifying current which develops after the onset of depolarisation (Kunze et al., 1997). However, when nerve cell bodies located close to the base of internodal strands were targeted, the majority of cells were slowly accommodating (Smith, Burke and Shuttleworth, 1999). Many of these cells were longitudinal muscle motor neurones, which are only infrequently recorded during random impalements (Bornstein et al., 1991a). However, care must be taken when comparing data recorded under different conditions. Slow excitatory synaptic inputs, evoked electrically, convert rapidly accommodating neurones into slowly accommodating (Kunze et al., 1997). If the mucosa of the preparation is left attached to the preparation, spontaneously occurring slow excitatory synaptic inputs, probably arising from enteric primary afferent neurones, cause most S cells to be in the slowly adapting state (Kunze et al., 1997). Since enteric primary afferent neurones are also powerfully activated by distension (Kunze et al., 1998, 1999), it is likely that the state of excitability of S cells will critically depend on the degree of stretch of the preparation in which they are recorded (Kunze et al., 1998). For this reason, it is important that cells be characterised under identical recording conditions, to establish whether or not there are differences in the electrophysiological characteristics. We have compared electrophysiological characteristics of retrogradely labelled, excitatory and inhibitory circular muscle motor neurones in the guinea-pig ileum under identical recording conditions. Preliminary results confirmed that all motor neurones received prominent fast excitatory synaptic potentials. However, inhibitory motor neurones were consistently less excitable than excitatory motor neurones, when their responses to depolarising current pulses were recorded. Typically inhibitory motor neurones fired one or two action potentials at the onset of depolarisation whereas excitatory motor neurones fired up to 5–6 action potentials. In addition, excitatory motor neurones showed a characteristic inflection in their responses to depolarising current pulses which delayed the first action potential. An identical inflection has been shown, in another class of S cells; the ascending

420

INNERVATION OF THE GASTROINTESTINAL TRACT

cholinergic interneurones, to be due to activation of a transient outward current (Brookes et al., 1997). This current was consistently present in excitatory motor neurones but was never seen in inhibitory motor neurones. These results strongly indicate that there are differences in the electrical properties of different classes of motor neurones, recorded under identical conditions. Further study will be needed to understand the functional significance of these differences for the control of intestinal motility. Circular muscle motor neurones: other regions of the guinea-pig In the guinea-pig gastric corpus, a detailed analysis of the chemical coding of excitatory and inhibitory motor neurones revealed gross similarities to the small intestine, but with some notable differences (Michel, Reiche and Schemann, 2000). Excitatory motor neurones were immunoreactive for ChAT with various combinations of tachykinin and ENK immunoreactivity, as seen in the small intestine, however a small proportion (just over 5%) were immunoreactive for NPY, a combination not seen in the ileum (Costa et al., 1996). Amongst the inhibitory motor neurones, NOS was colocalised with combinations of NPY and ENK immunoreactivity as for the short inhibitory motor neurones in the ileum. There is no data currently available about the distribution of NFP triplet or GRP, which are markers for the long inhibitory motor neurones, so it is not currently clear whether the distinction between long and short neurones exists in the stomach. However, differences in the distribution of cells between the lesser and greater curvatures were reported (Michel, Reiche and Schemann, 2000). In the guinea-pig distal colon, immunohistochemical study of the chemical coding of the varicose endings of the axons of motor neurones again reveals a similar situation (Lomax and Furness, 2000). VIP and NOS coexist completely in a set of descending inhibitory motor neurones, some of which also contain NPY. Unlike the ileum, however, ENK immunoreactivity is confined to excitatory motor neurones where it colocalises with a cholinergic marker [the vesicular acetylcholine transporter, which is a better marker for varicosities than ChAT (Weihe et al., 1996)] and tachykinins. A similar situation pertains in the proximal colon of the guinea-pig (Messenger, 1993). Circular muscle motor neurones: other species The same basic scheme appears to account for circular muscle motor neurones as they have been studied in other species. In the rat small intestine, lesion studies demonstrated that VIP-containing axons had descending projections to the circular muscle and also contained NPY immunoreactivity (Ekblad et al., 1987) and NOS (Ekblad et al., 1994). However, in the rat colon, NPY immunoreactivity was present in few presumed inhibitory motor neurones (Ekblad et al., 1988). In motor neurones in both small and large intestine, ENK immunoreactivity was confined to an ascending pathway to the smooth muscle and was not affected anal to a lesion. Thus it is likely to be present mainly in ascending excitatory neuronal pathways, and probably in cholinergic and tachykinin-containing nerve fibres in the small intestine. The innervation of circular muscle of the mouse small intestine is similar to that of the guinea-pig and rat, with NOS-immunoreactive fibres of inhibitory motor neurones usually containing VIP and often colocalising with NPY.

CELLULAR ORGANISATION OF THE MAMMALIAN ENTERIC NERVOUS SYSTEM

421

Excitatory motor neurone axons contain ChAT (or VAChT) and tachykinins are also present, but unlike the guinea-pig small intestine, these often contained calretinin immunoreactivity (Sang and Young, 1996). The patterns of coexistence of ENK have not been quantitatively studied in mouse intestine to date. In the colon, the situation was very much more complex. While NOS and VIP sometimes colocalised, they were also found separately and NPY was very sparse. GABA immunoreactivity was present in subsets of both inhibitory and excitatory motor neurones. There appeared to be greater heterogeneity of coding of excitatory motor neurones in the mouse colon with various combinations of VAChT ± tachykinins ± calretinin, with some calretinin fibres apparently non-cholinergic. In the human, the projections of circular muscle motor neurones have been studied using retrograde tracing techniques. Nerve fibres, immunoreactive for ChAT were abundant in the circular muscle of the both small and large intestine (Porter et al., 1996). When motor neurones were retrogradely labelled from the circular muscle of the colon, cholinergic nerve cell bodies were typically located aboral to their field of innervation and made up a small majority of all filled neurones in the myenteric plexus (Porter et al., 1997). Inhibitory motor neurones, immunoreactive for NOS, but not ChAT made up 48% of the filled cells and were typically located oral to the circular muscle that they innervated. VIP was present in a subset of inhibitory motor neurones (Wattchow et al., 1997) whereas tachykinin immunoreactivity was present in a small subset of excitatory motor neurones (Wattchow et al., 1997). As reported elsewhere, all of the motor neurones that could be recognised were unipolar and generally belonged to the Dogiel type I morphological class. It has previously been shown that NPY is present in a subset of inhibitory (NADPHdiaphorase-reactive) motor neurones (Wattchow, Furness and Costa, 1988; Nichols, Staines and Krantis, 1994). ENK immunoreactivity, as in the rat colon is restricted to ascending excitatory motor neurones (Wattchow, Furness and Costa, 1988). In addition to the innervation from the myenteric plexus, the circular muscle of the human colon is also innervated by motor neurones with cell bodies in the submucous plexus (Porter et al., 1999). Of these cell bodies, 97% were located either in Schabadasch’s or the intermediate plexus: only 3% were located in the innermost plexus of Meissner. Of these cell bodies, only 11% were immunoreactive for NOS and most of these also contained VIP. This suggests that the remaining 89% of neurones labelled from the circular muscle contained neither of these markers and may be excitatory motor neurones, although this was not directly confirmed. It has been established in the canine colon that inhibitory motor neurones project from the submucous plexus to the circular muscle (Furness et al., 1990a). Physiological studies suggest that both excitatory and inhibitory motor neurones in the circular muscle may modulate pacemaker activity in the circular muscle of canine colon (Sanders and Smith, 1986). It seems likely that the presence of motor neurone cell bodies in the submucous plexus may be characteristic of the intestines of larger animals and humans, although this needs systematic testing.

INTERNEURONES The presence of varicose branching nerve fibres in enteric ganglia, which do not arise from enteric primary afferent neurones, indicates the existence of enteric interneurones.

422

INNERVATION OF THE GASTROINTESTINAL TRACT

Thus, in the guinea-pig ileum, these include varicosities immunoreactive for 5-HT (Costa et al., 1982), calretinin (Brookes, Steele and Costa, 1991a), ENK (Furness, Costa and Miller, 1983), SOM (Costa et al., 1980c), VIP (Costa et al., 1980b), NPY (Uemura, Pompolo and Furness, 1995), pituitary adenylyl cyclase-activating peptide (PACAP) (Portbury et al., 1995a), GAL (Furness et al., 1987), NOS (Costa et al., 1992), GRP (Costa et al., 1984) and perhaps some of those immunoreactive for a tachykinin (Costa et al., 1980a). A combination of immunohistochemical studies, combined with lesions and retrograde tracing in vitro have identified at least five classes of interneurones that account for these varicose endings in myenteric and submucous ganglia in the guineapig ileum. ASCENDING INTERNEURONES The first clue about the identity of ascending interneurones came from the observation that lesions to the myenteric plexus caused degeneration of ENK-immunoreactive nerve endings in myenteric ganglia for several millimetres orally, but not anally (Furness, Costa and Miller, 1983). Later it was realised that ENK-immunoreactive varicosities in myenteric ganglia were also immunoreactive for calretinin, which also disappeared oral to such lesions (Brookes, Steele and Costa, 1991a). Calretinin is present in two major classes of neurones in the myenteric plexus of the guinea-pig small intestine; the ascending interneurones, which are also immunoreactive for ENK, tachykinins, NFP triplet and ChAT, and in longitudinal muscle motor neurones which lack ENK and NFP immunoreactivity (Brookes, Steele and Costa, 1991a). Detailed study of the projections of this class of nerves, using retrograde tracing in vitro, led to the conclusion that there is just one functional class of ascending interneurone in the guinea-pig small intestine (Brookes et al., 1997). These cholinergic neurones project orally for up to 14 mm, giving rise to varicose side branches in myenteric ganglia more than 3–4 mm oral to their nerve cell bodies. They have typical Dogiel type I morphology, with lamellar dendrites and a single, orally directed axon and account for just 5% of all myenteric nerve cells (Costa et al., 1996). Targeted intracellular recordings made from these neurones showed that they have prominent fast excitatory synaptic potentials, which are often sufficient to drive action potentials. It is difficult to trace the sources of synaptic inputs using just electrical stimulation because one cannot distinguish between orthodromic and antidromic activation of nerve fibres. Given that there are axons up to 100 mm long running down the intestine, which give rise to multiple collaterals along their length, electrical stimulation at any point in the preparation (oral or anal to the recorded cell body) may activate synaptic inputs from this site. Nevertheless, ascending interneurones were seen to receive prominent fast synaptic inputs from anally located cells, presumably of the same class. This was confirmed by focally pressure ejecting the nicotinic agonist dimethyl phenylpiperazinium (DMPP) onto ganglia anal to the recorded cell. Unlike electrical stimulation, DMPP does not activate axons. This is readily demonstrated by the observation that spritzing DMPP onto internodal strands never evoked synaptic potentials. At some anally-located “hotspots” in ganglia, DMPP activated fast synaptic inputs which must have arisen from other ascending interneurones. This demonstrated, functionally, that ascending interneurones are organised into chains running up the intestine. This is

CELLULAR ORGANISATION OF THE MAMMALIAN ENTERIC NERVOUS SYSTEM

423

compatible with the report that calretinin immunoreactive nerve cell bodies are often surrounded by calretinin-immunoreactive varicosities, which must arise from this class of ascending interneurone (Pompolo and Furness, 1995). Ascending interneurones were highly excitable cells in the organ-cultured preparations in which they were studied (Brookes et al., 1997), responding to depolarising current pulses with slowly adapting trains of action potentials. However, the same class of cells in fresh preparations, which were not strongly stretched, have also been reported to have phasic, fast adapting responses to depolarising current (Kunze and Furness, 1999). It has recently been reported that in fresh stretched preparations, these neurones fire tonically (Smith, Burke and Shuttleworth, 1999). Thus it seems likely that the degree of stretch may alter the basal electrophysiological characteristics of these neurones from fast adapting to slow adapting, probably due to the activation of slow synaptic inputs from stretch-sensitive enteric primary afferent neurones. Ascending interneurones consistently showed (27 of 29 cells) a marked inflection in their responses to depolarising current pulses. This was due to a transient outward current, similar to the “A current”, that had previously only been reported in AH cells (Hirst, Johnson and van Helden, 1985b; Galligan, North and Tokimasa, 1989) but which has recently been characterised in S cells too (Starodub and Wood, 2000). While the presence of this current may influence the patterns of firing of these neurones, it is also notable for the fact that we were unable to detect it during targeted recordings of inhibitory motor neurones, under identical recording conditions. This suggests that it may be possible to identify electrophysiological subtypes of S cells in the myenteric plexus of the guinea-pig ileum. Ascending interneurones also displayed a strong inwardly-rectifying current when they were hyperpolarised. The morphological and electrophysiological characteristics of these cells are readily compatible with a number of observations on the reflex behaviour of specimens of guinea-pig ileum. Rapid distension evokes an ascending excitatory reflex up to 40 mm oral to the stimulus (Smith, Bornstein and Furness, 1990), which is considerably further than the sum of the projections of enteric primary afferent neurones and excitatory motor neurones. The direct demonstration that interneurones make functional chains, linked by cholinergic nicotinic synapses (Brookes et al., 1997), is compatible with the observation that ascending interneuronal pathways, studied in a divided organ bath, were substantially mediated via nicotinic receptors (Tonini and Costa, 1990). The entire reflex pathway is substantially inhibited by nicotinic blockers (Holzer, 1989; Smith, Bornstein and Furness, 1990; Tonini and Costa, 1990; Johnson et al., 1996). However, in the presence of nicotinic blockade, some ascending excitation persists (Holzer, 1989). It is possible that tachykinins release from ascending interneurones may contribute to this pathway. This reflex pathway probably also contributes to the contraction occurring during distension-induced peristalsis in this preparation (Waterman, Tonini and Costa, 1994; Brookes et al., 1999), which is also substantially inhibited by hexamethonium. However, naloxone, the opiate receptor antagonist can restore peristalsis in the continuing presence of hexamethonium (Bartho, Holzer and Lembeck, 1987; Bartho et al., 1989). It is tempting to speculate that some of this effect may be due to blockade of the actions of endogenous opioids, including ENK, which are present in both excitatory motor neurones and in ascending interneurones.

424

INNERVATION OF THE GASTROINTESTINAL TRACT

Ascending interneurones in other regions of the guinea-pig Immunohistochemical studies, in some cases combined with lesions, have identified a number of neurochemicals located in orally-directed pathways in other parts of the gastrointestinal tract of the guinea-pig. In the stomach, ascending interneurones have not been unequivocally identified, but are likely to be cholinergic, based on analysis of randomly dye filled myenteric neurones (Schemann and Schaaf, 1995). In the distal colon, some ascending interneuronal pathways, identified by varicosities in ganglia and the effects of lesions, were reported to contain immunoreactivity for NPY and this has been proposed to be a major difference in the chemical coding of ascending interneurones from the guineapig ileum (Messenger and Furness, 1990). However, in a recent study, it was reported that most of the immunoreactivity for NPY, particularly in ascending non-varicose axons, arises from extrinsic neurones (Browning et al., 1999). Sectioning the colonic nerves reduced NPY immunoreactivity substantially in the mid- to upper colon, whereas NPYimmunoreactive nerve cell bodies in the pelvic ganglia supply the distal colon and rectum (Browning et al., 1999). NPY-immunoreactive nerve fibres from these sources do not contain tyrosine hydroxylase immunoreactivity. It has previously been demonstrated in the colon of many species, including the guinea-pig, that large thick trunks, called “shunt fascicles” contain the orally directed axons of extrinsic neurones (Christensen et al., 1984). In addition, intrinsic enteric neurones, also appear to use them as a conduits over relatively long distances running both orally and anally (McRorie, Krier and Adams, 1991). A recent study has reported ascending interneurones in the guinea-pig distal colon having a more filamentous morphology than their counterparts in the small intestine (Lomax et al., 1999). Their chemical coding also appears to be rather variable since some contain either SOM or VIP immunoreactivity in addition to ChAT and ENK (Lomax and Furness, 2000). While this pattern of colocalisation is clearly different from the small intestine, it is not clear whether there are really two functional classes of ascending interneurone, or whether this represents variability within a single class. According to the criteria that we have proposed, (that two differentiating characteristics should co-vary) this remains open. In the guinea-pig proximal colon, the coding of ascending interneurones may be similar to that of the colon. Calretinin immunoreactive neurones were reported to have predominantly a descending projection, based on disappearance of immunoreactive nerve fibres from ganglia following myotomies (Messenger, 1993). However calretinin immunoreactivity was present in a substantial proportion of intracellularly dye filled neurones with ascending projections in the proximal colon (Messenger, Bornstein and Furness, 1994). This discrepancy may reflect inaccuracy in quantifying the loss of immunoreactive nerve fibres following lesions. It appears that SOM immunoreactivity may also be present in at least some ascending interneurones in the proximal colon (Messenger, 1993).

Ascending interneurones in other species In the rat small and large intestine, ENK immunoreactivity in ganglia was depleted on the oral side of a lesion to myenteric pathways, indicating that at least some of the ascending pathways contain ENK immunoreactivity. These neurones were estimated to project for 6–7 mm orally. Since fibres disappeared oral to the lesion in both ganglia and muscle, it is

CELLULAR ORGANISATION OF THE MAMMALIAN ENTERIC NERVOUS SYSTEM

425

likely that both ascending interneurones and excitatory motor neurones contain this peptide, or alternatively that the same neurones project to both targets (Ekblad et al., 1988). In addition, in the rat colon, SP immunoreactivity was also depleted oral to a myectomy (Ekblad et al., 1988), however this was not the case in the small intestine where SP-immunoreactive neurones appear to project primarily in the anal direction (Ekblad et al., 1987). In the mouse, the projections of ENK-immunoreactive neurones have not been determined, but lesion studies suggest that SP is present in ascending pathways in both the small and large intestine (Sang, Williamson and Young, 1997). Approximately 80% of all SP-immunoreactive terminals in myenteric ganglia are cholinergic in both small and large murine intestine (Sang and Young, 1998a). This suggests that there is a population of ascending cholinergic interneurones which also contains SP or related tachykinin but which rarely contains calretinin (Sang and Young, 1996). In the human colon, excitatory motor neurones had projections up to 10 mm long. However, application of DiI to myenteric ganglia, labelled orally projecting cells with axons up to 40 mm long. Thus a substantial population of ascending neurones, which do not project to either the circular or longitudinal muscle (Wattchow, Brookes and Costa, 1995; Wattchow et al., 1997) is present. These neurones are largely cholinergic, based on ChAT immunoreactivity (Porter et al., 2002) and about one quarter to one third of them are immunoreactive for tachykinins (Wattchow et al., 1997) but not for calretinin, which is present in descending pathways. The projections of ENK-immunoreactive neurones have not been determined in the human gut although ENK is likely to be present in a population of descending interneurones, immunoreactive for GABA transaminase and NOS (Krantis, Nichols and Staines, 1998). Whether or not it is also in ascending interneuronal pathways remains to be determined. DESCENDING INTERNEURONES Studies in the guinea-pig small intestine have shown that long descending interneuronal pathways are exclusively located in the myenteric plexus. Applying retrogradely transported tracers to submucous ganglia did not label cell bodies more than 3 mm oral or anal to the application site (Song et al., 1992). While it is possible that chains of short interneurones could be functionally important, there is no physiological evidence for their existence. In the myenteric plexus, it is clear that there are long descending non-motor pathways. Application of a retrograde tracer to a myenteric ganglion labels nerve cell bodies up to 136 mm orally. However, the longest motor neurone projection identified to date in the guinea-pig small intestine is about 25 mm long. Thus, all cells with projections longer than 25 mm must be either primary afferent neurones or interneurones. It is now well established that the situation with descending interneuronal pathways is considerably more complex than the ascending interneuronal pathway which appears to consist of a single cell type. At least four populations of descending interneurones have been characterised immunohistochemically (Costa et al., 1996) in addition to the “dendritic” Dogiel type II neurones with long descending projections (Brookes et al., 1995), which probably function as primary afferent neurones (see above). The characterisation of descending interneurones in the guinea-pig small intestine will be considered first, then compared to the populations of interneurones in other regions of the guinea-pig gut, and in other species.

426

INNERVATION OF THE GASTROINTESTINAL TRACT

Somatostatin-immunoreactive interneurones SOM was one of the first peptides to be localised immunohistochemically in enteric neurones (Hökfelt et al., 1975). While it is well established as a mediator of D cells (entero-endocrine cells of the stomach; Brand and Schmidt, 1991) it was also shown to be present in a subset of neurones in the guinea-pig small intestine (Costa et al., 1980c). Subsequent work has characterised these neurones in some detail and it is now clear that there are two classes of myenteric neurones with immunoreactivity for SOM (Costa et al., 1996). One class projects to the mucosa and has immunoreactivity for NPY and a large number of other peptides (see section below on submucous neurones; Furness et al., 1985) and the other is a class of interneurones which comprises about 4% of myenteric neurones. Immunoreactivity for ChAT has shown that these interneurones are also likely to be cholinergic (Steele, Brookes and Costa, 1991). Retrograde tracing studies have shown that these cholinergic interneurones project anally in the myenteric plexus for up to 70 mm, confirming and extending results obtained by lesioning pathways (Costa et al., 1980c). They give rise to numerous SOM-immunoreactive varicosities in myenteric ganglia which often surround the cell bodies of other neurones of the same class (Portbury et al., 1995b; Song et al., 1997a). In addition, apparently identical neurones can be retrogradely labelled from the submucous ganglia (Song et al., 1997a; Meedeniya et al., 1998). It is likely, but has not been shown directly, that individual SOM-containing interneurones project to ganglia in both plexuses. SOM-immunoreactive interneurones which project to submucous ganglia are responsible for the non-adrenergic inhibitory synaptic potentials that can be recorded in submucous neurones following extrinsic denervation (Shen and Surprenant, 1993). SOM-immunoreactive interneurones have a very characteristic morphology. They are typically located in the middle of ganglia, often in clusters of 3–7 neurones. They have a smooth, elongated cell body with long filamentous dendrites which are readily distinguishable from the processes of most Dogiel type I cells. They have a single long process which typically projects anally out of the ganglion, often giving rise to varicose collaterals as it passes through neighbouring ganglia (Portbury et al., 1995b; Song et al., 1997a). While their soma-dendritic morphology is distinctive, they also have very characteristic combinations of electrophysiological characteristics. They often have an unusually high input impedance when recorded with conventional “sharp” microelectrodes. Following depolarising current pulses, there is a marked “notch” on the rising phase of the membrane response, very similar to that caused by the transient outward current characterised in ascending interneurones (Brookes et al., 1997). Their soma action potentials often show an inflection on the falling phase, similar to that of AH cells, but the long after-hyperpolarisation is typically rather small. However when made to fire repeatedly during a prolonged depolarising current pulse, the after hyperpolarisation can summate to cause a marked reduction in excitability. These cells also appear to have a characteristic “sag” in their responses to hyperpolarising current pulses which has been shown to be due to the cation current IH (Galligan et al., 1990b). This current is very noticeable during recordings from enteric primary afferent neurones with Dogiel type II morphology and AH cell characteristics. All SOM interneurones receive prominent fast excitatory synaptic inputs. It seems likely that these may be amongst the cells that have been classified as AH cells with fast EPSPs.

CELLULAR ORGANISATION OF THE MAMMALIAN ENTERIC NERVOUS SYSTEM

427

Some studies have been carried out to determine the connectivity of SOM interneurones in the guinea-pig small intestine. In conventional fluorescence microscopy it is clear that clusters of SOM-immunoreactive nerve cell bodies in myenteric ganglia are surrounded by higher than average densities of SOM-immunoreactive varicosities. This has been studied quantitatively using electron microscopy, showing that SOM-immunoreactive varicosities often make synaptic contacts onto SOM interneurones, and that these synapses actually account for more than half of their inputs (Portbury et al., 1995b; Pompolo and Furness, 1998). Likewise, a high density of SOM-immunoreactive close appositions has been recorded using confocal microscopy (Mann et al., 1997; Song et al., 1997a). It is likely that these neurones make long functional chains running the length of the small intestine, probably connected primarily via cholinergic nicotinic synapses, given the electrophysiologically-characterised inputs to these cells (Song et al., 1997a). An interesting observation about the sources of inputs to these cells was made using ultrastructural methods, in which it was reported that calbindin-immunoreactive varicosities account for very few inputs to SOM-immunoreactive interneurones (Pompolo and Furness, 1998). This is significant because it suggests that SOM-immunoreactive interneurones are not likely to be strongly activated by enteric primary afferent neurones. They may be activated indirectly since they receive inputs from NOS or GRP-immunoreactive interneurones (Mann et al., 1997; Pompolo and Furness, 1998), but it is possible that the source of the synaptic drive to these cells originates elsewhere. The axons of SOM neurones must affect other parts of the enteric neuronal circuitry in order for them to play any role in physiology. It is not clear which other classes of enteric neuron they influence, although they probably make some contact with NOS immunoreactive motor neurones or interneurones. It is also possible that they may have a modulatory role, perhaps not entirely dependent on close synaptic contacts, and possibly even utilising volume transmission. Somatostatin 2a receptors have been localised immunohistochemically in the enteric nervous system of the rat (Sternini et al., 1997) and in the guinea-pig myenteric plexus SOM depolarises some neurones and hyperpolarises others (Katayama and North, 1980). Thus it is possible that the SOM-immunoreactive interneurones tend to be activated throughout the length of the small intestine and may modulate activity in other enteric neuronal circuits. One could speculate that activation of this circuit might predispose the enteric circuitry towards certain functional states, for example changing from the fasted state to the fed state after a meal. 5-HT immunoreactive interneurones Immunoreactivity for 5-HT is present in numerous neuronal varicosities in myenteric and submucous ganglia in freshly fixed preparations of guinea-pig ileum (Costa et al., 1982). The number of labelled varicosities can be substantially increased by pre-loading the preparations with 5-HT or with 5,7-dihydroxytryptamine, a 5-HT analogue which is resistant to breakdown by monoamine oxidase. With still higher concentrations of 5-HT analogues, nerve cell bodies are labelled. These account for about 1% of all myenteric neurones and have typical Dogiel type I morphological features, with lamellar dendrites and a single, anally directed axon. The cells are small to medium sized and often located near the origin of internodal strands, or pressed hard up against the edges of ganglia. All of

428

INNERVATION OF THE GASTROINTESTINAL TRACT

the 5-HT immunoreactive neurones also contain immunoreactivity for ChAT and hence are likely to be cholinergic. They are also strongly immunoreactive for NFP triplet (Costa et al., 1996). Based on lesion studies, it was estimated that these neurones have exclusively anal projections, up to about 24 mm long to myenteric ganglia and that they project up to 12 mm anally to submucous ganglia (Furness and Costa, 1982). Later retrograde tracing studies showed that these were underestimates, and that 5-HT-immunoreactive cholinergic interneurones project up to 100 mm anally to myenteric ganglia and up to 40 mm anally to submucous ganglia (Meedeniya et al., 1998). No studies have specifically identified any intracellularly recorded neurones as having 5-HT immunoreactivity, so their electrophysiological properties have not been characterised. It is likely however that they are S cells, given their characteristic Dogiel type I morphology. The connectivity of 5-HT interneurones has been studied both ultrastructurally and in light microscopy. It was reported that 5-HT varicosities were numerous around electrophysiologically identified AH cells which had subsequently been intracellularly dye filled (Erde, Sherman and Gershon, 1985). This was taken as evidence that 5-HT might be a mediator of the prominent slow EPSP in AH cells (see above). However, this result was not confirmed in a later study, where 5-HT immunoreactive varicosities were relatively sparse around Dogiel type II neurones and rarely formed synaptic contacts (Young and Furness, 1995). However, 5-HT varicosities were very dense, and frequently formed synaptic specialisations around the cell bodies of 5-HT immunoreactive nerve cell bodies, suggesting that these interneurones also form functional chains running along the guinea-pig small intestine (Young and Furness, 1995). Other classes of descending interneurones The varicose endings of 5-HT interneurones and SOM interneurones are readily identifiable by the single markers (5-HT and SOM) which define the classes. However varicosities with other combinations of neurochemicals have also been identified in myenteric and submucous ganglia of guinea-pig small intestine. A few of these types arise from sources extrinsic to the gut, including sympathetic nerve varicosities which are immunoreactive for tyrosine hydroxylase and spinal afferent nerves, immunoreactive for SP and CGRP (Gibbins et al., 1985). However, varicosities have been reported that are immunoreactive for: VIP (Costa et al., 1980b), NPY (Furness et al., 1983), GRP/bombesin (Costa et al., 1984), GAL (Furness et al., 1987), NOS (Costa et al., 1992), helospectin (Absood et al., 1992) alkaline phosphatase (Song, Brookes and Costa, 1994b) and PACAP (Portbury et al., 1995a), which are unlikely to arise from extrinsic sources. In most cases, lesion studies suggest that these markers are present in neurones with aboral projections in the myenteric plexus. In addition, they are not present in ascending pathways, nor in the circumferential pathways mediated by Dogiel type II cells. In double or triple labelling studies, many of these markers have been shown to coexist in the same neurones, however in most cases the overlap has not been complete (Uemura, Pompolo and Furness, 1995). It is not possible, at present, to state how many functional classes of descending interneurones are defined by the presence of one or more of these markers. The problem is compounded by the fact that all of these markers are also present in inhibitory motor neurones, with descending projections to the circular muscle (Costa et al., 1996).

CELLULAR ORGANISATION OF THE MAMMALIAN ENTERIC NERVOUS SYSTEM

429

A study on the distribution of ChAT immunoreactivity in myenteric neurones suggested that the presence of ChAT immunoreactivity might be used to distinguish interneurones from inhibitory motor neurones. Steele, Brookes and Costa (1991) demonstrated that 16% of VIP-immunoreactive nerve cell bodies were also immunoreactive for ChAT (Steele, Brookes and Costa, 1991). Since inhibitory motor neurones were immunoreactive for VIP but not ChAT (Brookes, Steele and Costa, 1991b), some VIP neurones were likely to be cholinergic descending interneurones in the myenteric plexus. This suggestion was supported by the observation that there were varicosities in ganglia that were immunoreactive for both ChAT and VIP (Steele, Brookes and Costa, 1991). It is known that VIP and NOS show nearly complete overlap in the myenteric plexus of the guinea-pig small intestine (Costa et al., 1992), the only exceptions being a tiny population of VIP-immunoreactive neurones with filamentous morphology which project to the mucosa (Song, Brookes and Costa, 1991) and very scarce viscerofugal neurones (Mann et al., 1995). Therefore it was to be expected that about 16% of NOS-immunoreactive neurones would contain ChAT. This was confirmed in two recent studies showing that 15–16% of NOS immunoreactive cells are apparently cholinergic (Li and Furness, 1998, 2000). It has recently been suggested by the same group that all VIP/NOS interneurones in the guinea-pig myenteric plexus are cholinergic (Li and Furness, 2000). This appears to be a deduction based on the presence of NOS/ChAT immunoreactive cell bodies in ganglia, which do not project to the muscle layers. To date, no quantitative studies of the coexistence of NOS or VIP and cholinergic markers (ChAT or the vesicular acetylcholine transporter, VAChT) in varicosities in myenteric ganglia have been reported. These would be essential to reach this conclusion. In the guinea-pig colon, it is clear that many VIP varicosities do not contain immunoreactivity for the vesicular acetylcholine transporter (see Lomax, Zhang and Furness, 2000). Preliminary studies in our laboratory, using confocal microscopy, have suggested that while some NOS-immunoreactive varicosities in myenteric ganglia of the guinea-pig small intestine also contain ChAT, many do not. This suggests that there are at least two functional classes of descending interneurones with NOS/VIP immunoreactivity: one cholinergic, the other non-cholinergic (Costa et al., 1996). It is well established, from a variety of physiological studies, that a significant proportion of electrically stimulated fast EPSPs in the myenteric plexus cannot be inhibited by the nicotinic blocker hexamethonium, and are probably mediated by ATP acting on P2X receptors (Galligan and Bertrand, 1994). The site at which purinergic fast transmission operates has been tested by examining the effects of lesions on the ratio of nicotinic: nonnicotinic fast EPSPs. This approach suggests that descending pathways, presumably mediated by descending interneurones, are responsible for a substantial proportion of nonnicotinic fast transmission (LePard and Galligan, 1999). Analysis of descending inhibitory reflexes has recently been reported in a multi-chambered organ bath, in which drugs could be selectively applied to the site of stimulation, the recording site, or an intermediate chamber where, presumably, interneuronal synapses were present. This study showed that descending inhibitory reflexes were more sensitive to P2X antagonist PPADS than were electrically evoked inhibitory junction potentials, strongly suggesting a major contribution by purinergic synapses to descending inhibitory pathways (Bian, Bertrand and Bornstein, 2000). It is well established that descending inhibitory pathways in the guinea-pig ileum are much less sensitive to nicotinic blockade, than are ascending excitatory pathways

430

INNERVATION OF THE GASTROINTESTINAL TRACT

(Smith and Furness, 1988; Smith, Bornstein and Furness, 1990). These observations are all compatible with the presence of non-cholinergic descending interneurones, although it has to be conceded that it is also equally possible that some descending interneurones release acetylcholine and ATP simultaneously. It has also been reported that NOS immunoreactive varicosities, which arise from VIP/NOS descending interneurones, make substantial number of synaptic and close contacts onto the cell bodies of other NOSimmunoreactive cells, which must include descending interneurones (Young, Furness and Povey, 1995). Electrophysiological recordings confirm that inhibitory motor neurones receive purinergic fast excitatory synaptic inputs (Johnson et al., 1999). In summary, VIP-immunoreactive descending interneurones cannot, as yet, be unequivocally distinguished from inhibitory motor neurones to the circular muscle on the basis of their chemical coding. While inhibitory motor neurones are not cholinergic, some, but probably not all of the descending interneurones contain immunoreactivity for ChAT. The most appropriate way to subdivide VIP-immunoreactive descending interneurones into functional classes is not yet apparent. Some of the VIP/NOS immunoreactive interneurones project to submucous ganglia where they make contacts with other nerve cell bodies (Li, Young and Furness, 1995). Descending interneurones: other regions In the guinea-pig proximal colon, VIP, GRP and GAL immunoreactivities are all present in descending pathways to myenteric ganglia, similar to the small intestine (Messenger, 1993). In addition, NADPH diaphorase, which usually corresponds closely with NOS immunoreactivity (Belai et al., 1992; Ward et al., 1992; Young et al., 1992), is also present in descending pathways. More detailed studies on colocalisation have not been carried out for VIP containing interneurones in the proximal colon. In the distal colon, there are descending interneurones immunoreactive for both VIP and ChAT and another, smaller population, immunoreactive for VIP, NOS and ChAT (which often contain GRP and sometimes calbindin). One other population is immunoreactive for NOS alone (without either ChAT or VIP) (Lomax and Furness, 2000). Again, it is not clear how to divide these populations into functional classes on the basis of these observations. Interestingly, SOM immunoreactivity is not obviously present in descending pathways in the proximal colon. SOM-immunoreactive varicosities disappear in myenteric ganglia on the oral side of a lesion, suggesting that SOM is present in ascending pathways. However, this may have masked a presence in descending pathways. It was noted that SOM immunoreactivity was present in sprouting nerve fibres on the oral side of lesions, suggesting that small numbers of axons may have descending projections (Messenger, 1993). In the distal colon, it is clear that SOM is present, along with calretinin in a class of ascending cholinergic interneurones, but does not appear to be present in descending pathways (Lomax and Furness, 2000). 5-HT-containing neurones of the guinea-pig distal colon are also descending interneurones, providing terminals in both myenteric and submucous ganglia aboral to their cell bodies (Wardell, Bornstein and Furness, 1994). In the stomach the situation is not entirely clear. 5-HT-containing neurones are rather more prevalent than in the small intestine and appear to give rise to varicosities essentially only in the myenteric ganglia and thus are

CELLULAR ORGANISATION OF THE MAMMALIAN ENTERIC NERVOUS SYSTEM

431

probably interneurones (Mawe et al., 1989). However, the polarity of their projections is not currently known, but it is clear that they have rather different chemical coding from the small intestine. While most appear to be cholinergic, about three quarters also contain SP immunoreactivity and about one quarter contain VIP immunoreactivity (Schemann et al., 1995). Descending interneurones: other species When comparing classes of cells between species, it is not possible to determine whether two classes are homologous (i.e. that they serve identical functions, having probably evolved from a common cell type) or whether they simply show convergence of a particular feature. This must be taken into consideration when comparing descending pathways in different species. Our understanding of the enteric neuronal circuitry in the guinea-pig ileum is more advanced than in any other species. In many cases there is, relatively speaking, a lack of information about pathways in other species; this is particularly true given the complexity of descending interneuronal pathways. In this section, some similarities and differences between descending interneuronal pathways in various species will be highlighted. The similarities may eventually be shown to reflect homologous classes, but this has not been established at the time of writing. 5-Hydroxytryptamine immunoreactivity has been studied in the enteric nervous system of many mammalian species including the ascending colon of the pig (Barbiers et al., 1995), rat stomach and duodenum (Kirchgessner and Gershon, 1989), rat colon (Nada and Toyohara, 1987), dog small intestine (Mann and Bell, 1993) and in the human gut (Griffith and Burnstock, 1983). In both pig and human occasional cells were present in the submucous plexus, especially the outer submucous plexus (Crowe et al., 1992), whereas in other species, 5-HT containing nerve cell bodies were found only in the myenteric plexus. It is not clear whether submucous 5-HT-containing neurones function as descending interneurones, although in the pig it has been shown that at least some of them project out of the gut to prevertebral sympathetic ganglia (Timmermans et al., 1993). In both the small and large intestine of the mouse, 5-HT is colocalised with ChAT (Sang and Young, 1996, 1998a) and appears to be present in varicose endings in myenteric and submucous ganglia, but not in other target layers. These interneurones have descending projections (Sang, Williamson and Young, 1997), thus the situation in the mouse is strikingly similar to that of the guinea-pig. Further studies of projections, co-localisation of other markers (especially ChAT) and connectivity of 5-HT containing neurones are required before it is possible to conclude that they perform similar roles in all species and regions of gut. SOM immunoreactivity is present in both nerve cell bodies and varicose fibres in many mammalian species. While it commonly occurs in secretomotor neurones which project to the mucosa, it is also present in varicose nerve fibres in the submucous and myenteric ganglia of the human gastrointestinal tract (Keast, Furness and Costa, 1984), presumably reflecting a role in interneuronal pathways. The polarity of these neurones has not been identified and the possibility that it is in ascending interneuronal pathways cannot be excluded. In the rat, SOM is present in varicose nerve endings in both myenteric and submucous ganglia and myenteric interneurones have been shown to have descending

432

INNERVATION OF THE GASTROINTESTINAL TRACT

projections in both the small and large intestine (Ekblad et al., 1987, 1988). A similar situation has been reported in the dog, with many immunoreactive myenteric nerve cell bodies, and some submucous neuronal somata, and many varicose fibres in the ganglia (Daniel et al., 1985). Myenteric neurones have anally-directed axons and do not innervate the smooth muscle layers. However, it seems unlikely that SOM interneurones form the same functional chains in the dog as reported in the guinea-pig, because immunoreactive nerve cell bodies were reported to be rarely surrounded by SOM-immunoreactive varicosities. In both the mouse colon (Heinicke and Kiernan, 1990), and rabbit small intestine (Keast, Furness and Costa, 1987) SOM varicose fibres are present in ganglia but are sparse in the muscle layers. The polarity and coding of these neurones in these species have not yet been studied in detail. In conclusion, SOM is expressed by enteric neurones in many species and regions of gut. At least some of these neurones are likely to function as interneurones and in several species they are likely to be descending interneurones. It would be premature, however, to conclude that all SOM-immunoreactive interneurones perform the same roles in different preparations.

VISCEROFUGAL NEURONES It has been known since the mid twentieth century that there are sympathetic reflex arcs that do not require connectivity with the central nervous system (Kuntz and Saccomanno, 1944). This suggests that there are afferent pathways arising in the gut wall that project and synapse within the prevertebral ganglia. While there are functional collaterals from extrinsic spinal afferent neurones which may mediate such effects, a type of enteric neurone, the viscerofugal neurone, has been shown to contribute to this pathway (Szurszewski and Miller, 1994). Viscerofugal neurones have nerve cell bodies in the myenteric or submucous ganglia and project, via the mesenteric nerves, to the prevertebral ganglia where they synapse with the sympathetic post-ganglionic neurones. In the guinea-pig small intestine, all viscerofugal neurones are located in the myenteric plexus. Their cell bodies tend to be located nearer to the mesenteric border of the intestine than to the anti-mesenteric border (Kuramoto and Furness, 1989) and they increase in number distally. They project to both the coeliac ganglion and the superior mesenteric ganglion (Kuramoto and Furness, 1989; Messenger and Furness, 1992, 1993), but not to the inferior mesenteric ganglion (Messenger and Furness, 1993). The morphology of these neurones was not entirely clear from studies using fast blue as a tracer, but using biotinamide it was shown that they consistently fit in the Dogiel type I class, having a single axon and short lamellar, or occasionally filamentous dendrites (Tassicker et al., 1999). In the small intestine these neurones are all immunoreactive for ChAT, suggesting that they are cholinergic, and are also immunoreactive for VIP and GRP (Mann et al., 1995) and some are also likely to be immunoreactive for NMU (Furness et al., 1989a). However they are not immunoreactive for NOS (Anderson et al., 1995; Mann et al., 1995). Because of their scarcity, the coding of these neurones cannot be directly studied in the myenteric ganglia without prior retrograde labelling, but it is clear that they comprise far fewer than 1% of all myenteric neurones in the guinea-pig small intestine (Costa et al., 1996). From their morphology, it would be expected that viscerofugal neurones in the small intestine

CELLULAR ORGANISATION OF THE MAMMALIAN ENTERIC NERVOUS SYSTEM

433

would have the electrophysiological characteristics of S cells, but no recordings have yet been reported. Viscerofugal neurones in other regions Viscerofugal neurones have been reported in the guinea-pig gastric antrum (but not fundus) and, in the oesophagus, projecting to the coeliac ganglion (Messenger and Furness, 1992). In addition, it has been reported that some nerve cell bodies in myenteric ganglia of the oesophagus can be retrogradely labelled from fine branches of the vagus nerve, but their target is not known (Zagorodnyuk and Brookes, 2000). The number of viscerofugal neurones in the stomach and oesophagus are very less, as in most of the small intestine, however there are many more viscerofugal neurones in the colon (Messenger and Furness, 1992, 1993). Viscerofugal neurone cell bodies in all regions of the guineapig gut are located in the myenteric ganglia. Colonic viscerofugal neurones have slightly different chemical coding from those of the ileum, with many of the neurones being immunoreactive for NOS and/or calbindin (Mann et al., 1995), in addition to VIP and GRP. Whether or not the variability in coding represents multiple functional classes of viscerofugal neurones from the colon, or whether it is just variability within a single class, is not currently clear. There is physiological evidence that GABA may also be released by endings of viscerofugal neurones projecting to the inferior mesenteric ganglion, since the GABAA antagonist, bicucculine, reduces the amplitude of both fast EPSPs of peripheral, but not central origin (Parkman et al., 1993; Stapelfeldt, Parkman and Szurszewski, 1993). The presence of GABA immunoreactivity in viscerofugal nerve cell bodies has not been studied quantitatively with immunohistochemical methods. Viscerofugal neurones in the distal colon, projecting to the inferior mesenteric ganglion have been recorded with intracellular microelectrodes following retrograde labelling and have been shown to correspond to S cells, with Dogiel type I morphological features (Sharkey et al., 1998). All of the neurones were immunoreactive for ChAT (Sharkey et al., 1998). Recorded colonic viscerofugal neurones received fast excitatory synaptic inputs (Sharkey et al., 1998), which arise, at least partly, from descending interneuronal pathways (Lomax, Zhang and Furness, 2000). The presence of fast synaptic inputs onto viscerofugal neurones raises interesting questions about how they might be physiologically activated. It has been reported that distension of a segment of intestine, bathed in Krebs solution with low [Ca2+], high [Mg2+], can activate viscerofugal neurones (Bywater, 1994; Stebbing and Bornstein, 1994). This has been taken as evidence that at least some viscerofugal neurones function as first order primary afferent neurones, since they do not require synaptic inputs to drive distension-evoked firing. However, it is possible that the use of large, rapid distensions, may have activated these neurones in a way that is not normally experienced physiologically. Whether or not such neurones can fire when synaptic transmission is blocked, in response to graded distension with a more physiological time course, is not currently clear. Viscerofugal neurones: other species Viscerofugal neurones from the oesophagus and stomach, projecting into the vagus nerve have been reported in the mouse (Sang and Young, 1998b) and in the rat (Berthoud,

434

INNERVATION OF THE GASTROINTESTINAL TRACT

Figure 10.3 Fields of innervation of enteric neurones. From quantitative data based on retrograde labelling studies it is possible to identify the fields of innervation of myenteric neurones. The outputs of 100 neurones found in a typical small ganglion (in black) are shown for enteric primary afferent neurones (EPANs), dendritic enteric primary afferent neurones (“dendritic EPANs”), ascending interneurones (AINs), excitatory motor neurones to the circular muscle (CM-EMNs), inhibitory motor neurones to the circular muscle (CM-IMNs), excitatory motor neurones to the longitudinal muscle (LM-EMNs) and for descending interneurones. Note that dendritic EPANs and descending interneurones have projections considerably longer than the 30 mm of tissue shown aboral to the ganglion. Numbers in brackets indicate the percentage of cells in each class projecting from the ganglion. (Adapted from Costa et al., 1996; Brookes, 2001).

Jedrzejewska and Powley, 1990), but their targets and roles are not known. There are viscerofugal neurones projecting from the rat stomach to the coeliac ganglion, which are immunoreactive for VIP (Lee et al., 1986) and a GRP or bombesin-like substance (Hamaji et al., 1987). The presence of NOS immunoreactivity in colonic, but not ileal viscerofugal neurones also appears to hold true for the rat (Domoto et al., 1995), however some major

CELLULAR ORGANISATION OF THE MAMMALIAN ENTERIC NERVOUS SYSTEM

435

Figure 10.4 Schematic modular arrangement of the innervation of one region of the small intestine. Myenteric ganglia each contain identical classes of neurones. Neurones innervating the central module and its muscle are shown in black. Note the descending inputs to the circular muscle from descending circular muscle inhibitory motor neurones (CM-IMNs). There are ascending inputs from circular muscle excitatory motor neurones (CM-EMNs) and local inputs to the longitudinal muscle from excitatory motor neurones (LM-EMNs). Descending interneurones (DINs) provide inputs (filled triangles) to other DINs, to CM-IMNs and to LM-EMNs. Ascending interneurones (AINs) make connections with other AINs, with CM-EMNs and LM-EMNs. In addition enteric primary afferent neurones (EPANs) make extensive connections with all other classes of neurones within the same region (shown as open triangles). Thus the gut can be represented quite simply as a series of overlapping, interacting, identical modules.

interspecies differences are apparent. In the dog colon, viscerofugal neurones projecting to the inferior mesenteric ganglia appear to have two types of chemical coding, being either immunoreactive for VIP/DYN/CGRP or for SP/ENK/Bombesin (or GRP) (Li and Masuko, 1997). While viscerofugal neurones are located exclusively in myenteric ganglia in the guinea-pig, and perhaps in other small animals, this is not the case in larger animals. In the pig colon, significant number of viscerofugal neurones projecting to the superior mesenteric ganglion are located in the submucous plexus (Barbiers et al., 1993), as reported in the cat small intestine (Feher, 1982). In the rat colon, it has been reported that axons of viscerofugal neurones project to the inferior mesenteric ganglion via the colonic nerves, but provide a weaker input to the major pelvic ganglia (Luckensmeyer and Keast, 1995). In most of the colon, viscerofugal neurones tended to be clustered near the mesenteric attachment, but where this was absent in the rectum they were evenly distributed around the circumference (Luckensmeyer and Keast, 1995). The viscerofugal neurones from the rectum to the inferior mesenteric ganglion usually contained calbindin and GRP, while those projecting to pelvic ganglia usually contained, in addition, VIP and NOS immunoreactivities (Luckensmeyer and Keast, 1996). A specialised type of viscerofugal neurone has been described which projects from the distal rectum of the rat, directly to the sacral spinal cord (Doerffler-Melly and Neuhuber, 1988), via the dorsal roots. Like other viscerofugal neurones in the distal colon and rectum of the rat, some of

436

INNERVATION OF THE GASTROINTESTINAL TRACT

the rectospinal neurones were immunoreactive for calbindin and VIP, but a second type contained CGRP immunoreactivity (Neuhuber et al., 1993). Whether there are two functionally distinct classes of rectospinal neurones remains to be determined. The major classes of myenteric neurones, and their fields of innervation are summarised in Figure 10.3. This schematically represents the projections of a representative sample of enteric neurones located in a single small myenteric ganglion, and gives a clear impression of the very different lengths and polarities of projections of different classes of myenteric neurones in the guinea-pig small intestine. This data is combined with information about major sources of of synaptic inputs to myenteric neurones in Figure 10.4. This shows a schematic representation of the highly ordered projections of myenteric neurones that subserve motor reflexes in the guinea-pig small intestine.

SUBMUCOSAL AND OTHER MUCOSALLY-PROJECTING NEURONES The submucous plexus of the guinea-pig ileum was extensively studied in the 1980s. Because of its relative simplicity, it was the first part of the enteric nervous system in which a comprehensive account of the different functional classes of neurones was obtained. It was reported that four classes of neurones could be distinguished on the basis of peptide and ChAT immunoreactivity (Furness, Costa and Keast, 1984) with one class of non-cholinergic neurones comprising approximately 45% of all submucous neurones, and three classes of cholinergic neurones accounting for the remainder. This conclusion, based solely on chemical coding, was later supported by retrograde tracing studies (Song et al., 1992). This scheme has withstood the test of time remarkably well, with one minor modification in the last 16 years. The enteric primary afferent neurones of the submucous plexus were described above and will not be considered further here.

SECRETOMOTOR NEURONES VIP-immunoreactive neurones VIP immunoreactive neurones in the submucous plexus of the guinea-pig ileum also contain immunoreactivity for GAL (Furness et al., 1987), NMU (Furness et al., 1989a) and for the opioid peptides dynorphin[1–8],dynorphin[1–1 7], dynorphin B and alpha neo endorphin (Steele and Costa, 1990) but they do not contain ChAT immunoreactivity (Furness, Costa and Keast, 1984). They make extensive projections to the subepithelial plexus of the mucosa and also contact submucous arterioles. These nerve cells have a slight tendency to project aborally within the submucous plexus, but their projections to the mucosa do not show a significant polarity (Song et al., 1992). A small population of myenteric neurones have similar chemical coding and also project to the mucosa (Song, Brookes and Costa, 1991). These may well be “displaced” secretomotor neurones with identical function to their submucous counterparts. The secretomotor effects of VIP have been extensively studied (for review see Keast, 1987) and it is clear that these populations of myenteric and submucous neurones are responsible for much of the neurally-

CELLULAR ORGANISATION OF THE MAMMALIAN ENTERIC NERVOUS SYSTEM

437

mediated non-cholinergic increase in secretion recorded in Ussing chamber studies (Cooke and Reddix, 1994). Electrophysiologically, VIP-immunoreactive secretomotor neurones, have been demonstrated to receive a combination of synaptic inputs, including fast excitatory synaptic potentials, slow excitatory synaptic potentials and slow inhibitory synaptic potentials (IPSPs) (Bornstein and Furness, 1988). It has been shown that the slow IPSPs are largely mediated by noradrenaline, released from the terminals of sympathetic neurones (North and Surprenant, 1985). However, following extrinsic denervation, slow IPSPs still persist (Hirst and McKirdy, 1975), and some slow IPSPs are resistant to adrenoceptor antagonists (Mihara et al., 1987) suggesting that a proportion is of intrinsic origin. It is likely that these are mediated by SOM released from the SOM-immunoreactive, cholinergic descending interneurones described above (Shen and Surprenant, 1993). The fast and slow excitatory synaptic potentials recorded in these, and other classes of submucous neurones, probably arise primarily from enteric neurones located in both the myenteric and submucous plexuses. Removal of the myenteric plexus several days prior to recording diminished the amplitude of fast and slow synaptic inputs that could be electrically stimulated in submucous neurones but did not abolish them (Bornstein, Furness and Costa, 1987). Confirmation that some synaptic potentials arise from submucous neurones was provided by the observation that such potentials persisted when the submucosa was isolated and maintained in organ culture for several days, giving time for the degeneration of severed nerve fibres (Song et al., 1997b). Besides the SOM-immunoreactive descending interneurones, submucous ganglia also receive varicose nerve endings from myenteric interneurones immunoreactive for NOS/VIP (Li, Young and Furness, 1995) and from 5-HT-containing interneurones (Meedeniya et al., 1998). VIP-immunoreactive secretomotor neurones in submucous ganglia, and their myenteric counterparts, are strongly activated by exogenous nitric oxide to produce cyclic guanosine monophosphate (Young et al., 1993b). Some of the SP-immunoreactive varicosities in submucous ganglia also arise from myenteric neurones (Costa et al., 1981), most probably the enteric primary afferent neurones of the myenteric plexus. In addition, it has been shown that some of the enteric primary afferent neurones in the submucous plexus make varicose endings around other nerve cell bodies in submucous ganglia (Evans, Jiang and Surprenant, 1994). Thus, VIP-immunoreactive submucous secretomotor neurones probably receive inhibitory synaptic input from both sympathetic and myenteric neurones, and receive excitatory synaptic inputs from both myenteric and submucous nerve cells. It is unlikely that they receive much direct innervation from the vagus nerve, since at least in the rat, vagal efferent nerve endings are very sparse in the submucous plexus (Holst, Kelly and Powley, 1997). Their electrophysiological characteristics have been described in more detail elsewhere (Bornstein and Furness, 1988; Mihara, 1993). NPY-immunoreactive neurones A class of cholinergic neurone contains immunoreactivity for NPY/CCK/CGRP/NMU/ SOM/GAL (Furness, Costa and Keast, 1984) and dynorphin [1–8] (Steele and Costa, 1990) and makes up between 29 and 33% of all submucous neurones in the guinea-pig small intestine (Song et al., 1992). These neurones give rise to varicose nerve endings in

438

INNERVATION OF THE GASTROINTESTINAL TRACT

the mucosa and are probably the major class of cholinergic secretomotor neurones. Again, they have no significant polarity in their projections, which are no more than 2 mm in the longitudinal axis of the gut. There is a class of cells in the myenteric plexus with similar chemical coding and which project directly to the mucosa (Furness et al., 1985; Song, Brookes and Costa, 1991). Again it seems likely that these are “displaced” cholinergic secretomotor cells which perform a similar role to their submucous counterparts.

VASOMOTOR NEURONES Calretinin-immunoreactive neurones A population of cells in the submucous plexus that was immunoreactive for ChAT but not for NPY, SP or VIP was described in the early 1980s (Furness, Costa and Keast, 1984). Later, it was shown that these neurones were immunoreactive for the calcium binding protein, calretinin (Brookes, Steele and Costa,1991a) and they were estimated to comprise approximately 12% of all submucous neurones. They project to submucous arterioles and are likely to contribute to the cholinergic vasodilation evoked by electrical stimulation of ganglia or single neurones (Neild et al., 1990). In addition, calretinin immunoreactive varicosities are found in the muscularis mucosa and around the base of the mucosal crypts, but do not appear to form endings in submucous ganglia. It is thus possible that calretinin immunoreactive neurones may also contribute to cholinergic secretion and possibly to motor activity of the muscularis mucosa.

Submucous interneurones and primary afferent neurones After these early studies on the chemical coding of submucous neurones, it has been established that there are probably two classes of VIP-immunoreactive cells in the submucous plexus of the guinea-pig small intestine. While the great majority are of the secretomotor type described above, a small proportion, probably fewer than 1%, function as interplexus interneurones (Song, Costa and Brookes, 1998). Retrograde tracers applied to the myenteric plexus of the guinea-pig ileum label submucous cell bodies, including some of the SP and calbindin immunoreactive primary afferent neurones described above (Kirchgessner and Gershon, 1988; Song, Costa and Brookes, 1998). However, they also label a substantially larger number of VIP-immunoreactive neurones. Detailed analysis of single cells suggests that these VIP-immunoreactive cells project directly to the myenteric plexus and do not have collaterals to the mucosa or submucous blood vessels (Song, Costa and Brookes, 1998).

Submucosal and other mucosally projecting neurones: other regions The submucosal plexus of the guinea-pig stomach contains fewer nerve cell bodies than elsewhere in the gastrointestinal tract, with many ganglia in the body of the stomach containing no neuronal somata. It appears that many of the neurones controlling secretion in the stomach are located in the myenteric plexus. It is well established that GRP plays an

CELLULAR ORGANISATION OF THE MAMMALIAN ENTERIC NERVOUS SYSTEM

439

important role, along with acetylcholine, in the neuronal control of gastric acid secretion. A GRP-like peptide has been immunohistochemically localised in the antral mucosa of the guinea-pig stomach (Miller, Furness and Costa, 1989). Retrograde labelling studies have recently revealed that GRP-containing neurones to the mucosa are mostly cholinergic and many contain NPY immunoreactivity (Pfannkuche et al., 2000). A separate class of neurones appears to innervate the muscularis mucosa, with immunoreactivity for ChAT, GRP and ENK, sometimes with SP immunoreactivity. Interestingly, cholinergic secretomotor pathways in the gastric corpus have a marked orally-directed polarity in their projections, unlike the situation in the small intestine. Another difference from the small intestine, is that the mucosa of the guinea-pig corpus is innervated by nerve fibres immunoreactive for NOS (Furness et al., 1994), which arise from nerve cell bodies in the myenteric plexus with an aborally polarised projections (Reiche and Schemann, 1998; Reiche and Schemann, 1999). Another significant difference is the colocalisation of VIP and NPY within the same nerve cells, some of which are cholinergic (Reiche and Schemann, 1999). Very few neurones projecting to the gastric mucosa contained immunoreactivity for calretinin, 5-HT or SOM. Some neurones were ENK-immunoreactive, but these have been shown to project primarily to the muscle layers, rather than to the mucosa (Pfannkuche et al., 1998a). The mucosa of the gastric fundus of the guinea-pig has also been demonstrated to be innervated by myenteric neurones with polarised cholinergic and nitrergic projections (Pfannkuche et al., 1998b). In both the corpus and fundus, cholinergic neurones comprise over 70% of all retrogradely labelled cells, compared to about 48% in the guinea-pig small intestine (Song et al., 1992). Further distally, in the proximal colon, cholinergic innervation makes up an even smaller proportion (23%) of retrogradely labelled neurones (Neunlist et al., 1998), with 32% being cholinergic in the distal colon (Neunlist and Schemann, 1998). The non-cholinergic innervation is by VIP-immunoreactive nerve cell bodies, as in the small intestine. However, these secretomotor neurones typically have a marked polarity to their projections in both proximal and distal colon (Neunlist et al., 1998; Neunlist and Schemann, 1998). This has been shown to have a functional correlate in the proximal colon. Changes in short circuit current evoked by electrical stimulation of descending pathways were more sensitive to a VIP antagonist than were ascending pathways. Ascending pathways were more sensitive to atropine than were descending pathways. This is clear evidence that functionally, the cholinergic and non-cholinergic secretomotor pathways are differentially polarised. At present it is neither clear how the two pathways interact nor what the physiological significance of the polarisation is. Throughout the colon and rectum, significant number of nerve cell bodies in the submucous ganglia were immunoreactive for NOS (Furness et al., 1994), which was not the case in the small intestine (Costa et al., 1992). SUBMUCOSAL AND OTHER MUCOSALLY-PROJECTING NEURONES: OTHER SPECIES The innervation of the gastric mucosa has been extensively studied in the rat and it is clear that many of the classes of nerve fibres bear close resemblance to those of the guinea-pig (Ekblad, Mei and Sundler, 2000). There are numerous cholinergic nerve fibres and nerve fibres containing VIP/NPY and sometimes PACAP and NOS, in addition to the extrinsic

440

INNERVATION OF THE GASTROINTESTINAL TRACT

innervation by spinal afferent neurones which play a prominent role in mucosal protection (see chapter by Peter Holzer in this volume). Some differences are apparent when compared to the guinea-pig. For example, much of the GRP immunoreactivity is present in VIP-containing nerve fibres and at least some of these are cholinergic (Ekblad, Mei and Sundler, 2000). As in the guinea-pig, ENK and SOM-immunoreactive nerve fibres are sparse in the lamina propria of the rat stomach. A major difference exists in the gross morphology of the submucous plexus in the small and large intestine of the guinea-pig and rat, and that of larger species such as pig and human. In smaller animals the submucous plexus appears to form essentially a single network, with ganglia all in approximately the same plane, although a plexus of nerve fibres at the submucosal border of the circular muscle has been described (Christensen and Rick, 1987). In contrast, in larger animals, the submucous plexus can be seen to be divided into at least two distinct layers. There is an inner submucous plexus (Meissner’s plexus) and an outer submucous plexus (Schabadasch’s or Henle’s plexus) in the pig small intestine (Timmermans et al., 1990), horse jejunum (Pearson, 1994) and human colon (Hoyle and Burnstock, 1989). At least in humans, there is also an intermediate submucous plexus. It is now clear that the proportions of different types of neurones, distinguished by their size (Hoyle and Burnstock, 1989) or chemical coding vary between the different components in the pig small intestine (Timmermans et al., 1990) and in human sigmoid colon (Crowe et al., 1992). Intracellular recordings from randomly impaled submucous neurones of the pig small intestine also showed consistent differences in the electrophysiological types of cells between the inner and outer plexuses (Thomsen et al., 1997). It seems that the outer submucous plexus (Schabadasch’s plexus) is more similar to the myenteric plexus than is Meissner’s plexus, since it contains markers such as SOM, 5-HT and ENK that are not present in the inner plexus. This is consistent with observations that some of the nerve cell bodies in the outer submucous plexus of the dog intestine (Furness et al., 1990a) and human intestine (Timmermans et al., 1994) project to the circular muscle layer. Retrograde tracing studies have confirmed that this is also the case for the human colon (Porter et al., 1999), since motor neurones could be retrogradely labelled by tracers applied to the circular muscle. In contrast to this, in the guinea-pig, all varicose nerve fibres in the circular muscle disappear following a myectomy, demonstrating that all motor neurones have their cell bodies in the myenteric plexus (Wilson et al., 1987). Quite apart from this major structural difference, the combinations and proportions of neurochemical markers present in submucous neurones have been shown to vary widely between species. In the rat small intestine, VIP and NPY co-exist in submucous neurones, unlike the guinea-pig ileum (Ekblad et al., 1987, 1988; Browning and Lees, 1994) and make up about 40% of all submucous neurones. In addition, it has been suggested that some of these neurones may project to the circular muscle since some varicose nerve fibres with this coding persist in the circular muscle following a myectomy operation (Ekblad et al., 1987). In addition, most VIP/NPY neurones are also immunoreactive for calbindin and for the NK3 receptor (Mann, Furness and Southwell, 1999). Nearly all submucous neurones in the rat small intestine were reported to contain ChAT immunoreactivity (Mann, Furness and Southwell, 1999), so the distinction between cholinergic and non-cholinergic secretomotor neurones in the small and large

CELLULAR ORGANISATION OF THE MAMMALIAN ENTERIC NERVOUS SYSTEM

441

intestines of the guinea-pig does not appear to apply in this species. There are also CGRP immunoreactive nerve cell bodies in the submucous ganglia which contain SOM immunoreactivity, making up approximately 18% of all cell bodies and an additional population of CGRP immunoreactive neurones without SOM, comprising approximately 25% of all cells. Some of these CGRP immunoreactive neurones also contained SP immunoreactivity (Pataky, Curtis and Buchan, 1990). In both small and large intestine, the effects of crushing the gut wall revealed polarised projections of VIP immunoreactive neurones, with a predominant oral projection, which contrasts with the slight aboral projection of VIP neurones in the guinea-pig small intestine (Song et al., 1992) and the marked aboral polarity in the proximal and distal colon (Neunlist et al., 1998; Neunlist and Schemann, 1998). In the mouse small and large intestine, the detailed chemical coding of submucous neurones has not been characterised in detail. However it has been reported that approximately 40% of all submucous neurones in the small intestine, and 20% of submucous neurones in the large intestine are immunoreactive for ChAT, and hence cholinergic. This is more similar to the situation in the guinea-pig than in the rat small intestine, but the subdivision of the classes remains to be determined. In the dog small intestine, populations of submucous neurones immunoreactive for VIP, NPY, SOM, ENK, GRP (Daniel et al., 1985) and CGRP (Sternini, de Giorgio and Furness, 1992) have been described, but the patterns of coexistence have not been worked out in detail.

SUMMARY ON ENTERIC NEURONE CLASSIFICATION Detailed studies on the enteric nervous system of the guinea-pig small intestine have provided a strong foundation for analysis of the neuronal circuitry that controls motility, secretion and blood flow of this preparation. This data is summarised in Figure 10.5. Enteric neurones have been classified using many different criteria, including their morphology, electrophysiological characteristics and ultrastructural features. However, in most cases these attempts have not led, on their own, to the identification of nonoverlapping, functionally identified classes of cells. The development of the concept of chemical coding provided a breakthrough in distinguishing types of neurones in both the peripheral and central nervous systems. When combined with lesions to pathways, or with neuronal tracing techniques, chemical coding is capable of distinguishing the multiple classes of neurones that have been deduced to exist from functional studies. Chemical coding provides sufficient degrees of freedom that neurones belonging to different functional classes are unlikely to be inappropriately lumped together. In contrast, electrophysiological classification has generally only been able to distinguish 2–4 types of cells in any preparation of enteric neurones. We suspect that damage and/or spontaneous release of modulators confound classification attempts. Careful analysis of somadendritic morphology has allowed multiple types of cells to be distinguished (Stach, 1989), but it is not clear whether each of these types is homogenous, nor what the functional role of each morphological type may be. In addition, the functional significance of small differences in dendritic topology is hard to interpret. This is especially true when dendrites are generally so short that cell bodies are likely to function as uni-compartmental

442

INNERVATION OF THE GASTROINTESTINAL TRACT

Figure 10.5 Schematic diagram of the classes of cells in the guinea-pig ileum. The diagram shows a schematic longitudinal section through the gut wall. Non-cholinergic neurones are shown with a grey tint. Enteric primary afferent neurones (EPANs) are shown in white and are present in both myenteric and submucous ganglia. Excitatory motor neurones to the circular muscle (CM-EMNs) and longitudinal muscle (LM-EMNs) are shown with rising oblique hatching. Inhibitory motor neurones to the circular muscle (CM-IMNs) and to the longitudinal muscle (LM-IMNs) are shown with crosshatching on a grey tint. Ascending interneurones (AIN are shown with horizontal hatching and descending interneurones (DINs) with vertical hatching. Four different types of descending interneurones are distinguishable, namely those immunoreactive for NOS (non cholinergic), and for VIP, 5-HT and SOM. Viscerofugal neurones are shown cross hatched and project out of the gut wall towards prevertebral ganglia. Secretomotor neurones are shown as grey tint (for VIP-immunoreactive, non cholinergic secretomotor neurones in both myenteric and submucous ganglia) and cholinergic secretomotor neurones as crosshatched (from both myenteric and submucous ganglia). Lastly there is a submucous neurones (horizontal hatching, grey tint) which is a VIP immunoreactive inter-plexus interneurone and a cholinergic vasomotor submucous neurone (diagonal crosshatching).

electrophysiological entities. However, the distinction between unipolar and multipolar neurones, may have profound physiological implications for transduction of incoming signals. Multiple labelling immunohistochemistry, combined with tract tracing or lesion studies, has proven to be the most powerful methodology for distinguishing and identifying functional classes of enteric neurones. In addition, it is compatible with other types of classification; electrophysiological and morphological data can be readily integrated into the classification scheme. This is particularly true of the guinea-pig ileum where an enormous number of electrophysiological studies have been carried out over the last 25 years. It is becoming increasingly true of other regions of the guinea-pig gastrointestinal tract and for preparations from other species, in particular the pig and the human. We expect to see more data from rat and mouse preparations in the near future, given the increasing availability and interest in transgenic and mutant animals of these species.

CELLULAR ORGANISATION OF THE MAMMALIAN ENTERIC NERVOUS SYSTEM

443

FUNCTIONAL SIGNIFICANCE OF CHEMICAL CODING In each region of gut studied, in each species that has been investigated, different functional classes of cells can be distinguished on the basis of their chemical coding, projections, soma-dendritic morphology and electrophysiological characteristics. What is more, under the controlled conditions in which laboratory animals are studied, the combinations of characteristics remain quite consistent from one individual to the next. The concept of chemical coding provides a powerful methodology to identify and distinguish different classes of enteric neurones. The nature of the markers that comprise the coding is of little relevance to this aim. However, immunohistochemical localisation of neurochemicals also provides information about the distribution of many neuroactive substances, from which deductions about the physiology of the nervous system can be made. A good example of this is the presence of NOS immunoreactivity, but not ChAT immunoreactivity in a subset of neurones retrogradely labelled from the circular muscle of the human colon (Porter et al., 1997), which allowed their identification as inhibitory motor neurones. Likewise the presence of ChAT in any class of interneurones is strongly suggestive that they use acetycholine as a primary transmitter. The question arises as to how far is reasonable to infer functional significance for the molecules localised immunohistochemically. Functional studies carried out in a wide range of preparations suggest that in every mammalian species studied to date, there are a number of well preserved functional classes of enteric neurons. These include excitatory and inhibitory motor neurons to the longitudinal and circular muscle layer, secretomotor neurons, enteric primary afferent neurons etc. Absence of one of these functional classes of neurones would be expected to result in severe malfunction of the digestive system, with clear adaptive disadvantage. It follows that molecules that are essential for the function of these classes of neurones are likely to be preserved throughout evolution and will be found to be common to different species. The expression of molecules such as acetylcholine and its associated synthetic and handling enzymes must be tightly controlled in enteric neurones. In contrast to this, there are other molecules, including many neuropeptides, which have rather more subtle modes of action. Often, these molecules are described as “neuromodulators”. In most cases their absence would be expected to have less dramatic functional consequences than blockade of cholinergic mechanisms. This suggests then that there is a hierarchy of functional significance of the molecules used for chemical coding of neurones. In other words, neurochemical markers are not all equally physiologically important. We can illustrate this point by comparing five well characterised neurochemical markers. ACETYLCHOLINE AND NITRIC OXIDE AS PRIMARY MOTOR NEURONE TRANSMITTERS Numerous functional studies over the last century have established the fundamental role of acetylcholine as a major ganglionic transmitter and as a transmitter of several classes of effector neurones (Furness and Costa, 1987). The distribution of ChAT in different classes of enteric neurones reflects this role. ChAT immunoreactivity is present in most classes of interneurones, in most presumed primary afferent neurones and in excitatory motor neurones, in all preparations wherever it has been tested. Similarly, nitric oxide is now

444

INNERVATION OF THE GASTROINTESTINAL TRACT

established as an important inhibitory neurotransmitter to the muscularis externa (Sanders and Ward, 1992; Brookes, 1993) and is quite consistently present in inhibitory motor neurones in all preparations studied to date. Mice with a knockout of the nNOS gene show alterations in gastrointestinal motility, particularly in gastric emptying, consistent with a major role for NO as an inhibitory transmitter (Mashimo and Goyal, 1999; Mashimo, Kjellin and Goyal, 2000). Thus ChAT and NOS enzymes are essential for the major transmitters of particular classes of enteric motor neurones and appear to show consistent patterns of expression. Another consistent feature is the polarised projection of inhibitory and excitatory motor neurones (Brookes et al., 1991, 1996, 1998; Porter et al., 1997; Pfannkuche et al., 1998b; Yuan and Brookes, 1999; Michel, Reiche and Schemann, 2000). The molecular basis of this polarity is not currently understood. This polarity probably has fundamental significance for the motility (see Chapter 1 of this volume). We would expect the molecules that underlie such polarisation to be consistently expressed across species. It is enlightening to compare the role of ChAT and NOS as primary markers for functionally significant enteric neurotransmitters, with some of the most thoroughly characterised neuropeptides localised in enteric neurones. Of these VIP and SP are currently best understood and seem to have a prominent role as co-transmitters of enteric inhibitory and excitatory motorneurons respectively. VIP AS A NEUROTRANSMITTER Although inhibitory motor neurons often contain VIP, there is greater variability in its expression in motor neurones than NOS and ChAT. For example, in the human colon, only a small proportion of inhibitory and excitatory motor neurones retrogradely labelled from the circular muscle contain VIP or SP, respectively (Wattchow et al., 1997) whereas nearly all contain either ChAT or NOS (Porter et al., 1997). In the guinea-pig ileum, nearly all circular muscle motor neurones contain either VIP or SP immunoreactivity (Llewellyn-Smith et al., 1988; Brookes, Steele and Costa, 1991b) and either NOS or ChAT (Costa et al., 1996). Many studies have shown that exogenous VIP causes inhibition and relaxation of gastrointestinal smooth muscle (see Chapter 6 of this volume) and there is good evidence that it can be released by enteric neurones. However, there are many fewer studies that show that specific antagonists (or neutralising antisera) block the effects of endogenous VIP. Most of the studies to date have been on in vitro preparations with powerful electrical stimulation parameters, often at relatively high frequency, and in the presence of drugs to block the action of other neurotransmitters which might otherwise occlude the effects of endogenous VIP. Under these conditions, VIP has been shown to be released and to have effects on smooth muscle that mimic endogenous inhibitory neurotransmitters, for example in the rat gastric fundus (Li and Rand, 1990), in the canine muscularis mucosa (Angel et al., 1983) and in the guinea-pig ileum (Crist, He and Goyal, 1992). However, in some cases, the effects of exogenous VIP may be mediated via release of other transmitters, particularly nitric oxide (Mashimo et al., 1996). Despite considerable study it is not yet clear whether VIP plays any role in normal physiology, in vivo, nor whether it may be preferentially released under particular pathophysiological conditions. To be fair, it should be pointed out that currently available

CELLULAR ORGANISATION OF THE MAMMALIAN ENTERIC NERVOUS SYSTEM

445

VIP antagonists are limited to a few peptide antagonists, with variable potency. But it is reasonable to conclude that, overall, the role of VIP as an inhibitory transmitter to the smooth muscle has been considerably more difficult to demonstrate than that of nitric oxide. VIP appears to contribute rather less to the physiological control of smooth muscle tone than does NO. SUBSTANCE P (OR TACHYKININS) AS NEUROTRANSMITTERS The same general point also appears to apply to the roles of tachykinins when compared with the dominant excitatory transmitter, acetylcholine. Several generations of highly specific, potent tachykinin antagonists are now available for assessing the roles of endogenous tachykinins in gastrointestinal function. There is extensive evidence supporting a role for tachykinins as excitatory neurotransmitters to smooth muscle (see review (Holzer Petsche, 1995; Holzer and Holzer-Petsche, 1997). This is largely based on studies carried out in vitro, under conditions where spontaneous activity is minimised, with intense physiological or electrical stimulation, and usually in the presence of cholinergic antagonists. In most reports in the literature, conditions have been carefully contrived to detect effects of endogenous tachykinins, avoiding their occlusion by the more prominent cholinergic transmission. There has been one report that antagonists to NK1 and NK3 receptors reduce distension-evoked reflex activity in the guinea-pig small intestine without pretreatment with cholinergic agonists (Johnson, Bornstein and Burcher, 1998; see also Chapter 6 of this volume). Surprisingly, there are still few studies documenting the effects of these antagonists on motor activity in vitro, without prior cholinergic blockade, and even fewer on their effects on motility in vivo. It has been shown that gastric emptying in the rat is slowed by a peptidergic SP antagonist (Holzer, Holzer-Petsche and Leander, 1986). It would seem that at present, tachykinins are involved in neurotransmission in reflex pathways and in neuromuscular transmission, but that their role is considerably less obvious than that of acetylcholine. OPIOID PEPTIDES AS ENTERIC TRANSMITTERS As described above, opioid peptides are abundantly expressed by enteric nerve cells (Furness, Costa and Miller, 1983; Steele and Costa, 1990). ENK is present in both excitatory and inhibitory motor neurons. Yet it has no direct action on the target muscle and thus play no major role in motor transmission in the guinea-pig ileum. However, it is able to inhibit transmitter release from motor neurons, acting in a neuromodulatory fashion, particularly during prolonged or high frequency electrical stimulation (Puig et al., 1977). It is difficult to demonstrate that opioids play any physiological role in normal motor activity. For example blockade of opioid receptors has relatively little effect on peristalsis in vitro (Kromer, 1990). It is only when the gut is forced to work against a greater resistance, as may occur in intestinal obstruction, that opioid appear to play a role as a “brake” by inhibiting the neural circuit underlying peristalsis (Waterman, Costa and Tonini, 1992). It would seem then that endogenous opioids may play little role in normal control of motility, but have an important neuromodulatory role under specific, perhaps pathophysiological conditions.

446

INNERVATION OF THE GASTROINTESTINAL TRACT

From this very brief account it is clear that the contribution of neurochemicals to functioning of the enteric nervous system can be arranged in a hierarchy. Transmitters such as acetylcholine and nitric oxide play a crucial role in moment-to-moment control. VIP and SP act as neurotransmitters, but play a less prominent role in normal neuromuscular transmission, perhaps having additive effects to the primary transmitters during periods of strong activation of motor neurones. In contrast, opioid peptides probably play a modulatory role that only becomes significant under even more extreme, perhaps pathophysiological conditions. For other peptides, such as GAL, NPY, SOM, CCK, GRP, GABA etc, the roles are even less clear and may be even more subtle. Lastly, there are a large number of other molecules such as the calcium-binding proteins, calbindin and calretinin and some structural proteins such as NFP triplet, which have been useful for distinguishing classes of neurones, but whose role is presently entirely unknown. In the future, we can predict that the roles will be identified for some of these molecules in enteric physiology. We have to be willing to entertain the possibility, however, that the presence of some molecules in enteric neurones may have no functional consequences. Just because a particular peptide or calcium binding protein is expressed in a particular enteric interneurone, does not indicate that it necessarily contributes to the behaviour of the neuronal circuitry. It is reasonable to assume that every peptide and protein in the body contributes in some way to the survival or “fitness” of the animal. However, there is no reason to believe that everywhere that peptide or protein is expressed, it must play an equally crucial role. There is a great deal of evidence about the effects of exogenous neuropeptides, but very much less about the roles of the endogenous stores of the same molecules. Inhibitory motor neurones in the guinea-pig ileum also use ATP as a transmitter substance (Crist, He and Goyal, 1992; see also Chapter by Hoyle et al., in this volume). The unavailability of a robust molecular marker for ATP-using neurones has so far prevented this being incorporated into chemical coding. The relative roles of the inhibitory mechanisms used by enteric inhibitory motor neurones varies across species and parts of the gut (Costa, Furness and Gibbins, 1986). This may explain some of the variability in expression of the corresponding molecular markers. The multiplicity of mechanisms by which enteric inhibitory motorneurones relax the muscle is also demonstrated by the nNOS knockout mice. WHY IS IT ALL SO VARIABLE? Equivalent neurones in different regions of gut, or in the same region of gut in different species, show enormous variability. We can no longer expect that the coding of a particular class of cells in one species would be identical to that of another species. In some ways that has been a considerable disappointment to workers in this field. It means that each preparation has to be studied de novo. It may become apparent, in future decades when many species have been characterised in detail, that some principles are universal. We can look forward to that day but we are not there yet. On the bright side, the techniques to characterise enteric neurones in each species are now tried and tested. Faced with the variability in the chemical coding between functionally similar neurones in different preparations of gut, the question arises as to how should we interpret these differences. There has been a great deal of speculation about the significance of variability

CELLULAR ORGANISATION OF THE MAMMALIAN ENTERIC NERVOUS SYSTEM

447

in chemical coding between neurones with similar functions in different preparations. The explanations themselves can be placed on a continuum. At one end there are explanations which dismiss the differences between preparations as being of little functional significance and reflecting “genetic drift”, “redundancy” or “limitations in the control of gene expression”. At the other end of the scale, explanations assume that “every difference must have a functional significance. If it is not yet obvious, then we just need to look harder, or in a different place”. No doubt, the truth lies somewhere between these extremes. If the idea is accepted that there is a hierarchy of importance of enteric neurochemicals, then the variability becomes more explicable. It would be expected that the molecules that are crucial for the effective action of a class of neurones should show little variability. Thus ChAT and NOS appear to be the most prominent excitatory and inhibitory neurotransmitters to the smooth muscle layers and appear to be the most consistent markers found in these neurones. VIP and SP also appear to play a role in neuromuscular transmission, albeit less prominent than that of acetylcholine and nitric oxide. Accordingly, there is more variability in the expression of these peptides in enteric motor neurones. The pure modulators, such as ENK and many of the other peptides (which appear to have no role in direct neuromuscular transmission, in most regions of gut) show even more variability in expression, being sometimes in excitatory motor neurones, sometimes in inhibitory motor neurones. At the far end of the scale, molecules like the calcium binding proteins appear to be present in different populations of cells in every preparation studied and often are not found consistently even all members of the same functional class within a single preparation. In the guinea-pig ileum, for example calretinin is expressed by one class of ascending interneurone (Brookes, Steele and Costa, 1991a) and by the majority, but not all, cholinergic motor neurones to the longitudinal muscle (Brookes et al., 1992). In contrast, in the mouse small intestine, calretinin is present in circular muscle motor neurones (Sang and Young, 1996), and in the human colon it is present in descending interneuronal pathways (Wattchow et al., 1997). Calbindin shows even greater variability in its expression. In the guinea-pig ileum it is found in about 80% of Dogiel type II neurones (Furness et al., 1988; Iyer et al., 1988). However, there is no evidence that the primary afferent neurones with and without calbindin have different physiology, projections or roles. In the guinea-pig colon, calbindin is also present in some descending interneurones with Dogiel type I morphology and in some viscerofugal neurones (Mann et al., 1995). This points to a large degree of variability in the patterns of expression of calcium binding proteins in particular. A factor that may potentially contribute to variability in expression is whether alternative or redundant mechanisms are available. There are no alternatives for acetylcholine and nitric oxide as primary motor neurone transmitters. However, PACAP, ATP and PHI can all be released by enteric motor neurones and are capable of relaxing gastrointestinal smooth muscle. Thus they represent molecular equivalents to VIP as a secondary inhibitory neurotransmitter. In some specialised areas, they may rival the importance of nitric oxide as the primary transmitter, for example in the guinea-pig taenia coli (Selemidis, Satchell and Cocks, 1997; Bartho, Lenard and Szigeti, 1998). Potential enteric neuromodulators which cause prejunctional or presynaptic inhibition, include NPY (Schemann and Tamura, 1992; Browning and Lees, 2000), GABA (Cherubini and North, 1984), 5-HT (North et al., 1980), dynorphin (Cherubini and North, 1985), adenosine (Christofi and

448

INNERVATION OF THE GASTROINTESTINAL TRACT

Wood, 1993b) and nitric oxide (when expressed in interneurones) (Tamura, Schemann and Wood, 1993). It is conceivable that any of these molecules may be able to substitute for ENKs in motor neurones and interneurones. Another modulatory effect of exogenous ENK is to hyperpolarise enteric nerve cell bodies (North, Katayama and Williams, 1979). A number of other enteric neuromodulators have the same effect, including 5-HT (Johnson, Katayama and North, 1980), SOM (Katayama and North, 1980), GAL (Tamura et al., 1988) and adenosine (Palmer, Wood and Zafirov, 1987). Potentially these substances may be functional alternatives to ENK acting at post-synaptic sites. Other neuromodulators cause slow excitation of enteric neurones. These include 5-HT (Wood and Mayer, 1978), tachykinins (Katayama and North, 1978), GRP and VIP (Zafirov et al., 1985a), CCK (Nemeth, Zafirov and Wood, 1985), CGRP (Palmer et al., 1986) and PACAP (Christofi and Wood, 1993a). Again, it is not inconceivable that these should be able to substitute functionally for one another in enteric neurones. Thus, even if neuromodulatory influences are important for the appropriate functioning of enteric neuronal circuits, there are multiple molecular mechanisms that can achieve it. This being the case, variability in expression may have little functional consequence. A HYPOTHESIS ABOUT VARIABILITY IN CHEMICAL CODING We can summarise these observations into a hypothesis that may explain some of the variability in chemical coding of functionally similar enteric neurones. Molecules which subserve the most essential functions, will show the least variability in expression. Similarly, molecules that play less essential roles in normal physiology may show more variability in patterns of expression. The hypothesis may be stated another way. Molecules which show the greatest degree of variability between preparations probably play the most subtle modulatory roles. In addition, the more molecular alternatives there are for a function, the more variability can be expected in their expression. Lastly, we should be willing to countenance the possibility that some molecules expressed by enteric neurones may have no role whatsoever in these particular cells. If a group of neurochemicals really have no functional role, we would expect a high degree of variability in their patterns of expression. This last point is unfortunately untestable, as would require an absence of evidence to be proven true, but perhaps it would be a reasonable starting point for discussions about the roles of neurochemicals. The hypothesis stated above, that “less important” neurochemicals show the greatest variability in expression (between functionally similar neurones in different preparations), has some explanatory power. However it also has a number of weaknesses. First, the decision about the relative importance of neurochemicals is very subjective.The lack of highly specific antagonists for many neurochemicals and peptides has led to a lack of data about their roles. One cannot assess the “importance” of a molecule without appropriate investigations. The aphorism “absence of evidence is not evidence of absence” is highly pertinent here. Lastly, we have to bear in mind that neurochemicals may play vitally important roles without being involved in neurotransmission and neuromodulation. Some neuropeptides may function as growth factors, as chemoattractants for white blood cells or as triggers for immunological responses. These roles may be difficult to detect if research is aimed at understanding moment-to-moment control of gut motility, secretion or blood flow.

CELLULAR ORGANISATION OF THE MAMMALIAN ENTERIC NERVOUS SYSTEM

449

ANIMAL MODELS FOR THE HUMAN ENTERIC NERVOUS SYSTEM It is clear from this review that there are considerable differences in the chemical coding, morphology, electrophysiological characteristics and every other feature used to classify enteric neurones in different regions of gut of the same species and between the same regions of gut in different species. This has led to some debate about the relative merits of different preparations. There is often an underlying assumption by many investigators that animal models have value only to the extent that they cast light directly on how the human enteric nervous system works. The question of which animal species is the best choice as a model for human diseases is an important one. It is clear from the variability surveyed in this review that it will not be possible to use any animal model to predict chemical coding of human neurones. For this aim it will be necessary to study human tissue directly. Animal models will be most valuable for understanding fundamental principles, particularly of disease states. The choice of model must take into account many factors including cost of animals, availability, size, diet, amount of connective tissue, thickness of tissue, the amount of existing data, availability of mutants or transgenic animals, well established disease models, etc. No single species combines all of these features. We desperately need more data from preparations other than the guinea-pig small intestine, in order to understand enteric neuronal function. However, it would be wasteful, in the extreme, to advocate turning our back on the most thoroughly characterised preparation. The guineapig small intestine will continue to provide an ideal preparation for understanding some aspects of enteric neuronal function and for developing new technical approaches. Other animal models will be more appropriate to answer other questions about gut physiology and pathophysiology.

LIMITATIONS OF CHEMICAL CODING Although chemical coding, in combination with other techniques, is a powerful methodology for characterising the structural organisation of the enteric nervous system; it has to be determined for each preparation. While the methodology to achieve this is available, it is not a simple task. Each week a new antiserum or histochemical marker becomes available, giving rise to a perception of increasing precision in our account of classes of neurones. However, characterising the presence of a new marker in some enteric neurones is a far cry from establishing the full chemical coding. It is only by being able to assign all neurones to functional classes that the full characterisation of a new marker becomes possible. Another major restriction on the power of chemical coding is the limited number of markers that can be simultaneously localised in enteric nerve cells. This becomes a serious problem when attempting to count nerve cells showing particular combinations of markers. Statistical sampling errors accumulate with the number of different combinations that have to be tested. One last limitation, mentioned above, is that the distribution of many functionally significant molecules cannot yet be determined. The purines, ATP and adenosine are good examples of this. It has become increasingly obvious that ATP, in particular, plays an important role in neuro-neuronal transmission in selected pathways in the enteric

450

INNERVATION OF THE GASTROINTESTINAL TRACT

nervous system (Galligan and Bertrand, 1994; Johnson et al., 1999; LePard and Galligan, 1999; Bian, Bertrand and Bornstein, 2000). However, at present there are no proven neurochemical markers for neurones that utilise ATP as a neurotransmitter. The same applies to adenosine. PLASTICITY OF THE ENTERIC NERVOUS SYSTEM It is now clear that the enteric nervous system has remarkable plasticity and can dramatically change its chemical coding under a wide range of circumstances. There are some problems with the interpretation of such changes. A major limitation of chemical coding studies is that they rely on a technique which determines the presence of markers at a single moment in time. Thus, the use of chemical coding to distinguish different types of neurones is complicated if the expression of markers changes over time. Another major limitation of detecting a neuronal marker histochemically is that the technique, per se, does not distinguish between changes in expression levels of a marker and the selective loss of cells. A few examples of plasticity are given below. For more thorough coverage, the reader is referred to recent excellent reviews (Collins, 1996; Giaroni et al., 1999). Developmental plasticity Neurones in the enteric nervous change in a number of ways during development. While catecholamine-synthesising neurones are rare in the adult enteric nervous system, a subset of nerve cells transiently expresses markers of catecholamine synthesis such as tyrosine hydroxylase during embryonic development. These neurones subsequently go on to become, amongst others, the 5-HT immunoreactive interneurones in adult life (Blaugrund et al., 1996). It has recently been reported that in the mouse small intestine, NOS-immunoreactive neurones are very rare in the submucous plexus. However, during embryonic development, nearly half of these neurones express NOS (Young and Ciampoli, 1998) but this reduces to fewer than 3% in the adult. The number of enteric neurones also show marked changes throughout development, with a decline in total number with age (Gabella, 1989). A comparable decline in cell density has also been reported in the human colon, with age (Gomes, Desouza and Liberti, 1997). Whether this involves the selective loss of particular classes of neurones, or a decrease in all classes is not currently clear. However, in the human small intestine, the overall proportions of neurones showing immunoreactivity for NOS and calretinin both increase substantially with age (Belai and Burnstock, 1999). This could either represent an increased expression of these markers by enteric neurones (i.e. a change in chemical coding) or a selective loss of neurones that lack either marker. Plasticity in response to denervation Extrinsic denervation of the gut wall causes a number of changes in enteric neurones. In the normal guinea-pig ileum, submucosal blood vessels are innervated, in part, by collaterals of spinal afferent neurones which are immunoreactive for both SP and CGRP. Following extrinsic denervation, nerve fibres immunoreactive for SP without CGRP innervated the

CELLULAR ORGANISATION OF THE MAMMALIAN ENTERIC NERVOUS SYSTEM

451

blood vessels. The appearance of these fibres coincided with the development of abnormal vasodilator responses which were largely insensitive to muscarinic blockade and which were apparently mediated by SP (Galligan et al., 1990a). The appearance of the new fibres is probably due to sprouting of new axon collaterals by myenteric neurones which do not normally innervate submucosal blood vessels, as it was prevented by removal of the underlying myenteric plexus (Jiang and Surprenant, 1992). It has been reported that extrinsic denervation of guinea-pig small intestine also increases the number of myenteric neurones which show reactivity for NADPH diaphorase (Yunker and Galligan, 1994) and for NOS immunoreactivity, but not VIP immunoreactivity (Yunker and Galligan, 1998). There was a concomitant increase in the amplitude of nerve-mediated relaxations of the longitudinal muscle layer which could be blocked by a NOS antagonist (Yunker and Galligan, 1998), but not of relaxations mediated by a nitric oxide donor. The effects of denervation were mimicked by capsaicin treatment but not by 6-hydroxydopamine treatment, suggesting that the loss of spinal afferent neurones was the trigger for NOS upregulation. In the rat jejunum, extrinsic denervation by removal of the coeliac/superior mesenteric ganglion also caused increases in NADPH diaphorase reactivity and NOS immunoreactivity. However this was mimicked by 6-hydroxydopamine and not by capsaicin, suggesting that it was the loss of sympathetic innervation which altered NOS expression (Nakao et al., 1998). Whether the discrepancy between the two studies represents a true interspecies difference, or is due to different methodologies, is not yet clear. Plasticity associated with disease states A number of studies have characterised changes in the enteric nervous system following experimentally induced diabetes in rats, by streptozotocin treatment. For example, VIP immunoreactivity in myenteric nerve cell bodies of the small and large intestine increased in intensity and VIP-immunoreactive nerve fibres changed in appearance (Belai et al., 1985). No changes in SP-immunoreactive fibres were noted in this study. In contrast, CGRP immunoreactive nerve cell bodies and varicose nerve fibres decreased in myenteric ganglia in streptozotocin-treated rats (Belai and Burnstock, 1987). Experimentally induced diabetes also reduced electrically evoked release of both VIP and CGRP from rat small intestine, but did not affect acetylcholine, 5-HT or SP release. It was speculated that diabetes may prevent normal release of VIP and hence cause accumulation in nerve cell bodies, whereas the reduction in CGRP release may be due to selective degeneration of the nerve cells that contain it (Belai, Lincoln and Burnstock, 1987). Differences in the effects of streptozotocin treatment were reported between the myenteric and submucous plexuses. For example, sympathetic fibres to myenteric ganglia were reduced in number, but not in submucous ganglia (Belai and Burnstock, 1990). In addition, differential effects have been reported between the small and large intestines, corresponding to changes in sympathetic neurones in the coeliac but not inferior mesenteric ganglion (Belai et al., 1991). The effects of streptozotocin on VIP levels appear to be mediated by the diabetes that it causes, rather than as a direct effect, because they can be prevented by insulin treatment or inhibition of aldose reductase (Belai et al., 1996). There is evidence from animal models of experimentally-induced inflammation, and in human patients with idiopathic inflammation, of alterations in the function of nerves in the

452

INNERVATION OF THE GASTROINTESTINAL TRACT

gut wall. For example, release of noradrenaline from sympathetic varicosities is reduced in Trichinella spiralis infected rat gut (Swain, Blennerhassett and Collins, 1991). VIP content of enteric neurones has been variously reported to be upregulated or downregulated (see (Collins, 1996) for review). In addition, in wholemount preparations of human small intestine, increased immunoreactivity for TH, NPY, 5-HT, VIP, NOS and PACAP were reported in Crohns disease (Belai et al., 1997) as was increased calretinin immunoreactivity (Belai and Burnstock, 1999). It is not clear from this and other studies whether the changes in immunoreactivity reflect an increase in immunoreactivity in neurones or whether neurones may start to express neurochemicals that they do not normally contain. The report that some myenteric nerve cell bodies expressed TH immunoreactivity in Crohns disease but not in control tissue, suggests that qualitative changes in chemical coding may occur (Belai et al., 1992). The examples listed above are intended to illustrate that the chemical coding of enteric neurones is not fixed, but naturally change during early development and can also alter in response to a number of insults and disease states. A challenge for the future will be to determine whether the actual classes of enteric neurons are different in number or if some of their molecular mechanisms have been upregulated or downregulated. This distinction has profound implication for the understanding of the processes underlying these changes and for the development of rational interventions to correct the changes that underlie dysfunction.

CONCLUDING REMARKS Chemical coding has proven to be a remarkably powerful concept for distinguishing the different classes of neurones in the enteric nervous system. The combinations of neurochemicals present in nerve cell bodies and fibres allow neurones with different functions to be distinguished. In many cases, the neurochemical markers have obvious physiological significance and can give an insight into the probable role of the class of neurone. In other cases, the significance of particular markers is not apparent at present. It is possible that some of the markers actually have very little physiological significance. This will become clearer as good antagonists and blockers become available for the large range of molecules localised in enteric nerve cells. From the extensive body of work carried out to determine the chemical coding of enteric neurones, it is now well established that there is a remarkable degree of organisation of the enteric nervous system throughout the gastrointestinal tract.

ACKNOWLEDGEMENTS SJHB was supported by the NH&MRC of Australia. We would like to thank Kate Brody, Bao Nan Chen, Sean Fitzgibbon, Wendy Bonner and Janine Falconer-Edwards for valuable library assistance.

CELLULAR ORGANISATION OF THE MAMMALIAN ENTERIC NERVOUS SYSTEM

453

REFERENCES Absood, A., Ekblad, E., Ekelund, M., Hakanson, R. and Sundler, F. (1992). Helospectin-like peptides in the gastrointestinal tract: immunocytochemical localization and immunochemical characterization. Neuroscience, 46, 431–438. Ambache, N. and Freeman, M.A. (1968). Atropine-resistant spasms due to excitation of non-cholinergic neurones in guinea-pig myenteric plexus. Journal of Physiology, 198, 92–94. Anderson, C.R., Furness, J.B., Woodman, H.L., Edwards, S.L., Crack, P.J. and Smith, A.I. (1995). Characterisation of neurons with nitric oxide synthase immunoreactivity that project to prevertebral ganglia. Journal of the Autonomic Nervous System, 52, 107–116. Angel, F., Go, V.L., Schmalz, P.F. and Szurszewski, J.H. (1983). Vasoactive intestinal polypeptide: a putative transmitter in the canine gastric muscularis mucosa. Journal of Physiology, 341, 641–654. Barbiers, M., Timmermans, J.P., Adriaensen, D., De Groodt-Lasseel, M.H. and Scheuermann, D.W. (1993). Topographical distribution and immunocytochemical features of colonic neurons that project to the cranial mesenteric ganglion in the pig. Journal of the Autonomic Nervous System, 44, 119–127. Barbiers, M., Timmermans, J.P., Adriansen, D., De Groodt-Lasseel, M.H. and Scheuermann, D.W. (1995). Projections of neurochemically specified neurons in the porcine colon. Histochemistry and Cell Biology, 103, 115–126. Bartho, L., Holzer, P. and Lembeck, F. (1987). Is ganglionic transmission through nicotinic receptors essential for the peristaltic reflex in the guinea-pig ileum? Neuropharmacology, 26, 1663–1666. Bartho, L., Holzer, P., Leander, S. and Lembeck, F. (1989). Evidence for an involvement of substance P, but not cholecystokinin-like peptides, in hexamethonium-resistant intestinal peristalsis. Neuroscience, 28, 211–217. Bartho, L., Lenard, Jr. L. and Szigeti, R. (1998). Nitric oxide and ATP co-mediate the NANC relaxant response in the guinea-pig taenia caeci. Naunyn Schmiedebergs Archives of Pharmacology, 358, 496–469. Bauer, V. and Kuriyama, H. (1982a). Evidence for non-cholinergic, non-adrenergic transmission in the guineapig ileum. Journal of Physiology, 330, 95–110. Bauer, V. and Kuriyama, H. (1982b). The nature of non-cholinergic, non-adrenergic transmission in longitudinal and circular muscles of the guinea-pig ileum. Journal of Physiology, 332, 375–391. Bayliss, W.M. and Starling, E.H. (1899). The movements and innervation of the small intestine. Journal of Physiology, 24, 99–143. Belai, A., Boulos, P.B., Robson, T. and Burnstock, G. (1997). Neurochemical coding in the small intestine of patients with Crohn’s disease. Gut, 40, 767–774. Belai, A. and Burnstock, G. (1987). Selective damage of intrinsic calcitonin gene-related peptide-like immunoreactive enteric nerve fibers in streptozotocin-induced diabetic rats. Gastroenterology, 92, 730–734. Belai, A. and Burnstock, G. (1990). Changes in adrenergic and peptidergic nerves in the submucous plexus of streptozocin-diabetic rat ileum. Gastroenterology, 98, 1427–1436. Belai, A. and Burnstock, G. (1999). Distribution and colocalization of nitric oxide synthase and calretinin in myenteric neurons of developing, aging, and Crohn’s disease human small intestine. Digestive Diseases and Sciences, 44, 1579–1587. Belai, A., Calcutt, N.A., Carrington, A.L., Diemel, L.T., Tomlinson, D.R. and Burnstock, G. (1996). Enteric Neuropeptides in Streptozotocin-Diabetic Rats – Effects of Insulin and Aldose Reductase Inhibition. Journal of the Autonomic Nervous System, 58, 163–169. Belai, A., Lincoln, J. and Burnstock, G. (1987). Lack of release of vasoactive intestinal polypeptide and calcitonin gene-related peptide during electrical stimulation of enteric nerves in streptozotocin-diabetic rats. Gastroenterology, 93, 1034–1040. Belai, A., Lincoln, J., Milner, P. and Burnstock, G. (1991). Differential effect of streptozotocin-induced diabetes on the innervation of the ileum and distal colon. Gastroenterology, 100, 1024–1032. Belai, A., Lincoln, J., Milner, P., Crowe, R., Loesch, A. and Burnstock, G. (1985). Enteric nerves in diabetic rats: increase in vasoactive intestinal polypeptide but not substance P. Gastroenterology, 89, 967–976. Belai, A., Schmidt, H.H., Hoyle, C.H., Hassall, C.J., Saffrey, M.J., Moss, J., et al. (1992). Colocalization of nitric oxide synthase and NADPH-diaphorase in the myenteric plexus of the rat gut. Neuroscience Letters, 143, 60–64. Berthoud, H.R., Jedrzejewska, A. and Powley, T.L. (1990). Simultaneous labeling of vagal innervation of the gut and afferent projections from the visceral forebrain with dil injected into the dorsal vagal complex in the rat. Journal of Comparative Neurology, 301, 65–79 Berthoud, H.R. and Powley, T.L. (1992). Vagal afferent innervation of the rat fundic stomach: morphological characterization of the gastric tension receptor. Journal of Comparative Neurology, 319, 261–276. Bertrand, P.P. and Galligan, J.J. (1994). Contribution of chloride conductance increase to slow EPSC and tachykinin current in guinea-pig myenteric neurones. Journal of Physiology, 481, 47–60.

454

INNERVATION OF THE GASTROINTESTINAL TRACT

Bertrand, P.P. and Galligan, J.J. (1995). Signal-transduction pathways causing slow synaptic excitation in guinea-pig myenteric AH neurons. American Journal of Physiology, 269, G710–G720. Bertrand, P.P., Kunze, W.A., Bornstein, J.C., Furness, J.B. and Smith, M.L. (1997). Analysis of the responses of myenteric neurons in the small intestine to chemical stimulation of the mucosa. American Journal of Physiology, 273, G422–G435. Bian, X.C., Bertrand, P.P. and Bornstein, J.C. (2000). Descending inhibitory reflexes involve P2X receptormediated transmission from interneurons to motor neurons in guinea-pig ileum. Journal of Physiology, 528, 551–560. Bjorkroth, U. (1983). Inhibition of smooth muscle contractions induced by capsaicin and electrical stimulation by a substance P antagonist. Acta Physiologica Scandinivica, 515 (Supplement), 11–16. Blaugrund, E., Pham, T.D., Tennyson, V.M., Lo, L., Sommer, L., Anderson, D.J. et al. (1996). Distinct subpopulations of enteric neuronal progenitors defined by time of development, sympathoadrenal lineage markers and mash-1-dependence. Development, 122, 309–320. Bornstein, J.C., Costa, M., Furness, J.B. and Lang, R.J. (1986). Electrophysiological analysis of projections of enteric inhibitory motoneurones in the guinea-pig small intestine. Journal of Physiology, 370, 61–74. Bornstein, J.C., Costa, M., Furness, J.B. and Lees, G.M. (1984a). Electrophysiology and enkephalin immunoreactivity of identified myenteric plexus neurones of guinea-pig small intestine. Journal of Physiology, 351, 313–325. Bornstein, J.C. and Furness, J.B. (1988). Correlated electrophysiological and histochemical studies of submucous neurons and their contribution to understanding enteric neural circuits. Journal of the Autonomic Nervous System, 25, 1–13. Bornstein, J.C., Furness, J.B. and Costa, M. (1987). Sources of excitatory synaptic inputs to neurochemically identified submucous neurons of guinea-pig small intestine. Journal of the Autonomic Nervous System, 18, 83–91. Bornstein, J.C., Furness, J.B. and Costa, M. (1989). An electrophysiological comparison of substance P-immunoreactive neurons with other neurons in the guinea-pig submucous plexus. Journal of the Autonomic Nervous System, 26, 113–120. Bornstein, J.C., Furness, J.B. and Kunze, W.A.A. (1994). Electrophysiological characterizatin of myenteric neurons – how do classification schemes relate. Journal of the Autonomic Nervous System, 48, 1–15. Bornstein, J.C., Furness, J.B., Smith, T.K. and Trussell, D.C. (1991a). Synaptic responses evoked by mechanical stimulation of the mucosa in morphologically characterized myenteric neurons of the guinea-pig ileum. Journal of Neuroscience, 11, 505–518. Bornstein, J.C., Hendriks, R., Furness, J.B. and Trussell, D.C. (1991b). Ramifications of the axons of AHneurons injected with the intracellular marker biocytin in the myenteric plexus of the guinea-pig small intestine. Journal of Comparative Neurology, 314, 437–451. Bornstein, J.C., North, R.A., Costa, M. and Furness, J.B. (1984b). Excitatory synaptic potentials due to activation of neurons with short projections in the myenteric plexus. Neuroscience, 11, 723–731. Brand, S.J. and Schmidt, W.E. (1991). Gastrointestinal Hormones. In Textbook of Gastroenterology, edited by T. Yamada, pp. 25–70. Philadelphia: J.B. Lippincott Co. Brookes, S.J.H. (1993). Neuronal nitric oxide in the gut. Journal of Gastroenterology and Hepatology, 8, 590–603. Brookes, S.J.H. (2001). Retrograde tracing of enteric neuronal pathways. Neurogastroenterology and Motility, 13, 1–18. Brookes, S.J.H. and Costa, M. (1990). Identification of enteric motor neurones which innervate the circular muscle of the guinea-pig small intestine. Neuroscience Letters, 118, 227–230. Brookes, S.J.H., Chen, B.N., Costa, M. and Humphreys, C.M.S. (1999). Initiation of peristalsis by circumferential stretch of flat sheets of guinea-pig ileum. Journal of Physiology, 516, 525–538. Brookes, S.J.H., Chen, B.N., Hodgson, W.M. and Costa, M. (1996). Characterization of excitatory and inhibitory motor neurons to the guinea-pig lower esophageal sphincter. Gastroenterology, 111, 108–117. Brookes, S.J.H., Ewart, W.R. and Wingate, D.L. (1987). Intracellular recordings from myenteric neurones in the human colon. Journal of Physiology, 390, 305–318. Brookes, S.J.H., Ewart, W.R. and Wingate, D.L. (1988). Intracellular recordings from cells in the myenteric plexus of the rat duodenum. Neuroscience, 24, 297–307. Brookes, S.J.H., Hennig, G. and Schemann, M. (1998). Identification of motor neurons to the circular muscle of the guinea-pig gastric corpus. Journal of Comparative Neurology, 397, 268–280. Brookes, S.J.H., Meedeniya, A.C., Jobling, P. and Costa, M. (1997). Orally projecting interneurones in the guinea-pig small intestine. Journal of Physiology, 505, 473–491. Brookes, S.J.H., Song, Z.M., Ramsay, G.A. and Costa, M. (1995). Long aboral projections of Dogiel type II, AH neurons within the myenteric plexus of the guinea-pig small intestine. Journal of Neuroscience, 15, 4013–4022.

CELLULAR ORGANISATION OF THE MAMMALIAN ENTERIC NERVOUS SYSTEM

455

Brookes, S.J.H., Song, Z.M., Steele, P.A. and Costa, M. (1992). Identification of motor neurons to the longitudinal muscle of the guinea-pig ileum. Gastroenterology, 103, 961–973. Brookes, S.J.H., Steele, P.A. and Costa, M. (1991a). Calretinin immunoreactivity in cholinergic motor neurones, interneurones and vasomotor neurones in the guinea-pig small intestine. Cell and Tissue Research, 263, 471–481. Brookes, S.J.H., Steele, P.A. and Costa, M. (1991b). Identification and immunohistochemistry of cholinergic and non-cholinergic circular muscle motor neurons in the guinea-pig small intestine. Neuroscience, 42, 863–878. Browning, K.N., Cunningham, S.M., Duncan, L., Timmermans, J. and Lees, G.M. (1999). Regional differences in the sympathetic innervation of the guinea-pig large intestine by neuropeptide Y- and tyrosine hydroxylase-immunoreactive nerves of divergent extrinsic origin. Journal of Comparative Neurology, 410, 515–530. Browning, K.N. and Lees, G.M. (1994). Reappraisal of the innervation of rat intestine by vasoactive intestinal polypeptide and neuropeptide Y-immunoreactive neurons. Neuroscience, 62, 1257–1266. Browning, K.N. and Lees, G.M. (1996). Myenteric neurons of the rat descending colon: electrophysiological and correlated morphological properties. Neuroscience, 73, 1029–1047. Browning, K.N. and Lees, G.M. (2000). Inhibitory effects of NPY on ganglionic transmission in myenteric neurones of the guinea-pig descending colon. Neurogastroenterology and Motility, 12, 33–41. Burnstock, G., Campbell, G., Bennett, M. and Holman, M.E. (1963). Inhibition of the smooth muscle of the taenia coli. Nature, 200, 581–582. Burnstock, G., Campbell, G. and Rand, M.J. (1966). The inhibitory innervation of the taenia of the guinea-pig caecum. Journal of Physiology, 182, 504–526. Bywater, R.A. (1994). Activity following colonic distension in enteric sensory fibres projecting to the inferior mesenteric ganglion in the guinea-pig. Journal of the Autonomic Nervous System, 46, 19–26. Bywater, R.A., Holman, M.E. and Taylor, G.S. (1981). Atropine-resistant depolarization in the guinea-pig small intestine. Journal of Physiology, 316, 369–378. Bywater, R.A. and Taylor, G.S. (1986). Non-cholinergic excitatory and inhibitory junction potentials in the circular smooth muscle of the guinea-pig ileum. Journal of Physiology, 374, 153–164. Cannon, W.B. (1912). Peristalsis, segmentation and the myenteric reflex. American Journal of Physiology, 30, 114–128. Cherubini, E. and North, R.A. (1984). Inhibition of calcium spikes and transmitter release by γ-aminobutyric acid in the guinea-pig myenteric plexus. British Journal of Pharmacology, 82, 101–105. Cherubini, E. and North, R.A. (1985). Mu and kappa opioids inhibit transmitter release by different mechanisms. Proceedings of the National Academy of Sciences USA, 82, 1860–1863. Christensen, J. and Rick, G.A. (1987). Intrinsic nerves in the mammalian colon: confirmation of a plexus at the circular muscle-submucosal interface. Journal of the Autonomic Nervous System, 21, 223–231. Christensen, J., Stiles, M.J., Rick, G.A. and Sutherland, J. (1984). Comparative anatomy of the myenteric plexus of the distal colon in eight mammals. Gastroenterology, 86, 706–713. Christofi, F.L. and Wood, J.D. (1993a). Effects of PACAP on morphologically identified myenteric neurons in guinea-pig small bowel. American Journal of Physiology, 264, G414–G421. Christofi, F.L. and Wood, J.D. (1993b). Presynaptic inhibition by adenosine A1 receptors on guinea-pig small intestinal myenteric neurons. Gastroenterology, 104, 1420–1429. Clerc, N., Furness, J.B., Li, Z.S., Bornstein, J.C. and Kunze, W.A. (1998). Morphological and immunohistochemical identification of neurons and their targets in the guinea-pig duodenum. Neuroscience, 86, 679–694. Collins, S.M. (1996). The immunomodulation of enteric neuromuscular function – implications for motility and inflammatory disorders. Gastroenterology, 111, 1683–1699. Cooke, H.J. and Reddix, R.A. (1994). Neural Regulation of Intestinal Electrolyte Transport. In Physiology of the Gastrointestinal Tract, edited by L.R. Johnson, pp. 2083–2132. New York: Raven Press. Coons, A.H. (1958). Fluorescent antibody methods. In General Cytochemical Methods, edited by J.F. Danielli, pp. 399–422. New York: Academic Press. Cornelissen, W., De Laet, A., Kroese, A.B., Van Bogaert, P.P., Scheuermann, D.W. and Timmermans, J.P. (2000). Electrophysiological features of morphological Dogiel type II neurons in the myenteric plexus of pig small intestine. Journal of Neurophysiology, 84, 102–111. Costa, M., Brookes, S.J.H., Steele, P.A., Gibbins, I., Burcher, E. and Kandiah, C.J. (1996). Neurochemical classification of myenteric neurons in the guinea-pig ileum. Neuroscience, 75, 949–967. Costa, M., Cuello, A.C., Furness, J.B. and Franco, R. (1980a). Distribution of enteric neurons showing immunoreactivity for substance P in the guinea-pig ileum. Neuroscience, 5, 323–331. Costa, M. and Furness, J.B. (1983). The origins, pathways and terminations of neurons with VIP-like immunoreactivity in the guinea-pig small intestine. Neuroscience, 8, 665–676.

456

INNERVATION OF THE GASTROINTESTINAL TRACT

Costa, M., Furness, J.B., Buffa, R. and Said, S.I. (1980b). Distribution of enteric nerve cell bodies and axons showing immunoreactivity for vasoactive intestinal polypeptide in the guinea-pig intestine. Neuroscience, 5, 587–596. Costa, M., Furness, J.B., Cuello, A.C.U., Verhofstad, A.A., Steinbusch, H.W. and Elde, R.P. (1982). Neurons with 5-hydroxytryptamine-like immunoreactivity in the enteric nervous system: their visualization and reactions to drug treatment. Neuroscience, 7, 351–363. Costa, M., Furness, J.B. and Gibbins, I.L. (1986). Chemical coding of enteric neurons. Progress in Brain Research, 68, 217–240. Costa, M., Furness, J.B. and Humphreys, C.M. (1986). Apamin distinguishes two types of relaxation mediated by enteric nerves in the guinea-pig gastrointestinal tract. Naunyn Schmiedebergs Archives of Pharmacology, 332, 79–88. Costa, M., Furness, J.B., Llewellyn Smith, I.J. and Cuello, A.C. (1981). Projections of substance P-containing neurons within the guinea-pig small intestine. Neuroscience, 6, 411–424. Costa, M., Furness, J.B., Pompolo, S., Brookes, S.J.H., Bornstein, J.C., Bredt, D.S., et al. (1992). Projections and chemical coding of neurons with immunoreactivity for nitric oxide synthase in the guinea-pig small intestine. Neuroscience Letters, 148, 121–125. Costa, M., Furness, J.B., Smith, I.J., Davies, B. and Oliver, J. (1980c). An immunohistochemical study of the projections of somatostatin-containing neurons in the guinea-pig intestine. Neuroscience, 5, 841–852. Costa, M., Furness, J.B., Yanaihara, N., Yanaihara, C. and Moody, T.W. (1984). Distribution and projections of neurons with immunoreactivity for both gastrin-releasing peptide and bombesin in the guinea-pig small intestine. Cell and Tissue Research, 235, 285–293. Cousins, H.M., Edwards, F.R., Hirst, G.D. and Wendt, I.R. (1993). Cholinergic neuromuscular transmission in the longitudinal muscle of the guinea-pig ileum. Journal of Physiology, 471, 61–86. Crist, J.R., He, X.D. and Goyal, R.K. (1992). Both ATP and the peptide VIP are inhibitory neurotransmitters in guinea-pig ileum circular muscle. Journal of Physiology, 447, 119–131. Crowe, R., Kamm, M.A., Burnstock, G. and Lennard-Jones, J.E. (1992). Peptide-containing neurons in different regions of the submucous plexus of human sigmoid colon. Gastroenterology, 102, 461–467. Daniel, E.E., Costa, M., Furness, J.B. and Keast, J.R. (1985). Peptide neurons in the canine small intestine. Journal of Comparative Neurology, 237, 227–238. Doerffler-Melly, J. and Neuhuber, W.L. (1988). Rectospinal neurons: evidence for a direct projection from the enteric to the central nervous system in the rat. Neuroscience Letters, 92, 121–125. Dogiel, A.S. (1899). Über den Bau der Ganglien in den Geflechten des Darmes und der Gallenblase des Menschen und der Säugethiere. Archiv für Anaomie und Physiogie Leipzig Anat. Abteil. Jahrgang 1899, 130–158. Domoto, T., Teramoto, M., Tanigawa, K., Tamura, K. and Yasui, Y. (1995). Origins of nerve fibers containing nitric oxide synthase in the rat celiac-superior mesenteric ganglion. Cell and Tissue Research, 281, 215–221. Ekblad, E., Ekman, R., Hakanson, R. and Sundler, F. (1988). Projections of peptide-containing neurons in rat colon. Neuroscience, 27, 655–674. Ekblad, E., Mei, Q. and Sundler, F. (2000). Innervation of the gastric mucosa. Microscopy Research and Technique, 48, 241–257. Ekblad, E., Mulder, H., Uddman, R. and Sundler, F. (1994). NOS-containing neurons in the rat gut and coeliac ganglia. Neuropharmacology, 33, 1323–1331. Ekblad, E., Winther, C., Ekman, R., Hakanson, R. and Sundler, F. (1987). Projections of peptide-containing neurons in rat small intestine. Neuroscience, 20, 169–188. Erde, S.M., Sherman, D. and Gershon, M.D. (1985). Morphology and serotonergic innervation of physiologically identified cells of the guinea-pig’s myenteric plexus. Journal of Neuroscience, 5, 617–633. Evans, R.J., Jiang, M.M. and Surprenant, A. (1994). Morphological properties and projections of electrophysiologically characterized neurons in the guinea-pig submucosal plexus. Neuroscience, 59, 1093–1110. Feher, E. (1982). Electron microscopic study of retrograde axonal transport of horseradish peroxidase in the wall of the small intestine in the cat. Acta Anatomica, 112, 69–78. Franco, R., Costa, M. and Furness, J.B. (1979). Evidence for the release of endogenous substance P from intestinal nerves. Naunyn Schmiedebergs Archives of Pharmacology, 306, 195–201. Furness, J.B., Bornstein, J.C. and Trussell, D.C. (1988). Shapes of nerve cells in the myenteric plexus of the guinea-pig small intestine revealed by the intracellular injection of dye. Cell and Tissue Research, 254, 561–571. Furness, J.B., Clerc, N., Lomax, A.E.G., Bornstein, J.C. and Kunze, W.A.A. (2000). Shapes and projections of tertiary plexus neurons of the guinea-pig small intestine. Cell and Tissue Research, 300, 383–387. Furness, J.B. and Costa, M. (1982). Neurons with 5-hydroxytryptamine-like immunoreactivity in the enteric nervous system: their projections in the guinea-pig small intestine. Neuroscience, 7, 341–349.

CELLULAR ORGANISATION OF THE MAMMALIAN ENTERIC NERVOUS SYSTEM

457

Furness, J.B. and Costa, M. (1987). The Enteric Nervous System. Edinburgh: Churchill-Livingstone. Furness, J.B., Costa, M., Emson, P.C., Hakanson, R., Moghimzadeh, E., Sundler, F. et al. (1983). Distribution, pathways and reactions to drug treatment of nerves with neuropeptide Y- and pancreatic polypeptide-like immunoreactivity in the guinea-pig digestive tract. Cell and Tissue Research, 234, 71–92. Furness, J.B., Costa, M., Gibbins, I.L., Llewellyn Smith, I.J. and Oliver, J.R. (1985). Neurochemically similar myenteric and submucous neurons directly traced to the mucosa of the small intestine. Cell and Tissue Research, 241, 155–163. Furness, J.B., Costa, M. and Keast, J.R. (1984). Choline acetyltransferase- and peptide immunoreactivity of submucous neurons in the small intestine of the guinea-pig. Cell and Tissue Research, 237, 329–336. Furness, J.B., Costa, M. and Miller, R.J. (1983). Distribution and projections of nerves with enkephalin-like immunoreactivity in the guinea-pig small intestine. Neuroscience, 8, 653–664. Furness, J.B., Costa, M., Rokaeus, A., McDonald, T.J. and Brooks, B. (1987). Galanin-immunoreactive neurons in the guinea-pig small intestine: their projections and relationships to other enteric neurons. Cell and Tissue Research, 250, 607–615. Furness, J.B., Johnson, P.J., Pompolo, S. and Bornstein, J.C. (1995). Evidence that enteric motility reflexes can be initiated through entirely intrinsic mechanisms in the guinea-pig small intestine. Neurogastroenterology and Motility, 7, 89–96. Furness, J.B., Keast, J.R., Pompolo, S., Bornstein, J.C., Costa, M., Emson, P.C. et al. (1988). Immunohistochemical evidence for the presence of calcium-binding proteins in enteric neurons. Cell and Tissue Research, 252, 79–87. Furness, J.B., Kunze, W.A., Bertrand, P.P., Clerc, N. and Bornstein, J.C. (1998). Intrinsic primary afferent neurons of the intestine. Progress in Neurobiology, 54, 1–18. Furness, J.B., Li, Z.S., Young, H.M. and Forstermann, U. (1994). Nitric oxide synthase in the enteric nervous system of the guinea-pig: a quantitative description. Cell and Tissue Research, 277, 139–149. Furness, J.B., Lloyd, K.C., Sternini, C. and Walsh, J.H. (1990a). Projections of substance P, vasoactive intestinal peptide and tyrosine hydroxylase immunoreactive nerve fibres in the canine intestine, with special reference to the innervation of the circular muscle. Archives of Histology and Cytology, 53, 129–140. Furness, J.B., Pompolo, S., Murphy, R. and Giraud, A. (1989a). Projections of neurons with neuromedin U-like immunoreactivity in the small intestine of the guinea-pig. Cell and Tissue Research, 257, 415–422. Furness, J.B., Trussell, D.C., Pompolo, S., Bornstein, J.C., Maley, B.E. and Storm-Mathisen, J. (1989b). Shapes and projections of neurons with immunoreactivity for γ-aminobutyric acid in the guinea-pig small intestine. Cell and Tissue Research, 256, 293–301. Furness, J.B., Trussell, D.C., Pompolo, S., Bornstein, J.C. and Smith, T.K. (1990b). Calbindin neurons of the guinea-pig small intestine: quantitative analysis of their numbers and projections. Cell and Tissue Research, 260, 261–272. Furukawa, K., Taylor, G.S. and Bywater, R.A. (1986). An intracellular study of myenteric neurons in the mouse colon. Journal of Neurophysiology, 55, 1395–1406. Gabella, G. (1987). The number of neurons in the small intestine of mice, guinea-pigs and sheep. Neuroscience, 22, 737–752. Gabella, G. (1989). Fall in the number of myenteric neurons in aging guinea-pigs Gastroenterology, 96, 1487– 1493. Gabella, G. (1994). Structure of muscles and nerves in the gastrointestinal tract. In Physiology of the Gastrointestinal Tract, Third Edition, edited by L.R. Johnson, pp. 751–793. New York: Raven Press. Gabella, G. and Trigg, P. (1984). Size of neurons and glial cells in the enteric ganglia of mice, guinea-pigs, rabbits and sheep. Journal of Neurocytology, 13, 49–71. Galligan, J.J. and Bertrand, P.P. (1994). ATP mediates fast synaptic potentials in enteric neurons. Journal of Neuroscience, 14, 7563–7571. Galligan, J.J., Jiang, M.M., Shen, K.Z. and Surprenant, A. (1990a). Substance P mediates neurogenic vasodilatation in extrinsically denervated guinea-pig submucosal arterioles. Journal of Physiology, 420, 267– 280. Galligan, J.J., North, R.A. and Tokimasa, T. (1989). Muscarinic agonists and potassium currents in guinea-pig myenteric neurones. British Journal of Pharmacology, 96, 193–203. Galligan, J.J., Tatsumi, H., Shen, K.Z., Surprenant, A. and North, R.A. (1990b). Cation current activated by hyperpolarization (IH) in guinea-pig enteric neurons. American Journal of Physiology, 259, G966–G972. Gershon, M.D. (1997). Genes and lineages in the formation of the enteric nervous system. Current Opinion in Neurobiology, 7, 101–109. Gershon, M.D. (1998). V. Genes, lineages, and tissue interactions in the development of the enteric nervous system. American Journal of Physiology, 275, G869–G873. Giaroni, C., De Ponti, F., Cosentino, M., Lecchini, S. and Frigo, G. (1999). Plasticity in the enteric nervous system. Gastroenterology, 117, 1438–1458.

458

INNERVATION OF THE GASTROINTESTINAL TRACT

Gibbins, I.L., Furness, J.B., Costa, M., MacIntyre, I., Hillyard, C.J. and Girgis, S. (1985). Co-localization of calcitonin gene-related peptide-like immunoreactivity with substance P in cutaneous, vascular and visceral sensory neurons of guinea-pigs. Neuroscience Letters, 57, 125–130. Gola, M., Niel, J.P., Bessone, R. and Fayolle, R. (1992). Single-channel and whole-cell recordings from nondissociated sympathetic neurones in rabbit coeliac ganglia. Journal of Neuroscience Methods, 43, 13–22. Gomes, O.A., Desouza, R.R. and Liberti, E.A. (1997). A preliminary investigation of the effects of aging on the nerve cell number in the myenteric ganglia of the human colon. Gerontology, 43, 210–217. Grafe, P., Mayer, C.J. and Wood, J.D. (1980). Synaptic modulation of calcium-dependent potassium conductance in myenteric neurones in the guinea-pig. Journal of Physiology, 305, 235–248. Grafe, P.G., Wood, J.D. and Mayer, C.J. (1979). Fast excitatory postsynaptic potentials in AH (Type 2) neurons of guinea-pig myenteric plexus. Brain Research, 163, 349–352. Grider, J.R. (1994). CGRP as a transmitter in the sensory pathway mediating peristaltic reflex. American Journal of Physiology, 266, 1139–1145. Grider, J.R. and Jin, J.G. (1994). Distinct populations of sensory neurons mediate the peristaltic reflex elicited by muscle stretch and mucosal stimulation. Journal of Neuroscience, 14, 2854–2860. Griffith, S.G. and Burnstock, G. (1983). Serotoninergic neurons in human fetal intestine: an immunohistochemical study. Gastroenterology, 85, 929–937. Hamaji, M., Kawai, Y., Kawashima, Y. and Tohyama, M. (1987). Projections of bombesin-like immunoreactive fibers from the rat stomach to the celiac ganglion revealed by a double-labeling technique. Brain Research, 416, 192–194. Heinicke, E.A. and Kiernan, J.A. (1990). An immunohistochemical study of the myenteric plexus of the colon in the rat and mouse. Journal of Anatomy, 170, 51–62. Hendriks, R., Bornstein, J.C. and Furness, J.B. (1990). An electrophysiological study of the projections of putative sensory neurons within the myenteric plexus of the guinea-pig ileum. Neuroscience Letters, 110, 286–290. Hennig, G.W., Costa, M., Chen, B.N. and Brookes, S.J.H. (1999). Quantitative analysis of peristalsis in the guinea-pig small intestine using spatio-temporal maps. Journal of Physiology, 517, 575–590. Hens, J., Schrodl, F., Brehmer, A., Adriaensen, D., Neuhuber, W., Scheuermann, D.W. et al. (2000). Mucosal projections of enteric neurons in the porcine small intestine. Journal of Comparative Neurology, 421, 429–436. Hidaka, T. and Kuriyama, H. (1969). Responses of the smooth muscle membrane in guinea-pig jejunum elicited by field stimulation. Journal of General Physiology, 53, 471–486. Hill, C.J. (1927). A contriubtion to our knowldege of the enteric plexuses. Philosophical Transactions of the Royal Society of London, Series B, 215, 355–387. Hirst, G.D., Johnson, S.M. and van Helden, D.F. (1985a). The calcium current in a myenteric neurone of the guinea-pig ileum. Journal of Physiology, 361, 297–314. Hirst, G.D., Johnson, S.M. and van Helden, D.F. (1985b). The slow calcium-dependent potassium current in a myenteric neurone of the guinea-pig ileum. Journal of Physiology, 361, 315–337. Hirst, G.D.S., Holman, M.E. and McKirdy, H.C. (1975). Two descending nerve pathways activated by distension of guinea-pig small intestine. Journal of Physiology, 244, 113–127. Hirst, G.D.S., Holman, M.E. and Spence, I. (1974). Two types of neurones in the myenteric plexus of duodenum in the guinea-pig. Journal of Physiology, 236, 303–326. Hirst, G.D.S. and McKirdy, H.C. (1975). Synaptic potentials recorded from neurones of the submucous plexus of guinea-pig small intestine. Journal of Physiology, 249, 369–385. Hodgkiss, J.P. and Lees, G.M. (1983). Morphological studies of electrophysiologically-identified myenteric plexus neurons of the guinea-pig ileum. Neuroscience, 8, 593–608. Hökfelt, T., Johansson, O., Efendic, S., Luft, A. and Arimura, A. (1975). Are there somatostain containing nerves in the rat gut? Experientia, 31, 852–854. Holman, M.E. and Weinrich, J.P. (1975). The effects of calcium and magnesium on inhibitory junction transmission in smooth muscle of guinea-pig small intestine. Pflügers Archiv, 360, 109–119. Holst, M.C., Kelly, J.B. and Powley, T.L. (1997). Vagal preganglionic projections to the enteric nervous system characterized with Phaseolus vulgaris-leucoagglutinin. Journal of Comparative Neurology, 381, 81–100. Holzer, P. (1989). Ascending enteric reflex: multiple neurotransmitter systems and interactions. American Journal of Physiology, 256, G540–G545. Holzer, P. and Holzer-Petsche, U. (1997). Tachykinins in the gut. Part I. Expression, release and motor function. Pharmacology and Therapeutics, 73, 173–217. Holzer, P., Holzer-Petsche, U. and Leander, S. (1986). A tachykinin antagonist inhibits gastric emptying and gastrointestinal transit in the rat. British Journal of Pharmacology, 89, 453–459. Holzer-Petsche, U. (1995). Tachykinin receptors in gastrointestinal motility. Regulatory Peptides, 57, 19–42.

CELLULAR ORGANISATION OF THE MAMMALIAN ENTERIC NERVOUS SYSTEM

459

Horikawa, K. and Armstrong, W.E. (1988). A versatile means of intracellular labeling: injection of biocytin and its detection with avidin conjugates. Journal of Neuroscience Methods, 25, 1–11. Hoyle, C. and Burnstock, G. (1989). Neuronal populations in the submucous plexus of the human colon. Journal of Anatomy, 166, 7–22. Iggo, A. (1955). Tension receptors in the stomach and the urinary bladder. Journal of Physiology, 128, 593–607. Iyer, V., Bornstein, J.C., Costa, M., Furness, J.B., Takahashi, Y. and Iwanaga, T. (1988). Electrophysiology of guinea-pig myenteric neurons correlated with immunoreactivity for calcium binding proteins. Journal of the Autonomic Nervous System, 22, 141–150. Jessen, K.R., Hills, J.M. and Saffrey, M.J. (1986). Immunohistochemical demonstration of GABAergic neurons in the enteric nervous system. Journal of Neuroscience, 6, 1628–1634. Jiang, M.M. and Surprenant, A. (1992). Re-innervation of submucosal arterioles by myenteric neurones following extrinsic denervation. Journal of the Autonomic Nervous System, 37, 145–154. Johnson, P.J., Bornstein, J.C. and Burcher, E. (1998). Roles of neuronal NK1 and NK3 receptors in synaptic transmission during motility reflexes in the guinea-pig ileum. British Journal of Pharmacology, 124, 1375–1384. Johnson, P.J., Bornstein, J.C., Yuan, S.Y. and Furness, J.B. (1996). Analysis of contributions of acetylcholine and tachykinins to neuro-neuronal transmission in motility reflexes in the guinea-pig ileum. British Journal of Pharmacology, 118, 973–983. Johnson, P.J., Shum, O.R., Thornton, P.D. and Bornstein, J.C. (1999). Evidence that inhibitory motor neurons of the guinea-pig small intestine exhibit fast excitatory synaptic potentials mediated via P2X receptors. Neuroscience Letters, 266, 169–172. Johnson, S.M., Katayama, Y. and North, R.A. (1980). Multiple actions of 5-hydroxytryptamine on myenteric neurones of the guinea-pig ileum. Journal of Physiology, 304, 459–470. Karaosmanoglu, T., Aygun, B., Wade, P.R. and Gershon, M.D. (1996). Regional differences in the number of neurons in the myenteric plexus of the guinea-pig small intestine and colon: an evaluation of markers used to count neurons. Anatomical Record, 244, 470–480. Katayama, Y. and North, R.A. (1978). Does substance P mediate slow synaptic excitation within the myenteric plexus? Nature, 274, 387–388. Katayama, Y. and North, R.A. (1980). The action of somatostatin on neurones of the myenteric plexus of the guinea-pig ileum. Journal of Physiology, 303, 315–323. Katayama, Y., Lees, G.M. and Pearson, G.T. (1986). Electrophysiology and morphology of vasoactiveintestinal-peptide-immunoreactive neurones of the guinea-pig ileum. Journal of Physiology, 378, 1–11. Keast, J.R. (1987). Mucosal innervation and control of water and ion transport in the intestine. Reviews of Physiology Biochemistry and Pharmacology, 109, 1–59. Keast, J.R., Furness, J.B. and Costa, M. (1984). Somatostatin in human enteric nerves. Distribution and characterization. Cell and Tissue Research, 237, 299–308. Keast, J.R., Furness, J.B. and Costa, M. (1987). Distribution of peptide-containing neurons and endocrine cells in the rabbit gastrointestinal tract, with particular reference to the mucosa. Cell and Tissue Research, 248, 565–577. Kirchgessner, A.L. and Gershon, M.D. (1988). Projections of submucosal neurons to the myenteric plexus of the guinea-pig intestine: in vitro tracing of microcircuits by retrograde and anterograde transport. Journal of Comparative Neurology, 277, 487–498. Kirchgessner, A.L. and Gershon, M.D. (1989). Identification of vagal efferent fibers and putative target neurons in the enteric nervous system of the rat. Journal of Comparative Neurology, 285, 38–53. Kirchgessner, A.L., Liu, M.T. and Gershon, M.D. (1996). In situ identification and visualization of neurons that mediate enteric and enteropancreatic reflexes. Journal of Comparative Neurology, 371, 270–286. Kirchgessner, A.L., Tamir, H. and Gershon, M.D. (1992). Identification and stimulation by serotonin of intrinsic sensory neurons of the submucosal plexus of the guinea-pig gut: activity-induced expression of Fos immunoreactivity. Journal of Neuroscience, 12, 235–248. Kromer, W. (1990). Endogenous opioids, the enteric nervous system and gut motility. Digestive Diseases, 8, 361–373. Kottegoda, S.R. (1970). Peristalsis of the small intestine. In Smooth Muscle, edited by E. Bülbring, B.A., A. Jones and T. Tomita, pp. 525–541. London: Edward Arnold. Krantis, A., Nichols, K. and Staines, W. (1998). Neurochemical characterization and distribution of enteric GABAergic neurons and nerve fibres in the human colon. Journal of the Autonomic Nervous System, 68, 33–42. Kuntz, A. and Saccomanno, G. (1944). Reflex inhibition of intestinal motility mediated through decentralized prevertebral ganglia. Journal of Neurophysiology, 7, 163–170. Kunze, W.A. (1998). A mobile intracellular microelectrode designed to record from neurons in contracting tissue. Brain Research, Brain Research Protocols, 3, 94–99.

460

INNERVATION OF THE GASTROINTESTINAL TRACT

Kunze, W.A., Bertrand, P.P., Furness, J.B. and Bornstein, J.C. (1997). Influence of the mucosa on the excitability of myenteric neurons. Neuroscience, 76, 619–634. Kunze, W.A., Bornstein, J.C. and Furness, J.B. (1995). Identification of sensory nerve cells in a peripheral organ (the intestine) of a mammal. Neuroscience, 66, 1–4. Kunze, W.A., Bornstein, J.C., Furness, J.B., Hendriks, R. and Stephenson, D.S. (1994). Charybdotoxin and iberiotoxin but not apamin abolish the slow after-hyperpolarization in myenteric plexus neurons. Pflugers Archiv European Journal of Physiology, 428, 300–306. Kunze, W.A., Clerc, N., Bertrand, P.P. and Furness, J.B. (1999). Contractile activity in intestinal muscle evokes action potential discharge in guinea-pig myenteric neurons. Journal of Physiology, 517, 547–613. Kunze, W.A., Clerc, N., Furness, J.B. and Gola, M. (2000). The soma and neurites of primary afferent neurons in the guinea-pig intestine respond differentially to deformation. Journal of Physiology, 526, 375–385. Kunze, W.A., Furness, J.B., Bertrand, P.P. and Bornstein, J.C. (1998). Intracellular recording from myenteric neurons of the guinea-pig ileum that respond to stretch. Journal of Physiology, 506, 827–842. Kunze, W.A.A. and Furness, J.B. (1999). The enteric nervous system and regulation of intestinal motility. Annual Review of Physiology, 61, 117–142. Kunze, W.A.A., Furness, J.B. and Bornstein, J.C. (1993). Simultaneous Intracellular Recordings From Enteric Neurons Reveal That Myenteric Ah Neurons Transmit Via Slow Excitatory Postsynaptic Potentials. Neuroscience, 55, 685–694. Kuramoto, H. and Furness, J.B. (1989). Distribution of enteric nerve cells that project from the small intestine to the coeliac ganglion in the guinea-pig. Journal of the Autonomic Nervous System, 27, 241–248. Langley, J.N. (1921). The autonomic nervous system. Heffer, Cambridge. Langley, J.N. and Magnus, R. (1905). Some observations of the movements of the intestine before and after degenerative section of the mesenteric nerves. Journal of Physiology, 33, 34–51. Lee, Y., Shiosaka, S., Hayashi, N. and Tohyama, M. (1986). The presence of vasoactive intestinal polypeptidelike immunoreactive structures projecting from the myenteric ganglion of the stomach to the celiac ganglion revealed by a double-labelling technique. Brain Research, 382, 392–394. LePard, K.J. and Galligan, J.J. (1999). Analysis of fast synaptic pathways in myenteric plexus of guinea-pig ileum. American Journal of Physiology, 276, G529–G538. Li, C.G. and Rand, M.J. (1990). Nitric oxide and vasoactive intestinal polypeptide mediate non-adrenergic, noncholinergic inhibitory transmission to smooth muscle of the rat gastric fundus. European Journal of Pharmacology, 191, 303–309. Li, M.Z. and Masuko, S. (1997). Neuronal circuitry between the inferior mesenteric ganglion and lower intestine of the dog. Archives of Histology and Cytology, 60, 391–404. Li, Z.S. and Furness, J.B. (1998). Immunohistochemical localisation of cholinergic markers in putative intrinsic primary afferent neurons of the guinea-pig small intestine. Cell and Tissue Research, 294, 35–43. Li, Z.S. and Furness, J.B. (2000). Inputs from intrinsic primary afferent neurons to nitric oxide synthaseimmunoreactive neurons in the myenteric plexus of guinea-pig ileum. Cell and Tissue Research, 299, 1–8. Li, Z.S., Young, H.M. and Furness, J.B. (1995). Do vasoactive intestinal peptide (VIP)- and nitric oxide synthase-immunoreactive terminals synapse exclusively with VIP cell bodies in the submucous plexus of the guinea-pig ileum? Cell and Tissue Research, 281, 485–491. Llewellyn-Smith, I.J., Costa, M., Furness, J.B. and Bornstein, J.C. (1993). Structure of the tertiary component of the myenteric plexus in the guinea-pig small intestine. Cell and Tissue Research, 272, 509–516. Llewellyn-Smith, I.J., Furness, J.B., Gibbins, I.L. and Costa, M. (1988). Quantitative ultrastructural analysis of enkephalin-, substance P-, and VIP-immunoreactive nerve fibers in the circular muscle of the guinea-pig small intestine. Journal of Comparative Neurology, 272, 139–148. Lomax, A.E. and Furness, J.B. (2000). Neurochemical classification of enteric neurons in the guinea-pig distal colon. Cell and Tissue Research, 302, 59–72. Lomax, A.E., Sharkey, K.A., Bertrand, P.P., Low, A.M., Bornstein, J.C. and Furness, J.B. (1999). Correlation of morphology, electrophysiology and chemistry of neurons in the myenteric plexus of the guinea-pig distal colon. Journal of the Autonomic Nervous System, 76, 45–61. Lomax, A.E., Zhang, J.Y. and Furness, J.B. (2000). Origins of cholinergic inputs to the cell bodies of intestinofugal neurons in the guinea-pig distal colon. Journal of Comparative Neurology, 416, 451–460. Luckensmeyer, G.B. and Keast, J.R. (1995). Distribution and morphological characterization of viscerofugal projections from the large intestine to the inferior mesenteric and pelvic ganglia of the male rat. Neuroscience, 66, 663–671. Luckensmeyer, G.B. and Keast, J.R. (1996). Immunohistochemical characterisation of viscerofugal neurons projecting to the inferior mesenteric and major pelvic ganglia in the male rat. Journal of the Autonomic Nervous System, 61, 6–16.

CELLULAR ORGANISATION OF THE MAMMALIAN ENTERIC NERVOUS SYSTEM

461

Maggi, C.A., Holzer, P. and Giuliani, S. (1994). Effect of omega-conotoxin on cholinergic and tachykininergic excitatory neurotransmission to the circular muscle of the guinea-pig colon. Naunyn Schmiedebergs Archives of Pharmacology, 350, 529–536. Mann, R. and Bell, C. (1993). Distribution and origin of aminergic neurones in dog small intestine. Journal of the Autonomic Nervous System, 43, 107–115. Mann, P.T., Furness, J.B., Pompolo, S. and Mader, M. (1995). Chemical coding of neurons that project from different regions of intestine to the coeliac ganglion of the guinea-pig. Journal of the Autonomic Nervous System, 56, 15–25. Mann, P.T., Furness, J.B. and Southwell, B.R. (1999). Choline acetyltransferase immunoreactivity of putative intrinsic primary afferent neurons in the rat ileum. Cell and Tissue Research, 297, 241–248. Mann, P.T., Southwell, B.R., Young, H.M. and Furness, J.B. (1997). Appositions made by axons of descending interneurons in the guinea-pig small intestine, investigated by confocal microscopy. Journal of Chemical Neuroanatomy, 12, 151–164. Mashimo, H. and Goyal, R.K. (1999). Lessons from genetically engineered animal models. IV. Nitric oxide synthase gene knockout mice. American Journal of Physiology, 277, G745–G750. Mashimo, H., He, X.D., Huang, P.L., Fishman, M.C. and Goyal, R.K. (1996). Neuronal constitutive nitric oxide synthase is involved in murine enteric inhibitory neurotransmission. Journal of Clinical Investigation, 98, 8–13. Mashimo, H., Kjellin, A. and Goyal, R.K. (2000). Gastric stasis in neuronal nitric oxide synthase-deficient knockout mice. Gastroenterology, 119, 766–773. Mawe, G.M., Schemann, M., Wood, J.D. and Gershon, M.D. (1989). Immunocytochemical analysis of potential neurotransmitters present in the myenteric plexus and muscular layers of the corpus of the guinea-pig stomach. Anatomical Record, 224, 431–442. McRorie, J., Krier, J. and Adams, T. (1991). Morphology and projections of myenteric neurons to colonic fiber bundles of the cat. Journal of the Autonomic Nervous System, 32, 205–215. Meedeniya, A.C.B., Brookes, S.J.H., Hennig, G.W. and Costa, M. (1998). The projections of 5-Hydroxytryptamine-accumulating neurones in the myenteric plexus of the small intestine of the guinea-pig. Cell and Tissue Research, 291, 375–384. Messenger, J.P. (1993). Immunohistochemical analysis of neurons and their projections in the proximal colon of the guinea-pig. Archives of Histology and Cytology, 56, 459–474. Messenger, J.P., Bornstein, J.C. and Furness, J.B. (1994). Electrophysiological and morphological classification of myenteric neurons in the proximal colon of the guinea-pig. Neuroscience, 60, 227–244. Messenger, J.P. and Furness, J.B. (1990). Projections of chemically-specified neurons in the guinea-pig colon. Archives of Histology and Cytology, 53, 467–495. Messenger, J.P. and Furness, J.B. (1992). Distribution of enteric nerve cells that project to the coeliac ganglion of the guinea-pig. Cell and Tissue Research, 269, 119–132. Messenger, J.P. and Furness, J.B. (1993). Distribution of enteric nerve cells projecting to the superior and inferior mesenteric ganglia of the guinea-pig. Cell and Tissue Research, 271, 333–339. Michel, K., Reiche, D. and Schemann, M. (2000). Projections and neurochemical coding of motor neurones to the circular and longitudinal muscle of the guinea pig gastric corpus. Pflügers Archiv – European Journal of Physiology, 440, 393–408. Mihara, S. (1993). Intracellular recordings from neurones of the submucous plexus. Progress in Neurobiology, 40, 529–572. Mihara, S., Nishi, S., North, R.A. and Surprenant, A. (1987). A non-adrenergic, non-cholinergic slow inhibitory post-synaptic potential in neurones of the guinea-pig submucous plexus. Journal of Physiology, 390, 357–365. Miller, A.S., Furness, J.B. and Costa, M. (1989). The relationship between gastrin cells and bombesin-like immunoreactive nerve fibres in the gastric antral mucosa of guinea-pig, rat, dog and man. Cell and Tissue Research, 257, 171–178. Morita, K. and Katayama, Y. (1992). Substance P inhibits activation of calcium-dependent potassium conductances in guinea-pig myenteric neurones. Journal of Physiology, 447, 293–308. Morita, K., North, R.A. and Katayama, Y. (1980). Evidence that substance P is a neurotransmitter in the myenteric plexus. Nature, 287, 151–152. Nada, O. and Toyohara, T. (1987). An immunohistochemical study of serotonin-containing nerves in the colon of rats. Histochemistry, 86, 229–232. Nakao, K., Takahashi, T., Utsunomiya, J. and Owyang, C. (1998). Extrinsic neural control of nitric oxide synthase expression in the myenteric plexus of rat jejunum. Journal of Physiology, 507, 549–560. Neild, T.O., Shen, K.Z. and Surprenant, A. (1990). Vasodilatation of arterioles by acetylcholine released from single neurones in the guinea-pig submucosal plexus. Journal of Physiology, 420, 247–265. Nemeth, P.R., Ort, C.A. and Wood, J.D. (1984). Intracellular study of effects of histamine on electrical behaviour of myenteric neurones in guinea-pig small intestine. Journal of Physiology, 355, 411–425.

462

INNERVATION OF THE GASTROINTESTINAL TRACT

Nemeth, P.R., Palmer, J.M., Wood, J.D. and Zafirov, D.H. (1986). Effects of forskolin on electrical behaviour of myenteric neurones in guinea-pig small intestine. Journal of Physiology, 376, 439–450. Nemeth, P.R., Zafirov, D.H. and Wood, J.D. (1985). Effects of cholecystokinin, caerulein and pentagastrin on electrical behavior of myenteric neurons. European Journal of Pharmacology, 116, 263–269. Neuhuber, W.L., Appelt, M., Polak, J.M., Baier-Kustermann, W., Abelli, L. and Ferri, G.L. (1993). Rectospinal neurons: cell bodies, pathways, immunocytochemistry and ultrastructure. Neuroscience, 56, 367–378. Neunlist, M., Dobreva, G. and Schemann, M. (1999). Characteristics of mucosally projecting myenteric neurones in the guinea-pig proximal colon. Journal of Physiology, 517, 533–546. Neunlist, M., Frieling, T., Rupprecht, C. and Schemann, M. (1998). Polarized enteric submucosal circuits involved in secretory responses of the guinea-pig proximal colon. Journal of Physiology, 506, 539–550. Neunlist, M. and Schemann, M. (1997). Projections and neurochemical coding of myenteric neurons innervating the mucosa of the guinea-pig proximal colon. Cell and Tissue Research, 287, 119–125. Neunlist, M. and Schemann, M. (1998). Polarised innervation pattern of the mucosa of the guinea-pig distal colon. Neuroscience Letters, 246, 161–164. Nichols, K., Staines, W. and Krantis, A. (1994). Neural sites of the human colon colocalize nitric oxide synthaserelated NADPH diaphorase activity and neuropeptide Y. Gastroenterology, 107, 968–975. Niel, J.P., Bywater, R.A. and Taylor, G.S. (1983). Apamin-resistant post-stimulus hyperpolarization in the circular muscle of the guinea-pig ileum. Journal of the Autonomic Nervous System, 9, 565–569. Nishi, S. and North, R.A. (1973). Intracellular recording from the myenteric plexus of the guinea-pig ileum. Journal of Physiology, 231, 471–491. North, R.A. (1973). The calcium-dependent slow after-hyperpolarization in myenteric plexus neurones with tetrodotoxin-resistant action potentials. British Journal of Pharmacology, 49, 709–711. North, R.A., Henderson, G., Katayama, Y. and Johnson, S.M. (1980). Electrophysiological evidence for presynaptic inhibition of acetylcholine release by 5-hydroxytryptamine in the enteric nervous system. Neuroscience, 5, 581–586. North, R.A., Katayama, Y. and Williams, J.T. (1979). On the mechanism and site of action of enkephalin on single myenteric neurons. Brain Research, 165, 67–77. North, R.A. and Surprenant, A. (1985). Inhibitory synaptic potentials resulting from alpha 2-adrenoceptor activation in guinea-pig submucous plexus neurones. Journal of Physiology, 358, 17–33. North, R.A. and Tokimasa, T. (1987). Persistent calcium-sensitive potassium current and the resting properties of guinea-pig myenteric neurones. Journal of Physiology, 386, 333–353. Osthaus, L.E. and Galligan, J.J. (1992). Antagonists of nitric oxide synthesis inhibit nerve-mediated relaxations of longitudinal muscle in guinea-pig ileum. Journal of Pharmacology and Experimental Therapeutics, 260, 140–145. Palmer, J.M., Schemann, M., Tamura, K. and Wood, J.D. (1986). Calcitonin gene-related peptide excites myenteric neurons. European Journal of Pharmacology, 132, 163–170. Palmer, J.M., Wood, J.D. and Zafirov, D.H. (1986). Elevation of adenosine 3′,5′-phosphate mimics slow synaptic excitation in myenteric neurones of the guinea-pig. Journal of Physiology, 376, 451–460. Palmer, J.M., Wood, J.D. and Zafirov, D.H. (1987). Purinergic inhibition in the small intestinal myenteric plexus of the guinea-pig. Journal of Physiology, 387, 357–369. Pan, H. and Gershon, M.D. (2000). Activation of intrinsic afferent pathways in submucosal ganglia of the guinea-pig small intestine. Journal of Neuroscience, 20, 3295–3309. Parkman, H.P., Stapelfeldt, W.H., Williams, C.L., Lennon, V.A. and Szurszewski, J.H. (1993). Enteric GABAcontaining nerves projecting to the guinea-pig inferior mesenteric ganglion modulate acetylcholine release. Journal of Physiology, 471, 191–207. Pataky, D.M., Curtis, S.B. and Buchan, A.M. (1990). The co-localization of neuropeptides in the submucosa of the small intestine of normal Wistar and non-diabetic BB rats. Neuroscience, 36, 247–254. Paton, W.D.M. (1955). The response of the guinea-pig ileum to electrical stimulation by coaxial electrodes. Journal of Physiology, 127, 40P–41P. Pearse, A.G.E. (1969). The cytochemistry and ultrastructure of polypeptide hormone-producing cells of the APUD series and the embyrologic, physiologic and pathologic implications of the concept. Journal of Histochemistry and Cytochemistry, 17, 303–313. Pearson, G.T. (1994). Structural organization and neuropeptide distributions in the equine enteric nervous system: an immunohistochemical study using whole-mount preparations from the small intestine. Cell and Tissue Research, 276, 523–534. Pfannkuche, H., Firzlaff, U., Sann, H., Reiche, D. and Schemann, M. (2000). Neurochemical coding and projection patterns of gastrin-releasing peptide-immunoreactive myenteric neurone subpopulations in the guineapig gastric fundus. Journal of Chemical Neuroanatomy, 19, 93–104. Pfannkuche, H., Reiche, D., Firzlaff, U., Sann, H. and Schemann, M. (1998a). Enkephalin-immunoreactive subpopulations in the myenteric plexus of the guinea-pig fundus project primarily to the muscle and not to the mucosa. Cell and Tissue Research, 294, 45–55.

CELLULAR ORGANISATION OF THE MAMMALIAN ENTERIC NERVOUS SYSTEM

463

Pfannkuche, H., Reiche, D., Sann, H. and Schemann, M. (1998b). Different subpopulations of cholinergic and nitrergic myenteric neurones project to mucosa and circular muscle of the guinea-pig gastric fundus. Cell and Tissue Research, 292, 463–475. Pompolo, S. and Furness, J.B. (1995). Sources of inputs to longitudinal muscle motor neurons and ascending interneurons in the guinea-pig small intestine. Cell and Tissue Research, 280, 549–560. Pompolo, S. and Furness, J.B. (1998). Quantitative analysis of inputs to somatostatin-immunoreactive descending interneurons in the myenteric plexus of the guinea-pig small intestine. Cell and Tissue Research, 294, 219–226. Portbury, A.L., McConalogue, K., Furness, J.B. and Young, H.M. (1995a). Distribution of pituitary adenylyl cyclase activating peptide (PACAP) immunoreactivity in neurons of the guinea-pig digestive tract and their projections in the ileum and colon. Cell and Tissue Research, 279, 385–392. Portbury, A.L., Pompolo, S., Furness, J.B., Stebbing, M.J., Kunze, W.A.A., Bornstein, J.C., et al. (1995b). Cholinergic, somatostatin-immunoreactive interneurons in the guinea-pig intestine – morphology, ultrastructure, connections and projections. Journal of Anatomy, 187, 303–321. Porter, A.J., Wattchow, D.A., Brookes, S.J.H., Schemann, M. and Costa, M. (1996). Choline acetyltransferase immunoreactivity in the human small and large intestine. Gastroenterology, 111, 401–408. Porter, A.J., Wattchow, D.A., Brookes, S.J. and Costa, M. (1997). The neurochemical coding and projections of circular muscle motor neurons in the human colon. Gastroenterology, 113, 1916–1923. Porter, A.J., Wattchow, D.A., Brookes, S.J.H. and Costa, M. (1999). Projections of nitric oxide synthase and vasoactive intestinal polypeptide-reactive submucosal neurons in the human colon. Journal of Gastroenterology and Hepatology, 14, 1180–1187. Porter, A.J., Wattchow, D.A., Brookes, S.J.H. and Costa, M. (2002). Cholinergic and nitrergic neurons in the myenteric plexus of the human colon. Gut (in press). Preiksaitis, H.G. and Diamant, N.E. (1997). Regional differences in cholinergic activity of muscle fibers from the human gastroesophageal junction. American Journal of Physiology, 35, G1321–G1327. Puig, M.M., Gascon, P., Craviso, G.L. and Musacchio, J.M. (1977). Endogenous opiate receptor ligand: electrically induced release in the guinea-pig ileum. Science, 195, 419–420. Reiche, D., Pfannkuche, H., Michel, K., Hoppe, S. and Schemann, M. (1999). Immunohistochemical evidence for the presence of calbindin containing neurones in the myenteric plexus of the guinea-pig stomach. Neuroscience Letters, 270, 71–74. Reiche, D. and Schemann, M. (1998). Ascending choline acetyltransferase and descending nitric oxide synthase immunoreactive neurones of the myenteric plexus project to the mucosa of the guinea-pig gastric corpus. Neuroscience Letters, 241, 61–64. Reiche, D. and Schemann, M. (1999). Mucosa of the guinea-pig gastric corpus is innervated by myenteric neurones with specific neurochemical coding and projection preferences. Journal of Comparative Neurology, 410, 489–502. Ritter, R.C., Costa, M. and Brookes, S.H. (1997). Nuclear fos immunoreactivity in guinea-pig myenteric neurons following activation of motor activity. American Journal of Physiology, 273, G498–G507. Sanders, K.M. and Smith, T.K. (1986). Motoneurones of the submucous plexus regulate electrical activity of the circular muscle of canine proximal colon. Journal of Physiology, 380, 293–310. Sanders, K.M. and Ward, S.M. (1992). Nitric oxide as a mediator of nonadrenergic noncholinergic neurotransmission. American Journal of Physiology, 262, G379–G392. Sang, Q., Williamson, S. and Young, H.M. (1997). Projections of chemically identified myenteric neurons of the small and large intestine of the mouse. Journal of Anatomy, 190, 209–222. Sang, Q. and Young, H.M. (1996). Chemical coding of neurons in the myenteric plexus and external muscle of the small and large intestine of the mouse. Cell and Tissue Research, 284, 39–53. Sang, Q. and Young, H.M. (1998a). The identification and chemical coding of cholinergic neurons in the small and large intestine of the mouse. Anatomical Record, 251, 185–199. Sang, Q. and Young, H.M. (1998b). The origin and development of the vagal and spinal innervation of the external muscle of the mouse esophagus. Brain Research, 809, 253–268. Schemann, M. and Schaaf, C. (1995). Differential projection of cholinergic and nitroxidergic neurons in the myenteric plexus of guinea-pig stomach. American Journal of Physiology, 269, G186–G195. Schemann, M., Schaaf, C. and Mader, M. (1995). Neurochemical coding of enteric neurons in the guinea-pig stomach. Journal of Comparative Neurology, 353, 161–178. Schemann, M. and Tamura, K. (1992). Presynaptic inhibitory effects of the peptides NPY, PYY and PP on nicotinic EPSPs in guinea-pig gastric myenteric neurones. Journal of Physiology, 451, 79–89. Schemann, M. and Wood, J.D. (1989). Electrical behaviour of myenteric neurones in the gastric corpus of the guinea-pig. Journal of Physiology, 417, 501–518. Scheuermann, D.W., Krammer, H.J., Timmermans, J.P., Stach, W., Adriaensen, D. and De Groodt-Lasseel, M.H. (1991). Fine structure of morphologically well-defined type II neurons in the enteric nervous system

464

INNERVATION OF THE GASTROINTESTINAL TRACT

of the porcine small intestine revealed by immunoreactivity for calcitonin gene-related peptide. Acta Anatomica, 142, 236–241. Schultzberg, M., Hökfelt, T., Nilsson, G., Terenius, L., Rehfeld, J.F., Brown, M., Eldel, R., Goldstein, M. and Said, S. (1980). Distribution of peptide- and catecholamine-containing neurons in the gastrointestinal tract of rat and guinea-pig: immunohistochemical studies with antisera to substance P, vasoactive intestinal polypeptide, enkephalins, somatostatin, gastrin/cholecystokinin, neurotensin and dopamine β-hydroxylase. Neuroscience, 5, 689–744. Schutte, I., Kroese, A. and Akkermans, L. (1995). Somal size and location within the ganglia for electrophysiologically identified myenteric neurons of the guinea-pig ileum. Journal of Comparative Neurology, 355, 563–572. Schutte, I.W., Hollestein, K.B., Akkermans, L.M. and Kroese, A.B. (1997). Evidence for a role of cholecystokinin as neurotransmitter in the guinea-pig enteric nervous system. Neuroscience Letters, 236, 155–158. Selemidis, S., Satchell, D.G. and Cocks, T.M. (1997). Evidence that NO acts as a redundant NANC inhibitory neurotransmitter in the guinea-pig isolated taenia coli. British Journal of Pharmacology, 121, 604–611. Sharkey, K.A., Lomax, A.E., Bertrand, P.P. and Furness, J.B. (1998). Electrophysiology, shape, and chemistry of neurons that project from guinea-pig colon to inferior mesenteric ganglia. Gastroenterology, 115, 909–918. Shen, K.Z. and Surprenant, A. (1993). Somatostatin-mediated inhibitory postsynaptic potential in sympathetically denervated guinea-pig submucosal neurones. Journal of Physiology, 470, 619–635. Shuba, M.F. and Vladimirova, I.A. (1980). Effect of apamin on the electrical responses of smooth muscle to adenosine 5-triphosphate and to non-adrenergic, non-cholinergic nerve stimulation. Neuroscience, 5, 853–859. Smith, T.K., Bornstein, J.C. and Furness, J.B. (1990). Distension-evoked ascending and descending reflexes in the circular muscle of guinea-pig ileum: an intracellular study. Journal of the Autonomic Nervous System, 29, 203–217. Smith, T.K., Burke, E.P. and Shuttleworth, C.W. (1999). Topographical and electrophysiological characteristics of highly excitable S neurones in the myenteric plexus of the guinea-pig ileum. Journal of Physiology, 517, 817–830. Smith, T.K. and Furness, J.B. (1988). Reflex changes in circular muscle activity elicited by stroking the mucosa: an electrophysiological analysis in the isolated guinea-pig ileum. Journal of the Autonomic Nervous System, 25, 205–218. Smith, T.K., Furness, J.B., Costa, M. and Bornstein, J.C. (1988). An electrophysiological study of the projections of motor neurones that mediate non-cholinergic excitation in the circular muscle of the guinea-pig small intestine. Journal of the Autonomic Nervous System, 22, 115–128. Smith, T.K. and Robertson, W.J. (1998). Synchronous movements of the longitudinal and circular muscle during peristalsis in the isolated guinea-pig distal colon. Journal of Physiology, 506, 563–577. Song, Z.M., Brookes, S.J. and Costa, M. (1991). Identification of myenteric neurons which project to the mucosa of the guinea-pig small intestine. Neuroscience Letters, 129, 294–298. Song, Z.M., Brookes, S.J.H., Steele, P.A. and Costa, M. (1992). Projections and pathways of submucous neurons to the mucosa of the guinea-pig small intestine. Cell and Tissue Research, 269, 87–98. Song, Z.M., Brookes, S.J.H. and Costa, M. (1994a). All calbindin-immunoreactive myenteric neurons project to the mucosa of the guinea-pig small intestine. Neuroscience Letters, 180, 219–222. Song, Z.M., Brookes, S.J.H. and Costa, M. (1994b). Characterization of alkaline phosphatase-reactive neurons in the guinea-pig small intestine. Neuroscience, 63, 1153–1167. Song, Z.M., Brookes, S.J.H. and Costa, M. (1996). Projections of specific morphological types of neurons within the myenteric plexus of the small intestine of the guinea-pig. Cell and Tissue Research, 285, 149–156. Song, Z.M., Brookes, S.J.H., Ramsay, G.A. and Costa, M. (1997a). Characterization of myenteric interneurons with somatostatin immunoreactivity in the guinea-pig small intestine. Neuroscience, 80, 907–923. Song, Z.M., Brookes, S.J.H., Neild, T.O. and Costa, M. (1997b). Immunohistochemical and electrophysiological characterization of submucous neurons from the guinea-pig small intestine in organ culture. Journal of the Autonomic Nervous System, 63, 161–171. Song, Z.M., Costa, M. and Brookes, S.J.H. (1998). Projections of submucous neurons to the myenteric plexus in the guinea-pig small intestine. Journal of Comparative Neurology, 399, 255–268. Spencer, N., Walsh, M. and Smith, T.K. (1999). Does the guinea-pig ileum obey the “law of the intestine”. Journal of Physiology, 517, 889–898. Stach, W. (1989). A revised morphological classification of neurons in the enteric nervous system. In Nerves and the Gastrointestinal Tract, edited by M.V. Singer and H. Goebell, pp. 29–45. Lancaster, UK: Kluwer Academic Publishers. Stapelfeldt, W.H., Parkman, H.P. and Szurszewski, J.H. (1993). The electrophysiological effects of endogenous GABA in the guinea-pig inferior mesenteric ganglion. Journal of Physiology, 471, 175–189. Starodub, A.M. and Wood, J.D. (2000). A-type potassium current in myenteric neurons from guinea-pig small intestine. Neuroscience, 99, 389–396.

CELLULAR ORGANISATION OF THE MAMMALIAN ENTERIC NERVOUS SYSTEM

465

Stebbing, M.J. and Bornstein, J.C. (1994). Electrophysiological analysis of the convergence of peripheral inputs onto neurons of the coeliac ganglion in the guinea-pig. Journal of the Autonomic Nervous System, 46, 93–105. Steele, P.A., Brookes, S.J. and Costa, M. (1991). Immunohistochemical identification of cholinergic neurons in the myenteric plexus of guinea-pig small intestine. Neuroscience, 45, 227–239. Steele, P.A. and Costa, M. (1990). Opioid-like immunoreactive neurons in secretomotor pathways of the guineapig ileum. Neuroscience, 38, 771–786. Sternini, C. and Anderson, K. (1992). Calcitonin gene-related peptide-containing neurons supplying the rat digestive system: differential distribution and expression pattern. Somatosensory and Motor Research, 9, 45–59. Sternini, C., de Giorgio, R. and Furness, J.B. (1992). Calcitonin gene-related peptide neurons innervating the canine digestive system. Regulatory Peptides, 42, 15–26. Sternini, C., Wong, H., Wu, S.V., de Giorgio, R., Yang, M., Reeve, Jr., J. et al. (1997). Somatostatin 2A receptor is expressed by enteric neurons, and by interstitial cells of Cajal and enterochromaffin-like cells of the gastrointestinal tract. Journal of Comparative Neurology, 386, 396–408. Su, H.C., Bishop, A.E., Power, R.F., Hamada, Y. and Polak, J.M. (1987). Dual intrinsic and extrinsic origins of CGRP- and NPY-immunoreactive nerves of rat gut and pancreas. Journal of Neuroscience, 7, 2674–2687. Swain, M.G., Blennerhassett, P.A. and Collins, S.M. (1991). Impaired sympathetic nerve function in the inflamed rat intestine. Gastroenterology, 100, 675–682. Szurszewski, J.H. and Miller, S.M. (1994). Physiology of prevertebral ganglia. In Physiology of the Gastrointestinal Tract, edited by L.R. Johnson, pp. 795–878. New York: Raven Press. Tack, J.F. and Wood, J.D. (1992). Electrical behaviour of myenteric neurones in the gastric antrum of the guineapig. Journal of Physiology, 447, 49–66. Tamura, K. (1992). Morphology of electrophysiologically identified myenteric neurons in the guinea-pig rectum. American Journal of Physiology, 262, G545–G552. Tamura, K. (1997). Synaptic inputs to morphologically identified myenteric neurons in guinea-pig rectum from pelvic nerves. American Journal of Physiology, 273, G49–G55. Tamura, K., Palmer, J.M., Winkelmann, C.K. and Wood, J.D. (1988). Mechanism of action of galanin on myenteric neurons. Journal of Neurophysiology, 60, 966–979. Tamura, K., Schemann, M. and Wood, J.D. (1993). Actions of nitric oxide-generating sodium nitroprusside in myenteric plexus of guinea-pig small intestine. American Journal of Physiology, 265, G887–G893. Tamura, K. and Wood, J.D. (1989). Electrical and synaptic properties of myenteric plexus neurones in the terminal large intestine of the guinea-pig. Journal of Physiology, 415, 275–298. Taraviras, S. and Pachnis, V. (1999). Development of the mammalian enteric nervous system. Current Opinion in Genetics and Development, 9, 321–327. Tassicker, B.C., Hennig, G.W., Costa, M. and Brookes, S.J. (1999). Rapid anterograde and retrograde tracing from mesenteric nerve trunks to the guinea-pig small intestine in vitro. Cell and Tissue Research, 295, 437– 452. Thomsen, L., Pearson, G.T., Larsen, E.H. and Skadhauge, E. (1997). Electrophysiological properties of neurones in the internal and external submucous plexuses of newborn pig small intestine. Journal of Physiology, 498, 773–785. Timmermans, J.P., Barbiers, M., Scheuermann, D.W., Bogers, J.J., Adriaensen, D., Fekete, E. et al. (1994). Nitric oxide synthase immunoreactivity in the enteric nervous system of the developing human digestive tract. Cell and Tissue Research, 275, 235–245. Timmermans, J.P., Barbiers, M., Scheuermann, D.W., Stach, W., Adriaensen, D. and De Groodt-Lasseel, M.H. (1993). Occurrence, distribution and neurochemical features of small intestinal neurons projecting to the cranial mesenteric ganglion in the pig. Cell and Tissue Research, 272, 49–58. Timmermans, J.P., Scheuermann, D.W., Barbiers, M., Adriaensen, D., Stach, W., Van Hee, R. et al. (1992). Calcitonin gene-related peptide-like immunoreactivity in the human small intestine. Acta Anatomica, 143, 48–53. Timmermans, J.P., Scheuermann, D.W., Stach, W., Adriaensen, D. and De Groodt-Lasseel, M.H. (1990). Distinct distribution of CGRP-, enkephalin-, galanin-, neuromedin U-, neuropeptide Y-, somatostatin-, substance P-, VIP- and serotonin-containing neurons in the two submucosal ganglionic neural networks of the porcine small intestine. Cell and Tissue Research, 260, 367–379. Tonini, M. and Costa, M. (1990). A pharmacological analysis of the neuronal circuitry involved in distensionevoked enteric excitatory reflex. Neuroscience, 38, 787–795. Trendelenburg, P. (1917). Physiologische und pharmakologische Versuche über die Dunndarmperistaltik. Naunyn-Schmiedeberg Archiv Experimental Pathologie und Pharmakologie, 81, 55–129. Tsuji, S., Anglade, P., Ozaki, T., Sazi, T. and Yokoyama, S. (1992). Peristaltic movement evoked in intestinal tube devoid of mucosa and submucosa. Japanese Journal of Physiology, 42, 363–375. Uemura, S., Pompolo, S. and Furness, J.B. (1995). Colocalization of neuropeptide y with other neurochemical markers in the guinea-pig small intestine. Archives of Histology and Cytology, 58, 523–536.

466

INNERVATION OF THE GASTROINTESTINAL TRACT

Wade, P.R. and Wood, J.D. (1988). Synaptic behavior of myenteric neurons in guinea-pig distal colon. American Journal of Physiology, 255, G184–G190. Ward, S.M., Xue, C., Shuttleworth, C.W., Bredt, D.S., Snyder, S.H. and Sanders, K.M. (1992). NADPH diaphorase and nitric oxide synthase colocalization in enteric neurons of canine proximal colon. American Journal of Physiology, 263, G277–G284. Wardell, C.F., Bornstein, J.C. and Furness, J.B. (1994). Projections of 5-hydroxytryptamine-immunoreactive neurons in guinea-pig distal colon. Cell and Tissue Research, 278, 379–387. Waterman, S.A., Costa, M. and Tonini, M. (1992). Modulation of peristalsis in the guinea-pig isolated small intestine by exogenous and endogenous opioids. British Journal of Pharmacology, 106, 1004–1010. Waterman, S.A., Tonini, M. and Costa, M. (1994). The role of ascending excitatory and descending inhibitory pathways in peristalsis in the isolated guinea-pig small intestine. Journal of Physiology, 481, 223–232. Wattchow, D.A., Brookes, S.J. and Costa, M. (1995). The morphology and projections of retrogradely labeled myenteric neurons in the human intestine. Gastroenterology, 109, 866–875. Wattchow, D.A., Furness, J.B. and Costa, M. (1988). Distribution and coexistence of peptides in nerve fibers of the external muscle of the human gastrointestinal tract. Gastroenterology, 95, 32–41. Wattchow, D.A., Porter, A.J., Brookes, S.J. and Costa, M. (1997). The polarity of neurochemically defined myenteric neurons in the human colon. Gastroenterology, 113, 497–506. Weihe, E., Tao Cheng, J.H., Schafer, M.K., Erickson, J.D. and Eiden, L.E. (1996). Visualization of the vesicular acetylcholine transporter in cholinergic nerve terminals and its targeting to a specific population of small synaptic vesicles. Proceedings of the National Academy of Sciences USA, 93, 3547–3552. Williamson, S., Pompolo, S. and Furness, J.B. (1996). GABA and nitric oxide synthase immunoreactivities are colocalized in a subset of inhibitory motor neurons of the guinea-pig small intestine. Cell and Tissue Research, 284, 29–37. Wilson, A.J., Llewellyn Smith, I.J., Furness, J.B. and Costa, M. (1987). The source of the nerve fibres forming the deep muscular and circular muscle plexuses in the small intestine of the guinea-pig. Cell and Tissue Research, 247, 497–504. Wood, J. (1989). Electrical and synaptic behavior of enteric neurons. In Handbook of Physiology, edited by J. Wood, pp. 465–517. Bethesda: Waverly Press. Wood, J.D. (1994a). Application of classification schemes to the enteric nervous system. Journal of the Autonomic Nervous System, 48, 17–29. Wood, J.D. (1994b). Physiology of the enteric nervous system. In Physiology of the Gastrointestinal Tract, edited by L.R. Johnson, pp. 423–482. New York: Raven Press. Wood, J.D. and Mayer, C.J. (1978). Slow synaptic excitation mediated by serotonin in Auerbach’s plexus. Nature, 276, 836–837. Wood, J.D. and Mayer, C.J. (1979). Serotonergic activation of tonic-type enteric neurons in guinea-pig small bowel. Journal of Neurophysiology, 42, 582–593. Wood, J.D., Mayer, C.J., Ninchoji, T. and Erwin, D.N. (1979). Effects of depleted calcium on electrical activity of neurons in Auerbach’s plexus. American Journal of Physiology, 236, C78–C86. Woodbury, J.W. and Brady, A.J. (1956). Intracellular recordings from moving tissues with a flexibly mounted microelectrode. Science, 123, 100–101. Xia, Y., Fertel, R.H. and Wood, J.D. (1997). Suppression of cAMP formation by adenosine in myenteric ganglia from guinea-pig small intestine. European Journal of Pharmacology, 320, 95–101. Yokoyama, S., Tsuji, S., Anglade, P. and Ozaki, T. (1990). Peristalsis evoked in the intestine tube devoid of mucosa and submucosa. Progress in Clinical and Biological Research, 327, 493–497. Young, H.M. and Ciampoli, D. (1998). Transient expression of neuronal nitric oxide synthase by neurons of the submucous plexus of the mouse small intestine. Cell and Tissue Research, 291, 395–401. Young, H.M. and Furness, J.B. (1995). Ultrastructural examination of the targets of serotonin-immunoreactive descending interneurons in the guinea-pig small intestine. Journal of Comparative Neurology, 356, 101–114. Young, H.M., Furness, J.B. and Povey, J.M. (1995). Analysis of connections between nitric oxide synthase neurons in the myenteric plexus of the guinea-pig small intestine. Journal of Neurocytology, 24, 257–263. Young, H.M., Furness, J.B., Sewell, P., Burcher, E.F. and Kandiah, C.J. (1993a). Total numbers of neurons in myenteric ganglia of the guinea-pig small intestine. Cell and Tissue Research, 272, 197–200. Young, H.M., Furness, J.B., Shuttleworth, C.W., Bredt, D.S. and Snyder, S.H. (1992). Co-localization of nitric oxide synthase immunoreactivity and NADPH diaphorase staining in neurons of the guinea-pig intestine. Histochemistry, 97, 375–373. Young, H.M., McConalogue, K., Furness, J.B. and Devente, J. (1993b). Nitric oxide targets in the guinea-pig intestine identified by induction of cyclic gmp immunoreactivity. Neuroscience, 55, 583–596. Yuan, S.Y. and Brookes, S.J.H. (1999). Neuronal control of the gastric sling muscle of the guinea-pig. Journal of Comparative Neurology, 412, 669–680.

CELLULAR ORGANISATION OF THE MAMMALIAN ENTERIC NERVOUS SYSTEM

467

Yunker, A.M. and Galligan, J.J. (1994). Extrinsic denervation increases NADPH diaphorase staining in myenteric nerves of guinea-pig ileum. Neuroscience Letters, 167, 51–54. Yunker, A.M. and Galligan, J.J. (1998). Extrinsic denervation increases myenteric nitric oxide synthase-containing neurons and inhibitory neuromuscular transmission in guinea-pig. Journal of the Autonomic Nervous System, 71, 148–158. Zafirov, D.H., Palmer, J.M., Nemeth, P.R. and Wood, J.D. (1985a). Bombesin, gastrin-releasing peptide and vasoactive intestinal peptide excite myenteric neurons. European Journal of Pharmacology, 115, 103–107. Zafirov, D.H., Palmer, J.M., Nemeth, P.R. and Wood, J.D. (1985b). Cyclic 3′,5′-adenosine monophosphate mimics slow synaptic excitation in myenteric plexus neurons. Brain Research, 347, 368–371. Zagorodnyuk, V.P. and Brookes, S.J. (2000). Transduction sites of vagal mechanoreceptors in the guinea-pig esophagus. Journal of Neuroscience, 20, 6249–6255. Zhou, X. and Galligan, J.J. (2000). GABAA receptors on calbindin-immunoreactive myenteric neurons of guinea-pig intestine. Journal of the Autonomic Nervous System, 78, 122–135.

11 Development of the Enteric Nervous System Michael D. Gershon Department of Anatomy and Cell Biology, Columbia University, College of Physicians and Surgeons, 630 W 168th Street, New York, NY 10032, USA During development, the gut is colonised by three populations of neuronal progenitors originating in the rostral truncal, vagal and sacral regions of the neural crest. Knockout of the mash-1 transcription factor leads to a complete lack of enteric ganglia in the oesophagus, by blocking colonisation by rostral truncal neural crest cells. In the rest of the gut, there is a selective loss of mash-1-sensitive serotonergic nerve cells but other cell types, for example those containing CGRP, are apparently unaffected. In contrast, knockout of GDNF, or its receptor, Ret, prevents enteric innervation of the entire gut below the oesophagus and forestomach. The neurotrophin, NT-3, also influences enteric neuronal development, at least in part by interacting with an unidentified cytokine that activates the CNTF receptorα. A localised aganglionosis of the distal colon occurs in mice lacking functional endothelin-3 or its receptor ETB. This may be due to an overproduction of the α subunit of Laminin-1 in this region, which promotes premature differentiation and failure to colonise by neuronal precursors. Hirschsprung’s disease in humans is sometimes associated with mutations of RET, ETB and ET-3, but other mechanisms are also likely to contribute. It is likely that the existence and causes of more subtle enteric neuropathies will become apparent as our understanding of enteric neuronal development continues to grow.

INTRODUCTION The unique nature of the enteric nervous system (ENS) is now well understood (Furness and Costa, 1987; Gershon, Kirchgessner and Wade, 1994). The many properties of the ENS that distinguish it from the remainder of the peripheral nervous system (PNS) have become known and are widely accepted. These properties include some that are structural, some that are chemical, and others that are functional. Among the unique structural properties of the ENS are the absence of collagen from the interiors of enteric ganglia, and the support of enteric neural elements by glia, rather than by Schwann cells (Gabella, 1971, 1972; Jessen and Mirsky, 1980; Gabella and Trigg, 1984; Gershon and Rothman, 1991). The ENS thus resembles the brain more than it does other elements of the PNS. The shapes of enteric neurons, moreover, are readily classifiable and characteristic of the bowel (Dogiel, 1899; Song, Brookes and Costa, 1996). Chemically, enteric neurons display the greatest degree of phenotypic diversity found in the PNS. The abundance of 469

470

INNERVATION OF THE GASTROINTESTINAL TRACT

enteric neurotransmitters and neuromodulators and the multiple combinations of the coincident storage of these molecules in enteric neurons have provided a means of coding various types of enteric neuron so that they can reproducibly be identified (Furness and Costa, 1987; Furness et al., 1989; Bornstein and Furness, 1992; Costa et al., 1992, 1996; Gershon, Kirchgessner and Wade, 1994; Schemann, Schaaf and Mader, 1995; Sang and Young, 1996). The chemical code and the shapes of enteric neurons have been related to their electrophysiological properties to provide the neurons of the ENS with an identification scheme that is unrivalled outside of the central nervous system (CNS) (Bornstein, Furness and Kunze, 1994; Wood, 1994; Costa et al., 1996). Functionally, the ability of the ENS to manifest motor and secretory reflex activity in the absence of input from the CNS is its most striking point of difference from the rest of the PNS (Furness and Costa, 1987; Cooke, 1989; Gershon, Kirchgessner and Wade, 1994). This capability implies that the ENS contains intrinsic sensory neurons (Song, Brookes and Costa, 1991, 1994; Kirchgessner, Tamir and Gershon, 1992; Kunze, Bornstein and Furness, 1995; Kirchgessner, Liu and Gershon, 1996) and interneurons (Brookes, Steele and Costa, 1991a; Kunze, Furness and Bornstein, 1993), in addition to the excitatory and inhibitory motor neurons (Brookes and Costa, 1990; Brookes, Mayo and Costa, 1992; Brookes et al., 1992) that innervate gastrointestinal effectors. The ENS also sends projections outside of the gut, to prevertebral sympathetic ganglia (Crowcroft, Holman and Szurszewski, 1971; Kreulen and Szurszewski, 1979; Szurszewski, 1981), as well as to ganglia in accessory digestive organs, such as the pancreas (Kirchgessner and Gershon, 1990, 1995) and gall bladder (Mawe and Gershon, 1989). The fact that the ENS is unlike any other region of the PNS became apparent many years ago, but then was eclipsed, probably by the fascination engendered by the discovery that acetylcholine (ACh) and noradrenaline (NA) are autonomic neurotransmitters (Kuntz, 1953). The initial knowledge of the special properties of the ENS was provided by the classical studies of Bayliss and Starling who first described the peristaltic reflex, which they called the “law of the intestine”, and showed that it was mediated by the “local nervous mechanism” of the bowel (Bayliss and Starling, 1899, 1900). Their work was confirmed by Trendelenburg (1917), who demonstrated that the peristaltic reflex could actually be elicited in a loop of guinea pig intestine suspended in vitro. The occurrence of the reflex in an isolated segment of gut, a preparation that clearly lacks any connection to the CNS, and which contains no dorsal root or cranial nerve ganglia, proves absolutely that the complete reflex arc is intrinsic to the ENS. Langley, who is responsible for the accepted definition of the autonomic nervous system, did not include the ENS in either its sympathetic or parasympathetic divisions (Langley, 1921). Instead, Langley considered the ENS a third autonomic division. Langley defined the sympathetic nervous system by its thoracic and lumbar outflows of preganglionic nerve fibres and the parasympathetic division by its cranial and sacral outflows. He realised that the number of motor fibres in the vagus nerves, as those nerves pass through the diaphragm is very small, while the number of ganglion cells in the ENS is very large. Langley did not think that it was likely that the small number of efferent vagal fibers could meaningfully innervate the much larger number of neurons in the vagally innervated portion of the bowel. The implication of this discrepancy between the number of vagal fibers and the number of ganglion cells is that many, if not most, enteric neurons do not receive a direct input from the CNS.

DEVELOPMENT OF THE ENTERIC NERVOUS SYSTEM

471

Neurons that do not receive input from one of the preganglionic outflows that define the sympathetic and parasympathetic divisions, cannot be defined as components of either. Subsequent studies have tended to confirm Langley’s supposition about the CNS innervation of enteric neurons (Kirchgessner and Gershon, 1989), although the vagus nerves appear to contact more neurons in the proximal than the distal bowel (Berthoud, Jedrzejewska and Powley, 1990; Berthoud and Powley, 1992; Fox and Powley, 1992; Schemann and Grundy, 1992; Powley et al., 1994). After Langley, a dark age of repression fell over the ENS. The idea that there were two postganglionic neurotransmitters in an autonomic nervous system that had two corresponding divisions appears to have been compelling. The separate classification of the ENS disappeared from use and enteric ganglia were thought of, if they were thought of at all, as the parasympathetic relay neurons in the vagal and sacral innervations of the bowel (Kuntz, 1913, 1953). The initial suggestion that 5-hydroxytryptamine (5-HT) (Gershon, Drakontides and Ross, 1965; Gershon and Ross, 1966a,b; Bulbring and Gershon, 1967; Rothman, Ross and Gershon, 1976), as well as ACh and NA, might be an enteric neurotransmitter was thus greeted sceptically (Costa and Furness, 1979). It took a revolution in immunocytochemistry that demonstrated that not just 5-HT (Costa et al., 1982; Furness and Costa, 1982), but an array of peptides were present in autonomic neurons (Dalsgaard et al., 1983; Furness and Costa, 1987), to make the phenotypic diversity of the ENS acceptable. The current realisation that the ENS is not a simple set of relay neurons and the recent flowering of research on the ENS thus represents, not a revolutionary new discovery, but a rediscovery of an old truth that had nearly been forgotten. Ecclesiastes was right, there is nothing new under the sun. For the ENS, therefore, this is the renaissance, not genesis. For one to appreciate the complex neurobiology of the development of the ENS, it is vital to understand the unique nature of this system. Since enteric neurons are different from other peripheral neurons, the ontogeny of these differences must be explained. Clearly, forces must be at work in the developing bowel and/or the sources of its neurons that are not at work elsewhere in the PNS. Given this imperative, the years during which knowledge of the nature of the ENS was in obscurity were years during which little useful could be learned about the development of the ENS. The modern explosion of information about enteric neuronal ontogeny thus roughly coincides with that of information about the mature ENS.

THE NEURAL CREST IS THE SOURCE OF ENTERIC NEURONAL AND GLIAL PROGENITORS The first step in understanding the development of a large and complex nervous system, such as the ENS, is to determine where the neurons and glia come from. It was not obvious to early workers in the field that enteric neurons and glia do not, in fact, arise de novo within the primitive digestive tube. Modern understanding of the development of the ENS thus dates from the work of Yntema and Hammond, who first showed that enteric ganglia are produced by émigrés, cells that migrate to the primordial gut from the neural crest (Yntema and Hammond, 1954, 1955). These investigators demonstrated that enteric

472

INNERVATION OF THE GASTROINTESTINAL TRACT

ganglia were missing from the bowel of chick embryos when what they called the “anterior” neural crest was deleted. More recently and precisely, the levels of the crest from which enteric neurons are derived were identified by Le Douarin and her colleagues using quail-chick interspecies chimeras (Le Douarin and Teillet, 1973, 1974). The nucleolar-associated heterochromatin in the nuclei of quail cells is so distinctive that it can be used as a stable marker that allows these cells to be accurately identified in the field of chick cells. In quail-chick chimeras, therefore, one can readily determine which cells are chick and which cells are quail simply by examining the nuclei in a microscopic field. To trace the migration of crest-derived cells, Le Douarin and her co-workers removed segments of the neural crest from chick embryos and replaced them with quail crest (or the reverse). When this was done, the foreign cells migrated in their new host and colonised organs in an apparently normal manner. The stably marked donor cells (chick or quail, depending on the particular experiment), however, are now identifiable in the targets of neural crest migration. Since only crest cells were transplanted to create the chimeras, the identification of donor cells in target organs revealed which of the cells in these organs were crest-derived. By transplanting small segments of crest at different axial levels, Le Douarin and her associates were able to map the regions of the crest that provide colonists for the gut. Le Douarin’s work with chimeras suggested that both the vagal (somites 1–7) and the sacral (caudal to somite 28) crest colonise the ENS. The entire bowel receives émigrés from the vagal crest, but only the post-umbilical gut receives them from the sacral crest. Le Douarin’s conclusions were challenged soon after they were published (Allan and Newgreen, 1980). In fact, it was pointed out that her experiments, by their nature, lacked an important control. Since the migrating crest cells in interspecies chimeras are foreign to the species in which they migrate, it is conceivable that they are more or less invasive than the host’s own cells. Only the foreign cells can be recognised in the chimeric embryos; therefore, there is no way to compare the behaviour of the donor’s crest cells to that of the host. Quail cells thus might reach destinations in a chick embryo that they would never have reached if they had been left to migrate in the quail (or, if experiments are done in the reverse order, chick cells might reach ectopic destinations in a quail embryo). A comparison with the movements of transplanted crest cells in the same species is thus the control that is missing in studies of interspecies chimeras. The participation of the sacral crest in the colonisation, even of the post-umbilical gut, was the most controversial of Le Douarin’s conclusions. Other investigators, attempting to follow the progression of crest-derived cells within the avian bowel, recognised only a single wave of “neuroblasts” that moved inexorably in an oral to anal direction (Allan and Newgreen, 1980). This directionality is compatible with the descent that would be expected of progenitors from the vagal crest, but it is inconsistent with the ascent that was expected of progenitors derived from the sacral crest. Of course, the morphological identification of cells that can be recognised as “neuroblasts” implies that the cells that are recognised have already differentiated sufficiently so that they can be seen to be in a neuronal lineage. “Neuroblasts” are thus, by definition, not undifferentiated precursor cells. The crest-derived progenitors of cells that can be identified as neurons because they exhibit neural markers are mesenchymal cells that cannot be distinguished from other cells of the embryonic mesenchyme unless special tracers, or markers, are employed

DEVELOPMENT OF THE ENTERIC NERVOUS SYSTEM

473

(Rothman and Gershon, 1982). The progression of “neuroblasts”, therefore, does not reveal the migration of crest-derived precursors, but the temporal and spatial order in which neurons differentiate. Neuronal progenitors have been demonstrated to be present in the gut before these cells give rise to progeny that express neuronal markers (Rothman and Gershon, 1982). The neuronal progenitors were first detected indirectly by observing the development of neurons in vitro in explants of bowel removed from fetal mice before neural markers could be visualised. The fact that neurons appear in vitro is conclusive evidence that neuronal precursors were present in the cultured tissue at the time of its explanation, whether or not these precursors were then morphologically recognisable. The interval between the arrival of neuronal progenitors and the time when they terminally differentiate is a period when crest-derived precursors have the opportunity to be influenced by the enteric microenvironment. The enteric microenvironment has been shown to play a major role in the development of enteric neurons and glia (Coulter, Gershon and Rothman, 1988; Mackey, Payette and Gershon, 1988; Gershon, Chalazonitis and Rothman, 1993). The oral to anal progression of cells that can be recognised as “neuroblasts”, therefore, may reflect the maturation of the enteric microenvironment, rather than the descent of crest-derived precursors. Recent studies have re-investigated the migration to the gut of cells from the neural crest by using markers that enable the endogenous crest cells to be followed (Figure 11.1). These experiments supply the control that is lacking from studies of interspecies chimeras. The endogenous crest cells have been labelled, prior to their migration with a vital dye or

Figure 11.1 Regions of the neural crest that contribute émigrés to the ENS. The largest contingent of cells originate in the vagal crest and colonise all of the bowel below the oesophagus and proximal stomach. The sacral crest colonises the post-umbilical gut. Note that the post-umbilical bowel contains crest-derived cells of both vagal and sacral origin.

474

INNERVATION OF THE GASTROINTESTINAL TRACT

a replication-deficient retrovirus (Pomeranz, Rothman and Gershon, 1991; Serbedzija et al., 1991). The experiments that have traced the movements of endogenous crest cells have confirmed that the gut of both chicks and mice is colonised by both vagal and sacral crest-derived cells (Pomeranz, Rothman and Gershon, 1991; Serbedzija et al., 1991). Evidence suggests that the human bowel is also colonised by cells from the sacral as well as the vagal crest (Toyohara et al., 1985; Tam and Lister, 1986). The tracing of cells labelled in avian embryos with a replication-deficient retrovirus has caused the concept of the vagal origin of enteric neurons to be slightly modified. It appears that the entire vagal crest does not contribute to the formation of the ENS. Rather, the majority of enteric neuronal progenitors depart only from the small region of the vagal crest lying between somites 3 and 6 (Epstein et al., 1994). In the mouse, studies with crest-derived cells that have been labelled prior to their migration with a dye that intercalates into their plasma membrane have unexpectedly revealed that a small region of the most anterior portion of the truncal crest also colonises part of the gut. This region participates in the formation of the ENS of the rostral-most foregut (oesophagus and adjacent stomach) (Durbec et al., 1996a). The ability of the vagal and the sacral regions of the crest to provide colonists for the gut is dramatically illustrated by back-transplantation experiments. This type of study involves the removal of tissue from an older embryo and transplanting it into a younger host. The procedure tests the developmental potential retained by cells in the graft by permitting this potential to become manifest in the permissive environment of the younger embryonic host. When the quail or mouse bowel is back-transplanted into a chick host embryo soon after it has been colonised by émigrés from the neural crest, crest-derived cells leave the back-grafted gut and re-migrate in the younger host embryo (Rothman et al., 1990). Where the re-migrating enteric crest-derived cells go, depends on where in the host embryos the back-transplants are placed. Crest-derived cells from the donor gut will only reach the bowel of their host if the donor tissue replaces the host’s vagal or sacral crest (Rothman et al., 1993). If the gut is placed in the region of the trunk of the host embryo, the donor cells do not re-colonise the bowel, but instead reach targets that are appropriate for the truncal crest. Crest-derived cells evidently remember nothing of their previous journey to the bowel when they re-migrate following back-transplantation. They retain instead the developmental potential of their ancestors to follow a neural crest migration pathway when they encounter one. Transgenic mice have also been used to trace the movements of crest-derived cells colonising the gut. Vagal crest-derived precursors of enteric neurons can be visually identified in transgenic mice that express lacz under the control of the promoter for dopamine β-hydroxylase (DBH) (Kapur, Yost and Palmiter, 1992). Although there are few catecholaminergic neurons in the adult mouse gut, enteric neurons, as well as sympathetic neurons, permanently express the DBH-lacZ transgene. The colonisation of the bowel by cells labelled with the DBH-lacZ transgene has been examined in detail in both normal and mutant mice in which the terminal colon becomes aganglionic (Kapur, Yost and Palmiter, 1993; Coventry et al., 1994). These important studies have directly revealed the location in the fetal gut where crest-derived cells migrate and they provide an accurate means of timing of the arrival of vagal crest-derived cells in various levels of the fetal mouse gut. There is reason to believe, however, that the DBH-lacZ transgene is probably expressed

DEVELOPMENT OF THE ENTERIC NERVOUS SYSTEM

475

only in a subset of vagal crest-derived cells and not in those of sacral origin. A subset of enteric neurons develop from transiently catecholaminergic precursors (Baetge and Gershon, 1989; Baetge, Pintar and Gershon, 1990; Baetge, Schneider and Gershon, 1990; Blaugrund et al., 1996). These progenitors, which express both tyrosine hydroxylase (TH) and DBH, have been called TC cells (transiently catecholaminergic cells). The successors of TC cells are not catecholaminergic; nevertheless, DBH is not completely repressed and can be detected in the neurons that develop from TC cell progenitors, even though other elements of the catecholaminergic phenotype are inactivated in these neurons (Baetge, Pintar and Gershon, 1990). It is thus probable that the cells in the fetal gut that express the DBH-lacZ transgene are TC cells and/or neurons that arise from TC cells. TC cells, however, do not give rise to the entire set of enteric neurons, but only to a subset that is probably smaller than that which arises from progenitors that never exhibit catecholaminergic properties (Blaugrund et al., 1996). As a result, many, and possibly even a majority, of enteric neuronal precursors are probably not detected by the expression of the DBH-lacZ transgene. The net result of the history of tracing studies is that it is now possible to conclude that in both mammals and birds, the ENS arises from more than a single axial level of the neural crest (Figure 11.1). These levels include the vagal, which is the major source of enteric neuronal progenitors and colonises the entire bowel, the most rostral truncal segments, and the sacral. The rostral truncal crest colonises only the primordial oesophagus and adjacent stomach, while the sacral crest colonises only the post-umbilical bowel. The regions of the crest that provide the gut with neural and glial precursors, however, are limited. This limitation, when it was discovered, raised the possibility, now rejected, that the pre-migratory crest cells in the enteric levels of the crest might be endowed with specific “homing” information, which permitted them to find their correct destinations in the bowel.

CREST-DERIVED CELLS FOLLOW DEFINED PATHWAYS AND ARE PLURIPOTENT WHEN THEY COLONISE THE GUT Experimental evidence has not supported the idea that cells in the pre-migratory crest are determined to migrate to the bowel before they set out to do so. Most studies have instead suggested that the migratory routes to be taken by crest-derived cells are not preprogrammed in the cell’s repertoire and that in fact, the pre-migratory crest cells are pluripotent. The earliest such evidence was provided by studies that involved the surgical interchange of levels of the crest in avian embryos (Rothman et al., 1986; Fontaine-Perus, Chanconie and Le Douarin, 1988). When a region of the crest that normally colonises the gut is replaced with one that normally does not, the transplanted crest cells still migrate to the bowel, despite their heterotopic location; furthermore, the heterotopic crest-derived cells not only colonise the gut, but after they do so, they give rise to a normal-appearing ENS. By forming enteric neurons, therefore, the heterotopic crest cells differentiate in a target-specific manner, even though this pattern of phenotypic expression may be inappropriate for their level of origin in the neural crest. Similarly, when grafted to replace crest cells at other axial levels, vagal or sacral crest cells will migrate to locations other than the

476

INNERVATION OF THE GASTROINTESTINAL TRACT

gut, and once there, they will give rise to non-enteric neurons. By itself, this type of experiment demonstrates only that the population of crest cells (taken as a set) is pluripotent. The experiments do not, by themselves, reveal anything about the developmental potential of the individual crest cells within the population. The population of crest-derived precursors may appear to be pluripotent, because it contains the committed precursors of every conceivable derivative of the neural crest. A target organ might then select certain of these precursor cells to develop, while the others die. In this case, therefore, individual cells would not be pluripotent, although the population, as a whole, would seem to be so. Alternatively, individual cells may actually be pluripotent and depend on signals from the target organ to instruct them as to which developmental program to follow. In the first case, the role of the target organ would be one of selection, in the second case the role of the target organ would be one of instruction. To distinguish between these possibilities, it is necessary to identify the phenotypes expressed by cells differentiating in clones originating from single crest-derived progenitors. Clones derived from individual crest cells have now been investigated. Many different phenotypes have been found to be represented in the progeny within these clones developing, either in vitro (Sieber-Blum and Cohen, 1980; Baroffio, Dupin and Le Douarin, 1988; Duff et al., 1991; Sextier-Sainte-Claire Deville, Ziller and Le Douarin, 1992; Ito, Morita and Sieber-Blum, 1993), or in vivo (Bronner-Fraser and Fraser, 1988, 1989; Fraser and Bronner-Fraser, 1991). Since a single crest cell can thus give rise to a wide variety of successors, the progenitors from which the clones arise must be pluripotent. Crest-derived cells, moreover, remain multipotent, with respect to their ability to give rise to neurons and glia, even after they have completed their migration to the bowel. This developmental potential has been revealed by back transplantation. In the critical back-transplantation experiments, colonised segments of bowel were placed in neural crest migration pathways at various axial levels (Rothman et al., 1990). When grafted to a truncal level of the host embryo, donor crest-derived cells left the graft and migrated to the host’s sympathetic ganglia, adrenal gland and peripheral nerves. All of these destinations are appropriate targets for the migration of truncal crest cells. None of the back-transplanted gut-derived crest cells, however, reached the host’s gut; moreover, within the host’s organs, the donor cells did not give rise to enteric neurons or glia. Instead, the terminally differentiated phenotypes of cells migrating from the donor gut were appropriate for the target organs in which they were found; thus, the donor crest cells, despite having previously migrated to the bowel, now developed as catecholaminergic neurons in sympathetic ganglia, chromaffin cells in the adrenals, and Schwann cells in peripheral nerves (Rothman et al., 1990). These observations show that the population of crest derived cells that colonises the gut is pluripotent when it gets there. In vitro studies of clones developing from single crest-derived cells of enteric origin have demonstrated that a variety of different phenotypes, including some that are not present in the normal ENS, can be found among the progeny within the clone (SextierSainte-Claire Deville, Ziller and Le Douarin, 1994). Despite their multipotent nature, however, crest-derived cells in the gut do not appear to be quite as pluripotent as their predecessors in the pre-migratory crest (Rothman et al., 1990; Sextier-Sainte-Claire Deville, Ziller and Le Douarin, 1994; Lo and Anderson, 1995). Pre-migratory crest cells are able to give rise to melanocytes and ectomesenchyme, derivatives that are not found, either in

DEVELOPMENT OF THE ENTERIC NERVOUS SYSTEM

477

clones developing from single enteric crest-derived progenitors (Sextier-Sainte-Claire Deville, Ziller and Le Douarin, 1994; Lo and Anderson, 1995), or in the destinations reached in host embryos by cells leaving back-grafts of bowel (Rothman et al., 1993). Studies of the sources of precursors of the ENS have thus set the ground rules for interpreting the phenomena of enteric neuronal development. The gut is colonised by pluripotent progenitors that migrate from restricted regions of the neural crest to the bowel because defined migratory pathways connect these sites to the gut (Figure 11.1). Similar migratory routes extending away from other regions of the crest lead to different target organs. The largest cohort of crest-derived émigrés is conveyed to the bowel by the vagal pathway, which, in avian embryos, colonises the gut from the proventriculus to the cloaca. The equivalent region in mammals extends from the corpus of the stomach to the rectum. The sacral pathway conveys a smaller cohort of crest-derived émigrés only to the post-umbilical bowel, which thus contains crest-derived cells from two axial levels. Remaining to be determined are the differences, if any, between the two crest-derived cell populations in the post-umbilical gut, and whether or not the two populations interact. The truncal pathway to the gut is quite small, has thus far been demonstrated only in mammals, and populates only the presumptive oesophagus and the most rostral portion of the stomach. This population appears to be different from the vagal crest, making it likely that the oesophagus is not colonised by any vagal crest cells at all (Durbec et al., 1996a). Still be identified are the molecular cues that cause crest-derived cells to follow the migratory pathways. Possibilities include chemoattractant or repellent molecules. Molecules that attract or repel growing axons have been identified in the vertebrate CNS (Tessier-Lavigne, 1994). Netrins, for example, either attract or repel growing axons (Kennedy et al., 1994; Serafini et al., 1994; Tessier-Lavigne, 1994; Colamarino and TessierLavigne, 1995), while semaphorins act as repellents (Luo, Raible and Raper, 1993; Kolodkin, 1995; Messersmith et al., 1995). The ability of axons to extend along a correct trajectory is analogous to the ability of crest-derived cells to migrate along a proper pathway (Perris, Paulsson and Bronner-Fraser, 1989; Lallier et al., 1994). There are two netrins, 1 and 2 and both are expressed in the developing bowel (2 to a greater extent than 1) (Kennedy et al., 1994; personal observation). The role, if any, played by the enteric netrins in the colonisation of the gut by cells from the neural crest has not yet been ascertained; nevertheless, the occurrence of netrins in the fetal bowel is intriguing in this regard, and is another example of the striking similarity of the ENS to the CNS. The function of the enteric netrins may be only to guide axons, but enteric netrins or semaphorins may also guide crest-derived cells to their destinations. Either function is at the moment purely speculative.

MULTIPLE PROGENITOR LINEAGES PARTICIPATE IN THE FORMATION OF THE ENS The progress of development is associated with an increasing degree of restriction of the developmental potential of crest-derived progenitors. As a result, crest-derived cells become sorted into recognisable lineages (Le Douarin, 1986; Anderson, 1989; Le Douarin and Dupin, 1993). Although the crest-derived cells that colonise the bowel are pluripotent,

478

INNERVATION OF THE GASTROINTESTINAL TRACT

their development potential is not absolute and at least two enteric lineages with different potentialities have been identified (Blaugrund et al., 1996). The sorting of precursors into different lineages is significant, because lineage-related properties of neuronal precursors interact with the enteric microenvironment to determine the fate of the crest-derived cells in the gut. None of the enteric crest-derived progenitors can be thought of as undifferentiated. In order for progenitors to respond to environmental cues, such as growth factors or molecules of the extracellular matrix, the responding cells must first express receptors that these signals can stimulate. The repertoire of receptors expressed by precursors, therefore, is just as important in determining their fate as the signals they encounter. The particular set of receptors that are expressed by crest-derived cells, and thus the factors to which these cells respond, depend on the lineage of the precursor cells. Lineage thus establishes which developmental options are open to precursors, while the environmental ligands determine which of those that are open are actually followed. The development of the ENS thus has a structure that can be compared to a Wagnerian opera. Lineages of crestderived precursors provide themes, the enteric microenvironment provides the counterpoint and the receptors and their ligands provide the repeating leit-motifs. The TC (transiently catecholaminergic) cells, which arise during the development of the ENS, provided the earliest indication that enteric neuronal precursors are sorted into more than a single lineage (Cochard, Goldstein and Black, 1978; Teitelman, Joh and Reis, 1978). These cells, which occur both in the fetal gut (Cochard, Goldstein and Black, 1978; Teitelman, Joh and Reis, 1978), and all along the pathway followed by vagal crest-derived cells migrating to the gut (Baetge and Gershon, 1989; Baetge, Schneider and Gershon, 1990), were discussed above in explaining the operation of the DBH-lacZ transgene (Kapur, Yost and Palmiter, 1992). TC cells express all of the catecholaminergic properties that are found in mature sympathetic neurons. For example, TC cells contain TH and DBH; they specifically take up NA, and they store a catecholamine (Jonakait et al., 1979; Gershon et al., 1984; Jonakait, Rosenthal and Morrell, 1989). Sympathetic neurons, however, like all neurons, are post-mitotic cells, while TC cells proliferate (Rothman et al., 1980; Teitelman et al., 1981; Baetge and Gershon, 1989; Baetge, Pintar and Gershon, 1990). By definition, therefore, because they are not post-mitotic, TC cells cannot be neurons; nevertheless, TC cells express neural markers. These markers include the cell type-specific intermediate filament proteins, peripherin and the neurofilament triplet (Baetge and Gershon, 1989; Baetge, Pintar and Gershon, 1990). TC cells, furthermore, give rise to neurons when they are explanted and grown in vitro (Baetge, Pintar and Gershon, 1990; Baetge, Schneider and Gershon, 1990). Since DBH continues to be expressed by the successors of TC cells, long after the transcription of TH has ceased, it is possible to identify the enteric neurons that develop from TC cell ancestors (Baetge, Pintar and Gershon, 1990). All enteric serotonergic neurons and at least some neurons that contain substance P or neuropeptide Y co-express DBH in the adult rat gut. The observation, therefore that DBH is found only in some, but not all mature enteric neurons, suggests that the transient expression of a catecholaminergic phenotype defines one lineage of enteric progenitor, and that the lack of catecholaminergic expression defines another. This suggestion has now been confirmed. The catecholaminergic phenotype is not the only characteristic that TC cells share with sympathoadrenal progenitor cells. Each of these progenitors also expresses the same cell

DEVELOPMENT OF THE ENTERIC NERVOUS SYSTEM

479

surface differentiation antigens, which change at the same developmental ages. Both TC cells and sympathoadrenal precursors express plasmalemmal proteins, recognised by a series of monoclonal antibodies, that are called the “SA” antigens (Anderson et al., 1991; Carnahan, Anderson and Patterson, 1991). The SA antigens appear transiently and are replaced by another transiently expressed surface antigen, “B2”, which is also recognised by monoclonal antibodies. The common expression of these cell surface antigens by both enteric and sympathoadrenal precursors caused Carnahan, Anderson and Patterson to suggest that enteric neurons are derived from the sympathoadrenal lineage (Carnahan, Anderson and Patterson, 1991). Because there are multiple lineages of crest-derived precursor participating in the formation of the ENS, this suggestion turned out to be only partially correct (see below); however, it is likely that the TC cells, if not all enteric neurons, are derived from a common sympathoadrenal-enteric progenitor (Blaugrund et al., 1996). The probable existence of a common sympathoadrenal-enteric precursor lineage in the developing avian ENS, where TC cells do not arise, has been revealed by studies of clones of enteric crest-derived progenitors (Sextier-Sainte-Claire Deville, Ziller and Le Douarin, 1994). These clones, differentiating in vitro, contain catecholaminergic cells, even though such cells do not arise in the avian gut in situ. Two types of experiment have been used to test the hypothesis that the ENS arises from a common sympathoadrenal-enteric progenitor (Blaugrund et al., 1996). In one set of experiments, all cells in the dissociated fetal rat gut that express the cell surface antigens common to sympathoadrenal and enteric neuronal precursors were destroyed in vitro by complement-mediated lysis. The B2 monoclonal antibodies were employed for this purpose. This treatment lysed all of the cells that expressed markers common to sympathetic and enteric neurons. The total destruction of such cells would be expected to completely prevent the in vitro development of neurons, if a common sympathoadrenal-enteric precursor exists and is the only source of enteric neurons. Complement-mediated lysis of all cells that expressed the B2 antigen did not, however, prevent enteric neurons from developing in vitro. Neurons continued to arise in the treated cultures, although their numbers were reduced. The types of neurons arising in the treated cultures, moreover, were revealed. Complement-mediated lysis eliminated all neurons that express TH, DBH, or B2, suggesting that it was effective in ablating the TC cell lineage. No serotonergic neurons developed in the treated cultures, a finding that is consistent with the idea that these cells are entirely TC-derived. In contrast, complement-mediated lysis did not prevent the development of neurons that contain calcitonin gene related peptide (CGRP). The effects of complement-mediated lysis thus are remarkably specific. The procedure prevents the development of some enteric neurons, those that are linked to TC cells, but it does not interfere with the development of others. These observations suggest that more than one precursor lineage must contribute to the development of the ENS; moreover, one of these lineages may be that which is postulated to be common to the sympathoadrenal system. The second set of experiments that demonstrated that multiple precursor lineages participate in the development of the ENS involved transgenic mice with a targeted deletion of mash-1, the mammalian analog of achaete-scute of Drosophila (Johnson et al., 1992). Mash-1 encodes a transcription factor and is normally expressed in fetal mice, both by cells in developing sympathetic and enteric ganglia (Lo et al., 1991; Guillemot and Joyner,

480

INNERVATION OF THE GASTROINTESTINAL TRACT

1993). When mash-1 is knocked out, the sympathetic nervous system fails to develop. Sympathetic neurons, therefore, are mash-1 dependent. The mash-1 dependence of cells in the sympathoadrenal lineage provides a test of the hypothesis that there is a common sympatho-adrenal-enteric progenitor. Unless the development of enteric neurons diverges from the sympathoadrenal system before neurons become mash-1 dependent, the knockout of mash-1 would be expected to eliminate enteric neurons, as well as sympathetic neurons, which arise from the putative common sympathoadrenal-enteric progenitor. The possibility that the enteric and sympathoadrenal systems diverge prior to becoming mash-1 dependent seems unlikely in view of the fact that mash-1 is expressed in the developing ENS. The ENS is not entirely absent in the mash-1 knockout mice (Guillemot et al., 1993). Instead, only the oesophagus is aganglionic. Neurons develop in the rest of the bowel, but they appear two days later than they normally do. The delay in the appearance of neurons could, in theory, simply be a reflection of a non-specific retardation of the development of all neurons; alternatively, it is also possible that a mash-1 dependent set of early-developing neurons is selectively eliminated in the knockout animals. An earlier observation lent credence to this latter possibility. Enteric neurons of different phenotypes are born in a sequential order (Pham, Gershon and Rothman, 1991). Among the earliest born are serotonergic neurons. Some serotonergic neurons arise from precursors that become postmitotic as early as embryonic day E8.5, an age that precedes the colonisation of the bowel by even the most pioneering of crest-derived cells. Other neurons, such as those that contain CGRP, are born much later than serotonergic neurons and continue to be born post-natally. The births of CGRP neurons commence at E16, about two days after the last serotonergic neuron has been born. These observations support the idea that lineages of enteric neurons might be distinguishable on the basis of birthdates. If so, then 5-HT and CGRP might be markers for early- and late born lineages respectively. If the mash-1 knockout selectively ablates an early-developing set of enteric neurons, therefore, then one might expect that the ENS of the knockout animals would lack serotonergic neurons, but contain those that express CGRP. The previously noted evidence that serotonergic neurons are TC cell-derived, and the common expression of many properties by TC cells and sympatho-enteric progenitors, suggested that TC cells might be mash-1 dependent (Blaugrund et al., 1996). This hypothesis was tested and confirmed. Antibodies to the Mash-1 protein were used to demonstrate the coincident expression of Mash-1 and TH immunoreactivities in normal mice. TC cells, therefore, express mash-1; moreover, no TC cells can be found in the bowel of mash-1 knockout mice. TC cells thus both express mash-1 and, like the precursors of sympathetic neurons, are mash-1 dependent. As predicted, there are also no serotonergic neurons in the late-developing ENS of mash-1 knockout mice (Figure 11.2). In contrast, the ENS of Mash-1 knockout mice does contain CGRP neurons. Two lines of data, one derived from complement-mediated lysis of enteric cells that express antigens in common with sympathoadrenal precursors, the other derived from an analysis of the defective ENS that arises in mash-1 knockout mice, converge to demonstrate that at least two precursor lineages participate in the formation of the ENS. Both lines of evidence imply that TC cells are the enteric representation of a mash-1-dependent sympathoadrenal-enteric progenitor. They also suggest that 5-HT is a marker for a neuron

DEVELOPMENT OF THE ENTERIC NERVOUS SYSTEM

481

Figure 11.2 The development of enteric serotonergic neurons is prevented in the mash-1 knockout mouse. The fetal mouse gut has been incubated with 3H-5-HT and the sites where the 5-HT transporter is active have been localised by radioautography. (A) In the normal bowel, the 5-HT transporter is active in the mucosa and in myenteric ganglia. (B) In the mash-1 knockout mouse, the transporter is active in the mucosa, but not in the myenteric ganglia. The normal transport of 3H-5-HT in the mucosa of the mash-1 knockout mouse indicates that the transporter molecule is normal in the mutant animals. The selective loss of the transporter from the ganglia of the mash-1 knockout mouse thus is due to the absence of enteric serotonergic neurons in these animals.

that is derived from this lineage and that CGRP is a marker for a neuron that is not. It can thus be concluded that the ENS is derived, in part, from a lineage that is common to sympathoadrenal precursors. The expression of mash-1 by cells in this lineage is obligatory. These cells also are born early in ontogeny, are transiently catecholaminergic in the gut, and give rise to a subset of enteric neurons. This subset includes all of the neurons of the oesophagus and all enteric serotonergic neurons. Cells that are not in this lineage have little in common with sympathoadrenal cells. They are never catecholaminergic, arise late in ontogeny, and neither express nor depend on mash-1. These cells, which could possibly be members of additional lineages that have yet to be defined, give rise to neurons in all regions of the bowel except the oesophagus. CGRP-containing neurons are representatives of this non-sympathoadrenal-related lineage.

SUBLINEAGES OF ENTERIC NEURONAL PRECURSORS ARE DEFINED BY RECEPTORS FOR DIFFERENT GROWTH/DIFFERENTIATION FACTORS The complexity of the ENS is so great, that it is simplistic to assume that enteric neuronal progenitors would be sorted only into two lineages. A great deal of evidence suggests that this is not the case. As development proceeds, the precursors of enteric neurons acquire receptors for different growth factors and become dependent on them. The expression of these growth factor receptors, and the consequent dependence of the neurons that express them, on the corresponding ligands, thus define sublineages of progenitor cells. The size

482

INNERVATION OF THE GASTROINTESTINAL TRACT

Figure 11.3 Multiple lineages of crest-derived precursors participate in the formation of the ENS. The fetal gut is colonised by a pluripotent precursor that depends on GDNF/Ret. Consequently, the knockout of GDNF or Ret (dashed line) will eliminate all enteric neurons below the oesophagus and proximal stomach. The pluripotent precursor gives rise to two subsequent lineages. One is mash-1-dependent, while the other is mash-1-independent. The knockout of mash-1 (dashed line) thus eliminates only those neurons developing in the mash-1-dependent lineage. This lineage includes serotonergic and motor neurons, not neurons that express CGRP. The knockout of CNTFRα appears to result in the loss of motor neurons to smooth muscle. This growth factor evidently operates at a later time in development and thus its loss is associated with a more restricted ENS defect.

of the lineage defined by a given receptor and ligand-dependence is determined by the degree to which the developmental potential of the precursor cells has been reduced at the time the receptor is expressed. For example, the dependence on a given growth factor of the early pluripotent progenitor that initially colonises the bowel would define the lineage of virtually the entire set of enteric neurons (Figure 11.3). Ablation of that factor, or its receptor, would prevent the development of any neurons in the gut, because all arise from the pluripotent precursor. As the successors of the pluripotent precursor become sorted into sublineages of more restricted developmental potential, however, the acquisition of growth factor dependence will come to define progressively smaller sets of enteric neurons. Ablation of the later-acquired factor, or its receptor, in contrast to that expressed by the pluripotent precursor, would not prevent the development of all enteric neurons, but only of those which arise from the subset of precursors that depend on this factor. Knockout mice provide insight into the progressive restriction of the developmental potential of enteric neuronal progenitors and permit sublineages to be identified.

DEVELOPMENT OF THE ENTERIC NERVOUS SYSTEM

483

THE PLURIPOTENT CREST-DERIVED PRECURSOR THAT COLONISES THE GUT EXPRESSES RET AND IS GLIAL CELL LINE-DERIVED NEUROTROPHIC FACTOR-DEPENDENT The pluripotent crest-derived émigrés that colonise most, although not all, of the bowel express the c-ret proto-oncogene and depend for their survival, on stimulation of the receptor, Ret, that this gene encodes (Pachnis, Mankoo and Costantini, 1993; Schuchardt et al, 1994; Durbec et al., 1996a). Ret is a receptor tyrosine kinase, the ligand for which has only recently been discovered. The functional ligand for Ret is glial cell line-derived neurotrophic factor (GDNF) (Durbec et al., 1996b; Jing et al., 1996; Trupp et al., 1996). GDNF was first found to be produced by a glial cell line (B49) and to promote the survival of midbrain dopaminergic neurons (Lin et al., 1993). Later, GDNF was observed to exert additional effects, including the enhancement of the survival of spinal motor neurons (Trupp et al., 1995). GDNF, which is distantly related to transforming growth factor-β (TGF-β), is a homodimer of two disulfide-linked peptides, each containing 134 amino acids. Mature GDNF is derived by the proteolytic cleavage of a larger 211 amino acid precursor, which occurs intracellularly prior to secretion. GDNF is very highly expressed in the developing gut and other peripheral organs (Choi-Lundberg and Bohn, 1995; Trupp et al., 1995). GDNF promotes not only the survival of CNS neurons and their extension of neurites (Trupp et al., 1996), but also the survival of sensory and sympathetic neurons as well (Trupp et al., 1995). Since GDNF affects sympathetic neurons, it would be expected also to affect at least those neurons of the ENS that arise from the sympathoadrenal-enteric lineage. Enteric and sympathetic neurons both, at least transiently, express c-ret (Pachnis, Mankoo and Costantini, 1993; Tsuzuki et al., 1995). The effects of GDNF and Ret in the gut are not limited to this one lineage, however (Figure 11.3). The ENS completely fails to develop in almost the entire bowel of c-ret knockout mice; neurons are found only in the rostral foregut (Schuchardt et al., 1994; Durbec et al., 1996a). The neurons that survive the knockout of Ret, therefore, are only those that migrate to the bowel from the rostral truncal crest. The entire set of vagal and sacral émigrés is lost. A similar lesion has recently been found to occur in the bowel of GDNF knockout mice (Moore et al., 1996; Pichel et al., 1996; Sanchez et al., 1996). GDNF, moreover, does not exert trophic effects on crest-derived cells obtained from c-ret knockout animals (Durbec et al., 1996b). The loss of neurons in the c-ret and the GDNF knockout mice can thus be thought of as different manifestations of the same phenomenon. The ablation of the receptor, Ret, has the same consequences in the developing ENS as the loss of the ligand, GDNF. The pluripotent crest-derived precursors of vagal and sacral origin must, therefore, be GDNF/Ret-dependent. GDNF does not itself appear to bind to the Ret receptor. Rather, GDNF binds to a protein called GDNFR-α, which is glycosylphosphatidylinositol-linked to cell surfaces and forms a complex with Ret to trigger autophosphorylation and other Ret-mediated effects (Jing et al., 1996; Treanor et al., 1996). The regions of the gut that develop from the most rostral foregut, in contrast to the majority of the bowel, do contain neurons, even in c-ret knockout mice (Schuchardt et al., 1994; Durbec et al., 1996a). With the exception of only the superior cervical ganglion, moreover, sympathetic ganglia develop in the c-ret knockout animals. These observations

484

INNERVATION OF THE GASTROINTESTINAL TRACT

suggest that the crest-derived cells that colonise the rostral foregut might have more in common with the precursors of the caudal sympathetic ganglia, than with those which give rise to the superior cervical ganglion. In contrast, the vagal and sacral crest-derived émigrés that colonise the bulk of the bowel, may have more in common with the precursors of the superior cervical than with those of other sympathetic ganglia. An intercalating fluorescent marker has been used to trace the migration of cells from the truncal crest in cultures from fetal mice. The labelled crest-derived cells that colonise the rostral foregut originate from the truncal crest cells as do the caudal sympathetic ganglia. In contrast, the post-otic population of vagal crest cells that colonises the remainder of the bowel also participates in forming the superior cervical ganglion. The common sympathoadrenalenteric lineage thus appears not to be a single entity. Those sympathoadrenal-enteric cells that participate in forming the ENS distal to the rostral foregut must be c-ret- and GDNFdependent, while those of the rostral foregut (presumptive oesophagus and adjacent stomach) are c-ret- and GDNF-independent. The vagal or truncal origin of the precursors thus seems to be more a determinant of GDNF/Ret-dependence than is the relationship of precursors to the sympathoadrenal system. Vagal precursors are all GDNF/Ret-dependent. Truncal precursors are all GDNF/Ret-independent. The mash-1-dependent and c-ret-dependent lineages have been proposed to be reciprocals of one another (Durbec et al., 1996a). The ENS of the oesophagus is mash-1-dependent, but c-ret-independent. The ENS of the rest of the bowel is c-ret-dependent, yet it contains neurons (some) that are mash-1-independent. This proposal, however, cannot be totally correct. As noted previously, the ENS below the oesophagus can hardly be considered to be mash-1 independent, even if some of the cells in this region are so. TC cells and all of the neurons that arise from them are lacking. It is not clear why the c-ret-independent crest-derived cells of the rostral foregut do not migrate distally in the bowel of c-ret or GDNF knockout mice. The evident failure of the c-ret-independent cells of the rostral foregut to expand their territory in c-ret knockout mice might be the result of an impairment of their ability to migrate, or possibly to proliferate. Alternatively, a compensatory factor (currently unknown) may be present only in the rostral foregut that allows neural precursors that would otherwise be GDNF/Ret-dependent to survive in this region, despite the absence of GDNF or Ret.

THE NEUROTROPHIN NT-3 PROBABLY AFFECTS THE DEVELOPMENT OF THE ENS Neurotrophins were not originally thought to play a significant role in enteric neuronal development. In contrast to developing sensory and sympathetic ganglia, which are nerve growth factor (NGF)-dependent, explanted enteric neurons thrive when cultured in the absence of NGF, or even in the presence of antibodies that neutralise NGF (Dreyfus and Bornstein, 1977; Dreyfus, Sherman and Gershon, 1977). Addition of NGF to the medium, moreover, does not stimulate the outgrowth of neurites from organotypic cultures of gut. The progeny of animals that produce auto antibodies to NGF exhibit severe sensory and sympathetic defects (Johnson et al., 1983; Pearson, Johnson and Brandeis, 1983); nevertheless, the same animals do not display ENS lesions (Gershon et al., 1983). The

DEVELOPMENT OF THE ENTERIC NERVOUS SYSTEM

485

development of the ENS, therefore, is independent of NGF. NGF, however, was only the first neurotrophin to be appreciated; it is not the only neurotrophin. The studies that indicate that enteric neuronal development does not require NGF were carried out before the other neurotrophins had been discovered. The neurotrophins are now known to constitute a family of small, very basic proteins, that in addition to NGF includes brain-derived neurotrophic factor, (BDNF), neurotrophin-3 (NT-3), NT-4/5 (Barbacid, 1993; Lindsay et al., 1994; Chao and Hempstead, 1995), and NT-6 (Gotz et al. 1994). There is a common neurotrophin receptor, p75NTR, to which all neurotrophins can bind, and three more specific receptor tyrosine kinases, or Trk molecules, which bind individual neurotrophins with higher affinity. TrkA selectively binds NGF, TrkB selectively binds BDNF and NT-4/5, and TrkC selectively binds NT-3. The neurotrophins can become promiscuous at high concentration and bind non-specifically to a Trk other than their primary receptor. Binding of a neurotrophin to a Trk appears to be critical for their specific effects, but the role of the common neurotrophin receptor, p75NTR, has been unclear. Many functions have putatively been assigned to p75NTR, including enhancing the affinity of a Trk for its neurotrophin, increasing the rate of binding of NGF to TrkA, and making Trk receptors more specific by decreasing their susceptibility to stimulation by an inappropriate neurotrophin. Other functions putatively assigned to p75NTR include effects on apoptosis, retrograde transport of neurotrophins, sphingomyelin hydrolysis, and cell migration (Chao and Hempstead, 1995; Bothwell, 1996). It has, however, only recently become clear that p75NTR really does play a role in development that is physiologically significant (Bothwell, 1996). The role played by p75NTR has been difficult to separate from that of a Trk receptor, because Trks can be functional, even in the absence of p75NTR. Schwann cells, however, do not express a Trk receptor, but they do express p75NTR. Despite the absence of a Trk, Schwann cells respond to neurotrophin stimulation. p75NTR is thus a neurotrophin receptor that can operate independently of a Trk. The transduction pathway activated by p75NTR involves the stimulation of the transcription factor, nuclear factor kappa-B (NF-κB) (Carter et al., 1996). Neurotrophins do not activate NF-κB in the Schwann cells of p75NTR knockout mice. Each of the trk genes may give rise to multiple transcripts (Barbacid, 1993; Tsoulfas et al., 1993; Lindsay et al., 1994; Chao and Hempstead, 1995), which in turn give rise to proteins that vary in their kinase domains. Variants of TrkB and TrkC also occur, which may be truncated so as to lack intracellular kinase domains. Other variants may contain inserts that inactivate their kinase activity. The discovery of multiple neurotrophins and their receptors led to the reinvestigation of the role(s) played by neurotrophins in enteric neuronal development. Since the common neurotrophin receptor, p75NTR is expressed by the crest-derived cells that colonise the fetal mouse and rat gut, it seemed likely that neurotrophins would be important (Baetge and Gershon, 1989; Baetge, Pintar and Gershon, 1990). Neurons and glia develop in vitro from these p75NTR-immunoreactive cells (Baetge, Schneider and Gershon, 1990). In fact, when antibodies to p75NTR are used to immunoselect crest-derived cells from the fetal bowel for growth in vitro, neurons and glia arise almost exclusively in cultures of the immunoselected population and very few develop in cultures of the p75NTR-depleted residual set (Pomeranz et al., 1993; Chalazonitis et al., 1994). It is conceivable that all of the crest-derived cells that colonise the gut express p75NTR. Thus far, no marker has revealed enteric crest-derived neural precursors that do not express p75NTR. Although

486

INNERVATION OF THE GASTROINTESTINAL TRACT

neurotrophins each activate a relatively specific Trk and may not require p75NTR, the expression of p75NTR by the crest-derived cells that colonise the gut certainly suggests that these cells should be neurotrophin-responsive. Another reason for suspecting that a neurotrophin plays a role in enteric neuronal development is that at least one lineage of enteric neurons is derived from a common sympathoadrenal-enteric progenitor. Sympathetic neurons are NGF-dependent, which enteric neurons are not: however, before this dependence develops, sympathetic neural precursors are supported by NT-3 (Black, 1978; Birren, Lo and Anderson, 1993; DiCicco-Bloom, Friedman and Black, 1993; Verdi and Anderson, 1994). When the NT-3 to NGF change in neurotrophin dependence occurs, sympathetic neuronal progenitors switch from expression of TrkC to TrkA (Birren, Lo and Anderson, 1993; DiCicco-Bloom, Friedman and Black, 1993; Wyatt and Davies, 1995). The TrkC to TrkA change may occur spontaneously (Wyatt and Davies, 1995), or it may have to be provoked by NT-3 (Verdi and Anderson, 1994). In any case, NT-3 promotes sympathoadrenal development (DiCicco-Bloom, Friedman and Black, 1993; Zhang, Winslow and Sieber-Blum, 1993; Verdi and Anderson, 1994); furthermore, the targeted knockout of NT-3 (Ernfors et al., 1994; Fariñas et al., 1994) and antibodies that neutralise NT-3 each interfere with the development of sympathetic neurons (Zhou and Rush, 1995). When NT-3 is absent during ontogeny, excessive apoptosis occurs in sympathetic neuroblasts (El Shamy et al., 1996). The early independence of sympathoadrenal progenitors on NT-3 suggests that this neurotrophin is likely to affect the cells in the enteric lineage that arise from the common sympathoadrenal-enteric progenitor in the gut. Divergence from the common pathway along specifically enteric lines, before the acquisition of TrkA and NGF-dependence, would explain the evident NGF-independence of virtually all enteric neurons. The enteric microenvironment might prevent the acquisition of NGF-dependence. This idea is reasonable, because the successors of TC cells lose their resemblance to sympathetic neurons and become non-catecholaminergic as they are induced to acquire other, gut-appropriate, phenotypes. The pre-divergent phase, however, when TC cells are still catecholaminergic, would be expected to be influenced by the early-acting neurotrophin, NT-3. If so, then NT-3 should support the development of the mash-1-dependent lineage of enteric neurons. Such an effect would be quite limited and thus vastly different from that of GDNF (Figure 11.3). Activation of Ret by GDNF is evidently required for the survival and/or development of an earlier precursor than the mash-1 dependent progenitor. Ablation of GDNF or Ret causes the total loss of neurons and glia, while the ablation of mash-1 leads only the loss of a limited set of neurons. The idea that NT-3 plays a role in enteric neuronal development is supported by the observation that TrkC, the neurotrophin receptor most selective for NT-3, is expressed by enteric neurons (Chalazonitis et al., 1994; Lamballe, Smeyne and Barbacid, 1994). Both the full-length (containing a kinase domain) and truncated forms of TrkC can be detected in newborn mice and fetal rats (Tsoulfas et al., 1993). Messenger RNA encoding TrkC has been found by in situ hybridisation in the developing and mature ENS (Tessarollo et al., 1993; Lamballe, Smeyne and Barbacid, 1994). The full-length TrkC is enriched in immunoselected crest-derived cells obtained from the fetal rat bowel (Chalazonitis et al., 1994). The binding of NT-3 to both full-length and truncated forms of TrkC has been observed in the E13.5 chick gut (Escandon et al., 1994), although such

DEVELOPMENT OF THE ENTERIC NERVOUS SYSTEM

487

binding was not detected by similar methods applied to the bowel of newborn mice (Escandon et al., 1994). NT-3 is expressed in the developing gut (Tojo et al., 1995). In transgenic mice, lacZ driven by the NT-3 promoter has been found to be expressed by cells in the outer mesenchyme of the fetal bowel. This is the layer within which crest-derived cells migrate and within which myenteric ganglia arise. The location of the lacZ-expressing cells thus suggests that NT-3 is probably produced in situ and secreted where it can readily affect the developing enteric neuronal precursors and/or neurons that express TrkC. NT-3 expression has not been detected in the submucosa, which is of interest because the submucosal plexus probably contains no neurons that arise from the mash-1-dependent precursors that are most likely to be the cells that respond to NT-3. Submucosal ganglia do not develop until long after myenteric ganglia have assembled (Payette, Bennett and Gershon, 1984; Payette et al., 1987) and no submucosal neurons arise from the early-born set of enteric neuronal precursors (Pham, Gershon and Rothman, 1991). Submucosal neurons, therefore, probably are derived from the mash-1-independent lineage (Figure 11.3), which gives rise to, among other cells, CGRP-containing neurons (Blaugrund et al., 1996). CGRP neurons are abundant in the submucosal plexus and much less common in the myenteric (Bornstein and Furness, 1988; Costa et al., 1996). These considerations are consistent with the idea that the development of the mash-1-dependent TC cell lineage, is influenced by NT-3.

NT-3 ENHANCES THE DEVELOPMENT OF ENTERIC NEURONS AND GLIA To study the effects of growth factors on the development of enteric neurons or glia in vitro, it was first necessary to develop a method to isolate crest-derived cells from the fetal bowel. Unless crest-derived cells are isolated, it is not possible to distinguish the effects of growth factors on neuronal and/or glial precursors from those mediated indirectly by other cells of the enteric mesenchyme. Enteric crest-derived cells can be isolated because, as noted above, they express cell-surface differentiation antigens that non-neuronal cells of the gut wall do not express. Antibodies decorate these plasmalemmal antigens in living cells and thus enable the crest-derived cells to be immunoselected. Immunoselection is analogous to the use, described earlier, of monoclonal antibodies to the B2 antigen for the complement-mediated lysis of TC cells (Blaugrund et al., 1996). In both cases, primary antibodies coat the outsides of cells that express the corresponding antigen. The addition of complement causes the antibody-coated cells to lyse. The addition of labelled secondary antibodies permits the cells to be selected. HNK-1 monoclonal antibodies were the first to be used for the immunoselection of crest-derived cells from the fetal gut of chicks and rats (Pomeranz et al., 1993; Chalazonitis et al., 1994). More recently, antibodies to p75NTR (Chalazonitis, Rothman and Gershon, 1995) and Ret (Lo and Anderson, 1995) have been employed. For immunoselection, the fetal (or, in avians, the embryonic) gut is dissociated. The suspended cells are first exposed to primary antibodies to label the cells that are crest-derived. These cells are then immunoselected, either magnetically, with secondary antibodies coupled to magnetic beads (Pomeranz et al., 1993; Chalazonitis et al.,

488

INNERVATION OF THE GASTROINTESTINAL TRACT

Figure 11.4 Crest-derived cells can be isolated from the fetal gut by immunoselection with antibodies to p75NTR, the common neurotrophin receptor. (A, B) Immunoselected (A) and residual (B) cells (those not selected by the antibodies have been immunostained with antibodies to PGP9.5, a neuronal marker. Neurons have developed in the culture of immunoselected cells (A) but not in the culture of residual cells (B). (C, D) Cultures of residual cells have been immunostained with antibodies to Kit (C) to demonstrate ICCs or desmin (D) to demonstrate smooth muscle. ICCs and smooth muscle develop specifically in cultures of residual cells.

1994) or by manual (Sextier-Sainte-Claire Deville, Ziller and Le Douarin, 1994) or automatic cell sorting (Lo and Anderson, 1995), utilising fluorescent secondary antibodies. Immunoselection thus provides two populations of cells, a purified crest-derived set, and a control set which has been depleted of cells that express the antigen used for the selection process. The two populations can be cultured separately and the behaviours of the cells compared. Alternatively, the populations can be compared chemically to determine whether factors or receptors of interest are selectively expressed by crest or non-crest derivatives. Neurons and glia develop selectively in the cultures that contain immunoselected cells, while terminally differentiated non-neuronal cells, such as the Kit-expressing interstitial cells of Cajal (ICCs) (Ward et al., 1994; Huizinga et al., 1995; Torihashi et al., 1995; Ward et al., 1995) and desmin-containing smooth muscle cells appear in the crestdepleted cultures of residual cells (Figure 11.4). The crest-derived population is still a proliferating one when it colonises the gut (Teitelman et al., 1981; Baetge and Gershon, 1989; Pham, Gershon, and Rothman, 1991); nevertheless, the residual cells continue to proliferate in vitro to a far greater extent than the crest-derived cells, which is understandable because the crest-derived cells withdraw from the cell cycle as they differentiate into neurons. NT-3 promotes the appearance of neurons and glia in cultures of isolated populations of crest-derived cells immunoselected from the fetal rat gut (Chalazonitis et al., 1994; Chalazonitis, 1996). NT-3, in contrast, does not affect the crest-depleted populations of residual cells that remain after immunoselection. To identify neurons in vitro, it is necessary that chemical markers be used, because the morphological appearance of cells in culture can be misleading. Effective neuronal markers include neurofilament proteins, peripherin, neuron specific enolase, and PGP9.5 (a neuronal form of ubiquitin hydrolase). These

DEVELOPMENT OF THE ENTERIC NERVOUS SYSTEM

489

markers can be visualised by immunocytochemistry. Glial markers include S100 and glial fibrillary acidic protein (GFAP). The promotion of neuronal and glial development by NT-3 is a concentration-dependent effect that peaks at 40 ng/ml. NT-3 also enhances neurite outgrowth. Interestingly, in contrast to pre-migratory crest cells, which are stimulated to proliferate by NT-3 (Kalcheim, Carmeli and Rosenthal, 1992; Pinco et al., 1993), NT-3 does not exert mitogenic effects on crest-derived cells immunoselected from the fetal gut. Its effect on these cells is to enhance their differentiation and thus withdrawal from the cell cycle. This action is similar to that of NT-3 on dorsal root ganglion cell precursors, which can even be induced by NT-3 to differentiate prematurely, thereby reducing their ultimate numbers (Ockel, Lewin and Barde, 1996). The action of NT-3 on post-migratory crest-derived cells, such as those that colonise the bowel and those that assemble in dorsal root ganglia, is thus different from its action on pre-migratory crest cells. In common with the effects of most growth factors, the action of NT-3 on cells immunoselected from the fetal gut, is associated with the transient induction of the c-fos protooncogene (Chalazonitis et al., 1994). Neurotrophins, other than NT-3, including NGF, BDNF, and NT4/5, do not affect either the in vitro development of enteric neurons and glia, or the in vitro proliferation or differentiation of the immunoselected enteric crest-derived cells. The in vitro data thus show that NT-3 can affect the development of enteric neurons and glia. Whether it actually does so physiologically has not yet been definitively established. Recent studies with transgenic mice have shown that NT-3 can exert an effect on the developing ENS in vivo (Pham et al., 1996). These studies have taken advantage of the ability of the DBH promoter to direct the expression of transgenes to enteric neurons (Kapur, Yost and Palmiter, 1992). When coupled to NT-3, the DBH promoter caused NT-3 to be overexpressed in the developing ENS. The DBH/NT-3 transgenic mice do not appear to be grossly abnormal. They survive, grow well, and mate; however, hyperganglionosis occurs in the myenteric plexus of the small and large intestines of the DBH/NT-3 transgenic animals. The number of neurons/ganglion, the number of neurons/unit length of gut, the packing density of neurons within ganglia, the proportional area of ganglia, and the mean size (maximal diameter and volume) of individual neurons are all significantly increased. In contrast, all of these parameters are unchanged in the submucosal plexus and the numbers of CGRP-containing neurons are unaltered. As noted above, CGRP-containing neurons, which are derived from cells in the mash-1-independent lineage, are a marker for that lineage, and are the latest-born of enteric neurons (Pham, Gershon and Rothman, 1991; Blaugrund et al., 1996). These data suggest that the earlydeveloping mash-1-dependent lineage of enteric neurons is probably selectively affected by the DBH/NT-3 transgene. That could be due to the fact that NT-3 overexpression is directed by the DBH promoter to the mash-1-dependent TC cells, which are the cells that normally express DBH. Alternatively, these cells may be the only cells in the bowel that can respond to NT-3. Knockout mice that lack genes encoding neurotrophins or neurotrophin receptors have now provided evidence that neurotrophins are essential for normal enteric neuronal development (Chalazonitis et al., 2001). Enteric neuronal defects have been identified in transgenic mice lacking TrkC (Klein et al., 1994), NT-3 (Ernfors et al., 1994; Farinas

490

INNERVATION OF THE GASTROINTESTINAL TRACT

et al., 1994; Tojo et al., 1995), but not p75NTR (Lee et al., 1992). The ENS, however, is a very complex nervous system. It is thus not enough to conclude from the mere fact of its presence that it is normal. In contrast to GDNF or Ret, NT-3 is clearly not essential for the survival of the pluripotent precursor cell that gives rise to the entire system (Figure 11.3). The loss of stimulation by NT-3 thus causes only a limited defect in the ENS which is limited to subsets, including intrinsic sensory (submucosal) neurons and interneurons. Laboratory animals are moderately perturbed. Even a highly abnormal ENS provides these animals with enough function to survive and gain weight under laboratory conditions. Other receptors or growth factors might compensate for the loss of NT-3 or TrkC. For example, a TrkC knockout does not decrease the numbers of sympathetic neurons; moreover, the knockout of TrkC also fails to prevent expression of TrkA and TH (Henderson, 1996), which are sympathoadrenal markers. NT-3, nevertheless, is still thought to play a physiologically important role in sympathoadrenal development. A receptor other than TrkC would have to be stimulated for it to do so. Detailed anatomical investigations and physiological analyses of motility or secretion have not been carried out in mice with knockouts of genes encoding TrkC, NT-3, or p75NTR. While a partially functional, if not necessarily normal, ENS can arise in the absence of NT-3, the relatively poor weight gain and limited survival of mice that lack NT-3 probably reflect the abnormal ENS that develops in the absence of NT-3 (Farinas et al., 1994).

A CYTOKINE PLAYS A ROLE IN THE DEVELOPMENT OF ENTERIC MOTOR NEURONS Ciliary neurotrophic factor (CNTF) was initially detected because it promotes the survival of chick ciliary ganglion neurons and could be extracted from the eye (Adler et al., 1979). A variety of developing and mature neurons were subsequently demonstrated to respond to CNTF, which has now been cloned and sequenced (Sendtner et al., 1994). CNTF is a member of the distantly related group of molecules that constitute the cytokine family, which includes leukaemia inhibitory factor (LIF), interleukin-6 (IL-6), interleukin-II (IL-II), oncostatin M (OSM), and cardiotrophin-1 (Sendtner et al., 1994; DeChiara et al., 1995; Pennica et al., 1995). CNTF differs from other cytokines of this family, with the possible exception of cardiotrophin-1, in that it primarily acts on neurons. There is a trimeric CNTF receptor (CNTFR), which can be activated by CNTF or LIF (Davis et al., 1991). This receptor is assembled from three molecular components. One of these, CNTFRα, binds CNTF, but is a peripheral membrane protein. The other three subunits of the receptor are transmembrane proteins. One, gp130, is common to the receptor for IL-6, while the other, LIFRβ, binds LIF (Gearing et al., 1991, 1992; Ip et al., 1992; Davis et al., 1993; Pennica et al., 1995). Gp130 and LIFRβ are signal-transducing molecules. A dimer of gp130 responds to IL-6, while a dimer of gp130 and LIFRβ responds to LIF. The three molecular subunits of the tripartite CNTFR are recruited to form a complex by CNTF stimulation. After CNTF binds to CNTFRα, the three components join to form the tripartite complex (Davis et al., 1993). LIF is thus able to mimic effects of CNTF, but CNTF is only able to mimic responses to LIF if CNTFRα is present. In the absence of CNTFRα,

DEVELOPMENT OF THE ENTERIC NERVOUS SYSTEM

491

CNTF neither binds to, nor activates the three subunits. The signal transduction events triggered by CNTF begin with the formation of the CNTFRα/LIFRβ/gp130 complex and involves the dimerisation of LIFRβ with gp130. The transducing regions of the transmembrane subunits then activate Jak tyrosine kinases, which are constitutively associated with the cytosolic domains of each of the three components (Kishomoto et al., 1994; Stahl et al., 1994). CNTFRα is restricted to the nervous system, which explains the neural specificity of the actions of CNTF (Kishimoto, Taga and Akira, 1994; Stahl et al., 1994). Expression of mRNA encoding CNTF can be detected in the developing bowel by using reverse transcriptase and the polymerase chain reaction (Rothman, Chen and Gershon, 1994); nevertheless, levels of CNTF in embryonic and fetal animals are very low (Sendtner et al., 1994). CNTFRα, however, is expressed by many cells of the developing nervous system, including the fetal ENS (Ip et al., 1993). Natural or targeted mutations that prevent CNTF expression, or the knockout of genes encoding CNTF, does not cause striking developmental abnormalities in either mice (Masu et al., 1993) or humans (Takahashi et al., 1994). It has been estimated that ~2.5% of the Japanese population may carry mutations that prevent expression of CNTF. In contrast to almost all other secreted proteins, furthermore, CNTF lacks a signal sequence. The protein should therefore not be transferred into the lumen of the cisternal space of cells that produce it and remain cytosolic. CNTF would thus not be expected to pass through the Golgi, be included in secretory vesicles, or be subject to secretion by exocytosis. If CNTF is secreted, therefore, the mechanism would have to be a novel one that has not yet been discovered. Because CNTF lacks a signal sequence, it has been proposed that CNTF may be an emergency factor that is released only by cells that are damaged or killed due to injury. Cell death occurs prominently during ontogeny; therefore, it is conceivable that CNTF could play a role in development even if it cannot be secreted by healthy cells. There is evidence, however, that suggests that CNTF itself may not be the developmentally important cytokine. Although development appears to proceed relatively normally in CNTF knockout mice, animals with targeted deletions of CNTFRα display profound motor and other defects at birth (DeChiara et al., 1995). These observations suggest that there may be another cytokine, yet to be discovered, that can activate CNTFRα. Mice that lack CNTFRα do not eat and die with a massively dilated bowel shortly after birth (DeChiara et al., 1995). The number of neurons that contain substance P (SP) or nitric oxide synthase (NOS) immunoreactivities are greatly reduced in the enteric plexuses of these animals (A. Kirchgessner, M.-T. Liu, T. DeChiara, and M. Gershon, unpublished data). Significantly, very few SP- or NOS-immunoreactive axons can be detected in the circular muscle layer of the bowel of CNTFRα knockout mice. Since motor neurons that excite smooth muscle contain SP (Brookes, Steele and Costa, 1991b; Costa et al., 1996) and motor neurons that inhibit smooth muscle contain NOS (Bult et al., 1990; Stark, Bauer and Szurszewski, 1991; Costa et al., 1992, 1996; Konturek et al., 1993; Young et al., 1993), it is likely that the knockout of CNTFRα interferes with the development of enteric motor neurons. Since LIFRβ is a component of the tripartite cytokine receptor that responds to CNTF, it is not surprising that an almost identical defect is also seen in the gut of LIFRβ knockout mice. These observations suggest that the still unknown ligand that is responsible for the activation of CNTFRα during development plays a critical role in the development

492

INNERVATION OF THE GASTROINTESTINAL TRACT

of the ENS, and especially in the ontogeny of motor neurons. While enteric ganglia can be detected in mice lacking CNTFRα or LIFRβ, the absence of motor neurons renders the ENS non-functional in these animals. The disparity between the syndromes induced by the knockout of CNTF and CNTFRα strongly implies that there is another endogenous ligand for CNTFRα in the fetal gut. LIF, OSM, IL-6, IL-II, and cardiotrophin-1 do not require CNTFRα and thus are probably not the yet-to-be-discovered CNTFRα ligand (DeChiara et al., 1995). Sympathetic neurons and their precursors respond to LIF (Ernsberger, Sendtner and Rohrer, 1989). CNTF (Saadat, Sendtner and Rohrer, 1989), and cardiotrophin-1 (Pennica et al., 1995), which promote the development of these cells and cause their phenotype to change from catecholaminergic to cholinergic. The ability of the cytokines to affect sympathetic neural precursors suggests that CNTF (and/or the postulated endogenous ligand for CNTFRα) might also act on enteric neurons that arise from the common sympathoadrenal-enteric progenitor. Cells that respond to CNTF would thus be anticipated to be in the mash-1-dependent TC cell lineage (Figure 11.3). Messenger RNA encoding CNTFRα has been found to be expressed in the fetal bowel and to be developmentally regulated (Rothman, Chen and Gershon, 1994). Messenger RNA encoding CNTFRα has also been demonstrated to be enriched in the crest-derived population of cells immunoselected with antibodies to p75NTR (Chalazonitis et al., 1998). When cultures of these p75NTR-immunoselected crest-derived cells are exposed to CNTF or LIF, the development of neurons and glia is strongly promoted. These actions of CNTF and LIF are concentration-dependent and additive, even at supramaximal concentrations, with those of NT-3 (Chalazonitis, Rothman and Gershon, 1995). The additive nature of the ability of CNTF/LIF and NT-3 to promote neuronal development is compatible with either the possibility that the cells that respond to cytokines are not the same as those that respond to neurotrophins or with the possibility that the same cells respond, but through different receptors and transduction pathways. The latter is known to be the case, but the former has not yet been excluded.

THE TERMINAL BOWEL OF ls/ls AND sl/sl MICE IS AGANGLIONIC We have seen, thus far, that the defects encountered in the gut of mice that lack GDNF/ Ret, Mash-1, CNTFRα, or LIFRβ are variable in their severity, but they all involve large regions of the bowel. Local effects, such as the loss of neurons in the oesophagus of mash-1 knockout mice or the sparing of the same neurons in c-ret knockout mice can be related to a defined lineage of enteric neuronal precursors, in this case, the cells that migrate to the bowel from the truncal level of the neural crest. A local defect, in which some of the neurons of a given lineage are missing, while others of the same lineage, in different regions of the bowel, develop normally, is more difficult to explain. The aganglionoses of the terminal colon seem, at first glance, to be particularly troubling. This region of the bowel receives both vagal and sacral crest-derived progenitors. Why neurons should fail to arise in the terminal colon, while other cells in the vagal and sacral territory of distribution

DEVELOPMENT OF THE ENTERIC NERVOUS SYSTEM

493

develop in the same animals, is not immediately obvious. This type of localised lesion in the ENS cannot be explained as the result of the absence of a factor, such as GDNF, which is required by neurons that are also found elsewhere in the bowel (Figure 11.3). If a missing factor is responsible for aganglionosis of the terminal colon, therefore, the factor would have to be necessary for the development of enteric neurons and glia only in the terminal gut. Such a factor might affect the crest-derived cells, but the factor could equally well influence the non-neuronal cells that establish the microenvironment of the terminal colon or both these cells and the neural crest derivatives. Congenital megacolon develops in the colon of lethal spotted (ls/ls) and piebaldlethal mutant mice (sl/sl) (Lane, 1966). This condition, inherited in both strains as an autosomal recessive, is due to aganglionosis of the terminal bowel and is quite similar to the aganglionosis and associated megacolon that is found in human patients with Hirschsprung’s disease (Bolande, 1971). Although the ls/ls and sl/sl mice were the first animals known to exhibit congenital megacolon, additional mutations that give rise to an analogous syndrome have now been discovered. These mutations include aganglionoses inherited as a recessive trait in species other than mice, such as the spotting lethal rat (sl/sl)(Ceccherini et al., 1995; Watanabe et al., 1995a,b; Gariepy, Cass and Yanagisawa, 1996; Karaki et al., 1996; Teramoto et al., 1996) and the spotted rabbit (en/en) (Bodeker et al., 1995). Megacolon can also be inherited as a dominant trait, as in the murine mutant, dominant megacolon (Dom) (Kapur et al., 1996; Puliti et al., 1996). In all of the animal mutations associated with aganglionosis, the affected region of the bowel is the terminal colon, and the coats of the animals are also spotted. It is not hard to understand why, in the affected animals, the relatively normally innervated region of the colon should become dilated proximal to the terminal aganglionic zone to give the appearance of megacolon. The ENS is essential for normal propulsive intestinal motility; therefore, when the ganglia of the ENS are not present, the bowel is as effectively obstructed as if a physical barrier were present. The reflexes normally mediated by the ENS do not occur in the aganglionic segment. Nerve fibers are present in the aganglionic bowel, which is not denervated; however, nerve fibers cannot, by themselves, take the place of the complex neuronal microcircuits that are present in enteric ganglia. The coats of affected animals become spotted, because there are no melanocytes in the white regions. Melanocytes, like enteric neurons, are formed by émigrés from the neural crest. Clearly, therefore, the genetic lesion in the congenital aganglionosis is brought about by a defect that affects the neural crest. The defect, however, does not affect all of the derivatives of the neural crest, nor does it affect all enteric neurons, or even all melanocytes. Enteric neurons and melanocytes can each develop in the face of the genetic lesion; however, in localised regions, the terminal bowel or scattered patches of skin, these cells fail to appear in the mutant animals. The reproducible correlation of an aganglionic terminal colon with spotted coats is compatible with the possibility that a common factor is required for completing the formation of the ENS and populating the skin with melanocytes. Since enteric neurons form elsewhere in the bowel and melanocytes arise outside the white spots, the putative common factor cannot be required for the differentiation of enteric neurons or melanocytes. It could however, be required for the completion of the process of colonisation of the bowel and skin by cells from the neural crest.

494

INNERVATION OF THE GASTROINTESTINAL TRACT

AGANGLIONOSIS OF THE TERMINAL COLON AND A SPOTTING COAT ARE CAUSED BY THE GENETIC ABSENCE OF THE ENDOTHELIN-B RECEPTOR OR ITS LIGAND, ENDOTHELIN-3 The discovery of the genetic abnormalities that occur in ls/ls and sl/sl mice resulted from unanticipated observations of the phenotypes encountered in knockout mice which had been prepared to investigate a different problem. Aganglionosis in ls/ls mice is caused by a mutation in the gene encoding big endothelin-3, the large and inactive precursor of the peptide hormone, endothelin-3 (ET-3) (Baynash et al., 1994). The genetic defect prevents the proteolytic cleavage of big ET-3 to the active hormone. The more severe aganglionosis of sl/sl mice (Hosoda et al., 1994), spotting lethal rats (Ceccherini et al., 1995; Gariepy, Cass and Yanagisawa, 1996; Karaki et al., 1996), and some Hirschsprung’s patients (Puffenberger et al., 1994), is caused by mutations of genes encoding the endothelin-B (ETB) receptor. This receptor has a high affinity for ET-3, although the affinity of ETB for ET-3 is not higher than that for endothelins 1 or 2 (ET-1, ET-2) (Yanagisawa, 1994). In contrast to ET-1 and ET-2, however, ET-3 is not an effective ligand for the endothelin A (ETA) receptor. ETB, therefore, is the only receptor that can physiologically be activated by ET-3. The three endothelins constitute a family, each of which is a 21 amino acid peptide (Sakamoto et al., 1993; Rubanyi and Polokoff, 1994). The endothelins stimulate ETA, ETB, or both. The endothelin receptors are serpentine (seven transmembrane domains) and G protein-coupled. The potency of each of the endothelins for stimulating ETB is about the same, but potencies differ greatly at ETA, where ET-1 > ET-2 >> ET-3 (Yanagisawa, 1994). The discovery of ET-1 resulted from its properties as a strong vasoconstrictor produced by vascular endothelial cells (Yanagisawa et al., 1988). The endothelins and the endothelin receptors, however, have now been found to be widely distributed (Rubanyi and Polokoff, 1994). The endothelins are synthesised with a signal sequence (as a preproendothelin) that enables the cell that secretes them to translocate the molecules across the membranes of the rough endoplasmic reticulum (RER) to the cisternal space. From the cisternal space of the RER, the proteins can be transported to the Golgi apparatus, and packaged into vesicles for secretion by exocytosis. The signal sequence is cleaved co-translationally, to convert the preproendothelins to the inactive big endothelins which are secreted. The big endothelins are split by a specific membrane-bound metalloprotease, the endothelinconverting enzyme-1 (ECE-1), to produce the active ET-1, ET-2, or ET-3 (Xu et al., 1994). Transgenic knockout mice that fail to produce ET-1 exhibit craniofacial defects arising from abnormal development of the first branchial arch (Kurihara et al., 1994). Missense mutations in the genes encoding ETB occur in sl/sl mice and so are associated with megacolon and coat spotting (Hosoda et al., 1994). Similar mutations can be found in the analogous human locus in a subset of human patients with Hirschsprung’s disease (Puffenberger et al., 1994). When the gene encoding ETB is knocked out in transgenic mice by homologous recombination, the colon becomes aganglionic, mimicking the megacolon seen in sl/sl mice, confirming the causal relationship between ETB and the aganglionosis of the terminal colon (Hosoda et al., 1994). Lethal spotting, the aganglionosis of the colon in rats, also arises as a result of a mutation that prevents expression of the rat

DEVELOPMENT OF THE ENTERIC NERVOUS SYSTEM

495

ETB; in this case the mutation is an interstitial deletion in an exon of the gene encoding ETB (Ceccherini et al., 1995; Karaki et al., 1996). An arginine is replaced by a tryptophan residue in the C-terminus of big ET-3 as a result of the mutation of the ET-3 gene in ls/ls mice (Baynash et al., 1994). The consequence of this mutation is that ECE-1 is unable to convert big ET-3 to ET-3. Since big endothelins are inactive, this mutation essentially ablates ET-3, suggesting that loss of the ligand, ET-3, in ls/ls mice has the same effect as the loss of the receptor, ETB is sl/sl mice. The suggestion is confirmed by the occurrence of an identical aganglionosis of the terminal colon in mice in which the gene encoding ET-3 has been knocked out by homologous recombination. While it is clear that ET-3 and ETB play essential roles in the development of the ENS of the terminal colon, it is not clear what these roles are. Since ETB can be stimulated by ET-1 and ET-2, as well as by ET-3 (Yanagisawa, 1994), the acquisition of aganglionosis in animals that lack only ET-3 indicates that the endothelins do not circulate in the fetus at times that are critical in the formation of the ENS (Baynash et al., 1994). On the contrary, the endothelins are likely to be factors that act only locally, so that the loss of the ligand, ET-3, will be manifest in the area where it is expressed. Conceivably, aganglionosis might result, when ET-3 is absent, because crest-derived neuronal precursors express ETB and require ET-3 stimulation. This possibility does not explain why ET-3 should only be required by those crest-derived precursors that develop in the terminal colon. A second possibility is that ET-3 might be required by a non-neuronal resident cell of the terminal colon, which provides a component of the local microenvironment that crest-derived precursors need to colonise the colon or develop there. Obviously, these two possibilities are not mutually exclusive. Both crest-derived and non-neuronal cells may express ETB and be influenced by ET-3. In any case, the issue of why enteric neurons, other than those in the colon, develop perfectly well in the absence of ET-3 or ETB is an issue that must be resolved.

AGANGLIONOSIS IS NOT EXPLAINED BY THE FAILURE OF ET-3 TO STIMULATE CREST-DERIVED NEURAL PRECURSORS IN THE COLON One hypothesis that has been advanced to account for the critical role of ET-3 in the formation of the ENS of the terminal colon is that ET-3 may be an autocrine growth factor produced and required by enteric crest-derived neural and glial precursors and by melanocytes (Baynash et al., 1994). This hypothesis nicely explains the universal correlation of aganglionosis of the colon with spotting of the coats in animals lacking either ET-3 or ETB. The autocrine hypothesis supposes that the migrating crest-derived cells that colonise the intestine express big ET-3, ECE-1, and ETB. The cells can thus convert big ET-3 to ET-3 and respond to the ET-3 they produce. If ET-3, were to be an autocrine growth factor required by cells migrating from the vagal crest, however, the congenital absence of either ET-3 or ETB would be expected to cause an aganglionosis of the entire bowel (like the defects induced by the knockouts of GDNF or Ret; Figure 11.3), not just the terminal colon, as is actually seen in ls/ls (Baynash et al., 1994), sl/sl mice (Hosoda et al., 1994; Kapur et al., 1995) and spotting lethal rats (Ceccherini et al., 1995; Gariepy, Cass and Yanagisawa, 1996; Karaki et al., 1996).

496

INNERVATION OF THE GASTROINTESTINAL TRACT

The vagal crest after all, colonises the entire gut below the oesophagus and rostral stomach (Figure 11.1). If the autocrine hypothesis applied only to sacral crest-derived cells, no part of the bowel would be expected to be completely aganglionic, because both vagal and sacral cells participate in colonising the post-umbilical gut and there is no region of the ENS that is entirely sacral in origin. Conceivably, the target of ET-3’s action might not be the colon. The target could instead be the pre-migratory crest (Lahav et al., 1996). For example, a critical mass of vagal crest cells might be needed in order for them to extend their colonising range to the terminal colon. The pre-migratory vagal crest cells may require stimulation by ET-3 in order to proliferate sufficiently to reach this critical mass. In the absence of the stimulation of their putative ETB receptors by ET-3, therefore, the population of vagal crest-derived cells might be too small for any of them to migrate as far as the terminal colon, which lies at the end of their migratory pathway. If this hypothesis were to be correct, then ET-3 could still be an autocrine growth factor that produces only localised effects. Applied to the skin, an analogous argument would account for coat spotting. The starting population of melanogenic precursors might have to be built up by ET-3-stimulated proliferation in order to reach a size that is big enough to spread throughout the skin. If the population is too small, patches devoid of melanocytes may occur, thereby accounting for the spotted coats of the affected animals. This hypothesis that ET-3 is a mitogen that is required by pre-migratory (or earlymigrating) crest cells has been supported by the demonstration that ET-3 actually does stimulate the proliferation of avian cells cultured from the pre-migratory neural crest (Lahav et al., 1996). These cells are induced by the addition of ET-3 in vitro to proliferate massively and, after having done so, to go on to develop primarily as melanocytes. These data thus only partially support the hypothesis that the production of a critical mass of crest-derived precursors depends on stimulation by ET-3. The hypothesis works well when applied to the skin, but the apparent action of ET-3 to skew development toward the generation of melanocytes is not compatible with the production of large number of enteric neuronal precursors; therefore, to explain why a loss of stimulation by ET-3, should deprive the gut of neuronal precursors, one would have to postulate that the action of ET-3 on pre-migratory crest cells in vivo is different from the effect that has been observed in vitro. That presumption could be valid, but there is, as yet no evidence to support it. In any case, if one assumes that the action of ET-3 on crest cells in vivo is different from that which it exerts in vitro, then the in vitro data cannot be used to support any hypotheses about the development of crest cells in the gut. The crest-derived cell population that colonises the bowel does not contain any cells with a melanogenic potential (Rothman et al., 1990, 1993) and is itself a proliferating population (Teitelman et al., 1981; Baetge and Gershon, 1989; Baetge, Pintar and Gershon, 1990). The ability of crestderived cells to proliferate as they migrate would argue that the putative effect of ET-3 as a mitogen is more likely to be manifest within the gut than in the pre-migratory crest. In this location, moreover, the loss of the option to develop as melanocytes might permit ET-3 to activate mitosis without pushing the crest cells counterproductively to develop along a melanocytic lineage. Crest-derived cells from the bowel are not identical to their predecessors in the pre-migratory crest. Relative to the pre-migratory crest, the developmental potential of

DEVELOPMENT OF THE ENTERIC NERVOUS SYSTEM

497

intra-enteric crest-derived cells has been reduced (Rothman et al., 1990; Lo and Anderson, 1995), enteric crest-derived cells respond differently from pre-migratory crest cells to NT-3 (Kalcheim, Carmeli and Rosenthal, 1992; Pinco et al., 1993; Chalazonitis et al., 1994), and the two cells express different cell surface proteins (Pomeranz et al., 1991). Crestderived cells immunoselected from the gut have recently also been found to respond differently than pre-migratory crest cells to ET-3. Whereas ET-3 is a mitogen for premigratory crest cells that increases the number of melanocytes differentiating in vitro (Lahav et al., 1996), the in vitro action of ET-3, when it is applied to enteric crest-derived cells, immunoselected from the developing murine bowel with antibodies to p75NTR, is actually to inhibit the generation of neurons (Wu et al., 1999). Exposure of these cultured crest-derived cells to ET-3 or to other ETB agonists decreases the number of neurons that arise in vitro. The addition of an ETB antagonist to the medium, moreover, does not have any visible effect at all on the differentiation of neurons, suggesting that neuronal differentiation in vitro cannot be influenced significantly by an autocrine effect of ET-3 on ETB receptors on crest-derived cells. Such an effect should be inhibited by an ETB antagonist. Since ET-3 inhibits the formation of enteric neurons from crest-derived precursors, it seems paradoxical that the absence of ET-3 or ETB is associated with aganglionosis of the terminal colon. The loss of an inhibitor of neuronal differentiation would seem at first glance to be an unlikely cause of the failure of neurons to develop in a given region of the bowel. The crest-derived precursors of neurons, however, are migratory, while neurons are sedentary cells. As a result, the differentiation of crest-derived precursors should also mean that neurogenic cells stop migrating. The implication of this consideration is that timing of neuronal differentiation is probably very important in the colonisation of the bowel by crest-derived cells. If neurons differentiate too soon, then the pool of migrating neurogenic cells will become depleted. Should this occur, then the bowel distal to the point where neurons differentiate prematurely will not become colonised by crest-derived precursors. ET-3 might thus be required to inhibit the premature development of crestderived cells as neurons and glia and to sustain them in a migratory (possibly still proliferating) state. This is a speculative hypothesis; however, the observation that ET-3 affects the development of enteric crest-derived cells in vitro, implies that crest-derived cells must express ETB receptors and be ET-3 responsive, even if the effect of ET-3 on them is not what it was first thought likely to be. The fact that ET-3 affects crest-derived cells, however, does not mean that these are the only cells in the bowel upon which ET-3 acts. THE AGANGLIONOSIS THAT ARISES IN ET-3 OR ETB-DEFICIENT ANIMALS IS NOT NEURAL CREST AUTONOMOUS Aganglionosis of the terminal colon could be induced if the absence of ET-3 or the loss of the ETB receptor were to make the enteric microenvironment of this region of the bowel inhospitable for colonisation by crest-derived cells (Kapur, Yost and Palmiter, 1993; Rothman, Goldowitz and Gershon, 1993; Coventry et al., 1994; Kapur et al., 1995). Expression of the ETB receptor by non-neuronal cells of the bowel wall might thus, as noted previously, be an indirect contributor to aganglionosis. Alternatively, crest-derived cells may secrete a factor in response to ET-3 that causes non-neuronal cells of the terminal colon to make the enteric microenvironment amenable for colonisation by émigrés

498

INNERVATION OF THE GASTROINTESTINAL TRACT

from the neural crest. The advancing wave of crest-derived pioneers migrating to and within the developing gut cannot be recognised morphologically, but the wave front can be detected indirectly by explanting the bowel and determining whether neurons develop in vitro (Rothman and Gershon, 1982, 1984; Rothman, Tennyson and Gershon, 1986). Neurons will arise in explants of the normal murine terminal colon removed from fetuses after stage 33 (Jacobs-Cohen et al., 1987). In contrast, cultures prepared from the terminal 2 mm of an ET-3 deficient (ls/ls) gut never contain neurons, no matter when they are explanted (Rothman and Gershon, 1984; Rothman, Tennyson and Gershon, 1986). The final segment of the ls/ls bowel, therefore, is presumptively aganglionic, suggesting that viable crest-derived cells may never enter this region. Crest-derived cells from a variety of co-cultured sources, including the ganglion-containing proximal gut of ls/ls mice, can enter explants of terminal bowel from control mice and give rise to neurons in vitro; however, in similar co-culture experiments, no source of crest-derived cells can colonise an ls/ls terminal colon (Jacobs-Cohen et al., 1987); the ls/ls colon also lacks the normal ability of the bowel, when co-cultured with crest cells, to promote their expression of gut-appropriate phenotypes (Coulter, Gershon and Rothman, 1988). These observations suggest that the wall of the terminal colon, not just crest-derived cells, is abnormal in mice that lack ET-3. The observations that normal crest-derived cells will not enter the presumptive aganglionic region of the ls/ls colon are inconsistent with the hypothesis that a deficiency of ET-3 causes aganglionosis only because ET-3 is an autocrine factor that crest-derived cells require. On the one hand, ls/ls crest-derived cells are unable to produce active ET-3, but they can give rise to neurons if they are allowed to colonise a normal colon. Moreover, wild-type crest-derived cells should be capable of synthesising active ET-3, but they cannot colonise the ls/ls colon. There may be additional sources of ET-3 in the colon and the crest-derived cells may not be the only cells in the wall of the colon that express ETB. Both neurogenic cells and non-neuronal cells could express ETB and each could play roles in enabling crest-derived cells to complete the colonisation of the bowel. The idea that ET-3 affects non-neuronal cells of the gut wall has been supported by data derived from studies of chimeric animals. For example, aganglionosis of the terminal colon does not develop in ls/ls-C3H aggregation chimeric mice, as long as some of the cells in the bowel wall are C3H; moreover, ls/ls neurons, identified with an endogenous marker (3-glucuronidase activity), are found throughout the colon of the chimeric animals (Rothman, Goldowitz and Gershon, 1993). Similarly, mutant neurons, marked by the expression of the DBH-lacZ transgene, develop in the terminal colon of aggregation chimeras constructed between wild-type and either ls/ls (Kapur, Yost and Palmiter, 1993; Coventry et al., 1994) or sl/sl embryos (Kapur et al., 1995). While it is conceivable, in chimeric embryos, that the autocrine secretion of ET-3 by their normal crest-derived neighbours, might rescue ET-3-deficient cells, it is inconceivable that such an effect could rescue crest-derived cells that do not express ETB. Since the cells of ls/ls mice lack ET-3, but not ETB, they might thus respond to ET-3 supplied by nearby wild-type cells in chimeric embryos. In contrast sl/sl cells lack ETB (Hosoda et al., 1994); nevertheless, they too colonise the terminal colon in chimeric mice. Cells that lack ETB should not be able to respond to ET-3 and thus are not amenable to rescue by nearby normal cells. The simple autocrine model, therefore, in which crest-derived cells are both the source and

DEVELOPMENT OF THE ENTERIC NERVOUS SYSTEM

499

target of ET-3, does not explain the development of aganglionosis in animals that lack ET-3 or ETB. If ET-3 is not an essential autocrine growth factor, why are enteric neurons missing from the terminal colon when either ET-3 or ETB are absent? It has been proposed that there are critical cells that are “downstream” from the cells that express ETB and that these “downstream” cells are required for the terminal gut to become colonised by crest-derived cells (Kapur et al., 1995). This hypothesis thus invokes the participation by another cell in the process by which crest-derived cells colonise the colon, thereby accounting for why the defect in ETB is not crest autonomous. In a chimeric embryo, the normal crest-derived cells presumably respond to ET-3 and signal the “downstream” cells, which then allow both normal crest-derived cells and their ET-3- or ETB-deficient neighbors to develop in the terminal colon. The hypothesis supposes, therefore, that an intercellular conversation is initiated by ET-3. An alternative hypothesis, which also invokes the participation of another cell, is less complex. This alternative possibility considers that the additional, nonneuronal cell, itself expresses ETB and is ET-3 responsive. Both hypotheses thus postulate a role for a non-neuronal cell in making it possible for the colon to be colonised by crestderived neuronal precursors; however, the first idea sees ET-3 as stimulating only the crest-derived cells, while the second idea dispenses with the intercellular conversation and assumes that the non-neuronal cell responds directly to ET-3. These ideas can each be tested. The location of ETB is important in these evaluations. This idea that the colon itself is abnormal in animals lacking ET-3 is supported by observations on the progression down the gut of vagal crest-derived cells, visualised by their expression of the DBH-LacZ transgene. The progression of these cells in ls/ls mice is entirely normal until the cells reach the colon. As soon as the cells cross the ileo-cecal junction, however, the migration of vagal crest-derived cells becomes abnormal and remains so until migration ceases short of the end of the colon (Kapur, Yost and Palmiter, 1992; Coventry et al., 1994). The ability of crest-derived cells to migrate in ET-3 deficient ls/ls mice is thus normal until the cells become exposed to the microenvironment of the colon, which by implication must itself be abnormal. The suggestion that the colonic microenvironment is intrinsically abnormal has been confirmed by experiments in which segments of wild-type or ls/ls mouse colon were back-transplanted into neural crest migration pathways of quail embryos (Rothman, Goldowitz and Gershon, 1993). The grafts each survive well; however, the behaviour of the quail crest-derived cells is different depending upon which of the two grafts they encounter (Figure 11.5). When quail crest-derived cells encounter a wild-type colon, they enter the grafts and/or pass on through (Figure 11.5A). The normal mouse colon thus does not obstruct the migration of cells from the quail crest. In marked contrast, quail crest-derived cells do not enteric backgrafts of ls/ls colon (Figure 11.5B). Instead, they form large ganglia just proximal to, but outside, the mouse gut. The quail crest-derived cells in these experiments are normal and thus should not be deficient either in ET-3 or in ETB. Their failure to enter the ls/ls colon, therefore cannot be explained by their loss of an essential autocrine growth factor. Rather, the absence of ET-3 in the mouse gut would appear to have prevented its colonisation by the presumably normal crest-derived cells of the quail. The simplest explanation of these data is that the absence of ET-3, which is confined in a transplantation experiment to the grafted tissue, affects non-neuronal elements of the bowel wall, causing them to render the

500

INNERVATION OF THE GASTROINTESTINAL TRACT

Figure 11.5 Quail crest-derived cells can migrate into backtransplanted segments of wild-type, but not ls/ls fetal mouse gut. (A) Quail crest-derived cells, identified by demonstrating HNK-1 immunoreactivity, migrate through the control mouse gut. Mouse cells can be distinguished from those of the quail, by their more scattered and abundant nuclear heterochromatin (Feulgen stain). (B) Quail cells fail to enter the ls/ls colon. Instead they form an immense ganglion at the border of the murine tissue.

bowel resistant to an influx of cells from the neural crest. The absence of active ET-3 in the ls/ls colon evidently makes the environment of the colon one that crest-derived cells do not enter. In summary, crest-derived cells appear to be capable of colonising the gut and giving rise to enteric neurons, whether or not they produce ET-3 or respond to ET B; however, the colon becomes abnormal in the absence of ET-3/ETB stimulation and resistant to colonisation by crest-derived cells, whether or not these crest-derived cells produce ET-3 or express ETB. In fact, abnormalities of smooth muscle-produced basal laminae have been described, both in the colon of ls/ls mice, and in human patients with Hirschsprung’s disease (see below).

BASAL LAMINAE ARE ABNORMAL IN THE COLON OF ls/ls MICE Extracellular matrix abnormalities occur in the aganglionic colon of ls/ls mice (Payette et al., 1988; Tennyson et al., 1990; Rothman et al., 1996). Similar abnormalities have

DEVELOPMENT OF THE ENTERIC NERVOUS SYSTEM

501

also been observed in human patients with Hirschsprung’s disease (Parikh et al., 1992; Parikh et al., 1995). All of these matrix defects include an overabundance and/or maldistribution of molecules that are normally found in basal laminae. Among the molecules that accumulate in the aganglionic colon are laminin, type IV collagen, nidogen, nonsulfated glycosaminoglycans, and proteoglycans. Although the mucosal basal lamina is thicker than normal in the developing colon of ls/ls mice, the most striking abnormality is in the diffuse distribution of these molecules in the mesenchyme of the colon and the surrounding pelvis (Payette et al., 1988; Tennyson et al., 1990; Rothman et al., 1996). This distribution places the accumulated matrix molecules directly in the migratory pathways followed by vagal crest-derived cells within the bowel (Tucker, Ciment and Thiery, 1986) and by sacral crest-derived cells within the pelvis (Pomeranz and Gershon, 1990). Contacts of HNK-1 immunoreactive crest-derived cells migrating in the enteric mesenchyme with laminin-immunoreactive tufts of electron opaque material have been demonstrated by double label electron microscopic immunocytochemistry (Pomeranz et al., 1991). The excessive accumulation of laminin and type IV collagen appears to arise in the ls/ls mouse hindgut before (Payette et al., 1988) crest-derived cells would be expected to colonise the terminal bowel (Jacobs-Cohen et al., 1987). This timing and the location of the matrix abnormalities make it possible that the abnormal extracellular matrix is causally related to the development of aganglionosis. If so, then the accumulation of basal lamina molecules would have to occur independently of the failure of crest-derived cells to develop in the colon. This independence is suggested by the timing of the appearance of the matrix abnormalities and the arrival of crest-derived cells, but the timing alone does not rule out the possibility that the matrix becomes abnormal because crest-derived cells are absent from the terminal bowel of ls/ls mice. Crestderived cells, for example, might secrete a factor that regulates the secretion of matrix components in advance of their migratory front. Direct evidence of whether the matrix abnormalities are a direct or indirect consequence of the ET-3 deficiency in ls/ls mice is thus needed. The accumulation of laminin and type IV collagen in the ls/ls fetal colon has recently been shown to be associated with an increase in the abundance of transcripts encoding the molecules (Rothman et al., 1996). Quantitative Northern analysis has revealed that mRNA encoding the β1 and γ1 chains of laminin, as well as that encoding the αl and α2 chains of type IV collagen, is increased in the ls/ls colon. Messenger RNA encoding laminin α1 is also increased, but because of its relatively low abundance, the increase in laminin αl mRNA had to be demonstrated by reverse transcription in combination with the competitive polymerase chain reaction (RT-cPCR), rather than by Northern analysis. The relative abundance of mRNA encoding laminin α1 is greatest at E11 and then declines as a function of age, until it stabilises at a very low level that is maintained in adult life. The production of laminin αl, therefore, is developmentally regulated. The laminin-1 isoform (α1-β1-γ1) thus is associated with development and is present in the fetal bowel while enteric ganglia are in the process of formation. The age-related decline in the abundance of laminin α1 transcripts occurs in the fetal colon of both wildtype and ls/ls mice; nevertheless, laminin α1 mRNA is significantly more abundant in the ls/ls fetal colon than in that of wild-type animals throughout development. Prior to day E15, cells that synthesise laminin α1 and β1 and those that produce the α2 chain of

502

INNERVATION OF THE GASTROINTESTINAL TRACT

collagen type IV have each been found, by in situ hybridisation, to be most concentrated in the endodermal epithelium. At later ages, however, cells that contain mRNA encoding these molecules are more abundant in the outer gut mesenchyme than in the epithelium, and more of this mRNA is found in the ls/ls than in the wild-type colonic and pelvic mesenchyme. The expression of laminin α1 in ls/ls and wild-type animals has been compared to that in age-matched E15 and newborn c-ret knockout mice (Rothman et al., 1996). As noted above, the entire gut is aganglionic below the oesophagus and proximal stomach of c-ret knockout mice (Figure 11.3); however, this pan-intestinal aganglionosis occurs because enteric neural and glial precursor cells are GDNF/Ret-dependent (Schuchardt et al., 1994; Durbec et al., 1996b; Jing et al., 1996; Sanchez et al., 1996; Trupp et al., 1996), not because of an abnormality in ET-3, the mutation that gives rise to the colonic aganglionosis in ls/ls mice (Baynash et al., 1994). If the increase in laminin αl mRNA that characterises the aganglionic colon of ls/ls is due to the absence of crest-derived cells, then the same defect should occur in the aganglionic gut of c-ret knockout mice. In contrast, this increase in mRNA should not be seen in the intestine of c-ret knockout mice if the defect in ls/ls is caused by the loss of an effect of ET-3 on non-neuronal cells of the colonic and pelvic mesenchyme. The abundance of mRNA encoding laminin αl was found to be the same in the intestines of wild-type and c-ret knockout mice both at E15 or in newborn animals (Rothman et al., 1996); moreover, the amount of laminin immunoreactivity in the colon of E15 and newborn c-ret knockout mice cannot be distinguished from that in agematched controls. These results suggest that the abnormal extracellular matrix of the ls/ls mouse colon is a primary effect of the ET-3 deficiency in these animals, and is not a secondary consequence of the absence of crest-derived cells from the ls/ls colon. The ls/ls defect probably includes an overabundance of the laminin-1 isoform, because the αl, β1, and γ1 subunit are all present in excess in the ls/ls mouse colon. The excess of laminin-1 occurs in the pelvic mesenchyme as well as in the bowel. An overabundance of this molecule, therefore, is present in the paths of each of the populations of crest-derived cell, vagal and sacral, that colonise the colon. The location of the extracellular matrix defect, in ls/ls mice is thus compatible with the idea that the matrix abnormalities are causally related to the pathogenesis of aganglionosis. The extracellular matrix in some regions of the embryo normally inhibits the migration of crest cells, which thus tend to avoid these locations. Matrix molecules that inhibit crest cell migration have been found between the ectoderm and the somites (Erickson, Duong and Tosney, 1992; Oakley et al., 1994), which coincides with the dorsolateral pathway of neural crest migration, the posterior sclerotome (Rickmann, Fawcett and Keynes, 1985; Norris, Stern and Keynes, 1989; Keynes et al., 1990), and the perinotochordal mesenchyme (Pettway, Guillory and Bronner-Fraser, 1990). Basal lamina components, however, have not been observed to accumulate in any of these regions (Norris, Stern and Keynes, 1989; Keynes et al., 1990; Oakley and Tosney, 1991; Oakley et al., 1994). The extracellular matrix in the zones that are normally avoided by migrating crest cells moreover, differs from that found in the aganglionic ls/ls (Payette et al., 1987) or Hirschsprung’s bowel (Holschneider et al., 1994; Miura et al., 1996) in that the regions that normally exclude crest-derived cells also inhibit the outgrowth of axons (Oakley and Tosney, 1991). In contrast, the aganglionic bowel is heavily innervated, both by axons of neurons from the more rostral

DEVELOPMENT OF THE ENTERIC NERVOUS SYSTEM

503

hypoganglionic gut and from extrinsic ganglia (Payette et al., 1987). The enteric aganglionosis, therefore, extends only to crest-derived cells, and does not include their axons. Crest cells normally adhere well to laminin (Lallier and Bronner-Fraser, 1991; Lallier et al., 1994); moreover, the migration of cells away from the neural crest is promoted by laminin-1 (Bilozur and Hay, 1988; Perris, Paulsson and Bronner-Fraser, 1989). The migration of cranial crest cells, furthermore, can be inhibited by the in vivo administration of antibodies to integrins, which antagonise the attachment of crest-derived cells to laminin (Bronner-Fraser, 1985, 1986), as well as by antibodies that recognise the complex formed in situ between laminin and proteoglycans (Bronner-Fraser and Lallier 1988). An overabundance of laminin, therefore, such as that which occurs in the ls/ls colon, should provide an excellent adhesive substrate for the crest-derived cells that colonise the bowel. Certainly, since laminin promotes neurite extension and axonal growth (Manthorpe et al., 1983; Calof and Reichardt, 1985; Lander, Fujii and Reichardt, 1985; Engvall et al., 1986; Kleinman et al., 1988; Liesi et al., 1989) the abundance of laminin-1 in the aganglionic colon of ls/ls mice and human patients with Hirschsprung’s disease probably accounts for the observations that the aganglionic tissue is very well supplied by the axons of neurons the cell bodies of which lie outside the aganglionic zone. The association of a laminin-rich substrate with aganglionosis, however, seems, at least superficially, counter-intuitive. If laminin interferes with the colonisation of the gut, then the response to laminin of crest-derived cells in the bowel would have to be different from those of pre-migratory crest cells, migrating crest cells that have not yet reached the target organ, or axons.

LAMININ-1 PROMOTES THE DEVELOPMENT OF ENTERIC NEURONS Extracellular matrix molecules are biologically active and can affect the terminally differentiated phenotypes expressed by the progeny of neural crest stem cells in vitro (Stemple and Anderson, 1992). The extracellular matrix, therefore, is not just the framework to which migrating crest-derived cells adhere; it is also a source of signalling information and is probably capable of influencing the fate of the crest-derived cells that differentiate in contact with it. Specifically, with regard to the matrix, one of the molecules that accumulates in the aganglionic colon, laminin-1, has been found to promote the in vitro development of cells that express neuronal markers, such as peripherin, neurofilament proteins, neuron specific enolase, or PGP9.5 (Gershon et al., 1993; Pomeranz et al., 1993; Tennyson et al., 1995; Chalazonitis et al., 1997). The promotion of enteric neuronal development by laminin-1 was first demonstrated by using crestderived cells immunoselected from the embryonic avian or fetal rat gut with antibodies to HNK-1. A similar effect is seen when cells are immunoselected from the mouse gut (Chalazonitis et al., 1992, 1997; Gershon et al., 1993; Tennyson et al., 1995) with antibodies to a cell-surface laminin-binding protein, known a LBP110 (Douville, Harvey and Carbonetto, 1988; Kleinman et al., 1988; Kleinman and Weeks, 1989; Jucker et al., 1991).

504

INNERVATION OF THE GASTROINTESTINAL TRACT

AN IKVAV SEQUENCE OF THE α1 SUBUNIT OF LAMININ-1 IS RESPONSIBLE FOR THE PROMOTION OF ENTERIC NEURONAL DEVELOPMENT LBP110 is a non-integrin laminin-binding protein that is similar to a β-amyloid precursor protein (Kibbey et al., 1993). The amino acid sequence, isoleucine-lysine-valine-alaninevaline (IKVAV), of the domain of laminin that binds to LBP110 is located on the laminin α1 chain, near its globular C-terminal end (Sephel et al., 1989a,b; Kleinman et al., 1991). PC12 cells express LBP110 and this expression can be down regulated by transfecting the cells with an antisense amyloid precursor protein cDNA (Kleinman et al., 1991; Kibbey et al., 1993). This antisense treatment inhibits NGF-induced neurite extension on a laminin-1 substrate. These observations have led Hynda Kleinman and her colleagues to suggest that LBP110 is a receptor for laminin-1, which mediates the effects of laminin-1 on neurite outgrowth (Kleinman et al., 1991; Kibbey et al., 1993, 1994). Kleinman and others have also implicated LBP110 as a laminin receptor that mediates many of the responses of non-neuronal cells to laminin (Haralabopoulos et al., 1994; Weeks et al., 1994; Bresalier et al., 1995; Corcoran et al., 1995; Nomizu et al., 1995). Crest-derived cells are the only ones in the developing bowel that express LBP110, which is thus co-localised in the fetal gut with crest markers (Pomeranz et al., 1991); moreover, neurons or glia preferentially differentiate in cultures of cells immunoselected from the fetal mouse gut with antibodies to LBP110 (Figure 11.6A) (Pomeranz et al., 1993). A synthetic peptide that contains the IKVAV sequence (IKVAV-peptide) of the domain of laminin α1 that binds to LBP110, competitively blocks the promotion of enteric neuronal and glial development in vitro by laminin-1 (Figure 11.6B) (Chalazonitis et al., 1992, 1997; Gershon et al., 1993). A nonsense peptide, a peptide with the same amino acids in a different sequence, or a peptide with a sequence found elsewhere in the laminin-1 molecule all fail to influence the response of enteric crest-derived cells to laminin-1. The development of enteric neurons and glia is not affected by the IKVAV-peptide when the crest-derived cells are cultured on poly-D-lysine or fibronectin. The action of the IKVAV-peptide, therefore, is both sequence-specific and laminin-specific. The IKVAVpeptide, furthermore, affects only the differentiation of crest-derived cells and does not reduce the total number of cells in culture. This observation suggests that the IKVAVpeptide does not interfere with the adhesion of cells to laminin-1. It seems likely that adhesion depends more on the binding of laminin by plasmalemmal integrins (BronnerFraser, 1985, 1986), than LBP110 (Kleinman et al., 1991). Antibodies to laminin αl, like the IKVAV-peptide, inhibit the ability of laminin-1 to promote the development of neurons (Chalazonitis et al., 1997). This effect of antibodies to laminin α1 is not shared either by pre-immune sera, nor is it shared by antibodies to the β1 chain of laminin. Neither the antibodies to laminin α1, nor those to laminin β1, cause cells to detach from a laminin-1 containing substrate. The observation that the IKVAV-peptide does not reduce the total number of cells in culture suggests that it does not block the adhesion of most of the cells in culture. The possibility remains, however, that the adherence of a neurogenic subset of crest-derived cells, the size of which is negligible in comparison to the total number of cells in cultures, is inhibited by the IKVAV-peptide. To investigate this possibility, laminin-1 was added in

DEVELOPMENT OF THE ENTERIC NERVOUS SYSTEM

505

Figure 11.6 The ability of laminin-1 to promote the development of enteric neurons in vitro is blocked by an IKVAV-peptide. Crest-derived cells were immunoselected from the fetal murine bowel with antibodies to p75NTR and cultured on a substrate of poly-D-lysine. The presence of neurons in the cultures is demonstrated by immunostaining with antibodies to PGP9.5. (A) The development of neurons is stimulated by the addition of soluble laminin. (B) When laminin is added, together with an IKVAV-peptide, the development of neurons is inhibited.

soluble form to cells that had previously been allowed to adhere for 24 h to poly-D-lysine (Chalazonitis et al., 1992, 1997; Gershon et al., 1993; Tennyson et al., 1995). When this was done, laminin-1 was equally effective in promoting the differentiation of neurons and glia as when it was used as a component of the substrate upon which cells were plated. This experimental paradigm rules out the explanation that laminin-1 promotes neuronal development because it selectively permits the adherence of cells able to differentiate as neurons. The observation, however, that because soluble laminin-1 is efficacious, does not establish that soluble laminin-1 is necessarily able to stimulate the receptors that promote enteric neuronal development. Soluble laminin-1might have to bind to the substrate before it can activate receptors on cell surfaces. Such a requirement could explain the observation that the IKVAV-peptide is a pure antagonist, and not an agonist at the LBP110 site. Still, whether laminin-1 activates LBP110 as a bound or soluble molecule, the fact that it is effective in promoting neuronal differentiation many hours after cells have become adherent indicates that promotion of neuronal development by laminin-1

506

INNERVATION OF THE GASTROINTESTINAL TRACT

is due to an effect on differentiation and not adherence. In this sense, therefore, laminin-1 is a growth factor for enteric neurons. As is also the case when crest-derived cells are exposed to other growth factors, a rapid, but transient induction of the expression of the c-fos proto-oncogene accompanies the response of immunoselected crest-derived cells to laminin-1 (Chalazonitis et al., 1992, 1997; Gershon et al., 1993; Tennyson et al., 1995). Expression of c-fos becomes apparent within one hour of the addition of laminin-1, but after 24 h c-fos expression can no longer be detected. The laminin-1-induced expression of c-fos, like laminin-1-promoted neuronal and glial development, is antagonised by the IKVAV-peptide, but not by control peptides. These observations suggest that both of these responses are LBP110-mediated. Since the IKVAV-peptide is an antagonist and not an agonist, it seems likely that the binding of the IKVAV domain to laminin-1 is necessary but not sufficient to activate the LBP110 receptor. LAMININ-1 EXERTS DIFFERENT EFFECTS ON PRIMARY CREST CELLS AND CREST-DERIVED CELLS IMMUNOSELECTED FROM THE FETAL BOWEL Pre-migratory and early-migrating crest cells express integrins and thus attach to laminin (Bronner-Fraser, 1986; Bilozur and Hay, 1988; Bronner-Fraser and Lallier, 1988; Lallier and Bronner-Fraser, 1991; Lallier et al., 1994), which these cells encounter in the embryonic mesenchyme and when they come into contact with basal laminae (Martins-Green and Erickson, 1987; Erickson, Loring and Lester, 1989; Pomeranz and Gershon, 1990; Pomeranz et al., 1991). In contrast to crest-derived cells immunoselected from the fetal gut, pre-migratory and early-migrating crest cells do not express LBP110 (Pomeranz et al., 1991). This receptor is not expressed by the crest-derived cells that colonise the bowel until after these cells enter the gut. Neural crest stem cells do not express LBP110 and exposure of these cells to laminin-1 does not induce them to differentiate as neurons (Stemple and Anderson, 1992, 1993). In fact, laminin stimulates the migration of crest cells, not their differentiation (Bilozur and Hay, 1988). The ability of crest-derived neuronal precursors to respond to laminin-1 must thus be a characteristic the cells acquire, either while migrating to the bowel, or after they enter it. It is likely that the difference between the responses to laminin-1 of neural crest stem cells on the one hand, and their enteric crest-derived successors on the other, can be accounted for by the failure of the stem cells to express LBP110 and the acquisition of LBP110 by the crest-derived neuronal precursors in the gut. The absence of LBP110 from crest-derived cells migrating to the bowel is probably important in enabling these cells to reach their destinations. Clearly, if laminin were to induce crest cells to differentiate as neurons as soon as such cells encountered it, differentiation would occur too soon and the bowel would never become colonised. Neurons do not migrate, their precursors do. Differentiation of crest-derived cells as neurons must thus be postponed until after the bowel has been colonised. The delay in LBP110 expression, therefore, means that enteric neuronal precursors can adhere to laminin-1 while they are migrating to the gut, without being induced by laminin α1 to turn into neurons before they reach the bowel. Within the gut, moreover, crest-derived cells

DEVELOPMENT OF THE ENTERIC NERVOUS SYSTEM

507

acquire LBP110 asynchronously. Presumably, the vagal crest-derived émigrés that express LBP110 as soon as they enter the proximal bowel are the precursors of neurons in foregut ganglia. Those that acquire LBP110 later are likely to be able to move more distally and differentiate into neurons in the midgut and eventually the hindgut (Pomeranz et al., 1991). The asynchrony in the timing of LBP110 expression within the bowel may thus make it possible for a subset of vagal crest-derived cells to delay differentiating and continue to migrate until they reach the terminal bowel. AN EXCESS OF LAMININ-1 IN THE COLON MAY CAUSE NEURONS TO DIFFERENTIATE PREMATURELY The seemingly paradoxical association of an overabundance of laminin-1 with aganglionosis in the terminal colon of mice that lack ET-3 and human patients with Hirschsprung’s disease may be causally related to the stimulation of the LBP110 on crest-derived cells by the excess laminin-1. Since the IKVAV domain of the laminin α1 subunit of laminin-1, acting on crest-derived cells that have acquired LBP110, would be expected to promote their differentiation as neurons, the laminin-1 rich environment of the ET-3-deficient colon and pelvis may provoke premature neuronal differentiation. ET-3, moreover, is itself an inhibitor of neuronal differentiation. The crest-derived cells migrating to and within the ET-3-deficient gut, therefore, are subjected simultaneously to positive and negative influences that are likely to combine to provide these cells with a very strong stimulus to differentiate along a neuronal pathway. Because ET-3 is absent, the crestderived cells are deprived of a factor that inhibits them from differentiating as neurons, while at the same time they encounter an environment rich in laminin-1 that stimulates them to do so. If crest-derived cells respond to these strong influences to differentiate as neurons before they have completed their task of colonising the bowel, the terminal colon, which is the last part of the gut to be colonised, will become aganglionic. A congenital deficiency of ET-3 or its receptor, ETB, may thus prevent the formation of ganglia in the terminal bowel by exerting direct and indirect effects, which synergise to promote neuronal differentiation at the expense of migration and proliferation. Since the overabundance of laminin α1 is present throughout the ls/ls colon (Rothman et al., 1996), this hypothesis predicts that the progression of the vagal crest-derived cells down the bowel of ls/ls mice would become abnormal for the first time when these cells enter the proximal colon. This prediction has been confirmed. The descent of crest-derived cells has been mapped in ls/ls mice that express the DBH-lacZ transgene (Coventry et al., 1994). The migration of these cells appears to be comparable in ls/ls and wild-type animals until the crest-derived cells cross the ileo-cecal junction. As soon as the cells enter the ls/ls colon, the rate at which they descend becomes slower and their progress become erratic. Laminin-1 accumulates not only in the colon of ls/ls mice, but also in the pelvic mesenchyme that surrounds it. The hypothesis that the excess of laminin-1 and the ET-3 deficiency of ls/ls mice combine to induce premature neuronal differentiation thus also predicts that sacral crest-derived precursors would stop short of the gut (Payette et al., 1988). Ectopic ganglia have been found outside the terminal bowel, in the pelvis of ls/ls mice (Rothman and Gershon, 1984; Payette et al., 1987). These ganglia actually fuse with myenteric ganglia in the hypoganglionic zone of the ls/ls colon. To do so, the ectopic ganglia have to penetrate through the

508

INNERVATION OF THE GASTROINTESTINAL TRACT

longitudinal muscle layer to form odd structures that are half in the bowel and half out. The ectopic ganglia, which are not present in wild-type mice, are likely to be sacral crestderived cells that have differentiated before reaching the colon. A new hypothesis to explain the aganglionosis of ET-3 or ETB deficiency is that the genetic defect causes the developmental regulation of the secretion of components of basal laminae to become delayed in the colon. This delay leads to an excess of laminin-1 in the matrix through which the crest-derived cells that colonise the colon must migrate. This excess drives inadequately resistant crest-derived progenitors to differentiate before reaching the terminal bowel. The hypothesis accounts for the fact that the aganglionosis of ET-3 deficiency is not neural crest autonomous.

ET-3 AFFECTS BOTH CREST-DERIVED AND NON-NEURONAL CELLS OF THE COLON Both crest-derived and non-crest-derived cells appear to be targets for ET-3 in the colon. Laminin is known, in the gut of normal mice, to be produced by epithelial cells and by one or more cells of the enteric mesenchyme (Simo et al., 1991). In situ hybridisation has revealed that mRNA transcripts encoding subunits of laminin are predominantly found in the epithelium early in development (Rothman et al., 1996). Later, however, about the time that ganglia begin to form, synthesis of these subunits switches to the mesenchyme. The locations of transcripts encoding subunits of lamini-1 and those encoding collagen IV are qualitatively similar in control and ls/ls mice, but quantitatively different. Messenger RNA encoding each of these components of basal lamina is more concentrated in the enteric mesenchyme of fetal ls/ls than in that of wild-type mice. Crest-derived cells are thus not the source of laminin-1 and are not responsible for its overproduction in the ls/ls bowel. The overproduction of laminin-1 that occurs in the aganglionic bowel of ls/ls mice does not occur in the aganglionic gut of c-ret knockout mice. The overproduction of laminin-1 in the ls/ls animals is thus explained most straightforwardly by the hypothesis that ET-3 down-regulates the secretion of laminin-1 by pelvic mesenchymal cells, including those of the fetal bowel. This hypothesis that ET-3 downregulates laminin-1 secretion implies that fetal mesenchymal cells and/or epithelial cells express ETB. Precursors of smooth muscle, fibroblasts, and ICCs are all found in the fetal enteric mesenchyme. ETB has been demonstrated on smooth muscle cells in the mature large (Okabe et al., 1995) and small intestines (Yoshinaga et al., 1992). Both of which respond to ET-3. When these receptors are first acquired by developing smooth muscle cells is unknown. The totally aganglionic bowel of c-ret knockout mice contains transcripts of mRNA encoding ET-3 and those encoding ETB (J. Chen, T. Rothman and M. Gershon, unpublished data). This observation confirms that the biosynthesis of these molecules in the bowel is not confined to crest-derived cells. In situ hybridisation carried out in mice in which the crest-derived cells are marked by their expression of the DBH-lacZ transgene (R. Kapur and M. Yanagisawa, reported at the 1996 meeting of the American Motility Society) has shown that ETB mRNA is present both in lacZ-expressing and non-lacZ-expressing cells. The location of the lacZ-negative cells that contain ETB transcripts is close to presumptive ganglia. In this location, the cells

DEVELOPMENT OF THE ENTERIC NERVOUS SYSTEM

509

could be ICCs or a subset of smooth muscle. This location is interesting, because ICCs have been found to be abnormal in patients with Hirschsprung’s disease (Yamataka et al., 1995; Vanderwinden et al., 1996).

INTERSTITIAL CELLS OF CAJAL ARE ABNORMAL IN THE AGANGLIONIC BOWEL OF PATIENTS WITH HIRSCHSPRUNG’S DISEASE The identity and origin of ICCs has been a contentious issue for a long time (Kobayashi et al., 1989; Sanders, 1996; Torihashi, Ward and Sanders, 1997). At one time, ICCs were thought to be fibroblasts (Cook and Burnstock, 1976). More recently, they have been postulated to be modified or primitive smooth muscle cells (Faussone-Pellegrini, 1985; Torihashi et al., 1993). This idea remains a current concept, but ICCs can be distinguished from banal smooth muscle, by their expression of, and dependence on, the c-kit protooncogene (Ward et al., 1994; Huizinga et al., 1995; Torihashi et al., 1995; Ward et al., 1995). This gene encodes a receptor tyrosine kinase, Kit, and is allelic with white spotting (W) (Besmer, 1991). The ligand for Kit has been given various names including Kit ligand (KL), stem cell factor and Steel factor. The gene encoding KL is allelic with steel (sl). A great deal of evidence indicates that Kit and KL are necessary for the development and/ or maintenance of ICCs. For example, both W (Ward et al., 1994; Huizinga et al., 1995) and sl (Ward et al., 1995) mutations inhibit the development of ICCs. Administration of antibodies that neutralise Kit is associated with the disappearance of ICCs from the mouse gut (Maeda et al., 1992; Torihashi et al., 1995). Finally, the in vitro development of Kit-expressing ICCs (Figures 11.1C, 11.7) is dependent on the presence of KL in the culture medium (Wu, Rothman and Gershon, 1996). Myogenic intestinal slow waves are impaired when the network of ICCs is lost or fails to develop (Maeda et al., 1992; Ward et al., 1994, 1995; Huizinga et al., 1995; Torihashi et al., 1995); therefore, ICCs are probably the pacemakers for these waves (Sanders, 1996). Intestinal motility becomes abnormal after the disruption of the ICC network causes slow waves to be lost and the bowel dilates in a manner that resembles that seen in aganglionosis. In the mature longitudinal muscle and elsewhere in the bowel wall during fetal life, ICCs and smooth muscle cells express markers in common (Torihashi, Ward and Sanders, 1997). Markers expressed by both ICCs and smooth muscle include the intermediate filament protein, desmin, and smooth muscle isoforms of actin and myosin. The crest-derived cell marker, Ret (Pachnis, Mankoo and Costantini, 1993; Schuchardt et al., 1994; Tsuzuki et al., 1995) is not expressed by ICCs (Torihashi, Ward and Sanders, 1997). These observations suggest that ICCs may be derived from a common smooth muscleICC progenitor. Studies of stably marked crest-derived cells in avian interspecies chimeras have also indicated that ICCs are mesenchymal derivatives and not cells of neural crest origin (Lecoin, Gabella and Le Douarin, 1996). The morphology of Kit-immunoreactive ICCs differs in various sites within the wall of the bowel (Figure 11.7) and may represent subtypes of ICC that diverge at different times from the putative common smooth muscleICC progenitor (Torihashi, Ward and Sanders, 1997). The evidently different subtypes of ICCs that surround myenteric ganglia and those that reside in the deep muscle plexus,

510

INNERVATION OF THE GASTROINTESTINAL TRACT

Figure 11.7 ICCs develop in vitro and can be demonstrated in the wall of the gut with antibodies to Kit. Note that some of the cells are thin and orientated parallel to muscle fibers, while others are thicker and more random in their orientation.

circular, and longitudinal muscle layers have been proposed to constitute functionally distinct cell classes. When crest-derived and non-crest-derived cells are separated by immunoselection with antibodies to p75NTR, ICCs develop in the cultures of non-crestderived cells (Figure 11.1C) (Wu, Rothman and Gershon, 1996). The abnormality of ICCs in the aganglionic region of the bowel of patients with Hirschsprung’s disease (Yamataka et al., 1995; Vanderwinden et al., 1996) demonstrates that the genetic defect in these patients is not restricted to cells of neural crest origin. Despite the fact that the normal pattern of ICCs is disrupted and their number reduced in the aganglionic colon in Hirschsprung’s disease, at least some ICCs are still present. ICCs can also be found in the terminal colon of ls/ls mice and in the aganglionic intestine of c-ret knockout mice. Again, in both cases the abundance of ICCs is less than that in wildtype mice; moreover, the network of ICCs is abnormal in the mutant bowel (Wu, Rothman and Gershon, 1996). ICCs thus can develop in the absence of ET-3 (in the ls/ls gut) and also in the absence of neurons (in the c-ret knockout mouse). The reduced numbers and the disruption of the network of ICCs in the aganglionic bowel of patients with Hirschsprung’s disease and ls/ls mice, however, may be the secondary result of the absence of neurons. Neurons may thus influence the development of at least some ICCs. This possibility is supported by the observation, made by in situ hybridisation, that enteric neurons

DEVELOPMENT OF THE ENTERIC NERVOUS SYSTEM

511

contain mRNA encoding KL (Torihashi, Ward and Sanders, 1997). Enteric neurons are probably, therefore, a source of KL. Since the isoform of KL that is physiologically active is membrane-bound (Miyazawa et al., 1995; Wehrle-Haller and Weston, 1995), neurons probably must make contact with the plasma membranes of ICCs to stimulate the Kit they express. A requirement for such cell-to-cell interaction might explain the close spatial relationship of one of the ICC subtypes to myenteric ganglia (Sanders, 1996). Stimulation of Kit on ICCs by KL expressed on the plasma membranes of neurons, therefore, could explain the ICC abnormalities associated with aganglionosis in ls/ls mice and patients with Hirschsprung’s disease. Since ICCs are not totally absent when the bowel is aganglionic, however, neurons must not be the only source of KL in the gut. In fact, mRNA encoding KL, as well as that encoding Kit, can be detected in the aganglionic bowel of c-ret knockout mice (J. Chen, T. Rothman and M. Gershon, unpublished data). Since ICCs develop in ls/ls mice, they clearly do not require ET-3 for their development; nevertheless, even if they do not need ET-3 to develop, they might still express ET B and be ET-3 responsive. The abnormal number and aberrant network of ICCs in the aganglionic Hirschsprung’s and ls/ls colon are consistent with this idea. The location of the non-neuronal cells in the DBH-lacZ mouse colon that contain ETB mRNA coincides with that of one subset of Kit-immunoreactive ICCs. ET-3 stimulates the development of smooth muscle in vitro (Wu et al., 1999). In the presence of ET-3 or an ETB agonist, the number of desmin-immunoreactive cells increases. ET-3 could thus promote the development of smooth muscle and/or ICCs from their common progenitor. Conceivably, this progenitor cell secretes more laminin-1 than does either of its successors. Laminin-1 is developmentally regulated (Rothman et al., 1996). The α1 chain decreases as a function of developmental age, while the β1 and γ1 subunits do not. Presumably, therefore, as laminin α1 declines it is replaced by another α subunit, such as α2. This putative shift in secretion from laminin-1 to laminin-2 would make the enteric mesenchyme far less likely to promote neurogenesis. If laminin-2 is secreted by mature smooth muscle and/or ICCs, then the shift from laminin-1 to laminin-2 secretion may reflect the development of smooth muscle and ICCs from their common progenitor, events that are likely to be promoted by ET-3. A delay in the development of smooth muscle and ICCs, resulting from the absence of ICCs, may thus prolong the secretion of laminin-1, causing excessive amounts of it to be present in the bowel at the time the colon is colonised by crest-derived cells.

DIFFERENT GENETIC ABNORMALITIES CAN GIVE RISE TO HIRSCHSPRUNG’S DISEASE Neuromuscular defects of the human gut are common perinatal problems. In addition to Hirschsprung’s disease, these disorders include a variety of related conditions such as hypoganglionosis, neuronal intestinal dysplasias (hyperganglionosis), ganglion cell immaturity, and dysganglionoses. Other neuromuscular defects that may involve ENS abnormalities include hypertrophic pyloric stenosis, volvulus, and intussusception. Hirschsprung’s disease itself is not uncommon and occurs in about 1/5000 live births (Angrist et al., 1995). A subset of patients with Hirschsprung’s disease have loss-of-function mutations in

512

INNERVATION OF THE GASTROINTESTINAL TRACT

the RET proto-oncogene (Edery et al., 1994; Romeo et al., 1994; Angrist et al., 1995; Borrello et al., 1995; Pasini et al., 1995). These patients account for only a relatively small fraction of the total set of Hirschsprung’s disease (Edery et al., 1994; Romeo et al., 1994; Angrist et al., 1995). Identical RET abnormalities can be associated with either long or short segment diseases; moreover, some patients not only have aganglionosis of the colon, but may also exhibit multiple endocrine neoplasia type A (which is more commonly associated with gain-of-function mutations in RET), maternal deafness, talipes, and malrotation of the gut. Distinctly different phenotypes can thus arise from identical mutations in RET. The relationship between the RET genotype and the Hirschsprung’s phenotype is, therefore, not entirely obvious; furthermore, the relatively low frequency of RET mutations in Hirschsprung’s disease indicates that additional genetic and/or environmental conditions must explain the majority of cases. Mutations in ETB have also been associated with Hirschsprung’s disease (Puffenberger et al., 1994). Again, however, Hirschsprung’s disease can occur in patients who exhibit mutations in neither ETB nor RET and mutations can occur in these genes without necessarily giving rise to the Hirschsprung’s disease phenotype (Puffenberger et al., 1994). Mutations of genes encoding ET-3, have also recently been linked to some cases of Hirschsprung’s disease. In the ET-3-deficient patients, the phenotype resembles that of ls/ls mice. Hirschsprung’s disease occurs in a setting of cutaneous pigmentary abnormalities and is combined with a Waardenburg type 2 phenotype (Shah-Waardenburg syndrome) (Edery et al., 1996; Hofstra et al., 1996). Since a wide variety of mutations (many of which are still to be identified) predispose toward Hirschsprung’s disease, the condition is a multigene abnormality (Puffenberger et al., 1994; Angrist et al., 1995). The genetic abnormalities found in patients with Hirschsprung’s disease are not completely comparable to the related mutations in animals. For example, c-ret knockout causes aganglionosis in mice only when the mutated gene is homozygous (Schuchardt et al., 1994; Durbec et al., 1996b). Even mice that are doubly heterozygous for both c-ret and ls do not become aganglionic (T. Rothman, M. Gershon and F. Costantini, unpublished observations). In contrast, patients with Hirschsprung’s disease who have RET mutations have only been heterozygous. The background and environment thus affect the penetrance of the aganglionic phenotype when Ret is mutated. The ET-3/ETB-deficient mice would seem to be useful models for studying the pathogenesis of Hirschsprung’s disease. From both a genetic and an anatomical point of view, these models strikingly resemble Hirschsprung’s disease. The molecular abnormalities that have been found in the extracellular matrix of the ET-3 deficient ls/ls mice (Rothman et al., 1996) also occur in patients with Hirschsprung’s disease (Parikh et al., 1992, 1995). These include a massive increase in the thickness of the muscularis mucosa and the excessive secretion of laminin and type IV collagen. Fujimoto and colleagues have concluded that laminin and type IV collagen accumulate at the sites where ganglia will form before neurogenesis begins, suggesting that the promotion of the development of neurons by laminin may determine where ganglia arise (Fujimoto et al., 1989). In any case, the sum of observations made from studies of ls/ls (Rothman and Gershon, 1984; Jacobs-Cohen et al., 1987; Payette et al., 1988; Kapur, Yost and Palmiter, 1993; Rothman, Goldowitz and Gershon, 1993; Rothman et al.,1993; Coventry et al., 1994), sl/sl (Kapur et al., 1995), and Dom (Kapur et al., 1996) mice have all indicated that the murine aganglionosis involves an intrinsic

DEVELOPMENT OF THE ENTERIC NERVOUS SYSTEM

513

abnormality of the colon in addition to, or instead of, defects of crest-derived neuronal precursors. It is thus highly likely that the pathogenesis of Hirschsprung’s disease will also be the result of abnormalities of both crest and non-crest-derived cells.

SUMMARY The ENS is a unique region of the PNS. It is larger, more complex, and more independent than any other. Its development must thus involve factors and/or mechanisms that are different from those which operate in the formation of extra-enteric ganglia. The ENS is formed by the progeny of immigrants that migrate to the bowel from vagal, rostral-most truncal, and sacral regions of the neural crest. Crest-derived progenitors are multipotent when they arrive in the gut, although their developmental potential is less than it was in the pre-migratory crest. Enteric crest-derived émigrés, for example, do not give rise to ectomesenchyme or melanocytes. Enteric crest-derived cells, however, can develop as sympathetic neurons or Schwann cells if they are experimentally induced to re-migrate and develop outside the bowel. The fate of crest-derived émigrés in the gut is thus powerfully influenced by the enteric microenvironment. The microenvironment does not, however, work alone. The effects exerted by the growth factors and matrix constituents of the bowel wall vary on different crest-derived cells because the receptors expressed by crest-derived cells also vary. The receptors expressed by crest-derived cells reflect the lineages and sublineages into which these cells have been sorted. Two such lineages have been identified. One, which gives rise to serotonergic neurons and other types of cell, is transiently catecholaminergic, born early, and mash-1-dependent. The other, which gives rise to CGRP-containing and other types of neuron, is never catecholaminergic born late, and mash-1-independent. Many signals that influence the differentiation and/or survival of enteric neurons have been identified, One is GDNF, which is the functional ligand that activates the Ret receptor. This factor acts early and is evidently required by the multipotent crest-derived cell that colonises the bowel distal to the oesophagus and most proximal stomach. A second factor is the neurotrophin, NT-3, which is additive with a cytokine that activates the CNTFRα, but which is not CNTF or LIF. The knockout of genes, such as those encoding GDNF or Ret that encode growth or transcription factors required by crestderived precursors while these cells are still multipotent produce large defects in the ENS. The knockout of genes, such as those encoding CNTFRα, which are required later in ontogeny, produce more limited neuronal abnormalities, although the defects may still be lethal if the cells that are lost, such as a motor neurons, are functionally vital. A different kind of abnormality, which involves the entire ENS, but in a restricted region, occurs in mice lacking ET-3 or ETB. The terminal colon of these animals is aganglionic. The mutation probably affects both the crest-derived precursors of enteric neurons and the nonneuronal cells that produce the matrix through which the crest-derived cells that colonise the gut must migrate. The precursors of smooth muscle and/or ICCs oversecrete the α1 subunit of laminin-1 which promotes neuronal differentiation. ET-3 inhibits the in vitro differentiation of crest-derived cells as neurons. The crest-derived cells that attempt to colonise the ET-3-deficient colon may thus encounter an overly strong drive to differentiate at the same time that they are deprived of a differentiation inhibitor. The result appears to

514

INNERVATION OF THE GASTROINTESTINAL TRACT

be premature differentiation at the expense of migration, leaving the terminal portion of the bowel uncolonised by crest-derived neuronal or glial precursors. Hirschsprung’s disease of human has been associated with mutations of RET, ETB, and ET-3, but is a multigene abnormality that has not yet been adequately explained. All of the factors that play roles in the development of the ENS are potential targets of mutations that causes birth defects in humans. The genetic bases of hypoganglionosis, neuronal intestinal dysplasias, and intestinal dysganglionoses, as well as additional genes that contribute to Hirschsprung’s disease are likely soon to be discovered. Understanding the pathogenesis of Hirschsprung’s disease and birth defects of the ENS can be expected to provide improved means of treating these conditions and, eventually, some hope of preventing them.

REFERENCES Adler, R., Landa, K.B., Manthorpe, M. and Varon, S. (1979). Cholinergic neuronotrophic factors: intraocular distribution of trophic activity for ciliary neurons. Science, 204, 1434–1436. Allan, I.J. and Newgreen, D.F. (1980). The origin and differentiation of enteric neurons of the intestine of the fowl embryo. American Journal of Anatomy, 157, 137–154. Anderson, D.J. (1989). The neural crest cell lineage problem: neuropoiesis? Neuron, 3, 1–12. Anderson, D.J., Carnahan, J.F., Michelsohn, A. and Patterson, P.H. (1991). Antibody markers identify a common progenitor to sympathetic neurons and chromaffin cells in vivo and reveal the timing of commitment to neuronal differentiation in the sympathoadrenal lineage. Journal of Neuroscience, 11, 3507–3519. Angrist, M., Bolk, S., Thiel, B., Puffenberger, E.G., Hofstra, R.M., Buys, C.H. et al. (1995). Mutation analysis of the RET receptor tyrosine kinase in Hirschsprung disease. Human Molecular Genetics, 4, 821–830. Baetge, G. and Gershon, M.D. (1989). Transient catecholaminergic (TC) cells in the vagus nerves and bowel of fetal mice: relationship to the development of enteric neurons. Developmental Biology, 132, 189–211. Baetge, G., Pintar, J.E. and Gershon, M.D. (1990). Transiently catecholaminergic (TC) cells in the bowel of the fetal rat: precursors of noncatecholaminergic enteric neurons. Developmental Biology, 141, 353–380. Baetge, G., Schneider, K.A. and Gershon, M.D. (1990). Development and persistence of catecholaminergic neurons in cultured explants of fetal murine vagus nerves and bowel. Development, 110, 689–701. Barbacid, M. (1993). The trk family of neurotrophin receptors: molecular characterisation and oncogenic activation in human tumors. In Molecular Genetics of Nervous System Tumors, edited by A.J. Levine and H.H. Schmidek, pp. 123–135. New York: Wiley and Sons. Baroffio, A., Dupin, E. and Le Douarin, N.M. (1988). Clone-forming ability and differentiation potential of migratory neural crest cells. Proceedings of the National Academy of Sciences of the USA, 85, 5325–5329. Bayliss, W.M. and Starling, E.H. (1899). The movements and innervation of the small intestine. Journal of Physiology (London), 24, 99–143. Bayliss, W.M. and Starling, E.H. (1900). The movements and the innervation of the large intestine. Journal of Physiology (London), 26, 107–118. Baynash, A.G., Hosoda, K., Giaid, A., Richardson, J.A., Emoto, N., Hammer, R.E. et al. (1994). Interaction of endothelin-3 with endothelin-B receptor is essential for development of epidermal melanocytes and enteric neurons. Cell, 79, 1277–1285. Berthoud, H.R., Jedrzejewska, A. and Powley, T.L. (1990). Simultaneous labeling of vagal innervation of the gut and afferent projections from the visceral forebrain with dil injected into the dorsal vagal complex in the rat. Journal of Comparative Neurology, 301, 65–79. Berthoud, H.R. and Powley, T.L. (1992). Vagal afferent innervation of the rat fundic stomach: morphological characterization of the gastric tension receptor. Journal of Comparative Neurology, 319, 261–276. Besmer, P. (1991). The kit ligand encoded at the murine Steel locus: a pleiotropic growth and differentiation factor. Current Opinion in Cell Biology, 3, 939–946. Bilozur, M.E. and Hay, E.D. (1988). Neural crest migration in 3D extracellular matrix utilizes laminin, fibronectin, or collagen. Developmental Biology, 125, 19–33. Birren, S.J., Lo, L. and Anderson, D.J. (1993). Sympathetic neuroblasts undergo a developmental switch in trophic dependence. Development, 119, 597–610. Black, I.B. (1978). Regulation of autonomic development. Annual Review of Neuroscience, 1, 183–214.

DEVELOPMENT OF THE ENTERIC NERVOUS SYSTEM

515

Blaugrund, E., Pham, T.D., Tennyson, V.M., Lo, L., Sommer, L., Anderson, D.J. et al. (1996). Distinct subpopulations of enteric neuronal progenitors defined by time of development, sympathoadrenal lineage markers and Mash-1-dependence. Development, 122, 309–320. Bodeker, D., Turck, O., Loven, E., Wieberneit, D. and Wegner, W. (1995). Pathophysiological and functional aspects of the megacolon-syndrome of homozygous spotted rabbits. Zentralblatt Fur Veterinarmedizin – Reihe A, 42, 549–559. Bolande, R.P. (1975). Hirschsprung’s disease, aganglionic or hypoganglionic megacolon. Animal model: aganglionic megacolon in piebald and spotted mutant mouse strains. American Journal of Pathology, 79, 189–192. Bornstein, J.C. and Furness, J.B. (1988). Correlated electrophysiological and histochemical studies of submucous neurons and their contribution to understanding enteric neural circuits. Journal of the Autonomic Nervous System, 25, 1–13. Bornstein, J.C. and Furness, J.B. (1992). Enteric neurons and their chemical coding. In Advances in the Innervation of the Gastrointestinal Tract, edited by G.E. Holle and J.D. Wood, pp. 101–114. New York: Elsevier Science B.V. Bornstein, J.C., Furness, J.B. and Kunze, W.A.A. (1994). Electrophysiological characterization of myenteric neurons – how do classification schemes relate? Journal of the Autonomic Nervous System, 48, 1–15. Borrello, M.G., Smith, D.P., Pasini, B., Bongarzone, I., Greco, A., Lorenzo, M.J. et al. (1995). RET activation by germline MEN2A and MEN2B mutations. Oncogene, 11, 2419–2427. Bothwell, M. (1996). p75NTR: a receptor after all. Science, 272, 506–507. Bresalier, R.S., Schwartz, B., Kim, Y.S., Duh, Q.Y., Kleinman, H.K. and Sullam, P.M. (1995). The laminin α1 chain Ile-Lys-Val-Ala-Val (IKVAV)-containing peptide promotes liver colonization by human colon cancer cells. Cancer Research, 55, 2476–2480. Bronner-Fraser, M. (1985). Alterations in neural crest migration by a monoclonal antibody that affects cell adhesion. Journal of Cell Biology, 101, 610–617. Bronner-Fraser, M. (1986). An antibody to a receptor for fibronectin and laminin perturbs cranial neural crest development in vivo. Developmental Biology, 117, 528–536. Bronner-Fraser, M. and Fraser, S.E. (1988). Cell lineage analysis reveals multipotency of some avian neural crest cells. Nature, 335, 161–164. Bronner-Fraser, M. and Fraser, S. (1989). Developmental potential of avian trunk neural crest cells in situ. Neuron, 3, 755–766. Bronner-Fraser, M. and Lallier, T. (1988). A monoclonal antibody against a laminin-heparan sulfate proteoglycan complex perturbs cranial neural crest migration in vivo. Journal of Cell Biology, 106, 1321–1329. Brookes, S.J.H. and Costa, M. (1990). Identification of enteric motor neurones which innervate the circular muscle of the guinea pig small intestine. Neuroscience Letters, 118, 227–230. Brookes, S.J.H., Mayo, R. and Costa, M. (1992). The neuronal basis of motor activity in the guinea-pig small intestine: a computer model. Gastroenterology, 102, A431 [Abstract]. Brookes, S.J.H., Song, Z.M., Steele, P.A. and Costa, M. (1992). Identification of motor neurons to the longitudinal muscle of the guinea pig ileum. Gastroenterology, 103, 961–973. Brookes, S.J.H., Steele, P.A. and Costa, M. (1991a). Calretinin immunoreactivity in cholinergic motor neurones, interneurones and vasomotor neurones in the guinea-pig small intestine. Cell and Tissue Research, 263, 471–481. Brookes, S.J.H., Steele, P.A. and Costa, M. (1991b). Identification and immunohistochemistry of cholinergic and non-cholinergic circular muscle motor neurons in the guinea-pig small intestine. Neuroscience, 42, 863–878. Bülbring, E. and Gershon, M.D. (1967). 5-Hydroxytryptamine participation in the vagal inhibitory innervation of the stomach. Journal of Physiology (London), 192, 823–846. Bult, H., Boeckxstaens, G.E., Pelckmans, P.A., Jordaens, F.H., Van Maercke, Y.M. and Herman, A.G. (1990). Nitric oxide as an inhibitory non-adrenergic non-cholinergic neurotransmitter. Nature, 345, 346–347. Calof, A.L. and Reichardt, L.F. (1985). Response of purified chick motoneurons to myotube conditioned medium: laminin is essential for the substratum-binding, neurite outgrowth-promoting activity. Neuroscience Letters, 59, 183–189. Carnahan, J.F., Anderson, D.J. and Patterson, P.H. (1991). Evidence that enteric neurons may derive from the sympathoadrenal lineage. Developmental Biology, 148, 552–561. Carter, B.D., Kaltschmidt, C., Kaltschmidt, B., Offenhauser, N., Bohm-Matthaei, R., Baeuerle, P.A. et al. (1996). Selective activation of NF-κB by nerve growth factor through the neurotrophin receptor p75. Science, 272, 542–545. Ceccherini, I., Zhang, A.L., Matera, I., Yang, G., Devoto, M., Romeo, G. et al. (1995). Interstitial deletion of the endothelin-B receptor gene in the spotting lethal (sl) rat. Human Molecular Genetics, 4, 2089–2096. Chalazonitis, A. (1996). Neurotrophin-3 as an essential signal for the developing nervous system. Molecular Neurobiology, 12, 39–53.

516

INNERVATION OF THE GASTROINTESTINAL TRACT

Chalazonitis, A., Rothman, T.P., Chen, J., Lamballe, F., Barbacid, M. and Gershon, M.D. (1994). Neurotrophin3 induces neural crest-derived cells from fetal rat gut to develop in vitro as neurons or glia. Journal of Neuroscience, 14, 6571–6584. Chalazonitis, A., Rothman, T.P., Chen, J., Vinson, E.N., MacLennan, A.J. and Gershon, M.D. (1998). Promotion of the development of enteric neurons and glia by neuropoietic cytokines: interactions with neurotrophin-3. Developmental Biology, 198, 343–365. Chalazonitis, A., Rothman, T.P. and Gershon, M.D. (1995). Ciliary neurotrophic factor (CNTF). and neurotrophin-3 (NT-3) potentiate one another in promoting the enteric neuronal development. Neuroscience, 25, 1545 [Abstract]. Chalazonitis, A., Tennyson, V.M., Kibbey, M.C., Rothman, T.P. and Gershon, M.D. (1997). The α1 subunit of laminin-1 promotes the development of neurons by interacting with LBP110 expressed by neural crestderived cells immunoselected from the fetal mouse gut. Journal of Neurobiology, 33, 118–138. Chalazonitis, A., Tennyson, V.M., Rothman, T.P. and Gershon, M.D. (1992). Selective isolation of neural and glial precursors from the developing murine bowel with antibodies to a 110 kDa cell surface laminin binding protein[Abstract]. Neuroscience, 18, 1109. Chalazonitis, A., Pham, T.D., Rothman, T.P., Distephano, P.S., Bothwell, M., Blair-Flynn, J., Tessarollo, L. and Gershon, M.D. (2001). Neurotrophin-3 is required for the survival differention of subsets of developing enteric neurons. J. Neuroscience, 21, 5620–5636. Chao, M.V. and Hempstead, B.L. (1995). p75 and Trk: a two-receptor system. Trends in Neurosciences, 18, 321–326. Choi-Lundberg, D.L. and Bohn, M.C. (1995). Ontogeny and distribution of glial cell line-derived neurotrophic factor (GDNF) mRNA in rat. Brain Research. Developmental Brain Research, 85, 80–88. Cochard, P., Goldstein, M. and Black, I.B. (1978). Ontogenetic appearance and disappearance of tyrosine hydroxylase and catecholamines in the rat embryo. Proceedings of the National Academy of Sciences of the USA, 75, 2986–2990. Colamarino, S.A. and Tessier-Lavigne, M. (1995). The axonal chemoattractant netrin-1 is also a chemorepellent for trochlear motor axons. Cell, 81, 621–629. Cook, R.D. and Burnstock, G. (1976). The ultrastructure of Auerbach’s plexus in the guinea-pig. II. Non-neuronal elements. Journal of Neurocytology, 5, 195–206. Cooke, H.J. (1989). Role of the “little brain” in the gut in water and electrolyte homeostasis. FASEB Journal, 3, 127–138. Corcoran, M.L., Kibbey, M.C., Kleinman, H.K. and Wahl, L.M. (1995). Laminin SIKVAV peptide induction of monocyte/macrophage prostaglandin E2 and matrix metalloproteinases. Journal of Biological Chemistry, 270, 10365–10368. Costa, M., Brookes, S.J.H., Steele, P.A., Gibbins, I., Burcher, E. and Kandiah, C.J. (1996). Neurochemical classification of myenteric neurons in the guinea-pig ileum. Neuroscience, 75, 949–967. Costa, M. and Furness, J.B. (1979). On the possibility that an indoleamine is a neurotransmitter in the gastrointestinal tract. Biochemical Pharmacology, 28, 565–571. Costa, M., Furness, J.B., Cuello, A.C.U., Verhofstad, A.A., Steinbusch, H.W. and Elde, R.P. (1982). Neurons with 5-hydroxytryptamine-like immunoreactivity in the enteric nervous system: their visualization and reactions to drug treatment. Neuroscience, 7, 351–363. Costa, M., Furness, J.B., Pompolo, S., Brookes, S.J.H., Bornstein, J.C., Bredt, D.S. et al. (1992). Projections and chemical coding of neurons with immunoreactivity for nitric oxide synthase in the guinea-pig small intestine. Neuroscience Letters, 148, 121–125. Coulter, H.D., Gershon, M.D. and Rothman, T.P. (1988). Neural and glial phenotypic expression by neural crest cells in culture: effects of control and presumptive aganglionic bowel from ls/ls mice. Journal of Neurobiology, 19, 507–531. Coventry, S., Yost, C., Palmiter, R.D. and Kapur, R.P. (1994). Migration of ganglion cell precursors in the ileoceca of normal and lethal spotted embryos, a murine model for Hirschsprung disease. Laboratory Investigation, 71, 82–93. Crowcroft, P.J., Holman, M.E. and Szurszewski, J.H. (1971). Excitatory input from the distal colon to the inferior mesenteric ganglion in the guinea-pig. Journal of Physiology, 219, 443–461. Dalsgaard, C.J., Hokfelt, T., Schultzberg, M., Lundberg, J.M., Terenius, L., Dockray, G.J. et al. (1983). Origin of peptide-containing fibers in the inferior mesenteric ganglion of the guinea-pig: immunohistochemical studies with antisera to substance P, enkephalin, vasoactive intestinal polypeptide, cholecystokinin and bombesin. Neuroscience, 9, 191–211. Davis, S., Aldrich, T.H., Stahl, N., Pan, L., Taga, T., Kishimoto, T. et al. (1993). LIFRβ and gp130 as heterodimerizing signal transducers of the tripartite CNTF receptor. Science, 260, 1805–1808. Davis, S., Aldrich, T.H., Valenzuela, D.M., Wong, V.V., Furth, M.E., Squinto, S.P. et al. (1991). The receptor for ciliary neurotrophic factor. Science, 253, 59–63.

DEVELOPMENT OF THE ENTERIC NERVOUS SYSTEM

517

DeChiara, T.M., Vejsada, R., Poueymirou, W.T., Acheson, A., Suri, C., Conover, J.C. et al. (1995). Mice lacking the CNTF receptor, unlike mice lacking CNTF, exhibit profound motor neuron deficits at birth. Cell, 83, 313–322. DiCicco-Bloom, E., Friedman, W.J. and Black, I.B. (1993). NT-3 stimulates sympathetic neuroblast proliferation by promoting precursor survival. Neuron, 11, 1101–1111. Dogiel, A.S. (1899). Über den bau ganglien in den gerflechten des darmes und der gallenblase des menschen und der säugetiere. Archiv für Anatomie und Physiologie Leipzig, Anatomische Abteilung Jahrgang 1899, 130–158. Douville, P.J., Harvey, W.J. and Carbonetto, S. (1988). Isolation and partial characterization of high affinity laminin receptors in neural cells. Journal of Biological Chemistry, 263, 14964–14969. Dreyfus, C.F. and Bornstein, M.B. (1977). Synthesis of serotonin by neurons of the myenteric plexus in situ and in organotypic tissue culture. Brain Research, 128, 125–139. Dreyfus, C.F., Sherman, D.L. and Gershon, M.D. (1977). Uptake of serotonin by intrinsic neurons of the myenteric plexus grown in organotypic tissue culture. Brain Research, 128, 109–123. Duff, R.S., Langtimm, C.J., Richardson, M.K. and Sieber-Blum, M. (1991). In vitro clonal analysis of progenitor cell patterns in dorsal root and sympathetic ganglia of the quail embryo. Developmental Biology, 147, 451–459. Durbec, P.L., Larssonblomberg, L.B., Schuchardt, A., Costantini, F. and Pachnis, V. (1996a). Common origin and developmental dependence on c-ret of subsets of enteric and sympathetic neuroblasts. Development, 122, 349–358. Durbec, P., Marcos-Gutierrez, C.V., Kilkenny, C., Grigoriou, M., Wartiowaara, K., Suvanto, P. et al. (1996b). GDNF signalling through the Ret receptor tyrosine kinase. Nature, 381, 789–793. Edery, P., Attie, T., Amiel, J., Pelet, A., Eng, C., Hofstra, R.M. et al. (1996). Mutation of the endothelin-3 gene in the Waardenburg-Hirschsprung disease (Shah-Waardenburg syndrome). Nature Genetics, 12, 442–444. Edery, P., Lyonnet, S., Mulligan, L.M., Pelet, A., Dow, E., Abel, L. et al. (1994). Mutations of the RET protooncogene in Hirschsprung’s disease. Nature, 367, 378–380. El Shamy, W.M., Linnarsson, S., Lee, K.F., Jaenisch, R. and Ernfors, P. (1996). Prenatal and postnatal requirements of NT-3 for sympathetic neuroblast survival and innervation of specific targets. Development, 122, 491–500. Engvall, E., Davis, G.E., Dickerson, K., Ruoslahti, E., Varon, S. and Manthorpe, M. (1986). Mapping of domains in human laminin using monoclonal antibodies: localization of the neurite-promoting site. Journal of Cell Biology, 103, 2457–2465. Epstein, M.L., Mikawa, T., Brown, A.M. and McFarlin, D.R. (1994). Mapping the origin of the avian enteric nervous system with a retroviral marker. Developmental Dynamics, 201, 236–244. Erickson, C.A., Loring, J.F. and Lester, S.M. (1989). Migratory pathways of HNK-1-immunoreactive neural crest cells in the rat embryo. Developmental Biology, 134, 112–118. Erickson, C.A., Duong, T.D. and Tosney, K.W. (1992). Descriptive and experimental analysis of the dispersion of neural crest cells along the dorsolateral path and their entry into ectoderm in the chick embryo. Developmental Biology, 151, 251–272. Ernfors, P., Lee, K.F., Kucera, J. and Jaenisch, R. (1994). Lack of neurotrophin-3 leads to deficiencies in the peripheral nervous system and loss of limb proprioceptive afferents. Cell, 77, 503–512. Ernsberger, U., Sendtner, M. and Rohrer, H. (1989). Proliferation and differentiation of embryonic chick sympathetic neurons: effects of ciliary neurotrophic factor. Neuron, 2, 1275–1284. Escandon, E., Soppet, D., Rosenthal, A., Mendoza-Ramirez, J.L., Szonyi, E., Burton, L.E. et al. (1994). Regulation of neurotrophin receptor expression during embryonic and postnatal development. Journal of Neuroscience, 14, 2054–2068. Farinas, I., Jones, K.R., Backus, C., Wang, X.Y. and Reichardt, L.F. (1994). Severe sensory and sympathetic deficits in mice lacking neurotrophin-3. Nature, 369, 658–661. Faussone-Pellegrini, M.S. (1985). Cytodifferentiation of the interstitial cells of Cajal related to the myenteric plexus of mouse intestinal muscle coat. An E.M. study from foetal to adult life. Anatomy & Embryology, 171, 163–169. Fontaine-Perus, J., Chanconie, M. and Le Douarin, N.M. (1988). Developmental potentialities in the nonneuronal population of quail sensory ganglia. Developmental Biology, 128, 359–375. Fox, E.A. and Powley, T.L. (1992). Morphology of identified preganglionic neurons in the dorsal motor nucleus of the vagus. Journal of Comparative Neurology, 322, 79–98. Fraser, S.E. and Bronner-Fraser, M. (1991). Migrating neural crest cells in the trunk of the avian embryo are multipotent. Development, 112, 913–920. Fujimoto, T., Hata, J., Yokoyama, S. and Mitomi, T. (1989). A study of the extracellular matrix protein as the migration pathway of neural crest cells in the gut: analysis in human embryos with special reference to the pathogenesis of Hirschsprung’s disease. Journal of Pediatric Surgery, 24, 550–556. Furness, J.B. and Costa, M. (1982). Neurons with 5-hydroxytryptamine-like immunoreactivity in the enteric nervous system: their projections in the guinea-pig small intestine. Neuroscience, 7, 341–349. Furness, J.B. and Costa, M. (1987). The enteric nervous system. Edinburgh: Churchill-Livingstone.

518

INNERVATION OF THE GASTROINTESTINAL TRACT

Furness, J.B., Morris, J.L., Gibbins, I.L. and Costa, M. (1989). Chemical coding of neurons and plurichemical transmission. Annual Review of Pharmacology and Toxicology, 29, 289–306. Gabella, G. (1971). Glial cells in the myenteric plexus. Zeitschrift für Naturforschung, 26B, 244–245. Gabella, G. (1972). Fine structure of the myenteric plexus in the guinea-pig ileum. Journal of Anatomy, 111, 69–97. Gabella, G. and Trigg, P. (1984). Size of neurons and glial cells in the enteric ganglia of mice, guinea-pigs, rabbits and sheep. Journal of Neurocytology, 13, 49–71. Gariepy, C.E., Cass, D.T. and Yanagisawa, M. (1996). Null mutation of endothelin receptor type B gene in spotting lethal rats causes aganglionic megacolon and white coat color. Proceedings of the National Academy of Sciences of the USA, 93, 867–872. Gearing, D.P., Thut, C.J., VandeBos, T., Gimpel, S.D., Delaney, P.B., King, J. et al. (1991). Leukemia inhibitory factor receptor is structurally related to the IL-6 signal transducer, gp130. EMBO Journal, 10, 2839–2848. Gearing, D.P., Comeau, M.R., Friend, D.J., Gimpel, S.D., Thut, C.J., McGourty, J. et al. (1992). The IL-6 signal transducer, gp130: an oncostatin M receptor and affinity converter for the LIF receptor. Science, 255, 1434–1437. Gershon, M.D., Chalazonitis, A., Blaugrund, E., Tennson, V.M. and Rothman, T.P. (1993). Laminin and neurotrophin-3 (NT-3) in the formation of the enteric nervous system. Anatomical Record (Suppl. 1), 54. Gershon, M.D., Chalazonitis, A. and Rothman, T.P. (1993). From neural crest to bowel: development of the enteric nervous system. Journal of Neurobiology, 24, 199–214. Gershon, M.D., Drakontides, A.B. and Ross, L.L. (1965). Serotonin: synthesis and release from the myenteric plexus of the mouse intestine. Science, 149, 197–199. Gershon, M.D., Kirchgessner, A.L. and Wade, P.R. (1994). Functional Anatomy of the Enteric Nervous System. In Physiology of the Gastrointestinal Tract, edited by L.R. Johnson, pp. 381–422. New York: Raven Press. Gershon, M.D. and Ross, L.L. (1966a). Location of sites of 5-hydroxytryptamine storage and metabolism by radioautography. Journal of Physiology (London), 186, 477–492. Gershon, M.D. and Ross, L.L. (1966b). Radioisotopic studies of the binding, exchange, and distribution of 5-hydroxytryptamine synthesised from its radioactive precursor. Journal of Physiology (London), 186, 451–476. Gershon, M.D. and Rothman, T.P. (1991). Enteric glia. Glia, 4, 195–204. Gershon, M.D., Rothman, T.P., Joh, T.H. and Teitelman, G.N. (1984). Transient and differential expression of aspects of the catecholaminergic phenotype during development of the fetal bowel of rats and mice. Journal of Neuroscience, 4, 2269–2280. Gershon, M.D., Rothman, T.P., Sherman, D. and Johnson, E.M. (1983). Effect of prenatal exposure to anti-NGF on the enteric nervous system (ENS) of the guinea pig. Anatomical Record, 205, 62A. Gotz, R., Koster, R., Winkler, C., Raulf, F., Lottspeich, F., Schartl, M. et al. (1994). Neurotrophin-6 is a new member of the nerve growth factor family. Nature, 372, 266–269. Guillemot, F. and Joyner, A.L. (1993). Dynamic expression of the murine Achaete-Scute homologue mash-1 in the developing nervous system. Mechanisms of Development, 42, 171–185. Guillemot, F., Lo, L.C., Johnson, J.E., Auerbach, A., Anderson, D.J. and Joyner, A.L. (1993). Mammalian achaete-scute homolog 1 is required for the early development of olfactory and autonomic neurons. Cell, 75, 463–476. Haralabopoulos, G.C., Grant, D.S., Kleinman, H.K., Lelkes, P.I., Papaioannou, S.P. and Maragoudakis, M.E. (1994). Inhibitors of basement membrane collagen synthesis prevent endothelial cell alignment in matrigel in vitro and angiogenesis in vivo. Laboratory Investigation, 71, 575–582. Henderson, C.E. (1996). Role of neurotrophic factors in neuronal development. Current Opinion in Neurobiology, 6, 64–70. Hofstra, R.M., Osinga, J., Tan-Sindhunata, G., Wu, Y., Kamsteeg, E.J., Stulp, R.P. et al. (1996). A homozygous mutation in the endothelin-3 gene associated with a combined Waardenburg type 2 and Hirschsprung phenotype (Shah-Waardenburg syndrome). Nature Genetics, 12, 445–447. Holschneider, A.M., Meier-Ruge, W. and Ure, B.M. (1994). Hirschsprung’s disease and allied disorders – a review. European Journal of Pediatric Surgery, 4, 260–266. Hosoda, K., Hammer, R.E., Richardson, J.A., Baynash, A.G., Cheung, J.C., Giaid, A. et al. (1994). Targeted and natural (piebald-lethal) mutations of endothelin-b receptor gene produce megacolon associated with spotted coat color in mice. Cell, 79, 1267–1276. Huizinga, J.D., Thuneberg, L., Kluppel, M., Malysz, J., Mikkelsen, H.B. and Bernstein, A. (1995). W/kit gene required for interstitial cells of Cajal and for intestinal pacemaker activity. Nature, 373, 347–349. Ip, N.Y., McClain, J., Barrezueta, N.X., Aldrich, T.H., Pan, L., Li, Y. et al. (1993). The alpha component of the CNTF receptor is required for signaling and defines potential CNTF targets in the adult and during development. Neuron, 10, 89–102. Ip, N.Y., Nye, S.H., Boulton, T.G., Davis, S., Taga, T., Li, Y., et al. (1992). CNTF and LIF act on neuronal cells via shared signaling pathways that involve the IL-6 signal transducing receptor component gp130. Cell, 69, 1121–1132.

DEVELOPMENT OF THE ENTERIC NERVOUS SYSTEM

519

Ito, K., Morita, T. and Sieber-Blum, M. (1993). In vitro clonal analysis of mouse neural crest development. Developmental Biology, 157, 517–525. Jacobs Cohen, R.J., Payette, R.F., Gershon, M.D. and Rothman, T.P. (1987). Inability of neural crest cells to colonize the presumptive aganglionic bowel of ls/ls mutant mice: requirement for a permissive microenvironment. Journal of Comparative Neurology, 255, 425–438. Jessen, K.R. and Mirsky, R. (1980). Glial cells in the enteric nervous system contain glial fibrillary acidic protein. Nature, 286, 736–737. Jing, S., Wen, D., Yu, Y., Holst, P.L., Luo, Y., Fang, M. et al. (1996). GDNF-induced activation of the ret protein tyrosine kinase is mediated by GDNFR-α, a novel receptor for GDNF. Cell, 85, 1113–1124. Johnson, Jr. E.M., Osborne, P.A., Rydel, R.E., Schmidt, R.E. and Pearson, J. (1983). Characterization of the effects of autoimmune nerve growth factor deprivation in the developing guinea-pig. Neuroscience, 8, 631–642. Johnson, J.E., Birren, S.J., Saito, T. and Anderson, D.J. (1992). DNA binding and transcriptional regulatory activity of mammalian achaete-scute homologous (MASH) proteins revealed by interaction with a muscle-specific enhancer. Proceedings of the National Academy of Sciences of the USA, 89, 3596–3600. Jonakait, G.M., Rosenthal, M. and Morrell, J.I. (1989). Regulation of tyrosine hydroxylase mRNA in catecholaminergic cells of embryonic rat: analysis by in situ hybridisation. Journal of Histochemistry & Cytochemistry, 37, 1–5. Jonakait, G.M., Wolf, J., Cochard, P., Goldstein, M. and Black, I.B. (1979). Selective loss of noradrenergic phenotypic characters in neuroblasts of the rat embryo. Proceedings of the National Academy of Sciences of the USA, 76, 4683–4686. Jucker, M., Kleinman, H.K., Walker, L.C., Bialobok, P., Williams, L.R. and Ingram, D.K. (1991). Localization and characterization of a 110 KD laminin binding protein immunoreactivity in adult and lesioned brain [Abstract]. Neuroscience, 17, 207. Kalcheim, C., Carmeli, C. and Rosenthal, A. (1992). Neurotrophin 3 is a mitogen for cultured neural crest cells. Proceedings of the National Academy of Sciences of the USA, 89, 1661–1665. Kapur, R.P., Yost, C. and Palmiter, R.D. (1992). A transgenic model for studying development of the enteric nervous system in normal and aganglionic mice. Development, 116, 167–175. Kapur, R.P., Yost, C. and Palmiter, R.D. (1993). Aggregation chimeras demonstrate that the primary defect responsible for aganglionic megacolon in lethal spotted mice is not neuroblast autonomous. Development, 117, 993–999. Kapur, R.P., Sweetser, D.A., Doggett, B., Siebert, J.R. and Palmiter, R.D. (1995). Intercellular signals downstream of endothelin receptor-B mediate colonization of the large intestine by enteric neuroblasts. Development, 121, 3787–3795. Kapur, R.P., Livingston, R., Doggett, B., Sweetser, D.A., Siebert, J.R. and Palmiter, R.D. (1996). Abnormal microenvironmental signals underlie intestinal aganglionosis in dominant megacolon mutant mice. Developmental Biology, 174, 360–369. Karaki, H., Mitsui-Saito, M., Takimoto, M., Oda, K., Okada, T., Ozaki, T. et al. (1996). Lack of endothelin ETB receptor binding and function in the rat with a mutant ETB receptor gene. Biochemical & Biophysical Research Communications, 222, 139–143. Kennedy, T.E., Serafini, T., de la Torre, J.R. and Tessier-Lavigne, M. (1994). Netrins are diffusible chemotropic factors for commissural axons in the embryonic spinal cord. Cell, 78, 425–435. Keynes, R.J., Johnson, A.R., Picart, C.J., Dunin-Borkowski, O.M. and Cook, G.M.W. (1990). A glycoprotein fraction from adult chicken grey matter causes collapse of CNS and PNS growth cones in vitro [Abstract]. Neuroscience, 16, 169. Kibbey, M.C., Corcoran, M.L., Wahl, L.M. and Kleinman, H.K. (1994). Laminin SIKVAV peptide-induced angiogenesis in vivo is potentiated by neutrophils. Journal of Cellular Physiology, 160, 185–193. Kibbey, M.C., Jucker, M., Weeks, B.S., Neve, R.L., Van Nostrand, W.E. and Kleinman, H.K. (1993). β-Amyloid precursor protein binds to the neurite-promoting IKVAV site of laminin. Proceedings of the National Academy of Sciences of the USA, 90, 10150–10153. Kirchgessner, A.L. and Gershon, M.D. (1989). Identification of vagal efferent fibers and putative target neurons in the enteric nervous system of the rat. Journal of Comparative Neurology, 285, 38–53. Kirchgessner, A.L. and Gershon, M.D. (1990). Innervation of the pancreas by neurons in the gut. Journal of Neuroscience, 10, 1626–1642. Kirchgessner, A.L. and Gershon, M.D. (1995). Presynaptic inhibition by serotonin of nerve-mediated secretion of pancreatic amylase. American Journal of Physiology, 268, G339–G345. Kirchgessner, A.L., Liu, M.T. and Gershon, M.D. (1996). In situ identification and visualization of neurons that mediate enteric and enteropancreatic reflexes. Journal of Comparative Neurology, 371, 270–286. Kirchgessner, A.L., Tamir, H. and Gershon, M.D. (1992). Identification and stimulation by serotonin of intrinsic sensory neurons of the submucosal plexus of the guinea pig gut: activity-induced expression of Fos immunoreactivity. Journal of Neuroscience, 12, 235–248.

520

INNERVATION OF THE GASTROINTESTINAL TRACT

Kishimoto, T., Taga, T. and Akira, S. (1994). Cytokine signal transduction. Cell, 76, 253–262. Klein, R., Silos-Santiago, I., Smeyne, R.J., Lira, S.A., Brambilla, R. Bryant, S. et al. (1994). Disruption of the neurotrophin-3 receptor gene trkC eliminates la muscle afferents and results in abnormal movements. Nature, 368, 249–251. Kleinman, H.K. and Weeks, B.S. (1989). Laminin: structure, functions and receptors. Current Opinion in Cell Biology, 1, 964–967. Kleinman, H.K., Ogle, R.C., Cannon, F.B., Little, C.D., Sweeney, T.M. and Luckenbill-Edds, L. (1988). Laminin receptors for neurite formation. Proceedings of the National Academy of Sciences of the USA, 85, 1282–1286. Kleinman, H.K., Weeks, B.S., Cannon, F.B., Sweeney, T.M., Sephel, G.C., Clement, B. et al. (1991). Identification of a 110-kDa nonintegrin cell surface laminin-binding protein which recognises an A chain neuritepromoting peptide. Archives of Biochemistry and Biophysics, 290, 320–325. Kobayashi, S., Furness, J.B., Smith, T.K. and Pompolo, S. (1989). Histological identification of the interstitial cells of Cajal in the guinea-pig small intestine. Arch ives of Histology and Cytology, 52, 267–286. Kolodkin, A.L. (1995). Semaphorins: mediators of repulsive growth cone guidance. Trends in Cell Biology, 61, 15–22. Konturek, S.J., Bilski, J., Konturek, P.K., Cieszkowski, M. and Pawlik, W. (1993). Role of endogenous nitric oxide in the control of canine pancreatic secretion and blood flow. Gastroenterology, 104, 896–902. Kreulen, D.L. and Szurszewski, J.H. (1979). Reflex pathways in the abdominal prevertebral ganglia: evidence for a colo-colonic inhibitory reflex. Journal of Physiology (London), 295, 21–32. Kuntz, A. (1913). On the innervation of the digestive tube. Journal of Comparative Neurology, 23, 173–192. Kuntz, A. (1953). The autonomic Nervous System. Philadelphia: Lee and Febiger. Kunze, W.A.A., Furness, J.B. and Bornstein, J.C. (1993). Simultaneous intracellular recordings from enteric neurons reveal that myenteric AH neurons transmit via slow excitatory postsynaptic potentials. Neuroscience, 55, 685–694. Kunze, W.A., Bornstein, J.C. and Furness, J.B. (1995). Identification of sensory nerve cells in a peripheral organ (the intestine) of a mammal. Neuroscience, 66, 1–4. Kurihara, Y., Kurihara, H., Suzuki, H., Kodama, T., Maemura, K., Nagai, R. et al. (1994). Elevated blood pressure and craniofacial abnormalities in mice deficient in endothelin-1. Nature, 368, 703–710. Lahav, R., Ziller, C., Dupin, E. and Le Douarin, N.M. (1996). Endothelin 3 promotes neural crest cell proliferation and mediates a vast increase in melanocyte number in culture. Proceedings of the National Academy of Sciences of the USA, 93, 3892–3897. Lallier, T. and Bronner-Fraser, M. (1991). Avian neural crest cell attachment to laminin: involvement of divalent cation dependent and independent integrins. Development, 113, 1069–1084. Lallier, T., Deutzmann, R., Perris, R. and Bronner-Fraser, M. (1994). Neural crest cell interactions with laminin: structural requirements and localization of the binding site for alpha 1 beta 1 integrin. Developmental Biology, 162, 451–464. Lamballe, F., Smeyne, R.J. and Barbacid, M. (1994). Developmental expression of TrkC, the neurotrophin-3 receptor, in the mammalian nervous system. Journal of Neuroscience, 14, 14–28. Lander, A.D., Fujii, D.K. and Reichardt, L.F. (1985). Laminin is associated with the “neurite outgrowth-promoting factors” found in conditioned media. Proceedings of the National Academy of Sciences of the USA, 82, 2183–2187. Lane, P.W. (1966). Association of megacolon with two recessive spotting genes in the mouse. Journal of Heredity, 57, 29–31. Langley, J.N. (1921). The autonomic nervous system. Cambridge: Heffer. Lecoin, L., Gabella, G. and Le Douarin, N. (1996). Origin of the c-kit-positive interstitial cells in the avian bowel. Development, 122, 725–733. Le Douarin, N.M. (1986). Cell line segregation during peripheral nervous system ontogeny. Science, 231, 1515–1522. Le Douarin, N.M. and Dupin, E. (1993). Cell lineage analysis in neural crest ontogeny. Journal of Neurobiology, 24, 146–161. Le Douarin, N.M. and Teillet, M.A. (1973). The migration of neural crest cells to the wall of the digestive tract in avian embryo. Journal of Embryology and Experimental Morphology, 30, 31–48. Le Douarin, N.M. and Teillet, M.A. (1974). Experimental analysis of the migration and differentiation of neuroblasts of the autonomic nervous system and of neurectodermal mesenchymal derivatives, using a biological cell marking technique. Developmental Biology, 41, 162–184. Lee, K.F., Li, E., Huber, L.J., Landis, S.C., Sharpe, A.H., Chao, M.V. et al. (1992). Targeted mutation of the gene encoding the low affinity NGF receptor p75 leads to deficits in the peripheral sensory nervous system. Cell, 69, 737–749. Liesi, P., Narvanen, A., Soos, J., Sariola, H. and Snounou, G. (1989). Identification of a neurite outgrowthpromoting domain of laminin using synthetic peptides. FEBS Letters, 244, 141–148.

DEVELOPMENT OF THE ENTERIC NERVOUS SYSTEM

521

Lin, L.F., Doherty, D.H., Lile, J.D., Bektesh, S. and Collins, F. (1993). GDNF: a glial cell line-derived neurotrophic factor for midbrain dopaminergic neurons. Science, 260, 1130–1132. Lindsay, R.M., Wiegand, S.J., Altar, C.A. and DiStefano, P.S. (1994). Neurotrophic factors: from molecule to man. Trends in the Neurosciences, 17, 182–190. Lo, L. and Anderson, D.J. (1995). Postmigratory neural crest cells expressing c-RET display restricted developmental and proliferative capacities. Neuron, 15, 527–539. Lo, L.C., Johnson, J.E., Wuenschell, C.W., Saito, T. and Anderson, D.J. (1991). Mammalian achaete-scute homolog 1 is transiently expressed by spatially restricted subsets of early neuroepithelial and neural crest cells. Genes and Development, 5, 1524–1537. Luo, Y., Raible, D. and Raper, J.A. (1993). Collapsin: a protein in brain that induces the collapse and paralysis of neuronal growth cones. Cell, 75, 217–227. Mackey, H.M., Payette, R.F. and Gershon, M.D. (1988). Tissue effects on the expression of serotonin, tyrosine hydroxylase and GABA in cultures of neurogenic cells from the neuraxis and branchial arches. Development, 104, 205–217. Maeda, H., Yamagata, A., Nishikawa, S., Yoshinaga, K., Kobayashi, S. and Nishi, K. (1992). Requirement of c-kit for development of intestinal pacemaker system. Development, 116, 369–375. Manthorpe, M., Engvall, E., Ruoslahti, E., Longo, F.M., Davis, G.E. and Varon, S. (1983). Laminin promotes neuritic regeneration from cultured peripheral and central neurons. Journal of Cell Biology, 97, 1882–1890. Martins-Green, M. and Erickson, C.A. (1987). Basal lamina is not a barrier to neural crest cell emigration: documentation by TEM and by immunofluorescent and immunogold labelling. Development, 101, 517–533. Masu, Y., Wolf, E., Holtmann, B., Sendtner, M., Brem, G. and Thoenen, H. (1993). Disruption of the CNTF gene results in motor neuron degeneration. Nature, 365, 27–32. Mawe, G.M. and Gershon, M.D. (1989). Relationship of gallbladder ganglia to the enteric nervous ystemt: strucutre, putative neurotransmitters and direct neural connections. In Proceedings of the 50th Falk Symposium. Titisee: Kluwer Academic Publishers. Messersmith, E.K., Leonardo, E.D., Shatz, C.J., Tessier-Lavigne, M., Goodman, C.S. and Kolodkin, A.L. (1995). Semaphorin III can function as a selective chemorepellent to pattern sensory projections in the spinal cord. Neuron, 14, 949–959. Miura, H., Ohi, R., Tseng, S.W. and Takahashi, T. (1996). The structure of the transitional and aganglionic zones of Auerbach’s plexus in patients with Hirschsprung’s disease: a computer-assisted three-dimensional reconstruction study. Journal of Pediatric Surgery, 31, 420–426. Miyazawa, K., Williams, D.A., Gotoh, A., Nishimaki, J., Broxmeyer, H.E. and Toyama, K. (1995). Membranebound Steel factor induces more persistent tyrosine kinase activation and longer life span of c-kit geneencoded protein than its soluble form. Blood, 85, 641–649. Moore, M.W., Klein, R.D., Farinas, I., Sauer, H., Armanini, M., Phillips, H. et al. (1996). Renal and neuronal abnormalities in mice lacking GDNF. Nature, 382, 76–79. Nomizu, M., Weeks, B.S., Weston, C.A., Kim, W.H., Kleinman, H.K. and Yamada, Y. (1995). Structure-activity study of a laminin alpha 1 chain active peptide segment Ile-Lys-Val-Ala-Val (IKVAV). FEBS Letters, 365, 227–231. Norris, W.E., Stern, C.D. and Keynes, R.J. (1989). Molecular differences between the rostral and caudal halves of the sclerotome in the chick embryo. Development, 105, 541–548. Oakley, R.A. and Tosney, K.W. (1991). Peanut agglutinin and chondroitin-6-sulfate are molecular markers for tissues that act as barriers to axon advance in the avian embryo. Developmental Biology, 147, 187–206. Oakley, R.A., Lasky, C.J., Erickson, C.A. and Tosney, K.W. (1994). Glycoconjugates mark a transient barrier to neural crest migration in the chicken embryo. Development, 120, 103–114. Ockel, M., Lewin, G.R. and Barde, Y.A. (1996). In vivo effects of neurotrophin-3 during sensory neurogenesis. Development, 122, 301–307. Okabe, H., Chijiiwa, Y., Nakamura, K., Yoshinaga, M., Akiho, H., Harada, N. et al. (1995). Two endothelin receptors (ETA and ETB) expressed on circular smooth muscle cells of guinea pig cecum. Gastroenterology, 108, 51–57. Pachnis, V., Mankoo, B. and Costantini, F. (1993). Expression of the c-ret proto-oncogene during mouse embryogenesis. Development, 119, 1005–1017. Parikh, D.H., Leibl, M., Tam, P.K. and Edgar, D. (1995). Abnormal expression and distribution of nidogen in Hirschsprung’s disease. Journal of Pediatric Surgery, 30, 1687–1693. Parikh, D.H., Tam, P.K., Van Velzen, D. and Edgar, D. (1992). Abnormalities in the distribution of laminin and collagen type IV in Hirschsprung’s disease. Gastroenterology, 102, 1236–1241. Pasini, B., Borrello, M.G., Greco, A., Bongarzone, I., Luo, Y., Mondellini, P. et al. (1995). Loss of function effect of RET mutations causing Hirschsprung disease. Nature Genetics, 10, 35–40. Payette, R.F., Bennett, G.S. and Gershon, M.D. (1984). Neurofilament expression in vagal neural crest-derived precursors of enteric neurons. Developmental Biology, 105, 273–287.

522

INNERVATION OF THE GASTROINTESTINAL TRACT

Payette, R.F., Tennyson, V.M., Pham, T.D., Mawe, G.M., Pomeranz, H.D., Rothman, T.P. et al. (1987). Origin and morphology of nerve fibers in the aganglionic colon of the lethal spotted (ls/ls) mutant mouse. Journal of Comparative Neurology, 257, 237–252. Payette, R.F., Tennyson, V.M., Pomeranz, H.D., Pham, T.D., Rothman, T.P. and Gershon, M.D. (1988). Accumulation of components of basal laminae: association with the failure of neural crest cells to colonize the presumptive aganglionic bowel of ls/ls mutant mice. Developmental Biology, 125, 341–360. Pearson, J., Johnson, E.M. and Brandeis, L. (1983). Effects of antibodies to nerve growth factor on intrauterine development of derivatives of cranial neural crest and placode in the guinea pig. Developmental Biology, 96, 32–36. Pennica, D., Shaw, K.J., Swanson, T.A., Moore, M.W., Shelton, D.L., Zioncheck, K.A. et al. (1995). Cardiotrophin-1. Biological activities and binding to the leukemia inhibitory factor receptor/gp130 signaling complex. Journal of Biological Chemistry, 270, 10915–10922. Perris, R., Paulsson, M. and Bronner-Fraser, M. (1989). Molecular mechanisms of avian neural crest cell migration on fibronectin and laminin. Developmental Biology, 136, 222–238. Pettway, Z., Guillory, G. and Bronner-Fraser, M. (1990). Absence of neural crest cells from the region surrounding implanted notochords in situ. Developmental Biology, 142, 335–345. Pham, T.D., Gershon, M.D. and Rothman, T.P. (1991). Time of origin of neurons in the murine enteric nervous system: sequence in relation to phenotype. Journal of Comparative Neurology, 314, 789–798. Pham, T., Wade, A., Chalazonitis, A., Skirboll, S.L., Bothwell, M. and Gershon, M.D. (1996). Increased numbers of myenteric neurons arise in transgenic mice that overexpress neurotrophin-3 (NT-3) directed to the enteric nervous system (ENS) by the dopamin beta-hydroxylase promoter. Neuroscience, 12, 999 [Abstract]. Pichel, J.G., Shen, L., Sheng, H.Z., Granholm, A.C., Drago, J., Grinberg, A. et al. (1996). Defects in enteric innervation and kidney development in mice lacking GDNF. Nature, 382, 73–76. Pinco, O., Carmeli, C., Rosenthal, A. and Kalcheim, C. (1993). Neurotrophin-3 affects proliferation and differentiation of distinct neural crest cells and is present in the early neural tube of avian embryos. Journal of Neurobiology, 24, 1626–1641. Pomeranz, H.D. and Gershon, M.D. (1990). Colonization of the avian hindgut by cells derived from the sacral neural crest. Developmental Biology, 137, 378–394. Pomeranz, H.D., Rothman, T.P. and Gershon, M.D. (1991). Colonization of the post-umbilical bowel by cells derived from the sacral neural crest: direct tracing of cell migration using an intercalating probe and a replication-deficient retrovirus. Development, 111, 647–655. Pomeranz, H.D., Sherman, D.L., Smalheiser, N.R., Tennyson, V.M. and Gershon, M.D. (1991). Expression of a neurally related laminin binding protein by neural crest-derived cells that colonize the gut: relationship to the formation of enteric ganglia. Journal of Comparative Neurology, 313, 625–642. Pomeranz, H.D., Rothman, T.P., Chalazonitis, A., Tennyson, V.M. and Gershon, M.D. (1993). Neural crestderived cells isolated from the gut by immunoselection develop neuronal and glial phenotypes when cultured on laminin. Developmental Biology, 156, 341–361. Powley, T.L., Holst, M.C., Boyd, D.B. and Kelly, J.B. (1994). Three-dimensional reconstructions of autonomic projections to the gastrointestinal tract. Microscopy Research & Technique, 29, 297–309. Puffenberger, E.G., Hosoda, K., Washington, S.S., Nakao, K., deWit, D., Yanagisawa, M. et al. (1994). A missense mutation of the endothelin-B receptor gene in multigenic Hirschsprung’s disease. Cell, 79, 1257–1266. Puliti, A., Poirier, V., Goossens, M. and Simonneau, M. (1996). Neuronal defects in genotyped dominant megacolon (Dom) mouse embryos, a model for Hirschsprung disease. Neuroreport, 7, 489–492. Rickmann, M., Fawcett, J.W. and Keynes, R.J. (1985). The migration of neural crest cells and the growth of motor axons through the rostral half of the chick somite. Journal of Embryology & Experimental Morphology, 90, 437–455. Romeo, G., Ronchetto, P., Luo, Y., Barone, V., Seri, M., Ceccherini, I., et al. (1994). Point mutations affecting the tyrosine kinase domain of the RET proto-oncogene in Hirschsprung’s disease. Nature, 367, 377–378. Rothman, T.P. and Gershon, M.D. (1982). Phenotypic expression in the developing murine enteric nervous system. Journal of Neuroscience, 2, 381–393. Rothman, T.P. and Gershon, M.D. (1984). Regionally defective colonization of the terminal bowel by the precursors of enteric neurons in lethal spotted mutant mice. Neuroscience, 12, 1293–1311. Rothman, T.P., Ross, L.L. and Gershon, M.D. (1976). Separately developing axonal uptake of 5-hydroxytryptamine and noepinephrine in the fetal ileum of the rabbit. Brain Research, 115, 437–456. Rothman, T.P., Specht, L.A., Gershon, M.D., Joh, T.H., Teitelman, G., Pickel, V.M. et al. (1980). Catecholamine biosynthetic enzymes are expressed in replicating cells of the peripheral but not the central nervous system. Proceedings of the National Academy of Sciences of the USA, 77, 6221–6225. Rothman, T.P., Tennyson, V.M. and Gershon, M.D. (1986). Colonization of the bowel by the precursors of enteric glia: studies of normal and congenitally aganglionic mutant mice. Journal of Comparative Neurology, 252, 493–506.

DEVELOPMENT OF THE ENTERIC NERVOUS SYSTEM

523

Rothman, T.P., Sherman, D., Cochard, P. and Gershon, M.D. (1986). Development of the monoaminergic innervation of the avian gut: transient and permanent expression of phenotypic markers. Developmental Biology, 116, 357–380. Rothman, T.P., Le, D.N., Fontaine, P.J. and Gershon, M.D. (1990). Developmental potential of neural crest-derived cells migrating from segments of developing quail bowel back-grafted into younger chick host embryos. Development, 109, 411–423. Rothman, T.P., Goldowitz, D. and Gershon, M.D. (1993). Inhibition of migration of neural crest-derived cells by the abnormal mesenchyme of the presumptive aganglionic bowel of ls/ls mice: analysis with aggregation and interspecies chimeras. Developmental Biology, 159, 559–573. Rothman, T.P., Le, D.N., Fontaine, P.J. and Gershon, M.D. (1993). Colonization of the bowel by neural crest-derived cells re-migrating from foregut backtransplanted to vagal or sacral regions of host embryos. Developmental Dynamics, 196, 217–233. Rothman, T.P., Chen, J. and Gershon, M.D. (1994). Microenvironmental factors in the differentiation of enteric neurons from neural crest-derived precursors. Neuroscience, 20, 1492 [Abstract]. Rothman, T.P., Chen, J., Howard, M.J., Costantini, F., Schuchardt, A., Pachnis, V. and Gershon, M.D. (1996). Increased expression of laminin-1 and collagen (IV) subunits in the aganglionic bowel of ls/ls, but not c-ret-/mice. Developmental Biology, 178, 498–513. Rubanyi, G.M. and Polokoff, M.A. (1994). Endothelins: molecular biology, biochemistry, pharmacology, physiology, and pathophysiology. Pharmacological Reviews, 46, 325–415. Saadat, S., Sendtner, M. and Rohrer, H. (1989). Ciliary neurotrophic factor induces cholinergic differentiation of rat sympathetic neurons in culture. Journal of Cell Biology, 108, 1807–1816. Sakamoto, A., Yanagisawa, M., Sawamura, T., Enoki, T., Ohtani, T., Sakurai, T. et al. (1993). Distinct subdomains of human endothelin receptors determine their selectivity to endothelinA-selective antagonist and endothelinB-selective agonists. Journal of Biological Chemistry, 268, 8547–8553. Sanchez, M.P., Silos-Santiago, I., Frisen, J., He, B., Lira, S.A. and Barbacid, M. (1996). Renal agenesis and the absence of enteric neurons in mice lacking GDNF. Nature, 382, 70–73. Sanders, K.M. (1996). A case for interstitial cells of cajal as pacemakers and mediators of neurotransmission in the gastrointestinal tract. Gastroenterology, 111, 492–515. Sang, Q. and Young, H.M. (1996). Chemical coding of neurons in the myenteric plexus and external muscle of the small and large intestine of the mouse. Cell and Tissue Research, 284, 39–53. Schemann, M. and Grundy, D. (1992). Electrophysiological identification of vagally innervated enteric neurons in guinea pig stomach. American Journal of Physiology, 263, G709–G718. Schemann, M., Schaaf, C. and Mader, M. (1995). Neurochemical coding of enteric neurons in the guinea pig stomach. Journal of Comparative Neurology, 353, 161–178. Schuchardt, A., D’Agati, V., Larsson-Blomberg, L., Costantini, F. and Pachnis, V. (1994). Defects in the kidney and enteric nervous system of mice lacking the tyrosine kinase receptor Ret. Nature, 367, 380–383. Sendtner, M., Carroll, P., Holtmann, B., Hughes, R.A. and Thoenen, H. (1994). Ciliary neurotrophic factor. Journal of Neurobiology, 25, 1436–1453. Sephel, G.C., Tashiro, K., Sasaki, M., Kandel, S., Yamada, Y. and Kleinman, H.K. (1989a). A laminin-pepsin fragment with cell attachment and neurite outgrowth activity at distinct sites. Developmental Biology , 135, 172–181. Sephel, G.C., Tashiro, K.I., Sasaki, M., Greatorex, D., Martin, G.R., Yamada, Y. et al. (1989b). Laminin A chain synthetic peptide which supports neurite outgrowth. Biochemical & Biophysical Research Communications, 162, 821–829. Serafini, T., Kennedy, T.E., Galko, M.J., Mirzayan, C., Jessell, T.M. and Tessier-Lavigne, M. (1994). The netrins define a family of axon outgrowth-promoting proteins homologous to C. elegans UNC-6. Cell, 78, 409–424. Serbedzija, G.N., Burgan, S., Fraser, S.E. and Bronner, F.M. (1991). Vital dye labelling demonstrates a sacral neural crest contribution to the enteric nervous system of chick and mouse embryos. Development, 111, 857–866. Sextier-Sainte-Claire Deville, F., Ziller, C. and Le Douarin, N.M. (1992). Developmental potentialities of cells derived from the truncal neural crest in clonal cultures. Brain Research Developmental Brain Research, 66, 1–10. Sextier-Sainte-Claire Deville, F., Ziller, C. and Le Douarin, N.M. (1994). Developmental potentials of enteric neural crest-derived cells in clonal and mass cultures. Developmental Biology , 163, 141–151. Sieber-Blum, M. and Cohen, A.M. (1980). Clonal analysis of quail neural crest cells: they are pluripotent and differentiate in vitro in the absence of noncrest cells. Developmental Biology , 80, 96–106. Simo, P., Simon-Assmann, P., Bouziges, F., Leberquier, C., Kedinger, M., Ekblom, P., et al. (1991). Changes in the expression of laminin during intestinal development. Development, 112, 477–487. Song, Z.M., Brookes, S.J. and Costa, M. (1991). Identification of myenteric neurons which project to the mucosa of the guinea-pig small intestine. Neuroscience Letters, 129, 294–298.

524

INNERVATION OF THE GASTROINTESTINAL TRACT

Song, Z.M., Brookes, S.J.H. and Costa, M. (1994). All calbindin-immunoreactive myenteric neurons project to the mucosa of the guinea-pig small intestine. Neuroscience Letters, 180, 219–222. Song, Z.M., Brookes, S.J.H. and Costa, M. (1996). Projections of specific morphological types of neurons within the myenteric plexus of the small intestine of the guinea-pig. Cell and Tissue Research, 285, 149–156. Stahl, N., Boulton, T.G., Farruggella, T., Ip, N.Y., Davis, S., Witthuhn, B.A. et al. (1994). Association and activation of Jak-Tyk kinases by CNTF-LIF-OSM-IL-6β receptor components. Science, 263, 92–95. Stark, M.E., Bauer, A.J. and Szurszewski, J.H. (1991). Effect of nitric oxide on circular muscle of the canine small intestine. Journal of Physiology (London), 444, 743–761. Stemple, D.L. and Anderson, D.J. (1992). Isolation of a stem cell for neurons and glia from the mammalian neural crest. Cell, 71, 973–985. Stemple, D.L. and Anderson, D.J. (1993). Lineage diversification of the neural crest: in vitro investigations. Developmental Biology, 159, 12–23. Szurszewski, J.H. (1981). Physiology of mammalian prevertebral ganglia. Annual Review of Physiology, 43, 53–68. Takahashi, R., Yokoji, H., Misawa, H., Hayashi, M., Hu, J. and Deguchi, T. (1994). A null mutation in the human CNTF gene is not causally related to neurological diseases. Nature Genetics, 7, 79–84. Tam, P.K. and Lister, J. (1986). Development profile of neuron-specific enolase in human gut and its implications in Hirschsprung’s disease. Gastroenterology, 90, 1901–1906. Teitelman, G., Joh, T.H. and Reis, D.J. (1978). Transient expression of a noradrenergic phenotype in cells of the rat embryoinic gut. Brain Research, 158, 229–234. Teitelman, G., Gershon, M.D., Rothman, T.P., Joh, T.H. and Reis, D.J. (1981). Proliferation and distribution of cells that transiently express a catecholaminergic phenotype during development in mice and rats. Developmental Biology, 86, 348–355. Tennyson, V.M., Payette, R.F., Rothman, T.P. and Gershon, M.D. (1990). Distribution of hyaluronic acid and chondroitin sulfate proteoglycans in the presumptive aganglionic terminal bowel of ls/ls fetal mice: an ultrastructural analysis. Journal of Comparative Neurology, 291, 345–362. Tennyson, V.M., Chalazonitis, A., Kibbey, M.C. and Gershon, M.D. (1995). Laminin-1 stimulates a cell surface receptor (LBP 110). to promote the differentiation of enteric neurons in vitro independently of its role as an adhesion molecule. Neuroscience, 21, 788 [Absract]. Teramoto, M., Domoto, T., Tanigawa, K., Yasui, Y. and Tamura, K. (1996). Distribution of nitric oxide synthase-containing nerves in the aganglionic intestine of mutant rats: a histochemical study. Journal of Gastroenterology, 31, 214–223. Tessarollo, L., Tsoulfas, P., Martin-Zanca, D., Gilbert, D.J., Jenkins, N.A., Copeland, N.G. et al. (1993). TrkC, a receptor for neurotrophin-3, is widely expressed in the developing nervous system and in non-neuronal tissues. Development, 118, 463–475. Tessier-Lavigne, M. (1994). Axon guidance by diffusible repellants and attractants. Current Opinion in Genetics and Development, 4, 596–601. Tojo, H., Kaisho, Y., Nakata, M., Matsuoka, K., Kitagawa, M., Abe, T. et al. (1995). Targeted disruption of the neurotrophin-3 gene with lacz induces loss of trkC-positive neurons in sensory ganglia but not in spinal cords. Brain Research, 669, 163–175. Torihashi, S., Kobayashi, S., Gerthoffer, W.T. and Sanders, K.M. (1993). Interstitial cells in deep muscular plexus of canine small intestine may be specialized smooth muscle cells. American Journal of Physiology, 265, G638–G645. Torihashi, S., Ward, S.M., Nishikawa, S.I., Nishi, K., Kobayashi, S. and Sanders, K.M. (1995). C-kit-dependent development of interstitial cells and electrical activity in the murine gastrointestinal tract. Cell and Tissue Research, 280, 97–111. Torihashi, S., Ward, S.M. and Sanders, K.M. (1997). Development of C-kit-positive cells and the onset of electrical rhythmicity in murine small intestine. Gastroenterology, 112, 144–155. Toyohara, T., Nada, O., Nagasaki, A., Goto, S. and Ikeda, K. (1985). An immunohistochemical study of serotoninergic nerves in the colon and rectum of children with Hirschsprung’s disease. Acta Neuropathologica, 68, 306–310. Treanor, J.J., Goodman, L., de Sauvage, F., Stone, D.M., Poulsen, K.T., Beck, C.D. et al. (1996). Characterization of a multicomponent receptor for GDNF. Nature, 382, 80–83. Trendelenburg, P. (1917). Physiologische und pharmakologische Versuche Uber die Dunndarmperistaltik. Naunyn-Schmiedeberg Archiv fürExperimentale Pathologies und Pharmakologie, 81, 55–129. Trupp, M., Arenas, E., Fainzilber, M., Nilsson, A.S., Sieber, B.A., Grigoriou, M. et al. (1996). Functional receptor for GDNF encoded by the c-ret proto-oncogene. Nature, 381, 785–789. Trupp, M., Ryden, M., Jornvall, H., Funakoshi, H., Timmusk, T., Arenas, E. et al. (1995). Peripheral expression and biological activities of gdnf, a new neurotrophic factor for avian and mammalian peripheral neurons. Journal of Cell Biology, 130, 137–148.

DEVELOPMENT OF THE ENTERIC NERVOUS SYSTEM

525

Tsoulfas, P., Soppet, D., Escandon, E., Tessarollo, L., Mendoza-Ramirez, J.L., Rosenthal, A. et al. (1993). The rat TrkC locus encodes multiple neurogenic receptors that exhibit differential response to neurotrophin-3 in PC12 cells. Neuron, 10, 975–990. Tsuzuki, T., Takahashi, M., Asai, N., Iwashita, T., Matsuyama, M. and Asai, J.P. (1995). Spatial and temporal expression of the ret proto-oncogene product in embryonic, infant and adult rat tissues. Oncogene, 10, 191–198. Tucker, G.C., Ciment, G. and Thiery, J.P. (1986). Pathways of avian neural crest cell migration in the developing gut. Developmental Biology, 116, 439–450. Vanderwinden, J.M., Rumessen, J.J., Liu, H., Descamps, D., Delaet, M.H. and Vanderhaeghen, J.J. (1996). Interstitial cells of cajal in human colon and in hirschsprungs disease. Gastroenterology, 111, 901–910. Verdi, J.M. and Anderson, D.J. (1994). Neurotrophins regulate sequential changes in neurotrophin receptor expression by sympathetic neuroblasts. Neuron, 13, 1359–1372. Ward, S.M., Burns, A.J., Torihashi, S., Harney, S.C. and Sanders, K.M. (1995). Impaired development of interstitial cells and intestinal electrical rhythmicity in steel mutants. American Journal of Physiology Cell Physiology, 38, C1577–C1585. Ward, S.M., Burns, A.J., Torihashi, S. and Sanders, K.M. (1994). Mutation of the proto-oncogene c-kit blocks development of interstitial cells and electrical rhythmicity in murine intestine. Journal of Physiology, 480, 91–97. Watanabe, Y., Ito, T., Harada, T., Kobayashi, S., Ozaki, T. and Nimura, Y. (1995a). Spatial distribution and pattern of extrinsic nerve strands in the aganglionic segment of congenital aganglionosis – stereoscopic analysis in spotting lethal rats. Journal of Pediatric Surgery, 30, 1471–1476. Watanabe, Y., Ito, T., Harada, T., Takahashi, M., Kobayashi, S., Ozaki, T. et al. (1995b). Expression of ret protooncogene products in the hypoganglionic segment of the small intestine of congenital aganglionosis rats. Journal of Pediatric Surgery, 30, 641–645. Weeks, B.S., Holloway, E., Klotman, P.E., Akiyama, S.K., Schnaper, H.W. and Kleinman, H.K. (1994). 12-O-tetradecanoylphorbol 13-acetate stimulates human T-lymphocyte adherence to the fibronectin RGD domain and the laminin IKVAV domain. Cellular Immunology, 153, 94–104. Wehrle-Haller, B. and Weston, J.A. (1995). Soluble and cell-bound forms of steel factor activity play distinct roles in melanocyte precursor dispersal and survival on the lateral neural crest migration pathway. Development, 121, 731–742. Wood, J.D. (1994). Physiology of the Enteric Nervous System. In Physiology of the Gastrointestinal Tract, edited by L.R. Johnson, D.H. Alpers, E.D. Jacobson and J.H. Walsh, pp. 423–482. New York: Raven Press. Wu, J.J., Rothman, T.P. and Gershon, M.D. (1996). Development of the interstitial cells of cajal in the mouse gut: requirement for Kit ligand (KL) [Abstract]. Neuroscience, 22, 31. Wu, J.J., Chen, J.-X., Rothman, T.P. and Gershon, M.D. (1999). Inhibition of in vitro enteric neuronal development by endothelin-3: mediation by endothelin B receptors. Development 126, 1161–1173. Wyatt, S. and Davies, A.M. (1995). Regulation of nerve growth factor receptor gene expression in sympathetic neurons during development. Journal of Cell Biology, 130, 1435–1446. Xu, D., Emoto, N., Giaid, A., Slaughter, C., Kaw, S., deWit, D., et al. (1994). ECE-1: a membrane-bound metalloprotease that catalyzes the proteolytic activation of big endothelin-1. Cell, 78, 473–485. Yamataka, A., Kato, Y., Tibboel, D., Murata, Y., Sueyoshi, N., Fujimoto, T. et al. (1995). A lack of intestinal pacemaker (c-kit) in aganglionic bowel of patients with hirschsprungs disease. Journal of Pediatric Surgery, 30, 441–444. Yanagisawa, M., Kurihara, H., Kimura, S., Tomobe, Y., Kobayashi, M., Mitsui, Y. et al. (1988). A novel potent peptide produced by vascular endothelial cells. Nature, 332, 411–415. Yanagisawa, M. (1994). The endothelin system: a new target for therapeutic intervention. Circulation, 89, 1320– 1322. Yntema, C.L. and Hammond, W.S. (1954). The origin of intrinsic ganglia of trunk viscera from vagal neural crest in the chick embryo. Journal of Comparative Neurology, 101, 515–542. Yntema, C.L. and Hammond, W.S. (1955). Experiments on the origin and development of the sacral autonomic nerves in the chick enbryo. Journal of Experimental Zoology, 129, 375–414. Yoshinaga, M., Chijiiwa, Y., Misawa, T., Harada, N. and Nawata, H. (1992). Endothelin-B receptor on guinea pig small intestinal smooth muscle cells. American Journal of Physiology, 262, G308–G311. Young, H.M., McConalogue, K., Furness, J.B. and Devente, J. (1993). Nitric oxide targets in the guinea-pig intestine identified by induction of cyclic GMP immunoreactivity. Neuroscience, 55, 583–596. Zhang, J.M., Winslow, J.W. and Sieber-Blum, M. (1993). Role of neurotrophin-3 (NT-3) in the expression of the adrenergic phenotype by neural crest cells [Abstract]. Neuroscience, 19, 251. Zhou, X.F. and Rush, R. (1995). Sympathetic neurons in neonatal rats require endogenous neurotrophin-3 for survival. Journal of Neuroscience, 15, 6521–6530.

Index accommodatory reflex (intestine/colon), 8 acetylcholine (ACh), 1, 215–17, 300–1, pharmacology, 215–17 secretion, 315 acid-evoked gastric hyperaemia mediators, 146–7 model and species differences, 147–8 neural pathways, 144–6 physiological relevance, 148–9 somatic vasoconstriction associated with, 148 adenosine triphosphate (ATP), 232–3 inhibitory neuromuscular transmission, 304–7 secretion, 319–20 sympathetic co-transmitter in colon, 180–4 vasomotor effects, 352 adrenoreceptor subtypes postjunctional, 225–7 prejunctional, 224–5 afferent neurons acid-induced secretion of bicarbonate, 149 “autonomic” impact of, 122–7 control mediators, 127–8 motor effects and mediators, 128–9 pathological inhibition of gastric motor activity, 149–50 physiological implications in the gastric motor control, 129–32 afferent vasodilator nerve fibres in the stomach, 116–21 role in circulation, 116–21 vagal efferent activation, 121 after-hyperpolarizing potential (AHP), 11

aganglionosis of colon, 492–513 AH/type II cells, 10–12, 60 γ-aminobutyric acid (GABA), 22, 240–3 effects on neuromuscular preparations, 241 electrophysiological effects on neurons, 241–2 role of endogenous GABA, 242–3 α-amino-3-hydroxy-5methyl-4-isoxazole propionic acid (AMPA), 244 δ-aminovaleric acid (DAVA), 241 angiotensin, 263–4 anti-secretory transmitters, 320–1 neuropeptide Y and related peptides, 320–1 noradrenaline, 320 arylazidoaminopropionyl (ANAPP3), 305 ascending interneurons, 422–5 regions of gut, 424 species, not guinea-pig, 424–5 Auerbach’s (myenteric) plexus, 395 autonomic nervous system, plasticity in the, 321–3 chronic extrinsic denervation, 322 chronic sympathectomy, 322–3 autoregulatory escape, 120, 352, 353, 354 basal lamina in colon of ls/ls mice, 500–3 biliary tract, 189, 204, 206 blood vessels to gut, 341–58 anatomy, 342–3 autoregulatory escape, 120, 352–4 CGRP and related peptides, 113–14 enteric nervous system, 348–53 innervation, 344–55 nitric oxide, 114–15 pathways and mediators of neurogenic vasodilatation, 113–16

527

528

INDEX

blood vessels to gut (Continued) post-prandial vasodilation, 354 prostaglandins and mast cell mediators, 116 sympathetic control, 344–6 spinal afferent nerves, 346–8 tachykinins, 115–16 vasoactive intestinal polypeptide and PACAP, 116 vasodilation by afferent nerves, 108–13 vasodilator reflex, 354–5 bradykinin, 263 brain-derived neurotrophic factor (BDNF), 485 brain-in-the-gut concept, 365–6 brainstem reflexes, 84–8 convergent vagal inputs, 87–8 modulation of the vagal outflow, 86 spontaneous activity, 85–6 calcitonin gene-related peptide (CGRP), 39, 138–9, 256–8 in extrinsic afferent nerves, 105–8 gastric vasodilation, 113–14 gastric mucosal protection, 138 gastric hyperaemia, 146–7 capsaicin-sensitive afferents, 150–1 central reflex pathways, 83–91 chemical coding, 11, 32 acetylcholine and nitric oxide as motor neurotransmitters, 443–4 enteric nerves in the stomach, 63–9 functional significance of chemical coding gallbladder neurons, 196–9 limitations of chemical coding, 449–52 opioid peptides as enteric transmitters, 445–6 related to motor activity in the isolated stomach, 66–9 substance P as neurotransmitters, 445 vagal fibres, 66 variability hypothesis, 448 VIP as neurotransmitter, 444–5 cholecystokinin (CCK), 27 enteric nervous system, 254–5 gallbladder function, 195–6 sphincter or Oddi neurons, 202–3 choline acetyltransferase (ChAT), 22 ciliary neurotrophic factor (CNTF), 490 ciliary neurotrophic factor receptor (CNTFR), 490

circular muscle (CM) motor neurons, 417–21 regions of gut, 420 species, not guinea-pig, 420–1 command neurons, 88–90 constipation (idiopathic, chronic), 324 co-transmission, 295, 296, 345 c-ret proto-oncogene, 483–4 crest-derived cells in the gut, 475–7 Crohn’s disease, 357, 367 cyclic adenosine monophosphate (cAMP), 343 cytokine, 377–8 development of enteric motor neurons, 490–2 descending interneurons, 425–32 5-HT immunoreactive interneurons, 427–8 classes, 428–30 regions of gut, 430–1 somatostatin-immunoreactive interneurons, 426–7 species not guinea-pig, 431–2 development of enteric nervous system, 469–514 diabetes mellitus, 323–4, 356–7 DiI, 69, 205–6, 401, 414, 417, 419, 425 dimethylphenylpiperazinium (DMPP), 68 4-diphenylacetoxy-methylpiperidine (4-DAMP), 301 dopamine β-hydroxylase (DBH), 65 dorsal motor nucleus of the vagus (DMNV), 85 dynorphin (DYN), 214 electrical field stimulation (EFS), 68 endothelins, 132, 303–4, 494–500, 508 endothelin-B (ETB) receptor, 494–500 endothelin-converting enzyme-1 (ECE-1), 494 endothelin-like peptide, 264 endothelium, 347, 349–50 endothelium-derived hyperpolarizing factor (EDHF), 343 endothelium-derived relaxing factor (EDRF), 343 endotoxicosis, 68 enkephalin (ENK), 32 enteric immunology, 367 enteric mast cells, 367–70 enteric motor neurons, 413–21

INDEX

enteric nervous system (ENS), 58–77, 298 animal models for the human, 449 cellular organisation in mammalian, 393–452 chemical coding, 396–7 classification of enteric neurons, 10–12, 396, 441–3 description, 395–8 development, 469–514 plasticity, 321–7, 450–2 stomach, 59–62 enteric primary afferent neurons, see intrinsic primary afferent neurons enterochromaffin cells, 38, 127, 128, 217, 233, 238, 239, 240, 321, 355, 376, 400 EPSPs, see synaptic ET-3 affects both crest-derived and non-neuronal cells of the colon, 508–9 excitatory amino acids, 243–5 effects on neuromuscular preparations, 244–5 effects on neurons, 245 excitatory junction potential (EJP), 12 excitatory neuromuscular transmission, 300–4 acetylcholine, 300–1 endothelin, 303–4 noradrenaline, 303 substance P, 301–3 excitatory synaptic potentials (EPSPs), 1 extravasation caused by extrinsic afferent neurons, 121–1 extrinsic afferent nerves gastric mechanorecptors, 80–2 inputs to the CNS, 80 intestinal afferents and feedback mechanism, 82–3 mucosal afferents, 82 nerve fibres in the stomach, 105–8 extrinsic innervation, 77–83, 298 denervation, 322 fast excitatory post-synaptic potentials (fEPSP), 60 opioids, 248 Fos, 20, 30, 39, 74, 76, 84, 89, 154, 239, 400, 401, 489, 506

529

GABA, see γ-aminobutyric acid galanin (GAL), 62, 261–2 gallbladder ganglion and neurons chemical coding, 196–9 extrinsic sensory fibres, 194–5 hormonal cholecystokinin, 195–6 interactions with intestine, 204–5 interactions with sphincter of Oddi, 206 morphological properties, 190–1 regulatory inputs, 193–6 electrical properties, 191–3 sympathetic postganglionic fibres, 194 synaptic inputs, 192–3 vagal efferent fibres, 193–4 gastric enteric neurons chemical coding, 63–9 circuits, 69–74 electrophysiological properties, 60–1 neuropharmacology, 60–3 projections and circuits within gastric myenteric plexus, 69–74 synaptic properties, 61–2 gastric mucosal blood flow (GMBF), 108 gastric mucosal hyperaemia, 141 gastric motor reflexes (types), 57–8 permissive control of, 90–1 gastrin releasing peptide (GRP), 22 gastrin, cholecystokinin and caerulein, 253–5 effects of CCK-like peptides on neuromuscular preparations, 253–4 electrophysiological effects of CCK-like peptides on neurons, 254–5 role of endogenous CCK-like peptides, 255 gastrin-releasing peptide and neuromedin B, 260–1 giant migrating contraction (GMC), 4 glial and enteric neuronal progenitors, source of, 471–5 glial cell line-derived neurotrophic factor (GDNF), 482–4 glial fibrillary acidic protein (GFAP), 488 glutamic acid decarboxylase (GAD), 240 glyceryl trinitrate (GTN), 311 hexahydrosiladifenidol (HHSiD), 215 Hirschsprung’s disease, 295, 324–6, 357, 394, 493, 494, 500, 501, 503, 507, 509, 510, 511, 512, 513, 514

530

INDEX

Hirschsprung’s disease (Continued) abnormal intestitial cells of Cajal in aganglionic bowel, 509–13 histamine, 371 mimics slow synaptic excitation, 371–2 pattern generation evoked, 373–4 presynaptic inhibition, 375–6 receptors for slow excitatory responses, 373 signal transduction for slow EPSPs, 372 histidine decarboxylase (HDC), 107 5-hydroxytryptamine, see serotonin hyoscine, 31, 34, 215, 219, 228, 241, 247, 254, 258, 262–3 hyperaemia (gastric mucosal), 140–9 mediators, 146–7 mechanisms, 147–9 hyperaemia-independent mechanisms, 141–2 hypoganglionosis, 511, 514 ileus, 58, 68, 88, 103, 150, 363 immuno-physiology of gut, 363–89 effector behaviour, 386–7 neural behaviour, 385–6 sensitised intestine, 385–7 inflammatory bowel disease, 68, 357 inhibitory neuromuscular transmission, 304–14 ATP, nitric oxide and vasoactive intestinal peptide, 304–7 inhibitory junction potential (IJP), 12 nitric oxide and intestinal cells of Cajal, 311–12 nitric oxide, 307–11 noradrenaline, 313–14 pituitary adenylate cyclase-activating peptide, 313–12 inhibitory synaptic potentials (IPSPs), see synaptic innervation of intestinal blood vessels, 344–53 cholinergic transmission, 349–51 extrinsic sensory nerves, 346–8 extrinsic vasoconstrictor nerves, 344–6 intrinsic nerves, 348–9 non-cholinergic transmitters, 351–3 interleukin-1β (IL-1β), 150, 377–8 interleukin-6 (IL-6), 371, 377–8, 490 interleukin-II (IL-II), 490 interneurons, identities of, 22, 421–36 ascending (orally-projecting), 422–5 descending (aborally-projecting), 425–32

interstitial cells of Cajal (ICCs), 218, 509–11 nitric oxide, 311–12 intestinal transmission neuroepithelial, 295 neuromuscular, 295 intestinofugal neurons, see viscerofugal neurons intraganglionic laminar endings (IGLE), 78, 402 intrinsic primary afferent neurons (IPANs), 16–21, 399–413 chemical coding, 405 electrical properties, 402–3 identities of, 1, 16–21, 400–2 morphology, 404–5 motor activity, 40–4 muscle mechanoreceptors, 18–20, 401–2 mucosal mechanoreceptors, 20–1 neurotransmission between, 36–8, 382–4, 406–9 species differences, 409–12 stomach, 75–7 synaptic inputs, 406–9 ischaemia, 82, 151, 354, 355–7 Kit ligand, 488, 509 β-lactoglobulin, 385–7 laminin-1 promotes the development of enteric nerves, 503–8 excess in the colon, 507–8 IKVAV peptide effects, 504–6 laminin-binding protein, 110 (LBP-110), 504–10 primary crest cells and crest-derived cells, 506–7 Law of the Intestine, 3, 470 leucocyte recruitment afferent nerves, 122–3 mast cells, 370 leukaemia inhibitory factor (LIF), 490–2 leukotrienes B4, E4, D4 and C4 (LTB4, LTE4, LTD4 and LTC4), 378–9 Lissauer’s tract, 83 longitudinal muscle (LM) motor neurons, 1, 414–17 innervation in other regions, 415–16 other regions and other species, 416–17 longitudinal muscle/myenteric plexus (LM-MP) preparation, 213

INDEX

mash-1, 479–81, 487 mast cells (enteric), 367–79 adenosine, 377 cytokines, 377–8 histamine, 371–6 serotonin, 376 Meissner’s (submucous) plexus, 396, 440 Met-enkephalin, 174, 248, 249, 253 N-methyl-D-aspartate (NMDA), 244 α,β-methylene-ATP (α,β-MeATP), 182 β,γ-methylene-ATP (β,γ-MeATP), 182 2-methylthio-ATP (2-MeSATP), 182 migrating myoelectric complex (MMC), 4 monoaxonal neurons, 10, 11 Nw-monomethyl-L-arginine (L-NMMA), 174 monosynaptic reflex, 3 motilin, 259–60 motility, see motor patterns motor neurons, identities of, 22–3, 413–21 longitudinal muscle, 414–17 circular muscle, 417–21 motor patterns, programs and enteric reflexes, 2–3, 40–2, 363–5 ascending excitatory reflex, 1, 12–14, 32–4, 40, 61, 76, 172, 216, 239, 247, 256, 299, 423 descending inhibitory reflex, 12–14, 35, 75, 76, 216, 227, 232, 253, 255, 299, 313, 429 gastric motor reflexes (types), 57–8, 74–5, 90–1 identities of neurons in the reflex pathways, 16–23 in vivo, 3–5 intestinal reflexes, 1, 3, 6, 7, 16 isolated systems, 5–9 peptidergic afferent nerves, 128–32 peristalsis, 3, 6, 7, 40–4, 58, 68, 74, 91, 172, 204, 215–21, 233, 236, 242, 244, 248, 253, 255, 257, 302, 323–4, 399, 401, 415, 423, 445, 470 permissive control of, 90–1 physiological studies, 3–9 mucosa histamine, 379–80 mediators of protection, 138–40 peptidergic afferent nerves protect, 136–49 restitution, healing and afferent nerves, 143–4 muscarinic receptors, 215–16

531

myectomies, 397 myogenic tone, 57 myotomies, 397, 424 NANC transmission, 171, 213 natural IL-1 receptor antagonist (nIL-1ra), 377 nerve growth factor (NGF), 322 neural basis of reflex control, 58–9 neural circuits mediating reflexes evoked by stationary stimuli in vitro, 9–16 interactions evoked by different stimuli, 14–15 local distensions, 12–13 mechanical or chemical stimulation of the mucosa, 13–14 responses recorded from neurons, 15–16 neural crest, 471–5 source of enteric neuronal progenitors, 471–5 pluripotent cells, 475 neural interactions between gallbladder and the sphincter of Oddi, 206 gut and the gall bladder, 204–7 gut and the sphincter of Oddi, 205–6 neural networks and effector systems behaviour, 379–85 coordinated recruitment, 382–3 gating functions, 383–4 neural pattern generation, 380–1 presynaptic inhibition, 384–5 responses to histamine and leukotrienes, 379–80 slow synaptic excitation, 381–2 neurodegenerative disease, 298 neuroeffector dysfunction in pathological conditions, 323–7 diabetes mellitus, 323–4 Hirschsprung’s disease, 324–6 idiopathic chronic constipation, 324 ulcerative colitis, 326–7 neuroeffector transmission in the intestine, 295–327 autonomy of the enteric nervous system, 297–8 co-transmission and neuromodulation, 296–7 extrinsic innervation, 298 neuroepithelial transmission in the intestine, 314–21

532

INDEX

neurofilament protein (NFP), 398 neurokinin A (NKA), 103 neurokinin B (NKB), 195 neurokinin-3 (NK3), 195 neuromedin U (NMU), 262 neuromuscular transmission, 299–314 ATP, nitric oxide and vasoactive intestinal peptide, 304–7 autonomic neuromuscular junction, 299–300 acetylcholine, 300–1 endothelin, 303–4 intestinal smooth muscle contractility, 299 noradrenaline, 303 nitric oxide, 307–11 nitric oxide and intestinal cells of Cajal, 311–12 noradrenaline, 313–14 pituitary adenylate cyclase-activating peptide, 313–12 substance P, 301–3 neuronal intestinal dysplasia, 511, 514 neuropeptide Y (NPY) and peptide YY (PYY), 62, 255–6 effects on neuromuscular preparations, 256 electrophysiological effects of NPY and related peptides, 256 secretion, 320–1 vasoconstriction potentiation, 352 vasodilation, 352 neurotensin, 258–9 neurotransmitters, 213–65, 296–7, plasticity of expression of, 321–7 neurotrophins BDNF, 485 NGF, 484–6 NT-3 and development of the enteric nervous system, 484–90 NT4/5, 485 NT6, 485 nicotinamide adenine dinucleotide phosphate (NADPH)-diaphorase, 197, 343 nicotinic receptors, 215–17 nitric oxide (NO), 36, 229–32, 379 effects on neuromuscular preparations, 230–1 effects on neurons and enteric smooth muscle cells, 231 inhibitory neuromuscular transmission, 307–12 role of endogenous NO, 231–2

nitric oxide synthase (NOS), 22 interneurons, 428–31 non-adrenergic non-cholinergic (NANC) nerves, 67, 227, 295–6 noradrenaline (NA), 61, 221–7, 303 inhibitory neuromuscular transmission, 313–4 secretion, 320 nuclear factor kappa-B (NF-κB), 485 nucleus tractus solitarius (NTS), 85 oncostatin M (OSM), 490 opioid peptides, 246–51 effects on neuromuscular preparations, 246–7 electrophysiological effects on neurons, 247–8 endogenous opioids, 248–51 neurotransmitter role, 445–6 opioid receptor like (ORL), 246 p-75 neurotrophin receptor, 485–6, 489 pancreatic polypeptide (PP), 62 parasympathetic nervous system, 171, 344 brainstem reflexes to stomach, 84–8 convergent vagal inputs in brainstem, 87–8 control of the colon, 173–8 control of the gall bladder, 193 modulation of vagal outflow to stomach, 86 nitric oxide, 174–6 prostaglandins, 176–8 spontaneous activity and gastric function, 85–6 peptide histidine isoleucine (PHI), 227 peptidergic afferent nerves, see spinal afferent peripheral nervous system (PNS), 245 peristalsis, 3, 6, 7, 40–4 opioids, 248–50 pharmacology adrenergic transmission, 221–7 amino acids as enteric neurotransmitters, 240–5 cholinergic transmission, 215–17 enteric nervous system, 213–65 histaminergic transmission, 245–6 NANC inhibitory transmission, 227–33 non-cholinergic excitatory transmission, 217–21

INDEX

peptides as enteric neurotransmitters, 246–65 serotonergic transmission, 233–40 phenylethanolamine-N-methyltransferase (PNMT), 65 piebald lethal mouse, 493–4 pilocarpine, 6 pituitary adenylate cyclase-activating peptide (PACAP), 35, 422 inhibitory neuromuscular transmission, 312–13 plasma protein extravasation, 121–2 leucocyte recruitment, 122–3 plasticity of the enteric nervous system, 295, 321–7, 450–2 associated with disease states, 451–2 developmental plasticity, 450 response to denervation, 450–1 platelet activating factor (PAF), 119, 378–9 pluripotent crest-derived neural precursors, 483–4 progenitor lineages in formation of ENS, 477–81 prostaglandins gastric vasodilation, 116 gastric mucosal protection, 140 prostaglandin D2 (PGD2), 378 prostaglandin E2 (PGE2), 116 purinergic transmission and receptors colon, 180–4 inhibitory neuromuscular transmission, 304–7 secretion, 319–20 ret proto-oncogene (c-ret), 483–4 rough endoplasmic reticulum (RER), 494 S/type I neurons, 10–12, 60 Schabadasch’s or Henle’s plexus, 396, 421, 440 Schwann cells, 469, 476, 485, 512 secretion acetylcholine, 315 ATP, 319–20 gastric and peptidergic afferent nerves, 123–7 gastric mediators, 127–8 histamine, 379–81 leukotrienes, 380 neurotransmitters, 315–20

533

substance P, 317–19 vasoactive intestinal peptide and related peptides, 315–17 secretomotor neurons, 436–8 NPY-immunoreactive neurons, 436–7 VIP-immunoreactive neurons, 436–7 sensory nerves, see IPANs, spinal nerves, vagus nerve sensory transduction, 38–40 chemoreceptors, 39–40 mucosal mechanoreceptors, 38–9 stretch, 38 serotonin (5-HT), 18, 234–8, 376 effects on neuromuscular preparations, 233–8 electrophysiological effects on neurons, 238 interneurons, 427–8 role of endogenous, 5-HT, 238–40 serotonergic transmission, 213, 233 Shah-Waardenburg syndrome, 512 signal, mast cells to the brain-in-the-gut, 371–9 simple reflexes, motor patterns and complex behavior, 40–4 muscle tension changes, effects when, 41–4 sodium nitroprusside (SNP), 230 somatostatin, 22, 32, 251–3 effects on neuromuscular preparations, 251–2 electrophysiological effects on neurons, 252 interneurons, 426–7 role of endogenous somatostatin, 252–3 sphincter of Oddi (SO) ganglia and neurons, 189 actions of CCK, 202–3 chemical coding, 203–4 morphological properties, 199–200 regulatory inputs, 201–3 structural and electrical properties, 199–201 sympathetic inputs, 201–2 synaptic inputs, 201 spinal afferent nerves blood flow, 108–21 mechanisms of gastric mucosal protection, 141–2 mediators of gastric mucosal protection, 138–40 gallbladder, 194–5 gastric motor activity, 128–32 gastric mucosal blood flow, 108–21, 144–9

534

INDEX

spinal afferent nerves (Continued) gastric secretion, 123–8 pathways in gastric mucosal protection, 138 resistance of gastric mucosal injury, 132–5 restitution and healing of gastric lesions, 143–4 stomach, 77, 105–8, 132–50 vasodilation, 346–8 vascular permeability and leucocyte recruitment, 121–3 spinal reflexes, 83–4 splanchnic nerves, 57, 84, 121, 145, 194, 201 subdiaphragmatic vagotomy, 104 sublineages of enteric neuronal precursors, 481–2 submucous plexus neurons, 436–41 colon, 438–9 inter-plexus interneurons, 438 secretomotor neurons, 436–7 species, not guinea-pig, 439–41 vasomotor neurons, 438 submucous interneurons and primary afferent neurons, 438 substance P (SP, see tachykinins) suramin, 24, 32, 34, 183, 233, 307, 313 sympathetic nervous system colon, 178–84 gall bladder, 194 noradrenaline and ATP as co-transmitters, 180–2 purine receptors in colon, 182–4 sphincter of Oddi, 201–2 sympathectomy effects, 322–3 vasoconstriction, 344–6 synaptic interactions ascending interneurons to motor neurons, 33–4 ascending interneurons, 32–3 descending interneurons and motor neurons, 35–6 descending interneurons, 34–5 IPANs and motor neurons, 28–30 IPANs to interneurons, 30–2 IPANs, 36–8, 383–4 presynaptic inhibition, 384–5 retrograde transmission from interneurons to IPANs, 36

synaptic neurotransmission, 1, 17, 20, 30, 31, 36, 60, 61, 62, 79, 129, 232, 246, 371, 385, 433 synaptic potentials, types of, 23–8 fast excitatory post-synaptic potentials (fEPSP), 60 inhibitory synaptic potentials, 201 slow, 381–2, 407–9 tachykinins, 1, 20, 139–40 electrophysiological effects, 219 gastric hyperaemia, 147 gastric mucosal protection, 139–40 gastric vasodilation, 115–16 neuromuscular preparations, 218, 301–3 neurokinin A (NKA), 103 neurokinin B (NKB), 195 receptors, 217–21 role of endogenous tachykinins, 219–21, 445 secretion, 317–19 tetraethylammonium (TEA), 25 tetrodotoxin (TTX), 73 thyrotropin-releasing hormone (TRH), 66 transforming growth factor-β (TGF-β), 483 Trichinella spiralis, 385–7 tumour necrosis factor α (TNFα), 371, 377–8 tyrosine hydroxylase (TH), 194 tyrosine kinase receptors, 485–90 TrkA, 485–6 TrkB, 486 TrkC, 486, 489–90 ulcerative colitis, 295, 326–7, 357, 367 vagus nerve afferent nerves, 77–80, 80–2 brainstem reflexes, 84–8 chemical coding, 66 convergent vagal inputs, 87–8 intraganglionic laminar endings (IGLE), 78, 402 mechanoreceptors, 80–2 modulation of vagal outflow, 86 mucosal receptors, 82 spontaneous activity, 85–6 vago-enteric interactions in stomach, 62–3 vascular supply, see blood vessels

INDEX

vascular tone of intestine, factors that influence, 343–4 vasoactive intestinal constrictor (VIC), 303 vasodilatation by afferent nerve stimulation, 108–13 vasomotor neurons, 438 calretinin-immunoreactive neurons, 438 VIP-like peptides, 22, 227–9 effects on neuromuscular preparations, 228 effects on neurons and enteric smooth muscle cells, 228–9

gastric mucosal protection by afferent nerves, 116 interneurons, 428–32 role of endogenous VIP-like peptides, 229 vasodilation, 351–2 viscerofugal neurons, 432–6

yohimbine, 182, 194, 222, 223, 225, 263, 320

535