- Email: [email protected]
function relationship for vagal and splanchnic afferent endings supplying the gastrointestinal tract
Journal of the Autonomic Nervous System, 22 (1988) 175-180 Elsevier
175
JAN 00812
Review Article
Speculations on the structure/function relationship for vagal and splanchnic afferent endings supplying the gastrointestinal tract David Grundy Department of Physiology, The University, Western Bank, Sheffield (U.K.)
Key words: Vagal afferent; S p l a n c h n i c afferent; G a s t r o i n t e s t i n a l m e c h a n o r e c e p t o r ; Gastrointestinal chemoreceptor
Abstract This paper discusses some of the unsettled issues in the study of the afferent innervation of the gastrointestinal (GI) tract. Afferent fibres in the vagus and splanchnic nerves have been studied electrophysiologicallyand much has been learnt from single fibre recordings. Splanchnic afferent fibres generally terminate in multiple mechanosensitive endings in the mesentery and serosa where they are in a position to monitor tension on the mesenteric attachments. Other mechanoreceptors following a mainly vagal pathway behave as if they are functionally in-series with the muscle elements of the gut wall and signal muscle tension generated passively by distension and actively during contraction. A third group of afferent endings supply the GI mucosa where they are in a position to signal information on the physical and chemical environment of the gut lumen. A complex picture of mucosal sensitivity has emerged with subpopulations of receptors with polymodal sensitivity and quality-specific mechanoreceptors, thermoreceptors and chemoreceptors. Unfortunately, there is little concensus amongst the different research groups because of different experimental paradigms. One group describes specific chemoreceptors, other groups fail to find them. In this minireview I have speculated on the cause of the often conflicting data on GI afferents and the implications this has for the interpretation of visceral receptor mechanisms.
Introduction The m o t o r a n d secretory activities of the gastro-intestinal tract are regulated to optimize digestion a n d a b s o r p t i o n of nutrients. T h e c o n t r o l m e c h a n i s m s which regulate these activities are organized o n a hierarchical basis with local a n d extrinsic reflexes delivering a variety of n e u r o t r a n s m i t t e r substances to the secretory a n d m o t o r effectors. I n addition, e n d o c r i n e a n d p a r a c r i n e cells within the gastrointestinal m u c o s a ' t a s t e ' the l u m i n a l c o n t e n t s a n d release their c o n t e n t s u n d e r appropriate conditions. The gastrointestinal tract
Correspondence: Department of Physiology, The University, Western Bank, Sheffield, S10 2TN, U.K.
is therefore c o n s t a n t l y exposed to a c h a n g i n g chemical e n v i r o n m e n t o n its l u m i n a l surface, from w i t h i n its i n t e r n a l s t r u c t u r e a n d f r o m the b l o o d s t r e a m . A f f e r e n t e n d i n g s within the wall of the gastrointestinal tract a n d its mesenteric connections m o n i t o r this c h a n g i n g e n v i r o n m e n t a n d f u r t h e r m o r e are exposed to a n y effector responses (motor, secretory or e n d o c r i n e ) m e d i a t e d b y their activation.
Magnitude and morphology of the gastrointestinal afferent innervation G a s t r o i n t e s t i n a l afferent e n d i n g s send axons via the a u t o n o m i c nerves to the b r a i n s t e m a n d spinal cord. I n the vagus nerves of n o n - r u m i n a n t s
0165-1838/88/$03.50 © 1988 Elsevier Science Publishers B.V. (Biomedical Division)
176
afferent fibres projecting to the nucleus tractus solitarius (nTS) from the abdomen outnumber by 10 to one the preganglionic parasympathetic efferent fibres making the reverse connection between brainstem and gut: a total of some 20,000 to 40,000 unmyelinated afferent fibres [1]. While there are proportionally more preganglionic sympathetic efferent fibres than afferent fibres in the splanchnic nerves, the total number of afferents entering the spinal cord is of a similar order of magnitude to that of the vagal supply [21]. Morphological data on the receptor endings are sparse. They are, with the exception of the Pacinian corpuscle, considered to be bare nerve endings [15]. The title of this paper may therefore seem misleading since very little is known about the structure of gastrointestinal afferent endings. Because they are inaccessible to direct electrophysiological investigation there is even less information on impulse generation within these endings. However, the way an afferent fibre responds to applied stimuli allows one to speculate on the structure of afferent endings and the transduction processes which lead to afferent impulse traffic. It is this functional approach that I wish to develop here. Vagal afferent terminals have been visualized using autoradiographic techniques following the injection of tritiated leucine into the nodose ganglia and are seen terminating in the muscle and close to the mucosal epithelium [33]. It is their position in these sites which determines their functional specificity as muscle tension receptors and monitors of luminal contents, respectively. Afferent terminations have also been described in subserosal, myenteric and submucous plexuses where it was suggested that collaterals of vagal afferents participate in axon reflexes [33]. This suggestion arises because neuropeptides including substance P, somatostatin and CCK are transported against the direction of afferent impulses from the cell body and can be followed immunohistochemically on route to the periphery where they can be released by neural impulses [12]. Thus afferent fibres may not only serve a sensory function but could also regulate transmission through the intramural ganglia. Other sensory structures visualized in the oesophageal intramural ganglia have been called intraganglionic laminar endings [32].
These are terminations of vagal afferent fibres which appear as tape-like structures on the inner surface of the ganglia. While it is pure speculation that these are mechanoreceptors, receptors in such a location would be in a position to detect mechanical deformation of the oesophageal wall during the presence or passage of a bolus. Again, however, these may be terminations of afferent collaterals involved in axon reflexes. Both muscle and mucosal endings project predominantly via vagal afferents to the brainstem. A third group of afferents which run mainly in the splanchnic nerves are associated with the visceral peritoneum and terminate either in the serosa, mesentery or omentum as indicated by the distribution of their receptive fields. These splanchnic afferents terminate in multiple mechanosensitive points which monitor tension on the mesenteric attachments and are proposed to mediate visceral reflexes and also signal pain of visceral origin [29].
Visceral sensitivity The range of visceral sensitivity can be gauged from the type of stimuli which give rise to reflex evoked responses. In this respect responses to both mechanical (distension) and chemical stimuli (luminal acidity, osmolarity, etc.) have been described, for example during feedback regulation of gastric emptying [23]. However, only when electrophysiological techniques are used to record directly the nerve impulses in individual sensory nerves can the properties of gastrointestinal afferents be fully and accurately quantified in terms of adequate stimuli, thresholds to stimulation and receptive fields. Such studies have confirmed the existence of both gastrointestinal mechanoreceptors and chemoreceptors although the study of gut receptor mechanisms is still a long way behind similar studies of somatic sensory endings. Part of the problem relates to the technical difficulties of recording from this largely unmyelinated population of nerves whose receptive fields are relatively inaccessible. It is also difficult to deliver quantitatively precise stimuli to a receptor ending when local reflex responses and humoral mechanisms may perturb or potentiate the stimulus.
177
Serosal receptors Splanchnic afferent fibres have between 1 and 8 punctate mechanoreceptive sites ( < 10 mm 2) distributed along mesenteric arteries, at vascular divisions of the arteries and on the wall of the viscus where an artery pierces the serosal surface [29]. The afferent fibres respond to distortion of the receptor ending and as such can respond to distension and contraction of the bowel. Visceral pain in humans and pseudoaffective responses to noxious levels of stimulation in animals is mediated through splanchnic pathways and splanchnic mechanoreceptors are probably responsible [21]. The existence of specific nociceptors within the gastrointestinal tract is a matter for debate [5] with the alternative argument being that both noxious and innocuous events are encoded in the intensity of discharge from the same population of neurones [21]. Serosal receptors may thus provide physiological regulation at low levels of stimulation but signal a noxious event when the intensity of discharge is excessive as might occur during obstruction or spasm. Tension receptors The muscle endings with vagal afferent fibres detect tension within the muscle generated passively by distension or actively during contraction. These are described variably as 'in-series' tension receptors [from 20] or stretch receptors (with an in-parallel location) [see 31] with the latter implying that they respond only to distension. This terminology is ambiguous because only a single functional type of muscle receptor has been characterized electrophysiologically but whose properties differ according to the gastrointestinal region in which they occur. Thus in a readily distensible region like the gastric corpus receptors are primarily sensitive to distension while in the muscular antrum, contraction is the main stimulus [2]. Nevertheless, antral distension and corpus contraction are effective stimuli. Recent work in our laboratory has concentrated on the behaviour of these vagal mechanoreceptors in the gastric corpus during active relaxation and highlights the importance of muscle tension as the primary determinant of receptor activity [3]. Under isometric conditions vagal afferent discharge mirrored pressure
changes during both contraction and relaxation. Under isotonic conditions both pressure changes and afferent discharge were attenuated despite large changes in muscle length. Thus, during a meal receptive relaxation of the stomach would limit wall tension and the stomach would fill with minimal activation of vagal afferent endings. If these receptors contribute to the sensation of satiety as has been suggested many times, then clearly this can not develop until relaxation wanes and corpus tone increases. The classification of in-series and in-parallel receptors in skeletal muscle has an obvious structural basis in the tendon organ and muscle spindle. The distinction is less clear-cut in smooth muscle which is composed of bundles of densely packed cells which are mechanically and electrically coupled [15]. Within the bundles the individual smooth muscle cells are separated from their neighbours by only tens of nanometres. In such a situation it is difficult to envisage an afferent ending in-series with an individual smooth muscle cell. However, the functional unit is not the cell but the bundle of cells that are interconnected by collagen fibrils forming laminar intramuscular septa. These septa are suggested to serve as intramuscular tendons transmitting tension longitudinally from one bundle of cells to the next. The in-series tension receptor in skeletal tendons may therefore have its smooth muscle counterpart in the vagal mechanoreceptors if these were located within the connective tissue matrix. In such a location they would respond to muscle tension generated passively by distension and actively during contraction. However, because the intramuscular septa run parallel to the muscle cells these receptors may function as in-series receptors from an inparallel location. Mucosal receptors Research on mucosal receptors has proved less conclusive than that on muscle and serosal mechanoreceptors because of their heterogeneity. Much recent work in this field has concentrated on the description of quality-specific chemoreceptors detecting various chemical components of chyme in the stomach and small intestine of the cat. Specific glucoreceptors [13,24], amino-acid receptors [22],
178
receptors to long and short chain lipids [28], and acid receptors [13] have all been described by the Marseilles group headed by Mei; as have specific thermoreceptors [14]. The existence of specific chemoreceptors has been readily accepted in the literature because they correspond to the stimuli which have been shown, for example, to inhibit gastric emptying or mediate satiety [23]. However, the evidence for such receptors is not unequivocal. Firstly, all of these stimuli evoke local responses and the afferent discharge may be secondary to these. Secondly there are anomalies in the literature. While responses to luminal chemicals have been described by many workers using whole nerve or multiunit recording techniques [see 10 and 34 for reviews of multiunit literature], the value of such studies is limited because of the inability to distinguish between a population of afferent fibres capable of responding to many different stimuli and subpopulations which are specific to different types of stimuli. It is also impossible to pin-point receptive fields in such preparations. These problems are overcome when recordings are made from single afferent fibres. All descriptions of single mucosal afferent fibres from outside Marseilles, using a wide variety of animal models including the cat, describe them as multi- or polymodal. In addition to responding to a wide variety of unrelated chemicals, these receptors also respond to mechanical stimulation in the form of mucosal deformation produced by lightly stroking the mucosa overlying the circular receptive field [6,8,9,17,19]. The major difference between these studies and those in which specific receptors were identified relates to the preparation of the viscera for the application of stimuli. Polymodal receptors were identified in preparation in which the viscera was opened and the lumen exposed for probing and direct application of chemicals while specific receptors were found in closed perfused preparations of loops of bowel (usually about 40 cm long). By surgically exposing the luminal surface, the mucous barrier overlying the mucosal epithelium may dry out or it may become eroded. In such preparations the mucosal epithelium may be exposed to stimuli it would not normally encounter and receptor endings may respond in a nonspecific manner, i.e. polymodal sensitivity. How-
ever, the exquisite sensitivity of polymodal receptors to mechanical stimuli, responding for example to a thread of cotton trailed over the receptive field, would argue in favour of this being a primary sensitivity of the endings. Indeed some mucosal receptors respond only to mechanical stimulation [7,8]. These endings may therefore function as contact receptors detecting the particulate nature of chyme [10]. On the other hand, in closed loops it is impossible to pin-point the receptive fields of the afferents being recorded and as such they could not be tested with the preferred mechanical stimulus of mucosal stroking. While distension of the loop and digital compression of the wall failed to evoke responses from quality-specific receptors, these stimuli are not adequate to eliminate mechanical sensitivity of an area up to only 7 mm in diameter. This point was acknowledged in a recent paper by Mei and Garnier [27] describing osmosensitive vagal receptors. Twenty-two percent of the afferent units were investigated for mechanical sensitivity, 67% of which responded to mucosal stroking and only 20% responded to strong distension or compression. The authors point out that the failure to respond to distension or mucosal stroking might have been due to inadequacy of the stimulus in either intensity or location but leave open the possibility of specific osmoreceptors. Clearly one must absolutely eliminate the possibility of mechanical sensitivity before a receptor is considered to be specific for a particular chemical stimulus. Do specific chemoreceptors and polymodal receptors represent discrete subpopulations of the mucosal afferent innervation or are the differences due to experimental design? In recent reviews by Mei [25,26], polymodal and specific chemoreceptors are considered as separate subpopulations of the mucosal afferent innervation each with different functional roles: glucose receptors and insulin secretion; nutrient receptors and satiety; acid and the enterogastric reflex. However, one must question why, with few exceptions, papers which describe specific receptors do not throw up the odd non-specific response and vice versa, papers describing polymodal receptors do not find the occasional specific one?
179 The transduction process in afferent endings The importance of mechanical sensitivity becomes apparent when the transducer mechanism of the receptor is considered. A receptor sensitive to both mechanical and chemical stimuli may have a common transduction process-distortion of the receptive field. This may hold true with luminal osmotic stimuli where swelling and shrinking of the extracellular space is likely to distort the receptor ending. A specific chemoreceptor, on the other hand, must have a separate mechanism, possibly an acceptor molecule on the receptor membrane. A post-absorptive stimulus to afferent endings in the ileum is indicated from the abolition of the mesenteric afferent (albeit whole nerve recording) response to glucose by phlorhizin which prevents the transfer of glucose across the luminal membrane of the enterocyte [16]. However, if the stimulus to mucosal endings is post-absorptive, then one must question the existence of, for example, a glucoreceptor which responds equally well to a wide range of other non-digestible and non-absorbable di- and tri-saccharides [24]. An alternative explanation is that mucosal receptors detect pre-absorptive signals rather than post-absorptive ones. Thus the afferent responses would be secondary to release of a humoral agent (5-HT, prostaglandin, peptides, etc.) from cells diffusely spread throughout the gastrointestinal mucosal epithelium where they are in a position to sense luminal contents directly. There is some histological evidence to support such a view. Mucosal epithelial cells with apical finger-like projections into the gut lumen and basal cytoplasmic processes underneath adjacent epithelial cells have also been described [30]. These were associated with submucosal nerve terminals and, because of their similarity to neuroepithelial cells of the tastebud, are suggested to have a receptor function in the intestine. The concept of secondary sense cells in the epithelium with closely associated afferent terminations is, however, complicated by the high turnover of mucosal epithelial cells and their continuous migration to the villus tip. Examples of afferent chemosensitivity to locally released chemical substances are found elsewhere in the visceral domain, for example,
heart and lung [31]. Splanchnic mechanoreceptors supplying the gastrointestinal tract can also be activated by bradykinin, a response consistent with the role of mediating visceral pain [18], while vagal gastric mechanoreceptors have been shown to respond to systemic CCK which is proposed as a mechanism by which CCK may cause satiety [11]. Thus while humoral and neural control mechanisms are known to interact at the final effector it is likely that there is also interaction at the level of the mucosal afferent ending to modify sensory inputs and reflex mechanisms.
Functional considerations The most obvious role for gastrointestinal afferents is providing the pathway for visceral sensations and the behavioural aspects of visceral stimulation, for example, feeding and satiety, nausea and vomiting. As discussed above, pain of gastrointestinal origin is mediated through splanchnic pathways. Some sensations, such as the feeling of fullness after a heavy meal, may be mediated by a vagal pathway. In addition to these sensory functions, vagal and splanchnic afferent fibres provide the means for reflexes whose efferent limbs are also in the autonomic nerves and which play a role in the regulation of gastrointestinal function. Recent work on vagal reflex mechanisms has indicated that vagal afferent inputs to the brainstern are disseminated according to the information they carry [4]. There is a strong mechanoreceptive input to the brainstem circuitry which determines the vagal efferent outflow but only a weak input from luminal chemicals unless they trigger vomiting. One might conclude then that chemoreceptive information is poorly represented in vago-vagal reflexes and may be more important in mediating behavioural effects like nausea, vomiting and satiety. However, when investigating reflex mechanisms one makes certain assumptions about the effectiveness of certain luminal stimuli in evoking an afferent discharge. Whether these stimuli activate specific groups of receptors or act through a common afferent pool has important implications for the interpretation of data on the
180
processing of the afferent information in the brainstem and cord. Clearly, a better knowledge of gastrointestinal sensory mechanisms is crucial to the development of our understanding of gastrointestinal control mechanisms.
References l Andrews, P.L.R. Vagal afferent innervation of the gastrointestinal tract, Prog. Brain Res., 67 (1986) 65-86. 2 Andrews, P.L.R., Grundy, D. and Scratcherd, T., Vagal afferent discharge from mechanoreceptors in different regions of the ferret stomach. J. Physiol. (Lond.), 298 (1980) 513-524. 3 Blackshaw, UA., Grundy, D. and Scratcherd, T., Vagal afferent discharge from gastric mechanoreceptors during contraction and relaxation of the ferret corpus. J. Auton. Nero. Syst., 18 (1987) 19-24. 4 Blackshaw, L.A., Grundy, D. and Scratcherd, T., Involvement of gastrointestinal mechano- and intestinal chemoreceptors in vagal reflexes: an electrophysiological study, J. Auton. Nero. Syst., 18 (1987) 225-234. 5 Cervero, F., Afferent activity evoked by natural stimulation of the biliary system in the ferret, Pain, 13 (1982) 137-151. 6 Clarke, G.D. and Davison, J.S., Mucosal receptors in the gastric antrum and small intestine of the rat with afferent fibres in the cervical vagus, J. Physiol. (Lond.), 284 (1978) 55-67. 7 Clerc, N. and Mei, N., Vagal mechanoreceptors located in the lower oesophageal sphincter of the cat, J. Physiol. (Lond.), 336 (1983) 487-498. 8 Cottrell, D.F. and Iggo, A., Mucosal enteroreceptors with vagal afferent fibres in the proximal duodenum of sheep, J. Physiol. (Lond.), 354 (1984) 497-522. 9 Davison, J.S., Response of single vagal afferent fibres to mechanical and chemical stimulation of the gastric and duodenal mucosa in cats, Q. J. Exp. Physiol., 57 (1972) 405-416. 10 Davison, J.S., The electrophysiology of gastrointestinal chemoreceptors, Digestion, 7 (1972) 312-317. 11 Davison, J.S., Activation of vagal gastric mechanoreceptors by cholecystokinin, Proc. West. Pharmacol. Soc., 29 (1986) 363-366. 12 Dockray, G.J. and Sharkey, K.A., Neurochemistry of visceral afferent neurones, Prog. Brain Res., 67 (1986) 133-148. 13 El Ouazzani, T. and Mei, N., Acido- et glucorecepteurs vagaux de la region gastro-duodenale, Exp. Brain Res., 42 (1981) 442-452. 14 El Ouazzani, T. and Mei, N., Electrophysiological properties and role of the vagal thermoreceptors of lower oesophagus and stomach of cat, Gastroenterology, 83 (1982) 995-1001. 15 GabeUa, G., Structure of muscles and nerves in the gastrointestinal tract. In L.R. Johnson (Ed.), Physiology of the Gastrointestinal Tract, 2nd edn., Raven Press, New York, 1987, pp. 335-382.
16 Hardcastle, J., Hardcastle, P.T. and Sanford, P.A., Effect of actively transported hexoses on afferent nerve discharge from the small intestine, J. Physiol. (Lond.), 285 (1978) 71-84. 17 Harding, R. and Leek, B.F., Rapidly adapting mechanoreceptors in the reticulo-rumen which also respond to chemicals, J. Physiol. (Lond.), 223 (1972) 32-33P. 18 Haupt, P., J~inig, W. and Kohler, W., Response patterns of visceral afferent fibres, supplying the colon, upon chemical and mechanical stimuli, Pflugers Arch., 398 (1983) 41-47. 19 lggo, A., Gastric mucosal receptors with vagal afferent fibres in the cat, Q. J. Exp. Physiol., 42 (1957) 398-409. 20 lggo, A., Gastrointestinal tension receptors with unmyelinated afferent fibres in the vagus of the cat, Q. J. Exp. Physiol., 42 (1957) 130-143. 21 Janig, W. and Morrison, J.F.B., Functional Properties of spinal visceral afferents supplying abdominal and pelvic organs, with special emphasis on visceral nociception, Prog. Brain Res., 67 (1986) 87-114. 22 Jeanningros, R., Vagal unitary response to intestinal amino-acid infusion in the cat: a putative signal for protein induced satiety, Physiol. Behav. 28 (1982) 9-21. 23 Meyer, J.H., Motility of the stomach and gastroduodenal junction. In L.R. Johnson (Ed.), Physiology of the Gastrointestinal Tract, 2nd edn., Raven Press, New York, 1987, pp. 613-629. 24 Mei, N., Vagal glucoreceptors in the small intestine of the cat, J. PhysioL (Lond.), 282 (1978) 485-506. 25 Mei, N., Sensory structures in the viscera, Prog. Sens. Biol., 4 (1983) 1-42. 26 Mei, N., Intestinal chemosensitivity, Physiol. Rev., 65 (1985) 211-237. 27 Mei, N. and Gamier, L., Osmosensitive vagal receptors in the small intestine of the cat, J. Auton. Nero. Syst., 16 (1986) 159-170. 28 Melone, J., Vagal receptors sensitive to lipids in the small intestine of the cat, J. Auton. Nero. Syst., 17 (1986) 231-241. 29 Morrison, J.F.B., The afferent innervation of the gastrointestinal tract. In F. Brooks and P.W. Evers (Eds.), Nerves and the Gut, CB Slack, New Jersey, 1977, pp. 297-326. 30 Newson, B., Ahlman, H., Dahlstrom, A. and Nyhus, L.M., Ultrastrnctural observations in the rat ileal mucosa of possible epithelial "taste cells" and submucosal sensory neurons, Acta Physiol. Scand., 114 (1982) 161-164. 31 Paintal, A.S., The visceral sensations--some basic mechanisms, Prog. Brain Res., 67 (1986) 3-19. 32 Rodrigo, J., deFilipe, J., Robles-Chillida, E.M., Perez-Anton, J.A., Mayo, I. and Gomez, A., Sensory vagal nature and anatomical access paths to esophageal laminar nerve endings in myenteric ganglia. Determination by surgical degeneration methods, Acta Anat., 112 (1982) 47-57. 33 Sato, M. and Koyano, H., Autoradiographic study on the distribution of vagal afferent nerve fibres in the gastroduodenal wall of the rabbit, Brain Res., 400 (1987) 101-109. 34 Sharma, K.N., Receptor mechanisms in the alimentary tract: their excitation and functions. In C.F. Code (Ed.), Handbook of Physiology. Alimentary Canal. Control of Water and Food Intake, Am. Physiol. Soc., Washington, DC, 1967. sect. 6, Vol. 1, pp. 225-237.
175
JAN 00812
Review Article
Speculations on the structure/function relationship for vagal and splanchnic afferent endings supplying the gastrointestinal tract David Grundy Department of Physiology, The University, Western Bank, Sheffield (U.K.)
Key words: Vagal afferent; S p l a n c h n i c afferent; G a s t r o i n t e s t i n a l m e c h a n o r e c e p t o r ; Gastrointestinal chemoreceptor
Abstract This paper discusses some of the unsettled issues in the study of the afferent innervation of the gastrointestinal (GI) tract. Afferent fibres in the vagus and splanchnic nerves have been studied electrophysiologicallyand much has been learnt from single fibre recordings. Splanchnic afferent fibres generally terminate in multiple mechanosensitive endings in the mesentery and serosa where they are in a position to monitor tension on the mesenteric attachments. Other mechanoreceptors following a mainly vagal pathway behave as if they are functionally in-series with the muscle elements of the gut wall and signal muscle tension generated passively by distension and actively during contraction. A third group of afferent endings supply the GI mucosa where they are in a position to signal information on the physical and chemical environment of the gut lumen. A complex picture of mucosal sensitivity has emerged with subpopulations of receptors with polymodal sensitivity and quality-specific mechanoreceptors, thermoreceptors and chemoreceptors. Unfortunately, there is little concensus amongst the different research groups because of different experimental paradigms. One group describes specific chemoreceptors, other groups fail to find them. In this minireview I have speculated on the cause of the often conflicting data on GI afferents and the implications this has for the interpretation of visceral receptor mechanisms.
Introduction The m o t o r a n d secretory activities of the gastro-intestinal tract are regulated to optimize digestion a n d a b s o r p t i o n of nutrients. T h e c o n t r o l m e c h a n i s m s which regulate these activities are organized o n a hierarchical basis with local a n d extrinsic reflexes delivering a variety of n e u r o t r a n s m i t t e r substances to the secretory a n d m o t o r effectors. I n addition, e n d o c r i n e a n d p a r a c r i n e cells within the gastrointestinal m u c o s a ' t a s t e ' the l u m i n a l c o n t e n t s a n d release their c o n t e n t s u n d e r appropriate conditions. The gastrointestinal tract
Correspondence: Department of Physiology, The University, Western Bank, Sheffield, S10 2TN, U.K.
is therefore c o n s t a n t l y exposed to a c h a n g i n g chemical e n v i r o n m e n t o n its l u m i n a l surface, from w i t h i n its i n t e r n a l s t r u c t u r e a n d f r o m the b l o o d s t r e a m . A f f e r e n t e n d i n g s within the wall of the gastrointestinal tract a n d its mesenteric connections m o n i t o r this c h a n g i n g e n v i r o n m e n t a n d f u r t h e r m o r e are exposed to a n y effector responses (motor, secretory or e n d o c r i n e ) m e d i a t e d b y their activation.
Magnitude and morphology of the gastrointestinal afferent innervation G a s t r o i n t e s t i n a l afferent e n d i n g s send axons via the a u t o n o m i c nerves to the b r a i n s t e m a n d spinal cord. I n the vagus nerves of n o n - r u m i n a n t s
0165-1838/88/$03.50 © 1988 Elsevier Science Publishers B.V. (Biomedical Division)
176
afferent fibres projecting to the nucleus tractus solitarius (nTS) from the abdomen outnumber by 10 to one the preganglionic parasympathetic efferent fibres making the reverse connection between brainstem and gut: a total of some 20,000 to 40,000 unmyelinated afferent fibres [1]. While there are proportionally more preganglionic sympathetic efferent fibres than afferent fibres in the splanchnic nerves, the total number of afferents entering the spinal cord is of a similar order of magnitude to that of the vagal supply [21]. Morphological data on the receptor endings are sparse. They are, with the exception of the Pacinian corpuscle, considered to be bare nerve endings [15]. The title of this paper may therefore seem misleading since very little is known about the structure of gastrointestinal afferent endings. Because they are inaccessible to direct electrophysiological investigation there is even less information on impulse generation within these endings. However, the way an afferent fibre responds to applied stimuli allows one to speculate on the structure of afferent endings and the transduction processes which lead to afferent impulse traffic. It is this functional approach that I wish to develop here. Vagal afferent terminals have been visualized using autoradiographic techniques following the injection of tritiated leucine into the nodose ganglia and are seen terminating in the muscle and close to the mucosal epithelium [33]. It is their position in these sites which determines their functional specificity as muscle tension receptors and monitors of luminal contents, respectively. Afferent terminations have also been described in subserosal, myenteric and submucous plexuses where it was suggested that collaterals of vagal afferents participate in axon reflexes [33]. This suggestion arises because neuropeptides including substance P, somatostatin and CCK are transported against the direction of afferent impulses from the cell body and can be followed immunohistochemically on route to the periphery where they can be released by neural impulses [12]. Thus afferent fibres may not only serve a sensory function but could also regulate transmission through the intramural ganglia. Other sensory structures visualized in the oesophageal intramural ganglia have been called intraganglionic laminar endings [32].
These are terminations of vagal afferent fibres which appear as tape-like structures on the inner surface of the ganglia. While it is pure speculation that these are mechanoreceptors, receptors in such a location would be in a position to detect mechanical deformation of the oesophageal wall during the presence or passage of a bolus. Again, however, these may be terminations of afferent collaterals involved in axon reflexes. Both muscle and mucosal endings project predominantly via vagal afferents to the brainstem. A third group of afferents which run mainly in the splanchnic nerves are associated with the visceral peritoneum and terminate either in the serosa, mesentery or omentum as indicated by the distribution of their receptive fields. These splanchnic afferents terminate in multiple mechanosensitive points which monitor tension on the mesenteric attachments and are proposed to mediate visceral reflexes and also signal pain of visceral origin [29].
Visceral sensitivity The range of visceral sensitivity can be gauged from the type of stimuli which give rise to reflex evoked responses. In this respect responses to both mechanical (distension) and chemical stimuli (luminal acidity, osmolarity, etc.) have been described, for example during feedback regulation of gastric emptying [23]. However, only when electrophysiological techniques are used to record directly the nerve impulses in individual sensory nerves can the properties of gastrointestinal afferents be fully and accurately quantified in terms of adequate stimuli, thresholds to stimulation and receptive fields. Such studies have confirmed the existence of both gastrointestinal mechanoreceptors and chemoreceptors although the study of gut receptor mechanisms is still a long way behind similar studies of somatic sensory endings. Part of the problem relates to the technical difficulties of recording from this largely unmyelinated population of nerves whose receptive fields are relatively inaccessible. It is also difficult to deliver quantitatively precise stimuli to a receptor ending when local reflex responses and humoral mechanisms may perturb or potentiate the stimulus.
177
Serosal receptors Splanchnic afferent fibres have between 1 and 8 punctate mechanoreceptive sites ( < 10 mm 2) distributed along mesenteric arteries, at vascular divisions of the arteries and on the wall of the viscus where an artery pierces the serosal surface [29]. The afferent fibres respond to distortion of the receptor ending and as such can respond to distension and contraction of the bowel. Visceral pain in humans and pseudoaffective responses to noxious levels of stimulation in animals is mediated through splanchnic pathways and splanchnic mechanoreceptors are probably responsible [21]. The existence of specific nociceptors within the gastrointestinal tract is a matter for debate [5] with the alternative argument being that both noxious and innocuous events are encoded in the intensity of discharge from the same population of neurones [21]. Serosal receptors may thus provide physiological regulation at low levels of stimulation but signal a noxious event when the intensity of discharge is excessive as might occur during obstruction or spasm. Tension receptors The muscle endings with vagal afferent fibres detect tension within the muscle generated passively by distension or actively during contraction. These are described variably as 'in-series' tension receptors [from 20] or stretch receptors (with an in-parallel location) [see 31] with the latter implying that they respond only to distension. This terminology is ambiguous because only a single functional type of muscle receptor has been characterized electrophysiologically but whose properties differ according to the gastrointestinal region in which they occur. Thus in a readily distensible region like the gastric corpus receptors are primarily sensitive to distension while in the muscular antrum, contraction is the main stimulus [2]. Nevertheless, antral distension and corpus contraction are effective stimuli. Recent work in our laboratory has concentrated on the behaviour of these vagal mechanoreceptors in the gastric corpus during active relaxation and highlights the importance of muscle tension as the primary determinant of receptor activity [3]. Under isometric conditions vagal afferent discharge mirrored pressure
changes during both contraction and relaxation. Under isotonic conditions both pressure changes and afferent discharge were attenuated despite large changes in muscle length. Thus, during a meal receptive relaxation of the stomach would limit wall tension and the stomach would fill with minimal activation of vagal afferent endings. If these receptors contribute to the sensation of satiety as has been suggested many times, then clearly this can not develop until relaxation wanes and corpus tone increases. The classification of in-series and in-parallel receptors in skeletal muscle has an obvious structural basis in the tendon organ and muscle spindle. The distinction is less clear-cut in smooth muscle which is composed of bundles of densely packed cells which are mechanically and electrically coupled [15]. Within the bundles the individual smooth muscle cells are separated from their neighbours by only tens of nanometres. In such a situation it is difficult to envisage an afferent ending in-series with an individual smooth muscle cell. However, the functional unit is not the cell but the bundle of cells that are interconnected by collagen fibrils forming laminar intramuscular septa. These septa are suggested to serve as intramuscular tendons transmitting tension longitudinally from one bundle of cells to the next. The in-series tension receptor in skeletal tendons may therefore have its smooth muscle counterpart in the vagal mechanoreceptors if these were located within the connective tissue matrix. In such a location they would respond to muscle tension generated passively by distension and actively during contraction. However, because the intramuscular septa run parallel to the muscle cells these receptors may function as in-series receptors from an inparallel location. Mucosal receptors Research on mucosal receptors has proved less conclusive than that on muscle and serosal mechanoreceptors because of their heterogeneity. Much recent work in this field has concentrated on the description of quality-specific chemoreceptors detecting various chemical components of chyme in the stomach and small intestine of the cat. Specific glucoreceptors [13,24], amino-acid receptors [22],
178
receptors to long and short chain lipids [28], and acid receptors [13] have all been described by the Marseilles group headed by Mei; as have specific thermoreceptors [14]. The existence of specific chemoreceptors has been readily accepted in the literature because they correspond to the stimuli which have been shown, for example, to inhibit gastric emptying or mediate satiety [23]. However, the evidence for such receptors is not unequivocal. Firstly, all of these stimuli evoke local responses and the afferent discharge may be secondary to these. Secondly there are anomalies in the literature. While responses to luminal chemicals have been described by many workers using whole nerve or multiunit recording techniques [see 10 and 34 for reviews of multiunit literature], the value of such studies is limited because of the inability to distinguish between a population of afferent fibres capable of responding to many different stimuli and subpopulations which are specific to different types of stimuli. It is also impossible to pin-point receptive fields in such preparations. These problems are overcome when recordings are made from single afferent fibres. All descriptions of single mucosal afferent fibres from outside Marseilles, using a wide variety of animal models including the cat, describe them as multi- or polymodal. In addition to responding to a wide variety of unrelated chemicals, these receptors also respond to mechanical stimulation in the form of mucosal deformation produced by lightly stroking the mucosa overlying the circular receptive field [6,8,9,17,19]. The major difference between these studies and those in which specific receptors were identified relates to the preparation of the viscera for the application of stimuli. Polymodal receptors were identified in preparation in which the viscera was opened and the lumen exposed for probing and direct application of chemicals while specific receptors were found in closed perfused preparations of loops of bowel (usually about 40 cm long). By surgically exposing the luminal surface, the mucous barrier overlying the mucosal epithelium may dry out or it may become eroded. In such preparations the mucosal epithelium may be exposed to stimuli it would not normally encounter and receptor endings may respond in a nonspecific manner, i.e. polymodal sensitivity. How-
ever, the exquisite sensitivity of polymodal receptors to mechanical stimuli, responding for example to a thread of cotton trailed over the receptive field, would argue in favour of this being a primary sensitivity of the endings. Indeed some mucosal receptors respond only to mechanical stimulation [7,8]. These endings may therefore function as contact receptors detecting the particulate nature of chyme [10]. On the other hand, in closed loops it is impossible to pin-point the receptive fields of the afferents being recorded and as such they could not be tested with the preferred mechanical stimulus of mucosal stroking. While distension of the loop and digital compression of the wall failed to evoke responses from quality-specific receptors, these stimuli are not adequate to eliminate mechanical sensitivity of an area up to only 7 mm in diameter. This point was acknowledged in a recent paper by Mei and Garnier [27] describing osmosensitive vagal receptors. Twenty-two percent of the afferent units were investigated for mechanical sensitivity, 67% of which responded to mucosal stroking and only 20% responded to strong distension or compression. The authors point out that the failure to respond to distension or mucosal stroking might have been due to inadequacy of the stimulus in either intensity or location but leave open the possibility of specific osmoreceptors. Clearly one must absolutely eliminate the possibility of mechanical sensitivity before a receptor is considered to be specific for a particular chemical stimulus. Do specific chemoreceptors and polymodal receptors represent discrete subpopulations of the mucosal afferent innervation or are the differences due to experimental design? In recent reviews by Mei [25,26], polymodal and specific chemoreceptors are considered as separate subpopulations of the mucosal afferent innervation each with different functional roles: glucose receptors and insulin secretion; nutrient receptors and satiety; acid and the enterogastric reflex. However, one must question why, with few exceptions, papers which describe specific receptors do not throw up the odd non-specific response and vice versa, papers describing polymodal receptors do not find the occasional specific one?
179 The transduction process in afferent endings The importance of mechanical sensitivity becomes apparent when the transducer mechanism of the receptor is considered. A receptor sensitive to both mechanical and chemical stimuli may have a common transduction process-distortion of the receptive field. This may hold true with luminal osmotic stimuli where swelling and shrinking of the extracellular space is likely to distort the receptor ending. A specific chemoreceptor, on the other hand, must have a separate mechanism, possibly an acceptor molecule on the receptor membrane. A post-absorptive stimulus to afferent endings in the ileum is indicated from the abolition of the mesenteric afferent (albeit whole nerve recording) response to glucose by phlorhizin which prevents the transfer of glucose across the luminal membrane of the enterocyte [16]. However, if the stimulus to mucosal endings is post-absorptive, then one must question the existence of, for example, a glucoreceptor which responds equally well to a wide range of other non-digestible and non-absorbable di- and tri-saccharides [24]. An alternative explanation is that mucosal receptors detect pre-absorptive signals rather than post-absorptive ones. Thus the afferent responses would be secondary to release of a humoral agent (5-HT, prostaglandin, peptides, etc.) from cells diffusely spread throughout the gastrointestinal mucosal epithelium where they are in a position to sense luminal contents directly. There is some histological evidence to support such a view. Mucosal epithelial cells with apical finger-like projections into the gut lumen and basal cytoplasmic processes underneath adjacent epithelial cells have also been described [30]. These were associated with submucosal nerve terminals and, because of their similarity to neuroepithelial cells of the tastebud, are suggested to have a receptor function in the intestine. The concept of secondary sense cells in the epithelium with closely associated afferent terminations is, however, complicated by the high turnover of mucosal epithelial cells and their continuous migration to the villus tip. Examples of afferent chemosensitivity to locally released chemical substances are found elsewhere in the visceral domain, for example,
heart and lung [31]. Splanchnic mechanoreceptors supplying the gastrointestinal tract can also be activated by bradykinin, a response consistent with the role of mediating visceral pain [18], while vagal gastric mechanoreceptors have been shown to respond to systemic CCK which is proposed as a mechanism by which CCK may cause satiety [11]. Thus while humoral and neural control mechanisms are known to interact at the final effector it is likely that there is also interaction at the level of the mucosal afferent ending to modify sensory inputs and reflex mechanisms.
Functional considerations The most obvious role for gastrointestinal afferents is providing the pathway for visceral sensations and the behavioural aspects of visceral stimulation, for example, feeding and satiety, nausea and vomiting. As discussed above, pain of gastrointestinal origin is mediated through splanchnic pathways. Some sensations, such as the feeling of fullness after a heavy meal, may be mediated by a vagal pathway. In addition to these sensory functions, vagal and splanchnic afferent fibres provide the means for reflexes whose efferent limbs are also in the autonomic nerves and which play a role in the regulation of gastrointestinal function. Recent work on vagal reflex mechanisms has indicated that vagal afferent inputs to the brainstern are disseminated according to the information they carry [4]. There is a strong mechanoreceptive input to the brainstem circuitry which determines the vagal efferent outflow but only a weak input from luminal chemicals unless they trigger vomiting. One might conclude then that chemoreceptive information is poorly represented in vago-vagal reflexes and may be more important in mediating behavioural effects like nausea, vomiting and satiety. However, when investigating reflex mechanisms one makes certain assumptions about the effectiveness of certain luminal stimuli in evoking an afferent discharge. Whether these stimuli activate specific groups of receptors or act through a common afferent pool has important implications for the interpretation of data on the
180
processing of the afferent information in the brainstem and cord. Clearly, a better knowledge of gastrointestinal sensory mechanisms is crucial to the development of our understanding of gastrointestinal control mechanisms.
References l Andrews, P.L.R. Vagal afferent innervation of the gastrointestinal tract, Prog. Brain Res., 67 (1986) 65-86. 2 Andrews, P.L.R., Grundy, D. and Scratcherd, T., Vagal afferent discharge from mechanoreceptors in different regions of the ferret stomach. J. Physiol. (Lond.), 298 (1980) 513-524. 3 Blackshaw, UA., Grundy, D. and Scratcherd, T., Vagal afferent discharge from gastric mechanoreceptors during contraction and relaxation of the ferret corpus. J. Auton. Nero. Syst., 18 (1987) 19-24. 4 Blackshaw, L.A., Grundy, D. and Scratcherd, T., Involvement of gastrointestinal mechano- and intestinal chemoreceptors in vagal reflexes: an electrophysiological study, J. Auton. Nero. Syst., 18 (1987) 225-234. 5 Cervero, F., Afferent activity evoked by natural stimulation of the biliary system in the ferret, Pain, 13 (1982) 137-151. 6 Clarke, G.D. and Davison, J.S., Mucosal receptors in the gastric antrum and small intestine of the rat with afferent fibres in the cervical vagus, J. Physiol. (Lond.), 284 (1978) 55-67. 7 Clerc, N. and Mei, N., Vagal mechanoreceptors located in the lower oesophageal sphincter of the cat, J. Physiol. (Lond.), 336 (1983) 487-498. 8 Cottrell, D.F. and Iggo, A., Mucosal enteroreceptors with vagal afferent fibres in the proximal duodenum of sheep, J. Physiol. (Lond.), 354 (1984) 497-522. 9 Davison, J.S., Response of single vagal afferent fibres to mechanical and chemical stimulation of the gastric and duodenal mucosa in cats, Q. J. Exp. Physiol., 57 (1972) 405-416. 10 Davison, J.S., The electrophysiology of gastrointestinal chemoreceptors, Digestion, 7 (1972) 312-317. 11 Davison, J.S., Activation of vagal gastric mechanoreceptors by cholecystokinin, Proc. West. Pharmacol. Soc., 29 (1986) 363-366. 12 Dockray, G.J. and Sharkey, K.A., Neurochemistry of visceral afferent neurones, Prog. Brain Res., 67 (1986) 133-148. 13 El Ouazzani, T. and Mei, N., Acido- et glucorecepteurs vagaux de la region gastro-duodenale, Exp. Brain Res., 42 (1981) 442-452. 14 El Ouazzani, T. and Mei, N., Electrophysiological properties and role of the vagal thermoreceptors of lower oesophagus and stomach of cat, Gastroenterology, 83 (1982) 995-1001. 15 GabeUa, G., Structure of muscles and nerves in the gastrointestinal tract. In L.R. Johnson (Ed.), Physiology of the Gastrointestinal Tract, 2nd edn., Raven Press, New York, 1987, pp. 335-382.
16 Hardcastle, J., Hardcastle, P.T. and Sanford, P.A., Effect of actively transported hexoses on afferent nerve discharge from the small intestine, J. Physiol. (Lond.), 285 (1978) 71-84. 17 Harding, R. and Leek, B.F., Rapidly adapting mechanoreceptors in the reticulo-rumen which also respond to chemicals, J. Physiol. (Lond.), 223 (1972) 32-33P. 18 Haupt, P., J~inig, W. and Kohler, W., Response patterns of visceral afferent fibres, supplying the colon, upon chemical and mechanical stimuli, Pflugers Arch., 398 (1983) 41-47. 19 lggo, A., Gastric mucosal receptors with vagal afferent fibres in the cat, Q. J. Exp. Physiol., 42 (1957) 398-409. 20 lggo, A., Gastrointestinal tension receptors with unmyelinated afferent fibres in the vagus of the cat, Q. J. Exp. Physiol., 42 (1957) 130-143. 21 Janig, W. and Morrison, J.F.B., Functional Properties of spinal visceral afferents supplying abdominal and pelvic organs, with special emphasis on visceral nociception, Prog. Brain Res., 67 (1986) 87-114. 22 Jeanningros, R., Vagal unitary response to intestinal amino-acid infusion in the cat: a putative signal for protein induced satiety, Physiol. Behav. 28 (1982) 9-21. 23 Meyer, J.H., Motility of the stomach and gastroduodenal junction. In L.R. Johnson (Ed.), Physiology of the Gastrointestinal Tract, 2nd edn., Raven Press, New York, 1987, pp. 613-629. 24 Mei, N., Vagal glucoreceptors in the small intestine of the cat, J. PhysioL (Lond.), 282 (1978) 485-506. 25 Mei, N., Sensory structures in the viscera, Prog. Sens. Biol., 4 (1983) 1-42. 26 Mei, N., Intestinal chemosensitivity, Physiol. Rev., 65 (1985) 211-237. 27 Mei, N. and Gamier, L., Osmosensitive vagal receptors in the small intestine of the cat, J. Auton. Nero. Syst., 16 (1986) 159-170. 28 Melone, J., Vagal receptors sensitive to lipids in the small intestine of the cat, J. Auton. Nero. Syst., 17 (1986) 231-241. 29 Morrison, J.F.B., The afferent innervation of the gastrointestinal tract. In F. Brooks and P.W. Evers (Eds.), Nerves and the Gut, CB Slack, New Jersey, 1977, pp. 297-326. 30 Newson, B., Ahlman, H., Dahlstrom, A. and Nyhus, L.M., Ultrastrnctural observations in the rat ileal mucosa of possible epithelial "taste cells" and submucosal sensory neurons, Acta Physiol. Scand., 114 (1982) 161-164. 31 Paintal, A.S., The visceral sensations--some basic mechanisms, Prog. Brain Res., 67 (1986) 3-19. 32 Rodrigo, J., deFilipe, J., Robles-Chillida, E.M., Perez-Anton, J.A., Mayo, I. and Gomez, A., Sensory vagal nature and anatomical access paths to esophageal laminar nerve endings in myenteric ganglia. Determination by surgical degeneration methods, Acta Anat., 112 (1982) 47-57. 33 Sato, M. and Koyano, H., Autoradiographic study on the distribution of vagal afferent nerve fibres in the gastroduodenal wall of the rabbit, Brain Res., 400 (1987) 101-109. 34 Sharma, K.N., Receptor mechanisms in the alimentary tract: their excitation and functions. In C.F. Code (Ed.), Handbook of Physiology. Alimentary Canal. Control of Water and Food Intake, Am. Physiol. Soc., Washington, DC, 1967. sect. 6, Vol. 1, pp. 225-237.