- Email: [email protected]
Signalling the state of the digestive tract
Autonomic Neuroscience: Basic and Clinical 125 (2006) 76 – 80 www.elsevier.com/locate/autneu
Review
Signalling the state of the digestive tract David Grundy Department of Biomedical Science, University of Sheffield, Sheffield, S10 2TN, United Kingdom Received 14 January 2006; received in revised form 14 January 2006; accepted 14 January 2006
Abstract The gastrointestinal tract has a rich sensory innervation. Extrinsic afferents in vagal, splanchnic and pelvic nerves project to the CNS where gut reflex function is coordinated and integrated with behavioural responses (e.g. regulation of food intake) and mediate sensations. The afferent information conveyed by vagal and spinal mechanosensitive afferents can be very different. Vagal afferents have low thresholds of activation and reach maximal responses within physiological levels of distension. In contrast, spinal afferents, although many have corresponding thresholds for activation, are able to respond beyond the physiological range and encode both physiological and noxious levels of stimulation. However, mechanosensitivity is not fixed but can be influenced by a wide range of chemical mediators released as a consequence of ischemia, injury and inflammation. Indeed, previously mechanical insensitive afferents can develop mechanosensitivity during inflammation and a variety of chemical mediators are implicated in this sensitisation process. Chemosensitivity is also a property of vagal mucosal afferents that detect the chemical milieu for chemicals absorbed across the epithelium or released from enteroendocrine cells that are strategically positioned to ‘‘taste’’ luminal contents. Thus, there exists a complex interplay between immunomodulators, neurotransmitters and neuroendocrine factors that underlie gastrointestinal sensing mechanisms and enable orchestration of appropriate host responses. D 2006 Elsevier B.V. All rights reserved.
Contents 1.
Introduction . . . . . . . . . . . . . . . 1.1. Morphological features of sensory 1.1.1. Vagal afferents. . . . . . 1.1.2. Spinal afferents . . . . . 2. Mechanosensitivity . . . . . . . . . . . . 2.1. Vagal afferents . . . . . . . . . . 2.2. Spinal afferents . . . . . . . . . . 3. Peripheral sensitization . . . . . . . . . . 4. Sensing luminal contents . . . . . . . . . 4.1. Protons . . . . . . . . . . . . . . 5. Conclusions . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . endings in the gut wall . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1. Introduction The sensory innervation of the GI tract is responsible for monitoring the various aspects of the digestive process, including motility, secretion, absorption and blood flow. E-mail address: [email protected]. 1566-0702/$ - see front matter D 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.autneu.2006.01.009
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
76 77 77 77 78 78 78 78 79 79 79 79
However, sensory in this context does not necessary imply sensation since much of this information rarely impinges on consciousness being more involved in autonomic reflexes and regulating behavioural mechanisms associated with food intake. The sensory terminals of vagal and spinal afferents are found at various levels within the gut wall including muscle,
D. Grundy / Autonomic Neuroscience: Basic and Clinical 125 (2006) 76 – 80
mucosal epithelia and enteric ganglia. Other endings terminate in the serosa and mesenteric attachments and form a dense network around mesenteric blood vessels and their tributaries in the gut wall. These various ending maintain a steady flow of afferent traffic to the CNS relating information on activity both within and outside the gut wall. The sensory information is conveyed to the CNS by vagal and spinal pathways to the brainstem and spinal cord, from where information is processed and relayed to higher brains areas (Joshi and Gebhart, 2000). The vagal and spinal pathways predominantly transmit different aspects of sensory information. Vagal neurones generally process physiological information e.g. the nature and composition of the luminal contents, the presence and amplitude of ongoing motor activity of the gut. In contrast, spinal neurones process mainly pathophysiological information e.g. potentially noxious mechanical or chemical stimuli arising through tissue injury, ischaemia and inflammation. However, vagal and spinal pathways share a number of functional properties and there is evidence for interplay between these two pathways in reflex control, in behaviours and in sensation. 1.1. Morphological features of sensory endings in the gut wall Afferent nerves largely follow the same pathway to the CNS as the efferent fibres leaving it in parasympathetic and sympathetic autonomic nerves. The various populations of afferents are therefore defines as vagal, pelvic and splanchnic afferents. Afferent fibres outnumber efferent fibres by about 10 : 1 in the vagus nerves, so one might consider this a predominantly sensory nerve. Vagal cell bodies lie in the nodose ganglia and project centrally to the nucleus tractus solitarius in the brainstem. The density of spinal afferents is more scant with < 7% of sensory cell bodies in the dorsal root ganglia (DRG) projecting to the viscera. The peripheral terminals of vagal and spinal afferents have been localised within the GI wall using fibre tracing techniques (Berthoud and Neuhuber, 2000). Both vagal and spinal nerves innervate the entire length gastrointestinal tract, but vagal afferents are more prevalent in the proximal gut and spinal, particularly pelvic afferents, predominate in the distal gut. The location of different afferent endings in the mucosa, muscle layers, and in the serosal and mesenteric attachments is consistent with their ability to respond to stimuli acting at these different sites within the GI wall. Although some afferent terminals are specialised in the way their terminals arbourise in the gut wall there are non of the specialized endings as found in the skin but instead the GI sensory nerves are bare nerve endings. Thus the differential sensitivity of afferents arises from both their location in the gut wall and the receptors and ion channels that they express on their terminals. 1.1.1. Vagal afferents Different populations of vagal afferent neurones terminate in either the mucosa or the muscle. Afferent endings in
77
the mucosa are closely associated with the lamina propria in proximity to the mucosal epithelium, but are never exposed directly to the contents of the lumen. Vagal afferents have been visualised close to specialised epithelial cells such as enterochromaffin cells and mast cells which may be the actual sensory transducers that monitor the physical and chemical nature of luminal contents (Phillips and Powley, 2000). Vagal afferent endings in the muscle can be classified into two main types: intramuscular arrays (IMA) and intraganglionic laminar endings (IGLE). IMAs are distributed within the muscle sheets, especially in the longitudinal muscle, parallel to the long axes of muscle fibres (Berthoud and Neuhuber, 2000). They are interposed between bundles of smooth muscle fibres, but they also course on, and form appositions with, intramuscular interstitial cells of Cajal. In contrast IGLEs are basket-like structures surrounding myenteric ganglia. As IGLEs are located between the circular and longitudinal muscle layers, they are exposed to shearing forces generated during muscle stretch or contraction. Evidence showing they respond to such mechanical forces has been elaborated recently by mapping the receptor fields of vagal afferent endings in the oesophagus and showing morphologically that these Fhot-spots_ to mechanical probing corresponds to the location of IGLEs (Zagorodnyuk and Brookes, 2000). IGLEs are the primary candidates for conveying mechanosensory information relevant to distension and contraction of the bowel wall. However, the proximity between IGLEs and myenteric ganglion neurones raises the intriguing possibility that they may also detect chemicals in the synaptic neuropil and as such provide a morphological substrate for ‘‘crosstalk’’ between the enteric nervous system and extrinsic afferents. 1.1.2. Spinal afferents Spinal afferent nerve terminals are located primarily in the serosa and muscle layers of the gut. Spinal afferents are largely unmyelinated and have multiple branching punctate endings that correspond to multiple receptive fields. Their location and response characteristics suggest that spinal afferents respond to distortion of the viscera (Gebhart, 2000). Spinal afferents also innervate the mucosa and are implicated in mucosal protection (Holzer, 1998). This protective role arises because spinal afferents have collateral branches that supply blood vessels and innervate the enteric ganglia. Activation of an afferent terminal causes an action potential to be propagated centrally, but action potentials can also propagate down axon collaterals and stimulate the release of neurotransmitters in a local axon reflex. In this manner, the release of transmitters from their varicose nerve terminals can modulate blood flow and enteric reflex pathways. The main neurotransmitters present in spinal afferents are calcitonin gene related peptide (CGRP) and substance P (SP) (Sternini et al., 1992). Both of these peptides are implicated in GI neurogenic inflammation and
78
D. Grundy / Autonomic Neuroscience: Basic and Clinical 125 (2006) 76 – 80
so their release via axon reflexes may be involved in inflammatory responses (Keates et al., 1998).
2. Mechanosensitivity Both vagal and spinal afferents in the muscle and mesenteric connections are mechanosensitive. In response to distension vagal afferents have a lower threshold for activation than spinal afferents (Sengupta et al., 1990). Little is known about the precise elements within GI afferent endings that are activated by mechanical activity and transduce this stimulus into action potential firing. In other sensory systems, mechanosensitive ion channels have been described whereby a mechanical stress on a receptor molecule in the neuronal cell membrane distorts it in such a way to cause ion channel activation and action potential firing. It can only be speculated that mechanosensitive ion channels are expressed by vagal and spinal mechanosensitive afferent neurones. The differing thresholds for activation may be explained by the expression of different ion channels and/or the different locations of vagal and spinal afferent terminals. 2.1. Vagal afferents Direct electrophysiological recordings of vagal afferent traffic en route to the CNS demonstrate that vagal afferents have low thresholds of activation and responses plateau within physiological levels of distension. Vagal afferents are described as Fin-series tension receptors_ by analogy with Golgi tendon organs which are sensitive to tension developed in the gut wall by stretch during distension and during contraction. As discussed above IGLEs have been proposed to be the Fin-series tension receptor_, although situated in the myenteric ganglia they may do this from an in parallel location, presumably since contraction of either muscle layer will generate shear forces in the interposed ganglia. 2.2. Spinal afferents Spinal afferents generally have higher thresholds for activation and in many cases would only be activated by levels of distension or contraction that would be considered noxious (Sengupta et al., 1990). However, some spinal afferents, particular running in the pelvic nerves, respond to lower levels of stimulation within the physiological range and indeed IGLEs have been described in the rectum (Lynn et al., 2003). Other afferents also have low thresholds for activation but because these endings continue to encode into the noxious range these afferents are said to have a Fwide dynamic range_ by analogy with second order neurons in the spinal cord that also respond to low and high intensity stimulation. These afferent endings therefore would encode both physiological and noxious levels of stimulation. These
different stimulus– response profiles are consistent with the view that it is high threshold and wide-dynamic range afferents that provide an intensity-related code that conveys noxious information and is responsible for mediating pain. Mechanosensitive thresholds and the relationship between stimulus and response are not fixed but can be influenced by a wide range of chemical mediators released as a consequence of injury, ischaemia and inflammation. A wide range of chemical mediators may interact in a potentiating way to modulate the sensitivity of spinal afferent endings, reducing the threshold for activation to cause hypersensitivity (Kirkup et al., 2001). Previously insensitive afferents have also been shown to develop mechanosensitivity during inflammation, these are referred to as ‘‘sleeping’’ or ‘‘silent’’ nociceptors.
3. Peripheral sensitization The threshold for activation of sensory neurones is not fixed but can be modulated by a number of different mediators. Such plasticity in afferent neurones can lead to sensitisation and hyperalgesia. Sensitisation is characterised by a decrease in threshold for firing of neurones and an elevated response to suprathreshold stimuli. This sensitisation of neuronal endings can lead to hyperalgesia, which is the hallmark of irritable bowel syndrome (IBS). At a molecular level, sensitisation can occur through at least three different mechanisms: i) a change in receptor sensitivity or ion channel gating; ii) an upregulation in the number of receptors and channels expressed on afferent endings; iii) the expression of new receptors and channels (Kirkup et al., 2001). Some receptors can be activated and potentiated by more than one stimuli e.g. TRPV1 receptor activity is modulated by capsaicin, pH, cannabinoids and heat. Other receptors share second messenger pathways and hence can be modulated indirectly by feedback from different mediators (Wood, 2001). A variety of factors can trigger sensitising responses. Sensory modulation occurs during inflammation, injury, infection or ischaemia, which involves a plethora of different cell types e.g. platelets, leukocytes, lymphocytes, macrophage, mast cells. Each of these cell types can release a host of different neuromodulators. Regardless of the precise mechanism, the normal sensitivity of neurones is altered. This neuronal sensitisation is often a necessary and appropriate response in the short term. However, if sensitisation persists after the challenge to the GI tract has been eliminated, this can lead to inappropriate sensory processing and long-term hyperalgesia. One of the hypotheses for IBS is that a transient gut infection evokes an inappropriate long-lasting gut hypersensitivity (Barbara et al., 2002). Both spinal and vagal afferents can become sensitised, although it is primarily spinal afferents that have been implicated in hyperalgesia. However, there is some evidence
D. Grundy / Autonomic Neuroscience: Basic and Clinical 125 (2006) 76 – 80
that vagal afferents may play a role in hyperalgesia as some vagal afferents project into the cervical region of the spinal cord and these fibres may be involved in transmitting sensation to thalamic nuclei. In addition, vagal afferents have been shown to facilitate nociceptive transmission and these pathways are believed to be implicated in the hyperalgesia that arises as a consequence of illness behaviour triggered by some cytokines, especially interleukin-1h (Watkins and Maier, 2000).
79
of long chain fatty acids is dependent upon chylomicron formation and it is not clear the extent to which CCK release, and hence neuronal activation, is reliant upon chylomicron formation. Short chain fatty acids directly activate vagal afferents via a post-absorptive CCK-independent mechanism (Lal et al., 2001). CCK is also implicated in sensing luminal protein digestion products (Eastwood et al., 1998). 4.1. Protons
4. Sensing luminal contents Afferents nerves terminating near to the mucosa are in a position to monitor the composition of the luminal contents. As afferents do not project directly into the lumen, their activation depends on an intermediary step that could either be post-absorptive i.e. neuronal activation by absorbed substances from the lumen, or pre-absorptive i.e. neuronal activation by a secondary substance released from within the mucosal epithelium. In this respect a role for both CCK and 5-HT released from enteroendocrine cells and acting as paracrine agents on the terminals of vagal afferents has been implicated in responses to a number of luminal signals (Raybould, 1999). Enterochromaffin cells (EC cells) release 5-HT and are strategically positioned in the intestinal mucosa to ‘‘taste’’ luminal contents and release their mediators across the basolateral membrane in order to generate action potentials in the afferent nerve endings within the lamina propria (Gershon, 1999). 5-HT liberated from EC cells acts directly on vagal extrinsic afferent nerves in the mucosa through activation of 5-HT3 receptors expressed on the nerve terminal. The physiological stimuli for the release of 5-HT from EC cells may be both chemical and mechanical, suggesting a role in both chemotransduction and mechanotransduction. A large body of data implicates EC cells in the detection of bacterial enterotoxins e.g. cholera toxin. These toxins trigger release of 5-HT from EC cells, to bring about an orchestrated response to dilute and subsequently eliminate the pathogenic material from the body through diarrhoea and vomiting and preclude further consumption of the potentially harmful material by causing nausea. 5-hydroxytryptamine release from enterochromaffin (EC) cells is also implicated in nutrient signalling particularly in relation to luminal sugars (Raybould, 1999). In contrast, lipids activate vagal afferents through a mixture of both pre- and post-absorptive mechanisms, and this varies according to the type of fatty acids present in the lumen. Long chain fatty acids indirectly activate vagal afferents through a cholecystokinin (CCK) dependent mechanism. CCK is released from specialised enteroendocrine cells in the intestinal epithelia and activates vagal afferents, probably in a paracrine manner acting on CCK(A) receptors on vagal afferent terminals. Absorption
Responses to change in luminal pH are observed in vagal afferents. It is speculated that this sensitivity is due to the activation of receptors that can be modulated directly by protons. A number of vagal afferents supplying the gut are sensitive to capsaicin, the response of which is mediated by the TRPV1 receptor. This receptor is also known to be sensitive to pH, and may therefore provide the molecular transducer of changes in pH (Rong et al., 2004). However, other proton sensing channels exist including members of the ASIC family and P2X receptors.
5. Conclusions Afferent fibres convey a vast amount of sensory information to the brainstem and spinal cord, but the nature of this information differs between vagal and spinal pathways. Vagal afferents convey predominantly physiological information whilst spinal afferents additionally encode noxious events. However, there is evidence for overlap and interplay between the sensitivity and functions of vagal and spinal afferents. Spinal nociceptors are markedly influenced by peripherally acting chemicals which are released during inflammation and injury and they are thought to trigger the processes leading to sensitisation and increased nociceptive activity. Other chemicals act in a more selective way to activate vagal afferents and are implicated in nutrient signalling from the GI tract. References Barbara, G., De Giorgio, R., Stanghellini, V., Cremon, C., Corinaldesi, R., 2002. A role for inflammation in irritable bowel syndrome? Gut 51 (Suppl 1), i41 – i44. Berthoud, B., Neuhuber, W.L., 2000. Functional and chemical anatomy of the afferent vagal system. Auton. Neurosci. 85, 1 – 17. Eastwood, C., Maubach, K., Kirkup, A.J., et al., 1998. The role of endogenous cholecystokinin in the sensory transduction of luminal nutrient signals in the rat jejunum. Neurosci. Lett. 254, 145 – 148. Gebhart, G.F., 2000. Pathobiology of visceral pain: molecular mechanisms and therapeutic implications IV. Visceral afferent contributions the pathobiology of visceral pain. Am. J. Physiol.: Gastrointest. Liver Physiol. 278, G834 – G838. Gershon, M.D., 1999. Review article: roles played by 5-hydroxytryptamine in the physiology of the bowel. Aliment. Pharmacol. Ther. 13, 15 – 30. Holzer, P., 1998. Neural emergency system in the stomach. Gastroenterology 114, 823 – 839.
80
D. Grundy / Autonomic Neuroscience: Basic and Clinical 125 (2006) 76 – 80
Joshi, S.K., Gebhart, G.F., 2000. Visceral pain. Curr. Rev. Pain 4, 499 – 506. Keates, A.C., Castagliuolo, I., Qiu, B., et al., 1998. CGRP upregulation in dorsal root ganglia and ileal mucosa during Clostridium difficile toxin A-induced enteritis. Am. J. Physiol. 274, G196 – G202. Kirkup, A.J., Brunsden, A.M., Grundy, D.I., 2001. Receptors on visceral afferents. Am. J. Physiol.: Gastrointest. Liver Physiol. 280, G7878 – G7894. Lal, S., Kirkup, A.J., Brunsden, A., Thompson, D.G., Grundy, D., 2001. Vagal afferent responses to fatty acids of different chain length in the rat. Am. J. Physiol.: Gastrointest. Liver Physiol. 281, G907 – G915. Lynn, P.A., Olsson, C., Zagorodnyuk, V., Costa, M., Brookes, S.J., 2003. Rectal intraganglionic laminar endings are transduction sites of extrinsic mechanoreceptors in the guinea pig rectum. Gastroenterology 125, 786 – 794. Phillips, R.J., Powley, T.L., 2000. Tension and stretch receptors in gastrointestinal smooth muscle: re-evaluating vagal mechanoreceptor electrophysiology. Brain Res. Rev. 34, 1 – 26. Raybould, H.E., 1999. Nutrient tasting and signaling mechanisms in the gut. I. Sensing of lipid by the intestinal mucosa. Am. J. Physiol. 277, G751 – G755.
Rong, W., Hillsley, K., Davis, J.B., Hicks, G.A., Winchester, W.J., Grundy, D., 2004. Jejunal afferent nerve sensitivity in wild type and TRPV1 knockout mice. J. Physiol. Sengupta, J.N., Saha, J.K., Goyal, R.K., 1990. Stimulus – response function studies of esophageal mechanosensitive nociceptors in sympathetic afferents of opossum. J. Neurophysiol. 64, 796 – 812. Sternini, C., De Giorgio, R., Furness, J.B., 1992. Calcitonin gene-related peptide neurons innervating the cat digestive system. Regul. Pept. 42, 15 – 26. Watkins, L.R., Maier, S.F., 2000. The pain of being sick: implications of immune-to-brain communication for understanding pain. Annu. Rev. Psychol. 51, 29 – 57. Wood, J.N., 2001. Pathobiology of visceral pain: molecular mechanisms and therapeutic implications II. Genetic approaches to pain therapy. Am. J. Physiol.: Gastrointest. Liver Physiol. 278, G507 – G512. Zagorodnyuk, V.P., Brookes, S.J., 2000. Transduction sites of vagal mechanoreceptors in the guinea pig esophagus. J. Neurosci. 20, 6249 – 6255.
Review
Signalling the state of the digestive tract David Grundy Department of Biomedical Science, University of Sheffield, Sheffield, S10 2TN, United Kingdom Received 14 January 2006; received in revised form 14 January 2006; accepted 14 January 2006
Abstract The gastrointestinal tract has a rich sensory innervation. Extrinsic afferents in vagal, splanchnic and pelvic nerves project to the CNS where gut reflex function is coordinated and integrated with behavioural responses (e.g. regulation of food intake) and mediate sensations. The afferent information conveyed by vagal and spinal mechanosensitive afferents can be very different. Vagal afferents have low thresholds of activation and reach maximal responses within physiological levels of distension. In contrast, spinal afferents, although many have corresponding thresholds for activation, are able to respond beyond the physiological range and encode both physiological and noxious levels of stimulation. However, mechanosensitivity is not fixed but can be influenced by a wide range of chemical mediators released as a consequence of ischemia, injury and inflammation. Indeed, previously mechanical insensitive afferents can develop mechanosensitivity during inflammation and a variety of chemical mediators are implicated in this sensitisation process. Chemosensitivity is also a property of vagal mucosal afferents that detect the chemical milieu for chemicals absorbed across the epithelium or released from enteroendocrine cells that are strategically positioned to ‘‘taste’’ luminal contents. Thus, there exists a complex interplay between immunomodulators, neurotransmitters and neuroendocrine factors that underlie gastrointestinal sensing mechanisms and enable orchestration of appropriate host responses. D 2006 Elsevier B.V. All rights reserved.
Contents 1.
Introduction . . . . . . . . . . . . . . . 1.1. Morphological features of sensory 1.1.1. Vagal afferents. . . . . . 1.1.2. Spinal afferents . . . . . 2. Mechanosensitivity . . . . . . . . . . . . 2.1. Vagal afferents . . . . . . . . . . 2.2. Spinal afferents . . . . . . . . . . 3. Peripheral sensitization . . . . . . . . . . 4. Sensing luminal contents . . . . . . . . . 4.1. Protons . . . . . . . . . . . . . . 5. Conclusions . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . endings in the gut wall . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1. Introduction The sensory innervation of the GI tract is responsible for monitoring the various aspects of the digestive process, including motility, secretion, absorption and blood flow. E-mail address: [email protected]. 1566-0702/$ - see front matter D 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.autneu.2006.01.009
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
. . . . . . . . . . . .
76 77 77 77 78 78 78 78 79 79 79 79
However, sensory in this context does not necessary imply sensation since much of this information rarely impinges on consciousness being more involved in autonomic reflexes and regulating behavioural mechanisms associated with food intake. The sensory terminals of vagal and spinal afferents are found at various levels within the gut wall including muscle,
D. Grundy / Autonomic Neuroscience: Basic and Clinical 125 (2006) 76 – 80
mucosal epithelia and enteric ganglia. Other endings terminate in the serosa and mesenteric attachments and form a dense network around mesenteric blood vessels and their tributaries in the gut wall. These various ending maintain a steady flow of afferent traffic to the CNS relating information on activity both within and outside the gut wall. The sensory information is conveyed to the CNS by vagal and spinal pathways to the brainstem and spinal cord, from where information is processed and relayed to higher brains areas (Joshi and Gebhart, 2000). The vagal and spinal pathways predominantly transmit different aspects of sensory information. Vagal neurones generally process physiological information e.g. the nature and composition of the luminal contents, the presence and amplitude of ongoing motor activity of the gut. In contrast, spinal neurones process mainly pathophysiological information e.g. potentially noxious mechanical or chemical stimuli arising through tissue injury, ischaemia and inflammation. However, vagal and spinal pathways share a number of functional properties and there is evidence for interplay between these two pathways in reflex control, in behaviours and in sensation. 1.1. Morphological features of sensory endings in the gut wall Afferent nerves largely follow the same pathway to the CNS as the efferent fibres leaving it in parasympathetic and sympathetic autonomic nerves. The various populations of afferents are therefore defines as vagal, pelvic and splanchnic afferents. Afferent fibres outnumber efferent fibres by about 10 : 1 in the vagus nerves, so one might consider this a predominantly sensory nerve. Vagal cell bodies lie in the nodose ganglia and project centrally to the nucleus tractus solitarius in the brainstem. The density of spinal afferents is more scant with < 7% of sensory cell bodies in the dorsal root ganglia (DRG) projecting to the viscera. The peripheral terminals of vagal and spinal afferents have been localised within the GI wall using fibre tracing techniques (Berthoud and Neuhuber, 2000). Both vagal and spinal nerves innervate the entire length gastrointestinal tract, but vagal afferents are more prevalent in the proximal gut and spinal, particularly pelvic afferents, predominate in the distal gut. The location of different afferent endings in the mucosa, muscle layers, and in the serosal and mesenteric attachments is consistent with their ability to respond to stimuli acting at these different sites within the GI wall. Although some afferent terminals are specialised in the way their terminals arbourise in the gut wall there are non of the specialized endings as found in the skin but instead the GI sensory nerves are bare nerve endings. Thus the differential sensitivity of afferents arises from both their location in the gut wall and the receptors and ion channels that they express on their terminals. 1.1.1. Vagal afferents Different populations of vagal afferent neurones terminate in either the mucosa or the muscle. Afferent endings in
77
the mucosa are closely associated with the lamina propria in proximity to the mucosal epithelium, but are never exposed directly to the contents of the lumen. Vagal afferents have been visualised close to specialised epithelial cells such as enterochromaffin cells and mast cells which may be the actual sensory transducers that monitor the physical and chemical nature of luminal contents (Phillips and Powley, 2000). Vagal afferent endings in the muscle can be classified into two main types: intramuscular arrays (IMA) and intraganglionic laminar endings (IGLE). IMAs are distributed within the muscle sheets, especially in the longitudinal muscle, parallel to the long axes of muscle fibres (Berthoud and Neuhuber, 2000). They are interposed between bundles of smooth muscle fibres, but they also course on, and form appositions with, intramuscular interstitial cells of Cajal. In contrast IGLEs are basket-like structures surrounding myenteric ganglia. As IGLEs are located between the circular and longitudinal muscle layers, they are exposed to shearing forces generated during muscle stretch or contraction. Evidence showing they respond to such mechanical forces has been elaborated recently by mapping the receptor fields of vagal afferent endings in the oesophagus and showing morphologically that these Fhot-spots_ to mechanical probing corresponds to the location of IGLEs (Zagorodnyuk and Brookes, 2000). IGLEs are the primary candidates for conveying mechanosensory information relevant to distension and contraction of the bowel wall. However, the proximity between IGLEs and myenteric ganglion neurones raises the intriguing possibility that they may also detect chemicals in the synaptic neuropil and as such provide a morphological substrate for ‘‘crosstalk’’ between the enteric nervous system and extrinsic afferents. 1.1.2. Spinal afferents Spinal afferent nerve terminals are located primarily in the serosa and muscle layers of the gut. Spinal afferents are largely unmyelinated and have multiple branching punctate endings that correspond to multiple receptive fields. Their location and response characteristics suggest that spinal afferents respond to distortion of the viscera (Gebhart, 2000). Spinal afferents also innervate the mucosa and are implicated in mucosal protection (Holzer, 1998). This protective role arises because spinal afferents have collateral branches that supply blood vessels and innervate the enteric ganglia. Activation of an afferent terminal causes an action potential to be propagated centrally, but action potentials can also propagate down axon collaterals and stimulate the release of neurotransmitters in a local axon reflex. In this manner, the release of transmitters from their varicose nerve terminals can modulate blood flow and enteric reflex pathways. The main neurotransmitters present in spinal afferents are calcitonin gene related peptide (CGRP) and substance P (SP) (Sternini et al., 1992). Both of these peptides are implicated in GI neurogenic inflammation and
78
D. Grundy / Autonomic Neuroscience: Basic and Clinical 125 (2006) 76 – 80
so their release via axon reflexes may be involved in inflammatory responses (Keates et al., 1998).
2. Mechanosensitivity Both vagal and spinal afferents in the muscle and mesenteric connections are mechanosensitive. In response to distension vagal afferents have a lower threshold for activation than spinal afferents (Sengupta et al., 1990). Little is known about the precise elements within GI afferent endings that are activated by mechanical activity and transduce this stimulus into action potential firing. In other sensory systems, mechanosensitive ion channels have been described whereby a mechanical stress on a receptor molecule in the neuronal cell membrane distorts it in such a way to cause ion channel activation and action potential firing. It can only be speculated that mechanosensitive ion channels are expressed by vagal and spinal mechanosensitive afferent neurones. The differing thresholds for activation may be explained by the expression of different ion channels and/or the different locations of vagal and spinal afferent terminals. 2.1. Vagal afferents Direct electrophysiological recordings of vagal afferent traffic en route to the CNS demonstrate that vagal afferents have low thresholds of activation and responses plateau within physiological levels of distension. Vagal afferents are described as Fin-series tension receptors_ by analogy with Golgi tendon organs which are sensitive to tension developed in the gut wall by stretch during distension and during contraction. As discussed above IGLEs have been proposed to be the Fin-series tension receptor_, although situated in the myenteric ganglia they may do this from an in parallel location, presumably since contraction of either muscle layer will generate shear forces in the interposed ganglia. 2.2. Spinal afferents Spinal afferents generally have higher thresholds for activation and in many cases would only be activated by levels of distension or contraction that would be considered noxious (Sengupta et al., 1990). However, some spinal afferents, particular running in the pelvic nerves, respond to lower levels of stimulation within the physiological range and indeed IGLEs have been described in the rectum (Lynn et al., 2003). Other afferents also have low thresholds for activation but because these endings continue to encode into the noxious range these afferents are said to have a Fwide dynamic range_ by analogy with second order neurons in the spinal cord that also respond to low and high intensity stimulation. These afferent endings therefore would encode both physiological and noxious levels of stimulation. These
different stimulus– response profiles are consistent with the view that it is high threshold and wide-dynamic range afferents that provide an intensity-related code that conveys noxious information and is responsible for mediating pain. Mechanosensitive thresholds and the relationship between stimulus and response are not fixed but can be influenced by a wide range of chemical mediators released as a consequence of injury, ischaemia and inflammation. A wide range of chemical mediators may interact in a potentiating way to modulate the sensitivity of spinal afferent endings, reducing the threshold for activation to cause hypersensitivity (Kirkup et al., 2001). Previously insensitive afferents have also been shown to develop mechanosensitivity during inflammation, these are referred to as ‘‘sleeping’’ or ‘‘silent’’ nociceptors.
3. Peripheral sensitization The threshold for activation of sensory neurones is not fixed but can be modulated by a number of different mediators. Such plasticity in afferent neurones can lead to sensitisation and hyperalgesia. Sensitisation is characterised by a decrease in threshold for firing of neurones and an elevated response to suprathreshold stimuli. This sensitisation of neuronal endings can lead to hyperalgesia, which is the hallmark of irritable bowel syndrome (IBS). At a molecular level, sensitisation can occur through at least three different mechanisms: i) a change in receptor sensitivity or ion channel gating; ii) an upregulation in the number of receptors and channels expressed on afferent endings; iii) the expression of new receptors and channels (Kirkup et al., 2001). Some receptors can be activated and potentiated by more than one stimuli e.g. TRPV1 receptor activity is modulated by capsaicin, pH, cannabinoids and heat. Other receptors share second messenger pathways and hence can be modulated indirectly by feedback from different mediators (Wood, 2001). A variety of factors can trigger sensitising responses. Sensory modulation occurs during inflammation, injury, infection or ischaemia, which involves a plethora of different cell types e.g. platelets, leukocytes, lymphocytes, macrophage, mast cells. Each of these cell types can release a host of different neuromodulators. Regardless of the precise mechanism, the normal sensitivity of neurones is altered. This neuronal sensitisation is often a necessary and appropriate response in the short term. However, if sensitisation persists after the challenge to the GI tract has been eliminated, this can lead to inappropriate sensory processing and long-term hyperalgesia. One of the hypotheses for IBS is that a transient gut infection evokes an inappropriate long-lasting gut hypersensitivity (Barbara et al., 2002). Both spinal and vagal afferents can become sensitised, although it is primarily spinal afferents that have been implicated in hyperalgesia. However, there is some evidence
D. Grundy / Autonomic Neuroscience: Basic and Clinical 125 (2006) 76 – 80
that vagal afferents may play a role in hyperalgesia as some vagal afferents project into the cervical region of the spinal cord and these fibres may be involved in transmitting sensation to thalamic nuclei. In addition, vagal afferents have been shown to facilitate nociceptive transmission and these pathways are believed to be implicated in the hyperalgesia that arises as a consequence of illness behaviour triggered by some cytokines, especially interleukin-1h (Watkins and Maier, 2000).
79
of long chain fatty acids is dependent upon chylomicron formation and it is not clear the extent to which CCK release, and hence neuronal activation, is reliant upon chylomicron formation. Short chain fatty acids directly activate vagal afferents via a post-absorptive CCK-independent mechanism (Lal et al., 2001). CCK is also implicated in sensing luminal protein digestion products (Eastwood et al., 1998). 4.1. Protons
4. Sensing luminal contents Afferents nerves terminating near to the mucosa are in a position to monitor the composition of the luminal contents. As afferents do not project directly into the lumen, their activation depends on an intermediary step that could either be post-absorptive i.e. neuronal activation by absorbed substances from the lumen, or pre-absorptive i.e. neuronal activation by a secondary substance released from within the mucosal epithelium. In this respect a role for both CCK and 5-HT released from enteroendocrine cells and acting as paracrine agents on the terminals of vagal afferents has been implicated in responses to a number of luminal signals (Raybould, 1999). Enterochromaffin cells (EC cells) release 5-HT and are strategically positioned in the intestinal mucosa to ‘‘taste’’ luminal contents and release their mediators across the basolateral membrane in order to generate action potentials in the afferent nerve endings within the lamina propria (Gershon, 1999). 5-HT liberated from EC cells acts directly on vagal extrinsic afferent nerves in the mucosa through activation of 5-HT3 receptors expressed on the nerve terminal. The physiological stimuli for the release of 5-HT from EC cells may be both chemical and mechanical, suggesting a role in both chemotransduction and mechanotransduction. A large body of data implicates EC cells in the detection of bacterial enterotoxins e.g. cholera toxin. These toxins trigger release of 5-HT from EC cells, to bring about an orchestrated response to dilute and subsequently eliminate the pathogenic material from the body through diarrhoea and vomiting and preclude further consumption of the potentially harmful material by causing nausea. 5-hydroxytryptamine release from enterochromaffin (EC) cells is also implicated in nutrient signalling particularly in relation to luminal sugars (Raybould, 1999). In contrast, lipids activate vagal afferents through a mixture of both pre- and post-absorptive mechanisms, and this varies according to the type of fatty acids present in the lumen. Long chain fatty acids indirectly activate vagal afferents through a cholecystokinin (CCK) dependent mechanism. CCK is released from specialised enteroendocrine cells in the intestinal epithelia and activates vagal afferents, probably in a paracrine manner acting on CCK(A) receptors on vagal afferent terminals. Absorption
Responses to change in luminal pH are observed in vagal afferents. It is speculated that this sensitivity is due to the activation of receptors that can be modulated directly by protons. A number of vagal afferents supplying the gut are sensitive to capsaicin, the response of which is mediated by the TRPV1 receptor. This receptor is also known to be sensitive to pH, and may therefore provide the molecular transducer of changes in pH (Rong et al., 2004). However, other proton sensing channels exist including members of the ASIC family and P2X receptors.
5. Conclusions Afferent fibres convey a vast amount of sensory information to the brainstem and spinal cord, but the nature of this information differs between vagal and spinal pathways. Vagal afferents convey predominantly physiological information whilst spinal afferents additionally encode noxious events. However, there is evidence for overlap and interplay between the sensitivity and functions of vagal and spinal afferents. Spinal nociceptors are markedly influenced by peripherally acting chemicals which are released during inflammation and injury and they are thought to trigger the processes leading to sensitisation and increased nociceptive activity. Other chemicals act in a more selective way to activate vagal afferents and are implicated in nutrient signalling from the GI tract. References Barbara, G., De Giorgio, R., Stanghellini, V., Cremon, C., Corinaldesi, R., 2002. A role for inflammation in irritable bowel syndrome? Gut 51 (Suppl 1), i41 – i44. Berthoud, B., Neuhuber, W.L., 2000. Functional and chemical anatomy of the afferent vagal system. Auton. Neurosci. 85, 1 – 17. Eastwood, C., Maubach, K., Kirkup, A.J., et al., 1998. The role of endogenous cholecystokinin in the sensory transduction of luminal nutrient signals in the rat jejunum. Neurosci. Lett. 254, 145 – 148. Gebhart, G.F., 2000. Pathobiology of visceral pain: molecular mechanisms and therapeutic implications IV. Visceral afferent contributions the pathobiology of visceral pain. Am. J. Physiol.: Gastrointest. Liver Physiol. 278, G834 – G838. Gershon, M.D., 1999. Review article: roles played by 5-hydroxytryptamine in the physiology of the bowel. Aliment. Pharmacol. Ther. 13, 15 – 30. Holzer, P., 1998. Neural emergency system in the stomach. Gastroenterology 114, 823 – 839.
80
D. Grundy / Autonomic Neuroscience: Basic and Clinical 125 (2006) 76 – 80
Joshi, S.K., Gebhart, G.F., 2000. Visceral pain. Curr. Rev. Pain 4, 499 – 506. Keates, A.C., Castagliuolo, I., Qiu, B., et al., 1998. CGRP upregulation in dorsal root ganglia and ileal mucosa during Clostridium difficile toxin A-induced enteritis. Am. J. Physiol. 274, G196 – G202. Kirkup, A.J., Brunsden, A.M., Grundy, D.I., 2001. Receptors on visceral afferents. Am. J. Physiol.: Gastrointest. Liver Physiol. 280, G7878 – G7894. Lal, S., Kirkup, A.J., Brunsden, A., Thompson, D.G., Grundy, D., 2001. Vagal afferent responses to fatty acids of different chain length in the rat. Am. J. Physiol.: Gastrointest. Liver Physiol. 281, G907 – G915. Lynn, P.A., Olsson, C., Zagorodnyuk, V., Costa, M., Brookes, S.J., 2003. Rectal intraganglionic laminar endings are transduction sites of extrinsic mechanoreceptors in the guinea pig rectum. Gastroenterology 125, 786 – 794. Phillips, R.J., Powley, T.L., 2000. Tension and stretch receptors in gastrointestinal smooth muscle: re-evaluating vagal mechanoreceptor electrophysiology. Brain Res. Rev. 34, 1 – 26. Raybould, H.E., 1999. Nutrient tasting and signaling mechanisms in the gut. I. Sensing of lipid by the intestinal mucosa. Am. J. Physiol. 277, G751 – G755.
Rong, W., Hillsley, K., Davis, J.B., Hicks, G.A., Winchester, W.J., Grundy, D., 2004. Jejunal afferent nerve sensitivity in wild type and TRPV1 knockout mice. J. Physiol. Sengupta, J.N., Saha, J.K., Goyal, R.K., 1990. Stimulus – response function studies of esophageal mechanosensitive nociceptors in sympathetic afferents of opossum. J. Neurophysiol. 64, 796 – 812. Sternini, C., De Giorgio, R., Furness, J.B., 1992. Calcitonin gene-related peptide neurons innervating the cat digestive system. Regul. Pept. 42, 15 – 26. Watkins, L.R., Maier, S.F., 2000. The pain of being sick: implications of immune-to-brain communication for understanding pain. Annu. Rev. Psychol. 51, 29 – 57. Wood, J.N., 2001. Pathobiology of visceral pain: molecular mechanisms and therapeutic implications II. Genetic approaches to pain therapy. Am. J. Physiol.: Gastrointest. Liver Physiol. 278, G507 – G512. Zagorodnyuk, V.P., Brookes, S.J., 2000. Transduction sites of vagal mechanoreceptors in the guinea pig esophagus. J. Neurosci. 20, 6249 – 6255.