17.2.1

Vagal Afferents

The cell bodies of vagal afferents reside in either the bilateral nodose ganglia or the bilateral jugular ganglia, located at the base of the skull ( Fig. 17.1 ). The nodose and jugular ganglia are embryologically distinct, with nodose neurons arising from the epibranchial placodes, while jugular neurons derive from neural crest cells of the postotic hindbrain, more akin to spinal afferents. The central projections of gastric vagal afferents have their terminals predominantly within the subnucleus gelatinosus of the nucleus of the solitary tract (NTS). Within the NTS, esophageal afferents project to the subnucleus centralis, whereas afferents innervating the gastric mucosa and muscle project to overlapping regions of the NTS. Additionally, there is a small proportion of mucosal afferents that project into the medial subnucleus. Overall, there are diverse projections from the NTS into a large range of nuclei throughout the brain, which are involved with behavioral, cognitive, and autonomic processing. The axons of vagal afferents travel via the paired dorsal and ventral vagal nerve trunks before organ-specific branches project into thoracic and abdominal organs, including the esophagus, stomach, and small intestine. The density of vagal afferent endings is greatest within the stomach and esophagus, reduces in the small intestine, and are absent in the distal colon and rectum. Vagal afferents innervate all layers of the wall of the upper gastrointestinal tract. They include various types of mucosal endings, intraganglionic laminar endings (IGLEs), and intramuscular arrays (IMAs), which are discussed in detail below. In keeping with their distribution and function, vagal afferents are generally associated with signaling physiological events in the upper gastrointestinal tract, such as fullness and satiety.

Schematic representation of the vagal and spinal innervation of the gastrointestinal tract. CG , celiac ganglia; GSN , greater splanchnic nerve; HGN , hypogastric nerve; ICG , intermediate cervical ganglia; IMG , inferior mesenteric ganglia; IMN , intermesenteric nerve; LCN , lumbar colonic nerves; LSN , lumbar splanchnic nerve; MPG , major pelvic ganglia; NTS , nuclei tractussolitarii; PN , pelvic nerve; SCG , superior cervical ganglia; SG , stellate ganglia; SMG , superior mesenteric ganglia; T , thoracic.

17.2.2

Spinal Afferents

The cell bodies of spinal afferents are located within the cervical, thoracic, lumbar, and sacral DRG, located bilaterally adjacent to the spinal cord. The precise regional location of the DRG depends upon the gastrointestinal region. The spinal afferent innervation of the gastroesophageal region spans from the cervical to the thoracic and upper lumbar segment of the spinal cord via axons traveling through the thoracic spinal and greater splanchnic nerves ( Fig. 17.1 ). The spinal afferent innervation of the small intestine spans from the thoracic to the lumbar spinal cord via the greater splanchnic nerves ( Fig. 17.1 ). The proximal colon receives spinal innervation from the thoracic to the lumbar spinal cord via the lumbar splanchnic nerve, while the distal colon receives dual innervation from the lumbar splanchnic nerve and sacral pelvic nerves. The rectum also receives innervation from the lumbosacral spinal cord via the sacral pelvic nerves ( Fig. 17.1 ). Spinal afferents innervate all layers of the wall of the gastrointestinal tract. They include various types of mucosal endings, IGLEs, IMAs, and in particular afferents that wrap around blood vessels, which are discussed in detail below.

The central terminals of spinal afferents project into discrete laminae of the superficial dorsal horn and make synaptic contacts with second-order neurons within the dorsal horn of the spinal cord. Spinal afferents supplying the upper gastrointestinal tract, such as the esophagus and small intestine, are distributed along the cervical and thoracic spinal cord (C1-T12), with peaks of innervation present in C2-C6, T2-T4, and T8-T12. Lumbar splanchnic fibers innervating the colon terminate within the thoracolumbar spinal cord (T10-L1), whereas sacral pelvic afferents innervating the distal colon and rectum terminate within the lumbosacral spinal cord (L6-S2). The central terminals of afferents innervating the distal colon are localized within the thoracolumbar spinal cord. They project into the superficial dorsal horn lamina I and transverse along mid and lateral collateral pathways of Lissauer into lamina V. The projection pattern of pelvic nerves in the lumbosacral spinal cord differs from that observed in the thoracolumbar spinal cord for the splanchnic nerves, with studies showing terminals present in the lumbosacral superficial dorsal horn lamina I and more prominently in Lissauer’s lateral collateral tracts projecting into laminae V and VI and medial tracts projecting into the dorsal commissure and spinal dorsal columns. The types of dorsal horn neurons activated by visceral stimuli differ between spinal segments. Within the thoracolumbar spinal cord, lamina I projection neurons and interneurons are activated by visceral stimuli. In the lumbosacral spinal cord, a greater number of interneurons and dorsal column neurons are responsive to colorectal distension relative to that observed in the thoracolumbar spinal cord. From the spinal cord, dorsal horn lamina I projection neurons relay visceral specific sensory information to higher central nervous system (CNS) centers via the ascending spinothalamic tracts. In addition, lamina I spinoparabrachial projection neurons relay visceral input into the parabrachial nucleus (PBn), which transmits to higher-order centers for nociceptive processing. The dorsal column is also important in the transmission of visceral pain, with postsynaptic spinal dorsal column neurons activated by visceral stimuli and transferring input into the ventral posterolateral (VPL) nucleus of the thalamus. Dorsal column neurons responsive to visceral stimuli are prominent in the lumbosacral spinal cord, located in laminae III-IV, and around the central canal area. Given their distribution throughout the gastrointestinal tract and their function, spinal afferents are generally associated with higher-threshold sensations such as discomfort, bloating, urgency, and pain.

17.12.1

Protease-Activated Receptors

Protease-activated receptors (PARs) are a family of four (PAR _1–4 ) G-protein coupled (GPCR) receptors that typically are activated by proteolytic cleavage of their N-terminal extracellular domain. This releases a new motif on the N-terminal tail that binds to the receptor in an agonist-like fashion. However, many proteases, including trypsins and catepsins, can cleave PARs at divergent sites resulting in biased agonism that can lead to unique pathophysiological outcomes. Colon-innervating DRG neurons express PAR ₂ , which when activated affects visceral sensory signaling. Elevated luminal serine protease activity, as well as increased tryptase release, occurs in colonic samples from IBS patients. When administered into the colon of healthy mice, these proteases induced visceral hyperalgesia (increased pain) and allodynia (pain at innocuous stimulus intensities) via a PAR ₂ -dependent mechanism. This hypersensitivity can be prevented by a PAR ₂ antagonist, was absent in PAR ₂ knock-out ( ^−/− ) mice, and can be replicated by intracolonic PAR ₂ agonist administration. Such PAR ₂ -mediated effects are further corroborated by recent findings demonstrating that supernatants derived from IBS-D, and IBS-C biopsies increased the excitability of mouse colonic sensory DRG neurons from PAR ₂ wild-type ( ^+/+ ) mice, but not from PAR ₂ ^−/− mice. More recently, in IBS patients, it has been shown that the intestinal epithelium produces and releases the active protease trypsin-3, which is able to signal to human submucosal enteric neurons and mouse sensory neurons, and to induce visceral hypersensitivity in vivo, all by a PAR ₂ -dependent mechanism. In animal models of IBD, there is evidence for increased luminal levels of PAR ₂ -agonists, particularly cathepsin-S, resulting in increased PAR ₂ -mediated activation of spinal nociceptors and visceral hypersensitivity.

17.12.2

Histamine Receptors

Histamine, the predominant mediator released from mast cells, acts via four related G-protein coupled receptors (H ₁ , H ₂ , H ₃ , and H ₄ ). Histamine is a powerful modulator of sensory afferent activity at the level of the gut wall and contributes to visceral hypersensitivity. In vivo, selective histamine H ₁ receptor (H ₁ R) antagonists reduce visceral hypersensitivity in animal models of IBS and IBD. Correspondingly, IBS supernatant-induced excitation of rat jejunal afferents and DRG neurons is reduced by a H ₁ R antagonist. This indicates that the pro-nociceptive effect of histamine is, at least in part, mediated by H ₁ R expressed on sensory afferents. H ₁ R has been demonstrated to couple with both TRPV1 and TRPV4 on gut afferents, which enhances afferent sensory signaling to the spinal cord. Activation of H ₁ R is clinically relevant in IBS, as the selective H ₁ R antagonist, ebastine, reduced abdominal pain, and improved quality of life in IBS patients in a randomized, double-blind, placebo-controlled, and single-center clinical trial. In addition to H ₁ R, there is evidence for histamine H ₄ receptors (H ₄ R) modulation of visceral sensation. The H ₄ R agonist 4-methylhistamine excites human submucous plexus neurons, an effect that is inhibited by the selective H ₄ R antagonist JNJ7777120. In a postinflammatory rat model of IBS, this H ₄ R antagonist dose-dependently reduced visceral hypersensitivity. Interestingly, there appears to be a functional interaction between H ₁ R and H ₄ R receptor subtypes, as their simultaneous administration in vivo potentiates their antinociceptive actions.

17.12.3

Serotonin

The majority of 5-HT in the body is found in the gastrointestinal tract, primarily contained within enterochromaffin cells and is released by meals, toxins, and chemotherapeutic agents. In the upper gut, 5-HT release activates vagal mucosal afferent endings, which underlies chemotherapy-induced nausea and vomiting. More recently, it has been demonstrated that microbial metabolites, as well as norepinephrine, result in the release of 5-HT from enterochromaffin cells in the intestine and colon that activates afferents innervating the mucosa via a 5-HT ₃ receptor mechanism ( Fig. 17.8 ). These findings demonstrate that enterochromaffin cells are gut chemosensors that couple to sensory neural pathways.

Enterochromaffin cells modulate 5HT ₃ R-expressing colonic afferents to cause mechanical hypersensitivity. (A) Norepinephrine (NE) causes mechanical hypersensitivity of pelvic colonic mucosal afferents, effects that are blocked by the TRPC4 antagonist ML205 and by the 5-HT ₃ antagonist alosetron. (B) Similarly, isovalerate (IVL), a metabolite produced by the gut microbiome, causes mechanical hypersensitivity of pelvic colonic mucosal afferents, which is also blocked by alosetron. (C) Neither NE nor IVL directly activate isolated colon innervating DRG neurons, whereas the 5-HT ₃ agonist does, suggesting key mechanisms in the signaling pathway. (D) Schematic diagram showing the key signaling pathways between enterochromaffin cells and colonic afferents.

((A–D) Modified from Bellono NW, Bayrer JR, Leitch DB, Castro J, Zhang C, O’Donnell TA, et al. Enterochromaffin cells are gut chemosensors that couple to sensory neural pathways. Cell 2017; 170 :185–98.e16, with permission.)

Mucosal barrier function is crucial for the overall function of the gastrointestinal tract; and is disturbed during inflammation associated with IBD, while increased epithelial permeability is evident in the small intestine and colon of IBS patients. These changes allow bacteria to access the interstitial compartment of the gut, resulting in luminally applied LPS from Salmonella typhimurium activating jejunal afferents. These afferents also display an increased sensitivity to 5-HT ₃ receptor agonists.

In the rat, 56% of high-threshold splanchnic colonic afferents respond to 5-HT, via both 5-HT ₃ and non-5-HT ₃ receptors, which correlates with the percentage of thoracolumbar DRG cell bodies retrogradely labeled from the colon that display 5-HT ₃ receptors. In the rat, 5-HT ₁ , 5-HT ₂ , and 5-HT ₃ receptor subtypes have been demonstrated to modulate responses to noxious colorectal distension, and serotonergic activation of visceral sensory neurons may increase their sensitivity to other sensory modalities. Following intestinal inflammation, the contribution of 5-HT ₃ receptors is decreased, yet the responsiveness to 5-HT is increased, indicating that other 5-HT receptor subtypes play an increased role in disease states. Correspondingly, later studies demonstrated that peripherally administered 5-HT ₄ receptor agonists have antinociceptive actions, while also having prokinetic actions and reducing colonic inflammation. In terms of translation to humans, metabolism of 5-HT appears disrupted in both IBS and IBD patients. The involvement of 5-HT in IBS patient symptomatology is implicated by increased number of enterochromaffin cells, increased mast cell populations, increased postprandial 5-HT release, and a decrease in symptoms using 5-HT ₃ receptor or 5-HT ₄ receptor antagonists.

17.12.4

Voltage-Gated Sodium (Na _V ) Channels

Na _V channels are large complex structures composed of a primary α subunit as the pore-forming region along with one or more auxiliary β subunits. Na _V channels are essential for action potential generation, propagation and the overall control of membrane excitability. Nine different mammalian Na _V channels, Na _V 1.1-Na _V 1.9, have been identified and can be divided into two groups based on their sensitivity to tetrodotoxin (sensitive: Na _V 1.1, Na _V 1.2, Na _V 1.3, Na _V 1.4, Na _V 1.6, Na _V 1.7; resistant: Na _V 1.5, Na _V 1.8, and Na _V 1.9). From a translational perspective, intrarectal administration of lidocaine (a pan-Na _V channel blocker) in IBS patients reduces rectal sensitivity and abdominal pain. These findings suggest that Na _V channels and the activation of peripheral afferent endings in the colon play key roles in the CVP experienced by IBS patients. It also shows that treatments applied in the periphery can reduce abdominal pain in IBS patients, independent of CNS actions.

Recent studies show diverse roles for specific Na _V channels in sensation from the gastrointestinal tract. Functional studies of colon innervating afferents reveal that Na _V 1.1 plays a crucial role in signaling mechanical pain from the colon. Application of the highly selective Na _V 1.1-agonist, δ-theraphotoxin-Hm1a (Hm1a), enhances mechanically evoked firing in a subpopulation of high-threshold colonic nociceptors, which is blocked by incubation with the Na _V 1.1/Na _V 1.3 antagonist ICA-121431 ( Fig. 17.9 ). Furthermore, Hm1a also induces hyperexcitability of isolated colon-innervating DRG neurons from healthy control mice. Importantly, colon-innervating DRG neurons isolated from mice with CVH show significantly enhanced responsiveness to Hm1a compared to healthy control mice, suggesting that Na _V 1.1 is essential for the development and maintenance of chronic visceral pain conditions ( Fig. 17.9 ). Accordingly, antagonism of Na _V 1.1 may be a future target for the treatment of disorders accompanied by chronic visceral pain originating from the colon.

Increased contribution of the voltage-gated sodium channel Na _V 1.1 to colonic afferents in a model of chronic visceral hypersensitivity (CVH). (A) The Na _V 1.1-specific agonist Hm1a (100 nM) increases the mechanosensitivity of high-threshold colonic nociceptors (serosal afferents) from a healthy control mouse. This effect is blocked by a Na _V 1.1 antagonist. (B) Whole-cell patch clamp recordings showing that Hm1a increases the excitability of a population of colon-innervating DRG neurons from healthy control mice. (C) Colonic nociceptors from CVH mice display increased baseline mechanosensitivity, which is further increased by Na _V 1.1 activation. (D) Colon innervating DRG neurons from CVH mice display increased sensitivity to the Na _V 1.1 agonist, Hm1a, resulting in dramatically increased action potential firing relative to baseline and healthy states. * P <.05, ** P <.01, *** P <.001.

((A–D) Modified from Osteen JD, Herzig V, Gilchrist J, Emrick JJ, Zhang C, Wang X, et al. Selective spider toxins reveal a role for the Nav1.1 channel in mechanical pain. Nature 2016; 534 :494–99, with permission.)

Na _V 1.5 in the circular smooth muscle contributes to normal intestinal motility through the modulation of slow-wave activity and muscle contractility. Ranolazine, a treatment for chronic angina, is able to inhibit Na _V 1.5 currents in human colonic smooth muscle cells, which is likely to be responsible for the constipation seen during long-term ranolazine treatment. Several loss-of-function mutations in SCN5A , the gene encoding Na _V 1.5 channels, have been reported in IBS patients, which may be associated with their altered motility and abdominal pain.

Biopsies from patients with rectal hypersensitivity show increased expression of Na _V 1.7 channels in rectal sensory fibers. However, deletion or pharmacological blockade of Na _V 1.7 in mice does not affect colonic afferents in healthy states. There is evidence for increased Na _V 1.8 expression in the DRG of mice in models of colitis, while Na _V 1.8 plays a key role in ciguatoxin (a pan-Na _V channel activator)-induced activation of colonic nociceptors and visceral pain. Na _V 1.9 is required for responses to intense mechanical stimulation, contributes to inflammatory mechanical hypersensitivity, and is essential for activation by noxious inflammatory mediators, such as ATP, PGE2, bradykinin, ATP, histamine, and 5-HT.

17.12.5

Acid Sensing Ion Channels

Acid sensing ion channels (ASICs) are members of the DEG/ENaC family of ion channels, which contribute to the detection of pH and mechanical stimuli via entry mainly of sodium ions through four related channels, ASIC1, ASIC2, ASIC3, and ASIC4. These ASICs form heteromultimers, and the composition of these heteromultimers differs depending on the tissue studied. In terms of the gut, ASIC1, 2, and 3 mRNA expression has been detected in vagal and spinal pathways but their abundance varies greatly between gut innervating afferents. ASIC1 has a greater expression in gastroesophageal innervating nodose ganglion neurons compared with colon innervating neurons, similar levels of ASIC2 are in both sets of neurons, while colonic neurons contain more ASIC3 than gastroesophageal neurons. As little as 30% of colonic neurons express ASIC1, while 47% express ASIC2 and 73% express ASIC3. Correspondingly, studies utilizing ASIC ^−/− mice have demonstrated widespread differences in the mechanosensory mechanisms between different pathways. Deletion of ASIC1a increases the mechanical sensitivity of splanchnic colonic and vagal gastroesophageal afferents ( Fig. 17.10 ).

Significant contribution of TRPA1 and TRPV4 to normal colonic high-threshold afferent mechanosensory function. Deletion of TRPA1 or TRPV4 markedly reduces the mechanosensitivity of high-threshold colonic splanchnic (A) mesenteric and (B) serosal afferents. Deletion of ASIC3 also reduces afferent mechanosensitivity, but the effects are more modest. In stark contrast, ASIC1a and ASIC2 deletion either increases mechanosensitivity or causes no effect, indicating at best a modulatory role for these channels. Overall, by far the greatest deficits are apparent in the TRPV4 ^−/− and TRPA1 ^−/− afferents.

(TRPV4 modified from Modified from Brierley SM, Page AJ, Hughes PA, Adam B, Liebregts T, Cooper NJ, et al. Selective role for TRPV4 ion channels in visceral sensory pathways. Gastroenterology 2008; 134 :2059–69, with permission. TRPA1 modified from Brierley SM, Hughes PA, Page AJ, Kwan KY, Martin CM, O’Donnell TA, et al. The ion channel TRPA1 is required for normal mechanosensation and is modulated by algesic stimuli. Gastroenterology 2009; 137 :2084–95.e3, with permission. ASIC1a, ASIC2, and ASIC3 modified from Page AJ, Brierley SM, Martin CM, Price MP, Symonds E, Butler R,et al. Different contributions of ASIC channels 1a, 2, and 3 in gastrointestinal mechanosensory function. Gut 2005; 54 :1408–15, with permission.)

Disruption of ASIC2 has varied effects: increasing mechanosensitivity in gastroesophageal mucosal endings, decreasing gastroesophageal tension receptors, increasing colonic serosal endings, and causing no change in colonic mesenteric endings ( Fig. 17.10 ). In ASIC3 ^−/− mice, gastroesophageal tension afferents have markedly reduced mechanosensitivity, while mucosal afferents are unchanged. In the colon, splanchnic serosal and mesenteric afferents display mechanosensory deficits, and in the pelvic nerve muscular/mucosal afferents have significantly reduced mechanosensory responses ( Fig. 17.10 ). Low-threshold pelvic mucosal and muscular afferents remained unchanged. Although the deficits in mechanosensitivity observed in the respective ASIC ^−/− mice are fairly modest, particularly compared with those documented for TRPV4 and TRPA1 ( Fig. 17.10 ), the effects on afferent mechanosensitivity translate to changes in the whole animal. Observations of altered gastric emptying and fecal output are found in ASIC1a ^−/− and ASIC2 ^−/− mice, respectively, while reductions in colonic mechanosensory function in ASIC3 ^−/− mice translate to significantly reduced VMR to colorectal balloon distension (CRD). In humans, increased ASIC3 expression is observed in Crohn’s disease intestine. Differences between the upper gut and lower gut afferents are also apparent in the sensitivity to the nonselective DEG/ENaC blocker benzamil. Splanchnic colonic afferent mechanosensitivity is virtually abolished by benzamil application, whereas gastroesophageal afferents are only marginally inhibited. Deletion of ASIC2 or ASIC3 significantly reduces the ability of benzamil to reduce splanchnic serosal afferent mechanosensitivity, while deletion of ASIC1a has no effect on benzamil sensitivity. ASIC3 also contributes to visceral peripheral sensitization as demonstrated by an intracolonic zymosan induced increase in VMR in control mice, which is lost in ASIC3 ^−/− mice. Deletion of ASIC3 also prevents gastritis-induced acid hyper-responsiveness of the stomach-brainstem axis. In terms of potential therapeutic options, although potent ASIC3 blockers are available, they need to be extremely selective to avoid serious side effects given that ASIC1a and ASIC2 are involved in CNS and baroreceptor function.

17.12.6

TRPV1

TRP channels constitute a major superfamily of nonselective ion channels that can be characterized as ankyrin repeat (TRPA), canonical (TRPC), melastatin related (TRPM), mucolipin (TRPML), polycystin (TRPP), and vanilloid binding (TRPV). TRPV1 is probably the best-known member of the TRP channel family as it is the archetypal somatic pain sensor and is expressed on small diameter primary afferent neurons (C-fibers) in DRG. Transient receptor potential channel vanilloid 1 (TRPV1) acts as an immediate sensory alarm in response to low pH, heat, and noxious agents such as capsaicin. Therefore, TRPV1 has been strongly implicated in somatosensory nociception and pain, including thermal nociception and inflammatory hyperalgesia and allodynia, and neuropathic pain. Because of this link to pain, TRPV1 is one of the most heavily studied channels in the gastrointestinal tract. However, compared with cutaneous afferents, the role of TRPV1 in the viscera is not as straightforward. This is because its expression is not necessarily indicative of sensory neuron nociceptor function. First, in mouse, TRPV1 expression is greater in colon innervating neurons than those innervating the skin, with 82% of thoracolumbar and 50%–60% of lumbosacral colonic DRG neurons displaying TRPV1-LI. Correspondingly, all classes of low-threshold and high-threshold afferents in vagal and spinal pathways respond in varying degrees to capsaicin, with a high-functional expression of TRPV1 in all colonic afferent subtypes. Capsaicin also causes mechanical desensitization in vitro in gastroesophageal, jejunal, and splanchnic colonic preparations. Finally, TRPV1 is also expressed in nonneuronal gut tissues, such as the rat esophageal epithelium where there is a TRPV1 splice variant, which is significantly different (60 vs 95 kDa) to the sensory DRG form of TRPV1.

Corresponding with these differences in expression, stark contrasts also exist between somatic and visceral afferents when investigating mechanosensory function in TRPV1 ^−/− mice. Deletion of TRPV1 has no effect on somatic mechanosensory function ; however, deficits are apparent in intestinal afferents, specifically in certain types of low-threshold distension sensitive gastroesophageal, jejunal, and pelvic colonic afferents. By contrast, high-threshold gut afferents are unaffected by TRPV1 deletion. Nevertheless, TRPV1 has been strongly implicated in contributing to visceral pain. The ability of TRPV1 to respond to low pH places it in a prime position to sense esophageal acid reflux, giving rise to the sensation of heartburn, while intraesophageal capsaicin instillation induces symptoms of epigastric burning. In animal models, intracolonic capsaicin administration causes pronounced visceral pain, while TRPV1 ^−/− mice also display decreased VMRs to CRD, suggesting an involvement of TRPV1 in signaling behavioral aspects of colonic pain. A key interaction in this process appears to be via TRPV1 and the G protein-coupled receptor kinase 6 (GRK6). The proinflammatory cytokine IL-1β sensitizes TRPV1, which can be prevented by overexpressing GRK6. Following colitis, TRPV1-induced behavioral pain responses are more pronounced in GRK6 ^−/− mice than in GRK6 ^+/+ mice, suggesting that GRK6 can regulate inflammation-induced sensitization hyperalgesia.

An often-overlooked function of TRPV1 activation is its ability to cause neurogenic inflammation via the local release of substance P and CGRP from sensory afferent endings. Mice lacking TRPV1 develop significantly less esophagitis after acid exposure compared with TRPV1 ^+/+ mice, while administration of a TRPV1 antagonist before or after intracolonic TNBS administration significantly reduces colitis severity. Similarly, in rats and mice, TRPV1 antagonists attenuate disease severity in DSS-induced colitis. However, diametrically opposed to these observations there are also reports of TRPV1 having a protective role in inflammatory models, with TRPV1 activation improving DSS-induced colitis in mice. Furthermore, a dinitrobenzene sulfonic acid model causes more severe colitis in TRPV1 ^−/− mice than TRPV1 ^+/+ mice. In terms of translation to humans, expression of TRPV1 is increased in colonic nerve fibers of IBD patients, and in patients with rectal hypersensitivity.

As discussed above, numerous inflammatory mediators can interact with TRPV1 to modulate their functional properties. In colon innervating DRG neurons, 5-HT can sensitize TRPV1 function. Specifically, 5-HT incubation increases capsaicin-induced currents and decreases the activation threshold temperature of TRPV1 from 42°C to 38°C. These effects can be mimicked by 5-HT ₂ and 5-HT ₄ receptor agonists, while this sensitization of TRPV1 can be reduced by selective 5-HT ₂ and 5-HT ₄ receptor antagonists, but not by a 5-HT ₃ receptor antagonist.

17.12.7

TRPV4

TRP channel vanilloid 4 (TRPV4) is the mammalian homolog of the Caenorhabditis elegans gene Osm-9, which is critical in osmotic and mechanical avoidance, suggesting that it functions to detect noxious osmotic and mechanical stimuli. In gut innervating sensory DRG neurons, TRPV4 expression is more restricted than that observed for TRPV1. TRPV4 is localized in 38% of gastroesophageal vagal neurons, 65%–76% of splanchnic colonic DRG neurons, and 58% of pelvic colonic DRG neurons. Although TRPV4 expression is more restricted, it is actually enriched within colonic DRG neurons, with its expression 20-fold greater in splanchnic colonic DRG neurons than in whole DRG. In the periphery, TRPV4 protein colocalizes with CGRP in colonic nerve fibers in the serosal and mesenteric attachment. By contrast, TRPV4 expression is scarce in intramuscular or mucosal nerve fibers.

Consistent with the lack of TRPV4 expression in gastroesophageal vagal neurons deletion of TRPV4 has absolutely no effect on vagal afferent function. However, in colonic afferents, where TRPV4 is enriched, mechanosensory responses are dramatically reduced in TRPV4 ^−/− mice. This is the case for both splanchnic serosal and mesenteric afferents. TRPV4 deletion also increases the mechanosensory thresholds of these afferents, suggesting that TRPV4, like TRPA1 (discussed below), contributes to setting mechanical activation thresholds. Furthermore, pelvic serosal afferents display similar deficits in mechanosensory responses and thresholds to those seen in the splanchnic pathway ( Fig. 17.10 ). Consistent with the case in the vagal pathway, low-threshold pelvic afferents are completely unaffected by TRPV4 deletion. Taken together, these data indicate that TRPV4 makes a specific and major contribution to high-threshold colonic afferent mechanosensory function. The extent of these deficits in mechanosensitivity are some of the largest reported in any afferent population, somatic or visceral. These changes in colonic neuron function also translate to decreased VMR to CRD in TRPV4 ^−/− mice or in mice with downregulated TRPV4 expression (via siRNA intrathecal injection), the extent of which is particularly apparent at higher distension pressures.

The endogenous TRPV4 agonist 5,6-EET (an arachidonic acid metabolite) increased the mechanosensory responses of high-threshold colonic afferents in TRPV4 ^+/+ mice, but not in TRPV4 ^−/− mice. Another TRPV4 agonist, 4α-PDD caused significant TRPV4-mediated calcium influx in isolated colonic DRG neurons. Intracolonic administration of 4α-PDD in TRPV4 ^+/+ mice increased neuronal c-fos expression in the lumbosacral spinal cord and caused dose-dependent visceral hypersensitivity. Therefore, TRPV4’s role is not only linked to visceral mechanosensation but also to visceral hypersensitivity.

TRPV4 can be sensitized by a series of inflammatory mediators both in vivo and in vitro. Preexposure of colonic DRG neurons to histamine or 5-HT increased Ca ^{2 +} responses induced by 4αPDD via various second messenger pathways. 5-HT or histamine also enhances TRPV4 expression at the plasma membrane. Downregulation of TRPV4 via siRNA intrathecal injection significantly reduces the hypersensitivity induced by 5-HT or histamine. Although PAR ₂ interacts with TRPV1 and TRPA1 (see above), there is evidence for a strong interaction with TRPV4, which underlies visceral hypersensitivity. TRPV4 and PAR ₂ are anatomically and functionally coexpressed in a large percentage of colon innervating neurons. In isolated colonic DRG neurons, the PAR ₂ -activating peptide, PAR ₂ -AP, sensitizes TRPV4-mediated currents, while activating splanchnic serosal colonic afferent fibers by a TRPV4-dependent mechanism. Intracolonic administration of PAR ₂ -AP also causes enhanced VMR to CRD at higher distending pressures (> 40 mmHg) measured 6 h and 24 h after administration. This is not observed in TRPV4 ^−/− mice at either time point, suggesting that TRPV4 mediates both the acute and delayed hyperalgesia induced by PAR ₂ activation. Taken together, these data suggest that TRPV4 is a common mechanism to several known mediators of visceral hypersensitivity, including histamine, 5-HT, and PAR ₂ .

At face value, TRPV4 appears to be an ideal target for the treatment of visceral pain. TRPV4 has a high specificity for high-threshold spinal colonic afferents with one of the largest deficits in mechanosensory function documented in TRPV4 ^−/− mice. There appears to be little to no role in low-threshold spinal afferents or in vagal afferent fibers. TRPV4 is concentrated in the outer layers of the colon and is association with blood vessels, in particular during active inflammation of the colon associated with Crohn’s disease and ulcerative colitis. Serine protease levels, 5-HT signaling, and histamine are all known to be increased in IBS patients and TRPV4 interacts with these mediators. Finally, IBS patient biopsy samples have been shown to contain increased levels of specific polyunsaturated fatty acid metabolites, which stimulate extrinsic sensory neurons from mice and generate visceral hypersensitivity via a TRPV4-dependent mechanism.

17.12.8

TRPA1

TRP channel ankyrin 1 (TRPA1) has been implicated as a mechanosensor and a major mediator of inflammatory pain. In gut innervating sensory neurons, TRPA1 expression is more restricted than that observed for TRPV1. In these neurons, TRPA1 is localized in both vagal and spinal afferents, with 55% of vagal gastroesophageal afferent neurons, 54% of splanchnic colonic, and 58% of pelvic colon innervating DRG neurons expressing TRPA1. In peripheral tissue, TRPA1 is located in mucosal, serosal, and mesenteric nerve fibers, indicating that TRPA1 is well placed to participate in the function of these afferent subtypes. Correspondingly, TRPA1 agonists (mustard oil or cinnamaldehyde) universally evoke mechanical sensitization of serosal and mesenteric high-threshold colonic afferents and pelvic mucosal afferents. These compounds also activate colon innervating DRG neurons in isolation. The use of TRPA1 ^−/− mice has allowed specific determination of the afferent fiber types that utilize TRPA1 to detect mechanical stimuli. These studies show striking deficits occurring in high-threshold colonic afferents, specifically splanchnic mesenteric afferents and splanchnic and pelvic serosal afferents ( Fig. 17.10 ). Furthermore, TRPA1 makes a modest contribution to low-threshold mucosal afferent mechanosensitivity in both vagal and pelvic pathways innervating the esophagus/stomach and colon, respectively. In contrast, TRPA1 does not contribute to the mechanosensitivity of vagal tension receptors or pelvic muscular and muscular/mucosal afferents. Interestingly, these are the classes of afferent that are affected by TRPV1 deletion. Another important finding is the observation that TRPA1 deletion increases the activation thresholds of serosal and mesenteric afferents to mechanical stimuli, suggesting that TRPA1 is important in setting mechanical activation thresholds. The predominant mechanosensory role for TRPA1 in high-threshold serosal and mesenteric afferents combined with the ability of TRPA1 agonists to increase the mechanical responsiveness of these afferents is suggestive of an involvement in pain. Supporting this assertion, TRPA1 ^−/− mice also display decreased VMR to high intensity (80 mmHg) CRD, but not lower distension pressures (15–60 mmHg). The observation that intracolonic administration of TRPA1 agonists enhances VMR to the higher distending pressures (45 and 60 mmHg) within 2 h of administration suggests that TRPA1 can modulate visceral mechanical hyperalgesia (discussed below). Further inference can also be deduced from the observation that the classes of TRPA1 ^−/− afferent displaying deficits are exactly the same classes of TRPA1 ^+/+ afferent displaying mechanical hypersensitivity after TNBS-induced colitis, namely, splanchnic mesenteric and serosal afferents and pelvic serosal and mucosal afferents.

Mustard oil has long been used as a visceral inflammatory model to provoke tissue damage and sensitize nociceptors. We now know that mustard oil (allyl-isothiocynate) is a TRPA1 agonist, and recent findings provide us with a more complete understanding of how this compound causes these effects in the viscera. TRPA1 agonists induce neurogenic inflammation via the release of substance P and CGRP, with increases in cytokines associated with macrophage and neutrophil activation and recruitment. However, the contribution of TRPA1 to different types of inflammation may not be as prevalent as that of TRPV1, with recent reports suggesting that the severity of colitis induced by TNBS is unaffected by TRPA1 deletion.

TRPA1 is also directly activated by irritants such as the endogenous aldehyde, 4-hydroxynonenal (4-HNE). Intracolonic administration of mustard oil or 4-HNE increases the VMR to CRD at the highest distending pressures (45 and 60 mmHg, respectively) within 2–3 h. Correspondingly, these agonists also cause neuronal activation, as indicated by induced c-fos expression, in the superficial laminae (I and II) of thoracolumbar (splanchnic) and lumbosacral (pelvic) spinal cord in TRPA1 ^+/+ but not TRPA1 ^−/− mice. Thus, intracolonic administration of TRPA1 agonists activates spinal nociceptors and causes visceral hyperalgesia. TRPA1 is also indirectly activated by inflammatory mediators such as bradykinin and PARs. TRPA1 mediates bradykinin-induced mechanical hypersensitivity in the guinea pig esophagus and the bradykinin-induced mechanical hypersensitivity observed in splanchnic serosal afferents. TRPA1 is also required for PAR ₂ -induced visceral hyperalgesia as the delayed visceral mechanical hypersensitivity to CRD caused by intracolonic administration of the PAR ₂ -activating peptide in TRPA1 ^+/+ mice is completely absent in TRPA1 ^−/− mice.