2.2.1. Occurrence of TRPV1 in the alimentary canal

The existence of TRPV1, the specific sensor for capsaicin, has been envisaged more than 30 years ago, given that most pharmacological effects of capsaicin have been attributed to a specific action oncapsaicin-sensitive primary afferent neurons (Jancsó et al., 1987; Holzer, 1991; Szallasi & Blumberg, 1999; Szolcsányi, 2004). The role of capsaicin-sensitive afferent neurons and, by indirect inference, the distribution of capsaicin receptors have been traced with two experimental paradigms: stimulation of afferent neurons by acute application of capsaicin, and pretreatment of animals or tissues with high doses of capsaicin (“capsaicin desensitization”) to defunctionalize afferent neurons via excess influx of Ca2+ following stimulation of TRPV1. Capsaicin-sensitive nerve cells may even degenerate if, e.g., a high dose of capsaicin is administered to newborn rats (Jancsó et al., 1977) or micromolar concentrations of capsaicin are added to dorsal root ganglion (DRG) cultures (Wood et al., 1988). Since the whole sensory neuron is defunctionalized by this approach, referral to a capsaicin-sensitive structure or function in its historical connotation does not necessarily point to an implication of TRPV1. In addition, capsaicin at higher concentrations can act on ion channels unrelated to TRPV1 and on cellular structures lacking TRPV1 (Holzer, 1991) such as ICCs (Choi et al., 2010). Treatment of adult rats with a neurotoxic dose of capsaicin reduces the number of nodose ganglion neurons that express TRPV1 and project to the stomach as observed 10 days post-treatment. By 67 days post-treatment, much of the vagal afferent innervation of the stomach is restored, mainly because of sprouting of capsaicin-insensitive nerve fibers (Ryu et al., 2010).

In the alimentary canal, TRPV1 occurs in extrinsic sensory neurons, intrinsic enteric neurons as well as epithelial and endocrine cells. The major cellular sources of TRPV1 in the rat, guinea-pig and murine alimentary canal are spinal and vagal primary afferent neurons (Patterson et al., 2003; Ward et al., 2003; Holzer, 2004; Horie et al., 2004; Kadowaki et al., 2004; Robinson et al., 2004; Schicho et al., 2004; Banerjee et al., 2007; Ryu et al., 2010; Zhao et al., 2010). Both in the DRG and nodose ganglion cells, TRPV1 is restricted to small and medium-sized somata that are known to give rise to unmyelinated (C-) and thinly myelinated (Aδ-) fibers (Caterina et al., 1997; Guo et al., 1999; Michael & Priestley, 1999; Banerjee et al., 2007; Tan et al., 2008, 2009). There are distinct regional differences in the relative proportion of sensory neurons that stain positive for TRPV1. Thus, TRPV1-immunoreactive fibers are considerably more prevalent in visceral than in somatic afferent neurons (Robinson et al., 2004; Brierley et al., 2005; Hwang et al., 2005; Christianson et al., 2006a). In fact, the majority of nodose ganglion neurons projecting to the rodent stomach and of DRG neurons projecting to the rodent gut express TRPV1 (Robinson et al., 2004; Schicho et al., 2004; Zhang et al., 2004; Brierley et al., 2005; Hwang et al., 2005; Christianson et al., 2006a, 2006b; Tan et al., 2008; Zhong et al., 2008), while only 32% of the vagal afferent neurons supplying the murine jejunum contain TRPV1 (Tan et al., 2009).

Within the rodent and human alimentary canal, TRPV1-positive nerve fibers occur in the musculature, enteric nerve plexuses, arterioles and mucosa (Facer et al., 2001; Yiangou et al., 2001; Chan et al., 2003; Patterson et al., 2003; Ward et al., 2003; Holzer, 2004; Horie et al., 2004; Kadowaki et al., 2004; Matthews et al., 2004; Robinson et al., 2004; Schicho et al., 2004; Zhang et al., 2004; Faussone-Pellegrini et al., 2005; Bhat & Bielefeldt, 2006; Akbar et al., 2008; Matsumoto et al., 2009, 2011). In addition, TRPV1 is present in nerve fibers supplying the taste papillae of the tongue (Caterina et al., 1997; Tominaga et al., 1998; Ishida et al., 2002; DeSimone & Lyall, 2006).

There are regional and species differences in the chemical coding of primary afferent neurons expressing TRPV1. Thus, calcitonin gene-related peptide (CGRP), substance P, somatostatin and other neuropeptides are messenger molecules characteristic ofcapsaicin-sensitive DRG neurons (Green & Dockray, 1988; Holzer, 1991; Sternini, 1992; Szallasi & Blumberg, 1999; Hwang et al., 2005; Price & Flores, 2007), while CGRP is absent from vagal afferent neurons containing TRPV1 (Tan et al., 2009). DRG neurons can be largely differentiated by their binding of isolectin B4 and their responsiveness to different neurotrophins (Guo et al., 1999; Michael & Priestley, 1999; Liu et al., 2004; Hwang et al., 2005; Price & Flores, 2007). In adult rodents, the isolectin B4-negative cell population responds to nerve growth factor, while isolectin B4-positive cells respond to the glial cell line-derived family of neurotrophins. However, there is no clear distinction between these groups of DRG neurons in their expression of TRPV1 and the neuropeptides substance P, CGRP and somatostatin (Price & Flores, 2007). In the rat, TRPV1 is found in both populations of DRG neurons but is more prevalent in isolectin B4-positive cells (Guo et al., 1999; Michael & Priestley, 1999; Liu et al., 2004; Hwang et al., 2005; Banerjee et al., 2007; Price & Flores, 2007), whereas in the mouse TRPV1 is largely absent from isolectin B4-positive DRG cells (Zwick et al., 2002; Woodbury et al., 2004; Price & Flores, 2007). In both rat and mouse, however, TRPV1 abounds in visceral DRG neurons that bind little isolectin B4 but are rich in CGRP and substance P (Ward et al., 2003; Robinson et al., 2004; Schicho et al., 2004; Brierley et al., 2005; Hwang et al., 2005; Christianson et al., 2006a, 2006b; Tan et al., 2008).

The presence of TRPV1 in the enteric nervous system has been less extensively studied than that in primary afferent neurons and has remained a matter of controversy. While some authors have localized TRPV1 immunoreactivity to somata of the guinea-pig, porcine and human enteric nervous system (Nozawa et al., 2001; Poonyachoti et al., 2002; Anavi-Goffer & Coutts, 2003; Chan et al., 2003), other authors have failed to confirm this location (Patterson et al., 2003; Ward et al., 2003; Holzer, 2004; Schicho et al., 2004). Importantly, the chemical coding of TRPV1-positive DRG neurons innervating the murine gut and of TRPV1-positive nerve fibers in the murine colon and rectum is distinct from that of enteric neurons (Tan et al., 2008; Matsumoto et al., 2011). Furthermore, TRPV1 messenger ribonucleic acid (mRNA) disappears from the rat stomach following extrinsic denervation (Schicho et al., 2004). Enteric neurons that issue projections from the rat colon to the spinal cord are likewise TRPV1-negative (Suckow & Caudle, 2008). The inconsistencies in the localization of TRPV1 to enteric neurons could in part be explained by the existence of several splice variants of TRPV1 (Wang et al., 2004; Faussone-Pellegrini et al., 2005; Szallasi et al., 2007) that may substantially differ in their immunoreactivity. For instance, two splicing products of the murine TRPV1 gene have been identified in the DRGs and several tissues including the gut. One of them, TRPV1-α, is equivalent to the rat and human TRPV1, whereas the other one, TRPV1-β, encodes a dominant negative subunit of the TRPV1 channel (Wang et al., 2004).

In addition to its prominent location in neurons, TRPV1 has been localized to non-neural cells of the alimentary canal. Thus, TRPV1-like immunoreactivity is found in the human submandibular gland where it occurs in serous acinar and ductal cells, but not in mucous acinar cells (Ding et al., 2010). Furthermore, TRPV1-like immunoreactivity has been localized to gastrin and parietal cells of the stomach as well as to epithelial cells of the esophageal, gastric and small intestinal mucosa of several species including man (Nozawa et al., 2001; Kato et al., 2003; Zhang et al., 2004; Faussone-Pellegrini et al., 2005; Kechagias et al., 2005; Cheng et al., 2009; Ericson et al., 2009; Ma et al., 2010). With respect to the presence of TRPV1 in these tissues a word of caution is in place, given that TRPV1 antisera can yield false positive staining in non-neural tissues, which persists in TRPV1 knockout mice (Everaerts et al., 2009).

2.2.2. Functional implications of TRPV1 in the digestive system

2.2.3. Association of TRPV1 with gastrointestinal disease

Upregulation of TRPV1 tissue levels has been observed in a number of inflammatory and functional GI disorders (Table 2). Thus, both erosive and non-erosive reflux diseases are associated with increased levels of TRPV1 in the esophageal mucosa (Matthews et al., 2004; Bhat & Bielefeldt, 2006; Guarino et al., 2010; Shieh et al., 2010). The upregulation of TRPV1 in the mucosa of patients with erosive esophagitis is associated with elevated expression of nerve growth factor and glial cell line-derived neurotrophic factor (Shieh et al., 2010). Idiopathic rectal hypersensitivity and fecal urgency (Chan et al., 2003) and Hirschsprung’s disease (Facer et al., 2001) are instances of TRPV1 upregulation in the absence of inflammation. Patients with uninvestigated dyspepsia have been found hypersensitive to intrajejunal capsaicin infusion (Hammer, 2006b), and a proportion of patients with functional dyspepsia are more responsive to ingestion of capsaicin capsules than healthy controls (Hammer et al., 2008). A role of TRPV1 in upper GI pain is also suggested by the antinociceptive effect which prolonged treatment with capsaicin-containing capsules has in healthy volunteers and patients with functional dyspepsia (Bortolotti et al., 2002; Führer & Hammer, 2009). Furthermore, the homozygous G315C polymorphism of the TRPV1 gene has been found to be inversely related to symptom severity in functional dyspepsia patients (Tahara et al., 2010).

Patients suffering from diarrhea-predominant irritable bowel syndrome exhibit hypersensitivity to the painful and burning sensations which chili-containing capsules and food elicit in the GI tract (Gonlachanvit et al., 2009). The TRPV1-positive innervation of the mucosa in the rectosigmoid colon is increased in patients with inflammatory bowel disease, in patients with irritable bowel syndrome as well as in patients with quiescent inflammatory bowel disease who continue to complain of irritable bowel syndrome-like symptoms (Yiangou et al., 2001; Akbar et al., 2008, 2010). Importantly, the density of TRPV1-positive nerve fibers in the rectosigmoid colon correlates with pain severity in both patients with irritable bowel syndrome and patients with symptomatic but quiescent inflammatory bowel disease (Akbar et al., 2008, 2010).

2.2.4. Therapeutic potential of TRPV1 ligands in the digestive system

Recognition of TRPV1 as a multimodal nocisensor, its sensitization by a number of proinflammatory and proalgesic pathways, its upregulation under conditions of hyperalgesia and its apparent implication in experimental colitis have made this TRP channel an attractive target for novel antinociceptive and antiinflammatory drugs (Szallasi et al., 2007; Holzer, 2008a, 2008b; Voight & Kort, 2010). Pharmacologically, the function of TRPV1 can be manipulated by two principal approaches: stimulant/defunctionalizing TRPV1 agonists and TRPV1 antagonists (Roberts & Connor, 2006; Gharat & Szallasi, 2008; Gunthorpe & Szallasi, 2008; Baraldi et al., 2010). The mechanism and result of the two approaches are profoundly different. While TRPV1 antagonists specifically modify the function of the ion channel, stimulant/defunctionalizing TRPV1 agonists target the cellular function of capsaicin-sensitive afferent neurons (Holzer, 1991; Szallasi et al., 2007). The “desensitization” that is brought about by capsaicin, resiniferatoxin or non-pungent TRPV1 ligands reflects “defunctionalization” of the whole afferent neuron expressing TRPV1 for a prolonged period of time.

At present, there is only evidence for a beneficial effect of stimulant/defunctionalizing TRPV1 agonists in GI disease. Acute ingestion of capsaicin has a protective effect against aspirin-induced erosions in the gastric mucosa of human volunteers (Yeoh et al., 1995). In addition, capsaicin can improve esophageal motility in patients with gastro-esophageal reflux disease (Grossi et al., 2006). Prolonged intake of capsaicin appears to desensitize afferent nerve fibers against noxious stimulation of the upper GI tract. Thus, ingestion of capsaicin (0.25 mg) capsules by healthy volunteers 3 times per day for 4 weeks blunts the pain response evoked by duodenal capsaicin administration and balloon distension (Führer & Hammer, 2009). Similarly, treatment of patients suffering from functional dyspepsia with capsaicin-containing capsules for 5 weeks leads to a significant reduction of pain symptoms (Bortolotti et al., 2002). Since there is some information that the prevalence of heartburn symptoms is low in Asian countries in which the intake of chili-rich food is common, it has been proposed that chronic ingestion of capsaicin may be beneficial in patients with gastro-esophageal reflux disease and functional dyspepsia (Gonlachanvit, 2010).

A great deal of effort has been put into developing compounds that block TRPV1 activation in a competitive or noncompetitive manner (Gharat & Szallasi, 2008; Gunthorpe & Szallasi, 2008; Kym et al., 2009; Wong & Gavva, 2009; Voight & Kort, 2010). Some TRPV1 blockers exhibit species differences in their activity and/or act in a stimulus-specific manner, i.e., differentially inhibit the activation of TRPV1 by capsaicin, heat, and acid (Gavva et al., 2005; Lehto et al., 2008). The design of stimulus-specific blockers is of particular importance because a number of TRPV1 antagonists turned out to have a pronounced hyperthermic effect that limits their clinical usefulness (Caterina, 2008; Gavva, 2008; Holzer, 2008b; Romanovsky et al., 2009; Wong & Gavva, 2009). Apart from their thermoregulatory perils (Caterina, 2008), TRPV1 antagonists are liable to impair the perception of potentially injurious heat (Voight & Kort, 2010). Furthermore, blockade of TRPV1 will interfere with the physiological role of this nocicensor in surveying the physical and chemical environment and, if necessary, in initiating protective responses. Such a role is obvious in the GI tract in which capsaicin-sensitive afferent neurons constitute a neural alarm system which helps maintaining mucosal homeostasis in the face of pending injury (Holzer, 1998, 2004; Akiba et al., 2006b). Other caveats are posed by the incomplete analysis of the involvement of TRPV1 in GI disease. For instance, the upregulation of TRPV1 in gut disorders does not necessarily reflect a causal implication of TRPV1 but may equally well mirror an epiphenomenon of the disease process (Holzer, 2008a).

There are several opportunities to design TRPV1 blockers such that they can be specifically targeted at aberrantly expressed or aberrantly operating TRPV1 channels (Holzer, 2008b) while sparing their physiological function. These opportunities include the development of modality-specific TRPV1 blockers, of uncompetitive TRPV1 antagonists, of compounds that selectively address certain TRPV1 splice variants that may be particularly disease-relevant, compounds that prevent TRPV1 sensitization, and compounds that prevent the recruitment and trafficking of TRPV1 to the cell membrane (Gavva, 2008; Holzer, 2008b; Romanovsky et al., 2009). If these approaches turn out to be successful, TRPV1 blockers may enjoy a wide spectrum of usefulness in GI disease which includes, pain, hyperalgesia and inflammation associated with gastro-esophageal reflux disease, inflammatory bowel disease, functional Gi disorders including functional dyspepsia and irritable bowel syndrome, disturbances of GI motor activity, and nausea and emesis.

2.3.1. Occurrence of TRPV2 and TRPV3 in the alimentary canal

TRPV2 has been localized to rodent DRG and nodose ganglion neurons supplying the GI tract (Caterina et al., 1999; Kashiba et al., 2004; Zhang et al., 2004; Stotz et al., 2008; Zhao & Simasko, 2010; Zhao et al., 2010). In contrast to TRPV1, TRPV2 occurs primarily in medium and large diameter afferent neurons (Zhang et al., 2004). In the rat oral mucosa, TRPV2 is not only present in subepithelial nerve fibers, but also in junctional epithelial cells surrounding each tooth, Langerhans cells, dendritic cells, macrophages and endothelial cells of venules, but not in oral epithelial cells (Shimohira et al., 2009). In the rat and guinea-pig small intestine, TRPV2 is found in nerve fibers within the muscularis and myenteric plexus as well as in myenteric cell bodies (Kashiba et al., 2004; Zhang et al., 2004). In addition, TRPV2 occurs in the β-cells of murine pancreatic islets and in the insulinoma cell line MIN6 (Hisanaga et al., 2009). Immunohistochemistry of human tissues has localized TRPV2 to epithelial cells of the salivary glands, pancreatic duct, stomach, duodenum and colon (Kowase et al., 2002).

TRPV3 is highly expressed by epithelial cells in the nose and tongue (Xu et al., 2006). It likewise occurs in DRG and nodose ganglion neurons (Zhang et al., 2004; Stotz et al., 2008; Zhao & Simasko, 2010; Zhao et al., 2010), but retrograde labeling has failed to identify TRPV3 in vagal afferent neurons supplying the murine stomach (Zhang et al., 2004). However, the muscle and mucosa of the murine stomach and small intestine do contain TRPV3 mRNA (Zhang et al., 2004). In addition, TRPV3 is expressed by superficial epithelial cells of the murine distal colon, but not by superficial epithelial cells of the stomach, duodenum and proximal colon (Ueda et al., 2009).

2.3.2. Functional implications of TRPV2 and TRPV3 in the digestive system

2.3.3. Association of TRPV3 with gastrointestinal disease

A study analyzing a possible association between genetic variability (as examined for 392 single-nucleotide polymorphisms) in 43 fatty acid metabolism-related genes and risk for colorectal cancer in 1225 patients with cancer and 2032 controls has shown that TRPV3 is associated with a higher risk for development of colorectal cancer (Hoeft et al., 2010).

2.4.1. Occurrence of TRPV4 in the alimentary canal

In the GI tract, TRPV4 occurs primarily in fibers of extrinsic primary afferent neurons, although some epithelial and other cells have also been reported to stain positively for this TRPV channel subunit. TRPV4 is present in the nodose ganglion, DRG, stomach, small intestine and colon of rodents (Zhang et al., 2004; Grant et al., 2007; Cenac et al., 2008; Brierley et al., 2008; Zhao & Simasko, 2010; Zhao et al., 2010). Retrograde labeling shows that vagal afferent neurons projecting to the murine forestomach contain TRPV4 and that, in these neurons, TRPV4 is coexpressed with TRPV1, TRPV2 and TRPA1 to various degrees (Zhang et al., 2004). Similarly, DRG neurons projecting to the murine pancreas synthesize both TRPV4 and TRPA1 (Ceppa et al., 2010). The expression of TRPV4 in the DRG neurons that project to the murine and human intestine via mesenteric and pelvic nerves exhibits pronounced regional differences (Brierley et al., 2008; Cenac et al., 2008; Sipe et al., 2008). Thus, TRPV4 is particularly abundant in DRG neurons supplying the colon, in which it is coexpressed with CGRP (Brierley et al., 2008; Cenac et al., 2008; Sipe et al., 2008). In the human colon, TRPV4 is found in fine nerve fibers associated with blood vessels in the submucosa and serosa, while the myenteric plexus, the circular and the longitudinal smooth muscles are largely negative (Brierley et al., 2008).

In the murine colon, TRPV4 occurs in brush-bordered epithelial cells, but not in mucus-secreting epithelial cells, as well as in mucosal glial cells and unidentified cells of the submucosal and muscular layers (Cenac et al., 2008; D’Aldebert et al., 2011). TRPV4 is likewise expressed by epithelial cells of the human colon and by the human colon carcinoma cell line Caco-2 (D’Aldebert et al., 2011).

A further location of TRPV4 is in the hepatobiliary tract in which it is expressed by ciliated cholangiocytes, as shown for both the rat, murine and human liver and bile system (Gradilone et al., 2007, 2010). In these cells, TRPV4 occurs specifically in the apical membrane and the cilia which are thought to subserve a mechano- and osmosensory role (Gradilone et al., 2007).

2.4.2. Functional implications of TRPV4 in the digestive system

2.4.3. Association of TRPV4 with gastrointestinal disease

TRPV4 channelopathies are responsible for a number of hereditary sensory and motor neuropathies as well as diseases with defects in bone development (Nilius & Owsianik, 2010). Thus, mutations of TRPV4 are found in congenital distal spinal muscular atrophy and scapuloperoneal spinal muscular atrophy (two hereditary motor neuropathies), hereditary motor and sensory neuropathy of type 2C (also known as Charcot–Marie–Tooth disease type 2C) (Auer-Grumbach et al., 2010; Deng et al., 2010; Landouré et al., 2010) and other rare motor neuropathies (Zimoń et al., 2010). Diseases with defective bone development such as metatropic dysplasia, brachyolmia, spondylo-epiphyseal dysplasia of Kozlowski type, spondylo-epiphyseal dysplasia of Maroteaux type (pseudo-Morquio syndrome type 2) and parastremmatic dysplasia are also associated with TRPV4 channelopathies (Camacho et al., 2010; Loukin et al., 2010; Nishimura G. et al., 2010).

Although the GI tract is not the prime victim of these diseases, gastro-esophageal reflux has been observed in hereditary sensory neuropathy of type 1B (Auer-Grumbach, 2008). Little is known, however, whether the hereditary sensory neuropathies of types 1A, 1B and 1D (Auer-Grumbach, 2008) impact on GI sensation and nociception, although the available reports do not mention any conspicuous GI disease phenotype.

TRPV4 in sensory neurons is upregulated in inflammatory bowel disease, and serosal blood vessels in tissues with active colitis are more densely innervated by TRPV4-positive nerve fibers than in biopsies from healthy controls (Brierley et al., 2008). Expression of TRPV4 in colonic epithelial cells of mice with experimental colitis is likewise upregulated, while the levels of TRPV4 mRNA in mucosal biopsies taken from patients with inflammatory bowel disease have been found unaltered (D’Aldebert et al., 2011). However, the colonic mucosa of inflammatory bowel disease patients is infiltrated by TRPV4-positive CD45-positive leukocytes which are absent from the mucosa of healthy controls (D’Aldebert et al., 2011).

In patients with polycystic liver diseases (autosomal-recessive polycystic kidney disease and autosomal-dominant polycystic kidney disease) TRPV4 is overexpressed in cholangiocytes, a finding that is reproduced in polycystic kidney rats, a model for autosomal-recessive polycystic kidney disease (Gradilone et al., 2010).

2.4.4. Therapeutic potential of TRPV4 ligands in the digestive system

Taken together, the experimental findings related to TRPV4 function in the digestive system attribute this TRP channel a pathophysiological role in inflammation, pain, hyperalgesia and polycystic liver disease. This inference if supported by the overexpression of TRPV4 in experimental colitis, inflammatory bowel disease and polycystic liver disease. Since TRPV4 appears to be a common mediator of mechanical hyperalgesia in the colon, it presents itself as a novel therapeutic target in patients with abdominal pain. In addition to their antinociceptive and antihyperalgesic potential, TRPV4 blockers may be useful as antiinflammatory drugs. To the contrary, TRPV4 agonists seem to be beneficial in polycystic liver disease in which they can inhibit cholangiocyte proliferation and cyst formation (Gradilone et al., 2010). Indeed, TRPV4 agonists and blockers are already on the horizon (Vincent et al., 2009) and await to be assessed for their therapeutic efficacy and safety.

2.5.1. Occurrence of TRPV5 and TRPV6 in the alimentary canal

TRPV5 and TRPV6 are expressed in the mucosa of the human and rodent small and large intestine (Clapham et al., 2005; Nijenhuis et al., 2005; Schoeber et al., 2007; Benn et al., 2008; Suzuki et al., 2008; Fukushima et al., 2009; Peleg et al., 2010). While TRPV5 occurs primarily in the kidneys, TRPV6 is predominantly expressed in intestinal epithelial cells and has also been localized to human colon cancer (Caco-2) cells (Schoeber et al., 2007; Fukushima et al., 2009; Bartik et al., 2010). In addition, TRPV5 mRNA occurs in rat nodose ganglion neurons expressing cholecystokinin (CCK) CCK1 receptors (Zhao & Simasko, 2010), which are likely to project to the GI tract.

2.5.2. Functional implications of TRPV5 and TRPV6 in the digestive system

TRPV5 and TRPV6 are the most Ca2+-selective members of the TRP ion channel family, play an important role in intestinal Ca2+ absorption (Clapham et al., 2005; Nijenhuis et al., 2005; Schoeber et al., 2007; Benn et al., 2008; Suzuki et al., 2008), and undergo calcium-induced inactivation via a phospholipase C-mediated pathway (Thyagarajan et al., 2009). The functional roles of TRPV5 and TRPV6 are interconnected with each other, given that TRPV5 knockout mice upregulate their intestinal TRPV6 expression to compensate for the negative Ca2+ balance caused by the loss of TRPV5-mediated Ca2+ reabsorption in the kidney (Suzuki et al., 2008). In contrast, TRPV6 knockout mice do not display any compensatory mechanism, which results in secondary hyperparathyroidism (Benn et al., 2008; Suzuki et al., 2008).

The expression and function of both TRPV5 and TRPV6 are subject to regulation by dietary conditions and multiple calciotropic hormones (Nijenhuis et al., 2005; Schoeber et al., 2007; Suzuki et al., 2008). Calcium deficiency in the diet is an important factor that increases the expression of TRPV6 in the duodenal mucosa of mice (Ko et al., 2009). Short chain fatty acids, the fermentation products of fructooligosaccharides, likewise increase expression of TRPV6 and calcium absorption in the rat colorectal epithelium as well as in Caco-2 cells (Fukushima et al., 2009). Curcumin, a polyphenol present in turmeric, is a ligand of the nuclear vitamin D receptor, activation of which enhances the expression of TRPV6 mRNA in Caco-2 cells (Bartik et al., 2010).

Parathyroid hormone and 1,25-dihydroxyvitamin D3 are particularly relevant to the expression of TRPV5 and TRPV6 in the intestinal mucosa (Nijenhuis et al., 2005; Schoeber et al., 2007). In chicken intestinal epithelial cells, 1,25-dihydroxyvitamin D3 and parathyroid hormone increase calcium uptake by stimulating a protein kinase A-mediated pathway to release β-glucuronidase, which in turn activates TRPV6 (Khanal et al., 2008). Consistent with this secondary transducer role of TRPV6 is the finding that the 1,25-dihydroxyvitamin D3-induced calcium uptake is abolished by TRPV6 siRNA (Khanal et al., 2008). In normal human duodenal mucosa obtained at endoscopy, TRPV6 expression is likewise increased following exposure to 1,25-dihydroxyvitamin D3 (Balesaria et al., 2009). Although there is a close functional interconnection between vitamin D and intestinal TRPV6, these factors can also operate independently of each other. Thus, the 1,25-dihydroxyvitamin D3-induced intestinal calcium transport in the murine duodenum does not entirely depend on TRPV6, because intestinal calcium uptake also occurs in TRPV6 knockout mice (Benn et al., 2008). In addition, intestinal calcium absorption and duodenal TRPV6 expression are upregulated in pregnant and lactating mice even in the absence of the nuclear vitamin D receptor (Fudge & Kovacs, 2010). Other hormones that impact on intestinal calcium absorption include prolactin which induces TRPV6 mRNA in the rat duodenal epithelium, synergizes with 1,25-dihydroxyvitamin D3 in enhancing TRPV6 expression and stimulates duodenal calcium transport (Ajibade et al., 2010).

Calcium homeostasis is disturbed in patients with Crohn’s disease (Huybers et al., 2008). Similarly, heterozygous mice carrying a deletion in the adenine- and uridine-rich elements of the tumor necrosis factor gene, considered to be a model for Crohn’s disease, display an increase in bone resorption along with a decrease in the duodenal expression of TRPV6 mRNA (Huybers et al., 2008). TRPV6 in the intestinal epithelium also appears to have a bearing on colon cancer development. Thus, TRPV6 expression is significantly enhanced in the Citrobacter rodentium-induced transmissible murine colonic hyperplasia model (Peleg et al., 2010). Whereas in the normal colon TRPV6 is restricted to the apical membrane of absorptive enterocytes, in the hyperplasia model TRPV6 is also distributed to the proliferating zone of the colonic crypts, in which it occurs in the basolateral membrane and perinuclear area of many epithelial cells (Peleg et al., 2010). When the animals are fed with a calcium-rich diet, the overexpression of TRPV6 in the hyperplasia model is reversed.

2.5.3. Association of TRPV6 with gastrointestinal disease

Increased levels of colonic TRPV6 are associated with early-stage colon cancer. Thus, TRPV6 has been found to be overexpressed in 66% of stage I tumors and in 17% of stage II tumors but is barely detectable in stages III and IV tumors (Peleg et al., 2010). These results suggest that aberrant function of TRPV6 is associated with early colon carcinogenesis but not with frank malignancy (Peleg et al., 2010). This inference is supported by experiments with the human colon carcinoma cell line Caco-2, which constitutively express high levels of TRPV6. Treatment of these cells with TRPV6-specific siRNA significantly reduces TRPV6 mRNA levels by about 50%, which is associated with a 40% reduction of cell proliferation and a more than twofold increase in apoptosis (Peleg et al., 2010). Together with experiments in mice with colon crypt hyperplasia (Peleg et al., 2010), these findings indicate that TRPV6 promotes the proliferation of colonic epithelial cells and may be a pathogenetic factor in the early stages of colon cancer development.

2.5.4. Therapeutic potential of TRPV6 ligands in the digestive system

Pharmacological interference with TRPV6 in the digestive system is likely to interfere with calcium absorption and bone mineralization as well as with epithelial cell hyperplasia and early stages of malignancies. Thus, blockade of TRPV6 can be considered as a measure to prevent and inhibit the development of colon cancer, but this effect needs to be balanced against any negative impact on calcium homeostasis.

2.6.1. Occurrence of TRPA1 in the alimentary canal

In the GI tract, TRPA1 occurs in 3 distinct cellular systems: extrinsic primary afferent neurons, intrinsic enteric neurons and endocrine cells of the mucosa.

Primary sensory neurons constitute the most significant source of TRPA1 in the body, given that TRPA1 is abundantly expressed by trigeminal ganglion, nodose ganglion and DRG neurons (Story et al., 2003; Zhang et al., 2004; Kobayashi et al., 2005; Anand et al., 2008; Yang et al., 2008; Brierley et al., 2009; Kondo et al., 2009, 2010; Yu et al., 2009; Cattaruzza et al., 2010; Ceppa et al., 2010; Hondoh et al., 2010; Zhao et al., 2010). In rodent and human DRG neurons, especially in those projecting to the gut, TRPA1 is preferentially expressed by small and medium diameter cell bodies in which it is colocalized with TRPV1 and CGRP (Bautista et al., 2005; Kobayashi et al., 2005; Anand et al., 2008; Brierley et al., 2009; Kondo et al., 2009, 2010; Cattaruzza et al., 2010; Hondoh et al., 2010; Zhao et al., 2010). Similarly, TRPA1 on nodose ganglion neurons in the guinea-pig is primarily associated with small to medium diameter somata and co-expressed with PAR-2 (Yu et al., 2009). In the periphery, TRPA1 has been found in the stomach, small intestine, colon and pancreas of the rat and mouse, the major source being peripheral nerve fibers of DRG and nodose ganglion neurons (Zhang et al., 2004; Penuelas et al., 2007; Brierley et al., 2009; Kondo et al., 2009, 2010; Cattaruzza et al., 2010; Ceppa et al., 2010; Dong Y. et al., 2010). The distribution of TRPA1 in the rodent gut is similar to that in the canine gut, given that TRPA1 is strongly expressed in the stomach, pancreas, small intestine and colon of the dog (Doihara et al., 2009a).

The other sources of TRPA1 in the GI tract are neurons of the human and murine enteric nervous system (Penuelas et al., 2007; Anand et al., 2008) as well as 5-HT-releasing enterochromaffin cells and CCK-releasing endocrine cells of the human, rat and murine GI mucosa (Purhonen et al., 2008; Nozawa et al., 2009). The murine neuroendocrine cell line STC-1, the rat endocrine cell line RIN14B and the human pancreatic endocrine cell line QGP-1, which contain enterochromaffin cell markers, likewise express TRPA1 (Purhonen et al., 2008; Doihara et al., 2009c; Nozawa et al., 2009).

2.6.2. Functional implications of TRPA1 in the digestive system

2.6.3. Association of TRPA1 with gastrointestinal disease

A point mutation in the S4 transmembrane segment of the TRPA1 gene appears to be responsible for an autosomal-dominant familial episodic pain syndrome that is triggered by fasting and physical stress (Kremeyer et al., 2010). Functional analysis shows that this mutation in the TRPA1 channel is associated with a 5-fold increase in inward current following activation and with enhanced secondary hyperalgesia caused by AITC (Kremeyer et al., 2010). It awaits to be explored whether function-relevant mutations in the TRPA1 gene are relevant to GI disease.

2.6.4. Therapeutic potential of TRPA1 ligands in the digestive system

The properties of TRPA1 as a multimodal sensor for a huge variety of irritant chemicals (Table 1) indicate that TRPA1 blockers could be used to manage chemical irritation and pain. In addition, the emerging pathophysiological implications of TRPA1 suggest that both TRPA1 agonists and blockers hold therapeutic potential in particular disease entities. There is good reason to surmise that TRPA1 blockers could be of particular benefit in inflammation-induced hyperalgesia in the GI tract. TRPA1 blockers may also be of use in pancreatitis (Ceppa et al., 2010) and inflammatory bowel disease. The discovery that activation of TRPA1 by some drugs contributes to their adverse effect profile (Table 1), which may include the alimentary canal, provides a rational basis for developing drugs that lack this unwanted property.

TRPA1 blockers such as HC-030031 have already been characterized, and many other compounds are in the pipeline (Baraldi et al., 2010). The interaction between TRPA1 and TRPV1 in the excitation and/or sensitization of nociceptive nerve fibers (Salas et al., 2009; Patil et al., 2010) hints at the possibility that combined TRPA1 plus TRPV1 blockers may have superior efficacy compared with blockers that target TRPA1 or TRPV1 alone. It awaits to be explored whether interference with TRPA1 disturbs temperature homeostasis, as it has been found with TRPV1 blockers. This limitation cannot be neglected, given that intragastric administration of the TRPA1 agonists AITC and cinnamaldehyde increases thermogenesis and reduces heat diffusion in mice (Masamoto et al., 2009).

Apart from TRPA1 blockers, TRPA1 agonists also may have therapeutic potential. The large variety of TRPA1 agonists available to date (Table 1) can be roughly categorized in two classes: (1) agonists that stimulate TRPA1 via a transient and reversible interaction with the channel, and (2) agonists with an electrophilic carbon or sulfur that bind covalently to cysteine residues of the cytoplasmatic N-terminus of TRPA1 (Baraldi et al., 2010). A specific application of TRPA1 agonists is suggested by their prokinetic effect in the murine colon, which—if translated to humans—could be utilized in the management of both atonic and spastic motor stasis of the gut (Kojima et al., 2009). TRPA1 agonists that will be useful in therapy certainly require optimization in their pharmacological profile, since the utility of the currently available TRPA1 agonists is limited by their irritative effects and high reactivity.

2.7.1. Expression of TRPM4 and TRPM5 in the alimentary canal

Expression of TRPM4 and, particularly, TRPM5, is limited to taste, epithelial and endocrine cells. TRPM4 transcripts have been identified in pancreatic α- and β-cell lines of rodent origin (Cheng et al., 2007; Marigo et al., 2009), and TRPM5 has been localized to pancreatic islets of the mouse in which it is expressed by the insulin-secreting β cells (Colsoul et al., 2010). TRPM5 abounds on the receptor cells of the lingual taste buds as well as on other chemosensory cells, notably on epithelial taste cells of the gut (Zhang et al., 2003; Bezençon et al., 2007; Kaske et al., 2007; Kokrashvili et al., 2009). Both TRPM4 and TRPM5 mRNA are found in the human digestive system (Fonfria et al., 2006). TRPM5 occurs in the stomach, small intestine and colon of mice and humans, the channel occurring in solitary brush (also termed tufted or caveolated) cells at the surface epithelium of the small and large intestine and in endocrine cells of the duodenal glands (Fonfria et al., 2006; Bezençon et al., 2007, 2008; Kaske et al., 2007; Kokrashvili et al., 2009; Young et al., 2009). Endocrine cells isolated from the rat stomach also stain positively for TRPM5 (Nakamura et al., 2010).

2.7.2. Functional implications of TRPM4 and TRPM5 in the digestive system

Functional evidence suggests that TRPM4 contributes to the membrane depolarization and increase in the intracellular calcium concentration that are relevant to glucose-, arginine vasopressin- and glibenclamide-evoked secretion of insulin from pancreatic β-cells and perhaps glucagon from pancreatic α-cells (Cheng et al., 2007; Marigo et al., 2009). TRPM5 is likewise thought to play a major role in the glucose-dependent release of insulin from pancreatic β-cells (Brixel et al., 2010; Colsoul et al., 2010). Specifically, the effect of glucose to release insulin is significantly reduced in TRPM5 knockout mice, which impairs glucose tolerance (Brixel et al., 2010; Colsoul et al., 2010).

The Ca2+-gated TRPM5 channel on the receptor cells of the taste buds in the tongue has been established to contribute to signal transduction in the sweet, umami and bitter tastes (Zhang et al., 2003; Damak et al., 2006). Bitter, sweet, and umami tastants bind to GPCRs which initiate a second-messenger signaling cascade involving activation of phospholipase Cβ2 and hydrolysis of phosphatidylinositol 4,5-bisphosphate to diacylglycerol and inositol trisphosphate. The subsequent increase in the intracellular Ca2+ concentration gates TRPM5 and causes depolarization of the taste cells and chemical transmission to gustatory afferent neurons (Zhang Z. et al., 2007). Real life gustatory experiences are often evoked by taste mixtures and determined by multiple interactions between different taste stimuli (Talavera et al., 2008). One such interaction is the suppression of sweet taste transduction by bitter tastants such as quinine. Quinine and its stereoisomer quinidine have been found to block TRPM5 currents and to inhibit the nerve impulses of sucrose-stimulated gustatory nerves in mice (Talavera et al., 2008). Since this quinidine-evoked inhibition of sweet taste signaling is absent in TRPM5 knockout mice, it would appear that TRPM5 plays an important role in bitter–sweet taste interactions (Talavera et al., 2008).

Apart from their occurrence in lingual taste cells, TRPM5 is widely expressed by other chemosensory cells, notably by the tufted epithelial cells of the GI tract (Kaske et al., 2007; Rozengurt & Sternini, 2007). Given that the TRPM5-positive cells of the gut express most genes encoding proteins implicated in sweet, bitter, and umami taste signaling (Bezençon et al., 2008), it would appear that TRPM5 exert taste functions throughout the alimentary canal and contribute to the surveillance of the chemical composition in its lumen. This functional implication is corroborated by the finding that the expression of TRPM5 in the upper small intestine is inversely correlated with the blood glucose concentration in type 2 diabetes patients (Young et al., 2009).

Another task of TRPM5-positive taste cells in the digestive system may be to play local effector roles because they express markers of neuronal and inflammatory cells (Bezençon et al., 2008) and can release biologically active messengers in response to stimulation. Solitary chemosensory cells in the upper small intestine of the murine coexpress TRPM5, β-endorphin, methionine enkephalin and uroguanylin (Kokrashvili et al., 2009). Stimulation of these cells with hyperosmolar solutions or glucose causes release of β-endorphin into the intestinal lumen but not blood serum, an effect that is blunted in TRPM5 knockout mice (Kokrashvili et al., 2009). In contrast, the intestinal release of β-endorphin evoked by acidic solutions or lipid exposure appears to be independent of TRPM5 (Kokrashvili et al., 2009).

2.7.3. Therapeutic potential of TRPM5 ligands in the digestive system

Being a transducer of the sweet taste, TRPM5 has been suggested to have therapeutic potential in the management of obesity and related metabolic dysfunctions, given that taste signaling is a critical determinant of ingestive behavior (Sprous & Palmer, 2010). How efficacious such an approach may be is difficult to judge, however, because deletion of TRPM5 does not completely abolish the preference for sweets and energy-rich food. The reason for this finding is that there are TRPM5-independent transduction mechanisms for glucose tasting (Ohkuri et al., 2009) and sweet and calorie-rich nutrients can directly influence brain reward circuits that control food intake independently of functional taste transduction (de Araujo et al., 2008; Ren et al., 2010).

As TRPM5 appears to be involved in glucose-dependent insulin secretion, TRPM5 dysfunction has been suggested to be a factor in the etiology of some forms of type 2 diabetes (Brixel et al., 2010). If so, normalization of TRPM5 function may be envisaged as a target in diabetes management.

2.8.1. Occurrence of TRPM6 and TRPM7 in the alimentary canal

TRPM6 and TRPM7 occur in GI epithelial cells and, as is true for TRPM7, in ICCs. In the murine gut, TRPM6 and TRPM7 abound in the brush-border membrane of the colon (Schlingmann et al., 2002; Voets et al., 2004; Fonfria et al., 2006; Groenestege et al., 2006; Rondón et al., 2008a, 2008b; Woudenberg-Vrenken et al., 2010). Furthermore, TRPM7 mRNA and protein are expressed by AGS cells, the most common human gastric adenocarcinoma cell line (Kim et al., 2008). Apart from their association with epithelial cells, TRPM7 is present on ICCs of the myenteric plexus and circular muscle throughout the GI tract of mice and humans (Kim et al., 2005, 2006, 2009).

2.8.2. Functional implications of TRPM6 and TRPM7 in the digestive system

Several data suggest that the formation of functional TRPM6 channels in epithelial cell membranes requires co-expression with TRPM7. There is indeed evidence for the formation of heteromeric ion channels consisting of both TRPM6 and TRPM7 subunits, but there is still controversy as to whether TRPM6 can also function on its own and TRPM7 is required only for trafficking of TRPM6 to the plasma membrane (Schlingmann et al., 2007; Quamme, 2008). TRPM6 and TRPM7 possess an atypical α-kinase domain and are involved in Mg2+ homeostasis, given that genetic and functional data indicate that both channel kinases play a critical role in the cellular Mg2+ absorption within the GI mucosa (especially in the colon) and in the reabsorption of Mg2+ in the kidney (Schlingmann et al., 2007; Quamme, 2008; Woudenberg-Vrenken et al., 2010). Homozygous deletion of TRPM6 in mice is lethal, while heterozygous deletion of TRPM6 results in mild hypomagnesemia, a deficit that cannot be compensated for by a magnesium-rich diet (Woudenberg-Vrenken et al., 2010).

TRPM6 and TRPM7 channel function and colonic Mg2+ absorption are in fact controlled by many factors. For instance, the hypomagnesemia associated with dietary Mg2+ restriction in mice causes upregulation of TRPM6 expression both in the intestine and kidney (Rondón et al., 2008a). The expression of TRPM6, but not TRPM7, mRNA in the colon is likewise enhanced by a Mg2+-rich diet (Groenestege et al., 2006). Feeding of mice with long-chain inulin for 2 weeks, which increases Mg2+absorption and enhances bone stores of Mg2+, results in a rise of TRPM6 and TRPM7 expression in the hindgut, while TRPM7 in the kidney is down-regulated (Rondón et al., 2008b).

Mg2+ is critical for the growth and survival of certain cancer cells, notably the human gastric adenocarcinoma cell line AGS (Kim et al., 2008). Blockade of TRPM7 channels with La3+ or suppression of TRPM7 expression by siRNA inhibits the growth and survival of AGS cells (Kim et al., 2008).

In addition to its epithelial function, TRPM7 plays a role in controlling intestinal motor activity. Electrophysiological, molecular biological, and immunohistochemical evidence indicates that TRPM7 participates in the nonselective cation current that underlies the pacemaker activity of ICCs in the GI tract (Kim et al., 2006). Thus, the electrophysiological and pharmacological properties of the nonselective cation current in ICCs of the mouse are identical to those of TRPM7, and the pacemaker activity of ICCs is inhibited by TRPM7-specific siRNA (Kim et al., 2005). TRPM7 expressed on ICCs is also likely to contribute to pacemaker activity in the human gut, since slow waves in the isolated muscle of the human small and large intestine are attenuated by blockade of TRPM7 channels with La3+ (Kim et al., 2009).

2.8.3. Association of TRPM6 with gastrointestinal disease

Non-functional mutations of TRPM6 are responsible for a primary disorder termed hypomagnesemia with secondary hypocalcemia (Schlingmann et al., 2002; Voets et al., 2004). The expression of TRPM6 mRNA in the intestinal epithelium is reduced in patients suffering from cholera, a change that is thought to counteract the secretory response to the infection (Flach et al., 2007).

2.8.4. Therapeutic potential of TRPM7 blockers in the digestive system

The ability of TRPM7 channel blockade to inhibit the growth and survival of AGS cells suggests TRPM7 to be a potential target for the treatment of gastric cancer (Kim et al., 2008).

2.9.1. Occurrence of TRPM8 in the alimentary canal

TRPM8 is distributed to several cell types in the digestive system. Being menthol-sensitive, the TRPM8 subunit is expressed in the papillae of the tongue as well as by a distinct population of primary afferent neurons originating in the nodose-petrosal ganglion and DRG (McKemy et al., 2002; Peier et al., 2002; Zhang et al., 2004; Kobayashi et al., 2005; Hondoh et al., 2010; Zhao et al., 2010). Unlike that of TRPA1, the colocalization of TRPM8 with TRPV1 in sensory neurons is limited (Bautista et al., 2005; Kobayashi et al., 2005; Kondo et al., 2009; Hondoh et al., 2010; Zhao et al., 2010). Within the GI tract, TRPM8 has been localized to the muscle of the rat stomach and colon (Mustafa & Oriowo, 2005) as well as to the liver (Fonfria et al., 2006). The expression of TRPM8 in the murine stomach and small intestine remains controversial, given that both positive (Zhang et al., 2004) and negative (Penuelas et al., 2007) results have been reported.

2.9.2. Functional implications of TRPM8 in the digestive system

TRPM8 is activated by temperatures in the cool to cold range and makes an important contribution to cold-induced pain (Bautista et al., 2007; Colburn et al., 2007; Dhaka et al., 2007; Knowlton et al., 2010). Apart from low temperatures, TRPM8 is stimulated by various chemical entities including menthol, icilin, geraniol, L-carvone, isopulegol and linalool (Table 1). These sensory modalities of TRPM8 are likely to explain the refreshing taste of menthol. Also keeping with this property is the finding that intragastric administration of the TRPM8 agonist menthol induces thermogenesis in mice (Masamoto et al., 2009).

Expressed by extrinsic primary afferent neurons and intrinsic systems within the GI tract, TRPM8 may play a chemosensory role in the alimentary canal. This contention is supported by the observation that the mixed TRPA1/TRPM8 agonist icilin is able to stimulate murine nodose ganglion neurons in culture (Zhang et al., 2004). However, systematic studies of the chemosensory role of TRPM8-bearing primary afferent neurons in the gut have not yet been reported, and the precise functional implications of TRPM8 in the digestive system await to be explored. In the murine colon, menthol induces a long lasting relaxation of smooth muscle, which remains unaffected by tetrodotoxin (Penuelas et al., 2007). It is questionable if this motor effect of menthol is due to activation of TRPM8 because the presence of TRPM8 in the murine gut has not unequivocally been established (Zhang et al., 2004; Penuelas et al., 2007). Whether the cooling-induced contraction of the rat gastric fundus (Mustafa & Oriowo, 2005) and guinea-pig ileum (Holzer & Lembeck, 1979) involves TRPM8 has not conclusively been investigated.

2.10.1. Occurrence of TRPP2 in the alimentary canal

The polycystin TRP channel TRPP2 (also known as polycystic kidney disease-like ion channel, PKD2L1) and the related PKD1 polycystin L3 (PKD1L3) are expressed by a select class (type III) of taste chemoreceptor cells in the murine tongue, which are distinct from those mediating sweet, umami and bitter tastes (Huang et al., 2006; Ishimaru et al., 2006; LopezJimenez et al., 2006; Kataoka et al., 2008; Chandrashekar et al., 2009; Kawaguchi et al., 2010). While TRPP2 has been reported to occur in the circumvallate, foliate and fungiform papillae of the tongue and in the palate, PKD1L3 has been localized to the circumvallate and foliate papillae only (Huang et al., 2006; Ishimaru et al., 2006). PKD1, the gene that is mutated in ~85% of autosomal dominant polycystic kidney disease patients, has been described to occur in many tissues including the intestine (Geng et al., 1997).

2.10.2. Functional implications of TRPP2 in the digestive system

Although several members of the TRPV, TRPA and TRPC channel subfamilies are sensitive to alterations in extracellular pH (Holzer, 2009), TRPP2 (PKD2L1) is the TRP channel involved in sour taste transduction (Chandrashekar et al., 2006; Ishimaru & Matsunami, 2009). In order to form functional channels, TRPP2 needs to associate as heteromer with related proteins of the PKD1 polycystin family such as PKD1L3 (Ishimaru et al., 2006; Kawaguchi et al., 2010). The PKD1 polycystins are not included in the TRP channel family because they are large proteins with a very long N-terminal extracellular domain and 11 transmembrane domains that include a 6-transmembrane TRP-like channel domain at the C terminus (Clapham et al., 2005).

Both genetic and functional studies corroborate that TRPP2 plays an important role as sour taste receptor (Huang et al., 2006; Ishimaru et al., 2006; LopezJimenez et al., 2006; Kataoka et al., 2008; Kawaguchi et al., 2010). In the taste buds of the murine tongue, TRPP2 is accumulated at the taste pore region where taste chemicals are detected. Coexpression of TRPP2 and PKD1L3 is necessary for the positioning of functional sour taste receptors at the cell surface, which is the case in the circumvallate and foliate papillae of the murine tongue (Huang et al., 2006; Ishimaru et al., 2006; Chandrashekar et al., 2009; Kawaguchi et al., 2010). In native circumvallate taste cells expressing TRPP2 the pH threshold for acid-induced increases in the intracellular calcium concentration is around 5.0 (Kawaguchi et al., 2010). Targeted ablation of TRPP2-expressing taste receptor cells results in a specific and total loss of sour taste transduction, whereas responses to sweet, umami, bitter and salty tastants remain unchanged (Huang et al., 2006). The relevance of PKD1, which occurs in the intestine (Geng et al., 1997), to GI function remains unexplored.

The TRPP2-expressing sour taste cells also appear to contribute to the taste of carbonation. While the principal CO2 taste sensor is carbonic anhydrase-4 (Chandrashekar et al., 2009), it remains to be determined whether TRPP2 plays a role in the downstream signaling of the carbonation taste.

2.11.1. Occurrence of TRPC in the alimentary canal

Distinct members of the classical or canonical TRP channels (TRPC) are widely distributed in the GI tract, the cellular sources being vagal afferent neurons, enteric neurons, smooth muscle cells, ICCs and epithelial cells. TRPC1, TRPC3, TRPC5 and TRPC6 have been localized to nodose ganglion neurons in which TRPC1 and TRPC6 are, at least in part, coexpressed with TRPV1 and TRPA1. The evidence for this colocalization rests on the finding that systemic pretreatment of rats with a neurotoxic dose of capsaicin causes a loss of TRPV1, TRPA1, TRPC1 and TRPC6 from vagal afferent neurons (Zhao et al., 2010). In the mouse DRG, mRNA for all 7 types of TRPC subunits (TRPC1–TRPC7) has been found, with TRPC1, TRPC3 and TRPC6 being the most abundant (Elg et al., 2007).

All 7 types of TRPC subunits have likewise been identified in the murine stomach (Lee et al., 2003; So & Kim, 2003). TRPC6 is expressed in human esophageal and gastric epithelial cells (Cai et al., 2009; Shi et al., 2009), while TRPC1 and TRPC5 have been localized to intestinal epithelial cells (Rao et al., 2006). In addition, TRPC4, TRPC6 and TRPC7 as well as splice variants of TRPC4 (TRPC4-α and TRPV-β) and TRPC7 are present in GI and vascular smooth muscle of the murine and canine alimentary canal, TRPC4 being the most abundant subunit (Walker et al., 2001, 2002; Torihashi et al., 2002; Lee et al., 2005; Kim et al., 2006; Tsvilovskyy et al., 2009). TRPC1, TRPC3, TRPC4, TRPC5 and TRPC6 mRNA have likewise been localized to circular muscle cells of the human esophageal body and lower esophageal sphincter, the levels of TRPC3 and TRPC4 mRNA being higher in the sphincter than in the body (Wang et al., 2003).

The guinea-pig enteric nervous system expresses TRPC1, TRPC3, TRPC4 and TRPC6, which are differentially distributed to the myenteric and submucosal plexuses (Liu et al., 2008). While TRPC1 abounds in myenteric neurons with cholinergic and nitrergic phenotypes as well as in submucosal cholinergic and noncholinergic secretomotor neurons, TRPC3 is restricted to submucosal neurons containing neuropeptide Y (Liu et al., 2008). TRPC4 and TRPC6 are likewise preferentially expressed by noncholinergic secretomotor neurons of the submucosal plexus (Liu et al., 2008).

2.11.2. Functional implications of TRPC in the digestive system

The abundant distribution of TRPC channels to vagal afferent neurons, enteric neurons, smooth muscle cells, ICCs and epithelial cells in the gut points to multiple functional implications of this class of TRP channels. However, their role in the neural control of digestive functions has not yet been examined. There is, however, increasing evidence that TRPC channels are involved in signal transduction within the epithelium and smooth muscle of the GI tract. In addition, distinct TRPC channels play a physiological role in GI epithelial cell homeostasis and in the development of malignancies.

The phenotype of TRPC1 knockout mice is characterized by a pronounced impairment of neurotransmitter-stimulated salivary gland fluid secretion, which suggests that TRPV1 is relevant to the secretory process in salivary gland acinar cells (Liu et al., 2007). In human colonic epithelial cells TRPC1 is involved in the signal transduction of the extracellular calcium-sensing receptor which is relevant to both intracellular calcium oscillation and inhibition of proliferation (Rey et al., 2010). Wounding of the mucosa causes translocation of stromal interaction molecule-1 to the plasma membrane, which in turn activates TRPC1-mediated calcium influx and promotes intestinal epithelial restitution (Rao et al., 2010). Induced TRPC1 expression facilitates, while inhibition of its expression by TRPC1-directed siRNA inhibits epithelial cell migration after wounding (Rao et al., 2006, 2010). In addition, induced TRPC1 expression sensitizes intestinal epithelial cells to apoptosis via calcium influx-mediated inhibition of nuclear factor-κB activity (Marasa et al., 2006, 2008).

TRPC1 and TRPC6 play a role in pancreatic and GI cancer development. In the pancreatic cancer cell line BxPc3, TRPC1 is a transducer of the action of tumor growth factor-β to stimulate cell motility, a process that is of relevance to tumor cell invasion (Dong H. et al., 2010). TRPC6, which can be activated both by receptor tyrosine kinases and GPCRs, appears to have a role in human esophageal and gastric cancer development (Cai et al., 2009; Shi et al., 2009). Treatment of the human gastric cancer cell lines AGS and MKN45 with the TRPC channel blocker SKF-96,365 arrests the cell cycle in the G2/M phase and suppresses the formation of gastric tumors in nude mice (Cai et al., 2009). Inhibition of TRPC6 likewise arrests the cell cycle of human esophageal squamous carcinoma cells in the G2 phase and inhibits tumor formation in nude mice injected with esophageal squamous carcinoma cells (Shi et al., 2009).

In the exocrine pancreas, TRPC3-mediated calcium influx is attributed an adverse role in the acute inflammation induced by supramaximal stimulation or exposure to toxic bile acids and palmitoleic acid ethyl ester (Kim M.S. et al., 2009). Consequently, deletion of TRPC3 has been found to attenuate experimentally evoked acute pancreatitis (Kim M.S. et al., 2009).

TRPC channels have been proposed to participate in intestinal pacemaker activity of ICCs which through electrical coupling drive slow wave activity in intestinal smooth muscle. The pacemaker current that initiates each slow wave derives from a Ca2+-inhibited, voltage-independent, nonselective cation channel in the ICCs. This channel exhibits properties similar to that reported for TRPC channels, notably TRPC4 (Walker et al., 2002). However, slow waves in intestinal smooth muscle remain unchanged in TRPC4 knockout mice (Lee et al., 2005). Thus, there is no conclusive evidence that TRPC channels, notably TRPC4, play a physiological role in the pacemaker activity of ICCs.

In contrast, distinct TRPC channels have been shown to operate as downstream transducers (effectors) of stimulated GPCRs such as muscarinic acetylcholine receptors. This mechanism has been suggested to apply to TRPC4, TRPC5 and TRPC6 which are activated by cholinergic stimulation of ICCs and smooth muscle in the murine stomach and intestine (Lee et al., 2003, 2005; Kim et al., 2006, 2007; Tsvilovskyy et al., 2009). Based on electrophysiological and pharmacological similarities, both TRPC4 (Lee et al., 2005) and TRPC5 (Lee et al., 2003) have originally been proposed to mediate the nonselective cation current evoked by carbachol in the murine stomach. A role of TRPC4 has been firmly established by the finding that the nonselective cation current evoked by muscarinic receptor stimulation is abolished by genetic deletion of TRPC4 (Lee et al., 2005; Kim et al., 2006).

Further studies using both TRPC4 and TRPC6 knockout mice indicate that TRPC4 underlies more than 80% of the muscarinic receptor-induced cation current in intestinal smooth muscle, while the residual current is mediated by TRPC6 (Tsvilovskyy et al., 2009). Following TRPC4 knockout, the carbachol-induced membrane depolarization and the atropine-sensitive contraction elicited by acetylcholine release from excitatory motor neurons are inhibited, these effects being aggravated by additional deletion of TRPC6. In addition, intestinal transit is slowed down in mice lacking TRPC4 and TRPC6 (Tsvilovskyy et al., 2009). From these studies it is evident that TRPC4 and TRPC6 couple muscarinic acetylcholine receptors to depolarization and contraction of intestinal smooth muscle. This is true for both M2 and M3 muscarinic receptors which use different second messenger mechanisms owing to their coupling to Gi/o and Gq/11 signaling pathways, respectively (Tsvilovskyy et al., 2009).

2.11.3. Association of TRPC6 with gastrointestinal disease

An implication of TRPC6 in human GI motor control is exemplified by the observation that a single nucleotide polymorphism in the promoter region of the TRPC6 gene and a missense variant in exon 4 of the TRPC6 gene may contribute to infantile hypertrophic pyloric stenosis (Everett et al., in press). TRPC6 appears to have a role in cancer development, given that there is a marked upregulation of TRPC6 expression in esophageal squamous cell carcinoma (Shi et al., 2009) and in epithelial cells of human gastric cancers (Cai et al., 2009).

2.11.4. Therapeutic potential of TRPC blockers in the digestive system

Cholinergic contraction of smooth muscle is a key determinant of GI motility. Recognition that TRPC4 and, to a minor degree, TRPC6 are transducers of the depolarizing action of M2 and M3 muscarinic acetylcholine receptors in GI smooth muscle raises the possibility to attenuate cholinergic motor effects while sparing other cholinergically controlled tissue functions. This opportunity would be particularly useful if pathological hypermotility in the GI tract is shown to be due to exaggerated TRPC4 and TRPC6 transduction (Ambudkar, 2009).

Blockade of TRPC1 and TRPC6 also holds considerable potential in GI oncology, given that TRPC1 may promote pancreatic cancer cell invasion (Dong H. et al., 2010) and TRPC6 seems to play a significant role in human esophageal and gastric cancer development (Cai et al., 2009; Shi et al., 2009).