The demonstration of increased mucosal nerve fibres immunoreactive to TRPV1 in human colonic biopsies from IBS patients, and their correlation with the degree of abdominal pain, may provide a putative basis of IBS symptoms. The increased TRPV1 nerve fibres were seen throughout the IBS group, with no difference when the group was subclassified by Rome II definitions into IBS-D, IBS-C and IBS-A.
Winston et al16
have presented animal data which strengthen the role of TRPV1 in visceral hypersensitivity. Acetic acid colonic irrigation was used to induce a state of chronic visceral hypersensitivity in neonatal rats, and they found increased TRPV1 expression in dorsal root ganglia containing colonic afferent neurons. Treatment with a TRPV1 antagonist ameliorated sensitivity, when used both in the neonates prior to the acetic acid colonic irrigation, and in the adult hypersensitive rats. Their work suggests that TRPV1 is probably important both in initiating the process of visceral hypersensitivity and in maintaining it. In agreement with this, we have reported increased levels of TRPV1 in patients with idiopathic rectal hypersensitivity and faecal urgency: in this group of patients, the increased TRPV1 nerve fibres were correlated with rectal distension and heat thresholds (Chan et al12
Our biopsies were obtained from IBS patients who were having either a colonoscopy or a flexible sigmoidoscopy. All control patients had a colonoscopy. We use different methods of bowel preparation in these two procedures; a phosphate enema for flexible sigmoidoscopy and Kleanprep for colonoscopy. There have been reports that bowel preparations may be responsible for histological changes17
with both enemas18
and colonoscopy sodium phosphate-containing bowel preparations.19 20
However, despite the use of these two different preparations, all biopsies were reported histologically as normal so there were no apparent changes induced due the type of preparation used. Furthermore, the changes seen with immunohistochemical markers for immune cells were similar through the IBS group. Only seven of the 23 IBS patients underwent a flexible sigmoidoscopy, so the majority had a full colonoscopy with identical bowel preparation to the controls. There were no statistically significant differences in any staining markers between the patients who underwent a flexible sigmoidoscopy and those who had a colonoscopy in the IBS group.
Although our two groups differed slightly in age, multivariate regression analysis showed that this did not affect the results. Our IBS group had an age and sex spread similar to the IBS group published recently by Barbara et al21
investigating mast cell mediators in IBS. Furthermore, an age-matched subgroup analysis revealed that the TRPV1-immunoreactive nerve fibres were still significantly increased in the IBS group. The number of IBS patients and controls in the overall group and in this subgroup were powered at the 80% level to detect differences in TRPV1 fibre number of 0.5 and 1.5 fibres/mm2
Although both control and IBS biopsies were categorised as normal on standard histology and mucosal endoscopic views were normal, the CD3 count and c-kit expression were both significantly higher in the IBS group. Our findings are consistent with results from previous studies.22–24
These have reported increased mast cell numbers in colonic biopsies,22 23 25
as we have noted. Barbara et al
found that mast cells which were proximal to nerves correlated with abdominal pain severity, suggesting a role for mast cells and their mediators in the altered sensorimotor pathophysiology of IBS.22 26
Our results also showed that mast cell numbers/c-kit staining (but not mast cell tryptase staining—data not shown) correlated positively with abdominal pain scores. Dong et al27
took biopsies from patients with IBS from different ileocolonic sites during colonoscopy and, using immunohistochemistry techniques, demonstrated an increase in mast cell numbers and SP-staining immunoreactive fibers, and also a spatial correlation between the two. Increased T lymphocytes have been reported in both infectious and non-postinfectious IBS.28–30
These results implicate that a low-grade inflammatory process in IBS could be involved in the pathogenesis of IBS, and that mast cell interaction with nerve fibres including SP-positive fibres may be important, in accord with our findings. From 6% to 17% of unselected IBS patients cite an infectious trigger for their onset of IBS symptoms, and prospective studies suggest an incidence of postinfectious IBS which follows a bacterial gastroenteritis of 4–31%.31
Other inflammatory cells have also been linked to postinfectious IBS; Dunlop et al
reported 20% increased numbers of enterochromaffin (EC) cells30
which contain serotonin 3 months after the initial infective gastroenteritis when compared with controls. When EC cells are triggered, they release serotonin which acts on nearby receptors and nerve endings. Furthermore there have been reports of increased cytokines in peripheral blood of IBS patients.32 33
These findings strengthen the role of an inflammatory process in triggering IBS. Although there is increasing evidence that sensorimotor dysfunction in IBS is likely to result from an interaction between mucosal immune cell mediators, nerve fibres and muscle layers, further studies into the triggering and predisposing mechanisms are needed. Various factors proposed include infective gastroenteritis, genetic causes, food allergies and alterations in gut microflora.34
Interestingly, TRPV1 expression has been reported on mast cells.35
Stander et al35
reported VR1 expression on dermal mast cells, suggesting a role in activation of these cells and perpetuation of inflammation.
TRPV1 is likely to be activated by the products of inflammation in IBS, and, through its upregulation, may contribute to symptoms including pain. There was evidence of nerve fibre sprouting in IBS, as PGP9.5 nerve fibres were increased. Inflammation-mediated upregulation of TRPV1 is well established, and has been shown to involve various mechanisms including nerve growth factor (NGF) and p38MAP kinase, along with sensitisation of TRPV1 by bradykinin B2 via intracellular enzymatic pathways. NGF production in peripheral tissues is enhanced by inflammation, and NGF is taken up and transported in a retrograde manner by nerve fibres to their cell bodies, leading to nerve sprouting and increased expression of TRPV1 and SP.36
Not only does NGF sensitise TRPV1 receptors to protons, enhancing their effect, but it also increases expression of TRPV1. Increased NGF, and recently trk A, expression has been reported in acute inflammatory bowel disease.37 38
Ji et al38
have shown that the increase in TRPV1 levels which occurs 12–24 h after inflammation is by an NGF-mediated p38 kinase pathway. TRPV1 activity is modulated by inflammatory mediators including bradykinin and prostaglandins, probably by cAMP-dependent protein kinase (PKA)- or protein kinase C (PKC)-mediated phosphorylation of the receptor.39
Generally, protein kinase-mediated phoshorylation of the TRPV1 receptor results in sensitisation, and dephosphorylation by protein phosphatases results in desensitisation.40
NGF immunostaining has been difficult to obtain in gut specimens, especially mucosal biopsies as we have used here. Future studies would be useful to assay NGF, with different processing of tissues.
Our findings, of increased total nerve fibres, and nerve fibres immunoreactive to TRPV1 and SP in IBS, may thus all be mediated via the effects of NGF. There are other mechanisms, however, that may also modulate TRPV1 function—Sugiura et al41
reported data from a mouse model suggesting that 5-hydroxytryptamine (5-HT, serotonin) receptor activation may enhance the responsiveness of the TRPV1 receptor to acid and temperature, and thereby contribute to peripheral sensitisation. It is thought that this response is mediated via 5-HT2
receptors. Serotonin signalling alterations seen in IBS may act in part via TRPV1 sensitisation.
TRPV1 activation produces an influx of calcium and sodium ions, along with release of neuropeptides (SP, CGRP). This in turn triggers and promotes the process of neurogenic inflammation. The oral TRPV1 antagonist JNJ 10185734 has been reported to attenuate dextran sulfate sodium (DSS)-induced colitis in mice, suggesting a possible role for TRPV1 in initiation or maintenance of inflammation in the gut.42
Other previous animal studies strengthen the view that TRPV1 is involved in inflammation and hyperalgesia. In a rat model of DSS colitis,43
neonatal animals chemically denervated of TRPV1 fibres by treatment with capsaicin, and those given a TRPV1 antagonist in established DSS colitis, were both protected from the damaging effects of DSS.
Co-localisation studies would give further information as to the origin of the nerve fibres under study, subject to the availability of good antibodies for TRPV1 and SP raised in different species. Such studies may also provide insight into the relationship of these markers to abdominal pain score. The two primary antibodies used in this study are both anti-rabbit (), thereby precluding co-localisation by immunohistochemistry.
There have been reports of a higher incidence of anxiety and depressive symptoms in IBS patients than in controls,44 45
but our IBS patients had similar BDI and HADS scores to those of the control group. Central factors are likely to play an important role in the generation of pain and visceral hypersensitivity as well as peripheral mechanisms.34
In this study we focused on the peripheral sensitising mechanisms involving low-grade inflammation and neuronal interactions in the generation of pain.
In summary, we present evidence that TRPV1 nerve fibres are increased in the mucosa of IBS patients. Our results provide a mechanism which may contribute to the pathophysiology of pain in IBS, and clinical trials with TRPV1 antagonists are warranted in this condition.