Colonic mucosal mediators from patients with irritable bowel syndrome excite enteric cholinergic motor neurons


Address for Correspondence
Giovanni Barbara, MD, Department of Clinical Medicine, St. Orsola-Malpighi Hospital, Via Massarenti, 9 – Building #5, I-40138 Bologna, Italy.
Tel: +39 051 6364103; fax: +39 051 345864; e-mail:


Background  Mediators released in the mucosal milieu have been suggested to be involved in visceral hypersensitivity and abdominal pain in patients with irritable bowel syndrome (IBS). However, their impact on myenteric neurons remains unsettled.

Methods  Mucosal biopsies were obtained from the descending colon of patients with IBS and controls. Mucosal mast cells were identified immunohistochemically. The impact of spontaneously released mucosal mediators on guinea pig electrically stimulated longitudinal muscle myenteric plexus (LMMP) preparations was assessed in vitro by means of selective receptor antagonists and inhibitors.

Key Results  Patients with IBS showed an increased mast cell count compared with controls. Application of mucosal mediators of IBS to LMMPs potentiated cholinergic twitch contractions, an effect directly correlated with mast cell counts. Enhanced contractions were inhibited by 50.3% with the prostaglandin D2 antagonist BW A868C, by 31.3% and 39% with the TRPV1 antagonists capsazepine and HC-030031, respectively, and by 60.5% with purinergic P2X antagonist pyridoxalphosphate-6-azophenyl-2′,4′-disulfonic acid. Conversely, the serotonin1-4, histamine1-3, tachykinin1-3 receptor blockade, and serine protease inhibition had no significant effect.

Conclusions & Inferences  Colonic mucosal mediators from patients with IBS excite myenteric cholinergic motor neurons. These effects were correlated with mast cell counts and mediated by activation of prostanoid receptors, TRPV1, and P2X receptors. These results support the role of mucosal inflammatory mediators and mast cell activation in altered motor function of IBS.




bowel disease questionnaire


constipation-predominant irritable bowel syndrome


diarrhea-predominant irritable bowel syndrome


healthy controls


inflammatory bowel disease


irritable bowel syndrome


longitudinal muscle myenteric plexus




pyridoxalphosphate-6-azophenyl-2′,4′-disulfonic acid


standard deviation


transient receptor potential cation channel, subfamily V, member 1


The irritable bowel syndrome (IBS) is a common functional gastrointestinal (GI) disorder in which abdominal pain or discomfort is associated with defecation or a change in bowel habit, and with features of disordered defecation.1 It has been hypothesized that intestinal low-grade immune activation, in the absence of overt inflammation, participates in altered intestinal sensory and motor function of IBS.2 Accordingly, IBS symptoms may be experienced by patients with inflammatory bowel disease (IBD) in remission 3 and following an acute bout of infectious gastroenteritis.4 Furthermore, an increased number of immune cells can be identified in the colonic,5 ileal,6 and jejunal7 mucosa of patients with IBS compared with healthy controls (HC). Attention has been directed to mucosal mast cells because of the known effects of their mediators on the physiology of the enteric nervous system.8 Among the wide array of mediators released by mast cells, histamine, proteases (e.g., tryptase), and prostanoids [e.g., prostaglandin (PG) E2] are known to regulate smooth muscle cell contractility, myenteric neuronal function, and extrinsic sensory afferents.8

In previous work, we and others demonstrated that soluble factors spontaneously released by the colonic mucosa of patients with IBS excited rat nociceptive visceral sensory pathways9,10 and human submucosal neurons.11 Although many mediators (e.g., bacterial, immune, and enteroendocrine) could be involved in such changes, these studies identified proteases, histamine, and serotonin (5-hydroxytryptamine – 5-HT) as the major components affecting neuron activation. Interestingly, there is an increased number and state of activation of mucosal mast cells in the descending colon of patients with IBS5 and a close proximity of mast cells to mucosal colonic innervation which correlated with the frequency and severity of abdominal pain. Taken together, these data imply the involvement of mast cell mediators in the observed physiological changes. Others were unable to detect increased mast cell numbers in the colonic mucosa,9,12,13 which is reasonable for a heterogeneous condition such as IBS and possibly related to regional, dietary, genetic, experimental differences among studies. Interestingly, one negative study showed decreased colonic mast cell numbers in spite of a markedly positive effect of mast cell stabilizers on both visceral hypersensitivity and symptoms, confirming rather than discarding an implication of mast cell in IBS.10

Although there is limited availability of deeper intestinal tissue (myenteric plexus and ‘muscolaris propria’) of patients with functional disorders, a recent study suggest that in patients with slow transit constipation mast cells may also be increased in the external muscle layer.14 This suggests that mast cell and other mediators can impact on smooth muscle contractility or on intrinsic motor neurons. Recent data showing an implication of mast cells in postoperative ileus further supports the implication of immune activation and mast cells in gut motility disorders.15

Based on these observations, we assessed the number of colonic mast cell infiltrating the colonic mucosa of patients with IBS and the impact of mediators from the mucosa on cholinergic motor responses evoked by electrical field stimulation in guinea pig ileal longitudinal muscle myenteric plexus (LMMP) preparations.

Patients and Methods


Patients with IBS were all seen in the Department of Clinical Medicine of the University of Bologna and diagnosed according to the Rome II criteria.16 To maximize possible differences related to bowel habit on results, we deliberately decided to include only patients with diarrhea- (D-IBS) or constipation-predominant (C-IBS) IBS in the study. HC were recruited by public advertisement and included in the study after thorough exclusion of GI symptoms. In particular, in HC we excluded the presence of the following symptoms in the last 12 months: abdominal discomfort or pain, bloating, alteration of bowel habit. Exclusion criteria for both HC and patients with IBS included the use of non-steroidal anti-inflammatory drugs, corticosteroids, antihistaminics, and mast cell stabilizers, major abdominal surgery, celiac disease (excluded by detection of antitransglutaminase and antiendomysial antibodies), allergic diseases, including asthma (family and personal history and specific anti-IgE antibodies) and other organic or severe psychiatric disorders as assessed by history taking and appropriate consultations and laboratory tests. Patients and HC gave written informed consent and the study protocol was approved by the local Ethic Committee (no. 64/2004/O/Sper and no. EM 14/2006/O) and conducted in accordance with the Declaration of Helsinki.

All participants underwent left colonoscopy after cleansing of the distal colon with a 500 mL water enema performed the evening before and the morning of the procedure. In all cases, we obtained six mucosal biopsies from the proximal descending colon. Two biopsies were fixed in buffered 10% formalin and processed for H&E histology to exclude microscopic colitis and to perform immunohistochemistry for mucosal mast cells. Four biopsies were used to obtain mucosal mediator release (see below).

Symptom questionnaire

Each patient completed a modified version of the Bowel Disease Questionnaire (BDQ) to evaluate symptoms.17 An Italian version of the questionnaire has been previously used in IBS studies in our country.5 Patients were asked to score frequency and severity of their symptoms over the last 2 weeks before interview, as previously described.5 We assessed the following IBS symptoms: abdominal pain/discomfort and bloating. The severity and the frequency of these symptoms were graded 0–4, as previously described.5

Mast cell counts

Biopsy specimens were fixed in buffered 10% formalin and processed for either H&E histology or immunohistochemistry. All histological sections were evaluated by an expert pathologist who was unaware of the diagnosis, for exclusion of overt mucosal inflammation or microscopic colitis. For immunohistochemistry, following antigen unmasking, randomly selected sections from the two biopsies were incubated with mouse monoclonal antibodies directed against tryptase (mast cell marker; diluted at 1 : 2000; Dakopatts, Glostrup, Denmark) overnight at 4 °C, as previously described.5 Quantification of mast cells was performed on sections immunostained for tryptase with a Leitz Dialux microscope in blind fashion using a computer-assisted analysis system (Cytometrica@ software; C&V, Bologna, Italy), as previously detailed.5

Mucosal mediator release

Spontaneous release of mucosal mediators was obtained using a previously described method.5 Briefly, upon removal, four mucosal biopsies were rapidly immersed in hard plastic tubes containing 1 mL of Hank’s solution (37 °C) and continuously oxygenated with a mixture of 95% O2 and 5% CO2. After 25 min incubation, samples were centrifuged at 200 g for 10 min, and 400–600 μL of the bathing solution was removed and stored at −80 °C until assayed to assess their effect in the guinea pig LMMPs (see below). At the end of the release experiment, biopsies were blotted and weighed (tissue weight: 5.8 ± 3.7 mg vs 7.2 ± 3.4 mg, = 0.315, for HC and IBS, respectively).


Male albino guinea pigs (250–400 g; G. Bettinardi; Momo, Novara, Italy) were used. Care and handling of the animals were in accordance with all NIH recommendations for the human use of animals. All experimental procedures were reviewed and approved by the Animal Use Committee of the University of Pavia in accordance with animal care European Union directive 86/609. The number of animals used was kept to the minimum necessary for a meaningful interpretation of the data.

Tissue preparation

Segments of the distal ileum were dissected from guinea pig and then immersed in standard Tyrode solution. Longitudinal muscle-myenteric plexus were prepared by teasing the longitudinal muscle with the intact myenteric plexus from the underlying circular muscle.18 For the analysis of electrically induced neurogenic contractions (twitch contractions), 4 cm long strips were folded in half and mounted in organ baths containing 5 mL of Tyrode solution (bubbled with a mixture of 95% O2 and 5% CO2) at 37 °C, under tension of 5 mN. Isometric contractions were recorded with a force-displacement transducer (Grass FT03 Harvard Instruments, South Natick, MA, USA).

Experimental design

Following a 45-min equilibration period, strips were stimulated via two platinum electrodes with a Grass S44 stimulator (Harvard Instruments, South Natick, MA, USA). Submaximal contractions were evoked by rectangular pulses with the following parameters: 0.1 Hz, 25–35 V, 0.5-ms pulse duration. Under these conditions, electrically evoked contractions are nerve-mediated and cholinergic in nature.18 After at least 10 min of reproducible twitch contractions, a single concentration of the biopsy supernatant (volume range: 100–150 μL) was administered and then, once reached the maximum effect, the tissue was repeatedly washed to completely remove the biopsy supernatant. The volume of supernatant was normalized to the weight of biopsies to add similar concentrations of soluble mediators. A second administration of the same volume of each supernatant was tested in the presence of one of the following antagonists: (i) the prostanoid D2 receptor antagonist BW A868C, used at a concentration (3 μmol L−1) known to specifically block all the prostanoid receptors including the EP2 subtype,19,20); (ii) the antagonists at transient receptor potential cation channel, subfamily V, member 1 (TRPV1), HC-030031, and capsazepine, both used at 1 μmol L−1; (iii) the purinergic P2X antagonist pyridoxalphosphate-6-azophenyl-2′,4′-disulfonic acid (PPADS) tetrasodium salt, used at 10 μmol L−1; (iv) the serotonin 5-HT1/2, 5-HT3, and 5-HT4 receptor antagonists ketanserin, ondansetron, and RS-39604, each used at 1 μmol L−1; (v) the selective histamine H1, H2, and H3 receptor antagonists mepyramine, ranitidine, and thioperamide, each used at 1 μmol L−1; (vi) the selective tachykinin NK1, NK2, and NK3 receptor antagonists RP67580, used at 0.1 μmol L−1, MEN11420, used at 1 μmol L−1, and SR142801 used at 0.1 μmol L−1; and (vii) the serine protease inhibitor nafamostat (FUT-175) used at 1 μmol L−1.

The samples were coded with a progressive number and the researchers performing the in vitro experiments were unaware of sample allocation until the end of the set of experiments for each drug. The functional experiments were carried out by researchers who were unaware of results of mast cell density for each biopsy.

Antagonists were added in the bath and left in contact with LMMPs for 30 min before the addition of the biopsy supernatant. FUT-175 was added to the biopsy supernatants for 30 min before addition of the supernatants to the organ baths. In ancillary experiments, we have shown that vehicles at the volumes/concentrations used in the present study (see below) did not affect twitch amplitude. Supernatant-induced change in the amplitude of electrically induced contractions of LMMP was expressed in mN as difference with the basal twitch amplitude. Signals were recorded using a PowerLab data acquisition system (Analogue Digital Instruments, Oxford, UK) and analyzed using PowerLab Chart v4.1.1 software.

Solutions and drugs

The Tyrode solution had the following composition (mmol L−1): NaCl, 136.9; KCl, 2.7; CaCl2, 1.8; MgCl2, 1.04; NaH2PO4, 0.4; NaHCO3, 11.9; glucose, 5.5. Ondansetron hydrochloride was from Glaxo-Smithkline (Uxbridge, UK), RS-39604, mepyramine maleate, thioperamide maleate, RP67580, PPADS, and capsazepine were from Tocris (Northpoint, UK), nafamostat (FUT-175) was from Calbiochem (Darmstadt, Germany), MEN11420 was from Menarini Pharmaceuticals (Florence, Italy), SR142801 was from Sanofi Research (Montpellier, France), Hank’s solution, ketanserin, ranitidine, and BW A868C were purchased from Sigma Chemical (St. Louis, MO, USA), HC-030031 was purchased from Chiesi. All drugs were dissolved in distilled H2O and administered in volumes not exceeding 1% v/v of the final bath volume with the exception of BW A868C, RP 67580, SR 142801, and capsazepine which were dissolved in dimethyl-sulphoxide and then diluted in distilled H2O. Stock solutions of ketanserin were prepared in 1 : 100 v/v ethanol/H2O. Further dilutions were in distilled H2O.

Data expression and statistical analysis

Data are reported as mean values ± standard deviation (SD). The simple contrast analysis of variance (anova) was used for the comparisons of mast cell counts and supernatant-induced changes on LMMPs twitch responses among the groups of subjects included in the study (HC, IBS-C, and IBS-D). The effect of gender on the area of lamina propria occupied by total mast cells was assessed by means of the one-way anova. The difference in mN between the enhancing effects of IBS supernatants on LMMPs twitch responses and the enhancing effects of IBS supernatants in the presence of antagonists/inhibitors was evaluated by means of multivariate analysis of variance. Correlations were analyzed using the Spearman rank correlation test. The Fisher’s exact test was applied to dichotomous data. The SPSS/PC+ (Version 13.0 for Windows; SPSS Inc., Chicago, IL, USA) statistical package running on a personal computer was used to analyze data. Two-tailed P-values less than 0.05 were considered statistically significant.



In the present study, we screened 78 consecutive patients with IBS. Of these, 8 were excluded because further investigations revealed the presence of organic, metabolic, or systemic diseases, 9 declined to undergo the required investigations, and 24 with alternating bowel symptoms were not included in the study. The remaining 37 patients with IBS (aged: 21–70 years; mean 35.2 years; 27 females) entered the study. Results were compared to those obtained in 11 HC (aged: 20–30 years; mean 24.9 years; 6 females). None of HC discontinued the study and all the results were included. All patients complained of abdominal pain/discomfort; 16 were D-IBS and 21 C-IBS; the most frequently associated symptom was bloating (91.9%). Patients’ clinical characteristics are reported in Table 1.

Table 1.   Characteristics of the study patients
Total number of patients37
Age (years)35.2 ± 10.3
Gender (% females)73
Abdominal pain scores
 Severity (0–4)1.95 ± 0.79
 Frequency (0–4)2.26 ± 0.90
Bloating scores
 Severity (0–4)1.95 ± 0.84
 Frequency (0–4)2.57 ± 1.14
Bowel habit (%)

Mast cell counts

Biopsies from all the enrolled patients with IBS (n = 37) and controls (n = 11) were processed for immunohistochemistry and quantitative experiments relative to mast cell counts. Compared to HC, the mean area of lamina propria occupied by tryptase positive mast cells was increased by 103% in patients with IBS (3.56 ± 1.53%vs 7.22 ± 2.54%; < 0.001; Fig. 1). Although the increase of the mean area occupied by tryptase positive mast cells in patients with D-IBS was greater than that observed in patients with C-IBS, this difference did not reach statistical significance (7.49 ± 3.39%vs 6.95 ± 1.35%; = 0.475). The lamina propria area occupied by mast cells was not different in male in comparison with female patients with IBS (7.50 ± 2.30%vs 7.12 ± 2.41%; = 0.660).

Figure 1.

 Representative photomicrographs showing tryptase positive mast cells in the colonic mucosa of a healthy control (A) and a patient with IBS (B). Mast cells were identified by immunohistochemistry for tryptase. Note the higher number of positive mast cells in the patient with IBS in comparison with the control (bar = 20 μm). Individual data plots, allocated by gender and bowel habit, are showed in panel (C). Compared with HC, the mean area of lamina propria occupied by tryptase positive mast cells was significantly increased in patients with IBS (*< 0.001). Empty circle: male HC; empty triangle: female HC; gray circle: male IBS-C; black circle: male IBS-D. Gray triangle: female IBS-C; black triangle: female IBS-D.

Effects of mucosal IBS mediators on twitch contractions in LMMPs

The addition of the medium (Hanks buffer) used to collect mucosal mediators from IBS and HC biopsies to the organ baths had no effect on electrically stimulated contractions (Fig. 2A). Seventy-nine percent of IBS supernatants enhanced cholinergic twitch contractions compared to 58% evoked by control supernatants. Compared to HC supernatants, IBS supernatants evoked a significantly greater increase in twitch height (2.44 ± 1.26 mN and 4.82 ± 2.88 mN, P = 0.025) (Fig. 2). The difference between the potentiation of twitch amplitude observed for D-IBS and C-IBS supernatants did not reach statistical significance (= 0.313) (Fig. 2).

Figure 2.

 Representative tracing showing the effects of Hanks and supernatants from a healthy control and a patient with IBS, on the electrically induced submaximal cholinergic contractions (twitch) in the longitudinal muscle-myenteric plexus (LMMPs) preparations of the guinea pig ileum (A). Enhancing effect of HC, constipation- (C-IBS) and diarrhea-predominant (D-IBS) supernatants on the electrically induced twitch contractions in LMMPs preparations (B). Values represent means ± SD of 8, 21, and 16 preparations, respectively. *P < 0.01 compared with HC; no significance between C- and D-IBS groups.

The enhancement of twitch height evoked by IBS supernatants was significantly reduced by 50.3% (P = 0.022, n = 9) by the prostanoid receptor antagonist BW A868C (3 μmol L−1), by 31.3% (P = 0.011, n = 5) by the TRPV1 antagonist capsazepine (1 μmol L−1). Similarly to capsazepine, the selective antagonist of TRPV1 HC-030031 (1 μmol L−1) reduced increased twitch contraction by IBS supernatants by 39% (P = 0.011, n = 10). The P2X receptor antagonist PPADS (10 μmol L−1) reduced increased twitch contraction by 60.5% (< 0.001, n = 9). Conversely, blockade of serotonin 5-HT1/2,3,4 (n = 5), histamine H1-3 (n = 5), tachykinin NK1-3 (n = 8) receptors was ineffective. Antagonism of serine proteases with FUT 175 1 μmol L−1 was also ineffective (n = 7; Fig. 3A–H).

Figure 3.

 Characterization of the neuroactive components contained in the IBS supernatants. Open columns represent the enhancing effects of IBS supernatants on LMMPs twitch responses; filled columns the enhancing effects of IBS supernatants in the presence of the prostanoid receptor antagonist BW A868C (3 μmol L−1, n = 9, = 0.022) (A), the serine proteases antagonist FUT 175 (1 μmol L−1, n = 7, NS) (B), the selective vanilloid receptor, TRPV1, antagonist HC-030031 (1 μmol L−1, n = 10, = 0.011) (C), the TRPV1 antagonist capsazepine (1 μmol L−1, n = 5, = 0.011) (D), the P2X receptor antagonist PPADS (10 μmol L−1, n = 9, < 0.001) (E), the combined blockade of serotoninergic (5-HT) 1/2, 3, and 4 receptors by ketanserine, ondansetron, and RS-39604 (5-HT mix, each used at 1 μmol L−1, n = 5, NS) (F), the combined blockade of histamine (H) 1, 2, and 3 receptors by mepyramine, ranitidine, and thioperamide (His mix, each used at 1 μmol L−1, n = 5, NS) (G), and the combined blockade of tachykinin (NK) 1, 2, and 3 receptors by RP67580, MEN11420, and SR142801 (NK mix, used at 0.1–1–0.1 μmol L−1, respectively, n = 8, NS) (H). Values represent means ± SD.

Correlation between twitch contractions and mast cell counts

Mast cell counts in the colonic mucosa of patients with IBS were significantly correlated with the twitch enhancement evoked by IBS supernatants (rs = 0.266; < 0.049; n = 37). In contrast, there was no correlation between mast cell counts and twitch enhancement evoked by HC supernatants (rs = −0.036; = 0.932; n = 8).


In the present study, we have provided evidence that colonic mucosal mediators from patients with IBS enhanced electrically stimulated twitch contractions in the guinea pig, which was indicative of increased excitement of myenteric cholinergic motor neurons. The increase in twitch contractions caused by IBS supernatants was mediated by activation of purinergic P2X receptors, prostanoid receptors, and TRPV1. Protease activated receptors, 5-HT1/2-4, histamine1-3, and tachykinin1-3 did not play a significant role. Twitch increase was significantly correlated with mast cell counts in the colonic mucosa of patients with IBS. Taken together, these results support the concept that the mucosal milieu, at least in part via intestinal immune activation and possibly through mast cell infiltration, may contribute to colonic motor dysfunction of patients with IBS via interaction with myenteric neurons.

Although different mediators could be involved in the effect seen on twitch contractions,5,10 we speculate that at least part of these effects can be ascribed to immune system-mast cell activation. In previous studies, human intestinal mast cells mediators exerted excitatory action on human submucosal neuron.21 In addition, this effect was similar to that evoked by supernatants from IBS mucosal biopsies.11 In the present study, we showed that mast cell infiltration was significantly correlated with twitch enhancement caused by IBS but not control supernatants. We have previously shown a 150% significant increase in the percentage of degranulated mast cells (as assessed by quantitative electron microscopy) in patients with IBS as compared with controls.5 Thus, an increased activation of mast cells would release higher amounts of mediators in the milieu which in turn may affect nerve function. In the GI tract of various species including humans, mast cells are closely apposed to nerves and there is a positive significant correlation between nerve-to-mast cell proximity and rate of activation. These features suggest that nerves may activate mast cells via neurotransmitter input and provide the anatomical basis for the communication between mast cells and enteric nerves.5,8 Serine proteases have been suggested as important mediators of altered bowel physiology in patients with IBS.22 The fact that trypsin and tryptase transcription in colonic mucosal samples is increased,9 suggests that the origin of proteases is from host tissues, and likely involves mast cells. On the other hand, these results do not completely discredit the possibility that some of the excitatory effects seen here were evoked by bacterial signals contaminating our mucosal biopsies.23 Further work has to be done to exactly decipher the implication of mast cells and their mediators obtained from patients with IBS on myenteric plexus.

In our experimental model, we demonstrated that the twitch contractions were reduced by the P2X antagonist PPADS, suggesting that ATP contained in IBS supernatants excited cholinergic motor neurons via P2X receptors. ATP may act as a neurotransmitter released in the synaptic space from nerve terminals or as an autocrine or paracrine signaling molecule from different cellular sources, including immune cells.24 Interestingly, in a co-colture in vitro system, ATP released from stimulated mast cells mediated nerve-fiber activation. Mast cell-induced neurite activation was blocked by the purinergic receptor antagonist PPADS.25 At the site of tissue injury or inflammation, damaged or dying cells release large amounts of ATP, which can interact with P2X receptors on neighboring cells.24 Of interest for the present study, P2X receptors are expressed by both extrinsic sensory and enteric nerves and by immunocytes, including mast cells.26,27 P2X agonists activate immune cells and evoke the release of cytokines and other signaling molecules. These immune factors may activate sensory and enteric nerves.28 Indeed, ATP is involved in inflammatory nociception24,29 and afferent hypersensitivity.30 In addition, P2X receptors evoke fast excitatory postsynaptic potentials in guinea pig myenteric S neuron (i.e., interneurons and motor neurons)31 and activation of AH neurons.32

Potentiation of twitch contractions evoked by IBS supernatants were significantly inhibited by the TRPV1 antagonists, HC-030031 and capsazepine. TRPV-1 is mainly expressed by sensory neurons and activated by capsaicin, heat (>43 °C), acid (pH < 5.9), and inflammatory mediators.33 Of interest, Akbar et al. demonstrated an increase of TRPV1 immunoreactive nerve fibers in colonic mucosa of patients with IBS, which was associated with a low-grade inflammatory response.34 The inhibitory effect of HC-030031 and capsazepine may be related to the action of different mediators contained in supernatants of patients with IBS on TRPV1 receptors expressed by sensory nerves which in turn could release excitatory neuropeptides acting on cholinergic motorneurons. TRPV1 activation leads to local release of sensory neuropeptides including calcitonin gene-related peptide and substance P.35 These biologically active substances largely act by stimulating intrinsic myenteric neurons that release acetylcholine and, to a lesser extent, tachykinins as final mediators.36 In guinea pig intestinal muscle, capsaicin induces tetrodotoxin-sensitive acetylcholine release from myenteric nerves.37 Furthermore, guinea pig intestinal contractions to capsaicin are strongly inhibited by atropine suggesting that the capsaicin-induced release of substance P from sensory nerves in the guinea pig intestine activates enteric excitatory cholinergic nerves.38

In this study, we have shown a participation of PG receptors in the enhancement of cholinergic contractions evoked by IBS supernatants. This is in keeping with our previous demonstration of increased production of PGs (i.e., PGE2) in the colonic mucosa of patients with IBS.10 Prostaglandins evoke slow excitatory postsynaptic potential-like excitatory actions in enteric neurons.39 PGE2 may act as an excitatory neuromodulator through a direct action on neurons in the myenteric plexus of the guinea pig ileum.40 A similar direct action of PGE2 and PGD2 on submucosal neurons in the guinea pig colon has been reported.41 Our data showing that the twitch contractions were reduced by BW A868C, indicated that PGs contained in IBS supernatants activated cholinergic motor neurons via prostanoid receptors, including the EP2 subtype. In fact at the concentration used in the present study (3 μmol L−1), BW A868C is known to block all the prostanoid receptors including the EP2 subtype.19

Although histamine, 5-HT, tachykinins, and proteases are known to activate enteric neurons,39 the contribution of these mediators in our experiments was negligible because the 5-HT1/2-4, H1-3, and NK1-3 specific receptor antagonists and protease inhibitor have no significant effect in our experimental model. Although there is evidence for elevated levels of histamine, proteases, and 5-HT in the mucosa of patients with IBS,5,10,42 it is possible that in the present animal model other mediators (such as ATP, TRPV1 activators, and PGs) may play a prominent role. Furthermore, myenteric plexus neurons may respond differentially to submucosal neurons to IBS supernatants. Likely, this may be related to a differential expression profile of receptors in the submucous compared with the myenteric plexus. In keeping with this hypothesis, a preliminary report showed a differential spike discharge of submucous vs myenteric neurons as measured with a fast neuroimaging technique.43

We observed that not all supernatants increased cholinergic twitch contractions, but the proportion of IBS samples eliciting a response was significantly higher from IBS compared with HC samples. In addition, the increase of twitch height was significantly higher after the administration of IBS supernatant compared with controls, independently of the patients’ bowel habit. These results are in line with previous reports showing a similar degree of immune cell infiltration and mediator release in patients with either D-IBS or C-IBS.5,9,11,34,44 Similarly, previous studies have found that IBS supernatants increased spike discharge in visceral afferents independently of IBS subtype.9,10 Thus, our results showed that neuron activation by IBS supernatant was unrelated to bowel habit. A mechanistic interpretation as to how a similar pattern of mediators is involved in such contrasting bowel habit is currently lacking. However, the inflammatory hypothesis leading to contrasting bowel habits is in line with evidence that in IBD both diarrhea and constipation can occur.45 In addition, it is likely that other contributing factors are involved in the determination of bowel habit changes, including, duration of the disease, involvement of other factors such as psychological features, and pathophysiological factors (e.g., release of serotonin and neuroplastic changes).

Many mediators released by immune/inflammatory cells interact with receptors on enteric neurons. These receptors represent attractive therapeutic targets in IBS.46 Based on our findings, it can be hypothesized that the activation of enteric neurons is a result of a synergic action of several mucosal mediators, including ATP, PGs, and TRPV1 activators. This aspect has to be taken into account in the relative inefficacy of previous single-receptor based therapies developed for IBS.47 On the other hand, we propose that a more general approach targeting immune activation48 and mast cells in particular49 may turn to be of value in targeting subsets of patients with IBS. In line with this hypothesis, a study by Klooker et al.12 showed that the mast cell stabilizer ketotifen reduced visceral hypersensitivity and improved symptoms in patients with IBS. Further studies on this line are now required to confirm these data.

Although patients with IBS were slightly, but significantly, older than HC, this is unlikely a confounding factor. In support of this concept, previous studies in the healthy large bowel showed that mucosal mast cell counts decline, rather than increase, with age.50 The complex nature of the experiments carried out in the present work prejudiced study of large numbers of patients and particularly HC. We argue that higher n values would add further insight since (i) we found significant differences between HC and patients with IBS, and (ii) the standard deviation of healthy samples on neurophysiology is rather small and comparatively much smaller than that found in IBS suggesting that a smaller number of HC vs IBS would be acceptable. In contrast with our previous data showing higher mast cell counts in female compared with male IBS, here we found no gender related differences, however, this discrepancy was interpreted in light of the fact that the n values of the present study do not allow to explore gender-related differences.

In conclusion, we have shown that mediators from colonic mucosa of IBS patients excite enteric cholinergic motor neuron in vitro which correlates with mast cell count, but is independent of bowel habit. We also demonstrated that these effects were mediated by activation of P2X receptors, prostanoid receptors, and TRPV1. Our results provide novel mechanistic insight into the mechanism underlying altered motor function in IBS.


This study is dedicated to the memory of Marcello Tonini, who coordinated this work with great enthusiasm until his premature death.


This work was supported by the Italian Ministry of Education, University and Research [no. 2002052573 (FDP, GB, and MT) and no. 2007Z292XF (FDP, GB, RC, and MT], from the University of Bologna (GB, RG, and RC). GB is a recipient of grants from ‘Fondazione Cassa di Risparmio di Bologna,’ Bologna, Italy and IMA S.p.A., Bologna, Italy.


No competing interests declared.

Author Contribution

MT and GB planned the study, designed the protocol, and contributed to the writing of the manuscript; BB and RV performed the research, acquired experimental data and contributed with statistical analysis. CC and GB performed the colonoscopies and wrote the article. LZ, GD, and VV were involved in immunohistochemistry studies and collected all data. RDG, VS, RC, and FDP contributed to drafting and revising the study design and manuscript.