SEARCH

SEARCH BY CITATION

Keywords:

  • enteric nervous system;
  • irritable bowel syndrome;
  • neuro–immune interactions

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Funding
  8. Disclosure
  9. Authors’ Contribution
  10. References

Background  We previously showed that colonic mucosal biopsy supernatants from patients with irritable bowel syndrome (IBS) activate neurons of the human submucous plexus, an area with densely packed immune cells. Based on the concept that mucosa-nerve signaling is altered in IBS, we tested in this study whether the nerve sensitizing effect of IBS mucosal biopsy supernatants is more prominent in the submucous than myenteric plexus.

Methods  Fast neuroimaging with the voltage-sensitive dye Di-8-ANEPPS was used to record activity of guinea-pig submucous and myenteric neurons after application of constipation (C)- and diarrhea (D)-IBS supernatants (three each) and four supernatants from healthy control subjects. Results are based on recordings from 4731 neurons.

Key Results  Control supernatants did not evoke significant responses in submucous or myenteric neurons. In contrast, all IBS supernatants evoked a significant spike discharge (median 3.6 Hz) in 46% of submucous neurons. This activation was significantly stronger than in the myenteric plexus where even twice the amount of supernatants evoked a lower spike frequency (median 2.1 Hz) in only 8.5% of neurons. Pharmacological studies revealed serotonin, histamine, and proteases as components mediating neuronal activation. Individual application of these components revealed that only serotonin evoked a significantly stronger activation of submucous compared with myenteric neurons.

Conclusions & Inferences  Direct neuronal activation by IBS mucosal biopsy supernatants is primarily a feature of submucous rather than myenteric neurons. This is associated with a stronger excitation of submucous neurons by serotonin. The plexus-specific effects support the concept that altered mucosa-nerve signaling underlies disturbances in IBS.


Abbreviations:
ENS

Enteric Nervous System

fEPSP

fast Excitatory Post-synaptic Potential

H

Histamine

HC

healthy controls

IBS (C; D)

Irritable Bowel Syndrome (Constipation Diarrhea)

IQR

Interquartile range

MSORT

Multi-site Optical Recording Technique

PAR

protease-activated receptors

5-HT

5-hydroxytryptamine or Serotonin

Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Funding
  8. Disclosure
  9. Authors’ Contribution
  10. References

Altered immune-nerve or mucosa-nerve signaling is thought to be one of the major contributing factors in the pathophysiology of irritable bowel syndrome (IBS).1–3 This concept is supported by several findings from studies reporting increased density and/or even more important, increased reactivity of intestinal mucosal mast cells,3–7 lymphocytes,8–11 and enterochromaffin (EC) cells12–14 in IBS specimens which may contribute to changes in gut physiology. Indeed, supernatants from peripheral blood mononuclear cells from patients with diarrhea predominant IBS (D-IBS) activate visceral afferents.15 Moreover, colonic mucosal biopsy supernatants from D-IBS and constipation predominant IBS (C-IBS) also activate visceral afferents1,3 and enteric neurons involving immune mediators, such as histamine and proteases and serotonin.2 Nerve sensitization is independent of IBS subtype as has been observed following application of mucosal biopsy supernatants of both C- and D-IBS patients to human submucous neurons which are located in an area of densely packed immune cells.16,17 About 70% of intestinal mucosal mast cells are in direct contact with subepithelial nerves, and another 20% are located within 2 μm.18 In addition, cell bodies of submucous neurons are in closer proximity to epithelial cells and intestinal EC cells than myenteric neurons which are located between the longitudinal and circular muscle layers. It is therefore plausible that mediators released from epithelial cells or immune cells may have direct access to cell bodies of submucous rather than myenteric neurons.

Considering the close proximity of submucosal neurons and immune cells17 one would expect that a neuronal activation by mucosal factors or mediators is predominantly the feature of the submucous rather than the myenteric plexus. To study this question we used guinea-pig submucous or myenteric plexus preparations to analyze in detail neuronal effects of mucosal biopsy incubation supernatants from patients with IBS. In addition, we tested specifically the effects of the single components histamine, proteases, and serotonin on either plexus to identify possible different effects. The main finding of the present study was the markedly higher sensitivity of submucous compared with myenteric neurons to mucosal biopsy supernatants from patients with IBS.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Funding
  8. Disclosure
  9. Authors’ Contribution
  10. References

Study participants and biopsy supernatants

All protocols and procedures performed with human subjects were approved by the appropriate Human Research Ethics Committees at University of Bologna and Technische Universität München (Prot.n. 906/2006 and Prot.n 1925/2004). Informed consent was obtained in writing from all subjects and the studies conformed to the standards set by the Declaration of Helsinki. Patients were recruited at the Department of Clinical Medicine of the University of Bologna with symptoms meeting Rome II criteria for diagnosis of IBS (n = 6; four female, two male patients; mean age 39 years; range 27–68 years). For our study we included 3 D-IBS and 3 C-IBS. None of the patients had a known intestinal infectious episode preceding development of their IBS symptoms. Biopsies were taken when the patients were symptomatic. Healthy controls (HC; n = 5; three female, two male patients; mean age 38 years; range 22–61 years) were recruited by public advertisement. As described in our previous study,2 in which we used the same biopsy supernatants, the exclusion criteria for both patients and controls included the following: major abdominal surgery, any organic syndrome, celiac disease (excluded by detection of antitransglutaminase and antiendomysial antibodies), asthma, food allergy, or other allergic disorders. None of the IBS patients or HC were taking non-steroidal anti-inflammatory drugs or other anti-inflammatory drugs, tricyclic antidepressants, or serotonergic agents including serotonin selective reuptake inhibitors, 5-HT3 receptor antagonists (i.e., granisetron, ondansetron, or alosetron), or 5-HT4 receptor agonists (i.e., tegaserod).

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. Four mucosal biopsies, taken from the proximal descending colon, were used to obtain an incubation supernatant. The incubation of the mucosal biopsies was performed as previously published.1 Briefly, four colonic biopsies of each study participant were rapidly immersed in 1 mL carbogen-aerated (95% O2/5% CO2) Hank’s solution warmed to 37 °C (Sigma, St. Louis, MO, USA). After an incubation of 25 min, the supernatants were centrifuged at 200 g for 10 min, aliquoted and stored at −70 °C until use.

Neuroimaging of guinea-pig enteric neurons with Multi-Site Optical Recording Technique

All animal studies were carried out in accordance with the German guidelines for animal protection and animal welfare and with approval of the local animal ethical committee at district veterinary office of Munich. Male guinea pigs (Dunkin Hartley weighing 280–420 g; Harlan GmbH, Borchen, Germany) were killed by cervical dislocation followed by exsanguination.

To study plexus-specific effects, we applied IBS supernatants to guinea-pig myenteric and submucous neurons and analyzed their spike activity using voltage-sensitive dye recording. We elected to examine guinea-pig rather than human tissues because a reliable recording from human myenteric plexus is still not possible. Enteric Nervous System activity was assessed on two adjacent segments of distal colon (length 2 cm each), which were quickly removed after killing the animals. The segments were dissected in ice-cold carbogen-aerated Krebs solution (pH 7.4), containing in mmol L−1: 117 NaCl, 4.7 KCl, 1.2 MgCl2 6H2O, 1.2 NaH2PO4, 25 NaHCO3, 2.5 CaCl2 2H2O, and 11 glucose (all from Sigma, Steinheim, Germany). In one segment, the mucosa and the muscle layers were carefully removed for preparation of the submucous plexus. To obtain longitudinal muscle-myenteric plexus preparation in the second segment, the mucosa, submucosa, and circular muscle layers were removed. The final preparations (10 × 20 mm) were pinned on silicone rings and placed in recording chambers continuously superfused with carbogen-aerated 37 °C Krebs solution at a rate of 17 mL min−1. The recording chambers were mounted onto an epifluorescence Olympus IX50 microscope (Olympus, Hamburg, Germany) equipped with a 75 W xenon arc lamp (Optosource, Cairn Research Ltd, Faversham, UK).

The multi-site optical recording technique (MSORT) is a fast imaging technique that allowed us the recording of action potential discharge in submucous and myenteric plexus neurons. The details of this technique have been described previously.19–21 Individual ganglia were stained with 20 μmol L−1 of the fluorescent voltage-sensitive dye 1-(3-sulfonatopropyl)-4-[β[2-(di-n-octylamino)-6-naphthyl]vinyl]pyridinium betaine (Di-8-ANEPPS; Molecular Probes Mobitec, Göttingen, Germany) by local pressure application through a microejection pipette; dye staining of enteric neurons did not change their electrical or synaptic properties.20

Recordings from Di-8-ANEPPS stained neurons were made with a 40× oil immersion objective (UAPO/340; Olympus) using an appropriate filter cube.20 Signals were acquired with a frequency of 1.6 kHz and processed by an array of 464 photodiodes (RedShirt Imaging, Decatour, GA, USA). Using a 40× objective, the spatial resolution was 280 μm2 per diode ensuring membrane potential recordings from all neurons within a ganglion at a single cell level. The MSORT technique allows measurement of relative changes in fluorescence intensity (ΔF/F) which are linearly related to changes in membrane potential.20 Illumination of the preparation for 1300–5000 ms was controlled by a software operated shutter (Uniblitz D122; Vincent Associates, New York, NY, USA). Longer exposures were not used due to potential dye bleaching and/or phototoxicity that may compromise results.

All ganglia were always checked for spontaneous activity before application of supernatants or drugs. For the final analysis all data were corrected for spontaneous activity (see methods chapter ‘Data expression and statistical analysis’). In cases of no spontaneously active neurons the viability of the neurons were tested by evoking fast excitatory postsynaptic potentials with a single electrical stimulus applied to an interganglionic fiber tract. IBS and HC supernatants were diluted 1 : 1 with Krebs buffer and were applied onto single submucous or myenteric ganglia by pressure ejection from micropipettes (20 p.s.i., ejection speed 55 ± 27 nL s−1, 100–200 μm distance from the ganglion). According to previously published calibration curves, any substance applied via pressure ejection pulses will be diluted by about 1 : 10 once it reaches the ganglion.22 The duration of the spritz application was 200 ms for all submucous ganglia and 200 and 400 ms for myenteric ganglia.

Specific receptor antagonists added to the superfusing Krebs solution included the 5-HT3 receptor antagonist cilansetron (0.1 μmol L−1; Solvay Pharmaceuticals, Hannover, Germany) or a mixture of the histamine H1 receptor antagonist pyrilamine (1 μmol L−1; Sigma) and the H2 antagonist ranitidine (10 μmol L−1; Sigma).22 Antagonists were perfused for 20 min before reapplication of IBS supernatants. In some experiments, the serine protease inhibitor FUT-175 (50 μg mL−1; Calbiochem, Darmstadt, Germany) was added to the supernatants 10 min before application to the submucosal neurons.1,3

In addition, defined chemical stimulations with either histamine (100 μmol L−1; Sigma), serotonin (5-HT, 1 mmol L−1; Sigma), or the protease-activated receptor 2 (PAR2)-activating peptide SLIGRL-NH2 (100 μmol L−1, Peptide Synthesis Core Facility, University of Calgary, Calgary, AB, Canada) were used to study possible differences in neuronal activation between submucous and myenteric neurons. Thereby histamine and serotonin were applied by local pressure pulse application (400 ms) on the ganglion,19,22 and the PAR2-activating peptide was perfused locally over the ganglion (3 μl of 100 μmol L−1 SLIRGL with speed of 100 nL s−1). The perfusion was necessary because of the known late onset response of guinea-pig submucous neurons to PAR2-activating peptide.23 For histamine and serotonin the recordings started 200 ms before the application of the substances and in the case of the PAR2-activating peptide, recording started directly after the local perfusion and lasted for 1850 ms. For all drugs the application methods were identical for submucous and myenteric neurons.

Data expression and statistical analysis

To analyze the proportion of neurons responding to the supernatants we counted the number of dye labeled neurons per ganglion. Individual neurons can be visualized as the dye incorporates into the membrane revealing the outline of individual cell bodies. The overlay of signals with the ganglion image allowed us to analyze the response of individual neurons.19,20

In cases of spontaneously active neurons, the frequency of action potential discharge following application of supernatants or drugs had to exceed the activity by at least 10% to be counted as a genuine response. In such cases the baseline number of action potentials was subtracted from the number of subsequent supernatants or drug evoked action potentials. If not otherwise stated, data are expressed as the median with the 25th and 75th percentiles (interquartile range; IQR). The percentages of activated neurons and the frequency of action potential discharge were compared between IBS and HC supernatants using the Kruskall–Wallis one-way analysis of variance on ranks combined with the post hoc multiple comparison procedure of Dunn (SigmaStat 3.1; Systat Software Inc., Erkrath, Germany). Effects of blockers were compared by the Wilcoxon Signed Rank Test for paired data (SigmaStat 3.1). Differences in effects of drug or supernatants between submucous and myenteric neurons were tested with the Mann–Whitney Rank Sum Test (SigmaStat 3.1). For all tests a P value of <0.05 was considered significant. To improve readability, numbers of subjects, ganglia, and neurons are given in sequence without further specification, e.g., a result based on experiments from five tissues, seven ganglia, and 30 neurons is presented as (5/7/30).

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Funding
  8. Disclosure
  9. Authors’ Contribution
  10. References

Effects of the IBS supernatants

A pressure pulse application of HC supernatants did not evoke any significant activation of guinea-pig submucous and myenteric neurons (Fig. 1). In contrast, 200 ms application of supernatants from all IBS patients evoked a significant fast onset spike discharge (median 3.6 Hz) in 46% of guinea-pig submucous neurons (Fig. 1). The spike discharge persisted throughout the recording period of 1850 ms. In some experiments, we used recording periods of up to 5 s but did not record any late onset response. Strikingly, each sample evoked a stronger activation in the submucous compared with the myenteric plexus. Initially, we used 200 ms application also in the myenteric plexus but did not observe reliable responses. Only after doubling the pressure pulse duration we were able to record a response. Furthermore, the proportion of responding neurons and their response rate was significantly lower in myenteric compared with submucous neurons (Fig. 1). A median of only 8.5% of myenteric neurons responded to IBS supernatants with a rather low spike frequency (median 2.1 Hz). In addition, in the myenteric plexus, only two of six IBS samples evoked a significantly higher action potential discharge compared with the effects of the HC samples (Fig. 1).

image

Figure 1.  Irritable bowel syndrome (IBS) supernatants activate guinea-pig enteric neurons. (A) In a submucous neuron (upper two traces) spritz application of D-IBS sample 137 evoked a strong action potential discharge, whereas the control supernatant HC 178 had no effect. In addition, a myenteric neuron (lower two traces) did not respond to a control supernatant HC 177, but showed a moderate action potential discharge in response to the D-IBS sample 137. (B) Summary of the neuronal responses (spike frequency and % responding neurons) to supernatant application in the submucous and myenteric plexus. In the submucous plexus, all IBS supernatants significantly increased the spike frequency in 30–57% of the neurons per ganglion compared with HC (#: P < 0.05; Dunn′s method). In contrast, in the myenteric plexus only two IBS samples showed a significantly higher spike frequency than control (§P < 0.05; Dunn′s method); the % responding neurons did not differ. In addition, IBS samples evoked a stronger response in submucous than myenteric neurons (*P values are given in the figure; Mann–Whitney Rank Sum test). For C-IBS sample 141, only 2 of 543 myenteric neurons responded, which does not allow a meaningful statistical analysis of different spike frequencies between submucous and myenteric neurons. Nevertheless, the nerve-activating effect of C-IBS is dramatically lower in myenteric neurons as only 2 of 534 myenteric neurons responded to this sample. The data derived from healthy control samples were pooled; for the submucous plexus, the sample numbers were HC 4, 5, 177, and 178, and for the myenteric plexus, the sample numbers were HC 6, 177, and 178. T/G/N, numbers of tissues, ganglia, neurons; median values with 25th and 75th percentiles.

Download figure to PowerPoint

Effects evoked by C- and D-IBS supernatants were similar in submucous [median spike frequency evoked by D-IBS: 3.0 Hz (IQR: 2.3/4.7) vs C-IBS: 3.0 Hz (1.8/5.1); = 0.673] and myenteric neurons [D-IBS: 2.0 Hz (IQR: 1.3/3.0) vs C-IBS: 2.0 Hz (1.3/2.6); = 0.580].

Active components in the IBS supernatants

To define the neuroactive components contained in the IBS supernatants we used a pharmacological approach with selective antagonists to specific mediators. For this analysis, the data derived from the individual IBS samples were pooled as they did not vary significantly between each other (P < 0.05; Fig. 1). In agreement with our previously published results in the human submucous plexus,2 serotonin, histamine, and proteases were also involved in the mediation of the IBS supernatants evoked activation of guinea-pig submucous neurons. To demonstrate involvement of serotonin we reapplied the IBS supernatants after a 20 min perfusion with the 5 HT3 receptor antagonist cilansetron,19 which significantly reduced or blocked the spike discharge evoked by all IBS supernatants in almost all neurons per ganglion (80%–100%; Fig. 2). To reveal a histaminergic component, we perfused the tissues for 20 min with a combination of the H1 and H2 histamine receptor antagonists pyrilamine and ranitidine. In the guinea-pig submucous plexus these two receptors are responsible for mediating the postsynaptic excitatory action of histamine.22,24 The histamine receptor antagonists reduced the IBS supernatant-evoked spike discharge in 98% (67–100%) of the neurons (Fig. 2). Finally, the preincubation of IBS supernatants with the serine protease inhibitor FUT-175 abolished the responses in nearly all neurons (60–100%) compared with the application of non-FUT-175-treated IBS supernatants (Fig. 2).

image

Figure 2.  Nerve activation by IBS supernatants involves serotonin, histamine, and proteases. The spike frequency of submucosal neurons evoked by IBS supernatants was significantly reduced in the presence of the 5-HT3 antagonist cilansetron, histamine H1/H2 receptor antagonists, or serine protease inhibitor FUT-175. Data derived from the different IBS samples were pooled. T/G/N, numbers of tissues, ganglia, neurons; median values with 25th and 75th percentiles; *significant difference, Wilcoxon Signed Rank test.

Download figure to PowerPoint

Effects of the individual active components

Serotonin, histamine, and proteases contributed to the IBS supernatant-evoked neural activation in the guinea pig. We further examined whether a different sensitivity to one of these components may explain the different responses of submucous and myenteric neurons elicited by the IBS supernatants. Therefore, we applied serotonin, histamine, or the PAR2 ligand SLIGRL on neurons of either plexus. We used SLIGRL as previous evidence has indicated that in the guinea-pig submucous plexus proteases primarily signal through PAR2.23 Interestingly, individual application of these components revealed that only serotonin was able to induce a significantly stronger activation of submucous compared with myenteric neurons (Fig. 3). Stronger activation was reflected by a higher action potential frequency (8.5 Hz vs 6.4 Hz) and a larger proportion of neurons activated by serotonin (55%vs 39%). In contrast, histamine and SLIGRL had similar effects in the two plexuses (Fig. 3).

image

Figure 3.  Serotonin, but not histamine or the PAR-2 agonist SLIGLR, has stronger effects in submucous than myenteric neurons. (A) Serotonin, histamine, and SLIGLR activate submucous (left three traces) and myenteric neurons (right three traces). Bars below the traces mark the application periods. (B) Summary of the neuronal responses (spike frequency and % responding neurons) to application of the three substances in the submucous and myenteric plexus. Only serotonin evoked a significantly higher spike frequency in a larger proportion of submucous neurons. Histamine and SLIGLR evoked similar effects in submucous and myenteric neurons (P values for differences in spike frequency after histamine application: = 0.713 and for % responding neurons = 0.157; P values for differences in spike frequency after SLIGRL perfusion: = 0.550, % responding neurons = 0.133). T/G/N, numbers of tissues, ganglia, neurons; median values with 25th and 75th percentiles; *significant difference, Mann–Whitney Rank Sum test.

Download figure to PowerPoint

To exclude the possibility that ionotropic receptor activation could evoke stronger activation of submucous neurons we studied effects of nicotine (10 μmol L−1). In contrast to the findings with 5-HT, nicotine evoked a stronger activation in myenteric neurons than in submucous neurons [median spike frequency in myenteric neurons: 9.0 Hz (IQR: 7.5/11.6) vs 7.4 Hz (4.9/10.6) in submucous neurons; < 0.001]. The number of nicotine-sensitive neurons were in both plexuses similarly [myenteric plexus: 79.8% of neurons (IQR: 73.6/85.6); submucous plexus: 73.3% (66.6/79.0); = 0.086].

We also excluded different labeling efficiency in myenteric and submucous plexus as a possible reason for the observed differences in neuronal responses. After dye staining the resting light intensity in submucous (n = 58 from four preparations) and myenteric neurons (n = 54 from four preparations) was similar [expressed in mV as output signal of the CCD camera: 1432 mV (IQR: 1125/1841) vs 1272 mV (IQR:1112/1634), = 0.297]. In the same preparations, we found that 400 ms spritz application of high K+ Krebs solution (50 mmol L−1 compensated by decreased NaCl concentrations) evoked a comparable spike discharge in submucous and myenteric neurons [3.9 Hz (IQR:0/8.3) vs 3.3 Hz (IQR:1.4/5.8]; = 0.678).

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Funding
  8. Disclosure
  9. Authors’ Contribution
  10. References

The main finding of this study in guinea pig was the markedly higher sensitivity of submucous compared to myenteric neurons to mucosal biopsy supernatants from patients with IBS. Even application of twice the amount of IBS supernatants onto myenteric neurons revealed a much smaller activation. Similar to our previous findings in human submucous neurons2 we could identify serotonin, histamine, and proteases as components that are involved in neural activation by IBS supernatants.

It is important to note that our experiments were performed in preparations containing either the myenteric or the submucous plexus with no functional connections between the two plexuses. Although these types of preparations were suited to study direct actions of the supernatants on submucous or myenteric neurons, they did not allow us to study the interdependency between the two plexuses. Such interactions may happen in intact tissue where activation of submucous neurons may cause a secondary response in the myenteric plexus. Despite this possibility, we may safely conclude that direct activation of neurons in the myenteric plexus by IBS supernatants is much less pronounced than in the submucous plexus.

Our results further support the relevance of altered signaling between immune, epithelial cells and nerves in IBS patients, and suggest that submucous rather than myenteric neurons are the predominant targets for mucosal mediators. The finding that the IBS supernatants preferentially activate submucous neurons is consistent with the mucosal to serosal gradient in immune cell density. Indeed, most mast cells and lymphocytes are located in the mucosal and submucosal regions while they are rarely observed in the myenteric and muscle layers.16,17 Our results suggest that the functional neuro–immune interactions occur mainly in the submucous plexus area where immunocytes and nerve processes are in close contact. Moreover, our findings identified a pathophysiological role of the close vicinity between EC cells and submucous neurons. This was previously demonstrated after mechanical stimulation of the mucosa which triggered 5-HT release from EC cells.25 This stimulus evoked a c-fos expression in submucous and myenteric neurons, but only the activation in the submucous plexus was direct, whereas activation of myenteric neurons was indirect25 and probably triggered by submucous neurons projecting to the myenteric plexus.26,27 Of course, direct and fast activation of myenteric neurons may also occur as a result of 5-HT3 receptor activation of their mucosal terminals.28

Interestingly, we observed a similar pattern in response to spritz application of incubation supernatants collected from stimulated cultured human intestinal mast cells.21 It has been shown that in the guinea-pig colon a higher proportion of submucous (38%) compared with myenteric neurons (11%) responded to the mast cell supernatants suggesting different sensitivities of the plexuses to certain immune challenges. This may also apply to other immune cell mediators. Thus, a higher proportion of submucous compared with myenteric neurons expresses c-fos in response to IL-1β29 and PGE2,30 which were increased in mucosal biopsies from IBS patients.10,31–33

Activation of human submucous plexus by IBS supernatants involved serotonin, histamine, and proteases.2 These findings also apply for their effects in guinea-pig submucous neurons based on the result that the IBS supernatant-evoked nerve activation was reduced by 5-HT3 or histamine H1/H2 receptor blockade and by inhibition of protease activity by FUT-175. We conducted experiments to compare the actions of serotonin, histamine, or PAR-2-activating peptide between myenteric and submucous neurons, to provide a possible explanation for the plexus related efficacies of IBS supernatants. We found that only serotonin mimicked the preferential action of the IBS supernatants as it triggered a more prominent activation of submucous than myenteric neurons. The excitatory effects of histamine and PAR-2-activating peptide were comparable. We may rule out that stimulation of ionotropic receptors would cause stronger activation in submucous neurons because the nicotinic receptor agonist nicotine evoked a more powerful activation of myenteric than submucous neurons. In addition, we showed that labeling efficiency and responses to high K+ Krebs solution is similar in myenteric and submucous neurons, which ruled out experimental bias. There is no conclusive explanation for the stronger action of serotonin on submucous neurons, but previous studies offer some hints that have to be explored in future studies. Channel properties of myenteric 5-HT3 receptor differ from those expressed on submucous neurons.34,35 Also particular pharmacological properties of myenteric 5-HT3 receptors35 or different 5-HT3 receptor subunits may contribute to the differences found in our study.

The different sensitivity to serotonin may in part account for the strong excitatory effects of the IBS supernatants in the submucous and their rather marginal response in the myenteric plexus. In addition to the 5-HT3 receptor antagonist, histamine receptor antagonists and serine protease inhibitors also strongly reduced the effects of the IBS supernatants on submucous neurons. Therefore, histamine and proteases must contribute to the nerve-activating actions of the IBS supernatants. As receptors for histamine, proteases, or serotonin are functionally expressed in both plexuses,19,22,36–39 the molecular basis for the prominent activation of IBS supernatants in submucous rather than myenteric neurons remains speculative. As previously suggested, one possible reason may be a particular receptor clustering in the submucous plexus that favors synergistic actions between histamine, proteases, and serotonin.2 In this context, it is noteworthy that the concentrations of exogenously applied serotonin, histamine, and proteases necessary to evoke excitation of enteric neurons were much higher than those found in the IBS supernatants (values published in2). The concentrations were chosen according to our previous studies on the actions of histamine,19 serotonin,22 and proteases.23 Of course, neural activation of IBS supernatants may in addition involve synergistic actions of other mediators which are increased in the intestinal mucosa of IBS patients.10,31–33 Recently, it has been shown that colonic soluble mediators from an IBS stress model in rats activated submucous neurons via an IL-6-dependent mechanism.40 Interestingly, it required 1 mmol L−1 IL-6 to match the Ca2+ signals evoked by IBS supernatants; a concentration that was much higher than the IL-6 levels found in the supernatants.40 Synergistic effects have been reported also for IL-1β, strongly enhancing the Ca2+ responses evoked by serotonin, adenosine triphosphate, SP, and electrical stimulation in guinea-pig myenteric neurons.41

In summary, we found that mucosal IBS supernatants preferentially activated submucous rather than myenteric neurons and reason that this activation profile is a functional correlate of the intimate cross-talk between immune or epithelial cells and submucous neurons. Our findings suggest that targeting specifically the interactions between mucosal mediators and submucous nerves may be an attractive therapeutic option for IBS and likely other functional bowel disorders.

Funding

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Funding
  8. Disclosure
  9. Authors’ Contribution
  10. References

This study was supported by the EU FP7 – IPODD consortium and the Deutsche Forschungsgemeinschaft DFG Sche-267/7-2; Italian Ministry of University and Research (COFIN Projects to G.B. and R.De.G.), and R.F.O. funds from the University of Bologna (to R.De.G., V.S., G.B.). R.De.G. is the recipient of a grant from the ‘Fondazione Del Monte di Bologna e Ravenna’ (Bologna, Italy); QL was supported by the Science and Technology Development Projects of Shandong Province of China (Grant No 21300004011065), and Jinan Institutions of Higher Education self-innovation Program (Grant No 21300005081023).

Authors’ Contribution

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Funding
  8. Disclosure
  9. Authors’ Contribution
  10. References

SB: data acquisition & analysis, manuscript drafting; QL, ThB, SV: data acquisition & analysis. GB, RDG, VS: material support, manuscript revision, funding. MS: study concept, manuscript revision, funding.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Funding
  8. Disclosure
  9. Authors’ Contribution
  10. References