| Background and aimsIrritable bowel syndrome (IBS) is a common functional gastrointestinal disorder which could cause significant reductions in patient’s quality of life and increased healthcare costs. Abdominal pain/discomfort is the most common symptom of IBS, with visceral hypersensitivity as one of the most important pathophysiological alterations of abdominal pain/discomfort in IBS. However, the exact mechanism of visceral hypersensitivity is still unclear. Previous studies for visceral hypersensitivity focused mainly on the functions of neurons and fibers of nervous system. The colonic mucosa could directly contact the luminal environment and the submucosal tissues. These cells participate the interplay between signals from colonic lumen to the deeper enteric never systems, leading to the development and maintenance of digestive diseases. Therefore, the colonic mucosal cells are promising targets in the pathogenesis of visceral hypersensitivity in IBS.N-methyl-D-aspartate receptor (NMDAR) is a type of ionotropic glutamate receptor, with NR1 as the obligatory subunit of this ion channel. NMDAR is widely expressed in the nervous system, playing key roles in excitatory synaptic transmission. NMDAR in central and peripheral nervous systems has been known to participate in the pathogenesis of visceral hypersensitivity. Meanwhile, previous studies have revealed the expression of functional NMDAR in non-neural cells and tissues, such as macrophages, keratinocytes, bone cells and parathyroid glands. However, the expression and possible role of NMDAR in visceral hypersensitivity in the colonic mucosa have not been elucidated.Previous studies have revealed that activation of NMDAR in neural cells could lead to production of brain-derived neurotrophic factor (BDNF), a member of the neurotrophin family, mediated by the extracellular signal-regulated kinase (ERK) pathway. Moreover, BDNF has been confirmed to be abundantly expressed on colonic mucosa and epithelia, and participates in the pathogenesis of visceral hypersensitivity in IBS.Therefore, we hypothesized that colonic mucosal NMDAR participated in the pathogenesis of visceral hypersensitivity in IBS, which might be mediated by the production of BDNF by the epithelial cells, in an ERK dependent pathway.In our study we tried to explore and testify our hypothesis in the following two parts. In the first part, we aimed to investigate the expression status of NMDAR on colonic mucosal cells and its relationship with visceral hypersensitivity:1. to confirm the expression status of colonic mucosal NMDAR in IBS patients and healthy controls, and the relationship between NMDAR expression levels and abdominal symptom scores; 2. to establish an IBS-like visceral hypersensitivity mice model and to explore the expression status of colonic mucosal NMDAR in mice and the relationship between NMDAR expression levels and visceral sensitivity AWR scores. In the second part, we aimed to investigate the possible mechanisms and downstream pathway underlying the NMDAR activation and visceral hypersensitivity in IBS:1. to testify whether the activation of mucosal NMDAR could lead to the synthesis and release of BDNF and visceral hypersensitivity in vivo in mice; 2. to reveal the possible signal pathway between colonic NMDAR activation and production of BDNF in vitro in colonic epithelial cells.MethodsPart 1IBS patients were diagnosed according to the Rome III criteria and subclassified into IBS with diarrhea (IBS-D) and IBS with constipation (IBS-C). Controls were enrolled from patients undergoing colonoscopy examination for polyps and cancer surveillance and had a normal colon. All participants were asked to score the abdominal pain/discomfort over the last 2 weeks according to a validated questionnaire described previously. Biopsy specimens during colonoscopy were obtained from the rectosigmoid junction. Immunohistochemistry and western blot were performed to investigate the expression levels of NMDAR on colonic mucosa. The correlation between mucosal NMDAR levels and abdominal pain/discomfort scores were calculated using Spearman rank correlation analysis. As NR1 is the obligatory subunit to form a functional NMDAR, so the NR1 antibody could label all NMDAR subtypes. Therefore, in the immunohistochemistry and western blot experiments in our study we used NR1 antibody to detect the expression of NMDAR and used the abbreviation "NR1" to stand for the expression of NMDAR.Male C57BL/6 mice weighing 20-30 g were used to establish the IBS-like visceral hypersensitivity mice model according to previous study. TNBS was administered intrarectally via a polyethylene catheter inserted 3 cm from the anus. The mice in the control group received an equivalent volume of 50% ethanol. The DAI (disease activity index) was used to evaluate the severity of TNBS colitis. Four weeks after the administration of TNBS, visceral sensitivity in mice was evaluated by measuring the scores of their abdominal withdrawal reflex (AWR) to colorectal distention (CRD).Following the CRD experiments, the mice were sacrificed and distal colon was removed. HE stain was used to evaluate the overall colonic inflammation state of the TNBS-treated and control mice. Immunohistochemistry and western blot were performed to evaluate the expression of NR1 on colonic mucosa. The correlations between NR1 expression and AWR score was assessed by using the Spearman’s rank correlation.Part 2Fecal samples were obtained from each mouse 4 weeks after TNBS administration before the CRD protocol. Each sample was dissolved and homogenized in double distilled water (2 mL/g feces). Supernatants were collected after centrifugation (4500 rpm,4℃,10 minutes). Fecal glutamate levels were measured using a commercially available ELISA kit. Ammonia was detected and quantified using an enzymatic determination kit. Both glutamate and ammonia levels were normalized against fecal weight (expressed as nmol/g feces and μg/g feces, respectively).Visceral sensitivity was investigated in mice receiving an intracolonic administration of ammonia (NMDAR agonist) at 200 mmol/L,500 mmol/L and 1000mmol/L (0.1 mL for each mouse), respectively, for three consecutive days. To verify whether the effect of ammonia was NMDAR-specific, a group of mice received an intracolonic administration of NMDAR antagonist MK801 (at 0.1,1.0 and 10mg/kg) 30 minutes before each treatment with ammonia. The mice treated with saline were set as the control group.Mice were administered intracolonically with NMDA (the specific agonist of NMDAR,10 mg/kg) for 3 consecutive days. Mice treated with PBS were served as controls. To block NMDAR, ERK pathway and TrkB receptor (the BDNF high affinity receptor), we pretreated mice by intracolonically administration of MK801 (10 mg/kg), U0126 (10 mg/kg) and TrkB/Fc (10 ng/mouse) 30 minutes before NMDA treatment. CRD procedures started 1 hour after the administration of NMDA on day 3. Electromyography (EMG) activity in response to CRD were recorded to evaluate the visceromotor response of mice.After the behavioral studies, mice were killed and the distal colons were removed. Under a dissection microscope, colonic mucosa was separated by blunt dissection. Protein was extracted from the mucosal samples for western blot of BDNF, total ERK and phosphorylated ERK (pERK).After serum-starved for 24 hours, HT29 cells were stimulated by NMDA (50,100 and 200 μM) to activate NMDAR. To inhibit NMDAR and ERK pathway, cells were preincubated with MK801 (10 μM) or U0126 (10 μM) for 30 minutes, respectively. Cells treated with PBS were used as controls. The cell supernatants were collected and an enzyme immunoassay kit was used for the quantitative determination of BDNF secretion in cell supernatants. Cells were also harvested for the measurement of BDNF, ERK and pERK using western blot.ResultsPart 11. Clinical characteristics of participantsThere were no significant differences in age and sex ratio between IBS patients and control subjects. A total of 37 patients with IBS were included, with 21 (56.8%) classified as IBS-D and 16 (43.2%) classified as IBS-C. For severity of abdominal symptoms, the mean score of IBS patients was significantly higher than that of controls (2.41 ± 1.01 vs.0.17 ± 0.38, P<0.001). No significant difference was found between IBS-D and IBS-C (2.48±1.03 vs.2.31 ± 1.01, P=0.63). For frequency of abdominal symptoms, the mean score of IBS patients was also significantly higher than that of controls (2.30 ± 1.10 vs.0.22 ± 0.55, P<0.001). No significant difference was observed between IBS-D and IBS-C (2.33 ± 1.24 vs.2.25 ± 0.93, P=0.82).2. NMDAR expression in colonic mucosa of humanImmunohistochemistry confirmed the abundant expression of NR1 in human colonic mucosa, including epithelial cells and lamina propria cells.3. The measurement of NR1 using immunohistochemistry in colonic mucosa of human and its relationship with abdominal symptomsQuantification analysis revealed that the expression level of NR1 in colonic mucosa was significant higher in IBS patients compared with controls (median 33.51, IQR 26.66-46.70 for IBS and median 25.35, IQR 13.60-35.66 for control; P=0.004). There was no significant difference between IBS-D and IBS-C patients (median 33.97, IQR 26.66-47.30 for IBS-D and median 33.15, IQR 26.28-42.75 for IBS-C; P=0.705).Using Spearman rank correlation analysis, a significant correlation was found between mucosal NR1 levels detected by immunohistochemistry and severity of abdominal pain/discomfort in all patients groups (control:r=0.560, P=0.016; IBS: r=0.522, P=0.001; control and IBS:r=0.596, P<0.001). The colonic mucosal NR1 levels also significantly correlated with the frequency of abdominal pain/discomfort in all patient groups (control:r=0.568, P=0.014; IBS:r=0.632, P<0.001; control and IBS: r=0.663,P<0.001).4. The measurement of NR1 using western blot in colonic mucosa of human and its relationship with abdominal symptomsDensitometric analysis of the western blot bands showed that compared with controls, the NR1 levels were significantly higher in colonic mucosa of IBS-D patients (1.71 folds of control, P<0.05) and IBS-C patients (1.51 folds of control, P<0.05). There was no significant difference between IBS-D and IBS-C patients (P>0.05).A significant correlation was found between mucosal NR1 levels quantified by western blot and severity of abdominal pain/discomfort in all patients groups (control: r=0.646, P=0.004; IBS:r=0.586, P<0.001; control and IBS:r=0.669, P<0.001). The colonic mucosal NR1 levels also significantly correlated with the frequency of abdominal pain/discomfort in all patient groups (control:r=0.649, P=0.004; IBS: r=0.551, P<0.001; control and IBS:r=0.708, P<0.001).5. Intrarectal administration of TNBS resulted in visceral hypersensitivity in mice without histological changesThe DAI scores were highest in day 1-7 after TNBS administration and declined gradually until to almost 0 in day 21. In day 28, the mice were used for further experiments. Macroscopic assessment including hyperaemia, ulcers, bowel wall thickening, stricture, et al., demonstrated no significant difference between colons of TNBS mice and controls. Histological evaluation of colons showed almost no inflammation in both TNBS-treated and control groups.Visceral response to CRD did not differ between TNBS mice and controls at 15 mmHg. However, mice treated with TNBS showed statistically stronger responses to CRD at 30,45 and 60 mmHg, compared with controls.6. NMDAR expression in colonic mucosa of miceNR1-immunoreactive cells were found to be scattered throughout the colonic mucosa of mice in all the colonic specimens, and NR1-like immunoreactivity was abundant in both mucosal epithelial cells and lamina propria cells in the colonic mucosa.7. The measurement of NRl using immunohistochemistry in colonic mucosa of mice and its relationship with colonic sensitivityQuantitative immunohistochemical analysis of the colonic mucosa of mice showed a significantly higher NR1 expression in TNBS-treated mice than in the control group (TNBS mice:median 19.7, IQR 15.4-26.9; controls:median 11.7, IQR 9.8-14.5; P=0.019).We then investigated the correlation between NRl expression and AWR score at all four CRD pressures. In both TNBS-treated mice and controls, a significant correlation was found between mucosal NR1 expression using immunohistochemistry and the AWR scores, especially at the CRD pressure of 45 mmHg (r=0.831, P=0.003 for TNBS-treated mice and r=0.766, P<0.01 for the controls, respectively) and 60 mmHg of CRD pressure (r=0.668, P=0.035 for TNBS-treated mice and r=0.875, P<0.001 for the controls, respectively).Regardless of the differences between the groups, NR1 expression was also significantly correlated with AWR scores at the CRD pressures of 30 (r=0.741, P< 0.001),45 (r=0.841, P<0.001) and 60 mmHg (r=0.738, P<0.001), respectively.8. The measurement of NR1 using western blot in colonic mucosa of mice and its relationship with colonic sensitivityAs shown by western blot, TNBS-treated mice showed a 1.74-fold higher expression of NR1 protein in the colonic mucosa compared with the mice in the control group (P=0.014). A significant correlation was found between NR1 expression levels and the AWR scores in the TNBS-treated mice and controls, especially at a CRD pressure of 30 mmHg (r=0.828, P= 0.003 for TNBS-treated mice and r=0.654, P= 0.040 for the controls) and 45 mmHg (r=0.757, P=0.011 for TNBS-treated mice and r=0.883, P<0.001 for the controls).For all animals, including both the TNBS and control groups, NR1 expression was also correlated with the AWR scores, at all four CRD pressures (r=0.524, P=0.018 at 15mmHg; r=0.826, P<0.001 at 30mmHg; r=0.843, P<0.001 at 45 mmHg; and r=0.641, P-0.002 at 60 mmHg).Part 21. Fecal glutamate and ammonia levels in TNBS-treated mice and controlsThere was no difference in the fecal glutamate level between the TNBS-treated mice and controls (17.4 ±0.1 nmol/g vs.17.7±0.1 nmol/g, P=0.085). However, the ammonia level in the feces of the TNBS-treated mice were significantly higher than that in the controls (15.4±1.7 μg/g vs.10.2 ± 1.1 μg/g, P=0.017).2. Ammonia-induced visceral hypersensitivity was repressed by pretreatment with NMDAR antagonist.In order to confirm the functional profile of the NMDAR activated by fecal ammonia in mouse colonic sensitivity, AWR scores in response to CRD pressure after intracolonic infusion of ammonia or saline was recorded. We found that the administration of ammonia markedly increased the colonic response to CRD at 30 and 45 mmHg compared with saline, especially at a concentration of 1000 mmol/L. However, this effect was largely attenuated in mice pretreated with the NMDAR antagonist MK-801 before ammonia infusion, especially at a concentration of 10 mg/kg.3. NMDA-induced visceral hypersensitivity was repressed by the blockade of NMDAR, ERK pathway and TrkB receptor.Compared with control group, intracolonic administration of NMDA did not change the colorectal sensitivity at 15 mmHg CRD pressure. However, NMDA produced a significant increase in colorectal sensitivity at pressures of 30,45 and 60 mmHg. Blockade of NMDAR and TrkB receptor of colon by intracolonically administration of MK801 and TrkB/Fc significantly repressed the visceral hypersensitivity produced by NMDA, at pressures of 30,45 and 60 mmHg. Blockade of ERK pathway by U0126 also significantly repressed the visceral hypersensitivity produced by NMDA, at pressures of 45 and 60 mmHg.4. Mucosal BDNF expression and ERK pathway activation in miceWestern blot showed that mucosal BDNF of mice receiving intracolonic administration of NMDA was significantly elevated compared with control group (P< 0.001), which was inhibited by pretreatment of MK801 and U0126 (Both P<0.01).No difference was observed in the total ERK levels among groups, but NMDA administration produced significantly greater expression levels of pERK in mice mucosa (P<0.001), which was also inhibited by pretreatment of MK801 and U0126 (Both P<0.001).5. BDNF secretion in cell supernatantsELISA showed that NMDA treatment induced increased BDNF protein levels in a concentration-dependent manner in cultured supernatants. To investigate whether this effect was mediated by the NMDAR activation and the ERK pathway, we pretreated cells with MK801 and U0126 30 minutes before NMDA treatment. Both MK801 and U0126 could repress the increased BDNF levels in supernatants induced by 100 μM NMDA (Both P<0.001).6. BDNF expression and ERK pathway activation in cellsWestern blot revealed that BDNF levels in harvested cells were significantly increased to 1.47,2.43 and 2.60 folds of control respectively, after NMDA stimulation at 50,100 and 200 μM. Preincubation of MK801 and U0126 abolished the effect of NMDA on BDNF expression.Similar results were found for the increased levels of pERK at 50,100 and 200 μM of NMDA treatment in a concentration-dependent manner compared with control group. Pretreatment with MK801 and U0126 could also block the increased levels of pERK.ConclusionsPart 11. NMDAR was abundantly expressed on human colonic mucosa and elevated NMDAR levels were observed in IBS patients compared with healthy controls, and were positively associated with the severity and frequency of abdominal symptoms.2. A viscerally hypersensitive mouse model could be established by the intracolonic application of TNBS. Increased expression of NMDAR in the colonic mucosa from TNBS-treated mice and a significant association between the NMDAR levels and colonic sensitivity were demonstrated, indicating a link between mucosal NMDAR and visceral hypersensitivity.Part 21. Increased ammonia was found in fecal samples from TNBS-treated viscerally hypersensitive mice. Moreover, the intracolonic administration of ammonia can lead to the activation of mucosal NMDAR and induce visceral hypersensitivity.2. Activation of mucosal NMDAR by intracolonic administration of NMD A in normal mice produced increased expression of BDNF, activation of ERK pathwan and visceral hypersensitivity.3. Activation of colonic epithelial cells in vitro could also induce increased production of BDNF, which was mediated by the activation of ERK pathway. |
