| Background and AimsIrritable bowel syndrome (Irritable bowel syndrome, IBS) is a common chronic functional disease of digestive system. The main symptom of IBS is abdominal pain/ discomfort, accompanied by altered bowel habit and faeces character and laboratory examination cannot find an organic profile. Abdominal pain is one of the main manifestations of IBS patients, and the occurrence of severity and frequency significantly affect the quality of life. However, the mechanism of IBS abdominal pain is not entirely clear at present. A large number of studies have found that the increased visceral sensitivity is an important pathophysiological mechanisms of IBS abdominal pain. Visceral hypersensitivity may contain the central and peripheral sensitizations. They maybe separately or jointly participate in the pathogenesis of IBS. Until now, how central or peripheral sensitization occurs is still poorly understood.In recent years, a series of studies have found that patients with IBS have obvious changes in colonic mucosa, which is closely associated with IBS visceral hypersensitivity. These changes include local immune activation that causes an increased release of serotonin from intestinal chromaffin cells, as well as histamine overexpression of activated mast cells, etc. Brain derived neurotrophic factor (BDNF) is one of the main members of neurotrophin family, and widely expressed in central and peripheral nervous system. BDNF plays an important role on the neuronal differentiation and maturity. In recent years, studies have shown that BDNF is also a kind of pain modulator, with its abnormal expression in different pathophysiological conditions contributing to a variety of pathological pain processes. While expression of BDNF is also found to be expressed the intestinal tract. Our previous study has found that patients with IBS, especially diarrhoea-predominant IBS (IBS-D) patients, show a high expression level in colonic epithelium and lamina propria, which is significantly associated with pain severity and frequency, suggesting that intestinal BDNF may play an important role in the IBS abdominal pain. However, what triggers colonic epithelial BDNF overexpression and how the increased colonic BDNF may participate in visceral hypersensitivity remain unclear. Several studies have found that IBS-D patients have an increased serine protease activity than that of normal controls, which may contribute to the increased intestinal permeability by activating protease activated receptor-2 (PAR-2) thereby causing visceral hypersensitivity. While studies have shown that PAR-2 activation is also involved in regulation of epithelial cells secretion process. Thus, whether the increased expression of BDNF in colonic epithelium of IBS-D patients is attributed to the increase in serine protease activity of intestinal contents and the following activation of PAR-2 signaling pathways, is worthy of our in-depth study. Enteroglial cells (EGCs), with their morphology and function similar to astrocytes (expressing the same activation maker-glial fibrillary acidic protein, GFAP) in brain, form an extensive cell-network throughout the intestinal wall. The EGC network, connecting the intestinal epithelium and the sub-mucosal enteric nervous system (ENS), has been revealed to play an important role in the maintenance of intestinal homeostasis. Recent studies further showed that EGC function could be profoundly changed by many factors such as pro-inflammatory cytokines, bacteria, and neurotransmitters. Thus, EGCs may undergo dynamic processing under pathological conditions and serve overlapping functions such as modulating intestinal barrier function, mucosal immunity and enteric neurotransmission via releasing various substances. Thus, whether BDNF can act on EGCs and thereby contribute to IBS-like visceral hypersensitivity has yet to be examined. On the basis of above findings, we, in the first part of the current study, would detect:1. the expression of BDNF in colonic epithelial cells after stimulation by luminal contents from in patients with IBS-D; 2. the molecular mechanism that IBS-D luminal contents from IBS-D patients-induced alteration of expression of colonic epithelial BDNF. In the second and third part, we would detect:1. the expression of BDNF, tropomyosin receptor kinase B (TrkB), EGC-related molecues in intestinal mucosa of IBS patients, as well as the correlation between these proteins and abdominal pain (second part); 2. the relationship between BDNF and EGC activation and their role in visceral hypersensitivity by using IBS-D fecal supernatants-induced hypersensitivity mouse model, EGC dysfunction mouse model and BDNF+/- mice; 3. the molecular mechanism that BDNF acts on EGCs, and the direct role of BDNF alone or activated EGCs on excitability and mechanosensitivity of intestinal sensory afferent nerves (third part).MethodsPart 1Twenty-two patients with IBS-D and 17 healthy control subjects were enrolled. The diagnosis of and sub-classification IBS were according to Rome Ⅲ criteria. Colonic luminal contents were collected from each subject and further processed to extract the supernatant (Fecal supernatants, FSN). Total proteolytic enzyme activity was determined by application of azo-casein method. Serine protease inhibitor FUT-175 was used to incubate with IBS-D FSN to confirm whether the protease activity is serine dependent.Total RNA was extracted from cells after 6 or 24 hours’FSN stimulation by using Trizol method. After RNA concentration assay, reverse transcription was performed to synthesize cDNA, and then real-time fluorescent quantitative PCR was done to detect BDNF and PAR-2 mRNA expression. Supernatants were collected from culture medium of FSN-stimulated cells, and then we used BDNF ELISA kit to detect BDNF concentration. Total protein was extracted and western blotting was used to detect PAR-2 protein expression. Human colonic cancer-derived epithelial cell line, Caco-2, was used in this study. siRNA targeted for PAR-2 used to knockdown PAR-2 in Caco-2 cells. Cells were received stimulation by IBS-D FSN or control FSN or saline after transfection for 24 hours and then PAR-2 and BDNF protein levels were determined. Using antagonists of PAR-2, p38 MAPK or p65 NF-κB, the impact of IBS-D FSN on those signialling pathway proteins was examined.Male C57BL/6 mice were chosen for in vivo experiments. IBS-D FSN was used to induce colonic hypersensitivity. Colorectal distention was performed after colonic infusion. Visceral sensitivity was recorded by electromyograph (EMG) activities. Colonic tissue at infused segment was collected and processed. Total protein was extracted and the expression of BDNF was detected by western blotting and immunohistochemistry.Part 2IBS patients and age- and sex-matched control subjects were invited to participate in this study based on the Rome Ⅲ criteria. Each IBS patient was invited to complete an Italian modified version of the Bowel Disease Questionnaire (BDQ) to evaluate gastrointestinal symptoms. Fecal samples were collected from HCs and IBS patients before bowel preparation for colonoscopy. All participants underwent colonoscopy after a standard bowel preparation by using polyethylene glycol. We took 6 specimens for each participant from the rectosigmoid junction. Two biopsies were used for routine hematoxylin and eosin histology, immunohistochemistry and immunofluorescence; two biopsies were used for western blotting; and two specimens were used for transmission electron microscopy (TEM).Immunohistochemistry was used for detecting the distribution and overall expression of BDNF and GFAP. Double immunostaining was used for detecting TrkB and substance P (SP) expressed in EGCs, and TrkB expressed on nerve fibers. Western blotting was used for quantifying the expression of BDNF, GFAP and TrkB proteins in biopsy specimens. A correlation analysis was performed to examine the relationship between those above protein levels and abdominal pain scores.TEM was used to study the morphological changes of EGCs. Quantification of the ultrastructure changes in EGCs are as followings:the integrated optical density of heterochromatin, the numbers of mitochondrion and glycogenosome as well as other organelles.Part 3Male heterozygous BDNF+/- mice (C57BL/6 background) and BDNF+/+ littermates at 4 months of age were used for visceral sensitivity studies. After intracolonic infusion with FSN, CRD and EMG was performed to evaluate the visceromotor response of mice. Colonic samples (at infused area) were obtained from mice 4 hours after supernatants infusion. We used western blotting or immunohistochemistry to detect expressions of mucosal BDNF, GFAP, TrkB and SP levels.Rat EGC cell line was exposed to recombinant human BDNF (r-HuBDNF). We used phospholipase Cγ1 (PLCγ1) inhibitor or recombinant TrkB Fc chimera (TrkB/Fc) to previously block r-HuBDNF-TrkB pathway to determine whether TrkB-PLCγ1 is responsible for the process during BDNF induced EGC activation and the following SP production. SP and signaling pathway proteins in r-HuBDNF -stimulated EGCs were detected by western blotting.Mesenteric afferent nerve recording was performed to investigate the effects of supernatants collected from r-HuBDNF-stimulated CRL-2690 cells or r-HuBDNF alone on nerve excitability and mechanosensitivity.ResultsPart 11. IBS-D fecal supernatants (IBS-D FSN) had increased serine protease activity.Total protease activity of IBS-D FSN (1461 ±143.1 U/mg protein) was significantly higher than that of normal controls (438.6+70.2 U/mg protein). Application of serine protease inhibitors FUT-175 significantly decreased the total protease activity of IBS-D FSN, which suggested that the increased total protease activity in IBS-D FSN was mainly relies on the serine protease.2. BDNF mRNA and protein expression were upregulated in Caco-2 cells after IBS-D FSN stimulation.BDNF mRNA expression was significantly increased in Caco-2 cells after IBS-D FSN challenge. Preincubation of IBS-D FSN with FUT-175 significantly suppressed the effect of IBS-D FSN. Consistent with the results of mRNA levels, ELISA showed BDNF protein expression also remarkably elevated after IBS-D FSN stimulation, and FUT-175 was also significantly inhibited the effect of IBS-D FSN on BDNF protein production.3. PAR-2 gene knockdown significantly repressed IBS-D FSN inducing Caco-2 cells to overexpress BDNFPAR-2 expression was significantly upregulated in Caco-2 cells when stimulated by IBS-D FSN compared with control FSN or saline. In order to further investigate whether PAR-2 activation is involved in IBS-D FSN-triggered upreguation of BDNF in Caco-2, we silenced PAR-2 gene in Caco-2 by using liposome-mediated RNA interference. Results showed that PAR-2 gene silencing had no effect on basal expression of BDNF but significantly inhibited IBS-D FSN-induced increase of BDNF expression in Caco-2 cells.4. P38 MAPK but not p65 NF-κB mediated IBS-D FSN-triggered BDNF overexpressionTo further study downstream pathway of PAR-2-mediated BDNF overexpression in Caco-2, we examined MAPK p38 and NF-κB p65. Results showed that IBS-D FSN significantly induced an increase in phosphorylation of p38 MAPK, but had no obvious effect on p65. PAR-2 inhibitor ENMD-1068 significantly inhibited IBS-D FSN induced phosphorylation of p38. P38 inhibitor SB203580 significantly blocked the effect of IBS-D FSN on elevating BDNF expression in Caco-2 cells, however, the p65 inhibitor PDTC did not.5. Colorectal infusion with IBS-D FSN induced hypersensitivity in mice, with itsaction dependent on colonic PAR-2 activation and BDNF overexpression. Western blotting results showed that, compared with control FSN, intracolonicinfusion with IBS-D FSN significantly increased BDNF expression in colonicepithelium, and this effect could be inhibited by ENMD-1608 and FUT-175.Immunohistochemical staining verified the western blotting results, showing thatincreased BDNF was mainly distributed in the colon epithelium and mucosa. Functionally, abdominal muscle reactions under colorectal infusion with saline orcontrol FSN did not affect the colonic sensitivity of mice. By contrast, IBS-D FSNsignificantly enhanced abdominal muscle electrical response to CRD. Blocking BDNFor PAR-2, or elimination of the increased serine protease activity of IBS-D FSNsignificantly inhibited IBS-D FSN’s effect on colonic hypersensitivity in mice.Part 21. Characteristics of study subjects In the present study,30 IBS patients and 30 HCs entered this study. On thewhole, IBS patients had significantly higher BDQ pain severity (P<0.0001) andfrequency (P<0.0001) scores than HCs. No difference was found between IBSsubgroups in the aspect of the above BDQ scores. Fecal serine-protease activities wereelevated markedly in IBS-D patients (median (IQR) 1521 (905.3-1983) trypsin units/mgprotein) compared with HCs (465 (264.3-708.0) trypsin units/mg protein; P<0.0001).Aprotinin significantly reduced the protease activities of IBS-D FSN (245 (109.8-413.5)trypsin units/mg protein; P<0.001). In contrast, there were no significant differences inprotease activities between IBS-C FSN (455 (253-905.5) trypsin units/mg protein) andHCs.2. The expression of BDNF and TrkB was closely asscociated with EGCs in humancolonic mucosa Immunohistochemical staining showed that BDNF was mainly stained in colonicepithelial cells and lamina propria, while GFAP was only stained in lamina propria. Double immunofluorescent staining identified that TrkB was mainly expressed in EGCs and PGP9.5-immunoreactive nerve fibers. EGCs were frequently close to or directly contacted with mucosal nerve fibers, which provides a structural basis for mucosal glia-nerve interaction.3. The expression of BDNF, TrkB and GFAP were increased in colonic mucosa of IBS patientsMarkedly increased BDNF levels in colonic mucosa were confirmed in a different population of patients with IBS. Total TrkB protein levels were significantly upregulated in colonic mucosal biopsies of patients with IBS. Western blotting and immunostaining confirmed the increased total GFAP and co-expression of GFAP and TrkB in colonic mucosal specimens of IBS patients, indicating that TrkB-related EGC activation might exist in IBS patients.4. IBS patients had altered ultrastructure of EGCsUltrastructural alterations of EGCs further showed that colon-derived EGCs in IBS patients had a decrease in integrated optical density of heterochromatin that indicated less repression of nearby sequences and more transcriptions, an increase number of mitochondria and glycogenosome, which suggested an elevated protein synthesis. In addition, rough endoplasmic reticulum and polyribosomes also seemed to be increased in the EGC cell body in IBS group.5. EGCs expressed more SP in IBS patientsA big proportion of SP immunostaining was co-localized with GFAP and this co-expression was significantly increased in IBS patients. In addition, SP positive nerve fibers and other cells were also observed in luminal propria, around EGCs.6. The expression of GFAP and SP was correlated with IBS abdominal painRelative GFAP protein levels and SP-positive EGC immunostaining areas were categorized according to the corresponding BDQ scores of each IBS patient, respectively. As expected, the protein levels of GFAP correlated significantly with abdominal pain severity and frequency scores in IBS patients. Similar correlation was observed for SP expression.Part 31. IBS-D FSN-mediated visceral hypersensitivity mice over-expressed colonic BDNF, GFAP, TrkB and SPWe used the IBS-D FSN collected from the study of part 2 to induce hypersensitivity in wildtype mice. Western blotting and immunofluorescence staining were performed to detect expressions of colonic BDNF, TrkB, GFAP, total SP and EGC-expressed SP. We found that the expressions of those above proteins were significantly higher in IBS-D FSN treated group than control FSN or aprotinin-preteated FSN group.2. BDNF gene knockdown inhibited the effect of IBS-D FSN on visceral sensitivityIn order to verify whether BDNF levels are associated with the activation of EGC and whether the inducing effect of IBS-D FSN on visceral sensitivity can be repressed by BDNF knockdown, we used the BDNF gene knockout heterozygous mice (BDNF+/-) as a negative control, and found that the EMG reaction of BDNF+/- mice receiving IBS-D FSN was significantly lower than that of wild-type mice. BDNF knockdown by half had a certain influence on the basal visceral sensitivity, which was more obvious during CRD with the pressure over 60 mmHg.3. BDNF gene knockdown inhibited the effect of IBS-D FSN on expression of BDNF, GFAP, TrkB and SPTo further clarify whether EGC activation in the colon is dependent on BDNF levels, we used BDNF+/- mice as the control to compare the effect of the IBS-D FSN or control FSN on expression of GFAP, TrkB and SP in BDNF+/+ mice and BDNF+/- mice. Western blotting showed that, in BDNF+/- mice, there was no significant difference in the expression of GFAP and TrkB between the effect of IBS-D FSN and control FSN. GFAP and TrkB expression levels were similar between BDNF+/+ mice and BDNF+/ mice under control FSN stimulation. Immunofluorescence dual staining showed that total SP and EGC-expressed SP levels in BDNF+/- mice did not differ between IBS-D FSN stimulation and Control FSN stimulation, but were significantly lower than BDNF+/+ mice under both stimulations.4. EGC dysfunction prevented effect of IBS-D FSN-induced visceral hypersensitivity and SP production in micePrevious administration of fluorocitrate in mice significantly decreased the elevated visceromotor response (VMR) scores evoked by IBS-D FSN. Furthermore, IBS-D FSN failed to evoke an increase in SP immunostaining in EGCs.5. r-HuBDNF directly activated rat EGC cell line (CRL-2690 cells) to produce SP via TrkB-dependent activation of PLCyl signaling pathwayr-HuBDNF significantly increased expression of GFAP in a dose dependent manner. TrkB/Fc completely blocked r-HuBDNF-induced expression of GFAP and TrkB as well as phosphorylation of PLCγ1. U73122 was also found to markedly inhibit r-HuBDNF-induced phosphorylation of PLCγ1 and GFAP expression. Untreated CRL-2690 cells expressed a very low level of SP. In contrast, r-HuBDNF significantly upregulated SP protein levels compared with control in CRL-2690 cells, and U73122 and TrkB/Fc could completely inhibit the expression of SP.6. Supernatants from r-HuBDNF-treated rat EGCs culture medium but not r-HuBDNF itself, enhanced discharge of mesenteric afferent nervesControl culture supernatants had no effect on the nerve discharge rate at any duration of time. Concentration-response curve showed that r-HuBDNF alone (10～200 ng/ml) had no excitatory role on afferent discharge compared with baseline. In sharp contrast, supernatants collected from r-HuBDNF (50～200 ng/ml)-treated CRL-2690 cells significantly increased the afferent discharge. When intervened with U73122 or TrkB/Fc, supernatants from r-HuBDNF-treated EGCs could not induce any augmentation of afferent discharge on the basis of baseline. The SP receptor NK1 antagonist FK888 significantly inhibited excitatory effect of r-HuBDNF-treated EGC supernatants. Interestingly, a residual effect of r-HuBDNF-treated EGC supernatants remained and was significantly greater than baseline measurements.7. r-HuBDNF increased mechanosensitivity of mesenteric afferent nerves, which was significantly enhanced by EGC activation.With mesenteric afferent nerve recording assay, we found that supernatants from r-HuBDNF-treated EGCs induced the greatest increase in mechanosensitivity of jejunal mesenteric afferent nerves. Modest, but significant increases in mechanosensitivity were also observed following pretreatment with r-HuBDNF alone. By contrast, preincubation of the nerve with TrkB/Fc partially repressed the increase induced by r-HuBDNF-stimulated EGC supernatants, while preincubation with TrkB/Fc and FK888 together remarkably inhibited the increase in mechanosensitivity caused by rHuBDNF-stimulated EGC supernatants. Collectively, these data suggest that r-HuBDNF alone can cause mechanosensitivity of intestinal afferent nerves although not directly induce nerve discharge. r-HuBDNF-activated EGCs also can increase nerve mechanosensitivity through secreting multiple mediators including SP. Thus, r-HuBDNF and activated EGCs may exert a synergistic effect in enhancing the mechanical sensitivity of afferent nerves.ConclusionsPart 11. Fecal supernatants from IBS-D patients can significantly promote the expression of intestinal epithelial BDNF through activating the PAR-2-p38 MAPK signaling pathway, and this effect is dependent on the elevated serine protease activity.2. IBS-D fecal supernatants can induce colonic hypersensitivity in vivo, and this process depends on the colonic epithelial PAR-2 receptor activation and the increased BDNF expression.Part 21. EGCs were probably activated in colon of IBS patients, with increased expression of GFAP and SP.2. There was a strong association between colonic mucosal BDNF, EGCs and sensory nerve fibers, and a enhanced interaction was expected.3. The alterations of BDNF levels and EGCs were significantly correlated with IBS abdominal pain.Part 31. BDNF could increase the chemical and mechanical sensitivity of sensory afferent nerves, but could not trigger excitatory discharge.2. BDNF could directly activate EGCs to secret SP.3. Activated EGCs could increase the chemical and mechanical sensitivity of sensory afferent nerves through releasing SP, directly induce discharge of nerves, and synergistically contribute to IBS-like visceral hypersensitivity induced by BDNF. |
PDF Full Text Request