Evaluation Of Intestinal Mucosal Barrier Function In Patients With Irritable Bowel Syndrome

BackgroundIrritable bowel syndrome (IBS) is a bowel disorder characterized by chronic lower abdominal pain or discomfort relieved after defecation in association with disordered defecation. IBS may be classified into three symptoms subgroups according to the Romeâ…¡criteria: constipation-predominant IBS, diarrhea-predominant IBS and IBS with alternating bowel movements. The concise etiology and pathophysiology of IBS remain largely unknown until now. It is considered to be a multifactorial disease, involving abnormal gastrointestinal motility, visceral hypersensitivity, psychosocial factors and mucosal inflammation. Recently, studies from several research groups support a view that the impaired intestinal mucosal barrier function, as measured by an increase of the intestinal permeability, may be implicated in the pathogenesis of diarrhea-predominant IBS. It is unknown that how many factors contribute to the increased intestinal permeability and whether it is related to the clinical symptoms. Although appeared as normal by conventional criteria, a persistent low-grade immune activation in the intestine has been shown in patients with diarrhea-predominant IBS using specific staining techniques and may contribute to the pathogenesis of IBS.In normal condition, gut epithelial lining forms a relatively impermeable barrier between luminal contents and submucosa required for intestinal homeostasis. This barrier is determined by complexes of proteins composing the junctional complexes. Tight junctions are the most apical organelle of the epithelial junctional complexes and are crucial for the formation and function of epithelial barriers. Numerous studies have proven the inflammation have a detrimental effect on epithelial barrier integrity in human diseases such as acute gastroenteritis, celiac disease and inflammatory bowel disease. Therefore, it is possible that the increased small bowel permeability in patients with diarrhea-predominant IBS was due to the persistent intestinal inflammation. Tight junctions comprise numerous proteins, with the best characterized being zonula occludens (ZO)-l and occludin which are critical elements in the formation of tight junctions. There were no published studies about whether the two main tight junction molecules (ZO-1 and occludin) were abnormal in diarrhea-predominant IBS in our knowledge.Preservation of the impaired mucosal barrier function and attenuation of the inflammation may be an attractive therapy for diarrhea-predominant IBS. A promising alternative is to use probiotic bacteria that interact with the host epithelium to resolve inflammation and preserve the barrier function. Probiotics were defined as 'living microorganisms that (when ingested) have a beneficial effect in the prevention and treatment of specific pathologic conditions". Currently, the most-studied probiotics are the lactic acid bacteria and some lactic acid bacteria strains used for fermented products such as Lactobacillus acidophilus alone has been shown to be effective in relieving IBS symptoms of pain and diarrhea. Bifidobacterium infantis 35624 has been shown to alleviate symptoms in IBS and this effect was associated with the rebalance of interleukin (IL)-10/IL-12 ratio produced by peripheral blood mononuclear cells. Viable Lactobacillus acidophilus or Streptococcus thermophilus has been shown to play a protective role in epithelial barrier integrity in cultured human intestinal Caco-2 monolayers. However, whether active lactic acid bacteria used for fermented products could improve the intestinal mucosal barrier function in diarrhea-predominant IBS is unknown. Viable lactic acid bacteria can't be added to food because they induce fermentation which change the taste, texture and freshness of food and are not easy to store. For these reasons, the living bacteria have only been used in a very narrow range of products such as yogurt or fermented milk. So, the fermented milk or yogurt is qualified for related researches in human.Objectives1. To evaluate whether the abnormal intestinal mucosal barrier function is related to the severity of colonic inflammation and clinical symptoms; to observe alterations of the colonic epithelial junctional complex and two main tight junction molecule (ZO-1 and occludin) in patients with diarrhea-predominant IBS.2. To determine the possible efficacy of active lactic acid bacteria on the increased intestinal permeability in patients with diarrhea-predominant IBS.3. To observe the possible effect of the heat-killed mixture of lactic acid bacteria (Mix-LAB) (L. bulgaricus, S. thermophilus, L. acidophilus) on epithelial barrier dysfunction induced by proinflammatory cytokines in human intestinal Caco-2 monolayers; to observe whether the heat-killed Mix-LAB could preserve the proinflammatory cytokines-induced alterations of ZO-1 and occludin in Caco-2 monolayers; to explore the possible effect of heat-killed Mix-LAB on proinflammatory cytokines-induced the increase nitric oxide (NO) production and inducible nitric oxide synthase (iNOS) mRNA expression.Methods1. Thirty patients who met the Rome II criteria for diarrhea-predominant IBS were investigated. Twelve healthy controls without any digestive complaints were recruited. The gut mucosa barrier integrity was evaluated by the intestinal permeability. Intestinal permeability was measured using a triple sugar test (lactulose, mannitol and sucralose). After an overnight fast, patients and healthy volunteers were asked to empty their bladders and then were given a test solution containing 10 g lactulose, 5 g mannitol and 5 g sucralose in 100 ml tap water. Urine was then collected for the following 24 hours. The total volumes of urine collected during the first 5h and the 24h were measured and recorded. The urinary sugar concentrations were determined by gas chromatography. The ratio of lactulose and mannitol recovery (L/M) during the first 5 h was used as the index of the small intestinal permeability and the total mass of sucralose excretion (mg) during the 24 h was as the index of colonic permeability. Colonoscopic biopsies were collected from the rectosigmoid for analysis the alterations of distribution and expressions of ZO-1 and occludin proteins, and epithelial junctional complex visualized by transmission electron microscopy. The severity of colonic inflammation was evaluated by the numbers of CD3 positive and CD25 positive cells. The global IBS scores in this study were calculated by the sum of scores of two subscales of Abdominal pain and Diarrhea in Gastrointestinal Symptom Rating Scale. To avoid possible adverse effect of purgation on permeability test, the triple sugar test came first and then the colonoscopy examination.2. Treatment was employed in a randomized single blind placebo controlled study with 30 patients with diarrhea-predominant IBS for 4 weeks. Patients took either fermented milk (AB100 Bright Dairy, China) 190g twice daily or placebo milk beverage(Bright Dairy, China) 200ml twice daily. The fermented milk contains 4 strains of active lactic acid bacteria Streptococcus thermophilus, Lactobacillus bulgaricus, Lactobacillus acidophilus and Bifidobacterium and the milk beverage has identical color and consistency to fermented milk but has no bacteria. One of the researchers performed general lactobacillus counts on a sample of the probiotic fermented milk to confirm they were active. The clinical symptoms and intestinal permeability were evaluated at week 0 and week 4. The intestinal permeability which has the ability to assess the gut mucosa integrity was measured by a triple sugar test. After an overnight fast, patients and healthy volunteers were asked to empty their bladders and then were given a test solution containing 10 g lactulose, 5 g mannitol and 5 g sucralose in 100 ml tap water. Urine was then collected for the following 24 hours. The total volumes of urine collected during the first 5h and the 24h were measured and recorded. The urinary sugar concentrations were determined by gas chromatography. The ratio of lactulose and mannitol recovery (L/M) during the first 5 h was used as the index of the small intestinal permeability and the total mass of sucralose excretion (mg) during the 24 h was as the index of colonic permeability. The IBS symptoms were assessed by the Gastrointestinal Symptom Rating Scale (GSRS) questionnaire and the global IBS scores in this study were calculated by the sum of scores of two subscales of Abdominal pain and Diarrhea in GSRS. The clinical symptoms were assessed and agreed by at least two investigators. Patients were also asked to complete a daily diary to evaluate symptoms (abdominal pain, abdominal bloating and sensation of flatulence) on 100mm visual analogue scale (VAS) scales. The primary endpoint was the improvement of proportions of patients with abnormal intestinal permeability at the end of four-week treatment. The proportions in two treatment groups were compared with baseline. The secondary endpoint was the improvement of IBS symptoms. Twelve healthy volunteers without any digestive complaints were recruited in this study as normal controls. 3. Three LAB strains of L. bulgaricus, S. thermophilus and L. acidophilus were heat killed in a water bath at 65℃for 60 min and then mixed and diluted to the appropriate concentration in fetal calf serum and antibiotics free Dulbecco's modified essential medium(DMEM). The concentration of Mix-LAB was the whole bacterial cells with cell counts of S. thermophilus, L. bulgaricus and L. acidophilus combined in a ratio of 1: 1: 1. Human intestinal Caco-2 monolayers were used 14-21 days after reaching confluence. Cytomix was a mixture containing 10 ng/ml interferon-γ, 10 ng/ml tumor necrosis factor-a and 1 ng/ml interleukin-1β. Caco-2 monolayers were treated with control culture medium, or with cytomix or cytomix plus heat-killed Mix-LAB (10⁹/ml) for 72h. The control culture medium is fetal calf serum and antibiotics free DMEM. At the end of the 72h experimental period, the cells were harvested for analysis the expressions and distribution of ZO-1 and occludin proteins, and the NO production was expressed by nitrate (NO3^-) plus nitrite (NO2) concentration in the culture supernatants. The epithelial barrier function was evaluated by permeability assay of small molecular weight fluorescein sodium after 24h, 48h and 72h of the initial incubation in Transwell compartment. Monolayer permeability was expressed as a clearance calculated using the formula [L]×v/([A]×A×t) (nl/cm²/h) where [L] is the concentration of fluorescein sodium in basolateral compartment, V is the volume of medium in basolateral compartment, [A] is the concentration of fluorescein sodium in apical compartment (0.2g/ml), t is time (h) and A is the area of filter membrane(cm²). To determine the dose dependent effect on epithelial permeability, the graded concentrations of Mix-LAB (10⁷-10⁹/ml) were tested.Results1. The increased small bowel permeability was defined as the ratio of the L/M > 0.025 and the upper limit of normal colonic permeability was 42.1 mg, defined as the P95 of total urinary sucralose excretion in 12 healthy volunteers in the present study. The proportion of patients with increased small bowel permeability was 60% (18/30) and the proportion of patients with increased colonic permeability was 53.3% (16/30). In the patients with diarrhea-predominant IBS, the small intestinal permeability was increased at 0.0377 (0.026) compared with controls, 0.0182 (0.006) (P=0.002); the colonic permeability was increased at 44.3(43.7) compared with controls, 31.36(10.7) (P=0.028). The small bowel or colonic permeability did not correlated with the number of intraepithelial lymphocytes (r=-0.07, P=0.714; r=-0.143, P=0.450, respectively), or lamina propria CD₃⁺ cells (r=-0.325, P=0.08; r=-0.056, P=0.768, respectively) or CD₂₅⁺ scores (r=0.302, P=0.105; r=-0.242, P=0.197, respectively). The small bowel or colonic permeability did not correlated with the total IBS scores (r=0.153, P=0.250; r=0.025, P=0.895, respectively), or the number of bowel movements everyday (r=0.01, P=0. 941; r=0.106, P=0.579, respectively). The transcription levels of ZO-1 and occludin in patients with diarrhea-predominant IBS decreased compared with normal controls (0.573 (0.734) versus 1.016 (0.47), P=0.038; 0.393 (1.32) versus 0.856 (1.46), P=0.027, respectively). The proportions of patients with different staining intensity of ZO-1 and occludin were significantly different from the normal controls (Chi Square=9.905, P=0.042; Chi Square=13.238, P=0.010, respectively). However, the distributions of ZO-1 and occludin proteins did not alter. There were 9 patients with strongly positive staining intensity of ZO-1 protein, 8 patients with positive staining intensity, and 13 patients with moderate positive staining intensity. There were 6 patients with strongly positive staining intensity of occludin protein, 7 patients with positive staining intensity, and 17 patients with moderate positive staining intensity. The staining of junctional complex among colonic enterocytes visualized was strong and continuous in normal controls; however, the staining was faint and discontinuous but not widened in 33.3 percent of 30 patients with D-IBS and it appeared continuous and strong in the other 66.7 percent patients.2. At week 4, the proportion of patients with increased small intestinal permeability in the probiotics group significantly decreased compared with baseline (28.6% versus 64.3%, P=0.023) while the proportion of patients with increased colonic permeability did not changed (57.1% versus 50.0%, P=0.705). After treatment with probiotics, the median of small bowel permeability significantly decreased from 0.0380(0.024) at baseline to 0.0228(0.020) (P=0.004). However, the increased colonic permeability improved neither in probiotics group nor in placebo group at week 4. Treatment with probiotics significantly decreased the mean global IBS scores in GSRS compared with baseline (9.62±1.05 versus 7.64±1.24, P<0.001). Analysis of the diary VAS data in the probiotics group showed that the mean scores of abdominal pain and flatulence were significantly lower than baseline (37.76±5.87 versus 30.11±7.71, P<0.001; 36.61±6.04 versus 32.50±8.11, P=0.010, respectively) and the sensation of bloating failed to show any improvement. At the end of probiotics treatment, the evolution value of small bowel permeability did not correlated with the evolution value of the total IBS scores in GSRS and the evolution of abdominal pain in VAS (r=0.343, P=0.229; r=0.385, P=0.202, respectively). There were no reported adverse events related to the study drinks.3. The paracellular permeability of Caco-2 monolayers to fluoresceinated dextran was significantly increased following incubation with cytomix for 24h (56.32±5.04), 48h (76.03±5.08), or 72h (85.08±3.11) when compared with controls at the respective same time point. The protective effect of heat-killed Mix-LAB on increased permeability induced by cytomix was concentration-dependent, being statistically significant for concentrations equal to or greater than 10⁸/ml after 48-h incubation (52.34±1.85 versus 76.03±5.08 , P<0.001) and 10⁹/ml after 24-h incubation (26.4±2.56 versus 56.32±5.04, P<0.001). Following 72 h of incubation with cytomix in Caco-2 monolayers, the mRNA expression levels of ZO-1 and occludin significantly decreased compared with controls (0.065(0.072) versus 1.016(2.625), P<0.001; 0.013(0.019) versus 0.908(1.672), P<0.001, respectively); the expression of ZO-1 and occludin proteins decreased to about 27%, 25% of the control levels, respectively. With the addition of heat-killed Mix-LAB (10⁹/ml), normal mRNA and protein expression levels of both ZO-1 and occludin were largely preserved. Cytomix altered distribution of ZO-1 and occludin proteins, but this effect was prevented by the heat-killed Mix-LAB(10⁹/ml). After coincubated with cytomix for 72h, the relative mRNA expression of iNOS and the production of NO in Caco-2 were higher normal conrtols (25.01(47.2) versus 1.22(2.0), P<0.001; 10.66±2.03 versus 3.38±1.31, P<0.001, respectively). The addition of Mix-LAB (10⁹/ml) in the medium normalized the relative mRNA expression of iNOS and production of NO.Conclusions1. In patients with diarrhea-predominant IBS, both the small bowel and colonic mucosal barrier function were abnormal. Both the small bowel and colonic permeability did not related to the severity of colonic inflammation and clinical symptoms.2.The colonic epithelial junctional complex was abnormal in some patients. The main tight junctional molecules expressions were down at transcription and protein levels in some patients with diarrhea-predominant IBS, but the distribution of ZO-1 and occludin proteins did not altered.3.Short term active lactic acid bacteria treatment for patients with diarrhea-predominant IBS improved small bowel mucosal barrier function and clinical symptoms, but failed to improve the colonic mucosal barrier function. There were no significant correlations between the improvement of small bowel permeability and IBS symptoms relief. This indicates that active lactic acid bacteria may exert beneficial effects on barrier function and symptoms through different mechanisms in diarrhea-predominant IBS.4.These data suggest that heat-killed Mix-LAB plays a protective role in proinflammatory cytokines-induced intestinal epithelial barrier dysfunction and this effect was dose-dependent. Heat-killed Mix-LAB restored proinflammatory cytokines-induced disruption of tight junction. The underlying mechanism includes heat-killed Mix-LAB inhibited proinflammatory cytokines-induced iNOS gene expression and increased NO production.