Iordanis Karagiannidis

Serum and Colonic Mucosal Immune Markers in Irritable Bowel Syndrome

OBJECTIVES:Low-grade colonic mucosal inflammation has been postulated to have an important role in the pathophysiology of irritable bowel syndrome (IBS). The objectives of this study were (i) to identify serum and tissue-based immunological and neuroendocrine markers associated with mucosal inflammation in male (M) and female (F) patients with non-post-infectious IBS (non-PI-IBS) compared with healthy controls and (ii) to assess possible correlations of such markers with IBS symptoms.METHODS:Sigmoid mucosal biopsies were obtained from 45 Rome II positive IBS patients without a history of PI-IBS (26 F, 35.5% IBS-C, 33.3% IBS-D, 31.1% IBS-A/M) and 41 healthy controls (22 F) in order to measure immunological markers (serum cytokine levels, colonic mucosal mRNA levels of cytokines, mucosal immune cell counts) and neuroendocrine markers associated with mucosal inflammation (corticotropin releasing factor- and neurokinin (NK)-related ligands and receptors, enterochromaffin cells). Symptoms were measured using validated questionnaires.RESULTS:Of all the serum and mucosal cytokines measured, only interleukin-10 (IL-10) mRNA expression showed a group difference, with female, but not male, patients showing lower levels compared with female controls (18.0±2.9 vs. 29.5±4.0, P=0.006). Mucosal mRNA expression of NK-1 receptor was significantly lower (1.15±0.19 vs. 2.66±0.56, P=0.008) in female, but not male, patients compared with healthy controls. No other significant differences were observed.CONCLUSIONS:Immune cell counts and levels of cytokines and neuropeptides that are associated with inflammation were not significantly elevated in the colonic mucosa of non-PI-IBS patients, and did not correlate with symptoms. Thus, these findings do not support that colonic mucosal inflammation consistently has a primary role in these patients. However, the finding of decreased IL-10 mRNA expression may be a possible biomarker of IBS and warrants further investigation.

547 Effects of Substance P on Pro and Anti-Inflammatory Responses of Human Mesenteric Preadipocytes Isolated From IBD Patients

Oxytocin (OT) is best known for its hormonal roles in milk let down and parturition. It has also been found in neurons that project within the brain and plays roles in mediating the central effects of nurturing stimuli. We have recently demonstrated that a subset of enteric neurons expresses OT and that about 70% of adult enteric neurons express the oxytocin receptor (OTR). OTR expression, which is developmentally regulated, also occurs in crypt epithelial cells where it concentrates at junctional complexes, particularly at the crypt-villus boundary. We tested hypotheses that endogenous enteric OT signaling affects intestinal motility, inflammation, and mucosal homeostasis. Dry stool mass and fecal water content were found to be greater in mice lacking OTR (OTRKO) than in WT mice. Total gastrointestinal transit time in OTRKO animals was also faster than that of WT animals; however, the time OTRKO mice required to expel glass beads inserted into their rectum was greater than that of WT animals. The severity of 2,4,6-trinitrobenzene sulfonic acid (TNBS)-induced colitis in OTRKO mice was more severe compared to that in WT animals. Clinical scores (aggregating weight loss, stool blood and consistency), histological damage scores, and abundance of transcripts encoding pro-inflammatory cytokines and chemokines, were all significantly greater in OTRKO than in WT mice. Administration of exogenous OT, moreover, ameliorated TNBS-induced colitis in WT mice. Absorption of FITC-dextran was greater in OTRKO than in WT, suggesting that intestinal macromolecular permeability is increased in the absence of OT signaling. Exudative enteropathy, however, did not occur in OTRKO mice and following its intravenous infusion, horseradish peroxidase was found to be trapped at tight junctions between enterocytes. Villus height and crypt depth were each smaller in the intestines of OTRKO than WT mice, suggesting that signaling by crypt epithelial OTR plays a role in mucosal homeostasis. The current study supports the idea that enteric oxytocinergic signaling is physiologically significant; it slows intestinal propulsion, modulates inflammation, dampens macromolecular permeability, and contributes to the maintenance mucosal homeostasis. Observations are consistent with an overarching hypothesis that oxytocinergic signaling counterbalances effects of other molecules, such as corticotrophin releasing hormone, which mediate effects of stress. If so, then congenital or acquired defects in enteric oxytocinergic signaling may decrease resistance to stress and thus contribute to the pathophysiology of intestinal disorders, such as irritable bowel syndrome or inflammatory bowel disease. Development of means of initiating or enhancing oxytocinergic signaling may also be useful in treating stress-related disorders.

Mo1609 Obesity Drives Colon Tumor Development in AOM-Treated Mice

Background/Aims: Obesity is an important risk factor for colon cancer in humans. Numerous studies in experimental animals demonstrated that a high fat diet enhances colon cancer development. A potential mechanism for this increased risk is that adipose tissue, a major source of adipokines, growth factors, and cytokines promotes tumor growth. Alternatively, a high-fat diet causes metabolic changes that can independently enhance tumor growth. Here we determined the independent effects of these factors on colon tumor development. Methods: C57BL/6J male mice were fed regular chow or high fat diet (HFD) for 8 weeks. Half of the animals on high fat diet were switched to regular chow, and half on regular chow were switched to the HFD with the resulting 4 groups being regular (R), HFD (H), regular switched to HFD (RH) and HFD switched to regular (HR). One week after the switch, 5 weekly doses of azoxymethane (AOM) were given to induce colon tumors. Mice were euthanized 1 day after the final dose of AOM to analyze gene expression, proliferation, and apoptosis in the colon or 25 weeks after AOM to analyze numbers of colon tumors and aberrant crypt foci (ACF). Body composition was measured by echo MRI. Results: We previously showed the significantly increased tumor multiplicities in all groups that received HFD (groups H, HR, and RH) relative to animals on regular diet (group R). In this study, an ELISA based array analysis of cytokines and growth factors in serum revealed that obesity was associated with differential expression of molecules involved in insulin-like growth factor (Igf) signaling. Relative to group R mice, group HR mice had ratios 1.5, 0.7, and 0.75 in serum concentrations of Igf-II, Igf binding protein (Igfbp) 5, and Igfbp6, respectively (p = 0.05 for each). These differences were also reflected in mRNA expression patterns in mesenteric adipose tissue, as all obese groups had significantly elevated expression of Igf-I and Igf-II and significantly depressed expression of Igfbp5 and 6. Group HR mice appeared to have a greater toxic response to AOM relative to all other groups as measured by apoptotic index and lengthening of the Ki67-positive proliferative zone in colonic crypts following AOM treatment. Conclusions: We have identified obesity itself as a tumor promoter. A mechanism for the increased tumor responses may involve heightened signaling through the Igf1-receptor due to increased expression of ligands and decreased expression of binding proteins in mesenteric adipocytes. Additionally, the greater tumor response of mice in HFD that were switched to regular diet likely involves a heightened toxic response to AOM. Supported by NIH RO-1 DK60729 (CP); A Pilot and Feasibility Study from CURE: Digestive Diseases Research Center 41301; and the Martin Blinder Fund for IBD Research

Colitis-Associated Adipose Tissue Responses to Corticotropin-Releasing Hormone Family of Neuropeptides

Background and Aims: Numerous studies have indicated that adipose tissue is an active endocrine organ and several adipose-associated molecules have been implicated as potential mediators of intestinal inflammation. Results from our group and others indicate that corticotrophin releasing hormone (CRH or CRF), and its peptide family members urocortins (UCNs) participate in the pathophysiology of intestinal inflammation and inflammatory bowel disease (IBD). Recent evidence also indicates that CRH receptors 1 and 2 are expressed by human adipocytes. Whether these ligands and receptors are expressed in the mesenteric fat and are modulated during colitis has never been studied. Here we compared levels of expression of the CRH peptide family at in the mesenteric fat depots of mice before and after experimental TNBS-induced colitis. We also exposed isolated human mesenteric preadipocytes to CRH and measured changes in cytokine production at the mRNA and protein levels. Methods: C57BL6 mice (n=10 per group) were injected intracolonically with either vehicle (40% ethanol, control) or TNBS (5 mg/20 grams) for 48 hr. Mesenteric adipose tissue was harvested from the mice and mRNA was isolated and converted to cDNA for real time RT PCR. Human mesenteric pre-adipocytes grown in culture were stimulated with 100nM CRH for 8 hours. Supernatants were collected for protein array and ELISA while mRNA was isolated for conversion to cDNA and use in real time RT PCR experiments. Data were analyzed using a two tailedMannWhitney test. Results: CRHmRNAwas significantly increased inmesenteric fat of TNBS-exposed mice compared to controls (by 112.9 fold, p<0.01). UCNs 2 and 3 were also significantly increased in these tissues (by 30.7 and 9.2 fold, p< 0.0016 and p< 0.0052, respectively). Protein arrays on CRH stimulated human mesenteric preadipocytes showed increased secretion of TIMP-2 and Rantes (CCL5 gene product) and decreased MIF, MCP-3 and PAI-1 levels. Changes in released MIF were also confirmed by ELISA in the medium of preadipocytes (p< 0.01). Conclusions: Increased expression of CRH and urocortins in adipose tissue during colitis together with the ability of CRH to induce changes in cytokine release from isolated adipocytes suggest that the CRH family of peptides may mediate inflammatory responses in mesenteric fat and affect the development of intestinal inflammation. These findings may represent a novel signaling pathway in colitis and could have important implications for inflammatory bowel disease treatment targets. Research Support: NIH NIDDK T32 DK07180-36 Gastroenterology Training Grant and P01 DK 33506

T2072 Colonic Mucosal Inflammation is Unlikely to Play a Primary Role in Irritable Bowel Syndrome

Introduction: An enhanced mucosal immune or inflammatory response has been postulated to play a key role in the pathophysiology of IBS. Previous studies have reported an increase in serum cytokines and colonic mucosal lymphocyte and mast cell counts in IBS patients vs. healthy controls (HCs), but data are inconsistent and limited. Aims: 1) To compare inflammatory markers in the blood and colonic mucosa in IBS and HCs, 2) To determine if these measures are affected by sex, and 3) To explore if these measures are related to IBS symptoms.Methods: Male (M) and female (F) Rome positive IBS patients with stable disease activity and HCs underwent a sigmoidoscopy with colon biopsies (taken at 30 cm from the anal verge). Blood samples were also collected. The following was measured: 1) serum levels of cytokines (IL-1β, IL-6, IL-8, IL-10, and TNF-α), 2) mucosal mRNA levels of the same cytokines, CRF, and CRF1 receptor using real-time PCR, and 3) lymphocyte and mast cell counts per total biopsy area using an automated system. All subjects completed questionnaires assessing GI and non-GI symptoms. Results: 45 IBS patients (26F, 19M) and 41 agematched HCs (22F, 19M) were studied. IBS subtypes were 34% IBS-C, 34% IBS-D, and 32% IBS-M/A. None of the IBS patients had documented post-infectious IBS. Overall, there were no statistically significant differences in any of the measured inflammatory markers between IBS patients and HCs. However, group differences were seen amongst female subjects. Compared to healthy women, female IBS had a small but significant mean increase in serum level of the pro-inflammatory cytokine TNF-α (4.4±0.08 vs. 4.1±0.06 pg/ml, p= 0.046). They also had lower mucosal mRNA levels of the anti-inflammatory cytokine IL-10 compared to HCs (4.9±0.1 vs. 5.5±0.2, p=0.02). However, neither finding maintained statistical significance when corrected for multiple comparisons. Serum cytokine levels did not correlate with their respective mucosal cytokine mRNA levels (p=NS). There were no significant differences in CRF colonic mucosal expression or cell counts between IBS and controls. Serum IL-10 levels (r=-0.44, p = 0.006) and colonic mucosal CRF expression (r=0.45, p=0.003) negatively correlated with current symptom ratings. Conclusion: Taken together, the lack of strong and significant differences in mucosal and serum immune markers between IBS andHCs argues against the presence of a predominant proinflammatory response within the colonic mucosa in IBS. Further studies are needed to determine if an immune disturbance plays a more significant role in women with IBS, and if immune markers are associated with symptom flares in IBS.