AGA Abstracts The mechanisms of berberine anti-inflammatory effects in experimental colitis are mediated by PLA2G4A and PNPLA8 in colonic epithelial and macrophage through JNK and p38 MAPK signaling pathways. Su1807 POST-OPERATIVE RECURRENCE AFTER ILEO-CAECAL RESECTION FOR CROHN'S DISEASE: TOWARDS AN ANTI-ADHERENT INVASIVE ESCHERICHIA COLI (AIEC) STRATEGY WITH RATIONALY SELECTED SACCHAROMYCES CEREVISIAE PROBIOTIC Caroline Valibouze, Silvia Speca, Thomas Lambin, Caroline Dubuquoy, Laurent Dubuquoy, Lena M'Ba, Lucil Schneider, Christel Rousseaux, Ballet Nathalie, Decherf Amelie, Marie Titecat, Benoit Foligne, Pierre Desreumaux, Christel Neut, Philippe Zerbib Background and aims: AIEC are found in 30% of patients with Crohn's disease (1). Their roles in post-surgical recurrence frequency and severity remain elusive. Saccharomyces cerevisiae CNCM I-3856 (SC) inhibits AIEC adhesion to epithelial cells and decreases AIEC colonization in mice (2). We investigated the influence of AIEC and SC on post-surgical anastomotic recurrence in transgenic (Tg) HLA-B27 and control non-transgenic rats. Meth- ods: Ileo-caecal resection with end to end anastomosis were performed at 12 weeks of life in 49 rats (28 Tg, 21 non Tg) receiving AIEC alone daily at 10 9 CFU/days (8 Tg, 5 non Tg), SC alone daily at 10 9 CFU/days (7 Tg, 7 non Tg) or AIEC + SC (both at 10 9 CFU/ days, 13 Tg and 9 non Tg). Animals were sacrificed 6 weeks after surgery. Recurrence was blindly assessed macroscopically (score from 0 to 4, recurrence for score ≥2) and histologically (score from 0 to 12) by 2 operators (mean ± SEM). Anastomotic mucosal AIEC was quantified at surgery and sacrifice in all animals (mean ± SEM). Results: Neither macroscopic nor histologic lesions were found in the 21 non Tg rats. AIEC given 1 week before surgery was associated with 87.5% macroscopic recurrence scored 3 ± 0.4 and histologic lesions scored 7.6 ± 0.5. Preventive treatment with SC begun 2 weeks before surgery strongly decreased AIEC-induced post-operative recurrence both macroscopically (46% of recurrence scored 1.4 ± 0.3, respectively p=0.08 and p=0.01) and histologically (3.8 ± 0.7, p=0.005). Mucosal anastomotic AIEC quantification was significantly lower at sacrifice compared to surgery in animals receiving AIEC + SC (Log 2.4 ± 0.06 CFU/g vs Log 3.2 ± 0.2, p=0.001) without any modification in Tg rats receiving AIEC alone (Log 3.5 ± 0.5 CFU/g vs Log 2.9 ± 0.2, p=0.64). No significant anastomotic lesion was observed in Tg rats receiving SC alone. Conclusion: Oral administration of AIEC induced postoperative recurrence in HLA-B27 Tg rats which was prevented by SC administration. Saccharomyces cerevisiae CNCM I-3856 could be a new therapeutic perspective in the prevention of post-operative recurrence in Crohn's disease patients. 1. Darfeuille-Michaud A et al. High prevalence of adherent-invasive Escherichia coli associated with ileal mucosa in Crohn's disease. Gastroenterology. 2004 Aug;127(2):412–21. 2. Sivignon A et al. Saccharomyces cerevisiae CNCM I-3856 prevents colitis induced by AIEC bacteria in the transgenic mouse model mimicking Crohn's disease. Inflamm Bowel Dis. 2015 Feb;21(2):276–86. Su1808 DOWNREGULATION OF RALGTPASE-ACTIVATING PROTEIN LEADS TO EXACERBATION OF MURINE DEXTRAN SULFATE SODIUM-INDUCED COLITIS Tomoya Iida, Naoki Minami, Minoru Matsuura, Ryutaro Shirakawa, Hisanori Horiuchi, Hiroshi Nakase Background and Aim: Small GTPases are molecular switches of cellular signaling. Normally, they are activated by GDP/GTP exchange mediated by the guanine nucleotide exchange factors (GEFs) and inactivated by GTP hydrolysis mediated by the GTPase activating proteins (GAPs). Ral is a member of Ras family proteins and known to regulate tumorigenesis and invasion/metastasis of various cancers. However, the contribution of Ral to intestinal inflammation has not been investigated. The objective of the study is to understand how Ral can contribute to intestinal inflammation. Methods: We used RalGAPα2 knockout (KO) mice (C57BL background) that exhibited Ral activation and wild-type (WT) mice. To induce acute colitis, 2% dextran sulfate sodium (DSS) in sterile water was administered for 5 days, and then regular drinking water was given for next 5 days. Mice were sacrificed on day 10 to evaluate the acute inflammatory phase. Also, we compared the difference of the cytokine production and phagocytosis of bone marrow-derived macrophages (BMDMs) stimulated with lipopolysaccharide between RalGAPα2 KO mice and WT mice. Next, to elucidate what kinds of cells mainly contribute to intestinal inflammation, bone marrow (BM) chimeric mice were generated as follows: WT→WT, WT→RalGAPα2 KO, RalGAPα2 KO→WT, RalGAPα2 KO→RalGAPα2 KO. These mice were exposed to DSS and the histology of the colon was examined. We drew the difference of the expression of matrix metalloproteinase (MMP)-9 in BMDMs between RalGAPα2 KO mice and WT mice and evaluated the effect of MMP-9 inhibitor treatment on DSS-induced colitis in both mice. Results: After induction of DSS, RalGAPα2 KO mice significantly had more disease severity and higher expression of TNF-α and IL-1β in the colonic tissues than WT mice. No difference in the cytokine production and the phagocytic ability of BMDMs between RalGAPα2 KO mice and WT mice was observed. The study with BM chimeric mice demonstrated that RalGAPα2 KO→Ral- GAPα2 KO chimeras had the most severe disease activity among all the chimeras and RalGAPα2 KO→WT chimeras had more disease severity than WT→RalGAPα2 KO chimeras and WT→WT chimeras. These data suggested that Ral activation of immune cells is domi- nantly involved in exacerbation of DSS-induced colitis in RalGAPα2 KO mice. The MMP- 9 gene expression and its activity in the colonic tissues of RalGAPα2 KO mice were signifi- cantly higher than WT mice and MMP-9 inhibitor treatment ameliorated disease severity of RalGAPα2 KO mice with DSS-induced colitis. Conclusions: Our data suggest that Ral activation contributes to intestinal inflammation by induction of MMP-9 expression and its activation. Therefore, blocking Ral activation could be a promising strategy for IBD treatment. S-620 AGA Abstracts Su1809 A COMPARISION STUDY OF SYSTEMIC VERSUS TARGETED INTRACECAL ANTI-TNFα MONOCLONAL ANTIBODY IN AN ADOPTIVE T-CELL TRANSFER MUINE MODEL OF COLITIS Shaoying N. Lee, Sharat Singh, Allison Luo, William J. Sandborn, Emil Chuang, Mitchell L. Jones Introduction: Systemically administered anti-TNFα antibodies, such as infliximab and adali- mumab, have revolutionized the treatment in ulcerative colitis ad Crohn's disease. However, recent publications suggest inadequate drug reaches diseased tissue, when compared to the TNFα burden present in the colon tissues of patients with active disease. It is suggested that this may explain the lack of efficacy observed in some subjects. The objective of this study was to assess whether intracecally (IC) delivered anti-mouse-TNFα monoclonal antibody might penetrate the mucosa and confer efficacy when compared with systemic intraperitoneal (IP) injection in an adoptive T-cell transfer induced chronic colitis murine model. Methods: All animals underwent surgical implantation of a cecal cannula for the ease of bolus topical delivery to the cecum. Colitis was induced by IP injection of CD44-/ CD62L+ T-cells isolated and purified from C57Bl/6 donor to the RAG2-/- recipient mice on Day 0. To minizine variation due to different routes of administration, animals were treated with both IP every third day (Q3D) and IC once daily (QD) of either the test article or controls (Vehicle or IgG) from Day 0 to 42. Mice underwent colonoscopy on Days 14, 28, 42 to assess colitis severity and stool consistency. Blood and tissues were collected for bioanalysis and histopathologic analysis at the Day 42 termination. Results: Significant body weight loss was observed in groups treated with vehicle or IgG control starting at Day 7 to 14. The Day 13 T-cell engraftment rate was 98% indicating successful induction of colitis. Treatment with either IP or IC of anti-TNFα antibody led to significant reduction of disease activity index (DAI), a combination score of body weight loss, colitis severity and stool consistency observed via colonoscopy, when compared to vehicle control (IP/IC). Similar and significant decrease of key inflammatory cytokines, including TNFα, IL-17A, IL-4 and IL-22, was observed with anti-TNFα treatment irrespective of the route of administration when compared to controls in colon tissue. Histopathology revealed effective induction of ileitis and colitis by adoptive T-cell transfer. Administration of anti-TNFα antibody through either IP or IC resulted in mean reductions in total histopathologic score, however only IC treated group showed a significant improvement when compared to vehicle control. Conclusion: In conclusion, this study has demonstrated targeted IC administration of anti-TNFα antibody was superior to traditional systemic IP administration at reducing histopathological damage of colon tissue in this chronic colitis model. These findings provide a proof of concept for topical delivery of anti-TNFα antibody to colonic tissue and show the potential for improved efficacy and safety in the treatment of inflammatory bowel disease (IBD). Figure 1: Shows Disease Activity Index (DAI) (mean % ± SD) (a combination of colitis score and stool score via video endoscopy plus body weight loss score) at study endpoint (42 day) in groups receiving T-Cell transfer (CD44-/CD62L+) and administration of anti- mouse TNFα antibody (IP or IC) or Antibody Control (IP or IC) or Vehicle control (IP and IC). Pair-wise comparisons by two-tailed Mann-Whitney U-Test for treatment effects; p<0.05*; p<0.01**, p<0.001***, and p<0.0001****