AGA Abstracts 0.002) and downregulation of Mmp-2 (FC=3.3, p=0.020), Mmp-9 (FC=3.1, p=0.030), Cldn1 (FC=3, p=0.040) compared to control mice. After chronic DSS administration, Xpnpep2 (FC=4.2, p=0.020) was upregulated in TIMP-1 KO mice compared to control mice. Young TIMP-1 KO mice (10-12 weeks) versus older animals (19-21 weeks) showed significantly increased expression of Reg3g (FC=16, p=0.030), Reg3b (FC=12.6, p=0.040) and Ido1 (FC= 3.5, p=0.008), whereas the expression levels of Mir200b (FC=2.2, p=0.009), Ctse (FC=1.3, p=0.030) and Wnt2b (FC=1.1, p=0.002) were decreased. Conclusion: TIMP-1 deficiency leads to upregulation of anti-bacterial and innate immunity genes, resulting in an attenuated development of acute colitis. In a chronic setting of inflammation, TIMP-1 KO mice have less remodeling and fibrosis. Unraveling the role of TIMP-1 in extracellular matrix remodeling will be necessary to understand the biology of intestinal wound healing and fibrosis in IBD. Mo1757 Evidence for a Role of Superantigens in the Pathogenesis of Ulcerative Colitis- Associated Primary Sclerosing Cholangitis Joel Pekow, Kelli Williams, Dustin Shaw, Katherine Meckel, Kathryn Lesko, David T. Rubin, Russell D. Cohen, Atsushi Sakuraba, Ira M. Hanan, Sushila Dalal, Kelly Monroe, Xiuliang Bao, Jianzhong Hu, Anabella G. Castillo, Joseph Odin, Jennifer Kwon, Rodney Newberry, Steven H. Itzkowitz, Jean-Frederic Colombel, Bana Jabri Introduction: Primary sclerosing cholangitis (PSC) patients who have inflammatory bowel disease (IBD) exhibit a homogeneous phenotype of IBD with mild colonic disease activity and a high risk of colon cancer suggesting a distinct pathogenesis. The association of HLA genotypes with PSC implies that a common antigen may be influencing disease development. Gut T cells are highly clonal effector memory cells thought to expand from microbial antigen exposure. We hypothesized that antigenic exposure drives expansion of a T-cell population unique to IBD patients with PSC, and our aim was to identify potential antigens driving disease pathogenesis in PSC using the colonic mucosal T cell receptor (TCR) repertoire as a tool. Methods: Ascending colon biopsies were obtained and placed in RPMI1640 medium in 5 subjects with ulcerative colitis and PSC (UC+PSC), 2 with Crohn's disease and PSC (CD+PSC), 2 with UC without PSC (UC), and 4 normal controls (NC). Lamina propria lymphocyte (LPL) cells were isolated, and TCRVβ2+, TCRVβ5.1+, and TCRVβ2(-)b5.1(-) T cell populations were sorted by FACS and stimulated in short-term cultures. After T cell expansion, 0.5x10 6 TCRVβ5.1+ or TCRVβ2+ cells were lysed and RNA extracted. PCR and TA cloning were performed using standard techniques. TCRVβ5.1 and TCRVβ2 sequences were analyzed using Gene Codes Sequencher software and IMGT/V-Quest. Results: Prelimi- nary analysis of TCR Vβ classifications from LPLs revealed that Vβ5.1 and Vβ2 were highly expressed in IBD patients with PSC (>5%). To determine if a particular antigen was driving expansion of T cells containing TCRVβ5.1 and TCRVβ2, CDR3 sequencing was performed in these cell lines. In contrast to the highly clonal TCR repertoire normally found in the intestine and as seen in NC, UC without PSC, and CD+PSC, sequencing of TCRVβ5.1+ cells demonstrated that UC patients with PSC have increased diversity of the CDR3 region. The percentage of the most abundant transcript by sequencing compared to the total number of transcripts was 22.8% in UC+PSC, 72.5% in CD+PSC, 70.5% in UC without PSC, and 73.5% in NC (p=0.009). When the J-segment of the TCRVβ5.1 chain was analyzed, the greatest diversity was also seen in UC+PSC. Additionally, a high percentage of J-segment usage in PSC patients (UC+PSC and CD+PSC) had the TRBJ2-7 segment. In contrast, analysis of TCR Vβ2 did not reveal differences in TCR diversity between the groups. Conclusion: UC patients with PSC demonstrate increased CDR3 diversity of TCRVβ5.1 and a higher percentage of usage of TRBJ2-7 J segment compared to IBD without PSC and normal controls. This finding suggests that exposure to an as yet identified superantigen is driving the T cell expansion unique to UC patients with PSC and that it likely binds to the TCRVβ5.1 CDR3 region, consistent with type IV superantigens. S-704 AGA Abstracts Mo1758 Impact of Intestinal Inflammation on FOXP3 Regulatory T-Cells Gilles Boschetti, Remi Duclaux-loras, Reem Kanjarawi, Bernard Flourie, Dominique Kaiserlian, Stephane Nancey Background : Regulatory Foxp3 + T cells (Treg) maintain immune tolerance by controlling systemic autoimmune and allergic diseases and are also crucial for intestinal homeostasis and oral tolerance to dietary antigens and to microbiota. Breakdown of intestinal homeostasis in chronic inflammatory bowel diseases is associated with inability of Treg to control inflam- mation, suggesting that the inflamed intestinal microenvironment is deleterious for Treg. Aims: To investigate the impact of intestinal inflammation on the number, survival, phenotype and suppressive function of Treg and on the conversion from naïve CD4 + T cells into Treg. Methods : Colitis was induced in B6 mice fed with 3% DSS in drinking water during 6 days. Severity of colitis was assessed by body weight loss, macroscopic and microscopic colon analysis. Number, phenotype, recruitment and conversion of Treg into mesenteric lymph nodes (MLN) and colon lamina propria (LP) were analyzed by flow cytometry. MLN and LP Treg ex vivo suppressive function was assessed by the ability of sorted CD4 + CD25 + Foxp3-egfp + Treg to inhibit the proliferation of allogenic CD4 + CD25 - Foxp3- egfp - Teff. Conversion from naïve CD4 + T cells into Treg was analysed by adoptive transfer experiments. Results : Intestinal inflammation enhanced the number of Foxp3 + Treg in MLN, although their frequency decreased due to the dense inflammatory infiltrate. Colitis was associated with increased proliferation of Foxp3 + Ki67 + Treg in MLN and LP. MLN Foxp3 + Treg at the steady state comprised both Helios + Neuropilin + (natural (nTreg)) and Helios - Neuropilin - (induced (iTreg)). Colonic inflammation decreased the frequency of nTreg in MLN. Ex vivo Treg suppression assays showed that MLN Treg from colitic mice maintain their suppressive function, when removed from the inflamed intestine. Adoptive transfer of naïve CD4 + T cells (containing nTreg) in congenic recipient mice showed that colon inflamma- tion exacerbated recruitment of CD4 + T cells into the LP. However, in vivo conversion of CD4 + CD25 + Foxp3 - pre-Treg (sorted from Foxp3-egfp transgenic mice) into Foxp3 + Treg was impaired in colitic recipient mice. Finally, transfer of Foxp3-egfp + cells in DSS mice showed a decrease of colitis severity. Conclusion : Taken together our data illustrate that Treg can be activated and recruited into the inflamed intestine and that intestinal inflammation impairs Treg conversion from CD4 + T cells. Mo1759 Reciprocal IL-22 Expression Masks IL-17 Activities in T-Cell Transfer Colitis Model in IL-17A-/- Mice Chang Soo Eun, Dong Soo Han, A. Reum Lee, Kyo-Sang Yoo Background: Th17/IL-23 pathway has been linked to the pathogenesis of several chronic inflammatory disorders, including inflammatory bowel disease. However, the function of IL-17A in inflammatory bowel disease remains unclear. We investigated the role of IL-17A on the intestinal inflammation using dextran sodium sulfate (DSS) and trinitrobenzene sulfonic acid (TNBS) induced colitis model and T-cell transfer induced mouse colitis model. Methods: Acute colitis was induced in IL-17A-/- and C57BL/6 wild-type (WT) mice by administering 2.7% DSS orally in drinking water for five days or by introducing 3.2 mg of TNBS in 50% ethanol into the rectum. In the T-cell transfer induced colitis model, purified CD4+ CD45RB high T cells (1x10 6 ) from IL-17A-/- or WT mice were injected intraperitoneally into Rag2-/- recipients. Clinical activities including weight loss and histologic findings of colonic segments were examined. Proinflammatory cytokine levels (IFN-γ, IL-12p40) were measured by ELISA in the supernatants of colonic tissue explants. Results: After DSS or TNBS treatment, clinical and histological activities of colitis, and pro-inflammatory cytokine levels in colonic tissue explants were attenuated in chemical-treated IL-17A-/- mice compared to chemical-treated WT mice. On the contrary, there were no significant differences in weight loss, histological scores, and proinflammatory cytokine levels (IFN-γ, IL-12p40) between recipients of IL-17A-/- CD45RB high T cells and recipients of WT CD45RB high T cells, although CD45RB high T cells from both donor mice induced significant colitis in the Rag2-/- recipients. However, IL-22 and SOCS-3 mRNA expression levels of colonic tissues were increased in the recipients of IL-17A-/- CD45RB high T cells compared to those of WT CD45RB high T cells. Conclusion: In chemically induced colitis models, IL-17A appear to be pathogenic, whereas in the T-cell transfer colitis model, IL-22 may have a compensatory pathogenic role in the colitis induction of the recipients of IL-17A-/- CD45RB high T cells.