S1700 Smurf2 Activity in T Lymphocytes Protects Mice from Experimental Colon Cancer Heike Dornhoff, Christoph Becker, Peter R. Galle, Markus F. Neurath Introduction: TGF-beta is a pleiotropic cytokine with wide ranging effects in proliferation, differentiation, immune suppression and apoptosis. Recent work from our group has shown that TGF-beta signalling in T cells is protective in a mouse model of colitis associated cancer. Smad ubiquitin regulating factors (Smurf) are ubiquitin ligases that are involved in the regulation of TGF-beta signalling. The aim of this study was to determine the function of Smurf2 expression in T-cells on the pathogenesis of experimental colitis associated colon cancer. Methodology: We could isolate a known splice variant of Smurf2 lacking an Exon in the C2-domain. To analyse whether this form has a regulatory role in colon associated cancer we generated a transgenic mouse strain that overexpresses Smurf2 in T-cells. Smurf2 expression were analysed by qPCR. Wild type (WT) and Transgenic (TG) mice were treated once with the mutagenic agent Azoxymethan (AOM) followed by three cycles of Dextran Sodiumsulfate (DSS). After each cycle, the inflammation of the gut and the tumor growth and size of every mouse were monitored by colonoscopy. Results: Smurf2 expression was upregulated by TGF-beta stimulation in T-cells and Smurf2 was markedly upregulated in tumor infiltrating CD4+ lymphocytes in AOM/DSS treated mice. Whereas WT mice suffered from severe colitis resulting in colon tumors beginning at day 35, Smurf2 transgenic mice had less colitis and were significantly protected from tumor development. Interestingly, T- lymphocytes overexpressing Smurf2 showed an upregulation of Smurf2 target genes including the TGFbRII and an activation of Smad3 as compared to wild-type T-lymphocytes, which were previously described as typical Smurf2 targets for degradation. In addition the transfection of Smurf2 and a CAGA-Luc Plasmid into Cos-cells for Smad3-Promotor studies yielded the same effect as shown by an upregulation of the Smad3 activity. Conclusion: Although, WT- Smurf has been described as a negative regulator of the TGF-beta signalling pathway, our data show surprisingly that a Smurf2 splice variant upregulates the TGF-beta receptor expression and increases TGF-beta signalling effects. Due to immunosuppressive effects on T-cells Smurf2 has beneficial effects on mucosal inflammation and colon tumor development. These findings are probably due to autoubiquitination of WTSmurf2 via the splice variant of Smurf2. Smurf2 thus emerges as an attractive target for modulation of chronic intestinal inflammation and colitis associated carcinogenesis. S1701 A Crucial Role of Rorgamma Expressing TH17 Cells in Chronic Intestinal Inflammation Moritz Leppkes, Christoph Becker, Stefan J. Wirtz, Ivaylo I. Ivanov, Clemens Neufert, Andrew J. Murphy, David Valenzuela, George Yancopoulos, Erin Nesbitt, Peter R. Galle, Dan R. Littman, Markus F. Neurath [Introduction]: IL-17 producing CD4+ T helper cells (Th17) have been described as a major contributor to chronic and autoimmune inflammation in mouse models such as experimental autoimmune encephalitis (EAE) or collagen-induced arthritis (CIA). Th17 cells produce the effector cytokines IL-17A, IL-17F and IL-22. However, to date, it is not fully understood, whether Th17 cells play a regulatory role in experimental colitis. [Methods]: T cells from IL-17A, IL-17F, IL-22 and RORgamma deficient mice were used in adoptive transfer models of colitis as well as neutralizing IL-17A antibodies. Readout parameters included histology, high resolution mini-endoscopy, cytokine analysis and studies on barrier function. [Results]: Transfer of IL-17A - deficient T lymphocytes resulted in severe colitis activity in reconstituted animals that was indistinguishable from wild-type controls. In contrast, adoptive transfer of RORgamma-/- T-cells failed to induce colitis, indicating a crucial role of Th17 cells for disease development. Transfer of IL-17F-/- and IL-22-/- deficient T cells showed similar colitis severity as compared to wild-type cells. However, when mice reconstituted with IL- 17F-/- T cells were additionally treated with a neutralizing anti-IL17A antibody, mice were significantly protected from disease development. This finding was associated with marked protective effects on the intestinal barrier, as shown by FITC dextran studies, and correlated with differential expression of tight junction proteins, such as pore-forming Claudin-2. [Conclusion]: In summary, we demonstrate a crucial role of RORgamma expressing Th17 cells in chronic intestinal inflammation. RORgamma controls IL-17A and IL-17F production and these cytokines play a redundant but highly pathogenic role in gut inflammation and barrier function. S1702 Essential Role of HSP70 to Prevent AOM/DSS Induced Inflammatory Colon Cancer Yun Tao, Lev Lichtenstein, John Hart, Marc Bissonnette, Eugene B. Chang Background: Hsp70 is a highly conserved inducible protein that inhibits stress-induced denaturation of target proteins and is an essential component of several signaling complexes. Since in prior studies we demonstrated that Hsp70 exerts anti-inflammatory and immunomo- dulatory effects, we asked if Hsp70 loss would alter colitis induced by dextran sulfate sodium (DSS), or neoplastic transformation in an azoxymethane (AOM, mutagen)/DSS model of colitis-associated colon cancer. Methods: Hsp70 double knock out mice (Hsp70.1-/-/70.3-/- ) were backcrossed 6 generations to C57Bl6/J. For DSS studies, mice received 2.5% DSS (5 days DSS, 3 wks off) x 4 cycles and were sacrificed 24 wks later. For AOM/DSS studies, mice (wt ~ 20g) received AOM (7.5 mg/kg body wt) and 3 cycles of 2.5% DSS beginning 1 wk after AOM. Mice were sacrificed 24 wks later. Tissues were graded for histology and examined by immunostaining. Results: Four cycles of DSS alone caused persistent severe colitis in 10/10 Hsp70 deleted mice, but resolved in 11/11 WT C57Bl6/J mice within 4 wks of DSS (p <0.001). In AOM/DSS treated mice, 25% (4/10) WT mice developed polypoid colonic carcinomas that resembled sporadic human colon cancers. In contrast, 87.5% (7/ 8) Hsp70 KO mice developed multifocal areas of dysplasia and flat invasive carcinomas that mimicked ulcerative colitis (UC)-associated carcinogenesis. These tumor incidences were significantly different (p <0.05). To begin to dissect molecular features of this model we A-253 AGA Abstracts stained tumors for p53, APC and Cox-2. Expression of Cox-2 was higher in Hsp70 KO dysplastic mucosa and tumors compared to WT tumors. Nuclear p53 was also higher in Hsp70 KO compared to WT tumors, suggesting p53 might be mutated in these tumors, an early event in UC colon cancer. Staining levels with carboxyterminal directed APC antibodies were lower in WT than Hsp70 KO tumors, suggesting Hsp70 loss also influences APC expression in colitis-induced colon cancer. Conclusions: These studies demonstrate Hsp70 plays a critical role in maintaining intestinal homeostasis. With loss of this essential protective chaperone, AOM/DSS induces persistent colitis and inflammatory colon cancer with high penetrance that mimics many features of human UC colon cancer. Furthermore, AOM/DSS induced alterations in p53 and APC expression in Hsp70 deleted mice resemble changes in UC-colonic carcinogenesis. This model will be useful to dissect the role of Hsp70 in IBD colitis and colon cancer. S1703 Distinct Role of Toll-Like Receptor 2 (TLR 2) in Development of Murine Colitis-Associated Colorectal Neoplasia Jae Geun Hyun, Yasmin Hernandez, David Hsu, Junske Maki, Keith J. Breglio, Daisy E. Conduah, Anli Chen, Ruliang Xu, Maria T. Abreu, Masayuki Fukata Background: The up-regulation of TLR4 and TLR2 in the mucosa has been reported in inflammatory bowel disease. We have described the importance of TLR4 signaling in the development of colorectal neoplasia in a mouse model of colitis-associated cancer. In this study, we examined the role of TLR2 signaling in the development of neoplasia in a mouse model of colitis-associated cancer. Methods: Colitis associated neoplasia was induced by AOM injection (7.3 mg/kg) followed by 2 cycles of 2.5% Dextran Sulfate Sodium (DSS) treatment for 7 days in TLR2-/- (n= 10) and TLR4-/-mice (n= 6) and wild type (WT) mice (n= 10). The number and size of dysplasia were measured under the microscope. Epithelial cell proliferation was examined by BrdU labeling. Expression of COX-2, iNOS, and phos- phorylated epidermal growth factor receptor (pEGFR) in the colonic mucosa were analyzed by Western blot analysis and immunofluorescence. Results: The number of dysplastic lesions in TLR2-/- was lower than WT mice but higher than TLR4-/- mice (1.5±1.5 vs. 6.0±2.1 vs. 0.4±0.2 , p< 0.001). In addition, the size of dysplastic lesions was smaller in TLR2-/- mice compared with WT mice but still greater than that seen in TLR4-/- mice (2.1±1.0 mm vs. 3.7±3.5 mm vs. 1.3±0.3, p= 0.02). Epithelial cell proliferation (number of BrdU positive cells per crypt) was lower in TLR2-/- mice than WT mice (13.3±5.2 vs. 20.4±6.2, p<0.001), but similar to TLR4-/- mice (14.3±3.8). Western blot showed decreased mucosal expression of COX-2, iNOS and pEGFR in TLR2-/- mice compared to WT mice; COX-2 expression in TLR2-/- mice was greater than TLR4-/- mice. The decrease in pEGFR was similar between TLR2-/- mice and TLR4-/- mice. We have shown decreased expression of COX-2 in intestinal epithelial cells and lamina propria macrophages in TLR4-/- mice. Consistent with the Western blot results, immunofluorescence showed a decrease in COX-2 signal in colonic epithelial cells of TLR2-/- mice compared to WT mice. But TLR2-/- mice have more COX-2 expressing lamina propria cells when compared to TLR4-/- mice, suggesting distinct roles of TLRs within different cell types in mucosal COX-2 expression. Conclusions: Mice lacking TLR2 showed an intermediate phenotype between TLR4-/- mice and WT mice during the develop- ment of colitis-associated neoplasia, suggesting a distinct role of TLR2 signaling in inflamma- tion induced tumorigenesis. Blocking specific TLR signals in the mucosa may be useful in chemoprevention of colorectal neoplasm in the setting of chronic inflammation. S1704 Fundamental Difference in Hemopoietic Growth Factors (Flt3l, GM-CSF) On Experimental Mouse Colitis and the Innate Immune Response Konrad Aden, Satheesh K. Sainathan, Kumar S. Bishnupuri, Qizhi Luo, Stefan Schreiber, Brian K. Dieckgraefe Background: Disordered innate immune clearance may play a major role in the etiology of IBD. Plasmacytoid dendritic cells (pDC) are the major source of type I interferons, which may regulate the mucosal immune response. We previously showed that treatment with GM-CSF dramatically reduces the severity of DSS-colitis in a mechanism involving dendritic cell populations, including the pDC. Flt3 Ligand (Flt3L) is a growth factor that similarly expands dendritic cells (DC) populations. Aim: To examine the differential effects of Flt3L and GM-CSF on DSS-induced colitis and the innate immune response. Method: Colitis in mice was induced with 5% DSS in their drinking water for 7 days. Mice were administered with Flt3L (20μg/mice) daily and compared to control groups. Disease activity index (DAI), histological analysis and immunohistochemistry was performed. Real Time RT-PCR was used for quantitative analysis of inflammatory genes. Flow cytometry was performed using anti-CD11c(DC), -440c(pDC). Results: Flt3L aggravates DSS-induced colitis in mice, by all disease end point measured. Histopathological analysis showed the trend towards worse inflammation in DSS+Flt3L treated mice compared to the DSS control group. Real Time RT-PCR exhibited a significant induction of the proinflammatory genes like TNFalpha and IL-1beta, which was worse in Flt3L+DSS treated animals. In order to examine the effects of Flt3L and GM-CSF on interferon production following CpG treatment, mice were pre- treated with GM-CSF, Flt3L or left untreated for 5 days. 4h prior to scarification each group was stimulated with CpG, a TLR9 agonist. Flow cytometry showed similar effects of both, GM-CSF and Flt3L, on dendritic cell populations. Quantitative Real Time RT-PCR analysis of colon and spleen tissue exhibited a drastic increase of type I-IFN gene induction in GM- CSF , but not in Flt3L pre-treated animals following CpG stimulation. A number of other genes were identified that were differentially expressed following GM-CSF or Flt3L treatment, including Cathelicidin, an antimicrobial peptide which is known to play a major part in innate immune defense. Conclusion: Flt3L and GM-CSF are hemopoietic growth factors, both capable of increasing pDC population. While GM-CSF ameliorates, Flt3L dramatically aggravates DSS-induced colitis in mice. Our data provide for the first time evidence on the fundamental differences between Flt3L and GM-CSF on TLR-driven innate immune response in experimental mouse colitis. AGA Abstracts