s / Osteoarthritis and Cartilage 22 (2014) S57–S489 S452 MMP1 and ADAMTS4 in FLSs at the gene expression level. Importantly, we also demonstrate that insulin blunts release of MMP1 and MMP13 protein into the culture media, counteracting the effect of TNFa and IL-1b. Thus, insulin appears to play a selective protective role in the diarthrodial joint by suppressing release of catabolic enzymes into the synovial fluid via its effects on FLSs. Further study is needed to determine if insulin resistance in obesity/diabetes impairs this critical role for insulin in protecting cartilage matrix in synovial joints. Although the mechanism of selective suppression of MMP1 and MMP13 by insulin is still unknown, insulin and insulinsensitizing agents could be new disease-modifying interventions in OA treatment. 822 EXPRESSION OF INTERLEUKIN-17 AND INTERLEUKIN-22 IN NON INFLAMMATORY AND INFLAMMATORY SYNOVIAL MEMBRANES FROM OSTEOARTHRITIS PATIENTS C. Deligne y,z, S. Casulli x,z, A. Pigenet y,z, C. Bougault y,z, L. Campillo-Gimenez x,z, F. Berenbaum k,y, C. Elbim x,z, X. Houard y,z. y INSERM UMRS938, Paris, France; zUPMC Univ Paris 06, Paris, France; x INSERM UMRS945, Paris, France; kDept. of Rheumatology, AP-HP SaintAntoine Hosp., Labex Transimmunom, Inflammation-ImmunopathologyBiotherapy Dept. (i2B), Paris, France Purpose: IL-17 and IL-22 are inflammatory cytokines classically involved in chronic inflammation in various diseases. Here, we investigated the expression and the release of IL-17 and IL-22 by OA synovial membranes, in relation to the inflammatory status of synovium. Methods: Synovial membranes from OA knee patients (n 1⁄4 32) were collected at surgery and inflammatory (Infl) and non-inflammatory (NI) areas were separated according to the macroscopic evaluation of inflammation that appeared as hypervascularized areas with hypertrophic and hyperaemic villi. A piece of each tissuewas frozen for mRNA extraction, fixed in paraformaldehyde for histology and incubated in serum free culture medium in order to obtain conditioned media containing soluble factors released by synovial tissues. IL-1b, IL-6, IL-8, IL18, IL-17, IL-22, IL-23, TNF-a, TGF-b1, myeloperoxidase (MPO) andMMP9were analyzed for mRNA expression by quantitative RT-PCR and/or for protein level by ELISA and gelatin zymography. Immunohistochemistry for endothelial cells (CD31) and leukocytes (CD45) among them macrophages (CD68), neutrophils (CD15) and Tand B-lymphocytes (CD3 and CD20, respectively) was performed. Results: Inflammatory areas of OA synovial membranes were characterized by increased CD45þ inflammatory cell infiltration and vessel area/tissue area as compared to non-inflammatory areas (p1⁄4 0.001 and p 1⁄4 0.009, respectively). Macrophages were present in the intimal layer of both NI and Infl OA synovial membranes with higher accumulation in Infl as compared to NI. Only the subintimal layer of Infl areas of OA synovial membranes contained macrophages, Tand B-lymphocytes and some neutrophils. Consistently, the inflammatory markers MMP-9 and MPO were released in significantly higher concentrations by Infl than by NI areas (p 1⁄4 0.026 and p 1⁄4 0.001, respectively). IL-17 and IL-22 were both expressed and released by OA synovial membranes. A stronger mRNA expression was found in Infl for both IL-17 and IL-22. Infl also released significantly higher levels of IL-22 than NI (p 1⁄4 0.046). Strong positive correlations were found between IL-17 and IL-22 at mRNA and protein levels. The expression of IL-17 and IL-22 is controlled by a subset of cytokines, including IL-1b, IL-6, IL-8, IL-18, IL-23, TGFb-1 and TNF-a. With the exception of TNF-a, all were released in significantly higher concentration by Infl as compared to NI areas. IL-12 was not detected in conditioned media of synovial tissues. Conclusions: Our results show an increased infiltration of lymphocytes as well as an increased release of inflammatory cytokines, including ILSynovial gene expression at day 1 after transection (as fold-change compared to healt