Anti-arthritic effect of E3 ubiquitin ligase, c-MIR, expression in the joints

International Immunology, Vol. 23, No. 3, pp. 177–183 doi:10.1093/intimm/dxq470 ª The Japanese Society for Immunology. 2011. All rights reserved. For permissions, please e-mail: Anti-arthritic effect of E3 ubiquitin ligase, c-MIR, expression in the joints Masayasu Toyomoto1,2,3, Satoshi Ishido4, Nobuyuki Miyasaka1,2, Hachiro Sugimoto3 and Hitoshi Kohsaka1,2,4 1 Department of Medicine and Rheumatology, Graduate School of Medical and Dental Sciences, Tokyo Medical and Dental University, Tokyo 113-8519, Japan 2 Global Center of Excellence (GCOE) Program, International Research Center for Molecular Science in Tooth and Bone Diseases, Tokyo Medical and Dental University, Tokyo 113-8519, Japan 3 World-Leading Drug Discovery Research Center, Graduate School of Pharmaceutical Sciences, Kyoto University, Kyoto 606-8501, Japan 4 RIKEN Research Center for Allergy and Immunology, Kanagawa 230-0045, Japan Correspondence to: H. Kohsaka; E-mail: Received 14 February 2010, accepted 14 December 2010 Abstract Cellular modulator of immune recognition (c-MIR) is an E3 ubiquitin ligase that ubiquitinates MHC class II and CD86 for their endocytosis and subsequent lysosomal degradation. In accordance with their importance in antigen presentation, systemic c-MIR over-expression downmodulates adaptive immune responses. Rheumatoid arthritis (RA) is a chronic synovitis driven by autoimmunity in the joints. Since antigen-presenting cells, such as macrophages, dendritic cells (DCs) and rheumatoid factor-positive B cells are abundant in the rheumatoid synovial tissues, autoantigens released by tissue damage should be presented locally, leading to amplification of systemic arthritogenic immune responses. Assuming that inhibition of the antigen presentation in the synovial tissues should suppress systemic arthritis, we transferred the c-MIR gene to the hind leg synovial tissues from mice with type II collagen (CII)-induced arthritis, an animal model of RA. The gene was transferred adenovirally because adenoviruses can infect DC and macrophages in vivo. Unexpectedly, therapeutic effect was observed only in the treated joints. Splenocyte responses and serum antibodies against CII were not suppressed. Moreover, in vitro studies disclosed that c-MIR gene transfer suppressed IL-6 production from synovial fibroblasts stimulated with tumor necrosis factor (TNF)-a or IL-1b. Bone marrow-derived macrophages and DC from c-MIR transgenic mice were impaired in IL-6 and TNF-a production when stimulated with LPS. This suppression was controlled at the post-transcriptional level since their mRNA was not affected. These results have disclosed a new function of c-MIR, inhibition of inflammatory cytokine production. Induction of c-MIR in the joints could be a new therapeutic approach to the treatment of RA. Keywords: arthritis, cytokine, inflammation Introduction Rheumatoid arthritis (RA) is a chronic autoimmune inflammatory disease that affects primarily the joints. The invasive granulomatous synovial tissues in the rheumatoid joints are called pannus and contain lymphocytes, dendritic cells (DCs), macrophages and synovial fibroblasts in an activated state. Since activated DCs, macrophages and rheumatoid factor-positive B cells can act as antigen-presenting cells (APC) (1–3), articular autoantigens could be presented to lymphocytes to induce their activation instead of anergy in the rheumatoid synovial tissues. Resulting joint tissue damage in turn promotes further release of the articular autoantigens, which should fuel the autoimmune circle of the arthritis. The T-cell infiltration in the rheumatoid synovial tissues and presence of autoreactive T cells against type II collagen (CII) (4) show that cellular immune responses against articular antigens indeed play a prominent role in RA. Antigen presentation in the normal immune system is controlled by multiple regulatory components to avoid development of autoreactivity. Cellular modulator of immune recognition (c-MIR) is an E3 ubiquitin ligase that ubiquitinates CD86 and MHC class II to induce their surface down-regulation by endocytosis and following lysosomal degradation (5, 6). The mRNA analyses of mouse splenocytes showed 178 Anti-arthritic effect of E3 ubiquitin ligase c-MIR that c-MIR is expressed primarily by macrophages and DC (6). The c-MIR expressing cells were present in multiple organs, including heart, brain, placenta, kidney, liver, lung, muscle, pancreas, spleen, thymus, peripheral blood lymphocyte, lymph node, tonsil, fetal liver and bone marrow (BM) (5, 7). Nevertheless, transgenic c-MIR over-expression impaired development of CD4 T-cell generation in the thymus and in the periphery (6). The transgenic mice were resistant to induction of experimental autoimmune encephalomyelitis, showing that over-expression of c-MIR gene has an impact on in vivo immune reactions. Structurally, c-MIR is among membrane-associated RINGCH (really interesting new gene-CH) (MARCH) gene family (7), which consists of 11 gene members, and also called MARCH-VIII. It was reported that c-MIR over-expression reduced expression of Fas, transferrin receptor, BM stromal cell antigen 2 (BST2; CD317) and syntaxin-4 in addition to CD86 and MHC class II (7, 8) although c-MIR-mediated ubiquitination was demonstrated only for CD86 and MHC class II. Over-expression of MARCH-I also down-regulated CD86, MHC class II, Fas and transferrin receptor (7, 9, 10). MARCH-II over-expression down-regulated CD86 and transferrin receptor while MARCH-IV and -IX over-expression down-regulated MHC class I and CD4 (7). Thus, members of the MARCH family appear to modulate antigen presentation in multiple aspects. However, it is still to be elucidated how these molecules cooperate to control the entire immune system. We assumed that over-expression of c-MIR in synovial cells, especially APC in the arthritic joints, would suppress aberrant presentation of autoantigens liberated by inflammation and suppress immune responses that amplify the arthritogenic immune circle. We addressed this question by transferring the c-MIR gene to the joints of mice with CII-induced arthritis (CIA), which is an animal model of RA. Adenoviral gene transfer was employed because recombinant adenovirus vectors can infect DC and macrophages as well as fibroblasts in vivo (11, 12). This local gene therapy failed to suppress systemic arthritic reactions. Instead, it exerted the local therapeutic effects, revealing a new anti-inflammatory function of c-MIR. Methods Reagents and cells Antibodies against CD86 (BD Biosciences, San Jose, CA, USA), MHC class II I-A b-chain (clone KL295, ATCC) and actin (Santa Cruz Biotechnology, Santa Cruz, CA, USA) were used for immunoblotting. Recombinant mouse IFN-c, TNF-a and human IL-1b were purchased from PeproTech (Rocky Hill, NJ, USA). Fibroblast-like synoviocytes (FLS) were isolated from the synovial tissues of mice with CIA and grown in DMEM supplemented with 10% fetal bovine serum (FB (...truncated)