Antibodies against retinal S-antigen in patients with juvenile chronic arthritis-associated uveitis.

Antibodies Against Retinal S-antigen in Patients with Juvenile Chronic Arthritis-Associated Uveitis SIR—Uveitis associated with juvenile chronic arthritis (JCA) is the only type of uveitis that is consistently associated with a high prevalence of autoantibodies. The specificity of the antinuclear antibodies that are present include nucleosomal proteins [1], low-molecular weight antigens [2] and histone autoantibodies [3]. 0stensen found that antibodies to histone H3 were particularly associated with the presence of uveitis; in an earlier study, we confirmed that the prevalence of autoantibodies to five histone proteins is raised in JCA and that H3 histone antibodies were the most common [4]. Uveitis is thought to have an autoimmune patho-genesis: strong evidence for this lies in the models of autoimmune uveitis that can be produced in experimental animals using a variety of retinal autoantigens. The most widely studied experimental uveitogen has been retinal S-antigen, a photoreceptor protein, and it has excited particular interest that there are sequence homologies between retinal S-antigen and histone H3. Histone peptides sharing sequences with the uveito-genic peptides of S-antigen can induce an identical experimental autoimmune uveoretinitis. It has been proposed that homologies between retinal antigens and widely conserved nucleoproteins may allow an immune response directed against microbial antigens to trigger autoimmune uveitis through the mechanism of molecular mimicry [5]. Antibodies to bovine retinal S-antigen have been reported in patients with JCA-associated uveitis [6-8] and anti-photoreceptor antibodies have been found by immunofluorescence [9]. This is unexpected, as the uveitis of JCA does not typically involve the retina where S-antigen is found. Recent ELISA studies in adult uveitis patients, using human S-antigen, have failed to demonstrate raised levels of antibodies in patients compared to controls [10, 11], even when the retina is inflamed. In a previous study, we were unable to detect any association between antibodies to histone H3 and the presence of uveitis, but antibodies to N-and C-terminal peptides were associated with the presence of uveitis. We therefore sought to confirm whether raised levels of human S-antigen antibodies occur in children with JCA-associated uveitis, and whether they correlate with the presence of antibodies to histone H3 or peptides derived from it. Standard ELISA assays were performed using retinal S-antigen prepared from human and bovine eyes. Antigen-free wells were used for controls, and normal sera from children and adults, as well as adults with uveitis were used as control populations. Raised levels were defined as sera with levels >2 S.D. …