Emergence of proinflammatory autoreactive T-cell responses in preclinical rheumatoid arthritis

Abstract

Background5 million people suffer from rheumatoid arthritis (RA) and every year 25 000 people will develop this disease in the UK. In an emerging model, the pathogenesis of RA can be understood as a development of sequential phases that precede clinically apparent disease. Autoimmunity with anti-immunoglobulin G rheumatoid factors (RF) and anti-citrullinated protein antibodies (ACPA) can be detected in the blood up to 13 years before onset of clinical synovitis (joint swelling). Regulatory T cells (Tregs) are an important component of the immune system that dampen and limit damage caused by excessive immune activity, and their dysfunction is found in RA. We aimed to identify autoreactive T cells to a known RA-associated antigen and to determine whether the quality of the response (regulatory vs inflammatory) is associated with health, preclinical RA, and RA. Methods41 patients were recruited from studies in the Division of Rheumatic & Musculoskeletal Disease, University of Leeds. 21 patients had preclinical RA (ACPA-positive with arthralgia but no evidence of clinical synovitis) and 20 had RA (ACPA-positive RA). Healthy volunteers (19 employees of the University of Leeds and Leeds Teaching Hospitals Trust) were controls. We measured T-cell responses to the RA-associated autoantigen, citrullinated vimentin (CitVim): peripheral blood mononuclear cells were incubated with peptide pools covering the entire sequence of CitVim for 72 h. Interleukin (IL) 10 (regulatory) and interferon (IFN) γ (inflammatory) responses were measured by ELISPOT. The study has ethics approval from the Leeds (West) Research Ethics (references 06/Q1205/169 and 04/Q1206/107). FindingsAutoreactive T-cell reactivity was detected in healthy controls, preclinical RA, and RA. However, the quality of this response differed between the groups: health was associated with IL-10 responses, whereas in preclinical RA there was an emergence of IFN-γ responses, which were also present in a proportion of RA patients (healthy volunteers, preclinical RA, and RA with IFN-γ responses were 11% [2/19], 32% [7/21], and 45% [9/20] and with IL-10 responses were 89% [17/19], 68% [14/21], and 55% [11/20], respectively; χ2 7·516, df=2, p=0·0233). Looking at the balance of regulation and inflammatory responses, we found a sequential reduction in the ratio of IL10 to IFNγ from health to preclinical RA to RA. There was a significant effect of disease status on the ratios with significant difference between health and preclinical RA and health and RA. The IL10:IFNγ ratio for healthy controls was significantly different from the preclinical RA and the RA groups (mean 1·255 [SD 1·383] vs 0·3892 [1·548] vs 0·1943 [1·395], ANOVA and Tukey test F=5·275, p=0·0062. We also found that the magnitude of the IFN-γ CitVim response increased after neutralising IL10 in four patients with preclinical RA. InterpretationWe report that a distinction can be made between health, preclinical RA, and RA in the context of the quality of autoimmune T-cell responses. We have found a sequential change in the balance of IL-10 and IFN-γ responses from health (which favours a regulatory phenotype) to preclinical RA and then to established RA (which favours a relatively inflammatory phenotype). We have also shown that autoreactive IL-10 Treg responses are actively inhibiting IFN-γ responses in patients with preclinical RA. It is plausible that such a regulatory T-cell response plays an important part in preventing disease in health and also progression to disease in preclinical RA. Perhaps approaches to strengthen this regulatory response in people with preclinical disease could prevent progression to disease. Such immunomodulatory approaches are used in allergy practice and in studies in the prevention of diabetes. FundingNone.