IL-27-producing CD14+ cells infiltrate inflamed joints of rheumatoid arthritis and regulate inflammation and chemotactic migration

Interleukin (IL)-27, a heterodimeric cytokine, has been reported to be involved in the pathogenesis of autoimmune diseases through mediating differentiation of Th1 or Th17 cells and immune cell activity or survival. However, the origin and effects of IL-27 in joints of rheumatoid arthritis (RA) remain unclear. In this study, we investigated the distribution and anti-inflammatory roles of IL-27 in RA synovium. The IL-27 levels in plasma of RA patients, osteoarthritis (OA) patients, or healthy volunteers (n = 15 per group) were equivalent and were at most 1 ng/ml, but the IL-27 level in synovial fluid of RA patients (n = 15, mean 0.13 ng/ml; range 0.017-0.37 ng/ml) was significantly higher than that in synovial fluid of OA patients (n = 15, mean 0.003 ng/ml; range 0-0.033 ng/ml) and potentially lower than in plasma. We analyzed the protein level of IL-27 produced by RA fibroblast-like synoviocytes (FLSs) or mononuclear cells (MNCs) from RA or OA synovial fluid or peripheral blood and showed that IL-27 in RA joints was derived from MNCs but not from FLSs. We also found by flow cytometry that IL-27-producing MNCs were CD14(+), and that these CD14(+)IL-27(+) cells were clearly detected in RA synovium but rarely in OA synovium by immunohistochemistry. Furthermore, we demonstrated that a relatively physiological concentration of IL-27 below 10 ng/ml suppressed the production of IL-6 and CCL20 from RA FLSs induced by proinflammatory cytokines through the IL-27/IL-27R axis. In the synovial fluid of RA, the IL-27 level interestingly had positive correlation with the IFN-gamma level (r = 0.56, p = 0.03), but weak negative correlation with the IL-17A level (r = -0.30, p = 0.27), implying that IL-27 in inflammatory joints of RA induces Th1 differentiation and suppresses the development or the migration of Th17 cells. These findings indicate that circulating IL-27-producing CD14(+) cells significantly infiltrate into inflamed regions such as RA synovium and have anti-inflammatory effects in several ways: both directly through the reduction of IL-6 production, and possibly through the induction of Th1 development and the suppression of Th17 development; and indirectly by regulation of recruitment of CCR6(+) cells, such as Th17 cells, through the suppression of CCL20 production. Our results suggest that such a serial negative feedback system could be applied to RA therapy. (C) 2011 Elsevier Ltd. All rights reserved.