AB0053 ACTIVIN A AND FOLLISTATIN AFFECT THE INTERACTION OF ENDOTHELIAL CELLS AND RHEUMATOID ARTHRITIS SYNOVIAL FIBROBLASTS

Abstract

Background: Activin A and its antagonist follistatin are part of an autoregulatory cycle, which is well known in the hypothalamic-pituitary-gonadal axis. Activins also have an important function in autoimmune diseases, such as rheumatoid arthritis (RA). Due to inflammation, activin A is released systemically, causing an induction of its antagonist follistatin. The negative feedback mechanism is well described for hepatocytes, but seems to be inactive in synovial fibroblasts from patients with rheumatoid arthritis (RASF). Neoangiogenesis, which is mediated partially by local fibroblasts, is increased due to inflammation and tissue hyperplasia in RA synovium. Despite the fact that RASF contribute to cartilage destruction in RA and RASF are able to interact with endothelial cells, less is known about the effect of activin and follistatin in this context. Objectives: The aim of this study was to examine the effect of activin A and follistatin on the interaction of RASF and endothelial cells. Methods: Endothelial cells (HUVEC) were commercially obtained and RASF were isolated from synovial tissue of patients with RA undergoing joint replacement surgery, RASF and HUVEC were stimulated in mono-, or coculture with activin A (15ng/ml), follistatin (500ng/ml) and/or IL-1β (1ng/ml). The concentrations of activin A, follistatin, VEGF and IL-6 were measured by ELISA. Results: IL-1β induced the release of activin A 8-fold in RASF alone (p In HUVECs, the IL-6 release was reduced by 37.6% after stimulation with activin A and IL-1β (n=5,p The release of VEGF was induced in RASF with IL-1β (89%), activin A (55%), activin A combined with IL-1β (148%), follistatin and IL-1β (84%) compared to unstimulated control. In coculture with HUVECs, the induction was less distinct than in monoculture (IL-1β: 75%, activin A: 22%, activin A and IL-1β: 101%, follistatin and IL-1β: 67%, n=4). Conclusion: The autoregulatory cycle of activin A and follistatin is active in endothelial cells and inactive in RASF. Due to the interaction of endothelial cells and RASF, the proinflammatory response of the RASF is weakened. This was shown in direct coculture with no induction in coculture compared to stimulation with activin A and IL-1β in RASF monoculture. Interestingly, in direct coculture, the effects of HUVECs appear to dominate resulting in a significant reduction of the activin A concentration in the presence of follistatin and IL-1β in comparison to RASF monoculture.