Our data may support the assumption of TNF alpha playing an important role in joint destruction in RA. Possible alternative conclusions are discussed.

The objective of this study was to quantify insulin-like growth factor (IGF) binding proteins (IGFBPs) in the synovial fluid (SF) and plasma of patients with rheumatic diseases and to study the role of these proteins in the regulation of cartilage proteoglycan (PG) synthesis. Immunological determination of IGFBP-2, IGFBP-3, IGF-I, IGF-II, interleukin-1 beta (IL-1 beta) and tumour necrosis factor alpha (TNF alpha) was undertaken in the SF and plasma of 115 patients with rheumatoid arthritis (RA; n = 53), osteoarthritis (OA; n = 44) and other rheumatic disorders. We also determined the effects of SF on bovine cartilage PG synthesis in culture. IGFBP-2 and IGFBP-3 were elevated in the plasma (by 38% and 28%, respectively) and SF (by 56% and 59%, respectively) of patients with RA compared to age- and sex-matched OA controls (determined by RIA and confirmed by Western ligand blot). IGF-I and IGF-II did not differ significantly between the two groups. OA SF, and, to a lesser extent, RA SF stimulated cartilage PG synthesis in culture, and more than 60% of this activity was neutralised by a specific monoclonal anti-IGF-I antibody. Human IGFBP-3 dose-dependently inhibited the stimulation of cartilage PG synthesis effected by SF or human IGF-I. In RA patients, the SF concentration of IGFBP-3 was positively correlated with SF levels of IL-1 beta and TNF alpha, with the serum levels of C-reactive protein and with the erythrocyte sedimentation rate. We concluded that IGF-I is, under the conditions studied, the most important anabolic factor in human SF with respect to articular cartilage PG synthesis. The bioactivity of IGF-I in joints is modulated by IGFBP-3, which is elevated in RA SF compared to OA SF. Elevated IGFBP-3 in RA SF may reduce the availability of IGF-I to articular chondrocytes, thus interfering with cartilage PG synthesis in RA.

The determination of IL-1 beta and IL-1ra in the SF of patients with joint disorders as examined in this study seems to allow to a certain extent a prediction of the collagenase- and caseinase-activity contained in the diseased joint. We would favor.