Abstract

OBJECTIVE--To investigate the role of tumour necrosis factor alpha (TNF alpha) in the development of antigen induced arthritis (AIA) in rabbits. METHODS--Monoclonal antibodies to rabbit TNF alpha were developed in rats and were used to detect TNF alpha in synovial fluid by enzyme linked immunosorbent assay and to localise it in tissue sections of synovium and cartilage from rabbits up to 21 days after induction of AIA. An antibody which neutralised TNF alpha activity in vitro was injected into rabbits to block TNF alpha action in vivo in AIA. Joint swelling, leucocyte infiltration into synovium and proteoglycan loss from cartilage were measured and compared with a control group, which were injected with sterile saline. RESULTS--Monoclonal antibodies to purified rabbit TNF alpha were prepared in rats and two were selected which were able to neutralise rabbit TNF alpha in a cytotoxicity bioassay. TNF alpha was detected in significant concentrations (21.7 (SE 0.5) pg/ml) in the arthritic joint fluid of rabbits with AIA only at one day after induction and it was then also sparsely localised in cells of the synovium, but from day 3 onwards it was localised more strongly in the deep zone of articular cartilage. Injection of anti-TNF monoclonal antibody R6 over three days into rabbits with AIA reduced joint swelling and leucocyte infiltration into joint fluid and decreased the expression of CD11b and CD18 on cells in the joint fluid. However, there was no significant reduction in the loss of proteoglycan from articular cartilage, although the joint fluid at three days contained a lower glycosaminoglycan content. The antibody R6 gave most effect at a dose of 0.6 mg/kg and there was no increase in its effectiveness at a fivefold greater dose (3.0 mg/kg). Treatment over 10 days gave a more complete suppression of joint swelling, but did not result in any less proteoglycan loss from cartilage. Treatment for five days with a 16 day follow up gave a significant reduction in swelling for several days beyond the treatment, but the swelling then slowly returned, until by day 21 there was no significant difference in joint swelling and there was also no recovery of cartilage proteoglycan content. A rabbit anti-rat immunoglobulin response was detected at 21 days, which may have limited the long term effectiveness of the antibody. CONCLUSIONS--In AIA in rabbits, TNF alpha was only detected in synovial fluid at one day after induction and there was only limited cellular localisation of TNF alpha in synovium and cartilage from three days. However, neutralising TNF alpha with a monoclonal antibody was effective in suppressing inflammatory changes in the joint during the acute onset of AIA, but it had little effect on the loss of proteoglycan from cartilage. The results suggest that blocking inflammation and synovitis with anti-TNF alpha may be more easily achieved than preventing damage to articular cartilage.