The effect of bee venom (BVA) on the development of type II collagen (CII)-induced arthritis (CIA) in rats has been studied. Male rats were immunized with an emulsion of 200 microg of CII and complete Freund's adjuvant (CFA). The rats were then given intraperitoneally (i.p.) injection of a suspension of BVA or saline during the experiment. The effect of BVA on cellular responses to CII was examined. In the control rats, the onset of arthritis was observed at the 24th day after the CII-immunization, and the severity of CIA was developed gradually. As compared with rats treated with saline, BVA i.p. injected at doses of more than 20 microl/100g mouse once a day for 14 days inhibited the ability of inguinal lymph node cells to produce T cell cytokines interleukin-1beta, -2, -6, tumor necrosis factor-alpha and interferon-gamma when the cells were obtained from rats 24 days after immunization and cultured in vitro with CII. When rats were injected i.p. with sheep red blood cells, hemagglutination titers in BVA-treated and control rats did not differ significantly when low doses of BVA was given to rats. However, i.p. injection of BVA at doses of more than 10 microl/100g/day suppressed antibody production. Pretreatment of rats with BVA could inhibit the development of collagen arthritis even when 10-20 microl/100g/day of the BVA were used for pretreatment. Interestingly, higher doses than 10 microlBVA/100g mouse were much effective for arthritis incidence. Treatment of rats with BVA prevented the development of collagen arthritis in a dose-dependent manner. Doses of BVA (15 and 20 microl/100g) resulted in decreased incidence of arthritis. In conclusion, therapeutic i.p injection with BVA improved the clinical course of the disease and the immune response to CII.