1 We have investigated the mechanism of bradykinin (BK)-induced plasma extravasation into the knee joint of the anaesthetized rat. Accumulation of [I-125]-human serum albumin within the synovial cavity was used as a marker of increased vascular permeability. 2 Perfusion with BK (1 mu M) produced significant plasma extravasation into the knee which was inhibited by co-perfusion of the selective bradykinin B-2 receptor antagonist D-Arg-[Hyp(3),Thi(5),D-Tic(7),Oic(8)]-bradykinin (Hoe 140, 200 nM). 3 The bradykinin B-1 receptor agonist, [des-Arg(9)]-BK (up to 100 mM), did not induce plasma extravasation into the knee joint over this time period. 4 Chemical sympathectomy by chronically administered 6-hydroxydopamine (6-OHDA) did not inhibit bradykinin-induced plasma extravasation. Acute intra-articular perfusion with 6-OHDA (to stimulate transmitter release from sympathetic nerve terminals) at concentrations up to 50 mM did not induce significant plasma extravasation. Intra-articular perfusion of 100 mM 6-OHDA. induced significant plasma extravasation but produced severe systemic toxicity. 5 The selective neurokinin(1) (NK1) receptor antagonist, RP67580 (230 nmol kg(-1)), or receptor antagonists for the mast cell products histamine and 5-hydroxytryptamine did not significantly inhibit BK-induced plasma extravasation. 6 Co-perfusion of the NO synthase inhibitor, NG-nitro-L-arginine methyl ester (L-NAME) (1 mM) did not significantly inhibit the response to BK. Xe-133 clearance from L-NAME (1 mM)-injected joints was significantly (P < 0.05) reduced compared to D-NAME injected joints, suggesting a reduction in blood flow as a result of decreased basal NO production. Systemic administration of L-NAME at doses sufficient to produce significant and sustained elevation of blood pressure (5 or 30 mg kg(-1), i.v. 15 min prior to BK perfusion) also failed to significantly inhibit the BK-induced response. 7 We conclude that, in normal joints, BK induces plasma extravasation by acting on bradykinin B-2 receptors and that this response is not dependent on secondary release of mediators from sympathetic nerve terminals, sensory nerves, mast cells or on generation of NO.