Resveratrol has been shown to induce vasorelaxation. In this study, we investigated the mechanism(s) of resveratrol-induced vasorelaxation in resistance mesenteric arteries from male lean and dietary-induced obese rats. Compared with lean rats, arteries from dietary-obese rats showed significant (P<0.001) endothelial dysfunction, as indicated by a decrease (>20%) in maximal acetylcholine-induced vasorelaxation. Resveratrol (5-35 micromol/l) induced concentration-dependent relaxation of mesenteric arteries preconstricted with noradrenaline (8 micromol/l) or KCl (125 mmol/l) from both lean and dietary-obese rats. There were no significant differences between the two groups, achieving a maximum relaxation of >95% at a concentration of 35 micromol/l. In noradrenaline-preconstricted arteries from lean rats, N(G)-nitro-L-arginine methyl ester (L-NAME; 100 and 300 micromol/l) caused a significant (P<0.01) concentration-dependent rightward shift in reseveratrol activity, with no effect on maximal responses. However, L-NAME (100 and 300 micromol/l) did not alter the effects of reseveratrol on arteries from dietary-obese rats, giving superimposed concentration-responses curves. Indomethacin was also ineffective in altering resveratrol activity in arteries from both lean and dietary-obese rats. In noradrenaline-precontracted arteries from dietary-obese rats, responses to resveratrol were not attenuated by endothelial denudation, indicating an action independent of the endothelium. This study indicates that: (a) the maximal effects of resveratrol on resistance arteries from lean and dietary-obese rats are not effected by endothelial dysfunction, and (b) the effects of resveratrol in lean animals (where endothelial function is not impaired), but not in dietary-obese rats, are mediated via NO.