The aim of this study was to investigate the vasorelaxant effect induced by diosgenin in superior mesenteric rings. In rings pre-contracted with phenylephrine (10 microM), diosgenin caused concentration-dependent relaxations [EC(50) = (3.3 +/- 1.2) x 10(- 4)M, E(max) = 94.2 +/- 2.6 %]. Vascular relaxation induced by diosgenin was significantly inhibited after removal of the endothelium (E(max) = 46 +/- 8.8%, p < 0.001) or after pre-treatment of the rings with N-nitro-L-arginine methyl esther (l-NAME) 100 or 300 microM (E(max) = 35.3 +/- 4%; 28.1 +/- 3.3%, respectively, p < 0.001), atropine 1 microM (E(max) = 24.6 +/- 3.4%, p < 0.001), hydroxocobalamin 30 microM (E(max) = 54.0 +/- 9.6%, p < 0.001), 1H-[1,2,4]oxadiazolo-[4,3-a]quinoxalin-1-one (ODQ) 10 microM (E(max) = 46.0 +/- 8.0%, p < 0.001) or indomethacin 1 microM (E(max) = 22.6 +/- 11.8%, p < 0.001). Vasorelaxation evoked by diosgenin was significantly inhibited after pre-treatment of preparations with both selective and non-selective inhibitors of large conductance Ca(2+)-activated K(+) (BK(Ca)) channels, iberiotoxin 100 nM or tetraethylammonium (TEA) 1mM, respectively (E(max) = 62.5 +/- 9.1%; 65.7 +/- 1.1%, p < 0.001). Conversely, in endothelium-denuded vessels, none of BK(Ca) channel blockers modified the relaxant effect induced by diosgenin. In mesenteric endothelial cells loaded with FURA-2 diosgenin was able to increase intracellular calcium concentrations, which were significantly decreased by atropine 1 microM. In addition, in isolated mesenteric rings, diosgenin induced marked increase in nitric oxide (NO) levels, which was completely abolished after functional endothelium removal. The results obtained here demonstrated that diosgenin-induced relaxation appears to involve endothelial muscarinic receptor activation with increase in intracellular calcium concentrations and consequent release of endothelium-derived relaxing factors (EDRFs), mainly NO and cyclooxygenase derivatives, which activate BK(Ca) channels. Nevertheless, further studies are necessary to clearly elucidate residual endothelium-independent relaxation induced by diosgenin.