The contribution of ATP-sensitive potassium channel (KATP channel) blockade in the vasoconstriction produced by vasopressin was studied. All experiments were performed using rat thoracic aorta cut in 4-mm rings, denuded from their endothelium and mounted into 20-ml organ baths. Glibenclamide (0.01-10 microM), a KATP channel antagonist, did not induce any measurable contraction, nor did it reduce the maximum contraction induced by vasopressin and phenylephrine. The specific inhibition of lemakalim-induced (a KATP channel activator) relaxation by vasopressin was investigated. Lemakalim (0.01-0.3 microM) relaxed both vasopressin (0.1 microM) and phenylephrine (0.3 microM) preconstricted vessels. However, in contrast to what would be expected from KATP blockade by vasopressin, rings preconstricted with vasopressin were more sensitive to the relaxant action of lemakalim, compared to phenylephrine preconstricted vessels (log[EC50] of -7.82 +/- 0.01 and -7.10 +/- 0.02, respectively, p < 0.05). Dose-response curves to papaverine (3-30 microM) in rings preconstricted with vasopressin and phenylephrine were comparable. When aortic rings were pretreated with lemakalim (0.1 microM), the maximum active tension induced by vasopressin was reduced (2.68 +/- 0.23 in control conditions vs. 0.62 +/- 0.08 g on pretreated vessels, p < 0.001), whereas that by phenylephrine was slightly increased. In order to explain the stronger relaxant action of lemakalim against vasopressin-induced constriction, the contribution of calcium influx through L-type calcium channels in the constriction of aortic rings to vasopressin and phenylephrine was compared.(ABSTRACT TRUNCATED AT 250 WORDS)