Experiments were designed to study the potential mechanisms underlying the vasodilator effect of celiprolol. Rings of canine left circumflex coronary artery and rat mesenteric artery, with and without endothelium, were suspended in organ chambers for isometric tension recording. In both blood vessels, celiprolol (10(-9)-10(-4) M) failed to produce relaxation in rings with and without endothelium; these same tissues relaxed in an endothelium-dependent manner to acetylcholine (10(-6) M). All tissues relaxed completely in the presence of papaverine (10(-4) M). In the coronary artery, isoproterenol (10(-9)-10(-4) M) produced endothelium-independent relaxations which were inhibited in a competitive fashion by celiprolol (pA2 = 7.52 +/- 0.14; slope = 0.98, 95% confidence limits = 0.80-1.15). In other experiments, strips of canine saphenous veins were incubated with [3H]norepinephrine [( 3H]NE) and suspended for superfusion. Electrical stimulation (2 Hz, 4 V, 2 ms for 6 min) produced an increase in [3H]NE overflow. Isoproterenol (2 X 10(-6) M) augmented the evoked release of [3H]NE. Treatment of the strips with celiprolol (up to 5 X 10(-6) M) did not inhibit isoproterenol-induced facilitation of [3H]NE release. Thus, although celiprolol is a potent antagonist of postjunctional beta-adrenoceptors in the coronary artery, no evidence was obtained for a direct or indirect vasodilator effect of celiprolol on isolated blood vessels.