The vascular effects of endothelin-1 (ET) in humans were investigated by brachial artery infusions of ET into 25 healthy volunteers. Forearm blood flow increased from a mean +/- SD value of 2.3 +/- 1.5 to 2.5 +/- 1.5 ml/min/100 ml forearm tissue (n = 25, p less than 0.05) in response to low dose (0.5 ng/min/100 ml forearm tissue) ET infusion and decreased to 1.78 +/- 1.3 and 1.1 +/- 0.9 ml/min/100 ml forearm tissue (p less than 0.001) during higher dosages (25 and 50 ng/min/100 ml forearm tissue). Sodium nitroprusside (0.6 micrograms/min/100 ml forearm tissue, n = 6), acetylcholine (16 micrograms/min/100 ml forearm tissue, n = 7), nifedipine (6 micrograms/min/100 ml forearm tissue, n = 6), and verapamil (80 micrograms/min/100 ml forearm tissue, n = 6) were infused alone and in combination with ET to evaluate the interactions between ET-induced vasoconstriction and stimulation of vascular muscle cyclic GMP levels by sodium nitroprusside, release of endothelium-derived relaxing factor by acetylcholine, and blockade of voltage-operated calcium channels by nifedipine and verapamil. Neither the vasodilator nor the vasoconstrictor response to ET was influenced by sodium nitroprusside or acetylcholine. In contrast, both calcium antagonists converted ET-induced vasoconstriction (e.g., delta forearm vascular resistance to ET 50 ng/min/100 ml forearm tissue, 151 +/- 100% and 164 +/- 92% in verapamil and nifedipine groups, respectively) to vasodilation (-35 +/- 12% and -21 +/- 16%, p less than 0.05). Our results demonstrate both ET-induced vasodilation (at low dosages) and vasoconstriction (at high dosages) in resistance vessels of normal humans. Blockade of voltage-operated calcium channels prevented ET-induced vasoconstriction and unmasked the vasodilator effect of high ET dosages. In human resistance vessels, blockade of voltage-operated Ca2+ channels but not cyclic GMP-dependent vasodilation may be an effective tool to inhibit ET-induced vasoconstriction.