We evaluated the antioxidant property of melatonin as it relates to the vasospastic effect of hydrogen peroxide (H2O2) on the human umbilical artery. Helical sections of umbilical arteries were obtained from healthy pregnant women who were delivered between weeks 37 and 39 of gestation. Changes in maximal potassium chloride (KCl)-induced tension were measured in arterial segments with intact endothelium. Segments were treated with H2O2 alone, or were pretreated either with an H2O2 scavenger (catalase, 2000 i.u.), a hydroxyl radical scavenger (mannitol, 10(-2) M), a nitric oxide-synthesis inhibitor (L-NG-monomethyl arginine, LNMA, 2 x 10(-4) M), or melatonin (10(-6) M to 10(-4) M). The effect of H2O2 (10(-4) M) on the relaxation induced by the calcium ionophore A23187 was also determined in arterial segments, with or without pretreatment with melatonin (10(-6) M, 10(-4) M). H2O2 (10(-6) M to 10(-4) M) potentiated vascular tension in a concentration-dependent manner (P < 0.0001). Pretreatment with LNMA significantly suppressed the vasospastic effect of H2O2 (P < 0.0001). Pretreatment with either catalase or mannitol significantly reduced the vasospastic effect of H2O2 (P < 0.005, P < 0.002, respectively). Melatonin also significantly reduced the vasospastic effect of H2O2 in a concentration-dependent manner (H2O2 10(-6) M, P < 0.0001 : H2O2 10(-5) M, P < 0.0001 : H2O2 10(-4) M, P < 0.00001). Pretreatment with H2O2 significantly inhibited the relaxation induced by the calcium ionophore A23187 (P < 0.005). Treatment with melatonin prior to exposure to H2O2 significantly restored the relaxation induced by A23187 (P < 0.005). Results suggest that H2O2 potentiates vascular tension in the human umbilical artery, perhaps by suppressing the endothelial synthesis of nitric oxide. Melatonin significantly suppressed the vasospastic effect of H2O2, possibly due to its ability to scavenge the hydroxyl radical.