We evaluated the hydroxyl radical scavenging effect of melatonin on the vasospastic action induced by hydrogen peroxide in human umbilical artery. Helical sections were made of umbilical arteries obtained from healthy pregnant women who were delivered between 37 and 39 weeks of gestation. Changes in maximal potassium chloride (KCl, 10[-2] M)-induced tension were measured in umbilical artery segments with intact endothelium. Segments were treated with H(2)O(2) (10[-9] M to 10[-7] M) only, or were pretreated with an H(2)O(2) scavenger (catalase, 2,000 IU), a hydroxyl radical scavenger (mannitol, 10[-2] M), or melatonin (10[-8] M to 10[-6] M). The effect of H(2)O(2) on the response of the segments of umbilical artery to external calcium was determined. Changes in KCl-induced contraction were also determined in segments pretreated with an inhibition of intracellular calcium release (ryanodine, 10[-4] M) prior to exposure to H(2)O(2). Pretreating the segments of umbilical arteries with H(2)O(2) (10[-8] M, 10[-7] M) significantly potentiated the maximal contraction induced by KCl (P < 0.0001, P < 0.03, respectively). Pretreatment with either catalase or mannitol significantly reduced the vasospastic effect of H(2)O(2) (10[-8] M) (P < 0.0001, P < 0.0001, respectively). Melatonin also significantly reduced the vasospastic effect of H(2)O(2) (10[-8] M), in a concentration-dependent manner (P < 0.0001). H(2)O(2) (10[-8] M) significantly increased the contractile response to external calcium. Melatonin pretreatment significantly suppressed the contractile response to external calcium. Treatment with ryanodine prior to exposure to H(2)O(2) did not affect KCl-induced contraction. Results suggest that H(2)O(2) potentiates the KCl-induced maximal contraction of the human umbilical artery, perhaps by increasing calcium influx via activation of the voltage-dependent calcium channel. Melatonin significantly suppresses the vasospastic effect of H(2)O(2), probably due to its scavenging of the hydroxyl radical.