The dose- and time dependence of melatonin and the effective window of melatonin administration were determined in a mouse model of myocardial infarction. When mouse hearts were subjected to 60 min of occlusion of the left anterior descending artery (LAD) followed by 4 h of reperfusion, melatonin pretreatment for 30 min significantly reduced the infarct size/risk area. The most effective dose was found to be 150 microg/kg intraperitoneally, and the effective period of protection lasted up to 2 h after melatonin administration. Melatonin administration 45 min after LAD ligation or right before reperfusion was as effective as administration 30 min before ligation; however, melatonin administered after the release of occlusion was not protective. Melatonin's effect was still present in mice deficient for the Mel1a melatonin receptor. 8-Methoxy-2-propionamidotetralin, a melatonin receptor agonist with no antioxidant activity, offered no protection, suggesting a lack of involvement of melatonin receptors. Finally, the effects of melatonin were similar in rats and mice. Our results demonstrate that melatonin is an effective cardioprotective agent when administered either before or during coronary occlusion at a very low dose.