To clarify the role of serotonin in eliciting myocardial ischemia, the effects of serotonin on coronary vasculature and cardiac performance were examined in anesthetized open chest dogs. Without coronary stenosis, intracoronary infusion of serotonin (0.001 to 1.0 micrograms/kg per min) for 40 s caused dose-dependent increases in coronary blood flow up to 175 +/- 8.3%. During fixed coronary stenosis produced by a metal constrictor that restricted active vasomotion of the stenosed segment, only the highest dose of serotonin increased coronary blood flow by 23 +/- 5.7% and decreased distal coronary pressure without any changes in stenosis resistance and systemic hemodynamic variables. During dynamic coronary stenosis created by inflation of an intraluminal microballoon that preserved active vasomotion of the stenosed segment, intracoronary serotonin (0.1 and 1.0 micrograms/kg per min) evoked marked decreases in coronary blood flow and distal coronary pressure and an increase in stenosis resistance, resulting in an elevation of left ventricular end-diastolic pressure and worsening of left ventricular dP/dt. These deleterious effects were reversed by additive intracoronary infusion of nitroglycerin (10 micrograms/min). The detrimental effects of serotonin were not attenuated by pretreatment with aspirin (10 mg/kg, intravenously), which suppressed adenosine disphosphate- and arachidonic acid-induced platelet aggregation in vitro. Furthermore, intracoronary infusion of serotonin (0.1 and 1.0 micrograms/kg per min) from the site distal to coronary stenosis failed to decrease coronary blood flow during dynamic coronary stenosis. Therefore, these detrimental effects of serotonin during dynamic coronary stenosis could be caused by constriction of a large coronary artery but not by platelet aggregation. Vascular effects of serotonin without coronary stenosis and with dynamic coronary stenosis were inhibited by pretreatment with methysergide (0.3 mg/kg, intravenously), but not with ketanserin (0.5 mg/kg, intravenously) which inhibited serotonin-induced platelet aggregation in vitro. Thus, serotonin produced opposing effects on coronary vasculature, that is, small coronary artery dilation and large coronary artery constriction. Both vascular effects were mediated by non-S2-serotonergic receptor activation of the coronary artery. The data obtained in this study suggest that serotonin-elicited constriction of a large coronary artery may play a crucial role in the pathogenesis of myocardial ischemia during pliable coronary stenosis.