The goal of this study was to test the hypotheses that endogenous ADP plays an important role in vivo in mediating platelet aggregation and cyclic coronary artery blood flow variations (CFVs) in stenosed and endothelium-injured coronary arteries in an experimental canine model. Anesthetized animals were studied and coronary blood flow velocities monitored by a pulsed Doppler flow probe positioned around the left anterior descending coronary artery. CFVs were established by an external constrictor positioned at sites with injured endothelium. Apyrase, an ADP-removing enzyme, was infused into the left anterior descending coronary artery (0.3-1.8 units/min) 30 minutes or 2 hours after the establishment of CFVs. Complete abolition of CFVs was achieved in 81% (13/16) of dogs with 30-minute CFVs and in 83% (five of six) of dogs with 2-hour CFVs. In other dogs, a potent inhibitor of ADP-induced platelet aggregation, clopidogrel, was administered as a 10 mg/kg i.v. bolus and a 2.5 mg/kg/hr infusion 30 minutes and 3 hours after the establishment of CFVs. This treatment resulted in complete abolition of CFVs in 14 dogs (100%) with either 30-minute or 3-hour CFVs. Epinephrine was infused into some dogs after CFVs had ceased as a result of either apyrase or clopidogrel administration and into some dogs in whom SQ29548, a thromboxane A2 receptor antagonist, had been given when apyrase failed to abolish CFVs. Epinephrine restored CFVs in all dogs treated with apyrase alone, 67% (four of six) of dogs treated with the combination of apyrase and SQ29548, and 29% (two of seven) of dogs treated with clopidogrel. The plasma epinephrine levels required for CFV restoration were 20 times higher than baseline values in dogs receiving apyrase alone, 100 times higher when a combination of apyrase and SQ29548 had been given, and more than 5,000 times higher in dogs receiving clopidogrel. In vitro studies showed that apyrase only inhibited ADP-induced platelet aggregation, whereas clopidogrel not only inhibited ADP-induced platelet aggregation, but also reduced platelet aggregation induced by the thromboxane mimetic U46619 and serotonin. These data suggest that 1) ADP is an important mediator of platelet aggregation and CFVs in vivo and 2) combined inhibition of thromboxane A2 and ADP's effects provides marked protection against CFVs in experimentally stenosed and endothelium-injured canine coronary arteries. These data and our previous observations are consistent with the possibility that specific antagonists of thromboxane A2, serotonin, and ADP, alone and together, may provide substantial protection against platelet aggregation leading to CFVs at sites of endothelial injury and coronary artery stenosis.