Coronary microvascular diameter is significantly influenced by adenosine and flow. However, the interaction between these two regulatory mechanisms in the control of coronary microvascular tone remains unknown. Because adenosine can activate ATP-sensitive K+ (KATP) channels and these channels are located on the endothelium in addition to vascular smooth muscle, we hypothesized that adenosine can potentiate flow-induced vasodilation by activating endothelial KATP channels in the coronary microcirculation. To test this hypothesis, experiments were performed in porcine subepicardial coronary arterioles (50-150 microns) using isolated, cannulated vessel techniques to allow intraluminal pressure and flow to be independently controlled. All vessels developed active tone, approximately 67-73% of maximum diameter, at 60 cmH2O intraluminal pressure and showed graded dilation to stepwise increases in flow. The magnitude of flow-induced dilation was potentiated by a threshold dose of adenosine (10(-10) M) but not by nitroprusside (10(-10) M). Luminal application of a high K+ concentration ([K+]) (40 mM) completely blocked flow-induced arteriolar dilation. In addition, luminal glibenclamide (10(-6) M) abolished the adenosine-potentiated component of flow-induced response. Indomethacin (10(-5) M) did not alter the dose-dependent dilation to adenosine. However, endothelial denudation, NG-monomethyl-L-arginine (10(-5) M), and luminal administration of a high [K+] or glibenclamide each produced identical inhibition of adenosine-induced vasodilation by shifting the 50% effective dose to the right by an order of magnitude. In contrast, vasodilation in response to nitroprusside was not altered by these pharmacological interventions.(ABSTRACT TRUNCATED AT 250 WORDS)