BACKGROUND Adenosine has been shown to reduce infarct size predominantly during reperfusion by adenosine A2-receptor-mediated processes. This cardioprotection may involve inhibition of events in the vascular compartment, such as adherence-independent and adherence-dependent actions of neutrophils. This study tested the hypothesis that adenosine exerts its cardioprotection during reperfusion by targeting effectors in the vascular compartment. METHODS Polyadenylic acid (molecular weight, 230,000 daltons) was used as an intravascularly confined adenosine mimetic. In anesthetized New Zealand white rabbits, the left coronary artery was occluded for 30 minutes and reperfused for 120 minutes. RESULTS Polyadenylic acid (1 mg/kg bolus, 0.5 mg kg-1 h-1) given 5 minutes before reperfusion significantly (p < 0.05) reduced infarct size compared with vehicle (23% +/- 2% versus 37% +/- 2% area at risk). The A1-antagonist KW-3902 had no effect on this polyadenylic acid-induced protection (17% +/- 3%), whereas the A1-A2 antagonist sulfophenytheophylline blocked this infarct size reduction (41% +/- 2%). In vitro adherence of platelet-activating factor-activated neutrophils to thoracic aortic endothelium was significantly diminished by polyadenylic acid (185 +/- 12 neutrophils/mm2 versus 36 +/- 4 neutrophils/mm2 endothelial surface). Sulfophenytheophylline inhibited this effect (280 +/- 6 neutrophils/mm2), whereas KW-3902 did not (31 +/- 7 neutrophils/mm2). CONCLUSIONS An intravascular adenosine mimetic agent exerts cardioprotection during reperfusion by targeting receptor-mediated mechanisms in the intravascular compartment, possibly involving inhibition of neutrophil-related processes.