SUMMARY We employed intracoronary infusions of calf intenstine adenosine deaminase (ADA) to test the hypothesis that adenosine regulates coronary blood flow during myocardial reactive hyperemia (RH). Infusions of 4.5 U ADA/min per kg body weight into the left circumflex coronary artery of 10 open-chest dogs reversibly reduced repayment of flow debt by 30-39% ( P < 0.05) following 5-, 10-, 15-, 20-, and 30-second coronary occlusions, the percentage reduction being independent of occlusion length. ADA reduced peak RH flow rate (17%, P < 0.05) only after 5-second occlusions. Intracoronary infusions of ["']ADA in seven dogs produced interstitial ADA concentrations between 1.2 and 13.1 U/ ml in perfused myocardium and, In five of these dogs, '"I activity in the cardiac node was 1.8-35 times that of contiguous mediastinal tissue. Theophylline, a specific adenosine antagonist, reduced repayment of flow debt by 27-38% {P < 0.02) in eight dogs, an effect similar to that of ADA, In six other dogs, ADA plus theophylline did not reduce RH flow debt repayment below that produced by ADA alone. This experiment confirms the contribution of adenosine to myocardial RH but shows that this nucleoside accounts for but a third of volume flow. Other, as yet unidentified, factors are collectively more important. Ore Res 49: 1383Res 49: -1267 TWO lines of evidence support the hypothesis that adenosine regulates coronary blood flow during myocardial reactive hyperemia: (1) adenosine appears in coronary venous blood during reactive hyperemia in quantities paralleling the duration of the preceding ischemic interval (Rubio et al., 1969), and (2) coronary flow rate changes concordantly during reactive hyperemia with changes in myocardial adenosine content (Olsson et al., 1978).However, other evidence challenges this view. Methylxanthines strongly antagonize the coronary effects of exogenous adenoaine (Afonso, 1970), yet only modestly reduce the reactive hyperemic response (Bittar and Pauly, 1971;Juhran et al., 1971; Cumish et al., 1972;Eikens and Wilcken, 1977). Coronary occlusions of one heart beat or less have no detectable effect on myocardial metabolism, yet elicit hyperemic responses (Eikens and Wilcken, 1974; Schwartz et el, 1980). Other vasodilatory substances such as K + and CO 2 (or H + ) accumulate in ischemic hearts and also could participate in reactive hyperemia (Raberger et al., 1975;Case, 1976;Murray and Sparks, 1978).This study estimates the contribution of adenosine to canine myocardial reactive hyperemia by employing adenosine deaminase (ADA) to destroy adenosine in the cardiac interstitium. This enzyme reduces but does not abolish this response, confirming that adenosine contributes to, but is not solely responsible for, post-ischemic coronary vasodilation. MethodsHealthy mongrel dogs weighing 15.5-30 kg were anesthetized by sodium pentobarbital (30 mg/kg, iv) and artificially ventilated at rates maintaining arterial blood Poj and Pco 2 in the normal range (Feigl and D'Alecy, 1971). Catheters in the right femoral vein a...

Analysis of the relationships between cardiac oxygen usage (MVO2), cardiac muscle adenosine levels([Ado]), and coronary vascular resistance (R) in open-chest, anesthetized dogs tested the hypothesis that adenosine is a physiological regulator of coronary flow. Experiments using each dog as its own control showed that [Ado] varied directly with MVO2 as the latter changed spontaneously or in response to atrial pacing, paired pacing, aortic constriction, of beta-adrenergic blockade. In turn, R varied inversely with changes in [Ado]. Stimulating MVO2 with isoproterenol significantly increased the slope of the regression of [Ado] on MVO2 as well as of R-1 on [Ado]. The effect of beta-adrenergic stimulation on [Ado] is unexplained, but its effect on R-1 seems to reflect the combined effects of adenosine and direct beta-adrenergic coronary relaxation. These results support the hypothesis that adenosine mediates the coronary flow responses to changes in MVO2.

Several experiments challenge the view that adenosine released during an interval of coronary occlusion (7) mediates the ensuing myocardial reactive hyperemia response (8,10). Cornary occlusions lasting 1 or 2 heartbeats (3) or less than 1 heart beat (11) probably do not cause myocardial ischemia yet elicit hyperemic responses, evidence that myogenic coronary relaxation may contribute to reactive hyperemia. Isobaric coronary perfusion with deoxygenated blood (5) or buffer solutions (12) in lieu of ischemia elicits a flow response that is significantly smaller than the hyperemic response to an equivalent interval of coronary occlusion. This evidence for the participation of myogenic relaxation and/or diffusible vasodilatory metabolites is augmented by experiments which tentatively identify CO2 (or H +) as one such metabolite. The intensity of the hyperemic response varies concordantly with arterial pCO2 (9). Finally, methylxanthines, which strongly inhibit the coronary vasoactivity of exogenous adenosine (1), curtail reactive hyperemic responses only modestly (6).We have begun to assess the individual contributions of adenosine, myogenic relaxation and other vasodilatory responses to myocardial reactive hyperemia. The studies reported herein test the hypothesis that adenosine deaminase, by destroying adenosine released into the cardiac interstitium, will reduce the intensity of reactive hyperemia. Materials and methodsCalf intestine adenosine deaminase, Type I, was purchased from Sigma Chemical Company. The experiments employed mongrel dogs anesthetized with chloralose (110 mg/kg i.v.) after premedication with morphine (2 mg/kg s.c.). Left thoracotomy exposed the left circumflex coronary artery, which was dissected for the implantation of an electromagnetic flowmeter, pneumatic occlusive cuff and an intracoronary catheter (4). A polyvinyl catheter advanced into the aortic root via the left carotid artery served for blood pressure monitoring.The experimental design called for comparisons of the hyperemic responses to 5, 10, 15, 20 and 30 sec coronary occlusion, before, during and after the intracoronary infusion of adenosine deaminase. Control observations consisted of a randomized series of two coronary occlusions of each duration during the intraeoronary infu-*) Supported by a Grant-In-Aid from the Suncoast Chapter, Florida AHA Affiliate, and NIH HL 20782-02. 955