Adenosine inhibition of hormone-sensitive adenylate cyclase activity was investigated using isolated myocardial membranes prepared from rat hearts. When cyclase activity was determined in membranes, using [alpha-32P]ATP as substrate, 10(-5) M adenosine inhibited isoproterenol-stimulated adenylate cyclase activity by 25% but did not inhibit basal activity or fluoride (5 mM) activation of the enzyme. The adenosine reduction of isoproterenol-sensitive cyclase activity was dependent on GTP but was not prevented by 10(-3) M theophylline. Adenosine neither appeared to compete with ATP for the substrate converting site of the enzyme nor reduced 5'-guanylyl imidodiphosphate activation of the enzyme. Inasmuch as lower concentrations of adenosine had no influence on enzyme activity, endogenous adenosine may be present in the adenylate cyclase assay. To obviate the effects of endogenous adenosine, the adenylate cyclase assay was then modified to a 2'-deoxy system with [alpha-32P]dATP used as the substrate in the presence of adenosine deaminase. With this assay system, the 15% inhibition of isoproterenol-stimulated adenylate cyclase activity produced by the adenosine receptor agonists, 10(-8) M 2-chloroadenosine or phenylisopropyladenosine, was prevented by 10(-4) M 8-phenyltheophylline or isobutylmethylxanthine (IBMX), respectively. While under these assay conditions, 10(-7) M 2',5'-dideoxyadenosine, a P-site analogue, did not influence the hormone-sensitive cyclase activity. The 35% reduction of the hormone-sensitive enzyme produced by this analogue at 10(-5) M was not prevented by IBMX. These results suggest that nanomolar concentrations of adenosine analogues interact with a methylxanthine-sensitive adenosine receptor that mediates the attention of membrane hormone-sensitive adenylate cyclase activity.