Cardiac sarcolemmal purity is essential for the verification of adenylate cyclase inhibition via A1-adenosine receptors.

Inhibition of cardiac adenylate cyclase by adenosine receptor agonists was reinvestigated in a more homogeneous sarcolemmal vesicular preparation than used in a previous study. Microsomal particles obtained by differential centrifugation were further fractionated on a shallow density gradient of Percoll. Two populations of plasma membrane vesicles were partially resolved. Identical peaks were identified for adenylate cyclase activity and [3H]ouabain binding, whereas 5'-nucleotidase activity and beta-adrenoceptor binding displayed an additional peak at higher density, where angiotensin converting enzyme, a marker for endothelial plasma membranes, was at maximal activity. Significant inhibition by N6-cyclohexyladenosine (CHA), as measured in each fractionation step following homogenization, was observed only at the activity peak of adenylate cyclase. Moreover, analysis of the degree and rank order of potency of several adenosine analogs was indicative for interaction with A1-adenosine receptors. Accordingly, the peak in adenosine receptor binding, using (-)[125I]iodo-N6-hydroxyphenyl-isopropyladenosine as the radioligand, coincided with CHA-inhibitable adenylate cyclase activity. By contrast, adenylate cyclase was slightly stimulated by CHA in the higher density range, an action suggested to be mediated via A2-adenosine receptors, which recently have been demonstrated to exist on guinea-pig coronary endothelium. It is concluded that the full extent of adenosine receptor-mediated adenylate cyclase inhibition in the heart is only to be demonstrated if contamination of the sarcolemmal preparation with endothelial membrane components is kept to a minimum.