The aim of the present study was to investigate the influence of reactive oxygen species (ROS) on the contractile responses of rat isolated left atria to muscarinic receptor stimulation. ROS were generated by means of electrolysis (30 mA, 75 s) of the organ bath fluid. Twenty minutes after the electrolysis period, the electrically paced atria (3 Hz) were stimulated with the adenylyl cyclase activator forskolin (1 microM). Subsequently, cumulative acetylcholine concentration-response curves were constructed (0.01 nM-10 microM). In addition, phosphoinositide turnover and adenylyl cyclase activity under basal and stimulated conditions were measured. For these biochemical experiments we used the stable acetylcholine analogue carbachol. The atria exposed to reactive oxygen species were influenced more potently (pD2 control: 6.2 vs. 7.1 for electrolysis-treated atria, P<0.05) and more effectively (Emax control: 40% vs. 90% reduction of the initial amplitude, P<0.05) by acetylcholine. In contrast, ROS exposure did not alter the responses to adenosine, whose receptor is also coupled via a Gi-protein to adenylyl cyclase. The basal (40% vs. control, P<0.05) as well as the carbachol-stimulated (-85% vs. control, P<0.05) inositol-phosphate formation was reduced in atria exposed to ROS. The forskolin-stimulated adenylyl cyclase activity was identical in both groups but carbachol stimulation induced a more pronounced reduction in adenylyl cyclase activity in the electrolysis-treated atria. Accordingly we may conclude that ROS enhance the negative inotropic response of isolated rat atria to acetylcholine by both a reduction of the positive (inositide turnover) and increase of the negative (adenylyl cyclase inhibition) inotropic components of cardiac muscarinic receptor stimulation. This phenomenon is most likely M2-receptor specific, since the negative inotropic response to adenosine is unaltered by ROS exposure.