Adenosine is a possible mediator of cardiac pain during myocardial ischemia; however, little is known about the influence of adenosine on cardiac sympathetic afferent activity and thereby on its algogenic mechanism. In 20 anaesthetized, decerebrated, curarized and artificially ventilated cats, we studied the impulse activity of 20 single afferent sympathetic fibers with a left ventricular receptive field in relation to epicardial applications of adenosine, coronary artery occlusions and arterial pressure rises. All fibers increased their impulse activity (from 1.2 +/- 0.2 to 2.6 +/- 0.5 imp/s; P < 0.001) during slight (20 +/- 8%) rises in aortic pressure, thus exhibiting low-threshold receptor characteristics. In 10 cats, epicardial applications of three different doses of adenosine (0.1, 1 and 10 mg/ml) caused a brief increase in neural activity with dose-related responses. This response was abolished by aminophylline, a P1 purinergic inhibitor. In the other group of 10 cats, four subsequent 30-s occlusions of the coronary arterial vessel supplying the receptive fields of the fibers were performed, in control conditions and 30 s, 3 and 7 min, respectively, after the end of excitation induced by adenosine (1 mg/ml) application. During the control coronary occlusion the impulse activity increased from 1.1 +/- 0.1 to 5.5 +/- 0.7 imp/s (P < 0.0001). A similar activation was present during the second occlusion initiated 30 s after the end of adenosine-induced activation. In contrast, a significant potentiation of the response was observed (8.8 +/- 1.2 vs. 5.3 +/- 0.9 imp/s; P < 0.001) during the occlusion initiated 3 min after the end of excitation by adenosine. This effect was no longer present during the last occlusion performed after 7 min. When the protocol was repeated substituting adenosine with saline (n = 5) or after i.v. administration of aminophylline (n = 5), no potentiation was observed, even though the excitatory response to coronary occlusion was preserved. These data show that adenosine can activate cardiac sympathetic afferent fibers in a dose-related manner, and potentiate their responses to coronary occlusion, while leaving unaffected the responsiveness to a hemodynamic stimulus. The excitatory effects are likely to involve the P1 purinergic receptors. The potentiation phenomenon might play a role in the genesis of an algogenic code.