The release of the neurotransmitter norepinephrine (NE) is modulated by presynaptic adenosine receptors. In the present study we investigated the effect of a partial activation of this feedback mechanism. We hypothesized that partial agonism would have differential effects on NE release in isolated hearts as well as on heart rate in vivo depending on the genetic background and baseline sympathetic activity. In isolated perfused hearts of Wistar and Spontaneously Hypertensive Rats (SHR), NE release was induced by electrical stimulation under control conditions (S1), and with capadenoson 6 · 10(-8) M (30 µg/l), 6 · 10(-7) M (300 µg/l) or 2-chloro-N(6)-cyclopentyladenosine (CCPA) 10(-6) M (S2). Under control conditions (S1), NE release was significantly higher in SHR hearts compared to Wistar (766+/-87 pmol/g vs. 173+/-18 pmol/g, p<0.01). Capadenoson led to a concentration-dependent decrease of the stimulation-induced NE release in SHR (S2/S1 = 0.90 ± 0.08 with capadenoson 6 · 10(-8) M, 0.54 ± 0.02 with 6 · 10(-7) M), but not in Wistar hearts (S2/S1 = 1.05 ± 0.12 with 6 · 10(-8) M, 1.03 ± 0.09 with 6 · 10(-7) M). CCPA reduced NE release to a similar degree in hearts from both strains. In vivo capadenoson did not alter resting heart rate in Wistar rats or SHR. Restraint stress induced a significantly greater increase of heart rate in SHR than in Wistar rats. Capadenoson blunted this stress-induced tachycardia by 45% in SHR, but not in Wistar rats. Using a [(35)S]GTP?S assay we demonstrated that capadenoson is a partial agonist compared to the full agonist CCPA (74+/-2% A(1)-receptor stimulation). These results suggest that partial adenosine A(1)-agonism dampens stress-induced tachycardia selectively in rats susceptible to strong increases in sympathetic activity, most likely due to a presynaptic attenuation of NE release.