Purpose: Adenosine fine-tuning regulates cardiovascular function. The negative chronotropic effect of adenosine depends on the activation of the most abundant adenosine receptor in the heart, the A1AR, leading to GIRK/Kir3.1/3.4 channels opening in the SA node. The A1AR-mediated responses in the atrial muscle also depend on KCa2/SK channel function (unpublished observations). Since these channels are involved in atrial repolarization and in human rhythm disturbances, we aimed at studying the interplay between A1AR and SK channels both in situ and single-cell preparations of rat atria. Methods: Isometric contraction experiments were performed on isolated spontaneously beating atria from Wistar rats. For whole-cell voltage-clamp recordings, atrial myocytes were enzymatically isolated by retrograde perfusion. Immunolocalization studies (A1AR, GIRK1, KCa2.2 and KCa2.3) were evaluated by confocal microscopy. The SA node was identified as a connexin-43 low / neurofilament-160 high immunoreactive region. Results: The selective A1AR agonist R-PIA (0.001-1μM) decreased atrial chronotropism and inotropism in concentration-dependent manner. However, the negative chronotropic effect of R-PIA (IC50~0.03 μM) was evidenced at much lower (~30 times) concentrations than the negative inotropic action (IC50~1 μM). Selective blockade of A1AR and of GIRK/Kir3.1/3.4 channels, respectively with DPCPX (100 nM) and tertiapin Q (300 nM), significantly attenuated the effects R-PIA. Blockade of KCa2/SK channels with apamin (30 nM) potentiated the negative inotropic response of R-PIA, without affecting atrial rate. Apamin-induced atrial sensitization to R-PIA was not observed with other cardiodepressant agents, namely oxotremorine (0.01-3 μM). Voltage-clamp experiments demonstrated that adenosine, via A1AR, plays a dual role in atrial cardiomyocytes by activating inwardly rectifying GIRK/Kir3.1/3.4 and inactivating outward Ca2+-activated KCa2/SK currents upon cell depolarization. Immunolabelling of A1AR, KIR3.1, KCa2.2 and KCa2.3 was observed throughout the right atria. Conclusion: Data demonstrate for the first time that A1AR has opposing effects on distinct K+ currents operating atrial repolarization. While A1AR activation favors the opening of inwardly rectifying GIRK/KIR3.1/3.4 channels modulating atrial automatism, it may also inactivate outward KCa2/SK currents. Taking into consideration that KCa2/SK channel dysfunction has pro-arrhythmic effects drugs targeting the A1AR for supraventricular tachycardia must be used with caution as they might precipitate atrial arrhythmias.