Chloride secretion in the apical bath of A6 epithelia was recorded under short circuit conditions in the absence of apical Na⁺. Treatment of the apical surface with adenosine (ADO, 3 μM) evoked a transient current increase within 60 s of 8.53±0.57 μA/cm². The peak current declined to a steady plateau of 2.49±0.24 μA/cm² (n=9). The currents were inhibited by NPPB (100 μM) and abolished after removal of Cl^– from the basolateral bath and thus identified as Cl^– secretion. A similar Cl^– secretion was observed by apical addition of a specific A1 agonist (N6-cyclopentyladenosine, CPA, 1 μM). Peak and plateau currents obtained with CPA were 3.19±0.30 and 1.56±0.15 μA/cm² (n=19), respectively. The idea of an involvement of an A1 receptor was supported by the complete inhibitory effect of an A1 antagonist 1,3-dipropyl-8-cyclopentylxanthine (DPCPX, 0.1 μM). Because intracellular Ca²⁺ concentration is hardly affected by ADO or CPA stimulation, Cl^– secretion does not seem to occur through Ca²⁺ activated Cl^– channels, but G-protein-coupled Cl^– channels could be involved. Amiloride (AMI) exerted a dose-dependent inhibitory effect on ADO and CPA activation of the Cl^– currents. With CPA=1 μM the AMI concentration for half maximal inhibition (K_A) was 1.26 μM. Increasing CPA to 2 μM augmented K_A to 2.45 μM. The shift of K_A indicates a competition of AMI with CPA at the adenosine-binding site. Further characterization of the A1 receptor was achieved by cloning, sequencing and functional expression in Xenopus laevis oocytes. A cDNA encoding the Xenopus A1 adenosine receptor (xA1) was obtained by RT-PCR starting from total RNA from A6 cells. Oligonucleotide primers were designed based on the cDNA sequence for the xA1 (GenBank/EMBL/DDBJ database: accession number AJ249842). The cloned receptor was functionally tested in Xenopus laevis oocytes by co-injection of xA1 cRNA and cRNA′s encoding mGIRK1 and mGIRK2 which form a bona fide G-protein-coupled inward rectifying K⁺ channel (Kobayashi et al. 2002). Injected and control oocytes were exposed to high K⁺ solutions (90 mM) and inward currents (I_K) were recorded at −70 mV. Basal I_K values were 210±64 nA (n=9). 100 nM ADO elicited a 3.5-fold stimulation of I_K. The addition of 50 μM AMI to the bath reduced the ADO-stimulated I_K component by 57%. These observations are consistent with a competitive inhibition between ADO and AMI at the xA1 receptor and indicate that the A1 receptor can activate G-protein-coupled ion channels.