Coupling of the human A1 adenosine receptor to different heterotrimeric G proteins: evidence for agonist-specific G protein activation

1 The present study investigates the effect of varying ligand structure on the ability of agonists to activate guanine nucleotide-binding proteins of the Gi, Gs and Gq families via the A(1) adenosine receptor. 2 In CHO cells expressing this receptor, inhibition or potentiation of forskolin-stimulated cAMP accumulation was used as an end point to measure the activation of Gi and, in Pertussis toxin (PTX)-treated cells, Gs, respectively. Stimulation of inositol phosphate accumulation in PTX-treated cells was used as an index of Gq activation. 3 CPA (N-6-cyclopentyladenosine), NECA (5'-N-ethyl-carboxyamidoadenosine) and eight analogues of these ligands presented a range of guanine nucleotide-binding protein (G-protein)-activating profiles. Some ligands could only activate Gi (e.g. 2'deoxyCPA), some primarily Gi and Gs (and only weakly Gq) (e.g. 3'deoxyCPA), highlighting the importance of the ribose hydroxyls in agonist activation of multiple G proteins. CHA (N-6-cyclohexyladenosine) activated Gi, Gs and Gq, but was more efficacious than CPA in activating Gs. 4 The NECA analogues 5'-N-cyclopropyl-carboxamidoadenosine, 5'-N-cyclobutyl-carboxamidoadenosine and 5'-N-cyclopentyl-carboxamidoadenosine (CPeCA) also activated all three G proteins, although their ability to activate Gs and Gq (relative to CPA) was reduced with increasing substituent size, such that CPeCA produced only a small stimulation (at 100 muM) at Gq, but was a full agonist, relative to CPA, at Gi and Gs. 5 This study suggests that the A(1) adenosine receptor can adopt agonist-specific conformations, arising from small changes in ligand structure, which lead to the differential activation of Gi, Gs and Gq.