Suramin acts as a G protein inhibitor because it inhibits the rate-limiting step in activation of the G_α subunit, i.e., the exchange of GDP for GTP. Here, we have searched for analogues that are selective for G_sα. Two compounds have been identified: NF449 (4,4′,4",4′"-[carbonyl-bis[imino-5,1,3-benzenetriyl bis-(carbonylimino)]]tetrakis-(benzene-1,3-disulfonate) and NF503 (4,4′-[carbonylbis[imino-3,1-phenylene-(2,5-benzimidazolylene)carbonylimino]]bis-benzenesulfonate). These compounds (i) suppress the association rate of guanosine 5′-[γ-thio]triphosphate ([³⁵S]GTP[γS]) binding to G_sα-s but not to G_iα-1, (ii) inhibit stimulation of adenylyl cyclase activity in S49 cyc⁻ membranes (deficient in endogenous G_sα) by exogenously added G_sα-s, and (iii) block the coupling of β-adrenergic receptors to G_s with half-maximum effects in the low micromolar range. In contrast to suramin, which is not selective, NF503 and NF449 disrupt the interaction of the A₁-adenosine receptor with its cognate G proteins (G_i/G_o) at concentrations that are >30-fold higher than those required for uncoupling of β-adrenergic receptor/G_s tandems; similarly, the angiotensin II type-1 receptor (a prototypical G_q-coupled receptor) is barely affected by the compounds. Thus, NF503 and NF449 fulfill essential criteria for G_sα-selective antagonists. The observations demonstrate the feasibility of subtype-selective G protein inhibition.