DRUG EFFICACY AT GUANINE NUCLEOTIDE-BINDING REGULATORY PROTEIN-LINKEDRECEPTORS - THERMODYNAMIC INTERPRETATION OF NEGATIVE ANTAGONISM AND OF RECEPTOR ACTIVITY IN THE ABSENCE OF LIGAND

The mutual effects that a hormonal ligand (H) and a guanine nucleotide regulatory protein (G protein) exert on each other when simultaneousl y occupying distinct sites of the receptor molecule (R) can be viewed as the molecular mechanism of drug efficacy. These effects are predict able on the basis of a model assuming that the ternary complex between the three partners (HRG) reaches equilibrium in the membrane [J. Biol . Chem. 255:7108-7117 (1980)]. Ligands can be classified as agonists, neutral antagonists, or negative antagonists, depending on whether the y enhance, leave unchanged, or reduce, respectively, the spontaneous t endency of R to interact with G. Using this model and the assumption t hat the G protein response observed in membranes reflects the sum of l igand-independent (RG) and ligand-dependent (HRG) receptor-G protein c omplexes, we can explain virtually all the phenomenology reported earl ier for opioid receptor-mediated stimulation of GTPase, i.e., 1) exist ence of ligands with both "positive" and "negative" intrinsic activity (the latter termed negative antagonists), 2) equipotency of neutral a ntagonists for the competitive blockade of the responses elicited both by agonists and by negative antagonists, and 3) apparent heterogeneit y of binding sites for the binding isotherms of negative antagonists. The ternary complex model can also explain the differential effects of sodium on ligand binding and ligand-dependent GTPase activity, if we assume that this ion reduces the stability constant between receptor a nd G protein in membranes. Computer simulations predict that a negativ e antagonist exhibits a discrepancy between "biological" K(i) (obtaine d by Schild plots) and true dissociation constant for the receptor, wh ich increases as the fraction of "precoupled" receptors in the membran e increases. The demonstration of negative antagonism is definitive ev idence for the existence of receptor coupling (hence activity) in the absence of ligand. Using this experimental paradigm, we show here that spontaneous receptor activity occurs in isolated membranes but not in intact NG108-15 cells.