The effects of propofol, pentobarbital, alphaxalone, etomidate and diazepam on t-[S-35]butylbicyclophosphorothionate ([S-35]TBPS) binding to membrane preparations from rat cerebral cortex were studied in the absence of gamma-aminobutyric acid (GABA). Addition of low concentrations (3-10 mu M) of propofol to washed membrane preparations (devoid of GABA) markedly enhanced [S-35]TBPS binding (maximal enhancement, 85%), whereas higher concentrations (50-100 mu M) inhibited this parameter. Diazepam also enhanced [S-35]TBPS binding in this preparation (maximal enhancement, 38%). In contrast, pentobarbital, alphaxalone and etomidate decreased [S-35]TBPS binding in a concentration-dependent manner. The propofol-induced increase in [S-35]TBPS binding in washed membranes was completely reversed by the addition of GABA at a concentration (0.3 mu M) that alone did not modify [S-35]TBPS binding (78% increase with 10 mu M propofol alone, 33% decrease in the additional presence of GABA). The ability of GABA to reverse the effect of propofol on [S-35]TBPS binding in washed membranes was shared by pentobarbital (200 mu M) and alphaxalone (3 mu M); etomidate (20 mu M) only partially antagonized the effect of propofol. Diazepam at a concentration (30 mu M) that alone had no effect on [S-35]TBPS binding failed to modify the propofol-induced increase in [S-35]TBPS binding, whereas at a concentration (3 mu M) that alone increased [S-35]TBPS binding the effect of diazepam was additive with that of propofol. The addition of bicuculline to washed membranes failed to abolish the increase in [S-35]TBPS binding induced by propofol or diazepam, but completely antagonized the effects of pentobarbital, alphaxalone and etomidate. Moreover, propofol and diazepam further enhanced the bicuculline-induced increase in [S-35]TBPS binding in unwashed membranes. Our data thus show that, by assaying [S-35]TBPS binding in washed membranes, it is possible to differentiate the molecular action of propofol from that of other general anesthetics.