In the present study the influence of in vivo administration, or in vitro addition, of the prototypic 5-HT1A receptor agonist 8-OH-DPAT on in vitro characteristics of GABA(A)/benzodiazepine receptor complexes was examined. In vivo administration of 8-OH-DPAT at a dose (32 mu g/kg, s.c. -10') that has been reported to produce anxiolytic-like effects in the elevated plus-maze doubled the K-d for in vitro binding of H-3-flunitrazepam to rat cortical membranes (B-max was unchanged) and enhanced GABA-stimulated (3, 10, 30 and 100 mu M) Cl-36(-) influx in corticohippocampal synaptoneurosomes. In synaptoneurosomes from vehicle treated rats, diazepam (1, 3 and 10 mu M) potentiated GABA-stimulated (3 mu M) Cl-36(-) influx. No such effect was observed in tissue from 8-OH-DPAT treated rats, in which the GABA-stimulated (3 mu M) Cl-36(-) influx was similar to that caused by GABA + diazepam in tissue from vehicle treated rats. When added in vitro, 8-OH-DPAT failed to alter basal or GABA-stimulated Cl-36(-) uptake. In vivo administration of a low ''anxiolytic'' dose of 8-OH-DPAT thus appears to interfere with GABA(A)/benzodiazepine receptor complexes, whereas in vitro application does not. The underlying mechanism remains to be elucidated but could involve in vivo release of positive modulators of GABA(A)/benzodiazepine receptor complexes, e.g. GABA, endozepines or neurosteroids. (C) 1997 Elsevier Science Ltd.