In vivo microdialysis was used to study the effects of the locally applied GABA(A) receptor agonist muscimol and GABA(A) receptor antagonist bicuculline on the basal dopamine efflux as well as on the endomorphin-1- and endomorphin-2-induced dopamine efflux in the nucleus accumbens of freely moving rats. Muscimol (2500 pmol) and bicuculline (5 and 10 nmol) increased basal dopamine efflux. Bicuculline (50 pmol) inhibited the muscimol (2500 pmol)-induced dopamine efflux. Muscimol (250 pmol), but not bicuculline (50 and 500 pmol), enhanced the endomorphin-1 (25 nmol)-induced dopamine efflux. Bicuculline (50 pmol) counteracted the muscimol (250 pmol)-induced increase of the endomorphin-1-elicited dopamine efflux. Neither muscimol (25 and 250 pmol) nor bicuculline (50 and 500 pmol) affected the endomorphin-2 (25 nmol)-induced dopamine efflux. The doses mentioned are the total amount of drug over the infusion period (25 or 50 min) that varied across the drugs. The finding that muscimol and bicuculline increased basal dopamine efflux may imply that these drugs acted at different sites. It is suggested that (1) muscimol acts at GABA(A) receptors on GABA-ergic neurons that exert an inhibitory control of dopaminergic neurons and, accordingly, disinhibits these dopaminergic neurons, and that (2) bicuculline acts directly at GABA(A) receptors on dopaminergic neurons and, accordingly, removes the inhibitory control of these dopaminergic neurons. The finding that an agonist, but not antagonist, of GABA(A) receptors enhanced the endomorphin-1's effects might indicate that endomorphin-1 produced a floor effect at the level of GABA(A) receptors located on presynaptic, dopaminergic terminals. Finally, the present results support our earlier reported notion that endomorphin-1 and endomorphin-2 increase accumbal dopamine efflux by different mechanisms.