The pathophysiology of idiopathic dystonias is still unknown, but it is regarded as a basal ganglia disorder. Previous experiments in the dt(sz) hamster, a model of primary paroxysmal dystonia, demonstrated reduced discharge rates and an abnormal pattern within the entopeduncular nucleus (EPN), a basal ganglia output structure. To clarify if this is based on abnormal gamma-aminobutyric acid(GABA)ergic or glutamatergic input, microinjections into the EPN were done in mutant hamsters in the present study. The GABA(A) receptor antagonists pentylenetetrazole and bicuculline exerted moderate antidystonic effects, while previous systemic administrations worsened dystonia in the dt(sz) mutant. GABA-potentiating drugs, i.e., the GABA(A) receptor agonist muscimol and the GABA transporter inhibitor 1,2,5,6-tetrahydro-1-[2-[[(diphenylmethylene)amino]oxy]ethyl]-3-pyridinecarboxy-lic acid (NNC-711), which are known to improve dystonia after systemic treatment in mutant hamsters, did not exert significant effects after EPN injections, but NNC-711 tended to increase the severity at the highest dose (2.5 ng bilateral). The NMDA receptor antagonist D(-)-2-amino-5-phosphopentanoic acid (AP-5) retarded the onset of a dystonic attack. However, this effect was not dose dependent and the AMPA receptor antagonist 2,3-dihydroxy-6-nitro-7-sulfamoylbenzol(f)quinoxaline (NBQX) alone or in combination with AP-5 and NNC-711, also failed to show any effects on dystonia. The present data do not provide clear evidence for an enhanced striatal GABAergic input or a reduced glutamatergic activation of the EPN via the subthalamic nucleus, i.e., more pronounced antidystonic effects of GABA(A) receptor antagonists and stronger prodystonic effects of GABAmimetics and glutamate receptor antagonists were expected. Nevertheless, previously found changes in entopeduncular activity probably play a critical pathophysiological role in dystonic hamsters.