Treatment with haloperidol, a dopamine receptor D-2 antagonist, for one month resulted in an increase in the mean percentage of asymmetric synapses containing a discontinuous, or perforated, postsynaptic density (PSD) [Meshul et al. (1994) Brain Res., 648:181-195] and a change in the density of striatal glutamate immunoreactivity within those presynaptic terminals [Meshul and Tan (1994) Synapse, 18:205-217]. We speculated that this haloperidol-induced change in glutamate density might be due to an activation of the corticostriatal pathway. To determine if activation of this pathway leads to similar morphological changes previously described following haloperidol treatment, GABA (10(-5) M, 0.5 microliters) was injected into the thalamic motor (VL/VM) nuclei daily for 3 weeks. This treatment resulted in an increase in the mean percentage of striatal asymmetric synapses containing a perforated PSD and an increase in the density of glutamate immunoreactivity within nerve terminals of asymmetric synapses containing a perforated or non-perforated PSD. Subchronic injections of GABA into the thalamic somatosensory nuclei (VPM/VPL) had no effect on the mean percentage of synapses with perforated PSDs but resulted in a small, but significant, increase in density of glutamate immunoreactivity. Using in vivo microdialysis, an acute injection of GABA (10(-5) M, 15 microliters) into VL/VM resulted in a prolonged rise in the extracellular level of striatal glutamate. The increase in asymmetric synapses with perforated PSDs and in glutamate immunoreactivity within nerve terminals of the striatum following either subchronic haloperidol treatment or GABA injections into VL/VM suggest that an increase in glutamate release may be a common factor in these two experiments. It is possible that the extrapyramidal side effects associated with haloperidol treatment may be due, in part, to an increase in release of glutamate within the corticostriatal pathway.