The effect of chronic levodopa administration on the functional activity of the basal ganglia and its output regions was evaluated by means of the 2-deoxyglucose (2-DG) autoradiographic technique in rats with a unilateral 6-hydroxydopamine lesion of the nigrostriatal pathway. The rates of local cerebral glucose utilization were studied under basal conditions as well as in response to challenge with a selective D1 or D2 dopamine-receptor agonist. Levodopa (100 mg/kg/d, i.p.) was administered for 19 d either continuously via infusion with an osmotic pump or intermittently by twice-daily injections. Following a 3-d washout, glucose utilization was found to be decreased by both levodopa regimens in the nucleus accumbens; intermittent levodopa also decreased glucose utilization in the entopeduncular nucleus, subthalamic nucleus, ventrolateral thalamus, ventromedial thalamus, ventroposterolateral thalamus, and lateral habenula. In control (lesioned and treated chronically with saline) rats, the D1 agonist SKF 38393 (5 mg/kg, i.v.) increased 2-DG uptake in the substantia nigra pars reticulata and entopeduncular nucleus ipsilateral to the lesion by 84% and 56%, respectively. Both continuous and intermittent levodopa blunted the SKF 38393-induced elevation in glucose metabolism in the substantia nigra pars reticulata, while intermittent levodopa also attenuated the increase in the entopeduncular nucleus. The S2 agonist quinpirole (0.4 mg/kg, i.v.) did not increase glucose utilization in any brain region in control animals; following intermittent levodopa treatment, however, quinpirole increased 2-DG uptake by 64% in the subthalamic nucleus and by 39% in the deep layers of the superior colliculus on the ipsilateral side. These findings indicate that chronic levodopa treatment has long-lasting effects on the functional activity of brain regions within the basal ganglia as well as in regions that are targets of basal ganglia output. The effects of chronic levodopa are dependent on the treatment regimen employed: Administration of levodopa on a intermittent basis causes alterations in glucose utilization that are more pronounced and widespread than those of the same daily dose of levodopa given by continuous infusion. Our results also suggest that chronic levodopa treatment differentially alters D1 and D2 receptor-mediated striatal output, decreasing D1 output through the striatonigral and striatoentopeduncular pathways and increasing D2 output through the striatopallidal pathway.