Background: Parkinson's disease (PD) patients with long-term levodopa (L-DOPA) treatment are suffering from severe circadian dysfunction. However, it is hard to distinguish that the circadian disturbance in patients is due to the disease progression itself, or is affected by L-DOPA replacement therapy. This study was to investigate the role of L-DOPA on the circadian dysfunction in a rat model of PD. Methods: The rat model of PD was constructed by a bilateral striatal injection with 6-hydroxydopamine (6-OHDA), followed by administration of saline or 25 mg/kg L-DOPA for 21 consecutive days. Rotarod test, footprint test, and open-field test were carried out to evaluate the motor function. Striatum, suprachiasmatic nucleus (SCN), liver, and plasma were collected at 6:00, 12:00, 18:00, and 24:00. Quantitative real-time polymerase chain reaction was used to examine the expression of clock genes. Enzyme-linked immunosorbent assay was used to determine the secretion level of cortisol and melatonin. High-performance liquid chromatography was used to measure the neurotransmitters. Analysis of variance was used for data analysis. Results: L-DOPA alleviated the motor deficits induced by 6-OHDA lesions in the footprint and open-field test (P < 0.01, P < 0.001, respectively). After L-DOPA treatment, Bmall decreased in the SCN compared with 6-OHDA group at 12:00 (P < 0.01) and 24:00 (P < 0.001). In the striatum, the expression ofBmall, Rora was lower than that in the 6-OHDA group at 18:00 (P < 0.05) and L-DOPA seemed to delay the peak of Per2 to 24:00. In liver, L-DOPA did not affect the rhythmicity and expression of these clock genes (P > 0.05). In addition, the cortisol secretion was increased (P > 0.05), but melatonin was further inhibited after L-DOPAtreatment at 6:00 (P < 0.01). Conclusions: In the circadian system of advanced PD rat models, circadian dysfunction is not only contributed by the degeneration of the disease itself but also long-term L-DOPA therapy may further aggravate it.