Involvement of Autophagy in Levodopa-Induced Dyskinesia

Abstract

Background: Autophagy is intensively studied in cancer, metabolic and neurodegenerative diseases, but little is known about its role in pathological conditions linked to altered neurotransmission. We examined the involvement of autophagy in levodopa (l-dopa)-induced dyskinesia, a frequent motor complication developed in response to standard dopamine replacement therapy in parkinsonian patients.

Methods: We used mouse and non-human primate models of Parkinson's disease to examine changes in autophagy associated with chronic l-dopa administration and to establish a causative link between impaired autophagy and dyskinesia.

Results: We found that l-dopa-induced dyskinesia is associated with accumulation of the autophagy-specific substrate p62, a marker of autophagy deficiency. Increased p62 was observed in a subset of projection neurons located in the striatum and depended on l-dopa-mediated activation of dopamine D1 receptors, and mammalian target of rapamycin. Inhibition of mammalian target of rapamycin complex 1 with rapamycin counteracted the impairment of autophagy produced by l-dopa, and reduced dyskinesia. The anti-dyskinetic effect of rapamycin was lost when autophagy was constitutively suppressed in D1 receptor-expressing striatal neurons, through inactivation of the autophagy-related gene protein 7.

Conclusions: These findings indicate that augmented responsiveness at D1 receptors leads to dysregulated autophagy, and results in the emergence of l-dopa-induced dyskinesia. They further suggest the enhancement of autophagy as a therapeutic strategy against dyskinesia. © 2021 The Authors. Movement Disorders published by Wiley Periodicals LLC on behalf of International Parkinson and Movement Disorder Society.

Keywords: l-dopa; Parkinson's disease; autophagy; p62; striatum.

FIG. 1

Chronic administration of levodopa (l‐dopa) reduces autophagy in the striata of parkinsonian mice and non‐human primates. C57BL/6 mice with a unilateral 6‐hydroxydopamine (6‐OHDA) lesion were treated with l‐dopa (10 mg/kg) as described, and the levels of p62 were measured by Western blotting in the striata contralateral (control) or ipsilateral (6‐OHDA) to the lesion. (A) Mice were divided in four groups (n = 4–10) and treated with vehicle, or with l‐dopa for 1, 4, and 9 days, and p62 was measured 4 hours after the last injection. **P < 0.01 versus control, Welch two‐sample t‐test. (B) Four groups (n = 6–7) of mice were treated with l‐dopa for 9 days and p62 was measured 2, 4, 8, or 24 hours after the last injection. Two‐way ANOVA showed a significant effect of 6‐OHDA lesion (F_1,45 = 86.01, P < 0.001) and a significant effect of time (F_3,45 = 8.16, P < 0.001), but no lesion × time interaction (F_3,45 = 1.61, P = 0.199). *P < 0.05 and ***P < 0.001 versus control, Tukey post hoc test. (C) Two groups (n = 6–9) of mice were treated with vehicle or l‐dopa for 9 days and mRNA for p62 was measured 4 hours after the last injection. (D) p62 was measured in the striata of three experimental groups (n = 5) of non‐human primates: normal (control), parkinsonian (1‐methyl‐4‐phenyl‐1,2,3,6‐tetrahydropyridine; MPTP), and parkinsonian treated with l‐dopa (MPTP+l‐dopa). *P < 0.05 versus control group, one‐way ANOVA and Tukey's multiple comparison test.

FIG. 2

The impairment of autophagy induced by levodopa (l‐dopa) is caused by dopamine D1 receptor (D1R)‐mediated activation of mTOR complex 1 (mTORC1) in the striatal projection neurons of the direct pathway (dSPN). (A) Mice (n = 8–9 per group) with a unilateral 6‐hydroxydopamine (6‐OHDA) lesion received chronic (9 days) pretreatment with l‐dopa (10 mg/kg) alone (vehicle), or in combination with SCH23390 or raclopride (0.125 mg/kg and 0.25 mg/kg, respectively, administered 15 minutes before l‐dopa). The levels of p62 were measured by Western blotting 4 hours after the last drug administration in the striata contralateral (control) or ipsilateral (6‐OHDA) to the lesion. Two‐way ANOVA showed a significant effect of pretreatment (F_2,46 = 22.59, P < 0.001), 6‐OHDA lesion (F_1,46 = 70.37, P < 0.001), and pretreatment × 6‐OHDA lesion interaction (F_2,46 = 7.44, P = 0.01). ***P < 0.001 versus control and †P < 0.001 versus 6‐OHDA pretreated with vehicle or raclopride, Tukey's post hoc test. (B) Left panels: immunofluorescence analysis of p62 and enhanced green fluorescent protein (EGFP) in a Drd1a‐EGFP mouse with a unilateral 6‐OHDA lesion treated for 9 days with l‐dopa (10 mg/kg) and perfused 4 hours after the last injection. DARPP‐32 immunoreactivity indicates that accumulation of p62 occurs in striatal projection neurons. Note the increased levels of p62 in the striatal projection neurons of the 6‐OHDA lesion striatum (arrowheads). Right panel: summary of data from 5 Drd1a‐EGFP mice showing the number of EGFP‐positive dSPN and EGFP‐negative iSPN with high levels of p62 immunoreactivity. Two‐way ANOVA showed significant effect of cell type (F_1,16 = 52.03, P < 0.001), 6‐OHDA lesion (F_1,16 = 100.8, P < 0.001), and cell type × 6‐OHDA lesion (F_1,16 = 52.03, P < 0.001). ***P < 0.001 versus control, Sidak's multiple comparison test. (C) Two groups (n = 9) of mice with a unilateral 6‐OHDA lesion were treated for 9 days with l‐dopa (10 mg/kg) alone (vehicle) or l‐dopa plus SCH23390 (0.125 mg/kg), and the levels of total and phosphorylated (S757) Ulk1 (P‐Ulk1) were determined by Western blotting 4 hours after the last drug administration in the striata contralateral (control) or ipsilateral (6‐OHDA) to the lesion. Two‐way ANOVA showed a significant SCH23390 pretreatment × 6‐OHDA lesion interaction (F_1,64 = 5.29, P < 0.05). ***P < 0.001 versus control and †P < 0.001 versus 6‐OHDA pretreated with vehicle, Tukey's post hoc test. (D, E) Mice with a unilateral 6‐OHDA lesion were treated for 9 days with l‐dopa (10 mg/kg) alone or in combination with rapamycin (2 or 5 mg/kg, administered 45 minutes before l‐dopa) and the levels of p62 (D, n = 6 per group) and total or P‐Ulk1 (E, n = 6–11 per group) were determined by Western blotting 4 hours after the last drug administration in the striata contralateral (control) or ipsilateral (6‐OHDA) to the lesion. (D) Two‐way ANOVA showed a significant effect of treatment (F_2,30 = 11.37, P < 0.001), 6‐OHDA lesion (F_1,30 = 68.86, P < 0.001), and no significant treatment × 6‐OHDA lesion interaction. ***P < 0.001 and **P < 0.01 versus respective control, †P < 0.05 and ††P < 0.01 versus 6‐OHDA/vehicle; Tukey's post hoc test. (E) Two‐way ANOVA showed a significant effect of treatment (F_1,30 = 12.57, P < 0.001), 6‐OHDA lesion (F_1,30 = 5.24, P < 0.05), and no significant treatment × 6‐OHDA lesion interaction. *P < 0.05 versus control, ††P < 0.01 versus 6‐OHDA/vehicle, Tukey's post hoc test.

FIG. 3

The autophagy promoting effect of rapamycin is occluded in Atg7 ^F/F ;Drd1a‐Cre ^+/− mice. (A) Atg7 ^F/F ;Drd1a‐Cre ^+/− and Atg7 ^F/F mice were injected unilaterally with 6‐hydroxydopamine (6‐OHDA) and the levels of p62 was measured by Western blotting in the striata ipsilateral (control) and contralateral (6‐OHDA) to the lesion. Note the large accumulation of p62 in Atg7 ^F/F ;Drd1a‐Cre ^+/− mice indicative of impaired autophagy in the striatal projection neurons of the direct pathway (dSPN). 6‐OHDA did not affect striatal p62 levels (control 100.0 ± 6.3 vs. 6‐OHDA 100.7 ± 5.4 in Atg7 ^F/F mice and control 2400 ± 137.1 vs. 6‐OHDA 2446 ± 220.2 in Atg7 ^F/F ;Drd1a‐Cre ^+/− mice) (n = 5–6 per group). (B, C) Atg7 ^F/F ;Drd1a‐Cre ^+/− and Atg7 ^F/F mice with a 6‐OHDA lesion were treated with vehicle, levodopa (l‐dopa) (10 mg/kg), or l‐dopa plus rapamycin (5 mg/kg, administered 45 minutes before l‐dopa) for 9 days and the levels of p62 (B), total S6 and S6 phosphorylated at S240/244 (C) were measured by Western blotting 8 hours after the last injection. ***P < 0.001 and **P < 0.01 versus vehicle, one‐way ANOVA followed by Dunnett's multiple comparison test (n = 4–9 per group).

FIG. 4

The anti‐dyskinetic action of rapamycin is prevented in Atg7 ^F/F ;Drd1a‐Cre ^+/− mice. Atg7 ^F/F ;Drd1a‐Cre ^+/− and Atg7 ^F/F mice (n = 4–6 per group) with a unilateral 6‐hydroxydopamine (6‐OHDA) lesion were treated for 9 consecutive days with levodopa (l‐dopa) (10 mg/kg) plus vehicle or l‐dopa plus rapamycin (5 mg/kg). Abnormal involuntary movements (AIMs) were assessed for 1 minute every 20 minutes, starting immediately after the last injection. (A) Cumulative effect on axial, limb, orofacial (ALO) AIMs during the entire observation period (120 minutes). Two‐way ANOVA showed a significant effect of treatment (F_1,7 = 26.62, P < 0.01), no effect of genotype (F_1,9 = 0.01, P > 0.9), and a significant effect of genotype × treatment interaction (F_1,7 = 8.28, P < 0.05). **P < 0.01 versus l‐dopa/vehicle, Sidak's multiple comparison test. (B) Sum of locomotive AIMs scored during the 120‐minute observation period. (C, D) Time course of total AIMs scored in Atg7 ^F/F (C) and Atg7 ^F/F ;Drd1a‐Cre ^+/− mice (D) every 20 minutes during the 120‐minute observation period. (C) Two‐way ANOVA indicated a significant effect of rapamycin treatment (F_1,9 = 69.2; P < 0.001), time (F_3,31 = 108.6; P < 0.001), and treatment × time interaction (F_5,45 = 6.4; P < 0.001). (D) Two‐way ANOVA indicated a significant effect of time (F_2,17 = 25.6; P < 0.001).