Inhibition of adenylyl cyclase type 5 prevents L-DOPA-induced dyskinesia in an animal model of Parkinson's disease

Abstract

The dopamine precursor L-3,4-dihydroxyphenylalanine (L-DOPA) is widely used as a therapeutic choice for the treatment of patients with Parkinson's disease. However, the long-term use of L-DOPA leads to the development of debilitating involuntary movements, called L-DOPA-induced dyskinesia (LID). The cAMP/protein kinase A (PKA) signaling in the striatum is known to play a role in LID. However, from among the nine known adenylyl cyclases (ACs) present in the striatum, the AC that mediates LID remains unknown. To address this issue, we prepared an animal model with unilateral 6-hydroxydopamine lesions in the substantia nigra in wild-type and AC5-knock-out (KO) mice, and examined behavioral responses to short-term or long-term treatment with L-DOPA. Compared with the behavioral responses of wild-type mice, LID was profoundly reduced in AC5-KO mice. The behavioral protection of long-term treatment with L-DOPA in AC5-KO mice was preceded by a decrease in the phosphorylation levels of PKA substrates ERK (extracellular signal-regulated kinase) 1/2, MSK1 (mitogen- and stress-activated protein kinase 1), and histone H3, levels of which were all increased in the lesioned striatum of wild-type mice. Consistently, FosB/ΔFosB expression, which was induced by long-term L-DOPA treatment in the lesioned striatum, was also decreased in AC5-KO mice. Moreover, suppression of AC5 in the dorsal striatum with lentivirus-shRNA-AC5 was sufficient to attenuate LID, suggesting that the AC5-regulated signaling cascade in the striatum mediates LID. These results identify the AC5/cAMP system in the dorsal striatum as a therapeutic target for the treatment of LID in patients with Parkinson's disease.

Keywords: Parkinson's disease; adenylyl cyclase; dyskinesia; l-DOPA.

Figure 1.

Effects of 6-OHDA lesions in AC5^+/+ and AC5^−/− mice. A, TH immunoreactivity on the 6-OHDA-lesioned and unlesioned sides in the striatum and SNc of AC5^+/+ and AC5^−/− mice. Scale bar, 500 μm. B, Percentage loss of SNc TH-positive cells on the lesioned side compared with the unlesioned side of AC5^+/+ and AC5^−/− mice (n = 13–14; Student's t test, p = 0.9604, t = 0.4655). C, Forelimb grip strength (g) of AC5^+/+ and AC5^−/− mice before and after 6-OHDA administration (n = 13–14). **p < 0.01 versus before and after 6-OHDA lesions. Two-way ANOVA: effect of genotypes, F_(1,25) = 0.26, p = 0.6115; effect of surgery, F_(1,25) = 36.30, p < 0.0001; interaction, F_(1,25) = 0.16, p = 0.6898, followed by Bonferroni test. D, Right forelimb use of AC5^+/+ and AC5^−/− mice in the cylinder test before and after 6-OHDA administration, and after the first treatment of l-DOPA (n = 13–14). **p < 0.01. One-way ANOVA followed by a Tukey post hoc test.

Figure 2.

LID is attenuated in 6-OHDA-lesioned AC5^−/− mice. A, Sum of ALO AIMs that were scored during the 120 min period after the last l-DOPA administration. Comparison between AC5^+/+ (n = 14) and AC5^−/− (n = 13) mice. Student's t test, p < 0.01. B, Sum of LOC AIMs that were scored during the 120 min period after l-DOPA administration in the same animals. Student's t test, p < 0.01. C, Total AIMs (sum of LOC and ALO AIMs) that were scored during the 120 min period after the last l-DOPA administration. Student's t test, p < 0.01. D, Time course of total AIMs that were scored every 20 min over a period of 120 min after the last l-DOPA administration (n = 13–14). Two-way ANOVA: effect of genotypes, F_(1,125) = 60.43, p < 0.0001; effect of time, F_(5,125) = 19.26, p < 0.0001; interaction, F_(5,125) = 13.25, p < 0.0001. E, F, Simple linear regression analysis showing the absence of a correlation between the depletion of TH and AIM scores in 6-OHDA-lesioned AC5^−/− mice, although TH immunoreactivity in the lesioned SNc was reduced by >80% (AC5^+/+ mice: r = 0.184, p = 0.1259; AC5^−/− mice: r = 0.007, p = 0.7791).

Figure 3.

6-OHDA-induced signaling changes were significantly reduced in the 6-OHDA-lesioned dorsal striatum of AC5^−/− mice. A, Phospho-PKA substrate, pERK1/2-, pACH3, and pMSK1 immunoreactivity in the 6-OHDA-lesioned dorsal striata of AC5^+/+ and AC5^−/− mice 30 min after the last injection of l-DOPA/benserazide. B, Number of phospho-PKA substrate-positive cells in the dorsolateral striata of AC5^+/+ and AC5^−/− mice (n = 13–14). Two-way ANOVA: effect of genotypes, F_(1,50) = 17.36, p < 0.0001; effect of the lesion, F_(1,50) = 39.17, p < 0.0001; interaction, F_(1,50) = 3.92, p > 0.05. Post hoc comparison (with Bonferroni test): **p < 0.01, 6-OHDA-lesioned versus unlesioned striata; ##p < 0.01, AC5^+/+ versus AC5^−/− mice. Scale bar, 100 μm. C, Number of phospho-ERK1/2-positive cells in the dorsolateral striata of AC5^+/+ and AC5^−/− mice (n = 13–14). Two-way ANOVA: effect of genotypes, F_(1,50) = 27.73, p < 0.001; effect of the lesion, F_(1,50) = 39.73, p < 0.001; interaction, F_(1,50) = 8.65, p < 0.01. Post hoc comparison (with Bonferroni test): **p < 0.01, 6-OHDA-lesioned versus unlesioned; ^##p < 0.01, AC5^+/+ versus AC5^−/−. D, Number of phospho-MSK1-positive cells in the dorsolateral striata of AC5^+/+ and AC5^−/− mice (n = 13–14). Two-way ANOVA: effect of genotypes, F_(1,50) = 1.68, p = 0.200; effect of the lesion, F_(1,50) = 4.05, p < 0.05; interaction, F_(1,50) = 5.63, p < 0.05. Post hoc comparison (with Bonferroni test): *p < 0.05, 6-OHDA-lesioned versus unlesioned striata; #p < 0.01, AC5^+/+ versus AC5^−/− mice. E, Number of phospho-ACH3-positive cells in the dorsolateral striata of AC5^+/+ and AC5^−/− mice (n = 13–14). Two-way ANOVA: effect of genotypes, F_(1,48) = 12.64, p < 0.0001; effect of the lesion, F_(1,48) = 18.68, p < 0.0001; interaction, F_(1,48) = 21.26, p < 0.0001. Post hoc comparison (with Bonferroni test): **p < 0.01, 6-OHDA-lesioned versus unlesioned striata; ##p < 0.01, AC5^+/+ versus AC5^−/− mice. F, G, Western blots showing the levels of p-Ser845-GluA1 (F) and p-Thr34-DARPP32 (G) in 6-OHDA-lesioned and unlesioned dorsal striata of AC5^+/+ and AC5^−/− mice 30 min after the last injection of l-DOPA/benserazide (n = 9–12). Two-way ANOVA with Bonferroni test: **p < 0.01, 6-OHDA-lesioned versus unlesioned striata; #p < 0.05 and ##p < 0.01, AC5^+/+ versus AC5^−/− mice. A significant interaction between genotype and treatment (F, F_(1,40) = 7.76, p < 0.01; G, F_(1,34) = 4.65, p = 0.0383). H, I, Western blots showing the levels of p-Thr202/Tyr204-ERK1 (H) and p-Thr202/Tyr204-ERK2 (I) in 6-OHDA-lesioned and unlesioned dorsal striata of AC5^+/+ and AC5^−/− mice 30 min after the last injection of l-DOPA/benserazide (n = 10–12). Two-way ANOVA with Bonferroni test: *p < 0.05, 6-OHDA-lesioned versus unlesioned striata. H, pERK1: effect of lesion, F_(1,40) = 11.45, p < 0.01; effect of genotype, F_(1,40) = 0.1153, p > 0.05; interaction, F_(1,40) = 0.57; p > 0.05. I, pERK2: effect of lesion, F_(1,40) = 14.37, p < 0.001; effect of genotype, F_(1,40) = 2.80, p > 0.05; interaction, F_(1,40) = 0.45; p > 0.05.

Figure 4.

cAMP/PKA and ERK signaling in short-term l-DOPA treatment. A–D, The number of cells positive to pPKA substrates (A), p-ERK1/2 (B), pMSK1 (C), and pACH3 (D) neurons in the lesioned and unlesioned dorsolateral striata of AC5^+/+ and AC5^−/− mice (n = 6–7). **p < 0.01, 6-OHDA-lesioned versus unlesioned striata; ##p < 0.01, AC5^+/+ versus AC5^−/− mice. A significant interaction was found between genotype and treatment (A, F_(1,22) = 35.81, p < 0.001; B, F_(1,22) = 9.79, p < 0.01; C, F_(1,22) = 5.86, p < 0.05; D, F_(1,22) = 26.16, p < 0.0001).

Figure 5.

Reduction of FosB/ΔFosB-positive cells in the 6-OHDA-lesioned dorsal striatum after long-term treatment with l-DOPA in AC5^−/− mice. A, Expression levels of FosB/ΔFosB immunoreactivity in the 6-OHDA-lesioned dorsal striata of AC5^+/+ and AC5^−/− mice 30 min after the first treatment (short-term treatment) and the last treatment (long-term treatment, 11 d) with l-DOPA. FosB/ΔFosB-positive cells were counted in the lesioned and unlesioned striata of AC5^+/+ and AC5^−/− mice. B, The number of FosB/ΔFosB-positive cells after the short-term administration of l-DOPA in the unlesioned and lesioned striata of AC5^+/+ and AC5^−/− mice (n = 7–5). Two-way ANOVA with Bonferroni test: effect of genotypes, F_(1,20) = 15.62, p = 0.0008; effect of the lesion, F_(1,20) = 24.44, p < 0.0001; interaction, F_(1,20) = 0.009, p = 0.923. C, The number of FosB/ΔFosB-positive cells after long-term l-DOPA treatment in the unlesioned and lesioned striata of AC5^+/+ and AC5^−/− mice (n = 13–14). Two-way ANOVA with Bonferroni test: effect of genotypes, F_(1,50) = 35.64, p < 0.0001; effect of the lesion, F_(1,50) = 132.51, p < 0.0001; interaction, F_(1,50) = 45.17, p < 0.0001: *p < 0.05 and **p < 0.01, 6-OHDA-lesioned versus unlesioned striata; ##p < 0.01, AC5^+/+ versus AC5^−/− mice. Scale bar, 100 μm.

Figure 6.

Suppression of AC5 in the dorsal striatum using lenti-AC5 shRNA decreased LID. A, Experimental design for treatment of 6-OHDA lesion, infusion of lenti-AC5 shRNA, following l-DOPA injection and behavioral tests. Lentivirus was unilaterally infused into 6-OHDA-lesioned striatum. B, Photomicrograph showing GFP fluorescence in the dorsal striatum injected with lenti-GFP. cc, Corpus callosum. C, d-AMPH-induced ipsilateral rotations in the two groups (Student's t test, p > 0.05). D, Right forelimb use levels in the cylinder test in mice injected with lenti-GFP or lenti-AC5 shRNA before and after the first treatment with l-DOPA. One-way ANOVA followed by Tukey post hoc test, **p < 0.01. E–G, ALO AIMs (E), LOC AIMs (F), and total AIM score (G) in mice injected with 6-OHDA followed by injection with lenti-GFP or lenti-AC5 shRNA (n = 5 each). Student's t test, *p < 0.05. H, Percentage loss of TH-positive cells in the lesioned SNc in mice injected with 6-OHDA followed by injection with lenti-GFP and lenti-AC5 shRNA. I, Real-time RT-PCR data showing the expression levels of AC5 in the dorsal striatum in wild-type mice injected with lenti-GFP control or lenti-AC5 shRNA. Expression levels were examined 18 d after viral injection (n = 6 each). Student's t test, **p < 0.01.