PINK1 and Parkin complementarily protect dopaminergic neurons in vertebrates

Abstract

Parkinson's disease (PD) is a common neurodegenerative disorder characterized by selective dopaminergic cell loss in the substantia nigra, but its pathogenesis remains unclear. The recessively inherited familial PD genes PARK2 and PARK6 have been attributed to mutations in the Parkin and PTEN-induced kinase 1 (PINK1) genes, respectively. Recent reports suggest that PINK1 works upstream of Parkin in the same pathway to regulate mitochondrial dynamics and/or conduct autophagic clearance of damaged mitochondria. This phenomenon is preserved from Drosophila to human cell lines but has not been demonstrated in a vertebrate animal model in vivo. Here, we developed a medaka fish (Oryzias latipes) model that is deficient in Pink1 and Parkin. We found that despite the lack of a conspicuous phenotype in single mutants for Pink1 or Parkin, medaka that are deficient in both genes developed phenotypes similar to that of human PD: late-onset locomotor dysfunction, a decrease in dopamine levels and a selective degeneration of dopaminergic neurons. Further analysis also revealed defects in mitochondrial enzymatic activity as well as cell death. Consistently, PINK1 and Parkin double-deficient MEF showed a further decrease in mitochondrial membrane potential and mitochondrial complex I activity as well as apoptosis compared with single-deficient MEF. Interestingly, these mitochondrial abnormalities in Parkin-deficient MEF were compensated by exogenous PINK1, but not by disease-related mutants. These results suggest that PINK1 and Parkin work in a complementary way to protect dopaminergic neurons by maintaining mitochondrial function in vertebrates.

Figure 1.

Body weight and spontaneous swimming movement of medaka. (A) Body weight (g). (B) Total swimming distance (cm). (C) Duration of swimming movement (s). (D) Swimming velocity (cm/s). Pink1+: Pink1 wild-type medaka, Pink1−: Pink1-deficient medaka, Parkin+: Parkin wild-type medaka, Parkin−: Parkin-deficient medaka. *P < 0.05, **P < 0.01, ***P < 0.001. Error bars represent SEM. n = 12 (A), and n = 15 for each group (B–D).

Figure 2.

Number of TH+ neurons in the middle diencephalon and medulla oblongata. (A) Western blot analysis of TH and TPH proteins at 12 months. (B) Ratio of TH/β-actin proteins and TPH/β-actin proteins. (C) Representative photographs of middle diencephalic dopaminergic neurons in Pink1+/Parkin+, Pink1−/Parkin+, Pink1+/Parkin− and Pink1−/Parkin−. (D) Number of TH+ neurons in the middle diencephalon at 4 and 12 months (top). Number of TH+ neurons in the medulla oblongata at 12 months (bottom). *P < 0.05, **P < 0.01, ***: P < 0.001. Error bars represent SEM. n = 8 for each group (B and D).

Figure 3.

Dopamine, noradrenaline and serotonin levels in medaka whole brain. (A) Amount of dopamine. (B) Amount of noradrenaline. (C) Amount of serotonin. All values are expressed as a percentage of the amount (ng) per protein weight (mg) for Pink1+/Parkin+. *: P < 0.05, **: P < 0.01, ***: P < 0.001. Error bars represent SEM. n = 12 for each group (A–C).

Figure 4.

Mitochondrial complex I–IV activity, transmission electron microscopic images and TUNEL assay of medaka brains. (A) Mitochondrial complex I–activities. Two micrograms of medaka brain lysate was used for complex I and II assays, and 10 µg for complex III and IV assays. n = 8 for each group. (B) B1: Elongated mitochondria in single- and double-deficient medaka. B2: Damaged mitochondria in Pink1−/Parkin− brains with an abnormal cristae structure. In the left panel, images on the right are magnified from areas indicated on the left. n = 4 (brains) for each group, 10 different regions per brain were used for the statistics. (C) Inclusion bodies containing filamentous materials in Pink1−/Parkin− brains. (D) TUNEL assay in Pink1−/Parkin− brains. White arrows indicate TH and TUNEL double-positive neurons in the middle diencephalon. *P < 0.05, **P < 0.01, ***P < 0.001. Error bars represent SEM.

Figure 5.

Mitochondrial function and apoptosis in MEFs. (A) RT–PCR showing PINK1 knockdown in WT and Parkin^−/− MEF (left) and mitochondrial complex I activity (right). (B) Mitochondrial complex I activity in double-deficient MEFs. (C) TMRE fluorescence. (D) Annexin V staining. (E) Caspase 3 and cleaved caspase 3 immunoblotting. (F) Representative photographs of ssDNA staining in MEF (left). Number of ssDNA-positive cells over the total number of cells (right). More than 160 cells were analyzed per group. *P < 0.05, **P < 0.01, ***P < 0.005, ****P < 0.001. Error bars represent SD (A—C) and SEM (D). n = 3 for (A and B), n ≥ 5 for (C).

Figure 6.

PINK1 rescue of Parkin-deficient mitochondrial phenotypes. (A) Immunoblotting of overexpressed Parkin, PINK1, L347P PINK1 and D362A; D384A PINK1 in Parkin-deficient MEF. (B) Mitochondrial complex I activity. (C) TMRE fluorescence. *P < 0.05. Error bars represent SD. n = 3 for (B) and n ≥ 5 for (C).

Figure 7.

Proposed roles of Parkin and PINK1 in protecting mitochondrial function in Drosophila and vertebrates.