PINK1 stabilized by mitochondrial depolarization recruits Parkin to damaged mitochondria and activates latent Parkin for mitophagy

Abstract

Parkinson's disease (PD) is a prevalent neurodegenerative disorder. Recent identification of genes linked to familial forms of PD such as Parkin and PINK1 (PTEN-induced putative kinase 1) has revealed that ubiquitylation and mitochondrial integrity are key factors in disease pathogenesis. However, the exact mechanism underlying the functional interplay between Parkin-catalyzed ubiquitylation and PINK1-regulated mitochondrial quality control remains an enigma. In this study, we show that PINK1 is rapidly and constitutively degraded under steady-state conditions in a mitochondrial membrane potential-dependent manner and that a loss in mitochondrial membrane potential stabilizes PINK1 mitochondrial accumulation. Furthermore, PINK1 recruits Parkin from the cytoplasm to mitochondria with low membrane potential to initiate the autophagic degradation of damaged mitochondria. Interestingly, the ubiquitin ligase activity of Parkin is repressed in the cytoplasm under steady-state conditions; however, PINK1-dependent mitochondrial localization liberates the latent enzymatic activity of Parkin. Some pathogenic mutations of PINK1 and Parkin interfere with the aforementioned events, suggesting an etiological importance. These results provide crucial insight into the pathogenic mechanisms of PD.

Figure 1.

Mitochondrial localization of Parkin is etiologically important. (A) HeLa cells expressing HA-Parkin were treated with CCCP or DMSO (control) and then immunostained with the indicated antibodies. (B) HeLa cells stably expressing HA-Parkin or intact SH-SY5Y cells were treated with CCCP or DMSO and subjected to fractionation experiments. I, C, and M indicate input, cytosol-rich supernatant, and mitochondria-rich membrane pellet, respectively. (C) Schematic diagram of disease-relevant mutants of Parkin used in this study. IBR, in between RING; Ubl, ubiquitin like. (D) Polarized mitochondria stained with MitoTracker red (red arrowheads) were not labeled by Parkin. In contrast, damaged mitochondria marked by Parkin (green arrowheads) were not stained with MitoTracker red. (E) HeLa cells expressing HA-Parkin with various pathogenic mutations were treated with CCCP, followed by immunocytochemistry. (A, D, and E) Higher magnification views of the boxed areas are shown in the insets. (F) Parkin colocalization with mitochondria was analyzed in >100 cells per mutation. Example figures indicative of robust colocalization (counted as 1), weak colocalization (counted as 0.5), and no colocalization (counted as 0) are shown. Error bars represent the mean ± SD values of at least three experiments. Bars: (A, E, and F) 10 µm; (D) 30 µm.

Figure 2.

Parkin exerts E3 activity only when the mitochondrial membrane potential is decreased. (A) HeLa cells expressing wild-type Parkin or E3-inactivating mutations were treated with CCCP and then immunostained with the indicated antibodies. When E3-inactivating mutations were introduced into Parkin, the mitochondrial ubiquitylation signal disappeared. (B and C) HeLa cells expressing HA-Parkin (B) or expressing both Mt-GFP and HA-Parkin (C) were treated with CCCP or DMSO (control) and then immunostained with the indicated antibodies. (D) Localization of GFP-Parkin to the mitochondria after CCCP treatment. (A–D) Higher magnification views of the boxed areas are shown in the insets. (E) HeLa cell lysates expressing GFP-Parkin were immunoprecipitated by anti-GFP antibody, followed by immunoblotting with the indicated antibodies. (F) Straight immunoblotting of HA- and GFP-Parkin in the absence or presence of CCCP. Note the slower migrating ladders derived from ubiquitylation (Ub) in only the GFP-Parkin with CCCP lane. (G) GFP-Parkin–expressing HeLa cells with various pathogenic mutations (Fig. 1 C) were treated with CCCP and subjected to immunoblotting. Asterisks show ubiquitylation of GFP-Parkin. Vertical black lines indicate that intervening lanes have been spliced out. IB, immunoblot; IP, immunoprecipitation. Bars, 10 µm.

Figure 3.

PINK1 is constitutively degraded in a mitochondrial membrane potential–dependent manner and localizes to depolarized mitochondria. (A) The number of HeLa cells with N-terminal– or C-terminal–tagged PINK1 localized to the mitochondria was counted in >100 cells. (B and C) Exogenous nontagged PINK1 (B) or endogenous PINK1 (C) in HeLa cells was immunostained with the indicated antibodies. (D) The number of HeLa cells with endogenous PINK1 localized to the mitochondria was counted as in A. (A and D) Error bars represent the mean ± SD values of least three experiments. (E) Endogenous PINK1 gradually accumulated after CCCP treatment. The first through the fifth lanes show endogenous PINK1, and the sixth lane shows overexpressed untagged PINK1. Note that the asterisk indicates a cross-reacting band because it was not affected by overproduction of untagged PINK1. (F) Subcellular fractionation of endogenous PINK1. Intact SH-SY5Y cells were treated with CCCP or DMSO and subjected to fractionation experiments (same sample as Fig. 1 B). I, C, and M indicate input, cytosol-rich supernatant, and the mitochondria-rich membrane pellet, respectively. (G) HeLa cells were treated with CCCP and cycloheximide as depicted, followed by immunoblotting with the indicated antibodies. LDH, lactate dehydrogenase. (F and G) Asterisks indicate a cross-reacting band. (H) N-terminal 34 aa of PINK1 recruited GFP to the mitochondria both in the absence and presence of CCCP. The top panel shows control HeLa cells expressing only GFP. (B, C, and H) Higher magnification views of the boxed areas are shown in the insets. Bars, 10 µm.

Figure 4.

PINK1 recruits cytoplasmic Parkin to damaged mitochondria. (A) PINK1 knockout (PINK1^−/−) or control (PINK1^+/+) MEFs were transfected with HA-Parkin, treated with CCCP, and subjected to immunocytochemistry with the indicated antibodies. Higher magnification views of the boxed areas are shown in the insets. (B) The number of MEFs with Parkin localized to the mitochondria was counted as in Fig. 3 A. (C and D) Neither activation of Parkin nor mitochondrial degradation was observed in PINK1^−/− MEFs. MEFs stably expressing GFP-Parkin were treated with CCCP for 4 h and subjected to immunoblotting (C) or for 24 h, followed by immunocytochemistry (D). IB, immunoblot; Ub, ubiquitylation. (E) The number of MEFs without anti-Tom20 antibody–detectable mitochondria was counted as in Fig. 3 A. In the example figure (left), blue arrowheads indicate cells without anti-Tom20 antibody–detectable mitochondria, and red arrowheads indicate cells harboring anti-Tom20 antibody–detectable mitochondria. (B and E) Error bars represent the mean ± SD values of least three experiments. Bars: (A) 10 µm; (D) 50 µm; (E) 150 µm.

Figure 5.

Kinase activity and mitochondrial targeting of PINK1 is imperative for mitochondrial localization of Parkin. (A) Schematic depiction of pathogenic and deletion mutants of PINK1 used in this study. MTS, mitochondria-targeting sequence; TMD, transmembrane domain. (B) Subcellular localization of Parkin in PINK1^−/− cells complemented by various pathogenic and deletion mutants of PINK1-Flag. Cells were treated with CCCP. Higher magnification views of the boxed areas are shown in the insets. (C) The number of cells with Parkin-positive mitochondria was counted as in Fig. 3 A. Error bars represent the mean ± SD values of least three experiments. (D) PINK1^−/− MEFs complemented by various PINK1 mutants were treated with CCCP and subjected to immunoblotting using anti-Parkin or anti-Flag (tag of PINK1) antibodies. IB, immunoblot; Ub, ubiquitylation. Bars, 10 µm.