Parkin Deficiency Impairs ER-Mitochondria Associations and calcium homeostasis via IP3R-Grp75-VDAC1 Complex

Abstract

Disruption of mitochondria-associated endoplasmic reticulum membranes (MAMs) and calcium homeostasis has been implicated in the pathogenesis of Parkinson's disease (PD). Parkin, a PD-associated E3 ubiquitin ligase, has been shown to regulate MAM integrity and calcium dynamics. However, the mechanisms of Parkin recruitment and its substrate specificity have not been well understood. This investigation has demonstrated that loss of Parkin enhances ER-mitochondria associations and leads to excessive calcium flux in MAM, resulting in abnormal mitochondrial permeability transition pore (mPTP) opening and decreased cell viability. Further, Parkin physically interacts with IP3R-Grp75-VDAC1 complex at ER-mitochondria contact sites, where it is recruited by IP3R-mediated calcium flux and mitophagy. More importantly, Parkin deficiency leads to the accumulation of IP3R levels, particularly in MAM region. In addition, Parkin fine-tunes the stability of the complex and ubiquitinates IP3R for degradation via the ubiquitin-proteasomal system, ensuring suitable calcium transfer. Taken together, our study reveals a novel role of Parkin in regulating ER-mitochondria contacts, providing insights into PD pathogenesis and potential therapeutic strategies targeting MAMs.

Keywords: IP3R; Parkin; calcium; mitochondria-associated ER membrane; ubiquitination.

Figure 1

Loss of parkin increases endoplasmic reticulum (ER)-mitochondria contact sites. (A) Confocal microscopy was used for live cell imaging of both control cells (upper) and Parkin KO M17 cells (bottom). ER was stained with an ER-Tracker (green) and mitochondria were stained with MitoTracker (red) (Scale bar, 20 μm). Quantification of ER-mitochondria association was performed using ImageJ (bar graphs on the right). (B) Confocal microscopy images of proximity ligation assays (PLA) of ER-mitochondria association. PLA signals (red dots) were determined by interaction between IP3R and VDAC1 in control cells (upper) and Parkin KO M17 cells (bottom) (Scale bar, 20 μm). PLA red fluorescent dots were present as the number of positive interactions per nucleus (bar graphs on the right). (C) Electron micrographs of mitochondria connected to the ER (pseudocolored green) in control and Parkin KO M17 cells (Scale bar, 1 μm). (D-F) Ultrastructural analysis of ER-mitochondria association. Quantitative analysis of the percentage of the mitochondria-associated endoplasmic reticulum membrane (MAM) to mitochondria (D), type of MAMs apposition (F) and average length (I) of ER-mitochondria association in control cells (n = 7) and Parkin KO M17 cells (n = 8). (G) Electron micrographs of mitochondria connected to the ER (pseudocolored green) in neurons from wild-type (WT) and Parkin KO mouse brains. (H-J) Ultrastructural analysis of ER-mitochondria association. Quantitative analysis of the percentage of the MAM to mitochondria (H), type of MAM apposition (I), and average length (J) of ER-mitochondria association (n = 5 mice per group, with 4 neurons per mouse). Data are expressed as means ± SEM based on three independent experiments. Data were analyzed using a two-tailed unpaired Student's t-test. *P < 0.05; **P<0.01; ***P < 0.001.

Figure 2

Parkin knockdown induces mPTP opening via MAM-mediated Ca²⁺ influx. (A) Mitochondrial calcium levels were evaluated using Rhod-2 staining and quantified by mean fluorescence intensity (MFI) through flow cytometry comparisons. (B) The mPTP functionality was gauged with calcein-AM staining in conjunction with CoCl2. (C and D) Measurement of calcium modulation in mitochondria (C) and the ER (D) in control cells (black) and Parkin KO M17 cells (red) was conducted using confocal microscopy. Thapsigargin (TG) was used to initiate calcium release. The right bar graphs indicate the quantification of maximal mitochondrial (C) or ER (D) calcium peak fluorescence during TG treatment (n = 68-83 cells). (E) Treatment with TG (2.5μm, 24 hours) in both genotypes was followed by an analysis of mPTP functionality using calcein-AM fluorescence. (F) CCK-8 assay was used to examine cell viability across four groups after treatment with TG (2.5μm, 24 hours). (G) Apoptosis was analyzed by flow cytometry using Annexin V/PI (propidium iodide) staining, following treatment with TG (2.5 μM, 24 hours). Data are expressed as means ± SEM based on three independent experiments. Data were analyzed using two-tailed unpaired Student's t-test (A-D) and one-way analysis of variance (ANOVA) with Tukey's multiple comparisons test (E-F). *P < 0.05; **P<0.01; ***P < 0.001.

Figure 3

Parkin resides in the MAM and interacts with the IP3R-Grp75-VDAC1 complex. (A-B) Subcellular fractionation and immunoblot characterization of parkin in the MAM fraction in wild-type C57BL6 mouse brain (A) and parkin-overexpressing M17 cells (B). Cells were fractionated into whole lysates (WL), crude mitochondria (CM), pure mitochondria (PM), ER, and MAM. Three independent experiments were conducted. (C) In situ close association between Parkin/IP3R (left), Parkin/Grp75 (middle), and Parkin/VDAC1 (right) were determined by proximity ligation assay (PLA) in normal M17 cells (Scale bar, 20 μm). (D-F) Immunoblot analysis of IP3R, Grp75, and VDAC1 in the parkin immunoprecipitates of total lysates from parkin-overexpressing M17 cells (D) and ventral midbrain of mouse brain (E). MAM fractions were prepared from the ventral midbrain of the mouse brain (F). Three independent experiments were conducted. (G-H) Blue native-polyacrylamide gel electrophoresis (BN-PAGE) and immunoblot analysis of CM fractions prepared from parkin-overexpressing M17 cells (G) and ventral midbrain of mouse brain (H) showed a large protein complex. The red box in G-H highlights the macrocomplex that contains all four components. Three independent experiments were conducted. (I) Two-dimensional separation and immunoblot analysis of CM fraction from parkin-overexpressing M17 cells showed a large protein complex. The red box in I highlights the macrocomplex that contains all four components. Three independent experiments were conducted.

Figure 4

Calcium-Dependent Localization of Parkin at the MAM. (A) Representative images of PLA targeting IP3R-Parkin interactions after treatment with 2-APB (50 μm, 2 hours) in M17 cells (Scale bar, 20 μm). (B) Representative immunoblot and quantitative analyses of IP3R coimmunoprecipitated with parkin after treatment with 2-APB (50 μm, 2 hours) in overexpressing WT parkin M17 cells. Three independent experiments were conducted and were quantified. (C) BN-PAGE of the complex after treatment with 2-APB (50 µm, 2 hours). Representative immunoblot images were detected by parkin. Three independent experiments were conducted. (D) Immunoblot analysis of MAM proteins in the WL and MAM fractions from control and 2-APB treatment (50 µm, 2 hours). Proteins were normalized to calreticulin. Three independent experiments were conducted and were quantified. Data are expressed as means ± SEM based on three independent experiments. Data were analyzed using a two-tailed unpaired Student's t-test. *P < 0.05; **P<0.01; ***P < 0.001.

Figure 5

Parkin regulates the stability of the IP3R-Grp75-VDAC1 complex. (A) Immunoblot analysis of chosen proteins in control and Parkin KO M17 cells. Control and Parkin KO cells were fractionated into WL and CM. Cytochrome c oxidase subunit IV (COX IV) was used as a loading control for CM proteins. Three independent experiments were conducted and were quantified. (B) Immunoblot analysis of MAM proteins in the WL and MAM fractions from WT and Parkin KO mice (n = 6/group). Proteins were normalized to calreticulin. (C) Immunoblot analyses of time-dependent degradation of IP3R were detected after treatment with cycloheximide (CHX, 100 μg/ml, 0-4 hours) in control and Parkin KO M17 cells. Three independent experiments were conducted and were quantified. (D) BN-PAGE analysis of the macrocomplex detected by IP3R in crude mitochondria from control and Parkin KO M17 cells. SDS-PAGE immunoblot was used to analyze parkin in CM. Data are quantified from three independent experiments. (E) BN-PAGE and immunoblot analysis of the macrocomplex detected by IP3R in the crude mitochondria fraction from mouse brain homogenates (left). Representative immunoblot images detected by parkin (n = 5/group, right). (F) Immunoblot analysis of Grp75/VDAC1 was coimmunoprecipitated with IP3R antibody in control and Parkin KO M17 cells. Three independent experiments were conducted and were quantified. (G) Representative images and quantification of IP3R-VDAC1 PLA signals revealed an association in tyrosine hydroxylase positive (TH+) neurons in the substantia nigra of Parkin KO mice. IP3R-VDAC1 PLA (red) was performed in brain sections and co-stained with TH (green) (n = 5/group), (Scale bar, 20 µm). (H) Immunoblot analysis of Grp75 and VDAC1 was coimmunoprecipitated with IP3R antibody in the brain homogenates of wild-type control and Parkin KO mice (n = 5/group). Data are expressed as means ± SEM. Data were analyzed using two-tailed unpaired Student's t-test; *P < 0.05; **P < 0.01; ***P < 0.001.

Figure 6

Parkin tunes the degradation of IP3R through K48-linked ubiquitination. (A) Immunoblot analyses of IP3R protein showed its reduced stability by Parkin-dependent expression in M17 cells. Three independent experiments were conducted and were quantified. (B) Immunoblot analysis was performed to assess IP3R protein expression in M17 cells overexpressing wild-type (WT) parkin, treated with MG132 (20 µM) or 3-MA (2.5 mM) for 24 hours. Three independent experiments were conducted and were quantified. (C) Representative immunoblot and quantitative analyses of the degradation of IP3R from overexpressing vector, WT parkin, and parkin C431S mutant M17 cells. Three independent experiments were conducted and were quantified. (D) Ubiquitination levels of IP3R were assessed in M17 cells expressing WT or C431S parkin plasmid. Cells were treated with 20 µM of MG132 for 4 hours. Three independent experiments were conducted. (E) Ubiquitination levels of IP3R were assessed in M17 cells expressing wild-type (WT) parkin, HA-Ub, and its mutant plasmids. Cells were treated with 20 µM of MG132 for 4 hours. Three independent experiments were conducted. Data are expressed as means ± SEM based on three independent experiments. Data were analyzed using ANOVA with Tukey's multiple comparisons test (A-C). *P < 0.05, **P < 0.01, ***P < 0.001.

Figure 7

IP3R is ubiquitinated by Parkin upon induction of E3 activities. (A) Immunoblot analyses of time-dependent degradation of IP3R were detected after treatment with carbonyl cyanide m-chlorophenyl hydrazine (CCCP, 20 µm, 0-6 hours) in control and Parkin KO M17 cells. Three independent experiments were conducted and were quantified. (B) Representative immunoblot and quantitative analyses of IP3R degradation after treatment with CCCP (20 µm, 2 hours) and proteasome inhibitor MG-132 (20 µm, 0.5 hour before CCCP treatment) in M17 cells. Three independent experiments were conducted and were quantified. (C) Representative images of PLA targeting IP3R-Parkin interactions after treatment with CCCP (20 µm, 1 hour) in M17 cells (n = 166-199 cells). (D) Representative immunoblot and quantitative analyses of IP3R coimmunoprecipitated with parkin after treatment with CCCP (20 µm, 1 hour) in overexpressing WT parkin M17 cells. Three independent experiments were conducted and were quantified. (E) BN-PAGE of the complex after treatment with CCCP (20 µm, 2 hours). Representative immunoblot images were detected by IP3R (left) and parkin (right). Three independent experiments were conducted. (F) Ubiquitination levels of IP3R after treatment with CCCP (20 µm, 2 hours) and MG-132 (20 µm, 1 hour before CCCP treatment) in M17 cells. Three independent experiments were conducted. Data are expressed as means ± SEM from three independent experiments. Data were analyzed using a two-tailed unpaired Student's t-test (A, C-F) and ANOVA with Tukey's multiple comparisons test (B). *P < 0.05; **P < 0.01; ***P < 0.0001.

Figure 8

Schematic diagram illustrating Parkin-mediated regulation of ER-mitochondria contacts and calcium homeostasis. Under normal conditions, IP3R-mediated Ca²⁺ flux recruits Parkin. Through its E3 ubiquitin ligase activity, Parkin ubiquitinates and degrades IP3R channels. This process leads to disassembly of the IP3R complex and helps regulate mitochondrial calcium homeostasis in mitochondria-associated membranes (MAMs). In the absence of functional Parkin, IP3R complexes abnormally accumulate at MAM regions. This increases cellular sensitivity to calcium stimuli and leads to mitochondrial calcium overload, triggering abnormal opening of the mitochondrial permeability transition pore (mPTP) and ultimately resulting in apoptosis.