Alternative mitochondrial quality control mediated by extracellular release

Abstract

Mitochondrial quality control, which is crucial for maintaining cellular homeostasis, has been considered to be achieved exclusively through mitophagy. Here we report an alternative mitochondrial quality control pathway mediated by extracellular mitochondria release. By performing time-lapse confocal imaging on a stable cell line with fluorescent-labeled mitochondria, we observed release of mitochondria from cells into the extracellular space. Correlative light-electron microscopy revealed that majority of the extracellular mitochondria are in free form and, on rare occasions, some are enclosed in membrane-surrounded vesicles. Rotenone- and carbonyl cyanide m-chlorophenylhydrazone-induced mitochondrial quality impairment promotes the extracellular release of depolarized mitochondria. Overexpression of PRKN (parkin RBR E3 ubiquitin protein ligase), which has a pivotal role in mitophagy regulation, suppresses the extracellular mitochondria release under basal and stress condition, whereas its knockdown exacerbates it. Correspondingly, overexpression of PRKN-independent mitophagy regulators, BNIP3 (BCL2 interacting protein 3) and BNIP3L/NIX (BCL2 interacting protein 3 like), suppress extracellular mitochondria release. Autophagy-deficient cell lines show elevated extracellular mitochondria release. These results imply that perturbation of mitophagy pathway prompts mitochondria expulsion. Presence of mitochondrial protein can also be detected in mouse sera. Sera of PRKN-deficient mice contain higher level of mitochondrial protein compared to that of wild-type mice. More importantly, fibroblasts and cerebrospinal fluid samples from Parkinson disease patients carrying loss-of-function PRKN mutations show increased extracellular mitochondria compared to control subjects, providing evidence in a clinical context. Taken together, our findings suggest that extracellular mitochondria release is a comparable yet distinct quality control pathway from conventional mitophagy.Abbreviations: ACTB: actin beta; ANXA5: annexin A5; ATP5F1A/ATP5A: ATP synthase F1 subunit alpha; ATG: autophagy related; BNIP3: BCL2 interacting protein 3; BNIP3L/NIX: BCL2 interacting protein 3 like; CCCP: carbonyl cyanide m-chlorophenylhydrazone; CM: conditioned media; CSF: cerebrospinal fluid; DMSO: dimethyl sulfoxide; EM: electron microscopy; HSPD1/Hsp60: heat shock protein family D (Hsp60) member 1; KD: knockdown; KO: knockout; MAP1LC3A/LC3: microtubule associated protein 1 light chain 3 alpha; MT-CO1: mitochondrially encoded cytochrome c oxidase I; NDUFB8: NADH:ubiquinone oxidoreductase subunit B8; OE: overexpression; OPA1: OPA1 mitochondrial dynamin like GTPase; OXPHOS: oxidative phosphorylation; PBS: phosphate-buffered saline; PB: phosphate buffer; PD: Parkinson disease; PINK1: PTEN induced kinase 1; PRKN: parkin RBR E3 ubiquitin protein ligase; RB1CC1/FIP200: RB1 inducible coiled-coil 1; SDHB: succinate dehydrogenase complex iron sulfur subunit B; TOMM20: translocase of outer mitochondrial membrane 20; TOMM40: translocase of outer mitochondrial membrane 40; UQCRC2: ubiquinol-cytochrome c reductase core protein 2; WT: wild-type.

Keywords: Mitochondria; Parkinson disease; mitochondrial quality control; mitophagy; parkin.

Figure 1.

Extracellular release of mitochondria. (A) Time-lapse confocal imaging on PC12^DsRed2mito showed release of extracellular mitochondria from cells. (B) Extracellular mitochondria studied by CLEM. (i–ii) Merged image of fluorescence and brightfield was acquired to visualize the grid and locate the DsRed2mito-labeled particle (arrowhead) at EM level. (ii) Scanned area (arrowhead) became dark. (iii) Zoomed-in view of the scanned area showing most extracellular DsRed2mito-labeled particles are free mitochondria. (iv) Magnified view showing typical cristae and small matrix granules in a morphological intact mitochondrion (boxed area in Biii). (C)(I) Immuno EM showing some particles are TOMM20-positive (arrowhead) as indicated by 10 nm colloidal gold particles. (ii) Immuno-gold detection of mRFP (arrowhead, 6 nm colloidal gold) and TOMM20 (black arrow, 12 nm colloidal gold) showing double-positive labeling. (D) Western blot analysis confirming presence of TOMM20, HSPD1, OPA1, CI-NDUFB8, CII-SDHB, CIII-UQCRC2, CIV-MT-CO1, CV-ATP5F1A in CM-derived pellet. (E) Effect of apoptosis inhibitors on the TOMM20 level in CM-derived pellet. One-way ANOVA, Dunnett’s test, n = 3 for each group

Figure 2.

Rotenone- and CCCP-induced mitochondrial quality impairment promotes extracellular release of mitochondria. (A) Microscopy analysis of number of extracellular mitochondria upon DMSO, rotenone and CCCP treatment. (B) Western blotting analysis showing mitochondrial proteins TOMM20, CV-ATP5F1A and HSPD1 relative to ACTB and ANXA5 level in pelleted samples and cell lysates. (C) Confocal images of JC10 fluorescence in control, rotenone- and CCCP-treated cells. (D) Images were processed for determination of aggregate:monomer (red:green) fluorescent intensity ratio, reflective of Ψm. Rotenone and CCCP caused mitochondrial depolarization. (E) JC10 fluorescence of extracellular mitochondria released upon DMSO, rotenone and CCCP treatment. (F) Quantification of total number of healthy and depolarized mitochondria released by DMSO-, rotenone- and CCCP-treated cells. Rotenone and CCCP increased release of depolarized mitochondria in a selective manner. One-way ANOVA, Dunnett’s test, n ≥ 3 for each group, * p < 0.05. (G) Electron micrographs revealing rotenone and CCCP samples contained more damaged mitochondria with disorganization or disappearance of cristae compared to control

Figure 3.

Mitophagy regulators modulate extracellular mitochondria release. (A) Representative western blot of TOMM20, HSPD1, OPA1, CI-NDUFB8, CII-SDHB, CIV-MT-CO1 and CV-ATP5F1A relative to ACTB in CM-derived pellet collected following extended non-treated condition (24 h). (B) Representative western blot of TOMM20 in CM-derived pellet collected following brief DMSO and rotenone-treated condition (4 h). PRKN overexpression suppresses extracellular mitochondria release under extended non-treated condition (24 h) and brief DMSO and rotenone-treated condition (4 h). (C) PRKN knockdown elevates extracellular mitochondria release under extended non-treated condition (24 h) and (D) brief DMSO and rotenone-treated condition (4 h). (E) Quantification of total number of healthy and depolarized mitochondria released by DMSO and rotenone-treated control and PRKN knockdown cells. One-way ANOVA, Tukey’s test, n = 3 for each group, * p < 0.05. (F) Representative western blot and densitometry showing TOMM20 relative to ACTB in CM-derived pellet of control, BNIP3L/NIX and BNIP3 overexpressing PC12 cells following extended non-treated condition. (G) Representative western blot and densitometry showing TOMM20 relative to ACTB in CM-derived pellet of HeLa WT, ATG3 KO, ATG5 KO, ATG7 KO, ATG9 KO, ATG13 KO, ATG14 KO, RB1CC1/FIP200 KO cells following extended non-treated condition. One-way ANOVA, Dunnett’s test, n ≥ 3 for each group, * p < 0.05

Figure 4.

Increased extracellular mitochondrial protein in the fibroblasts and CSF derived from PD patient with PRKN mutations and the sera from PRKN-deficient mice. (A) Representative western blot image of PRKN protein expression in primary skin fibroblasts of healthy control subjects (HC8404, HC8405, HC8546) and PD patients with PRKN mutations (PRKN 008, 010, 011). PRKN levels were undetectable in patient-derived PRKN mutant fibroblasts. (B and C) Western blotting and densitometry analysis showing increased level of extracellular mitochondrial protein TOMM20 relative to ACTB in patient-derived PRKN mutant fibroblasts. Mann Whitney Test, n = 3 for each group, * p < 0.05. (D and E) Western blotting and densitometry analysis showing increased level of mitochondrial protein TOMM40 in the CSF of patients with PRKN mutations. Mann Whitney Test, n = 5 for control group and n = 3 for patients with PRKN mutations group, * p < 0.05. (F and G) Western blotting and densitometry analysis showing increased level of mitochondrial protein TOMM20 in the sera of prkn^−/- mice. Mann Whitney Test, n = 3 for WT group and n = 5 for prkn^−/- group, * p < 0.05