mitochondrial autophagy in cells with mtdna mutations results from

Mitochondrial autophagy in cells with mtDNAmutations results from synergistic lossof transmembrane potential and mTORC1 inhibition

Robert W. Gilkerson1, Rosa L.A. De Vries3, Paul Lebot3, Jakob D. Wikstrom4, Edina Torgyekes1,

Orian S. Shirihai4, Serge Przedborski1,3 and Eric A. Schon1,2,∗

1Department of Neurology, 2Department of Genetics and Development and 3Department of Pathology and Cell

Biology, Columbia University Medical Center, New York, NY, USA and 4Evans Biomedical Research Center,

Boston University, Boston, MA, USA

Received August 8, 2011; Revised and Accepted November 7, 2011

Autophagy has emerged as a key cellular process for organellar quality control, yet this pathway apparentlyfails to eliminate mitochondria containing pathogenic mutations in mitochondrial DNA (mtDNA) in patientswith a variety of human diseases. In order to explore how mtDNA-mediated mitochondrial dysfunctioninteracts with endogenous autophagic pathways, we examined autophagic status in a panel of humancytoplasmic hybrid (cybrid) cell lines carrying a variety of pathogenic mtDNA mutations. We found thatboth genetic- and chemically induced loss of mitochondrial transmembrane potential (Dcm) caused recruit-ment of the pro-mitophagic factor Parkin to mitochondria. Strikingly, however, the loss of Dcm alone wasinsufficient to prompt delivery of mitochondria to the autophagosome (mitophagy). We found that mitophagycould be induced following treatment with the mTORC1 inhibitor rapamycin in cybrids carrying either large-scale partial deletions of mtDNA or complete depletion of mtDNA. Further, we found that the level ofendogenous Parkin is a crucial determinant of mitophagy. These results suggest a two-hit model, in whichthe synergistic induction of both (i) mitochondrial recruitment of Parkin following the loss of Dcm and (ii)mTORC1-controlled general macroautophagy is required for mitophagy. It appears that mitophagy can beaccomplished by the endogenous autophagic machinery, but requires the full engagement of both ofthese pathways.

INTRODUCTION

The human mitochondrial genome is a 16.6 kb circle ofdouble-stranded DNA (mtDNA) (1). Point mutations inmtDNA cause maternally inherited diseases, including neur-opathy, ataxia and retinitis pigmentosa (NARP), and mito-chondrial encephalomyopathy, lactic acidosis and stroke-likeepisodes (MELAS) (reviewed in 2). Large-scale (kilobase-sized) partial deletions of mtDNA (D-mtDNAs), in which aportion of the circular mtDNA is lost, are typically associatedwith Kearns–Sayre syndrome (KSS) (2). High levels ofD-mtDNAs are also found in the substantia nigra of patientswith Parkinson disease (PD) and in normal aging (3,4).

Autophagy is a major cellular quality control mechanism forthe selective degradation of large-scale protein aggregates and

organelles, including mitochondria (5). Despite the ability ofautophagy to degrade non-functional mitochondria (5), thevery existence, persistence and even accumulation of patho-genic mtDNA mutations in human patients imply that select-ive removal of mitochondria containing pathogenic mtDNAsis not occurring to any appreciable extent. The reason forthe failure of the autophagic machinery to recognize theseorganelles is unknown.

Macroautophagy is the process by which an isolation mem-brane forms around cellular components, engulfing themwithin the autophagosome, which subsequently fuses withhydrolase-containing lysosomes that break down the engulfedmaterials (6). This process is controlled directly by the mam-malian target of rapamycin (mTOR): active mTOR is com-plexed with its interacting partner Raptor and associated

∗To whom correspondence should be addressed at: Department of Neurology, Room P&S 4-449, Columbia University Medical Center,630 West 168th Street, New York, NY 10032, USA. Tel: +1 2123051665; Fax: +1 2123053986; Email: [email protected]

# The Author 2011. Published by Oxford University Press. All rights reserved.For Permissions, please email: [email protected]

Human Molecular Genetics, 2012, Vol. 21, No. 5 978–990doi:10.1093/hmg/ddr529Advance Access published on November 11, 2011

Dow

nloaded from https://academ

ic.oup.com/hm

g/article/21/5/978/584007 by guest on 01 Decem

ber 2021

factors (7) [termed mTORC1, differentiating it from themTOR/Rictor (8), or mTORC2, complex], activating akinase signaling cascade that promotes cellular proliferationwhile simultaneously inhibiting autophagy (reviewed in9,10). Under autophagic conditions, however, mTORC1assumes a kinase-inhibited conformation (7), initiating auto-phagosome formation (11). Macroautophagy can thus resultin a general, nonspecific sequestration of cytoplasm withinautophagosomes, or in the specific selective degradation ofdifferent organelles (12).

Selective targeting of mitochondria for autophagy involvesPARK6 [protein PINK1 (PTEN-induced kinase 1)] andPARK2 (Parkin, an E3 ubiquitin ligase), both of which can bemutated in familial PD (13,14). Experimental chemical uncoup-ling of the mitochondrial transmembrane potential (Dcm) causesthe translocation of PINK1 to the mitochondrial outer mem-brane (MOM) (15–17), leading to recruitment of Parkin fromthe cytosol to the MOM (15). Following the Parkin-mediatedubiquitination of mitochondrial porin (18) and mitofusins 1and 2 (16), these substrates are then bound by p62, a keyfactor in selective autophagy (19), which then deliversmitochondria to the autophagosome via interaction withmicrotubule-associated light-chain 3 (LC3) (18).

Despite the known pathogenic nature of mtDNA mutations,it is unknown why intrinsic autophagic pathways do not elim-inate these naturally occurring forms of dysfunctional mito-chondria, although the fact that patient cells typically harbora mixture of wild-type and mutant mtDNAs (heteroplasmy)might explain, at least in part, the lack of an autophagic re-sponse. If true, cells carrying only mutated mtDNAs (homo-plasmy) should have increased steady-state autophagyrelative to WT cells. We therefore assayed autophagy in abattery of homoplasmic cybrid cell lines repopulated with dif-ferent patient-derived pathogenic mutant mtDNAs.Surprisingly, the genetic loss of mitochondrial function inthese cells did not increase intrinsic steady-state levels ofautophagy, but we discovered that widespread mitophagy(defined here as delivery of mitochondria to the autophago-some) could be induced in response to two conditions: (i)the loss of Dcm, resulting in recruitment of Parkin to the or-ganelle, and (ii) the activation of macroautophagy, throughthe inhibition of mTORC1 signaling.

RESULTS

Cells carrying mtDNA mutations do not have elevatedlevels of macroautophagy

We employed a panel of cybrid cell lines, in which a human143B osteosarcoma nuclear background was depleted of allendogenous mtDNA (20) and repopulated with patient-derivedmtDNAs (21). The WT (FLP6a39.2),D (FLP6139.32), A3243G(RN164), T8993G (JCP261) and r0 (143B206) cell lines havebeen described previously (21–23). Briefly, the WT cell linecarries homoplasmic WT-mtDNA; the D cell line is homoplas-mic for a 1.9 kb D-mtDNA from a patient with KSS; theA3243G cell line is homoplasmic for its respective pointmutation derived from a patient with MELAS; the T8993Gcell line is homoplasmic for a point mutation derived from a

patient with NARP; and the r0 cell line is depleted of allmtDNA (Table 1).

We infected each cell line with a lentivirus containingLC3-GFP (24) and established clonal cultures to ensure con-sistent LC3-GFP expression. Cytosolic LC3-I is recruited tothe autophagosome upon the induction of autophagy via con-jugation to phosphatidylethanolamine, producing the autopha-gic LC3-II form (6). When fused to GFP (�27 kDa), the twoLC3 forms migrate on gels with apparent molecular weights of�43 and �41 kDa, respectively. In western blots of cellsexpressing LC3-GFP-II, a band of �27 kDa is also observed,as fusion of the autophagosome with the lysosome results incleavage of LC3-GFP-II, producing ‘free’ GFP as anadditional marker of autophagy (25). When visualized byfluorescence microscopy, LC3-GFP-I is present diffusely inthe cytosol; upon autophagic induction, LC3-GFP-I isprocessed to LC3-GFP-II and recruited to autophagosomesas discrete punctae, providing a useful visual marker ofautophagosomes (26,27).

Upon anti-GFP western blotting of lysates from the cybridcell lines, we observed a predominant band for cytosolicLC3-GFP-I at �43 kDa in all samples (Fig. 1A). We alsoobserved bands corresponding to the autophagic LC3-GFP-II(�41 kDa) and free GFP (�27 kDa) species, but at lowerlevels [i.e. a ratio of (LC3-GFP-II + free GFP)/(LC3-GFP-I)of �0.2–0.5], with D cells showing a (non-significant)2.5-fold higher ratio than WT (Fig. 1B, gray bars). In agree-ment with the western blotting, live-cell fluorescence micros-copy revealed that none of the cell lines examined had anyappreciable proliferation of autophagosomes (SupplementaryMaterial, Fig. S1A, upper panels). Even though the D cellshad increased amounts of autophagic LC3-GFP-II and freeGFP (Fig. 1A), confocal fluorescence microscopy showedthat there was no difference between WT and D cells in thenumber or size of the LC3-GFP punctae (Fig. 1C and D).Thus, our data indicate that there is no increase in autophagyin cells carrying mtDNA mutations relative to WT cells.Western blotting against endogenous LC3 revealed that bothD and r0 cells actually have decreased endogenous LC3levels, compared with two independently derived WT lines(Supplementary Material, Fig. S1D).

Rapamycin challenge reveals increased macroautophagyin D cells

As none of the cloned cybrid lines had elevated levels ofautophagosomes, we asked whether they could respond to an

Table 1. Homoplasmic cybrid lines used in this studya

Cell line Full code name Disease Reference

WT FLP6a39.32 None 23D FLP6a39.32 KSS 23A3243G RN164 MELAS 30T8993G JCP261 NARP 21r0 143B206 In vitro mtDNA depletion 20WT2 CW420-116 None 21D 2 KAF4EB12.49 KSS 49

aSee also reference 21.

Human Molecular Genetics, 2012, Vol. 21, No. 5 979

Dow


ic.oup.com/hm


ber 2021

http://hmg.oxfordjournals.org/lookup/suppl/doi:10.1093/hmg/ddr529/-/DC1



autophagic stimulus. We therefore treated them with themTORC1 inhibitor rapamycin (7), a canonical inducer ofmacroautophagy.

Treatment with 250 nM rapamycin for 48 h (SupplementaryMaterial, Fig. S1B) caused no appreciable increase inLC3-GFP-II in WT, A3243G, T8993G or r0 cells, but D celllysates revealed a striking �4-fold increase in theLC3-GFP-II and free GFP bands (Fig. 1A and B, blackbars). We obtained similar results both by LC3-GFP fluores-cence microscopy (Supplementary Material, Fig. S1A, lowerpanels) and by immunolabeling against endogenous LC3-II(Supplementary Material, Fig. S1C). Using confocal fluores-cence microscopy, compared with untreated cells, neitherrapamycin-treated WT nor D cells showed significantincreases in the number of autophagosomes per cell, butrapamycin-treated D cells displayed an increase in theaverage size of autophagosomes (61+ 5 versus 27+ 10pixels in untreated D cells) (Fig. 1D), reflecting aproliferation of numerous large LC3-GFP punctae (Fig. 1C).Similar results were obtained using uncloned mass culture ofLC3-GFP-infected WT and D cybrids (Supplementary Mater-ial, Fig. S1E), indicating that these results were not confined toany particular clonal cell line. To determine the nature of thisincreased autophagosomal content, we examined D cell re-sponse to chloroquine, an inhibitor of autophagosomal degrad-ation (28) (Supplementary Material, Fig. S2A–C), as well aslysosomal acidification in response to rapamycin (Supplemen-tary Material, Fig. S2D). These experiments demonstrated thatthe observed proliferation of autophagosomes in rapamycin-treated D cells was due to a bona fide increase in autophagicflux, rather than to a defect in autophagosome maturationcaused by the loss of mitochondrial function in D cells.Thus, rapamycin preferentially induces macroautophagy in Dcells, but not in other cybrid cell lines.

Autophagosomes within rapamycin-treated D cells containmitochondria with severely decreased mitochondrialtransmembrane potential

To determine whether the autophagic induction observed inrapamycin-treated D cells resulted in mitophagy, we examinedLC3-GFP-infected cells that had been counterstained withMitoTracker Red. As expected from Figure 1, only rapamycin-treated D cells showed a proliferation of large autophago-somes, as demonstrated by the �2-fold increase in theaverage size of autophagosomes (Fig. 1D); strikingly, manyof these abnormally large structures were MitoTracker-positive (Fig. 2A, arrowheads), indicating that these autopha-gosomes had indeed engulfed mitochondria. Quantitativeimage analysis revealed that the rapamycin-treated D cellshad, on average, 23+ 7% of their total mitochondria coloca-lizing with autophagosomes, compared with just 4+ 2% foruntreated D cells, which was similar to the value in rapamycin-treated WT cells (5+ 1%) (Fig. 2B). Collectively, thesefindings demonstrate robust uptake of mitochondria byautophagosomes in rapamycin-treated D cells.

We examined the involvement of the mitochondrial-specificParkin and p62 autophagic factors in this mitophagy. Follow-ing treatment with rapamycin, GFP-Parkin was recruited tomitochondria of D cells (Fig. 3A). WT cells showed adiffuse GFP fluorescence throughout the cell that was essen-tially unchanged upon treatment with rapamycin. UntreatedD cells showed a mosaic distribution of fluorescence, with

Figure 1. Autophagic status of cybrid cell lines. (A) Western blot analysis oftotal protein (20 mg per lane), probed with anti-GFP (n ¼ 4); lower panelshows anti-tubulin loading control. (B) Quantification of bands fromanti-GFP western blotting using ImageJ. Band intensities of autophagicLC3-GFP-II and free GFP bands were combined and expressed as a ratiowith cytosolic LC3-GFP-I (means of n ¼ 4+SD). The asterisk indicates dif-ference from untreated cells (P , 0.001, the Newman–Keuls post hoc test fol-lowing a two-way repeated measure ANOVA). (C) Confocal fluorescencemicroscopy of fixed cells from LC3-GFP-infected WT and D

cybrid cell lines grown in the absence or presence of 250 nM rapamycin for48 h (n ¼ 4). Scale bar ¼ 20 mm. (D) Quantitation of the average number ofLC3-GFP autophagosome punctae per cell and the average size of LC3-GFPautophagosome punctae (means of n ¼ 4+SD). The asterisk indicates differ-ence from untreated cell lines (P ¼ 0.034, the Newman–Keuls post hoc testfollowing a two-way repeated measure ANOVA).

980 Human Molecular Genetics, 2012, Vol. 21, No. 5

Dow


ic.oup.com/hm


ber 2021












some cells showing punctate signals that colocalized withMitoTracker (Fig. 3A, arrows), whereas rapamycin-treated Dcells displayed a striking recruitment of GFP-Parkin to largemitochondria (Fig. 3A, arrowheads). Anti-p62

immunolabeling showed a strong signal only in rapamycin-treated D cells, appearing as large punctae which frequentlycolocalized with mitochondria (Fig. 3B, arrowheads). More-over, p62 immunolabeling also showed colocalization with

Figure 2. Confocal microscopy of mitochondria and autophagosomes from WT and D cells. (A) WT and D cybrid cell lines incubated in the absence or presenceof 250 nM rapamycin for 48 h, followed by staining with MitoTracker Red (MTR) (n ¼ 4). Scale bar ¼ 20 mm. Note colocalization of mitochondria withLC3-GFP-positive autophagosomes (arrowheads). (B) Quantification of mitochondria colocalizing with LC3-GFP punctae (means of n ¼ 4+SD). The asteriskindicates difference from all other cell lines (P ¼ 0.005, the Newman–Keuls post hoc test following a two-way repeated measure ANOVA).


Dow


ic.oup.com/hm


ber 2021

LC3-GFP autophagosomes in rapamycin-treated D cells(Fig. 3C, arrowheads). These experiments demonstrate thatthe autophagic uptake of mitochondria in rapamycin-treatedD cells was mediated by Parkin and p62.

Since Parkin-mediated mitophagy would presumably requirethe loss of Dcm (15–17), we examined basal Dcm in the cybridlines using tetramethyl rhodamine ester (TMRE). Notably, D

cells had the lowest TMRE fluorescence of all the cell linesstudied (Fig. 4A and B). WT cells maintained an average peakfluorescence intensity of 446+ 194 U, whereas D and r0 cells[both with essentially no respiratory chain function (21)] main-tained low peak TMRE values (138+ 53 and 234+ 70, re-spectively), similar to values in which Dcm was dissipatedcompletely by the protonophore uncoupler carbonyl cyanide

Figure 3. Confocal fluorescence microscopy of Parkin and p62 in mitochondrial autophagy. (A) WT and D cells were transiently transfected with GFP-Parkin,grown in the absence or presence of rapamycin, and stained with MitoTracker Red (n ¼ 2). Scale bar ¼ 20 mm. (B) WT and D cells were grown in the absence orpresence of rapamycin, immunolabeled with anti-p62 antibody and stained with MitoTracker Red. Note recruitment of p62 to large, swollen mitochondria(arrowheads) (n ¼ 2). (C) Confocal fluorescence microscopy of LC3-GFP-infected WT and D cells grown in the absence or presence of rapamycin and immu-nolabeled with anti-p62 antibody (n ¼ 2). Scale bar ¼ 20 mm.


Dow


ic.oup.com/hm


ber 2021

m-chlorophenylhydrazone (CCCP) (148+ 85 U). On the otherhand, T8993G cells [with relatively intact respiratory chainfunction (29)] and A3243G cells [with �30% residual respira-tory chain function (30)] exhibited more WT-like peak TMREvalues (280+ 76 and 448+ 43, respectively). Live-cellimaging essentially recapitulated these results (Fig. 4C). Thus,the strong autophagic induction observed in D cells coincidedwith their low steady-state Dcm.

mTOR inhibition and loss of Dcm are both necessaryfor mitophagy

Rapamycin appears to induce mitophagy in D cells by a mech-anism involving low Dcm. But if the loss of Dcm were thesole requirement for mitophagy, why did D cells show no ele-vated basal level of macroautophagy? We therefore askedwhether the loss of Dcm, in and of itself, was sufficient toinduce autophagy, by treating WT cells with low levels of

CCCP, which is sufficient to collapse Dcm, inducing recruit-ment of Parkin to mitochondria (16,17) without adverselyaffecting other cellular functions (31). In WT cells incubatedwith increasing concentrations of CCCP (up to 10 mM), weobserved little or no change in the levels of LC3-GFP-II byanti-GFP western blotting, relative to untreated WT cells(Fig. 5A). Thus, CCCP treatment alone was insufficient toinduce macroautophagy. [On the other hand, treatment of WTcells with exceedingly high levels of CCCP (100 mM) resultedin the proliferation of autophagosomes and striking increasesin LC3-GFP-II relative to untreated WT cells (SupplementaryMaterial, Fig. S3A); however, these concentrations of CCCPaffect microtubule dynamics (31) and lysosomal permeability(32), and as such are not specific to mitochondrial Dcm.]

This lack of autophagic induction prompted us to considerother potential signaling requirements for mitochondrialautophagy beyond the collapse of Dcm. Although D cells con-stitutively maintain a low Dcm, they only induced mitophagy

Figure 4. Analysis of mitochondrial transmembrane potential in cybrid cell lines. (A) Average peak TMRE fluorescence values of cybrid cell lines (means of n ¼3+SD). Peak fluorescence values were normalized to unstained cells for each experiment. The asterisk indicates difference from WT (P , 0.001, the Dunnettpost hoc test following a one-way ANOVA). (B) Representative flow cytometry experiment of cybrid cell lines incubated with 2.5 nM TMRE. A total of 50 000events per cell line. Histogram (Y-axis) is expressed as % of cells at maximum values. (C) Live-cell fluorescence microscopy of cybrid cell lines incubated with2.5 nM TMRE. Note low TMRE fluorescence of D cells. Scale bar ¼ 20 mm.


Dow


ic.oup.com/hm


ber 2021



when challenged with rapamycin. Accordingly, we examinedmTORC1 signaling, which controls macroautophagy, in ourcells. When mTORC1 kinase signaling is active (i.e. macroau-tophagy is repressed), downstream targets are phosphorylated,including the ribosomal S6 protein (33). In all CCCP-treatedWT cells, we observed a strong band for phospho-S6(Fig. 5A), consistent with the lack of autophagic inductionin response to CCCP treatment observed above.

We then monitored mTORC1 signaling in our panel ofmutant cybrids, including the rapamycin-treated D cells thathad displayed robust autophagic induction. In agreementwith our previous results (Fig. 1), all untreated cell linesmaintained a strong phospho-S6 signal, indicating thatmTORC1-mediated signaling was active (Fig. 5B), precludingthe induction of macroautophagy. However, in rapamycin-treated cells, none of the samples exhibited a signal forphospho-S6, indicating a complete inhibition of mTORC1 sig-naling, allowing for the induction of macroautophagy(Fig. 5B). In spite of the derepression of mTORC1 in all ofour cell lines, autophagy was induced only in the D cells,but not in the WT, T8993G, A3243G or r0 cells, indicatingthat mTORC1 inhibition alone is not sufficient to induce mito-chondrial autophagy.

Thus, neither the loss of Dcm alone nor the inhibition ofmTORC1 alone was sufficient to induce mitophagy in Dcells; both were required.

Chemical uncoupling does not fully recapitulate the geneticloss of Dcm

These data demonstrate that although D cells have a severelyimpaired basal Dcm, mitophagy is induced only upon the add-ition of rapamycin. Accordingly, we hypothesized that theautophagic induction in rapamycin-treated D cells was dueto the synergistic combination of the loss of Dcm coupledwith mTORC1 inhibition. To test this hypothesis, we exam-ined whether CCCP-mediated loss of Dcm, combined with

Figure 5. Autophagy and mTORC1 signaling. (A) Western blot analysis oflysates from WT cells (20 mg) incubated for 1 h with increasing concentrationsof CCCP, probed with anti-GFP, anti-phosophoS6 and anti-tubulin antibodies(n ¼ 2). (B) Western blot analysis of lysates from cybrid cell lines incubated inthe absence or presence of 250 nM rapamycin for 48 h, probed with anti-phosphoS6 and anti-tubulin antibodies (n ¼ 2).

Figure 6. CCCP-mediated uncoupling and autophagy. (A) Top, schematic ofexperiment. Confocal fluorescence microscopy of WT cells grown in theabsence or presence of 10 mM CCCP for 1 h, without or with pretreatmentwith rapamycin for 48 h; (n ¼ 3). Size bar ¼ 20 mm. (B) Quantitation of theaverage number of LC3-GFP autophagosome punctae per cell and averagesize of LC3-GFP autophagosome punctae of the cells in A (means of n ¼3+SD). D cell data from Figure 1. (C) Top, schematic of experiment. Con-focal fluorescence microscopy of WT cells grown in the absence orpresence of 10 mM CCCP for 49 h, in the absence or presence of rapamycinfor 48 h (n ¼ 3). (D) Quantitation as in (B); (n ¼ 3).


Dow


ic.oup.com/hm


ber 2021

rapamycin treatment, might recapitulate the autophagic pheno-type in a WT mtDNA background. As expected, neither treat-ment of WT cells with 10 mM CCCP alone nor with 250 nM

rapamycin alone yielded the proliferation of largeautophagosomes that we saw in rapamycin-treated D cells(Fig. 6A and B). Contrary to our expectations, however, treat-ment first with rapamycin for 48 h followed by CCCP for 1 halso did not show autophagic induction (Fig. 6A and B). Wetherefore attempted to mimic more closely the constitutiveloss of Dcm in D cells by incubating WT cells first withCCCP for 1 h and then treating the cells with rapamycin for48 h in the presence of CCCP, but in spite of the presenceof both agents, autophagic induction was still minimal(Fig. 6C and D). When we examined the WT cells that hadbeen treated with CCCP for short (1 h) or long (49 h)periods for the colocalization of autophagosomes with mito-chondria (Supplementary Material, Fig. S3B), they displayedfew colocalizing LC3-GFP punctae (Supplementary Material,Fig. S3C). Furthermore, CCCP did not induce the formation ofautophagosomes in either of two independent WT cell lines, asassayed by anti-LC3-II immunolabeling (SupplementaryMaterial, Fig. S3D).

Thus, in no case did CCCP treatment, whether short or longterm, either singly or in combination with rapamycin, elicit anautophagic induction in WT cells equivalent to that seen inrapamycin-treated D cells. Thus, there appears to be an essen-tial difference in the 143B osteosarcoma background betweenchemically induced dissipation of Dcm by CCCP and naturallyoccurring loss of Dcm due to authentic mtDNA mutations.

Parkin is necessary for mitophagy in cells carrying mtDNAmutations

The inability of CCCP to induce mitochondrial autophagy inWT cells raised a mechanistic question: what were the relativelevels of PINK1 and Parkin in our cell lines? If PINK1 andParkin were greatly decreased or absent in WT or r0 cells,no amount of uncoupling would have induced mitochondrialautophagy; conversely, if D cells had much greater expressionof PINK1 and Parkin than other cells, this might explain theirdifferential propensity to undergo rapamycin-induced mito-chondrial autophagy.

We therefore used quantitative reverse transcription PCR(qRT-PCR) to examine PINK1 and Parkin mRNA expressionin our panel of cybrids (endogenous levels of these proteinsare very low and thus difficult to observe with antibodies),both constitutively and in response to rapamycin. WT cellscontained ample amounts of PINK1 and Parkin [more thanfour times as much as in human SH-SY5Y or HeLa cells(34,35)]. However, basal endogenous transcript levels ofPINK1 were significantly lower than WT in several of themtDNA mutant cybrids: an average of 15% of WT inT8993G, 26% in A3243G and 36% in r0 cells (Fig. 7A). Simi-larly, basal expression of Parkin was frequently lower: 47% inD, 35% in T8993G and only 21% in r0 cells (Fig. 7A). Treat-ment with rapamycin did not result in significant changes inthese levels (Supplementary Material, Fig. S4A and B).

Thus, the lack of autophagy in WT cells was probably duenot to low levels of PINK1 or Parkin but rather to high Dcm.However, the lack of autophagy in r0 cells (with low Dcm)

could indeed have been due to low levels of PINK1 andParkin expression, as r0 cells had dramatically decreasedlevels of these factors. We therefore examined the ability ofthe different cybrid lines to recruit Parkin to mitochondria(Supplementary Material, Fig. S4C). Upon transfection withGFP-Parkin, we found that both D and r0 cells had increasedproportions of cells with mitochondrial recruitment of Parkin(39 and 37%, respectively) compared with WT cells trans-fected in a similar manner (1%) (Fig. 7B). As expected,there was little recruitment of GFP-Parkin in the A3243Gand T8993G cells (Fig. 7A), as they had essentially WTlevels of Dcm. Thus, reduced Dcm in r0 cells causes recruit-ment of Parkin to mitochondria in a manner identical to thatin D cells when exogenous Parkin is introduced. Upon the add-ition of CCCP, WT, r0, D and A3243G cells showed dramaticrecruitment of Parkin to mitochondria, as expected, given thetotal collapse of Dcm.

As Parkin is genetically downstream of PINK1 (36), wetherefore overexpressed GFP-Parkin in r0 cells to askwhether increasing Parkin levels could induce the formationof autophagosomes like that seen in rapamycin-treated Dcells. As expected, in the absence of exogenous Parkin,rapamycin-treated WT and r0 cells showed very few cytoplas-mic LC3-GFP autophagosomes, whereas rapamycin-treated Dcells showed a characteristic proliferation of large autophago-somes (Fig. 7C). Strikingly, however, Parkin-transfected r0

cells showed a marked proliferation of LC3-GFP autophago-somes in response to rapamycin treatment (Fig. 7C). Quantita-tion of LC3-GFP punctae per cell and average puncta size(Fig. 7D) confirmed that the addition of Parkin to r0 cellsresults in an autophagic phenotype identical to that seen inrapamycin-treated D cells. Microscopic analysis of theLC3-GFP autophagosomes in Parkin-expressing r0 cellsrevealed that these autophagosomes frequently colocalized tomitochondria, with concomitant Parkin recruitment to thoseorganelles (Fig. 7E). In agreement with Suen et al. (37),these results suggest that r0 cells are essentially similar to Dcells in terms of mitophagic induction; they merely haveless endogenous Parkin.

DISCUSSION

Our data indicate that widespread mitophagy (defined here asmitochondria sequestered within autophagosomes) can beachieved in cells carrying mtDNA mutations, and that thelevel of Parkin is critically important to the cell’s ability toundergo mitophagy. Strikingly, the loss of Dcm, in and ofitself, is not sufficient to activate mitophagy: the mitophagic re-sponse requires the synergistic activation of (i) the loss of Dcm

(activating recruitment of Parkin to mitochondria) and (ii)rapamycin-mediated inhibition of mTORC1 signaling (activat-ing general macroautophagy), in order to accomplish robustautophagic sequestration of dysfunctional mitochondria. Ourresults indicate that a crucial distinction exists between recruit-ment of Parkin to mitochondria (due to the loss of Dcm) anduptake of mitochondria within autophagosomes (true mito-phagy). In other words, although the loss of Dcm is necessaryfor mitophagy, it is not sufficient.


Dow


ic.oup.com/hm


ber 2021








We find that mtDNA mutations resulting in the loss of Dcm

cause increased recruitment of Parkin to mitochondria, con-sistent with and extending recent studies demonstrating thatchemical dissipation of Dcm results in recruitment of Parkinto mitochondria (15,17,35). However, our studies suggestthat the loss of Dcm alone, whether chemically induced oras a result of mtDNA mutation, does not induce the formationof autophagosomes to any appreciable extent. Basally, cells

carrying D-mtDNAs did not have a significantly greaternumber or size of autophagosomes than did WT cells. Wealso found that the addition of CCCP to WT cells failed toinduce the proliferation of autophagosomes, mirroring findingsin cortical neurons (38). These findings are consistent with theobservation that D-mtDNAs, which cause reduced Dcm, canexist in patients without being degraded [in fact, they areoften amplified preferentially to those containing

Figure 7. (A) PINK1 and Parkin mRNAs in cybrid cell lines. qRT-PCR analysis of endogenous levels of PINK1 (gray) and Parkin (black) transcripts in cybridcell lines (means of n ¼ 3+SD). The asterisk indicates difference from WT (P , 0.05, the Dunnett post hoc test following a one-way ANOVA). (B) Parkinrecruitment in cybrid cell lines. Indicated cell lines were transiently transfected with GFP-Parkin and scored for their recruitment of Parkin to mitochondria(Supplementary Material, Fig. S4C) in the absence or presence of 10 mM CCCP. The asterisk indicates difference from untreated cells (P , 0.001, theNewman–Keuls post hoc test following a two-way repeated measure ANOVA; n ¼ 3). (C) LC3-GFP-expressing WT, D and r0 cells were incubated with rapa-mycin for 48 h. Lower panels show cells transiently transfected with mCherry-Parkin and provide representative illustration of three independent experiments.Asterisks denote cells expressing mCherry-Parkin (not shown in figure). Size bar ¼ 20 mm. (D) Quantitation of LC3-GFP punctae numbers and puncta sizes percell, as in Figures 1D, 6B and D. n ¼ 3 experiments. The asterisk indicates difference from untransfected cells (P ¼ 0.031, the Newman–Keuls post hoc testfollowing a two-way repeated measure ANOVA). (E) LC3-GFP-expressing r0 cells were transiently transfected with mCherry-Parkin (red) and incubated withrapamycin for 48 h, followed by staining with MitoTracker DeepRed (blue). Inset: note colocalization of all three signals. Scale bar ¼ 20 mm.


Dow


ic.oup.com/hm


ber 2021


WT-mtDNA (2)]. These data argue that an additional signal-ing requirement exists for autophagic degradation of mito-chondria in vivo.

One such requirement appears to be the activation of macro-autophagy, via the inhibition of mTORC1 signaling. It wasonly when mTORC1 was inhibited via rapamycin that weobserved a proliferation of autophagosomes containing mito-chondria. Interestingly, although we examined a series of dif-ferent pathogenic mtDNA mutations, only D cells (with lowDcm), but not A3243G or T8993G cells (with relatively highDcm), showed a robust induction in response to rapamycin.Thus, the loss of mitochondrial function per se [i.e. impairedrespiratory chain function (A43243G) or reduced ATP synthe-sis (T8993G)] is apparently not an important determinant inrecruiting pro-mitophagic factors to mitochondria, consistentwith recent findings showing a similar lack of mitochondrialParkin recruitment in mice with transgenic loss of mitochon-drial function (39).

A challenge to the ‘two-hit’ model proposed above (i.e. theloss of Dcm plus derepression of mTORC1) is our finding thatthe reduction of Dc via CCCP treatment of WT cells in con-junction with rapamycin treatment failed to result in a robustmitophagic phenotype, as had been seen in rapamycin-treatedD cells. There are at least two reasons that might explain thisdiscrepancy. First, protonophores like CCCP dissipate Dcacross all polarized cellular membranes, not just mitochondrialmembranes. In particular, besides depolarizing the plasmamembrane (40) and mitochondria, protonophores have beenshown to depolarize lysosomes and to inhibit lysosomal acid-ification (32,41), which is necessary for autophagic degrad-ation. Second, mtDNA mutations represent a constitutive,mitochondrial-specific loss of Dcm: these cells lack fullyassembled respiratory complexes (42) and show increasedmitochondrial ubiquitination (43), whereas CCCP-treatedcells do not (44). Thus, we feel that this discrepancy under-scores a distinction between the pharmacological dissipationof overall membrane potential by a protonophore and themitochondrial-specific loss of Dcm due to authentic mtDNAmutations.

In support of this view, although it has been elegantly andcomprehensively shown that CCCP treatment of various celllines causes recruitment of Parkin to otherwise-normal mito-chondria (15–17), it has not, in fact, been demonstrated con-vincingly that such treatment results in widespread deliveryof those mitochondria to autophagosomes. In the one studyin which quantitative results were reported (15), CCCP treat-ment of Parkin-overexpressing HeLa cells [containing ap-proximately 400–800 mitochondria (45)] resulted in only�1–2% of the organelles colocalizing with LC3, a valuethat contrasts with the �23% of mitochondria colocalizingwith LC3 in D cells (Fig. 2B).

Our results also demonstrate the importance of Parkin levelsto the cell’s ability to undergo mitophagy. Initially, we hadexpected that r0 cells (with a low Dcm, similar to that in Dcells) would display mitophagy when challenged with rapamy-cin, asD cells did. However, we subsequently determined that r0

cells had the lowest basal level of Parkin transcripts of all exam-ined cell lines. When exogenous Parkin was expressed in r0

cells, we found that they were able to recruit Parkin to mitochon-dria and they responded to rapamycin treatment with a

mitophagic phenotype identical to that found in rapamycin-treated D cells. These experiments highlight the importance ofParkin expression to the cell’s ability to undergo mitophagy.

Most mechanistic studies of mitochondrial autophagy todate have relied on overexpression of PINK1 and Parkin(37), but relatively little is known about the intrinsic biologyof these two factors. In our analysis of the relative levels ofPINK1 and Parkin mRNAs in cybrids, mtDNA mutant celllines, regardless of the specific mtDNA defect, had decreasedsteady-state levels of these transcripts relative to WT cells.Similarly, we found that cells carrying D-mtDNAs, as wellas r0 cells, showed decreased levels of endogenous LC3.Knockdown of PINK1 results in mitochondrial dysfunction(46). Our findings suggest that the converse may also betrue: mtDNA-mutant cells have depressed PINK1 and Parkinexpression, as well as decreased endogenous LC3 levels.These data suggest that mtDNA-based mitochondrial dysfunc-tion represses transcription of pro-mitophagic factors, al-though the mechanism by which this ‘retrograde’ signalingoccurs is currently unknown.

These experiments demonstrate the physiological potentialof mitophagy as a therapeutic strategy to treat mitochondrialdiseases, such as KSS, and neurodegenerative disorders,such as PD. High levels of D-mtDNAs in substantia nigraneurons of both normal aged individuals (4) and PD patients(3) suggest that the accumulation of D-mtDNAs maycombine with environmental insults and other nuclear muta-tions to elicit neuronal cell death. We note that gene profilingof dopaminergic neurons from patients with sporadic PDrevealed a decrease in PINK1 and Parkin expression (47), mir-roring the expression of these factors observed here. This lossof mitophagic factors may contribute to the persistence ofpathogenic mtDNAs in patients. Thus, therapeutic approachesin PD employing pro-autophagic agents such as rapamycinmay induce autophagic degradation of mitochondria carryingD-mtDNAs (48).

In summary, our findings suggest a ‘two-hit’ model formitochondrial autophagy: (i) the activation of mitophagic tar-geting (via the loss of Dcm) and (ii) the activation of generalmacroautophagy (via mTORC1 inhibition). These studies indi-cate that mtDNA-mediated mitochondrial dysfunction hasretrograde signaling effects on pro-mitophagic factors suchas PINK1, Parkin and LC3, and that the endogenous autopha-gic machinery is sufficient for autophagy of mitochondria car-rying D-mtDNAs. These findings provide a mechanistic basisfor mitochondrial autophagy, suggesting that pro-autophagicstrategies may be effective against diseases associated withD-mtDNAs, such as KSS and PD, using the activation of en-dogenous autophagic pathways against physiological formsof mitochondrial dysfunction.

MATERIALS AND METHODS

Cell culture

Cells were grown in Dulbecco’s high-glucose MEM (Gibco)with 10% FBS (Invitrogen) and supplemented with 10 mg/mluridine (Sigma) at 378C and 5% CO2. Lentiviral LC3-GFPinfection was performed as described previously (24).LC3-GFP-expressing clonal cybrid lines were isolated using


Dow


ic.oup.com/hm


ber 2021

cloning rings (Scienceware). Rapamycin, chloroquine diphos-phate salt and CCCP were from Sigma. Cells were seeded at50 000 cells per well in six-well dishes and allowed to recoverovernight prior to treatment. Unless otherwise indicated, allrapamycin treatments were for 48 h at 250 nM. For flow cytome-try, 500 000 cells were seeded to 6 cm-diameter dishes over-night, followed by 20 min incubation with 2.5 nM TMRE,washed and analyzed using an LSRII Cell Analyzer (50 000events per trial).

Fluorescence microscopy

Cells were incubated with MitoTracker Red CMXRos orLysoTracker Red (Invitrogen) for 30 min, washed and fixedwith 4% paraformaldehyde in PBS for 30 min. Immunofluor-escence microscopy used anti-p62 monoclonal antibody(Abcam 56416), anti-TOM20 polyclonal antibody (SantaCruz sc-11415) or anti-LC3B polyclonal antibody (SigmaL7543) in conjunction with AlexaFluor-conjugated goat anti-mouse and goat anti-rabbit secondary antibodies (Invitrogen).Cover slips were mounted with 50% glycerol in PBS. Con-focal microscopy utilized a Zeiss LSM510 META invertedconfocal microscope with a 40× oil immersion objective.Confocal images were adjusted equally using ImageJ. Conven-tional oil immersion and live-cell imaging used an OlympusIX70 inverted fluorescence microscope and SPOT RT digitalcamera. Image quantitation was performed using MetaMorph.Images were processed equally. For TMRE, live-cell imagingwas conducted by incubating cells in 2.5 nM TMRE for20 min, followed by direct visualization.

Western blotting

Cell lysates were washed with PBS and lysed in cold RIPAbuffer (50 mM Tris–HCl, 150 mM NaCl, 1 mM EDTA, 0.1%SDS) on ice for 5 min. Lysates (20 mg protein) were run on a10% SDS–PAGE gel overnight at 25 V, 48C. Proteins weretransferred to ZetaProbe PVDF (Bio-Rad) for 1 h at 100 V,48C. Blots were probed with anti-GFP monoclonal antibodyJL-8 (Clontech), anti-phospho ribosomal S6 monoclonal anti-body (Cell Signaling, Catalog No. 2215), or anti-tubulin(Sigma) at 1:1000 dilution in PBS overnight at 48C, followedby incubation with horseradish peroxidase-conjugated goatanti-mouse antibody (Amersham). Blots were developed usingWestDura Extended Duration Substrate (Thermo Scientific)and exposed films developed on a Kodak X-Omat. Westernblot quantification was performed using ImageJ.

Quantitative reverse transcription

cDNAs were produced using SuperScriptw First-Strand Syn-thesis System (Invitrogen). An amount of 50 ng of cDNAswere run in reactions using PINK1, Parkin and actinprimers from Qiagen (Hs_PINK1_2_SG QT01670459,Hs_PARK2_va.1_SG QT01005571, Hs_ACTB_2_SGQT01680476) with Power SYBRw Green PCR Master Mix(Applied Biosystems) for 40 cycles.

Statistical analysis

Differences among means were analyzed using one- ortwo-way repeated measure analysis of variance (ANOVA).When ANOVA showed significant differences, pair-wise com-parisons between means were tested by the Newman–Keulspost hoc testing, whereas comparisons against the controlgroup were tested by the Dunnett post hoc testing. In all ana-lyses, the null hypothesis was rejected at the 0.05 level. Allstatistical analyses were performed using SigmaStat forWindows 3.5.

SUPPLEMENTARY MATERIAL

Supplementary Material is available at HMG online.

ACKNOWLEDGEMENTS

We thank Theresa Swayne, Hua Yang, Estela Area-Gomezand Guy Las for technical advice.

Conflict of Interest statement: The authors declare they haveno conflicting interest.

FUNDING

This work was supported by grants from the National Institutesof Health [HD32062 (to E.A.S.) and DK074778 (to O.S.S)], theMuscular Dystrophy Association, Edison Pharmaceuticals, theEllison Medical Foundation and the Marriott MitochondrialDisorder Clinical Research Fund (MMDCRF). S.P. is supportedby grants from the National Institutes of Health (NS042269,NS064191, NS38370, NS070276 and NS072182), the US De-partment of Defense (W81XWH-08-1-0522, W81XWH-08-1-0465 and W81XWH-09-1-0245), the ParkinsonDisease Foundation, the Thomas Hartman Foundation For Par-kinson’s Research, Project A.L.S., the Muscular Dystrophy As-sociation and P2ALS. Work at the Confocal and SpecializedMicroscopy Facility at Columbia University was supported byNIH grants P30 CA13696 and S10 RR017885.

REFERENCES

1. Anderson, S., Bankier, A.T., Barrell, B.G., de Bruijn, M.H., Coulson,A.R., Drouin, J., Eperon, I.C., Nierlich, D.P., Roe, B.A., Sanger, F. et al.(1981) Sequence and organization of the human mitochondrial genome.Nature, 290, 457–465.

2. DiMauro, S. and Schon, E.A. (2003) Mitochondrial respiratory-chaindiseases. N. Engl. J. Med., 348, 2656–2668.

3. Bender, A., Krishnan, K.J., Morris, C.M., Taylor, G.A., Reeve, A.K.,Perry, R.H., Jaros, E., Hersheson, J.S., Betts, J., Klopstock, T. et al. (2006)High levels of mitochondrial DNA deletions in substantia nigra neurons inaging and Parkinson disease. Nat. Genet., 38, 515–517.

4. Kraytsberg, Y., Kudryavtseva, E., McKee, A.C., Geula, C., Kowall, N.W.and Khrapko, K. (2006) Mitochondrial DNA deletions are abundant andcause functional impairment in aged human substantia nigra neurons. Nat.Genet., 38, 518–520.

5. Twig, G., Hyde, B. and Shirihai, O.S. (2008) Mitochondrial fusion, fissionand autophagy as a quality control axis: the bioenergetic view. Biochim.Biophys. Acta, 1777, 1092–1097.

6. Boya, P., Gonzalez-Polo, R.A., Casares, N., Perfettini, J.L., Dessen, P.,Larochette, N., Metivier, D., Meley, D., Souquere, S., Yoshimori, T. et al.


Dow


ic.oup.com/hm


ber 2021


(2005) Inhibition of macroautophagy triggers apoptosis. Mol. Cell. Biol.,25, 1025–1040.

7. Kim, D.H., Sarbassov, D.D., Ali, S.M., King, J.E., Latek, R.R.,Erdjument-Bromage, H., Tempst, P. and Sabatini, D.M. (2002) mTORinteracts with raptor to form a nutrient-sensitive complex that signals tothe cell growth machinery. Cell, 110, 163–175.

8. Sarbassov, D.D., Ali, S.M., Kim, D.H., Guertin, D.A., Latek, R.R.,Erdjument-Bromage, H., Tempst, P. and Sabatini, D.M. (2004) Rictor, anovel binding partner of mTOR, defines a rapamycin-insensitive andraptor-independent pathway that regulates the cytoskeleton. Curr. Biol.,14, 1296–1302.

9. Sarbassov, D.D., Ali, S.M. and Sabatini, D.M. (2005) Growing roles forthe mTOR pathway. Curr. Opin. Cell Biol., 17, 596–603.

10. Sengupta, S., Peterson, T.R. and Sabatini, D.M. (2010) Regulation of themTOR complex 1 pathway by nutrients, growth factors, and stress. Mol.

Cell, 40, 310–322.

11. Jung, C.H., Jun, C.B., Ro, S.H., Kim, Y.M., Otto, N.M., Cao, J.,Kundu, M. and Kim, D.H. (2009) ULK-Atg13-FIP200 complexesmediate mTOR signaling to the autophagy machinery. Mol. Biol. Cell,20, 1992–2003.

12. Kundu, M. and Thompson, C.B. (2005) Macroautophagy versusmitochondrial autophagy: a question of fate? Cell Death Differ., 12(Suppl.2), 1484–1489.

13. Kitada, T., Asakawa, S., Hattori, N., Matsumine, H., Yamamura, Y.,Minoshima, S., Yokochi, M., Mizuno, Y. and Shimizu, N. (1998)Mutations in the parkin gene cause autosomal recessive juvenileparkinsonism. Nature, 392, 605–608.

14. Valente, E.M., Abou-Sleiman, P.M., Caputo, V., Muqit, M.M., Harvey,K., Gispert, S., Ali, Z., Del Turco, D., Bentivoglio, A.R., Healy, D.G.et al. (2004) Hereditary early-onset Parkinson’s disease caused bymutations in PINK1. Science, 304, 1158–1160.

15. Narendra, D., Tanaka, A., Suen, D.F. and Youle, R.J. (2008) Parkin isrecruited selectively to impaired mitochondria and promotes theirautophagy. J. Cell Biol., 183, 795–803.

16. Narendra, D.P., Jin, S.M., Tanaka, A., Suen, D.F., Gautier, C.A., Shen, J.,Cookson, M.R. and Youle, R.J. (2010) PINK1 is selectively stabilized onimpaired mitochondria to activate Parkin. PLoS Biol., 8, e1000298.

17. Vives-Bauza, C., Zhou, C., Huang, Y., Cui, M., de Vries, R.L., Kim, J.,May, J., Tocilescu, M.A., Liu, W., Ko, H.S. et al. (2010) PINK1-dependent recruitment of Parkin to mitochondria in mitophagy. Proc. Natl

Acad. Sci. USA, 107, 378–383.

18. Geisler, S., Holmstrom, K.M., Skujat, D., Fiesel, F.C., Rothfuss, O.C.,Kahle, P.J. and Springer, W. (2010) PINK1/Parkin-mediated mitophagy isdependent on VDAC1 and p62/SQSTM1. Nat. Cell Biol., 12, 119–131.

19. Kraft, C., Peter, M. and Hofmann, K. (2010) Selective autophagy:ubiquitin-mediated recognition and beyond. Nat. Cell Biol., 12, 836–841.

20. King, M.P. and Attardi, G. (1989) Human cells lacking mtDNA:repopulation with exogenous mitochondria by complementation. Science,246, 500–503.

21. Pallotti, F., Baracca, A., Hernandez-Rosa, E., Walker, W.F., Solaini, G.,Lenaz, G., Melzi D’Eril, G.V., Dimauro, S., Schon, E.A. and Davidson,M.M. (2004) Biochemical analysis of respiratory function in cybridcell lines harbouring mitochondrial DNA mutations. Biochem. J., 384,287–293.

22. Gilkerson, R.W., Schon, E.A., Hernandez, E. and Davidson, M.M. (2008)Mitochondrial nucleoids maintain genetic autonomy but allow forfunctional complementation. J. Cell Biol., 181, 1117–1128.

23. Santra, S., Gilkerson, R.W., Davidson, M. and Schon, E.A. (2004)Ketogenic treatment reduces deleted mitochondrial DNAs in culturedhuman cells. Ann. Neurol., 56, 662–669.

24. Twig, G., Elorza, A., Molina, A.J., Mohamed, H., Wikstrom, J.D., Walzer,G., Stiles, L., Haigh, S.E., Katz, S., Las, G. et al. (2008) Fission andselective fusion govern mitochondrial segregation and elimination byautophagy. EMBO J., 27, 433–446.

25. Mizushima, N., Yoshimori, T. and Levine, B. (2010) Methods inmammalian autophagy research. Cell, 140, 313–326.

26. Kim, I. and Lemasters, J.J. (2011) Mitochondrial degradation byautophagy (mitophagy) in GFP-LC3 transgenic hepatocytes duringnutrient deprivation. Am. J. Physiol. Cell Physiol., 300, C308–C317.

27. Shvets, E. and Elazar, Z. (2009) Flow cytometric analysis of autophagy inliving mammalian cells. Methods Enzymol., 452, 131–141.

28. Degtyarev, M., De Maziere, A., Orr, C., Lin, J., Lee, B.B., Tien, J.Y.,Prior, W.W., van Dijk, S., Wu, H., Gray, D.C. et al. (2008) Akt inhibitionpromotes autophagy and sensitizes PTEN-null tumors to lysosomotropicagents. J. Cell Biol., 183, 101–116.

29. Mattiazzi, M., Vijayvergiya, C., Gajewski, C.D., DeVivo, D.C., Lenaz,G., Wiedmann, M. and Manfredi, G. (2004) The mtDNA T8993G(NARP) mutation results in an impairment of oxidativephosphorylation that can be improved by antioxidants. Hum. Mol.

Genet., 13, 869–879.30. King, M.P., Koga, Y., Davidson, M. and Schon, E.A. (1992) Defects in

mitochondrial protein synthesis and respiratory chain activity segregatewith the tRNALeu(UUR) mutation associated with mitochondrial myopathy,encephalopathy, lactic acidosis, and strokelike episodes. Mol. Cell. Biol.,12, 480–490.

31. Maro, B., Marty, M.C. and Bornens, M. (1982) In vivo and in vitro effectsof the mitochondrial uncoupler FCCP on microtubules. EMBO J., 1,1347–1352.

32. Zhang, G.J., Liu, H.W., Yang, L., Zhong, Y.G. and Zheng, Y.Z. (2000)Influence of membrane physical state on the lysosomal protonpermeability. J. Membr. Biol., 175, 53–62.

33. Schieke, S.M., Phillips, D., McCoy, J.P., Jr, Aponte, A.M., Shen, R.F.,Balaban, R.S. and Finkel, T. (2006) The mammalian target of rapamycin(mTOR) pathway regulates mitochondrial oxygen consumption andoxidative capacity. J. Biol. Chem., 281, 27643–27652.

34. Chan, N.C., Salazar, A.M., Pham, A.H., Sweredoski, M.J., Kolawa, N.J.,Graham, R.L., Hess, S. and Chan, D.C. (2011) Broad activation of theubiquitin-proteasome system by Parkin is critical for mitophagy. Hum.

Mol. Genet., 20, 1726–1737.35. Rakovic, A., Grunewald, A., Kottwitz, J., Bruggemann, N., Pramstaller,

P.P., Lohmann, K. and Klein, C. (2011) Mutations in PINK1 and Parkinimpair ubiquitination of Mitofusins in human fibroblasts. PLoS One, 6,e16746.

36. Park, J., Lee, S.B., Lee, S., Kim, Y., Song, S., Kim, S., Bae, E., Kim, J.,Shong, M., Kim, J.M. et al. (2006) Mitochondrial dysfunction inDrosophila PINK1 mutants is complemented by parkin. Nature, 441,1157–1161.

37. Suen, D.F., Narendra, D.P., Tanaka, A., Manfredi, G. and Youle, R.J.(2010) Parkin overexpression selects against a deleterious mtDNAmutation in heteroplasmic cybrid cells. Proc. Natl Acad. Sci. USA, 107,11835–11840.

38. Van Laar, V.S., Arnold, B., Cassady, S.J., Chu, C.T., Burton, E.A. andBerman, S.B. (2011) Bioenergetics of neurons inhibit the translocationresponse of Parkin following rapid mitochondrial depolarization. Hum.

Mol. Genet., 20, 927–940.39. Sterky, F.H., Lee, S., Wibom, R., Olson, L. and Larsson, N.G. (2011)

Impaired mitochondrial transport and Parkin-independent degeneration ofrespiratory chain-deficient dopamine neurons in vivo. Proc. Natl Acad.

Sci. USA, 108, 12937–12942.40. Friedberg, I., Weisman, G.A. and De, B.K. (1985) Permeability change in

transformed mouse fibroblasts caused by ionophores, and its relationshipto membrane permeabilization by exogenous ATP. J. Membr. Biol., 83,251–259.

41. Connop, B.P., Thies, R.L., Beyreuther, K., Ida, N. and Reiner, P.B. (1999)Novel effects of FCCP [carbonyl cyanide p-(trifluoromethoxy)phenylhydrazone] on amyloid precursor protein processing.J. Neurochem., 72, 1457–1465.

42. Garcia, J.J., Ogilvie, I., Robinson, B.H. and Capaldi, R.A. (2000)Structure, functioning, and assembly of the ATP synthase in cells frompatients with the T8993G mitochondrial DNA mutation. Comparison withthe enzyme in r0 cells completely lacking mtDNA. J. Biol. Chem., 275,11075–11081.

43. Domingues, A.F., Arduino, D.M., Esteves, A.R., Swerdlow, R.H.,Oliveira, C.R. and Cardoso, S.M. (2008) Mitochondria andubiquitin-proteasomal system interplay: relevance to Parkinson’s disease.Free Radic. Biol. Med., 45, 820–825.

44. Gegg, M.E., Cooper, J.M., Schapira, A.H. and Taanman, J.W. (2009)Silencing of PINK1 expression affects mitochondrial DNA and oxidativephosphorylation in dopaminergic cells. PLoS One, 4, e4756.


Dow


ic.oup.com/hm


ber 2021

45. Garcia-Rodriguez, L.J. (2007) Appendix 1. Basic properties ofmitochondria. Methods Cell Biol., 80, 809–812.

46. Dagda, R.K., Cherra, S.J., III, Kulich, S.M., Tandon, A., Park, D. and Chu,C.T. (2009) Loss of PINK1 function promotes mitophagy through effectson oxidative stress and mitochondrial fission. J. Biol. Chem., 284, 13843–13855.

47. Simunovic, F., Yi, M., Wang, Y., Macey, L., Brown, L.T., Krichevsky,A.M., Andersen, S.L., Stephens, R.M., Benes, F.M. and Sonntag, K.C.(2009) Gene expression profiling of substantia nigra dopamine neurons:

further insights into Parkinson’s disease pathology. Brain, 132, 1795–1809.

48. Malagelada, C., Jin, Z.H., Jackson-Lewis, V., Przedborski, S. andGreene, L.A. (2010) Rapamycin protects against neuron death in invitro and in vivo models of Parkinson’s disease. J. Neurosci., 30,1166–1175.

49. Tang, Y., Manfredi, G., Hirano, M. and Schon, E.A. (2000) Maintenanceof human rearranged mitochondrial DNAs in long-term culturedtransmitochondrial cell lines. Mol. Biol. Cell, 11, 2349–2358.


Dow


ic.oup.com/hm


ber 2021

mitochondrial autophagy in cells with mtdna mutations results from

Documents