henhaobio
Sign in

The EMBO JournalCell BiologyPeer-reviewed journal article

Novel reporter of the PINK1-Parkin mitophagy pathway identifies its damage sensor in the import gate

Julia A. ThayeriD, Jennifer D. PeterseniD, Xiaoping HuangiD, Luiza M Gruel Budet, James HawrotiD, Daniel M. RamosiD, Shiori SekineiD, Yan LiiD, Michael Emmerson WardiD, Derek P. NarendraiD

PDFDOIPreprintCode

In-depth summary

Claude Sonnet 4.6 · xhigh effort · 8 Nov 2026 · from the full text of version 1 (21 Feb 2025) · CC0 1.0

A HaloTagged MFN2 reporter enabled CRISPRi screens showing diverse mitochondrial damage activates PINK1-Parkin via membrane potential loss, with TOM retaining PINK1 when TIM23 is blocked.

How PINK1-Parkin senses mitochondrial damage remained unresolved, notably the roles of the TOM-TIM23 import path and its two driving forces (membrane potential versus the ATP-dependent PAM motor). This work adds a reporter complementary to mt-Keima and mito-QC and indicates the membrane potential is the main force, with TOM, not TIM23, holding PINK1.

Key results

  1. ShownEndogenous MFN2-Halo works as a quantitative single-cell readout of PINK1-Parkin activity, with degradation dependent on PINK1 in HeLa and, more weakly, in HEK293 cells with endogenous Parkin.

    MFN2 sgRNA caused a ∼5-fold decrease in MFN2-Halo fluorescence by flow cytometry (N = 3 experiments, p ≤ 0.0001); PINK1 KD blocked degradation; HEK293 degradation after 10 µM CCCP was lower and slower than in HeLa with mCh-Parkin. · MFN2-Halo is a sensitive single-cell reporter of…; Figure legends

  2. ShownSix FACS-based genome-wide CRISPRi screens in three reporter lines found facilitators (PINK1, TOMM7, OPTN, NPLOC4, UFD1L, glycolytic enzymes) and activators, most of them mitochondrial genes, including TIMM23.

    Screens run in duplicate with top and bottom 30% MFN2-Halo bins (6 million cells per bin); gene scores were higher in HeLa with mCh-Parkin than in HEK293; Parkin and NRF1 appeared only in the HEK293 screen. · Genome-wide CRISPRi screens with MFN2-Halo…; Glycolysis facilitates activation of the…; Genome-wide CRISPRi Screen; Figure legends

  3. ShownFunctionally diverse activators converge on loss of membrane potential (MMP) as the trigger for PINK1 accumulation; TIMM23 knockdown is the exception, stabilizing PINK1 first on high-MMP mitochondria.

    PINK1-YFP rose sharply once MMP dye fell to ∼50% of control (18 of 20 activators tested). Confocal N = 48 cells: PINK1-YFP higher on low-MMP mitochondria (p ≤ 0.0001). TIMM23 KD gave a distinct high-MMP, high-PINK1-YFP population. · Diverse forms of mitochondrial damage activate…; Figure legends

  4. ShownThe TOM complex, not TIM23, holds PINK1 on the outer membrane. TOMM5, TOMM7 and TOMM20 are needed, whereas TOMM70 is dispensable for endogenous PINK1.

    TIMM23 KD or KO (analyzed 7 to 8 days) put PINK1-YFP on mitochondria with increased pS65-Ub; TOMM22 or TOMM40 KD placed it near lipid droplets. After TIMM23 KD, CN-PAGE shifted with a TOMM20 antibody (N = 2), and with TOMM22 and TOMM40 antibodies in one of those experiments. · PINK1 is stabilized in TOM complex following…; TOMM20 and TOMM5 are required for PINK1…; Figure legends

  5. ShownGlycolytic ATP is needed for new PINK1 synthesis during OXPHOS impairment, and the PAM motor plays a supporting role in PINK1 import compared with MMP.

    10 mM 2-DG + 10 µM CCCP blocked translation (SUNset) and PINK1 accumulation; oligomycin lowered the CCCP threshold. DNAJA3 KD but not PAM16 KD stabilized endogenous PINK1; TIMM23 KO gave more PINK1-YFP than TIMM44 KO. · Glycolysis facilitates activation of the…; PAM import motor facilitates but is not required…

Methods, data and reproducibility

Endogenous MFN2-Halo flow cytometry reporter used in pooled dual-guide CRISPRi FACS screens, followed by arrayed knockdown/knockout with flow cytometry, TEM/CLEM, confocal imaging, immunoblotting, CN-PAGE and proteomics/AP-MS.

  • Whole-genome dual-guide CRISPRi screens (Weissman library), MFN2-Halo reporter, HeLa with/without mCh-Parkin and HEK293, untreated or with OXPHOS inhibitors · human (HeLa, HEK293) · n = six screens, each in duplicate; 6 million cells per sorted bin
  • Proteomics (whole cell, soluble/insoluble fractions) and PINK1-YFP AP-MS after knockdowns · human (HeLa) · n = 4 replicates/sgRNA on 1 occasion
  • TEM of mitochondrial ultrastructure after knockdown of top activators, APEX-ER marked cells · human (HeLa) · n = at least ten cells per condition
  • Confocal quantification of PINK1-YFP on high vs low MMP mitochondria (NDUFAB1, PMPCB, DNAJA3, TIMM23 knockdowns) · human (HeLa) · n = N = 48 to 91 cells per analysis
  • i3Neurons (WTC11 iPSC-derived) treated with glycolysis inhibitors and CCCP · human · n = N ≥ 6 total replicates from at least 2 differentiations
Methods & tools
MAGeCK-Vispr robust ranked algorithm, MitoCarta3.0, Spectronaut 19.5 (directDIA), Proteome Discoverer 3.1, Perseus, SciPy 1.14.0, statsmodels 0.14.1, MTSviewer, Fiji 1.54m, TrakEM2, GraphPad Prism 10, Cytek Amnis CellStream flow cytometer
Software & environment
Spectronaut 19.5, Proteome Discoverer 3.1, SciPy 1.14.0, statsmodels 0.14.1, Fiji 1.54m, GraphPad Prism 10 and Pandas/NumPy/Seaborn custom Python scripts (versions of the latter not given).
Reproducibility gaps
Screen sequencing data to be deposited in GEO (accession listed as TBD); cell lines and constructs only on request with a Material Transfer Agreement.
Limitations (4)
  • The YFP tag may block degradation of full-length PINK1, so PINK1-YFP but not endogenous PINK1 accumulated after TOM loss; this explanation is a hypothesis. · stated by the authors
  • PAM motor may facilitate PINK1 import when PINK1-YFP is overexpressed, and TOMM70 bound excess PINK1-YFP, so some exogenous-expression findings may not reflect endogenous behavior. · stated by the authors
  • TEM scoring used single transductions with at least ten cells per knockdown, and proteomics used 4 replicates on 1 occasion, limiting statistical power for ultrastructure and proteome claims. · noted in this summary
  • Mechanistic work is mostly in HeLa cells with exogenous Parkin or PINK1-YFP; neuronal data are limited to i3Neurons treated with glycolysis inhibitors, so generality to neurons beyond that is untested. · noted in this summary

Read this if you study PINK1-Parkin mitophagy or mitochondrial protein import, or want a quantitative single-cell reporter for CRISPRi screening.

Citations
32
FWCI
22.31
Field percentile
top 1%

Abstract

Damaged mitochondria can be cleared from the cell by mitophagy, using a pathway formed by the recessive Parkinsons disease genes PINK1 and Parkin. How mitochondrial damage is sensed by the PINK1-Parkin pathway, however, remains uncertain. Here, using a Parkin substrate-based reporter in genome-wide screens, we identified that diverse forms of mitochondrial damage converge on loss of mitochondrial membrane potential (MMP) to activate PINK1. Consistently, the MMP but not the presequence translocase-associated motor (PAM) import motor provided the essential driving force for endogenous PINK1 import through the inner membrane translocase TIM23. In the absence of TIM23, PINK1 arrested in the translocase of the outer membrane (TOM) during import. The energy-state outside of the mitochondria further modulated the pathway by controlling the rate of new PINK1 synthesis. Our results identify separation of PINK1 from TOM by the MMP, as the key damage-sensing switch in the PINK1-Parkin mitophagy pathway. HighlightsO_LIMFN2-Halo is a quantitative single-cell reporter of PINK1-Parkin activation. C_LIO_LIDiverse forms of mitochondrial damage, identified in whole-genome screens, activate the PINK1-Parkin pathway by disrupting the mitochondrial membrane potential (MMP). C_LIO_LIThe primary driving force for endogenous PINK1 import through the TIM23 translocase is the MMP with the PAM import motor playing a supporting role. C_LIO_LILoss of TIM23 is sufficient to stabilize PINK1 in the TOM complex and activate Parkin. C_LI

Reproducibility ledger

Nobody has reported trying to reproduce this paper yet. Did you replicate an experiment or run the code? Your report saves the next lab a week.

Sign in to file a reproducibility report.

Related papers

Similarity Map
SelectedSame fieldCross-field
THISWaters '22Eldeeb '23Conti '22Maruszczak '22Marchesan '23Ye '22
82% semantic matchSystems Biology
Christopher S Waters, Sigurd Angenent, Steven J. Altschuler+1 more·bioRxiv·Oct 7, 2022