Parkin‐mediated ubiquitylation redistributes MITOL/March5 from mitochondria to peroxisomes

• DOI: 10.15252/embr.201947728
• Publication Date: 2019-11-28T15:38:18Z
• Authors: Fumika Koyano; Koji Yamano; Hidetaka Kosako; Yoko Kimura; Mayumi Kimura; Yukio Fujiki; Keiji Tanaka; Noriyuki Matsuda
• Keywords: Peroxisome proliferator

Article10 October 2019Open Access Transparent process Parkin-mediated ubiquitylation redistributes MITOL/March5 from mitochondria to peroxisomes Fumika Koyano Ubiquitin Project, Tokyo Metropolitan Institute of Medical Science, Tokyo, Japan Search for more papers by this author Koji Yamano Hidetaka Kosako Division Cell Signaling, Fujii Memorial Sciences, Tokushima University, Tokushima, Yoko Kimura Department Agriculture Graduate School Integrated Science and Technology, Shizuoka Shizuoka, Mayumi Yukio Fujiki Bioregulation, Kyushu Higashi-ku, Fukuoka, Keiji Tanaka Laboratory Protein Metabolism, Noriyuki Matsuda Corresponding Author [email protected] orcid.org/0000-0001-8199-952X Information Koyano1, Yamano1, Kosako2, Kimura3, Kimura1, Fujiki4, Tanaka5 *,1 1Ubiquitin 2Division 3Department 4Medical 5Laboratory *Corresponding author. Tel: [phone]; Fax: 3152; E-mail: EMBO Reports (2019)20:e47728https://doi.org/10.15252/embr.201947728 PDFDownload PDF article text main figures. Peer ReviewDownload a summary the editorial decision including letters, reviewer comments responses feedback. ToolsAdd favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info Abstract Ubiquitylation outer mitochondrial membrane (OMM) proteins is closely related onset familial Parkinson's disease. Typically, reduction in potential results OMM proteins, which are then targeted proteasomal mitophagic degradation. The role with non-degradative fates, however, remains poorly understood. In study, we find that E3 ubiquitin ligase translocates depolarized following mitophagy stimulation. This unusual redistribution mediated peroxins (peroxisomal biogenesis factors) Pex3/16 requires activity Parkin, ubiquitylates K268 MITOL C-terminus, essential p97/VCP-dependent extraction MITOL. These findings imply directs peroxisomal translocation upon stimulation reveal novel as sorting signal allows certain specialized escape damaged mitochondria. Synopsis Mitochondrial at Parkin required ubiquitylated PINK1/Parkin-dependent manner response depolarization. Ubiquitylated instead undergoing Specific (Pex3 Pex16) p97/VCP peroxisomes. may affect size abundance. Introduction well-organized organelle. When decreased, PINK1 causal gene products inherited disease, play pivotal roles selective degradation via proteasome autophagy pathway (the called mitophagy). Ser/Thr kinase, an RBR-type (ubiquitin ligase) catalyzes conjugation protein substrates ubiquitin-adducted E2 (ubiquitin-conjugating enzyme) 1-9. accumulates on subsequently phosphorylates Ser65 both ubiquitin-like (Ubl) domain 10-15. Through interaction phosphorylated ubiquitin, also becomes phosphorylated, triggers auto-activation subsequent conformational changes 16-18. various activated leads elimination 19-31. RING-type four transmembrane domains integrated into 32-34. It has been reported regulates morphology association fission factors, Fis1 Drp1 34. Indeed, siRNA-mediated knockdown resulted fragmentation. known ubiquitylate mitofusin2 (MFN2), tethering endoplasmic reticulum (ER) 33, 35, 36. investigated whether E3(s) cooperate eliminate 37, noticed showed ectopic localization during ubiquitylation. finding interest disconnected had not previously reported. found overexpressed endogenous translocated Notably, was completely dependent triggering manner. Furthermore, peroxin-dependent observed, suggesting p97/VCP–peroxin axis Results depolarization Parkin-dependent To understand molecular details when C-terminal HA-tagged Flag-Parkin were transiently expressed HeLa cells. Cells treated or without uncoupler CCCP (carbonyl cyanide m-chlorophenylhydrazone) stained anti-HA anti-Tom20/TOMM20 antibodies (Fig 1A). Without treatment, MITOL-HA overlapped Tom20, indicating its localization. Interestingly, after h large population clustered dot-like structures did co-localize assumed localized another organelle rather than cytosol. clarify stimulation, first performed co-immunostaining experiments using markers. MITOL-HA-positive coincident immunostaining patterns Sec61β (ER marker) LAMP1 (lysosome marker; Fig 1B C), but largely catalase (peroxisome treatment 1D F). Next, co-localization between each marker assessed Pearson correlation coefficient. quantitative analysis supported our aforementioned Tom20 significantly decreased 1E, columns 1 2). Although small percentage merged absence markedly increased 7 8). Peroxisomal PMP34 fused FusionRed C-terminus (PMP34-FusionRed) co-localized MITOL-GFP Parkin-expressing cells 1G), confirmed 1H). addition, dissipation unaffected addition 3Flag epitope (Appendix S1A). exposed potassium ionophore valinomycin (another type uncoupler) exhibited co-localization, artifact S1B). indicate moves tested possibility halfway point Following extended stably expressing HA-Parkin 3Flag-MITOL Pex14 S2A), S2B), S2C: because substantially degraded > 12 h, Tom20-positive selected be shown). peaked within dispersed over time. However, even no evidence Sec61β. Figure 1. loss A–C. formed treatment. 15 μM subjected immunocytochemistry. (mitochondrial A) signals well whereas MITOL-positive dots B), (lysosomal C) Scale bars, 10 μm. D. Higher magnification images boxed regions shown panel. Arrowheads representative examples catalase. E. Correlation statistics Sec61β, LAMP1, Dots individual coefficient data points. box-plots, center lines medians, box limits 25th 75th percentiles determined R software package, whiskers extend 1.5 times interquartile range percentiles. Means number samples X-axis, respectively. Statistical significance calculated one-tailed Welch's t-test. F. line graph shows scan fluorescence through three (red bar D) clearly indicates (green line) (magenta line). G. (PMP) 34-FusionRed H. PMP34-FusionRed. Download figure PowerPoint impaired For purpose, transfected lack functional PARK2 genomic mutation 4, 38. As expected, moved GFP-Parkin merge 2A, bottom panel). contrast, retained lacking expression upper Valinomycin-treated same phenomena S1C), 2B). relocation 2. move peroxisomes, Parkin. Wild-type MITOL-HA, immunocytochemistry anti-Tom20 antibodies. presence GFP-Parkin. plasmid pEGFP-C1 vector, ± MG132 immunoblotted anti-Flag, anti-ubiquitin, anti-tubulin Black arrowheads panel mono-ubiquitin middle panel, total cellular amount dramatically reduced Immunoblotting combined fractionation (MFN2) underwent rapid particular 3,000 g pellet (mitochondria-rich fractions). Cytochrome c oxidase subunit 2 (MTCO2, inner protein) 24 contrast those two minimal. Note chemical apoptosis inhibitor Z-VAD-FMK (10 μM) added along prevent cell death. Quantification 3Flag-MITOL, MFN2, MTCO2 levels PNS fraction + indicated times. Data represent mean fold change s.e.m. relative untreated biological replicates. Pre-existing doxycycline induce expression, washed fresh medium stop synthesis new After immunostained anti-Pex14 protein), anti-Hsp60 sought demonstrate pre-existing depolarization, direct targeting newly synthesized simplest experiment would suggest use cycloheximide (CHX), blocks synthesis. cannot CHX depends accumulation thus, block consequently impede translocation/activation 39. Instead CHX, utilized induction/repression system. pTRE3G-3Flag-MITOL pCMV-Tet3G plasmids. Before repressed observed 2F, top induced 2F). repeatedly remove induction (i.e., synthesis) additional h. Examination these revealed 2F), We wanted trivial (i) exists distinct organellar states, one predominantly smaller grouping (ii) very (within h) mitochondria-localized leaves peroxisome-localized dominant grouping. could lead erroneous conclusion If true, should decrease rapidly; level drastically altered rapidly 2C). monitored period (6 conjunction analysis. Mitofusin2 22, 2D E), mitochondria-rich fractions. MTCO2, 40, E). persisted throughout 6 only slight confirming decreased. modest it seen MFN2 destined Ub- mitophagy-mediated Given mitochondria-enriched contains some (as 11), consistent hypothesis migrates occurs downstream Because ATG5 indispensable progression autophagy, knockout used investigate inhibition affects analyzed immunocytochemistry, PMP70 similar wild-type EV1). does further refute under steady-state conditions apparent CCCP-induced mitophagy. Click here expand figure. EV1. Autophagy prerequisite peroxisomesThe anti-Flag anti-PMP70 cells, (including mitophagy) Collectively, unlikely directly treatment; rather, support analyze dependency, examined described above. (C431S) mutant catalytic Cys critical mutated 3A, panels), activity. redistribution. (KO) 41, despite lower substantiated 3C). determine sufficient transition SH-SY5Y human neuroblastoma expresses immunofluorescence (note included incubation sensitive detach easily). Appendix S3, once again redistributed frequency appeared overexpression think difference arises experimental conditions. minor damaged, physiological manage them. causes all mitochondria, result, overwhelmed. conditions, might insufficient assist molecules do observe exogenous expression. thus conclude dissipation, 3. Both C431S mutant, CCCP, anti-catalase catalytically inactive KO damaged. MITOL–peroxisome require own E3-inactive Cys65Ser/Cys68Ser (CS) H43W mutants HA-Parkin, CS mutant. unable specifically relocates Since E3, next itself impair activity, Cys65/Cys68 RING Ser conserved Zn-binding His43 Trp (H43W) 34, 35. Despite higher background signals, localize 3B). Quantitative cytochemical 3D), limited MITOL, other immunostaining. N-terminal anchored MitoNEET/CISD1 Tom70, Miro1/2 4A). No either 4B). matrix protein, Hsp60, likewise undergo Drp1, Fis1, Mitofusin (MFN) 1/2, MiD49, Mcl1 interact 42-44. subcellular detect av