A novel mitochondrial ubiquitin ligase plays a critical role in mitochondrial dynamics
2006; Springer Nature; Volume: 25; Issue: 15 Linguagem: Inglês
10.1038/sj.emboj.7601249
ISSN1460-2075
AutoresRyo Yonashiro, Satoshi Ishido, Shinkou Kyo, Toshifumi Fukuda, Eiji Gotō, Yohei Matsuki, Mari Ohmura‐Hoshino, Kiyonao Sada, Hak Hotta, Hirohei Yamamura, Ryoko Inatome, Shigeru Yanagi,
Tópico(s)RNA modifications and cancer
ResumoArticle27 July 2006free access A novel mitochondrial ubiquitin ligase plays a critical role in mitochondrial dynamics Ryo Yonashiro Ryo Yonashiro Laboratory of Molecular Biochemistry, School of Life Science, Tokyo University of Pharmacy and Life Science, Hachioji, Tokyo, Japan Department of Genome Science, Kobe University Graduate School of Medicine, Chuo-Ku, Kobe, Japan Search for more papers by this author Satoshi Ishido Corresponding Author Satoshi Ishido Laboratory for Infectious Immunity, RIKEN Research Center for Allergy and Immunology, Tsurumi-ku, Yokohama, Kanagawa, Japan Search for more papers by this author Shinkou Kyo Shinkou Kyo Department of Genome Science, Kobe University Graduate School of Medicine, Chuo-Ku, Kobe, Japan Search for more papers by this author Toshifumi Fukuda Toshifumi Fukuda Laboratory of Molecular Biochemistry, School of Life Science, Tokyo University of Pharmacy and Life Science, Hachioji, Tokyo, Japan Search for more papers by this author Eiji Goto Eiji Goto Department of Genome Science, Kobe University Graduate School of Medicine, Chuo-Ku, Kobe, Japan Laboratory for Infectious Immunity, RIKEN Research Center for Allergy and Immunology, Tsurumi-ku, Yokohama, Kanagawa, Japan Search for more papers by this author Yohei Matsuki Yohei Matsuki Department of Genome Science, Kobe University Graduate School of Medicine, Chuo-Ku, Kobe, Japan Laboratory for Infectious Immunity, RIKEN Research Center for Allergy and Immunology, Tsurumi-ku, Yokohama, Kanagawa, Japan Search for more papers by this author Mari Ohmura-Hoshino Mari Ohmura-Hoshino Laboratory for Infectious Immunity, RIKEN Research Center for Allergy and Immunology, Tsurumi-ku, Yokohama, Kanagawa, Japan Search for more papers by this author Kiyonao Sada Kiyonao Sada Department of Genome Science, Kobe University Graduate School of Medicine, Chuo-Ku, Kobe, Japan Search for more papers by this author Hak Hotta Hak Hotta Department of Genome Science, Kobe University Graduate School of Medicine, Chuo-Ku, Kobe, Japan Search for more papers by this author Hirohei Yamamura Hirohei Yamamura Department of Genome Science, Kobe University Graduate School of Medicine, Chuo-Ku, Kobe, Japan Search for more papers by this author Ryoko Inatome Ryoko Inatome Laboratory of Molecular Biochemistry, School of Life Science, Tokyo University of Pharmacy and Life Science, Hachioji, Tokyo, Japan Department of Genome Science, Kobe University Graduate School of Medicine, Chuo-Ku, Kobe, Japan Search for more papers by this author Shigeru Yanagi Corresponding Author Shigeru Yanagi Laboratory of Molecular Biochemistry, School of Life Science, Tokyo University of Pharmacy and Life Science, Hachioji, Tokyo, Japan PRESTO, Japan Science and Technology Agency (JST), Kawaguchi, Saitama, Japan Search for more papers by this author Ryo Yonashiro Ryo Yonashiro Laboratory of Molecular Biochemistry, School of Life Science, Tokyo University of Pharmacy and Life Science, Hachioji, Tokyo, Japan Department of Genome Science, Kobe University Graduate School of Medicine, Chuo-Ku, Kobe, Japan Search for more papers by this author Satoshi Ishido Corresponding Author Satoshi Ishido Laboratory for Infectious Immunity, RIKEN Research Center for Allergy and Immunology, Tsurumi-ku, Yokohama, Kanagawa, Japan Search for more papers by this author Shinkou Kyo Shinkou Kyo Department of Genome Science, Kobe University Graduate School of Medicine, Chuo-Ku, Kobe, Japan Search for more papers by this author Toshifumi Fukuda Toshifumi Fukuda Laboratory of Molecular Biochemistry, School of Life Science, Tokyo University of Pharmacy and Life Science, Hachioji, Tokyo, Japan Search for more papers by this author Eiji Goto Eiji Goto Department of Genome Science, Kobe University Graduate School of Medicine, Chuo-Ku, Kobe, Japan Laboratory for Infectious Immunity, RIKEN Research Center for Allergy and Immunology, Tsurumi-ku, Yokohama, Kanagawa, Japan Search for more papers by this author Yohei Matsuki Yohei Matsuki Department of Genome Science, Kobe University Graduate School of Medicine, Chuo-Ku, Kobe, Japan Laboratory for Infectious Immunity, RIKEN Research Center for Allergy and Immunology, Tsurumi-ku, Yokohama, Kanagawa, Japan Search for more papers by this author Mari Ohmura-Hoshino Mari Ohmura-Hoshino Laboratory for Infectious Immunity, RIKEN Research Center for Allergy and Immunology, Tsurumi-ku, Yokohama, Kanagawa, Japan Search for more papers by this author Kiyonao Sada Kiyonao Sada Department of Genome Science, Kobe University Graduate School of Medicine, Chuo-Ku, Kobe, Japan Search for more papers by this author Hak Hotta Hak Hotta Department of Genome Science, Kobe University Graduate School of Medicine, Chuo-Ku, Kobe, Japan Search for more papers by this author Hirohei Yamamura Hirohei Yamamura Department of Genome Science, Kobe University Graduate School of Medicine, Chuo-Ku, Kobe, Japan Search for more papers by this author Ryoko Inatome Ryoko Inatome Laboratory of Molecular Biochemistry, School of Life Science, Tokyo University of Pharmacy and Life Science, Hachioji, Tokyo, Japan Department of Genome Science, Kobe University Graduate School of Medicine, Chuo-Ku, Kobe, Japan Search for more papers by this author Shigeru Yanagi Corresponding Author Shigeru Yanagi Laboratory of Molecular Biochemistry, School of Life Science, Tokyo University of Pharmacy and Life Science, Hachioji, Tokyo, Japan PRESTO, Japan Science and Technology Agency (JST), Kawaguchi, Saitama, Japan Search for more papers by this author Author Information Ryo Yonashiro1,2, Satoshi Ishido 3, Shinkou Kyo2, Toshifumi Fukuda1, Eiji Goto2,3, Yohei Matsuki2,3, Mari Ohmura-Hoshino3, Kiyonao Sada2, Hak Hotta2, Hirohei Yamamura2, Ryoko Inatome1,2 and Shigeru Yanagi 1,4 1Laboratory of Molecular Biochemistry, School of Life Science, Tokyo University of Pharmacy and Life Science, Hachioji, Tokyo, Japan 2Department of Genome Science, Kobe University Graduate School of Medicine, Chuo-Ku, Kobe, Japan 3Laboratory for Infectious Immunity, RIKEN Research Center for Allergy and Immunology, Tsurumi-ku, Yokohama, Kanagawa, Japan 4PRESTO, Japan Science and Technology Agency (JST), Kawaguchi, Saitama, Japan *Corresponding authors: Laboratory of Molecular Biochemistry, School of Life Science, Tokyo University of Pharmacy and Life Science, 1432-1 Horinouchi, Hachioji, Tokyo 192-0392, Japan. Tel.: +81 42 676 7146; Fax: +81 42 676 4149; E-mail: [email protected] or Tel.: +81 45 503 7022; Fax: +81 45 503 7021; E-mail: [email protected] The EMBO Journal (2006)25:3618-3626https://doi.org/10.1038/sj.emboj.7601249 PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info In this study, we have identified a novel mitochondrial ubiquitin ligase, designated MITOL, which is localized in the mitochondrial outer membrane. MITOL possesses a Plant Homeo-Domain (PHD) motif responsible for E3 ubiquitin ligase activity and predicted four-transmembrane domains. MITOL displayed a rapid degradation by autoubiquitination activity in a PHD-dependent manner. HeLa cells stably expressing a MITOL mutant lacking ubiquitin ligase activity or MITOL-deficient cells by small interfering RNA showed an aberrant mitochondrial morphology such as fragmentation, suggesting the enhancement of mitochondrial fission by MITOL dysfunction. Indeed, a dominant-negative expression of Drp1 mutant blocked mitochondrial fragmentation induced by MITOL depletion. We found that MITOL associated with and ubiquitinated mitochondrial fission protein hFis1 and Drp1. Pulse–chase experiment showed that MITOL overexpression increased turnover of these fission proteins. In addition, overexpression phenotype of hFis1 could be reverted by MITOL co-overexpression. Our finding indicates that MITOL plays a critical role in mitochondrial dynamics through the control of mitochondrial fission proteins. Introduction Ubiquitination is a post-translational modification in which a small conserved peptide, ubiquitin, is appended to target proteins in the cell through a series of complex enzymatic reactions (Hershko and Ciechanover, 1992). The ubiquitin–proteasome system plays an important role in controlling the levels of various cellular proteins and regulates basic cellular processes such as cell cycle progression, signal transduction, cell transformation and immune recognition (Hershko and Ciechanover, 1998). This system is also involved in the protein quality control, which maintains the health of the cell (Sitia and Braakman, 2003). E3 ubiquitin ligases are a large family of proteins that can be classified into three major structurally distinct types: N-end rule E3, E3 containing the HECT (Homology to E6AP C-Terminus) domain and E3 with the RING (Really Interesting New Gene) finger, including its derivatives, the U-Box and the PHD (Plant Homeo-Domain) (Jackson et al, 2000; Pickart, 2001; Patterson, 2002). PHD constitutes a widely distributed subfamily of zinc fingers and had been initially believed to be involved in chromatin-mediated transcriptional regulation. But recent studies have demonstrated that various proteins containing PHD function as E3 ubiquitin ligases including MEKK1 and Doa10 (Coscoy and Ganem, 2003). Subsequently, several PHD-containing viral proteins have been identified that promote immune evasion by downregulating proteins that govern immune recognition (Ishido et al, 2000a). For example, Kaposi's sarcoma-associated herpesvirus encodes two proteins, MIR (modulator of immune recognition) 1 and 2, which have been shown to contain PHD and function as E3 ubiquitin ligases for immune recognition-related molecules such as MHC class I (Ishido et al, 2000b; Means et al, 2002). Thus, these viral regulators lead to ubiquitination of their targets by functioning as E3 ubiquitin ligases that require the PHD motif. We have recently demonstrated that human genome also encodes a novel PHD-containing protein that functions as an E3 ubiquitin ligase designated as c-MIR (Goto et al, 2003). Furthermore, in a genome screen for related mammalian proteins containing PHD like c-MIR, we discovered a novel membrane-bound E3 ubiquitin ligase designated MITOL, which is specifically localized in mitochondria. Mitochondria are the center of cellular energy production and essential metabolic reactions, and are involved in the induction of apoptosis by release of cytochrome c. Mitochondria are remarkably dynamic organelles. Time-lapse microscopy of living cells reveals that mitochondria undergo constant migration and morphological changes. The mitochondrial fission and fusion machinery plays an essential role in the dynamics, division, distribution and morphology of the organelle (Yaffe, 1999; Jensen et al, 2000; Griparic and van der Bliek, 2001; Shaw and Nunnari, 2002). Evolutionarily conserved large GTPases Dnm1/Drp1/Dlp1 and Fzo1/mitofusin are involved in mitochondrial fission and fusion, respectively (Hales and Fuller, 1997; Hermann et al, 1998; Labrousse et al, 1999; Sesaki and Jensen, 1999). Mammalian protein hFis1 is the orthologue of the yeast Fis1p known to participate in yeast mitochondrial division (James et al, 2003). However, their regulatory mechanisms are largely unknown. In this study, we demonstrate that MITOL is involved in the regulation of mitochondrial dynamics through the downregulation of mitochondrial fission proteins. We discuss a crucial role of MITOL in mitochondrial biology. Results MITOL is a novel mitochondria-specific E3 ubiquitin ligase In a human genome screening for MIR family, we identified a novel mitochondrial ubiquitin ligase and named it MITOL. Figure 1A shows the amino-acid sequence of MITOL. MITOL contains a PHD motif at its N-terminus and predicted four-transmembrane domains. We first examined whether PHD in MITOL exhibited E3 ubiquitin ligase activity. For this purpose, two constructs, GST-fused PHD (2–72 amino acids of MITOL) and GST-fused PHD mutant (CS mutant; C65S, C68S), which lacks Zn-binding ability, were generated. In vitro ubiquitination assay using lysate of rabbit reticulocytes containing E1 and E2s revealed a clear polyubiquitin signal in GST-PHD but not in GST or GST-PHD CS mutant (Figure 1B). This indicated that PHD in MITOL has an E3 ubiquitin ligase activity. Northern blot analysis showed the ubiquitous expression of MITOL in various human tissues (Figure 1C). To ascertain expression of MITOL a specific anti-MITOL antibody was raised against a peptide corresponding to amino acids 257–279 of MITOL. Western blot analysis with anti-MITOL antibody demonstrated the expression of MITOL as an ≈30 kDa band in both COS-7 and HeLa cells, but not with control, preimmune serum (Figure 1D). In addition, an enhanced MITOL signal in MITOL expression vector-transfected cells verified the specificity of this antibody and the position of MITOL protein band. Figure 1.Identification of MITOL, a novel membrane-bound E3 ubiquitin ligase. (A) Amino-acid sequence of human MITOL. PHD motif and predicted four-transmembrane domains are shown by an underline and shadow lines, respectively. Anti-MITOL antibody was raised against the peptide indicated by dots above the amino acids. (B) E3 ubiquitin ligase activity in GST-fused MITOL-PHD, but not CS mutant (C65S, C68S). In vitro ubiquitin ligase assay using purified GST, GST-PHD and GST-PHD CS mutants was performed as described in Materials and methods. (C) Northern blot analysis of MITOL in various human tissues. A full-length human MITOL cDNA was used as a probe. (D) Expression of MITOL protein in COS-7 and HeLa cell lines. Lysates of COS-7, HeLa and MITOL expression vector-transfected HeLa cells were immunoblotted with control preimmune or immune serum containing anti-MITOL antibody. Download figure Download PowerPoint In order to examine subcellular distribution of MITOL in HeLa cells, immunohistochemical analysis using anti-MITOL antibody was performed and is shown in Figure 2A (upper panels). Endogenous MITOL was colocalized with mitochondrial marker MitoTracker, suggesting that MITOL is a mitochondrial protein. Likewise, HeLa cells expressing Myc-tagged MITOL were immunostained with anti-Myc antibody and MitoTracker. Myc-tagged MITOL was also colocalized with mitochondria (Figure 2A, lower panels). To confirm the mitochondrial localization of MITOL, cytosolic and mitochondrial fractions of normal HeLa or HeLa cells expressing Myc-tagged MITOL were isolated and MITOL localization was determined by immunoblotting using anti-MITOL (Figure 2B, left) or anti-Myc antibodies (Figure 2B, right). As expected, MITOL was localized in the mitochondrial fraction. As the mitochondrial fraction is known to contain a part of endoplasmic reticulum (ER) proteins, mitochondrial and ER-rich fractions were separated and MITOL localization was examined using anti-PDI (ER marker) and anti-Tom20 antibodies (mitochondria marker). As shown in Figure 2C, MITOL was not detected in the ER-rich fraction, indicating the specific localization of MITOL in mitochondria. Carbonate extraction assay showed that MITOL behaved like an integral membrane protein. This result was consistent with predicted four-transmembrane structure (Figure 2D). We next examined whether MITOL localized at the outer or inner membrane of mitochondria. Purified intact mitochondria prepared from HeLa cells were treated with trypsin, which attacked the cytosolic domain of the outer membrane proteins. Treatment with trypsin did not affect the inner membrane protein Tim23, whereas it degraded MITOL as well as the outer membrane protein Tom20 (Figure 2E). Mitochondrial outer membrane proteins were removed from mitochondria by nonionic detergent digitonin treatment. The digitonin solubility of MITOL was consistent with that of Tom20 but not Tim23 (Figure 2E). To determine the topology of the N-terminus, trypsin protection assay on mitochondria isolated from cells expressing an N-terminally HA-tagged MITOL was performed. As shown in Figure 2E (right panel), the N-terminal PHD domain rapidly disappeared from mitochondria by trypsin treatment. These results suggested that MITOL was localized at mitochondrial outer membranes and the N-terminal PHD domain is exposed to the cytoplasm. Figure 2.Specific localization of MITOL in the outer mitochondrial membrane. (A) Colocalization of MITOL with mitochondria. (Upper panels) HeLa cells were immunostained with anti-MITOL (green) antibody and MitoTracker (red). (Lower panels) HeLa cells expressing Myc-tagged MITOL were immunostained with anti-Myc antibody (green) and MitoTracker (red). (B) Subcellular fractionation indicates the localization of MITOL in mitochondria. Whole lysates (W) and cytosolic (C) and mitochondrial fractions (M) isolated from HeLa (left) or HeLa cells expressing Myc-MITOL (right) were immunoblotted with anti-MITOL, anti-Myc, anti-tubulin or anti-Tim23 mitochondria marker antibodies. (C) MITOL is not localized in ER. The mitochondrial fraction (M) and the ER-rich fraction (ER) isolated from HeLa cells were immunoblotted with anti-MITOL, anti-PDI (ER marker) or anti-Tom20 (mitochondria marker) antibodies. (D) Carbonate extraction assay indicates that MITOL is an integral membrane protein. Whole lysates (W), supernatant (S; peripheral protein) and pellet (P; integral protein) fractions after carbonate extraction from isolated mitochondria were immunoblotted with anti-MITOL, anti-Tom20 or anti-cytochrome c antibodies. (E) Sensitivity of MITOL to trypsin or digitonin treatment indicates MITOL localization in the mitochondrial outer membrane. Purified mitochondria from HeLa cells (left and middle panels) or cells expressing an N-terminally HA-tagged MITOL (right panel) were treated with trypsin or digitonin, and disappearance of MITOL along with mitochondrial markers was monitored by immunoblotting with anti-MITOL, anti-HA, anti-Tim23 or anti-Tom20 antibodies. Download figure Download PowerPoint Rapid degradation of MITOL by autoubiquitination activity To understand the metabolic steady-state level of MITOL expression, the effect of protein synthesis inhibitor cycloheximide (CHX) on MITOL expression was examined. As shown in Figure 3A (left panel), CHX treatment downregulated MITOL expression and this was blocked by addition of proteasome inhibitor MG132. A similar result was obtained in HeLa cells expressing FLAG-MITOL (Figure 3A, right panel). The half-life of FLAG-MITOL was about 60 min and it became difficult to detect the MITOL protein band after 3 h treatment with CHX. On the other hand, treatment with MG132 blocked the CHX-induced downregulation of MITOL (lane 7). Equal amount of tubulin staining demonstrated that the same amount of protein was loaded in each lane. Compared to FLAG-MITOL, endogenous MITOL showed a slow degradation after CHX treatment, suggesting that endogenous MITOL may be regulated by an unknown modification. Figure 3.Rapid degradation of MITOL by autoubiquitination activity. (A) Effect of protein synthesis inhibitor CHX on MITOL protein expression. HeLa cells (left panel) or HeLa cells expressing FLAG-MITOL WT (right panel) were treated with or without CHX (10 μg/ml) for indicated times in the presence or absence of MG132 (50 μM) and cell lysates were immunoblotted with anti-MITOL, anti-FLAG or anti-tubulin antibodies. (B) No degradation of MITOL CS mutant. HeLa cells or HeLa cells expressing FLAG-MITOL WT or FLAG-MITOL CS mutant were treated with or without CHX (10 μg/ml) for 3 h and immunoblotted with anti-FLAG or anti-tubulin antibodies. (C) Polyubiquitination of MITOL WT but not MITOL CS mutant. Lysates of HeLa cells cotransfected with control vector, FLAG-MITOL WT or FLAG-MITOL CS mutant with HA-tagged ubiquitin (HA-Ub) were immunoprecipitated with anti-FLAG antibody agarose beads and immunoblotted with anti-HA or anti-FLAG antibodies. (D) Mitochondrial aggregation by MITOL CS mutant. HeLa cells expressing Myc-MITOL WT and CS mutant were stained with anti-Myc antibody and MitoTracker. Bar, 10 μm. Mitochondrial aggregation induced by MITOL WT or CS mutant was measured by counting at least 100 cells (right panel). Error bars represent s.d. n=3. (E) Accumulation of MITOL CS mutant in the insoluble fraction. Whole lysates (W) and soluble (S) and insoluble (I) fractions were isolated from HeLa cells or HeLa cells expressing Myc-MITOL WT or CS mutants and immunoblotted with anti-Myc antibody. Bars, 10 μm. (F) Electron microscopic analysis indicated severe mitochondrial damage in MITOL CS mutant-overexpressing cells. Mitochondrial morphology in control vector, MITOL WT or CS mutant-transfected cells was examined by electron microscopic analysis at 27 600-fold magnification. Download figure Download PowerPoint To determine whether ubiquitin ligase activity in MITOL-PHD was required for this degradation of MITOL, the effect of CHX treatment on the degradation of CS mutant was similarly studied. Figure 3B shows that CHX treatment failed to degrade the CS mutant. This suggested that the degradation of MITOL was PHD-dependent. We therefore investigated whether MITOL exhibited autoubiquitination activity. Immunoprecipitation of FLAG-MITOL with anti-FLAG antibody from HeLa cells coexpressing FLAG-MITOL and HA-ubiquitin revealed significant polyubiquitination of FLAG-MITOL wild type (WT) but not CS mutant (Figure 3C). This result indicated that MITOL has an autoubiquitination activity. Taken together, these results suggested that MITOL strictly controls its protein expression level by rapid degradation of MITOL through the PHD-dependent autoubiquitination activity. To investigate the physiological relevance of this rapid turnover of MITOL by autoubiquitination activity, the effects of CS mutant on mitochondrial function and morphology were examined using a transient expression system. We found that compared to MITOL WT, CS mutant significantly induced mitochondrial aggregation (Figure 3D). Statistical analysis showed that over 80% of CS mutant-expressing mitochondria formed aggregates, whereas only 10–20% of WT-expressing mitochondria were aggregated. Western blot analysis indicated an accumulation of CS mutant but not WT in the insoluble fraction of mitochondria (Figure 3E). Electron microscopic analysis revealed the mitochondrial swelling and collapse in CS mutant-expressing cells (Figure 3F). This phenomenon was not observed in WT-expressing mitochondria. This mitochondrial dysfunction may be caused by the accumulation of insoluble or unfolded CS mutant. Taken together, these results suggested that ubiquitin ligase activity of MITOL was necessary for inhibiting accumulation of degenerated MITOL, including other proteins, in mitochondria. Requirement of MITOL for mitochondrial dynamics through interaction with and ubiquitination of mitochondrial fission factors To assess the role of MITOL in mitochondria, we established several HeLa cell lines stably expressing FLAG-MITOL WT or FLAG-MITOL CS mutant. We found that the stable expression level of FLAG-MITOL WT or CS mutant was greatly lower than its transient expression (Figure 4A, right panel). This may be owing to the downregulation of FLAG-MITOL through an unknown mechanism to protect cells from toxicity, such as mitochondrial aggregation induced by overexpression of FLAG-MITOL WT or CS mutant. Indeed, no significant mitochondrial aggregation was observed in stable clones of HeLa cell line expressing FLAG-MITOL WT or CS mutant. Instead, HeLa cell lines expressing CS mutant (but not WT) frequently showed an abnormal mitochondrial morphology, such as fragmentation (Figure 4A, left panel). About 30–50% of mitochondria in CS mutant cell line revealed mitochondrial fragmentation. Other two CS mutants also showed a similar mitochondrial fragmentation. To verify this phenomenon, the effects of MITOL knockdown on mitochondrial morphology were examined. HeLa cells transfected with scramble small interfering RNA (siRNA) as control or two siRNAs against MITOL were incubated for 48 h and the expression level of endogenous MITOL was monitored by immunoblot analysis probed with anti-MITOL antibody. As shown in Figure 4B, both the siRNAs against MITOL, but not scramble, reduced MITOL expression by about 70–90%. Using this system, we compared the mitochondrial morphology between control and MITOL-deficient mitochondria and found an increase in mitochondrial fragmentation in MITOL-deficient mitochondria. Statistical analysis showed that over 50% of mitochondria were fragmented in cells transfected with siRNA1 or siRNA2. We generated other three siRNAs against MITOL and found that cells transfected with these three siRNAs showed a similar mitochondrial fragmentation (not shown). This result suggested the enhancement of mitochondrial fission by MITOL depletion. To determine whether mitochondrial fragmentation caused by overexpression of siRNA-mediated depletion of endogenous MITOL is specific to hFis1/Drp1-dependent fission, we examined the effect of dominant-negative Drp1 GTPase-deficient mutant (Drp1K38A) expression on the MITOL-deficient mitochondrial morphology. As shown in Figure 4C, after cotransfection of siRNA and Drp1K38A, the dominant-negative Drp1 mutant blocked mitochondrial fragmentation by siRNA1-mediated MITOL depletion. Statistical analysis demonstrated that Drp1 mutant induced mitochondrial elongation in over 90% of MITOL-depleted cells. These results indicated that MITOL was required for normal mitochondrial morphology and was involved in mitochondrial dynamics probably through the regulation of mitochondrial fission. Figure 4.Critical role of MITOL in mitochondrial morphology. (A) Mitochondrial fragmentation in HeLa cell lines stably expressing FLAG-MITOL CS mutant. HeLa cell lines stably expressing control vector, FLAG-MITOL WT or CS mutant were stained with MitoTracker and mitochondrial morphologies were compared. The expression levels of FLAG-MITOL after stable and transient transfection relative to mock-transfected cells were shown by immunoblotting using anti-FLAG antibody (right panel). Bars, 10 μm. (B) Aberrant mitochondrial morphology in MITOL-deficient cells. HeLa cells transfected with scramble, MITOL siRNA1 or siRNA2 were immunoblotted with anti-FLAG and anti-vinculin antibodies (right) or stained with MitoTracker (left). Bars, 10 μm. Percentages of cells showing each mitochondrial morphology were calculated from 100 scramble- or MITOL siRNA-transfected cells. Error bars represent s.d. n=3. (C) Dominant-negative Drp1 mutant blocked mitochondrial fragmentation by MITOL depletion. HeLa cells transfected with FLAG-Drp1 mutant (K38A) and siRNA1 were stained with MitoTracker and anti-FLAG antibody. An arrowhead and asterisks indicate siRNA1-mediated MITOL-depleted cells with or without Drp1 mutant expression, respectively. Bars, 10 μm. Percentages of cells showing each mitochondrial morphology were calculated from 100 MITOL-depleted cells with or without Drp1 mutant expression. Error bars represent s.d. n=3. Download figure Download PowerPoint Experiments with siRNA-mediated knockdown of MITOL suggested that mitochondrial fission factors such as hFis1 or Drp1 may be substrate for MITOL. To explore this possibility, we investigated whether MITOL could ubiquitinate hFis1. As shown in Figure 5, MITOL WT, but not CS mutant, was found to ubiquitinate hFis1. In addition, co-immunoprecipitation assay revealed that MITOL associated with hFis1 (Figure 5B). To examine the role of MITOL in the control of endogenous hFis1, adenovirus-based constructs expressing MITOL WT and CS mutant were generated. Compared to cells expressing LacZ as control, endogenous hFis1 was decreased in cells expressing MITOL WT. On the other hand, an increase in endogenous hFis1 was detected in cells expressing MITOL CS or MITOL-deficient cells by siRNA1 (Figure 5C). CHX treatment suggested the delayed degradation of endogenous hFis1 in MITOL-deficient cells. To further demonstrate MITOL-dependent turnover of hFis1, pulse–chase experiment was performed. As shown in Figure 5D and E, MITOL overexpression promoted hFis1 turnover and this effect was inhibited by MG132 treatment. Immunoblot analysis probed with anti-FLAG antibody indicated that an equal amount of hFis1 was immunoprecipitated (not shown). Thus, MITOL may control the protein expression level of hFis1 through the ubiquitin–proteasome pathway. Indeed, overexpression of hFis1 causes mitochondrial fragmentation (Figure 6A). Therefore, we next tested whether overexpression phenotype of hFis1 can be reverted by co-overexpression of MITOL. Expectedly, co-overexpression of MITOL WT, but not CS mutant, blocked hFis1-induced mitochondrial fragmentation (Figure 6B and C). Coexpression of MITOL CS mutant induced mitochondrial aggregation consisting of fragmented mitochondria. Figure 5.MITOL associates with and ubiquitinates mitochondrial fission protein hFis1. (A) Ubiquitination of hFis1 by MITOL. Lysates of HeLa cells transfected with indicated vectors were immunoprecipitated with anti-FLAG antibody and immunoprecipitates were immunoblotted with anti-HA or anti-FLAG antibodies. Whole lysates were immunoblotted with anti-Myc antibody. (B) Interaction of MITOL with hFis1. Lysates of HeLa cells transfected with indicated vectors were immunoprecipitated with anti-Myc antibody and immunoprecipitates were immunoblotted with anti-FLAG or anti-Myc antibodies. To demonstrate the specificity of the co-immunoprecipitation, anti-HA antibody was used as a negative control. Whole lysates were immunoblotted with anti-FLAG antibody to confirm the expression of FLAG-hFis1. (C) Accumulation of endogenous hFis1 by MITOL dysfunction. Lysates of HeLa cells
Referência(s)