MTP18 (also known as MTFP1), an inner mitochondrial membrane protein, plays a vital role in maintaining mitochondrial morphology by regulating mitochondrial fission. Here, we found that MTP18 functions as a mitophagy receptor that targets dysfunctional mitochondria into autophagosomes for elimination. Interestingly, MTP18 interacts with members of the LC3 (also known as MAP1LC3) family through its LC3-interacting region (LIR) to induce mitochondrial autophagy. Mutation in the LIR motif (mLIR) inhibited that interaction, thus suppressing mitophagy. Moreover, Parkin or PINK1 deficiency abrogated mitophagy in MTP18-overexpressing human oral cancer-derived FaDu cells. Upon exposure to the mitochondrial oxidative phosphorylation uncoupler CCCP, MTP18[mLIR]-FaDu cells showed decreased TOM20 levels without affecting COX IV levels. Conversely, loss of Parkin or PINK1 resulted in inhibition of TOM20 and COX IV degradation in MTP18[mLIR]-FaDu cells exposed to CCCP, establishing Parkin-mediated proteasomal degradation of outer mitochondrial membrane as essential for effective mitophagy. We also found that MTP18 provides a survival advantage to oral cancer cells exposed to cellular stress and that inhibition of MTP18-dependent mitophagy induced cell death in oral cancer cells. These findings demonstrate that MTP18 is a novel mitophagy receptor and that MTP18-dependent mitophagy has pathophysiologic implications for oral cancer progression, indicating inhibition of MTP18-mitophagy could thus be a promising cancer therapy strategy.

Mitochondria are essential organelles that play a vital role in maintaining cellular homeostasis, energy metabolism and fate. Mitochondrial dynamics refers to the balance between mitochondrial fission and mitochondrial fusion, which together maintain mitochondrial homeostasis and quality (Wang et al., 2020). Mitochondrial dynamics are essential to the health of the cell, and defects in this process lead to consequences such as metabolic diseases, neurodegenerative disorders (Flippo and Strack, 2017), aging (Annesley and Fisher, 2019) and cancers (Dai and Jiang, 2019). Mitochondria are highly dynamic, and they cope with stress by activating the fusion–fission cycle, mitophagy and controlled apoptosis for maintenance (Youle and van der Bliek, 2012). The mitofusins Mfn1 and Mfn2 in the outer mitochondrial membrane (OMM), and OPA1 in the inner mitochondrial membrane (IMM), regulate mitochondrial fusion and cristae remodeling; the Drp1 (also known as DNM1L) and human (h)Fis1 proteins in the OMM and MTP18 (also known as MTFP1) in the IMM are responsible for mitochondrial fission (Fonseca et al., 2019; Yu et al., 2019). In addition, mitochondrial autophagy selectively eliminates damaged mitochondria to maintain mitochondrial health and homeostasis. Dysfunctional mitochondria stabilize mitochondrial kinase on the OMM, thus recruiting Parkin to initiate mitophagy. Parkin promotes ubiquitination of OMM proteins, leading to their binding with autophagic adaptor proteins, which in turn interact with the autophagic protein LC3 to engulf the superfluous mitochondria. In another way, many proteins especially present on the OMM function as mitophagy receptors, directly bind with LC3 through the LC3-interacting region (LIR) to select damaged mitochondria to an autophagosome.

MTP18, a novel human nuclear-encoded protein with a molecular mass of 18 kDa, is a downstream molecule of phosphatidylinositol 3-kinase (PI3K) signaling (Tondera et al., 2005). This protein, which is localized to the IMM, plays a vital role in maintaining mitochondrial morphology and cell survival (Tondera et al., 2004). Overexpression of MTP18 changes mitochondrial morphologic structure from filamentous to fragmented spherical. Knockdown of MTP18 results in highly fused mitochondria. MTP18-regulated mitochondrial fission has been identified as an essential mediator of various pathologic conditions, including acute inflammation and kidney, heart and neuronal injury (Aung et al., 2019; Kreymerman et al., 2019; Wei et al., 2018). For example, miR-668 has been reported to inhibit MTP18, restoring mitochondrial dynamics by preventing mitochondrial fission from protecting renal tubular cells against apoptosis from ischemic acute kidney injury (Wei et al., 2018). Moreover, the oncogenic role of MTP18 has been reported in various cancer types. Interestingly, MTP18 is commonly overexpressed in hepatocellular carcinoma, and it contributes to the growth and metastasis of liver cancer cells by regulating cell cycle progression, epithelial-to-mesenchymal transition, MMP-9 activation and inhibition of apoptosis (Zhang et al., 2018). MTP18 is also frequently overexpressed in oral squamous cell carcinoma, and MTP18-dependent mitochondrial fission promotes the generation of reactive oxygen species (ROS) to activate cell cycle progression and inhibit apoptosis, leading to cancer growth and survival (Xiao et al., 2020). Moreover, MTP18 regulates doxorubicin-induced cell death in gastric cancer cells, revealing the potential role of MTP18 in chemoresistance (Aung et al., 2017). However, the role of MTP18 in controlling the mitochondrial-dependent fission that contributes to mitophagy induction, and its significance in cell death in oral squamous cell carcinoma, remains to be identified.

In the present study, we established MTP18 as an IMM receptor activating mitophagy for the clearance of dysfunctional mitochondria. Furthermore, we showed that MTP18-induced mitochondrial fission is followed by mitophagy through interaction with LC3 (also known as MAP1LC3) family proteins (hereafter, just donated LC3), which are present on autophagosomes. Interestingly, we identified that the LC3-interacting region (LIR) in MTP18 binds with LC3 to induce mitochondrial autophagy for the engulfment of dysfunctional mitochondria. Importantly, Parkin-mediated proteasomal rupture of the OMM is necessary for LC3 and MTP18 to cooperate for successful mitophagy. We also found that MTP18-mediated mitophagy permits the survival of cancer cells. In that context, we have documented how MTP18 promotes cancer cell survival and inhibits apoptosis through mitophagy, demonstrating the critical pathophysiologic implications of MTP18-dependent mitophagy in oral cancer.

MTP18 in mitochondrial fragmentation and fission

Previous studies have reported that MTP18 is involved in mitochondrial fission to control mitochondrial dynamics (Tondera et al., 2005). Here, we examined the role of MTP18 in mitochondrial fission-related proteins and mitochondrial morphology by determining assessing these after its gain and loss of function in oral squamous cell carcinoma. Our western blot analysis demonstrated that the mitochondrial fission protein Fis1 shows no significant changes in its level in MTP18-overexpressing cells; however, the level of the primary regulator for mitochondrial fission, Drp1 phosphorylated at Ser616 (Fonseca et al., 2019), increased in MTP18-overexpressing human oral cancer-derived FaDu cells (Fig. S1A,B). Furthermore, an examination of mitochondrial fragmentation by TOM20 (TOMM20) staining under confocal microscopy (Fig. 1A) showed that, compared with the filamentous mitochondria seen in a control group of FaDu cells, MTP18-overexpressing cells contained fragmented mitochondria. Interestingly, in the presence of the mitochondrial oxidative phosphorylation uncoupler carbonyl cyanide 3-chlorophenylhydrazone (CCCP), extensive mitochondrial fragmentation was observed in MTP18-overexpressing cells compared with that in control cells. The use of an ImageJ macro tool to analyze the mitochondrial skeleton and mean branch length in magnified images of the TOM20 staining (Fig. 1B) showed that, in MTP18-overexpressing cells, the mitochondrial network was fragmented and formed many small, round mitochondrial branches. Similarly, MTP18 loss of function inhibited mitochondrial fission and accelerated mitochondrial fusion in Cal33 cells, an oral and pharyngeal cancer cell line. Our data show that MTP18 knockdown diminishes the levels of mitochondrial fission protein Fis1 and phosphorylation of Drp1 at Ser616 in Cal33 cells (Fig. S1C,D). Furthermore, compared with control cells, MTP18-knockdown cells showed elongated mitochondrial morphology (Fig. 1C). In the presence of CCCP (Fig. 1D), the mitochondrial skeleton and mean branch length in MTP18-knockdown Cal33 cells demonstrated that there was a more extensive mitochondrial network than that seen in control cells, confirming that MTP18 controls mitochondrial fission in oral cancer cells.

Fig. 1.

MTP18 induces mitochondrial fission and mitophagy. (A) Mitochondrial morphology was analyzed in MTP18-overexpressing cells under confocal microscopy. After CCCP (10 μM, 6 h) treatment, the cells were immunostained for TOM20, and the outer mitochondrial membrane protein and nuclei were counterstained with DAPI. The ZOOM indicated the area highlight by the dashed box, the SKELETON shows mitochondrial branch lengths. (B) The ImageJ macro tool was used to analyze the mitochondrial skeleton morphologies. (C) Cal33 cells were transfected with siMTP18, the cells were immunostained for TOM20, and nuclei were counterstained with DAPI. The ZOOM panel shows the area highlighted by the dashed box, the SKELETON shows mitochondrial branch lengths. (D) The ImageJ macro tool was used to analyze the mitochondrial skeleton morphologies. (E–G) FaDu cells expressing MTP18 were incubated with 10 μM CCCP for 6 h, after which (E) the levels of COX IV was analyzed by western blotting, and (F) immunofluorescence of TOM20 was determined by confocal microscopy. (G) The relative fluorescence intensity was then quantified using the ImageJ macro tool. (H–J) siMTP18-FaDu cells were incubated with CCCP, (H) levels of COX IV was analyzed by western blotting and (I) immunofluorescence of TOM20 was determined by confocal microscopy. (J) The relative fluorescence intensity was then quantified. (K) The levels of LC3, and COX IV were analyzed by western blotting in CCCP-treated MTP18-FaDu cells in the presence of 5 mM wortmannin (2 h). (L,M) Immunofluorescence of TOM20 was (L) determined by confocal microscopy, and (M) the relative fluorescence intensity was then quantified. (N–P) After MTP18-FaDu cells were transfected with siBeclin 1, (N) levels of Beclin 1, COX IV and LC3 was analyzed by western blotting, and (O) immunofluorescence of TOM20 was determined by confocal microscopy, and (P) the relative fluorescence intensity was then quantified. For all panels, n=3; >25 cells counted per condition for B, D, G, J, M and P. Quantitative results are mean±s.d. ***P<0.001; **P<0.01; *P<0.05; ns, not significant (one-way ANOVA with Šidák’s post test). Scale bars: 25 μm.

Fig. 1.

MTP18 induces mitochondrial fission and mitophagy. (A) Mitochondrial morphology was analyzed in MTP18-overexpressing cells under confocal microscopy. After CCCP (10 μM, 6 h) treatment, the cells were immunostained for TOM20, and the outer mitochondrial membrane protein and nuclei were counterstained with DAPI. The ZOOM indicated the area highlight by the dashed box, the SKELETON shows mitochondrial branch lengths. (B) The ImageJ macro tool was used to analyze the mitochondrial skeleton morphologies. (C) Cal33 cells were transfected with siMTP18, the cells were immunostained for TOM20, and nuclei were counterstained with DAPI. The ZOOM panel shows the area highlighted by the dashed box, the SKELETON shows mitochondrial branch lengths. (D) The ImageJ macro tool was used to analyze the mitochondrial skeleton morphologies. (E–G) FaDu cells expressing MTP18 were incubated with 10 μM CCCP for 6 h, after which (E) the levels of COX IV was analyzed by western blotting, and (F) immunofluorescence of TOM20 was determined by confocal microscopy. (G) The relative fluorescence intensity was then quantified using the ImageJ macro tool. (H–J) siMTP18-FaDu cells were incubated with CCCP, (H) levels of COX IV was analyzed by western blotting and (I) immunofluorescence of TOM20 was determined by confocal microscopy. (J) The relative fluorescence intensity was then quantified. (K) The levels of LC3, and COX IV were analyzed by western blotting in CCCP-treated MTP18-FaDu cells in the presence of 5 mM wortmannin (2 h). (L,M) Immunofluorescence of TOM20 was (L) determined by confocal microscopy, and (M) the relative fluorescence intensity was then quantified. (N–P) After MTP18-FaDu cells were transfected with siBeclin 1, (N) levels of Beclin 1, COX IV and LC3 was analyzed by western blotting, and (O) immunofluorescence of TOM20 was determined by confocal microscopy, and (P) the relative fluorescence intensity was then quantified. For all panels, n=3; >25 cells counted per condition for B, D, G, J, M and P. Quantitative results are mean±s.d. ***P<0.001; **P<0.01; *P<0.05; ns, not significant (one-way ANOVA with Šidák’s post test). Scale bars: 25 μm.

MTP18 promotes mitophagy to eliminate dysfunctional mitochondria

MTP18 is localized to the IMM, serves as an essential mitochondrial fission protein and undergoes degradation during mitophagy (Fig. S1E,F). Mitochondrial fission is necessary for mitophagy, a process that eliminates dysfunctional mitochondria for cellular homeostasis. To understand whether increased mitochondrial fission is linked to mitophagy to eliminate damaged or depolarized mitochondria, we hypothesized that MTP18-induced mitochondrial fission might play a role in clearing damaged mitochondria through mitophagy. To study that role, we examined mitophagy status using western blotting and confocal microscopy to analyze changes in the expression of the OMM protein TOM20 and the IMM protein COX IV in MTP18-overexpressing FaDu cells. For mitophagy induction, we treated MTP18-expressing FaDu cells with CCCP and observed that COX IV levels were significantly decreased in MTP18-expressing cells compared to in control cells (Fig. 1E; Fig. S1G). Moreover, the immunostaining analysis revealed that MTP18-overexpressing FaDu cells displayed extensive mitophagy induction, as indicated by a greater decrease in TOM20 level upon CCCP exposure (Fig. 1F,G). Similarly, MTP18 knockdown by siRNA against MTP18 (siMTP18) resulted in decreased mitophagy in FaDu cells. The western blot data showed that, upon CCCP treatment, the reduction in COX IV levels in MTP18-knockdown cells were restored (Fig. 1H; Fig. S1H); however, the level of COX IV remained lower in cells exposed only to CCCP. TOM20 immunostaining analysis confirmed that mitophagy was abrogated in MTP18-knockdown cells, as indicated by the increased levels of TOM20 (Fig. 1I,J). Furthermore, the loss of MTP18 expression in Cal33 cells blocked mitophagy. The western blot data and immunostaining analysis showed that, upon CCCP treatment, the reduction in mitochondrial proteins COX IV (Fig. S1I,J) and TOM20 (Fig. S1K,L) levels was restored in Cal33 cells exposed to siMTP18.

To validate the role of MTP18 in mitochondrial degradation, we used western blotting and immunostaining to examine mitophagy status in MTP18-overexpressing cells in the presence of the autophagy inhibitor wortmannin. In a comparison of MTP18-overexpressing cells with control cells after exposure to CCCP, wortmannin treatment resulted in the inhibition of autophagy in the MTP18-overexpressing cells (Fig. 1K; Fig. S2A), as shown by a decrease in the LC3 II (lipidated LC3) level leading to accumulation of COX IV. Moreover, the reduction in TOM20 levels in MTP18-overexpressing cells after exposure to CCCP was blocked in the presence of wortmannin, but not in its absence (Fig. 1L,M). Additionally, knockdown of Beclin 1 (a key protein in the regulation of autophagy initiation) in MTP18 overexpressing FaDu cells by siRNA inhibited autophagy and subsequently reduced MTP18-induced mitochondrial clearance. As shown in Fig. 1N and Fig. S2B, COX IV accumulation was found in the Beclin 1-knockdown condition in MTP18-overexpressing cells in the presence of CCCP compared to in cells only overexpressing MTP18. Similarly, the decrease in TOM20 levels was restored in the Beclin 1-knockdown condition in MTP18-overexpressing cells (Fig. 1O,P), confirming that MTP18 induces mitophagy in oral cancer cells.

Next, we examined mitochondrial functionality during mitophagy induction in MTP18-overexpressing cells exposed to CCCP. In that setting, we treated MTP18-overexpressing FaDu cells with CCCP, and we used flow cytometry to label the total (MitoTracker Green FM) and functional (MitoTracker Red CMXRos) mitochondria. In MTP18-overexpressing FaDu cells, exposure to CCCP, compared with exposure to the empty vector, was associated with an increase in the percentage of damaged mitochondria cleared for mitophagy (Fig. 2A,B). In a further investigation into the activation of class III PI3K around depolarized mitochondria in MTP18-overexpressing FaDu cells, we transfected cells with a plasmid expressing a p40phoxPX–EGFP fusion protein (p40phox is also known as NCF4). The PX domain of p40phox PX–EGFP is known to interact specifically with the product of class III PI3K – phosphatidylinositol 3-phosphate. Thus, p40phoxPX–EGFP serves as a probe for evaluating the subcellular quantity and the distribution of PI3K-produced phosphatidylinositol 3-phosphate (Kanai et al., 2001). The MTP18-overexpressing FaDu cells transfected with p40phox were exposed to CCCP, stained for TOM20 and analyzed by confocal microscopy. Mitochondrial staining was observed to be reduced in CCCP-treated MTP18-overexpressing FaDu cells compared with that in FaDu cells transfected with empty vector. The remaining mitochondria were concentrated in perinuclear clusters that were apposed to p40phoxPX–EGFP hotspots (Fig. 2C,D). Next, we examined whether mitophagosomes fuse with lysosomes for mitochondrial degradation. To examine mitophagy flux, we assessed the colocalization of mitochondrial protein TOM20 and lysosomal-associated membrane protein 1 (LAMP1) in empty vector and MTP18-overexpressing FaDu cells. As expected, yellow fluorescence was considerably increased in CCCP-treated MTP18-FaDu cells compared with vector-FaDu cells (Fig. 2E,F), confirming that MTP18 promotes mitophagy to eliminate dysfunctional mitochondria through lysosome. Furthermore, to verify the role of MTP18 in mitophagy, we assessed the mitophagy flux by using mito-Keima in vector and MTP18-overexpressing FaDu cells. We found that MTP18-overexpressing FaDu cells showed green and red puncta upon CCCP treatment, whereas the vector cells showed only green puncta indicating the mitophagy induction Additionally, the mt-Keima red/green ratio (mitolysosome/mitochondrion) increases in MTP18-overexpressing FaDu cells upon exposure to CCCP, showing an increase in mitophagy flux (Fig. 2G,H).

Fig. 2.

MTP18 induces mitophagy by selecting dysfunctional mitochondria. (A) MTP18-FaDu cells were treated with CCCP (10 μM, 6 h) and then incubated with MitoTracker Green FM (MTG) and MitoTracker Red CMXRos for 30 min. Mitochondrial dysfunction was then determined by flow cytometry. The flow cytometry results show the proportion of healthy and damaged mitochondria with the percentage of cells in upper and lower area, respectively. (B) The percentage of cells with damaged mitochondria. (C,D) After MTP18-FaDu cells were transfected with p40phox–EGFP and treated with CCCP (10 μM, 6 h), cells were stained for TOM20 and analyzed by confocal microscopy (C). The ZOOM panel shows the area highlighted by the dashed box. The graph represents the percentage of mitochondrial clusters apposed to p40 (D). (E,F) Fusion of mitochondria and lysosomes in MTP18-FaDu cells was studied by staining for LAMP1 and TOM20 after 6 h of CCCP treatment; colocalization was determined by confocal microscopy (E). The percentage of LAMP1 colocalized with TOM20 was then quantified (F). (G,H) The vector- and MTP18-expressing FaDu cells were transfected with mKeima-Red-Mito7 for 48 h, and the expression was observed by confocal microscopy after CCCP exposure (G), the ratio of mito-Keima signal denoting 561 nm fluorescence (red) and 488 nm fluorescence (green) was then quantified (H). n=3, >25 cells counted per conditions. For all panels, n=3; >25 cells counted per condition for D, F and H. Quantitative results are mean±s.d. ***P<0.001; **P<0.01; *P<0.05; ns, not significant (one-way ANOVA with Šidák’s post test). Scale bars: 18.4 μm.

Fig. 2.

MTP18 induces mitophagy by selecting dysfunctional mitochondria. (A) MTP18-FaDu cells were treated with CCCP (10 μM, 6 h) and then incubated with MitoTracker Green FM (MTG) and MitoTracker Red CMXRos for 30 min. Mitochondrial dysfunction was then determined by flow cytometry. The flow cytometry results show the proportion of healthy and damaged mitochondria with the percentage of cells in upper and lower area, respectively. (B) The percentage of cells with damaged mitochondria. (C,D) After MTP18-FaDu cells were transfected with p40phox–EGFP and treated with CCCP (10 μM, 6 h), cells were stained for TOM20 and analyzed by confocal microscopy (C). The ZOOM panel shows the area highlighted by the dashed box. The graph represents the percentage of mitochondrial clusters apposed to p40 (D). (E,F) Fusion of mitochondria and lysosomes in MTP18-FaDu cells was studied by staining for LAMP1 and TOM20 after 6 h of CCCP treatment; colocalization was determined by confocal microscopy (E). The percentage of LAMP1 colocalized with TOM20 was then quantified (F). (G,H) The vector- and MTP18-expressing FaDu cells were transfected with mKeima-Red-Mito7 for 48 h, and the expression was observed by confocal microscopy after CCCP exposure (G), the ratio of mito-Keima signal denoting 561 nm fluorescence (red) and 488 nm fluorescence (green) was then quantified (H). n=3, >25 cells counted per conditions. For all panels, n=3; >25 cells counted per condition for D, F and H. Quantitative results are mean±s.d. ***P<0.001; **P<0.01; *P<0.05; ns, not significant (one-way ANOVA with Šidák’s post test). Scale bars: 18.4 μm.

MTP18-mediated mitochondrial fission is essential to induce mitophagy

Asymmetric mitochondrial fission generates dysfunctional mitochondrial fragments for turnover by mitophagy. The role of MTP18 in mitochondrial fission has already been established, and our study found that MTP18 induces mitophagy. To verify the role of mitochondrial fission in mitophagy, we exposed MTP18-overexpressing FaDu cells to the Drp1 inhibitor Mdivi1 (Cassidy-Stone et al., 2008). Exposure to Mdivi1 in the presence of CCCP was associated with decreased phosphorylation of Drp1 (Ser616) in MTP18-FaDu cells (Fig. 3A; Fig. S2C). Another mitochondrial fission protein, Fis1, showed no significant changes. Intriguingly, mitophagy as analyzed by COX IV levels in western blotting, revealed that MTP18-induced mitophagy was inhibited by Mdivi1 after exposure to CCCP. We also measured COX IV intensity in FLAG–MTP18-overexpressing cells by confocal microscopy. In MTP18-FaDu cells, the fluorescence intensity of COX IV was stronger in the Mdivi1 group, and COX IV intensity decreased only in the presence of CCCP (Fig. 3B). The relative fluorescence intensity of COX IV was quantified using ImageJ (Fig. 3C), suggesting that Mdivi1 significantly blocked MTP18-induced mitochondrial fission and mitophagy in MTP18-expressing FaDu cells. Furthermore, to study the effect of Drp1 on MTP18-induced mitochondrial fission and mitophagy, we transiently transfected MTP18-FaDu cells with Drp1 siRNA, which interfered with unopposed mitochondrial fission. In this setting, CCCP did not affect mitochondrial morphology, mitochondrial fission or mitophagy in the Drp1 siRNA group. In contrast, with control siRNA (siControl) there was an induction of mitochondrial fission and subsequent mitophagy in response to CCCP, as quantified by western blotting (Fig. 3D; Fig. S2D) and immunostaining analysis (Fig. 3E). Furthermore, the relative fluorescence intensity of COX IV was decreased in FLAG–MTP18-overexpressing cells in response to CCCP, whereas Drp1-knockdown MTP18-overexpressing FaDu cells had restored COX IV levels in response to CCCP (Fig. 3F). Consequently, to understand the importance of MTP18-induced mitochondrial fission in mitophagy, we used siRNA to knockdown MTP18 in Drp1-overexpressing FaDu cells. The result showed a significant decrease in the expression of Fis1 in MTP18-knockdown Drp1-overexpressing FaDu cells compared with the Drp1-overexpressing cells. (Fig. 3G; Fig. S2E). Interestingly, the decreased COX IV levels in Drp1-overexpressing FaDu cells were restored in MTP18-knockdown Drp1-overexpressing cells. TOM20 staining subsequently revealed that loss of MTP18 expression in Drp1-overexpressing FaDu cells primes the loss of fission activity, resulting in hyper-mitochondrial fusion and loss of mitophagy (Fig. 3H–J), confirming that MTP18-induced mitochondrial fission is essential for mitophagy.

Fig. 3.

MTP18 induces mitophagy through mitochondrial fission. (A–C) MTP18-FaDu cells were treated with CCCP (10 μM, 6 h) in the presence of Mdivi1 (3 h), and the levels of COX IV was analyzed by (A) western blotting and (B) confocal microscopy. (C) The relative fluorescence intensity of COX IV levels in MTP18-overexpressing cells was quantified. (D–F) MTP18-FaDu cells were transiently transfected with siDrp1 for 48 h, after that the cells were treated with 10 μM CCCP for 6 h and then analyzed for levels of COX IV by (D) western blotting and (E) confocal microscopy. (F) The relative fluorescence intensity of COX IV levels in MTP18-overexpressing cells was quantified. (G–J) After FaDu cells were co-transfected with Drp1 and siMTP18 for 48 h, the cells were incubated with 10 μM CCCP for 6 h and then analyzed for COX IV levels by (G) western blotting and (H) confocal microscopy. The ZOOM panel in H shows the area highlighted by the dashed box. The ImageJ macro tool was used to analyze (I) relative TOM20 intensity and (J) mitochondrial branch length. For all panels, n=3; >25 cells counted per condition for C, F, I and J. Quantitative results are mean±s.d. ***P<0.001; **P<0.01; *P<0.05; ns, not significant (one-way ANOVA with Šidák's post test). Scale bars: 18.4 μm.

Fig. 3.

MTP18 induces mitophagy through mitochondrial fission. (A–C) MTP18-FaDu cells were treated with CCCP (10 μM, 6 h) in the presence of Mdivi1 (3 h), and the levels of COX IV was analyzed by (A) western blotting and (B) confocal microscopy. (C) The relative fluorescence intensity of COX IV levels in MTP18-overexpressing cells was quantified. (D–F) MTP18-FaDu cells were transiently transfected with siDrp1 for 48 h, after that the cells were treated with 10 μM CCCP for 6 h and then analyzed for levels of COX IV by (D) western blotting and (E) confocal microscopy. (F) The relative fluorescence intensity of COX IV levels in MTP18-overexpressing cells was quantified. (G–J) After FaDu cells were co-transfected with Drp1 and siMTP18 for 48 h, the cells were incubated with 10 μM CCCP for 6 h and then analyzed for COX IV levels by (G) western blotting and (H) confocal microscopy. The ZOOM panel in H shows the area highlighted by the dashed box. The ImageJ macro tool was used to analyze (I) relative TOM20 intensity and (J) mitochondrial branch length. For all panels, n=3; >25 cells counted per condition for C, F, I and J. Quantitative results are mean±s.d. ***P<0.001; **P<0.01; *P<0.05; ns, not significant (one-way ANOVA with Šidák's post test). Scale bars: 18.4 μm.

The LC3-interacting region of MTP18 is necessary for it to interact with LC3 for mitophagy induction

We hypothesized that MTP18 might be a mitophagy receptor that interacts with LC3 to eliminate dysfunctional mitochondria. We used the amino acid sequence analysis to identify the LIRs of MTP18 and to determine whether they are essential for mitophagy. Three possible LIR domains (W/F/YxxL/I/V motifs) are present in MTP18. They are the amino acids 32–35, 87–90 and 157–160, two of which (32–35 and 87–90) might be conserved in the transmembrane domain of mitochondria (Tondera et al., 2005). Given that MTP18 is an IMM protein localized to the transmembrane domain, the two possible LIRs (32–35 and 87–90) found in the transmembrane domain are less likely to participate in mitophagy. We, therefore, assumed that Tyr157, Pro158, Thr159 and Val160 residues of MTP18 could be functioning as the LIR motif (Fig. 4A) for the interaction of MTP18 with LC3.

Fig. 4.

MTP18 interacts with LC3 during mitophagy. (A) The sequence of MTP18 and the predicted LC3-interacting region (LIR) sequence of MTP18 (red). (B) MTP18[WT] and MTP18[mLIR]-FaDu cells were treated with 10 μM CCCP for 6 h, immunoprecipitated with anti-FLAG, and immunoblotted with anti-LC3 antibodies. (C–F) After FaDu cells expressing vector, MTP18[WT], and MTP18[mLIR] were treated with CCCP for 6 h, cell lysates were analyzed for labeled antibodies by western blot analysis (C). FaDu cells transfected with MTP18[WT] and MTP18[mLIR] were treated with 10 μM CCCP for 6 h and then immunostained with the indicated antibodies; colocalization was subsequently assessed by confocal microscopy (D). SKELETON shows mitochondrial branch lengths. (E) The relative fluorescence intensity of COX IV expression was determined. (F) The ImageJ macro tool was used to analyze mitochondrial branch length. (G,H) After FaDu cells expressing vector, MTP18[WT], and MTP18[mLIR] were treated with CCCP (10 μM, 6 h), cells were analyzed through confocal microscopy. The ZOOM panel in G shows the area highlighted by the dashed box. (H) The LC3 dots per cell colocalized with MTP18 were quantified. (I) The percentage of LAMP1 colocalized with TOM20 in FLAG–MTP18-expressing cells was quantified. (J) The ratio of wild-type and mutant MTP18-FaDu cells expressing mito-Keima signal denoting 561 nm fluorescence (red) and 488 nm fluorescence (green) was then quantified. For B–J, n=3; >25 cells counted per condition for E–J. Quantitative results are mean±s.d. ***P<0.001; **P<0.01; *P<0.05; ns, not significant (one-way ANOVA with Šidák's post test). Scale bars: 18.4 μm.

Fig. 4.

MTP18 interacts with LC3 during mitophagy. (A) The sequence of MTP18 and the predicted LC3-interacting region (LIR) sequence of MTP18 (red). (B) MTP18[WT] and MTP18[mLIR]-FaDu cells were treated with 10 μM CCCP for 6 h, immunoprecipitated with anti-FLAG, and immunoblotted with anti-LC3 antibodies. (C–F) After FaDu cells expressing vector, MTP18[WT], and MTP18[mLIR] were treated with CCCP for 6 h, cell lysates were analyzed for labeled antibodies by western blot analysis (C). FaDu cells transfected with MTP18[WT] and MTP18[mLIR] were treated with 10 μM CCCP for 6 h and then immunostained with the indicated antibodies; colocalization was subsequently assessed by confocal microscopy (D). SKELETON shows mitochondrial branch lengths. (E) The relative fluorescence intensity of COX IV expression was determined. (F) The ImageJ macro tool was used to analyze mitochondrial branch length. (G,H) After FaDu cells expressing vector, MTP18[WT], and MTP18[mLIR] were treated with CCCP (10 μM, 6 h), cells were analyzed through confocal microscopy. The ZOOM panel in G shows the area highlighted by the dashed box. (H) The LC3 dots per cell colocalized with MTP18 were quantified. (I) The percentage of LAMP1 colocalized with TOM20 in FLAG–MTP18-expressing cells was quantified. (J) The ratio of wild-type and mutant MTP18-FaDu cells expressing mito-Keima signal denoting 561 nm fluorescence (red) and 488 nm fluorescence (green) was then quantified. For B–J, n=3; >25 cells counted per condition for E–J. Quantitative results are mean±s.d. ***P<0.001; **P<0.01; *P<0.05; ns, not significant (one-way ANOVA with Šidák's post test). Scale bars: 18.4 μm.

We performed an immunoprecipitation and colocalization study to confirm the interaction between MTP18 and LC3 in FaDu cells. Upon treatment with the mitochondrial uncoupling agent CCCP, we identified that LC3 interacted with MTP18–FLAG through FLAG immunoprecipitation and western blot analysis confirming the presence of MTP18 (Fig. 4B). Moreover, the number of LC3 dots per cell for MTP18–FLAG-positive cells was significantly higher in the CCCP-treated than in control cells (Fig. S3A,B), indicating that MTP18 interacts with LC3 to induce mitophagy. Next, we used site-directed mutagenesis to create substitution mutations in the LIR domain (Y157A and V160A) in MTP18, developing a mutant MTP18Y157A/V160A, denoted MTP18[mLIR]. After overexpression of MTP18[mLIR], which lacks the ability to bind to LC3 (Fig. 4B), we studied mitophagy in FaDu cells exposed to CCCP. We analyzed mitophagy status by western blotting (Fig. 4C; Fig. S3C) and immunofluorescence analysis (Fig. 4D,E) of COX IV levels. We observed no significant changes in COX IV levels with CCCP exposure, meaning that inhibition of mitophagy is associated with the mutation in the LIR domain. By contrast, in the presence of CCCP, the mitochondrial skeleton and mean branch length in MTP18 wild-type and MTP18 LIR mutant FaDu cells had no significant changes, confirming that LIR mutation does not affect mitochondrial fission (Fig. 4F). Interestingly, in a colocalization study, the overexpressed MTP18[mLIR] did not bind to LC3 and thus blocked mitophagy (Fig. 4G,H), confirming that the MTP18–LC3 interaction is essential for mitochondrial clearance by mitophagy. Subsequently, we observed mitophagy flux by assessing colocalization of mitochondrial proteins TOM20 and LAMP1 in MTP18[mLIR]-FaDu cells. However, significantly fewer colocalized MTP18[mLIR]-FaDu cells were found, with or without CCCP treatment (Fig. S3D; Fig. 4I). In addition, we observed mitophagy status using mito-Keima in MTP18- and MTP18[mLIR]-overexpressing FaDu cells. Remarkably, we found that the mt-Keima red/green ratio in MTP18-FaDu cells was significantly increased compared to the MTP18[mLIR]-FaDu cells confirming that the LIR motif is essential for the mitophagy induction (Fig. S3E; Fig. 4J).

Outer mitochondrial membrane rupture is essential for the MTP18 and LC3 interaction to induce mitophagy

Given that MTP18 is an IMM protein, OMM rupture is necessary for MTP18 to bind to LC3 to promote mitophagy. OMM protein degradation occurs in two possible settings. First, upon CCCP-induced Parkin translocation, OMM proteins might be extracted from the mitochondrial membrane and delivered to the proteasomes. Second, proteasomes might be recruited to damaged mitochondria (Durcan and Fon, 2015; Harper et al., 2018; Yoshii et al., 2011). To examine the role of proteasome-dependent OMM rupture in MTP18 exposure to induce mitophagy, COX IV levels was analyzed in the presence of the proteasomal inhibitor epoxomicin (Wei et al., 2017) in MTP18-FaDu cells. Epoxomicin was observed to suppress the degradation of COX IV on exposure of those cells to CCCP; however, in the control group in MTP18-FaDu cells, COX IV was degraded upon CCCP treatment (Fig. 5A–C; Fig. S3F), indicating that proteasomal degradation of the OMM is necessary for MTP18-induced mitophagy. We also examined the interaction of MTP18 and LC3 in the presence of epoxomicin and found that epoxomicin hindered the colocalization of MTP18 and LC3 due to loss of OMM degradation in MTP18-overexpressing cells (Fig. 5D,E). Furthermore, to examine autophagosome formation around dysfunctional mitochondria, we transfected MTP18-FaDu cells with p40phox, exposed them to CCCP in the presence of epoxomicin, and stained them for TOM20. Mitochondrial staining was observed to be reduced in CCCP-treated MTP18-overexpressing FaDu cells with mitochondrial clusters that had been exposed to p40phox. In contrast, epoxomicin-treated MTP18-FaDu cells with or without CCCP exposure showed an increase in TOM20 levels, and the mitochondria were concentrated in perinuclear clusters that were not apposed to p40phox PX–EGFP hotspots (Fig. 5F,G), confirming that proteasomal degradation to rupture the OMM is essential for mitophagy induction.

Fig. 5.

Proteasomal degradation of the mitochondrial outer membrane is essential for the MTP18–LC3 interaction to induce mitophagy. (A–C) After MTP18-FaDu cells were treated with CCCP (10 μM, 6 h) in the presence of 100 nM epoxomicin (EPOX; 3 h), the levels of COX IV was analyzed by (A) western blotting and (B) immunostaining assay. (C) The total relative fluorescence intensity of COX IV was determined. (D,E) After MTP18-overexpressing FaDu cells were exposed to CCCP in the presence of epoxomicin and immunostained for FLAG and LC3 antibodies, colocalization was analyzed (D) by confocal microscopy. (E) The number of LC3 dots colocalized with MTP18-FLAG–positive cell were quantified. (F,G) After MTP18-FaDu cells were transfected with p40phox–EGFP, treated with CCCP in the presence of epoxomicin, and stained for TOM20, (F) they were analyzed by confocal microscopy. (G) The graph represents mitochondrial clusters apposed to PI3K class III (p40phox). For all panels, n=3; >25 cells counted per condition for C, E and G. Quantitative results are mean±s.d. ***P<0.001; **P<0.01; *P<0.05; ns, not significant (one-way ANOVA with Šidák's post test). Scale bars: 18.4 μm.

Fig. 5.

Proteasomal degradation of the mitochondrial outer membrane is essential for the MTP18–LC3 interaction to induce mitophagy. (A–C) After MTP18-FaDu cells were treated with CCCP (10 μM, 6 h) in the presence of 100 nM epoxomicin (EPOX; 3 h), the levels of COX IV was analyzed by (A) western blotting and (B) immunostaining assay. (C) The total relative fluorescence intensity of COX IV was determined. (D,E) After MTP18-overexpressing FaDu cells were exposed to CCCP in the presence of epoxomicin and immunostained for FLAG and LC3 antibodies, colocalization was analyzed (D) by confocal microscopy. (E) The number of LC3 dots colocalized with MTP18-FLAG–positive cell were quantified. (F,G) After MTP18-FaDu cells were transfected with p40phox–EGFP, treated with CCCP in the presence of epoxomicin, and stained for TOM20, (F) they were analyzed by confocal microscopy. (G) The graph represents mitochondrial clusters apposed to PI3K class III (p40phox). For all panels, n=3; >25 cells counted per condition for C, E and G. Quantitative results are mean±s.d. ***P<0.001; **P<0.01; *P<0.05; ns, not significant (one-way ANOVA with Šidák's post test). Scale bars: 18.4 μm.

Previous studies have reported that Parkin is essential for proteasome-dependent OMM rupture (Yoshii et al., 2011). We, therefore, sought to determine whether proteasome-dependent OMM rupture is required for MTP18–LC3 interaction. To evaluate the role of Parkin in OMM rupture for MTP18-mediated mitophagy in HeLa cells (which are Parkin deficient; Burman et al., 2017), we co-transfected HeLa cells with MTP18 and Parkin overexpression plasmids and then used western blotting and immunostaining to analyze the mitophagy status in those cells. Western blot analysis showed that COX IV levels was reduced in the MTP18 and Parkin group but was unchanged in the MTP18-only group after exposure to CCCP, suggesting that Parkin might be required for OMM rupture during MTP18-mediated mitophagy in HeLa cells (Fig. 6A; Fig. S3G). Furthermore, MTP18 overexpression directly reduced COX IV levels in the presence of CCCP in EGFP–Parkin-expressing HeLa cells, as measured by quantitative image analysis (Fig. 6B,C). We validated the importance of Parkin by transiently transfecting siRNA against Parkin (siParkin) into MTP18-overexpressing FaDu cells. Parkin knockdown was observed to block COX IV degradation and abrogate mitophagy upon CCCP treatment in MTP18-overexpressing FaDu cells (Fig. 6D–F; Fig. S3H).

Fig. 6.

Parkin and PINK1 are required for MTP18-dependent mitophagy. (A–C) After HeLa cells were co-transfected with MTP18 and GFP–Parkin and treated with CCCP (10 μM, 6 h), COX IV levels was analyzed by (A) western blotting and (B) confocal microscopy. Scale bars: 25 μm. (C) The total relative fluorescence intensity of COX IV levels was determined. (D–F) After MTP18-FaDu cells were transfected with siParkin and treated with 10 μM CCCP for 6 h, COX IV levels was analyzed by (D) western blotting and (E) confocal microscopy. (F) The total relative fluorescence intensity of COX IV levels in the absence of Parkin was determined. (G–I) MTP18-FaDu cells were transfected with siPINK1 and treated with 10 μM CCCP for 6 h, COX IV levels was analyzed by (G) western blot and (H) confocal microscopy. (I) The total relative fluorescence intensity of COX IV levels in the absence of PINK1 was determined. For all panels, n=3; >25 cells counted per condition for C, F and I. Quantitative results are mean±s.d. ***P<0.001; **P<0.01; *P<0.05; ns, not significant (one-way ANOVA with Šidák's post test). Scale bars: 18.4 μm.

Fig. 6.

Parkin and PINK1 are required for MTP18-dependent mitophagy. (A–C) After HeLa cells were co-transfected with MTP18 and GFP–Parkin and treated with CCCP (10 μM, 6 h), COX IV levels was analyzed by (A) western blotting and (B) confocal microscopy. Scale bars: 25 μm. (C) The total relative fluorescence intensity of COX IV levels was determined. (D–F) After MTP18-FaDu cells were transfected with siParkin and treated with 10 μM CCCP for 6 h, COX IV levels was analyzed by (D) western blotting and (E) confocal microscopy. (F) The total relative fluorescence intensity of COX IV levels in the absence of Parkin was determined. (G–I) MTP18-FaDu cells were transfected with siPINK1 and treated with 10 μM CCCP for 6 h, COX IV levels was analyzed by (G) western blot and (H) confocal microscopy. (I) The total relative fluorescence intensity of COX IV levels in the absence of PINK1 was determined. For all panels, n=3; >25 cells counted per condition for C, F and I. Quantitative results are mean±s.d. ***P<0.001; **P<0.01; *P<0.05; ns, not significant (one-way ANOVA with Šidák's post test). Scale bars: 18.4 μm.

PINK1 has been recognized for sensing damaged mitochondria associated with Parkin recruitment to the mitochondria for proteasomal degradation (Durcan and Fon, 2015). To study the role of PINK1 in the recruitment of Parkin, siRNA against PINK1 (siPINK1) was transiently transfected into MTP18-overexpressing FaDu cells, and then western blotting and immunostaining were used to examine the expression status of Parkin and the IMM protein COX IV. Loss of PINK1 was observed to revoke MTP18-induced mitophagy (Fig. 6G–I; Fig. S3I), as evidenced by the lack of degradation of the IMM proteins, confirming that PINK1 is essential for degrading the OMM to induce MTP18-dependent mitophagy. We validated the role of Parkin in OMM rupture by transiently transfecting siParkin into wild-type MTP18-FaDu cells and mutant MTP18[mLIR]-FaDu cells. It observed that upon CCCP treatment in MTP18 FaDu cells, that both the OMM protein TOM20 and IMM protein COX IV were degraded but Parkin-knockdown MTP18-FaDu cells showed intact OMM and IMM proteins on CCCP treatment. Again, loss of Parkin was observed to inhibit the degradation of OMM protein TOM20 and IMM protein COX IV upon CCCP treatment in MTP18[mLIR]-expressing FaDu cells. In contrast, TOM20 levels was reduced, and COX IV levels remained intact after CCCP treatment in MTP18[mLIR]-overexpressing FaDu cells, suggesting that Parkin is involved in OMM rupture (Fig. 7A,B). Similarly, we observed that loss of PINK1 expression was associated with inhibition of TOM20 and COX IV degradation after CCCP treatment in MTP18-expressing FaDu cells. However, only MTP18[mLIR]-expressing FaDu cells showed reduced TOM20 and intact COX IV levels (Fig. 7C,D), signifying that PINK1 knockdown limits OMM degradation and mitophagy, confirming that Parkin is vital for degrading the OMM and not for mitophagy in MTP18-induced mitophagy.

Fig. 7.

Parkin and PINK1 are vital for degrading the outer mitochondrial membrane but not for mitophagy. (A,B) After FaDu cells expressing MTP18[WT] and MTP18[mLIR] were transfected with siParkin and then treated with 10 μM CCCP for 6 h, (A) levels of COX IV and TOM20 was analyzed by confocal microscopy. (B) The relative fluorescence intensity of COX IV and TOM20 levels was determined. (C,D) Further MTP18[WT] and MTP18[mLIR] overexpressing FaDu cells were transfected with siPINK and treated with 10 μM CCCP for 6 h, and (C) levels of COX IV and TOM20 were analyzed by confocal microscopy. (D) The relative fluorescence intensity of COX IV and TOM20 was determined. For all panels, n=3; >25 cells counted per condition for B and D. Quantitative results are mean±s.d. ***P<0.001; **P<0.01; *P<0.05 (one-way ANOVA with Šidák's post test). Scale bars: 18.4 μm.

Fig. 7.

Parkin and PINK1 are vital for degrading the outer mitochondrial membrane but not for mitophagy. (A,B) After FaDu cells expressing MTP18[WT] and MTP18[mLIR] were transfected with siParkin and then treated with 10 μM CCCP for 6 h, (A) levels of COX IV and TOM20 was analyzed by confocal microscopy. (B) The relative fluorescence intensity of COX IV and TOM20 levels was determined. (C,D) Further MTP18[WT] and MTP18[mLIR] overexpressing FaDu cells were transfected with siPINK and treated with 10 μM CCCP for 6 h, and (C) levels of COX IV and TOM20 were analyzed by confocal microscopy. (D) The relative fluorescence intensity of COX IV and TOM20 was determined. For all panels, n=3; >25 cells counted per condition for B and D. Quantitative results are mean±s.d. ***P<0.001; **P<0.01; *P<0.05 (one-way ANOVA with Šidák's post test). Scale bars: 18.4 μm.

MTP18 promotes cell survival and inhibits apoptosis through the mitophagy pathway

A previous reports has shown that MTP18 facilitates cell survival in various cancers (Zhang et al., 2018). To examine the role of MTP18 in cell survival in oral cancer, we studied cell death in MTP18-overexpressing FaDu cells after starvation and exposure to various anticancer drugs. A 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium (MTT) assay in MTP18-expressing cells after 72 h of nutrient starvation found greater viability in those cells than in control cells similarly nutrient-starved (Fig. S4A). The MTT assay was repeated with various anticancer drugs, including cis-dichlorodiammine platinum (II) (CDDP), doxorubicin and 5-fluorouracil. After MTP18-overexpressing FaDu cells were treated with those drugs at various doses for 24 h, the expected response of a decrease in cell viability was observed to be significantly rescued by MTP18 expression (Fig. 8A; Fig. S4B,C). Next, annexin V–Propidium Iodide (PI) analysis by flow cytometry was used to determine cell apoptosis status after CDDP exposure (Fig. S4D; Fig. 8B), confirming that MTP18 confers survival to oral cancer cells undergoing chemotherapeutic stress. Furthermore, in MTP18-FaDu cells exposed to CDDP, western blot analysis of the expression of anti-apoptotic and pro-apoptotic proteins was used to quantify cell death, revealing that MTP18 overexpression moderates CDDP-induced apoptosis (Fig. S4E,F). Moreover, the caspase 3/7 glo assay showed that activation of the caspase cascade was higher in the CDDP-treated control group than in the MTP18 group (Fig. S4G), confirming that MTP18 serves a survival function in oral cancer cells. Furthermore, we used siRNA to knockdown MTP18 in Cal33 cells, and we performed the MTT assay using a different dose of CDDP for 24 h. The MTP18 knockdown was observed to induce significant apoptosis in response to CDDP (Fig. 8C). Flow cytometry analysis by annexin V–PI confirmed the apoptosis induction after CDDP treatment in the MTP18 loss-of-function group (Fig. S4H; Fig. 8D). Apoptosis analysis of siMTP18-transfected Cal33 cells by western blotting showed that CDDP promoted apoptosis by increasing the expression of pro-apoptotic proteins and decreasing the levels of anti-apoptotic proteins, with activation of caspase 3 and 7 during loss of MTP18 levels (Fig. S4I–K). Taken together, these results confirm that MTP18 provides a survival advantage to cancer cells exposed to stress.

Fig. 8.

MTP18 promotes mitophagy to enhance cell survival in oral cancer cells. MTP18-FaDu cells were treated with varying concentrations of cisplatin (CDDP; 0, 5, 10 and 20 µM for 24 h), and (A) cell survival was analyzed with an MTT assay. After MTP18-FaDu cells were treated with 20 µM CDDP (24 h), (B) apoptosis was quantified with annexin V–PI by flow cytometry. After Cal33 cells were transfected with siMTP18 and treated with varying concentrations of CDDP (0, 5, 10, 20 and 40 µM for 24 h), (C) survival was analyzed with an MTT assay; (D) apoptosis was quantified with annexin V–PI by flow cytometry. After MTP18-overexpressing FaDu cells were treated with 20 µM CDDP for 24 h in the presence of wortmannin (WORT), (E) survival was analyzed by MTT assay; (F) apoptosis was quantified with annexin V–PI staining by flow cytometry. After MTP18-expressing FaDu cells were treated with 20 µM CDDP (24 h) in the presence of 50 µM Mdivi1 (3 h), (G) survival was analyzed by MTT assay; (H) apoptosis was quantified with annexin V–PI staining by flow cytometry. MTP18[WT] and MTP18[mLIR]-FaDu cells were analyzed (I) cell survival by MTT assay, and (J) apoptosis was quantified by annexin V–PI staining through flow cytometry. (K) MTP18-FaDu cells were transfected with siParkin and were treated with 20 µM CDDP (24 h) and analyzed cell survival with an MTT assay, (L) apoptosis was quantified by annexin V–PI staining through flow cytometry. (M) PINK1-knockdown MTP18-FaDu cells were treated with 20 µM CDDP (24 h), cell survival was analyzed through MTT assay and (N) apoptosis was quantified by annexin V–PI staining through flow cytometry. n=3. Quantitative results are mean±s.d. ***P<0.001; **P<0.01; *P<0.05 (one-way ANOVA with Šidák's post test).

Fig. 8.

MTP18 promotes mitophagy to enhance cell survival in oral cancer cells. MTP18-FaDu cells were treated with varying concentrations of cisplatin (CDDP; 0, 5, 10 and 20 µM for 24 h), and (A) cell survival was analyzed with an MTT assay. After MTP18-FaDu cells were treated with 20 µM CDDP (24 h), (B) apoptosis was quantified with annexin V–PI by flow cytometry. After Cal33 cells were transfected with siMTP18 and treated with varying concentrations of CDDP (0, 5, 10, 20 and 40 µM for 24 h), (C) survival was analyzed with an MTT assay; (D) apoptosis was quantified with annexin V–PI by flow cytometry. After MTP18-overexpressing FaDu cells were treated with 20 µM CDDP for 24 h in the presence of wortmannin (WORT), (E) survival was analyzed by MTT assay; (F) apoptosis was quantified with annexin V–PI staining by flow cytometry. After MTP18-expressing FaDu cells were treated with 20 µM CDDP (24 h) in the presence of 50 µM Mdivi1 (3 h), (G) survival was analyzed by MTT assay; (H) apoptosis was quantified with annexin V–PI staining by flow cytometry. MTP18[WT] and MTP18[mLIR]-FaDu cells were analyzed (I) cell survival by MTT assay, and (J) apoptosis was quantified by annexin V–PI staining through flow cytometry. (K) MTP18-FaDu cells were transfected with siParkin and were treated with 20 µM CDDP (24 h) and analyzed cell survival with an MTT assay, (L) apoptosis was quantified by annexin V–PI staining through flow cytometry. (M) PINK1-knockdown MTP18-FaDu cells were treated with 20 µM CDDP (24 h), cell survival was analyzed through MTT assay and (N) apoptosis was quantified by annexin V–PI staining through flow cytometry. n=3. Quantitative results are mean±s.d. ***P<0.001; **P<0.01; *P<0.05 (one-way ANOVA with Šidák's post test).

Next, we hypothesized that MTP18-induced mitophagy might support cancer cell survival in stress conditions. To explore this mechanism, we investigated the apoptosis status of MTP18-overexpressing FaDu cells after using wortmannin to inhibit their MTP18-induced mitophagy. The MTT assay performed to examine cell viability during inhibition of autophagy and mitophagy in response to CDDP indicated that inhibition of mitophagy significantly blocked cell growth (Fig. 8E). Moreover, annexin V–PI assays showed that inhibition of MTP18-mediated mitophagy significantly increased apoptosis during CDDP treatment and that MTP18 overexpression established MTP18-mediated mitophagy, which facilitated cell growth (Fig. S5A; Fig. 8F). Similarly, caspase 3 and 7 activity increased on inhibition of autophagy in MTP18-overexpressing FaDu cells (Fig. S5B). We also pre-treated MTP18-FaDu cells with Mdivi1 (50 µM, 3 h), exposed them to CDDP treatment and used the MTT assay to validate decreased cell viability (Fig. 8G). Annexin V–PI assays showed that Mdivi1 induced significant apoptosis during CDDP treatment (Fig. S5C; Fig. 8H). Inhibition of mitochondrial fission in MTP18-FaDu cells promoted caspase 3 and 7 activity to induce apoptosis (Fig. S5D), as validated by overexpressing MTP18[mLIR] in FaDu cells, thus blocking MTP18-mediated mitophagy. Upon CDDP treatment, the growth of oral cancer cells was significantly inhibited in the MTP18[mLIR] group, as shown by MTT assay (Fig. 8I). The annexin V–PI analysis indicated that the apoptosis induced in the MTP18[mLIR] group was significantly reduced compared with that in the MTP18-FaDu group in response to CDDP (Fig. S5E; Fig. 8J). Furthermore, to understand the role of Parkin and PINK1 in MTP18-driven mitophagy on oral cancer cell survival, we knocked down Parkin and PINK1 using siRNA in vector control and MTP18-FaDu cells and observed the cell survival through MTT assay and apoptosis through the annexin V–PI assay in response to CDDP treatment. The data showed that Parkin or PINK knockdown in MTP18-FaDu cells significantly decreased cell viability and increased apoptosis (Fig. 8K–N; Fig. S5F,G) compared to MTP18-FaDu cells during CDDP treatment, establishing that MTP18 promotes cell survival and inhibits apoptosis through mitophagy in oral cancer cells.

Several proteins – including BNIP3, BINIP3L, FUNDC1, Bcl2L-13, PUMA and FKBP8 – that are present on the OMM, are mitochondrial receptors for the clearance of dysfunctional mitochondria through autophagy (Panigrahi et al., 2020). Similarly, prohibitin 2, an IMM protein, functions as a mitophagy receptor to remove unwanted mitochondria and to regulate cellular physiology (Wei et al., 2017). Results from the present study reveal that MTP18 induces IMM fission and then interacts with LC3 through its LIR motif to initiate mitophagy, establishing it as an IMM mitophagy receptor.

Mitochondria maintain their dynamic nature through a coordinated fission–fusion cycle for proper shape and size to fulfill the energy demands of cells (Yu and Pekkurnaz, 2018). A few conserved proteins, including Drp1 and Fis1, have been reported to be crucial mediators of OMM division (Losón et al., 2013). However, MTP18 has been reported to be the only IMM protein that plays an essential role in IMM fission (Tondera et al., 2005). Our results show higher mitochondrial fission in MTP18-overexpressing cells, establishing MTP18 control of the mitochondrial fission–fusion balance. Interestingly, the knockdown of MTP18 in Drp1- overexpressing cells reduces mitochondrial fission. Furthermore, we found neither MTP18 overexpression in Drp1-knockdown cells nor Drp1 overexpression in MTP18-knockdown cells induced mitochondrial fission, suggesting that Drp1 and MTP18 are both necessary – Drp1 for OMM division and MTP18 for IMM division.

Uneven mitochondrial fission generates dysfunctional mitochondria, followed by mitophagy, in which small, elliptical mitochondria shaped during the mitochondrial fission process are digested and cleared (Ashrafi and Schwarz, 2013; Burman et al., 2017). In that context, our results suggest that MTP18 overexpression is associated with mitochondrial fission, resulting in fragmented mitochondria, thus increasing mitophagy activity. By contrast, MTP18 deficiency leads to a mitochondrial network that inhibits mitophagy. Furthermore, reduced levels of Drp1 and its fission activity combine to prevent mitophagy. Intriguingly, Drp1-overexpressing MTP18-knockdown cells have reduced mitophagy, suggesting that MTP18 plays a central role in the regulation of mitochondrial fission (followed by mitophagy), although the precise function of MTP18 and its associated signaling in mitochondrial fission has yet to be established.

In mitochondrial cargo selection, proteins localized to the OMM bind either directly to LC3 via the LIR motif, which acts as a mitophagy receptor, or indirectly via ubiquitylation using an adaptor protein that contains both ubiquitin-binding domains and LIR domains (Onishi and Yamano, 2021; Panigrahi et al., 2020). During mitophagy, MTP18 binds to LC3 via an established LIR motif essential for MTP18-mediated mitophagy. Thus, we report for the first time that MTP18 is a mitophagy receptor required for mitophagy in mammalian cells. Interestingly, MTP18 is the only mitochondrial fission protein that has a LIR motif and that serves as a mitophagy receptor. That observation proposes a new model for the mitophagy mechanism, in which an IMM fission protein can play a crucial role in cargo selection for autophagy. Moreover, our findings show that the mutant LIR domain in MTP18 does not affect the previously described function of MTP18, including mitochondrial fission, suggesting that MTP18-mediated mitochondrial fission induces mitophagy, but that inhibition of mitophagy does not affect fission activity.

Proteasomes are recruited to the mitochondria upon mitochondrial depolarization in the perinuclear region, an event that is essential for the exposure of the IMM to the cytosol (Yoshii et al., 2011). Proteasomal activity has been reported to be stimulated during the accumulation of mitochondrial precursors in the cytosol because of defective mitochondrial protein localization. Mitochondrial stress-induced proteasomal activity and proteasomes degrade the OMM and expose the IMM to the cytosol (Lavie et al., 2018; Wrobel et al., 2015). Interestingly, apart from its role in mitophagy, Parkin induces OMM degradation of depolarized mitochondria, depending on proteasomal activity. In fact, a previous study has established that prohibitin 2 acts as an IMM mitophagy receptor through Parkin-dependent OMM rupture in mammalian cells. Furthermore, prohibitin 2 stabilizes PINK1 through the PARL–PGAM5–PINK1 axis and causes mitochondrial recruitment of Parkin to trigger mitophagy (Yan et al., 2020). Our results reveal that inhibition of proteasomal activity reduces the MTP18–LC3 interaction, leading to a decrease in MTP18-mediated mitophagy, suggesting that MTP18 necessitates proteasomal degradation of the OMM to expose the IMM to the cytosol so it can interact with LC3. Our data also show that Parkin or PINK1 knockdown inhibits MTP18-dependent mitophagy, establishing the critical role of Parkin and PINK1 in MTP18-induced mitophagy, although whether MTP18 requires Parkin or PINK1 for OMM rupture or ubiquitin-dependent mitophagy has to be identified. Previously, Yoshii et al. (2011) reported that Parkin regulates the degradation of an OMM protein by the proteasome and IMM proteins by mitophagy. Our present study established that levels of TOM20 was reduced in mLIR–MTP18-overexpressing cells, but not in Parkin- or PINK1-knockdown MTP18[mLIR]-overexpressing cells, which accords with Parkin being essential for OMM rupture, but not for mitophagy during MTP18-induced mitophagy, although detailed studies have to be conducted.

Mitochondrial fission induces cristae remodeling and amplifies caspase-dependent apoptosis (Oettinghaus et al., 2016). Burman et al. (2017) reported that mitochondrial fission protects healthy mitochondria from elimination by mitophagy. Interestingly, blocking Drp1-mediated mitochondrial fission prevents mitophagy and increases apoptosis (Lin et al., 2020). Moreover, Fis1 independently regulates mitochondrial fission and apoptosis (Alirol et al., 2006; Yu et al., 2005). Initially, Tondera et al. (2004) reported that MTP18 knockdown leads to apoptosis, as indicated by the release of cytochrome c from mitochondria and activation of caspases in both HaCaT and COS-7 cells. MTP18 is overexpressed and plays a crucial oncogenic role in carcinogenesis (Zhang et al., 2018). Our data show that MTP18 confers a survival benefit to oral cancer cells, and MTP18 deficiency triggers apoptosis in response to various stresses and anticancer drugs. In addition, inhibition of MTP18-induced mitophagy results in increased apoptosis, indicating that MTP18 controls cell survival through mitochondrial fission and mitophagy in oral cancer cells. Thus, targeting MTP18 and its associated signaling could serve as a potential therapeutic pathway for the treatment of oral cancer and its malignancy.

Chemicals

Carbonyl cyanide 3-chlorophenylhydrazone (CCCP) (no. C2759), wortmannin (no. 681675), Mdivi1 (no. M0199), epoximicine (no. E3652), DAPI (no. D9542), DMSO (no. D2650), cis-dichlorodiammine platinum (II) (CDDP) (no. P4394), 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium (MTT) (no. M2128) and protease inhibitor (no. P8340) were purchased from Sigma-Aldrich (St Louis, MO, USA). Fetal bovine serum (sterile-filtered, South American origin; no. 10270106) and Opti-MEM (no. 51985034) were purchased from Thermo Fisher Scientific. Antibiotic-antimycotic (no. A002A), Trypsin-EDTA (no. TCL-007), Dulbecco's modified Eagle's medium (no. AL151A) and minimum essential medium Eagle (no. AL047S) were purchased from HiMedia Laboratories (Mumbai, India). Lipofectamine 2000 (no. 11668019), MitoTracker Green (no. M7514), and MitoTracker Red CMXRos (no. M7512) were purchased from Invitrogen (Carlsbad, CA, USA). Lysis buffer (no. 9803S) was purchased from Cell Signaling Technology (Danvers, MA, USA). The annexin V–PI kit (no. 1001K) was purchased from Abgenex (Bhubaneswar, India), and the Caspase-Glo 3/7 Assay kit (no. G8090), from Promega (Madison, WI, USA).

Antibodies

The antibodies against the following proteins were used in the study: β-actin [Abgenex, 11-13012; 1:5000 for western blotting (WB)], FLAG M2 [Sigma-Aldrich, F1804; 1:5000 for WB and 1:1000 for immunofluorescence (IF)], MTP18 (Sigma-Aldrich, SAB4301167; 1:1000), LC3 (Cell Signaling Technology, 83506S; 1:1000 for WB and 1:500 for IF), LC3 (Sigma-Aldrich, L7543; 1:500 for IF), COX IV (Cell Signaling Technology, 4850S; 1:6,000 for WB and 1:500 for IF), DRP1 (Santa Cruz Biotechnology, sc-271583; 1:1000 for WB), p-DRP1 (Ser616; Cell Signaling Technology, 3455S; 1:1000 for WB), LAMP1 (Sigma-Aldrich, SAB3500285; 1:500 for IF), Fis1 (Santa Cruz Biotechnology, sc-376469; 1:1000 for WB), Beclin 1 (Santa Cruz Biotechnology, sc-48341; 1:1000 for WB), PINK1 (Santa Cruz Biotechnology, sc-517353; 1:500 for WB), Parkin (Santa Cruz Biotechnology, sc-32282; 1:500 for WB), TOM20 (BD Biosciences, 612278; 1:6000 for WB and 1:3000 for IF), BAX (Santa Cruz Biotechnology, sc-20067; 1:500 for WB), cytochrome c (BD Biosciences, 556433; 1:1,000 for WB), PARP (Cell Signaling Technology, 9542S; 1:1000 for WB), anti-mouse IgG Alexa Fluor 568 (Thermo Fisher Scientific, A-11004; 1:500), anti-rabbit IgG Alexa Fluor 568 (Thermo Fisher Scientific, A-11011; 1:500), anti-rabbit IgG Alexa Fluor 488 (Thermo Fisher Scientific, A-11008; 1:500), anti-mouse IgG Alexa Fluor 488 (Thermo Fisher Scientific, A-11001; 1:500), anti-mouse IgG Alexa Fluor 647 (Abcam, ab150115; 1:500), and anti-rabbit IgG Alexa Fluor 700 (Thermo Fisher Scientific, A-21038; 1:500), anti-mouse IgG secondary conjugated to HRP (Abgenex, 11-301; 1:5000) and rabbit IgG conjugated to secondary HRP (Abgenex, 11-315; 1:5000) are commercially available.

Cell culture

Human oral cancer cell lines (FaDu and Cal33) were obtained from American Type Culture Collection (Manassas, VA, USA), and were cultured in, respectively, minimum essential medium Eagle's medium and Dulbecco's modified Eagle's medium with 10% fetal bovine serum and 1× antibiotic-antimycotic. Cultures were incubated at 37°C in a humidified incubator (95% air, 5% CO2).

Plasmids and siRNA

FaDu cells were transfected with wild-type MTP18 plasmid (no. RC205922) purchased from Origene (Rockville, MD, USA). To study the accumulation of autophagosomes, vector- and MTP18-FaDu cells were transfected with p40PX-EGFP plasmid (Addgene plasmid no. 19010, deposited by Michael Yaffe). Similarly, for the mitochondrial fission study, FaDu cells were transiently transfected with Drp1 (Addgene plasmid no. 34706, deposited by David Chan). In addition, HeLa cells were transfected with pEGFP-Parkin WT (Addgene plasmid no. 45875, deposited by Edward Fon). A corresponding empty backbone vector was transfected into each plasmid of interest. siRNAs for MTP18 (no. sc-75842), Drp1 (no. sc-43732), Beclin 1 (no. sc-29797), PINK1 (no. sc-37153) and Parkin (no. sc-42158), and their corresponding siControl were obtained from Santa Cruz Biotechnology.

Immunofluorescence staining and colocalization studies

MTP18-transfected FaDu cells were treated with or without CCCP (10 μM, 6 h), fixed in 10% formaldehyde, and then washed with phosphate-buffered saline. The cells were permeabilized by the addition of 0.3% Triton X-100 for 30 min at room temperature, followed by incubation with primary antibody overnight (14–16 h). The cells were then washed with phosphate-buffered saline and incubated with secondary antibody for 4–6 h, followed by DAPI counterstaining. The expression patterns of proteins and the colocalization studies were conducted using a confocal laser scanning microscope (Leica TCS SP8: Leica Microsystems, Wetzlar, Germany).

Western blotting and immunoprecipitation analysis

Vector- and MTP18-FaDu cells were treated with CCCP (10 μM, 6 h) and then underwent protein extraction. The cells were lysed with lysis buffer, and 50 μg protein was subjected to SDS-PAGE. The resulting protein gel was transferred to a nitrocellulose membrane and blocked with 5% skimmed milk (in phosphate-buffered saline with 0.001% Tween 20) at room temperature for 1 h. The blots were subsequently incubated with the corresponding primary antibodies for MTP18, β-actin, phosphorylated (p)Drp1, Drp1, Fis1, COX IV, LC3, Beclin 1, TOM20, PINK1, Parkin, Bax, cytochrome c and PARP overnight (12–14 h) and then with the HRP-conjugated secondary antibody for 2 h. Protein expression was determined by chemiluminescence analysis using an Image Quant LAS500 chemiluminescence camera (GE Healthcare, Chicago, IL, USA). For the immunoprecipitation assay, cell lysates were incubated overnight at 4°C with their respective antibodies and then coupled with protein A magnetic beads. Afterwards, western blotting analysis with the desired antibody was performed.

Analysis of dysfunctional mitochondria

Dysfunctional mitochondria that accumulate because of cellular stress are reserved for mitophagy. Mitochondrial mass was therefore quantified by flow cytometry using MitoTracker Green FM and MitoTracker Red CMXRos at 100 nM for 30 min at 37°C. After staining, the cells were trypsinized and harvested for analysis. The functional/healthy and dysfunctional mitochondrial populations were quantified as the proportion of dysfunctional mitochondria that contributed to the total mitochondrial population (Mukhopadhyay et al., 2016).

Generation of MTP18[mLIR] variant by site-directed mutagenesis

To prepare the MTP18[mLIR] variant (Y157AxxV160A) from the wild-type MTP18 plasmid, the QuikChange II site-directed mutagenesis kit (Stratagene-Agilent, Santa Clara, CA, USA) was used as described in the manufacturer's protocol. The oligonucleotide primers for site-directed mutagenesis (forward: 5′-CTGGGCTTCCCCGCTGTTGGGGCGAGCTTGCGCAG-3′; reverse: 5′-CTGCGCAAGCTCGCCCCAACAGCGGGGAAGCCCAG-3′) were obtained from IDT (Integrated DNA Technologies, Redwood City, CA, USA). To confirm the point mutations Y157A and V160A, plasmid DNA sequencing was performed at the DNA sequencing facility of the Institute of Life Sciences, Bhubaneswar.

Cell viability assay

The role of MTP18 in cellular survivability after CDDP treatment was analyzed using an MTT assay. FaDu cells were harvested, and 2×104 cells per well were cultured in a 96-well plate at 37°C. The cells were transfected with empty vector (pCMV6; no PS100001, Origene), MTP18, MTP18[mLIR], and then treated with 20 µM CDDP for 24 h in the presence or absence of the inhibitor wortmannin (5 mM, 2 h). MTT solution (5 mg ml−1) was then added, and the cells were incubated for 3 h. The resultant formazan crystals were dissolved in DMSO, and the absorbance was measured at 595 nm in a microplate reader (PerkinElmer, Waltham, MA, USA).

Caspase activity assays

FaDu and Cal33 cells with overexpressed and knockdown MTP18 were treated with CDDP (20 μM, 24 h) for 24 h in the presence of inhibitors. The whole-cell lysate of the cells was analyzed for caspase 3 and 7 activity using the Caspase-Glo 3/7 assay kit as directed in the kit's manual (Promega, Madison, WI, USA).

Flow cytometry

Vector-FaDu, MTP18-FaDu, and MTP18[mLIR]-FaDu cells were treated with CDDP (20 μM, 24 h) for 24 h in the presence of wortmannin (5 M, 2 h) and Mdivi1 (50 μM, 3 h). The cells were then collected by trypsinization and washed with phosphate-buffered saline by centrifuging at 1000 g for 5 min at 37°C. The cells were then incubated with annexin V binding buffer, followed by washing and incubation with annexin V and PI in binding buffer as instructed in the user manual and analyzed by flow cytometry (BD Biosciences).

Image acquisition

High-resolution Z-stack images were acquired to analyze the expression of mitochondrial membrane proteins [Z-stack across the intensity of the cell provides a constant state of 3D information about the mitochondrial morphology and protein expression]. The expression pattern of the desired mitochondrial proteins was accomplished using a confocal laser scanning microscope (Leica TCS SP8). The experimental cells were treated and stained with the required antibody per the experimental setup, followed by image acquisition. The images were taken randomly and multiple patches of cells containing single and multiple cells (∼20 to 30 cells) for each experiment, and every experiment was repeated three times. The confocal microscopy settings were adjusted to give the brightest unsaturated images and maintained for each experimental set up during the analysis.

Image analysis

The expression pattern of proteins was assessed using a confocal laser scanning microscope (Leica TCS SP8). After the random acquisition, the image analysis was undertaken by a researcher who was not aware of the conditions for each experimental setting. The total cellular signal was measured for each image, and the average of the total cellular signal of the protein was analyzed by using appropriate statistical analysis. For expression analysis of the 3D images, the total cellular signal of the desire protein was analyzed, in which the 3D images were converted to 8-bit type, and then the total cellular 3D intensity was measured through the 3D Suite plugin available in ImageJ macro tool. The mean intensity was taken to plot the graph. The raw images were binarized, and the mitochondrial network/branch length were analyzed by using plugin Skeleton as previously described previously (Valente et al., 2017).

Statistical analysis

The Prism software application (version 7.0: GraphPad, San Diego, CA, USA) was used to analyze statistical significance. Error bars denote s.d. of the means. Statistical differences between experimental subgroups were determined using an unpaired two-tailed Student's t-test. One-way ANOVA with Šidák's post-hoc test was performed to analyze nonparametric tests between the groups. All P-values were obtained from a number of individual experiments.

We are grateful to Central Research Facility (CRF), National Institute of Technology Rourkela for providing the confocal microscopy facility for this research. We thank Sushanta Pradhan for technical support and expertise in confocal microscopy for data acquisition for the research work and Dr Rabindra kumar Behera for helping us in generating the mutant MTP18.

Author contributions

Conceptualization: D.P.P., P.P.P., S.K.B.; Methodology: D.P.P., P.P.P., S. Patra, B.S.P., S.K.B.; Validation: D.P.P., P.P.P., S.K.B.; Formal analysis: D.P.P., P.P.P., S. Patil, B.S.P., S.K.B.; Investigation: D.P.P., P.P.P., B.P.B., S.K.B.; Data curation: D.P.P., P.P.P.; Writing - original draft: D.P.P., B.P.B., S. Patra, S.K.B.; Writing - review & editing: D.P.P., S. Patra, S. Patil, B.S.P., S.K.B.; Visualization: D.P.P., P.P.P., S. Patil, B.S.P., S.K.B.; Supervision: S.K.B.; Funding acquisition: S.K.B.

Funding

This work was partly supported by the Science and Engineering Research Board (number: CRG/2021/000053), Department of Science and Technology (DST), and the Board of Research in Nuclear Sciences (BRNS) [Number: 54/14/07/2022-BRNS/10913], Department of Atomic Energy (DAE), Government of India. Research infrastructure was partly provided by Fund for Improvement of S&T infrastructure in universities & higher educational institutions (FIST) (number: SR/FST/ LSI-025/2014), Department of Science and Technology, Government of India.

Data availability

All relevant data can be found within the article and its supplementary information.

Alirol
,
E.
,
James
,
D.
,
Huber
,
D.
,
Marchetto
,
A.
,
Vergani
,
L.
,
Martinou
,
J.-C.
and
Scorrano
,
L.
(
2006
).
The mitochondrial fission protein hFis1 requires the endoplasmic reticulum gateway to induce apoptosis
.
Mol. Biol. Cell
17
,
4593
-
4605
.
Annesley
,
S. J.
and
Fisher
,
P. R.
(
2019
).
Mitochondria in health and disease
.
Cells
8
,
680
.
Ashrafi
,
G.
and
Schwarz
,
T. L.
(
2013
).
The pathways of mitophagy for quality control and clearance of mitochondria
.
Cell Death Differ.
20
,
31
-
42
.
Aung
,
L. H. H.
,
Li
,
R.
,
Prabhakar
,
B. S.
,
Maker
,
A. V.
and
Li
,
P.
(
2017
).
Mitochondrial protein 18 (MTP18) plays a pro-apoptotic role in chemotherapy-induced gastric cancer cell apoptosis
.
Oncotarget
8
,
56582
-
56597
.
Aung
,
L. H. H.
,
Li
,
Y. Z.
,
Yu
,
H.
,
Chen
,
X.
,
Yu
,
Z.
,
Gao
,
J.
and
Li
,
P.
(
2019
).
Mitochondrial protein 18 is a positive apoptotic regulator in cardiomyocytes under oxidative stress
.
Clin. Sci.
133
,
1067
-
1084
.
Burman
,
J. L.
,
Pickles
,
S.
,
Wang
,
C.
,
Sekine
,
S.
,
Vargas
,
J. N. S.
,
Zhang
,
Z.
,
Youle
,
A. M.
,
Nezich
,
C. L.
,
Wu
,
X.
,
Hammer
,
J. A.
et al. 
(
2017
).
Mitochondrial fission facilitates the selective mitophagy of protein aggregates
.
J. Cell Biol.
216
,
3231
-
3247
.
Cassidy-Stone
,
A.
,
Chipuk
,
J. E.
,
Ingerman
,
E.
,
Song
,
C.
,
Yoo
,
C.
,
Kuwana
,
T.
,
Kurth
,
M. J.
,
Shaw
,
J. T.
,
Hinshaw
,
J. E.
,
Green
,
D. R.
et al. 
(
2008
).
Chemical inhibition of the mitochondrial division dynamin reveals its role in Bax/Bak-dependent mitochondrial outer membrane permeabilization
.
Dev. Cell
14
,
193
-
204
.
Dai
,
W.
and
Jiang
,
L.
(
2019
).
Dysregulated mitochondrial dynamics and metabolism in obesity, diabetes, and cancer
.
Front. Endocrinol. (Lausanne)
10
,
570
.
Durcan
,
T. M.
and
Fon
,
E. A.
(
2015
).
The three ‘P's of mitophagy: PARKIN, PINK1, and post-translational modifications
.
Genes Dev.
29
,
989
-
999
.
Flippo
,
K. H.
and
Strack
,
S.
(
2017
).
Mitochondrial dynamics in neuronal injury, development and plasticity
.
J. Cell Sci.
130
,
671
-
681
.
Fonseca
,
T. B.
,
Sánchez-Guerrero
,
Á.
,
Milosevic
,
I.
and
Raimundo
,
N.
(
2019
).
Mitochondrial fission requires DRP1 but not dynamins
.
Nature
570
,
E34
-
E42
.
Harper
,
J. W.
,
Ordureau
,
A.
and
Heo
,
J.-M.
(
2018
).
Building and decoding ubiquitin chains for mitophagy
.
Nat. Rev. Mol. Cell Biol.
19
,
93
-
108
.
Kanai
,
F.
,
Liu
,
H.
,
Field
,
S. J.
,
Akbary
,
H.
,
Matsuo
,
T.
,
Brown
,
G. E.
,
Cantley
,
L. C.
and
Yaffe
,
M. B.
(
2001
).
The PX domains of p47phox and p40phox bind to lipid products of PI(3)K
.
Nat. Cell Biol.
3
,
675
-
678
.
Kreymerman
,
A.
,
Buickians
,
D. N.
,
Nahmou
,
M. M.
,
Tran
,
T.
,
Galvao
,
J.
,
Wang
,
Y.
,
Sun
,
N.
,
Bazik
,
L.
,
Huynh
,
S. K.
,
Cho
,
I. J.
et al. 
(
2019
).
MTP18 is a novel regulator of mitochondrial fission in CNS neuron development, axonal growth, and injury responses
.
Sci. Rep.
9
,
10669
.
Lavie
,
J.
,
De Belvalet
,
H.
,
Sonon
,
S.
,
Ion
,
A. M.
,
Dumon
,
E.
,
Melser
,
S.
,
Lacombe
,
D.
,
Dupuy
,
J.-W.
,
Lalou
,
C.
and
Bénard
,
G.
(
2018
).
Ubiquitin-dependent degradation of mitochondrial proteins regulates energy metabolism
.
Cell Rep.
23
,
2852
-
2863
.
Lin
,
X.-H.
,
Qiu
,
B.-Q.
,
Ma
,
M.
,
Zhang
,
R.
,
Hsu
,
S.-J.
,
Liu
,
H.-H.
,
Chen
,
J.
,
Gao
,
D.-M.
,
Cui
,
J.-F.
,
Ren
,
Z.-G.
et al. 
(
2020
).
Suppressing DRP1-mediated mitochondrial fission and mitophagy increases mitochondrial apoptosis of hepatocellular carcinoma cells in the setting of hypoxia
.
Oncogenesis
9
,
67
.
Losón
,
O. C.
,
Song
,
Z.
,
Chen
,
H.
and
Chan
,
D. C.
(
2013
).
Fis1, Mff, MiD49, and MiD51 mediate Drp1 recruitment in mitochondrial fission
.
Mol. Biol. Cell
24
,
659
-
667
.
Mukhopadhyay
,
S.
,
Naik
,
P. P.
,
Panda
,
P. K.
,
Sinha
,
N.
,
Das
,
D. N.
and
Bhutia
,
S. K.
(
2016
).
Serum starvation induces anti-apoptotic cIAP1 to promote mitophagy through ubiquitination
.
Biochem. Biophys. Res. Commun.
479
,
940
-
946
.
Oettinghaus
,
B.
,
D'Alonzo
,
D.
,
Barbieri
,
E.
,
Restelli
,
L. M.
,
Savoia
,
C.
,
Licci
,
M.
,
Tolnay
,
M.
,
Frank
,
S.
and
Scorrano
,
L.
(
2016
).
DRP1-dependent apoptotic mitochondrial fission occurs independently of BAX, BAK and APAF1 to amplify cell death by BID and oxidative stress
.
Biochim. Biophys. Acta
1857
,
1267
-
1276
.
Onishi
,
M.
,
Yamano
,
K.
,
Sato
,
M.
,
Matsuda
,
N.
and
Okamoto
,
K.
(
2021
).
Molecular mechanisms and physiological functions of mitophagy
.
EMBO J.
40
,
e104705
.
Panigrahi
,
D. P.
,
Praharaj
,
P. P.
,
Bhol
,
C. S.
,
Mahapatra
,
K. K.
,
Patra
,
S.
,
Behera
,
B. P.
,
Mishra
,
S. R.
and
Bhutia
,
S. K.
(
2020
).
The emerging, multifaceted role of mitophagy in cancer and cancer therapeutics
.
Semin. Cancer Biol.
66
,
45
-
58
.
Tondera
,
D.
,
Santel
,
A.
,
Schwarzer
,
R.
,
Dames
,
S.
,
Giese
,
K.
,
Klippel
,
A.
and
Kaufmann
,
J.
(
2004
).
Knockdown of MTP18, a novel phosphatidylinositol 3-kinase-dependent protein, affects mitochondrial morphology and induces apoptosis
.
J. Biol. Chem.
279
,
31544
-
31555
.
Tondera
,
D.
,
Czauderna
,
F.
,
Paulick
,
K.
,
Schwarzer
,
R.
,
Kaufmann
,
J.
and
Santel
,
A.
(
2005
).
The mitochondrial protein MTP18 contributes to mitochondrial fission in mammalian cells
.
J. Cell Sci.
118
,
3049
-
3059
.
Valente
,
A. J.
,
Maddalena
,
L. A.
,
Robb
,
E. L.
,
Moradi
,
F.
and
Stuart
,
J. A.
(
2017
).
A simple ImageJ macro tool for analyzing mitochondrial network morphology in mammalian cell culture
.
Acta Histochem.
119
,
315
-
326
.
Wang
,
Y.
,
Liu
,
H.-H.
,
Cao
,
Y.-T.
,
Zhang
,
L.-L.
,
Huang
,
F.
and
Yi
,
C.
(
2020
).
The role of mitochondrial dynamics and mitophagy in carcinogenesis, metastasis and therapy
.
Front. Cell Dev. Biol.
8
,
413
.
Wei
,
Y.
,
Chiang
,
W.-C.
,
Sumpter
,
R.
, Jr
,
Mishra
,
P.
and
Levine
,
B.
(
2017
).
Prohibitin 2 is an inner mitochondrial membrane mitophagy receptor
.
Cell
168
,
224
-
238.e10
.
Wei
,
Q.
,
Sun
,
H.
,
Song
,
S.
,
Liu
,
Y.
,
Liu
,
P.
,
Livingston
,
M. J.
,
Wang
,
J.
,
Liang
,
M.
,
Mi
,
Q.-S.
,
Huo
,
Y.
et al. 
(
2018
).
MicroRNA-668 represses MTP18 to preserve mitochondrial dynamics in ischemic acute kidney injury
.
J. Clin. Invest.
128
,
5448
-
5464
.
Wrobel
,
L.
,
Topf
,
U.
,
Bragoszewski
,
P.
,
Wiese
,
S.
,
Sztolsztener
,
M. E.
,
Oeljeklaus
,
S.
,
Varabyova
,
A.
,
Lirski
,
M.
,
Chroscicki
,
P.
,
Mroczek
,
S.
et al. 
(
2015
).
Mistargeted mitochondrial proteins activate a proteostatic response in the cytosol
.
Nature
524
,
485
-
488
.
Xiao
,
T.
,
Sun
,
J.
,
Xing
,
Z.
,
Xie
,
F.
,
Yang
,
L.
and
Ding
,
W.
(
2020
).
MTFP1 overexpression promotes the growth of oral squamous cell carcinoma by inducing ROS production
.
Cell Biol. Int.
44
,
821
-
829
.
Yan
,
C.
,
Gong
,
L.
,
Chen
,
L.
,
Xu
,
M.
,
Abou-Hamdan
,
H.
,
Tang
,
M.
,
Désaubry
,
L.
and
Song
,
Z.
(
2020
).
PHB2 (prohibitin 2) promotes PINK1-PRKN/Parkin-dependent mitophagy by the PARL-PGAM5-PINK1 axis
.
Autophagy
16
,
419
-
434
.
Yoshii
,
S. R.
,
Kishi
,
C.
,
Ishihara
,
N.
and
Mizushima
,
N.
(
2011
).
Parkin mediates proteasome-dependent protein degradation and rupture of the outer mitochondrial membrane
.
J. Biol. Chem.
286
,
19630
-
19640
.
Youle
,
R. J.
and
Van Der Bliek
,
A. M.
(
2012
).
Mitochondrial fission, fusion, and stress
.
Science
337
,
1062
-
1065
.
Yu
,
S. B.
and
Pekkurnaz
,
G.
(
2018
).
Mechanisms orchestrating mitochondrial dynamics for energy homeostasis
.
J. Mol. Biol.
430
,
3922
-
3941
.
Yu
,
T.
,
Fox
,
R. J.
,
Burwell
,
L. S.
and
Yoon
,
Y.
(
2005
).
Regulation of mitochondrial fission and apoptosis by the mitochondrial outer membrane protein hFis1
.
J. Cell Sci.
118
,
4141
-
4151
.
Yu
,
R.
,
Jin
,
S.-B.
,
Lendahl
,
U.
,
Nistér
,
M.
and
Zhao
,
J.
(
2019
).
Human Fis1 regulates mitochondrial dynamics through inhibition of the fusion machinery
.
EMBO J.
38
,
e99748
.
Zhang
,
Y.
,
Li
,
H.
,
Chang
,
H.
,
Du
,
L.
,
Hai
,
J.
,
Geng
,
X.
and
Yan
,
X.
(
2018
).
MTP18 overexpression contributes to tumor growth and metastasis and associates with poor survival in hepatocellular carcinoma
.
Cell Death Dis.
9
,
956
.

Competing interests

The authors declare no competing or financial interests.

Supplementary information