Endogenous Drp1 Mediates Mitochondrial Autophagy and Protects the Heart Against Energy StressNovelty and Significance
Rationale: Both fusion and fission contribute to mitochondrial quality control. How unopposed fusion affects survival of cardiomyocytes and left ventricular function in the heart is poorly understood.
Objective: We investigated the role of dynamin-related protein 1 (Drp1), a GTPase that mediates mitochondrial fission, in mediating mitochondrial autophagy, ventricular function, and stress resistance in the heart.
Methods and Results: Drp1 downregulation induced mitochondrial elongation, accumulation of damaged mitochondria, and increased apoptosis in cardiomyocytes at baseline. Drp1 downregulation also suppressed autophagosome formation and autophagic flux at baseline and in response to glucose deprivation in cardiomyocytes. The lack of lysosomal translocation of mitochondrially targeted Keima indicates that Drp1 downregulation suppressed mitochondrial autophagy. Mitochondrial elongation and accumulation of damaged mitochondria were also observed in tamoxifen-inducible cardiac-specific Drp1 knockout mice. After Drp1 downregulation, cardiac-specific Drp1 knockout mice developed left ventricular dysfunction, preceded by mitochondrial dysfunction, and died within 13 weeks. Autophagic flux is significantly suppressed in cardiac-specific Drp1 knockout mice. Although left ventricular function in cardiac-specific Drp1 heterozygous knockout mice was normal at 12 weeks of age, left ventricular function decreased more severely after 48 hours of fasting, and the infarct size/area at risk after ischemia/reperfusion was significantly greater in cardiac-specific Drp1 heterozygous knockout than in control mice.
Conclusions: Disruption of Drp1 induces mitochondrial elongation, inhibits mitochondrial autophagy, and causes mitochondrial dysfunction, thereby promoting cardiac dysfunction and increased susceptibility to ischemia/reperfusion.
The heart muscle is characterized by a large volume of mitochondria because of its high energy demand.1–3 Mitochondria produce ATP primarily by using the electrochemical gradient formed by electron transfer via the electron transport chain located on the inner mitochondrial membrane. However, electron leakage from the electron transport chain and production of O2− and H2O2, which arises from dismutation of O2−, occur constantly as byproducts of ATP synthesis, making mitochondria a major source of reactive oxygen species in cardiomyocytes. Although reactive oxygen species at physiological levels act as signaling molecules to induce adaptive responses,4 dysregulated reactive oxygen species production in response to stress damages mitochondrial proteins, stimulating a feed-forward mechanism for reactive oxygen species production, mitochondrial dysfunction, and cell death, including apoptosis triggered by cytochrome c release and necrosis triggered by mitochondrial permeability transition pore (mPTP) opening. To protect against these catastrophic events, cells have intrinsic quality control mechanisms to maintain the overall health of mitochondria, including fusion, fission, and mitochondrial autophagy.5
Editorial, see p 225
Mitochondria are dynamic organelles that constantly undergo fusion and fission5 to adapt to changes in the cellular environment. Although mitochondrial fusion allows mitochondria to maintain membrane potential by fusing depolarized mitochondria to intact ones, fission allows the segregation of unrecoverable mitochondria so that they can be eliminated by autophagy or mitophagy, a specialized form of autophagy.6 Mitochondrial fusion is critically regulated by mitofusin 1 and mitofusin 2, specialized proteins localized on the outer mitochondrial membrane, and by Opa1, a protein localized on the inner mitochondrial membrane, whereas mitochondrial fission is regulated by mitochondrial fission 1 and mitochondrial fission factor, localized on the outer mitochondrial membrane, and by recruitment of a cytoplasmic GTPase, dynamin-related protein 1 (Drp1), to mitochondrial fission sites.7,8
Although the presence of fission and fusion has not been documented in adult ventricular myocytes in an unequivocal manner, previous studies have suggested that mitochondrial quality control plays an essential role in protecting the heart against stress.2 For example, downregulation of mitofusin 1 and mitofusin 2 promotes cardiac dysfunction at baseline and in response to stress because of lack of mitochondrial fusion.9,10 In contrast, Drp1-mediated mitochondrial fission seems to promote cell death during ischemia/reperfusion (I/R).11 Because suppression of Drp1 induces mitochondrial fusion, these results have led to a general belief that mitochondrial fusion is protective.12,13 However, experiments investigating the role of fission in the heart were conducted using mdivi-1, a chemical inhibitor of Drp1.14
Drp1 plays an essential role in mediating Parkin-induced mitochondria selective autophagy, namely mitophagy in mouse embryonic fibroblast (MEF) cells.15 Drp1 also mediates Bnip3-induced autophagy in adult cardiomyocytes.16 However, whether Drp1 is involved in general autophagy that can remove mitochondria (which we here referred to as mitochondrial autophagy) at baseline and in response to stress in cardiomyocytes awaits further investigation using specific interventions. We reasoned that loss-of-function experiments should be conducted using shRNA or a mouse model of genetic deletion of Drp117 to address the role of endogenous Drp1 in regulating mitochondrial autophagy and consequent cardiomyocyte survival and death. In this study, we asked the following: (1) whether endogenous Drp1 plays a protective or detrimental role in the heart at baseline and in response to stress and (2) whether Drp1 mediates mitochondrial autophagy in response to energy stress in cardiomyocytes.
An expanded Methods section is available in the online Data Supplement.
Generation of Drp1 flox homo (fl/fl) mice has been described.17 Cardiac-specific conditional Drp1 knockout (Drp1-CKO) mice were generated by crossing Drp1 fl/fl and alpha myosin heavy chain–MerCreMer (αMHC-MCM) mice, and expression of Drp1 was downregulated by tamoxifen injection (TI, 20 mg/kg, IP) for 5 days. Cardiac-specific heterozygous Drp1 KO (Drp1-hetCKO) mice were generated by crossing Drp1 flox hetero (fl/+) mice and αMHC-MCM mice. Transgenic mice expressing monomeric red fluorescent protein (mRFP)-green fluorescent protein (GFP)-LC3 have been described.18 All experiments involving animals were approved by the Rutgers–New Jersey Medical School’s Institutional Animal Care and Use Committee.
Keima With Mitochondrial Localization Signal
Keima is a fluorescent protein that emits different colored signals at acidic and neutral pHs. Keima with mitochondrial localization signal (Keima-MLS) is a mitochondrially localized pH-indicator protein described by Katayama et al.19 We generated adenovirus harboring Keima-MLS. The method used to detect lysosomal delivery of Keima-MLS has been described.19
Data are expressed as mean±SEM. The difference in means between 2 groups was evaluated using the t test. One-way ANOVA was used to compare multiple groups. Post hoc comparisons of considered pairs were performed using the Bonferroni post hoc test. P values of <0.05 were considered statistically significant. In figure legends, n indicates the number of experiments.
Drp1 Downregulation Stimulates Apoptosis in Cardiomyocytes
To evaluate the role of endogenous Drp1 in regulating mitochondrial morphology in cardiomyocytes, we constructed adenovirus harboring Drp1 shRNA (Ad-shDrp1) and confirmed that Ad-shDrp1 decreases Drp1 in cardiomyocytes compared with adenovirus harboring scramble shRNA (Ad-shScr; Online Figure IA). To observe the morphology of mitochondria, cultured cardiomyocytes were cotransduced with adenovirus harboring mitochondrially targeted DsRed2. Ninety-six hours after transduction, mitochondria in Ad-shDrp1–transduced cardiomyocytes were elongated compared with those in Ad-shScr–transduced cardiomyocytes (Figure 1A). The proportion of cardiomyocytes with elongated mitochondria, as defined by an average mitochondrion length >2 sarcomere units (Online Figure IB), was significantly greater in Ad-shDrp1–transduced cardiomyocytes than in Ad-shScr–transduced cardiomyocytes (Figure 1A). On the contrary, cardiomyocytes with foreshortened mitochondria, as defined by an average mitochondrion length <1 sarcomere unit (Online Figure IB), were markedly reduced in Ad-shDrp1–transduced cardiomyocytes. These results suggest that Drp1 is required for mitochondrial foreshortening in cardiomyocytes at baseline.
Transduction with Ad-shDrp1 significantly increased the number of terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL)-positive cardiomyocytes (Figure 1B) and the amount of cleaved caspase 3 compared with transduction with Ad-shScr (Figure 1C), suggesting that endogenous Drp1 is essential in protection against apoptosis in cardiomyocytes. To exclude the possibility that our timing prevented observation of a period during which Drp1 downregulation-induced fusion is protective, we evaluated cardiomyocyte viability at various time points after transduction of Ad-shDrp1 and Ad-shScr. Viability time-dependently decreased between 0 and 96 hours after transduction of Ad-shDrp1 into cardiomyocytes and was significantly lower in Ad-shDrp1–transduced cardiomyocytes than in Ad-shScr–transduced cardiomyocytes after 72 hours (Figure 1D), indicating that Drp1 downregulation is persistently detrimental.
Endogenous Drp1 Mediates Autophagy and Mitochondrial Quality Control
We next examined the role of Drp1 in cardiomyocyte autophagy. Drp1 downregulation with Ad-shDrp1 significantly reduced LC3-II and increased p62/sequestosome 1 (SQSTM1), a protein degraded by autophagy (Figure 2A). To evaluate autophagic flux, cardiomyocytes were treated with chloroquine, which inhibits fusion of autophagosomes with lysosomes.20 Ad-shDrp1 significantly depressed chloroquine-induced accumulation of LC3-II, and accumulation of p62/SQSTM1 did not change significantly after chloroquine treatment in the presence of Ad-shDrp1 (Figure 2A), suggesting that Drp1 downregulation suppresses autophagic flux in cardiomyocytes at baseline.
To further evaluate the effect of Drp1 downregulation on autophagosome formation, cardiomyocytes were transduced with Ad-GFP-LC3. The number of GFP-LC3 dots was low at baseline, and there was no statistically significant difference between Ad-shScr– and Ad-shDrp1–transduced cardiomyocytes (Figure 2B). However, there was significantly less chloroquine-induced accumulation of GFP-LC3 dots in Ad-shDrp1–transduced cardiomyocytes than in Ad-shScr–transduced cardiomyocytes (Figure 2B), further supporting the idea that Drp1 downregulation suppresses autophagic flux at baseline.
There were significantly more mitochondria, as evaluated with real-time polymerase chain reaction of cytochrome b DNA and immunoblotting of cytochrome c oxidase subunit IV, in Ad-shDrp1–transduced cardiomyocytes than in control cardiomyocytes (Figure 2C and 2D), suggesting that suppression of autophagy caused by Drp1 downregulation leads to accumulation of mitochondria. Peroxisome proliferator-activated receptor-gamma coactivator 1α expression was not significantly altered in Ad-shDrp1–transduced cardiomyocytes (Online Figure IC), suggesting that mitochondrial biogenesis was not affected. Because suppression of autophagy may impair mitochondrial quality control, we evaluated the effect of Drp1 downregulation on mitochondrial function.
Mitochondrial ATP production was significantly lower in cardiomyocytes transduced with Ad-shDrp1 than in those with Ad-shScr (Figure 2E). The effect of Drp1 downregulation on mitochondrial membrane potential was evaluated with JC-1. Drp1 knockdown led to the appearance of green JC-1 staining, indicating depolarization of the mitochondrial membrane potential in cardiomyocytes (Figure 2F). Furthermore, decreases in absorbance at 540 nm in mitochondrial swelling assays, indicative of mPTP opening, were significantly greater in cardiomyocytes with Drp1 knockdown than in control cardiomyocytes (Figure 2G), suggesting that mPTP opening is accelerated by Drp1 downregulation. Cyclosporin A attenuated cardiomyocyte death as evaluated with CellTiter Blue assays, suggesting that mPTP opening contributes to cardiomyocyte death in response to Drp1 downregulation (Figure 2H).
We also evaluated the rate of oxidative phosphorylation in cardiomyocytes, using a Seahorse analyzer (Online Figure IIA). We normalized the oxygen consumption rate (OCR) with either mitochondrial DNA content or cell viability to compensate for potential cell loss caused by cell death. The basal OCR was significantly lower in Drp1-downregulated cardiomyocytes than in control cardiomyocytes (Online Figure IIB). The OCR-linked ATP synthesis, as evaluated with oligomycin treatment, and the maximum respiratory rate, as determined by FCCP (trifluorocarbonylcyanide phenylhydrazone) uncoupling, were also significantly lower in Drp1-downregulated cardiomyocytes than in control cardiomyocytes (Online Figure IIC and IID). The level of proton leak, determined by subtracting OCR-linked ATP synthesis from basal OCR, did not significantly differ between Drp1-downregulated and control cardiomyocytes (Online Figure IIE). Together, these data indicate that Drp1 downregulation induces accumulation of mitochondria accompanied by mitochondrial dysfunction in cardiomyocytes.
Drp1 Mediates Mitochondrial Foreshortening in Response to Glucose Deprivation
We next investigated the involvement of Drp1 in mitochondrial dynamics in response to energy stress. Drp1 was localized primarily in the cytosol in control cardiomyocytes (Figure 3A). Glucose deprivation (GD), which is known to affect mitochondrial dynamics in other cell types,12,13 induced modest mitochondrial accumulation of Drp1 in cultured cardiomyocytes within 4 hours (Figure 3A), accompanied by a slight decrease in cytosolic Drp1, although the reduction did not reach statistical significance. GD-induced mitochondrial expression of Drp1 was also observed with anti-Drp1 immunostaining in mitochondrially targeted DsRed2 expressing cardiomyocytes (Figure 3B). These results suggest that GD increases Drp1 translocation from the cytosol to mitochondria in cardiomyocytes.
After 1 hour of GD, the proportion of cardiomyocytes with elongated mitochondria was increased, but that of cardiomyocytes with foreshortened mitochondria was also increased slightly in Ad-shScr–transduced cardiomyocytes (Figure 3C). A similar result was obtained in cardiomyocytes transduced with adenovirus harboring LacZ (Ad-LacZ; not shown). More than 50% of Ad-shDrp1–transduced cardiomyocytes exhibited elongated mitochondria after 1 hour of GD. After 4 hours of GD, however, ≈15% of Ad-shScr– or Ad-LacZ–transduced cardiomyocytes exhibited foreshortened mitochondria, whereas >50% still showed elongated mitochondria and <1% exhibited foreshortened mitochondria in Ad-shDrp1–transduced cardiomyocytes (Figure 3C).Thus, although GD induces transient mitochondrial elongation followed by foreshortening, Drp1 downregulation induces persistent increases in elongation irrespective of GD. These results suggest that Drp1 plays an essential role in mitochondrial foreshortening at baseline and during GD. Transduction with Ad-shDrp1 significantly increased TUNEL-positive cardiomyocytes after 1 and 4 hours of GD compared with transduction with Ad-shScr (Figure 3D), suggesting that endogenous Drp1 protects cardiomyocytes against apoptosis during GD.
We evaluated the role of endogenous Drp1 in mediating autophagy in response to GD. Four hours of GD significantly increased the number of GFP-LC3 dots in Ad-shScr–transduced cardiomyocytes, but this increase was significantly attenuated in Ad-shDrp1–transduced cardiomyocytes (Figure 3E). To evaluate autophagic flux, cardiomyocytes were cotransduced with adenovirus harboring tandem fluorescent mRFP-GFP-LC3.21 mRFP retains its fluorescence, whereas GFP loses its fluorescence in the acidic environment of lysosomes. In merged images, the red puncta that overlay green puncta and appear yellow indicate autophagosomes, whereas free red puncta indicate autolysosomes. After 4 hours of GD, the numbers of both yellow and free red dots were increased in Ad-shScr–transduced cardiomyocytes, indicating stimulation of autophagic flux. However, the GD-induced increases were attenuated in Ad-shDrp1–transduced cardiomyocytes (Figure 3F), suggesting that Drp1 downregulation inhibits GD-induced autophagic flux. Atg7 increases autophagic flux in cardiomyocytes.22,23 Drp1 downregulation significantly reduced Atg7-induced increases in autophagosomes and autolysosomes at baseline and in response to GD in cardiomyocytes (Figure 3G).Together, the data indicate that endogenous Drp1 plays an essential role in mediating mitochondrial foreshortening, autophagy, and cell survival during GD in cardiomyocytes.
Because Drp1 physically interacts with Bcl-xL in neurons24 and Bcl-xL inhibits autophagy through its binding to Beclin1, we investigated the involvement of Bcl-xL in the suppression of autophagy by Drp1. Using coimmunoprecipitation assays, we confirmed that Drp1 physically interacts with Bcl-2 and Bcl-xL in cardiomyocytes in the presence of Drp1 overexpression (Online Figure IIIA). Increased expression of Drp1 inhibited, whereas downregulation of Drp1 augmented the physical interaction between Beclin1 and Bcl-2 or Bcl-xL (Online Figure IIIB). Downregulation of Drp1 decreased the number of GFP-LC3 dots at baseline and in response to 4 hours of GD. However, the number of GFP-LC3 dots increased significantly when Drp1 was downregulated in the presence of Bcl-xL downregulation with or without chloroquine20 (Online Figure IIIC and IIID). These results suggests that downregulation of Drp1 inhibits autophagy through a Bcl-xL–dependent mechanism, most likely by enhancing interaction between Beclin1 and Bcl-xL.
Prolonged Treatment With Mdivi-1 Mimics the Effect of Drp1 Downregulation
Because previous studies showed that suppression of Drp1 by mdivi-1 protects cardiomyocytes from cell death,11 we investigated the effect of mdivi-1 on mitochondrial morphology and cell death. Single treatment with mdivi-1 at 50 or 100 μmol/L for 1 hour significantly increased the number of cardiomyocytes with elongated mitochondria at baseline. Mdivi-1 at 100 μmol/L also prevented foreshortening of mitochondria after 4 hours of GD (Online Figure IVA). To compare the effects of Drp1 downregulation and mdivi-1 on survival of cardiomyocytes side-by-side, cardiomyocytes were treated with chelerythrine (10 μmol/L), an inducer of apoptosis,25 in the presence or absence of either mdivi-1 or Ad-shDrp1. Although Ad-shDrp1 transduction for 96 hours decreased cardiomyocyte survival at baseline and in response to chelerythrine, mdivi-1 treatment for 1 hour increased cardiomyocyte survival at baseline and in response to chelerythrine compared with vehicle treatment (Online Figure IVB). Mdivi-1 treatment at 50 μmol/L did not significantly affect cardiomyocyte viability in response to GD but the treatment at 100 μmol/L significantly reduced it (Online Figure IVC). Taken together, these results suggest that a single treatment with mdivi-1 has direct cell-protective effects on cardiomyocytes independent of Drp1. However, a higher dose of mdivi-1 partially mimics the effect of Drp1 downregulation even after a single application.
Treatment of cardiomyocytes with 50 μmol/L mdivi-1 every 24 hours for 1 week induced elongation of mitochondria at baseline and inhibited foreshortening of mitochondria even after 4 hours of GD (Online Figure IVD). Prolonged treatment with mdivi-1 time-dependently decreased cardiomyocyte viability compared with vehicle treatment (Online Figure IVE) and significantly suppressed GD-induced autophagic flux as evaluated with mRFP-GFP-LC3 (Online Figure IVF). Thus, prolonged treatment with mdivi-1 mimics the effect of Drp1 downregulation.
Drp1 Mediates Autophagic Removal of Mitochondria
We investigated whether clearance of mitochondria is regulated by Drp1 using mitochondria-targeted Keima fluorescence.19 Keima has a bimodal excitation spectrum peaking at 438±12 and 550±15 nm, corresponding to neutral and acidic pH states, respectively.19 Because fusion of autophagosomes with lysosomes exposes the autophagosome contents to acidic pH, the maturation of autophagosomes to autolysosomes can be monitored by measuring Keima fluorescence.19 We confirmed that Keima-MLS is expressed in cardiomyocyte mitochondria (Figure 4A). Puncta with a high ratio of excitation at 560/440 nm (high 560/440) colocalized with Alexa 488 Dextran, reflecting increased lysosomal localization of Keima-MLS, after treatment with 25 μmol/L of cyanide 3-chlorophenylhydrazone, a mitochondrial uncoupler, for 16 hours to induce mitochondrial autophagy26 (Figure 4B; Online Figure VA), confirming that Keima-MLS works as expected in cardiomyocytes. Puncta with high 560/440, indicating the presence of mitochondria in lysosomes, were significantly increased after 4 hours of GD in cardiomyocytes transduced with Ad-shScr, but not in cardiomyocytes transduced with Ad-shDrp1 (Figure 4C). This increase was abolished in the presence of Ad-shBeclin1-mediated Beclin1 downregulation (Online Figure VB), suggesting that it is mediated by autophagy and that Drp1 is necessary for stimulating autophagic mitochondrial degradation. Interestingly, downregulation of Beclin1 did not affect GD-induced increases in mitochondrial foreshortening (Online Figure VC) but significantly increased GD-induced cell death (Online Figure VD). Thus, although evidence suggests that unopposed fusion of mitochondria alone can induce cell death,27 suppression of autophagy alone may also induce cardiomyocyte death even when mitochondrial foreshortening is not affected.
Atg7 overexpression, which is known to stimulate autophagy,22,23 failed to increase puncta with high 560/440 in Drp1-downregulated cardiomyocytes (Figure 4D), even though it increased autophagosomes and autolysosomes in this condition (Figure 3F and 3G), nor did it inhibit Drp1 knockdown-induced cell death (Figure 4E).
To elucidate the role of endogenous Drp1 in autophagy further, cardiomyocytes were subjected to GD in the presence or absence of Drp1 knockdown and electron microscopic (EM) analyses were conducted (Figure 4F). Drp1 downregulation significantly reduced the number of mitochondria and increased relative mitochondria mass at baseline (Figure 4G). Drp1 downregulation also decreased the total number of autophagosomes at baseline and in response to GD and decreased the number of autophagosomes selectively containing mitochondria (Figure 4H). These results suggest that endogenous Drp1 is important in mediating both general autophagies, including mitochondrial autophagy.
Forced Drp1 Overexpression Induces Apoptosis in Cardiomyocytes
Adenovirus-mediated overexpression of Drp1, which is higher than the level induced by GD, induced foreshortening of mitochondria in cardiomyocytes at baseline by 5-fold (Online Figure VIA and VIB). Under this condition, increases in apoptosis, decreases in mitochondrial DNA, and decreases in mitochondrial membrane potential were observed (Online Figure VIC–VIE). These results suggest that persistent and high-level expression of Drp1 induces mitochondrial dysfunction and apoptosis in cardiomyocytes. Drp1 overexpression significantly increased Keima-MLS puncta with high 560/440 in cardiomyocytes both at baseline and after 4 hours of GD (Online Figure VIF), indicating stimulation of lysosomal removal of mitochondria. Interestingly, suppression of autophagy by Ad-shAtg7 attenuated the increased cell death induced by Drp1 overexpression (Online Figure VIG), suggesting that excessive activation of autophagy by Drp1 may induce cell death.
Basal Characterization of Drp1-CKO Mice
To evaluate the role of endogenous Drp1 in vivo, we used loss-of-function mouse models. No homozygous mice were born during attempts to generate Drp1-CKO mice using αMHC-MCM mice. Therefore, to examine the effect of Drp1 on cardiac function in adult mice in vivo, we generated Drp1-CKO mice by crossing Drp1 flox homo (fl/fl) and αMHC-MCM mice, and Drp1 expression was downregulated in a tamoxifen-dependent manner. We used Drp1-CKO without TI and Drp1 fl/fl with or without TI as controls. Fifteen-week-old male mice were subjected to TI (20 mg/kg, IP) for 5 days. Four and 8 weeks after TI, we measured cardiac function and performed biochemical and histological analyses (Online Figure VIIA). Immunoblot analyses confirmed that cardiac Drp1 levels were significantly lower in Drp1-CKO mice than in control mice (Figure 5A) and that Drp1 was downregulated in a heart-specific manner in Drp1-CKO mice (Online Figure VIIB). Cardiac levels of other proteins involved in mitochondrial dynamics, such as mitofusin 1, mitofusin 2, OPA1, and fission 1, were unaltered in Drp1-CKO mice compared with control mice (Online Figure VIIC). Drp1-CKO mice started to die 8 weeks after TI and all died by 13 weeks after injection, whereas no control mice died during the observation period of 16 weeks after TI. Kaplan–Meier analysis revealed that the survival rate was significantly lower in Drp1-CKO mice than in control mice (Online Figure VIID). Four weeks after TI, the hearts of Drp1-CKO mice were enlarged compared with control hearts (Figure 5B). Postmortem assessment showed that both left ventricular (LV) weight/tibial length, an index of LV hypertrophy, and lung weight/tibial length, an index of lung congestion, were significantly greater in Drp1-CKO than in control mice 4 and 8 weeks after TI (Online Tables I and II). Wheat germ agglutinin staining of LV sections 4 and 8 weeks after TI showed that cardiomyocyte cross-sectional area was significantly greater in Drp1-CKO than in control mice (Figure 5C; Online Figure VIIE). Myocardial fibrosis, as evaluated with Picric Acid Sirius Red and Masson’s Trichrome staining, was also significantly greater in Drp1-CKO mice than in control mice (Figure 5D; Online Figure VIIF and VIIG). Echocardiographic measurements 4 and 8 weeks after TI showed that the LV diastolic dimension was significantly greater, and the LV ejection fraction, an indicator of LV systolic function, was lower in Drp1-CKO mice than in control mice (Online Tables III and IV). Hemodynamic measurements at 4 weeks after TI showed that LV +dP/dt was decreased, whereas LV end-diastolic pressure was significantly elevated in Drp1-CKO compared with control mice (Online Table II). We confirmed that αMHC-MCM alone did not influence cardiac function or histology in either the presence or absence of tamoxifen (Online Figure VIIH–VIIJ). Taken together, these results suggest that Drp1 downregulation induces LV dysfunction and cardiac hypertrophy at baseline.
Drp1 Downregulation Induces Mitochondrial Elongation and Dysfunction
To examine how Drp1 deletion affects mitochondrial morphology in the heart, EM analysis was performed. At baseline, mitochondria in control mice hearts were primarily rectangular or spherical in shape, whereas tubular mitochondria were observed less frequently. On the contrary, mitochondria in Drp1-CKO mice were mostly elongated/enlarged 4 and 8 weeks after TI (Figure 6A; Online Figure VIIIA). After 48 hours of fasting, mitochondria in control mice hearts became smaller and spherical. In contrast, mitochondria in Drp1-CKO mice remained elongated even after fasting (Figure 6A). Autophagosomes containing mitochondria were observed in control mice hearts after 48 hours of fasting, but not in Drp1-CKO mice hearts (Figure 6A). Quantitative analysis revealed that mitochondrial mass was significantly greater in Drp1-CKO mice hearts than in control mice hearts at baseline and after fasting (Figure 6A). These results suggest that endogenous Drp1 plays an essential role in mediating mitochondrial foreshortening at baseline and during fasting in mouse heart in vivo.
Four or 8 weeks after TI, Drp1 depletion increased the cytochrome c oxidase subunit IV protein level (Figure 6B; Online Figure VIIIB), and mitochondrial DNA content, evaluated with real-time polymerase chain reaction of cytochrome b DNA, was significantly greater in Drp1-CKO mice than in control mice (Figure 6C). These results suggest that the mitochondrial content is increased by Drp1 downregulation. Mitochondrial biogenesis was evaluated by immunoblot analyses of proliferator-activated receptor-gamma coactivator-1α and mitochondrial transcription factor A. Cardiac protein expressions of proliferator-activated receptor-gamma coactivator-1α and mitochondrial transcription factor A did not differ between Drp1-CKO and control mice 4 and 8 weeks after TI, suggesting that Drp1 depletion did not affect mitochondrial biogenesis (Figure 6D; Online Figure VIIIC). However, mitochondrial ATP production was significantly attenuated in Drp1-CKO mice hearts 4 and 8 weeks after TI, compared with control mice hearts (Figure 6E; Online Figure VIIID). The activity of mitochondrial complexes I, II + III, and IV was also significantly attenuated in Drp1-CKO mice hearts 8 weeks after TI, compared with control mice hearts (Online Figure VIIIE). The extent of mPTP opening, as evaluated by the decrease in absorbance at 540 nm in mitochondrial swelling assays, was significantly greater in Drp1-CKO mice hearts 4 weeks after TI than in controls (Figure 6F), suggesting that mPTP opening is accelerated in Drp1-CKO mice. The cardiac levels of 4-hydroxynonenal, a marker of oxidative stress, and mitochondrial production of H2O2, evaluated with Amplex Red assays, were also significantly higher in Drp1-CKO mice 4 weeks after TI than in control mice (Figure 6G and 6H). Thus, Drp1 depletion results in mitochondrial dysfunction and oxidative stress in the heart.
Because the initial assessment of mitochondrial function was conducted using hearts harvested 4 to 8 weeks after TI, when both hypertrophy and LV dysfunction are obvious in Drp1-CKO mice, mitochondrial dysfunction could be secondary to pathological hypertrophy. We, therefore, also investigated an earlier time point. Echocardiographic analyses revealed no significant difference in LV ejection fraction between control and Drp1-CKO mice 10 days after TI (Online Table V), nor was there a significant difference in cardiomyocyte cross-sectional area or cardiac fibrosis (Online Figure VIIIF and VIIIG), confirming that this time point precedes the development of pathological hypertrophy. Nevertheless, mitochondrial function, as assessed by ATP production and mitochondrial swelling assays, was already severely attenuated in Drp1-CKO mice compared with control mice 10 days after TI (Figure 6E and 6F). Together with the observation that Drp1 downregulation directly induces mitochondrial dysfunction in cultured cardiomyocytes (Figure 2E to 2H), these results suggest that Drp1 depletion induces mitochondrial dysfunction in the heart even before manifestation of hypertrophy and LV dysfunction.
We investigated whether Drp1 downregulation in the heart affects apoptosis. There were significantly more TUNEL-positive nuclei in Drp1-CKO mice hearts than in controls 4 and 8 weeks after TI (Figure 7A; Online Figure VIIIH). Cleaved caspase-3 and cytochrome c release into the cytosolic fraction were also significantly elevated in Drp1-CKO mice hearts 4 weeks after TI (Figure 7B), as was the serum HMGB1 (high mobility group box 1) level, an indicator of necrosis (Online Figure VIIII). These results suggest that endogenous Drp1 is required for protection against the mitochondrial mechanisms of apoptosis and necrosis in cardiomyocytes.
Autophagy Is Inhibited in Drp1-CKO Mice
We next investigated the role of Drp1 in mediating autophagy in the heart in vivo. There was significantly less LC3-II and significantly more p62 in Drp1-CKO mice hearts than in controls, 4 weeks after TI (Figure 7C). To examine whether Drp1 downregulation attenuates autophagic flux, we evaluated the effect of chloroquine injection on autophagosome accumulation.20 LC3-II accumulation was suppressed even in the presence of chloroquine, whereas p62/SQSTM1 accumulation did not change significantly after chloroquine treatment in the Drp1-CKO mouse heart (Figure 7D). To further evaluate the level of autophagic flux in cardiomyocytes in vivo, we crossed Drp1-CKO and Drp1 fl/fl with cardiac-specific mRFP-GFP-LC3 (tf-LC3) transgenic mice. Both Drp1-CKO × tf-LC3 and Drp1 fl/fl × tf-LC3 (control tf-LC3) were injected with tamoxifen for 5 days. Fasting increased the number of LC3 dots with both green and red color (appearing yellow in merged images), representing autophagosomes, as well as the number of dots with only red color, representing autolysosomes, in control tf-LC3 mice, indicating increased autophagic flux (Figure 7E). In contrast, the number of yellow and free red dots did not increase in response to fasting in Drp1-CKO × tf-LC3 mice (Figure 7E). Taken together, these results suggest that Drp1 downregulation suppresses autophagic flux at baseline.
Drp1 Depletion Induces Stress Intolerance and Enhances I/R Injury
Although mitochondrial fission and fusion are essential for maintaining mitochondrial quality control, their role in cardiac development and stress resistance remains unknown. To address this question, we crossed Drp1 fl/fl mice with αMHC-MCM mice. Although no mice with Drp1-CKO were born, mice with Drp1-hetCKO were viable at 12 weeks, suggesting that Drp1 is required for normal prenatal development but that 1 functional allele is sufficient during this period. Cardiac Drp1 expression was 40% lower in Drp1-hetCKO mice than in control (Drp1 flox/+) mice (Figure 8A). LV function, assessed by LV ejection fraction, in Drp1-hetCKO mice was normal at 12 weeks of age (Figure 8B). Neither LV weight/tibial length nor lung weight/tibial length differed between 12-week-old Drp1-hetCKO and control mice (Online Table VI). Histological analyses showed that the cardiomyocyte cross-sectional area and myocardial fibrosis also did not differ between 12-week-old Drp1-hetCKO and control mice (Online Figure IXA and IXB). However, ATP production was significantly lower in 12-week-old Drp1-hetCKO mice (Figure 8C), suggesting that mitochondrial dysfunction develops before histological and hemodynamic changes in Drp1-hetCKO mice. The fact that LV function is maintained at 12 weeks of age in Drp1-hetCKO mice allowed us to use these mice to examine the role of Drp1 during stress in the heart.
Mitochondrial Drp1 was significantly increased in response to 48-hour fasting or I/R but not in Drp1-hetCKO mice (Figure 8D). To evaluate the role of endogenous Drp1 in protection against stress in vivo, 12-week-old Drp1-hetCKO and control mice underwent 48-hour fasting. The LV ejection fraction was significantly lower in Drp1-hetCKO mice than in control mice after fasting, suggesting that endogenous Drp1 acts to preserve LV function during fasting (Figure 8E). Similar results were observed in Drp1-CKO mice with tamoxifen treatment (Online Figure IXC). To evaluate the role of endogenous Drp1 in protection against I/R, 12-week-old Drp1-hetCKO and control mice were subjected to 30 minutes of myocardial ischemia followed by 24 hours of reperfusion. EM analyses showed that I/R increased the number of smaller and spherical mitochondria in control mice, suggesting that mitochondrial fission was induced. However, these changes were significantly attenuated in Drp1-hetCKO mice (Figure 8F), suggesting that endogenous Drp1 mediates mitochondrial fission after I/R. Autophagosomes containing mitochondria were observed in control mice hearts but not in Drp1-hetCKO mice hearts after I/R (Figure 8F). There was also significantly less LC3-II and more p62 in Drp1-hetCKO mice hearts than in control hearts at baseline and after I/R (Figure 8G), suggesting that autophagy is suppressed by heterozygous Drp1 downregulation. The infarct size/area at risk after I/R, as evaluated with Alcian Blue and tetrazolium chloride staining, was not affected by αMHC-MCM alone (Online Figure IXD) but was significantly greater in Drp1-hetCKO mice than in control mice (55.2±3.0% versus 40.2±1.6%; P<0.05; n=3 per group; Figure 8H). Similar results were observed in Drp1-CKO mice with tamoxifen treatment (Online Figure IXE).Taken together, these results suggest that inhibition of mitochondrial fission through Drp1 downregulation enhances myocardial injury in response to I/R.
We also evaluated the effect of mdivi-1 on I/R injury. One-time treatment with mdivi-1 just before I/R significantly reduced the infarct size/area at risk (Online Figure XA), confirming previous observations by others.11 However, the same treatment also reduced the infarct size/area at risk in Drp1-hetCKO mice, suggesting that short-term treatment with mdivi-1 protects the heart through Drp1-independent mechanisms (Online Figure XB). Although repetitive applications of mdivi-1 (1.2 mg/kg per day for 7 days) did not significantly reduce LV systolic function (Online Figure XC and Table VII), it significantly increased mitochondrial mass, as determined by EM (Online Figure XD), reduced mitochondrial function to a similar extent as heterozygous Drp1 downregulation, as determined by mitochondrial swelling assays and ATP production (Online Figure XE and XF), and significantly enhanced the infarct size/area at risk after I/R (Online Figure XG), thereby mimicking the effect of Drp1-hetCKO. Thus, although the effects of long-term treatment with mdivi-1 are similar to those of Drp1 downregulation with regard to I/R injury enhancement, albeit weaker, 1-time treatment with mdivi-1 seems to have protective effects, which are most likely independent of Drp1.
Our results suggest that endogenous Drp1 induces mitochondrial foreshortening at baseline and in response to stress in the heart and the cardiomyocytes therein. Contrary to previous reports,11,28 downregulation of endogenous Drp1 in cardiomyocytes induces mitochondrial dysfunction and apoptosis, despite significant induction of mitochondrial elongation, thereby inducing cardiac dysfunction at baseline and exacerbating myocardial injury in response to I/R. Using Keima-MLS, we show that Drp1 plays an essential role in mediating lysosomal removal of mitochondria in cardiomyocytes. Thus, our results suggest that endogenous Drp1 contributes to mitochondrial quality control.
Although it is generally thought that fused mitochondria function better, Drp1 downregulation significantly increased the number of cardiomyocytes with depolarized mitochondria even at baseline. Although Drp1 is localized primarily in the cytosol in unstimulated cardiomyocytes, a low level of mitochondrial turnover mediated by Drp1 seems essential to maintain mitochondrial function in cardiomyocytes. Given that even heterozygous Drp1 downregulation induces mitochondrial dysfunction and heart failure in mice, it seems that endogenous Drp1 plays an essential role in mitochondrial quality control in the heart in vivo as well.
Whether mitochondria undergo fusion or fission during stress may depend on cell type and stress. In MEF cells,12,13 fasting induces mitochondrial fusion induced by phosphorylation of Drp1 at Ser637 by protein kinase A and translocation of Drp1 to the cytoplasm, which allows mitochondria to maintain ATP synthesis and escape autophagic destruction.12,13 On the contrary, in HL1 cells in vitro and cardiomyocytes in the heart in vivo,11,29 fasting and hypoxia stimulate mitochondrial fission. Regardless of whether fusion or fission is stimulated by stress, these studies showed that suppression of fission and stimulation of fusion through Drp1 downregulation, expression of dominant-negative Drp1, mdivi-1, or expression of mitofusin 1/2 promotes ATP production and cell survival. Here, we show that mitochondria in cardiomyocytes transiently undergo elongation during GD, but that the number of mitochondria with foreshortening also increases thereafter, accompanied by accumulation of Drp1 in mitochondria. Drp1 downregulation in this scenario blunted foreshortening of mitochondria and exacerbated cell death, suggesting that the induction of foreshortening is adaptive in cardiomyocytes.
Our results suggest that endogenous Drp1 is important in mediating autophagy in cardiomyocytes. Drp1 controls autophagic flux at least at the level of autophagosome formation because there were fewer GFP-LC3 puncta when Drp1 was downregulated in the presence of chloroquine, an inhibitor of autophagosome–lysosome fusion or autophagic flux.20 The suppressive effect of Drp1 downregulation upon global autophagy, rather than its specific effect on mitochondria-specific autophagy, was unexpected. We here show that Drp1 physically interacts with Bcl-2/Bcl-xL and that downregulation of Drp1 promotes interaction between Beclin1 and Bcl-2/Bcl-xL. Because Bcl-2 and Bcl-xL are endogenous inhibitors of Beclin1 (Pattingre), increased interaction between Beclin1 and Bcl-2/Bcl-xL in the presence of Drp1 downregulation should lead to suppression of autophagy. In fact, the suppression of general autophagy by Drp1 downregulation was rescued by downregulation of Bcl-xL, indicating the critical role of the Bcl-2 family proteins in this process.
We here show that a GD-induced increase in lysosomal localization of Keima-MLS19 is attenuated in the presence of Drp1 downregulation. Given the mitochondrial localization of Keima-MLS and that Keima-MLS puncta with high 560/440, indicating acidic pH, are localized in lysosomes and are abolished when Beclin1 is downregulated, increases in Keima-MLS puncta with high 560/440 presumably reflect autophagic degradation of mitochondria. Thus, the significant reduction in lysosomal Keima-MLS puncta, together with EM images showing a significant reduction in autophagosomes primarily containing mitochondria, in Drp1 knockdown cardiomyocytes indicates that endogenous Drp1 plays an essential role in mediating GD-induced increases in mitochondrial autophagy. The Keima-MLS analysis was not sensitive enough to demonstrate a reduction in lysosomal removal of mitochondria at baseline when Drp1 is downregulated. However, given that dysfunctional mitochondria accumulate in Drp1-downregulated cardiomyocytes, it is likely that Drp1 also mediates autophagic mitochondrial degradation at baseline.
In this work, we used the term “mitochondrial autophagy” to describe the clearance of mitochondria by autophagy. Although our results suggest that Drp1 regulates mitochondrial clearance through general autophagy, whether or not Drp1 also affects mitochondria-selective autophagy, namely mitophagy, could not be evaluated because of technical limitations. To this end, specific assays to accurately evaluate the presence of mitophagy and specific interventions to modulate mitophagy seem essential.
Conditional Drp1 downregulation leads to decreases in cardiac function within 4 weeks, and all animals died within 13 weeks caused by heart failure. Histological analyses showed that the Drp1 deficiency induces hypertrophy and fibrosis in the heart and increases cardiomyocyte apoptosis. The fact that conditional cardiac-specific combined downregulation of mitofusin 1 and mitofusin 2 also leads to rapid development of cardiac dysfunction within 2 weeks9 indicates that both unopposed fission and unopposed fusion of mitochondria may cause cardiac dysfunction and suggests the critical importance of mitochondrial remodeling in the heart.
There are some differences between the cardiac phenotypes of Drp1-CKO and cardiac-specific combined downregulation of mitofusin 1 and mitofusin 2 knockout mice.9,10 For example, neither cardiac hypertrophy nor the increased cardiomyocyte apoptosis observed in Drp1-CKO were apparent in cardiac-specific combined downregulation of mitofusin 1 and mitofusin 2 knockout mice. This suggests that ATP depletion may be more profound in the absence of fusion than in the absence of fission.
The reason for the opposite effects of Drp1 downregulation by genetic deletion and Drp1 suppression with mdivi-1 in response to I/R remains to be elucidated. One possibility is that our shRNA treatment may have induced stronger, more prolonged suppression of mitochondrial fission than a single dose of mdivi-1 at 50 μmol/L, the concentration used by others.10 We observed modest cell death–promoting effects when cardiomyocytes were treated with mdivi-1 at a higher concentration (100 μmol/L) or multiple times. A second possibility is that Drp1 downregulation may induce more potent suppression of general autophagy to even below physiological levels than mdivi-1. Although suppression of excessive autophagy may be salutary, suppression below physiological levels may be harmful. Third, mdivi-1 may more strongly suppress cell death than Drp1 downregulation by directly acting on apoptosis mechanisms. Mdivi-1 blocks Bax/Bak-dependent release of both Smac/Diablo and cytochrome c in HeLa cells,14 and we found that mdivi-1 inhibited chelerythrine-induced apoptosis in cardiomyocytes, which Drp1 downregulation did not. Furthermore, 1-time treatment with mdivi-1 reduced I/R injury even in Drp1-hetCKO mice, suggesting that mdivi-1 most likely has a Drp1-independent antiapoptotic function. Along this same line, mdivi-1 affects other molecules besides Drp1, including delayed rectifier K+ channels,27 raising the issue of specificity of the chemical inhibitor. Fourth, mitochondrial localization of Drp1 is positively regulated by protein kinase A,30 calcineurin,30 PUMA,31 Bax/Bak,32 ceramide,33 and O-linked-β-N-acetylglucosamine modification,34 and is negatively regulated by miR-49935 and Pim1.36 Thus, some experimental conditions may induce excessive Drp1 activation/upregulation, which may in turn induce deleterious effects in cardiomyocytes. In fact, Drp1 overexpression in cardiomyocytes above the level caused by GD induced cell death. Drp1 suppression by mdivi-1 may be protective under such experimental conditions.
We have shown previously that Beclin1 haploinsufficiency inhibits I/R injury and suppresses autophagy.37 Here, we show that Drp1 haploinsufficiency exacerbates I/R injury but is also accompanied by suppression of autophagy. At present, mechanisms explaining the difference remain to be clarified. Drp1 downregulation may have a more pronounced effect on general autophagy and mitochondrial autophagy than Beclin1 downregulation, thereby suppressing autophagy below physiological levels. Another possibility is that Drp1 downregulation may more globally affect mitochondrial quality control mechanisms, including inducing unopposed mitochondrial elongation and suppression of global autophagy, rather than being limited to suppression of autophagic mitochondrial degradation. Further investigation is required to address this issue.
In summary, persistent Drp1 downregulation inhibits clearance of mitochondria by autophagy and causes mitochondrial dysfunction and consequent cell death in the heart and in the cardiomyocytes therein, both at baseline and under stress conditions. Drp1 plays an important role in mediating mitochondrial foreshortening and autophagic mitochondrial degradation in cardiomyocytes.
We thank Christopher D. Brady and Daniela Zablocki for critical reading of the article and Luke Fritzky for technical assistance.
Sources of Funding
This work was supported in part by US Public Health Service Grants HL102738, HL67724, HL112330, HL91469, and AG23039. This work was also supported by the Fondation Leducq Transatlantic Networks of Excellence. Dr Ikeda has been supported by a Postdoctoral Fellowship from the Founders Affiliate, American Heart Association, and by a grant from the Rotary Foundation Ambassadorial Scholarship.
In September, 2014, the average time from submission to first decision for all original research papers submitted to Circulation Research was 14.29 days.
The online-only Data Supplement is available with this article at http://circres.ahajournals.org/lookup/suppl/doi:10.1161/CIRCRESAHA.116.303356/-/DC1.
- Nonstandard Abbreviations and Acronyms
- adenovirus harboring Drp1 shRNA
- adenovirus harboring scramble shRNA
- alpha myosin heavy chain
- dynamin-related protein 1
- cardiac-specific conditional Drp1 knockout
- cardiac-specific heterozygous Drp1 knockout
- electron microscopic/microscopy
- glucose deprivation
- Keima with mitochondrial localization signal
- left ventricular
- mitochondrial permeability transition pore
- oxygen consumption rate
- tamoxifen injection
- Received January 19, 2014.
- Revision received October 13, 2014.
- Accepted October 17, 2014.
- © 2014 American Heart Association, Inc.
- Song M,
- Chen Y,
- Gong G,
- Murphy E,
- Rabinovitch PS,
- Dorn GW 2nd.
- Youle RJ,
- van der Bliek AM
- Twig G,
- Elorza A,
- Molina AJ,
- et al
- Otera H,
- Mihara K
- Chen Y,
- Liu Y,
- Dorn GW II.
- Papanicolaou KN,
- Kikuchi R,
- Ngoh GA,
- Coughlan KA,
- Dominguez I,
- Stanley WC,
- Walsh K
- Ong SB,
- Subrayan S,
- Lim SY,
- Yellon DM,
- Davidson SM,
- Hausenloy DJ
- Rambold AS,
- Kostelecky B,
- Elia N,
- Lippincott-Schwartz J
- Cassidy-Stone A,
- Chipuk JE,
- Ingerman E,
- Song C,
- Yoo C,
- Kuwana T,
- Kurth MJ,
- Shaw JT,
- Hinshaw JE,
- Green DR,
- Nunnari J
- Tanaka A,
- Cleland MM,
- Xu S,
- Narendra DP,
- Suen DF,
- Karbowski M,
- Youle RJ
- Lee Y,
- Lee HY,
- Hanna RA,
- Gustafsson ÅB
- Hariharan N,
- Maejima Y,
- Nakae J,
- Paik J,
- Depinho RA,
- Sadoshima J
- Pattison JS,
- Osinska H,
- Robbins J
- Sciarretta S,
- Zhai P,
- Shao D,
- Maejima Y,
- Robbins J,
- Volpe M,
- Condorelli G,
- Sadoshima J
- Matsuda N,
- Sato S,
- Shiba K,
- Okatsu K,
- Saisho K,
- Gautier CA,
- Sou YS,
- Saiki S,
- Kawajiri S,
- Sato F,
- Kimura M,
- Komatsu M,
- Hattori N,
- Tanaka K
- Bhandari P,
- Song M,
- Chen Y,
- Burelle Y,
- Dorn GW 2nd.
- Disatnik MH,
- Ferreira JC,
- Campos JC,
- Gomes KS,
- Dourado PM,
- Qi X,
- Mochly-Rosen D
- Cribbs JT,
- Strack S
- Wang JX,
- Li Q,
- Li PF
- Wasiak S,
- Zunino R,
- McBride HM
- Parra V,
- Eisner V,
- Chiong M,
- Criollo A,
- Moraga F,
- Garcia A,
- Härtel S,
- Jaimovich E,
- Zorzano A,
- Hidalgo C,
- Lavandero S
- Gawlowski T,
- Suarez J,
- Scott B,
- Torres-Gonzalez M,
- Wang H,
- Schwappacher R,
- Han X,
- Yates JR 3rd.,
- Hoshijima M,
- Dillmann W
- Din S,
- Mason M,
- Völkers M,
- Johnson B,
- Cottage CT,
- Wang Z,
- Joyo AY,
- Quijada P,
- Erhardt P,
- Magnuson NS,
- Konstandin MH,
- Sussman MA
- Matsui Y,
- Takagi H,
- Qu X,
- Abdellatif M,
- Sakoda H,
- Asano T,
- Levine B,
- Sadoshima J
Novelty and Significance
What Is Known?
Combined genetic downregulation of mitofusin 1 and mitofusin 2, mitochondrial outer membrane proteins regulating mitochondrial fusion, causes mitochondrial fragmentation and dysfunction and heart failure in mice.
Dynamin-related protein 1 (Drp1) is a GTPase that mediates mitochondrial fission in noncardiac cells.
Pharmacological suppression of Drp1 with mdivi-1 attenuates myocardial injury in response to ischemia/reperfusion.
What New Information Does This Article Contribute?
Chronic downregulation of Drp1 induces elongation of mitochondria, mitochondrial dysfunction, heart failure and premature death in mice.
Downregulation of Drp1 inhibits general autophagy and movement of mitochondrial proteins into lysosomes in cardiomyocytes.
Chronic downregulation of Drp1 enhances myocardial injury in response to ischemia/reperfusion.
Mitochondria have the ability to remove damaged parts through the process of fission and fusion and consequent degradation through autophagy. Using a loss-of-function mouse model, we show that chronic downregulation of Drp1, a GTPase known to induce mitochondrial fission, causes mitochondrial dysfunction, myocardial cell death, heart failure, and the death of the animal. In vitro analyses show that genetic downregulation of Drp1 directly inhibits general autophagy in cardiomyocytes through Bcl-xL–dependent mechanisms. Furthermore, using mito-Keima, a pH-sensitive protein, we show that endogenous Drp1 is essential for mitochondrial autophagy in response to glucose starvation in cardiomyocytes. Downregulation of Drp1 in turn causes accumulation of dysfunctional mitochondria and increased cell death. Furthermore, downregulation of Drp1 exacerbates ischemia/reperfusion injury in the mouse heart in vivo. These results suggest that endogenous Drp1 plays an important role in mediating mitochondrial autophagy and maintaining mitochondrial function in response to stress.