Restoration of mitophagy ameliorates cardiomyopathy in Barth syndrome

ABSTRACT

Barth syndrome (BTHS) is an X-linked genetic disorder caused by mutations in the TAFAZZIN/Taz gene which encodes a transacylase required for cardiolipin remodeling. Cardiolipin is a mitochondrial signature phospholipid that plays a pivotal role in maintaining mitochondrial membrane structure, respiration, mtDNA biogenesis, and mitophagy. Mutations in the TAFAZZIN gene deplete mature cardiolipin, leading to mitochondrial dysfunction, dilated cardiomyopathy, and premature death in BTHS patients. Currently, there is no effective treatment for this debilitating condition. In this study, we showed that TAFAZZIN deficiency caused hyperactivation of MTORC1 signaling and defective mitophagy, leading to accumulation of autophagic vacuoles and dysfunctional mitochondria in the heart of Tafazzin knockdown mice, a rodent model of BTHS. Consequently, treatment of TAFAZZIN knockdown mice with rapamycin, a potent inhibitor of MTORC1, not only restored mitophagy, but also mitigated mitochondrial dysfunction and dilated cardiomyopathy. Taken together, these findings identify MTORC1 as a novel therapeutic target for BTHS, suggesting that pharmacological restoration of mitophagy may provide a novel treatment for BTHS.

Abbreviations: BTHS: Barth syndrome; CCCP: carbonyl cyanide 3-chlorophenylhydrazone; CL: cardiolipin; EIF4EBP1/4E-BP1: eukaryotic translation initiation factor 4E binding protein 1; GAPDH: glyceraldehyde-3-phosphate dehydrogenase; KD: knockdown; KO: knockout; LAMP1: lysosomal-associated membrane protein 1; LV: left ventricle; MAP1LC3/LC3: microtubule-associated protein 1 light chain 3; MEFs: mouse embryonic fibroblasts; MTORC1: mechanistic target of rapamycin kinase complex 1; OCR: oxygen consumption rate; PE: phosphatidylethanolamine; PIK3C3/VPS34: phosphatidylinositol 3-kinase catalytic subunit type 3; PINK1: PTEN induced putative kinase 1; PRKN/Parkin: parkin RBR E3 ubiquitin protein ligase; qRT-PCR: quantitative real-time polymerase chain reaction; RPS6KB/S6K: ribosomal protein S6 kinase beta; SQSTM1/p62: sequestosome 1; TLCL: tetralinoleoyl cardiolipin; WT: wild-type.

KEYWORDS:

• BTHS
• cardiolipin
• mitophagy
• MTORC1
• rapamycin
• TAFAZZIN

Introduction

Barth syndrome (BTHS) is a lethal, X-linked genetic disorder caused by mutations in the TAFAZZIN/Taz gene and characterized by dilated cardiomyopathy, exercise intolerance, chronic fatigue, delayed growth, and neutropenia [1]. TAFAZZIN, a mitochondrial protein located in membranes lining the intermembrane space [2,3], is a phospholipid transacylase required for cardiolipin (CL) remodeling, resulting in its mature acyl composition with predominance of tetralinoleoyl cardiolipin (TLCL) which contains four linoleolic acid chains (C18:2) in heart, liver, and skeletal muscle [3,4]. Mutations in the TAFAZZIN gene result in TLCL depletion and accumulation of monolysocardiolipin, which is believed to cause mitochondrial dysfunction, cardiomyopathy, neutropenia, and growth defects in BTHS [1]. However, the molecular mechanisms by which TAFAZZIN deficiency and loss of TLCL cause mitochondrial dysfunction in BTHS remain poorly defined. Currently, there are no targeted therapies for BTHS.

CL is a mitochondrial phospholipid which plays a critical role in maintaining respiratory chain enzyme function, membrane stability, mitochondrial dynamics, mtDNA biogenesis, and mitochondrial quality control [5–8]. Depletion of TLCL is also associated with aging and various aging-related diseases [7,9–11]. Recent studies have implicated a role of CL in mitophagy, a selective autophagy of mitochondria that plays a critical role in mitochondrial quality control and cell survival by removing damaged mitochondria [12]. CL externalization to the outer mitochondrial membrane serves as a signal for elimination of damaged mitochondria from neuronal cells [12]. Defective mitophagy causes inefficient mitochondrial respiration and a high level of oxidative stress [13]. Previous studies have shown that TAFAZZIN deficiency caused defective mitophagy in mouse embryonic fibroblasts (MEFs) and HeLa cells, leading to impaired oxidative phosphorylation and oxidative stress [14,15]. TAFAZZIN deficiency also caused accumulation of structurally abnormal mitochondria with inefficient oxidative phosphorylation in BTHS patient tissues and experimental mouse models [16–19], suggesting that the clearance of damaged mitochondria via mitophagy is impaired in BTHS. However, molecular mechanisms by which TAFAZZIN deficiency causes defective mitophagy as well as the potential contribution of defective mitophagy to the onset of cardiomyopathy in BTHS remain poorly understood.

Using a mouse model with inducible Tafazzin knockdown (KD) as an animal model of BTHS, we showed in this study that TAFAZZIN deficiency caused hyperactivation of MTOR (mechanistic target of rapamycin kinase) complex 1 (MTORC1), which significantly impaired mitophagy in the heart and skeletal muscle. Consequently, inhibition of MTORC1 by rapamycin not only mitigated the associated cardiomyopathy, but also restored mitophagy, mitochondrial morphology and respiration in the heart of TAFAZZIN KD mice, implicating MTORC1 and mitophagy as novel therapeutic targets for the treatment of BTHS.

Results

Inducible Tafazzin knockdown causes hyperactivation of MTORC1 in mice heart

BTHS is caused by the mutations in the TAFAZZIN gene on X-chromosome. We used mice with inducible Tafazzin knockdown previously generated by Khuchua et al. [20] as a rodent model of BTHS to identify the underlying causes of mitochondrial dysfunction in BTHS. The wild-type (WT) and heterozygotes transgenic male mice were continuously fed with water containing doxycycline (2 mg/mL) for 8 months after weaning (21-days-old) to induce knockdown of Tafazzin mRNA. As previously reported, doxycycline efficiently induced knockdown of Tafazzin mRNA level in the heart, skeletal muscle, and liver of transgenic mice (TAFAZZIN KD), as evidenced by results from the quantitative real-time polymerase chain reaction (qRT-PCR) analysis of Tafazzin mRNA expression (Figure 1A), and Western blot analysis of TAFAZZIN protein expression (Figure 1B). Previous studies demonstrated that TAFAZZIN deficiency induced hypertrophy in both primary neonatal ventricular fibroblasts and cardiomyocytes [21,22], and impaired mitophagy in BTHS [14,15], but the underlying causes for these defects remain poorly understood. To answer these questions, we first determined the signaling pathways changed by TAFAZZIN deficiency in BTHS mice. Interestingly, we found that TAFAZZIN deficiency induced MTORC1 hyperactivation in the heart of TAFAZZIN KD mice, as revealed by up-regulated phosphorylation levels of RPS6KB/S6K (ribosomal protein S6 kinase beta) and EIF4EBP1/4E-BP1 (eukaryotic translation initiation factor 4E binding protein 1), two downstream targets of MTORC1 signaling (Figure 1C, quantified in Figure 1D and E). Consistent with the findings, TAFAZZIN deficiency also significantly enhanced phosphorylation levels of AKT, RPS6KB, and EIF4EBP1 in the heart of TAFAZZIN KD mice relative to the WT control mice in response to insulin stimulation (Figure 1F). We next investigated the effect of TAFAZZIN deficiency on MTORC1 signaling in MEFs isolated from WT and TAFAZZIN KD mice and cultured the MEFs in medium containing doxycycline. The knockdown efficiency of TAFAZZIN was confirmed by Western blot analysis (Figure S1A). Consistent with findings from the heart, TAFAZZIN deficiency also significantly increased phosphorylation levels of RPS6KB and EIF4EBP1 in isolated MEFs cultured in medium containing doxycycline (Figure 1G), which were significantly inhibited in response to treatment with rapamycin and torin1, two potent inhibitors of MTORC1, as indicated by results from the Western blot analysis (Figure 1G). In contrast to torin1 which inhibited the phosphorylation of both RPS6KB and EIF4EBP1, rapamycin selectively inhibited the phosphorylation of RPS6KB, but not EIF4EBP1.

Figure 1. TAFAZZIN deficiency causes MTORC1 hyperactivation. (A) qRT-PCR analysis of Tafazzin mRNA expression in the heart, skeletal muscle, and liver of male WT and TAFAZZIN KD mice fed with water containing doxycycline (2 mg/mL) after weaning (21-days-old) for 2 months. (B) Western blot analysis of TAFAZZIN protein expression in the heart, skeletal muscle, and liver of WT and TAFAZZIN KD mice. (C) Western blot analysis of MTORC1 signaling pathways in the heart of WT and TAFAZZIN KD mice. (D and E) Statistical analysis of phosphorylated levels of RPS6KB (D) and EIF4EBP1 (E) in the heart of WT and TAFAZZIN KD mice. (F) Western blot analysis of AKT-MTORC1 signaling pathway in the heart of WT and TAFAZZIN KD mice in response to insulin stimulation. (G) Western blot analysis of MTORC1 activity in primary MEFs isolated from WT and TAFAZZIN KD mice. MEFs were cultured in medium containing doxycycline (1 μg/mL) for 3 days, and then treated with rapamycin (1 μM) or torin1 (1 μM) for 30 min. Data are represented as mean ± SD. **p < 0.01, ***p < 0.001 by Student’s t test.

TAFAZZIN deficiency causes defective mitophagy in mice heart

Our previous study showed that knockdown of TAFAZZIN in MEFs caused defects in the initiation of mitophagy [14]. However, the underlying mechanisms are not fully understood. Under nutrient-rich condition, MTORC1 activation promotes cell growth and inhibits autophagy [23]. The hyperactivation of MTORC1 in the heart of TAFAZZIN KD mice prompted us to investigate whether TAFAZZIN deficiency leads to defective mitophagy in the heart. We first analyzed the mitochondrial morphology in the heart of WT and TAFAZZIN KD mice by electron microscopy analysis. As shown in Figure 2A, TAFAZZIN deficiency caused disarray of mitochondrial morphology, including fragmentation, irregular shape and distribution, and a loss in mitochondrial cristae structure. Remarkably, TAFAZZIN deficiency also led to accumulation of autophagic vacuoles wrapped by mitochondria (Figure 2A, right panel, highlighted by arrows, quantified in Figure 2B), indicating defects in mitophagy. To identify the underlying causes for the defects, we next analyzed expression levels of several biomarkers involved in autophagy and mitophagy in the heart of WT and TAFAZZIN KD mice by Western blot analysis. During autophagic initiation, a cytosolic form of MAP1LC3/LC3 (microtubule-associated protein 1 light chain 3; LC3-I) is conjugated to phosphatidylethanolamine (PE) to form the LC3–PE conjugate (LC3-II), which is recruited to phagophore membranes, a key step required for autophagic initiation. In support of this notion, the ratio of LC3-II:LC3-I, a key indicator of autophagosome biogenesis, was significantly decreased in the heart of TAFAZZIN KD mice (Figure 2C, quantified in Figure 2D). In contrast, the expression level of SQSTM1/p62 (sequestosome 1), a receptor protein of autophagy, was significantly increased in the heart of the TAFAZZIN KD mice (Figure 2C, quantified in Figure 2E). Additionally, PIK3C3/VPS34 (phosphatidylinositol 3-kinase catalytic subunit type 3), the only known class III phosphatidylinositol 3-kinase which plays an essential role in autophagic initiation in eukaryotic cells, was dramatically increased by TAFAZZIN deficiency in the heart of TAFAZZIN KD mice (Figure 2C, quantified in Figure 2F). PINK1 (PTEN induced putative kinase 1) and PRKN/PARKIN (parkin RBR E3 ubiquitin protein ligase) are two important biomarkers of mitophagy. In the process of mitophagy, PINK1 accumulates on defective mitochondria, eliciting the translocation of PRKN from the cytosol to mitochondria, a key step required for the clearance of damaged mitochondria. Consistent with the defective mitophagy, we further found that both the expression levels of PINK1 and PRKN were significantly increased in the heart of TAFAZZIN KD mice (Figure 2C, quantified in Figure 2G and H, respectively), suggesting a downstream defect in autophagic consumption. In support of this notion, the glycosylated form of LAMP1 (lysosomal-associated membrane protein 1), a key regulator of lysosomal integrity, pH, and catabolism [24], was significantly activated, which is evidenced by upward band shift in the heart of the TAFAZZIN KD mice (Figure 2C, highlighted by an arrow, quantified in Figure 2I), suggesting that TAFAZZIN deficiency may also impair lysosomal function.

Figure 2. TAFAZZIN deficiency leads to defective mitophagy in mice heart. (A) Electron microscopy analysis of cardiac mitochondrial morphology in WT and TAFAZZIN KD mice. Scale bar: 2 μm. (B) Quantification of autophagic vacuoles in the heart of WT and TAFAZZIN KD mice. n = 3 mice per group. (C) Western blot analysis of the expression of glycosylated LAMP1 and autophagic biomarkers, including PIK3C3, LC3, SQSTM1, PINK1, and PRKN, in the heart of WT and TAFAZZIN KD mice. (D) Statistical analysis of LC3-II:LC3-I ratio in the heart of WT and TAFAZZIN KD mice. (E-I) Quantification of the protein expression levels, including SQSTM1 (E), PIK3C3 (F), PINK1 (G), PRKN (H) and glycosylated LAMP1 (I), in the heart of WT and TAFAZZIN KD mice. Data are represented as mean ± SD. *p < 0.05, **p < 0.01, ***p < 0.001 by Student’s t test.

Rapamycin treatment restores left ventricle function in TAFAZZIN KD mice

The MTORC1 signaling pathway integrates mitogen and nutrient signals to control cell growth and proliferation [25], as well as autophagy [23]. Hyperactivation of MTORC1 is implicated in the pathogenesis of cardiomyopathy [26]. We next determined whether attenuation of MTORC1 signaling by rapamycin would prevent the development of left ventricle (LV) dysfunction in TAFAZZIN KD mice, a major metabolic defect in BTHS that leads to premature death. The WT and TAFAZZIN KD male mice were fed with water containing doxycycline starting from weaning (21-days-old) to induce Tafazzin mRNA depletion in the TAFAZZIN KD mice. At 3 months of age, the TAFAZZIN KD mice were randomly divided into two groups. One group of TAFAZZIN KD mice were fed with diet containing 14 ppm active rapamycin encapsulated in eudragit, a dosage previously showed to extend lifespan in mice [27] and restore cardiac and skeletal muscle function in LMNA/Lamin A/C-deficient mice [28], while the other group of TAFAZZIN KD mice, as well as the WT control mice, were fed with control eudragit diet for 6 consecutive months. At the end of rapamycin treatment, mice were subjected to cardiac functional analysis by echocardiography (Figure 3A). In contrast to the WT mice, the TAFAZZIN KD mice fed with control diet exhibited LV dilation and ejection dysfunction, as evidenced by the representative images from echocardiography (Figure 3B), reduced ejection fraction (Figure 3C), and fractional shortening (Figure 3D). In contrast, rapamycin treatment not only mitigated dilated cardiomyopathy, but also restored the LV ejection fraction and fractional shortening in TAFAZZIN KD mice (Figure 3B–D). Moreover, results from echocardiographic analysis also revealed that rapamycin treatment significantly attenuated dilated cardiomyopathy, as evidenced by decreased LV internal diameter at end systole (LVID-s, Figure 3E), but not at end diastole (LVID-d, Figure 3F). Furthermore, rapamycin administration restored the thickness of systolic LV interventricular septum (IVS-s) (Figure 3G) without significant effect on diastolic LV interventricular septum (IVS-d, Figure 3H). However, rapamycin treatment failed to restore thickness of LV posterior wall (LVPW) both at end systole and end diastole, which were significantly reduced by TAFAZZIN deficiency (Figure 3I–J).

Figure 3. Rapamycin attenuates cardiomyopathy and LV dysfunction in TAFAZZIN KD mice. (A) Experimental outline of rapamycin administration in TAFAZZIN KD mice. (B) Representative images of the echocardiographic analysis of cardiac function of WT and TAFAZZIN KD mice fed with control or rapamycin diet. LVID-s and LVID-d, LV internal diameter at end systole and at end diastole, respectively. (C and D) Echocardiographic analysis of LV ejection fraction (EF, C) and fractional shortening (FS, D) in WT and TAFAZZIN KD mice. (E-J) Echocardiographic analysis of LVID-s (E), LVID-d (F), interventricular septum thickness at end systole (IVS-s, G) and at end diastole (IVS-d, H), and LV posterior wall thickness at end systole (LVPW-s, I) and at end diastole (LVPW-d, J) of WT and TAFAZZIN KD mice fed with control or rapamycin diet. n = 5–8. Data are represented as mean ± SD. *p < 0.05, **p < 0.01, ***p < 0.001 by one-way ANOVA. n.s, no significance.

Rapamycin mitigates cardiomyocyte hypertrophy in TAFAZZIN KD mice

We next investigated whether rapamycin administration would attenuate the hyperactivation of MTORC1 in the heart of TAFAZZIN KD mice. Consistent with findings from echocardiography, rapamycin significantly inhibited the hyperactivation of MTORC1 in the heart of TAFAZZIN KD mice, as evidenced by the down-regulated phosphorylation of MTOR and its downstream targets, including RPS6KB and EIF4EBP1 (Figure 4A). In addition, rapamycin significantly attenuated the hyperactivation of MTORC1 in the skeletal muscle of TAFAZZIN KD mice (Figure S2A-S2C). Moreover, TAFAZZIN deficiency caused hypertrophy of cardiomyocytes, as shown by hematoxylin and eosin (H&E) staining (Figure 4B) and quantitative analysis of the average cell size of cardiomyocytes in LV (Figure 4C). In further support of the role of hyperactivation of MTORC1 in dilated cardiomyopathy in TAFAZZIN KD mice, these defects were significantly attenuated by treatment with rapamycin (Figure 4B–C). Furthermore, TAFAZZIN deficiency significantly increased the expression of hypertrophic biomarkers, including Nppb/Bnp (natriuretic peptide type B), Myh7/β-mhc (myosin heavy chain 7), Nppa/Anf (natriuretic peptide type A), and Acta1 (actin alpha 1, skeletal muscle) in the heart of TAFAZZIN KD mice, as evidenced by results from qRT-PCR analysis (Figure 4D–G). Again, rapamycin treatment significantly attenuated mRNA expression levels of these hypertrophic biomarkers (Figure 4D–G).

Figure 4. Rapamycin attenuates cardiomyocyte hypertrophy through inhibition of MTORC1 signaling. (A) Western blot analysis of MTORC1 signaling pathways in the heart of WT and TAFAZZIN KD mice fed with control or rapamycin diet. (B) H&E staining of the LV sections of WT and TAFAZZIN KD mice. Scale bar: 40 μm. (C) Quantitative analysis of cardiomyocytes size of WT and TAFAZZIN KD mice fed with control or rapamycin diet. n = 3 mice per group, 100 cells per mouse were used for quantification analysis. (D-G) qRT-PCR analysis of mRNA levels of major biomarkers associated with hypertrophic cardiomyopathy, including Nppb (D), Myh7 (E), Nppa (F), and Acta1 (G) in the heart of WT and TAFAZZIN KD mice fed with control or rapamycin diet. Data are represented as mean ± SD. *p < 0.05, **p < 0.01, ***p < 0.001 by one-way ANOVA.

Rapamycin restores mitophagy in the heart of TAFAZZIN KD mice

We next examined the effect of rapamycin treatment on mitochondrial morphology in the heart by electron microscopy analysis. Compared with the WT mice, TAFAZZIN deficiency caused disarray of mitochondrial morphology, mitochondrial fragmentation, and irregular distribution between sarcomeres (Figure 5A, mitochondrial size was quantified in Figure 5B). In addition, TAFAZZIN deficiency caused significant enlargement of the sarcomeres (Figure 5A, quantified in Figure 5C). Consistent with a defect in mitophagy, TAFAZZIN deficiency also caused massive accumulation of autophagic vacuoles (Figure 5A, highlighted by arrows, quantified in Figure 5D). In contrast, these defects were significantly mitigated by treatment with rapamycin. Consequently, rapamycin treatment not only mitigated mitochondrial fragmentation, but also eliminated autophagic vacuoles, leading to normalization of sarcomere width in the heart of TAFAZZIN KD mice (Figure 5A, quantified in Figure 5B–D). The findings further suggested that mitochondrial vacuolization may be caused by defective mitophagy in the heart of TAFAZZIN KD mice. This notion is supported by results from Western blot analysis of the major autophagic biomarkers, including PIK3C3, LC3, SQSTM1, PINK1 and PRKN. TAFAZZIN deficiency significantly increased SQSTM1 expression and LC3-I:LC3-II ratio, two key negative indicators of autophagy (Figure 5E, quantified in 5 F and 5 G), and upregulated the expression of both PINK1 and PRKN, suggesting a downstream defect in autophagic consumption (Figure 5E, quantified in H and I). In support of this notion, TAFAZZIN deficiency also promoted glycosylation of LAMP1, a defect implicated in lysosomal dysfunction associated with Niemann-Pick disease [29]. In contrast, rapamycin treatment not only normalized SQSTM1 expression level and the LC3-I:LC3-II ratio, but also the expression levels of PINK1 and PRKN in the heart of TAFAZZIN KD mice (Figure 5E, quantified in Figure 5F–I). Furthermore, rapamycin treatment significantly attenuated the abnormal glycosylation of LAMP1, as evidenced by a decrease in band size in comparison with the vehicle treated TAFAZZIN KD mice, without significant effect on the expression level of glycosylated LAMP1 (Figure 5E, highlighted by an arrow). In addition to heart, rapamycin partially mitigated defective mitophagy in the skeletal muscle of TAFAZZIN KD mice, as evidenced by the increased ratio of LC3-II:LC3-I bands on the Western blot analysis (Figure S2A and S2D).

Figure 5. Rapamycin mitigates autophagic vacuoles and restores mitophagy in the heart of TAFAZZIN KD mice. (A) Electron microscopy analysis of cardiac mitochondrial morphology in the WT and TAFAZZIN KD mice fed with control or rapamycin diet. Scale bar: 4 μm. Arrows highlight the mitochondria with vacuole. S, sarcomere. (B) Quantitative analysis of cardiac mitochondrial size. n = 3 mice per group, and 100 mitochondria per mouse were counted. (C) Statistical analysis of sarcomere width in the heart of WT and TAFAZZIN KD mice fed with control or rapamycin diet. (D) Quantitative analysis of autophagic vacuoles in the heart of WT and TAFAZZIN KD mice fed with control or rapamycin diet. n = 3 mice per group, and 100 mitochondria per mouse were counted. (E) Western blot analysis of LAMP1 and biomarkers associated with autophagy and mitophagy, including PIK3C3, LC3, SQSTM1, PINK1 and PRKN, in the heart of WT and TAFAZZIN KD mice fed with control or rapamycin diet. (F-I) Statistical analysis of protein expression levels, including SQSTM1 (F), LC3-II:LC3-I ratio (G), PINK1 (H), and PRKN (I), in the heart of WT and TAFAZZIN KD mice fed with control or rapamycin diet. n = 4. Data are represented as mean ± SD. *p < 0.05, **p < 0.01, ***p < 0.001 by one-way ANOVA.

Rapamycin inhibits mitochondrial oxidative stress and restores mitochondrial respiration in TAFAZZIN-deficient cells

Previous studies showed that TAFAZZIN deficiency induced mitochondrial superoxide anions production in cells [15,30]. We next determined whether rapamycin could also mitigate this defect in TAFAZZIN-deficient MEFs stained with MitoSOX, a mitochondrial targeted, fluorescent reactive oxygen species indicator. As shown in Figure 6A, TAFAZZIN deficiency dramatically increased the mitochondrial superoxide anion levels in MEFs, which was significantly attenuated by rapamycin (Figure 6A, the MitoSOX fluorescence intensity was quantified in Figure 6B). We next determined the effect of rapamycin on mitochondrial respiration in isolated mitochondria from heart samples and in primary MEFs by Seahorse analysis. Mitochondria were isolated from frozen heart samples, followed by analysis for mitochondrial respiration by Seahorse flux analyzer according to the protocol as previously described [31,32]. The results showed that TAFAZZIN deficiency significantly decreased mitochondrial oxygen consumption rate (OCR) in response to treatment with the succinate/rotenone and TMPD/ascorbate in isolated mitochondria from the heart of TAFAZZIN KD mice (Figure 6C, quantified in Figure 6D). Likewise, TAFAZZIN deficiency also significantly impaired mitochondrial respiration, including basal, maximal, and ATP-linked respiration, as well as the spare capacity in MEFs (Figure 6E, quantified in Figure 6F). Consistent with normalization in mitochondrial morphology, rapamycin treatment also significantly restored the mitochondrial respiration in both TAFAZZIN KD cardiac mitochondria and MEFs (Figure 6C–F). We next verified the findings in C2C12 cells, a skeletal myocyte cell line, with CRISPR-Cas9-mediated knockout of the Tafazzin gene. The knockout of TAFAZZIN in C2C12 cells was confirmed by Western blot analysis (Figure S1B). Consistent with findings from cardiomyocytes and MEFs, rapamycin partially restored the mitochondrial respiration, including basal, maximal, and ATP-linked respiration, as well as the spare capacity, all of which were significantly reduced in TAFAZZIN-deficient C2C12 cells (Figure S3).

Figure 6. Rapamycin attenuates mitochondrial superoxide anions production and restores mitochondrial respiration in TAFAZZIN-deficient cells. (A) Confocal imaging analysis of the mitochondrial superoxide anions in primary WT and TAFAZZIN KD MEFs. MEFs were cultured in the presence or absence of rapamycin (1 μM) or vehicle for 4 h, followed by staining with MitoTracker Green and MitoSOX Red. Scale bar: 20 μm. (B) Quantitative analysis of MitoSOX Red fluorescence intensity. All images were taken at the same settings, and the MitoSOX Red intensity was analyzed using ImageJ software. n > 30 cells per group. (C) Seahorse analysis of OCR in isolated mitochondria from the heart of WT and TAFAZZIN KD mice. The OCRs were measured from equal amount of mitochondria isolated from frozen heart samples in response to succinate/rotenone (Succ/Rot), antimycin A (AA), TMPD/ascorbate (TPMD/Asc), and sodium azide treatment. n = 6–7 per group. (D) Quantitative analysis of Succ/Rot- and TMPD/Asc (complex IV, CIV)-dependent respiration in isolated mitochondria from frozen heart samples shown in panel C. (E) Seahorse analysis of mitochondrial OCR in primary WT and TAFAZZIN KD MEFs in response to treatment with rapamycin (1 μM) or vehicle for 4 h. The OCR was normalized by protein concentration in each sample. (F) Quantitative analysis of OCR from basal, maximal, and ATP-linked respiration, as well as spare capacity in primary MEFs shown in panel E. Data are represented as mean ± SD. *p < 0.05, **p < 0.01, ***p < 0.001 by one-way ANOVA.

Rapamycin restores lysosomal function and mitophagy in TAFAZZIN-deficient cells

The findings that rapamycin treatment prevented TAFAZZIN deficiency induced abnormal glycosylation of LAMP1 prompted us to investigate whether TAFAZZIN deficiency also caused lysosomal dysfunction, and if so, whether rapamycin treatment would restore lysosomal function in the TAFAZZIN KD MEFs. Using confocal imaging analysis, we next determined a role of TAFAZZIN and rapamycin in regulating lysosomal morphology in MEFs isolated from TAFAZZIN KD and WT control mice. The MEFs were cultured in the presence or absence of rapamycin, and stained with LysoTracker Green and MitoTracker Red to label lysosomes and mitochondria with green and red fluorescence, respectively. Consistent with abnormal glycosylation of LAMP1, TAFAZZIN deficiency led to significant enlargement of lysosomes (Figure 7A, Lyso panel, highlighted by arrow heads, lysosomal size was quantified in Figure 7B and C). Treatment with rapamycin significantly normalized lysosomal morphology in TAFAZZIN KD MEFs (Figure 7A–C). We next determined the effect of rapamycin on mitophagy by analyzing the co-localization of mitochondria and lysosomes, an indirect measurement of mitophagy in isolated MEFs. The MEFs were first stained with LysoTracker Green and MitoTracker Red, and then treated with carbonyl cyanide m-chlorophenylhydrazone (CCCP), a potent mitochondrial uncoupler that stimulates mitophagy, in the presence or absence of rapamycin, followed by confocal imaging analysis of lysosomal morphology and co-localization with mitochondria. Treatment with CCCP failed to normalize lysosomal size in TAFAZZIN-deficient MEFs (Figure 7D, enlarged lysosomes were highlighted by arrow heads). Moreover, treatment with CCCP significantly induced mitophagy in WT MEFs, but not in TAFAZZIN KD MEFs, as shown by a lack of co-localization of MitoTracker Red with LysoTracker Green (Figure 7D, highlighted by arrows, and quantified in Figure 7E). In final support of a key role of MTORC1 hyperactivation in the pathogenesis of BTHS, rapamycin treatment not only normalized lysosomal size, but also restored mitophagy in TAFAZZIN KD MEFs in response to treatment with CCCP (Figure 7D and E). Together, these findings suggest that rapamycin treatment mitigates BTHS in TAFAZZIN KD mice in part by restoring mitophagy.

Figure 7. Rapamycin restores lysosomal morphology and mitophagy in TAFAZZIN KD MEFs. (A) Confocal imaging analysis of lysosomal morphology in MEFs. Primary MEFs from TAFAZZIN KD and WT mice were treated with rapamycin (1 μM) or vehicle for 4 h, followed by staining with MitoTracker Red and LysoTracker Green to label mitochondria and lysosomes, respectively. Scale bar: 10 μm. Arrow heads highlight the enlarged lysosomes. (B and C) Quantitative analysis of the average size of lysosomes (B) and the percentage of enlarged lysosomes (>1 μm²) (C) in WT and TAFAZZIN KD MEFs in response to rapamycin treatment. n > 15 cells per group. (D) Confocal imaging analysis of the mitolysosomes formation in MEFs in response to induction of mitophagy by CCCP (20 μM) for 1 h. For rapamycin treatment, cells were pre-treated with rapamycin (1 μM) for 3 h, followed by CCCP treatment. Scale bar: 20 μm. Arrow heads highlight the enlarged lysosomes, and arrows highlight the co-localization of MitoTracker Red with LysoTracker Green. (E) Quantitative analysis of the co-localization of MitoTracker Red with LysoTracker Green in MEFs in response to CCCP and rapamycin treatment. n = 50 cells per group. Data are represented as mean ± SD. *p < 0.05, **p < 0.01, ***p < 0.001 by two-way ANOVA. n.s, no significance.

Discussion

BTHS is a rare X-linked genetic disorder caused by mutations in the TAFAZZIN gene, resulting in CL depletion and mitochondrial dysfunction, which contribute to the development of dilated cardiomyopathy and early lethality in BTHS patients [1,16–19,33]. As the mitochondrial signature phospholipid, CL plays a critical role in mitophagy, from autophagosome initiation to lysosomal degradation, as demonstrated by previous studies [12,14,15]. Additionally, externalization of CL from the mitochondrial inner membrane to the outer membrane surface serves as an important recognition signal by autophagosomes that directs damaged mitochondria to mitophagy [12,34]. Externalized CL has been shown to directly interact with BECN1/Beclin 1, a central regulator of autophagy, on the mitochondrial outer membrane [35]. Moreover, externalized CL plays a critical role in the binding of PE conjugated LC3 (LC3-II), a critical step in autophagosome biogenesis and cargo recognition [12]. Consequently, defective mitophagy leads to an accumulation of damaged mitochondria in cells, which is implicated in the pathogenesis of a number of chronic conditions including cancer, cardiovascular diseases, fatty liver disease, and neurodegenerative diseases [9–11,36]. In support of the critical role of CL in mitophagy, we showed in this study that TAFAZZIN deficiency caused a massive accumulation of dysfunctional mitochondria in the heart of TAFAZZIN KD mice. Mitochondria go through periodic cycles of fusion and fission, which plays a key role in eliminating damaged mitochondria through mitophagy. CL also plays a critical role in supporting various proteins and enzymes involved in the mitochondrial fission and fusion process [8,33,34]. In support of this notion, we further showed in this study that TAFAZZIN deficiency led to mitochondrial disarray, a loss of cisterna structure, and fragmentation in the heart. As one of the most mitochondrion-enriched tissues, heart is highly sensitive to mitochondrial dysfunction which likely contributes to the onset of dilated cardiomyopathy in the BTHS patients.

The MTORC1 signaling pathway plays a powerful role as a negative regulator of autophagy by downregulating the ULK1 (unc-51 like autophagy activating kinase 1) complex [37]. In support of defective mitophagy associated with BTHS, we showed in this study that the MTORC1 signaling pathway was hyperactivated in the heart of TAFAZZIN KD mice. It has been reported that phosphatidic acid (PA), a key intermediate metabolite of CL, interacts with MTOR in a manner that is competitive with rapamycin, which is required for the stability of both MTORC1 and MTORC2 complexes [38,39]. In support of this notion, PA content was elevated in TAFAZZIN KD mice [40]. In addition, TAFAZZIN deficiency significantly increased the activity of hydroxymethylglutaryl-coenzyme A reductase [41], a rate-limiting enzyme of the mevalonate pathway in the biosynthesis of cholesterol. Consistently, TAFAZZIN deficiency significantly increased the free cholesterol level in the heart [42], which may contributes to the hyperactivation of MTORC1 in BTHS, since cholesterol is an essential activator of MTORC1 and promotes MTORC1 activation on lysosomal membrane [43]. Upon activation, MTORC1 is recruited to the lysosomal membrane by Rag GTPases in a Ragulator and v-ATPase-dependent manner [44]. Hyperactivation of MTORC1 has been shown to cause lysosomal dysfunction [45,46]. Furthermore, MTORC1 negatively regulates the fusion of the autophagosome with lysosomes through interacting and phosphorylation of UVRAG (UV radiation resistance-associated gene protein) [47]. Consistent with hyperactivation of MTORC1, we showed in this study that TAFAZZIN deficiency also caused a massive enlargement of lysosomes. Treatment with rapamycin not only normalized lysosomes size, but also restored mitophagy and mitochondrial respiration in the heart and in isolated MEFs from TAFAZZIN KD mice.

Accumulation of autophagic vacuoles is associated with onset of dilated cardiomyopathy in human patients with heart failure [48]. In further support of a causative role of defective mitophagy in the pathogenesis of BTHS, we showed in this study that TAFAZZIN deficiency caused massive accumulation of autophagic vacuoles in the heart of TAFAZZIN KD mice. Remarkably, most of the autophagic vacuoles were wrapped by mitochondria, suggesting a defect in mitophagic consumption step. In support of this notion, we further showed that TAFAZZIN deficiency not only led to accumulation of SQSTM1 and increased LC3-I:LC3-II ratio, but also significantly upregulated protein expression of PINK1 and PRKN. Both PINK1 and PRKN play a critical role in the initiation of mitophagy. Phosphorylation of ubiquitin by PINK1 activates the ubiquitin ligase PRKN on the outer mitochondrial membrane where they act to recruit autophagy receptors [49]. In line with these observations, we further demonstrated that treatment with rapamycin not only normalized protein expression of PINK1 and PRKN, but also prevented accumulation of autophagic vacuoles in the heart of TAFAZZIN KD mice.

LAMP1 is a major lysosomal membrane protein that plays an important role in maintaining the structural integrity and acidic pH of the lysosomal compartment. Together with LAMP2, they contribute to about 50% of the all proteins of the lysosomal membrane [24]. Disruption of both LAMPs causes Danon disease which is characterized by accumulation of autophagic vacuoles in heart and skeletal muscle [24]. The defect is highly reminiscent of what we observed in TAFAZZIN KD cardiomyocytes. Additionally, both LAMP1 and LAMP2 are heavily glycosylated on their luminal domains. A recent study showed that abnormal LAMP1 glycosylation was implicated in the pathogenesis of Niemann-Pick disease, type C (NPC), a fatal neurodegenerative disorder caused by autosomal recessive mutations in either NPC1 or NPC2 [29]. Attenuation of LAMP1 glycosylation is associated with significant improvement of NPC. NPC1 mutation causes cholesterol accumulation within lysosomes, leading to MTORC1 hyperactivation and defective mitophagy in NPC [43,50]. Accordingly, genetic or pharmacological inhibition of MTORC1 restored lysosomal function and ameliorated mitochondrial dysfunction in a neuronal model of NPC [50]. In line with hyperactivation of MTORC1, we showed in this study that TAFAZZIN deficiency significantly promoted abnormal glycosylation of LAMP1 in the heart of TAFAZZIN KD mice. In final support of defective mitophagy in the pathogenesis of BTHS, treatment with rapamycin not only prevented abnormal LAMP1 glycosylation concurrently with normalization of lysosome morphology, but also restored mitophagy in the heart of TAFAZZIN KD mice and in the isolated primary TAFAZZIN-deficient MEFs.

Despite of intensive research efforts in recent years, there are no direct and effective treatments for BTHS to date. A widely studied strategy is to use adeno-associated virus (AAV) gene therapy to replace mutant TAFAZZIN. AAV-mediated human TAFAZZIN gene replacement normalized cardiac and skeletal muscle function in the TAFAZZIN deficient rodent models [22,51]. While promising, discovering an effective pharmacological therapy target is important for the development of first-in-line treatment for BTHS. A previous study showed that decreasing cytosolic translation restored oxidative phosphorylation in a yeast model of BTHS without changing the CL profile, indicating that cytosolic translation as a potential therapeutic target for the treatment of BTHS [52]. Although MTORC1 activity is necessary for normal regulation of cardiomyocytes homeostasis and growth, chronic hyperactivation of MTORC1 signaling appears to accelerate the cardiac aging and the development of cardiomyopathy [53]. Genetic or pharmacological inhibition of MTORC1 mitigated cardiomyopathy in several human diseases, including obesity [54], LEOPARD syndrome [55] and LMNA deficiency induced diseases [28]. In addition, MTORC1 is an established drug target and rapamycin has been used in organ transplantation patients during the last few decades without serious side effects [56]. Finally, our current study demonstrated that inhibition of MTORC1 by rapamycin not only restored mitophagy, but also significantly attenuated dilated cardiomyopathy in TAFAZZIN KD mice. Taken together, our findings revealed for the first time that hyperactivation of MTORC1, which contributes to the defective mitophagy in BTHS, plays a critical role in the pathogenesis of BTHS. Since rapamycin is a FDA approved drug, our findings suggest that treatment of human BTHS patient with rapamycin or other MTORC1 inhibitors may significantly attenuate the pathogenesis of this lethal disease.

Materials and methods

Reagents

Rapamycin (37094), carbonyl cyanide 3-chlorophenylhydrazone (CCCP; C2759), carbonyl cyanide 4-(trifluoromethoxy)phenylhydrazone (FCCP; C2920), oligomycin A (75351), rotenone (R8875), antimycin A (A8674), N,N,N’,N’-tetramethyl-p-phenylenediamine (TMPD; T7394), ascorbic acid (1043003), sodium azide (S8032), succinate (S9512), phosphatase inhibitor cocktail 2 (P5726), and phosphatase inhibitor cocktail 3 (P0044) were purchased from MilliporeSigma. EDTA-free protease inhibitor cocktail (11873580001) was from Roche. MitoTracker™ Red CMXRos (M7512), MitoTracker™ Green FM (M7514), LysoTracker™ Green DND-26 (L7526), MitoSOX Red (M36008), Hoechst 33342 solution (62249), TRIzol (15596018), SuperScript™ IV first-strand synthesis system (18091200), and SYBR™ green PCR master mix (4309155) were from Thermo Fisher Scientific. Doxycycline hyclate (446061000) was from Acros Organics. Torin1 (inh-tor1) was from InvivoGen. LabDiet 5LG6 with 152 ppm encapsulated rapamycin (14 ppm active rapamycin) (9G7F) and the LabDiet 5LG6 with eudragit (9G7G, control diet) were ordered from Lab Supply.

Anti-phospho-RPS6KB/S6K (T389, 9205), RPS6KB/S6K (9202), phospho-MTOR (S2448, 2971), MTOR (2972), phospho-EIF4EBP1/4E-BP1 (T37/46, 2855), EIF4EBP1/4E-BP1 (9644), phospho-AKT (T308, 13038), AKT (9272), PIK3C3/VPS34 (4263), and MAP1LC3/LC3 (2775) antibodies were purchased from Cell Signaling Technology. Anti-LAMP1 (ab24170) and PRKN/PARKIN (ab15954) antibodies were from Abcam. Anti-PINK1 (23274-1-AP) antibody was from Proteintech. Anti-SQSTM1/p62 (P0067) antibody was from MilliporeSigma. Anti-GAPDH (sc-32233) antibody was from Santa Cruz Biotechnology. Anti-TAFAZZIN antibody was a gift from Dr. Steven M. Claypool at the Johns Hopkins University School of Medicine [2]. HRP conjugated goat anti-rabbit IgG Fc secondary antibody (31463) and HRP conjugated goat anti-mouse IgG (H + L) secondary antibody (31430) were from Thermo Fisher Scientific.

Mice care

B6.Cg-Gt(ROSA)26Sor^{tm37(H1/tetO-RNAi:Tafazzin)Arte}/ZkhuJ mouse, a tetracycline inducible shRNA-mediated Tafazzin knockdown mouse model of BTHS generated by Dr. Zaza Khuchua’s lab [20], was purchased from The Jackson Laboratory (Stock No: 014648). All animals were maintained in an environmentally controlled condition under a 12 h light/12 h dark cycle with free access to food and water. Experimental animals were generated by mating WT male C57BL/6 J mice with heterozygous transgenic female mice. WT and transgenic male mice were fed with drinking water containing doxycycline (2 mg/mL) starting at the weaning date (21-days-old) and continued to the end of the experiments to induce the knockdown of Tafazzin. For rapamycin administration, TAFAZZIN KD mice fed with water containing doxycycline were randomly divided into two groups at the age of 3-months. One group of TAFAZZIN KD mice was fed with low-fat rodent diet containing 152 ppm encapsulated rapamycin (14 ppm active rapamycin). The other group of TAFAZZIN KD mice, as well as the WT mice, was fed with control diet with eudragit for another 6 months. All experiments involving animals were approved by the Institutional Animal Care and Use Committee, and use protocols according to the NIH guidelines (NIH publication no. 86–23 [1985]).

Echocardiography

Echocardiographic analysis of mice heart was performed by using a VisualSonics Vevo 2100 Imaging System and an MS550D transducer (VisualSonics, Toronto, ON, Canada). M-mode short axis and B-mode long axis images of the LV were analyzed to measure the following parameters, interventricular septal end-diastole and end-systole (IVS-d and IVS-s), LV internal diameter end-diastole and end-systole (LVID-d and LVID-s), LV posterior wall end-diastole and end-systole (LVPW-d and LVPW-s), LV ejection fraction (LVEF), and LV fractional shortening (LVFS).

Histological analysis

Heart samples were isolated and fixed with 4% paraformaldehyde for 48 h. Fixed hearts were dehydrated and embedded in paraffin, and 5 μm sections were cut with a Leica RM-2162 (Leica, Bensheim, Germany). H&E staining was performed as previously described [9]. The size of cardiomyocytes was quantified using ImageJ software (NIH).

Transmission electron microscopy

The mitochondrial ultrastructure in mouse cardiomyocytes was evaluated by using electron microscopy. Heart samples from the same site of LV were fixed in 5% glutaraldehyde and 4% paraformaldehyde in 0.1 M sodium cacodylate buffer (pH 7.4) with 0.05% CaCl₂ for 24 h. After washing in 0.1 M sodium cacodylate buffer, tissues were fixed overnight in 0.1 M cacodylate buffer containing 1% OsO₄, dehydrated, and embedded in EMbed-812 resin (Electron Microscopy Sciences, 14121). The sections were stained with 2% uranyl acetate, followed by 0.4% lead citrate, and viewed with a JEOL JEM-2200FS 200kV electron microscope (Electron Microscopy Sciences Core at the UT Health San Antonio).

Isolation and culture of MEFs

MEFs were isolated at E12.5–13.5 from the time of copulation of female transgenic B6.Cg-Gt(ROSA)26Sor^{tm37(H1/tetO-RNAi:Tafazzin)Arte}/ZkhuJ mice bred with male C57BL/6 J mice. Each embryo was dissected out on a dish with Hank’s balanced salt solution (Gibco, 14175) separately. Head, limbs, and internal organs were removed. The head was collected and genomic DNA was isolated for genotyping using primers and PCR conditions previously described [20]. The remaining embryo was minced with a scalpel or razor blade and transferred to a 15 mL tube with 4 mL of collagenase solution (2 mg/mL collagenase IV [MilliporeSigma, C5138], 0.7 mg/mL DNase I [MilliporeSigma, D5025] and 10 mg/mL hyaluronidase [MilliporeSigma, H6254]), followed by incubation for 30–60 min in a 37°C until all of the tissue chunks were gone. The digested solution was filtered through 100 μm mesh and washed with culture medium before seeding in culture dishes. MEFs were cultured in Dulbecco’s Modified Eagle’s Medium (MilliporeSigma, D5796) supplemented with 10% fetal bovine serum (Atlanta Biologicals, S11550H), 1x nonessential amino acids (Gibco, 11140–050), 1x essential amino acids (Gibco, 11130–051), 100 U/mL penicillin-streptomycin (Corning, 30–002-Cl), and 1.25 mg/mL amphotericin B (Corning, 30–003-CF). MEFs were cultured in completed medium containing doxycycline (1 μg/mL) for 3 days to induce Tafazzin knockdown in transgenic MEFs, and used in the following experiments.

Generation of Tafazzin gene knockout C2C12 cells using CRISPR-Cas9 gene editing

Tafazzin gene was knocked out in C2C12 cells by transfecting CRISPR-Cas9 mouse plasmids from Santa Cruz Biotechnology (sc-426211 and sc-426211-HDR) using Viafect transfection reagent (Promega, E4982), and then purified through green fluorescent protein and red fluorescent protein fluorescence at the UT Health San Antonio Flow Cytometry Core. Cells were then kept under selection with 10 µg/mL puromycin (Santa Cruz Biotechnology, sc-205821). Vector control cells were created by transfecting C2C12 cells with an empty pBABE-puro vector backbone (Addgene, 1764; deposited by Jay Morgenstern and Hartmut Land), selected and kept in culture medium with 10 µg/mL puromycin.

Oxygen consumption rate (OCR) measurement

OCR was measured using a Seahorse XF24 analyzer (Seahorse Bioscience, North Billerica, MA, USA). In brief, MEFs or C2C12 cells were seeded in a XF24 microplate and cultured overnight. The TAFAZZIN KD MEFs or TAFAZZIN KO C2C12 cells were treated with rapamycin (1 μM) or vehicle for 4 h. The OCR was then measured in response to treatment with oligomycin (1 μM), FCCP (1.5 μM), and a mixture of rotenone (1 μM) and antimycin A (1 μM). Real-time OCR was recorded three times during each conditional cycle. After the measurements, the cells were lysed and the protein concentration was measured by using a BCA protein assay kit (Thermo Fisher Scientific, 23225). The OCR was normalized with the protein concentration in each well. OCR in isolated mitochondria from frozen heart samples was analyzed using the same method as previously described [31,32]. In brief, an equal amount of mitochondria (4 μg protein/well) isolated from frozen heart samples was loaded to XF24 microplate, and the OCR was then measured in response to the treatment with a mixture of succinate (5 mM) and rotenone (2 μM), antimycin A (4 μM), a mixture of TMPD (0.5 mM) and ascorbate (1 mM), and sodium azide (50 mM). Real-time OCR was recorded two times during each conditional cycle.

Quantitative real time-PCR (qRT-PCR) analysis

Total RNA was extracted from mice hearts using TRIzol reagent, and RNA quantity was determined at 260 nm (NanoDrop 1000 Spectrometer, Thermal Scientific). Single-stranded cDNA was synthesized using SuperScript IV reverse transcriptase system, and gene expression was determined by qRT-PCR, which was performed by using SYBR green PCR master mix and the 7300 Real-Time PCR System (Applied Biosystems, Waltham, MA, USA). The mRNA levels were determined using a standard curve method and normalized to the level of Gapdh which was used as an internal control. The primers used in PCR are shown below:

Tafazzin:Forward, 5ʹ-CCCTCCATGTGAAGTGGCCATTCC-3ʹ

Reverse, 5ʹ-TGGTGGTTGGAGACGGTGATAAGG-3ʹ

Myh7:Forward, 5ʹ-AGGGCGACCTCAACGAGAT-3ʹ

Reverse, 5ʹ-CAGCAGACTCTGGAGGCTCTT-3ʹ

Nppb:Forward, 5ʹ-GCTGCTTTGGGCACAAGATAG-3ʹ

Reverse, 5ʹ-GGAGCTCTTCCTACAACAACTT-3ʹ

Nppa:Forward, 5ʹ-GTGTACAGTGCGGTGTCCAA-3ʹ

Reverse, 5ʹ-ACCTCATCTTCTACCGGATC-3ʹ

Acta1:Forward, 5ʹ-GTTCGCGCTCTCTCTCCTCA-3ʹ

Reverse, 5ʹ-GCAACCACAGCACGATTGTC-3ʹ

Gapdh:Forward, 5ʹ-AATGGTGAAGGTCGGTGTG-3ʹ

Reverse, 5ʹ-GTGGAGTCATACTGGAACATGTAG-3ʹ

Confocal imaging analysis

Live MEFs were stained with MitoTracker Red CMXRos (50 nM) and LysoTracker Green DND-26 (100 nM) to visualize mitochondria and lysosomes, respectively. To induced mitophagy, MEFs were treated with CCCP (20 μM) for 1 h in complete medium. Mitochondria were stained with MitoTracker Red CMXRos before CCCP treatment. For rapamycin treatment, MEFs were pre-treated with rapamycin (1 μM) for 3 h in complete medium, followed by mitochondrial staining and CCCP treatment. For detecting the mitochondrial superoxide anions, MEFs were stained with MitoTracker Green FM (100 nM) and MitoSOX Red (2 μM) for 15 min at 37°C. Nuclei were labeled with Hoechst 33342 (500 nM). Images were captured using a Zeiss LSM 780 confocal microscope (Carl Zeiss AG, Oberkochen, Germany) with a 63x/1.4 NA oil immersion objective. Multiple fields of view were selected at random and imaged for further analysis using Zeiss ZEN Blue or ImageJ software.

Statistical analysis

Data were routinely represented as mean ± SD. Statistical significance was assessed by Student’s t-test, one-way ANOVA or two-way ANOVA using GraphPad Prism 7.0. Biological significance was considered statistically significant at p < 0.05. *p < 0.05; **p < 0.01; ***p < 0.001.

Acknowledgments

We would like to thank Dr. Steven M. Claypool at the Johns Hopkins University School of Medicine for providing us the anti-TAFAZZIN antibody and providing insightful suggestions on revising our manuscript, and Dr. John-Paul Andersen for proof-reading the manuscript.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Supplementary material

Supplemental data for this article can be accessed here.

Additional information

Funding

This work was supported in part by funding from the NIH [R01AG055747, Y.S.], American Diabetes Association [1-18 IBS-329, Y.S.], Barth Syndrome Foundation (Y.S.), and an endowment from Joe R. and Teresa Lozano Long Distinguished Chair in Metabolic Biology (Y.S.).