TBC1D15/RAB7-regulated mitochondria-lysosome interaction confers cardioprotection against acute myocardial infarction-induced cardiac injury

Rationale: Ischemic heart disease remains a primary threat to human health, while its precise etiopathogenesis is still unclear. TBC domain family member 15 (TBC1D15) is a RAB7 GTPase-activating protein participating in the regulation of mitochondrial dynamics. This study was designed to explore the role of TBC1D15 in acute myocardial infarction (MI)-induced cardiac injury and the possible mechanism(s) involved. Methods: Mitochondria-lysosome interaction was evaluated using transmission electron microscopy and live cell time-lapse imaging. Mitophagy flux was measured by fluorescence and western blotting. Adult mice were transfected with adenoviral TBC1D15 through intra-myocardium injection prior to a 3-day MI procedure. Cardiac morphology and function were evaluated at the levels of whole-heart, cardiomyocytes, intracellular organelles and cell signaling transduction. Results: Our results revealed downregulated level of TBC1D15, reduced systolic function, overt infarct area and myocardial interstitial fibrosis, elevated cardiomyocyte apoptosis and mitochondrial damage 3 days after MI. Overexpression of TBC1D15 restored cardiac systolic function, alleviated infarct area and myocardial interstitial fibrosis, reduced cardiomyocyte apoptosis and mitochondrial damage although TBC1D15 itself did not exert any myocardial effect in the absence of MI. Further examination revealed that 3-day MI-induced accumulation of damaged mitochondria was associated with blockade of mitochondrial clearance because of enlarged defective lysosomes and subsequent interrupted mitophagy flux, which were attenuated by TBC1D15 overexpression. Mechanistic studies showed that 3-day MI provoked abnormal mitochondria-lysosome contacts, leading to lysosomal enlargement and subsequently disabled lysosomal clearance of damaged mitochondria. TBC1D15 loosened the abnormal mitochondria-lysosome contacts through both the Fis1 binding and the RAB7 GAPase-activating domain of TBC1D15, as TBC1D15-dependent beneficial responses were reversed by interference with either of these two domains both in vitro and in vivo. Conclusions: Our findings indicated a pivotal role of TBC1D15 in acute MI-induced cardiac anomalies through Fis1/RAB7 regulated mitochondria-lysosome contacts and subsequent lysosome-dependent mitophagy flux activation, which may provide a new target in the clinical treatment of acute MI.


Introduction
Although much progress has been made in the clinical therapeutics of heart diseases to reduce cardiovascular mortality, ischemic heart disease, especially myocardial infarction (MI), still remains a major threat to human health with somewhat dismal reperfusion effectiveness [1,2]. The course of MI can be clinically staged into hyperacute, acute, subacute and chronic phases [3,4]. Adequate care and Ivyspring International Publisher treatment during acute phase are critical for the prognosis of patients following MI insult [5,6]. It is thus pertinent to further elucidate the precise pathogenesis of acute MI in an effort to engage effective clinical management regimen. Among plethora of mechanisms postulated for the onset and progression of acute MI, mitochondria have drawn the most attention [2,[7][8][9][10]. Mitochondria serve as the energy powerhouse for cells with ATP production [8,11]. Upon injury, mitochondria lose their membrane potential and release harmful byproducts of ATP production such as reactive oxygen species (ROS) and pro-apoptotic factors such as cytochrome C, which provoke detrimental effect to cellular function and cell fate [12][13][14]. Indeed, mitochondrial dysfunction has been well consolidated to function as a precursor to cell death [15]. Therefore, it is pertinent to maintain stringent mitochondrial quality to control ROS production and promote cell survival.
Mitochondrial quality control involves a dynamic process of fission, fusion, mitophagy, and biogenesis, among which, mitophagy serves as the main safeguard for mitochondrial quality control [16]. Mitophagy (selective mitochondrial autophagy) is an important and highly conserved dynamic cytosolic process for removal and recycling of long-lived or damaged mitochondria by lysosomes through PINK1/Parkin-or mitophagy receptors-dependent pathway [14,17,18]. Although findings from genetically engineered mice have depicted a protective role for mitophagy in various cardiac pathological settings [19][20][21][22], a controversy recently emerged for the precise role of autophagy in the setting of cardiac post-infarction [23][24][25]. In particular, autophagic activity is upregulated in ischemic hearts during early stage of infarction at various reported time, while autophagy seems to be compromised during the late stage when all autophagosomes and lysosomes were fused to form autolysosomes. In this context, autolysosomes may accumulate during late stage of acute MI challenge. However, the precise mechanism(s) involved in the accumulation of autolysosomes remains largely unknown.
Autolysosomes accumulation is resulted from blockade of autolysosomes degradation, which may be associated with lysosomal dysfunction [26,27]. In the last few years, functional interactions of intracellular organelles like mitochondria and lysosomes, has drawn the most increasing attention [28]. Recent evidence has depicted a critical role for mitochondria-lysosome contacts in the regulation of lysosomal dynamics and maintenance of cellular homeostasis [29]. Under normal condition, mitochondria and lysosomes joint together to form mitochondria-lysosome contacts dynamically in healthy cells. The TBC1D15/Fis1/RAB7 signaling cascade is deemed a critical process regulating the mitochondria-lysosome contacts. Active GTP-bound RAB7, a small GTPase from the Rab family, facilitates mitochondria-lysosome contacts and regulates lysosomal transport, fusion and maturation. TBC1D15, a member of the TBC (Tre2/Bub2/Cdc16)domain-containing protein family, is the GTPaseactivating protein (GAP) for RAB7. TBC1D15 regulates RAB7 from an active GTP-bound state into an inactive GDP-bound state upon GTP hydrolysis. TBC1D15 is recruited to mitochondria courtesy of binding with Fis1, an outer mitochondrial membrane protein. Mitochondrial TBC1D15 drives lysosomal RAB7 GTP hydrolysis at mitochondria-lysosome contact sites and then untethers contacts, providing a mechanism for mitochondria to modulate lysosomal dynamics. Although the mitochondria-lysosome contacts are proven to be distinct from damaged mitochondria targeting to lysosomes for degradation, expression of mutant TBC1D15 (Fis1 binding or RAB7-GAP domain, regulating mitochondrialysosome contacts) may induce abnormally large lysosomes in HeLa cells. This observation denoted the involvement of TBC1D15 in the regulation of lysosome-dependent mitochondrial autophagy under pathological conditions. However, the precise role of TBC1D15 and TBC1D15-regulated mitochondrialysosome contacts in the heart remains to be elucidated. To this end, this work was designed to determine (i) whether TBC1D15 imposes any effect on cardiac function and mitochondrial quality control following acute MI insult, if any; (ii) whether TBC1D15-induced changes in cardiac structure and function are mediated by alteration in TBC1D15/Fis1/RAB7-mediated mitochondrialysosome contacts and lysosome-dependent mitophagy.

Animals and materials
All experimental animal protocols were performed in accordance with the Guide for the Care and Use of Laboratory Animals (NIH Publication No. 85-23, revised 1996 by the US National Institutes of Health) and approved by the Institutional Animal Care and Use Committee of the Zhongshan Hospital Fudan University, Shanghai, China (approval number: 2019-309). Briefly, male C57BL/6J mice (6-8-week-old) were purchased from Shanghai Model Organism Center, Inc. All animals were housed following regular circadian cycle with free access to sterilized water and food.  [30].

Experimental model
Following 3 days of acclimation, animals were anesthetized with continuously 2% isoflurane inhalation and myocardial infarction (MI) was established as previously described [31]. Briefly, to exactly expose the heart, left thoracotomy was performed in the fifth intercostal space. The left anterior descending branch (LAD) of coronary artery was permanently ligated with a 6-0 silk suture. After ligation, the skin was sutured by 4-0 nylon sutures. Mice were carefully placed in a temperaturecontrolled cage at 37 °C until recovery. To explore the role of TBC1D15, mice were randomized into four groups: Sham-Ad-LacZ, MI-Ad-LacZ, Sham-Ad-TBC1D15 and MI-Ad-TBC1D15 groups. To discern specific domain(s) involved in TBC1D15-offered response in MI, a cohort of sham or MI challenged mice received mutant TBC1D15 (R400K or Δ231-240) adenovirus. Sham-operated mice were subjected to the same procedure except ligation of LAD coronary artery. To achieve cardiac-specific overexpression, the adenovirus-targeted TBC1D15 (wild-type, R400K or Δ231-240 mutant) was directly injected onto intra-myocardium at a dosage of 1.2×10 8 pfu/ul with a 1:10 dilution using a 29G needle at three distinct sites of the ischemic zone 3 days prior to LAD ligation per the manufacturer's manual [32]. Mice transfected with an adenoviral vector encoding β-galactosidase (Ad-LacZ, control viral vector) while receiving sham procedure were used as control group [30]. Mice were sacrificed 3 days after MI for further experimentation [33].

Isolation and culture of primary neonatal mouse cardiomyocytes (NMCMs)
The protocol to isolate primary NMCMs was described previously [34]. In brief, neonatal mouse hearts were quickly excised and minced into debris followed by enzymatic digestion with collagenase Ⅰ. After digesting for several times, the digested fragments were placed to sediment for several min, and the cells in supernatants were preplated for 90 min to remove fibroblasts and endothelial cells. Then the residual supernatants with abundant cardiomyocytes were replanted in collagen-coated dishes. NMCMs were cultured in complete DMEM with 4500 mg/L glucose, 4 mM L-glutamine, 110 mg/L sodium pyruvate, 1% (v/v) penicillin/ streptomycin and 10% (v/v) FBS, and were then incubated at 95% air, 5% CO2 and 37 °C. After incubated for 48 h, the medium was refreshed by complete medium. NMCMs were transfected with adenoviruses encoding LacZ, TBC1D15 (wild-type, R400K or Δ231-240 mutant) and mRFP-GFP-LC3 or mito-Keima at the indicated MOI according to the corresponding instructions.

Hypoxia protocol of NMCMs
Hypoxia was employed to simulate the MI procedure [35]. Briefly, the culture medium of NMCMs was switched to a glucose-free DMEM and cells were incubated in a chamber containing 95% N 2 and 5% CO 2 gas mixture (Columbus Instruments) for 9 h before further experimentation. Cells transfected with an adenoviral vector encoding LacZ but not encoding TBC1D15 (wild-type, R400K or Δ231-240 mutant) at the same MOI under normoxic condition were used as the control group. All in vitro studies were performed using 3 or 6 independent experiments.

Measurement of infarct size
After 3 days of MI, mice were anesthetized with 2% isoflurane and injected with 1.5% Evans blue from the aortic root. The hearts were rapidly excised and cut into five 1-mm slices on frozen ice. Then the slices were incubated in 1% 2,3,5-triphenyltetrazolium chloride (TTC) solution for 15 min at 37 ℃ to demarcate the viable and non-viable myocardium. The staining was stopped by ice-cold sterile saline and the slices were fixed in 10% formalin solution. Five consecutive slices from each sample were sequentially imaged using a Canon camera with macro lens. Each slice was weighed. Total left ventricular area, area at risk (AAR) and infarct area from individual slices were measured using the Image J software (National Institute of Health, Version 1.8). The weights of infarct area and AAR from the same slice were calculated as weight of the slice multiplied by the ratio of each area. The overall weights of total left ventricle area, infarct area and AAR from the same heart were summed by five consecutive slices. Ratios of overall AAR/left ventricle area and infarct area/AAR of the heart were calculated. A Pearson coefficient between weight of infarct area and AAR was analyzed and was shown with a ratio parameter (k) [37].

Determination of interstitial fibrosis
Mouse hearts were arrested in diastole by injection of 10% potassium chloride after anesthesia. Hearts were then rapidly excised and fully fixed in 4% paraformaldehyde at 4 °C. After fixation for at least 24 h, each heart was cut transversely into five consecutive 1-mm slices, embedded in paraffin and sliced into 4 μm thick sections for Masson Trichrome staining. Individual single section was obtained using a light microscope (Leica, Germany). Left ventricular and collagen-positive areas were measured using the Image J software. The ratio of overall interstitial fibrosis/left ventricle from the same heart was calculated by total collagen-positive area normalized to total left ventricle area [32].

Determination of apoptosis
For study at the cellular or subcellular levels, mice were injected with 1.5% Evans blue from aortic root. Non-ischemic region of the heart showed blue while ischemic region displayed white (necrosis, infarct region) and light red (alive, peri-infarct region) in color. Blue and white tissues were wiped away and the light red tissues about 1 mm away from the border of white tissues were obtained for further study. Apoptosis of myocardium and primary cardiomyocytes were determined using TUNEL staining. After washing with PBS for several times and permeating with 0.3% Triton X-100 for 20 min, tissues or cells were incubated with FITC conjugated dUTP solution for 1 h at 37 °C and were then stained with DAPI for an additional 20 min. Micrographs of TUNEL-positive and DAPI-stained nuclei were captured randomly at 400× magnification using a fluorescence microscope (Leica, Germany) and were counted using Image J. For in vivo study, 3 fields per heart, 5 hearts per group were analyzed. For in vitro study, 12 fields from 3 independent experiments per group were counted. At least 100 cells per field were counted. The percentage of apoptotic cells was calculated as the ratio of the number of TUNELpositive cells to that of total DAPI-stained cells [36].

Transmission electron microscopy (TEM)
Murine heart tissues were fixed in 2% glutaraldehyde for at least 24 h. The tissues were then immersed in 2% osmium tetroxide and 1% aqueous uranyl acetate for 1 h. After washed with a series of ethanol solutions (50%, 70%, 90% and 100%), tissues were transferred to propylene oxide, incubated in a 1:1 mixture of propylene oxide and EMbed 812 (Electron Microscopy Sciences) for 1 h and were then placed in a 70 °C oven to polymerize. Sections (75-80 nm) were cut using an ultramicrotome (Leica, Germany) equipped with a Diatome diamond knife and were collected on 200-mesh copper grids. After poststained in 5% uranyl acetate for 10 min and in Reynold's lead citrate for 5 min, sections were observed using a 40-120 kV transmission electron microscope (Hitachi H600 Electron Microscope, Hitachi, Japan) [36]. Ten to twenty microscopic fields from 5 hearts per group were analyzed with at least 1000 mitochondria included per group.

Determination of mitochondrial membrane potential (MMP)
Primary cardiomyocytes were cultured in disposable confocal dishes. After respective treatments, cells were rinsed with PBS and incubated with 5 μM JC-1 dye at 37 °C for 20 min. Fluorescent cells with a ProLong Live Antifade Reagent were visualized at 630× magnification using a confocal microscope (Leica, Germany). Micrographs of 10 fields from 3 independent experiments per group were counted. In JC-1 assay, the originally green fluorescent dye forms red fluorescent aggregates when they encountered energized mitochondria with higher membrane potential. The fluorescence intensity of red and green was measured using Image J and MMP was calculated as red fluorescence intensity/green fluorescence intensity [36].

Reactive oxygen species (ROS) detection
For cellular ROS detection, living cardiomyocytes were stained with DCFH-DA fluorescence probe. Cells were rinsed using 37 °C pre-warmed PBS and incubated with 10 μM DCFH-DA (1:1000, no serum) for 20 min at 37 °C. Then, cells were washed with 37 °C DMEM without serum for 3 times and were incubated with a ProLong Live Antifade Reagent. Micrographs of 10 fields from 3 independent experiments per group were obtained at 630× magnification using a Leica confocal microscope. DCFH-DA fluorescence (green) were measured using the Image J software.

Oxygen consumption rate (OCR)
The oxygen consumption rate (OCR) was analyzed using an XFe96 extracellular flux analyzer (Seahorese Bioscience). Firstly, Seahorse 96-well plates were attached by poly-d-lysine (PDL) (Sigma-Aldrich) for at least 2 h before NMCMs isolation. Then, isolated NMCMs were planted on PDL attached plates at a density of 5×10 5 cells/well and were analyzed to detect the OCR of NMCMs by sequentially adding the following metabolic regulators including oligomycin A (1 μM), FCCP (1 μM), antimycin A (1 μM) and rotenone (1 μM). OCR was measured using cells from 16 wells per group.

Mitochondria isolation
Mitochondria were isolated from NMCMs using Abcam's benchtop mitochondria isolation kit. Cells were frozen and thawed to weaken membranes, suspended in Reagent A at 5.0 mg/ml, incubated on ice for 10 min. Then NMCMs were homogenized with a glass Dounce homogenizer, and were then centrifuged at 1000× g for 10 min at 4 °C twice. The supernatant was centrifuged at 12000× g for 10 min at 4 °C, and the remaining pellet was resuspended in Reagent C supplemented with phosphatase/protease inhibitor and frozen at -80 °C. Mitochondrial protein concentration was determined by colorimetry using Enhanced BCA Protein Assay Kit (BCA).

Immunofluorescence
Murine heart sections and primary cardiomyocytes cultured in confocal dishes were fixed in 4% paraformaldehyde. After washing with PBS for several times and incubating with goat serum containing 0.3% Triton X-100 for 1 h, they were incubated overnight with primary antibody at 4 °C. Then the sections or cells were incubated with the fluorescence-conjugated secondary antibody for 1 h. Cells were further stained with DAPI containing anti-fade reagent. Micrographs (10 fields from 3 independent experiments per group) were obtained using a Leica confocal microscope and fluorescence colocalization (Pearson correlation and Mander coefficient) was analyzed using Image J via the Coloc 2 modality.

Co-immunoprecipitation and western blot
NMCMs were lysed with NP-40 buffer (50 mM Tris-HCl pH 7.5, 150 mM NaCl, NP-40) containing protease and phosphatase inhibitor cocktails. After centrifugation, the supernatants fractions were subjected to immunoprecipitation with monoclonal IgG and monoclonal FLAG antibodies with protein-A/G beads. The precipitants were analyzed by western blot.
For western blot, protein samples were extracted with a RIPA buffer containing phosphatase and protease inhibitor. Equal quantities of proteins (20 μg) were separated by 10-12% SDS-PAGE, and were then electrophoretic transferred to 0.22 μm PVDF membranes. After blocking with 5% non-fat milk in TBST buffer for 60 min, membranes were incubated with primary antibodies overnight at 4 ˚C. Blots were rinsed three times for 10 min in TBST and incubated with the HRP-conjugated secondary antibody for 1 h at room temperature. Densitometry was detected with the enhanced chemiluminescence (ECL) reagent. All the primary antibodies used for co-immunoprecipitation and western blotting are shown in Table  S1.

Real-time polymerase chain reaction (real-time PCR)
Real-time PCR was conducted as described previously [38]. Total RNA of cells or heart tissues was harvested using a TRIzol Reagent (Thermo Fisher Scientific, Waltham, USA) and a RNeasy Total RNA Isolation Kit (Qiagen, Hilden, Germany). Isolated RNA was reverse-transcribed using an iScript cDNA Synthesis Kit (Takara BIO, Otsu, Japan). Real-time PCR was performed using iQ SYBR Green Supermix (Bio-Rad Laboratories, Hercules, CA) in the CFX96TM Real-Time System (Bio-Rad Laboratories, Hercules, CA). All primers were purchased from Sangon Biotech Corporation (Shanghai, China) and were presented in Table S2. Levels of miR-1 were measured using the mirVana qRT-PCR miRNA Detection Kit (Thermo Fisher Scientific, Waltham, USA) in conjunction with real-time PCR. The mimics and antagomir of miR-1 were both purchased from Thermo Fisher Scientific (Waltham, USA). Mutated nucleotides in the TBC1D15-3'UTR were conducted according to a previous report [39]. Luciferase activities were detected using a dual luciferase reporter assay kit (Promega) with a luminometer after 1 g PGL3-target DNA and 0.1 g PRL-TK transfection with lipofectamine 3000 (Thermo Fisher Scientific, Waltham, USA) for 48 h [40].

Live cell time-lapse imaging
Live cells were imaged using a Leica confocal microscope. For mitochondria-lysosome contacts, cells were incubated with Mito-Tracker Red (mitochondria, red) and Lyso-Tracker Green DND-26 (lysosome, green) for 2 h at the concentrations of 100 nM and 50 nM in combination with a ProLong Live Antifade Reagent for the indicated duration according to the manufacturer's protocol. Images were dynamically recorded at 561-nm for Mito-Tracker Red and 488-nm for Lyso-Tracker Green DND-26. Images were analyzed using the Image J software [29].

Lysosomal acidification and size
Lysosomal acidification was measured in neonatal cardiomyocytes loaded with 1 mg/ml Lyso-Sensor Green DND-189 for 1 h at 37 °C [41]. The Lyso-Sensor dye is a probe that produces green fluorescence in acidic environments and is greener when exposed to more acidic environments. Cells were incubated with a ProLong Live Antifade Reagent to delay fluorescence auto-bleaching. The pictures were captured by a Leica confocal microscope (20-25 fields from 3 independent experiments per group). The fluorescence intensity of Lyso-Sensor Green DND-189 was calculated using the Image J software. The Image J software was used to evaluate the size and number of lysosomes in each cell. The average size of lysosomes was normalized to the number of lysosomes [42,43].

Statistical analysis
Data were analyzed with Graph-Pad Prism 8 (GraphPad Software, LLC, San Diego, CA, USA) and expressed as Mean ± SEM. Data distribution was examined using the Shapiro-Wilk normality test. Two groups were compared by Student's t test (two-tailed). Multiple groups were compared by one-way ANOVA followed with the Tukey post hoc test. For groups receiving different secondary treatments, two-way ANOVA together with Tukey test were performed to compare the difference. A value of P < 0.05 was considered statistically significant.

Fis1 binding and RAB7-GAP domains are involved in TBC1D15-dependent lysosomal acidification restoration and subsequent mitophagy regulation
Given the decreased autophagosome formation in hypoxic cardiomyocytes, it is plausible to speculate that autolysosomes accumulation could be due to the blockade of autolysosome degradation, which may be associated with endosomal or lysosomal dysfunction. We went on to examine endosomal and lysosomal function following long-term hypoxia in cardiomyocytes.

Lysosomal acidification restoration may be associated with TBC1D15/Fis1/RAB7dependent mitochondria-lysosome contacts untethering
Both Fis1 binding and RAB7-GAP domains are involved in mitochondria-lysosome contacts, critical for both mitochondrial and lysosomal dynamics [29]. To this end, we examined the mitochondria-lysosome contacts in cardiomyocytes following acute MI or long-term hypoxia using TEM and confocal live time-lapse imaging. Static results from TEM imaging revealed that the distance of mitochondria-lysosome contacts was decreased (8.23 ± 0.86 vs. 28.43 ± 2.65 from Sham, P < 0.05) while the length of mitochondria-lysosome contacts was increased (341.8 ± 17.1 vs. 151.3 ± 9.7 from Sham, P < 0.05) in peri-infarct myocardium 3 days after MI, indicating elevated tight mitochondria-lysosome contacts following acute MI. TBC1D15 (WT) increased distance (18.13 ± 0.72 vs. 8.23 ± 0.86 from MI, P < 0.05) and decreased length (239.8 ± 11.1 vs. 341.8 ± 17.1 from MI, P < 0.05) of mitochondria-lysosome contacts while TBC1D15 (R400K and Δ231-240 mutants) failed to attenuate the elevation of mitochondria-lysosome contacts induced by 3-day MI ( Figure 6A-C). Dynamic observation from confocal live cell time-lapse imaging revealed that mitochondria (Mito-Tracker, red) tended to be in contact with extremely enlarged lysosomes (Lyso-Tracker, green) for an excessively prolonged duration under 9-h hypoxic condition (264.3 ± 11.8 vs. 38.7 ± 3.7 from Normoxia, P < 0.05), and even after belated contacts release, these enlarged lysosomes were unlikely to timely recover to prehypoxia levels, the effect of which was abrogated by TBC1D15 (WT) (110.0 ± 4.7 vs. 264.3 ± 11.8 from Hypoxia, P < 0.05) while not by TBC1D15 (R400K and Δ231-240 mutants) ( Figure 6D-E and Video S1-8). In light of previous results, these findings depicted an essential role for mitochondria-lysosome contacts in TBC1D15-mediated lysosomal regulation and mitophagy flux activation.

Discussion
The salient findings from our study depicted that TBC1D15 preserved cardiac function, decreased infarct area, myocardial fibrosis and cardiomyocyte apoptosis following acute MI. TBC1D15 levels were dramatically downregulated due to elevation of miR-1 in response to acute MI, leading to RAB7mediated accumulation of abnormal mitochondrialysosome contacts and subsequently contacts-induced enlargement of lysosomes. The enlarged lysosomes could fuse with autophagosomes to form autolysosomes although cargos inside failed to be degraded within the autolysosomes due to defective acidification. As a result, dysfunctional autolysosomes and ROS accumulation ultimately resulted in cardiac dysfunction. However, TBC1D15 overexpression could translocate onto mitochondria through binding with mitochondrial Fis1 and promote untethering of mitochondria-lysosome contacts via activation of RAB7 GTP hydrolysis, thus preventing enlargement of lysosomes and restoring lysosomal degradation. In consequence, TBC1D15 is capable of promoting clearance of damaged mitochondria for quality control, resulting in preserved myocardial mitochondrial function, mitochondrial integrity and cardiac contractile function in the face of acute MI challenge, as summarized in our scheme (Figure 8). To the best of our knowledge, this is the first study to investigate the role of TBC1D15 in the heart and describe the relationship between mitochondria-lysosome contacts and mitophagy flux in long-term ischemic/hypoxic settings.
The key finding in our study was that TBC1D15 functioned as the critical bridge molecule between mitochondria-lysosome contacts and mitophagy. As a member of TBC-domain-containing protein family, TBC1D15 is known to participate in multiple cellular processes [29,[49][50][51][52][53][54][55][56][57] and neurological disorders [58]. Collectively, TBC1D15 is involved in mitochondrial fission, autophagosome modification, mitochondrialysosome contacts and transport of cellular substance via endosomal system. However, little is known about the role of TBC1D15 in the heart. In our hands, TBC1D15 level was reduced in ischemic heart while overexpression of TBC1D15 preserved cardiac systolic function and improved myocardial morphology following acute MI, similar with its role in neurological disorders [58]. Ample evidence has revealed the deleterious role of mitochondrial damage in the onset and progression of acute MI [7], which correlates well with our results. Furthermore, our study also proved that TBC1D15 exerted protective effects on acute MI-induced mitochondrial abnormalities and ROS accumulation, similar with reported actions of TBC1D15 on mitochondrial morphology [29,54]. Mitochondria in adult cardiomyocytes can be categorized into 3 distinct populations, peri-nuclear, interfibrillar, and subsarcolemmal mitochondria, with geographical distinction in the regulation of mitochondrial metabolism [59]. Our data showed no difference in the relative proportion of these 3 mitochondrial populations in all experimental groups. In addition, our results also revealed lack of effect for TBC1D15 on total number of mitochondria and PGC1α levels. These findings have indicated a beneficial role of TBC1D15 on mitochondria possibly through regulation of quality control instead of mitochondrial biogenesis and distribution. Seminal findings from Ong and colleagues depicted the importance of inhibiting excessive mitochondrial fragmentation in conferring cardioprotection against I/R injury [60][61][62]. Either excessive large or small content of mitochondria can be detrimental for cell homeostasis. Although our findings have shed some lights towards a role for TBC1D15 in mitochondrial morphology, further study is still warranted to fully elucidate the role of TBC1D15 in the governance of mitochondrial homeostasis in particular mitochondrial dynamics.  Under ischemia or hypoxic stress, TBC1D15 is downregulated as a result of TBC1D15 mRNA degradation driven by miR-1, leading to compromised recruitment of TBC1D15 to mitochondria via binding with Fis1 and delayed release of lysosomes from mitochondria by RAB7 GTP hydrolysis at the mitochondria-lysosome contact sites. The abnormal mitochondria-lysosome contacts may prolong excessive metabolic exchanges between the two organelles, and elevate lysosomal osmotic pressure, resulting in enlargement, poor acidification and dysfunction of lysosomes, ultimately blockade of lysosome-dependent degradation of damaged mitochondria in ischemia or long-term hypoxia. Accumulation of damaged mitochondria results in loss of mitochondrial membrane potential, excessive ROS production, cardiomyocyte apoptosis, and cardiac dysfunction. Overexpression of TBC1D15 loosens tethering of mitochondria-lysosome contacts, restores lysosome size and acidification, and promotes the ability of lysosome-dependent digestion of depolarized mitochondria.
Mitophagy is a highly conserved lysosomedependent process through which damaged mitochondria are degraded and recycled for mitochondrial quality control. Dysregulation of mitophagy participates in the pathophysiology of cardiovascular disease [26]. Autophagy has been examined in MI-induced myocardial abnormalities but whether autophagy activity is upregulated or downregulated during the different phases of MI has remained a question of debate [63]. Here, our findings revealed that following coronary artery ligation, level of LC3II was initially increased, but then returned to a sub-normal level. On the other hand, p62 level declined 2 h following infarction, and then raised beyond normal level at 72 h post-infarction. These findings indicate activation of autophagy during early acute phase of MI followed by dampened autophagy in late acute period. Further assessment of autophagy or mitophagy flux using mRFP-GFP-LC3, mito-Keima and bafilomycin A1 treatment revealed that long-term hypoxia induced accumulation of autolysosomes instead of autophagosomes, indicating the blockage of autophagy flux, similar with the finding of impaired autophagosomes clearance in myocardial I/R injury [48]. Accumulation of autolysosomes depends on decreased efficiency of degradation and/or increased rate of uptake (fusion). In fact, the decrease in autophagosomes or LC3II level may indirectly signify a decreased rate of uptake while lysosomal dysfunction or p62 level may indirectly reveal the decreased efficiency of digestion. However, the exact rate of uptake or efficiency of digestion needs more comprehensive dynamic investigation. Both immunofluorescence of COXIV or LC3 with LAMP1 and Lyso-Sensor detection confirmed the existence of lysosomes enlargement with disabled acidification following long-term hypoxia. However, the activity of autophagy may be a little different from previous reports as different phases investigated, similar with the tendency of autophagy activity in transverse aortic constriction (TAC) mouse model [30]. Overexpression of TBC1D15 effectively dampened number of enlarged lysosomes, restored lysosomal acidification and activated mitophagy flux, coinciding with TBC1D15-regulated lysosomal morphology [55] and lysosomal activation in autophagy flux regulation [41].
Recent findings have proved the functional interaction between mitochondria and lysosomes in a neuropathological context [64]. To date, different pathways of interaction between mitochondria and lysosomes have been described including mitophagy, transfer of mitochondria-derived vesicles (MDVs) and mitochondria-derived compartments (MDCs), and direct physical membrane contact between mitochondria and lysosomes [28]. A more recent study denoted that TBC1D15 untethers mitochondrialysosome contacts through binding with Fis1 and activating RAB7 GTP hydrolysis, and subsequently regulates mitochondrial and lysosomal morphology [65]. Here, we demonstrated that loss of TBC1D15 (as in the case of acute MI)-induced abnormal mitochondria-lysosome contacts were observed in both peri-infarct myocardium and long-term hypoxic cardiomyocytes which was ameliorated by TBC1D15 re-expression. Furthermore, either TBC1D15 mutant lacking Fis1 binding or RAB7 GAPase activation failed to modulate lysosomal acidification, mitophagy flux activation and cardioprotective effects mediated by TBC1D15, indicating an essential role of mitochondria-lysosome contacts in acute MI-induced lysosomal dysregulation, mitophagy flux blockage and cardiac abnormalities. In this way, this work also provides the possibility that TBC1D15 promotes the shift between different pathways of mitochondria and lysosome interaction (from mitochondria-lysosome contacts to mitophagy). Mitochondria-lysosome contacts may function as platforms for metabolic exchanges between the two organelles, as interorganelle communication is usually mediated by membrane contact sites [66,67]. Dysregulation of these metabolic exchange processes may lead to excessive accumulation of substance, enlargement of lysosomes and subsequent lysosomal dysfunction. However, more work would be taken to explore the specific mechanism(s) involved in RAB7-mediated mitochondria-lysosome contacts and lysosomal enlargement.
In summary, TBC1D15 exerts protective effects on the function and morphology of infarct heart through untethering mitochondria-lysosome contacts via Fis1/RAB7 pathway and subsequently keeping good mitochondrial quality control by restoring lysosomal acidification. This work may provide a new target in the clinical treatment of acute MI. However, more in-depth scrutiny should be engaged towards understanding the role of TBC1D15 in other pathological conditions especially for cardiovascular diseases.