H9c2 cells, which are rat embryonic ventricular cardiomyocytes, were obtained from American Type Culture Collection (ATCC, VA, USA). The H9c2 cells were cultured in Dulbecco’s modified Eagle’s medium (Gibco, MA, USA) containing 10% fetal bovine serum [Biological Industries (BI), CT, USA] and supplemented with 1% penicillin/streptomycin/amphotericin B and 2 mM L-glutamine (BI). The cells were maintained in a humidified incubator at 37 °C in 5% CO₂ and 95% air.

H9c2 cells were pretreated for 4 h with or without 100 nM 25-hydroxyvitamin D3 (Vit D) (Cayman, UM, USA) or 10 nM MitoTEMPO (Santa Cruz, TX, USA). In addition, to evaluate the effect of mitochondrial fission or autophagy on mitophagy, H9c2 cells were pretreated for 4 h with or without the mitochondrial fission inhibitor, Mdivi-1 (10 µM) or the autophagy inhibitor, bafilomycin A1 (10 µM) (Cayman). Hypoxia and reoxygenation (H/R) was carried out based on a previously described method (Yao et al., 2019). In brief, the hypoxic cell culture medium lacked serum. The H9c2 cells were then incubated for 6 h at 37 °C in an anaerobic chamber under hypoxic conditions (1% O₂). The cells were then transferred to a conventional incubator for 12 h. The corresponding control cells were incubated under normoxic conditions for the same duration.

The cytotoxic effects of H/R and Vit D were assessed by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) (Bionovas, ON, CA). After various treatments, MTT solution was added at a final concentration of 0.5 mg/ml and incubated for 4 h in 5% CO₂ at 37 °C. The MTT-containing media were then removed, and the formazan crystals were dissolved by adding dimethyl sulfoxide (DMSO, 150 μL/well), followed by incubation for 15 min with mild shaking at room temperature (RT). The optical density was measured spectrophotometrically at 550 nm using a microplate reader (Biotek, VT, USA). The cell viability was expressed relative to that of the control.

A TUNEL apoptosis assay was used to detect DNA fragmentation using an in situ cell death detection kit (Roche, CA, USA) according to the manufacturer’s instructions. Briefly, H9c2 cells were fixed in 4% paraformaldehyde for 20 min after H/R stimulation for the indicated time points. The samples were then incubated with the TUNEL reagent in a dark, humidified chamber at 37 °C for 1 h. As a negative control, cells and tissues were treated only with the labeling solution. Nuclear counterstaining with 4′,6-diamidino-2-phenylindole (DAPI) (Southern Biotech, AL, USA) was performed, and the stained cells were examined using a fluorescence microscope (Leica, Wetzlar, Germany). The number of TUNEL-positive nuclei was counted under a high-power field in six different non-overlapping fields from each slide.

Apoptotic and necrotic cells were quantified by annexin V-FITC binding and propidium iodide (PI) uptake (BioLegend, CA, USA) according to the provided protocols. Briefly, after the cells were treated as indicated, the cells were harvested, resuspended in 100 μL binding buffer containing 2.5 μL FITC-annexin V and 5 μL PI solution (100 μg/ml), and incubated for 15 min in the dark at 4 °C. The cellular fluorescence was measured using a FACSCalibur flow cytometer (BD, NJ, USA). The cells that were considered viable were negative for both dyes, while the cells that were in the early phase of apoptosis were annexin V-positive and PI-negative, the cells that were in late phase of apoptosis were annexin V/PI-positive, and the cells that were in necrosis annexin V-negative and PI-positive.

Western blot was performed as previously described (Pu et al., 2017). Cardiac tissues and cells were homogenized in RIPA lysis buffer [50 mM Tris, pH 7.4, 150 mM NaCl, 1% NP-40, 0.5% sodium deoxycholate, 0.1% sodium dodecyl sulfate (SDS)]. Samples with equal amounts of protein (20 μg) were electrophoresed in an SDS-polyacrylamide gel and transferred to polyvinylidene fluoride (PVDF) membranes (Millipore, MA, USA). These membranes were probed overnight at 4 °C with the following primary antibodies: caspase 3, cytochrome c, p-Drp1, Mff, and LC3B, which were purchased from Cell Signaling (MA, USA), HIF-1α and Bax, which were purchased from GeneTex (Hsinchu city, Taiwan), Bcl-2, which was purchase from BD, and BNIP3, which was purchased from Aviva Systems Biology (CA, USA). Then, the membranes were incubated with a horseradish peroxidase-conjugated goat anti-mouse or anti-rabbit IgG secondary antibody (Jackson, PA, USA). The bound antibodies were detected using enhanced chemiluminescence (ECL) (Merck, NJ, USA). The intensity of the bands was quantified using ImageJ software (NIH, MD, USA). β-actin (Abcam, MA, USA) was used as the internal standard.

For co-immunoprecipitation, cells were collected and lyzed with lysis buffer. The supernatant fractions were collected and incubated with 1 μg of the appropriate antibody and precipitated overnight with protein A/G Sepharose beads (G-Bioscience, MO, USA) at 4°C. The beads were washed 3 times with wash buffer by centrifugation at 2,500 g at 4 °C. The precipitated proteins were collected by centrifugation at 2,500 g for 5 min. The immunoprecipitated proteins were separated by SDS-PAGE and subjected to Western blot as described above. The primary antibodies were as follows: Drp1 (Cell Signaling) and LC3B antibodies. The precipitation purity was also evaluated with Mff and BNIP3 antibodies.

The levels of mitochondrial ROS were detected using the mitochondrial superoxide indicator MitoSox Red (Invitrogen, MA, USA). H9c2 cells were treated with 1 μM MitoSox Red for 15 minat 37 °C. Fluorescent images were captured using a fluorescence microscope. TO-PRO-3 (100 nM, Thermo, MA, USA), a dead cell indicator, was added before MitoSox Red analysis by an LSRFortessa flow cytometer (BD). 2′,7′-Dichlorodihydrofluorescein diacetate (DCFH-DA) (Thermo) and dihydroethidium (DHE) (Invitrogen) were used to detect the levels of intracellular oxidative free radicals and superoxide anions, respectively. Cardiomyocytes on coverslips were treated with or without H/R and then incubated with serum-free medium containing DCFH-DA (10 μM) in the dark at 37 °C for 30 min. After incubation, the conversion of DCFH-DA to the fluorescent product DCF was detected by fluorescence microscopy, and 5 μL PI solution (100 μg/ml) was added before DCFH-DA analysis by flow cytometry. In addition, cardiomyocytes were incubated in serum-free medium containing DHE (5 μM) for 15 min at 37 °C. Superoxide anions oxidize DHE, yielding ethidium, which emits red fluorescence at 535 nm, and images were captured by fluorescence microscopy. DiOC₆(3) (90 nM, Thermo), a lipophilic dye that is selective for the mitochondria of live cells, was added before DHE analysis by flow cytometry.

5,5′,6,6′-Tetrachloro-1,1′,3,3′-tetraethylbenzimidazolyl-carbocyanine iodide (JC-1, BD) was used to analyze changes in mitochondrial transmembrane potential. After the various treatments, the cells were incubated with JC-1 (2 μM) at 37 °C for 30 min in the dark. JC-1 monomers emit green fluorescence and indicate the dissipation of the ΔψM, whereas JC-1 aggregates emit red and indicate an intact ΔψM. Cell fluorescence was monitored using an LSM 510 confocal microscope (Zeiss, Oberkochen, Germany) and a FACSCalibur flow cytometer.

The levels of cellular ATP were evaluated using an ATP assay kit (Molecular Probe, OR, USA). The cells were lyzed and then centrifuged (12,000 g) for 5 min at 4 °C. Subsequently, the supernatants were collected, and the ATP levels were analyzed. The ATP working reagent was added to the 96-well plate for 5 min, and 20 μL of each sample was also added to each well. The ATP levels were measured using a microplate luminometer (Berthold, Bad Wildbad, Germany) and calculated according to standard ATP curves.

Cells were briefly washed with PBS and incubated with 1 μg/ml acridine orange hydrochloride solution (Invitrogen) in PBS for 20 min at RT. Then, the cells were analyzed using confocal microscopy and flow cytometry. The acidic autophagic vacuoles emitted bright red fluorescence, while green fluorescence was observed in the cytoplasm and nucleus.

After the indicated treatments, H9c2 cells on sterilized coverslips were washed with PBS, fixed with 4% paraformaldehyde for 15 min at RT, and then permeabilized with 0.1% Triton X-100 for 10 min at RT. After washing with PBS, the cells were blocked with 3% bovine serum albumin (BSA) (Novagen, Darmstadt, Germany) in PBS for 1 h at RT. The cells were incubated with primary anti-Drp1 (Cell Signaling) or anti-LC3B (Cell Signaling) (1:250 dilution in PBS containing 1% BSA) antibodies overnight at 4 °C. After washing three times with PBS, the cells were incubated with anti-rabbit Alexa Fluor® 488 (Invitrogen) (1:250 dilution in PBS containing 1% BSA) for 1 h at RT. After washing three times with PBS, the cells were incubated with COX IV (Thermo) (1:250 dilution in PBS containing 1% BSA) overnight at 4°C. After washing three times with PBS, the cells were incubated with anti-mouse Alexa Fluor® 647 (Invitrogen) antibodies for 1 h at RT. The cells were then counterstained with 1 μg/ml DAPI for 3 min. Analysis and photomicrography were performed using an LSM 510 inverted confocal microscope.

A mitochondria isolation kit (Thermo) was used to prepare mitochondrial and cytoplasmic fractions from whole H9c2 cell lysates. The experimental procedures were performed according to the manufacturer’s instructions. The mitochondrial and cytoplasmic fractions were stored at −80 °C for further studies.

Mitochondrial morphology was observed using MitoTracker staining (Invitrogen). After the different treatments, cardiomyocytes were incubated with 400 nM MitoTracker for 1 h at 37 °C in the dark. Visualization was performed using a total internal reflection fluorescence microscope (TIRF, Zeiss) along with DIC optics and epifluorescence illumination, and the length of mitochondria was measured with ImageJ software. The length of more than 50 mitochondria per cell was measured, and measurements were collected from 20 cells. Three replicates were performed for each biological sample.

Adult male C57BL/6 mice (∼25 g) were obtained from National Taiwan University and housed at 25 °C ± 5 °C with 12-h light/dark cycles. All the animal experiments were performed in accordance with the National Institutes of Health guidelines for the use of laboratory animals and approved by the Taiwan University Animal Ethics Committee. The mice were randomly divided into the following groups: sham surgery, I/R, I/R + Vit D, and Vit D. One week before the surgery, Vit D was dissolved in PBS and intraperitoneally administered 3 times at 30 ng/mouse every 2 days. The sham group received PBS injections on the same schedule. The pharmacological dose was selected based on previous reports (Bodyak et al., 2007; Bae et al., 2011). In addition, bafilomycin (2.5 mg/kg) was intraperitoneally administered 1 day before I/R operation to inhibit the occurrence of autophagy. Myocardial I/R injury was performed according to the previously described method (Yang et al., 2017). Briefly, a mouse was anesthetized with 2% isoflurane, and the heart was manually exposed through a small incision without the need for intubation. Ischemia was induced by ligating the left anterior descending coronary artery (LAD) using an 8-0 nylon suture with a section of PE-10 tubing placed over the LAD at a position 1 mm from the tip of the normally positioned left atrium. After occlusion for 30 min, reperfusion was initiated by releasing the ligature and removing the PE-10 tubing. After reperfusion for 3 h, the heart was removed and immediately placed in ice-cold PBS. Hearts exhibiting infarcts involving the anterior and apical regions were rapidly frozen and stored in liquid nitrogen (N₂) for Western blot analysis. In some cases, an entire cross-section was fixed with 4% paraformaldehyde solution for haematoxylin and eosin staining and immunohistochemistry.

Cells and hearts from all the groups were imaged by transmission electron microscopy (TEM) to observe mitochondrial ultrastructure and mitophagy. The samples were immediately fixed with 2% glutaraldehyde and 2% paraformaldehyde in 0.1 M PBS at 4 °C overnight. After washing with PBS, the samples were postfixed with 2% osmium tetroxide. The fixed samples were dehydrated in a graded alcohol series and embedded in epoxy resin. Ultrathin sections were double-stained with uranyl acetate and lead citrate and then examined with a Hitachi H700 electron microscope (Hitachi, Tokyo, JP).

The data are presented as the mean ± standard error of the mean (SEM) from at least three independent experiments. One-way analysis of variance (ANOVA) was performed between groups, followed by Dunnett’s post hoc test. p < 0.05 indicated statistical significance.

To evaluate whether Vit D pretreatment improves myocardial cell damage under I/R conditions, H9c2 cells were pretreated with Vit D or PBS and then subjected to H/R in vitro to simulate the physiological stress experienced by cardiomyocytes during ischemic infarct. Various concentrations of Vit D were used to treat H9c2 cells for 24 h, and the cell viability, which was evaluated by MTT assay, was not reduced (data not shown). Then, H9c2 cells were incubated with or without 100 nM Vit D for 4 h, exposed to hypoxic conditions for 6 h and treated with reoxygenation for 12 h under normoxic conditions (H/R). The MTT assay showed that H/R treatment significantly decreased H9c2 cell viability, while Vit D pretreatment reduced this effect (Figure 1A). Apoptotic cells were detected using the TUNEL assay and annexin V-FITC/PI staining assay. As shown by TUNEL analysis, H/R treatment significantly increased apoptosis, whereas Vit D pretreatment significantly decreased apoptosis (Figures 1B,C). H/R increased the annexin V-positive apoptotic cell populations (annexin V positive/PI positive and annexin V positive/PI negative), whereas Vit D pretreatment significantly decreased these cell populations (Figures 1D,E). To further confirm that the effects of Vit D on apoptosis are due to its antioxidant activity, MitoTEMPO, a mitochondrial ROS scavenger, was used, and its effect on H/R-induced apoptosis was evaluated. Similar to Vit D, the flow cytometry results indicated that MitoTEMPO also reduced apoptosis in H/R-treated cardiomyocytes (Figures 1D,E). Cleaved caspase three is considered an indicator of apoptosis. Compared to the control, H/R treatment significantly increased the cleaved caspase three levels, while Vit D decreased the levels of cleaved caspase 3 (Figures 1F,G). The expression balance of the BCL-2 families, including pro-apoptotic proteins (such as Bax) and anti-apoptotic proteins (such as Bcl-2), determines cell fate (Siddiqui et al., 2015). We detected the levels of Bcl-2 and Bax expression using Western blot (Figures 1F,G). H/R treatment significantly increased the level of the Bax/Bcl-2 ratio, while Vit D pretreatment decreased it.

To investigate whether Vit D can affect mitochondrial ROS production in cardiomyocytes under H/R conditions, MitoSox Red was used to detect mitochondrial ROS. The results showed that H/R treatment significantly increased the production of mitochondrial ROS, while Vit D pretreatment decreased this effect (Figure 2A). Similar to pretreatment with Vit D, treatment with MitoTEMPO, a mitochondria-specific ROS scavenger, significantly decreased the mitochondrial ROS production in viable cells (MitoSox Red-positive and TO-PRO-3-negative) after H/R injury (Figures 2B,C). In addition, the cytoplasmic superoxide anion and H₂O₂ levels were detected using fluorescent protein-based redox probes, namely, DCFH-DA and DHE, respectively. H/R treatment significantly increased the cellular production of cytoplasmic superoxide anions and H₂O₂, as determined by fluorescence microscopy, while Vit D pretreatment reversed these effects (Figures 2D,G). In addition, H/R treatment increased the production of cytoplasmic superoxide anions in viable cells (PI-negative and DCFH-DA-positive), as shown by flow cytometry, while Vit D significantly decreased it (Figures 2E,F). Interestingly, the Vit D-treated group had lower DCFH-DA levels than the control group, and the H/R + Vit D group also had lower DCFH-DA levels than the control group. Additionally, the DHE overlay showed that H/R increased cytoplasmic H₂O₂, while Vit D decreased cytoplasmic H₂O₂ (Figure 2H). H/R increased the cytoplasmic H₂O₂ in viable cells (DHE-positive and DiOC₆(3)-positive), as shown by flow cytometry, whereas Vit D decreased it (Figure 2I). These results indicate that Vit D is an effective antioxidant in cardiomyocytes.

The ΔΨm and ATP levels serve as critical parameters for cell apoptosis and mitochondrial function. In most cardiomyocytes, H/R induced marked changes in terms of a shift in fluorescence emission from red to green, as shown by the JC-1 assay; this shift indicated ΔΨm dissipation. The emission of green fluorescence by cardiomyocytes pretreated with Vit D was reduced, and the distribution pattern was similar to that of the control group. In addition, the results of JC-1 analysis by flow cytometry revealed that H/R treatment increased the populations of cells with low ΔΨm (JC-1 green positive and red negative), whereas Vit D reversed this effect (Figures 3A–C). The ATP content was decreased after H/R injury, while Vit D treatment protected cardiomyocytes from this H/R-induced effect (Figure 3D). The ATP concentration of the H/R + Vit D group was higher than that of the control group. The dissipation of ΔΨm can lead to mitochondria-dependent apoptosis. In addition, H/R induced the expression of cytochrome c, a marker of mitochondrial apoptosis, whereas this upregulation of cytochrome c was reduced by Vit D pretreatment (Figures 3E,F). These results indicate that the protective effect of exogenous Vit D on cardiomyocytes occurs via the modulation of mitochondrial function.

Mitochondria are highly dynamic organelles that undergo fission and fusion, and these processes are closely related to mitochondria function and ROS production (Westermann, 2010). MitoTracker was used to examine the effects of Vit D on the mitochondrial morphology in H/R-treated cardiomyocytes. The results showed that the length of the mitochondria in H/R-treated cells was significantly shorter than that in control-treated cells, and Vit D reversed this effect (Figures 4A,B). To further investigate how Vit D regulates mitochondrial fission, p-Drp1 and Mff, two regulatory proteins associated with mitochondrial fission, were examined. Our results showed that H/R treatment increased the expression of p-Drp1 and Mff, whereas the expression of these proteins was decreased by Vit D (Figures 4C–E). To further confirm the effects of mitochondrial ROS on mitochondrial fission after H/R injury, we examined the expression of p-Drp1 and Mff in Mito TEMPO and H/R-treated cardiomyocytes. We found that MitoTEMPO reduced the expression of both p-Drp1 and Mff (Figures 4F–H). Next, we investigated whether Drp1 interacts with Mff. We found that Drp1 was co-immunoprecipitated with Mff following H/R treatment, while Vit D decreased the interaction (Figure 4I). Our results suggest that the protective effect of Vit D against H/R-induced mitochondrial fission is attributed to its antioxidant capacity.

It has been reported that mitophagy plays a regulatory role in apoptosis (Yussman et al., 2002). Acridine orange (AO) staining was used to detect autophagic vacuoles. H/R-treated cells exhibited increased autophagosome formation, which was inhibited when cells were pretreated with Vit D; these results indicated that Vit D reduced the accumulation of autophagosomes (Figure 5A). In addition, the results of AO analysis by flow cytometry suggested that H/R injury significantly increased the population of cells undergoing autophagy (AO green positive and red positive), while Vit D decreased this population. AO green and red fluorescence overlay showed that H/R treatment enhanced AO staining, whereas Vit D reduced AO staining (Figure 5B). To investigate how Vit D affects mitophagy, we analyzed the expression of BNIP3 and the conversion of LC3I to LC3II by Western blot. The results showed that H/R increased BNIP3 expression and the LC3II/LC3I ratio. Nevertheless, Vit D reduced the expression of both BNIP3 and LC3BII (Figures 5C–E). To further confirm the effects of mitochondrial ROS on mitophagy after H/R injury, we examined the expression of BNIP3 and the conversion of LC3I to LC3II in MitoTEMPO and H/R-treated cardiomyocytes. We found that MitoTEMPO reduced the expression of both BNIP3 and LC3BII/LC3BI ratio (Figures 5F–H). Next, we investigated whether BNIP3 interacts with LC3B. Strong interactions between BNIP3 and LC3B were observed in H/R-treated cardiomyocytes, and the addition of Vit D markedly inhibited this interaction (Figure 5I). Taken together, these results indicate that Vit D can attenuate H/R-induced mitophagy by reducing mitochondrial ROS production.

We then further evaluated the mitochondrial translocation of Drp1, Mff, BNIP3, and LC3B in response to H/R treatment. H/R treatment induced a significant increase in Drp1, Mff, BNIP3 and LC3B translocation to the mitochondria. However, Vit D treatment reduced the translocation of these proteins (Figure 6A). Using double immunofluorescence staining, we further investigated changes in the mitochondria undergoing mitophagy. We observed that co-localization of COX IV (a mitochondrial marker) with Drp1 or LC3B was more clearly present in the H/R group than in the control group, while Vit D pretreatment reduced these phenomena (Figures 6B,C). Transmission electron microscopy images showed that Vit D reduced the presence of autophagic vacuoles in H/R-treated cardiomyocytes (Figure 6D). To further evaluate the effect of the mitochondrial fission on H/R-induce mitophagy, we performed the treatment of mitochondrial fission inhibitor (Mdivi-1) on the expression of mitophagy-related proteins in H/R-treated H9c2 cells. Cells with Mdivi-1 treatment significantly reduced the levels of H/R-induced BNIP3 and LC3BII/LC3BI ratio expression (Figure 6E,F). These results indicated that mitochondrial fission was involved in H/R-induced mitophagy. Furthermore, we examined the effects of Vit D on HR-induced cardiomyocyte injury via mitophagy, we performed the effect of bafilomyocin A1, an autophagy inhibitor, on the expression of mitophagy-related proteins. Cells with bafilomyocin A1 treatment significantly reduced H/R-induced BNIP3 and LC3BII/LC3BI expression (Figures 6G,H).

Compared with sham surgery, I/R induced the infiltration of inflammatory cells into the infarct zone of cardiac tissues, and Vit D ameliorated this inflammatory infiltration (Figure 7A). I/R increased the intracellular levels of superoxide anions, as shown by DHE staining, while Vit D decreased the intercellular levels of superoxide anions in the heart after I/R injury (Figure 7B). Ultrastructure images were examined by TEM. The mice administered Vit D treatment showed complete mitochondrial structures with well-aligned cristae (Figure 7C). I/R treatment significantly increased the number of apoptotic cells, while Vit D pretreatment reduced the effect (Figure 7D). The levels of HIF-1α, Mff, p-Drp 1, BNIP3, LC3B and cleaved caspase three in the I/R-treated mice were significantly elevated compared to those in the control-treated mice. Furthermore, Vit D treatment significantly reduced the levels of these proteins during I/R damage (Figure 7E). Furthermore, we examine the effects of Vit D on I/R-induced mitophagy, we performed the effect of bafilomyocin A1 on the expression of mitophagy-related proteins. Mice with bafilomyocin A1 treatment significantly reduced I/R-induced BNIP3 and LC3BII/I expression (Figure 7F).