Introduction

Myocardial infarction (MI) is a cardiovascular disease which is mainly characterized by myocardial necrosis caused by acute or persistent ischemia and anoxia. Although survival following a first myocardial infarct approaches 90% today, re-infarction occurs commonly and carries a high mortality (Dutta et al.2012). In a population study, 1-yr risk of re-infarction reached 17.4% (Christos et al.2010). Timely reperfusion after the onset of ischemia is the current standard therapy for myocardial ischemia (Gerczuk and Kloner 2012; Members et al.2013). However, reperfusion strategies may result in another unpleasant injury, ischemia/reperfusion (I/R) (Yellon and Hausenloy 2007; Hausenloy and Yellon 2015). Reperfusion injury is mainly caused by some cellular alterations such as oxidative stress and inflammation (Gong et al.2013; Braunwald 2015). Thus, theoretical research on preventing myocyte death following I/R is necessary to develop the novel strategies on treating patients with MI.

Oxidative stress induced by I/R always related to autophagy in cardiomyocyte (Hariharan et al.2011). Recent studies have shown that levels of autophagy were enhanced in the I/R-injured cardiomyocyte (Jian et al.2011; Przyklenk et al.2012). However, a study has reported that autophagy has a dual, opposite role in the heart, depending on the stimulus (Gustafsson and Gottlieb 2009). There is general agreement that upregulation of autophagy is cardioprotective and profoundly attenuates myocardial ischemia/reperfusion injury (Kanamori et al.2011; Loos et al.2011; Cui et al.2014). But excessive autophagy also could induce apoptosis of cardiomyocyte (Huang et al.2015; Huang et al.2017).

Mitophagy, the specific autophagic elimination of mitochondria, is regulated in many metazoan cell types by Parkin and PINK1 (Youle and Narendra 2011). Parkin protein deficiency could promote cardiac injury and reduce mice survival following myocardial infarction (Kubli et al.2013). PINK1 played as a cardioprotective kinase (Siddall et al.2008), and a study has reported that PINK1-mediated autophagy was regulated by PTEN (Roe et al.2015). Robust evidence also proved PTEN plays an important role in autophagy (Ueno et al.2008). PTEN nuclear translocation could promote autophagy in response to DNA damage (Chen et al. 2015a). However, little was known about PTEN regulating mitophagy in I/R-injured cardiomyocyte.

APE1, an oxidative repair protein, can transfer from nuclear to mitochondrion. Reactive oxygen species (ROS) could induce the activation of APE1 (Pines et al.2005). Its main function is repairing spontaneous DNA damage (Whitaker et al.2018), and this function also existed in mitochondrion (Li et al.2010). However, it remains unclear whether APE1 influence mitochondrial function during I/R injuring in cardiomyocyte.

In this study, we cultured H9c2 cells and conducted hypoxia/reoxygenation (H/R) treatment. We showed that inhibiting of PTEN function and APE1 overexpression may play a regulatory role in mitophagy, which may be regulate number of mitochondrion and maintain its quality control. PTEN-mediated mitophagy may protect against cardiac H/R injury.

Materials and Methods

Cell culture

H9c2 cells (rat cardiomyocytes) were purchased from American Type Culture Collection (ATCC, Manassas, VA) Cells resuscitated from liquid nitrogen refrigeration were cultured in high glucose-Dulbecco Minimum Essential Medium (H-DMEM) (Solarbio, Beiking, China) with 10% fetal bovine serum (FBS) (Thermo Fisher Scientific, Waltham, MA) and 1% streptomycin/penicillin, at 37°C with 5% CO2.

Hypoxia/reoxygenation

Cells were cultured with H-DMEM/serum-free starvation, 2 h before exposure to hypoxia. To conduct hypoxia, H9c2 cells were placed in a 37°C airtight chamber saturated with 95% N2/5% CO2 for 24 h. Reoxygenation was performed in a 37°C/5% CO2 incubator under H-DMEM/10% FBS for 30 min (Webster et al.1999). To inhibit function of PTEN protein, 5 μM of bpv (phen) (Enzo Life Sciences, Farmingdale, NY) was added in the above process. After H/R treatment, cells were harvested and used in other assays.

Cell transfection

To overexpress APE1 in H9c2 cells, the coding sequence of APE1 was cloned through primers (forward: 5′-GGG GTA CCA TGC CGA AGC GTG GGA AAA AG-3′; reverse: 5′-CCG CTC GAG TCA CAG TGC TAG GTA TAG GGT GAT AGG AC-3′) and inserted into the expression plasmid pcDNA3.0 (Invitrogen, Carlsbad, CA). Recombinant plasmid was transfected into H9c2 cells through Lipofectamine 2000 Reagent (Invitrogen). Cells transfected with the wild-type plasmid pcDNA3.0 were used as control. At 24 h after transfection, cells were harvested and used to perform H/R treating.

Cell activity assay

Equal numbers of H9c2 cells (1 × 104) suffered H/R treating in each group. Subsequently, cells were washed 3 times and resuspended by phosphate buffer saline. Then cell activities were detected through the Cell Counting Kit-8 (Dojindo, Japan). Briefly, 100 μl of CCK-8 solution was added into each well and then it was incubated for 4 h at 37°C with 5% CO2. Finally, the optical density at 450 nm was recorded by a microplate reader (Thermo Scientific).

Flow cytometry

The apoptosis induced by H/R injury was measured by flow cytometry assay using an Annexin V-FITC/PI Apoptosis Kit (MultiSciences, Hangzhou, China). The H/R-treated cells (5 × 105) were washed 3 times by phosphate buffer saline and moved to flow tubes after trypsin digestion. Then they were resuspended by 200 μl of Annexin V-FITC/PI solution and incubated for 15 min at dark. Finally, the fluorescence signal value was measured by a BD FACS Calibur™ flow cytometry (BD Biosciences, Franklin, NJ).

Hypoxia/reoxygenation

Cells were cultured with H-DMEM/serum-free starvation, 2 h before exposure to hypoxia. To conduct hypoxia, H9c2 cells were placed in a 37°C airtight chamber saturated with 95% N2/5% CO2 for 24 h. Reoxygenation was performed in a 37°C/5% CO2 incubator under H-DMEM/10% FBS for 30 min (Webster et al.1999). To inhibit function of PTEN protein, 5 μM of bpv (phen) (Enzo Life Sciences, Farmingdale, NY) was added in the above process. After H/R treatment, cells were harvested and used in other assays.

Autophagy detection

The level of autophagy in the H/R-treated H9c2 cells was detected by the Ad-GFP-LC3B, adenovirus expressing GFP-LC3B fusion protein (Beyotime, Shanghai, China). Briefly, cells (1 × 105) were transfected with Ad-GFP-LC3B (20 MOI) at 24 h before the H/R treating. After H/R treating, the green fluorescence signal was visualized by a Leica DMI6000 B Microscope (Leica, Germany). Furthermore, we used the mt-Keima assay (Bingol et al.2014) to observe the mitophagy in the H9c2 cells. Briefly, cells (1 × 105) were transfected with lentivirus-packaged mt-Keima-COX8 (50 MOI), which leads to overexpression of Keima protein in mitochondrion. As the Keima protein releases different fluorescence signals under acidic and neutral conditions, the mitophagy can be observed when the mitochondrial autophagy is fused with acid lysosome. Twenty-four hours after transfection, cells suffered H/R treating or other administration. Finally, the green fluorescence signal (458 nm) and red fluorescence signal (543 nm) were visualized by a confocal microscopy (TCS SP8 X, Leica, Germany).

Western blotting

Cell total proteins were extracted with lysis buffer containing 50 mM Tris (pH 7.6), 150 mM NaCl, 1% TritonX-100, 1% deoxycholate, 0.1% SDS, 1 mM PMSF, and 0.2% aprotinin (Sigma-Aldrich, Saint Louis, MO). The cytoplasmic and mitochondrial protein was extracted through a Cytoplasmic and Mitochondrial Protein Extraction Kit (Sangon Biotech, Shanghai, China). The concentrations of extracted protein were determined with the BCA Protein Assay kit (Beyotime). Then 20-μg protein samples were separated by a SDS-PAGE electrophoresis and transferred to a polyvinylidene fluoride membrane. The membrane was blocked with 5% evaporated skimmed milk tris-buffered saline (TBS) at room temperature overnight; washed 3 times by TBS; and subsequently incubated with anti-Cytochrome C (Abcam, diluted 1:800), anti-APE1 (Abcam, diluted 1:2000), anti-PTEN (Abcam, diluted 1:400), anti-PINK1 (Abcam, diluted 1:500), anti-Parkin (Abcam, diluted 1:400), anti-LC3 (Abcam, diluted 1:1500), or anti-P62 (Abcam, diluted 1:1000) at 4°C overnight. GAPDH (Abcam, diluted 1:5000) and COX (Abcam, diluted 1:1000) were used as an internal control, washed 3 times again, and incubated with HRP Goat anti-Rabbit IgG (BOSTER, Wuhan, China) for 40 min. Finally, the protein bands were visualized with ECL-Plus reagent (Millipore, Billerica, MA) and scanned using a bio-imaging analyzer (Bio-Rad, Hercules, CA).

Statistical analysis

All data were obtained from at least three independent experiments and represented as mean ± standard deviation. An ANOVA analysis with Tukey’s multiple comparisons test was used to analyze differences between multi-groups in GraphPad Prism 6 (GraphPad Software, San Diego, CA). P values < 0.05 were considered statistically significant.

Results

Loss of PTEN function promotes cardiac H/R injury

As PTEN plays a promotion role in autophagy (Arico et al.2001; Ueno et al.2008), this study initially aimed to explore its role in the H/R-injured cardiomyocyte. To inhibit function of PTEN protein, 5 μM of bpv (phen) was added into medium throughout the process of H/R injury. CCK-8 detection found that the cell viability in the H/R group is significantly decreased compared with the control group, and in the H/R + bpv (phen) group, it is also significantly lower than that in the H/R group (Fig. 1A). In addition, results of flow cytometry assay showed that H/R treating significantly promotes apoptosis of H9c2 cells and this effect is significantly enhanced by stimulation of bpv (phen), especially 19.3% of early apoptosis in the H/R group increased to 31.9% in the H/R + bpv (phen) group (Fig. 1B). As release of Cytochrome C from mitochondria induced caspase activation and apoptosis (Jiang and Wang 2004), we measured protein level of Cytochrome C in H/R-treated H9c2 cells by western blotting. Result showed that protein level of cytoplasmic Cytochrome C in the H/R group is significantly higher than that in the control group, and the stimulation with bpv (phen) significantly promotes the accumulation of cytoplasmic Cytochrome C (Fig. 1C, D). These results suggested that loss of PTEN function promotes the apoptosis and release of Cytochrome C from mitochondria in H/R-injured H9c2 cells.

Figure 1.
figure 1

Bpv (phen) exacerbated cardiac H/R injury. H/R-injured H9c2 cells were simulated with 5 μM of bpv (phen). (A) Result of CCK-8 assay showed bpv (phen) caused a further decrease of cell viability. (B) Bpv (phen) increased the apoptosis of H/R-injured H9c2 cells. (C, D) Western blot analysis with cytoplasmic protein showed an increased release of Cytochrome C in H/R + bpv (phen) group. GAPDH was used as reference protein. *P < 0.05 and ***P < 0.001 vs. control; #P < 0.05, ##P < 0.01, and ###P < 0.001 vs. H/R.

Full size image

Stimulation with bpv (phen) inhibits mitophagy in H/R-injured H9c2 cells

To illuminate whether bpv (phen) influences the release of Cytochrome C from mitochondria by regulating mitophagy, we detected protein level of LC3B, a key factor indicating autophagy. Results showed that H/R treating significantly induced expression of LC3B proteins, including LC3B I, LC3B II, and the ratio of LC3B II/LC3B I, while they were significantly suppressed by the stimulation of bpv (phen), and it appeared the opposite change for the levels of P62, a substrate of autophagy (Fig. 2A–C). Meanwhile, results showed by the Ad-GFP-LC3B also proved that H/R treating increased the autophagy level of H9c2 cells, while bpv (phen) decreased it (Fig. 2E). The mitophagy showed by mt-Keima assay has a same tendency (Fig. 2D). Western blotting revealed that total levels of PTEN and APE1 were both significantly increased in the H/R group (Fig. 2F, G). Protein of mitochondrial PINK1 and Parkin was both significantly accumulated in the H/R group and significantly reduced by the stimulation of bpv (phen) (Fig. 2H, I). Of note, the level of mitochondrial APE1 was only increased in the H/R group. These results suggested that loss of PTEN function induces a level decline of autophagy in H/R-injured H9c2 cells, which may be a potential reason for increased release of Cytochrome C from mitochondria.

Figure 2.
figure 2

Bpv (phen) inhibited PINK1/Parkin-mediated mitophagy in H/R-injured H9c2 cells. H/R-injured H9c2 cells were simulated with 5 μM of bpv (phen). (A, B) Western blot analysis with cytoplasmic protein. Bpv (phen) significantly decreased the level of LC3B I and LC3B II, while increased level of P62. (C) The ratio of LC3B II/LC3B I. (D) The inhibited autophagy in bpv (phen)-treated group was determined with Ad-GFP-LC3B. Scale bar = 20 μm. (E) The mitophagy was shown by mt-Keima assay. Scale bar = 20 μm. (F, G) Western blot analysis with mitochondrial protein. The level change of total PTEN and APE1. (H, I) Western blot analysis showed the level change of mitochondrial PINK1, Parkin, and APE1. GAPDH and COX-4 were used as reference protein. *P < 0.05 vs. control; #P < 0.05 vs. H/R.

Full size image

Overexpression of APE1 attenuates cardiac H/R injury

Above results showed level of mitochondrial APE1 was changed in H/R-injured H9c2 cells. To explore the role of APE1 in cardiac H/R injury, we overexpressed APE1 in H9c2 cells. Then it was found that overexpression of APE1 significantly increased the cell viability of H/R-injured H9c2 cells when compared with the H/R group (transfected with blank empty vector) (Fig. 3A). The percentage of early apoptosis in APE1-overexpressed H9c2 cells was 11.4%, which was significantly lower than 20.9% of the H/R group (Fig. 3B). Overexpression of APE1 attenuated the apoptosis induced by H/R injury in H9c2 cells. In addition, the bands of western blotting showed that overexpression of APE1 also significantly decreased the protein level of cytoplasmic Cytochrome C (Fig. 3C, D). These revealed that APE1 plays a protective role in cardiac H/R injury.

Figure 3.
figure 3

Overexpression of APE1 protected against cardiac H/R injury. (A) APE1-overexpressed H9c2 cells showed an increased cell viability. (B) Overexpression of APE1 inhibited apoptosis in H/R-injured H9c2 cells. (C, D) Western blot analysis with cytoplasmic protein showed an inhibited release of Cytochrome C in H/R + APE1 group. GAPDH was used as reference protein. **P < 0.01 vs. control; #P < 0.05 and ##P < 0.01 vs. H/R.

Full size image

APE1 promotes PINK1/Parkin-mediated mitophagy in H/R-injured H9c2 cells

Studies reported that APE1 mainly functions as a DNA repair enzyme (Tell et al.2009; Whitaker et al.2018). In this study, we aimed to discuss its effect on mitophagy. In APE1-overexpressed and H/R-injured H9c2 cells, we found the protein level of LC3B II and the ratio of LC3B II/LC3B I was significantly increased compared with that of only H/R-injured H9c2 cells, while it appeared the opposite change for the levels of P62 (Fig. 4A–C). The increased autophagy level in APE1-overexpressed and H/R-injured H9c2 cells was also proved by the more quantity of fluorescent spots showed with the Ad-GFP-LC3B (Fig. 4D). The mitophagy showed a same tendency (Fig. 4E). Overexpression of APE1 was subsequently validated by western blotting, which showed both total cell APE1 and mitochondrial APE1 were significantly increased (Fig. 4F–I). Meanwhile, PINK1 and Parkin, two related proteins involved in mitophagy, were also accumulated in mitochondrion in APE1-overexpressed cells. These means overexpression of APE1 promotes the process of PINK1/Parkin-mediated mitophagy.

Figure 4.
figure 4

Overexpression of APE1 promoted PINK1/Parkin-mediated mitophagy in H/R-injured H9c2 cells. (A, B) Western blot analysis with cytoplasmic protein. In H/R-injured H9c2 cells, overexpression of APE1 significantly increased the level of LC3B I and LC3B II, while decreased the level of P62. (C) The ratio of LC3B II/LC3B I. (D) The enhanced autophagy in APE1 overexpression group was determined with Ad-GFP-LC3B. Scale bar = 20 μm. (E) The mitophagy was shown by mt-Keima assay. Scale bar = 20 μm. (F, G) Western blot analysis with mitochondrial protein. It showed the level change of total PTEN and APE1. (H, I) Western blot analysis showed the level change of mitochondrial PINK1, Parkin, and APE1. GAPDH and COX-4 were used as reference protein. *P < 0.05 vs. control; #P < 0.05 vs. H/R.

Full size image

Discussion

Mitochondrial dysfunction is one of the damage induced by I/R or H/R injury. Mitochondria are the primary sites for oxidative stress within cardiac cells. In this study, we performed H/R injury treatment in H9c2 cells. In H/R-injured cells, Cytochrome C was accumulated and the apoptosis rate was increased. Release of Cytochrome C induced by H/R could activate caspase-dependent apoptosis (Zou et al.1997). Furthermore, level of autophagy and protein levels of mitochondrial PINK1 and Parkin were increased after the injury. H/R-induced autophagy has been reported in previous studies (Zhang et al.2012; Liu et al.2016). Activated PINK1/Parkin signal pathway in H9c2 cells indicated that the specific autophagy, mitophagy, was also induced by H/R injury (Sun et al.2015). Considering that upregulation of autophagy is cardioprotective in cardiac H/R injury (Kanamori et al.2011; Loos et al.2011), it can be speculated that activated mitophagy may be a mechanism of self-repair of H9c2 cells, which could play an important role in the clearance of dysfunctional mitochondria and consequent decrease of Cytochrome C release.

To inhibit the function of PTEN, we stimulated H9c2 cells with the bpv (phen). Subsequently, we found that bpv (phen) exacerbated the cardiac H/R injury. Apoptosis rate was significantly increased under the stimulation of bpv (phen) plus H/R injury. A study has reported that bpv (phen) induced apoptosis through influencing autophagy (Chen et al. 2015b). However, our results were inconsistent with the study of Tian et al. (Tian et al. 2012), and we speculated that it may be mainly because of the different H/R-processing time or different concentration of bpv (phen). In our study, results showed that bpv (phen) decreased the protein levels of mitochondrial PINK1 and Parkin. PINK1/Parkin-mediated mitophagy was also inhibited by bpv (phen), which may subsequently induce the augment of dysfunctional mitochondria. The study of Yin et al. also proved that PTENα-deficiency impaired mitophagy-dependent mitochondrial clearance and exacerbated the cardiac H/R injury (Yin et al.2018).

In the study, we found another mitochondrial protein, APE1, was also accumulated in H/R-injured H9c2 cells. APE1 overexpression partially decreased apoptosis rate and level of Cytochrome C. The author reported that APE1 can hinder the apoptosis of cardiac progenitor cells (Aonuma et al.2016). Meanwhile, APE1 also overcome the multifunctional oxidative damage in nerve cells (Kaur et al.2015). Of note, our results showed that APE1 overexpression also influenced the level of autophagy, which may be caused by the increase of mitochondrial PINK1 and Parkin. Parkin, a tumor suppressor and Parkinson’s disease (PD)–associated gene, has been found to play a role in maintaining the quality of mitochondrial APE1 (Scott et al.2016). We suspected that the accumulation of multifunctional APE1 promoted the increase of mitochondrial Parkin, which was also shown by results of western blotting, and it subsequently induced high level of mitophagy. However, the mechanism of APE1 overexpression influencing autophagy still needs further experiments to explore.

Conclusion

In summary, our results found inhibition of PTEN function induced by bpv (phen)-suppressed PINK1/Parkin-mediated mitophagy, which resulted in an increased apoptosis and release of mitochondrial Cytochrome C in H/R-injured H9c2 cells. Meanwhile, overexpression of APE1 also influenced the level of mitophagy and protected against cardiac hypoxia/reoxygenation injury. This study may provide a little theoretical basis for developing the novel strategies on treating patients with myocardial ischemia disease.