A total of 20 patients with CHD, 1–28 years old (mean, 6.02 years old), were eligible for inclusion in the present study. All patients were admitted to the Institute of Cardiovascular Surgery, Xinqiao Hospital of the Third Military Medical University (Chonqing, China). Of these, 10 patients presented with a cyanotic defect and 10 with an acyanotic defect. The local Ethics Review Board of the Third Military Medical University Affiliated Hospital approved the study protocol and the informed consent forms. This investigation conforms to the principles outlined in the Declaration of Helsinki. Written informed consent was obtained from the parents of all participants.

Human samples were sectioned and embedded as previously described (14). Briefly, a myocardial biopsy specimen was obtained from the right ventricular outflow tracts during right ventricular outflow tract reconstruction. The myocardial samples intended for western blotting and quantitative polymerase chain reaction (qPCR) analysis were immediately snap-frozen in liquid nitrogen and stored at −80°C until analysis.

The embryonic rat heart-derived H9c2 cells purchased from the American Type Culture Collection (CRL-1466) were grown in Dulbecco's modified Eagle's medium (Gibco; Thermo Fisher Scientific, Inc. Waltham, MA, USA) supplemented with 10% fetal bovine serum (Biological Industries, Beti-Haemek, Israel) at 37°C and exposed to 21% O₂, 5% CO₂ and 74% N₂. Following serum starvation overnight, cells in the hypoxic group were placed in an in vivo 200 cultivator (Ruskinn Technology, Ltd., Bridgend, UK) containing a gaseous mixture of 94% N₂, 5% CO₂ and 1% O₂ at 37°C for 48 h. In addition, the AMPK activation group was treated with AICAR (1 mM; Sigma-Aldrich; Merck KGaA, Darmstadt, Germany), an AMPK agonist, while the inhibition group was exposed to compound C (1 μM; Merck KGaA), an AMPK inhibitor. Cells in the normoxic groups were incubated under the same conditions, except with 21% O₂ concentrations.

Total DNA was isolated from H9c2 cells and myocardial samples using DNAiso Plus (Takara Biotechnology Co., Ltd., Dalian, China) according to the manufacturer's instructions. For the quantitative determination of mitochondrial DNA (mtDNA) content relative to nuclear DNA (nDNA), primers specific for amplification of mtDNA Cyt and nDNA-encoded GAPDH were selected. The following primer sequences were used: Cyt sense, 5′-CCGGAGCAATCCAGGTCGGTT-3′ and antisense, 5′-TGGTTGGGAGCACTTATGGTAAGGA-3′; and GAPDH sense, 5′-TGGGGAAGGTGAAGGTCGGAGT-3′ and antisense, 5′-CCCGTTCTCAGCCTTGACGGTG-3′. SYBR^® Premix Ex Taq™ II (Takara Biotechnology Co., Ltd.) was used according to the manufacturer's instructions, and qRT-PCR was performed using a StepOnePlus™ Real-Time PCR system (Applied Biosystems; Thermo Fisher Scientific, Inc.). The reaction conditions were as follows: 95°C for 30 sec, followed by 45 cycles of 95°C for 5 sec and 60°C for 30 sec. Relative expression of target mtDNA was normalized to GAPDH using the ΔΔCq method (15).

The cells were treated with or without chloroquine (Chl; 10 μM; Sigma-Aldrich; Merck KGaA), which inhibits fusion of autophagosomes with lysosomes (16), for 4 h prior to incubation at 37°C for 30 min in media containing MitoTracker (1:4,000; Life Technologies; Thermo Fisher Scientific, Inc.), then fixed with 4% formaldehyde for 15 min at room temperature and blocked with PBS containing 10% goat serum, 0.3 M glycine, 1% BSA and 0.3% Triton X-100 (Beyotime Institute of Biotechnology, Beijing, China) for 1 h at room temperature. Staining of the treated cells with rabbit anti-rat microtubule-associated proteins 1B light chain 3B (LC3B) polyclonal antibody (1:200; L7543; Sigma-Aldrich) was performed overnight at 4°C in PBS containing 1% BSA and 0.3% Triton X-100. Alexa Fluor 488 anti-rabbit antibody (1:500; A0423; Beyotime Institute of Biotechnology) was used as the secondary antibody incubated for 1 h at room temperature. A confocal laser scanning microscope (Leica TCS SP2; Leica Microsystems, Wetzlar, Germany) was used to image stained cells, and the co-localization dots were analyzed using ImageJ software (v1.6.0; National Institutes of Health, Bethesda, MD, USA). Each group contained three dependent samples, and every sample was analyzed in five random fields.

The adherent cells were incubated with MitoTracker for 30 min as aforementioned, followed by treatment with rhodamine (5 μM; Beyotime Institute of Biotechnology) for 20 min. A confocal laser scanning microscope was used to image the stained cells. Five random fields were analyzed per sample.

JC-1 staining was also used to detect alterations of mitochondrial transmembrane potential (ΛΨm). Cells were collected, washed and suspended in JC-1 binding buffer (Beyotime Institute of Biotechnology). JC-1 dye (1:100; Beyotime Institute of Biotechnology) was added and the cells were incubated for 20 min at 37°C in the dark, followed by washing 2 times with binding buffer. The cell suspensions were analyzed by flow cytometry.

Cell apoptosis was determined by Annexin V-fluorescein 5-isothiocyanate (Annexin V-FITC) assay. H9c2 cells (1×10⁷ cells in 100 μl) were collected, washed and suspended in Annexin V binding buffer (Wanleibio Co., Ltd., Shanghai, China). FITC-conjugated Annexin V and propidium iodide (PI) (1:100; Wanleibio Co., Ltd.) were added to the binding buffer that contained suspended cells, both incubated for 15 min following the manufacturer's protocols. Following incubation, the cells were analyzed by flow cytometry (MoFloTMXDP; Beckman Coulter, Co., Fullerton, CA, USA).

H9c2 cells were collected from the cultures and replated into a 96-well plate (1×10⁵ cells/ml). The cells in the hypoxic group, the AMPK activation group and the AMPK inhibition group were treated as described above. Cell proliferation was measured by Cell Counting Kit-8 (CCK-8; Wuhan Bioengineering Institute, Wuhan, China) assays according to the manufacturer's protocols. Prior to reading the absorbance at 450 nm, 10 μl CCK-8 solution and 100 μl complete medium were sequentially added to each well.

H9c2 cells and myocardial samples were lysed in ice-cold lysis buffer (radioimmunoprecipitation assay), with a protease inhibitor (0.5 mM PMSF) (both Beyotime Institute of Biotechnology) and phosphatase inhibitor (1 tablet/10ml; Roche Diagnostics GmbH, Mannheim, Germany). The lysates were further centrifuged at 15,000 × g for 15 min at 4°C. Proteins from the H9c2 cells and myocardial samples lysates were measured using the Bradford method. Cell proteins (20 μg) were extracted with SDS lysis buffer and separated by 10% SDS-PAGE gel (both from Beyotime Institute of Biotechnology). Subsequently, protein samples were transferred to 0.22-μm polyvinylidene difluoride membranes (Roche Diagnostics GmbH), which were blocked with 5% BSA for 1 h at room temperature, followed by incubation with primary antibodies at 4°C overnight. The primary antibodies included the following: Rabbit anti-mouse AMPKα polyclonal antibody (1:1,000; #5831), rabbit ant-mouse p-AMPKα polyclonal antibody (1:1,000; #2535) (both Sigma-Aldrich; Merck KGaA) and rabbit anti-mouse β-actin monoclonal antibody (1:1,000; bs-0061R; Bioss, Beijing, China). Next, the membranes were incubated with horseradish peroxidase-conjugated secondary antibodies (1:5,000; A0208) for 1 h at 37°C and detected using an enhanced chemiluminescence kit (both from Beyotime Institute of Biotechnology), and densitometry signals were quantified using ImageQuant TL software (General Electric, Co., Fairfield, CT, USA).

Data from at least three independent experiments are expressed as the mean ± SD. SPSS 16.0 software (SPSS, Inc., Chicago, IL, USA) was used for the statistical analysis. The difference in means between 2 groups was evaluated using Student's t-test, and one-way analysis of variance was used for comparisons between ≥2 groups. P<0.05 was considered to indicate a statistically significant difference.

The epidemiological data of the patients is summarized in Table I. Patients with cyanotic heart disease had significantly lower preoperative arterial oxygen saturation (SaO₂) than patients with acyanotic heart disease. Comparing the cyanotic patients with the acyanotic patients, the relative mtDNA/nDNA ratio was found to be higher in the cyanotic group. Similarly, the ratio of H9c2 cells in the hypoxic group was higher than that in the H9c2 cells cultured under normoxic conditions. The copies of mtDNA Cyt were significantly increased in the cyanotic patients and the H9c2 cells of hypoxic group when compared with the acyanotic patients and the H9c2 cells of the normoxic group, respectively (Fig. 1A). Fluorescence microscopy was performed to detect ΛΨm in the cells cultivated under hypoxic and normoxic conditions. Cells were co-stained with rhodamine 123 and MitoTracker, and it was found that the transmembrane potential was decreased in hypoxic cardiomyocytes, as the co-localization of rhodamine 123 and MitoTracker was decreased (Fig. 1B). Moreover, after the cells were stained with JC-I, which more specifically reduces mitochondrial ΛΨm, the loss of ΛΨm was increased and the number of H9c2 cells with intact ΛΨm was reduced compared with the normoxic H9c2 cells (Fig. 1C). Loss of the ΛΨm is a key step in the intrinsic apoptotic pathway (17). Therefore, the apoptosis of H9c2 cells was evaluated using flow cytometry following staining of the cells with Annexin V-FITC in combination with PI. A significantly higher percentage of apoptotic cells were detected in the hypoxic group (Fig. 1D). The CCK-8 assay showed that the proliferation of the H9c2 cells was significantly decreased in the hypoxic group compared with that in the normoxic group (Fig. 1E). The results revealed that compared with that in acyanotic patients, mitochondrial biogenesis was upregulated in patients with cyanotic heart disease and in the H9c2 cells cultivated in the chronic hypoxic condition. In addition to the upregulated mitochondrial biogenesis, the loss of ΛΨm and the apoptosis of the cells were also significantly increased compared with the normoxic cells. These data suggested that chronic hypoxia promoted mitochondrial biogenesis and induced mitochondrial damage and cell apoptosis at the same time.

As the damaged and dysfunctional mitochondria increased in the H9c2 cells exposed to long periods of hypoxia, selective mitochondrial autophagy was detected in cells exposed to hypoxia. After staining the cells with antibody to LC3B and immunofluorescence assessment, it was found that LC3B rarely localized in the mitochondria, confirming the presence of selective mitochondrial autophagy, also known as mitophagy, in the cardiomyocytes. Notably, the expression of LC3 was increased in the mitochondria when the H9c2 cells were subjected to long periods of hypoxia compared with that in cardiomyocytes cultured under normoxic conditions. Moreover, when the cardiomyocytes were treated with Chl, an inhibitor of autophagosome-lysosome fusion, the expression of LC3 in the mitochondria was significantly increased (Fig. 2). These results suggested that mitophagosomes were increased in the H9c2 cells under hypoxic conditions and that the mitophagy was induced by chronic hypoxia.

Since AMPK plays a significant role in maintaining intracellular homoeostasis during hypoxia, the present study first investigated the impact of chronic hypoxia on AMPK activation. As expected, western blotting results confirmed that the level of phosphorylation of AMPK was increased in the ventricular myocardium of cyanotic patients and in the H9c2 cells cultured under hypoxic conditions (Fig. 3A and B). To test further whether AMPK was necessary in regulating mitophagy, the activation of AMPK was determined following different treatments, including AMPK activator AICAR and AMPK inhibitor compound C, and it was found that the level of phosphorylation of AMPK was increased in the AMPK activation group and decreased in the AMPK inhibition group compared with that in the hypoxic control group (Fig. 3B). Immunofluorescence showed that the AMPK activator increased the co-localization of LC3 dots (green) and mitochondria (red) in the Chl-treated and untreated H9c2 cells, while the AMPK inhibitor presented the opposite results (Fig. 3C and D). These results demonstrated that the activity of AMPK was increased in the myocardium of infants with cyanotic cardiac defects and in the H9c2 cells under hypoxia. In addition, mitophagy was promoted in the cells of the AMPK activation group and suppressed in the AMPK inhibition group, thereby suggesting that activation of AMPK is a critical factor for mitophagy under chronic hypoxia.

To determine whether AMPK contributes to mitochondrial quality control and protects cardiomyocytes during chronic hypoxia, ΛΨm was detected using fluorescence microscopy following rhodamine 123 and MitoTracker staining. Compared with that of the hypoxic control group, the percentage of dysfunctional mitochondria was decreased in the AMPK activation group, but increased in the AMPK inhibition group (Fig. 4A). Following the JC-1 staining, flow cytometry revealed that the number of H9c2 cells with reduced ΛΨm was decreased in the AMPK activation group and increased in the AMPK inhibition group compared with that of the hypoxic group (Fig. 4B). The apoptosis of H9c2 cells was detected using flow cytometry following staining with Annexin V-FITC in combination with PI. It was observed that, compared with that in the control group, the ratio of early apoptotic cells was decreased following treatment of AICAR, but increased following treatment with compound C (Fig. 4C). The proliferation of H9c2 cells that was detected by CCK-8 assay was increased by AMPK activator, but decreased by AMPK inhibitor (Fig. 4D). These data suggested that under chronic hypoxic conditions, activation of AMPK contributed to the mitochondrial quality and has a protective role for cell survival.