Rat embryonic cardiomyoblast-derived H9c2 cells (TCR-C607; Cas9X, Suzhou, China) were maintained in high-glucose Dulbecco's modified Eagle's medium (DMEM; Boster, Wuhan, China) containing 10% fetal bovine serum (FBS, CF01S02; CellBox, Hunan, China) and 1% penicillin–streptomycin antibiotic cocktail (PYG0016; Boster, Wuhan, China). Cells were incubated at a constant temperature of 37°C under a humidified atmosphere of 5% CO2 and 95% air.

Isoproterenol (ISO, I5627; Sigma-Aldrich, St. Louis, MO, USA) was first dissolved in water and diluted to 25 μM in culture medium. Dapagliflozin (DAPA, HY-10450; MedChemExpress, Monmouth Junction, NJ, USA), provided as a 10 mM stock solution in DMSO, was thawed and diluted to 10 μM. The AMPK agonist 5-aminoimidazole-4-carboxamide ribonucleotide (AICAR, HY-13417; MedChemExpress, Monmouth Junction, NJ, USA) was initially dissolved in water, followed by dilution to 500 μM. All pharmacological treatments were administered to H9c2 cells for 24 hours under standard culture conditions (37°C, 5% CO2).

Upon reaching 80–100% confluence, cells were seeded into six-well plates and allowed to adhere until 30–50% confluence was achieved. Transfected with 100 nM AMPKα2-specific siRNA (Tsingke, Shanghai, China) was performed using GenMute™ siRNA Transfection Reagent (SL100568; SignaGen, Shanghai, China) following the manufacturer's protocol. Forty-eight hours post-transfection, total protein was extracted and subjected to Western blot analysis to assess AMPKα2 knockdown efficiency.

H9c2 cells were seeded in 96-well plates at an optimal density of 3×103 cells/well in 100 μL culture medium. Following 24-hour ISO exposure, CCK-8 reagent (BS350B; Biosharp, AnHui, China) was added to each well at 10% of the total culture volume (10 μL/well), followed by incubation for 90 min at 37°C. Absorbance measurements were then performed at 450 nm using a multiskan skyhigh microplate reader (Thermo Fisher Scientific, Waltham, MA, USA).

The cultured H9c2 cells were fixed with 4% paraformaldehyde for 15 min, washed twice in PBS, and permeabilized with 0.05% Triton X-100 (9002-93-1; Solarbio, Beijing, China) for five minutes. They were subsequently incubated with 100 nM TRITC-phalloidin (CA1610; Solarbio, Beijing, China) in 1% BSA for 30 min, followed by three 5-min PBS washes. Nuclei were counterstained with 1 μg/mL Hoechst 33342 (C1028; Beyotime, Shanghai, China) for five minutes, then washed twice with PBS. Stained cells were captured with an inverted fluorescent microscope (ECLIPSE Ts2R, Shanghai, China). Cell surface area was quantified using ImageJ software (version 1.5.3; National Institutes of Health, Bethesda, USA) by analyzing ≥30 cells per condition.

For ultrastructural analysis, treated H9c2 cells were fixed in 2.5% glutaraldehyde (G1102; Servicebio, Wuhan, China) at 4°C overnight, then post-fixed with 1% osmium tetroxide (18456; Ted Pella, Redding, CA, USA) for 2 hours at room temperature (RT). After graded dehydration by ethanol series, the cells were infiltrated with EMBed 812 resin (90529-77-4; SPI, West Chester, PA, USA). Subsequently, the resin blocks were cut to 60–80 nm thin sections, and then double-stained with 2% uranium acetate (02624-AB; SPI, West Chester, PA, USA) and 2.6% lead citrate. Ultimately, the mitochondrial ultrastructure was examined and images were captured by TEM (HITACHI, H-7500, Japan).

H9c2 cells were seeded into six well plates at a density of 1×105 cells/well (30–50% confluence). Configure the JC-1 solution (C2006; Beyotime, Shanghai, China) according to the manufacturer's instructions, 1 mL of JC-1 working solution was added to H9c2 cells, which were then incubated in a 37°C incubator for 30 min. Following incubation, cells were gently rinsed twice with ice-cold JC-1 staining buffer. Finally, fluorescence images were acquired using a fluorescence microscope (ECLIPSE Ts2R, Shanghai, China). The change in the ratio of red fluorescence (J-aggregate) to green fluorescence (monomer) represented an alteration in the level of mitochondrial membrane potential.

H9c2 cells were seeded in confocal dishes and 6-well plates and cultured overnight. For autophagosome visualization, confocal dish-cultured cells in confocal dishes were transduced with mCherry-LC3 lentivirus particles (G120A; Hanbio, Shanghai, China), strictly adhering to manufacturer's 1/2 smal-volume infection instructions. After 48 hours, mCherry-LC3-expressing cells were loaded with 100 nM MitoTracker™ Green FM (C1048; Beyotime, Shanghai, China) to label mitochondria, whereas cells in 6-well plates received combinatorial staining with 100 nM MitoTracker™ Green FM and 75 nM LysoTracker Red (C1046; Beyotime, Shanghai, China) for mitochondrial and lysosome labeling for 15 min at 37°C. Nuclear counterstaining was performed with 10 μg/mL Hoechst 33342 (C1028, Beyotime, Shanghai, China) for five minutes at 37°C. Lastly, co-localization images of mitochondria with autophagosomes (mCherry-LC3 puncta) and lysosomes (LysoTracker Red) were acquired using confocal microscopy (Zeiss LSM, Baden-Württemberg, Germany) and fluorescence microscopy (ECLIPSE Ts2R, Shanghai, China), respectively.

Total RNA was isolated from H9c2 cells using the Trizol reagent (G3013; Servicebio, Wuhan, China). Subsequently, the extracted RNA was reverse-transcribed into complementary DNA (cDNA) using a qPCR reverse transcription kit (MF166; Mei5 Biotechnology, Beijing, China) equipped with a genomic DNA (gDNA) removal functionality. Finally, cDNA amplification was performed using the CFX96 Touch™ Real-Time PCR Detection System (Bio-Rad, Hercules, CA, USA) with HiPer SYBR Premix EsTaq (MF787; Mei5 Biotechnology, Beijing, China) under optimized cycling conditions. All qPCR assays were conducted in triplicate and normalized to the reference gene GAPDH. Relative gene expression levels were calculated using the 2^-ΔΔCT method. The gene-specific primer sequences used in this study were as follows:

ANP Forward: 5′-CCT GGA CTG GGG AAG TCA AC-3′; Reverse: 5′-GTC AAT CCT ACC CCC GAA GC-3′;

BNP Forward: 5′-ATT CCA AGA TGG CAC ATA GT-3′; Reverse: 5′-AGG AGG TCT TCC TAA AAC AAC-3′;

GAPDH Forward: 5′-GAG ACA GCC GCA TCT TCT TG-3′; Reverse: 5′-TGA CTG TGC CGT TGA ACT TG-3′.

Total cellular proteins were lysed from H9c2 cells using RIPA lysis buffer (AR0102; Boster, Wuhan, China) supplemented with complete protease and phosphatase inhibitors. Protein concentrations were quantified using a bicinchoninic acid (BCA) assay kit (AR0146; Boster, Wuhan, China). Following quantification, protein lysates were mixed with 5X protein loading buffer (AR1112; Boster, Wuhan, China) at a 4:1 ratio. After denaturation at 95°C for five minutes, 10 μg of proteins lysates were resolved on 8–12% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and subsequently electrotransferred onto a polyvinylidene fluoride (PVDF) membrane (ISEQ00010; Merck Millipore, Billerica, MA, USA). Membranes were blocked with rapid blocking buffer (G2052; Servicebio, Wuhan, China) for 15 min, then incubated overnight at 4°C with the following primary antibodies: anti-LC3B (CY5992; Abways, Shanghai, China), anti-P62/SQSTM1 (CY5546; Abways, Shanghai, China), anti-AMPKα1+AMPKα2 (ab207442; Abcam, Cambridge, MA, USA), anti-AMPKα1 (phospho T183)+AMPKα2 (phospho T172) (ab133448; Abcam, Cambridge, MA, USA), anti-AMPKα2 (ab97275; Abcam, Cambridge, MA, USA) and GAPDH (AB0037; Abways, Shanghai, China). After incubation with horseradish peroxidase (HRP)-conjugated secondary antibodies (BA1054; Boster, Wuhan, China) for 90 min at RT, protein bands were detected using enhanced chemiluminescence (ECL) substrate (MA0186; MeilunBio, Dalian, China) with one minute exposure. Eventually, protein band intensities were quantified by densitometric analysis using Image Lab™ software (version 6.1.0; Bio-Rad, Hercules, CA, USA).

All experiments were conducted independently at least three times. Data are presented as mean±standard deviation (SD) and were statistically analyzed using GraphPad Prism Software (version 9.0; GraphPad Software, San Diego, CA, USA). For comparisons among three or more groups, statistical significance was determined by one-way analysis of variance ith Tukey's honestly significant difference post hoc test. A probability value of p<0.05 was considered statistically significant for all analyses.

Cardiac hypertrophy represents a critical physiopathological mechanism underlying heart failure (HF) progression we therefore employed an in vitro cardiomyocyte hypertrophy model to investigate the pharmacological effects of dapagliflozin (DAPA) on HF. A concentration gradient of isoproterenol (ISO; 0–100 μM) was used to assess BNP mRNA expression, a key hypertrophic marker, in H9c2 cells. BNP mRNA expression began to increase at 25 μM ISO, with no futher significant changes observed up to 75 μM (Figure 1A). To further determine the optimal ISO concentration, we evaluated the H9c2 cell viability using the CCK-8 assays across the ISO concentration range. Relative to the untreated control group, the cell viability decreased significantly at 50 μM (Figure 1B). We therefore selected 25 μM ISO for subsequent experiments. The DAPA concentration (10 μM) was chosen based on previous work by Yang et al.12

Figure 1.

The effect of isoproterenol and dapagliflozin on hypertrophy in H9c2 cells. (A) Dose-dependent effects of isoproterenol (0–100 μM) on BNP mRNA expression in H9c2 cells. (B) CCK-8 analysis was used to detect the effect of isoproterenol (0–100 μM) on cell viability. Black arrow denotes the optimal isoproterenol concentration (25 μM) used for subsequent experiments. (C and D) mRNA expression levels of ANP and BNP were determined by quantitative real-time polymerase chain reaction. (E and F) TRITC-phalloidin staining was used to detect the cell surface area. Results were presented as mean±SD (n=3 for each group). *p<0.05, **p<0.01, ***p<0.001; ns, not significant.

To evaluate DAPA's effects on ISO-induced H9c2 hypertrophy cardiomyocytes, we assessed cell surface area and hypertrophic marker (ANP and BNP) expression. ISO treatment significantly increased both ANP and BNP expression (Figure 1C and D) and cell surface area (Figure 1E and F). DAPA alone showed no adverse effects, but co-treatment with ISO significantly attenuated ISO-induced hypertrophy in H9c2 cells (Figure 1C–F). These findings demonstrate that DAPA effectively attenuates ISO-induced cardiac hypertrophy.

Given the established role of mitochondrial dysfunction in HF progression, we further investigated DAPA's effects on ISO-induced mitochondrial damage. Transmission electron microscopy revealed that ISO-induced mitochondrial ultrastructural abnormalities in H9c2 cells, characterized by swelling, cristae reduction, shortening and focal vacuolization (Figure 2A). Conversely, the mitochondrial ultrastructure remained intact in the DAPA-treated group (Figure 2A). Furthermore, mitochondrial membrane potential (MMP) showed a notable reduction in the JC-1 red/green fluorescence ratio in the ISO-treated group, which was reversed back to normal levels in the DAPA-treated group, as evidenced by the normalized JC-1 red/green ratio (Figure 2B and C). These data collectively indicate DAPA's protective against ISO-induced mitochondrial dysfunction, preserving both structural and functional integrity.

Figure 2.

Dapagliflozin attenuated mitochondrial damage induced by isoproterenol in H9c2 cells. (A) Transmission electron microscopy was used to evaluate the effect of dapagliflozin on abnormal mitochondrial ultrastructure in isoproterenol-treated H9c2 cells. Single black arrow: normal mitochondria; double black arrows: damaged mitochondria. (B and C) Mitochondrial membrane potential was assessed using the JC-1 fluorescence. Results were presented as mean±SD (n=3 for each group). *p<0.05, ***p<0.001; ns, not significant.

Mitophagy serves as a crucial quality control mechanism that eliminates damaged mitochondria to maintain normal mitochondrial functions. To investigate DAPA's mitophagic protection against ISO-induced damage, we examined autophagy markers (LC3II/LC3I ratio and P62) by Western blotting across treatment groups. The LC3II/LC3I ratio was reduced and P62 levels were elevated in the ISO-treated group compared to normal control cells, suggesting that ISO inhibited autophagy (Figure 3A–C). In contrast, DAPA treatment restored autophagic flux, as illustrated by an elevated LC3II/LC3I ratio and reduced P62 levels (Figure 3A–C). To further validate DAPA's mitophagic regulation, we performed mCherry-LC3 transfection combined with MitoTracker™ Green staining in H9c2 cells. DAPA treatment increased autophagosome-mitochondria co-localization (yellow puncta), indicating initial mitophagosome formation (Figure 3D and E). Furthermore, increased mitochondria-lysosome co-localization (yellow fluorescence) confirmed subsequent degradation, provides strong evidence of DAPA's mitophagy induction (Figure 3F and G). In summary, all these findings demonstrate that DAPA activates mitophagy in ISO-stimulated H9c2 cells.

Figure 3.

Dapagliflozin promoted levels of mitophagy in isoproterenol-induced H9c2 cells. (A) Representative Western blots of LC3B and P62 in H9c2 cells. (B and C) Quantitative analysis of the expression levels of LC3B and P62. (D and E) The co-localization of autophagosomes with mitochondria was detected using confocal microscope. (F and G) Mitochondria-lysosome co-localization was assessed via fluorescence microscopy. Results were presented as mean±SD (n=3 for each group). *p<0.05, ***p<0.001; ns, not significant.

We next explored DAPA's mechanism in promoting mitophagy in ISO-treated H9c2 cells. AMP-activated protein kinase (AMPK) activation regulates various cellular processes, including mitophagy. We therefore assessed AMPKα phosphorylation by Western blotting. The p-AMPKα/AMPKα ratios were decreased in ISO-treated H9c2 cells but reversed by DAPA treatment (Figure 4A and B). Meanwhile, total AMPKα levels remained unchanged (Figure 4A and C). These results demonstrate that DAPA enhances AMPKα phosphorylation without altering its expression. We further examined the expression level of AMPKα2, the predominant cardiac AMPKα isoform. Western blotting results revealed that DAPA upregulated AMPKα2 in cells treated with ISO (Figure 4D and E). Collectively, DAPA activated AMPKα and upregulated AMPKα2 in ISO-treated H9c2 cells.

Figure 4.

Dapagliflozin enhanced AMPKα phosphorylation and AMPKα2 expression. (A) Representative Western blots of p-AMPKα and AMPKα. (B and C) Quantitative analysis of the expression levels of p-AMPKα and AMPKα. (D) Representative Western blots of AMPKα2. (E) Quantitative analysis of the expression levels of AMPKα2. Results were presented as mean±SD (n=3 for each group). *p<0.05, **p<0.01, ***p<0.001; ns, not significant.

To investigate the role of AMPKα in DAPA's cardioprotection against ISO-induced injury, we used 5-aminoimidazole-4-carboxamide ribonucleotide (AICAR; 500 μM), an established AMPK agonist known for its cardioprotective effects in H9c2 cells.13 Similar to DAPA, AICAR treatment significantly attenuated ISO-induced hypertrophic responses, as evidenced by reduced ANP and BNP expression (Figure 5A and B) and normalized cell surface area (Figure 5C and D). Furthermore, AICAR preserved MMP, counteracting ISO-induced depolarization (Figure 5E and F). AICAR treatment also recapitulated DAPA-mediated restoration of autophagic flux, as evidenced by normalized LC3II conversion and p62 degradation (Figure 6A–C). Collectively, these findings demonstrate that DAPA attenuates ISO-induced cardiomyocyte hypertrophy and mitochondrial dysfunction through AMPKα activation.

Figure 5.

The AMPK agonist 5-aminoimidazole-4-carboxamide ribonucleotide conferred cardioprotection that paralleled the therapeutic efficacy of dapagliflozin. (A and B) Hypertrophic markers ANP and BNP mRNA levels were quantified using quantitative real-time polymerase chain reaction. (C and D) Cardiomyocyte hypertrophy was assessed via TRITC-phalloidin staining of H9c2 cell surface area. (E and F) Mitochondrial membrane potential was evaluated using JC-1 fluorescence. Results were presented as mean±SD (n=3 for each group). *p<0.05, **p<0.01, ***p<0.001; ns, not significant.

Figure 6.

The AMPK agonist 5-aminoimidazole-4-carboxamide ribonucleotide conferred cardioprotection that paralleled the therapeutic efficacy of dapagliflozin. (A) Representative Western blots of LC3B and P62. (B and C) Quantitative analysis of the expression levels of LC3B and P62. Results were presented as mean±SD (n=3 for each group). *p<0.05, **p<0.01, ***p<0.001; ns, not significant.

To establish AMPKα2-dependence of DAPA's effects, we knocked down AMPKα2 expression in H9c2 cells using AMPKα2 small interfering RNA (siRNA). Western blotting revealed a marked decrease in AMPKα2 levels in H9c2 cells transfected with AMPKα2 siRNA (Figure 8A and B). In comparison to the DAPA-treated group, co-treatment with AMPKα2 siRNA abolished DAPA's anti-hypertrophic effects in ISO-stimulated cells (Figure 7A–D). Meanwhile, DAPA's ameliorative effect on MMP was attenuated following AMPKα2 siRNA transfection (Figure 7E and F). Furthermore, DAPA-induced LC3 conversion and P62 degradation were both impaired by AMPKα2 deficiency (Figure 8C–E). Collectively, these results suggest that DAPA attenuates cardiac hypertrophy, promotes mitophagy and protectes mitochondrial integrity partially via the AMPKα2 pathway in ISO-stimulated cells.

Figure 7.

AMPKα2 interference abrogated the protective effects of dapagliflozin. (A and B) The mRNA expression levels of ANP and BNP after transfection with AMPKα2 siRNA were determined by quantitative real-time polymerase chain reaction. (C and D) Cell surface area alterations in AMPKα2 siRNA-transfected H9c2 cells were detected using TRITC-phalloidin staining. (E and F) Changes in mitochondrial membrane potential following AMPKα2 siRNA interference. Results were presented as mean±SD (n=3 for each group). ***p<0.001; ns, not significant.

Figure 8.

AMPKα2 knockdown attenuated the promotion of mitophagy by dapagliflozin. (A) Representative Western blots of AMPKα2. (B) Quantitative analysis of the expression levels of AMPKα2. (C) Representative Western blots of LC3B and P62. (D and E) Quantitative analysis of the expression levels of LC3B and P62. Results were presented as mean±SD (n=3 for each group). *p<0.05, **p<0.01, ***p<0.001; ns, not significant.