This article is mentioned in:

Introduction

Despite advances in the treatment of coronary artery disease, acute myocardial infarction (AMI) remains the leading cause of human mortality worldwide (1). AMI is characterized by a sudden reduction in blood and oxygen supply to the heart, irreversible muscle damage and cardiomyocyte death, resulting in the formation of an infarct zone containing nonfunctional myocytes, which are remodeled into scar tissue (2). The limited ability of the damaged heart to regenerate the damaged myocardium leads to the progression of cardiac decompensation and heart failure (3). AMI causes the death of a majority of cardiomyocytes which are replaced by fibrotic tissue, leading to cardiac malfunction. Unlike myocardial muscle, fibrotic tissue fails to contract and therefore is unable to pump blood to the whole body (4,5). Therefore, a reduction in fibrosis and cardiomyocyte death is necessary to repair the infarcted heart following AMI (6). Additionally, the death of cardiomyocytes results in fibrous hyperplasia. Therapeutic strategies for abating cardiomyocyte death are necessary to delay or prevent the onset of fibrosis and heart failure following AMI.

Recent studies have identified a linear association between mitochondrial fate and cardiac dysfunction (7,8). Mitochondria may be associated with cardiomyocyte survival and, therefore, their contribution has a marked impact on the repair of the infarcted heart post-AMI (9,10). A growing body of evidence suggests that cardiomyocytes require well-structured mitochondria to sustain the cardiac function in the post-infarcted heart; this process requires well-orchestrated mitophagy to maintain mitochondrial quantity and quality (11). However, an imbalance in the mitochondrial degradation process results in alterations in cellular homeostasis which, in turn, may aggravate cardiomyocyte death and the development of cardiac dysfunction (12,13). These previous studies indicated that mitophagy may be a potential target for repairing the infarcted heart following AMI. In yeast and mammals, three receptors have been identified to activate mitophagy: Parkin, FUN14 domain-containing protein 1 (FUNDC1) and BCL2/adenovirus E1B 19 kDa protein-interacting protein 3 (Bnip3). Previous studies have established the role of FUNDC1 and Bnip3 in cardiac injury (14,15). However, the role of Parkin in chronic cardiac damage remains unclear. Parkin is activated by serine/threonine-protein kinase PINK1 and primarily exerts its influence in neurological disorders (16). Parkin is able to interact with damaged mitochondria and traffic them to the lysosome (17). These previous data illustrate that Parkin-mediated mitophagy is important for the cellular regulation of steady-state mitochondrial turnover, and determines the number of healthy mitochondria. Accordingly, the identification of a novel drug to activate Parkin-mediated mitophagy may provide increased beneficial effects for patients following AMI.

Liraglutide is an antidiabetic agent which reduces hyperglycemia through increased glucose-dependent insulin secretion, glucagon suppression, delayed gastric emptying and appetite suppression (18–21). It has been demonstrated that liraglutide is important for cardioprotection, including reducing infarct size, improving the left ventricular ejection fraction and reversing cardiac remodeling (22–24). However, whether liraglutide protects against cardiomyocyte death and repairs the post-infarction heart via mitophagy remains unclear, in addition to the potential molecular links between liraglutide and mitophagy.

NAD-dependent protein deacetylase sirtuin 1 (SIRT1) is one of the seven mammalian homologs (SIRT1-SIRT7) of the yeast silent information regulator 2 (25). SIRT1 is an NAD+-dependent protein deacetylase which has been reported to serve a number of roles in cells, including longevity, apoptosis, DNA repair, inflammation and mitochondrial regulation (26–28). The regulatory effect of SIRT1 on the activation of mitophagy has gained attention, although the underlying mechanisms have not been completely elucidated. Notably, a previous study demonstrated that liraglutide was able to activate SIRT1 to trigger autophagy, ameliorating non-alcoholic fatty liver disease (29). These previous data suggest a potential association between liraglutide and SIRT1. Therefore, the present study posited and analyzed the following three hypotheses: i) Whether liraglutide has the ability to repair or improve the infarcted heart following AMI; ii) whether Parkin-mediated mitophagy is responsible for the protective action of liraglutide on the injured heart; and iii) whether liraglutide regulates mitophagy via SIRT1 in cardiomyocytes.

Materials and methods

Ethics statement

The present study was performed in accordance with the Declaration of Helsinki and the guidelines of the Ethics Committee of No. 1 People's Hospital (Suzhou, China). All experimental protocols were approved by the Ethics Committee of No.1 People's Hospital. The experiments on animals were performed in accordance with the Guide to the Care and Use of Experimental Animals (Vol. 1, 2nd ed., 1993; and Vol. 2, 1984; Canadian Council on Animal Care, Ottawa, ON, Canada; www.ccac.ca), or the Guide for the Care and Use of Laboratory Animals (1996; National Academy Press, Washington, DC, USA).

Myocardial infarction model

Male Sprague-Dawley rats (n=150, 8–10 weeks, weight, 250±10 g) were purchased from the Experimental Animal Center, Academy of Military Medical Science and were housed under standard laboratory conditions (27°C, 40–60% humidity, a 12-h light and dark cycle). A commercial pellet diet and fresh drinking water were given ad libitum. Rats were randomly divided into the following groups, with n=6 in each: i) Sham group; ii) PBS; iii) low dose of liraglutide; and iv) high dose of liraglutide. Rats were intraperitoneally anaesthetized with sodium pentobarbital (30 mg/kg). The animals were subsequently incubated and ventilated with a volume-regulated respirator during surgery. Subsequent to a left lateral thoracotomy and pericardectomy, the left coronary artery was identified and ligated with a 6.0 prolene suture. Freshly prepared liraglutide (Sigma-Aldrich; Merck KGaA, Darmstadt, Germany) was administered via the caudal vein at doses of 0.09 mg/kg (low dose group) or 0.18 mg/kg (high dose group) for 28 successive days post-infarction. These concentrations of drugs were selected as they were previously reported to be effective in reducing cardiac damage (10).

Sample preparation and histological analysis

The hearts were excised and rapidly frozen in Optimal Cutting Temperature medium at −20°C (Agar Scientific, Ltd., Stansted, UK) for the preparation of frozen sections (4-µm thickness). A total of 10 sections were prepared at 10 different transversal levels at the site of tissue necrosis, equally distributed from base to apex. Sirius red staining was performed at room temperature in sections to quantify the cardiac fibrosis and was observed using an inverted microscope (magnification, ×40; BX51; Olympus Corp., Tokyo, Japan). The degrees of collagen fiber accumulation in the infarcted area were evaluated by measuring the fluorescence density of fibrotic region in the left ventricular area, which was calculated using RS Image Pro, version 4.5 (Media Cybernetics, Inc., Rockville, MD, USA). Terminal deoxynucleotidyl-transferase-mediated dUTP nick end labeling (TUNEL) staining was performed to assess cellular apoptosis according to the protocol described below. The degree of apoptosis was calculated as the number of TUNEL-positive cells per 500 cellular nuclei. The nuclei were stained with the chromatin dye DAPI. Briefly, cells were fixed for 1 h in 4% (w/v) paraformaldehyde at room temperature. Following labeling with TUNEL for ~30 min at room temperature, the cells were exposed to DAPI in the dark for 5 min. Then, the stained cells were observed using an inverted microscope (magnification, ×100; BX51; Olympus Corp.). In order to determine the inflammatory response, heart sections were stained with anti-tumor necrosis factor (TNF) α (1:1,000; Cell Signaling Technology, Inc., Danvers, MA, USA; cat. no. 6945). In brief, samples were fixed with 4% paraformaldehyde for 10 min at room temperature, permeabilized with 0.3% Triton X-100 for 5 min, and blocked with 10% goat serum albumin (Invitrogen; Thermo Fisher Scientific, Inc., Waltham, MA, USA) for 1 h at room temperature. Specimens were subsequently incubated with transforming growth factor (TGF) β (1:1,000; Cell Signaling Technology, Inc.; cat. no. 3711) overnight at 4°C, then washed with PBS three times, and incubated with horseradish peroxidase AffiniPure Goat Anti-Rabbit secondary antibody (1:2,000; Beyotime Institute of Biotechnology, Haimen, China; cat. no. A0277) for 45 min at room temperature. Images were captured using an inverted microscope (magnification, ×40; BX51; Olympus Corp.).

Chronic hypoxia model in vitro

In vitro, chronic hypoxia of cardiomyocytes was used to mimic the infarcted heart following AMI. Hypoxic conditions were produced using fresh Hank's solution (Beyotime Institute of Biotechnology) with 95% N2 and 5% CO2. The dishes (1×106 cells) were placed into a hypoxic incubator at 37°C, that was equilibrated with 95% N2 and 5% CO2 and the actual oxygen concentration was 0. Ambient O2 levels in the hypoxic incubator were monitored with an O2 analyzer (series-2000; Alpha Omega Instruments, Lincoln, RI, USA). Chronic hypoxia of cardiomyocytes cell line H9C2 (purchased from the American Type Culture Collection, Manassas, VA, USA) was performed for ~48 h. For liraglutide treatment, H9C2 were treated with liraglutide (0–50 nM) for 48 h in the presence of hypoxia. To activate SIRT1, pretreatment with SRT1720 (SRT; 10 µM; Sigma-Aldrich; Merck KGaA) was performed for ~4 h. To inhibit the SIRT1, Selisistat (10 µM; Sigma-Aldrich; Merck KGaA) was used for 6 h. To suppress the role of liraglutide, Exendin 9–39 (Ex9-39; 10 nM; Sigma-Aldrich; Merck KGaA) was used to block the glucagon-like peptide 1 receptor (GLP1R) under hypoxic conditions.

Immunofluorescence staining

The 1×106 cells were seeded into the dishes then were washed in PBS and permeabilized for 10 min at 4°C in a solution of 0.1% Triton X-100 and 0.1% sodium citrate in PBS. Then, samples were blocked with 10% goat serum albumin for 1 h at room temperature and subsequently were incubated with primary antibodies against overnight at 4°C (30). Following three rinses with PBS, the secondary antibody, Alexa Fluor 488 donkey anti-rabbit antibody (1:1,000; cat. no. A-21,206; Invitrogen; Thermo Fisher Scientific, Inc.) was added to the samples for 1 h at room temperature. The primary antibodies used in the present study were as follows: Mitochondrial import receptor subunit TOM20 homolog (1:1,000; Abcam, Cambridge, UK; cat. no. ab78547), lysosome-associated membrane glycoprotein 1 (1:1,000; Abcam; cat. no. ab24170), cytochrome-c (1:1,000; cyt-c; Abcam; cat. no. ab133504), Parkin (1:1,000; Cell Signaling Technology, Inc.; cat. no. 2132) and Sirt1 (1:1,000; Cell Signaling Technology, Inc.; cat. no. 9475). Images were taken using an inverted microscope (magnification, ×40; BX51; Olympus Corp.).

Western blotting

Cells were washed with PBS and lysed in Laemmli Sample Buffer (Bio-Rad Laboratories, Inc., Hercules, CA, USA), and further homogenized with a rotor-stator homogenizer. Proteins were isolated and concentrations were determined using the Bicinchoninic Acid Protein Assay kit (Thermo Fisher Scientific, Inc.) (31). A mass of 20–80 µg proteins was loaded on a 12–15% SDS-PAGE gel. Following electrophoresis, proteins were transferred to a polyvinylidene fluoride western blotting membrane (Roche Applied Science, Penzberg, Germany). Membranes were blocked with 5% nonfat dried milk [in TBS-Tween 20 (TBST)] for 2 h at room temperature and incubated overnight at 4°C with primary antibodies. The membrane was subsequently washed with TBST (5 min; three times) and incubated with horseradish peroxidase-conjugated secondary antibodies (Cell Signaling Technology, Inc.) for 1 h at room temperature. Following washing with TBST (5 min; three times), bands were detected using an enhanced chemiluminescence substrate (Applygen Technologies, Inc., Beijing, China). Band intensities were normalized to the respective internal standard signal intensity (β-actin antibody; 1:2,000; Abcam; cat. no. ab8224). The experiment was repeated three times. The primaries antibodies used in the present study were against the following proteins: Pro-caspase3 (1:1,000; Cell Signaling Technology, Inc.; cat. no. 9662), cleaved caspase3 (1:1,000; Cell Signaling Technology, Inc.; cat. no. 9664), baculoviral IAP repeat-containing protein 2 (c-IAP1; 1:2,000; Cell Signaling Technology, Inc. cat. no. 7065), caspase9 (1:1,000; Abcam; cat. no. ab32539), Parkin (1:1,000; Cell Signaling Technology, Inc.; cat. no. 2132), microtubule-associated protein light chain (LC) 3II (1:1,000; Cell Signaling Technology, Inc.; cat. no. 3868), sequestome-1 (p62; 1:1,000; Abcam; cat. no. ab56416), Beclin1 (1:1,000; Cell Signaling Technology, Inc.; cat. no. 3495), autophagy protein 5 (Atg5; 1:1,000; Cell Signaling Technology, Inc.; cat. no. 12,994), Sirt1 (1:1,000; Cell Signaling Technology, Inc., cat. no. 9475), TGFβ (1:1,000; Cell Signaling Technology, Inc.; cat. no. 3711), matrix metalloproteinase (MMP) 9 (1:1,000; Cell Signaling Technology, Inc.; cat. no. 13667) and poly (ADP ribose) polymerase 1 (PARP; 1:1,000; Cell Signaling Technology, Inc.; cat. no. 9532) The mean densities of the bands were represented as the optical density in units/mm2 and normalized to that of β-actin (Quantity One, version 4.6.2; Bio-Rad Laboratories, Inc.).

Reactive oxygen species (ROS) detection, JC-1 staining and adenosine 5′-triphosphate (ATP) detection

Cells were incubated in serum-free Dulbecco's modified Eagle's medium (DMEM; Gibco; Thermo Fisher Scientific, Inc.) containing dichloro-dihydro-fluorescein diacetate (10 µM) for 30 min (32). Cells were subsequently washed with PBS. Flow cytometric analyses were performed using a BD FACSCalibur™ flow cytometer (BD Biosciences, Franklin Lakes, NJ, USA) using BD CellQuest Pro software (version 5.1; BD Biosciences).

The mitochondrial potential was assessed using the probe JC-1, a sensitive fluorescent dye used to detect alterations in mitochondrial potential. Following treatment, cells were incubated with 10 mg/ml JC-1 for 10 min at 37°C in the dark and monitored with a fluorescence microscope (magnification, ×100; BX51; Olympus Corp.) (33). Red-orange fluorescence was attributable to a potential-dependent aggregation in the mitochondria. Green fluorescence, reflecting the monomeric form of JC-1, appeared in the cytosol following mitochondrial membrane depolarization.

The level of ATP in cells was determined using an ATP Bioluminescence Assay kit (Beyotime Institute of Biotechnology) Harvested cultured cells (1×106) were lysed with a PBS (Sigma-Aldrich; Merck KGaA), followed by centrifugation at 10,000 × g for 2 min at 4°C. The level of ATP was determined by mixing 50 µl supernatant with 50 µl luciferase reagent, which catalyzed the light production from ATP and luciferin. The emitted light was linearly compared with the ATP concentration and measured using a microplate luminometer.

MTT and TUNEL assays

MTT experiments were performed in 96-well plates. The cells were applied to the scaffold at a density of 104 cells/well. Following 2–3 days of culture, the samples were washed 3 times with PBS, and 50 µl MTT was added to each well. The samples were subsequently incubated for 4 h at 37°C in a humid atmosphere containing 5% CO2. The MTT solution was removed and 200 µl dimethyl sulfoxide was added to each sample and incubated for 10 min. Following the addition of Sorensen's buffer, the absorbance was determined at a wavelength of 570 nm (34).

A TUNEL assay to detect DNA fragmentation in the cell nuclei was performed using an In Situ Cell Death Detection kit (Roche Diagnostics GmbH, Mannheim, Germany) (35), according to the manufacturer's protocol. DAPI was used to label the nuclei (at room temperature for ~30 min). The results are expressed as apoptotic cells per tubule cross-section.

ELISA analysis

Cells (1×106) in the treatment groups were lysed, centrifuged for 10 min at 1,600 × g at 4°C, and the supernatant (10 µl) was mixed with the pre-mixed reagent (according to the manufacturer's protocol) using a vortex mixer. This mixture was placed in a thermostatic water bath for 20 min at 37°C. The mixture was incubated with the color-developing agent for 10 min at room temperature. The absorbance value was determined at 560 nm using a microplate reader. The superoxide dismutase (SOD), glutathione (GSH) and malondialdehyde (MDA) levels were calculated using the standard formula, according to the protocols of the commercially-available ELISA kits (Beyotime Institute of Biotechnology, Haimen, China; cat. nos. S0086, S0131 and S0058, respectively).

RNA isolation and reverse transcription-quantitative polymerase chain reaction (RT-qPCR) analysis

Total RNA was isolated from cells using TRIzol™ reagent (Invitrogen; Thermo Fisher Scientific, Inc.). A total of 1 µg RNA from each sample was reverse-transcribed into cDNA using an RT kit (Eurogentec, Liege, Belgium) (36). The qPCR was performed with primers and matched probes from the Universal Fluorescence-labeled Probe Library (Roche Diagnostics GmbH). Quantification of gene expression was performed using an ABI PRISM 7500 Sequence Detection system (Applied Biosystems; Thermo Fisher Scientific, Inc.) with SYBR® Green (Beiijing Transgen Biotech Co., Ltd., Beijing, China). The relative mRNA expression levels were normalized to that of β-actin using the 2−ΔΔCq method (12). The thermocycling conditions were as follows: 95°C for 10 min, followed by 40 cycles of 95°C for 15 sec and 72°C for 35 sec, for PCR. The experiments were repeated three times with triplicates. The primers used in the present study were as follows: Interleukin (IL) 6, forward primer 5′-TCACCGATGTCTACCTGCTG-3′, reverse primer 5′-CACAGGGTTGAGCCAAAAGT-3′, C-C motif chemokine 2 (MCP1), forward primer 5′-CTCAACATCATGAAGGTCTC-3′, reverse primer 5′-GGCATTCAGTTCCAGGTCAG-3′, and GAPDH, forward primer 5′-ACGACATAGACGGCATCCA-3′, reverse primer 5′-GCTGTGGTTCAGTTGTGGTG-3′.

Caspase3/9 activity and lactate dehydrogenase (LDH) release assay

Caspase-3/9 activity kits (Beyotime Institute of Biotechnology) were used, according to the manufacturer's protocols (34). The relative caspase-3/9 activity was calculated from the ratio of treated cells to untreated cells (1×106). The assays were repeated three times.

LDH is a fairly stable enzyme that is released from the cytosol into the culture medium as a consequence of cellular integrity damage. Thus, we used an LDH assay (Beyotime Institute of Biotechnology) to evaluate the presence of cell injury or damage. The level of LDH released was expressed as a percentage of the control group.

Small RNA interference assay

In order to evaluate the functional role of Parkin, small interfering (si)RNA was used to reduce its expression. The selective siRNA duplex (5′-GAGGAUGACAACGACAUAATT-3′, antisense, 5′-UUAUGUCGUUGUCAUCCUCTT-3′) and a nonspecific control duplex (5′-UUCUCCGAACGUGUCACGUTT-3′; antisense, 5′-ACGUGACACGUUCGGAGAATT-3′). These were obtained from YangZhou Ruibo Biotech Co., Ltd. (Yangzhou, China). Transfection of siRNA into cells was performed using Lipofectamine® 2000 transfection reagent (Thermo Fisher Scientific, Inc.), according to the manufacturer's protocol (37). Cultured cells were washed with Opti-Minimal Essential Medium (Invitrogen; Thermo Fisher Scientific, Inc.) without serum or antibiotics and seeded in 6-well plates to 30–40% confluence. The transfection reagent and siRNA were diluted separately in serum-free medium, mixed and incubated for 10 min at room temperature to form the siRNA/lipid complex. This complex was added to each well at a final concentration of 70 nM/well siRNA. At 48 h post-transfection, cells were harvested to determine Parkin protein expression levels by western blot analysis according to the protocol described above.

Statistical analysis

All data are expressed as the mean ± standard deviation. Statistical analyses were performed with SPSS software (version 17.0; SPSS, Inc., Chicago, IL, USA). Statistical significance between two groups was determined by Student's t-test. Results for more than two groups were evaluated by one-way analysis of variance with the least significant difference test. P<0.05 was considered to indicate a statistically significant difference.

Results

Liraglutide reduces fibrosis, inflammatory responses and cardiomyocyte apoptosis in the post-infarcted heart

To examine the protective role of liraglutide in repairing the infarcted heart, fibrosis and cardiomyocytes death were assessed. Compared with the control group, the post-infarcted heart (28 days following myocardial infarction) exhibited extensive fibrosis, as evidenced by Sirius red staining (Fig. 1A and B). However, liraglutide had the ability to reduce cardiac fibrosis in a dose-dependent manner. To provide further evidence for the role of liraglutide in cardiac fibrosis in the post-infarcted heart, the expression of TGFβ and MMP9 was examined via western blotting. TGFβ and MMP9 are key feature of fibrosis. Compared with the control group, the post-infarcted heart exhibited more TGFβ and MMP9 expression (Fig. 1C-E). However, these phenotypic alterations were rescued by liraglutide in a concentration-dependent manner.

Figure 1. Liraglutide repairs the infarcted heart. Chronic heart damage was induced via myocardial infarction. Freshly prepared liraglutide was administered via the caudal vein at doses of 0.09 mg/kg (low-dose group) or 0.18 mg/kg (high-dose group) for 28 successive days post-infarction. (A) Cardiac fibrosis was evaluated via Sirius red staining and (B) the results were quantified. Liraglutide was applied at high (0.18 mg/kg) and low (0.09 mg/kg) dosages. (C) Western blotting was used to evaluate the expression of (D) TGFβ and (E) MMP9, which are key feature of cardiac fibrosis following myocardial infarction. (F) The TUNEL assay was used to evaluate the apoptotic rate of heart 28 days after post-infarction, and (G) the results were quantified. (H-J) The alteration in apoptotic protein expression was assessed via (H) western blotting in the heart tissue at 28 days post-infarction, and the expression levels of (I) caspase3 and (J) PARP were analyzed by densitometry. (K) Immunohistochemical analysis of TNFα was applied to evaluate the cardiac inflammatory response. The alteration in (L) IL-6 and (M) MCP1 mRNA expression was assessed. Magnification, ×40. *P<0.05 vs. sham group; #P<0.05 vs. PBS group. Lir, liraglutide; TUNEL, terminal deoxynucleotidyl-transferase-mediated dUTP nick end labeling; TGFβ, transforming growth factor-β; MMP9, matrix metalloproteinase 9; Cle, cleaved; PARP, poly (ADP ribose) phosphate polymerase 1; IL-6, interleukin-6; MCP1, C-C motif chemokine 2; TNFα, tumor necrosis factor-α.

Since cardiomyocyte death is the primary reason for the development of cardiac fibrosis and heart failure following myocardial infarction, a TUNEL assay was used to observe cardiomyocyte apoptosis. As presented in Fig. 1F and G, the post-infarcted heart had an increased number of TUNEL-positive cells. However, treatment with liraglutide was able to reduce the ratio of TUNEL-positive cells. In order to provide further evidence for cellular death during cardiac remodeling, apoptotic protein expression was detected. Compared with the control group, caspase3 and its primary cleavage target PARP were increased in the post-infarcted heart (Fig. 1H-J), indicative of myocardial death in the post-infarcted heart. However, these alterations were reversed by treatment with liraglutide.

Apart from fibrosis and cellular death, the present study additionally examined the inflammatory response which is associated with cardiomyocyte death and fibrosis accumulation. Compared with the control group, increased expression of TNFα was observed in the myocardial tissue in the post-infarcted heart (Fig. 1K). Additionally, the transcriptional levels of IL6 and MCP1 were increased in the post-infarcted heart (Fig. 1L-M). However, treatment with liraglutide was able to repress the expression of the inflammatory markers. These data suggested that liraglutide had the ability to repair the damaged heart following myocardial infarction.

Liraglutide sustains H9C2 survival by abating mitochondrial apoptosis

To gain an insight into the mechanism through which liraglutide maintained cardiomyocyte viability, a chronic hypoxia model with H9C2 cells was used to mimic the infarcted heart, and the molecular signaling responsible for the anti-apoptotic properties of liraglutide was examined (Fig. 2). Following 48 h of hypoxia stimulation, H9C2 apoptosis was detected via MTT and LDH release assays. Chronic hypoxia reduced cell viability, as evidenced by a reduced MTT value (Fig. 2A) and increased LDH release (Fig. 2B). However, treatment with liraglutide was able to inhibit the apoptotic signaling in a dose-dependent manner. Considering that the minimum protective concentration of liraglutide was 5 nM, this concentration was used in the following experiments. To further quantify H9C2 apoptosis under chronic hypoxia, a TUNEL assay was used. It was observed that hypoxia increased the number of TUNEL-positive cells (Fig. 2C and D), which was inhibited by liraglutide. Since liraglutide, a type of GLP1R agonist, exerts protective action on cells via GLP1R, Ex9-39 was used to block the GLP1R. Following the blockade of GLP1R, the protective effect of GLP1 on H9C2 apoptosis disappeared (Fig. 2C and D). In order to investigate whether mitochondrial apoptosis was involved in H9C2 apoptosis, cyt-c was assayed. In response to mitochondrial apoptosis, cyt-c was released from the mitochondria into the cytoplasm, where it interacts with caspase9 and activates caspase3. Therefore, immunofluorescence was used to observe the cyt-c cellular location. In the normal cells, cyt-c was observed in the cellular cytoplasm (Fig. 2G). However, hypoxia induced the diffusion of cyt-c into the cytoplasm and nucleus (Fig. 2G), indicating the activation of mitochondrial apoptosis. By contrast, liraglutide had the ability to reverse such alterations. Notably, the GLP1R blocker was able to reverse the inhibitory action of liraglutide on cyt-c diffusion.

Figure 2. Liraglutide reduces cardiomyocyte death via inhibition of mitochondrial apoptosis. The hypoxia model of H9C2 in vitro was used to mimic chronic cardiac injury. H9C2 cells were cultured under hypoxic condition for ~48 h. (A) MTT and (B) LDH assays were used to detect the cell viability. Liraglutide was able to sustain H9C2 viability in a dose-dependent manner under 48 h of hypoxia. *P<0.05 vs. Ctrl group; #P<0.05 vs. Lir (0 nM) group. Since the minimum concentration of liraglutide, which significantly promoted H9C2 viability under hypoxia was 5 nM, this concentration was used in the following experiments. (C) A TUNEL assay was used to observe the cellular apoptosis and (D) representative images are presented. Ex9-39, a water-soluble GLP-1 receptor antagonist was used to inhibit the action of liraglutide. *P<0.05 vs. Ctrl group; #P<0.05 vs. hypoxia group; @P<0.05 vs. hypoxia + Lir group. (E) Western blotting was used to detect alterations in the expression of proteins associated with mitochondrial apoptosis. (F) The protein expression of Caspase3 was quantified. (G) Immunofluorescence analysis of cyt-c localization. Chronic hypoxia induced the cyt-c leakage from mitochondria and trafficking into the nucleus. The protein expression levels of (H) caspase9, (I) Bax and (J) c-IAP1 were quantified. Liraglutide was able to reduce the expression of apoptotic proteins, and this effect was reversed by Ex9-39. *P<0.05 vs. Ctrl group; #P<0.05 vs. hypoxia group; @P<0.05 vs. hypoxia + Lir group. Lir, liraglutide; Ctrl, control; cle, cleaved; LDH, lactate dehydrogenase; TUNEL, terminal deoxynucleotidyl-transferas-mediated dUTP nick end labeling; Ex9-39, exendin 9–39; Bax, apoptosis regulator BAX; c-IAP1, baculoviral IAP repeat-containing protein 2; cyt-c, cytochrome c.

To provide further evidence for mitochondrial apoptosis, alterations in the expression of proteins associated with cyt-c leakage were assessed. As presented in Fig. 2F and H-J, hypoxia elevated the expression of caspase3, caspase9 and Bax, and reduced the expression of c-IAP1. However, liraglutide was able to promote anti-apoptotic protein expression and limit pro-apoptotic protein expression (Fig. 2F and H-J). Notably, the anti-apoptotic effect of liraglutide was mediated by GLP1R. These data indicated that liraglutide sustained H9C2 survival in the context of chronic hypoxia damage by abating mitochondrial apoptosis.

Mitophagy is activated by liraglutide via Parkin

As mitophagy is important for the maintenance of mitochondrial homeostasis, it was hypothesized that liraglutide may preserve mitochondrial balance via mitophagy (38). To validate this hypothesis, the expression of mitophagy markers was examined. Compared with the control group, hypoxia induced a downregulation of mitophagy, as evidenced by decreased LC3II, p62, Beclin1 and Atg5 expression (Fig. 3). However, liraglutide was able to reverse the expression of mitophagy parameters, and this effect was nullified by Ex9-39 (Fig. 3A, B and D-G). Notably, as Parkin is an important receptor in mitophagy, alterations in Parkin expression were examined. It was observed that hypoxia progressively reduced the expression of Parkin (Fig. 3A and B), which was rescued by treatment with liraglutide. However, with the blockade of GLP1R, Parkin expression was decreased (Fig. 3A and B). These data indicated that liraglutide triggered Parkin-associated mitophagy under hypoxia.

Figure 3. Liraglutide activates mitophagy. (A) Western blotting was used to measure the alterations in the expression of mitophagy-associated proteins. (B) Parkin expression was quantified. (C) Co-staining of mitochondria and lysosomes was performed. In the enlarged panels, orange immunofluorescence indicates mitophagy. Chronic hypoxic stimulation reduced the number of foci of mitochondria and lysosomes. However, liraglutide was able to increase the evidence of mitophagy, and this effect was nullified by Ex9-39. (D) LC3II, (E) p62, (F) Atg5 and (G) Beclin-1 expression levels were quantified. *P<0.05 vs. Ctrl group; #P<0.05 vs. hypoxia group; @P<0.05 vs. hypoxia + Lir group. LC, microtubule-associated protein light chain; p62, sequestome-1; Atg5, autophagy protein 5; Ctrl, control; Lir, liraglutide; Ex9-39, exendin 9–39; mito, mitochondrial.

Additionally, to provide direct evidence for the role of liraglutide in the activation of mitophagy, immunofluorescence analysis was used to observe the overlap of mitochondria and lysosomes. As presented in Fig. 3C, in the control group, a number of mitochondria were tagged with lysosomes. However, following chronic hypoxic stimulation, the majority of mitochondria were separated from lysosomes (Fig. 3C), indicative of mitophagy inhibition. However, treatment with liraglutide was able to reverse the interaction between mitochondria and lysosomes, and contributed to the overlap of mitochondria and lysosomes. However, with blockade of GLP1R, mitophagy was inhibited despite treatment with liraglutide (Fig. 3C). These data demonstrated that liraglutide had the ability to reverse mitophagy activity under chronic hypoxic conditions.

Loss of Parkin-associated mitophagy inhibits the anti-apoptotic effect of liraglutide

To examine the consequences of mitophagy activation under treatment with liraglutide, mitochondrial function was evaluated. Since the generation of energy is the principal role of mitochondria, the cellular ATP content was detected (39). As presented in Fig. 4A, hypoxia repressed ATP production when compared with the control group. However, treatment with liragutide reversed the effect on ATP content. To examine whether Parkin-associated mitophagy was responsible for the protective role of liraglutide on ATP production, siRNA was used to knock down the expression of Parkin. Following silencing of Parkin in liraglutide-treated H9C2 cells, the ATP content was decreased (Fig. 4A). These data indicated that Parkin-associated mitophagy was involved in liraglutide-mediated ATP production.

Figure 4. Liraglutide protects the mitochondrial function via Parkin-associated mitophagy. siRNA was used to knock down the expression of Parkin in liraglutide-treated H9C2 cells. (A) ATP production was detected, and liraglutide preserved the ATP concentration in a Parkin-dependent manner. (B) Mitochondrial potential was detected via JC-1 staining and (C) representative images are presented. Red fluorescence indicates the normal or healthy mitochondria and green fluorescence indicates the damaged mitochondrial potential. (D) Cellular ROS was detected via flow cytometry and (E) the results were analyzed. Alterations in the expression of cellular antioxidant factors were detected. (F) GSH, (G) MDA and (H) SOD concentrations were detected via ELISA analysis. (I) Caspase activity detection was used to demonstrate the role of Parkin-associated mitophagy in cellular apoptosis. *P<0.05 vs. Ctrl group; #P<0.05 vs. hypoxia group; @P<0.05 vs. hypoxia + Lir group. ATP, adenosine 5′-triphosphate; siRNA, small interfering RNA; ROS, reactive oxygen species; Lir, liraglutide; Ctrl, control; GSH, glutathione; MDA, malondialdehyde; SOD, superoxide dismutase.

ATP generation results from mitochondrial membrane potential (38). Mitochondria convert the proton gradient of mitochondrial membrane potential to chemical energy. Accordingly, the alteration in membrane potential was investigated. Through JC1 staining, it was observed that hypoxia reduced the mitochondrial membrane potential, as revealed by increased green fluorescence (Fig. 4B and C). However, liraglutide had the ability to maintain the membrane potential and this effect was reversed by Parkin knockdown. Aside from the collapse of mitochondrial membrane potential, it was observed that hypoxia induced excessive cellular oxidative stress, as evidenced by elevated ROS (Fig. 4D and E). Liraglutide had the ability to repress ROS release, whereas this effect was neutralized by Parkin silencing (Fig. 4D and E). As a consequence of ROS overproduction, a marked decline in the expression of antioxidant factors was observed, including GSH and SOD (Fig. 4F-H). By contrast, the expression of MDA, the lipid peroxide, was increased. Liraglutide had the ability to reverse the effect on GSH and SOD, although it repressed MDA in a Parkin-dependent manner (Fig. 4F-H). In order to examine whether mitophagy was associated with cellular death under chronic hypoxia, caspase3 activity was detected. As presented in Fig. 4I, the caspase3 activity was elevated under treatment with hypoxia and was reduced in response to liraglutide application. Notably, loss of Parkin increased caspase3 activity despite treatment with liraglutide. These data indicated that Parkin-associated mitophagy was responsible for the beneficial effects of liraglutide on mitochondria involving energy metabolism, mitochondrial membrane potential, cellular oxidative stress, redox biology and cellular apoptosis.

SIRT1 is activated by liraglutide and contributes to Parkin-dependent mitophagy

The molecular mechanism through which liraglutide regulated Parkin-dependent mitophagy was investigated in the present study (Fig. 5). SIRT1 has been identified to be a novel factor for mitochondrial homeostasis by sustaining mitochondrial membrane potential, reducing oxidative injury and promoting mitochondrial biogenesis (40). Based on this, it was hypothesized that SIRT1 may be the upstream signaling molecule for Parkin-dependent mitophagy. Western blotting was performed in the infarcted heart to examine the alterations in SIRT1 expression (Fig. 5A and B). It was observed that SIRT1 expression was downregulated in the infarcted heart. However, treatment with liraglutide was able to could increase SIRT1 expression in a dose dependent manner (Fig. 5A and B). Similarly, the downregulation of SIRT1 was also observed in H9C2 cells under chronic hypoxia (Fig. 5C-E), and this tendency was reversed by liraglutide treatment.

Figure 5. Liraglutide activates Parkin via the SIRT1 pathway. (A) Western blotting and (B) densitometric analysis was used to detect the alterations in SIRT1 in the infarcted heart. Liraglutide was able to sustain the expression level of SIRT1 in a dose-dependent manner. In vitro, total proteins were isolated and analyzed via (C) western blotting to measure the alterations in (D) SIRT1 and (E) Parkin expression. SRT, the activator of SIRT1, was applied in the control group under hypoxia conditions. SRT elevated the expression levels of SIRT1 and Parkin in the context of hypoxic stimulation. Selisistat, the inhibitor of SIRT1, was used in the liraglutide-treated cells. Selisistat blocked the promotive effect of liraglutide on SIRT1 and Parkin expression. (F) Caspase9 activity and (G) LDH release assays were used to detect cell viability and mitochondrial apoptosis, respectively. (H) Co-staining of SIRT1 and Parkin. Liraglutide enhanced the level of Parkin expression via SIRT1. Blockade of SIRT1 inhibited the role of liraglutide on Parkin. By contrast, activation of SIRT1 was able to reverse the alteration in Parkin expression under hypoxic condition. *P<0.05 vs. Ctrl group; #P<0.05 vs. hypoxia group; @P<0.05 vs. hypoxia + Lir group. SIRT1, NAD-dependent protein deacetylase sirtuin-1; SRT, SRT1720; Lir, liraglutide; Ctrl, control; LDH, lactate dehydrogenase.

To demonstrate whether SIRT1 was involved in Parkin-dependent mitophagy, SRT (10 µM for 4 h), an activator of SIRT1, was used to reverse the decrease in SIRT1 expression under treatment with hypoxia. The SIRT1 inhibitor Selisistat (10 µM for 6 h) was applied to reduce SIRT1 expression in liraglutide-treated cells. As presented in Fig. 5C-E, activation of SIRT1 under hypoxic conditions was able to reverse the effect on Parkin expression, consistent with the results from the liraglutide group. However, in the liraglutide-treated cells, Selisistat not alleviated the SIRT1 expression and repressed Parkin expression (Fig. 5C-E). In order to provide further evidence for the regulatory role of SIRT1 in Parkin, co-staining for SIRT1 and Parkin was performed. It was observed that hypoxia reduced the level of SIRT1 and Parkin expression (Fig. 5H). However, these alterations were rescued by treatment with liraglutide or SRT (Fig. 5H). By contrast, following inhibition of SIRT1 with Selisistat, the beneficial effect of liraglutide on SIRT1 and Parkin disappeared.

Caspase9 activity and LDH release assays were performed to investigate whether SIRT1 was involved in mitochondrial protection and cellular survival. It was observed that caspase9 activity and LDH content were increased in the hypoxia group (Fig. 5F and G). However, treatment with liraglutide or SRT application was able to abate the caspase9 activity and LDH release (Fig. 5F and G). Notably, following the blockade of SIRT1 with Selisistat, caspase9 activity and LDH content were increased despite treatment with liraglutide. These data illustrated that SIRT1 was the upstream molecule for Parkin-dependent mitophagy, and that SIRT1 contributed to mitochondrial protection and cellular survival. The beneficial role of liraglutide in the post-infarcted heart through regulation of SIRT1-mediated mitophagy was summarized in Fig. 6.

Figure 6. Illustrative model of the action of liraglutide on the infarcted heart via Parkin-dependent mitophagy, through the activation of SIRT1. Liraglutide signals SIRT1 which enhances the expression of Parkin, the receptor of mitophagy. Through the SIRT1/Parkin pathway, liraglutide activates mitophagy which is vital for the improvement of the infarcted heart. Protective mitophagy blocks mitochondrial apoptosis, promoting cardiomyocyte survival. In addition, mitophagy alleviates oxidative stress and preserves the mitochondrial function. These beneficial effects reduce cardiac fibrosis and the inflammatory responses, leading to the improvement of the infarcted heart. SIRT1, NAD-dependent protein deacetylase sirtuin-1.

Discussion

In the present study, it was observed that: i) Liraglutide had the ability to repair the post-infarcted heart; ii) liraglutide reduced fibrosis, the inflammatory response and myocardial death; iii) mechanistically, chronic hypoxic stimulation caused cardiomyocyte death via mitochondrial apoptosis, as evidenced by reduced mitochondrial membrane potential, cyt-c leakage and increased pro-apoptotic protein expression; iv) liraglutide inhibited mitochondrial apoptosis via mitophagy; v) liraglutide elevated SITR1 expression which increased Parkin, leading to mitophagy activation; vi) protective mitophagy reversed cellular ATP production, reduced cellular oxidative stress and balanced the redox response; and vii) the beneficial effect of liraglutide on cardiomyocyte apoptosis was dependent on GLP1R. To the best of our knowledge, this is the first study to describe the role of liraglutide in the post-infarcted heart, Parkin-dependent mitophagy and the SIRT1 signaling pathway.

Liraglutide is a novel GLP-1 receptor agonist that acts as a hypoglycemic agent by activating the GLP1R. A previous study has reported that liraglutide holds cardioprotective properties, and that this cardioprotective effect is independent of its hypoglycemic action (21). Previous studies have demonstrated that the application of liraglutide for 3 months reduced myocardial injury and increased cardiac function in patients with AMI (20,24). These findings suggested that liraglutide may have the ability to repair the infarcted heart. However, little evidence is available to explain the mechanism underlying this. The present study demonstrated that treatment with liraglutide for 1 month following myocardial infarction reduced cardiac fibrosis, inflammatory responses and cardiomyocyte death. These observations may explain the protective effect of liraglutide on the infarcted heart. Notably, liraglutide was previously demonstrated to improve hypertension, dyslipidemia, vascular endothelial dysfunction and other cardiovascular disease risk factors that serve as the pathological foundations of AMI (19,22,41). These beneficial actions of liraglutide may be involved in the improvement of the post-infarcted heart. Considering that few strategies are accessible to delay and repair the post-infarcted heart, the use of liraglutide may bring further clinical benefits for patients with AMI.

In the present study, it was determined that the SIRT1/Parkin/mitophagy pathway was the primary mechanism of defense that was enhanced by liraglutide to preserve cardiomyocyte survival in a state of chronic hypoxia. Mitochondria are the center of energy metabolism and are the regulators of cellular signaling (40,42). Previous studies have demonstrated that mitochondrial dynamics, particularly mitophagy, serve an important role in reducing myocardial and endothelial cell damage (13,43). Mitophagy is a selective form of general autophagy in which mitochondria are specifically targeted for degradation at the autophagolysosome (44). Through mitophagy, the damaged mitochondria are separated, leading to the balance of mitochondrial quality and quantity (45,46). Notably, the process of mitophagy relies on a growing cadre of ‘mitophagy adaptors’ or ‘mitophagy receptors’. There are three receptors which have been identified to be associated with mitophagy activation: FUNDC1, Bnip3 and Parkin (47,48). In the present study, it was observed that Parkin-associated mitophagy was inactivated under chronic hypoxia treatment. Notably, a previous study suggested that acute hypoxia, including ischemia reperfusion injury, may activate mitophagy (11). Brief ischemia or hypoxia may trigger mitophagy to consume damaged mitochondria and sustain cellular homeostasis. However, long-term hypoxia is the inhibitory signal for mitophagy, according to the present data and a previous study (49). Previous studies have demonstrated that mitophagy activity is downregulated during the progression of cardiac malfunction following AMI (50–52), which is consistent with the results of the present study. Therefore, identifying a means by which to activate mitophagy is important for the repair of the infarcted heart. In the present study, liraglutide activated mitophagy via the SIRT1/Parkin pathway. SIRT1 is one of the seven mammalian homologs (SIRT1-SIRT7) of yeast silent information regulator 2. SIRT1 is an NAD+-dependent protein deacetylase. SIRT1 serves a number of roles in cells, including longevity, apoptosis, DNA repair, inflammation and mitochondrial regulation (25). As an important protein in cellular metabolism, the regulatory effect of SIRT1 on mitochondrial dynamics has gained much attention. In the present study, it was demonstrated that SIRT1 was responsible for Parkin upregulation and mitophagy activation. These findings enrich the current understanding of the role of SIRT1 in mitochondrial homeostasis.

As a consequence of mitophagy activation in the present study, cellular oxidative and mitochondrial damage was rescued. Considering that excessive oxidative injury and mitochondrial energy disorders are the principal feature of cardiac failure following AMI (53,54), mitophagy may be a potential therapeutic target for restoring mitochondrial homeostasis and cardiac energy metabolism. Notably, previous studies have indicated that excessive mitophagy may shift cellular survival towards death signaling. It has been argued that excessive mitophagy may consume the majority of mitochondria, leading to the loss of the majority of the mitochondrial mass (13,55). As a consequence of mitochondrial loss, cells fail to generate sufficient ATP to fuel the cellular biological functions, including contraction, division and mobilization. Therefore, it was suggested that moderate mitophagy activity is required to maintain cellular homeostasis. However, whether excessive mitophagy is harmful for myocardial structure and function remains unclear. Further studies are required to examine this question.

In conclusion, the results of the present study illustrated the important role of liraglutide in repairing the infarcted heart via mitophagy. Liraglutide amplified the SIRT1/Parkin pathway leading to the activation of protective mitophagy, which was the essential element of mitochondrial protection. Mitophagy reduced cellular oxidative stress, alleviated mitochondrial damage and abated cellular apoptosis under chronic hypoxic conditions. The present findings elucidated the mechanism underling cardiac dysfunction. In addition, the results of the present study identified a convenient and effective method of repairing the infarcted heart. Therefore, the enhancement of mitophagy by liraglutide may be a practical and efficient adjuvant to the treatment of cardiac failure following AMI.

Acknowledgements

The authors of the present study would like to thank the Institute of Basic Medicine Science of Suzhou No. 1 People's Hospital (Suzhou, China). The present study was supported by grants from the Jiangsu Medical and Health Science and Technology Development Plan (grant no. 2016WS0259).

References