Metformin alleviates ethanol-induced cardiomyocyte injury by activating AKT/Nrf2 signaling in an ErbB2-dependent manner

Metformin, a first-line oral anti-diabetic drug, has recently been reported to exert protective effect on various cardiovascular diseases. However, the potential role of metformin in ethanol-induced cardiomyocyte injury is still unknown. Therefore, this study was aimed to investigate the effect of metformin on ethanol-induced cardiomyocyte injury and its underlying mechanism. H9c2 cardiomyocytes were exposed to ethanol for 24 h to establish an ethanol-induced cardiomyocyte injury model, and followed by treatment with metformin in the presence or absence of Lapatinib (an ErbB2 inhibition). CCK8 and LDH assays demonstrated that metformin improved cell viability in cardiomyocytes exposed to ethanol. Furthermore, metformin suppressed cardiomyocyte apoptosis and reduced the expressions of apoptosis-related proteins (Bax and C-CAS-3). In addition, our results showed that metformin activated the AKT/Nrf2 pathway, and then promoted Nrf2 nuclear translocation and the transcription of its downstream antioxidant genes (HO-1, CAT and SOD2), thereby inhibiting oxidative stress. Interestingly, we found that ErbB2 protein expression was significantly inhibited in ethanol-treated cardiomyocytes, which was markedly reversed by metformin. In contrast, Lapatinib largely abrogated the activation of AKT/Nrf2 signaling by metformin, accompanied by the increases in oxidative stress and cardiomyocyte apoptosis, indicating that metformin prevented ethanol-induced cardiomyocyte injury in an ErbB2-dependent manner. In summary, our study provides the first evidence that metformin protects cardiomyocyte against ethanol-induced oxidative stress and apoptosis by activating ErbB2-mediated AKT/Nrf2 signaling. Thus, metformin may be a potential novel treatment approach for alcoholic cardiomyopathy.


Introduction
Excessive alcohol consumption has been associated with an increased risk of heart disease, particularly alcoholic cardiomyopathy (ACM), which is a specific heart muscle disease [1,2]. Evidence from previous research has shown that alcohol can induce cardiomyocyte apoptosis and myocardial damage, leading to heart chamber dilatation, reduced myocardial contractility, cardiac dysfunction, and ultimately to ACM and even heart failure [3,4]. It is worth noting that mortality and morbidity of ACM are increasing on a yearly basis [5]. Thus, there is an urgent need to uncover the effective strategies for the treatment of ACM.
The balance between reactive oxygen species (ROS) generation and intrinsic antioxidant defense systems is critical for cardiomyocyte homeostasis under physiological conditions [6]. Oxidative stress, excessive ROS production exceeding intracellular antioxidant capacity, has been observed in ethanol-exposed cardiomyocytes [7]. In fact, the overproduction of ROS is tightly related with alcohol metabolism, which leads to cardiomyocyte apoptosis, ultimately causing cardiac structural and functional abnormalities [5,8]. Therefore, the discovery and validation of drugs that can effectively improve oxidative stress state of cardiomyocytes caused by ethanol may be a promising therapeutic strategy.
Metformin, a first-line blood glucose-lowering drug, is most widely used for the treatment of type II diabetes mellitus [9]. Interestingly, the potential role of metformin in heart diseases has recently attracted substantial scientific attention. Notably, several lines of evidence have demonstrated that metformin protects against a variety of heart diseases, including cardiac hypertrophy, myocardial infarction, myocardial ischemia-reperfusion and heart failure [10][11][12][13]. However, there is no evidence that metformin has a potential effect on ethanol-induced cardiomyocyte injury.
Nuclear factor erythroid-2 related factor 2 (Nrf2) is a redox-sensitive transcription factor that considered to be one of the most important intracellular defense against oxidative stress [14]. Further study have demonstrated that Nrf2 binds to the antioxidant-response element (ARE) and then promotes the transcription of antioxidant genes, such as, Heme Oxygenase-1 (HO-1), Catalase (CAT) and Superoxide Dismutase 2 (SOD2) [15]. Of note, Nrf2 has been reported to protect heart from damage by upregulating protective antioxidant response element genes to reduce oxidative stress [16]. In addition, Nrf2 plays a vital role in ethanol-induced cellular damage. For instance, Nrf2 activation ameliorated ethanol-induced oxidative stress and liver injury in mice [17].
ErbB2 (Her2/Neu), a member of the epidermal growth factor receptor (EGFR) family, was originally identified owing to its potent oncogenic activity [18]. The importance of ErbB2 in cardiac homeostasis soon became obvious by a discovery that some ErbB2-overexpressing breast cancer patients treating with trastuzumab or herceptin (targeting ErbB2 antibodies) developed cardiac dysfunction [19,20]. In fact, ErbB2 plays essential role in heart development and postnatal tissue homeostasis [21]. ErbB2-mutant mice displayed cardiac defects during embryogenesis and died at embryonic day 10.5 [22]. Additionally, ErbB2 conditionalknockout (ErbB2-CKO) mice developed heart failure, and cardiomyocytes isolated from ErbB2-CKO mice were more susceptible to doxorubicin insult [23]. Meanwhile, emerging evidences have suggested that ErbB2 can protect against dilated cardiomyopathy, myocardial infarction and heart failure [24,25]. Moreover, the protective effects of ErbB2 are largely mediated by the activation of multiple intracellular signal transduction pathways, especially Protein kinase B (AKT) [26]. For example, D'Uva et al. demonstrated that ErbB2 overexpression effectively activated AKT and then promoted cardiomyocyte proliferation, which eventually improved myocardial infarction [27]. Hence, ErbB2 may be a therapeutic target for the treatment of ACM.
Therefore, the aim of the present study is to explore the potential role and mechanisms of metformin in ethanolinduced cardiomyocyte injury. In addition, whether ErbB2 mediates the cardioprotective effect of metformin was investigated in this study.

H9c2 cell culture and treatment
The rat H9c2 cardiomyocyte cell line was purchased from the American Type Cell Collection (ATCC) and cultured in Dulbecco's modified Eagle's medium (DMEM; Gibco, C11995500BT) supplemented with 10% fetal bovine serum (FBS; Gibco, 10270-106) and 1% penicillin/streptomycin (Solarbio, P1400) in an incubator containing 95% air and 5% CO 2 at 37 °C. An ethanol-induced cardiomyocyte injury model was established as previously described [5]. In Brief, H9c2 cardiomyocytes were exposed to ethanol (200 mM) for 24 h and followed by treatment with a range of concentrations of metformin.
To further determine whether ErbB2 played a role in protective effect of metformin against ethanol treatment, H9c2 cardiomyocytes were pretreated with Lapatinib (an inhibitor of ErbB2; Selleck, GW-572,016) for 2 h before ethanol and metformin treatment.

Cell viability assay
Cell viability was assessed using the cell counting kit-8 (CCK-8) (Biosharp, BS350B), according to the manufacturer's protocol. A total of 5 × 10 3 cells/well were seeded in 96-well plates and then exposed to ethanol treated with metformin for 24 h. Following treatment, 10 µL CCK-8 solution was added to each well, and H9c2 cardiomyocytes were further incubated for 2 h. The absorbance was measured at 450 nm using a microplate reader (BioTek, USA).

Lactate dehydrogenase release assay (LDH) assay
When cardiomyocyte damage occurs, LDH is released into the culture medium. H9c2 cardiomyocytes were seeded on 96-well plates and treated as described above. LDH release in the cell culture supernatant was collected and subsequently detected by LDH cytotoxicity assay detection kit (Beyotime, P0028), according to the manufacturer's instructions. The absorbance was measured at 490 nm using a microplate reader (BioTek, USA).

Dihydroethidium (DHE) assay
To detect the ROS level in cardiomyocytes, dihydroethidium (DHE, Sigma-Aldrich, D7008) staining was performed. H9c2 cardiomyocytes were seeded on 6-well plates and treated as described above, and then incubated with 5 µM DHE (in PBS) for 30 min at 37 °C in a dark chamber. Subsequently, the fluorescence intensity was observed with a computer-assisted microscope (Nikon, Japan).

Western blot analysis
Total proteins from H9c2 cardiomyocytes were homogenized or lysed with RIPA buffer (Solarbio, R0010) and quantified by the Bicinchoninic Acid (BCA) protein assay kit (Thermo Scientific, 23,223). The amount of 30 µg protein lysate were separated by SDS-PAGE gels and then transferred to polyvinylidene fluoride (PVDF) membranes (Merck Millipore, IPVH00010). After blocking blocked with 5% BSA in Tris-buffered saline containing 0.1% (v/v) Tween 20 (TBST) for 2 h at room temperature, the membrane were incubated with appropriate primary antibodies overnight at 4 °C. Then, membranes were incubated with either HRPgoat-anti-mouse (Abcam, ab6789) or HRP-goat-anti-rabbit (Abcam, ab6721), to bind the primary antibodies for 2 h at room temperature. Next, immunoreactive bands were visualized by ChemiDoc MP Imaging System using Western-Bright ECL HRP Substrate (Advansta, K-12,045-D50) and quantitatively analyzed with ImageJ software.

Extraction of cytoplasm and nucleus
For nuclear Nrf2 accumulation assays, H9c2 cardiomyocytes were harvested and lysed to obtain cytoplasmic and nuclear lysates using nuclear and cytoplasmic extraction reagents kit (Beyotime, P0028), according to the manufacturer's instructions. GAPDH and LaminB (Proteintech, 12987-1-AP) were used as the loading control for the cytoplasmic and the nuclear fraction, respectively.

Real-time quantitative PCR (RT-qPCR) analysis
Total RNA was extracted from H9c2 cardiomyocytes using TRIzol reagent (Takara Bro Inc, 9108), as described by the manufacturer's instructions. Next, 1 µg total RNA was reversely transcribed to complementary DNA (cDNA) by the Hiscript ® III Reverse Transcriptase kit (Vazyme, R223-01). The qPCR reactions was performed on an ABI7500 Real-Time PCR System (Applied Biosystems, USA) using ChamQ Universal SYBR qPCR Master Mix (Vazyme, Q711-02) with specific primers. The amplification procedure was performed at 95 °C for 30 s; 40 cycles at 95 °C for 5 s, 60 °C for 30 s; followed by a melting curve at 95 °C for 15 s, 60 °C for 1 min, 95 °C for 15 s, according to the manufacturer's instructions. The relative expression levels of each gene were quantitated using the 2 −∆∆Cq method and GAPDH was used as a control. The sequences of specific primers used for RT-qPCR in this study are listed as below: purchased from Sangon Biotech (Shanghai, China).

Statistical analysis
Data were presented as mean ± standard deviation (SD). Differences of each sample were evaluated using the unpaired Student's two-tailed t test for two experimental groups or one-way analysis of variance (ANOVA) for multiple groups. If P value < 0.05, the result was considered significant different. All experiments were repeated at least three independent times.

Metformin attenuates ethanol-induced cardiomyocyte apoptosis
To examine the effects of metformin on ethanol-triggered cardiomyocyte injury, H9c2 cardiomyocytes were treat with different concentration of metformin for 24 h. Firstly, the CCK-8 analysis showed that 0-10 mM metformin had no obvious inhibitory effects on cardiomyocyte viability (Fig. 1A). The result shown in Fig. 1B-C illustrated that metformin treatment (0.5, 1, 2.5, and 5 mM) remarkably relieved the decrease in cell viability and the increase in LDH release induced by ethanol stimulation. Furthermore, we found that treatment with 1 mM metformin had a stronger effect than 0.5 mM metformin; however, there was no significant difference among treatment with 1, 2.5 and 5 mM. Therefore, 1mM metformin was selected as the working concentration for further experiments in this study. Meanwhile, western blot assay revealed that ethanol exposure significantly increased the protein levels of apoptosisrelated factors, Bax and cleaved-caspase-3 (C-CAS-3), which were both inhibited by metformin treatment (Fig. 1D). These above results demonstrated that metformin effectively suppressed ethanol-induced cardiomyocyte apoptosis.

Metformin protects cardiomyocyte against ethanol-induced oxidative stress by activating AKT/Nrf2 signaling
It well accepted that oxidative stress triggered by ethanol administration is a major contributor to the apoptosis of cardiomyocytes. Therefore, DHE staining was used to determine the level of the ROS; the result showed that ethanol treatment caused ROS accumulation in cardiomyocytes compared with the control group, which was suppressed by metformin ( Fig. 2A). Meanwhile, we found that ethanol significantly inhibited the phosphorylation of AKT (p-AKT/ AKT ratio), and then suppressed the protein level of Nrf2 in the nucleus. However, metformin attenuated ethanol-induced inhibition of AKT phosphorylation and subsequently promoted nuclear accumulation of Nrf2 (Fig. 2B-C). In addition, the ethanol-induced the decreases in the expressions of Nrf2 downstream genes, HO-1, CAT and SOD2, were also up-regulated by metformin (Fig. 2D). However, metformin treatment alone had no effect on cardiomyocytes under basal conditions. Collectively, these data suggested that metformin ameliorated ethanol-induced oxidative stress via regulating Akt/Nrf2 pathway.

Metformin upregulates the protein expression of ErbB2 in ethanol-treated cardiomyocyte
ErbB2 plays a critical role in regulating pathophysiological processes in the heart. To explore the role of ErbB2 in ethanol-triggered cardiomyocyte injury, we detected the protein level of ErbB2 in cardiomyocyte after ethanol exposure. As shown in Fig. 3A, compared with the control group, ErbB2 expression in ethanol group was significantly suppressed at the protein level (Fig. 3A). Further study revealed that ErbB2 expression was significantly down-regulated at 3, 6, 12 and 24 h but not significantly changed at 1 h after ethanol exposure ( Fig. 3B-C). However, metformin treatment (0.5 and 1 mM) elevated expression of ErbB2 in a dose-dependent manner (Fig. 3D). Taken together, these results strongly supported the view that metformin increased the protein expression of ErbB2 in ethanol-treated cardiomyocytes.

Inhibition of ErbB2 restricts the protective effect of metformin on ethanol-induced cardiomyocyte injury
To further investigated whether ErbB2 was responsible for the beneficial effect of metformin on ethanol-induced cardiomyocyte injury. Thus, we treated H9c2 cardiomyocytes with Lapatinib, an inhibitor of ErbB2. Firstly, CCK8 analysis revealed that the cell viability was not affect by Lapatinib Fig. 1 Metformin attenuates ethanol-induced cardiomyocyte apoptosis. A H9c2 cardiomyocytes were subjected to 0.25, 1, 2.5, 5 or 20 mM metformin for 24 h at the basic level, and CCK8 assay was performed to examine the cell viability. B H9c2 cardiomyocytes were subjected to 0.25, 0.5, 1, 2.5 or 5 µM metformin under ethanol expo-sure, and CCK8 assay was performed to examine the cell viability. C LDH assay was performed to measure the LDH release from H9c2 cardiomyocytes. D Western blot was performed to detect the protein levels of Bax and C-CAS-3. Data are represented as the mean ± SD, # P < 0.05 versus con group, * P < 0.05 versus ethanol alone group in a range of 0.5-2.5 µΜ, but was notably decreased at 5 µΜ and 20 µΜ. Based on our results and other similar study, 2.5 µΜ Lapatinib was selected in the subsequent experiments. As depicted in Fig. 4B-C, Lapatinib restricted the anti-apoptotic effect of metformin following ethanol exposure, evidenced by the decreased cell viability and increased LDH release in cardiomyocytes. Consistently, metformin markedly attenuated the ethanol-induced increase in Bax and C-CAS-3 protein levels, which were dramatically reversed by Lapatinib (Fig. 4D).
In addition, the effect of ErbB2 inhibition on the antioxidative activity of metformin was also investigated. Consistent with the observation in Fig. 2A, the ROS level was elevated after ethanol exposure but was reduced by metformin administration. In contrast, Lapatinib greatly blunted the metformin-mediated suppression of ROS overproduction induced by ethanol exposure (Fig. 5A). Next, the activation of the ErbB2 downstream cascade, AkT, was detected by western blot analysis. We found that metformin-elicited the phosphorylation of AkT was notably blocked by Lapatinib in ethanol-treated cardiomyocytes. At the molecular level, Lapatinib significantly decreased protein levels of total Nrf2 and nuclear Nrf2, which were increase by metformin followed by ethanol stimulation. Moreover, Lapatinib substantially abrogated the capacity of metformin to promote transcription of Nrf2 downstream antioxidant genes, as confirmed by the decreased HO-1, CAT and SOD2 at the mRNA level. Overall, these data implied that ErbB2 mediated the protective effect of metformin on ethanol-induced cardiomyocyte injury.

Discussion
Previous studies have shown that metformin can alleviate myocardial damage in various heart diseases [10]. It is also well accepted that both Nrf2 and ErbB2 play crucial roles in cardiac homeostasis [26,28]. However, there was no evidence to show the role of metformin in ethanol-induced cardiomyocyte injury. Furthermore, the underlying mechanisms connecting between ErbB2 and Nrf2 in ethanol-induced cardiomyocyte injury remain unknown. To the best of our knowledge, the present study reveals for the first time that: (i) metformin can prevent ethanol-induced cardiomyocyte injury by reducing oxidative stress; (ii) the expression of ErbB2 is decreased in response to ethanol; (iii) metformin upregulated AKT/Nrf2 signaling via activating ErbB2. Overall, the present study indicated that metformin might be an effective strategy to prevent ethanol-induced myocardial injury.
Growing evidence indicated that cardiomyocyte apoptosis triggered by oxidative stress contributed to the progression of ACM [29]. Similarly, our results also showed that ethanol exposure caused ROS overproduction, accompanied by cardiomyocyte apoptosis. In recent years, metformin has been proven to exhibit a protective effect on heart diseases, and the cardioprotection of metformin is strongly associated with the inhibition of oxidative stress [30,31]. More importantly, metformin also exhibited beneficial effects on ethanolinduced oxidative stress and cellular damage. In adolescent rats, metformin improved ethanol-induced learning and memory impairment by reducing oxidative damage in the hippocampal and cortical tissue [32]. Moreover, metformin protected against ethanol-induced hepatic injury and gastric injury via its antioxidant property [33,34]. Considering that metformin has powerful anti-oxidative effects, we speculated ErbB2 inhibitor restricted anti-apoptotic effect of metformin in ethanol-treated cardiomyocyte. A H9c2 cardiomyocytes were subjected to 0.25, 1, 2.5, 5 or 20 µM metformin for 24 h at the basic level, and CCK8 assay was performed to examine the cell viability, # P < 0.05 versus con group. H9c2 cardiomyocytes were pretreated with or without 2.5 µM Lapatinib for 2 h and then incubated with or without 1 mM metformin for 24 h in the presence or absence of ethanol.
B CCK8 assay was performed to examine the cell viability. C LDH assay was performed to measure the LDH release from H9c2 cardiomyocytes. D Western blot was performed to detect the protein levels of Bax and C-CAS-3. # P < 0.05 versus con group, * P < 0.05 versus ethanol alone group, & P < 0.05 versus ethanol + metformin group. Data are represented as the mean ± SD that metformin might exhibit potential therapeutic value in improving ethanol-induced myocardial injury. As expected, our data showed that metformin alleviated ethanol-induced oxidative stress and subsequent cardiomyocyte apoptosis.
Then, the question arises of how metformin alleviates oxidative stress in ethanol-treated cardiomyocyte. Numerous studies have provided evidence that Nrf2 is a well-established master redox regulator and Nrf2 activation protects cardiomyocyte against ethanol-induced oxidative stress and apoptosis [35,36]. Subsequent experiments revealed that AKT was a well-known up-stream regulator of Nrf2 and mediated the cardioprotection of Nrf2 [37,38]. Our previous study also showed that Nrf2 activation attenuated cardiac damage after myocardial infarction injury, whereas AKT inhibition almost abolish the effect of Nrf2 [39]. It is worth noting that metformin is an effective Nrf2 activator, and the effect is tightly related to activation of AKT [40,41]. These results lead us to hypothesize that metformin might elicit similar effect in ethanol-treated cardiomyocytes. Similarly, we found that Nrf2 signaling was significantly inhibited by ethanol exposure. In contrast, metformin markedly phosphorylated AKT and then promoted Nrf2 nuclear translocation, thereby increasing the expression of Nrf2 downstream antioxidative enzymes, which is consistent with our hypothesis. Taken together, our results suggested that metformin alleviated ethanol-induced cardiomyocyte injury by reducing oxidative stress through AKT/Nrf2 signaling.
Although ErbB2 is well-known for its oncogenic activity and anti-ErbB2 therapy is widely used in clinical treatment for breast cancer, ErbB2 appears to be a protective factor for cardiomyocyte with poor proliferative capacity [18,42]. ErbB2 was highly expressed in neonatal cardiomyocytes during embryonic cardiac development but progressively declined after birth [24]. Furthermore, ErbB2 expression was suppressed in pathological cardiac tissue [21,43]. Consistently, in this study, ErbB2 protein expression was significantly down-regulated in cardiomyocytes in response to ethanol stimulation, suggesting that ErbB2 played a prominent role in ACM.
Of note, activation and agonists of ErbB2 have been proposed as therapeutic approaches for heart disease. ErbB2 overexpression and activator (Neuregulin 1, NRG1) promote cardiac function and repair, particularly in myocardial infarction and heart failure [27,44,45]. More meaningfully, Fig. 5 ErbB2 inhibitor weakened anti-oxidative effect of metformin in ethanol-treated cardiomyocyte. H9c2 cardiomyocytes were pretreated with or without 1 µM Lapatinib for 2 h and then incubated with or without 1 mM metformin for 24 h in the presence or absence of ethanol. A DHE staining was performed to measure intracellular ROS (left) and then quantitatively analyzed the intensity of DHE fluorescence (right) (Scale bar = 100 μm). B Western blot was performed to detect the protein levels of p-AKT/AKT ratio, t-Nrf2, n-Nrf2 and HO-1. C RT-qPCR analysis was performed to determine the mRNA levels of HO-1, CAT and SOD2. Data are represented as the mean ± SD, # P < 0.05 versus con group, * P < 0.05 versus ethanol group, & P < 0.05 versus ethanol + metformin group a phase II clinical trial demonstrated that NRG1 effectively improved cardiac function in patients with stable congestive heart failure [46]. Interestingly, Belmonte et al. pointed out that ErbB2 functioned as a cardiac protective factor by reducing oxidative stress, who reported that ErbB2 markedly upregulated antioxidant enzymes to scavenge ROS accumulation and consequently prevented doxorubicin-induced cardiotoxicity [20]. Besides, AKT has been reported to be a key substrate of ErbB2, and blockage of AKT reversed the protective function of ErbB2 in the heart [27,47]. Furthermore, a recent literature implicated that ErbB2 silence resulted in inactivation of AKT [48]. Importantly, Zhu et al. demonstrated that ErbB2 was an important downstream effector of metformin [49]. Therefore, we speculated that ErbB2 might be involved in the protection against ethanolinduced cardiomyocyte injury and the regulation of AKT/ Nrf2 signaling by metformin. In our result, metformin significantly up-regulated the protein expression of ErbB2 in ethanol-treated cardiomyocytes. However, ErbB2 inhibitor largely abrogated metformin-mediated activation of AKT/ Nrf2 signaling, which resulted in elevated ROS accumulation and cardiomyocyte apoptosis. Collectively, our aforementioned results indicated that metformin protected against ethanol-induced cardiomyocyte injury via activating ErbB2mediated AKT/Nrf2 signaling.
There were limitations in our study. Firstly, although the H9c2 cell culture model has proved the protective effect and mechanism of metformin on cardiomyocyte under ethanol stimulation; however, H9c2 cells are cardiomyoblast-derived embryonic rat heart and may be characterized by higher expression of ErbB2 than adult mature cardiomyocytes, this notion needs to be further validated using other cardiomyocyte cell line and animal model. Secondly, ErbB2 is a receptor tyrosine kinase, and its phosphorylation widely participates in cellular signal transduction and biological processes, therefore, it is something worth investigating the potential role of ErbB2 phosphorylation in the protection of metformin against ethanol-induced cardiomyocyte injury. Thirdly, AMPK plays an important role in the development of alcoholic cardiomyopathy and the biological effect of metformin, therefore, the effect of AMPK on the cardioprotection of metformin against ethanol-induced cardiomyocyte injury should be systematically investigated in the follow up studies. Finally, further studies are expected to systematically investigate the underlying mechanism of how ethanol reduced and metformin increases ErbB2 expression in cardiomyocytes.
In conclusion, our present study uncover a previously unknown protective effect of metformin on ethanol-induced cardiomyocyte injury. Mechanistically, metformin up-regulates ErbB2 expression and then activates AKT/Nrf2 signaling; this effect promotes Nrf2 nuclear translocation and the transcription of its downstream antioxidant genes, thereby inhibiting oxidative stress and cardiomyocyte apoptosis. Thus, metformin may become a potential therapeutic agent against ethanol-induced cardiac injury.