Therapeutic Potential of Metformin in Diabetes Mellitus-Related Cardiovascular Complications

Abstract

The diabetic population continues to grow worldwide, resulting in many chronic cardiovascular complications, including atherosclerosis and diabetic cardiomyopathy, as well as an increase in the incidence of heart failure. Metformin, as the first-line oral therapy for type 2 diabetes, lowers blood glucose and reduces the incidence of diabetes mellitus (DM)-related cardiovascular events, such as myocardial infarction. The cardiovascular protective effect of metformin is due not only to the relief of insulin resistance and the improvement of glucose and lipid metabolism but also to the inhibition of oxidation and inflammation. Metformin exerts its multiple effects primarily through AMPK-dependent and AMPK-independent mechanisms. This chapter reviews the beneficial effects of metformin on DM-related cardiovascular complications and dissects the potential molecular mechanisms.

1. Introduction

Cardiovascular disease (CVD) is the leading cause of death in adults with diabetes mellitus (DM) [1]. DM caused 3.7 million deaths in individuals aged 70 and under, the majority of which were due to cardiovascular disease (43% of this 3.7 million) [2]. According to the action in diabetes and vascular disease: preterax and diamicron modified release controlled evaluation (ADVANCE) trial, adults with diabetes are two to three times more likely than adults without diabetes to develop cardiovascular disease. Adjusted for age at diagnosis, the risk of macrovascular events, including cardiovascular death, nonfatal myocardial infarction, or non-fatal stroke, increased by 49% for each additional 5-year duration of type 2 diabetes in adults. However, more severe hypoglycemic events occurred after intensive glycemic control, and the type of glycemic control had no significant effect on cardiovascular-related mortality [3, 4].

Metformin was not the first blood-glucose-lowering agent to be developed, and the process has an interesting history that begins with the withdrawal of phenformin and buformin (the only other biguanides used to lower blood glucose at that time) after they were found to have harmful effects in clinical trials [5]. Although the evidence of cardiovascular benefits was published in the United Kingdom prospective diabetes study (UKPDS) in 1998 [6], metformin emerged as the first-choice and most-prescribed oral medication for lower blood glucose in the USA in 1995 [7]. Recent studies have found that metformin has cardiovascular benefits as well, lowering mortality and morbidity associated with cardiovascular diseases, such as stable coronary artery disease and myocardial infarction [8, 9]. This chapter provides an overview of the cardiovascular benefits of metformin and its underlying mechanisms.

2. Metformin and atherosclerosis (AS)

Over the last 30 years, accumulating evidence has shown that metformin reduces atherosclerotic plaque formation in animals fed a high-cholesterol diet [10, 11]. Metformin significantly reduced atherosclerotic plaque and serum high-sensitivity C-reactive protein while inhibiting the activation of the nuclear factor κ-B (NF-κB) pathway in the vessel wall in a high cholesterol diet-induced atherosclerotic rabbit [11]. The same observation was also made in an atherosclerotic mouse model as well. Metformin treatment decreased plaque formation in a high-cholesterol diet-induced apolipoprotein E knockout (ApoE^−/−) mouse model [12]. It should be noted that metformin had an effect on plaque instability as well. Recent studies have found that metformin [100 mg/(kg·day)] significantly reduced calcification plaques in ApoE^−/− mice fed a high-fat diet (HFD) [13]. These findings suggest that metformin may not only alleviate atherosclerotic plaque formation but also improve plaque stability.

2.1 Metformin and endothelial injury

Vascular endothelial dysfunction is one of the most important pathological alterations and is considered an initial event for the development of AS. Metformin has been shown in clinical studies to improve endothelium-dependent vasodilation in diabetic patients [14]. In a preclinical study, metformin improved nitric oxide (NO)-mediated vasodilation in endothelial cells (ECs) in vitro by increasing NO production via the Adenosine 5′-monophosphate (AMP)-activated protein kinase (AMPK) pathway [15, 16]. Dong et al. demonstrated that activating AMPKα2 attenuates endoplasmic reticulum stress in vascular ECs [17], and metformin, an AMPKα2 agonist, protects human coronary artery ECs from diabetic lipoapoptosis by activating AMPK [18].

While investigating the effects of metformin on hyperglycemia and associated insulin resistance, clinical studies have addressed its effects on endothelial dysfunction as well. A study conducted in 2001 on the effects of metformin on endothelial function in type 2 diabetes (T2D) patients demonstrated that metformin improves insulin resistance and endothelium-dependent vasodilation; however, no significant effect of endothelium-independent or nitrate-independent vasodilation was observed in this study [19]. These findings suggest that vasodilatory effects are closely associated with metformin-induced insulin sensitivity. AS is a chronic inflammatory pathological process, and anti-inflammatory therapy is considered a functional approach for alleviating the atherosclerotic process. Long-term treatment (4.3 years) with metformin added to insulin therapy decreased the levels of several inflammatory biomarkers of endothelial dysfunction, such as von Willebrand factor (vWF) and soluble vascular cell adhesion molecule-1 (sVCAM-1) in T2D patients. These data indicate the potential of metformin in reducing cardiovascular events [20]. A similar study was conducted by Vitale et al. in 2005. They recruited 65 patients with metabolic syndrome to investigate the effects of metformin (500 mg, b.i.d, for 3 months) on endothelial function. Metformin treatment increased brachial artery endothelial-dependent flow-mediated dilation. It was shown that insulin resistance and endothelial dysfunction are interrelated and can be affected by metformin [21]. These effects of metformin have been demonstrated to benefit patients with type 1 diabetes as well [22]. In the REMOVAL trial in type 1 diabetes, metformin was shown to reduce cardiovascular risk, but no effect of glycemic control on endothelial dysfunction was observed [23]. Other studies involving women with polycystic ovary syndrome (PCOS) treated with metformin or rosiglitazone have shown that metformin is associated with improved flow-mediated vasodilation and does not affect endothelial relaxation independently, making it more effective than rosiglitazone [24, 25]. Metformin treatment for 3 months improved arterial stiffness and associated endothelial dysfunction and reduced carotid intima-media thickness in women with PCOS [26, 27]. Clinical trials involving metformin monotherapy (NCT00169624) or metformin in combination with other antidiabetic drugs in T2D patients (NCT00169624) and PCOS (NCT01459445) can be found online (https://clinicaltrials.gov). All of the aforementioned findings indicate that metformin monotherapy or in combination with other antidiabetic strategies can be a viable option for treating endothelial dysfunction in patients with hyperglycemia and insulin resistance.

2.1.1 Metformin and inflammatory response in ECs

Metformin (0.1–2.5 μg/ml) has been shown to inhibit advanced glycation end products (AGE)-induced monocyte adhesion by inhibiting endothelial cell adhesion molecules [28]. Metformin (2–10 mM) inhibits tumor necrosis factor (TNF)-α-induced gene expression of vascular cell adhesion molecule 1 (VCAM1), E-selectin, intercellular cell adhesion molecule-1 (ICAM-1), and monocyte chemoattractant protein-1 (MCP-1), thereby facilitating monocyte adhesion to activated ECs and the IKK/IKBα/NF-KB signaling pathway [29]. Poly ADP-ribose polymerase-1 (PARP-1) inhibits the expression of BCL6 (zinc finger protein 51) by binding to its intron 1, and it subsequently increases the expression of pro-inflammatory cytokines, such as VCAM-1 and MCP1. The AMPK activator or metformin (1 mM) induces PARP-1 dissociation from the Bcl-6 intron 1, thereby increasing the expression of Bcl-6, and inhibiting the expression of inflammatory mediators [30]. These findings indicate that metformin may be an effective drug for preventing monocyte adhesion. Additionally, metformin (20 μM) alleviated intracellular oxidative stress by up-regulating the expression of lectin-like oxidized low-density lipoprotein receptor 1 (LOX1) [31].

2.1.2 Metformin and NO production in ECs

NO is crucial for vascular physiological function and maintaining vascular tone. Endothelial nitric oxide synthase (eNOS)-dependent NO production increases endothelium-dependent vasodilation. Metformin (60 mg/kg/d) restores arterial endothelial function in Goto-Kakizaki (GK) rats (a spontaneous T2D animal model) and rats treated with streptozotocin (STZ) (an induced T2D animal model) by increasing NO bioavailability and restoring endothelium-dependent vasodilation [32]. Metformin (50–500 μM) also increases eNOS phosphorylation and its interaction with HSP90, resulting in an increase in NO production. High glucose levels indicate impaired eNOS-HSP90 interaction, which can be restored with metformin [16]. Metformin (250 mg/kg/d) has also been found to activate eNOS in endothelial progenitor cells isolated from STZ-induced diabetic mice [33]. Tetrahydrobiopterin (BH4) is a cofactor for the biological function of eNOS and regulates the NO level, which is important for vascular physiology. Cyclohydrolase 1 (GCH1) is a rate-limiting enzyme in BH4 biosynthesis whose deficiency affects BH4 level. The eNOS-activating effect of metformin is also related to guanosine triphosphate GCH1. Metformin (300 mg/kg/d) increases the protein expression of GCH at a posttranslational level [34]. In diabetic and obese mice, this protective effect increases endothelial-dependent vasodilation [35].

2.1.3 Metformin and vascular integrity in ECs

Another leading cause of diabetic vascular change is abnormal vascular integrity. Increased vascular permeability promotes monocyte extravasation. Metformin (0.1–1 mM) has also been shown to decrease vascular permeability of ECs in the brain. Metformin-induced AMPK was demonstrated to be involved in the prevention of endothelial dysfunction [36]. Glycocalyx is a matrix structure that prevents vascular permeability; metformin (0.33 mg/ml) can improve glycocalyx barrier function in db/db mice [37]. Metformin (400 mg/kg, bid) reduced lung endothelial hyperpermeability and systemic inflammatory response by activating AMPK in STZ-induced diabetic mice and db/db diabetic mice. Metformin administration improved survival, which was associated with suppression of VE-cadherin phosphorylation (pVE-cadherin) [38].

2.2 Metformin and macrophage

2.2.1 Metformin and inflammation

Chronic inflammation is an important pathological process in the onset and progression of AS [39]. Cholesterol metabolism homeostasis in macrophages can be disrupted in the process of lipid degradation [40]. The inflammatory response can be activated during the development of AS by free cholesterol (FC), which induces the formation of cholesterol crystals and, subsequently, the formation of foam cells [41]. Excessive chemotaxis retention factors released by foam cells caused pro-inflammatory cytokines and the inflammatory response. Plaque rupture can be caused by excessive inflammation and increases the risk of coronary thrombosis [42]. Currently, targeting cytokines (such as interlukin (IL)-1β, IL-1β; IL-17, and TNF) has been shown to slow the progression of CVD in some cases [43]. Treatment with metformin has been shown to be beneficial in the treatment of macrovascular complications in addition to its hypoglycemic effect. Metformin (0.09 U) can reduce the log-transformed neutrophil to lymphocyte-ratio after 8 to 16 months compared to sulfonylurea therapy. Metformin also inhibited blood cytokines, including the C-C motif chemokine 11 (CCL11) in a non-diabetic heart failure trial (registration: NCT00473876). Methionine increased the levels of homocysteine (Hcy), TNF-α, H₂S, and IL-1β in C57BL/6 mice while decreasing the level of cystathionine g-lyase (CSE). Hcy increased the expression of DNA methyltransferase and the methylation of the CSE promoter in macrophage, whereas metformin treatment can reduce the deleterious effects of methionine [44]. Phosphatase and tensin homolog (PTEN)-deficient macrophages create a persistent inflammatory microenvironment in which they induce nitric oxide synthase (iNOS)/NO and cyclooxygenase-2 (COX)-2/Prostaglandin E2 (PGE2), which can be inhibited by metformin by inhibiting reactive oxygen species (ROS) production and Akt activation [45]. These anti-inflammatory effects of metformin are also the result of AMPK activation, but even after treatment with compound C, an AMPK inhibitor, residual metformin activity remains significant [46]. Our previous study also demonstrated that metformin inhibits NOD-like receptor thermal protein domain associated protein 3 (NLRP3) inflammasome and suppresses DM-accelerated AS in ApoE^−/− mice through AMPK activation [47]. Metformin treatment of mouse or human alveolar macrophages prevents particle-induced production of complex III mitochondrial ROS, thereby inhibiting calcium release-activated channel (CRAC) activation and IL-6 release [48]. These studies suggest that metformin reduces oxidative stress in macrophages.

2.2.2 Metformin and foam cell formation

Foam cells are an important element of atherosclerotic plaque. Macrophages absorb modified plasma-derived lipoproteins, leading to the formation of lipid-filled foam cells and atherosclerotic lesions [39]. Metformin inhibits the accumulation of cholesterol in macrophages under various inducible conditions. The ATP-binding cassette transporter G1 (ABCG1)-mediated cholesterol efflux cannot effectively regulate glycated high-density lipoprotein (HDL) particles. Metformin can restore cholesterol efflux mediated by glycated HDL [49]. This is also observed in oxidized low-density lipoprotein (oxLDL)-stimulated cholesterol accumulation and in foam cell formation, which may be associated with ABCG1 up-regulation. Metformin also reduced the rate of cholesterol biosynthesis from acetate in the J774A.1 macrophage cell line [50]. Metformin inhibits foam cell formation and reduces the expression of adipogenic differentiation-associated protein (ADRP) in LPS-induced THP-1 macrophages [51].

In addition, when metformin is combined with other drugs, its efficacy increases while adverse reactions are reduced. Coadministration of metformin and T317 (liver X receptor agonist) with an HFD alleviated the development of AS in ApoE−/− mice. This was due to a decrease in monocyte adhesion and macrophage cell proliferation, which was associated with an increase in ATP-binding cassette transporter A1(ABCA1)/ ABCG1 expression [52].

2.2.3 Metformin and macrophage polarization

In the pathological process, tissue injury and pro-inflammatory cytokines secretion promote the differentiation of monocytes into macrophages and further exacerbate inflammation [53]. M1 macrophages are traditionally involved in pro-inflammatory and bactericidal activity, whereas M2 macrophages produce anti-inflammatory mediators and growth factors in response to the resolution of inflammation [54]. In a clinical study involving 30 healthy volunteers and 30 obese volunteers (20 with diabetes), the obese volunteers were treated with metformin and peripheral blood mononuclear cells (PBMCs) were isolated to measure polarization markers. The results revealed that the level of the CD68 marker was increased in diabetic patients, while CD11b, CD11c, CD163, and CD169 were decreased. The elevated expression levels of TNF-α, iNOS, IL-6, CD16, CD36, and CD206 indicated the presence of macrophages with an M1-like phenotype. After treatment with metformin, the levels of TNF-α, iNOS, IL-6, CD11c, CD36, CD169, and CD206 showed no significant difference between healthy volunteers and diabetic patients. These results suggest a close relationship between metformin treatment and distinct patterns of phenotypic markers [55]. Metformin treatment decreased not only M1 macrophage marker levels (MCP1 and CD11c) but also blood levels of TNF-α and IL-6 in male C57/6 J mice fed an HFD for 7 weeks. Therefore, metformin-regulated AMPK may be involved in M2 phenotype macrophages and decreases inflammation in obesity [56]. Chronic metformin administration also promoted microglia/macrophage M2 polarization by regulating AMPK, which may improve functional recovery after middle cerebral artery occlusion (MCAO) in mice [57]. Metformin-activated AMPK also induced macrophage M2 polarization and inflammatory cytokines expression, alleviating the formation of atherosclerotic plaque [57].

Taken together, metformin has been shown to have a great anti-AS effect by regulating different molecular signaling pathways related to vascular homeostases, such as increasing NO production, maintaining endothelial integrity, decreasing oxidative stress, preventing M1 type polarization, inhibiting inflammation, and so on (Figure 1).

Figure 1.

Schematic of the molecular signaling pathways involved in anti-AS effect of metformin.

3. Metformin and diabetic cardiomyopathy

Diabetic cardiomyopathy (DCM) is typically characterized by left ventricular (LV) dysfunction and the absence of a history of coronary artery disease or hypertension [58, 59]. The molecular mechanism initiating DCM is unclear, but the major clinical and biochemical dysfunctions in DCM development have been identified, such as hyperglycemia, systemic insulin resistance, and impaired cardiac insulin signaling [60, 61, 62]. Emerging evidence highlights the significance of abnormal mitochondrial function and energy metabolism in causing pathological remodeling of the cardiac structure [63, 64].

Metformin has been shown to be cardioprotective in ongoing clinical and basic research and it has been recommended as first-line therapy in diabetic patients with heart failure (HF) based on clinical practice guidelines [65]. Numerous studies have shown that monotherapy or combination therapy with metformin reduces mortality and/or hospitalizations in patients with diabetes and/or HF [66]. Similarly, a 10-year follow-up study revealed that metformin reduced the incidence of myocardial infarction from all causes [67]. In basic research, it was found that metformin protects against oxidative stress [68, 69]. It attenuates pro-inflammatory responses induced by LPS or oxidative stress [70, 71]. However, there are contradictory analyses of these positive reports as well [72, 73].

3.1 Metformin and free fatty acid (FFA) metabolism

In diabetic patients, increasing FFA released from adipose tissue leads to an increase in circulating levels, especially in patients with visceral adiposity. Abnormal circulating levels of FFAs cause a shift from glucose to FFA uptake and utilization, impairing myocardial energy metabolism [74, 75]. This elevated FFA in circulation also increases the expression of the nuclear receptor peroxisome proliferator-activated receptor-α (PPAR-α) [76]. Activated PPAR-α improves the expression of FFA oxidation-related genes (pyruvate dehydrogenase kinase 4, PDK4; CD36), thereby increasing mitochondrial FFA uptake [77]. Myocardial FFA uptake that exceeds its FFA β-oxidation capacity results in lipid accumulation [78]. This metabolic disorder protects against subsequent oxidation and toxic lipid metabolites at first, but this metabolic imbalance eventually leads to mechanical dysfunction and organ failure. Lipid accumulation also promotes the production of toxic lipid intermediates (diacylglycerol, DAG; ceramides) which contribute to oxidative stress and cardiomyocyte apoptosis. Research has shown that lipid accumulation in myocytes is closely related to the severity of ischemia–reperfusion (IR) and CVD [79, 80].

Metformin treatment was associated with a decrease in fasting plasma glucose (P < 0.05), insulin (P < 0.05), triglyceride [TG] (P < 0.05), and FFA (P < 0.03) levels, according to a 2004 study involving 120 overweight T2D patients treated with placebo + diet (n = 60) or metformin (850 mg twice daily) + diet (n = 60) for 4 months [81]. Jeppesen also demonstrated that metformin treatment lowers postprandial glucose, insulin, FFA, and TG levels (P < 0.001) in patients with non-insulin-dependent DM (NIDDM) [82]. During the early stages of hypertension, spontaneously hypertensive rats (SHR) exhibited myocardial metabolic changes including FFA. Metformin-treated SHR exhibited normalization of FFA levels, mean arterial blood pressure, cardiac glucose uptake rates, left ventricular mass/tibia length, and wall thickness. Metformin may exert its effects mechanically by increasing fatty acid oxidation and decreasing oxidative stress by activating AMPK [83]. These findings suggest that the cardioprotective effect of metformin is attributable to its ability to lower FFA levels.

3.2 Metformin and AGE

The nonenzymatic glycation and oxidation of proteins and lipids result in the formation of AGE, and this is an important pathogenetic change in DCM [84]. AGE can covalently bind to each other or to extracellular proteins, such as collagen, laminin, and elastin, rendering the AGE complex vulnerable [84]. This can impair the degradation of collagen leading to collagen accumulation and fibrosis, which causes local tissue stiffness and increases myocardial stiffness [85].

In a clinical trial, metformin (1700 mg daily) was administered to 15 T2D patients for 3 months; the results indicated that patients treated with metformin had lower levels of circulating AGE, which was associated with fasting plasma glucose and glycated hemoglobin after 3 months of treatment [86]. Although the exact molecular mechanism is unknown, metformin is thought to bind its guanidine group to the α-dicarbonyl group of methylglyoxal, which is an intermediate in the formation of AGE, leading to the neutralization of reactive dicarbonyls. Furthermore, metformin activates glyoxolase, promoting the conversion of methylglyoxal to D-lactate [87, 88].

3.3 Metformin and mitochondrial dysfunction

The heart is an energy-dependent organ with abundant mitochondria that provide cellular energy through oxidative phosphorylation. In a diabetic model, abnormal dynamic mitochondrial structure was observed, as well as impaired function and decreased glutathione concentration in mitochondria, all of which promote DCM [89]. On the other hand, proteomic research proteins from mouse cardiomyocytes of the type 1 diabetes model revealed that a large portion of the altered proteins was related to mitochondrial origin and FFA oxidation [90]. This decrease in mitochondrial respiration and expression of the proteins involved in oxidative phosphorylation was also observed in an obese T2D mouse model, suggesting that impaired mitochondrial oxidative capacity and reduced ATP production promote cardiac dysfunction and subsequently limit myocardial contractility [91]. A recent study by Anderson et al. indicates that reducing mitochondrial dysfunction and oxidative stress in diabetic patients may promote the pathogenesis of HF [92].

Metformin has been reported to have a potent effect on improving mitochondria function and regulating the metabolism of cardiomyocytes in a failing heart. Metformin-induced activation of AMPK upregulated the expression of eNOS and peroxisome proliferator-activated receptor-gamma coactivator 1-alpha (PGC-1α), which are important in regulating the biogenesis and function of mitochondria [93, 94]. Metformin increased the expression of stimulated NADH dehydrogenase (ubiquinone) alpha subcomplex-1 (NDUFA1), NDUFA2, NDUFA13, and manganese superoxide dismutase, indicating that it promotes mitochondrial biogenesis under hyperglycemic conditions, as well as mitochondrial biogenesis-related transcription factors (nuclear respiratory factor; NRF-1 and NRF-2). Metformin also upregulated mitochondrial NDUFA13 protein expression by activating AMPK signaling [95].

3.4 Metformin and oxidative stress

The intracellular imbalance between ROS production and antioxidant regulation leads to oxidative stress. The pathological process in both type 1 and type 2 rodent diabetic models is associated with excessive ROS production in the mitochondria [96]. Increased ROS production from both non-mitochondrial (activating NADPH oxidase) and mitochondrial (modulating mitochondrial electron chain to generate superoxide) sources as a result of various risk factors may lead to cardiac dysfunction by directly damaging proteins or DNA or inducing apoptosis [97]. It is notable that a reaction between ROS and NO produces peroxynitrite species, which triggers a range of pathogenic events including impaired coagulation and inflammation [98]. The progression of oxidative damage and antioxidant defect leads to a number of responses that promote ventricular remodeling, including cardiomyocyte hypertrophy [99, 100].

Metformin has been studied for its antioxidative capacity in animal models. Previously, activation of protein phosphatase 2A (PP2A) was shown to attenuate myocardial cell apoptosis in vitro under high glucose conditions by deactivating NF-κB signaling [101]. Metformin has been shown to reduce ROS production, pro-inflammatory mediator release, and NF-κB signaling in human and rat primary cardiomyocytes in response to high glucose levels by activating PP2A [102]. Metformin has been reported to prevent mitochondrial dysfunction, alleviate intracellular oxidative stress, and prevent the loss of aconitase activity that was linked to the development of diabetes in the aorta and kidney of Goto-Kakizaki rats. However, this effect was not observed in the heart during hyperglycemia or after administration of metformin [103].

3.5 Metformin and calcium homeostasis

Ca²⁺ in cardiomyocytes is a crucial regulator of cardiac contractility. Diabetic myocytes exhibited abnormal protein modifications, including a shift in myosin isoenzyme composition (from V1 to V3 isoforms) and an abnormal predominance of the β myosin heavy chain (MHC), which resulted in a decrease in contractile protein Ca-ATPase activity and shortening velocity [104]. The diabetic heart was also observed to have abnormal intracellular Ca²⁺ handling during the contractile cycle, as a result of ryanodine receptor-mediated Ca²⁺ release reduction from the sarcoplasmic reticulum (SR) and a decrease in the upstroke phase of the Ca²⁺ transient in streptozotocin-induced diabetic rats. This may result in cardiac dysfunction [105]. Ca²⁺ dysregulation was determined in human cardiomyocytes from T2D patients, and decreased Ca²⁺ sensitivity in myofilaments was observed [106]. The sarco- (endo-) plasmic reticulum Ca²⁺ ATPase (SERCA) 2 is a key enzyme in the regulation of Ca²⁺ localization in SR and has a close relationship with contractile dysfunction. Decreased activity of SERCA was observed in cardiomyocytes incubated in high-glucose medium, in which abnormal Ca²⁺ signaling has been reported [107]. There is also a hypothesis that the slower transient Ca²⁺ kinetics in the diabetic heart results from a slower action potential and reduced SERCA2a expression [108]. The dysfunctional calcium homeostasis and its clinical consequences require further confirmation.

Metformin was reported to have a novel therapeutic perspective role in regulating abnormal mitochondrial Ca²⁺ content in dystrophin-deficient mice, which is a mitochondrial dysfunction model, and mitochondrial Ca²⁺ uptake kinetics significantly increased the mitochondrial Ca²⁺ content. In metformin-treated cardiomyocytes from 3-month-old dystrophin-deficient mice, decreased mitochondrial Ca²⁺ content and increased complex I-driven respiration were observed [109]. However, there is no direct evidence that metformin regulates Ca²⁺ homeostasis in cardiomyocytes.

In conclusion, Metformin alleviates DCM development in different approaches, such as promoting FFA oxidation, decreasing FFA and AGE levels, reducing toxic lipid metabolite, and protecting against oxidative damage (Figure 2).

Figure 2.

Schematic of the cardioprotective mechanisms of metformin on DCM.

4. Metformin and cardiac ischemia: Reperfusion injury

In 1998, the UKPDS study reported that metformin treatment was associated with a 39% reduction in the risk of myocardial infarction (MI), and a 36% and 42% reduction, respectively, in 10-year overall mortality and diabetes-related mortality, compared to insulin or sulfonylureas (SUDs) therapy [6, 67, 110]. According to data from the reduction of atherothrombosis for continued health registry, metformin use is associated with a 24% reduction in all-cause mortality [111]. In a meta-analysis of observational studies, Pladevall et al. found that metformin is associated with a lower risk of MI when compared to SUDs [112]. In the rat models of MI, metformin administration was found to reduce infarct size, preserve myocardial function, and attenuate myocardial remodeling [113, 114].

In a study conducted in 1988, it was demonstrated that oral administration of metformin reduces infarct size 48 h after coronary artery ligation in a mammalian model [115, 116]. Solskov et al. found that a single dose of metformin (250 mg/kg, orally) 24 h before coronary occlusion reduced infarct size in the Langen-dorff isolated rat heart perfused model [113]. Research has shown that a much lower dose of metformin (125 ug/kg) administered intraperitoneal 18 h before ischemia reduced infarct size by more than 50% [117]. In the other isolated working rat heart model, metformin treatment appeared to promote the recovery of contractile function before ischemia and during reperfusion. This finding reflects an effect on cardiac pumping [118].

Clinical and animal studies have shown that metformin has an effective role in ischemia–reperfusion injury, and mechanical studies have provided a new perspective on metformin. Calvert et al. reported that phosphorylation of AMPK occurs after the onset of myocardial ischemia and remains active for 24 hours or more following reperfusion in a murine model of coronary artery occlusion. In rat hearts exposed or not to ischemia, a very low dose of metformin augmented the phosphorylation of AMPK. This implies that the beneficial effect of metformin on ischemia–reperfusion injury is AMPK-dependent. Recent research has shown that metformin-induced eNOS phosphorylation activates AMPK, which is also required for cardioprotective action [117, 119]. One of the first things that happen is mitochondrial dysfunction, which leads to regulated cell death, oxidative stress, and the release of inflammatory cytokines. Mitochondrial dysfunction is caused by an abnormal opening of the mitochondrial permeability transition pore (mPTP), which is also seen in ischemia–reperfusion injury. Immediately after reperfusion, the mPTP opening causes the release of pro-apoptotic factors, such as cytochrome C [120, 121]. It has been reported that metformin activates several kinases of the Reperfusion Injury Salvage Kinase (RISK) pathway like phosphatidylinositol-3-kinase (PI3K) and Akt, thereby restoring abnormal opening of the mPTP (Figure 3). In a mammalian study (nondiabetic Wistar rats and type 2 diabetic Goto-Kakizaki rats), metformin upregulated the level of Akt phosphorylation after reperfusion, while concurrent administration of a PI3K inhibitor prevented Akt phosphorylation and abrogated the protective effect of metformin [122].

Figure 3.

Schematic of the signaling pathways of metformin in regulating eNOS level and mPTP.

5. Metformin and heart failure

The therapeutic effects of metformin in HF appear beneficial based on clinical trials, as it reduces all-cause mortality and improves cardiac function, as confirmed by the American diabetes association [65]. In an observational study involving 10,920 HF patients taking metformin, sulfonylureas, and/or insulin, the sulfonylurea monotherapy group served as a reference group. This study demonstrated that metformin in monotherapy and in combination with sulfonylureas was associated with a reduction in all-cause mortality compared to sulfonylureas in monotherapy; These findings were consistent with those of a separate analysis of patients treated with or without insulin [72]. Another observational study involved 16,417 T2D patients discharged with a major diagnosis of HF. These patients were treated with thiazolidinediones (n = 2226), metformin (n = 1861), or both thiazolidinediones and metformin (n = 261). Although patients treated with metformin had a lower risk of hospitalization for HF, there was no significant difference in all-cause hospitalization between the three groups [9]. In an analysis of the PL-ASC registry (Polish registry of acute coronary syndromes), admitted diabetic patients with acute coronary syndrome who underwent percutaneous coronary intervention (PCI) were accessed. Two groups of patients were compared: those who used metformin and those who did not. Therefore, metformin treatment has been shown to benefit HF patients and discharges [123].

In vivo or in vitro basic research also determined the therapeutic role of metformin in cardiomyopathy, which may lead to HF progression. Metformin has been reported as an anti-inflammatory agent and an antioxidant in cardiomyopathy via AMPK-dependent and AMPK-independent mechanisms. Metformin inhibited high glucose-induced oxidative stress in H9C2 cardiomyocytes by promoting mitochondrial biogenesis, mitochondrial biogenesis-related transcription factors, and AMPK activation [95]. Yang et al. demonstrated that metformin improved diabetic heart function by inhibiting the NLRP3 inflammasome; however, AMPK inhibitors can counteract this effect. Furthermore, metformin inhibited NLRP3 inflammasome-induced cell death by inhibiting mTOR and promoting autophagy [124]. Lai et al. established the 2-hit pulmonary hypertension associated with heart failure with preserved ejection fraction (PH-HFpEF) model (double-leptin receptor defect (obese ZSF1) with the combined treatment of vascular endothelial growth factor receptor blocker SU5416) in rats with multiple features of metabolic syndrome. The glucose-lowering effect of the agent was abrogated in Sirtuin3 (SIRT3)-deficient human skeletal muscle cells and in SIRT3 knockout mice fed an HFD. Additionally, oral metformin treatment on model rats reduced right ventricular systolic pressure and media well thickness, which was associated with activated AMPK and SIRT3 levels. These findings indicate that metformin normalizes PH-HFpEF in a rat model involving AMPK and SIRT3 activation [125]. Hence, these findings demonstrated the role of metformin diabetic cardiomyopathy.

6. Conclusion

Over the course of 50 years of clinical experience, metformin has been well-established as the first-line hypoglycemic drug for the management of diabetes. Metformin inhibits hepatic gluconeogenesis to exert its antihyperglycemic effect. Metformin inhibits the respiratory-chain complex 1, resulting in a decrease in energy charge and a subsequent reduction in hepatic glucose output. These lipid-lowering effects and improvements in insulin sensitivity are primarily dependent on the AMPK signaling pathway.

It is evident that metformin plays a significant role in diabetic cardiomyopathy at various sites. Metformin promotes FFA oxidation and decreases FFA levels, thereby preventing oxidative stress and toxic lipid metabolites. Metformin decreases AGE levels and may alleviate collagen accumulation and fibrosis, which subsequently reduces myocardial stiffness. Importantly, metformin restores mitochondrial function, decreasing intracellular ROS levels and oxidative stress. The chronic administration of metformin to patients has cardioprotective effects, reducing the risk of heart failure and myocardial ischemia–reperfusion injury. Various molecular mechanisms are implicated in the beneficial effects of metformin, such as reducing oxidative stress and cell death via AMPK-dependent and AMPK-independent pathways. Metformin influences vascular physiology by increasing NO production, maintaining endothelial integrity, decreasing oxidative stress, and inhibiting inflammation, all of which contribute to its anti-AS effect. Metformin is regarded by clinicians as an established drug with a significant role in the treatment of T2D. However, there are still many significant unanswered questions from a mechanistic perspective. Understanding the pharmacological mechanisms of metformin may aid in the treatment of not only diabetes but also cardiovascular diseases.

Acknowledgments

This work was supported by the National Natural Science Foundation of China (grant numbers 81873514 and 82170357) and Natural Science Foundation of Guangdong province (grant numbers 2021A1515011766).

Conflict of interest

The authors declare no conflict of interest.