MicroRNA-mediated control of myocardial infarction in diabetes

Abstract

Diabetes mellitus is a global public health problem whose cases will continue to rise along with the progressive increase in obesity and the aging of the population. People with diabetes exhibit higher risk of cardiovascular complications, especially myocardial infarction (MI). microRNAs (miRNAs) are evolutionary conserved small non-coding RNAs involved in the regulation of biological processes by interfering in gene expression at the post-transcriptional level. Accumulating studies in the last two decades have uncovered the role of stage-specific miRNAs associated with key pathobiological events observed in the hearts of people with diabetes and MI, including cardiomyocyte death, angiogenesis, inflammatory response, myocardial remodeling, and myocardial lipotoxicity. A better understanding of the importance of these miRNAs and their targets may provide novel opportunities for RNA-based therapeutic interventions to address the increased risk of MI in diabetes.

Introduction

Diabetes mellitus, affecting over 500 million people worldwide, is a major public health problem characterized by high blood glucose due to lack of insulin production from pancreatic β-cells (type 1 diabetes mellitus) or a reduced insulin-stimulated response to glucose disposal in the tissues or insulin resistance (type 2 diabetes mellitus (T2D)). This chronic disease is associated with a 2- to 4-fold increased risk of cardiovascular complications, especially myocardial infarction (MI), the leading cause of death among patients with diabetes mellitus (1). Cardiovascular complications in patients with diabetes are mainly attributed to atheroma plaque formation and ventricular dysfunction (2). In addition, different structural and functional changes in the heart of patients with diabetes have established diabetic cardiomyopathy as a distinct entity independent of other cardiac risk factors. Increased cardiac hypertrophy and fibrosis are structural changes observed in people with diabetes, and these changes are associated with increased oxidative stress, cardiomyocyte death, and impaired angiogenesis and microvascular function (3). However, the pathophysiological mediators underlying the heart injury linked to insulin resistance and diabetes remain poorly understood.

Regulatory non-coding RNAs are key players in the regulation of most biological processes. Among them, microRNAs (miRNAs) are small (~ 20–22 nt) evolutionarily conserved single-stranded non-coding RNAs that regulate gene expression at the post-transcriptional level. The miRNAs bind to target mRNAs through base complementarity with their 3’UTR region, causing the inhibition of their translation or promoting their degradation. It is estimated that miRNAs regulate around 60% of mRNAs in humans (4). To date, over 2000 miRNAs have been identified, whose function has been related to both physiological and pathophysiological conditions. Emerging studies reveal distinct roles for miRNAs in the regulation of insulin resistance, diabetes, and its cardiovascular complications (5, 6). Indeed, expression analysis of diabetic heart biopsies demonstrate a dysregulated miRNA profile that is implicated in the mechanism of the disease (7).

Here, we summarize the role of miRNAs in specific aspects implicated in cardiac damage in diabetes and myocardial infarction, such as cardiomyocyte death, inflammation, angiogenesis, myocardial remodeling, and myocardial lipotoxicity. Understanding the implication of these miRNAs will not only shed new insights in the pathobiology underlying the increased risk of MI, but also may provide novel opportunities for therapeutic intervention.

Cardiomyocyte death

Patients with diabetes show increased rates of cell death in the heart compared to non-diabetic patients (8). The initiation of signaling through cell death pathways, including necrosis (non-programmed cell death) and apoptosis (programmed cell death), observed in the diabetic heart has been attributed to different events such as hyperglycemia, fatty acids, oxidative stress, and mitochondrial dysfunction (9). In addition, myocardial ischemic injury involves activation of cell death pathways in response to impaired coronary perfusion, including apoptosis and necrosis but also triggers autophagy and pyroptosis pathways. Autophagy, which has been commonly associated to a cell survival mechanism by the degradation of cell components, is considered cardioprotective since its inhibition increase infarct zone in MI models. However, it overactivation of autophagy signaling can also lead to programmed cell death (10). Pyroptosis, a caspase 1-dependent cell death, is another mechanism involved after ischemia-induced cardiomyocyte injury and is associated with the inflammatory response (11). Different miRNAs such as miR-21, miR-1, miR-133, miR-17, miR-302, miR-30 have been implicated in cardiomyocyte death.

miR-21, one of the most abundantly expressed miRNAs, is involved in many pathological processes, including diabetes and CVD. Different experimental models of heart ischemic disease showed increased expression of miR-21 in the infarcted area, and has been related to cardiac cell survival, hypertrophy, and myocardial fibrosis. In addition, decreased miR-21 expression is found in cardiomyocytes of experimental models of diabetes (12). miR-21 reduction in db/db mice hearts increased its target gelsolin (Gsn), an actin binding protein that acts as a transcription factor, and is associated with cellular oxidative stress, cardiac hypertrophy, and impaired cardiac dysfunction (12). In addition, Gsn induces cardiomyocytes apoptosis after MI since apoptotic cells are reduced in the infarct zone of Gsn -KO mice after LAD ligation (13), reducing cardiac remodeling after injury. Other studies have also described the pro-survival effect of miR-21, since locally delivered antagomiR-21 reversed protective effects of ischemic postconditioning by preventing myocardial apoptosis, infarct size, and ventricular function via its target genes programmed cell death 4 (PDCD4) (14) and PTEN (15).

miR-1 and miR-133, which are highly abundant in the heart, are transcribed from the same loci, and has been related to heart development and function (16). While circulating miR-1 and miR-133 expression is increased in patients with myocardial damage, their expression is decreased in the infarcted myocardium (17). In addition, miR-1 and miR-133 expression are decreased in heart tissue in different experimental. models of diabetic mice (18, 19). Both miRNAs are oxygen and glucose sensitive. Indeed, miR-1 and miR-133a expression is down-regulated in cardiomyocyte under hypoxic conditions. However, it is interesting to note that miR-133a and miR-1 may have opposite roles controlling cell death, since miR-133 over-expression suppresses hypoxia-induced apoptosis by targeting TAGLN2 (20), while miR-1 over-expression reduced its target NOTCH3 promoting apoptosis and reducing autophagy (21). Moreover, an inverse correlation among miR-133a expression and autophagy markers has been described in the left ventricle of diabetic heart failure patients (22). Glucose stimulation up-regulated miR-1 expression in cardiomyocytes both in vitro and in streptozotocin (STZ)-stimulated mice increasing apoptosis through directly targeting Hsp60 expression (23). Finally, plasma levels of miR-1 and miR-133a have been proposed as biomarkers for the diagnosis of cardiac-related complication in patients with type 2 diabetes (24).

The miR-17–92 cluster, which is known to play a role in cardiac development (25) and angiogenesis (26), regulates post-MI injury and cardiac function in rapamycin-treated diabetic mice (27). Indeed, high-fat diet fed cardiac-specific miR-17–92-cluster deficient mice showed enhanced infarct size after I/R injury. Mechanistically, researchers described a STAT3-dependent induction of miR-17–92 cluster that regulates cardiomyocyte apoptosis by targeting the pro-apoptotic protein prolyl hydroxylase (Egln3/PHD3), modifying infarct size and cardiac function (27).

miR-302a expression is suppressed in post-ischemic alloxan-induced diabetic rabbit mice and in cardiomyocytes under simulated ischemia and high glucose conditions (28). Moreover, injection of miR-302 mimic conjugated with hyaluronic acid hydrogel to the heart promoted cardiomyocyte proliferation in the border zone after MI injury, improving cardiac function (29). Mechanistically, changes in miR-302 expression due to miRNA delivery led to decreased expression of its targets Lats2, Mob1, and Mst1, all components of the Hippo signaling pathway (29). In addition, Samidurai et al. described an increased expression of the miR-302a target PTEN in the heart after ischemia reperfusion. miR-302a mimic transfection reduced PTEN expression and cell death in pluripotent-stem-cells-derived cardiomyocytes under hyperglycemic-ischemic conditions. miR-302a expression was also found regulated by mTOR activity, since rapamycin-treated diabetic heart following ischemia reperfusion injury showed restoration of miR-302a and reduced infarct size, suggesting a cardioprotective role to miR-302a (28).

Another miRNA whose expression levels are reduced in patients with T2D is miR-24, an insulin-sensitive miRNA. Using different experimental mouse models, Wang et al. demonstrated the cardioprotective role of miR-24. For example, miR-24 overexpression reduced infarct size after myocardial ischemia/reperfusion injury in diabetic mice. Similar results were observed in a cardiomyocyte-specific knock-in miR-24 transgenic mouse, suggesting a cardiac-specific effect. Mechanistically, the authors propose OGT (O-GlcNAc transferase), ATG4A (an autophagy-related protein), and BIM (apoptosis-related protein) as miR-24 targets (30). Indeed, inhibition of BIM expression by miR-24 reduced cardiomyocyte apoptosis and infarct size in a mouse model of MI (31).

In addition, other miRNAs including miR-103/107 and miR-2861, have been associated with the necrosis pathway in ischemic heart disease and diabetes. miR-103/107 are found to be negative regulators of insulin sensitivity and involved in cardiac function (32). miR-103/107 expression is elevated in the infarct zone of the ischemic heart. Down-regulation of these miRNAs attenuated myocardial necrosis and cardiac function in an ischemia/reperfusion mouse model. Mechanistically, miR-103/107 regulates the necrosis pathway by targeting Fas-associated protein with death domain (FADD), a negative regulator of programmed necrosis (33). On the other hand, miR-2861 participates in cardiomyocyte necrosis through targeting the adenine nucleotide translocase 1 (ANT1), which is involved in mitochondrial function. miR-2861 inhibition increased ANT1 expression in the heart and reduced myocardial necrosis in the ischemic heart (34). In addition, cardiomyocyte-specific overexpression of ANT1 prevented cardiomyopathy in STZ-induced diabetic mice (35).

Finally, the glucose-sensitive miR-30d, whose expression is increased in rat model of diabetic cardiomyopathy, promoted cardiomyocyte pyroptosis (36), a programmed cell death associated with pro-inflammatory cytokine release. Elevated miR-30d expression is associated with caspase-1 activation and IL-1β and IL-18 expression in STZ-induced diabetic hearts. In addition, in vitro studies elucidated that high miR-30d-induced pyroptosis is mediated through downregulation of the apoptosis repressor with caspase recruitment domain (ARC) by directly targeting its transcription factor foxo3a in cardiomyocytes under hyperglycemic conditions. Conversely, MI-induced hypoxia in non-diabetic rats showed increased miR-30d expression in cardiomyocytes, protecting against apoptosis by the inhibition of MAP4K4 signaling (37).

Inflammation

Damaged myocardium triggers an inflammatory response with intracellular signaling and increased expression of adhesion molecules that results in the migration of macrophages, monocytes, and neutrophils into the infarct zone. Activation of resident macrophages in addition to the infiltration of other immune cells together with the stimulation of fibroblast activation and proliferation is responsible for the subsequent healing process and myocardial remodeling.

In addition to its role in cell survival and cardiac hypertrophy, miR-21 regulates inflammation after ischemic injury in the myocardium. miR-21 overexpression after MI is implicated in cardioprotective attenuation of inflammation. Indeed, miR-21 deficiency facilitates inflammatory cytokine production in macrophages, enhancing infarct size and cardiac dysfunction (38).

miR-146 is involved in many inflammatory diseases and an inverse association has been described between its expression and diabetes cardiovascular complications (39). Feng et al. observed that the levels of miR-146a are reduced in hearts of diabetic mice, and examined the role of endothelial miR-146a in mediating inflammation in diabetic heart by using endothelial-specific miR-146a transgenic mice, and demonstrated that the cardiac abnormalities in diabetic mice were improved in the diabetic mice overexpressing miR-146a. Mechanistically, endothelial miR-146a regulates the expression of inflammatory cytokines (IL-6, IL-1β, TNFα and MCP-1) in the diabetic hearts through directly targeting TRAF6 and IRAK1, two regulators of NF-κB pathway, which in turn is reflected in improved cardiac function (40).

Macrophages are important regulators of inflammation following myocardial injury, which are involved in the clearance of debris and in the tissue regeneration process. Although macrophage polarization has been described as a dynamic process, macrophages are classically divided into pro-inflammatory M1-like phenotype or anti-inflammatory M2-like phenotype. Different miRNAs have been related to macrophage polarization (41). Among them, miR-155 is up-regulated in inflammatory diseases and promoted M1-like polarization by regulating NF-κB, STAT, and PI3K/AKT pathways through its targets BCL6, SOCS1, and INPP5D, respectively (42), and is increased in cardiac tissue after myocardial infarction in db/db mice (43). The use of strategies to inhibit miR-155 expression has been proposed as a strategy to restore cardiac function in diabetic mice. For example, systemic delivery of antago-miR-155 conjugated with gold nanoparticles reduced miR-155 expression in macrophages and reduced M1/M2 ratio and cell apoptosis, improving heart function (44). In contrast, intramyocardial injection of bone marrow-derived progenitor cells, which has been described as cardioprotective after myocardial infarction injury, reduced miR-155 expression in the infarcted heart from diabetic mice, improving cardiac fibrosis and function (43).

Angiogenesis

After an ischemic event in the heart, the myocardium responds by inducing proangiogenic factors trying to restore blood flow within the infarcted zone. However, endothelial function is altered in patients with diabetes, which impairs this compensatory mechanism (45). Accumulating studies have identified miRNAs as regulators of signaling pathways implicated in diabetes-induced vascular impairment (46). It has been described that some of the angiogenesis-related miRNAs are down-regulated in the heart of patients with diabetes (47, 48). For example, the endothelial-enriched miR-126 is considered a master regulator of angiogenesis in both development and pathological processes. Its expression is markedly down-regulated in both circulation and myocardium of patients with diabetes (47). In addition, proangiogenic capacity of circulating endothelial progenitor cells is dependent on miR-126 expression and secretion, which is decreased in STZ-treated mice and patients with diabetes (48, 49), and mediated by its target Spred-1, an enhancer of the VEGF signaling pathway (49). Conversely, miR-195 expression is increased in the heart of STZ-induced diabetic mice. miR-195 knockdown by anti-miR-195 delivery restored coronary blood flow in diabetic mice, suggesting its role in the vascular dysfunction in diabetic heart. Mechanistically, anti-miR-195 administration increased Sirt1 and Bcl2 expression, both targets of miR-195, and reduced oxidative stress, apoptosis, and increase myocardial capillary density in diabetic hearts (50). In addition, Yan et al. (51) showed that injection of miR-17 antagomiRs abrogated infarct size in STZ-induced diabetic mice after MI surgery. In this study, authors showed how miR-17 inhibition increased the expression of VEGF, a direct target of miR-17, enhancing capillarity density in diabetic mice after MI.

Myocardial remodeling

Myocardial remodeling has been commonly divided into an early phase, which involves an initial enlargement of infarct size due to the activity of matrix metalloproteinases (MMPs) resulting in wall thinning and ventricular dilatation, and a late phase, in which different signaling pathways are induced to develop an adaptive response that leads to cardiomyocyte hypertrophy and formation of collagen scar. Some miRNAs have been described as crucial in the remodeling process of the heart under insulin resistance conditions, including miR-21, miR-133, and miR-134.

Although miR-21 triggers a pro-survival response after an ischemic event preventing heart hypertrophy, miR-21 response after I/R injury has been described as a “double-edge sword” (52). Indeed, miR-21 is protective not only in the myocardium but also in other tissues in the early-stages after ischemia by inducing pro-survival signaling. However, long-term actions of miR-21 can be detrimental by promoting. fibrosis, for example by targeting sprout homolog 1 (Spry1) (53). In the context of diabetes, miR-21 under high glucose conditions promoted proliferation and collagen production through targeting dual specific phosphatase 8 (DUSP8) in cardiac fibroblasts (54). Moreover, inhibition of miR-21 prevented perivascular fibrosis and myocardial function in STZ-induced diabetic mice. Mechanistically, miR-21 inhibition in STZ-injected mice increased SMAD7 expression in endothelial cells and suppressed endothelial-mesenchymal transition (EndMT), reducing collagen overproduction in diabetic cardiomyopathy (55). Because of these promising results, translational efforts are emerging that leverage the antifibrotic effects of antimiR-21 in ischemic hearts. Recently, Hinkel et al. described the first evidence of inhibition of remodeling-associated to miR-21 by intracoronary delivering of anitmiR-21 in pigs after left coronary artery occlusion (56).

In addition to its anti-apoptotic role, miR-133a has also anti-hypertrophic and anti-fibrotic functions. Decreased expression of miR-133 in the heart is found in the Ins2^+/− diabetic Akita mice that has been associated with heart dysfunction and cardiac remodeling. Indeed, restoration of miR-133 expression in this model attenuated diabetes-induced cardiac fibrosis, hypertrophy, and impaired contractility as demonstrated in miR-133a mimic-treated mice (19) and cardiac-specific miR-133a transgenic mice (57). Mechanistically, the role of miR-133a in attenuating diabetes-induced cardiac fibrosis has been related to its effect on the expression of some targets involved in the remodeling of extracellular matrix, such as connective tissue growth factor (CTGF) (58) and TGFβ1 (59).

miR-34 family members are senescence-related miRNAs whose expression is increased in the heart of patients with T2D (60). Circulating miR-34a expression positively correlated with HbA1c levels, suggesting glucose-dependent expression. Sirtuin 1 (Sirt1), a direct target of miR-34 associated with cell aging and glucose metabolism, was down-regulated in heart tissue. Indeed, miR-34 inhibition in the heart reduced cardiac hypertrophy in aged mice (61), and restored cardiac function by improving left ventricular remodeling after MI (62). In addition, glucose-induced miR-34 expression is reduced in bone marrow mesenchymal stem cells (BMSCs) exposed to resveratrol, a Sirt1 stimulator (63). In the same study, Zhang et al. also demonstrated that resveratrol-treated BMSCs injected intramuscularly improved cardiac function following infarction in diabetic rats.

Myocardial lipotoxicity

Finally, emerging studies highlight the contribution of lipotoxicity, or the accumulation of free fatty acids (FFA), to diabetic cardiomyopathy and myocardial infarction. Excessive FFA in diabetic hearts impairs the ability to switch energy source between FFA and glucose. Overreliance on FFA oxidation as the main source of energy is highly inefficient for ATP production compared to utilization of glucose leading to cardiac functional inefficiency and cardiac failure or adverse cardiac remodeling (64). miR-451 expression was found to be increased in hearts from mice with insulin resistance and in neonatal rat cardiac myocytes treated with palmitic acid (65). Cardiac-specific deletion of miR-451 ameliorated high-fat diet-induced cardiac hypertrophy, an effect mediated in part by the target calcium-binding protein 39 (Cab39), a scaffold protein of liver kinase B1 (LKB1) and was associated with increased expression of phosphorylated AMPK and reduced expression of mammalian target of rapamycin (mTOR). Another miRNA, miR-320, mediated diabetes-associated lipotoxicity and cardiac dysfunction by serving as an activating RNA in the nucleus thereby regulating fatty acid metabolic gene targets including CD36 (66). In addition, miR-30c overexpression by using an adenoviral associated viral system improved glucose utilization, reduced reactive oxygen species generation, myocardial lipid accumulation, cardiac apoptosis, and improved cardiac function in db/db mice in part by targeting PGC-1b. (67). Collectively, these studies raise the possibility that modulation of cardiac lipotoxicity holds promise for improving post-MI cardiac remodeling.

Challenges and Future Directions

As described above, several miRNAs highlighted in this review could be used as promising therapeutic tools in the management of the cardiometabolic diseases. The modulation of miRNA expression in preclinical models demonstrates the potential to improve heart function after MI in the context of insulin resistance and diabetes. However, challenges remain to translate these findings therapeutically to human patients. One of the major challenges in miRNA-based therapies is their delivery and appropriate distribution in a specific tissue (68). The two prominent modes of delivery are systemic (intravenous) and local (intracoronary or intramyocardial). Although local delivery would have the most direct effect on ischemic tissue after MI, systemic administration may be more beneficial when aiming to target inflammatory cells. For example, systemic delivery of endothelial-related miRNAs, such as miR-126, miR-146, or miR-195, are effective modulators of myocardial capillarity density or proinflammatory cytokines after cardiac injury. However, the need for serial administrations or accumulation of the delivered oligonucleotide-based products in non-targeted organs are limitations of the systemic delivery. Conversely, other strategies have been attempted to locally deliver miRNA mimics and inhibitors, including local injection of anti-miR-21 into the LAD using a balloon catheter approach, which resulted in accumulation throughout the left ventricular myocardium with sustained reduction of miR-21 in both cardiomyocytes and fibroblasts (53). Delivery of different combinations of miRNAs concomitantly should also be considered in future studies. Most of the studies presented in this review used liposomal-base techniques to deliver oligonucleotides targeting miRNAs. In this regard, despite cardiac-specific delivery, it has been reported that local administration of high concentrations of miRNA mimics or inhibitors can also affect lung and kidney (56), which may cause side effects in these tissues. In addition, other tissue-specific delivery modalities may open new avenues for miRNA-base therapies to treat patients, including nanoparticles with cell-specific ligands or adeno-associated virus (AAV) vectors with cardiac-specific promoters, thus reducing off-target effects. For example, based on the reported early stage overexpression of angiotensin II type 1 (AT1) receptor in cardiomyocytes after MI, Xue et al. developed nanoparticles anchored with AT1 peptides loaded with miR-1 inhibitor (69). Intravenously delivery of these nanoparticles selectively accumulated in the ischemic cardiac tissue and significantly reduced miR-1 expression, which attenuated cardiomyocyte apoptosis, infarct size, and cardiac function after MI. In another study, recombinant adeno-associated virus cardiomyocyte-specific promoter troponin T vectors were used to modify miR-320 expression in diabetic hearts which rescued the cardiac dysfunction in diabetic mice (66).

One of the features of miRNA regulation is their ability to bind to multiple targets. Indeed, each miRNA can regulate hundreds of targets, and often mRNAs regulate targets from the same pathway, increasing efficiency in modulating biological processes. On the other hand, different miRNAs can also target the same mRNAs, ensuring repression by this cooperative phenomenon. Thus, multiplicity and cooperativity of the miRNA-mediated regulation results in a redundancy that can buffer possible transcriptomic imbalance, thereby allowing fine-tuning of a specific pathway of interest. For example, miR-21 and miR-133 target key regulators of TGFβ signaling in cardiac fibroblasts. Therefore, administration of miRNA inhibitors against these miRNAs can have an additive effect on preventing cardiac fibrosis.

As we have described throughout this review, changes in miRNA expression modify cell behavior affecting different pathobiological aspects of the disease. In addition, miRNAs are released from the cell encapsulated within vesicles or associated to proteins and can be taken up by other cells, which has been described as mode of cell-to-cell communication, or “miracrine” effect. Although the full mechanism of miRNA release remains unclear, a recent study identified unique sequences involved in the retention or secretion of the miRNAs (70). This interesting finding provides insights for the miRNA sorting machinery and may provide new therapeutic strategies focused on the regulation of miRNA retention or release in order to improve clinical outcomes in patients.

miRNAs found in the circulation are also proposed as an accessible diagnostic tool. Accumulating efforts have been exploring the use of miRNA expression in pathological conditions in which the diagnosis remains a challenge. In a recent work, Blanco-Domínguez et al. identified a novel miRNA that can be used to identify patients with myocarditis (71), who are frequently misdiagnosed as patients with MI with non-obstructive coronary arteries. Currently, myocarditis diagnosis typically requires either an invasive biopsy or magnetic resonance imaging. The authors use experimental mouse models of myocarditis and myocardial infarction and performed miRNA profiling in Th17 cells, which are increased at the acute phase of myocarditis but not after MI, and identified mmu-miR-721 as a characteristic marker of myocarditis. The ability of the human homologue hsa-miR-Chr8:96 to discriminate patients with myocarditis from patients with MI was confirmed in four different cohorts. This finding raises the possibility of exploring miRNAs as new biomarkers for the diagnosis of cardiac-related complication in patients with diabetes, or even as prognosticators of increased cardiovascular risk post-MI.

Conclusion

The recent development and success of mRNA vaccines has drawn worldwide attention to the potential use of RNA-based therapies. Studies highlighted in this review demonstrate the importance of miRNAs in the different pathobiological processes observed in the hearts of people with diabetes and myocardial infarction, including cardiomyocyte death, angiogenesis, inflammation response, myocardial remodeling and myocardial lipotoxicity. Approaches that can leverage stage-specific miRNA targeting may provide a novel strategy to overcome the increased risk of MI in cardiometabolic disease.

Figure 1.

miRNAs and their associated targets involved in different pathobiological 502 processes important in myocardial infarction in diabetes, including cardiomyocyte death, angiogenesis, inflammation, myocardial remodeling, and myocardial lipotoxicity.