Overview of MicroRNAs in Cardiac Hypertrophy, Fibrosis, and Apoptosis

MicroRNAs (miRNAs) are non-coding RNAs that play essential roles in modulating the gene expression in almost all biological events. In the past decade, the involvement of miRNAs in various cardiovascular disorders has been explored in numerous in vitro and in vivo studies. In this paper, studies focused upon the discovery of miRNAs, their target genes, and functionality are reviewed. The selected miRNAs discussed herein have regulatory effects on target gene expression as demonstrated by miRNA/3′ end untranslated region (3′UTR) interaction assay and/or gain/loss-of-function approaches. The listed miRNA entities are categorized according to the biological relevance of their target genes in relation to three cardiovascular pathologies, namely cardiac hypertrophy, fibrosis, and apoptosis. Furthermore, comparison across 86 studies identified several candidate miRNAs that might be of particular importance in the ontogenesis of cardiovascular diseases as they modulate the expression of clusters of target genes involved in the progression of multiple adverse cardiovascular events. This review illustrates the involvement of miRNAs in diverse biological signaling pathways and provides an overview of current understanding of, and progress of research into, of the roles of miRNAs in cardiovascular health and disease.


Introduction
MicroRNAs (miRNAs) are small non-coding RNAs that have emerged as essential regulators for almost all aspects of intracellular signaling pathways in eukaryotic cells. MiRNAs originate from long primary transcripts (pri-miRNA) and are subsequently cleaved by the nuclear ribonuclease, Drosha, to form microRNA precursors (pre-miRNA) of approximately 70-nucleotides. Pre-miRNAs are exported to the cytoplasm and further processed into approximately 22-nucleotide double-stranded miRNAs by the endonuclease, Dicer. One of the strands, namely the guide strand or mature miRNA, binds to Argonaute proteins within the RNA-induced silencing complex (RISC). The RISC utilizes the guide strand to target the 3 1 end untranslated regions (3 1 UTR) of gene transcripts by Watson-Crick base pairing. The action of miRNAs usually results in downregulation of their target genes by triggering messenger RNA degradation or by inhibiting the translational machinery [1]. The importance of miRNAs in the cardiovascular system is demonstrated by the profound structural and functional abnormalities observed in cardiac-specific Dicer deletion in animal models [2][3][4]. For the past decade, studies using various cellular and animal disease and transgenic models have allowed identification of miRNAs and their target genes, in various cardiovascular disease states.

Hypertrophy-Related miRNAs
Biomechanical stress and other pathological stimuli potentially trigger cardiac remodeling manifested as cardiac hypertrophy with an increase in myocyte size and myofibrillar volume, but not myocyte cell number facilitating sustained cardiac output in the face of functional insult. However, whilst such an adaptive mechanism is beneficial in the short term, prolonged hypertrophy has potentially disastrous consequences leading ultimately to heart failure and death. Studies profiling dysregulated miRNA expression in animal models of cardiac hypertrophy induced by thoracic-aortic banding or via constitutively activated calcineurin signaling have revealed both up-and down-regulated miRNAs. Functional analyses of these dysregulated miRNAs have demonstrated that miRNAs may exert either positive or negative regulatory effects on cardiac hypertrophic pathways.
The role of one of the most abundant cardiac miRNAs, miR-1, in hypertrophy was suggested by the inverse relationship between its expression and progression of cardiac hypertrophy in a transverse aortic constriction (TAC) rodent model. Gain-of-function miRNA studies in neonatal myocyte culture demonstrated downregulation of several hypertrophic target genes of miR-1 predicted in silico, including Ras GTPase-activating protein (RasGAP), cyclin-dependent kinase 9 (Cdk9), Ras homolog enriched in brain (Rheb), and fibronectin [9]. Further studies demonstrate that miR-1 prevents cardiac hypertrophy by suppressing heart and neural crest derivatives expressed 2 (Hand2), and by inhibiting the activity of insulin-like growth factor (Igf1) and extracellular matrix remodeling factor, twinfilin 1 (Twf1) [10,11]. miR-101 was shown to regulate the expression of ras-related protein-1A (Rab1A) in the TAC-induced hypertrophy rodent model and in angiotensin II induced cellular hypertrophy. Antagonism of miR-101 attenuated the hypertrophic phenotype in both models [12]. The role of miR-133 in regulating cardiac hypertrophy was demonstrated by the induction of cardiac hypertrophy in miR-133 null mice. Molecular investigations revealed that miR-133 targets multiple anti-hypertrophic genes, including guanosine triphosphate-guanosine diphosphate (GDP-GTP) exchange protein, Rhoa, signal transduction kinase cell division control protein 42 (Cdc42), and the nuclear factor, negative elongation factor complex member A (Nelfa/Whsc2) [13]. miR-145 negatively regulates the expression of GATA binding protein 6 (GATA6) and subsequently attenuates hypertrophy in isoproterenol-induced cellular hypertrophy models [14]. Cardiomyocyte hypertrophy induced by high glucose is modulated by miR-150 via its regulatory effect on the expression of the transcriptional co-activator, p300 [15]. Downregulation of miR-185 in a TAC-induced hypertrophy rodent model was found to produce anti-hypertrophic effects by targeting multiple genes in calcium signaling pathways, including calcium/calmodulin-dependent protein kinase II δ (Camk2d), sodium/calcium exchanger 1 (Ncx1), nuclear factor of activated T cells, cytoplasmic, and calcineurin dependent 3 (Nfatc3). miR-185 also attenuates endothelin 1-induced hypertrophy in neonatal rat myocytes, as evidenced by reductions in cellular size and expression of hypertrophic markers [16]. miR-223 antagonizes hypertrophy, induced by endothelin-1 or TAC, through targeting of cardiac troponin I (cTnI)-interacting kinase (Tnni3k) [17]. In murine models, miR-26 inhibits expression of its predicted target gene, GATA4, in TAC-induced mouse cardiac hypertrophy [18]. Similarly, down-regulation of miR-34a was observed in angiotensin II-induced hypertrophy in myocardial cells resulting in upregulation of the expression of an autophagy related gene, Agt9a [19]. Another anti-hypertrophic miRNA, miR-378, controls cardiac hypertrophy in the TAC model via repression of mitogen activated protein kinase (MAPK) signaling by targeting Mapk1, insulin-like growth factor receptor 1 (Igf1r), growth factor receptor-bound protein 2 (Grb2), and kinase suppressor of ras 1 (Ksr1) [20]. miR-9 opposes cardiac hypertrophy by targeting myocardin and reducing expression of nuclear factor of activated T cells c3 (Nfatc3). Overexpression of miR-9 suppresses cardiac hypertrophy in animal models [21]. Similarly, miR-98 was demonstrated to be an anti-hypertrophic miRNA by targeting Cyclin D2 [22].

miRNAs in Cardiac Fibrosis
Cardiac fibrosis is a common phenotype, found in several cardiac diseases including myocardial infarction and heart failure, characterized by the adverse accumulation of collagens and other extracellular matrix (ECM) proteins. Early work from Thum et al. [48] demonstrated that miR-21 promotes cardiac fibrosis by targeting extracellular regulated kinase inhibitor sprouty homolog 1 (Spry1) with stimulation of MAPK signaling in cardiac fibroblasts. In a model of myocardial ischemia-reperfusion, miR-21 was found to target Pten and subsequently leading to an increase in matrix metalloprotease 2 (Mmp2). In line with this finding, antagonism of miR-21 results in an increase of Pten in cardiac fibroblasts [49]. In a pioneer study by Olson's group, expression of miR-29 family members (miR-29a, miR-29b and miR-29c) negatively correlated with expression of genes involved in ECM production and fibrosis after experimental myocardial infarction. Furthermore, the miR-29 family regulates the expression of pro-fibrotic genes, including extracellular matrix genes elastin [50], fibrillin 1 (Fbn1), collagen type I, α 1 and 2 (Col1α1, Col1α2) and collagen type III, α 1 (Col3α1) [51]. In addition, miR-29b targeting Tgfβ1 to play a role in transforming growth factor β (TGFβ)/SMAD3 signaling [52]. Several other miRNAs have been identified as also targeting collagens and TGFβ signaling to play a role in fibrogenesis. For instance, Let-7i and miR-26a attenuate collagen deposition and exert their function through targeting Col1α2 and Col1α1, respectively [23,53,54]. The role of miR-133a in myocardial fibrosis and electrical repolarization in pressure-overloaded adult hearts may reflect its regulatory effects on the expression of Col1α1, Serca2a and calcineurin [55,56]. Recently, miR-101and miR-101a were found to negatively regulate TGFβ signaling through targeting of Tgfβ receptor type 1 (Tgfβr1) and c-Fos. Cardiac fibrosis is significantly suppressed by upregulation of miR-101a in the rat myocardial infarction model [57,58]. miR-125b is pro-fibrotic via targeting of apelin, one of the key fibrogenesis repressors in the heart [59]. Wang et al. [60] showed that miR-24 could interfere with TGFβ signaling by targeting the pro-protein convertase, furin, and subsequently downregulate the level of TGFβ in cardiac fibroblasts. By targeting connective tissue growth factor (Ctgf), miR-133 and miR-30 are suggested to be important for collagen synthesis [61]. Figure 2 provides a schematic representation of the involvement of known miRNAs in classical fibrogenesis signaling pathways. ischemia-reperfusion, miR-21 was found to target Pten and subsequently leading to an increase in matrix metalloprotease 2 (Mmp2). In line with this finding, antagonism of miR-21 results in an increase of Pten in cardiac fibroblasts [49]. In a pioneer study by Olson's group, expression of miR-29 family members (miR-29a, miR-29b and miR-29c) negatively correlated with expression of genes involved in ECM production and fibrosis after experimental myocardial infarction. Furthermore, the miR-29 family regulates the expression of pro-fibrotic genes, including extracellular matrix genes elastin [50], fibrillin 1 (Fbn1), collagen type I, α 1 and 2 (Col1α1, Col1α2) and collagen type III, α 1 (Col3α1) [51]. In addition, miR-29b targeting Tgfβ1 to play a role in transforming growth factor β (TGFβ)/SMAD3 signaling [52]. Several other miRNAs have been identified as also targeting collagens and TGFβ signaling to play a role in fibrogenesis. For instance, Let-7i and miR-26a attenuate collagen deposition and exert their function through targeting Col1α2 and Col1α1, respectively [23,53,54]. The role of miR-133a in myocardial fibrosis and electrical repolarization in pressure-overloaded adult hearts may reflect its regulatory effects on the expression of Col1α1, Serca2a and calcineurin [55,56].

miRNAs in Cardiomyocyte Apoptosis
Apoptosis is programmed cell death which normally ensures proper functional and metabolic homeostasis in multicellular organisms. As with other eukaryotic cells, apoptosis in cardiomyocytes may be triggered by the activation of two main pathways, an extrinsic pathway that is mediated by death receptors, and an intrinsic pathway that involves mitochondrial permeability/transmembrane potential. Both pathways are known to activate initiator caspases (caspase 8, 9, 10) and finally the apoptosis executioner, caspase 3 (Casp3) [63]. Results from one recent study demonstrated that miR-133 and miR-874 negatively regulate the expression of Casp3 and Casp8, respectively, and protect cardiomyocytes from cell death induced by oxidative stress [64]. MiR-133 also regulates expression of Casp9 with subsequent reduction in Casp3 levels in nicotine-induced cell death [65]. Casp3 is also targeted by miR-378 with consequent attenuation of ischemia-induced apoptosis [66]. Srivastava and colleagues used an ischemia-reperfusion injury rat model to show the beneficial effect of ischemic post-conditioning may be mediated by the regulatory effect of miR-133a on Casp9 [67]. In addition to caspases, miRNAs have also been shown to target the B-cell lymphoma 2 (Bcl2) family within the intrinsic mitochondrial apoptosis pathway. Functionally, these miRNAs could be divided into anti-apoptosis and pro-apoptosis sub-families. At present, miR-1, miR-15b, miR-30b, miR-34a, and miR-497 are reported to promote cardiomyocyte cell death through repression of Bcl2 (anti-apoptotic) gene expression whereas miR-149 and miR-24 repress apoptosis by targeting the pro-apoptotic genes Puma and Bim [68][69][70][71][72][73][74].
In addition to targeting Bcl2 or caspase families, miRNAs were reported to target various components of the upstream apoptosis-related signaling cascades. For instance, miR-1 mediates apoptosis of H9c2 cells induced by high glucose, by suppressing the expression of anti-apoptotic genes such as insulin-like growth factor (Igf1). In a transgenic mouse overexpression of miR-1 exacerbated cardiac apoptosis secondary to ischemia-reperfusion. Further mechanistic studies revealed that protein kinase C ε (Pkcε) and heat shock protein 60 (Hsp60) are targets for miR-1 [75,76]. Similarly, miR-320 promotes apoptosis of cardiomyocytes by reducing the level of heat shock protein 20 (Hsp20) [77].
In contrast to be the positive regulators of cardiac apoptosis, several miRNAs such as miR-21, miR-199a, and miR-30 have been reported to antagonize apoptosis of cardiac cells. Multiple lines of evidence demonstrate that miR-21 attenuates cardiac apoptosis secondary to oxidative stress-induced cardiac apoptosis by suppressing expression of programmed cell death 4 (Pdcd4) [78,79]. The protective role of miR-21 against cardiac apoptosis is further substantiated by the observation that during hydrogen peroxide-induced injury in cardiac myocytes knockdown of miR-21 induces, while upregulation of miR-21 protects against, cell death [79]. MiR-21 also suppresses apoptosis induced by hypoxia through targeting of the PTEN/AKT pathway [80]. Like miR-21, miR-199a regulates the expression of hypoxia-inducible factor 1 α (Hif1α) and a class II histone deacetylase, sirtuin 1 (Sirt1). Overexpression of miR-199a under hypoxic conditions downregulates Hif1α and Sirt1 with subsequent suppression of p53 and attenuation of hypoxia-induced cardiomyocyte apoptosis [81]. Interestingly, one study demonstrated that p53 is also targeted by the miR-30 family (miR-30a-e). In apoptosis induced by oxidative stress, downregulation of miR-30 upregulates p53 and dynamin-related protein 1 (Drp1), an initiator of mitochondrial fission, finally culminating in apoptosis [82]. The miR-30 family was found to mediate doxorubicin-induced apoptosis in adult rat cardiomyocytes. Additional experiments demonstrated that miR-30 targets multiple genes in the β-adrenergic signaling pathway including β1 and β2 adrenoceptors and G protein α i subunit (G iα 2) [83]. Several other miRNAs have been demonstrated to play critical roles in hypoxia-induced apoptosis in cardiomyocytes. Among them, miR-100, miR-101, miR-132, miR-15b, miR-20a, and miR-92a were found to enhance apoptosis by targeting negative regulators/factors in the intrinsic apoptosis pathway [69,81,[84][85][86][87][88]. In contrast, miR-133a, miR-138, and miR-146b were demonstrated to attenuate hypoxia-induced apoptosis by targeting different upstream signaling pathways [80,[89][90][91][92]. Numerous other miRNAs have been demonstrated to exert pro-apoptotic or anti-apoptotic effects by targeting various signaling networks upstream of the intrinsic apoptosis pathway. Figure 3 illustrates the involvement of known miRNAs in classical signaling pathways regulating apoptosis.

Future Perspectives
MiRNAs have emerged as vital post-transcriptional regulators involved in almost all biological processes ranging from development, such as stem cell differentiation, to maintenance of cell and tissue integrity for multiple physiological functions. The roles of miRNAs in cardiovascular pathophysiological events and the therapeutic potential of targeting specific miRNA in ameliorating progression towards cardiac hypertrophy, fibrosis, and apoptosis are under ongoing exploration in various in vitro and in vivo disease models. Clearly, as discussed in this review, each cardiac pathophysiological event is regulated by multiple microRNAs. As illustrated in Table 4, a number of known microRNAs are not functionally restricted but have been shown to participate in multiple aspects of cardiac pathophysiology. Since the predominant cardiovascular diseases such as heart failure or hypertension result from complex pathophysiology the elucidation of the roles of multifunctional disease-related miRNAs will provide targets for developing therapeutics able to simultaneously suppress multiple disease-mediating pathological pathways. Recent technological advances in unraveling the complexity of miRNA biology and function, particularly in relation to the cardiovascular system, is crucial for accurate translation of knowledge to alter specific disease-causing signaling cascades, making the development of miRNA-mediated clinical intervention a reachable reality in the near future. However, the challenge remains particularly in integrating laboratory and clinical findings to design innovative research platforms to translate research findings into clinical applications such as biomarkers and/or therapeutic intervention. Cardiovascular diseases, such as heart failure (HF), are complex with contributions from various genomic, genetic, and environmental factors. To date, although numerous genome-wide studies have revealed differential expression patterns of miRNAs in HF compared to controls, signature miRNA profiles for HF have yet to be definitively identified. By compiling reports from different HF cohorts it is apparent a number of apoptosis-related miRNAs display consistent differential expression patterns in various biological specimens including serum, plasma and cardiac tissue. For instance, miR-1 and miR-21 were originally demonstrated to play roles in cardiac remodeling and apoptosis-related signaling in cellular and animal models. A later genome-wide miRNA study revealed that these two miRNAs together with several other cardiac hypertrophy-related miRNA (miR-23) and apoptosis-related miRNAs (miR-195 and miR-199), were differentially expressed in myocardial biopsies from HF patients compared with controls, suggesting the pervasive involvement of these miRNAs in the pathology of HF [117]. Similarly, another cluster of apoptosis-related miRNAs (miR-145, miR-214, miR-378, and miR-7), as well as miR-1 were found differentially expressed in HF cardiac tissues in another study [118]. One recent HF cohort study revealed differential expression of a cluster of hypertrophy-related miRNAs (miR-101 and miR-22), a cluster of apoptosis-related miRNAs (miR-17, miR-30a, and miR-92a) and miR-21 in serum [119]. Further comparisons of 10 selected miRNAs listed in Table 4 with data originating from 9 HF cohorts [117][118][119][120][121][122][123][124] revealed that most of these multifunctional miRNAs are reported as dysregulated in more than one cohort study, suggesting clinical significance. Amongst these, miR-1 was shown to be differentially expressed in four cohorts [117,118,122,124], and miR-21 was reported to differentially expressed in two cohort studies [117,119]. Comparative evaluation across different data bases, particularly in multiple sample types, investigative models, patient cohorts and disease phenotypes, is necessary to shortlist candidate miRNAs for development of putative therapeutic applications in the future.
Over the past decade the essential roles of miRNAs in the biology and cellular dysfunction of cardiovascular pathologies have been revealed at an increasing pace. In this article, by compiling miRNAs associated with cardiac hypertrophy, fibrosis, and apoptosis, we hope to bring insights for the roles of miRNAs in cardiac remodeling pathways to the readers. This body of information provides the foundation for further innovative research towards development of miRNA-based diagnostic methods and therapeutics.