microRNAs are an abundant class of endogenous 22 nucleotide non-coding RNAs that post-transcriptionally control gene expression, interfering with mRNA stability and/or translational rate, by base-pairing to the 3´UTR of protein-coding mRNAs [1-3]. microRNAs are nucleus-encoded non-coding genes that are transcribed by RNA polymerase II and processed by Drosha-Dgcr8 complexes into pre-miRNAs. Subsequently, within the cytoplasm pre-miRNAs are further processed by Dicer endonucleases, leading to double-stranded small RNAs. Eventually the mature miRNA (guide strand) will be loaded into an RNA-induced silencing complex (RISC), which contains Argonaute family members, and within RISC miRNAs mediate mRNA translational repression and/or mRNA instability and degradation [4-5] (Figure 1). Thus, microRNAs might provide a way to diminish the timing and extent of gene expression changes triggered by discrete signaling pathways. The molecular mechanisms of canonical microRNA biogenesis are rather well-established [4-5]. Nevertheless, additional mechanisms of microRNA processing, Drosha and7or Dicer independent have also been described [6-9]. The concept of tissue-specific expression of discrete sets of microRNAs [10] and their corresponding functional consequences is currently emerging. Several hundreds of miRNAs have been identified in humans [www.miRBase.org]. Bioinformatic althgoritms of microRNA seed sequence base-pairing predicts that thousand of messages are under selection [1-3] [http://www.targetscan.org, http://www.microrna.org, http://pictar.mdc-berlin.de/], putatively controlling thus multiple cellular, tissue specific and developmental processes. Over the last years, emerging roles for microRNAs have been identified during both cardiac and skeletal myogenesis. Technical tools have been developed to analyze the expression profile of microRNAs by state-of-the-art means, such as in situ hybridization (http://geisha.arizona.edu/geisha/; Exiqon), qRT-PCR (Exiqon) and microarrays (Ambion, Exiqon), despite the small size of these transcripts. In addition, classical transgenesis has been used to over-express microRNA precursors and similarly genetic engineering has provided specific microRNA knock-outs to dissect the functional role of distinct microRNAs in vivo. Moreover, commercial pre-microRNAs (agonists) and anti-miR/antagomiRs (antagonism) molecules have provided a means to discern the functional capabilities of microRNAs in vitro as well as in vivo (antagomiRs). In this review we will briefly highlight the similar and divergent roles of microRNAs during cardiac and skeletal myogenesis. Before entering the microRNA arena, we will first describe the essential morphogenetic and transcription regulatory events that occur during cardiac and skeletal muscle development.
Figure 1. Schematic representation of microRNA biogenesis. Transcription of microRNA gene leads to a precursor sequence which can contain a single pre-miRNA or multiple pre-microRNAs (pri-miRNAs), resulting from transcription of clustered microRNAs at a single locus (<10 kb). Pri-miRNAs are excised into pre-miRNAs in the nucleus by the action of Drosha. Pre-miRNAs are transferred to the cytoplasm by exportin-5 and recognized by the exonuclease Dicer, which will further process pre-miRNAs into mature microRNAs and will help loading these non-coding RNAs into the RISC complex, providing an environment for sequence complementary with target mRNAs followed by selective translational repression and/or mRNA cleavage. At present it is not fully establish if these mechanisms are mutually exclusive; recent evidence suggest that mRNA cleavage may follow initial transcriptional repression
The developing heart
The development of the heart is a complex process that leads to the formation of four-chambered heart with synchronous contraction from a linear peristaltic contracting tube [11]. The development of the heart initiates soon after gastrulation, when precardiac progenitor cells emerge from the primitive streak and migrate to form the cardiac crescent [11]. Soon afterwards, the linear heart tube forms and bends rightwards, thereby displaying the first sign of morphological asymmetry during embryonic development [12]. Concomitantly, the appearance of distinct atrial and ventricular cardiac chambers takes place. Soon thereafter, separation of left and right embryonic chambers, as well as of their connecting counterparts, will lead to the formation of a four-chambered heart with distinct systemic (left) and pulmonary (right) circuitries [11]. During these morphogenetic processes, gene expression is tightly controlled both spatially and temporally [12-13]. Cardiomyocyte specification and determination results from a wide variety of signaling pathways, such as Bmp, Tgf-beta and Wnt, which in concert and at different developmental stages progressively shape the heart. A wide array of studies has provided fundamental understanding of the transcriptional pathways involved in cardiac morphogenesis. In this context, a large variety of transcription factor are progressively activated, including Nkx2.5, Isl1, T-box, Mef2 and Hand genes, conferring distinct capacities to the developing cardiac cells [11-13]. Currently, we are experiencing a new revolution in our understanding of cardiac development through a series of seminal papers that report the expression and functional roles of microRNAs at distinct stages of cardiac development [14-17].
The formation of skeletal muscles
Trunk and cranial skeletal muscles have different developmental origins. Skeletal muscles of the trunk are derived from the myotomal compartment of the somites, whereas cranial skeletal muscles are derived from cranial mesoderm [18-20]. Nonetheless in both cases, skeletal myogenesis requires the occurrence of specific coordinated events, including exit from the cell cycle, transcription of muscle-specific proteins, fusion into polynucleated fibers and assembly of the contractile apparatus. Such complex processes are regulated at multiple levels [20]. In trunk muscle, proliferative Pax3/7 positive cells, which have stem cell-like properties, undergo activation of the myogenic determination genes, such as Myf5 and MyoD, driving the cell to enter the myogenic programme and become myoblasts with limited proliferation capacity [20]. Myoblasts then exit the cell cycle and acquire a post-mitotic state, with activation of several genes, such as Mrf4, followed rapidly by differentiation and terminal differentiation [20]. Cranial skeletal muscles display similar differentiation events, but are not dependent on Pax3/Pax7 activation [19-20]. Their hierarchical transcriptional activation and differentiation remains largely to be elucidated. A new layer of complexity has been recently added to skeletal muscle development by studies revealing the role of microRNAs at crucial steps of myogenesis [21-22].
At present, a number of microRNAs have been demonstrated to play pivotal roles during cardiac and skeletal muscle development, while many others remain to be explored in more detail. In this review we shall briefly highlight the role of several microRNAs with pivotal roles during cardiac and skeletal myogenesis and discuss their role under pathological conditions.
microRNAs in cardiac and skeletal muscle development
miR-1 and miR-133 microRNA family members play an important role in the control of differentiation and proliferation of muscle cells [25]. The miR-1 subfamily consists of two closely related transcripts, miR-1-1 and miR-1-2, encoded by distinct genes. Bi-cistronic transcripts are generated i.e. miR-1-1/miR-133a-2 andmiR-1-2/miR-133a-1, respectively. The miR1-1/miR133a-2 gene is intergenic whereas miR1-2/miR133a-1 gene is located within an intron of the Mindbomb (Mib) gene [26]. miR-1 is expressed at low levels in the embryonic heart and skeletal muscle but progressively increases in neonatal cardiac and skeletal muscles, remaining expressed at substantially lower levels than in adult tissues. Similar expression profiles have been documented for miR-133 (D. Franco, unpublished data). Importantly, miR-1 also displays a dynamic expression profile during cardiogenesis. miR-1-1 becomes confined to the inner curvature of the heart and atrial appendages, whereas at later stages it is ubiquitously expressed in the entire myocardium. miR-1-2, in contrast, is expressed only within the developing ventricles [14].
Gain- and loss-of function functional analysis of miR-1 genes has demonstrated the pivotal role of this microRNA during cardiogenesis. For example, transgenic mice overexpressing miR-1 have demonstrated that miR-1 is able to promote cardiomyocyte arrest through inhibition of the cardiac-enriched transcription factor Hand2 [23]. Impaired Hand2 expression provokes decreased ventricular cardiomyocyte expansion, resulting in a thin-walled ventricular phenotype [23]. In line with these findings, miR-1-2 knockout mice display an early embryonic lethality, showing ventricular septal defects, thus demonstrating the importance of miR-1 function during heart development [14]. Interestingly, adult miR-1-2 deficient mice show a thickened chamber wall which might be interpreted to result from a continuous process of cardiomyocyte hyperplasia from embryonic stages. The effect of miR-1 in the heart is consistent with its function in skeletal muscle, since miR-1 overexpression in skeletal myoblasts leads to a decrease in cell proliferation and promotes skeletal muscle differentiation [23]. C2C12 myoblasts transfected with miR-1 show strongly enhanced expression of early and late myogenic markers including Pax7, MyoD, myogenin and skeletal α-actin [27]. In this context, recent studies have showed that miR-1 promotes myogenesis by targeting histone deacetylase 4 (Hdac4), a transcriptional repressor of muscle gene expression; accelerated myogenic differentiation induced by miR-1 is accompanied by a marked decrease of phosphorylated histone H3 [28-29]. Gene expression profiling of zebrafish Dicer mutants and miR-1 or miR-133 knockdown embryos has revealed that these microRNAs contribute to over 54% of microRNA-regulated gene silencing during development of the embryonic skeletal musculature [30]. Importantly, miR-1 and miR-133 targets identified in this study were mainly involved in sarcomeric actin organization.
The miR-133 family contains three members, i.e. miR-133a-1, miR-133a-2 and miR-133b, two of which are expressed as bi-cistronic units with miR-1 family members, as previously mentioned [25]. The third member of the family, miR-133b, is located 3.8 kilobases from miR-206, suggesting thus that these two microRNAs might be transcribed as a bi-cistronic unit. Interestingly, miR-1 and miR-206 show high sequence similarity and therefore might interact with similar mRNA targets. Microarray analyses in cardiac and skeletal muscles have shown similar expression profiles of miR-133b and miR-206 (D. Franco, unpublished data). miR-133 is exclusively expressed in developing cardiac and skeletal muscle. Microarray analysis has revealed increased miR-133 expression in developing mouse hearts from day embryonic day (E)12.5 through to at least E18.5 [31-32]. miR-133a inhibits differentiation and promotes proliferation of myoblasts and, therefore, has opposite effects to miR-1. Overexpression of miR-133 promotes myoblast proliferation by repressing serum response factor (Srf) and thus the expression of myogenin and myosin heavy chain isoforms [23, 25, 33]. Surprisingly, loss-of-function of miR-133a-1 or miR-133a-2 results in viable mice with structurally and functionally normal hearts, suggesting redundant roles, possibly also by miR-133b [34]. Consistent with this hypothesis, double mutants (miR-133a-1/miR-133a-2) lead to ventricular septal defects in half of the developing embryos, and those surviving to adulthood display dilated cardiomyopathy and heart failure. This phenotype might result from elevated Srf and cyclinD2 (ccnd2) expression [34].
miR-206 has also been recently characterized as a regulator of skeletal muscle [35]. Overexpression of miR-206 inhibits a component of the DNA polymerase, as well as follistatin (Fstl1) and utrophin (Utrn), genes that are indispensable for differentiation of skeletal muscle cells [35-36]. In co-operation with miR-133 and/or miR-1, miR-206 can repress myoblast fusion by targeting connexin 43 (Cx43; Gja1) gap junction channels without altering levels of Cx43 mRNA [37]. Curiously, no evidence has been reported concerning the regulation of Gja1 by miR-206 in cardiac muscle, although both are co-expressed in cardiomyocytes.
Several microRNAs have been found to be processed from intronic regions of myosin genes which are important for contractility. MiR-208a is encoded within intron 27 of the human and mouse alpha-myosin heavy chain (MHC) gene (Myh6) while miR-208b resides within Myh7/ beta-MHC. Another microRNA, miR-499 can be found in the Myh7b gene. The expression of these myomiRs is associated with the cardiac-muscle-specific expression their host myosin gene [38-39] and are suspected to be controlled by common regulatory elements, including Gata4, Mef2 and thyroid hormone receptor signaling [40]. Targeted deletion of all myomiRs displayed no overt cardiac phenotype. However, under stress these microRNAs have been shown to be part of a regulatory network making a switch from the fast to a slow muscle fiber phenotype [41].
Members of another cluster of microRNAs, the miR-23/24/27 cluster, have been described to be associated to cardiac and/ or skeletal muscle function.
miR-24 is upregulated on myoblast differentiation. Transcriptional regulation of miR-24 has been reported to be controlled by TGF-β/Smad signaling. Knockdown of miR-24 led to reduced expression of myogenic differentiation markers including myogenin and myosin heavy chain genes in C2C12 cells, while ectopic expression of miR-24 enhanced differentiation, and partially rescued myogenic inhibition by TGF-β [22]. These data support a role of miR-24 in skeletal myogenesis.
Expression of miR-27 has been reported both in skeletal and cardiac muscle. It has been shown that miR-27 can target Pax3 expression during myotome development, as well as in skeletal muscle satellite cells [22], whereas miR-27 controls Mef2c expression in both, cardiac and skeletal muscle cells [42].
Several other microRNAs have also been reported to play conspicuous roles in skeletal muscle development, such as miR-181 which is required for efficient myoblast differentiation by regulating Hox-A11 [43] or miR-221/miR-222 which are required for myoblast cell fusion [44]. Similarly, a subset of microRNAs has been reported to play noticeable roles in cardiac muscle, such as miR-98, miR-199 and miR-21 [45-49]. In this important to highlight that complex regulatory mechanisms involving regulation of miRNA-17-92 by bone morphogenetic factors (Bmp2 and Bmp4) results in modulation of transcription factors such as islet-1 and Tbx1 [17]. As summarized in Figure 2, these data illustrate the increasing evidence of crucial roles of microRNAs during cardiac and skeletal muscle development. Our current knowledge, however, is just incipient, suggesting that microRNA regulation is set to be an important and challenging research topic in myogenesis in the coming years.
Schematic representation of the most representative microRNAs involved in cardiac development and disease. Multiple roles have been reported for miR-1 and miR-133 in both cardiac development and disease, mainly controlling cell proliferation/apoptosis, epigenetic remodeling and ion channel regulation, whereas the role of other microRNAs, such as miR-27, miR-30, miR-29, miR-208 and miR-328 is just beginning to be understood. Transcription of microRNA gene leads to a precursor sequence which can contain a single pre-miRNA or multiple pre-microRNAs (pri-miRNAs), resulting from transcription of clustered microRNAs at a single locus (<10 kb). Pri-miRNAs are excised into pre-miRNAs in the nucleus by the action of Drosha. Pre-miRNAs are transferred to the cytoplasm by exportin-5 and recognized by the exonuclease Dicer, which will further process pre-miRNAs into mature microRNAs and will help loading these non-coding RNAs into the RISC complex, providing an environment for sequence complementary with target mRNAs followed by selective translational repression and/or mRNA cleavage. At present it is not fully establish if these mechanisms are mutually exclusive; recent evidence suggest that mRNA cleavage may follow initial transcriptional repression.
microRNAs in cardiac diseases
Several studies have provided evidence of novel links between miRNA function and cardiac disease. Differential microRNA expression has been reported in distinct clinical and experimental models of cardiac diseases, including dilated cardiomyopathy [50], ischemic heart [49-51], heart failure [52], hypertophic cardiomyopathy [31, 38-39] and atrial fibrillation [53]. These data support the involvement of microRNAs in cardiac diseases. Functional studies to date have focused mainly on two cardiac physiopathological conditions, the hypertrophic and arrhthymogenic heart, as illustrated in Figure 2. Functional assessment of the role of these microRNAs in cardiac hyperthrophy was demonstrated by van Rooij et al. [38,39] who overexpressed of individual stress-inducible miRNAs in isolated cardiomyocytes and found this was sufficient to induce hypertrophic growth. In addition, Sayed et al [48] demonstrated that miR-1 overexpression inhibits protein synthesis and cell growth in part by inhibiting Ras guanosine-triphosphatase-activating protein (Rasgap), cyclin-dependent kinase 9 (cdk9), fibronectin (fn1) and Ras homolog enriched in brain (Rheb). Based on these observations, it has been suggested that downregulation of miR-1 is necessary to relieve of growth-related target genes from its repressive influence and induction of hypertrophy. More recently it has been reported that decreased levels of miR-1 can provoke cardiac hypertrophy by regulating hsp60, hsp70 and caspase-9 [54] and Bcl-2 [55].
Care et al. [31] have also reported a critical role of miR133 in cardiac hypertrophy. In vitro manipulation of miR-133 blocked cardiomyocyte hypertrophy. Moreover, in vivo miR-133 antagomiR injection led to sustained cardiac hypertrophy, probably as consequence of targeting RhoA, Cdc42 and Whsc2 [31]. Similar to miR-1, miR-133 has also being reported to regulate hsp60, hsp70 and caspase-9 [54], thus contributing to the hyperthrophic response. In this context, modulation of miR-1 expression by propanolol has been reported to be an adequate therapeutical approach to enhance ischaemic cardioprotection, since it can rescue the expression of Kir2.1 and Gja1 [56].
Other microRNAs have also been reported to play a role in cardiac hypertrophy, yet their functional role remains poorly understood. For example, miR-21 is up-regulated in hypertrophic murine hearts [39-40, 49]. Owing to its capacity to simultaneously regulate multiple pro-apoptotic and anti-apoptotic genes [57], miR21 has been suggested to play a pivotal role in cardiac hypertrophy [57-58]. Consistent with this hypothesis, cultured neonatal rat cardiomyocyte exposed to angiotensin II or phenylephrine markedly up-regulated miR-21 expression and miR-21 antisense oligonucleotides were sufficient to inhibit cardiomyocyte hypertrophy induced by either angiotensin II or phenylephrine [59]. Transgenic mice overexpressing miR-195 under the control of the α-myosin heavy chain (MHC) promoter is sufficient to induce pathological cardiac growth and heart failure [39]. However, the mechanism by which miR-195 is relevant to hypertrophy has not been studied to date.
In addition to the role of microRNAs in cardiac hypertrophy, new evidence has revealed that microRNA function is also critical for cardiac electrophysiology. Gain-of-function miR-1 transfection into healthy hearts significantly widened the QRS complex and prolonged the QT interval, indicating cardiac conduction arrest [60], while miR-1 loss-of-function narrowed the QRS complex. In addition, miR-1 knockdown prevents cardiac arrhythmias, while miR-1 overexpression leads to cardiac arrhythmias in normal and infarcted rat hearts, respectively. At the molecular level, it has been demonstrated that miR-1 regulates Irx5, connexin43 (Gja1) and Kir2.1 (Kcnj2) expression [60-61], suggesting that a tight regulation of miR-1 levels is crucial to maintain normal cardiac conduction. Furthermore, overexpression of miR-1 enhances cardiac excitation-contraction coupling by selectively increasing L-type calcium and ryanodine receptor channel phosphorylation and intefering with protein phosphatase PP2A subcellular localization to those channels [62]. Aberrant miR-1 expression has been documented in atrial fibrillation patients [63], further supporting the role of this microRNA in cardiac arrhythmias.
miR-133 is capable of regulating cardiac electrical function in the diabetic myocardium, by affecting Herg (Kcnh2) expression [64]. Diabetic rabbit hearts display increased miR-133 expression and delivery of exogenous miR-133 into rabbit cardiomyocytes produces posttranscriptional repression of Kcnh2, and thus induced substantial IKr current decrease. Slower repolarization of the action potential in diabetic hearts results in QT prolongation. Furthermore, miR-1 and miR-133 also regulate the expression of hyperpolarization- activated cyclic nucleotide-gated channels (Hcn), Hcn2 and Hcn4, thereby reducing pace-maker activity in diabetic hearts [65].
In addition to these microRNAs, several other have also been recently implicated in cardiac diseases, yet their functional consequences remain poorly discerned. A role for miR-29 has been reported in cardiac fibrosis [66], miR-30 and miR-133 play a role in cardiac extracellular matrix remodeling [67], while miR-328 has been demonstrated to be increased in atrial fibrillation, further demonstrated by a functional role for miR-328 in AF vulnerability [59].
microRNAs in skeletal muscle diseases
Differential expression of microRNAs has been reported in several muscle diseases. A comparative study across a panel of ten different groups of muscle disorders and unaffected human skeletal muscle provided evidence that miR-30b, miR-92, miR-361, miR-423, miR-29ab are aberrantly expressed in several skeletal muscle disorders including Duchenne muscular dystrophy (DMD), Nemaline myopathy (NM), Fasciocapulohumeral muscular dystrophy (FSHD) and Limb girdle muscular dystrophies type 2B (LGMD2B) [68]. Similarly, miR-21 expression is impaired in ischemic skeletal muscle [69], miR-206 in myotonic dystrophy type 1 [70] amyotrophic lateral sclerosis [71] and within the muscular dystrophy experimental model, mdx [72] and miR-1 and miR-133a expression are decreased during skeletal muscle hypertrophy [73]. Yet to date, evidence concerning the functional role of these microRNAs is scarce.
Some insights have been gained in different dystrophies and inflammatory myopathies regarding NF-kB modulation of the immune response, mediated by microRNAs, [74-75]. Activated NF-kB inducing miR-146 expression has been recently demonstrated by Taganov et al [75], supporting the hypothesis that miR-146 might play a role in the onset of this myopathy, yet it will be important to identify and analyze miR-146 downstream targets genes. In addition, insights have been obtained concerning the generation of amyotrophic lateral sclerosis (ALS). Mice deficient in miR-206 display normal neuromuscular synapses during muscle formation but lack of miR-206 accelerates amyotrophic lateral sclerosis in an ALS diseased model. These effects are, in part mediated by regulating fibroblast growth factor and histone deacetylase 4 signaling pathways [75].
Conclusion:
In summary, cardiac-specific and skeletal muscle-specific miRNAs have been identified and essential functions in controlling skeletal muscle proliferation and differentiation have been reported. Notably, it has been shown that microRNAs clustered at the same chromosomal loci and transcribed together as a single transcripts can produce two independent, mature miRNAs with distinct biological functions, achieved by inhibiting different target genes. One of the major challenges of microRNA regulation is to understand their transcriptional and post-transcriptional regulation in the context of cell type specificity. Similarly, the identification of target genes providing tissue and functional versatility remains to be fully elucidated. Thus, over the next years we will be able to uncover the functional roles of microRNAs in muscle development and disease, providing the bases for the use of microRNA expression as biomarkers and for the design and exploration of the therapeutical capabilities of these small RNA molecules, alone or in combination with cell regenerative approaches, as has been recently reported concerning the use of miRNA-treated skeletal myoblast transplantation to repair hearts damaged after myocardial infarction [76].
Our mission: To reduce the burden of cardiovascular disease.