- Email: [email protected]
MicroRNAs flex their muscles
Review
MicroRNAs flex their muscles Eva van Rooij, Ning Liu and Eric N. Olson Department of Molecular Biology, University of Texas Southwestern Medical Center, Dallas, TX 75390-9148, USA
MicroRNAs negatively regulate gene expression by promoting mRNA degradation and inhibiting mRNA translation. Recent studies have uncovered a cadre of muscle-specific microRNAs that regulate diverse aspects of muscle function, including myoblast proliferation, differentiation, contractility and stress responsiveness. These myogenic microRNAs, which are encoded by bicistronic transcripts or are nestled within introns of myosin genes, modulate muscle functions by fine-tuning gene expression patterns or acting as ‘on-off’ switches. Muscle-specific microRNAs also participate in numerous diseases, including cardiac hypertrophy, heart failure, cardiac arrhythmias, congenital heart disease and muscular dystrophy. The myriad roles of microRNAs in muscle biology pose interesting prospects for their therapeutic manipulation in muscle disease. Myriad roles of microRNAs in muscle biology Muscle cells provide a powerful model for understanding the transcriptional circuitry and signaling systems that regulate cell differentiation and organogenesis. The process of muscle development begins when mesodermal stem cells become committed to a muscle cell fate in response to extracellular signals, giving rise to immature muscle cells or myoblasts. The signals and transcription factors that control the specification and differentiation of cardiac and skeletal muscle cells are different, but the differentiation process of both types of muscle is accompanied by the activation of vast arrays of muscle-specific genes encoding the proteins required for the specialized functions of striated muscles, including components of the contractile apparatus, enzymes, receptors and ion channels. Although the transcriptional networks involved in muscle differentiation are well defined, recent studies have shown a new layer of regulation in muscle cell proliferation, differentiation, contractility and stress responsiveness that is mediated by a collection of muscle-specific microRNAs (miRNAs) that are highly conserved in vertebrates (Table 1). Here we discuss the roles of miRNAs in muscle biology, the prospects for manipulating their expression and activities in muscle diseases and important questions for the future. miRNA biogenesis miRNAs are 22 nucleotides long and inhibit translation or promote mRNA degradation by annealing to complementary sequences in the 30 untranslated regions (UTRs) of specific target mRNAs [1]. There are estimated to be as many as 1000 miRNAs encoded by the human genome [2], roughly equaling the number of transcription factors. Corresponding author: Olson, E.N. ([email protected]).
Individual miRNAs can target dozens of mRNAs, and individual mRNAs can be targeted by multiple miRNAs, allowing for enormous complexity and regulatory potential of gene expression. In most cases, miRNAs fine-tune gene expression patterns, but there are also examples in which they act as on-off switches in gene expression [3,4]. The power of miRNAs as regulators of gene expression is also underscored by a recent study demonstrating their ability to up-regulate translation of specific targets [5]. miRNAs are transcribed as long pri-miRNAs that encode one or more miRNAs or are encoded by introns of protein-coding genes. Pri-miRNAs are processed in the nucleus by the RNase Drosha, yielding stem-loop structures of 70 nucleotides. These pre-mRNAs are transported to the cytoplasm where they are further processed by the RNase Dicer, giving rise to the mature miRNA and its complementary strand from the stem-loop, referred to as the ‘star’ strand. The mature miRNA is incorporated into the RNA-induced silencing complex (RISC), where the miRNA strand anneals to the 30 UTRs of target mRNAs, promoting translational repression or mRNA degradation [6–10]. Identification of bona fide mRNA targets for individual miRNAs represents one of the biggest challenges in the field. Nucleotides 2 through 8 at the 50 end of the miRNA, termed the seed sequence, are the most important determinants of mRNA target selection [11]. However, other nucleotides and mRNA secondary structure in the regions surrounding the target sequence also influence the association of miRNAs with their targets [12]. Because of the plethora of potential targets for individual miRNAs, gain and loss of function miRNA phenotypes likely reflect the aggregate effects of perturbing the expression of many mRNA targets rather than a single target. The MADS–miR-1/133 system: a balancing act Skeletal muscle formation requires the commitment of multipotential stem cells to the skeletal muscle lineage [13]. In skeletal muscle cells, the processes of proliferation and differentiation are mutually exclusive. In response to growth factor depletion, proliferating myoblasts exit the cell cycle, fuse to form multinucleated myotubes and activate the transcription of muscle differentiation genes. miRNAs, by modulating the balance between the antagonistic processes of myoblast growth and differentiation, are integral components of the regulatory circuitry for muscle development. The importance of miRNAs in the normal development of vertebrates is evidenced by Dicer loss of function. Dicer-deficient mice die at gastrulation and lack multipotent stem cells [14]. To study the requirement of miRNAs specifically during skeletal muscle development, a conditional Dicer allele was removed by expressing Cre
0168-9525/$ – see front matter ß 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.tig.2008.01.007 Available online 5 March 2008
159
Review
Trends in Genetics Vol.24 No.4
Table 1. Muscle-specific microRNAs miR miR-1
miR-133
miR-206
miR-208
Functions Inhibition of proliferation Control of cardiac conductance Myogenesis Cardiogenesis Inhibition of cardiac hypertrophy Promotion of proliferation Control of cardiac conductance Myogenesis
b-Mhc expression Stress-dependent cardiac remodeling
Validated target a Hand2 lrx5, KCND2
Refs [17] [21]
HDAC4 Delta RhoA, Cdc-42, WHSC2
[20] [49] [46]
SRF HERG
[20] [30]
Cx43 Fst11, Utrn PolA1
[50] [51] [22]
THRAP1
[32]
a
Cdc-42, cell division cycle 42; Cx43, gap junction protein, a 1; Fstl1, follistatin-like 1; HERG, potassium voltage-gated channel, subfamily H, member 2; RhoA, ras homolog gene family, member A; Utrn, utrophin; WHSC2, Wolf-Hirschhorn syndrome candidate 2.
recombinase under control of MyoD regulatory elements, which are expressed in skeletal muscle from day E9.75 onward. Skeletal muscle deletion of Dicer perturbs embryonic and postnatal muscle development, causing skeletal muscle hypoplasia [15]. The MADS (MCM1, agamous, deficiens, serum response factor) box transcription factors, serum response factor (SRF) and myocyte enhancer factor-2 (MEF2), regulate muscle cell proliferation and differentiation through combinatorial interactions with other transcriptional regulators. MEF2 and SRF activate skeletal muscle gene expression in collaboration with myogenic bHLH (basic helix loop helix) proteins, including myogenin and MyoD, whereas these MADS box factors activate cardiac gene expression in conjunction with GATA transcription factors, the homeodomain protein Nkx2–5, myocardin, a transcriptional coactivator and others [16]. In addition to regulating the expression of muscle structural genes, SRF and MEF2 control the expression of two pairs of muscle-specific miRNAs, miR-1–1/133a-2 and miR1–2/133a-1, which are encoded by bicistronic pre-miRNAs (Figure 1). Muscle-specific enhancers located upstream
and within the introns of miR-1–1/133a-2 and miR-1–2/ 133a-1 control transcription in differentiated muscle cells [17–19]. SRF and MEF2 cooperate with MyoD to activate transcription of these miRNAs in skeletal muscle. In the heart, the upstream enhancer relies on SRF for expression in ventricular and atrial myocytes, whereas MEF2 directs ventricular expression from the intronic enhancer (Figure 2) [17,19]. A third pair of related muscle-specific miRNAs, miR206/133b, is expressed specifically in skeletal muscle [20]. miR-1–1, miR-1–2 and miR-206 differ at only three nucleotides, and miR-133a-2, miR-133a-1 and miR133b are identical except for one nucleotide at the 30 end of miR-133b, indicating that each of these miRNA trios are likely to target the same mRNAs (Figure 1). Such redundancy complicates functional analysis by traditional gene knockouts in the mouse. miR-1 and miR-133 modulate muscle growth and differentiation by regulating SRF and MEF2 activity, thereby establishing negative feedbacks loops within muscle cell lineages [20] (Figure 2). miR-1 represses the expression of histone deacetylase 4 (HDAC4) [20], which acts as a signaldependent repressor of muscle differentiation together with MEF2 [16]. Thus, miR-1 up-regulation during differentiation is a mechanism to dampen HDAC4 expression and to potentiate MEF2 pro-myogenic activity (Figure 2). Forced miR-1 overexpression in the embryonic heart results in lethality during mid-embryogenesis because of a deficiency of cardiomyocytes and consequent cardiac failure [17]. These findings suggest that miR-1 inhibits cardiomyocyte proliferation, consistent with the presence of excessive cardiomyocytes in miR-1–2/ mice [21]. Among its many predicted targets, miR-1 represses the translation of Hand2 [17], a bHLH transcription factor that is required for cardiac growth during embryogenesis (Figure 3). miR-206 promotes skeletal muscle differentiation, whereas miR-206 knockdown, through antisense oligonucleotides, prevents myoblasts from exiting the cell cycle and blocks differentiation [18,22]. Consistent with these findings, miR-206 negatively regulates DNA polymerase a (polA1) translation, thereby inhibiting DNA synthesis [22].
Figure 1. Bicistronic muscle-specific microRNA (miRNA) clusters. The figure shows schematic diagrams of bicistronic muscle-specific miRNA genes and their sequence homologies. (a) Three bicistronic miRNA genes encoding miR-1–1/miR-133a-2, miR-1–2/miR-133a-1 and miR-206/miR-133b are shown. Each pair of miRNAs is transcribed from left to right. Individual miRNAs, designated by colored rectangles, are generated by processing of the bicistronic miRNA transcripts. Chromosomal locations of each miRNA gene and the muscle tissues in which they are expressed are shown. (b) Sequences homologies among muscle-specific miRNAs are shown. Nonhomology is indicated by black letters.
160
Review
Trends in Genetics
Vol.24 No.4
Figure 2. Control of skeletal muscle proliferation and differentiation by miR-1 and miR-133. In skeletal muscle cells, the process of proliferation is antagonistic to the process of myoblast differentiation. miR-1 and miR-133 function at the center of a network of transcription factors to regulate skeletal myoblast proliferation and differentiation. (a) Myocyte enhancer factor-2 (MEF2) and myogenic basic helix loop helix (bHLH) proteins, including MyoD (green), regulate their own expression, as well as the expression of downstream muscle structural genes. Additionally, these transcription factors use upstream and intragenic enhancers to activate transcription of bicistronic pri-miRNAs encoding miR-1 (blue rectangle) and miR-133 (black rectangle) in differentiated skeletal muscle. miR-1 represses expression of HDAC4 (purple), a signal-dependent repressor of MEF2 (red) activity, thereby establishing a negative feedback loop to modulate miR-1 and miR-133 expression and promoting myoblast differentiation. miR-133 represses expression of serum response factor (SRF; yellow), a positive regulator of miR-1/133 expression and repressor of myoblast proliferation. (b) E11.5 transgenic mouse embryos, which harbor a lacZ transgene controlled by the miR-1–2/133a-1 intragenic enhancer, are shown. The wildtype (WT) enhancer directs transgene expression in heart and skeletal muscle (note the blue color), whereas an enhancer with a mutation in the MEF2 site is not expressed in skeletal muscle or ventricular myocardium.
By contrast, miR-133 has been proposed to promote myoblast proliferation, a role opposite to that of miR-1 [20]. The ability of miR-133 to promote proliferation has been ascribed to the repression of SRF, an essential regulator of muscle differentiation (Figure 2). miR-133 also represses translation of the polypyrimidine tract-binding protein (nPTB), which promotes differential splicing of a variety of transcripts that influence the muscle differentiation program [23]. It seems counter-intuitive that miR-1 and miR-133, which are encoded by the same MEF2regulated bicistronic transcripts, would exert opposing effects on muscle growth and differentiation; these miRs also play opposing roles in cardiomyocyte apoptosis, with miR-1 being pro-apoptotic and miR-133 being anti-apoptotic [24]. However, both miR-1 and miR-133 fine-tune key regulatory pathways in an antagonistic manner, with the balance being tipped one way or the other by additional transcription factors and regulatory pathways. Further evidence for the involvement of muscle-specific miRs in the control of skeletal muscle growth comes from the observation that functional overload of skeletal muscle, a stimulus for hypertrophy, is accompanied by miR-1–2 and miR-133a-2 down-regulation and miR-206 up-regulation [25]. Moreover, a mutation that causes dramatic muscularity in the Texel strain of sheep has been mapped to a single G-to-A mutation within the 30 UTR of the mRNA
encoding myostatin, a member of the transforming growth factor b (TGF-b) family of growth factors that represses muscle growth. This mutation creates a target site for miR1 and miR-206, resulting in myostatin translational repression [26], which phenocopies the ‘muscling’ phenotype that results from Myostatin loss of function mutations in mice, cattle and humans [27,28]. miR-1 and miR-133 have also been implicated in the modulation of electrical conductance within the heart. miR-1–2/ mice that survive to adulthood display cardiac conduction abnormalities, accompanied by dysregulation of the potassium channel, Kcnd2 [21]. These defects are likely attributable to developmental abnormalities, because antisense-mediated miR-1 knockdown in the adult heart suppressed arrhythmias that resulted from myocardial infarction [29]. Conversely, miR-1 overexpression in the myocardium promoted ischemic arrhythmias [29]. miR-133 has also been linked to cardiac arrhythmias. Exogenous miR-133 expression in cardiomyocytes inhibits expression of ether-a-go-go related gene (HERG), which encodes a K+ channel responsible for rapid delayed rectifier K+ current (IKr) [30]. Together, these findings suggest that the heart is exquisitely sensitive to the level of miR-1 or miR-133 expression such that an increase or decrease in their expression renders the heart susceptible to arrhythmogenic events. 161
Review
Trends in Genetics Vol.24 No.4
Figure 3. Control of cardiac growth, differentiation and conductance by miR-1 and miR-133. miR-1 and miR-133 function at the center of a network of transcription factors to regulate ventricular growth, cardiomyocyte differentiation and conductance of the heart. (a) miR-1(blue) and miR-133 (black) are expressed in cardiac muscle from a bicistronic pri-miRNA. miR-1 represses expression of HDAC4 (purple), a signal-dependent repressor of myocyte enhancer factor-2 (MEF2; red) activity, thereby establishing a negative feedback loop to modulate miR-1 and miR-133 expression in ventricular myocardium. miR-1 also represses expression of Hand2 (green), a positive regulator of proliferation, and Irx5 (Iroquois homeobox 5; red), which regulates the expression of Kcdn2, a K+ channel involved in cardiac repolarization. miR-133 represses expression of serum response factor (SRF; yellow), a positive regulator of miR-1 and miR-133 expression. MEF2 and SRF rely on additional cofactors (orange) that are not yet identified, to activate transcription of miR-1 and miR-133 and other target genes in the heart. (b) Hearts from E11.5 transgenic mouse embryos, which harbor a lacZ transgene controlled by the miR-1–2/133a-1 intragenic enhancer, are shown. The wildtype (WT) enhancer directs expression in atrium and ventricles (note blue color), whereas an enhancer with a mutation in the MEF2 site is not expressed in the ventricular myocardium. la, left atrium; lv, left ventricle; ra, right atrium; rv, right ventricle.
Regulation of stress-responsiveness and muscle cell identity by miR-208 Acute and chronic injury to the heart results in hypertrophic growth, which frequently culminates in a loss of pump function, fibrosis and lethal arrhythmias [31]. A hallmark of heart disease is a switch in expression of myosin heavy chain (MHC) genes from a-MHC, a fast contracting myosin, to b-MHC, a slow, embryonic myosin. Intriguingly, intron 29 of the gene that encodes a-MHC gives rise to the sole cardiac-specific miRNA, miR-208, which plays a central role in the regulation of the a- to b-MHC switch, and in other aspects of the cardiac stress response, such as cardiomyocyte hypertrophy and fibrosis [32] (Figure 4). Mice with a homozygous deletion of the miR-208 coding region are viable, but they fail to upregulate b-MHC expression in response to pressure overload or activated calcineurin (a potent inducer of cardiac 162
remodeling). a- and b-MHC are also antithetically regulated by thyroid hormone (T3): T3 induces a-MHC, whereas hypothyroidism results in a dramatic increase in b-MHC expression [33]. In the absence of miR-208, b-MHC expression is potently repressed during the hypothyroid state. By contrast, b-MHC expression during embryogenesis, when it is still responsive to embryonic transcription factors like Nkx2.5 and MEF2 [34], is unaffected in miR208/ mice. Thus, miR-208 is dedicated specifically to the activation of b-MHC in response to stress and hypothyroidism in the adult heart and acts as an on-off switch for bMHC expression rather than a fine-tuner of gene expression. Among the top predicted miR-208 targets is the thyroid hormone receptor coregulator THRAP1, which is a cofactor of the thyroid hormone receptor and can either act as an activator or repressor [34]. Because of the repressive effect
Review
Trends in Genetics
Vol.24 No.4
Figure 4. Modulation of cardiac muscle phenotypes by miR-208. The a- and b-myosin heavy chain (MHC) genes are linked in the genome. a-MHC promotes fast cardiac muscle contraction, whereas b-MHC promotes slow contraction. miR-208, encoded by the a-MHC gene, negatively regulates the expression of THRAP1 and other targets. In the absence of miR-208, b-MHC expression becomes nonresponsive to stress signaling and hypothyroidism. miR-208 absence also results in the up-regulation of fast skeletal muscle genes in the heart and a block in pathological cardiac remodeling. The short-term consequences of the lack of miR-208 are the maintenance of cardiac function after stress. However, miR-208–null mice eventually display abnormalities in cardiac contractility, which might reflect the inappropriate expression of skeletal muscle genes in the heart.
on b-MHC transcription specifically after birth, we proposed that miR-208 normally modulates THRAP1 expression, such that, in the absence of miR-208, THRAP1 expression is up-regulated along with consequent enhancement of its actions, whether positive or negative [32]. By example, increased THRAP1 expression is predicted to enhance the repressive influence of the TR on the negative thyroid hormone response element (TRE) in the b-MHC promoter. Further research is needed to determine how other potential miR-208 targets regulate cardiac physiology. Intriguingly, the genome contains an additional copy of miR-208, miR-208b, which is located within the b-MHC gene [2,35]. miR-208 and miR-208b share the same seed
sequence but differ at three nucleotides in their 30 region (Figure 5). The expression and functions of miR-208b have not yet been reported, but it is likely that miR-208b, like miR-208, is co-expressed with its host myosin gene, i.e. predominantly in slow skeletal muscle and in the heart after stress. Because a- and b-MHC are counterbalanced in the heart, the expression of miR-208/miR-208b is likely maintained at relatively constant levels. Thus, although aMHC is the dominant MHC isoform in mice and b-MHC dominates in humans [36], the relative expression of miR208 versus miR-208b will differ accordingly in mice versus humans. Because miR-208 is required for cardiac b-MHC expression in response to stress and hypothyroidism in mice, it is tempting to speculate that miR-208b might play
Figure 5. A family of MyomiRs. Three microRNAs (miRNAs) encoded by introns of myosin heavy chain (MHC) genes share homology, suggesting they target overlapping sets of mRNAs. We refer to these miRNAs as Myomirs because they are encoded by myosin genes and are expressed specifically in striated muscle cells. These miRNAs and their host MHC genes are expressed specifically in slow skeletal muscle and/or heart. (a) miR-208 is encoded by intron 29 of the a-MHC gene (Myh6), miR-208b by intron 31 of b-MHC (Myh7) and miR-499 by intron 19 of Myh7b. Each myosin gene is transcribed from left to right; both the muscle tissues in which they are expressed and their chromosomal locations are shown. (b) Sequences homologies among muscle-specific miRNAs are shown. Nonhomology is indicated by black letters.
163
Review a more important role in the human shift toward b-MHC during heart disease. Nonetheless, these data indicate that there must exist an intimate form of cross-talk between these miRNAs and the myosin genes in which they are embedded. Although miR-208 is cardiac specific, fast skeletal muscle genes are inappropriately expressed in miR208/ hearts, suggesting that one function of miR-208 is to repress the expression of a transcriptional activator of the fast skeletal muscle gene program [32] (Figure 4). It is intriguing in this regard that thyroid hormone promotes the formation of fast myofibers and represses slow myofiber development [37]. Whether miR-208 acts through THRAP1 to control thyroid hormone signaling, stress-dependent cardiac growth and myofiber specification or whether additional targets are involved remains to be determined. Because the sequence homology between miR-208 and miR-208b indicates that they share common mRNA targets, it will be interesting to see whether miR208b fulfills a comparable function to miR-208 in repressing fast skeletal genes. It is tempting to speculate that bMHC gene expression plays a dual role in slow skeletal muscle, not only by regulating the speed and efficiency of contraction, but also by encoding miR-208b, which represses the expression of fast skeletal genes and thereby determines slow myofiber identity. Another more ancient myosin, Myh7b, like b-MHC, is expressed specifically in the heart and slow skeletal muscles [38]. An intron within Myh7b encodes another miRNA, miR-499. Although the sequence of the mature miR-499 miRNA differs substantially from miR-208 and miR-208b, the seed region contains six overlapping bases, hinting at possible shared functions (Figure 5). We refer to these miRs as MyomiRs, because of their location within myosin genes, their specific expression in myogenic cells and the ability of miR-208 to regulate myosin switching and myofiber identity. The existence of a family of closely related musclespecific miRNAs encoded by MHC genes raises many interesting questions. Because it seems counterintuitive that miR-208 regulates the expression of b-MHC, which itself encodes a miRNA closely related to miR-208, miR208b, it will be intriguing to determine to what extent the functions and targets of the MyomiRs actually are redundant. Another interesting facet of the MyomiR network is the hierarchical relationships among these miRs; the regulation of b-MHC and thereby miR-208b, is miR-208 dependent, which might be both species specific and stress dependent. Another interesting point is how and why striated muscle evolved a regulatory network in which miRNAs are embedded in their target myosin genes, and what are the mechanisms whereby these miRNAs regulate the cardiac stress response and myofiber identity? The answers to these questions promise to provide new and important insights into the fundamental logic of muscle gene expression and development, as well as the regulatory circuits that drive cardiac disease. Looking toward miRNA-based therapeutics The key roles of miRNAs in such a diversity of muscle functions raises interesting prospects for therapeutic 164
Trends in Genetics Vol.24 No.4
manipulation of miRNA-based mechanisms in the settings of muscle diseases. For example, the diminished hypertrophy, fibrosis and fetal gene activation observed in miR208/ mice in response to cardiac stress raises the possibility that therapeutic suppression of miR-208 expression might enhance cardiac function after acute or chronic injury [32]. Similarly, numerous stress-responsive miRNAs, including miR-21 and miR-195, are up-regulated during heart disease, some of which have been shown to be sufficient and necessary for cardiac dysfunction [39–44]. Diminishing the expression of these pathogenic miRNAs could provide therapeutic benefit. However, in contrast to miR-208, which is cardiac-specific, the latter miRNAs are widely expressed, which poses challenges in preventing undesired effects on noncardiac tissues. Specific miRNA expression patterns are also associated with muscular dystrophies and other myopathies [45]. It will be of interest to determine whether these disease-associated miRNAs participate directly in the disease process and, if so, whether their therapeutic modulation might influence disease severity or progression. It is also conceivable that therapeutic elevation of specific miRs could enhance muscle function. miR-133, for example, has been implicated in the blockade to cardiac hypertrophy [46], suggesting that its elevation could be beneficial under conditions of cardiac stress. However, it should also be kept in mind that elevation or diminution of miR-133 expression has been linked to cardiac arrhythmias [30], demonstrating the sensitivity of cardiac function to fluctuations in this miRNA and pointing to the possible adverse effects of manipulating miRNA expression in vivo. Similarly, modulation of miR-1 expression, either up or down, causes dysregulation of cardiac ion channels and consequent conduction defects [21,29]. Ventricular-septal and atrial-septal defects in miR-1 mutant mice indicate an important role for this miRNA, and probably many others, in heart development [21]. In this regard, DiGeorge syndrome, a chromosomal deletion disorder resulting in cardiac and craniofacial abnormalities, is associated with deletion of the DiGeorge syndrome critical region 8 gene (DGCR8), which encodes a component of the RISC complex [47]. Whether DGCR8 deletion contributes to congenital abnormalities by perturbing the functions of miRNAs that direct these processes remains to be determined. How might the biology of miRNAs be exploited in future clinical applications? One can imagine inhibiting the expression of miRNAs involved in disease pathogenesis using antimiRs. Several promising approaches involving chemically modified oligonucleotides complementary to miRNAs have recently been described [48]. It is also conceivable that miRNA mimics could be developed to enhance the expression of beneficial miRNAs. Customized anti-miRs or miR mimics directed against specific mRNA targets can also be envisioned. Another potentially powerful therapeutic approach will be the development of drugs that modulate the expression or activities of disease-related miRNAs. Not withstanding these opportunities, there are several important hurdles that remain to be overcome, such as efficient delivery and specificity of anti-miRs or miR mimics in vivo. This is of particular concern given that
Review individual miRNAs can have dozens or even hundreds of targets, such that blockade of a miRNA against one set of targets could be salutary whereas eliminating repression of another set of targets could be pathological. Off-target effects also represent a significant challenge, especially considering that miRNA-mediated repression often requires homology of only six to seven nucleotides in the seed region of the miRNA and the mRNA target. Toxicity of chemical modifications used to facilitate cellular uptake and prevent degradation also represents an important consideration. The modification of antagomirs with cholesterol, for example, poses problems of liver toxicity. Finally, it is apparent that miRNA expression patterns are dynamically regulated during disease and provide molecular signatures for disease progression. Thus, it is likely that miRNA expression patterns will be used for disease diagnosis and prognosis, as shown recently for numerous forms of heart disease [31]. Concluding remarks Our understanding of micro RNA (miRNA) biology is still in its infancy. It has been estimated that at least one third of mammalian genes are regulated by as many as a thousand miRs, only a few of which have been studied in any detail. An important challenge for the future will be to identify the downstream targets that mediate the actions of miRNAs in development and disease. Ascribing the actions of a particular miRNA to specific targets is especially challenging, given the plethora of mRNAs of both high and low affinity that are generally targeted by a miRNA. Nevertheless, identifying the targets of miRNAs with causal roles in disease should lead to new and unanticipated therapeutic targets and provide new hypotheses for the mechanistic basis of disease. In addition, methods for manipulating miRNAs are being rapidly developed and will be instrumental in the therapeutic modulation of miRNAs. The ability of mutations or single nucleotide polymorphisms to destroy, alter or create new target sequences for miRNAs represents an intriguing source of phenotypic variation and a potential cause of disease, as exemplified by the dramatic skeletal muscle hypertrophy in Texel sheep that results from the creation of a miR-1 and miR-206 target site [26]. Such polymorphisms will likely be difficult to identify, given the degeneracy within miRNA– mRNA interactions and the relatively short sequences of miRNAs and their targets. Mutations within miRNAs and their target sequences also seem likely to contribute to subtle phenotypic variations by augmenting or diminishing expression of target mRNAs rather than having complete on-off effects. Finally, although muscle has been among the most intensely studied cell type with respect to the regulation and mechanisms of action of miRNAs, the principles learned from muscle will undoubtedly apply to other cell types. Acknowledgements Work in E.O.’s laboratory was supported by grants from the National Institutes of Health, the Donald W. Reynolds Cardiovascular Clinical Research Center and the Robert A. Welch Foundation. E.v.R. and N.L. were supported by grants from the American Heart Association.
Trends in Genetics
Vol.24 No.4
References 1 Bartel, D.P. (2004) MicroRNAs: genomics, biogenesis, mechanism, and function. Cell 116, 281–297 2 Berezikov, E. et al. (2005) Phylogenetic shadowing and computational identification of human microRNA genes. Cell 120, 21–24 3 Lee, C.T. et al. (2007) Evolutionary conservation of microRNA regulatory circuits: an examination of microRNA gene complexity and conserved microRNA-target interactions through metazoan phylogeny. DNA Cell Biol. 26, 209–218 4 Stark, A. et al. (2005) Animal MicroRNAs confer robustness to gene expression and have a significant impact on 30 UTR evolution. Cell 123, 1133–1146 5 Vasudevan, S. et al. (2007) Switching from repression to activation: microRNAs can up-regulate translation. Science 318, 1931–1934 6 Denli, A.M. et al. (2004) Processing of primary microRNAs by the Microprocessor complex. Nature 432, 231–235 7 Gregory, R.I. et al. (2004) The Microprocessor complex mediates the genesis of microRNAs. Nature 432, 235–240 8 Lee, Y. et al. (2003) The nuclear RNase III Drosha initiates microRNA processing. Nature 425, 415–419 9 Pillai, R.S. et al. (2007) Repression of protein synthesis by miRNAs: how many mechanisms? Trends Cell Biol. 17, 118–126 10 Valencia-Sanchez, M.A. et al. (2006) Control of translation and mRNA degradation by miRNAs and siRNAs. Genes Dev. 20, 515–524 11 Lewis, B.P. et al. (2005) Conserved seed pairing, often flanked by adenosines, indicates that thousands of human genes are microRNA targets. Cell 120, 15–20 12 Brennecke, J. et al. (2005) Principles of microRNA-target recognition. PLoS Biol. 3, e85 13 Pownall, M.E. et al. (2002) Myogenic regulatory factors and the specification of muscle progenitors in vertebrate embryos. Annu. Rev. Cell Dev. Biol. 18, 747–783 14 Bernstein, E. et al. (2003) Dicer is essential for mouse development. Nat. Genet. 35, 215–217 15 O’Rourke, J.R. et al. (2007) Essential role for Dicer during skeletal muscle development. Dev. Biol. 311, 359–368 16 Potthoff, M.J. and Olson, E.N. (2007) MEF2: a central regulator of diverse developmental programs. Development 134, 4131–4140 17 Zhao, Y. et al. (2005) Serum response factor regulates a muscle-specific microRNA that targets Hand2 during cardiogenesis. Nature 436, 214– 220 18 Rao, P.K. et al. (2006) Myogenic factors that regulate expression of muscle-specific microRNAs. Proc. Natl. Acad. Sci. U. S. A. 103, 8721– 8726 19 Liu, N. et al. (2007) An intragenic MEF2-dependent enhancer directs muscle-specific expression of microRNAs 1 and 133. Proc. Natl. Acad. Sci. U. S. A. 104, 20844–20849 20 Chen, J.F. et al. (2006) The role of microRNA-1 and microRNA-133 in skeletal muscle proliferation and differentiation. Nat. Genet. 38, 228–233 21 Zhao, Y. et al. (2007) Dysregulation of cardiogenesis, cardiac conduction, and cell cycle in mice lacking miRNA-1-2. Cell 129, 303–317 22 Kim, H.K. et al. (2006) Muscle-specific microRNA miR-206 promotes muscle differentiation. J. Cell Biol. 174, 677–687 23 Boutz, P.L. et al. (2007) MicroRNAs regulate the expression of the alternative splicing factor nPTB during muscle development. Genes Dev. 21, 71–84 24 Xu, C. et al. (2007) The muscle-specific microRNAs miR-1 and miR-133 produce opposing effects on apoptosis by targeting HSP60, HSP70 and caspase-9 in cardiomyocytes. J. Cell Sci. 120, 3045–3052 25 McCarthy, J.J. and Esser, K.A. (2007) MicroRNA-1 and microRNA133a expression are decreased during skeletal muscle hypertrophy. J. Appl. Physiol. 102, 306–313 26 Clop, A. et al. (2006) A mutation creating a potential illegitimate microRNA target site in the myostatin gene affects muscularity in sheep. Nat. Genet. 38, 813–818 27 Lee, S.J. and McPherron, A.C. (1999) Myostatin and the control of skeletal muscle mass. Curr. Opin. Genet. Dev. 9, 604–607 28 Tobin, J.F. and Celeste, A.J. (2005) Myostatin, a negative regulator of muscle mass: implications for muscle degenerative diseases. Curr. Opin. Pharmacol. 5, 328–332 29 Yang, B. et al. (2007) The muscle-specific microRNA miR-1 regulates cardiac arrhythmogenic potential by targeting GJA1 and KCNJ2. Nat. Med. 13, 486–491
165
Review
Trends in Genetics Vol.24 No.4
30 Xiao, J. et al. (2007) MicroRNA miR-133 represses HERG K+ channel expression contributing to QT prolongation in diabetic hearts. J. Biol. Chem. 282, 12363–12367 31 van Rooij, E. and Olson, E.N. (2007) MicroRNAs: powerful new regulators of heart disease and provocative therapeutic targets. J. Clin. Invest. 117, 2369–2376 32 van Rooij, E. et al. (2007) Control of stress-dependent cardiac growth and gene expression by a microRNA. Science 316, 575–579 33 Morkin, E. (2000) Control of cardiac myosin heavy chain gene expression. Microsc. Res. Tech. 50, 522–531 34 Park, S.W. et al. (2005) Thyroid hormone-induced juxtaposition of regulatory elements/factors and chromatin remodeling of Crabp1 dependent on MED1/TRAP220. Mol. Cell 19, 643–653 35 Landgraf, P. et al. (2007) A mammalian microRNA expression atlas based on small RNA library sequencing. Cell 129, 1401–1414 36 Miyata, S. et al. (2000) Myosin heavy chain isoform expression in the failing and nonfailing human heart. Circ. Res. 86, 386–390 37 Vadaszova, A. et al. (2004) Influence of thyroid status on the differentiation of slow and fast muscle phenotypes. Physiol. Res. 53 (Suppl 1), S57–S61 38 McGuigan, K. et al. (2004) Evolution of sarcomeric myosin heavy chain genes: evidence from fish. Mol. Biol. Evol. 21, 1042–1056 39 Cheng, Y. et al. (2007) MicroRNAs are aberrantly expressed in hypertrophic heart: do they play a role in cardiac hypertrophy? Am. J. Pathol. 170, 1831–1840 40 Ikeda, S. et al. (2007) Altered microRNA expression in human heart disease. Physiol. Genomics 31, 367–373
41 Sayed, D. et al. (2007) MicroRNAs play an essential role in the development of cardiac hypertrophy. Circ. Res. 100, 416–424 42 Tatsuguchi, M. et al. (2007) Expression of microRNAs is dynamically regulated during cardiomyocyte hypertrophy. J. Mol. Cell. Cardiol. 42, 1137–1141 43 Thum, T. et al. (2007) MicroRNAs in the human heart. A clue to fetal gene reprogramming in heart failure. Circulation 116, 258–267 44 van Rooij, E. et al. (2006) A signature pattern of stress-responsive microRNAs that can evoke cardiac hypertrophy and heart failure. Proc. Natl. Acad. Sci. U. S. A. 103, 18255–18260 45 Eisenberg, I. et al. (2007) Distinctive patterns of microRNA expression in primary muscular disorders. Proc. Natl. Acad. Sci. U. S. A. 104, 17016–17021 46 Care, A. et al. (2007) MicroRNA-133 controls cardiac hypertrophy. Nat. Med. 13, 613–618 47 Landthaler, M. et al. (2004) The human DiGeorge syndrome critical region gene 8 and Its D. melanogaster homolog are required for miRNA biogenesis. Curr. Biol. 14, 2162–2167 48 Krutzfeldt, J. et al. (2005) Silencing of microRNAs in vivo with ‘antagomirs’. Nature 438, 685–689 49 Kwon, C. et al. (2005) MicroRNA1 influences cardiac differentiation in Drosophila and regulates Notch signaling. Proc. Natl. Acad. Sci. U. S. A. 102, 18986–18991 50 Anderson, C. et al. (2006) MIR-206 regulates connexin43 expression during skeletal muscle development. Nucleic Acids Res. 34, 5863–5871 51 Rosenberg, M.I. et al. (2006) MyoD inhibits Fstl1 and Utrn expression by inducing transcription of miR-206. J. Cell Biol. 175, 77–85
Elsevier.com – linking scientists to new research and thinking Designed for scientists’ information needs, Elsevier.com is powered by the latest technology with customer-focused navigation and an intuitive architecture for an improved user experience and greater productivity. The easy-to-use navigational tools and structure connect scientists with vital information – all from one entry point. Users can perform rapid and precise searches with our advanced search functionality, using the FAST technology of Scirus.com, the free science search engine. Users can define their searches by any number of criteria to pinpoint information and resources. Search by a specific author or editor, book publication date, subject area – life sciences, health sciences, physical sciences and social sciences – or by product type. Elsevier’s portfolio includes more than 1800 Elsevier journals, 2200 new books every year and a range of innovative electronic products. In addition, tailored content for authors, editors and librarians provides timely news and updates on new products and services. Elsevier is proud to be a partner with the scientific and medical community. Find out more about our mission and values at Elsevier.com. Discover how we support the scientific, technical and medical communities worldwide through partnerships with libraries and other publishers, and grant awards from The Elsevier Foundation. As a world-leading publisher of scientific, technical and health information, Elsevier is dedicated to linking researchers and professionals to the best thinking in their fields. We offer the widest and deepest coverage in a range of media types to enhance cross-pollination of information, breakthroughs in research and discovery, and the sharing and preservation of knowledge.
Elsevier. Building insights. Breaking boundaries. www.elsevier.com 166
MicroRNAs flex their muscles Eva van Rooij, Ning Liu and Eric N. Olson Department of Molecular Biology, University of Texas Southwestern Medical Center, Dallas, TX 75390-9148, USA
MicroRNAs negatively regulate gene expression by promoting mRNA degradation and inhibiting mRNA translation. Recent studies have uncovered a cadre of muscle-specific microRNAs that regulate diverse aspects of muscle function, including myoblast proliferation, differentiation, contractility and stress responsiveness. These myogenic microRNAs, which are encoded by bicistronic transcripts or are nestled within introns of myosin genes, modulate muscle functions by fine-tuning gene expression patterns or acting as ‘on-off’ switches. Muscle-specific microRNAs also participate in numerous diseases, including cardiac hypertrophy, heart failure, cardiac arrhythmias, congenital heart disease and muscular dystrophy. The myriad roles of microRNAs in muscle biology pose interesting prospects for their therapeutic manipulation in muscle disease. Myriad roles of microRNAs in muscle biology Muscle cells provide a powerful model for understanding the transcriptional circuitry and signaling systems that regulate cell differentiation and organogenesis. The process of muscle development begins when mesodermal stem cells become committed to a muscle cell fate in response to extracellular signals, giving rise to immature muscle cells or myoblasts. The signals and transcription factors that control the specification and differentiation of cardiac and skeletal muscle cells are different, but the differentiation process of both types of muscle is accompanied by the activation of vast arrays of muscle-specific genes encoding the proteins required for the specialized functions of striated muscles, including components of the contractile apparatus, enzymes, receptors and ion channels. Although the transcriptional networks involved in muscle differentiation are well defined, recent studies have shown a new layer of regulation in muscle cell proliferation, differentiation, contractility and stress responsiveness that is mediated by a collection of muscle-specific microRNAs (miRNAs) that are highly conserved in vertebrates (Table 1). Here we discuss the roles of miRNAs in muscle biology, the prospects for manipulating their expression and activities in muscle diseases and important questions for the future. miRNA biogenesis miRNAs are 22 nucleotides long and inhibit translation or promote mRNA degradation by annealing to complementary sequences in the 30 untranslated regions (UTRs) of specific target mRNAs [1]. There are estimated to be as many as 1000 miRNAs encoded by the human genome [2], roughly equaling the number of transcription factors. Corresponding author: Olson, E.N. ([email protected]).
Individual miRNAs can target dozens of mRNAs, and individual mRNAs can be targeted by multiple miRNAs, allowing for enormous complexity and regulatory potential of gene expression. In most cases, miRNAs fine-tune gene expression patterns, but there are also examples in which they act as on-off switches in gene expression [3,4]. The power of miRNAs as regulators of gene expression is also underscored by a recent study demonstrating their ability to up-regulate translation of specific targets [5]. miRNAs are transcribed as long pri-miRNAs that encode one or more miRNAs or are encoded by introns of protein-coding genes. Pri-miRNAs are processed in the nucleus by the RNase Drosha, yielding stem-loop structures of 70 nucleotides. These pre-mRNAs are transported to the cytoplasm where they are further processed by the RNase Dicer, giving rise to the mature miRNA and its complementary strand from the stem-loop, referred to as the ‘star’ strand. The mature miRNA is incorporated into the RNA-induced silencing complex (RISC), where the miRNA strand anneals to the 30 UTRs of target mRNAs, promoting translational repression or mRNA degradation [6–10]. Identification of bona fide mRNA targets for individual miRNAs represents one of the biggest challenges in the field. Nucleotides 2 through 8 at the 50 end of the miRNA, termed the seed sequence, are the most important determinants of mRNA target selection [11]. However, other nucleotides and mRNA secondary structure in the regions surrounding the target sequence also influence the association of miRNAs with their targets [12]. Because of the plethora of potential targets for individual miRNAs, gain and loss of function miRNA phenotypes likely reflect the aggregate effects of perturbing the expression of many mRNA targets rather than a single target. The MADS–miR-1/133 system: a balancing act Skeletal muscle formation requires the commitment of multipotential stem cells to the skeletal muscle lineage [13]. In skeletal muscle cells, the processes of proliferation and differentiation are mutually exclusive. In response to growth factor depletion, proliferating myoblasts exit the cell cycle, fuse to form multinucleated myotubes and activate the transcription of muscle differentiation genes. miRNAs, by modulating the balance between the antagonistic processes of myoblast growth and differentiation, are integral components of the regulatory circuitry for muscle development. The importance of miRNAs in the normal development of vertebrates is evidenced by Dicer loss of function. Dicer-deficient mice die at gastrulation and lack multipotent stem cells [14]. To study the requirement of miRNAs specifically during skeletal muscle development, a conditional Dicer allele was removed by expressing Cre
0168-9525/$ – see front matter ß 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.tig.2008.01.007 Available online 5 March 2008
159
Review
Trends in Genetics Vol.24 No.4
Table 1. Muscle-specific microRNAs miR miR-1
miR-133
miR-206
miR-208
Functions Inhibition of proliferation Control of cardiac conductance Myogenesis Cardiogenesis Inhibition of cardiac hypertrophy Promotion of proliferation Control of cardiac conductance Myogenesis
b-Mhc expression Stress-dependent cardiac remodeling
Validated target a Hand2 lrx5, KCND2
Refs [17] [21]
HDAC4 Delta RhoA, Cdc-42, WHSC2
[20] [49] [46]
SRF HERG
[20] [30]
Cx43 Fst11, Utrn PolA1
[50] [51] [22]
THRAP1
[32]
a
Cdc-42, cell division cycle 42; Cx43, gap junction protein, a 1; Fstl1, follistatin-like 1; HERG, potassium voltage-gated channel, subfamily H, member 2; RhoA, ras homolog gene family, member A; Utrn, utrophin; WHSC2, Wolf-Hirschhorn syndrome candidate 2.
recombinase under control of MyoD regulatory elements, which are expressed in skeletal muscle from day E9.75 onward. Skeletal muscle deletion of Dicer perturbs embryonic and postnatal muscle development, causing skeletal muscle hypoplasia [15]. The MADS (MCM1, agamous, deficiens, serum response factor) box transcription factors, serum response factor (SRF) and myocyte enhancer factor-2 (MEF2), regulate muscle cell proliferation and differentiation through combinatorial interactions with other transcriptional regulators. MEF2 and SRF activate skeletal muscle gene expression in collaboration with myogenic bHLH (basic helix loop helix) proteins, including myogenin and MyoD, whereas these MADS box factors activate cardiac gene expression in conjunction with GATA transcription factors, the homeodomain protein Nkx2–5, myocardin, a transcriptional coactivator and others [16]. In addition to regulating the expression of muscle structural genes, SRF and MEF2 control the expression of two pairs of muscle-specific miRNAs, miR-1–1/133a-2 and miR1–2/133a-1, which are encoded by bicistronic pre-miRNAs (Figure 1). Muscle-specific enhancers located upstream
and within the introns of miR-1–1/133a-2 and miR-1–2/ 133a-1 control transcription in differentiated muscle cells [17–19]. SRF and MEF2 cooperate with MyoD to activate transcription of these miRNAs in skeletal muscle. In the heart, the upstream enhancer relies on SRF for expression in ventricular and atrial myocytes, whereas MEF2 directs ventricular expression from the intronic enhancer (Figure 2) [17,19]. A third pair of related muscle-specific miRNAs, miR206/133b, is expressed specifically in skeletal muscle [20]. miR-1–1, miR-1–2 and miR-206 differ at only three nucleotides, and miR-133a-2, miR-133a-1 and miR133b are identical except for one nucleotide at the 30 end of miR-133b, indicating that each of these miRNA trios are likely to target the same mRNAs (Figure 1). Such redundancy complicates functional analysis by traditional gene knockouts in the mouse. miR-1 and miR-133 modulate muscle growth and differentiation by regulating SRF and MEF2 activity, thereby establishing negative feedbacks loops within muscle cell lineages [20] (Figure 2). miR-1 represses the expression of histone deacetylase 4 (HDAC4) [20], which acts as a signaldependent repressor of muscle differentiation together with MEF2 [16]. Thus, miR-1 up-regulation during differentiation is a mechanism to dampen HDAC4 expression and to potentiate MEF2 pro-myogenic activity (Figure 2). Forced miR-1 overexpression in the embryonic heart results in lethality during mid-embryogenesis because of a deficiency of cardiomyocytes and consequent cardiac failure [17]. These findings suggest that miR-1 inhibits cardiomyocyte proliferation, consistent with the presence of excessive cardiomyocytes in miR-1–2/ mice [21]. Among its many predicted targets, miR-1 represses the translation of Hand2 [17], a bHLH transcription factor that is required for cardiac growth during embryogenesis (Figure 3). miR-206 promotes skeletal muscle differentiation, whereas miR-206 knockdown, through antisense oligonucleotides, prevents myoblasts from exiting the cell cycle and blocks differentiation [18,22]. Consistent with these findings, miR-206 negatively regulates DNA polymerase a (polA1) translation, thereby inhibiting DNA synthesis [22].
Figure 1. Bicistronic muscle-specific microRNA (miRNA) clusters. The figure shows schematic diagrams of bicistronic muscle-specific miRNA genes and their sequence homologies. (a) Three bicistronic miRNA genes encoding miR-1–1/miR-133a-2, miR-1–2/miR-133a-1 and miR-206/miR-133b are shown. Each pair of miRNAs is transcribed from left to right. Individual miRNAs, designated by colored rectangles, are generated by processing of the bicistronic miRNA transcripts. Chromosomal locations of each miRNA gene and the muscle tissues in which they are expressed are shown. (b) Sequences homologies among muscle-specific miRNAs are shown. Nonhomology is indicated by black letters.
160
Review
Trends in Genetics
Vol.24 No.4
Figure 2. Control of skeletal muscle proliferation and differentiation by miR-1 and miR-133. In skeletal muscle cells, the process of proliferation is antagonistic to the process of myoblast differentiation. miR-1 and miR-133 function at the center of a network of transcription factors to regulate skeletal myoblast proliferation and differentiation. (a) Myocyte enhancer factor-2 (MEF2) and myogenic basic helix loop helix (bHLH) proteins, including MyoD (green), regulate their own expression, as well as the expression of downstream muscle structural genes. Additionally, these transcription factors use upstream and intragenic enhancers to activate transcription of bicistronic pri-miRNAs encoding miR-1 (blue rectangle) and miR-133 (black rectangle) in differentiated skeletal muscle. miR-1 represses expression of HDAC4 (purple), a signal-dependent repressor of MEF2 (red) activity, thereby establishing a negative feedback loop to modulate miR-1 and miR-133 expression and promoting myoblast differentiation. miR-133 represses expression of serum response factor (SRF; yellow), a positive regulator of miR-1/133 expression and repressor of myoblast proliferation. (b) E11.5 transgenic mouse embryos, which harbor a lacZ transgene controlled by the miR-1–2/133a-1 intragenic enhancer, are shown. The wildtype (WT) enhancer directs transgene expression in heart and skeletal muscle (note the blue color), whereas an enhancer with a mutation in the MEF2 site is not expressed in skeletal muscle or ventricular myocardium.
By contrast, miR-133 has been proposed to promote myoblast proliferation, a role opposite to that of miR-1 [20]. The ability of miR-133 to promote proliferation has been ascribed to the repression of SRF, an essential regulator of muscle differentiation (Figure 2). miR-133 also represses translation of the polypyrimidine tract-binding protein (nPTB), which promotes differential splicing of a variety of transcripts that influence the muscle differentiation program [23]. It seems counter-intuitive that miR-1 and miR-133, which are encoded by the same MEF2regulated bicistronic transcripts, would exert opposing effects on muscle growth and differentiation; these miRs also play opposing roles in cardiomyocyte apoptosis, with miR-1 being pro-apoptotic and miR-133 being anti-apoptotic [24]. However, both miR-1 and miR-133 fine-tune key regulatory pathways in an antagonistic manner, with the balance being tipped one way or the other by additional transcription factors and regulatory pathways. Further evidence for the involvement of muscle-specific miRs in the control of skeletal muscle growth comes from the observation that functional overload of skeletal muscle, a stimulus for hypertrophy, is accompanied by miR-1–2 and miR-133a-2 down-regulation and miR-206 up-regulation [25]. Moreover, a mutation that causes dramatic muscularity in the Texel strain of sheep has been mapped to a single G-to-A mutation within the 30 UTR of the mRNA
encoding myostatin, a member of the transforming growth factor b (TGF-b) family of growth factors that represses muscle growth. This mutation creates a target site for miR1 and miR-206, resulting in myostatin translational repression [26], which phenocopies the ‘muscling’ phenotype that results from Myostatin loss of function mutations in mice, cattle and humans [27,28]. miR-1 and miR-133 have also been implicated in the modulation of electrical conductance within the heart. miR-1–2/ mice that survive to adulthood display cardiac conduction abnormalities, accompanied by dysregulation of the potassium channel, Kcnd2 [21]. These defects are likely attributable to developmental abnormalities, because antisense-mediated miR-1 knockdown in the adult heart suppressed arrhythmias that resulted from myocardial infarction [29]. Conversely, miR-1 overexpression in the myocardium promoted ischemic arrhythmias [29]. miR-133 has also been linked to cardiac arrhythmias. Exogenous miR-133 expression in cardiomyocytes inhibits expression of ether-a-go-go related gene (HERG), which encodes a K+ channel responsible for rapid delayed rectifier K+ current (IKr) [30]. Together, these findings suggest that the heart is exquisitely sensitive to the level of miR-1 or miR-133 expression such that an increase or decrease in their expression renders the heart susceptible to arrhythmogenic events. 161
Review
Trends in Genetics Vol.24 No.4
Figure 3. Control of cardiac growth, differentiation and conductance by miR-1 and miR-133. miR-1 and miR-133 function at the center of a network of transcription factors to regulate ventricular growth, cardiomyocyte differentiation and conductance of the heart. (a) miR-1(blue) and miR-133 (black) are expressed in cardiac muscle from a bicistronic pri-miRNA. miR-1 represses expression of HDAC4 (purple), a signal-dependent repressor of myocyte enhancer factor-2 (MEF2; red) activity, thereby establishing a negative feedback loop to modulate miR-1 and miR-133 expression in ventricular myocardium. miR-1 also represses expression of Hand2 (green), a positive regulator of proliferation, and Irx5 (Iroquois homeobox 5; red), which regulates the expression of Kcdn2, a K+ channel involved in cardiac repolarization. miR-133 represses expression of serum response factor (SRF; yellow), a positive regulator of miR-1 and miR-133 expression. MEF2 and SRF rely on additional cofactors (orange) that are not yet identified, to activate transcription of miR-1 and miR-133 and other target genes in the heart. (b) Hearts from E11.5 transgenic mouse embryos, which harbor a lacZ transgene controlled by the miR-1–2/133a-1 intragenic enhancer, are shown. The wildtype (WT) enhancer directs expression in atrium and ventricles (note blue color), whereas an enhancer with a mutation in the MEF2 site is not expressed in the ventricular myocardium. la, left atrium; lv, left ventricle; ra, right atrium; rv, right ventricle.
Regulation of stress-responsiveness and muscle cell identity by miR-208 Acute and chronic injury to the heart results in hypertrophic growth, which frequently culminates in a loss of pump function, fibrosis and lethal arrhythmias [31]. A hallmark of heart disease is a switch in expression of myosin heavy chain (MHC) genes from a-MHC, a fast contracting myosin, to b-MHC, a slow, embryonic myosin. Intriguingly, intron 29 of the gene that encodes a-MHC gives rise to the sole cardiac-specific miRNA, miR-208, which plays a central role in the regulation of the a- to b-MHC switch, and in other aspects of the cardiac stress response, such as cardiomyocyte hypertrophy and fibrosis [32] (Figure 4). Mice with a homozygous deletion of the miR-208 coding region are viable, but they fail to upregulate b-MHC expression in response to pressure overload or activated calcineurin (a potent inducer of cardiac 162
remodeling). a- and b-MHC are also antithetically regulated by thyroid hormone (T3): T3 induces a-MHC, whereas hypothyroidism results in a dramatic increase in b-MHC expression [33]. In the absence of miR-208, b-MHC expression is potently repressed during the hypothyroid state. By contrast, b-MHC expression during embryogenesis, when it is still responsive to embryonic transcription factors like Nkx2.5 and MEF2 [34], is unaffected in miR208/ mice. Thus, miR-208 is dedicated specifically to the activation of b-MHC in response to stress and hypothyroidism in the adult heart and acts as an on-off switch for bMHC expression rather than a fine-tuner of gene expression. Among the top predicted miR-208 targets is the thyroid hormone receptor coregulator THRAP1, which is a cofactor of the thyroid hormone receptor and can either act as an activator or repressor [34]. Because of the repressive effect
Review
Trends in Genetics
Vol.24 No.4
Figure 4. Modulation of cardiac muscle phenotypes by miR-208. The a- and b-myosin heavy chain (MHC) genes are linked in the genome. a-MHC promotes fast cardiac muscle contraction, whereas b-MHC promotes slow contraction. miR-208, encoded by the a-MHC gene, negatively regulates the expression of THRAP1 and other targets. In the absence of miR-208, b-MHC expression becomes nonresponsive to stress signaling and hypothyroidism. miR-208 absence also results in the up-regulation of fast skeletal muscle genes in the heart and a block in pathological cardiac remodeling. The short-term consequences of the lack of miR-208 are the maintenance of cardiac function after stress. However, miR-208–null mice eventually display abnormalities in cardiac contractility, which might reflect the inappropriate expression of skeletal muscle genes in the heart.
on b-MHC transcription specifically after birth, we proposed that miR-208 normally modulates THRAP1 expression, such that, in the absence of miR-208, THRAP1 expression is up-regulated along with consequent enhancement of its actions, whether positive or negative [32]. By example, increased THRAP1 expression is predicted to enhance the repressive influence of the TR on the negative thyroid hormone response element (TRE) in the b-MHC promoter. Further research is needed to determine how other potential miR-208 targets regulate cardiac physiology. Intriguingly, the genome contains an additional copy of miR-208, miR-208b, which is located within the b-MHC gene [2,35]. miR-208 and miR-208b share the same seed
sequence but differ at three nucleotides in their 30 region (Figure 5). The expression and functions of miR-208b have not yet been reported, but it is likely that miR-208b, like miR-208, is co-expressed with its host myosin gene, i.e. predominantly in slow skeletal muscle and in the heart after stress. Because a- and b-MHC are counterbalanced in the heart, the expression of miR-208/miR-208b is likely maintained at relatively constant levels. Thus, although aMHC is the dominant MHC isoform in mice and b-MHC dominates in humans [36], the relative expression of miR208 versus miR-208b will differ accordingly in mice versus humans. Because miR-208 is required for cardiac b-MHC expression in response to stress and hypothyroidism in mice, it is tempting to speculate that miR-208b might play
Figure 5. A family of MyomiRs. Three microRNAs (miRNAs) encoded by introns of myosin heavy chain (MHC) genes share homology, suggesting they target overlapping sets of mRNAs. We refer to these miRNAs as Myomirs because they are encoded by myosin genes and are expressed specifically in striated muscle cells. These miRNAs and their host MHC genes are expressed specifically in slow skeletal muscle and/or heart. (a) miR-208 is encoded by intron 29 of the a-MHC gene (Myh6), miR-208b by intron 31 of b-MHC (Myh7) and miR-499 by intron 19 of Myh7b. Each myosin gene is transcribed from left to right; both the muscle tissues in which they are expressed and their chromosomal locations are shown. (b) Sequences homologies among muscle-specific miRNAs are shown. Nonhomology is indicated by black letters.
163
Review a more important role in the human shift toward b-MHC during heart disease. Nonetheless, these data indicate that there must exist an intimate form of cross-talk between these miRNAs and the myosin genes in which they are embedded. Although miR-208 is cardiac specific, fast skeletal muscle genes are inappropriately expressed in miR208/ hearts, suggesting that one function of miR-208 is to repress the expression of a transcriptional activator of the fast skeletal muscle gene program [32] (Figure 4). It is intriguing in this regard that thyroid hormone promotes the formation of fast myofibers and represses slow myofiber development [37]. Whether miR-208 acts through THRAP1 to control thyroid hormone signaling, stress-dependent cardiac growth and myofiber specification or whether additional targets are involved remains to be determined. Because the sequence homology between miR-208 and miR-208b indicates that they share common mRNA targets, it will be interesting to see whether miR208b fulfills a comparable function to miR-208 in repressing fast skeletal genes. It is tempting to speculate that bMHC gene expression plays a dual role in slow skeletal muscle, not only by regulating the speed and efficiency of contraction, but also by encoding miR-208b, which represses the expression of fast skeletal genes and thereby determines slow myofiber identity. Another more ancient myosin, Myh7b, like b-MHC, is expressed specifically in the heart and slow skeletal muscles [38]. An intron within Myh7b encodes another miRNA, miR-499. Although the sequence of the mature miR-499 miRNA differs substantially from miR-208 and miR-208b, the seed region contains six overlapping bases, hinting at possible shared functions (Figure 5). We refer to these miRs as MyomiRs, because of their location within myosin genes, their specific expression in myogenic cells and the ability of miR-208 to regulate myosin switching and myofiber identity. The existence of a family of closely related musclespecific miRNAs encoded by MHC genes raises many interesting questions. Because it seems counterintuitive that miR-208 regulates the expression of b-MHC, which itself encodes a miRNA closely related to miR-208, miR208b, it will be intriguing to determine to what extent the functions and targets of the MyomiRs actually are redundant. Another interesting facet of the MyomiR network is the hierarchical relationships among these miRs; the regulation of b-MHC and thereby miR-208b, is miR-208 dependent, which might be both species specific and stress dependent. Another interesting point is how and why striated muscle evolved a regulatory network in which miRNAs are embedded in their target myosin genes, and what are the mechanisms whereby these miRNAs regulate the cardiac stress response and myofiber identity? The answers to these questions promise to provide new and important insights into the fundamental logic of muscle gene expression and development, as well as the regulatory circuits that drive cardiac disease. Looking toward miRNA-based therapeutics The key roles of miRNAs in such a diversity of muscle functions raises interesting prospects for therapeutic 164
Trends in Genetics Vol.24 No.4
manipulation of miRNA-based mechanisms in the settings of muscle diseases. For example, the diminished hypertrophy, fibrosis and fetal gene activation observed in miR208/ mice in response to cardiac stress raises the possibility that therapeutic suppression of miR-208 expression might enhance cardiac function after acute or chronic injury [32]. Similarly, numerous stress-responsive miRNAs, including miR-21 and miR-195, are up-regulated during heart disease, some of which have been shown to be sufficient and necessary for cardiac dysfunction [39–44]. Diminishing the expression of these pathogenic miRNAs could provide therapeutic benefit. However, in contrast to miR-208, which is cardiac-specific, the latter miRNAs are widely expressed, which poses challenges in preventing undesired effects on noncardiac tissues. Specific miRNA expression patterns are also associated with muscular dystrophies and other myopathies [45]. It will be of interest to determine whether these disease-associated miRNAs participate directly in the disease process and, if so, whether their therapeutic modulation might influence disease severity or progression. It is also conceivable that therapeutic elevation of specific miRs could enhance muscle function. miR-133, for example, has been implicated in the blockade to cardiac hypertrophy [46], suggesting that its elevation could be beneficial under conditions of cardiac stress. However, it should also be kept in mind that elevation or diminution of miR-133 expression has been linked to cardiac arrhythmias [30], demonstrating the sensitivity of cardiac function to fluctuations in this miRNA and pointing to the possible adverse effects of manipulating miRNA expression in vivo. Similarly, modulation of miR-1 expression, either up or down, causes dysregulation of cardiac ion channels and consequent conduction defects [21,29]. Ventricular-septal and atrial-septal defects in miR-1 mutant mice indicate an important role for this miRNA, and probably many others, in heart development [21]. In this regard, DiGeorge syndrome, a chromosomal deletion disorder resulting in cardiac and craniofacial abnormalities, is associated with deletion of the DiGeorge syndrome critical region 8 gene (DGCR8), which encodes a component of the RISC complex [47]. Whether DGCR8 deletion contributes to congenital abnormalities by perturbing the functions of miRNAs that direct these processes remains to be determined. How might the biology of miRNAs be exploited in future clinical applications? One can imagine inhibiting the expression of miRNAs involved in disease pathogenesis using antimiRs. Several promising approaches involving chemically modified oligonucleotides complementary to miRNAs have recently been described [48]. It is also conceivable that miRNA mimics could be developed to enhance the expression of beneficial miRNAs. Customized anti-miRs or miR mimics directed against specific mRNA targets can also be envisioned. Another potentially powerful therapeutic approach will be the development of drugs that modulate the expression or activities of disease-related miRNAs. Not withstanding these opportunities, there are several important hurdles that remain to be overcome, such as efficient delivery and specificity of anti-miRs or miR mimics in vivo. This is of particular concern given that
Review individual miRNAs can have dozens or even hundreds of targets, such that blockade of a miRNA against one set of targets could be salutary whereas eliminating repression of another set of targets could be pathological. Off-target effects also represent a significant challenge, especially considering that miRNA-mediated repression often requires homology of only six to seven nucleotides in the seed region of the miRNA and the mRNA target. Toxicity of chemical modifications used to facilitate cellular uptake and prevent degradation also represents an important consideration. The modification of antagomirs with cholesterol, for example, poses problems of liver toxicity. Finally, it is apparent that miRNA expression patterns are dynamically regulated during disease and provide molecular signatures for disease progression. Thus, it is likely that miRNA expression patterns will be used for disease diagnosis and prognosis, as shown recently for numerous forms of heart disease [31]. Concluding remarks Our understanding of micro RNA (miRNA) biology is still in its infancy. It has been estimated that at least one third of mammalian genes are regulated by as many as a thousand miRs, only a few of which have been studied in any detail. An important challenge for the future will be to identify the downstream targets that mediate the actions of miRNAs in development and disease. Ascribing the actions of a particular miRNA to specific targets is especially challenging, given the plethora of mRNAs of both high and low affinity that are generally targeted by a miRNA. Nevertheless, identifying the targets of miRNAs with causal roles in disease should lead to new and unanticipated therapeutic targets and provide new hypotheses for the mechanistic basis of disease. In addition, methods for manipulating miRNAs are being rapidly developed and will be instrumental in the therapeutic modulation of miRNAs. The ability of mutations or single nucleotide polymorphisms to destroy, alter or create new target sequences for miRNAs represents an intriguing source of phenotypic variation and a potential cause of disease, as exemplified by the dramatic skeletal muscle hypertrophy in Texel sheep that results from the creation of a miR-1 and miR-206 target site [26]. Such polymorphisms will likely be difficult to identify, given the degeneracy within miRNA– mRNA interactions and the relatively short sequences of miRNAs and their targets. Mutations within miRNAs and their target sequences also seem likely to contribute to subtle phenotypic variations by augmenting or diminishing expression of target mRNAs rather than having complete on-off effects. Finally, although muscle has been among the most intensely studied cell type with respect to the regulation and mechanisms of action of miRNAs, the principles learned from muscle will undoubtedly apply to other cell types. Acknowledgements Work in E.O.’s laboratory was supported by grants from the National Institutes of Health, the Donald W. Reynolds Cardiovascular Clinical Research Center and the Robert A. Welch Foundation. E.v.R. and N.L. were supported by grants from the American Heart Association.
Trends in Genetics
Vol.24 No.4
References 1 Bartel, D.P. (2004) MicroRNAs: genomics, biogenesis, mechanism, and function. Cell 116, 281–297 2 Berezikov, E. et al. (2005) Phylogenetic shadowing and computational identification of human microRNA genes. Cell 120, 21–24 3 Lee, C.T. et al. (2007) Evolutionary conservation of microRNA regulatory circuits: an examination of microRNA gene complexity and conserved microRNA-target interactions through metazoan phylogeny. DNA Cell Biol. 26, 209–218 4 Stark, A. et al. (2005) Animal MicroRNAs confer robustness to gene expression and have a significant impact on 30 UTR evolution. Cell 123, 1133–1146 5 Vasudevan, S. et al. (2007) Switching from repression to activation: microRNAs can up-regulate translation. Science 318, 1931–1934 6 Denli, A.M. et al. (2004) Processing of primary microRNAs by the Microprocessor complex. Nature 432, 231–235 7 Gregory, R.I. et al. (2004) The Microprocessor complex mediates the genesis of microRNAs. Nature 432, 235–240 8 Lee, Y. et al. (2003) The nuclear RNase III Drosha initiates microRNA processing. Nature 425, 415–419 9 Pillai, R.S. et al. (2007) Repression of protein synthesis by miRNAs: how many mechanisms? Trends Cell Biol. 17, 118–126 10 Valencia-Sanchez, M.A. et al. (2006) Control of translation and mRNA degradation by miRNAs and siRNAs. Genes Dev. 20, 515–524 11 Lewis, B.P. et al. (2005) Conserved seed pairing, often flanked by adenosines, indicates that thousands of human genes are microRNA targets. Cell 120, 15–20 12 Brennecke, J. et al. (2005) Principles of microRNA-target recognition. PLoS Biol. 3, e85 13 Pownall, M.E. et al. (2002) Myogenic regulatory factors and the specification of muscle progenitors in vertebrate embryos. Annu. Rev. Cell Dev. Biol. 18, 747–783 14 Bernstein, E. et al. (2003) Dicer is essential for mouse development. Nat. Genet. 35, 215–217 15 O’Rourke, J.R. et al. (2007) Essential role for Dicer during skeletal muscle development. Dev. Biol. 311, 359–368 16 Potthoff, M.J. and Olson, E.N. (2007) MEF2: a central regulator of diverse developmental programs. Development 134, 4131–4140 17 Zhao, Y. et al. (2005) Serum response factor regulates a muscle-specific microRNA that targets Hand2 during cardiogenesis. Nature 436, 214– 220 18 Rao, P.K. et al. (2006) Myogenic factors that regulate expression of muscle-specific microRNAs. Proc. Natl. Acad. Sci. U. S. A. 103, 8721– 8726 19 Liu, N. et al. (2007) An intragenic MEF2-dependent enhancer directs muscle-specific expression of microRNAs 1 and 133. Proc. Natl. Acad. Sci. U. S. A. 104, 20844–20849 20 Chen, J.F. et al. (2006) The role of microRNA-1 and microRNA-133 in skeletal muscle proliferation and differentiation. Nat. Genet. 38, 228–233 21 Zhao, Y. et al. (2007) Dysregulation of cardiogenesis, cardiac conduction, and cell cycle in mice lacking miRNA-1-2. Cell 129, 303–317 22 Kim, H.K. et al. (2006) Muscle-specific microRNA miR-206 promotes muscle differentiation. J. Cell Biol. 174, 677–687 23 Boutz, P.L. et al. (2007) MicroRNAs regulate the expression of the alternative splicing factor nPTB during muscle development. Genes Dev. 21, 71–84 24 Xu, C. et al. (2007) The muscle-specific microRNAs miR-1 and miR-133 produce opposing effects on apoptosis by targeting HSP60, HSP70 and caspase-9 in cardiomyocytes. J. Cell Sci. 120, 3045–3052 25 McCarthy, J.J. and Esser, K.A. (2007) MicroRNA-1 and microRNA133a expression are decreased during skeletal muscle hypertrophy. J. Appl. Physiol. 102, 306–313 26 Clop, A. et al. (2006) A mutation creating a potential illegitimate microRNA target site in the myostatin gene affects muscularity in sheep. Nat. Genet. 38, 813–818 27 Lee, S.J. and McPherron, A.C. (1999) Myostatin and the control of skeletal muscle mass. Curr. Opin. Genet. Dev. 9, 604–607 28 Tobin, J.F. and Celeste, A.J. (2005) Myostatin, a negative regulator of muscle mass: implications for muscle degenerative diseases. Curr. Opin. Pharmacol. 5, 328–332 29 Yang, B. et al. (2007) The muscle-specific microRNA miR-1 regulates cardiac arrhythmogenic potential by targeting GJA1 and KCNJ2. Nat. Med. 13, 486–491
165
Review
Trends in Genetics Vol.24 No.4
30 Xiao, J. et al. (2007) MicroRNA miR-133 represses HERG K+ channel expression contributing to QT prolongation in diabetic hearts. J. Biol. Chem. 282, 12363–12367 31 van Rooij, E. and Olson, E.N. (2007) MicroRNAs: powerful new regulators of heart disease and provocative therapeutic targets. J. Clin. Invest. 117, 2369–2376 32 van Rooij, E. et al. (2007) Control of stress-dependent cardiac growth and gene expression by a microRNA. Science 316, 575–579 33 Morkin, E. (2000) Control of cardiac myosin heavy chain gene expression. Microsc. Res. Tech. 50, 522–531 34 Park, S.W. et al. (2005) Thyroid hormone-induced juxtaposition of regulatory elements/factors and chromatin remodeling of Crabp1 dependent on MED1/TRAP220. Mol. Cell 19, 643–653 35 Landgraf, P. et al. (2007) A mammalian microRNA expression atlas based on small RNA library sequencing. Cell 129, 1401–1414 36 Miyata, S. et al. (2000) Myosin heavy chain isoform expression in the failing and nonfailing human heart. Circ. Res. 86, 386–390 37 Vadaszova, A. et al. (2004) Influence of thyroid status on the differentiation of slow and fast muscle phenotypes. Physiol. Res. 53 (Suppl 1), S57–S61 38 McGuigan, K. et al. (2004) Evolution of sarcomeric myosin heavy chain genes: evidence from fish. Mol. Biol. Evol. 21, 1042–1056 39 Cheng, Y. et al. (2007) MicroRNAs are aberrantly expressed in hypertrophic heart: do they play a role in cardiac hypertrophy? Am. J. Pathol. 170, 1831–1840 40 Ikeda, S. et al. (2007) Altered microRNA expression in human heart disease. Physiol. Genomics 31, 367–373
41 Sayed, D. et al. (2007) MicroRNAs play an essential role in the development of cardiac hypertrophy. Circ. Res. 100, 416–424 42 Tatsuguchi, M. et al. (2007) Expression of microRNAs is dynamically regulated during cardiomyocyte hypertrophy. J. Mol. Cell. Cardiol. 42, 1137–1141 43 Thum, T. et al. (2007) MicroRNAs in the human heart. A clue to fetal gene reprogramming in heart failure. Circulation 116, 258–267 44 van Rooij, E. et al. (2006) A signature pattern of stress-responsive microRNAs that can evoke cardiac hypertrophy and heart failure. Proc. Natl. Acad. Sci. U. S. A. 103, 18255–18260 45 Eisenberg, I. et al. (2007) Distinctive patterns of microRNA expression in primary muscular disorders. Proc. Natl. Acad. Sci. U. S. A. 104, 17016–17021 46 Care, A. et al. (2007) MicroRNA-133 controls cardiac hypertrophy. Nat. Med. 13, 613–618 47 Landthaler, M. et al. (2004) The human DiGeorge syndrome critical region gene 8 and Its D. melanogaster homolog are required for miRNA biogenesis. Curr. Biol. 14, 2162–2167 48 Krutzfeldt, J. et al. (2005) Silencing of microRNAs in vivo with ‘antagomirs’. Nature 438, 685–689 49 Kwon, C. et al. (2005) MicroRNA1 influences cardiac differentiation in Drosophila and regulates Notch signaling. Proc. Natl. Acad. Sci. U. S. A. 102, 18986–18991 50 Anderson, C. et al. (2006) MIR-206 regulates connexin43 expression during skeletal muscle development. Nucleic Acids Res. 34, 5863–5871 51 Rosenberg, M.I. et al. (2006) MyoD inhibits Fstl1 and Utrn expression by inducing transcription of miR-206. J. Cell Biol. 175, 77–85
Elsevier.com – linking scientists to new research and thinking Designed for scientists’ information needs, Elsevier.com is powered by the latest technology with customer-focused navigation and an intuitive architecture for an improved user experience and greater productivity. The easy-to-use navigational tools and structure connect scientists with vital information – all from one entry point. Users can perform rapid and precise searches with our advanced search functionality, using the FAST technology of Scirus.com, the free science search engine. Users can define their searches by any number of criteria to pinpoint information and resources. Search by a specific author or editor, book publication date, subject area – life sciences, health sciences, physical sciences and social sciences – or by product type. Elsevier’s portfolio includes more than 1800 Elsevier journals, 2200 new books every year and a range of innovative electronic products. In addition, tailored content for authors, editors and librarians provides timely news and updates on new products and services. Elsevier is proud to be a partner with the scientific and medical community. Find out more about our mission and values at Elsevier.com. Discover how we support the scientific, technical and medical communities worldwide through partnerships with libraries and other publishers, and grant awards from The Elsevier Foundation. As a world-leading publisher of scientific, technical and health information, Elsevier is dedicated to linking researchers and professionals to the best thinking in their fields. We offer the widest and deepest coverage in a range of media types to enhance cross-pollination of information, breakthroughs in research and discovery, and the sharing and preservation of knowledge.
Elsevier. Building insights. Breaking boundaries. www.elsevier.com 166