Roles of MicroRNAs and Myocardial Cell Differentiation

As drug therapy is of limited efficacy in the treatment of heart diseases related to loss of cardiomyocytes, which have very poor division potential, regenerative medicine is expected to be a new strategy to address regenerative treatment in cardiac diseases. To achieve myocardial regeneration, elucidation of the mechanism of myocardial differentiation from stem cells is essential. Myocardial differentiation from embryonic pluripotent stem cells has been investigated worldwide, and remarkable developments such as establishment of induced pluripotent stem cells and transformation of somatic cells to cardiomyocytes have recently been made, markedly changing the strategy of regenerative medicine. At the same time, the close involvement of microRNA in the maintenance, proliferation, differentiation, and reprogramming of these stem cells has been revealed. In this report, microRNA is outlined, focusing on its role in myocardial differentiation.

Introduction

Cardiomyocytes proliferate and form the heart in the embryonic period, but proliferation stops soon after birth. Therefore, it is difficult to cure systolic dysfunction (heart failure), particularly in cases accompanied by myocardial injury, such as myocardial infarction and cardiomyopathy. Drug therapy for heart failure using β-blockers and angiotensin-converting enzyme inhibitors improves the short-term prognosis, but the 5-year survival rate of patients with severe heart failure is less than 50%. To overcome this situation, regenerative therapy has been investigated worldwide wherein cardiovascular cell differentiation of stem cells has been induced and applied to supplement and regenerate the affected regions. As a source of cells for regenerative medicine, somatic stem cells such as bone-marrow-derived cells and vascular endothelial precursor cells, and pluripotent stem cells, such as embryonic stem (ES) cells and induced pluripotent stem (iPS) cells, have been considered. Studies on direct transformation of somatic cells to the target cells, such as cardiomyocytes, have also been progressing.

In parallel with this trend, an important role of microRNA (miRNA or miR) in the maintenance, proliferation, differentiation, and reprogramming of stem cells has recently been revealed. The small RNA present in C. elegans was the miRNA initially discovered and reported in Cell in 1993 by Ambros and Ruvkun,1, 2 in which the phenomenon currently accepted widely had already been pointed out: this RNA has an antisense sequence complementary to a specific gene sequence and performs the posttranscriptional regulation of gene expression. But the fact that this small RNA is present in plants through to humans and is universally involved in biological activity, that is, the name “microRNA” and its concept, was not established until Fire and Mello et al. reported RNA interference in 1998, which led to their being later awarded the Nobel Prize.³ After the establishment of the concept of microRNAs, the importance of their functions in mammals has been revealed in various fields, and their involvement in cardiogenesis and development of cardiac hypertrophy, heart failure, and arrhythmia has been reported. In mice, impairment of the synthesis of all miRNAs in the heart leads to cardiac hypofunction and subsequently death. Therefore, miRNAs are essential for the heart, and changes in their expressions may also influence pathophysiology in humans. The association between miRNA and myocardial differentiation.

ES cells derived from the inner cell mass of fertilized ova before implantation (blastocysts) have the potential to differentiate into all three germ layers (pluripotency) and infinitely self-replicate while maintaining the undifferentiated condition (self-renewal).⁴ Accordingly, if appropriate induction of ES cell differentiation in a mass culture becomes possible, the cells can be used as a cell source for regenerative medicine in various fields.

To investigate the function of miRNA in development and differentiation, mice and ES cells in which miRNA synthesis was abolished by knocking out (KO) RNase III gene Dicer1 and double-stranded RNA-binding protein gene Dgcr8 have been prepared. Dicer1 KO mice undergo embryonic death by E7.5, and pluripotency factor Oct4 is not expressed.⁵ Similarly, transgenic mice in which heart-muscle-specific KO of Dicer1 starts on E8.5 die because of cardiac-hypoplasia-associated heart failure by E12.5.⁶ In Dicer1 KO ES cells, Oct4 is expressed, but their division potential is markedly defective; no endo- and ectodermal markers are expressed even after embryoid body (EB) formation-induced differentiation, and proliferation stops on day 8.7, 8 These findings reveal that miRNA is essential for proliferation and differentiation of cells including cardiogenesis. The phenotype of Dgcr8 KO ES cells is slightly different from that of Dicer1 KO, and proliferation and differentiation continue after 16 days of EB formation. Oct4 and Nanog are expressed in Dgcr8 KO ES cells, suggesting that several miRNAs promote ES cell differentiation by repressing pluripotency factors.

Retinoic acid used to induce myocardial differentiation enhances miR-134, miR-296, and miR-470 expressions. These miRNAs promote ES cell differentiation by targeting Sox2, Nanog, and Oct4.⁹ miR-145 directly targets Oct4, Sox2, and Klf4 and inhibits self-replication of ES cells, promoting differentiation.⁸ miR-145 is inhibited by Oct4 in undifferentiated cells, but its expression is enhanced with progression of differentiation.

Inversely, the expression levels of many miRNAs specifically expressed in undifferentiated ES cells decrease with EB-formation-induced differentiation.¹⁰ For example, the ES-cell-specific miR-290 cluster indirectly regulates DNA methylation by targeting Rbl-2, but the levels of this miRNA decrease as differentiation progresses.11, 12

To efficiently induce stem cell differentiation into target cells, it may be necessary to cancel miRNA-dependent controlled self-replication and multipotency (Fig. 1).

Many transcription factors, such as Nkx2.5, GATA4, Tbx5, myocyte enhancer factor-2 (MEF2), and serum response factor (SRF), have been identified as cardiomyocyte precursor cells or essential genes for morphogenesis of the heart in previous studies on cardiogenesis.¹³ They are also known to play important roles in myocardial differentiation of ES cells, but no master gene, such as MyoD in skeletal muscle cells, has been discovered. It has been suggested that the development/differentiation program is operated by gene expression control by networks of several transcription and regulatory factors in the heart and heart muscle. In addition, involvement of miRNA in cardiogenesis and myocardial differentiation has recently been clarified (Fig. 2).

miRNAs are distinguished on the basis of their sequence. The presence of several hundred species has been reported in mammals so far, but, basically, they are synthesized through a common pathway. They are transcribed from genomic DNA, go through two precursors, primary- and precursor-miRNAs (pri- and pre-miRNAs, respectively), and finally become mature miRNAs. When the function of a gene is analyzed, experimental systems of acquisition or inhibition of function, or both, are constructed, but to evaluate the overall function of miRNAs, it is practically impossible to simultaneously knock out/inhibit or overexpress all miRNAs, and the inhibition of all miRNAs by knocking out molecules essential for the synthesis process may be the most practical method.

Dicer is a ribonuclease (RNase) essential for the processing of pre-miRNA to mature miRNA, although there is an exception. Dicer-knockout mice have been prepared and the phenotypes have been analyzed. Fetal death occurs in homozygous whole-body knockout mice,⁷ while heterozygous mice are normal; hence, the function in the heart is analyzed in homozygous heart-muscle-specific knockout mice employing the Cre-loxP system. It is possible that Dicer has a function other than miRNA processing and its loss has an influence, but, at present, the phenotypes are considered to be due to the inhibition of all miRNA functions. The phenotypes in the heart are outlined in the following sections.

NKx2.5 is a heart-specific transcription factor essential for cardiogenesis. As it is expressed in cardiomyocyte precursors at 8.5 days of embryonic age, Dicer can be deleted from the heart in the early embryonic period by expressing Cre recombinase under the promoter of NKx2.5. Fetal death from heart failure occurs at 12.5 days of embryonic age, and edema and hypoplasia of the ventricular muscle are observed.¹⁴

The α-myosin heavy chain (MHC) is a heart-muscle-specific constrictive protein forming two isoforms with β-MHC. Heart-muscle-specific Dicer deletion can be induced in the mid-embryonic period by expressing Cre recombinase under the α-MHC promoter. In the fetal heart at 14.5 days of embryonic age, Dicer is knocked out and the total miRNA expression level is reduced, but fetal death does not occur. A dilated cardiomyopathy-like condition occurs within 4 days after birth, and the animals die from heart failure.¹⁵

Tamoxifen administration induces Cre recombinase expression under the α-MHC promoter, similar to that in the above mice, which enables heart-muscle-specific Dicer deletion at a specific timing. When Dicer deletion is induced in young mice at 3 weeks of age, mild ventricular remodeling and marked atrial dilation develop, and animals start to suddenly die after 1 week. When it is induced in adult mice at 8 weeks of age, marked bilateral ventricular dilatation accompanied by cardiomyocyte enlargement, muscle fiber disarray, and ventricular fibrosis develop and the systolic function decreases.¹⁶

The most important factor for cardiogenesis is the miR-1/133 family specifically expressed at a high level in heart and skeletal muscle cells.

miR-1 is encoded by two genes (miR-1-1 and miR-1-2), and each gene is simultaneously transcribed with one of two miR-133a genes (miR-133a-1 and miR-133a-2). Their expressions are limited to the heart and skeletal muscle, and their transcriptions are regulated by transcription factors SRF and MEF2/MyoD, respectively.¹⁷ In cardiogenesis, miR-1 is considered to control the balance between the differentiation and proliferation of cardiomyocyte precursors through Hand2.¹⁷ When miR-1 is overexpressed under the β-MHC promoter in fetal cardiomyocytes, the ventricle is thinned at 13.5 days of embryonic age and cardiogenesis stops. Half of homozygous miR-1-2-knockout mice die between the late embryonic period and birth because of a ventricular septal defect.⁶ In most of the other half, heart failure develops after birth and sudden cardiac death occurs as a result of arrhythmia without apparent morphological abnormality in the heart.⁶

There are three genes of miR-133 (miR-133a-1/miR-133a-2/133b), and these are bicistronically transcribed with miR-1-2, miR-1, and miR-206, respectively. Expression of the two miR-1/133a clusters is regulated by MEF2 and SRF and limited to heart and skeletal muscles. Expression of the miR-206/133b cluster is limited to skeletal muscle.¹⁸

There are two polycistronically transcribed miR-1/133a clusters: miR-1-1/133a-2 is abundant in the precursor atria and miR-1-2/133a-1 is specifically present in the ventricle. miR-1/133a cluster expression in cardiomyocytes is directly regulated by MEF2 and SRF. There is also the miR-206/133b cluster transcribed from other genetic loci, but its function is skeletal-muscle-specific.

miR-1 targets Hand2 (dHand) to negatively control cardiogenesis, in addition to forming a feedback loop through targeting histone deacetylase (HDAC) 4 to inhibit downstream MEF2. In addition to feedback through targeting SRF, miR-133 inhibits myocardial proliferation by targeting cyclin D2.16, 17 Both miR-1-2-defective mice and miR-133a-1/2 double KO mice have ventricular septal defects and die in the late embryonic or neonatal period.6, 19 In mice overexpressing miR-1 or miR-133a under the β-MHC promoter, myocardial proliferation is inhibited, resulting in embryonic death.19, 20 These findings show that miR-1/133 is essential for division/proliferation of cardiomyocytes and for heart formation in the developmental process.