SEARCH

SEARCH BY CITATION

Keywords:

  • MicroRNA;
  • Mesenchymal stem cells;
  • Differentiation;
  • Cardiomyocytes

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Clinical Implications and Limitation
  8. Conclusions
  9. Acknowledgements
  10. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  11. REFERENCES

Accumulating evidence demonstrated that bone marrow-derived mesenchymal stem cells (BMSCs) may transdifferentiate into cardiomyocytes and replace apoptotic myocardium so as to improve functions of damaged hearts. However, little information is known about molecular mechanisms underlying myogenic conversion of BMSCs. microRNAs as endogenous noncoding small molecules function to inhibit protein translation post-transcriptionally by binding to complementary sequences of targeted mRNAs. Here, we reported that miR-124 was remarkably downregulated during cardiomyocyte differentiation of BMSCs induced by coculture with cardiomyocytes. Forced expression of miR-124 led to a significant downregulation of cardiac-specific markers—ANP, TNT, and α-MHC proteins as well as reduction of cardiac potassium channel currents in cocultured BMSCs. On the contrary, the inhibition of endogenous miR-124 with its antisense oligonucleotide AMO-124 obviously reversed the changes of ANP, TNT, and α-MHC proteins and increased cardiac potassium channel currents. Further study revealed that miR-124 targeted the 3′UTR of STAT3 gene so as to suppress the expression of STAT3 protein but did not affect its mRNA level. STAT3 inhibitors AG490, WP1066, and S3I-201 were shown to attenuate the augmented expression of ANP, TNT, α-MHC, GATA-4 proteins, and mRNAs in cocultured BMSCs with AMO-124 transfection. Moreover, GATA-4 siRNA reduced the expression of ANP, TNT, α-MHC, and GATA-4 proteins but did not impact STAT3 protein in cocultured BMSCs, indicating GATA-4 serves as an effector of STAT3. In summary, we found that miR-124 regulated myogenic differentiation of BMSCs via targeting STAT3 mRNA, which provides new insights into molecular mechanisms of cardiomyogenesis of BMSCs. STEM CELLS2012;30:1746–1755


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Clinical Implications and Limitation
  8. Conclusions
  9. Acknowledgements
  10. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  11. REFERENCES

Stem cell's transplantation has recently been proposed as a promising strategy to repair damaged myocardium in cardiovascular diseases [1–5]. Although embryonic stem cells (ESCs) are able to differentiate into cardiac cells and exhibit strong therapeutic potential for heart diseases [6], the ethical controversies surrounding the origin of ESCs hinder its broad application in patients. Bone marrow-derived mesenchymal stem cells (BMSCs) were previously considered to play only a supportive role for bone marrow hematopoiesis, but recently BMSCs were shown to possess lots of important biological actions, including transdifferentiation, angiogenesis, antiapoptosis, antifibrosis, anti-inflammatory, immunosuppression, and so forth [5, 7–9]. Particularly, the multipotent differentiation potential of BMSCs enables it to turn into a variety of cell types such as cardiomyocytes, endothelial cells, neurons, osteoblasts, and fat cells [8, 9]. It was uncovered that BMSCs were induced to transdifferentiate into cardiomyocyte-like cells after being exposed to 5-azatypine, GSK-3β, GATA-4, or difluoromethylornithine [10–13]. Furthermore, a series of studies confirmed the feasibility and safety of BMSCs-based cell therapy for ischemic, failing and dilated cardiomyopathy via transendocardial, intracoronary, or other delivery routes [3–5, 7–9]. Despite the growing attention paid to BMSCs in cardiac repair [14–16], the detailed molecular mechanisms contributing to cardiomyocyte differentiation of BMSCs remain incompletely understood.

microRNAs (miRNAs) are one kind of endogenous noncoding small RNAs with approximately 22 nucleotides, which function to regulate the expression of proteins by acting on the 3′UTR of target genes encoding these proteins. It is well documented that miRNAs participate in a variety of biological process, including cellular proliferation, apoptosis, differentiation, aging, and so forth [17, 18]. Meanwhile, the increasing evidence also shows that miRNAs are involved in many pathological conditions, such as cancer, cardiac infarction, arrhythmias, virus infection, and Alzheimer's disease [19–21], which has been proposed as a novel target to treat these disorders. Recently, several miRNAs were suggested to associate with cardiomyocyte differentiation of pluripotent stem cells. The serum response factor-dependent muscle-specific miR-1 was reported to modulate cardiomyogenesis and maintain the expression of muscle genes via downregulation of the Notch signaling pathway [22]. In cardiac progenitor cells, the overexpression of miR-1 and miR-499 reduced cellular proliferation rate and enhanced its differentiation into cardiomyocytes via the repression of HDAC4 or Sox6 [23]. Both miR-1 and miR-133 levels were increased in differentiating ESCs, but miR-1 was demonstrated to promote differentiation of ESCs into cardiac lineage. Conversely, miR-133 may block the differentiation of myogenic precursors [24]. Nevertheless, the roles of miRNAs in cardiomyocyte differentiation of BMSCs are not fully clarified yet. This study aimed to uncover the role of miRNAs in myogenic conversion of BMSCs.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Clinical Implications and Limitation
  8. Conclusions
  9. Acknowledgements
  10. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  11. REFERENCES

Animals

Male Sprague-Dawley rats were bought from the Experimental Animal Center of the Affiliated Second Hospital of Harbin Medical University. All the procedures and experimental protocols were approved by the Experimental Animal Committee of Harbin Medical University. All the rats were treated in accordance to the guideline for the Care and Use of Laboratory Animals published by the U.S. National Institute of Health.

BMSCs Culture and Expansion In Vitro

The procedure for BMSCs isolation and culture from immature SD rats (100 ± 10 g) was just as described previously [25]. In brief, after anesthetized, the femurs and tibias of SD rats were quickly taken out, and bone marrow cells were flushed from the bone marrow cavities into the beaker with MesenCult basal medium supplemented with Mesenchymal Stem Cell Stimulatory Supplement (Stem Cell Technologies, Vancouver, BC, Canada). Bone marrow cells were harvested and plated into the dishes and then incubated with the same medium supplemented with penicillin (100 U/ml)/streptomycin (100 μg/ml) (Sigma-Aldrich, St. Louis, MO) in a humidified atmosphere of 5% CO2 at 37°C. The culture medium was changed twice per week. The cells were washed with phosphate buffered saline (PBS), trypsinized using 0.05% Trypsin/EDTA solution (Sigma-Aldrich, St. Louis, MO), and subcultured under the same conditions. Cultured BMSCs between passages 3–5 were used for the following experiments.

Ventricular Myocytes Culture In Vitro

Neonatal rat ventricular myocytes (NRVMs) were isolated and cultured from the hearts of 3-day-old SD rats by a method described previously [25]. Briefly, the hearts were dissociated from newborn rats after anesthetization, and then the ventricular samples were cut into 1–2 mm3 pieces. Cardiac tissues were then trypsinized at 37°C for 1–2 minutes and neutralized with the culture medium after the suspensions were discarded. The tissues were continuously trypsinized to disappear and then cell suspensions were collected. After centrifugation at 2,000 rpm for 180 seconds, the harvested cells were resuspended in Dulbecco's modified Eagle's medium (DMEM) (HyClone Laboratories Inc., Logan, UT) supplemented with 10% fetal bovine serum (Gibco BRL, Grand Island, NY) and penicillin (100 U/ml)/streptomycin (100 μg/ml) (Sigma-Aldrich, St. Louis, MO) and cultured in the flask at 37°C in humid air with 5% CO2. Two hours later, the isolated cells were plated on six-well plate at a density of 5 × 105 cells per well. Three days later, the medium was changed and the cells were cocultured with transfected BMSCs.

Coculture Model of BMSCs and NRVMs In Vitro

A coculture model of BMSCs and NRVMs in vitro was established just as described previously with some modifications [26]. BMSCs and ventricular myocytes were indirectly cocultured at a ratio of 1:10 in the Transwell Chamber with a semipermeable polycarbonate membrane. The culture medium used is DMEM (HyClone Laboratories Inc., Logan, UT) contains 10% fetal bovine serum (Gibco BRL, Grand Island, NY) and penicillin (100 U/ml)/streptomycin (100 μg/ml) (Sigma-Aldrich, St. Louis, MO).

Transfection of miRNAs and siRNA

In vitro gene transfection was performed using miR-124, its antisense oligonucleotide AMO-124, or scramble control mixed with Lipofectamine 2000 (Invitrogen, CA) according to instructions provided by the manufacturer. rno-miR-124 (5′-UAAGGCACGCGGUGAAUGCC-3′), antisense oligonucleotide AMO-124 (5′-GGCATTCACCGCGUGCCUUAUU-3′), and a scrambled RNA (5′-UUCUCCGAACGUGUCACGUTT-3′) (GenePharma, Shanghai, China). The siRNA designed for GATA-4 mRNA (sense: 5′-GCCCAAGAAUCUGAAUAAATT-3′, antisense: UUUAUUCAGAUUCUUGGGCTT) was purchased from GenePharma (Shanghai, China).

Quantitative Real-Time RT-PCR

Total RNA samples were extracted using Trizol method (Invitrogen, CA) according to the manufacturer's instructions. The real-time quantitative PCR analysis was performed with Applied Biosystems 7500 Real-Time PCR Systems (Applied Biosystems, Foster City, CA). The expression level of glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was used as an internal control of mRNAs, and the U6 level was regarded as internal control of miRNAs. Primers used in real-time quantitative PCR analysis: U6 (forward: 5′-GCTTCGGCAGCACATATACTAAAAT-3′, reverse: 5′- CGCTTCACGAATTTGCGTGTCAT-3′); GAPDH (forward: 5′-AAGAAGGTGGTGAAGCAGGC-3′, reverse: 5′-TCCACCACCCTGTTGCTGTA-3′); miR-124 (RT primer: 5′-GTCGTATCCAGTGCGTGTCGTGGAGTCGGCAATTGCACTGGATACGACGGCAT-3′, forward: 5′-GGTAAGGCACGCGGTGAATG-3′, reverse: 5′-GTGCGTGTCGTGGAGTC-3′); atrial natriuretic peptide (ANP) (forward: 5′-TCAGAGAGATGGAGGTGCT-3′, reverse: 5′-CCAATCCTGTCAATCCTAC-3′); troponin T (TNT) (forward: 5′-CGAGGAAGAGGTGGTGGAGTAC-3′ reverse: 5′-ACCATCCTCCTCCTCCTCCACTG-3′); GATA-4 (forward: 5′-CGATATGTTTGATGACTTCTC-3′, reverse: 5′-CGTTGCATAGGTAGTGTCC-3′); α-myosin heavy chain (α-MHC) (forward: 5′-CACCGTGGACTACAACATC-3′, reverse: 5′-AGCCTTTCTTCTTGCCTC-3′).

Immunofluorescence Staining

Cultured BMSCs were fixed in 4% paraformaldehyde for 20 minutes. Then the cell membrane was penetrated by 0.4% Triton X-100 for 1 hour and blocked by goat serum for 1 hour, followed by primary antibody staining at 4°C over night. Primary antibody included ANP (1:100, Santa Cruz Biotechnology Inc, Santa Cruz, CA), TNT (1:100, Santa Cruz Biotechnology Inc, Santa Cruz, CA), GATA-4 (1:200, Abcam, Cambridge, UK), and α-MHC (1:200, Abcam, Cambridge, UK). Afterward, cultured cells were incubated with FITC-conjugated secondary antibody for 1 hour. Secondary antibodies were used to detect unconjugated primary antibodies containing goat anti-rabbit IgG (Alexa Fluor® 488: Molecular Probes, Eugene, OR) for ANP, TNT, and GATA-4 and goat anti-mouse IgG (Alexa Fluor® 594: Molecular Probes, Eugene, OR) for α-MHC. The cells were stained using 4′,6-diamidino-2-phenylindole (DAPI) (Beyotime, China) for 5 minutes, and then cover slides were mounted on glass slides with Cytoseal XYL (Richard-Allan Scientific, Kalamazoo, MI) and observed under Confocal microscope.

Electrophysiological Recordings

The whole-cell patch-clamp techniques were used to record potassium channel currents using an Axopatch 200B amplifier (Axon Instruments, Foster City, CA) in the voltage-clamp mode. The detailed procedure to perform electrophysiological recording is just as decreased previously [25]. In brief, the resistance of glass electrodes was modified between 2 and 4 MΩ after being filled with pipette solution. The voltage command pulses were produced by a 12-bit digital-to-analog converter controlled by pClamp 9.0 software (Axon Instruments, Foster City, CA). After the gigaseal formation, the cellular membrane was disrupted by gentle suction to establish the whole-cell patch-clamp configuration. The standard external solution for potassium channel currents recording contained (in mM) NaCl 136, KCl 5.4, CaCl2 1.8, MgCl2 1.0, NaH2PO4 0.33, HEPES 10, and glucose 10; pH adjusted to 7.4 with NaOH. The pipette solution for potassium channel currents recordings contained (in mM) KCl 20, K-aspartate 110, MgCl2 1.0, HEPES 5, EGTA 10, and Na2ATP 5; pH adjusted to 7.2 with KOH. In order to avoid the intervention from other channels currents, sodium currents were inactivated by adding tetrodotoxin (TTX, 30 μM; Sigma-Aldrich, St. Louis, MO) and calcium currents were omitted by the use of CdCl2 (0.2 mM; Sigma-Aldrich, St. Louis, MO) in extracellular solution.

Construction of Chimeric miRNA-Target Site-Luciferase Reporter Vectors

To construct reporter vectors carrying miRNA-target sites, we synthesized the gene fragments containing the predicted target sites (Invitrogen, CA) for rat STAT3 3′UTR and inserted the fragment of STAT3 3′UTR into the multiple cloning sites in the pMIR-REPORT luciferase miRNA expression reporter vector (Ambion, Inc., Austin, TX), as described previously [27, 28].

Dual Luciferase Reporter Assay

Luciferase reporter transfection was performed in HEK293 cells after starvation using Opti-MEM (Invitrogen, CA). A mixture of pMir-STAT3 (Ambion, Inc, Austin, TX), thymidine kinase (TK) (Promega, Madison, WI), miRNA (GenePharma, Shanghai, China), Lipofectamine 2000 (Invitrogen, CA), and Opti-MEM (Invitrogen, CA) was added to each well of 24-well plate. Cell lysates were harvested at 48 hours after transfection, and the reporter activity was measured with the Dual Luciferase Assay (Promega, Madison, WI) by Luminometer (GloMaxTM 20/20, Promega, Madison, WI). It was normalized by the coexpressed TK Luciferase. All Luciferase assays were repeated three times and the activities were normalized to control luciferase activity.

Data Analysis

Group data are expressed as mean ± SEM. Statistical comparisons among multiple groups were performed by analysis of variance (ANOVA). A two-tailed p < .05 was taken to indicate a statistically significant difference. Statistical values were calculated using the SPSS software and illustrated using the GraphPad Prism 4.0.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Clinical Implications and Limitation
  8. Conclusions
  9. Acknowledgements
  10. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  11. REFERENCES

miRNAs Were Altered During Cardiomyocyte Differentiation of BMSCs

To examine whether miRNAs are involved in myogenic conversion of BMSCs, we established an in vitro coculture model of BMSCs and NRVMs at a ratio of 1:10 which mimics cardiac microenvironment and induces BMSCs to differentiate into cardiomyocytes. After 7-day coculture, immunofluorescence staining was used to detect cardiac-specific makers including ANP, TNT, α-MHC, and GATA-4 proteins in cocultured and uncocultured BMSCs. As illustrated in Figure 1A, cocultured BMSCs displayed the positive staining of ANP, TNT, α-MHC, and GATA-4 proteins compared with uncocultured BMSCs, indicating that the cocultured BMSCs were induced to turn into cardiomyocytes in vitro. The percentage of BMSCs with the positive staining of ANP, TNT, α-MHC, and GATA-4 proteins gradually increases with the increase of coculture time (Fig. 1B).

thumbnail image

Figure 1. miRNAs were altered during cardiomyocyte differentiation of BMSCs. (A): Immunofluorescence detection of cardiac-specific markers ANP, TNT, α-MHC, and GATA-4 proteins in BMSCs cocultured with neonatal rat ventricular myocytes (NRVM) at day 7. (B): The number of BMSCs with ANP, TNT, α-MHC, and GATA-4 positive staining after coculture with NRVMs. (C): miR-124 was altered in cocultured BMSCs compared with uncocultured BMSCs at day 7 by miRNAs array screening. (D): The real-time quantitative RT-PCR verification of the difference of miR-124 between uncocultured BMSCs and cocultured BMSCs. Data are expressed as mean ± SEM from five individual experiments. The scale bar represents 30 μm. *, p < .05 versus uncocultured BMSCs. Abbreviations: ANP, atrial natriuretic peptide; BMSCs, bone marrow-derived mesenchymal stem cells; MHC, myosin heavy chain; TNT, troponin T.

Download figure to PowerPoint

Then, we conducted the miRNA array to screen out dysregulated miRNAs during cardiomyocyte differentiation of BMSCs. Seven miRNAs such as miR-30-5p, miR-449a, miR-125b-5p, and miR-674-5p were found increased by >1.5-fold, and 10 miRNAs containing miR-124, miR-140, miR-143, miR-23a, miR-31, and so forth, were decreased by at least 50%. Among these miR-124 level was shown significantly decreased in cocultured BMSCs compared with uncocultured BMSCs (Fig. 1C). The real-time quantitative RT-PCR analysis confirmed a significant downregulation of miR-124 level in cocultured BMSCs at day 7 (Fig. 1D). These findings prompted us to focus on exploring the role of miR-124 in the differentiation of BMSCs into cardiomyocytes.

Overexpression of MiR-124 Inhibited Cardiomyocyte Differentiation of BMSCs

Based on the above observations, we decided to investigate whether forced expression of miR-124 is able to regulate myogenic conversion of cocultured BMSCs. Figure 2A–2C showed that in vitro gene transfer of miR-124 obviously inhibited the expression of ANP, TNT, and α-MHC during cardiomyocyte differentiation of BMSCs under the condition mimicking cardiac microenvironment. On the contrary, the sequence-specific inhibition of miR-124 with its antisense oligonucleotide (AMO-124) significantly reversed the decreased expression of ANP, TNT, and α-MHC proteins and promoted cardiomyocyte differentiation of BMSCs. The gene deliver of miR-124 together with AMO-124 was able to abolish both the inhibitory role of miR-124 and the facilitated action of AMO-124 in cardiomyogenesis of BMSCs. The scrambled miRNA did not influence the expressions of ANP, TNT, and α-MHC proteins. In consistent with the changes of ANP, TNT, and α-MHC proteins, their mRNA levels were also markedly decreased or increased in cocultured BMSCs transfected with miR-124 or AMO-124, respectively (Fig. 2D). These results suggested that in vitro gene transfer of miR-124 obviously inhibited BMSCs to obtain cardiomyocyte phenotypes, but the loss of endogenous miR-124 was capable to promote cardiomyocyte differentiation of BMSCs after cocultured with NRVMs.

thumbnail image

Figure 2. The regulatory role of miR-124 in the expression of ANP, TNT, and α-MHC in differentiated bone marrow-derived mesenchymal stem cells (BMSCs). (A): The detection of α-MHC expression in cocultured BMSCs transfected with miR-124, AMO-124, and scrambled RNA by immunofluorescence staining at days 3 and 7. (B): The expression of ANP protein was detected by Confocal microscope in cocultured BMSCs transfected with miR-124, AMO-124, and scrambled RNA at days 3 and 7. (C): The influence of miR-124 on the expression of TNT in cocultured BMSCs at days 3 and 7. (D): Quantitative analysis of the effects of miR-124 on the levels of ANP, TNT, and α-MHC mRNAs in cocultured BMSCs at day 7. Data were obtained from five independent experiments. The scale bar represents 50 μm. *, p < .05 versus Control; #, p < .05 versus miR-124; and +, p < .05 versus AMO-124. Abbreviations: ANP, atrial natriuretic peptide; MHC, myosin heavy chain; TNT, troponin T.

Download figure to PowerPoint

Regulatory Roles of MiR-124 in Electrophysiological Properties of BMSCs

In addition to cardiac-specific proteins, we further observed the influence of miR-124 on cardiac potassium channels of BMSCs. As displayed in Figure 3A, miR-124 overexpression led to a considerable reduction of KCNJ1, KCND2, KCND3, and KCNA5 mRNAs in cocultured BMSCs which contributes to form inward rectifier and transient outward potassium currents in cardiac cells. In contrast, these mRNAs levels were obviously reversed in cocultured BMSCs transfected with AMO-124. Figure 3B showed that inward rectifier potassium currents in cocultured BMSCs transfected with miR-124 were smaller than that in cocultured BMSCs with AMO-124 transfection. Likewise, BMSCs with miR-124 overexpression exhibited a considerable reduction of transient outward potassium currents compared with BMSCs transfected with AMO-124 (Fig. 3C). These data suggested that miR-124 not only regulated the expression of cardiac-specific proteins but also affected electrophysiological characteristics of cocultured BMSCs.

thumbnail image

Figure 3. The regulation of cardiac electrophysiological properties by miR-124 in bone marrow-derived mesenchymal stem cells (BMSCs). (A): The real-time quantitative RT-PCR analysis of the effects of miR-124 on KCNJ1, KCND2, KCND3, and KCNA5 mRNA levels in cocultured BMSCs. (B): Inward rectifier potassium currents carried by KCNJ1 gene were recorded in cocultured BMSCs with miR-124, AMO-124, miR-124+AMO-124, and scrambled RNA transfection. (C): The effects of miR-124 on transient outward potassium currents encoded by KCND2, KCND3, and KCNA5 in cocultured BMSCs. The results were collected from three independent experiments. Data are expressed as mean ± SEM. *, p < .05 versus control and #, p < .05 versus AMO-124.

Download figure to PowerPoint

STAT3 Gene as the Target Site of MiR-124

Signal transducer and activator of transcription 3 also known as STAT3 is a transcription factor that is encoded by the STAT3 gene. A large number of studies revealed an essential role of STAT3 in the induction of transdifferentiation. We performed a computational analysis on Targetscan (http://www.targetscan.org/) to predict the possible target gene of miR-124, and then STAT3 was selected because it contains the “seed site” of miR-124 in its 3′UTR (Fig. 4A). To confirm STAT3 as the target gene of miR-124, we engineered the fragment of STAT3 3′UTR containing the binding sites for miR-124 into luciferase reporter vector and performed luciferase reporter activity assay. As illustrated in Figure 4B, transfection of miR-124 caused a substantial reduction of luciferase activity on the luciferase expression constructs carrying the target fragment. The inhibition of luciferase activity by miR-124 may be efficiently reversed by AMO-124. Furthermore, a negative control construct did not affect the luciferase activity. These results indicated STAT3 gene as a target site of miR-124.

thumbnail image

Figure 4. STAT3 as a target gene for miR-124 in differentiated BMSCs. (A): The predicted binding site of miR-124 on the 3′UTR of STAT3 gene. (B): Luciferase assay in HEK293 cells for the post-transcriptional repression of STAT3 by miR-124. HEK293 cells were transfected with miR-124, AMO-124, miR-124+AMO-124, and a scrambled miRNA for negative control with Lipofectamine 2000, and control cells were treated with Lipofectamine 2000 alone. (C): The influences of miR-124 and AMO-124 alone or together on the mRNA level of STAT3 gene during cardiomyogenesis of BMSCs. (D): Immunocytofluorescence staining verification of the post-transcriptional repression of STAT3 by miR-124 in the cytoplasm and the nucleus of cocultured BMSCs. BMSCs without neonatal rat ventricular myocytes (NRVMs) coculture is in the left upper panel of Figure 4D. (E): The expression of p-STAT3 protein determined by Western blotting in cocultured BMSCs. These figures are representative from three individual experiments. The scale bar represents 50 μm. *, p < .05 versus Control; #, p < .05 versus miR-124. Abbreviations: BMSCs, bone marrow-derived mesenchymal stem cells; GAPDH, glyceraldehyde 3-phosphate dehydrogenase.

Download figure to PowerPoint

The real-time quantitative RT-PCR analysis was further used to observe the effect of miR-124 on the mRNA level of STAT3 gene. As shown in Figure 4C, in vitro transfection with miR-124 or AMO-124 alone or together did not obviously alter the expression of STAT3 mRNA in cocultured BMSCs at day 7. Furthermore, the ability of miR-124 to regulate the expression of STAT3 protein in cocultured BMSCs was verified by immunofluorescence staining and Western blotting (Fig. 4D and 4E). Positive staining of STAT3 protein in the cytoplasm and the nucleus was both significantly reduced in cocultured BMSCs with miR-124 overexpression. On the contrary, in vitro AMO-124 transfection was able to enhance the positive staining of STAT3 protein in the cytoplasm and the nucleus of cocultured BMSCs. The phosphorylation of STAT3 protein in cocultured BMSCs was obviously inhibited after miR-124 transfection but upregulated after AMO-124 transfection. These indicate the regulation of cardiomyocyte differentiation of BMSCs by miR-124 via targeting STAT3.

STAT3 Contributed to the Regulation of Cardiomyogenesis of BMSCs by MiR-124

To confirm that STAT3 was involved in the regulation of cardiomyogenesis of BMSCs by miR-124, three STAT3 inhibitors AG490, WP1066, and S3I-201 were used to block the activation of STAT3 in this study. We found that AG490 50 μM markedly attenuated the facilitated role of AMO-124 in cardiomyocyte differentiation of cocultured BMSCs, which demonstrated as the reduced expression of ANP, TNT, and α-MHC in the cytoplasm and GATA-4 in the nucleus of cocultured BMSCs (Fig. 5A). Similarly, pretreatment with WP1066 3.8 μM and S3I-201 100 μM also decreased the expression of ANP, TNT, and α-MHC proteins in the cytoplasm and GATA-4 in the nucleus of cocultured BMSCs transfected with AMO-124. These results further implicate that STAT3 as a target of miR-124 played a central role in myogenic conversion of cocultured BMSCs, and cardiomyogenesis of BMSCs could be effectively abrogated by the blockage of STAT3. In parallel to the data from immunofluorescence staining, real-time quantitative RT-PCR analysis also confirmed that cocultured BMSCs treated by three STAT3 inhibitors exhibited a distinct reduction of ANP, TNT, and α-MHC mRNA levels, compared with cocultured BMSCs without STAT3 inhibitors treatment (Fig. 5B).

thumbnail image

Figure 5. STAT3 inhibitors blocked myogenic conversion of bone marrow-derived mesenchymal stem cells (BMSCs). (A): The effects of AG490, WP1066, and S3I-201 on cardiac differentiation of BMSCs. Green or red indicates positive staining. (B): The effects of AG490, WP1066, and S3I-201 on the mRNA levels of ANP, TNT, and α-MHC in differentiated BMSCs. Data were from three independent experiments. The scale bar represents 50 μm. *, p < .05 versus AMO-124. Abbreviations: ANP, atrial natriuretic peptide; MHC, myosin heavy chain; TNT, troponin T.

Download figure to PowerPoint

GATA-4 Functioned as an Effector of STAT3 During Cardiomyogenesis of BMSCs

STAT3 can activate the transcription of GATA-4 by binding to the promoter of GATA-4 gene. Activated GATA-4 is translocated into nucleus and enhances the expression of ANP, TNT, and α-MHC mRNAs and proteins. We therefore investigated whether STAT3 can regulate the expression of cardiac-specific makers in cocultured BMSCs via mediating GATA-4 expression. As illustrated in Figure 6, the gene-specific inhibition of GATA-4 with siRNA can reduce the expression of ANP, TNT, and α-MHC proteins in cocultured BMSCs but have no influences on the expression of STAT3 in the cytoplasm and the nucleus. Thus, these results suggest that GATA-4 serves as an effector of STAT3 in cardiomyocyte differentiation of BMSCs.

thumbnail image

Figure 6. Silencing of GATA-4 by siRNA inhibited cardiomyocyte differentiation of cocultured bone marrow-derived mesenchymal stem cells (BMSCs). Immunofluorescence staining of ANP, TNT, α-MHC, GATA-4, and STAT3 expressions in cocultured BMSCs with AMO-124 transfection at day 7. Cocultured BMSCs transfected with AMO-124 plus GATA-4 siRNA showed the reduced expression of ANP, TNT, α-MHC, and GATA-4 but not affected STAT3. Nuclei were stained with 4′,6-diamidino-2-phenylindole (DAPI). The scale bar represents 50 μm. Abbreviations: ANP, atrial natriuretic peptide; MHC, myosin heavy chain; TNT, troponin T.

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Clinical Implications and Limitation
  8. Conclusions
  9. Acknowledgements
  10. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  11. REFERENCES

This study for the first time uncovered that miR-124 was able to regulate differentiation of BMSCs into cardiomyocytes via targeting the 3′UTR of STAT3, which in turn affected the expression of ANP, TNT, α-MHC, and GATA-4. These findings offer novel insights into our understanding of molecular mechanisms underlying transformation of BMSCs into cardiac cells.

It has been well documented that BMSCs are capable to turn into cardiomyocytes, endothelial cells, and vascular smooth muscle cells [8–11]. The potential molecular pathways involved in stem cell's transdifferentiation were also introduced in some investigations. It was reported that Jagged1 protein activated Notch signal and enhanced the differentiation of BMSCs into cardiomyocyte in vitro and in vivo [29]. GSK-3β overexpression caused differentiation of BMSCs into cardiomyocytes via inhibiting the expression of β-catenin [11]. Translocation of Nkx2.5 and GATA-4 to the nucleus was also shown to drive BMSCs to get a cardiac phenotype [13].

Lately, miRNAs are also identified to regulate the multilineage differentiation potential of ESCs and other progenitor cells [30–32]. miR-1 and miR-133 played significant roles in the myocardial differentiation of mouse ESCs, and Cdk9 may be involved in this process as a target of miR-1 [30]. miRNA-145 regulated OCT4, SOX2, and KLF4 and in turn repressed expression of pluripotency genes, and induced lineage-restricted differentiation of human ESCs [31]. However, the information about the roles of miRNAs in myogenic conversion of BMSCs remains unclear. This study aimed to investigate whether miRNAs regulated cardiomyocyte differentiation of BMSCs and explore its potential mechanism.

Cardiomyogenesis of BMSCs in vitro was induced by coculture of BMSCs with NRVMs at a ratio of 1:10, which was reported and confirmed in a series of studies [25, 26]. In this study, we found that miR-124 was strongly reduced in cardiac differentiation of BMSCs caused by coculture with NRVMs. Several studies have uncovered that miR-124 in the adult brain positively modulates the transitory progression of adult subventricular zone neurogenesis through the repression of Sox9 [33–35]. Likewise, the interaction of Sox9 and miR-124 also regulated the neurogenesis of embryonic neuroepithelial cells in the spinal cord [36]. Accordingly, the increase of miR-124 level is considered as an important trigger of the transition from proliferation to neural differentiation [37]. Interestingly, we found for the first time that BMSCs transfected with miR-124 displayed the reduced expression of cardiac-specific markers and potassium channel currents. Conversely, the depletion of miR-124 with AMO-124 supported myogenic conversion of BMSCs. These results implicated that miR-124 was likely to serve as a switch for differentiation of BMSCs into cardiac and neural lineage.

STAT3 plays a crucial role in self-renew, transdifferentiation, and paracrine actions of BMSCs [13, 38–40]. Activation of STAT protein could increase the survival and differentiation of transplanted BMSCs and resulted in better functional recovery of infarcted myocardium [38]. Inhibition of STAT3 activation in engrafted BMSCs exerted deleterious effects on BMSCs transplantation-mediated myocardial recovery after ischemic/reperfusion injury in an isolated rat heart model [39]. In contrast, STAT inhibitor WP1066 abrogated BMSCs-mediated growth factor production and functional improvement [40]. All these suggested STAT3 as a key molecular regulator for mesenchymal stem cell proliferation and differentiation [41].

In this study, STAT3 was established as the target gene of miR-124 by a computer prediction and luciferase activity assay. BMSCs transfected with miR-124 demonstrated a significant reduction of STAT3 expression, compared with untransfected BMSCs. This suggests that miR-124 surely acts on the 3′UTR region of STAT3 gene. Consistently, previous studies also reported that the forced overexpression of miR-124 might lead to a considerable reduction of STAT3 level in mouse ESCs and induced the inhibition of astrocytic lineage differentiation [42, 43]. Our data revealed the 3′UTR of STAT3 mRNA as the binding site of miR-124 in BMSCs. Furthermore, application of STAT3 inhibitors could block the potential of BMSCs to acquire a cardiomyogenic phenotype in cocultured BMSCs. In accordance, it was previously displayed that the inhibition of STAT3 activation by specific inhibitors drastically suppressed the differentiation of mouse ESCs into cardiomyocytes induced by coapplication of leukemia inhibitory factor and bone morphogenic protein-2 [44, 45]. Similarly, the activation of STAT3 is also required for differentiation of P19CL6 cells into beating cardiomyocytes via direct transcriptional regulation of GATA-4, Nkx2.5, and Tbx5 [46, 47]. The involvement of activated STAT3 in both differentiation and paracrine action of BMSCs seem contradictory. But, it should be noted that the upstream and/or downstream pathway of STAT3 in two kind of biological behaviors of BMSCs are different. In this study, we found that miR-124 regulated the activation of STAT3 and in turn affected myogenic differentiation of BMSCs via activating GATA-4. During the release of soluble protective factors by BMSCs, STAT3 was activated by p38, tumor necrosis factor (TNF), or other factor and then targeted at the promoter of vascular endothelial growth factor (VEGF), ANP, or other genes [7, 40]. Thus, although STAT3 was activated in both two kinds of behaviors of BMSCs, their signaling pathways are different. In our studies, siRNA-mediated inhibition of GATA-4 obviously blocked the cardiomyocyte differentiation of BMSCs promoted by AMO-124, demonstrated as the decreased expression of ANP, TNT, and α-MHC, but did not affect the upregulation of STAT3, indicating that GATA-4 is severing as an effector of STAT-3 in the regulation of BMSCs differentiation.

Clinical Implications and Limitation

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Clinical Implications and Limitation
  8. Conclusions
  9. Acknowledgements
  10. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  11. REFERENCES

Although a larger number of evidence has confirmed that BMSCs transplantation was capable to regenerate the damaged cardiac cells, only a small part of transplanted BMSCs was able to form functional cardiomyocytes both in vivo and in vitro [48]. Reasonably, transplantation of BMSCs engineered by miRNAs will be more efficacious treatment strategy for injured myocardium in heart infarction or failure [49]. In this study, we did not observe the contractility of differentiated BMSCs at day 7 or even day 13, which may be explained as follows: (a) It is well known that the pacemaker cells with autodepolarization property produce the electrical pacing for atrial and ventricular myocytes in the whole hearts. Without the stimulation from pacemaker cells, differentiated BMSCs may not exhibit beating function because they have no autodepolarization currents; (b) The mechanical stretch and electrical stimuli are two important factors that play a crucial role in the development of contractile functions of cardiomyocytes. The lack of mechanical and electrical stimuli is also an important reason for cardiac cells without beating appearance in this study; (c) Approximately, 2 weeks of coculture might be not long enough for them to develop into beating cells. In addition, the connexin43 expression between BMSCs and NRVMs was not detected, because BMSCs and NRVMs were indirectly cocultured in a Transwell chamber and could not touch each other. Furthermore, whether other signal pathways also did not contribute to the regulation of cardiomyogenesis of BMSCs by miR-124 needs to be further investigated in the future.

Conclusions

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Clinical Implications and Limitation
  8. Conclusions
  9. Acknowledgements
  10. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  11. REFERENCES

In summary, for the first time, miR-124 was shown downregulated in myogenic conversion of BMSCs, which regulated the expression of STAT3 by targeting at 3′UTR of STAT3 and in turn affect the expression of cardiac-specific markers including ANP, TNT, α-MHC, and GATA-4. These findings will greatly improve our understanding of the roles of miRNAs in cardiomyocyte differentiation of stem cells and improve the effects of BMSCs-based therapy for damaged myocardial repair and regeneration.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Clinical Implications and Limitation
  8. Conclusions
  9. Acknowledgements
  10. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  11. REFERENCES

This work was supported by the Funds for Creative Research Groups of The National Natural Science Foundation of China (81121003), the Major Program of National Natural Science Foundation of China (81130088), and the National Natural Science Fund of China (30900601/81170096).

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Clinical Implications and Limitation
  8. Conclusions
  9. Acknowledgements
  10. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  11. REFERENCES