miRNAs in Extracellular Vesicles from iPS-Derived Cardiac Progenitor Cells Effectively Reduce Fibrosis and Promote Angiogenesis in Infarcted Heart

Cardiac stem cell therapy offers the potential to ameliorate postinfarction remodeling and development of heart failure but requires optimization of cell-based approaches. Cardiac progenitor cells (CPCs) induction by ISX-9, a small molecule possessing antioxidant, prosurvival, and regenerative properties, represents an attractive potential approach for cell-based cardiac regenerative therapy. Here, we report that extracellular vesicles (EV) secreted by ISX-9-induced CPCs (EV-CPCISX-9) faithfully recapitulate the beneficial effects of their parent CPCs with regard to postinfarction remodeling. These EV contain a distinct repertoire of biologically active miRNAs that promoted angiogenesis and proliferation of cardiomyocytes while ameliorating fibrosis in the infarcted heart. Amongst the highly enriched miRNAs, miR-373 was strongly antifibrotic, targeting 2 key fibrogenic genes, GDF-11 and ROCK-2. miR-373 mimic itself was highly efficacious in preventing scar formation in the infarcted myocardium. Together, these novel findings have important implications with regard to prevention of postinfarction remodeling.


Introduction
Myocardial infarction (MI) and subsequent heart failure are a leading cause of death worldwide [1]. Despite advances in medical and device therapies, heart failure continues to be associated with a 5-year mortality of~50%. Stem cell therapy thus offers a great potential for cardiac tissue repair and regeneration, which might ultimately improve symptoms and longevity [2].
Notably, the beneficial effects of cardiac stem cell therapy are largely attributed to a paracrine mechanism of action that involves the release of cellular factors from the transplanted stem cells [3][4][5]. More recent studies show that these factors are packed into small membrane-bound vesicles known as extracellular vesicles (EV, 30-200 nm), which can invoke a multitude of signals [6,7]. The EV contents vary amongst stem cells. Cardiac progenitor cells (CPCs) are of particular interest due to their inherent properties of cell protection, cell development, differentiation, and desirable effects imparted into the host tissue [8][9][10]. EV from newborns improved ventricular remodeling post-MI significantly more than those derived from aging patients [11]. Similarly, EV secreted from young cardiosphere-derived cells exerted stronger antisenescent effects than those derived from aged animals [12].
Recent studies demonstrated that effects of CPCs on cardiac repair and regeneration can be faithfully recapitulated by their EV [6,13]. Multiple miRNAs in EV act as mediators of cell-cell communication within the cardiovascular system [2] and can be transferred into recipient cells to regulate gene expression, thus leading to cardioprotection [11,13,14]. We reported that a small molecule, ISX-9, could render CPCs (CPC ISX-9 ) highly resistant to oxidative stress, thus permitting better survival and engraftment in the infarcted myocardium [15]. Development of CPC ISX-9 may represent a significant advance in the cardiac stem cell field, as ISX-9 treatment circumvents the need to genetically reprogram the cells in order to enhance their function. Since CPC ISX-9 are well positioned for therapeutic application in humans, characterizing EV secreted from these cells is not only important to provide insight into their mechanisms of action, but may also help to identify novel miRNAs involved in cardioprotection.
Since the EV cargo contents are unique to each cell type, consequently, their effectiveness is variable. Considering this limitation, we have generated multipotent CPCs from human-induced pluripotent stem cells (hiPSCs) using a unique small molecule with antioxidant and regenerative properties capable of successfully propagating in the infarcted myocardium. Since CPCs are the cells of choice for regeneration, their EV would be considered to be more effective in cardiac repair than EV from non-CPC. Therefore, the purpose of the study was to exploit EV from hiPSC-CPC induced with ISX-9 and not the role of ISX-9 per se on EV release from CPC. Here, we tested the hypothesis that EV secreted by ISX-9-induced CPCs (EV-CPC ISX-9 ) will be highly efficacious in cardiac repair owing to the unique properties of their parent cells. EV-CPC ISX-9 exerted strong effects on fibrosis and angiogenesis in the infarcted myocardium of mice. Mechanistically, we identified miR-373 enriched in EV-CPC ISX-9 , which elicited strong antifibrotic effects by targeting two genes, growth differentiation factor 11  and Rho-associated coiled-coil containing kinase-2 (ROCK-2), and showed that the miR-373 mimic effectively inhibits postinfarct cardiac remodeling.

Materials and Methods
2.1. Cell Culture. Human iPSC cell line (ACS-1021, ATCC, USA) was maintained in mTeSR1 media (Stem Cell Technology) on vitronectin-coated six-well plates with daily medium changes. Cells were passaged with ReLeSR™ reagent every 4-7 days according to the manufacturer's protocol (Stem Cell Technology). For CPC generation, briefly, hiPSCs maintained on vitronectin-coated six-well plates in mTeSR1 media (Stem Cell Technology) were dissociated into single cells using Accutase (Invitrogen) at 37°C for 10 min and then were seeded on to vitronectin-coated six-well plates at 1 × 10 6 cells/well in mTeSR1 supplemented with 5 μM ROCK inhibitor (Y-27632, Stem Cell Technology) for 24 h. The following day, cells were cultured in mTesR1 with daily medium change for 3 days. Afterwards, the medium was switched to RPMI/B27 minus insulin supplemented with ISX-9 (20 μM, dissolved in DMSO, Stem Cell Technology) for 7 days. Embryoid bodies (EBs) were generated using the hanging drop method in RPMI/B27 minus insulin medium. Human dermal fibroblast cell line (CC-2511) and lung fibroblast cell line (CC-2512) were obtained from the Lonza Company. Briefly, fibroblasts were maintained in FibroGRO™ Complete Media (Millipore Sigma). Cells were passaged with Accutase; passages 2-4 were used for experiments. Commercial human CPCs (control-CPC) derived from human iPS cells (catalog: R1093, Cellular Dynamics International) were maintained in serum-free William's E Medium supplemented with Cocktail B (CM400, Life Technologies). Passage 2 was used for experiments.

Isolation of EV.
Human iPSC cell line ACS-1021 (ATCC, USA) and CPCs induced by ISX-9 were cultured as described [15]. In some cases, EBs and commercial human CPCs were also cultured. Conditioned media was collected and EVs were isolated by centrifugation at 3000 rpm for 30 min to remove cells and debris, followed by filtration through a 0.22 μm filter to remove the remaining debris. Then, the medium was further concentrated to 500 μl using Amicon Ultra-15 100 kDa centrifugal filter units (Millipore). Isolation of EV in the concentrated medium was carried out through qEV size exclusion columns (Izon Science). EV fractions were collected and concentrated by Amicon Ultra-4 10 kDa centrifugal filter units to a final volume of <100 μl. The purified EVs were stored at -80°C and subsequently characterized by particle size, EV markers, and electron microscopy.

Particle Size and Concentration Distribution
Measurement with Tunable Resistive Pulse Sensing. Particle size and concentration distribution were performed using tunable resistive pulse sensing (TRPS) technique with a qNano instrument (Izon Science). Briefly, the number of particles were counted (at least 600 to 1000 events) using 20 mbar pressure and NP200 nanopore membranes stretched between 46.5 and 47.5 mm. Calibration was performed using known concentration of CPC200 beads (diameter: 210 nm). Data were processed using Izon Control Suite software.

Transmission Electron
Microscopy. EV pellets were fixed with 4% paraformaldehyde (PFA). Following a total of 8 washes using PBS, grids were contrasted with a uranyloxalate solution for 5 minutes and transferred to methylcellulose-uranyl acetate for 10 minutes on ice as previously described [16]. Samples were examined on a JEOL JEM-1220 transmission electron microscope (TEM) (JEOL USA, Inc.).

EV Uptake by Fibroblasts.
To track EV uptake by cultured fibroblasts, purified EV were labeled with PKH26, a red membrane dye (Sigma-Aldrich), according to the manufacturer's protocol. Briefly, 300 μl of EV was suspended into 100 μl of Diluent C, which was mixed with 1.4 μl of PKH26 dye. The labeling reaction was stopped by adding an equal volume of EV-free FBS. Exosome Spin Columns (Cat. 4484449, Thermo Fisher Scientific) were used to remove unincorporated PKH26. The cultured fibroblasts in the slide chamber were incubated with labeled EV at 37°C for 24 h. After incubation, cells were stained with Calcein AM (5 μM). Cells were fixed with 2% formaldehyde for 5 min and mounted with DAPI containing ProLong Gold Antifade medium (Thermo Fisher Scientific). Images were taken with an FV1000 confocal microscope (Olympus, Japan).
2.7. miRNA Array Analysis. The NanoString nCounter Human v3 miRNA Expression Assay was used to perform the microRNA profiling analysis. The assay allows measurement of 800 different microRNAs at the same time for each sample. 3.5 μl of suspension RNA was annealed with multiplexed DNA tags (miR-tag) and bridges target specifics. Mature microRNAs were bonded to specific miR-tags using a ligase enzyme, and excess tags were removed by enzyme clean-up step. The tagged microRNA product was diluted 1 to 5, and 5 μl was combined with 20 μl of reporter probes in a hybridization buffer and 5 μl of capture probes overnight (17 hours) at 65°C to permit hybridization of probes with specific target sequences. Excess probes were removed using two-step magnetic bead-based purification on an automated fluidic handling system (nCounter Prep Station), and target/probe complexes were immobilized on the cartridge for data collection. The nCounter Digital Analyzer took images of immobilized fluorescent reporters in the sample cartridge with a CCD camera through a microscope objective lens. For each cartridge, a high-density scan encompassing 325 fields of view was performed. NanoString raw data was analyzed with nSolver™ software, provided by NanoString Technologies. The mean plus 2 times the standard deviation of negative control probes was used to perform background subtraction; positives were used to perform technical normalization to adjust lane by lane variability due to differences in hybridization, purification, or binding. Data was then normalized by calculating the geometric mean of the spikes present in each sample, as recommended by NanoString. One-way ANOVA was used to calculate the P value; targets with P < 0:05 were selected.

miRNA Target Gene Prediction, Gene Ontology (GO)
Analysis, and Luciferase Activity Assay. miRNA target gene prediction and gene ontology analysis were carried out using DIANA MR-microT and mirPath software. Differential miRNA target genes in significant GO and pathway categories, obtained from GO and pathway analyses, were analyzed with mirPath v.3 software. GO biological process includes biological processes, molecular function, and cellular component of upregulated and downregulated genes.

Myocardial Infarction Model.
Animal experiments were carried out both at the University of Illinois at Chicago and Augusta University according to experimental protocols approved by the University of Illinois at Chicago and Augusta University Animal Care and Use Committee, and the methods were performed in accordance with the guide for the Care and Use of Laboratory Animals by the Institute of Animal Resources. MI model was generated as previously described [15]. Briefly, MI was induced in 8-9-week-old NOD/SCID mice (The Jackson Laboratory) or C57/B6 mice which were anaesthetized with 2% isoflurane (isoflurane USP, Henry Schein), intubated, and ventilated. The left anterior descending (LAD) coronary artery was permanently ligated with a Prolene #8-0 suture. 10 mins after LAD ligation, EV (1 * 10 12 /ml) from hiPSC or CPC ISX-9 were injected into the myocardium along the border zone with a total of 20 μl. The same volume of PBS was injected in the control group. miR-373 mimic in vivo transfection was performed in C57/B6 mice. in vivo-jetPEI™ system (Polyplus-transfection SA) was used for intracardiac miRNA delivery. 200 pmoles of miR-373 mimic complexed with in vivo-jetPEI at a N/P ratio of 7 in a volume of 20 μl was injected into the myocardium along the border zone approximately 10 mins after LAD ligation.
2.11. Histology. Histological analysis was performed in randomly selected hearts from mice subjected to MI and treatment groups (PBS, EV-hiPSC, or EV-CPC ISX-9 (n = 3 per group)). Mice were sacrificed after 1 month of treatment with EV. For immunostaining, hearts were fixed with 4% PFA for 1 hour at room temperature and replaced by 30% sucrose overnight at 4°C. Afterwards, hearts were cryopreserved in an optical cutting temperature (OCT) compound (Tissue-Tek) at -80°C. Hearts were sliced into 5 μm thick frozen sections and incubated with primary antibodies including α-sarcomeric actinin (A7811, Sigma, 1 : 200), Ki67 (ab16667, Abcam, 1 : 500), cTnT (13-11, Thermo fisher Scientific, 1 : 300), and SMA (ab5694, Abcam, 1 : 300). Signals were visualized with Alexa Fluor 647 and Alexa Fluor 488 secondary antibodies (Life Technologies). Images were recorded on a confocal microscope (FV1000, Olympus, Japan). For fibrosis analysis, hearts were embedded in paraffin and cut at 5 μm thick sections. Masson's trichrome staining was performed according to the manufacturer's protocol (HT-15, Sigma). The size of the LV area and the scar area was measured using the ImageJ software. 4 sections (EV-treated mice) and 6 sections (miR-373 mimic-treated mice) were analyzed per heart. The fibrosis area was determined as the ratio of the scar area to the LV area and expressed as a percentage. Vessel density was assessed in 9 animals (3 in each group) in NOD/SCID mice and 6 animals in C57/B6 mice (3 in each group) which were sacrificed at 1 M after MI. The number of vessels was blindly counted on 27 sections (3 sections per heart) in NOD/SCID mice or 18 sections (3 sections per heart) in C57/B6 mice in the infarct and border areas of all mice after staining with an antibody α-SMA using a fluorescence microscope at a 400x magnification. Vascular density was determined by counting α-SMA-positive vascular structures. The number of vessels in each section was averaged and expressed as the number of vessels per field (0.2 mm 2 ).
2.13. RNA Extraction and Real-Time PCR. Total RNA from EV was isolated using miRNeasy Micro Kit (QIAGEN). Reverse transcription was performed using a miScript II RT Kit (QIAGEN). Quantification of mRNA and selected miRNAs was performed by the real-time system Quant-Studio 3 (ABI) using a miScript SYBR Green PCR Kit (QIAGEN). miRNA primer sequences are shown in Table S1, and mRNA primer sequences are shown in Table S2. Expression levels of selected miRNAs were quantified and validated with RT-PCR, and values are expressed as 2 -ΔΔCT with respect to the expression of the reference U6. The primer of U6 was provided in the PCR kit.
2.14. Statistical Analysis. Data are expressed as mean ± SD. Test for normality of data was performed. Statistical analysis of differences was compared by ANOVA with Bonferroni's correction for multiple comparisons. Comparisons between two groups were evaluated with Student's t-test. A probability value of P < 0:05 was considered statistically significant. Statistical analyses were performed using GraphPad Prism 6.0 (Chicago, USA).
3.2. EV-CPC ISX-9 Exhibit a Unique miRNA Profile. Next, we performed a miRNA array to determine whether miRNA cargo content of EV-CPC ISX-9 differs from that of hiPSCs, EBs, and commercial CPCs (Figure 3(a)). Global miRNA profiling showed that EV-CPC ISX-9 had a unique miRNA expression signature very different from that of the other derived EVs. miR-520/-373 family members, including miR-371, miR-302, miR-372, miR-373, and miR-520, as well as miR-512, miR-548, and miR-367, were significantly upregulated in EV-CPC ISX-9 compared to EV from other parent cells (Figure 3(b)) (GSE126347). Furthermore, the expression of these enriched miRNAs (miR-373/miR-548/miR-367/miR-520) was validated with real-time PCR (Figure 3(d)). The target genes of the differentially expressed miRNAs control a broad range of biological functions. The biological process of Gene Ontology (GO) enrichment analysis based on enriched miRNA-targeted genes demonstrated that some target genes were significantly enriched in responses to stress and cell cycle (Figure 3(c)).     . We hypothesized that the enriched miR-373 EVs from CPC ISX-9 exert antifibrotic effects. First, using PKH26 labeling, we confirmed that EV were internalized by fibroblasts and localized in the perinuclear region ( Figure S1). The miR-373 expression level was markedly higher in EV than in their donor cells, CPC ISX-9 ( Figure S2). Inhibition of miR-373 in CPC ISX-9 resulted in decreased miR-373 expression in EV-CPC ISX-9 ( Figure S3A-B), and reduced miR-373 expression in fibroblasts incubated with these EV compared to those from control cells ( Figure S3C). Stimulation of fibroblasts with TGF-β led to significant upregulation of fibrotic genes (MMP-2, TIMP-2, TIMP-1, FN1, CTGF, and MMP-9). Upon pretreatment of fibroblasts with EV from control CPC ISX-9 , upregulation of these fibrotic genes by TGF-β was inhibited. However, inhibition of miR-373 in CPC ISX-9 abrogated the capacity of the EV to inhibit fibrotic gene expression. Conversely, pretreatment of fibroblasts with miR-373 mimic inhibited TGF-β-induced expression of fibrotic genes (Figure 1(a)). We also performed experiments in a second cellular model of fibrosis by exposing the fibroblasts to hypoxia. Figure 1(b) shows that with 72 h exposure in a hypoxic environment, both lung and dermal fibroblasts differentiated into myofibroblasts expressing α-SMA. As expected, the miR-373 mimic significantly inhibited fibroblast transdifferentiation into myofibroblasts under hypoxia. Similarly, fibroblasts failed to transdifferentiate into myofibroblasts when they were pretreated with EV from CPC ISX-9 . Taken together, these results suggest that the miR-373 contained in EV-CPC ISX-9 suppresses fibrosis both in in vitro and in vivo levels.
Moreover, under hypoxic conditions, the expression of GDF-11 and ROCK-2 was increased in lung fibroblasts (Figure 1(f)), while pretreatment with the miR-373 mimic or EV-CPC ISX-9 significantly inhibited expression of these two genes, supporting the contention that miR-373 targets GDF-11 and ROCK-2 (Figure 1(g)).

Discussion
Stem cell-based therapy has been well recognized to improve cardiac function following MI. While this therapy has merits, it also suffers from several limitations, particularly lack of suitable stem cell type and their insufficient engraftment and growth, ranging from no new cell formation to sparse newly formed cells in the infarcted tissue [18][19][20]. Cellular therapy has been propelled by the invention of iPS cells, which have the ability to transform into different progenitor cell types. The cardiac progenitors derived from iPS cells and their counter parts have been used both in animal models of MI [21,22] and in humans [23] with promising results. While the underlying mechanisms of beneficial effects of stem cell therapy remain a point of debate, increasing evidence suggests that paracrine factors play a key role by reducing cell death and stimulating cell migration and proliferation [24,25]. This paracrine signaling involves the secretion of small vesicles or EV harboring multiple miRNAs, proteins, and other factors that mediate protection in the heart. Secreted extracellular vesicles (EV) are packed with potent prorepair proteins and RNA cargo that are both cell type-specific and differentially produced and secreted according to the cellular environment. Additionally, miRNA profiles of EV might be distinct from cellular miRNA patterns [26].
In this study, EV derived from CPC ISX-9 were found to be highly cardioprotective, and the effect can in part be attributed to their specific miRNA content. CPC ISX-9 -derived EV were highly enriched with miR-520/-373 family members including miR-371, miR-372, miR-373, and miR-520, as well as miR-512, miR-548, and miR-367, compared to EV derived from other parent cells. miR-373, which was particularly highly enriched in EV-CPC ISX-9 , was first identified as a human embryonic stem cell-(ESC-) specific miRNA, implicated in the regulation of cell proliferation, apoptosis, senescence, migration, and invasion, as well as DNA damage repair following hypoxia stress [27].
Little has been published regarding the putative role of miR-373 in regulating cardiac pathology or function. In a mouse model of type 1 diabetic cardiomyopathy, miR-373 was found to be significantly downregulated, and application of a miR-373 mimic to neonatal cardiomyocytes exposed to elevated glucose in vitro suppressed cell hypertrophy [28]. Fibrosis is also an important pathological feature of diabetic cardiomyopathy, but effects of miR-373 on fibrosis were not investigated in that study. Fibrosis plays a prominent role in ventricular remodeling and ultimately in the pathogenesis of heart failure after MI. A previous study reported that miR-373 targeted the members of TGF-β signaling including TGF-β receptor 2 and Smad2 and promoted mesoderm differentiation in human embryonic stem cells [17]. miR-373-3p expression was low in the hypertrophic myocardium with diffuse myocardial fibrosis [29], suggesting that miR-373 may function as an antifibrotic miRNA. Thus, we hypothesized that because of their enrichment in miR-373, EV-CPC ISX-9 might produce strong antifibrotic effects to modulate cardiac remodeling.
Our results indicate that both EV-CPC ISX-9 and the miR-373 mimic inhibited TGF-β-and hypoxia-induced fibrotic gene expression in vitro. With inhibition of miR-373 in EV-CPC ISX-9 , or treatment with miR-373 inhibitor, the effects on fibrotic gene expression were abrogated. The luciferase activity assay confirmed that miR-373 targeted GDF-11 and ROCK-2, both known to be involved in fibrosis. An iso-form of Rho-associated coiled-coil forming protein kinase 2, ROCK-2 is reportedly a critical mediator of organ fibrosis. Inhibition of ROCK-2 protected ROCK-2-haploinsufficient mice from bleomycin-induced myofibroblast differentiation and pulmonary fibrosis [30], while its activation was implicated in the development of idiopathic pulmonary fibrosis [31]. Additionally, fibroblast-specific ROCK-2 was reported to promote cardiac hypertrophy, fibrosis, and diastolic dysfunction due to upregulation of the profibrotic gene (CTGF) and promyofibroblast differentiation (α-SMA) genes [32]. Mutant mice with elevated fibroblast ROCK activity exhibited enhanced Ang II-stimulated cardiac hypertrophy and fibrosis [32]. The role of the second identified target gene, GDF-11, is more controversial. It was reported to beneficially reverse age-related cardiac hypertrophy and skeletal muscle dysfunction [33,34], while other reports suggest that it promotes cardiac and skeletal muscle dysfunction and wasting [35], inhibits skeletal muscle regeneration [36], exerts profibrotic effects [37], and promotes renal failure and interstitial fibrosis [38]. Therefore, our data suggest that miR-373 inhibited profibrotic gene upregulation and myofibroblast differentiation in fibroblasts by targeting GDF-11 and ROCK-2.
In vivo data showed that compared with EV-hiPSC and PBS, EV-CPC ISX-9 treatment reduced fibrosis and improved cardiac function, thus supporting a therapeutic role for EV-CPC ISX-9 in cardiac remodeling. Given that EV-CPC ISX-9 were found to be highly enriched in miR-373, we tested the effects of EV-CPC ISX-9 injection in the heart and miR-373 mimic treatment in vivo on the miR-373 expression level and found that the miR-373 expression level in the heart was increased ( Figure S10) and it significantly decreased fibrosis and improved cardiac function post-MI. Moreover, the miR-373 mimic also promoted angiogenesis, which was likely mediated by its ability to activate HIF downstream signaling [39]. These findings suggest that the antifibrotic effects of EV-CPC ISX-9 are, at least in part, mediated by miR-373, and they also support the notion that the miR-373 mimic might represent a novel therapeutic strategy for controlling fibrosis and cardiac remodeling post infarction and perhaps in other disorders, such as diabetic cardiomyopathy.
The second major effect of EV-CPC ISX-9 was on cardiomyocyte proliferation in the infarcted myocardium. A previous study reported that miR-294 (miR-290 cluster), the mouse homolog of human miR-371/372/373 cluster, had a strong effect on cardiac progenitor cell proliferation [40] and that its overexpression led to differentiation towards the mesendoderm lineage [17]. It should be borne in mind, however, that these effects could also be attributed to other miRNAs present in the EV, including miR-302, miR-548, miR-512, and miR-367. Further studies are required to dissect the role of individual EV-CPC ISX-9 miRNAs in regulating cardiac fibrosis, cardiomyocyte proliferation, and other pathological events in the context of postinfarction remodeling.

Conclusion
Several clinical and investigational reports have demonstrated the therapeutic applications of cardiac progenitor cells for the treatment of the ischemic heart. Consequently, these studies led to advance new cell-free (EV) strategies to overcome the limitations of cell-based approaches with the same effectiveness and outcomes. The intracoronary administration of EV eliminates the need for open heart surgery for intramyocardial administration of stem cells. However, the promise of EV does not establish the fact whether their effect is continuous and permanent or if future efforts should continue on strategies directed towards successful engraftment and survival of iPSCderived cardiac progenitors as a source for new myofiber growth and EV for paracrine effects as well.
In summary (Figure 7), we report that EV-CPC ISX-9 exhibit a unique miRNA profile and that miR-373 is particularly highly enriched in these EV. EV-CPC ISX-9 elicit strong antifibrotic effects, which is attributed at least in part to their enrichment in miR-373. Treatment with EV-CPC ISX-9 in vivo reduced postinfarction fibrosis and remodeling, promoted cardiomyocyte proliferation and angiogenesis, and improved cardiac function, findings which were at least in part recapitulated by direct  application of miR-373 mimic. These findings have important implications for understanding the paracrine mechanisms of stem cell function and advancing the field of cardiac stem cell therapeutics.

Data Availability
The raw data of miRNA array is deposited in the GEO database (GSE126347). Other data are available from the authors upon reasonable request.