Recent evidence has shown that cardiomyocytes (CMs) can proliferate at a low level after myocardial infarction (MI), but it is insufficient to reestablish heart function. Several microRNAs (miRNAs) have been proven to sufficiently induce rodent CM proliferation. However, whether miRNAs identified in rodents can promote human CM proliferation is unknown due to the poorly conserved functions of miRNAs among species. In the present study, we demonstrate that i) expression of microRNA-302d (miR-302d) decreased significantly during CM differentiation from human pluripotent stem cells (hPSCs) from day 4 to day 18; ii) miR-302d efficiently promoted proliferation of hPSC-derived CMs; iii) miR-302d promoted CM proliferation by targeting LATS2 in the Hippo pathway; and iv) RNA-sequencing analysis revealed that overexpression of miR-302d induced changes in gene expression, which mainly converged on the cell cycle. Our study provides further evidence for the therapeutic potential of miR-302d.
Despite decades of research and drug development, cardiovascular diseases (CVD) remain the top cause of death worldwide . Injury to the heart caused by CVD usually results in a loss of cardiomyocytes (CMs). Adult CMs are generally believed to be a terminally differentiated cell type, with only a negligible capacity of regeneration and proliferation . Several methods have been developed to promote cardiac regeneration . Primarily, cell-based therapies have attracted much attention. These therapies include treatments with skeletal myoblasts , mesenchymal stem cells , cardiac-derived cells  and pluripotent stem cells . Alternatively, secretory factors released by stem cells have been considered potential treatments to promote cardiac repair. For instance, the activation of growth factor neuregulin 1 (NRG1) successfully prevented cardiac remodeling in a clinical trial , and exosomes carrying miR-451 prevented CM apoptosis in a mouse myocardial ischemia/reperfusion model . Direct reprogramming of fibroblasts into CMs by forced expression of Gata4, Mef2c and Tbx5 (GMT) has proven to be effective in inducing cardiac repair [10,11]. Nevertheless, the cocktail of transcription factors might cause incomplete reprogramming in vivo, which could lead to unfavorable consequences .
Recent evidence has shown that division of pre-existing CMs is the main source of new CMs in mice during normal aging, and the new CMs are necessary to maintain myocardial homeostasis . CMs could also reenter the cell cycle and proliferate at a low level in pathological conditions, such as myocardial infarction (MI), but it is insufficient to reestablish heart function . Currently, several manipulations have been proven effective in promoting CM proliferation. Overexpressing several cell-cycle regulators, including CDK1, CCNB, CDK4 and CCND, could significantly promote adult mammalian CM proliferation . Also, exogenous administration of certain growth factors such as NRG1 , FGF1 and p38i  could promote CM proliferation in injured mammalian hearts. Modulation of some intrinsic signaling pathways could also promote CM proliferation. For instance, overexpressing Tbx20 induces mouse CM proliferation by activating Akt, Hippo and BMP signaling pathways, and knocking out Sav1 or overexpressing Yap to manipulate Hippo signaling pathway could enhance CM proliferation . Activation of the NRG1/ErbB4 signaling pathway by adding NRG1 or overexpressing ErbB4 could stimulate cell-cycle re-entry and division of CMs . Moreover, previous researchers have demonstrated that forced overexpression of certain microRNA (miRNAs) promotes CM proliferation in mice. Hsa-miR-590 and hsa-miR-199a were shown to promote cell-cycle re-entry of mouse CMs both ex vivo and in vivo . The miRNAs which induce CM proliferation in vitro mainly converge on the Hippo signaling pathway . Several miRNAs, including miR-302d, miR-373 and miR-590, have been proven to promote rat CM proliferation by activating the transcriptional cofactor Yap .
miRNAs are a group of small noncoding RNAs containing ∼22 nucleotides that function post-transcriptionally to regulate gene expression . Several miRNAs have been proven to sufficiently promote mammalian CM proliferation [23–25]. However, the functions and targets of miRNAs are poorly conserved in CMs among species . Human pluripotent stem cells (hPSCs), including both human-induced pluripotent stem cells (hiPSCs) and human embryonic stem cells (hESCs), can be efficiently differentiated into CMs, allowing investigation of miRNA functions in human CMs [27–29]. A recent large-scale screen revealed that miRNAs stimulating hESC-derived CM (hESC-CM) proliferation overlapped only minimally with those previously shown to stimulate rodent CM proliferation . Therefore, it is necessary to investigate whether miRNAs identified in rodents can efficiently promote human CM proliferation.
The miR-302/367 cluster, generally consisting of five members, miR-367, miR-302d, miR-302a, miR-302c and miR-302b, is ubiquitously distributed in vertebrates and occupies an intragenic cluster located in the gene La-related protein 7 (LARP7) . The cluster was revealed to play an important role in cell proliferation. In mouse embryonic stem cells (mESCs), miR-302 members promoted cell proliferation by directly silencing the expression of Cdkn1a and hence promoting the G1–S cell-cycle transition . In hESCs, however, miR-302 family members inhibited cell proliferation by simultaneously suppressing CDK2, CCND1/2 and BMI-1 to block the G1–S cell-cycle transition , indicating that miR-302 members may play opposite roles in mice and humans.
The miR-302/367 cluster is also important for CM proliferation in mice. This cluster regulates the Hippo pathway by targeting macrophage stimulating 1 (Mst1), large tumor suppressor 2 (Lats2) and MOB kinase activator 1B (Mob1b) in the heart . Overexpression of this cluster in the developing heart induced CM proliferation and cardiomegaly, while transient overexpression of the miR-302/367 cluster promoted cardiac regeneration, leading to improved cardiac contractile function after MI . Whether and how miR-302/367 members can promote human CM proliferation remains unknown. In the present study, we demonstrated that miR-302d overexpression could induce the proliferation of hPSC-derived CMs (hPSC-CMs) by targeting LATS2, thus offering an additional candidate molecule for cardiac regeneration in humans.
Materials and methods
Maintenance and direct differentiation of hESC-CMs and hiPSC-CMs
The H9 human ESC line (Wicell) and human iPSC line (kindly provided by Cellapy®) were seeded on Matrigel Matrix (Corning)-coated cell culture plates and maintained with mTeSR1™ medium (Stem Cell Technologies). The cells were regularly tested for mycoplasma contamination. Direct cardiac differentiation of hESCs and hiPSCs into beating CMs was conducted by sequentially modulating the Wnt pathway, according to a previously modified protocol . Briefly, the differentiation procedure was initiated when the cells reached 80–90% convergence by changing the culture medium to RPMI 1640 medium with l-glutamine (Thermo Fisher Scientific), supplemented with B27 minus insulin (Thermo Fisher Scientific) and the GSK3β inhibitor CHIR99021 (10 μM, Millipore). Twenty-four hours later (day 1), the medium was changed to fresh RPMI 1640 medium supplemented with B-27 minus insulin. On day 3, the medium was replaced with RPMI 1640 media with B-27 minus insulin and the Wnt pathway inhibitor IWP2 (5 mM, Millipore) for 48 h. From day 7, the cells were cultured in RPMI 1640 medium with B-27 supplement (Thermo Fisher Scientific), and spontaneous contraction of myocytes was usually observed after day 10. The hPSC-CMs were later purified using RPMI 1640 medium with B-27 supplement but without L-glutamine. The medium was replaced every 2 days, and the cells were maintained at 37°C and 5% CO2. Functional analyses of cells were performed on day 30 of induction unless indicated otherwise.
miRNA mimic transfection
The synthetic human miRNA-302d-3p mimics (5′-UAAGUGCUUCCAUGUUUGAGUGU-3′ and 5′-ACUCAAACAUGGAAGCACUUAUU-3′) and the negative control mimics (5′-UUCUCCGAACGUGUCACGUTT-3′ and 5′-ACGUGACACGUUCGGAGAATT-3′) were purchased from GenePharma. The miRNA-302d-3p mimics were transfected into hESC-CMs using a standard reverse transfection protocol at a final miRNA mimic concentration of 40 nM. Briefly, the transfection reagent (Lipofectamine 3000 Reagent, Thermo Fisher Scientific) was diluted in Opti-MEM (Thermo Fisher Scientific) and added to the medium. Five hours after transfection, the culture medium was replaced with fresh RPMI 1640 + B27 medium. The medium was replaced every 24 h. RNA was collected 48 h after transfection. Protein collection and the immunofluorescence (IF) imaging were conducted 72 h after transfection.
The hESC-CMs and hiPSC-CMs were fixed with 4% PFA for 20 min and permeabilized with 0.3% Triton X-100 in PBS solution for 20 min. Subsequently, the cells were blocked for 60 min in 3% BSA (Sigma) diluted with PBS. Later, the cells were stained for 1 h at room temperature with primary antibodies, mouse monoclonal antibody against sarcomeric α-actinin (Abcam) alone or together with rabbit antibody against Ki-67 (Cell Signaling Technology), diluted in 3% BSA. The cells were washed with PBS three times and incubated for 1 h with the secondary antibody conjugated to Alexa Fluor-594 goat anti-mouse IgG (Thermo Fisher Scientific).
For 5-ethynyl-2′-deoxyuridine (EdU) incorporation assay, the medium was replaced with medium containing 20 μM EdU (Thermo Fisher Scientific) 48 h after miRNA mimics or siRNA transfection. After 24 h of EdU incubation, the cells were fixed and processed for IF imaging following the above protocols. Cells were processed with the Click-IT EdU 488 Imaging kit (Thermo Fisher Scientific) according to the manufacturer’s protocols.
For cell apoptosis test, cells were processed to TUNEL staining with a one-step TUNEL apoptosis assay kit (Beyotime) following the manufacturer’s instructions after the IF imaging procedure. Cell nuclei were stained with DAPI (Thermo Fisher Scientific) diluted in PBS solution for 15 min.
mRNA/miRNA isolation and reverse transcription
The RNAsimple Total RNA Kit (Tiangen) was utilized to isolate total mRNA from cultured cells, and total miRNA was extracted with the miRcute miRNA Isolation Kit (Tiangen) following the manufacturer’s instructions. For the quantitative detection of mRNA, first-strand cDNA synthesis was conducted with the isolated RNA (100 ng) using the 5× All-In-One RT MasterMix (Applied Biological Materials). The miScript Reverse Transcription Kit (Qiagen) was used for the reverse transcription of miRNAs. Samples were run in biological triplicates. The concentration and quality of RNA and miRNA were assessed using a NanoDrop 2000 Instrument (Thermo Fisher Scientific).
Quantitative real-time PCR analysis
To quantitatively detect the expression of mRNA, 400 nM primers and 1×SYBR Green mix (Bimake) were mixed with 2 μl of cDNA template, and GAPDH was used to normalize the gene-specific expression levels. For the detection of miRNA, the miScript SYBR Green PCR kit (Qiagen) was used with specific primers for miR-302d-3p according to the manufacturer’s protocols. hU6 (TIANGEN) was utilized as a control for normalization of miRNAs. Quantitative real-time PCR (qRT-PCR) was carried out using the CFX96 Real-Time PCR Detection Systems (Bio-Rad Laboratories). The mRNA and miRNA relative expression was quantified by measuring the cycle threshold (Ct) values and standardized using the 2−ΔΔCt method relative to GAPDH. All primers are available in Supplementary Table S1.
The siRNAs were purchased from GenePharma. Transfection of siRNA into hESC-CMs and hiPSC-CMs was performed using Lipofectamine 3000 Reagent (Thermo Fisher Scientific) at a final concentration of 40 nM as described for the miRNA mimic transfection protocol. The negative control siRNA is the same with the negative control miRNA mimics. The culture medium was changed to fresh RPMI 1640 medium supplemented with B-27 for 5 h and was then replaced every 24 h. RNA was collected 48 h after transfection. Protein collection and the IF imaging were conducted 72 h after transfection. All siRNA sequences are available in Supplementary Table S2.
Luciferase reporter assays
To construct luciferase reporter plasmids for LATS2, a gene fragment containing the predicted target binding sites within the 3′-untranslated region (3′UTR) of LATS2 from human genomic DNA was amplified by polymerase chain reaction and inserted into the 3′UTR of the psi-CHECK2 luciferase plasmid. The predicted binding sites (positions 262-269 and 362-369) in the 3′UTR region of LATS2 were mutated and employed as contrast. Later, cotransfection of HEK293T cells was conducted in 24-well plates with the different luciferase reporter plasmids and miRNA-302d mimics (40 nM) using Lipofectamine 2000 Reagent (Thermo Fisher Scientific). The cells were cultured in DMEM for 36–48 h after cotransfection. Firefly and Renilla luciferase activities were measured using a Dual Luciferase Reporter Assay System (Promega), and the relative luciferase expression was measured on a GloMax 96 Microplate Luminometer (Promega). The luciferase reporter assays were conducted in biological triplicates.
Western blot analysis
Seventy-two hours after transfection of miRNA mimics or siRNA, cells were lysed with ice-cold NP-40 Lysis Buffer (Beyotime) following the manufacturer’s instructions. Extracted proteins were separated by SDS-PAGE and transferred to polyvinylidene fluoride (PVDF) membranes (Millipore), blocked in 5% BSA in 0.05% Triton X-100/TBS for 1 h, and incubated overnight with the following primary antibodies at indicated dilutions: YAP/TAZ (1:1000; #8418), Phospho-YAP (Ser127) (1:1000; #13008), and GAPDH (1:3000; #5147) obtained from Cell Signaling Technology. PVDF membranes were exposed to horseradish peroxidase (HRP)-conjugated secondary antibodies (Cell Signaling Technology), and signals were detected with the High-sig ECL Western Blotting Substrate (Tanon).
Plasmid construction and gene overexpression
To construct plasmids overexpressing CDK1, human CDK1 cDNA (kindly provided by Professor Jiahuai Han’s Lab) were cloned into pCDH-CMV-MCS-EF1-Puro using PCR/restriction digest-based cloning. The recombinant plasmid was extracted using the EndoFree Mini Plasmid Kit II (Tiangen), and the insertion was confirmed by Sanger sequencing (Genewiz). To overexpress CDK1 in hESC-CMs, the recombinant plasmid or vector were transfected to hESC-CMs using Lipofectamine 3000 Reagent (Thermo Fisher Scientific) following the manufacturer’s instructions.
Electrophysiology studies were carried out on the 40th day of differentiation. A multielectrode array (hPSC-derived CM) (Multi Channel Systems) was utilized to record the electrophysiology features, including beat period, field potential duration (FPD) and conduction velocity. FPD was subsequently corrected (FPDc) using Fridericia’s formula (FDP/3 √ RR, where RR = interspike interval). All experiments were performed in DMEM without FBS or antibiotics.
RNA-sequencing (RNA-seq) and bioinformatics analysis
RNA-sequencing (RNA-seq) experiments were performed by Shanghai Biotechnology Corporation (SBC). In brief, total RNA was isolated with an RNeasy Micro Kit (Qiagen) following the manufacturer’s instructions and assessed for an RIN number to inspect RNA integrity by an Agilent Bioanalyzer 2100 (Agilent Technologies). Total RNA was quantified with a Qubit Fluorometer. The cDNA library was constructed using a Truseq RNA Sample Preparation Guide (Illumina, Inc.) and sequenced on an Illumina HiSeq-2000 system (Illumina, Inc.). High-quality reads were aligned to the human mRNA reference sequence (RefSeq) for GRCh37/hg19 in the UCSC genome browser. The number of transcripts assigned to each gene was tallied and normalized to fragments per kilobase of exon model per million mapped reads (FPKM). Kyoto Encyclopedia of Genes and Genomes (KEGG) and Gene Ontology (GO) were utilized for the analyses of differentially expressed genes (DEGs). The RNA-seq data has been deposited to SRA database (accession number: PRJNA531900).
All results are presented as the mean ± SEM. Statistical analysis was performed with an unpaired two-tailed t test for comparisons between two groups, and one-way ANOVA followed by the Dunnett’s post-hoc test was used for experiments involving more than two groups. GraphPad Prism7 software (GraphPad) was adopted for statistical analysis. P-values below 0.05 (*), 0.01 (**) or 0.001 (***) were considered significant, and P-values are indicated in the figures.
miR-302d expression was down-regulated during hESC-CM differentiation
Early hPSC-CMs proliferate efficiently, similar to embryonic or fetal mammalian CMs, but the proliferative capacity decreases over time [32,33], allowing us an opportunity to study how miRNAs regulate CM proliferation during this process. The hESCs and hiPSCs could be efficiently differentiated into beating CMs by temporally manipulating the canonical Wnt signaling pathway (Figure 1A). Seven days after differentiation, the cells started contracting rhythmically. Thirty days after differentiation, CMs showed normal sarcomeric structure, as illustrated by IF staining of α-actinin, cardiac troponin T (cTnT) and DAPI (Figure 1B). To identify miRNAs that regulate CM proliferation, we profiled the temporal changes of miRNA expression at the genome-wide level during CM differentiation from hESCs at four stages: pluripotent stem cells (day 0), mesoderm (day 2), cardiac mesoderm (day 4), CM progenitor cells (day 6) and differentiated CMs (day 18) (Figure 1A) . We focused on miRNAs highly expressed on day 0 and day 18 as revealed by miRNA-seq analysis (FPKM value > 2000) and identified a total of 46 miRNAs (Figure 1C). Among these miRNAs, ten miRNAs were up-regulated (fold change > 2), and eight miRNAs were down-regulated (fold change < 0.5) from day 4 to day 18 (Figure 1C). miR-302d decreased most significantly during CM development (Figure 1D), which was confirmed by qRT-PCR (Figure 1E). Because the miR-302/367 members have been reported to regulate CM proliferation in mice, we tested whether miR-302d can regulate hPSC-CM proliferation.
miR-302d expression decreases during the in vitro differentiation of hESC-CMs
Overexpression of miRNA-302d was sufficient to induce proliferation of hPSC-CMs
To investigate the potential roles of miR-302d in CM proliferation, we transfected 27-day-old hESC-CMs with miR-302d mimics or a negative control (Supplementary Figure S1A). Transfection of miR-302d mimics led to miR-302d overexpression (Supplementary Figure S1B) and CM proliferation, as shown by cell quantification (Figure 2A,B). The proliferation of CMs was confirmed by EdU cell-proliferation assays. miR-302d transfection increased the percentage of EdU-positive hESC-CMs from 2.4 to 14.4%, while transfection with the negative control did not increase EdU-positive hESC-CMs (Figure 2C,D). We also used Ki-67 immunostaining to investigate the proliferation of CMs. The Ki-67-positive cells increased from 4.9 to 12.4% after miR-302d transfection (Figure 2E,F). In addition to hESC-CMs, the EdU assay revealed that miR-302d could promote hiPSC-derived CM (hiPSC-CM) proliferation. EdU-positive hiPSC-CMs increased from 2.6 to 5.7% after miR-302d transfection (Supplementary Figure S2). In summary, overexpression of miR-302d could promote proliferation of both hESC-CMs and hiPSC-CMs.
miR-302d overexpression promotes the proliferation of hESC-CMs
We further examined whether overexpression of miR-302d influenced other properties of CMs. Overexpression of miR-302d did not influence field potential duration (FPD, Fridericia corrected) (Supplementary Figure S3A), beat period (Supplementary Figure S3B), traces of field potentials (Supplementary Figure S3C) or cell size (Supplementary Figure S3D). Moreover, miR-302d mimics transfection did not influence hESC-CM apoptosis, according to TUNEL staining (Supplementary Figure S4). To test whether miR-302d influenced CM maturation, we counted the proportion of binucleated and multinucleated cells , and tested several CM maturation-related genes [36,37]. Overexpression of miR-302d increased binucleated cells from 5.8 to 7.2%, while the number of multinucleated cells remained unchanged (Supplementary Figure S5A). The expression pattern of the genes encoding troponin (TNNI1, TNNI3 and TNNT2), myosin (MYH6, MYH7, MYL2, MYL4 and MYL7) and ion channels (SCN5A and KCNJ2) was not changed by miR-302d mimics transfection (Supplementary Figure S5B). Together, these results indicate that miR-302d does not have a significant impact on CM maturation.
miR-302d induced hESC-CM proliferation by inhibiting the LATS2 gene
Three genes, Mst1, Lats2 and Mob1b, have been identified as targets of miR-302d in mice . We used TargetScan to predict the binding sites of miR-302d on these three genes and identified two binding sites in the 3′UTR of LATS2, located at positions 262–269 and 362–369 (Figure 3A). We failed to identify miR-302d-binding sites on human MST1 and MOB1B genes, a contrast to those in mice. We performed luciferase assays to test whether LATS2 is a direct target of miR-302d in HEK293T cells. We constructed reporter constructs containing luciferase fused to wild-type (WT) LATS2 3′UTR, a 3′UTR with a single binding site mutated (MT1/MT2) and a 3′UTR with both binding sites mutated (MT1+MT2) (Figure 3B). Compared with cotransfection of the negative control with the WT-3′UTR, cotransfection of miR-302d mimics with WT-3′UTR, MT1 or MT2 resulted in 58.4, 75.1 and 74.1% luciferase activity, respectively (Figure 3C). Mutation of both binding sites (MT1+MT2) completely abolished the inhibitory effect (Figure 3C). In addition, we analyzed LATS2 mRNA expression of hESC-CMs overexpressing miR-302d and found that LATS2 expression decreased to 45% of that of cells transfected with a negative control (Figure 3D). These data demonstrated that LATS2 is a direct target of miR-302d.
miR-302d induces CM proliferation by inhibiting LATS2
Moreover, we investigated whether down-regulation of LATS2 could promote CM proliferation. Transfection of siRNA targeting LATS2 (si-LATS2) resulted in down-regulation of LATS2 expression compared with transfection with negative control siRNA (Figure 3E). The proliferation of hiPSC-CMs transfected with si-LATS2 alone or together with miR-302d mimics was investigated by EdU assays. The percentage of EdU-positive hiPSC-CMs in the si-LATS2 group was approximately two times higher than that in the control group (Figure 3F,G), indicating that down-regulation of LATS2 could promote CM proliferation. Transfection of si-LATS2 together with miR-302d mimics could not further increase the EdU-positive hiPSC-CMs (Figure 3F,G). After transfection of hESC-CMs with si-LATS2, the percentage of EdU-positive cells was approximately six times higher than that in the control group (Supplementary Figure S6), while the percentage of TUNEL-positive cells did not change (Supplementary Figure S4), indicating that cell apoptosis was not significantly affected. Conclusively, these data demonstrated that miR-302d promotes CM proliferation by inhibiting LATS2 expression.
In mammals, LATS2 is the main member of the Hippo pathway that phosphorylates and inhibits the downstream effector YAP, thereby preventing its nuclear entry . To test whether miR-302d-mediated inhibition of LATS2 expression influences Yap/Taz signaling pathway, phosphorylated (Ser127) YAP (p-YAP) was analyzed by Western blot. In the hESC-CMs transfected with si-LATS2 or miR-302d mimics, the level of p-YAP decreased (Figure 3H). Moreover, expression of three YAP-regulated genes, AREG, CTGF and CXCL5, was significantly up-regulated (Figure 3I), indicating an increased nuclear activity of YAP. Taken together, miR-302d influenced Yap/Taz signaling pathway by inhibiting LATS2 expression.
miR-302d overexpression influenced the expression of cell cycle-related genes in hESC-CMs
To investigate the effects of miR-302d in hESC-CMs on molecular, cellular and signaling pathways, we processed hESC-CMs overexpressing miR-302d for RNA-seq analysis. A total of 711 DEGs (q-value < 0.05) were identified, including 459 up-regulated genes and 252 down-regulated genes (Figure 4A,B). KEGG enrichment analysis showed that 24 up-regulated genes belonged to the PI3K-Akt signaling pathway, 20 genes belonged to pathways related to the cell cycle, and 17 genes belonged to oocyte meiosis (Figure 4C). KEGG classification analysis showed that the most affected classes of genes are related to signal transduction, cancers and cell growth and death (Figure 4D). GO analysis revealed that the DEGs enriched in cellular components are related to proliferation, including condensed chromosome kinetochore and biological processes such as mitotic spindle midzone assembly (Supplementary Figure S7). We also performed qRT-PCR to test the expression of seven cell cycle-related genes, five of which were influenced by miR-302d overexpression (Figure 4E). These results indicated that miR-302d regulated CM proliferation by influencing multiple cell cycle-related genes.
miR-302d overexpression drives cell-cycle progression in hESC-CMs
To further explore the possible mechanism of these DEGs on CM proliferation, we knocked down four cell cycle-related genes (CCND1, CDKN1, PTEN and TP53. Supplementary Figure S8A–D), or overexpressed CDK1 gene (Supplementary Figure S8E). qRT-PCR results revealed that knockdown of CCND1, CDKN1 or PTEN in hESC-CMs decreased LATS2 expression (Figure 4F), while knockdown of TP53 or overexpression of CDK1 had no impact on LATS2 expression (Figure 4G). These results indicated that some differentially expressed cell cycle-related genes could inhibit LATS2 expression, and thus partially contribute to the CM proliferation.
Adult mammalian CMs have a limited regenerative capacity. Several methods have been developed to stimulate CM proliferation. Overexpressing cell-cycle regulators  or exogenous administration of certain growth factors [16,17] are pro-proliferative in injured mammalian hearts. Recently, a number of intrinsic signaling pathways were confirmed to be involved in CM proliferation and regeneration. CM proliferation could be enhanced by activating Akt, Hippo and BMP signaling pathways through overexpressing Tbx20 , and knocking out Sav1 or overexpressing Yap to modulate Hippo signaling pathway are also pro-proliferative . Adding NRG1 or overexpressing ErbB4 could activate the NRG1/ErbB4 signaling pathway, thus stimulate CM proliferation .
Several miRNAs have been proven to sufficiently induce rodent CM proliferation [23–25], and most of these pro-proliferative miRNAs converge on Hippo pathway [20,21]. However, whether these miRNAs could promote human CM proliferation is unclear. Therefore, investigating these miRNAs in human CMs is crucial. The miR-302/367 cluster has been shown to regulate the proliferation of several cell types, but the target genes and effects are different depending on the organs and species. Studies in mice have shown that the miR-302/367 cluster promotes proliferation of both mESCs and CMs [25,28]. The miR-302/367 cluster targets CDKN1A and hence promotes the G1–S cell-cycle transition in mESCs , while it targets Mst1, Lats2 and Mob1b, three genes involved in the Hippo pathway in mouse CMs .
In addition to those down-regulated miRNAs such as miR-302d, during hESC-CM differentiation, we observed a number of significantly up-regulated miRNAs, including miR-30c, miR-143, miR-30d and miR-22. miR-30 family participates in CM ischemic injury and autophagy . Recent studies have demonstrated that inhibiting miR-30 or miR-22 are beneficial for injured CM [41,42]. miR-143 regulates apoptosis pathways in the heart by repressing multiple anti-apoptosis targets . Interestingly, down-regulation of miR-143 is protective against oxidative stress in CMs .
In contrast to mESCs, the miR-302/367 cluster displays opposite effects on hESC proliferation because its target genes are different from those in mESCs. The miR-302/367 cluster inhibits hESC proliferation by targeting CDK2, CCND1/2 and BMI-1, blocking the G1–S cell-cycle transition . Whether the miR-302/367 cluster promotes or inhibits human CM proliferation and its target genes in human CMs remain unknown. Our study provides further evidence for the therapeutic potential of miR-302d. We demonstrated that miRNA-302d, a member of the miR-302/367 cluster, could efficiently promote hPSC-CM proliferation. We further demonstrated that miRNA-302d promoted hPSC-CM proliferation by targeting LATS2. Therefore, we provide further evidence for the therapeutic potential of the miR-302/367 cluster. In future work, whether other members of the miR-302/367 cluster can promote human CM proliferation should be investigated.
Our RNA-seq results showed that overexpressing miR-302d in hESC-CMs could influence the relative expression of multiple cell-cycle genes, including CCND1, CDKN1 and PTEN, while knocking down those genes in hESC-CMs has a repression effect on LATS2. These results indicate that the interaction between miR-302d and LATS2 is not the only mechanism by which miR-302d promotes the proliferation of hESC-CMs. Moreover, in human neuroblastoma SK-N-MC cells, miR-302 could activate PI3K-Akt signaling pathway by silencing PTEN, an important regulator of the PI3K-Akt signaling pathway. Therefore, PTEN could be an alternative target of miR-302d in CMs, while more research needs to be done in this field.
Adult CMs can proliferate at a low level after injury, and several methods have been developed to promote cardiac regeneration. Although several miRNAs have been proven to induce rodent CM proliferation, whether miRNAs identified in rodents could promote human CM proliferation is unknown, due to the poorly conserved functions of miRNAs among species.
We conducted an RNA-seq in hESC-CMs, revealing that the expression of miR-302d decreased significantly during CM differentiation from hESCs. Further, we demonstrated that miR-302d overexpression could induce the proliferation of hPSC-CMs by targeting LATS2, and overexpression of miR-302d induced changes in gene expression, which mainly converged on the cell cycle.
We provided evidence for the therapeutic potential of miR-302d in the human heart, which could be further utilized in the treatment of diseases such as myocardial infarction.
This work was supported by the National Natural Science Foundation of China [grant number 81870199]; the National Basic Research Program of China [grant number 2015CB943300]; the Foundation for Innovative Research Group of the National Natural Science Foundation of China [grant number 31521003]; and the Opening Program 2018 of the State Key Laboratory of Genetic Engineering [grant number SKLGE1809].
The authors declare that there are no competing interests associated with the manuscript.
F.X. designed and performed the experiments; J.-C.Y. and J.S. analyzed the miRNA-seq and RNA-seq data; M.-M.L. performed parts of the experiments; L,-M.S. revised the manuscript; F.L., Y.-M.W., L.S. and J.-B.G. supervised the project. All authors read and approved the final manuscript.
Dulbecco’s Modified Eagle Medium
fetal bovine serum
fragments per kilobase of exon model per million mapped read
human embryonic stem cell
human-induced pluripotent stem
human pluripotent stem cell
kyoto encyclopedia of genes and genome
large tumor suppressor 2
mouse embryonic stem cell
quantitative real-time PCR
sodium dodecyl sulphate polyacrylamide gel electrophoresis
TdT-mediated dUTP nick end labeling