Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2017 Oct 27.
Published in final edited form as: Cell Rep. 2017 Sep 26;20(13):3014–3024. doi: 10.1016/j.celrep.2017.09.005

Comparative gene expression analyses reveal distinct molecular signature between differentially reprogrammed cardiomyocytes

Yang Zhou 1,2, Li Wang 1,2, Ziqing Liu 1,2, Sahar Alimohamadi 1,2, Chaoying Yin 1,2, Jiandong Liu 1,2, Li Qian 1,2,3
PMCID: PMC5659840  NIHMSID: NIHMS904665  PMID: 28954220

SUMMARY

Cardiomyocytes derived from induced pluripotent stem cells (iPSC-CMs) or directly reprogrammed from non-myocytes (induced cardiomyocytes, iCMs) are promising sources for heart regeneration or disease modeling. However, the similarities and differences between iPSC-CM and iCM are still unknown. Here we performed transcriptome analyses of beating iPSC-CMs and iCMs generated from cardiac fibroblasts (CFs) of the same origin. Although both iPSC-CMs and iCMs establish CM-like molecular features globally, iPSC-CMs exhibit a relatively hyperdynamic epigenetic status while iCMs exhibit maturation status that more resemble adult CMs. Based on gene expression of metabolic enzymes, iPSC-CMs primarily employ glycolysis while iCMs utilize fatty acid oxidation as the main pathway. Importantly, iPSC-CMs and iCMs exhibit different cell cycle status, alteration of which influenced their maturation. Therefore, our study provides a foundation for understanding the pros and cons of different reprogramming approaches.

Keywords: cardiac reprogramming, cardiomyocyte, iCM, iPSC, cardiac differentiation, gene expression profiling

eTOC

Zhou et al. performed gene expression profiling of cardiomyocytes derived from induced pluripotent stem cells (iPSC-CMs) or generated through direct reprogramming (induced cardiomyocytes, iCMs) from cardiac fibroblasts of the same origin. Comparative analyses revealed similarities and differences in the molecular signatures of iPSC-CMs and iCMs.

graphic file with name nihms904665u1.jpg

INTRODUCTION

Heart disease continues to be the leading cause of death worldwide, partly due to the limited therapeutic options. Because mammalian adult heart has little regenerative potential upon injury (Laflamme and Murry, 2011; Xin et al., 2013), generation of functional cardiomyocytes (CMs) or CM-like cells for replacement therapies offers alternative strategies for repairing damaged hearts. Recent advances in cellular reprogramming have made it possible to generate autologous CMs by cardiac differentiation of induced pluripotent stem cells (iPSC) (Takahashi and Yamanaka, 2006; Zhang et al., 2009) or by iCM reprogramming directly from fibroblasts (Ieda et al., 2010; Nam et al., 2013; Qian et al., 2012; Song et al., 2012). Efficient differentiation of iPSCs into functional CMs has been achieved by stepwise induction that partially recapitulates cardiac differentiation during embryonic development (Kattman et al., 2011). These in vitro generated iPSC-CMs are relatively immature, reflected at both functional and molecular levels (Lundy et al., 2013; Uosaki et al., 2015; Yang et al., 2014). Direct reprogramming of non-myocytes into iCMs has been achieved by different combination of factors (Addis et al., 2013; Fu et al., 2013; Hirai et al., 2013; Jayawardena et al., 2012; Wada et al., 2013). Considerable progress has been made towards improving iCM reprogramming efficiency and understanding the underlying molecular mechanisms (Abad et al., 2017; Dal-Pra et al., 2017; Liu et al., 2016a, 2016b; Mohamed et al., 2016; Muraoka et al., 2014; Nam et al., 2014; Wang et al., 2015a; Yamakawa et al., 2015; Zhao et al., 2015; Zhou et al., 2015, 2016). However, it remains largely unknown about the cellular and molecular details of iCMs, neither is known about the phenotypic similarities or differences between iCMs and iPSC-CMs.

In this study, we performed a comparative analysis of the molecular characteristics of reprogrammed contractile CMs generated by direct reprogramming or differentiation of iPSCs from cardiac fibroblasts (CFs) of the same origin. Although both reprogrammed CMs possess cardiac features at the transcriptional level, in-depth analyses revealed that iPSC-CMs more closely resembled embryonic/fetal CMs and possessed a relative hyperdynamic epigenetic status when compared to iCMs. Furthermore, a focused analysis of metabolic genes and cell cycle regulators indicated that iCMs upregulated genes related to fatty acid oxidation and inactive cell cycle status. In contrast, iPSC-CMs exhibited higher expression in genes involved in glycolysis and active cell cycle status. We also showed that inhibition of cell cycle progression facilitated certain maturation features in iPSC-CMs. Taken together, our comparative analyses between iPSC-CMs and iCMs derived from CFs of the same origin suggested that reprogrammed CMs generated by different approaches are distinct at their molecular characteristics, thereby providing guidance for future clinical applications of various cellular reprogramming approaches.

RESULTS

Generation of iPSC-CM and iCM from CF of the Same Origin

To minimize the effects of genetic background, line-to-line variations and epigenetic memories of tissue specific fibroblasts, we used CFs of the same origin from αMHC-GFP transgenic pups (P1.5) (Ieda et al., 2010). After transduction with OSKM (Oct4, Sox2, Klf4 and c-Myc) (Takahashi and Yamanaka, 2006), two independent iPSC lines were established and subsequently differentiated into CMs. Meanwhile, iCMs were generated by infecting CFs with retroviruses expressing a polycistronic Mef2c, Gata4 and Tbx5 transgene (in short MGT) (Wang et al., 2015a). Since cardiac differentiation from iPSCs and direct reprogramming of iCMs are two different processes, it is difficult to identify CMs of comparable stage by marker expression. Furthermore, there was no single marker that exclusively labels reprogrammed cardiomyocytes for enrichment; selection based on one marker (i.e. αMHC-GFP) may result in exclusion of more fully reprogrammed or differentiated myocytes. Therefore, we decided to use all cells that have undergone reprogramming/differentiation process at the stage when contractile phenotypes can be readily observed for comparative analysis. To this end, we collected beating iPSC-CMs and iCMs for microarray analysis and also included endogenous neonatal CFs (neoCFs) and CMs (neoCMs) isolated from the same transgenic pups (P1.5) as controls (Figure 1A).

Figure 1. Generation, Characterization and Genome-wide Comparison of iCMs and iPSC-CMs.

Figure 1

Open in a new tab

(A) Schematic of comparative transcriptome analyses of iCMs and iPSC-CMs derived from CFs of the same origin.

(B) iPSC lines established from CFs.

(C) Immunostaining of iPSCs for SSEA1 and Oct4.

(D) qPCR of pluripotent genes in CF-derived iPSCs. Embryonic stem cell line E14 was used as a positive control. n=3, error bars indicated SEM.

(E) ICC for MAP2 (ectoderm), cTnT (mesoderm), αSMA (mesoderm) and PECAM-1 (endoderm) in differentiated cells derived from iPSCs.

(F) ICC of day 14 iCMs against GFP and αActinin.

(G) Pearson’s correlation heatmap of indicated samples.

(H,I) PCA of whole genome expression profiles from microarray experiments as listed, with a scatter plot of PC1 versus PC2 (H) and a scatter plot of PC2 versus PC3 (I). Samples were grouped by different colors as indicated.

All scale bars are 100 μm except in (C) are 200 μm.

See also Figures S1 and S2, Movie S1.

Two iPSC lines named as iPSC1 and iPSC2 with typical pluripotent stem cell morphology were manually selected and expanded in feeder free culture system (Figure 1B). Pluripotency of both lines is further confirmed by immunocytochemistry (ICC) (Figure 1C) and qPCR (Figure 1D) of pluripotent markers. To evaluate the differentiation potential of iPSCs, we generated embryoid bodies (EBs) that subsequently differentiated into three germ layers expressing germ-layer specific markers (Figures 1E and S1A). We then used the well-established protocol to differentiate iPSCs into CMs (Kattman et al., 2011), and collected beating iPSC-CMs in maturation medium (Movie S1). To obtain beating iCMs, retroviruses expressing polycistronic MGT were transduced into primary CFs. These iCMs turned on αMHC-GFP reporter, αActinin and cardiac Troponin T (cTnT) at day 14 (Figures 1F and S1B), and then showed spontaneous contractile activity after another 14 days in culture of maturation medium (Movie S1). Thus, both types of reprogrammed CMs derived from CFs of the same origin were harvested at beating stage from the same maturation medium.

Similar and Different Molecular Features of iPSC-CMs and iCMs

To investigate the global transcription profiles of reprogrammed CMs, we performed microarray experiments of iPSC-CMs, iCMs, neoCMs and neoCFs on Agilent mouse 8x60K GE 1-color platform. After background correction and normalization, samples were clustered by correlation coefficient. According to the heatmap of correlation coefficient, biological duplicates of iCMs and iPSC-CMs were highly correlated and grouped together, while iCMs and iPSC-CMs showed distinct gene expression profiles (Figure 1G and S2A). Then, principal component analysis (PCA) was used to discern variations among samples (Figure S2B). PCA analyses showed that PC1 and PC2 reflected the divergence among different CMs (Figure 1H), while PC3 mainly accounted for the variations between CMs and CFs (Figure 1I) and PC4 revealed the differences between iPSC-CMs derived from the two iPSC lines (Figure S2C). Thus, these data indicated that the three types of CMs share common gene expression patterns when compared to CFs, but also exhibit specific gene signatures for each of them.

To further interrogate the similarities and differences between iPSC-CMs and iCMs, we applied Gene Ontology (GO) analysis to characterize differentially expressed genes among these cell types. First, we selected significantly up-regulated and down-regulated genes (at least 2-fold changes, p < 0.05) in all three types of CMs when compared to CFs (Figure 2A). GO results showed that biological processes related to CM were enriched in iCMs, while fibroblast related biological processes were suppressed (Figure 2B). Next, to determine the unique molecular features of iPSC-CMs or iCMs among three types of CMs, we focused on the uniquely expressed genes that displayed at least 2-fold changes (p < 0.05) in expression levels when compared to the other two types of CMs and hierarchically clustered these genes into groups (Figures 2C and 2E). We found that some categories of iCM-high genes partially overlapped with those of fibroblast-high genes (Figures 2B and 2D), suggesting residual trace of CF molecular signature in reprogrammed iCMs. Importantly, metabolic-related genes were uniquely upregulated in iCMs while cell cycle-related genes were markedly suppressed in iCMs, suggesting unique metabolic and cell cycle features of iCMs (Figure 2D, also see below for more details). Interestingly, we found that the major GO terms enriched for iPSC-CM-high genes were “regulation of transcription DNA-templated”, “mRNA transport”, “RNA splicing” and “covalent chromatin modification” (Figure 2F, also see below for more details), indicating that iPSC-CMs might undergo active transcription and have hyperdynamic epigenetic regulation after shortly differentiated from iPSCs (Meshorer et al., 2006). Taken together, these data revealed that although reprogrammed CMs possess the identity of endogenous neonatal CMs, each type exhibits unique molecular features.

Figure 2. Genetic and Epigenetic Differences between iCMs and iPSC-CMs.

Figure 2

Open in a new tab

(A,C,E) Differentially expressed genes with at least 2-fold changes in all types of CMs when compared with CFs were grouped and clustered as CM-specific genes (A), with at least 2-fold changes in iCMs when compared with the rest of CMs were grouped and clustered as iCM-specific genes (C), with at least 2-fold changes in iPSC-CMs when compared with the rest of CMs were grouped and clustered as iPSC-CM-specific genes (E).

(B,D,F) GO terms of biological process enriched in CM-specific (B), iCM-specific (D), iPSC-CM-specific (E) genes. Numbers in parenthesis indicate the gene number of each GO term. Top panel plot represents enriched functional categories for upregulated genes; bottom panel represents enriched functional categories for downregulated genes.

(G) GSEA shows positive correlation of chromatin modification (left) and chromatin remodeling (right) genes in iPSC-CMs relative to iCMs (top) or neoCMs (bottom).

(H) Western blot of H3K4me3, H3K27me3, H3K9me2, H3K27ac on iPSCs, iPSC-CMs, iCMs and neoCFs. H3 was used as the loading control. Quantification was shown on the right.

(I, J) ChIP-qPCR for H3K4me3 (I) or H3K27me3 (J) at indicated gene loci on iCMs and iPSC-CMs. Chr8 and Actb were used as negative and positive control respectively in (I). Actb was used as a negative control in (J).

n=3 for (H, I, J), error bars indicated SEM; *p < 0.5, **p < 0.01.

Epigenetic Differences between iPSC-CMs and iCMs

GO analysis revealed that genes highly expressed in iPSC-CMs were enriched in epigenetic processes including covalent chromatin modification and methylation (Figure 2F), it is thus intriguing to determine the epigenetic status between iPSC-CMs and iCMs. To further confirm the enrichment of chromatin regulation genes in iPSC-CMs compared to iCMs or neoCMs, we performed gene set enrichment analysis (GSEA) based on GO terms “chromatin modification” and “chromatin remodeling”. Consistently, genes in both GO terms were positively enriched in iPSC-CMs (Figure 2G). Moreover, western blot analysis of a set of histone marks demonstrated higher level of active mark H3K4me3 and lower level of repressive mark H3K27me3 in iPSC-CMs compared to iCMs, suggesting the relatively active global epigenetic status in iPSC-CMs (Figure 2H). Additionally, we found opposite trends in epigenetic re-patterning of H3K4me3 and H3K27me3 along iPSC-CM differentiation and iCM reprogramming, when comparing the two differentially reprogrammed CMs with their starting cells (Figure 2H). Next, we also determined the physical occupancy of H3K4me3 and H3K27me3 at selected loci of cardiac, fibroblast and pluripotent genes in iPSC-CMs and iCMs by ChIP-qPCR. At cardiac loci, a higher H3K4me3 level was observed in iPSC-CMs (Figure 2I). Examination of fibroblast genes revealed higher H3K4me3 in iCMs than in iPSC-CMs (Figure 2I). Meanwhile, ChIP-qPCR at pluripotent loci in iPSC-CMs revealed significant increases in both H3K4me3 and H3K27me3 levels associated with bivalent modification in iPSCs (Figures 2I and 2J). Together, our data are suggestive of distinct patterns of chromatin modifications in reprogrammed CMs via the two different routes.

Differed Maturation Statuses of iPSC-CMs and iCMs

Uosaki and colleagues studied transcriptional landscape of CM maturation based on a number of published microarray data and identified sets of differentially regulated genes during CM maturation (Uosaki et al., 2015). This study prompted us to determine whether reprogrammed iCMs and iPSC-CMs represent CMs at different maturation stages. To this end, we evaluated the maturation status of reprogrammed CMs with GSEA based on the CM-maturation-related gene sets identified by Uosaki et al. Notably, we found that gene sets associated with early embryonic (E8–E11) CMs were significantly enriched in iPSC-CMs. In contrast, gene sets associated with mature CMs were underrepresented in iPSC-CMs, but highly enriched in iCMs (Figure 3A). As expected, gene expression of the control neoCMs was positively correlated with that of late embryonic (E16–E18)/neonatal (P3–P10) CMs, suggesting its relative mature stage when compared with iPSC-CMs (Figure 3B). Interestingly, GSEA of neoCMs and iCMs showed that iCMs displayed a higher correlation with gene sets of mature CMs (Adult) than neoCMs (Figure 3C). Based on the pairwise comparison of GSEA, we concluded that, among all three types of CMs, iPSC-CMs mostly resemble early embryonic CMs and iCMs exhibit molecular signature more similar to that of adult CMs (Figure 3D). Next, we asked if iCM reprogramming is a progressing process towards or a more direct conversion to adult-like CMs. Thus, MGT- and LacZ (as controls)- infected reprogrammed cells at different time points (day 0, 3, 5, 7, 10, 14) were harvested and subjected to microarray analysis. Overall PCA plot indicated distinct dynamic changes in expression pattern of reprogrammed and control cells over time (Figure S2D). Furthermore, gene set enrichment score of early embryonic CM-related genes decreased and that of mature CM-associated genes increased along iCM reprogramming, suggesting that rapid maturation might occur during iCM induction (Figure 3E). The heatmap of genes involved in early and mature gene sets also demonstrated similar changes (Figure 3F). In conclusion, we found that unlike what has been reported for iPSC-CMs, the directly reprogrammed iCMs seem to rapidly gain the adult-like CM features at the transcription level.

Figure 3. Evaluation of Maturation Status of iCMs and iPSC-CMs.

Figure 3

Open in a new tab

(A–C) GSEA shows enrichment of indicated gene sets associated with early (E8–E11), middle (E12–E14), late (E16–E18)/neonatal (P3–P10) and mature (adult heart) stages (Uosaki et al., 2015) of heart development (from left to right) in iPSC-CMs versus iCMs (A), in iPSC-CMs versus neoCMs (B), in iCMs versus neoCMs (C).

(D) Illustration of distinct maturation stages of heart development corresponding to iPSC-CMs, iCMs and neoCMs.

(E) Lined scatter plot of the enrichment scores of early, middle, late, mature gene sets highlighted by different colors corresponding to time points during iCM reprogramming.

(F) Hierarchical clustering and heatmap of the expression of genes related to different developmental stages (green, early genes; purple, mature genes) from D0,3,5,7,10,14 and beating iCMs.

(G) Hierarchical clustering of long-term cultured iPSC-CMs with other types of CMs based on the enrichment scores of different gene sets indicated in (A). ST, short-term cultured iPSC-CMs; LT, long-term cultured iPSC-CMs.

(H) Representative ICC images of sarcomere structure labeled by αActinin in iCMs, iPSC-CM-ST and iPSC-CM-LT. Top panel is a high-magnification image of the area highlighted by rectangle in bottom panel. Scale bars, 10 μm.

(I) Representative calcium transient images and quantitative traces of fluorescence (below) in iCMs and iPSC-CMs as indicated. RFU, Relative Fluorescence Units, was calculated by dividing background fluorescence intensity.

See also Figures S2 and S3, Movie S2.

Because recent studies demonstrated that prolonged culturing could enhance PSC-CM maturation (Lundy et al., 2013; Uosaki et al., 2015), we intended to further compare long-term cultured iPSC-CMs to iCMs. We cultured beating iPSC-CMs for additional 20 days and collected them for comparative transcriptome analysis. The unsupervised hierarchical clustering based on GSEA enrichment score showed that long-term cultured iPSC-CMs are more similar to short-term cultured iPSC-CMs in maturation status than iCMs or neoCMs (Figure 3G). Of note, long-term culturing strengthened the enrichment score of mature gene set, but failed to diminish the enrichment score of early gene set. Consistently, hierarchical clustering of maturation-related genes demonstrated similar results (Figure S3A), indicating that different maturation routes occur during iCM reprogramming and long-term culture of iPSC-CMs.

To further explore distinct features of CMs related to maturation, we evaluated the sarcomere structure by ICC of a cardiac Z-disc protein, αActinin, on beating iCMs and iPSC-CMs (Figure 4H). Beating iCMs demonstrated repetitive band alignment of organized sarcomeres with clear registration across the cell, while beating iPSC-CMs showed less organized sarcomeres. Moreover, long-term cultured iPSC-CMs exhibited sarcomere structures similar to those in iCMs, indicating that beating iPSC-CMs are maturing in a progressive manner that is different from the rapid acquisition of maturation in iCMs. In addition, calcium imaging was performed to assess physiological features of iCMs and iPSC-CMs. Calcium flux was measured in representative beating cells from each type of CMs (Figure 3I, Movie S2). It is noticeable that a higher magnitude of fluorescent intensity changes was detected in iCMs than in iPSC-CMs. Taken together, our data demonstrate distinct maturation statuses of iCMs and iPSC-CMs that are in parallel exemplified with different sarcomere structures and calcium oscillation characteristics. However, long term culturing could enhance certain maturation characteristics of iPSC-CMs.

Figure 4. Differences in Metabolic and Cell Cycle Status of iCMs and iPSC-CMs.

Figure 4

Open in a new tab

(A) Maps of major metabolic pathways listing key genes encoding metabolic enzymes and summarizing relative expression changes of these genes in different CMs. Genes with at least 1.5-fold changes in mean expression value when compared between iCMs and iPSC-CMs were defined as highly expressed genes in iCMs (red) or iPSC-CMs (blue). Unchanged genes were marked in white.

(B) qPCR of indicated genes related to glycolysis or FA oxidation/PPP. Fold changes (FC) were presented as log2 of expression values in iCMs compared to iPSC1-CMs. n=3, error bars indicated SEM.

(C) Heatmap representation of KEGG_CELL_CYCLE gene set shows its suppression in iCMs compared to iPSC-CMs.

(D) GSEA shows positive correlation of cell cycle genes in iPSC-CMs compared to iCMs.

(E) Differentially expressed cyclins and cyclin-dependent kinases when compared between iCMs and iPSC-CMs. Bar graph shows log2 fold changes of mean expression value from microarray samples. n=4, error bars indicated SEM.

(F) Representative ICC images and quantification for % of αActinin+ cells and % of Ki67+ cells out of αActinin+ iPSC-CMs and iCMs. Bottom panels are high magnification images for areas indicated in top panels. n=14 for iPSC-CMs and n=7 for iCMs, error bars indicated SEM.

(G) Representative ICC images and quantification for % of Ki67+ cells out of αActinin+ iPSC-CMs, % of αActinin+ cells and % of cells with sarcomeres out of αActinin+ iPSC-CMs at day 18 with or without 2-day treatment of 10 μM MMC. n=10, error bars indicated SEM; **p < 0.01.

All scale bars are 100 μm.

See also Figures S3 and S4, Movie S1.

Unique Metabolic and Cell Cycle Signatures in iPSC-CMs and iCMs

According to our GO results, we then compared the expression of genes involved in metabolism and cell cycle regulation among various CM types. To explore the expression pattern of metabolic genes in iPSC-CMs and iCMs, we selected key genes involved in energy metabolism and performed hierarchical clustering (Figure S3B). It is noticeable that genes with higher expression in iCMs were mostly grouped in fatty acid (FA) oxidation, while most glycolytic genes were expressed at a higher level in iPSC-CMs. Furthermore, to identify the metabolic pathway preferentially utilized by each type of CMs, we determined the relative expression changes (defined as the fold change of mean expression in iCMs versus that in iPSC-CMs) of genes involved in glycolysis, FA oxidation, pentose phosphate pathway (PPP) and TCA and highlighted in different colors (Figures 4A and S3C). Genes involved in glycolysis, which is the primary means of energy production in early embryonic hearts (Lopaschuk and Jaswal, 2010), were expressed at a lower level in iCMs. In contrast, genes associated with PPP and FA oxidation were expressed at a higher level in iCMs than those in iPSC-CMs (Figure 4A). In adult CMs, PPP is essential for the preservation of contractile function through its role of oxidative defense and FA oxidation is the major energy source (Jain et al., 2003; Lopaschuk and Jaswal, 2010). The expression changes of key enzymes in these pathways were further validated by qRT-PCR (Figure 4B). We also discovered the progressive upregulation of FA oxidation-related genes and downregulation of glycolysis genes along iCM induction (Figure S3D). Finally, we determined metabolic changes in long-term cultured iPSC-CMs (Figure S3E). Interestingly, we found that long-term cultured iPSC-CM is grouped together with iCM and neoCM, suggesting enhanced maturation at metabolic level resulted from long-term culturing.

Furthermore, we performed gene expression analysis of cell cycle gene set from Kyoto Encyclopedia of Genes and Genomes (KEGG), and demonstrated that many of cell cycle-related genes, including cell-division cycle (Cdc), cyclin and cyclin-dependent kinase (Cdk) genes, were repressed in iCMs while positively enriched in iPSC-CMs (Figure 4C). Similar results were obtained by GSEA (Figure 4D). In addition, stepwise gene expression changes of cell cycle genes were observed along iCM reprogramming (Figure S4A). The cyclins and cyclin-dependent kinases are known to be down-regulated in developing CMs (Kang and Koh, 1997). Thus, we compared the expression of representative Cdc and Cdk genes between iCMs and iPSC-CMs (Figure 4E). Consistently, genes related to G1/S and G2/M were repressed in iCMs, suggesting inactive cell cycle in iCMs, reminiscent of the quiescent status in adult CMs. In addition, the hierarchical clustering of cell cycle genes demonstrated that long-term cultured iPSC-CMs appeared to be similar to iCMs and neoCMs (Figure S4B). This observation is consistent with expression changes of the genes related to maturation and metabolism.

We were intrigued by the finding that alteration in transcription level of cell cycle genes was correlated with changes in expression of maturation genes when comparing iCMs to iPSC-CMs. Therefore, we sought to determine if cell cycle status of various CMs is related to their respective maturation status and if the alteration in cell cycle could influence maturation. First, we measured the expression of active cell cycle marker Ki67 in cardiac marker αActinin+ iCMs and iPSC-CMs. As expected, 20% of αActinin+ iPSC-CMs expressed Ki67, while none of iCMs were Ki67 positive (Figure 4F). To determine the effect of alteration in cell cycle on CM maturation, we used Mitomycin C (MMC) to block cell division of iPSC-CMs at the time they initiated contraction. After MMC treatment, no Ki67+ cells were found in αActinin+ iPSC-CMs (Figure 4G). Importantly, MMC treatment increased the percentage of αActinin + cells and promoted the assembly of sarcomeres in iPSC-CMs (Figure 4G). In addition, we observed faster beating rate in iPSC-CMs after MMC treatment (Movie S1). These data suggest that inhibition of cell cycle progression promotes the maturation of iPSC-CMs, at least the phenotypes we examined here. Therefore, the distinct cell cycle statuses of iCMs and iPSC-CMs might be one of the mechanisms underlying their differed maturation status.

DISCUSSION

iPSC-CMs and iCMs have emerged as appealing cardiomyocyte-like cells for disease modeling and potential therapeutic application. Reprogrammed iCMs from CFs showed some cellular features resembling adult-like ventricular cells (Ieda et al., 2010). In our study, we found that beating iPSC-CMs are more similar to the early embryonic CMs. Although long-term culture promoted organization of sarcomeres and switch of metabolic pathway (Figures 3H and S3E), consistent with a previous report (Uosaki et al., 2015), long-term cultured iPSC-CMs are limited to an immature stage according to our GSEA results (Figures 3A and 3G), suggesting that roadblocks exist to block further maturation of iPSC-CMs during late stage of iPSC differentiation.

The distinct molecular features between iCMs and iPSC-CMs imply that cardiac reprogramming and CM differentiation from iPSC take distinct routes and might also involve different signaling cascades to acquire cardiomyocyte fate in a non-myocyte cell. The relative rapid acquisition of early maturation features in iCM at reprogramming day 3 (Figure 3E) might result from the transient high expression of exogenous M/G/T (Ieda et al., 2010; Wang et al., 2015a). In addition, it is noticeable that iPSC-CMs might possess carry-over covalent chromatin modifications from iPSCs, suggesting that iPSC-CMs and iCMs might carry distinct epigenetic memories (Figures 2G–2J). These traces of epigenetic memories in CMs reprogrammed via different routes may be barriers to CM maturation. Interestingly, among the differentially expressed genes between iPSC-CMs and iCMs, we also identified a large number of noncoding RNAs (Table S1) that could contribute to the differed CM features. In spite of the heterogeneity of reprogrammed cells at any stage of such unsynchronized process, our genome-wide study using pooled cells still identified major molecular features of iCMs and iPSC-CMs, further indicating that the differences in their molecular features are significant enough that were not masked by averaging-out of the signals. Therefore, our comparative transcriptome profiling of iCMs and iPSC-CMs will be useful to help determine under which scenario one or the other type of derived CM is more desired for applications such as disease modeling, drug screen or in vivo repair.

EXPERIMENTAL PROCEDURES

Mouse Lines

αMHC promoter driven-GFP transgenic mice were described previously (Ieda et al., 2010). Animal care was performed in accordance with the guidelines established by the University of North Carolina, Chapel Hill. All mouse protocols were approved by the Institutional Animal Care and Use Committee (IACUC), University of North Carolina, Chapel Hill.

Plasmids and Viral Packaging

Retroviral constructs pMXs-Oct4, pMXs-Sox2, pMXs-Klf4 and pMXs-cMyc were used for iPSC generation (Takahashi and Yamanaka, 2006). To generate beating iCMs, single retroviral construct expressing polycistronic MGT was used (Wang et al., 2015b). The protocol of viral packaging is provided in the Supplemental Information.

Isolation of neoCFs, neoCMs, and iCM Reprogramming

Cardiac fibroblast isolation was performed as previously described (Wang et al., 2015b). For the isolation of neoCMs, digested heart cells were applied to MACS Dissociator (Miltenyi Biotec) and underwent MACS by using Neonatal Cardiomyocyte Isolation Cocktail (Miltenyi Biotec) to collect flow through containing unlabeled CMs. Detailed protocols of primary cell culture are available in the Supplemental Information. iCM reprogramming was performed as previously described (Zhou et al., 2016). 28 days after viral infection, spontaneously beating cells were observed and harvested for RNA extraction, ICC, ChIP, Calcium Imaging and Western.

iPSC Generation and Cardiac Differentiation

iPSC lines were generated as previously described (Takahashi et al., 2007) with minor modification as described in the Supplemental Information. Cardiac differentiation of iPSCs was performed according to a previously described but slightly modified protocol (Kattman et al., 2011). Detailed experimental procedures related to iPSC and iPSC-CM are provided in the Supplemental Information.

Microarray Analysis

Isolated and purified RNA from indicated samples was applied to Agilent Mouse Gene Expression 8x60K Microarray. Data were processed using the limma package from Bioconductor (Ritchie et al., 2015). PCA analysis was performed using the ‘prcomp’ package in R (www.r-project.org) with the normalized gene expression data. GSEA was performed using GSEA 2-2.2.0 software for testing specific gene sets as previously described (Lu et al., 2016). For GSEA on more than 2 groups, the enrichment scores were calculated using the R package ‘GSVA’ with ‘gsva’ method and default settings (Hänzelmann et al., 2013). The gene expression clustering and heatmaps were generated by the ‘pheatmap’ package in R. All microarray data reported in this paper have been deposited into Gene Expression Omnibus (GEO) with the accession number: GSE 99814

RT-qPCR, ChIP-qPCR and Western Blot

RT-qPCR was performed as described previously (Zhou et al., 2016). ChIP was performed using MAGnify ChIP System (ThermoFisher) according to the manufacturers’ instructions with α-H3K4me3/α-H3K27me3/α-IgG antibodies. The primer sequences used are provided in Supplemental Table S2. Western blotting was performed as described previously (Wang et al., 2015a). Antibody information is provided in the Supplemental Information.

ICC, Flow Cytometry and Calcium Imaging

Reprogrammed iCMs, iPSCs and differentiated iPSC-CMs were stained for ICC and flow cytometry according to a previously described protocol (Wang et al., 2015a). Calcium signals were imaged with Rhod-3 Calcium Imaging Kit (Life Technologies) according to the manufacturer’s instructions. Additional details are available in the Supplemental Information.

Statistical Analysis

Values were presented as means ± SEM. The unpaired t-test was used to determine the significance of differences between two groups. A value of p < 0.05 was considered statistically significant (*), a p value of < 0.01 was considered highly significant (**). All data are representative of multiple repeated experiments.

Supplementary Material

1. Movie S1, related to Figures 1 and 4. Spontaneous contraction of iCMs and iPSC-CMs.

Sequentially showing spontaneously contractile cells generated from CF-derived iPSC1 and iPSC2 after cardiac differentiation in cardiac maturation medium for 13 days, spontaneously contractile cells generated from CFs via direct reprogramming in cardiac maturation medium for 28 days, and replated beating iPSC-CMs with or without treatment of MMC for 2 days at differentiation day 18.

Download video file (40.4MB, mp4)
2. Movie S2, related to Figure 3. Calcium oscillation of iCMs and iPSC-CMs.

Spontaneous calcium transients were labeled with Rhod-3 (red) in iCMs at reprogramming day 30 or in iPSC-CMs replated at differentiation day 12 as indicated.

Download video file (27.6MB, mp4)
3

Table S1, related to Figure 3. Differentially expressed lncRNAs between neoCFs, neoCMs, iCMs and iPSC-CMs

4

Highlights.

  • Gene expression profiling of iPSC-CM and iCM derived from CF of the same origin

  • Comparative analyses revealed distinct molecular features between iPSC-CM and iCM

  • iPSC-CM and iCM differ in gene signature of cell cycle, metabolism and chromatin status

  • iCM more closely resembles adult CM when compared to iPSC-CM

Acknowledgments

We are grateful for the expert technical assistance from the UNC Flow Cytometry Core, Microscopy Core and Genomics Core. We thank Dr. Rui Lu for the assistance of data analysis. This study was supported by NIH/NHLBI R00 HL109079 grant and Association (AHA) 15GRNT25530005 to J.L., AHA 13SDG17060010, the Ellison Medical Foundation (EMF) New Scholar Grant AG-NS-1064-13 and NIH/NHLBI R01HL128331 to L.Q.

Footnotes

AUTHOR CONTRIBUTIONS

Conceptualization, Y.Z. and L.Q.; Methodology, Y.Z., L.W. and L.Q.; Formal Analysis, Y.Z.; Investigation, Y.Z., L.W., Z.L., S.A. and C.Y.; Writing – Original Draft, Y.Z. and L.Q.; Writing – Review & Editing, Y.Z., J.L. and L.Q.; Visualization, Y.Z.; Supervision, L.Q.; Funding Acquisition, J.L. and L.Q.

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  1. Abad M, Hashimoto H, Zhou H, Morales MG, Chen B, Bassel-Duby R, Olson EN. Notch Inhibition Enhances Cardiac Reprogramming by Increasing MEF2C Transcriptional Activity. Stem Cell Reports. 2017;8:548–560. doi: 10.1016/j.stemcr.2017.01.025. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Addis RC, Ifkovits JL, Pinto F, Kellam LD, Esteso P, Rentschler S, Christoforou N, Epstein Ja, Gearhart JD. Optimization of direct fibroblast reprogramming to cardiomyocytes using calcium activity as a functional measure of success. J Mol Cell Cardiol. 2013;60:97–106. doi: 10.1016/j.yjmcc.2013.04.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Dal-Pra S, Hodgkinson CP, Mirotsou M, Kirste I, Dzau VJ. Demethylation of H3K27 Is Essential for the Induction of Direct Cardiac Reprogramming by miR Combo. Circ Res. 2017 doi: 10.1161/CIRCRESAHA.116.308741. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Fu JD, Stone NR, Liu L, Spencer CI, Qian L, Hayashi Y, Delgado-Olguin P, Ding S, Bruneau BG, Srivastava D. Direct reprogramming of human fibroblasts toward a cardiomyocyte-like state. Stem Cell Reports. 2013;1:235–247. doi: 10.1016/j.stemcr.2013.07.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Hänzelmann S, Castelo R, Guinney J. GSVA: gene set variation analysis for microarray and RNA-seq data. BMC Bioinformatics. 2013;14:7. doi: 10.1186/1471-2105-14-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Hirai H, Katoku-Kikyo N, Keirstead SA, Kikyo N. Accelerated direct reprogramming of fibroblasts into cardiomyocyte-like cells with the MyoD transactivation domain. Cardiovasc Res. 2013;100:105–113. doi: 10.1093/cvr/cvt167. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Ieda M, Fu JD, Delgado-Olguin P, Vedantham V, Hayashi Y, Bruneau BG, Srivastava D. Direct reprogramming of fibroblasts into functional cardiomyocytes by defined factors. Cell. 2010;142:375–386. doi: 10.1016/j.cell.2010.07.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Jain M, Brenner DA, Cui L, Lim CC, Wang B, Pimentel DR, Koh S, Sawyer DB, Leopold JA, Handy DE, et al. Glucose-6-Phosphate Dehydrogenase Modulates Cytosolic Redox Status and Contractile Phenotype in Adult Cardiomyocytes. Circ Res. 2003;93:9e–16. doi: 10.1161/01.RES.0000083489.83704.76. [DOI] [PubMed] [Google Scholar]
  9. Jayawardena TM, Egemnazarov B, Finch EA, Zhang L, Alan Payne J, Pandya K, Zhang Z, Rosenberg P, Mirotsou M, Dzau VJ. MicroRNA-mediated in vitro and in vivo direct reprogramming of cardiac fibroblasts to cardiomyocytes. Circ Res. 2012;110:1465–1473. doi: 10.1161/CIRCRESAHA.112.269035. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Kang MJ, Koh GY. Differential and dramatic changes of cyclin-dependent kinase activities in cardiomyocytes during the neonatal period. J Mol Cell Cardiol. 1997;29:1767–1777. doi: 10.1006/jmcc.1997.0450. [DOI] [PubMed] [Google Scholar]
  11. Kattman SJ, Witty AD, Gagliardi M, Dubois NC, Niapour M, Hotta A, Ellis J, Keller G. Stage-specific optimization of activin/nodal and BMP signaling promotes cardiac differentiation of mouse and human pluripotent stem cell lines. Cell Stem Cell. 2011;8:228–240. doi: 10.1016/j.stem.2010.12.008. [DOI] [PubMed] [Google Scholar]
  12. Laflamme MA, Murry CE. Heart regeneration. Nature. 2011;473:326–335. doi: 10.1038/nature10147. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Liu L, Lei I, Karatas H, Li Y, Wang L, Gnatovskiy L, Dou Y, Wang S, Qian L, Wang Z. Targeting Mll1 H3K4 methyltransferase activity to guide cardiac lineage specific reprogramming of fibroblasts. Cell Discov. 2016a;2:16036. doi: 10.1038/celldisc.2016.36. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Liu Z, Chen O, Zheng M, Wang L, Zhou Y, Yin C, Liu J, Qian L. Re-patterning of H3K27me3, H3K4me3 and DNA methylation during fibroblast conversion into induced cardiomyocytes. Stem Cell Res. 2016b;16:507–518. doi: 10.1016/j.scr.2016.02.037. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Lopaschuk GD, Jaswal JS. Energy metabolic phenotype of the cardiomyocyte during development, differentiation, and postnatal maturation. J Cardiovasc Pharmacol. 2010;56:130–140. doi: 10.1097/FJC.0b013e3181e74a14. [DOI] [PubMed] [Google Scholar]
  16. Lu R, Wang P, Parton T, Zhou Y, Chrysovergis K, Rockowitz S, Chen WY, Abdel-Wahab O, Wade PA, Zheng D, et al. Epigenetic Perturbations by Arg882-Mutated DNMT3A Potentiate Aberrant Stem Cell Gene-Expression Program and Acute Leukemia Development. Cancer Cell. 2016 doi: 10.1016/j.ccell.2016.05.008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Lundy SD, Zhu WZ, Regnier M, Laflamme MA. Structural and Functional Maturation of Cardiomyocytes Derived from Human Pluripotent Stem Cells. Stem Cells Dev. 2013;22:1991–2002. doi: 10.1089/scd.2012.0490. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Meshorer E, Yellajoshula D, George E, Scambler PJ, Brown DT, Misteli T. Hyperdynamic plasticity of chromatin proteins in pluripotent embryonic stem cells. Dev Cell. 2006;10:105–116. doi: 10.1016/j.devcel.2005.10.017. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Mohamed TMA, Stone NR, Berry EC, Radzinsky E, Huang Y, Pratt K, Ang Y-S, Yu P, Wang H, Tang S, et al. Chemical Enhancement of In Vitro and In Vivo Direct Cardiac Reprogramming. Circulation. 2016 doi: 10.1161/CIRCULATIONAHA.116.024692. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Muraoka N, Yamakawa H, Miyamoto K, Sadahiro T, Umei T, Isomi M, Nakashima H, Akiyama M, Wada R, Inagawa K, et al. MiR-133 promotes cardiac reprogramming by directly repressing Snai1 and silencing fibroblast signatures. EMBO J. 2014;33:1565–1581. doi: 10.15252/embj.201387605. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Nam Y, Song K, Luo X, Daniel E, Lambeth K, West K, Hill Ja. Reprogramming of human fi broblasts toward a cardiac fate. 2013 doi: 10.1073/pnas.1301019110. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Nam Y, Lubczyk C, Bhakta M, Zang T, Fernandez-perez A, Mcanally J, Bassel-duby R, Olson EN, Munshi NV. Induction of diverse cardiac cell types by reprogramming fibroblasts with cardiac transcription factors. Development. 2014:4267–4278. doi: 10.1242/dev.114025. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Qian L, Huang Y, Spencer CI, Foley A, Vedantham V, Liu L, Conway SJ, Fu J, Srivastava D. In vivo reprogramming of murine cardiac fibroblasts into induced cardiomyocytes. Nature. 2012;485:593–598. doi: 10.1038/nature11044. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Ritchie ME, Phipson B, Wu D, Hu Y, Law CW, Shi W, Smyth GK. Limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res. 2015;43:e47. doi: 10.1093/nar/gkv007. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Song K, Nam Y, Luo X, Qi X, Tan W, Huang GN, Acharya A, Smith CL, Tallquist MD, Neilson EG, et al. Heart repair by reprogramming non-myocytes with cardiac transcription factors. Nature. 2012;485:599–604. doi: 10.1038/nature11139. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Takahashi K, Yamanaka S. Induction of pluripotent stem cells from mouse embryonic and adult fibroblast cultures by defined factors. Cell. 2006;126:663–676. doi: 10.1016/j.cell.2006.07.024. [DOI] [PubMed] [Google Scholar]
  27. Takahashi K, Okita K, Nakagawa M, Yamanaka S. Induction of pluripotent stem cells from fibroblast cultures. Nat Protoc. 2007;2:3081–3089. doi: 10.1038/nprot.2007.418. [DOI] [PubMed] [Google Scholar]
  28. Uosaki H, Cahan P, Lee DI, Wang S, Miyamoto M, Fernandez L, Kass DA, Kwon C. Transcriptional Landscape of Cardiomyocyte Maturation. Cell Rep. 2015;13:1705–1716. doi: 10.1016/j.celrep.2015.10.032. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Wada R, Muraoka N, Inagawa K, Yamakawa H, Miyamoto K, Sadahiro T, Umei T, Kaneda R, Suzuki T, Kamiya K, et al. Induction of human cardiomyocyte-like cells from fibroblasts by defined factors. Proc Natl Acad Sci U S A. 2013;110:12667–12672. doi: 10.1073/pnas.1304053110. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Wang L, Liu Z, Yin C, Asfour H, Chen O, Li Y, Bursac N, Liu J, Qian L. Stoichiometry of Gata4, Mef2c, and Tbx5 influences the efficiency and quality of induced cardiac myocyte reprogramming. Circ Res. 2015a;116:237–244. doi: 10.1161/CIRCRESAHA.116.305547. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Wang L, Liu Z, Yin C, Zhou Y, Liu J, Qian L. Improved Generation of Induced Cardiomyocytes Using a Polycistronic Construct Expressing Optimal Ratio of Gata4, Mef2c and Tbx5. J Vis Exp. 2015b:e53426. doi: 10.3791/53426. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Xin M, Olson EN, Bassel-Duby R. Mending broken hearts: cardiac development as a basis for adult heart regeneration and repair. Nat Rev Mol Cell Biol. 2013;14:529–541. doi: 10.1038/nrm3619. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Yamakawa H, Muraoka N, Miyamoto K, Sadahiro T, Isomi M, Haginiwa S, Kojima H, Umei T, Akiyama M, Kuishi Y, et al. Fibroblast Growth Factors and Vascular Endothelial Growth Factor Promote Cardiac Reprogramming under Defined Conditions. 2015 doi: 10.1016/j.stemcr.2015.10.019. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Yang X, Pabon L, Murry CE. Engineering adolescence: Maturation of human pluripotent stem cell-derived cardiomyocytes. Circ Res. 2014;114:511–523. doi: 10.1161/CIRCRESAHA.114.300558. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Zhang J, Wilson GF, Soerens AG, Koonce CH, Yu J, Palecek SP, Thomson JA, Kamp TJ. Functional cardiomyocytes derived from human induced pluripotent stem cells. Circ Res. 2009;104 doi: 10.1161/CIRCRESAHA.108.192237. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Zhao Y, Londono P, Cao Y, Sharpe EJ, Proenza C, O’Rourke R, Jones KL, Jeong MY, Walker LA, Buttrick PM, et al. High-efficiency reprogramming of fibroblasts into cardiomyocytes requires suppression of pro-fibrotic signalling. Nat Commun. 2015;6:8243. doi: 10.1038/ncomms9243. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Zhou H, Dickson ME, Kim MS, Bassel-Duby R, Olson EN. Akt1/protein kinase B enhances transcriptional reprogramming of fibroblasts to functional cardiomyocytes. Proc Natl Acad Sci U S A. 2015 doi: 10.1073/pnas.1516237112. 201516237. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Zhou Y, Wang L, Vaseghi HR, Liu Z, Lu R, Alimohamadi S, Yin C, Fu JD, Wang GG, Liu J, et al. Bmi1 Is a Key Epigenetic Barrier to Direct Cardiac Reprogramming. Cell Stem Cell. 2016;18:382–395. doi: 10.1016/j.stem.2016.02.003. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1. Movie S1, related to Figures 1 and 4. Spontaneous contraction of iCMs and iPSC-CMs.

Sequentially showing spontaneously contractile cells generated from CF-derived iPSC1 and iPSC2 after cardiac differentiation in cardiac maturation medium for 13 days, spontaneously contractile cells generated from CFs via direct reprogramming in cardiac maturation medium for 28 days, and replated beating iPSC-CMs with or without treatment of MMC for 2 days at differentiation day 18.

Download video file (40.4MB, mp4)
2. Movie S2, related to Figure 3. Calcium oscillation of iCMs and iPSC-CMs.

Spontaneous calcium transients were labeled with Rhod-3 (red) in iCMs at reprogramming day 30 or in iPSC-CMs replated at differentiation day 12 as indicated.

Download video file (27.6MB, mp4)
3

Table S1, related to Figure 3. Differentially expressed lncRNAs between neoCFs, neoCMs, iCMs and iPSC-CMs

NIHMS904665-supplement-3.xlsx (489.5KB, xlsx)
4
NIHMS904665-supplement-4.pdf (1.2MB, pdf)

RESOURCES