Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2014 Jul 15;23(14):1704-16.
doi: 10.1089/scd.2013.0509. Epub 2014 Apr 22.

A Simple, Cost-Effective but Highly Efficient System for Deriving Ventricular Cardiomyocytes From Human Pluripotent Stem Cells

Free PMC article

A Simple, Cost-Effective but Highly Efficient System for Deriving Ventricular Cardiomyocytes From Human Pluripotent Stem Cells

Zhihui Weng et al. Stem Cells Dev. .
Free PMC article

Erratum in


Self-renewable human pluripotent stem cells (hPSCs) serve as a potential unlimited ex vivo source of human cardiomyocytes (CMs) for cell-based disease modeling and therapies. Although recent advances in directed differentiation protocols have enabled more efficient derivation of hPSC-derived CMs with an efficiency of ∼50%-80% CMs and a final yield of ∼1-20 CMs per starting undifferentiated hPSC, these protocols are often not readily transferrable across lines without first optimizing multiple parameters. Further, the resultant populations are undefined for chamber specificity or heterogeneous containing mixtures of atrial, ventricular (V), and pacemaker derivatives. Here we report a highly cost-effective and reproducibly efficient system for deriving hPSC-ventricular cardiomyocytes (VCMs) from all five human embryonic stem cell (HES2, H7, and H9) and human induced PSC (hiPSC) (reprogrammed from human adult peripheral blood CD34(+) cells using nonintegrating episomal vectors) lines tested. Cardiogenic embryoid bodies could be formed by the sequential addition of BMP4, Rho kinase inhibitor, activin-A, and IWR-1. Spontaneously contracting clusters appeared as early as day 8. At day 16, up to 95% of cells were cTnT(+). Of which, 93%, 94%, 100%, 92%, and 92% of cardiac derivatives from HES2, H7, H9, and two iPSC lines, respectively, were VCMs as gauged by signature ventricular action potential and ionic currents (INa(+)/ICa,L(+)/IKr(+)/IKATP(+)); Ca(2+) transients showed positive chronotropic responses to β-adrenergic stimulation. Our simple, cost-effective protocol required the least amounts of reagents and time compared with others. While the purity and percentage of PSC-VCMs were comparable to a recently published protocol, the present yield and efficiency with a final output of up to 70 hPSC-VCMs per hPSC was up to 5-fold higher and without the need of performing line-specific optimization. These differences were discussed. The results may lead to mass production of hPSC-VCMs in bioreactors.


<b>FIG. 1.</b>
FIG. 1.
A highly efficient ventricular cardiomyocyte (VCM) differentiation system of human pluripotent stem cells (hPSCs). (A) Schematic of a direct differentiation protocol used for the differentiation of hPSCs to VCMs under non-feeder and non-xenogeneic conditions. Spontaneously contracting clusters were detected as early as day 8. (B) Images demonstrate typical culture morphology at different time points of differentiation. Scale bar=200 μm. Color images available online at
<b>FIG. 2.</b>
FIG. 2.
Quantitative PCR-based gene expression analysis of embryoid bodies (EBs) at different stages of cardiac differentiation. HES2 were subjected to the cardiac differentiation protocol described in Figure 1. Expression levels of pluripotent, cardiac mesoderm markers, and cardiac specific genes were normalized and expressed relative to housekeeping gene GAPDH. Plots shown are representative of three independent experiments.
<b>FIG. 3.</b>
FIG. 3.
Specification of cardiac differentiation in multiple hPSC lines. (A) Representative flow cytometry plots of cardiac troponin T (cTNT)+ cells differentiated from hPSCs. The entire well of one 6-well plate was enzymatically digested into a single-cell suspension and 100,000 cells were analyzed. (B) Percentages of cTnT+ cells in day 16–18 EBs of HES2, H7, H9DF, and two nonviral CD34+-derived human induced PSC (hiPSC) lines. Mean±SEM.
<b>FIG. 4.</b>
FIG. 4.
Structural characterization of HES2-cardiomyocyte (CM) by immunostaining. (A) HES2-EBs were digested into single cells at day 20 of differentiation and then costained for MLC2v/MF20, MLC2a/α-actinin, and connexin 43/α-actinin. Nuclei were counterstained with DAPI (blue). Representative immunostaining images for MLC2a and MLC2v in HES2-CMs examined at (B) day 20 and (C) day 40 postdifferentiation. Color images available online at
<b>FIG. 5.</b>
FIG. 5.
Specification of ventricular differentiation of hPSCs. (A) Representative action potential (AP) profiles of HES2-CMs, H7-CMs, H9DF-CMs, iPSC-PB1-CMs, and iPSC-PB2-CMs measured by patch-clamp recordings. (B) Immunostaining of nonselected HES2-CMs for MLC2v and cTnT expression. HES2-EBs were digested into single cells and then reseeded for immunostaining without any purification. Scale bar=200 μm. (C) Representative Ca2+ transient tracings of HES2-VCMs and iPSC-PB1-VCMs paced by electrical field stimulation (1 Hz, 40 V/cm). All cells examined were >35 days old. (D) Percentages of cells expressing VCM and atrial cardiomyocyte (ACM) AP profiles as detected by patch-clamping. Color images available online at
<b>FIG. 6.</b>
FIG. 6.
Ionic currents (INa, ICaL, IKr, and IKATP) in HES2-CMs. Representative (A) INa and (B) ICaL traces elicited by voltage protocol shown in inset. Middle: peak I–V plots. Right: steady-state inactivation and activation relations. (n=7). (C) Representative IKr traces after the subtraction of E4031-insensitive current elicited by voltage protocol shown in inset. (D) Activation relation of Ikr (n=11). (E) Sarcolemmal IKATP channels in HES2-VCMs. Left panel: representative tracings of sarcolemmal IKATP in HES2-VCMs at 0 mV under control conditions (blue line), with sodium cyanide (CN, 2 mM) alone (black line), or with CN and glibenclamide (GLI, 10 μM; red line). Right panel: summary of averaged current densities under the same conditions. Cells were stimulated to 0 mV for 1,000 ms from a holding potential of −80 mV preceded by a 100-ms prepulse to −10 mV (n=5). **P<0.01 compared with control group; ##P<0.01 compared with CN group. Color images available online at
<b>FIG. 7.</b>
FIG. 7.
Immature mitochondrial structure of HES2-VCMs. (A) Lentivirus (LV)-MLC2v-Tdtomato-T2A-Zeo–transduced and Zeo-selected HES2-VCMs (30–50 days old) were immunostained for α-actinin and mitochondrial cytochrome c oxidase (COX)IV. Nuclei were counterstained with DAPI (blue). Scale bar=25 μm. (B) HES2-VCMs were stained using the potential-sensitive JC-1 dye. Left panel shows representative image of control cells with both green and red fluorescence in the HES2-VCMs. Right panel shows representative image of HES2-VCMs subjected to 100 μM hydrogen peroxide (H2O2) (n=3). Scale bar=100 μm. Color images available online at
<b>FIG. 8.</b>
FIG. 8.
In vivo survival of H9-VCMs under kidney capsule transplantation. (A) H9DF was used to differentiate VCMs. Differentiation efficiency was estimated based on cTnT protein expression by flow cytometry. Representative FACS plot is shown. (B) In vivo bioluminescence imaging (BLI) signal measured from animals in which H9DF-VCMs were transplanted into the kidney capsule. (C) A representative animal imaged following transplantation of 2×105 H9DF-VCMs into the kidney capsule. Color scale bar value represents photons/sec/cm2/steradian (p/s/cm2/sr). (D) Immunohistochemistry for cTNT-positive cells in the transplanted kidneys harvested on d28. Scale bar=100 μm. Color images available online at

Similar articles

See all similar articles

Cited by 36 articles

See all "Cited by" articles

Publication types

MeSH terms


LinkOut - more resources