Dierentiation (2006) 74:149159 DOI: 10.1111/j.1432-0436.2006.00063.

r 2006, Copyright the Authors

Journal compilation r 2006, Interntaional Society of Differentiation

O RI G INA L AR T I C L E

Byung Sun Yoon . Seung Jun Yoo . Jeoung Eun Lee .

Seungkwon You . Hoon Taek Lee . Hyun Soo Yoon

Enhanced differentiation of human embryonic stem cells into

cardiomyocytes by combining hanging drop culture and
5-azacytidine treatment

Received November 1, 2005; accepted in revised form January 23, 2006

Abstract
Cell replacement therapy is a promising approach for the treatment of cardiac diseases. It is, however, challenged by a limited supply of appropriate cells.
Therefore, we have investigated whether functional cardiomyocytes can be efciently generated from human embryonic stem cells (hESCs). In this study, we developed
an efcient protocol for the generation of functional
cardiomyocytes from hESCs by combining hanging drop
culture and 5-azacytidine, a well-known demethylating
agent, and then evaluated the expression of cardiac-specific markers. hESCs were cultured both in the medium
without or with 0.1, 1, or 10 mM of 5-azacytidine under a
hanging drop culture. The expression of several cardiacspecific markers was determined by real-time PCR, RTPCR, immunouorescence, and confocal microscopy. To
verify the structural and functional properties of hESCderived cardiomyocytes, we performed electron micros-

Key words human embryonic stem cell

cardiomyocyte
5-azacytidine
hanging drop culture

Byung Sun Yoon
Hoon Taek Lee

Department of Animal Science
Konkuk University
1 Hwayang-dong, Gwangjin-gu
Seoul 143-701, Korea

Introduction

Seung Jun Yoo
Seungkwon You

Division of Biotechnology and Genetic Engineering
College of Life and Environmental Sciences
Korea University
Seoul, Korea
. )
Jeoung Eun Lee
Hyun Soo Yoon (*
Department of Anatomy & Cell Biology
College of Medicine
Hanyang University
17 Haengdang-dong, Seongdong-gu
Seoul 133-792, Korea
Tel: 182-2-2220-0602
Fax: 182-2-2281-7841
E-mail: hsyoon@hanyang.ac.kr
U.S. Copyright Clearance Center Code Statement:

copy and electrophysiological recording. The efciency

of beating cell generation was significantly improved in
the hanging drop culture compared with that in suspension culture. Treatment of hESCs with 0.1 mM of 5azacytidine for 13 days significantly increased the
number of beating cells and simultaneously enhanced
the expression of cardiac-specific markers. Transmission
electron microscopy and electrophysiological recording
showed that hESC-derived cardiomyocytes acquired
structural and functional properties of cardiomyocytes.
In conclusion, these results suggest that differentiation of
hESCs into cardiomyocytes can be enhanced by the
combination of hanging drop culture and 5-azacytidine
treatment. Also the methylation status of genes related to
cardiomyocyte development may play an important role
in the differentiation of hESCs into cardiomyocytes.

Human embryonic stem cells (hESCs) have been successfully derived from the inner cell mass (ICM) of
blastocysts and can be maintained in vitro for prolonged
periods (Thomson et al., 1998; Reubino et al., 2000;
Baharvand et al., 2004). Embryonic stem (ES) cells can
form embryoid bodies (EBs, strictly endoderm surrounding an inner core of ectoderm) that can give rise
to cells found in all three embryonic germ layers
(ectoderm, mesoderm, and endoderm), which are embryonic sources of all the cells in the body (ItskovitzEldor et al., 2000). A number of studies suggest that
hESCs are pluripotentthey can differentiate into a
variety of cell types, including pancreatic b-cells, neural
cells, cardiomyocytes, and blood cells. These features of

03014681/2006/7404149 $ 15.00/0

150

hESCs suggest that they have the potential to provide

an unlimited supply of various cell types for cell replacement therapy.
The pluripotency of hESCs has the great advantage
that it could be used in cell replacement therapy for
cardiac failure as well as for the study on the mechanisms of heart development in vitro (Klug et al., 1996;
Johkura et al., 2003). Cardiomyocytes have also
been successfully derived from various stem cells, such
as bone marrow stem cells (Orlic et al., 2001),
hematopoietic stem cells (Jackson et al., 2001), and embryonic carcinoma cells (Rudnicki et al., 1990).
Recent studies have suggested that hESCs can form
EBs in vitro, some of which initiate spontaneous beating
(Kehat et al., 2001, 2002). Based on the expression
of cardiac-specific genes, extracellular electrical
activity, and cellular ultrastructure, these beating cells
contain cardiomyocytes (Kehat et al., 2001; Snir et al.,
2003). The potential of hESCs to differentiate into
cardiomyocytes was rst reported by Kehat et al.
(2001), followed by other groups (He et al., 2003;
Snir et al., 2003). This differentiation of hESCs
into cardiomyocytes is aided by co-culture with
END2 visceral endoderm-like cells (Mummery et al.,
2003).
Several groups have reported that 5-azacytidine, a
demethylating agent, induced the differentiation of
mesenchymal stem cells into cardiomyocytes in vitro
(Makino et al., 1999; Hakuno et al., 2002; Fukuda,
2003). Xu et al. (2002) reported that 5-azacytidine
induced the differentiation of hESCs into cardiomyocytes. This compound can cause extensive demethylation of 5-methylcytosine and reduce DNA
methyltransferase activity in the cell (Haaf and Schmid, 2000). Recently, 5-azacytidine was reported to reverse the differentiation status of EBs back to ES cells
(Tsuji-Takayama et al., 2004). 5-Azacytidine has been
useful for studying the roles of DNA methylation in the
mechanisms of gene activation and cell differentiation.
In the present study, therefore, we developed the
protocol for the enhanced differentiation of hESCs into
functional cardiomyocytes. We also evaluated the effect
of hanging drop culture and 5-azacytidine on the differentiation of hESCs into cardiomyocytes as well as the
gene expression of cardiac-specific markers.

Materials and methods

Culture of hESCs
Two hESC lines (Miz-hES2 and HSF-6) were maintained on mouse
embryonic broblasts (MEFs) established from day-13.5-postcoitum fetuses of CF1 mice. Feeder cells were cultured in
Dulbeccos modied Eagles medium (DMEM) (high glucose, Invitrogen, Carlsbad, CA), supplemented with 10% FBS (HyClone,
Logan, UT), 0.1 mM b-mercaptoethanol, and 0.1 mM non-essential
amino acids. They were treated with 10 mg/ml mitomycin C (Sigma,

St. Louis, MO) for 1.5 hr to arrest mitosis and replated at a concentration of 6.1  104 cells/well in gelatin-coated 4-well culture
dishes. The hESCs were cultured in DMEM/F12 (without pyruvate) (DMEM/F12), supplemented with 20% knock-out serum replacement (SR) (Gibco/BRL, Invitrogen, Carlsbad, CA), 0.1 mM
b-mercaptoethanol, 1% non-essential amino acids (Gibco/BRL),
100 U/ml penicillin G, 100 mg/ml streptomycin, and 4 ng/ml human
recombinant basic broblast growth factor (Invitrogen). The hESC
colonies were transferred onto newly prepared CF1 feeder layers
and mechanically dissociated using a micropipette every 57 days.
We used hESCs with passage numbers between 50 and 70.

Differentiation of hESCs into cardiomyocytes

To induce differentiation of cardiomyocytes, hESCs were mechanically dissociated into small clumps. EBs were formed using EB
medium, which consisted of DMEM/F12 with 20% fetal bovine
serum (FBS), 1 mM glutamine, 0.1 mM b-mercaptoethanol, and
1% non-essential amino acids, and were cultured by using a hanging drop method (Maltsev et al., 1994). The effect of 5-azacytidine
(Sigma) on the differentiation of hESCs was examined by either
suspension culture or hanging drop culture following dierentiation for 13 days. After 3 days of differentiation in the hanging
drop or suspension culture, EBs were plated on gelatin (0.1%, Sigma)-coated 4-well plates (Nunc, Roskilde, Denmark) with four to
six EBs per well. Daily microscopic observations were conducted to
detect beating EBs and to determine the beating rate. Contracting
areas were mechanically dissected using a micropipette and were
continuously subcultured.

Reverse transcription (RT)-polymerase chain reaction (PCR)

Total RNAs were prepared by using Trizol reagent (Invitrogen).
Standard RT was performed using 500 ng of total RNA, oligo
d(T)1218 primer (Invitrogen), and AMV reverse transcriptase
(Roche Molecular Biochemicals, Nutley, NJ). The RT-PCR was
carried out with 1 ml of cDNA template, 10 pmol of primers, and
PCR premix (1 U Tag DNA polymerase, 250 mM dNTPs, 10 mM
Tris-HCl, 40 mM KCl, and 1.5 mM MgCl2; Bioneer, Korea). Primer sequences for PCR are shown in Table 1. All primer sets had a
calculated annealing temperature of 621C. The PCR was carried
out in a GeneAmp 9600 (Perkin Elmer, Boston, MA) using the
following: a 5 min denaturation at 941C; 30 cycles of 941C for
30 sec, 621C for 30 sec and 721C for 30 sec; and a nal extension for
10 min at 721C.

Real-time quantitative PCR

Real-time PCR using the iCycler Optical Detection System (BioRad) was carried out in a nal reaction volume of 25 ml with SYBR
Green (Bio-Rad, Hercules, CA). Briefly, 500 ng of total RNA was
transcribed to cDNA. One microliter of cDNA template was then
added to 12.5 ml of the 2  SYBR green PCR master mix and
10 pmol of each primer. The temperature proles were the same as
those for RT-PCR with the exception of the annealing temperature,
which was 651C. The thermal denaturation protocol was run at the
end of the PCR to determine the amplication of the specific
products. The cycle number at which the reaction crossed an arbitrarily placed threshold (Ct) was determined for each gene. Data
were analyzed by using the 2DDCt method to obtain the relative
expression level and by using that of b-actin as a normalization
control in each sample. The relative amount of target 5 2DDCt ,
where Ct is the threshold cycle for target amplication,
DCt Cttarget gene  Ctinternal reference , and DDCt DCtsample  DCtcalibrator .

151
Table 1 Primer sequence information for RT-PCR
Gene

Accession no.

Primer sequence (5 0 ! 3 0 )

Annealing
temperature (1C)

Product
size (bp)

b-actin

BC002409

62

260

GATA-4

NM_002052

62

422

Nkx2.5

AB021133

62

322

MLC-2A

BC027915

62

239

MLC-2V

BC031006

62

200

a-MHC

NM_002471

62

413

b-MHC

X06976

62

396

ANP

NM_006172

62

406

cTnT

BC002653

62

152

cTnI

X54163

62

218

bIII-tubulin

AF427491

62

246

MAP2

BC038857

62

256

Oct-4

NM_002701

62

455

Nanog

NM_024865

Forward: AGCAAGCAGGAGTATGACGA
Reverse: TGTGAACTTTGGGGGATG
Forward: CTCCCCTGGCAAAACAAGAG
Reverse: TGCCGTGTCTTAGCAGTCGT
Forward: GGTCTATGAACTGGAGCGGC
Reverse: ATAGGCGGGGTAGGCGTTAT
Forward: GCTCTTTGGGGAGAAGCTCA
Reverse: CGTCTCCATGGGTGATGATG
Forward: GGCGCGTGAACGTGAAAAAT
Reverse: CAGCATTTCCCGAACGTAAT
Forward: GTCATTGCTGAAACCGAGAATG
Reverse: GCAAAGTACTGGATGACACGCT
Forward: AGATGGATGCTGACCTGTCC
Reverse: GGTTTTTCCTGTCCTCCTCC
Forward: GAACCAGAGGGGAGAGACAGAG
Reverse: CCCTCAGCTTGCTTTTTAGGAG
Forward: GGCAGCGGAAGAGGATGCTGAA
Reverse: GAGGCACCAAGTTGGGCATGAACGA
Forward: CCTGCGGAGAGTGAGGATCT
Reverse: TAGGCAGGAAGGCTCAGCTC
Forward: CTCAAGATGTCCTCCACCTTCAT
Reverse: CTCCTCGTCGTCTTCGTACATCT
Forward: TAGCAGTCCTGAAAGGTGAACAAG
Reverse: GACTTTGTGCTACCCTGTAAAGCA
Forward: AAGAACATGTGTAAGCTGCGGCCC
Reverse: GGAAAGGCTTCCCCCTCAGGGAAAGG
Forward: ATAGCAATGGTGTGACGCAG
Reverse: GATTGTTCCAGGATTGGGTG

62

219

a-MHC, a-myosin heavy chain; b-MHC, b-myosin heavy chain; ANP, atrial natriuretic peptide; cTnT, cardiac troponin T; cTnI, cardiac
troponin I.

Immunouorescence and confocal microscopy

Beating areas were mechanically dissected using a micropipette.
These beating cells were then enzymatically dispersed using
0.05% trypsin-EDTA for 15 min at 371C. The hESC-derived cardiomyocytes and enzymatically dispersed cells were xed for 30 min
at room temperature in phosphate-buered saline (PBS) with Ca21
and Mg21 containing 4% paraformaldehyde and then permeabilized for 1 hr at room temperature with PBS containing 0.1% Triton
X-100. The cells were then blocked with 3% bovine serum albumin
and incubated with the corresponding primary antibodies for 1 hr
at room temperature. Primary antibodies (1:100 dilutions) included
anti-rabbit antibody against human GATA-4, anti-goat antibody
against human cardiac myosin light chain, and anti-goat antibody
against human atrial natriuretic peptide (ANP), anti-mouse antibody against human desmin (Santa Cruz Biotechnology, Santa
Cruz, CA) and anti-mouse antibody against human a-actinin
(Sigma). After washing three times with PBS, cells were exposed for
45 min at room temperature to the corresponding uorescent
isothiocyanate- or rhodamine-conjugated secondary antibodies at a
dilution of 1:200 and then observed under uorescence and confocal microscopes.

Transmission electron microscopy (TEM)

For TEM, the beating cells were mechanically dissected and xed
for 3 hr in 2.5% glutaraldehyde, post-xed in PBS containing 1%
osmium tetroxide, dehydrated in ascending concentrations of ethanol, and embedded in Epon 812. Semi-thin sections were obtained using a glass knife and thin sections with a diamond knife on

an ultramicrotome (LKB, Kent City, MI). Semi-thin sections were

placed on glass slides, stained with 1% toluidine blue, and examined under a light microscope. Thin sections were placed on copper
grids, stained for 10 min with 2% uranyl acetate and lead citrate,
and examined under an H-7600 transmission electron microscope
(Hitachi, Tokyo, Japan).

Electrophysiological recording
Cells were isolated from beating hESC-derived cardiomyocytes by
treating them with 0.05% trypsin-EDTA for 15 min at 371C. After
dissociation, cells were replated on glass coverslips for 57 days.
Electrical activities of dissociated single cells were measured at 371C
by whole-cell patch-clamp techniques. In the voltage-clamp mode,
whole-cell currents were recorded by the application of a family of
voltage steps (from  90 to 40 mV with a series of 10 mV increments) for 40 ms. The initial holding potential was  70 mV. In the
current-clamp mode, the action potentials were elicited by a series
of current injections (from 0.1 to 3.10 nA in 0.5 nA increments) for
9 ms. The initial current was held at  0.15 nA to generate a membrane potential of approximately  90 mV. The bath solution consisted of 147 mM NaCl, 5 mM KCl, 10 mM HEPES, 1.5 mM CaCl2,
1 mM MgCl2, and 5 mM glucose (pH 7.4). The internal pipette
solution contained 155 mM KCl, 100 mM aspartic acid, 2 mM
MgCl2, 7 mM Na2ATP, 10 mM EGTA, 20 mM HEPES, and
10 mM phosphocreatine (pH 7.2). Patch pipettes resistances were
58 MO. The sampling rates were 10 kHz (voltage-clamp mode) and
3.3 kHz (current-clamp mode). The data were acquired using an
Axopatch 1D patch, a Digidata 1200B interface, and pCLAMP 6
software (Axon instruments, Union City, CA).

152
Statistical analysis
A w2-test was used for statistical analyses, and, in all cases, a value
with a probability of po0.05 was considered as an indication of
statistical significance. Results were expressed as the mean  standard error of the mean (SEM).

Results
Effects of hanging drop culture and 5-azacytidine
treatment on the differentiation of hESCs into
cardiomyocytes
The EBs were formed by dissociating hESC colonies
(Fig. 1A) and by growing them in a hanging drop culture (Fig. 1B). Differentiated cardiomyocytes were cultured up to 4050 days (Figs. 1C, 1D, respectively). We
observed that efciency of beating cell generation was
higher in hanging drop culture than in a suspension
culture. We then used a demethylating agent, 5-azacytidine, which has been shown to induce the dierentiation of bone marrow-derived mesenchymal stem cells
and hESCs into cardiomyocytes (Makino et al., 1999;
Hakuno et al., 2002; Xu et al., 2002; Fukuda, 2003).
Differentiation of hESCs (Miz-hES2 and HSF-6) into
cardiomyocytes was enhanced by the combination of
hanging drop culture and 5-azacytidine treatment (Fig.
2A). The effect of 5-azacytidine on hESC differentiation
into cardiomyocytes was examined by treating EBs
either in suspension or in hanging drop cultures for 3
days with different doses (0.1, 1, and 10 mM). We found
that beating cells appeared following 4 days of dierentiation and remained viable for several months in
culture. Treatment with 0.1 mM of 5-azacytidine for 13

Fig. 1 Differentiation of human embryonic stem cells (hESCs) into

cardiomyocytes. (A) Undifferentiated hESCs. (B) Embryoid bodies
(EBs) after 3 days of hanging drop culture. (C) Differentiated
cardiomyocytes from hESCs on day 40. (D) Expansion of the
beating area in EBs from hESCs at day 50. Scale bars 5 100 mm.

days significantly increased the rate of differentiation

into cardiomyocytes (Miz-hES2: 41.88% for treated
cells versus 22.37% for untreated control cells, HSF-6:
34.89% for treated group versus 18.55% for control
group). On the other hand, high doses of 5-azacytidine
(1 and 10 mM) decreased the rate of differentiation into
cardiomyocytes (Table 2). In HSF-6 cell line, the cells
were all lysed with the treatment of 10 mM 5-azacytidine. The generation of beating cells in the EBs generally increased with time in culture and treatment with
0.1 mM of 5-azacytidine significantly increased the percentage of beating cells when they were in the tissue
culture plates (Fig. 2B).

Expression of cardiomyocyte-specific markers from

hESC-derived cardiomyocytes
The expression of several cardiac-specific markers was
evaluated by RT-PCR in hESC-derived cardiomyocytes, undifferentiated hESCs, EBs after 3 days
of differentiation, and differentiated non-beating cells.
Complementary DNA generated from commercially
available human fetal heart RNA was used as a positive
control. As shown in Fig. 3A, hESC-derived cardiomyocytes expressed the cardiac transcription factors
GATA-4 and Nkx2.5, as well as the cardiac-specific
genes; a-myosin heavy chain (a-MHC), b-MHC, cardiac troponin T (cTnT), cardiac troponin I (cTnI), ANP,
myosin light chain-2A (MLC-2A), and MLC-2V. The
expression of cardiac-specific genes was, however, barely detectable in the control group (undifferentiated
hESCs, EBs and non-beating cells). Furthermore, we
found cTnI expression, which was used as a cardiacspecific markers (Kehat et al., 2004), in the undierentiated hESCs and non-beating cells (non-cardiomyocytes), as well as cardiomyocytes. To clarify its
expression, the amplied cTnI RT-PCR product was
sequenced and was veried as human cTnI (data not
shown). Undifferentiated markers, such as Oct-4 and
Nanog, and neuronal markers, MAP2 and bIII-tubulin,
were not expressed in hESC-derived cardiomyocytes either. Immunouorescence and confocal microscopy
were performed to conrm the presence of cardiac-specific proteins in beating cells. Figure 3B shows that
hESC-derived cardiomyocytes were positively immunostained with anti-ANP (a1), anti-desmin (b), anti-MLC
(c), and anti-a-actinin (d). We found that these cardiacspecific proteins were localized only in the beating regions. Most of the differentiated cells expressed markers
of cardiomyocytes. The hESC-derived cardiomyocytes
also specifically expressed the cardiac transcription
factor, GATA-4 (Fig. 3Ba2). Additionally, a-actinin
and GATA-4 expression was also detected in single cells
isolated from beating EBs (Figs. 3Be, f1, f3). Undierentiated hESCs were used as a negative control and they
did not react with cardiac muscle-specific antibodies

153

Fig. 2 Enhanced differentiation of human embryonic stem cells

(hESCs) into cardiomyocytes. (A) Efciency of beating embryoid
body (EB) generation in the presence or absence of 5-azacytidine in
a hanging drop culture and in a suspension culture. denotes statistical significance at po0.05. denotes statistical significance at

po0.01. (B) Cumulative percentage of cardiomyocytes derived

from hESCs with various doses of 5-azacytidine. Days of plating
indicates the number of days that the cells were cultured on a tissue
culture plate after the cells had been taken from the hanging drop
culture. Left panel: Miz-hES2; right panel: HSF-6.

(Fig. 3Bg1). As an additional negative control for antibody specicity, we incubated cells with only secondary antibodies to monitor the non-specific binding of
secondary antibodies to the cells (Fig. 3Bh1).

hESC-derived cardiomyocytes showed the typical striation pattern of the sarcomeres (Figs. 4A4D).
Notably, the hESC-derived cardiomyocytes contained
Z-band (arrow, Z), which is a major characteristic of
cardiac muscles containing fascia adherens (arrow, FA),
desmosomes (arrow, D), gap junctions (arrow, G) (Fig.
4Aa), and granules of ANP (A) (Fig. 4Ad).

Ultrastructural analysis of hESC-derived

cardiomyocytes
TEM was performed to visualize the ultrastructure of
hESC-derived cardiomyocytes. TEM revealed that
Table 2 Efciency of beating EB generation at various doses of 5azacytidine
Cell lines

5-azacytidine
(mM)

Number of
EB

Number of beating
(%) (mean  SEM)

Miz-hES2

0
0.1
1
10

206
217
201
168

45 (22.37
91 (41.88
30 (15.25
10 (5.44






1.15)
0.92)
1.34)
0.55)

HSF-6

0
0.1
1
10

199
192
192
155

36 (18.55
67 (34.89
14 (7.29
0 (0






0.59)
1.29)
0.66)
0)

po0.05 by w2-test. (n 5 6).

EB, embryoid body; SEM, standard error of the mean.

Electrophysiology of hESC-derived cardiomyocytes

To further determine whether hESC-derived cardiomyocytes possess the electrical properties of mature
cardiomyocytes, we performed electrophysiological
studies using a whole-cell patch-clamp technique.
All electrical activities were obtained from dissociated
single cells. Whole-cell currents elicited by the application of voltage steps showed active voltage-dependent
responses (Fig. 4Ba). Voltage dependence of whole-cell
currents is common to all excitable cells including
cardiac muscle cells. Typical inward sodium currents
were evident, and outward currents were also detected,
such as A-type and delayed potassium channels,
which suggest the existence of voltage-gated inward
and outward channels (data not shown). The activities
of these voltage-gated sodium and potassium
channels were more evident when the action potentials

154

Fig. 3 Characterization of cardiomyocytes

derived from human embryonic stem cells
(hESCs). (A) Expression of cardiac-specific markers in the hESC-derived cardiomyocytes. Lane 1, undifferentiated hESCs;
lane 2, embryoid bodies (EBs) at day 3;
lane 3, non-beating cells; lane 4, beating cells; lane 5, human fetal heart as a
positive control; lane 6, negative control
(no RT). (B) Confocal microscopy for
cardiac-specific markers in hESC-derived
cardiomyocytes. The hESC-derived cardiomyocytes (a) were stained with antibodies against atrial natriuretic peptide (ANP)
(a1) and GATA-4 (a2). The hESC-derived
cardiomyocytes stained with anti-desmin
(b), anti-myosin light chain (c), and antia-actinin antibodies (d). Single cells were
isolated from beating cells and were
stained with anti-a-actinin (e) and antiGATA-4 antibodies (f1, f3). (g) Undierentiated hESCs. (g1) Undifferentiated hESCs stained with anti-ANP antibody. As a
negative control for antibody specicity,
the hESC-derived cardiomyocytes were
stained only with secondary antibodies
(h1). Scale bars 5 100 mm (e, f2 5 25 mm).

were evoked by current injections. Currents that

were sufciently strong to overcome a threshold potential (around  45 mV) generated the action potentials
(Fig. 4Bb). Furthermore, the complete depletion of
the action potential by the application of tetrodotoxin,
which is a sodium channel blocker, in the bath solution
indicates that voltage-gated sodium channels are responsible for the rising phase of the action
potential (Fig. 4Bc). Therefore, it appears that hESCderived cardiomyocytes have at least some functional electrical properties of the differentiated cardiomyocytes.

Real-time quantitative PCR

Real-time PCR experiments showed that 5-azacytidine
up-regulated the expression of cardiac-specific markers.
The highest expression of the cardiomyocyte transcription factors GATA-4 and Nkx2.5 was detected in the
group treated with 1 mM of 5-azacytidine: the cardiacspecific markers were most highly expressed in the
0.1 mM-treated groups (Table 3: Miz-hES2). The higher
percentage of beating cells in the 0.1 mM-treated group
was correlated with the higher expression of cardiacspecific markers rather than that of the cardiac tran-

155

Fig. 4 Ultrastructural and electrophysiological analyses of human

embryonic stem cell (hESC)-derived cardiomyocytes on day 70. (A)
Transmission electron microscopy (TEM) image of cardiomyocytes. (a) TEM image showing the presence of a desmosome
(D), gap junction (G), and fascia adherens (FA). (bd). A, granule
of atrial natriuretic peptide; N, nucleus; Z, Z-bands; S, sarcomere.
Scale bars 5 0.5 mm (panels a, c, and d) or 10 mm (panel b). (B)
Representative whole-cell currents and evoked action potentials
obtained from a single hESC-derived cardiomyocyte. (a) Whole-cell

current traces were measured by the application of a series of voltage steps. Active voltage responses and inward sodium currents are
evident by depolarizing potential steps. (b) The action potentials
were evoked by a series of current injections. Only currents enough
to overcome the threshold potential (approximately  45 mV in
this gure) can generate the action potentials. (c). The action potential was blocked by the application of tetrodotoxin (200 nM) in
bath solution. An action potential trace from B(b) was superimposed for comparison.

scription factors, compared with that of normalized

control with no 5 azacytidine treatment group. In HSF6 cell line, the higher expression of both cardiac transcription factors and cardiac-specific markers were,
however, detected in the 0.1 mM-treated group. During
cardiac development, the expression of cardiac transcription factors and cardiac-specific markers increased
continuously until 15 days of differentiation and decreased slightly thereafter, compared with that of
normalized control with day 5 differentiation group
(Table 4.).

transplantation (Kehat et al., 2004). When a typical

myocardial infarction occurs, more than 25% of the
cells in the heart are lost, which leads to heart failure.
For example, the left ventricle of the human heart contains approximately 5.8  109 myocytes (Kajstura et al.,
1998). Therefore, a large-scale culture is needed to supply enough cells for clinical applications. Generation
of a number of cells can be achieved with an efcient,
directed-differentiation system coupled with an enrichment method.
Kehat et al. (2001) rst reported that hESCs could
be differentiated into cardiomyocytes with 8.1%
efciency. Xu et al. (2002) reported that the addition
of 5-aza-2 0 -deoxycytidine (5-aza-dC) to developing EBs
at later days of differentiation (68 days) has been
shown to be more effective than the addition at earlier
days (15 days). It was also found that the 5-aza-dCtreated beating unit contained 44  2% cTnI-positive
cells, which were not clearly associated with the
functional cardiomyocytes (Messner et al., 2000; Xu
et al., 2002).
In the present study, we sought to determine an
efcient protocol for the generation of functional

Discussion
The generation of cardiomyocytes from hESCs has several potential applications, including transplantation
for curing heart failure. Cardiomyocytes derived from
hESCs have been successfully transplanted to generate
pacemaker cells in a swine model of atrioventricular
block. In this case, the source of the new ventricular
ectopic rhythm was conrmed to be the site of cell

EBs formed from a hanging drop culture were treated with various doses of 5-azacytidine (mM). Expressions of the various mRNAs were normalized by the expression of b-actin.
Denotes statistical significance when po0.05 by w2-test.
(po0.01). (n 5 3).
EB, embryoid body; hESC, human embryonic stem cells; a-MHC, a-myosin heavy chain; b-MHC, b-myosin heavy chain; MLC-2V, myosin light chain-2V; ANP, atrial natriuretic
peptide.

1.000
8.980
3.249
1.000
1.000
16.000 14.929
8.980 18.809
1.000 1.000
7.127 4.925
7.464 5.040
0.00 1.000
 3.17 9.624
 1.70 3.647
0.00
 3.90
 4.23
0.00
 4.00
 3.17
0.00
 2.83
 2.90
4.07  1.83 6.90  0.87 4.00  1.18 6.77  0.64 5.67  0.64 4.23  0.97 0.00
0.80  0.44 4.07  0.86 1.70  0.82 2.77  0.38 1.77  1.77 1.07  0.64  3.27
2.20  1.57 4.00  2.41 1.67  1.69 3.60  0.26 1.43  0.61 2.53  1.83  1.87
0
0.1
1
HSF-6

0.00
 2.30
 2.33

1.000
1.782
0.933
1.097
1.000
1.350
1.072
1.551
1.000
2.895
4.702
3.564
1.000 1.000
1.954 7.639
10.079 2.297
6.650 1.097
0.00 1.000
 0.83 13.611
0.10 24.251
 0.13 12.409
0.00
 0.43
 0.10
 0.63
0.00
 1.53
 2.23
 1.83
0.00
 2.93
 1.20
 0.13
0.00
 0.97
 3.33
 2.73
0.40 0.00
0.15  3.77
0.40  4.60
0.42  3.63





2.34 8.13
0.21 7.17
0.35 4.80
1.04 5.40






0.35 3.53
1.17 0.60
0.10 2.33
1.00 3.40






0.31 3.70
0.40 2.17
0.15 1.47
2.00 1.87






1.00 6.93
0.23 6.50
0.35 6.83
1.42 6.30






0.49 2.07
0.36 1.23
0.31 2.17
0.61 1.93





5.20
1.43
0.60
1.57

GATA-4 Nkx2.5

Miz-hES2 0
0.1
1
10

GATA-4 Nkx2.5 a-MHC b-MHC MLC-2V ANP GATA-4 Nkx2.5 a-MHC b-MHC MLC-2V ANP
a-MHC b-MHC

MLC-2V ANP

DDCt
DCt

5-azacytidine
(mM)
Cell
lines

Table 3 Expression of cardiac-specific transcription factors and markers in hESC-derived cardiomyocytes by real-time PCR

2  DDCt

156

cardiomyocytes using different lines of hESCs. In comparison with previous reports (Kehat et al., 2001; Xu
et al., 2002), we found that the hanging drop culture
was more efcient than the suspension culture for the
generation of beating cells. Also, there was no effect
when 5-azacytidine was treated after differentiation day
4 in our study (data not shown). These apparent differences in the efciency of differentiation may be due
to differences in the methods used for EB formation and
the culture conditions for the differentiated EBs. In our
study, EBs were formed by mechanical dissociation of
hESCs into clumps and then cultured clumps in hanging
drops containing DMEM/F12 with 20% FBS. We observed that EBs in suspension culture formed a lower
percentage of beating cardiomyocytes than EBs in
hanging drop culture. Therefore, we used the hanging
drop culture to form the EBs, and attached the EBs
onto plates for cardiomyocyte differentiation after 3day culture of EBs in hanging drop instead of 710 days
culture of EBs in suspension, as previously described
(Kehat et al., 2001). In addition, we applied a single EB
per 20 ml of hanging drop containing DMEM/F12 medium with 20% FBS. We also addressed the importance
of the number of EBs per well for the differentiation
into cardiomyocytes. We found that ve EBs per
1.9 cm2/well were more optimal for the cardiomyocyte
differentiation (data not shown). With the dierentiation protocol, by combining hanging drop culture and
5-azacytidine treatment, we were able to enhance the
rate of beating EBs generation by up to 41.88  0.92%
(Miz-hES2) and 34.89  1.29% (HSF-6), which is much
more efcient than the 8.1% originally reported by
Kehat and coworkers (2001) [9].
Furthermore, various hESC lines (HSF-6 and MizhES1, 2, 4, and 6) (Park et al., 2003; Kim et al., 2005)
can be efciently induced to differentiate into cardiomyocytes in vitro with 5-azacytidine treatment. Although spontaneous beating has never been observed in
Miz-hES1, 4, and 6 hESC cell lines, we could, however,
generate beating cells from Miz-hES1, 4, and 6 hESC
cell lines by combining hanging drop culture and 5azacytidine treatment up to 17.6%, 11.1%, and 5%,
respectively.
The number of beating cells significantly increased
when hESCs were treated with 0.1 mM of 5-azacytidine
during differentiation days 13. In addition, 0.1 mM of
5-azacytidine increased the expression of cardiac-specific transcription factors and marker genes, including
GATA-4, Nkx2.5, ANP, cardiac a/b-MHC, and MLC2V, compared with those in the no 5-azacytidine
treatment group, while the expression of the cardiac
transcription factors was even higher in 1 mM of the
5-azacytidine treatment group. Therefore, it appeared
that the concentration of 5-azacytidine was critical in
the differentiation of hESCs into cardiomyocytes. In
fact, in the group treated with 1 mM of 5-azacytidine,
half of the cells were dispersed to death, and the re-

1.000
2.000
8.574
5.401
1.000
1.149
2.000
6.350
1.000
8.378
7.294
6.650
1.000
1.447
2.764
2.579
4.90
4.47
2.23
2.17
5
10
15
20
HSF-6

Denotes statistical significance when po0.05 by w2-test. (n 5 3).

Expressions of the various mRNAs were normalized by the expression of b-actin.
a-MHC, a-myosin heavy chain; b-MHC, b-myosin heavy chain; MLC-2V, myosin light chain-2V; ANP, atrial natriuretic peptide.

1.000
6.350
5.924
2.406
0.00 1.000
 1.00 1.350
 3.10 6.350
 2.43 6.650
0.00
 0.20
 1.00
 2.67
0.00
 3.07
 2.87
 2.73
0.00
 0.53
 1.47
 1.37
0.00
 2.67
 2.57
 1.27
0.36 0.00
2.07  0.43
0.87  2.67
1.42  2.73




0.21 5.20
1.05 4.20
1.62 2.10
2.06 2.77




0.52 7.33
0.15 7.13
0.32 6.33
0.81 4.67




0.36 9.70
0.85 6.63
0.21 6.83
1.36 6.97




0.62 7.90
0.51 7.37
0.65 6.43
0.74 6.53
0.35 11.40
1.31 8.73
0.90 8.83
1.71 10.13










1.000
0.741
2.639
1.866
1.000
1.260
3.482
1.350
1.000
1.414
2.144
3.403
1.000
1.587
3.647
3.564
1.000
1.176
2.047
3.249
0.00 1.000
0.43 2.144
 1.40 5.401
 0.90 3.564
0.00
 0.33
 1.80
 0.43
0.00
 0.50
 1.10
 1.77
0.00
 0.67
 1.87
 1.83
0.00
 0.23
 1.03
 1.70
0.25 0.00
2.73  1.10
1.19  2.43
1.55  1.83




1.11 3.03
1.45 3.47
0.40 1.63
0.95 2.13




0.61 2.20
1.15 1.87
1.85 0.40
1.18 1.77




0.76 3.20
1.61 2.70
0.92 2.10
1.25 1.43




0.80 3.87
1.01 3.20
0.96 2.00
1.80 2.03




4.33
4.10
3.30
2.63
0.51
2.80
0.87
1.17




3.47
2.37
1.03
1.63
Miz-hES2 5
10
15
20

GATA-4 Nkx2.5 a-MHC b-MHC MLC-2V ANP GATA-4 Nkx2.5 a-MHC b-MHC MLC-2V ANP
GATA-4 Nkx2.5

a-MHC b-MHC MLC-2V ANP

DDCt
DCt

Days of
differentiation
Cell
lines

Table 4 Expression of cardiac-specific transcription factors and cardiac-specific markers during cardiomyocyte development by real-time PCR

2DDCt

157

sulting EBs decreased in size and had a lower chance to

form beating EBs. Also, when we treated the EBs with
5-azacytidine for a longer time (72 hr), we observed cell
death in the EBs, which resulted in failure of beating
EBs generation. In HSF-6 cell line, most of the cells
were lysed in 1 mM 5-azacytidine treatment and showed
less efciency for beating cell generation. On the other
hand, 10 mM 5-azacytidine-treated cells were all lysed
and no beating cells were generated. Therefore, the
toxic effect of 5-azacytidine was higher in HSF-6 cell
line than in Miz-hES2 cell line. On the other hand,
treatment with 0.1 mM of 5-azacytidine increased the
expression of both cardiac transcription factors and
cardiac-specific markers in HSF-6 cell line.
Therefore, it appeared that the cardiomyogenic effect
of 1 mM 5-azacytidine was somewhat compromised by
its toxic effect, and the resulting optimal dose of 5azacytidine for cardiomyogenic effect could be 0.1 mM.
In Miz-hES2 cell line, the highest expression of the
transcription factors, GATA-4 and Nkx2.5, was observed in the 1 mM 5-azacytidine treatment group. On
the other hand, a positive correlation of expression levels was observed only between two transcription factors
(GATA-4 and Nkx2.5) and b-MHC. Therefore, it is
possible that cardiac-specific markers are regulated by
additional transcription factors, such as myocyte enhancer factor 2 or T-box-containing transcription factor-5 (Morin et al., 2000; Hiroi et al., 2001). In HSF-6
cell line, however, the expression levels of both cardiac
transcription factors and cardiac-specific markers were
higher in 0.1 mM. Electrophysiological proles of spontaneously differentiated beating cells showed that there
are three distinct types of cardiomyocytes, including
atrial-, ventricular-, and nodal-like cells (He et al.,
2003). In a recent report, high-resolution activation
maps using microelectrode array (MEA) suggested the
presence of a functional syncytium in the hESC-derived
cardiomyocytes with stable focal activation and conduction properties (Kehat et al., 2002).
In the present study, we comprehensively characterized the cardiomyocytes derived from hESCs by RTPCR, immunouorescence, and confocal microscopy as
well as by TEM and electrophysiological recording. The
results of these studies conrmed that the differentiated
cells were structurally and functionally equivalent to
cardiomyocytes. These cells expressed cardiac-specific
genes, including transcription factors GATA-4 and
Nkx2.5, which play a significant role in cardiac development (Grepin et al., 1997; Lien et al., 1999). In
addition, cardiac-specific genes were also detected, including a-MHC, b-MHC, cTnT, cTnI, ANP, MLC-2A,
and MLC-2V, even though cTnI was expressed in
undifferentiated hESC, EBs, and non-beating cells.
Furthermore, it has recently been reported that cTnI
was not appropriate for the cardiac muscle specific
marker (Messner et al., 2000). Therefore, it is not
enough to conclude that 5-aza-dC had the effect on the

158

cardiomyocyte differentiation by performing immunostaining using antibody against cTnI (Xu et al., 2002).
Therefore, identication of hESC-derived cardiomyocytes should be conrmed by performing structural and functional analyses, such as electron microscopy
and electrophysiology.
In the present study, we performed sequence analysis
of cTnI RT-PCR product to eliminate the false positive
result and veried that the amplied cTnI RT-PCR
product was human cTnI and was expressed in undifferentiated hESCs, non-beating EBs, and beating EBs
(data not shown). TEM and electrophysiological recording further veried that our hESC-derived cardiomyocytes had the structural and functional properties
of cardiomyocytes.
In conclusion, differentiation of hESCs into cardiomyocytes can be enhanced by applying hanging drop
culture and 5-azacytidine treatment with the different
eciencies of cardiomyogenesis among hESC lines tested in this study, which suggests that each hESC line
may respond differently to the cardiomyogenic stimuli.
Although the mechanism by which 5-azaytidine induces
differentiation into cardiomyocytes remains unclear, the
results suggest that the genes required for cardiogenesis
may be silenced by methylation. Therefore, the methylation status of genes related to cardiomyocyte development may play an important role in the differentiation
of hESCs into cardiomyocytes. We are currently
conducting transplantation of hESC-derived cardiomyocytes into large animal disease models, such as pigs
and primates, which will be necessary to conrm the
function of these cells in vivo. Additional challenges
will be the enrichment of cardiomyocytes by directed
differentiation of hESCs and the elucidation of the
mechanism by which hESCs differentiate into cardiomyocytes.
Acknowledgments This research was supported by grants from the
Stem Cell Research Center of the 21st Century Frontier Research
Program of the Korean Ministry of Science and Technology
(SC2200). This project was also supported by the Korean Ministry
of Education and Human Resources (2005).

References
Baharvand, H., Ashtiani, S.K., Valojerdi, M.R., Shahverdi, A.,
Taee, A. and Sabour, D. (2004) Establishment and in vitro differentiation of a new embryonic stem cell line from human
blastocyst. Differentiation 72:224229.
Fukuda, K. (2003) Use of adult marrow mesenchymal stem cells for
regeneration of cardiomyocytes. Bone Marrow Transplant
32:S25S27.
Grepin, C., Nemer, G. and Nemer, M. (1997) Enhanced cardiogenesis in embryonic stem cells overexpressing the GATA-4
transcription factor. Development 124:23872395.
Haaf, T. and Schmid, M. (2000) Experimental condensation inhibition in constitutive and facultative heterochromatin of mammalian chromosomes. Cytogenet Cell Genet 91:113123.
Hakuno, D., Fukuda, K., Makino, S., Konishi, F., Tomita, Y.,
Manabe, T., Suzuki, Y., Umezawa, A. and Ogawa, S. (2002)

Bone marrow-derived cardiomyocytes (CMG cells) express functional adrenergic and muscarinic receptors. Circulation 105:
380386.
He, J.Q., Ma, Y., Lee, Y., Thomson, J.A. and Kamp, T.J. (2003)
Human embryonic stem cells develop into multiple types of cardiac myocytes: action potential characterization. Circ Res 93:
3239.
Hiroi, Y., Kudoh, S., Monzen, K., Ikeda, Y., Yazaki, Y., Nagai, R.
and Komuro, I. (2001) Tbx5 associates with Nkx2-5 and
synergistically promotes cardiomyocyte differentiation. Nat
Genet 28:276280.
Itskovitz-Eldor, J., Schuldiner, M., Karsenti, D., Eden, A.,
Yanuka, O., Amit, M., Soreq, H. and Benvensity, N. (2000)
Differentiation of human embryonic stem cells into embryoid
bodies compromising the three embryonic germ layers. Mol Med
6:8895.
Jackson, K.A., Majka, S.M., Wang, H., Pocius, J., Hartley, C.J.,
Majesky, M.W., Entman, M.L., Michael, L.H., Hirschi, K.K.
and Goodell, M.A. (2001) Regeneration of ischemic cardiac
muscle and vascular endothelium by adult stem cells. J Clin Invest 107:13951402.
Johkura, K., Cui, L., Suzuki, A., Teng, R., Kamiyoshi, A., Okamura, S., Kubota, S., Zhao, X., Asanuma, K., Okouchi, Y.,
Ogiwara, N., Tagawa, Y. and Sasaki, K. (2003) Survival and
function of mouse embryonic stem cell-derived cardiomyocytes
in ectopic transplant. Cardiovasc Res 58:435443.
Kajstura, J., Leri, A., Finato, N., Di Loreto, C., Beltrami, C.A.
and Anversa, P. (1998) Myocyte proliferation in end-stage
cardiac failure in humans. Proc Natl Acad Sci USA 95:
88018805.
Kehat, I., Gepstein, A., Spira, A., Itskovitz-Eldor, J. and Gepstein,
L. (2002) High-resolution electrophysiological assessment of
human embryonic stem cell-derived cardiomyocytes: a novel
in vitro model for the study of conduction. Circ Res 91:
659661.
Kehat, I., Kenyagin-Karsenti, D., Snir, M., Segev, H., Amit, M.,
Gepstein, A., Livne, E., Binah, O., Itskovitz-Eldor, J. and Gepstein, L. (2001) Human embryonic stem cells can differentiate
into myocytes with structural and functional properties of cardiomyocytes. J Clin Invest 108:407414.
Kehat, I., Khimovich, L., Caspi, O., Gepstein, A., Shofti, R., Arbel,
G., Huber, I., Satin, J., Itskovitz-Eldor, J. and Gepstein, L.
(2004) Electromechanical integration of cardiomyocytes
derived from human embryonic stem cells. Nat Biotechnol 22:
12821289.
Kim, S.J., Lee, J.E., Park, J.H., Lee, J.B., Kim, J.M., Yoon, B.S.,
Song, J.M., Roh, S.I., Kim, C.G. and Yoon, H.S. (2005) Efcient
derivation of new human embryonic stem cell lines. Mol Cells
19:4653.
Klug, M.G., Soonpaa, M.H., Koh, G.Y. and Field, L.J. (1996)
Genetically selected cardiomyocytes from differentiation embryonic stem cells form stable intracardiac grafts. J Clin Invest
98:216224.
Lien, C.L., Wu, C., Mercer, B., Webb, R., Richardson, J.A. and
Olson, E.N. (1999) Control of early cardiac-specific transcription
of Nkx2-5 by a GATA-dependent enhancer. Development
126:7584.
Makino, S., Fukuda, K., Miyoshi, S., Konishi, F., Kodama, H.,
Pan, J., Sano, M., Takahashi, T., Hori, S., Abe, H., Hata, J.,
Umezawa, A. and Ogawa, S. (1999) Cardiomyocytes can be
generated from marrow stromal cells in vitro. J Clin Invest
103:697705.
Maltsev, V.A., Wobus, A.M., Rohwedel, J., Bader, M. and Hescheler, J. (1994) Cardiomyocytes differentiated in vitro from
embryonic stem cells developmentally express cardiac-specific
genes and ionic currents. Circ Res 75:233244.
Messner, B., Baum, H., Fischer, P., Quastho, S. and Neumeier, D.
(2000) Expression of messenger RNA of the cardiac isoforms of
troponin T and I in myopathic skeletal muscle. Am J Clin Pathol
114:544549.

159
Morin, S., Charron, F., Robitaille, L. and Nemer, M. (2000)
GATA-dependent recruitment of MEF2 proteins to target promoters. EMBO J 19:20462055.
Mummery, C., Ward-van Oostwaard, D., Doevendans, P., Spijker,
R., van den Brink, S., Hassink, R., van der Heyden, M., Opthof,
T., Pera, M., de la Riviere, A.B., Passier, R. and Tertoolen, L.
(2003) Differentiation of human embryonic stem cells to cardiomyocytes: role of coculture with visceral endoderm-like cells.
Circulation 107:27332740.
Orlic, D., Kajstura, J., Chimenti, S., Jakoniuk, I., Anderson, S.M.,
Li, B., Pickel, J., McKay, R., Nadal-Ginard, B., Bodine, D.M.
and Anversa, P. (2001) Bone marrow cells regenerate infarcted
myocardium. Nature 410:701705.
Park, J.H., Kim, S.J., Oh, E.J., Moon, S.Y., Roh, S.I. and Yoon,
H.S. (2003) Establishment and maintenance of human embryonic
stem cells on STO, a permanently growing cell line. Biol Reprod
69:20072014.
Reubino, B.E., Pera, M.F., Fong, C-Y., Trounson, A. and
Bongso, A. (2000) Embryonic stem cell lines from human blastocysts: somatic differentiation in vitro. Nat Biotechnol 18:399
404.

Rudnicki, M.A., Jackowski, G., Saggin, L. and McBurney, M.W.

(1990) Actin and myosin expression during development of cardiac muscle from cultured embryonal carcinoma cells. Dev Biol
138:348358.
Snir, M., Kehat, I., Gepstein, A., Coleman, R., Itskovitz-Eldor, J.,
Livne, E. and Gepstein, L. (2003) Assessment of the ultrastructural and proliferative properties of human embryonic stem cellderived cardiomyocytes. Am J Physiol Heart Circ Physiol
285:H2355H2363.
Thomson, J.A., Itskovitz-Eldor, J., Shapiro, S.S., Waknitz, M.A.,
Swiergiel, J.J., Marshall, V.S. and Jones, J.M. (1998) Embryonic
stem cell lines derived from human blastocysts. Science 282:1145
1147.
Tsuji-Takayama, K., Inoue, T., Ijiri, Y., Otani, T., Motoda, R.,
Nakamura, S. and Orita, K. (2004) Demethylating agent, 5azacytidine, reverses differentiation of embryonic stem cells.
Biochem Biophys Res Commun 323:8690.
Xu, C., Police, S., Rao, N. and Carpenter, M.K. (2002) Characterization and enrichment of cardiomyocytes derived from human embryonic stem cells. Circ Res 91:501508.