Am J Physiol Heart Circ Physiol 285: H2355–H2363, 2003;
10.1152/ajpheart.00020.2003.
Assessment of the ultrastructural and proliferative properties
of human embryonic stem cell-derived cardiomyocytes
Mirit Snir,1 Izhak Kehat,1 Amira Gepstein,1 Raymond Coleman,2
Joseph Itskovitz-Eldor,3 Erella Livne,3 and Lior Gepstein1
1
Cardiovascular Research Laboratory, Department of Physiology and Biophysics, 2Department
of Anatomy, and 3Department of Obstetrics and Gynecology, Rambam Medical Center,
The Bruce Rappaport Faculty of Medicine and the 1Rappaport Family Institute for Research
in the Medical Sciences, Technion-Israel Institute of Technology, 31096 Haifa, Israel
Submitted 8 January 2003; accepted in final form 14 July 2003
DESPITE THE DEVELOPMENTS made in different areas of
molecular cardiology, several questions relating to the
origin, differentiation, and maturation of early-stage
human cardiac tissue remain unanswered. Similarly,
the mechanisms that are involved in human cardiomyocyte cell-cycle regulation during embryonic and adult
life are also largely unknown. Research in these areas
has been partially hampered by the lack of a suitable
human in vitro model.
Embryonic stem (ES) cells are pluripotent cell lines
that were initially derived from the inner cell mass of
mouse blastocyst-stage embryos (9, 23) and recently
also from human blastocysts (28, 33). These unique cell
lines are characterized by their capacity for prolonged
undifferentiated proliferation in culture while retaining the capability to differentiate into derivatives of all
three germ layers. During in vitro differentiation, the
mouse ES cells were demonstrated to develop into
specialized somatic tissues, including cardiomyocytes,
and were shown to recapitulate many of the processes
of early in vivo cardiac development (4, 12). Recently, a
similar reproducible cardiomyocyte differentiating system was also established in the human model (16, 35).
Dispersed cells isolated from spontaneous beating areas generated within the differentiating embryoid bodies (EBs) were demonstrated to possess the ultrastructural, molecular, and functional properties of earlystage cardiomyocytes (16). More recently, we (15) have
demonstrated that this differentiating system is not
limited to the derivation of individual cardiomyocytes
but rather a functional syncytium is generated with
spontaneous pacemaker activity and action potential
propagation.
In the present study, we attempted to further characterize the ultrastructural and morphological changes
associated with the differentiation and maturation of
the human ES cell-derived cardiomyocytes. We also
aimed to establish this unique differentiating system
as an in vitro model to study human cardiomyocyte
cell-cycle regulation. Our results show a reproducible
temporal pattern of early cardiomyocyte proliferation,
cell-cycle withdrawal, and ultrastructural maturation
Address for reprint requests and other correspondence: L. Gepstein,
Cardiovascular Research Laboratory, The Bruce Rappaport Faculty of
Medicine, Technion-Israel Institute of Technology, 2 Efron St., PO Box
9649, 31096 Haifa, Israel (E-mail: [email protected]).
The costs of publication of this article were defrayed in part by the
payment of page charges. The article must therefore be hereby
marked ‘‘advertisement’’ in accordance with 18 U.S.C. Section 1734
solely to indicate this fact.
cell-cycle withdrawal; DNA synthesis; embryoid bodies; sarcomeres
http://www.ajpheart.org
0363-6135/03 $5.00 Copyright © 2003 the American Physiological Society
H2355
Downloaded from http://ajpheart.physiology.org/ by 10.220.33.5 on June 14, 2017
Snir, Mirit, Izhak Kehat, Amira Gepstein, Raymond
Coleman, Joseph Itskovitz-Eldor, Erella Livne, and
Lior Gepstein. Assessment of the ultrastructural and proliferative properties of human embryonic stem cell-derived
cardiomyocytes. Am J Physiol Heart Circ Physiol 285:
H2355–H2363, 2003; 10.1152/ajpheart.00020.2003.—Assessment of early ultrastructural development and cell-cycle regulation in human cardiac tissue is significantly hampered by the
lack of a suitable in vitro model. Here we describe the possible
utilization of human embryonic stem cell (ES) lines for investigation of these processes. With the use of the embryoid body
(EB) differentiation system, human ES cell-derived cardiomyocytes at different developmental stages were isolated and their
histomorphometric, ultrastructural, and proliferative properties were characterized. Histomorphometric analysis revealed
an increase in cell length, area, and length-to-width ratio in
late-stage EBs (⬎35 days) compared with early (10–21 days)
and intermediate (21–35 days) stages. This was coupled with a
progressive ultrastructural development from an irregular
myofibrillar distribution to an organized sarcomeric pattern.
Cardiomyocyte proliferation, assessed by double labeling with
cardiac-specific antibodies and either [3H]thymidine incorporation or Ki-67 immunolabeling, demonstrated a gradual withdrawal from cell cycle. Hence, the percentage of positively
stained nuclei in early-stage cardiomyocytes ([3H]thymidine:
60 ⫾ 10%, Ki-67: 54 ⫾ 23%) decreased to 36 ⫾ 7% and 9 ⫾ 16%
in intermediate-stage EBs and to ⬍1% in late-stage cardiomyocytes. In conclusion, a reproducible temporal pattern of early
cardiomyocyte proliferation, cell-cycle withdrawal, and ultrastructural maturation was noted in this model. Establishment
of this unique in vitro surrogate system may allow to examine
the molecular mechanisms underlying these processes and to
assess interventions aiming to modify these properties. Moreover, the detailed characterization of the ES cell-derived cardiomyocyte may be crucial for the development of future cell
replacement strategies aiming to regenerate functional myocardium.
H2356
PROPERTIES OF HUMAN ES-DERIVED CARDIOMYOCYTES
in this model. The detailed characterization of the
human ES cell-derived cardiomyocytes cells may also
possess an important clinical value because these
unique cells represent an attractive source for the
newly emerging cell therapy strategies aiming to regenerate functional myocardium.
METHODS
AJP-Heart Circ Physiol • VOL
RESULTS
Ultrastructural analysis. The earliest contracting
tissue could be noted in EBs at 7 days postplating.
Cardiomyocytes at this stage (Fig. 1A) were relatively
small with a large nucleus-to-cytoplasm ratio. The cells
contained few delicate myofibrils that lacked alignment or any organized sarcomeric pattern and were
distributed throughout the cytoplasm in a disoriented
285 • DECEMBER 2003 •
www.ajpheart.org
Downloaded from http://ajpheart.physiology.org/ by 10.220.33.5 on June 14, 2017
Human ES cell propagation and in vitro differentiation.
Human undifferentiated ES cells of the single-cell clone H9.2
(1) were cultured on the mitotically inactivated (mitomycin
C) mouse embryonic fibroblast feeder layer as previously
described (16). The culture medium consisted of 80% knockout DMEM (no-pyruvate, high-glucose formulation; Life
Technologies, Rockville, MD) supplemented with 20% FBS
(HyClone; Logan, UT), 1 mM L-glutamine, 0.1 mM mercaptoethanol, and 1% nonessential amino acid stock (all from
Life Technologies).
The human ES cell line used in the present study was the
National Institutes of Health-approved human ES cell clone
(H9.2). This line was obtained from Prof. Joseph ItskovitzEldor (Technion University, Haifa, Israel) and has been used
in several previous studies (1, 15, 16, 33).
To induce differentiation, the ES cells were dispersed to
small clamps (3–20 cells) with the use of collagenase IV (Life
Technologies, 1 mg/ml). The cells were then transferred to
plastic petri dishes, where they were cultured in suspension
for 7–10 days. During this period, the cells aggregated to
form EBs, which were then plated on 0.1% gelatin-coated
culture dishes and observed microscopically for the appearance of spontaneous contractions.
Preparation of samples for light and transmission electron
microscopy. Spontaneously contracting foci within the EBs
were mechanically dissected and fixed in 3% glutaraldehyde
in a 0.1 M sodium cacodylate buffer (pH 7.4) at 4°C for 24 h.
After fixation, samples were rinsed with 7.5% sucrose in the
0.1 M sodium cacodylate buffer (pH 7.4), postfixed in 1%
OsO4 in the same buffer for 1 h, dehydrated in graded
ethanol, and embedded in Epon 812.
For histomorphometric analysis, semithin epoxy resin-embedded sections (0.5 ␮m) were generated and stained with
0.1% toluidine blue in 1% sodium tetraborate (pH 12). Morphometric analysis was subsequently carried out using a
computerized image-analysis system composed of an Olympus CH40 microscope fitted with a charge-coupled device
camera (model Wat-202D, Watec; Tokyo, Japan) and using
the Analysis DocU3 software (Soft Imaging System; Munster, Germany). The boundary of each myocyte was traced,
and the cells’ longest diameter, width, length-to-width ratio
and area were determined. Several sections from seven EBs
were examined for each of the three developmental groups.
For transmission electron microscopy, thin (60–90 nm)
sections were prepared with the use of a diamond knife on a
ultramicrotome (Nova; Bromma, Sweden). Sections were
mounted on copper grids, stained with saturated uranyl
acetate and 1% lead citrate (Merck), and examined with the
use of a transmission electron microscope (model 100 SX,
JEOL; Peabody, MA) operating at 80 kV.
Immunohistochemistry and autoradiography. Incorporation of [3H]thymidine by the human ES cell-derived cardiomyocytes was used to assess DNA synthesis by these cells.
[3H]thymidine (New Life Science; Boston, MA) was added to
culture medium of beating EBs at a final concentration of 10
␮Ci/ml. After incubation of 17 h at 37°C, the contracting EBs
were rinsed three times with DMEM medium. The beating
areas within the EBs were mechanically dissected and fixed
immediately in 10% neutral-buffered paraformaldehyde
overnight. Preparations were then dehydrated in graded
alcohols (70–100%), cleared in chloroform, and embedded in
paraplast (Sherwood Medical).
The 5-␮m sections were deparaffinized and treated with
3% H2O2 in methanol for 20 min. Sections were then stained
with the Histostain-SP kit (Zymed) according to the manufacturer’s instructions. The primary antibodies that were
used included mouse anti-cardiac troponin I (anti-cTnI;
Chemicon, Temecula, CA), anti-sarcomeric ␣-actinin mAbs
(Sigma), and anti-nebulin antibodies (Sigma) at dilutions of
1:5,000, 1:50, and 1:100, respectively. For autoradiography,
after immunohistological processing, the sections were immersed at 42°C in an autoradiographic emulsion solution
(LM-1, Amersham) for 4 min. The sections were then air
dried and kept in light-tight boxes for 12 days. The emulsions
were developed with the use of a Kodak D-170 developer
(18°C, 7 min). The sections were then fixed in 25% sodium
thiosulfate (Sigma) for 10 min and examined microscopically.
Cardiomyocytes were identified in the sectioned EBs by the
positive cytoplasmatic staining with sarcomeric ␣-actinin. All
cardiomyocyte nuclei were clearly identified and then scored
(labeling index) for the percentage of nuclei that contained
silver grains. The procedure was repeated for EBs at various
developmental stages (multiple sections from seven different
EBs sampled from at least three different experiments for
each developmental group).
To further evaluate cardiomyocyte cell-cycle activity, we
utilized an additional cell-cycle marker, Ki-67. Whole EBs
were fixed in 4% paraformaldehyde and blocked with PBS
containing 3% normal goat serum for 60 min. Immunostaining was then performed using mouse anti-cTnI antibody
(Chemicon) at a dilution of 1:5,000 and rabbit polyclonal
anti-human Ki-67 antibody (Santa Cruz) at a dilution of
1:100 overnight at 4°C. Preparations were then incubated
with Cy-2-conjugated anti-rabbit IgG and Cy-3-conjugated
anti-mouse IgG (both from Chemicon) at a dilution of 1:100
for 1 h at room temperature. Cell nuclei were counterstained
with the DNA dye To-Pro-3 (Molecular Probes; Eugene, OR).
Confocal microscopy was performed using a Nikon Eclipse
E600 microscope and Bio-Rad Radiance 2000 scanning system. The cardiomyocyte labeling index was then determined
as the percentage of cardiomyocyte nuclei that were positively stained for Ki-67.
Statistical analysis. Data are expressed as means ⫾ SD.
The EBs were divided into three groups: 1) early stage: 7–20
days postplating, 2) intermediate stage: 21–35 days postplating, and 3) late stage: 36–60 days postplating. Multiple
sections from seven EBs from at lease three different experiments were analyzed in each group. Differences between the
groups in the morphometric measurements and labeling indexes were tested with the use of a Kruskal-Wallis one-way
analysis of variance on ranks test. If found to be statistically
significant, then post hoc analysis was performed with the
use of the unequal Tukey’s test. P ⬍ 0.05 was considered to be
statistically significant.
PROPERTIES OF HUMAN ES-DERIVED CARDIOMYOCYTES
AJP-Heart Circ Physiol • VOL
and random fashion (Fig. 1, A and B). Many free
ribosomes, some arranged in rosettes, were observed
throughout the cytoplasm in close proximity to the
myofibrils, suggesting that these cells are actively engaged in protein synthesis (Fig. 1B). During early development (17 days postplating), the number of myofibrils gradually increased and while developing myofibrillar stacks continued to be randomly oriented in
some cells, in others a shift to a more organized structure was noted (Fig. 1C). This was manifested by organization of the myofibrils around an amorphous electron-dense material representing the precursor of the Z
band (Z body). In some areas, a more developed pattern
was noted in which the fibrils were loosely held between two or three successive Z bodies.
At 27 days postplating, ultrastructural maturation
continued and was characterized by the development of
a sarcomeric organization. This was manifested by
fusion of the round clusters of the electron-dense material (Z bodies) to form a continuous line, the nascent
Z band (Fig. 1D). Parallel Z bands were demonstrated
to confine the myofibrils in the typical sarcomeric pattern. The Z bands also served as anchoring points for
branching sarcomeres. Despite this increased level of
organization, the number of sarcomeres within the
cells was still relatively low.
At 40 days postplating, there was a gradual increase
in the number of myofibrils and in their degree of
spatial organization (Fig. 2, A and B). The cells become
elongated as their cytoplasm filled with sarcomeres
(Fig. 2A). There was a clear increase in the amount of
the electron-dense material in the Z band (Fig. 2B).
However, even at this stage, different degrees of organization patterns could exist simultaneously in neighboring cells in the same EB and even within the same
cell (Fig. 2A).
At 60 days of postplating, we could note an additional increase in the amount and degree of organization of the sarcomeres within the cytoplasm (Fig. 2C).
This can also be viewed in both longitudinal and transverse sections (Fig. 2D). Mitochondria were seen
packed in close association with the myofibrils (Fig.
2C). At this stage, discrete A and I bands in addition to
the Z bands could also be noted within the sarcomeres
(Fig. 2C) and, in a minority of the cells, also M bands.
Interestingly, the system of transverse (T) tubules,
considered to develop very late during cardiomyocyte
development, was not seen in any of the cells examined.
Morphometric analysis. Immunohistochemical studies demonstrated that the human ES cell-derived cardiomyocytes were generally concentrated in a single
area that occupied a relatively small portion of the EB
(Fig. 3, A and B). Within these beating regions, however, cardiomyocytes dominated and consisted of 30–
60% of all cells (Fig. 3, A–D).
Morphological analysis of the ES cell-derived cardiomyocytes revealed a developmental maturation process. The early-stage cardiomyocytes were relatively
small, round, or rod shaped and overwhelmingly mono-
285 • DECEMBER 2003 •
www.ajpheart.org
Downloaded from http://ajpheart.physiology.org/ by 10.220.33.5 on June 14, 2017
Fig. 1. Transmission electron microscopy images of the human embryonic stem (ES) cell-derived cardiomyocytes in early- and intermediate-stage embryoid bodies (EBs). A and B: representative electron
micrographs of early-stage cardiomyocytes (7 days postplating). Note
that cardiomyocytes were relatively small with a large nucleus
(Nu)-to-cytoplasm ratio and abundant distribution of ribosomes (free
ribosomes, FR). Nascent myofibrils (MF) were present and distributed randomly within the cytoplasm. Also present were developing
intercalated discs (ID). C: electron micrograph of a beating EB 17
days postplating. Note the development of electron dense spots (Z
bodies, arrows) anchoring the myofibrils. D: ultrastructural characterization of a beating EB 27 days postplating. Note the development
of a sarcomeric ultrastructural pattern and fusion of the electron
dense spots to form the nascent Z band (arrow).
H2357
H2358
PROPERTIES OF HUMAN ES-DERIVED CARDIOMYOCYTES
nucleated (Figs. 3C). With maturation, the overall appearance of the human ES-derived cardiomyocytes
changed to strands of more elongated cardiomyocytes
(Figs. 3D). This was also accompanied by an increase in
the cytoplasm-to-nuclear ratio. Interestingly, in contrast to the murine ES model (18), almost all cells
(⬎99%) remained mononucleated.
Morphometric analysis of the ES cell-derived cardiomyocytes was accomplished by measuring the longest
length, width, length-to-width ratio, and area of individual cells. The general tendency (Fig. 4) was a significant increase in the length and area of the cardiomyocytes in late-stage EBs (n ⫽ 94) compared (P ⬍
0.05) with cells from the early (n ⫽ 87) and intermediate (n ⫽ 102) stages. Because no significant changes
were noted in the cells’ width, this tendency was also
coupled with a significant increase in the cells’ lengthto-width ratio in late-stage EBs (Fig. 4).
Proliferation and DNA synthesis. Assessment of the
incorporation of [3H]thymidine was used to evaluate
the DNA synthesis ability of the ES cell-derived cardiomyocytes during in vitro development. To ensure that
the labeling index measurements were performed in
cardiomyocytes, we initially had to develop a reliable
assay capable of distinguishing these cells from other
ES-cell derivatives. Preliminary experiments indicated
that this could be readily accomplished by initial mechanical microdissection of the spontaneously contracting foci within the EBs to achieve cardiomyocyte-enriched specimens. Cardiomyocytes could then be identified by the positive staining of sarcomeric ␣-actinin
AJP-Heart Circ Physiol • VOL
and absence of Nebulin (a skeletal muscle marker)
immunosignal.
Figure 5, A–D, displays typical examples of simultaneous autoradiographic and immunohistochemistry
studies of undifferentiated stem cells (Fig. 5A) and of
contracting EBs of different developmental stages
(Figs. 5, B–D). Cardiomyocytes could be identified by
the positive ␣-actinin immunoreactive signal (red-orange areas in Figs. 5, B–D). Note the lack of staining at
the initial stem cell phase (Fig. 5A) and the gradual
increase in the intensity of the signal from early-stage
(Fig. 5B) to intermediate- (Fig. 5C) and late-stage cardiomyocytes (Fig. 5D). Incorporation of [3H]thymidine
was be identified by the presence of silver grains over
the cells’ nuclei (Fig. 5, A–D). Positive stained (silverblack) nuclei were present in the majority of cells at the
undifferentiated stem cell phase (Fig. 5A) and their
number gradually decreased from early-stage (Fig. 5B)
to intermediate-stage (Fig. 5C) cardiomyocytes. DNA
synthesis was almost completely absent (lack of
[3H]thymidine uptake) in late-stage cardiomyocytes
(Figs. 5D).
The observed developmental pattern in the labeling
index during in vitro cardiomyogenesis was observed
in all EB studied (Fig. 6). This analysis revealed that
undifferentiated stem cells as well as EBs in suspension before cardiomyocyte induction were actively synthesizing DNA (labeling indexes ⬎60%). A similar
high-labeling index was also noted in early-stage cardiomyocytes with 60 ⫾ 10% of the cells showing
[3H]thymidine uptake. The labeling index gradually
285 • DECEMBER 2003 •
www.ajpheart.org
Downloaded from http://ajpheart.physiology.org/ by 10.220.33.5 on June 14, 2017
Fig. 2. Ultrastructural characterization of late-stage
ES-derived cardiomyocytes. A and B: electron micrographs of a beating EB 40 days postplating. Note the
increase in the quantity and the degree of organization
of the contractile proteins (A) and in the amount of the
electron-dense material forming the Z band (B). C and
D: ultrastructural characterization of an EB at 60 days
postplating. Note the increase in the number and degree of maturation of the sarcomeres. Also present are
mitochondria (Mit) packed between the sarcomeres.
Myofibrils can be appreciated in both transverse and
longitudinal sections (D).
PROPERTIES OF HUMAN ES-DERIVED CARDIOMYOCYTES
declined in intermediate-stage (21–35 days postplating) contracting EBs (36 ⫾ 7%) and was reduced to
⬍1% in late-stage EBs (over 36 days postplating).
To further evaluate cell-cycle activity during in vitro
cardiomyocyte development we utilized an additional
marker of cell-cycle activity, Ki-67. Expression of Ki-67
is a prerequisite for cells to transverse the cell cycle
and undergo cell division. Figure 5, E and F, shows
typical confocal images of EBs at various developmental stages costained with anti-cTnI antibodies (red) and
H2359
anti-Ki-67 antibodies (green nuclear staining). Note
the high percentage of positively stained nuclei for
Ki-67 in early-stage cardiomyocytes (Fig. 5E) and the
absence of any significant staining in late-stage cardiomyocytes (Fig. 5F). Hence, similar to the [3H]thymidine experiments, the labeling index for Ki-67 was
55 ⫾ 23% in early-stage cardiomyocytes (20 days postplating) and was reduced to 10 ⫾ 16% (P ⬍ 0.05) in
intermediate-stage cardiomyocytes (30-day EBs) and
to 2 ⫾ 2% (P ⬍ 0.05) in late-stage EBs (38 days
postplating).
DISCUSSION
Fig. 3. A: immunostaining of a sectioned EB (shown at a low magnification) with anti-sarcomeric ␣-actinin antibodies. Note that the
positively stained cardiomyocytes (brown cells) are grouped together
(white arrow) and occupy only a limited area within the EB. B:
confocal immunoflurescent image of a contracting EB with the use of
anti-cardiac tropinin I (cTnI) antibodies (red) and ToPro3 (nuclear
staining, blue). The positively stained cardiomyocytes (red cells)
were grouped in a single area (white arrow) within the EB, whereas
the rest of the EB was composed of noncardiomyocytes cells [identified by the positive nuclear staining with TroPro3 (blue) and the lack
of cytoplasmatic staining with anti-cTnI antibodies]. Note that the
ES cell-derived cardiomyocytes display an early-striated cytoplasmatic staining and are organized in isotropic pattern within the EB.
C and D: immunostaining of sectioned EBs with the use of antisarcomeric ␣-actinin antibodies. The micrographs show high-resolution
views only of the contracting area within the EB and consequentially
all cells are stained (brown). Note the morphological changes of the
ES cell-derived cardiomyocytes during maturation from small and
round cell in early-stage cardiomyocytes (9 days after plating, C) to
larger and more elongated cells in late-stage EBs (51 days postplating, D). Almost all cardiomyocytes were mononucleated.
AJP-Heart Circ Physiol • VOL
285 • DECEMBER 2003 •
www.ajpheart.org
Downloaded from http://ajpheart.physiology.org/ by 10.220.33.5 on June 14, 2017
Cell replacement therapy is emerging as a novel
therapeutic paradigm for myocardial repair but has
been hampered by the lack of cell sources for human
cardiac tissue (27). Human ES cells may provide an
attractive cell source for this strategy because of their
ability to be propagated in vitro in mass and to differentiate into the desired cardiac lineage. As a first step
in the development of such a clinical strategy, however,
detailed characterization of the phenotypic properties
of these cells should be performed. Of special interest is
to determine the proliferative capacity and ultrastructural characteristics of these cardiomyocytes because
these properties may have a significant impact on the
ultimate graft success (27).
In this study, we examined the developmental
changes associated with the in vitro differentiation and
maturation of human ES cell-derived cardiomyocytes.
Our results demonstrate a reproducible temporal pattern of cardiomyocyte proliferation, cell-cycle withdrawal, and ultrastructural and morphological maturation. In addition to the significance of these results in
characterizing this promising cell source for future cell
therapy procedures, the current report also presents a
new in vitro surrogate system to study the mechanisms
underlying cell-cycle regulation and ultrastructural
development as well as possible interventions aiming
to modify these processes.
H2360
PROPERTIES OF HUMAN ES-DERIVED CARDIOMYOCYTES
Fig. 4. Morphometric changes in the
cell’s length, width, area, and lengthto-width ratio during in vitro cardiomyocyte maturation. *P ⬍ 0.05 compared with baseline measurements.
AJP-Heart Circ Physiol • VOL
long enough to achieve full ultrastructural maturity.
To this end, it is interesting to note that human cardiomyocytes derived from fetal hearts as late as after
midterm still did not achieve full ultrastructural maturity and that myofibrillar development continued
throughout the entire fetal period (17). Other possible
factors may be related to the in vitro culturing conditions of the model and the lack of in vivo signals for
hypertrophic growth. In particular, the lack of hemodynamic workload typical of in vivo working myocytes
may be of key importance because cardiomyocytes
grown in vivo in the absence of this workload were
shown to lack appropriate cell alignment and ultrastrctural development (3).
The mechanisms controlling the complex process of
myofibril assembly and sarcomeric organization are
still not fully understood (10). Several models have
been used to study these processes, including different
primary cell cultures and in situ embryonic models,
mostly in chicks and rats. The development of the
human ES differentiating system may prove valuable
for these types of studies because of the relatively
reproducible ultrastructural developmental process observed here, the prolonged culturing capability of the
model, and the ability to modify the ES cells genetically.
The latter property may be of significant value because
the genetic modification of these important structural
proteins would otherwise lead to embryonic lethality.
Cardiomyocyte proliferation and DNA synthesis.
Whereas numerous animal studies examined cardiomyocyte DNA synthesis and cell cycle activity during
embryonic, fetal, and neonatal development (2, 25, 31),
there is a paucity of similar data regarding during
early human cardiac development (14). In the current
study, the DNA synthesis ability of human ES cellderived cardiomyocytes and their capability to undergo
reduplication and cell-cycle withdrawal were examined. Our results demonstrate a high level of [3H]thy-
285 • DECEMBER 2003 •
www.ajpheart.org
Downloaded from http://ajpheart.physiology.org/ by 10.220.33.5 on June 14, 2017
Ultrastructural and morphological development. In
our previous report (16), we have demonstrated that
human ES cells can differentiate to generate contracting cells with structural and functional properties typical of early-stage cardiomyocytes. In the current
study, we expanded these observations by examining
the process of cardiomyocyte ultrastructural and morphometric maturation in this in vitro model during a
period of up to 60 days in culture. Our results show
that the human ES cell-derived cardiomyocytes follow
a maturation process that is roughly similar to that
reported in vivo (5, 7, 17, 20, 22) and in the in vitro
murine ES model (11, 12, 34). This process was characterized by gross morphological changes in which the
relatively small and round cells situated in round accumulations in early-stage EBs gradually developed to
larger and more elongated cells that tended to accumulate in strands in late-stage EBs. These morphological
changes were coupled with a similar ultrastructural
maturation process characterized by a progressive increase in the amount and organization of the contractile material. Hence, the initial pattern of irregular
myofibrillar distribution in early-stage myocytes gradually organized into parallel myofibril arrays in more
developed cells and ultimately resulted in the generation of well-defined sarcomeres with recognizable A, I,
and Z bands in late-stage myocytes. Yet it is interesting to note that despite this developmental pattern the
human ES cell-derived cardiomyocytes still did not
reach the level of maturity, typical of adult cardiomyocytes. This was manifested, for example, by the lack of
developed T tubule system.
The inability of the human ES-derived cardiomyocytes to reach the same degree of ultrastructural maturity as noted in the vivo adult heart may be related to
a number of factors. First, although human ES-derived
cardiomyocytes were followed in this study for a relatively long period (up to 60 days), it might still not be
PROPERTIES OF HUMAN ES-DERIVED CARDIOMYOCYTES
H2361
midine uptake in early-stage cardiomyocytes which
was gradually reduced in intermediate-stage EBs and
was almost completely absent in late-stage EBs. Cessation of DNA synthesis in our model and cell cycle
withdrawal was followed by a stage characterized by
morphological and ultrastructural maturation. A similar pattern was also noted when cycling myocytes
AJP-Heart Circ Physiol • VOL
were identified by the expression of Ki-67 in the nuclei.
Although the exact function of Ki-67 is not clear, it is
expressed in all proliferating cells in late G1, S, and
G2M phases of the cell cycle (21, 29). One advantage of
this marker is that it is not involved in DNA repair.
The cell-cycle changes observed in the human ES
cell-derived cardiomyocytes follow a roughly similar
285 • DECEMBER 2003 •
www.ajpheart.org
Downloaded from http://ajpheart.physiology.org/ by 10.220.33.5 on June 14, 2017
Fig. 5. Changes in the cell cycle activity and DNA synthesis during in vitro
maturation of ES cell-derived cardiomyocytes. A–D: autoradiography and
immunohistochemistry (using anti-sarcomeric ␣-actinin antibodies) of 10- to
51-day-old contracting EBs. Undifferentiated ES cells and contracting EBs
at different developmental stages were
first labeled with [3H]thymidine and
then processed for autoradiography
and immunohistochemistry. The results of the immunostaining of undifferentiated stem cells (A) and of ES
cell-derived cardiomyocytes at 10 days
(B), 30 days (C), and 50 days (D) postplating are shown. Note the absence of
staining for sarcomeric ␣-actinin at the
stem cell phase and the gradual increase in the intensity of staining in
the ES-cell-derived cardiomyocytes (orange-brown cells filling the entire highmagnification field) during maturation.
Incorporation of [3H]thymidine was
used to identify cells with DNA synthesis, which were identified by the presence of positively silver-black stained
nuclei. Note that the majority of cells at
the undifferentiated ES stage were actively synthesizing DNA. Similarly, silver grains were present in the nuclei of
cardiomyocytes in early-stage (B) and
intermediate-stage (C) EBs but were
almost completely absent in late-stage
EBs (D). E and F: confocal images
showing the results of double staining
of whole EBs with anti-cTnI antibodies
(red) and anti-human Ki-67 antibodies
(a marker of cycling cells, green nuclear staining). In addition, all nuclei
(of both cycling and noncycling cells)
can be viewed in the right panels by the
blue nuclear staining with To-Pro-3.
Note the presence of cycling myocytes
(red cells with positively stained green
nuclei) in early-stage EB (18 days postplating, E) and lack of any such cycling
cells (absence of nuclear staining in all
cardiomyocytes) in late-stage EBs (38
days postplating).
H2362
PROPERTIES OF HUMAN ES-DERIVED CARDIOMYOCYTES
pattern to that found in the murine ES model (18) with
a temporal difference related most probably to the
interspecies differences. The highly reproducible results obtained in both ES models and also in several in
vivo animal studies (31) raises the possibility that
cardiomyocytes may be intrinsically programmed with
respect to the number of cell divisions that can occur in
between cardiogenic induction and cell-cycle withdrawal or by an “intrinsic timer” (6) that may control
cell-cycle withdrawal.
In the mouse, there is a transition from hyperplastic
to hypertropic growth shortly after birth. This transition is coupled with a marked increase in myofibrillar
density and the formation of binucleated cardiomyocytes. Several studies (8, 32) documented during this
period a gradual decrease in radiolabeled thymidine
incorporation leading to the suggestion that binucleation results from genomic duplication and kariokinesis in the absence of cytokinesis. Our study provides
some conflicting results with respect to these findings.
First, we noted cessation of DNA synthesis in the
human ES cell-derived cardiomyocytes after ⬃40–50
days from the initiation of differentiation (7–10 days in
suspension plus ⬎36 days postplating). Although this
process was longer than that observed in the mouse ES
model, it is significantly shorter than the length of the
human gestational period. Second, the number of
binucleated cardiomyocytes observed after cessation of
DNA synthesis was very low (⬍1% in a similar range
as in the undifferentiated ES cells).
Although these differences may be inherent to the in
vitro nature of the model, there is some evidence to
suggest similar in vivo phenomena. It is interesting to
note that in a study that examined cardiomyocyte DNA
synthesis during murine fetal and neonatal development, two temporally distinct phases of DNA synthesis
were documented (32). The first phase occurred during
early fetal life and was associated exclusively with
cardiomyocyte proliferation and was followed by cessation of cardiomyocytes reduplication in late fetal life.
During the neonatal period a second phase of DNA
synthesis was observed, which was associated with the
process of binucleation. The termination of DNA synAJP-Heart Circ Physiol • VOL
285 • DECEMBER 2003 •
www.ajpheart.org
Downloaded from http://ajpheart.physiology.org/ by 10.220.33.5 on June 14, 2017
Fig. 6. Changes in the labeling index (percentage of positively radiolabeled nuclei) of undifferentiated stem cells (n ⫽ 2,172), EBs in
suspension before cardiogenic induction (n ⫽ 1,439), and early- (n ⫽
2,350), intermediate- (n ⫽ 2,193), and late (n ⫽ 1,715)-stage human
ES cell-derived cardiomyocytes. *P ⬍ 0.05 compared with baseline
measurements.
thesis in our model may thus relate to the first phase in
the above study. The paucity of binucleation in the
human ES-derived cardiomyocytes may also relate to
the first phase of DNA synthesis that was associated
exclusively with reduplication in the murine study
(32). It is also interesting to note that in general, the
percentage of binucleated or multinucleated adult human cardiomyocytes (⬃20%) (24) is much lower than
that observed in the mouse heart (30) and that the
number of binucleated cells in the mouse ES model was
lower than that observed in vivo (18).
The reproducible results observed in this study also
suggests that the human ES differentiating system
may be utilized as a possible in vitro surrogate system
to study the mechanisms involved in cardiomyocyte
DNA synthesis, reduplication, and cell-cycle regulation
as well as of molecular interventions aiming to modify
these processes. The ability to intervene in these processes may be of considerable therapeutic value given
the limited regenerative capacity of the adult heart. To
this end, it is interesting to note that the mouse ES
model has already been utilized to study the possible
effects of such interventions, taking advantage of its
ability to generate pure cardiomyocytes, its reproducible proliferative capacity, and the fact that it is readily
amenable to cotransfection with multiple expression
constructs (13, 26).
Another important implication of the current study
relates to the possible utilization of the human ES
cell-derived cardiomyocytes in future cell replacement
strategies. The proliferative capabilities of the ES cellderived cardiomyocytes may have a major impact on
the ultimate success of this strategy because it may
directly affect the number of cells available for transplantation as well as the size, survival, and possible
integration of the in vivo cell graft (27). Combining the
recently described human ES technology with gene
transfer strategies aiming to modify cell-cycle properties may provide a particular attractive approach for
improving the ultimate success of these newly emerging therapeutic paradigms.
Limitations of described model. Although the results
of the current study suggest that the human ES cardiomyocyte differentiating system may serve as a
unique in vitro model to study human cardiac tissue, it
also holds some limitations. First, cardiomyocytes, although being highly concentrated in a single area,
consist of only a minority of the cells within the entire
EB. In addition, based on data from the mouse model,
the ES cell-derived cardiomyocytes are expected to
represent a mixed group of atrial- and ventricular-like
cells (12). These limitations may be partially overcome
in the future by the use of double labeling with well
characterized antibodies or by establishing genetically
modified ES clones expressing a selectable marker
(antibiotic resistance gene, green fluorescent protein,
etc.) under the control of a cardiac-specific promoter to
generate pure cardiomyocyte cultures (19).
The second limitation of the model stems from its in
vitro nature, which may be different from the in vivo
settings, and may thus influence ultrastructural mat-
PROPERTIES OF HUMAN ES-DERIVED CARDIOMYOCYTES
The authors thank Pessia Shentzer for valuable technical support.
DISCLOSURES
This study was supported in part by Israel Science Foundation
Grant 520/01, the Johnson & Johnson focused given grant, and the
Nahum Guzik Research fund.
REFERENCES
1. Amit M, Carpenter MK, Inokuma MS, Chiu CP, Harris CP,
Waknitz MA, Itskovitz-Eldor J, and Thomson JA. Clonally
derived human embryonic stem cell lines maintain pluripotency
and proliferative potential for prolonged periods of culture. Dev
Biol 227: 271–278, 2000.
2. Anversa P and Kajstura J. Ventricular myocytes are not
terminally differentiated in the adult mammalian heart. Circ
Res 83: 1–14, 1998.
3. Bishop SP, Anderson PG, and Tucker DC. Morphological
development of the rat heart growing in oculo in the absence of
hemodynamic work load. Circ Res 66: 84–102, 1990.
4. Boheler KR, Czyz J, Tweedie D, Yang HT, Anisimov SV,
and Wobus AM. Differentiation of pluripotent embryonic stem
cells into cardiomyocytes. Circ Res 91: 189–201, 2002.
5. Brook WH, Connell S, Cannata J, Maloney JE, and Walker
AM. Ultrastructure of the myocardium during development from
early fetal life to adult life in sheep. J Anat 137: 729–741, 1983.
6. Burton PB, Raff MC, Kerr P, Yacoub MH, and Barton PJ.
An intrinsic timer that controls cell-cycle withdrawal in cultured
cardiac myocytes. Dev Biol 216: 659–670, 1999.
7. Chacko KJ. Observations on the ultrastructure of developing
myocardium of rat embryos. J Morphol 150: 681–709, 1976.
8. Clubb FJ Jr and Bishop SP. Formation of binucleated myocardial cells in the neonatal rat. An index for growth hypertrophy. Lab Invest 50: 571–577, 1984.
9. Evans MJ and Kaufman MH. Establishment in culture of
pluripotential cells from mouse embryos. Nature 292: 154–156,
1981.
10. Gregorio CC and Antin PB. To the heart of myofibril assembly. Trends Cell Biol 10: 355–362, 2000.
11. Guan K, Furst DO, and Wobus AM. Modulation of sarcomere
organization during embryonic stem cell-derived cardiomyocyte
differentiation. Eur J Cell Biol 78: 813–823, 1999.
12. Hescheler J, Fleischmann BK, Lentini S, Maltsev VA, Rohwedel J, Wobus AM, and Addicks K. Embryonic stem cells: a
model to study structural and functional properties in cardiomyogenesis. Cardiovasc Res 36: 149–162, 1997.
13. Huh NE, Pasumarthi KB, Soonpaa MH, Jing S, Patton B,
and Field LJ. Functional abrogation of p53 is required for T-Ag
induced proliferation in cardiomyocytes. J Mol Cell Cardiol 33:
1405–1419, 2001.
AJP-Heart Circ Physiol • VOL
14. Huttenbach Y, Ostrowski ML, Thaller D, and Kim HS. Cell
proliferation in the growing human heart: MIB-1 immunostaining in preterm and term infants at autopsy. Cardiovasc Pathol
10: 119–123, 2001.
15. Kehat I, Gepstein A, Spira A, Itskovitz-Eldor J, and Gepstein L. High-resolution electrophysiological assessment of human embryonic stem cell-derived cardiomyocytes: a novel in
vitro model for the study of conduction. Circ Res 91: 659–661,
2002.
16. Kehat I, Kenyagin-Karsenti D, Snir M, Segev H, Amit M,
Gepstein A, Livne E, Binah O, Itskovitz-Eldor J, and Gepstein L. Human embryonic stem cells can differentiate into
myocytes with structural and functional properties of cardiomyocytes. J Clin Invest 108: 407–414, 2001.
17. Kim HD, Kim DJ, Lee IJ, Rah BJ, Sawa Y, and Schaper J.
Human fetal heart development after mid-term: morphometry
and ultrastructural study. J Mol Cell Cardiol 24: 949–965, 1992.
18. Klug MG, Soonpaa MH, and Field LJ. DNA synthesis and
multinucleation in embryonic stem cell- derived cardiomyocytes.
Am J Physiol Heart Circ Physiol 269: H1913–H1921, 1995.
19. Klug MG, Soonpaa MH, Koh GY, and Field LJ. Genetically
selected cardiomyocytes from differentiating embronic stem cells
form stable intracardiac grafts. J Clin Invest 98: 216–224, 1996.
20. Legato MJ. Sarcomerogenesis in human myocardium. J Mol
Cell Cardiol 1: 425–437, 1970.
21. MacCallum DE and Hall PA. The biochemical characterization of the DNA binding activity of pKi67. J Pathol 191: 286–
298, 2000.
22. Manasek FJ. Embryonic development of the heart. I. A light
and electron microscopic study of myocardial development in the
early chick embryo. J Morphol 125: 329–365, 1968.
23. Martin G. Isolation of a pluripotent cell line from early mouse
embryos cultured in medium conditioned by teratocarcinoma
stem cells. Proc Natl Acad Sci USA 78: 7635, 1981.
24. Olivetti G, Cigola E, Maestri R, Corradi D, Lagrasta C,
Gambert SR, and Anversa P. Aging, cardiac hypertrophy and
ischemic cardiomyopathy do not affect the proportion of mononucleated and multinucleated myocytes in the human heart. J
Mol Cell Cardiol 28: 1463–1477, 1996.
25. Pasumarthi KB and Field LJ. Cardiomyocyte cell cycle regulation. Circ Res 90: 1044–1054, 2002.
26. Pasumarthi KB and Field LJ. Cardiomyocyte enrichment in
differentiating ES cell cultures: strategies and applications.
Methods Mol Biol 185: 157–168, 2002.
27. Reinlib L and Field L. Cell transplantation as future therapy
for cardiovascular disease? A workshop of the National Heart,
Lung, and Blood Institute. Circulation 101: E182-E187, 2000.
28. Reubinoff BE, Pera MF, Fong CY, Trounson A, and
Bongso A. Embryonic stem cell lines from human blastocysts:
somatic differentiation in vitro. Nat Biotechnol 18: 399–404,
2000.
29. Scholzen T and Gerdes J. The Ki-67 protein: from the known
and the unknown. J Cell Physiol 182: 311–322, 2000.
30. Soonpaa MH and Field LJ. Assessment of cardiomyocyte DNA
synthesis during hypertrophy in adult mice. Am J Physiol Heart
Circ Physiol 266: H1439–H1445, 1994.
31. Soonpaa MH and Field LJ. Survey of studies examining
mammalian cardiomyocyte DNA synthesis. Circ Res 83: 15–26,
1998.
32. Soonpaa MH, Kim KK, Pajak L, Franklin M, and Field LJ.
Cardiomyocyte DNA synthesis and binucleation during murine
development. Am J Physiol Heart Circ Physiol 271: H2183–
H2189, 1996.
33. Thomson JA, Itskovitz-Eldor J, Shapiro SS, Waknitz MA,
Swiergiel JJ, Marshall VS, and Jones JM. Embryonic stem
cell lines derived from human blastocysts. Science 282: 1145–
1147, 1998.
34. Westfall MV, Pasyk KA, Yule DI, Samuelson LC, and
Metzger JM. Ultrastructure and cell-cell coupling of cardiac
myocytes differentiating in embryonic stem cell cultures. Cell
Motil Cytoskeleton 36: 43–54, 1997.
35. Xu C, Police S, Rao N, and Carpenter MK. Characterization
and enrichment of cardiomyocytes derived from human embryonic stem cells. Circ Res 91: 501–508, 2002.
285 • DECEMBER 2003 •
www.ajpheart.org
Downloaded from http://ajpheart.physiology.org/ by 10.220.33.5 on June 14, 2017
uration and cell cycle activity. Some properties that
may differ in the in vitro settings include the different
availability of cell nutrients within the contracting EB,
the presence of adjacent nonmyocyte proliferating
cells, the lack of possible paracrine, humoral, neuronal,
and other in vivo signals, and the lack of hemodynamic
load.
Despite these limitations, the current study provides
the first characterization of the developmental changes
in the cell cycle activity and ultrastructural properties
during the in vitro differentiation and maturation of
human ES cell-derived cardiomyocytes. Our results
demonstrate a reproducible temporal pattern of early
cardiomyocyte proliferation, cell cycle withdrawal, and
ultrastructural maturation in this model. The results
of the current study may have important implications
for future cell replacement and tissue engineering
strategies using these unique cells and for the in vitro
modeling of human cardiac tissue.
H2363