Journal of the American College of Cardiology
© 2008 by the American College of Cardiology Foundation
Published by Elsevier Inc.
Vol. 51, No. 9, 2008
ISSN 0735-1097/08/$34.00
doi:10.1016/j.jacc.2007.11.040
PRECLINICAL RESEARCH
Pre-Treatment of Mesenchymal Stem Cells With
a Combination of Growth Factors Enhances Gap Junction
Formation, Cytoprotective Effect on Cardiomyocytes,
and Therapeutic Efficacy for Myocardial Infarction
Joo-Yong Hahn, MD, PHD,*† Hyun-Ju Cho, MS,* Hyun-Jae Kang, MD, PHD,*†
Tack-Seung Kim, MS,‡ Mi-Hyung Kim, PHD,‡ Jung-Hwa Chung, MD,* Jang-Whan Bae, MD, PHD,*
Byung-Hee Oh, MD, PHD,*† Young-Bae Park, MD, PHD,*† Hyo-Soo Kim, MD, PHD*†
Seoul, Republic of Korea
Objectives
The goal of this study was to investigate the effect of pre-treatment of mesenchymal stem cells (MSCs) with
growth factors (GFs) on cardiomyogenic differentiation, cytoprotective action on cardiomyocytes (CMCs), and
their therapeutic efficacy in myocardial infarction.
Background
Mechanisms of myocardial repair with MSC transplantation have not been fully elucidated, and therapeutic efficacy needs to be enhanced.
Methods
The MSCs obtained from the bone marrow of Fisher344 rats were treated with fibroblast growth factor-2, insulinlike growth factor-1, and bone morphogenetic protein-2. The expression of cardiac specific markers and the cytoprotective effect of MSCs with its mechanism were evaluated. Efficacy of MSCs transplantation was studied in
rat myocardial infarction model.
Results
Treatment of MSCs with cocktails of GFs enhanced expression of cardiac transcription factors and survival. Induction of cardiac specific markers by coculture with CMCs and gap junctional communication with CMCs was
more active in GF-treated MSCs than untreated MSCs. The GF-treated MSCs reduced apoptosis of neighboring
CMCs in a hypoxic condition and enhanced the phosphorylated Akt and phosphorylated c-AMP response element
binding protein expression of CMCs, which was markedly reduced by gap junction blockade. In a rat myocardial
infarction model, transplantation of GF-treated MSCs resulted in smaller infarct size and better cardiac function than
transplantation of untreated MSCs. Additionally, GF treatment enhanced gap junction formation of transplanted MSCs,
which did not aggravate arrhythmia.
Conclusions
Pre-treatment of MSCs with GFs enhanced cytoprotective effects on neighboring CMCs through gap junction and
improved the therapeutic efficacy of MSC transplantation for myocardial repair. “Priming of MSCs with GFs” before transplantation might improve the therapeutic efficacy of cell therapy. (J Am Coll Cardiol 2008;51:
933–43) © 2008 by the American College of Cardiology Foundation
Mesenchymal stem cells (MSCs) have been reported to
repair damaged myocardium and improve cardiac function
after myocardial infarction (MI) in pre-clinical studies (1,2).
Bone marrow cells containing MSCs were most widely
From the *National Research Laboratory for Cardiovascular Stem Cells, Seoul
National University College of Medicine, Seoul, Republic of Korea; †Innovative
Research Institute for Cell Therapy, Seoul National University Hospital, Seoul,
Republic of Korea; and ‡Anterogen Corp., Seoul, Republic of Korea. This study was
supported by a grant from Stem Cell Research Center (SC3150) and from Innovative
Research Institute for Cell Therapy, Republic of Korea. Drs. Hahn and Cho
contributed equally to this work.
Manuscript received August 20, 2007; revised manuscript received October 26,
2007, accepted November 27, 2007.
studied in clinical trials; however, their therapeutic benefit
needs to be improved (3,4). Although the underlying
mechanisms have not been fully elucidated, the differentiation of MSCs into cardiomyocytes (CMCs) might explain,
at least partly, their therapeutic effect. Additionally, MSCs
can exert cytoprotective effects on CMCs in a paracrine way.
A substantial portion of the salutary effects of MSC implantation might be attributable to salvage of endangered
ischemic myocardium (5). Although MSCs secrete large
amounts of angiogenic and anti-apoptotic factors (6), the
mechanisms of these cytoprotective effects are not fully understood. Gap junctions allow exchange of small molecules,
including secondary messengers between adjacent cells (7), and
934
Hahn et al.
Mesenchymal Stem Cells and Growth Factors
might be a potential route for
MSCs to exert cytoprotective
effects.
BMP ⴝ bone
To improve efficacy of MSC
morphogenetic protein
transplantation,
interventions that
CMC ⴝ cardiomyocyte
facilitate differentiation and enDMEM ⴝ Dulbecco’s
hance the cytoprotective effects
Modified Eagle Medium
of MSC can be a rational apFACS ⴝ fluorescenceproach. Recently, the role of
activated cell sorter
growth factors (GFs) in the difFBS ⴝ fetal bovine serum
ferentiation of stem cells into
FGF ⴝ fibroblast growth
CMCs has been studied, infactor
cluding fibroblast growth factor
FS ⴝ fractional shortening
(FGF)-2, insulin-like growth facGF ⴝ growth factor
tor (IGF)-1, and bone morphogeIGF ⴝ insulin-like growth
netic protein (BMP)-2 (8 –11).
factor
Combination of these GFs might
LV ⴝ left
facilitate differentiation of MSCs
ventricle/ventricular
into CMCs. Furthermore, many
LVEDD ⴝ left ventricular
GFs stimulate an anti-apoptotic
end-diastolic dimension
signal, and FGF-2 and IGF-1
LVESD ⴝ left ventricular
have been reported to increase the
end-systolic dimension
expression of connexin-43 (12,13).
MEF ⴝ myocyte enhancer
Pre-treatment of MSCs with GFs
factor
might enhance transfer of antiMI ⴝ myocardial infarction
apoptotic signal to CMCs.
MSC ⴝ mesenchymal stem
In this study, we investigated
cell
whether treatment of MSCs with
RT-PCR ⴝ reverse
a combination of GFs (FGF-2,
transcriptase-polymerase
chain reaction
IGF-1, and BMP-2) enhances
VF ⴝ ventricular fibrillation
cardiomyogenic differentiation
and the cytoprotective effect of
MSCs, elucidating the potential
role of gap junction in cytoprotective action of MSCs.
Furthermore, to confirm the therapeutic applicability of
GF-primed MSCs, we examined therapeutic efficacy of
GF-treated MSCs in improving myocardial repair after
transplantation in a rat MI model.
Abbreviations
and Acronyms
Methods
Animal care. All animal experiments were performed under
approval from the Institutional Animal Care and Use Committee of Seoul National University Hospital and complied
with the National Research Council’s “Guidelines for the Care
and Use of Laboratory Animals” (revised 1996).
Isolation and culture of MSCs from rat bone marrow. The MSCs were isolated from the femur of 8-weekold male Fischer 344 rats (200 to approximately 250 g,
Daehan Biolink Co., Chungbuk, Korea). Bone marrow
plugs were extracted and suspended in MSC culture medium (Dulbecco’s Modified Eagle Medium [DMEM, Invitrogen, Carlsbad, California], 10% fetal bovine serum
[FBS, Hyclone, Logan, Utah], ascorbic acid [Sigma, St.
Louis, Missouri], dexamethasone [Sigma], and mouse leu-
JACC Vol. 51, No. 9, 2008
March 4, 2008:933–43
kemia inhibitory factor [Sigma]) and then incubated at
37°C for 3 days before the first medium change. The
mesenchymal population was isolated on the basis of its
ability to adhere to the culture plate (14). At 90% confluence, the cells were trypsinized and subcultured. Second
passage MSCs were used in all experiments. The MSCs
were labeled with vibrant 1.1=-dioctadecyl-3,3,3=,3=tetramethylindocarbocyanine perchlorate (DiI, Molecular
Probes, Eugene, Oregon) or with 4=,6-diamino-2phenylindole (DAPI, Sigma) for detection and labeled with
vibrant DiO (Molecular Probes) for separation after coculture with CMCs.
Isolation of neonatal CMCs. The CMCs were isolated
from the hearts of neonatal Fisher 344 rats as previously
described with minor modifications (15). Briefly, 2- to
4-day-old rats were killed, and then hearts were excised.
Ventricles were minced and trypsinized, and supernatant
was obtained. And then CMCs were separated, placed in
serum-free DMEM for 24 h, and then cultured in 10%
FBS-DMEM.
Viability assay of MSCs. Viability of MSCs was measured
after exposure to 0.5% hypoxia for 30 h with trypan blue
exclusion assay.
Differentiation of MSCs into CMCs: treatment with
GFs and coculture with CMCs. To define optimal treatment condition, we tested differentiation efficacy of BMP-2,
FGF-2, IGF-1, and their combination. First, differentiation
efficacy was measured by degree of myocyte enhancer factor
(MEF)-2 expression (MEF2-positive cells/total MSCs) on
immunofluorescent staining with anti-MEF2 (1:200, Santa
Cruz Biotechnology, Santa Cruz, California). Treatment of
MSCs with a combination of BMP-2, FGF-2, and IGF-1
resulted in the higher percentage of MEF2-positive cells
compared with BMP-2 or FGF-2 or IGF-1 alone or any
combination of 2 GFs. Expression of MEF2 reached a peak
at day 7 and then showed a plateau (Online Figs. I and II).
Second, in western blot analysis, expression of CMCspecific markers such as Nkx2.5, GATA4, and cTnI was
stronger with combination of BMP-2, FGF-2, and IGF-1
compared with BMP-2 or FGF-2 or IGF-1 alone or any
combination of 2 GFs (Online Fig. III). Again, expression
of CMC-specific markers increased until day 7 and then
showed a plateau. On the basis of the results of these
experiments, differentiation was induced by 7 days’ incubation with differentiation media (DMEM supplemented
with 2% FBS, ascorbic acid, and dexamethasone) and after
2 days’ coculture with CMCs. For GF treatment, FGF-2,
IGF-1, and BMP-2 (all from R&D Systems, Minneapolis,
Minnesota) were added to the differentiation medium. The
concentration was 50 ng/ml for FGF-2, 2 ng/ml for IGF-1,
and 10 ng/ml for BMP-2. To evaluate the expression of
cardiac transcription factors and cardiac specific markers,
immunoblot analysis was performed at 1, 3, 5, and 7 days
after differentiation induction. After coculture with CMCs,
immunofluorescence staining was performed to define the
phenotype of MSCs with anticardiac troponin I (Santa
JACC Vol. 51, No. 9, 2008
March 4, 2008:933–43
Cruz Biotechnology) and anticonnexin-43 (Santa Cruz
Biotechnology), which were detected with goat antirabbit
IgG antibodies (Molecular Probes) conjugated with fluorescein isothiocyanate. Meanwhile, in another experiment,
DiO-labeled MSCs were separated with fluorescenceactivated cell sorter (FACS) from CMCs by detecting DiO
in MSCs. Immunoblot was performed as previously described (16). The MSCs were harvested in lysis buffer.
Protein was separated on SDS-polyacrylamide electrophoresis gel and transferred to a polyvinylidene fluoride
membrane (Millipore, Billerica, Massachusetts). The membrane was blocked and incubated with primary antibodies
against GATA-4, NKx-2.5, cardiac troponin-I, connexin43, and antialpha tubulin (all from Oncogene, Cambridge,
Massachusetts) were used at a dilution of 1:500. As secondary
antibody, antimouse IgG HRP (Promega, Madison, Wisconsin) or antirabbit IgG HRP (Promega) was used at a dilution
of 1:2,500. A chemiluminescent detection reagent ECL (Amersham, Piscataway, New Jersey) was used for detection. To
characterize the cell type of MSCs after GF treatment and
coculture with CMCs, reverse transcriptase polymerase chain
reaction (RT-PCR) was done as previously described (17).
Total ribonucleic acid (RNA) was isolated by Trizol (Invitrogen) method and reverse transcribed with reverse transcription system (Clontech, Mountain View, California),
and complementary deoxyribonucleic acid (cDNA) was
amplified with CMC or myocyte-specific markers
(connexin-43, cardiac troponin-I, alpha- and beta-MHC,
alpha-sarcomeric actin, and atrial natriuretic peptide) and
glyceraldehyde 3-phosphate dehydrogenase (GAPDH) by
PCR. Primers used were as follows:
Forward primer 5=-TCCTTGGTGTCTCTCGCTTT-3=
and reverse primer 5=-GTTCACCCAATGCGACTCTT-3= for connexin-43
Forward primer 5=-CACCCTTCTAAGACCCTCCA-3=
and reverse primer 5=-CCTCCTTCTTCACCTGCTTG-3= for cardiac troponin-I
Forward primer 5=-CATAGGGGACCGTAGCAAGA-3=
and reverse primer 5=-CTGCCCCTTGGTGACATACT-3= for alpha-MHC
Forward primer 5=-GCACTGGCCAAGTCAGTGTA-3=
and reverse primer 5=-GGACACGGTCTGAAAGGATG-3= for beta-MHC
Forward primer 5=-GACCACAGCTGAACGTGAGA-3=
and reverse primer 5=-CATAGCACGATGGTCGATTG-3= for alpha-sarcomeric actin
Forward primer 5=-AGGCCATATTGGAGCAAATC-3=
and reverse primer 5=-CCTTAATATGCAGAGTGGGAGA-3= for atrial natriuretic peptide
Forward primer 5=-CGTGGAAGGACTCATGAC-3=
and reverse primer 5=-CAAATTCGTTGTCATACCAG-3= for GAPDH
Hahn et al.
Mesenchymal Stem Cells and Growth Factors
935
Intercellular communication between MSCs and CMCs
through gap junction: dye transfer experiment. The
MSCs were labeled with DiI, which cannot pass through
the gap junction, and calcein-AM (2.5 ␮mol/l; Calbiochem,
San Diego, California), which can spread only through the
gap junction. And then labeled MSCs were cocultured with
unlabeled CMCs. At 48 h after coculture, dye transfer was
evaluated under fluorescence microscope. To selectively
block gap junctional communication, heptanol (0.5 mmol/
ml) was used.
Hypoxia exposure to CMCs with or without MSCs
coculture. Hypoxic conditions were created by incubating
CMCs at 37°C in airtight chambers with 0.5% oxygen
concentration under coculture with DiI-labeled MSCs.
After 24-h hypoxic exposure, cocultured MSCs were eliminated by FACS. Then, apoptosis of CMCs was evaluated
by flow cytometry (FACStar plus, Becton Dickinson,
Franklin Lakes, New Jersey) as described previously (18).
The CMCs exposed to hypoxia without coculture were used
as a control. To evaluate the role of gap junction in the
cytoprotective effect of MSCs, heptanol (0.5 mmol/ml) was
added during hypoxia. To investigate underlying mechanism of cytoprotective effect, the expression of phosphorylated Akt and phosphorylated c-AMP response element
binding protein (CREB) in CMCs were evaluated with
immunoblot analysis.
Assay of IGF-1 and Hepatocyte GF for evaluating
paracrine function of MSCs. After 7 days’ incubation of
MSCs in differentiation medium followed by 30 h of
incubation in conditioned medium (low glucose DMED
with 10% FBS), IGF-1 and hepatocyte growth factor
(HGF) levels in conditioned medium were measured by
enzyme-linked immunosorbent assay (rat IGF-1 enzyme
immunoassay, Diagnostic Systems Laboratory, Webster,
Texas; and rat HGF enzyme immunoassay, Institute of
Immunology, Tokyo, Japan). To exclude the possibility of
cross reaction, antirat IGF-1 antibody that has no reactivity
to human IGF-1 was used for this assay.
Rat MI model and MSCs transplantation. Rats were
anesthetized with ketamine hydrochloride (100 mg/kg,
Yuhan Corp., Seoul, Korea) and xylazine (10 mg/kg, Bayer,
Shawnee Mission, Kansas) by intraperitoneal injection.
Experimental MI was induced by temporary ligation of the
left anterior descending artery for 45 min, as previously
described (19). Development of MI was confirmed by
echocardiography 7 days after the procedure. Then rats were
randomized into 3 groups: rats receiving control media only
(control group), untreated MSCs (untreated MSC group),
and GF-treated MSCs (treated MSC group). Before transplantation, MSCs were labeled with red DiI for detection at
later time points. Under anesthesia, each rat received 3
injections (total of 106 cells/heart) into the border zone of
MI, approximately 1 to 2 mm apart.
Functional assessment of the infarcted rat heart by
echocardiography. Transthoracic echocardiography was
performed at just before coronary ligation and 1, 4, and 8
936
Hahn et al.
Mesenchymal Stem Cells and Growth Factors
weeks after coronary ligation, as described previously (19)
with an echocardiographic system (Acuson 128-XP system,
Acuson, Florida) equipped with a 7-MHz linear–array
transducer. Left ventricular end-diastolic and end-systolic
dimensions (LVEDD and LVESD) were measured according to the leading-edge method of the American
Society of Echocardiography. The LV percent fractional
shortening (%FS) was calculated as 100 ⫻ (LVEDD ⫺
LVESD)/LVEDD.
Induction of arrhythmia. Arrhythmia was induced by
aconitine (Sigma) infusion. At 8 weeks after coronary
ligation, rats were anesthetized and 24-gauge intravenous
catheter was inserted into the femoral vein for drug administration. After a 30-min period of stabilization, aconitine
was intravenously infused at a rate of 0.1 ml/min by infusion
pump for 510 s. After aconitine infusion, rats were observed
for 20 min. Electrocardiography (ECG, Surgivet, Waukesha, Wisconsin) was continuously recorded throughout the
experiment. The incidence of arrhythmias was analyzed in
accordance with the Lambeth Conventions (20). An arrhythmia score was used to indicate the incidence and
duration of arrhythmias, where 0 indicates no arrhythmia; 1:
an arrhythmia duration of ⬍10 s; 2: an arrhythmia duration
of 11 to 30 s; 3: an arrhythmia duration of 91 to 180 s or
Figure 1
JACC Vol. 51, No. 9, 2008
March 4, 2008:933–43
reversible ventricular fibrillation (VF) for ⬍10 s; 5: an
arrhythmia duration longer than 180 s or reversible VF for
more than 10 s; and 6: an irreversible VF or death of the
animal.
Histological analysis of the infarcted rat heart with or
without MSCs transplantation. Rats were killed at 8
weeks after coronary ligation for the assessment. Heart
sections embedded in paraffin were cut into 4-␮m slices,
and the size of area of fibrosis was determined by image
analysis system (Image Pro version 4.5; MediaCybernetics,
Bethesda, Maryland) on Masson’s trichrome-stained slides.
The measurements of area of fibrosis were performed on 2
separate sections of each heart, and the averages were used
for statistical analysis. DiI-positive area was measured with
automated color detection software (Image Pro version 4.5,
MediaCybernetics). For immuno-histochemical analysis,
the rat hearts were cryopreserved in OCT compound
(Tissue-Teck, Sakura, Torrance, California). To evaluate
apoptosis, the terminal deoxynucleotidyl transferasemediated dUTP nick end-labeling (TUNEL) assay
(Chemicon S7100 kit, Chemicon, Temecula, California)
and immunohistochemistry for caspase-3 (Abcam 1:50,
Abcam, Cambridge, Massachusetts) were performed. The
TUNEL or caspase-3 positive cells were counted in 10
Pre-Treatment of MSCs With GFs Augmented the Expression of Cardiac
Specific Genes and Connexin-43 in MSCs, Which Was Induced by Coculture With CMCs
(A) Reverse transcriptase-polymerase chain reaction. In the single mesenchymal stem cell (MSC) culture condition, treatment of MSCs with cocktails of growth factors
(GFs) does not affect the expression of cardiac specific genes except for the alpha-sarcomeric actin. But in coculture condition with cardiomyocytes (CMCs), treatment of
GFs significantly augmented the induction of cardiac specific genes in MSCs, such as atrial natriuretic peptide, connexin-43 (Cx43), cardiac troponin-I (cTnI), alpha and
beta myosin heavy chain (MHC) genes. (B) Western blot analysis in the single MSC culture condition. The expression of GATA-4 and NKx-2.5 were distinctively induced
only in GF-treated MSCs. (C) Western blot analysis after 2 days of coculture with CMCs. The GF treatment augmented the induction of cTnI and Cx43 protein expression
in MSC by coculture with CMCs. (D) The percentage of MSCs expressing Cx43 was significantly higher in the GF-treated MSCs than in naïve ones. Data were derived
from 5 randomly chosen microscopic fields of 3 separate experiments. (E) Representative images after 2 days of coculture of MSCs with CMCs. The MSCs were labeled
with DAPI (blue). The CMCs were stained with TnI (red) without DAPI. Connexin-43 (green) was expressed more frequently in the treated MSCs compared with the
untreated ones. Connexin-43, which indicates the formation of gap junction (arrows), was expressed linearly between MSCs and CMCs (TnI positive cells without DAPI).
Arrowheads indicate MSCs expressing both TnI (red) and Cx43. *The MSCs expressing only Cx43 (magnification ⫻200). T ⫽ treated; UT ⫽ untreated.
JACC Vol. 51, No. 9, 2008
March 4, 2008:933–43
different microscopic fields of at least 3 different sections
from each animal. Immunofluorescence staining was performed with anticardiac troponin I (Santa Cruz Biotechnology) or anticonnexin-43 (Santa Cruz Biotechnology),
which was detected with goat antirabbit IgG antibodies
(Molecular Probes) conjugated with green fluorescein isothiocyanate. Blue DAPI staining was done to localize the
nucleus. Colocalization of the red DiI-labeled MSCs and
green connexin-43 or cardiac troponin I expression were
examined with a confocal microscope (LSM 510 META,
Carl Zeiss, Peabody, Massachusetts).
Statistical analysis. All data are presented as mean ⫾
SD. Continuous variables were compared by the Student
t test, and multiple comparisons were performed by
analysis of variance with a Bonferroni correction. A value
of p ⬍ 0.05 was considered significant, and all analyses
were performed with SPSS version 11.0 (SPSS, Chicago,
Illinois).
Results
Pre-treatment of MSCs with GFs potentiated their
viability in hypoxic condition. The effects of pretreatment with GFs combination of FGF-2, IGF-1, and
BMP-2 on the viability of MSCs were evaluated. Viability
of MSCs was measured after exposure to 0.5% hypoxia for
30 h. The GF-treated MSCs showed better survival than
untreated MSCs under hypoxia (viable cell proportion in
Figure 2
Hahn et al.
Mesenchymal Stem Cells and Growth Factors
937
trypan blue exclusion assay: 67 ⫾ 11% vs. 47 ⫾ 14%, p ⬍
0.05).
Pre-treatment of MSCs with GFs augmented the expression of various cardiac specific markers and connexin-43,
which was induced by coculture with CMCs. We evaluated the effects of GF treatment on differentiation of MSCs
into CMCs. After 7-day treatment with GFs, RT-PCR
showed that expression of alpha-sarcomeric actin was higher
in GF-treated MSCs than untreated MSCs. However, in
the single MSCs culture condition, most of the markers did
not express irrespective of treatment with GFs. Although
connexin-43 expressed weakly in both GF-treated MSCs
and untreated ones, there was no significant difference between
the 2 groups. But, after coculture with CMCs, the messenger
RNA (mRNA) expression of CMC- or myocyte-specific
markers such as atrial natriuretic peptide, troponin-I, alpha
and beta myosin heavy chain as well as connexin-43 was
augmented in GF-treated MSCs compared with untreated
ones (Fig. 1A).
Immunoblot analysis showed that treatment with GFs
increased the expression of cardiac transcription factors.
The protein expression of GATA-4 and NKx-2.5 were
induced distinctively during 7 days of culture with GFs
although not without GFs (Fig. 1B). After an additional
2 days of coculture with CMCs, connexin-43 protein was
expressed, which was enhanced in GF-treated MSCs
compared with untreated ones (Figs. 1C and 1D). In
immunofluorescent staining, connexin-43 was expressed
Pre-Treatment of MSCs With GFs Enhanced the Formation of Functioning Gap Junction With CMCs
(A and B) The MSCs labeled with 1.1=-dioctadecyl-3,3,3=,3=-tetramethylindocarbocyanine perchlorate (DiI) (red) and calcein-AM (green) were cocultured with unlabeled
CMCs. Only green calcein-AM is transmittable through functioning gap junction. Cells having green fluorescence without red denote CMCs that are connected with labeled
MSCs through functioning gap junction. These cells were scarcely observed in the coculture with untreated MSCs (A), whereas they were frequently observed in the
coculture with GF-treated MSCs (B). (C) After addition of heptanol into coculture condition with treated MSC and CMCs, there are no CMCs having green fluorescence
alone, corroborating that calcein transfer was mediated through gap junction (magnification ⫻200). (D) Treatment with GFs significantly increased intercellular communication between MSCs and CMCs. Data were derived from 5 randomly chosen microscopic fields of 3 separate experiments. HPF ⫽ high power field; other abbreviations
as in Figure 1.
938
Hahn et al.
Mesenchymal Stem Cells and Growth Factors
linearly between MSCs and CMCs, which suggests the
formation of gap junction. It was more frequently observed
in the GF-treated MSCs (Fig. 1E). Although western blot
showed that the expression of cardiac troponin I was
increased in GF-treated MSCs, the frequency of troponin I
positive-cells in immunofluorescent staining was not significantly different between GF-treated and untreated MSCs.
Pre-treatment of MSCs with GFs enhanced gap junctional communication between MSCs and CMCs. To
evaluate gap junctional communication, we examined direct
dye transfer from MSCs to adjacent CMCs. We labeled
MSCs both with red DiI and green calcein-AM and
cocultured them with the unlabeled CMCs (Fig. 2). Only
calcein-AM, not DiI, is transmittable through gap junction.
We observed cells stained with green calcein-AM without
red DiI, which indicated CMCs that have taken up
calcein-AM via functioning gap junctions from MSCs.
Quantitative analysis showed that transfer of calcein-AM
from MSCs to the surrounding CMCs through the gap
junction was more frequent in coculture of CMCs with
GF-treated MSCs than with untreated MSCs (Fig. 2D).
After the addition of a gap junction blocker, calcein-AM
transfer was blocked (Fig. 2C). This suggested that MSCs
were linked to CMCs via functioning gap junctions. Therefore, pre-treatment with GFs enhanced not only the expression of connexin-43 in MSCs but also the gap junctional
communication with CMCs.
Figure 3
JACC Vol. 51, No. 9, 2008
March 4, 2008:933–43
GF-treated MSCs protect CMCs through a gap
junction-mediated mechanism. We investigated whether
GFs potentiated the cytoprotective effect of MSCs and
whether this effect is mediated through the gap junction
(Fig. 3). After 24 h of hypoxia, the apoptotic fraction of
CMCs was significantly reduced in coculture condition
with MSCs compared with CMCs single culture (Figs.
3A and 3B). Such antiapoptotic action of MSCs on
CMCs was significantly potentiated in the coculture with
GF-treated MSCs (Fig. 3C). Treatment of heptanol
partially inhibited antiapoptotic effects of treated MSCs
on CMCs (Fig. 3D). These data suggest that MSCs exert
a protective effect on CMCs via gap junctions, which is
augmented by GF treatment of MSCs. Additionally,
expression of phospho-Akt and phospho-CREB increased in CMCs cocultured with GF-treated MSCs
compared with untreated MSCs, which was also reversed
by heptanol (Fig. 3E). However, there was no significant
difference in the capacity to secrete IGF-1 or HGF
between untreated and treated MSCs (Fig. 3F). Collectively, these results suggest that, at least partially, cytoprotective effect of MSCs was mediated through gap
junction and that GF pre-treatment of MSCs potentiated
the gap junction-mediated cytoprotective effect of MSCs.
The increase of intercellular communication through gap
junction activated Akt and CREB pathway in CMCs.
Pre-Treatment of MSCs With GFs Potentiated
Their Cytoprotective Effects on CMCs, Which Was Mediated Through Gap Junction
(A to D) Fluorescence-activated cell sorter (FACS) for evaluating apoptosis. The CMCs were exposed to hypoxia alone as a control (A), under coculture with untreated
MSCs (B), and GF-treated MSCs (C). Hypoxia-induced apoptosis of CMCs was reduced by coculture with MSCs, which was more notable in coculture condition with GFtreated MSCs. Heptanol partially inhibited cytoprotective effects of GF-treated MSCs on CMCs (D). (E) Western blot analysis. The expression of phospho-Akt (p-Akt) and
phospho-c-AMP response element binding protein (p-CREB) increased in CMCs cocultured with GF-treated MSCs under hypoxia, which was reversed by gap junctional
blocker. (F) Enzyme-linked immunosorbent assay (ELISA) for insulin-like GF (IGF)-1 and hepatocyte GF (HGF). There was no significant difference in secretion of IGF-1 and
HGF between untreated and GF-treated MSCs. Abbreviations as in Figure 1.
Hahn et al.
Mesenchymal Stem Cells and Growth Factors
JACC Vol. 51, No. 9, 2008
March 4, 2008:933–43
Transplantation of GF–pre-treated MSCs showed a superior myocardial salvage of rat infarcted heart to that of
naïve ones. The area of fibrotic scar in the infarcted rat
heart was significantly smaller when GF-treated MSCs
were transplanted than when untreated MSCs were at 8
weeks after MI (Figs. 4A and 4B). Left ventricular %FS,
LVESD, and LVEDD were similar among all 3 groups at
baseline. At 8 weeks, the motion of the LV anterior wall was
obviously better when transplanted with the GF-treated
MSCs than with the untreated MSCs or with control media
(Fig. 4C). Although LVEDD was not significantly different
among 3 groups (Fig. 4D), %FS were significantly higher
when transplanted with the GF-treated MSCs than with
the untreated MSCs or with control media at 4 and 8 weeks
after MI (Fig. 4E).
Pre-treatment of MSCs with GFs enhanced engraftment
of MSCs and gap junction formation with CMCs and
reduced apoptosis in infarcted myocardium. Transplanted MSCs were detected until 8 weeks after MI in the
infarct border zone. At 2 and 8 weeks, the proportion of
MSCs expressing connexin-43 was higher in the heart
transplanted with GF–pre-treated MSC than with untreated ones. The gap junction formation was more fre-
Figure 4
939
quently observed between MSCs themselves or MSCs and
surrounding CMCs in the GF-treated MSC group compared with the untreated MSC group (Figs. 5A to 5C). In
addition to quantitative increase in the expression of
connexin-43, expression pattern was more organized in the
heart transplanted with GF–pre-treated MSC than with
untreated ones (Online Fig. IV). However, the expression of
troponin-I in MSC was similar in the treated MSC group
and untreated ones (Online Fig. V). Area of surviving
MSCs was significantly greater in the GF-treated MSC
group compared with the untreated group at 2 weeks (Figs.
5D and 5E). Apoptosis was significantly reduced in the
infarcted heart when transplanted with the GF-treated
MSCs than with the untreated ones at 2 weeks after MI (1
week after MSC implantation) (Figs. 5F to 5H). Taken
together, pre-treatment of MSCs with GFs enhances gap
junction formation and cell survival in the infarcted heart.
Transplantation of GF–pre-treated MSCs did not aggravate susceptibility to drug-induced arrhythmia after
MI. The arrhythmia induction experiment was performed
in 20 rats (10 rats transplanted with the GF-treated MSCs
and with untreated MSCs, respectively) at 8 weeks after MI.
During and after aconitine infusion, we were able to observe
Transplantation of GF–Pre-Treated MSCs Showed a Greater
Myocardial Salvage in Rat Infarcted Heart Than That of Untreated MSCs
(A) Representative images of infarcted rat heart. Heart obtained 8 weeks after myocardial infarction (MI), and transplantation of GF-treated MSC showed smaller infarct
than that with transplantation of untreated MSCs (Masson’s trichrome staining: blue indicates infarcted scar tissue). (B) The area of fibrosis was significantly smaller in
the infarcted heart transplanted with GF-treated MSCs compared with that with untreated MSCs (n ⫽ 7 in each group). (C) Representative figures of echocardiographic
finding. The anterior wall motion was preserved better in the GF-treated MSC group. (D and E) Although there was no significant difference at baseline, percent fractional
shortening (%FS) was higher in the group transplanted with GF-treated MSC than with untreated MSC or with media at 8 weeks after MI. Left ventricular end-diastolic
dimension (LVEDD) was significantly smaller in groups with MSC transplantation than with media at 4 weeks, but there was no significant difference at 8 weeks. The MI
with media injection (n ⫽ 8), MI with transplantation of untreated MSC (n ⫽ 10), and MI with GF-treated MSC (n ⫽ 9). *p ⬍ 0.05 versus media injection group; #p ⬍
0.05 versus untreated MSC group. Abbreviations as in Figure 1.
940
Figure 5
Hahn et al.
Mesenchymal Stem Cells and Growth Factors
JACC Vol. 51, No. 9, 2008
March 4, 2008:933–43
Treatment of MSCs With GFs Enhanced Their Engraftment and Gap
Junction Formation With CMCs and Reduced Apoptosis in the Infarcted Myocardium
(A and B) Representative figures of immunostaining at 2 and 8 weeks after MI. Transplanted MSCs, labeled with DiI (red) before injection and with DAPI (blue) at
nuclei, were engrafted in the myocardium (unlabeled CMCs with DAPI nuclear staining only). There was significantly higher connexin-43 expression in the infarcted
heart transplanted with the GF–pre-treated MSCs than with the untreated MSCs. Connexin-43 (green) was expressed between MSCs (arrows) or between MSCs and
CMCs (arrowheads). (C) The percentage of MSCs expressing connexin-43 was significantly higher in the GF–pre-treated MSCs than in the untreated ones. (D) DiIpositive area, which might reflect the engraftment area of transplanted MSCs, was significantly greater in the heart transplanted with GF–pre-treated MSCs than with the
untreated ones at 2 weeks after MI. The statistical significance was lost at 8 weeks. (E) Low-power views, which showed that DiI-positive area at 2 weeks was greater
in the infarcted myocardium transplanted with GF–pre-treated MSCs than with naïve ones. (F) Terminal deoxynucleotidyl transferase-mediated dUTP nick end-labeling
(TUNEL) staining. (G) Immunohistochemistry for caspase-3. At 2 weeks after MI, apoptosis (arrows) was significantly lower in the infarcted heart transplanted with GFpretreated MSCs than with untreated ones. (H) Three randomly selected fields in 2 separate slides/animal (n ⫽ 5 in each group) were counted. Abbreviations as in Figures 1, 2, and 5.
premature ventricular contraction (PVC), ventricular tachycardia, and VF (Fig. 6A). There were 5 rats in the
GF-treated MSC group and 8 in the untreated MSC group
showing any kind of arrhythmia. Ventricular tachycardia
was induced in 2 in the GF-treated MSC group and 3 in the
untreated group. One rat in each group died, owing to
irreversible VF. Rats transplanted with the treated MSCs
had a tendency toward a lower arrhythmia score and less
frequent PVCs, which did not reach statistical significance
(Figs. 6B and 6C). These data suggest that enhanced
expression of connexin-43 in MSCs and gap junctional
communication with CMCs by GF pre-treatment of MSCs
at least did not aggravate arrhythmia after transplantation of
MSCs to the infarcted myocardium.
Discussion
In this study, we showed that pre-treatment of MSCs with
a combination of GFs (FGF-2 ⫹ IGF-1 ⫹ BMP-2)
enhanced the expression of cardiac transcription factors and
that these primed MSCs, compared with naïve MSCs,
showed the augmented induction of cardiac specific genes
by coculture with CMCs. With the greater expression of
connexin-43 and gap junction formation between CMCs,
the primed MSCs showed better cytoprotective effects on
CMCs than naïve ones. Furthermore, the cytoprotective
effect on CMCs was mediated through gap junctions, which
is a novel mechanism to explain the therapeutic effect of
MSCs. In rat MI model, transplantation of GF–pre-treated
MSCs after MI enhanced gap junction formation and cell
survival in the infarcted myocardium and reduced infarct
size, leading to the improved LV function.
GF treatment and cardiomyogenic differentiation of
MSCs. Bone morphogenetic protein, a member of the
transforming growth factor (TGF) family, and FGF have
been reported to play important roles in induction of cardiac
transcription factors in developing embryos (21) and to
induce cardiomyogenic differentiation from noncardiac mesodermal cells (11,22). The BMP-2 in vivo induced expression of cardiac-specific transcription factors (8,11) and is
known to play an important role in the process of heart
JACC Vol. 51, No. 9, 2008
March 4, 2008:933–43
Figure 6
Pre-Treatment of MSCs With GFs Did Not Aggravate
Arrhythmogenecity of Transplanted MSCs
(A) Representative figures of electrocardiographic recording in rats. (B) The
arrhythmia score according to Lambeth Convention showed a tendency to
be lower in the infarcted heart transplanted with GF–pre-treated MSCs than in
the heart with the untreated MSCs (1.1 ⫾ 0.6 vs. 2.4 ⫾ 0.7, p ⫽ 0.09).
(C) The difference of premature ventricular complex counts between the GFtreated MSC and the untreated group did not reach statistical significance
(402 ⫾ 288 vs. 753 ⫾ 305, p ⫽ 0.11). PVC ⫽ premature ventricular contraction.; other abbreviations as in Figure 1.
development (23,24). The FGF-2 also plays a pivotal role in
the differentiation process of resident cardiac progenitor
cells into functional CMCs (10). The IGF-1 was shown to
enhance cellular engraftment and host organ-specific differentiation of embryonic stem cells after injection in the area
of acute myocardial injury (9) and to enhance survival and
proliferation of cardiac stem cells (25).
In the present study, we demonstrated that a combined
treatment with FGF-1, IGF-1, and BMP-2 increased the
expression of cardiac transcription factors in MSCs. The
expression of CMC-specific genes, however, was not induced by treatment with GFs in the MSCs single culture
condition. But in coculture condition with CMCs, GFtreated MSCs showed a greater expression of cardiac specific
genes including connexin-43 and a better gap junction formation with CMCs than naïve MSCs. We think that MSCs
after treatment with GFs are “primed” MSCs expressing
cardiac transcription factors or “committed” MSCs into
myocyte lineage but not mature or immature CMCs. But
these “primed” MSCs have a capability to differentiate to
CMC-like cells in coculture with CMCs in vitro, to form
gap junction, and to survive in the infarcted myocardium in
vivo more than naïve MSCs. Our data suggested that GFs
Hahn et al.
Mesenchymal Stem Cells and Growth Factors
941
have the potential to induce MSCs to express CMCspecific genes and connexin-43 and to form functioning gap
junction, in response to appropriate stimuli such as coculture with CMCs. Although the underlying mechanism
responsible for increased expression of connexin-43 by GFs
in MSCs remains unclear, FGF-2 and IGF-1 have been
shown to increase connexin-43 expression and intercellular
communication (12,13).
Our data suggest that contact with CMCs would be
needed to differentiate MSCs into CMC lineage. It was
reported that differentiation of bone marrow stromal cells
into cells with cardiac phenotype requires intercellular
communication with myocytes (26). However, whether
intercellular communication is essential for cardiomyogenic
differentiation is unknown. Li et al. (27) reported that
TGF-beta induced the myogenic differentiation of stem
cells without coculture.
Gap junction and cytoprotective effect. In this study,
GF–pre-treated MSCs exerted better protective effects on
CMCs than untreated MSCs, and these protective effects
were mediated through gap junctions. Currently, the cytoprotective effect of MSCs is regarded as one of the mechanisms to explain the therapeutic effects of MSCs. Previous
studies have focused on paracrine effects of MSCs. The
MSCs secrete large amounts of various angiogenic and
anti-apoptotic factors (6), and the conditioned medium
from cultured MSCs were shown to reduce apoptosis of
CMCs exposed to hypoxia (5). We also observed that
MSCs secreted IGF-1 and HGF in this study. However,
the capacity to secrete cytokines is not different between
GF-treated and untreated MSCs in this study. Therefore,
paracrine effects can not explain the difference in cytoprotective effects between GF-treated and untreated MSCs.
In this study, formation of functional gap junctions was
more efficient in GF-treated MSCs compared with untreated MSCs, and the blockade of gap junctions reversed
the protective effects of MSCs on CMCs and abolished the
difference in protective effects on CMCs between GFtreated and untreated MSCs. These data suggest that the
cytoprotective effects of MSCs are mediated not only by
paracrine actions but also by direct cell-to-cell communication through gap junctions and that treatment with GFs
might potentiate gap junction-mediated cytoprotective effect. Recent data support the hypothesis that gap junction
might propagate cell survival and death signals (28). In our
study, connexin-43 expression that indicated the formation
of gap junction was more prominent in GF-treated MSC
group in vitro and vivo. And dye transfer experiments
demonstrated that metabolic coupling with CMCs was
more efficient in GF-treated MSCs than in untreated
MSCs. Although we could not confirm which material was
transferred from MSCs to CMCs, second messengers,
which have been known to pass through the gap junction,
are potential candidates. Cyclic adenosine monophosphate
is a well-known second messenger that can pass through gap
junction (7) and phosphorylate CREB, which in turn exerts
942
Hahn et al.
Mesenchymal Stem Cells and Growth Factors
an antiapoptotic effect (29). In the present study, we showed
increased phosphorylation of CREB in CMCs cocultured
with treated MSCs. Moreover, the Akt pathway was activated in CMCs cocultured with treated MSCs. Enhanced
expression of CREB and Akt in CMCs cocultured with
GF-treated MSCs were inhibited by blocking gap junction.
In short, MSCs have a cytoprotective effect on CMCs
through the gap junction, which is potentiated by treatment of
MSCs with GFs that facilitate the expression of connexin-43
and formation of functional gap junctions.
Reduced infarct size and improved systolic function by
transplantation of GF–pre-treated MSCs. There are several possible mechanisms that could explain the difference
between GF-treated MSCs and untreated ones in reduction
of infarct size and preservation of LV systolic function.
First, GF treatment might make MSCs stronger against
apoptosis. The GF–pre-treated MSCs showed better survival against hypoxia. The DiI-positive area was greater in
the infarcted myocardium transplanted with GF–pretreated MSCs than with naïve MSCs in the early posttransplantation period. Surviving MSC can contribute to
cardiac repair through direct differentiation and paracrine
and cytoprotective action. Thus the more MSCs survive, the
bigger improvement of cardiac function can be expected
(30). Anti-apoptotic effects of GF treatment on MSCs can
be explained by at least 2 mechanisms: 1) FGF-2 prolonged
telomerase length and life span of MSCs and is useful in
obtaining a large number of cells with preserved differentiation potential (31); and 2) FGF and IGF-1 activate
PI3-kinase/Akt pathway, which is known to mediate antiapoptotic signaling (32,33). However, the survival benefit of
MSCs was attenuated at 8 weeks, and absolute amounts of
surviving MSCs were also markedly decreased. Thus, this
mechanism can not solely explain the beneficial effects of
GF treatment observed at 8 weeks. Second, as discussed
previously, GF treatment enhanced gap junction-mediated
cytoprotective effect of MSCs on CMCs. Finally, the better
LV systolic function in the treated MSC group might be
due to synchronized contraction. Enhanced intercellular
communication through gap junction might lead to electrical synchronization and then synchronous contraction
(34,35). Therefore, better-coordinated LV systolic contraction can be expected when GF–pre-treated MSCs are
transplanted.
Effects of MSC transplantation on arrhythmogenecity of
infarcted myocardium. Since proarrhythmic effects of
skeletal myoblasts have been reported, there have been
concerns about proarrhythmic potential of stem cell transplantation. Mixtures of MSCs and neonatal rat ventricular
myocytes can produce an arrhythmogenic substrate (36).
Whether increased gap junctional communication is favorable remains uncertain. However, increased expression of
connexin-43 has been reported to reduce inducible arrhythmia in infarcted heart (37). Concordant with the previous
study, transplantation of GF–pre-treated MSCs, which can
make higher expression of connexin-43 and functional gap
JACC Vol. 51, No. 9, 2008
March 4, 2008:933–43
junctions, did not aggravate arrhythmic risk after MI in our
study.
Study limitations. There are several limitations to this
study. First, we used human GFs for pre-treatment of
MSCs isolated from rat bone marrow. However, structural
homology of these GFs is very high between humans and
rats (38,39). Second, because we treated MSCs with a
combination of 3 GFs, the effect of individual GF and the
underlying mechanism of individual effect were not understood. Third, DiI is a membrane dye and might have
limitation for detection of transplanted MSCs 8 weeks later.
However, Dai et al. (40) reported that the DiI-positive cells
were observed at 3 and 6 months after transplantation of
DiI-labeled MSCs into the scar of a 1-week-old infarcted
myocardium. And DiI diffuses within the plasma membrane, resulting in staining of the entire cell, which was
observed in our study as well as a previous one (40).
Conclusions
We found that the priming of MSCs with a combination of
FGF-2, IGF-1, and BMP-2 enhanced commitment of
MSCs to CMC lineage in the coculture with CMCs. These
GF-primed MSCs with better expression of connexin-43
showed a greater cytoprotective effect on CMCs, which was
mediated through gap junction. Moreover, coupling
through gap junction might explain the improvement of LV
systolic function after transplantation of primed MSCs to
the infarcted myocardium compared with naïve MSCs.
Treatment of MSCs with GFs might be a physiological and
feasible strategy to increase therapeutic efficacy of MSC
transplantation to the infarcted myocardium, which has a
significant clinical impact.
Reprint requests and correspondence: Dr. Hyo-Soo Kim Dr.
Hyun-Jae Kang, Department of Internal Medicine, Seoul National
University College of Medicine, 28 Yeongun-dong, Chongro-gu,
Seoul, 110-744, Korea. E-mail: [email protected] or nowkang@
snu.ac.kr.
REFERENCES
1. Orlic D, Kajstura J, Chimenti S, et al. Bone marrow cells regenerate
infarcted myocardium. Nature 2001;410:701–5.
2. Shake JG, Gruber PJ, Baumgartner WA, et al. Mesenchymal stem cell
implantation in a swine myocardial infarct model: engraftment and
functional effects. Ann Thorac Surg 2002;73:1919 –25.
3. Janssens S, Dubois C, Bogaert J, et al. Autologous bone marrowderived stem-cell transfer in patients with ST-segment elevation
myocardial infarction: double-blind, randomised controlled trial.
Lancet 2006;367:113–21.
4. Meyer GP, Wollert KC, Lotz J, et al. Intracoronary bone marrow cell
transfer after myocardial infarction: eighteen months’ follow-up data
from the randomized, controlled BOOST (BOne marrOw transfer to
enhance ST-elevation infarct regeneration) trial. Circulation 2006;113:
1287–94.
5. Gnecchi M, He H, Liang OD, et al. Paracrine action accounts for
marked protection of ischemic heart by Akt-modified mesenchymal
stem cells. Nat Med 2005;11:367– 8.
Hahn et al.
Mesenchymal Stem Cells and Growth Factors
JACC Vol. 51, No. 9, 2008
March 4, 2008:933–43
6. Nagaya N, Kangawa K, Itoh T, et al. Transplantation of mesenchymal
stem cells improves cardiac function in a rat model of dilated
cardiomyopathy. Circulation 2005;112:1128 –35.
7. Lawrence TS, Beers WH, Gilula NB. Transmission of hormonal
stimulation by cell-to-cell communication. Nature 1978;272:501– 6.
8. Ghosh-Choudhury N, Abboud SL, Mahimainathan L, Chandrasekar
B, Choudhury GG. Phosphatidylinositol 3-kinase regulates bone
morphogenetic protein-2 (BMP-2)-induced myocyte enhancer factor
2A-dependent transcription of BMP-2 gene in cardiomyocyte precursor cells. J Biol Chem 2003;278:21998 –2005.
9. Kofidis T, de Bruin JL, Yamane T, et al. Insulin-like growth factor
promotes engraftment, differentiation, and functional improvement
after transfer of embryonic stem cells for myocardial restoration. Stem
Cells 2004;22:1239 – 45.
10. Rosenblatt-Velin N, Lepore MG, Cartoni C, Beermann F, Pedrazzini
T. FGF-2 controls the differentiation of resident cardiac precursors
into functional cardiomyocytes. J Clin Invest 2005;115:1724 –33.
11. Schultheiss TM, Burch JB, Lassar AB. A role for bone morphogenetic
proteins in the induction of cardiac myogenesis. Genes Dev 1997;11:
451– 62.
12. Aberg ND, Blomstrand F, Aberg MA, et al. Insulin-like growth
factor-I increases astrocyte intercellular gap junctional communication
and connexin43 expression in vitro. J Neurosci Res 2003;74:12–22.
13. Doble BW, Kardami E. Basic fibroblast growth factor stimulates
connexin-43 expression and intercellular communication of cardiac
fibroblasts. Mol Cell Biochem 1995;143:81–7.
14. Tomita S, Li RK, Weisel RD, et al. Autologous transplantation of
bone marrow cells improves damaged heart function. Circulation
1999;100:II247–56.
15. Thaik CM, Calderone A, Takahashi N, Colucci WS. Interleukin-1
beta modulates the growth and phenotype of neonatal rat cardiac
myocytes. J Clin Invest 1995;96:1093–9.
16. Cho HJ, Youn SW, Cheon SI, et al. Regulation of endothelial cell and
endothelial progenitor cell survival and vasculogenesis by integrinlinked kinase. Arterioscler Thromb Vasc Biol 2005;25:1154 – 60.
17. Lee HY, Chung JW, Youn SW, et al. Forkhead transcription factor
FOXO3a is a negative regulator of angiogenic immediate early gene
CYR61, leading to inhibition of vascular smooth muscle cell proliferation and neointimal hyperplasia. Circ Res 2007;100:372– 80.
18. Yang HM, Kim HS, Park KW, et al. Celecoxib, a cyclooxygenase-2
inhibitor, reduces neointimal hyperplasia through inhibition of Akt
signaling. Circulation 2004;110:301– 8.
19. Hahn JY, Cho HJ, Bae JW, et al. Beta-catenin overexpression reduces
myocardial infarct size through differential effects on cardiomyocytes
and cardiac fibroblasts. J Biol Chem 2006;281:30979 – 89.
20. Walker MJ, Curtis MJ, Hearse DJ, et al. The Lambeth Conventions:
guidelines for the study of arrhythmias in ischaemia infarction, and
reperfusion. Cardiovasc Res 1988;22:447–55.
21. Olson EN, Schneider MD. Sizing up the heart: development redux in
disease. Genes Dev 2003;17:1937–56.
22. Alsan BH, Schultheiss TM. Regulation of avian cardiogenesis by Fgf8
signaling. Development 2002;129:1935– 43.
23. Knochel S, Dillinger K, Koster M, Knochel W. Structure and
expression of Xenopus tropicalis BMP-2 and BMP-4 genes. Mech
Dev 2001;109:79 – 82.
24. Zhang H, Bradley A. Mice deficient for BMP2 are nonviable and have
defects in amnion/chorion and cardiac development. Development
1996;122:2977– 86.
943
25. Urbanek K, Rota M, Cascapera S, et al. Cardiac stem cells possess
growth factor-receptor systems that after activation regenerate the
infarcted myocardium, improving ventricular function and long-term
survival. Circ Res 2005;97:663–73.
26. Xu M, Wani M, Dai YS, et al. Differentiation of bone marrow stromal
cells into the cardiac phenotype requires intercellular communication
with myocytes. Circulation 2004;110:2658 – 65.
27. Li TS, Hayashi M, Ito H, et al. Regeneration of infarcted myocardium
by intramyocardial implantation of ex vivo transforming growth
factor-beta-preprogrammed bone marrow stem cells. Circulation
2005;111:2438 – 45.
28. Krysko DV, Leybaert L, Vandenabeele P, D’Herde K. Gap junctions
and the propagation of cell survival and cell death signals. Apoptosis
2005;10:459 – 69.
29. Alexaki VI, Charalampopoulos I, Kampa M, et al. Activation of
membrane estrogen receptors induce pro-survival kinases. J Steroid
Biochem Mol Biol 2006;98:97–110.
30. Tang YL, Tang Y, Zhang YC, Qian K, Shen L, Phillips MI.
Improved graft mesenchymal stem cell survival in ischemic heart with
a hypoxia-regulated heme oxygenase-1 vector. J Am Coll Cardiol
2005;46:1339 –50.
31. Bianchi G, Banfi A, Mastrogiacomo M, et al. Ex vivo enrichment of
mesenchymal cell progenitors by fibroblast growth factor 2. Exp Cell
Res 2003;287:98 –105.
32. Debiais F, Lefevre G, Lemonnier J, et al. Fibroblast growth factor-2
induces osteoblast survival through a phosphatidylinositol 3-kinasedependent, -beta-catenin-independent signaling pathway. Exp Cell
Res 2004;297:235– 46.
33. Zheng WW, Quirion RR. Insulin-like growth factor-1 (IGF-1)
induces the activation/phosphorylation of Akt kinase and cAMP
response element-binding protein (CREB) by activating different
signaling pathways in PC12 cells. BMC Neurosci 2006;7:51.
34. Reinecke H, MacDonald GH, Hauschka SD, Murry CE. Electromechanical coupling between skeletal and cardiac muscle. Implications for
infarct repair. J Cell Biol 2000;149:731– 40.
35. Beeres SL, Atsma DE, van der Laarse A, et al. Human adult bone
marrow mesenchymal stem cells repair experimental conduction block
in rat cardiomyocyte cultures. J Am Coll Cardiol 2005;46:1943–52.
36. Chang MG, Tung L, Sekar RB, et al. Proarrhythmic potential of
mesenchymal stem cell transplantation revealed in an in vitro coculture
model. Circulation 2006;113:1832– 41.
37. Kuhlmann MT, Kirchhof P, Klocke R, et al. G-CSF/SCF reduces
inducible arrhythmias in the infarcted heart potentially via increased
connexin43 expression and arteriogenesis. J Exp Med 2006;203:87–97.
38. Frenz DA, Liu W, Williams JD, et al. Induction of chondrogenesis:
requirement for synergistic interaction of basic fibroblast growth factor
and transforming growth factor-beta. Development 1994;120:415–24.
39. Rubin JS, Mariz I, Jacobs JW, Daughaday WH, Bradshaw RA.
Isolation and partial sequence analysis of rat basic somatomedin.
Endocrinology 1982;110:734 – 40.
40. Dai W, Hale SL, Martin BJ, et al. Allogeneic mesenchymal stem cell
transplantation in postinfarcted rat myocardium: short- and long-term
effects. Circulation 2005;112:214 –23.
APPENDIX
For supplementary figures, please see the online version of this article.