OUP user menu

Purified cardiomyocytes from bone marrow mesenchymal stem cells produce stable intracardiac grafts in mice

Naoichiro Hattan, Haruko Kawaguchi, Kiyoshi Ando, Eriko Kuwabara, Jun Fujita, Mitsushige Murata, Makoto Suematsu, Hidezo Mori, Keiichi Fukuda
DOI: http://dx.doi.org/10.1016/j.cardiores.2004.10.004 334-344 First published online: 1 February 2005


Objective: We have previously isolated cardiomyogenic cells from murine bone marrow (CMG cells). Regenerated cardiomyocytes are important candidates for cell transplantation, but as they are stem cell derived, they can be contaminated with various cell types, thereby requiring characterization and purification. Our objectives were to increase the efficiency of cell transplantation and to protect the recipients from possible adverse effects using an efficient and effective purification process as well as to characterize regenerated cardiomyocytes.

Methods: Noncardiomyocytes were eliminated from a mixture of stem-cell-derived cells using a fluorescence-activated cell sorter to specifically isolate CMG cells transfected with a recombinant plasmid containing enhanced green fluorescent protein (EGFP) cDNA under the control of the myosin light chain-2v (MLC-2v) promoter. Gene expression and the action potential were investigated, and purified cells were transplanted into the heart of adult mice.

Results: Six percent to 24% of transfected CMG cells expressed EGFP after differentiation was induced, and a strong EGFP-positive fraction was selected. All the sorted cells began spontaneous beating after 3 weeks. These cells expressed cardiomyocyte-specific genes such as α-skeletal actin, β-myosin heavy chain, MLC-2v, and CaV1.2 and incorporated bromodeoxyuridine for 5 days. The isolated EGFP-positive cells were expanded for 5 days and then transplanted into the left ventricle of adult mouse hearts. The transplanted cells survived for at least 3 months and were oriented in parallel to the cardiomyocytes of the recipient heart.

Conclusions: The purification and transplantation of differentiated cardiomyocytes from adult stem cells provides a viable model of tissue engineering for the treatment of heart failure.

  • Cardiomyocytes
  • Heart failure
  • Transplantation
  • Stem cell
  • Bone marrow

1. Introduction

Necrotic cardiomyocytes in infarcted ventricular tissue are progressively replaced by fibroblasts leading to the formation of scar tissue and this loss of cardiomyocytes leads to regional contractile dysfunction. Transplanted fetal cardiomyocytes can survive in heart scar tissue, thereby limiting scar expansion and preventing post-infarction heart failure [1–3]. The transplantation of cultured cardiomyocytes into damaged myocardium has been proposed as a novel method for treating heart failure. While this is a revolutionary idea, it remains clinically unfeasible due to the difficulty in obtaining donor fetal hearts. For this reason, research has focused on the development of a cardiomyogenic cell line to treat heart failure by transplantation therapy.

Advances in regenerative medicine have enabled the generation of various cell types from embryonic stem (ES) cells or adult stem cells [4,5]. We recently reported the generation of cardiomyocytes from marrow mesenchymal stem cells in vitro (CMG cells) and demonstrated that these cells spontaneously beat, express atrial natriuretic factors, and possess a fetal ventricular cardiomyocyte-like phenotype [6]. We also reported that cardiomyocytes regenerated from marrow mesenchymal stem cells express α1A, α1B, α1D, β1, and β2 adrenergic receptors and M1 and M2 muscarinic receptors [7]. Stimulation of the α1 receptors with phenylephrine caused cardiomyocyte hypertrophy, and stimulation of the β receptors with isoproterenol increased the beating rate and contractility of the regenerated cardiomyocytes. These findings demonstrate the suitability of bone-marrow-derived regenerated cardiomyocytes as a candidate for use in cell transplantation therapy.

Purification of regenerated cardiomyocytes is required prior to use for cardiomyocyte transplantation. The population of cardiomyocytes in ES-cell-derived embryoid bodies is less than 10%, and the population of cardiomyocytes in 5-azacytidine-exposed CMG cells is less than 10–30%. To increase the efficiency of transplantation and protect recipients from possible adverse effects, regenerated cardiomyocytes need to be purified from the population of differentiated cell types prior to cell transplantation. Klug [8] and Muller [9] independently reported that embryonic stem-cell-derived cardiomyocytes could be purified using a cardiomyocyte-specific gene promoter–drug-resistant gene expression system. In this study, we purified bone-marrow-derived cardiomyocytes using a recombinant plasmid containing enhanced green fluorescent protein (EGFP) cDNA under the control of the myosin light chain-2v (MLC-2v) promoter. Purified cells were then transplanted into recipient mice hearts and the success of transplantation was analyzed histologically.

2. Methods

All experimental procedures and protocols were approved by the Animal Care and Use Committees of the Keio University, Japan, and the investigation conforms to the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH Publication No. 85–23 revised 1996).

2.1. Preparation of bone marrow-derived regenerated cardiomyocytes

Murine bone-marrow-derived mesenchymal stem cells (CMG cells) were cultured in Iscove's modified Dulbecco's medium (IMDM) supplemented with 20% FBS as previously described [6,7]. The cells were exposed to 3 μmol/l of 5-azacytidine for 24 h to induce cell differentiation [6].

2.2. Construction of myosin light chain 2v-promoted EGFP plasmid

An expression vector, pMLC2v-EGFP, was constructed by cloning a 2.7-kb HindIII–EcoRI fragment of the rat MLC-2v promoter region [10,11] into the HindIII–EcoRI site of pEGFP-1 (Clontech, Palo Alto, CA), so that EGFP would be expressed under the control of MLC-2v promoter (Fig. 1a). This plasmid also contains the neomycin-resistance gene to enable selection of permanently transfected clones. MLC-2v is specifically expressed in ventricular cardiomyocytes.

Fig. 1

Construction of pMLC-2v-EGFP and expression of EGFP in differentiated CMG cells. (a) Restriction map of pMLC-2v-EGFP. (b–g) Microscopy of the pMLC-2v-EGFP stably transfected CMG cells. b, d, and f shows phase contrast microscopy of the CMG cells after differentiation, and c, e, and g represent fluorescent microscopic views of the same field in b, d, and f. (b, c) 3 days, (d, e) 7 days, and (f, g) 4 weeks after the 5-azacytidine exposure. Bars indicate 100 μm.

2.3. Transfection of MLC2v-EGFP expression plasmid and cell selection

The MLC2v-EGFP plasmid was transfected into CMG cells by liposomal transfection. After 24 h when cells are about 20% confluent, a mixture containing 2 μg of plasmid DNA and 4 μl of LT1 TransIT Polyamine Transfection Reagent (Mirus Corporation) in OPTI-MEM (Life Technologies, Gaithersburg, MD) were added to each 35-mm culture dish. After selection with 1000 μg/ml of G418 for 4 weeks, stably transfected colonies derived from single cells were cloned and pooled. EGFP fluorescence was observed under a fluorescence microscope (Olympus TMD300, Tokyo, Japan).

2.4. Flow cytometry and cell sorting

Flow cytometry and sorting of EGFP(+) cells were performed on a FACS Vantage (Becton Dickinson, Cockeysville, MD). Cells were analyzed by light forward and side scatter and for EGFP fluorescence through a 530 nm band pass filter as they traversed the beam of an argon ion laser (488 nm, 100 mW). Nontransfected control cells were used to set the background fluorescence. Cell sorting was performed 3 days after 5-azacytidine exposure at 500 cells/s as EGFP(+) cells displaying fluorescence higher than the background level were observed at this time point.

2.5. Infection of recombinant adenovirus vectors

Replication-deficient recombinant adenovirus vector, pAdex-LacZ, was constructed by cloning LacZ cDNA into the SwaI site of pAdex1CAwt as previously described [12]. In this vector, E. coli β-galactosidase is expressed under the control of a strong, ubiquitously expressed, promoter-derived from the cytomegalovirus enhancer-chicken β-actin hybrid [13]. On day 3 after seeding, EGFP(+) cells isolated by FACS were incubated with PBS containing Adex-LacZ virus at 10 MOI for 100 min. The cells were washed three times to remove virus remaining on the cell surface. Prior to transplantation, the cells were incubated in IMDM with 20% FBS for 2 days.

2.6. Transmission electron microscopy

Cells were washed three times with PBS (pH 7.4) prior to transmission electron microscopy. Cells were initially fixed with PBS containing 2.5% glutaraldehyde for 2 h. The cells were then embedded in epoxy resin. Ultra-thin sections cut horizontally to the growing surface were double stained in uranyl acetate and lead citrate, and viewed under a JEM-1200EX transmission electron microscope.

2.7. Bromodeoxyuridine (BrdU) incorporation

To detect nuclei undergoing DNA synthesis, cells were incubated with BrdU (10 μM) for 5 h, rinsed with PBS, and then fixed in methanol for 20 min at 4 °C. Immunofluorescence microscopy using a monoclonal antibody against BrdU was performed as described previously [14]. The percentage of BrdU-positive cells was estimated by counting cells on photographs of randomly chosen fields.

2.8. Gene expression analysis

Total RNA was extracted from EGFP(+) cells isolated at 7 days following transfection. RT-PCR was performed to detect α-myosin heavy chain (MHC), β-MHC, α-skeletal actin, α-cardiac actin, myosin light chain-2v (MLC-2v), MLC-2a, Cav1.2, myoD, calponin, and α-smooth muscle actin genes. The primers and PCR cycles used were as described previously [6,15,16]. Primers for Cav1.2 were CTGCAGGTGATGATGAGGTC for the forward primer and GCGGTGTTGTTGGCGTTGTT for the reverse primer.

2.9. Immunostaining

Cells were attached to gelatin-coated glass slides, fixed in 4% paraformaldehyde, and then stained with primary antibodies against anti-GATA4, anti-troponin I, and anti-MEF2C antibodies (all from Santa Cruz Biotechnology), or anti-connexin43 antibody (Sigma). Anti-goat-IgG conjugated with Texas red or anti-rabbit IgG conjugated with Rhodamine (1:500, Pharmingen) was use as a secondary antibody.

2.10. Action potential recording

Electrophysiological studies were performed in IMDM containing (mmol/L) CaCl2 1.49, KCl 4.23, and HEPES 25 (pH 7.4). Cultured cells were placed on the stage of an inverted phase contrast optic (Diaphoto-300, Nikon) at 23 °C. Action potentials were recorded using conventional microelectrodes as described previously [8]. Intracellular recordings were taken from MLC2v-EGFP-purified cells 3 weeks following transfection.

2.11. Cell transplantation

Animal Care and Use Committees of Keio University approved all experimental procedures and protocols. Female scid mice (12 weeks) were anesthetized initially with ether and placed on a warm pad maintained at 37 °C. The trachea was cannulated with a polyethylene tube connected to a respirator (Shinano, Tokyo, Japan) with a tidal volume set at 0.6 ml and a rate set at 110/min. Mice were then anesthetized with 0.5–1.5% isoflurene under controlled ventilation with a respirator for the remainder of the surgical procedure. A left thoracotomy was performed between ribs 4 and 5, and the pericardial sac was removed. Isolated EGFP(+) cells that had been expanded for 5 days were resuspended in PBS at a concentration of 5 × 107 cells/ml. A total cell suspension volume of 50 μl was drawn into a 50 μl Hamilton syringe with a 31-gauge needle, and 10 μl was injected into the anterior wall of the left ventricle. Following the transplantation, residual cells in the syringe were collected and stained with trypan blue. The total and living cell numbers were counted. The number of living cells to inject was calculated by the following formula. (The injected living cells)=[(Total injected cells)−(Residual cells in the syringe)](Percent of living cells). Injection of PBS was used as a control.

2.12. Histological studies

The mice were sacrificed, and the hearts were dissected and fixed in 2% formaldehyde and 0.2% glutaraldehyde in PBS at room temperature for 5 min. The hearts were then washed in PBS and then incubated overnight in X-gal solution (1 mg/ml X-gal, 15 mmol/L potassium ferricyanide, 15 mmol/L potassium ferrocyanide, and 2 mmol/L MgCl2 in PBS). The hearts were refixed in the same fix solution, embedded in paraffin, and sectioned into 6-μm-thick slices for hematoxylin–eosin staining. The numbers of X-gal-stained CMG cells were counted using serial sections of the transplanted heart (more than 200 slices/mouse), and an estimate of total transplanted cell survival was obtained using the following formula. (Percent of cells surviving in the recipient heart)=[(Total surviving cells in the recipient heart)/(Injected living cells)]100.

To observe EGFP fluorescence, the hearts were embedded in OCT compound and frozen with liquid nitrogen. A cryostat was used to generate 6-μm-thick sections. The samples were examined with a confocal LASER microscope (LSM510; Carl Zeiss, Jena, Germany). The GFP signal was confirmed by emission finger printing, using the LSM 510 Meta spectrometer (Carl Zeiss).

2.13. Electrocardiography (ECG) recording

ECG recordings were performed 2 and 4 weeks after transplantation. Mice were anesthetized with ether, needle limb leads were fixed, and the ECG was recorded for 1 h.

2.14. Statistics

Values are presented as mean ± SD. The significance of differences among mean values was determined by ANOVA. Statistical comparison of the control and treated groups was carried out using the nonparametric Fisher's multiple comparison tests. The level accepted for significance was p<0.05.

3. Results

3.1. Regenerated cardiomyocytes, but not other cell types, express EGFP

G418-resistant cells were exposed to 5-azacytidine and after 3 days EGFP(+) cells exhibited a fibroblast-like morphology (Fig. 1b,c), and were difficult to distinguish from other cell types. After 7 days, the EGFP(+) cells displayed a spindle-like morphology (Fig. 1d,e), but did not spontaneously beat at this stage. After 3 weeks, the EGFP(+) cells began to appear more rod-like and form inter-cell connections and after 4 weeks spontaneous beating was observed (Fig. 1f,g). Some fractions of the EGFP(−) cells differentiated into adipocytes, but other EGFP(−) cells did not display any specific morphology. These findings indicate that the MLC2v–EGFP system may be a useful method for distinguishing regenerated cardiomyocytes from other cell types at an early stage.

3.2. Fluorescence-activated cell sorting (FACS) analysis

FACS analysis was performed 3 day after 5-azacytidine exposure to isolate regenerated cardiomyocytes. Control cells (before 5-azacytidine exposure) showed no detectable fluorescence (Fig. 2a), whereas 3 days after 5-azacytidine exposure the cells stable transfected with the MLC2v-EGFP expression plasmid generated sufficient EGFP signal for cell sorting (Fig. 2b). The EGFP(+) fraction ranged from 6–24%. Fig. 2c,d shows the cells 4 days after cell sorting (7 days after 5-azacytidine exposure) displaying a fibroblast-like morphology. The percentage of EGFP-positive cells was calculated by comparing cell counts from phase contrast microscopy with EGFP(+) cell counts using fluorescence microscopy, 3 days after cell sorting. More than 99% of the sorted cells expressed EGFP fluorescence. After 3 weeks, these cells had a spindle-like appearance and began spontaneous beating (Fig. 2e,f).

Fig. 2

FACS analysis of the pMLC-2v-EGFP-transfected cells and microscopy of the sorted cells. (a, b) FACS analysis of the pMLC-2v-EGFP-transfected cells. The horizontal axis indicates the intensity of EGFP fluorescence. (a) Control cells, (b) cells 3 days after exposure to 5-azacytidine exposure. d and f are fluorescence microscopy images of the EGFP signal. c and e are phase contrast microscopy views of the same field. (c, d) 4 days, and (e, f) 3 weeks after cell sorting. Note that all the cells display EGFP fluorescence, and that the EGFP(+) CMG cells exhibit a cardiomyocyte-like appearance and spontaneously beat after 3 weeks. Bars in c,d and e,f indicate 100 and 500 μm, respectively.

3.3. Character of the sorted EGFP(+) regenerated cardiomyocytes

A total of 768 single EGFP(+) cell clones were isolated using FACS analysis. Although EGFP(+) cells undergo cell division after 5-azacytidine exposure, a cardiomyocyte cell line could not be generated as cells stop proliferating after several cell divisions. The cells were exposed to BrdU to confirm their mitogenicity, and double immunostaining was performed with antisarcomeric myosin and anti-BrdU antibodies. Myosin-positive cells incorporated BrdU until day 5, but stopped incorporating it after day 7 (Fig. 3a). This finding shows that the mitogenicity of the isolated EGFP(+) CMG cells is limited, so it can be assumed that the risk of cardiomyosarcoma formation is negligible.

Fig. 3

Characteristics of the sorted CMG cardiomyocytes. (a) BrdU incorporation of EGFP(+) CMG cells after cell sorting. BrdU was loaded for 5 h, and its incorporation was detected. BrdU incorporation was observed until 5 days after cell sorting (8 days after 5-azacytidine exposure). (b) Phenotype of the EGFP(+) CMG cells. RT-PCR was performed for α-MHC, β-MHC, MLC-2v, MLC-2a, α-skeletal actin, α-cardiac actin, and cardiac α1c Ca2+ channel. The expression pattern of the cardiac contractile protein indicated that these cells had the fetal ventricular phenotype. MLC-2v-EGFP selected cells did not express myoD, calponin, and α-smooth muscle actin genes. Femoral muscle, which includes vascular smooth muscle cells, were used as a positive control. M: 1-kb DNA ladder. RT: reverse transcription. (c) The representative tracing of the action potentials at 3 weeks after cell sorting was shown. These action potentials show ventricular cardiomyocyte-like action potentials.

RT-PCR analysis of cardiac contractile proteins revealed that the isolated EGFP(+) CMG predominantly express the β-myosin heavy chain, α-skeletal-actin, and MLC-2v, indicating that the phenotype of these cells represents fetal ventricular cardiomyocytes. These cells also express cardiac L-type Ca2+ channels but did not express myogenic genes such as myoD, or smooth-muscle-specific genes, such as calponin or α-smooth muscle actin genes (Fig. 3b).

3.4. Action potential recording

MLC2v-EGFP-selected cells showed regular spontaneous beating 3 weeks following selection. The action potentials of these cells had a relatively shallow resting membrane potential with a late diastolic slow depolarization, like a pacemaker potential. They also displayed peak-notch-plateau characteristics representative of ventricular cardiomyocyte-like action potentials (Fig. 3c).

3.5. Immunostaining and transmission electron microscopy

Immunostaining revealed that EGFP(+) but not EGFP(−) CMG cells express cardiac troponin I (Fig. 4a–d). EGFP(+) CMG cells express both GATA4 and MEF2C, respectively (Fig. 4e,f). Interestingly, EGFP(−) CMG cells express GATA4 and Nkx2.5. These findings are consistent with the previous report that these cardiac transcription factors are expressed before final 5-azacytidine exposure [6]. EGFP(+) CMG cells also express connexin43 (Fig. 4g).

Fig. 4

Photograph of immunofluorescence and transmission electron micrograph of CMG cells. (a–d) EGFP(+) and EGFP(−) CMG cells were stained with anti-troponin I antibodies (a) and DAPI (c). EGFP(+) CMG cells expressed troponin I, but EGFP(−) CMG cell did not express troponin I. (e) Immunofluorescent staining with GATA4. Both EGFP(+) and EGFP(−) CMG cells expressed GATA4. (f) Immunofluorescent staining with MEF2C. Both EGFP(+) and EGFP(−) CMG cells expressed MEF2C. (g) Immunofluorescent staining with connexin43. EGFP(+) CMG cells expressed connexin43. (h) Transmission electron microscopy of the CMG cells showed typical contractile apparatus.

The sorted GFP(+) cells were cultured for 2 weeks, fixed, and processed for transmission electron microscopy. The typical contractile apparatus of the sarcomeres, including striation pattern, was observed (Fig. 4h).

3.6. Cell transplantation study

Animals with transplanted EGFP(+) cells were sacrificed at 2, 4, 8, and 12 weeks. Confocal LASER microscopy revealed that the EGFP(+) transplanted cardiomyocytes survived in the recipient heart (Fig. 5a–c). The control experiment revealed no EGFP(+) transplanted cardiomyocytes (data not shown) [17]. The orientation of the transplanted cells was consistent with the cardiomyocytes of the recipient heart. The EGFP(+) cells were observed only at the site of injection in the left ventricle and in no other parts of the heart. We also confirmed that these green signals were not due to nonspecific background fluorescence, but due to the EGFP itself, using absorbance frequency analysis on a LSM510 Meta spectrometer (Fig. 5d).

Fig. 5

Histological analysis of the transplanted CMG cells. (a–c) Confocal microscopy of the recipient heart transplanted with the sorted cardiomyocytes at 4 weeks. The transplanted cells could be clearly identified by EGFP signals. a and b show the transverse section of the transplanted cardiomyocytes, and c shows the longitudinal section. Bars indicate 50 μm. (d) The emission profile of the green signal in GFP+ cells was investigated by CLSM. The emission peak existed at 510–530 nm, and the profile was ascertained to be that of GFP, and not arising from nonspecific background. The inset shows the GFP+ transplanted cells, and the cross indicates the site of the emission profile. (e–m) The bone-marrow-derived cardiomyocytes were sorted and marked with adenovirus-mediated LacZ gene. e is a photograph of the whole heart, and f shows an enlarged photograph of the injected site. The transplanted LacZ-positive cells were identified from the surface. g and h show the microscopy of the injected site of the left ventricle. The samples were stained with LacZ and hematoxylin–eosin. i–k show higher magnification of the same fields: i shows the transverse section, and j,k show the longitudinal section. (l, m) Transplanted CMG cells were stained with anticonnexin43 antibody and DAPI. Connexin43 was expressed at the both ends of the transplanted EGFP(+) CMG cells. Green: EGFP, Blue: DAPI, Red: connexin43.

Fig. 5e shows the entire murine heart stained with LacZ at 4 weeks after transplantation, and Fig. 5f showed an enlarged photograph of the site of injection. Cells were transplanted into the anterior free wall of the left ventricle and were observed to be rectangular in shape and located at the surface of the heart. Fig. 5g,h shows the site of injection in a transverse section of the left ventricle stained with LacZ and hematoxylin–eosin 4 weeks after transplantation. The scar of the injection needle is shown in Fig. 5h. Granulomatous tissue was also observed around the site of injection. The LacZ-stained transplanted cells were clearly visible, and were located throughout the site of injection. Fig. 5i–k shows transverse and longitudinal sections of the transplanted cardiomyocytes at higher magnifications. This figure clearly shows the arrangement of the transplanted cells parallel to the cardiomyocytes of the recipient heart. Fig. 5l,m shows expression of connexin43 at the longitudinal border between transplanted EGFP(+) CMG cells and adjacent cardiomyocytes of the recipient heart.

Transplanted cardiomyocytes survived in the recipient heart for more than 3 months and the estimated percentage of cells surviving transplantation was 6.5 ± 3.2%. Table 1 shows the diameter of transplanted cardiomyocytes in transverse section. The diameter increased to almost the same size as the cardiomyocytes in the recipient heart over 4 weeks after which time no further increase was observed.

View this table:
Table 1

Diameter of the transplanted bone-marrow-derived cardiomyocyte

Time after transplantation (weeks)24812Recipient cardiomyocytes
Diameter (μm)10.5 ± 3.6*19.0 ± 4.8$19.1 ± 5.0$19.1 ± 4.9$19.5 ± 5.1
  • The diameter of the transplanted cardiomyocytes was measured by the transverse section of the recipient hearts. Each data was obtained by measuring 200 cells. Mean ± SD. $: not significant vs. recipient cardiomyocytes.

  • * p<0.01 vs. 4 weeks and recipient cardiomyocytes.

3.7. ECG recording and survival curve

Of the 35 mice that had undergone cell transplantation, 5 died within 24 h. This is most likely a result of the surgical procedure. The remaining 30 mice survived the duration of the observation period. ECGs recordings in 5 mice at 2 and 4 weeks, respectively, showed no evidence of arrhythmia (ventricular premature beats, ventricular tachycardia) during the recording period (data not shown). This finding suggests that survival of recipients in this model is not affected by arrhythmia.

4. Discussion

Since our report that cardiomyocytes can be regenerated from bone marrow stem cells [6,7], several studies have shown that transplantation of bone-marrow mononuclear cells or bone marrow stem cell fractions into the heart can improve cardiac function. Although the direct transplantation of these cells omits prior differentiation or purification, and thereby shortens the therapeutic period, it remains undetermined whether transplanted cells differentiate into the desired cardiomyocytes or endothelial cells, and not into other cell types including osteoblasts, chondroblasts, or adipocytes. The establishment of a reliable method to repair injured myocardium using cardiomyocyte transplantation requires the preparation of a sufficient number of well-characterized, purified regenerated cardiomyocytes, and an estimation of the survival rate of the transplanted cells.

Tomita [18] reported that the transplantation of 5-azacytidine-treated primary cultured marrow-stromal cells improved the function of the infarcted myocardium. Since the population of mesenchymal stem cells in primary cultured mice marrow stromal cells is less than 0.01%, it is likely that most of the cells transplanted do not differentiate into cardiomyocytes, and that the observed improvement in cardiac function is caused by an improvement in ventricular remodeling or stimulation of angiogenesis.

Jackson transplanted adult stem cells [CD34(−)/low, c-Kit(+), Sca-1(+)] into lethally irradiated mice subsequently rendered ischemic by coronary artery occlusion followed by reperfusion, and reported that the engrafted cells migrated into ischemic cardiac muscle and blood vessels, differentiated to cardiomyocytes and endothelial cells, and contributed to the formation of functional tissue [19]. They found that the donor-derived endothelial cells were present at around 3.3%, primarily in small vessels adjacent to the infarct, and that donor-derived cardiomyocytes were present at around 0.02% and were found primarily in the peri-infarct region. Taken together, these findings show that differentiation from marrow stromal cells to cardiomyocytes in vivo is possible, but that their prevalence is less than other cell types. Condrelli [20] reported that neural stem cells differentiated into heart muscle cells when mixed with heart muscle cells from newborn rats in a process known as transdifferentiation. The mechanism of in vitro transdifferentiation is based on the idea that the developmental limitations of tissue specific stem cells are dictated by the environment, and that new signals that relax these limitations may be provided by cells from a different tissue [20].

It is most likely that the direct transplantation of stem cells into the heart does not facilitate their differentiation into cardiomyocytes, but merely results in their fusion with residual cardiomyocytes. We propose a more rigorous method to achieve repair of damaged tissue by first differentiation adult stem cells cardiomyocytes in vitro, and then transplanting a sufficient number of differentiated cardiomyocytes into the damaged heart tissue. To avoid possible adverse effects, we emphasize the importance of thoroughly investigating the molecular and electrophysiological characteristics of the stem cell-derived regenerated cardiomyocytes prior to transplantation.

In the present study, we used an EGFP reporter gene under the control of the MLC-2v promoter to tag isolated cardiomyocytes. Following FACS analysis, 99% of the isolated cardiomyocytes expressed EGFP, and when transplanted into the recipient heart they survived for at least 4 weeks. We observed no other cell types in the transplanted area, but this may have been because we only used a strongly expressing EGFP(+) fraction.

A plasmid encoding reporter genes and cardiac specific gene promoters was used in a previous study to isolate cardiomyocytes from ES cells or embryonic carcinoma cells (EC cell) [21]. Klug et al. [8] transfected a fusion gene containing the α-myosin heavy chain (α-MHC) promoter and aminoglycoside phosphotransferase (NeoR) into pluripotent ES cells, then differentiated these cells in vitro prior to G418 selection. They reported high purification (>99%) and a survival period in the recipient heart of at least 7 weeks following transplantation. Zweigerdt et al. [22] and Zandstra et al. [23] reported a lab-scale protocol to generate cultures of highly enrich cardiomyocyte from ES cells transfected with a α-MHC-NeoR containing plasmid, and suggest its application to a larger-scale process for the supply of stem cell based cardiomyocytes. Muller et al. [9] isolated a subpopulation of ventricular-like cardiomyocytes from ES cells by transfecting the EGFP gene under the control of the MLC-2v promoter and cytomegalovirus enhancer. Moore et al. [24] reported that EC cell (P19Cl6)-derived cardiomyocytes could be isolated using an EGFP reporter under the control of 250 bp of the MLC-2v promoter. They enzymatically digested embryoid bodies, then isolated a population of cardiomyocytes (97% pure) using Percoll gradient centrifugation and FACS analysis. Kolossov et al. [25] reported the use of EGFP under the control of the cardiac α-actin promoter to isolate ES cell-derived cardiomyocytes. The present study confirmed the efficiency of this strategy for the isolation and purification of cardiomyocytes from bone-marrow-derived stem cells.

Reinecke and Murry [26] and Zhang et al. [27] highlighted the importance of a quantitative analysis of grafted cardiomyocytes, since a large number of fetal or neonatal cardiomyocytes often display apoptosis within several days of transplantation. They reported that only a small percentage of cardiomyocytes survive in the cryoinjured recipient heart, and that heat shock or adenoviral transfer of constitutive active Akt genes could increase their survival. In comparison, the present study reports a slightly higher survival rate for bone marrow-derived cardiomyocytes. One possible reason is the difference in the experimental models as the present study used a mouse uninjured model and not a rat cryoinjured heart model. Another reason is the small size of our not fully differentiated transplanted cells compared with fetal or neonatal cardiomyocytes. A small size may allow transplanted cells to go deep into the recipient heart without mechanical injury.

Recently, Takeda et al. [28] reported that the life span of human bone marrow mesenchymal stem cells could be prolonged by infecting the cells with the retrovirus encoding oncogene bmi-1, human papilloma virus E6 and E7, and human telomerase reverse transcriptase over 150 population doublings, and that these cells could be induced to differentiate into cardiomyocyte using 5-azacytidine and co-culture with the rat cardiomyocytes. Although this procedure is not suitable for clinical application at the present stage, the findings provide valuable information on the use of human bone marrow stem cells for the regeneration of cardiomyocytes.

In summary, the present study provides a new model for tissue engineering. Further studies are required to improve cardiomyocyte differentiation and to increase the efficiency of the transplantation procedure.


This study was supported in part by the research grants (10B-1) of “Nervous and Mental Disorders from the Ministry of Health and Welfare”, Japan, the research grants from the Ministry of Education, Science and Culture, Japan, and the research grants from Health Science Research Grants for Advanced Medical Technology from the Ministry of Welfare, Japan.


  • 1 Naoichiro Hattan and Haruko Kawaguchi contributed equally to this paper.

  • Time for primary review 21 days


  1. [1]
  2. [2]
  3. [3]
  4. [4]
  5. [5]
  6. [6]
  7. [7]
  8. [8]
  9. [9]
  10. [10]
  11. [11]
  12. [12]
  13. [13]
  14. [14]
  15. [15]
  16. [16]
  17. [17]
  18. [18]
  19. [19]
  20. [20]
  21. [21]
  22. [22]
  23. [23]
  24. [24]
  25. [25]
  26. [26]
  27. [27]
  28. [28]
View Abstract