Adult Cardiac Sca-1-positive Cells Differentiate into Beating Cardiomyocytes

Although somatic stem cells have been reported to exist in various adult organs, there have been few reports concerning stem cells in the heart. We here demonstrate that Sca-1-positive (Sca-1+) cells in adult hearts have some of the features of stem cells. Sca-1+ cells were isolated from adult murine hearts by a magnetic cell sorting system and cultured on gelatin-coated dishes. A fraction of Sca-1+ cells stuck to the culture dish and proliferated slowly. When treated with oxytocin, Sca-1+ cells expressed genes of cardiac transcription factors and contractile proteins and showed sarcomeric structure and spontaneous beating. Isoproterenol treatment increased the beating rate, which was accompanied by the intracellular Ca2+ transients. The cardiac Sca-1+ cells expressed oxytocin receptor mRNA, and the expression was up-regulated after oxytocin treatment. Some of the Sca-1+ cells expressed alkaline phosphatase after osteogenic induction and were stained with Oil-Red O after adipogenic induction. These results suggest that Sca-1+ cells in the adult murine heart have potential as stem cells and may contribute to the regeneration of injured hearts.

The heart has long been thought to adapt to increased work and loss of cardiomyocytes by the cellular hypertrophy of residual cardiomyocytes, but not by the proliferation of mature cardiomyocytes or the differentiation of undifferentiated cells. However, recent reports have suggested that adult cardiomyocytes can proliferate under certain pathologic conditions and that there are cells expressing stem cell markers in the adult heart (1)(2)(3)(4). It has been reported that Sca-1-and c-kit-positive (ϩ) cells exist in the adult heart (5) and that adult murine hearts contain potential stem cells; side population (SP) 1 cells (6,7). However, it remains to be clarified whether these cells have the characteristics of stem cells such as abilities of selfrenewal and differentiation into various types of cells including mature cardiomyocytes.
Sca-1 is a member of the Ly-6 family and has first reported as one of the cell surface markers of hematopoietic stem cells (8). Recently many reports have demonstrated that multipotential stem cells derived from bone marrow and skeletal muscle express Sca-1. Okumoto et al. (9) have reported that Sca-1ϩ cells from bone marrow differentiate into hepatocyte when treated with hepatic growth factor. Gojo et al. (10) have reported that adult mesenchymal stem cells from bone marrow abundantly express Sca-1 and differentiate into cardiomyocyte in vivo. Qu-Petersen et al. (11) have shown that skeletal muscle-derived stem cells, which highly express Sca-1, contribute to the regeneration of the skeletal muscle in a mouse model of Duchenne muscle dystrophy. They also demonstrated that the skeletal muscle-derived stem cells were able to differentiate into neural cells and endothelial cells. Asakura et al. (12) have reported that ϳ90% of SP cells in skeletal muscle express Sca-1. It has been reported that skeletal muscle-derived Sca-1ϩ and CD34ϩ cells restore dystrophin in mdx mice (13) and that CD34ϩ and CD45Ϫ cells in the interstitial spaces of skeletal muscle, which highly express Sca-1, differentiate into adipocytes, endothelial, and myogenic cells (14). These findings suggest that Sca-1 might be important evidence for somatic stem cells.
Currently little is known about the humoral or growth factors that induce cardiomyogenic differentiation. It has been shown that ectopic application of bone morphological protein (BMP) 2 and 4 elicits cardiogenic responses in the chick in vivo system (15), and fibroblast growth factor (FGF) 2 and 4, combined with BMP-2 or BMP-4 can induce cardiogenesis in chick non-precardiac mesoderm (16). The non-canonical Wnt/c-Jun-N-terminal kinase pathways have been reported to be essential for cardiac induction in frog and chick embryo systems (17,18). These factors are prerequisites for early cardiac differentiation but are not sufficient for accomplishing differentiation into mature beating cardiomyocytes. Recently Paquin et al. (19) have reported that oxytocin induces differentiation of P19 embryonic carcinoma cells to beating cardiomyocytes. In support of the role of oxytocin in cardiac development, the oxytocin receptor is increased at the protein level in the murine heart from day 7 of gestation, when cardiac differentiation starts (20). Although the precise mechanism of the effect of oxytocin is not clear, oxytocin may play an important role in the differentiation into cardiomyocytes from primitive cells including adult somatic stem cells. Here, we first report that a novel population from Sca-1ϩ cells derived from the adult murine heart proliferates and differentiates into beating cardiomyocytes with oxytocin treatment.
Isolation and Culture of Sca-1ϩ Cells from the Adult Murine Heart-A heart of adult C57Bl/6 mouse (10 -12 weeks old) was enzymatically dissociated into a single cell suspension as described previously (21). Enrichment of Sca-1ϩ cells was achieved by sorting using the Magnetic Cell Sorting (MACS) system (Miltenyl Biotec, Sunnyvale, CA). Whole primary cell suspension was incubated with PE-conjugated anti-Sca-1 antibody for 10 min on ice, washed in PBS supplemented with 3% FBS, incubated with anti-PE micro beads for 15 min at 4°C, and washed with PBS supplemented with 3% FBS. The samples were passed through a MACS column set up in a Miltenyl magnet and the Sca-1ϩ cells were eluted from the column by washing with PBS supplemented with 3% FBS. To increase the purity of the Sca-1ϩ cells, magnetic sorting was performed one more time. The Sca-1ϩ cells were cultured on 1% gelatin-coated dishes with Iscove's Modified Dulbecco's Medium (IMDM) supplemented with 10% FBS, 100 g/ml of penicillin, and 250 g/ml of streptomycin at 37°C in humid air with 5% CO 2 . Twenty-four hours after seeding, the cells were treated with 10 M 5Ј-azacytizine for the initial 72 h or 100 nM oxytocin (WAKO, Japan). After treatment, the medium was changed every 3 days.
Characterization of Cardiac Muscle-derived Stem Cells for Flow Cytometric Analysis-Sca-1ϩ cells were isolated by the MACS system with biotin-conjugated anti-Sca-1 antibody and anti-biotin micro beads. Magnetic sorting was repeated twice, and the cells were incubated with PE-conjugated anti-CD45 antibody, PE-conjugated anti-CD34 antibody, and PE-conjugated anti-c-kit antibody, respectively for 10 min on ice and washed with PBS supplemented with 3% FBS. The percentages of CD45ϩ, CD34ϩ, and c-kitϩ cells were analyzed by the EPICS ALTRA flow cytometer using EXPO32 software (Beckman Coulter, Miami, FL).
Immunocytochemistry-Cells were fixed with 4% paraformaldehyde for 15 min at room temperature. After preblocking with PBS containing 2% donkey serum, 2% bovine serum albumin and 0.2% Nonidet P-40 for 30 min, primary antibodies in PBS containing 2% donkey serum, 2% bovine serum albumin and 0.1% Nonidet P-40 were applied overnight in 4°C. Subsequently cells were washed three times in PBS, and then fluorescein isothiocyanate-or Cy5-conjugated secondary antibodies were applied to visualize expression of specific proteins. Nuclear staining was performed with TO-PRO-3 (Molecular Probes, Eugene, OR). Images of cells were taken by laser confocal microscopy (Radiance2000, Bio-Rad, Hercules, CA).
Phase Contrast Live Imaging-Live images were taken by a Zeiss inverted microscope (Carl Zeiss, Jena, Germany) equipped with phasecontrast objectives and an AxioCam camera. Live image of beating cells were obtained by a chilled CCD camera (Hamamatsu) using I-O DATA Videorecorder software.
Measurement of Intracellular Ca 2ϩ Concentration-Intracellular Ca 2ϩ concentration ([Ca 2ϩ ] i ) in beating cells derived from cardiac Sca-1ϩ cells was measured as previously described (31). The beating cells on gelatin-coated glass coverslips were incubated in HEPES (load- mM ␤-glycerophosphate as described previously (32). For detection of osteocytes, alkaline phosphatase staining (leukocyte alkaline phosphatase assay kit, Sigma-Aldrich) was used. Adipogenic differentiation was induced as described previously (28). Briefly the cells were cultured with Dulbecco's Modified Eagle's Medium (DMEM) supplemented with 5% horse serum with MDI-I mixture; 0.5 mM methyl-isobutylxanthine, 1 M dexamethasone, 100 mM indomethacin, and 10 g/ml insulin for 2 days and then cultured with Dulbecco's modified Eagle's medium supplemented with 5% horse serum and 10 g/ml of insulin for 1 day. Treatment with MDI-I followed by insulin was repeated four times. For detection of accumulated oil droplets, Oil-Red O staining was performed followed by nuclear hematoxylin counterstaining.
Statistical Analysis-Values are presented as mean Ϯ S.E. The significance of differences among mean values was determined by analysis of variance. The accepted level of significance was p Ͻ 0.05.

RESULTS
Cell Surface Antigens of Sca-1ϩ Cells Derived from the Adult Murine Heart-Flow cytometric analysis revealed that Sca-1ϩ cells were enriched to over 90% when adult murine cardiac cells were sorted twice with the MACS system using PE-conjugated anti-Sca-1 antibody and anti-PE micro beads (Fig. 1A). The number of purified Sca-1ϩ cells was ϳ1 ϫ 10 4 cells. Limana et al. (33) have estimated the number of cardiomyocytes in an adult murine heart as ϳ3 ϫ 10 6 . Therefore the percentage of cardiac Sca-1ϩ cells was ϳ0.3% of the total number of cardiomyocytes. Next we examined other cell surface antigens such as CD45, CD34, and c-kit in Sca-1ϩ cells. After repeated magnetic sorting with biotin-conjugated anti-Sca-1 antibody and anti-biotin micro beads, enriched Sca-1ϩ cells were incubated with PE-conjugated anti-CD45, CD34 and c-kit antibodies and analyzed by flow cytometry. In enriched Sca-1ϩ cells, ϳ40% of the cells expressed CD45 (Fig. 1B), and ϳ10% of the cells expressed CD34 (Fig. 1C) and c-kit (Fig. 1D).
Sca-1ϩ Cells from the Adult Murine Heart Differentiate into Beating Cardiomyocytes-In order to induce differentiation into cardiomyocytes, Sca-1ϩ cells were treated with either 5Ј-azacytizine or oxytocin. When Sca-1ϩ cells were cultured with medium containing FBS, they showed various cell shapes, and spindle-like and elongated shapes were predominant ( Fig.  2A). Two weeks after treatment with oxytocin, small round cells with prominent nucleus and little cytoplasm appeared (arrowheads in Fig. 2B). These round cells rapidly proliferated, formed clusters, and detached from the culture dish so that spindle-shaped cells were left. The cells were re-plated when reached to confluence. Four weeks after starting treatment with oxytocin, some spontaneously beating cells were recognized among spindle-shaped cells (arrow in Fig. 2, C and D and Supplemental Data). Spontaneous beating was observed at ϳ1% cells. On the other hand, cells after treatment with 5Јazacytizine or vehicles showed fibroblast-like morphology, and never exhibited round or spindle-shaped morphology, or spontaneous beating.
Next we examined the gene expression of cardiac transcription factors and cardiac structural proteins in Sca-1ϩ cells by RT-PCR. Before treatment with 5Ј-azacytizine or oxytocin, only Csx/Nkx-2.5 and GATA4 were slightly expressed (Fig. 3, lane  P). Four weeks after treatment with 5Ј-azacytizine or oxytocin, all genes of cardiac transcription factors including Csx/Nkx-2.5, GATA4, and MEF-2C and structural proteins such as ␣and ␤-MHC, MLC-2a, MLC-2v, and cardiac ␣-actin were expressed (Fig. 3, lane A for 5Ј-azacytizine and lane OT for oxytocin). Treatment with 100 nM oxytocin antagonist (OTA, [d(CH 2 ) 5 -1,Tyr(Me)-2,Thr-4,Orn-8,Tyr-NH 2 -9] vasotocin, Wako, Japan) completely inhibited oxytocin-induced expression of cardiac genes (Fig. 3, lane OTϩOTA). Total RNA obtained from the adult murine heart and liver were used as positive and negative controls (Fig. 3, lane H for heart and lane L for liver). Loading of equal amounts of RNA was confirmed by expression of the ␤-actin gene. Cardiac gene expression was not observed in cells cultured with vehicle (Fig. 3, lane V).
To examine the expression and localization of cardiac proteins, the Sca-1ϩ cells treated with oxytocin and 5Ј-azacytizine were stained with specific antibodies against cardiac proteins. The cells treated with oxytocin expressed GATA4 (Fig. 4A), ANF (Fig. 4B), and cardiac troponin T (Fig. 4, A and B). MLC-2v (Fig. 4C), sarcomeric myosin heavy chain (Fig. 4D), and tropomyosin (data not shown) were also expressed. Notably, staining of each contractile protein showed a fine striated pattern. Connexin 43 was expressed at the junction between two cardiac troponin T-expressing cells (Fig. 4E). These findings indicate that treatment with oxytocin induced differentiation of Sca-1ϩ cells, derived from the adult murine heart, into mature cardiomyocytes, which had well-organized structures and electrical junctions. After treatment with 5Ј-azacytizine, a fraction of cells expressed sarcomeric myosin heavy chain in fibrillar pattern (Fig. 4F), but not cardiac troponin T (data not shown). Next we sorted cardiac Sca-1ϩ cells on the basis of CD45 expression and cultured with oxytocin. Some of the Sca-1ϩ/CD45Ϫ cells expressed sarcomeric myosin after oxytocin treatment, but none of the Sca-1ϩ/CD45ϩ cells expressed myosin (data not shown), suggesting that Sca-1ϩ cells that can differentiate into cardiomyocytes are in CD45Ϫ population.
Cardiac contraction is regulated by beat to beat change in [Ca 2ϩ ] i . To ascertain that the spontaneous beating of differentiated cardiac Sca-1ϩ cells depends on intracellular level of Ca 2ϩ , we analyzed [Ca 2ϩ ] i transients of the beating cells. As shown in the upper panel of Fig. 5A, the spontaneous beating of differentiated Sca-1ϩ cells was accompanied with [Ca 2ϩ ] i transients. After treatment with 10 Ϫ7 M isoproterenol for 5 min, the frequency of [Ca 2ϩ ] i transients was increased in comparison with control (Fig. 5A, upper panel versus lower panel). Next we examined the predominant subtype of ␤ receptors, which mediates changes in beating rate. Differentiated cardiac Sca-1ϩ cells were treated with vehicle (PBS), propranolol, CGP20712A (␤ 1 -selective blocker), or ICI118551 (␤ 2 -selective blocker) for 30 min and then stimulated with isoproterenol alone for 5 min. Isoproterenol significantly increased the beating rate of the control cells (control: 131.9 Ϯ 5.6, n ϭ 10 versus isoproterenol: 228.9 Ϯ 7.3, n ϭ 10, p Ͻ 0.01, Fig. 5B). The pretreatment with propranolol (average 196.4 Ϯ 5.6, n ϭ 10, p Ͻ 0.05 versus isoproterenol) and CGP20712A (average 188.9 Ϯ 7.5, n ϭ 10, p Ͻ 0.05 versus isoproterenol) reduced the increase in beating rate in response to isoproterenol significantly (Fig. 5B). ICI118551 had no effect on the isoproterenol-induced increase in beating rate.
Sca-1ϩ Cells from the Adult Murine Heart Express Oxytocin Receptor mRNA-To elucidate the role of oxytocin receptor in cardiomyogenesis of Sca-1ϩ cells, we examined the expression of oxytocin receptor in Sca-1ϩ cells. Oxytocin receptor mRNA was present at low levels in Sca-1ϩ cells before oxytocin treatment (Fig. 6, lane P). Expression levels of oxytocin receptor remained low in cells cultured with vehicle (Fig. 6, lane V). After treatment with oxytocin, expression levels of oxytocin receptor were up-regulated (Fig. 6, lane OT). In accordance with the inhibitory effect of oxytocin antagonist on oxytocin-induced cardiac differentiation, oxytocin antagonist inhibited oxytocin-induced up-regulation of the oxytocin receptor (Fig. 6, lane OTϩOTA). These results suggest that the positive feedback mechanism, namely oxytocin-induced up-regulation of oxytocin receptor, plays an important role in oxytocin-induced cardiomyocyte differentiation of cardiac Sca-1ϩ cells.
Sca-1ϩ Cells Can Differentiate into Osteocytes and Adipocytes-It has been reported that Sca-1ϩ cells from skeletal muscle and bone marrow differentiate into various types of cells such as adipocytes, endothelial cells, muscle, neural, and hepatic cells (9,11,14). To determine whether the Sca-1ϩ cells from the adult murine heart have pluripotency, we examined whether these cells could differentiate into cells other than cardiomyocytes. When treated with osteogenic inducers, some of Sca-1ϩ cells were stained with alkaline phosphatase, one of the early markers of osteogenesis (Fig. 7A). RT-PCR clearly revealed that osteogenic marker mRNAs such as alkaline phosphatase and osteocalcin were induced in Sca-1ϩ cells after treatment with osteogenic inducers (Fig. 7B). On the other hand, Sca-1ϩ cells treated with oxytocin never expressed alkaline phosphatase and osteocalcin. When Sca-1ϩ cells were cultured with MDI-I mixture for twelve days, some of Sca-1ϩ cells showed cytoplasmic accumulation of oil droplets stained with Oil-Red O, indicating that Sca-1ϩ cells differentiated into adipocytes (Fig. 7C). DISCUSSION In this report, we have first demonstrated that adult cardiac Sca-1ϩ cells can differentiate into beating cardiomyocytes in vitro by treatment with oxytocin. When treated with oxytocin, the Sca-1ϩ cells expressed cardiac genes including Csx/Nkx-2.5, GATA4, MEF-2C, ␣-MHC, ␤-MHC, MLC-2a, MLC-2v, and cardiac ␣-actin, and cardiac proteins including GATA4, cardiac troponin T, tropomyosin, MLC-2v, sarcomeric myosin heavy chain, ANF, and connexin 43. Furthermore, some of Sca-1ϩ cells showed well organized sarcomere and spontaneous beating. Although transient treatment with 5Ј-azacytizine also induced expression of cardiac genes in Sca-1ϩ cells, it did not induce expression of cardiac troponin T, assembly of sarcomere or spontaneous beating. These results suggest that treatment with 5Ј-azacytizine induces differentiation of Sca-1ϩ cells into cardiomyocytes incompletely and that oxytocin is a more potent inducer of cardiac differentiation than 5Ј-azacytizine.
P19 teratocarcinoma cells differentiate into beating cardiomyocytes after treatment with Me 2 SO and have been considered as a good model of in vitro cardiogenesis (34,35). Several  (19) have reported that oxytocin induces P19 embryonic carcinoma cells to differentiate into cardiomyocytes. Treatment with oxytocin as well as with Me 2 SO induced colony formation of beating cardiomyocytes, expression of cardiac proteins, and oxytocin receptor proteins. In this study, cardiac Sca-1ϩ cells expressed low levels of oxytocin receptor mRNA that were positively regulated by oxytocin itself, and pretreatment with oxytocin antagonist completely inhibited oxytocin-induced expression of cardiac genes. These results suggest that oxytocin induces cardiomyocyte differentiation of cardiac Sca-1ϩ cells through oxytocin receptors. Furthermore Sca-1ϩ cells treated with oxytocin did not express osteogenic marker mRNAs, suggesting that oxytocin is not a nonspecific inducer like 5Ј-azacytizine but has some specificity for cardiac lineage.
Oxytocin receptors are coupled to G q/11 class GTP-binding proteins and stimulate the generation of inositol trisphosphate and diacylglycerol, leading to Ca 2ϩ release and activation of protein kinase C (37). Oxytocin stimulates cell proliferation through calcium (38,39) and protein kinase C pathways (38). proliferation through oxytocin receptors that lead to an increase in intracellular Ca 2ϩ and tyrosine phosphorylation. Tyrosine phosphorylation in oxytocin signaling has been reported to activate both p38 mitogen-activated protein kinase and extracellular signal-regulated kinase 2 (41,42). The mechanism by which oxytocin stimulates tyrosine phosphorylation has not been elucidated, but may be mediated by G␤␥ subunit dissociating from G ␣ subunit. Oxytocin inhibits the proliferation of human brain tumors (43), breast cancer cells (44), and adenocarcinoma of endometrium (45) via the cyclic adenosine monophosphate-protein kinase A pathway. Tahara et al. (46) have reported that the RhoA/Rho-kinase cascade is involved in oxytocin-induced rat uterine contraction. Among the considerable diversity of oxytocin-mediated signaling pathways, the specific pathway that activates cardiogenesis is currently unknown. Recently post-translational modification of cardiac transcription factors has been reported to be important for their transcriptional activities. Rho-like GTPases can phosphorylate GATA4 via activation of the p38 mitogen-activated protein kinase pathway, which enhances the potency of GATA4 (47). MEF2 is stimulated by calmodulin kinase activation in the heart (48). It remains to be determined which oxytocin signaling pathways are important for differentiation of cardiomyocytes.
It has been reported that c-kitϩ, Sca-1ϩ, lineage-, and CD34-/low fraction of bone marrow cells contain hematopoietic stem cells, which contribute to long term multilineage reconstitution of the blood system in mice (49). Orlic et al. (50) and Gojo et al. (10) have reported that c-kitϩ bone marrow cells and c-kitϩ bone marrow-derived mesenchymal cells transdifferentiate into cardiomyocytes in vivo, suggesting that c-kit is one of the cell surface markers of multipotent stem cells in bone marrow. The multipotential stem cells also reside in skeletal muscle, although the origin of the stem cells is still controversial (51). Skeletal muscle-derived stem cells reported by Qu-Petersen et al. (11) and Torrente et al. (13) highly express CD34 and Sca-1 but not c-kit and CD45 and differentiate into neural and endothelial cells. In our study, cardiac Sca-1ϩ cells expressed low levels of c-kit, suggesting that the features of stem cell markers on cardiac stem cells is distinct from bone marrowderived stem cells and rather similar to skeletal muscle-derived stem cells.
Tamaki et al. (14) isolated CD34ϩ and CD45Ϫ cells from the interstitial space of skeletal muscle, which highly expressed Sca-1 but not other endothelial progenitor cell markers. The CD34ϩ/CD45Ϫ cells differentiated into adipocytes, endothelial and myogenic cells and expressed Bcrp1/ABCG2 gene mRNA, which is an important determinant of the SP phenotype. Recently Polesskaya et al. (52) have reported that CD45ϩ/Sca-1ϩ cells from injured skeletal muscle differentiate into myoblasts much more than CD45Ϫ/Sca-1 cells. Because of the hematopoietic restricted expression of CD45 antigen, skeletal myogenic CD45ϩ/Sca-1ϩ cells might be of hematopoietic origin. In our study, cardiac Sca-1ϩ cells expressed low levels of CD34 and ϳ40% of the cardiac Sca-1ϩ cells expressed CD45, one of hematopoietic cell markers. We sorted cardiac Sca-1ϩ cells on the basis of CD45 expression and cultured them with oxytocin. Some Sca-1ϩ/CD45Ϫ cells expressed sarcomeric myosin after oxytocin treatment, but no Sca-1ϩ/CD45ϩ cells expressed myosin (data not shown), suggesting that Sca-1ϩ cells that can differentiate into cardiomyocytes are in the CD45Ϫ population. Therefore, in terms of the expression of CD34 and CD45, the cardiac muscle stem cells are distinct from the previously reported skeletal muscle-derived stem cells.
Sca-1ϩ cells from the adult heart expressed GATA4 and Csx/Nkx-2.5, but not Oct-3/4 before treatment with oxytocin (data not shown), suggesting that the Sca-1ϩ cells are committed to cardiomyocytes to some degree. Makino et al. (24) have reported that mouse bone marrow-derived mesenchymal stem cells (CMG cells) differentiate into cardiomyocyte after 5Ј-azacytizine treatment. Although the cell surface antigens of CMG cells were not analyzed, the bone marrow-derived mesenchymal stem cells, which differentiated into cardiomyocytes after 5Ј-azacytizine treatment in vivo, expressed Sca-1, c-kit, and CD34 (10), suggesting that the cardiac Sca-1ϩ cells are different from bone marrow-derived mesenchymal stem cells. Cardiac Sca-1ϩ cells differentiated into osteocytes and adipocytes in appropriate conditions, suggesting that cardiac Sca-1ϩ cells have the intragerm layer multipotency. It remains to be determined whether the cardiac Sca-1ϩ cell population contains stem cells capable of differentiating to extra germ layer lineage.
The spontaneously beating differentiated cardiac Sca-1ϩ cells showed [Ca 2ϩ ] i transients and treatment with isoproterenol increased the frequency of [Ca 2ϩ ] i transients and beating rate. The similar response to isoproterenol has been reported in adult murine cardiomyocytes (53), embryonic stem cells-derived cardiomyocytes (54), and CMG cells (55). The ␤ 1 -selective blocker, CGP20712A, significantly reduced isoproterenol-induced increase in beating rate to the same extent as the nonselective ␤-blocker, propranolol, but the ␤ 2 -selective blocker, ICI118551, did not. These results suggest that the ␤ 1 receptor is the predominant subtype that mediates the changes in beating rate of cardiomyocytes derived from Sca-1ϩ cells.
During the preparation of this article, two studies on cardiac stem cells were reported (56,57). They have shown that c-kitϩ or Sca-1ϩ cells derived from the adult murine heart express cardiac genes and proteins after the cardiogenic induction. We showed for the first time that there are potential adult cardiac stem cells that have an ability to proliferate and differentiate into various types of cells including beating cardiomyocytes in vitro. Although the role of cardiac stem cells is uncertain, our results suggest their possible role in cardiac repair. In addition, the understanding of precise molecular mechanisms of the differentiating process of cultured cardiac stem cells may provide us with new insights into cardiac development and regeneration.