Cardiomyogenic Potential of C-Kit+–Expressing Cells Derived From Neonatal and Adult Mouse Hearts
Background— C-kit is a receptor tyrosine kinase family member expressed in hematopoietic stem cells. C-kit is also transiently expressed in cardiomyocyte precursors during development and in a rare cell population in the normal adult heart. In the present study, the cardiomyogenic potential of c-kit+ cells isolated from normal neonatal, normal adult, and infarcted adult mouse hearts was evaluated.
Methods and Results— Magnetic activated cell sorting was used to prepare c-kit+ cells from the hearts of ACT-EGFP/MHC-nLAC double transgenic mice. These animals exhibit widespread enhanced green fluorescent protein (EGFP) expression and cardiomyocyte-restricted nuclear β-galactosidase activity, thus permitting simultaneous tracking of cell survival and differentiation. A subset of the c-kit+ cells from double transgenic neonatal hearts acquired a cardiomyogenic phenotype when cocultured with fetal cardiomyocytes (2.4% of all EGFP+ cells screened) but rarely when cultured alone or when cocultured with mouse fibroblasts (0.03% and 0.05% of the EGFP+ cells screened, respectively). In contrast, c-kit+ cells from normal adult double transgenic hearts failed to undergo cardiomyogenic differentiation when cocultured with nontransgenic fetal cardiomyocytes (>18 000 EGFP+ cells screened) or when transplanted into normal or infarcted adult mouse hearts (14 EGFP+ grafts examined). A single c-kit+ cell from an infarcted double transgenic adult heart was observed to acquire a cardiomyogenic phenotype in coculture (>37 000 EGFP+ cells screened).
Conclusions— These data suggest that the ability of cardiac-resident c-kit+ cells to acquire a cardiomyogenic phenotype is subject to temporal limitations or, alternatively, that the cardiomyogenic population is lost. Elucidation of the underlying molecular basis may permit robust cardiomyogenic induction in adult-derived cardiac c-kit+ cells.
Received September 13, 2009; accepted February 22, 2010.
Transplantation of donor myocytes or myogenic stem cells is emerging as a potential therapeutic intervention for the treatment of heart failure.1 Consequently, considerable effort is being invested to identify markers for stem cells with cardiomyogenic activity. C-kit (also known as CD117) is a member of the receptor tyrosine kinase family that is expressed at high levels in a number of hematopoietic progenitors, including hematopoietic stem cells, multipotent progenitors, and common myeloid progenitors.2–4 C-kit is also expressed in early thymocyte progenitors, mast cells, melanocytes, interstitial cells of Cajal, and prostate stem cells.
Editorial see p 1981
Clinical Perspective on p 2000
C-kit expression has been observed recently in cardiovascular progenitors. For example, Wu and colleagues5 used a reporter transgene to isolate cells expressing the cardiomyogenic transcription factor Nkx-2.5 in differentiating embryonic stem cell cultures. A subset of these cells expressed modest levels of c-kit and gave rise to both cardiomyocytes and smooth muscle cells in vitro, suggesting that c-kit marked bipotent cardiovascular progenitors. A similar approach was used by Christoforou and colleagues,6 who further demonstrated that cardiomyocytes, smooth muscle cells, and endothelial cells could be derived from the subpopulation of cells expressing c-kit, Nkx2–5, and Flk-1. Flk-1 was previously reported to be expressed in cardiovascular progenitors derived from embryonic stem cells.7
C-kit expression has also been reported in cardiovascular precursors during in vivo development. Using a BAC reporter transgene expressing enhanced green fluorescent protein (EGFP) under the regulation of the c-kit promoter, Tallini and colleagues8 demonstrated that neonatal hearts contain cells coexpressing c-kit and Flk-1. C-kit reporter transgene expression was also observed in neonatal cardiomyocytes with α-actinin immune reactivity, and expression levels appeared to be inversely related to the level of differentiation. These data are consistent with the notion that c-kit expression marks cardiomyogenic precursors and that expression is extinguished with terminal differentiation. In support of this, Tallini and colleagues further demonstrated that clonally amplified EGFP-expressing cells from neonatal hearts carrying the c-kit reporter transgene gave rise to cardiomyocytes, smooth muscle cells, and endothelial cells, similar to observations for embryonic stem cell–derived c-kit+ cells.8 Transient c-kit expression in neonatal cardiomyocytes was also observed by Li and colleagues9 via immune cytological analyses and was thought to be critical for the termination of cardiomyocyte cell cycle activity.
The role of c-kit–expressing cells in the adult heart is less clear. Experiments with adult mice with diminished levels of c-kit activity10 or with mice carrying reporter transgenes8 suggested that c-kit–expressing cells are predominantly involved with postinjury revascularization and beneficial myofibroblast-mediated remodeling. Other studies suggest that c-kit immune reactivity in the adult heart is limited to mast cells.11 In contrast, transplantation of in vitro amplified c-kit+ cells from adult rat12 or human13 hearts was thought to result in overt myocardial regeneration, with the transplanted c-kit+ cells giving rise to endothelial cells, smooth muscle cells, and cardiomyocytes. In support of this, Kubo and colleagues14 demonstrated that adenovirus-transduced c-kit+ cells from failing human hearts could give rise to cardiomyocytes in vitro. Unfortunately, differences in the experimental approaches and readouts employed in these various studies make it difficult to retrospectively evaluate the relative cardiomyogenic potential of cardiac-resident c-kit+ cells at different stages of development.
Coculture with cardiomyocytes has been used previously to characterize the cardiomyogenic potential of a number of progenitor cell populations.14–16 In the present study, coculture with fetal mouse cardiomyocytes was employed to directly compare the cardiomyogenic potential of cardiac-resident c-kit+ cells isolated from normal neonatal, normal adult, and infarcted adult mouse hearts. The experiments utilized a combination of reporter transgenes and immune cytology to monitor cell survival and cardiomyogenic potential. A subset of the c-kit+ cells isolated from neonatal hearts acquired a cardiomyogenic phenotype in coculture, independent of cell fusion events. However, this activity was markedly diminished in c-kit+ cells isolated from normal or infarcted adult hearts. Moreover, c-kit+ cells from adult hearts failed to undergo cardiomyogenic differentiation when transplanted into either normal or infarcted adult hearts. These data suggest that the ability of cardiac-resident c-kit+ cells to acquire a cardiomyogenic phenotype is subject to temporal limitations or, alternatively, that the cardiomyogenic population is lost.
See Methods in the online-only Data Supplement for a complete description.
ACT-EGFP (C57BL/6-Tg[ACTB-EGFP]1Osb/J) and (DBA/2J×C57Bl/6J)f1 mice were from the Jackson Laboratory (Bar Harbor, Me). MHC-nLAC mice17 utilize the mouse α-cardiac myosin heavy chain promoter to target expression of a nuclear β-galactosidase reporter. NOD/SCID-IL2-γ null mice were from the Indiana University Simon Cancer Center.
Flow Cytometry of C-Kit+ Cells
Cardiac cells were isolated by mincing whole neonatal or adult hearts in 0.1% collagenase IV, incubating for 45 minutes at 37°C,18 and then filtering through a 40-μm mesh. Alternatively, whole hearts were minced in Liberase/Blendzyme for 45 minutes at 37°C. Cells were stained with PerCP-conjugated antibody against CD45 or anti-c-kit-PE antibody for 45 minutes. Cells were analyzed with the use of a FACSCalibur flow cytometer.
Magnetic-Activated Cell Sorting Isolation of C-Kit+ Cells From the Heart
Cells from enzymatically dispersed whole hearts (normal or infarcted) were reacted with magnetic beads conjugated with mouse anti-c-kit antibody for 20 minutes and subsequently separated by magnetic-activated cell sorting (MACS) into c-kit+ and c-kit− fractions.
Coculture of C-Kit+ Cells With Fetal Mouse Cardiomyocytes
Embryonic day 15 whole hearts were dissociated with collagenase I for 60 minutes at 37°C, plated on gelatin-coated dishes at a density of 25 000 to 50 000 cells per cm2, and cultured for 24 hours in Dulbecco’s modified Eagle’s medium supplemented with 10% fetal bovine serum and antibiotics. The medium was replaced with Dulbecco’s modified Eagle’s medium supplemented with 10% fetal bovine serum, 10 μg/mL insulin, 10 μg/mL transferrin, and antibiotics and seeded with freshly isolated c-kit+ cells at a density of 2500 to 5000 cells per cm2.
Myocardial Infarction and Intracardiac Grafting
Myocardial infarction was performed in 12-week-old mice as described previously.19 For intracardiac grafting, donor cells suspended in 3 μL phosphate-buffered saline were injected into the anterior and posterior infarct border zones of the ischemic myocardium after coronary ligation or in normal hearts.
Characterization and Isolation of Myocardial C-Kit+ Cells
Fluorescence-activated cell sorting (FACS) was used to quantify and characterize the c-kit+ cells in normal neonatal and adult hearts, as well as in infarcted adult hearts. C-kit+ cells from adult bone marrow were used as a reference. Isotype-matched antibodies were used to set threshold levels for detecting c-kit and CD45 immune reactivity. C-kit+/CD45− and c-kit+/CD45+ populations were readily observed in representative dot plots from each source (Figure 1A). Quantitative analyses (Figure 1B and 1C) revealed that ≈9% of the mononucleated cells from the bone marrow expressed c-kit, the majority of which (>90%) coexpressed the hematopoietic lineage marker CD45. In the neonatal heart, 0.65% of the cells were c-kit+, and of these, ≈10% were CD45+. In the adult heart, ≈0.5% of the cells expressed c-kit, of which ≈15% coexpressed CD45. Myocardial infarction via permanent coronary artery occlusion resulted in an ≈3-fold increase in the number of cells expressing c-kit in adult hearts, of which the majority (76%) coexpressed CD45 when analyzed at 7 days after infarction, in agreement with published observations.10
MACS was employed to isolate c-kit+ cells from mice carrying the ACT-EGFP reporter transgene (which targets EGFP expression under the β-actin promoter).20 The vast majority of cells obtained from neonatal or normal adult hearts after 2 cycles of MACS purification exhibited EGFP epifluorescence (Figure 2A and 2B, respectively). Immune cytology analyses revealed that >90% of cells with EGFP epifluorescence also exhibited c-kit immune reactivity. Similar results were obtained from infarcted adult hearts (not shown).
C-Kit+ Cells Derived From Neonatal Hearts Acquire a Cardiac Phenotype After Coculture With Fetal Cardiomyocytes
Coculture with cardiomyocytes was used to quantify the cardiomyogenic potential of cardiac-resident c-kit+ from various stages of development. Cells were prepared from double transgenic mice carrying the ACT-EGFP and MHC-nLAC reporter transgenes. As shown above, the ACT-EGFP reporter targets EGFP expression to most cardiac-resident c-kit+ cells. The MHC-nLAC reporter targets cardiomyocyte-restricted nuclear β-galactosidase activity, which is detected by reacting with the chromogenic β-galactosidase substrate X-GAL.17 Thus, the presence of c-kit+ cells in cocultures could be readily quantified via the presence of EGFP epifluorescence or anti-EGFP immune reactivity, and their cardiomyogenic differentiation could be quantified by colocalization of anti-EGFP and anti-α-actinin immune reactivity or by induction of nuclear β-galactosidase activity. Control experiments using fetal cardiomyocytes confirmed the fidelity of the reporter transgenes for these analyses (Figure I in the online-only Data Supplement).
Fetal cardiomyocytes from nontransgenic mice were plated, and 24 hours later the cultures were seeded with MACS-isolated c-kit+ cells prepared from the hearts of neonatal ACT-EGFP/MHC-nLAC double transgenic mice. After 7 days of coculture, a subset of the cells with EGFP epifluorescence exhibited robust contractile activity (Movie I in the online-only Data Supplement). The samples were fixed, reacted with X-GAL, and processed for EGFP (Alexa488-conjugated antibody) and α-actinin (rhodamine-conjugated secondary antibody) immune reactivity. A subset of the EGFP-expressing cells exhibited highly mature sarcomeric structure as evidenced by α-actinin immune reactivity (Figure 3); moreover, all of the EGFP-positive cells with α-actinin immune reactivity also exhibited nuclear β-galactosidase activity (Figure 3, inset). Two percent to 3% of the neonatal heart–derived c-kit+/EGFP+ cells exhibited a cardiomyogenic phenotype when cocultured with nontransgenic fetal cardiomyocytes (Table). In contrast, EGFP-expressing, β-galactosidase–positive cardiomyocytes were only detected rarely when the neonatal heart–derived c-kit+ cells were cultured alone or when cocultured with NIH-3T3 cells (Table).
To determine whether cell fusion contributed to the apparent cardiomyogenic events, fetal cardiomyocyte cultures were prepared from single transgenic MHC-nLAC mice. Twenty-four hours later, the cultures were seeded with MACS-purified c-kit+ cells prepared from the hearts of single transgenic neonatal ACT-EGFP mice. The cultures were fixed 7 days later and processed as described above. Two percent to 3% of the cells with EGFP immune reactivity also exhibited α-actinin immune reactivity (Figure 4). However, the preponderance of these cells did not exhibit MHC-nLAC reporter transgene activity (which was detected readily in cardiomyocytes lacking EGFP epifluorescence; Figure 4, inset). Indeed, only 2 cells with MHC-nLAC reporter transgene activity were observed when >3500 EGFP-expressing cells were screened (Table), suggesting a cardiomyocyte/c-kit+ cell fusion rate of 0.05% under these assay conditions.
C-Kit+ Cells Derived From Adult Hearts Have Very Limited Cardiomyogenic Potential After Coculture With Fetal Cardiomyocytes or Intracardiac Engraftment
Next, c-kit+ cells prepared from normal adult mice were tested for cardiomyogenic activity. Fetal cardiomyocytes from nontransgenic mice were plated, and 24 hours later the cultures were seeded with MACS-isolated c-kit+ cells prepared from the hearts of adult ACT-EGFP/MHC-nLAC double transgenic mice. After 7 days of coculture, the samples were fixed, reacted with X-GAL, and processed for EGFP and α-actinin immune reactivity. Cells with green EGFP immune reactivity were readily identified, indicating that the adult cardiac-resident c-kit+ cells survived in coculture. However, these cells lacked detectable α-actinin immune reactivity (Figure 5) and failed to activate the MHC-nLAC reporter transgene (Figure 5, inset). No cardiomyogenic events were detected when >18 000 EGFP-expressing c-kit+ cells were screened (Table).
To determine whether the adult heart provides an environment more conducive to their cardiomyogenic differentiation, MACS-isolated c-kit+ cells prepared from normal adult ACT-EGFP/MHC-nLAC double transgenic mice were engrafted into normal or infarcted (via permanent coronary artery ligation) recipient hearts. Syngeneic (C57Bl/6J×DBA/2J)F1 or immunocompromised NOD/SCID-γ recipients were used. The hearts were harvested at 10 to 21 days after engraftment, fixed, and sectioned. The sections were then reacted with X-GAL and processed for EGFP and α-actinin immune reactivity. Although cells with EGFP immune reactivity were detected within the engrafted hearts, they lacked α-actinin immune reactivity (Figure 6) and failed to activate the MHC-nLAC reporter (Figure 6, inset). No cardiomyogenic events were detected in 14 hearts successfully engrafted with adult heart–derived c-kit+ cells (Table I in the online-only Data Supplement). Control experiments with double transgenic fetal cardiomyocyte donor cells confirmed the fidelity of the reporter transgenes for these analyses (Figure II in the online-only Data Supplement).
Finally, to determine whether myocardial infarction enhances the cardiomyogenic potential of cardiac-resident c-kit+ cells, adult ACT-EGFP/MHC-nLAC double transgenic mice were subjected to permanent coronary artery ligation. Seven days later, c-kit+ cells were prepared via MACS purification and seeded onto cultures of fetal cardiomyocytes from nontransgenic mice that were prepared 24 hours earlier. After 7 days of coculture, the samples were fixed and processed as described above. Once again, the EGFP-expressing c-kit+ cells readily survived in coculture. A single EGFP+ cell exhibited α-actinin immune reactivity, and MHC-nLAC reporter activity was observed (Figure 7). The rest of the >37 000 c-kit+ cells examined in the cocultures lacked α-actinin immune reactivity and failed to activate the MHC-nLAC reporter (Figure 8 and Table).
The data presented here indicate that a subpopulation of the cardiac-resident c-kit+ cells isolated from neonatal mouse heart exhibits an overt cardiomyogenic phenotype when cocultured with fetal cardiomyocytes. This phenotype was largely absent in cardiac-resident c-kit+ cells isolated from normal or infarcted adult mouse hearts. Importantly, the use of reporter transgenes permitted straightforward quantification of c-kit+ cell survival and differentiation, and there was extremely high concordance between the results obtained via X-GAL reaction (ie, monitoring induction of the MHC-nLAC reporter) and α-actinin immune reactivity (Table). Thus, the fidelity of the reporter transgene readout is quite high for identifying well-differentiated cardiomyocytes. Collectively, these data suggest a developmental loss of cardiomyogenic activity in c-kit+ cells in the transition from neonatal to adult life in mice or, alternatively, a loss of the cardiomyogenic population.
Previous studies suggested that cell fusion could contribute to apparent transdifferentiation events in vitro and in vivo.21,22 Although cell fusion events could be detected when MHC-nLAC fetal cardiomyocytes were cocultured with neonatal ACT-EGFP c-kit+ cells, the frequency of these events was >50-fold lower than the frequency of putative cardiomyogenic events (0.05% versus 3%; Table). Carryover of differentiated cells from the neonatal c-kit+ heart preparations could also contribute to the apparent cardiomyogenic events. However, immune cytology analyses immediately after MACS isolation indicated that only 0.11% of the cells exhibited α-actinin immune reactivity (9.82% exhibited smooth muscle actin immune reactivity, and 0.24% exhibited isolectin 4B immune reactivity; Figure IX and Table II in the online-only Data Supplement). The observation that the percentage of cells with α-actinin immune reactivity after MACS (0.11%) was 22- to 27-fold lower than the percentage of α-actinin–positive cells expressing the reporter transgenes after coculture with nontransgenic fetal cardiomyocytes (2.4% to 3%; Table) suggests that simple carryover of neonatal cardiomyocytes does not account for the observed results. Moreover, the presence of a roughly similar content of α-actinin+ cells immediately after MACS isolation versus 7 days of single culture or NIH-3T3 coculture (Table) argues against a role of enhanced survival and/or seeding of carryover neonatal cardiomyocytes in the fetal cardiomyocyte cocultures. Collectively, the data are consistent with the notion that a small portion of cardiac-resident c-kit+ cells from neonatal hearts can undergo cardiomyogenic differentiation when cocultured with fetal cardiomyocytes. The data do not distinguish whether these cells represent a true cardiomyogenic stem cell or, alternatively, a c-kit+ developmental intermediate that requires a coculture environment to acquire a cardiomyogenic phenotype. FACS followed by coculture assay (Figure X and Table III in the online-only Data Supplement) revealed that the apparent cardiomyogenic activity is present in the c-kit+/CD45− subpopulation of cells isolated from the neonatal hearts.
In contrast, the ability of c-kit+ cells isolated from normal or infarcted adult mouse hearts to acquire a cardiomyogenic phenotype was limited, with only a single potential cardiomyogenic event detected when >56 000 cardiac-resident c-kit+ cells were screened. Moreover, the origin of the single EGFP+/α-actinin+/β-galactosidase+ cardiomyocyte observed in cocultures with c-kit+ cells from infarcted adult heart is not clear. It is equally likely that this cell originated from carryover of a border zone cardiomyocyte that reinduced c-kit expression,8 from a fusion event between a c-kit+ cell and a fetal cardiomyocyte, or from a bona fide cardiomyogenic differentiation event. Unfortunately, the rarity of the event precludes a systematic assessment of its origin. These data suggest that the cardiomyogenic c-kit+ subpopulation present in neonatal hearts is lost on maturation or, alternatively, loses its ability to undergo cardiomyogenic conversion when cocultured with fetal cardiomyocytes.
These results differ from those of several in vitro studies using adult heart–derived c-kit+ cells from rats and humans. Although single culture of adult heart–derived c-kit+ cells resulted in only a rudimentary cardiomyocyte phenotype (as evidenced by the induction of a limited number of myocyte markers and the absence of myofiber structure),12 coculture with cardiomyocytes resulted in more robust cardiomyogenic differentiation.13,14,23 It is noteworthy that before the establishment of cocultures, these latter experiments all employed varying degrees of in vitro manipulation of the c-kit+ cells, including prolonged amplification of the cells or exposure to adenoviruses in suspension culture. It is possible that these manipulations imparted a certain degree of reprogramming that enhanced cardiomyogenic potential. Indeed, increased expression of GATA-4 was observed in long-term cultures of adult heart–derived c-kit+ cells.23 Although subtle differences in methodologies might have altered our ability to observe overt cardiomyogenic induction in adult c-kit+ cells, the observation that neonatal c-kit+ cells were cardiomyogenic when subjected to the same protocols underscores a fundamental difference between heart-derived c-kit+ cell populations prepared from different developmental stages. Adult heart–derived c-kit+ cells also failed to give rise to overt cardiomyogenesis after engraftment into normal or infarcted mouse hearts in our experiments, despite survival of the donor cells. These results contrast with previous data reporting extensive cardiac regeneration after direct injection of c-kit+–derived cells from rat or human hearts.12,13,24 Once again, in vitro manipulation of the cells in these latter studies may have contributed to the discrepant results.
The observations reported here are supported by other studies. For example, Tallini and colleagues8 showed that c-kit+ cells from neonatal mouse hearts could be clonally amplified and give rise to cardiomyocytes, smooth muscle cells, and endothelial cells. In addition, c-kit expression was observed to mark cardiomyocyte and smooth muscle precursors in fetal hearts and differentiating embryonic stem cell cultures.5,6 Thus, the apparent cardiomyogenic activity in c-kit+ cells derived from neonatal hearts observed here is perhaps not surprising. It is likely that the coculture experiments used in the present study mimicked the cardiomyogenic conditions developed by Tallini and colleagues. Although these authors did not examine the cardiomyogenic potential of c-kit+ cells isolated from adult hearts, they demonstrated that expression of a c-kit–promoted reporter transgene was predominantly associated with smooth muscle and endothelial cells after cryoinjury of the heart. Reporter gene expression observed in border zone cardiomyocytes was attributed to a previously described reactivation of c-kit expression after myocardial injury.9 The absence of overt cardiomyogenic activity in adult heart–derived c-kit+ cells after coculture or intracardiac injection in the present study is consistent with these results. These results are also consistent with the low levels of cumulative cardiomyocyte renewal observed in individuals alive during above-ground nuclear testing.25 Presently, it is not clear whether the neonatal c-kit+ cells with apparent cardiomyogenic activity are cardiac resident or whether they populate the heart after migration from an extracardiac site. The absence of cardiomyogenic induction in cocultures with c-kit+ cells isolated from neonatal day 2 (Figure XI and Table III in the online-only Data Supplement) or adult19 bone marrow suggests that the cells are not from the marrow, with the caveat that if residence in a cardiac niche was required to acquire cardiomyogenic activity in marrow-derived cells, this assay would not detect it.
The normal adult mouse heart contains ≈1.5×106 cardiomyocytes, which comprise 14.2% of the total cell number (Reference 26 and M.S., unpublished data, 2009); consequently, there are ≈9×106 noncardiomyocytes in the adult mouse heart. FACS analyses revealed that ≈0.5% of the noncardiomyocytes are c-kit+ (Figure 1) in the uninjured adult heart. Thus, there should be ≈45 000 c-kit+ cells per heart. Because we failed to see any cardiomyogenic events when 18 866 c-kit+ cells from normal hearts were screened, these data suggest that there would likely be very few (ie, <3) cardiomyogenic c-kit+ stem cells in the normal adult mouse heart. In infarcted mouse hearts, the c-kit+ population increases to 1.4% at 7 days after injury. If one assumes a roughly similar number of total cells, this would translate to 126 000 c-kit+ cells per infarcted adult heart. Because we observed only a single potential cardiomyogenic event (aforementioned caveats notwithstanding) when 37 372 cells were screened, these data suggest that there would likely be <4 cardiomyogenic c-kit+ cells per adult mouse heart at 7 days after infarction.
Limitations of the Study
The present study does not exclude the possibility of a cardiomyogenic c-kit+ population in vivo in the normal or injured adult mouse heart. However, the quantitative analyses suggest that if such cells do exist, they are extremely rare (within the limits of the cell isolation protocols and cardiomyogenic assays employed). It is possible that long-standing heart failure is required to obtain c-kit+ cells with cardiomyogenic activity. Although Kubo and colleagues14 demonstrated that c-kit+ cells from failing human hearts were cardiomyogenic in coculture, it is not clear whether cells from nonfailing hearts were similarly examined. It is possible that our results reflect an intrinsic species difference because all of the in vitro differentiation and cell transplantation studies reported to date used rat or human cells; thus far, only studies with circumstantial data implying cardiomyogenic activity in cardiac-resident c-kit+ cells from adult mouse hearts in the absence of genetic enhancement have been reported.27–29 It is also possible that our fetal cardiomyocyte cultures may lack the factors needed to promote differentiation of adult cardiac-resident c-kit+ cells, despite being sufficient to induce differentiation of neonatal-derived cells (although coculture with neonatal cardiomyocyte preparations also failed to promote differentiation; Figure XII in the online-only Data Supplement). Finally, our experiments did not address benefits of c-kit+ cells on ventricular function after infarction that result from noncardiomyogenic activity.
We thank Dorothy Field for technical support.
Sources of Funding
This study was supported by the National Institutes of Health (HL083126) and the German Research Foundation (DFG ZA 575/1-1).
Rubart M, Field LJ. Cardiac regeneration: repopulating the heart. Ann RevPhysiol. 2006; 68: 29–49.
Tallini YN, Greene KS, Craven M, Spealman A, Breitbach M, Smith J, Fisher PJ, Steffey M, Hesse M, Doran RM, Woods A, Singh B, Yen A, Fleischmann BK, Kotlikoff MI. c-kit expression identifies cardiovascular precursors in the neonatal heart. Proc Natl Acad Sci U S A. 2009; 106: 1808–1813.
Li M, Naqvi N, Yahiro E, Liu K, Powell PC, Bradley WE, Martin DIK, Graham RM, Dell'Italia LJ, Husain A. c-kit is required for cardiomyocyte terminal differentiation. Circ Res. 2008; 102: 677–685.
Bearzi C, Rota M, Hosoda T, Tillmanns J, Nascimbene A, De Angelis A, Yasuzawa-Amano S, Trofimova I, Siggins RW, Lecapitaine N, Cascapera S, Beltrami AP, D'Alessandro DA, Zias E, Quaini F, Urbanek K, Michler RE, Bolli R, Kajstura J, Leri A, Anversa P. Human cardiac stem cells. Proc Natl Acad Sci U S A. 2007; 104: 14068–14073.
Kubo H, Jaleel N, Kumarapeli A, Berretta RM, Bratinov G, Shan X, Wang H, Houser SR, Margulies KB. Increased cardiac myocyte progenitors in failing human hearts. Circulation. 2008; 118: 649–657.
Pfister O, Mouquet F, Jain M, Summer R, Helmes M, Fine A, Colucci WS, Liao R. CD31- but not CD31+ cardiac side population cells exhibit functional cardiomyogenic differentiation. Circ Res. 2005; 97: 52–61.
Soonpaa MH, Koh GY, Klug MG, Field LJ. Formation of nascent intercalated disks between grafted fetal cardiomyocytes and host myocardium. Science. 1994; 264: 98–101.
Zaruba MM, Theiss HD, Vallaster M, Mehl U, Brunner S, David R, Fischer R, Krieg L, Hirsch E, Huber B, Nathan P, Israel L, Imhof A, Herbach N, Assmann G, Wanke R, Mueller-Hoecker J, Steinbeck G, Franz WM. Synergy between CD26/DPP-IV inhibition and G-CSF improves cardiac function after acute myocardial infarction. Cell Stem Cell. 2009; 4: 313–323.
Murry CE, Soonpaa MH, Reinecke H, Nakajima H, Nakajima HO, Rubart M, Pasumarthi KB, Virag JI, Bartelmez SH, Poppa V, Bradford G, Dowell JD, Williams DA, Field LJ. Haematopoietic stem cells do not transdifferentiate into cardiac myocytes in myocardial infarcts. Nature. 2004; 428: 664–668.
Rota M, Padin-Iruegas ME, Misao Y, De Angelis A, Maestroni S, Ferreira-Martins J, Fiumana E, Rastaldo R, Arcarese ML, Mitchell TS, Boni A, Bolli R, Urbanek K, Hosoda T, Anversa P, Leri A, Kajstura J. Local activation or implantation of cardiac progenitor cells rescues scarred infarcted myocardium improving cardiac function. Circ Res. 2008; 103: 107–116.
Bergmann O, Bhardwaj RD, Bernard S, Zdunek S, Barnabé-Heider F, Walsh S, Zupicich J, Alkass K, Buchholz BA, Druid H, Jovinge S, Frisén J. Evidence for cardiomyocyte renewal in humans. Science. 2009; 324: 98–102.
Gude N, Muraski J, Rubio M, Kajstura J, Schaefer E, Anversa P, Sussman MA. Akt promotes increased cardiomyocyte cycling and expansion of the cardiac progenitor cell population. Circ Res. 2006; 99: 381–388.
Limana F, Zacheo A, Mocini D, Mangoni A, Borsellino G, Diamantini A, De Mori R, Battistini L, Vigna E, Santini M, Loiaconi V, Pompilio G, Germani A, Capogrossi MC. Identification of myocardial and vascular precursor cells in human and mouse epicardium. Circ Res. 2007; 101: 1255–1265.
In recent years, intracoronary cell delivery of stem/progenitor cells has emerged as a potential therapy to restore impaired cardiac function. However, the efficacy of this treatment in improving ejection fraction has been moderate at best. The quest for a more potent stem/progenitor cell has identified cardiac-resident c-kit+ cells as a potential candidate. During embryonic heart development, primitive Nkx2.5+/c-kit+/Flk1+ cardiac progenitor cells have been shown to undergo cardiomyogenic differentiation. However, the role of c-kit+ cell populations in the adult heart is less clear. The study by Zaruba and colleagues utilized reporter transgenes in conjunction with in vitro and in vivo assays to monitor the survival and differentiation of cardiac-resident c-kit+ cell populations isolated from different stages of postnatal development. The data suggest that the neonatal heart harbors a resident c-kit+ cell population with cardiomyogenic potential. However, this activity is greatly reduced in the adult heart, suggesting that the cardiomyogenic activity is subject to temporal limitations or, alternatively, that the cardiomyogenic subpopulation is largely absent in the adult heart. Uncovering the molecular basis for these limitations may permit robust cardiomyogenic induction in adult-derived cardiac c-kit+ cells. With respect to clinical trials utilizing endogenous c-kit+ stem cell populations, this report underscores the importance of determining the extent to which secondary effects such as enhanced neoangiogenesis or inhibition of cardiomyocyte apoptosis contribute to functional improvement.
The online-only Data Supplement is available with this article at http://circ.ahajournals.org/cgi/content/full/CIRCULATIONAHA.109.909093/DC1.