|
|
||||||||
J Thorac Cardiovasc Surg 2008;135:673-678
© 2008 The American Association for Thoracic Surgery
Cardiothoracic Transplantation |
a Assistance Publique-Hôpitaux de Paris, Hôpital Européen Georges Pompidou, Department of Cardiovascular Surgery; University Paris-5, Faculty of Medicine, Paris, France
b Assistance Publique-Hôpitaux de Paris, Hôpital Européen Georges Pompidou, Department of Pathology; University Paris-5, Faculty of Medicine, INSERM U 872, Paris, France
c INSERM U 633, Laboratory of Biosurgical Research, Paris, France
d Assistance Publique-Hôpitaux de Paris, Hôpital Européen Georges Pompidou, Epidemiology and Clinical Research Unit, INSERM CIE4, Paris, France
e Assistance Publique-Hôpitaux de Paris, Hôpital Européen Georges Pompidou, Department of Anesthesiology, Paris, France
f Assistance Publique-Hôpitaux de Paris, Hôpital Européen Georges Pompidou, Department of Cardiovascular Surgery; University Paris-5, Faculty of Medicine, INSERM U 633, Paris, France
g Assistance Publique-Hôpitaux de Paris, Hôpital Européen Georges Pompidou, Department of Cardiovascular Surgery; University Paris-5, Faculty of Medicine, Paris, France
Received for publication July 31, 2007; revisions received September 26, 2007; accepted for publication October 26, 2007. * Address for reprints: Philippe Menasché, MD, PhD, Department of Cardiovascular Surgery, Hôpital Européen Georges Pompidou, 20, rue Leblanc, 75015 Paris, France. (Email: philippe.menasche{at}egp.aphp.fr).
| Abstract |
|---|
|
|
|---|
Methods: We prospectively screened 32 endomyocardial biopsies harvested from heart transplant recipients (off rejection episodes) and 18 right appendage biopsies collected during coronary artery bypass surgery, and processed the tissue specimens for the immunohistochemical detection of markers of stemness (c-kit, MDR-1, Isl-1), hematopoietic origin (CD45), mast cells (tryptase), endothelial cells (CD105), and cardiac lineage (Nkx2.5). Confocal microscopy was used for colocalization experiments. Three right appendage biopsies were also cultured for 2 to 3 weeks, at the completion of which c-kit–positive cells were sorted by flow cytometry.
Results: In endomyocardial biopsies, a median number of 2.7 (1.8–4) c-kit–positive cells/mm2 were found, and this number was even significantly smaller in right appendage biopsies (1 [0.5–1.8] c-kit–positive cell/mm2, P = .01). All of these c-kit–positive cells co-stained for CD45 and were more specifically identified as mast cells by their positive staining for the specific tryptase marker. However, none of the c-kit–positive cells expressed the markers of stemness MDR-1 and Isl-1 or colocalized with CD105. Flow cytometry confirmed the small number of c-kit–positive cells in cultured right atrial appendages.
Conclusion: These data raise a cautionary note on the therapeutic exploitation of cardiac stem cells in patients with ischemic cardiomyopathy, who may be the elective candidates for regenerative therapy.
| Introduction |
|---|
|
|
|---|
To the present, studies on CSCs have primarily focused on animal models, and few clinical data are available. The present study was designed to assess whether these cells could be identified in the myocardium in patients (particularly those with ischemic heart disease) in such a way that their subsequent isolation, expansion, and reinjection could become a clinically relevant procedure.
| Materials and Methods |
|---|
|
|
|---|
Immunohistochemistry
Samples were processed as previously described3
using antibodies recognizing markers of stemness (c-kit, dilution 1/50, Dako, Trappes, France; MDR-1, 1/250, Proteogenix, Oberhausbergen, France; Isl-1, 1/50, Developmental Studies Hybridoma Bank, Iowa City, Iowa), hematopoietic origin (CD45, 1/100, Dako), mast cells (tryptase, 1/200, Dako), endothelial cells (CD105, 1/100, Abcam, Cambridge, UK), and cardiac lineage (Nkx2.5, 1/50, R&D, Minneapolis, Minn). Both negative and positive controls were used to assess the specificity of the labeling. Negative controls were obtained by replacing primary mouse monoclonal antibodies against MDR-1, Isl-1, CD45, tryptase, Nkx2.5, and CD105 by normal mouse serum and rabbit polyclonal antibody against c-kit by normal rabbit serum. Positive controls were as follows: c-kit–positive tumor cells in a gastrointestinal stromal tumor; MDR-1–positive tubules in a human kidney obtained from the normal part of a nephrectomy specimen for cancer; Isl-1–positive nuclei in Langerhans islet cells obtained from a pancreas removed for neonatal hypoglycemia; tryptase-positive mast cells in an inflammatory skin lesion; CD105-positive endothelial cells in cardiac tissue; and Nkx2.5-positive nuclei in normal adult cardiac myocytes.
Cell Counts
Whole labeled tissue sections were scanned using an eyepiece grid with a microscope at x20 magnification. The number of c-kit–positive cells was normalized per square millimeter to account for the variable number of examined fields per patient, which in turn reflected the variable size of the tissue samples available for analysis.
Combined Immunofluorescence and Confocal Laser Microscopy
These procedures were performed as previously described.4
In brief, the rabbit polyclonal anti-c-kit antibody was first incubated and revealed by using a biotinylated anti-rabbit antibody (Dako) and streptavidin-cyanin-2 (Amersham, Les Ulis, France) or by using an Alexa 555-labeled anti-rabbit antibody (Invitrogen, Cergy-Pontoise, France); monoclonal antibodies against tryptase or CD45 were then incubated and revealed using a cyanin-3–labeled antimouse antibody, and Nkx2.5 and CD105 antibodies were revealed using a biotin-labeled antimouse antibody and streptavidin-cyanin-2. Double immunofluorescence was assessed with a confocal laser microscope (Leica, Rueil-Malmaison, France).
Cell Cultures
Three intraoperatively harvested right appendages were cultured according to previously described techniques developed to grow CSCs5,13,14
and assessed for the expression of c-kit by flow cytometry. Data were collected with CellQuest software and analyzed with ModFitLT2.0 software (BD Biosciences, Le-Pont-de-Claix, France).
Statistics
The results are expressed as means ± 1 standard deviation or median values (interquartile range). Comparison of c-kit–positive cell numbers between EMB and right appendage samples was performed by the Wilcoxon test, and the level of statistical significance was set at the .05 level.
| Results |
|---|
|
|
|---|
The median number of microscopic fields assessed per patient was 14 (range: 10.5–17.5) for EMBs and 30 (17–45) for right appendage biopsies. In EMBs, a median number of 2.7 [1.8;4] c-kit–positive cells/mm2 were found, regardless of the age of the donor and the time interval between transplantation and biopsy (
Figure 1, A). This number was significantly smaller in right appendage biopsies (1 [0.5;1.8] c-kit–positive cell/mm2, P = .01 vs EMB, Figure 1, B). Colocalization experiments demonstrated that all these c-kit–positive cells were of hematopoietic origin, as evidenced by their positive staining for CD45 (Figure 1, C). Their perivascular location suggested that the c-kit–positive cells could be mast cells, and this phenotype was confirmed by a consistently positive staining for the specific tryptase marker (Figure 1, D). However, none of the c-kit–positive cells identified in these tissue samples expressed the markers of stemness (MDR-1, Isl-1), whereas the antibodies used strongly stained control tissues (
Figure 2, A and B). Furthermore, the c-kit–positive cells did not colocalize with the cardiac lineage marker Nkx2.5 (Figure 2, C). Likewise, the CD105 marker stained endothelial cells in both EMBs and right atrial biopsies, but there was no overlay between this staining and the expression of the stem cell marker c-kit (Figure 2, D). The culture data tended to support the scarcity of c-kit–positive cells in right appendage tissue because they represented only 0.58% ± 0.67% of cells identified by flow cytometry after a 3- to 4-week period of cultures.
|
|
| Discussion |
|---|
|
|
|---|
Methodologic Issues
One major issue raised by CSCs is the multiplicity of markers that have been described to characterize them. At least 4 different markers, which are at times mutually exclusive, have been described: c-kit,5,6
Sca-1,7
side population,8,9
and Isl-1.10
It is difficult to reconcile the multiplicity of these phenotypic patterns with the limited (if any) self-regenerative capacity of the heart. Because the CSC population is not characterized in a more consistent way, we selected to identify these cells on the basis of c-kit, the receptor of stem cell factor, because experimentally clonogenic c-kit–positive cells have been reported to feature a greater ability to generate myocardial cells.5
Our results show that there are few c-kit–positive cells in the right septal wall of presumably normal adult hearts (those used for transplantation) and that this number is still lower in patients undergoing CABG for ischemic heart disease, who are actually better representatives of the target population for regenerative therapy. The perivascular location of these c-kit–positive cells was consistent with a mast cell phenotype, which was confirmed by their expression of the panhematopoietic marker CD45, the mast cell-specific marker tryptase (and an additional positive staining for toluidine blue; data not shown). By itself, this finding is not unexpected because the right appendage has been shown to harbor 0.5% to 1.5% of mast cells,11
and both cyclosporine and FK506 have been reported to up-regulate the expression of c-kit on a mast cell line,12
which could account for the greater number of the c-kit–positive cells in our EMB-derived tissue samples. The important observation is that none of these c-kit–positive cells expressed any of the other tested markers of stemness (MDR-1 and Isl-1) or of the more specific cardiac differentiation pathway (Nkx 2.5), thereby raising serious doubts about the presence (or the persistence) of a reservoir of CSCs in the adult myocardium. These findings are consistent with those of Yamabi and colleagues,13
who reported that c-kit–positive cells isolated from human fetal hearts were negative for cardiac markers. Their results differ from the experimental observation that cardiac c-kit–positive cells are lineage-negative but overall fit the paradigm that these progenitors are rare, on the order of 1 per 1 x 104 myocytes.14
Whether the hematopoietic c-kit–positive cells that we have identified normally reside in cardiac niches or dynamically migrate from the bone marrow cannot be determined from the present study. Likewise, our findings do not exclude that an initially present pool of CSCs was subsequently exhausted because of senescence or after its mobilization for repairing ischemic injuries because cardiac failure in humans has been reported to cause a 10-fold increase in the number of dividing cardiomyocytes compared with healthy hearts.15
This compensatory phenomenon may have emptied the reservoir of CSCs. The hypothesis that cardiac progenitors may disappear over time is also supported by the finding made in tissue samples retrieved during pediatric heart operations that Isl-1–positive cells have only been detected up to the age of 8 weeks.10
Comparison With Previous Studies
Studies performed in rat,6
dog,5
and pig16
have demonstrated that CSCs differentiated into cardiomyocytes after engraftment into infarcted myocardium, reduced infarct size, attenuated remodeling, and improved left ventricular function. In contrast, human data are limited to 3 clinical studies. In the study by Urbanek and colleagues,17
samples were retrieved from the hypertrophied left ventricular wall of patients undergoing aortic valve surgery, which makes the comparison difficult with our data. Conversely, the study by Messina and colleagues18
is more similar to ours in that it entailed derivation of human tissue from atrial or ventricular biopsy specimens taken during cardiac operations and subsequent culture that resulted in self-adherent clusters called cardiospheres. These cardiospheres expressed endothelial and stem cell markers (including c-kit), and cells derived from them displayed morphologic and phenotypical patterns characteristic of cardiomyocytes. A similar approach was subsequently described by Smith and colleagues,16
who used human right ventricular biopsy specimens to grow cardiospheres made of a mix population of cardiac progenitors, fibroblasts, and mesenchymal cells. These cardiospheres were primarily defined by a consistent expression of c-kit and CD105, which contrasts with our finding that these 2 markers were not coexpressed by the same cells and suggests that CD105 expression in cardiospheres may have reflected a contamination by endothelial cells. Although the putative CSCs did not spontaneously beat unless they were cocultured with rat cardiomyocytes in these 2 studies, their injection of the border zones of acute myocardial infarction in immunodeficient mice resulted in improved function and histologic patterns of cardiac regeneration. Altogether, these data have raised the appealing hypothesis that CSCs could be retrieved from an EMB16
or pharmacologically mobilized by growth factors5
and, after in vitro expansion, reinjected to effect myocardial regeneration.
However, the results of these studies should be interpreted cautiously because of the possibility of phenotypic changes induced by repeat passages, which could explain why the antigen expression pattern of these cultured cells was different from that of the unprocessed cells examined in the present study and possibly closer to the biological reality. Furthermore, the fact that CSCs only acquired their cardiomyocyte phenotype after co-cultures with rat cardiomyocytes makes it difficult to exclude the possibility that the cardiomyocytic differentiation of the engrafted cardiosphere-derived CSCs was not a mere consequence of fusion events.7
It is noteworthy that the patient demographics were not clearly detailed in any of the above mentioned studies (in one study, it was simply reported that the patients' ages ranged from 1 month to 80 years18
).
Study Limitations
This study was primarily based on immunostaining, which may always be fraught with technical artifacts, particularly when colocalization experiments are performed. Efforts were made to minimize these shortcomings with the use of confocal microscopy. Furthermore, the unequivocal expression of the cardiac markers Nkx2.5, MDR-1, and Isl-1 in our positive controls makes it unlikely that these markers would have been missed if they had been expressed by the c-kit–positive cells. We also acknowledge that deriving quantitative data from immunohistochemical observations may not be accurate, but it is unlikely that it may have led to a gross underestimation of the number of CSCs, given our method of counting. Second, one could argue that we did not test the whole set of stemness or cardiac differentiation markers. We did not assess the expression of Sca-1 because the cardiac differentiation of Sca-1–positive cells requires exposure to 5-azacytidine,7
which is clinically irrelevant. Furthermore, the ability of cells expressing this marker to give rise to cardiomyocytes is 6 times less than that of c-kit–positive cells.5
We did not screen the tissue samples for GATA-4 or MEF-2 because of the lack of specific monoclonal antibodies that were required for the co-staining experiments with our anti-c-kit polyclonal antibody. A third limitation of the results obtained in the surgical group of patients is that sampling was performed in the right atrium. This was intentionally done to mimic a clinical scenario whereby CSCs would be harvested by an EMB and does not allow the exclusion of the presence of resident cardiac progenitors in other areas of the heart, for which the accessibility would then be more technically challenging. Finally, one could argue that we only provided a snapshot of the c-kit–positive cell content in biopsies without subsequent sorting and single cell clonal expansion and that (although initially small) the number of these cells could be expanded in vitro. This is conceivable, although not supported by our flow cytometry data that demonstrated only a small proportion of c-kit–positive cells after culture. Given the minute amount of starting materials, sorting of the candidate cells and their expansion (without loss of the cardiac differentiation potential) under Good Manufacturing Practice are likely to be technically challenging and time-consuming procedures, notwithstanding the delay in onset of cell therapy, which may be a concern in these often critically ill patients. The individual variability in cell functionality and the cost of customized quality controls may further complicate the clinical practicality of this patient-specific approach.
| Conclusions |
|---|
|
|
|---|
| Acknowledgments |
|---|
| Footnotes |
|---|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
G. M. Ellison, D. Torella, S. Dellegrottaglie, C. Perez-Martinez, A. Perez de Prado, C. Vicinanza, S. Purushothaman, V. Galuppo, C. Iaconetti, C. D. Waring, et al. Endogenous Cardiac Stem Cell Activation by Insulin-Like Growth Factor-1/Hepatocyte Growth Factor Intracoronary Injection Fosters Survival and Regeneration of the Infarcted Pig Heart J. Am. Coll. Cardiol., August 23, 2011; 58(9): 977 - 986. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. C. Sturzu and S. M. Wu Developmental and Regenerative Biology of Multipotent Cardiovascular Progenitor Cells Circ. Res., February 4, 2011; 108(3): 353 - 364. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. Mishra, K. Vijayan, E. J. Colletti, D. A. Harrington, T. S. Matthiesen, D. Simpson, S. K. Goh, B. L. Walker, G. Almeida-Porada, D. Wang, et al. Characterization and Functionality of Cardiac Progenitor Cells in Congenital Heart Patients Circulation, February 1, 2011; 123(4): 364 - 373. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Bel, V. Planat-Bernard, A. Saito, L. Bonnevie, V. Bellamy, L. Sabbah, L. Bellabas, B. Brinon, V. Vanneaux, P. Pradeau, et al. Composite Cell Sheets: A Further Step Toward Safe and Effective Myocardial Regeneration by Cardiac Progenitors Derived From Embryonic Stem Cells Circulation, September 14, 2010; 122(11_suppl_1): S118 - S123. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. P. Opolski, M. Niewada, and A. Witkowski Letter by Opolski et al Regarding Article, "Patient Characteristics and Cell Source Determine the Number of Isolated Human Cardiac Progenitor Cells" Circulation, August 10, 2010; 122(6): e422 - e422. [Full Text] [PDF] |
||||
![]() |
M. Omatsu-Kanbe, T. Yamamoto, Y. Mori, and H. Matsuura Self-beating Atypically Shaped Cardiomyocytes Survive a Long-term Postnatal Development While Preserving the Expression of Fetal Cardiac Genes in Mice Journal of Histochemistry & Cytochemistry, June 1, 2010; 58(6): 543 - 551. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. M. Zaruba, M. Soonpaa, S. Reuter, and L. J. Field Cardiomyogenic Potential of C-Kit+-Expressing Cells Derived From Neonatal and Adult Mouse Hearts Circulation, May 11, 2010; 121(18): 1992 - 2000. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. L. Mummery, R. P. Davis, and J. E. Krieger Challenges in Using Stem Cells for Cardiac Repair Science Translational Medicine, April 14, 2010; 2(27): 27ps17 - 27ps17. [Full Text] [PDF] |
||||
![]() |
Y. Zhou, P. Pan, L. Yao, M. Su, P. He, N. Niu, M. A. McNutt, and J. Gu CD117-positive Cells of the Heart: Progenitor Cells or Mast Cells? Journal of Histochemistry & Cytochemistry, April 1, 2010; 58(4): 309 - 316. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Itzhaki-Alfia, J. Leor, E. Raanani, L. Sternik, D. Spiegelstein, S. Netser, R. Holbova, M. Pevsner-Fischer, J. Lavee, and I. M. Barbash Patient Characteristics and Cell Source Determine the Number of Isolated Human Cardiac Progenitor Cells Circulation, December 22, 2009; 120(25): 2559 - 2566. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Ferreira-Martins, C. Rondon-Clavo, D. Tugal, J. A. Korn, R. Rizzi, M. E. Padin-Iruegas, S. Ottolenghi, A. De Angelis, K. Urbanek, N. Ide-Iwata, et al. Spontaneous Calcium Oscillations Regulate Human Cardiac Progenitor Cell Growth Circ. Res., October 9, 2009; 105(8): 764 - 774. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. Stamm, Y.-H. Choi, B. Nasseri, and R. Hetzer A heart full of stem cells: the spectrum of myocardial progenitor cells in the postnatal heart Therapeutic Advances in Cardiovascular Disease, June 1, 2009; 3(3): 215 - 229. [Abstract] [PDF] |
||||
![]() |
P. Menasche Skeletal Myoblasts for Cardiac Repair: Act II? J. Am. Coll. Cardiol., December 2, 2008; 52(23): 1881 - 1883. [Full Text] [PDF] |
||||
![]() |
H. Reinecke, E. Minami, W.-Z. Zhu, and M. A. Laflamme Cardiogenic Differentiation and Transdifferentiation of Progenitor Cells Circ. Res., November 7, 2008; 103(10): 1058 - 1071. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. Koninckx, K. Hensen, J.-L. Rummens, and M. Hendrikx Cardiac stem cells in the real world J. Thorac. Cardiovasc. Surg., September 1, 2008; 136(3): 797 - 798. [Full Text] [PDF] |
||||
![]() |
P. Menasche Towards the second generation of skeletal myoblasts? Cardiovasc Res, August 1, 2008; 79(3): 355 - 356. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |