| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Department of Cardiothoracic Surgery, Pediatric Division, Stanford University School of Medicine, Stanford, California
b Department of Anesthesia, Stanford University School of Medicine, Stanford, California
Accepted for publication June 16, 2008.
* Address correspondence to Dr Riemer, Department of Cardiothoracic Surgery, Stanford University School of Medicine, CV116C, Stanford, CA 94305-5407 (Email: riemerk{at}stanford.edu).
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and MethodsResultsCommentFootnotesAcknowledgmentsReferences |
|---|
Methods: Immunologic markers of cell lineage (stage-specific embryonic antigen 4 [SSEA-4], islet cell antigen 1 [Isl1], c-kit, Nkx2.5, sarcoplasmic reticulum calcium-regulated ATPase type 2 [SERCA2]) and proliferation (Ki67) were localized in right ventricular biopsies from 32 CHD patients aged 2 to 93 days.
Results: Neonatal myocardium contains progenitor cells and transitional cells expressing progenitor and differentiated myocyte marker proteins. Some cells expressed the pluripotent cell marker c-kit and also coexpressed the myocyte marker SERCA2. Multipotent progenitor cells, identified by the expression of Isl1, were found. Ki67 was expressed in some myocytes and in nonmyocyte cells. A few cells expressing SSEA-4 and Isl1 were observed during the early postnatal period. Cells expressing c-kit, the premyocyte marker Nkx2.5, and Ki67 were found throughout the first postnatal month. A progressive decline in cell density during the first postnatal month was observed for c-kit+ cells (p = 0.0013) and Nkx2.5+ cells (p = 0.0001). The percentage of cells expressing Ki67 declined during the first 3 postnatal months (p = 0.0030).
Conclusions: Cells in an incomplete state of cardiomyocyte differentiation continue to reside in the infant heart. However, the relative density of progenitor cells declines during the first postnatal month.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentFootnotesAcknowledgmentsReferences |
|---|
Several recent studies have reported both success and failure in attempts to use various types of stem cells therapeutically for amelioration of myocardial infarction and cardiomyopathies [3, 5, 6]. The still incomplete nature of our basic understanding of stem cell biology is a major barrier to the successful harnessing of the promising power of stem cells. The discovery of resident CPCs in the myocardium provides the opportunity to approach therapeutics by using cells that are already predirected toward this organ and perhaps more easily coaxed into a desired supportive role.
The focus of our research is to define the population of proliferating cells in neonatal myocardium, including pluripotent cells capable of regenerating all cardiac cell types. Recent reports suggest that a hierarch progression of stem cells from primitive pluripotent, to multipotent, to lineage-designated (ie, CPCs), and finally to fully differentiated (eg, cardiomyocyte) exists in myocardium and other tissues [7].
We believe that resident cells such as CPCs, and perhaps their precursors, may potentially have a role as a therapeutic adjunct to cardiac operations. Our specific interest is the potential of CPCs to augment postsurgical recovery of pediatric patients with congenital heart diseases (CHD). Children with complex CHD frequently require surgical intervention early in life and often, several reoperations to replace tissue or materials that they have either outgrown or that have deteriorated [8]. The controlled manipulation of CPCs offers the potential to provide surgical therapies using biomaterials that grow with the patient, and because they are self-renewing tissues, potentially do not deteriorate; hence, the possibility of engineering customized cardiac tissues from CPCs is particularly attractive.
This study sought to determine whether CPCs, recently demonstrated in adult human myocardium, [9, 10] were present in the neonatal heart and to determine the changes in this cell population that occur with age. We described changes in their distribution through the first month of life, identified putative resident CPCs in the neonatal myocardium, and demonstrated cells exhibiting phenotypes transitional between progenitor and cardiac muscle cells. We also evaluated changes in the proliferative capacity of cardiac cells in the immediate postnatal period, when many children with CHD undergoing surgery are critically dependent on rapid functional recovery of myocardium to avoid early and late morbidity. Although many studies of rodent myocardium have documented a decline in the population of mitotic cardiocytes [11–13], few studies have focused on human myocardium, the subject of the present study.
|
Material and Methods |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentFootnotesAcknowledgmentsReferences |
|---|
|
Neonatal tissue specimens were transported from the operating suite to the laboratory in cold saline, immersed in ice. Portions were immediately immersion-fixed in 10% neutral-buffered formalin (Fisher Scientific, Pittsburgh, PA) and stored at 4°C until they were ready for immunohistochemistry analysis, at which time they were embedded in paraffin.
Immunohistochemistry
After dehydration, heat-induced antigen retrieval was performed by boiling the sections in citrate buffer (pH 6.0; BD Biosciences, San Diego, CA). To block endogenous peroxidase activity, sections were incubated in an aqueous solution of 3% hydrogen peroxide for 10 minutes at room temperature and then washed with phosphate-buffered saline.
Sections were subsequently incubated with primary antibodies. Cell proliferation was assessed on the basis of positive nuclear staining of Ki67 antigen [14] (BD Biosciences). Antisera to specific cellular antigens (Table 2) [15–24] were used to assess cell lineage or other stated aspects of specificity. Embryonic stem cells were identified by stage-specific embryonic antigen 4 (SSEA-4; Chemicon International Millipore, Danvers, MA; Abcam Inc, Cambridge, MA) and Oct-3/4 (Santa Cruz Biotechnology) antigen staining [15]. We tested SSEA-4 antibody samples from two commercial sources, and both had the same localization. However, only a single clone of this antibody hybridoma exists [16], so both antibody sources were from the same original culture.
|
Cardiac cell lineage was assessed using antibodies to Nkx2.5 (R & D systems, Minneapolis, MN), Mef-2 (muscle-specific DNA binding protein, Santa Cruz Biotechnology), and sarcoplasmic reticulum calcium regulated ATPase type 2 [20] (SERCA2; Sigma, St. Louis, MO). The homeodomain transcription factor Nkx2.5 is a well-characterized marker of early cardiocyte lineage that, together with the zinc finger transcription factor GATA4, is expressed throughout the precardiac mesoderm of the primary heart field [21, 22]. Early myocardial lineage differentiation was indicated by Nkx2.5 expression, and advanced differentiation was indicated by troponin-T or SERCA2 expression. Troponin-T and atrial natriuretic peptide [23] were also used to localize myocardial and right atrial tissue, respectively.
Primary antisera were either used alone or, for colocalization, sequentially with an intervening blocking step between the first and second antiserum to prevent nonspecific cross-labeling of antigens. Detection of staining was achieved using the Vectastain Elite system (Vector Labs, Burlingame, CA), biotinylated secondary antisera to the host of the primary antisera, and either avidin-conjugated peroxidase and diaminobenzidine (DAB) substrate, which forms a brown precipitate, or avidin-conjugated fluorochromes (green, fluorescein DCS; and red, Texas Red, both from Vector Labs). Nuclei were stained using either hematoxylin (light microscopy), or fluorescent DNA Hoechst 33342 (blue; Molecular Probes, Eugene, OR). Fluorescent images were captured at x63 original magnification using a DM 6000 microscope (Leica Microsystems, Bannockburn, IL) equipped with a Nuance Multispectral Imaging System (CRI Inc, Cambridge, MA). Confocal images were acquired using a LSM 510 microscope (Carl Zeiss, Oberkochen, Germany), and Volocity three-dimensional (3D) imaging software (Improvision, Lexington, MA) was used for reconstruction and 3D rendering of image stacks for colocalization of dual-labeled specimens.
Statistical Analysis
Quantitation of stained cells in postnatal specimens (identification blinded to the observer) was by counting at least 10 randomly chosen high-power fields on each slide, computing the mean number of positive cells, and expressing it as a percentage of the total cells in the field. Each data point in the plotted arrays is the result from a single specimen at the age indicated and shows the average number of positively stained cardiomyocytes expressed as a percentage of total cells per high-power field. The mean cell counts per field were compared across postnatal age using linear regression and appropriate post hoc analysis of variance (ANOVA) of the slope assisted by the JMP statistical software (SAS Inc, Carey, NC). A value of p = 0.05 (defined as the t statistic predicting a change in expression with age) was the minimum value considered for statistical significance of comparisons. Owing to the very limited size of most biopsy specimens, not all specimens were available for all quantitative assays, with "N" indicating the number of specimens analyzed. The magnifications indicated for all images are the objective multiplied by the eyepiece power.
|
Results |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentFootnotesAcknowledgmentsReferences |
|---|
|
Primitive and CPCs Resident in Human Heart Tissue
To determine whether primitive, pluripotent stem cells were present in the human neonatal heart, we examined the expression of two embryonic stem cell markers, SSEA-4 and Oct-3/4. Cells expressing SSEA-4 were scattered throughout the atrial and ventricular myocardium of the fetal heart. The antigen was localized to the cell membrane and in the cytoplasm. A rounded morphology and large nucleus-to-cytoplasm ratio in the SSEA-4+ cells were consistent with stem cell morphology. Some areas near the right atrium showed a greater density of SSEA-4+ cells (Fig 2a). SSEA-4+ cells were occasionally found expressed in neonatal myocardium. Five of 32 specimens contained SSEA-4+ cells (Fig 2b and c). As with fetal myocardium, SSEA-4+ cells in neonates were smaller, more rounded in shape, and generally had limited cytoplasmic area compared with cardiomyocytes. Although SSEA-4 is recognized as an integral membrane protein and surface antigen, we observed staining throughout the cell rather than being restricted to the cell surface, as is frequently reported for isolated human embryonic stem cells.
|
We next examined the expression of Isll, a transcription factor previously reported to identify cardiac-lineage progenitor cells within myocardial specimens. Cells expressing Isll were identified in both fetal and newborn myocardium (Fig 2d to j). In fetal heart specimens (Fig 2d to h), cells expressing Isll were found clustered in the right atrial wall (Fig 2f and g), which was localized by staining for myocardium (troponin-T, Fig 2d) and right atrial myocardial cells (atrial natriuretic peptide, Fig 2e). We were unable to find similar clusters of Isll+ cells in ventricular regions of the fetal heart, which did contain occasional Isll+ cells (data not shown). In neonatal heart biopsy specimens (ie, RVOT region), very few cells expressing Isll were found, often limited to 1 to 2 cells per specimen (Fig 2i and j). We were unable to find any Isll+ cells in specimens from infants aged more than 1 postnatal week. Isll+ cells comprised a subpopulation of cells smaller than myocytes, essentially limited in size to the stained nucleus with a thin layer of cytoplasm, and symmetrically rounded.
Cells expressing the CPC and hematopoietic marker c-kit were identified within human fetal and postnatal myocardial specimens. Cells expressing c-kit were of similar size but more symmetrically rounded than myocytes, and the staining exhibited the membrane localization appropriate for an integral membrane receptor (Fig 3). Cells positive for c-kit were located within connective tissue adjacent to muscle bundles and occasionally within muscle bundles. The positive cells were fewer in number in postnatal specimens compared with fetal hearts (Fig3a to c). The percentage of cells expressing c-kit declined over the first postnatal month, by the end of which their density had fallen fourfold (Fig 3d, p = 0.0013). As with Ki67 expression, age rather than type of congenital lesion appeared to be the determinant of the number of cells expressing c-kit. Dual labeling studies showed evidence that some cells expressing c-kit also expressed Ki67 (Fig 3e and f) and SERCA2 (Fig 3g). Intriguingly, such transitional cells expressing both markers were generally found adjacent to the peripheral margins of muscle fiber bundles.
|
|
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentFootnotesAcknowledgmentsReferences |
|---|
Moreover, our studies demonstrate that nonfibroblast cells, that are likely capable of differentiating into myocytes (ie, CPCs), persist in neonatal myocardium. Colocalization studies revealed that some cells expressing a differentiated myocyte phenotype marker (SERCA2) coexpressed markers of the CPC phenotype or a proliferation marker. Although such transitional cells were rare, their presence indicates a high degree of phenotypic plasticity among cardiac myocytes in the neonatal heart. The proliferative state of some of the CPCs was demonstrated by colocalization of c-kit with Ki67 in neonatal hearts. However, whether the c-kit+ cells were entering or exiting mitosis cannot be determined.
The proliferation data in this study are consistent with previous reports that most human myocardial muscle cells become terminally differentiated with ageing. Zak [13] compared mitotic activity within the myocardium of humans and several animal species and concluded that myocyte division falls rapidly after birth and that cellular hypertrophy was the primary basis of cardiac enlargement in adults. Consistent with these studies, we found that the mitotic process reaches a very low level within 1 month of birth in human neonatal myocardium. However, as noted by the studies by Zak, reported nearly 3 decades ago, DNA synthesis in myocardial cells can be reactivated under "rare conditions," and cardiac myoblasts are unusual in their ability to simultaneously divide and synthesize proteins associated with differentiated function [13]. Our observations that some Ki67+ myocytes express SERCA2 support the general concept of continued myocardial cell expansion on a limited basis as a fundamental mechanism of regenerative repair.
The observed coexpression of stem cell (c-kit) and differentiated myocyte (SERCA2) marker proteins by some myocytes is noteworthy. Hence in addition to probable mitosis of "differentiated" cardiomyocytes as noted, the coexpression data demonstrate that resident stem cells clearly transition to the cardiomyocyte fate. Considering myocardial lineage alone, we found that distinct populations of cells in an early (Nkx2.5+) and late (SERCA2) state of myocardiocyte differentiation reside in the neonatal heart. The population of the incompletely differentiated Nkx2.5+ cells appears to decline to a very low level within the first month of postnatal life, consistent with an expected maturation of the myocyte population. Although speculative, it is conceivable that such cells could more easily reenter a proliferative state to assist in the healing process in a manner that would facilitate rapid recovery.
We identified three populations of CPCs in the myocardium: pluripotent SSEA-4+ cells, a population of CPCs indicated by c-kit expression, and also the presence of another, possibly distinct, population of CPCs in the very early neonatal heart as evidenced by positive staining for Isl1. Although we found them in the neonatal RVOT, no Isl1+ cells were detected in neonatal hearts beyond the first postnatal week.
The presence of precursor cardiomyocytes in adult mouse and human heart has been previously reported [9, 24, 27]. Since we initiated our study, we are aware of one report [27] of the presence of myocardial precursor cell populations in human pediatric heart specimens. In that study [27], Isl1+ cells were found in neonatal human atrial tissue, an observation that we have also made (data not shown). We also examined human fetal heart, primarily as a positive control for the neonate. As anticipated, we found that fetal myocardium contains a higher density of CPCs than postnatal samples. This confirms that fetal heart tissue would therefore be a rich source of CPCs for therapeutics as well as for further investigation of CPC biology.
Homeostatic expansion of incompletely differentiated precursor cell populations may normally facilitate recovery from operation or injury as a significant component of the healing process. Compelling examples of such expansion in human heart are the observations that myocardial cells expressing the Y-chromosome have been found to appear in female heart grafts of male recipients [28], and myocyte hyperplasia in the presence of aortic stenosis [24]. Therapies that promote the in vivo expansion of precursor cell populations may prove to be a useful means to accelerate recovery from cardiac operations, to treat failing myocardium, or possibly to reduce scar tissue formation in the heart postoperatively. Their potential contribution may be even greater in the early newborn period, before these putative precursor populations decline. Ex vivo expansion of CPCs resident in neonatal myocardium may be useful in the future treatment of pediatric patients undergoing corrective procedures for their CHD. Many of these patients require multiple procedures to replace homograft tissues, which deteriorate and calcify over time [8, 29]. Tissue engineering of allografts that have growth potential may be beneficial for these patients.
In this study we assessed specimens of the RVOT and did not extensively survey other neonatal heart regions because the tissue was not available. We cannot yet determine how representative the RVOT region of myocardium is of the heart in general. Indeed, others have suggested that atrial tissue may be a particularly rich site of precursor cell populations [27], and our Isl1 localization data from fetal heart is consistent with this observation. In an ongoing study we are assessing other myocardial regions from neonatal heart specimens as they become available.
A second limitation is the RVOT tissue was primarily from CHD patients and thus cannot be considered entirely normal. The only expectedly normal myocardial tissue specimens we had access to for analysis were fetal hearts from elective terminations conducted for noncardiac indications, although the reason for termination was unknown for one specimen with grossly normal heart morphology.
We identified SSEA-4+ cells in the human myocardium, which has also been reported in rat heart. [7]. Although this antigen is used as marker of pluripotent stem cells, we are aware of only a single monoclonal antibody to this antigen. We observed apparent cytoplasmic as well as membranous localization of SSEA-4, whereas unfixed embryonic stem cells usually display membranous localization of this antigen. The different localization in fixed tissue may simply reflect alteration of the antigen by fixation and paraffin embedding, but the staining was selective and immuno-specific, and we did not observe staining for SSEA-1, a closely related antigen, in this tissue (data not shown).
The high Ki67 expression we occasionally observed in older infants may be an indication of dynamic remodeling driven by presence of congenital disease. Although this suggests a possible departure from normality within the specimen, it also presents an intriguing opportunity to investigate how ongoing disease may activate endogenous pathways for awakening the expansion and differentiation of CPCs to mediate the self-repair process.
The presence of binucleated myocardiocytes assessed by electron microscopic morphometry is often used as an indication of terminal differentiation of myocardiocytes and other cells. We chose instead to use the expression of cell cycle–associated genes to assess the mitotic status of the cells because of its expectedly greater reliability based on higher sampling efficiency. Although we did not systematically examine the number of binucleated cells in these specimens, their presence was observed more frequently in specimens from older (>1 week) neonates.
Our studies demonstrate that cells in an incomplete state of cardiomyocyte differentiation continue to reside in the heart after birth, but their relative density declines steeply during the first month of life. Is it possible to control the expansion of this pluripotent cell population in vivo to facilitate recovery from morbidities or enhance surgical repair at any age? Many investigators have approached this question through approaches using ex vivo cell expansion and forced differentiation in cell-based therapies, which, unfortunately, has provided largely disappointing results. Recognition of the persistence of CPCs in the myocardium provides the opportunity to discover the basis for their normal control so that directed expansion and differentiation may be approached for therapy.
|
Acknowledgments |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentFootnotesAcknowledgmentsReferences |
|---|
|
Footnotes |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentFootnotesAcknowledgmentsReferences |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentFootnotesAcknowledgmentsReferences |
|---|
Related Article
This article has been cited by other articles:
|
|
 |
|
  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]
|
|
|
|
 |
|
 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. A. Thistlethwaite Invited Commentary Ann. Thorac. Surg., October 1, 2008; 86(4): 1320 - 1320. [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 |
