Scalable Production of Embryonic Stem Cell-Derived ... - CiteSeerX

Download PDF
3 downloads 0 Views 3MB Size Report
Comment
This study describes a novel approach to generate scal- able quantities of ..... In vitro chamber specification during embryonic stem cell cardiogenesis.
TISSUE ENGINEERING Volume 9, Number 4, 2003 © Mary Ann Liebert, Inc.

Scalable Production of Embryonic Stem Cell-Derived Cardiomyocytes P.W. ZANDSTRA, Ph.D.,1,2 C. BAUWENS, B.Eng.,1,2 T. YIN, M.Sc.,1 Q. LIU, Ph.D.,1 H. SCHILLER, Ph.D.,3 R. ZWEIGERDT, Ph.D.,3 K.B.S. PASUMARTHI, Ph.D.,4 and L.J. FIELD, Ph.D.4

ABSTRACT Cardiomyocyte transplantation could offer a new approach to replace scarred, nonfunctional myocardium in a diseased heart. Clinical application of this approach would require the ability to generate large numbers of donor cells. The purpose of this study was to develop a scalable, robust, and reproducible process to derive purified cardiomyocytes from genetically engineered embryonic stem (ES) cells. ES cells transfected with a fusion gene consisting of the a-cardiac myosin heavy chain (MHC) promoter driving the aminoglycoside phosphotransferase (neomycin resistance) gene were used for cardiomyocyte enrichment. The transfected cells were aggregated into embyroid bodies (EBs), inoculated into stirred suspension cultures, and differentiated for 9 days before selection of cardiomyocytes by the addition of G418 with or without retinoic acid (RA). Throughout the culture period, EB and viable cell numbers were measured. In addition, flow cytometric analysis was performed to monitor sarcomeric myosin (a marker for cardiomyocytes) and Oct-4 (a marker for undifferentiated ES cells) expression. Enrichment of cardiomyocytes was achieved in cultures treated with either G418 and retinoic acid (RA) or with G418 alone. Eighteen days after differentiation, G418-selected flasks treated with RA contained approximately twice as many cells as the nontreated flasks, as well as undetectable levels of Oct-4 expression, suggesting that RA may promote cardiac differentiation and/or survival. Immunohistological and electron microscopic analysis showed that the harvested cardiomyocytes displayed many features characteristic of native cardiomyocytes. Our results demonstrate the feasibility of large-scale production of viable, ES cell-derived cardiomyocytes for tissue engineering and/or implantation, an approach that should be transferable to other ES cell derived lineages, as well as to adult stem cells with in vitro cardiomyogenic activity. INTRODUCTION

G

during fetal development results from the proliferation of differentiated, contractile cardiomyocytes. During neonatal life the proliferative capacity of cardiomyocytes is dramatically reduced and subsequent increases in cardiac mass arise ROWTH OF THE HEART

largely from cardiomyocyte hypertrophy. Although there is some debate regarding the absolute rate at which adult cardiomyocytes are able to enter the cell cycle,1,2 it is generally accepted that this occurs only infrequently.3 Many forms of cardiovascular disease are accompanied by acute or chronic cardiomyocyte loss. Given their limited intrinsic regenerative capacity, cardiomyocyte death

1 Institute

of Biomaterials and Biomedical Engineering, University of Toronto, Toronto, Ontario, Canada. of Chemical Engineering and Applied Chemistry, University of Toronto, Toronto, Ontario, Canada. 3 Cardion AG, Erkrath, Germany. 4 Krannert Institute of Cardiology, Indiana University School of Medicine, Indianapolis, Indiana. 2 Department

767

ZANDSTRA ET AL.

768

in the adult heart leads to formation of scar tissue with a concomitant stiffening of the myocardium. This process triggers a progressive sequence of cellular and physiological processes that can ultimately lead to left ventricular dilatation and heart failure.4 Despite advances in the treatment of acute myocardial infarction, the ability to repair extensive myocardial damage and to treat heart failure is limited.5 Cellular transplantation has emerged as a potential strategy to replace cardiomyocytes that are lost as a consequence of disease progression.6 To date, a number of different cell types have been tested in animal models, including fetal cardiomyocytes,7–16 embryonic stem cellderived cardiomyocytes,17 stem cells with apparent cardiomyogenic potential,18,19 and skeletal myoblasts.20–24 The observation that cell transplantation can have a positive effect on the pathophysiologic cardiac remodeling in response to myocardial infarction has prompted several phase I clinical trials using autologous skeletal myoblasts as donor cells. However, a priori, cardiomyocytes would appear to be the logical choice as a donor cell type if the ultimate goal of the intervention is to restore lost contractile activity (as opposed to alter the remodeling process). Indeed, a study from the Kloner laboratory supports this notion.25 In light of this, clinical application of cardiomyocyte cell transplantation will require the ability for large-scale production of suitable donor cells. Embryonic stem (ES) cells are pluripotent cells derived from the inner cell mass of blastocysts. ES cells have the ability to differentiate into a variety of cell lineages in vitro.26 Cardiomyogenic induction during ES cell differentiation is manifested by the appearance of spontaneously and rhythmically contracting myocytes.27 ES-derived cardiomyocytes express many of the landmarks of adult cardiomyocytes including a- and b-cardiac myosin heavy chain (MHC),28 a-tropomyosin,29 myosin light chain 2v (MLC-2v),30 atrial natriuretic factor,30 phospholamban,31 and type B natriuretic factor,32 as well as exhibit normal contractile sensitivity to calcium.33 Electrophysiological analyses have identified action potentials typical of atrial, ventricular, and conduction system cardiomyocytes from ES-isolated cell preparations.34,35 Furthermore, cell cycle withdrawal and multinucleation in ES-derived cardiomyocytes follow a temporal program analogous to that observed during cardiogenesis in vivo.36 These attributes suggested that differentiating ES cells might constitute a renewable source of donor cardiomyocytes suitable for cellular transplantation into the heart. Accordingly, we undertook a series of studies to establish an approach for producing and harvesting enriched cardiomyocytes from differentiating mouse ES cells in a scalable culture system, with the notion that these processes could be utilized for the generation of therapeutic donor cells. Our approach utilized ES cells transfected with a vector containing two transcriptional

units (designated MHC-neor/pGK-hygror). The first unit consists of the phosphoglycerate kinase promoter and a cDNA encoding resistance to hygromycin (used to select stably transfected ES cells before differentiation).37 The second unit comprises the cardiomyocyte-restricted a-cardiac myosin heavy chain promoter driving the neomycin resistance gene (used for cardiomyocyte enrichment).17,37 The expression of the fusion gene in ESderived cardiomyocytes facilitates their selection with G418 during in vitro differentiation. In this study, kinetic changes of the number in embryoid bodies (EBs), total cells, and cardiomyocytes were measured during the time course of differentiation of ES cells into cardiomyocytes in vitro. In addition, a combinational method to enrich for and dissociate the desired target cells from the resulting EBs in spinner flasks was developed. Using this culture system, as many as 1.5 3 107 dispersed and viable ES-derived cardiomyocytes can be generated in a 250-mL bioreactor. Fluorescence-activated cell sorting (FACS) analysis showed that more than 70% of the harvested cells displayed myocyte-specific immunoreactivity. Immunohistological and electron microscopic analysis showed that the harvested cardiomyocytes exhibited the expected cardiac markers and morphology. This approach should be a useful procedure to generate a large quantity of cardiomyocytes from in vitro differentiation of ES cells.

MATERIALS AND METHODS Cells R138, D326, CCE,39 and J140 mouse ES cell lines were used in the study. All ES cells were maintained on gelatin (Sigma, Oakville, CA) (0.2%)-coated tissue culture flasks (Sarstedt, Montreal, CA) in high-glucose Dulbecco’s modified Eagle’s medium (DMEM; Invitrogen, Burlington, CA) supplemented with 15% defined fetal bovine serum (HyClone, Logan, UT), 2 mM L -glutamine (Invitrogen), 0.1 mM 2-mercaptoethanol (Sigma), 0.1 mM nonessential amino acids (Invitrogen), 1 mM sodium pyruvate, penicillin (50 IU/mL), and streptomycin (50 mg/mL) (Invitrogen) in a humidified environment containing 5% CO2 at 37°C. The cells were kept in the undifferentiated state by the addition of leukemia inhibitory factor (LIF, 1000 units/mL; Chemicon, Temecula, CA) and subcultured every second day. The undifferentiated cells were transfected, via electroporation, with plasmid carrying both an a-myosin heavy chain promoter in front of the neomycin phosphotransferase gene (MHC-neor) and a phosphoglycerate kinase in front of the hygromycin resistance gene (PGK-hygror) in a common pUCBM20 vector backbone (Boehringer, Mannheim, Germany). For transfection, the cells were suspended in 800 mL of phosphate-buffered saline (PBS) at a density of 5 3 106

SCALABLE PRODUCTION OF ES CELL-DERIVED CARDIOMYOCYTES cells/mL and electroporated with 50 mg of linearized (HindIII) DNA, using a GenePulser II system (800 V, 3 mF; Bio-Rad, Hercules, CA). Electroporated cells were cultivated and selected by growth for at least 7 days in culture medium containing hygromycin B (200 mg/mL; Invitrogen) at 37°C in humidified air with 5% CO2 . The CM7/1 clone was derived from the J1 ES cell line. CM7/1 was generated by expanding single clones in hygromycin B-supplemented medium for an additional 3 weeks and selecting for cells that exhibited robust growth as undifferentiated cells, as well as the ability to reproducibly generate large numbers of cells on G418 addition. The mouse cardiomyocyte cell line HL-141 was used as a positive control in this study. These cells were cultured on gelatin/fibronectin (1 mg of fibronectin in 80 mL of 0.02% gelatin)-coated flasks in Claycomb medium (JRH Biosciences, Lenexa, KS) supplemented with 10% defined fetal bovine serum (JRH Biosciences), 2 mM L glutamine (Invitrogen), penicillin–streptomycin (100 U/mL–100 mg/mL; Invitrogen), and 0.1 mM norepinephrine (Sigma) in a 5% CO2 humid environment. HL1 cells exhibit a differentiated cardiomyocye phenotype while proliferating in vitro.

ES cell differentiation cultures To initiate differentiation, ES cells were dispersed into individual cells with trypsin and resuspended in ES cell differentiation medium (ES culture medium without LIF). The cells were allowed to aggregate in hanging drops (3000 cells/30 mL) for 2 days as described.37 The resulting embryoid bodies (EBs) were transferred to gelatin-coated 25-cm2 flasks for analysis or were seeded individually onto gelatin-coated 24-well tissue culture plates and examined daily after day 4 under a phase-contrast microscope for the presence of contracting cardiomyocytes. For the bulk stirred suspension cultures, differentiation was initiated by seeding 4 3 106 cells into 100-mm bacterial petri dishes containing 10 mL of ES differentiation medium as described above. After 4 days in suspension culture, the EBs were transferred at a density of 5 3 105 cells/mL into a 250-mL spinner flask (Bellco, Vineland, NJ) with 250 mL of fresh medium. The spinner flask was stirred at 60 rpm with a paddle-type impeller (Bellco) and a half-medium exchange was performed every 2 days. For ES-derived cardiomyocyte selection, on day 9 after the initiation of differentiation, medium supplemented with G418 (400 mg/mL; Invitrogen) was added to eliminate the cells not expressing the neomycin resistance gene (i.e., noncardiomyocytes). For cultures supplemented with retinoic acid (RA), all-trans-RA (Sigma) was used at a final concentration of 1029 M from day 9 (same day as the start of G418 treatment) after the initiation of differentiation.

769

Enzymatic dissociation of EBs growing in spinner flask Aliquots of EBs harvested from the spinner flasks, enumerated visually by phase-contrast microscopy, and resuspended in 100 mL of collagenase B (1 mg/mL; Roche Diagnostics, Laval, PQ, Canada), dispase I (2 U/mL; Roche Diagnostics), and 2% fetal bovine serum (FBS) in 13 PBS. After incubation at 37°C for 45 min, 900 mL of ES cell differentiation medium was added. EBs were dissociated into single cells by trituration. Dispersed cells were resuspended with fresh culture medium for replating on precoated 0.2% gelatin plates for immunocytochemistry. Alternatively, the cells were washed with PBS and processed for RNA isolation or FACS analyses. EBs attached on tissue culture flasks were dissociated by adding 1 mL of enzyme mixture (collagenase B, dispase I, and FBS), and incubating for 45 min at 37°C. Enzyme mixture was diluted by adding 6 mL of medium. EBs were dissociated by trituration.

Antibodies, immunocytochemistry, and flow cytometry Monoclonal anti-mouse sarcomeric myosin heavy chain antibody (MF-20) was obtained from the Developmental Studies Hybridoma Bank (DSHB, Iowa City, IA). In situ immunohistochemisry was performed on cells grown on gelatin (0.2%)/fibronectin coated cell culture plates. Plates were rinsed with cold PBS and fixed with a 1:1 mixture of methanol and acetone for 20 min on ice. After blocking with 1% bovine serum albumin (Sigma) in PBS for 1 h at room temperature, the cells were incubated with primary antibody (MF-20) for 1 h at 37°C. After three washes in PBS for 5 min each, a fluorescein isothiocyamate (FITC)- or horseradish peroxidase-conjugated goat anti-mouse IgG antibody (Sigma) was applied (as indicated) for 1 h at a dilution of 1:100 (where indicated, nuclei were counterstained with hematoxylin). Visualization was achieved with a fluorescence microscope (CK40-RFL; Olympus, Tokyo, Japan) equipped with a Coolpix 990 digital camera (Nikon, Tokyo, Japan). Undifferentiated ES cells, HL-1-positive control cells,41 and dissociated EBs were analyzed for intracellular protein expression by flow cytometry. Dispersed cells (5 3 105 –1 3 106) were washed with PBS and aliquotted into 100 mL of IntraPrep permeabilization reagent 1 (Immunotech, Marseille, France) for 15 min at room temperature. After washing once with 1 mL of Hanks’ buffered saline solution (HBSS) with 2% FBS (HF), the cells were permeabilized for 5 min with IntraPrep permeabilization reagent 2 and incubated with a 1:10 dilution of mouse anti-sarcomeric myosin heavy chain (MF-20) or mouse anti-mouse Oct-442 antibody (BD Bioscience, Missisauga, Canada) for 15 min at room temperature. After washing once with PBS, secondary

ZANDSTRA ET AL.

770

FITC-conjugated goat anti-mouse IgG antibody was added at a 1:100 dilution and the cells were incubated for 15 min at room temperature. Test samples were also tested for viability by adding 7-amino-actinomycin D (7AAD, 1 mg/mL; Molecular Probes, Eugene, OR) for 15 min at 4°C. Cells were analyzed with a flow cytometer (XL; Beckman-Coulter, Mississauga, CA), using EXpoADCXL4 software (Beckman-Coulter). Positive staining was defined as the emission of a level of fluorescence that exceeded levels obtained by .99.9% of cells from the control population stained with only the secondary antibody.

PCR and Western analysis Total RNA was isolated from test cells, using a GenElute mammalian total RNA kit (Sigma). RNA was quantified by UV spectrophotometer and 0.5 mg of RNA was used in each reverse transcriptase-polymerase chain reaction (RT-PCR) (One-Step RT-PCR kit; Qiagen, Valencia, CA). The oligonucleotide primers for Oct-4 amplification were 59-GGCGTTCTCTTTGGAAAGGTGTTC-39 and 59-CAAAGCTCCAGGTTCTCTTG-39. The oligonucleotide primers for MHC amplification were 59-CTGATGGCACAGAAGATGCT-39 and 59-GTTCAGGATGCGATACCTCT-39. The oligonucleotide primers for b-actin amplification were 59-AGGGGCCGGACTCATCGTACTC-39 and 59-GTGACGAGGCCCAGAGCAAGAG-39. The PCR amplification conditions were 30 cycles of 30 s at 94°C, 30 s at 52°C, and 60 sec at 72°C. The sizes of the anticipated RT-PCR amplification products were 636 bp for Oct-4 transcripts and 1058 bp for MHC transcripts. Equivalent loading was verified by amplification of b-actin, and primer specificity was verified by RT-free amplification in the second reaction (data not shown). Protein for Western analysis for Oct-4 protein expression was obtained as previously described.42 Equivalent cell numbers and protein amounts (verfied with a MicrBSA protein assay reagent kit; Pierce, Rockford, IL) were loaded onto gels to measure Oct-4 protein levels under test conditions. STO fibroblast cells cultured and loaded under identical conditions as ES cells served as negative controls.

Electron microscopy Electron microscopy was performed in the Hospital for Sick Children (Toronto, ON, Canada). Samples were prepared by sequential dehydration in 70, 80, 90, and 95% ethanol for 1 h, followed by immersion in 100% ethanol for 3 h. Samples were then critical point dried, gold sputter coated with a Polaron SC515 SEM coating system, and then examined on a Hitachi S-2000 scanning electron microscope (accelerating voltage, 15 kV).

Statistics All numerical analysis was performed with Origin 6.1 (OriginLab, Northampton, MA) graphing and data analysis software. All results, generated from at least three independent experiments (or as indicated), were analyzed using a significance level p 5 0.05. Means were considered equivalent for p values less than 0.05.

RESULTS Flow cytometric quantitation of cardiomyogenic induction Our goal was to develop a robust protocol for the generation of scalable numbers of functional cardiac myocytes. To kinetically evaluate cardiomyocyte development during ES cell differentiation, we first developed a protocol to measure myosin heavy chain (MHC) expression by flow cytometry using MF-20, a monoclonal antibody that recognizes sarcomeric myosin, which under the conditions employed screens for cardiomyocyte induction. Figure 1A shows a representative (from at least three experiments) flow cytometric histogram for MF-20 expression in the HL-1 cardiomyocyte cell line.41 As expected, the majority (89% positive relative to staining with secondary antibody alone) of HL-1 cells express MF-20. A representative (n 5 3) dot plot of MF-20 expression on D3 ES cells differentiated for 4 days (Fig. 1B, 0.2% MF-201 cells) or 9 days (Fig. 1C, 3.9% MF201 cells) shows the low level of cardiomyogenic induction achieved in differentiating ES cells under typical (default) differentiation culture conditions (i.e., EBs generated in hanging drop cultures, plated into 25-cm3 flasks after 2 days of aggregation, and examined after 4 and 9 days of differentiation). No additional enrichment of cardiomyogentic potential was detectable after day 9 under these conditions (data not shown). Despite these low levels of cardiomyogenic induction (when analyzed on a per-cell basis), nearly 100% of the EBs were spontaneously contracting by day 12 (data not shown).

Comparison of cardiomyogenic induction from different ES cell lines The potential of various ES cell lines to undergo cardiomyogenesis was evaluated by comparing both the percentage of MF-20-positive cells and the total yield of MF20-positive cells generated after differentiation. EBs generated from the same numbers of input ES cells, using the CCE, D3, J1, R1, and CM7/1 (a J1 subclone stably transfected with the pBM20-MHC-neor/pGK-hygromycin transgene) lines, were formed by the hanging drop method. The resulting EBs were transferred to 75cm2 flasks and analyzed after a total of 9 days of differ-

SCALABLE PRODUCTION OF ES CELL-DERIVED CARDIOMYOCYTES

771

FIG. 1. MF-20 expression on control cardiac myocytes (HL-1 cells) (A), and well as on ES cells differentiated as EBs for 4 (B) and 9 (C) days. (D) A representative (of more than five experiments) series of dot plots of MF-20 expression during selection in stirred suspension cultures. Selection was initiated on day 9 of differentiation.

entiation. As shown in Fig. 2A, the D3 cell line reproducibly yielded a significantly higher (4.9 6 0.6%) percentage of MF-20 positive cells than the other tested ES lines. In contrast, the R1 and CM7/1 cell lines showed the greatest 9-day total cell yield (Fig. 2B). Taken together, the greatest number of total MF-201 cells was produced from the CM7/1 cell line (p , 0.05 versus all other cell lines). This result is surprising given that the J1 parent of the CM7/1 clone exhibited the lowest intrinsic capacity for cardiogenic differentiation, and suggests that the selection of clones from the other ES cell lines (i.e., D3) may improve default cardiac cell yields further.

Generation of cardiomyocytes in stirred suspension cultures We next tested the utility of stirred suspension cultures of EBs for the generation of scalable numbers of cardiac cells. On the basis of our previous observations that ES cells added directly into stirred cultures clump together (resulting in poor growth and differentiation),43 we differentiated EBs in dishes for 4 days37 before adding them to 250-mL stirred cultures. Because the CM7/1 ES cells carry the MHC-neor transgene, G418 was added to the cultures on day 9 of differentiation in order to enrich for cardiomyocytes. Figure 1D shows a representative series

of dot plots of MF-20 expression during selection in stirred suspension cultures. In addition, it has been previously reported that the addition of RA at a concentration of 1029 M promotes the generation of embryonic stem cell-derived cardiac cells, possibly by inducing their differentiation from more primitive progenitors.27 Accordingly, the effect of RA on cardiomyogenic differentiation in stirred suspension cultures was assessed. Figure 3A and B shows the kinetics of EB numbers and cell growth in the stirred suspension culture system from day 4 (when cells are transferred from dishes to spinner flasks) to day 18 of differentiation. Little change in EB numbers was observed during the study (Fig. 3A), indicating that addition of EBs to the spinner culture after day 4 of differentiation prevented most aggregation.43 Before RA and G418 addition, there was significant cell growth from day 4 (approximately 108 total cells; 4 3 105 cells/mL) to day 9 (approximately 1.6 3 108 total cells) in the stirred suspension cultures (Fig. 3B). G418 6 RA addition on day 9 resulted in a rapid reduction of total cell numbers. By day 18, 1.9 3 107 6 4 3 106 and 9 3 106 6 3 3 106 total selected cells were produced in the RA 1 G418 and the G418-treated cultures, respectively. Figure 3C tracks the frequency of MF-20 expression in the stirred suspension cultures during the course of differentiation. The percentage of MF-201 cells rose grad-

ZANDSTRA ET AL.

772

The progressive enrichment of cardiomyocytes (and the concomitant loss of undifferentiated ES cells) observed by flow cytometry were corroborated by Western and RT-PCR analysis of MHC and Oct-4 expression, both in response to RA addition and as a function of time during the differentiation process (Fig. 4). Together, these data indicate that after selection by G418 and RA all undifferentiated ES cells have been removed from the culture system. The data in Fig. 3 can be used to calculate (total MF201 cells 5 total cell number 3 percent MF-201 cells) the changes in absolute numbers of cardiomyocytes during culture. As shown in Fig. 5, RA supplementation resulted in a significantly (p , 0.01) greater number of MF201 cells generated by day 18. Interestingly in cultures supplemented with RA, greater numbers of cells were enumerated on day 18 than on day 14; in contrast, cultures grown in the absence of RA addition had the maximum number of MF-201 cells detected on day 9 (the day that selection was initiated).

Light and electron microscopy

FIG. 2. Comparison of cardiomyocyte development and cell proliferation of various ES cell lines. (A) MF-20 expression, (B) total cell numbers, and (C) total numbers of MF-201 cells on day 9 after differentiation.

ually, up to approximately 5.5% (as expected from the control cultures; Fig. 2) before treatment with G418. After selection by G418, the percentage of MF-20-expressing cells detected by flow cytometry in both RA-treated and non-RA-treated spinner flasks rose to 71 6 8 and 65 6 4%, respectively (no significant differences were observed between RA- and non-RA-treated cultures). Flow cytometric analyses of Oct-4 expression were performed simultaneously with MF-20 analyses in order to characterize the frequency of multipotent cells remaining in the cultures as differentiation progressed. As expected, the Oct-4 expression levels decreased significantly (from an average of 75% on day 4; Fig. 3D) during the culture. Interestingly, the levels of Oct-41 cells detected by flow cytometry were greater in the non-RAtreated cultures (0.4 6 0.2% on days 14 and 18) than in the RA-treated cultures (0.2 6 0.1% on day 14 and undetectable on day 18). Importantly, analysis of intracellular protein expression by flow cytometry at these low levels, although performed relative to the isotype controls, is difficult to quantify.

Consistent with previous reports,17,37 selected cells produced in the stirred suspension cultures had the phenotypic and functional characteristics of cardiomyocytes. ES cell-derived cardiomyocytes were dissociated after 18 days of differentiation and replated onto fibronectin/collagen-coated flasks to generate clusters of spontaneously beating cells. Immunofluorescence staining revealed the presence of cardiac-specific proteins at the expected spatial organization (Fig. 6A and B). Immunohistochemical analysis of selected and replated cells revealed that a range of MF-20 expression was detectable among cells that exhibited morphological characteristics of cardiac myocytes (Fig. 6C). Ultrastructural analysis of the cardiac bodies showed mononuclear cells, with parallel arrays of myofibrillar bundles oriented in an irregular manner in some of the cells, whereas more mature sarcomeric organization was apparent in others. The formation of early and more developed Z bands, a cellular structure that characteristically appears during in vivo cardiomyocyte differentiation, could be observed in many of the ES cell-derived cardiomyocytes (Fig. 6D). The differentiated cardiomyocytes maintained these patterns and continued to contract vigorously for at least 6 weeks.

DISCUSSION This study describes a novel approach to generate scalable quantities of cardiomyocytes in vitro by the differentiation of ES cells in stirred suspension cultures. This approach overcomes many of the limitations of the widely utilized static culture systems (e.g., lack of mixing resulting in gradients of physicochemical parameters

SCALABLE PRODUCTION OF ES CELL-DERIVED CARDIOMYOCYTES

773

FIG. 3. Kinetic changes in the total number of EBs (A), total cells (B), MF-201 cells (C), and Oct-41 cells (D) during ES cell differentiation in stirred suspension cultures in the presence (solid symbols) and absence (open symbols) of RA. G418 was added to the cultures on day 9 to initiate cardiac myocyte selection (see arrows). Results are from three independent experiments. Significant differences (p , 0.05) are noted by an asterisk.

important in ES cell differentiation,44,45 difficulties in medium supplementation/control, etc.) to enable the production of scalable numbers of target cells. The scalable production of relevant numbers of cardiomyocytes should enable rigorously designed studies to test the in vivo function of ES-derived cardiac cells. By combinatorial use of stirred cultures and the MHC neor /G418 selection system, relatively pure large-scale populations of cardiomyocytes were readily generated without the need for extensive growth factor/cytokine treatment. The resulting cardiomyocytes displayed vigorous spontaneous beating activity and showed the expected cardiomyocyte morphology. Even without any optimization for parameters such as ES cell line, cell density, medium supplementation, and so on, this combination of technologies was able to produce as many as 1.5 3 107 cardiomyocytes in 250mL stirred cultures. Flow cytometry with MF-20 (an anti-sarcomeric myosin heavy chain antibody) was used as the primary readout to quantitatively measure cardiomyogenic differentiation. Although MF-20 reacts with both cardiac and skeletal muscle, previous studies have shown that skeletal myocytes are not present at the stages of ES dif-

ferentiation examined here.17,46,47 This view is further supported by the results from many studies on contractile protein expression in stem cell-derived EBs30,47 and by the observation that that expression of the skeletal lineage determination factors MyoD and myogenin were not detected by RT-PCR before day 16 of EB development.46 On the basis of these facts, it was assumed that all of the MF-201 cells scored in our study were cardiomyocytes. The HL-1 cell line41 was used as a positive control for MF-20 staining. This is a mouse atrial cardiomyocyte cell line derived from transgenic mice expressing the simian virus 40 (SV40) large T oncoprotein in the heart. Flow cytomentric analysis of these cells revealed that at least 89% of the cells were MF-20 positive when cells were allowed to reach confluence. The fact that neither the cardiomyocyte cell line nor the selected ES cell-derived cardiomyocytes exhibited 100% MF-20 expression may be due to a variety of factors including the developmental stage of the cells, the proliferative status of the cells being analyzed, difficulties in measuring cytosolic proteins by flow cytometry, or the FACS gating parameters (i.e, cells were considered MF-20 positive only if they were brighter than 99.5% of the isotype control population).

ZANDSTRA ET AL.

774

FIG. 4. Molecular analysis of MHC gene expression and Oct-4 protein expression during stirred suspension culture. (A) Oct4 protein expression pre- and postselection in the presence (1) or absence (2) of RA. (B) MHC and Oct-4 gene expression (with b-actin control lanes) at input, immediately preselection (day 8) and postselection (day 18).

Supporting this last possibility, photomicrographs of cells stained for MF-20 by immunohistochemisty suggested that the vast majority (all cells analyzed) of the bioreactor-generated cells were positive for MF-20 expression, albeit at different levels of expression (Fig. 6D). Although these issues make determination of the absolute number of cardiomyocytes generated problematic, it is clear that MF-20 FACS is useful to track cardiomyocyte enrichment during culture, as well as in determining the

FIG. 5.

effects of culture manipulations of ES cell-derived cardiomyogenesis (unpublished data). A more in-depth functional and phenotypic analysis, including markers against specific types of cardiac cells,45 is now required in order to characterize the effect of these culture manipulations on the generation of cardiac cells for tissueengineering applications. The data in Fig. 3 were used to calculate (using MF20 expression and total cell number) the number of

Calculated (using the values in Fig. 3) total numbers of MF-201 cells generated in the stirred suspension cultures.

SCALABLE PRODUCTION OF ES CELL-DERIVED CARDIOMYOCYTES cardiomyocytes in culture on days 9, 14, and 18. It is of interest to note that in the RA-treated cultures, cardiomyocyte yield rose from ,8 3 106 cells to ,1.4 3 107 cells; in contrast, there were no significant changes in the numbers of cardiomyocytes detected during this time period in the non-RA-treated cultures (,8 3 106 cells). These observations are consistent with tritiated thymidine incorporation studies of cultured ES-derived cardiomyocytes.36 Although relatively high rates of cardiomyocyte DNA synthesis were detected after 11 days of differentiation, levels were markedly reduced by 21 days of differentiation. A similar pattern of cardiomyocyte proliferation can be inferred from studies with human ES cells.48 In these latter studies, human ES-derived cardiomyocytes plated on multielectrode arrays displayed extremely stable two-dimensional action potential propagation maps. The experiments described here also confirm and extend earlier studies showing a positive effect of RA on cardiac development in vitro.27,49,50 The addition of RA to our cultures induced the differentiation of any remaining detectable Oct-4-positive ES cells and increased the total numbers of cardiomyocytes. It is not clear from our data whether the increases in MF-201 cells detected in the RA-treated cultures were due to increased cardiomyocyte generation from upstream progenitors, or were due

775

to the proliferation of MF-201 cells detected on day 9. Previous studies on the proliferation of ES cell-derived cardiomyocytes,36 and the effect of RA on Oct-41 cells, suggest that RA may act to induce the differentiation of ES cells and/or their progeny. Definitive elaboration of the mechanism by which RA induces the production of greater amounts of cadiomyocytes, including an analysis of whether the effect of RA is specific to the cardiomyocyte lineage or whether RA acts on multiple cell lineages, will depend on our ability to quantitatively measure both the fraction of MF-20-positive cells and the proliferative status of these cells at different stages in culture with and without selection and RA supplementation; these studies are underway. The rationale of cellular cardiomyocyte transplantation as a potential therapy for heart disease has been extensively demonstrated.7–17 Although several types of cells have been transplanted into the hearts of experimental animals (reviewed in Reinlib and Field6 ), cardiac myocytes are theoretically the best candidate donor cell to restore lost cardiac function because of their intrinsic electrophysiological, structural, and contractile properties, which collectively allows them to functionally integrate with the host myocardium7,8. There are nonetheless several potential drawbacks limiting the use of cardiomy-

FIG. 6. Morphologic and structural analysis of ES cell-derived cardiomyocytes. Immunofluorescence staining of MF-20 expression of (A) dissociated cardiomyocytes generated from day 18 RA 1 G418-treated suspension cultures replated on gelatin and fibronectin-coated plates, or (B) non-dissociated “cardiac bodies” from the same culture similarly plated. (C) Similarly replated nondissociated “cardiac bodies” stained with primary MF-20 antibody and detected with a horseradish peroxidase-labeled secondary antibody. (D) An electron micrograph image of cells in “cardiac body” showing mature sarcomeric organization and Z-banding (arrow) (field width, 20 mm).

ZANDSTRA ET AL.

776

ocytes for cellular transplantation, relating mainly to the problems associated with cell sourcing. At present, the best-established source of potential donor human cardiomyocytes for therapeutic intracardiac transplantation is from differentiating ES or EG cells.48,51,52 It is also possible that in the future adult-derived stem cells with cardiomyogenic activity might be exploitable for therapeutic transplantation18,19,53–59; however, in most instances the fidelity of cardiomyogenesis in adult stem cells has not been rigorously established, or cardiomyogenic activity in the human counterpart has not yet been validated. Regardless of the stem cell source, the approach described here provides a useful tool to generate enriched and relative pure cardiomyocytes for cell transplantation. In addition to enabling the potential for therapeutic cell transplantation, the availability of large quantities of cardiomyocytes will give rise to the opportunity to discover and identify new growth factors, to evaluate drugs, and to perform toxicological in vitro studies. In summary, in this study, a procedure was designed for the scalable generation of cardiomyocytes differentiated from murine ES cells. The generated myocytes were shown to display functional (contractile) and structural properties consistent with early-stage cardiomyocytes. The establishment of this cultivation system may have a useful impact on the understanding of mammalian cardiac development and function, and provide a powerful research and clinical tool in several fields such as pharmacological and toxicological testing, functional genomics, early cardiomyogenesis, cell transplantation, and tissue engineering.

4. 5.

6.

7.

8.

9.

10.

11.

12.

13.

ACKNOWLEDGMENTS Funding for this work was provided from Cardion AG (Erkrath, Germany). Celine Bauwens is supported by a fellowship from the Stem Cell Network, a Canadian Center of Excellence. P.W.Z. is a Canada Research Chair in Stem Cell Bioengineering. R.K. Li (University Health Network, Toronto, ON, Canada) is acknowledged for the use of laboratory space at the beginning of this project.

14.

15.

16.

REFERENCES 1. Soonpaa, M.H., and Field, L.J. Survey of studies examining mammalian cardiomyocyte DNA synthesis. Circ. Res. 83, 15, 1998. 2. Anversa, P., and Kajstura, J. Ventricular myocytes are not terminally differentiated in the adult mammalian heart. Circ. Res. 83, 1, 1998. 3. Erokhina, I.L., and Rumyantsev, P.P. Proliferation and biosynthetic activities of myocytes from conductive system and working myocardium of the developing mouse heart:

17.

18.

Light microscopic autoradiographic study. Acta Histochem. 84, 51, 1988. Swynghedauw, B. Molecular mechanisms of myocardial remodeling. Physiol. Rev. 79, 215, 1999. Anversa, P., Fitzpareick, D., Argani, S., and Capasso, J.M. Myocyte mitotic division in the aging mammalian rate heart. Circ. Res. 69, 1159, 1991. Reinlib, L., and Field, L. Cell transplantation as future therapy for cardiovascular disease? A workshop of the National Heart, Lung, and Blood Institute. Circulation 101, E182, 2000. Soonpaa, M.H., Koh, G.Y., Klug, M.G., and Field, L.J. Formation of nascent intercalated disks between grafted fetal cardiomyocytes and host myocardium. Science 264, 98, 1994. Koh, G.Y., Soonpaa, M.H., Klug, M.G., Pride, H.P., Cooper, B.J., Zipes, D.P., and Field, L.J. Stable fetal cardiomyocyte grafts in the hearts of dystrophic mice and dogs. J. Clin. Invest. 96, 2034, 1995. Leor, J., Patterson, M., Quinones, M.J., Kedes, L.H., and Kloner, R.A. Transplantation of fetal myocardial tissue into the infarcted myocardium of rat: A potential method for repair of infarcted myocardium? Circulation 94, II332, 1996. Li, R.K., Mickle, D.A., Weisel, R.D., Mohabeer, M.K., Zhang, J., Rao, V., Li, G., Merante, F., and Jia, Z.Q. Natural history of fetal rat cardiomyocytes transplanted into adult rat myocardial scar tissue. Circulation 96, (9 Suppl.), II-304, 1997. Jia, Z.Q., Mickle, D.A., Weisel, R.D., Mohabeer, M.K., Merante, F., Rao, V., Li, G., and Li, R.K. Transplanted cardiomyocytes survive in scar tissue and improve heart function. Transplant. Proc. 29, 2093, 1997. Scorsin, M., Hagege, A.A., Dolizy, I., Marotte, F., Mirochnik, N., Copin, H., Barnoux, M., le Bert, M., Samuel, J.L., Rappaport, L., and Menasche, P. Can cellular transplantation improve function in doxorubicin-induced heart failure? Circulation 98, II151, 1998. Watanabe, E., Smith, D.M., Jr., Delcarpio, J.B., Sun, J., Smart, F.W., Van Meter, C.H., Jr., and Claycomb, W.C. Cardiomyocyte transplantation in a porcine myocardial infarction model. Cell Transplant 7, 239, 1998. Gojo, S., Kitamura, S., Germeraad, W.T., Yoshida, Y., Niwaya, K., and Kawachi, K. Ex vivo gene transfer into myocardium using replication-defective retrovirus. Cell Transplant. 5, S81, 1996. Gojo, S., Kitamura, S., Hatano, O., Takakusu, A., Hashimoto, K., Kanegae, Y., and Saito, I. Transplantation of genetically marked cardiac muscle cells. J. Thorac. Cardiovasc. Surg. 113, 10, 1997. Reinecke, H., Zhang, M., Bartosek, T., and Murry, C.E. Survival, integration, and differentiation of cardiomyocyte grafts: A study in normal and injured rat hearts. Circulation 100, 193, 1999. Klug, M.G., Soonpaa, M.H., Koh, G.Y., and Field, L.J. Genetically selected cardiomyocytes from differentiating embronic stem cells form stable intracardiac grafts. J. Clin. Invest. 98, 216, 1996. Shake, J.G., Gruber, P.J., Baumgartner, W.A., Senechal, G., Meyers, J., Redmond, J.M., Pittenger, M.F., and Martin, B.J. Mesenchymal stem cell implantation in a swine myocardial infarct model: Engraftment and functional effects. Ann. Thorac. Surg. 73, 1919, 2002.

SCALABLE PRODUCTION OF ES CELL-DERIVED CARDIOMYOCYTES 19. Orlic, D., Kajstura, J., Chimenti, S., Jakoniuk, I., Anderson, S.M., Li, B., Pickel, J., McKay, R., Nadal-Ginard, B., Bodine, D.M., Leri, A., and Anversa, P. Bone marrow cells regenerate infarcted myocardium. Nature 410, 701, 2001. 20. Koh, G.Y., Klug, M.G., Soonpaa, M.H., and Field, L.J. Differentiation and long-term survival of C2C12 myoblast grafts in heart. J. Clin. Invest. 92, 1548, 1993. 21. Robinson, S.W., Cho, P.W., Levitsky, H.I., Olson, J.L., Hruban, R.H., Acker, M.A., and Kessler, P.D. Arterial delivery of genetically labelled skeletal myoblasts to the murine heart: Long-term survival and phenotypic modification of implanted myoblasts. Cell Transplant. 5, 77, 1996. 22. Murry, C.E., Wiseman, R.W., Schwartz, S.M., and Hauschka, S.D. Skeletal myoblast transplantation for repair of myocardial necrosis. J. Clin. Invest. 98, 2512, 1996. 23. Taylor, D.A., Atkins, B.Z., Hungspreugs, P., Jones, T.R., Reedy, M.C., Hutcheson, K.A., Glowerm D.D., and Kraus, W.E. Regenerating functional myocardium: Improved performance after skeletal myoblast transplantation. Nat. Med. 4, 929, 1998. 24. Chiu, R.C., Zibaitis, A., and Kao, R.L. Cellular cardiomyoplasty: Myocardial regeneration with satellite cell implantation. Ann. Thorac. Surg. 60, 12, 1995. 25. Muller-Ehmsen, J., Peterson, K.L., Kedes, L., Whittaker, P., Dow, J.S., Long, T.I., Laird, P.W., and Kloner, R.A. Rebuilding a damaged heart: Long-term survival of transplanted neonatal rat cardiomyocytes after myocardial infarction and effect on cardiac function. Circulation 105, 1720, 2002. 26. Doetschman, T.C., Eistetter, H., Katz, M., Schmidt, W., and Kemler, R. The in vitro development of blastocyst-derived embryonic stem cell lines: Formation of visceral yolk sac, blood islands and myocardium. J. Embryol. Exp. Morphol. 87, 27, 1985. 27. Wobus, A.M., Kaomei, G., Shan, J., Wellner, M.C., Rohwedel, J., Ji, G., Fleischmanm, B.K., Katus, H.A., Hescheler, J., and Franze, W.M. Retinoic acid accelerates embryonic stem cell-derived cardiac differentiation and enhances development of ventricular cardiomyocytes. J. Mol. Cell Cardiol. 29, 1529, 1997. 28. Sanchez, A., Jones, W.K., Gulick, J., Doetschman, T., and Robbins, J. Myosin heavy chain gene expression in mouse embryoid bodies: An in vitro developmental study. J. Biol. Chem. 266, 22419, 1991. 29. Muthuchamy, M., Pajak, L., Howles, L., Doetschman, T., and Wieczorek, D.F. Developmental analysis of tropomyosin gene expression in embryonic stem cells and mouse embryos. Mol. Cell. Biol. 13, 3311, 1993. 30. Miller-Hance, W.C., LaCorbiere, M., Fuller, S.J., Evans, S.M., Lyons, G., Schmidt, C., Robbins, J., and Chien, K.R. In vitro chamber specification during embryonic stem cell cardiogenesis. Expression of the ventricular myosin light chain-2 gene is independent of heart tube formation. J. Biol. Chem. 268, 25244, 1993. 31. Ganim, J.R., Luo, W., Ponniah, S., Grupp, I., Kim, H.W., Ferguson, D.G., Kadambi, V., Neumann, J.C., Doetschman, T., and Kranias, E.G. Mouse phospholamban gene expression during development in vivo and in vitro. Circ. Res. 71, 1021, 1992. 32. Boer, P.H. Activation of the gene for type-b natriuretic fac-

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

777

tor in mouse stem cell cultures induced for cardiac myogenesis. Biochem. Biophys. Res. Commun. 199, 954, 1994. Metzger, J.M., Lin, W.-I. and Samuelson, L.C. Transition in cardiac contractile sensitivity to calcium during the in vitro differentiation of mouse embryonic stem cells. J. Cell. Biol. 126, 701, 1994. Wobus, A.M., Wallukat, G., and Hescheler, J. Pluripotent mouse embryonic stem cells are able to differentiate into cardiomyocytes expressing chronotropic responses to adrenergic and cholinergic agents and Ca plus plus channel blockers. Differentiation 48, 173, 1991. Maltsev, V.A., Rohwedel, J., Hescheler, J., and Wobus, A.M. Embryonic stem cells differentiate in vitro into cardiomyocytes representing sinus-nodal, atrial and ventricular cell types. Mech. Dev. 44, 41, 1993. Klug, M.G., Soonpaa, M.H., and Field, L.J. DNA synthesis and multinucleation in embryonic stem cell-derived cardiomyocytes. Am. J. Physiol. 269, H1913, 1995. Pasumarthi, K.B., and Field, L.J. Cardiomyocyte enrichment in differentiating ES cell cultures: Strategies and applications. Methods Mol. Biol. 185, 157, 2002. Nagy, A., Rossant, J., Nagy, R., Abramow-Newerly, W. and Roder, J.C. Derivation of completely cell culture-derived mice from early-passage embryonic stem cells. Proc. Natl. Acad. Sci. U.S.A. 90, 8424, 1993. Soudais, C., Heikinheimo, B.M., MacArthur, C.A., Narita, N., Saffitz, J.E., Simon, M.C., Leiden, J.M., and Wilson, D.B. Targeted mutagenesis of the transcription factor GATA-4 gene in mounse embryonic stem cells disrupts visceral endoderm differentiation in vitro. Development 121, 3877, 1995. Zijlstra, M., Li, E., Sajjadi, F., Subramani, S., and Jaenisch, R. Germ-line transmission of a disrupted b2-microglobulin gene produced by homologous recombination in embryonic stem cells. Nature 342, 435, 1989. Claycomb, W.C., Lanson, N.A., Jr., Stallworth, B.S., Egeland, D.B., Delcarpio, J.B., Bahinski, A., and Izzo, N.J., Jr. HL-1 cells: A cardiac muscle cell line that contracts and retains phenotypic characteristics of the adult cardiomyocyte. Proc. Natl. Acad. Sci. U.S.A. 95, 2979, 1998. Viswanathan, S., Benatar, T., Rose-John, S., Lauffenburger, D.A., and Zandstra, P.W. Ligand/receptor signaling threshold (LIST) model accounts for gp130-mediated embryonic stem cell self-renewal responses to LIF and HIL-6. Stem Cells 20, 119, 2002. Dang, S.M., Kyba, M., Perlingeiro, R., Daley, G.Q., and Zandstra, P.W. Efficiency of embryoid body formation and hematopoietic development from embryonic stem cells in different culture systems. Biotechnol. Bioeng. 78, 442, 2002. Czyz, J., and Wobus, A. Embryonic stem cell differentiation: The role of extracellular factors. Differentiation 68, 167, 2001. Boheler, K.R., Czyz, J., Tweedie, D., Yang, H.T., Anisimov, S.V., and Wobus, A.M. Differentiation of pluripotent embryonic stem cells into cardiomyocytes. Circ. Res. 91, 189, 2002. Doevendans, P.A., Kubalak, S.W., An, R.H., et al. Differentiation of cardiomyocytes in floating embryonic bodies is comparable to fetal cardiomyocytes. J. Mol. Cell. Cardiol. 32, 839, 2000.

778 47. Doevendans, P.A., Kubalak, S.W., An, R.H., Becker, D.K., Chien, K.R., and Kass, R.S. Differentiation of cardiomyocytes in floating embryoid bodies is comparable to fetal cardiomyocytes. J. Mol. Cell Cardiol. 32, 839, 2000. 48. Kehat, I., Kenyagin-Karsenti, D., Snir, M., Segev, H., Amit, M., Gepstein, A., Livne, E., Binah, O., Itskovitz-Eldor, J., and Gepstein, L. Human embryonic stem cells can differentiate into myocytes with structural and functional properties of cardiomyocytes. J. Clin. Invest. 108, 407, 2001. 49. Tran, C.M., and Sucov, H.M. The RXRa gene functions in a non-cell-autonomous manner during mouse cardiac morphogenesis. Development 125, 1951, 1998. 50. Gajovic, S., St-Onge, L., Yokota, Y., and Gruss, P. Retinoic acid mediates Pax6 expression during in vitro differentiation of embryonic stem cells. Differentiation 62, 187, 1997. 51. Shamblott, M.J., Axelman, J., Littlefield, J.W., Blumenthal, P.D., Huggins, G.R., Cui, Y., Cheng, L., and Gearhart, J.D. Human embryonic germ cell derivatives express a broad range of developmentally distinct markers and proliferate extensively in vitro. Proc. Natl. Acad. Sci. U.S.A. 98, 113, 2001. 52. Thomson, J.A., Itskovitz-Eldor, J., Shapiro, S.S., Waknitz, M.A., Swiergiel, J.J., Marshall, V.S., and Jones, J.M. Embryonic stem cell lines derived from human blastocysts. Science 282, 1145, 1998. 53. Clarke, D.L., Johansson, C.B., Wilbertz, J., Veress, B., Nilsson, E., Karlstrom, H., Lendahl, U., and Frisen, J. Generalized potential of adult neural stem cells. Science 288, 1660, 2000. 54. Jackson, K.A., Majka, S.M., Wang, H., Pocius, J., Hartley, C.J., Majesky, M.W., Entman, M.L., Michael, L.H., Hirschi, K.K., and Goodell, M.A. Regeneration of ischemic cardiac muscle and vascular endothelium by adult stem cells. J. Clin. Invest. 107, 1395, 2001.

ZANDSTRA ET AL. 55. Malouf, N.N., Coleman, W.B., Grisham, J.W., Lininger, R.A., Madden, V.J., Sproul, M., and Anderson, P.A. Adultderived stem cells from the liver become myocytes in the heart in vivo. Am. J. Pathol. 158, 1929, 2001. 56. Quaini, F., Urbanek, K., Beltrami, A.P., Finato, N., Beltrami, C.A., Nadal-Ginard, B., Kajstura, J., Leri, A., and Anversa, P. Chimerism of the transplanted heart. N. Engl. J. Med. 346, 5, 2002. 57. Laflamme, M.A., Myerson, D., Saffitz, J.E., and Murry, C.E. Evidence for cardiomyocyte repopulation by extracardiac progenitors in transplanted human hearts. Circ. Res. 90, 634, 2002. 58. Hruban, R.H., Long, P.P., Perlman, E.J., Hutchins, G.M., Baumgartner, W.A., Baughman, K.L., and Griffin, C.A. Fluorescence in situ hybridization for the Y-chromosome can be used to detect cells of recipient origin in allografted hearts following cardiac transplantation. Am. J. Pathol. 142, 975, 1993. 59. Jiang, Y., Jahagirdar, B.N., Reinhardt, R.L., Schwartz, R.E., Keene, C.D., Ortiz-Gonzalez, X.R., Reyes, M., Lenvik, T., Lund, T., Blackstad, M., Du, J., Aldrich, S., Lisberg, A., Low, W.C., Largaespada, D.A., and Verfaillie, C.M. Pluripotency of mesenchymal stem cells derived from adult marrow. Nature 418, 41, 2002.

Address reprint requests to: Peter W. Zandstra, Ph.D. Institute of Biomaterials and Biomedical Engineering Room 407, Rosebrugh Building 4 Taddle Creek Road Toronto, ON M5S-3E3, Canada E-mail: [email protected]