Improved cardiac function in infarcted mice after treatment with pluripotent embryonic stem cells


  • Timothy J. Nelson,

    1. Department of Cell Biology, Neurobiology, and Anatomy and the Cardiovascular Center Medical College of Wisconsin, Milwaukee, Wisconsin
    Search for more papers by this author
  • Zhi-Dong Ge,

    1. Department of Pharmacology and Toxicology and the Cardiovascular Center Medical College of Wisconsin, Milwaukee, Wisconsin
    Search for more papers by this author
  • Jordan Van Orman,

    1. Department of Cell Biology, Neurobiology, and Anatomy and the Cardiovascular Center Medical College of Wisconsin, Milwaukee, Wisconsin
    Search for more papers by this author
  • Matthew Barron,

    1. Department of Cell Biology, Neurobiology, and Anatomy and the Cardiovascular Center Medical College of Wisconsin, Milwaukee, Wisconsin
    Search for more papers by this author
  • Diane Rudy-Reil,

    1. Department of Cell Biology, Neurobiology, and Anatomy and the Cardiovascular Center Medical College of Wisconsin, Milwaukee, Wisconsin
    Search for more papers by this author
  • Timothy A. Hacker,

    1. Department of Medicine, Section of Cardiovascular Medicine, University of Wisconsin, Madison, Wisconsin
    Search for more papers by this author
  • Ravi Misra,

    1. Department of Biochemistry and the Cardiovascular Center Medical Collegeof Wisconsin, Milwaukee, Wisconsin
    Search for more papers by this author
  • Stephen A. Duncan,

    1. Department of Cell Biology, Neurobiology, and Anatomy and the Cardiovascular Center Medical College of Wisconsin, Milwaukee, Wisconsin
    Search for more papers by this author
  • John A. Auchampach,

    1. Department of Pharmacology and Toxicology and the Cardiovascular Center Medical College of Wisconsin, Milwaukee, Wisconsin
    Search for more papers by this author
  • John W. Lough

    Corresponding author
    1. Department of Cell Biology, Neurobiology, and Anatomy and the Cardiovascular Center Medical College of Wisconsin, Milwaukee, Wisconsin
    • Department of Cell Biology, Neurobiology and Anatomy, Medical College of Wisconsin, 8701 Watertown Plank Road, Milwaukee, WI 53226
    Search for more papers by this author
    • Fax: 414-456-6517


Because pluripotent embryonic stem cells (ESCs) are able to differentiate into any tissue, they are attractive agents for tissue regeneration. Although improvement of cardiac function has been observed after transplantation of pluripotent ESCs, the extent to which these effects reflect ESC-mediated remuscularization, revascularization, or paracrine mechanisms is unknown. Moreover, because ESCs may generate teratomas, the ability to predict the outcome of cellular differentiation, especially when transplanting pluripotent ESCs, is essential; conversely, a requirement to use predifferentiated ESCs would limit their application to highly characterized subsets that are available in limited numbers. In the experiments reported here, we transplanted low numbers of two murine ESC lines, respectively engineered to express a β-galactosidase gene from either a constitutive (elongation factor) or a cardiac-specific (α-myosin heavy chain) promoter, into infarcted mouse myocardium. Although ESC-derived tumors formed within the pericardial space in 21% of injected hearts, lacZ histochemistry revealed that engraftment of ESC was restricted to the ischemic myocardium. Echocardiographic monitoring of ESC-injected hearts that did not form tumors revealed functional improvements by 4 weeks postinfarction, including significant increases in ejection fraction, circumferential fiber shortening velocity, and peak mitral blood flow velocity. These experiments indicate that the infarcted myocardial environment can support engraftment and cardiomyogenic differentiation of pluripotent ESCs, concomitant with partial functional recovery. Anat Rec Part A, 288A:1216–1224, 2006. © 2006 Wiley-Liss, Inc.

The potential of stem cells to regenerate adult tissues has caused extraordinary interest in their therapeutic application. Stem cells are classified as adult or embryonic, according to their respective origins from within niches of mature adult tissues and bone marrow, or from blastocyst-stage embryos. The possibility that adult stem cells can repair infarcted myocardium was indicated in experiments using animal models (Quaini et al., 2002; Beltrami et al., 2003; Mangi et al., 2003). Although initial results from trials in human patients using autologous bone marrow cells were encouraging (Perin et al., 2004; Schachinger et al., 2004; for review, see Muller et al., 2005), it was recently reported that improvements in cardiac function were only temporary (Meyer et al., 2006). Moreover, because the basis for improved recovery is unlikely mediated by remuscularization of damaged myocardium, the need to evaluate cells capable of differentiating into contractile tissue has been emphasized (for review, see Wollert and Drexler, 2006).

By contrast with adult stem cells, the robust ability of embryonic stem cells (ESCs) to differentiate into cardiac myocytes both in vitro (Kanno et al., 2004; Rudy-Reil and Lough, 2004) and in vivo (Hodgson et al., 2004) warrants their evaluation as therapeutic agents. In this regard, recent reports have indicated that transplanted pluripotent ESCs confer modest functional improvement of infarcted myocardium in mouse (Kofidis et al., 2005; Singla et al., 2006) and rat (Behfar et al., 2002; Min et al., 2002, 2003; Hodgson et al., 2004) models. Although monitoring of differentiated ESC outcomes has indicated that pluripotent donor cells can differentiate into cell types present in the adult heart (Behfar et al., 2002; Hodgson et al., 2004; Singla et al., 2006), the reliance on fluorescent signals to establish identities of the transplanted cells has been questioned because infarcted myocardium is autofluorescent (for reviews, see Laflamme and Murry, 2005; Terman and Brunk, 2005). To circumvent this potential problem, we developed two lines of ESCs that enable lacZ-based histochemical distinction between donor cells that constitutively express the elongation factor (EF) promoter and donor cells that differentiate into cardiomyocytes per cardiac-specific expression of the α-myosin heavy chain (α-MHC) promoter.

We have used these ESC lines to evaluate cardiac regeneration in a mouse model of myocardial infarction in which the left anterior descending (LAD) and the first proximal branch of the left circumflex (LCx) arteries are permanently ligated, followed 1 hr later by injection of a relatively low number (50,000) of pluripotent ESCs. Eight weeks later, donor cells were identified in ischemic areas only based on expression of constitutively expressed β-galactosidase. Evidence that at least some of the transplanted cells differentiated into cardiomyocytes was indicated by expression of α-MHC-dependent β-galactosidase. Although ESC-derived teratomas developed in the pericardium of 21% of the transplanted hearts, the absence of engrafted donor cells within the myocardium of these hearts suggested that infarcted myocardium does not support tumor growth. Echocardiographic evaluation of infarcted/transplanted hearts that did not contain tumors revealed modest albeit significant improvements in several parameters of cardiac function, including ejection fraction, circumferential fiber shortening velocity (VCF), and peak mitral blood flow velocity (peak-E). These findings add to the weight of evidence that pluripotent ESCs, even in small numbers, can confer significant functional improvement to infarcted myocardium and indicate that this environment can support the differentiation of pluripotent ESCs into cardiomyocytes, a requirement for myocardial remuscularization.


Genetically Engineered ESCs

The line of murine embryonic stem cells used in these experiments, R1 ES cells, was prepared from a male E3.5 blastocyst resultant from crossing 129X1/SvJ and 129S1 mice. To create the ES-R1-F3 subline (hereafter F3 cells), the proximal promoter and first exon of the X-linked endogenous HPRT gene were replaced with a neomycin-resistance cassette, causing this cell line to be highly sensitive to growth in medium containing hypoxanthine, aminopterin, and thymidine (HAT) (Misra et al., 2001). F3 cells can be rescued by homologous recombination with an HPRT targeting vector containing the missing HPRT gene sequences, thereby conferring resistance to HAT treatment and effectively selecting for single-copy transgenic ESCs. Construction of the expression cassette containing the α-MHC promoter (kindly provided by Dr. Jeffrey Robbins of Cincinnati Children's Hospital Medical Center) (Gulick et al., 1991) upstream of β-galactosidase cDNA was described previously (Misra et al., 2001); this construct is hereafter referred to as α-MHC-lacZ. The expression cassette containing the elongation factor promoter was kindly provided by Dr. Sangmi Chung of Harvard University; this promoter is constitutively expressed in pluripotent and differentiated ESCs (Chung et al., 2002). To prepare this cassette, the ∼ 1 kbp EF promoter was removed from pEF-hrGFP (Chung et al., 2002) by Not1/Nsi1 digestion, blunted, and ligated into a shuttle vector, pMB105lacZ (Misra et al., 2001), which contains a nuclear β-galactosidase cDNA (from pLacF plasmid) and 3′-UTR. The resultant cassette, hereafter termed EF-lacZ, was subcloned into the Not1 site of pMP8NEB-lacZ, from which 100 μg were purified, linearized with restriction endonuclease PME1, and resuspended along with 1.5 × 108 F3-ESCs in 3.1 ml growth medium. The mixture (600 μl) was electroporated in 4 mm gap cuvettes (BioRad) at 250 volts/500 μf/r8 using a BTX Electro Cell Manipulator 600. The F3-ESCs were plated on mouse embryonic fibroblast (MEF) feeders in standard culture medium containing leukocyte inhibitory factor (LIF). HAT (Sigma) selection was started 2 days after electroporation and 10–20 colonies of putative F3-ESC-EF-lacZ cells were isolated within 10–16 days. Transgenic cell lines were genotyped by PCR amplification of the lacZ gene using the primer pair 5′-CACCGATCGCCCTTCCCAACAGTT-3′ and 5′-TGCCGCTCATCCGCCACATA-3′. Accuracy of targeting into the HPRT locus was determined by Southern blotting (Misra et al., 2001).

Preparation of F3 Cells for Injection

All cell culture components were purchased from Specialty Media (Phillipsburg, NJ) except where indicated. To prepare F3 cells for injection, aliquots were expanded without feeders in ES cell culture medium consisting of DMEM supplemented with 1 mM sodium pyruvate, 0.1 M nonessential amino acids, 2 mM glutamine, 0.2 mM β-mercaptoethanol, 1 × pen/strep antiobiotics, pretested 15% fetal calf serum (FCS; Hyclone), and LIF at a concentration (i.e., 1 × LIF) that was empirically determined to preserve pluripotency of ESCs for a minimum of four population doublings. F3 cells were harvested with trypsin/EDTA to form a cell suspension that was counted with a hemocytometer. Cells for intracardiac injection were washed by gentle centrifugation and resuspended in ES medium (without FCS or LIF) and adjusted to a final concentration of 50,000 cells/10 μl.

Myocardial Infarction and ESC Transplantation

To prepare for infarction, 10- to 14-week-old male C57BL/6 mice were anesthetized with sodium pentobarbital (100 mg/kg i.p.) and placed on a warm heating pad to maintain body temperature at 37°C. A PE-60 polyethylene tube was inserted into the trachea and connected to a mouse ventilator (Hugo Sachs Elektronic Model 845, Germany). Mice were respirated (tidal volume = 225 μL; rate, 100 strokes/min) with room air supplemented with 100% oxygen to maintain blood gases within normal physiological limits. An electrocardiogram was obtained with needle electrodes using the limb lead II configuration. A thoracotomy was performed at the fourth left intercostal space to expose the heart, the pericardium was opened, and an 8-0 nylon suture was ligated around the LAD coronary artery ∼ 3 mm from the tip of the normally positioned left atrium with the aid of a dissecting microscope. A second ligature was placed around the first proximal branch of the LCx coronary artery supplying the anterior/lateral wall of the heart.

Experimental mice were injected with either 50 × 103 α-MHC-lacZ cells or 50 × 103 EF-lacZ cells, suspended in 10 μl ES medium without FCS or LIF. In both cases, control mice received 10 μl ES medium only. Ten sham-operated mice were prepared identical to the experimental mice described in Table 2, except that the ligatures in these mice were not tightened (i.e., the mice were not subjected to ischemia) and they did not receive cells or medium. The injections were performed 1 hr after ligation using a Hamilton syringe to an area along the inferior border of the infarcted area. After injection, the chest wall was closed with a 7-0 polypropylene suture with one layer through the chest wall and muscle, and a second layer through the skin and subcutaneous tissue. Mice were removed from the ventilator and kept in a warm chamber in which room air was supplemented with 100% oxygen. The endotracheal tube was withdrawn after 30 min.


Echocardiography of infarcted mice injected with EF-lacZ cells and parallel ES medium controls was performed at the University of Wisconsin Department of Medicine Cardiovascular Physiology Core facility on mice lightly anesthetized with ketamine chloride (35 mg/kg i.p.) using a Sonos 5500 echocardiograph (Agilent Technologies, Palo Alto, CA) equipped with a 15 MHz pediatric transducer. Echocardiography of infarcted mice injected with α-MHC-lacZ cells (and parallel ES medium controls) was performed at the Medical College of Wisconsin Cardiovascular Center core facility on mice anesthetized with 1% isoflurane using a VisualSonics Vevo 770 high-frequency ultrasound rodent imaging system equipped with a 30 MHz mouse probe designed for cardiac imaging (VS-RMV-707). In both instances, parasternal short-axis, long-axis, and apical views were recorded. The parasternal short-axis view was used to measure anteroposterior internal diameter (D), anterior wall thickness (AW), and posterior wall thickness (PW) at end-diastole (D) and end-systole (S) at the mid-papillary level. The apical position was used to monitor inflow through the mitral valve and outflow through the aortic valve by pulsed Doppler to measure isovolumic relaxation, isovolumic contraction, and ejection times. Left ventricular systolic function was assessed by fractional shortening {FS, % = [(LVDd − LVDs)/LVEDd] × 100}, mean velocity of circumferential fiber shortening (Vcf, circumferences/second = FS/ejection time), and the myocardial performance index [MPI = (isovolumic contraction time + isovolumic relaxation time)/ejection time].


After 9 weeks, mice were anesthetized with sodium pentobarbital (100 mg/kg i.p.) and hearts were rapidly exposed and perfused with ice-cold cardioplegic solution (25 mM KCl/5% dextrose) to arrest during diastolic relaxation. Hearts were rapidly removed, atria with AV valves were excised, prior to placing ventricles in 0.25% glutaraldehyde fixative for 30 min. X-gal staining was performed as previously described (Nelson et al., 2004) overnight at 37°C. After photographing intact ventricles, these were transversely sectioned through the plane at the site of ligation and embedded by rapid freezing in OCT medium for subsequent sectioning at 7–10 micrometer thickness to detect lacZ+ cells. Parallel sections were stained with Masson's trichrome (catalog number HT15; Sigma-Aldrich, St. Louis, MO) to detect areas of collagen deposition.


Derivation of ESC Lines That Express β-Galactosidase in Constitutive and in Tissue-Specific Fashion

To identify transplanted donor cells, we generated two lines of ESCs, termed EF-lacZ and α-MHC-lacZ cells, that respectively express nuclear β-galactosidase under control of the constitutive elongation factor (EF) and cardiac-specific α-MHC promoters. Either promoter was cloned upstream of the lacZ expression cassette, subcloned into the mouse HPRT targeting vector, and electroporated for site-specific homologous recombination with the HPRT locus in R1-derived F3 ES cells. These procedures are schematically summarized in Figure 1A. This targeting strategy enables highly efficient screening in HAT medium, resulting in greater than 95% correctly targeted clones that contain a single, nonepigenetically inserted transgene (Misra et al., 2001; Nelson et al., 2004). By avoiding random integration of multiple gene copies, the resulting ESCs are genetically identical to the parental cells except for the transgene. Figure 1B demonstrates high levels of β-galactosidase in the EF-lacZ line, which expresses this marker regardless of differentiative state, in cultured pluripotent ESCs. Figure 1C shows that the α-MHC-lacZ construct is exclusively expressed in the hearts of embryos prepared from these cells by tetraploid aggregation per the technique previously described (Misra et al., 2001); note that because tetraploid aggregation produces embryos that are completely derived from ESCs, all cells within the embryo contain the same genome.

Figure 1.

Constitutive expression of EF-lacZ, and cardiac-specific expression of α-MHC-lacZ, enables respective identification of donor cells and cardiomyocytes. A: Scheme of transgene insertion into the HPRT locus of F3-ESCs, enabling constitutive expression of lacZ from the elongation factor promoter or cardiac-specific expression from the α-MHC promoter. F3-ESCs have a Neo gene replacing the native HPRT promoter (P) and exon 1. The HPRT targeting vector, pMP8NEB-lacZ, removes the Neo gene, and reconstitutes the HPRT promoter and exon 1, while introducing the transgene. B: Transgenic ESCs stained with X-gal, showing expression of lacZ in cultured EF-lacZ cells and (C) in the hearts of embryos derived solely from α-MHC-lacZ cells via tetraploid aggregation. D: To induce infarction, ligatures were placed around the LAD and first branch of the LCx arteries. EF-lacZ or α-MHC-lacZ cells were transplanted 60 min later, at the inferior border of the infarction site (red circle). ECG changes were noted in all animals after placing the two ligatures; after 60 min, a Q-wave and blanching of the infarcted myocardium were observed. E: Photograph showing that this protocol produced infarction of the anterolateral free wall of the left ventricle, resulting in significant thinning of the myocardial wall between the ligatures.

Transplantation and Survival of EF-lacZ and α-MHC-lacZ Cells in Acutely Infarcted Mouse Myocardium

To assess whether small numbers of pluripotent ESCs can be used in a mouse model of myocardial infarction, conditions were established that permit reproducible large infarctions without causing high mortality. A double-ligation procedure was utilized; distal ligation of the LAD artery and a branch of the LCx artery (Fig. 1D) caused infarction of nearly the entire anterior and lateral walls of the left ventricle (Fig. 1E). Although this two-vessel ligation caused ECG changes within 60 min (Fig. 1D) and significant thinning of the anterior/lateral free wall (Fig. 1E), the perioperative mortality rate was only 11%, followed by a 95% survival rate after 9 weeks. One hour after ligation, a single injection of 10 μl containing 50,000 ESCs, or vehicle only, was introduced to the infarction site as illustrated in Figure 1E.

Eight weeks after injecting either cell line, mice were subjected to echocardiography and hearts were histochemically reacted with X-gal to detect transplanted ESCs (Fig. 2). Among 12 hearts injected with EF-LacZ cells, 11 exhibited β-gal-positive cells that were essentially homogeneously distributed throughout the scar tissue; β-gal-positive cells were not detected outside of the infarcted area, which is indicated by the broken line in Figure 2A. The single heart that did not exhibit β-gal-positive cells in the myocardium exhibited a large (5 mm diameter) teratoma that appeared to be confined to the pericardial sac (Fig. 2C). Figure 2B shows sections from the ischemic area of X-gal-stained hearts that had been infarcted and injected with medium only (Fig. 2B, left panels) or EF-lacZ cells (Fig. 2B, right panels); in the latter, β-gal-positive transplanted cells detected by X-gal staining were accompanied by a relatively extensive deposition of collagen fibers as revealed by Masson trichrome staining. Figure 2D shows a section removed from a similar location of an infarcted X-gal-reacted heart that had been injected with α-MHC-LacZ cells; in this instance, β-gal expression indicates differentiation into cardiomyocytes (Misra et al., 2001). Hearts injected with MHC-LacZ cells also exhibited tumor formation, at an incidence of 4 in 12; hence, the overall tumor incidence using both cell lines was 5 in 24, an occurrence rate of ∼ 21%.

Figure 2.

Allogeneic ESCs survive after transplantation into acutely infarcted myocardium. A: After 9 weeks, whole hearts stained with X-gal revealed transplanted EF-lacZ cells throughout the infarcted tissue, boundaries of which are denoted by the broken line; red circle denote site of injection. B: Top: whole mount of X-gal-stained heart showing the presence of EF-lacZ donor cells in infarcted area and the absence of staining in control hearts. Scale bar = 750 μm. Middle: Cross-sections of the X-gal-stained hearts shown in the top showing homogeneous distribution of EF-lacZ cells. Scale bar = 200 μm. Bottom: Comparison of collagen deposition, assessed by Masson's trichrome staining, between hearts injected with EF-lacZ cells or with medium only (controls). Scale bar = 200 μm. β-gal+ cells were present only in the scar tissue observed in the anterior/lateral free wall. C: Heart injected with EF-lacZ cells containing a β-gal+ tumor located between epicardium and pericardium. This mass, which is magnified in the lower part of C, was not incorporated into the myocardium. The blue color indicates the origin of cells in this tumor from the EF-lacZ cells; however, whether they are cardiomyocytes cannot be discerned. Scale bar = 500 μm. D: Cross-section of X-gal-reacted whole mount heart that had been infarcted and injected with α-MHC-lacZ cells; the blue color denotes de novo cardiogenesis. Scale bar = 200 μm.

Although it was not possible to enumerate the relative numbers of β-gal-positive cells from each cell line in this study, results indicate that ischemic myocardium can support the survival of transplanted pluripotent ESCs (Fig. 2A and B), and that some of these differentiate into cardiomyocytes (Fig. 2D). Moreover, because ESC-derived tumors have been observed only in the pericardial sac (Fig. 2C), not in healthy or ischemic myocardium, the possibility is suggested that growth factors and/or cytokines limited to the latter environment may nurture the differentiation of ESCs without promoting proliferation signals that lead to unregulated cellular growth.

Transplantation of Only 50,000 Pluripotent EF-lacZ or α-MHC-lacZ Cells During Acute Myocardial Infarction ImprovesCardiac Function

Cardiac function in animals injected with medium only (controls) or 50,000 cells of either ESC line 1 hr after infarction was assessed by echocardiography 8 weeks later. Tables 1 and 2 present summaries of the echocardiographic results using EF-lacZ and α-MHC-lacZ cells, respectively. Using EF-lacZ cells (Table 1), although there were no differences in left ventricular dimensions during diastole (AWd, PWd, and LVDd) between the two experimental groups, mice transplanted with EF-lacZ ESCs exhibited significant improvements in indexes of systolic function including LVDs, fractional shortening, and velocity of circumferential shortening. Moreover, the myocardial performance index (MPI), a Doppler-derived index of global left ventricular function (Broberg et al., 2003), was increased to borderline statistical significance (P = 0.064). Representative M-mode echocardiograms of a vehicle-treated mouse heart and a heart injected with EF-lacZ cells are shown in Figure 3, indicating the improvement in anterior wall motion.

Figure 3.

Echocardiography demonstrates significant functional improvements in animals transplanted with EF-lacZ or α-MHC-lacZ cells. A and B: Representative M-mode echocardiographic tracings of parasternal short axis views of (A) infarcted hearts injected with medium only, which exhibit a significantly thin akinetic anterior LV wall and abnormal ventricular diameters during diastole and systole. Yellow arrow denotes luminal edge of anterior wall; red arrow, luminal edge of posterior wall; purple bar, distance between anterior wall and posterior wall during systole. B: Infarcted hearts injected with EF-lacZ cells, which displayed improved anterior wall motion and thickness, as well as normalized ventricular diameters. C: Left: Ejection fractions of noninfarcted sham-mice compared to medium controls and α-MHC-lacZ-injected hearts 4 and 8 weeks after infarction/transplantation, indicating significant improvement after ESC treatment. Right: The peak-E/peak-A ratio, which assesses diastolic function/ventricular filling, shows complete normalization 4 weeks after ESC transplantation. Error bars in C indicate the standard error of the mean (SEM). D: The extent of improvement in selected functional parameters conferred by either ESC line: LVDs, fractional shortening, and shortening velocity; by contrast, neither ESC line improved PWd or LVDd. The P values of < 0.05 above each pair of bars denote the probability that the extent of improvement is different from medium-injected controls.

Table 1. Cell Therapy with pluripotent EF-lacZ ES cells
 Medium(N = 12)Stem Cells(N = 11)P-Value(<0.05***)
  1. Statistical analysis of functional improvement in hearts injected with EF-lacZ cells (N = 11) or with medium only (N = 12), after eight weeks. All values are means ± SEM; significance was determined using Student's t-Test (P < 0.05).

  2. PWd, posterior wall in diastole; AWd, anterior wall (interventricular septum) in diastole; LVDd, left ventricular diameter in diastole; LVDs, left ventricular diameter in systole; RWth, relative wall thickness = 2*PWd/LVDd; % Fractional Shortening, (LVDd-LVDs)/LVDd; LV mass/BW, left ventricular mass in grams/body weight in grams; Peak E, peak velocity of blood through the mitral valve; IVRT, isovolumic relaxation time in seconds; MPI (Broberg et al., 2003), myocardial performance index or the ratio of isovolumic contraction and relaxation times to ejection time; Vcf, mean velocity of circumferential fiber shortening or shortening velocity (circumferences/s).

heart rate (beats/min)638±60668±340.171
PWd (mm)0.88±0.130.88±0.110.949
AWd (mm)0.67±0.160.73±0.120.388
LVDd (mm)3.85±0.333.62±0.440.176
LVDs (mm)2.75±0.492.29±0.470.038***
(%) fractional shortening29±837±70.026***
peak E (cm/sec)84±1091±120.179
IVRT (msec)15±114±20.084
Table 2. Cell Therapy with pluripotent α-MHC-lacZ ES cells
 Medium (N = 12)Stem Cells (N = 12)p-Value (<0.05***)Sham (N = 10)
  1. Statistical analysis of functional improvement in hearts injected with α-MHC-lacZ cells (N = 12) or with medium only (N = 12), after eight weeks. All values are means ± SEM; significance was determined using Student's t-Test (p < 0.05). Abbreviations are described in the legend for Table 1.

heart rate (beats/min)535±5534±40.917545±9
PWd (mm)0.88±0.070.88±0.060.9680.95±0.06
AWd (mm)0.47±0.050.71±0.050.001***0.85±0.07
LVDd (mm)4.61±0.214.18±0.100.0843.81±0.09
LVDs (mm)3.66±0.243.06±0.110.040***2.37±0.13
(%) fractional shortening22±227±10.040***38±3
ejection fraction (%)43±452±20.041***76±3
peak E (cm/sec)65±579±30.035***85±4
IVRT (msec)12.9±0.815.0±0.60.048***15.3±0.7

Infarcted hearts injected with 50,000 α-MHC-lacZ cells exhibited statistically significant improvements in the same parameters after 8 weeks (Table 2); in addition, other parameters that were not statistically improved in EF-lacZ-injected hearts were improved in hearts injected with α-MHC-lacZ-injected cells, including anterior wall diameter (diastole), the peak velocity of blood flow through the mitral valve (peak-E), and the isovolumic relaxation time (IVRT). Echocardiography performed on the same animals only 4 weeks after infarction/transplantation indicated that these functional parameters were improved by that time (4-week data are shown only for ejection fraction; Fig. 3C). The extent of functional improvement in selected parameters mediated by EF-lacZ and α-MHC-lacZ cells is compared in Figure 3D. Hence, cell therapy with α-MHC-lacZ ESCs confers functional improvements that are detectable by 4 weeks and are maintained throughout at least the next 4 weeks.


In these experiments, we modified two ESC lines to express a single-copy β-galactosidase transgene, activated by either a constitutive promoter (EF-lacZ) or by a cardiac-specific promoter (α-MHC-lacZ). Using this approach, we were able to monitor the fate of transplanted cells via histochemistry, rather than fluorescent markers, thereby avoiding the possibility of confounding false positive signals that are reportedly caused by autofluorescence generated by ischemic myocardium (Terman and Brunk, 2005). Because the expression of β-galactosidase reveals the presence of all injected EF-lacZ donor cells regardless of their differentiative state, while the expression of β-galactosidase by transplanted α-MHC-lacZ cells indicates their differentiation into cardiac myocytes, we were able to distinguish generic donor cells from donor cells that actually differentiated into cardiomyocytes. These results indicate that transplanted pluripotent ESCs can survive in ischemic myocardium, and that at least some of these are able to differentiate into cardiomyocytes, which are theoretically capable of contracting (Rudy-Reil and Lough, 2004). Using the approach employed in this study, coupled with refinements in histological processing, quantitative assessment of the percentages of EF-lacZ donor cells that differentiate into α-MHC-lacZ-positive cardiomyocytes should be possible.

Our observations that pluripotent ESCs engraft and survive only within ischemic myocardium imply that these transplanted cells cannot migrate into areas of viable myocardium, and that the ischemic environment uniquely provides trophic factors sufficient for the survival and differentiation of pluripotent ESCs. Regarding the observation that ESC-derived tumors occurred only in the pericardial sac, we speculate that this resulted from inadvertent injection of ESCs into this area, wherein pluripotent cells, in the absence of cardio-inductive molecules in the infarcted environment (Behfar et al., 2002), may undergo teratogenesis. This interpretation is consistent with findings recently reported by Behfar et al. (2005), who established that a therapeutic load of as many as 105 ESCs (i.e., 103 ESCs/mg heart weight) should not cause tumorigenesis.

That significant functional recovery of mouse myocardium is mediated by either pluripotent ESC line (Tables 1 and 2) confirms previous findings using relatively large numbers of pluripotent ESCs in this small animal model [Hodgson et al. (2004): 3 × 105 ESCs; Kofidis et al. (2005): 106 ESCs], as well as very recent findings in which low numbers of transplanted ESCs similar to those reported here were observed to confer functional benefit [Singla et al. (2006): 3 × 104 ESCs]. In comparing results in Tables 1 and 2, it is noted that because different drugs were used to induce mild anesthesia during echocardiography (ketamine for Table 1 and isoflurane, which is more cardio-depressive, in Table 2), it is speculated that this is the major cause of baseline differences in these data. In particular, the extent of improved fractional shortening, a parameter that approximates ejection fraction, reported in the cited studies is similar to the extent of improvement reported here. The aggregate of these findings provide strong evidence that small numbers of ESCs can confer functional benefit, which should enhance cell therapy paradigms using predifferentiated ESCs, pure cohorts of which are obtainable in relatively small numbers. This is important in view of our observations that at least some of the transplanted pluripotent ESCs differentiate into cardiomyocytes, suggesting that improved function, at least in part, results from de novo cardiomyogenesis.

Although these studies indicate that pluripotent ESCs confer improved function after transplantation, normative values were not reestablished. Although the benefit of transplanting α-MHC-lacZ cells was observed after 4 weeks (animals transplanted with EF-lacZ cells were not echocardiographed at that time) and was stable through at least 8 weeks, the extent of improvement between these two time points did not differ for any parameter. Nonetheless, whether infarcted hearts transplanted with pluripotent ESCs continue to improve toward normative values during longer-term recovery must be investigated. Moreover, the extent to which transplanted pluripotent ESCs differentiate and integrate into healthy myocardium, as well as whether predifferentiation of donor ESCs confers improvements in these parameters as well as function, is also being investigated.


Supported by NIH HL079277 (to J.L.) and a Jenkins Cardiovascular Research Fellowship Award (to T.J.N.).