INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

MYOCARDIAL INFARCTION (MI) is a life-threatening event that may cause sudden cardiac death and heart failure. Despite considerable advances in the diagnosis and treatment of heart disease, cardiac dysfunction after MI is still the major worldwide cardiovascular disorder (5). After acute MI, the damaged myocardium is gradually replaced by fibrotic noncontractile cells. The developing ventricular dysfunction is primarily due to a massive loss of cardiomyocytes. A recent study showed that mild proliferating myocytes derived from resident cardiomyocytes or circulating stem cells may contribute to the increase in muscle mass of the infarcted human myocardium (1). However, hypertrophy and mild proliferation without effective therapy do not attenuate the onset and progress of cardiac dysfunction after MI. Therefore, finding effective new approaches to improve cardiac dysfunction after MI remains a major therapeutic challenge.

Cell transplantation has emerged as a potential new approach for repairing damaged myocardium in recent years. Transplanted cardiomyocytes have been shown to survive, proliferate, and connect with the host myocardium in murine models (28). Li and co-workers (15-17) demonstrated that transplanted fetal cardiomyocytes could form new cardiac tissue within the myocardial scar induced by cryoinjury and improve heart function. Bishop et al. (2) reported that embryonic myocardium of rats can be implanted and cultured in oculo and demonstrated that the engrafted embryonic cardiomyocytes proliferated and differentiated. In a recent review, Hescheler et al. (7) pointed out that pluripotent embryonic stem (ES) cells cultivated within embryonic bodies reproduced highly specialized phenotypes of cardiac tissue. Most of the biological and pharmacological properties of cardiac-specific ion currents were expressed in cardiomyocytes developed in vitro from pluripotent ES cells (7, 9). Recently, Etzion et al. (6) transplanted cardiomyocytes isolated from 15-day-old embryos into rat MI hearts and found that cell-transplanted MI animals had attenuated left ventricular (LV) dilatation, infarct thinning, and myocardial dysfunction. However, the significance of ES cell transplantation in postinfarcted failing hearts remains to be determined. This study was designed to determine whether transplanted ES cells could survive in injured myocardium and improve cardiac function in postinfarcted rats.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Preparation of ES cells. The mouse ES cell line, ES-D3, was obtained from the American Type Culture Collection (Manassas, VA) and maintained with methods previously described (27). Briefly, ES-D3 cells were cultured in DMEM on mitotically inactive mouse embryonic fibroblast feeder cells (American Type Culture Collection). The medium was supplemented with 15% fetal bovine serum, 0.1 mM -mercaptoethanol (Sigma Chemical, St. Louis, MO), and 10³ units/ml of leukemia inhibitory factor conditioned medium (BRL, Gaithersburg, MD) to suppress differentiation. To initiate differentiation, ES cells were dispersed with trypsin and resuspended in the medium without supplemental leukemia inhibitory factor and cultured with the hanging drops (approximate 400 cells per 20 µl) method for 5 days. They were then seeded into 100-mm cell culture dishes. Spontaneously beating clusters were dissected by use of a sterile micropipette and recultured for another 2-3 days at 37°C in a humidified atmosphere with 5% CO₂.

View larger version (10K): [in this window] [in a new window]   Fig. 1.   Action potentials in cultured embryonic stem (ES) cells and neonatal mouse cardiomyocytes. The action potentials were recorded from a spontaneously beating ES cell (A), which was cultured 9 days after withdrawal of leukemia inhibitory factor from the medium. The whole cell zero-current clamp technique was applied to record spontaneous action potentials. The shape of the action potentials is very similar to that in mammalian neonatal cardiomyocytes (B). The spontaneously beating rate of this cell was 86 beats/min at room temperature.

ES cell transplantation. The experiments were performed in male Wistar rats (Charles River, Wilmington, MA) with an initial body weight of ~250 g. The investigation conformed to the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH Publication No. 85-23, revised 1996), and the protocol was approved by the Institutional Animal Care Committee. MI was induced by ligation of the left anterior descending coronary artery under anesthesia with pentobarbital sodium (60 mg/kg ip). The method to create a MI model in rats was described previously (20). Cell transplantation was performed within 30 min after induction of MI. ES cell suspension (30 µl) was injected into three sites, one within the infarct area and two in the myocardium bordering the ischemic area. Each injection was 10 µl of the medium containing beating ES cells (10⁴ cells). Medium-treated animals received the same MI operation but were only injected with the equivalent volume of the cell-free medium. The sham group underwent an identical surgery with neither ligation of the coronary artery nor cell transplantation.

Measurements of hemodynamics and isometric contraction. Hemodynamics were measured before (baseline) and 6 wk after MI induction. Rats were anesthetized with pentobarbital sodium (60 mg/kg ip). A carotid artery was isolated and cannulated with a 3-Fr high-fidelity microtip catheter connected to a pressure transducer (Millar Instruments, Houston, TX). The Millar Mikro-Tip catheter was advanced into the left ventricle to record ventricular pressure for a brief period of time. LV systolic and end-diastolic pressures, the maximum rate of LV systolic pressure rise (+dP/dt_max), mean arterial pressure, and heart rate were monitored and recorded on a chart-strip recorder (Gould Series 2000). Analog signals were digitized by use of a data translation series (model DI-220) analog-digital converter (Data Instruments, Akron, OH) and then stored and analyzed on a Windaq data-acquisition system.

7

6

5

4

Echocardiographic studies. Six weeks after MI or sham operation, rats were anesthetized with pentobarbital sodium. The echocardiographic procedure was performed by a method as described previously (18, 19). A commercially available echocardiographic system (Agilgent Sonos 5500) was used for all studies. LV mass was calculated using a standard cube formula. Relative anterior wall thickness, relative posterior wall thickness, and LV internal dimensions were measured from at least three consecutive cardiac cycles. We also used endocardial fractional shortening and midwall fractional shortening as indexes to estimate LV systolic function. The results of M-mode tracings were analyzed from the data recorded on an optical disk.

Measurement of infarct size. The MI was restricted to the left ventricle and occurred in all operated rat hearts. Infarct sizes were quantified according to the method described previously (12). In brief, several 2- to 3-mm-thick transverse sections were made consecutively in the left ventricle and ventricular sections were gently pressed flat. The epicardial and endocardial circumferences of the infarct area and entire flattened left ventricle were outlined on a transparent sheet. Infarct size (%) was calculated from the ratio of the surface area of infarct wall and the entire surface area of the left ventricle.

Identification of transplanted cells and histological studies. Subsets of animals were killed at 6 wk post-MI, and their hearts were quickly removed for single cell isolation and histological determination. The method for cell isolation was described in detail elsewhere (31). The hearts for pathological study were transversely sectioned. Selected sections from the free wall of the left ventricle, including infarct and peri-infarct regions, were embedded in tissue-freezing medium (Fisher Scientific, Fair Lawn, NJ). Frozen sections (10 µm in thickness) of LV tissues were made and stained with hematoxylin and eosin. The hearts from a few animals were also fixed with 10% formalin overnight, paraffin embedded, sectioned at 5 µm thickness, and stained with hematoxylin and eosin. Survival of engrafted cells was confirmed by identification of GFP expression under fluorescent microscopy. To identify differentiated myocytes from engrafted ES cells, we used an immunofluorescent method to identify sarcomeric -actin muscle isoform (which is present in fetal cardiomyocytes, but not in normal adult myocytes), cardiac -myosin heavy chain (-MHC) and cardiac troponin I (cTnI). Frozen sections were fixed with acetone at 4°C for 10 min and then incubated with a monoclonal anti--actin antibody (Sigma Chemical) for 45 min at room temperature. Sections were washed three times in PBS and incubated with Cy3-conjugated goat anti-mouse IgG antibody (Sigma Chemical) for 45 min. After extensive washing in PBS, slides prepared for -actin staining were mounted with DAPI/Antifade (Oncor, Gaithersburg, MD). Nonspecific binding was blocked by incubation with 1% bovine serum in remaining sections. Different frozen sections were stained immunohistochemically with a mouse monoclonal anti-GFP antibody (Zymed, San Francisco, CA), a goat polyclonal IgG anti-cTnI antibody (Santa Cruz Biotechnology, Santa Cruz, CA), or a mouse anti--MHC monoclonal antibody (Berkeley Antibody, Richmond, CA) for 60 min. After washing with PBS, sections were incubated with a rabbit anti-goat conjugated rhodamine IgG (for cTnI) or a goat anti-mouse conjugated fluorescein IgG (for -MHC and GFP) antibody (Pierce Chemical, Rockford, IL). Finally, fluorescent staining for -actin, -MHC, and cTnI was detected and photographed under fluorescent microscopy.

Data analysis. All values are presented as means ± SD. Data derived from three groups or more were evaluated by ANOVA with repeated measurements. Difference between two groups was compared by unpaired Student's t-test. A level of P < 0.05 was considered as significant difference.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Improvement of LV function after ES cell transplantation. Six weeks after infarction, the LV weight and ratio of LV weight to body weight were significantly increased in MI rats (P < 0.05 vs. the sham-operated group, Table 1). Intramyocardial transplantation of ES cells in MI rats significantly attenuated the severity of LV hypertrophy (Table 1). The area of infarcted myocardium was reduced from 40 ± 2% for MI rats injected with the cell-free medium to 35 ± 1% for MI rats transplanted with ES cells (P < 0.05).

View larger version (13K): [in this window] [in a new window]   Fig. 2.   ES cell transplantation improved left ventricular (LV) function [LV pressure (LVP) and peak LV systolic pressure (+dP/dt)] in post-myocardial infarction (MI) rats. Continuous chart strip recordings of hemodynamic measurement in anesthetized animals [sham-operated rats (A), postinfarcted rats injected with cell-free medium (B), and postinfarcted rats transplanted with ES cells (C)]. Measurement was conducted in rats 6 wk after the MI operation.

Improvement of isometric contractility in papillary muscle after ES cell transplantation. At baseline, papillary muscles isolated from MI rats injected with the cell-free medium showed a significant decrease in developed tension (Fig. 3). In animals with intramyocardial injection of ES cells, developed tension appeared to be significantly preserved. Elevation of extracellular Ca²⁺ levels increased developed tension of papillary muscles isolated from all three groups of rats. The increase in developed tension was concentration dependent. However, the concentration-response curve of developed tension in MI rats injected with the cell-free medium was shifted downward significantly (Fig. 3). -Adrenergic stimulation with cumulative concentrations of isoproterenol induced a pronounced increase in developed tension in papillary muscles isolated from sham-operated rats (Fig. 3). In contrast, papillary muscles isolated from MI rats injected with the cell-free medium had no positive inotropic response to isoproterenol stimulation. It is surprising that ES cell transplantation significantly partially restored the inotropic response to isoproterenol stimulation (Fig. 3).

View larger version (21K): [in this window] [in a new window]   Fig. 3.   Inotropic response of papillary muscles to Ca2+ and isoproterenol (Iso) stimulation after ES cell transplantation. Original representative recordings show inotropic responsiveness during isoproterenol stimulation in papillary muscles isolated from a sham-operated rat (A), a postinfarcted rat injected with cell-free medium (B), and a postinfarcted rat transplanted with ES cells (C). Experiments were carried out in animals 6 wk after the MI operation. The average data show the effects of extracellular Ca2+ levels (D) and isoproterenol (E) on developed tension (DT) produced by the stimulated muscle of isolated papillary muscles. Sham, sham-operated group (n = 6); MI + Medium, MI group injected with cell-free medium (n = 6); MI + ES Cell, MI group transplanted with ES cells (n = 7). *P < 0.05, **P < 0.01 vs. Sham; #P < 0.05 vs. MI + Medium.

Identification of transplanted ES cells. Frozen sections prepared from MI areas after 6 wk of cell transplantation showed GFP-positive spots under fluorescent microscopy (Fig. 4A). In contrast, sections from sham-operated hearts or MI control areas had no such GFP-positive tissue. In addition, single GFP-positive cells were detected under fluorescent microscopy in cells isolated from MI hearts 6 wk after transplantation (Fig. 4B). The rod-shaped GFP-positive cells had clear striations, which are characteristic of adult cardiomyocytes (Fig. 4, B and C). The isolated host cardiomyocytes were GFP negative (Fig. 4, B and C). In addition, cells isolated from the sham-operated or MI control hearts were unable to detect GFP-positive cells. In cell-transplanted animals, the average GFP-positive cells were 1,158,574 ± 100,894 (n = 5) out of a total 15,949,206 ± 406,249 myocytes isolated from each MI left ventricle. The percentage of GFP-positive cells was 7.3%. According to our calculation in culture, the efficiency of transfection of GFP was ~80-90% in cultured ES cells. Although we did not measure the exact size of GFP-positive cells, we found that the shape and size of mature GFP-positive myocytes did not significantly differ from those of host cardiomyocytes under a microscope (Fig. 4, B and C). However, the small round cell with GFP staining in Fig. 4B might be an immature cardiac cell, as reported by others (6). These results suggest that implanted ES cells not only survived in injured myocardium but also were able to differentiate into mature cardiomyocytes.

View larger version (46K): [in this window] [in a new window]   Fig. 4.   Presence of GFP-positive cells in injured myocardium and in isolated single cardiomyocytes. A: the frozen section was prepared from a MI heart 6 wk after ES cell transplantation. GFP-positive spots were detected under fluorescent microscopy (×200). B: under fluorescent microscopy (×200), single GFP-positive cells with rod shape and striations were detected in cells isolated from a post-MI heart 6 wk after ES cell transplantation. C: this panel corresponds to B. Isolated single cardiomyocytes from host myocardium were GFP negative with rod shaped and clear striations under light-contrast microscopy (×200). Insets in B and C are further enlarged from the areas pointed to by the arrows.

Histology and immunostaining of infarcted myocardium. Hematoxylin-eosin staining shows fibrosis in the infarction area in MI control hearts (Fig. 5, B and E) injected with the cell-free medium 6 wk after the MI operation. In ES cell-transplanted MI rats, characteristic phenotype of engrafted cells without evidence of immunorejection was detected in infarct areas (Fig. 5, C and F). Immunostaining confirmed that engrafted cells were distinct from host cardiomyocytes and infarct tissue. Implanted cells stained positively with the monoclonal antibody of antisarcomeric -actin (data not shown). However, the anti--actin staining was negative in sham-operated and control MI hearts. In addition, engrafted cells stained positively to cardiac -MHC in cell-transplanted myocardium (Fig. 6C). Compared with MI hearts injected with cell-free medium (Fig. 6B), the sections from a MI cell-transplanted heart show much higher intensity of immunostaining for -MHC. To further confirm differentiation of engrafted cells into cardiomyocytes, we did double staining for GFP and cTnI in cell-transplanted myocardial sections. Figure 7 shows GFP- and cTnI-positive staining and their overlaps in numerous areas. These results confirm the survival and differentiation of implanted ES cells in injured myocardium.

View larger version (81K): [in this window] [in a new window]   Fig. 5.   Photomicrograms of transverse sections from rat left ventricles at the level of the posterior papillary muscle. Endocardium is on the left and epicardium is on the right for each image. Images were taken at low (A-C, ×40) or high magnification (D-F, ×200). A and D are from a rat with sham operation and show normal architecture and histology of myocardium. B and E are from a rat surviving acute MI 6 wk after ligation of the left anterior descending coronary artery and intramyocardial injection with the cell-free medium. The submyocardium was largely replaced by scar tissue, and there was granulation formation on the pericardium. The subepicardium was partially preserved. C and F are from a MI rat 6 wk after transplantation of ES cells. There was disruption of the myocardial architecture, and the subendocardial scar tissue was largely replaced by immature, hypertrophic cardiomyocytes, which might have differentiated from the implanted cells.

View larger version (77K): [in this window] [in a new window]   Fig. 6.   Immunostaining for cardiac -myosin heavy chain. The -myosin heavy chain staining of the myocardial sections is shown for a sham-operated rat heart (A), a MI heart with injection of the cell-free medium (B), and a MI heart with intramyocardial transplantation of ES cells (C).

View larger version (51K): [in this window] [in a new window]   Fig. 7.   Double staining for green fluorescent protein (GFP) and cardiac TnI (cTnI) of the myocardium transplanted with ES cells. A and B: staining of GFP by a monoclonal anti-GFP antibody and of cTnI by a polyclonal antibody, respectively. Overlaps of GFP and cTnI staining in C demonstrate that engrafted GFP-positive cells differentiated into cardiac myocytes. Magnification is ×200.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The main findings of the present study are that 1) embryonic stem cells can be implanted and survive in injured rat myocardium and 2) transplantation of ES cells improves global cardiac function and myocardial contractility.

Several studies have demonstrated the feasibility of engrafting exogenous cells into host myocardium, including fetal cardiomyocytes (26, 28), atrial tumor (11), satellite cells (3), or bone marrow cells (29). These engrafted cells have been histologically identified in normal myocardium up to 4 mo after transplantation (3). Gap junctions have been found between the engrafted and host myocardium (8, 10, 28). Recently, myocyte transplantation has been extended into ischemically damaged myocardium with coronary artery occlusion in rats (29) or with cryoinjury in rats (15, 16) and dogs (3). ES cells are pluripotent cells derived from the early embryo and retain the ability to differentiate into all cell types, including cardiomyocytes (24, 25). One of the advantages of using ES cells is to reduce immunoreactivity, because ES cells express less immune-related cell-surface proteins (22). This study provides the evidence for survival of ES cells in injured myocardium after transplantation. Further, the isolated single GFP-positive cells were rod-shaped with clear striations. These characteristics of GFP-positive cells indicate that implanted ES cells, at least part of them, not only survived in injured myocardium, but also differentiated into mature cardiomyocytes after 6 wk of cell transplantation. Therefore, the improvement of ventricular function may result, at least partially, from cardiogenesis of implanted ES cells. This result is consistent with the recent findings of regeneration of infarcted myocardium in mice with transplantation of bone marrow cells (21, 29).

Large MI induced by permanent ligation of the left anterior descending coronary artery results in remarkable impairment of cardiac function (23). We found that infarct size was reduced in MI rats implanted with ES cells. In addition, ES cell transplantation significantly improved LV function and isometric contractility in MI rats. One possibility for the improvement of cardiac function is a reduction of infarcted area by regeneration of myocardium after transplantation of ES cells. Reduction of the infarct size could prevent overstretching of the ventricle and preserve normal contractile function (Frank-Starling law). This is consistent with a previous report that reduction of chamber size improved heart performance (15, 16). In addition, myocardial regeneration by ES cells may improve global function of infarcted hearts, which provides beneficial effects on papillary muscle contractility. In contrast, papillary muscle contractility was decreased in untreated MI rat hearts, because enlarged failing hearts might overstretch and damage papillary muscles. Our morphological data confirm that engrafted ES cells survived in infarcted myocardium by identification of GFP-positive cells in cell-implanted hearts 6 wk after cell transplantation. GFP-positive cells isolated from post-MI hearts with ES cell transplantation were rod-shaped with clear striations that mimicked adult cardiomyocytes. The viability of engrafted cells was also identified by positive stains to sarcomeric -actin, which is rarely seen in adult rat myocardium (13). Furthermore, engrafted cells stained positively to -MHC and double-stained to GFP and cTnI. In contrast, staining for -actin was negative, and the intensity of MHC staining was lower in MI myocardium without ES cell transplantation. Calculation of the number of single GFP-positive cells from total isolated LV myocytes indicates that engrafted cells accounted for 7.3% of LV myocardium 6 wk after MI induction and cell transplantation. These data strongly suggest that cardiogenesis occurred in the infarcted myocardium after ES cell transplantation. Regeneration of myocardium and improvement of cardiac function also occurred in MI mice after intramyocardial implantation of bone marrow cells (21). Approximately 68% of the myocardium in the infarcted portion of the ventricle was newly formed in MI mice 9 days after cell transplantation (21). In addition, Beltrami et al. (1) found that there was myocyte proliferation in human heart after MI and that regeneration of myocytes may contribute to the increase in muscle mass of the myocardium. Therefore, myocardial regeneration by implanted stem cells may play a primary role in the improvement of ventricular function of infarcted failing hearts.

Another possible explanation of the beneficial effects of ES cell transplantation is that ES cells serve as platforms for the release of cardioprotective factors such as vascular endothelial growth factors. In normal porcine hearts, Van Meter et al. (30) showed that transplantation of either human atrial cardiomyocytes or fetal human ventricular cardiomyocytes can induce nascent blood vessel formation in grafted areas and the host ventricle. The increase in microcirculation provides the grafted cells with blood supply and is also an avenue for removal of cellular debris due to primary injury. Subsequently, attenuation of infarct size was also found in the present study. More recently, Tomita et al. (29) found that the number of capillaries in bone marrow cell-transplanted animals was significantly larger than that of the control. Therefore, the reduction of infarct size in MI rats transplanted with ES cells in our experiments may partially result from an improvement of blood supply. If the ischemic zone was reperfused, additional growth factors may reach other regions of the heart. The functional benefits of ES cell transplantation on papillary muscle contractility may relate to the release of growth factors whose positive effects on myocardial contractility have been reported by our laboratory (4) and others (14). Thus the improvement of ventricular function in postinfarcted hearts with ES cell transplantation may result from an increase in the pool of cardiomyocytes and from the paracrine effects of engrafted cells which facilitate the repair of injured cardiac tissue.

In conclusion, this study demonstrates the feasibility of transplanting ES cells into injured myocardium in rats. Transplanted ES cells were able to form stable intramyocardial grafts and to improve cardiac function in postinfarcted failing hearts. Our results raise the possibility that ES cell transplantation may provide a new and novel approach to improve cardiac function after a massive MI.