This study assessed the potential therapeutic efficacy of adipose-derived stem cells (ASCs) on infarcted hearts. Myocardial infarction was induced in rat hearts by occlusion of the left anterior descending artery (LAD). One week after LAD occlusion, the rats were divided into three groups and subjected to transplantation of ASCs or transplantation of cell culture medium (CCM) or remained untreated. During a 1-mo recovery period, magnetic resonance imaging showed that the ASC-treated hearts had a significantly greater left ventricular (LV) ejection fraction and LV wall thickening than did the CCM-treated and untreated hearts. The capillary density in infarct border zone was significantly higher in the ASC-treated hearts than in the CCM-treated and untreated hearts. However, only 0.5% of the ASCs recovered from the ASC-treated hearts were stained positive for cardiac-specific fibril proteins. It was also found that ASCs under a normal culture condition secreted three cardiac protective growth factors: vascular endothelial growth factor, hepatocyte growth factor, and insulin-like growth factor-1. Results of this study suggest that ASCs were able to improve cardiac function of infarcted rat hearts. Paracrine effect may be the mechanism underlying the improved cardiac function and increased capillary density.
- myocardial infarction
- magnetic resonance
- cardiogenic transdifferentiation
a cumulative loss of the cardiomyocytes is the fundamental reason for heart failure in ischemic heart disease (29). Replacement of infarcted myocardium with new functional cardiomyocytes and vessel-constituting cells has been proposed as a potential therapy for myocardial infarction and heart failure (1, 24). Adult-derived stem cells may hold promise for the treatment of heart failure since they can be used for autologous transplantation (5). Several types of adult stem cells such as bone marrow stem cells (BMSCs), skeletal myoblasts, and cardiac stem cells have been studied extensively for their capacity of cardiac repair (7, 18, 20, 21, 34). A number of animal studies have demonstrated therapeutic benefits of the adult stem cells for myocardial infarction (5, 7, 18, 20, 21, 34). Human studies, on the other hand, have generated very mixed results (26, 29, 33). Thus investigators continue to search for more curative types of adult stem cells.
Adipose tissue consists of mature adipocytes and a stromal vascular fraction. The latter is a heterogeneous cell population containing endothelial cells, smooth muscle cells, blood cells, and fibroblast-like cells (8, 37). The latter have the potential to differentiate into various mesodermic lineages and are referred to as adipose-derived stem cells (ASCs) (8, 37). Adipose tissue has several advantages over other tissues as a source of adult stem cells. First, adipose tissue is abundant in most of the population that have ischemic heart disease and can be harvested repetitively from the same subjects with minimum morbidity. Adipose tissue contains a high density of ASCs with approximately a half million ASCs per one milliliter of lipoaspirate (2). Thus a clinically relevant dose of ASCs may be extracted from a small amount of subcutaneous adipose tissue, potentially eliminating the need for ex vivo expansion. Second, ASCs have a potential to differentiate into several cell lineages (8, 16, 31, 40). These desirable characteristics of ASCs demonstrate its utility as a good cell candidate for cell therapy of myocardial infarction and heart failure. There are, however, very few studies that have focused on the efficacy of ASCs for the improvement of heart function and the reduction of cardiac remodeling (23, 25, 28, 39). Our study was designed to evaluate the capacity of ASCs in the prevention of heart failure.
MATERIAL AND METHODS
The animals used in this study received humane care in compliance with the Guide to the Care and Use of Experimental Animals formulated by the Canadian Council on Animal Care. The Animal Care Committee of the Institute for Biodiagnostics approved the experimental protocols of this study.
Preparation of ASCs.
ASCs were prepared according to the method developed by Zuk and colleagues (40) with some modifications. In brief, the subcutaneous adipose tissue was obtained from the abdominal and inguinal regions of 10 inbred male Lewis rats. The isolated adipose tissue was washed extensively with phosphate-buffered saline (PBS) to remove contaminating debris and blood cells. The adipose tissue was minced and digested with collagenase I (2 mg/ml; Worthington Biochemical, Lakewood, NJ) at 37°C for 20–30 minutes. Collagenase activity was neutralized by DMEM-F12 (HyClone, Logan, UT) containing 15% fetal bovine serum (FBS; HyClone, Logan, UT). Digested adipose tissue was filtered twice with a 100-μm and then with a 25-μm nylon membrane to eliminate the undigested fragments. The cellular suspension was centrifuged at 1,000 g for 10 min. The cell pellets were resuspended in cell culture medium (CCM) and cultivated for 24 h at 37°C in 5% CO2. Unattached cells and debris were removed, and fresh CCM containing 15% FBS was added to the adherent cells, which were cultured at 37°C in 5% CO2 until 70–80% confluent. Identity of the isolated cells was determined by assessing their surface markers using flow-cytometric-analysis with FITC-conjugated antibodies and FITC-conjugated isotype-matched control antibodies (Biolegend, San Diego, CA). The cells used for cytometric analysis were at passage 2.
Lentiviral vectors encoding a gene for emerald green fluorescent protein (eGFP) were produced by cotransfection of 293FT cells with plasmids pLenti6.2-GW/eGFP, pLP1, pLP2, and pLP/VSVG with lipofectamine 2000 (Invitrogen, Carlsbad, CA). When the ASCs were at ∼60% confluence, the lenti-eGFP vectors were added to ASC dishes at 10–20 multiples of infection. Viral infection was carried out at 37°C and 5% CO2 for 2 h. It was found that most of the ASCs (>98%) expressed eGFP immediately after the transfection. To track the ASCs in vivo with magnetic resonance (MR) imaging, the eGFP-labeled ASCs were incubated for 2 days in a CCM containing 50 μg/ml superparamagnetic iron oxide (SPIO) nanoparticles (Feridex; Bayer) and 6 μg/ml protamine sulfate. The latter was used as a transfecting agent.
On the day of cell transplantation, the ASCs were trypsinized and the detached cells were then centrifuged. The supernatant was removed, and FBS-free medium was added to the cell pellet. The cell suspension was filtered with a 25-μm sieve to eliminate the clumped cells. Before injection, the cell preparation was examined under a microscope to ensure no evident cell clumping in the ASC preparation. The ASCs used for transplantation were within five passages.
Induction of transdifferentiation of ASCs in vitro.
To ensure that the isolated cells were ASCs and to estimate the purity of the ASC preparation, the isolated cells were subjected to four types of induction. Adipogenic transdifferentiation was induced by incubation of the cells for 12 days in DMEM containing 5% FBS, 33 mM biotin, 17 mM pantothenate, 10 μM bovine insulin, 250 isobutyl-methylxanthine, 200 μM indomethacin, 1 μM dexamethasone, 5 μg/ml streptomycin, and 5 U/ml penicillin. Osteogenic induction was carried out by cultivating the ASCs for 28 days in DMEM containing 10% FBS, 10 mM β-glycerophosphate, 50 μM ascorbate-2-phosphate, 10 nM 1,25(OH)2 vitamin D3, 5 μg/ml streptomycin, and 5 U/ml penicillin. Myogenic induction was conducted by 28-day incubation in DMEM containing 8% FBS, 7% horse serum, 50 μM hydrocortisone, 5 μg/ml streptomycin, and 5 U/ml penicillin.
Cardiogenic induction of ASCs was conducted by coculture of ASCs with the neonatal cardiomyocytes in an indirect cell-cell contact system. Briefly, freshly isolated neonatal cardiomyocytes were seeded on the bottom of a 6-well plate with a final volume of 3 ml D/F12 (10% FBS in DMEM + F12 in 1:1 ratio). ASCs were seeded on a coverslip that was placed on the upper surface of Transwell membrane insert with a pore size of 0.4 μm (Corning-Costar, Acton, MA). The Transwell inserts were then placed into the 6-well plate and allowed to submerge in the CCM. The cocultures were incubated for 7, 14, and 21 days, respectively. At the end of the three coculture periods, ASCs were collected for RT-PCR analysis of cardiac-specific markers.
Expression of the genes specific for the four types of transdifferentiation was determined using RT-PCR. Total mRNA from the induced and control (noninduced) ASCs was extracted using the TRIzol Reagent (Invitrogen, Carlsbad, CA) protocol. One microgram of mRNA was reversely transcribed using SuperScript III reverse transcriptase (Invitrogen).
The specific genes for assessment of adipogenic differentiation were lipoprotein lipase (LPL) and peroxisome proliferator-activated receptor-γ (PPAR-γ). The genes for osteogenic transdifferentiation were bone sialoprotein (BSP) and osterix (OSX). Myogenin and smooth muscle α-actin (α-SMA) were markers for skeletal myogenic transdifferentiation. GATA4, nkx2.5, and myosin light chain 2v (MLC-2v) were three marker genes for assessment of cardiogenic transdifferentiation. GAPDH was used as an internal control for RNA extraction and RT-PCR assay.
Expression of VEGF, hepatocyte growth factor (HGF), and IGF-I by ASCs was also determined with RT-PCR. Sequences for all the primers used in this study are detailed in Table 1.
Animal model and experimental protocol.
Inbred female Lewis rats with an average body weight of 200 g underwent an open-chest surgery to permanently occlude the left anterior descending coronary artery (LAD). The surgery was performed under isoflurane anesthesia during which respiration was controlled with a rodent ventilator at a rate of 60–70 breaths/min and tidal volume of 2 to 3 ml. A left anterior thoracotomy was made through the fourth intercostal space. The LAD was permanently occluded at ∼2 mm from its origin using a 6-0 silk suture. The chest was then closed.
One week after LAD occlusion, the animals were randomly divided into three groups (ASC treated, n = 11; CCM treated, n = 10; and untreated, n = 10). The animal's chest was reopened under the same anesthetic conditions described above. In the ASC-treated group, four injections of ∼1.25 ×106 eGFP/SPIO-labeled ASCs in 10 μl with total volume of 40 μl (4 × 10 μl) were made into the infarct border using a 30-gauge needle. Rats in the CCM-treated group received four injections of 10-μl CCM in the same regions. Rats in the untreated group underwent a repeat thoracotomy only. All animals were allowed to recover for 4 wk.
All rats underwent MR imaging for assessment of cardiac function at 1 and 4 wk after the second surgery (cell and medium transplantation). At end of the 4-wk recovery period, the animals were euthanized and the hearts were excised.
Control values of cardiac function were obtained from six healthy rats at an equivalent body weight and age.
Cardiac cine MR imaging.
Cardiac MR imaging was used to track the implanted ASCs and to assess their effects on cardiac function and structure. Anesthetized rats were positioned prone in a cradle. Standard limb leads were constructed from electrodes placed on both forepaws and the left hind paw. The cradle was inserted into an in-house manufactured quadrature MR coil. The coil was positioned in the center of the horizontal bore of a 7 Tesla Bruker magnet interfaced to Bruker Biospec console (Bruker, Karlsruhe, Germany). MR imaging was triggered with both respiratory and ECG signals.
Cardiac cine imaging was performed using a gradient-echo sequence with a repetition time of 9.2 ms, echo time of 3.5 ms, field of view of 8 × 8 cm2, and matrix size of 256 × 256. The cine images were acquired from five consecutive slices along the short cardiac axis with a slice thickness of 2.0 mm and no gap between the slices. Depending on the heart rate, 15–18 frames were typically fit into a cardiac cycle. With 4-signal averages, the total acquisition time of the cine images was ∼43 min.
At end of the protocol, the hearts were excised from the animals under anesthesia. Six ASC-treated, 10 CCM-treated, and 10 untreated hearts were transversely cryosectioned into 6-μm thick slices from the apex to the base. Prussian blue, which stains SPIO blue, was used to localize the SPIO nanoparticles. A fluorescence microscope was used to monitor eGFP-positive (eGFPpos) cells.
To assess cardiogenic transdifferentiation of the engrafted ASCs, the tissue sections of the ASC-treated rat hearts were first incubated for 1 h in media containing one of two primary antibodies: mouse anti-α-myosin heavy chain (α-MHC; ABR-Affinity BioReagents, Golden, CO) and rabbit anti-α-sarcomeric actinin (Sigma, St. Louis, MO). The tissue sections were then rinsed three times with PBS, followed by 1-h incubation in the secondary antibody Alexa-Fluor594-conjugated goat anti-mouse/rabbit IgG (Invitrogen Canada, Burlington, ON, Canada). Cell nuclei were stained with 4',6-diamidino-2-phenylindole (DAPI; Sigma).
As a further step to assess cardiogenic transdifferentiation of the engrafted ASCs, five ASC-treated hearts were perfused in a Langendorff apparatus with Krebs-Henseleit solution containing 0.5% collagenase II (Worthington Biochemical, Freehold, NJ) for 20–30 min. The engrafted ASCs (eGFP-positive cells) were recovered with fluorescence-activated cell sorting technique. Expression of sarcomeric α-actinin and α-MHC in the recovered ASCs was then determined.
The number of capillaries in the infarct border zone (infarcted region adjacent to the viable myocardium) was determined by staining the endothelial cells on the heart tissue sections with biotinylated Lectin (Sigma-Aldrich). A size criterion of 10 μm was used to exclude small arterioles and venules. For the obliquely sectioned capillaries, the shortest diameter was used as the selection criterion. Three tissue sections from each animal were used for the measurement of capillary density at ×400 magnification. All measurements were done in a blinded fashion. Capillary density was expressed as the mean of capillaries per squared millimeter.
Some heart tissue sections were stained with Masson's Trichrome to delineate the infarct region. The percentage of viable myocardium in infarct region was estimated using ImageJ software (National Institutes of Health, Bethesda, MD).
Image processing and data analysis.
Cine MR images were analyzed using an image processing software Marevisi (Institute for Biodiagnostics, Winnipeg, Canada). In a set of cine images, the image with the smallest cavity was considered to represent the end systole (ES), whereas the one with the largest cavity was considered to reflect end diastole (ED). Left ventricular (LV) endocardial and epicardial contours were drawn semiautomatically on serial short-axis slices of both ED and ES images. Manual adjustments were made when required. LV wall thickness at ED (LVTED) and LV wall thickness at ES (LVTES) was measured in six segments: three in the infarct region and three in a remote region. LV wall thickening (LVth) was computed with the function: (LVTES − LVTED)/LVTED × 100%. LV area at ED (LVAED) and LV area at ES (LVAES) were measured. LV ejection fraction (LVEF) was calculated using the function [∑LVAED − ∑LVAES]/∑LVAED × 100%, where ∑LVAED and ∑LVAES are the sum of LV area in all slices measured at ED and ES, respectively.
Statistical analyses were performed using Statistica (Statsoft, Tulsa, OK). ANOVA was used to determine significant difference in contractile parameters between the groups of animals.
All numerical results are expressed as means ± SD. A value of P < 0.05 indicates a significant difference.
Surface marker profile and growth factor secretion of the rat ASCs.
The rat ASCs isolated from the subcutaneous adipose tissue expressed CD29, CD44, CD90.1, and CD49d and lacked CD34 and CD106 (Table 2). The characteristics of surface marker expression of the ASCs found in this study are in agreement with those previously reported by others (2, 10, 35).
In addition, it was found that under normal culture conditions ASCs expressed mRNA for three cardioprotective growth factors (VEGF, HGF, and IGF-I; Fig. 1).
Transdifferentiation potential of ASCs.
Under the adipogenic, myogenic, and osteogenic inductive conditions described above, the ASCs expressed mRNAs for LPL and PPAR-γ, α-SMA and myogein, and BSP and OSX, respectively (Fig. 2). In contrast, ASCs under control conditions (noninduction) did not express the marker genes (Fig. 2). Moreover, 78 ± 3% of the adipogenically induced ASCs were stained positive for lipid drops and 76 ± 1% of the osteogenically induced ASCs were positive for alkaline phosphatase (Fig. 3). Taken together with the surface marker profile (Table 1), the histochemical results suggest that more than 50% of the cells in our ASC preparation were adipose-derived stromal stem cells.
RT-PCR analysis showed that the ASCs started to express mRNA for GATA4, nkx2.5, and MLC-2v after 3 wk of the cocultivation with the neonatal cardiomyocytes (Fig. 4). However, there was no evident beating of the ASCs with or without electrical stimulation. This indicates that the ASCs expressing the cardiac marker genes were not definite cardiomyocytes.
Effects of ASCs on LV function.
Cine MR imaging showed that the average LVEF of the ASC-treated rats was significantly (P < 0.05) higher than that of the CCM-treated and untreated rats at both 1 (54.1 ± 4.2% vs. 47.8 ± 4.7% and 46.4 ± 2.3%) and 4 wk (55.8 ± 5.4% vs. 45.7 ± 6.1% and 44.4% ± 1.4%) after cell transplantation (Fig. 5). However, LVEF of the ASC-treated animals was still significantly lower than that (67 ± 3%) of the healthy control rat hearts under general anesthesia (Fig. 5), indicating that ASCs under our experimental conditions could not completely restore global function of the infarct hearts. The functional results demonstrate that ASCs could at least prevent infarction-associated functional deterioration.
Figure 6 shows three stacks of representative MR images acquired at the end of the 4-wk recovery period. The ASC-treated hearts had preservation of the LV wall thickness. In contrast, the CCM-treated and untreated hearts displayed evident thinning of LV anterior wall. These thinning areas in the CCM-treated and untreated hearts did not exhibit systolic wall thickening during a cardiac cycle.
Furthermore, at 1 wk after cell transplantation, LVTED and LVTES in the infarct region were slightly (P > 0.05) greater in the ASC-treated rats than in the CCM-treated rats and untreated hearts (Table 3). LVth in the infarct region, on the other hand, was significantly (P < 0.05) greater in the ASC-treated hearts (43.4 ± 8.4%) than in the CCM-treated (18.4 ± 4.1%) and untreated hearts (18.2 ± 3.7%; Table 3), suggesting that improvement in regional function in the ASC-treated hearts may not be related to regeneration of the myocardium. At 4 wk after cell transplantation, LVTED and LVTES were significantly (P < 0.05) greater in the ASC-treated rats than in the CCM-treated rats and untreated hearts (Table 3). Likewise, LVth was statistically greater in the ASC-treated hearts (39.8 ± 3.8%) than in the CCM-treated hearts (12.8 ± 0.5%). LVth in remote region among the three groups was comparable at both 1 and 4 wk after cell transplantation (Table 3). No significant differences in LVTED, LVTES, and LVth were detected between the CCM-treated and untreated hearts (Table 3).
LVTED and LVTES of the six healthy rats were 2.7 ± 0.2 mm and 1.8 ± 0.1 mm, respectively, giving rise to LVth of 33 ± 4% (Table 3), which was not significantly different from those of the remote myocardium in the three groups of hearts.
Potential cardiogenic transdifferentiation of ASCs.
The tissue sections obtained from six ASC-treated hearts showed a number of eGFPpos cells in the infarct region (Fig. 7). However, very few (∼0.5%) eGFPpos cells were stained positive for cardiac-specific fibril proteins, sarcomeric α-actinin, and α-MHC (Fig. 7). Moreover, the ASCs recovered from the five ASC-treated rat hearts at the end of recovery showed a similar positive rate for the cardiac specific markers (Fig. 8).
Effect of ASCs on capillary density and myocardial remodeling.
Capillary density in the infarct border zone was significantly (P < 0.05) higher in the ASC-treated rats (4,571 ± 162 capillaries/mm2) than in the CCM-treated rats (3,183 ± 194 capillaries/mm2) and untreated hearts (3,088 ± 313 capillaries/mm2; Fig. 9). Figure 10 shows a significantly greater percentage of viable myocardium in infarct region in the ASC-treated hearts than in the CCM-treated and untreated hearts.
Imaging of the engrafted ASCs.
MR images of ASC-treated hearts showed one or more signal voids in the anterior wall of the LV (Fig. 11A). Heart sections taken from the ASC-treated animals showed a number of SPIO nanoparticles in the signal void regions (Fig. 11B), indicating that the signal voids on MR images were the result of the SPIO nanoparticles. Under a fluorescence microscope, the same tissue sections displayed many green spots, presumably being eGFPpos ASCs (Fig. 11C). Overlaying the two pictures, it was found that the SPIO nanoparticles overlapped with the green dots. In all the tissue sections analyzed, almost 100% of the SPIO nanoparticles were overlapped with green dots, suggesting that the signal voids on the MR images came from the eGFP-expressing cells (Fig. 11D). We also found that ∼5% of green dots were not overlapped with SPIO nanoparticles. Taken together, the imaging results suggest that ratio of eGFP to SPIO overlaying is about 95%.
The in vitro experiments in this study showed that the ASCs under control culture condition expressed three cardiac-protective growth factors: VEGF, HGF, and IGF-I. The ASCs expressed cardiac marker genes (GATA4, nkx2.5, and MLC-2v) following 21 days of coculture with the neonatal cardiomyocytes. Our in vivo experiments showed that injection of ASCs into the infarct border zone prevented deterioration of global (LVEF) and regional (LVth) function of the infarcted rat hearts. Transplantation of ASCs also led to a significant increase in capillary density in the infarct border zone. However, only 0.5% of the engrafted ASCs were stained positive for sarcomeric α-actinin and α-MHC. Finally, the engrafted ASCs could be monitored using SPIO-enhanced MR imaging for at least 4 wk after cell transplantation.
ASCs prevented functional deterioration of infarcted rat hearts.
In this study, intramyocardial injection of ASCs was performed at 1 wk after LAD occlusion and found to significantly improve both regional and global cardiac function, suggesting that ASCs provided an effective therapeutic strategy for heart failure in subacute myocardial infarction. The degree of functional improvement observed in this study was comparable with those observed with BMSCs (3) and cardiac stem cells (CSCs) (7), indicating that ASCs are at least equally effective for cardiac repair relative to BMSCs and CSCs. In addition, Mazo and colleagues (25) found that intramyocardial injection of ASCs at 1 mo after LAD occlusion also resulted in a significant improvement in cardiac function in comparison with control. Taken together, the studies suggest that ASCs could be an effective therapy for heart failure in both subacute and chronic myocardial infarction.
Importantly, the improvement in LVEF and LVth were observed as early as 1 wk after ASC transplantation, suggesting that ASCs provided therapeutic benefits in a timely fashion. On the other hand, this rapid response in cardiac function suggests that the functional improvement was not related to cardiac transdifferentiation of the ASCs or myocardial regeneration.
Although the intergroup LVEF of the CCM-treated and untreated hearts measured at 1 and 4 wk of recovery period was not significantly different, the intragroup LVth measurements at the two time points was statistically (P < 0.05) different (18.4 ± 4.1% vs. 12.8 ± 0.5% in the CCM-treated hearts; and 18.2 ± 3.7% vs. 13.0 ± 0.6% in the untreated hearts; Table 3). This demonstrates that the CCM-treated and untreated hearts were continuously deteriorating during the recovery period, whereas the ASC-treated animals did not. Likewise, although LVEF of the ASC-treated hearts was significantly higher than those of the CCM-treated hearts and untreated hearts, LVEF of the ASC-treated hearts at 1 and 4 wk after cell injection was comparable (Fig. 5). This seems to suggest that ASCs mainly prevented functional deterioration.
In addition, LVTED, LVTES, and LVth in the remote areas among the ASC-treated, CCM-treated, and untreated hearts were not significantly different (Table 3), suggesting that the improvement in LVEF observed in the ASC-treated rats was not due to a compensatory increase of contractile function in the remote regions.
Potential mechanism for the improvement in contractile function.
Several mechanisms may be involved in the improvement of cardiac function observed in the ASC-treated rats, such as cardiogenic transdifferentiation and paracrine action. In this study, we injected ∼5 × 106 ASCs into the infarct rim. According to the rate of cell retention (<10%) reported by others, approximately a half million ASCs might be retained in the region of interest (12). Moreover, a significant percentage of the retained ASCs might die shortly after cell transplantation due to apoptosis or other mechanisms (15). The actual number of the ASCs that survived and engrafted in the infarct region might be very limited relative to the number of cardiomyocytes in a healthy heart (4 to 5 × 107 cardiomyocytes/gram of myocardium). Moreover, it was found that only 0.5% of eGFPpos cells were stained positive for the cardiac-specific markers. Even though the 0.5% of eGFPpos cells were fully cardiodifferentiated, they were not sufficient for the functional improvement observed in this study. As mentioned above, the functional improvement in the ASC-treated rats was observed 1 wk after cell transplantation. Evidently, 1 wk was not sufficiently long to allow full development of the contractile apparatus. In addition, our cell culture experiments showed that ASCs did not express cardiac marker genes (GATA4, nkx2.5, or MLC-2v) until 3 wk of coculture with the neonatal cardiomyocytes. These data indicate that the functional improvement observed in the ASC-treated rats cannot be attributed to cardiogenic transdifferentiation of the engrafted ASCs.
On the other hand, VEGF, HGF, and IGF-I have been suggested to play important roles in cardiac repair. VEGF can enhance angiogenesis, stem cell mobilization, and cardiomyocyte proliferation (17, 36). HGF is believed to be involved in regeneration of various tissues, including myocardium (13, 30). IGF-I is able to prevent apoptosis and to enhance cell proliferation by activating multiple signal transduction pathways (6, 11). Direct injections of plasmid DNA encoding each of the growth factors into the ischemic myocardium have been shown to increase capillary density, reduce infarct size, and improve cardiac contractile function. In agreement with others (27, 32), this study shows that ASCs under normal culture conditions expressed the three growth factors (Fig. 1). We also found that the infarct border zone was stained significantly stronger for VEGF in the ASC-treated hearts than in the CCM-treated and untreated hearts (data not shown). All these suggest that secretion of the growth factors may be responsible for the observed benefits, particularly angiogenesis, in the ASC-treated hearts.
Potential of cardiac transdifferentiation of ASCs.
As mentioned above, ASCs expressed some cardiac-specific genes when cocultured with the neonatal cardiomyocytes, suggesting that ASCs have a potential to take cardiomyocyte genotype. This result agrees with the findings of Gaustad et al. (14), who showed that human ASCs expressed cardiomyocyte marker genes following exposure to the extracts of rat cardiomyocytes. However, we did not observe any beating activity of the ASCs under the coculture conditions, indicating that the ASCs expressing cardiac marker genes were not definite cardiomyocytes yet. In addition, only a very limited percentage (∼0.5%) of eGFPpos cells on ASC-treated hearts were stained positive for sarcomeric α-actinin and α-MHC (Figs. 7 and 8). Similarly, Mazo and colleagues (25) did not find any engrafted ASCs stained positive for cardiac fibril proteins. All these suggest that ASCs may have a limited capacity of cardiac transdifferentiation. Moreover, it remains to be determined whether the positive staining for the cardiac fibril proteins was due to cardiac transdifferentiation or fusion of the engrafted ASCs with the native cardiomyocytes. Importantly, expression of some cardiac marker genes and positive staining of cardiac proteins do not indicate a definite cardiac transdifferentiation. The benefits observed in this study could not be mainly related to the potential cardiac trasndifferentiation of the engrafted ASCs.
Tracking of engrafted cells using MR imaging.
Due to the proapoptotic and cytotoxic microenvironment in infarct region, a significant percentage of eGFP/SPIO-labeled ASCs may die shortly after transplantation (9). The SPIO nanoparticles may remain in the dead cells or in the interstitial compartment or be taken up by macrophages. Thus signal voids in MR images may not necessarily represent living stem cells (19, 22). In this study, the ASC-treated hearts showed MR signal void(s) in the anterior wall of the LV. Tissue sections of the ASC-treated hearts showed SPIO nanoparticles and green spots (eGFP-expressing ASCs; Fig. 11). Spatially, the SPIO and green spots overlapped very well, indicating that signal voids on MR images were most likely the result of the eGFP/SPIO-labeled ASCs. To ensure that the fluorescing cells were alive, we recently monitored the green fluorescence of dead eGFP-labeled ASCs. The eGFP-labeled ASCs were killed with four cycles of freezing/thawing, and cell death was then confirmed. We found that the dead eGFP-labeled ASCs could continue fluoresce for 1 wk after cell death (data not shown). This suggests that some of the green dots on the heart tissue sections could come from dead eGFP-labeled ASCs. At present, it is unclear what percentage of the green dots was from living ASCs. However, it is expected that the majority of the green dots were from living ASCs. Since all of the SPIO nanoparticles were superimposed with green spots, we believe that the signal voids on MR imaging mainly represented living implanted cells.
In addition, as shown in Fig. 11, some eGFPpos ASCs stained negatively for SPIO. This could be related to a cell proliferation-associated decrease in the intracellular concentration of SPIO nanoparticles. As such, SPIO-enhanced MR imaging could not significantly overestimate living engrafted cells. However, the area of a signal void may be larger than the distribution area of the SPIO-labeled cells due to an amplification effect of ferrous contrast agents (4, 38). In this context, SPIO-enhanced MRI may falsely enlarge the area of stem cell distribution.
In this study therapeutic benefits of ASCs were assessed in small groups of healthy young rodents. Heart failure was induced by a sudden occlusion of the LAD. In contrast, human heart failure usually develops over a relatively long period in elders who often have atherosclerosis and suffer from obesity, diabetes, or hypertension. The animal model used in this study did not completely simulate human heart failure. Therefore, more studies, particularly with a large animal model, are needed to confirm the therapeutic benefits of ASCs transplantation.
The present study was to determine whether intramyocardial transplantation of ASCs improves contractile function and repairs injured myocardium. The results of this study do not provide direct insight on potential mechanisms for improvement of cardiac function and angiogenesis. It is also unclear why ASCs only resulted in a limited improvement in cardiac function compared with healthy controls and whether it is possible to further enhance the therapeutic benefits of ASCs.
Finally, cardiogenic transdifferentiation of ASCs was assessed only by monitoring expression of few cardiac transcriptional factors and fibril proteins. No electrophysiological or morphological studies were performed. As mentioned above, expression of some cardiac genes does not equal to a definite cardiac transdifferentiation. It remains to be determined whether ASCs can transdifferentiate into cardiomyocytes and vessel-constituting cells in vivo.
Intramyocardial injection of ASCs at 1 wk after LAD occlusion prevents deterioration in cardiac contractile function and enhances angiogenesis (capillary density) of infarcted rat hearts, indicating that ASCs may be a potential therapeutic option for myocardial infarction and heart failure. Under appropriate conditions ASCs express cardiac marker genes. Finally, SPIO-enhanced MR imaging seems a reliable technique to monitor the engrafted cells.
This work was supported by the National Research Council Canada, the Canadian Institutes of Health Research, Manitoba Health Research Council, and the Cardiac Sciences Program of St. Boniface General Hospital.
↵* L. Wang and J. Deng made equal contributions to this study.
- Copyright © 2009 the American Physiological Society