EMG7 mouse ESCs, which have an αMHC promoter-driven enhanced GFP gene, E14Tg2a ESCs, and OP9 cells were generated as described previously[12-14] and transferred to KAIST.

Lentiviruses were generated by transfecting FUtdTW (Addgene plasmid 22478)[15] with pMD2.G (Addgene plasmid 12259), pMDLg/pRRE (Addgene plasmid 12251) and pRSV-Rev (Addgene plasmid 12253)[16] in 293T cells using jetPEI (Polypus-transfection). Supernatants were collected 48 h after transfection, filtered through a 0.45 μm filter, and concentrated by Lenti-X concentrator (Clontech). Viral particles were resuspended in ESC medium with 4 mg/mL polybrene. EMG7 mouse ESCs were incubated in this medium for 24 h. Selection of ESCs was performed by FACS sorting.

For the induction of Flk1⁺ mesodermal precursor cells (MPCs), ESCs were cultured without leukemia inhibitory factor (LIF, Millipore) and plated on a 0.1% gelatin-coated dish at a cell density between 1 × 10³ and 1.5 × 10³ cells cm² in the differentiation medium, which is αMEM (Invitrogen) containing 10% fetal bovine serum (FBS, Welgene), 0.1 mmol/L of 2-mercaptoethanol (Invitrogen), 2 mmol/L of L-glutamine (Invitrogen) and 50 U/mL of penicillin-streptomycin (Invitrogen). Medium was changed every other day for 4.5 d. At day 4.5, differentiated ESCs were harvested with 0.25% trypsin-EDTA (Invitrogen), and antigen retrieval was performed in the differentiation medium for 30 min in an incubator. Then, cells were washed using 2% FBS in phosphate buffered saline (PBS) and incubated with biotinconjugated anti–mouse Flk1 antibody (clone AVAS12a1, eBioscience) and anti-streptavidin MicroBeads (Miltenyi Biotec). Flk1⁺ MPCs were sorted by AutoMACS Pro Separator (Miltenyi Biotec). For induction of CLCs, sorted Flk1⁺ MPCs were plated onto the mitomycin C (AG Scientific)-treated confluent OP9 cells at a density of 5-10 × 10³ cells cm² in the medium containing 3 μg/mL of CsA, 10 μmol/L of Y27632, 400 μmol/L of Trolox, and 1 μg/mL of EW7197 (CsAYTE)[11,17]. The medium was refreshed every other day. Live images of differentiation process of CLCs and CMs were obtained using Axiovert 200M microscope (Carl Zeiss) equipped with AxioCam MRm (Carl Zeiss). Phase contrast images including beating CMs were obtained using an Infinity X digital camera and DpxView LE software (DeltaPix).

The cells were harvested with 0.25% trypsin-EDTA or dissociation buffer (Invitrogen). To analyze live cells, antigen retrieval was performed in the differentiation medium for 30 min in an incubator and the cells were incubated for 20 min with the following antibodies: Allophycocyanin-conjugated anti–mouse PDGFRα (eBioscience, 17-1401, clone APA5, 1:100) and phycoerythrin/Cy7-conjugated anti–mouse Flk1 (BioLegend, 136414, clone AVAS12a1, 1:50). In live cell analysis and sorting, dead cells were excluded using 4,6-diamidino-2-phenylindole (DAPI, Sigma, D8417, 1:1000), and OP9 cells were excluded from Flk1⁺ MPC by gating in flow cytometry. The differentiated CMs were sorted using αMHC-GFP. Analyses and sorting were performed by FACS Aria II (Beckton Dickinson). Data were analyzed using FlowJo Version 7.5.4 software (TreeStar).

Twenty eight male 9-wk-old BALB/c nude mice were kept in the specific pathogen free before the experiment under a 12:12 h light/dark cycle with lights on at 8:00 AM. They were deprived of food for 18 h but permitted water ad libitum before surgery. Animal care and experimental procedures were performed to conform the NIH guidelines (Guide for the care and use of laboratory animals) and approved by the Animal Care Committee of KAIST (KA2013-40).

All mice were anesthetized through an intraperitoneal injection of a combination of anesthetics (80 mg/kg ketamine, 12 mg/kg xylazine) before any procedures. After intubation, the mice were ventilated with room air (SomnoSuiteTM, Kent scientific). MI was induced by exposing the heart by left thoracotomy and permanently ligating the proximal portion of left anterior descending coronary artery with an 8-0 prolene thread under respiratory support. After ligating the proximal portion of left anterior descending coronary artery, infarction of the anterior wall of left ventricle was confirmed in each mouse by the presence of a pale anterior wall and myocardial hypokinesis. Immediately after ligation of coronary artery and the confirmation of infarction, 100 μL PBS containing 1 x 10⁶ PDGFRα⁺ CLCs or αMHC⁺ CMs were intramyocardially injected with a 31-gauge (0.25 mm) insulin syringe into the 3 different sites along the borderline of the infarcted area.

Transthoracic echocardiography (TTE) studies were performed (VIVID 7 dimension system, General Electric-Vingmed Ultrasound) 15 d after MI surgery and cell implantation. Images were obtained using an i13L transducer (5.3-14.0 MHz, GE Healthcare) with high temporal and spatial resolution. Two-dimensionally targeted M-mode parameters were measured at a level of papillary muscle in parasternal short axis view during 6 consecutive cardiac beats. All measurements were performed in a blind fashion according to the guidelines of American Society for Echocardiography.

Before sacrifice, mice were anesthetized with a mixture of ketamine (80 mg/kg) and xylazine (10 mg/kg). For hematoxylin and eosin (H and E) staining, samples were fixed overnight with 4% paraformaldehyde and embedded in paraffin after tissue processing. For immunofluorescence staining, samples were fixed in 4% paraformaldehyde, dehydrated in 20% sucrose solution overnight, and embedded in tissue freezing medium (Leica). Samples were blocked with 5% goat (or donkey) serum in 0.01% Trition X-100 in PBS and then incubated overnight at 4 °C with the following primary antibodies: Mouse anti-α-actinin monoclonal antibody (Sigma Aldrich, A7811, clone EA-53, 1:100) or rabbit anti-α-actinin polyclonal antibody (Abcam, ab68167, clone EP2529Y, 1:100), rabbit anti-Ki-67 polyclonal antibody (Abcam, ab15580, 1:200), mouse anti-α-SMA monoclonal antibody (Sigma Aldrich, A2547, clone 1A4, 1:500), hamster anti-CD31 monoclonal antibody (Millipore, MAB1398Z, clone 2H8, 1:400), and rabbit anti-GFP polyclonal antibody (Millipore, AB3080, 1:200). After several washes, the samples were incubated for 2 h at RT with the following secondary antibodies: Cy5-conjugated anti-mouse IgG (Jackson ImmunoResearch, 715-175-150, 1:1000) and Cy3-, Cy5-, FITC-conjugated anti-rabbit IgG antibodies (Jackson ImmunoResearch, 711-165-152, 711-175-152, 711-095-152, 1:1000). Then the samples were mounted with fluorescent mounting medium (DAKO) and immunofluorescent images were acquired using a Zeiss LSM780 confocal microscope (Carl Zeiss). To calculate capillary density, number of CD31⁺ endothelial cells was counted per random 0.5 mm² area in the infarcted myocardium at 15 d after cell implantation. To analyze the regenerative effects of host myocardium, number of Ki-67⁺/α-actinin⁺ CMs was counted per 104 nuclei in the peri-infarcted area, ranged within 200 μm from infarcted region, at 3 and 15 d after cell implantation. Images were analyzed using ImageJ software (http://imagej.nih.gov/ij/, 1.47V, NIH, United States).

Values are presented as mean ± SD. For continuous data, statistical significance was determined with the Mann-Whitney U test between 2 groups and the Kruskal-Wallis test followed by Tukey’s honest significant difference test with ranks or multiple-group comparison. Statistical analysis was performed with SAS 9.4 (SAS Institute Inc). Statistical significance was set at P < 0.05 or 0.01.

To investigate the regenerative potential of PDGFRα⁺ CLCs and αMHC⁺ CMs, cells were sorted, and approximately 1 × 10⁶ of each were implanted into the left ventricular myocardium after inducing acute MI. The results of the recipient groups were compared with those of MI hearts without implantation. Analyses were performed at 2 wk after implantation of cells (Figure 1A). To trace the implanted cells in the infarcted heart, we induced PDGFRα⁺ CLCs from ESCs expressing tdTomato fluorescence (Figure 1B). As shown in Figure 1C, the implanted cells were mainly distributed along several myocardial cavities 1 h after implantation.

Open in New Tab   Full Size Figure   Download Figure

Figure 1 Implantations of platelet-derived growth factor receptor-α⁺ cardiac lineage-committed cells and αMHC⁺ cardiomyocytes in the acute myocardial infarction murine model. A: Experimental scheme of implanting either platelet-derived growth factor receptor-α (PDGFRα)⁺ cardiac lineage-committed cells (CLCs) or αMHC⁺ cardiomyocytes (CMs) into acute myocardial infarction (MI) murine model. Analyses were performed at 2 wk after implantation of approximately 1 × 10⁶ cells of PDGFRα⁺ CLCs or αMHC⁺ CMs into the left ventricular myocardium of acute MI murine model; B: Live cell image showing tdTomato⁺ cells during induction of PDGFRα⁺ CLCs from embryonic stem cells. Scale bars, 100 μm; C: Representative confocal image showing implanted tdTomato⁺ PDGFRα⁺ CLCs in the myocardial spaces (arrowheads) of the infarcted zone (dotted line and arrow), which was formed by ligation of coronary artery 1 h prior to the implantation. Scale bar, 500 μm. CLCs: Cardiac lineage-committed cells; CMs: Cardiomyocytes; PSCs: Pluripotent stem cells; ESCs: Embryonic stem cells; LIF: Leukemia inhibitory factor.

First, to evaluate the functional recovery of infarcted hearts after cell implantation, we performed TTE 14 d after implantation. Compared with that in untreated MI hearts, the anterior and septal regional wall motion was notably and similarly improved (see arrowheads in Figure 2A) in the MI hearts implanted with PDGFRα⁺ CLCs (hereafter designated as MI+PDGFRα⁺ CLCs) and αMHC⁺ CMs (designated as MI+αMHC⁺ CMs). Moreover, the left ventricular internal dimension during systole of both MI+PDGFRα⁺ CLCs and MI+αMHC⁺ CMs was approximately 12%–23% less compared with that of untreated MI hearts (Figure 2B). Both MI+PDGFRα⁺ CLCs and MI+αMHC⁺ CMs also showed significant and similar improvements in systolic functional parameters, which included an ejection fraction increased by 20.0/15.6% and fractional shortening increased by 9.5/7.2%, respectively, compared with those of untreated MI hearts (Figure 2C). All TTE parameters are summarized in Table 1. These findings indicate that the implantation of PDGFRα⁺ CLCs and αMHC⁺ CMs had similar beneficial effects in the functional recovery of acutely infarcted hearts. Next, to confirm whether the implanted cells were properly engrafted to the infarcted myocardium, we performed histologic analyses at 15 d after implantation. Overall, the gross sizes of MI+PDGFRα⁺ CLCs and MI+αMHC⁺ CMs were smaller than that of untreated MI hearts (Figure 2D). Hematoxylin and eosin staining showed that untreated MI hearts had a thinner ventricular wall (0.19 mm) than did controls, while the ventricular walls of MI+PDGFRα⁺ CLCs and MI+αMHC⁺ CMs were similarly thicker (0.47 mm and 0.39 mm, respectively) compared with that of untreated MI hearts (Figures 2E and F).

Open in New Tab   Full Size Figure   Download Figure

Figure 2 Implantations of platelet-derived growth factor receptor-α⁺ cardiac lineage-committed cells and αMHC⁺ cardiomyocytes equally improves contractile function and structure in the infarcted heart. A: Representative M-mode transthoracic echocardiography views of control, myocardial infarction (MI), MI+ platelet-derived growth factor receptor-α (PDGFRα)⁺ cardiac lineage-committed cells (CLCs), and MI+αMHC⁺ cardiomyocytes (CMs). Improved anterior (white arrowheads) and septal (yellow arrowheads) regional wall motion are observed in the left ventricles of MI+PDGFRα⁺ CLCs and MI+αMHC⁺ CMs; B and C: Quantifications of left ventricular internal dimension in systole (mm), ejection fraction (%) and fractional shortening (%). Each group, n = 7. ^aP < 0.05 and ^bP < 0.01 vs Con; ^cP < 0.01 vs MI; D: Gross images of hearts in control, MI, MI+PDGFRα⁺ CLCs, and MI+αMHC⁺ CMs. Scale bars, 2.5 mm; E: H and E staining of mid-sectioned hearts of control, MI, MI+PDGFRα⁺ CLCs, and MI+αMHC⁺ CMs. Scale bars, 1 mm and 50 μm in the upper and lower panels, respectively; F: Quantifications of the thickness (mm) of left ventricle in the infarcted region. Each group, n = 7. ^bP < 0.01 vs Con; ^cP < 0.01 vs MI or MI+PDGFRα⁺ CLCs. CLCs: Cardiac lineage-committed cells; CMs: Cardiomyocytes; MI: Myocardial infarction.

Importantly, implanted PDGFRα⁺ CLCs and αMHC⁺ CMs were visible as tdTomato⁺/α-MHC-GFP⁺ cells aligned and integrated with host CMs (Figure 3A). Implanted PDGFRα⁺ CLCs and αMHC⁺ CMs were mostly differentiated into α-actinin⁺ CMs, and they did not convert into CD31⁺ endothelial cells or αSMA⁺ mural cells (Figures 3B and C). Moreover, CD31⁺ blood vessels in the infracted area increased by 2.1- and 1.8-fold in MI+PDGFRα⁺ CLCs and MI+αMHC⁺ CMs at day 15 after implantation (Figures 4A and B), while the numbers of Ki-67⁺ CMs also transiently increased equally by 2.4-fold at day 3; no such increases were detected at day 15 in both groups (Figures 4C and D). Thus, in addition to integration of implanted MI+PDGFRα⁺ CLCs and MI+αMHC⁺ CMs into the host myocardium, paracrine effects of MI+PDGFRα⁺ CLCs and MI+αMHC⁺ CMs appeared to be involved in the functional recovery of acutely infarcted hearts. Both types of implanted cells persisted up to 60 d after implantation (Figure 4E), which was the longest observation period in this study.

Open in New Tab   Full Size Figure   Download Figure

Figure 3 Integration and differentiation of implanted platelet-derived growth factor receptor-α⁺ cardiac lineage-committed cells in the infarcted heart. A: tdTomato-tagged platelet-derived growth factor receptor-α (PDGFRα)⁺ cardiac lineage-committed cells (CLCs) or αMHC⁺ cardiomyocytes (CMs) were implanted into the infracted myocardium and integration was confirmed by immunostaining. tdTomato⁺/α-MHC-GFP⁺ cells (white arrowheads) are implanted PDGFRα⁺ CLCs and αMHC⁺ CMs. Scale bars, 100 μm; B: Representative confocal images showing differentiation of tdTomato⁺ PDGFRα⁺ CLCs and αMHC⁺ CMs into cardiomyocytes 15 d after the implantation. αSMA-expressing cells were negative for tdTomato or α-MHC-GFP signal as shown in the inlet. Scale bars, 25 μm; C: Percentages of α-actinin⁺ CMs, CD31 endothelial cells and αSMA⁺ smooth muscle cells of the implanted PDGFRα⁺ CLCs and αMHC⁺ CMs. Each group, n = 4. Scale bars, 20 μm. CLCs: Cardiac lineage-committed cells; CMs: Cardiomyocytes; MI: Myocardial infarction; ECs: Endothelial cells; SMCs: Smooth muscle cells.

Open in New Tab   Full Size Figure   Download Figure

Figure 4 Proliferation and survival of implanted platelet-derived growth factor receptor-α⁺ cardiac lineage-committed cells in the infarcted heart. A: Representative confocal image of Ki-67⁺ host cardiomyocytes in the peri-infarcted region 3 d after cell injection. Dotted-lined rectangular region is magnified in right. Scale bar, 50 μm; B: Quantifications of Ki-67⁺α-actinin⁺ cardiomyocytes per 10⁴ nuclei in the peri-infarcted region 3 and 15 d after the implantation. Each group, n = 4. ^aP < 0.05 vs myocardial infarction (MI); C: Representative confocal images of revascularization within the infarcted areas 15 d after the implantation. Scale bars, 100 μm; D: Quantifications of capillary density (No. of CD31⁺ ECs/mm²) within the infarcted areas. Each group, n = 4. ^bP < 0.01 vs MI; ^cP < 0.01 vs MI+ platelet-derived growth factor receptor-α (PDGFRα)⁺ cardiac lineage-committed cells (CLCs); E: Representative confocal images of tdTomato⁺ PDGFRα⁺ CLCs and αMHC⁺ CMs 60 d after the implantation. Three independent experiments showed similar findings. Scale bars, 20 μm; F: Schematic diagram illustrating the regenerative potential of PDGFRα⁺ CLCs in the infarcted heart. CLCs: Cardiac lineage-committed cells; CMs: Cardiomyocytes; MI: Myocardial infarction.

Pluripotent stem cell (PSC)-derived cardiomyocytes (CMs) have become one of the most attractive cellular resources for cell-based therapy to rescue damaged cardiac tissue.

The proliferative capacity of PSC-derived CMs is decreased after beating and terminal differentiation. Furthermore, there is no definite surface marker of differentiated PSC-derived CMs to facilitate purification.

We investigated the regenerative potential of mouse embryonic stem cell-derived PDGFRα⁺ cardiac lineage-committed cells (CLCs) in a murine myocardial infarction (MI) model and compared their efficacy with differentiated CMs.

We implanted platelet-derived growth factor receptor-α (PDGFRα)⁺ CLCs and differentiated αMHC⁺ CMs into a MI murine model and performed functional analysis using transthoracic echocardiography (TTE) and histologic analysis.

Compared with the untreated MI hearts, the anterior and septal regional wall motion and systolic functional parameters were notably and similarly improved in the MI hearts implanted with PDGFRα⁺ CLCs and αMHC⁺ CMs based on TTE. In histologic analysis, the untreated MI hearts contained a thinner ventricular wall than did the controls, while the ventricular walls of MI hearts implanted with PDGFRα⁺ CLCs and αMHC⁺ CMs were similarly thicker compared with that of the untreated MI hearts. Furthermore, implanted PDGFRα⁺ CLCs aligned and integrated with host CMs and were mostly differentiated into α-actinin⁺ CMs, and they did not convert into CD31⁺ endothelial cells or αSMA⁺ mural cells.

PDGFRα⁺ CLCs from mouse ESCs exhibiting proliferative capacity showed a regenerative effect in infarcted myocardium. Therefore, mouse ESC-derived PDGFRα⁺ CLCs may represent a potential cellular resource for cardiac regeneration.

PDGFRα⁺ CLCs served as the potential donor population for cardiac regeneration, and our findings provide conceptual and technical advances in stem cell therapy for cardiac regeneration.