Embryonic Stem Cells and Released Factors Stimulate c-kit + /FLK-1 + Progenitor Cells and Promote Neovascularization in Doxorubicin-Induced Cardiomyopathy

ES Cells and Preparation of CM

ES cells (CGR8, a mouse embryonic stem cell line provided by Michel Puceat, Centre de Researches, INSERM, U390 France) were cultured and maintained on gelatin-coated plates (Sigma-Aldrich, St. Louis, MO, USA and BD Biosciences, San Jose, CA, USA, respectively) in Dulbecco's modified Eagle medium (Invitrogen, Grand Island, NY, USA) containing leukemia inhibitory factor (LIF; Millipore, Billerica, MA, USA), glutamine (Millipore), penicillin/streptomycin (Millipore), sodium pyruvate (Millipore), β-mercaptoethanol (Sigma-Aldrich), and fetal bovine serum (Invitrogen). ES cell-derived CM was generated as previously described (28,30). In brief, 5 × 10⁵ ES cells were cultured in a 100-mm Petri dish for 24 h followed by a media exchange with LIF-excluded media. After 48 h, cell media containing ES cell-released factors was removed, filtered (0.2-um filter; Millipore), labeled ES-CM, and retained for future use.

Animals and Experimental Protocol

All protocols involving animals were reviewed and approved by the University of Central Florida Animal Care Committee. Male and female C57BL6 mice, with no sex discrimination, between 8 and 10 weeks of age, were divided into the following five study groups: Control (C), DOX (D), DOX + cell culture (CC) media (D + CC), DOX + conditioned media (D + CM), and DOX + ES cells (D + ES Cells); n = 6—8 mice/group.

Intraperitoneal (IP) DOX (Sigma-Aldrich) injections were administered on day 1 (D1), D3, and D5 with each mouse receiving a cumulative dose of 12 mg/kg divided into three separate, equal injections, each containing 4 mg/kg. For treatment groups, CC media (400 μl/injection × three injections), ES-CM (400 μl/injection × three injections), or ES cells (5 × 10⁵ ES cells/injection × three injections) were injected IP on D2, D4, and D6. Twenty-eight days post-D6, heart function was assessed using echocardiography (Sonos 5500 Ultrasound System with a 15-6 L hockey-stick transducer; Philips Sonos, Andover, MA, USA), and animals were sacrificed by pentobarbitol (40 mg/kg IP; Henry Schein, Lake Mary, FL, USA) and cervical dislocation.

Preparation of Sections and Histopathology

Upon removal, hearts were fixed in 4% paraformaldehyde (Fisher Scientific, Pittsburgh, PA, USA) and embedded in paraffin. Hearts were cut into 5-μm serial sections, deparaffinized in xylene (Fisher Scientific), and rehydrated by sequential incubation in 100%, 95%, 70%, and 30% ethanol (Fisher Scientific). To visualize the vessels and quantify vascular fibrosis, hearts were stained with Masson's trichrome (Fisher Scientific) as previously described (3). Next, collagen accumulation around vessels [vascular fibrosis (VF)], as well as total vessel area (VA), were measured using ImageJ software (NIH, Bethesda, MD, USA), and a ratio of VF/VA was obtained.

Apoptotic Vessels

To identify apoptotic vasculature, heart sections were deparaffinized as aforementioned and permeabilized with proteinase K (Sigma-Aldrich; 25 μg/ml in 100 mM Tris HCl; Fisher Scientific). Apoptotic nuclei were identified using a terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling (TUNEL) assay (TMR red; Roche Applied Bio Sciences, Indianapolis, IN, USA) as previously published (3). Heart sections stained with TUNEL were costained with primary antibodies against rabbit anti-cluster of differentiation 31 (CD31; Cat# 250589, 1:100 dilution; Abbiotec, San Diego, CA, USA) followed by incubation with goat anti-rabbit IgG-fluorescein isothiocyanate secondary antibodies (1:30 dliution; Santa Cruz, Santa Cruz, CA, USA). Sections were then mounted with Vectashield (Vector Laboratories, Burlingame, CA, USA) and viewed under Olympus (Center Valley, PA, USA) fluorescent and Leica (Buffalo Grove, IL, USA) confocal microscopes. Left ventricular vessels containing colocalized CD31 and TUNEL were considered apoptotic. Apoptotic vasculature was quantified in one to two heart sections from n = 5—6 hearts.

c-kit, FLK-1, and CD31 Immunostaining

Heart sections were deparaffinized in xylene, rehydrated in ethanol, and washed in distilled water and PBS (Fisher Scientific) as described above. c-kit, FLK-1, and CD31 immunostainings were performed according to previously published protocols (3). In brief, nonspecific binding sites were blocked by incubation in 10% normal goat serum (NGS; Vector Laboratories) for 1 h. Sections were incubated with mouse anti-c-kit, anti-FLK-1, or anti-CD31 monoclonal antibodies (Santa Cruz Biotechnologies, Santa Cruz, CA) diluted 1:25 in 10% NGS for 60 min followed by three PBS washes. Next, sections were incubated with Alexa Fluor 568- or 488-conjugated goat anti-mouse IgG secondary antibodies (Invitrogen) diluted 1:50 in PBS for 60 min and mounted with antifade Vectashield medium containing 4′,6-diamidino-2-phenylindole (DAPI) (Vector Laboratories) for nuclear visualization. Quantitative analysis of c-kit⁺, FLK-1⁺, and CD31⁺ cells was performed on heart sections from three to five different hearts from each group under Olympus and Leica confocal microscopes. In brief, the total number of c-kit⁺ cells per section was corrected with DAPI and quantified under an Olympus microscope. Next, c-kit cells positive for FLK-1 and CD31 were quantified, and a ratio of double c-kit⁺/FLK-1⁺ and c-kit⁺/ CD31⁺ cells over total c-kit⁺ cells were obtained and multiplied by 100. Finally, CD31⁺ cells were quantified in the same fashion as c-kit and corrected with total DAPI to obtain a ratio of percent CD31⁺ cells per section.

Vascular Smooth Muscle Immunostaining

To quantify the number of vessels per heart section, deparaffinized and rehydrated sections were incubated overnight with mouse anti-smooth muscle α-actin antibody (1:100; Sigma-Aldrich), followed by anti-mouse antibody (4:1,000, M.O.M. kit; Vector Laboratories) for 1 h over an orbital shaker. Sections were then mounted with antifade Vectashield mounting medium containing DAPI to detect total nuclei. Total number of vessels per slide was quantified under Olympus microscope, and representative pictures for each group were obtained using a Leica scanning confocal microscope.

Echocardiographic Analysis

Cardiac function was assessed at 4 weeks post-D6 by transthoracic echocardiography in n = 6—8 mice/group as previously published (8). Mice were anesthetized with 2% isoflurane (Butler Animal Health Supply, Dublin, OH, USA) and placed in the supine position on a temperature-controlled heating pad (Gaymar Industries, Inc., Orchard Park, NY, USA). Echocardiography was performed using a Philips Sonos 5500 Ultrasound system with a 15-6 L hockey-stick transducer. Left ventricular internal dimension-diastole (LVIDd), left ventricular internal dimension-systole (LVIDs), fractional shortening [FS, (LVIDd − LVIDs)/LVIDd × 100], left ventricular volume at end diastole [EDV = 7/(2.4 + LVIDd) × LVIDd³], left ventricular volume at end systole [ESV = 7/(2.4 + LVIDs) × LVIDs³], and ejection fraction [EF = (EDV —ESV)/EDV × 100] were measured in the short-axis view at the midpapillary muscle level.

Data Analysis

Statistical analysis was performed using SigmaStat 3.5 software (Statcon, Witzenhausen, Germany). All values are presented as mean ± SEM. Analysis of data was performed using the one-way ANOVA and the Tukey ad hoc test. Statistical relevance was assigned when p < 0.05.

CM and ES Cells Prevent Vessel Apoptosis in DIC

Representative photomicrographs demonstrating vascular apoptosis are shown in Figure 1A with CD31⁺ cells stained red (A—E), TUNEL⁺ nuclei in green (F—J), and total nuclei stained blue with DAPI (K—O). The lower panels in Figure 1A (P—T) are merged images from each group. The smaller boxes (U—Y) are enhanced images shown to demonstrate the colocalization of CD31, TUNEL, and DAPI within a single vessel. Quantification of apoptotic vessels suggest vascular apoptosis is significantly elevated in D and D + CC hearts relative to controls (p < 0.05) (Fig. 1B). Importantly, CM and ES cell transplantation dramatically mitigate vessel apoptosis relative to D and D + CC groups (p < 0.05) (Fig. 1B). However, no statistical significance was obtained between CM and ES cell-transplanted hearts (Fig. 1B).

OPEN IN VIEWER

Figure 1.

Effects of transplanted CM and ES cells on vascular cell apoptosis in the DIC injured heart. (A) Representative photomicrographs demonstrating vascular apoptosis are shown with CD31⁺ cells stained red (A—E), TUNEL⁺ nuclei in green (F—J), and total nuclei stained blue with DAPI (K—O). The lower panels (P—T) are merged images from each group. The smaller boxes (U—Y) are enhanced images shown to demonstrate the colocalization of CD31, TUNEL, and DAPI within a single vessel. Scale bar: 100 μm. (B) Right histogram demonstrates quantitative vessel apoptotic data from n = 5—6 animals per group. *p < 0.05 versus C, #p < 0.05 versus D, and $p < 0.05 versus D + CC. D + CM: DOX + CM; D + ES cells: DOX + ES cells. DIC, doxorubicin-induced cardiomyopathy.

CM and ES Cells Activate c-kit⁺/FLK-1⁺ Cells in Chronic DIC

To determine the impact of transplanted CM and ES cells on cardiac endogenous vascular progenitor cells and their contribution to neovascularization in chronic DIC, sections were double-stained with c-kit and FLK-1. Representative images in Figure 2A demonstrate c-kit⁺ cells in green (A, F, K, P, and U), FLK-1⁺ cells in red (B, G, L, Q, and V), total nuclei in blue (C, H, M, R, and W), and merged images (D, I, N, S, and X). The smaller boxes (E, J, O, T, and Y) are enlarged images shown to demonstrate the colocalization of c-kit, FLK-1, and DAPI. Statistical analysis revealed that a significant decline in the number of c-kit⁺/FLK-1⁺ cells resulted in the D and D + CC groups relative to the non-DIC hearts (p < 0.05) (Fig. 2B). Importantly, a significant increase in the number of cells with combined c-kit and FLK-1 expression was quantitated in the CM and ES cell-transplanted groups suggesting that CM and ES cells may promote new artery formation in the DIC injured heart (p < 0.05) (Fig. 2B).

OPEN IN VIEWER

Figure 2.

Effects of transplanted CM and ES cells on c-kit and FLK-1 activation. (A) Representative images demonstrate v-kit Hardy—Zuckerman 4 feline sarcoma viral oncogene homolog-positive (c-kit⁺) cells in green (A, F, K, P, and U), fetal liver kinasepositive (FLK-1⁺) cells in red (B, G, L, Q, and V), total nuclei in blue (C, H, M, R, and W), and merged images (D, I, N, S, and X). Enlarged merged images are presented in E, J, O, T, and Y. Scale bar: 80 μm. (B) Right histogram reveals a significant increase in c-kit⁺/FLK-1⁺ activation following CM and ES cell transplantation. n = 5—6 animals/groups. *p < 0.05 versus C, #p < 0.05 versus D, and $p < 0.05 versus D + CC.

CM and ES Cells Enhance c-kit⁺/CD31⁺ Cells in the Chronic DIC Myocardium

To determine whether c-kit⁺ cells were present in mature coronary artery sites, sections were double-labeled with c-kit and CD31, a mature endothelial marker. Representative photomicrographs in Figure 3A show c-kit⁺ cells in green (A, F, K, P, and U), CD31⁺ cells in red (B, G, L, Q, and V), total nuclei stained with DAPI in blue (C, H, M, R, and W), and merged images (D, L, N, S, and X). Enlarged images demonstrating colocalization of c-kit, CD31, and DAPI are presented in (E, J, O, T, and Y). Data provided suggest that the amount of c-kit⁺ cells localized with CD31⁺ cells is significantly diminished in the D and D + CC groups compared to the control group (p < 0.05) (Fig. 3B). However, following CM and ES cell transplantation, c-kit⁺ cells colocalized with CD31⁺ cells were significantly enhanced relative to the D and D + CC groups, implying CM and ES cells activate c-kit⁺ cells, which, in turn, differentiate into mature endothelial cells (p < 0.05) (Fig. 3B).

OPEN IN VIEWER

Figure 3.

Transplanted CM and ES cells enhance c-kit⁺/CD31⁺ cells in DIC injured heart. (A) Representative photomicrographs show c-kit⁺ cells in green (A, F, K, P, and U), CD31⁺ cells in red (B, G, L, Q, and V), total nuclei stained with DAPI in blue (C, H, M, R, and W), and merged images (D, I, N, S, and X). Enlarged merged images are presented in E, J, O,T, and Y. Scale bar: 100 μm. (B) Right histogram demonstrates c-kit⁺/CD31⁺ cell quantitative data. n = 5—6 animals/groups. *p < 0.05 versus C, #p < 0.05 versus D, and $p < 0.05 versus D + CC.

CM and ES Cells Enhance CD31⁺ Cells in the DIC Heart

Representative images are shown in Figure 4A demonstrating CD31⁺ cells in red (A, E, I, M, and Q), total nuclei in blue (B, F, J, N, and R), and merged images (C, G, K, O, and S). Enlarged images of CD31⁺ cells are depicted in (D, H, L, P, and T). The number of CD31⁺ cells in heart sections from each group was quantified, and our data reveal a significant decrease in the number of these cells in the untreated DIC hearts compared to non-DIC myocardium (p < 0.05) (Fig. 4B). Notably, CM and ES cell-transplanted groups contained dramatically more CD31⁺ cells compared to the D and D + CC groups (p < 0.05) (Fig. 4B). Although not statistically significant, a trend of increased CD31 expression in D + CM and D + ES cell groups relative to control was observed, suggesting CM and ES cells promote vessel formation in response to acute DIC injury (Fig. 4B).

OPEN IN VIEWER

Figure 4.

Effects of CM and ES cell transplantation on cardiac CD31 expression. (A) Representative images are shown demonstrating CD31⁺ cells in red (A, E, I, M, and Q), total nuclei in blue (B, F, J, N, and R), and merged images (C, G, K, O, and S). Enlarged merged images are presented in D, H, L, P, and T. Scale bar: 100 μm. (B) Right histogram reveals a significant increase in CD31 expression in CM and ES cell groups compared to D and D + CC groups. n = 5—6 animals/groups. *p < 0.05 versus C, #p < 0.05 versus D, and $p < 0.05 versus D + CC.

Transplanted CM and ES Cells Promote Vessel Formation in the Injured DIC Heart

To determine the effects of transplanted CM and ES cells on coronary vessel formation, total myocardial vessels were identified using smooth muscle α-actin immunostaining and Masson's trichrome staining as depicted in Figure 5A and B, respectively. Quantitative analysis revealed a significant increase in the number of vessels in the CM and ES cell group myocardium relative to the DOX-alone injured heart (p < 0.05) (Fig. 5C). However, no statistical significance (p > 0.05) was observed between the CM and ES cell groups (Fig. 5C).

OPEN IN VIEWER

Figure 5.

Effects of transplanted CM and ES cells on coronary artery formation in the DIC-injured myocardium. (A) Representative images are shown demonstrating vascular smooth muscle cells (A, E, I, M, and Q), total nuclei in blue (B, F, J, N, and R), and merged images (C, G, K, O, and S). Enlarged merged images are presented in D, H, L, P, and T. (B) Representative photomicrographs demonstrating Masson's trichrome-stained sections from control and experimental groups (A—E). Enlarged vessels from (A—E) are depicted in (F—J). Arrows represent the areas enlarged from (A—E) to (F—J), respectively. (C) Histogram demonstrating quantitative analysis of total vessel counts. n = 5—6 animals/groups. Scale bar: 100 μm. *p < 0.05 versus C, #p < 0.05 versus D, and $p < 0.05 versus D + CC.

Transplanted CM and ES Cells Reduce Vascular Fibrosis in the Injured DIC Heart

Collagen accumulation around the vessels was detected on Masson's trichrome-stained heart sections (Fig. 6). The ratio of VF over VA demonstrated significantly increased fibrosis around vessels in DOX alone and DOX-CC animals (p < 0.05). Nonetheless, this increase was significantly reduced following CM and ES cell transplantation (p < 0.05).

CM and ES Cells Preserve Cardiac Function in Chronic DIC

Echocardiographic analysis was used to examine the effects of transplanted CM and ES cells on cardiac function at 4 weeks post-D6. LVIDd, LVIDs, FS, EDV, ESV, and EF were significantly dysregulated in the D and D + CC groups compared to the nonpathological mice (Fig. 7A—F). However, in CM and ES cell-transplanted mice, heart function for all parameters evaluated were significantly improved including 1) decreased LVIDd (Fig. 7A); 2) decreased LVIDs (Fig. 7B); 3) increased FS (Fig. 7C); 4) decreased EDV (Fig. 7D); 5) decreased ESV (Fig. 7E); and 6) increased EF (Fig. 7F) relative to D and D + CC mice (p < 0.05). Overall, the echocardiographic data suggest significant left ventricular functional improvement following CM and ES cell transplantation consistent with neovascularization.

OPEN IN VIEWER

Figure 6.

Transplanted CM and ES cells reduce vascular fibrosis following DIC. (A) (A—E) Representative images of VF for all experimental groups. (F) Histogram demonstrating quantitative analysis of VF/VA percent. n = 5—6 animals/groups. *p < 0.05 versus C, #p < 0.05 versus D, and $p < 0.05 versus D + CC.

OPEN IN VIEWER

Figure 7.

Transplanted CM and ES cells improve cardiac function following DIC. (A) LVIDd, (B) LVIDs, (C) FS, (D) EDV, (E) ESV, and (F) EF. LVIDd, left ventricular internal dimension-diastole; LVIDs, left ventricular internal dimension-systole; EDV, left ventricular volume at end diastole; ESV, left ventricular volume at end systole. n = 6—8 animals/groups. *p < 0.05 versus C, #p < 0.05 versus D, and $p < 0.05 versus D + CC.