Regulation of Transplanted Cell Homing by FGF1 and PDGFB after Doxorubicin Myocardial Injury

Abstract

There is no effective treatment for the total recovery of myocardial injury caused by an anticancer drug, doxorubicin (Dox). In this study, using a Dox-induced cardiac injury model, we compared the cardioprotective effects of ventricular cells harvested from 11.5-day old embryonic mice (E11.5) with those from E14.5 embryos. Our results indicate that tail-vein-infused E11.5 ventricular cells are more efficient at homing into the injured adult myocardium, and are more angiogenic, than E14.5 ventricular cells. In addition, E11.5 cells were shown to mitigate the cardiomyopathic effects of Dox. In vitro, E11.5 ventricular cells were more migratory than E14.5 cells, and RT-qPCR analysis revealed that they express significantly higher levels of cytokine receptors Fgfr1, Fgfr2, Pdgfra, Pdgfrb and Kit. Remarkably, mRNA levels for Fgf1, Fgf2, Pdgfa and Pdgfb were also found to be elevated in the Dox-injured adult heart, as were the FGF1 and PDGFB protein levels. Addition of exogenous FGF1 or PDGFB was able to enhance E11.5 ventricular cell migration in vitro, and, whereas their neutralizing antibodies decreased cell migration. These results indicate that therapies raising the levels of FGF1 and PDGFB receptors in donor cells and or corresponding ligands in an injured heart could improve the efficacy of cell-based interventions for myocardial repair.

Keywords: cardiac dysfunction; cardiomyopathy; doxorubicin; growth factor and chemokine receptors; ventricular cell migration.

Figure 1

Comparison of homing efficiencies of E11.5 and E14.5 ventricular cells in the host myocardium by tail vein injection method. (A–D) Representative micrographs from X-Gal stained sections of cell injection experiments. Recipient mice were treated with (A) saline or (B) doxorubicin (Dox) for 3 days, donor cells were injected 3 days later in the tail veins and recipient hearts were processed after 3 days for X-Gal, or (C) Sirius Red and Fast Green (SR + FG) and (D) hematoxylin and eosin staining (H&E). Arrows indicate engrafted X-Gal positive donor cells in panels A and B. (C) Fibrotic myocardium after Dox treatment is revealed by a red stain. (D) Engrafted cells were found frequently in areas of inflammation (arrows indicate small inflammatory cells and arrowheads indicate X-Gal positive donor cells). (E) Representative cryosection sequentially processed for von Willebrand factor (vWF) and X-Gal staining. Note absence of X-Gal positive donor cells in blood vessel compartments indicated by the arrows. (F) Quantification of engraftment efficiency of X-Gal positive cells in saline (S) or Dox (D) treated mice. * p < 0.05 vs. all other groups, $ p < 0.05 vs. Dox + 3-days post E11.5 cell injections, # p < 0.05 vs. Dox + 3- or 7-days post E11.5 cell injections, One-way ANOVA with Tukey’s multiple comparison test. Results are mean ± SEM of 3 experiments/group. (G) Quantification of blood vessels per section in recipient hearts after systemic delivery of donor cells. * p < 0.05 saline vs. all other groups, # p < 0.05 vs. Dox + 3- or 7-days post E14.5 cell injections and saline group, One-way ANOVA with Tukey’s multiple comparison test. Results are mean ± SEM of 30 sections from 3 recipient hearts/group.

Figure 2

Cell migration (A) and invasion (B) properties of embryonic ventricular cells. 300 μL of 0.5 × 10⁶ cells/mL serum-free cell suspensions were incubated for 22 h at 37 °C in polycarbonate membrane inserts with 8 μm pores with surrounding wells containing 10% FBS DMEM. The bottom surface of the polycarbonate inserts was coated with type I collagen for the cell invasion assay (B), but this coating was absent in cell migration assay (A). The inserts were then removed, migratory cells that had passed through the pores were washed and stained, and the cell stain was solubilized for measurement of absorbance. Asterisk indicates that cell migration was significantly lower in the E14.5 cells p < 0.05, N = 6 independent experiments for each group).

Figure 3

Age-dependent change in cytokine receptor mRNAs in embryonic ventricular cells. Equal amounts of total RNA from E11.5 and E14.5 ventricles were reverse transcribed using oligo-dT primers, and real-time quantitative polymerase chain reaction (RT-qPCR) was used to amplify and quantify the product, with the expression intensity normalized against GAPDH. The log2 fold changes for each transcript in E11.5 vs. E14.5 ventricles were plotted on y-axis. * p < 0.005 vs. E14.5 ventricles, # p < 0.05 vs. E14.5 ventricles. Students unpaired t-test. N = 3-4 independent experiments for each group).

Figure 4

Comparison of chemokine ligand mRNA levels in ventricular cells from adult hearts following doxorubicin treatment. Mice were given intraperitoneal injections of saline (Control) or doxorubicin (Dox), and their hearts were harvested after 3 days. Equal amounts of total RNA were reverse transcribed using oligo-dT primers, and real-time quantitative polymerase chain reaction (RT-qPCR) was used to amplify and quantify the product, with the expression intensity normalized against GAPDH. The log2 fold changes for each transcript in Dox vs. saline ventricles were plotted on y-axis. * p < 0.005 vs. saline treated group, # p < 0.05 vs. saline treated group. Students unpaired t-test. N = 3 independent experiments for each group).

Figure 5

Doxorubicin increases FGF-1 and PDGF-B protein levels in adult heart ventricular cells. Mice were given intraperitoneal injections of saline (S) or doxorubicin (D) and their hearts were harvested after 24 h (Day 1) or 72 h (Day 3). Using Western blotting, lysates containing equal amounts of total protein from the ventricular tissue of the differently treated animals were tested for FGF-1 and PDGF-B protein levels. When standardized against GAPDH, the level of both proteins (FGF-1 and PDGF-B) was found to be significantly higher in the Dox-treated ventricles than in the saline treated ventricles by Day 3. * p < 0.05 vs. saline (S), One-way ANOVA with Tukey’s multiple comparisons test, N = 3 independent experiments for each group.

Figure 6

FGF-1 and PDGF-B increase embryonic ventricular cell migration, whereas their neutralizing antibodies decrease migration. 300 μL of 0.5 × 10⁶ cells/mL serum-free cell suspensions were incubated for 22 h at 37 °C in polycarbonate membrane inserts with 8 μm pores, with surrounding wells containing 1% FBS DMEM and FGF-1, its antibody, or both (A) or PDGF-B, its antibody, or both (B). The inserts were then removed, migratory cells that had passed through the pores were washed and stained, and the cell stain was solubilized for measurement of absorbance. * p < 0.05 vs. all other groups; # p < 0.05 vs. all other groups, One-way ANOVA with Tukey’s multiple comparisons test, N = 3 independent experiments for each group.

Figure 7

Cardiac function defects can be corrected by embryonic ventricular cell injections. Electrocardiographic analysis of saline or doxorubicin injected C57BL/6 mice treated with or without tail injection of E11.5 ventricular cells. (A) Electrocardiogram (ECG) traces of different treatment groups of recipient mice. (B–F) Quantification of ECG parameters, (B) heart rate (HR), (C) PR interval, (D) QRS interval, (E) RR interval, and (F) QTc of saline and doxorubicin (Dox)-treated mice injected with and without embryonic day (E) 11.5 ventricular cells. (D) # p < 0.05 vs. other groups, One-way ANOVA with Tukey’s multiple comparison test. Results are mean ± SEM of 3–4 experiments/group. Panels B,C and E,F: No significant differences observed between groups.

Figure 8

Echocardiographic analysis of Saline or Doxorubicin injected mice treated with or without tail injection of E11.5 ventricular cells. (A) Representative short axis M-mode image of left ventricle (LV) illustrating dimensions of LV wall, cavity and cardiac functional measurements. The y-axis indicates the distance from transducer (in cm) and the x-axis indicates time (in ms). M-Mode images illustrate LV anterior wall (AW) and posterior wall (PW) along with the ventricular chamber, through systole (s; green dotted line) and diastole (d; red dotted line). LVIDs, LV internal diameter during systole; LVIDd, LV internal diameter during diastole. (B–F) Quantification of short-axis M-mode echocardiogram parameters of untreated and doxorubicin (Dox)-treated BL6 mice treated with and without tail vein injection of E11.5 ventricular cells. (B) EF, ejection fraction, (C) %FS, percent fractional shortening, (D) SV, stroke volume, (E) EDV, end diastolic volume (F) ESV, end systolic volume. # p < 0.05 vs. saline, and & p < 0.05 vs. Dox + E11.5 cells. One-way ANOVA with Tukey’s multiple comparison test. Results are mean ± SEM of 3–4 experiments/group.