Deciphering the Cardiovascular Potential of Human CD34+ Stem Cells

Abstract

Ex vivo monitored human CD34⁺ stem cells (SCs) injected into myocardium scar tissue have shown real benefits for the recovery of patients with myocardial infarctions. They have been used previously in clinical trials with hopeful results and are expected to be promising for cardiac regenerative medicine following severe acute myocardial infarctions. However, some debates on their potential efficacy in cardiac regenerative therapies remain to be clarified. To elucidate the levels of CD34⁺ SC implication and contribution in cardiac regeneration, better identification of the main regulators, pathways, and genes involved in their potential cardiovascular differentiation and paracrine secretion needs to be determined. We first developed a protocol thought to commit human CD34⁺ SCs purified from cord blood toward an early cardiovascular lineage. Then, by using a microarray-based approach, we followed their gene expression during differentiation. We compared the transcriptome of undifferentiated CD34⁺ cells to those induced at two stages of differentiation (i.e., day three and day fourteen), with human cardiomyocyte progenitor cells (CMPCs), as well as cardiomyocytes as controls. Interestingly, in the treated cells, we observed an increase in the expressions of the main regulators usually present in cardiovascular cells. We identified cell surface markers of the cardiac mesoderm, such as kinase insert domain receptor (KDR) and the cardiogenic surface receptor Frizzled 4 (FZD4), induced in the differentiated cells in comparison to undifferentiated CD34⁺ cells. The Wnt and TGF-β pathways appeared to be involved in this activation. This study underlined the real capacity of effectively stimulated CD34⁺ SCs to express cardiac markers and, once induced, allowed the identification of markers that are known to be involved in vascular and early cardiogenesis, demonstrating their potential priming towards cardiovascular cells. These findings could complement their paracrine positive effects known in cell therapy for heart disease and may help improve the efficacy and safety of using ex vivo expanded CD34⁺ SCs.

Keywords: CD34+ cells; cardiovascular differentiation; cell therapy; gene expression profiling; umbilical cord blood.

Figure 1

Cardiovascular differentiation of human CD34⁺ SCs and CMPCs: procedures and morphologies. (A) Schematic description of protocol used for cardiovascular differentiation of human CD34⁺ SCs to study gene expression profiles at day 0 (d0), d3, and d14 of differentiation. (B) Morphology of CD34⁺ SCs purified from human cord blood (H-UCB), as observed by optical microscopy on d3 and d14. (C) Cardiac differentiation procedure of cardiomyocyte progenitor cells (CMPCs) (upper panel) and their morphologies at d1 and d29 (lower panel). Bar scale: 100 µm.

Figure 2

Cardiovascular gene expression. Relative expression analysis of indicated genes as determined by real time RT-PCR in CD34⁺ SCs, controls, and cells differentiated at d3 and d14, as well as CMPCs and cardiomyocytes. Data are presented as means ± SD of four individual experiments; Unpaired t-test, * p < 0.05, ** p < 0.01, *** p < 0.005 and **** p < 0.001 when compared to CD34⁺ SCs; ns: not significant.

Figure 3

Global gene expression profiling of human CD34⁺ cells at different stages of differentiation. (A) Numbers of genes with expression changes significantly (p < 0.001) affected in the indicated samples versus CD34⁺ SCs. (B) Hierarchical clustering of differential expression profiles among human CD34⁺ cells and those at different stages of differentiation (Diff d3 and Diff d14), CMPCs and cardiomyocytes, based on Pearson correlation of significant (p < 0.001) differentially expressed genes among 20 samples. Upregulated genes and downregulated genes are, respectively, represented in red and green. (C) Differential expression profiles of gene expressions determined by microarray represented as scatter plot of the means of total gene expressions versus log₂ fold-changes of Diff d14 cells compared to CD34⁺ SCs (upper panel) and of CMPCs compared to Diff d14 cells (lower panel) calculated from the means of four samples (Red represents the number of upregulated genes and green those downregulated). Statistical significance was calculated by ANOVA using the following parameters, p < 0.001, Benjamini-Hochberg’s correction and SNK test.

Figure 4

Heat map showing normalized expression variations in the indicated sets of genes in treated cells and differentiated CMPCs and cardiomyocytes relative to CD34⁺ SCs (green is decrease, and red is increase, relative to control). The color scale is shown at the bottom. Heat map shows average of normalized intensities for one condition (n = 4).

Figure 5

Differential gene expression profiling of Diff d14 vs. CD34 ⁺ SCs. (A) Hierarchical clustering carried out from 6837 probes by t-test (p < 0.001) on eight samples and based on Pearson correlation analysis. (B) Gene ontology analysis of the significantly up-regulated genes in Diff d14 cells versus CD34⁺ SCs and their classification according to their biological functions. The gene related to angiogenesis, cardiovascular system development, and heart development are overexpressed significantly in Diff d14 using t-test (significance was assigned for p < 0.001) with Benjamini–Hochberg’s correction. (C–E) Plots (Log₂ FC) of top unregulated genes in Diff d14 vs. CD34⁺ SC for cardiac cell differentiation (C), endothelial differentiation (D), and cardiovascular cell development (E). A pAdj value < 0.001 is indicated for each gene and is calculated by adjusting the p-value with Benjamini Hochberg false discovery rate (FDR).

Figure 6

Heat maps showing expression levels of sets of genes involved in cardiovascular differentiation and Wnt (A) or TGF-β (B) signaling in CD34⁺ SCs, treated cells Diff d14, CMPCs, and cardiomyocytes (green is decrease and red is increase relative to control). The color scale is shown at the bottom. Heat maps show averages of normalized values for one condition (n = 4).

Figure 7

FCM analysis of cardiovascular receptor expressions in CD34-positive population of untreated CD34⁺ SCs (blue) and CD105-positive population of Diff d14 cells (red). The black histograms indicate the corresponding isotype signals.

Figure 8

Immunofluorescence staining of endothelial and cardiac markers, FZD4, CD31, vWF, and sarcomeric α-actin (SαA) in Diff d14 cells. The left panels show the controls (CTL) with the secondary antibodies alone. Green is Alexa Fluor 488-stained, and red is Alexa Fluor 555-stained. DAPI was used to visualize nuclei (blue). The pictures display representative merged immunofluorescence images. Scale bar, 20 µm. (Original magnification, ×20 (left and middle panels), ×40 (right panel)).

Figure 9

Proposed mechanism involved in CD34⁺ stem cell cardiac regenerative medicine.