Genetic module and miRNome trait analyses reflect the distinct biological features of endothelial progenitor cells from different anatomic locations

Abstract

Background: Endothelial progenitor cells (EPCs) play a fundamental role in post-natal vascular repair, yet EPCs from different anatomic locations possess unique biological properties. The underlying mechanisms are unclear.

Results: EPCs from CB expressed abundant genes involved in cell cycle, hypoxia signalling and blood vessel development, correlating with the phenotypes that CB-EPCs proliferated more rapidly, migrated faster, and formed tubule structure more efficiently. smRNA-seq further deciphered miRNome patterns in EPCs isolated from CB or PB: 54 miRNAs were enriched in CB-EPCs, while another 50 in PB-EPCs. Specifically, CB-EPCs expressed more angiogenic miRNAs such as miR-31, while PB-EPCs possessed more tumor suppressive miRNAs including miR-10a. Knocking down miR-31 levels in CB-EPCs suppressed cell migration and microtubule formation, while overexpressing miR-31 in PB-EPCs helped to recapitulate some of CB-EPC functions.

Conclusions: Our results show the foundation for a more detailed understanding of EPCs from different anatomic sources. Stimulating the expression of angiogenic microRNAs or genes in EPCs of low activity (such as those from patients with cardiovascular diseases) might allow the development of novel therapeutic strategies.

Figure 1

Different angiogenic abilities between EPCs isolated from different anatomic locations. ( A) EPCs from cord blood (CB-EPCs) migrate faster than those from adult peripheral blood (PB-EPCs). EPCs from different sources were subjected to Transwell cell migration assays, and migrated cells were stained (representative pictures are shown) and counted (left panel, n = 3). *:P < 0.05 ( B) CB-EPCs form better microvasculature structures in vitro. EPCs were subjected onto MatriGel for tube formation assays (representative pictures are shown). Tube lengths of formed microvascular structure were counted (left panel, n = 3). *:P < 0.05 ( C) Cell proliferation assays show CB-EPCs grow faster in vitro. Cultured EPCs were subjected into MTT assays for monitoring cell proliferation rate.

Figure 2

Distinct gene expression patterns of different EPCs. ( A) A heat map showing genes more abundant in CB-EPCs. Columns represent EPC samples from different donors, while rows represent probe sets. Genes in red: increased expression; in blue: decreased. ( B) A principle component analysis (PCA) plot using genes differentially expressed between different EPCs. Each spot represents a single array sample. ( C) The unique biological functions of CB-EPCs. CB-EPC-enriched genes were subjected to a Gene Ontology (GO) database search. These categories were selected from the “Biological Process” organizing principle in the GO database (http://www.geneontology.org/). The number of genes, gene symbols, and p values for each category that are significantly enriched are listed (p < 0.05). (D) Validation of mRNA array data by qPCR. Mean gene expression levels of EPC genes were compared to the average CT values of GAPDH and beta-actin controls. Results are expressed as mean ± standard deviation. *:P < 0.05 ( E) Canonical pathways enriched in CB-EPCs according to the analysis of the Ingenuity Pathway Analysis (IPA) web tool. ( F) Schematic representation of the “FLT3 Signaling in Hematopoietic Progenitor Cells” pathway. CB-EPC genes assigned to this pathway are indicated and in red.

Figure 3

Differentially expressed miRNAs between CB- and PB-EPCs discovered by smRNA-Seq. ( A) A table summarizes reads number ( upper) and a flowchart describes the data analysis pipeline for quantification of known miRNAs from smRNA-Seq data ( lower). ( B) Differential expressed miRNAs between cord blood EPCs and adult peripheral blood EPCs. ( C-D) qPCR validation of smRNA-Seq data. Mean miRNA expression levels were compared to the average CT values of U6 snRNA + U48 snRNA + 5S rRNA controls. miRNAs more abundant in CB-EPC ( C) or PB-EPC ( D) were verified. *:P < 0.05 ( E) A major functional genetic network composed of multiple PB-EPC microRNAs (in green) and CB-EPC genes (in red). This network is displayed graphically as nodes (gene products) and edges (biological relationships between nodes) mapped by the Ingenuity Pathway Analysis (IPA) tool. The intensity of the node color indicates the degree of differential expression. Hub miRNAs in this genetic network are shown.

Figure 4

miR-31 is involved in EPC activities. ( A-C) Knocking down endogenous miR-31 in CB-EPCs. Anti-miR-31 antagomiR or the siGFP control (Ctrl) was introduced into CB-EPCs by electroporation, and 2 days later EPCs were subjected to tube formation ( A) and Transwell cell migration assays ( B). Cellular miR-31 levels were detected by RT-qPCR ( C, left panel; n = 3), and migration assay and tube formation assay results were also quantified ( C, middle and right panels; n = 3, using cells from 3 batches of donors). ( D) Overexpressing miR-31 in PB-EPCs stimulates EPC angiogenic abilities. Cellular miR-31 levels were detected by RT-qPCR (left panel; n = 3). Cellular migration assays and tube formation assays were done, and results were quantified (middle and right panels; n = 3, using cells from 3 batches of donors).