Derivation and Characterization of Endothelial Cells from Porcine Induced Pluripotent Stem Cells

Abstract

Although the study on the regulatory mechanism of endothelial differentiation from the perspective of development provides references for endothelial cell (EC) derivation from pluripotent stem cells, incomplete reprogramming and donor-specific epigenetic memory are still thought to be the obstacles of iPSCs for clinical application. Thus, it is necessary to establish a stable iPSC-EC induction system and investigate the regulatory mechanism of endothelial differentiation. Based on a single-layer culture system, we successfully obtained ECs from porcine iPSCs (piPSCs). In vitro, the derived piPSC-ECs formed microvessel-like structures along 3D gelatin scaffolds. Under pathological conditions, the piPSC-ECs functioned on hindlimb ischemia repair by promoting blood vessel formation. To elucidate the molecular events essential for endothelial differentiation in our model, genome-wide transcriptional profile analysis was conducted, and we found that during piPSC-EC derivation, the synthesis and secretion level of TGF-β as well as the phosphorylation level of Smad2/3 changed dynamically. TGF-β-Smad2/3 signaling activation promoted mesoderm formation and prevented endothelial differentiation. Understanding the regulatory mechanism of iPSC-EC derivation not only paves the way for further optimization, but also provides reference for establishing a cardiovascular drug screening platform and revealing the molecular mechanism of endothelial dysfunction.

Keywords: TGF-β; angiogenesis; endothelial cells; hindlimb ischemia; pig.

Figure 1

Generation and characterization of piPSC-ECs. (A) The schematic diagram for piPSC-EC derivation. (B) Morphology of piPSC-ECs. (C) Tube-like structures were formed by piPS-ECs. (D) Uptake of Dil-Ac-LDL by piPSC-ECs. (E–G) Endothelial-specific markers were detected via immunofluorescence staining. piPSC-ECs were positive for CD144 (E), eNOS (F) and vWF (G). Blue fluorescence represents the nuclei labeled by Hoechst33342. Scale bar = 50 μm.

Figure 2

piPSC-ECs cultured on 3D gelfoam scaffolds. (A–D) Confocal images of Dio-labeled piPSC-ECs on gelfoam scaffolds after 1 day (A-A’,B-B′) or 7 days of cultivation (C-C′,D-D′). (B-B′,D-D′) are the zoomed-in versions of (A-A′,C-C′) within the white rectangles. Scale bar = 50 µm. (E) Cell density assay was performed at the indicated treatments and time points. Mean ± SD. n = 4. The bars without the same letter are significantly different (p < 0.05).

Figure 3

Histomorphology and angiogenesis of adductors after cell transplantation. (A–C) Histomorphological observation of mice treated with EGM-2 (A), PFFs (B) and porcine iPS-ECs (C). Scale bar = 100 μm. (D–O) Angiogenesis of mice treated with EGM-2 (D–G), PFFs (H–K) and porcine iPSC-ECs (L–O). Blue fluorescence represents the nuclei labeled by Hoechst33342 (D,H,L). Green fluorescence represents CD31 expression (E,I,M). Red fluorescence represents porcine-specific vimentin expression (F,J,N). Merged images of (E,F), (I,J), and (M,N) (G,K,O). Scale bar = 100 μm. (P,Q) Quantitative analysis of angiogenesis in different groups. Total capillary density (P). Percentage of capillary from porcine cells (Q). Mean ± SD, n = 6. The bars labeled with an asterisk are significantly different (p < 0.05).

Figure 4

Genome−wide transcriptional profile analysis for the regulation of piPSC−EC differentiation. (A) Hierarchical clustering analysis of the genes differentially expressed between iPS cells and AOCs at p < 0.001, q < 0.001, |log2 fold change| > 3. Columns indicate samples and rows indicate genes. Blue indicates repression and red indicates promotion. (B) Principal component analysis (PCA) with the differentially expressed genes using R package ade4 (version 1.7-18). (C) The shift in each PC during the process of differentiation. A relative coordinate was calculated by normalizing the coordinates of all cells within each PC with a range from 0 to 1. The results are presented as the mean ± SD, n = 3. (D) Trajectory visualization in 2D using any two of the top four PCs. (E) Heatmap of the high contribution genes of the top four PCs. Columns indicate PCs and rows indicate genes. Brown indicates high contribution and white indicates low contribution. Pathway analysis of indicated genes was performed (p < 0.05).

Figure 5

Effects of TGF-β-Smad2/3 signaling on the differentiation of piPSC-ECs. (A) Detection of secretion levels of TGF-β1 by ELISA. Mean ± SD. n = 4. (B) Detection of expression levels of TGF-β by Western blot. (C) Detection of Smad2/3 phosphorylation levels by Western blot. (D,E) Expression levels of mesoderm marker genes (D) and endothelial marker genes (E) at indicated time points by RT-PCR. Mean ± SD. n = 4. The bars without the same letter are significantly different (p < 0.05). (F) Flow cytometric analysis of CD31-positive cells in different treatment groups on day 12.