Heterogeneity of proangiogenic features in mesenchymal stem cells derived from bone marrow, adipose tissue, umbilical cord, and placenta

Abstract

Background: Mesenchymal stem cells (MSCs) have been widely proven effective for therapeutic angiogenesis in ischemia animal models as well as clinical vascular diseases. Because of the invasive method, limited resources, and aging problems of adult tissue-derived MSCs, more perinatal tissue-derived MSCs have been isolated and studied as promising substitutable MSCs for cell transplantation. However, fewer studies have comparatively studied the angiogenic efficacy of MSCs derived from different tissues sources. Here, we evaluated whether the in-situ environment would affect the angiogenic potential of MSCs.

Methods: We harvested MSCs from adult bone marrow (BMSCs), adipose tissue (AMSCs), perinatal umbilical cord (UMSCs), and placental chorionic villi (PMSCs), and studied their "MSC identity" by flow cytometry and in-vitro trilineage differentiation assay. Then we comparatively studied their endothelial differentiation capabilities and paracrine actions side by side in vitro.

Results: Our data showed that UMSCs and PMSCs fitted well with the minimum standard of MSCs as well as BMSCs and AMSCs. Interestingly, we found that MSCs regardless of their tissue origins could develop similar endothelial-relevant functions in vitro, including producing eNOS and uptaking ac-LDL during endothelial differentiation in spite of their feeble expression of endothelial-related genes and proteins. Additionally, we surprisingly found that BMSCs and PMSCs could directly form tubular structures in vitro on Matrigel and their conditioned medium showed significant proangiogenic bioactivities on endothelial cells in vitro compared with those of AMSCs and UMSCs. Besides, several angiogenic genes were upregulated in BMSCs and PMSCs in comparison with AMSCs and UMSCs. Moreover, enzyme-linked immunosorbent assay further confirmed that BMSCs secreted much more VEGF, and PMSCs secreted much more HGF and PGE2.

Conclusions: Our study demonstrated the heterogeneous proangiogenic properties of MSCs derived from different tissue origins, and the in vivo isolated environment might contribute to these differences. Our study suggested that MSCs derived from bone marrow and placental chorionic villi might be preferred in clinical application for therapeutic angiogenesis.

Keywords: Adipose tissue; Bone marrow; Heterogeneity; Mesenchymal stem cells; Placental chorionic villi; Pro-angiogenic features; Umbilical cord.

Fig. 1

Endothelial differentiation potential of different MSC populations is heterogeneous and limited. a Relative expression levels of CD31, CD34, Flt1, vWF, VE-cadherin, and Tie-2 were investigated in undifferentiated and EC-differentiated BMSCs, AMSCs, UMSCs, and PMSCs. The candidate gene expression in undifferentiated MSCs was normalized to 1, and the relative fold-change in the corresponding EC-differentiated cells was shown as 2^–ΔΔCT. BMSCs derived from two donors, and AMSCs, UMSCs, and PMSCs derived from three individuals were used. b Confocal microscope was used to investigate the expression of vWF and CD31 in EC-differentiated MSCs and HUVECs. To identify whether the EC-differentiated cells displayed similar functions to endothelial cells, immunostaining of eNOS and acLDL-uptaking assay were respectively performed. All photographs were captured at × 200 magnification. HUVECs served as positive control. MSC mesenchymal stem cells, HUVEC human umbilical vein endothelial cells, BMSC bone marrow-derived MSCs, AMSC adipose tissue-derived MSCs, UMSC umbilical cord-derived MSCs, PMSC placental chorionic villi-derived MSCs, vWF von Willebrand Factor, eNOS endothelial nitric oxide synthase, acLDL acetylated-low density lipoprotein

Fig. 2

MSCs derived from different tissue sources display distinct tube formation capacities on Matrigel in vitro. MSCs were seeded on Matrigel in vitro at 2 × 10⁴ cells/well. MSCs could form tube-like structures on Matrigel temporarily (a). Representative photographs were taken after 12 hours of incubation (scale bar = 500 μm). b Numbers of tube structures were counted and analyzed (BMSC, n = 2; AMSC, UMSC, and PMSC, n = 3). Each sample was performed in triplicate (*p < 0.05, ****p < 0.0001). MSC mesenchymal stem cells, BMSC bone marrow-derived MSCs, AMSC adipose tissue-derived MSCs, UMSC umbilical cord-derived MSCs, PMSC placental chorionic villi-derived MSCs

Fig. 3

Conditioned medium (CM) of BMSCs, AMSCs, UMSCs, and PMSCs exerts various proangiogenic effects on endothelial cells in vitro. CMs were generated from 10⁶ cells after 48 hours of incubation with EBM2. CMs supplemented with 2 % FBS, EBM2 supplemented with 2 % FBS (negative control), and EGM2-MV (containing 2 % FBS, positive control) were used to culture endothelial cells and further comparatively analyze their effects on endothelial cell proliferation. a Similar to EGM2-MV, BMSC^CM and PMSC^CM significantly promoted endothelial cell proliferation in comparison with EBM2 after 48 hours of culture; however, AMSC^CM and UMSC^CM did not produce such a promotion effect. Each sample was performed in triplicate (n = 3, *p < 0.05, **p < 0.01). To determine the bioactivity of secreted factors from MSCs on endothelial cells, an in-vitro Matrigel tube formation assay was performed. Representative photographs were taken after 9 hours of incubation (scale bar = 500 μm). b Number of tube structures (c) was counted; total tube area (d) and total tube length (e) in each well were measured and analyzed using ImageJ software. Each sample was performed in duplicate (n = 3, *p < 0.05, **p < 0.01, ***p < 0.001). MSC mesenchymal stem cells, BMSC bone marrow-derived MSCs, AMSC adipose tissue-derived MSCs, UMSC umbilical cord-derived MSCs, PMSC placental chorionic villi-derived MSCs

Fig. 4

Gene and protein expression of selected paracrine factors in different MSC populations. a Relative mRNA expression of selected angiogenic cytokines was assessed, including VEGF-A, VEGF-C, HGF, bFGF, NGF, ANG, TGF-β, IL-6, IL-8, IL-1α, IL-1β, and Cox-2. The relative expression level of the candidate cytokine in one sample of BMSCs or PMSCs was normalized to 1, and all other samples for corresponding gene expression were shown as 2^–ΔΔCT fold-change. Each sample was tested in triplicate (n = 3–5). b Concentrations of bFGF, VEGF, and HGF in 48-hour supernatants and the PGE2 concentration in 72-hour supernatants of BMSCs, AMSCs, UMSCs, and PMSCs were measured using corresponding ELISA kits (n = 3, *p < 0.05, **p < 0.01). c The relative mRNA expression of IL-8, HGF, TGF-β, IL-1α, IL-1β, IL-6, and Cox-2 was separately tested on one donor-derived UMSC and PMSC. The relative expression level of the candidate gene in UMSCs was normalized to 1, and the fold-change in PMSCs was expressed as 2^–ΔΔCT. To better show the different expression of angiogenic genes in UMSCs and PMSCs, data were shown in log₂. MSC mesenchymal stem cells, BMSC bone marrow-derived MSCs, AMSC adipose tissue-derived MSCs, UMSC umbilical cord-derived MSCs, PMSC placental chorionic villi-derived MSCs, VEGF vascular endothelial growth factor, HGF hepatocyte growth factor, bFGF basic fibroblast growth factor, NGF nerve growth factor, ANG angiogenin, TGF-β transforming growth factor beta, IL interleukin