Improved survival, vascular differentiation and wound healing potential of stem cells co-cultured with endothelial cells

Abstract

In this study, we developed a methodology to improve the survival, vascular differentiation and regenerative potential of umbilical cord blood (UCB)-derived hematopoietic stem cells (CD34(+) cells), by co-culturing the stem cells in a 3D fibrin gel with CD34(+)-derived endothelial cells (ECs). ECs differentiated from CD34(+) cells appear to have superior angiogenic properties to fully differentiated ECs, such as human umbilical vein endothelial cells (HUVECs). Our results indicate that the pro-survival effect of CD34(+)-derived ECs on CD34(+) cells is mediated, at least in part, by bioactive factors released from ECs. This effect likely involves the secretion of novel cytokines, including interleukin-17 (IL-17) and interleukin-10 (IL-10), and the activation of the ERK 1/2 pathway in CD34(+) cells. We also show that the endothelial differentiation of CD34(+) cells in co-culture with CD34(+)-derived ECs is mediated by a combination of soluble and insoluble factors. The regenerative potential of this co-culture system was demonstrated in a chronic wound diabetic animal model. The co-transplantation of CD34(+) cells with CD34(+)-derived ECs improved the wound healing relatively to controls, by decreasing the inflammatory reaction and increasing the neovascularization of the wound.

Figure 1. Protein expression in undifferentiated CD34⁺ cells and their endothelial progenies.

A) Immunofluorescence results for undifferentiated CD34⁺ cells. Bar corresponds to 20 µm. B, C) FACS analysis of (B) undifferentiated CD34⁺ cells and (C) CD34⁺-derived ECs at passage 1 (1^st P) and 3 (3^rd P). Percentages of positive cells were calculated based on the isotype controls (grey plot) and are shown in the histogram plots.

Figure 2. Characterization of CD34⁺-derived ECs.

A) Immunofluorescence analysis. Bar corresponds to 40 µm. B) Quantitative RT-PCR analysis. Results are average ± SD, n = 4. C) Quantification of cord length and number of sprouts at 15 h and 48 h. Counts were performed using a magnification of ×100. Results are average ± SD, n = 3. *, **, *** denote statistical significance (P<0.05, P<0.01, P<0.001, respectively).

Figure 3. Adhesion of CD34⁺ cells to different substrates.

CD34⁺ cells were seeded on 24-well plates coated with fibrin gel, type I collagen gel (2.5 mg/mL) and fibronectin (50 µg/mL), and incubated for 3 h at 37°C. Polystyrene culture wells were used as control. After that time, the cells were washed and the attached ones were counted in six random microscope fields (×200 magnification) for each replicate. Results are average ± SD, n = 6. * and ** denote statistical significance (P<0.05 and P<0.01, respectively).

Figure 4. Viability of encapsulated cells.

A) Viability of encapsulated cells, as assessed by a LIVE/DEAD assay. Results are average ± SD, n = 3. B) Mitochondrial metabolic activity of encapsulated cells. Results are average ± SD, n = 6. Each absorbance at 540 nm was normalized by day 2 absorbance. C) Viability of encapsulated cells, as quantified by a LIVE/DEAD assay. Results are average ± SD, n = 3. In all figures, * denote statistical significance within time group: * P<0.05, **P<0.01, *** P<0.001. # denotes statistical significance between time groups comparing the respective control/treatment groups: # P<0.05, ## P<0.01. D) Phosphorylated Akt/total Akt and phosphorylated ERK/total ERK ratios assessed by a Bio-Plex phosphoprotein detection assay. Results are average ± SD, n = 3.

Figure 5. Multiplex cytokine analysis.

Analysis of cytokines secreted by (A) CD34⁺-derived ECs grown on tissue culture polystyrene or encapsulated in fibrin gels for 2 or 10 days, (B) CD34⁺ cells grown on tissue culture polystyrene or encapsulated in fibrin gels for 2 or 10 days and (C) co-culture of CD34⁺ and CD34⁺-derived ECs encapsulated in fibrin gels for 2 or 10 days. Cell media were collected after being in contact with the cells for 2 days. Results are average ± SD, n = 3.

Figure 6. Viability and differentiation of CD34⁺ cells cultured on top of fibrin gels and gene expression analysis of encapsulated cells.

A) Apoptosis and necrosis of CD34⁺ cells cultured for 7 days on top of fibrin gels, in EGM-2 medium or EGM-2 medium conditioned by CD34⁺-derived ECs. Results are average ± SD, n = 3 (magnification of ×100). B) CD34⁺ cell differentiation on top of fibrin gels for 7 days in the absence (B.1) or in the presence (B.2) of CD34⁺-derived ECs. B.3 shows the percentage of cells able to uptake Ac-LDL. Results are average ± SD, n = 18. C) Quantitative RT-PCR analysis of CD34⁺, CD34⁺-derived ECs, or a co-culture of both cells, encapsulated in fibrin gels for 10 days. Results are average ± SD, n = 4. In all figures, * denote statistical significance: * P<0.05, **P<0.01, *** P<0.001.

Figure 7. Regenerative effect of encapsulated CD34⁺ cells and CD34⁺-derived ECs on diabetic wounds.

A) Wound closure (relatively to initial wound area) in diabetic mice treated by topical application of 1×10⁵ CD34⁺ cells and 0.35×10⁵ CD34⁺-derived ECs (A.1), CD34⁺ cells (A.2) or CD34⁺-derived ECs (A.3), encapsulated in fibrin gels. Control wounds received a saline solution (PBS) only. Results are average ± SEM, n = 6. * denotes statistical significance (P<0.05). B) Representative images of vWF immunostaining of wounds at day 3. Scale bar represents 50 µm. C) Quantification of wound capillary density (number of capillaries per mm²), based on the vWF immunostaining results, for 3 and 10 day-old wounds in the different experimental groups. Results are average ± SEM, n = 3. ** denotes statistical significance (P<0.001). D) Cytokine expression on mouse wound samples excised 3 days post-wounding, as determined by a Bio-Plex mouse cytokine assay. Wounds had been treated by topical application of CD34⁺ cells or CD34⁺ cells plus CD34⁺-derived ECs encapsulated in fibrin gels. Control wounds received fibrin gel alone. Results are average ± SD, n = 3.