Isolation of CD248-expressing stromal vascular fraction for targeted improvement of wound healing

Abstract

Wound healing remains a global issue of disability, cost, and health. Addition of cells from the stromal vascular fraction (SVF) of adipose tissue has been shown to increase the rate of full thickness wound closure. This study aimed to investigate the angiogenic mechanisms of CD248+ SVF cells in the context of full thickness excisional wounds. Single cell transcriptional analysis was used to identify and cluster angiogenic gene-expressing cells, which was then correlated with surface marker expression. SVF cells isolated from human lipoaspirate were FACS sorted based on the presence of CD248. Cells were analyzed for angiogenic gene expression and ability to promote microvascular tubule formation in vitro. Following this, 6mm full thickness dermal wounds were created on the dorsa of immunocompromised mice and then treated with CD248+, CD248-, or unsorted SVF cells delivered in a pullalan-collagen hydrogel or the hydrogel alone. Wounds were measured every other day photometrically until closure. Wounds were also evaluated histologically at 7 and 14 days post-wounding and when fully healed to assess for reepithelialization and development of neovasculature. Wounds treated with CD248+ cells healed significantly faster than other treatment groups, and at 7 days, had quantitatively more reepithelialization. Concurrently, immunohistochemistry of CD31 revealed a much higher presence of vascularity in the CD248+ SVF cells treated group at the time of healing and at 14 days post-op, consistent with a pro-angiogenic effect of CD248+ cells in vivo. Therefore, using CD248+ pro-angiogenic cells obtained from SVF presents a viable strategy in wound healing by promoting increased vessel growth in the wound.

Figure 1

(A) Heat maps obtained from single cell transcriptional analysis show clustering based on pro-angiogenic genes (VEGF, FGF2, PDGFRA, and PDGFRB). (B) Linear discriminant analysis revealed CD248 as the marker whose expression most significantly correlated with cluster identification. (C) Flow cytometry plot shows prevalence of CD248 positive cells obtained from SVF (79.1% negative, 14.8% positive).

Figure 2

(A) qRT-PCR results of HGF and VEGF reveal a significant upregulation of both genes in the CD248+ populations when compared to CD248- and unsorted groups. (*p < 0.05, **p < 0.01). (B) Micrographs show results from endothelial tubule formation assay, with exogenous VEGF 10ng/ml alone serving as a positive control. Top row shows tubules stained with calcein AM, bottom row shows the computed quantities of vessel formation. (C) Graphs show quantification of the stained tubules. CD248+ cells show highest percent mesh area (*p < 0.05), and highest number of master junctions and segments (*p < 0.05).

Figure 3

(A) Representative images show photometric analysis of full thickness excisional wounds across 4 groups (CD248+, CD248-, unsorted, and hydrogel alone). Black boxes indicate wound closure. (B) Quantification of wound area was performed using Image J. Graph represents numeric value of wound area across the 4 test groups from day 0 until time of wound closure. CD248+ cells are seen to accelerate wound healing significantly compared to CD248-, unsorted cells, and hydrogel alone on days 7, 9, and 11. CD248+ treated wounds healed by day 13, CD248- and unsorted cell-treated wounds healed by day 15, and hydrogel alone-treated wounds healed by day 16 (*p < 0.05 for CD248+ vs. all other groups at days 7, 9, and 11).

Figure 4

A) Immunohistochemical stains of wounds for VEGF (red), CD248 (green), and DAPI (blue), with merged image on right. Scale bar = 100µm. (B) Quantification of immunofluorescent images (*p < 0.05).

Figure 5

(A) CD31 immunohistochemical staining of wounds excised 7 days post-op. Scale bar = 100 µm. (B) H&E images of full wounds. Dotted lines represent wound borders, and solid perpendicular lines highlight length of epithelial cover. Re-epithelialization is expressed as a percentage of the distance between the dotted lines. (C) Quantification of re-epithelialization (top), and quantification of CD31 staining (bottom). (*p < 0.05).

Figure 6

(A) Haematoxylin and eosin staining of excised wounds at day 14 (top row). Scale bar = 80µm. Anti-CD31 (red) immunofluorescent staining (bottom row) Scale bar = 100 µm. (B) Quantification of CD31 immunofluorescent images shown in 6A revealed significantly more staining in CD248+ treated wounds compared to other groups (*p < 0.05). (C) H&E staining at completion of wound healing (top row). Scale bar = 80 µm. Anti-CD31 (red) immunofluorescent staining (bottom row). Scale bar = 100µm. (D) Quantification of CD31 immunofluorescent images shown in 6C demonstrated more nascent vessels in the wounds of CD248+ group when compared to the other 3 test groups. (*p < 0.05).