Compromised angiogenesis and vascular Integrity in impaired diabetic wound healing

Abstract

Vascular deficits are a fundamental contributing factor of diabetes-associated diseases. Although previous studies have demonstrated that the pro-angiogenic phase of wound healing is blunted in diabetes, a comprehensive understanding of the mechanisms that regulate skin revascularization and capillary stabilization in diabetic wounds is lacking. Using a mouse model of diabetic wound healing, we performed microCT analysis of the 3-dimensional architecture of the capillary bed. As compared to wild type, vessel surface area, branch junction number, total vessel length, and total branch number were significantly decreased in wounds of diabetic mice as compared to WT mice. Diabetic mouse wounds also had significantly increased capillary permeability and decreased pericyte coverage of capillaries. Diabetic wounds exhibited significant perturbations in the expression of factors that affect vascular regrowth, maturation and stability. Specifically, the expression of VEGF-A, Sprouty2, PEDF, LRP6, Thrombospondin 1, CXCL10, CXCR3, PDGFR-β, HB-EGF, EGFR, TGF-β1, Semaphorin3a, Neuropilin 1, angiopoietin 2, NG2, and RGS5 were down-regulated in diabetic wounds. Together, these studies provide novel information about the complexity of the perturbation of angiogenesis in diabetic wounds. Targeting factors responsible for wound resolution and vascular pruning, as well those that affect pericyte recruitment, maturation, and stability may have the potential to improve diabetic skin wound healing.

Fig 1. Diabetic wounds show altered 2-D and 3-D vascular architecture.

Endothelial cells were identified by CD31 staining in baseline uninjured skin (day 0) and skin wounds at days 7, 10, 14, 18 and 22 post-wounding. A) Representative fluorescence photomicrographs of CD31 staining in the wounds. Scale bar = 100μm. B). Wound vascularity. The percent area of the wound edge and wound bed occupied by CD31-positive staining. Mean ±SD, n = 5. #p<0.01. C). MicroCT images of blood vessels in day 10 wounds of WT and diabetic mice. Scale bar = 1mm. D-I). Total vessel surface area, number of branch junctions, total branch length in (UNIT OF MEASUREMENT), total number of branches, total vessel volume, and the number of branches grouped by branch length, respectively, based on microCT image analysis. *p<0.05. n = 4 in WT group and 6 in diabetic group.

Fig 2. Diabetic wounds show increased amounts of vessel leakage and decreased pericyte coverage.

A). Representative fluorescence photomicrographs of CD31- and DAPI-stained FITC-dextran perfused wounds. Scale bar = 100μm. B). The percent of extravascular FITC-dextran. *p>0.05, n = 5. C). The percent of intravascular FITC-dextran. n = 5. D). Comparative levels of endothelial cell—pericyte association, expressed as the percent of CD31+ vessels not colocalized with desmin, p<0.01, n = 5. E). Representative fluorescence photomicrographs of CD31- and desmin-stained uninjured skin (US), days 10 and 18 wounds. Scale bar = 100μm.

Fig 3

A-G) Semi-quantitative RT-PCR analysis of pro- and anti-angiogenic factors over the time course of wound healing. n = 5, *p<0.05, #p<0.01. H) Heatmap summarizing findings of A-G, showing genes with no difference (gray boxes)significant downregulation (white boxes) or upregulation (black boxes) over the time course of healing in db/db versus WT mice. In the figure, factors are identified using common protein name abbreviations; for gene symbols see Table 3.

Fig 4

A-L) Semi-quantitative RT-PCR analysis of capillary pericyte recruitment and stabilization factors in wounds over the time course of wound healing. n = 5, *p>0.05, #p<0.01. M) Heatmap summarizing findings of A-L, showing genes with no difference (gray boxes) significant downregulation (white boxes) or upregulation (black boxes) over the time course of healing in db/db versus WT mice. In the figure, factors are identified using common protein name abbreviations; for gene symbols see Table 3.

Fig 5

A-D) Semi-quantitative RT-PCR analysis of pericyte markers in mouse diabetic wounds over the time course of wound healing. n = 5, *p>0.05, #p<0.01. E) Heatmap summarizing findings of A-D, showing genes with no difference (gray boxes) significant downregulation (white boxes) or upregulation (black boxes) over the time course of healing in db/db versus WT mice. In the figure, factors are identified using common protein name abbreviations; for gene symbols see Table 3.