VEGF-A, PDGF-BB and HB-EGF engineered for promiscuous super affinity to the extracellular matrix improve wound healing in a model of type 1 diabetes

Abstract

Chronic non-healing wounds, frequently caused by diabetes, lead to lower quality of life, infection, and amputation. These wounds have limited treatment options. We have previously engineered growth factors to bind to exposed extracellular matrix (ECM) in the wound environment using the heparin-binding domain of placental growth factor-2 (PlGF-2_123-144), which binds promiscuously to ECM proteins. Here, in the type 1 diabetic (T1D) NOD mouse model, engineered growth factors (eGFs) improved both re-epithelialization and granulation tissue formation. eGFs were even more potent in combination, and the "triple therapy" of vascular endothelial growth factor-A (VEGF-PlGF-2_123-144), platelet-derived growth factor-BB (PDGF-BB-PlGF-2_123-144), and heparin-binding epidermal growth factor (HB-EGF-PlGF-2_123-144) both improved wound healing and remained at the site of administration for significantly longer than wild-type growth factors. In addition, we also found that changes in the cellular milieu of a wound, including changing amounts of M1 macrophages, M2 macrophages and effector T cells, are most predictive of wound-healing success in the NOD mouse model. These results suggest that the triple therapy of VEGF-PlGF-2_123-144, PDGF-BB-PlGF-2_123-144, and HB-EGF-PlGF-2_123-144 may be an effective therapy for chronic non-healing wounds in that occur as a complication of diabetes.

Fig. 1. The triple therapy of VEGF-PlGF-2_123–144, PDGF-BB-PlGF-2_123–144, and HB-EGF-PlGF-2_123–144 improves wound healing in the T1D NOD mouse model.

Given as measured by increased granulation-tissue formation (a, c) and wound extent (diameter of unhealed wound normalized to the known length of the resection biopsy punch (12 mm) to account for folding or contracture of wound tissue (b, d)). Time points assayed are 3 days (a, b) and 7 days (c, d). For wound extent, smaller numbers indicate more healing. e Representative images of healed wounds from 7 days, stained with hematoxylin and eosin. Black arrows indicate initial wound extent, red arrows indicate healed extent. Scale bar is 1 mm. n ranges from 3 to 20. * Denotes comparison to the diabetic wound. # Denotes a comparison between untreated NOD wound and untreated NOR wound, and the WT GF(s) are compared to their counterpart -PLGF-2_123–144 variant(s), e.g., VEGF vs VEGF- PlGF-2_123–144. ^*p < 0.05, p < 0.01, p < 0.001, ANOVA + Student’s t-test for post hoc for wound extent, Kruskal-Wallis + Mann-Whitney post hoc test for granulation tissue. ^#p < 0.05, p < 0.01, p < 0.001, Mann-Whitney for granulation tissue, Student’s t-test for wound extent. Student’s t-test for comparison between VEGF-PlGF_123–144 + PDGF-PlGF_123–144 and VEGF-PlGF_123–144 + PDGF-PlGF_123–144 + HB-EGF-PlGF_123–144. Error bars are SEM.

Fig. 2. Growth factors fused with the -PLGF-2_123–144 activate their cognate receptors at levels comparable to WT GF and remain in wound tissue longer than their wild-type (WT) counterpart.

a 100 ng of VEGF-A-PlGF-2_123–144, PDGF-BB-PlGF-2_123–144, and HB-EGF-PlGF-2_123–144 caused corresponding receptor phosphorylation (VEGFR2, PDGFR, and EGFR, respectively) at the indicated amounts over 10 min on HUVEC cells (for VEGF-A165) or fibroblasts (for PDGF-BB and HB-EGF). Concentrations in wounded skin tissue (ng/mg) for (b) VEGF-A-PlGF-2_123–144, (c) PDGF-BB-PlGF-2_123–144, and (d) HB-EGF-PlGF-2_123–144 after the indicated time points. n = 7, ^*p < 0.05, Mann-Whitney. Error bars are SEM.

Fig. 3. Neutrophil (CD45+, CD11b+, CD11c−, Ly6G+) count in wounds after 3–7 days of healing.

a 3 days, b 7 days. Comparison includes unwounded skin from both the NOD and NOR mice. n ranges from 3 to 20. * Denotes comparison to the diabetic wound. # Denotes a comparison between untreated NOD wound and untreated NOR wound, and the WT GF(s) are compared to their counterpart placental growth factor (-PLGF-2_123–144) variant(s), e.g., VEGF vs VEGF- PlGF-2_123–144. ^*p < 0.05, p < 0.01, p < 0.001, Kruskal-Wallis + Mann-Whitney post hoc test. ^#p < 0.05, p < 0.01, p < 0.001, Mann-Whitney. Error bars are SEM.

Fig. 4. Macrophage cell count in wounds after 3–7 days of healing.

Inflammatory M1 (CD45+, CD11b+, CD11c−, Ly6G−, Ly6C^low, SSC^high, MHC class II+) vs alternatively activated M2 (CD45+, CD11b+, CD11c−, Ly6G−, Ly6C^high, SSC^low, CD206+) macrophages after (a) 3 days or (b) 7 days. Arginase+ alternatively activated macrophages after (c) 3 days or (d) 7 days. Arginase+ classically activated macrophages after (e) 3 days or (f) 7 days. Comparison includes unwounded skin from both the NOD and NOR mice. n ranges from 3 to 20. * Denotes comparison to the diabetic wound. # Denotes a comparison between untreated NOD wound and untreated NOR wound, and the WT GF(s) are compared to their counterpart -PlGF-2_123–144 variant(s), e.g., VEGF vs VEGF- PlGF-2_123–144. ^*p < 0.05, p < 0.01, p < 0.001, Kruskal-Wallis + Mann-Whitney post hoc test. ^#p < 0.05, p < 0.01, p < 0.001, Mann-Whitney. Error bars are SEM.

Fig. 5. Non-hematopoietic cell count in wounds after 3–7 days of healing.

Mesenchymal stem cells (CD45−, CD44+, CD29+, CD90+, SCA-1+) after (a) 3 days or (b) 7 days. Endothelial cells (CD45−, CD31+) after (c) 3 days or (d) 7 days. Smooth muscle cells CD45−, SMA+) after (e) 3 days or (f) 7 days. Comparison includes unwounded skin from both the NOD and NOR mice. n ranges from 3 to 20. * Denotes comparison to the diabetic wound. # Denotes a comparison between untreated NOD wound and untreated NOR wound, and the WT GF(s) are compared to their counterpart -PLGF-2_123–144 variant(s), e.g., VEGF vs VEGF- PlGF-2_123–144. ^*p < 0.05, p < 0.01, p < 0.001, Kruskal-Wallis + Mann-Whitney post hoc test. ^#p < 0.05, p < 0.01, p < 0.001, Mann-Whitney. Error bars are SEM.

Fig. 6. CD4+ T cell count in wounds after 3–7 days of healing.

CD4+ T cells (CD45+, CD3+, CD4+) at (a) 3 days or (c) 7 days. Subset of CD4+ T cells that are amphiregulin positive at (b) 3 days or (d) 7 days. Comparison includes unwounded skin from both the NOD and NOR mice. n ranges from 3 to 20. * Denotes comparison to the diabetic wound. # Denotes a comparison between untreated NOD wound and untreated NOR wound, and the WT GF(s) are compared to their counterpart -PLGF-2_123–144 variant(s), e.g., VEGF vs VEGF- PlGF-2_123–144. ^*p < 0.05, p < 0.01, p < 0.001, Kruskal-Wallis + Mann-Whitney post hoc test. ^#p < 0.05, p < 0.01, p < 0.001, Mann-Whitney. Error bars are SEM.

Fig. 7. Effector T cell count in wounds after 3–7 days of healing.

Effector T cells (CD45+, CD3+, CD44+, CD62L−) at (a) 3 days or (c) 7 days. Subset of effector T cells that are amphiregulin positive at (b) 3 days or (d) 7 days. Comparison includes unwounded skin from both the NOD and NOR mice. n ranges from 3 to 20. * Denotes comparison to the diabetic wound. # Denotes a comparison between untreated NOD wound and untreated NOR wound, and the WT GF(s) are compared to their counterpart -PLGF-2_123–144 variant(s), e.g., VEGF vs VEGF-PlGF-2_123–144. ^*p < 0.05, p < 0.01, p < 0.001, ANOVA + Student t-test for post hoc. ^#p < 0.05, p < 0.01, p < 0.001, Student’s t-test. Error bars are SEM.

Fig. 8. T regulatory cell count in wounds after 3–7 days of healing.

T-regulatory cells (CD45+, CD3+, CD4+, CD25 high, Foxp3+) at (a) 3 days or (c) 7 days. Subset of effector T cells that are amphiregulin positive at (b) 3 days or (d) 7 days. Comparison includes unwounded skin from both the NOD and NOR mice. n ranges from 3 to 20. * Denotes comparison to the diabetic wound. # Denotes a comparison between untreated NOD wound and untreated NOR wound, and the WT GF(s) are compared to their counterpart -PLGF-2_123–144 variant(s), e.g., VEGF vs VEGF-PlGF-2_123–144. ^*p < 0.05, p < 0.01, p < 0.001, Kruskal-Wallis + Mann-Whitney post hoc test. ^#p < 0.05, p < 0.01, P < 0.001, Mann-Whitney. Error bars are SEM.