Restoration of the healing microenvironment in diabetic wounds with matrix-binding IL-1 receptor antagonist

Abstract

Chronic wounds are a major clinical problem where wound closure is prevented by pathologic factors, including immune dysregulation. To design efficient immunotherapies, an understanding of the key molecular pathways by which immunity impairs wound healing is needed. Interleukin-1 (IL-1) plays a central role in regulating the immune response to tissue injury through IL-1 receptor (IL-1R1). Generating a knockout mouse model, we demonstrate that the IL-1-IL-1R1 axis delays wound closure in diabetic conditions. We used a protein engineering approach to deliver IL-1 receptor antagonist (IL-1Ra) in a localised and sustained manner through binding extracellular matrix components. We demonstrate that matrix-binding IL-1Ra improves wound healing in diabetic mice by re-establishing a pro-healing microenvironment characterised by lower levels of pro-inflammatory cells, cytokines and senescent fibroblasts, and higher levels of anti-inflammatory cytokines and growth factors. Engineered IL-1Ra has translational potential for chronic wounds and other inflammatory conditions where IL-1R1 signalling should be dampened.

Fig. 1. IL-1R1 signalling delays wound healing in diabetic mice.

a Full-thickness wounds (5 mm) were created in diabetic mice (Lepr^db/db) and non-diabetic littermates (Lepr^db/+). Concentrations of IL-1β and IL-1Ra in wounds harvested at various time points. n = 4 wounds per time point. b Concentrations of IL-1β detected in media of unstimulated or stimulated (LPS + ATP) bone marrow-derived macrophages (MΦs) from Lepr^db/+ and Lepr^db/db mice. c Lepr^db/db mice were crossed with Il1r1^−/− mice to generate diabetic mice deficient for IL-1R1. Representative pictures of 14-week-old mice are shown. Scale bar = 1 cm. d, e Full-thickness wounds were created in Lepr^db/db and Lepr^db/db-Il1r1^−/− mice. Graph in d shows wound closure kinetics evaluated by histomorphometric analysis of tissue sections. n = 12 wounds per time point. Representative histology (haematoxylin and eosin staining) after 9 days are shown in (e). Black arrows indicate wound edges and red arrows indicate tips of epithelium tongue. The epithelium (if any) appears in purple as a homogeneous keratinocyte layer on top of the wounds. The granulation tissue under the epithelium contains granulocytes with dark-purple nuclei. Fat tissue appears as transparent bubbles. Scale bar = 1 mm. For (a), (b) and (c), data are means ± SEM. For (a) and (d), two-way ANOVA with Bonferroni post hoc test for pair-wise comparisons. For (b), two-tailed Student’s t test. **P ≤ 0.01, ***P ≤ 0.001.

Fig. 2. PlGF_123–141-fused IL-1Ra and PDGF-BB strongly bind ECM and are retained in skin tissue.

a Amino-acid sequences of PlGF fragments fused to glutathione S-transferase (GST). b ELISA plates were coated with ECM proteins and incubated with PlGF fragments. Graphs show signals given when detecting GST. PlGF_123–141 is shown in red. n = 4. c PlGF_123–141 (in red) was added to the C terminus of IL-1Ra (in pink) or PDGF-BB (in blue) to generate IL-1Ra/PlGF_123–141 and PDGF-BB/PlGF_123–141. PDGF-BB occurs as a dimer. d Schematic representation of the ECM-mimetic hydrogel and skin endogenous ECM. Fg fibrinogen, Fn fibronectin, Vn vitronectin, TnC tenascin C, HS heparan sulfate. e ECM-mimetic hydrogels were generated with IL-1Ra or PDGF-BB variants and incubated in ten times volume of buffer (with or without plasmin) that was changed every 24 h. The graphs show the cumulative release of IL-1Ra or PDGF-BB variants in buffer. n = 4. f The percentage of IL-1Ra and PDGF-BB variants remaining in skin after intradermal injection was measured at various time points. n = 4 per time point. For (b), (e) and (f), data are means ± SEM. For (b), one-way ANOVA with Bonferroni post hoc test for pair-wise comparisons. For (e) and (f), two-way ANOVA with Bonferroni post hoc test for pair-wise comparisons. *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001.

Fig. 3. Super-affinity IL-1Ra accelerates wound healing in diabetic mice.

a, b Full-thickness wounds in Lepr^db/db were treated with IL-1Ra or PDGF-BB variants (0.5 μg of wild types, equimolar of engineered versions). Wound closure evaluated by histomorphometric analysis of tissue sections in (a). n = 8 wounds per condition. Representative histology (haematoxylin and eosin staining) 9 days post treatment shown in (b). Black arrows indicate wound edges and red arrows indicate tips of epithelium tongue. The epithelium (if any) appears in purple as a homogeneous keratinocyte layer on top of the wounds. The granulation tissue under the epithelium contains granulocytes with dark-purple nuclei. Fat tissue appears as transparent bubbles. Scale bar = 1 mm. c, d Angiogenesis at day 9 assessed by immunostaining of wound sections for CD31 (endothelial cells, red) and desmin (smooth muscle cells, green). Representative images in (c). Dashed lines indicate separation between epidermis (indicated as e) and dermis (indicated as d). Scale bar = 0.2 mm. Quantification of CD31 in (d). n = 8. e Schematic representation of hypothesised IL-1Ra/PlGF_123–141 effects in diabetic wounds. MΦs macrophages, MMPs matrix metallopeptidases, SASP senescence-associated secretory phenotype; grey arrows represent induction; blue lines represent inhibition. f Neutrophil and macrophage populations in wounds measured by flow cytometry at various time points post wounding. Percentages were calculated over total live wound cells. Median fluorescence intensity (MFI) for CD206 was measured in macrophages (F4/80⁺, CD11b⁺ cells). n = 8 wounds. In (a), (d) and (f), data are means ± SEM. One-way ANOVA with Bonferroni post hoc test for pair-wise comparisons (significances shown are between saline and the other groups, unless indicated otherwise). *P ≤ 0.05, **P ≤ 0.001, ***P ≤ 0.001. ns non-significant.

Fig. 4. Super-affinity IL-1Ra leads to a pro-healing microenvironment.

a Full-thickness wounds were created in Lepr^db/db mice and treated with saline control or IL-1Ra variants (0.5 μg of wild type, equimolar IL-1Ra/PlGF_123–141). After 6, 9 and 12 days wound concentrations of cytokines, MMPs, TIMP-1 and growth factors were measured by ELISA. The heat map shows fold change in log₂ over treatment with saline. n = 4. b Dermal fibroblasts were cultured with IL-1β (1 ng/ml) or saline control. After 9 days, SA-β-gal activity was assessed using senescence green probe (SGP). Graph shows median fluorescence intensity (MFI). n = 6. c Full-thickness wounds in Lepr^db/db mice were treated with saline or IL-1Ra variants. After 9 days, SA-β-gal activity in wound fibroblast was assessed by flow cytometry using SGP MFI. SA-β-gal activities were compared to fibroblasts in non-injured skin samples. n = 4. For all panels, data are means ± SEM. For (a), two-way ANOVA with Bonferroni post hoc test for pair-wise comparisons between IL-1Ra and IL-1Ra/PlGF_123–141. For (b), two-tailed Student’s t test. For (c), one-way ANOVA with Bonferroni post hoc test for pair-wise comparisons. * P ≤ 0.05, ** P ≤ 0.01, ***P ≤ 0.001. ns non-significant.