Systemic Delivery of Anti-Integrin αL Antibodies Reduces Early Macrophage Recruitment, Inflammation, and Scar Formation in Murine Burn Wounds

Abstract

Objective: Increased macrophage recruitment in the early stages of wound healing leads to an excessive inflammatory response associated with elevated fibrosis and scarring. This recruitment relies upon integrins on the surface of monocytes that regulate their migration and extravasation from the circulation into the wound site, where they differentiate into macrophages. The aim of this study was to determine if inhibiting monocyte extravasation from the circulation into burns would reduce macrophages numbers in burns and lead to reduced inflammation and scar formation.

Approach: Scald burns were created on mice and treated with integrin alpha L (αL) function blocking antibody via intravenous delivery day 1 after injury. The effect of inhibiting macrophage recruitment into the burn was assessed using macro- and microscopic wound parameters as well as immunohistochemistry for inflammatory cell markers, cytokines, and collagen deposition.

Results: Burn wound-associated macrophages were reduced by 54.7% at day 3 following treatment with integrin αL antibody, with levels returning to normal by day 7. This reduction in macrophages led to a concomitant reduction in inflammatory mediators, including tumor necrosis factor-alpha (TNFα) and Il-10 as well as a reduction in proscarring transforming growth factor beta 1 (TGFβ1). This reduced inflammatory response was also associated with less alpha smooth muscle actin (αSMA) expression and an overall trend toward reduced scar formation with a lower collagen I/III ratio.

Innovation: Treatment of burns with integrin αL function blocking antibodies reduces inflammation in burn wounds.

Conclusion: These results suggest that reducing macrophage infiltration into burn wounds may lead to a reduced early inflammatory response and less scar formation following burn injury.

Allison J. Cowin, PhD

Introduction

Despite advances in burns research, management of burn injuries remains a challenge and there are limited options available for improving healing and reducing scar formation.¹ While inflammation is a key stage in normal tissue repair, prolonged or excessive inflammation contributes to increased fibrosis and scarring.² Burn injuries stimulate the rapid recruitment of circulating immune cells, including neutrophils and monocytes, into the burn wound site.^3,4 Once in the wound, monocytes differentiate into macrophages that regulate a wide range of processes that are necessary for repair, including the deposition of new extracellular matrix, wound vascularization and closure, by releasing cytokines and growth factors such as transforming growth factor beta 1 (TGFβ1).^5,6 They also recruit more immune cells leading to an amplification of the inflammatory response.⁷ In the early stages of repair, macrophages predominantly present with a proinflammatory M1 phenotype and are prolific secretors of proinflammatory cytokines, including tumor necrosis factor-alpha (TNFα) and interleukin-6 (IL-6).^8,9 In the later stages of repair, M2 macrophages predominate and secrete anti-inflammatory cytokines that shut down the inflammatory response, suggesting that macrophages play different roles in repair at distinct time points.

A number of studies using mice that lack or have greatly depleted levels of macrophages show that they are key regulators of fibrosis and scar formation.^10–12 Studies targeting the depletion of macrophages at distinct stages (early, mid, and late) of the repair process, using a diphtheria toxin-driven lysozyme M-specific ablation protocol, showed that macrophages play different roles at distinct stages of wound healing.¹³ Importantly, it was found that proinflammatory macrophages present in the early stages of repair (days 1–5) are responsible for regulating the degree of scar formation. Although a significant delay in reepithelialization was seen, repopulation of the wound with macrophages rescued this delay suggesting that reducing the level of early wound macrophages might produce a more favorable wound outcome in terms of scar formation. The same diphtheria toxin-driven lysozyme M-specific ablation strategy cannot be used in human patients and so new strategies to reduce macrophage numbers are required.

Monocyte tissue infiltration from blood requires the cells to adhere, spread, laterally migrate, and cross the endothelial barrier to reach the site of injury.¹⁴ Integrins located on the surface of cells regulate this adhesion and migration of leukocytes during extravasation from the blood and their subsequent migration into sites of inflammation.¹⁵ To date, 18α and 8β integrin subunits have been identified, which can form 24 distinct α-β subunit heterodimer permutations. These integrin complexes are differentially expressed depending on the cell type and some are immune specific.¹⁶ In particular, the lymphocyte function-associated antigen-1 (LFA-1) complex consisting of integrin alpha L (αL) and integrin beta 2 (β2) is found only on immune cells and is responsible for monocyte adherence, migration, and extravasation from blood vessels.¹⁷ In diseases such as psoriasis, immune cell recruitment to skin is reduced in patients administered anti-integrin αL function blocking antibodies.¹⁸ This has led to our hypothesis that treatment of burns with anti-integrin αL antibodies would lead to reduced immune cell recruitment into the site of injury leading to reduced numbers of macrophages, inflammation, and scar formation.

Clinical Problem Addressed

Severe burn injuries can lead to excessive fibrosis and scar formation. Inflammation is a major contributing factor to scarring in burns patients and strategies aimed at reducing excessive immune responses are likely to improve healing outcomes. Integrins regulate the adhesion and migration of leukocytes from the blood and into sites of injury. The integrin αL function blocking antibody, Efalizumab, has previously been used clinically to reduce inflammation in people with psoriasis. Using a mouse model of scald burn injury, the effect of integrin αL antibody treatment on inflammation and healing was assessed. While there are distinct differences between mouse burn wound repair and the human situation, these studies provide the first evidence that reducing monocyte extravasation from blood vessels into burn injuries reduces inflammatory responses and is an important first step toward the development of new approaches to improve burn injury repair.

Materials and Methods

Human studies

Human monocytes and macrophages were purified from the blood and burns of five patients 9 days postburn injury, respectively, with approval from Westmead Children's Hospital Human Ethics Committee and in compliance with the ethical rules for human experimentation that are stated in the 1975 Declaration of Helsinki, as per established protocols.¹⁹ The expression of integrin αL (CD11a) was detected using western analysis using Anti-Human CD11a antibody (610826, BD Biosciences, San Jose, CA). Quantification of bands was performed using ImageJ software (U.S. National Institutes of Health, Bethesda, MD).

Animal studies

All animal procedures were approved by the Women's and Children's Health Network Animal Ethics Committee (AE1025/2/2018) and carried out in accordance with the Australian code of practice for the care and use of animals for scientific purposes. For the burn studies, 10-week-old male BALB/c mice (Animal Resource Centre, Perth, Australia) were randomly assigned to treatment and timepoint groups (n = 8/group) and subjected to a partial thickness scald burn injury according to previously published protocols.²⁰ In brief, circular partial thickness scald burns were induced via exposure of 69 mm² of the shaved dorsum to 65°C circulating water for 45 s followed by immersion of the burn in a cold-water bath for 45 s to stop the burning process. Integrin αL function-blocking antibody (anti-αL: 50 μg in PBS (functional grade purified rat anti-mouse CD11a M17/4 ES-16-0111-85, eBioscience, San Diego, CA) or isotype-matched immunoglobulin control (IgG: 50 μg IgG in PBS (functional grade purified rat IgG2a κ ES-16-4321-85, eBioscience) were administered via a single 50 μL injection into the tail vein of the mice 24 h after scald burn injury. This integrin αL function-blocking M17/4 clone has previously been used to functionally block LFA-1 (integrin αL and integrin β2) and has been characterized extensively in the literature.^21,22 At day 3, 7, 14, or 28 postburn injury, the mice were euthanized, digital images acquired, and the burn and surrounding skin surgically excised to the fascia and one half of each wound fixed in formalin for histological assessment of burn wound closure over time. Four micrometers sections were cut from paraffin-embedded formalin-fixed tissue samples and subjected to hemotoxylin and eosin staining to visualize the skin structures and measure the distance between normal dermal tissue (overall wound length) and the epidermal edges (epithelial gap). All analyses were conducted by assessors who were blinded to the treatments.

Burn wound assessment

Burns were analyzed using morphometric analysis of using digital images taken under normal light at each time point. The appearance of the scars at day 28 was assessed using a semiquantitative macroscopic ranking analysis by eight independent assessors as previously described.²³ In brief, burn wound images were arranged in a PowerPoint presentation and assessors were blinded to which images represented treatment and control. Assessors were asked to give a score for five different parameters: Visual Analog Scale (rank from “excellent” to “poor”), Color (Score from “perfect” to “gross mismatch”), Skin Texture (Matte or Shiny), Margins (Distinct or Indistinct), and Vascularity (score from normal to purple) to give a total score ranging from 5 (best) to 18 (worst). The scars were also analyzed using a DermaLab Combo (CyberDERM, Broomall, PA) spectrophotometer, which measured skin color based on the principle of narrow-band reflectance spectrophotometry of 550 ± 30 nm for hemoglobin [erythema or vascularity], giving a single reading using Commission Internationale de l'Eclairage (CIE)—luminance (L), red-green axis (a) and blue-yellow axis (b) [CIELab] values.

Immunohistochemistry

Paraffin-embedded skin samples were sectioned (4 μm) before antigen retrieval by heat in a Decloaking Chamber (Biocare Medical, Inc., Pacheco, CA) at 90°C for 10 min, followed by enzymatic digestion with 0.0625 g Trypsin from Porcine Pancreas (Sigma Aldrich, St. Louis, MO) in 250 mL 37°C PBS for 3 min. Nonspecific antibody binding was blocked with 3% normal goat or horse serum in PBS for 30 min at room temperature followed by overnight primary antibody incubation in a moist air-tight box at 4°C. Primary antibodies against neutrophils (NIMP-R14, rat sc-59338; Santa Cruz Biotechnology, Dallas, TX; 100 μg/mL @ 1:200), macrophages (F4/80, rat MCA497G; Bio-Rad Laboratories, Hercules, CA; 0.5 mg/mL @ 1:200), M2 macrophages (YM-1, rabbit 60130; StemCell Technologies, Vancouver, Canada; 1:200), αSMA (mouse M0851; DAKO, Carpinteria, CA; 71 μg/mL @ 1:200), CD31 (rabbit ab28364; Abcam, Cambridge, UK; @ 1:200), collagen I (rabbit #600-401-103-0.5; Rockland, Limerick, PA; 1 mg/mL @ 1:200), and collagen III (rabbit #600-401-105-0.5; Rockland; 1 mg/mL @ 1:200) were used. Species-specific goat Alexa Fluor anti-rat 488 A11006, goat Alexa Fluor anti-rabbit 488 A11008, goat Alexa Fluor anti-rabbit 568 A11011, goat Alexa Fluor anti-mouse 633 A21050, goat Alexa Fluor anti-rabbit 635 A31577; all Invitrogen, Carlsbad, CA; 2 mg/mL @ 1:200 were diluted in phosphate-buffered saline and applied at room temperature for 1 h followed by nuclear counter stain with 2 μg/mL 4′,6-diamidino-2-phenylindole dihydrochloride (DAPI; Sigma Aldrich) for 2 min. Control slides were included in each run without primary antibody to control for nonspecific secondary antibody binding (Supplementary Fig. S1). Immunofluorescence was acquired using an Olympus IX83 microscope and DP80 camera (Olympus, Tokyo, Japan). Images were captured using a 20 × Objective and using 0.5 × gain unless otherwise stated. Images were then processed using the cellSENS Microscope Imaging Software (Olympus) to measure either positive cells per area or mean fluorescence intensity, which is automatically calculated as intensity per area. Ratio of collagen I to collagen III signal was calculated as the mean collagen I intensity/mean collagen III intensity.

Cytokine measurement

Surrounding normal tissue was removed from the remaining, unfixed half of each burn wound, which was snap frozen in liquid nitrogen and used for protein isolation. Total protein was isolated using AllPrep RNA/Protein Kit (Cat. no. 80404; Qiagen, Germany) according to manufacturer's instructions after homogenization using MoBio BeadMill tubes (Ceramic Bead Tubes, 1.4 mm, Cat. no. 13113-50; MoBio, Carlsbad, CA) in the Bullet Blender Storm 24 (Next Advance, Troy, NY). Cytokine expression was analyzed using ELISA MAX™ Deluxe Mouse TNF-α (#430905), ELISA MAX Deluxe Set Mouse IL-10 (#431414) and LEGEND MAX™ Total TGF-β1 ELISA Kit (#436707) (all BioLegend, San Diego, CA), according to manufacturer's instructions and normalized to mg total protein within wound sample.

Collagen assessment

Collagen was analyzed in histological sections using two-photon second harmonic generation (SHG) microscopy. Histological samples were imaged using a 20 × 1.0 NA water immersion objective on an upright fixed-stage two-photon laser scanning microscope system (Zeiss, Oberkochen, Germany). The excitation source was a Ti:Sapphire femtosecond laser cavity (Newport Mai Tai, Irvine, CA), coupled into a LSM 710 scan module (Zeiss). An excitation wavelength of 890 nm was used to collect SHG signal (435 ± 20 nm) from collagen. Signal was acquired from three separate areas measuring 320 × 320 μm² across each sample. Transmission images were coacquired with SHG data. ImageJ (National Institute of Health, Bethesda, MD) was used to calculate percentage area covered by SHG signal per image, after conversion to a binary image based upon a single manually determined threshold value applied across all images as previously described.^24–26

Statistical analysis

Power studies for the burn mouse model showed that a sample size of eight would give 90% power using a 5% test level and a two-tailed test. Eight mice treatment group per time point were used for the experiments. Statistical differences were determined using a two-tailed Mann–Whitney t-test or Student's t-test. A p-value of less than 0.05 was considered significant.

Results

Integrin αL is highly expressed on circulating human monocytes

Integrin αL/CD11a is expressed on monocytes from healthy individuals at much higher levels compared to macrophages.²⁷ To confirm that integrin αL was expressed on monocytes and macrophages from burn patients, monocytes were purified from the blood and macrophages from burns tissue of five patients 9 days postburn injury. These cells were then examined for the expression of integrin αL (Fig. 1a). When compared to burn wound-associated macrophages isolated from the same patient, the expression of integrin αL was elevated in the monocytes (Fig. 1a, b), suggesting that integrin αL would be a potential target for reducing monocyte infiltration into burn wounds in the early stages of repair.

Figure 1.

Integrin-αL is highly expressed by circulating monocytes from burn wound patients and anti-αL treatment results in an initial delay in wound healing of partial thickness scald burns in mice. (a) A representative blot of integrin αL expression in burns patients' blood-derived monocytes (M) and burn-wound associated macrophages (ø) from the same patient at day 9 postburn. (b) Quantification of relative integrin αL expression in burn wound-associated macrophages (ø) compared to same patients' blood-derived monocytes (M) at day 9 postburn. Mean ± SEM. n = 5. (c) Representative scald burns at days 0, 3, 7, 14, and 28. (d) Macroscopic measurements of the wound area expressed as the % of initial wound area. (e) Representative H&E stained burns at day 3, 7, 14, and 28 postburn injury treated with IgG control or integrin αL antibody. (f) Histological wound length. (g) Histological epithelial gap. Scale bar for (c) 10 mm and (e) 500 μm. Wound length and epithelial gap in (e) denoted by white arrow heads and bars. Mean ± SEM. n = 8. *p < 0.05. αL, alpha L; H&E, hemotoxylin and eosin; IgG, immunoglobulin control; SEM, standard error of the mean.

Anti-αL treatment delays the early stages of healing in murine burn wounds

Macroscopic analysis of the murine burns showed an initial delay in burn wound closure at day 3 with the burns being 14.38% larger in the anti-αL-treated mice (Fig. 1c, d) compared with those treated with isotype-matched IgG control. This increase in burn area at this early stage of healing was statistically significant (p = 0.021). However, by day 7, both treated and control burns were similar in size and both followed the same rate of healing over the 28 day period. The burns were also assessed histologically (Fig. 1e), but the difference in wound closure was not as obvious. No significant difference in epithelial gap, nor wound contraction, as assessed by wound length was observed (Fig. 1f, g).

Anti-αL treatment dampens the early inflammatory response in burn wounds

The effect of the anti-αL treatment on the initial inflammatory response was assessed using immunohistochemistry for NIMP R14 (neutrophils) and F4/80 (macrophages) (Fig. 2a). Neutrophil numbers were not significantly altered in the burns in response to anti-αL treatment at day 3 or 7 postburn compared to controls (Fig. 2b). At day 3, a 54.7% reduction (p = 0.036) in macrophage numbers was seen in the anti-αL antibody-treated mice compared with controls (Fig. 2b). However, by day 7, the numbers of macrophages were similar in both control and anti-αL antibody-treated burns. Other integrin αL/CD11a expressing leukocytes such as lymphocytes, eosinophils, and basophils were not assessed in this study.

Figure 2.

Anti-αL treatment reduces macrophage numbers within the wound site 3 days postburn. Immune cell presence was detected by fluorescent immunohistochemical analysis of (a) NIMP R-14 (neutrophils) and F4/80 (macrophages) (green) expression within the wound area of 4 μm sections of formalin-fixed and paraffin-embedded scald burns excised at day 3 and 7 postburn injury. Sections were counterstained with DAPI (blue) to visualize the nucleus. Scale bar = 10 μm. Measurements were taken of the number of (b) neutrophils and macrophages within the wound dermis presented as mean number of positive cells ± SEM within the wound area. n = 6. *p < 0.05. DAPI, 4′,6-diamidino-2-phenylindole dihydrochloride.

To determine if the reduction in macrophage numbers seen after anti-αL antibody treatment altered cytokine levels in the burns, total protein was isolated from tissue sections and levels of proinflammatory cytokine TNFα and anti-inflammatory cytokine IL-10 detected by enzyme-linked immunosorbent assay (ELISA). A nonsignificant reduction in TNFα expression was observed in the anti-αL-treated mice burns at day 3 (p = 0.074) compared to controls, which was significantly reduced at day 7 (p = 0.035) (Fig. 3a). IL-10 expression was also significantly reduced at 7 days postburn compared to control-treated mice (p = 0.026) (Fig. 3b).

Figure 3.

Anti-αL treatment alters inflammatory profile in the wound site postburn. (a) Proinflammatory cytokine TNFα and (b) anti-inflammatory cytokine IL-10 expression was assessed by ELISA within the wound tissue of scald burns excised at day 3 and 7 postburn injury. Measurements were normalized to mg wound tissue and expressed as mean pg/mg ± SEM. n = 6. *p < 0.05, ^#p = 0.074. (c) Macrophage phenotype was determined by fluorescent immunohistochemical analysis of F4/80 (macrophages, green) and YM-1 (M2 macrophages, red) colocalization (merge, orange) within the wound area of 4 μm sections of formalin-fixed and paraffin-embedded scald burns excised at day 3 and 7 postburn injury. Sections were counterstained with DAPI (blue) to visualize the nucleus. Scale bar = 20 μm. Measurements were taken of the number of M2 macrophages (dual F4/80+YM-1+), presented as (d) mean number of dual-positive cells ± SEM within the wound area, presented alongside total macrophage numbers. Solid bar = F4/80+ macrophages, Hatched bar = F4/80+YM-1+ M2 macrophages. n = 6. *p < 0.05. (e) Proscarring TGFβ1 expression was assessed by ELISA within the wound tissue of scald burns excised at day 3 and 7 postburn injury. Measurements were normalized to mg wound tissue and expressed as mean pg/mg ± SEM. n = 6. *p < 0.05. ELISA, enzyme-linked immunosorbent assay; IL-10, interleukin-10; TGFβ1, transforming growth factor beta 1; TNFα, tumor necrosis factor-alpha.

To determine the proportion of macrophages that were anti-inflammatory (M2), the numbers of dual positive YM-1 (M2 macrophage marker) and total F4/80 cells were assessed during early and late inflammatory stages of wound healing at day 3 and 7. At day 3, while the total numbers of macrophages were significantly decreased following anti-αL treatment, the numbers of M2-positive cells remained constant (Fig. 3c, d). Whereas, at day 7, an increase in M2 macrophages was observed in the anti-αL-treated mice (p = 0.045) (Fig. 3c, d).

Macrophages also secrete growth factors during the repair process. To investigate if proscarring TGFβ1 was altered posttreatment with anti-αL antibodies, the burns were also assessed for expression of TGFβ1 by ELISA. While a slight initial reduction in TGFβ1 was observed at day 3 in response to treatment with the anti-αL antibody (Fig. 3e), this was only significant at day 7 (p = 0.037).

The effect of anti-αL antibody treatment on the differentiation of fibroblasts into proscarring myofibroblasts was assessed by looking at the numbers of myofibroblast marker, alpha smooth muscle actin (αSMA) positive, and endothelial cell marker, CD31 negative cells. Dual-positive cells which identify blood vessels were excluded from analysis. A significant reduction in αSMA⁺/CD31⁻ cells was observed in the anti-αL-treated mice at day 7 (p = 0.012) and day 28 postburn (p = 0.019) compared to control-treated mice (Fig. 4a, b).

Figure 4.

Anti-αL treatment reduces myofibroblasts within the wound. (a) Separate images of fluorescent immunohistochemical detection of αSMA (red) and endothelial cell marker CD31 (orange) demonstrating αSMA⁺ CD31⁻ myofibroblasts within the wound area of scald burns excised at day 3, 7, and 28 postburn injury and αSMA⁺ CD31⁺ blood vessels. Sections were counterstained with DAPI (blue) to visualize the nucleus, with no primary control images included. Scale bar = 20 μm. (b) αSMA⁺ CD31⁻ cells were counted and expressed as mean cells/mm² ± SEM. n = 6. *p < 0.05. αSMA, alpha smooth muscle actin.

Reduced burn wound macrophage numbers, following anti-αL treatment, improves the appearance of the scar following burn injury

The murine burns were assessed at day 28 for maturation of collagen bundles and scarring. Scar tissue was immunostained to look at the extracellular reorganization of collagen I (Fig. 5a) and collagen III (Fig. 5b). The ratio of collagen I to collagen III (Fig. 5d) in the scar area of day 28 burn wounds treated with anti- αL antibody was similar to that found within unwounded tissue, while the Collagen I/III ratio of IgG control-treated burns was significantly higher than both the anti-αL antibody-treated burn wounds (p = 0.038) and unwounded tissue (p = 0.0044). Quantitative SHG microscopy of fibrillar collagen within the scar area (Fig. 5c) revealed no significant differences in the mean signal intensity (Fig. 5e) nor percent area coverage of this signal (Fig. 5f), however, a trend for less SHG signal in the scar area of burns treated with anti- αL antibody was observed (p = 0.054 compared with unwounded tissue). The redness of the scar was assessed using spectrophotometry, and the overall trend was that scars from mice treated with anti-αL antibody were less red compared with IgG control-treated mice (Fig. 5g, h). Macroscopic analysis of the same scars also revealed a trend for scars to be smaller in area (14.3% vs. 12.2% of initial burn wound area) in anti-αL antibody-treated mice compared with control-treated mice (Fig. 5g, i). Images of the day 28 scars were scored by eight assessors blinded to the treatment group to determine the impact of anti-αL treatment upon the overall appearance of the scar. Comparing the average scar score of the anti-αL group compared with the IgG control group for all assessors revealed a subtle but statistically significant improvement in scar appearance (p = 0.042) (Fig. 5j).

Figure 5.

Anti-αL treatment decreases the collagen I/III ratio and scar appearance in scald burn scars. (a) Collagen I and (b) collagen III (green) expression was detected by fluorescent immunohistochemistry 28 days following burn and treatment with anti-αL or IgG control and compared to unwounded tissue. Sections were counterstained with DAPI (blue) to visualize nucleus. (c) Collagen fibers were visualized using quantitative SHG microscopy. Scale bar = 50 μm. (d) The ratio of Collagen I/III was calculated from the mean fluorescence intensity of collagen I and III staining and expressed as mean ratio ± SEM. n = 8. (e) Collagen fiber assessment was performed from SHG images and expressed as mean SHG signal intensity (f) and % area coverage of the SHG signal. (g) Scar appearance was assessed 28 days post-scald burn and treatment with anti-αL or IgG control, with measurements taken of (h) redness/erythema of the skin within the scar area as well as normal unwounded skin and (i) total scar area expressed as % of initial wound area. Scale bar = 10 mm. Mean ± SEM. n = 8. (j) Eight independent assessors also scored the overall appearance of the day 28 scars from macroscopic images giving each scar a score ranging from 5 (best) to 18 (worst). Box plots of the pooled average scores of all eight assessors for scars treated with anti-αL or IgG control, showing the lower (Q1) and upper (Q3) quartile with whiskers representing the Minimum and Maximum average scores, the line within the box representing Median and + representing Mean. *p < 0.05, **p < 0.01. SHG, second harmonic generation.

Discussion

Excessive inflammation that is typically seen in burn injuries often leads to fibrosis, hypertrophic scarring, and contracture, which can be debilitating for patients.¹³ Studies suggest that the number of macrophages in a wound, particularly in the early stages of repair, dictate the level inflammation which contributes to the resulting fibrosis and scarring.²⁸ Therapeutic approaches that reduce the numbers of macrophages within burns would therefore be expected to lead to better burn injury outcomes. Anti-integrin therapies have been used in the clinic to reduce inflammation in a range of diseases, including psoriasis, due to their ability to block immune cell migration.¹⁸ This study therefore tested the ability of function-blocking antibodies to integrin αL, a key adhesion molecule in monocyte migration and extravasation from blood vessels,¹⁷ to reduce inflammation and improve burn injury repair using a mouse model of burn injury repair. While differences do exist between mouse and human skin; human skin has a thicker epidermis and lacks a panniculus carnosus, which results in mice tending to heal by contraction, however, they also share many similarities, including homeostasis, inflammation, and proliferation, and studies using mouse models can provide important information about new therapeutic approaches.²⁹

Integrin αL was found to be highly expressed on circulating monocytes in the blood of burn patients making it an ideal candidate for targeting to reduce monocyte extravasation from the blood into the burn wounds. Treatment of mice with a function blocking anti-αL antibody 24 h after scald injury did indeed lead to a reduction in burn wound-associated macrophages in the early stages of repair, which resulted in a decrease in inflammation and lower levels of proscarring TGFβ1 and αSMA expression. Overall, there was also a trend toward better resolution of the scar by 28 days postburn, including smaller, less red scars, with a more normal Collagen I/III ratio.

The timing of the treatment of the burns with the αL antibody was important. Integrin αL is expressed by all leukocytes and is expressed on both monocytes and neutrophils, two of the key players in regulating inflammation.¹⁷ Neutrophils are the first immune cell type recruited to the wound within hours of injury and their numbers peak around 1–2 days following injury.⁹ As one of the first responders to an injury, their function is beneficial. They remove potential harmful pathogens, perform an early debridement of the wound, and recruit other immune cells through the secretion of soluble mediators. Monocytes migrate into the site of injury around 24 h later and their numbers increase until around day 2 and remain high until day 5. This timing leaves a window of opportunity to target early macrophages in the early inflammatory phase without affecting neutrophil numbers. To this end, anti-αL antibodies were delivered systemically 1 day postburn injury, after the initial influx of neutrophils into the wound. Indeed, the anti-αL treatment had no significant effect on the overall numbers of neutrophil that were present within the burn wounds compared with controls.

The systemic delivery of αL antibodies 1 day after burn injury was able to half the number of wound-associated macrophages on day 3, suggesting that the antibody had successfully reduced monocyte migration into the wound. Whether the residual population of macrophages seen in the burn is due to macrophages that have entered before the delivery of the antibody, the antibody not fully blocking migration due to compensation by another integrin family member such as MAC-1¹⁷ or the dose of antibody being too low is not clear. Future studies are needed to determine if a higher dose or even a second dose of function blocking antibodies could further reduce macrophage numbers and improve the overall scar outcome.

Nevertheless, the ability to transiently prevent immune cells from migrating into tissue is of importance as the anti-LFA-1 antibody efalizumab, was withdrawn after three cases of progressive multifocal leukoencephalopathy were described in patients on long-term (>3 years) efalizumab therapy.³⁰ We have used only 1 dose of anti-LFA-1 integrin antibody in this mouse study and this treatment would not be supported for long-term use. In studies where LFA-1's partner ICAM-1 is blocked, short-term (1 day) in a human phase I clinical trial³¹ looking at targeting immune cell migration after burn injury has shown that the approach of reducing immune cell migration is feasible with none of the side effects seen with long-term inhibition of immune cell migration.

Previous studies have also shown that when the early population of macrophages is ablated, macrophages are able to repopulate a burn wound at later time points.¹³ Although there was slight delay in healing initially, this repopulation ultimately led to wound closure at a similar rate to the control wounds. The detrimental role that early macrophages play appears to be linked, in part, to their secretion of proinflammatory cytokines, such as TNFα.^11,32 In impaired wound healing models, Ashcroft et al. showed that delivering anti-TNFα dampens the excessive inflammatory response and results in improved healing rates.³³ Delivering anti-TNFα to full thickness excisional wounds significantly delayed wound healing at day 3 with no significant difference in wound size by day 7.³⁴ In patients receiving anti-TNFα to treat rheumatoid arthritis, there were three cases (6%) where a delay of wound healing of 1–2 weeks was observed following surgery.³⁵ The reduction in early macrophage numbers seen in anti-αL-treated mice led to the reduced expression of TNFα at day 3 and 7 postburn. Interestingly IL-10 levels within the burns were also significantly reduced at day 7 suggesting a potential dampening of both pro and anti-inflammatory mediators of inflammation by the treatment. In addition, while the total number of macrophages was reduced at day 3 following anti-αL treatment, similar numbers of anti-inflammatory M2 macrophages were found to be present within the burns, and by day 7, the numbers of these anti-inflammatory macrophages were significantly increased compared with control-treated burns. These results suggest that the anti-integrin approach reduces macrophage migration and dampens early inflammation within burns.

Macrophages are one of the major cell types that secrete TGFβ1,¹³ which drives αSMA-positive myofibroblast differentiation and wound contraction.³⁶ Treatment with αL functional blocking antibodies also resulted in a significant reduction in TGFβ1 at day 7 in the burns; however, and this was paralleled with a significant reduction in the expression of αSMA at this time. This reduction in profibrotic TGFβ1 may, in part, be responsible for the trend that was observed toward reduced scar formation in the αL antibody-treated mice. Close inspection of the burns showed that at day 28, the anti-αL treatment group scars more closely resemble unwounded skin than their control counterparts. While the abundance of both collagen types was reduced in the wounded tissue of both groups compared to unwounded tissue, the area covered by collagen fibers was similar. The anti-αL treatment group displayed a lower collagen I/III ratio, more similar to that of unwounded tissue. Collagen III is less abundant in normal skin, but is initially found in wounds and later replaced by collagen I, and therefore, a lowered collagen I/III ratio is often associated with a reduction in scarring and fibrosis.³⁷ Burn wounds result in a fibrotic wound area and a higher collagen I to III ratio. By day 28, the IgG control group already displays a skew toward excess collagen I deposition. The scars following anti-αL treatment, however, show a ratio much more similar to unwounded tissue, such that although remodeling is still progressing, the resulting scar appears to be compositionally more similar to unwounded tissue. Visually, the scars treated with anti-αL antibodies were scored better by independent observers than control-treated burns.

Overall, these results suggest that a targeted integrin αL therapy reduces macrophage infiltration into burns leading to reduced inflammation during the early phase of healing and a trend toward reduced scar formation. Future studies will determine whether the reduction in scarring can be improved by a modified integrin αL therapy regime.

Innovation

Despite advances in burns research, management of burn injuries remains a challenge, and there are limited options available for improving healing and reducing scar formation. While inflammation is a key stage in normal tissue repair, prolonged or excessive inflammation contributes to increased fibrosis and scarring. These studies have identified that anti-integrin therapies have the potential to reduce inflammation and thereby potentially improve burn injury outcomes. Integrin αL function-blocking antibodies have previously been used to reduce inflammation in psoriatic skin. Repurposing an existing therapy for the treatment of burns could lead to rapid development of a new antiscarring approach.

Key Findings

• Excessive inflammation contributes to elevated fibrosis and scarring in burn wounds.

• Integrins modulate monocyte extravasation from the circulation into burns

• Blocking monocyte entry into murine burn wounds using integrin αL function-blocking antibodies reduces the numbers of macrophages.

• Decreased proinflammatory TNFα and profibrotic TGFβ1 are observed following treatment with αL antibody with a trend toward reduced scar formation.

• An integrin αL therapy for the treatment of burns could potentially lead to reduced macrophage infiltration into burns leading to reduced inflammation and less scarring.

Abbreviations and Acronyms

αL

alpha L

αSMA

alpha smooth muscle actin

β2

beta 2

DAPI

4′,6-diamidino-2-phenylindole dihydrochloride

ELISA

enzyme-linked immunosorbent assay

IgG

immunoglobulin

IL-6

interleukin-6

LFA-1

lymphocyte function-associated antigen-1

SHG

second harmonic generation

TGFβ1

transforming growth factor beta 1

TNFα

tumor necrosis factor alpha

Acknowledgments and Funding Sources

This research was supported, in part, by the Channel 7 Children's Research Foundation. AJC is supported by NHMRC Senior Research Fellowship (GNT#1102617)

Authors' Contributions

A.J.C. and R.Z.M. conceived and designed the experiments; X.S., D.A., and R.Z.M. performed the experiments; X.S. analyzed the data; N.P. and M.S. contributed analysis tools; X.S., R.Z.M., and A.J.C. wrote the article.

Author Disclosure and Ghostwriting

No competing financial interests exist. The content of this article was expressly written by the author(s) listed. No ghostwriters were used to write this article.

About the Authors

Xanthe L. Strudwick, PhD, is a research associate at the Future Industries Institute, University of South Australia. She has research expertise in wound healing and cell biology. Damian H. Adams, BBiotech(Hons), is a research assistant at the Future Industries Institute, University of South Australia with extensive wound healing expertise. Natasha T. Pyne, BSc(Hons), was a research assistant within the Centre for Cancer Biology at the University of South Australia and SA Pathology, Adelaide, Australia. She is currently an operations manager at the South Australia Health and Medical Research Institute. Michael S. Samuel, PhD, is Associate Professor and head of the Tumour Microenvironment Laboratory at the Centre for Cancer Biology, University of South Australia and SA Pathology, Adelaide, Australia. Rachael Z. Murray, PhD, is Senior lecturer in the School of Biomedical Sciences and Head of the Inflammation and Tissue Repair group at Queensland University of Technology. Allison J. Cowin, PhD, is Professor of Regenerative Medicine at Future Industries Institute, University of South Australia. She is an NHMRC Senior Research Fellow, Editor of Wound Practice and Research, Board member of Australasian Society for Dermatology Research (ASDR) and was founding president of the Australasian Wound and Tissue Repair Society.

Supplementary Material

Supplementary Figure S1