Wound Healing Immediately Post-Thermal Injury Is Improved by Fat and Adipose Derived Stem Cell Isografts

Abstract

Objectives

Patients with severe burns suffer functional, structural, and aesthetic complications. It is important to explore reconstructive options given that no ideal treatment exists. Transfer of adipose and adipose-derived stem cells (ASCs) has been shown to improve healing in various models. We hypothesize that use of fat isografts and/or ASCs will improve healing in a mouse model of burn injury.

Methods

Twenty 6–8 week old C57BL/6 male mice received a 30% surface area partial-thickness scald burn. Adipose tissue and ASCs from inguinal fat pads were harvested from a second group of C57BL/6 mice. Burned mice received 500μl subcutaneous injection at burn site of 1) processed adipose, 2) ASCs, 3) mixed adipose (adipose and ASCs), or 4) sham (saline) injection (n=5/group) on the first day post-injury. Mice were followed by serial photography until sacrifice at days 5 and 14. Wounds were assessed for burn depth and healing by Hematoxylin and Eosin (H&E) and immunohistochemistry.

Results

All treated groups showed improved healing over controls defined by decreased wound depth, area, and apoptotic activity. After 5 days, mice receiving ASCs or mixed adipose displayed a non-significant improvement in vascularization. No significant changes in proliferation were noted at 5 days.

Conclusions

Adipose isografts improve some early markers of healing post-burn injury. We demonstrate that addition of these grafts improve specific structural markers of healing. This improvement may be due to an increase in early wound vascularity post-graft. Further studies are needed to optimize use of fat or ASC grafts in acute and reconstructive surgery.

INTRODUCTION

With over 400,000 clinically significant burns in the United States in 2012 and an estimated 11 million people worldwide requiring medical attention for burns each year, there remains a need for improvements in the treatment of burn injuries.^1,2 These injuries present a clinical challenge requiring significant resources for both initial post-burn care and chronic management. Long term consequences of thermal injuries including scar and joint contracture, fibrosis, and soft tissue contour deformity cause significant morbidity for patients if such sequelae are not treated appropriately.³

Severe burns lead to deficits in the epidermis, dermis, and underlying soft tissue with significant injury to surrounding tissues that can go unrecognized in the initial post-burn period.⁴ These injuries further disrupt the biochemical milieu of the wound causing both delay and disorganization of the initial regenerative processes required for successful healing.^5–7 The combination of disorganized wound healing and tissue deficits promote the characteristic abnormal scarring of thermal injuries.⁸ Without intervention these scars develop into tight, thick, fibrotic bands that lack normal elasticity, color, and texture.⁹ When burn scars are present on normally mobile anatomic structures such as the face and extremities there is the potential for significant functional and aesthetic impairment.

There are currently few options to address the functional and aesthetic sequelae of burn injuries. Non-operative management strategies include corticosteroids, dermabrasion, and pressure dressings.^10–1 A major limitation of these strategies is that they do not prevent the formation of the scar but rather attempt to mitigate its consequences. Currently, full- and split-thickness skin grafts are the mainstay of surgical treatment. These grafts are able to protect the wound, improve appearance, and even improve function in the setting of severe wound contracture. However, skin grafts are not without risks and significant cosmetic deformities. Skin grafts have limited durability and are susceptible to recurrent infection. Further, skin grafting carries all of the risks associated with a formal operation. Although full-thickness skin grafts have been touted to decrease contractures and provide improved cosmetic outcomes, patients with large surface area burns may not have enough healthy tissue to provide full-thickness grafts. Split-thickness skin grafts, although available in large quantities, can lead to soft tissue contracture, contour irregularities, asymmetries, and color mismatch with adjacent tissues.¹² Thus there is need for an option that provides both more durable coverage and a better cosmetic outcome including better color match, better contour and thickness match, and better symmetry.

Autologous fat grafts may provide a therapeutic option for these defects either as an adjunctive procedure to current surgical approaches or as a stand-alone therapy.¹³ Adipose tissue is available in large quantities, is easy to manipulate, and is able to fill soft tissue deficits. Autologous fat transfer is already in use by surgeons for patients with lipoatrophy of the face associated with HIV or facial paralysis, radiation injury, chest wall defects, breast capsular contracture, chronic ulceration, and any other contour irregularities on the body.^13–4 While most fat grafting techniques rely on the homologous function of the graft as a space filler, there is a growing body of evidence that the paracrine effects of the graft alter the biochemical milieu of the wound leading to improvement in healing as well as in the functional and aesthetic resolution of the injury.^14–6 Recent evidence indicates that syngeneic fat has the capacity to both accelerate revascularization and reduce fibrosis in thermal injuries.¹⁷ Furthermore, adipose tissue is a major source of mesenchymal stem cells.¹⁸ These adipose derived stem cells (ASCs) are under investigation for their regenerative potential, their pro-angiogenic stimulus, and have been shown in various injury models to improve wound healing.^16,19,20 Autologous fat thus may have the potential to address both the tissue defect and poor wound profile of thermal injury.

Given the need and shortcomings of currently available treatment options, it is important to investigate additional options for burn patients, such as autologous fat transfer. In this present study we investigate the potential for processed adipose tissue, ASCs, and mixed adipose isografts to enhance the reconstruction of burn scars. We hypothesized that all three types of grafts would improve certain aspects of wound healing in our mouse model of burn injury.

METHODS

Mice

This study was approved by the University of Michigan University Committee on the Use and Care of Animals (UCUCA PRO00001553). Young-adult (6–8 week) C57BL/6 mice (n=20) were anesthetized under 3% isoflurane inhalation. The dorsum was shaved using an electric clipper. The animals were divided into four groups: 1) processed adipose (n=5), 2) ASCs (n=5), 3) mixed adipose (n=5), or 4) sham (n=5). All animals were housed in UCUCA supervised facilities with no restrictions to diet or water.

Thermal Injury

All animals underwent a 30% surface area partial-thickness scald injury by exposure to 60C water for 17 seconds and received analgesia with subcutaneous buprenorphine per standardized protocol. One mouse from the ASC group was sacrificed on post-injury day 0 for scald injury greater than 30%. Animals were allowed to recover under heat lamp and were given subcutaneous buprenorphine analgesia for three additional days.

Isolation of ASCs

The inguinal fat pad was harvested from C57BL/6 mice. The adipose tissue was minced finely with mechanical disruption. This adipose tissue was then serially washed in phosphate-buffered solution (PBS) and placed in collagenase as previously described.²¹ After 30 minutes of digestion the collagenase was neutralized with fetal bovine serum (FBS). The tissue was then centrifuged at 1,000 rpm for five minutes. After centrifugation, the supernatant, consisting of plasma and adipocytes, was discarded and the stromal vascular component (cell pellet) was plated in 10% Dulbecco’s Modified Eagle Medium (DMEM). Cell homogeneity was assured by passaging the cells at least 3 times prior to use. This has previously been shown to enrich for a 97% homogenous population of ASCs that are CD105+, CD90+, and CD45−.²²

Isolation of Fat Isografts

Inguinal fat pads were harvested from C57BL/6 mice. The adipose tissues were finely minced with mechanical disruption. This adipose tissue was then centrifuged for 3 minutes at 3,000 rpm per the Coleman protocol²³. After centrifugation, the cell pellet and supernatant were discarded and the adipose tissue was used for grafting.

Fat and ASC Isografts

One day post-burn, mice received a 500ul subcutaneous injection at the wound site of either: 1) processed adipose tissue (Coleman protocol), 2) 1×10⁶ ASCs suspended in PBS, 3) mixed adipose tissue and ASCs (1×10⁶ in 500uL), or 4) 500 uL of PBS solution alone.

Serial Photography

All mice were followed with serial photography daily to assess for wound healing. Healing was assessed by quantifying burn area until sacrifice. Seven mice were followed to sacrifice at post-injury day 5. The remaining mice (n=12) were followed for 2 weeks to sacrifice. One mouse was sacrificed at the time of burn due to larger than desired burn surface area (>30%).

Immunostaining

Paraffin embedding and immunohistochemistry were performed. Selected slides were stained with hematoxylin and eosin (H&E). Immunohistochemistry staining was performed using CD-31/PECAM (1/100, b-CM303B, Biocare Medical, Concord, CA), Ki-67(1/500, ab16667, Abcam, Cambridge, MA), and Caspase-3 (1/1000, 9661, Cell Signaling, Danvers, MA). Peroxidase activity was quenched with 3% hydrogen peroxide. Slides were blocked with 5% bovine serum albumin (BSA) in PBS and appropriate biotinylated secondary antibodies were applied. 3,3′-Diaminobenzidine was used for visualization.

Microscopy and Tissue Analysis

Skin from the center of the thermal injury site, including underlying adipose tissue grafts (if applicable) were harvested for analysis. Skin was selected to provide a cross section containing both burned and unburned tissue across the length of the sample. Skin was fixed in formalin, embedded in paraffin, and sectioned for H&E and immunohistochemistry staining. Tissues from mice sacrificed at day 5 were analyzed for markers of early healing including histologic wound depth (H&E staining; depth of necrotic tissue), angiogenesis (CD-31/PECAM; endothelial cell count), apoptosis (Caspase; stained cells), and proliferation (Ki67, stained cells). Tissues distal and proximal to injection were harvested from mice sacrificed at day 14 and analyzed for depth and viability by H&E stain. All tissues were evaluated for necrosis based on loss of cellular architecture, cellular disruption, and vacuolization²⁴. Following staining, H&E and immunohistochemistry images were taken using a Nikon E-800 upright microscope (Nikon, Melville, NY). Slides were imaged at 10x using bright field microscopy at site of injury, site of adipose tissue graft, and immediate periphery. Images were taken at equal distance from wound center as assessed by burn diameter. Composite images were formed from overlapping images at each burn site. Depth measurements were assessed at six points per sample. Depth of burn was assessed qualitatively by evaluation of deepest necrotic margin as above. Immunohistochemistry staining by CD-31, Ki67, and caspase was visually quantified for each mouse treatment type with NIS-Elements (Nikon, Melville, NY). Five equivalent non-overlapping fields were chosen at random from the composite images. Ki67 and caspase stains were assessed by direct count of positive-staining pixels. CD-31 was assessed by evaluation of positive-stained capillary rings.

Data Analysis

Unless otherwise indicated, data are expressed as means +/− SD. Values presented are means of all fields evaluated per treatment group. Differences in treatment groups were tested for significance versus sham controls using Student’s T-test. Fat, ASC, and mixed isografts were likewise compared against each other. Differences of p<0.05 were considered statistically significant.

RESULTS

Fat Isografts Remain Viable

H&E images of fat grafts and overlying dermis were examined for evidence of necrosis or inflammation consistent with acute rejection at both 5 day and 14 day time points (Fig 1A & 2A). By H&E and gross examination at time of sacrifice, viable dermis and subcutaneous tissue was identified in all mice receiving grafts and all grafts were well attached in the subcutaneous space based on gross examination on day of sacrifice.

Figure 1. 5-Day Burn depth following partial thickness thermal injury.

Thickness (5d) was measured at six points spread across each injury. A. Representative images collected from each experimental subset. B. Average of measurements of depth of burns. Error bars indicate SD. Samples for burn depth stained with H&E.

Figure 2. 14-Day Burn depth following partial thickness thermal injury.

Thickness (14d) was measured at six points spread across each injury. A. Representative images collected from each experimental subset. B. Average of measurements of depth of burns. Error bars indicate SD. Samples for burn depth stained with H&E.

All Grafts Reduce Wound Depth and Area

Next we evaluated the wound sites for histologic depth and gross evidence of recovery by serial photography. Histologic examination revealed that groups receiving ASCs alone (122.9±11.1 μm) or combined fat isografts and ASCs (126.1±9.7 μm) displayed significantly decreased wound depth on day 5 compared to untreated controls (193.6±18.3 μm: p<0.01; Fig. 1B). There was no evidence of difference in wound thickness at day 5 for fat grafts alone versus control (203.4±16.3 μm p=0.35; Fig. 1B). By day 14 the depth of injury was significantly reduced for all treatment groups (Fat 154.8±25.5 μm vs. ASC 157.1±22.0 μm vs. mixed 141.6±24.4 μm) compared to untreated burns (215.2±19.2 μm: p<0.05; Fig. 2B).

We also examined whether the changes in wound depth were a purely local effect at injection/transplant site or whether structural changes could be identified away from the graft site. Comparison of burn depth at injection site and distal to the burn showed no difference at day 14 for all groups except for mice which received fat isograft only (Fig. 3).

Figure 3. 14-Day Burn depth proximal and distal to treatment site.

Skin samples were taken at sites distal and proximal to injection of treatment. Burn depth was measured at six points spread across each tissue sample. Error bars indicate SD. Samples for burn depth stained with H&E.

Initial burn area assessed by size of eschar on days 1 and 2 was noted to be similar across all groups (Fig. 4, Columns 1&2). Likewise, no difference in wound area was noted at 5 days when comparing each of the treatment groups with the sham control (Fig. 4, Column 3). However, decreased burn area compared to sham control was observed in all treatment groups by day 14 (Fig. 4, Column 5). However, there were no differences in burn area between any of the treatment groups by day 14 (Fig. 4).

Figure 4. Burn area progression in mice.

Thermal injuries days 1,2,5,9, and 14. Representative images collected from each experimental subset.

Fat and ASC Grafts Improve Early Markers of Wound Healing but do not Effect Cell Proliferation

We next explored the effect of fat isograft and/or ASC treatment on apoptosis (caspase), angiogenesis (CD31), and proliferation (Ki67) using immunohistochemistry. We found that after 5 days, mice receiving either ASCs alone or fat isograft with ASCs showed a significant reduction in apoptosis when compared to controls (p<0.05; Fig. 5). There was no significant difference in caspase staining for apoptosis between fat alone and control. Likewise, there was no difference between either of the two groups containing ASCs when compared with each other (Fig. 5).

Figure 5. Apoptosis following partial thickness scald injury.

Cell death (5d) was determined by direct count of Caspase staining cells deep to the burn and to the immediate periphery. A. Representative images collected from each experimental subset. B. Counts were measured in four equivalent regions and averages are represented above. Error bars indicate SD. Samples stained by IHC for Caspase (Arrows).

Both groups receiving ASCs showed a non-significant increase in vascularization by CD31 staining versus controls (Fig. 6). No significant change in vascularization was noted between the fat graft and control groups. We found no significant difference in proliferation between any of the groups tested by Ki67 staining.

Figure 6. Vascularization following partial thickness scald injury.

Angiogenesis (5d) was determined by direct count of CD-31 staining rings (capillary count) deep to the burn and to the immediate periphery. A. Representative images collected from each experimental subset. B. Counts were measured in six equivalent regions and averages are represented above. Error bars indicate SD. Samples stained by IHC for CD-31 (Arrows).

DISCUSSION

Here we explore the impact of fat and ASC isografts on healing following thermal injury. Our model for fat and/or ASC transfer is easily reproducible. By transferring fat and ASCs from donor mice that match the immunologic profile of the recipients (both are C57BL/6), we attempted to replicate the clinical picture of adipose autografting. First, we demonstrate that fat isografts remain viable through the early period of wound healing when placed at the burn site. Burn wound depth and area was reduced in mice which have been treated with fat and/or ASCs. Additionally, we showed that apoptotic markers are significantly reduced in mice receiving treatments that included ASCs.

A major clinical challenge presented by burn injuries is the associated delay and disorganization of cell proliferation, revascularization, and selective apoptosis to regenerate the original healthy structure.^4–6 The traditional model of wound healing is divided into three stages: inflammation, proliferation, and maturation. Successful wound healing requires early activation of both distant phagocytic cells and local endothelial cells to prepare the wound environment for recovery. Early markers of this process include formation of new capillaries, phagocytic debridement of damaged cells, and enhanced mitotic activity in local proliferative tissues.²⁵ In optimal conditions, granulation tissue will form and the wound defect will be filled with collagen, which is later remodeled to restore form and function to the site of injury. In burn injuries, this process is both delayed and disrupted by structural damage and the presence of the burn eschar.²⁶ Burn healing is also limited by damage distal to the immediate site of injury by thermal energy leading to the phenomenon of burn conversion.⁴ A potent inflammatory response to burned tissues further delays healing and damages local tissues disrupting capillary formation and preventing distal cells necessary for recovery from arriving.^4,5,26 This capillary damage and delay in revascularization contributes to many of the acute or subacute sequelae of thermal injury including increased risk of infection and fluid loss.²⁶

Treatments that included ASCs were found to decrease regional apoptosis in the early period post-burn. Likewise, improvement in vascularization was found to approach significance in groups receiving ASCs. These results were not found in groups receiving fat alone. Previous studies have examined the biochemical profile of ASCs in the wound site and found that they secrete angiogenic cytokines including VEGF and FGF2.²⁷ There is mounting evidence that adipose transfer also improves neovascularization at the burn site.^14,17,19,20 Although we did not identify the same improvement in vascularity reported previously, our study does have a limited duration of two weeks. It is believed that ASCs are the primary functional components of fat in terms of improved wound healing and it is possible that our study ended too early to examine the more subtle effect of fat isografts on the wound environment.¹⁹

Spatial contour defects following burn injury are also difficult to address in the clinical setting. Current therapeutic options include biologic and artificial skin substitutes. Fat grafting has become increasingly popular in the treatment and remodeling of burn scars.^13,14 Autologous fat transfer has been shown to dramatically improve the texture and contour of the burn scar, however, the mechanisms for this change are currently poorly understood. ^13,14 Using our model we have shown that addition of fat isografts, ASCs, or a combination of these two may improve structural markers of healing including burn depth and area, suggesting that fat grafting may not only improve biochemical markers of healing but may also contribute to an actual improvement in burn wound outcome.

The most significant consequences of thermal injury come from damage to the mobile anatomic structures of the extremities and face. Our dorsal burn model benefits from ease of use and reproducibility but does not address functional outcomes in these structures. Mouse skin heals differently from human tissue and is subject to rapid contraction following full-thickness injuries unlike human skin which heals primarily through re-epithelialization and granulation tissue formation.²⁸ Furthermore, the time-course of our results limits the conclusions we can draw on long-term functional and aesthetic outcomes. Despite these limitations, our model serves as a starting point to further explore the contribution of fat and ASC composite grafts to wound healing.

It is currently unknown what effects early intervention with fat and/or ASCs will have on long-term outcomes post-thermal injury. Our study with early time-points suggests that these grafts may impart some change on the wound healing. We will next carry out long-term studies to address final outcomes with treatment. Given the significance of burns to mobile anatomic sites we will evaluate the effect of our treatment using a hind-limb burn model. Finally, we will also continue to optimize our fat transfer process. Previous studies have shown that fat grafts can remain viable in mouse subcutaneous tissue as long as eight to twelve weeks, however, ASCs usually significantly decrease in number after two weeks.^20,29 Finally, any effects from the selection of the donor site, the processing of the fat, or the time and method of delivery are currently poorly understood and it is likely that there is room for improvement in the therapeutic delivery of our grafts to the wound site.