Allogeneic Transplantation of an Adipose-Derived Stem Cell Sheet Combined With Artificial Skin Accelerates Wound Healing in a Rat Wound Model of Type 2 Diabetes and Obesity

The population of patients with diabetes is growing worldwide and reached ∼400 million in 2013 (1). An estimated 15–25% of patients with diabetes are at risk for developing a lower-extremity diabetic ulcer in their lifetime (2). Among patients with diabetic foot ulcers, 7–20% will subsequently require an amputation, and 85% of lower-extremity amputations in diabetic patients are caused by foot ulcers (3). An urgent need exists to develop new therapies for the treatment of diabetic wounds to prevent foot ulcers from leading to amputations.

Artificial skin is one of the commercially available treatments for full-thickness skin defects after debridement. Artificial skin typically comprises two layers: an outer silicone sheet layer and an inner collagen sponge layer that act as the epidermis and dermis, respectively (4). However, difficulties often are experienced with the use of artificial skin treatments for diabetic wounds in patients with neuropathy, impaired blood flow, or relatively large wounds. In these cases, the formation of neodermal tissue is delayed, thus prolonging the treatment period (5).

Recombinant basic fibroblast growth factor has been widely used in wound healing to promote angiogenesis and granulation. Although recombinant basic fibroblast growth factor successfully accelerates wound healing, it requires frequent administration because of its short half-life (6,7).

Cell-based therapy has emerged as a new application for the treatment of ulcers (8,9). In particular, mesenchymal stem cells (MSCs) exhibit an excellent potential for increasing the rate of wound healing given their self-renewal capacity, immunomodulatory effects, and ability to differentiate into various cell lineages (10). Adipose-derived stem cells (ASCs), a type of MSC, exhibit various advantageous properties, such as paracrine activity and angiogenic potential (11,12). ASCs have been used experimentally in wound-healing applications (13,14).

Single-cell suspensions of MSCs have been directly injected around wounds in numerous studies, and accelerated wound healing has been observed (15,16). Despite reports of improved wound healing in diabetic ulcer models following the injection of single-cell suspensions, the residence time of transplanted cells at the wound site is unclear. Yang et al. (17) reported that the injection of single-cell suspensions result in the formation of island-like aggregates and visible necrosis of the injected cells.

Cell sheet engineering improves the efficacy of cell transplantation. Okano et al. (18) reported that temperature-responsive culture dishes can be prepared by covalently grafting the temperature-responsive polymer N-isopropylacrylamide onto a culture dish surface. The grafted polymer layer permits temperature-controlled cell adhesion/detachment on the culture dish surface. Specifically, the surface is hydrophobic at 37°C, allowing cells to adhere and proliferate. In contrast, the surface becomes hydrophilic at <32°C, causing cells to spontaneously detach from the surface. These cells can be subsequently harvested as a contiguous cell sheet with intact cell-cell junctions and extracellular matrix (ECM). This approach avoids the use of proteolytic enzymes, such as trypsin, that damage the ECM, and the cell sheets can be immediately transplanted to the wound site (19).

Artificial skin has been used to achieve great therapeutic results, even in chronic wounds such as diabetic wounds (20). In addition, cellular artificial skin has been developed, and a significant body of evidence supports its use in the treatment of diabetic wounds (21). As previously stated, ASCs exhibit therapeutic potential; therefore, ASC sheets with artificial skin have the potential to accelerate chronic wound healing and offer a feasible therapeutic strategy for diabetic wound treatment. The efficacy of ASC sheets with artificial skin requires further evaluation in diabetic wounds with exposed bone. In this study, the artificial skin used provided a three-dimensional framework for ASCs, maintaining transplanted ASC sheets and wounds in a moist environment, preventing wounds from spontaneous contraction, and protecting wounds from infection and external forces (22). The aim of the current study was to evaluate the efficacy of allogeneic ASC sheet transplantation with artificial skin to promote the healing and vascularization of full-thickness skin defects in a rat diabetic wound model.

All experimental protocols were approved by the Animal Welfare Committee of Tokyo Women’s Medical University School of Medicine. Zucker diabetic fatty (ZDF) rats (ZDF-Lepr^fa/CrlCrlj) were used as a type 2 diabetic obesity model. Adipose tissue was isolated from Lewis rats (LEW/CrlCrlj) and EGFP rats [SD-Tg (CAG-EGFP)] and used to prepare cell sheets.

Rat ASCs (rASCs) were isolated from the inguinal adipose tissue of Lewis rats (20–33 weeks old, male), which was processed according to a previously reported method (23). Briefly, the isolated adipose tissue was enzymatically digested with 0.1% type A collagenase (Roche Diagnostics, Mannheim, Germany) at 37°C for 1 h. The stromal vascular fraction was collected after centrifugation at 700g for 5 min. Cells in the stromal vascular fraction were plated on a 60-cm² Primaria tissue culture dish (BD Biosciences, Franklin Lakes, NJ) and cultured in complete culture medium (MEM α with GlutaMAX [Invitrogen, Carlsbad, CA] with 20% FBS [Moregate Biotech, Bulimba, QLD, Australia] and 1% penicillin/streptomycin [Sigma-Aldrich, St. Louis, MO]) at 37°C in a 5% CO₂ incubator. After 24 h, debris was removed by washing with PBS (Life Technologies, Grand Island, NY), and fresh medium was added. The cells were passaged with 0.25% trypsin-EDTA (Life Technologies) on day 3 and transferred to a new dish. Subcultures were plated at a density of 1.7 × 10³ cells/cm² every 3 days until passage 3.

rASCs were characterized by measuring their colony-forming, adipogenic, and osteogenic abilities using previously reported methods (24). For each assay, 100 rASCs at passage 3 were plated in a 60-cm² dish and cultured in complete medium for 7 days. For the colony-forming assay, the cells were fixed with 4% paraformaldehyde (PFA) (Muto Pure Chemicals, Tokyo, Japan) and strained with 0.5% crystal violet in methanol for 5 min.

For adipogenesis, the medium was switched to an adipogenic medium consisting of complete medium supplemented with 0.5 μmol/L dexamethasone (Fuji Pharma, Tokyo, Japan), 0.5 mmol/L isobutyl-1-methyl xanthine (Sigma-Aldrich), and 50 μmol/L indomethacin (Wako Pure Chemical Industries, Osaka, Japan). After 14 days, the cells were fixed with 4% PFA for at least 1 h and stained for 2 h with fresh Oil Red O solution (Wako). For osteogenesis, the medium was switched to calcification medium consisting of complete medium supplemented with 100 nmol/L dexamethasone, 10 mmol/L β-glycerophosphate (Sigma-Aldrich), and 50 μmol/L ascorbic acid (Wako). The cells were incubated for 21 days and then stained with 1% alizarin red S solution. The number of stained colonies was counted for each assay (n = 3).

One million rASCs at passage 4 were suspended in 100 μL PBS containing 10 μg/mL fluorescein isothiocyanate–conjugated primary antibodies to characterize their surface marker expression (Table 1). After incubation for 30 min at 4°C, the cells were washed with PBS and then suspended in 1 mL PBS for analysis. Cell fluorescence was evaluated using a Gallios flow cytometer (Beckman Coulter, Tokyo, Japan), and data were analyzed using Kaluza for Gallios software (Beckman Coulter).

rASCs derived from Lewis rats or EGFP rats at passage 3 were seeded on 35-mm diameter temperature-responsive culture dishes (UpCell; CellSeed, Tokyo, Japan) at a density of 1.5 × 10⁵ cells/dish and cultured in complete medium for 3 days. The cells were then cultured in complete medium with 16.4 μg/mL ascorbic acid for an additional 4–5 days. After reducing the temperature to room temperature, the cells spontaneously detached as contiguous cell sheets and were harvested from the dishes with forceps.

ZDF rats (n = 48; 16 weeks old; male; 520–600 g) were used as a rat wound-healing model for type 2 diabetes and obesity. rASC sheets from Lewis rats were used for transplantation (Fig. 1A–D). The blood glucose levels of the ZDF rats were measured using a blood glucose monitor (Glutest Neo Sensor; Sanwa Kagaku Kenkyusho, Nagoya, Japan), and their body weights were monitored both before the operation and at kill.

ZDF rats were anesthetized by inhalation of 4% isoflurane (Pfizer Japan, Tokyo, Japan) and ventilated with a rodent mechanical ventilator (Stoelting, Wood Dale, IL). Then, 15 × 10-mm full-thickness skin defects were created on the heads of the ZDF rats by removing the cutaneous tissue from the epidermis to the periosteum (25), and the wound healing of the defects was observed at predetermined time points. ZDF rats were randomly divided into two groups—a control group and an rASC sheet transplantation group—to investigate the efficacy of cell sheets for wound healing. In the transplantation group, an rASC sheet was placed on the defect. The defects with or without an rASC sheet were covered with 15 × 10 mm of artificial skin (PELNAC; Smith & Nephew, Tokyo, Japan). Defects were closed with 10 stitches using 5-0 silk sutures (Alfresa, Osaka, Japan) (Fig. 1D). To protect the wound, 20 × 15 mm of nonadhesive dressing (Hydrosite Plus [known as ALLEVYN Non-Adhesive in the U.S.]; Smith & Nephew) was placed on top of the artificial skin with 5-0 silk sutures to maintain a moist wound environment and absorb exudate.

EGFP-expressing rASCs were obtained from EGFP rats to track the fate of the transplanted cells. EGFP-expressing rASC sheets were created and transplanted into full-thickness defects in ZDF rats (16–18 weeks old; male; n = 2) using the aforementioned procedure.

Gross wounds were observed and photographs taken at 0, 3, 7, 10, and 14 days (n = 12) after the operation and every 3–7 days thereafter (n = 6) until complete wound closure was observed (42 days). The wound area was measured by tracing the wound margin on the photograph and calculating the pixel data using ImageJ software (National Institutes of Health, Bethesda, MD). The time to complete wound closure was assessed (n = 6).

Three rats from each group were killed 3, 7, 10, and 14 days after the operation. The wound areas were excised for histological analysis, and the blood vessel density was analyzed immunohistochemically. After fixation with 4% PFA for 24 h at 4°C, the trimmed specimens were decalcified with Morse’s solution (10% sodium citrate [Wako] and 22.5% formic acid [Wako]) (26) for 3 days at 4°C with gentle agitation. The Morse’s solution was exchanged daily. After decalcification, the specimens were rinsed with PBS, embedded in paraffin, cut into 6-μm–thick sagittal sections, and stained with hematoxylin-eosin (H-E). The sections of the wound area were then observed for histological analysis.

To quantify vascularization within the wound, blood vessel endothelial cells were immunohistochemically stained with an anti-CD31 antibody (rabbit polyclonal antibody; Thermo Fisher Scientific Anatomical Pathology, Fremont, CA). The specimens were pretreated through heating followed by blocking with 1% BSA (Sigma-Aldrich) and then incubated with the anti-CD31 antibody at 4°C overnight. After primary antibody staining, the specimens were washed with PBS, incubated with a secondary antibody (Alexa Fluor 488 Goat Anti-Rabbit IgG [H+L] Antibody; Life Technologies), mounted on coverslips with Prolong Gold, and stained with DAPI (Invitrogen). Serial sections of the specimens were observed with a fluorescence microscope (U-RFL-T; Olympus, Tokyo, Japan), and the images were analyzed with application software (DP2-BSW; Olympus).

The blood vessel densities of six animals in each group were determined 14 days after the operation by measuring the area of CD31-positive vessels in the wound area of the specimen. The center of the wound in each section was selected. The vessel area in the selected field of each specimen was then observed with a microscope (Eclipse E800; Nikon, Tokyo, Japan), and the area of CD31-positive vessels was quantified using ImageJ software. The relative area of CD31-positive vessels (VA) was calculated using Eq. 1:

where VAact and Af are the actual area of CD31-positive vessels and the total area of the field, respectively.

EGFP-expressing rASC sheets were transplanted to trace the fate of the transplanted rASC sheets. At 3, 5, 7, and 14 days after the transplantation, the rats’ chests were opened, and their vasculature was perfused from the left ventricle with 100 mL of 2% PFA in 0.01 mol/L PBS (pH 7.4) at a pressure of 120 mmHg. The wound tissue, including the skull bone and surrounding cutaneous tissue, was then removed and soaked in 4% PFA for 24 h at 4°C. The specimens were subsequently washed with PBS and decalcified with 10% EDTA weight for volume (Wako) in PBS for 5 days at 4°C with gentle agitation. The EDTA solution was changed daily. After decalcification, the specimens were rinsed >10 times with PBS, cryoprotected in an ascending series of sucrose-PBS (15 and 30 weight for volume percent), embedded in optimal cutting temperature compound (Sakura Finetek Japan, Tokyo, Japan), and snap-frozen in liquid nitrogen. Then, 14-μm–thick frozen sections were prepared with a cryostat (Leica CM1850; Finetek) and processed for immunohistochemistry. The frozen sections were first incubated with 4% Block-Ace (Dainippon Sumitomo Seiyaku, Osaka, Japan) and then incubated with rat antiendothelial cell antibody-1 (RECA-1) (Abcam, Cambridge, U.K.) as the primary antibody at a dilution of 1:200 in PBS containing 1% BSA (Sigma-Aldrich) at 4°C overnight. After several PBS washes, the specimens were incubated with a Cy3-conjugated secondary antibody at a 1:200 dilution (Jackson ImmunoResearch Laboratories, West Grove, PA) for 3 h at room temperature. Finally, the specimens were observed and photographed with a Keyence BZ-9000 fluorescence microscope (Keyence Corp., Osaka, Japan).

rASCs at passage 3 were seeded onto temperature-responsive culture dishes at a density of 1.5 × 10⁵ cells/dish and cultured at 37°C for 72 h. The cells achieved subconfluence (80–90%) at 72 h after seeding, and this time point was defined as day 0. The medium was then collected and replaced with complete medium containing 16.4 μg/mL ascorbic acid on days 0, 2, and 4. The rASC sheets were transplanted on day 4 in this study. After the collected medium was centrifuged at 300g for 3 min at 4°C, the supernatant was collected, immediately frozen, and maintained at −80°C for further experiments.

The number of cultured cells per dish was assessed on days 0, 2, and 4 (n = 4). The levels of vascular endothelial growth factor (VEGF), hepatocyte growth factor (HGF), transforming growth factor-β1 (TGF-β1), IGF-I, epidermal growth factor (EGF), and keratinocyte growth factor (KGF) were determined using RRV00, MHG00, MB100B, MG100 (R&D Systems, Minneapolis, MN), SE560Ra (Cloud-Clone Corp., Houston, TX), and CSB-E12905r (CUSABIO, Wuhan, China) quantitative ELISA kits, respectively. The protein concentrations were measured by duplicate ELISA assays, which were performed according to the manufacturer’s instructions. The amounts of the various growth factors in the supernatant were calculated to examine the amount secreted from the rASC sheets at each time point.

Data are expressed as the mean ± SD. Comparisons between the two groups (i.e., transplant, control) were performed using Student t test. P < 0.05 was considered statistically significant.

To evaluate the characteristics of the rASCs, crystal violet, Oil Red O, and alizarin red S staining were performed to confirm rASC self-renewal, adipogenesis, and osteogenesis, respectively. rASCs exhibited colony-forming potential (Fig. 2A) as well as adipogenic and osteogenic differentiation potential in vitro (Fig. 2B and C), suggesting that the rASCs used in this study possessed MSC-like properties.

Flow cytometry was used to characterize cell surface markers (Fig. 2D–H). rASCs strongly expressed CD29 and CD90. In contrast, low expression levels of CD11b, CD31, and CD45 were observed.

ZDF rats in each group exhibited an average blood glucose level >250 mg/dL, and these levels were ∼2.5-fold higher than those observed in 16-week-old nondiabetic rats. ZDF rats exhibited an average body weight of 578.2 g in the transplantation group and 582.6 g in the control group. These body weights were ∼1.5-fold higher than those observed in 16-week-old nondiabetic rats.

Digital photographs were obtained 0, 3, 7, 10, and 14 days after the operation and every 3–7 days thereafter until complete wound closure (Fig. 3A–L). The average wound area was significantly smaller in the transplantation group than in the control group beginning on the 7th day (Fig. 4A). The wounds in both groups were digitally photographed and observed until complete wound closure (n = 6). The average observed wound closure time was significantly reduced in the transplantation group compared with the control group (Fig. 4B).

In low-magnification microphotographs of sagittal sections of H-E–stained specimens collected 14 days after the operation (Fig. 5A–D), a thin dermal layer was observed for the control group (Fig. 5A). In contrast, dense connective tissue was observed in the transplantation group (Fig. 5B). Higher-magnification microphotographs indicated the same results (Fig. 5C and D).

Blood vessel density was quantified 14 days after the operation (Fig. 5E–G). The blood vessel density in the wound was increased by ∼2.5-fold in the transplantation group compared with the control group after 14 days, and the difference was statistically significant (n = 6) (Fig. 5G).

The average number of cultured cells per dish was 8.76 × 10⁵ cells on day 4 at the time of transplantation (Fig. 6A).

VEGF, HGF, TGF-β1, IGF-I, EGF, and KGF were present in the conditioned medium from cultured rASCs at passage 3 (Fig. 6B–G). Specifically, the levels of VEGF, HGF, TGF-β1, IGF-I, EGF, and KGF per rASC sheet were 18.60, 3.27, 0.76, 2.78, 0.44, and 0.42 ng/sheet/day at day 4 (n = 4), respectively, as determined using ELISA kits for each protein. These results confirmed that the rASC sheets were secreting growth factors at the time of transplantation.

EGFP-expressing rASC sheets were transplanted into the wounds (Fig. 7A and B). Fourteen days after the transplantation, the transplanted EGFP-expressing rASCs (stained green [Fig. 7C]) and endothelial cells of regenerated blood vessels (stained red with RECA-1 [Fig. 7D]) were visualized. Overlaid microphotographs revealed that EGFP-expressing rASCs were located on the periphery of the RECA-1–positive blood vessels (Fig. 7E).

Although streptozotocin-induced type 1 diabetic rats without obesity are most frequently used for diabetic wound-healing studies, this study used a wound-healing model with type 2 diabetes, which is known to affect >90% of patients with diabetes. Specifically, ZDF rats were used as a wound-healing model of type 2 diabetes and obesity in this study. These rats spontaneously develop obesity at ∼4 weeks of age and then develop type 2 diabetes at 8–12 weeks, exhibiting hyperglycemia combined with insulin resistance, dyslipidemia, and hypertriglyceridemia (27). Delayed wound healing, impeded blood flow in peripheral blood vessels, and diabetic nephropathy are also observed (28–30).

ASC sheets have been used to accelerate wound healing in previous studies. Lin et al. (31) and McLaughlin and Marra (32) reported that transplantation of human ASC sheets with fibrin-coated membranes accelerates wound healing in a 12-mm–diameter full-thickness skin defect on the backs of 6-week-old athymic nude mice. Cerqueira et al. (33) reported that human ASC sheet transplantation promotes wound regeneration in 10-mm–diameter full-thickness skin defects on the backs of 5-week-old normal mice. In contrast to previous studies, the current study created a wider full-thickness skin defect with exposed bone (wound area 15 mm²) on the parietal region of 16-week-old ZDF rats. In addition, the periosteum, which may support the regeneration of the skin, was removed to increase the severity of the wound (25). Diabetic wounds with exposed bone are often observed clinically after debridement. To assess the clinical applicability of the method for severe diabetic foot ulcers, this study attempted to produce a wider full-thickness skin defect model with exposed bone and removed periosteum in rats with type 2 diabetes and obesity.

The differences in wound-healing mechanisms noted between humans and rodents are attributed to anatomical differences of the skin (e.g., panniculus carnosus layer). Normal rats exhibit contraction-based wound healing, whereas humans display re-epithelialization and granulation tissue formation–based wound healing. Typically, wound splinting in rodent models helps to minimize wound contracture, which allows for the gradual formation of granulation tissue (34). Slavkovsky et al. (28) reported that the contraction of diabetic wounds in ZDF rats is impaired, whereas nondiabetic wounds are almost entirely closed by contraction. Epithelialization also contributes to repair in this diabetic wound model in ZDF rats. Furthermore, the bilayer artificial skin we used in the current study prevented wound contraction and promoted new connective tissue matrix synthesis, resembling the true dermis (35). In the current study, artificial skin was placed on a recipient wound and fixed with nylon threads to reduce wound contraction and prevent wound enlargement as a result of the rat’s loose skin. Even if considerable contraction of the wound was observed, it appeared similar across all conditions between the control and the transplantation groups except in the presence or absence of ASC sheet transplantation. Under these conditions, significant differences in wound area and vascular density were observed between the groups. This experiment may be useful for evaluating ASC sheets.

Estrogen deficiency is associated with impaired cutaneous wound healing. Female rats exhibit a faster rate of wound healing with a higher rate of wound contraction because of their thinner skin (36). Accordingly, to reduce potential hormonal influences on wound healing, only male rats were used in this study.

ASCs exhibit several advantages compared with MSCs derived from other tissues (12). Adipose tissue is relatively abundant in the human body. These tissues are accessible and can be collected using a minimally invasive liposuction procedure. ASCs are also abundant in adipose tissue (11). Moreover, many reports have suggested that ASCs play an important role in accelerating wound healing (37,38). Thus, ASCs are considered a good cell source for stem cell therapies.

This study investigated whether allogeneic ASC sheets together with artificial skin are effective at accelerating wound healing. Xing et al. (39) reported that postoperative wound healing after laparotomy is impaired in obese rats, suggesting that the risk of laparotomy dehiscence may be increased by obesity. In clinical practice, patients with diabetic ulcers often exhibit severe diabetes complications, including uncontrolled high blood glucose and a high BMI. The collection of adipose tissue from patients with acute diabetic ulcers is difficult. Furthermore, after obtaining autologous ASCs, ASC sheet preparation requires several weeks. rASCs from diabetic animals exhibit altered properties and impaired functions (40). In fact, our group attempted to create rASC sheets using rASCs from ZDF rats, but cell proliferation was slow (data not shown). Given that MSCs, including ASCs, suppress immune responses (41), rASC sheets prepared from normal rats were transplanted into defects in ZDF rats to study allogeneic transplantation. In the current study, accelerated wound healing was observed in the transplantation group, and no obvious clinical signs of immune rejection, such as erythema, other local inflammatory signs, or visible signs of necrosis as described by Falanga et al. (42) and Briscoe et al. (43), were observed in the transplantation group during the experimental period.

rASC sheets were successfully created from Lewis rats exhibiting a wide range of ages (8, 10, 16, 20, and 33 weeks) (data not shown), and rASC sheets from 20–33-week-old rats were used for the transplantations. This study showed that rASC sheets from 33-week-old rats secreted growth factors and accelerated wound healing compared with the control group, suggesting that adipose tissue from middle-aged rats created functional rASC sheets. ASC sheets from patients over a wide range of ages could potentially be useful for treatment in clinical settings. In clinical practice, the age of patients undergoing liposuction operations ranges from young to elderly, and most patients are adults. This study intentionally used cells from middle-aged rats to simulate actual clinical practice.

In this study, rASC sheets contained ∼9 × 10⁵ rASCs at the time of transplantation. A previous study based on the topical administration of MSCs in a collagen scaffold reported that 1 × 10⁶ MSCs were transplanted into a 6-mm–diameter wound, which was the highest dose in that experiment. The highest percentage of wound closure and angiogenesis was achieved in 1 week with this cell dose compared with lower doses, suggesting that the wound-healing effects of MSCs may be dose dependent (44). In the current study, approximately the same number of rASCs was transplanted into larger defects (15 × 10 mm) as a cell sheet, and wound closure in the transplantation group was significantly accelerated 2 weeks after the operation. These results suggest that cell sheet technology might enhance the cellular activity and efficacy of cell transplantation because cell sheets are harvested with intact cell-cell junctions and undamaged ECM and can be immediately transplanted into the defect site (19). Moreover, a cell sheet is thin and exhibits sufficient flexibility to allow it to adhere well to the rough surface of a wound. In fact, cell sheet technologies have been used for the reconstruction of various tissues, with curative effects (45,46).

In this study, the time to complete wound closure was significantly faster in the transplantation group, and increased blood vessel density was confirmed in the transplantation group 14 days after transplantation compared with the control group. The regeneration and introduction of capillaries inside the wound are important because tissue regeneration commonly requires blood flow to supply oxygen and nutrition and remove waste. Increased blood vessel density at the wound site may help to promote wound healing.

Wound healing requires complex processes, including a coordinated interplay between cells and growth factors; it also requires multiple growth factors acting synergistically (47). A previous study reported that the rate of VEGF secretion from surrounding cells, such as macrophages and fibroblasts, is reduced in diabetic wounds (48). Numerous physiological factors contribute to wound-healing deficiency in individuals with diabetes (48).

In this study, various secreted growth factors (VEGF, HGF, TGF-β1, IGF-I, EGF, and KGF) played important roles in wound repair and were detected in large quantities in the conditioned medium from rASC sheets. ASCs secrete angiogenic growth factors (49), such as VEGF and HGF (50), and these growth factors could contribute to neovascularization (51,52) and accelerate wound healing in both normal and diabetic rats (15). These findings, which are consistent with the current results, suggest that the paracrine effects of transplanted rASCs might enhance neovascularization and wound epithelialization.

In this study, EGFP-expressing rASCs were surrounded with blood vessel endothelial cells 14 days after transplantation, suggesting that the rASCs remained at the wound site for at least 14 days and were able to differentiate into mural cells at the wound site and support the reconstitution of new blood vessels (53,54). MSCs originate from and are natively associated with blood vessel walls, suggesting that MSCs belong to a subset of perivascular cells (55,56). Some studies reported that ASCs have the potential to differentiate into endothelial cells (57,58). Also previously reported was that the majority of transplanted ASCs are adjacent to microvessels and differentiate into pericytes (59,60).

The results from this animal study demonstrate a promising approach for the application of ASC sheets together with artificial skin to accelerate diabetic wound healing. However, additional studies in larger animals are necessary to confirm and validate the efficacy of ASC sheets with artificial skin for clinical trials.

In conclusion, this study demonstrated that allogeneic transplantation of ASC sheets in combination with artificial skin accelerates wound healing in a rat model of type 2 diabetes and obesity. The rASC sheets displayed effective paracrine activity because they secreted angiogenic growth factors and potentially differentiated into perivascular cells to support the construction of new blood vessels. Thus, ASC sheets exhibited therapeutic value in this rat model and may be useful for accelerating the healing of diabetic ulcers in patients with type 2 diabetes and obesity.

Acknowledgments. The authors thank Toshiyuki Yoshida and Kaoru Washio of the Institute of Advanced Biomedical Engineering and Science, Tokyo Women’s Medical University, for valuable advice and suggestions; Hozue Kuroda of the Institute of Advanced Biomedical Engineering and Science, Tokyo Women’s Medical University, for excellent technical support; Taichi Ezaki of the Department of Anatomy and Developmental Biology, Tokyo Women’s Medical University School of Medicine, for expert technical advice and support in immunohistochemistry; Yukiko Koga of the Department of Plastic and Reconstructive Surgery, Juntendo University School of Medicine, for practical advice; and Norio Ueno, of the Institute of Advanced Biomedical Engineering and Science, Tokyo Women’s Medical University, for English editing and valuable advice. They also thank Yuka Tsumanuma and Supreda Suphanantachat of Tokyo Medical and Dental University; Issei Komatsu and Daisuke Fujisawa of the Institute of Advanced Biomedical Engineering and Science, Tokyo Women’s Medical University; Nana Mori of the Department of Odontology, Periodontology Section, Fukuoka Dental College; and Aya Matsui of the Department of Anatomy and Developmental Biology, Tokyo Women’s Medical University School of Medicine, for technical support. Finally, the authors thank Kazuki Ikura of the Diabetic Center, Tokyo Women’s Medical University School of Medicine, for expert advice on diabetic feet.

Funding. This study was supported by the Creation of Innovation Centers for Advanced Interdisciplinary Research Areas Program of the Project for Developing Innovation Systems “Cell Sheet Tissue Engineering Center (CSTEC)” from the Ministry of Education, Culture, Sports, Science and Technology (MEXT), Japan.

Duality of Interest. T.O. is a founder and director of the board of CellSeed, Inc., and holds technology licensing and patents from Tokyo Women’s Medical University. T.O. also is a stakeholder of CellSeed, Inc. Tokyo Women’s Medical University receives research funds from CellSeed, Inc. No other potential conflicts of interest relevant to this article were reported.

Author Contributions. Y.K. wrote the manuscript and researched data. T.I. and S.M. researched data, contributed to the discussions, and reviewed and edited the manuscript. M.Y., T.O., and Y.U. participated in discussions and reviewed and edited the manuscript. T.I. and T.O. are the guarantors of this work and, as such, had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Prior Presentation. Parts of this study were presented in abstract form at the 74th Scientific Sessions of the American Diabetes Association, San Francisco, CA, 13–17 June 2014.