The Co-Culture of ASCs and EPCs Promotes Wound Healing by Promoting Angiogenesis in Full Thickness Skin Defects


 Background The repair of large-scale full-thickness skin defects represents a challenging obstacle in skin tissue engineering. To address the most important problem in skin defect repair, namely insufficient blood supply, this study aimed to find a method that could promote the formation of vascularized skin tissue. Method The phenotypes of ASCs and EPCs were identified respectively, and ASCs/EPCs were co-cultured in vitro to detect the expression of dermal and angiogenic genes. Furthermore, the co-culture system combined with dermal extracellular matrix hydrogel was used to repair the full-scale skin defects in rats.Result The co-culture of ASCs/EPCs could increase skin and angiogenesis-related gene expression in vitro. The results of in vivo animal experiments demonstrated that the ASCs/EPCs group could significantly accelerate the repair of skin defects by promoting the regeneration of vascularized skin.Conclusion It is feasible to replace traditional single seed cells with ASC/EPC co-culture system for vascularized skin regeneration. This system could ultimately enable clinicians to better repair the full-thickness skin defects and avoid donor site morbidity.


Introduction
The repair of large-scale full-thickness skin defects represents a challenging problem in skin tissue engineering [1,2]. At present, skin tissue engineering technology has made great progress, but there are still some obstacles, such as obvious scarring of new skin, large amount of tissue shrinkage, slow skinization speed, and di culty in constructing full-thickness skin tissue [3][4][5]. The main reason is that the implanted tissue-engineered skin lacks angiogenesis and is unable to obtain su cient nutrition, which limits the repair of large-scale full-thickness skin defects. Previous studies have con rmed that angiogenesis is an important factor affecting the process of skin formation in the body, and capillaries are involved in the process of skin regeneration [6]. At the site of new skin formation, broblasts were found near vascular endothelial cells, suggesting that angiogenesis and skin regeneration are interdependent [7]. Meanwhile, in the process of skin formation, restricted angiogenesis will lead to poor skin healing, increased scar tissue, and delayed wound healing. In the repair of small-scale skin defects, skin tissue has a strong ability to regenerate. This regenerative ability is achieved by the vascular system around the defect area, which continuously recruits nearby broblasts and related bioactive factors for skin repair and provides nutritional support for the repair process [8]. However, in the case of large-scale skin defects, severe damage to the surrounding vascular system will affect the recruitment of nearby broblasts and the transportation of nutrients, resulting in the limitation of the regeneration process of skin tissues and failure to complete the repair of large-scale skin defects [9]. Therefore, in the process of repairing skin defects, it is essential to achieve vascularized skin regeneration.
The application of endothelial progenitor cells (EPCs) is a potential method to achieve vascularization [10]. EPCs are precursor cells of vascular endothelial cells and have the ability to proliferate, migrate and differentiate into cells arranged along the lumen of blood vessels [11]. Previous studies have found that EPCs participate in the process of skin formation and repair, and vascular endothelial cells enhance the activity of pre-broblasts and play an important role in the process of skin regeneration [12,13]. Adipose-derived stem cells (ASCs) are extracted from adipose tissue and have high proliferation and growth characteristics and multi-differentiation potential [14]. Compared with other tissue-derived stem cells, ASCs have the unique advantages of extensive tissue sources, easy availability, less damage to the donor, and high safety. As a type of seed cell for skin tissue engineering, ASCs has been proven to promote skin tissue regeneration [15]. As con rmed in previous studies, ASCs can accelerate the recruitment of EPCs, enhance their angiogenesis capability, and promote the formation of new blood vessels. Besides, ASCs can secrete a variety of biologically active factors such as vascular endothelial growth factor (VEGF) and epidermal growth factor (EGF), which are bene cial to enhance the viability of endothelial cells to promote angiogenesis [16,17]. Therefore, co-culture of ASCs/EPCs may promote the formation of vascularized skin through cell-cell interaction.
To study the effect of co-culture of ASCs and EPCs on the construction of vascularized skin tissue, the phenotype of ASCs and EPCs was rst identi ed, and then it was determined whether the co-culture of ASCs and EPCs could promote the differentiation of skin and vascular endothelial cells in vitro.
Furthermore, ASCs/EPCs co-culture system combined with dermal ECM gel material was used to repair severe skin defects in rats, and analyzed by gross, histology and immuno uorescence assays. The results indicated that the co-culture of ASCs and EPCs could enhance skin regeneration and angiogenesis, and signi cantly promote the repair of large-scale skin defects. It is feasible to use ASCs/EPCs co-culture system to replace traditional single-seed cells for skin tissue engineering.
Materials And Method

Preparation and culture of ASCs and EPCs
Animal procedures were conducted in accordance with the protocol approved by IACUC of China Medical University. Sprague-Dawley (SD) rats, 4-week-old, were euthanized by CO 2 . The subcutaneous white adipose tissue of the inguinal region was harvested, washed with phosphate buffer saline (PBS), and minced by sterile surgical scissors. The minced tissue was digested with 0.1% collagenase type (Gibco) at 37℃ for 1 h and then centrifuged at 1000 rpm for 5 min. The upper layer of undigested fat and brous connective tissue was removed, and the cell pellet in the lower layer was resuspended and ltered through a 70 µm lter. After another centrifugation for 5 min, the cell pellet was resuspended in the culture medium consisting of low-glucose Dulbecco's Modi ed Eagle Medium (DMEM) (Hyclone), 10% fetal bovine serum (FBS) (Gibco), and 1% penicillin/streptomycin. The cell suspension was seeded into a 25 cm 2 ask and cultured at 37℃ with 5% CO 2 in a humidity atmosphere [13]. The medium was changed 24 h after inoculation, and then replaced every 2-3 days, and the cells were passaged after 80% con uence. The ASCs at passage 3 were collected for the following experiments.
After sacri ce with CO 2 , the bone marrow in the femurs of 4-week-old SD rats was blown out and beaten evenly. Following the centrifugation with Histopaque-1083 (Sigma), the cell pellets were collected and resuspended in mononuclear cell separation solution. Then CD34 + cells were separated and harvested by immunomagnetic bead method. After resuspension in EGM medium( Lonza, Cologne, Germany), the cells were seeded into asks and cultured at 37℃ with 5% CO 2 in a humidity atmosphere. The culture medium was replaced 24 h to remove nonadherent cells. After culture and passage, EPCs at passage 3 were used for the further experiments [18,19]. 1×10 6 ASCs and EPCs at passage 3 were harvested, washed with 10% FBS/PBS and centrifuged at 1000 rpm for 5 min to collect cell pellets.

Flow Cytometry Analysis and Immuno uorescence Staining
Identi cation of ASCs surface markers: FITC-labeled antibody was added to the cell suspension, incubated, centrifuged, washed, and resuspend the cells. Then, ow cytometry was used to detect surface markers including CD73, CD90, CD45, and CD34. Moreover, the immuno uorescence staining was performed to observe the cell surface markers CD73 and CD90 of ASCs. The samples were xed and processed with 0.3% TritonX-100. After blocking with goat serum at room temperature, a uorescentlabeled primary antibody was added, then incubated, washed, and nally counterstained with DAPI [15].
Identi cation of EPCs surface markers: EPCs were added with FITC-labeled primary antibodies followed by incubation, centrifugation and wash. Then, surface markers including CD133, CD11b, CD34, CD31 were detected by ow cytometry. In addition, for lectin staining of EPCs, antibody was added, then incubated in the dark, xed, washed with PBS, and observed under a uorescence microscopy.

Preparation of dermal extracellular matrix (dECM) hydrogel
dECM was prepared as previously described. In brief, full thickness skin was obtained within 1 hour following the slaughtering of the healthy pigs (average age 6-8 months and weight 110-120 kg) at a local slaughter house located in Shenyang, China. The epidermis, subcutaneous fat and connective tissue were removed by mechanical layering to separate the dermis. Dermis was treated with the following solutions under constant agitation on an orbital shaker at 300 RPM:0.25% trypsin at 4 °C for 6 h ,three washes with deionized water for 15 min, 70% ethanol for 10 h, 3% H 2 O 2 for 15 min, three washes with deionized water for 15 min, 1% Triton X-100 in 0.26% EDTA/0.69% Tris for 22 h with a fresh change at the 6th hour, another three washes with deionized water, 0.1% peracetic acid/4% ethanol for 2 h, two washes with PBS, and three washes with deionized water at end.
Subsequently, dECM were frozen and lyophilized for 24 h to prepare the hydrogel. After grinding and sieving through a 40 mesh screen, the powdered ECM was then enzymatically digested in a solution of 1 mg/ml pepsin (Sigma) in 0.01 M hydrochloric acid under a constant stirring for 48 h at room temperature. Then, the acidic digest solution with a concentration of 10 mg/ml was prepared for subsequent experiments. Prior to the gelation, neutralization was performed by the addition of one-tenth the digest volume of 0.1 M sodium hydroxide and one-ninth the volume of 10⋅PBS on ice, and the neutralized solution obtained was the pre-gel. The pre-gel was placed at 37 °C for 30 minutes for gelation to occur.

Live/Dead staining
In the live/dead cell staining, calcein AM is capable of permeating the membrane of viable cells, where it is cleaved by intracellular esterase and produces a green uorescence. Ethidium bromide homodimer-1 is able to enter cells with damaged membranes and bind to fragmented nucleic acids, thereby producing red uorescence in dead cells. On days 1, 3, and 7, the cells were gently rinsed with sterile PBS, and incubated with a Live/Dead Assay Kit for 20 minutes in an incubator. Sections were immediately transferred and imaged with an inverted uorescence microscope with a confocal imaging system.

Co-culture of ASCs and EPCs in vitro
To determine the optimal ratio of EPCs and ASCs in skin tissue regeneration, seven groups were divided for the next experimental observation, including ASCs alone, EPCs alone, and EPCs to ASCs at ratios of 2:1, 1: 1, 1: 2, 1: 5, and 1:10. Cells of each groups were seeded in 12-well plates at the density of 1 × 10 5 cells per well and induced with dECM medium (cells inoculated in dECM material and cultured in LDMEM containing 10% FBS) or L-DMEM (cells inoculated and cultured in L-DMEM containing 10% FBS) for 7 days, which was prepared detecting the expression of CK19 and vimentin by immuno uorescence.

Quantitative reverse transcription-polymerase chain reaction (qRT-PCR)
Total RNA was extracted from the cells of different groups using Trizol reagent (Invitrogen) according to manufacturer's instructions. Reverse transcription of total RNA into cDNA was performed by RT-PCR (Invitrogen) using the reverse transcription rst chain synthesis system. PCR ampli cation with speci c primers was employed to analyze the expression of epidermal cell marker genes including keratin (CK5, CK19), Col I, Col III, vascular endothelial growth factor (VEGF), von Willebrand factor (vWF). Following the manufacturer's instructions, real-time PCR was performed using SYBR GREEN PCR Master Mix on ABI PRISM 7500 sequence detection system. PCR conditions were 94 °C 1 min, 95 °C 30 s, 58 °C 40 s, and 72 °C 1 min 30 s with a total of 40 cycles. All reactions were repeated three times and normalized to GAPDH. Comparative CT was calculated to evaluate the relative difference of PCR results of each group.

Matrigel tubule formation assay
Prior to the assay, the Matrigel (BD Corporation, USA) was moved from − 20 °C to 4 °C to fully melt. Then 50 µL of Matrigel was added to each well of a 96-well plate and placed at 37 °C for 2 h to coagulate. The cells of seven groups were seeded to the 96-well plate containing matrigel (5 duplicate wells per group), placed in a 37 °C, 5% CO2 incubator, and observed under a light microscope every 3 h. When obvious blood vessels were formed, photographs were taken immediately to record. The number and relative length of each component tube were analyzed by Image-Pro Plus 6.0 software.

Establishment of wound model
Forty-eight male Sprague Dawley rats (8 weeks old) weighing 200-250 g were anesthetized by intraperitoneal injection of 1% sodium pentobarbital solution (40 mg/kg body weight). Rats were randomly divided into four groups (n = 10 each group) according to the construct used to repair the wound: Black, ECM + EPCs (ECM + E), ECM + ASCs (ECM + A), ECM + ASCs + EPCs (ECM + A + E). The cells of each group were mixed with sterilized dECM hydrogel at a concentration of 1⋅10 6 /ml to prepare an injectable gel scaffold for use. Each wound was covered with a mixed gel material of 2 ml hydrogel and cells, and bandaged after surgery. The rats were given intramuscular injection of anti-in ammatory and analgesic drugs and were kept separately for observation to avoid biting each other. To evaluate the therapeutic effect, the wound closure of rats was observed and recorded. At 7 and 14 days after operation, 6 rats were taken from each group. After anesthesia, the wound was covered with a sterile plastic lm and the size was drawn with a pencil. The drawn lm was scanned into a computer, and Image-Pro Plus 6.0 image analysis software was used to analyze the contour size and calculate the wound closure rate according to the following formula: (original wound area − wound area after treatment)/original wound area × 100%.

Histological Analysis
On the 14th day after operation, the rats in four groups were anesthetized and sacri ced. The middle part of repair area in each group was trimmed for histological analysis. Following xation in 4% paraformaldehyde for 24 h, dehydrated in a graded series of ethanol, embedded in para n and cut into 4 µm sections. After depara nization,hematoxylin and eosin (HE) staining was performed. A nearsurface eld of view was randomly selected, and low-magni cation eld (10 × 10) was used to choose the area with a large number of in ammatory cells (neutrophils, eosinophils, basophils, lymphocytes and monocytes). Then 5 elds in the high-magni cation eld (10 × 40) were selected to observe and count the number of in ammatory cells in the eld, and the average value was calculated. In addition, Masson trichrome staining was used to analyze morphological characteristics of collagen bers and the healing of skin wounds.

Immuno uorescence analysis of vascularization
Tissue sections on the 14th day after surgery were depara nized with xylene, dehydrated by ethanol, antigen retrieval and serum blocking, and the sections were added with rabbit anti-CD31 polyclonal antibody (Sigma) to label vascular endothelial cells. 3 areas with high microvessel density marked by CD31 were randomly selected in the low magni cation eld of view (10 × 10), and then the microvessel density in 5 elds in each area was counted in the high magni cation eld of view (10 × 20), and the average value was calculated.

Statistical analysis
The test results were expressed as mean ± standard deviation, and statistical analysis was performed by SPSS 21.0 software (IBM, Armonk, New York, USA). T test was used for the comparison of the data of two groups and one-way ANOVA was for the samples of multiple groups. p < 0.05 was considered as statistically signi cant.

Detection and characterization of ASCs and EPCs
The ow cytometry of ASCs demonstrated that ASCs are positive for CD73 and CD90 and negative for CD45 and CD34 (Fig. 1A). The immuno uorescence staining also exhibited that CD73 and CD90 were positive in ASCs (Fig. 1B). Meanwhile, EPCs are positive for CD133 and CD34 and negative for CD31 and CD11b in ow cytometry (Fig. 1C). To further identify the EPC phenotype, lectin staining of EPCs was performed after the cells were cultured for 7 days which was shown in Fig. 1D that lectin staining was positive.

dECM hydrogel promotes cell proliferation and has thermosensitive properties
Stem cells were inoculated and cultured with dECM as a scaffold, and cell viability was detected at three time points, 3, 7, and 14 days. The results shown in Fig. 2A demonstrated that the cell viability was good, indicating that dECM has no cytotoxicity. Meanwhile, it was found that dECM has thermosensitive properties. The dECM pre-gel is liquid at low temperatures and transformed into a gel state at 37˚C, which made it an injectable scaffold material for skin defect repair.

Determination of the optimal ratio of ASCs/EPCs coculture in vitro
To determine the optimal ratio of ASCs/EPCs in vitro co-culture, the expression of epidermal cell markers CK19 and vimentin in different ratios of ASCs/EPCs co-culture system was compared after culture in dECM medium and L-DMEM (Table 1). The comparison results showed that the expression of CK19 and vimentin in ASCs/EPCs co-cultured at a cell ratio of 1:1 in dECM medium were signi cantly higher than that of other groups. With the same cell ratio, the expression of CK19 and vimentin in the dECM medium was higher than that of the L-DMEM group.

ASCs/EPCs co-culture in vitro promotes the expression of dermal tissue markers and angiogenesis-related genes
In order to further study whether the co-culture of ASCs and EPCs affects the expression of epidermal cell marker and endothelial cell marker genes, RT-PCR was used to analyze the main components of dermal tissues including epidermal cell and collagen corresponding markers Ck5, Ck19, Col I, Col III and the expression of endothelial cell marker genes, such as VEGF and vWF. The results showed that the expression levels of Ck5, Ck19, Col I and Col III in the ASC/EPC group (A + E group) were signi cantly higher than those in the ASC group (A group) and EPC group (E group) (Fig. 2B), suggesting that the coculture of ASCs/EPCs could increase the expression of epidermal cell and collagen marker genes. In addition, the mRNA levels of endothelial cell marker genes VEGF and vWF were dramatically higher in the A + E group than in the A group and E group (Fig. 2B), which indicated that the co-culture of ASCs/EPCs could increase the expression of angiogenesis-related genes. These above results demonstrated that the co-culture of ASCs/EPCs could enhance the expression of dermal tissue markers and vascular endothelial cell genes.
Matrigel tubule formation assay demonstrated that the density of tube structure in the ASC/EPC coculture group was dramatically higher than that in the ASC alone or EPC alone group (Fig. 2C). Moreover, the amount of tubular structures increased signi cantly in the ASC/EPC co-culture group compared with the single cell groups (p < 0.05) (Fig. 2D). And the length of the tubular structure formed in the co-culture group was also signi cantly longer than that of the other two groups (p < 0.05) (Fig. 2E).

Co-culture of ASCs and EPCs enhances the repair of full-thickness skin defects in rats
In the comparison of wound healing at different time points after operation, it was observed that the wound healing rate was faster in the ECM + A + E group. (Fig. 3A) (Fig. 3B). In terms of healing rate, ECM + A group is higher than Blank group and ECM + E group, but lower than ECM + A + E group. In addition, compared with other groups, the newly formed skin color and shape of the ECM + A + E group are remarkably closer to normal skin. On the contrary, other groups, especially the Blank group and ECM + E group, had obvious scar tissue formation and contraction.
The HE staining results of the regenerated skin tissue showed that the thickness of the newly formed epithelium in the ECM + A + E group was greater than that of the other groups, and there were skin appendages formed as shown in the Fig. 4A. Meanwhile, it was found that there were different degrees of angiogenesis in each group, and the ECM + A + E group was signi cantly more than the other groups (Fig. 4A). In terms of the thickness of scar tissue in regenerated skin, the ECM + A + E group was signi cantly smaller than the other groups as shown in Fig. 4B (P < 0.05).
The regeneration of collagen tissue was evaluated by masson staining, which indicated that the ECM + A + E group was signi cantly more than other groups as shown in Fig. 5A. The OD values of Col I and Col III in each group were quantitatively measured, suggesting that the average collagen content of ECM + A + E group was signi cantly higher than that of the other groups as shown in Fig. 5B (P < 0.05). The total content of Col I and Col III was measured for comparison and analysis, which demonstrated that the result in the ECM + A + E group was also dramatically higher than that of the other groups as shown in Fig. 5C (P < 0.05).
The blood vessel immuno uorescence assay was performed on the regenerated skin tissue, which showed that the blood vessel uorescence staining range in the ECM + A + E group was larger than that in the other groups as shown in Fig. 6A (the red color in the gure is the blood vessel morphological staining). According to quantitative analysis shown in Fig. 6B, the number of newly formed blood vessels in the ECM + A + E group was signi cantly higher than that in the other groups (P < 0.05).

Discussion
There are still some problems to be solved in the reconstruction of large-scale full-thickness skin defects, including slow wound healing, obvious scars, and obvious skin tissue shrinkage, which is attribute to the lack of vascularized skin tissue regeneration, affecting the repair effect of skin defects [20]. To construct skin tissues accompanying the vascular system, our team co-cultured ASCs and EPCs to establish a dual stem cell system. This study found that the ASCs/EPCs co-culture system could enhance the expression of skin and angiogenesis genes in vitro, and by supporting the regeneration of vascularized skin, it signi cantly accelerated the wound healing of skin defects in vivo.
EPCs are the precursor cells of vascular endothelial cells, which have the ability to proliferate, migrate and differentiate into cells arranged along the vascular cavity, and can be isolated from peripheral blood and spleen. ASCs also have high proliferation and growth characteristics and multi-differentiation potential, which can be extracted from autologous subcutaneous fat and are the most easily obtained autologous stem cells. But so far, speci c markers for each cell type are still lacking. Many surface proteins have been used to identify rat ASCs, including CD73, CD90, CD105, CD44, etc. [21,22]. In this study, CD73 and CD90 are used as positive markers for ASCs. The results showed that ASCs expressed CD73 and CD90 positive cell surface proteins, and CD45 and CD34 negative cell surface proteins. In EPCs, CD34 and CD133 were detected, which were highly expressed in EPCs, but not expressed after EPCs differentiated into mature vascular endothelial cells [23]. In the meantime, CD31, which is not expressed in EPCs but highly expressed in mature endothelial cells, and CD11b, which is expressed in monocytes but not in EPCs, were also detected [24][25][26][27]. It was shown that EPCs expressed CD133 and CD34 with positive surface protein pro les, and CD11b and CD31 had negative surface protein pro les. These results con rmed the phenotype of ASCs and EPC, suggesting that the above-mentioned markers could be used for the identi cation of ASCs and EPCs.
In this study, a porcine acellular dermal ECM was prepared according to the protocol proposed by Matthew T. Wolf, as an injectable thermosensitive gel scaffold material for repairing skin defects. It was found that the dermal ECM can be used as a stem cell induction medium without cytotoxicity, in which cells could survive continuously for more than 14 days and exhibit proliferation and differentiation. To further determine the optimal ratio of ASCs/EPCs co-culture, 7 groups of different ratios of cell co-culture experiments in vitro were designed and grouped according to the ratio of EPCs in cells from less to more. The purpose is to study the effect of different EPCs ratios on wound healing to determine the best coculture ratio. It was found that when the ratio of ASCs/EPCs was 1:1, the expression of CK19 and vimentin was signi cantly higher than that of the other groups. Moreover, the expression of CK19 and vimentin in the ECM medium group was also dramatically higher than that of the L-DMEM group, indicating that the co-culture of ASCs/EPCs induced by ECM at a ratio of 1:1 is more conducive to the differentiation of stem cells into epidermal-like cells. After determining the optimal ratio, RT-PCR analyzed the expression of ASCs skin marker genes CK5, CK19, Col I, Col III, and endothelial marker genes VEGF, vWF, which showed that the ASCs/EPCs co-culture system could increase the expression of skin-related and angiogenesis-related genes. In addition, it was found in Matrigel tubule formation assay that ASCs/EPCs co-culture could improve the angiogenesis ability of cell lines.
On the basis of in vitro experiments, in vivo experiments on repairing rat skin defects were further carried out. The animal experiment was designed into 4 groups: blank control group, ECM + E group, ECM + A group, ECM + A + E group. The healing area was measured and analyzed at 7 and 14 days after the operation. The results clearly showed that the repair speed of the skin defect area in the ECM + A + E group was signi cantly faster than the other groups. In terms of healing rate, the ECM + A group was higher than the Blank group and the ECM + E group, but lower than the ECM + A + E group. In addition, color and shape of the newly formed skin in ECM + A + E group was signi cantly closer to normal skin, while there were obvious scar tissue formation and contraction in other groups, especially the Blank group and ECM + E group. HE staining showed that the angiogenesis density in the ECM + A + E group was higher than that in the other three groups, and a small amount of skin appendage formation was also detected, which was rarely seen in other groups. The thickness of the new skin tissue in the ECM + A + E group was greater than that of the other groups, and the scar thickness was also signi cantly less than that of the other groups. Masson staining con rmed that the production of Col I and Col III in the ECM + A + E group was greater than that in the other groups. Furthermore, it was con rmed in CD31 immuno uorescence analysis that the number and length of new blood vessels in the defect area of the ECM + A + E group were greater than those of the other three groups. These results demonstrated that ECM + A + E group achieved more effective vascularized skin regeneration. The possible reason for this phenomenon is that the vascularization of the regenerated skin in the ECM + A group and ECM + E group is relatively small, and it only depends on the ingrowth of host blood vessels. However, the distance between the blood vessels of the host tissue and the center of the skin defect is far and insu cient to achieve skin regeneration, especially for large-scale skin defects. Therefore, nutrients, metabolites and other molecules are unable to be transported to the central area of the defect, which severely hinders skin regeneration [28]. In the ASCs/EPCs group, EPCs directly increased vascular invasion and promoted the differentiation of ASCs into dermal direction. ASCs could promote the recruitment of EPCs and enhance the ability of EPCs to form blood vessels. This interaction accelerated the formation of vascularized skin, making nutrients, cytokines and other molecular factors involved in the skin healing process easier to access to the wound site. The results also showed that the implanted EPCs/ASCs co-culture system had a synergistic effect in comparison with the individual ASCs. The co-culture group signi cantly promoted skin tissue formation, which may be attributed to increased vascularization, leading to better recruitment of skin progenitor cells and other cytokines involved in skin healing. Vascularization is considered a necessary condition for skin formation, and insu cient vascularization leads to impaired skin formation or delayed skin healing [29]. Previous studies have shown that ASCs could produce soluble cytokines such as vascular endothelial growth factor (VEGF) and promote endothelial cell migration through autocrine or paracrine action, which contributes to cell rearrangement and tubular network formation [30].
VEGF is recognized as a main promoter of endothelial cell migration and tubular network formation [31]. The activation of VEGF can induce phosphorylation, participate in signal transduction, promote endothelial cell migration, and promote angiogenesis. In this process, the PI3K-Akt and MAPK pathways are the most important signal pathways for cytokine-cytokine receptor interaction, which are essential for damage repair [32][33][34]. Moreover, it was detected that the expression of VEGF and its receptors in the ASCs/EPCs co-culture group was up-regulated. This may be due to the production of VEGF and other cytokines and activation of the corresponding signal pathways by ASCs in the co-culture group, which facilitated the migration of vascular endothelial cells and the formation of tubular structures by through autocrine and paracrine, thereby increasing the formation of blood vessels. However, the molecular mechanism of this synergistic signaling pathway in the formation of vascularized skin in the ASCs/EPCs co-culture system is still unclear, which is also the focus of the next step of our study group

Conclusion
In summary, this study proved that the ASCs/EPCs co-culture cell system could synergistically promote the formation of vascularized skin, thereby realizing the repair of large-scale skin defects, which is unable to be achieved by other single seed cells. In addition, the co-culture cell system has the advantages of easy availability and low cost compared with stem cell angiogenesis-related gene transfection and other pro-vascularization methods, so it has a good application perspective. Based on the above results, we believe that it is feasible to use ASCs/EPCs co-culture system to replace traditional single-seed cells as a potential method of vascularized skin regeneration.
ASCs : Adipose-derived mesenchymal stem cells.  The red arrows show the newly-formed blood vessels in each group, and the number of new blood vessels in the ECM+A+E group is signi cantly more than that in the other groups. B: Quantitative analysis of regenerated skin scar thickness, *p <005, ECM+A+E group compared with all other groups.