Adipose Mesenchymal Stem Cell Derived Exosomes Promote Keratinocytes and Fibroblasts Embedded in Collagen/Platelet‐Rich Plasma Scaffold and Accelerate Wound Healing

Engineered skin substitutes derived from human skin significantly reduce inflammatory reactions mediated by foreign/artificial materials and are consequently easier to use for clinical application. Type I collagen is a main component of the extracellular matrix during wound healing and has excellent biocompatibility, and platelet‐rich plasma can be used as the initiator of the healing cascade. Adipose mesenchymal stem cell derived exosomes are crucial for tissue repair and play key roles in enhancing cell regeneration, promoting angiogenesis, regulating inflammation, and remodeling extracellular matrix. Herein, Type I collagen and platelet‐rich plasma, which provide natural supports for keratinocyte and fibroblast adhesion, migration, and proliferation, are mixed to form a stable 3D scaffold. Adipose mesenchymal stem cell derived exosomes are added to the scaffold to improve the performance of the engineered skin. The physicochemical properties of this cellular scaffold are analyzed, and the repair effect is evaluated in a full‐thickness skin defect mouse model. The cellular scaffold reduces the level of inflammation and promotes cell proliferation and angiogenesis to accelerate wound healing. Proteomic analysis shows that exosomes exhibit excellent anti‐inflammatory and proangiogenic effects in collagen/platelet‐rich plasma scaffolds. The proposed method provides a new therapeutic strategy and theoretical basis for tissue regeneration and wound repair.


Introduction
The development of tissue-engineered skin has required the testing of various new materials; however, almost all foreign tissues have a risk of infection.The degradation products are often engulfed by macrophages and cause inflammatory reactions that can impair wound repair cells, such as keratinocytes and fibroblasts, thus hindering the wound-healing process.Collagen (COL) is the main organic component that maintains the biological properties [1] and structural integrity of the natural extracellular matrix (ECM), regulates cell behavior, and fills defect sites during wound repair; [2] additionally, COL is the most widely used basic material in bioscaffolds. [3,4]7] Nevertheless, obtaining bioscaffold materials with sufficient regenerative potential remains challenging for clinicians. [8]nability to revascularize is another limitation of the current skin substitutes, and cells in the substitute die and slough away from the scaffolds with limited blood supply.Platelet-rich plasma (PRP) contains various growth factors, including plateletderived growth factor (PDGF), transforming growth factor (TGF), insulin-like growth factor, epidermal growth factor, and vascular endothelial growth factor (VEGF).These growth factors directly participate in the initiation process of tissue repair and reconstruction and promote the formation of blood vessels; furthermore, they can accelerate cell proliferation and division, accelerate cell growth, and shorten wound healing time. [9,10]In this study, COL-I and PRP were mixed, and biological scaffold materials were obtained by adding Thrombin.The components involved in the scaffold were derived from humans, which would improve clinical translation.
We also added adipose mesenchymal stem cell derived exosomes (ADSC-exos) to the scaffold in the expectation that these would improve the performance of the scaffold.Exosomes, such as nanoscale vesicles, are rich in bioactive substances, including proteins, ribonucleic acids, enzymes, and lipids.Exosomes can be transmitted to other cells as signaling molecules that exert biological functions, and consequently, exosomes have been widely studied as a biological therapy and are expected to be used in the clinic in the future. [11]Stem-cell-derived exosomes retain the characteristics of stem cells, including promoting tissue repair and regeneration, and immune regulation.Compared with stem cells, exosomes are safer, lack tumorigenicity, and have a low embolism risk.[17] Therefore, we explored whether exosomes can further improve the performance of scaffolds in our study.
Wound re-epithelialization is a multifactor cooperative process dominated by epidermal cells in wound healing, and during this, keratinocytes not only migrate from the periwound area to the wound and proliferate to cover the wound but also secrete various cytokines and growth factors to regulate the entire healing process. [18]The interaction of fibroblasts with keratinocytes during skin wound healing is a highly coordinated process. [19,20]Fibroblasts produce ECM, glycoproteins, adhesion molecules, and various cytokines, and fibroblast migration, proliferation, and ECM production are critical for functional dermal regeneration. [21,22]Keratinocytes and fibroblasts are critical repair cells in the wound-healing process. [23,24]In this study, keratinocytes and fibroblasts were embedded into COL-I scaffold with PRP (COL/PRP) to construct a double-layer tissueengineered skin.The effects of COL/PRP scaffolds on cell adhesion and growth and wound healing were investigated; additionally, the underlying mechanism was explored.The aim of this study was to construct a new cellular-bioscaffold material to promote wound repair.
Herein, we used a biological scaffold comprising COL-I, PRP, and ADSC-exos to simulate the ECM environment; in addition, fibroblasts and keratinocytes were loaded into the scaffold as essential components in the natural wound repair process and can be absorbed and utilized in vivo, avoiding inflammation caused by the degradation of artificial materials.Exosomes can further promote the growth and migration of repair cells, thereby accelerating wound healing.The structural and physicochemical properties of this cellular scaffold were analyzed via morphology, elastic modulus, and adhesion properties.Keratinocytes and fibroblasts were loaded in the scaffold, and the exosomes secreted by ADSCs were also added; cell proliferation and migration in the scaffold were then assessed, and the specific mechanism of proliferation and migration was studied through proteomics.Finally, the COL/PRP scaffold loaded with repair cells and ADSC-exos was applied in a full-thickness skin defect mouse model to verify the wound healing effect.We expect this cell bioscaffold to provide a fast and sophisticated solution for wound healing.

Design and Characterization of the COL/PRP Scaffold
After a wound occurs, periwound plasma is exuded and platelets are immediately activated to release PDGF to promote thrombosis, forming a gel-like mass to cover the wound, which is the initial stage of the wound healing response.The extraction and application of PRP can be used to simulate the initiation of the healing cascade.Additionally, fibroblasts recruited by chemokines repair wounds through collagen deposition. [25,26][29] For simulation of the wound healing microenvironment, a bioscaffold was generated using COL-I and PRP, with fibroblasts and keratinocytes as added layers.The schematic diagram of the synthesis is shown in Figure 1A.PRP and COL-I were mixed at different mass ratios, and 10 μL mL −1 Thrombin was added to form the main network of the scaffold.The material properties of the scaffold were tested.After the optimal mass ratio of PRP was selected based on 3D pore size, elastic modulus, and tensile strength, fibroblasts, and exosomes were added, after which keratinocytes were seeded to cover the scaffold surface.The effect of the COL/PRP scaffold on wound healing was then evaluated using four groups: a pure PRP group and PRP/COL-I groups with mass ratios of 16:1, 8:1, and 4:1.Scaffolds with a highly porous structure are crucial in realizing cell migration and crawling, interactions between different cell layers, and nutrient flow. [30]First, the microstructures of COL/PRP scaffold with different mass ratios were characterized.COL/PRP scaffolds with different mass ratios showed uniform 3D interconnected porous structures (Figure 1B), with pore sizes ranging from 20 to 120 μm, and with the decrease in the mass ratio of PRP, the pore diameter gradually decreased.For fibroblasts with a diameter of 17-20 μm, this pore size can provide sufficient area and space for cell growth. [31,32]This study confirmed that COL/PRP can form a 3D stereoscopic bioscaffold structure, providing environmental support for cell adhesion, migration, and proliferation.The elastic modulus of the COL/PRP scaffold, i.e., the strongest resistance to deformation, was largest at a mass ratio of 8:1, reaching 463 ± 0.252 kPa (Figure 1C).The tissue adhesive nature of the scaffold enabled this to maintain binding to the wound, reducing the risk of falling off during use. [33]Lap-shear tests were used to evaluate the tissue adhesion properties of the COL/PRP scaffold at different mass ratios.The adhesion strength in each group was ≈5 kPa (Figure 1D) with no significant difference between the groups, confirming that the COL/PRP scaffold met the wound surface adhesion requirements.The tensile strength of the 8:1 group was highest, reaching 220.6 ± 3.725 kPa (Figure 1E).The hematoxylin-eosin (H&E) staining also indicated a uniform pore structure, and with the decrease in the mass ratio of PRP, the structure became denser, and the pore diameter gradually decreased (Figure 1F).
Furthermore, to evaluate whether the COL/PRP scaffolds had favorable biocompatibility, a live/dead assay was performed; the scaffolds in each group were cocultured with fibroblasts (Figure 1G) and keratinocytes (Figure 2A) for 3 days.The live/dead cell staining of fibroblasts and keratinocytes identified only a few dead cells in each experimental group, which was similar to the result for the normal control group, and the quantitative analysis of live cells was not significantly different (Figures 1H  and 2D), confirming that the scaffold has favorable cell compatibility.Ki67 is the main marker used to assess cell proliferation. [34]he COL/PRP scaffolds with different mass ratios were cocultured with fibroblasts and keratinocytes and stained with Ki67 (Figure 2B, C).Immunofluorescence results indicated that the 8:1 mass ratio group had the highest abundance of fibroblasts and keratinocytes expressing Ki67 (Figure 2E,F).These data confirmed that the COL/PRP scaffolds with a mass ratio of 8:1 had better performance and a greater cell proliferation-promoting effect than those in the COL/PRP scaffolds in the other groups.Therefore, the mass ratio of 8:1 was used for subsequent experiments.Following the design process (Figure 1A), PRP and COL-I were mixed at a mass ratio of 8:1, and 10 5 cm −2 fibroblasts were added to a scaffold with a 1-mm thickness.After mixing, 10 μL mL −1 Thrombin was added to solidify the scaffold.Af-ter the cells were grown for 2 days, 10 5 cm −2 keratinocytes were seeded on the surface.As seen in Figure 2G, the keratinocytes were evenly distributed on the surface of the scaffold, with a satisfactory shape and cobblestone-like appearance.Additionally, staining for the keratinocyte cell-specific marker K14 and the fibroblast-specific marker -SMA in the scaffolds revealed a uniform hierarchical distribution of cells (Figure 2H).The 3D-reconstructed image obtained after scanning along the z-axis with a confocal laser microscope (Figure 2I) showed that PKH26labeled keratinocytes and PKH67-labeled fibroblasts on the bottom were distributed in a uniform hierarchical pattern on the top and bottom, respectively.These results confirmed that the two cell types could grow effectively on the COL/PRP scaffold.

ADSC-Exos Induce Fibroblast Migration and Proliferation in Scaffolds
ADSC-exos play important roles in wound healing, are nanoscale in size, immune tolerant, and can transfer proteins, messenger RNAs (mRNAs), microRNAs (miRNAs), and other small molecules into recipient cells to activate regeneration-related signaling pathways, thus exerting biological effects. [35,36]To verify whether ADSC-exos can affect the growth of cells in the COL/PRP scaffold, thereby promoting the wound healing effect of the COL/PRP scaffold, we extracted human AD-SCs from adipose tissue and then cultured them in vitro.After the supernatant was collected, the ADSC-exos were obtained by differential centrifugation (Figure 3A); a typical cupshaped double-layered membrane structure was observed under SEM (Figure 3B), and high-sensitivity flow cytometry for nanoparticle analysis showed that the size of the ADSCexos was between 50 to 120 nm (Figure 3C).Western blotting showed that the exosome surface markers CD9, CD63, and TSG101 were highly expressed (Figure 3D), indicating that ADSC-exos were successfully extracted.The ADSC-exos and fibroblasts were then simultaneously added to verify the proliferation and migration ability of the cells.
First, COL/PRP scaffolds with and without ADSC-exos were examined by SEM.We found that cells could grow into the scaffold and effectively fill the scaffold (Figure 3E, yellow arrows).The SEM results for the scaffolds with ADSC-exos showed that the exosomes were effectively encapsulated in the COL/PRP scaffold (Figure 3F, yellow box).Laser confocal microscopy showed that PKH26-labeled exosomes (red fluorescence) were evenly distributed around fibroblasts (green fluorescence), indicating that the cell/exosome-scaffold model was successfully established (Figure 3G).Next, the ability of ADSC-exos to promote the migration of fibroblasts in vitro was verified.The fibroblasts were labeled with calcein-AM and incubated with the scaffold for 72 h to observe the migration distance of the two groups of cells on the scaffold.Compared with cells in scaffolds without ADSCexos, the cells in scaffolds with ADSC-exos migrated farther at each observation time point (Figure 3H,J).Immunofluorescence staining for the proliferation-related marker Ki67 was performed for the scaffolds and showed that those with ADSC-exos had the most fibroblasts expressing Ki67 (Figure 3I,K).These results indicated that exosomes enhanced the migration and proliferation of fibroblasts in the COL/PRP scaffolds.

Growth of Keratinocytes in COL/PRP Scaffolds
Keratinocytes were seeded on the surface of fibroblasts/COL/PRP scaffolds with or without ADSC-exos, and surface coverage, proliferation, and migration were analyzed (Figure 4).The fibroblasts were first embedded on the COL/PRP scaffolds for 48 h, and then keratinocytes were uniformly seeded on these at 10 5 cm −2 , after which the coverage of keratinocytes was observed at different time points.Keratinocyte coverage on the COL/PRP scaffolds with ADSC-exos was greater than that for the control group at each time point, forming a monolayer that completely covered the surface within 72 h (Figure 4A).Ki67 immunofluorescence staining showed that the rate of proliferation of keratinocytes was higher in the group with ADSC-exos (Figure 4B,D), suggesting that the high surface coverage by keratinocytes is closely related to their higher proliferation rate. [32,37]The cell scratch assay was used to simulate the crawling process of epithelial cells at the edge of the skin. [38]DSC-exos significantly accelerated keratinocyte migration.The migration distance of keratinocytes in the group without ADSC-exos was shorter than that in the group with ADSCexos (Figure 4C,E).[41] The addition of exosomes further promoted the proliferation and migration of keratinocytes.Therefore, we speculated that the excellent keratinocyte coverage effect in the group with ADSC-exos occurred because exosomes and fibroblasts played a synergistic role that accelerated the rapid proliferation of keratinocytes (Figure 4F).

Proteomic Analysis of the Effects of ADSC-Exos on Scaffold
Evaluation of the six samples of COL/PRP scaffolds with or without exosomes (n = 3 per group) afforded the number of 6594 differentially expressed proteins (DEPs) in the two groups.The results were normally distributed, indicating that the samples are reliable and the results are stable (Figure 5A).Proteomic features were investigated using liquid chromatography with tandem mass spectrometry (LC-MS/MS).Differential expression analysis was performed for 6594 proteins detected in the samples.Pie charts were used to visualize different proportions of DEPs in the cytoplasm and extracellular space (21.76% and 9.41%, respectively), indicating that the DEPs may be related to exosomes (Figure 5B).As shown in the quantitative volcano plot (Figure 5C), 139 DEPs were identified, of which 81 proteins were upregulated (red dots) and 58 proteins were downregulated (blue dots); histograms are shown in Figure 5D.The protein-protein interaction network showed that interleukin (IL)−1, intercellular adhesion molecule 1 (ICAM1), thymidine phosphorylase (TYMP), lysyl oxidase-like 4 (LOXL4), and TNIP1 were correlated and interacted with each other in the exosome group (Figure 5E).A cluster analysis heatmap revealed DEP enrichment in cells.Several of these DEPs were closely related to anti-inflammatory effects in the exosome group; for example, IL-1 and ICAM1, and several cytokines, such as VEGF and angiogenesis-related TYMP, were closely related to angiogenesis and vascular integrity in the exosome group (Figure 5G).The increase in LOXL4 suggests that exosomes may be involved in the regulation of collagen deposition of ECM.Gene Ontology (GO) enrichment analysis and Kyoto Encyclopedia of Genomes (KEGG) analysis were used to assess all identified DEPs.GO enrichment analysis revealed that DEPs were associated with extracellular structure and function (Figure 5F).KEGG analysis indicated that several signaling pathways associated with inflammatory responses, such as the NF-kappa B (NK-kB), Notch, and TNF signaling pathways, and angiogenesis-related pathways, such as the VEGF signaling pathway, were significantly different, suggesting that ADSC-exos may exert physiological activities in these aspects (Figure 5H).The COL/PRP scaffold provides a basic microenvironment for cell migration and proliferation, and the addition of ADSC-exos exerts its biological activity.Inflammation-related molecules in the samples were detected using ELISA, and angiogenesis-related cytokines were detected using western blotting to verify the proteomics results.Compared with that of the non-exosome group, the expression of IL-1 and ICAM1 in the exo-some group was significantly decreased (Figure 5I), and the VEGF and TYMP protein levels were significantly increased (Figure 5J,K).These results confirmed that ADSCexos played an important biological role in reducing inflammation and promoting angiogenesis in COL/PRP scaffolds.

Effect of the Exosome/Cell-Loaded COL/PRP Scaffold on Wound Healing
Finally, the healing effect of the exosome/cell-loaded COL/PRP scaffold was verified using a mouse wound model.A round fullthickness skin defect model (1 cm diameter) was created on the back of mice; the experimental scheme is shown in Figure 6A.Mice were randomly divided into three groups: COL/PRP scaffolds without cells (the control group), COL/PRP scaffolds with fibroblasts and keratinocytes, and COL/PRP scaffolds with two cell types and exosomes.Figure 6B shows the wound images in each group at various time points, demonstrating the gradual decrease in the size of the wound area.After 7 days, the wound area in the experimental groups was significantly smaller than that in the control group.The group with exosomes showed the fastest wound healing rate; where the wound closed within 12 days, demonstrating that the addition of exosomes helped promote wound healing in mice.The group with additional cells also exhibited a healing effect compared with that of the control group, but this was not as significant as that in the group with exosomes (Figure 6D).H&E staining and Masson staining were used to histologically assess wound epithelialization and granulation processes on day 12. H&E staining results showed that the granulation tissue in the group with exosomes was in a favorable healing condition, and exhibited the shortest wound length and complete formation of epithelial tissue (Figure 6C, left).Collagen arrangement and distribution are important criteria for evaluating wound healing.Appropriate collagen deposition is beneficial to wound repair, but excessive and disordered deposition can cause scarring, which is not conducive to repair. [42,43]Masson staining showed collagen in wounds in different groups (Figure 6C, right).The COL fibers in the group with exosomes were neatly and regularly arranged, indicating that the quality of wound healing was better than that in the other groups.H&E and Masson's staining showed that the COL/PRP scaffolds with exosomes and repaired cells had the most satisfactory healing effect.
Immunofluorescence staining for Ki67 was performed with skin tissue.The group with exosomes showed the strongest and most significant staining for Ki67, indicating that the group with exosomes promoted the proliferation of wound cells and accelerated wound healing (Figure 6E,G).K14 is a specific marker of keratinocytes, [44] and the K14 fluorescence staining of skin tissue showed that the group with exosomes had increasingly complete epidermal formation (Figure 4F,H).These results indicate that the COL/PRP scaffolds with exosomes and cells played a role in promoting efficient cell proliferation, which is conducive to accelerating wound healing.

Anti-Inflammatory and Proangiogenic Effects of the Exosome/Cell-Loaded COL/PRP Scaffold on Wounds
The inflammatory environment of wounds can impede wound healing, [19,45] and the level of TNF- and IL-6 can be used to characterize the inflammatory response of wounds. [46]On the 3rd day after wounding, the expression levels of TNF- and IL-6 in each group were detected using immunohistochemical staining.Compared with the control group, fewer inflammatory fac-tors were present in the group with exosomes than in the other groups (Figure 7A), and the expression level of TNF- and IL-6 was significantly lower than that in other groups, confirming that the COL/PRP scaffolds with exosomes and cells alleviated the inflammatory response in the wound, thereby improving the microenvironment for the wound repair process (Figure 7B,C).Immunofluorescence staining for macrophages (F4/80) and for M1 (CD80) and M2 (CD206) with paraffin-embedded mice skin wound specimens on indicating time were shown in Figure S1 (Supporting Information).Immunofluorescence analysis revealed that CD80 M1 macrophages were more abundant on day 3 in scaffold groups and scaffold with cell groups than in exosome/cell-loaded scaffold groups.For application of an exosome/cell-loaded scaffold, the average ratios of CD80/F4/80+ M1 macrophages were lower than other groups.Besides, M2 macrophages were prominently detected in exosome/cell-loaded scaffold groups on day 12. Blood vessels provide nutrients and oxygen support throughout the wound-healing process and are essential for wound repair. [47,48]Laser Doppler perfusion imaging was used to analyze blood perfusion in wounds at each time point.Blood perfusion in the wounds of mice in each group gradually increased (Figure 7D,E).At each observation time point, the group with exosomes had the strongest blood signal compared with that in the other two groups, indicating that the COL/PRP scaffolds with exosomes and cells effectively promoted the formation of microvessels during the wound healing process and that the degree of vascularization was higher.CD31 indicates new blood vessel growth in wounds, and -SMA indicates the presence of mature blood vessels. [49]On day 12, double SMA/CD31 fluorescence results indicated that the amount of blood vessels in the group with exosomes was higher than that in the other groups; additionally, the blood vessel density was the highest among the three groups, and the lumen was large, with a complete structure (Figure 7F,I).Western blotting for the angiogenesis-related factor VEGF confirmed the effect of additional exosomes on angiogenesis and maturation.VE-cadherin is an important adhesion molecule for endothelial cells and is crucial in maintaining vascular integrity and stability. [50,51]The expression of VE-cadherin protein in the group with exosomes was higher than that in other two groups (Figure 7G,H).These results suggest that COL/PRP scaffolds loaded with exosomes and cells can accelerate wound healing by improving the wound inflammatory microenvironment and promoting wound angiogenesis and maturation.

Discussion
In recent years, the field of skin tissue engineering has made several advancements to facilitate wound healing for skin regeneration and has considerable potential for improving the rate and quality of wound healing.However, many challenges remain for tissue-engineered skin in skin regeneration.Among these, the primary factors in determining the quality of skin tissue engineering are how to rapidly establish the blood supply and regulate the immune microenvironment.Vascularization is critical for the prolonged function and survival of tissueengineered skin substitutes.Rapid blood supply establishment can provide nutrients for cells in tissue-engineered skin, avoiding cell necrosis. [52]The immune microenvironment also plays  an important role in wound healing.The accumulation of M1 macrophages and the release of inflammatory cytokines induce an excessive inflammatory response that will deteriorate the wound microenvironment, impairing the establishment of a reliable connection between tissue-engineered skin and the wound base.
Herein, we developed a 3D mesh scaffold containing ADSCexos.The scaffold is formed by mixing COL-I and PRP.Ac-tivation of PRP releases a large amount of fibrin, fibronectin and fibronectin, and forms a fiber network with COL-I.The porous structure can provide support for cell adhesion and crawling, preventing cell loss. [25]Furthermore, the scaffold releases a large number of growth factors such as TGF-, PDGF, FGF, HGF, and VEGF, which, as a strong vascular growth factor, play a critical role in wound healing and vascularization.
ADSC-exos play an important role in promoting tissue repair, regeneration, and angiogenesis.In multifunctional wound dressings, continuous release of exosomes can improve ROS damage, induce angiogenesis, enhance proliferation, granulation tissue formation, and collagen accumulation in diabetic wounds. [53]Here, in vitro experiments confirmed the promoting effect of ADSC-exos on cell proliferation and migration.In vivo experiments showed that the scaffold based on ADSC-exos could increase the number of blood vessels during wound repair, promote the proliferation of wound repair cells, and accelerate the healing process, which confirmed the considerable advantages of ADSC-exos in vascularization and healing.Thus, ADSC-exos are critical for the successful development of tissue-engineered skin.
The important role of immune microenvironment in wound healing should also be considered.Yuan et al. [54] designed a hydrogel that packaged engineered small extracellular vesicles (sEVs) and applied them to diabetic wounds, showing that sustained release of sEVs induced macrophages to switch from M1 to M2 phenotype and thereby exert prohealing effects throughout the inflammation and proliferation stages of diabetic wound healing.In our study, the ADSC-exos loaded in the COL/PRP scaffold material can play a similar effect through sustained release.We found that ADSC-exos promoted the transformation of macrophages to M2 type and reduced the expression of IL-6 and TNF- in the wound.
In addition, fibroblasts and keratinocytes were added to the COL/PRP scaffold to construct a double-layer tissue-engineered skin.All components are essential in the repair process of wounds and can be directly used during wound healing.Compared with artificial materials, the COL/PRP scaffold greatly reduces the influence of foreign-body-induced inflammatory response on the local immune microenvironment of the wound, which is an advantage of the engineered skin constructed in this study.
The COL/PRP scaffold exhibited ideal healing-promoting effects both in vivo and in vitro.Proteomic analysis showed that exosomes exhibited excellent anti-inflammatory and proangiogenic effects in COL/PRP scaffolds.However, in the in vivo and in vitro experiments, we observed that ADSC-exos played an important role in promoting cell proliferation.We considered that the proteomic analysis derived from vitro experiments showed the most apparent changes in the levels of cell proteins, suggesting that ADSC-exos could promote the expression of anti-inflammatory and angiogenesis-related proteins in cells, although the proteins that play a role in promoting growth were not prominent, which could indicate that exos do not have a growth-promoting effect.However, in an in vivo environment, the mechanisms of wound healing are more complex and anti-inflammatory and promoting vascularization proteins could promote cell proliferation by improving the local microenvironment.

Conclusion
We have designed a tissue-engineered skin that contained essential components in the natural repair process of the wound and could be directly used in the wound healing process.The COL/PRP scaffold has a uniform 3D interconnected porous structure, with excellent elastic modulus and biological adhe-sion.Compared with most current tissue-engineered skin, our COL/PRP scaffolds loaded with keratinocytes, fibroblasts, and ADSC-exos had an advantageous role in promoting angiogenesis and improving the immune microenvironment.We verified the therapeutic effect of our tissue-engineered skin in promoting wound closure in mice.In addition, we elucidated the potential mechanism whereby tissue-engineered skin promotes cell proliferation and migration due to the role played by ADSC-exos in this system.The various components involved in our tissueengineered skin are derived from the human body itself and are easy to prepare.Thus, the tissue-engineered skin designed and evaluated in this study is easy to be transformed for clinical use and could have considerable commercial potential as a wound dressing.

Experimental Section
PRP Preparation: A total of 30 mL of blood was extracted from an adult volunteer after obtaining their informed consent.Blood samples were then centrifuged at 1500 rpm for 10 min and separated into three phases: platelet-poor plasma (top), PRP (middle), and erythrocytes (bottom).The top and middle layers were transferred to new tubes and centrifuged again at 3000 rpm for another 10 min.The supernatant plasma was discarded, and the remaining plasma was designated as PRP.Five samples of PRP were then stored at −80 °C.
Preparation of COL/PRP Scaffolds: COL-I and Thrombin were purchased from Macklin (Shanghai, China).COL-I was lysed in RPR at the indicated mass ratio: 62.5, 125, or 250 mg COL-I were each mixed with 1 mL PRP and 10 μL Thrombin and defined as the 16:1, 8:1, and 4:1 groups, respectively; 1 mL PRP and 10 μL Thrombin was defined as the PRP group.The mixed solutions were stirred uniformly.
Preparation of Exosome/Cell-Loaded COL/PRP Scaffolds: After 25 mg COL-I and 200 μL of PRP were mixed, and then 10 5 of fibroblasts were added into the mixture in wells of a 24-well plate; 1 μg μL −1 exosome and 10 μL mL −1 Thrombin were added, and the sample was stirred uniformly and incubated at 37 °C with 5% CO 2 .After 3 days of coculture, 10 5 of keratinocytes were seeded on the surface of scaffold.
Isolation, Culture, and Identification of ADSC-Exos: Human adipose tissues were acquired from patients undergoing selective liposuction.Adipose tissues were fragmented and digested with 1 mg mL −1 collagenase type I (Gibco, Thermo Fisher Scientific, Inc.) for 60 min and then filtered using a 100 μm mesh.After centrifugation at 200 × g for 5 min, ADSC pellets were obtained and resuspended in special culture media (Gibco; Thermo Fisher Scientific, Inc.).Samples were cultured at 37 °C, and the culture medium was changed every 3 days.Exosomes were obtained from ADSCs supernatants.Transmission electron microscopy, nanoparticle tracking analysis, and western blotting were used to analyze the exosomes.A BCA protein assay kit (Beyotime, Shanghai, China) was used to measure the exosomal protein concentration (1 μg μL −1 ), and then the samples were stored at −80 °C.
The cytotoxicity of the COL/PRP scaffold was evaluated using a Viability/Cytotoxicity Kit (Beyotime, Shanghai, China).After coculture for 72 h, fluorescence stains of the live and dead cells were observed according to manufacturer's instructions, and an FSX100 microscope and ImageJ software were used for analysis.Ki67 immunofluorescent staining was employed to evaluate the proliferation of keratinocytes and fibroblasts.Cells in the scaffold were fixed with 4% paraformaldehyde, permeabilized with 0.1% Triton X-100, stained with anti-Ki67 antibodies (1:200, Abcam, Cambridge, UK) at 4 °C for one night, then incubated with AF488-conjugated secondary antibody (1:100, Abcam, Cambridge, UK) for 1 h at room temperature.DAPI (Beyotime, Shanghai, China) was used to detect the nuclei in immunofluorescence staining.The number of Ki67-positive cells was counted using ImageJ software.
The migration of fibroblasts was recorded using an FSX100 microscope.Cells were labeled with a Viability/Cytotoxicity Kit (Beyotime, Shanghai, China).The migration distance of fibroblasts at every indication time was observed and analyzed as a ratio of migrated area to original area.A 10 μL pipette tip was used to induce a scratch in the keratinocyte layer that on the surface of scaffold at a density of 1 × 10 6 cells per well.Cell closures were recorded by microscope (Olympus, Tokyo, Japan) at each time point.The migration rate was determined as the ratio of migrated area to original area.The fusion of keratinocytes was recorded by a microscope with the initial cell density of 0.5 × 10 5 , and photos were taken at every time point in each group.
Morphology Analysis of the COL/PRP-Series Scaffolds: The COL/PRPseries scaffolds were freeze-dried to evaluate their microstructural morphology using a scanning electron microscope (Zeiss EVO-MA 10 model EDS 250; Zeiss, Jena, Germany).
Mechanical Tests: The mechanical properties of COL/PRP-series scaffolds were tested by uniaxial tensile testing with the Bose Endura TEC ELF 3200 system (Bose, USA).All the samples were trimmed to a size of 1 × 2 × 10 mm 3 before being gripped with clamps.Uniaxial tensile force was applied at a rate of 2 mm min −1 with an initial force of 0.05 N until rupture, and peak stress was defined as UTS.The load and elongation of the sample were recorded by the biomechanical analyzer.
Tissue Adhesion Capacity of the COL/PRP-Series Scaffolds: The adhesive strength of the scaffolds was evaluated as follows.Briefly, fresh porcine skin tissue (purchased from supermarket) was sectioned into a 6 × 30 mm rectangle and immersed in PBS before use.Subsequently, 100 μL of COL/PRP-series scaffold prepolymer solutions was placed onto the surface of the porcine skin.Another piece of skin with an adhesive area of 8 × 10 mm was applied to the scaffold solution.Then, the porcine skin was kept at 25 °C for 30 min.A lap-shear test on Materials Test System (MTS Criterion 43, MTS Criterion) equipped with a 50 N load cell at a rate of 2 mm min −1 was used to evaluate the adhesion properties.All tests were performed five times.
Proteomics Analysis: Label-free LC-MS/MS was used for proteomics analysis of the cell-loaded COL/PRP scaffold and the exosome/cell-loaded COL/PRP scaffold (n = 3 in each group).After protein extraction and trypsin digestion, liquid chromatography mobile phase A was used to dissolve the peptides, and NanoElute ultra-performance liquid chromatography system was used to separate the peptides, which were then injected into the capillary ion source for ionization analyses via tims-TOF Pro mass spectrometry.The Maxquant search engine (v.1.5.2.8) was used for processing the MS/MS data.KEGG online service tools were used to annotate the DEPs, which were then matched with the database through the KEGG Mapper.GO annotation proteome was screened with the UniProt-GOA database (http://www.ebi.ac.Uk/GOA/).Hierarchical clustering based on DEP functional classification was visualized by a heat map.A corrected p-value of <0.05 was considered significant.
ELISA: Proteins extracted from the two groups were collected and examined with ELISA kits for IL-1 and ICAM1 following the manufacturer's guidance.The absorbance at 450 nm was measured using a microplate reader, and the concentration was calculated according to the standard curve.
In Vivo Wound Healing in Mouse Models: Male BALB/c mice aged 6 weeks were purchased from The Fourth Military Medical University Laboratory Animals Center.A single round full-thickness skin wound ≈10 mm in diameter was created after animals were anesthetized.Scaffolds were administered onto the wound surface.All mice were randomly assigned into three groups (n = 10 per group): 1) control group (treated with the COL/PRP scaffold); 2) model group (treated with the cell-loaded COL/PRP scaffold); 3) exosomes model group (treated with the exosome/cell-loaded COL/PRP scaffold).The wounded area was measured and photographed on days 0, 3, 7, 10, and 12 after treatment.Subsequently, mice were euthanized, and the wound tissues were harvested at the indicated times for analysis.
Blood Perfusion Imaging: On day 12 after the operation, a laser Doppler perfusion imaging analyzer was used to evaluate the blood perfusion of the wound.Mice were anesthetized with isoflurane.The average blood perfusion in the wound area of the mice (ROI-1) and the wound edge skin blood perfusion (ROI-2) were detected.ROI-1/ROI-2 was the average perfusion ratio.
Western Blot Analysis: Sample proteins were separated using sodium dodecyl sulfate-polyacrylamide gel electrophoresis and then transferred to polyvinylidene fluoride membranes.After the membranes were blocked with 5% nonfat dry milk for 2 h, they were incubated with primary anti-VEGF, anti-VE-cadherin, or anti-TYMP antibodies diluted 1:1000 (Abcam, Cambridge, UK) at 4 °C overnight; anti--actin (1:3000; Abcam, Cambridge, UK) was used a control.Membranes were incubated with horseradish peroxidase-conjugated secondary anti-rabbit IgG antibodies (1:2000, Boster, Wuhan, China) for 1 h at 37 °C.An ECL detection system (Millipore, USA) was used to measure the intensity of protein expression via Fluor Chem system (Alpha Innotech).Results were analyzed using Im-ageJ software.
Statistical Analysis: All data were analyzed using SPSS 17.0 software and were presented as the mean ± standard deviation (SD).A Student's t-test was used for comparisons between two groups, and analysis of variance was used for multigroup comparisons.In all cases, p < 0.05 indicated a significant difference.
Ethics Approval and Consent to Participate: Human normal subcutaneous adipose tissues were acquired from patients undergoing selective liposuction at Xijing Hospital (Xi'an, China).The patients provided informed consent before the study.The human and animal studies in this work were approved by the Ethics Committee of Xijing Hospital affiliated with Fourth Military Medical University.. experiments and analyzed study data; K.S. and K.W. provided study materials, reagents, and replicated the results.C.T. provided laboratory samples and animals.All authors read and approved the final manuscript.

Figure 1 .
Figure 1.Characterization of the COL/PRP-series scaffolds.A) Schematic of the preparation of the exosome/cell-loaded COL/PRP scaffold.B) Morphology of all COL/PRP scaffolds; scale bar: 200 μm.C) The elastic modulus was strongest in the 8:1 group.D) Tissue adhesion properties for all scaffolds.E) Stress-strain curves of all scaffolds at axial direction showed that the 8:1 group had the highest ultimate tensile strength (UTS).F) Hematoxylin and eosin staining of all scaffolds; scale bar: 1000 μm.G) Representative images of live/dead staining of fibroblasts in each group; scale bar: 1250 μm.H) Quantitative statistical data of fibroblast viability.Data are shown as the mean ± SD (n = 6 per group, *p < 0.05, **p < 0.01, ***p < 0.001).

Figure 2 .
Figure 2. Characterization of cells in COL/PRP-series scaffolds.A) Representative images of live/dead staining of keratinocytes in each group; scale bar: 650 μm.B) Effect of the COL/PRP-series scaffolds on fibroblast proliferation were determined using Ki67 immunofluorescence assays; scale bar: 650 μm.C) Effect of the COL/PRP-series scaffolds on keratinocyte proliferation was determined using Ki67 immunofluorescence; scale bar: 650 μm.D) Quantitative data of keratinocyte viability.E) Relative expression of Ki67 of fibroblasts.F) Statistical data of the relative expression of Ki67 of keratinocytes.G) Image of keratinocytes seeded on the surface of the COL/PRP scaffold; scale bar: 650 μm.H) Immunofluorescence assays of keratinocytes and fibroblasts in the COL/PRP scaffold.K14 was used to characterize keratinocytes and SMA was used to characterize fibroblasts; scale bar: 1000 μm.I) Confocal 3D image of keratinocytes and fibroblasts in the COL/PRP scaffold.PKH26 was used to characterize keratinocytes (green) and PKH67 was used to characterize fibroblasts (red).Data are shown as the mean ± SD (n = 6 per group, *p < 0.05, **p < 0.01, ***p < 0.001).

Figure 4 .
Figure 4. Effects of ADSC-exos on keratinocytes in the COL/PRP scaffold.A) Representative images of keratinocytes on the COL/PRP scaffold of each group; scale bar: 650 μm.B) Effect of the ADSC-exos on keratinocyte proliferation was determined using Ki67 immunofluorescence assays; scale bar: 650 μm.C) Effect of ADSC-exos on keratinocytes migration at every indicating time; scale bar: 650 μm.D) Relative expression of Ki67 of keratinocytes.E) Relative migration of keratinocytes in each group.F) Schematic diagram indicating the proposed mechanisms underlying the effect of ADSC-exos on cells in the COL/PRP scaffold.(n = 6 per group, *p < 0.05, **p < 0.01, ***p < 0.001).

Figure 5 .
Figure 5. Proteomics analyses of the biological importance of the cell-loaded COL/PRP scaffold with or without exosomes (n = 3 per group).A) The number of differentially expressed proteins (DEPs) and normal distribution phenotype of samples from the two groups.B) Subcellular localization classification showed the DEPs were mainly concentrated in the cytoplasm and extracellular space C,D) A volcano plot showed the DEPs between two groups (red dots: upregulated; blue dots: downregulated), which were quantified with a histogram.E) Protein-protein interaction results in the exosome/cell-loaded COL/PRP scaffold.F) GO annotation of DEPs showed their different biological roles of exosomes in the cell-loaded COL/PRP scaffold.G) Heatmap representation of the cluster analysis of DEPs in the exosome/cell-loaded COL/PRP scaffold.H) KEGG pathway enrichment analysis suggested that DEPs were annotated in the cell inflammation and angiogenesis, such as IL-1, and ICAM1 and VEGF signaling pathways.I) The IL-1 content measured via ELISA was significantly lower in the exosome/cell-loaded COL/PRP scaffold (n = 6 per group).J) The ICAM1 content measured via ELISA was significantly lower in the exosome/cell-loaded COL/PRP scaffold (n = 6 per group).K) Change in VEGF and TYMP expression (n = 6 per group).L) Relative expression of VEGF and TYMP.Data are shown as the mean ± SD (*p < 0.05, **p < 0.01, ***p < 0.001).

Figure 6 .
Figure 6.Effects of the exosome/cell-loaded COL/PRP scaffold on wound healing in a mouse wound model.A) Schematic diagram of the experimental processes.B) Digital photographs of wound areas dressed with the COL/PRP scaffold, cell-loaded COL/PRP scaffold, and exosome/cell-loaded COL/PRP scaffold on days 0, 3, 7, 10, and 12. C) Representative hematoxylin and eosin and Masson staining of the wound tissue in each group on day 12; scale bar: 1 mm.D) Analysis of the wound closure rate in each group at each time point.E) Representative photographs of Ki67 immunofluorescence staining of skin wound tissue on day 7 after injury; scale bar: 1 mm.F) Representative photographs of K14 immunofluorescence staining of skin wound tissue on day 7 after injury; scale bar: 1 mm.G) Statistical data of relative Ki67 expression of the different groups.H) Statistical data of relative K14 expression of the different groups.Data are shown as the mean ± SD (n = 5 per group, *p < 0.05, **p < 0.01, ***p < 0.001).

Figure 7 .
Figure 7. Impact of the exosome/cell-loaded COL/PRP scaffold on wound healing in a mouse wound model.A) Representative photographs of immunohistochemistry staining of TNF and IL-6 of skin wound tissue on day 3 after injury; scale bar: 50 μm.B,C) Relative expression of TNF (B) and IL-6 (C).D) Typical images of laser Doppler perfusion imaging and quantitative analysis results I of each group on days 0, 3, 7, and 12. E,F) Representative images of -SMA and CD31 immunofluorescence staining of skin wound tissue on day 7 after injury; scale bar: 100 μm upper; scale bar: 100 μm under.G) Change in VEGF and VE-cadherin expression.H) Relative expression of VEGF and Ve-cadherin.I) Relative expression of -SMA and CD31.(n = 5 per group, *p < 0.05, **p < 0.01, ***p < 0.001).