Development of a stem cell spheroid‐laden patch with high retention at skin wound site

Abstract Mesenchymal stem cells such as human adipose tissue‐derived stem cells (hADSCs) have been used as a representative therapeutic agent for tissue regeneration because of their high proliferation and paracrine factor‐secreting abilities. However, certain points regarding conventional ADSC delivery systems, such as low cell density, secreted cytokine levels, and cell viability, still need to be addressed for treating severe wounds. In this study, we developed a three‐dimensional (3D) cavity‐structured stem cell‐laden system for overdense delivery of cells into severe wound sites. Our system includes a hydrophobic surface and cavities that can enhance the efficiency of cell delivery to the wound site. In particular, the cavities in the system facilitate hADSC spheroid formation, increasing therapeutic growth factor expression compared with 2D cultured cells. Our hADSC spheroid‐loaded patch exhibited remarkably improved cell localization at the wound site and dramatic therapeutic efficacy compared to the conventional cell injection method. Taken together, the hADSC spheroid delivery system focused on cell delivery, and stem cell homing effect at the wound site showed a significantly enhanced wound healing effect. By overcoming the limitations of conventional cell delivery methods, our overdense cell delivery system can contribute to biomedical and clinical applications.

suitable candidate for therapeutic applications because of their high proliferative potential, exosome secretion, autocrine and paracrine factors, and adipose tissue accessibility as a cell source. 4,5 Despite their strong potential as a donor cell source for stem cell-based therapy, ADSCs implanted into lesions such as ulcers, ischemia, and wound areas have shown limited therapeutic efficacy due to their low survival rate at the wound site. 6,7 The major reason for this low survival rate is due to the harsh microenvironment at the wound site, caused by hypoxia, lack of nutrients, proinflammation, and loss of cell adhesion. To improve the engraftment of ADSCs in the wound area, several strategies have been developed. These strategies include the genetic modification of stem cells, 8 delivery of stem cells with growth factors, 9 use of tissue engineering scaffolds, 10 and delivery of threedimensional (3D) stem cell cluster/aggregate. 11 Although these strategies showed improvements at certain degrees, more efficient approaches are still required for the treatment of severe wounds.
Conventional treatments are focused on the management of skin wounds, which include removing necrotic tissue and reducing infection by replacing wound dressings and applying different kinds of ointments. 12 However, for hard-to-heal wounds, including chronic wounds, major burn injuries (>20% of total body surface area burned), and full-thickness severe skin wounds, comprehensive treatment is required rather than management. 13 Recent advances in severe skin wound regeneration are mostly based on biomedical engineering approaches, such as stem cell delivery, cell-free scaffolds, cellular skin substitutes, and 3D bioprinting. [14][15][16] In particular, the stem cell delivery method for wound healing treatment is a promising method because of its immune regulatory function and regenerative factor secretion. Although stem cells (including ADSCs) have great potential for the treatment of severe wounds, enhanced and efficient stem cell delivery methods are necessary for clinical treatment.
To overcome some of the limitations of previous studies, we developed a 3D cavity-structured cell-laden system for overdense delivery of cells into severe wound sites. The 3D structure of the patch was designed to simplify the formulation of ADSC spheroids.
We fabricated a hydrophobic surface patch and many cavities to enhance the number of loaded cells into the patch, which complemented the weakness of the conventional patch system and low delivery density of cells. The ADSC spheroids in the patch had an increased expression of therapeutic growth factors compared to 2D cultured cells in vitro. Previous reports have shown that the increasing diameter of 3D cell aggregate can lead the upregulation of hypoxia inducible factor 1 alpha (HIF-1α) and HIF-1α-related angiogenic factors expression. [17][18][19] Based on the enhanced therapeutic gene expression from 3D cell aggregate, various attempted to treat intractable diseases with 3D cell aggregates have been developed. Recent report has shown that the angiogenic paracrine factor secretion from 3D cell aggregate, spheroid, can be optimized by controlling cell culture period and diameter range. 17 Moreover, combining the spheroids with new materials such as hydrogel to enhance the therapeutic efficacy of the cell has been reported. 18 However, conventional methods required time consuming and labor-intensive procedure for preparing the spheroids. Also, injecting spheroids with syringes can induce shear stress to the spheroids during the injection process. The stress leads low cell viability and structural damage that can decrease the therapeutic effect of spheroid. Therefore, to increase the therapeutic efficacy of spheroids, it is necessary to develop easy and fast spheroid transplantation method without structural damage. Here, we have suggested a patch with a specific 3D structure that can induce and include spheroid easily and quickly. Moreover, our patch was designed to deliver spheroids directly to the wound sites without cellular or structural damages. When the hADSCs were applied with the 3D-structured patch, the cells were localized in the wound site and had enhanced therapeutic efficacy compared to the conventional cell injection method. Particularly, we focused on the cellular efficacy of delivery and homing to the wound site, which illustrated the increased efficacy of our patch system. Our overdense cell delivery system represents an effective alternative to conventional cell injection methods and thus, could contribute to biomedical and clinical applications. Gibco BRL) and centrifuged (400g, 5 min) to remove any bubbles in the 3DP cavities. PBS in the cell plate was subsequently removed, and detached ADSCs were loaded into the 3DP cavities by centrifugation (300g, 5 min for 1.5 Â 10 6 cells per patch). After centrifugation, the plate was incubated for 24 h in a 5% (vol/vol) CO 2 incubator at 37 C.

| Cell viability assay
Live and dead cells were imaged after staining with fluorescein diacetate (FDA; Sigma) and ethidium bromide (EB; Sigma). Dead cells were stained red because of the nuclear permeability of EB. Viable cells, which can convert the nonfluorescent FDA into fluorescein, were stained green. To compare the cell viability of the 2D surface and 3DP, identically sized PDMS compared to the 3DP patch without a 3D structure (flat PDMS) was prepared. Next, the same number of cells applied to the CLP was loaded onto the flat PDMS. The cells were stained with FDA and EB following a 24-h incubation.

| Wound treatment
Eight-week-old male athymic mice (20-25 g body weight; Orient) were anesthetized by inhalation of 2.5% isoflurane in oxygen gas. The skin wound area was marked using a 1.8 Â 1.8 cm square shaped stamp on the back of each mouse. The epidermis, dermis, and stratum corneum in the marked areas were surgically removed. After skin incision and removal, four surgical sutures (6-0 silk suture; AILEE Co.) F I G U R E 1 A schematic diagram of 3DP fabrication, CLP preparation, and in vivo experiments. (a) A schematic representation of 3DP fabrication. (b) Scanning electron microscopy image of fabricated 3DP and its structural specification. (c) Sequential diagram for CLP preparation and in vivo experiments. 3DP was assembled to customize cell-loading mold. ADSCs were loaded onto patch by centrifugation. After incubation for 24 h, ADSCs formed spheroids in 3DP (CLP for in vivo experiment). The developed CLPs were subsequently applied to mouse skin wound model for 3 days after skin wound modeling. 3DP, 3D cavity-structured patch; ADSC, adipose tissue-derived mesenchymal stem cell; CLP, stem cell-laden 3D cavity-structured patch were made at the center of each side of the square-shaped wound to prevent skin contracture. 20 Immediately after skin wound modeling, the mice were randomly divided into four groups: (1)

| In vivo wound imaging and wound closure quantification
The wounds were photographed at each time point. Wound closure was quantified and calculated based on the pixel-to-centimeter ratio.
The wound size was measured using an imaging software (ImageJ).
The wound margin was defined as a grossly visible margin of epithelial tissue migration toward the center of the wound and over the granulation tissue bed. Wound closure was calculated as the percentage of the initial wound area [(wound area at time point)/(initial wound area) Â 100%]. 21

| Live imaging of transplanted cells
To evaluate hADSC engraftment and retention after treatment, hADSCs were labeled with VivoTrack 680 (PerkinElmer) or DiI (Thermo Fisher Scientific) and injected or delivered with CLP at the wound site. Luminescence intensity was monitored (the images were taken after Days 0, 1, 2, 3, 4, 9, and 14 days after surgery and each treatment) and quantified for 14 days using an IVIS Spectrum Live Imaging System (PerkinElmer).

| Quantitative reverse transcriptionpolymerase chain reaction
Quantitative reverse transcription-polymerase chain reaction (qRT-PCR) analysis was used to quantify relative messenger RNA expression. Total β-actin was used as the internal control.

| Statistical analysis
Quantitative data are expressed as the mean ± SD. Prism 9 software (GraphPad Software) was used to perform one-way analysis of variance. Tukey's test was performed when significant differences were detected. p < 0.05 were considered as statistically significant. Data shown in this study are from a representative experiment repeated three times with similar results. Compared with the flat PDMS group, VEGF gene expression levels were significantly higher in the CLP group. However, FGF2 gene expression levels were not significantly different between the two groups.

| Deposition of delivered stem cells after CLP treatment
The deposition of delivered stem cells was examined using in vivo live imaging ( Figure 4). VivoTrack 680 labeled 1.5 Â 10 6   The CLP group showed a higher wound closure rate from Days 5 to 14 (Figure 5a,b). Hematoxylin and eosin (H&E) staining and quantification of the re-epithelized wound ratio revealed enhanced skin wound regeneration in the CLP group (Figure 5c,e). To confirm epidermis (the outermost part of the skin) regeneration, immunohistochemistry with involucrin was performed. Similar to the H&E staining results, involucrin expression was much higher in the outermost part of the wound site in the CLP group than in the other groups (Figure 5d). In addition to histological results, CD31 gene expression was enhanced in the CLP group compared with other groups, whereas α-SMA gene expression was significantly enhanced in the CLP group compared to NT and 3DP groups (Figure 5f).

| DISCUSSION
Skin wounds are one of the most demanding medical issues worldwide. 22 Severe skin wounds, including chronic and large wounds, are difficult to heal using conventional wound care methods. 12  (e) Quantification of epithelized area ratio as the percentage of total wound area. Significant differences between two groups are indicated as **p < 0.01 or ***p < 0.001. (f) Relative expression of CD31 and α-SMA. Significant differences between two groups are indicated as *p < 0.05, **p < 0.01, or ***p < 0.001. CLP, stem cell-laden 3D cavity-structured patch boarder of the wound site. Therefore, injected cells cannot participate in the wound healing process after a certain point. In contrast, cells delivered using the CLP were successfully localized at the wound site and remained localized until Day 14, suggesting that the delivered stem cells have a better chance to be involved in the wound healing process by this method compared to others. As a result of the effective delivery of hADSCs with CLP, the wound healing rate and vascularization markers were significantly increased. Furthermore, an establishment of well-developed epithelial cell layer was confirmed using involucrin staining ( Figure 5). 28

| CONCLUSION
In this study, we fabricated an overdensely CLP for the treatment of severe wounds. This stem cell delivery patch system can be easily transferred to the clinical setup for the treatment of chronic or slowhealing wounds by extending the size of the cell patch. However, the material that we used (PDMS) should be substituted for skin wound favorable materials that have characteristics including gas and liquid permeability and biodegradability. To improve the wound healing effect, different cell types (e.g., keratinocytes and fibroblasts) can be codelivered with stem cells. Overall, the CLP developed in this study is an easy and effective method for the treatment of hard-to-heal wounds.