Umbilical cord mesenchymal-stem-cell-derived nanovesicles as a novel strategy to promote wound healing in diabetes

Abstract

Objective

Recent studies found that exosomes (Exo) derived from mesenchymal stem cells (MSC) (MSC-Exo) accelerated diabetic wound healing. However, the low yield during exosome extraction is still a major barrier to their clinical utility.

Methods

We constructed a method to produce umbilical cord MSC-derived nanovesicles (UCMSC-NV) by serial extrusion through filters and investigated the effects of UCMSC-NV on wound healing in vivo and in vitro, as well as the potential mechanisms.

Results

We found that the characteristics of UCMSC-NV were similar to those of exsome (UCMSC-Exo) but with much higher production yields. Further analysis showed that UCMSC-NV promoted the migration of fibroblasts and angiogenesis in vitro, and both UCMSC-NV and UCMSC-Exo showed similar therapeutic capacities for wound healing in vivo. Sequencing analysis revealed that UCMSC-NV and UCMSC-Exo had similar miRNA compositions, and the target genes of the differentially expressed miRNAs in UCMSC-NV were enriched in pathways of inflammation and damage-repair-related functions. Mass spectrometry analysis showed that UCMSC-NV encapsulated functional proteins that may achieve therapeutic effect equally as good as UCMSC-Exo.

Conclusions

UCMSC-NV are more efficacious and can be obtained at a higher yield than UCMSC-Exo and are a promising therapeutic strategy to improve wound healing in diabetes patients.

INTRODUCTION

Diabetes is a metabolic syndrome characterized by chronic hyperglycemia, and up to 25% of people with diabetes will acquire a non-healing diabetic foot ulcer in their lifetime.^[1,2] Despite the availability of improved therapies, more than 50% patients with diabetic foot ulcers have ulcers that fail to heal, which can impact their ability to perform daily activities and even lead to limb amputation.^[1] Therefore, wound healing in diabetes still poses an immense problem in modern medicine, leading to a lot of interest in this area of diabetes research.

To date, several treatment strategies for diabetic wounds are available, such as surgical debridement, negative-pressure wound treatment, artificial skin replacement, and growth factor therapy based on the control of blood glucose.^[3] However, most treatments are effective only for mild to moderate wounds, and none of the above treatments achieve 100% effectiveness in reducing the risk of amputation or recovering full skin functionalities.^[4] Traditional methods to accelerate healing are skin transplantation and laser therapy. For large-area skin damage, normal skin transplantation may only include epidermis and epithelium, resulting in delayed healing and scar formation. Furthermore, considering the wound healing process is a complex interaction process, the use of a single growth factor or drug may not reach the ideal aim.^[5]

In the past 10 years, there have been encouraging reports on the treatment of diabetic skin wounds with cell therapy.^[6] As multipotent stromal cells can differentiate into mesenchymal lineages, mesenchymal stem cells (MSCs) have been used to control several steps of the wound healing and regeneration process in disease models and have become a new option for wound treatment.^[7] These cells can migrate to injured sites and induce anti-inflammatory phenotypes via their interaction with immune system cells.^[8,9] In addition, they may promote neoangiogenesis via their secretion of angiogenic factors and collagen.^[10,11] Therefore, the use of MSCs is an attractive therapeutic strategy for contributing to tissue repair and regeneration.^[12–14] However, this method has some limitations that greatly limit its clinical application, including ethical considerations for embryonic stem cells. Moreover, the complex environment of diabetic wounds also affects the maintenance and survival of stem cells through mechanisms such as immunological rejection.^[15]

In recent years, studies have shown that MSCs act in wound healing and tissue regeneration mainly through paracrine factors, such as various functional proteins, RNA, and soluble cytokines, to improve the regeneration ability of injured wounds rather than differentiating into skin cells at the sites of injury.^[16] It has been reported that exosomes are the main effective components of MSCs in tissue repair.^[17] In addition, compared with MSCs cells, exosomes greatly reduce the possibility of immune rejection and avoid blocking pulmonary capillaries owing to their small particle size, and exosomes derived from MSCs were reported to accelerate diabetic wound healing by enhancing angiogenesis.^[18] Hou et al. also found that MSC derived extracellular vesicle as a bioactive material showed therapeutic potential for wound healing.^[5] However, the rate of exosome production by MSCs is very low.^[19] Furthermore, in late passage, due to the significantly reduced expression of MSC-related growth factor genes and proteins, the amounts of therapeutic growth factors and their mRNAs are greatly reduced in secreted exosomes.

Therefore, in this study, we constructed a method to produce umbilical cord MSC-derived nanovesicles (UCMSC-NV) and investigated their effects on skin wounds of db/db mice. The effects of UCMSC-derived exosomes (UCMSC-Exo) were also observed as a control. We compared the potential roles of UCMSC-NV and UCMSC-Exo in wound healing in diabetic mice, and further determined their effectiveness and yield. The goal of the study was to provide new insights into the clinical applications of MSCs in wound healing for diabetes patients.

METHODS Cell isolation, identification, and culture

Human UCMSCs were isolated from the umbilical cord of a full-term pregnancy. The umbilical cords of full-term fetuses were cut into tissue blocks of about 1 mm³. The tissue blocks were washed with PBS and centrifuged at 2000 rpm for 5 minutes. Then, they were resuspended in 25 mL serum-free cell culture medium (Lonza, Walkersville, MD, USA), plated into a T175 flask, and incubated at 37°C with 5% CO₂. Primary cells were harvested when the confluence reached 80%. Trilineage-induced differentiation experiments of UCMSC were carried out using oil red O staining to confirm adipogenic differentiation, Alizarin red staining to verify osteogenic differentiation, and Alcian blue staining to examine chondrogenic differentiation. The related cell surface markers CD90, CD11b, CD73, CD19, CD29, CD105, CD34, CD44, CD45, and HLA-DR (Biolegend, USA) were detected by flow cytometry.

Second- to third-generation UCMSCs in good condition were cultured in 10 cm² Petri dishes. When the cell fusion degree reached 70%, the cells were starved in serum-free medium for 48 hours. Then the supernatant was collected and prepared for exosome isolation, and the cells were collected for nanovesicle extraction.

Isolation and extraction of exosomes

The supernatant was collected and centrifuged at 4°C and 2000 × g for 20 minutes, followed by 4°C, 10,000 × g, angular centrifugation for 30 minutes, and collected and filtered through a 0.22-µm filter. The supernatant was subjected to horizontal high-speed centrifugation at 100,000 × g, 4°C, for 70 minutes, the resulting supernatant was discarded, and the exosome precipitate and protein in the culture medium were obtained. Then precooled PBS was added to the precipitate and centrifuged at 100,000 × g and 4°C for 70 minutes. To the precipitate, we added 50 µl of precooled PBS for re-suspension, and the fluid was stored at −80°C.

Preparation, isolation, and extraction of nanovesicles

The collected cells were adjusted to a concentration of 2 to 3 × 10⁶ mL⁻¹, then passed through 10-µm, 5-µm, 1-µm, 400-nm, and 200-nm polycarbonate films serially by extrusion (11 or 21 times). The extruded vesicles were centrifuged horizontally at 4°C for 10 minutes, and the supernatant was collected. After centrifugation at 4°C, 2000 × g for 20 minutes and 4°C, 10,000 × g, angular centrifugation for 30 minutes, the supernatant was collected and filtered through a 0.22-µm filter. This was followed by 100,000 × g, 4°C, horizontal high-speed centrifugation for 70 minutes, and removal of the supernatant. Then, 50 µl pre-cooled PBS was added to the tube to suspend the nanovesicles, which were then stored at −80°C.

Characteristic identification

Morphological observation was by transmission electron microscope, and particle size distribution and potential were measured on a Malver Zetasizer Nano ZS instrument. Surface marker proteins were detected by western blot. Briefly, total protein was extracted by RIPA lysis buffer and then separated by SDS-PAGE and transferred to polyvinylidene fluoride (PVDF) membranes. After blocking via 5% skimmed milk, membranes were incubated (overnight, 4°C) with specific primary anti-CD63 (ab217345, 1:1500, Abcam, Cambridge, UK), CD81 (ab109201, 1:1000, Abcam, Cambridge, UK) and TSG101 (ab125011, 1:1000, Abcam, Cambridge, UK). Then the protein intensity was determined and measured by Image Lab software (5.2 Version, Bio-Rad Laboratories Co. Ltd, CA, USA).

The ability of vesicles to enter cells by endocytosis was assessed. First, the exosomes and vesicles were dyed with 10 µl DiI for 10 minutes at room temperature. Human umbilical cord endothelial cells (HUVEC) were cultured in 24-well plates (500 µl per well) at 1 × 10⁵ cells mL⁻¹. Then pre-dyed UCMSC-NV and UCMSC-Exo were added and the cells cultured for 6 hours. The cells were then fixed with 4% paraformaldehyde, stained with DAPI, and observed by confocal microscopy (TCS-SP8, Leica, Germany).

Cell migration analysis by wound-healing assay and Transwell analysis

For the wound-healing assay, after the HUVEC wells were almost full of cells, they were scratched with a plastic pipette tip, and 2 mL of serum-free DMEM medium containing UCMSC-NV was added. Pictures were taken 0, 12, 24, 36, and 48 hours later under a microscope (Nikon, Japan).

For the Transwell analysis, HUVEC cells were cultured in 12-well plates, and 1 mL of DMEM complete medium containing UCMSC-NV (extruded NV suspension at 10 µg mL⁻¹) was added. After 24 hours, the cells were re-suspended in serum-free culture medium and added to the upper chamber of a Transwell, then 900 µl of complete medium was added to the lower chamber. After 12 hours of culture (the culture time depended on the specific cells), the cells were washed with PBS, fixed with ethanol, stained with crystal violet, and observed under a microscope (Nikon, Japan).

Angiogenesis assay in vitro

µicro-slides in wells were coated with precooled Matrigel matrix glue at 4°C, 10 µl per well, and solidified in a 37°C incubator for 30 minutes. HUVEC cells were pretreated with UCMSC-NV or UCMSC-Exo for 24 hours, and 1 × 10⁵ cells mL⁻¹ (diluted with serum-free DMEM medium) was added to the µ-slides. After being cultured for 6–8 hours, the degree of tubule formation was observed under a microscope. At least five photos per well were taken, and the number of tubules, total length, and number of tubule connections were counted.

Animal experiments

Seven-week-old C57 BL/KS db/db male mice (Lepr-KO/KO, n = 10) were purchased from Shanghai SLAC Laboratory Animal Co., Ltd. (China). All experimental protocols and animal handling procedures were conducted according to the recommendations in Guide for the Care and Use of Laboratory Animals, published by the National Institutes of Health (Publication No. 80–23, revised in 1996). The procedures complied with the Declaration of Helsinki and the study was approved by the Experimental Animal Committee of Zhejiang Province People's Hospital (Hangzhou, China).

A skin injury model was established after 1 week of feeding. First, diabetic mice were anesthetized with CO₂, and a sterile punch was used to make a round skin injury of a diameter of 8 mm on both sides of the spine. Then, mice with full-thickness skin defects were randomly divided into three groups: control group (treated with normal saline), NV group (treated with 10 µg UCMSC-NV), and Exo group (treated with 10 µg of UCMSC-Exo).

Wound photographs were taken, and the percentage wound healing area was analyzed using the Image-Proplus image analysis system. The wound healing rate was calculated with the formula (wound healing, %) = (64 mm²-detection area)/64 mm² × 100%. For molecular and histologic analyses, the wounded skin of the mice was resected, fixed in 10% neutral buffered formalin for at least 48 hours, and embedded in paraffin. Skin sections (4 µm) were deparaffinized, rehydrated in graded ethanol, and stained with hematoxylin and eosin (HE) and Masson's trichrome. Inflammatory cell infiltration and the epidermal thickness of the HE-stained skin in the center of the wound stained were evaluated by Image J software. The degrees of dermalization and scar formation were also evaluated. Collagen formation at the skin wound was observed by Masson's staining. The level of type I/III collagen in the wound tissue was detected by immunohistochemical labeling. Immunofluorescence labeling was used to determine the number of CD31-positive and CD31/α-SMA-double-positive cells in the skin wound to detect the formation of new blood vessels.

RNA sequencing and bioinformatics

After the miRNA-seq data were standardized, the miRNA count expression matrix was identified using the “limma” package of R software. Then the differentially expressed miRNA were obtained. To avoid the loss of important miRNAs, we set the thresholds to a log2|fold change| > 0.5 and a p-value < 0.05. Target genes of miRNA were predicted using the ENCORI database (Encyclopedia of RNA Interactomes, http://starbase.sysu.edu.cn/index.php). The “org.Hs.eg.db” and “clusterProfiler” packages of R software were used for gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) functional enrichment analysis of potential miRNA targeted genes.

Mass spectrometric analysis

Dithiothreitol (DTT) was added into 20 µg protein solution samples at a final concentration of 10 mM and vortexed for 45 minutes at room temperature. Then, indoleacetic acid (IAA) was added to a final concentration of 20 mM, and the solution incubated for 30 minutes at room temperature in the dark. Pre-cooled acetone was added into the samples 6–9 times and incubated at 4°C for 2 hours for protein precipitation. Then, the samples were centrifuged at 10,000 × g for 10 minutes and the supernatant discarded. The protein pellet was resuspended in 100 mM ABC solution (protein concentration, 1 µg µl⁻¹), trypsin was added at a ratio of 1:50, and the pellet was hydrolyzed in a 37°C water bath overnight. Then, the demineralized peptide solution was dried in a rotary vacuum. The peptide was resuspended in 20 µl of 2% acetonitrile (ACN) + 0.1% formic acid (FA) and analyzed by LC-MS/MS. To eliminate the very low-scoring random peptide matches automatically, uncertain proteins were eliminated, and the target data was analyzed.

For the standardized proteomics data, proteins with relatively high expression in the UCMSC-NV versus exo, NV versus MSC, and EXO versus MSC groups were selected. The “org.HS.Eg.db” and “clusterProfiler” packages of R software were used to transform and annotate Gene ID, and functional enrichment analysis was carried out. The “pheatmap” and “ComplexHeatmap” packages were used to produce heatmaps to display the differential protein expression, and the “ggplot2” package was used for visualization.

Statistics and data analysis

All statistical analyses were performed using the Statistical Package for the Social Sciences (version 13.0; SPSS Inc, Chicago, IL, USA). Data presented in bar graphs are shown as the mean (± SEM) of replicates. A Student t-test was used for comparison of two datasets if they showed normal distribution, and multiple comparisons were performed by analysis of variance (ANOVA). For all tests, p-values were obtained from two-tailed statistical tests, and p-values less than 0.05 were considered statistically significant.

RESULTS Identification of UCMSCs

After subculture for one generation, cells that had adhered to the dish bottom in a monolayer presented fibroblast-like morphology (Figure 1A). The differentiation potential of UCMSCs was verified, and the isolated and purified UCMSCs presented multidirectional differentiation potential (Figure 1A). Osteogenic differentiation results showed that calcium nodules were stained with Alizarin red. Lipid droplets (with oil-red O staining) were observed in UCMSCs after adipogenic differentiation induction. Chondrogenic differentiation results revealed that acid mucopolysaccharides were stained with Alcian blue, representing cartilage differentiation.

FIGURE 1. Characterization of UCMSC by multi-lineage differentiation and flow cytometry analysis. A, Typical photos of passage four (P4) cells from primary UCMSC culture; Alizarin red staining showing mineralized nodule formation in UCMSC after 4 weeks of osteogenic induction. After cartilage formation was induced for 3 weeks, the cell was stained with toluidine blue. Positive staining with Alcian blue showed the formation of chondrocyte-like cells. Intracellular oil red-O staining showed the formation of lipid-rich vacuoles in UCMSC after induction for 2 weeks. B, Flow cytometry analysis using UCMSC biomarkers. The white plot showed the negative peak of control, and the red peak showed the protein expression of sample

Flow cytometry showed that UCMSCs were positive for the stem-cell markers CD29, CD105, CD90, and CD44 and negative for CD11b, CD73, CD45, CD34, CD19, and HLA-DR (Figure 1B). These results suggested that the UCMSCs originated from MSCs.

Characterization of UCMSC-NV and UCMSC-Exo

The morphology and particle size of UCMSC-NV and UCMSC-Exo were observed by transmission electron microscopy, and the particle size distribution and potential were measured with a Malver Zetasizer Nano ZS instrument. We found that UCMSC-NV and UCMSC-Exo both exhibited circular film-like structures under transmission electron microscope (Figure 2A). The size distribution of UCMSC-NV and UCMSC-Exo were relatively consistent, and all were approximately 100 nm in diameter (Figure 2B). The potentials were all negative, with UCMSC-NV having lower values than UCMSC-Exo (Figure 2C). Cell endocytosis experiments showed that both UCMSC-NV and UCMSC-Exo (red DiI staining) were absorbed by HUVEC cells within 6 hours, and there was no significant difference in their degree of uptake (Figure 2D). Western blot results identified CD63 and TSG101 in both UCMSC-NV and UCMSC-Exo, while CD81 was identified in UCMSC-NV and cells but not in UCMSC-Exo (Figure 2E).

FIGURE 2. Characterization of UCMSC-NV and UCMSC-Exo. A, Morphological observation under transmission electron microscope. B, Particle size distribution showed UCMSC-NV and UCMSC-Exo particles were similar in size. C, Zeta potential of UCMSC-NV was greater than that of UCMSC-Exo. D, Endocytosis of UCMSC-NV and UCMSC-Exo by HUVEC was observed under fluorescence microscope. E, Biomarkers were detected by western blotting. E, Yield of UCMSC-NV was higher than that of UCMSC-Exo. *p [less than] 0.05, ***p [less than] 0.001

Yields of UCMSC-NV and UCMSC-Exo

UCMSC-NV and UCMSC-Exo were extracted when the total number of cells was 1 × 10⁷. The cells were then resuspended in 10 mL PBS, and 7 mL solution remained after extrusion. After ultracentrifugation, they were resuspended in 210 µl PBS. The protein concentration was determined by BCA method, and the yields of UCMSC-NV and UCMSC-Exo were 239.9 ± 48.5 and 12.0 ± 3.3 µg 10⁻⁷ cells, respectively; that is, the yield of UCMSC-NV was about 20 times that of UCMSC-Exo (Figure 2F).

UCMSC-NV effect on migration of fibroblasts and angiogenesis in vitro

The effects of UCMSC-NV on the migration of fibroblast L929 and HUVEC were detected by wound healing assay and Transwell analysis, as shown in Figure 3A,B. Both UCMSC-NV and UCMSC-Exo promoted the migration of HUVEC and L929, and the effect of UCMSC-NV is significantly greater than that of UCMSC-Exo. Angiogenesis also contributes to wound healing. We, therefore, examined the effects of UCMSC-NV on HUVEC in a tube-formation assay. The tube length and total number of loops of HUVEC treated with UCMSC-NV or UCMSC-Exo for 8 hours significantly increased compared with those in the control group, with a better effect seen in the UCMSC-NV group than the UCMSC-Exo group (Figure 3E–G). These data demonstrate that the beneficial effects of UCMSC-NV on fibroblasts and endothelial cells may contribute to promoting wound healing.

FIGURE 3. UCMSC-NV promoted migration of fibroblast and angiogenesis of HUVEC. A, Transwell results showing both UCMSC-NV and UCMSC-Exo promoted the migration of HUVEC. B, Wound healing assay showing both UCMSC-NV and UCMSC-Exo promoted the migration of L92. C, Tube-formation assay showing both UCMSC-NV and UCMSC-Exo promoted tube-formation by HUVEC. *p [less than] 0.05 compared with control group

Therapeutic effect of UCMSC-NV on wound healing in diabetes

To explore the effect of UCMSC-NV on cutaneous wound healing efficiency, UCMSC-NV and UCMSC-Exo were applied to full-thickness diabetic wounds, and saline was used in the blank control treatment group. Figure 4 illustrates the size changes of diabetic wounds from the three groups on days 0, 3, 14, and 21 post-injury. Gross observation of wound closure in mice showed that all wounds treated with UCMSC-NV and UCMSC-Exo displayed a remarkable decrease in size by days 14 and 21, while the negative control wounds exhibited a slow decrease in size during the experimental time. Consistent with the gross observation, the quantitative wound closure rates of the UCMSC-NV and UCMSC-Exo groups indicated their faster healing rates compared to the control group during the whole healing process, but the difference between UCMSC-NV and UCMSC-Exo groups was not significant (Figure 4B).

FIGURE 4. Therapeutic effects of UCMSC-NV and UCMSC-Exo on wound healing in diabetic mice. A, Representative photos of wounds on days 0, 3, 14, and 21 after UCMSC-NV and UCMSC-Exo treatment. B, Statistical histogram of wound closure at different time points. C, Representative pictures of hematoxylin-eosin staining of wound tissue sections 14 and 21 days after treatment. D, Representative pictures of Masson's staining of wound tissue sections after UCMSC-NV or UCMSC-Exo treatment. *p [less than] 0.05

HE and Masson staining showed that both UCMSC-NV and UCMSC-Exo treatments reduced inflammation and promoted the growth of fiber and collagen cells, which are beneficial for wound healing in diabetic mice, and there was no significant difference between the two groups (Figure 4C,D). There was no significant difference in the effect on collagen deposition in wounded tissues between the UCMSC-NV and UCMSC-Exo groups. These results suggested that both UCMSC-NV and UCMSC-Exo had similar therapeutic capacities for wound healing in diabetic mice.

UCMSC-NV effect on angiogenesis and collagenous fiber formation in vivo

Immunofluorescence results showed that both UCMSC-NV and UCMSC-Exo significantly upregulated CD31 and α-SMA expression (Figure 5A), and there was no significant difference between the two groups. The immunohistochemical results showed that, compared with the control treatment, UCMSC-NV significantly increased the expression of collagen III to a higher degree than UCMSC-Exo (Figure 5B). These results suggest that both UCMSC-NV and UCMSC-Exo promoted wound healing by inducing angiogenesis and collagenous fiber formation in diabetic skin wounds.

FIGURE 5. Effects of UCMSC-NV and UCMSC-Exo on angiogenesis and collagen deposition in wound tissue of diabetic mice. A, After UCMSC-NV and UCMSC-Exo treatment for 21 days, wound tissues were labeled with immunofluorescent antibodies for detecting CD31 and α-SMA, and the average intensity of expression levels was analyzed. *(CD31) or # (α-SMA), p [less than] 0.05 compared with control group. B, Wound tissues were stained immunohistochemically to detect collagen I and III levels, and the average intensity of expression is shown in the statistical histogram. *p [less than] 0.05 compared with control group

Effect of UCMSC-NV MiRNA cargo on wound healing in diabetes

After standardizing the expression profile count data, PCA cluster analysis was carried out to explore and visualize the differences in total miRNA (Figure 6A). Both the UCMSC-NV group and UCMSC-Exo group were plotted a distance away from the UCMSC group, which showed that their miRNA composition was significantly different from that of the UCMSC group. The distance between the UCMSC-NV group and UCMSC-Exo group was small, suggesting that they may have similar miRNA compositions.

FIGURE 6. RNA sequencing and mass spectrometric analysis showed that the components of UCMSC-NV are beneficial for wound healing. A, Principal component analysis (PCA) of miRNA in UCMSC-NV, UCMSC-Exo, and UCMSC groups; Volcano map of differential miRNA in UCMSC-NV; GO and KEGG pathway analysis of targeted genes of differential expression miRNA. B, Heatmap of proteins in UCMSC-NV, UCMSC-Exo and UCMSC; GO and KEGG pathway analysis of upregulated proteins in UCMSC-NV

The differential miRNA expression of the UCMSC-NV group versus UCMSC-Exo group was identified by the R software “limma” package and showed as Volcano map. As a result, 657 differentially expressed miRNAs were found, including 233 that were up-regulated and 424 down-regulated in the UCMSC-NV group compared with the UCMSC-Exo group.

After filtering these miRNAs, the target genes of significant miRNAs were predicted with the ENCORI database (http://starbase.sysu.edu.cn/index.php), then analyzed by gene enrichment analysis. We found that the PI3K-Akt signaling pathway, p53 signaling pathway, and MAPK signaling pathway were significantly enriched in both the UCMSC-NV and UCMSC-Exo groups. Most of the enriched pathways are involved in inflammation and damage-repair-related functions, which suggests that UCMSC-derived nanovesicles, with a high production rate, can be used to functionally replace UCMSC-derived exosomes.

Identification of proteins in UCMSC-NV

According to the annotations of UniProt ID, A heatmap was used to visualize the differential expression of the proteins (Figure 6B). Then 2891 proteins were found to be upregulated in UCMSC-NV compared to UCMSC-Exo. Enrichment analysis was showed that diabetic cardiomyopathy, chemical carcinogenesis reactive oxygen species, and the oxidative phosphorylation pathway were significantly enriched in the UCMSC-NV group.

DISCUSSION

MSCs are reported to play important roles in the continuous regeneration of the epidermis of the skin.^[20] At present, UCMSCs are the cell type of choice in cell therapy, and the evidence accumulated thus far indicates that allogeneic UCMSC administration is a safe way to provide low expression levels of class I and class II human leukocyte antigens.^[21,22] Thus, the fewer ethical issues, the lack of pain involved in obtaining abandoned umbilical cords, and the lack of immune-response complications are the prominent benefits of UCMSCs compared to other MSC resources.^[23]

In the present study, we constructed a method to produce UCMSC-NV and analyzed the efficacy of UCMSC-NV in wound healing in mice with diabetes. First, we confirmed that our cells were UCMSCs based on morphological observations and supported these with marker analysis results. Then we produced UCMSC-NV via an extrusion method and analyzed their morphology and particle size distribution. The zeta potential value of UCMSC-NV was more negative than that of UCMSC-Exo suggesting that most of the UCMSC-NV originated from cell membrane, and may be more stable than UCMSC-Exo. UCMSC-NV and UCMSC-Exo both exhibited circular film-like structures and an approximate 100-nm particle size and could be endocytosed into HUVECs. The potential mechanism of UCMSC-NV may attribute to the special properties of cell membrane. When the cell membrane is broken, it will spontaneously form a kind of composition with a closed bilayer structure. Meanwhile, the main advantages of the method were easy operation and save time compared the traditional ultracentrifugation. Most notably, the yield of UCMSC-NV was more than 20 times that of UCMSC-Exo, suggesting that UCMSC-NV may be beneficial when clinical administration reaches a therapeutically effective dosage.

The effects of UCMSC-NV on the migration of fibroblast L929 was detected by wound healing assay and Transwell analysis. Both UCMSC-NV and UCMSC-Exo promoted the migration of fibroblast L929 cells, and UCMSC-NV were significantly more effective than UCMSC-Exo. Recent studies also suggested that MSCs stimulated human dermal fibroblasts via paracrine effects and enhanced cutaneous wound healing.^[24,25] The effect of UCMSC-NV on HUVEC tube-formation also revealed that UCMSC-NV promoted angiogenesis, suggesting that UCMSC-NV may affect wound healing by promoting fibroblast migration and angiogenesis.

The effect of UCMSC-NV on cutaneous wound repair was also confirmed in vivo, and UCMSC-NV-treated db/db mice showed significantly increased expression of CD31, α-SMA, and collagen I and III, and faster healing rates than the control mice. These results showed that UCMSC-NV promoted fibroblast growth and angiogenesis as well as collagen formation. Human MSCs have been reported to secrete several other extracellular matrix-regulating materials (collagen, elastin, and fibronectin).^[26,27] Two major collagenous components of the dermis are collagen types I and III.^[28] As an important of the ECM components of the skin, collagen plays a key role in growth and determines skin's elasticity.^[29]

Like exosomes, UCMSC-NV carry various cargo molecules, such as functional proteins, miRNAs, and signal lipids, that mediate cell-to-cell communication by initiating a series of biological responses in recipient cells. Sequencing analysis of miRNA within the particles revealed that UCMSC-NV and UCMSC-Exo may have similar miRNA compositions. Furthermore, the target genes of the differentially expressed miRNAs in UCMSC-NV were enriched in pathways of inflammation and damage-repair-related functions. Mass spectrometry analysis also showed that the functional proteins encapsulated in UCMSC-NV have additional function in repair after damage, suggesting that UCMSC-NV may achieve therapeutic effects as good as those in UCMSC-Exo.

CONCLUSION

We constructed a method to produce UCMSC-NV by serial extrusion through filters, with advantages of easy operation and save time compared the traditional ultracentrifugation. Furthermore, we revealed that the therapeutic effects of UCMSC-Exo on wound healing in diabetes could be achieved using UCMSC-NV, which can be obtained at a greater yield than exosomes. Additionally, we discovered the components of UCMSC-NV were similar to those of UCMSC-Exo, suggesting that the function and role of UCMSC-NV may be analogous to those of UCMSC-Exo. These results suggested that UCMSC-NV represent a novel strategy to improve wound healing in diabetic patients.

CONFLICT OF INTEREST

The authors declare no conflict of interest.

DATA AVAILABILITY STATEMENT

Research data are not shared.