Wound healing effects of exosomes from hypoxia- and normoxia-expanded adipose-derived stem cells of infant and adult origin

Abstract

Background

Exosomes derived from adipose-derived stem cells (ADSC-Exos) have emerged as promising therapeutic agents for promoting wound healing through paracrine mechanisms. While adult ADSC-Exos have been extensively studied, those derived from infant ADSCs have received limited attention, despite their potentially superior regenerative capacity. In addition, hypoxia preconditioning has been shown to enhance the therapeutic efficacy of stem cells and their secretome; however, its influence on exosome function across different donor ages remains unclear. This study aimed to evaluate and compare the wound healing effects of ADSC-Exos from adult and infant sources expanded under hypoxic and normoxic conditions in a diabetic mouse model.

Methods

ADSC-Exos were isolated from adult and infant donors cultured under normoxic and hypoxic conditions. Exosomes were characterized by morphology, transmission electron microscopy (TEM) imaging, nanoparticle tracking analysis (NTA), and Western blotting. RNA sequencing was performed to profile miRNAs encapsulated in the exosomes. In vitro assays assessed the effects of exosomes on proliferation, migration, and wound healing in human dermal fibroblasts (HDFs) under high-glucose conditions. In vivo efficacy was evaluated using full-thickness wounds in db/db diabetic mice to assess the therapeutic effects of the exosomes.

Results

Normoxic infant ADSC-Exos exhibited a broader size distribution and a larger mean particle size. Several wound-healing-related miRNAs were upregulated in hypoxic ADSC-Exos from both adult and infant sources. Normoxic infant ADSC-Exos significantly enhanced HDF proliferation and migration, whereas hypoxic adult ADSC-Exos exhibited superior early wound closure. In vivo, normoxic adult ADSC-Exos achieved the fastest wound closure at day 7, and normoxic infant ADSC-Exos showed significantly greater wound closure by day 10. All ADSC-Exos promoted dermal regeneration, with hypoxic adult ADSC-Exos showing the most pronounced effects.

Conclusion

Infant ADSC-Exos demonstrated inherent advantages under normoxic conditions, including enhanced effects on cell proliferation, migration and wound regeneration. While hypoxic preconditioning enhanced the efficacy of adult ADSC-Exos, it provided limited additional benefit to infant-derived exosomes, highlighting their intrinsic suitability for regenerative therapies.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13287-025-04636-4.

Background

Wound healing is a complex and dynamic process that involves a series of overlapping stages: hemostasis, inflammation, proliferation, and remodeling. Effective wound healing requires coordinated cellular and molecular responses to restore the integrity and function of damaged tissues [1]. In chronic conditions such as diabetes, wound healing is often impaired, leading to persistent wounds that pose significant clinical challenges and increase the risk of severe complications, including infections and amputations [2]. In contrast, evaluating the therapeutic effect on wound healing in healthy animals can be challenging, as their rapid self-healing often obscures the significance of treatments. As a result, healing-impaired diabetic mice serve as a widely used model for delayed wound healing [3]. The critical characteristics of non-healing chronic diabetic wounds would include reduced vascularization, hypoxia, elevated oxidative stress, and infection. To address these challenges, adipose-derived stem cells (ADSCs)-derived exosomes (ADSC-Exos) loaded with oxygen-releasing OxOBand have been developed for the treatment and management of infected diabetic wounds [4].

ADSCs have garnered significant attention in regenerative medicine due to their multipotent capabilities and their potential to differentiate into various cell types, including keratinocytes [5] and dermal fibroblasts [6], both of which are critical for skin repair and regeneration. In addition to their differentiation potential, ADSCs contribute to maintaining skin tissue structure by physiologically responding to local injuries and seeding younger cells into the outer epidermis, thereby inducing rejuvenating mechanisms [7]. Infant ADSCs, which can be extracted from excised polydactyly fat tissue typically considered surgical waste, have demonstrated superior stemness, enhanced proliferation, and lineage-specific differentiation potential, making them an attractive ADSC source for cell-based therapies [8, 9]. Moreover, ADSC-conditioned medium has been shown to promote human dermal fibroblast proliferation, migration, and synthesis of type Ⅰ and Ⅲ collagen [10], further highlighting their regenerative potential. Recently, exosomes derived from ADSCs have emerged as a promising acellular therapeutic strategy. These nano-sized vesicles can modulate inflammation, promote keratinocyte proliferation and collagen deposition, and accelerate epithelialization, in preclinical wound healing models [11, 12]. While infant ADSC exhibit enhanced regenerative potential, identifying shared bioactive components—such as exosomal miRNAs—between infant and adult ADSCs may reveal conserved mechanisms underlying their therapeutic effects. Therefore, the present study attempted to evaluate the wound healing potential of exosomes derived from infant ADSCs and to compare them with adult-derived counterparts.

Emerging evidence has highlighted the role of exosomes, small extracellular vesicles secreted by cells, in mediating the therapeutic effects of mesenchymal stem cells (MSCs) [13]. Among various types of MSCs, ADSCs exhibit key defining characteristics, including the expression of CD73, CD90, and CD105, while lacking hematopoietic markers such as CD45 and CD34, in accordance with the criteria defined by the International Society for Cell & Gene Therapy (ISCT) [14]. ADSC-Exos carry bioactive molecules such as proteins, lipids, and RNAs, which can modulate various cellular processes essential for wound healing [15]. Notably, the therapeutic potential of ADSC-Exos can be significantly influenced by the environmental conditions under which the parent ADSCs are cultured [16], with hypoxia shown to enhance their wound healing potential [17–19]. While the biological effects of adult ADSC-Exos have been extensively studied, the therapeutic capacity of exosomes derived from infant ADSCs-known to possess higher stemness-remains largely unexplored.

Studies have demonstrated that hypoxia-conditioned ADSC-Exos embedded in hydrogels can enhance the healing of diabetic wounds by delivering circular RNAs such as circ-snhg11 and promoting the proliferation of M2-like macrophages [18]. Additionally, hypoxic ADSC-Exos have been found to improve cell proliferation, migration, and angiogenesis in H₂O₂-treated human umbilical vein endothelial cells by activating the USP22/HIF-1α axis and promoting H19 expression [19]. Furthermore, these exosomes can mitigate UV-induced skin damage through the delivery of circ-Ash1l and inhibition of ferroptosis [20]. Collectively, these findings highlight the multifaceted regenerative capabilities of hypoxia-induced ADSC-Exos in diverse pathological conditions, supporting their potential in therapeutic applications for skin repair.

In our previous study, we found that expanding ADSCs under hypoxic conditions could enhance the neuronal [21] and chondrogenic differentiation [22] potential in vitro. This suggests that preserving ADSCs in a niche-like hypoxic environment may reduce cellular senescence and help maintain their stemness. In addition, ADSCs have also been reported to improve the healing of irradiated skin tissue [23]. Building upon these findings, the present study aims to investigate the comparative efficacy of exosomes derived from normoxic and hypoxic infant ADSCs, as well as adult ADSCs, in promoting wound healing in diabetic mice. By comparing exosomal effects across different oxygen conditions and developmental stages, we seek to characterize age- and environment-dependent differences in functional and therapeutic properties of ADSC-Exos. Furthermore, this study employs the db/db mouse model, which closely mimics the impaired wound healing observed in human diabetic ulcers due to its leptin receptor deficiency, making it a suitable platform for evaluating efficacy. These results are expected to clarify how developmental stage and hypoxic conditioning influence the regenerative potential of ADSC-Exos, thereby guiding the development of more effective exosome-based therapies for chronic wound healing.

Materials and methods

Preparation of infant and adult adscs, and human dermal fibroblasts (HDFs)

The infant ADSCs were collected from fat tissue obtained from excised redundant thumbs after surgical reconstruction for 5 children with polydactyly. The procedures were Institutional Review Board (IRB)-approved. The adult ADSCs were isolated from lipoaspirate obtained from 6 donors undergoing thigh liposuction under local anesthesia. Once soft tissues will be infiltrated with a solution of saline plus epinephrine (1:1,000,000), subcutaneous fat was suctioned (− 500 to − 700 mmHg) using a conventional liposuction machine equipped with a 2.5-mm (inner diameter) cannula. The study protocol was also approved by the IRB. The harvested fat was enzymatically digested using 1 mg/mL collagenase (Wako, Osaka, Japan) in Hank’s balanced salt solution (HBSS; Gibco, Carlsbad, CA, USA) by incubation in a shaking water bath (37 °C, 30 min). The resulting stromal vascular fraction (SVF) was washed with PBS three times to remove residual collagenase, then filtered and centrifuged. The SVF pellet was resuspended with an appropriate amount of medium for seeding (2–3 × 10⁶ /10-cm dish). ADSCs were isolated based on their adherence to tissue-culture plastic and expanded in Dulbecco’s Modified Eagle Media (DMEM; Gibco), supplemented with 10% fetal bovine serum (FBS; Invitrogen, Carlsbad, CA, USA) and antibiotic-antimycotic solution (Corning Life Science, Corning, NY, USA). Medium was changed every two days, and cell were passaged every four days at a 1:5 ratio before reaching 80% confluence. For hypoxic expansion, cells were cultured in 94% N₂, 5% CO₂, and 1% O₂ (38); normoxic conditions were maintained in 5% CO₂ and 20% O₂.

HDFs were purchased from ScienCell (Carlsbad, CA, USA) and cultured in DMEM containing 10% FBS at 37℃ with 5% CO₂.

ADSCs-exos preparation

When the cells reached approximately 90% confluence, they were washed 1–2 times with PBS and cultured in 7 mL of DMEM supplemented with 10% exosome-depleted FBS for an additional 48 h. The exosome-depleted FBS was prepared by filtering standard FBS through 100-kDa molecular weight cutoff (MWCO) centrifugal filter (Amicon Ultra, Merck Millipore, Billerica, MA, USA), and the filtrate was used for subsequent culture. After 48 h, the conditioned medium was collected and centrifuged at 300×g for 5 min to remove floating cells. The supernatant was transferred to a 50-mL centrifuge tube and centrifuged again at 2000×g for 30 min at 4 °C to remove cell debris. The resulting supernatant was filtered through a 0.22 μm pore membrane filter (Merck Millipore). The filtered solution was then concentrated using Amicon Ultra 15-mL Centrifugal Filters (MWCO = 100 kDa; Merck Millipore) by centrifugation at 4000 rpm for about 10 min at 4 °C. The filtrate was discarded, and this concentration step was repeated once to ensure efficient removal of low-molecular-weight contaminants. After the second concentration step, 15 mL of PBS was added to the retentate, followed by another round of centrifugation to wash the retained exosomes, thereby removing residual contaminants and ensuring exosome purity. The final exosomes concentrate was collected and stored at − 80 °C until use.

For protein quantification, the exosome samples were diluted 1:1 with RIPA Lysis Buffer (Merck Millipore), and protein concentration was determined using a BCA assay kit (Thermo Scientific Pierce, Rockford, IL, USA) .

Transmission electron microscope (TEM)

Exosomes derived from hypoxic or normoxic ADSCs, collected from the cell culture medium using ultrafiltration centrifugal filters, were resuspended in PBS. The exosome suspension was adsorbed onto formvar/carbon-coated copper EM grids (Sigma-Aldrich) at room temperature for 10 min. Excess liquid was removed using filter paper, and the grids were dried in a desiccator overnight. Finally, the attachment of exosomes to the grids was observed using a TEM (JEM-1400plus; JOEL, Japan) at 100 kV.

Nanoparticle tracking analysis (NTA)

NTA (Zetaview, Diessen, Germany) was used to determine the concentration and size distribution of exosomes derived from hypoxic or normoxic ADSCs. For instrument calibration, 100 nm polystyrene latex beads were used. The exosomes were then diluted to 1 ml, loaded into a syringe, and introduced at a flow rate of 25 µl/s using a syringe pump, with visual capture performed by a camera. Each video analyzed at least 2000–4000 exosomes.

Western blot analysis

Isolated ADSC-derived exosomes were confirmed for protein expression by Western blotting. The protein extraction reagent (M-PER; Pierce, Rockford, IL) was mixed with Halt™ protease inhibitor cocktail (Pierce) on ice, and proteins were extracted for 5 min. Protein concentrations were determined using the BCA assay kit (Thermo Scientific Pierce). Equal amounts of protein samples were mixed with sample buffer, heated at 95 °C for 5 min, and analyzed by 10% SDS–polyacrylamide gel electrophoresis. The separated proteins were then transferred onto PVDF membrane filters. The membranes were blocked for 1 h and incubated overnight at 4 °C with primary antibodies against CD63 and CD81. After washing, the membranes were incubated for 1 h with horseradish peroxidase (HRP) -conjugated goat anti-mouse or anti-rabbit IgG (BD Bioscience, San Jose, CA, USA). Finally, the proteins were visualized using a chemiluminescence detection assay (Perkin-Elmer, Waltham, MA, USA).

Dil stained exosomes uptake by HDFs

HDF cells (passage 7–9) were cultured in 8-well chamber slides (Ibidi, Gräfelfing, Germany), with each well seeded with 6.37 × 10⁴ cells in 500 µl of medium. After overnight incubation to allow cell attachment, Vybrant™ Dil Cell-Labeling Solution (Thermo Fisher Scientific, Waltham, MA, USA) was used to label the exosomes. The exosomes were mixed with Dil dye at a ratio of 20 µg exosomes per 1 × 10⁵ cells, using a solution of 1000 µl of exosome suspension to 5 µl of Dil dye [24]. The mixture was incubated in the dark for 30 min, then added to a 100 kDa centrifugal filter unit, with 4 ml of PBS, and centrifuged at 3000 ×g for 15 min. This washing process was repeated three times to remove any excess dye. A control group was processed in the same way [24, 25].

Before adding the labeled exosomes, the cells were washed once with PBS. The labeled exosomes were then resuspended in DMEM with a D-glucose concentration of 4.5 g/L (DMEM-HG; Cytiva, Marlborough, MA, USA), with a final volume of 500 µl. The labeled exosomes were added to the cells and co-cultured for 24 h. After incubation, the cells were washed with PBS to remove any unbound exosomes, fixed with 4% PFA for 15 min, and stained with 4’, 6-diamidino-2-phenylindole (DAPI, Sigma-Aldrich) to label the nuclei. The cells were then mounted and observed under an Olympus BX43 microscope (Tokyo, Japan).

Cell proliferation assay

HDFs (passage 7 to 9) were seeded into a 96-well plate at a density of 2,000 cells per well in 100 µl of DMEM supplemented with 10% FBS. After 24 h of incubation, the medium was replaced with DMEM-HG supplemented with 1% FBS and exosomes at a concentration of 100 µg/ml. The cells were cultured for 24 and 48 h. After each time point, both the control and experimental groups were switched to a medium containing cell counting kit-8 (CCK-8, Dojindo, Japan) and incubated for 3 h. Absorbance was measured at 450 nm, and data analysis was presented as the ratio of absorbance values of the experimental group relative to the control group.

Transwell migration assay

HDFs (passages 7 to 9) were used in this experiment. A 6.5 mm Transwell^® with an 8.0 μm pore polycarbonate membrane insert (Corning) was placed into a Transwell 24-well plate. In the upper chamber of each insert, 4 × 10⁴ cells were seeded, suspended in low-serum (5% FBS) DMEM-HG medium. Simultaneously, the lower chamber was filled with DMEM-HG complete medium (containing 10% FBS) with or without 100 µg/ml exosomes. After 12 h, the cells were washed once with PBS and fixed with 4% PFA. The cells on both sides of the membrane were stained with 0.4% crystal violet for 1 h, followed by two PBS washes. The cells adhered to the upper side of the membrane were removed using a cotton swab, and the cells that migrated to the lower side were photographed and observed [26]. At 10× magnification, 10 fields of view were selected from each group to quantify the crystal violet-stained area. The data were presented as the ratio of stained areas between the experimental and control groups.

In vitro wound healing assay

HDFs (passages 7 to 9) were used in this experiment. To create a 500 μm scratch, the culture-insert 2 well (Ibidi) was placed in a 12-well plate, and 1 × 10⁴ cells were seeded into each well of the culture insert. After 24 h of incubation, the culture insert was removed, and the cells were washed once with 1 ml of PBS. To simulate a high-glucose environment, the medium was replaced with 1 ml of serum-free DMEM-HG containing 100 µg/ml of exosomes. The control group was treated with PBS only, with the volume of PBS adjusted to be consistent across all groups. Images were taken at 24 and 48 h to observe the wound healing process. The scratch area was quantified using ImageJ software. Migration area (%)= (A₀-A_n)/A₀ × 100. A₀ represents the initial wound area (0 h), and A_n represents the residual area of the wound at the time of measurement.

db/db mice model of wound healing

A study protocol, including the research question, key design features, and analysis plan, was prepared before the experiment to ensure methodological consistency and adherence to ethical guidelines. All procedures were conducted following the approved guidelines of the animal experimentation committee of Taipei Veterans General Hospital and adhered to the “Guide for the Care and Use of Laboratory Animals” by the National Academy of Sciences of the United States. All experiments were operated in Taipei Veterans General Hospital Animal Laboratory. Nine-week-old male db/db mice (C57BLKS/J Iar -+Lepr^db/+Lepr^db; National Laboratory Animal Center, Taipei, Taiwan) were used in the study. The average weight of the mice was approximately 20–30 g. The db/db mouse model is a well-established genetic model of type 2 diabetes, characterized by a mutation in the Lepr gene, leading to defective leptin signaling. The genetic predisposition to diabetes and impaired wound healing, making it a relevant model for chronic wound studies. Mice were required to be in good health at the time of the experiment, without signs of infection, severe weight loss, or abnormal behavior that could interfere with wound healing. Mice that exhibited severe complications, such as uncontrolled infections, excessive distress, or failure to recover from anesthesia, were excluded from the study. Additionally, any data points with technical errors, such as improper injection, wound closure failure, or significant deviations from experimental protocols, were excluded from the final analysis. These criteria were established a priori to ensure the reliability of the data and the ethical treatment of the animals. All mice included in the study completed the experiment, and no animals, experimental units, or data points were excluded from the analysis. All samples were processed according to the established protocol, and no technical issues or complications met the predefined exclusion criteria.

The study included five experimental groups: (1) PBS, (2) normoxic adult ADSC-Exos, (3) hypoxic adult ADSC-Exos, (4) normoxic infant ADSC-Exos and (5) hypoxic infant ADSC-Exos. Mice were randomly assigned to five experimental groups to ensure unbiased allocation. Randomization was performed using a simple randomization method, where each mouse was assigned to a group using a random number generator. This approach minimized potential confounding factors and ensured balanced distribution across treatment and control groups. During the allocation phase, the researchers responsible for randomizing the mice were aware of group assignments. However, to minimize bias, the individuals conducting the experiment (e.g., administering injections, performing wound care) were partially blinded to group allocation, as exosome treatments and PBS were prepared and coded by a separate researcher. For outcome assessment, wound measurements and imaging were performed in a blinded manner, where the evaluator was unaware of the treatment groups. Similarly, histological and molecular analyses were conducted with coded samples to ensure unbiased interpretation. Blinding was not applied during data analysis, as group identities were known to the researchers performing statistical analyses. All mice were housed under identical environmental conditions, with cages placed in randomized positions within the facility to prevent location-based variability. Treatments and measurements were performed in a randomized order to avoid systematic bias. Additionally, all procedures, including wound creation, injections, and imaging, were conducted by the same trained researchers to ensure consistency. Despite these efforts, some potential confounders, such as individual variability in wound healing, could not be entirely eliminated.

Before the surgery, mice were anesthetized with 5% isoflurane (Attane; Panion & BF Biotech Inc., Taiwan) for 1.5–2 min to induce anesthesia. Isoflurane, which allows for rapid induction and recovery, could maintain anesthesia for up to 12 h. Once the mice lost consciousness and ceased movement, an anesthesia mask was placed over the mouse’s head, and 2% isoflurane was administered. Surgery commenced after confirming the absence of eyelid reflexes and pain responses. The surgical site was shaved, disinfected with alcohol, and a 6-mm punch biopsy (Kai Medical, Tokyo, Japen) was used to create a wound at the level of the mouse’s spine by pressing and rotating the tool. The excised skin was lifted with forceps and removed with scissors. A circular silicone splint (inner diameter 9 mm) was placed to prevent wound contraction and secured with fast-drying glue and 5 -0 nylon sutures. Intraoperative wound care included covering the wound with a SI-Mesh dressing, followed by a transparent waterproof film to cover the entire wound. Postoperatively, the wound was carefully sutured and disinfected. Mice were housed in accordance with the animal center’s guidelines, with bedding changed every two days and daily observations of postoperative and experimental conditions. They were maintained in specific pathogen-free (SPF) conditions at a controlled temperature (22–24 °C) and humidity (40–60%), with a 12-hour light-dark cycle. Mice were acclimated to their environment for at least one week before experiments.

On day 0, using a 25G needle, subcutaneous injections were administered at four sites around the wound with either 200 µL of PBS containing 100 µg/ml normoxic adult ADSCs-Exos (n = 4), hypoxic adult ADSCs-Exos (n = 4), normoxic infant ADSC-Exos (n = 4), hypoxic infant ADSCs-Exos (n = 4), or PBS alone (n = 4). Considering statistical power, feasibility, and ethical guidelines. A total of 20 mice were used, with four animals per group, to ensure sufficient data for meaningful comparisons. Each cage housed four to five mice, and ear notching was used for identification. To minimize the pain from wound creation, only one wound was made on each mouse. Wounds were opened and photographed on days 0, 3, 7, 10, and 14 to assess wound area reduction over time. The primary outcome was used to determine the sample size, ensuring sufficient statistical power to detect differences between treatment groups. The effect size (Cohen’s d) was calculated to assess the magnitude of differences between groups. The 95% confidence intervals (CI) for Cohen’s d were also reported. On day 14, mice were sacrificed by CO₂ asphyxiation in a chamber, in accordance with the Institutional Animal Care and Use Committee (IACUC) guidelines of Taipei Veterans General Hospital. No chemical anesthetics were used for euthanasia. Skin specimens (~ 12 × 12 mm) were then collected and fixed in either 4% paraformaldehyde (PFA) for 30 min or in Zinc Fixative (BD Biosciences) for 1–2 days, followed by dehydration and paraffin embedding. Tissue sections (4–5 μm thick) were prepared for hematoxylin and eosin (HE) staining and Masson’s trichrome staining to evaluate tissue regeneration and collagen deposition.

This study has been conducted and reported in accordance with the Animal Research: Reporting of In Vivo Experiments (ARRIVE) guidelines 2.0 to ensure transparency, reproducibility, and ethical standards in animal research.

Hematoxylin-eosin (HE) staining

After deparaffinization of the paraffin sections, the nuclei were stained with hematoxylin (Sigma-Aldrich) for 2.5 to 5 min, and the staining was monitored under a microscope. Once the nuclei appeared blue-purple, the sections were rinsed under running tap water for 5 min. The sections were then briefly immersed in distilled water, the excess water was removed by gently shaking the slides, and the cytoplasm was stained with eosin (Sigma-Aldrich) for 7 s. This was followed by another 5-minute rinse under running tap water and a briefly immersion in distilled water. After drying at 45 °C, the staining was observed under a microscope to confirm that the nuclei appeared blue and the cytoplasm pink. Finally, the slides were mounted using Micromount mounting medium (Leica, Nussloch, Germany) and examined using an Olympus BX43 microscope (Tokyo, Japan).

Masson’s trichrome staining

After deparaffinizing the paraffin sections, Masson’s trichrome staining was performed using the Trichrome Stain Kit (Fine Test Biotech, Wuhan, China) following the manufacturer’s instructions. The sections were first stained with iron hematoxylin solution for 5 min. Acid alcohol differentiation solution was then applied for 10 s. After differentiation, the sections were immersed in bluing solution for 5 min and rinsed with distilled water. Next, the sections were stained with ponceau-acid fuchsin solution for 10 min, followed by a rinse with distilled water. To differentiate the collagen, the sections were treated with phosphomolybdic acid solution for 1–2 min or until the collagen no longer appeared red, typically between 10 and 15 min. Aniline blue solution was applied for 2 min to stain the collagen. Subsequently, the sections were immersed in acetic acid solution for 1 min. The sections were quickly dehydrated in 95% ethanol for 2–3 s, followed by dehydration in absolute ethanol for 5–10 s. This step was repeated three times. The sections were then cleared in xylene, mounted using micromount mounting medium (Leica), and observed under an Olympus BX43 microscope (Tokyo, Japan). The thickness of regenerated dermis was determined by Image-Pro PLUS software (Media Cybernetics, Silver Spring, MD, USA).

Library preparation and small RNA sequencing

A total of 100 ng of RNA was utilized to prepare small RNA libraries using the QIAseq^® miRNA Library Kit (QIAGEN, Hilden, Germany) as per the manufacturer’s instructions. Briefly, specific 3’ and 5’ adaptors were ligated to the respective ends of the small RNA molecules. First-strand cDNA synthesis was carried out with QIAseq miRNA NGS RT Enzyme and an RT primer. After PCR amplification, the amplified products were size-selected to isolate fragments of 170–200 bp using QIAseq beads. Library quality was assessed using the Qsep400 system. Libraries meeting quality criteria were then sequenced on the Illumina NovaSeq 6000 platform, using 75 bp single-end reads. Sequencing was performed by Genomics, BioSci & Tech Co., New Taipei City, Taiwan.

Bioinformatics analysis of sequencing data

The raw sequencing data underwent adapter trimming using TrimGalore! (v0.6.6) to remove unwanted sequences. Cleaned reads were then mapped to the reference genome, as well as mature and hairpin miRNAs from miRBase (v.22.1v) [27] using Bowtie (v1.3.0) [28] for accurate identification of miRNA reads. Post-alignment, samtools (v1.12) [29] was used to process the BAM files, followed by the calculation and normalization of miRNA expression levels via edgeR (v3.26.5) [30]. Differentially expressed miRNAs (DEmiRNAs) were identified using DEGSeq (v1.48.0) [31]. To predict miRNA targets, three different algorithms—miRanda, TargetScan, and PITA—were employed. The predicted targets were then subjected to Gene Ontology (GO) [32] and Kyoto Encyclopedia of Genes and Genomes (KEGG) [33] functional enrichment analysis using the R package clusterProfiler (v3.14.0) [34, 35].

Statistical analysis

Quantitative analysis was performed using Prism (version 5.03, GraphPad, La Jolla, California, USA). Data were presented as mean ± standard error. Statistical significance was assessed using the Mann-Whitney U test for comparisons between two groups, and two-way analysis of variance (ANOVA) for comparisons among multiple groups. Results were considered significant when the p-value was less than 0.05.

Results

Characterization of exosomes derived from ADSCs

The morphology of adult and infant ADSC cells at the same passage and timepoint was observed under normoxic and hypoxic culture conditions (Fig. 1A–D). Under the microscope, all ADSCs exhibited the typical spindle-shaped morphology. Exosomes isolated from ADSCs appeared round and displayed a bilayer vesicular structure under TEM examination (Fig. 1E–H). Further NTA analysis showed size differences among exosomes secreted by the four ADSC groups (Fig. 1I–L).

Fig. 1.

Characterization of adult and infant adipose derived stem cell-exosomes (ADSC-Exos) under hypoxic and normoxic conditions. Phase-contrast microscopy was used to image the cellular morphology of adult ADSCs expanded under normoxic (Nor) (A) and hypoxic (Hyp) (B) conditions, and infant ADSCs expanded under normoxic (C) and hypoxic (D) conditions (magnification ×40; scale bar = 500 μm). Transmission electron microscopy was used to image isolated exosomes derived from normoxic adult ADSCs E, hypoxic adult ADSCs (F), normoxic infant ADSCs (G), and hypoxic infant ADSCs (H) (magnification ×30,000; scale bar = 100 nm). Nanoparticle tracking analysis was performed to determine the particle size distribution of normoxic adult ADSC-Exos (I), hypoxic adult ADSC-Exos (J), normoxic infant ADSC-Exos (K), and hypoxic infant ADSC-Exos (L). M Quantitative analysis of the mean and mode particle sizes from these four groups is shown. Data are presented as mean ± SD from 3–5 experimental replicates. Statistical significance between groups was determined using one-way ANOVA, where “*” represents p < 0.05. N The cropped blots showed the presence of exosomal markers, including CD81 and CD63, as detected by Western blot analysis.O Dil-labeled exosomes were incubated with human dermal fibroblasts (HDFs) for 24 h and visualized under an Olympus BX43 microscope. The upper panels showed images at ×200 magnification with scale bars of 100 μm. The lower panels presented enlarged views at ×400 magnification with scale bars of 50 μm. Yellow arrows indicated the Dil-labeled exosomes (red) localized around DAPI-stained nuclei (blue)

In Fig. 1M, quantitative analysis from 3 to 5 independent experiments, analyzed using one-way ANOVA, indicated that exosomes derived from adult ADSCs were significantly smaller than those from infant ADSCs under normoxic conditions (p < 0.05). For hypoxic adult ADSC-Exos, the mean size was 92.22 ± 3.69 nm, and the mode was 63.14 ± 5.19 nm. In the normoxic adult ADSC group, the mean size was 96.86 ± 3.21 nm, and the mode was 69.4 ± 8.51 nm. Both groups’ exosomes were smaller than those of the infant ADSC-Exos. The hypoxic infant ADSCs had a mean size of 98.5 ± 0.6 nm and a mode of 64.6 ± 2.13 nm, while the normoxic infant ADSCs exhibited significantly larger exosomes, with a mean size of 106.13 ± 5.28 nm and a mode of 79.25 ± 22.18 nm. The larger mean particle size of normoxic infant ADSC-Exos indicates a broader particle size distribution, meaning there is greater size variability among exosomes, potentially including a higher proportion of larger particles.

Tetraspanins CD63 and CD81, widely recognized as biomarkers for exosomes, are abundantly expressed in these vesicles [36]. To confirm their presence, we performed Western blot analysis of the exosomal membrane proteins CD63 and CD81. The results showed positive expression in all exosome groups (Fig. 1N). Full-length blots are provided in Supplementary Figure S1, confirming the successful isolation of exosomes from ADSCs. The uptake of ADSC-Exos by HDFs was also confirmed, with Dil-labeled ADSC-Exos from all four groups showing localization around the nuclei of HDFs (Fig. 1O).

Identification of miRNAs in hypoxic and normoxic ADSC-Exos

The ADSC-Exos related miRNA were identified and compared to evaluate the potentials of wound healing. The top 100 miRNA screening criteria were |log2(FC)|>2.5 (FC: fold change) and P < 0.001, and cluster heatmap analysis revealed significantly differentially expressed miRNAs in the hypoxia- and normoxia-expanded adult and infant ADSC-Exos groups (Fig. 2A).

Fig. 2.

Micro RNA (miRNA) expression in adult and infant adipose derived stem cell-exosomes (ADSC-Exos) under hypoxic and normoxic conditions. A Heatmap showing the top 100 differentially expressed miRNAs in 3 independent samples of adult and infant ADSC-Exos under normoxic and hypoxic conditions. B, C Bar plots representing the results of Gene Ontology (GO) over-representation analysis of biological functions in hypoxia- and normoxia-expanded adult ADSC-Exos (B) and infant ADSC-Exos (C). D, E Dot plot showing enrichment of signaling pathways based on Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis in exosomes from hypoxia- and normoxia-expanded adult ADSCs (D) and infant ADSCs (E)

The functional annotations of miRNA-targeted genes were evaluated to assess significantly enriched KEGG pathways and GO terms. The differentially expressed miRNAs were mainly associated with GO biological processes such as stem cell population maintenance, cell number maintenance, and lymphangiogenesis-biological functions that are relevant to wound healing and upregulated in both hypoxic adult and infant ADSC-Exos groups (Fig. 2B and C). Rather than differentiating directly, ADSC-derived exosomes contribute to tissue repair by modulating the local microenvironment and regulating the activity of resident or recruited stem cells. Through the delivery of bioactive molecule such as miRNAs, these exosomes can promote the maintain of stem cell populations, support cell survival and proliferation, and facilitate lymphatic vessel formation—all of which are critical for effective wound healing process [37]. These findings highlight the paracrine and regulatory roles of exosomes in orchestrating complex repair process.

All significantly affected KEGG pathways in the hypoxic adult and infant ADSC-Exos groups (Fig. 2D and E) were identified by comparing signaling pathways involved in wound healing. Among them, the chemokine signaling pathway, mTOR signaling pathway, VEGF signaling pathway, focal adhesion, and leukocyte transendothelial migration were linked to promoting regeneration during wound healing. These results suggest that ADSC-Exos may participate in tissue regeneration by regulating immune cell recruitment, enhancing angiogenesis, and facilitating cell adhesion and migration at the wound site—emphasizing their functional contribution to key processes involved in wound repair.

Proliferation, in vitro migration and wound healing

The normoxic infant ADSC-Exos showed the greatest enhancement in HDF proliferation under high glucose condition, with cell viability in the normoxic infant ADSC-Exos group being significantly higher than in the control, normoxic adult ADSC-Exos, and hypoxic infant ADSC-Exos groups at 24 h (Fig. 3A). However, no significant differences were observed between the groups after 48 h. The results of the Transwell assay, observing cells that migrated to the lower side in response to exosome treatments (Fig. 3B and C), were quantified using ImageJ software (Fig. 3D). The data showed an overall trend of enhanced cell migration in all exosome-treated groups compared to the control. Specifically, the normoxic infant ADSC-Exos group demonstrated a significantly greater ability to promote cell migration compared to the control group, as shown in Fig. 3D.

Fig. 3.

Proliferation-enhancing potential and in vitro migration assay of human dermal fibroblasts (HDFs) treated with adult and infant adipose derived stem cell-exosomes (ADSC-Exos) under hypoxic and normoxic conditions in high glucose medium. A Cell proliferation of HDFs in high glucose medium treated with 100 µg/ml normoxic (Nor) adult ADSC-Exos, hypoxic (Hyp) adult ADSC-Exos, normoxic infant ADSC-Exos, hypoxic infant ADSC-Exos, and PBS (control) for 24 and 48 h was determined by CCK-8 analysis. Mean ± SD with three experimental replicates were expressed. Statistical significance between groups was determined by two-way ANOVA. B A Transwell migration assay was performed on human dermal fibroblasts (HDFs) in high-glucose medium treated with 100 µg/ml normoxic adult ADSC-Exos, hypoxic adult ADSC-Exos, normoxic infant ADSC-Exos, hypoxic infant ADSC-Exos, and PBS (control) for 12 h. HDFs were suspended in high-glucose medium supplemented with 5% FBS in the upper chamber, while the lower chamber contained high-glucose medium supplemented with 10% FBS, with or without exosomes. After 12 h, cells that had migrated to the lower side of the membrane were fixed and stained with 0.4% crystal violet. C The morphology of migrated cells was imaged using a Nikon Eclipse TS100 microscope (magnification ×100; scale bar = 25 μm). D Migrated cells were quantified using ImageJ software, and relative counts were normalized to the control group. Data are presented as mean ± SD from three independent experiments. Statistical significance between groups was determined using one-way ANOVA; “*” indicates p < 0.05

The hypoxic adult ADSC-Exos demonstrated the most effective in vitro wound healing (Fig. 4A). Quantification of the wound closure rate revealed that at 24 h, the hypoxic adult ADSC-Exos-treated group had significantly higher wound closure compared to the control group. At 48 h, wound closure in the hypoxic adult ADSC-Exos group was significantly greater than in the control, normoxic adult ADSC-Exos, and hypoxic infant ADSC-Exos groups. Additionally, the normoxic infant ADSC-Exos group showed significantly higher wound closure than the control group at 48 h (Fig. 4B). These findings suggest that normoxic infant ADSC-Exos may possess a greater capacity to promote HDF proliferation and transwell (vertical) migration, whereas hypoxic adult ADSC-Exos appear to enhanced lateral (horizontal) migration in wound healing assays.

Fig. 4.

In vitro wound healing of human dermal fibroblasts (HDFs) in high glucose condition treated with normoxic and hypoxic adult and infant adipose derived stem cell-exosomes (ADSC-Exos). A In vitro wound healing assay of HDFs in high glucose medium treated with 100 µg/ml normoxic adult ADSC-Exos, hypoxic adult ADSC-Exos, normoxic infant ADSC-Exos, hypoxic infant ADSC-Exos, and PBS (control) for 24 and 48 h was imaged. (magnification ×40; scale bar = 500 μm) B Wound closure (%) was calculated as (A0 - An) / A0 × 100, where A0 represents the initial wound area (0 h) and An represents the residual wound area at the time of measurement. Data are expressed as Mean ± SD from 3–5 experimental replicates. Statistical significance between groups was determined by two-way ANOVA, where “*” indicates p < 0.05, “**” indicates p < 0.01, and “***” indicates p < 0.0001

In vivo db/db mice model wound healing

In the animal experiments, a single full-thickness wound was created on the back of each db/db mouse to minimize pain from the procedure. A circular silicone splint was applied over the wound to prevent contraction. Wound conditions were monitored on days 0, 3, 7, 10, and 14 (Fig. 5A), and wound sizes were measured and expressed as percentages of wound closure (Fig. 5B). In this study, a healed wound was defined as one exhibiting 100% closure of the wound area. By day 14, some mice in the treatment groups achieved complete wound closure, while others showed substantial improvement. Among all groups, the PBS control group exhibited the slowest wound closure. In contrast, the normoxic adult ADSC-Exos group demonstrated the fastest healing response, with the highest wound closure percentage observed on day 7 (69.3% ± 17.3, mean ± SD), which was significantly higher than that of the control group (26.2% ± 6.3). The hypoxic adult ADSC-Exos (63.7% ± 29.2) and hypoxic infant ADSC-Exos groups (58% ± 21.4) groups also achieved significantly greater wound closure than the control group on day 7. Notably, the normoxic infant ADSC-Exos group exhibited a slower wound healing rate, showing significantly improved wound closure compared to the control only on day 10 (91.0% ± 10.0 vs. 61.0% ± 37.6, p < 0.05). The effect sizes (Cohen’s d) and 95% confidence intervals for comparisons between treatment groups and the control are summarized in Supplementary Table S1. On Day 3, the normoxic adult ADSC-Exos group demonstrated a large effect size compared to the control group (Cohen’s d = 8.68, 95% CI: − 0.08, 17.43), while the hypoxic adult ADSC-Exos group showed a moderate effect size (d = 2.50, 95% CI: − 0.28, 5.28). On Day 7, the effect sizes remained substantial, with the hypoxic adult ADSC-Exos group exhibiting a strong effect (d = 4.60, 95% CI: − 0.16, 9.35), followed by the hypoxic infant ADSC-Exos group (d = 2.01, 95% CI: − 0.34, 4.36). By Day 10 and Day 14, both adult and infant ADSC-Exos treatments continued to show moderate to large effect sizes, with Cohen’s d values ranging from 1.56 to 4.24. However, some confidence intervals were broad, suggesting variability in treatment responses. Collectively, these results indicate that normoxic adult ADSC-Exos may accelerate early-stage wound healing, while hypoxic infant ADSC-Exos tend to promote a delayed yet sustained healing effect. This temporal distinction highlights how donor age and culture conditions influence the kinetics and efficacy of exosome-based wound therapies.

Fig. 5.

Evaluation of diabetic wound healing in db/db mice model treated with adult and infant adipose derived stem cell-exosomes (ADSC-Exos) under hypoxic and normoxic conditions. A A wound was created on the back of each db/db mouse using a 6 mm punch biopsy (Kai Medical, Tokyo, Japan). To prevent wound contraction, a circular silicone splint with an inner diameter of 9 mm was placed over the wound. The entire wound was then covered with an SI-Mesh dressing. Postoperatively, the wound was carefully sutured and disinfected. On day 0, subcutaneous injections were administered at four sites surrounding the wound, delivering either 200 µL of PBS containing 100 µg/mL normoxic (Nor) adult ADSC-derived exosomes (n = 4), hypoxic (Hyp) adult ADSC-derived exosomes (n = 4), normoxic infant ADSC-derived exosomes (n = 4), hypoxic infant ADSC-derived exosomes (n = 4), or PBS alone (Control, n = 4). Wounds were photographed on days 0, 3, 7, 10, and 14. B The wound sizes were analyzed, and the percentage of wound closure was calculated. Data are presented as the mean ± SD from four independent experiments. Statistical significance between groups was determined using two-way ANOVA, where “*” indicates p < 0.05, “**” indicates p < 0.01, and “***” indicates p < 0.001

The wound skin samples from the back of each db/db mouse were collected on day 14 and sectioned for analysis. HE and Masson’s trichrome staining were used to visualize and differentiate the distribution of collagen proteins in the dermal layer (Fig. 6A). After 14 days of recovery, the average thickness of the regenerated dermis was measured. The mean thicknesses for the control group, normoxic adult ADSC-Exos group, hypoxic adult ADSC-Exos group, normoxic infant ADSC-Exos group, and hypoxic infant ADSC-Exos group were 10.9 μm, 57 μm, 73.86 μm, 64.86 μm, and 82.41 μm, respectively. However, only the hypoxic adult ADSC-Exos group showed a significant difference compared to the control group (Fig. 6B). These findings suggest that hypoxic adult ADSC-Exos may provide enhanced promotion of dermal tissue regeneration, as evidenced by the significantly increased thickness of the regenerated dermis compared to the control group.

Fig. 6.

Histological analysis of wound healing in db/db mice treated with adult and infant adipose derived stem cell-exosomes (ADSC-Exos) under hypoxic and normoxic conditions. A Skin wound sections from the backs of db/db mice were collected on day 14 post-treatment and stained with Hematoxylin-Eosin (HE) and Masson ‘s trichrome. A representative image of normal, uninjured db/db mouse skin was shown for reference and tissue structure comparison; this group was not included in quantitative analysis. The regenerated dermis is indicated by yellow double arrows and quantitatively measured. Scale bar = 100 µm. B The average thickness of regenerated dermis was measured using Image-Pro PLUS software, with data presented as the mean ± SD from four independent samples. Statistical significance between groups was assessed using one-way ANOVA, with “*” indicates p < 0.05

Discussion

ADSC-Exos were successfully isolated from both adult and infant ADSCs under normoxic and hypoxic conditions, as confirmed by morphological observations, TEM imaging, NTA analysis, and Western blot detection of exosomal markers CD63 and CD81. Exosomal size analysis revealed that normoxic infant ADSC-Exos had a broader size distribution and a larger mean particle size compared to other groups, potentially reflecting a higher proportion of larger particles. Cellular internalization and perinuclear localization were demonstrated by the uptake of Dil-labeled ADSC-Exos by HDFs.

The heterogeneity in the sizes of normoxic infant ADSC-Exos, characterized by a larger mean particle size but no significant difference in mode compared to other groups, suggests a broader particle size distribution. This may indicate a higher concentration of bioactive substances, such as proteins and nucleic acids, potentially contributing to their enhanced capacity to promote HDF proliferation. Larger exosomes are known to facilitate cellular uptake and influence functional outcomes in target cells [38]. However, since the mode remains similar across groups, the predominant particle size is comparable, suggesting that size alone does not fully account for the proliferation-promoting effects. Instead, the functional impact of exosomes is likely depending on their specific molecular cargo rather than size alone. While a broader size distribution may imply increase cargo capacity, enhanced functionality is dependent on the enrichment of bioactive molecules [39]. Therefore, the composition of exosomal contents, such as specific proteins or RNAs concentrations, in conjunction with particle size heterogeneity, likely plays a pivotal role in promoting cell proliferation.

In in vitro experiments, normoxic infant ADSC-Exos showed the greatest enhancement of HDF proliferation at 24 h, although this effect was not sustained at 48 h. They also significantly promoted wound closure at 48 h compared to the control and demonstrated the most pronounced effect on Transwell migration, indicating an enhanced capacity to promote vertical cell movement. Conversely, hypoxic adult ADSC-Exos exhibited the most effective in vitro wound closure, with significantly greater wound closure at both 24 and 48 h compared to other groups. The larger mean particle size of normoxic infant ADSC-Exos suggests their potential suitability for applications requiring sustained proliferation or cellular activation. The comparable mode sizes indicate similar permeability and distribution profiles to those of other groups, supporting their potential use across diverse tissue or cellular targets.

Several upregulated miRNAs enclosed within the exosomes derived from hypoxic adult and infant ADSCs targeted genes involved in the pathways related to wound healing, such as the chemokine signaling pathway [40], mTOR signaling pathway [41], VEGF signaling pathway [42], focal adhesion [43], and leukocyte transendothelial migration [44]. Additionally, miRNAs associated with neuron development pathways—such as those regulating neuron projection development, telencephalon development, and positive regulation of neurogenesis—were significantly enriched in hypoxic ADSC-Exos compared to their normoxic counterparts. These findings suggest that hypoxia enhances the neuronal regeneration potential of ADSC-Exos. However, this enrichment in neuron-related miRNA did not translate into a superior wound healing effect in vitro, particularly for exosomes derived from infant ADSCs, suggesting that hypoxia may selectively favor neural over dermal regenerative pathways.

In in vivo experiments, normoxic adult ADSC-Exos achieved the fastest wound closure, followed by hypoxic adult ADSC-Exos, hypoxic infant ADSC-Exos, and normoxic infant ADSC-Exos on day 7. Interestingly, by day 10, only normoxic infant ADSC-Exos group demonstrated significantly higher wound closure than control group. This suggests that adult ADSC-Exos primarily enhance early-stage wound closure, while provided infant ADSC-Exos may contribute more to later-stage healing. Although direct comparisons between adult and infant ADSC-Exos in wound healing are limited, according to in vitro assays, hypoxic adult ADSC-Exos significantly enhanced lateral migration in scratch assays, a process closely linked to early-stage wound closure, were keratinocytes and fibroblasts migrate across the wound bed [45, 46]. In contrast, normoxic infant ADSC-Exos more effectively promoted vertical migration in Transwell assays, which may correspond to deeper tissue remodeling and dermal infiltration occurring at later stages of wound healing [45]. While further validation is needed, this observation suggests that the differential effects of ADSC-Exos on vertical and lateral cell migration may underline their distinct temporal contributions to wound healing in vivo. The effect size analysis further supports the significant impact of ADSC-Exos treatments on wound healing. The large Cohen’s d values observed, particularly in the hypoxic adult ADSC-Exos and hypoxic infant ADSC-Exos groups, suggest a strong therapeutic effect. However, the relatively wide confidence intervals indicate potential variability, highlighting the need for further studies with larger sample sizes to confirm these findings.

Previous studies have noted that wound closure does not always correlate with dermal regeneration [23]. In diabetes, ulcerogenic cells in wound can inhibit healing, leading to a thicker epidermis and cornified layer [47]. Our results showed reduced dermis regeneration in the control group, but all ADSC-Exos treatments enhanced dermal thickness. Hypoxia preconditioning significantly improved the regenerative effects of adult ADSC-Exos as evidenced by thicker regenerated dermis in diabetic mice. Although both normoxic and hypoxic infant ADSC groups showed a trend toward increased dermal thickness, the large standard deviation observed in those groups suggest high variability among samples, which may have limited the ability to achieve statistical significance. Nonetheless, given the conserved nature of exosome-mediated signaling mechanisms, these results offer insights applicable to other preclinical models and support future investigations with larger sample sizes to further evaluate their therapeutic relevance in humans.

Infant ADSCs inherently exhibit superior proliferation, anti-senescence, antioxidative abilities, and differentiation potential compared to adult ADSCs [8]. Although the cited study did not directly evaluate the function of exosomes, it is plausible that such enhanced stemness of infant ADSCs may be partially reflected in the bioactive composition of their secreted exosomes. Unlike adult ADSCs, which often require hypoxic stimulation to improve the therapeutic efficacy of their exosomes, the intrinsic cellular advantages of infant ADSCs could potentially explain why their exosomes remain functionally competent under normoxic conditions. Nevertheless, further studies are necessary to confirm whether these cellular-level benefits consistently translate into enhanced exosomal activity and therapeutic efficacy.

This study has several limitations. Particle size heterogeneity may influence functionality, including cellular uptake efficiency, biodistribution, and the release kinetics of bioactive contents [48]. Future studies should examine how particle size affects exosome transport, permeability, and content release refine their therapeutic applications. In animal experiment, high variability and limited sample sizes reduced the ability to detect significant differences between groups, underscoring the need for larger cohorts or more controlled conditions. Given the small sample size (n = 4 per group), effect sizes (Cohen’s d) and confidence intervals were reported to provide an estimate of the magnitude and variability of the treatment effects (Supplementary Table S1). While statistical significance may not always be achieved, the observed effect sizes suggest potential biological relevance. In the context of human biology, differences in immune responses, skin structure, and wound healing dynamics between mice and humans necessitate further validation in larger animal models and clinical trials. Additionally, variations in exosome uptake and function under different pathological conditions should be considered when extrapolating these findings to diverse clinical scenarios. Furthermore, the underlying mechanisms and contributing factors were not fully elucidated in this study. Although KEGG pathway analysis of exosomal miRNA was performed, specific gene-level changes in recipient cells following treatment were not directly assessed. Moreover, the composition and function of ADSC-derived exosomes may be influenced by donor variability, environmental preconditioning (e.g., hypoxia vs., normoxia), and tissue origin (e.g. visceral vs. subcutaneous fat). These biological and physiological variables may contribute to heterogeneity in therapeutic outcomes and should be further investigated in future studies.

Conclusion

Hypoxic preconditioning enhanced the wound-healing effects of adult ADSC-Exos, while normoxic infant ADSC-Exos already exhibited superior regenerative capacity without hypoxic stimulation. These findings suggest that the therapeutic benefits of hypoxia may be more pronounced in adult ADSC-Exos. Additionally, the healing mechanisms may differ, with adult ADSC-Exos promoting lateral migration and early wound closure, and infant ADSC-Exos supporting vertical migration and deeper tissue remodeling.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

ADSCs

Adipose-derived mesenchymal stem cells

Exos

Exosomes

TEM

Transmission electron microscopy

NTA

Nanoparticle tracking analysis

HDF

Human dermal fibroblast

USP22

Ubiquitin-specific peptidase 22

HIF-1α

Hypoxia-inducible factor 1-alpha

IRB

Institutional Review Board

SVF

Stromal-vascular fraction

DMEM

Dulbecco’s Modified Eagle Medium

MWCO

Molecular weight cut-off

PBS

Phosphate-buffered saline

HBSS

Hank’s balanced salt solution

HG-DMEM

High glucose Dulbecco’s Modified Eagle Medium

FBS

fetal bovine serum

SDS

Sodium dodecyl sulfate

PVDF

Polyvinylidene fluoride

HRP

horseradish peroxidase

DAPI

4,6-diamidino-2-phenylindole

CCK-8

cell counting kit-8

RNA

Ribonucleic acid

miRNA

Micro ribonucleic acid

HE

Hematoxylin-eosin

NGS

Next-generation sequencing

DEmiRNAs

Differentially expressed miRNAs

GO

Gene Ontology

KEGG

Kyoto Encyclopedia of Genes and Genomes

SPF

Specific pathogen-free

IACUC

Institutional Animal Care and Use Committee

ARRIVE

Animal Research: Reporting of In Vivo Experiments

ANOVA

One-way analysis of variance

Author contributions

Conception and design of study: Szu-Hsien Wu, Kuang-Kai Hsueh, and Jung-Pan Wang. Acquisition of data and/or analysis and interpretation of data: Yu-Ting Liao, Zi-Xuan Tao and Ming-Te Cheng. Funding acquisition: Szu-Hsien Wu and Kuang-Kai Hsueh. And all authors participated in drafting the article or revising it critically for important intellectual content, and gave final approval of the version to be submitted and any revised version.

Funding

This study was partially supported by grants from the Taipei Veterans General Hospital (V113C-118), the Ministry of Science and Technology (MOST 111-2314-B-075-057-MY3), the Childhood Burn Foundation of The Republic of China (for the years 2023 and 2024), and the Taoyuan General Hospital, Ministry of Health and Welfare (PTH112030).

Data availability

All data generated or analyzed during this study are presented within this published article. The miRNA sequencing data supporting the findings of this study are provided as supplementary files (see Appendix).

Declarations

Ethics approval and consent to participate

The protocol title for human ethics is " Comparison of therapeutic effects of exosomes derived from human adipose-derived mesenchymal stem cell from hypoxic and normoxic culture on wound healing in impaired db/db mice and the mechanism”. The collection of human ADSCs was approved by the IRB of the Taipei Veterans General Hospital, with approval number 2021-02-012B. This approval was granted on February 28, 2021. Fat tissue was obtained with informed consent from participants or, in the case of children under 16, from their parent or legal guardian. The animal ethics protocol, titled “Comparison of Therapeutic Effects of Exosomes Derived from Human Adipose-Derived Mesenchymal Stem Cells from Hypoxic and Normoxic Cultures on Wound Healing in Impaired db/db Mice and the Mechanism,” was approved by the IACUC of Taipei Veterans General Hospital. The approval (IACUC No. 2023 − 160) was granted on July 18, 2024. HDFs were purchased from ScienCell, which confirmed that the collection of human cells had received ethical approval and that donors provided informed consent. This declaration is outlined on their website (https://sciencellonline.com/en/technical-support/ethical-statement/?srsltid=AfmBOorE7-c3OTL0o0ioQxlprNeagvV-BjZOb0h8PncVALYFYE3y2zCU).

Consent for publication

This manuscript does not contain any individual person’s data in any form (including individual details, images, or videos) that requires consent for publication.

Competing interests

Szu-Hsien Wu received funding from the Taipei Veterans General Hospital, the Ministry of Science and Technology and the Childhood Burn Foundation of the Republic of China to support the experimental procedures. Kuang-Kai Hsueh received funding from the Taoyuan General Hospital, Ministry of Health and Welfare to partially support the miRNA analysis.