iPSC-derived exosomes promote diabetic wound healing by attenuating inflammatory responses

Abstract

Background

Owing to impaired glucose metabolism, the high-glucose microenvironment in diabetic patients disrupts a series of biological reactions that hinder the wound healing process, resulting in a significant cost to the health care system and an urgent need for new and advanced therapies.

Methods

In this study, induced pluripotent stem cell-derived exosomes (iPSC-Exos) were isolated from iPSC culture supernatant via centrifugation and ultrafiltration. We evaluated the therapeutic effects of iPSC-Exos on diabetic wound healing through two clinically relevant animal models (spontaneous genetic diabetic mouse model and streptozotocin (STZ)-induced diabetic mouse model). iPSC-Exos were topically administered to full-thickness cutaneous wounds in diabetic mice. The therapeutic effects were systematically assessed by measuring wound closure rates, conducting comprehensive histopathological evaluations, and performing quantitative analysis of inflammatory mediators via ELISA.

Results

We demonstrated that iPSC-Exos can significantly accelerate diabetic wound healing through two clinically relevant animal models (spontaneous genetic diabetic mouse model and STZ-induced diabetic mouse model) for the first time. The multifaceted therapeutic mechanisms include: (ⅰ) Direct activation of tissue regeneration (promotion of re-epithelialization, tissue remodeling and scar attenuation); (ⅱ) Modulation of the inflammatory microenvironment (promoting macrophage polarization toward anti-inflammatory M2 phenotype/suppressing inflammation).

Conclusions

This dual-animal model approach, which closely recapitulates key pathophysiological features of human diabetic wounds, offers superior clinical translatability compared to single-animal model studies. Our findings not only establish a robust scientific foundation for clinical development of iPSC-Exos therapy, but also present a transformative, cell-free treatment paradigm for chronic diabetic wounds that addresses critical unmet medical needs.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13287-026-05005-5.

Introduction

Diabetes mellitus is a metabolic disease characterized by a relative or absolute lack of insulin secretion leading to chronically increased blood glucose levels [1–3]. The core mechanism of diabetic wound healing impairment is governed by a vicious cycle, driven by hyperglycemia and involving multiple interconnected components. This cycle, initiated by systemic pathophysiological disturbances, propagates through the disruption of the local microenvironment, and ultimately leads to a halt in cellular repair processes. Chronic hyperglycemia, as the initial driving factor, exerts its effects through two primary pathways: On one hand, it activates the polyol, hexosamine, and protein kinase C (PKC) pathways, triggering oxidative stress and accumulating advanced glycation end products (AGEs). This cascade further damages nerves and blood vessels: neuropathy impairs wound sensory perception and disrupts the skin barrier, increasing the risk of occult injuries [4–6]; meanwhile, angiopathy leads to insufficient blood perfusion, laying the groundwork for subsequent hypoxia and ischemia [7, 8]. On the other hand, hyperglycemia directly impairs the chemotactic and phagocytic capabilities of immune cells, reducing the efficiency of bacterial clearance and creating favorable conditions for infection. Against this backdrop, the wound microenvironment deteriorates further: hyperglycemia upregulates the expression of pro-inflammatory factors through epigenetic modifications, promoting the sustained polarization of M1-type macrophages. Concurrently, it delays the clearance of neutrophils, forming an “inflammation-oxidative stress” cycle that induces senescence in repair cells such as fibroblasts and endothelial cells [9]. Angiopathy, combined with the high oxygen consumption of inflammatory cells, exacerbates hypoxia; however, hyperglycemia downregulates the hypoxia-inducible factor-1 (HIF-1)/vascular endothelial growth factor (VEGF) axis, inhibiting angiogenesis. Ischemia and hypoxia further worsen energy supply, hindering the proliferative phase of wound healing [7, 8, 10]. Additionally, the hyperglycemic environment facilitates bacterial colonization and biofilm formation, which stimulate the secretion of pro-inflammatory factors and exogenous proteases; this, together with the imbalance between upregulated expression of endogenous matrix metalloproteinases (MMPs) and downregulated expression of tissue inhibitors of metalloproteinases (TIMPs), collectively degrades the extracellular matrix (ECM) and growth factors, leading to tissue necrosis [11]. Additionally, chronic inflammation and infection also render the wound microenvironment alkaline (elevated pH), which further enhances MMP activity and inhibits fibroblast function [12–14], forming a self-perpetuating “microenvironment deterioration loop“ [7, 8]. Eventually, the wound remains stagnant in the inflammatory phase for an extended period, unable to progress through the normal proliferative and remodeling phases of healing, thus developing into a chronic non-healing wound [15]. Traditional wound treatments, such as local dressings, blood glucose control, skin grafting, and laser therapy, have shown limited therapeutic efficacy and may carry various risks, including poor adherence to blood glucose control, donor-site damage and hyperpigmentation issues [16, 17]. Given the serious health problems associated with diabetic skin injuries and the limitations of existing treatments, the development of new and more effective technologies and strategies is of significant clinical importance for effectively promoting diabetic wound healing in light of current health trends.

Stem cells are a group of undifferentiated cells (including adult stem cells and embryonic stem/induced pluripotent stem cells) with the ability to self-renew and rebuild functional tissues [18–20]. In recent years, stem cell therapy has garnered significant attention due to its remarkable capabilities in tissue repair and regeneration. However, its clinical application faces numerous challenges, including immune rejection, potential risk of tumorigenesis, as well as ethical controversies [21, 22]. With further research, emerging studies have shown that the ability of stem cells to repair tissue damage is dependent not only on their migration and differentiation abilities but also on their strong paracrine abilities [23, 24]. Exosomes, as an important component of stem cell paracrine secretion, have gradually become a research hotspot in cell-free therapeutic strategies. Exosomes are nanoscale vesicles secreted by cells, rich in bioactive molecules such as lipids, proteins, and nucleic acids [25, 26]. They play a critical role in tissue repair and regeneration by regulating the phenotype and function of recipient cells.

Exosomes derived from induced pluripotent stem cells (iPSCs) have garnered significant attention due to their unique advantages. iPSCs can be derived from somatic cells with the ability to differentiate into any cell type, possessing unlimited self-renewal and multipotent differentiation capabilities [27–30]. Moreover, they can be generated from cells of patients themselves, thereby avoiding immune rejection [31]. Studies have shown that iPSC-derived exosomes not only inherit the regenerative and anti-inflammatory properties of iPSCs [32], but also exhibit excellent stability, low immunogenicity, and the potential for large-scale production. These characteristics make them ideal candidates for the treatment of diabetic wound healing. Therefore, as an innovative cell-free therapeutic strategy, the use of iPSC-derived exosomes avoids the potential risk of tumorigenesis and ethical controversies associated with cell therapy, representing a safer therapeutic approach that extends stem cell research from cell-based therapy to exosome-based therapy. However, the specific mechanisms by which iPSC-derived exosomes promote diabetic wound healing, particularly their role in modulating inflammatory responses, still require further in-depth investigation.

In this study, we investigated the role and mechanisms of iPSC-derived exosomes (iPSC-Exos) in diabetic wound healing. We established two diabetic mouse models with full-thickness cutaneous wounds: a spontaneous genetic diabetic mouse model and an streptozotocin (STZ)-induced diabetic wound model. The effects of iPSC-Exos on wound healing were evaluated, with a focus on their roles in extracellular matrix remodeling, promotion of angiogenesis, and modulation of the wound immune microenvironment.

Materials and methods

Culture and characterization of iPSCs

The iPSCs were prepared by Guidon Pharmaceutics (Beijing, China). Briefly, iPSCs were cultured in TeSR-AOF medium (Stemcell Technologies) on plates coated with vitronectin (Thermo Fisher Scientific) and the medium was changed daily. The cells were passaged using Versene Solution (Thermo Fisher Scientific) at a ratio of 1:4 − 1:6 every 4–5 day as clumps regularly to maintain cell viability and undifferentiated status.

The morphology of iPSC clone clusters and individual cells was observed using EVOS™ imaging system (EVOS^TMXL core, Thermo). The pluripotency marker TRA-1–60 was quantitatively analyzed by flow cytometry system (CYTEK Biosciences, USA). Alkaline phosphatase (AP) staining was performed with BCIP/NBT Alkaline Phosphatase Colour Development Kit (Beyotime) according to manufacturer’s instructions.

The biomarker proteins of iPSCs, including NANOG, SSEA4, SOX2 and OCT4 were evaluated by immunocytochemistry analyses. Cells were fixed using a 4% paraformaldehyde for 15 min before washing three times with PBS. The cells were permeabilized with 0.1% Triton X-100 (Solarbio, China) for 15 min at room temperature. The cells were washed twice with PBS, then blocked with 2% BSA solution for 1 h in the dark. Subsequently, the cells were incubated overnight at 4℃ with primary antibodies against OCT4 (653702; Biolegend, China), SSEA4 (4755 S; Cell Signaling Technology, USA), Nanog (4903 S; Cell Signaling Technology, USA), and SOX2 (3579 S; Cell Signaling Technology, USA), each diluted at 1:200. The following day, the cells were incubated with the corresponding secondary antibodies for 2 h in the dark at room temperature. After two additional PBS washes, the cells were counterstained with Hoechst (1:1000 dilution) for 10 min in the dark.

Isolation and characterization of iPSC-Exos

iPSC-Exos were isolated from iPSC culture supernatants by centrifugation and ultrafiltration. Briefly, when the coverage of iPSCs reached 80%, the culture medium was substituted daily with GDEV medium (Guidon Pharmaceutics), and the culture supernatant was collected after 48 h. The supernatant of iPSC cell culture was centrifuged at 300 g for 10 min to remove cells and large cellular debris. The supernatant was then filtered using 0.22 μm syringe filters to eliminate larger particles and impurities. Subsequently, the filtered supernatant was filtered and concentrated with Amicon Ultra-15 Centrifugal Filters (Ultracel-100 kDa, Merck Millipore, Burlington, MA, USA). The liquid obtained from the previous step was resuspended by adding phosphate-buffered saline (PBS) to remove any residual medium components and other small molecules. The isolated iPSC-Exos were aliquoted and stored at −80 °C in PBS to preserve their stability and bioactivity until use.

The morphology of iPSC-Exos was examined using transmission electron microscopy (TEM) (JEM-1200EX, JEOL, Japan). The diameter was analyzed using a Flow NanoAnalyzer (NanoFCM U30, China). Exosomal marker proteins, including Alix, TSG101, CD9, CD63, and Calnexin, were detected by Western blot analysis. The protein concentration of iPSC-Exos was quantified using a BCA protein assay kit (23225, Thermo Fisher, USA).

Animal study

Male diabetic mice (C57BLKS/J-leprdb/leprdb, db/db, 11–12 weeks of age) were purchased from Beijing Vital River Laboratory Animal Technology Co., Ltd. (Beijing, China), C57BL/6J male mice (4 weeks of age) were purchased from SiPeiFu Biotechnology Co., Ltd. (Beijing, China), and all animals were maintained under specific pathogen-free (SPF) conditions with controlled environmental parameters: ambient temperature of 20–26 °C, relative humidity of 40–70%, and a 12-hour light/dark cycle. This study has been reported in line with the ARRIVE guidelines 2.0. The experimental unit for wound healing assessment was defined as individual wounds, with two full-thickness wounds created per mouse dorsum. Consequently, the total sample size (n = X wounds) represents twice the number of animals (n = X/2 mice). The mice were coded and randomly divided into experimental and control groups with a random number generator. All cages were randomly assigned to different positions on the rearing rack and were systematically rotated during weekly cage changes to minimize potential positional effects. Fasting blood glucose measurements were performed during the same time window each day. The order of handling and testing for each group of mice was randomized each day to avoid biases introduced by diurnal fluctuations and order effects. Only the staff that performed the experiments were aware of the allocation groups. Sample size was based on literature review and compliance with the 3Rs principles. Postoperatively, mice received subcutaneous buprenorphine (0.1 mg/kg) once on the day of surgery and twice on the following day (postoperative day 2). After completion of the experiment, mice were euthanized by cervical dislocation performed by trained personnel, in compliance with the guidelines of the Institutional Animal Care and Use Committee (IACUC) of the Chinese Academy of Medical Sciences and School of Basic Medicine (Approval number: ACUC-XMSB-2024-069). Based on animal welfare principles, we have established predefined criteria for humane endpoints and euthanasia to alleviate the suffering of laboratory animals. The humane endpoint criteria are provided in the Supplementary Materials and Methods. If an animal exhibits one or more of the signs, it will be humanely euthanized.

Wound healing assessment in diabetic model mice

To evaluate the wound healing efficacy in a spontaneous genetic diabetic mouse model in vivo, we established a full-thickness cutaneous skin wound model. Ten db/db mice were randomly divided into the NS group (control group, treated with normal saline, n = 10 wounds from 5 mice) and the EXO group (treated with iPSC-Exos, n = 10 wounds from 5 mice) with a random number generator. Before establishing the dorsal wound model and during the experiment, fasting blood glucose levels were measured in C57BL/KsJ-^db+/db+ mice to confirm their diabetic status. Mice with fasting blood glucose levels exceeding 16.7 mmol/L were included in the study [33]. Animals that meet this standard are subsequently included in the experiment. They were anesthetized in a chamber (RWD Life Science Co., Ltd., China) containing 1.5% isoflurane and 60% oxygen, and their hair on the dorsal region were shaved. A disposable biopsy punch was used to create two full-thickness cutaneous skin wounds (8-mm diameter) on the dorsum of each mouse. Then, 25 µL of normal saline (control group) or iPSC-Exos (containing 10 µg of protein) were placed on the wound on Day 0, 1, 2, 3, 4, 5, 7 and 10 postoperative. Finally, the wounds were covered with Tegaderm™ Film (3 M Inc., USA) and bandages to avoid dehydration due to water evaporation and to prevent biting of the sample site. The dressings were refreshed after the wound was exposed for treatment and photography. The wounds were photographed, and blood flow perfusion was measured by a laser speckle flow imaging system (RFLSI III, RWD Life Science Co., Ltd., China) on postoperative Day 0, 3, 7 and 14. Blinded measurements of wound size were performed using ImageJ software (NIH, USA). The percentage of the original wound area was calculated via the following equation:

where S_current and S_initial are the calculated areas of each wound on Day 0 and Day n, respectively.

To evaluate the efficacy of iPSC-Exos in achieving complete recovery in diabetic mice and determining the total time required, we established an induced diabetic mouse model [34]. A total of 20 mice were allocated to groups randomly using a random number generator, including 5 in the control group (negative control) and 15 in the diabetic modeling group. Under specific pathogen-free conditions, the 5 control mice were fed a standard diet, while the 15 diabetic modeling mice were continuously fed a high-fat diet for one month. Then the 5 control mice received daily intraperitoneal injections of sodium citrate buffer for 5 consecutive days as a control treatment. For the 15 diabetic modeling mice, STZ (S8050, Solarbio) in citric acid‒sodium citrate buffer once a day at a dosage of 50 mg/kg/day was continuously injected intraperitoneally into the mice for 5 days to induce diabetes. Two weeks after STZ injection, the mice whose blood glucose levels stably exceeded 16.7 mmol/L were considered diabetic [33]. To evaluate the wound healing effect, C57BL/6J diabetic mice were anesthetized with isoflurane, and 6-mm diameter full-thickness cutaneous wounds were created on both sides of the back via a skin biopsy punch on the shaved and sterilized dorsum. The diabetic mice were then randomly divided into two groups with a random number generator, and each wound was given 25 µL of normal saline (positive control group, D-NS,) or exosomes (D-EXO, containing 10 µg of protein) every day. In addition, round normal wounds without diabetes were used as the normal control group (C-NS) and were given 25 µL of normal saline (D-NS). Wound images were captured on postoperative Day 0, 3, 7, 14, and 21, followed by imaging at 2-day intervals from Day 21 onward until complete wound healing was achieved in all groups. The relative wound area was calculated using ImageJ software. The other operations were the same as those described above.

To research the mechanism by which iPSC-Exos facilitate wound healing in diabetes, a total of 10 mice were allocated to groups randomly using a random number generator, including 3 in the control group and 7 in the diabetic modeling group. The diabetic mice were then randomly divided into two groups with a random number generator, and each wound (6-mm diameter) was given 25 µL of normal saline (positive control group, D-NS,) or exosomes (D-EXO, containing 10 µg of protein) on Day 0, 1, 2, 3, 4, 5, 7, 10 and photographed on postoperative Day 0, 3, 7, 10 and 14. The mice were sacrificed, and the wounds (including surrounding tissue) were harvested on Day 14. One half of the tissue samples were fixed in 4% paraformaldehyde (PFA), embedded in paraffin, and subsequently subjected to histological analyses, including Hematoxylin and Eosin (H&E) staining, Masson’s Trichrome staining, and immunofluorescence (IF) analysis. The other half were homogenized for protein extraction to measure the total protein content and the expression levels of pro-inflammatory cytokines, namely TNF-α, IL-6, and IL-1β.

The calculations for the percentage of re-epithelialization, the epithelial thickness index (ETI), and the scar elevation index (SEI) were performed according to established methodologies from previous work [35]. The specific procedures were as follows:

The percentage of re-epithelialization was quantified by measuring the length of the newly regenerated epithelium and expressing it as a percentage of the total distance between the original wound margins. The epithelial thickness index (ETI) was calculated to assess the maturity of the neo-epithelium (only wounds that achieve complete re epithelialization will be included in the analysis), derived from measurements at the wound center (P₂) and adjacent uninjured skin (P₁). The scar elevation index (SEI) was used to evaluate scar hypertrophy by comparing the total cross-sectional area of the wound site to the area of the underlying normal dermis. The formulas for the indices are as follows:

where ‘a’ and ‘b’ are the lengths of the newly regenerated epithelium from both wound edges and ‘c’ is the total wound gap length.

where P₁ is the epidermal thickness of the normal adjacent skin and P₂ is the maximum epidermal thickness within the regenerated wound area.

where P₁ is the dermal thickness of the normal adjacent skin and P₂ is the maximum dermal thickness within the central wound area.

Histological staining

For histological immunofluorescence analyses, the wound tissues were fixed with 4% paraformaldehyde solution, dehydrated with gradient ethanol and embedded in paraffin. The sections were cut into slices and subjected to H&E staining, which was performed via a standard protocol and Masson staining (according to the Masson’s Trichrome Stain Kit standard protocol of Solarbio). For immunofluorescence analyses, after rinsing in PBS, the sections were sealed with goat serum at room temperature for 60 min. The sections were then incubated with primary antibodies against α-SMA (19245; Cell Signaling Technology, USA), a CD206 polyclonal antibody (18704-1-AP; Proteintech, China), a purified anti-mouse F4/80 antibody (123101; Biolegend, China) and a CD86 polyclonal antibody (13395-1-AP; Proteintech, China) overnight at 4 °C and then inoculated with the corresponding secondary antibodies. The nuclei were stained with DAPI. Quantification of the staining was performed using ImageJ software (NIH, USA).

ELISA analysis of inflammatory cytokines

The extraction of tissue proteins was carried out according to the following specific steps: the harvested tissue was added to cold PBS containing a mixture of protease inhibitors at a ratio of 1 g of tissue to 9 mL of lysate and then placed in a tissue homogenizer at 60 Hz for 10 s. The procedure was repeated thirty times. The resulting tissue homogenate was centrifuged at 16,000 g for 10 min at 4 °C and the upper supernatant was aspirated. The total protein concentration in the tissues was measured via a Pierce™ BCA protein assay kit (23225, Thermo Fisher, USA). The expression of the relevant inflammatory factors TNF-α (DY410-05, R&D Systems, China), IL-6 (DY406-05, R&D Systems, China) and IL-1β (DY401-05, R&D Systems, China) was measured via ELISA kits according to the manufacturer’s instructions.

Western blot analysis

Exosomes were quantified via a Micro BCA™ protein assay kit (23235, Thermo Fisher, USA). Equal amounts of protein samples were loaded onto a 10% SDS-PAGE gel for separation and then transferred to a nitrocellulose (NC) membrane. The membrane was incubated with 5% skim milk powder solution (D8340, Solarbio, China) at room temperature for 1 h and then incubated overnight at 4 °C with primary antibodies against ALIX (ab186429, Abcam, USA), TSG101 (ab125011, Abcam, USA), CD9 (20597-1-AP, Proteintech, China), CD63 (52090 S, Cell Signaling Technology, USA), and calnexin (2679, Cell Signaling Technology, USA) and further incubated with the appropriate horseradish peroxidase (HRP)-conjugated secondary goat anti-rabbit IgG-HRP antibodies (M21002S, Abmart, China) for 1 h at room temperature. The blots were visualized using a chemiluminescence imaging system (ChemiScope 6100, Clinx, China) after SuperSignal™ West Pico Luminol/Enhancer Solution (34580, Thermo Fisher, USA) was added to the NC film.

Statistical analysis

Statistical analysis was performed using GraphPad Prism software. All statistical tests were assessed for assumptions of normality and homogeneity of variances. For data satisfying parametric assumptions, a two-tailed Student’s t-test was employed for comparisons between two groups, and one-way analysis of variance (ANOVA) followed by post hoc Bonferroni correction was applied for comparisons among multiple groups. If the data violate the normality assumption, nonparametric statistical tests are applied. p < 0.05 was taken as statistically significant; *p < 0.05, **p < 0.01, ***p < 0.001. Data from at least three individual experiments are listed as mean ± standard deviation (mean ± SD).

Results

Isolation and characterization of iPSC-Exos

iPSC-Exos were isolated from iPSC culture supernatants by centrifugation and ultrafiltration (Fig. 1a). Microscope images showed that the clone clusters vary in size, with clear edges and tight intercellular connections, formed by the tight aggregation of multiple iPSCs. iPSCs were oval in shape with clear cell boundaries (Fig. 1b). Immunocytochemistry analyses revealed strong expression of key pluripotency markers, including NANOG, SSEA4, SOX2, and OCT4 (Fig. 1c). Alkaline phosphatase (AP) staining further demonstrated high AP activity in iPSCs (Fig. 1d). Flow cytometry analyses revealed strong expression of the surface biomarker TRA-1–60 (Fig. 1e). These findings collectively confirmed that the iPSCs maintain an undifferentiated state and retain robust pluripotent characteristics. To identify the exosomes derived from iPSCs, we employed TEM, western blotting, and nanoflow analysis. TEM imaging revealed that the iPSC-Exos exhibited the typical “saucer-like” morphology of exosomes, with a complete bilayer membrane structure (Fig. 1f). Western blot analysis further exhibited the presence of exosomal markers, including Alix, TSG101, CD9, and CD63, as well as the negative marker calnexin (Fig. 1g). Full-length gels are provided in Supplementary Fig. 1. Size distribution analysis using a Flow NanoAnalyzer (NanoFCM, Xiamen, China) showed that iPSC-Exos predominantly ranged in diameter from 40 to 80 nm (Fig. 1h). Collectively, these results demonstrated the successful harvesting and characterization of iPSC-Exos.

Fig. 1.

Characterization of iPSCs and iPSC-Exos. a Schematic representation of iPSCs and iPSC-Exos preparation. b Optical microscope images of iPSCs. c Flow cytometry of iPSCs biomarkers. d AP staining of iPSCs. e Flow cytometry analyses of the surface biomarker TRA-1–60. f TEM analysis of iPSC-Exos. g Western blotting analysis of iPSC-Exos surface markers. h Nanoflow cytometry for analyzing the size distribution of iPSC-Exos

Wound healing assessment in a spontaneous genetic diabetic mouse model

To evaluate the in vivo efficacy of iPSC-Exos in diabetic wound healing, we used db/db mice. This model lacks the leptin receptor and exhibits characteristics similar to human type 2 diabetes. A full-thickness cutaneous wound model was established in these mice to mimic the wound healing process observed in diabetic patients. Throughout the experimental period, after excluding mice that were humanely euthanized due to persistent hyperglycemia, the db/db mice included in the final data analysis maintained stable and severe hyperglycemia (blood glucose > 20 mmol/L; see Supplementary Table) during the entire study. During the intervention period, three mice in the normal saline (NS) group (control group) and one mouse in the exosome-treated (EXO) group were humanely euthanized due to persistent hyperglycemia. Therefore, the results of the control group with 2 mice (n = 4 wounds from 2 mice) and the EXO group with 4 mice (n = 8 wounds from 4 mice) were included in the data analysis. As illustrated in Fig. 2a, two full-thickness cutaneous wounds with a diameter of 8 mm were created on the backs of each mouse using a round punch, followed by the local application of normal saline (NS) or iPSC-Exos (EXO) on Day 0, 1, 2, 3, 4, 5, 7 and 10. The wounds were monitored to analyze the therapeutic effects of iPSC-Exos on diabetic wound healing (Fig. 2b). Compared with the NS group, the EXO group promoted wound healing more strongly at Day 7 and 14 (Fig. 2c). Angiogenesis was monitored using a laser speckle flow imaging system, which revealed a time-dependent in blood flow from Day 0 to Day 14 as the wounds healed (Fig. 2d). Nevertheless, analysis of blood flow perfusion values (Fig. 2e) indicated that the EXO group presented greater blood perfusion during the proliferative phase of angiogenesis compared to the NS group, suggesting the potential of iPSC-Exos to promote vascularization.

Fig. 2.

iPSC-Exos promoted cutaneous wound healing in a spontaneous diabetes model (2 mice were assigned to control group (n = 4), and 4 mice to EXO group (n = 8)). a Schematic diagram of wound healing after different treatments. b Representative digital images of the wound healing process. c Quantification of the wound area relative to that on Day 0. d Laser Doppler image of the dorsal skin and representative images of the different groups during the wound healing process. e Relative blood flow perfusion of wounds to that of Day 0. Error bars were defined as the standard deviation and significance levels are indicated as follows: *p < 0.05; **p < 0.01; ***p < 0.001

Time-course analysis of wound healing in an STZ-induced type 2 diabetic mouse model

In the human-like full-thickness cutaneous wound model using db/db mice, we confirmed that iPSC-Exos significantly promote wound healing in diabetic mice. To further validate these findings, we employed the STZ-induced diabetic mouse model, a widely used system for studying diabetic wound healing (Fig. 3a). Diabetic conditions were induced in C57BL/6J mice by intraperitoneal injection of STZ for 5 consecutive days. Full-thickness cutaneous wounds, 6 mm in diameter, were created using a round punch and subsequently treated. Figure 3b shows the fasting blood glucose levels of the control group and successfully induced diabetic mice. Before STZ injection (Pre-D0), all experimental animals exhibited fasting blood glucose levels within the normal range. Approximately one week after STZ administration, 12 out of the 15 diabetic modeling mice showed a significant increase in blood glucose levels exceeding 20 mmol/L, confirming the successful induction of diabetes. Therefore, among the 20 mice, the control group (negative control) was selected into 5 mice (n = 10), and the diabetic modeling group was selected into 12 mice (6 mice to each D-NS and D-EXO group (n = 12)). As shown in Fig. 3c-e, nondiabetic healthy control mice (C-NS) exhibited faster wound healing compared to diabetic control mice at all postoperative time points, further validating the successful establishment of diabetic wound model. Notably, diabetic mice treated with iPSC-Exos (D-EXO) achieved complete wound healing 8 days earlier than diabetic mice treated with normal saline (D-NS), demonstrating the potential of iPSC-Exos to shorten the time required for full recovery.

Fig. 3.

iPSC-Exos accelerated the complete healing time of diabetic wounds in vivo (5 mice were assigned to C-NS group (n = 10), and 6 mice to each D-NS and D-EXO group (n = 12)). a Schematic diagram of the time points for establishing the model and performing the treatments. b Fasting blood glucose levels in mice before the full-thickness cutaneous wound model was created after the intraperitoneal injection of STZ. c Representative digital images of the wound healing process. d Quantification of the wound area relative to that on Day 0. e Complete healing time of the different groups. Error bars were defined as the standard deviation and significance levels are indicated as follows: *p < 0.05; **p < 0.01; ***p < 0.001

Wound healing assessment in an STZ-induced type 2 diabetes mouse model

Using both the human-like full-thickness cutaneous wound model in db/db mice and the STZ-induced diabetic C57BL/6J mouse model, we confirmed that iPSC-Exos significantly accelerate wound healing in diabetic mice. To further explore the underlying mechanisms by which iPSC-Exos promote diabetic wound healing, we employed the STZ-induced diabetic mouse model (Fig. 4a). In the diabetic modeling group, 6 out of the 7 mice exhibited a significant increase in fasting blood glucose levels exceeding 20 mmol/L approximately one week after STZ injection. Therefore, among the 10 mice, the control group (negative control) was selected into 3 mice (n = 6), and the diabetic modeling group was selected into 6 mice (3 mice to each D-NS and D-EXO group (n = 6)). Fasting blood glucose levels of the enrolled mice are shown in Fig. 4b, confirming the stability of the diabetic model. Representative images and quantitative analysis of wound closure during the healing process (Fig. 4c, d) revealed that the percentage of the original wound area in the iPSC-Exos-treated diabetic group (D-EXO) was significantly reduced compared to the saline-treated diabetic group (D-NS) and nearly reached levels comparable to the nondiabetic control group (C-NS) (Fig. 4e). These findings suggested that iPSC-Exos play a crucial role in accelerating the healing of diabetic wounds.

Fig. 4.

iPSC-Exos enhanced cutaneous wound healing in an in vivo mouse model (3 mice per group with 6 wounds (n = 6)). a Schematic diagram of the time points for establishing the model and performing the treatments. b Fasting blood glucose levels in mice before the full-thickness cutaneous wound model was created after the intraperitoneal injection of STZ. c Representative digital images of the wound healing process. d Schematic diagram of the wound healing of different treatment groups. e Quantification of the wound area relative to that on Day 0. Error bars were defined as the standard deviation and significance levels are indicated as follows: *p < 0.05; **p < 0.01; ***p < 0.001

Enhanced re-epithelialization and tissue remodeling by exosome-based therapy for superior wound healing with scar attenuation

The healing of a wound progresses through a series of continuous and overlapping stages [36]. Following initial hemostasis and inflammatory responses, the wound transforms into a proliferative phase, during which it becomes enveloped by new skin and blood vessels. As fresh granulation tissue begins to fill the damaged area, the structural integrity of the skin is incrementally restored. Once the granulation tissue matures into connective tissue, the wound progresses to the final remodeling phase. To thoroughly evaluate wound healing, this study employed re-epithelialization percentage, Epithelial Thickness Index (ETI), Scar Elevation Index (SEI), and granulation tissue gap as key metrics, complemented by in-depth histological staining and immunofluorescence analysis (Fig. 5). Representative histological sections illustrating the measurement methods for re-epithelialization percentage, ETI, and SEI are presented in Fig. 5a, while Fig. 5b depicts the method for measuring granulation tissue gaps across groups. Re-epithelialization is crucial for establishing an early functional barrier, which helps prevent excessive water loss and protects against infection [37, 38]. On postoperative day 14, the proportions of completely re-epithelialized wounds in the control, exosome (D-Exo), and PBS (D-NS) groups were 50%, 66.7%, and 50%, respectively, with corresponding overall mean re-epithelialization percentages of 74%, 86%, and 50% (Fig. 5c). Furthermore, the granulation tissue gap in the D-Exo group was significantly smaller than that in the D-NS group (Fig. 5d). As standardized quantitative indicators for assessing the microscopic quality of wound healing, the ETI results showed that all groups had ETI values exceeding 100%, but the D-NS group exhibited a significantly higher ETI than both the control and D-Exo groups (Fig. 5e). The SEI results demonstrated that compared to the D-NS group, the D-Exo group had a significantly reduced SEI, approaching levels observed in normal tissue (Fig. 5f). Together, these findings indicate that iPSC-Exos effectively promote wound re-epithelialization and granulation tissue formation while reducing the risk of scarring, ultimately optimizing healing quality.

Fig. 5.

iPSC-Exo promoted wound healing and reduce scar formation by activating tissue regeneration (3 mice per group with 6 wounds (n = 6)). Representative histology sections illustrating the key morphometric parameters of a Percentage of re-epithelization, epithelial thickness index, scar elevation index and b Granulation tissue gaps. c Percentage of re-epithelization. d Granulation tissue gaps. e Epithelial thickness index. f Scar elevation index. g Masson’s trichrome-stained section images of C-NS, D-NS and D-EXO, reflecting collagen deposition. h Collagen deposition density. i α-SMA (red) immunofluorescence staining image reflecting neovascularization in the wound tissue. j α-SMA fluorescence intensity. Error bars were defined as the standard deviation and significance levels are indicated as follows: *p < 0.05; **p < 0.01; ***p < 0.001

During the wound healing phase, the deposition and remodeling of collagen facilitate tissue repair and regeneration. Adequate collagen deposition and remodeling can improve the tensile strength of the tissue, leading to more effective healing outcomes [39]. Masson’s trichrome staining revealed sparse collagen deposition in the wounds of the D-NS group. In contrast, the D-Exo group showed significantly increased deposition of newly synthesized collagen, forming large, wavy collagen fibers organized into a matrix resembling that of the non-diabetic C-NS group (Fig. 5g, h). This demonstrates that iPSC-Exos promote collagen deposition and remodeling, contributing to enhanced tissue tensile strength.

Angiogenesis is a critical process in tissue healing, as blood vessels deliver oxygen and nutrients to cells in and around the wound [40]. To evaluate the condition of mature vessels post-treatment, α-smooth muscle actin (α-SMA) expression was assessed via immunofluorescence staining on postoperative Day 14 (Fig. 5i). Immunofluorescence staining for α-smooth muscle actin (α-SMA) revealed higher expression in the D-Exo group compared to the D-NS group (Fig. 5j), indicating the promotion of mature vessel formation. This ensures adequate delivery of oxygen and nutrients to cells within the wound, supporting the healing process.

In summary, in the diabetic mouse model, iPSC-Exos significantly enhanced wound healing outcomes by promoting re-epithelialization, granulation tissue regeneration, and angiogenesis, increasing collagen deposition and remodeling, while simultaneously inhibiting scar formation.

iPSC-Exos-mediated attenuation of inflammatory responses in in vivo wound healing

Wound healing typically progresses through the phases of hemostasis, inflammation, proliferation, and remodeling [41]. Nevertheless, in diabetic wounds, this orderly sequence is disrupted, often characterized by prolonged low-grade inflammation that hampers or halts the healing process. Specifically, macrophages play an important role in regulating inflammation and facilitating wound repair by differentiating into either the M1 (inflammatory) phenotype or the M2 (anti-inflammatory) phenotype. However, diabetes causes a dysfunctional macrophage response and impaired phenotypic transition from M1 to M2 [42]. In the diabetic wound model, D-EXOs exhibited a pro-healing effect, potentially due to the role of iPSC-Exos in facilitating the shift in wound healing from the inflammatory phase to the proliferative phase. Consequently, the impact of iPSC-Exos on macrophage polarization was further explored by examining the expression of M1 macrophages and M2 macrophages to evaluate the extent of inflammation modulation in diabetic wounds following various treatments. F4/80 was chosen as a general macrophage marker, while CD86 and CD206 served as specific markers for M1 and M2 macrophages, respectively. Compared with the wounds of nondiabetic healthy control mice (C-NS group), the wounds of diabetic mice treated with normal saline (D-NS group) presented an increased proportion of M1 macrophages and a decreased proportion of M2 macrophages, indicating that the wounds of the D-NS group were still in a prolonged inflammatory state on Day 14, which hindered the wound healing process. Compared with those in the D-NS group, treatment with iPSC-Exos (D-EXO group) significantly reduced the proportion of M1 macrophages (CD86) and increased the number of M2 macrophages (CD206), reaching overall M1 and M2 macrophage levels comparable to those in the wounds of nondiabetic healthy control mice (Fig. 6a-d).

Fig. 6.

iPSC-Exos regulated inflammation in vivo (3 mice per group with 6 wounds (n = 6)). a Images of F4/80 (green), M1 phenotype macrophage (CD86, red) and nuclei (DAPI, blue) immunostaining in wound tissue. b Representative quantitative immunofluorescence analysis of M1 macrophages. c Images of F4/80 (green), M2 phenotype macrophage (CD206, red) and nuclei (DAPI, blue) immunostaining in wound tissue. d Representative quantitative immunofluorescence analysis of M2 macrophages. Quantitative analysis of the levels of the inflammation-associated cytokines e TNF-α, f IL6, and g IL1β in tissue wounds via ELISA. Error bars were defined as the standard deviation and significance levels are indicated as follows: *p < 0.05; **p < 0.01; ***p < 0.001

To further validate the transition to an effective healing phase following iPSC-Exos treatment, the cytokine levels in the wound tissues were also measured by enzyme-linked immunosorbent assay (ELISA). Compared with those in the D-NS group, the expression levels of proinflammatory factors, such as TNF-α, IL-6, and IL-1β, in diabetic wounds were lower in the D-EXO group (Fig. 6e-g).

These findings indicate that iPSC-Exos can modulate the inflammatory microenvironment within wound tissue by reducing the expression of inflammatory cytokines and promoting the polarization of macrophages toward the M2 anti-inflammatory phenotype. This dual action facilitates the transition from inflammation to proliferation, thereby enhancing the overall wound healing process in diabetic conditions.

Discussion

As mentioned in the Introduction of this study, our objective was to investigate the reparative effects and mechanisms of iPSC-Exos on diabetic skin injuries. The present study demonstrates the therapeutic potential of iPSC-Exos in promoting diabetic wound healing through multiple mechanisms, including the promotion of re-epithelialization, enhancement of tissue remodeling, attenuation of scarring, and modulation of inflammation (Fig. 7). Our findings provide compelling evidence that iPSC-Exos can effectively address the impaired healing process in diabetic wounds.

Fig. 7.

Mechanisms of iPSC-Exos in promoting diabetic wound healing. Exosomes promote wound healing by facilitating re-epithelialization, granulation tissue formation, and angiogenesis, as well as by increasing collagen deposition while simultaneously inhibiting scar formation

The fundamental process of wound healing typically begins with hemostasis and inflammation, subsequently enters a proliferative phase characterized by granulation tissue formation to fill the defect [43, 44] and re-epithelialization to establish a protective barrier, and ultimately culminates in a remodeling phase that restores skin structure and function. Collagen, a major component of the extracellular matrix, is critical for tissue repair, particularly during the remodeling phase. Its deposition and organization are key determinants of skin strength and appearance [39]. The histological analysis of wound tissues supports the regenerative potential of iPSC-Exos. Treatment with iPSC-Exos dramatically improved wound healing outcomes by promoting re-epithelialization, enhancing tissue remodeling, and reducing scar formation. The significant reduction in wound area and the earlier completion of healing in iPSC-Exos-treated diabetic mice highlight their ability to promote tissue regeneration and repair. Studies by Teng and Lee et al. have demonstrated that exosomes derived from human umbilical cord mesenchymal stem cells and adipose-derived mesenchymal stem cells can accelerate diabetic wound healing by promoting re-epithelialization and enhancing collagen deposition [45, 46]. Our research aligns with their findings and further expands this perspective by demonstrating the efficacy of iPSC-Exos in treating diabetic skin injuries. The presence of well-organized collagen fibers in the D-EXO group, resembling those in non-diabetic controls, further underscores the restorative capacity of iPSC-Exos.

Angiogenesis is a pivotal process in wound healing, as it ensures the delivery of oxygen and nutrients to the wound site. Poor angiogenesis is a key factor for chronic wound healing in diabetes, so enhancing angiogenesis and blood supply reconstruction in diabetic wounds is crucial for accelerating chronic wound healing [47]. Studies by Teng, Lee, et al. have found that exosomes derived from human umbilical cord mesenchymal stem cells and adipose-derived mesenchymal stem cells can promote angiogenesis and accelerate diabetic wound healing. Research by Li, et al. has found that enhancing blood perfusion can promote wound healing [45, 46, 48]. Our data revealed that iPSC-Exos significantly enhanced blood perfusion and promoted the formation of mature blood vessels, as evidenced by increased α-SMA expression. This pro-angiogenic effect is particularly important in diabetic wounds, where impaired vascularization is a major contributor to delayed healing. The observed increase in blood flow and vessel maturation in iPSC-Exos-treated wounds suggested that iPSC-Exos can not only effectively promote vascular maturation but also improve blood perfusion by reestablishing vascular function, thereby accelerating healing.

The important role of the immune microenvironment in wound healing should also be considered. Increasing evidence shows that the key to accelerating wound healing in diabetes is to avoid excessive inflammation, including the secretion of inflammatory factors and inappropriate differentiation of macrophages in chronic wounds [49–51]. Macrophage polarization plays an important role in wound healing in diabetes [51, 52]. The classically activated macrophages (M1) and alternatively activated macrophages (M2) are two typical phenotypes of macrophages [53]. M1 macrophages are characterized by the production of proinflammatory cytokines, such as IL-1β and TNF-α [54]. The subsequent inflammatory response may lead to organ dysfunction, whereas M2 macrophages can secrete cytokines related to anti-inflammatory effects, thereby reducing the inflammatory response [55]. If macrophages cannot effectively transform from M1 macrophages to M2 macrophages, nonhealing wounds can occur [56]. Additionally, studies have confirmed that exosomes derived from mesenchymal stem cells can promote the polarization of macrophages toward the M2 phenotype and reduce inflammatory responses, thereby facilitating wound healing [45, 46, 52, 57]. We detected many M1 macrophages and fewer M2 macrophages in the D-NS group, revealing a highly inflammatory environment in the wounds, which was attributed to high glucose impairing macrophage phenotypic shifts, leading to an abnormally large accumulation of M1 macrophages. In contrast, due to the anti-inflammatory effect of iPSC-Exos, the number of M1 macrophages in the wounds of the D-EXO group was significantly reduced, and the number of M2 macrophages was increased. This shift is critical for transitioning from the inflammatory phase to the proliferative phase of wound healing, demonstrating that iPSC-Exos can restore a balanced inflammatory microenvironment. Moreover, the reduction in pro-inflammatory cytokines (TNF-α, IL-6, and IL-1β) in iPSC-Exos-treated wounds underscores their ability to mitigate chronic inflammation, a hallmark of diabetic wounds. These findings highlight the role of iPSC-Exos in modulating immune responses and promoting tissue repair.

The findings from this study have important implications for the treatment of diabetic wounds, which are a major clinical challenge due to their chronic nature and poor response to conventional therapies. The capacity of iPSC-Exos to promote re-epithelialization and tissue remodeling, regulate inflammation, and reduce scar formation makes them a promising therapeutic agent for diabetic wound healing. Moreover, the use of iPSC-Exos offers several advantages, including their non-immunogenic nature, ease of isolation, and potential for large-scale production, making them a viable option for clinical translation.

Conclusions

In summary, this preclinical study conducted in a mouse model of diabetic wounds suggests that iPSC-derived exosomes (iPSC-Exos) may facilitate wound healing through multiple mechanisms. Key findings include enhanced tissue regeneration, as reflected by promoted re-epithelialization, granulation tissue formation, increased collagen deposition, and a concurrent reduction in scar formation; improved angiogenesis, characterized by increased local blood perfusion and higher density of mature vessels; as well as immunomodulatory effects, evidenced by reduced infiltration of pro-inflammatory M1 macrophages, decreased levels of inflammatory cytokines (TNF-α, IL-6, IL-1β), and an increase in anti-inflammatory M2 macrophages. These results indicate that iPSC-Exos hold potential as a cell-free therapeutic strategy for diabetic wound repair, supported further by their non-immunogenic nature and scalable production feasibility. The findings provide a preclinical basis for the design of an ongoing investigator-initiated trial (IIT, Ethics Approval ID: 2025BJYYEC-KY097-01). This clinical trial incorporates key mechanisms identified in our wound healing model, particularly pro-angiogenic and immunomodulatory effects, with the aim of facilitating future clinical translation of iPSC-Exos in diabetic tissue regeneration.

However, several limitations warrant consideration: the specific bioactive components within iPSC-Exos responsible for the observed therapeutic effects require further elucidation; optimal dosing regimens and delivery strategies remain to be established in future studies.

Abbreviations

iPSC-Exos

Induced pluripotent stem cell-derived exosomes

STZ

Streptozotocin

iPSCs

Induced pluripotent stem cells

AP

Alkaline phosphatase

OCT4

Octamer-binding transcription factor 4

SSEA4

Stage-specific embryonic antigen 4

SOX2

SRY-box transcription factor 2

PBS

Phosphate-buffered saline

TEM

Transmission electron microscopy

SPF

Specific pathogen-free

PFA

Paraformaldehyde

H&E

Hematoxylin and Eosin

IF

Immunofluorescence

BCA

Bicinchoninic acid

TNF-α

Tumor necrosis factor-alpha

IL-6

Interleukin-6

IL-1β

Interleukin-1 beta

ELISA

Enzyme-linked immunosorbent assay

NC

Nitrocellulose

ALIX

ALG-2-interacting protein X

TSG101

Tumor susceptibility gene 101 protein

CD9

Cluster of differentiation 9

CD63

Cluster of differentiation 63

HRP

Horseradish peroxidase

IgG

Immunoglobulin G

NS

Normal saline

α-SMA

α-smooth muscle actin

CD86

Cluster of differentiation 86

CD206

Cluster of differentiation 206

M1

Classically ctivated macrophages

M2

Alternatively activated macrophages

Author contributions

L.L. conceived and designed the study, collected and assembled data, performed data analysis and interpretation, and wrote the manuscript. W. H. and J.Q. provided administrative support and study materials. L.W. contributed to conception and design and provided administrative support. Y.W. contributed to conception and design and provided financial support. Y.X., H.C., H.J., and X.H. provided administrative support, study materials, and financial support. J.Z. conceived and designed the study, provided financial and administrative support, supplied study materials, and gave final approval of the manuscript. All authors reviewed the manuscript.

Funding

This research was supported by the National Key Research and Development Program of China (2022YFC3602000), the CAMS Initiative for Innovative Medicine (2021-I2M-1-005, 2023-I2M-2-005, 2021-I2M-1-035 and 2021-I2M-1-053), the National Natural Science Foundation of China (82402139, 8240055254), China Postdoctoral Science Foundation (GZB20240072), National High Level Hospital Clinical Research Funding (2022-PUMCH-B-092, 2022-PUMCH-C-059), and Beijing Key Clinical Specialty Construction Project. Chinese Academy of Medical Sciences (CAMS) Clinical and Translational Medicine Research Fund (BJ-2024-312).

Data availability

The datasets used during this study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

All mouse breeding and handling procedures were approved by the Institutional Animal Care and Use Committee (IACUC) of Chinese Academy of Medical Sciences and School of Basic Medicine. (Title of the approved project: Investigation of the therapeutic effects and the mechanisms of iPSC-derived exosomes in diabetic skin wound healing; Name of the institutional approval committee: Institutional Animal Care and Use Committee (IACUC) of Chinese Academy of Medical Sciences and School of Basic Medicine; Approval number: ACUC-XMSB-2024-069; Date of approval: March 18th 2024). All experimental procedures will strictly comply with the requirements of the Ethics Committee and the Laboratory Animal Welfare Committee. The animal study has been reported in line with the ARRIVE guidelines 2.0. Peripheral blood sample from healthy volunteer was obtained from agreeable donor with the approval of the Ethical Committee for Clinical Trials of Acrospace Center Hospital (Title of the approved project: Procurement of peripheral blood from healthy donors for the purpose of building an induced pluripotent stem cell (iPSC) repository for use in clinical studies; Approval number: EC2023-38; Date of approval: March 17th 2023). The iPSC used in this study were generated by one of our research team members (Guidon Pharmaceutics, Beijing, China) from the peripheral blood samples obtained under the above ethical approval. Written informed consent was obtained from the donor prior to participation. The consent form explicitly covered the use of biological samples for scientific research.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.