Engineered mesenchymal stem cells-laden chondroitin sulfate hydrogel promotes diabetic wound healing through ROS scavenging and macrophage polarization

Abstract

The treatment of diabetic wound complications has long been a formidable challenge in the clinic, and a multifunctional biomaterial dressing holds great promise as an ideal approach for promoting diabetic wound healing. In this study, we designed a crosslinked chondroitin sulfate (CS) and polyethylene glycol (PEG) hydrogel (CS gel) as a reservoir via a Michael addition reaction. To further enhance the biological functions, surface-engineered mesenchymal stem cells (MSCs) modified with bilirubin-modified chondroitin sulfate-mediated mixed micellar backpacks possessing ROS scavenging/responsive and CD44-targeting abilities (DCMMs) were incorporated into the hydrogel system. The hydrogel served as a physical carrier, providing a sustained-release platform that ensured the long-term presence and controlled delivery of MSCs-tethered micellar backpacks at the wound site. In vitro experiments revealed the remarkable capacity of the hydrogel for ROS scavenging and regulating macrophage polarization. In vivo investigations in diabetic mouse models further confirmed the therapeutic potential, revealing a significant acceleration of wound closure, attenuation of inflammation, and augmentation of angiogenesis. Overall, this innovative hydrogel system that encapsulates surface-engineered mesenchymal stem cells (MSCs-DCMMs) represents a promising remedy and a framework for stem-cell-based therapeutic systems to treat diabetic wounds.

Graphical abstract

Highlights

• •Engineered MSC-laden hydrogel enhances diabetic wound healing.
• •Bioorthogonal conjugation of ROS-responsive CD44-targeting DCMMs to MSCs surfaces.
• •Hydrogel extends wound-site retention of MSCs-DCMMs for localized drug delivery.
• •Synergistic effects: ROS clearance, M2 polarization, angiogenesis, ECM remodeling.

1. Introduction

Diabetes is a chronic metabolic disease with a high incidence worldwide [1,2]. Diabetic wounds such as diabetic foot ulcers are common complications of diabetes. They can cause severe consequences, including amputation and loss of life. Diabetic wounds are usually nonhealing and accompanied by prolonged inflammation, severe infection and impaired angiogenesis, which seriously affect the health of patients and may even be life-threatening [[3], [4], [5]]. The wound-healing process of diabetic patients is impeded by the accumulation of reactive oxygen species (ROS), an abnormal macrophage phenotype, dysregulation of proinflammatory cytokines, impaired cell proliferation, and angiogenesis during the inflammatory phase [4,6,7]. In addition, chronic wounds are particularly vulnerable to bacterial infection when there is a hyperglycemic environment [8]. Therefore, scavenging ROS, regulating the macrophage phenotype and inflammation, inducing angiogenesis, and eliminating bacterial infection are the keys to effectively treating chronic diabetic wounds. The development of multifunctional injectable hydrogels (dressings with ROS-scavenging ability and anti-inflammatory effects) holds great promise for clinical diabetic wound healing.

Gauze, sponges, nanofibers, polymer bandages, and other 3D scaffolds with similar structures have been used in clinical dressings [[9], [10], [11]]. Because extracellular matrices (ECMs) provide significant physical support and biomechanical intervention for tissue morphogenesis, hydrogel dressings with 3D networks composed of natural components of ECMs have great potential in diabetic wound healing [12]. Chondroitin sulfate (CS) is a natural polysaccharide that widely exists in ECMs and is conducive to cell adhesion and proliferation [13]. It also has biological characteristics such as anti-inflammatory activity and stem cell niche maintenance [12], thus making it an ideal matrix for hydrogel dressings and helping restore the function of the inflammatory microenvironment during diabetic wound healing [14]. Polyethylene glycol (PEG) is an FDA-approved synthetic polymer with good biocompatibility, biodegradability and nontoxicity [15]. It is one of the most ideal inert polymers for adjusting the mechanical properties of hydrogels [16]. The hydrogel formed from CS and PEG through cross-linking can combine the advantages of natural and synthetic materials, significantly increasing the degradation time. Additionally, hydrogels are injectable and highly adaptable to irregular wounds. Studies have reported that hyperbranched PEG- and sulfhydryl-modified CS can be used to synthesize hydrogels through Michael addition for cartilage regeneration [17]. Nevertheless, the therapeutic requirements for diabetic wounds are complicated, and CS hydrogels without functional modifications do not have enough biological activity and function for therapy. Owing to the porous 3D structure of the hydrogel, it can carry drugs, micelles, and even cells to achieve long-term release, enhanced anti-inflammatory activity and antibacterial properties, so injectable hydrogels loaded with functional carriers could be ideal dressings for diabetic wound healing [11].

Mesenchymal stem cells (MSCs), which are multipotent stem cells, have been widely studied because of their low immunogenicity, self-renewal, immunoregulatory and anti-inflammatory effects [18]. Research has demonstrated that MSCs can promote diabetic wound healing by enhancing angiogenesis and promoting cell proliferation and ECM secretion without serious complications or adverse reactions related to treatment [[19], [20], [21], [22]]. In addition, MSCs can promote the polarization of activated M1 macrophages (proinflammatory subtypes) to the M2 anti-inflammatory phenotype, thereby alleviating diabetic wounds [23], which shows that MSCs have superior potential in the treatment of diabetic wounds. However, during the inflammatory phase of diabetic wounds, the entire wound environment is filled with excessive ROS, which may lead to difficulties in maintaining the vitality and function of MSCs in environments characterized by high oxidative stress [[24], [25], [26]]. Combining ROS-responsive/scavenging units with MSCs could achieve synergistic effects, while the ROS-responsive/scavenging units clear ROS and reduce the damage caused by oxidative stress to stem cells to maintain the viability of MSCs, and MSCs play a role in repairing diabetic wounds [27].

Compared with exogenous antioxidants, bilirubin (BR), an endogenous antioxidant, could be an ideal ROS-responsive/scavenging compound because it can be completely metabolized and eliminated in the body [[28], [29], [30], [31]], but its insolubility limits its combination with MSCs. Since proinflammatory M1 macrophages overexpress CD44 receptors [32,33], CS-modified materials are often used for targeted therapy and drug delivery. Therefore, we speculate that the combined delivery of biocompatible materials using the ROS-responsive compounds BR and CD44-targeting structural unit CS (CS-BR) and MSCs could play a role in synergistic treatment for diabetic wound healing, in which not only can CS-BR target M1 macrophages in the inflammatory microenvironment so that MSCs can promote the polarization of macrophages but also scavenge ROS, protecting MSCs from damage caused by oxidative stress so that MSCs can repair diabetic wounds more efficiently.

Based on the above theories and our previous research results [34], in this study, we designed a multifunctional injectable hydrogel composed of thiolated chondroitin sulfate (CS-SH) and polyethylene glycol diacrylate (PEG-DA) loaded with surface-engineered MSCs with CS-BR mixed micellar backpacks to treat diabetic wounds (Scheme 1). First, mixed micelles with ROS scavenging/responsive and CD44-targeting abilities (DCMMs) were bioorthogonally coupled on the surface of MSCs to obtain MSCs-DCMMs. Then, CS-SH was crosslinked with PEG-DA by a mild Michael addition reaction to form a multifunctional injectable hydrogel, which was used to carry the micellar knapsack to fabricate the MSCs-DCMMs@CS Gel. We hypothesized that the hydrogel matrix network, as the depot of the MSCs-DCMMs, would degrade and release the MSCs-DCMMs slowly, thus protecting the micellar knapsack, preventing its rapid removal, prolonging the residence time of the MSCs and micellar knapsack at the wound site, and promoting controllable and sustained local drug delivery. The synergistic combination of these components facilitates therapeutic outcomes through multiple mechanisms, including promoting angiogenesis, anti-inflammatory effects, and extracellular matrix regeneration. This study aims to provide an effective and promising therapeutic strategy and a framework for stem cell-based therapeutic systems to treat diabetic wound healing.

Scheme 1.

Schematic illustration of the preparation and therapeutic effects of the MSCs-DCMMs@CS Gel.

2. Materials and methods

2.1. Materials

DBCO-PEG_2k-PCL_3.5k (≥98 %) was purchased from Xi'an Ruixi Biological Technology Co., Ltd. (Xi'an, China). Ac4GalNAz (Cat No: 1086–100) was purchased from Click Chemistry Tools (Scottsdale, USA). Chondroitin 4-sulfate (CS, 33 kDa, ≥90 %, bovine bone) was obtained from Solarbio Co., Ltd. (Beijing, China); PEG₅₀₀₀-DA was obtained from MACKLIN reagent (Shanghai, China).

The RAW264.7 mouse macrophage line was purchased from Wuhan Punosai Life Technology Co., Ltd. (Wuhan, China) and cultured in DMEM containing 1 % penicillin‒streptomycin and 10 % fetal bovine serum. Human umbilical vein endothelial cells (HUVECs) and L929 fibroblasts were obtained from the Cell Bank of Chinese Academy of Sciences (Shanghai, China). Human umbilical cord mesenchymal stem cells (MSCs) and culture medium were obtained from Fuyuan Biotechnology (Fuyuan Biotechnology Co., Ltd. Shanghai, China).

2.2. Synthesis of hydrogel precursors

Chondroitin sulfate-bilirubin (CS-BR) was obtained as described in our previous study [34]. Thiolated chondroitin sulfate (CS-SH) was synthesized according to Scheme 1 [35]. Briefly, CS (2 mmol), NHS (4 mmol) and EDC·HCl (4 mmol) were dissolved in 50 mL of ultrapure water and stirred for 2 h. CSA·HCl (6 mmol) was dissolved in 10 mL of HCL solution (pH 3.0) and then added to the mixture. The whole solution system was stirred for another 24 h under nitrogen protection. Finally, the mixture was dialyzed against HCL solution (pH 3.5) containing 0.1 M NaCl for 2 days. CS-SH was frozen and lyophilized, and the structure was analyzed by ¹H NMR spectroscopy.

2.3. Preparation and characterization of DBCO-PEG_2k-PCL_3.5k/CS-BR mixed micelles (DCMMs)

DCMMs were prepared via the thin-film hydration method [34]. Briefly, 20 mg of egg yolk lecithin and 5 mg of DBCO-PEG_2k-PCL_3.5k were dissolved in 10 mL of chloroform, rotary evaporated to form a film, and then hydrated with PBS, in which 0, 1, 3, or 5 mg of CS-BR was dissolved to select the optimal ratio. Next, the mixture was sonicated for 10 min (400 W, 5 s on and 5 s off) in an ice bath to obtain DCMMs. In addition, we also prepared DCMM/C6 micelles loaded with coumarin 6 (C6) to label the mixed micelles by adding C6 with egg yolk lecithin and DBCO-PEG_2k-PCL_3.5k together, while the other procedures were the same.

The particle size and dispersity (Ð) of the DCMMs were measured via dynamic light scattering (DLS) (Malven, Nano ZS90, Malven, UK), and the morphology of the DCMMs was observed via transmission electron microscopy (TEM) (Talos L120C G2, Thermo Fisher, USA). The stability of the DCMMs was studied daily to investigate the effects of intravenous injection and protein adsorption on the liposomes via DLS. In general, DCMMs were diluted with 10 % fetal bovine serum (FBS), PBS, or water and incubated at 37 °C for seven days. The particle size was measured via DLS.

The critical micelle concentration (CMC) was detected via the pyrene fluorescence probe technique [36]. In brief, a pyrene solution of 1.5 × 10⁻⁵ M was dissolved in tetrahydrofuran, and 40 μL of pyrene solution was added to a 1 mL brown volumetric flask and placed in a fume hood for 6 h to dry. Next, 0.7, 0.3, 0.1, 0.07, 0.03, 0.01, 0.007, 0.003, 0.001, 0.0007, and 0.0003 mg/mL DCMMs were added to the volumetric flask, and ultrasonication was used to dissolve the pyrene solution. After that, the flask was sealed with sealing film and incubated at 37 °C for 24 h with oscillation, after which the fluorescence intensity of the solution at 333 nm and 328 nm was measured via a fluorescence spectrophotometer (Shimadzu, Japan).

2.4. Determination of the optimal micelle concentration for constructing MSCs-DCMMs

To determine the optimal micelle concentration for constructing MSCs-DCMMs [27], the MSCs were incubated with 0.02 mg/mL Ac₄GalNAz for 72 h to obtain N₃-MSCs. DCMMs were diluted to 0.005, 0.01, 0.02, 0.04, 0.08, and 0.1 mg/mL in serum-free DME/F12 medium and incubated with N₃-MSC for 2 h to link the micelle backpacks, thus obtaining MSCs-DCMMs. DCMMs/C6 instead of DCMMs were used to prepare MSCs-DCMMs/C6 for other experiments when the other procedures were the same. The MSCs-DCMMs were then stimulated with 100 μM H₂O₂ for 2 h. The ROS level and apoptosis rate of the MSCs-DCMMs were studied via flow cytometry (Beckman, USA), and the cell viability was determined via the MTT assay.

2.5. Characterization of the MSCs-DCMMs

The morphology of the MSCs-DCMMs was observed by scanning electron microscopy (SEM) (ZEISS GeminiSEM 300, Germany) [27]. Moreover, MSCs-DCMMs/C6 were fixed with 4 % paraformaldehyde, and the nuclei were stained with Hoechst. MSCs-DCMMs/C6 were observed under a fluorescence microscope.

To investigate the stability of the MSCs-DCMMs, MSCs-DCMMs/C6 were prepared by incubating N₃-MSCs with 0.08 mg/mL DCMMs/C6 for 2 h. MSCs that were not preincubated with Ac₄GalNAz to form N₃-MSCs and incubated with 0.08 mg/mL DCMMs/C6 for 2 h (MSCs + DCMMs/C6) were used as controls. Two groups of cells were washed and cultured in serum-free DMEM/F12 for 1, 3 and 5 days. Finally, the fluorescence intensity of the cells was measured via flow cytometry (BD FACS Verse, USA) to evaluate the stability of the MSCs-DCMMs [27].

2.6. Preparation and characterization of the MSCs-DCMMs@CS gel

The procedure for preparing the injectable hydrogel samples consisted of the following steps: First, CS-SH was dissolved in water (10 %, w/v), and PEG-DA was dissolved in 0.5 M HEPES buffer (7.5 %, w/v, pH 8.0). The two solutions were vortex-mixed in equal volumes and cross-linked at 37 °C to form a blank CS hydrogel. To prepare DCMMs@CS Gel and MSCs-DCMMs@CS Gel, DCMMs were prepared with 0.5 M HEPES buffer (pH 8.0), and PEG-DA was dissolved in a DCMM solution or an MSCs-DCMM solution, while the other procedures were the same. The MSCs were subsequently resuspended in the PEG-DA solution to obtain the MSCs@CS gel.

The morphologies of the hydrogels were assessed by scanning electron microscopy (SEM) (ZEISS GeminiSEM 300, German). Briefly, the hydrogels were flash-frozen in liquid nitrogen for 1 min and then subjected to freeze-drying for 24 h. Afterward, the samples were sectioned into small pieces, mounted onto sample studs, and coated with gold by sputtering. The hydrogel structure was then visualized via SEM.

For the rheological characterization of the hydrogels, the storage modulus (G′), loss modulus (G″), gel point and viscosity were determined via a rotational rheometer (MCR 302; Anton paar; Graz, Austria) [35]. Swelling properties were also evaluated. The lyophilized hydrogels were weighed (W0), swelled in PBS at 37 °C and weighed (Wt) at predetermined times after the removal of excess PBS. The weight of the initial dry hydrogel was recorded before the experiment, after which the hydrogel was soaked in PBS at 37 °C. The swelling ratio (SR) was calculated via the following formula: SR (g/g) = (Wt−W0)/W0.

2.7. In vitro biocompatibility

2.7.1. In vitro cytocompatibility

Moreover, the Calcein-AM/PI staining kit (Keygenbio, Cat: KGAF001, Jiangsu, China) was employed to visualize cell viability following hydrogel co-culture.

The leaching liquor of the hydrogels was collected to investigate their cytocompatibility in vitro [12]. After washing with PBS, the hydrogels were soaked in DMEM (containing 1 % penicillin‒streptomycin and 10 % FBS), extracted for 24 h and then filtered through a 0.22 μm membrane for use. RAW264.7 cells, HUVECs and L929 cells were seeded in 96-well plates at 5000 cells/well and incubated for 24 h. Next, the culture medium was replaced with the leach liquor of the hydrogels, and the cell viability was measured via the MTT assay to determine the in vitro cytocompatibility of the hydrogels. Moreover, the Calcein-AM/PI staining kit was employed to visualize cell viability with L929 cells following hydrogel co-culture [4].

2.7.2. Hemolysis assay

Blood from SD rats was collected for hemolysis assays, and H₂O and 0.9 % NaCl were used as positive and negative controls, respectively [4,34,37]. The prerequisite solutions of different hydrogel groups were added to the erythrocyte dilution solution and incubated at 37 °C for 2 h. After incubation, the supernatant was centrifuged, and the hemoglobin content was determined at 541 nm via a microplate apparatus as follows [34]:

$$\text{Hemolytic}\text{ratio}(\%)=\left(\text{ODc}-\text{ODn}\right)/\left(\text{ODp}-\text{ODn}\right)\times 100\%$$

where ODc is the absorbance of the sample, ODn is the absorbance of the negative control and ODp is the absorbance of the positive control.

2.8. In vitro antioxidant capacity

To evaluate the in vitro antioxidant capacity of the hydrogels, a DPPH free radical scavenging assay was performed first. An appropriate amount of pregel solution was added to the solution of DPPH, and an equal amount of PBS was substituted for the control group. The hydrogels were incubated at 37 °C for 30 min under light-free conditions, and the absorbance of the DPPH solution at 517 nm was measured with a microplate reader (Biotek, USA).

To investigate the ability of DCMMs to scavenge ROS, DCFH-DA was used as an intracellular ROS probe. MSCs were stimulated with 100 μM H₂O₂ for 2 h and then incubated with DCMMs for 24 h. Subsequently, the MSCs were incubated with DCFH-DA dye for 30 min, and the fluorescence intensity was determined via laser confocal microscopy (LSCM, LSM800, Oberstein Zeiss, Germany) and analyzed semiquantitatively. Following identical procedures, after a 48-h incubation period, the culture medium was aspirated, and the cells were lysed following a rinse with precooled PBS. The SOD content in the cells was then determined via a total SOD activity assay kit.

To evaluate whether MSCs-DCMMs@CS Gels have the ability to clear ROS, RAW264.7 cells were stimulated with 500 ng/mL lipopolysaccharide (LPS) and 10 ng/mL interferon (IFN) to establish an in vitro cellular inflammatory model and then incubated with different sets of hydrogels for 48 h. After incubation with DCFH-DA dye for 30 min, the fluorescence intensity was analyzed via flow cytometry (BD FACS Verse, USA).

To explore the protective effect of the MSCs-DCMMs@CS Gel on mitochondria, RAW264.7 cells inoculated in 12-well plates were stimulated with 500 ng/mL LPS and 10 ng/mL IFN and then incubated with the CS Gel, MSCs@CS Gel, DCMMs@CS Gel or MSCs-DCMMs@CS Gel for 48 h. JC-1 monomers and JC-1 aggregates in the mitochondrial matrix of RAW264.7 cells were fluorescently labeled with a Mitochondrial Membrane Potential Kit [34]. The fluorescence intensity was subsequently measured via flow cytometry.

2.9. Phenotypic effects of the MSCs-DCMMs@CS Gel on macrophages in vitro

To assess the effect of the MSCs-DCMMs@CS Gel hydrogel on macrophage polarization in vitro, RAW264.7 cells were inoculated in 12-well plates at a density of 1 × 10⁵ cells per well. After overnight incubation, 500 ng/mL LPS and 10 ng/mL IFN were added, and the hydrogels were cocultured for 24 h. Normal cells without any treatment were used as the negative control group, and cells treated with only LPS were used as the positive control group. After incubation, the cells were washed, collected and incubated with an anti-CD86-APC (M1) antibody and an anti-CD163-PE (M2) antibody for 30 min, and the fluorescence intensity was measured via flow cytometry.

To detect the macrophage phenotype in the skin, the skin was cut in a Petri dish, added to DMEM containing 10 % FBS, collagenase IV (1 mg/mL) and deoxyribonuclease (50 μg/mL), and the mixture was incubated for 90 min. The digestion was terminated with 10 ml of DMEM containing 10 % DMEM, after which the mixture was passed through a 70 μm cell strainer twice, washed, and then closed and processed. The cells were incubated with an anti-CD86-APC (M1) antibody, an anti-CD163-PE (M2) antibody, an anti-CD45 antibody, or an anti-F4/80 (M0) antibody for 30 min and analyzed via flow cytometry (BD FACS Verse, USA).

2.10. Diabetic skin wound healing effect of the MSCs-DCMMs@CS Gel in vivo

Male C57BL/6J mice (6–8 weeks old) were fasted for 12 h and then intraperitoneally injected with STZ (60 mg/kg) for 5 days to establish a diabetic mouse model. After 72 h of normal feeding, mice with fasting blood glucose concentrations greater than 16.1 mmol/L were selected for the experiments. All the animal studies were conducted according to the National Research Council Guidelines for the Care and Use of Laboratory Animals and approved by the Animal Experimentation Ethics Committee of Anhui Medical University (LLSC20241705).

The mice were randomly divided into 5 groups (n = 5): the blank group, CS gel group, MSCs@CS gel group, DCMMs@CS gel group, and MSCs-DCMMs@CS gel group. The mice were anesthetized with isoflurane, and the dorsal hair of each mouse was shaved and disinfected with 75 % ethanol. Full-thickness wounds were created on the shaved backs via 10-mm biopsy perforations. Next, the defects were covered with CS gel, MSCs@CS gel, DCMMs@CS gel, or MSCs-DCMMs@CS gel. All the wounds were photographed on postoperative days 0, 2, 4, 6, 8, and 10, and the area of the wounds was calculated via ImageJ software. The mice were then sacrificed, and full-thickness skin samples were collected for subsequent experiments. In addition, different wound sizes were recorded at different times, and their wound areas were counted. The wound closure rate (WCR) was calculated using the following equation:

$$\text{WCR}=\left({\mathrm{A}}_0-{\mathrm{A}}_{\mathrm{t}}\right)/{\mathrm{A}}_0$$

While WCR refers to wound closure rate, A₀ and A_t denote the wound area on days 0 and t, respectively.

2.11. Histological and immunohistochemical analysis

For histological analysis, full-skin wound samples were collected and fixed on the 3rd and 7th postoperative days. All the samples were subjected to hematoxylin‒eosin (H&E) and Sirius red staining to assess epithelial regeneration, collagen deposition, the inflammatory response, and vascularization of the wounds [38,39]. Full-skin wound samples were collected on day 10 for immunofluorescence and immunohistochemical staining of CD86, CD163, α-SMA, CD31, HIF-1α, TNF-α, and IL-10.

Following sample detachment, it was fixed in 4 % paraformaldehyde overnight, dehydrated using an automated dehydrator, embedded in paraffin, and sectioned into 4 μm slices. Subsequently, the sections were subjected to immunofluorescence staining for CD86, CD163, collagen I/collagen III, CD31 and α-SMA, immunohistochemical staining for VEGFA, HIF-1α, IL-1β, IL-10 and TNF-α. The sections were then observed and photographed using a Slide scanner (3DHISTECH, Hungary).

Proteomic analysis was used to evaluate protein expression in the wound tissues of diabetic mice on day 10, and the protein expression in the MSCs-DCMMs@CS Gel hydrogel-treated group was compared with the control group (MSCs-DCMMs@CS-Gel vs control). The extraction, separation, and analysis of proteins were conducted by Shanghai Biopharma Biotechnology Co., Ltd. (Shanghai, China). Bioinformatic analysis was carried out with Microsoft Excel and R statistical computing software. Hierarchical clustering analysis and volcano plots were generated with the statistical language R. To annotate the sequences, information was extracted from UniProtKB/SwissProt, Kyoto Encyclopedia of Genes and Genomes (KEGG), and Gene Ontology (GO). GO and KEGG enrichment analyses were carried out with Fisher's exact test, and FDR correction for multiple testing was also performed. The GO terms were grouped into three categories: biological process (BP), molecular function (MF), and cellular component (CC). According to Fisher's exact test, enriched GO and KEGG pathways were nominally statistically significant (p < 0.01). The construction of protein–protein interaction (PPI) networks was also conducted via the STRING database with Cytoscape software.

2.12. Safety assessment

The major organs of the mice, including the heart, liver, spleen, lungs, and kidneys, were collected on day 10, fixed, and embedded in paraffin wax to obtain H&E-stained sections. These sections were analyzed histologically to assess the treatment's systemic toxicity.

2.13. Statistical analysis

The data were expressed as means ± SDs. Statistical analysis was performed via a t-test or one-way analysis of variance (ANOVA). P values less than 0.05 were considered to indicate significant differences. (∗p < 0.05).

3. Results and discussion

3.1. Optimization and characterization of DCMMs

The method used to fabricate the DCMMs is shown in Fig. 1A. First, CS-BR conjugates were successfully constructed and confirmed in our previous study [34]. The optimized material mass ratios of the DCMMs were screened, as shown in Fig. S1A–D. The mass ratios of DBCO-PEG_2k-PCL_3.5k:CS-BR, including 5:0, 5:1, 5:3 and 5:5, were used to fabricate DCMMs, and among all the ratios, when the mass ratio of DBCO-PEG_2k-PCL_3.5k:CS-BR was 5:5, the DCMMs had a particle size of ∼127.9 nm, a zeta potential of ∼ −14.2 mV, and a polydispersity index (PDI) of ∼0.289. Additionally, with this structure, the DCMMs had good stability for seven days in water, PBS and serum (Fig. S1E). The morphology of the DBCO micelles (when the mass ratio of DBCO-PEG2k-PCL3.5k:CS-BR was 5:0) and DCMMs was characterized via TEM. Following the DLS results, the TEM results revealed that the particle size of the DCMMs increased significantly compared with that of the DBCO micelles, indicating that adding CS-BR conjugates could increase the size of the micelles, and the morphology revealed that both the DBCO micelles and DCMMs had homogeneous spherical shapes (Fig. 1B). The CMC of the DCMMs was determined to be 0.0478 mg/mL, which was sufficient low to maintain the micellar structure (Fig. S2).

Fig. 1.

Characterization of the MSCs-DCMMs. (A) Formulation of MSCs-DCMMs. (B) Size distribution and TEM images of DBCO micelles and DCMMs. (C) SEM images of MSCs, MSCs + DCMMs and MSCs-DCMMs. (D) Representative CLSM images of MSCs with or without Ac4GalNAz pretreatment coincubated with DCMMs. Scale bar: 100 μm. (E) Quantification of coumarin fluorescence via CLSM. (F) Representative CLSM images of ROS in MSCs cells through DCFH-DA fluorescence. (G) Semiquantification of DCFH-DA fluorescence. Scale bars: 100 μm and 50 μm. The data represent the means ± SDs, n = 3. ∗∗∗p < 0.001.

3.2. Optimization and characterization of the MSCs-DCMMs

To confirm the optimal micellar concentration for constructing DCMMs tied to MSCs (MSCs-DCMMs), first, MSCs conjugated with the azide group (N₃-MSCs) were obtained by preincubating the MSCs with Ac4GalNAz. N₃-MSCs were then incubated with different concentrations of DCMMs for 2 h to construct MSCs-DCMMs. The micellar backpacks were tied on MSCs by bioorthogonal click chemistry, which could provide stable and safe covalent labeling between MSCs and micelles [27,40]. Then, the MSC-DCMMs were incubated with H₂O_2, and the optimal micellar concentration was investigated by evaluating and comparing the apoptosis rate, viability and ROS level of the MSCs. The data revealed that the viability of the MSCs decreased to 60.68 % after H₂O₂ stimulation (p < 0.001), whereas the viability of the MSCs increased to different degrees after they were exposed to different concentrations of DCMMs (Fig. S3C). In addition, all concentrations of DCMMs inhibited H₂O₂-induced MSCs apoptosis, especially 0.04 and 0.08 mg/mL DCMMs (Fig. S3A and B). When the ROS levels were compared, 0.08 and 0.005 mg/mL DCMMs significantly decreased the ROS production of H₂O₂-induced MSCs (p < 0.05) (Fig. S3D and E). Based on these results, 0.08 mg/mL DCMMs was ultimately selected as the optimal concentration for incubation with MSCs.

To confirm that DCMMs, not DBCO micelles, can protect MSCs from oxidative stress damage, the antioxidant capacity of DCMMs against elevated oxidative stress in H₂O₂-stimulated MSCs was determined. A 2-7-dichlorofluorescin diacetate (DCFH-DA) probe was used to detect intracellular ROS levels. As shown in Fig. 1F and G, H₂O₂-stimulated MSCs produced greater amounts of ROS than untreated cells did, and compared with DBCO micelles, DCMMs significantly reduced the ROS levels in stimulated H₂O₂-MSCs (p < 0.001), indicating that CS-BR conjugates with ROS-responsive/scavenging ability could protect MSCs from damage caused by oxidative stress.

To verify whether DCMMs were attached to the surface of the MSCs, the morphology of the MSCs-DCMMs was observed by scanning electron microscopy, and large spherical-like particles could be observed on the surface of the MSCs, demonstrating that the DCMMs were successfully attached to the MSCs (Fig. S5 and Fig. 1C). In addition, we also prepared DCMMs loaded with coumarin 6 (DCMMs/C6) to observe whether DCMMs were attached to the surface of the MSCs visually, and CLSM images revealed weak fluorescence in the MSCs without preincubating the cross-linking agent Ac4GalNAz, which was considered to indicate the internalization of DCMMs by the MSCs. Nevertheless, the fluorescence intensity was significantly stronger when DCMMs were co-incubated with MSCs after they were preincubated with Ac4GalNAz than when they were not preincubated with Ac4GalNAz (p < 0.001) (Fig. 1D and E), demonstrating that after being preincubated with Ac4GalNAz, DCMMs were successfully conjugated with MSCs via click chemistry. To determine the stability of DCMMs on surface MSCs via click chemistry, DCMMs co-incubated with MSCs but without preincubation with Ac4GalNAz were used as a control group (MSCs + DCMMs/C6), and their fluorescence intensity over time was detected. As shown in Fig. S6A and Fig. S6B, the fluorescence intensity of the MSCs + DCMMs/C6 mixture decreased rapidly with time, whereas that of the MSCs- DCMMs/C6 mixture remained stable over 3 days. This demonstrated the good stability of micellar backpacks on the surface of MSCs via bioorthogonal click chemistry [27]. Overall, these results confirmed that DCMMs, which are mixed micelles with ROS scavenging/responsive abilities, were successfully bioorthogonally coupled on the surface of MSCs, forming MSCs-DCMMs.

3.3. Preparation and characterization of the MSCs-DCMMs@CS Gel

First, CS-SH was successfully synthesized, and the structure was confirmed by 1H NMR spectroscopy (Fig. S7). The fundamental formulation of the CS gel was 10 % CS-SH dissolved in water mixed with 7.5 % PEG-DA dissolved in 0.5 M HEPES buffer (pH 8.0), which achieved complete gelation within approximately 200 s at 37 °C (Fig. 2A), which was consistent with the rheological results (Fig. 2E). The MSCs@CS gel, DCMMs@CS gel, and MSCs-DCMMs@CS gel all exhibited similar gelation times as the CS gel did, which was suitable for injection under mild conditions. SEM revealed a typical porous three-dimensional structure of the MSCs-DCMMs@CS gel (Fig. 2H), and such a stereoscopic three-dimensional structure provided suitable space for the incorporation of MSCs, DCMMs, and MSCs-DCMMs.

Fig. 2.

Characterization of the MSCs-DCMMs@CS Gel. (A) Gelation time of the hydrogels, including the CS Gel, MSCs@CS Gel, DCMMs@CS Gel and MSCs-DCMMs@CS Gel. (B) The swelling ratio of the hydrogels. (C) Dependence of G′ and G″ on the frequency of hydrogels. (D) Viscosity changes in the hydrogels with time. (E) Storage modulus G′ and loss modulus G″ curves of the hydrogels. (F) Representative self-healing images of the MSCs-DCMMs@CS gel. (G) Autonomous self-healing property of the MSCs-DCMMs@CS gel. (H) SEM image of the MSCs-DCMMs@CS gel. (I) Live/dead staining of MSCs in the hydrogel. Scale bar: 50 μm. The data represent the means ± SDs, n = 3.

According to the swelling results (Fig. 2B), all the hydrogels rapidly swelled and reached swelling equilibrium within 24 h. Their equilibrium swelling rate reached 300 %, which could be due to the porous 3D structure of the CS-based networks [41]. The rheological results revealed that the MSCs-DCMMs@CS gel had a certain elasticity and viscosity, and its storage modulus (G′) was greater than the loss modulus (G″) within a certain frequency range (Fig. 2C), indicating the elastic properties of the hydrogels. The viscosities of each group of hydrogels were similar and remained stable after reaching a certain time point, indicating that the addition of MSCs and DCMMs did not affect the structure of the hydrogels (Fig. 2D).

To further verify the self-healing ability of the MSCs-DCMMs@CS Gel in practical applications, the hydrogel was cut into two halves to simulate fracture during the use process. Next, the two parts of the ruptured hydrogel were contacted for 5 min, and the results revealed the complete healing of the hydrogel (Fig. 2F), demonstrating the good self-healing capability of the MSCs-DCMMs@CS Gel. The continuous strain sweep results revealed that the G′ of the hydrogel was lower than the G″ when a high dynamic strain (100 %) was used (Fig. 2G), indicating that the network of the MSCs-DCMMs@CS gel collapsed. When a lower strain (1 %) was used, the collapsed structure of the hydrogel recovered promptly, indicating that the hydrogel had a great self-healing ability [42,43].

3.4. Biocompatibility of the MSCs-DCMMs@CS Gel

Since hydrogel dressings directly contact diabetic wounds, one of the basic properties of the MSCs-DCMMs@CS gel is good biocompatibility [42]. First, the hydrogels' hemolysis in the animal experiments was studied. As shown in Fig. S4A and Fig. S4B, the hemolysis rate of all the hydrogels was less than 1 %, which was by the ASTM biomaterial standard (HR < 5 %) [44,45], indicating that the hydrogels could safely contact blood while being applied to wounds. Next, calcein-AM/PI staining was used to evaluate the viability of the encapsulated MSCs within the hydrogel. Fig. 2I showed the relatively uniform distribution of MSCs within the hydrogel. Green fluorescence (indicating live cells) with similar intensity lasted for at least 5 days, whereas a limited number of red fluorescent dots (representing dead cells) were observed on day 3, with some increase on day 5. The results showed that most MSCs could survive for a short time after encapsulation in the hydrogels, demonstrating the good biocompatibility of the hydrogel.

During the wound healing process, endothelial and inflammatory cells play important roles in the regeneration and repair of the epidermis and internal environment, respectively. Therefore, human umbilical vein endothelial cells (HUVECs) and RAW264.7 cells were used to investigate cell viability after treatment with DCMMs@CS gel extract. MTT was used to detect cell viability, and the results revealed that after the cells were incubated with 0.1 g/mL hydrogel extracts for 48 h, the extracts did not significantly affect the viability of either HUVECs, RAW264.7 cells or L929 cells, thus guaranteeing the safety of the hydrogels (Fig. S8). L929 fibroblasts, as key cells in skin tissue, were also brought in to be co-cultured with hydrogels and calcein-AM/PI staining was used to evaluate the compatibility of MSCs-DCMMs@CS gel [37]. As shown in Fig. S9, Green fluorescence with similar intensity lasted for 3 days. On the contrast, only limited red fluorescence was observed on day 1, with a little increase on day 3. The results showed that L929 fibroblasts co-cultured with hydrogels exhibited robust growth, demonstrating the good biocompatibility of the hydrogel.

Furthermore, to verify the biosafety of the hydrogels, we collected different organs, such as the heart, liver, spleen, lungs and kidneys, after the pharmacodynamic experiments and observed the physiological and structural changes in these organs via H&E staining. The results revealed that all the hydrogel groups, including the CS Gel, MSCs@CS Gel, DCMMs@CS Gel and MSCs-DCMMs@CS Gel groups, had no obvious pathological changes or abnormalities in these organs, verifying that the CS-based hydrogels had no systemic toxicity in the diabetic mouse model. Taken together, these results indicate that MSCs-DCMMs@CS gel has good biocompatibility, which is promising for clinical applications [45].

3.5. Pharmacodynamic evaluation of the MSCs-DCMMs@CS gel

To investigate the therapeutic effect of the MSCs-DCMMs@CS gel on diabetic wounds, a diabetic mouse model was first established, following anesthesia induction, circular sections of skin lesions measuring 10 mm in diameter were created on the dorsum using a skin biopsy punch, and the CS gel, MSCs@CS gel, DCMMs@CS gel, and MSCs-DCMMs@CS gel were used as experimental groups. Mice that were not subjected to this model were used as control group. The entire process only involved one incubation, and a sterile breathable waterproof wound dressing was used for fixation to prevent the gel from falling off. Considering that active immunization occurs in the early stage of tissue injury, the wound healing process was assessed at preset times, and skin samples were collected on 3 and 7 days after treatment (Fig. 3A). The wound healing process was recorded photographically in Fig. 3B, which showed that all the hydrogel groups presented significant wound reduction and closure on day 10, while the wounds of the control group remained largely unclosed, indicating that diabetic wound covered with hydrogels positively affected wound healing and could promote wound closure [46]. Among all the groups, the MSC-DCMMs@CS Gel group presented the smallest wound area by day 10 with the wound healing rate reaching 91.10 ± 2.13 % (Fig. 3B and C). The results suggest the clear advantages of MSC-DCMMs@CS Gel in promoting wound healing.

Fig. 3.

Pharmacodynamic study of the MSCs-DCMMs@CS gel. (A) Schematic diagram of the dosing schedule. Gelation time of the hydrogels, including the CS gel, MSCs@CS gel, DCMM@CS gel and MSCs-DDM@CS gel. (B) Typical images of wounds subjected to different treatments and a schematic illustration of wound morphological changes in the different groups over 10 days. (C) Relative wound area curves of mice within 10 days after different treatments. (D) H&E staining images of tissues from wound samples at different time intervals (the width of double-headed arrow represents the width of the wound). (E) Sirius red-stained images of tissues from wound samples at different time intervals. n = 6.

Histologic analysis was used to further explore the effects on wound healing. The hematoxylin and eosin (H&E) staining results revealed that on days 3 and 7, the MSCs-DCMMs@CS Gel group presented the minimum wound width, which was consistent with the representative images of wound closure (Fig. 3D). In addition, H&E analysis revealed an increase in the amount of granulation tissue after 7 days of treatment with MSCs-DCMMs@CS Gel, suggesting that the wound healing effect of the MSCs-DCMMs@CS Gel hydrogel was significant. As an important component of skin tissue, the content and distribution of collagen fibers in regenerated tissue can be used to assess the wound healing process. Sirius red staining also revealed that, compared with the other hydrogel groups, the MSCs-DCMMs@CS Gel group presented dense and oriented collagen deposition after 7 days, which indicated that the MSCs-DCMMs@CS Gel could significantly promote the regeneration of skin tissues (Fig. 3E). Following completion of the treatment regimen, we conducted a comprehensive safety evaluation of the formulation. H&E staining results (Fig. S10) revealed no evidence of toxicity or pathological alterations in major organ systems, demonstrating favorable biosafety profiles of the hydrogel formulation.

The fact that it is difficult for diabetic patients to recover from inflammatory diseases such as diabetic foot disease is widely recognized because immune dysfunction caused by diabetes tends to cause chronic inflammation [3,4]. The use of immune-modulating biomaterials to reduce inflammation and avoid long-term diseases offers new ideas for promoting tissue repair, such as diabetic wounds, in the clinic. Pharmacodynamic evaluation of the MSCs-DCMMs@CS gel demonstrated that the designed multifunctional injectable hydrogel is desirable for clinical diabetic wound healing.

3.6. Ex vivo and in vivo antioxidant capacity of the MSCs-DCMMs@CS Gel

Since the wound healing effect of the MSCs-DCMMs@CS Gel was confirmed above, the mechanism of MSCs-DCMMs@CS Gel treating diabetic wounds was then explored. A high-glucose environment leads to excessive reactive oxygen species (ROS) generation, which can damage normal cells and lead to oxidative stress, resulting in the progression of wounds to a complex chronic inflammatory phase [47]. Therefore, the antioxidant capacity of the MSCs-DCMMs@CS gel ex vivo was evaluated by a DPPH free radical scavenging assay. DPPH, as a stable free radical, can be captured when there is a free radical scavenger present, resulting in a lighter color. As shown in Fig. 4A, the color of DPPH after treatment with the MSCs-DCMMs@CS Gel extract was intuitively the lightest, and the quantitative analysis revealed that the MSCs-DCMMs@CS Gel had the highest DPPH scavenging ratio (Fig. 4B). 2′,7′ dichlorodihydrofluorescein diacetate (DCFH-DA), an ROS probe, was used to assess intracellular ROS levels in LPS-stimulated RAW264.7 cells, which are usually used as an in vitro inflammatory model. Intracellular ROS can oxidize nonfluorescent DCFH to generate fluorescent DCF, so the detection of DCF fluorescence could be used to evaluate the level of intracellular ROS. Compared with that in the control group, the green fluorescence in the MSCs-DCMMs@CS Gel group was markedly weaker than that in all the treatment groups (p < 0.05) (Fig. 4E and F), revealing the strong antioxidant effect of the MSCs-DCMMs@CS Gel. We also evaluated the ROS scavenging ability of hydrogels in cells via a superoxide dismutase (SOD) assay kit. SOD is a metalloenzyme with antioxidant properties present in living organisms. SOD catalyzes the differentiation of superoxide anion radicals into oxygen and hydrogen peroxide, which play crucial roles in maintaining the oxidative and antioxidant balance of the organism. The results revealed that both the DCMM@CS Gel and MSCs-DCMMs@CS Gel clearly increased the expression of cellular antioxidant enzymes and directly scavenged ROS, and among all the hydrogel groups, the MSCs-DCMMs@CS Gel group presented the strongest SOD activity (Fig. S11), thereby protecting cells from oxidative stress-mediated cellular damage.

Fig. 4.

Ex vivo and in vivo antioxidant capacity of the MSCs-DCMMs@CS Gel. (A) Visual images of the DPPH scavenging test. (B) DPPH scavenging activity of the hydrogels. (C) Determination of the mitochondrial membrane potential of RAW264.7 cells labeled with JC-1 by flow cytometry. (D) Quantitative analysis of the mitochondrial membrane potential via flow cytometry. (E, F) Quantitative analysis of ROS via DCFH-DA fluorescence intensity determined by flow cytometry in RAW264.7 cells. (G) Semiquantitative analysis of HIF-1α levels by immunohistochemical staining. (H) Representative images of immunohistochemical staining of HIF-1α in wound sections on day 10. Data represent the mean ± SD, n = 3, ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001 vs the MSCs-DCMMs@CS Gel group.

The ability of the MSCs-DCMMs@CS Gel to protect mitochondria from oxidative stress damage under oxidative stress conditions was then explored to evaluate the antioxidant capacity since excessive ROS can cause mitochondrial damage. JC-1 staining was used to determine the membrane potential (Δψm) of LPS-stimulated RAW264.7 cells [34]. After LPS stimulation, the Δψm of RAW264.7 cells was significantly lower than that of control cells, indicating the disruption of mitochondrial function (Fig. 4C and D). After treatment with the hydrogels, the Δψm of all the groups increased, and the Δψm of the MSCs-DCMMs@CS Gel-treated group was significantly greater than that of the other treatment groups (p < 0.05), indicating that the MSCs-DCMMs@CS Gel had the greatest protective effect on mitochondrial function and could effectively reduce mitochondrial damage under oxidative stress.

The pathogenesis of diabetic wounds leads to a diverse microenvironment that can induce hyperglycemia, resulting in severe oxidative stress [48]. Hypoxia-inducible factor (HIF-1α) is a key factor that regulates oxidative stress in mitochondrial tissues. The expression of HIF-1α is upregulated in a hypoxia-dependent manner in the diabetic environment, so the level of oxidative stress in skin tissues is assessed by HIF-1α. As shown in Fig. 4G and H, the MSCs-DCMM@CS gel effectively ameliorated oxidative stress in diabetic wounds and decreased the level of ROS in the skin. Despite the diabetic wound environment being hypoxic, studies have shown that the function and stability of HIF-1α are impaired by the presence of ROS [49].The upregulated HIF-1α expression in the MSCs-DCMMs@CS hydrogel group indirectly reflects reduced ROS levels, which is consistent with the bidirectional regulatory relationship between HIF-1α and ROS.

3.7. Ex vivo and in vivo immunomodulation and anti-inflammatory capacity of the MSCs-DCMMs@CS Gel

Chronic inflammation at diabetic wound sites represents a critical pathological factor contributing to impaired wound healing [42]. Macrophages are responsible for the regulation of inflammation and can be divided into two phenotypes: classically activated (M1) cells and selectively activated (M2) cells. These two types of macrophages exhibit proinflammatory and anti-inflammatory properties, respectively. The balance between these two subtypes of macrophages is critical for the regulation of tissue inflammation and wound repair. First, we explored whether the MSCs-DCMMs@CS gel could regulate the macrophage repolarization of RAW264.7 cells. The flow cytometry results showed (Fig. 5A) that MSCs-DCMMs@CS Gel was able to significantly decrease the percentage of CD86-labeled M1-type macrophages compared with that in the LPS-stimulated groups (p < 0.001) and markedly increase the percentage of CD163-labeled M2-type macrophages compared with that in all the treatment groups (p < 0.001), which indicated that MSCs-DCMMs@CS Gel could modulate the macrophage phenotype into the anti-inflammatory M2 subtype in vitro most effectively.

Fig. 5.

Ex vivo and in vivo immunomodulation and anti-inflammatory capacity of the MSCs-DCMMs@CS Gel. (A) Flow cytometric analysis and quantification of CD86-labeled M1-type and CD163-labeled M2-type macrophages in RAW264.7 cells. (B) Flow cytometric analysis and quantification of CD86-labeled M1-type and CD163-labeled M2-type macrophages in skin tissues. (C) Representative LSCM images of CD86-labeled M1-type and CD163-labeled M2-type macrophages in skin tissues. (D) Representative images of immunohistochemical staining for IL-10, IL-1β and TNF-α in wound sections. Scale bar: 50 μm. Data represent the mean ± SD, n = 3, ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001 vs the +LPS/Control group.

To further illustrate the anti-inflammatory effect in vivo, skin tissues from wounds were collected after pharmacodynamic evaluation and labeled with CD86 (M1) or CD163 (M2) to determine the macrophage phenotype. The results revealed a significant decrease in M1-type macrophages and a significant increase in M2-type macrophages in the MSCs-DCMMs@CS Gel group compared with those in the control group (p < 0.05) (Fig. 5B). Additionally, immunofluorescence staining of skin tissues revealed more M2 macrophages (red fluorescence) and fewer M1 macrophages (green fluorescence) in the MSCs-DCMMs@CS Gel group than in all the treatment groups, which was consistent with the results above (Fig. 5C). The introduction of MSCs and DCMMs effectively regulated the M1/M2 ratio.

Furthermore, immunohistochemical analysis of inflammatory factors (IL-10, IL-1β and TNF-α) in wound tissue was also performed. MSCs-DCMMs@CS gel treatment significantly downregulated the expression of proinflammatory cytokines and increased the expression of anti-inflammatory mediators (Fig. 5D and Fig. S12). These results suggest that the MSCs-DCMMs@CS gel coordinates immune regulation through multiple mechanisms, including macrophage phenotype regulation and cytokine regulation, to achieve therapeutic effects.

Our latest research has demonstrated that bilirubin-modified chondroitin sulfate-mediated liposomes could significantly ameliorate acute kidney injury by regulating macrophage polarization-induced mitophagy [34]. Interestingly, to some extent, the healing process of diabetic wounds shares similar pathological mechanisms with acute kidney injury, such as the accumulation of ROS and dysregulation of proinflammatory cytokines. The polarization of macrophages is also crucial for the healing process of diabetic wounds, which is similar to acute kidney injury. The imbalance of M1/M2 macrophages during the course of diabetic influence ROS levels, and alleviating oxidative stress by scavenging ROS could reduce the inflammatory state of macrophages and exhibit anti-inflammatory properties in return [50,51]. Liu et al. reported Glycopeptide-based multifunctional nanofibrous hydrogel had the ability to alleviate oxidative stress by scavenging ROS, thereby reducing the inflammatory state of macrophages and exhibiting anti-inflammatory properties [52]. Similarly, our research showed that MSCs-DCMMs@Gel could directly scavenge ROS and reduce mitochondrial damage under oxidative stress in macrophages, thereby regulating the M1-to-M2 phenotype switch and pro-inflammatory cytokines in the diabetic wound sites.

3.8. Angiogenesis-promoting ability of the MSCs-DCMMs@CS Gel

The integrity of blood vessels and neovascularization are essential for normal wound healing [4]. To assess the angiogenic effects of MSCs-DCMMs@CS Gel, we continued to perform immunofluorescence and immunohistochemical analyses of several key markers associated with angiogenesis and extracellular matrix (ECM) remodeling. As shown in Fig. 6A, immunofluorescence staining results showed that α-SMA (mature vascular marker) and CD31 (endothelial cell marker) expression were significantly increased in the MSCs-DCMMs@CS Gel treatment group compared with the control group [53]. The results of semi-quantitative analysis showed (Fig. 6B and C) that α-SMA and CD31 levels were significantly increased in the MSCs-DCMMs@CS Gel treatment group (p < 0.001), which promoted angiogenesis.

Fig. 6.

Angiogenesis-promoting ability of the MSCs-DCMMs@CS Gel. (A) Representative LSCM images of α-SMA and CD31 immunofluorescence staining in diabetic wounds. (B) Semiquantitative analysis of α-SMA levels via immunofluorescence staining. (C) Semiquantitative analysis of CD31 levels via immunofluorescence staining. (D) Representative images of immunofluorescent Col I/Col III staining and immunohistochemical staining of VEGF in wound sections. (E, F) Semiquantitative analysis of Col I and Col III levels via immunofluorescence staining. (G) Semiquantitative analysis of VEGF levels via immunohistochemical staining. Scale bar: 20 μm. Data represent the mean ± SD, n = 3, ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001 vs the MSCs-DCMMs@CS Gel group.

Furthermore, we evaluated the expression of collagen I (Col I) and collagen III (Col III), which are critical components of the ECM [54]. In the MSCs-DCMMs@CS Gel group, the expression of collagen I (Col I) was markedly elevated, whereas the expression of collagen III (Col III) was notably reduced, as demonstrated by immunofluorescence staining (Fig. 6D) and semiquantitative analysis (Fig. 6E and F). The upregulation of Col I and downregulation of Col III suggest enhanced extracellular matrix (ECM) deposition and progressive maturation of skin tissue. In addition, the results of VEGF immunohistochemical staining (Fig. 6D–G) also revealed that VEGF levels in the MSCs-DCMMs@CS Gel treatment group were significantly increased (p < 0.01), which further confirmed the ability of the gel to promote angiogenesis by activating VEGF signaling pathways [55].

These results suggested that MSCs-DCMMs@CS Gel promoted angiogenesis by increasing the expression of key vascular markers (α-SMA and CD31), enhancing ECM remodeling (Col I and Col III), and stimulating VEGF-mediated signaling pathways. This therapeutic effect may be mediated by the synergistic effect of MSCs and DCMMs, which regulate the wound microenvironment and promote the formation of vascular networks and tissue regeneration.

3.9. Proteomic analysis of the therapeutic effects of MSCs-DCMMs@CS Gel on periwound skin tissue

To better investigate the therapeutic mechanism of this gel, we investigated the therapeutic effect of MSCs-DCMMs@CS gel on the skin tissue around the wound via proteomic analysis. As shown in Fig. 7A and B, 359 proteins were downregulated and 750 proteins were upregulated in MSCs-DCMMs@CS gel group compared to control group (e.g, KRT77 and CKM were upregulated, Fabp9 and ftl1 were downregulated), indicating that the gel had a regulatory effect on protein expression. The volcano plot in Fig. 7B shows the differentially expressed proteins (red, upregulated; blue, downregulated) (p < 0.05). These findings suggest that proteins associated with inflammation, cell proliferation, and tissue repair, such as GPX5, Cstdc5, and Cstdc5, may be involved in the therapeutic mechanism of the gel. The heatmap of hierarchical cluster analysis in Fig. 7C showed different protein expression patterns among the groups, which confirmed the regulatory role of the MSCs-DCMMs@CS Gel in the tissue around the wound, indicating that it has multiple targets and multiple pathways. In addition, we also performed GO and KEGG pathway enrichment analyses. The expression of differentially expressed proteins increased during extracellular matrix organization, inflammation regulation, cell proliferation, and apoptosis, resulting in improved antioxidant activity and peroxidase activity (Fig. 7D and E). The KEGG analysis results (Fig. 7F and G) revealed associations with extracellular matrix-receptor interactions and the PI3K-Akt and TNF signaling pathways. These results suggest that the MSCs-DCMMs@CS Gel can regulate the skin tissue around the wound to promote wound healing.

Fig. 7.

Proteomic Analysis of the Therapeutic Effects of MSCs-DCMMs@CS Gel on Peri-Wound Skin Tissue. (A) Differential statistics of upregulated and downregulated proteins. (B) Volcano plots of important proteomic results (p value < 0.05, absolute fold change (FC) > 2). (C) Hierarchical clustering tree heatmap of the differentially expressed proteins. (D) Control.vs. MDC group GO analyses of molecular function (MF)-enriched bubble plots (top 20). (E) Control.vs. MDC group GO analyses of molecular function (MF) functional circular maps (top 10). (F) KEGG pathway enrichment bubble plot of whole protein (top 20) analysis. (G) KEGG pathway enrichment histogram of the results of the total protein (top 30) analysis, n = 3.

3.10. Limitations

While this study demonstrates the potential therapeutic effects of MSCs-DCMMs@Gel on diabetic wound healing in mice, several limitations should be acknowledged. Firstly, the diabetic wound model was established using streptozotocin (STZ)-induced C57BL/6J mice. Although the model is widely used, it has inherent differences from human type 1 or type 2 diabetes. For example, mice have a faster wound healing rate compared to humans. These disparities limit the direct translation of the gel dressing's efficacy from mice to human diabetic patients. Secondly, the observation period of wound healing was only 10 days, mainly focusing on the acute wound healing process, and it was unable to assess long-term outcomes such as scar formation or recurrence. These long-term data are critical for determining the gel's clinical applicability. Moreover, although proteomics analysis identified enriched pathways, the specific molecular targets mediating the therapeutic effect of MSCs-DCMMs@CS Gel have not been verified yet, and further confirmation of causal relationships through gene knockdown or overexpression experiments is needed. Despite these limitations, our findings still provide a foundational basis for the gel dressing's potential in diabetic wound treatment.

4. Conclusion

In conclusion, we have successfully developed an engineered MSCs-laden chondroitin sulfate hydrogel (MSCs-DCMMs@CS Gel) that promotes diabetic wound healing. The system integrates chondroitin sulfate-polyethylene glycol hydrogel with ROS-scavenging, CD44-targeting micellar backpacks, enabling sustained release of MSCs-DCMMs at the wound site. In vitro results showed MSCs-DCMMs@CS Gel owned the ability of ROS scavenging and regulation of macrophage polarization, while in vivo studies confirmed promoted wound healing, reduced inflammation, and enhanced angiogenesis. Overall, this innovative hydrogel system that encapsulates surface-engineered MSCs (MSCs-DCMMs) represents a promising and comprehensive strategy as a wound dressing for diabetic wound treatment, with substantial potential for translation into clinical applications.

CRediT authorship contribution statement

Xi Cao: Writing – original draft, Visualization, Funding acquisition, Conceptualization. Ziqi Shen: Writing – original draft, Visualization, Conceptualization. Xiaohua Wang: Software, Investigation. Yuanyuan Chu: Software, Investigation. Runkong Wang: Software, Data curation. Liyang Zhu: Software, Data curation. Ruixue Zhong: Software, Data curation. Zhong Zhang: Software, Investigation, Formal analysis. Mingquan Wu: Software, Investigation, Data curation. Xu Zhou: Writing – review & editing, Supervision, Investigation, Formal analysis. Lei Zhang: Supervision, Resources, Funding acquisition, Conceptualization.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

The following is the Supplementary data to this article:

Data availability

Data will be made available on request.