Multifunctional chito-oligosaccharide-and deferoxamine-modified GelMA-OHA hydrogel dressing with antibacterial, antioxidant, and pro-angiogenic activities for infected wound healing

Abstract

Infected wounds resist healing due to multidrug-resistant bacterial colonization and persistent inflammation, which render conventional antibiotic therapies largely ineffective. Consequently, non-antibiotic strategies that coordinate infection control with tissue regeneration are urgently required. We design a dual-crosslinked hydrogel composed of gelatin methacryloyl (GelMA) and oxidized hyaluronic acid (OHA), further functionalized with chitosan oligosaccharide (COS) and deferoxamine (DFO), yielding the multifunctional GOCD hydrogel. Dynamic Schiff base linkages confer controlled release of COS and DFO, enabling coordinated antibacterial, antioxidant, and pro-regenerative activities. Experiments show that the GOCD hydrogel achieves an antibacterial rate of 88.03% against methicillin-resistant Staphylococcus aureus (MRSA). Meanwhile, its DPPH and ABTS radical-scavenging rates are 77.92% and 89.44%, respectively. Moreover, the hydrogel upregulates Nrf2 expression, effectively scavenges intracellular reactive oxygen species (ROS), and significantly enhances the migration of human umbilical vein endothelial cells (HUVECs) and tube-forming capacity. In infected rat wounds, GOCD hydrogel treatment leads to a wound closure rate of 92.97% by day 14, accompanied by organized collagen deposition and a significant increase in vascular density, with no observed systemic toxicity. Collectively, the GOCD hydrogel constitutes a safe and effective non-antibiotic healing platform that coordinates antibacterial, antioxidant, and pro-angiogenic functions for the repair of infected wounds.

Schematic diagram of GOCD hydrogel synthesis and its promotion of skin lesion healing.

1. Introduction

The skin functions as the body's main shield, defending against microbial invasion and mechanical insults while regulating fluid balance, body temperature, sensory perception, and immune surveillance.¹ Trauma, burns, infection, or metabolic disorders compromise this barrier, rendering wounds susceptible to microorganisms such as Staphylococcus aureus and Pseudomonas aeruginosa, which adhere, proliferate, and form biofilms, sustaining inflammation and impairing tissue regeneration.^2,3 Infected wounds are characterized by excessive levels of ROS, dysregulated macrophage polarization, local hypoxia, impaired angiogenesis, and defective extracellular matrix remodeling, collectively maintaining a chronic inflammatory state.^4,5 While antibiotic therapy, mechanical debridement, and traditional dressings can alleviate infection or provide physical protection, these approaches often fail to regulate the complex wound microenvironment comprehensively, constrained by their singular mode of action, frequent replacement needs, inadequate drug penetration, and risk of inducing bacterial resistance, highlighting the need for advanced strategies that simultaneously control infection, inflammation, and tissue repair.^6–10

Recent material engineering efforts have sought to overcome these limitations. Natural polysaccharide hydrogels, metal ions, nanoparticles, and antioxidants have been incorporated to counter infection, oxidative stress, and impaired angiogenesis.^11–14 Despite these advances, such systems are frequently hampered by insufficient mechanical strength, a separation of antibacterial and pro-angiogenic functions, potential biocompatibility concerns regarding residual metal ions, and complex, variable fabrication processes. Therefore, achieving integrated regulation of antibacterial, anti-inflammatory, antioxidant, and angiogenic effects in infected wounds remains a significant challenge.^15–17 In this context, GelMA and hyaluronic acid (HA) derivatives provide an ideal foundation for constructing robust multifunctional systems. GelMA contains RGD motifs that promote cell adhesion, migration, and proliferation. Furthermore, its photocrosslinkable nature enables precise control over architecture, mechanical properties, and shape adaptability.^18,19 HA regulates cell motility, inflammation, and tissue hydration.²⁰ Its oxidized form, OHA, introduces aldehyde groups that form reversible Schiff base linkages with amino groups, conferring self-adhesive, self-healing, and dynamically stable networks suitable for irregular and highly exudative wounds.²¹ Through the combination of GelMA's tunable structural framework with OHA's dynamic reversibility, multifunctional platforms can be designed to incorporate diverse bioactive molecules and holistically regulate the complex wound microenvironment.

Furthermore, multiple bioactive molecules can synergistically modulate the pathological microenvironment of infected wounds.²² COS has garnered significant attention for this purpose due to its specific physicochemical features and bioactivities. COS is produced by the enzymatic or chemical degradation of chitin, typically exhibiting a molecular weight below 5 kDa and retaining the β-1,4-linked d-glucosamine backbone of chitosan.²³ Compared to high-molecular-weight chitosan, COS offers superior water solubility, pH responsiveness, and antibacterial functionality, making it a preferred material for forming biocomposite hydrogels with potent antibacterial activity.²⁴ DFO, an iron chelator with potent angiogenic and antioxidant effects, also shows significant promise in wound therapy by alleviating hypoxia and reducing oxidative stress.^25,26 However, both COS and DFO face limitations such as short in vivo retention and difficulty in maintaining effective local concentrations, which necessitate the development of controlled-release carriers to translate their therapeutic potential into scalable and reproducible clinical outcomes.

Here, we propose a dual-crosslinked GelMA/OHA hydrogel incorporating COS and DFO to modulate the wound microenvironment. By integrating GelMA's tunable structural framework with OHA's dynamic reversibility, this hydrogel collectively enhances mechanical stability, tissue adhesion, antibacterial activity, inflammation modulation, and angiogenesis. This rationally designed platform offers a multifunctional strategy for the comprehensive repair of infected wounds, presenting a promising approach to accelerate tissue regeneration.

2. Materials and methods

2.1. Materials

Sodium hyaluronate, ethylene glycol, and COS were bought from Shanghai Macklin Biochemical Technology Co., Ltd Deferoxamine and sodium periodate were obtained from Shanghai Aladdin Biochemical Technology Co., Ltd Gelatin Methacryloyl (GelMA, EFL-GM-60) was purchased from EFL-Tech (SuZhou) Co., Ltd DMEM and fetal bovine serum (FBS) were purchased from Gibco (USA). The CCK-8 assay was sourced from Shanghai Taoshu Biotechnology Co., Ltd The ROS assay kit, live/dead cell staining kit (Calcein-AM/PI), hematoxylin and eosin (H&E) staining kit, and DAPI staining solution were procured from Beyotime Biotechnology (China). Primary and secondary antibodies for immunohistochemistry and immunofluorescence were purchased from ABclonal Technology Co., Ltd (China).

2.2. Hydrogel synthesis and characterization

2.2.1. Hydrogel preparation

Following previously reported methods,²⁷ OHA was first prepared. Briefly, HA (1 g) was dissolved in 100 mL of deionized water, and then sodium periodate (0.85 g) was added. to the solution, and the mixture was reacted in the dark at room temperature for 24 h and then quenched with 2 mL of ethylene glycol for 1 h. The reactants were dialyzed against deionized water using a dialysis membrane with a molecular weight cut-off of 10 000 Da for 4 days. The resulting solution was freeze-dried to yield white, fluffy OHA. The synthesis of OHA was verified through Fourier-transform infrared (FTIR) and proton nuclear magnetic resonance (¹H NMR) analyses.

GelMA and OHA solids were accurately weighed and dissolved in a 0.25 wt% LAP solution, yielding working concentrations of 10% (w/v) and 1% (w/v), respectively. COS or DFO was then incorporated into the hydrogel precursor solution. Based on antibacterial assays and CCK-8 cytotoxicity results (Fig. S1 and S2), the final concentrations were set at 3 mg mL⁻¹ for COS and 30 mg mL⁻¹ for DFO. The mixture was cast into molds and exposed to ultraviolet light (365 nm, 20 mW cm⁻², 1 cm distance) for 30 s to induce crosslinking, forming the hydrogel. The hydrogel extract was obtained by incubating the hydrogel with cell and bacteria culture medium at 37 °C for 24 h.

2.2.2. Hydrogel characterization

Lyophilized hydrogel samples were characterized to assess their chemical composition, morphology, and mechanical properties. For FTIR analysis, freeze-dried hydrogels were ground with KBr (1 : 100, w/w) and pressed into pellets. Spectra were recorded over 400–4000 cm⁻¹ with a resolution of 4 cm⁻¹ after background correction. The hydrogel microstructure was examined by scanning electron microscopy (SEM). Lyophilized specimens were sectioned to size, affixed to conductive tape, and coated with a 5–10 nm gold layer prior to observation of their surface and internal porous structure. For compressive mechanical testing, cylindrical hydrogel samples were first equilibrated in phosphate-buffered saline (PBS) at 37 °C for 24 h. Each sample was then placed between the plates of a mechanical testing system fitted with a 0.1 N load cell and compressed uniaxially at a constant rate of 3 mm min⁻¹ until either a predetermined strain was achieved or the sample failed. Load and displacement data were recorded in real time to calculate stress (σ = F/A) and strain (ε = ΔL/L₀), from which stress–strain curves were generated. Each measurement was performed in triplicate.

2.2.3. Self-healing and adhesion properties

The hydrogels' self-healing performance was evaluated with two sets of circular samples, one containing rhodamine 101 and the other unlabeled. Each hydrogel was cut in half, and semicircular halves from different sets were recombined to monitor the healing process. To evaluate adhesion, the hydrogel precursor solution was then coated onto fresh pig skin that had been pre-wetted with PBS. Following 30 s of UV curing, the resulting hydrogel was folded 180° and twisted 360° to assess its adhesion strength and deformation resistance.

2.2.4. Swelling behavior

The initial weight (W₀) of each hydrogel sample was recorded prior to immersion in PBS at 37 °C. At designated time points, samples were removed, gently blotted with filter paper, and weighed (W_t). Each measurement was repeated three times. The swelling percentage (%) was calculated as follows:Swelling percentage (%) = (W_t − W₀)/W₀ × 100%

2.2.5. Degradation behavior

Hydrogels were equilibrated to their swelling state, and their initial mass (W_i) was measured. The samples were then maintained at 37 °C in PBS supplemented with 1 U mL⁻¹ collagenase. At predetermined intervals, the hydrogels were retrieved, surface moisture was blotted with filter paper, and their remaining mass (W_t) was determined. Each time point was assessed in triplicate. Degradation percentage was calculated according to the formula below:Degradation rate (%) = (W_i − W_t)/W_i × 100%

2.2.6. In vitro drug release

GOCD hydrogels were incubated in 2 mL (V_r) of deionized water at 37 °C. At scheduled intervals, 200 µL (V_s) of supernatant was removed and replenished with an equal amount of fresh deionized water. DFO quantitative method: Standard solutions for DFO quantification were prepared at concentrations of 0, 5, 10, 25, 50, 100, 150, 200, and 250 µg mL⁻¹ and mixed with an equal volume of 6 mM ferric chloride solution. After incubating at room temperature for 30 min, absorbance at 485 nm was measured against a deionized water blank to establish a calibration curve. COS quantitative method: Standard solutions of 0, 400, 800, 1200, 1600, and 2000 µg mL⁻¹ were prepared. These solutions were combined with DNS reagent in a 1 : 2 volume ratio, maintained in boiling water for 5 min, and then cooled in an ice-water bath for 2 min to stop the reaction. Absorbance at 517 nm was measured against a deionized water blank to generate a corresponding standard curve. Cumulative release rate calculation: The concentrations of DFO and COS in the sampled aliquots were quantified using their respective standard curves, and cumulative release profiles were subsequently plotted.

C _n : measured supernatant drug concentration at the n-th sampling time-point; C_i: drug concentrations from all preceding sampling points; m_total: total loaded amount of DFO or COS in GOCD hydrogels.

2.2.7. In vitro antioxidant activity

DPPH Assay: DPPH radical scavenging activity was assessed by mixing hydrogel precursor solution with DPPH solution (final concentration 2.5 mmol L⁻¹) and incubating in the dark for 30 min at room temperature. The absorbance was measured at 517 nm.DPPH scavenging (%) = (A₀ − A₁)/A₀ × 100%A₀: the absorbance value of the control group; A₁: the absorbance value of the experimental group.

ABTS Assay: ABTS⁺ solution was generated by mixing 7.4 mM ABTS with 2.6 mM potassium persulfate (1 : 1, v/v) and incubating in the dark for 12 h at room temperature. Samples (0.5 mL) were mixed with ABTS⁺ solution (4 mL) for 6 min, and absorbance was measured at 734 nm in triplicate.ABTS scavenging (%) = (A₀ − A₁)/A₀ × 100%A₀ is the absorbance value of the control group and A₁ is the absorbance value of the experimental group.

2.3. In vitro biocompatibility

Mouse connective tissue fibroblast cells (L929) were obtained from Pronosai company (Wuhan, China). Human umbilical vein endothelial cells (HUVECs) were acquired from Cell Bank/Stem Cell Bank, CEMCS, CAS. Unless otherwise specified, the experimental groups were designated as follows: CON, GO (GelMA-OHA), GOC (GelMA-OHA + COS), and GOCD (GelMA-OHA + COS + DFO).

2.3.1. CCK-8 assay for cell viability

L929 cells and HUVECs were seeded into 96-well plates at a density of 5 × 10³ cells per well and incubated overnight. The blank control group received complete culture medium, while the experimental groups were treated with 100 µL of hydrogel extract. Cells were then cultured for 1, 3, or 5 days. At each time point, 10 µL of CCK-8 reagent was added to each well, and the plates were incubated at 37 °C in the dark for 2 h. Absorbance was subsequently measured at 450 nm. Experiments were conducted in triplicate.

2.3.2. Live/dead cell staining

L929 cells and HUVECs were dispensed into 24-well plates at 1 × 10⁵ cells per well and incubated at 37 °C overnight. The culture medium was subsequently replaced with the respective hydrogel extracts, and the cells were incubated for a further 1, 3, or 5 days. At the end of every incubation interval, the medium was aspirated, followed by rinsing the cells with PBS. The cells were then stained with 100 µL of diluted Calcein-AM/PI solution per well at 37 °C for 30 min in the dark. After staining, the cells were washed with PBS and imaged.

2.3.3. Hemolysis assay

Fresh whole blood was collected from rats into EDTA-anticoagulant tubes. Red blood cells (RBCs) were isolated by subjecting the samples to centrifugation at 300×g for 15 min and were then washed 2–3 times with PBS until the supernatant appeared clear. The washed RBCs were resuspended in saline to a final 5% (v/v) concentration. 0.2 mL of each hydrogel extract solution was combined with 1.8 mL of the RBC suspension in a 5 mL tube and incubated at 37 °C on a shaker for 1 h. Negative and positive controls consisted of 0.2 mL of saline and 0.2 mL of 1% Triton X-100, respectively. Following incubation, the samples were mixed thoroughly, transferred to 2 mL tubes, and centrifuged at 1000×g for 5 min. The supernatant was photographed, and the absorbance was measured at 540 nm. Hemolysis (%) was calculated using the following formula:Hemolysis (%) = (A_s − A_b)/(A_c − A_b) × 100%where A_s, A_c, and A_b represent the absorbance values of the hydrogel group, the 1% Triton X-100 group, and the NaCl group, respectively.

2.4. Intracellular antioxidant activity

2.4.1. DCFH-DA staining

L929 cells in good growth condition were seeded into 24-well plates at a density of 1 × 10⁵ cells per well and allowed to attach to the plate surface overnight. The experimental groups were designated as follows: a blank control cultured in complete DMEM for 6 h; an LPS-stimulated group incubated with DMEM containing 1 mg mL⁻¹ LPS for 6 h; and hydrogel-treated groups incubated with different hydrogel extracts in the presence of 1 mg mL⁻¹ LPS for 6 h. After treatment, the cells were gently washed twice with PBS and incubated with 1 mL of 10 µM DCFH-DA working solution at 37 °C in the dark for 30 min. The dye solution was then removed, the cells were washed twice with PBS, and fluorescence images were captured using a fluorescence microscope.

2.4.2. Nrf2 immunofluorescence staining

L929 cells were seeded on coverslips in 24-well plates at a density of 1 × 10⁵ cells per well and cultured overnight. The processing conditions were the same as in 2.4.1. After medium aspiration, cells were washed with PBS, fixed in 4% paraformaldehyde for 15 min at room temperature, rinsed, permeabilized with 0.2% Triton X-100 (5 min) and given three PBS washes. Cells were blocked with 3% BSA for 60 min at room temperature, followed by overnight incubation at 4 °C with first antibody diluted 1 : 200 in a humidified chamber. Following three PBS washes, samples were treated with a fluorescent secondary antibody and subjected to incubation for 1 h at room temperature in darkness. The nucleus was stained with DAPI for 5 min, and the samples received three final PBS washes. Finally, the slides were sealed with an anti-fluorescence quencher. Nrf2-positive cells were quantified by fluorescence microscopy.

2.4.3. RT-qPCR

L929 cells were cultured in 6-well plates and allowed to attach overnight. After treatment, the cells were lysed with Trizol reagent for total RNA extraction. Reverse transcription was conducted according to the manufacturer's guidelines. The 2^−ΔΔCt method was applied to determine relative mRNA levels, normalized to GAPDH. Corresponding primer sequences are shown in SI Table S1.

2.5. Antibacterial activity

2.5.1. Dilution plate assay

S. aureus and MRSA strains were used at a concentration of 1 × 10⁶ CFU mL⁻¹ and co-incubated with hydrogel extracts in culture medium at 37 °C for 24 h. Following incubation, 50 µL of the bacterial suspension was collected, diluted 100-fold, and spread evenly onto the center of an agar plate. Sterile glass beads were added, and the plate was gently shaken horizontally to distribute the suspension uniformly. The plates were subsequently incubated at 37 °C for 24 h before bacterial colony growth was observed and documented.

2.5.2. SEM imaging of bacteria

S. aureus and MRSA suspensions (1 × 10⁶ CFU mL⁻¹) were co-cultured with hydrogel extracts in medium at 37 °C for 24 h, using a medium-only control. After incubation, the medium was aspirated, and the samples were gently washed with PBS. Bacterial fixation was performed using 2.5% glutaraldehyde for 40 min, after which the samples were washed three times with PBS for 5 min per wash. Dehydration was carried out through a graded ethanol series (30%, 50%, 70%, 90%, and 100%), incubating for 10 min at each concentration. A 20 µL aliquot of the prepared suspension was deposited onto a silicon wafer and allowed to air-dry. Finally, the wafer was sputter-coated with gold for 60 s at 10 mA to achieve a layer of approximately 5 nm prior to scanning electron microscopy.

2.5.3. MRSA biofilm assay

A bacterial suspension (1 × 10⁶ CFU mL⁻¹) was mixed with hydrogel extracts and transferred to 12-well plates (500 µL per well), with untreated wells serving as controls. Following 24 h of incubation at 37 °C to allow biofilm formation, the planktonic bacteria and medium were aspirated. After gentle washing with PBS, the biofilms were fixed in 500 µL paraformaldehyde for 20 min, washed with PBS, and incubated with 500 µL of 0.1% crystal violet in the dark for 15 min. After removing excess dye, the wells were repeatedly rinsed with PBS until the wash solution was clear. The plates were air-dried and photographed. For biomass quantification, 1 mL of 95% ethanol was added to each well and incubated for 5 min and thoroughly mixed. The solution was transferred to a 96-well plate, and absorbance at 570 nm was determined with a microplate reader.

2.6. Migration and angiogenic activity

2.6.1. Cell migration assay

Healthy HUVECs were plated in 6-well plates at a density of 2 × 10⁵ cells per well and maintained in complete DMEM until they reached near-confluence (90–100%). A sterile 200 µL pipette tip was used to create a straight scratch across the well bottom. Following PBS washing to eliminate detached cells, the medium was replaced with serum-free DMEM containing hydrogel extracts. The same field was imaged at 0 h and 24 h with an inverted microscope. The cell migration rate was quantified with ImageJ software using the following formula:Migration rate (%) = (A₀ − A_t)/A₀ × 100%A₀ is the initial area at 0 h, and A_t corresponds to the area at time t.

2.6.2. Assessment of tube formation

After thawing overnight at 4 °C, 20 µL of Matrigel was added to each well of a 24-well plate carefully to avoid bubble formation. The plate was then incubated at 37 °C under 5% CO₂ for 30–60 min to allow the Matrigel to solidify. HUVECs (1 × 10⁵ cells/well) were treated with hydrogel extracts corresponding to each experimental group. After 6 h, the tube-like structures were observed. Cells were stained with Calcein-AM and kept in the dark for 30–60 min and subsequently washed three times with PBS. Fluorescence images were captured, and tube formation was quantified using ImageJ software.

2.6.3. RT-qPCR for angiogenesis-related genes

To evaluate angiogenic potential, HUVECs were seeded in 6-well plates at a density of 3 × 105 cells per well and allowed to adhere overnight. The cells were then treated for 24 h, with the blank control group receiving complete DMEM and the hydrogel-treated groups receiving various hydrogel extracts in DMEM. RNA was extracted from the processed cells using the same procedure as described in 2.4.3. Corresponding primer sequences are shown in SI Table S2.

2.7. In vivo wound healing assay

The wound healing capacity of the hydrogels was assessed in healthy SD rats with a full-thickness infected skin injury model. All animal experiments were approved by the Institutional Animal Care and Use Committee of Harbin Medical University (Approval No. SYDW2025-144). Twenty-four rats were randomly assigned to four treatment groups: a blank control, a GO hydrogel group, a GOC hydrogel group, and a GOCD hydrogel group. After shaving and disinfecting the dorsal area, a circular full-thickness skin defect with a diameter of 1 cm was surgically created. Each wound was inoculated with an MRSA suspension to a final concentration of 1 × 10⁶ CFU mL⁻¹. Following infection, the wounds exhibited marked erythema, edema, exudation, and purulent discharge, indicating successful establishment of the infected wound model. We photographed the wounds on days 0, 3, 7, and 14 to calculate the wound closure rates. On days 7 and 14, skin tissues were collected, fixed, dehydrated through a graded ethanol series, and embedded in paraffin for sectioning. Sections were stained with Masson, hematoxylin and eosin (H&E), and immunohistochemistry was performed for markers including interleukin-6 (IL-6) and cluster of differentiation 31 (CD31). To evaluate potential systemic toxicity, the major organs—the heart, liver, spleen, lung, and kidney—were also harvested for histopathological analysis.

2.8. Statistical analysis

Statistical analysis was performed using GraphPad Prism software. For inter-group difference analysis, an independent samples t-test was used for comparison between two groups, one-way ANOVA or two-way ANOVA followed by Tukey's post-hoc test for multiple comparisons among multiple groups. Data are presented as mean ± standard deviation (SD). Statistical significance was defined as P < 0.05 (*P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001).

3. Results and discussion

3.1. Physical and chemical properties of hydrogels

The synthesis of OHA and GOCD hydrogels is depicted in Scheme 1. FTIR spectroscopy (Fig. 1A) showed a characteristic peak at 1731 cm⁻¹, which is attributed to the C O stretching vibration of the aldehyde group.²⁸ Analysis of the ¹H NMR spectrum (Fig. 1B) clearly revealed the proton signals of the sugar ring backbone (3.0–4.5 ppm) and the N-acetyl methyl singlet (δ ≈ 2.0 ppm), demonstrating that the oxidation process left the main chemical structure of HA intact. New proton signals in the OHA spectrum also emerged in the 4.8–5.2 ppm region, arising from the aldehyde groups generated at the C2/C3 positions and their hydrated or hemiacetal forms.²⁹ These signals further confirm the successful introduction of aldehyde groups, which exist in dynamic equilibrium. The FTIR spectra in Fig. 1C showed that the GO, GOC, and GOCD hydrogels retained the principal structural features of GelMA and OHA, including the O–H/N–H stretching vibration near 3300 cm⁻¹, the amide I band at 1630 cm⁻¹, and the C–O–C stretching vibration near 1030 cm⁻¹. The disappearance of the peak at 1731 cm⁻¹ after gelation suggests that the aldehyde groups underwent condensation reactions with amino groups and participated in the crosslinking process.³⁰ For GOC, the enhanced absorption in the 3300 cm⁻¹ region relative to GO suggests that incorporating COS strengthened hydrogen bonding interactions.³¹ In GOCD, the slight shift of the amide I band and the altered peak shape in the O–H/N–H region suggest the potential participation of DFO hydroxamic acid groups in non-covalent interactions within the hydrogel network, which may contribute to a more homogeneous internal structure among the multiple components.³² The stress–strain curves for the GO, GOC, and GOCD hydrogels are shown in Fig. 1D. All hydrogels displayed a characteristic nonlinear stress increase with strain, consistent with a network structure progressively resisting deformation.³³ The GOCD hydrogel exhibited the highest maximum compressive stress (63.34 kPa), compared with GOC (50.97 kPa) and GO (29.52 kPa), indicating that incorporating COS and DFO enhanced network compactness and mechanical strength. This enhanced load-bearing capacity allowed the structure to remain stable under relatively high strain. Macroscopic morphology observations (Fig. 1E) revealed that both the GO and GOCD hydrogels were uniformly fluid prior to photocuring, demonstrating injectability. Following 30 s of light exposure, they solidified rapidly, confirming a sufficient crosslinking reaction and a short gelation time potentially suitable for rapid shaping and conformal coverage of clinical wounds. Fig. 1F shows that the GOCD hydrogel maintained conformal contact with moist porcine skin without obvious detachment after bending or twisting, demonstrating its stable attachment to tissue surfaces. This dual functionality helps resist exogenous bacterial invasion and maintain wound microenvironment homeostasis.

Scheme 1. Schematic diagram of GOCD hydrogel synthesis and its promotion of skin lesion healing.

Fig. 1. (A) FTIR spectra of HA and OHA, and (B) ¹H NMR spectra confirming HA oxidation. (C) FTIR spectra of GO, GOC, and GOCD hydrogels. (D) Stress–strain curves and (E) images of the hydrogel before and after photocuring. (F) Adhesion performance of the hydrogel. (G–I) Self-healing behavior of GOCD.

Hydrogel self-healing and interfacial adhesion characteristics are critical for wound management.³⁴ As shown in Fig. 1G, the cross-section of the hydrogel re-bonds within two hours, demonstrating that dynamic imine bonds rapidly reorganize after fracture to repair the network microscopically. This dynamic interaction improves hydrogel durability under mechanical disturbance.³⁵ Rheological tests further confirmed the favorable mechanical resilience and self-healing potential of the hydrogel. Amplitude sweep measurements identified a critical rupture strain of approximately 100%, indicating that the hydrogel can tolerate large-magnitude deformation before the onset of gel-to-sol transition (Fig. 1H). Moreover, step-strain cyclic tests confirmed that the storage modulus could fully recover to its initial state after multiple damage–recovery cycles, indicating highly reversible dynamic network reconstruction and superior self-healing performance (Fig. 1I).

SEM images of the lyophilized hydrogels (Fig. 2A) revealed a porous, interconnected architecture across all groups. This indicates that photo-crosslinking and Schiff base crosslinking cooperatively formed a stable three-dimensional network. Quantitative image analysis showed that the porosity decreased progressively from 44.56% (GO) to 40.21% (GOC), 30.21% (GOD), and 28.46% (GOCD). These pore channels can absorb wound exudate and provide favorable pathways for nutrient diffusion and cell infiltration, thereby facilitating regulation of the early wound microenvironment during repair.³⁶ Swelling properties influence drug diffusion and absorption capacity.³⁷ As shown in Fig. 2B, all three hydrogels reached swelling equilibrium in PBS (37 °C, pH 7.4) within approximately 12 h. After 48 h, the GO hydrogel exhibited the highest swelling ratio (152.40%), while the GOC (148.08%) and GOCD (140.93%) hydrogels showed slightly lower values. This reduction in swelling ratio is likely attributable to the enhanced compactness of the network structure following the incorporation of COS and DFO, which is consistent with the decreased porosity observed by SEM. This range of swelling remains appropriate, however, as it provides necessary fluid absorption capacity without the excessive expansion that could cause edge lifting or detachment in highly exudative wounds. Enzymatic degradation experiments (Fig. 2C) showed that both GOC and GOCD hydrogels degraded more slowly than GO, a result attributed to their higher crosslinking density and more compact network. This dense structure physically impedes the penetration of enzyme molecules, retarding their access to the degradable polymer backbone. Consequently, the material maintains its structural integrity longer at the wound site, which supports sustained drug release and an enduring physical barrier. The release of COS and DFO from the hydrogel followed a sustained, biphasic profile (Fig. 2D and E). COS release was relatively gradual, with cumulative rates of 16.11% at 24 h and 44.16% at 144 h, indicating that a portion of the COS was constrained within the Schiff base network, limiting its diffusion. DFO release was initially faster, reaching 19.1% by 24 h, before transitioning to a stable sustained-release phase with a cumulative release of 49.0% at 144 h. This pattern delivers an effective initial dose early while maintaining a stable concentration later, thereby supporting prolonged antibacterial, antioxidant, and pro-healing functions. As shown in Fig. 2F–I, the GOCD hydrogel exhibited high scavenging ratios against both ABTS and DPPH free radicals in chemical assays, reaching 89.44% and 77.92%, respectively, indicating potent free radical-scavenging capacity.

Fig. 2. (A) SEM images illustrating the internal microstructure. (B) Swelling behavior, and (C) enzymatic degradation profiles of the hydrogels. (D) COS release profile and (E) DFO release profile from the composite hydrogel. (F–I) ABTS and DPPH radical-scavenging capacities with corresponding experimental images. n = 3.

3.2. Biocompatibility

As a wound repair material, the hydrogel must exhibit good biocompatibility.³⁸ We systematically evaluated its biocompatibility at the cellular, blood, and in vivo levels. Cytocompatibility was assessed using L929 cells and HUVECs. Live/dead staining demonstrated that cells in the hydrogel-treated groups were predominantly viable, with no evident increase in dead cells compared to the control group (Fig. 3A and B and S3). Cell morphology showed no significant difference from the control group, confirming excellent cytocompatibility. Furthermore, cells were co-cultured with hydrogel extracts, and viability was measured indirectly via the CCK-8 assay at 450 nm absorbance. After 1, 3, and 5 days of co-culture, cell survival rates in all groups exceeded 80%, consistent with the live/dead staining results (Fig. 3C and E). For hemocompatibility, the hemolysis rate of biomedical materials must be below 5%.³⁹ We tested the hydrogels using fresh rat blood, with 1% Triton X-100 and 0.9% NaCl solution serving as positive and negative controls, respectively. The supernatants of the hydrogel-treated groups were nearly colorless (Fig. 3D), with calculated hemolysis rates of 2.02% for GO, 1.86% for GOC, and 1.84% for GOCD, all well below the safety threshold. These results indicate that the hydrogels did not induce significant hemolysis and possessed good hemocompatibility. In vivo experiments provided further confirmation (Fig. S4). Compared with the control group, H&E-stained sections of various organs showed no structural or morphological alterations under light microscopy. These findings demonstrated the biocompatibility of the hydrogel in vivo.

Fig. 3. Live/dead staining of L929 cells (A) and HUVECs (B) cultured with the hydrogels. CCK-8 assays of L929 cells (C) and HUVECs (E). (D) Hemolysis assay of the hydrogels. n = 3.

3.3. In vivo antioxidant capacity

Following skin injury, significant levels of ROS are generated locally due to the infiltration of inflammatory cells, rupture of cell membranes, and mitochondrial dysfunction.⁴⁰ This oxidative stress subsequently induces lipid peroxidation and DNA damage, which amplifies the inflammatory response.⁴¹ Consequently, the shift from inflammation to proliferation in wound healing is severely impeded.⁴²

Intracellular ROS levels were first measured with a DCFH-DA probe to evaluate the impact of different hydrogel systems on oxidative stress. As shown in Fig. 4A and B, intense green fluorescence signals in the LPS-stimulated positive control group indicated significant intracellular ROS accumulation. While the GO group showed a slight decrease relative to the LPS group, no significant difference was observed, demonstrating that the GelMA-OHA matrix alone could not substantially ameliorate oxidative stress. In comparison, the fluorescence signal of the cells in the GOC and GOCD groups showed a decreasing trend. Given the close correlation between oxidative stress and inflammatory response, we further analyzed the effects of different treatments on the inflammatory status of L929. RT-qPCR analysis (Fig. 4C and D) showed that the expression of inflammatory markers IL-6 and IL-1β were significantly reduced in the GOC and GOCD groups.

Fig. 4. (A) ROS staining and (B) corresponding quantitative analysis. (C) IL-6 expression level. (D) IL-1β expression level. n = 3.

Immunofluorescence staining (Fig. 5A, B) revealed group differences in NRF2 expression. Compared with the LPS control, the GOC and GOCD groups displayed markedly enhanced NRF2 nuclear translocation. RT-qPCR analysis further revealed (Fig. 5C and D) that the expression levels of Nrf2 and HO-1 were significantly elevated in the GOC and GOCD groups. These findings suggest that GOC and GOCD treatments may activate the Keap1-Nrf2-ARE antioxidant pathway by promoting the dissociation of Nrf2 from Keap1 and its subsequent nuclear translocation, which initiates transcription of downstream antioxidant genes such as HO-1 and ultimately enhances cellular antioxidant capacity.⁴³ In summary, the GOC and GOCD hydrogels not only scavenge free radicals efficiently in vitro but also exert significant intracellular antioxidant effects, thereby supporting their potential for mitigating inflammatory responses and accelerating tissue healing.

Fig. 5. (A) Immunofluorescent staining of NRF2 (red) and (B) its quantitative analysis. (C) Nrf2 expression level. (D) HO-1 expression level. n = 3.

3.4. Antibacterial properties

Bacteria delay wound healing by prolonging the inflammatory phase, disrupting granulation tissue formation, and elevating protease activity.¹⁷S. aureus is the primary causative agent of such infections.⁴⁴ The emergence of MRSA, with its distinct drug resistance and high pathogenicity, has made it a critical pathogen in infected wound management. Consequently, biomaterials intended for chronic wound healing require intrinsic or incorporated antibacterial functionality.⁴⁵

First, the survival of S. aureus and MRSA was assessed via plate counting. As shown in Fig. 6A and C, abundant bacterial colonies grew on the agar plates in the control and GO groups, whereas only a small number appeared in the GOC and GOCD groups. Statistical analysis of the antibacterial rates (Fig. 6F and G) revealed that both the GOC and GOCD groups exhibited rates exceeding 80%, showing a significant increase compared with the control and GO groups. Biofilms are a key factor in the persistence of infection, as their dense polysaccharide matrix can severely limit antimicrobial penetration and establish a persistent bacterial reservoir. The anti-biofilm formation assay and its quantification (Fig. 6E and H) demonstrated that GOC and GOCD significantly inhibited MRSA biofilm formation during the initial 24 h colonization period.

Fig. 6. Plate images of S. aureus (A) and corresponding antibacterial rates (F), and plate images of MRSA (C) with antibacterial rates (G). MRSA biofilm inhibition assay (E) and its quantitative analysis (H). Morphological changes of S. aureus (B) and MRSA (D) observed by SEM. n = 3.

To further elucidate the antibacterial mechanism, morphological changes in S. aureus and MRSA across treatment groups were examined by SEM. The results (Fig. 6B and D) showed that bacteria in the control and GO groups maintained smooth surfaces and intact structures. In contrast, treatment of bacteria with GOC and GOCD resulted in distinct cellular damage, including obvious shrinkage, disrupted membrane structures, and leakage of intracellular contents. These results demonstrate that COS is the primary contributor to the observed antibacterial activity, while DFO provides limited direct antibacterial effects. This is consistent with previously reported mechanisms, in which COS adsorbs onto negatively charged bacterial membrane components through electrostatic interactions, increasing membrane permeability and inducing leakage of intracellular contents.⁴⁶

3.5. Migration and angiogenesis

Angiogenesis plays a vital role in wound repair, facilitating the transport of oxygen and nutrients and regulating fibroblast growth, collagen deposition, inflammation resolution, and epithelial regeneration.⁴⁷ Consequently, assessing the pro-angiogenic capacity of a material provides a critical measure of its potential to effectively promote wound healing.

Since active migration is a prerequisite for endothelial cells to align, extend, and form tubular structures during early angiogenesis, we first conducted scratch assays to assess the effects of different treatments on cell migration. After 24 h (Fig. 7A and B), the GOCD group exhibited a significantly higher cell migration rate (62.15%) than the control group (47.88%), indicating that GOCD may improve the local microenvironment to enhance endothelial cell motility. Subsequent in vitro angiogenesis assays showed that HUVECs formed more extensive tubular networks following GOCD treatment (Fig. 7C). The mesh density increased approximately 3.12-fold relative to the control, while the numbers of segments, nodes, and branches increased by about 4.01-, 3.14-, and 1.69-fold, respectively (Fig. 7D–G), confirming that the GOCD group effectively promotes tube formation.

Fig. 7. (A) Migration images of HUVECs at 0 h and 24 h under different treatments and (B) the corresponding quantitative analysis. (C) Tube formation assay of HUVECs after different treatments, with quantitative analyses of (D) total tube meshes, (E) number of vessel segments, (F) number of nodes, and (G) number of branches. n = 3.

Moreover, immunofluorescence analysis demonstrated that the average fluorescence signal strength of the adhesion molecule CD31 was markedly higher in the GOCD group than in the control (Fig. 8A and B). CD31, a key marker of endothelial adhesion and angiogenesis, is associated with lumen stabilization and vascular network maturation.⁴⁸ Collectively, these data indicate that the GOCD group may promote angiogenesis by modulating the HIF-1α/VEGF/PDGF axis and upregulating key molecules like CD31, providing a mechanistic basis for its potential in wound healing. RT-qPCR analysis further validated these findings at the transcriptional level. The expression levels of VEGF, platelet-derived growth factor (PDGF), and hypoxia-inducible factor-1α (HIF-1α) were upregulated in the GOCD group by 1.51-, 2.89-, and 1.28-fold, respectively, compared to the control group (Fig. 8C–E). This upregulation likely stems from DFO-induced stabilization of HIF-1α, which activates downstream pro-angiogenic genes.⁴⁹

Fig. 8. (A) Immunofluorescence staining of CD31 and (B) its quantitative analysis. (C) VEGF expression level. (D) PDGF expression level. (E) HIF-1α expression level. n = 3.

3.6. In vivo healing

Following a comprehensive evaluation of its multifunctional properties, the GOCD hydrogel was tested in a full-thickness infected skin defect model (1 cm diameter) in SD rats to assess its in vivo performance, biosafety, and therapeutic potential. Representative wound images from each group at various time points are shown in Fig. 9A. Although the wound area decreased over time in all groups, healing rates differed significantly. As shown in Fig. 9B, on day 7, the wound healing rates were 46.93%, 51.65%, 67.77%, and 78.08% for the control, GO, GOC, and GOCD groups, respectively. By day 14, healing in the GOCD group had further increased to 92.97%, a rate significantly higher than all other groups, indicating that the hydrogel effectively modulates the complex wound microenvironment to promote closure. This enhanced healing likely results from the synergistic antibacterial, antioxidant, and pro-angiogenic functions of the GOCD hydrogel. Histological analysis corroborated these results. H&E and Mssson staining (Fig. 9C and 10A) showed that on day 7, wounds in the control and GO groups had an incomplete epidermis, disorganized tissue, loosely arranged collagen fibers, and substantial neutrophil and lymphocyte infiltration. In contrast, the GOC and GOCD groups exhibited more orderly collagen deposition, a reduced inflammatory response, and a more continuous epidermal layer, with the most pronounced improvement in the GOCD group. By day 14, collagen fibers in the GOCD group were denser and more organized (Fig. 10D), with a collagen content of 80.83%, markedly greater than in the control group (48.08%). Vascular structures and skin appendages in the GOCD group also appeared clearer and more complete, demonstrating an accelerated tissue remodeling process.

Fig. 9. (A) Representative wound images and schematic illustrations on days 0, 3, 7, and 14. (B) Wound closure rates of the Control, GO, GOC, and GOCD groups on days 3, 7, and 14. (C) H&E staining images. n = 3.

Fig. 10. (A) Images of Masson staining. (B) Immunohistochemical staining of IL-6. (C) Immunohistochemical staining of CD31. (D) Quantification of collagen deposition on day 14. (E) Quantification of IL-6 on day 7 and (F) Quantification of CD31 on day 14. n = 3.

Concurrently, while inflammation is an essential phase in tissue healing, excessive activation can impede the process.⁵⁰ IL-6, a multifunctional cytokine, is an important inflammatory regulator whose overexpression exacerbates inflammation and hinders wound healing.⁵¹ Immunohistochemical analysis (Fig. 10B and E) demonstrated that on day 7, the IL-6-positive area in the GOCD group (5.61%) was significantly lower than in the control (25.53%) and GO (18.03%) groups, indicating that this material effectively modulates the inflammatory microenvironment. This modulation promotes the orderly progression of the wound from inflammation to proliferation, thereby accelerating tissue healing. Immunohistochemical staining for CD31 on day 14 (Fig. 10C and F) further confirmed a significant increase in new blood vessels in the GOCD group, which exhibited elevated vascular density and enlarged lumen diameters, suggesting a pro-angiogenic capability.

4. Conclusion

In this study, a multifunctional GOCD hydrogel loaded with COS and DFO was developed for the repair of infected wounds. The hydrogel can release active molecules COS and DFO continuously at the wound site, thereby overcoming the characteristics of short-term retention and limited duration of action in vivo. Moreover, the GOCD hydrogel exhibited favorable biocompatibility, antibacterial activity against S. aureus and MRSA, as well as antioxidant and pro-angiogenic capacities. In a rat model of infected full-thickness skin defect, GOCD significantly accelerated wound closure, promoted collagen accumulation and tissue restructuring, enhanced angiogenesis, and mitigated inflammation, facilitating the transition from the inflammatory phase to the proliferative phase. By integrating antibacterial, antioxidant, and tissue-regenerative functions, the GOCD hydrogel offers a promising treatment strategy for infected and persistent skin wounds and provides a rational framework for the design of next-generation intelligent wound dressings.

Author contributions

Na Zhou: writing-original draft, conceptualization, data curation, methodology. Jie Huang: visualization, validation. Mengmeng Cao: visualization, validation. Xiaohan Zhang: methodology, validation. Shujian Zhang: methodology, validation. Wenxuan Zhang: validation, visualization. Xinrui Luan: methodology, validation. Xuancheng Wei: methodology, validation. Zansheng Huang: validation, methodology. Han Jin: writing-review and editing, conceptualization. Bin Zhang: writing-review and editing, conceptualization. Ying Li: writing-review and editing, supervision, project administration, funding acquisition, conceptualization.

Conflicts of interest

The authors declare no conflict of interest.

Data availability

The data that support the findings of this study are available from the corresponding authors.

Supplementary information (SI) is available. See DOI: https://doi.org/10.1039/d6ra02045c.