Dynamically responsive hydrogel with mechanical stimulation enhances diabetic wound healing via activation of Piezo1-mediated efferocytosis

Abstract

Correcting the disordered metabolism and achieving dynamic, comprehensive management of chronic diabetic wounds remains a significant challenge. This study presents a double-network dynamic hydrogel exhibiting long-term anti-inflammatory, antioxidant properties, and tunable mechanical strength. The hydrogel is primarily composed of modified chitosan, hyaluronic acid, sodium alginate, and ZnO₂/Fe³⁺ nanoparticles. The incorporated ZnO₂/Fe³⁺ nanoparticles enable microenvironmental regulation by responding to H⁺ or reactive oxygen species (ROS), while the released Fe³⁺ ions drive hydrogel network reconstruction, thereby enhancing mechanical properties. In vitro studies demonstrate the hydrogel's efficacy in efficiently scavenging ROS and enhancing Piezo1-mediated macrophage efferocytosis through cell-matrix interactions, accelerating macrophage polarization towards the M2 phenotype and resolving inflammation. In vivo experiments further confirm that the CHS@ZnO₂/Fe³⁺ hydrogel significantly promotes re-epithelialization. Mechanical stimulation provided by the hydrogel recruited abundant fibroblasts and endothelial cells to the wound site, facilitating collagen deposition and angiogenesis. This novel hydrogel dressing, combining mechanical and biochemical dual-regulation, provides an advanced therapeutic strategy for the efficient repair of diabetic chronic wounds.

Graphical abstract

Highlights

• •Here, we develop a novel CHS@ZnO₂/Fe³⁺ double-network hydrogel based on chitosan-hyaluronic acid-sodium alginate, integrating the biocompatibility, adhesiveness, and biodegradability inherent to natural polysaccharides. Key innovations include:
• •Metabolic Reprogramming & Microenvironment Remodeling: The hydrogel effectively scavenges reactive oxygen species (ROS) and facilitates lactate clearance.
• •Immunomodulation & Antibacterial Action: Zn²⁺ ions promote macrophage M2 polarization and confer robust antibacterial properties.
• •Fe³⁺ ions trigger a dynamic reorganization of the alginate network via coordination interactions, leading to a self-stiffening effect that provides sustained mechanical support for tissue regeneration.
• •Enhanced Efferocytosis: Piezo1-mediated augmentation of macrophage phagocytosis clears apoptotic cells, alleviates inflammation, and further drives polarization toward the pro-regenerative M2 phenotype.
• •Mechanotransduction via Piezo1/YAP Pathway: The enhanced matrix stiffness provides sustained mechanical signaling, activating the Piezo1/YAP pathway to stimulate fibroblast proliferation, adhesion, and migration.
• •Self-Contained Therapeutic System:The hydrogel accomplishes these functions without relying on any exogenous biological agents, highlighting its self-sufficient nature and translational potential.

1. Introduction

Diabetic wounds, particularly diabetic foot ulcers (DFUs), have emerged as a major clinical challenge due to their high prevalence—affecting approximately 2–10 % of the global diabetic population—and the substantial risk of lower limb amputation [1]. In these wounds, the chronic hyperglycemic milieu induces persistent inflammation, excessive accumulation of reactive oxygen species (ROS), and impaired angiogenesis, which collectively disrupt the wound healing microenvironment. This dysregulation leads to spatial disorganization of tissue architecture and significantly impedes re-epithelialization and collagen remodeling [2]. Moreover, bacterial infection further exacerbates local hypoxia and intensifies inflammatory responses, compounding the complexity of the healing process [3,4]. Although conventional therapeutic approaches, such as debridement and hyperbaric oxygen therapy, can provide temporary symptomatic relief [5], they often fail to coordinate the dynamic progression through the inflammatory, proliferative, and remodeling phases, and thus exhibit limited efficacy, particularly in large or non-healing wounds. As a result, therapeutic strategies that simultaneously modulate inflammation and promote cellular proliferation to facilitate re-epithelialization have become an area of intense research interest. Among these, hydrogel-based dressings have emerged as highly promising biomaterials due to their excellent biocompatibility, ECM-mimetic viscoelasticity, and multifunctional integration capacity. Hydrogels can be engineered to deliver drugs, cells, or bioactive molecules in a controlled manner, thereby offering a versatile platform for wound repair [[6], [7], [8]]. However, many current hydrogel systems still suffer from limited healing efficiency, high production costs, and transient therapeutic effects. In addition, the fabrication of multifunctional hydrogels often involves complex synthesis processes and exhibits suboptimal synergy between components. Therefore, there remains a pressing clinical need for the development of simplified, programmable hydrogel dressings capable of orchestrating the wound healing process in a dynamic and responsive manner.

Cells possess the ability to sense and respond to physical cues—such as substrate stiffness and topographical features—by converting mechanical stimuli into biochemical signals that regulate gene expression and orchestrate diverse cellular processes [9,10]. However, the static nature of conventional hydrogels fails to accommodate the dynamic remodeling of the wound microenvironment, thereby limiting their therapeutic efficacy. Matrix mechanics, or the dynamic biophysical properties of the extracellular matrix, play a pivotal role in guiding cellular behaviors during tissue repair [11]. In particular, mechanical stimulation has been shown to enhance fibroblast proliferation and promote the secretion of extracellular matrix macromolecules, both of which are essential for effective wound healing [12]. Current therapeutic strategies underutilize mechanotransduction, which limits their capacity to promote effective wound contraction and organized collagen deposition. For instance, a novel glycyrrhizic acid-based hybrid hydrogel dressing induced by inorganic zinc ions demonstrates favorable immunomodulatory properties in diabetic wound management [13]. However, its design primarily focuses on biochemical regulation, leaving the mechanotransduction mechanism mediated by the intrinsic mechanical properties of the hydrogel largely unexplored. Recently, a contractile hydrogel dressing was developed that applies active mechanical tension to approximate wound edges, thereby promoting fibroblast proliferation and migration to facilitate wound closure [14]. Nevertheless, the increased stiffness following contraction may restrict further deformation and compromise mechanical compliance during healing. Moreover, conventional hydrogels often assume a fixed morphology after formation, which hinders their adaptability to the dynamically changing demands of the wound healing process. Sun et al. developed an antioxidant-active contractile hydrogel by integrating mechanical stimulation with biochemical modulation, offering a combined strategy for chronic wound closure [15]. While these single-function hydrogels show efficacy in addressing specific issues, their overall performance remains limited when confronting the complex and dynamic pathological microenvironment of chronic diabetic wounds, characterized by recurrent infection, persistent hypoxia, and immune dysfunction. Therefore, there is a pressing need to develop multifunctional hydrogel dressings capable of dynamically modulating the wound microenvironment and autonomously adapting their mechanical properties according to the distinct stages of tissue repair.

In this study, we developed a novel dual-network dynamic hydrogel system that integrates static and responsive components, achieving a balance between structural stability and matrix adaptability. This system is based on a chemically modified static hydrogel network (CHS), composed of chitosan, hyaluronic acid, and sodium alginate. By incorporating ZnO₂/Fe³⁺ nanoparticle complexes, the hydrogel responds to acidic or ROS-rich environments, where ZnO₂/Fe³⁺ decomposes and gradually releases Zn²⁺ and Fe³⁺ ions. The released Zn²⁺ promotes macrophage M2 polarization and exerts antibacterial effects, thereby reducing wound infection. Concurrently, Fe³⁺ ions undergo secondary cross-linking with carboxyl groups of alginate, forming a new dynamic hydrogel network that reinforces the overall structure and enhances matrix stiffness, thereby preserving mechanical robustness under physiological conditions (CHS@ZnO₂/Fe³⁺). The increased stiffness of the matrix provides sustained mechanical cues that activate the Piezo1/YAP signaling pathway, promoting fibroblast proliferation, adhesion, and migration. Concurrently, Piezo1-mediated enhancement of macrophage efferocytosis facilitates the clearance of apoptotic cells, mitigates inflammation, and further drives polarization toward the pro-regenerative M2 phenotype. In vitro and in vivo studies demonstrated that this dynamic hydrogel effectively modulates the metabolic microenvironment and integrates mechanical and biochemical signaling to shorten the inflammatory phase, accelerate the M1-to-M2 macrophage transition, and significantly improve collagen deposition and neovascularization. These effects collectively promote the progression from the proliferative to the remodeling phase, ultimately leading to rapid re-epithelialization and scarless healing in diabetic wounds. Importantly, this hydrogel system achieves these therapeutic effects without the need for exogenous biological agents, relying instead on intrinsic responsiveness and biomechanical modulation. This dual-modality ‘mechano–biochemical’ regulatory strategy represents a new paradigm for the efficient repair of chronic diabetic wounds.

2. Results and discussion

2.1. Differential gene expression signatures in diabetic foot ulcers

Analysis of foot skin samples from healthy non-diabetic individuals, healed diabetic foot ulcer (DFU) patients, and non-healing DFU patients using single-cell RNA sequencing (dataset GSE165816) revealed a significantly higher proportion of monocytes/macrophages and lower proportion of fibroblasts/smooth muscle cells in non-healing DFUs compared to other groups (Fig. 1a and b). KEGG enrichment analysis demonstrated marked dysregulation in macrophages from non-healing DFUs versus healthy controls, particularly in oxidative stress, phagocytosis, and efferocytosis pathways (Fig. 1c). Differential gene expression analysis of fibroblasts between healed and non-healing DFUs identified significant alterations in non-healers, including transcriptional regulators (FOS, JUN), extracellular matrix (ECM) remodeling/cell migration genes (S100A10, CTSB, MMP2, COL6A1, MMP11), and epithelial differentiation marker KRT10 (Fig. 1d). Notably, elevated TGFB1 expression in healed DFUs suggested TGF-β pathway involvement in Wnt-mediated healing (Fig. S1) [16,17]. Comparative analysis across all three groups revealed significant upregulation of a pro-healing gene cluster—AXL, GAS6, SLC7A11 (efferocytosis axis), COL5A1 (ECM structural protein), and GNAI2 (G-protein signaling)—in healed DFUs, whereas non-healing DFUs exhibited elevated APCDD1 and IL13RA1 (indicating oxidative stress/chronic inflammation) alongside aberrant expression of SFRP2 (Wnt inhibitor) and COL23A1 (ECM assembly), suggesting fibroblast dysfunction (Fig. 1e). Collectively, these results indicate that dysregulated efferocytosis coupled with imbalanced fibroblast transcriptional control and ECM remodeling constitutes a core mechanism driving persistent inflammation and impaired healing in chronic diabetic wounds.

Fig. 1.

The characteristic distribution of macrophages and fibroblasts in human DFU. (a) Cluster analysis using the Uniform Manifold Approximation and Projection (UMAP) technique of single-cell sequencing from healthy subjects, DFU-healers and DFU non-healers skin samples revealed 15 distinct cell clusters. (b) The stacked bar plots show the proportions of different cell types in samples from different groups. (c) KEGG pathway enrichment analysis of macrophages in healthy and DFU non-healer. (d) The heatmap illustrates the significantly differentially expressed genes between fibroblast clusters in DFU-healers and DFU non-healers. (e) Violin plots showing expression levels of pro-healing and inflammation-related gene clusters in each group. (f) Representative immunofluorescence images of TUNEL (green), macrophages (red), and nuclei (blue) in normal and diabetic wounds. Arrowheads indicate TUNEL⁺ apoptotic cells. (g) Immunofluorescence staining assessed macrophage infiltration in wound tissues. (h) Immunostaining of COL 1 in fibroblasts. M1: iNOS⁺ positive cells, M2: CD206⁺ positive cells. The white dashed line represents skin epidermal tissue in the wound center. The results were presented as means ± S.D. ∗p < 0.05; ∗∗p < 0.01; DFU: Diabetic foot ulceration.

Persistent inflammation and imbalanced M1/M2 macrophage polarization represent hallmark pathological features of diabetic wounds, with macrophage dysfunction emerging as a common consequence of uncontrolled inflammation and impaired efferocytosis or phagocytosis [[18], [19], [20]]. To investigate the inflammatory response and macrophage dysfunction in diabetic wounds, we first examined the infiltration of neutrophils in wound tissues from diabetic and non-diabetic rats. Immunostaining revealed a significant increase in Ly6g⁺ neutrophils in diabetic wounds compared to normal wounds (Fig. S3a and b). We further assessed the clearance of apoptotic cells by macrophages using immunofluorescence co-localization. The number of TUNEL⁺ apoptotic cells was significantly elevated in diabetic wounds. However, the proportion of apoptotic cells internalized by macrophages was markedly reduced under diabetic conditions, indicating impaired efferocytosis (Fig. 1f). To determine whether defective clearance of apoptotic neutrophils correlates with altered macrophage polarization, immunofluorescence staining of macrophage phenotypes was performed. Diabetic wounds exhibited significantly greater infiltration of iNOS⁺ M1 macrophages than normal wounds, whereas CD206⁺ M2 macrophage levels showed no significant intergroup difference; however, the M2/M1 ratio was markedly reduced in diabetic wounds (Fig. 1g and Fig. S3d). Additionally, immunofluorescence analysis revealed that the expression of collagen type I (COL1) was significantly reduced in diabetic wounds compared to normal wounds (Fig. 1h). In summary, impaired clearance of apoptotic cells is associated with disrupted macrophage polarization, while fibroblast dysfunction contributes to diminished collagen deposition, collectively leading to delayed wound healing in diabetes.

2.2. CHS@ZnO₂/Fe³⁺ preparation and characterization

In this research, we developed a self-healing, dynamically responsive composite hydrogel dressing. To endow the hydrogel with intrinsic antioxidative properties, ZnO₂/Fe³⁺ nanoparticles were first synthesized. These nanoparticles were subsequently incorporated into an active self-healing hydrogel matrix (CHS) via a conventional free-radical polymerization strategy, resulting in the formation of a pH/ROS-responsive dynamic hydrogel system. Briefly, chitosan, hyaluronic acid, and sodium alginate were chemically modified to generate Aldehyde-Modified Hyaluronic Acid (AHA), Aldehyde-Modified Sodium Alginate (ASA), and Hydroxypropyl Chitosan (HPC) respectively. These functionalized polymers were crosslinked via dynamic covalent interactions to construct a robust hydrogel network. ZnO₂/Fe³⁺ nanoparticles were uniformly dispersed within the hydrogel precursor solution prior to gelation, ensuring homogeneous distribution throughout the matrix (Fig. 2).

Fig. 2.

Schematic depicting the therapeutic mechanism of CHS@ZnO₂/Fe³⁺ hydrogel for wound healing. Schematic illustration of the preparation process of CHS@ZnO₂/Fe³⁺ hydrogels. The CHS@ZnO₂/Fe³⁺ hydrogel dynamically responds to pathological pH/ROS fluctuations in diabetic wound microenvironments, delivering mechanobiotic cues that coordinately modulate cellular behaviors and reprogram immune responses to promote chronic wound healing.

We systematically characterized the structural composition and physicochemical properties of both ZnO₂/Fe³⁺ nanoparticles and the CHS@ZnO₂/Fe³⁺ hydrogel. Transmission electron microscopy (TEM) was first employed to assess the morphology of ZnO₂ and ZnO₂/Fe³⁺ nanoparticles (Fig. 3a). Dynamic light scattering (DLS) analysis revealed a slightly increase in the hydrodynamic diameter of ZnO₂/Fe³⁺ composite particles (93.46 ± 27.14 nm) compared to pristine ZnO₂ particles (85.33 ± 13.64 nm). However, the difference in particle size was not statistically significant. This result suggests that Fe³⁺ ions likely modified the ZnO₂ surface through the formation of aggregates or a thin coating, while largely preserving the original dispersion state and structural stability of the particles (Fig. 3b). Furthermore, zeta potential measurements showed a marked alteration in surface charge upon Fe³⁺ complexation, suggesting that coordination or electrostatic interactions between Fe³⁺ ions and the ZnO₂ surface led to a change in surface charge density (Fig. 3c). These results collectively confirm the successful integration of Fe³⁺ onto the surface of ZnO₂ nanoparticles.

Fig. 3.

Preparation and characterization of CHS@ZnO₂/Fe³⁺. (a) TEM image of ZnO₂ and ZnO₂/Fe³⁺. Scale bar: 50 nm. (b–c) The hydrodynamic particle size distributions and zeta potentials of the nanoparticles. (d) The microstructure of the different hydrogels was observed using SEM. Scale bar: 40 μm. (e) Macroscopic demonstration of the self-healing property of hydrogels. (f) Quantitative analysis of the pore size of different hydrogels. (g) Swelling ratios of different hydrogels. (h) Live/dead staining images of macrophages co-cultured on different hydrogels for 1 day and 3 days (n = 3). Data are representatives of independent experiments and all data are given as means ± SD.

The Scanning electron microscopy (SEM) images demonstrated that CHS, CHS@ZnO₂, and CHS@ZnO₂/Fe³⁺ hydrogels all exhibited a uniform and well-organized three-dimensional porous network structure (Fig. 3d). The results of elemental analysis demonstrate a uniform distribution of iron throughout the hydrogel network (Fig. S6). Quantitative analysis indicated no significant difference in average pore size among the groups, implying that incorporation of nanoparticles did not disrupt the fundamental architecture of the CHS hydrogel scaffold (Fig. 3f). As shown in Fig. 3e, the CHS hydrogel displayed characteristic self-healing behavior, with rejoining at the interface leading to structural reconstitution. Moreover, equilibrium swelling ratios of all hydrogel formulations were measured and found to be comparable, indicating that neither Fe³⁺ coordination nor nanoparticle incorporation significantly affected the swelling behavior of the hydrogel (Fig. 3g).

In addition to functional performance, good biocompatibility is a fundamental prerequisite for hydrogel-based wound dressings [21]. To assess the cytocompatibility of the composite hydrogel, we conducted live/dead cell staining and CCK-8 assays. Rat-derived fibroblasts were co-cultured with each hydrogel formulation, and cell viability was evaluated after 1 and 3 days of incubation. As shown in Fig. 3h, all groups—including the control, CHS, CHS@ZnO₂, and CHS@ZnO₂/Fe³⁺ hydrogels—exhibited predominantly green fluorescence (live cells), with minimal red fluorescence (dead cells), indicating high cell viability across conditions. Quantitative analysis using the CCK-8 assay further confirmed these observations. Fibroblasts cultured with the various hydrogels for 1, 3, and 5 days showed no significant differences in cell viability among groups. Moreover, cell proliferation increased over time in all groups, suggesting that the hydrogels supported cell adhesion and growth (Fig. S7). Collectively, these results demonstrate that the CHS@ZnO₂/Fe³⁺ composite hydrogel exhibits excellent biocompatibility and poses no detectable cytotoxic effects on fibroblasts.

2.3. Degradation behavior and mechanical reinforcement of CHS@ZnO₂/Fe³⁺ hydrogel

Building upon the established structural stability and biocompatibility of CHS@ZnO₂/Fe³⁺ hydrogels, we further evaluated their mechanical performance and degradation characteristics under different environmental conditions. Specifically, we examined ion release kinetics and network integrity in both neutral (pH 7.4) and mildly acidic (pH 5.5) media. As shown in Fig. 4a, under physiological pH, the release profiles of Zn²⁺ and Fe³⁺ remained relatively stable over a 144-h period, indicating minimal structural disruption and excellent hydrogel stability. In contrast, under acidic conditions (pH 5.5), Zn²⁺ exhibited a characteristic burst release pattern, with a rapid increase in ion concentration reaching a peak within 8 h, followed by a plateau phase. Fe³⁺ release, however, followed a biphasic trend: a slow initial release during the first 8 h, followed by a sustained linear release phase. This difference may be attributed to the acid-sensitive dissociation of coordination bonds between ZnO₂/Fe³⁺ complexes and the hydrogel matrix, facilitating the rapid release of Zn²⁺ and Fe³⁺. Notably, the released Fe³⁺ can subsequently re-coordinate with the alginate chains in the hydrogel, leading to secondary ionic crosslinking and reinforcing the network structure. SEM further revealed distinct microstructural changes under acidic conditions (Fig. 4b). Compared with hydrogels incubated at pH 7.4, those exposed to pH 5.5 displayed altered pore morphology, fiber alignment, and increased network density—likely a result of ion-triggered matrix degradation and re-crosslinking dynamics. These observations provide microstructural evidence supporting the hypothesis of a degradation–reinforcement coupling mechanism.

Fig. 4.

The microenvironment responsiveness and mechanical properties of CHS@ZnO₂/Fe³⁺ hydrogel. (a) Ion release curves of CHS@ZnO₂/Fe³⁺ hydrogel under different pH. (b) SEM images demonstrating the microtopography of CHS@ZnO₂/Fe³⁺ hydrogel under different pH. Scale bars: 50 μm and 100 μm. (c) Strain-stress curves of hydrogels. (d) Storage modulus of hydrogels. (e) Weight loss curve of hydrogels. (f) Detection of lactic acid secretion by macrophages under various stimuli (n = 3). (g) Lactic acid secretion by macrophages co-cultured with hydrogels (n = 3). (h) Fluorescence images and quantitative analysis of intracellular ROS detection in macrophages co-cultured with different hydrogels. Data are representatives of independent experiments and all data are given as means ± SD; n = 5 per group.(ns: non-significant, P > 0.05; ∗P < 0.05; ∗∗P < 0.01; ∗∗∗P < 0.001; ∗∗∗∗P < 0.0001).

To validate this hypothesis, we conducted mechanical testing, including compressive stress–strain analysis and rheological measurements. As shown in Fig. 4c and d, incorporation of ZnO₂/Fe³⁺ nanoparticles significantly enhanced the elastic modulus and storage modulus (G′) of the hydrogels, indicating that nanoparticles served as effective physical crosslinkers. Interestingly, when subjected to acidic conditions, both elastic and storage moduli increased further. Combined with the ion release data, these results suggest that the sustained release of Fe³⁺ under acidic stimulation drives secondary crosslinking with SA, forming additional coordination bonds that restructure the hydrogel network into a denser, mechanically reinforced three-dimensional matrix. Such dynamic mechanical remodeling allows cells to sense and respond to evolving matrix stiffness through deformation and remodeling of the hydrogel, potentially enhancing mechano-transduction and supporting tissue regeneration [22]. During the 14-day incubation period, all hydrogel formulations exhibited gradual weight loss when immersed in acidic solution, indicating progressive degradation (Fig. 4e). On day 14, the percentage of weight loss for CHS, CHS@ZnO₂, and CHS@ZnO₂/Fe³⁺ hydrogels was 71.8 ± 7.01 %, 72.6 ± 4.19 %, and 64.0 ± 4.32 %, respectively. These results demonstrate that the CHS-based hydrogels possess favorable biodegradability, with the presence of Fe³⁺ contributing to a slightly more sustained degradation profile due to its role in secondary crosslinking.

In the diabetic wound microenvironment, impaired microcirculation leads to hypoxic stress and sustained inflammation, forming a vicious cycle that significantly enhances glycolytic flux and results in an imbalance between lactate production and clearance [23,24]. Aberrant lactate metabolism further perpetuates chronic inflammation by modulating immune cell phenotypes and cytokine networks [25]. Through in vitro experiments, we investigated the dynamic changes in lactate metabolism of macrophages and its association with microenvironmental stimuli. Time-course analysis revealed a progressive accumulation of lactate in macrophages under standard culture conditions. Upon exposure to oxidative stress (H₂O₂, 100 μM) and inflammatory stimuli (LPS, 1 μg/mL) for 24 h—mimicking key features of the diabetic wound niche—lactate production increased by 1.13-fold and 1.38-fold, respectively, compared to the untreated control group. These findings suggest a metabolic shift toward a glycolysis-dominant, pro-inflammatory (M1-like) phenotype in response to oxidative and inflammatory stressors (Fig. 4f). To evaluate the potential of our hydrogel system in reversing this metabolic reprogramming, macrophages were co-cultured with CHS, CHS@ZnO₂, or CHS@ZnO₂/Fe³⁺ hydrogels for 24 h. As shown in Fig. 4g, all hydrogel-treated groups exhibited significantly reduced extracellular lactate levels compared to both the stimulated and control groups. This result indicates that the hydrogel mitigates lactate accumulation through a synergistic mechanism involving H⁺-responsive buffering and modulation of macrophage metabolic activity, thereby contributing to microenvironmental normalization.

In the chronic wound microenvironment, ROS play a pivotal role in modulating inflammatory signaling in macrophages. The excessive ROS accumulation can lead to oxidative cellular damage and impaired tissue regeneration [26,27]. To further investigate the ROS-scavenging capacity of the composite hydrogel, we employed H₂O₂ to simulate an inflammatory wound environment, and assessed intracellular ROS levels using the DCFH-DA fluorescent probe. As shown in Fig. 4h, H₂O₂-treated macrophages exhibited a pronounced increase in fluorescence intensity of 2′,7′-dichlorofluorescein (DCF), indicating elevated intracellular ROS levels. In contrast, macrophages co-cultured with CHS@ZnO₂ and CHS@ZnO₂/Fe³⁺ hydrogels demonstrated significantly reduced DCF fluorescence, corresponding to only 2.05-fold and 1.55-fold of the untreated baseline, respectively. These findings suggest that the composite hydrogel effectively attenuates oxidative stress by scavenging excess ROS, thereby protecting macrophages from oxidative damage. This ROS-scavenging effect is primarily attributed to the redox activity of ZnO₂, whose peroxide bonds enable rapid neutralization of free radicals through redox cycling mechanisms.

2.4. Anti-inflammatory polarization and efferocytosis enhancement in vitro

Macrophage metabolism plays a critical role in orchestrating wound healing and tissue remodeling by regulating the transition between pro-inflammatory (M1) and anti-inflammatory (M2) phenotypes [28]. However, impaired phenotypic switching—characterized by sustained M1 activation—remains a major barrier to the resolution of inflammation and contributes to the chronicity of diabetic wounds [29]. Given the potent ROS-scavenging capacity of ZnO₂, we hypothesized that the CHS@ZnO₂/Fe³⁺ hydrogel may exert an anti-inflammatory effect by modulating macrophage polarization. To evaluate this, we used LPS to induce the polarization of M0 macrophages into the M1 phenotype, serving as a positive control. Immunofluorescence staining was performed to identify iNOS⁺ macrophages as an indicator of pro-inflammatory activation. As shown in Fig. 5a, LPS stimulation markedly increased the proportion of iNOS⁺ macrophages compared to the control group, indicating robust pro-inflammatory activation. In contrast, co-culture with CHS@ZnO₂ or CHS@ZnO₂/Fe³⁺ hydrogels resulted in significantly reduced iNOS fluorescence intensity, suggesting suppression of M1 polarization (Fig. 5b). Consistently, ELISA results revealed elevated levels of pro-inflammatory cytokines TNF-α and IL-6 in the LPS group, whereas hydrogel-treated macrophages exhibited significantly decreased expression of these cytokines (Fig. 5c and d).

Fig. 5.

Macrophage polarization in vitro in response to CHS@ZnO₂/Fe³⁺ hydrogel. (a–b) iNOS/F-actin fluorescence staining was used to analyze the anti-inflammatory performance of different hydrogels, and the positive cellular iNOS expression rate statistics. (c–d) The expression of TNF-α and IL-6 was detected in the cell supernatants of each group by using Elisa kit (n = 3). (e) Western blot analysis of Piezo1 and YAP expression in macrophages grown on different hydrogel substrates (n = 3). (f) HL-60 cells were induced to differentiate into neutrophils, and paclitaxel was added to induce apoptosis. (g) Giemsa staining revealed apoptotic neutrophils. Scale bars: 50 μm and 25 μm. (h) Representative fluorescence image of macrophages engulfing neutrophils. Macrophages were co-cultured with CFSE-labeled apoptotic neutrophils (green) for 2 h. (i–j) Immunofluorescence images demonstrating CHS@ZnO₂/Fe³⁺ hydrogel-mediated modulation of M2 macrophage polarization and quantitative analysis. (k–l) The expression of IL-10 and TGF-β was detected in the cell supernatants of each group by using Elisa kit (n = 3). Data are representatives of independent experiments and all data are given as means ± SD; n = 5 per group.(ns: non-significant, P > 0.05; ∗P < 0.05; ∗∗P < 0.01; ∗∗∗P < 0.001; ∗∗∗∗P < 0.0001).

Cells can sense and respond to mechanical cues through deformation or remodeling of the surrounding hydrogel matrix during its degradation. This mechano-sensing process enables cells to transduce physical signals into intracellular responses [22]. Macrophages possess an efficient efferocytosis, whereby they clear apoptotic cells to promote the resolution of inflammation [30]. However, in diabetic wounds, this phagocytic capacity is markedly impaired, contributing to persistent inflammation and delayed healing [31]. Previous studies have demonstrated that increased matrix stiffness enhances macrophage phagocytosis, and this stiffness-mediated improvement in efferocytosis is dependent on the mechanosensitive ion channel Piezo1 [32,33]. We previously demonstrated that the embedded ZnO₂/Fe³⁺ nanoparticles undergo degradation under acidic conditions, releasing Fe³⁺ ions that re-crosslink with the CHS-based network, thereby enhancing the mechanical strength of CHS@ZnO₂/Fe³⁺ hydrogels (Fig. 4c and d). To elucidate the underlying mechanism, Western blot analysis was performed to quantify the expression of Piezo1 and YAP proteins. The CHS@ZnO₂/Fe³⁺ group showed markedly elevated levels of both proteins, suggesting that Fe³⁺-mediated hydrogel network remodeling enhances the matrix stiffness of hydrogels and effectively activates the Piezo1/YAP mechanosignal transduction pathway. (Fig. 5e). To investigate the impact of this hydrogel on macrophage efferocytosis, we co-cultured macrophages with apoptotic neutrophils labeled with Carboxyfluorescein Succinimidyl Ester (CFSE) and assessed the clearance efficiency. Briefly, neutrophil-like cells were generated by treating HL-60 cells with DMSO (1.25 %) for five days to induce differentiation, as verified by nuclear morphology (lobulated nuclei), followed by induction of apoptosis using paclitaxel (0.1 nM) (Fig. 5f and g). Apoptotic neutrophils were then co-cultured with macrophages pretreated with CHS, CHS@ZnO₂, or CHS@ZnO₂/Fe³⁺ hydrogels for 2 h. As shown in Fig. 5h, macrophages exposed to CHS@ZnO₂/Fe³⁺ hydrogels exhibited the highest level of phagocytic activity, with significantly enhanced clearance of apoptotic cells compared to other groups.

Moreover, immunofluorescence staining of CD206 revealed an increased proportion of CD206⁺ macrophages in both CHS@ZnO₂ and CHS@ZnO₂/Fe³⁺ groups, indicative of enhanced M2 polarization (Fig. 5i and j). The ELISA results revealed that the secretion of anti-inflammatory cytokine IL-10 and the regenerative factor TGF-β was markedly upregulated following hydrogel treatment (Fig. 5k and l), indicating that CHS@ZnO₂ and CHS@ZnO₂/Fe³⁺ hydrogels effectively restored macrophage homeostasis and promoted an anti-inflammatory, pro-regenerative phenotype. Notably, the CHS@ZnO₂/Fe³⁺ group exhibited a higher abundance of CD206⁺ macrophages than the CHS@ZnO₂ group, which may be attributed to Piezo1/YAP-mediated enhancement of efferocytosis driven by improved mechanical properties, thereby facilitating the phenotypic switch from M1 to M2 macrophages.

2.5. Activation of fibroblasts and crosstalk between macrophages, fibroblasts and endothelial cells

This study elucidates a H⁺/ROS-responsive “ion release–network reconstruction” coupling mechanism, whereby rapid dissociation of ZnO₂/Fe³⁺ nanoparticles under acidic conditions triggers initial ion release, while the sustained liberation of Fe³⁺ ions engages in coordination crosslinking with the hydrogel matrix, thereby reinforcing its mechanical strength. Such dynamic re-crosslinking endows the hydrogel with intelligent responsiveness, enabling adaptive mechanical reinforcement via ion-mediated secondary crosslinking, which in turn facilitates cellular sensing and response to matrix stiffness modulation.

To distinguish the direct influence of ionic release from hydrogel-mediated mechanical signaling, we established two distinct in vitro culture models (Fig. 6a). When macrophages were indirectly co-cultured with hydrogels, the CHS@ZnO₂ and CHS@ZnO₂/Fe³⁺ groups exhibited an increased proportion of CD206⁺ cells and reduced iNOS⁺ macrophages compared to the LPS group, which is likely attributable to the intrinsic anti-oxidative and anti-inflammatory properties of ZnO₂ nanoparticles (Fig. S9). However, in contrast to the results observed from direct surface culturing on the hydrogel substrates, no significant differences in macrophage polarization were detected between the CHS@ZnO₂ and CHS@ZnO₂/Fe³⁺ groups under indirect co-culture conditions. Consistently, the efferocytosis of macrophage toward CFSE-labeled apoptotic cells did not significantly differ among hydrogel groups in the indirect culture system, suggesting that direct mechanotransductive signaling, rather than soluble ionic factors alone, is essential for promoting macrophage efferocytosis (Fig. S10). Loss of Piezo1 impairs calcium influx and disrupts downstream cytoskeletal organization in macrophages, thereby compromising their efferocytosis [34,35]. The enhancement of efferocytosis in macrophages is critically dependent on substrate stiffness–induced activation of Piezo1 [33]. These findings collectively demonstrate that the CHS@ZnO₂/Fe³⁺ hydrogel exerts its immunomodulatory effects primarily through mechanotransductive signaling, in which hydrogel-mediated matrix stiffening directs macrophage polarization by engaging the Piezo1-dependent pathway.

Fig. 6.

In vitro modulation of fibroblast functions by CHS@ZnO₂/Fe³⁺ hydrogel. (a) Schematic diagrams of two cell-hydrogel co-culture models: Cells co-culture on hydrogel or co-cultured with hydrogel. (b) Representative immunofluorescence images of YAP localization (green) with DAPI-counterstained nuclei (blue) and quantitative analysis of nuclear/cytoplasmic YAP fluorescence intensity ratios. (c) F-actin staining was used to analyze the fibroblasts spreading on the surface of the hydrogel. (d, e) Images and quantification of the migration of fibroblasts. (f) Western blot analysis of YAP, COL 1 and α-SMA expression in fibroblasts co-cultured on different hydrogel substrates. (g) Experimental design to study macrophage association with fibroblasts and endothelial cells. (h) The expression levels of COL 1, COL 3, α-SMA and CD31 proteins after macrophage supernatant treatment. (i) Representative images of α-SMA and COL 1 fluorescence staining. Data are representatives of independent experiments and all data are given as means ± SD; n = 5 per group.(ns: non-significant, P > 0.05; ∗P < 0.05; ∗∗P < 0.01; ∗∗∗P < 0.001; ∗∗∗∗P < 0.0001).

Fibroblast activation and their subsequent differentiation into myofibroblasts play a central role in extracellular matrix (ECM) remodeling and wound healing [36,37]. To evaluate the regulatory effects of CHS@ZnO₂/Fe³⁺ hydrogels on fibroblast behavior, we cultured fibroblasts directly on the hydrogel surfaces for 72 h and assessed the nuclear translocation of Yes-associated protein (YAP) via immunofluorescence staining. As shown in Fig. 6b, cells cultured on CHS@ZnO₂/Fe³⁺ hydrogels exhibited significantly enhanced YAP nuclear localization compared to the control group, as evidenced by an elevated nuclear-to-cytoplasmic fluorescence intensity ratio. This nuclear accumulation of YAP—a key effector of the Hippo signaling pathway—indicates its activation and suggests involvement in downstream transcriptional programs related to cell adhesion, spreading, and migration [9]. Furthermore, phalloidin staining of filamentous actin (F-actin) revealed that fibroblast spreading behavior was markedly influenced by hydrogel stiffness. Cells on stiffer CHS@ZnO₂/Fe³⁺ hydrogels displayed more pronounced and organized actin stress fibers, indicating enhanced cytoskeletal tension and mechanical sensing activity (Fig. 6c). In addition, fibroblasts cultured on CHS@ZnO₂/Fe³⁺ hydrogels exhibited markedly elongated lamellipodia, suggesting that the stress-relaxation properties of the hydrogel matrix facilitated dynamic assembly of protrusive structures at the leading edge. Lamellipodia extension is a prerequisite for effective cell spreading and directional migration. The dynamic network reconfiguration of the hydrogel likely alleviates mechanical constraints on the cytoskeleton, thereby promoting lamellipodial protrusion and enhancing cellular motility [38]. Interestingly, fibroblasts cultured on rigid glass substrates exhibited a larger spreading area, yet no significant increase in YAP nuclear translocation was observed. This phenomenon may be attributed to the inability of static, non-adaptive substrates to provide sustained mechanical signaling. Although high initial stiffness supplies mechanical cues, the absence of dynamic remodeling likely triggers compensatory responses such as focal adhesion internalization and actin cytoskeleton depolymerization, allowing cells to mechanically adapt. Mechanical cues derived from ECM stiffness are sensed through integrin-mediated transmembrane mechanotransduction pathways, which directly regulate the dynamic assembly of focal adhesions (FAs), thereby influencing cell spreading and migration [39]. The ionically crosslinked dynamic hydrogel modulates fibroblast proliferation, filopodia extension, and spreading behavior through activation of the integrin β1/YAP signaling axis [40]. Scratch wound assays revealed that fibroblasts cultured directly on CHS@ZnO₂/Fe³⁺ hydrogels exhibited a significantly enhanced migration rate, reaching 73.6 ± 8.8 % closure at 24 h, compared to 37.5 ± 12.0 % in the control group (Fig. 6d and e). In contrast, when fibroblasts were co-cultured with the hydrogels, no significant differences in cell migration rates were observed among the groups within 24 h (Fig. S12). Furthermore, Western blot analysis showed that under co-culture conditions, YAP protein expression remained at baseline levels, and the expression of myofibroblast markers α-SMA and COL1 was not significantly altered (Fig. 6f), indicating that direct mechanical engagement with the hydrogel substrate is essential for YAP activation and differentiation of myofibroblasts.

Intercellular communication plays a pivotal role in orchestrating tissue repair. To investigate the influence of macrophage polarization on fibroblast activation and endothelial angiogenic behavior, we employed a conditioned medium co-culture model. Specifically, the supernatants collected from macrophages cultured on different hydrogel substrates were applied to fibroblasts or endothelial cells (Fig. 6g). Using western blotting and immunofluorescence staining, we systematically assessed the phenotypic and functional changes in these recipient cells. As shown in Fig. 6h, fibroblasts exposed to CHS@ZnO₂/Fe³⁺-conditioned medium exhibited a 2.3-fold increase in α-SMA expression compared to the control group. Additionally, the secretion of COL1 and COL3 was elevated by 2.8- and 5.9-fold (Fig. S16b), respectively, indicating a shift toward a myofibroblast phenotype and enhanced ECM deposition. Immunofluorescence further confirmed that α-SMA–positive cells displayed bundled alignment, accompanied by markedly intensified COL1 fluorescence, highlighting improved ECM remodeling (Fig. 6i). Functional assays of endothelial cells revealed enhanced migration capacity and more CD31 expression (Figs. S14 and S15). This enhancement is likely attributable to the CHS@ZnO₂/Fe³⁺ hydrogel's ability to promote M2 macrophage polarization, which in turn facilitates the paracrine secretion of TGF-β. This secreted factor mediates intercellular communication that synergistically regulates fibroblast-to-myofibroblast differentiation and angiogenic programming in endothelial cells, ultimately promoting ECM synthesis and vascularization [41,42]. Together, these findings suggest that CHS@ZnO₂/Fe³⁺ hydrogels mediate mechanotransduction to enhance cell adhesion and migration while simultaneously fostering M2 macrophage-derived TGF-β secretion, thereby accelerating collagen deposition and neovascularization—key steps in advancing wound healing.

2.6. In vivo evaluation of wound healing efficacy

To assess the therapeutic potential of CHS@ZnO₂/Fe³⁺ hydrogel in chronic wound healing, full-thickness excisional wounds (10 mm in diameter) were created on the dorsal skin of diabetic rats using a biopsy punch, followed by different treatment regimens. Gross morphological observation revealed that wound closure in the CHS, CHS@ZnO₂, and CHS@ZnO₂/Fe³⁺ hydrogel groups progressed significantly faster than in the untreated control group (Fig. 7a and b). Notably, wounds treated with CHS@ZnO₂ and CHS@ZnO₂/Fe³⁺ hydrogels demonstrated accelerated healing as early as day 3 (Fig. 7c). As wound healing advanced, the CHS@ZnO₂/Fe³⁺ group exhibited the highest closure rate, reaching approximately 95 % by day 12, compared to ∼78 % in the control group.

Fig. 7.

Therapeutic effects of CHS@ZnO₂/Fe³⁺ hydrogel in promoting diabetic wound healing in rats. (a, b) Photographs of representative wounds from various treatment groups and quantitative analysis. (c) Wound closure rate of each treatment group. (d, e) H&E and Masson staining of diabetic wound tissues on days 7 and 14. (f, g) Relative quantitative analysis of epidermal thickness and re-epithelialization rate on day 14. (h, i) Statistical analysis of wound length and collagen deposition on days 7 and 14. Data are representatives of independent experiments and all data are given as means ± SD; n = 5 per group.(ns: non-significant, P > 0.05; ∗P < 0.05; ∗∗P < 0.01; ∗∗∗P < 0.001; ∗∗∗∗P < 0.0001).

Histological evaluation by hematoxylin and eosin (H&E) staining on day 7 revealed more abundant granulation tissue in the CHS@ZnO₂/Fe³⁺ group than in the control, indicating faster re-epithelialization (Fig. 7d and g). Quantitative measurement of the unhealed wound length further confirmed that hydrogel treatment significantly reduced wound size, with the most pronounced effect observed in the CHS@ZnO₂/Fe³⁺ group (Fig. 7h). After 14 days, all groups showed varying degrees of granulation tissue formation and re-epithelialization. Strikingly, the CHS@ZnO₂/Fe³⁺-treated wounds exhibited near-complete closure with minimal scarring and regeneration of skin appendages (Fig. 7f). Masson's trichrome staining further demonstrated enhanced collagen deposition in the CHS@ZnO₂/Fe³⁺ group at both day 7 and day 14, surpassing all other groups (Fig. 7e–i). Collectively, these findings confirm that CHS@ZnO₂/Fe³⁺ hydrogel significantly accelerates wound closure by promoting epithelial regeneration and collagen deposition, offering an effective strategy for chronic wound management.

2.7. RNA-seq analysis

To further elucidate the underlying mechanisms by which CHS@ZnO₂/Fe³⁺ hydrogels promote diabetic wound healing, we conducted transcriptomic profiling of wound tissues at day 14 post-treatment. A total of 3659 genes were detected across the control, CHS@ZnO₂, and CHS@ZnO₂/Fe³⁺ groups. Among them, 1643 shared genes were identified and quantified using a Venn diagram (Fig. 8a). Volcano plots revealed 300 differentially expressed genes (DEGs) between the control and CHS@ZnO₂ groups, 1250 DEGs between CHS@ZnO₂/Fe³⁺ and control, and 449 DEGs between CHS@ZnO₂/Fe³⁺ and CHS@ZnO₂ (Fig. 8b).

Fig. 8.

Transcriptomic research on the treatment of diabetic chronic wounds. (a, b) Venn diagram and volcano plot analysis of the three treatment groups. (c) Heatmap of differential gene expression. (d) Annotation of GO for biological process category. (e) The bar chart shows the results of GO enrichment analysis of differential genes between the control group and the CHS@ZnO₂/Fe³⁺ group. (f) Circos plot displaying the association between differential genes and associated GO terms between the Control group and the CHS@ZnO₂/Fe³⁺ group. (g–j) GSEA analysis of differentially expressed genes related to wound healing in control and CHS@ZnO₂/Fe³⁺ groups.

Based on the wound healing timeline, the major DEGs were classified into stages corresponding to inflammation (immune response), proliferation (angiogenesis, adhesion, and cell proliferation), and remodeling (ECM deposition and tissue reconstruction), as illustrated in the heatmap (Fig. 8c and Fig. S26). Gene Ontology (GO) enrichment analysis revealed that these DEGs were primarily involved in skin development, epithelial regeneration, and phagocytic processes (Fig. 8d). Further GO comparisons between the CHS@ZnO₂/Fe³⁺ and control groups highlighted significant enrichment in pathways associated with skin development, extracellular matrix organization, collagen biosynthesis, immune responses, and cell adhesion (Fig. 8e and f). Gene Set Enrichment Analysis (GSEA) further confirmed that CHS@ZnO₂/Fe³⁺ treatment positively regulated pathways crucial for wound repair, including integrin-mediated adhesion, smooth muscle cell proliferation, ECM remodeling, and angiogenesis (Fig. 8g). In addition, CHS@ZnO₂/Fe³⁺ hydrogels exhibited intrinsic antibacterial activity, which may contribute to improved wound outcomes (Fig. S25).

2.8. In vivo regulation of macrophage polarization and fibroblast function

Macrophages play a pivotal role in tissue repair through phagocytosis and the secretion of cytokines; however, dysfunction in these processes is a key contributor to delayed wound healing [43]. WikiPathways enrichment analysis revealed that inflammation-related pathways were significantly enriched in both CHS@ZnO₂ and CHS@ZnO₂/Fe³⁺ groups compared to the control, indicating that the composite hydrogels have modulatory effects on inflammatory responses. This effect is likely attributed to the anti-inflammatory properties of ZnO₂ nanoparticles (Fig. 9a and c). Subsequently, immunostaining was performed to further verify whether inflammation alleviation was associated with macrophage phenotype switching. Immunofluorescence images showed that on days 7 and 14 post-treatment, the CHS@ZnO₂/Fe³⁺ hydrogel group exhibited the highest M2/M1 macrophage ratio (Fig. 9d and Fig. S17). This indicates that treatment with CHS@ZnO₂ hydrogels significantly downregulated the proportion of iNOS ⁺ macrophages while promoting the distribution of CD206⁺ macrophages at the wound site. Notably, compared to the control and CHS@ZnO₂ groups, focal adhesion-related pathways were significantly enriched in the CHS@ZnO₂/Fe³⁺ group, suggesting that this hydrogel may enhance the expression of focal adhesion-associated genes (Fig. 9b). Interactions between cells and the ECM are critical for maintaining tissue function and regulating cellular behavior. Focal adhesions sense and respond to mechanical tension, thereby modulating cell activities. These findings further elucidate the mechanisms by which CHS@ZnO₂/Fe³⁺ dynamic hydrogels regulate cellular behavior both in vitro and in vivo. The Fe³⁺-mediated secondary crosslinking, induced by the decomposition of ZnO₂/Fe³⁺ nanoparticles, enhances hydrogel structural stability and mechanical properties, thereby promoting fibroblast proliferation, migration, and adhesion.

Fig. 9.

CHS@ZnO₂/Fe³⁺ hydrogel modulates macrophage polarization and enhances efferocytosis (apoptotic cell clearance). (a) Wikipathway enrichment analysis of differentially expressed genes between control and CHS@ZnO₂ groups, (b) control and CHS@ZnO₂/Fe³⁺ groups, (c) CHS@ZnO₂ and CHS@ZnO₂/Fe³⁺ groups. (d) Immunofluorescence staining of macrophage polarization markers in different treatment groups on day 7. iNOS (M1 marker, red), CD206 (M2 marker, green), and DAPI (nuclear stain, blue). (e) GO enrichment analysis of differentially expressed genes between CHS@ZnO₂ and CHS@ZnO₂/Fe³⁺ groups. (f) Representative immunofluorescence images of CD68 (macrophage marker, red), TUNEL (apoptosis marker, green), and DAPI (nuclear stain, blue) in different treatment groups on day 7. White arrows indicate colocalized CD68⁺/TUNEL⁺ cells. Data are representatives of independent experiments and all data are given as means ± SD; n = 5 per group.

Phagocytosis, the process of clearing apoptotic cells, is a critical factor in promoting wound healing [44]. Persistent apoptotic cells can exacerbate inflammation and inhibit tissue repair. Transcriptomic analysis revealed that treatment with CHS@ZnO₂/Fe³⁺ hydrogel, compared to CHS@ZnO₂ hydrogel, significantly enhanced macrophage phagocytic activity (Fig. 9e). This finding was further validated by fluorescence staining, which showed increased clearance of apoptotic cells by macrophages after 7 days of CHS@ZnO₂ hydrogel treatment (Fig. 9f). Moreover, compared with the control group, the CHS@ZnO₂/Fe³⁺ hydrogel group showed increased Piezo1 expression and a significantly higher ratio of YAP nuclear translocation in the wound area (Fig. S18). Our study demonstrates that CHS@ZnO₂/Fe³⁺ hydrogel effectively facilitates the transition from the inflammatory phase to the proliferative and remodeling phases during wound healing. This effect is mediated through dynamic responsive changes that enhance mechanical properties, further modulating macrophage polarization.

To further assess myofibroblast differentiation and neovascularization, immunofluorescence staining of α-SMA and CD31 was conducted. On day 7, the CHS@ZnO₂/Fe³⁺ group exhibited widespread vascular structures across the wound area, with high α-SMA expression intensity and a well-organized, mature vascular network (Fig. 10a and Fig. S19). These findings are consistent with the H&E and Masson's trichrome staining results, where a greater number of capillary-like vascular structures (well-defined lumens) and a more prominent red-stained area were observed in the CHS@ZnO₂/Fe³⁺ group compared to other treatments. Excessive neovascularization, however, can potentially contribute to fibrosis at the wound site, underscoring the need for controlled angiogenesis during tissue repair [45]. In additiation, Ki67 immunostaining was performed as a nuclear proliferation marker to investigate the proliferative response of fibroblasts to hydrogel treatment. As shown in Fig. 10b, the proportion of Ki67-positive fibroblasts in the CHS@ZnO₂/Fe³⁺ group reached approximately 78 %, markedly higher than that in the control group (∼30 %). This enhanced proliferative capacity may be attributed to mechanical activation of fibroblasts induced by Fe³⁺ mediated secondary crosslinking within the hydrogel matrix, which led to increased matrix stiffness. Moreover, robust expression of vimentin—a mesenchymal marker associated with fibroblast recruitment—was observed in the wounds treated with CHS@ZnO₂/Fe³⁺ for 7 or 14 days, whereas only sparse expression was detected in the control group (Fig. 10c and Fig. S21), suggesting that untreated wounds remained in a persistent inflammatory phase. Notably, by day 14, CD31 expression in the CHS@ZnO₂/Fe³⁺ group was significantly reduced compared to earlier time points, indicating that this hydrogel dressing not only promotes neovascularization during the reparative phase but also prevents excessive angiogenesis during tissue remodeling (Fig. S22). Collagen deposition is a critical determinant of diabetic wound healing during the repair and remodeling stages. Immunofluorescence staining for COL1 and COL3 revealed low collagen expression in the control group, whereas wounds treated with CHS@ZnO₂/Fe³⁺ hydrogels exhibited markedly higher levels of both COL1 and COL3, with fibers showing aligned orientation (Fig. 10d, e and Fig. S23). These results demonstrate that CHS@ZnO₂/Fe³⁺ hydrogels possess superior pro-angiogenic capacity in the early healing phase and effectively recruit and mechanically activate fibroblasts, thereby enhancing collagen deposition and remodeling to accelerate diabetic wound closure.

Fig. 10.

CHS@ZnO₂/Fe³⁺ hydrogel accelerates wound healing by promoting angiogenesis and collagen deposition. (a) The expression of α-SMA and (c)Vimentinon day 7 under various treatments, indicating the adhesion and differentiation of fibroblasts. (b) Representative immunofluorescent images of Ki67 on day 7 and quantification analysis, indicating the proliferative activity of fibroblasts. (d, e) The expression of COL 1 and COL 3 on day 7, indicating the collagen deposition. Data are representatives of independent experiments and all data are given as means ± SD; n = 5 per group.(ns: non-significant, P > 0.05; ∗P < 0.05; ∗∗P < 0.01; ∗∗∗P < 0.001; ∗∗∗∗P < 0.0001).

In this study, we developed a double-network hydrogel composed of chitosan, hyaluronic acid, and sodium alginate incorporated with ZnO₂/Fe³⁺ nanoparticles (CHS@ZnO₂/Fe³⁺), which exhibits H⁺ and ROS-responsive microenvironment regulation and enhanced mechanical reinforcement. Compared to previously reported water-initiated enhancement hydrogels based on glycine-modified supramolecular thioctic acid and γ-Fe₂O₃ nanoparticles—which promote re-epithelialization through sustained mechanical stimulation and immunomodulation—our system offers several distinct advantages [46]. Unlike single Fe³⁺ releasing systems, the CHS@ZnO₂/Fe³⁺ hydrogel provides broader microenvironment responsiveness and multifunctional outputs. Beyond sole Fe³⁺ ions release, the ZnO₂/Fe³⁺ complex efficiently scavenges ROS and modulates the metabolic microenvironment by facilitating lactate clearance. The released Zn²⁺ further contributes to antibacterial activity and macrophage M2 polarization, while Fe³⁺ ions are concurrently released to reconstruct the hydrogel network via coordination bonds, thereby improving mechanical properties. More importantly, the hydrogel enhances Piezo1-mediated macrophage efferocytosis through cell-matrix interactions, directly addressing a key bottleneck in inflammatory resolution. Furthermore, by combining mechanical stimulation with biochemical signaling, our hydrogel synergistically promotes re-epithelialization, collagen remodeling, and angiogenesis. This integrated approach—simultaneously regulating the wound microenvironment, modulating immune responses, and providing mechanical support—makes the CHS@ZnO₂/Fe³⁺ hydrogel particularly suitable for complex healing scenarios such as diabetic chronic wounds, which are often characterized by infection, hypoxia, and uncontrolled inflammation.

While this study demonstrates that the CHS@ZnO₂/Fe³⁺ hydrogel effectively promotes wound healing through dual mechano-biochemical regulation in an acute wound model, certain limitations should be acknowledged. First, the acute full-thickness skin defect model used here cannot fully recapitulate the complex pathological microenvironment of clinical diabetic chronic wounds, which is typically characterized by sustained hypoxia, severe infection, accumulation of advanced glycation end products (AGEs), and dysregulated chronic inflammation. Therefore, the efficacy and long-term safety of this hydrogel in genuine diabetic chronic wounds require further systematic validation in more clinically relevant animal models, such as composite models established in db/db diabetic mice combined with high-fat diet induction, wound infection with drug-resistant bacteria, and hypoxic conditions. Second, this work primarily focuses on early cellular events and mechanotransduction mechanisms, leaving the potential influence of the hydrogel on collagen metabolism and scar formation during the complete healing cycle, especially the remodeling phase, insufficiently explored. Addressing these questions would deepen the theoretical understanding of this study and provide clearer guidance for the subsequent precise optimization of the hydrogel. In summary, the present study provides foundational data and a theoretical basis for the application of the CHS@ZnO₂/Fe³⁺ hydrogel in wound repair. The identified limitations also highlight the need for further research that more closely mimics clinical scenarios. Future work will focus on these aspects to facilitate the translation of this smart dressing toward clinical application.

3. Conclusion

In summary, we developed a dynamic, ROS/H⁺-responsive CHS@ZnO₂/Fe³⁺ hydrogel with enhanced mechanical properties and multifunctional bioactivity for the treatment of diabetic wounds. The incorporation of ZnO₂/Fe³⁺ nanocomposites endowed the hydrogel with stimuli-responsiveness, enabling microenvironmental modulation through ROS scavenging and pH-triggered ion release. The subsequent release of Fe³⁺ facilitated secondary crosslinking within the hydrogel matrix, thereby reinforcing its structural integrity and stiffness. In vitro experiments demonstrated that the enhanced mechanical cues mediated by CHS@ZnO₂/Fe³⁺ hydrogel activated Piezo1-dependent efferocytosis and promoted macrophage polarization toward a pro-regenerative M2 phenotype. In a full-thickness diabetic wound model, the hydrogel significantly accelerated wound closure by promoting re-epithelialization, collagen deposition, fibroblast recruitment, and myofibroblast differentiation under sustained mechanical stimulation. Transcriptomic analysis further revealed that CHS@ZnO₂/Fe³⁺ hydrogel modulated key biological pathways associated with skin development, ECM organization, inflammation resolution, phagocytosis, angiogenesis, and antimicrobial defense. Collectively, these findings highlight the therapeutic potential of CHS@ZnO₂/Fe³⁺ as an intelligent, biomechanically active wound dressing that orchestrates immune modulation and tissue remodeling, offering a promising strategy for the management of chronic diabetic wounds.

4. Experimental section

4.1. Synthesis of ZnO₂ NPs and ZnO₂/Fe³⁺

100 mg of Zinc acetate (Zn (CH₃COO)₂, 99.99 %) and 100 mg of Polyvinylpyrrolidone (PVP, K15) were dissolved in 5 mL of water. Then 1 mL of H₂O₂ was added quickly under ultrasound. After reaction for 24 h, the resulting PVP-modified ZnO₂ nanoparticles were collected by centrifugation (8000 rpm,10 min). To synthesize ZnO₂/Fe³⁺, the aqueous solution of ZnO₂ was mixed with the same volume of 200 mg/ml FeCl₃. After stirring at room temperature for 4 h, the obtained ZnO₂/Fe³⁺ nanoparticles were collected by centrifugation (12000 rpm, 15 min).

4.2. Synthesis of Aldehyde-Modified Hyaluronic Acid (AHA), Aldehyde-Modified Sodium Alginate (ASA), and Hydroxypropyl Chitosan (HPC)

Aldehyde-modified hyaluronic acid (AHA) was synthesized by periodate oxidation of hyaluronic acid (HA, 50 kDa, 1.0 g) in 100 mL ultrapure water. Sodium periodate (NaIO₄, 0.5 g) was added, and the mixture was stirred in the dark at 25 °C for 4 h. The reaction was quenched with 10 mL ethylene glycol, dialyzed (3.5 kDa MWCO) for 72 h, and freeze-dried to obtain AHA (30 % aldehyde modification, determined by hydroxylamine hydrochloride titration).

Aldehyde-modified sodium alginate (ASA) was prepared via analogous oxidation: sodium alginate (SA, 80 kDa, 1.0 g) dissolved in 100 mL water at 40 °C was reacted with NaIO₄ (0.6 g) in the dark at 25 °C for 6 h. After quenching with 10 mL ethylene glycol, the mixture was dialyzed (3.5 kDa MWCO) for 72 h and freeze-dried, yielding ASA with 25 % aldehyde modification.

Hydroxypropyl chitosan (HPC) was synthesized by reacting chitosan (100 kDa, 90 % deacetylation, 1.0 g) in 50 mL 10 % NaOH with 5 mL propylene oxide at 60 °C under reflux for 12 h. The mixture was neutralized to pH 7.0, dialyzed (8 kDa MWCO) for 72 h, and freeze-dried to obtain HPC.

4.3. Preparation of self-healing hydrogels

Self-healing hydrogels were fabricated through Schiff base reactions between aldehyde groups in AHA/ASA and amino groups in HPC, with dynamic imine bonds enabling self-healing properties.

CHS hydrogel: AHA (0.1 g) and ASA (0.1 g) were dissolved in 5 mL ultrapure water to form a mixed aldehyde polymer solution. Separately, HPC (0.2 g) was dissolved in 5 mL ultrapure water. The two solutions were rapidly mixed under magnetic stirring at room temperature, and gelation occurred within 5 min. The resulting CHS hydrogel was allowed to stand for 1 h to complete crosslinking.

CHS@ZnO₂ hydrogel: ZnO₂ nanoparticles (0.02 g) were dispersed in 5 mL ultrapure water via ultrasonic treatment for 30 min to form a uniform suspension. AHA (0.1 g) and ASA (0.1 g) were then added to the ZnO₂ suspension and stirred until fully dissolved. This mixture was mixed with 5 mL HPC solution (0.2 g) under stirring, leading to gelation within 6 min. The CHS@ZnO₂ hydrogel was aged for 1 h at room temperature.

CHS@ZnO₂/Fe³⁺ hydrogel: ZnO₂/Fe³⁺ nanoparticles (0.02 g) were dispersed in 5 mL ultrapure water via ultrasonic treatment for 30 min to form a uniform suspension. AHA (0.1 g) and ASA (0.1 g) were dissolved in this suspension, and the mixture was combined with 5 mL HPC solution (0.2 g) under stirring. Gelation was observed within 4 min, and the CHS@ZnO₂/Fe³⁺ hydrogel was aged for 1 h to stabilize the network structure, with Fe³⁺ ions potentially enhancing crosslinking via coordination with aldehyde/amino groups.

4.4. Characterization of nanoparticles

TEM observations were carried out using a (JEM-2100) transmission electron microscope, which has a column structure with an electron gun at the top, followed by condenser lenses, specimen stage, objective lens, intermediate lenses, and projector lenses, and is equipped with a digital camera for image capture. The instrument operates at an accelerating voltage of 200 kV. For sample preparation, nanoparticle suspensions (0.1 mg/mL in ultrapure water) were drop-cast onto carbon-coated copper grids. The grids were then placed in a desiccator and dried at room temperature to remove water before being loaded into the TEM for observation.

The hydrodynamic particle size distributions and zeta potentials of the nanoparticles were determined using a (Malvern Zetasizer Nano ZS) dynamic light scattering (DLS) instrument. This instrument consists of a laser source, a sample cell holder, a detector, and a computer with data analysis software. Measurements were conducted at 25 °C with a scattering angle of 90°. Nanoparticle suspensions (0.1 mg/mL in ultrapure water) were prepared and ultrasonicated for 10 min to ensure uniform dispersion. The suspensions were then transferred to a disposable cuvette (for particle size measurement) or a zeta potential cell (for zeta potential measurement) and placed in the instrument for testing.

4.5. Characterization of hydrogels

Fourier Transform Infrared (FTIR) Spectroscopy analysis was performed using a Fourier transform infrared spectrometer (Nicolet iS50, Thermo Fisher Scientific) in the range of 4000–400 cm⁻¹. Freeze-dried hydrogel samples were mixed with potassium bromide (KBr) at a mass ratio of 1:100, ground into a fine powder, and pressed into pellets for testing. Each sample was scanned 32 times with a resolution of 4 cm⁻¹. The microstructure of the hydrogels was observed using a scanning electron microscope (SEM, SU8010, Hitachi). Freeze-dried hydrogel samples were fractured under liquid nitrogen to expose the cross-section, then sputter-coated with a thin layer of gold (thickness: ∼10 nm) to improve conductivity. SEM images were captured at an accelerating voltage of 5 kV.

4.6. Degradation performance test

Hydrogel samples (weight: ∼0.2 g) were immersed in 10 mL of phosphate-buffered saline (PBS, pH 7.4) and incubated at 37 °C in a constant temperature shaking incubator (100 rpm). At predetermined time intervals (1, 3, 5, 7, 10, and 14 days), the residual hydrogels were retrieved, rinsed with ultrapure water, freeze-dried, and weighed. The degradation rate was calculated using the formula:

Degradation rate (%) = [(Initial weight - Residual weight)/Initial weight] × 100 %. Each group was tested in triplicate.

4.7. Rheological analysis

Rheological measurements were conducted using a rheometer (Anton Paar MCR 302) with a parallel plate geometry (diameter: 20 mm, gap size: 1 mm). All tests were performed at 37 °C. For dynamic frequency sweep, the storage modulus (G′) and loss modulus (G″) were recorded in the frequency range of 0.1–10 Hz under a constant strain of 1 % (within the linear viscoelastic region, pre-determined by strain sweep test). For the CHS@ZnO₂/Fe³⁺ hydrogel group, an additional rheological test was performed after adding 50 μL of dilute hydrochloric acid (0.1 M) to the hydrogel surface to adjust the local pH to ∼5.0. The dynamic frequency sweep was repeated under the same conditions (frequency: 0.1–10 Hz, strain: 1 %, temperature: 37 °C) immediately after acid addition.

4.8. Compressive mechanical testing

Compressive mechanical properties of the hydrogels were measured using a universal testing machine (Instron 5967) equipped with a 500 N load cell. Hydrogel samples were prepared into cylindrical shapes with a diameter of 8 mm and a height of 10 mm using a custom mold. Prior to testing, the samples were equilibrated in phosphate-buffered saline (PBS, pH 7.4). In contrast, the CHS@ZnO₂/Fe³⁺ hydrogel was equilibrated in PBS (pH 5.4) at 37 °C for 24 h to achieve complete hydration. The compressive test was conducted at a crosshead speed of 1 mm/min until the sample reached 70 % strain. The compressive stress–strain curves were recorded, and key parameters such as compressive strength (maximum stress at 70 % strain) and compressive modulus (slope of the linear region in the stress–strain curve, typically between 10 % and 30 % strain) were calculated. Each group was tested with at least five replicate samples to ensure data reliability.

4.9. Antioxidant experiment

NIH-3T3 cell line (3T3 cell), human umbilical vein endothelial cells line (HUVEC) and RAW264.7 cell line were purchased from Procell Life Science & Technology Co., Ltd. These cell lines were cultured in DMEM supplemented with 1 % streptomycin/penicillin and 10 % fetal bovine serum (FBS, Gibco) in a cell incubator (37 °C, humidified atmosphere of 5 % CO2).

RAW264.7 cells (5 × 10⁴ cells) were seeded on the surfaces of various hydrogels. The cells were then stimulated with H₂O₂ (100 μM) for 12 h. Subsequently, the cells were stained using a ROS fluorescent probe detection kit (Beyotime, China) for 30 min. Fluorescent images were observed and captured under a fluorescence microscope.

4.10. In vitro biocompatibility assay

RAW264.7 cells were seeded on the surfaces of various hydrogels at a density of 5 × 10⁴ cells per well. After culturing for 1 and 3 days, the culture medium was removed, and the cells were gently washed once with PBS. Following the manufacturer's instructions, a live/dead viability staining solution was added, and the cells were incubated in the dark at 37 °C for 20 min. Stained cells were observed and imaged using a DM8 fluorescence microscope. Viable cells exhibited green fluorescence due to staining with Calcein-AM, whereas dead cells exhibited red fluorescence due to staining with PI (Beyotime, China).

Additionally, cell viability was assessed using the Cell Counting Kit-8 (CCK-8) assay. RAW264.7 cells were co-cultured with the different hydrogels for 1, 3, and 5 days. Subsequently, CCK-8 reagent was added to each well, and the plates were incubated in the dark at 37 °C for 1 h. The optical density (OD) value of the solution in each well was then measured at a wavelength of 450 nm using a microplate reader.

4.11. Efferocytosis assay

In accordance with previous studies, RAW264.7 cells were seeded on the surfaces of various hydrogels at a density of 5 × 10⁴ cells per well and cultured for 24 h. The cells were then treated with LPS (1 μg/ml) for 24 h. Apoptotic cells were labeled with CFSE (MCE, HY-D0938) according to the manufacturer's instructions. Subsequently, the CFSE-labeled apoptotic cells were co-cultured with macrophages at a ratio of 5:1 (apoptotic cells to macrophages) for 2 h. After three gentle washes with PBS, the cells were fixed with 4 % paraformaldehyde at room temperature for 15 min. Immunofluorescence staining was performed using an anti-F4/80 antibody (Biolegend, 123109) overnight at 4 °C. Phagocytosis of apoptotic cells was finally assessed by fluorescence microscopy.

4.12. Immunofluorescence staining for macrophage polarization

RAW264.7 macrophages were seeded onto the surfaces of various hydrogels at a density of 5 × 10⁴ cells per well and stimulated with LPS (1 μg/ml) for 12 h. After 24 h of total co-culture with the hydrogels, cells were washed with PBS, fixed, and permeabilized. Cells were then incubated overnight at 4 °C with primary antibodies specific for either iNOS Rabbit mAb (Abclonal, A3774) or CD206 Rabbit mAb (Proteintech, 18704-1-AP). Following PBS washes, corresponding fluorescently labeled secondary antibodies were applied and incubated at room temperature for 2 h. Finally, F-actin was stained with FITC-conjugated phalloidin, and nuclei were counterstained with DAPI. Fluorescence microscopy was employed to visualize iNOS⁺ or CD206⁺ macrophages, with simultaneous assessment of cellular morphology.

4.13. Western blot analysis

Protein extraction from RAW264.7 cells was performed using RIPA lysis buffer (Beyotime). Samples containing equal protein amounts underwent electrophoretic separation on 4 %–20 % gradient SDS-PAGE gels at 120 V. Following electrophoresis, proteins were electrotransferred onto PVDF membranes via wet transfer methodology. Membranes were then immersed in blocking buffer for 2 h at ambient temperature. Primary antibody incubation proceeded overnight at 4 °C according to manufacturer-recommended dilutions. Finally, membranes were treated with horseradish peroxidase-conjugated secondary antibodies (Beyotime) for 1 h at room temperature. The specific antibody information is as follows: Piezo1 (1:1000, Proteintech, 28511-1-AP), YAP (1:5000, Proteintech, 13584-1-AP), α-SMA (1:5000, Proteintech, 14395-1-AP), COL1 (1:2000, Abclonal, A16891), COL3 (1:800, Proteintech, 22734-1-AP), GAPDH (1:50000, Proteintech, 60004-1-Ig).

4.14. Animal model

All experimental protocols involving animals were reviewed and approved by the Animal Ethics Committee of Shanghai Jiao Tong University. Male Sprague-Dawley (SD) rats weighing 200–250 g were administered a single intraperitoneal injection of freshly prepared streptozotocin (STZ) solution (60 mg/kg body weight) to induce type 1 diabetes mellitus. One week post-induction, rats exhibiting non-fasting blood glucose levels ≥16.7 mmol/L were considered diabetic model successes. Three full-thickness excisional wounds per rat were created on the dorsal skin using a sterile 1-mm biopsy punch. The wounds were randomly assigned to receive intradermal injections of one of the following hydrogels: CHS, CHS@ZnO₂, or CHS@ZnO₂/Fe³⁺. Wounds in the control group received injections of PBS.

4.15. Wound healing assessment

Wound areas were photographed on days 3, 5, 7, 10, and 12 post-treatment. The wound area was quantified using Image J software and expressed as a percentage of the initial wound area. On days 7 and 14 post-wounding, rats were euthanized, and wound tissues along with surrounding skin were harvested, fixed in 4 % paraformaldehyde, paraffin-embedded, and sectioned. Sections were stained with H&E to assess overall wound closure, re-epithelialization, granulation tissue formation, and inflammatory cell infiltration. Collagen deposition and organization within the granulation tissue were visualized and quantified using Masson's trichrome staining. Tissue sections underwent immunofluorescence staining to evaluate specific markers of the healing process and macrophage polarization, and the expression of related healing indicators.

4.16. RNA sequencing and bioinformatics analysis

Full-thickness skin wound tissues were harvested 14 days post-surgery. Total RNA was extracted from the tissues using TRIzol™ reagent (Invitrogen) following the manufacturer's instructions. Qualified RNA samples were used to construct sequencing libraries according to standard protocols. Gene Ontology (GO) enrichment analysis and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis were performed on the significantly DEGs. Terms and pathways with an FDR-adjusted p-value <0.05 were considered statistically significant.

4.17. Statistical analysis

Results were expressed as mean ± standard deviation (SD) from at least three independent experiments. One-way analysis of variance with Tukey's test (Origin 8.0) was used for statistical analysis. The significance of differences was treated as follows: ∗ represents p < 0.05 and ∗∗ represents p < 0.01.

CRediT authorship contribution statement

Zaijin Tao: Writing – original draft, Methodology, Data curation, Conceptualization. Ziyun Li: Writing – original draft, Methodology. Yanxuan Shao: Writing – review & editing, Methodology, Formal analysis, Data curation. Yang Xiao: Writing – review & editing. Xinbin Fan: Methodology, Formal analysis, Data curation. Liuqing Yang: Validation. Zhenyu Sun: Methodology, Conceptualization. Tairong Cui: Validation. Zehou Sun: Methodology, Investigation. Jia Jiang: Supervision, Resources, Conceptualization. Xiaofeng Lian: Writing – review & editing, Resources, Project administration, Data curation. Xuran Guo: Validation, Methodology, Data curation, Conceptualization. Shen Liu: Writing – review & editing, Investigation, Funding acquisition, Conceptualization. Xin Ma: Project administration, Funding acquisition, Conceptualization.

Ethics approval and consent to participate

All animal procedures were approved by institutional review committee of Shanghai Jiao Tong University (SYXK (Hu)2023–0042).

Data and materials availability

All data are available in the main text or the supplementary materials.

Declaration of competing interest

The authors declare no competing interests.

Appendix A. Supplementary data

The following is the Supplementary data to this article: