Bidirectional redox-modulating vanadium carbide-integrated smart microneedles for infected wound healing

Abstract

Wound healing is plagued by an inherent reactive oxygen species (ROS) paradox: elevated ROS are critical for antibacterial defense, yet excessive ROS levels impede inflammation resolution and tissue repair. Conventional single-component nanomaterials lack the capacity for bidirectional ROS regulation, while multi-drug combination strategies suffer from complex regulatory modalities, poor stability, and antimicrobial resistance issues—all of which limit clinical translation. To address these challenges, we develop a vanadium carbide (V₄C₃)-iodinene (Ine)@polyvinyl alcohol (PVA) smart microneedle system. Notably, V₄C₃ within the system exhibits intrinsic bifunctionality: it not only generates ROS under light irradiation to enable photodynamic antibacterial therapy but also scavenges excess ROS after irradiation. Further enhanced by Ine and PVA microneedles, which boost functional adaptability and biocompatibility, this system forms a therapeutic platform with excellent mechanical strength, adaptive wound conformity, and laser-controlled switchable ROS regulation. In vitro and in vivo experiments demonstrate its potent antibacterial activity, its ability to induce M2 macrophage polarization, and its ability to reduce pro-inflammatory cytokine expression, thereby alleviating inflammation and accelerating wound healing. By leveraging V₄C₃'s bifunctionality and combining it with microneedle delivery and laser regulation, the simplified yet intelligent wound-healing platform resolves the ROS paradox, avoids issues such as drug resistance, and holds great potential for clinical translation.

Graphical abstract

Based on the bidirectional ROS-regulating capability of V₄C₃, the V₄C₃-Ine@PVA smart microneedle system, paired with microneedle self-softening and laser control, enables on-demand infected wound therapy. It resolves the ROS demand paradox in antibacterial and anti-inflammatory, while circumventing potential flaws such as insufficient regulatory precision and drug resistance induction of conventional antibacterial-antiinflammatory platforms.

Highlights

• •First demonstration of a single material with bidirectional ROS regulation for both antibacterial and anti-inflammatory functions.
• •Hydrogel mechanical tuning is integrated into the therapeutic workflow for efficient delivery and adaptation.
• •Iodinene serves as a core bridging component to synergize all functional units into one integrated system.

1. Introduction

Wound healing is a highly complex biological process that sequentially undergoes three phases: pathogen clearance (a core component of the inflammatory phase), anti-inflammatory regulation, and tissue regeneration [1]. Among these, pathogen clearance and anti-inflammatory regulation are particularly critical, yet they exhibit contradictory requirements for ROS levels: elevated ROS levels are essential for efficient pathogen eradication [[2], [3], [4]], whereas ROS elimination is required in the anti-inflammatory phase to suppress pro-inflammatory signaling and prevent excessive tissue damage [5,6]. This intrinsic contradiction renders traditional single-component nanomaterials unable to simultaneously achieve efficient antibacterial and anti-inflammatory effects [7], ultimately limiting the overall efficiency of wound healing.

To overcome this bottleneck, research has gradually shifted toward multi-component composite systems, aiming to coordinate antibacterial and anti-inflammatory functions through combined drug administration or sequential release modalities. However, such systems generally suffer from drawbacks, including inadequate regulatory precision [8,9] and the potential induction of bacterial resistance upon long-term use [10,11], which severely hinder their clinical translation. Furthermore, existing multi-component systems often rely on the simple superposition of multiple functional units [[12], [13], [14]], which increases the difficulty and cost of fabrication [15], further compromising therapeutic efficacy.

Relative to these systems, multifunctional materials featuring structural simplicity and straightforward, controllable mechanisms of action offer a novel solution for advanced wound therapy strategies. Among redox-active inorganic materials, conventional metal nitride systems and metal carbides represent two related but structurally distinct platforms. For example, a recent nitride-based wound dressing has demonstrated temporally coordinated ROS generation and removal for infected diabetic wound repair [16]. Compared with such nitride systems, metal carbides possess conductive metal-carbon frameworks and abundant surface terminations, which are particularly favorable for light-triggered electron transfer and enzyme-like ROS regulation [17,18]. Within the carbide category, carbide MXenes represent an important class of two-dimensional metal carbides in the broader MXene family [19] and have garnered considerable attention in biomedicine owing to their unique layered structures [20,21], abundant surface functional groups [22,23], and excellent photothermal and electrical properties [24,25]. Specifically, V₄C₃ exhibits promising bifunctional characteristics for bidirectional ROS regulation—under near-infrared (NIR) laser irradiation, its robust electron transfer capability enables efficient generation of ROS such as superoxide anions (O₂·^-), achieving photodynamic antibacterial activity without relying on conventional antibiotics [12,26,27], thereby fundamentally avoiding the risk of drug resistance. In the absence of laser irradiation, V₄C₃ inherently possesses enzyme-like properties to continuously scavenge various types of ROS and regulate the expression of inflammatory factors, providing support for anti-inflammatory repair [21,28,29]. This bifunctional feature positions it as a potential core material for constructing bidirectional ROS regulation platforms.

However, despite the exciting tunable bifunctional properties of V₄C₃, its efficacy in the complex wound microenvironment is still limited by the slight inherent biotoxicity of nanomaterials [30] and the instability of catalytic surfaces [31]. To address this issue, combining V₄C₃ with suitable auxiliary functional materials to construct heterostructures has been proven to be a promising optimization strategy [[32], [33], [34]]. Among numerous biomedical materials with wound microenvironment-responsive characteristics, Ine stands out due to its unique properties. As a nanosheet form of elemental iodine, Ine can easily form a stable heterostructure with V₄C₃ [35] and effectively circumvent the drawbacks of traditional antibacterial agents prone to inducing drug resistance or causing off-target toxicity through wound microenvironment responsiveness [36]. Through H₂O₂-responsive iodine redox reactions, Ine-associated iodine/iodide species can consume excessive H₂O₂ in the infected wound microenvironment and generate bactericidal iodine species, mainly iodine and hypoiodous acid (HIO). These iodine species can oxidize bacterial membrane components and intracellular biomolecules, thereby disrupting bacterial integrity and enhancing antibacterial activity. In addition, Ine also contributes to alleviating local oxidative stress damage in wounds. The synergistic integration of V₄C₃ and Ine is expected to significantly enhance the overall therapeutic efficacy for infected wound treatment.

Notably, without an appropriate delivery platform, it is challenging for therapeutic agents to exert their efficacy [37,38]. In recent years, hydrogel microneedles, as emerging transdermal drug delivery platforms, possess precise delivery capabilities and unique potential for wound adaptation [39]. However, their clinical translation is limited by the inherent mechanical strength defects of traditional hydrogel materials, which hinder effective penetration of the skin barrier. Fortunately, studies have confirmed that PVA hydrogels can undergo salting-out modification with various ions to adjust their mechanical strength on demand [40,41]. This effectively addresses the technical bottlenecks of traditional hydrogel microneedles, such as fragility during insertion and low transdermal efficiency, while still retaining the function of softening to conform to wounds. Through precise regulation of the dynamic strength of PVA, it is promising to construct an intelligent drug delivery system suitable for complex wounds, laying a robust foundation for the effective exertion of functional components.

Herein, we constructed a V₄C₃-Ine@PVA smart microneedle system, which achieved precise regulation of ROS through dual-functional components and thereby efficiently resolved the long-standing ROS paradox in the field of wound healing. Initially, V₄C₃-Ine served as the core ROS-regulating material in the platform, enabling laser-controllable ROS modulation. Under laser irradiation, V₄C₃ efficiently generated ROS to exert potent bactericidal activity, while Ine responded to the high H₂O₂ levels in the wound microenvironment and in situ produced antibacterial active species such as elemental iodine and hypoiodous acid, synergistically reinforcing bactericidal efficacy. In the absence of laser stimulation, V₄C₃-Ine spontaneously eliminated excessive ROS through multiple antioxidant pathways, including the superoxide dismutase (SOD)-like and catalase (CAT)-like enzyme-mimicking activities of V₄C₃, its direct hydroxyl radical-scavenging capability, and the H₂O₂-responsive redox reaction of Ine, thereby alleviating inflammatory responses. Notably, the iodine generated during this responsive process could effectively soften the PVA hydrogel matrix, enabling the microneedles (MNs) to achieve intimate contact with irregular wound beds and thereby promoting the localized retention and sustained release of functional components. Both in vitro and in vivo experiments corroborated that V₄C₃-Ine MNs exhibited efficient and controllable antibacterial and anti-inflammatory activities and significantly promoted collagen deposition, reduced inflammatory infiltration, and accelerated wound closure. This intelligent platform overcame the limitation of unbalanced ROS regulation in traditional wound healing strategies and provided a novel therapeutic regimen with promising translational potential for clinical wound management.

2. Experimental section

2.1. Materials

The Aluminum vanadium carbide (V₄AlC₃), 3,3′,5,5′-Tetramethylbenzidine dihydrochloride (TMB·2HCl), Tetramethylammonium hydroxide solution (TMAOH), Iodine, 2 - Phenyl - 4,4,5,5 - tetramethylimidazoline - 1 - oxyl - 3 - oxide (PTIO), 2,2′-Azino-Bis(3-Ethylbenzothiazoline-6-Sulfonic Acid) Diammonium Salt (ABTS) and PVA were purchased from Adamas-Beta Co., Ltd. (Shanghai, China). Hydrofluoric acid (HF) was obtained from Macklin Biochemical Co., Ltd. (Shanghai, China). H₂O₂ was purchased from China National Pharmaceutical Group Chemical Reagent Co., Ltd. (Shanghai, China). Horseradish peroxidase (HRP) was purchased from Absin Bioscience Inc. (Shanghai, China). Polydimethylsiloxane microneedle (MN) molds were purchased from Shiling Laike Die Business Co., In (Guangzhou, China).

2.2. Characterization

The morphology and microstructure of the samples were characterized using scanning electron microscopy (SEM) and transmission electron microscopy (TEM). The crystal structure was analyzed by X-ray diffraction (XRD). The elemental composition and chemical valence states were determined by X-ray photoelectron spectroscopy (XPS). The functional groups were identified by Fourier transform infrared spectroscopy (FT-IR) and Raman spectroscopy. The elemental concentrations were quantified using Inductively Coupled Plasma Optical Emission Spectroscopy (ICP-OES) analysis. The photothermal performance of the samples was evaluated using a 1064 nm laser and a handheld infrared thermal imaging camera.

2.3. Synthesis of V₄C₃ MXene

V₄C₃ MXene was prepared by selectively etching V₄AlC₃ powders. First, 1 g of V₄AlC₃ was immersed in 45 mL of hydrofluoric acid (HF, 40%) in a sealed reaction vessel and stirred at room temperature for 4 days to remove the aluminum layers. The resulting product was then repeatedly washed with deionized water by centrifugation (8000 rpm, 6 min) until the pH of the supernatant exceeded 6.

To achieve delamination, 0.5 g of the obtained V₄C₃ was dispersed in 30 mL of tetramethylammonium hydroxide (TMAOH, 40% in water) and stirred at 25 °C for 3 days. The suspension was centrifuged at 8000 rpm to remove unexfoliated particles, followed by washing with deionized water.

The dispersion was subsequently sonicated under an argon atmosphere for 90 min and centrifuged at 10000 rpm to collect few-layered V₄C₃ nanosheets. Finally, the purified nanosheets were further processed using an ultrasonic cell disruptor to obtain thinner nanosheets for subsequent applications.

2.4. Synthesis of iodine nanosheets

Iodine nanosheets were synthesized via liquid-phase exfoliation. Bulk iodine powder (500 mg) was first ground and dispersed in 100 mL of deionized water, followed by ultrasonic treatment for 4 h using an ultrasonic cleaner. The suspension was then left standing overnight to promote sedimentation.

After standing, the upper supernatant was collected and centrifuged twice at 3000 rpm for 5 min to remove residual unexfoliated particles. The iodine nanosheets suspended in the upper layer were harvested for further use.

2.5. Synthesis of V₄C₃-Ine

The prepared V₄C₃ nanosheets were dispersed in 20 mL of deionized water and sonicated for 1 h. Subsequently, 40 mL of iodine nanosheet suspension was added to the dispersion, followed by another 2 h of ultrasonic treatment to ensure uniform mixing.

The resulting mixture was stirred at 400 rpm for 12 h to form a stable and homogeneous heterostructure dispersion. After completion, the dispersion was centrifuged at 3000 rpm for 10 min, and the collected precipitate was designated as the V₄C₃-Ine heterostructure.

2.6. Fabrication of V₄C₃-Ine Hydrogel Microneedles

Initially, PVA powder was dissolved into the previously prepared V₄C₃-Ine heterostructure dispersion under vigorous stirring and heating at 80 °C to prepare a 10 wt% PVA solution. The mixture was then subjected to ultrasonic treatment for 1 h to obtain a homogeneous solution.

Subsequently, the resulting solution was poured into a microneedle mold and filled into the cavities by vacuum-assisted degassing. The filled mold was frozen at −20 °C for 2 h and then thawed at room temperature for 1 h. This freeze-thaw cycle was repeated three times to form the hydrogel microneedles.

After demolding, the microneedles were immersed in 1 M Na₂SO₄ solution and kept static for 12 h to enhance their mechanical strength.

2.7. Characterization of V₄C₃-Ine Hydrogel Microneedles

The morphology of the microneedle patches was examined using a macroscopic camera, optical microscope, and SEM. The patches were then immersed in a solution containing 100 μM H₂O₂ to mimic the hydrogen peroxide-rich wound microenvironment, which induced microneedle softening and subsequent penetration into porcine skin. skin staining experiments were conducted to directly verify the penetration performance. Finally, the mechanical strength of the microneedles was assessed using a universal testing machine to obtain pressure-displacement and stress-time curves. The patches were placed on the bottom fixed stage, while the upper movable stage equipped with a force sensor was driven downward at a constant speed of 0.1 mm min⁻¹ to compress the microneedles, and the real-time variations of force with respect to displacement and time were continuously recorded.

2.8. Evaluation of ABTS·⁺, PTIO· and H₂O₂ scavenging activity

To evaluate the antioxidant capacity of V₄C₃-Ine nanosheets, ABTS·⁺, PTIO· radical and H₂O₂ scavenging assays were conducted.

For the ABTS·⁺ assay, 3 mg of ABTS powder was dissolved in 0.735 mL of deionized water, followed by the addition of 1.43 mL of potassium persulfate (K₂S₂O₈, 1 mg/mL) solution. The mixture was stirred thoroughly and incubated in the dark for 12 h to generate ABTS·⁺ radicals. Prior to use, the reaction solution was diluted with deionized water to an appropriate absorbance at its maximum absorption wavelength.

V₄C₃-Ine sample solutions with different concentration gradients were prepared. In each test, 100 μL of the sample solution was mixed with 900 μL of ABTS·⁺ solution, followed by incubation at room temperature for 6 min. The absorbance at 734 nm was then recorded using a UV-Vis spectrophotometer.

For the PTIO· assay, 3 mg of PTIO· powder was dissolved in 20 mL of deionized water and subjected to ultrasonic treatment for 5 min to prepare the PTIO· stock solution. The stock solution was then diluted and adjusted to an appropriate absorbance at its maximum absorption wavelength to obtain the PTIO· working solution.

Similarly, V₄C₃-Ine sample solutions with different concentration gradients were prepared. In each test, 200 μL of the sample solution was mixed with 800 μL of PTIO· working solution, followed by incubation at 37 °C for 30 min. After incubation, the absorbance at 557 nm was measured to evaluate the PTIO· radical scavenging ability.

Finally, the H₂O₂ scavenging capacity was evaluated using a strategy involving H₂O₂-induced oxidation of TMB. A 0.5% hydrogen peroxide solution was prepared to mimic the H₂O₂ concentration in the wound microenvironment. Different concentrations of V₄C₃-Ine were added to the H₂O₂ solution and co-incubated for 5 min. Subsequently, high-speed centrifugation was performed to remove V₄C₃-Ine particles. TMB was added to the collected supernatant, and after a 2 min incubation, the absorbance at 652 nm was measured using a UV-Vis spectrophotometer to assess the H₂O₂ scavenging efficiency.

2.9. Detection of ROS generation ability of V₄C₃-Ine

The ROS generation ability of V₄C₃-Ine nanosheets was evaluated using a superoxide anion radical assay kit (sulfanilamide colorimetric method). According to the manufacturer's instructions, sample solutions were first mixed with the assay buffer provided in the kit, followed by the addition of the substrate solution. After sufficient reaction, the sulfanilamide reagent was added. The mixture was incubated at room temperature for 10 min in the dark, and the absorbance was then measured at 530 nm. The superoxide anion generation ability was assessed by comparing the absorbance changes among different sample groups. All reagents were freshly prepared, and all procedures strictly followed the assay kit protocol.

2.10. Redox switching cycles of V₄C₃-Ine

To further evaluate the repeatability and controllability of the laser-regulated ROS transition, an alternating ROS generation/scavenging cycling assay was performed. Briefly, V₄C₃-Ine was first incubated with ABTS·⁺ solution under laser-off conditions, and the residual absorbance at 734 nm was recorded to evaluate its ROS scavenging activity. After the ROS-scavenging assay, V₄C₃-Ine was separated by centrifugation and reused for the subsequent ROS-generation assay. Under laser irradiation, the collected V₄C₃-Ine was resuspended in the O₂·^- detection reagent, and the absorbance at 530 nm was recorded to assess laser-triggered ROS generation. After the ROS generation assay, V₄C₃-Ine was collected again by centrifugation and subjected to the next ROS generation/scavenging cycle. This alternating procedure was repeated for five consecutive cycles, and the peak absorbance values at 734 and 530 nm were plotted to monitor the redox-switching behavior.

2.11. Photothermal test

To evaluate the photothermal properties of V₄C₃-Ine nanosheets, aqueous dispersions were prepared and subjected to 1064 nm laser irradiation. For the concentration-dependent experiment, dispersions with different concentrations (1, 2 and 4 mg/mL) were placed in 2 mL centrifuge tubes, and irradiated with a laser at a fixed power density of 1.5 W/cm² for 5 min. For the power-dependent experiment, dispersions with a fixed concentration of 1 mg/mL were irradiated under different power densities (0.25, 1.0 and 1.5 W/cm²) for 5 min.

During irradiation, the temperature change of each sample was monitored and recorded every 30 s using an infrared thermal imaging camera. In addition, the photothermal stability of the V₄C₃-Ine nanosheets was assessed by subjecting the samples to five on/off laser irradiation cycles at 1.5 W/cm². The photothermal conversion efficiency (η) was measured through the following formula [42]:

$$\begin{array}{cc}\theta =\frac{T-T_{\mathit{min}}}{T_{\mathit{max}}-T_{\mathit{min}}}& \tau =\frac{t}{\mathit{ln}\theta }\\ Q=\frac{m\ast c\ast T_{\mathit{max}}-T_{\mathit{min}}}{\tau }& \eta =\frac{Q_{material}-Q_{water}}{I\ast (1-{10}^{-A_{1064}})}\end{array}$$

(t: time, T: temperature, m: quality, c: specific heat capacity, A1064: absorbance at 1064 nm, I: laser power density)

2.12. In vitro evaluation of cytotoxicity, cell proliferation, and ROS-scavenging ability

For cytotoxicity analysis, different dilutions of PVA, V₄C₃-PVA, Ine-PVA, and V₄C₃-Ine-PVA were co-incubated with mouse embryonic fibroblast cell line (NIH3T3) for 24 h, and cell viability was assessed using the CCK-8 assay. Cells treated with 50 μg/mL of each solution were further stained with a Live/Dead Cell Staining Kit (SYTO9-propidium iodide, PI) and observed by confocal laser scanning microscopy (CLSM). In addition, hemocompatibility was evaluated by co-incubating 50 μg/mL of each solution with mouse red blood cells, using PBS and H₂O₂ as the negative and positive controls, respectively.

Cell migration under ROS scavenging was indirectly evaluated by scratch assay, where a straight-line scratch was created using a 200 μL pipette tip. After co-incubation with 50 μg/mL solutions for 24 h, the scratch width and cell proliferation were examined. Intracellular ROS elimination was assessed using the DCFH-DA method by co-culturing NIH3T3 cells with 50 μg/mL of each solution. Fluorescence changes inside cells were continuously monitored using CLSM and flow cytometry (FCM). Furthermore, under the same experimental conditions, mitochondrial membrane potential was measured using the JC-1 assay kit.

2.13. Evaluation of inflammatory cytokine regulation and macrophage polarization

RAW264.7 macrophages were divided into PBS control, LPS-treated, and LPS + V₄C₃-Ine MNs-treated groups. After treatments, total RNA was extracted and reverse-transcribed into cDNA, and qPCR was conducted to determine the mRNA expression levels of inflammatory cytokines (IL-6, IL-1β, TNF-α). To further assess the immunoregulatory effects, cells were collected and subjected to surface antibody staining, followed by flow cytometry to analyze macrophage phenotype markers (F4/80, CD86, CD206) and evaluate the changes in M1/M2 polarization [43].

2.14. In vitro evaluation of antibacterial and antibiofilm performances

MRSA suspensions were prepared and subjected to ten different treatments: (1) PBS, (2) PBS + Laser, (3) PVA, (4) PVA + Laser, (5) V₄C₃-PVA, (6) V₄C₃-PVA + Laser, (7) Ine-PVA, (8) Ine-PVA + Laser, (9) V₄C₃-Ine-PVA, and (10) V₄C₃-Ine-PVA + Laser. The concentration of all solutions was fixed at 50 μg/mL, and laser irradiation was carried out at 1064 nm, 1 W cm⁻² for 5 min.

After treatment, bacterial survival was first evaluated by colony counting assay, while SYTO9-PI staining and crystal violet assay were employed to assess bactericidal efficacy and antibiofilm activity, respectively. To further examine biosafety, the treated bacterial suspensions were co-incubated with mouse erythrocytes, and hemolysis was determined.

The morphology of MRSA from five laser-treated groups was then observed by SEM following PBS washing, fixation with 2.5% glutaraldehyde at 4 °C for 12 h, graded ethanol dehydration (30%, 50%, 70%, 90%, 100%, 10 min each), drying, and gold sputtering.

To probe the antibacterial mechanism, DCFH-DA staining (30 min) was performed on the five laser groups, and intracellular ROS production was analyzed by fluorescence imaging and flow cytometry. Finally, to verify the universality of the antibacterial performance, the same plate-counting and live/dead staining assays were conducted against Pseudomonas aeruginosa.

Additionally, an inhibition-based comparative experiment was performed to distinguish the respective contributions of photodynamic therapy (PDT) and photothermal therapy (PTT) to the antibacterial activity. Briefly, MRSA suspensions were incubated with V₄C₃-Ine samples and divided into four groups: Control, PDT, PTT and PDT + PTT. For the PDT group, the samples were placed on an ice-water bath during 1064 nm laser irradiation. For the PTT group, the bacterial suspensions were pretreated with NaN₃ for 20 min before laser irradiation, thereby suppressing ROS-mediated photodynamic antibacterial effects while retaining the photothermal effect. The PDT + PTT group was irradiated under the same laser conditions without cooling or ROS scavenging. After treatment, bacterial viability was evaluated by SYTO9/PI live/dead staining and MTT assay.

2.15. In vivo therapeutic evaluation

All animal procedures were reviewed and approved by the Experimental Animal Management and Ethics Committee of Shanghai Tenth People's Hospital (approval number: [SHDSYY-2024-4840]). Female BALB/c mice (5-6 weeks old) were employed to investigate the wound healing performance of microneedle (MN) patches in vivo. Circular full-thickness wounds of approximately 1 cm in diameter were generated and inoculated with MRSA for 24 h. The animals were then randomly allocated into five groups (n = 5 per group): (1) PBS, (2) PVA MNs, (3) V₄C₃ MNs, (4) Ine MNs, and (5) V₄C₃-Ine MNs. On day 1, all groups received MN patch administration followed by laser treatment (1064 nm, 1 W cm⁻², 5 min).

Photographs of the wounds were taken on days 0, 1, 3, 5, 7, 9 and 11, and the wound areas were quantified using ImageJ. On day 3, bacterial colonies from the wound tissues were enumerated on LB agar plates. The harvested wound samples were further subjected to histological and immunohistochemical assessments, including H&E staining, Masson's trichrome staining, TNF-α and CD31 immunohistochemistry [44]. In addition, the polarization of macrophages was detected by flow cytometry. On day 7, similar histological and immunological analyses were carried out, and reactive oxygen species (ROS) levels at the wound sites were determined using dihydroethidium staining. Finally, the major organs were examined by H&E staining, and blood samples were collected for serum biochemical analyses.

2.16. Statistical analysis

Data analysis was carried out using Microsoft Excel 2021, Origin 2024, and IBM SPSS Statistics 27. All results are presented as mean ± standard deviation (SD). Statistical differences between groups were evaluated with a two-tailed Student's t-test, where ∗p < 0.05, ∗∗p < 0.01 and ∗∗∗p < 0.001 were considered statistically significant.

3. Results and discussion

3.1. Synthesis and Characterizations of V₄C₃-Ine Hydrogel Microneedles

The V₄C₃-Ine heterostructure was fabricated via a solution-phase self-assembly strategy (Scheme 1): First, few-layered V₄C₃ nanosheets were obtained by Tetramethylammonium hydroxide solution (TMAOH)-assisted delamination of HF-etched Aluminum vanadium carbide (V₄AlC₃) powder, followed by sonication and gradient centrifugation. Subsequently, bulk iodine was exfoliated in deionized water via sonication to yield iodine nanosheets. The as-prepared V₄C₃ and iodine nanosheets were mixed at a volume ratio of 1:2, a stable heterostructure dispersion was formed via sonication and magnetic stirring, and V₄C₃-Ine was harvested for subsequent characterizations after centrifugation and purification.

Scheme 1.

Schematic illustration of the preparation of V₄C₃-Ine microneedles and their laser-triggered bidirectional redox regulation for integrated wound therapy, including V₄C₃-mediated ROS generation/scavenging and H₂O₂-responsive Ine antibacterial activity through iodine species formation.

Transmission electron microscopy (TEM) images clearly revealed the morphologies of the intermediate products and the final heterostructure, demonstrating that V₄C₃-Ine was exhibited as a sheet-like thin-layer structure (Fig. 1B and C, Fig. S1). The corresponding elemental mapping images further confirmed the homogeneous distribution of V, C, and I, indicating the successful assembly of iodine nanosheets onto the V₄C₃ surface (Fig. 1D). Scanning electron microscopy (SEM) images showed that the lamellar morphology was preserved, with a submicron-scale lateral size (Figs. S2–S5). Atomic force microscopy (AFM) characterization demonstrated that V₄C₃-Ine existed as nanosheets with a thickness ranging from 1 to 4 nm (Fig. 1E), which further corroborated the successful synthesis of the target material. To further assess the morphological stability of V₄C₃-Ine after long-term storage, the aged sample was characterized by TEM. The images showed that V₄C₃-Ine still maintained a typical sheet-like morphology without obvious structural collapse or severe aggregation, indicating the favorable morphological stability of the heterostructure (Fig. S6).

Fig. 1.

Synthesis and Characterizations of V₄C₃-Ine Hydrogel Microneedles. (A) Fabrication process of V₄C₃-Ine Hydrogel microneedles; (B) TEM image of V₄C₃ (C) TEM image of V₄C₃-Ine; (D) Elemental mapping image of V₄C₃-Ine; (E) AFM image of V₄C₃-Ine; (F) Full XPS spectrum of V₄C₃-Ine; (G) high-resolution I 3d spectrum of V₄C₃-Ine; (H) high-resolution V 2p spectrum of V₄C₃-Ine; (I) XRD pattern of V₄C₃-Ine; (J-L) SEM images of V₄C₃-Ine MNs from different perspectives. (M) Optical image showing the bending strength of V₄C₃-Ine MNs (N) H&E-stained histological section of porcine skin after insertion of V₄C₃-Ine MNs before Na₂SO₄ treatment. (O) H&E-stained section after Na₂SO₄-induced ionic crosslinking. (P) H&E-stained section after treatment with low-concentration H₂O₂. (Q) Stress-displacement curves of the three types of MNs.

X-ray diffraction (XRD) analysis indicated that the V₄C₃-Ine composite retained the characteristic diffraction peaks of V₄C₃. Additional diffraction peaks, which were attributed to crystalline iodine, were also detected, thus confirming the structural integrity of both phases (Fig. 1I). X-ray photoelectron spectroscopy (XPS) analysis was conducted to elucidate the valence states of I and V (Fig. 1F). Specifically, the high-resolution I 3d spectrum (Fig. 1G) verified the presence of I⁰ whereas the V 2p spectrum (Fig. 1H) displayed mixed oxidation states. These mixed states were ascribed to partial surface oxidation, a phenomenon that was anticipated to enhance the ROS scavenging capacity of V₄C₃. Additionally, Fourier transform infrared spectroscopy (FT-IR) and Raman spectroscopy were performed to verify the surface chemistry and structural integrity of V₄C₃-Ine. The broad FT-IR band at approximately 3300 cm⁻¹ was attributed to surface –OH groups, while the characteristic absorption around 550-600 cm⁻¹ corresponded to V–C vibrations, suggesting that the V₄C₃ framework was retained after Ine loading. Raman spectra further showed that the characteristic vibration signals of V₄C₃ were preserved in V₄C₃-Ine, confirming the structural integrity of the fabricated heterostructure (Fig. S7). Collectively, these findings corroborate the successful construction of the V₄C₃-Ine heterostructure, which featured well-preserved phase components and robust interfacial interactions.

To achieve precise delivery and improve wound adaptability, PVA was selected as the matrix for loading V₄C₃-Ine to fabricate hydrogel microneedles: Experimental validation demonstrated that PVA effectively improved the dispersibility of V₄C₃-Ine (Fig. S8), and the mechanical strength of the resulting hydrogel microneedles can be regulated via ionic interactions. Four types of microneedle patches were fabricated in this work, with a focus on characterizing V₄C₃-Ine-loaded microneedles; their fabrication process is illustrated in Fig. 1A. Each patch consisted of a 10∗10 microneedle array on a 12.5 mm∗12.5 mm base, with conical needles (600 μm in height, 230 μm in basal diameter, and 1 mm in inter-needle spacing); SEM images further confirmed the uniform conical morphology (Fig. 1J–L). Inductively Coupled Plasma Optical Emission Spectroscopy (ICP-OES) analysis showed that the average loading amount of V₄C₃-Ine in each microneedle patch was 60.212 μg. Based on element-specific quantification, the mass fractions of V₄C₃ and Ine were calculated to be 68.655% and 31.345%, respectively.

Regarding mechanical performance, pristine PVA hydrogel microneedles exhibited insufficient strength and failed to penetrate the skin effectively (Fig. 1N); after soaking in Na₂SO₄ solution, ionic crosslinking substantially enhanced their mechanical strength, enabling successful insertion into porcine skin (Fig. 1O). Notably, attenuation of the mechanical strength of the microneedle tips was observed following skin insertion (Fig. 1P), which was ascribed to the inhibitory effect of iodide ions on PVA's salting-out process and the resultant reduction in ionic interactions of the hydrogel network. This property endowed the microneedles with initial rigidity (skin penetration) and post-insertion flexibility (in vivo drug release). Compression tests were performed on microneedles subjected to distinct treatment protocols, yielding corresponding stress-displacement curves. These curves not only provided direct evidence for the tunable mechanical strength of the microneedles but also corroborated the efficacy of the rigid-soft transition design strategy (Fig. 1Q). Furthermore, cyclic compression-recovery tests and strain-dependent apparent compressive modulus analysis were performed to further verify the mechanically tunable behavior of the MNs (Fig. S9).

3.2. ROS Modulation and Photothermal Properties of V₄C₃-Ine Heterostructures

To comprehensively characterize the redox behavior and photothermal performance of V₄C₃-Ine, its ROS-modulating activities and photothermal conversion efficiency were systematically probed. First, V₄C₃-Ine composite was confirmed to exhibit strong light absorption capacity via ultraviolet absorption spectroscopy measurements, and this property endowed the material with both photothermal and photodynamic responsiveness under laser irradiation (Fig. S10). The O₂·^- generation capability of V₄C₃-Ine nanosheets was then assessed using a sulfonamide-based colorimetric assay kit. As the concentration increased from 0.1 to 1.0 mg/mL, the absorbance at 530 nm gradually intensified, indicating efficient superoxide production upon laser irradiation (Fig. 2A). Furthermore, the radical scavenging ability of V₄C₃-Ine was evaluated using ABTS·⁺ and PTIO· systems. As shown in Fig. 2B, the absorbance of ABTS·⁺ solution at ∼750 nm decreased progressively with increasing V₄C₃-Ine concentration (0.25, 0.5, and 1 mg/mL), suggesting excellent antioxidant properties. Similarly, PTIO· absorption at ∼560 nm diminished in a dose-dependent manner (Fig. 2C), further confirming the effective ROS scavenging capability. Additionally, TMB was employed to verify the scavenging effect on H₂O₂. As the concentration of V₄C₃-Ine increased, the absorbance of the solution at 652 nm gradually decreased, demonstrating the concentration dependence of its H₂O₂-scavenging activity (Fig. 2D). To further verify the repeatability and controllability of ROS switching, an alternating ROS generation/scavenging cycling assay was performed, and the corresponding data are provided in Fig. S11. V₄C₃-Ine maintained relatively stable ROS generation and scavenging capacities over five consecutive cycles, and the slight signal decrease during cycling was likely associated with minor sample loss during repeated centrifugation and recovery processes. Moreover, high-resolution V 2p XPS spectra of V₄C₃-Ine obtained without laser irradiation, immediately following laser irradiation, and at 24 h post-irradiation revealed laser-induced redistribution and subsequent partial recovery of V valence states, indicating that multivalent V sites dynamically participated in the electron-transfer process and supported the nanozyme-like redox-regulating behavior of V₄C₃-Ine (Fig. S12). Beyond cyclic controllability, the long-term stability of this ROS-regulating function was further evaluated after storage. The aged V₄C₃-Ine maintained substantial ROS-generation capability and retained effective scavenging activity toward multiple radical species (Fig. S13). Together with the above results, these findings demonstrate that V₄C₃-Ine features synergistic and repeatable ROS regulatory capability.

Fig. 2.

Reactive Oxygen Species (ROS) Modulation and Photothermal Properties of V₄C₃-Ine Heterostructures (A) Detection of O₂·^- generation ability; (B) ABTS·⁺ scavenging assay; (C) PTIO· scavenging assay; (D) H₂O₂-scavenging assay based on TMB oxidation (E) Thermal images of solutions with different concentrations under irradiation over time; (F) Temperature elevation profiles of solutions with different concentrations; (G) Heating/cooling cycles of V₄C₃-Ine dispersions; (H) Photothermal conversion efficiency.

In addition, the photothermal conversion performance of V₄C₃-Ine under laser irradiation was systematically evaluated using thermal imaging. The dispersions exhibited pronounced photothermal heating effects (Fig. 2E), with clear concentration-dependent (Fig. 2F) and power-dependent (Fig. S14) behaviors. The heating rate increased proportionally with both concentration and laser power. Notably, repeated on/off irradiation cycles (Fig. 2G) yielded consistent heating profiles, and the calculated photothermal conversion efficiency reached 37.99% (Fig. 2H), demonstrating the robust photothermal performance and stability of the heterostructure. It was verified through the above experiments that V₄C₃-Ine featured synergistic ROS regulatory capability and high-efficiency photothermal performance, which endowed it with promising prospects in the construction of multifunctional therapeutic nanoplatforms.

3.3. In Vitro Biosafety, antioxidant, and anti-inflammatory properties of V₄C₃-Ine MNs

Considering the practical application of V₄C₃-Ine MNs, their biosafety was evaluated. NIH3T3 cells were incubated with different concentrations of V₄C₃-Ine-PVA solutions for 24 h. Cell Counting Kit-8 (CCK-8) assays showed that high concentrations of V₄C₃ MNs resulted in decreased cell viability, while the V₄C₃-Ine MNs group exhibited negligible cytotoxicity. This was likely attributed to the enhanced stability of V₄C₃ after complexing with iodine nanosheets, which reduced the toxicity from degradation (Fig. 3A). Additionally, the Live/Dead assay showed prominent green fluorescence, confirming the excellent biocompatibility of V₄C₃-Ine MNs (Fig. 3B). Moreover, co-incubation of red blood cells with MN leaching solutions did not cause notable hemolysis (Fig. S15). In the scratch assay, exogenous H₂O₂ markedly inhibited cell migration. Upon treatment with V₄C₃-Ine MNs, this oxidative stress-induced inhibition was effectively alleviated, restoring cell migratory capacity and leading to the narrowest wound width, thereby accelerating scratch wound healing. These findings further highlight the potent ROS-scavenging capability of V₄C₃-Ine MNs (Fig. 3C).

Fig. 3.

In Vitro Biosafety, Antioxidant, and Anti-inflammatory Properties of V₄C₃-Ine MNs. (A) Cell viability of NIH3T3 after incubation with different concentrations of solution for 24 h. (B) Live/dead cell assay. (C) Scratch width. (D) Intracellular reactive oxygen species (ROS) level image and mitochondrial membrane potential image. (E) Relative fluorescence intensity. (F) Relative fluorescence area of J-aggregate. (G) Expression levels of IL-6, TNF-α, and IL-1β analyzed by qRT-PCR. (H) FCM of CD80 (M1 marker) and CD206 (M2 marker) levels in RAW264.7. Data are presented as mean ± SD (n = 3), ∗p < 0.05, ∗∗p < 0.01, and ∗∗∗p < 0.001, ns: no significance.

Further investigation into the ability of V₄C₃-Ine MNs to scavenge intracellular reactive oxygen species (ROS) revealed that compared to the hydrogen peroxide group, the green fluorescence intensity in the V₄C₃-Ine MNs group was markedly reduced, demonstrating a marked decrease in ROS levels in NIH3T3 cells (Fig. 3D–F). This reduction in ROS was accompanied by a prominent increase in the red fluorescence intensity, suggesting that V₄C₃-Ine MNs could stabilize mitochondrial membrane potential and protect cells from oxidative stress.

As ROS are crucial signaling molecules, elevated levels were found to activate inflammatory signaling pathways and induce increased expression of pro-inflammatory cytokines such as TNF-α, IL-1β and IL-6. Given the efficient ROS-scavenging capacity of V₄C₃-Ine MNs, the expression levels of these inflammatory cytokines were assessed via quantitative real-time PCR (qRT-PCR). The results demonstrated that V₄C₃-Ine MNs significantly suppressed the production of major pro-inflammatory cytokines in RAW264.7 cells (Fig. 3G). Additionally, to verify the impact of V₄C₃-Ine MNs on macrophage polarization, multi-color flow cytometry (FCM) was used to evaluate the polarization status of RAW264.7 cells. Specifically, V₄C₃-Ine MNs effectively promoted the M2 polarization of macrophages (Fig. 3H), which was beneficial for reducing inflammation and promoting tissue repair.

Taken together, V₄C₃-Ine MNs demonstrated excellent biocompatibility, antioxidant, and anti-inflammatory properties in vitro. They were able to modulate the reduction of inflammatory cytokines and promote macrophage polarization, which highlighted their potential for therapeutic applications aimed at tissue repair and inflammation regulation.

3.4. In Vitro Antibacterial Activity Test

In addition to its anti-inflammatory properties, V₄C₃-Ine MNs also demonstrate an efficient ability to generate ROS under laser irradiation, enhancing its antibacterial potential. In this study, Methicillin-Resistant Staphylococcus aureus (MRSA) was selected as the primary pathogen to assess antibacterial and antibiofilm activity (Fig. 4A). Standard colony counting results (Fig. 4B and C) and live/dead fluorescence images (Fig. 4I) revealed that PVA had negligible antibacterial activity, while measurable antibacterial effects were observed for V₄C₃ MNs, Ine MNs, and V₄C₃-Ine MNs. After laser irradiation, the V₄C₃ MNs + Laser group and V₄C₃-Ine MNs + Laser group showed a significant increase in bacterial death, indicating that V₄C₃ nanosheets had a pronounced photodynamic sterilization effect. Importantly, under combined antibacterial treatment (V₄C₃-Ine MNs + Laser group), the antibacterial performance exceeded that of individual oxidative antibacterial therapy (Ine MNs group) or PDT (V₄C₃ MNs + Laser group).

Fig. 4.

In Vitro Antibacterial Activity Test (A) A schematic illustration showing the antibacterial and antibiofilm activities of V₄C₃-Ine MNs. (B) Representative images of MRSA colonies on agar plates after different treatments, (C) corresponding bacterial viability. (D) Hemolysis of red blood cells after co-incubation with MRSA in different groups. (E) Hemolysis levels. (F) Macroscopic MRSA biofilm images with crystal violet staining. (G) 3D reconstruction images of bacterial live (green fluorescence)/dead (red fluorescence) staining of MRSA biofilm. (H) SEM images of MRSA under different treatments. (I) Live/dead staining. (J) ROS levels. (K) Fluorescence images of DCFH-DA-stained MRSA after different treatments. Data are presented as mean ± SD (n = 3), ∗p < 0.05, ∗∗p < 0.01, and ∗∗∗p < 0.001.

Further validation of antibiofilm activity was performed using 3D live/dead fluorescence (Fig. 4G) and crystal violet staining (Fig. 4F). The results showed that biofilms in the Control, PVA MNs, and Ine MNs groups remained mostly intact and viable, even after laser irradiation. In contrast, the V₄C₃ MNs + Laser group and V₄C₃-Ine MNs + Laser group disrupted the tight biofilm structure, with more dead bacteria observed. Notably, the V₄C₃-Ine MNs group with laser irradiation exhibited the best antibiofilm activity.

Hemolysin, a major pathogenic factor, serves as a representative indicator of antibacterial activity by reflecting MRSA toxicity. After centrifugation of treated bacteria, the hemolysin levels in the supernatant were assessed for their ability to lyse red blood cells. Encouragingly, both qualitative (Fig. 4D) and quantitative (Fig. 4E) evaluations showed that hemolysin secretion by MRSA was substantially suppressed in the V₄C₃-Ine MNs + Laser group, further highlighting the antibacterial potential of V₄C₃-Ine MNs. Scanning electron microscopy (SEM) imaging was used to visualize the morphological changes of MRSA. As shown in Fig. 4H, MRSA in the Control and PVA MNs groups displayed relatively smooth surfaces with intact membrane structures, indicating that neither laser irradiation nor the matrix material of the MNs alone produced antibacterial effects. In contrast, after laser treatment, the other three groups exhibited shrunken membrane structures, confirming the damage or death of MRSA.

Additionally, the V₄C₃-Ine MNs + Laser group exhibited the highest number of dead bacteria and the elevated ROS levels (Fig. 4J and K), further validating the enhanced photodynamic therapy (PDT) efficacy. To investigate the broad-spectrum antibacterial activity of the microneedles against other bacterial strains, standard colony counting and live/dead fluorescence imaging were performed on Pseudomonas aeruginosa (Fig. S16). The results were consistent with those obtained for MRSA, providing further evidence of the system's high-efficiency antibacterial properties.

To further clarify whether the antibacterial effect was mainly derived from PDT or PTT, we performed an additional mechanism-dissection experiment by selectively suppressing photothermal or photodynamic effects (Fig. S17). Live/dead staining revealed that red fluorescence increased in the PDT-only and PTT-only groups, although to different extents, demonstrating that both therapeutic modalities were involved in bacterial killing. Compared with the PTT-only group (74.9%), the PDT-only group (62.4%) exhibited lower bacterial viability, indicating that PDT made a greater contribution to the antibacterial effect under these conditions. Moreover, the combined PDT + PTT group showed the most pronounced red fluorescence and the lowest bacterial viability of 7.5%, confirming the synergistic antibacterial efficacy of photodynamic ROS generation and photothermal injury. These results demonstrate that both PDT and PTT contribute to the antibacterial effect of V₄C₃-Ine. PDT accounts for the major contribution, whereas PTT provides an auxiliary and synergistic effect. This finding is consistent with our original design rationale of V₄C₃-Ine for light-controlled redox regulation.

Overall, these results underscore the effectiveness of V₄C₃-Ine MNs in employing a self-reactive and laser-excited ROS amplification mode for efficient sterilization and biofilm elimination.

3.5. In vivo assessment of therapeutic efficacy

After the remarkable antibacterial and anti-inflammatory effects in vitro were confirmed, systematic in vivo investigations were further conducted. A murine skin wound model infected with MRSA biofilms was established to evaluate the therapeutic efficacy of V₄C₃-Ine MNs (Fig. 5A). Five groups were designed: PBS, PVA MNs, V₄C₃ MNs, Ine MNs and V₄C₃-Ine MNs. All groups were subjected to laser irradiation (1 W/cm², 5 min), and wound healing progression was monitored on days 1, 3, 5, 7, 9 and 11. Representative wound photographs (Fig. 5B), quantitative wound area measurements (Fig. 5E), and bacterial survival assays (Fig. 5C–F) consistently indicated that the V₄C₃-Ine MNs group achieved faster wound closure and more pronounced antibacterial efficacy than the other groups.

Fig. 5.

(A) Schematic illustration of MRSA-infected wound model establishment and treatment timeline. (B) Representative images of wound healing progression in different treatment groups (PBS, PVA MNs, V₄C₃ MNs, Ine MNs, and V₄C₃-Ine MNs) over 11 days, along with corresponding wound area measurements. (C) Photographs of bacterial colonies from wound tissues after treatment. (D) Body weight changes of mice during the experimental period. (E) Quantitative analysis of wound area reduction at different time points. (F) Relative bacterial survival rates in wound tissues across treatment groups. (G) ROS fluorescence staining of wound tissue sections on day 7. (H) Flow cytometry analysis of macrophage polarization on days 3 and 7. Data are presented as mean ± SD (n = 5), ∗p < 0.05, ∗∗p < 0.01, and ∗∗∗p < 0.001.

To probe the anti-inflammatory mechanism more comprehensively, wound tissues were harvested on day 7 for ROS level assessment using dihydroethidium staining. As expected, the red fluorescence intensity in the V₄C₃-Ine MNs group was markedly lower than in all other groups (Fig. 5G), indicating that ROS levels were effectively suppressed during the post-sterilization healing phase, thereby highlighting a strong anti-inflammatory effect.

In addition, flow cytometric analysis was performed on days 3 and 7 to evaluate macrophage polarization (Fig. 5H). On day 3 (antibacterial phase), the V₄C₃-Ine MNs group contained the highest proportion of M1 macrophages (Q3), whereas on day 7 (anti-inflammatory phase), this group exhibited the largest population of M2 macrophages. This dynamic transition clearly demonstrated that V₄C₃-Ine MNs possessed dual antibacterial and anti-inflammatory functions.

To better elucidate the mechanisms by which V₄C₃-Ine MNs promote wound healing, additional histological analyses were conducted on day 7, focusing on epidermal regeneration, collagen deposition, inflammatory factor expression, and angiogenesis. H&E staining revealed a significantly thickened epidermis and reduced inflammatory cell infiltration in the V₄C₃-Ine MNs group, suggesting effective suppression of local inflammation and accelerated wound repair (Fig. 6A–F). Subsequent Masson's trichrome staining further confirmed that V₄C₃-Ine MNs elicited the most prominent collagen fiber deposition within the regenerative tissue microenvironment (Fig. 6B–G). In parallel with enhanced tissue remodeling, CD31 immunohistochemistry revealed the densest microvascular networks in wounds treated with V₄C₃-Ine MNs, indicating markedly promoted angiogenesis during the repair process (Fig. 6C–H). Moreover, TNF-α immunohistochemical staining showed the weakest brown signal in the V₄C₃-Ine MNs group, demonstrating a notable reduction in pro-inflammatory cytokine expression and the establishment of a microenvironment favorable for wound healing (Fig. 6D).

Fig. 6.

Histological and immunohistochemical staining of wound tissues at day 7, including (A) H&E staining, (B) Masson's trichrome staining, (C) CD31 immunohistochemical staining, and (D) TNF-α immunohistochemical staining. Quantitative analyses include (E) macrophage polarization, (F) epidermal thickness, (G) collagen fiber deposition, and (H) blood vessel density in wound tissues. Data are presented as mean ± SD (n = 5), ∗p < 0.05, ∗∗p < 0.01 and ∗∗∗p < 0.001.

On this basis, histological analyses were also performed on day 3 wound tissues (Figs. S18 and S19). The overall group differences were consistent with the day 7 results, with V₄C₃-Ine MNs showing the most favorable outcomes across H&E, Masson, TNF-α, and CD31 staining. Comparison between day 3 and day 7 results revealed that the V₄C₃-Ine MNs group exhibited lower inflammatory factor levels at later stages, accompanied by greater collagen deposition and enhanced neovascularization. Collectively, these findings confirm that V₄C₃-Ine MNs exhibit the capacity to accelerate the wound healing process by persistently suppressing inflammation while promoting collagen synthesis and angiogenesis.

Lastly, systemic safety was corroborated through biochemical analysis and histological examination of major organs (heart, liver, spleen, lung, and kidney), with no apparent hematological toxicity or pathological abnormalities identified (Figs. S20–22). Taken together, these results demonstrate that V₄C₃-Ine MNs, under laser irradiation, achieve efficient antibacterial activity and sustained anti-inflammatory effects, while also exhibiting excellent biocompatibility and minimal systemic toxicity.

4. Conclusion

In this study, targeting the contradictory requirements for ROS during the antibacterial and anti-inflammatory phases of wound healing, we constructed an intelligent microneedle therapeutic platform based on bifunctional V₄C₃. By leveraging the intrinsic properties of their key components, this platform achieved bidirectional ROS regulation under NIR light irradiation switching, overcoming the inherent limitations of traditional multi-component systems—including complex regulatory modalities, insufficient stability, and the risk of inducing antimicrobial resistance. The V₄C₃-Ine heterostructure fully exploited the inherent bifunctionality of V₄C₃: under NIR light irradiation, it enabled efficient ROS generation to exert potent antibacterial effects, while upon cessation of irradiation, it scavenged excess ROS to suppress inflammatory responses. Complementing this core bifunctionality, the incorporation of Ine further enhanced therapeutic efficacy by consuming ROS to generate bactericidal HIO and I₂. Meanwhile, PVA hydrogel microneedles served as an optimal delivery system, realizing precise therapeutic agent release and excellent adaptability to irregular wound surfaces. Both in vitro and in vivo experiments consistently confirmed that the V₄C₃-Ine microneedles not only exhibited broad-spectrum antibacterial activity and efficient biofilm eradication capability but also modulated inflammatory responses, promoted collagen deposition, and accelerated angiogenesis—ultimately leading to significant enhancement of wound healing. More importantly, the platform demonstrated excellent biocompatibility and systemic safety, laying a solid foundation for potential clinical translation. In summary, this study proposed a novel bidirectional ROS regulatory therapeutic strategy based on bifunctional vanadium carbide, highlighting its great application potential in advancing wound healing treatments.

Data availability statement

The data that support the findings of this study are available in the supplementary material of this article.

CRediT authorship contribution statement

Zhichao Yao: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Supervision, Validation, Writing – original draft. Yue Zhang: Data curation, Formal analysis, Investigation, Methodology, Validation, Visualization. Shiyang Lin: Formal analysis, Investigation, Validation. Yan Gao: Conceptualization, Formal analysis, Investigation. Dandan Shen: Supervision, Validation. Zijiu Sun: Resources, Supervision. Fenyong Sun: Resources, Supervision. Zhongqi Cui: Data curation, Formal analysis, Funding acquisition, Investigation, Supervision. Shuo Shi: Conceptualization, Data curation, Formal analysis, Funding acquisition, Methodology, Resources, Supervision, Writing – review & editing.

Declaration of competing interest

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

Appendix A. Supplementary data

The following is the Supplementary data to this article: