Coronavesicle-engineered hydrogel sequential photothermal-immunometabolic therapy reprograms tissue microenvironment for diabetic-infected wounds healing Authors: Bodeng Wu, Shan Lin, Qi Xiu, Mingzhen Zhong, Yiyao Huang, Zhenxun Wang, Xiuhua Wu, Jiangang Mei, Shuhan Liu, Liping Liu, Bo Li, Weilun Pan, Lei Zheng Published: 2025 ABSTRACT AbstractPersistent infection and dysregulated tissue microenvironment constitute the primary obstacles hindering the healing of infectious diabetic wounds. Herein, we developed an NIR-responsive hydrogel dressing incorporating polyphenol-engineered plant-derived nanovesicles (termed Coronavesicles) via an in-situ deposition strategy, designed to implement sequential photothermal-immunometabolic therapy for infectious diabetic wounds. The coronavesicles within the hydrogel preferentially adhere to bacterial surfaces to deliver localized photothermal therapy (PTT), wherein subsequent heat-triggered rupture releases intrinsic bioactive constituents to enhance antibacterial efficacy. Following bacterial clearance, the released components from coronavesicles synergistically modulate fibroblast proliferation, alleviate oxidative stress, and reverse macrophage-mediated inflammatory immunity. This multi-pronged therapeutic strategy demonstrates desirable treatment outcomes both in vitro and in vivo. This work contributes to the understanding and future development of engineering plant-derived nanovesicles for diabetic wound management.Graphical abstract Supplementary InformationThe online version contains supplementary material available at 10.1186/s12951-025-03918-6.Keywords: Diabetic-infected wound, Plant-derived nanovesicles, Sequential therapy, Engineered nanovesicles, Photothermal therapy, Polyphenol, Hydrogel, Antibacterial Introduction Infectious diabetic wounds constitute a significant clinical healthcare challenge, with 40–50% of diabetic patients experiencing delayed healing, amputation, or even mortality [1, 2]. Although clinically used silver ion dressings serve as the primary treatment for infectious diabetic wounds, they exhibit negligible efficacy in promoting tissue repair within the disordered and highly oxidative stressed wound environment [3, 4]. Consequently, the current “clearance and heal” therapeutic strategy emphasizes rapid bacterial eradication coupled with synergistic tissue regeneration [5, 6]. While artificial organic molecules, metal nanomaterials, and microorganisms have been explored for treating infectious diabetic wounds, their application is frequently limited by metabolic challenges and biocompatibility concerns [7]. Plant-derived nanovesicles (PDNVs) have recently gained attention as a novel class of natural nanomedicines with intrinsic biocompatibility and bioactivity [8, 9]. Turmeric-derived nanovesicles (TDNVs), in particular, are enriched with curcuminoid and other antioxidative metabolites that modulate macrophage polarization, attenuate oxidative stress, and promote fibroblast proliferation [10, 11]. However, their intrinsic antibacterial effect activity has not been fully elucidated, particularly in the context of infection-associated chronic wounds. To address this limitation, surface engineering strategies have been introduced to augment vesicle functionality [12]. Among them, polyphenols have attracted considerable attention due to their strong adhesion capability, antioxidant properties, and excellent biocompatibility [13]. Such coatings can effectively enhance vesicle stability, improve photothermal conversion efficiency, and promote interaction with negatively charged bacterial membranes through hydrogen bonding and electrostatic interactions [14, 15]. However, the therapeutic potential and mechanisms of polyphenol-engineered TDNVs in infection-associated diabetic wounds remain largely unexplored. Hydrogel-based therapeutic systems have emerged as promising candidates for diabetic wound management due to their excellent biocompatibility, moisture retention, and capability to provide a conducive microenvironment for tissue regeneration [16]. Hydrogels not only maintain wound hydration and gas exchange but also serve as multifunctional carriers for sustained and localized release of therapeutic agents. Moreover, their tunable network structure enables the incorporation of photothermal agents, allowing NIR-responsive antibacterial therapy that eradicates antibiotic-resistant microorganisms through localized hyperthermia while minimizing damage to surrounding healthy tissues [17]. The integrated systems can simultaneously eliminate bacterial infection and accelerate tissue repair, thereby offering a comprehensive approach for treating infectious diabetic wounds. Herein, we developed polyphenol-coronated turmeric-derived nanovesicles (termed coronavesicles, CVs) via an in-situ PDA deposition strategy and integrated them into a PLGA-PEG-PLGA hydrogel (CV@Gel) for spatiotemporally controlled therapy (Fig. 1A). Upon NIR irradiation, polyphenol-mediated adhesion of coronavesicles to bacterial membranes enables localized photothermal ablation, which simultaneously triggers vesicle disruption and cargo release to synergize antibacterial effects. Following bacterial clearance, the released bioactive components regulate fibroblast proliferation, reduce oxidative stress, and reprogram macrophage-mediated inflammatory immunity, collectively promoting wound regeneration (Fig. 1B). This sequential photothermal-immunometabolic cascade provides a rational and biocompatible strategy for eliminating antibiotic-resistant infections while restoring the dysregulated diabetic wound microenvironment. Preparation and application of CV@Gel hydrogel for diabetic infected wound healing. (A) Construction of CV@Gel. TDNVs were obtained from turmeric root, coated with polyphenols, combined with self-assembled PLGA-PEG-PLGA micelles, and dissolved in a thermal-sensitive hydrogel to form CV@Gel. (B) Mechanism and therapeutic effects of CV@Gel in diabetic infected wound healing Materials and methods Reagents Dopamine hydrochloride (C8H11NO2·HCl) and streptozotocin (STZ) were obtained from Macklin Biochemical Technology Co., LTD., Shanghai, China. The PLGA-PEG-PLGA hydrogel (PLGA1600(75:25)-PEG1500-PLGA1600(75:25), molecular weight 4700; PLGA: PEG: PLGA molar ratio 2:9:2, with deionized water as solvent) was sourced from Tansh Technology Co., LTD., Guangzhou, China. 2′,7′-dichlorofluorescin diacetate (DCFH-DA), 4’,6-diamidino-2-phenylindole (DAPI) and Hoechst 33,342 were bought from Solarbio Technology Co., LTD., Beijing, China. Dulbecco’s modified eagle medium (DMEM), fetal bovine serum (FBS), phosphate buffer solution (1×PBS, pH = 7.4), penicillin − streptomycin, and LIVE/DEAD Bacterial Viability Kit were bought from Thermo Fisher Scientific Co., LTD., Shanghai, China. The RIPA lysis buffer, bicinchoninic acid assay, protein loading buffer, bovine serum albumin (BSA), protease inhibitor cocktail, Coomassie blue, cell count kit-8 (CCK8) assay, cell viability/cytotoxicity assay kit and EdU cell proliferation kit, were purchased from Beyotime Biotechnology Co., Ltd., Guangzhou, China. Isolation and purification of TDNVs This protocol for extracting and purifying TDNVs follows the established methods described in the literature [18]. Briefly, TDNVs were isolated and purified from turmeric using the previously mentioned method. Fresh turmeric from farmers’ markets in Guangxi province was washed with deionized water, minced, soaked in cold PBS, homogenized, and filtered through 212 μm mesh to remove large particles. Sequential centrifugation steps were then performed at different speeds, including 1000 g for 20 min, 3000 g for 20 min, 10,000 g for 30 min, and 70,000 g for 1 h, aiming to separate large particles and fibrous debris. The supernatant was then subjected to ultracentrifugation at 135,000 g for 70 min using a Beckman Optima XP-100 ultracentrifuge (Beckman, USA). The pellet was resuspended in PBS and loaded onto an OptiPrep (iodixanol) density gradient medium with layers of 5%, 10%, 20%, 40%, and 60% (w/v). After centrifugation at 150,000 g for 2 h, the 20%/40% interface layer was carefully collected, washed with PBS, and centrifuged again at 135,000 g for 70 min at 4 °C. The purified TDNVs were stored at − 80 °C for subsequent use. Metabolomics analysis of TDNVs The metabolomic dataset of TDNVs used in this study was obtained from our previously published work [11]. Data were acquired in both positive and negative ESI modes and processed using ProteoWizard and XCMS for feature detection and alignment. Synthesis of coronavesicles Briefly, 5 mg dopamine hydrochloride was dispersed in 5 mL Tris-Hcl buffer (100 mM Nacl) to obtain a 1 mg/ml polydopamine (PDA) solution. The prepared TDNVs (1 mg) were added introduced into the solution and shaken at room temperature for 90 min. Afterward, the CVs was collected through centrifugation at 6000 g for 10 min and resuspended in PBS. Characterization of the coronavesicles The morphology of TDNVs and CVs was analyzed using transmission electron microscopy (TEM, H-7650, HITACHI). The size and zeta potential of the CVs were measured using a Zetasizer Nano ZS (Malvern). Their UV-vis absorption spectra were recorded with a Lambda 25 UV/Visible spectrophotometer (PerkinElmer). Photothermal properties were evaluated by irradiating the solutions with an 808 nm laser and tracking temperature variations using a thermal imager (H16, Hikmicro). Hemolytic test of coronavesicles The hemolytic properties of CVs were assessed using red blood cells (RBCs) as a model. RBCs were separated from whole blood and washed three times with PBS at 3000 g for 3 min to remove plasma and buffy layers. The RBCs were then divided into centrifuge tubes, each containing 100 µL, and treated with 2% Triton-X100 (positive control), PBS (negative control), or CVs at varying concentrations (25, 50, 75, 100, 150, 200, 300, 400, 500 µg/mL). All samples were kept stationary for 1 h before being centrifuged at 3000 g for 5 min. Subsequently, 100 µL of the supernatant was collected into a 96-well plate, and the absorbance at 575 nm was recorded. Synthesis of CV@Gel The prepared CVs solution was subjected to centrifugation at 6000 g for 10 min at 4 °C to collect the precipitate. The resulting CVs nanohybrids were then dispersed into a PLGA-PEG-PLGA hydrogel solution (1 mg/mL) to form CV@Gel. Assessment of Sol-Gel transition characteristics of CV@Gel The gelation temperatures of CV@Gel hydrogel were evaluated using a tube inversion method. Centrifuge tubes (2 mL) containing hydrogels at volumetric concentrations of 15%, 17.5%, and 20% were gradually heated in 0.5 °C intervals starting from 25 °C until gelation was achieved. The transition temperatures were recorded when the hydrogels ceased flowing during the inversion test. Cell culture The L929 mouse fibroblast and RAW264.7 macrophage cell lines were obtained from the American Type Culture Collection (ATCC). These cells were maintained in DMEM supplemented with 10% FBS and 100 U/mL penicillin-streptomycin. All cells were incubated at 37 °C with 5% CO2. In vitro cytotoxicity of CV@GEL L929 mouse fibroblasts were plated in 96-well plates at a density of 5 × 103 cells per well and incubated overnight in DMEM containing 10% FBS and 1% penicillin-streptomycin at 37 °C under 5% CO2. Following the initial incubation, cells were treated with varying concentrations of CV@Gel (0–200 µg/mL, TDNVs-equivalent) and incubated for an additional 48 h. The cells were then rinsed with PBS and supplemented with 90 µL of fresh DMEM medium along with CCK-8 reagent (10 µL) for 2 h without light exposure. Absorbance at 450 nm was recorded using a microplate reader. A similar approach was used to evaluate the photocytotoxicity effects of CV@Gel on L929 cells. In 96-well plates, 3 × 10⁴ L929 cells per well were seeded and incubated with CV@Gel (100 µg/mL, TDNVs-equivalent) or equivalent concentrations of other treatment groups for 6 h. The cells were then rinsed with PBS and provided with 90 µL of fresh DMEM before laser exposure at 808 nm (1 W/cm²) for 5 min. After irradiation, 10 µL of CCK-8 reagent was introduced, and the cells were incubated in darkness for 2 h. Absorbance was subsequently recorded at 450 nm using a microplate reader. For the live/dead fluorescence assay, L929 cells were plated in glass culture dishes and treated with various nanohybrids for 24 h. The cells were washed three times with PBS, exposed to an 808 nm laser at 1 W/cm² for 5 min, and subsequently stained with LIVE/DEAD fluorescence dye following the manufacturer’s protocol. A fluorescence microscope (Nikon) was used to observe the cells and assess the biocompatibility of CV@Gel. Bacterial culture MDRPA, MRSA, and CRE strains were obtained from clinical samples and grown in LB broth with shaking at 200 rpm and 37 °C. Bacterial cells were collected at the logarithmic growth stage through centrifugation at 8000 g for 5 min and washed three times with saline. The concentration of bacteria was measured using a nephelometer (BD Biosciences). CV@Gel uptake by bacteria To evaluate bacterial uptake of vesicles from hydrogel formulations, TDNVs and CVs were labelled with the fluorescent dye Did and incorporated into hydrogel. MDRPA suspensions were co-incubated with each formation (Blank, Gel, TDNVs@Gel, and CV@Gel) at 37 °C for 30 min. After incubation, bacteria were stained with SYTO 9 (V22887, Invitrogen) and analyzed by flow cytometry (LSRFortessa™, BD) to quantify vesicle uptake. The colocalization of Did-labeled vesicles with bacteria was further visualized using CLSM (FV3000, Olympus). In vitro antibacterial evaluation The bacterial suspensions were prepared at high concentration (107 CFU/mL) or low concentration (105 CFU/mL) and resuspended in sterile saline. To ensure homogeneous mixing and to avoid premature gelation, CV@Gel and bacteria were combined in the sol state (liquid form) and incubated for 30 min at room temperature. Control groups included PBS, blank Gel, TDNVs@Gel, and CV@Gel (TDNVs-equivalent) without NIR treatment. The CV@Gel + NIR group was then exposed to an 808 nm laser (1 W/cm², 3 min). A 100 µL aliquot from the bacterial suspension solution was plated on LB plates at 37 °C for 18 h. The number of CFU were counted on each plate to assess bacterial viability and evaluate the antibacterial effects of different treatments. Additionally, bacterial viability was monitored in real time using the LIVE/DEAD Bacterial Viability Kit (L7012, Invitrogen). For time-dependent bacterial growth curves, 100 µL of bacterial suspension was transferred to a 96-well plate and treated with different formulations, followed by incubation for 24 h. Optical density at 600 nm (OD600) was measured at intervals of 4, 8, 12, 16, 20, and 24 h using a microplate reader (N34227, TECAN). Biofilm elimination by CV@Gel The bacterial biofilm was generated by incubating bacteria in LB broth with a 96-well plate at 37 °C for 48 h without agitation. Subsequently, the culture medium was discarded, and the biofilms were rinsed gently with PBS. The established biofilms were then subjected to treatment with PBS, Gel, TDNVs@Gel, CV@Gel-NIR, and CV@Gel + NIR, respectively. For staining, the biofilms were incubated with crystal violet for 15 min, followed by PBS washing and air-drying at room temperature. The bound crystal violet was subsequently dissolved in 30% glacial acetic acid, and the absorbance at 570 nm was recorded using a microplate reader. In addition, the remaining bacterial biofilms were also stained with SYTO 9 (V22887, Invitrogen) before observation of confocal microscopy (Nikon). Bacterial biofilms were formed by incubating bacteria in LB broth within 96-well plates at 37 °C for 48 h without shaking. After incubation, the culture medium was discarded, and the biofilms were rinsed gently with PBS. The biofilms were treated with PBS, Gel, TDNVs@Gel, CV@Gel-NIR, or CV@Gel + NIR (TDNVs-equivalent). For staining, the biofilms were incubated with crystal violet for 15 min, followed by PBS washing and air-drying at room temperature. The retained crystal violet was dissolved in 30% glacial acetic acid, and the absorbance at 570 nm was recorded using a microplate reader. Additionally, residual biofilms were stained with SYTO 9 (V22887, Invitrogen) for confocal microscopy imaging using a Nikon microscope. Intercellular ROS detection The generation of reactive oxygen species (ROS) in L929 cells was quantified using DCFH-DA. The cells were treated with CV@Gel (20 µg/mL, TDNVs-equivalent), TDNVs@Gel (20 µg/mL), Gel, and PBS for 8 h, followed by washing with PBS. Subsequently, DCFH-DA (10 µmol/L) was applied to the cells for 20 min, followed by treatment with H2O2 (1 mM) for 4 h. After a final PBS wash, the cells were subjected to analysis utilizing fluorescence microscopy (Nikon) and flow cytometry (BD Biosciences). Antioxidant ability of CV@Gel Oxidative stress was induced in L929 fibroblast using H2O2. The cells were plated at a density of 2 × 10⁴ cells per well in 96-well plates and incubated overnight to allow attachment. Subsequently, the cells were exposed to CV@Gel (10 µg/mL and 20 µg/mL, based on TDNVs protein content), TDNVs@Gel (20 µg/mL), Gel, and PBS for 8 h. The cells were then rinsed with PBS and subjected to H2O2 (1 mM) in fresh DMEM for an additional 12 h. Cell viability was evaluated through the CCK8 assay. In vitro macrophage polarization The RAW 264.7 murine macrophage cell was plated at a density of 1 × 105 cells per well in 12-well plates and incubated overnight for adherence. The cells were then exposed to CV@Gel (10 µg/mL and 20 µg/mL, based on TDNVs protein content), TDNVs@Gel (20 µg/mL), Gel, and PBS as a control for 24 h. After washing with PBS, the cells were stained with CD206-PE antibodies (BioLegend) and analyzed by flow cytometry to evaluate macrophage polarization. Immunoblotting Protein were lysed in RIPA buffer containing protease and phosphatase inhibitors. Samples were separated on 5–12% SDS-PAGE and transferred onto PVDF membranes. Membranes were blocked with 5% bovine serum albumin and incubated overnight at 4 °C with primary antibodies. Subsequently, membranes underwent three washes with TBS-0.1% tween-20 for 5 min and exposed to horseradish peroxidase-conjugated secondary antibodies (1:5000) for 1 h. Protein detection was carried out using the Fdbio-Dura ECL Kit (Fudebiotech, China) and visualized on photographic films (Biostep, Germany). Primary antibodies used for western blotting included Bcl-2 (Ab182858, Abcam), BAX (Ab32503, Abcam), HO-1 (Ab68477, Abcam), Nrf2 (AF0639, Affinity), p-Nrf2 (Ab76026, Abcam), β-actin (AF7018, Affinity). Goat anti-rabbit IgG (H + L) HRP (S0001, Affinity) served as the secondary antibody. In vivo toxicity assay C57BL/6J mice (6–8 weeks old) were utilized to assess the in vivo toxicity of TDNVs@Gel and CV@Gel. The experimental groups of mice (n = 3) received subcutaneous injections of PBS, Gel, TDNVs@Gel, CV@Gel and CV@Gel + NIR (2 mg/mL, 100 µL), respectively. On day 20, the primary organs and skin tissue were collected and subjected to hematoxylin and eosin (H&E) staining. Infected diabetic wound healing evaluation The animal procedures were authorized by the Laboratory Animal Management Committee of Southern Medical University (Approval No. SMUL2022179). C57BL/6J male mice (6–8 weeks old) were obtained from HFK Bioscience Co., LTD (Beijing). After fasting for 12 h, the mice were given an intraperitoneal injection of streptozotocin (150 mg/kg in a citric acid-sodium citrate buffer, pH 4.2–4.5) to induce diabetes. Mice with non-fasting glucose levels above 16.7 mmol/L one-week post-injection were classified as diabetic. These mice were anesthetized with 2% pentobarbital sodium (50 mg/kg) via intraperitoneal injection, and full-thickness wounds were made on their dorsal surfaces using sterile biopsy punches. Infection was induced by injecting 50 µL of MDRPA (10⁷ CFU/mL) into the wound, creating an infected diabetic wound model. The mice were divided into five groups (n = 4 per group) and treated with PBS, Gel, TDNVs@Gel, CV@Gel-NIR, or CV@Gel + NIR. A single 100 µL dose of hydrogel (2 mg/mL, equivalent to 200 µg TDNVs protein per mouse) was applied per wound, with no further interventions. Temperature variations in NIR-irradiated groups were recorded using a digital NIR photothermal imaging system (H16, Hikmicro). Wound areas were photographed on days 1, 4, 8, 12, and 16. On day 16, the mice were euthanized, and wound tissues were collected for H&E and Masson’s trichrome staining. Macrophage polarization in the tissue samples was evaluated using CD86 markers. Histology analysis Tissue samples were fixed in 10% formalin and embedded in paraffin. Sections were cut to a thickness of 5 μm, followed by deparaffinization and rehydration. The samples were stained with H&E and Masson’s trichrome, and the stained sections were observed under a microscope. For immunofluorescence analysis, tissues were embedded in an OCT compound (Sakura Finetek) and cryosectioned at a thickness of 5 μm at −20 °C. Blocking was performed for 1 h at −20 °C using 5% BSA in PBS. The slides were incubated with primary antibodies overnight at 4 °C. After washing, the sections were incubated with DAPI and fluorochrome-conjugated secondary antibodies for 60 min. Imaging was performed with a confocal microscope and analyzed using Image J software (National Institutes of Health). ELISA To evaluate the effect of CV@Gel on macrophage-mediated inflammation, RAW 264.7 cells were exposed to LPS (200 ng/mL) for 12 h to induce M1 polarization. Subsequently, the cells were treated with CV@Gel (20 µg/mL), TDNVs@Gel (20 µg/mL), Gel, and PBS for another 12 h. The cytokines (TNF-α and IL-1β) in the cell supernatant were further quantified using an ELISA assay. Fresh skin tissues were collected for protein extraction. Mouse TNF-α (GEM0003) and IL-1β (GEM0002) ELISA kits were obtained from Servicebio Biosciences. The ELISA assays were conducted following the manufacturer’s instructions. Transcriptomic evaluation of CV@Gel for diabetic wound healing Wound tissue samples were collected on day 8 for RNA sequencing experiments. Total RNA was extracted from wound tissue using TRIzol® Reagent (Magen), and quality was assessed with a Nanodrop ND-2000 spectrophotometer (Thermo Scientific, USA) and Agilent Bioanalyzer 4150 (Agilent Technologies, CA, USA). Library preparation was carried out using the ABclonal mRNA-seq Lib Prep Kit (ABclonal, China) in accordance with the manufacturer’s instructions, followed by sequencing on an Illumina Novaseq 6000/MGISEQ-T7 platform. For differential expression analysis, the DESeq2 R package was used with |log2FC| >1 and Padj < 0.05 as criteria. GO and KEGG pathway enrichment analyses were performed through the clusterProfiler R package, with significance set at P < 0.05. Statistical analysis Data are expressed as mean ± SD. Details on data preprocessing and the sample sizes (n) is provided in the corresponding figure legends. For normally distributed data, a normal distribution, independent two-sample t-tests were applied for two-group comparisons, while one-way ANOVA with Tukey’s post-hoc test was used for multiple groups. Non-normally distributed data were analyzed using the Mann-Whitney U test for two groups or the Wilcoxon rank-sum test for multiple groups (Kruskal–Wallis test for group comparisons). All statistical analyses were performed with GraphPad Prism version 9.0 and Origin2019b. Statistical significance was considered at P-value < 0.05. ‘ns’ denotes not significant; significance levels are indicated as * p < 0.05, ** p < 0.01, *** p < 0.001, and **** p < 0.0001. Results Preparation and characterization of coronavesicles The fabrication of coronavesicles is described in Fig. 1, the photothermally responsive polymeric nanovesicles were prepared by taking advantage of the surface chemistry [19]. Briefly, the TDNVs were naturally sourced nanovesicles obtained from turmeric through differential density centrifugation. The TDNVs displayed a uniform teacup-like morphology with a size ranging from 100 to 200 nm visualized by transmission electron microscopy (TEM) (Fig. 2A). Dopamine was utilized to coat TDNVs through incubation and agitation, resulting in the formation of CVs possibly facilitated by charge interactions or covalent attachment. The morphology of CVs exhibited a “coronavesicle”-like structure in comparison to TDNVs, indicating the successful incorporation of PDA onto the membrane of TDNVs (Fig. 2B). Dynamic light scattering measurements showed that introducing dopamine increased the size of particles from 194.87 ± 10.84 nm for bare TDNVs to 240.63 ± 3.95 nm for CVs (Fig. 2C), with the corresponding polydispersity index (PDI) values of 0.124 ± 0.022 and 0.196 ± 0.016, respectively (Fig. S1). The zeta potentials of TDNVs and CVs were − 29.28 ± 2.53 mV and − 32.5 ± 1.85 mV, respectively (Fig. 2D). Characterization of the NIR-responsive nanohybrid CVs and CV@Gel. (A-B) TEM images of TDNVs (A) and CVs (B). (C) Size distribution analysis of TDNVs and CVs. (D) Zeta potential analysis of TDNVs and CVs (n = 5). (E) UV-vis spectra of TDNVs and CVs. (F) UV-vis spectra of CVs at different concentrations. (G) Temperature variations of CVs solution under laser irradiation at different power levels. (H) Temperature variations of CVs with varying concentrations under laser irradiation. (I) Temperature variation over time in the cyclic heating-cooling experiment. (J) Photograph of Gel, TDNVs@Gel, and CV@Gel hydrogel loaded in separate medical bottles. (K) Sol–gel behavior of CV@Gel, upright at RT, inverted at RT, and under NIR irradiation. (L) Temperature changes of CV@Gel, CVs, TDNVs@Gel, Gel, and PBS solution. (M) Thermal infrared images of CV@Gel, CVs, TDNVs@Gel, Gel, and PBS solution under laser irradiation. Data are expressed as mean ± SD Due to the highly controllable nature of PDA surface modification, we optimized the PDA coating by adjusting dopamine concentration, reaction time, and temperature to achieve improved CVs stability, photothermal effect, and biocompatibility. Ultraviolet-visible (UV-Vis) spectrum showed a distinct absorbance of CVs relative to TDNVs at an NIR range from 750 to 900 nm (Fig. 2E). In addition, FT-IR spectra revealed both PDA and TDNVs characteristic peaks in CVs, indicating the effective formation of CVs (Fig.S2) [20]. As concentration increased from 0.35 to 1.25 mg/ml, CVs exhibited an increased absorbance peak at 808 nm (Fig. 2F). Since PDA readily absorbs NIR and has notable photothermal conversion properties, CVs modified with PDA was endowed with considerable photothermal properties. CVs exhibited photothermal conversion dependent on both laser density and concentration (Fig. 2G and H), indicating its adjustable photothermal response for various applications. The photothermal conversion efficiency of CVs was measured at 17.64% (Fig. S3). Meanwhile, Fig. 2I demonstrated the high photostability of CVs, as evidenced by its consistent temperature increase during 5 cycles of NIR laser irradiation. CVs was rapidly heated to approximately 58 °C under NIR irradiation and maintained this temperature consistently through all 5 cycles, demonstrating its efficient photothermal conversion and precise controllability under NIR exposure. In addition, TEM analysis revealed partial morphological alterations of CVs after repeated irradiation, where slight rupture of the core TDNVs was observed (Fig. S4). However, the PDA coating remained tightly associated with the vesicular membrane, thereby preserving the overall photothermal stability of the CVs. The biosafety plays a critical role in biomedical applications. The hemolytic assay revealed that even at a concentration of CVs (500 µg/mL), the hemolysis rate of red blood cells remained below the general requirement threshold of 5% [21] (Fig. S5). These results indicate the successful synthesis of CVs with NIR absorption, suggesting its potential for safe use in biomedical applications. Synthesis and characterization of CV@Gel This study investigated the potential of CVs to address multiple challenges, including antibiotic-resistant bacterial infections, oxidative stress, and chronic inflammation in diabetic wound management [22]. A biodegradable and biocompatible temperature-responsive hydrogel, PLGA-PEG-PLGA (Gel), was selected as the matrix to encapsulate CVs, forming CV@Gel [23]. Based on the solution − gelation (sol − gel) phase transition diagram (Fig. S6), CV@Gel (17.5 wt%) demonstrated a gelation temperature of approximately 36.5°C and thus was selected for the subsequent experiments. The stability of CV@Gel was assessed through frequency sweep analysis. Both the storage modulus (G’) and loss modulus (G’’) increased as angular frequency rose (Fig. S7A). Furthermore, viscosity measurements were evaluated by subjecting it to varying shear rates varying between 0.01 and 100 s^(−1) (Fig. S7B). The results revealed that the Gel exhibited decreasing viscosity with an increasing shear rate, indicating shear thinning behavior. The UV − vis spectra of Gel, CVs, and CV@Gel were analyzed (Fig. S8), revealing distinct absorption peaks in the NIR region for both CVs and CV@Gel. Figure 2J displayed Gel, TDNVs@Gel, and CV@Gel separately loaded in a 5 mL medical bottle. Upon NIR irradiation, when the temperature increased (>36.5 °C), the CV@Gel solution transformed into a gel state, whereas at room temperature, CV@Gel remained in a sol state (Fig. 2K). Upon 808 nm laser irradiation, the CV@Gel solution heated from 28.0 °C to 58.1 °C, whereas the CVs solution reached 45.2 °C (Fig. 2L-M). However, the temperature changes of Ctrl, Gel, and TDNVs@Gel groups were negligible. Interestingly, the temperature rose more rapidly in CV@Gel than in pure CVs after surpassing the LCST (36.5 °C), likely due to changes in the hydrophilic/hydrophobic properties of CVs and the Gel. Therefore, the hydrogel matrix may contribute to antibacterial efficacy by enhancing local photothermal retention and maintaining an elevated temperature at the bacterial-hydrogel interface. Furthermore, the biocompatibility analysis revealed that CV@Gel (0–200 µg/mL) showed no significant cytotoxicity toward L929 fibroblast cells (Fig. S9 and S10). Taken together, these findings demonstrate the successful preparation of a thermal-sensitive CV@Gel hydrogel with NIR reactivity, synthesized from natural plant-derived nanovesicles. Uptake of CV@Gel by bacterial Given the adhesive properties of polyphenols, which facilitate bacterial interaction and uptake, we next examined the internalization of Did-labeled vesicles from different formulations by MDRPA [24, 25]. As shown in Fig. 3A, flow cytometry analysis revealed markedly higher fluorescence intensity in the CV@Gel-treated group compared with Blank, Gel, and TDNVs@Gel groups, indicating more efficient vesicles association with bacterial. Quantitative analysis further confirmed that the uptake of Did-labeled vesicles by MDRPA was significantly enhanced in the CV@Gel group (Fig. 3B), suggesting that PDA modification promotes bacterial adhesion and internalization of TDNVs. Moreover, confocal laser scanning microscopy (CLSM) was utilized for direct visualization of CVs localization. CVs was detected surrounding and within the bacterial cells, suggesting its entry into the cytoplasm through the membrane (Fig. 3C-D). Next, we compared the bactericidal ability of CV@Gel, CVs and PDA (equal PDA content). Under the same excitation power at 808 nm wavelength, CV@Gel and CVs induced greater bacterial death than pure PDA (Fig. 3E-F), suggesting that CVs may facilitate PDA entry into bacterial cells (Fig. 3G). These results confirm the rapid internalization of CVs from CV@Gel by MDRPA, ensuring targeted accumulation and strong interaction with bacteria, ultimately boosting photothermal and antibacterial efficacy. In vitro uptake and interaction of CV@Gel with MDRPA. (A) Flow cytometry analysis of MDRPA after 30 min incubation with different treatment, including PBS (Blank), Gel, TDNVs@Gel, and CV@Gel. (B) Statistical analysis of Did-labeled vesicles uptake by MDRPA in each group (n = 3). (D) Quantification of Did fluorescence intensity in MDRPA treated with CV@Gel and Ctrl (n = 3). (E) Colony images on Luria-Bertani broth (LB) agar plates of MDRPA subjected to various treatments under equal PDA content. (F) Statistical analysis of bacterial viability based on the colony number of MDRPA after various treatments (n = 3). (G) Schematic illustration of the uptake and antibacterial effect of CV@Gel. Data are expressed as mean ± SD. **p < 0.01, ***p < 0.001, and ****p < 0.0001 Antibacterial activity of CV@Gel Drug-resistant bacterial infection is a major hindrance to wound healing in diabetes mellitus [26]. This study investigated the antibacterial ability of CV@Gel against three clinically relevant drug-resistant bacteria, including Gram-negative MDRPA and Carbapenem-Resistant Escherichia coli (CRE), as well as Gram-positive Methicillin-Resistant Staphylococcus aureus (MRSA). According to Fig. 3E and our previous research [17], NIR light alone could not cause damage to bacteria, and comparisons with PDA, CVs, and CV@Gel under identical NIR irradiation demonstrated that the potent antibacterial effect is primarily driven by CVs and CV@Gel rather than PDA or NIR alone. Thus, Ctrl, Gel, TDNVs@Gel, CV@Gel-NIR, and CV@Gel + NIR groups were set to evaluate bactericidal ability. CV@Gel + NIR exhibited demonstrated effective elimination of MDRPA (Fig. 4A-B), CRE (Fig. S11A-B), and MRSA (Fig. S12A-B), with no colonies observed on the culture plates. Real-time dual fluorescent live/dead staining further validated the bactericidal effect of CV@Gel at high concentrations (107 CFU/mL). Notably, the TDNVs@Gel and CV@Gel-NIR groups reduced MDRPA levels by approximately 20% and 40%, respectively, at a low bacterial concentration (105 CFU/mL) (Fig. S13). This partial antibacterial effect may be attributed to bioactive metabolites contained in TDNVs, which may inhibit bacterial growth. However, at higher bacterial concentrations, no significant differences were observed among the Ctrl, TDNVs@Gel, and CV@Gel-NIR groups, likely due to the inherent antibacterial properties of the endogenous components of TDNVs. Antibacterial activities of CV@Gel in vitro. (A) Colony images on LB agar plates and CLSM images of MDRPA stained with live/dead dyes after various treatments. (B) Statistical analysis of bacterial viability at a high bacterial concentration (107 CFU/mL) based on MDRPA colony counts after various treatments (n = 5). (C) Growth inhibition assay of MDRPA after treatments of different formulations in 24 h. (D) SEM images of bacteria before and after the treatment with CV@Gel + NIR. (E) Images of residual MDRPA biofilm subjected to different treatments and stained with crystal violet. (F) Statistical analysis of residual stained biofilms calculated via OD570 nm (n = 5). (G) 3D scanning of MDRPA biofilms after different treatments. Data are expressed as mean ± SD. *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001 The growth inhibition assay showed complete suppression of bacterial growth by CV@Gel + NIR, with noticeable recurrence observed at 12 and 24 h in other treatment groups (Fig. 4C, Fig. S11C, and Fig. S12C). In addition, a scanning electron microscope (SEM) was utilized to analyze the bacterial morphology. The morphology and structure of bacterial cells in the CV@Gel + NIR group showed cracks and shrinkage, whereas those in the Ctrl group remained intact with a smooth surface (Fig. 4D). Biofilms are microbial communities that resist antibiotic treatment and play a key role in chronic wound infections [27]. Therefore, the influence of CV@Gel hydrogel on biofilm was assessed using crystal violet staining (OD values at 570 nm) and fluorescence imaging. CV@Gel + NIR significantly disrupted MDRPA biofilms. In contrast, TDNVs@Gel and CV@Gel-NIR treatments showed a slight biofilm reduction compared to Ctrl (Fig. 4E-G). Furthermore, the CV@Gel + NIR group exhibited comparable biofilm eradication effects when applied to treat CRE and MRSA (Fig. S11D and Fig. S12D). The above results showed that TDNVs can inhibit bacterial growth and delay bacterial biofilm formation at low bacterial concentrations, while CV@Gel + NIR highlights the superiority of synergistic sterilization due to its photothermal effect and endogenous sterilization effect. We further incubated CV@Gel with cells and removed the supernatant after incubation. The results showed that CV@Gel exhibited low photocytotoxicity to L929 fibroblast cells (Fig. S14), possibly due to its preferential uptake by bacteria within the same time frame. CV@Gel has demonstrated significant antibacterial ability against common clinical drug-resistant bacteria, indicating that CV@Gel might be a broad-spectrum biomaterial for wound infections. CV@Gel enhances antioxidant defense in fibroblasts and inhibits macrophage-mediated inflammation Excessive accumulation of reactive oxygen species (ROS) represents a significant impediment to the healing of diabetic wounds [28], as reducing the superoxide state in fibroblasts is beneficial for attenuating apoptosis and accelerating wound repair [29]. The in vitro release profile demonstrated that CV@Gel achieved almost complete CVs release within 1 h under cell culture conditions, ensuring adequate exposure during subsequent cellular experiments (Fig. S15). Therefore, we further investigated the antioxidative effects of CV@Gel on murine fibroblast L929 cells. Initially, L929 cells were pre-treated with CV@Gel, TDNVs@Gel, Gel, and PBS for 8 h, followed by stimulation with H2O2 for 4 h to simulate oxidative stress, with an untreated group as the blank control. Intracellular ROS levels were assessed via the DCFH-DA probe, which exhibits green fluorescence upon observation. Significant green fluorescence was observed in the PBS and Gel groups, whereas fluorescence intensity markedly decreased in the TDNVs@Gel and CV@Gel groups. (Fig. 5A and S16). Flow cytometry results further demonstrated that CV@Gel treatment reduced the proportion of cells with excessive ROS from 49.8% in the PBS group to 21.2%, while the TDNVs@Gel treatment resulted in a reduction to 34.2% (Fig. 5B). Correspondingly, the viability of L929 cells was preserved by both CV@Gel and TDNVs@Gel treatments, even at 1 mM H2O2 exposure (Fig. 5C). The CV@Gel group exhibited superior antioxidative effects compared to the TDNVs@Gel group, likely due to the PDA loaded in the CV@Gel group, which possesses the capability to scavenge exogenous ROS. These results indicate that CV@Gel can participate in the clearance of ROS in fibroblasts in vitro, while Gel alone exhibited limited antioxidative capacity in comparison to CV@Gel and TDNVs@Gel. Evaluation of CV@Gel in antioxidant defense of fibroblasts and modulation of macrophage-mediated inflammation. (A) Representative images of ROS levels in fibroblast detected using DCFH-DA probe after various treatments. (B) ROS levels in L929 fibroblast cells measured following different treatments. (C) Cell viability of L929 fibroblasts assessed after various treatments (n = 5). (D) Apoptosis markers in L929 fibroblasts after various treatments under oxidative stress. (E) Protein expression levels of BAX/Bcl-2, normalized to control values (n = 3). (F) Antioxidant markers in L929 fibroblasts after various treatments. (G-H) Protein expression levels of p-Nrf2/Nrf2 (G) and HO-1 (H), normalized to control values (n = 3). (I) Flow cytometry analysis of M0-type macrophage polarization after various treatments. (J-K) ELISA analysis of relative TNF-α and IL-1β levels in cell supernatants after various treatments (n = 3). (L) Representative metabolite structures from each subgroup. (M) Potential target gene network of TDNVs and DFI. (N) Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses of the target genes. Data are expressed as mean ± SD. *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001 To identify the mechanism associated with the protective effects of CV@Gel and TDNVs@Gel, the expression of apoptosis-related and antioxidant-related proteins was evaluated. The immunoblotting results showed that TDNVs@Gel and CV@Gel treatments both reduced the BAX/Bcl-2 ratio, a key indicator of cell apoptosis, during oxidative stress-induced apoptosis (Fig. 5D-E). We further explored the impact of CV@Gel on the Nrf2/HO-1 antioxidant pathway in fibroblasts. After 8-hour incubation with different treatment groups, western blot analysis revealed that both TDNVs@Gel and CV@Gel treatments significantly upregulated the levels of p-Nrf2/Nrf2 and HO-1 proteins, known regulators of the antioxidant response (Fig. 5F). Furthermore, CV@Gel enhanced the oxidative stress capacity of fibroblasts in a concentration-dependent manner, with 20 µg/mL showing a more remarkable ability to improve oxidative stress compared to 10 µg/mL (Fig. 5G-H). In contrast, treatment with Gel alone did not alter the characteristic changes in the antioxidant pathway. Our findings indicate that CV@Gel protects L929 cells from ROS-induced damage by enhancing the endogenous antioxidant Nrf2/HO-1 pathway and eliminating exogenous ROS accumulation. Additionally, PDA modification does not affect the ability of TDNVs to enhance endogenous antioxidant capacity. Pathogenic infections that polarize inflammatory macrophages to continually release cytokines cause excessive inflammation, a major impediment to healing diabetic infectious wounds [30]. Consequently, we explored the effects of CV@Gel on macrophage polarization. After 24 h of co-incubation with RAW 264.7 macrophages, CV@Gel increased M2-type macrophages (CD206 marker) from 0.56% in the Ctrl group to 73.2%. In comparison, TDNVs@Gel achieved 57.2%, while the Gel group (0.78%) showed no polarization change (Fig. 5I). This suggests that TDNVs can polarize M0-type macrophages to the M2-type macrophages. CV@Gel achieved more effective M2-type macrophage polarization than TDNVs@Gel, likely due to the synergistic effects of PDA and TDNVs, as previous studies have reported that PDA treatment can induce differentiation into M2-type macrophages [31]. ELISA analysis revealed that CV@Gel significantly inhibited the secretion of pro-inflammatory cytokines TNF-α and IL-1β in LPS-induced macrophages following different treatments (Fig. 5J-K). These results indicate that CV@Gel has could serve as a promising candidate to modulate the diabetic wound microenvironment in vivo, thereby promoting effective wound healing. To further elucidate the material basis of TDNVs and their synergistic interaction with PDA in CV@Gel, thereby facilitating the antioxidant and macrophage-modulating functions of TDNVs, we performed an analysis of their untargeted metabolomics data. Metabolomic analysis of TDNVs revealed that the top three metabolite subclasses were lipids and lipid-like molecules (29.8%), organic nitrogen compounds (15.4%), and organic acids and derivatives (14.1%) (Fig. S17A). Furthermore, a bioactivity annotation of the top 20 most abundant metabolites (Table S1) indicated potential roles in antioxidation, anti-inflammation, immune modulation, and nutritional support. The most abundant representative metabolites from each subclass were identified as ar-turmerone, choline, citrate, curcumin, and adenosine (Fig. S17B), and their molecular structures are shown in Fig. 5L. A potential drug–target–disease interaction network constructed for these five metabolites (Fig. 5M) revealed enriched targets predominantly involved in antioxidant defense and apoptosis-related pathways. Additionally, KEGG pathway enrichment analysis of the associated genes revealed a close relationship with lipid metabolism and diabetic complications (Fig. 5N), while GO enrichment analysis indicated involvement in biological processes such as bacterial response, oxidative stress, and nutrient levels (Fig. S18). Collectively, these results reveal that the identified metabolites constitute the material basis for TDNVs’ antioxidant and immunomodulatory functions, and further support that their synergistic interplay with PDA in CV@Gel contributes to the combinatorial therapeutic efficacy observed in diabetic wound healing. In vivo diabetic infected wound healing by CV@Gel In vivo antibacterial properties and wound healing effects of CV@Gel were assessed in a diabetic full-thickness wound model infected with MDRPA (Fig. 6A). Before the experiment, biocompatibility assessments showed no significant damage to major organs or skin tissue in vivo (Fig. S19 and S20), and the MDRPA infected wound model was successfully established, as evidenced by clear bacterial colonization and elevated inflammatory cytokines (Fig. S21 and S22). Mice were divided into five groups and treated on day 1 post-wounding with PBS, Gel, TDNV@Gel, CV@Gel-NIR, or CV@Gel + NIR. No noticeable variations in body weight were recorded among the groups throughout the treatment period, indicating the safety of treatments (Fig. S23). In addition, CV@Gel exhibited good tissue adhesion and flexibility on both porcine and mouse skin under NIR irradiation, indicating its suitability for wound repair applications (Fig. S24). The wound temperature in mice treated with CV@Gel increased from 29.1 °C to 52.1 °C after 3 min of irradiation under an 808 nm laser, whereas the temperature changes in other groups were minimal (Fig. 6B-C). Consistent with the in vitro results, bacterial colonies collected from the wounds were markedly reduced after CV@Gel treatment, especially with NIR irradiation (Fig. S25). The CV@Gel + NIR group showed minimal bacterial biofilm at the wound site on day 4, compared to the Ctrl group, which exhibited substantial biofilm and exudate (Fig. 6D). By day 12, biofilm remnants were still observed in the Ctrl and Gel groups, while none were detected in the TDNVs@Gel and CV@Gel-NIR groups, aligning with the in vitro antibiofilm formation assays (Fig. 5F). The formation of bacterial biofilms is a staged process, beginning with the initial adhesion of bacteria to solid surfaces, followed by microcolony formation and secretion of an extracellular matrix. This allows bacteria to thrive within a favorable microenvironment and resist attacks from antibiotics and the host immune system [32]. Current literature indicates that nanoparticles can effectively inhibit the development and formation of biofilms by disrupting bacterial cell membranes, interfering with intercellular communication, and blocking essential biochemical signals [33]. TDNVs@Gel or CV@Gel may impede the initial stages of biofilm adhesion or disrupt the structure of biofilms, thereby reducing their protective capabilities. In vivo antibacterial properties and healing effects of CV@Gel on diabetic infected wounds. (A) Flowchart of the in vivo intervention experiment for diabetic infected wounds. (B) Representative thermal infrared images of mice from different groups. (C) Quantification of the temperature increase in various groups. (D) Representative images of wound healing on days 1, 4, 8, 12, and 16 after treatments. (E) Statistical analysis of wound contraction efficiency over time for different treatments (n = 4). (F) Schematic illustrations of diabetic infected wound contraction in mice during 16 days of various treatments. (G) H&E and Masson’s trichrome staining images of wound tissues collected on day 16. (H) Representative CD86 immunofluorescence staining images in wound tissue with varied treatments. (I) Quantified fluorescence intensity of CD86 in wound tissue (n = 3). (J-K) Statistical analysis of TNF-α (J) and IL-1β (K) levels in skin tissues by ELISA (n = 4). Data are presented as mean ± SD. *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001 During the healing process, CV@Gel + NIR exhibited the highest rate of wound contraction. CV@Gel-NIR and TDNVs@Gel also demonstrated faster healing rates compared to the Ctrl and Gel groups, likely due to the antibiofilm and antioxidant properties of TDNVs or CVs (Fig. 6E-F). H&E and Masson’s trichrome staining further confirmed the superior wound-healing effects of CV@Gel + NIR. Histological analysis revealed that the CV@Gel + NIR group exhibited near-complete re-epithelization and well-organized collagen deposition by day 16, suggesting accelerated tissue remodeling and wound closure compared to the Ctrl and Gel groups, which displayed incomplete epithelial regeneration and disorganized collagen fibers (Fig. 6G). In contrast, the CV@Gel-NIR and TDNVs@Gel groups also showed improved healing, with moderate collagen deposition and a more defined epidermal layer. Immunofluorescent staining for CD86, a marker of pro-inflammatory M1 macrophages, showed significantly reduced CD86 expression in the CV@Gel + NIR and CV@Gel-NIR groups compared to the Gel and Ctrl groups (Fig. 6H-I). These suggest that CV@Gel, particularly with NIR irradiation, may have contributed to a shift in the macrophage phenotype toward a less inflammatory state, likely facilitated by the earlier clearance of bacterial infection. Furthermore, CV@Gel demonstrated its in vivo ability to suppress macrophage-associated inflammatory responses by reducing pro-inflammatory cytokines TNF-α and IL-1β (Fig. 6J-K). These results demonstrated the effectiveness of CV@Gel in promoting wound re-epithelialization and collagen deposition while reducing inflammation to accelerate overall healing, especially with NIR irradiation. Transcriptomic evaluation of antibacterial mechanism of CV@Gel hydrogel To investigate the global host response to anti-infective therapy, RNA-seq was utilized to analyze transcriptomes of MDRPA-infected diabetic wound tissues on day 8 post-treatment, comparing the CV@Gel + NIR group with the Ctrl group. Principal Component Analysis (PCA) plot demonstrated distinct separation between the Ctrl and CV@Gel + NIR groups, indicating clear differentiation between these two groups (Fig. 7A). A total of 674 differentially expressed genes (DEGs) were identified for CV@Gel + NIR group compared to Ctrl, with 460 down-regulated genes and 214 up-regulated genes (Fig. 7B). Gene Ontology (GO) analysis of down-regulated DEGs (Fig. 7C) revealed significant enrichment in pathways associated with inflammatory response, defense response, and response to other organisms, indicating that CV@Gel + NIR treatment likely led to effective bacterial clearance along with reduced inflammation and immune response compared to the Ctrl group. Conversely, up-regulated DEGs (Fig. 7D) were enriched in pathways related to tissue development, muscle adaptation, and collagen network formation, suggesting that CV@Gel + NIR treatment accelerated processes of tissue repair and structural organization in infected wounds. For KEGG pathway enrichment analysis, down-regulated DEGs (Fig. 7E) were significantly enriched in pathways associated with inflammatory and immune responses, including cytokine-cytokine receptor interaction and the TNF signaling pathway, while up-regulated DEGs (Fig. 7F) were primarily enriched in pathways involved in tissue repair and structural organization, notably in ECM-receptor interaction. To further elucidate the effects of CV@Gel + NIR treatment on gene expression, a heatmap was generated to display the expression patterns of selected DEGs associated with bacterial response, inflammatory response, and tissue development (Fig. 7G). Notably, genes involved in response to bacterium and inflammatory response, such as Trem1, TLR4, TNF, and IL-1β, were markedly down-regulated in the CV@Gel + NIR group compared to Ctrl, while anti-inflammatory related genes like Il1f10 and Orm1 were significantly up-regulated. In contrast, genes associated with tissue repair and development, including Gata3, Bmp4, and Wnt10b, exhibited increased expression in the CV@Gel + NIR group. Overall, the down-regulated pathways were primarily involved in resolving excessive inflammation, while the up-regulated pathways supported tissue repair and regeneration, illustrating a coordinated response of pathogen clearance and tissue restoration. These results emphasized the dual role of CV@Gel + NIR treatment in reducing inflammation and promoting wound healing in MDRPA-infected diabetic wounds, underscoring its potential as an effective strategy for treating infected chronic wounds. Transcriptomic analysis of MDRPA-infected wounds on day 8 post-therapy (n = 3). (A) PCA plot of clustered datasets between the CV@Gel + NIR and Ctrl groups. (B) Volcano plot of DEGs between CV@Gel + NIR and Ctrl groups. (C-D) Go Term enrichment analysis for (C) down-DEGs and up-DEGs (D). (E-F) KEGG pathway enrichment analysis for down-DEGs (E) and up-DEGs (F). (G) Heatmap of selected DEGs related to bacterial response, inflammatory response, and tissue development pathways across CV@Gel + NIR and Ctrl groups Discussion CV@Gel addresses critical limitations in diabetic wound care through a spatiotemporally regulated “clearance and heal” strategy. Unlike conventional silver ion dressings that lack tissue-repair functionality in oxidative microenvironments, CV@Gel synergistically integrates bacterial eradication with immunometabolic reprogramming through its polyphenol-coronated architecture. This design improves upon the antibacterial limitations of artificial nanomaterials and unmodified PDNVs while mitigating biocompatibility concerns associated with metal-based agents. The sequential therapeutic cascade begins with polyphenol-mediated adhesion of coronavesicles to bacterial membranes. The catechol and amine groups of PDA may form multiple hydrogen bonds and electrostatic interactions with negatively charged bacterial membranes, which could enhance vesicle anchoring, whereas mammalian membranes composed mainly of zwitterionic phospholipids likely exhibit weaker affinity toward polyphenols [15, 34]. This enables targeted photothermal ablation under NIR, where localized heating simultaneously disrupts vesicles to release intrinsic bioactive constituents. While previous studies highlighted the intrinsic antimicrobial activity of PDNVs, our findings reveal that TDNVs@Gel exhibits effective bacterial inhibition at low pathogen burdens, with diminished efficacy under high bacterial loads-likely due to limited endogenous antimicrobial molecules [35]. In our system, the synergistic interaction among PDA modification, TDNVs-derived bioactive metabolites, and the hydrogel framework markedly enhances antibacterial efficacy, effectively compensating for the limited intrinsic activity of unmodified TDNVs. Importantly, this targeted strategy minimizes off-target effects compared to systemic antimicrobials or non-degradable metal nanoparticles. Following bacterial burden reduction, CV@Gel reprograms the wound microenvironment by activating fibroblasts through Nrf2/HO-1 signaling and inducing macrophage polarization toward an M2-like anti-inflammatory phenotype. This spatiotemporally coordinated strategy not only ensures effective infection control but also fosters regenerative microenvironment remodeling. Importantly, CV@Gel retains the functional bioactivity of TDNVs after PDA modification, maintaining its capacity to modulate fibroblasts and macrophages-two key cell types in wound healing [36]. Beyond influencing individual cell behavior, CV@Gel facilitates pro-healing crosstalk between fibroblasts and macrophages, enabling coordinated antioxidant defense, matrix remodeling, and resolution of inflammation. This synergistic cellular interaction likely underpins the transcriptional changes observed in vivo, where CV@Gel treatment significantly suppressed genes related to bacterial response and inflammatory activation, while upregulating pathways involved in extracellular matrix organization and collagen fiber synthesis. Although no significant differences in oxidative stress levels were detected among treatment groups, this may reflect the advanced healing status in CV@Gel + NIR-treated wounds, combined with the inherent ROS-scavenging capacity of PDA [37]. CV@Gel offers engineering advantages for the clinical management of diabetic-infected wounds, addressing common limitations of conventional hydrogels, including poor efficacy against antibiotic-resistant bacteria and insufficient regulation of multicellular dynamics [17]. Furthermore, its plant-derived vesicle components provided a more accessible and scalable source than mammalian cell derived EVs, enhancing practicality for clinical applications. Limitations of CV@Gel include its reliance on NIR activation, requiring validation in larger models or clinical samples, and the need for optimal dose control. Future research should focus on evaluating the effectiveness and appropriate dosing of CV@Gel in larger models and clinical settings, as well as developing NIR activation methods suitable for clinical application. Conclusion Our study developed a biomimetic hydrogel system, CV@Gel, that integrates spatiotemporally coordinated antibacterial and regenerative functions for the treatment of diabetic-infected wounds. By leveraging a plant-derived nanovesicle-based interface engineering strategy, CV@Gel combines the photoresponsive capabilities of polydopamine with the preserved bioactivity of TDNVs. This strategy enables CV@Gel to effectively eliminate multidrug-resistant pathogens—including MDRPA, MRSA, and CRE—through targeted photothermal therapy, while subsequently reprogramming the wound microenvironment to promote fibroblast activation and macrophage polarization. The system reduces inflammation, modulates oxidative stress, and fosters a regenerative milieu characterized by enhanced re-epithelialization and collagen deposition, demonstrating its potential as a therapeutic approach in regenerative medicine for diabetic-infected wound care. Supplementary Information Supplementary Material 1: Supplementary data to this article can be found in the appendix Acknowledgements We would like to thank the team of Dr. Mengchen Zou from the Department of Endocrinology at Nanfang Hospital, Southern Medical University, for their valuable advice on the treatment and care of diabetic foot infections. Some figures were created by BioRender.com. Abbreviations Coronavesicles Coronavesicles-engineered hydrogel Turmeric-derived nanovesicles Polydopamine Multidrug-resistant pseudomonas aeruginosa Methicillin-resistant staphylococcus aureus Carbapenem-resistant escherichia coli Photothermal therapy Lower critical solution temperature Extracellular vesicles Plant-derived nanovesicles Streptozocin 2’,7’-Dichlorofluorescin diacetate 4’,6-diamidino-2-phenylindole Dulbecco’s modified eagle medium Fetal bovine serum Phosphate buffer solution Bovine serum albumin Cell counting kit-8 Red blood cells American type culture collection Reactive oxygen species Transmission electron microscopy Near-infrared Phosphatidic acid Confocal laser scanning microscopy Scanning electron microscope Principal component analysis Differentially expressed genes Gene ontology Kyoto encyclopedia of genes and genomes Extracellular matrix Author contributions Bodeng Wu, Shan Lin and Qi Xiu contributed equally to this work.Conceptualization: ZL, PWL, WBD, LB. Method: WBD, PWL, LS, XQ, ZMZ, HYY, WZX, LCC, MJG. Investigation: WBD, LS, XQ, WXH, ZMZ, LSH. Visualization: PWL, WBD, XQ, WXH, ZMZ, LLP. Capital Acquisition: ZL, LB, PWL. Project management: ZL, LB, PWL, WBD. Supervision: ZL, LB, PWL, WBD, LS, HHY. Writing – First Draft: WBD, LS, XQ. Writing – Reviewing and Editing: ZL, LB, PWL, WBD. Funding This work was supported by the National Science Fund for Distinguished Young Scholars of China (82025024), National Key R&D Program of China (2021YFA1300604), National Natural Science Foundation of China (82302639 and 82172371), and Guangdong Provincial Clinical Research Center for Laboratory Medicine (2023B110008). Data availability The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request. Declarations Ethics approval and consent to participate The animal study was approved by the Laboratory Animal Management Committee of Southern Medical University (Approval No. SMUL2022179). Consent for publication All authors have approved the manuscript and agree for the submission. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Bodeng Wu, Shan Lin and Qi Xiu contributed equally to this work. Contributor Information Bo Li, Email: nfyylibo2012@smu.edu.cn. Weilun Pan, Email: 1772672757@qq.com. Lei Zheng, Email: nfyyzhenglei@smu.edu.cn. References Associated Data Supplementary Materials Supplementary Material 1: Supplementary data to this article can be found in the appendix Data Availability Statement The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.