Polyhexamethylene guanidine derived carbon dots with antibacterial and immunoregulatory properties loaded on carboxymethyl chitosan and ε-poly-L-lysine hydrogel to achieve sustained drug release for treating Pseudomonas aeruginosa infected wound

Abstract

The treatment of infectious wounds is confronted with the challenges of antibiotic-resistant bacterial infections and excessive inflammatory responses, significantly impacting the healing process. A multifunctional hydrogel dressing system used as local strategy is needed to address this clinical dilemma. Herein, a kind of carbon dots (P-CDs) were synthesized using polyhexamethylene guanidine (PHMG) and citric acid (CA) through a hydrothermal method, exhibiting excellent broad-spectrum antibacterial activity, including effective inhibition of clinically isolated resistant bacteria. The antibacterial mechanism of P-CDs involved the generation of reactive oxygen species, inhibition of ATP synthesis, and disruption of bacterial membrane integrity with the leakage of protein. Additionally, P-CDs demonstrate pronounced immunomodulatory properties by inducing macrophage polarization toward the M2 phenotype, thus modulating the inflammatory microenvironment. These functional components were incorporated into an intelligent hydrogel system (P-CDs@EPCS) constructed from ε-poly-L-lysine and carboxymethyl chitosan, resulting in a wound dressing with tissue adhesion, resilient maintenance of mechanical integrity, and favorable mechanical properties. P-CDs@EPCS exhibited a sustained pH-responsive drug release profile for up to 10 days, maintaining the stability of active constituents throughout the release period. In vivo experiments confirmed that P-CDs@EPCS significantly accelerated wound healing in infections caused by Pseudomonas aeruginosa, demonstrating a synergistic effect of antibacterial, anti-inflammatory, and repair-promoting actions. This study provides a promising solution for treating infectious wounds, integrating key advancements in the field of hydrogels, including dynamic crosslinking design and smart drug delivery systems.

Graphical abstract

Integration of polyhexamethylene guanidine derived carbon dots with antibacterial and immunoregulatory properties into carboxymethyl chitosan and ε-poly-L-lysine hydrogel for controllable and effective accelerating infectious wound healing.

1. Introduction

The World Health Organization has reported that chronic wounds impact approximately 2 % of the global population, leading to prolonged physical and psychological burdens owing to impaired healing [1,2]. The repair of chronic wound is obstructed by multiple factors, including bacterial infections, sustained inflammation, oxidative stress, and dysregulated angiogenesis [3]. Notably, complicated infections in the wound microenvironment play a pivotal role in hindering granulation tissue formation and delaying healing [4]. Chronic wound microbiomes are dominated by 3–10 bacterial species, with Staphylococcus aureus (S. aureus) and Pseudomonas aeruginosa (P. aeruginosa) being the most prevalent [3,4]. S. aureus is in nearly all chronic wounds, while over 50 % are co-colonized by P. aeruginosa. P. aeruginosa-induced wound infections cause severe tissue necrosis, elevate systemic infection risks, and substantially complicate therapeutic interventions due to their susceptibility to multidrug resistance [5]. Although topical long-acting administration is a crucial means of wound repair, the emergence of multidrug-resistant strains significantly limits the efficacy of antibiotics and local antimicrobial therapies [6]. Moreover, emerging evidence highlights the importance of modulating inflammation to accelerate healing in infectious wounds [7]. Macrophages, as key innate immune cells, contribute to pathogen clearance, debris removal and regulation of wound inflammation [8]. Beyond phagocytosis, macrophages typically regulate immune cell activity and adapt their phenotypes to the various stages of wound healing through polarization. Macrophages exist in two primary polarization states of pro-inflammatory M1 and anti-inflammatory M2. M1 macrophages secrete cytokines like TNF-α and IL-1β to combat infections, whereas M2 macrophages promote tissue repair and angiogenesis via cytokines like IL-10 and IL-4. Early M1 activation is essential for infection control, but prolonged M1 dominance perpetuates inflammation and hinders healing. The orderly transformation of macrophages from M1 phenotype to M2 phenotype, which is disrupted by continuous infection, fails to alleviate inflammation and support tissue regeneration, leading to chronic wounds [9]. Consequently, strategies to continuously combat bacterial infection and induce macrophage anti-inflammatory reprogramming in the local wound area will offer promising therapeutic avenues for managing infected wound.

Local dressing is a significant means to achieve sustained and safe drugs with antibacterial and immunoregulatory properties for treating wounds [10,11]. Nanomedicines, as emerging alternatives in therapy, exhibit substantial potential to replace conventional antimicrobial agents and augment antibiotic efficacy [12,13]. Among them, carbon dots (CDs) are particularly notable in the biomedical field due to their unique stable physicochemical profiles, favorable biocompatibility, customizable surface functionalization, and efficient activity [14]. Antibacterial CDs prepared from small molecule compounds [15], polymers [16], and drugs [17] are regarded as promising candidates for treating infectious diseases, owing to their enhanced antibacterial properties and low susceptibility to drug resistance [18]. Although CDs have great potential in controlling infection, they still face limitations in the management of long-term intractable wounds. On the one hand, individual antibacterial properties are insufficient to support multifunctional requirements of wound healing. On the other hand, their water dispersibility can impair mechanical stability, skin adhesion, and controlled drug release, which are crucial for local treatment approaches [19,20]. Therefore, it is necessary to obtain multifunctional CDs with antibacterial properties and applicability to special carriers for high-performance local treating strategies. By adjusting the raw materials, CDs with both antibacterial and antioxidant properties [15,21], as well as antioxidant and immunoregulatory properties [22,23], were obtained through specialized synthesis strategies, indicating the possibility of preparing a certain new kind of CDs with both antibacterial and immunomodulatory properties, thereby improving the effectiveness of local wound treatment. While, the achievement of the CDs with antibacterial and immunomodulatory properties is still in the initial stage.

Hydrogels, as highly cross-linked, three-dimensional hydrophilic polymer networks, are known for their biocompatibility, biodegradability, controllable mechanical properties, and carrier ability [24]. Correspondingly, hydrogels are considered as ideal materials for wound dressings [25]. Carboxymethyl chitosan (CMCS) containing abundant carboxyl and amino groups exhibits pH responsive behavior, mucosal adhesion, and controllable degradation characteristics [26]. ε-poly-L-lysine (EPL) with its amino groups enhances the mechanical and drug-loading properties of hydrogels through electrostatic interactions [27]. Multifunctional hydrogels prepared from chitosan and EPL have attracted increasing attention for improving drug delivery efficiency and therapeutic effect. While, a current challenge is to fully utilize the inherent characteristics of CMCS and EPL to develop hydrogels that simultaneously achieve high drug loading efficiency and sustained release performance [28]. It is necessary to systematically study the preparation method, drug loading and release behavior of composite hydrogels using CMCS and EPL as raw materials, in order to establish basic principles and technical support for designing advanced intelligent drug delivery systems.

By binding to the cell membrane of bacteria and fungi, polyhexamethylene guanidine (PHMG) disrupts the integrity of the cell membrane and exerts its antibacterial effect [29]. Correspondingly, PHMG has been applied in sterilization and water treatment [30]. Nevertheless, the significant threats of PHMG to eucaryotic cells and hemolytic activity restrict the direct biomedical application of PHMG [31]. Recent studies have shown that CDs with enhanced antibacterial properties can be obtained from antibacterial materials such as antibiotics [32,33]. Combined with our previous work on CDs with specific response from citric acid (CA) [34], herein, as shown in Scheme 1, a type of CDs (P-CDs) with broad-spectrum antimicrobial activity was effectively synthesized via a one-step hydrothermal method using PHMG and CA as precursors. Furthermore, P-CDs played an immunomodulatory capability by downregulating pro-inflammatory cytokines of TNF-α and IL-6 in macrophages. Moreover, polyvinyl alcohol (PVA) was used as a chemical crosslinker to conjugate CMCS and EPL, forming a composite hydrogel (EPCS). Subsequent incorporation of P-CDs into EPCS resulted in the drug-loaded hydrogel system, designated as P-CDs@EPCS. P-CDs@EPCS serves as an effective dressing for treating infected wounds with three critical advantages: (1) enhanced drug delivery efficiency with pH-responsive release kinetics (sustained release over 10 days); (2) dual antibacterial and anti-inflammatory effects via macrophage M2 polarization, while overcoming bacterial resistance (including clinically isolated drug-resistant strains); and (3) significantly accelerated wound healing and tissue regeneration.

Scheme 1.

The preparation and encapsulation of P-CDs into EPCS to form P-CDs@EPCS for treating P. aeruginosa-infected wounds.

2. Experimental sections

2.1. Reagents

Polyhexamethylene guanidine hydrochloride (PHMG·HCl), CA, dimethyl diallyl ammonium chloride (DDA), PVA 1799, CMCS, Propidium iodide (PI) and 2′,7′-dichlorodihydrofluorescein diacetate (DCFH-DA) were purchased from Aladdin Industrial Corporation (Shanghai, China). ε-Poly-L-lysine hydrochloride (EPL, MW < 5000) was from Macklin Biochemical Technology Co., Ltd (Shanghai, China). Nutrient Agar (NA) and Mueller–Hinton (MH) broth were provided by Qingdao Hope Bio-technology Co., Ltd (Qingdao, China). Cellulose dialysis bags (500–1000) and 4 % rabbit red blood cell suspensions were provided by Shanghai Yuanye Bio-Technology Co., Ltd (Shanghai, China). The BCA kit was purchased from Beijing Zoman Biotechnology Co., Ltd (China). The ATP kit was purchased from Abbkine Biotechnology Co., Ltd (Wuhan, China). LPS from E. coli O111:B4 was purchased from Sigma Aldrich (USA). The CCK-8 kit was purchased from the Glpbio Technology Inc (USA). The Mouse TNF-α ELISA Kit, Mouse IL-10 ELISA Kit, Mouse IL-4 ELISA Kit and Mouse IL-6 ELISA Kit were purchased from Jianglai Biotechnology Co., Ltd (Shanghai, China). The standard strains of Methicillin-Resistant Staphylococcus aureus (MRSA) (ATCC43300), and Pseudomonas aeruginosa (ATCC27853) were purchased from SuZhou, BNCC Sci.&Tech. Co., Ltd (Shanghai, China). Human Umbilical Vein Endothelial Cells (HUVEC), Murine Fibroblast Cell Line (L929), and RAW264.7 cells were obtained from Procell Life Sci.&Tech. Co., Ltd (Wuhan, China).

Transmission electron microscopy (TEM) was operated on an FEI Talos F200S transmission electron microscope (Thermo Fisher Scientific, USA). Fluorescence spectra were recorded on a Cary Eclipse fluorescence spectrophotometer (Agilent Technologies, USA). UV–vis absorbance spectra were recorded on a UV-2450 UV–visible spectrophotometer (Shimadzu, Japan). Fourier transform infrared (FTIR) spectra were recorded on a Nicolet iS50 Fourier transform infrared spectrometer (Thermo Fisher Scientific, USA). Raman spectra were recorded on HORIBA HR Evolution (HORIBA Scientific, France). X-ray photoelectron spectroscopy (XPS) was recorded using an X-ray photoelectron spectrometer (Thermo K-alpha, Thermo Fisher Scientific, USA). Scanning electron microscopy (SEM) was recorded on an SU8600 (Hitachi, Japan). Rheometry was recorded on a rheometer (Discovery HR-1, TA Instruments, Waters Corporation). Cell fluorescence images and Flow cytometry were obtained using an inverted fluorescence microscope (Leica DMIRB, Leica Microsystems, USA) and an FACSVerse (BD Biosciences, USA), respectively. Confocal laser scanning images (CLSM) were acquired on a Leica SP5 (Leica, Germany).

2.2. Preparation and characterization of the P-CDs

Four hundreds mg of CA, 2 mL DDA and 2 mL 0.05 g/mL PHMG·HCl were dissolved in deionized water to achieve a total volume of 20 mL. The mixed solution was transferred to a Teflon reactor and subjected to hydrothermal reaction at 180 °C for 6 h. The resulting product was then dialyzed using a dialysis bag with a molecular weight cut-off (MWCO) of 500–1000 Da for 48 h, with the water changed every 2 h. The purified liquid in the dialysis bag was freeze-dried to obtain the solid P-CDs, which were stored at 4 °C for subsequent experiments.

2.3. Preparation of P-CDs@EPCS

Briefly, 1.2 g of PVA was completely dissolved in 15 mL water with magnetic stirring at 90 °C for 1 h. Subsequently, 0.6 g of CMCS was added, and the mixture was stirring for 1 h to obtain a homogeneous CMCS/PVA solution. Then, the 5 mL aqueous solution containing 0.15 g EPL was added to the CMCS/PVA solution, and the mixture was subsequently stirred for 4 h to ensure complete dissolution and form the hydrogel precursor solution. Following three freeze-thaw cycles, the hydrogel precursor solution was transformed into a homogeneous EPL/PVA/CMCS composite hydrogel (EPCS).

Two milliliters of 100 μg/mL P-CDs were introduced into the 3 mL hydrogel precursor solution. The resulting P-CDs incorporated EPL/PVA/CMCS composite hydrogels are designated as P-CDs@EPCS, in which the equivalent concentration of P-CDs was calculated to be 40 μg/mL. P-CDs@EPCS was poured into a mold and subjected to three freezing–thawing cycles (frozen at −20 °C for 6 h and then thawed at room temperature for 3 h) to obtain the hydrogel with desired shapes.

2.4. Rheological characterization

The viscoelastic properties of EPCS were evaluated using a TA Discovery HR-1 rheometer under physiological conditions (25 °C, pH 7.4). Time-dependent and frequency-dependent rheological behaviors were assessed through consecutive measurements of shear stress. For the time sweep analysis, the hydrogel was subjected to six alternating cycles of low-strain (1 %, 50s) and high-strain (300 %, 50s) at a frequency of 5 Hz over 1 h, with continuous monitoring of storage (G′) and loss (G″) moduli to evaluate structural recovery. Frequency sweep measurements were simultaneously performed from 0.1 to 100 rad/s at a strain of 1 % to determine the mechanical spectrum of the hydrogel.

2.5. Swelling performance test of P-CDs@EPCS

For the swelling test, dried EPCS and P-CDs@EPCS were immersed in PBS (pH 7.4). After predetermined time intervals, the samples were carefully removed from the PBS. Excess solution on the surface was blotted dry with filter paper, and the mass of the swollen hydrogels was precisely measured. The swelling ratio at each time point was calculated using the following equation:

$$\text{Swelling}\text{Radio}(\%)=\frac{\mathrm{W}1-\mathrm{W}0}{\mathrm{W}0}\times 100\%$$

W1 denotes the mass of the hydrogel following swelling, while W0 represents the initial mass of the hydrogel.

2.6. Degradation behavior of P-CDs@EPCS

Identically weighed EPCS samples were immersed in 10 mL of PBS at pH 5.5 and 7.4, respectively. The samples were incubated at 37 °C with shaking at 240 rpm for specified durations. At designated time points, samples were retrieved, dried, and weighed. The degradation ratio was calculated as follows:

$$\text{Degradation}\text{Radio}(\%)=\frac{\mathrm{W}0-\mathrm{W}1}{\mathrm{W}0}\times 100\%$$

W1 denotes the mass of the hydrogel following immersion, while W0 represents the initial mass of the hydrogel.

2.7. The release behavior of P-CDs@EPCS

Briefly, one g P-CDs@EPCS was soaked and continuously agitated at 37 °C in 10 mL PBS at pH 5.5 and 7.4, respectively. At predetermined time intervals, 200 μL of the release medium were extracted for release profile analysis and replenished with an equivalent volume of fresh PBS. The quantity of released P-CDs was measured using a microplate reader at 205 nm.

2.8. In vitro antibacterial experiments of P-CDs@EPCS

2.8.1. Evaluating the minimum inhibitory concentration (MIC) of P-CDs

A series of MH broth tubes containing gradient concentrations of P-CDs were prepared via broth microdilution. Cultured bacterial colonies were suspended in saline and adjusted to 0.5 McFarland standard (1.5 × 10⁸ CFU/mL), followed by a 100-fold dilution to 1.5 × 10⁶ CFU/mL. Each tube received 1 mL of P-CDs-containing MH broth and 1 mL of bacterial suspension (1.5 × 10⁸ CFU/mL). Negative controls contained 2 mL of pure MH broth, and positive controls contained 1 mL MH broth and 1 mL bacterial suspension. After 24-h incubation at 37 °C, bacterial growth was assessed visually. The MIC of P-CDs against each strain was defined as the lowest concentration that completely inhibited visible growth.

2.8.2. Standard plate counting assay and zone of inhibition assay

Bacterial treatments followed identical experimental conditions. After co-culturing different bacterial suspensions with 40 μg/mL P-CDs, 10 % (w/v) EPCS extract, or 10 % (w/v) P-CDs@EPCS extract at 37 °C for 24 h, the corresponding samples were centrifugated at 240 rpm. Then, each of the mixtures was diluted 1000-fold. Subsequently, 100 μL aliquots were spread onto MH agar plates and incubated at 37 °C for 24 h for photography. Zone of inhibition assay: Overnight cultures of different clinical isolates of P. aeruginosa were diluted to 1 × 10⁵ CFU/mL. Sterile swabs were moistened with the bacterial suspension and used to evenly streak the surface of MH agar plates. Susceptibility paper disks were soaked separately in 40 μg/mL P-CDs solution, 10 % (w/v) EPCS extract, and 10 % (w/v) P-CDs@EPCS extract for 12 h. Then, they were removed with sterile forceps and placed onto the inoculated plates. All plates were incubated at 37 °C for 24 h, after which the zones of inhibition were observed and their diameters measured for each group.

2.8.3. Bacterial growth kinetics assay

Fifty μL of P-CDs at varying concentrations were added to a 96-well plate (NEST, Wuxi, China), followed by 50 μL of different bacterial suspension (1.5 × 10⁶ CFU/mL). After mixing, the solution was incubated at 37 °C for 12 h. The OD₆₀₀ was monitored every 2 h during the initial 12 h of incubation. For the antibacterial assay of prolonged drug release, the procedure followed the same steps as the aforementioned drug release test. At specific time intervals, 1 mL of the cumulative drug release solution was collected. This solution was then combined with 1 mL of two standard bacterial strains (1.5 × 10⁶ CFU/mL) and co-cultured for 12 h. Then, the OD₆₀₀ was measured to evaluate bacterial growth.

2.8.4. Biofilm disruption assessment

P. aeruginosa and MRSA biofilms were grown in confocal dishes for 48 h using bacterial suspensions (1.5 × 10⁸ CFU/mL). After washing with PBS, biofilms were treated with P-CDs, EPCS, or P-CDs@EPCS for 24 h. Live/dead staining was performed with SYTO-9 (6 μM) and PI (30 μM) at 37 °C for 30 min in the dark. After washing with PBS three times, CLSM images of the biofilms were acquired using a Leica system (LAS X software, Germany). For quantitative analysis, biofilms were prepared and treated similarly. After 48-h incubation with agents, wells were washed with PBS, fixed with methanol (15 min), stained with 0.1 % (w/v) crystal violet (15 min), and re-washed. Images were captured and quantified via ImageJ.

2.8.5. BCA and ATP assays

P. aeruginosa and MRSA suspensions were prepared in saline with an OD₆₀₀ of 0.5. Subsequently, 700 μL of P-CDs, EPCS extract, or P-CDs@EPCS extract were added to 700 μL of bacterial suspension, followed by incubation at 37 °C with shaking for 4 h. After centrifugation (5000 rpm, 5 min), supernatants were collected for extracellular protein quantification using BCA and ATP assay kits according to the manufacturers’ protocols.

2.8.6. Bacterial ROS generation assay

Bacterial suspensions (1 mL, 1.5 × 10⁸ CFU/mL) were treated with 1 mL of 40 μg/mL P-CDs, 10 % (w/v) EPCS extract, or 10 % (w/v) P-CDs@EPCS extract at 37 °C for 4 h. After centrifugation at 5000 rpm for 5 min, the pellets were washed three times with saline and incubated with 10 μM DCFH-DA for 30 min. ROS production was monitored using inverted fluorescence microscopy.

2.9. In vitro cytotoxicity and effects of P-CDs@EPCS on cell functions

2.9.1. Cell viability

HUVEC and L929 at logarithmic growth phase were trypsinized, harvested, and centrifuged. Cells were seeded in 96-well plates at 1.0 × 10⁴ cells/well. DMEM medium containing PBS (control), 40 μg/mL P-CDs, 10 % (w/v) EPCS extract, or 10 % (w/v) P-CDs@EPCS extract was added to respective wells. After 24-h incubation (37 °C, 5 % CO₂), the medium was replaced with 10 % CCK-8 reagent. Absorbance at 450 nm was measured using a microplate reader. Cell proliferation activity was calculated as follows:

$$\text{Cell}\text{Proliferation}\text{Activity}(\%)=\frac{{OD}_s-{OD}_b}{{OD}_c-{OD}_b}\times 100\%$$

OD_b, OD_c, and OD_s represent the absorbance at 450 nm of the blank medium, untreated cells in medium, and the experimental group, respectively.

2.9.2. Living/dead cell double-staining

L929 and HUVEC cells were respectively seeded onto 24-well plates at a density of 1.0 × 10⁴ cells per well. According to the experimental grouping, each well was supplemented with DMEM medium containing PBS, 40 μg/mL P-CDs, 10 % (w/v) EPCS extract, or 10 % (w/v) P-CDs@EPCS extract. The cells were incubated in an incubator at 37 °C for 24 h with 5 % CO₂. After incubation, the cells were stained with 2 μM Calcein AM and 4 μM PI at 37 °C for 30 min to facilitate co-staining for imaging.

2.9.3. Hemolysis test of P-CDs@EPCS

To evaluate hemocompatibility, a 4 % rabbit red blood cell suspension (Yuanye Bio-Technology, Shanghai, China) was prepared and mixed with 40 μg/mL P-CDs, 10 % (w/v) EPCS extract, or 10 % (w/v) P-CDs@EPCS extract at 37 °C for 1 h in a 1:1 ratio. These complexes were then centrifuged at 2400 rpm for 5 min, using physiological saline treatment as a negative control and ddH₂O treatment as a positive control. Hemolysis percentage was calculated by measuring the OD₅₄₀ value of the supernatant.

$$\text{Hemolysis}\text{Rate}(\%)=\left(\frac{\left(Hd-Hb\right)}{\left(Hs-Hb\right)}\right)\times 100\%$$

where $Hd$ is the optical density of the sample group. $Hb$ is the optical density of the negative control group. $Hs$ is the optical density of the positive control group.

2.9.4. Cell scratch assay

L929 and HUVEC cells were uniformly seeded onto 6-well plates at a density of 1.0 × 10⁶ cells per well. Once the cells reached over 90 % confluence, the medium was removed, and a vertical scratch was made at the center using a 10 μL pipette tip. Wells were subsequently rinsed three times with PBS. Then, DMEM medium containing PBS, 40 μg/mL P-CDs, 10 % (w/v) EPCS extract, or 10 % (w/v) P-CDs@EPCS extract was added, respectively. Scratch closure was observed at 0 and 24 h using an inverted fluorescence microscope. The scratch closure rate was calculated using the formula:

The scratch healing rate (%) =(W₀-W₁)/W₀ × 100 %

W₀ and W₁ represents the scratch area at 0 and 24 h, respectively.

2.9.5. Transwell cell migration assay

In a 24-well transwell plate, 40 μg/mL P-CDs, 10 % (w/v) EPCS extract, or 10 % (w/v) P-CDs@EPCS extract were prepared in the lower chamber. L929 cells were seeded in the upper chamber at a density of 5.0 × 10³ cells per well. The lower chamber was filled with DMEM containing 1 % FBS, with blank wells serving as the control group. After 24 h of incubation, the upper chamber was collected, and the recruited L929 cells were fixed with 4 % paraformaldehyde for 30 min. The cells were then stained with 0.1 % (w/v) crystal violet for 15 min. Images of the recruited L929 cells were captured using an inverted microscope, with three random images taken for each upper chamber sample. The number of recruited L929 cells in each image was quantified for analysis.

2.10. In vitro anti-inflammatory assay

RAW264.7 cells were seeded at a density of 1.0 × 10⁵ cells per well in a 12-well plate using complete DMEM medium (containing 10 % fetal bovine serum and 1 % penicillin-streptomycin). The untreated RAW264.7 cells were designated as blank group, while the inflammatory RAW264.7 cells induced by LPS (1 μg/mL) and IFN-γ (20 ng/mL) were set as the control group. Cells were co-cultured for 24 h with either P-CDs, EPCS extract, or P-CDs@EPCS extract. After incubation, cells were collected, centrifuged, and washed with phosphate-buffered saline (PBS). Cells were then fixed with 4 % paraformaldehyde for 30 min, washed, and permeabilized with 0.1 % Triton-X100 for 10 min. CD86 and CD206 antibodies were added, and cells were incubated in the dark for 30 min. After washing, fluorescence intensity was analyzed using flow cytometry (FACSVerse, BD). Using the same methods, cell culture supernatants were collected post-incubation to measure TNF-α, IL-6, IL-4, and IL-10 using ELISA kits.

To assess the anti-inflammatory effects of different hydrogels on the interaction between RAW264.7 and HUVEC cells, an indirect co-culture model was established. RAW264.7 cells were cultured in DMEM containing 10 % FBS and 1 % antibiotics at a density of 1.0 × 10⁵ cells per well in 12-well plates. The cells were stimulated with 1 μg/mL LPS and 20 ng/mL IFN-γ to simulate an inflammatory microenvironment. After 24 h of co-incubation with P-CDs, EPCS, or P-CDs@EPCS, the supernatants were collected by centrifugation. HUVEC cells were seeded in the upper chamber of a Transwell at a density of 3.0 × 10⁴ cells per well, while 600 μL of the collected supernatants from each group was added to the lower chamber. The cells were incubated at 37 °C with 5 % CO₂ for 36 h. After removing the upper chamber cells, the migrated cells in the lower chamber were washed three times with PBS, fixed with 4 % paraformaldehyde at room temperature for 30 min, and stained with 0.1 % crystal violet for 15 min. The number of migrated cells was then observed and quantified using an inverted microscope.

2.11. P-CDs@EPCS accelerated P. aeruginosa-infected wounds healing

The in vivo experiments were approved by the Animal Ethics Committee of Fujian Medical University (Ethics Approval Number: IACUC FJMU 2024-0290) and were conducted in accordance with the “Guidelines for Animal Experiments.” SPF-grade SD rats purchased from Spibio Biotechnology Co., Ltd. (Beijing; Production License No. SCXK(Jing)2019-0030). The in vivo safety of P-CDs@EPCS was assessed in Sprague-Dawley (SD) rats. Six male SD rats (6–8 weeks old, 200 ± 20 g) were randomly divided into two groups: normal saline (control) and P-CDs@EPCS. Based on body weight, each rat received a topical application on the dorsal skin of 0.3 mL/100 g of either normal saline or P-CDs@EPCS hydrogel (20 wt%) once daily for ten consecutive days. Throughout the study, the rats were maintained under standard housing conditions and monitored for general health status. On Day 10, after the final treatment, the animals were euthanized, and major organs (heart, liver, spleen, lung, and kidneys) were collected, fixed in 4 % paraformaldehyde, and subjected to hematoxylin and eosin (H&E) staining to evaluate potential histopathological changes.

To evaluate the therapeutic efficacy of P-CDs@EPCS, 24 male SD rats (6–8 weeks old, 200 ± 20 g) were acclimated for one week and then randomly assigned into four groups: control (normal saline), P-CDs (80 μg/mL), EPCS (20 wt%), and P-CDs@EPCS (20 wt%). Following anesthesia by intraperitoneal injection of 4 % chloral hydrate (2 mL/kg), the dorsal area of each rat was shaved using depilatory cream and disinfected. A full-thickness skin defect with a diameter of approximately 8 mm (from epidermis to dermis) was created on the dorsal skin using sterile surgical scissors. For bacterial infection, 50 μL of P. aeruginosa suspension (1.5 × 10⁸ CFU/mL) was directly applied onto the wound surface and allowed to air-dry naturally, establishing the infected wound model. On the day of wound creation and infection, this time point was designated as Day 0. Twenty-four hours later (Day 1), the wounds of the four groups were treated with either physiological saline (control), dressings soaked with P-CDs solution, or a single application of hydrogel (EPCS or P-CDs@EPCS), respectively. Wound closure was recorded on days 0, 4, 8, and 10 using a digital camera under consistent settings and monitored by caliper measurement. The wound area was quantified from photographs using ImageJ software. On Day 10, wound tissues were harvested for histological and immunohistochemical analysis. Skin around the wound was excised, fixed in 4 % paraformaldehyde for 24 h, and embedded in paraffin. Sections (5 μm) were stained with H&E and Masson’s trichrome staining for histological evaluation. For immunohistochemistry, paraffin sections underwent citrate buffer (pH 6.0) microwave-mediated antigen retrieval, blocking with 5 % normal serum in PBS for 1 h, and incubation with primary antibodies against CD86, CD206, vascular endothelial growth factor (VEGF), and collagen I (COL-1) overnight at 4 °C. After washing, biotinylated secondary antibodies were applied for 1 h at room temperature, and signal detection was performed using a DAB substrate kit followed by hematoxylin counterstaining, dehydration, clearing, and mounting.

2.12. Statistical analysis

All experiments were performed at least three times. The statistical significance of the differences was analyzed using one-way analysis of variance (ANOVA). Asterisks indicate statistical significance: ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001. Graphs and statistical analyses were performed using GraphPad Prism software version 20 and OriginPro 2022 (Learning Edition).

3. Results and discussion

3.1. Characterization of P-CDs

First, the morphology of P-CDs was examined using TEM (Fig. 1A). The analysis revealed that P-CDs are torispherical particles with an average size of 9.94 ± 0.05 nm, demonstrating good dispersion and uniformity. The lattice spacing was measured at 0.22 nm, corresponding to the diffraction plane of graphitic carbon [35]. The UV–Vis absorbance spectrum of P-CDs (Fig. 1B) showed an absorption peak at 205 nm, attributing to the π→π∗ transition of C=C bonds within the P-CDs [36]. Fig. 1B (insert) demonstrated that P-CDs displayed excitation wavelength-dependent properties, with emission peaks gradually red-shifting as the excitation wavelengths increased from 300 to 380 nm. This behavior indicates the characteristic fluorescence properties of CDs [37]. Correspondingly, the P-CDs exhibited maximum emission intensity at 406 nm when excited at 350 nm (Fig. 1B).

Fig. 1.

Characterization of P-CDs. (A) The TEM, HRTEM (top right) of P-CDs and the size distribution of P-CDs (top left), scale bar = 20 nm. (B) The UV–vis absorption and fluorescence spectra of P-CDs, insert is the fluorescence emission spectra of P-CDs at 300–380 excitation. (C) The FTIR spectra of P-CDs, PHMG, CA and DDA. (D) Raman spectrum of P-CDs. The XPS spectrum (E), and high-resolution XPS spectra of C1s (F), N1s (G), O1s (H), and Cl2p (I) of P-CDs.

The surface functional groups of P-CDs were analyzed. As shown in Fig. 1C, the FTIR spectrum of CA showed absorption peaks at 3307 cm⁻¹ and 1705 cm⁻¹, corresponding to the stretching vibrations of O-H and C=O, respectively. Similarly, P-CDs exhibited a C=O absorption peak at 1714 cm⁻¹ and an O-H stretching vibration peak at 2927 cm⁻¹. The FTIR spectrum of DDA showed absorption peaks at 3470 cm⁻¹ and 959 cm⁻¹, which were attributed to the stretching vibrations of C-H and the bending vibrations of C-H out-of-plane on the ring, respectively. P-CDs also showed the peaks at 3456 cm⁻¹ and 1047 cm⁻¹ corresponding to the C-H stretching. The C=N stretching vibration of the guanidine group in PHMG shifted from 1671 cm⁻¹ to 1633 cm⁻¹ in P-CDs, indicating the preservation of the guanidine structure [36]. Additionally, the peaks at 1540 cm⁻¹ for PHMG and 1539 cm⁻¹ for P-CDs were assigned to the stretching vibration of N-H in the guanidine group. The retention of characteristic peaks associated with the guanidine group in P-CDs confirms the successful guanidinylation of P-CDs. The Raman spectrum (Fig. 1D) showed the characteristic peaks at 1349 cm⁻¹ (D peak) and 1569 cm⁻¹ (G peak). The D peak corresponds to disordered carbon structures, indicating sp³ doping by heteroatom and surface defects in the graphene structure. The G peak reflects the in-plane vibration of sp² hybridized carbon atoms. The intensity ratio (I_D/I_G = 0.36) suggests a significantly high degree of graphitization in the sp² carbon domains of P-CDs [38]. Additionally, the Raman spectrum showed higher peak positions for P-CDs at 1230 cm⁻¹ and 1368 cm⁻¹. According to precursor analysis, the peak at 1230 cm⁻¹ mainly arises from C-O bond vibrations in carboxylate groups of CA, while the peak at 1368 cm⁻¹ is associated with C-N bonds in guanidine groups of PHMG [39,40].

The elemental composition of P-CDs was analyzed using XPS. The full-range XPS spectrum of P-CDs showed four main peaks, corresponding to C1s, O1s, N1s, and Cl2p as shown in Fig. 1E. The binding energies for four peaks were 284.78 eV, 532.44 eV, 400.09 eV, and 198.3 eV, with atomic contents of 45.7 %, 22.62 %, 4.06 % and 5.25 %, respectively. The high-resolution XPS spectra of C1s (Fig. 1F) can be deconvoluted into three peaks, which were attributed to C-C/C-H (284.83 eV), C-N/C-O (286.08 eV), and C=O/C=N (288.10 eV). The high-resolution XPS spectra of N1s (Fig. 1G) illustrated two peaks at 399.42 eV and 401.77 eV, corresponding to pyrrolic N and quaternary N, suggesting the presence of positively protonated amines [41]. Moreover, the high-resolution XPS spectra of O1s (Fig. 1H) suggested the two peaks of 530.22 eV and 531.91 eV, which were assigned to C=O and C-O, respectively. The presence of carboxyl-related groups (O-C=O, C-O) can promote the participation of P-CDs in immune cell receptor for influencing macrophage polarization [42,43]. The high-resolution XPS spectra of Cl2p (Fig. 1I) exhibited two peaks at 198.2 eV (Cl2p_3/2) and 199.8 eV (Cl2p_1/2) with a spin-orbit splitting of approximately 1.6 eV [44]. This result indicates that Cl in the P-CDs mainly exists in a stable ionic form, most likely originating from the Cl-containing precursor of PHMG·HCl. The Cl species are presumed to be retained and uniformly distributed in the P-CDs structure or bound to surface functional groups. The presence of ionic chlorine likely plays a role in surface passivation, which contributes to the stability of the P-CDs. The integrated FTIR, Raman spectroscopy, and XPS analyses confirm the retention of guanidinium, carboxyl, and quaternary ammonium groups on P-CDs. While, the spatial distribution and charge density of surface functional groups are still needed further studied for better understanding of the molecular structure of P-CDs.

3.2. Preparation and characterization of P-CDs@EPCS

The P-CDs@EPCS were synthesized using a freeze-thaw method (Fig. 2A). PVA was dissolved in water at 90 °C, followed by sequential addition of EPL and CMCS at 50 °C to form a sol precursor (EPCS) for further incorporation of P-CDs. The sol-gel transition of EPCS was successfully achieved through cyclic freeze-thaw processing (Fig. 2B). SEM characterization revealed a three-dimensional and relatively regular porous network in EPCS (Fig. 2C), indicating that hydrogen bonding between EPL and PVA chains stabilized the architecture, preventing pore collapse while enhancing structural integrity. Meanwhile, the EPCS’s loose porous architecture facilitates nutrient exchange and exudate absorption while providing a 3D scaffold for cellular migration and tissue regeneration-essential features [22]. The crosslinked hydrogel network and loaded P-CDs were confirmed by FTIR. As shown in Fig. 2D, the absorption peaks at 3300 cm⁻¹ and 2930 cm⁻¹ indicated the stretching vibrations of O-H and C-H in PVA. CMCS exhibited an asymmetric stretching vibration of carboxylate at 1697 cm⁻¹ and an N-H bending vibration of the amino group around 1538 cm⁻¹ [45]. EPL, rich in ε-amino groups, displayed distinct -NH₂ stretching and bending vibrations at 3317 cm⁻¹ and 1612 cm⁻¹, respectively [46]. In the FTIR of EPCS, the broadened stretching vibration band of O-H at 3307 cm⁻¹ was attributed to intermolecular hydrogen bonding between the amino groups of EPL and hydroxyl groups from PVA [47]. Notably, compared with the CMCS, the carboxylate absorption band red shifted to 1679 cm⁻¹, accompanied by a concomitant displacement of the N-H bending vibration at 1540 cm⁻¹. These spectral shifts suggest the electrostatic interaction between the deprotonated carboxylate groups of CMCS and protonated amino groups from EPL, which collectively contribute to the formation of the hydrogel network [28]. Furthermore, P-CDs@EPCS from the freeze-thaw treatment of sol of EPCS and P-CDs was characterized by FTIR (Fig. 2E). The FTIR spectrum of pristine P-CDs revealed distinct characteristic peaks of the O-H stretching vibration (2927 cm⁻¹), C-H stretching vibration (3456 cm⁻¹), N–H stretching vibration (3420 cm⁻¹), C–N stretching vibration (1640 cm⁻¹), and guanidyl skeletal vibration (1539 cm⁻¹). P-CDs@EPCS retained the guanidyl skeletal vibration peak at 1550 cm⁻¹, confirming the structural integrity of P-CDs after encapsulation. Furthermore, the broadened stretching vibration band of O–H/N–H centered at 3320 cm⁻¹ indicated the enhanced hydrogen bonding interactions in P-CDs@EPCS, which were attributed to the overlapping contributions of guanidyl N–H and hydrogel matrix O–H groups [48]. The redshift of the asymmetric stretching vibration of -COO^- from 1679 cm⁻¹ in EPCS to 1632 cm⁻¹ in P-CDs@EPCS suggested the formation of ionic interactions between the negatively charged -COO^- groups of EPCS and the positively charged guanidyl moieties of P-CDs [20,49]. These spectral changes collectively confirm the successful encapsulation of P-CDs within the EPCS network with enhanced interaction while maintaining the intrinsic chemical functionalities of both components.

Fig. 2.

Preparation and characterization of P-CDs@EPCS. (A) Diagram of P-CDs@EPCS synthesis. (B) Morphological changes of EPCS before and after gelation. (C) SEM images of EPCS. (D) FTIR spectra of EPCS, CMCS, EPL and PVA. (E) FTIR spectra of P-CDs@EPCS, EPCS and P-CDs. (F) Strain sweep tests and (G) frequency sweep tests of the EPCS. (H) Rheological time sweep under cyclic low/high strain of EPCS. (I) Swelling ratio of EPCS and P-CDs@EPCS in PBS (pH 7.4). (K) Degradation ratio of EPCS and P-CDs@EPCS in PBS (pH 7.4) and PBS (pH 5.5) in 24-day period. (J) Cumulative release changes of P-CDs from the P-CDs@EPCS in PBS (pH 7.4) and PBS (pH 5.5) in 10-day period. Data were presented as mean ± SD (n = 3).

The mechanical properties and stability of the hydrogels were characterized through rheological testing. Initially, rheological strain sweeps measurements (Fig. 2F) illustrated that the gel-to-sol transition point of EPCS occurred at nearly 100 % strain, indicating solid-like behavior within this range and demonstrating good mechanical properties [50]. Subsequently, frequency sweep tests were conducted at 1 % strain, revealing that the storage modulus (G′’) was significantly higher than the loss modulus (G″) (Fig. 2G). This suggests a stable network architecture within EPCS that exhibits characteristic elastic solid behavior [45,51]. The cyclic low/high strain time sweep test (Fig. 2H) revealed that under low strain (1 %), the storage modulus (G′) was significantly higher than the loss modulus (G″), indicating that elastic behavior dominated and the hydrogel network remained intact. Upon application of high strain (300 %), G′ sharply decreased, reflecting the disruption of the network structure. Notably, when the strain returned to 1 %, G′ rapidly recovered to its initial value, and this recovery was reproducible over multiple cycles, demonstrating the excellent self-recovery ability and mechanical resilience of the EPCS hydrogel [52]. The possessed mechanical and self-healing properties ensure sufficient support and protection of the prepared hydrogels for treating wounds.

The equilibrium swelling ratio serves as a critical indicator of the internal crosslinking density and structural characteristics of hydrogels. The swelling properties of EPCS and P-CDs@EPCS were characterized. Both EPCS and P-CDs@EPCS rapidly absorbed water and swelled, reaching equilibrium at pH 7.4 (Fig. 2I), which should be due to their hydrophilic and interconnected porous structures. This swelling behavior is beneficial for exudate absorption in wounds [46]. After 6 h of swelling, EPCS and P-CDs@EPCS exhibited the highest swelling ratios of approximately 371 % and 418 % at pH 7.4, respectively. The incorporation of P-CDs introduced numerous -OH and -NH groups into the structure, allowing for the formation of more hydrogen bonds with water in the porous network. This increased the hydrogel’s capacity to absorb and retain water molecules within its three-dimensional structure, resulting in a higher swelling ratio [53]. The pH-responsive degradation behavior of P-CDs@EPCS was systematically evaluated in PBS solutions with pH 5.5 and 7.4 via gravimetric analysis (Fig. 2J). P-CDs@EPCS gradually degraded over time at both pH levels of 5.5 and 7.4, with a higher degradation rate observed at pH 5.5 compared to pH 7.4. By day 23, the degradation percentages of P-CDs@EPCS reached 76.56 % at pH 5.5, significantly higher than the 55.33 % observed at pH 7.4. Concurrently, the drug release profile demonstrated biphasic release kinetics under both pH conditions (Fig. 2K). P-CDs@EPCS showed a rapid release phase in the first 5 days, followed by sustained stabilization after 10 days. Notably, the weakly acidic environment of pH 5.5 enhanced cumulative drug release to 74.9 %, surpassing the 67.87 % release at pH 7.4. This pH-dependent degradation and release behavior mainly originates from the disruption of electrostatic interactions between P-CDs and EPCS, which facilitates P-CDs dissociation and subsequent payload liberation [54]. This pH-responsive characteristic enables rapid release of antibacterial drug during the initial wound healing phase with lower pH environment [55]. Moreover, adhesion is a key parameter for wound dressings, as shown in Fig. S1, P-CDs@EPCS demonstrated rapid and flexible adhesion to the human fingertip. Upon gentle contact, P-CDs@EPCS established a firm attachment. As the finger bended and straightened, P-CDs@EPCS remained elastic and adjusts its shape accordingly. A considerable force was required for the detachment of P-CDs@EPCS from finger. During the stripping process, P-CDs@EPCS maintained its structural integrity without leaving any visible residue. Such adhesion activities ensure stable fixation and close contact with wound surfaces, thus supporting effective healing.

3.3. Antimicrobial activity of P-CDs@EPCS

The antibacterial property of P-CDs was first evaluated. As shown in Table 1, the MICs of P-CDs against S. aureus, E. coli, and P. aeruginosa were 2.5, 5, and 10 μg/mL. Moreover, the MIC of P-CDs to MRSA was 2.5 μg/mL, implying the effectiveness of P-CDs against resistant bacteria. Furthermore, several clinically isolated drug-resistant bacteria (Fig. S2) were also used to evaluate the real antibacterial efficacy of P-CDs. As shown in Table 1, the MIC values of P-CDs against clinically isolated resistant strains ranged from 8 to 32 μg/mL. In addition, Fig. S3 and Table 1 demonstrated that P-CDs also possess antifungal activity against Candida albicans with the MIC value of 8 μg/mL. These results indicate that the P-CDs exhibit antimicrobial and antifungal activity. Additionally, the inhibition zone tests were performed on three different clinically isolated strains of P. aeruginosa (Fig. S4). Both the P-CDs and the P-CDs@EPCS demonstrated significant antibacterial effects, as evidenced by the clear inhibition zones. These results further confirm the great value of prepared P-CDs and P-CDs@EPCS against clinically resistant pathogens, suggesting the great promising application feasibility.

Table 1.

MIC values of P-CDs against common standard strains and clinically isolated strains.

Bacterial strains	Standard strain number/Label of clinical samples	MIC (μg/mL)
P. aeruginosa	ATCC27853	10
E. coli	ATCC25922	5
MRSA	ATCC43300	2.5
S. aureus	ATCC6538	2.5
Resistant-E. coli from clinic	6680	8
MRSA from clinic	6596	8
MRSA from clinic	6201	8
Resistant-P. aeruginosa from clinic	5917	16
Resistant-P. aeruginosa from clinic	6187	32
Resistant-P. aeruginosa from clinic	6602	32
Candida albicans from clinic	4391	8
Candida albicans from clinic	4297	8

Researches indicate that Staphylococcus is consistently found in chronic wounds, with over half of these wounds also colonized by P. aeruginosa [4]. Correspondingly, MRSA and P. aeruginosa were chosen as representative Gram-positive and Gram-negative bacteria for further evaluating the antimicrobial property of P-CDs@EPCS. The antibacterial efficacy of P-CDs@EPCS against MRSA and P. aeruginosa was evaluated by comparing bacterial growth inhibition among P-CDs, EPCS, and P-CDs@EPCS (Fig. 3A and B). P-CDs@EPCS exhibited the strongest antibacterial activity with the lowest OD₆₀₀, P-CDs exhibited close antibacterial activity to P-CDs@EPCS. In contrast, EPCS alone showed some level of inhibition. These results indicate that P-CDs primarily contribute to the antimicrobial activity, and EPCS synergistically enhances the antibacterial effect of P-CDs@EPCS. Furthermore, the temporal relationship between the release kinetics and antibacterial efficacy was systematically evaluated, incorporating a diffusion-swelling synergistic release mechanism [56]. Co-culture of cumulative drug release solutions (collected at specified time points) with two standard bacterial strains for 12 h revealed a time-dependent improvement in bacterial inhibition that coincided with the gradual accumulation of P-CDs (Fig. 3C). Maximum inhibition rates of 34.5 % (MRSA) and 33.4 % (P. aeruginosa) against both bacterial strains were achieved during the peak release period, further confirming the sustained release profile of P-CDs capable of maintaining effective therapeutic concentrations in wound environments to significantly suppress bacterial proliferation. The bacterial counts of treated P-CDs, EPCS, and P-CDs@EPCS were assessed on agar plates (Fig. 3D). Corresponding quantitative colony counts (Fig. 3E and F) revealed bacterial survival rates of 44.4 % and 43.2 % for the two strains treated with EPCS, highlighting its limited standalone efficacy. The incorporation of P-CDs significantly enhanced the antibacterial efficiency of P-CDs@EPCS, reducing survival rates to 3.3 % and 15.8 %, respectively, attributable to the intrinsic antimicrobial properties of P-CDs and their synergistic interaction within the composite. SEM images (Fig. 3G) revealed intact smooth membranes in untreated bacteria, whereas P-CDs and P-CDs@EPCS induced pronounced membrane wrinkling, rupture, and cytoplasmic leakage. Collectively, these findings demonstrate that P-CDs@EPCS effectively suppresses bacterial proliferation by integrating the bioactivity of P-CDs with the structural effects of the hydrogel matrix. Moreover, P-CDs@EPCS exhibited sustained antibacterial efficacy, positioning it as a promising candidate for infection management applications.

Fig. 3.

The antibacterial capacity of the P-CDs@EPCS. Bacteriostasis curves at 12 h (A) MRSA and (B) P. aeruginosa. (C) Bacterial growth inhibition percentage of MRSA and P. aeruginosa on P-CD@EPCS within 10 days. (D) Macroscopic images of growth of MRSA and P. aeruginosa on broth agar plates after co-culturing with saline (control), P-CDs, EPCS, and P-CDs@EPCS 24 h at 37 °C. (E–F) Statistical data of the colony densities of MRSA (E) and P. aeruginosa (F) from (D). (G)The SEM images of co-cultured MRSA and P. aeruginosa on saline (control), P-CDs, EPCS and P-CDs@EPCS after 8 h. Data in E, F were presented as mean ± SD (n = 3), ∗∗P < 0.01, ∗∗∗P < 0.001.

Bacterial biofilm formation significantly contributes to bacterial harm [57]. The effect of P-CDs@EPCS on mature biofilms was investigated using SYTO 9/PI staining. Untreated biofilms showed bright green fluorescence with little red signal (Fig. 4A and B). In contrast, treatments with P-CDs, EPCS, and P-CDs@EPCS displayed varying levels of red fluorescence, indicating different degrees of bacterial death and biofilm disruption. Compared to the relatively low red fluorescence observed in EPCS treated bacteria, those treated with P-CDs and P-CDs@EPCS demonstrated strong red fluorescence, confirming the high effectiveness of P-CDs and P-CDs@EPCS against both planktonic and colonized bacteria. Crystal violet quantification using standard bacterial strains and clinically isolated P. aeruginosa strains further verified the significant biofilm eradication capabilities of P-CDs and P-CDs@EPCS (Fig. 4C and D, and Fig. S5), consistent with SYTO 9/PI results. The similar degrees of biofilm eradication of P-CDs and P-CDs@EPCS suggest that P-CDs play a primary role in the antibacterial effect of P-CDs@EPCS.

Fig. 4.

Antimicrobial mechanism of P-CDs@EPCS using MRSA and P. aeruginosa as model with the treatment of saline (control), P-CDs, EPCS, and P-CDs@EPCS. (A–B) CLSM images of Live/Dead-stained bacterial biofilms of MRSA (A) and P. aeruginosa (B) (Scale bar = 100 μm). (C–D) Crystal violet staining images (insert) and the survival rate of the elimination of biofilm formation by MRSA (C) and P. aeruginosa (D). (E–F) PI staining degrees for membrane integrity assessment of MRSA (E) and P. aeruginosa (F). (G–H) Measurement of bacterial extracellular protein levels of MRSA (G) and P. aeruginosa (H). (I–J) ATP content in of MRSA (I) and P. aeruginosa (J). Data were presented as mean ± SD (n = 3), ns P > 0.05, ∗P < 0.05, ∗∗P < 0.01, and ∗∗∗P < 0.001. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

The results presented in Fig. 3, Fig. 4C, and D illustrated that the primary antibacterial activity was originated from P-CDs, while EPCS mainly served as a carrier with moderate antimicrobial properties. The antibacterial mechanism was investigated mainly based on P-CDs. The positively charged surface of P-CDs (Zeta potential: +5.86 mV, Fig. S6) exhibited enhanced efficacy against Gram-positive (G⁺) bacteria, likely due to electrostatic interactions between the positively charged P-CDs and the negatively charged bacterial membranes. Additionally, the PI assay showed a significant increase in fluorescence in treated bacteria compared to untreated bacteria (Fig. 4E and F), indicating enhanced membrane permeability. Protein quantification (Fig. 4G and H) and ATP measurements (Fig. 4I and J) revealed that treatment with P-CDs and P-CDs@EPCS resulted in increased leakage of both protein and ATP concentrations into the culture medium, consistent with the compromised membrane integrity observed in SEM images (Fig. 3G). DCFH-DA assays detected substantial ROS generation in bacteria treated with P-CDs, EPCS, and P-CDs@EPCS, as shown in Fig. S7. The results indicated that MRSA and P. aeruginosa treated with P-CDs and P-CDs@EPCS exhibited strong green fluorescence intensities, whereas untreated or EPCS-treated bacteria displayed weak green signals, suggesting the generation of ROS in bacteria treated with P-CDs or P-CDs@EPCS. In contrast, co-incubation of P-CDs with mammalian cells resulted in minimal green fluorescence intensity, showing significant differences compared to H₂O₂-treated cells (Fig. S8), suggesting the selective generation of ROS by P-CDs in prokaryotes. These findings demonstrate that P-CDs within P-CDs@EPCS exhibit antibacterial effects through structural membrane damage, cellular content leakage, and ROS generation [58].

3.4. Biocompatibility and cellular function of P-CDs@EPCS

The biocompatibility of P-CDs@EPCS was evaluated using CCK-8 assay. As shown in Fig. 5A and B, HUVEC and L929 cells cultured with P-CDs, EPCS extracts, or P-CDs@EPCS extracts for 1–2 days showed no significant proliferation inhibition. After 24 h, HUVEC and L929 cells treated with P-CDs exhibited cell viabilities of 82.45 % and 77.37 %, respectively. When treated by P-CDs@EPCS, HUVEC and L929 cells maintained higher viabilities at 92.67 % and 90.27 %, respectively. After 48 h, the viabilities of HUVEC and L929 cells treated with P-CDs were 79.03 % and 75.67 %, while P-CDs@EPCS still demonstrated excellent viabilities of HUVEC and L929 of 91.91 % and 86.67 %, respectively. Compared to the significant destruction of red blood cells treated with water, P-CDs, EPCS, and P-CDs@EPCS exhibited favorable hemocompatibility, with hemolysis rates below 5 % (Fig. 5C). Live/dead staining further confirmed high cell viability in the presence of P-CDs, EPCS, and P-CDs@EPCS (Fig. 5D), consistent with the CCK-8 results [52]. Moreover, P-CDs@EPCS were subcutaneously implanted in the dorsum of rats to assess their in vivo biocompatibility [59]. H&E staining of heart, liver, spleen, lungs and kidneys after 10 days of treatment revealed no inflammatory cell infiltration or significant histological differences among P-CDs@EPCS and control groups (Fig. S9), confirming the excellent systemic biocompatibility [60].

Fig. 5.

In vitro cytotoxicity and effects of P-CDs@EPCS. Viability of (A) HUVEC cells and (B) L929 cells treated with PBS (control), P-CDs, EPCS, and P-CDs@EPCS at 24 and 48 h. (C) Hemolysis rate and interacting images (insert) of red cells with water (control), P-CDs, EPCS, and P-CDs@EPCS. (D) Fluorescence images of Live/Dead staining of HUVEC and L929 cells on PBS (control), P-CDs, EPCS and P-CDs@EPCS (Scale bar = 100 μm). Data in A, B were presented as mean ± SD (n = 3), ns P > 0.05, and ∗P < 0.05. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

The inflammatory phase of wound healing is primarily mediated by macrophages, which clear damaged tissue, enhance host defense, and orchestrate the transition to tissue regeneration. In infected wounds, impaired phenotypic switching from pro-inflammatory M1 to reparative M2 macrophages contributes to destructive microenvironments and chronic wound formation [61]. To evaluate the immunomodulatory capability of P-CDs@EPCS, this study established an inflammatory microenvironment model using LPS/IFN-γ stimulation in RAW264.7 cells (control group). Flow cytometry results showed that the proportion of CD206⁺ M2 macrophages significantly increased in the P-CDs (16.5 %) and P-CDs@EPCS (17.7 %) treatment groups compared to the control group (2.28 %) (Fig. 6A), indicating that both treatments effectively promoted the polarization of macrophages from pro-inflammatory M1 to anti-inflammatory M2 type. Given that this polarization effect appeared to originate mainly from the P-CDs component, flow cytometry experiments were performed with P-CDs alone at graded concentrations (20–100 μg/mL). As shown in Fig. S10, P-CDs promoted a concentration-regulated effect on the polarization of macrophages. With the increased concentrations of P-CDs, P-CDs treatment generally shifted macrophage polarization from the pro-inflammatory M1 phenotype toward the anti-inflammatory/reparative M2 phenotype from 20 to 60 μg/mL. And then, the effect of P-CDs on the polarization of macrophages from M1 to M2 was gently decreased from 60 to 100 μg/mL. This result implies that P-CDs consistently produced robust immunoregulatory activity in combination with potent antibacterial properties.

Fig. 6.

Anti-inflammatory effects of hydrogels as well as promotion of macrophage polarization from M1 to M2 phenotype. (A) Flow cytometry was used to analyze the expression of CD206 and CD86 in RAW264.7 cells under LPS + IFN-γ-induced inflammation (control), following treatment with P-CDs, EPCS and P-CDs@EPCS. (B–D) ELISA analysis of the levels of TNF-α (B), IL-6 (C), IL-10 (D), and IL-4 (E) in RAW264.7 cells with LPS + IFN-γ treatment (control), LPS + IFN-γ treatment and P-CDs, LPS + IFN-γ treatment and EPCS and LPS + IFN-γ treatment and P-CDs@EPCS. (F) HUVEC migration was analyzed via Transwell assays using conditioned media (blank: PBS; control: LPS + IFN-γ; experimental: control group treated by P-CDs, EPCS, and P-CDs@EPCS, respectively) in lower chambers, by crystal violet staining, Scale bar = 100 μm. (G) Migrated cell quantification from F. Data were presented as mean ± SD (n = 3), ns P > 0.05, ∗P < 0.05, ∗∗P < 0.01, and ∗∗∗P < 0.001. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

TNF-α and IL-6 are classical pro-inflammatory mediators that sustain or amplify the inflammatory phase; reducing their levels can limit excessive tissue damage [62]. Conversely, IL-10 is a potent anti-inflammatory cytokine that not only suppresses excessive inflammation but also indirectly promote collagen synthesis and extracellular matrix remodeling by modulating fibroblast activity [63]. IL-4 is a well-recognized pro-angiogenic factor that can act directly on endothelial cells to accelerate neovascularization [64]. Furthermore, the secretion of key inflammatory cytokines in the RAW264.7 cell supernatant was assessed through ELISA (Fig. 6B–E), confirming the significant regulatory effects of the different treatment groups on cytokine levels and highlighting the immunomodulatory function of P-CDs@EPCS. Although the EPCS group exhibited moderate reductions in the pro-inflammatory cytokines of TNF-α and IL-6 compared to those of the control group, the EPCS group showed no significant variations in the anti-inflammatory cytokines of IL-10 and IL-4 compared to the control group. However, both the P-CDs and P-CDs@EPCS groups displayed markedly reduced levels of pro-inflammatory cytokines of TNF-α and IL-6 alongside elevated expression of anti-inflammatory cytokines of IL-10 and IL-4, as illustrated in Fig. 6B–E. Specifically, compared to the control group, the P-CDs@EPCS group reduced TNF-α and IL-6 expression by 2.4-fold and 1.8-fold, respectively. In contrast, IL-10 and IL-4 levels elevated 2.8-fold and 3.7-fold compared to the control group. These findings indicate that P-CDs and P-CDs@EPCS have a good ability to inhibit pro-inflammatory polarization and promote anti-inflammatory polarization targeting macrophages. EPCS only weakly inhibits inflammatory polarization, suggesting that the immunomodulatory properties of P-CDs@EPCS are primarily dominated by the encapsulated P-CDs with the carrier-assisted role of EPCS.

Infected wound microenvironments impose heightened metabolic demands on resident tissue cells, necessitating increased nutrient and oxygen supply to support reparative processes. Neovascularization serves as a critical pathway for delivering oxygenated, nutrient-rich circulation to the injury site, establishing the biochemical and biophysical conditions necessary for cellular repopulation, matrix remodeling, and functional tissue restoration. The directed polarization of macrophages can lead to distinct anti-inflammatory and tissue repair functions through the secretion of key cytokines and growth factors [65]. The macrophage-mediated endothelial responses were systematically evaluated using conditioned media from differentially treated RAW264.7 cells, as detailed in Fig. 6F and G. Transwell migration assays demonstrated a dose-dependent enhancement in endothelial chemotaxis, with the P-CDs group showing a 1.8-fold increase and the P-CDs@EPCS group achieving a 2.3-fold elevation in HUVEC recruitment relative to the blank and control group, both reaching statistical significance (p < 0.001). These results confirm the ability of P-CDs@EPCS to induce macrophages to polarize into repair promoting phenotypes, thereby potentiating the migratory competence of endothelial cells damaged by infectious inflammation.

Furthermore, the pro-migratory capacity serves as a critical determinant of wound healing efficacy [66]. To systematically evaluate the migration potential of P-CDs@EPCS, scratch assays were conducted on HUVEC and L929 cells (Fig. 7A–D). Quantitative analysis revealed that P-CDs@EPCS significantly enhanced the relative migration rates of both cell types, reaching 56.28 % and 64.00 %, which are 1.8 times and 1.6 times higher than the control group, respectively. This enhancement treated by P-CDs@EPCS was significantly greater than that observed with the individual components, P-CDs (38.65 %–41.27 %) or EPCS (49.57 %–58.48 %). The superior migration-promoting performance of P-CDs@EPCS may arise from the synergistic effect between the physicochemical properties of the EPCS hydrogel and the sustained release effect of P-CDs, thus enhancing the overall migration-promoting capability of the composite hydrogel beyond that of the individual components. Transwell migration assays with L929 fibroblasts demonstrate the superior permeability of the P-CDs@EPCS group, achieving the migration of 378 ± 8 cells after incubation at 37 °C for several hours. This performance notably exceeds that of the controls, including P-CDs (206 ± 10 cells), EPCS (363 ± 5 cells), and untreated controls (68 ± 3 cells), as illustrated in Fig. 7E and F. Importantly, all experimental groups exhibited varying degrees of enhanced motility in fibroblasts and endothelial cells, attributed to the excellent biocompatibility of the EPCS and its porous structure and mechanical strength, which support cell adhesion and migration. Moreover, P-CDs alone also display some reparative effects, thus, P-CDs@EPCS significantly elevates cell migration ability, offering crucial support for wound healing.

Fig. 7.

In vitro effects of P-CDs@EPCS on cell functions. (A, C) Scratch assay of HUVEC (A) and L929 (B) cells incubating in complete medium with saline (control), P-CDs, EPCS, and P-CDs@EPCS for 24 h, respectively (Scale bar = 100 μm). (B, D) Corresponding quantitative analysis of wound closure rates from A and B. (E) Transwell migration assay of L929 cells treated with saline (control), P-CDs, EPCS, and P-CDs@EPCS by crystal violet staining, (Scale bar = 100 μm). (F) Corresponding quantitative analysis of migrated cell numbers from E. Data in B, D, and F were presented as mean ± SD (n = 3), ns P > 0.05, ∗P < 0.05, and ∗∗∗P < 0.001. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

3.5. P-CDs@EPCS accelerated P. aeruginosa-infected wounds healing in vivo

An 8 mm full-thickness skin wound rat model infected with P. aeruginosa was applied to further evaluate the in vivo wound healing capabilities of P-CDs@EPCS (Fig. 8A). After a one-week acclimatization period (designated as Day 0), the infected skin wounds were assigned to four groups: Control (normal saline), P-CDs alone, EPCS alone, and P-CDs@EPCS. Treatments were applied and observed macroscopically on Day 0, 4, 8, and 10 (Fig. 8B and C). On day 0, all groups exhibited typical signs of infection, including redness and tissue edema. Over time, differences in the healing process began to emerge among the groups. By day 4, the control group showed significant purulent accumulation and partial scab formation; however, there were no significant differences in accelerated healing between the treatment groups. By day 8, the wounds in the control group were dry, with no obvious purulent accumulation, and the scabs had thickened. The wound areas of the P-CDs (19.94 %), EPCS (19.53 %), and P-CDs@EPCS groups (14.69 %) were significantly smaller than those of the control group (48.50 %), with the P-CDs@EPCS group showing the most pronounced reduction (Fig. 8B, C, and 8D), likely due to its antibacterial and anti-inflammatory properties. By day 10, the control group still exhibited a significant wound area (27.97 %), while most wounds in the treatment groups had significantly shrunk, and redness had diminished. Notably, all treatment groups demonstrated better healing effects compared to the control group, with the P-CDs@EPCS group having the smallest remaining wound area at just 4.3 %. The performance of the EPCS group (8.67 %) was superior to that of the P-CDs group (15.53 %), likely due to the supportive physical properties of the EPCS hydrogel facilitating wound repair [67,68]. These results indicate that P-CDs@EPCS exhibit superior biocompatibility and physicochemical properties, thereby effectively promoting the healing of infected wounds.

Fig. 8.

P-CDs@EPCS dressings accelerated P. aeruginosa-infected wounds healing in vivo. (A) Schematic diagram of the animal experimental procedure: mice of the four groups were infected with P. aeruginosa on dorsal wounds and respectively treated with saline (control), P-CDs, EPCS, and P-CDs@EPCS dressings according to the schedule shown. (B) Representative wound photographs on days 0, 4, 8, and 10 after infection of the four groups. (C) Simulated wound progression stack images at the same time points for each group. (D) Residual wound area (%) at days 4, 8, and 10. (E–F) H&E and Masson staining images of infected skin tissue at day 10 post-treatment; The black, blue, and red arrows indicate the epithelium, inflammatory cells, and hair follicles, respectively. (Scale bar = 500 μm). Data in D were presented as mean ± SD (n = 3), ∗∗∗P < 0.001. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

Tissue specimens from all experimental groups were collected on day 10 and analyzed histomorphologically using H&E staining and Masson’s Trichrome staining. H&E-stained sections (Fig. 8E) showed significant inflammatory cell infiltration at the wound periphery in the control group on day 10. In contrast, both the P-CDs and EPCS groups exhibited diminished inflammatory infiltration and relatively intact epidermal architectures. Notably, the P-CDs@EPCS group demonstrated the least inflammatory infiltration coupled with the most organized epidermal reconstruction, characterized by a significant presence of regenerated hair follicles and sebaceous glands. Masson’s trichrome staining (Fig. 8F) further corroborated these findings, showing reduced collagen voids and denser fiber arrangement in the P-CDs and EPCS groups compared to the control group. The P-CDs@EPCS group demonstrated superior tissue regeneration, as evidenced by a higher density of hair follicles, a structurally intact epidermis, and orderly collagen fiber deposition. These histological outcomes collectively underscore the therapeutic efficacy of P-CDs@EPCS in actively promoting healing in infected wounds.

Immunohistochemical (IHC) analyses of wound tissues were carried out to further elucidate the wound healing. Fig. 9A and B demonstrated that the control group exhibited the most pronounced CD86 expression alongside diminished CD206 levels in regenerated skin, indicative of a persistent pro-inflammatory microenvironment. In contrast, P-CDs treatment ameliorated inflammatory responses and oxidative stress, resulting in marked reductions in CD86 expression and enhanced CD206 expression in the P-CDs group (Fig. 9E and F). These findings align with the prior cellular experiments (Fig. 6), revealing that P-CDs effectively attenuated polarization toward pro-inflammatory M1 phenotypes while promoting partial polarization toward anti-inflammatory M2 phenotypes, thereby facilitating the transition from inflammatory to proliferative phases in wound healing. Furthermore, compared to the control group, the EPCS treatment group exhibited mild downregulation of CD86 and upregulation of CD206, confirming certain anti-inflammatory activity. The synergistic effect of EPCS and P-CDs resulted in P-CDs@EPCS demonstrating remarkable anti-inflammatory efficacy, characterized by the lowest CD86/CD206 ratio, indicating a complete resolution of the inflammatory phase in skin tissue. Parallel IHC evaluation of VEGF and COL-1 expression (Fig. 9C and D) revealed upregulated levels across all treatment groups compared to the control group (Fig. 9G and H). The COL-1 level of the EPCS group was slightly higher than that of the P-CDs group, which is in accord with the healing result (Fig. 8B), suggesting the importance of supportive physical properties of dressings for wounds [67,68]. Moreover, P-CDs@EPCS exhibited the highest expression intensities of VEGF and COL-1 (Fig. 9G and H). These results collectively underscore that P-CDs@EPCS exhibits spatiotemporal precision in therapeutic agent delivery, with profound therapeutic implications for orchestrating inflammatory microenvironment modulation, collagen neogenesis, and angiogenic activation to accelerate repair in infected wounds.

Fig. 9.

Immunohistochemical analysis. (A–D) Immunohistochemical images of CD86 (A), CD206 (B), VEGF (C), and COL-1 (D) in tissue samples from P. aeruginosa-infected wounds at day 10 post-treatment with saline (control), P-CDs, EPCS, and P-CDs@EPCS dressings, respectively (Scale bar = 50 μm). (E–H) Quantification of CD86 (E), CD206 (F), VEGF (G), and COL-1 (H) expression from the corresponding image. Data in E-H were presented as mean ± SD (n = 3), ∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001.

3.6. Limitation

While, there are still some limitations. The structural simulation of P-CDs with the group distribution has not yet been clarified, and the binding mode between P-CDs and hydrogels needs to be further clarified; the in vivo long-term safety of P-CDs@EPCS needs to be evaluated; the present main focus is on showcasing the phenomenological results of P-CDs@EPCS, the molecular mechanisms underlying its antibacterial and immune regulatory effects need to be further clarified; Animal experiments only use rats for modeling, and their effectiveness in other animal models and wound models with underlying diseases needs further validation.

4. Conclusion

In summary, P-CDs with antibacterial and immunoregulatory properties were synthesized from PHMG and CA. CMCS and EPL were crosslinked with PVA through freeze-thaw cycles to fabricate EPCS, which effectively encapsulated P-CDs to form P-CDs@EPCS. P-CDs@EPCS retains the sustained release characteristics of EPCS while preserving the antibacterial and immunomodulatory properties of P-CDs. The released P-CDs demonstrated significant synergistic antibacterial effects against clinical isolates of resistant P. aeruginosa and MRSA. The biological activity and immunomodulatory properties of P-CDs@EPCS could effectively enhance the activity of inflammatory induced cells, thereby promoting angiogenesis and fibroblast activity. In a P. aeruginosa-infected wound model, wound progression, histological analysis, and immunohistochemistry confirmed that P-CDs@EPCS accelerated angiogenesis, promoted granulation tissue formation, enhanced collagen deposition, and drove epidermal regeneration. This multifunctional P-CDs@EPCS exhibit significant potential in the field of infected wound healing.

CRediT authorship contribution statement

Shuangying Yang: Writing – original draft, Methodology, Formal analysis, Data curation, Conceptualization. Youjia Wu: Supervision, Funding acquisition, Formal analysis. Xiaoyuan Huang: Formal analysis, Data curation. Xiaoyu Li: Formal analysis, Data curation. Yao Wang: Methodology, Investigation. Jianfeng Pan: Validation, Resources, Funding acquisition. Zhengjun Huang: Resources. Shaohuang Weng: Writing – review & editing, Writing – original draft, Supervision, Project administration, Funding acquisition, Conceptualization. Jianyong Huang: Writing – review & editing, Supervision, Project administration, Funding acquisition.

Declaration of competing interest

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

Appendix A. Supplementary data

The following is the Supplementary data to this article:

Data availability

Data will be made available on request.