Carbomol hydrogels integrating chlorogenic acid-loaded exosomes promote diabetic wound healing through antibacterial and immunomodulatory activities

Abstract

Diabetic wounds represent a significant global clinical challenge, exacerbated by persistent bacterial infection and a dysregulated inflammatory microenvironment. Effective strategies for diabetic wounds are absent. Here, we found that NOD-like receptor thermal protein domain associated protein 3 (NLRP3) is an important regulating molecule associated with inflammatory microenvironment. We engineered Carbomol hydrogels integrating chlorogenic acid-loaded exosomes (Exo-CA@CB) to address both infection and NLRP3-mediated inflammation. The Carbomol hydrogel (CB) matrix provides biocompatibility, adhesion, self-healing properties, and sustained release, while the encapsulated Exo-CA deliver bioactive cargo. In vitro studies demonstrated the potent antibacterial activity of Exo-CA@CB against common wound pathogens. Crucially, Exo-CA@CB effectively promoted macrophage polarization towards the regenerative M2 phenotype. In the infected diabetic wound model, topical application of Exo-CA@CB hydrogels significantly accelerated healing. This study validates a new approach, initiated by single-cell analysis, for developing Exo-CA@CB hydrogels as a promising platform to manage infected diabetic wounds by concurrently tackling infection and NLRP3-driven inflammation.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13062-026-00818-z.

Introduction

Diabetes mellitus is a growing health problem in the aging population, adding immense socioeconomic burdens on global healthcare systems [1]. Characterized by impaired tissue repair and chronic non-healing ulcers, diabetic wounds is one of the most debilitating complications of diabetes mellitus [2, 3]. While the chronic diabetic wound is the convergent result of various pathophysiological abnormalities, such as vascular insufficiency, peripheral neuropathy, increased oxidative stress, and dysregulated immunity, macrophage-driven chronic inflammation is regarded as the core pathology leading to non-healing of diabetic wounds [4]. Notably, the inability of macrophages to transit from a pro-inflammatory M1 phenotype to a reparative M2 phenotype, which leads to excessive cytokine production and aggravated tissue damage, substantially compromises the healing of diabetic wounds [5].

Central to this inflammatory dysregulation is the aberrant activation of NLRP3 inflammasome, a key innate immune sensor responsive to sterile danger signals and microbial cues [6]. NLRP3 inflammasome activation triggers the maturation and secretion of IL-1β and IL-18, which are key pro-inflammatory cytokines that drive macrophages toward the M1 phenotype. The release of these cytokines creates a positive feedback loop that sustains the inflammatory microenvironment conducive to M1 polarization [7]. Furthermore, NLRP3-mediated caspase-1 activation can directly regulate metabolic reprogramming in macrophages, promoting glycolysis and thereby supporting the functional characteristics of M1 macrophages [8].

In the diabetic microenvironment, hyperglycemia-associated metabolic stress persistently activates NLRP3 which amplifies inflammatory cascades through excessive release of IL-1β and TNF [9]. Other than promoting chronic inflammation, NLRP3 hyperactivation also disrupts macrophage phenotypic plasticity, resulting in failed M1-to-M2 transition which is a prerequisite for wound healing [10].

Pharmacological inhibition of NLRP3, such as using disulfiram and metformin, has shown some therapeutic effects in preclinical studies but limited efficacy in clinically complex wounds where microbial infection is intertwined with immune dysregulation [11, 12]. Indeed, nearly a half of diabetic wounds are complicated by bacterial colonization, particularly methicillin-resistant Staphylococcus aureus (MRSA), which often further amplifies NLRP3 activities and inflammation [13, 14]. These observations underscore the need for therapeutic strategies that can simultaneously target bacterial infection and inflammasome-driven immune dysfunction.

Chlorogenic acid (CA), a plant-derived polyphenol, has emerged as a compelling therapeutic candidate due to its dual antibacterial and NLRP3-inhibitory properties [15, 16]. Yet, the intrinsic instability and low bioavailability of CA substantially limit its in vivo performance [17]. An efficient delivery platform is needed to protect CA from degradation, enhance its bioactivity in the inflammatory niche, and ultimately improve clinical efficacy. Macrophage-derived exosomes may be a unique candidate platform. Owing to their natural biocompatibility, and potent immunomodulatory properties, exosomes are able to encapsulate and deliver therapeutic cargo to inflamed tissues [18]. Moreover, the membrane architecture of exosomes ensure uptake by macrophages and can protect sensitive molecules from enzymatic degradation of CA [18].

Here we engineered CA-loaded macrophage-derived exosomes (Exo-CA) as “biological missiles” to enhance CA stability, boost CA accumulation within macrophages, and synergize its anti-inflammatory effects with the endogenous regulatory functions of exosomal cargo. We found Exo-CA effectively redirected macrophage polarization from an M1-dominant to an M2-regenerative phenotype to directly counteract NLRP3-driven pathology. To achieve localized and sustained therapeutic delivery, Exo-CA was further embedded into a Carbomol hydrogel matrix to engineer Exo-CA@CB. Due to excellent tissue adhesion and controlled-release capabilities, Carbomol hydrogels enable the hydrogel to have prolonged retention of exosomes at the wound surface and continuous therapeutic release in situ [19].

By integrating exosomal delivery with a bioadhesive hydrogel platform, the engineered Exo-CA@CB is able to break the self-perpetuating cycle of infection and inflammation in diabetic wounds. This dual-action, host-directed strategy simultaneously eradicates pathogenic bacteria and restores immune balance to create a microenvironment conducive to tissue repair. The combined mechanistic rationale and biomaterial design position Exo-CA@CB as a promising next-generation therapeutic platform for the treatment of infected diabetic wounds (Fig. 1).

Fig. 1.

Schematic diagram of Carbomol hydrogels integrating chlorogenic acid-loaded exosomes for promoting diabetic wound healing through antibacterial and immunomodulatory activities

Materials and methods

Cell culture

The RAW 264.7 murine macrophage cell line (American Type Culture Collection, ATCC) was cultured in Dulbecco’s Modified Eagle Medium (DMEM; Hyclone, USA) supplemented with 10% fetal bovine serum (FBS; Gibco, USA) and 1% penicillin streptomycin (PS). Cells were maintained in a humidified incubator at 37 °C with 5% CO₂.

Preparation and characterization of CA loaded exosomes (Exo CA)

CA was added to sub confluent RAW 264.7 culture (4 × 10⁵ cells) at 1 mg/mL and incubated for 24 h. The supernatant was collected and sequentially centrifuged (300 g, 10 min; 2,000 g, 10 min; 10,000 g, 30 min) to remove cells and debris. Exosomes were then isolated by two rounds of ultracentrifugation (110,000 g, 3 h, 4 °C; Beckman Coulter, USA) and stored at -80 °C until use. Size distribution, concentration and morphology of Exo CA were analyzed by nanoparticle tracking analysis (NTA) and transmission electron microscopy (TEM). Exosomal markers were measured using western blotting.

Isolated exosomes are stored short-term at 4 °C before being used for subsequent material preparation, and repeated freeze‑thaw cycles of exosomes are strictly avoided. If storage exceeds 7 days, fresh exosomes are newly isolated. Exosomes are prone to loss of bioactivity upon repeated freeze-thawing; therefore, we isolate and use exosomes freshly on demand.

As described previously, CA was loaded into exosomes. The mixture was then ultracentrifuged under 100,000×g for 1 h at 4 °C to separate CA-loaded exosomes from free, unencapsulated CA. The supernatant containing free CA was carefully collected and analyzed by UV-Vis, and the pellet (CA-loaded exosomes) was washed and resuspended in PBS.

Encapsulation efficiency (EE%) and loading efficiency (LE%) were calculated as follows:

For release kinetics, CA-loaded exosomes were dispersed in PBS and placed into a dialysis bag (MWCO 3.5 kDa). The dialysis system was incubated at 37 °C under gentle shaking. At predetermined time intervals, aliquots of the external medium were withdrawn, and the concentration of released CA was determined by UV-Vis. The cumulative release percentage was plotted against time to characterize the release profile.

Preparation and characterization of hydrogels

The 1 wt% Carbomol (CB) gel was prepared by dissolving Carbomol 940 (Macklin Reagent, China) in ddH₂O (1 g per 100 mL) under stirring at 60 °C for 30 min. The pH was adjusted to 6.5–7.5 with 0.25% triethanolamine (Adamas, China) and stirred for an additional 10 min. Exo CA powder was incorporated into the CB gel under stirring for 30 min, and the mixture was left overnight at room temperature to form the composite Exo CA@CB hydrogel. The crude hydrogel was washed 3 times with ddH₂O to remove unbound material and stored at 4 °C.

For scanning electron microscopy (SEM), hydrogels were frozen at -80 °C for 24 h, lyophilized (0.05 mbar, 24 h), and imaged using a Hitachi SU 3500 SEM at 5 kV and 10 µA.

Cell biocompatibility

Viability of RAW 264.7 cells treated with control, free CA or Exo CA was evaluated with a CCK 8 kit (Solarbio, China). After 24–48 h of treatment, cells were washed and incubated with 10% CCK 8 solution for 4 h. Absorbance at 450 nm was measured on a microplate reader (BioTek Synergy Neo2, USA).

Single cell RNA seq analysis

Public scRNA seq data (GEO accession GSE154557) [20]were processed with Seurat (v4.1.0, USA). Quality control assessed gene detection per cell, total molecular counts, and mitochondrial genome proportion, excluding genes detected in < 3 cells and cells with < 200 detected genes or high mitochondrial content. Counts were normalized to 10,000 UMIs per cell, log transformed and scaled. Batch effects were removed using PCA anchor points, with successful merging indicated by overlapping clusters in PC1-PC2 space. Dimensionality reduction used the top variable genes, with principal component selection guided by ElbowPlot inflection points and JackStrawPlot p-value significance (top 7 PCs). UMAP embedding visualized cell clusters, followed by clustering with the Find Clusters function and subsequent annotation based on top marker gene heatmaps. Cell proportion distributions across groups and selected gene expression patterns were also analyzed to characterize cellular heterogeneity.

Immunofluorescence (IF) staining

Cells or tissue sections were fixed with 4% paraformaldehyde (15 min), blocked with 4% BSA (1 h), and incubated overnight at 4 °C with primary antibodies against NLRP3 (1:400) and F actin (1:1000). After washing, samples were incubated with fluorescent secondary antibodies (Servicebio, China) for 1.5 h, counterstained with DAPI (15 min), and imaged with a confocal microscope (Olympus FV3000, Japan).

Western blotting

Cells were lysed in RIPA buffer (Beyotime, China) containing protease and phosphatase inhibitors. Protein concentration was determined by BCA assay (Beyotime). Samples (50 µg per lane) were separated by SDS PAGE, transferred to PVDF membranes (Millipore, USA), blocked with StartingBlock (Thermo, USA; 15 min), and incubated overnight at 4 °C with primary antibodies: Alix (1:1000), TSG101 (1:1000) and Calnexin (1:1000). HRP conjugated secondary antibodies (Solarbio, China) were applied for 2 h, and bands were visualized with Image Lab 3.0.

Flow cytometric

RAW 264.7 cells were treated for 48 h with control, CA or Exo CA, detached with Accutase (4 °C), washed and stained with fluorochrome conjugated anti mouse CD86 and CD206 antibodies (BioLegend, USA; 30 min, 4 °C, dark). Analysis was performed on a CytoFLEX flow cytometer (Beckman Coulter, USA).

In vitro antibacterial assays

Methicillin resistant Staphylococcus aureus (MRSA) was grown overnight in tryptic soy broth (TSB). For colony forming unit (CFU) assays, test samples were immersed in 1 mL of bacterial suspension (1 × 10⁶ CFU/mL) at 37 °C for 2 h. Supernatants were serially diluted, plated on blood agar, incubated overnight at 37 °C, and colonies were counted.

Bacterial viability was also assessed using the LIVE/DEAD BacLight™ kit (Invitrogen, USA). Diluted bacterial suspensions were stained with SYTO 9 and propidium iodide (PI) for 60 min in the dark, and fluorescence intensity was quantified semi quantitatively with ImageJ (v1.8.0, NIH, USA). For biofilm imaging, biofilms grown on discs were stained and observed by confocal microscopy.

Intracellular bacterial clearance was evaluated by co culturing macrophages with GFP tagged MRSA (MOI = 10) for 24 h. Internalized bacteria were visualized by confocal microscopy.

Rat model of diabetic wound and treatment

Female Sprague Dawley rats (200 g; purchased from the Laboratory Animal Center at The First Affiliated Hospital of Zhejiang University) were housed under specific pathogen free (SPF) conditions (22 ± 1 °C, 12 h light/dark cycle). All animal experiments were approved by the Ethics Committee of the First Affiliated Hospital of Zhejiang University (SYDW20221065) and adhered strictly to ethical guidelines for animal welfare. Diabetes was induced by intraperitoneal injection of streptozotocin (50 mg/kg in citrate buffer). Rats with fasting blood glucose > 16.67 mmol/L 1 week later were regarded as having diabetic mellitus and were used for experiments.

Under pentobarbital anesthesia (40 mg/kg, i.p.), the dorsal skin was shaved, disinfected, and a full thickness circular wound (10 mm diameter) was created. Each wound was inoculated with 10 µL MRSA suspension (1 × 10⁸ CFU/mL). After 48 h, approximately 200 µL of hydrogel was applied to the wound. Wounds were photographed on day 0, 3, 7 and 14. For bacterial analysis, 5 rats per group were randomly euthanized on day 3.

Histological and immunohistochemical analysis

Rats were euthanized on day 2 and 14 post-treatment. Wound specimens were fixed in 4% paraformaldehyde for 48 h, embedded, sectioned (5 μm), and stained with hematoxylin & eosin (H&E), Masson’s trichrome (under polarized light), and Immunofluorescence staining (iNOS, CD206, NLRP3) for Histological analysis.

Statistical analysis

All experiments were repeated at least 3 times. Data are expressed as mean ± SD. Differences were analyzed by one way ANOVA followed by Tukey’s post hoc test (SPSS 26, USA). P < 0.05 was considered statistically significant.

Results and discussion

Single-cell transcriptomics reveals NLRP3 hyperactivation in diabetic wound macrophages

Macrophages play an indispensable role in the orchestration of diabetic wound healing [20]. To delineate the immune landscape underlying diabetic tissue repair, we analyzed public single-cell RNA sequencing (scRNA-seq) datasets of biopsies from human diabetic wounds and normal skin (Fig. 2A). Based on the Human Primary Cell Atlas, unsupervised clustering identified 8 distinct cell populations, including major clusters of macrophages, T cells, and endothelial cells (Fig. 2B). As macrophage plasticity is critical for the transition from inflammation to proliferation, modulating macrophage phenotypes may offer a promising therapeutic avenue [21]. While systemic anti-inflammatory agents, such as dexamethasone, lack specificity and often have limited efficacy, strategies targeting macrophage plasticity may have better outcomes [22]. Echoing our previous findings, quantitative cellular analysis revealed a significant expansion of the macrophage compartment in diabetic wounds as compared to healthy skin tissues (Fig. 2C), suggesting a dominant role of macrophage in pathology of diabetic wound [23]. Given that chronic inflammation in diabetes is primarily driven by NLRP3 inflammasome dysregulation [24], we specifically interrogated NLRP3 expression level and confirmed a marked upregulation of NLRP3 in diabetic wound-associated macrophages (Fig. 2D, E). Findings suggested that NLRP3 inflammasome may be a critical therapeutic target to reverse immune dysfunction in diabetic wounds.

Fig. 2.

Single-cell transcriptomic landscape reveals immune dysregulation and macrophage expansion in diabetic wounds. (A) Schematic illustration of the scRNA-seq workflow comparing biopsies from chronic diabetic wounds and matched normal skin controls. (B) t-SNE visualization of integrated single-cell transcriptomes. Distinct cell clusters were annotated based on the Human Primary Cell Atlas. (C) Quantification of cellular composition showing a significant expansion of the macrophage compartment in diabetic wound tissues relative to controls. (D-E) Feature plots (D) and violin plots (E) demonstrating markedly upregulated expression of NLRP3 inflammasome components in diabetic wound macrophages compared to those in healthy skin (**P < 0.001)

Engineering and characterization of chlorogenic acid-loaded exosomes (Exo-CA)

Chlorogenic acid (CA), a bioactive polyphenol enriched in honeysuckle (Fig. 3A), exhibits potent antioxidant, anti-inflammatory, and antibacterial properties [25]. Clinical transition of CA, however, is compromised by poor solubility and rapid physiological degradation. Molecular docking studies theoretically validated the binding affinity of CA to the NLRP3 protein (Fig. 3B), supporting CA may be used as an inflammasome inhibitor [26].

Fig. 3.

Engineering and physicochemical characterization of Chlorogenic Acid-loaded Exosomes (Exo-CA). (A) Chemical structure of Chlorogenic Acid (CA). (B) Schematic and structural modeling illustrating the potential covalent interaction between CA and the NLRP3 protein, putatively via thiol-reactive moieties. (C) Schematic of Exo-CA preparation via co-incubation and ultracentrifugation. (D) TEM showing typical exosomal cup-shaped morphology. (E) TEM confirming uniform size distribution. (F) Confocal microscopy images showing the cellular uptake of PKH26-labeled Exo-CA (red) by RAW264.7 macrophages (F-actin: green; Nuclei: blue). Scale bar: 80 μm. (G) Western blot analysis validating the enrichment of exosomal markers (TSG101, Alix) and exclusion of the ER marker Calnexin in Exo-CA. (H) CCK-8 viability assay of macrophages treated with PBS, free CA, or Exo-CA for 24 h and 48 h. ns: not significant, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001

To overcome bioavailability barriers, we leveraged macrophage-derived exosomes as a biomimetic delivery system (Exo-CA), capitalizing on their intrinsic tropism for inflammatory sites and low immunogenicity [27]. Exo-CA was fabricated by co-incubating RAW 264.7 macrophages with CA, followed by differential ultracentrifugation. Physicochemical characterization via TEM and NTA confirmed that Exo-CA maintained a typical cup-shaped morphology with a uniform size distribution (Fig. 3C). As shown in Table S1, the encapsulation efficiency of CA was 27.48 ± 5.23%, and the loading efficiency was 11.35 ± 2.57%. Western blotting verified the retention of exosomal identity (enrichment of the markers TSG101 and Alix) and the absence of the ER contaminant Calnexin (Fig. 3E). CA displayed a rapid initial release from exosomes, with over 57.44% released within the first 3 days (Fig. S1), providing sufficiently high drug levels for early antibacterial activity. Thereafter, the release profile shifted to a sustained phase, reaching a cumulative release of approximately 94.06% by day 14. Moreover, uptake assays demonstrated that PKH26-labeled Exo-CA was efficiently internalized by recipient macrophages (Fig. 3D). Importantly, CCK-8 assays confirmed excellent biocompatibility of Exo-CA (Fig. 3F) for subsequent in vivo applications. The drug loading performance of CA-loaded exosomes was evaluated.

Exo-CA reprograms macrophage polarization via NLRP3

Aberrant NLRP3 inflammasome activation acts as a checkpoint that locks macrophages in a pro-inflammatory state. Previous studies have established that NLRP3 modulation, whether through XBP1 signaling or autophagy pathways, is pivotal for M2 polarization [10, 28]. Within this mechanistic framework, we investigated whether Exo-CA can reverse inflammatory polarization. Immunofluorescence analysis demonstrated that LPS stimulation induced enhanced NLRP3 expression in macrophages, which was effectively attenuated by Exo-CA treatment (Fig. 4A, B). The suppression of NLRP3 translated into a phenotypic shift: flow cytometry revealed that Exo-CA significantly downregulated the M1 marker CD86/iNOS while concurrently upregulated M2 marker CD206/Arg-1 (Fig. 4C, D and Fig. S2). This reciprocal modulation indicates that Exo-CA successfully reprogramed macrophages from a pro-inflammatory to a reparative phenotype.

Fig. 4.

Exo-CA inhibits NLRP3 inflammasome activation and reprograms macrophage polarization towards an anti-inflammatory phenotype. (A) Representative immunofluorescence images of NLRP3 (green) in macrophages following treatment with PBS, free CA, or Exo-CA. Nuclei: DAPI (blue) (n = 3). Scale bar: 20 μm. (B) Quantitative analysis of NLRP3 fluorescence intensity. (C) Representative flow cytometry histograms of surface markers CD86 (M1) and CD206 (M2) after 24 h treatment (n = 3). (D) Quantification of CD86⁺ and CD206⁺ macrophage populations, indicating a shift from M1 to M2 polarization (n = 3). ns: not significant, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001

Potent in vitro bactericidal and immunomodulatory activity against MRSA

Methicillin-resistant Staphylococcus aureus (MRSA) is a major pathogen complicates diabetic foot ulcers and often leads to prolonged chronic infection and amputation [29, 30]. To test therapeutic potential of Exo-CA, we used MRSA as a model pathogen.

Exo-CA exhibited strong direct bactericidal activity, as evidenced by significant reductions in colony formation (Fig. 5A, C) and a marked increase in dead/compromised bacteria in Live/Dead staining assays (Fig. 5B, D). Beyond direct killing, effective clearance of intracellular bacteria requires functional macrophages. In a macrophage-MRSA co-culture model, Exo-CA treatment significantly suppressed the survival of intracellular MRSA (Fig. 5E-G). Findings suggest a dual mechanism of action: Exo-CA not only attacks bacteria directly but also restores the phagocytic and bactericidal capacities of host macrophages. This host-directed strategy has advantages over the conventional antibiotics as it can mitigate the development of bacterial resistance.

Fig. 5.

Dual antibacterial mechanism of Exo-CA against methicillin-resistant Staphylococcus aureus (MRSA). (A) Representative images of bacterial colonies on sheep blood agar plates after 24 h treatment. (B) Live/Dead staining (SYTO 9/PI) of MRSA after 12 h exposure. Green: live; Red: dead. Scale bar: 200 μm. (C-D) Quantitative analysis of bacterial viability based on CFU counts (C) and the percentage of dead bacteria (D) (n = 3). (E) Schematic illustrating the dual mode of action: direct extracellular inhibition and suppression of intracellular MRSA survival within macrophages. (F) Confocal images of macrophages infected with GFP-MRSA (green) and treated for 4 h. Scale bar: 50 μm. (G) Quantification of the intracellular MRSA burden (n = 3). ns: not significant, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001

Development of exo-CA@CB hydrogel for treating diabetic wound healing

To translate these in vitro findings, we established a superficial MRSA-infected wound model in diabetic rats. Given that prolonged bacterial colonization and inflammation halt the transition of macrophages from the inflammatory to the proliferative phase [31, 32], a combined anti-infective and immunomodulatory strategy is critical [33]. Upon our previous work demonstrating Carbomol (CB) hydrogel as an effective delivery platform [19, 34], we incorporated Exo-CA into CB matrix to synthsize the Exo-CA@CB composite hydrogel (Fig. 6A, B) with the aims to provide sustained release of Exo-CA in the wound bed. In vivo assessments demonstrated that diabetic wounds treated with Exo-CA@CB exhibited significantly accelerated closure rates, compared to control and free CB groups, not only in the early 4 days but also in the final 12 days (Fig. 6C and Fig. S3). Evidence suggests that the hydrogel system effectively integrates the therapeutic bioactivity of Exo-CA with an optimal wound healing environment.

Fig. 6.

Fabrication of the Exo-CA@CB composite hydrogel and its therapeutic efficacy in diabetic infected wounds. (A) Schematic of Exo-CA@CB hydrogel fabrication and representative SEM images showing the porous microstructure. Scale bar: 2.5 μm. (B) Schematic diagram of Exo-CA@CB hydrogel for wound healing. (C) Representative photographs monitoring the wound healing process across different treatment groups on days 0, 4, 8, and 12. n = 6

In vivo infection control and tissue remodeling

Effective bacterial clearance is a prerequisite for tissue regeneration. We evaluated the in vivo antimicrobial efficacy of the hydrogel by quantifying bacterial burden in wound tissues. Consistent with our in vitro data, the Exo-CA@CB treatment significantly reduce bacterial load, as compared to other treatments (Fig. 7A). Histological evaluation further confirmed solid quality of wound healing. H&E staining on days 4 and 12 revealed that Exo-CA@CB treatment markedly alleviated inflammatory cell infiltration and promoted re-epithelialization (Fig. 7B). Moreover, Masson’s trichrome staining demonstrated enhanced collagen deposition and more organized fiber alignment in the Exo-CA@CB group than other groups (Fig. 7C). Quantitative analysis of the wound gap confirmed the superior therapeutic efficacy of the composite hydrogel (Fig. 7D). To verify the in vivo anti-inflammatory effect of Exo-CA@CB, we assessed iNOS expression in diabetic wounds via immunohistochemistry (Fig. S4). Control wounds showed extensive iNOS-positive staining, while CB treatment moderately reduced iNOS levels, and Exo-CA@CB treatment dramatically attenuated iNOS expression. Quantitative analysis confirmed these findings, with the Exo-CA@CB group exhibiting the lowest iNOS positive area percentage. To verify the in vivo anti-inflammatory effect of Exo-CA@CB, we assessed CD206, and NLRP3 expression in diabetic wounds via immunohistochemistry (Fig. S5). The number of CD206-positive macrophages in the Exo-CA@CB group was significantly higher than that in the other two groups, indicating that Exo-CA@CB promotes M2 macrophage polarization. Furthermore, NLRP3 immunofluorescence staining revealed markedly decreased expression following Exo-CA@CB treatment. Collectively, these results suggest that Exo-CA@CB reduces NLRP3 expression, thereby facilitating macrophage transition from the M1 to M2 phenotype and effectively suppressing the local inflammatory response.

Fig. 7.

Exo-CA@CB hydrogel accelerates wound repair by reducing bacterial burden and resolving inflammation in diabetic infected wounds. (A) Representative images of bacterial colonies isolated from wound tissues and corresponding CFU quantification. (B) H&E staining of wound sections on days 4 and 12, showing reduced inflammatory infiltration (blue arrows indicate wound edges/re-epithelialization). (C) Masson’s trichrome staining assessing collagen maturity and organization (red arrows) on days 4 and 12. Scale bar: 2000 μm. (D) Quantitative analysis of wound closure rates (wound length) over time. n = 6

Collectively, data indicate that Exo-CA@CB synergistically eliminates infection and resolves inflammation and thus, restores the structural integrity of diabetic wounds. The time-consuming preparation and isolation of exosomes represent the core bottleneck restricting its clinical application, large-scale production, and further clinical translation [35]. In contrast, Carbomol and chlorogenic acid are already commercially available products, which endows them with favorable translational potential.

Conclusions

This study demonstrates that macrophages in diabetic wounds exhibit significantly elevated NLRP3 expression, contributing to chronic inflammation and impaired healing. To address this, CA—a validated NLRP3 inhibitor with poor solubility—was encapsulated into macrophage-derived exosomes (Exo-CA). Exo-CA effectively suppressed NLRP3 activation in vitro, reprogramming macrophages from pro-inflammatory M1 to reparative M2 phenotypes. Crucially, Exo-CA also demonstrated potent antibacterial activity against MRSA, both directly and by enhancing macrophage-mediated bacterial clearance. Exo-CA was loaded into a Carbomol hydrogel (Exo-CA@CB). In vivo, Exo-CA@CB hydrogel significantly accelerated diabetic wound healing in infected rat models by synergistically reducing bacterial burden, suppressing inflammation, and promoting tissue repair. This host-directed, dual-action therapeutic strategy overcomes limitations of conventional treatments and holds significant promise for clinical translation in diabetic wound management.

Electronic Supplementary Material

Below is the link to the electronic supplementary material.

Author contributions

Hehui Wang designed the research, conducted in vitro and in vivo experiments; Ming Guan prepared the figures; Guangyu Chu analyzed the data; Hehui Wang and Yue Wang wrote the paper. All of the authors contributed to this article and approved the submitted version.

Funding

This work was supported by the National Key Research and Development Program of China (2022YFA1207503).

Data availability

Everyone can obtain datasets used and analyzed in this article from the corresponding author on sound reasons.

Declarations

Ethics approval and consent to participate

All animal experiments were approved by the Ethics Committee of the First Affiliated Hospital of Zhejiang University (SYDW20221065) and adhered strictly to ethical guidelines for animal welfare.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.