Hydrogen peroxide (H₂O₂, 30%) and DOX were purchased from Aladdin Chemistry (Shanghai) Co., Ltd. Phosphate-buffered saline (PBS) was obtained from Biosharp Biotechnology. Fresh ginger was sourced from a local market in Tianjin. The CCK-8 kit, sucrose, Calcein-AM/PI, DAPI staining solution, and Nile red staining solution were provided by Sangon Biotech (Shanghai). Crystal violet staining solution (0.1%) was supplied by Solarbio Life Sciences (Beijing, China). Dichlorodihydrofluorescein diacetate (DCFH-DA) and trypsin were purchased from Shanghai Shaoxin Biotechnology Co., Ltd. Fetal bovine serum (FBS), DMEM/high glucose medium, and RPMI-1640 medium were obtained from Thermo Fisher Scientific (China) Co., Ltd. Alizarin Red S (ARS) and alkaline phosphatase (ALP) staining kits were purchased from Cyagen (China). All other chemicals used were of analytical grade.

Fresh ginger was thoroughly washed with ultrapure water, dried, weighed, and cut into small pieces. The pieces were homogenized in 0.1 M PBS for 3 min using a homogenizer. The resulting homogenate was filtered, and the filtrate was collected and centrifuged at 4,000 × g for 60 min at 4°C. The supernatant was collected and subjected to successive centrifugation at 8,000 × g and 12,000 × g for 60 min each at 4°C. The final supernatant was then centrifuged at 16,000 × g for 60 min at 4°C, and the resulting precipitate was collected and resuspended in PBS.

To further purify the vesicles, 10% PEG6000 was added to the suspension, followed by centrifugation at 16,000 × g for 60 min at 4°C. The precipitate was collected and resuspended in PBS. A 25% sucrose solution was then added, and the mixture was centrifuged at 10,000 × g for 30 min at 4°C. The final precipitate was resuspended in PBS and filtered through a 0.45 μm membrane to obtain the purified GV solution.

GVs and DOX were mixed at a mass ratio of 1:2 and sonicated using an ultrasonic cell disruptor (JY92-IIDN, Ningbo Scientz Biotechnology Co., Ltd., China) at 1% amplitude for 10 cycles (2 s on/2 s off). Unloaded DOX was removed by ultrafiltration, and the GVs-loaded DOX was incubated at 37°C for 1 h. Zeta potential measurements before and after DOX loading were performed using a Nano-ZS 90 Nanosizer (Malvern Instruments Ltd., United Kingdom). Transmission electron microscopy (TEM) imaging was conducted with a JEM210 microscope (Japan) at 200 kV. A 5 μL droplet of the ginger vesicle solution was placed on a carbon-coated copper grid and air-dried at room temperature.

For GVs-DOX colocalization analysis, purified GVs (1 × 10¹⁰ particles/mL, quantified by nanoparticle tracking analysis) were labeled with PKH67 green fluorescent dye (2 μM in Diluent C, Sigma-Aldrich) via hydrophobic insertion at 25°C for 20 min under gentle agitation. Unbound dye was removed by ultracentrifugation (100,000×g, 4°C, 2 h) through a 30% sucrose cushion. DOX was fluorescently labeled with Cy5.5 NHS ester (Lumiprobe; excitation/emission: 675/694 nm) at a 1:5 M ratio via amine coupling. Excess dye was removed using Zeba™ Spin Desalting Columns (7K MWCO, Thermo Fisher). Fluorescent imaging was performed using a Leica TCS SP8 STED 3X confocal microscope.

The synthesis of CMCSMA was performed as described in previous studies and detailed in the Supplementary Material (Bucciarelli et al., 2024; Schuurmans et al., 2021). For hydrogel formation, freshly prepared solutions were used: 10% (w/v) CMCSMA, 5% (w/v) TA, 0.5 M HCl, and 4% (w/v) NaHCO₃. Specifically, 1 mL of 10% CMCSMA solution was mixed with 225 μL of 5% TA solution and 50 μL of 0.5 M HCl. After thorough mixing, 100 μL of 4% NaHCO₃ was added, and the solution was immediately shaken for approximately 1 min. Gelation was achieved through photocrosslinking for 5 min.

To characterize the hydrogel, samples were flash-frozen in liquid nitrogen and lyophilized. The surface morphology of the resulting hydrogels and nanoparticles was examined using scanning electron microscopy (SEM, Hitachi Regulus 8230). Fourier-transform infrared spectroscopy (FTIR, Thermo Fisher Nicolet 6700) was conducted in the 500–4,000 cm⁻¹ range to analyze chemical structures. The dynamic rheological properties, including storage modulus and loss modulus, were evaluated using a rotational rheometer (Anton Paar, Graz, Austria) as a function of temperature. To assess drug release kinetics, hydrogels were immersed in 1 mL of PBS (pH 7.4) and incubated at 37°C in a humid environment. At predetermined time intervals, the DOX concentration in the release medium was measured via ultraviolet-visible (UV-Vis) spectrophotometry. The entire PBS volume was replaced with fresh buffer each time to maintain sink conditions throughout the release study.

The hemostatic efficacy of the CMCSMA/TA hydrogel for deep, incompressible wounds was evaluated using a liver incision model, as described in previous studies (Jiang et al., 2024). In this model, a 0.5 cm deep incision was made in the liver using a scalpel to induce bleeding. Then, 100 µL of preformed hydrogel was applied directly to the wound site. Hemostasis time and total blood loss were recorded to assess efficacy. A no-treatment group served as the control.

To evaluate cellular uptake, Cy5-labeled DOX was loaded into GVs (Cy5-DOX@GVs) and visualized using fluorescence microscopy. MNNG/HOS and bone marrow-derived mesenchymal stem cells (BM-MSCs) were seeded in 6-well plates at a density of 5 × 10⁵ cells per well, incubated overnight, and treated with Cy5-DOX@GVs for 4 h. Cells were then washed, stained with 1 μM DAPI for 30 min, and imaged using a fluorescence microscope (Nikon, Japan). For cytotoxicity assessment, 2 × 10⁴ MNNG/HOS cells were seeded into 48-well plates with 1 mL of complete DMEM. After 12 h of incubation, 50 μL of various hydrogels or PBS (control) were added. Following a 24-h treatment, live/dead cell staining was performed using Calcein-AM and PI. To assess cell migration, a Transwell assay was conducted. A total of 2 × 10⁴ MNNG/HOS cells in 100 μL of serum-free DMEM were seeded into the upper chamber of 24-well Transwell inserts (8 μm pore size, BD Biosciences). The lower chambers were filled with DMEM containing 20% FBS as a chemoattractant. After 24 h of incubation, cells were fixed with 4% paraformaldehyde for 15 min, stained with 0.1% crystal violet for 30 min, and imaged under a light microscope. Migrated cells were quantified by counting the average number per field.

To assess the ROS scavenging ability and corresponding cytoprotective effects of the hydrogels, BM-MSCs cells were seeded at a density of 1 × 10⁵ cells per well in 48-well plates with 1 mL of complete DMEM. After 12 h of incubation, the medium was replaced with 1 mL of DMEM containing 1 mM H₂O₂ and 100 μL of the respective hydrogels or PBS (control). Following 2 h of incubation, the hydrogels were removed, and the wells were gently rinsed with PBS. Cells were then stained with DCFH-DA and Hoechst 33,342, and observed under a fluorescence microscope to evaluate intracellular ROS levels. To assess cell proliferation, BM-MSCs (1 × 10⁴ cells per well) were seeded in 96-well plates pretreated with different hydrogel formulations and cultured for 1, 3, and 7 days. Cell viability was measured using a CCK-8 assay, with absorbance recorded at 450 nm using a microplate reader. For quantitative analysis of anti-inflammatory, anti-senescence, and osteogenic gene expression, BM-MSCs were seeded at 2 × 10⁵ cells per well in 6-well plates and incubated overnight with or without hydrogels. The following day, cells were treated with or without 0.5 μg/mL lipopolysaccharide (LPS) for 24 h. After stimulation, total RNA was extracted, reverse transcribed, and subjected to quantitative real-time PCR (RT-PCR) using a SYBR Green Master Mix Kit. Primer sequences are provided in Supplementary Table S1.

Mouse bone marrow-derived macrophages (BMDMs) were isolated from the femurs of female mice (6–8 weeks old) and seeded in 24-well plates at a density of 1 × 10⁵ cells/well. After 24 h, the culture medium was replaced with GelA medium supplemented with RANKL (50 ng/mL), M-CSF (40 ng/mL), and TGF-β2 (1 ng/mL). The medium was refreshed every 2 days. After 7 days of induction, cells were fixed with 10% neutral-buffered formalin and stained with TRAP solution according to the manufacturer’s protocol to assess osteoclast formation. Cell viability was assessed using an MTT assay, and absorbance was measured with a microplate reader to calculate survival rates.

To evaluate the antitumor effect of the hydrogel, MNNG/HOS tumor-bearing nude mice were randomly divided into four groups (n = 4 per group): Control (PBS injection), GVs@Gel (blank hydrogel with ginger vesicles), and GVs-DOX@Gel (hydrogel loaded with both GVs and doxorubicin). Due to the localized delivery format, a free DOX-only group was not included, as unencapsulated DOX rapidly diffuses and fails to mimic clinical hydrogel-mediated retention. When tumors reached ∼50 mm³, mice were anesthetized using isoflurane (4% induction in 100% oxygen, followed by 2% maintenance via nose cone), and 100 μL of PBS, GVs@Gel, or GVs-DOX@Gel was injected peritumorally (single administration). Body weight and tumor volume were monitored daily. On day 10, mice were euthanized by intraperitoneal injection of sodium pentobarbital (150 mg/kg) followed by cervical dislocation to ensure irreversible loss of consciousness, in accordance with AVMA euthanasia guidelines, and tumors were harvested for histological analysis. Hematoxylin and eosin (H&E), Ki67, and TUNEL staining were performed on tumor sections to investigate the therapeutic mechanism of GVs-DOX@Gel. The sample size (n = 4) was selected based on preliminary power analysis and ethical guidelines to reduce animal use while allowing detection of statistically significant therapeutic effects.

A full-thickness excisional wound model was established in male Sprague-Dawley rats (6–8 weeks old) to assess the wound healing potential of the hydrogels. Anesthesia was induced via intraperitoneal injection of a mixture containing ketamine (80 mg/kg) and xylazine (10 mg/kg) in 0.9% saline (1 mL/kg), ensuring surgical plane anesthesia with loss of pedal withdrawal reflex. Animals were maintained on a heating pad to preserve body temperature (37°C ± 0.5°C) throughout the procedure. Under anesthesia, rats were shaved and subjected to an 8 mm diameter circular skin excision using a biopsy punch. Animals were randomly assigned to three groups (n = 3 per group): PBS, GVs@Gel, and GVs-DOX@Gel. At predetermined time points, skin tissues from the wound sites were collected, fixed in paraffin, and subjected to histological evaluation using H&E and Masson’s trichrome staining.

All data are presented as mean ± standard deviation (SD). Statistical comparisons between groups were performed using two-tailed t-tests in GraphPad Prism 6. Statistical significance was indicated as ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, and ∗∗∗∗p < 0.0001, respectively.

GVs were isolated via size exclusion chromatography and ultrafiltration (Figures 2A,B) and characterized using TEM (Figure 2C). GVs were predominantly concentrated in the fourth fraction, while negligible amounts were detected in other fractions or in unpurified samples. To confirm successful encapsulation of DOX, GVs were labeled with PKH67 (green fluorescence), and DOX was labeled with CY5 (red fluorescence). Fluorescence microscopy revealed a strong overlap of green and red signals in the merged images, confirming efficient colocalization of DOX within the GVs (Figure 2D). The hydrodynamic diameter and zeta potential of GVs-DOX were measured by dynamic light scattering (DLS). The particles exhibited an average diameter of 100.4 ± 25.52 nm and a zeta potential of −11.66 mV, indicating colloidal stability (Figures 2E,F). To characterize the CMCSMA/TA hydrogel encapsulating GVs-DOX, scanning electron microscopy was performed on lyophilized samples. The hydrogel exhibited a loose, highly porous architecture with interconnected pores ranging from 50 to 200 μm (Figure 2G), a structure favorable for nutrient diffusion, cell infiltration, and tissue regeneration. Fourier-transform infrared (FT-IR) spectroscopy was used to analyze the molecular interactions between CMCSMA and TA. Characteristic peaks of CMCSMA appeared at 1,652 cm⁻¹ (amide I, C=O stretch) and 1,550 cm⁻¹ (amide II, N–H bend), while TA showed a peak at 3,340 cm⁻¹ (–OH stretch). In the hydrogel spectrum, CMCSMA’s amide peaks were significantly weakened, and the–OH peak shifted to 3,205 cm⁻¹ with notable broadening. These changes suggest the formation of hydrogen bonds and potential ionic or covalent interactions between CMCSMA and TA, indicating successful hydrogel formation (Figure 2H). Rheological analysis revealed that the storage modulus (G′) exceeded the loss modulus (G″), confirming the formation of a stable, crosslinked hydrogel network (Supplementary figure S1). To evaluate the drug release behavior, an in vitro release study of DOX was conducted in PBS. After 24 h, only 39.46% of DOX was released from GVs-DOX@Gel, compared to 59.15% from DOX@Gel without GVs encapsulation (Figure 2I). This demonstrates that GVs encapsulation effectively slowed the release rate, supporting sustained drug delivery and prolonged retention at the tumor site. In summary, the CMCSMA/TA hydrogel successfully encapsulated DOX-loaded GVs, forming a stable, porous, and biocompatible structure with controlled and sustained drug release, ideal for localized tumor therapy.

Severe bleeding is a leading cause of mortality, particularly in combat and emergency trauma settings. Blood transfusions after massive hemorrhage can result in complications such as coagulopathy, infection, and multi-organ failure. Therefore, rapid and effective hemostasis is critical to reducing blood loss and improving survival rates (Gao et al., 2020). To assess the hemostatic performance of the CMCSMA/TA hydrogel, we employed a rat liver incision bleeding model (Figure 3). Both the medical sponge and the CMCSMA/TA hydrogel significantly reduced blood loss compared to the untreated control. In the control group, blood loss averaged 4.97 g, which was reduced to 3.75 g with the medical sponge and further decreased to 2.59 g with the hydrogel. Similarly, hemostatic time decreased from 516 s in the control group to 434 s with the sponge and 232 s with the hydrogel. These results indicate that the CMCSMA/TA hydrogel exhibits superior hemostatic efficacy compared to the medical sponge. This enhanced hemostatic performance is likely attributable to both the intrinsic bioactivity and the adhesive properties of the hydrogel. Chitosan’s amino groups become protonated in blood, forming -NH₃ ⁺ groups that attract negatively charged platelets, promoting red blood cell aggregation and thrombus formation (Yang et al., 2020). TA contributes through vasoconstrictive effects and its interaction with plasma proteins to facilitate clotting (Yang et al., 2020). Additionally, the hydrogel’s strong tissue adhesion enables it to effectively seal wound sites and block bleeding. In summary, the CMCSMA/TA hydrogel demonstrates potent hemostatic activity due to the synergistic effects of its bioactive components and adhesive capability, making it a promising candidate for intraoperative bleeding control.

The cellular uptake of GVs-DOX nanoparticles (NPs) was evaluated in MNNG/HOS osteosarcoma cells and bone marrow-derived mesenchymal stem cells (BM-MSCs) using fluorescence microscopy. To visualize uptake, DOX was replaced with Cy5-labeled DOX (red fluorescence), while cell nuclei were counterstained with DAPI (blue). As shown in Figure 4A, cells treated with Cy5-DOX@GVs exhibited strong red fluorescence, indicating enhanced cellular internalization of the nanoparticles. To examine the in vitro therapeutic effects, MNNG/HOS cells were co-cultured with various hydrogel formulations. Compared to the Control and Gel groups, the GVs-DOX@Gel group displayed a significantly higher proportion of dead cells, as evidenced by more intense red fluorescence in live/dead staining images (Figures 4B–D). Furthermore, Transwell migration assays revealed that GVs-DOX@Gel treatment substantially inhibited tumor cell migration compared to other groups (Figures 4E,F).

Good biocompatibility is essential for materials intended for biomedical applications. To evaluate the biocompatibility of the GVs-DOX@Gel hydrogel, bone marrow-derived mesenchymal stem cells (BM-MSCs) were used. Intracellular ROS levels were measured using the DCFH-DA fluorescent probe. As shown in Figure 5A, H₂O₂ treatment markedly increased ROS generation, whereas treatment with GVs-DOX@Gel resulted in minimal green fluorescence, indicating a significant reduction in cellular ROS levels. Cytotoxicity was assessed using the CCK-8 assay (Figure 5B), which showed that the hydrogel exhibited good cell viability, confirming its biocompatibility with BM-MSCs.

To further explore the hydrogel’s therapeutic potential, its anti-inflammatory, anti-senescent, and pro-osteogenic properties were evaluated. BM-MSCs were stimulated with H₂O₂ or lipopolysaccharide (LPS), a potent pro-inflammatory agent. Quantitative PCR analysis of gene expression (Figure 5C) revealed that GVs-DOX@Gel treatment significantly downregulated the expression of the pro-inflammatory gene MMP13 and the senescence marker P21, while maintaining the expression of osteogenic (OCN) and angiogenic (VEGF) markers, indicating a favorable pro-regenerative profile.

The osteogenic potential of GVs-DOX@Gel was further validated using alizarin red (ARS) and alkaline phosphatase (ALP) staining in MC3T3-E1 cells. As shown in Figure 5D, treatment with GVs-DOX@Gel significantly enhanced mineralization and ALP activity compared to the control, suggesting promotion of osteogenic differentiation. Additionally, tartrate-resistant acid phosphatase (TRAP) staining (Figure 5E) showed a notable reduction in TRAP-positive cells following treatment, indicating inhibition of osteoclastic differentiation in BMDMs. MTT assays further confirmed that GVs-DOX@Gel suppressed osteoclast proliferation and activity. Collectively, these findings demonstrate that GVs-DOX@Gel possesses excellent biocompatibility and exhibits multifunctional properties, including antioxidant, anti-inflammatory, anti-senescent, osteogenic, and anti-osteoclastic effects, making it a promising candidate for postoperative bone regeneration and osteosarcoma therapy.

To further evaluate the in vivo antitumor efficacy of the GVs-DOX@Gel hydrogel, we established an osteosarcoma xenograft model by subcutaneously injecting MNNG/HOS cells into Balb/c nude mice. Once tumor volumes reached approximately 50 mm³, mice received peri-tumoral injections of PBS (control), GVs@Gel, or GVs-DOX@Gel. Body weight and tumor volume were monitored daily for 21 days. As shown in Figures 6A,B, the GVs-DOX@Gel group exhibited significant tumor growth inhibition compared to both the PBS and GVs@Gel groups. Although the GVs@Gel group initially showed some suppression of tumor growth, tumors began to grow rapidly after day 6. In contrast, GVs-DOX@Gel maintained continuous tumor suppression throughout the observation period (Figures 6C,D). This sustained effect can be attributed to the hydrogel’s role as a localized drug reservoir, enabling gradual DOX release and maintaining therapeutic drug concentrations at the tumor site.

Histological analysis of excised tumors further confirmed the superior antitumor activity of GVs-DOX@Gel. H&E staining, Ki67 immunohistochemistry, and TUNEL fluorescence staining were conducted. Ki67 staining revealed significantly reduced tumor cell proliferation in the GVs-DOX@Gel group, while TUNEL staining showed markedly increased apoptosis, as indicated by stronger red fluorescence compared to the other groups (Figures 6E-G). Together, these results demonstrate that GVs-DOX@Gel hydrogel effectively inhibits tumor proliferation and promotes apoptosis in vivo by providing sustained DOX release, confirming its strong potential for postoperative osteosarcoma treatment.

Surgical resection remains a key treatment for osteosarcoma, yet challenges such as local tumor recurrence and delayed wound healing persist. To evaluate the in vivo therapeutic potential of GVs-DOX@Gel in addressing these issues, we established a full-thickness excisional skin wound model in rats to simulate postoperative healing. Rats were divided into three groups: Control (treated with PBS), Gel (treated with CMCSMA/TA hydrogel), and GVs-DOX@Gel. First, as shown in Figure 7A, a subcutaneous osteosarcoma xenograft model was created by injecting MNNG/HOS cells into immunodeficient rats. Seven days later, tumors were surgically resected to simulate postoperative trauma. Following resection, the wounds were treated with the respective formulations. As shown in Figure 7B, the GVs-DOX@Gel group exhibited significantly faster and more complete wound closure compared to both the Control and Gel groups. Quantitative analysis revealed that the GVs-DOX@Gel group consistently showed the smallest wound area on days 3, 7, and 12 post-treatment (Figures 7C,D). To further assess tissue regeneration, skin samples collected on day 14 were analyzed using H&E and Masson staining. H&E staining (Figure 7E) revealed that the GVs-DOX@Gel group had a thicker regenerated epidermis (green arrow) and more skin appendages, such as glands and hair follicles. Masson staining (Figure 7F) demonstrated enhanced collagen deposition in the GVs-DOX@Gel group, indicating improved dermal remodeling. Notably, both the Gel and GVs-DOX@Gel groups outperformed the Control group in promoting wound healing, suggesting that the CMCSMA/TA hydrogel contributes to wound sealing and tissue repair. However, the superior outcomes observed in the GVs-DOX@Gel group highlight the added therapeutic benefits of GVs-DOX, particularly in enhancing fibroblast activity and collagen synthesis. These results demonstrate that GVs-DOX@Gel not only supports effective postoperative tumor therapy but also significantly accelerates wound healing, making it a promising candidate for integrated osteosarcoma treatment.