PLGA₁₅₀₀₀-PEG₅₀₀₀ (GA: LA = 50:50) polymer was purchased from Xi’an Ruixi Biological Technology Co., Ltd. (Xi’an, China). Isopropyl β-D-thiogalactoside (IPTG), dichloromethane (DCM), 2-morpholinoethanesulfonic Acid (MES), acetone, ampicillin (AMP), imidazole were purchased from Sinopharm Chemical Reagent Co., Ltd. N-hydroxy succinimide (NHS) and 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide (EDC) were purchased from Shanghai Macklin Biochemical Technology Co., Ltd. Bovine serum albumin (BSA), 2,2′-azino-bis (3-ethylbenzothiazoline-6-sulfonic acid) (ABTS), horseradish peroxidase (HRP)-labeled goat anti-mouse IgG, and 0.45 µm sterile filters were purchased from Merck (Shanghai, China). Alhydrogel™ was purchased from Croda (UK). Ninety-six-well polystyrene plates for the enzyme-linked immunosorbent assay (ELISA) were purchased from Corning (New York, NY, USA). Enzyme-linked immunospot (ELISPOT) kits were purchased from Mabtech (Stockholm, Sweden). Escherichia coli (E. coil) BL21 (DE3) was purchased from TIANGEN BIOTECH (BEIJING) Co., Ltd. (Beijing, China). Ni-Sepharose was purchased from Cytiva (Shanghai, China). A High-Capacity Endotoxin Removal column was purchased from Xiamen Bioendo Technology Co., Ltd. (Xiamen, Chian). Bradford kits were purchased from Beijing Solarbio science&technology Co., Ltd. BALC/c mice were purchased from Air Force Millitary Universities. The S.aureus ATCC25923 strain was obtained from Xijing Hospital (Xi’an, China).

rEsxA and rEsxB were expressed and purified from E.coli. Briefly, the full-length DNA sequences of rEsxA and rEsxB were cloned to pET28a vector with a 6×His-tag. The construct was transformed into E.coli BL21(DE3). A single colony was amplified in 5 mL Luria-Bertani (LB) medium with AMP (100 μg ml^-1) for 3 h at 37°C while shaking. The pre-culture was diluted 1:100 in 1 L LB medium (100 µg ml⁻¹ AMP) and cultured at 180 rpm at 37°C, until OD₆₀₀ ~ 0.6. Protein expression was induced by adding IPTG, and cultivation continued under the same conditions (37°C, 180 rpm) for 4 h. The culture was centrifuged and the pellet was re-suspended in PBS. The cell suspension was homogenized and centrifuged at 15,000 × g for 10 min at 4°C. The supernatant was purified by Ni-Sepharose chromatography with 100 mmol/L imidazole wash buffer. The purified protein was dialysed and finally the buffer was replaced with PBS. The endotoxins were subsequently removed from the protein solution using the EtEraser™ HP High-Efficiency Endotoxin Removal Kit. The concentration, purity and specificity of these antigens were determined by BCA assay, SDS-PAGE and Western Blot.

25% NPs were synthesized using the emulsion solvent evaporation method, as described previously (19–21). Briefly, 100 mg PLGA-PEG polymer was dissolved in a mixture of 1.5 mL dichloromethane (DCM) and 1 mL acetone. This organic solution was then added to 10 mL of 5% (w/v) polyvinyl alcohol (PVA) aqueous solution. The mixture was emulsified using an ultrasonic cell disruptor for 2 min to form an oil-in-water (O/W) emulsion. The emulsion was subsequently poured into 50 mL of deionized (DI) water and stirred for 4 h at room temperature. The resulting 25% NPs were collected, filtered through a 0.45 µm membrane filter, centrifuged at 18,000 × g for 1 h at 4°C, washed three times with DI water, and lyophilized with trehalose for storage.

rEsxA and rEsxB were covalently conjugated to the 25% NPs via an acylation reaction, following our established protocol (19). 10 mg of 25% NPs were dispersed in 10 mL of 25 mmol/L 2-(N-morpholino)ethanesulfonic acid (MES) buffer. The carboxyl groups on the NPs were activated by sequentially adding 0.4 mL of 1 mol/L 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide (EDC) and 0.25 mL of 1 mol/L N-hydroxysuccinimide (NHS). The mixture was then incubated at room temperature for 3 h. The activated NPs were centrifuged at 18,000 × g for 1 h at 4°C and washed three times with DI water. The activated NP pellet was resuspended in 10 mL of PBS containing either rEsxA or rEsxB at a concentration of 1 mg/mL and incubated overnight at 4°C with gentle agitation. The resulting antigen-conjugated NPs (25% NPs-rEsxA or 25% NPs-rEsxB) were collected by centrifugation at 16,500 × g for 1 h at 4°C, washed three times with DI water, and lyophilized with trehalose for subsequent use.

The morphology of 25% NPs was confirmed through scanning electron microscope (SEM) (Quattro S, ThermoFisher, Massachusetts, US). The size distribution and zeta potential of 25% NPs, 25% NPs-rEsxA and 25% NPs-rEsxB were measured by dynamic light scattering (DLS) (Delsa™ Nano, Beckman-Coulter, High Wycombe, UK). Each sample was measured in triplicate. The successful synthesis of the 25% NPs-rEsxA or rEsxB were determined by Fourier Transform infrared spectroscopy (FTIR, Thermo Nicolet IS50, Thermo Fisher Scientific, Waltham, MA, USA). The Bradford assay was applied to quantify the total antigens loaded, and the loading efficiency was calculated using the following formula:

In vitro cytotoxicity of the 25% NPs nanovaccines (25% NPs-rEsxA and 25% NPs-rEsxB) against L929 fibroblast cells was assessed using the CCK-8 assay. Briefly, L929 cells were seeded into 96-well plates containing complete culture medium (RPMI 1640 supplemented with 10% fetal bovine serum (FBS) and 1% penicillin-streptomycin (Pen/Strep)) and incubated for 24 h. Subsequently, cells were treated with varying concentrations of the nanovaccines (ranging from 1 mg mL^-¹ to 7.8125 μg mL^-¹) for 12 h. Following treatment, the cells were washed twice with PBS, and CCK-8 solution was added to each well. The plates were then incubated at 37°C for 1–4 h. The optical density (OD) at 450 nm was measured using a microplate reader. Cell viability was calculated using the following formula:

In vivo biocompatibility of the nanovaccines was evaluated in BALB/c mice by monitoring body weight changes. Mouse body weights were recorded every seven days for a total of five measurements following the initial vaccination.

Additionally, hematoxylin and eosin (H&E) staining of major organs was performed according to a standard protocol described in our previous work. Organs were harvested from mice euthanised 42 days after the first vaccination for histological analysis.

Female BALB/c mice (six-week-old) were obtained from the Animal Experiment Center of Air Force Medical University. All animal procedures were conducted in accordance with the National Research Council’s Guide for the Care and Use of Laboratory Animals and were approved by the Animal Care and Ethics Committee of the Fourth Military Medical University (Approval No. KY20213144-1). Negative control groups received PBS, rEsxA (12.5 μg/mouse), rEsxB (12.5 μg/mouse), or rEsxA + rEsxB (12.5 μg each antigen, total 25 μg/mouse). Positive control groups (Alum-adjuvanted) received Alum mixed with rEsxA (12.5 μg/mouse), rEsxB (12.5 μg/mouse), or rEsxA + rEsxB (12.5 μg each, total 25 μg/mouse). Test groups received 25% NPs conjugated with rEsxA (12.5 μg antigen/mouse), rEsxB (12.5 μg antigen/mouse), or rEsxA + rEsxB (12.5 μg each antigen, total 25 μg/mouse). The mice were immunized once every two weeks for a total of three immunizations, with six mice per group, and each group received a volume of 200 μL. Blood samples were collected from the tail vein on days 0, 17, and 35 for subsequent analysis.

The rEsxA and rEsxB antibody responses were detected by enzyme-linked immunosorbent assay (ELISA) on days 35 after the first immunization. Typically, rEsxA or rEsxB protein was pre-immobilized on 96-well ELISA plate (Corning) at 10 μg/mL (PBS, 100 μL/well, incubated overnight at 4°C). And The plate was then washed 5 times with PBS containing 0.1% Tween (PBST). Twofold serially diluted mice serum ranging from 1:250 to 1: 512,000 (100 μL/well) at 37°C for 1 h. The plate was washed 5 times with PBST and incubated with HRP-labeled goat anti-mouse IgG (1:3300, 100 µL/well) at 37°C for 45 min. After washing 5 times with PBST, 100 µL of ABTS solution was added to each well for 20 min at room temperature and OD value was measured using a microplate reader (Biotek SYNERGY LX Instruments, Santa Clara, CA, USA) at 405 nm. For antibody responses analysis, the area under the curve (AUC) method was more accurate than the endpoint titer (ET) method. The antigen-specific antibody titer results were represented by the mean serum ELISA AUC with the S.D. of 6 mice in each group.

On day 7 after the third boost vaccination through subcutaneous (S.C.), mice were euthanized by exposure to a gradually rising concentration of carbon dioxide (CO°C), followed by cervical dislocation to ensure death. Cytokine IL-4 (representative of humoral immunity), IL-17A (representative of Th17 T cell activation) and IFN-γ (representative of cell-mediated immunity) were analyzed using a cytokine-specific enzyme-linked immunospot (ELISPOT) assay. Briefly, the ELISPOT plate (capture antibody precoated) were washed 4 times with PBS and blocked with a blocking solution (RPMI-1640 containing 10% FBS, 200 µL/well) for 30 min at room temperature. Splenocytes isolated from immunized mice were plated at a concentration of 1 × 10⁶ cells/well and induced with each antigen alone (4 µg/ml antigen) in triplicate and incubated for 24 h at 37°C. Phytohemagglutinin (PHA, 10 µg/mL) was used as a positive control. Splenocytes from naive mice stimulated with rEsxA or rEsxB, splenocytes from unstimulated mice, and RPMI 1640-treated splenocytes were used as negative controls. After washing 5 times with PBS, the detection antibodies into PBS containing 0.5% FBS (PBS-0.5% FBS) were added to each well and incubated for 2 h at room temperature, followed by streptavidin-HRP conjugate for 1 h at room temperature. And then the plates were washed as above, the color was developed with TMB solution and the spots were counted in an ELISPOT reader (Cellular Technology Limited, Cleveland, OH, USA).

Staphylococcus aureus strain ATCC 25923 was grown in LB medium at 37°C with shaking at 180 rpm until mid-exponential phase (OD₆₀₀ ~ 0.6). Bacterial cells were collected by centrifugation, washed twice, and resuspended in PBS. The bacterial concentration was adjusted spectrophotometrically based on OD₆₀₀ and verified by plating serial dilutions on LB agar for colony-forming unit (CFU) enumeration. The challenge inoculum was prepared at a target concentration of 5.26 × 10⁸ CFU/mL, which corresponded to an LD₁₀₀ dose based on preliminary virulence assessments. On day 42 after the first vaccination, immunized mice were intravenously injected via the tail vein with 5.26 × 10⁸ CFU (1 × LD₁₀₀) or 10.52 × 10⁸ CFU (2 × LD₁₀₀) of S. aureus ATCC 25923. Mice were monitored for 14 days for mortality and clinical signs. A humane endpoint protocol was strictly followed, under which any mouse exhibiting either ≥20% weight loss from baseline or severe clinical symptoms indicating irreversible distress or morbidity-such as pronounced lethargy, hunched posture, or inability to access food or water, was promptly euthanized by CO°C asphyxiation to prevent undue suffering. All surviving mice were euthanized at the conclusion of the experiment.

The bacteriolytic activity of antibodies in all experimental groups was evaluated using the standard Neisser-Wechsberg bacteriolysis assay (22, 23). Staphylococcus aureus ATCC 25923 was cultured to OD600 ~ 0.6. The bacterial suspension was then diluted 10-fold in sterile saline. A reaction mixture was prepared by combining 0.2 mL of the diluted bacteria, 0.2 mL of 10-fold diluted serum from immunized mice, 0.2 mL of 10-fold diluted guinea pig serum (as an exogenous complement source), and 1.4 mL of sterile saline. This mixture was incubated at 37°C for 30 minutes. After incubation, appropriate dilutions were spread onto LB agar plates and incubated overnight at 37°C to enumerate viable bacteria. Finally, the number of colony-forming units of S. aureus was counted and the percentage of lytic bacteria was calculated as the following formula:

The quantitative data are expressed as mean ± standard deviation (SD). Statistical analysis was evaluated by using an unpaired two-tailed Student’s t test and log-rank (Mantel–Cox) analysis was employed to determine the statistical significance of the survival rate. Results with * p< 0.05, ** p< 0.01, *** p< 0.001, and **** p< 0.0001 were considered statistically significant.

25% NPs were synthesized as previously described, with the synthetic route schematically illustrated in Figure 2A. The preparation and purification of the recombinant proteins rEsxA and rExsB were carried out following the standard protocols. The rEsxA and rExsB were then loaded on the NPs surface by covalent conjugation, respectively.

SDS-PAGE analysis was employed to testify the successful preparation and purification of the rEsxA and rExsB antigen. We then examined the NPs and NPs-antigen’s morphology, size distribution and zeta potential by SEM and DLS.

As shown in Figure 2B, the molecular weight of rEsxA and rEsxB were 11.5 kDa, confirmed by SDS-PAGE. SEM results showed spherical morphology of the 25% NPs (Figure 2C). The hydrodynamic diameter of bare 25% NPs was 182.4 ± 1.2 nm (Figure 2E), and the zeta potential was -19.73 ± 0.2 mV (Figure 2F). rEsxA and rEsxB were then conjugated onto the NPs surface by amide covalent bonding, confirmed by FT-IR spectra. The amide bond peaks at 3288.6 cm^-¹ (ν_N-H) and 1655.8 cm^-¹ (ν_C=O) for 25% NPs-rEsxA, and the peaks of 3300.8 cm^-¹ (ν_N-H) and 1657.9 cm^-¹ (ν_C=O) for 25% NPs-rEsxB, confirmed the successful conjugation of rEsxA or rEsxB to the 25% NPs, respectively (Figure 2D).

After conjugation with rEsxA or rEsxB, the hydrodynamic diameters increased to 209.7 ± 0.6 nm (25% NPs-rEsxA) and 205.8 ± 0.8 nm (25% NPs-rEsxB), indicating that the conjugation of antigen leads to a ~20 nm size increase (Figure 2E). The zeta potential became more negatively after antigen conjugation (Figure 2F), with -22.59 ± 1.1 mV for 25%NPs-rEsxA and -21.04 ± 1.4 mV for 25%NPs-rEsxB). BCA assay revealed that the loading efficiency of both 25% NPs-rEsxA and 25% NPs-rEsxB was ~3.0% (mass ratio) (Figure 2G).

Before the efficacy evaluation of the nanovacines, 25% NPs-rEsxA, 25% NPs-rEsxB and 25% NPs-rEsxA+25% NPs-rEsxB, the biosafety issues were first addressed in vitro and in vivo. CCK-8 assays, which employed to test the cytotoxicity of the nanovaccines, demonstrated the well cell viability (>85%) in all tested groups (Figure 3A).

Subcutaneous administration in mice caused no significant weight changes compared to controls (Figure 3B). Histopathological analysis via H&E staining revealed that no lesions in major organs (heart, liver, kidney, lung, spleen) were observed in immunized mice (Figure 3C; Supplementary Figure S1).

To evaluate the humoral immunity induced by the nanovaccines, antigen-specific IgG levels were measured after immunization in all groups. Vaccination was performed according to standard protocols (prime on day 0; boosts on days 14 and 28; Figure 4A) via subcutaneous administration. Blood samples were collected on days 17 and 35 after the first administration. Day 17 was selected to capture the early adaptive immune response, representing a critical time window for assessing the initial immunogenicity of the vaccine prior to the dominance of memory responses. An ELISA assay was used to assess humoral immune activation. The results showed that at day 17 after the first vaccination, IgG responses in the PBS, free antigen, and Alum-adjuvanted vaccine groups were barely detectable. In contrast, the nanovaccine groups exhibited a significant increase in IgG response. Moreover, the combined nanovaccine group (25% NPs-rEsxA + 25% NPs-rEsxB) showed an approximately 1.8-fold higher IgG response compared to the single-antigen nanovaccine groups (25% NPs-rEsxA or 25% NPs-rEsxB) (Figures 4B, C).

By day 35 after the first vaccination, the PBS and free antigen groups remained seronegative, while the Alum-adjuvanted vaccine groups produced significant IgG levels. The nanovaccine groups showed further increases in IgG antibody levels compared to day 17. When comparing the Alum-adjuvanted vaccines and the nanovaccines, the rEsxA-specific IgG response was comparable between the two groups (Figure 4D). However, the rEsxB-specific IgG response was significantly higher in the nanovaccine group than in the Alum-adjuvanted group. In the combined vaccination groups (Alum-rEsxA + Alum-rEsxB and 25% NPs-rEsxA + 25% NPs-rEsxB), IgG responses to rEsxA and rEsxB were evaluated separately (Figure 4E). The results demonstrated that for both rEsxA- and rEsxB-specific IgG, the AUC values of the 25% NPs-adjuvanted vaccine group were significantly higher than those of the Alum group, indicating that 25% NPs serves as a more effective adjuvant for the recombinant proteins rEsxA and rEsxB.

In order to test the tendency of the immune response, ELISPOT assay was employed to study the specific cytokines, IL-4 representative for Th2, IFN-γ for Th1/CTL and IL-17A for Th17/innate immune response.

Firstly, for IL-4 production, though there was no significant difference of IgG titer between Alum-rEsxA and 25% NPs-rEsxA, the IL-4 spots secreted by spleen cells of the vaccinated mice in 25% NPs-rEsxA group was significantly higher than that in Alum-rEsxA ones. The results indicated that 25% NPs served as the rEsxA adjuvant could activate the IL-4 associated Th2 immune response more efficiently (Figure 5B; Supplementary Figure S3). Similarly, we compared the IL-4 spots data of rEsxB between the Alum vaccine and the 25% NPs ones. The same trend was found with the antibody titer results in Figure 4E. The 25% NPs was more helpful for IL-4 secretion compared with Alum adjuvant, when the vaccine antigen is rEsxB (Figure 5C; Supplementary Figure S3). It was undoubtable that in the combined vaccination batch, 25% NPs vaccines performed better than the Alum ones, either for rEsxA or rEsxB.

We also compared the IFN-γ production in all tested and control groups, in order to investigate the cellular immune response activation issues by ELISPOT assay. For either rEsxA or rEsxB specific stimulation, the IFN-γ spots in 25% NPs adjuvant vaccine groups were more than that in Alum adjuvant vaccine groups, no matter for single antigen nor antigen combination cases (Figures 5D, E; Supplementary Figure S4). The results indicated that 25% NPs served as rEsxA or rEsxB adjuvant could better stimulate the Th1/CTL immune responses than Alum adjuvant.

As a critical effector of innate immunity and a bridge to adaptive immunity, IL-17A plays a pivotal role in host defense. Similar trend was found in the IL-17A ELISPOT assay compared with the IFN-γ secretion results. For single antigen loading, 25% NPs vaccines elicited 249 (25% NPs-rEsxA) and 203 (25% NPs-rEsxB) IL-17A spots, about 4.4-fold higher than Alum ones (Figures 5F, G, Supplementary Figure S5). Moreover, for combined vaccines treatment groups, 25% NPs vaccine groups further amplified this immune response, with IL-17A production 470 and 353 spots for rEsxA and rEsxB, respectively. Except for the humoral and cellular immune response stimulation, the enhancement of innate immune response by 25% NPs adjuvant may lead to better body protection against S. aureus than Alum vaccines.

Bacteriolysis assay was performed to evaluate the antibody-dependent bactericidal activity elicited by vaccination with 25% NPs-based vaccines or Alum-adjuvanted vaccines, using free antigen and PBS as controls. As shown in Figure 6, antibodies from mice immunized with the 25% NPs vaccines exhibited a significantly enhanced capacity to promote bacterial clearance compared to those from the Alum group.

In order to investigate the body protection ability of the nanovaccines, we challenged the mice with the lethal dose (1×LD₁₀₀) of S. aureus via vein tail administration at 35 days after the first immunization (Figure 7A). As respected, all mice in the PBS and free antigen groups died within 8 days (Supplementary Figure S6). In Alum adjuvant vaccine groups, the survival percentage was only 20% and 40% in Alum-rEsxA and Alum-rEsxB, respectively. Even in the Alum-rEsxA and Alum-rEsxB combined vaccination group, the survival percentage just reached 50% (Figures 7B–D). However, in 25% NPs adjuvant vaccines groups, no matter for single antigen loaded vaccine groups nor the combined vaccination group, the survival percentage was much higher than the corresponding Alum ones. Moreover, for 25% NPs-rEsxB and 25% NPs-rEsxA and 25% NPs-rEsxB combined vaccination groups, all animals survived in the lethal challenge experiment (Figure 7E).

Though significant differences were found in the results of antibody titer, IL-4, IFN-γ, IL-17A secretion and neutralizing capability of the antibody between the single antigen vaccine (25% NPs-rEsxB) and combine vaccine (25% NPs-rEsxA and 25% NPs-rEsxB) groups (Supplementary Figure S2), both of them got 100% survival percentage in the lethal challenge experiment. It was indicated by the results above that if the lethal dose increases, the protection ability between the single antigen vaccine and the combine one could be differentiated.

Thus, we further double the challenge dose of S. aureus to test the difference of protection ability between the single antigen loaded and the combined vaccine groups when 25% NPs served as adjuvants. Undoubtedly, the animals of all PBS and free antigen vaccination groups died out within 5 days after challenge (Supplementary Figure S7). Moreover, the Alum vaccine groups, either for single antigen loaded or the combined vaccines groups, the animals died out within 8 days (Figures 7F–H). In contrast, some of the animals in 25% NPs vaccine groups kept alive till the end of the challenge experiment. 20% and 30% of the animals in 25% NPs-rEsxA and 25% NPs-rEsxB were alive, while the survival percentage reached 80% in 25% NPs-rEsxA and 25% NPs-rEsxB combined vaccination group (Figure 7I).

The results of lethal challenge experiments reflected the superior body protection ability of the 25% NPs vaccines, especially for the 25% NPs-rEsxA and 25% NPs-rEsxB combined vaccination route. And these results were also consistent with the findings in the antibody titer and the cytokine secretion studies.