V. parvula cultures were prepared according to previously established protocols (Liu et al., 2020). OMVs were isolated utilizing a standardized methodology using the ExoBacteria OMV Isolation Kit (System Biosciences, Pal Alto, CA, USA) (Micoli and Maclennan, 2020). The bacterial concentration was determined with a spectrophotometer at an optical density of 600 nm. After the initial centrifugation at 6,000 × g for 30 min, the supernatant was collected and filtered through a 0.22-μm membrane to ensure complete bacterial cell removal. The filtrate was then transferred into an ultracentrifuge tube and subjected to ultracentrifugation at 100,000 × g for 80 min at 4°C, during which grayish-yellow OMVs were deposited on the bottom of the tube. The supernatant was discarded and the pellet resuspended in phosphate-buffered saline (PBS) before being centrifuged again under identical conditions. The purified OMVs were then resuspended in 1 ml pre-cooled sterile PBS, filtered through a 0.22-μm membrane for concentration, aliquoted, and stored at −80°C to prevent freeze–thaw damage. The protein concentrations of the OMV fractions were confirmed using a BCA protein assay kit (Thermo Fisher Scientific, Waltham, MA, USA). The morphology of the V. parvula OMVs was assessed using a transmission electron microscope (TEM) (Thermo Fisher Scientific), and the diameter was determined through dynamic light scattering analysis (Malvern Zetasizer Nano ZS90; Malvern Panalytical, Malvern, UK).

Neutrophils were extracted from the bone marrow of 8-week-old male C57BL/6J mice, which were obtained from Suzhou Charles River Laboratory Animal Co., Ltd. (SH9H-2020-A220-1; Suzhou, Jiangsu, China). The isolation process employed discontinuous Percoll gradients (GE Healthcare Co., Princeton, NJ, USA), as previously described (Hendricks et al., 2017). C57BL/6J mice were euthanized and the tibias/femurs were isolated. Marrow cavities were flushed with serum-free RPMI medium, filtered through a 70-μm mesh, and centrifuged. The cell pellets were resuspended before layering onto preformed Percoll gradients (78%, 65%, and 55%). After gradient centrifugation at 800 × g without braking, the 65%–78% interface was collected and washed and the erythrocytes were lysed. The final pellets were resuspended in supplemented RPMI medium and incubated under standard cell culture conditions.

The viability and purity of the neutrophils were assessed using Wright–Giemsa staining and flow cytometry (FCM) analysis. FCM analysis involved staining the cells with PerCP–cy5.5–CD11b antibody (Ab) (clone 1A8; BioLegend, San Diego, CA, USA) and PE-Ly6G Ab (clone 1A8; BioLegend) prior to any experimental procedures.

For the experimental setup, the neutrophils (5 × 10⁵/well) were seeded for a 0.5-h incubation before the addition of V. parvula OMVs (50 μg/ml). Subsequently, the neutrophils were cultured for 4 h under the same conditions. In specific experiments, the neutrophils were pretreated for 1 h with the NET inhibitor GSK484 (HY-100514; MCE, Monmouth Junction, NJ, USA) at 100 nmol/L or the CYBB inhibitor apocynin (HY-N0088, MCE) at 100 μmol/L prior to stimulation with V. parvula OMVs. The experimental groups were designated as follows: the control group, the V. parvula OMV group, the V. parvula OMV+GSK484 group, and the V. parvula OMV+apocynin group.

Neutrophils (5 × 10⁵/well) that were either stimulated with V. parvula OMVs or left unstimulated for 4 h were then seeded onto the Transwell upper chamber (3 μm; Corning Inc., Corning, NY, USA) for 2 h at 37°C with 5% CO₂ to evaluate their migration toward the lower chamber. Tissue culture media dissolved with interleukin 8 (IL-8; 20 ng/ml) were added to the basolateral side of each well. The schematic diagram is shown in Figure 1A . Following a 2-h incubation period, the non-migratory neutrophils in the upper chamber were cleaned using a small cotton swab. Neutrophils that migrated on the membrane were fixed with a dye fixing solution (Beyotime Co., Shanghai, China) and subsequently stained with a crystal violet staining solution (Beyotime). The resulting images were analyzed using optical microscopy, and the average number of migrated neutrophils per high-power field (HPF) was calculated from six different microscope views.

All V. parvula were labeled with fluorescein isothiocyanate (FITC) to facilitate detection by FCM. The chemical surface labeling process involved culturing bacteria as previously described (Fine et al., 2023), followed by centrifugation and labeling for 1 h at 37°C in 250 µg/ml FITC in an anaerobic chamber.

Neutrophils (1 × 10⁵ cells/ml) that were either stimulated with V. parvula OMVs or left unstimulated for 4 h were cultured with a final volume of 100 µl per well. Thereafter, 20 µl of fluorescent bacteria (1 × 10⁸ CFU/ml) was appended to achieve a bacteria-to-cell ratio of 100:1. The reaction mixture was cultured for an additional hour. The mean fluorescence intensity (MFI) of the FITC⁺ neutrophils was recorded. For each sample, 50,000 events were collected. All FCM data were quantified and analyzed using FlowJo version 10.9 software (BD FlowJo, Ashland, Oregon, USA).

Neutrophil apoptosis was evaluated using FCM. Apoptotic cells were identified using the Annexin V/PI Apoptosis Kit (Beyotime), as described in previous studies (Ou et al., 2022).

The neutrophils were washed three times after V. parvula OMV stimulation for 4 h and labeled with 1 ml propidium iodide (PI; 50 mg/L). The PI-stained neutrophils were subsequently tagged with 4 ml Annexin V. FCM analysis counted 10,000 events to determine the number of apoptotic cells. The data obtained from FCM were analyzed using FlowJo version 10.9 software (BD FlowJo).

Neutrophils (2 × 10⁵ cells/ml) that were either stimulated with V. parvula OMVs or left unstimulated for 4 h were assessed using a TUNEL kit (40308ES20; Yeasen Biotech, Shanghai, China) and SYTOX Green (KGA261; KeyGEN BioTECH, Nanjing, China) following the manufacturers’ instructions. The neutrophils were fixed with a dye fixing solution, treated with proteinase K to reactivate antigens, incubated with an equilibrate buffer, and then labeled with the TUNEL detection cocktail. Afterward, SYTOX Green and DAPI (Beyotime) were applied. After staining for 15 min, TUNEL and SYTOX Green staining was observed under Zeiss LSM880.

Following stimulation for 4 h, neutrophils (2 × 10⁵ cells/ml) that were either stimulated with V. parvula OMVs or left unstimulated were incubated with the ROS probe DCFH-DA (10 μmol/L; Beyotime) for 20 min. Subsequently, the neutrophils were washed three times with serum-free RPMI medium to thoroughly remove the DCFH-DA that had not entered the cells. The expression of ROS in neutrophils was directly observed using a Leica Microsystems microscope and analyzed with the LAS X software (Leica, Wetzlar, Germany). The neutrophils were pipetted to dislodge them from the surface, collected, and analyzed by FCM. The MFI was subsequently determined using FCM. The neutrophils in the V. parvula OMVs+GSK484 and V. parvula OMVs+apocynin groups were also incubated with the ROS probe DCFH-DA.

RNA sequencing was evaluated in neutrophils with or without V. parvula OMVs stimulation for 4 h by Majorbio Bio-Pharm Technology (Shanghai, China). The total neutrophil RNA was derived using the TRIzol reagent (Takara, Kusatsu, Shiga, Japan). Fragments per kilobase of exon per million mapped (FPKM) were applied to normalize the gene expression levels. The FPKM data were used to elucidate the relationship between the control group and the V. parvula OMVs group using principal component analysis (PCA) and hierarchical clustering analysis. Changes in the differentially expressed genes (DEGs) were identified using edgeR with the general filtering criteria of |log2 fold change| > 1.5 and p < 0.05 applied to the FPKM values. The DEGs were enriched using Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) annotations to identify the relevant pathways.

The neutrophil total RNA was extracted using the TRIzol reagent. The concentration and the purity of the extracted RNA were assessed to ensure quality. Complementary DNA (cDNA) was synthesized using the PrimeScript RT Master Mix (Takara) according to the manufacturer’s instructions. Reverse transcription polymerase chain reaction (RT-PCR) assays were performed by amplifying each cDNA sample with SYBR Premix Ex Taq (Yesen, Shanghai, China) on an ABI Real-Time PCR 7300 System (Roche, Basel, Switzerland). The relative target gene expression levels were normalized to that of β-actin. The primer sequences are presented in Table 1 .

Protein was extracted and was measured using a BCA protein assay kit. Equal cellular protein was subjected to SDS-PAGE (10%) and then transferred into polyvinylidene fluoride (PVDF) membranes (Millipore, Billerica, MA, USA). After blocking with 5% non-fat dry milk for 1 h as previously described (Qiu et al., 2021), the membranes were incubated overnight at 4°C with primary antibodies. Finally, the membranes were incubated with the corresponding horseradish peroxidase (HRP)-linked secondary antibodies and then scanned using an enhanced chemiluminescence reagent (Thermo Fisher Scientific).

To assess the ICAM-1 or the myeloperoxidase (MPO) expression in neutrophils, 1 × 10⁶ neutrophils with or without V. parvula OMVs stimulation for 4 h were washed and stained with the PerCP–cy5.5 anti-mouse–CD11b antibody, the PE anti-mouse Ly-6G antibody, the anti-rabbit ICAM-1 antibody (CD54; Proteintech, Wuhan, China), and/or the MPO antibody (HA601249; HUABIO Co., China; Cell Signaling Technology, Danvers, MA, USA). The neutrophils were subsequently incubated with fluorescent-labeled secondary antibodies. Unstained cells served as negative controls, and single-color positive controls were made respectively. FCM data were determined on 60,000 events and were analyzed using FlowJo version 10.9 software. The ICAM-1- or MPO-expressing neutrophils were determined as CD11b⁺Ly6G⁺ICAM-1⁺ or CD11b⁺Ly6G⁺MPO⁺ populations, respectively.

Neutrophils with or without V. parvula OMVw stimulation for 4 h were fixed, permeabilized for 15 min, and blocked using an immunostaining blocking solution (Beyotime). Subsequently, the neutrophils were stained with the anti-rabbit ICAM-1 antibody and the anti-goat MPO antibody (R&D Systems, Minneapolis, MN, USA) at 4°C overnight. Coverslips were mounted onto glass slides using an anti-fluorescence quencher containing DAPI (Beyotime) after incubation with the secondary antibodies with fluorescent labeling for 1 h. The images were observed using Zeiss LSM880.

The morphology of the NETs induced was visualized by incubating the cells with the primary anti-goat MPO antibody and the anti-rabbit citrullinated H3 antibody (CitH3; ab5103, Abcam, Waltham, MA, USA). The images were observed using Zeiss LSM880.

To identify NETs, the circulating cell-free DNA (cf-DNA), neutrophil elastase (NE), and MPO were quantified in the supernatant, as previously described (Wang et al., 2023). Following V. parvula OMVs stimulation of the neutrophils for 4 h, the supernatant fractions were collected into 1.5-ml tubes. The supernatant tubes were centrifuged at 450 × g for 5 min at 4°C to pellet any detached neutrophils. The supernatant was then transferred to another 1.5-ml tube. Quantification of cf-DNA was performed with the Quant-iT PicoGreen ds-DNA Quantification Kit (Invitrogen, Carlsbad, CA, USA). The NE concentrations were determined using an NE ELISA kit (Yinxing Lab, Shanghai, China). MPO activity was assessed with an MPO Detection Kit (Nanjing Jiancheng Bioengineering Institute, Nanjing, China).

After stimulation of the neutrophils with V. parvula OMVs for 4 h, the supernatant fractions were quantified using a PAD4-specific ELISA Kit (Cusabio, Wuhan, China) in strict accordance with the manufacturer’s instructions. The detailed procedure was as follows: 100 μl of the standard product and the sample were respectively added into the preset standard well and the sample well and then incubated at 37°C for 2 h. Once all the liquid had been aspirated, biotin antibody (1×) was added and incubated at 37°C for 1 h. Subsequently, the wells were washed twice and incubated with HRP–avidin (1×) at 37°C for 1 h. The wells were then washed five times and 90 μl of the TMB substrate was added to each well. The wells were incubated in the dark at 37°C for 15–30 min. Subsequently, 50 μl of the stop solution was added to each well. Finally, the optical density of each well was measured within 5 min using a microplate reader set at 450 nm.

Experimental data were expressed as the mean ± standard error and analyzed using GraphPad Prism software (GraphPad Software, San Diego, CA, USA). For comparison between two groups, Student’s t-test was used. For comparison of more than three groups, one-way analysis of variance (ANOVA) was employed. Values of *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001 were defined as statistically significant.

The V. parvula OMVs showed spherical, bilayer cell portions in TEM, which revealed vesicle-like structures ( Figure 2A ). The nanoparticle tracking analysis (NTA) results demonstrated that the average diameter of the V. parvula OMVs was 125.7 ± 5.6 nm, similar to the typical OMVs diameters of other Gram-negative bacteria ( Figure 2B ).

The FCM analysis confirmed that the purity of the CD11b⁺Ly6G⁺ neutrophils after Percoll gradient separation exceeded 90% ( Figure 2C ). The Wright–Giemsa staining results revealed the neutrophil morphology, which showed segmentation in the nuclei ( Figure 2D ).

Neutrophils exhibited robust migration in response to V. parvula OMVs stimulation ( Figure 1B , red arrow indicating migrated neutrophils). The number of migrating neutrophils increased significantly in the V. parvula OMVs group ( Figure 1C ). We also examined whether the V. parvula OMVs have an effect on the expression of chemokines. The results showed that the mRNA expression levels of Cxcl1, Cxcl2, and Icam-1 were elevated in the V. parvula OMVs group ( Figure 1D ). The protein expression levels of CXCL1 and ICAM-1 in the neutrophils of the V. parvula OMVs group were notably increased ( Figure 1M ).

The phagocytosis analysis was based on the MFI of the FITC⁺ neutrophil population. The results revealed no significant changes in the V. parvula OMVs group ( Figure 1E ).

The frequency of neutrophil apoptosis was significantly increased in the V. parvula OMVs group compared with the control group ( Figures 1F, G ), indicating that neutrophil apoptosis was extremely enhanced by the V. parvula OMVs. In the control group, the neutrophils exhibited an intact nuclei (DAPI, blue). In the V. parvula OMVs group, apoptotic cells (red) showed cell shrinkage and membrane blebbing, while NETotic cells (green) displayed chromatin decondensation, cell swelling, and membrane rupture ( Figure 1H ).

Neutrophils exhibited significantly enhanced ROS levels in the V. parvula OMVs group ( Figure 1I ). The V. parvula OMVs elicited higher ROS production in neutrophils ( Figure 1J ). The FCM analysis revealed a mean fluorescence intensity (MFI) of 343,007 ± 102,664 relative fluorescence units (RFU) in the V. parvula OMVs group, which was substantially higher than the 36,081 ± 2,125 RFU of the control group ( Figure 1K ).

The Cybb mRNA expression level was increased significantly in the V. parvula OMVs group ( Figure 1L ). Correspondingly, the CYBB protein expression in the neutrophils of the V. parvula OMVs group was notably elevated, consistent with the observed mRNA expression patterns ( Figure 1M ).

To elucidate the interaction between the V. parvula OMVs and the neutrophils, RNA sequencing was performed. The PCA demonstrated consistency in the data ( Figure 3A ). The DEGs with increased expression were enriched in inflammatory processes (e.g., Il-1β and Tnf) and neutrophil chemotactic responses to stimulation (e.g., Cxcl1, Cxcl2, and Ccl3). The volcano plot revealed 640 upregulated and 2,148 downregulated mRNAs in the neutrophils of the V. parvula OMVs group compared with the control group ( Figure 3B ). In particular, the Icam-1 mRNA expression in neutrophils was significantly enhanced in the V. parvula OMVs group. A heatmap illustrating the differential gene expression patterns was generated ( Figure 3C ). KEGG and GO pathway enrichment analyses of the Icam-1 gene expression were conducted. The top 10 most significantly enriched GO terms indicated that Icam-1 expression is associated with extrinsic apoptotic pathways ( Figure 3D ). The top 10 most significantly enriched KEGG pathways ( Figure 3E ) included TNF signaling and various infection-related pathways.

Both the mRNA and protein levels of ICAM-1 in the neutrophils of the V. parvula OMVs group were markedly upregulated ( Figures 1D, M ).

FCM analysis of the ICAM-1 expression in neutrophils was performed. The gating strategy is shown in Figure 4A . A notable increase in ICAM-1⁺ neutrophils was observed in the V. parvula OMVs group ( Figure 4B ). To characterize the ICAM-1⁺ neutrophils in terms of their NET formation capacity, the expression of MPO was analyzed in the V. parvula OMVs group using FCM. Stimulation with V. parvula OMVs resulted in significantly higher MPO expression levels ( Figure 4C ). Fluorescence microscopy revealed the higher expression of MPO and ICAM-1 in the neutrophils of the V. parvula OMVs group than that in the control group. ( Figure 4D ). These findings indicate that ICAM-1⁺ neutrophils exhibit enhanced NET formation. The gating strategy for the FCM analysis of the MPO expression in ICAM-1⁺ neutrophils is shown in Figure 4E . It was demonstrated that ICAM-1⁺ neutrophils had significantly higher levels of MPO expression compared with ICAM-1⁻ neutrophils ( Figure 4F ). Surface-bound extracellular MPO serves as a marker for NETs, suggesting that ICAM-1⁺ neutrophils improved NET formation.

The formation of NETs in response to V. parvula OMVs was quantified by the cf-DNA, NE, and MPO expression. V. parvula OMVs led to enhanced cf-DNA production. The control group exhibited lower cf-DNA concentrations (465 ± 55.15 ng/ml) compared with the V. parvula OMV group (1,021 ± 45.12 ng/ml) ( Figure 5A ). The NE levels were also elevated in the V. parvula OMV group (5.145 ± 1.861 mg/ml) relative to the control group (3.285 ± 0.2256 mg/ml) ( Figure 5B ). The MPO expression was markedly increased in the V. parvula OMVs group (11.13 ± 2.788 U/g protein) compared with the control group (4.523 ± 1.880 U/g protein) ( Figure 5C ). The immunofluorescence (IF) assays corroborated these findings, showing enhanced CitH3 and MPO expression, indicative of the increased NET formation in response to V. parvula OMVs stimulation ( Figure 5H ).

The formation of NETs is closely related to the NADPH oxidase-derived ROS. Figures 1I , 1J indicated that the V. parvula OMVs heightened the ROS production in active neutrophils. The Cybb gene encodes gp91-phox, the catalytic subunit of NADPH oxidase 2 (NOX2), which generates ROS. Notably, Figures 1L , 1M showed that both mRNA and protein expression of CYBB were significantly increased in neutrophils following stimulation with V. parvula OMVs. Moreover, the elevated expression levels of the CYBB protein ( Figure 5D ) and the Cybb mRNA ( Figure 5E ) induced by V. parvula OMVs were attenuated by the CYBB inhibitor apocynin. NET formation, including the cf-DNA, NE, and MPO expression ( Figures 5A–C ), and the increased ICAM-1⁺ neutrophils ( Figure 5G ) induced by V. parvula OMVs were also attenuated in the V. parvula OMVs+apocynin group, suggesting that the effect of V. parvula OMVs on ICAM-1⁺ neutrophils was mediated by ROS production. In addition, the higher ROS level in the V. parvula OMVs group substantially declined in the V. parvula OMVs+apocynin group (p < 0.001; Figures 5I, J ).

To verify the NET formation effect of the V. parvula OMVs on ICAM-1⁺ neutrophils, the NET inhibitor GSK484 was used. This resulted in a significant decrease of the cf-DNA, NE, and MPO expression in the V. parvula OMVs+GSK484 group compared with the V. parvula OMVs group ( Figures 5A–C ). The neutrophils in the V. parvula OMVs+GSK484 group demonstrated no elevated levels of the CitH3 protein expression compared with the control group ( Figure 5D ). Correspondingly, the unregulation of the Pad4 mRNA expression in neutrophils was attenuated in the V. parvula OMVs+GSK484 group ( Figure 5F ). The IF assays corroborated these findings, showing that the increased NET formation was reduced in the V. parvula OMVs+GSK484 group ( Figure 5H ). Notably, the administration of GSK484 reversed the V. parvula OMVs-induced increase in ICAM-1⁺ neutrophils, effectively reducing the frequency of this cell population in the V. parvula OMVs+GSK484 group ( Figure 5G ). Simultaneously, the expression levels of the CYBB protein ( Figure 5D ) and the Cybb mRNA ( Figure 5E ) induced by V. parvula OMVs were attenuated in the V. parvula OMVs+GSK484 group. The excessive ROS production in neutrophils after V. parvula OMVs stimulation was attenuated in the V. parvula OMVs+GSK484 and V. parvula OMVs+apocynin groups ( Figures 5I, J ). In addition, there was a significant decrease in the expression of PAD4 in the V. parvula OMVs+GSK484 and V. parvula OMVs+apocynin groups compared with the V. parvula OMVs group ( Figure 5K ). These results suggest that the possible mechanism through which V. parvula OMVs increase ICAM-1⁺ neutrophils along with elevated NET formation was mainly through the ROS–PAD4 signaling pathway.