Isogenic P.aeruginosa strains-wild-type PAO1, hyper-biofilm mutant PAO1ΔwspF (characterized by elevated intracellular c-di-GMP), and hypo-biofilm mutant PAO1/plac-yhjH (with reduced c-di-GMP levels) were obtained from the Centre for Environmental Life Sciences Engineering, Nanyang Technological University, Singapore (Chua et al., 2013, 2015). All strains were cryopreserved at -80 °C in lysogeny broth (LB; Guangdong Huakai Microbial Technology Co., Ltd., China) supplemented with 25% (v/v) glycerol. For plasmid maintenance, 60 μg/mL tetracycline (Sigma-Aldrich) was added to PAO1/plac-yhjH cultures. Frozen stocks were streaked onto LB agar plates (Beijing Lanqiao Technology Co., Ltd., China) using sterile loops and incubated at 37 °C for 24 h (HPX-400 incubator, Shanghai Yuejin Medical Equipment Co., Ltd.). Single colonies were then inoculated into LB broth and cultured aerobically at 37 °C with 220 rpm agitation for 18 h (THZ-82 shaker, Changzhou Zhibo Instrument Co., Ltd.). Bacterial cells were harvested by centrifugation at 6,000rpm for 3 min (Z366 centrifuge, HERMLE Labortechnik GmbH, Germany), followed by three washes with phosphate-buffered saline (PBS). Bacterial density was standardized to OD600 = 0.1 (10^8 CFU/mL) using a UV-Vis spectrophotometer (T6 model, Beijing Pushi General Instrument Co., Ltd.), with final inocula adjusted to 5×10^7 CFU/mL in sterile PBS for subsequent experiments (20).

P.aeruginosa reporter strains PAO1/pcdrA-gfp(PAO1 carrying the pcdrA-gfp report), PAO1ΔwspF/pcdrA-gfp, and PAO1/plac-yhjH/pcdrA-gfp were obtained from the Centre for Environmental Life Sciences Engineering at Nanyang Technological University, Singapore. As reported (21), GFP expression from the pcdrA-gfp reporter positively correlates with intracellular c-di-GMP levels: pcdrA-gfp reporters in P.aeruginosa, intracellular c-di-GMP levels positively correlate with GFP expression: high c-di-GMP boosts GFP expression, while low c-di-GMP decreases it. Strains were maintained as described previously, with antibiotic selection as follows (21, 22): 50µg/mL carbenicillin for PAO1/pcdrA-gfp and PAO1ΔwspF/pcdrA-gfp, and 50 µg/mL carbenicillin plus 10 µg/mL tetracycline for PAO1/plac-yhjH/pcdrA-gfp. Bacterial suspensions standardized to OD600 = 0.1 were aliquoted (200 µL/well) into 96-well microplates in six technical replicates and incubated statically at 37°C for 24 h. Biofilm formation was quantified by measuring both optical density (OD600) and GFP fluorescence intensity (ex/em: 485/535 nm) using a Tecan Spark multi-mode microplate reader (Tecan Group Ltd., Austria). Relative fluorescence intensity (RFI) was calculated as RFI/OD600 to normalize reporter activity against bacterial biomass with three biologically independent experiments performed to ensure reproducibility.

As previously described, P.aeruginosa strains PAO1, PAO1ΔwspF, and PAO1/plac-yhjH were adjusted to 1×10^8 CFU/mL in sterile PBS, and plate coating was performed to verify the colony count. The standardized bacterial suspension (200 µL/well) was aliquoted into 96-well plates (6 technical replicates per strain) and incubated statically at 37°C for 24 h. After incubation, the culture medium in each well was discarded, and wells were washed 3 times with 200 µL sterile PBS, then air-dried at room temperature. Each well was stained with 210 µL of 0.1% crystal violet solution for 30 min, followed by gentle rinsing with deionized water to remove excess stain. After re-drying, 220 µL of absolute ethanol was added to each well, and the plate was incubated with gentle shaking for 30 min to elute the stain. Absorbance was measured at OD595.

Nine male C57BL/6J mice (8–10 weeks old, with a body weight of 22–25 g) were obtained from the Laboratory Animal Center of Guangxi Medical University and maintained under specific pathogen-free (SPF) conditions. Mice were randomly allocated into three experimental groups (n=3/group) using a stratified randomization protocol based on body weight: PAO1, PAO1ΔwspF, and PAO1/plac-yhjH. Mice were anesthetized with isoflurane, and aseptic techniques were strictly followed during the subsequent intrathoracic injection procedure: the injection site on the thoracic skin was disinfected with 75% ethanol, and a sterile 30G needle was used to inoculate 100 μL bacterial suspension (5×10^7 CFU, colony-forming unit) at the 7th-8th intercostal space. Using autoclaved surgical instruments, the thoracic cavity was opened gently. Thoracic lavage fluid was collected 24 h post-infection following euthanasia: sterile PBS (0.5 mL) was injected and aspirated twice to maximize fluid recovery, and immediately processed for subsequent analyses. Thoracic lavage fluid samples were serially diluted (10⁵ and 10⁶ in PBS), and 20μL aliquots from each dilution were plated in triplicate on LB agar. Plates were incubated at 37 °C for 18–24 h. P.aeruginosa colonization was confirmed by colony morphology. All procedures were approved by the Animal Ethics Committee of Guangxi Medical University (Approval No. 202312014). Mice were allocated into three experimental groups (n=3/group) using a stratified randomization protocol based on body weight: PAO1, PAO1ΔwspF, and PAO1/plac-yhjH. Mice were anesthetized with isoflurane and inoculated via intrathoracic injection of 100μL bacterial suspension (5×10^7 CFU) using a 30G needle at the 7th-8th intercostal space. Thoracic lavage fluid was collected 24 h post-infection following euthanasia: PBS (0.5 mL) was injected and aspirated twice to maximize effusion recovery, and immediately processed for subsequent analyses. Thoracic effusion samples were serially diluted (10⁵and 10⁶ in PBS), and 20μL aliquots from each dilution were plated in triplicate on LB agar. Plates were incubated at 37°C for 18–24 h. The presence of P.aeruginosa in the thoracic cavity was confirmed by enumeration of CFU following plating of thoracic lavage fluid. All procedures were approved by the Animal Ethics Committee of Guangxi Medical University (Approval No. 202312014).

At 24 h post-P.aeruginosa pleural infection, mice were euthanized by excessive anesthesia, and intercostal parietal pleural tissues were collected. Tissues were fixed in 4% paraformaldehyde (4°C, 24 h), dehydrated in gradient ethanol (70%, 80%, 90%, 100%, 15 min each), cleared in xylene (twice, 10 min each), and paraffin-embedded into 5 μm serial sections. Post-dewaxing to water, sections were stained with hematoxylin (5 min), rinsed to blue with tap water, differentiated in 1% hydrochloric acid-ethanol (30 s), re-rinsed, then eosin-stained (2 min); followed by gradient ethanol dehydration (90%, 100%, 5 min each), xylene clearing (twice, 10 min each), and neutral gum mounting. Pleural pathological changes (inflammatory cell infiltration, tissue damage) were observed via a Nikon Eclipse E100 light microscope, with 5 random fields per group photographed.

For detecting inflammatory cytokines (IL-6, IL-1β, TNF-α) in pleural lavage fluid, specific ELISA kits (JONLNBIO, Mouse IL-6: JL20268; IL-1β: JL18442; TNF-α: JL10484) were used per manufacturers’ protocols. Briefly, pleural lavage fluid was centrifuged at 1000×g for 20 min at 4°C to remove debris, and supernatants were collected. Standards and samples were added to pre-coated microplates, incubated at 37°C, washed 5 times, then HRP-conjugated secondary antibody was added, incubated and re-washed. TMB triggered color development, terminated with termination solution. OD values were measured at 450 nm, and cytokine concentrations were calculated via standard curves.

To assess biofilm formation in vivo and vitro, three P. aeruginosa strains were adjusted to an optical density at 600 nm (OD₆₀₀) of 0.1, then statically incubated in a 37 °C incubator for 24 h. Thoracic lavage fluid from P. aeruginosa-infected mice (24 h post-infection) was centrifuged at 500 × g for 5 min to remove large debris; 200 μL of the supernatant was incubated on 13 mm×13 mm sterile glass coverslips for 1 h at 37 °C, gently washed with 0.9% NaCl (Solarbio, China) 3 times for 10min, stained with Viable/Dead Kit 30 (Solarbio, China; green: viable, red: dead), imaged using a Leica TCS SP8 STED confocal microscope (oil objective) with Nucgreen (green, Ex 503 nm, Em 530 nm) and EthD-III (red, Ex 530 nm, Em 620 nm) parameters, Z-stack images acquired from base to top, and thickness quantified via LAS X Office 3D plugin (3 fields/sample, 3 replicates, average calculated).

At 24 h post-P.aeruginosa pleural infection, mice were euthanized by excessive anesthesia; intercostal parietal pleura was excised with autoclaved instruments, immediately fixed in 2.5% glutaraldehyde (4°C, 24 h), washed thrice with 0.1 mol/L phosphate buffer (pH 7.4, 15 min each), secondarily fixed in 1% osmium tetroxide (room temperature, 1–2 h) and re-washed, dehydrated in gradient ethanol (30%-100%, 15–20 min each; 100% twice), replaced with tert-butanol, frozen at -20°C for 30 min, dried via freeze dryer (SCD-350M, Shi’anJia, China), coated with 5–10 nm gold film (ion sputtering coater JY-S100, Guangzhou Jingying, China), and observed under SEM4000X (5 kV) to analyze bacterial aggregation and extracellular matrix deposition.

Mouse pleural lavage fluid was centrifuged (3000rpm, 15 min, 4°C) to separate pellets, which were then washed three times with sterile PBS; RNA was extracted from the pellets using Trizol (Servicebio, China), and its quality was determined (RIN>7.0, OD260/280≥1.8). RNA sequencing of mouse pleural lavage fluid cells (n=3 mice/group) was carried out using Illumina NovaSeq X Plus (150 bp paired-end, Gene Denovo, Guangzhou). Raw reads were quality-controlled first, and clean reads were subjected to bioinformatics analysis via Omicsmart, a dynamic real-time interactive online platform(http://www.omicsmart.com). Gene expression was quantified (FPKM) and differentially expressed genes analyzed (DESeq2, v1.38.3; |log2FC| >2, FDR <0.05). Differentially expressed genes were subjected to clustering analysis using the Mfuzz package, retaining genes with membership values >0.2. These genes were further classified via Mfuzz (fuzzy C-means clustering) to identify co-expression modules.

Using Mfuzz package, we identified gene expression trend modules based on dynamic patterns across different c-di-GMP expression levels of P.aeruginosa. Persistently downregulated modules were selected, and their genes were used for subsequent PPI networks analysis. These networks were built via STRING (score>0.4). After removing isolated nodes, the network was imported into Cytoscape 3.10.3. The CytoHubba plugin was used to calculate node centrality via Maximum Clique Centrality (MCC), and the top 20 genes by descending scores were selected to build the core subnetwork. Gene clusters exhibiting consistent expression trajectories were functionally annotated using DAVID database for Gene Ontology (GO) terms and Kyoto Encyclopedia of Genes and Genomes (KEGG) database for pathway enrichment. False Discovery Rate (FDR) less than 0.05 was considered statistically significant.

Candidate genes from differential expression analysis were validated via quantitative reverse transcription PCR (qRT-PCR). pleural lavage fluid was centrifuged (3000rpm, 15 min, 4°C) to separate pellets, the cell pellet was washed three times with PBS to remove residual supernatant. Briefly, total RNA was extracted from pleural lavage fluid cells using Trizol (Servicebio, China), with quality verified via Nanodrop (Thermo Fisher Scientific, USA; A260/280: 1.8–2.0). cDNA synthesis was performed using a reverse transcription kit (Takara Bio, Inc.).Gene-specific primers were designed and synthesized by Gensys Biotechnology (Shanghai, China). Reactions were conducted in 20μL reactions with 2×SYBR Green Fast qPCR Master Mix (Yifeixue, Nanjing, China), and the thermal cycling protocol followed the manufacturer’s instructions. Gapdh was used as the endogenous control for normalization, and relative expression levels were calculated via the 2−ΔΔCt method. Primer sequences are provided in Supplementary Table S1.

As previously described, bone marrow neutrophils were isolated from C57BL/6J wild-type mice (23, 24). Briefly, bone marrow cells were isolated from the femurs and tibias via density gradient centrifugation using Histopaque 1119 (Sigma, St. Louis, MO, USA) and Histopaque 1077 (Sigma, St. Louis, MO, USA). The intermediate layer between Histopaque 1119 and Histopaque 1077 was collected, followed by lysis of erythrocytes with RBC lysis buffer (Solarbio, China). All procedures were performed carefully and rapidly on ice under sterile conditions to maintain cell viability and prevent cell activation.

Freshly isolated primary mouse bone marrow neutrophils were resuspended in RPMI 1640(Gibco, USA)medium supplemented with 10% fetal bovine serum(Cellbox, China) and seeded into poly-L-lysine-precoated confocal(Servicebio, China) dishes at 1×10⁶ cells/well. After incubation at 37°C with 5% CO₂ for 30 min to allow adhesion, three P.aeruginosa strains (PAO1 wild-type, PAO1△wspF, PAO1/plac-yhjH) were adjusted to OD600 = 0.1 and added at a multiplicity of infection (MOI) of 100:1; following stimulation at 37 °C with 5% CO₂ for 3 h, non-adherent bacteria were removed by PBS washing (5 min×3). For immunofluorescence staining (applied to both the above in vitro-stimulated neutrophils and in vivo mouse pleural effusion samples), cells were fixed with 4% paraformaldehyde (room temperature, 15 min), blocked with 10% goat serum (Solarbio, China; 37 °C, 1 h), and incubated sequentially with primary antibodies (Abcam: anti-Histone H3 (citrulline R2+R8+R17, ab281584, 1:2000); anti-Myeloperoxidase, ab90810, 1:50) at 4°C overnight and fluorescent secondary antibodies (Abcam: goat anti-rabbit IgG H&L (Alexa Fluor^® 594, ab150084); goat anti-mouse IgG H&L (Alexa Fluor^® 488, ab150117); both 1:400) at 37°C for 1 h in the dark. After PBS washing (5 min×3), cells were stained with DAPI (Solarbio, China) for 10 min in the dark, and images were acquired using a Leica TCS SP8 STED confocal microscope (excitation wavelengths: 405 nm for DAPI, 488 nm for Alexa Fluor^® 488, 594 nm for Alexa Fluor^® 594). Under a laser confocal microscope with an oil immersion objective, 5 non-overlapping fields were randomly selected per group. Total neutrophils were defined as MPO⁺DAPI⁺ cells. NETs-positive cells were identified as cells with nuclear fragmentation and extracellular fibrillar citH3⁺MPO⁺ colocalization (excluding intracellular-only signals). The percentage of NET-releasing cells was calculated as (number of NETs-positive cells/total neutrophils) × 100%.

The concentration of myeloperoxidase-DNA (MPO-DNA) complexes in murine pleural effusion supernatants was quantified using a Mouse MPO-DNA Complex commercial enzyme-linked immunosorbent assay (ELISA) kit (F31066-A, FANKEW, Shanghai, China), Cell-free DNA (cfDNA) was quantified using the Quant-iT™ PicoGreen™ dsDNA Assay Kit (Cat. No. P7589, Invitrogen, USA) according to the manufacturer’s instructions.

All quantitative data are presented as mean ± standard deviation (SD). Intergroup differences were analyzed using one-way analysis of variance (ANOVA) followed by Tukey’s post hoc test for normally distributed data with homogeneous variance; for non-normal data, the Kruskal-Wallis H test was used, with Dunn’s test (Bonferroni-corrected) for pairwise comparisons when significant overall differences were found. All analyses were implemented in GraphPad Prism version 9.5. Statistical significance was defined as p <0.05.

In vitro analyses of three P. aeruginosa strains (Supplementary Figures S1A, B) showed: via pcdrA-gfp reporter, PAO1△wspF had the strongest GFP fluorescence and highest c-di-GMP, followed by wild-type PAO1 and PAO1/plac-yhjH; via crystal violet staining (Supplementary Figure S1C), their biofilm biomass exhibited the same hierarchy (PAO1△wspF > PAO1 > PAO1/plac-yhjH), with significant pairwise differences for both parameters. These trends matched confocal microscopy results of their in vitro biofilms (Supplementary Figure S2): PAO1△wspF formed the densest biofilms, followed by wild-type PAO1, and PAO1/plac-yhjH the least.

Figure 1 shows representative thoracic cavity dissection images of mice in different groups: the P. aeruginosa-infected groups (including PAO1, PAO1△wspF, and PAO1/plac-yhjH strains) had obvious purulent secretions, while the PBS control group had none. In HE-stained sections (Figure 2), compared with the PBS control group, the P. aeruginosa-infected groups showed obvious parietal pleura thickening, accompanied by extensive inflammatory cell infiltrations. The control group exhibited a clear and thin pleural structure with regularly arranged cells and no obvious inflammatory cell aggregation; in contrast, the infected groups presented increased pleural tissue thickness, with massive inflammatory cell infiltrations in the interstitium, suggesting that P. aeruginosa infection could induce pleural inflammatory responses and histomorphological changes. However, no morphologic differences identifiable by naked eye were observed among the three P. aeruginosa-infected groups.

Quantitative analysis via CLSM (Figure 3) revealed obvious hierarchical differences in biofilm thickness among the tested strains: the PAO1ΔwspF strain formed the thickest biofilms, followed by the wild-type PAO1 strain, while the PAO1/plac-yhjH strain formed the thinnest biofilms. This result was further confirmed by SEM observation of pleural tissues (Figure 4): the degree of bacterial aggregation and the pattern of extracellular matrix deposition in SEM images were completely consistent with the order of biofilm formation quantified by CLSM (PAO1ΔwspF > PAO1 > PAO1/plac-yhjH). Specifically, the PAO1ΔwspF strain formed dense bacterial aggregates on the surface of pleural tissues, surrounded by a large amount of flocculent extracellular matrix; in contrast, only scattered bacteria were observed in the PAO1/plac-yhjH strain group, with extremely low content of extracellular matrix. Collectively, the morphological evidence from CLSM and SEM confirmed that the infection process of P.aeruginosa in the mouse pleural cavity completed the transition from a planktonic state to a biofilm-associated infection state within 24 hours.

At 24 h post P. aeruginosa infection, ELISA quantified IL-6, IL-1β, and TNF-α in pleural effusion (Supplementary Figure S3). Compared to the Control group, all infected groups (PAO1△wspF, PAO1, PAO1/plac-yhjH) showed significantly elevated IL-6, with a gradient of PAO1/plac-yhjH > PAO1 > PAO1△wspF. IL-1β was negligible in Controls but markedly increased in all infected groups; PAO1 and PAO1/plac-yhjH had similar IL-1β levels but both were significantly higher than PAO1△wspF (P < 0.0001). For TNF-α, Controls showed no significant differences from PAO1△wspF or PAO1/plac-yhjH, while PAO1 had significantly higher TNF-α than both, with no difference between PAO1△wspF and PAO1/plac-yhjH.

Through integrated RNA sequencing and bioinformatics analyses, we systematically characterized host transcriptomic responses using principal component analysis (PCA), volcano plots, and functional enrichment strategies. Comparative transcriptome profiling identified c-di-GMP-dependent regulatory networks modulating host-pathogen interactions, with particular emphasis on biofilm-associated immune evasion mechanisms and metabolic reprogramming events. PCA was performed to assess the overall gene expression profiles of P.aeruginosa strains in mice with varying levels of c-di-GMP. The three experimental groups, namely PAO1, PAO1ΔwspF, and PAO1/plac-yhjH, showed distinct clustering in the 3D PCA plot (Figure 5A). The first principal component (PC1) accounted for 88% of the variance, indicating a strong separation between the strains. PAO1 (yellow) clustered separately from PAO1ΔwspF (blue) and PAO1/plac-yhjH (red), suggesting significant differences in gene expression across these strains. Differential gene expression analysis (Figure 6B) revealed a significant number of differentially expressed genes (DEGs) between the experimental groups. In the comparison between PAO1ΔwspF and PAO1 (Panel 1), 98 genes were upregulated (orange dots) and 31 genes were downregulated (green dots). In the comparison between PAO1 and PAO1/plac-yhjH (Panel 2), 5 genes were significantly upregulated, while 3 genes were downregulated. Finally, the comparison between PAO1ΔwspF and PAO1/plac-yhjH (Panel 3) identified a substantial number of DEGs, with 1376 genes upregulated and 2733 genes downregulated (FDR < 0.05, |log2FC| > 2). Volcano plots display differential gene expression between WT-wspF vs WT-PAO1 (Figure 5C), WT-PAO1 vs WT-yhjH (6D), and WT-wspF vs WT-yhjH (6E) (FDR < 0.05, |log₂FC| > 2), with the top 10 upregulated and downregulated genes labeled. These results indicate that varying levels of c-di-GMP in P. aeruginosa significantly influence the host gene expression profile in mouse thoracic lavage fluid, highlighting the regulatory role of P. aeruginosa’s c-di-GMP in shaping host gene expression under pleural infection conditions.

Differential gene expression analysis revealed significant transcriptional remodeling across PAO1ΔwspF, PAO1, and PAO1/plac-yhjH groups. To further investigate the interrelationships among the three groups, we conducted Mfuzz clustering analysis to delineate dynamic gene expression patterns. Mfuzz clustering identified 10 distinct gene clusters across the three infection groups (PAO1△wspF, PAO1, and PAO1/plac-yhjH) (Figure 6). Differential gene expression analysis revealed significant transcriptional remodeling across the PAO1ΔwspF, PAO1, and PAO1/plac-yhjH groups. To further investigate the interrelationships among these three groups, we conducted Mfuzz clustering analysis (membership>0.2) to delineate dynamic gene expression patterns. This analysis identified 10 distinct gene clusters, among which 5 clusters (Clusters 2, 3, 6, 7, 9) exhibited consistent and sustained downregulated expression trends-results that highlight dynamic changes in mRNA expression profiles driven by differential c-di-GMP levels.

To understand how host cells isolated from thoracic lavage fluid respond to c-di-GMP modulation, we went on to carry out functional enrichment analysis on five modules that were coordinately downregulated - Clusters 2, 3, 6, 7, and 9 from the Mfuzz Analysis. These modules showed a sustained downward trend, which aligned with the decreasing c-di-GMP levels. Figure 7 depicts the KEGG pathway and Gene Ontology (GO) enrichment analyses of genes associated with neutrophil extracellular traps (NETs) formation, exploring the impact of P.aeruginosa c-di-GMP concentration on mouse NETs stimulation. In the KEGG Pathway Enrichment results (Figure 7A), the ‘Neutrophil extracellular trap formation’ pathway was significantly enriched, with 48 contributing gene. This finding highlights its importance in the host’s immune response to P. aeruginosa infection. Additionally, ‘Ribosome biogenesis in ‘eukaryotes’ and ‘Lysosome’ pathways were enriched, indicating the complexity of the underlying regulatory network. A circular plot was used to illustrate the gene count and percentage of different pathways. The GO enrichment results (Figure 7B) further support the structural and metabolic mechanisms of NETs: enrichment of terms like ‘intracellular membrane-bounded organelle’ indicates the involvement of organelles such as lysosomes (key for storing and releasing granular enzymes essential for chromatin decondensation and NETs structural integrity) and the endoplasmic reticulum (involved in post-translational modification of NETs-related proteins) in NETs formation. Enrichment of ‘metabolic pathways’ and ‘cellular metabolic process’ suggests metabolic reprogramming fuels NETs biogenesis, as neutrophils undergo rapid metabolic shifts to meet the energy and biosynthetic demands of chromatin remodeling, granular protein synthesis, and membrane trafficking during NETs extrusion. The NETs formation pathway plays a pivotal role in regulating the immune response and the dynamic interaction between bacterial c-di-GMP expression and host defense mechanisms in the mouse infection model. The KEGG and GO enrichment analyses have uncovered important functional genes, biological processes, and cellular components associated with NETs formation, emphasizing its significance in immune regulation during P.aeruginosa infection.

Among these five coordinately downregulated modules from Mfuzz analysis (Clusters 2, 3, 6, 7, 9), the top 10 hub genes (Pik3cd, Ncf1, Pik3r2, Fcgr4, Fcgr1, Plcg2, Casp1, Mapk14, Cybb, and Tlr4) are highlighted in orange in Figure 8. These genes are mainly involved in key pathways that regulate inflammatory responses and immune activation, including NF-κB, MAPK, and PI3K-Akt signaling. Representative genes such as Mapk14, Tlr4, and Cybb, which have high connectivity, play central roles in immune response and signal transduction. The remaining 10 key genes (Mapk11, Clcn5, Slc25a5, Atg7, Selp, Selplg, C3, Hdac2, Plcb1, Plcb3) are highlighted in purple, with involvement in complementary pathways including p38 MAPK signaling, autophagy, complement activation, calcium signaling, and epigenetic regulation, which collectively support processes like immune cell recruitment, metabolic homeostasis, and transcriptional control of NETs formation. Collectively, these results shed light on the critical molecular interactions driving NETs formation and identify potential targets for follow-up investigations.

Gene expression levels were analyzed using one-way ANOVA to identify significant differences among experimental groups. qRT-PCR validation of prioritized candidate genes (orange-highlighted) revealed Cybb, Mapk14, and Pik3cd as significantly differentially expressed (p < 0.05) across pairwise comparisons of P.aeruginosa infection groups (PAO1, PAO1△wspF, PAO1/plac-yhjH) (Figure 9). One-way ANOVA confirmed group-wide transcriptional dynamics, with expression patterns tightly correlated to bacterial c-di-GMP levels, implicating their roles in c-di-GMP-dependent neutrophil activation and immune regulation during host-pathogen interplay.

Immunofluorescence analysis, both in vitro (primary bone marrow neutrophils) and in vivo (pleural effusion samples) (Figure 10A), revealed distinct patterns of NETs formation among different P.aeruginosa strains. NETs were visualized by co-staining extracellular DNA (DAPI, blue), citrullinated histone H3 (red), and myeloperoxidase (MPO, green).Notably, the order of NETs formation intensity was consistent in both in vitro and in vivo models: PAO1△wspF > PAO1 > PAO1/plac-yhjH. Quantitative analysis (Figure 10B) further confirmed statistically significant differences among the groups (p < 0.05), with the PAO1△wspF strain known to form more biofilm, inducing the strongest NETs response. Additionally, cell-free DNA (cf-DNA) levels (Figure 10C) and MPO concentrations (Figure 10D) in the supernatants of centrifuged pleural lavage fluids from infected mice showed a parallel trend: the PAO1△wspF group exhibited the highest levels, followed by the PAO1 group, while the PAO1/plac-yhjH group had the lowest. These findings collectively indicate a direct correlation between bacterial c-di-GMP levels and NETs induction, suggesting that c-di-GMP in P.aeruginosa modulates the activation of host neutrophil-mediated immune responses. It should be noted that due to the scarcity of immune cells in pleural lavage fluids from non-infected mice, a PBS control group was not included in the in vivo immunofluorescence analysis.