Eight-week-old C57BL/6 male mice were ordered from Beijing Vital River Laboratories (Beijing, China). The mice were housed in a specific pathogen-free (SPF) facility. Mice aged 10–12 weeks were used in the study. PD was established in the mice as previously described (Bai et al., 2022). The bilateral maxillary second molars of the mice were ligatured with 5–0 silk. Then, mice were orally inoculated with subgingival plaques (1x10⁹ colony-forming units per mouse) from PD patients every other day for 2 weeks. Patients with moderate to severe PD were recruited and those who smoke, suffer systemic diseases, have less than 8 teeth, take antibiotics or probiotics and undergo periodontal treatments within the last 6 months were excluded. Subgingival plaques were then collected from PD patients and mixed well and stored in the -80°C freezer in advance. Subsequently, pulmonary fibrosis was induced in the mice with BLM (HY-17565A, MedChemExpress, Shanghai, China). Briefly, mice were first anaesthetized with isoflurane gas (4%) and then intratracheally administered BLM (0.5 mg/kg body weight, 30 μl per mouse) or saline control by an atomizing needle (Bio Jane Trading Co., Ltd., Shanghai, China). Mice were sacrificed 3 weeks later or at the indicated time points for further analysis.

The Pg-induced PD mouse model was established by ligation of bilateral maxillary second molars with 5–0 silk sutures as previously described and then orally inoculated with Pg (ATCC 33277, 1x10⁹ colony-forming units per mouse) every other day for 2 weeks. Then BLM was intratracheally administered, and the sutures were maintained until the end of the experiment. The Institute of Animal Care and Use Committee (IACUC) of Cyagen Biosciences, China (Approval No. ICU21-0080), approved the conduct of all the animal experiments. Collecting subgingival plaques from PD patients was approved by the Institutional Review and Ethics Board of Shanghai Ninth People’s Hospital, Shanghai Jiao Tong University School of Medicine (No. SH9H-2021-T115–1). All participants provided informed consent.

To deplete neutrophils, mice were intraperitoneally injected with 200 µg of anti-mouse Ly6G antibody (BE0075-1, BioXCell) every 2 days since BLM administration to the end of the experiment. To neutralize IL-17A, mice were intraperitoneally injected with 200 µg of anti-mouse IL-17A antibody (BE0173, BioXCell) every 3 days since BLM administration until the end of the experiment.

To determine PD-induced alveolar bone loss as previously described (Abe and Hajishengallis, 2013), maxillae of mice were acquired, and the gingiva were removed, followed by bleaching and then rinsing. 1% methylene blue was then used to stain the maxillae. The staining maxillae were captured by a Leica digital camera (Leica, Germany). The distance between the cementoenamel junction (CEJ) and alveolar bone crest (ABC) was measured by ImageJ software (National Institutes of Health, Bethesda, USA).

Lung lobes were fixed by 4% paraformaldehyde, dehydrated and then embedded in paraffin. Paraffin-embedded sections were subsequently performed with H&E staining or Masson’s trichrome staining. All images were obtained with a Leica DMi8 microscope (Leica, Wetzlar, Germany). The level of interstitial fibrosis was assessed as acknowledged protocols (Hübner et al., 2008). The hydroxyproline level was quantified by a hydroxyproline assay kit (A030-3-1, Nanjing Jiancheng, China).

RNA of lung tissues was isolated by TRIzol reagent (15596018, Invitrogen, USA). RNA was then reverse-transcribed to cDNA by a reverse transcription kit (Takara, Shiga, Japan). A LightCycler 480 II (Roche) was used to perform real-time PCR. Primers used in this study were listed in Supplementary Table S1 .

Lung tissues were homogenized using RIPA lysis buffer containing phenylmethylsulfonylfluoride (PMSF, Sigma–Aldrich) and protease inhibitor (Med Chem Express). Protein concentration was then measured by the BCA protein assay kit (88512, Thermo Fisher Scientific). The 10% SDS–PAGE was used to separate proteins. Detection was performed with ECL Western Blotting Substrates (Thermo Fisher Scientific). The antibodies used in this study were anti-TGF-β1 (ab215715, Abcam), anti-Fibronectin (ab2413, Abcam) and anti-GAPDH (14c10, Cell Signaling Technology).

Single-cell suspensions from the lungs were prepared according to previous protocol (Jungblut et al., 2009). The lungs were put in a gentleMACS C Tube (130-093-237, Miltenyi Biotec) containing 5 mL of HEPES buffer supplemented with DNase I (80 U/mL, A37780050, applichem) and collagenase IV (300 U/ml, A004186-0001, worthington). The C Tube was attached to the gentleMACS Dissociator (130-093-235, Miltenyi Biotec) to run the program “m_lung_01”. Then the lung tissues were digested for 30 minutes with agitation at 37°C and followingly run the program “m_lung_02” to obtain suspensions. PBS was added to terminate the digestion, and the suspensions were filtered by a cell strainer to acquire single lung cells. The cervical lymph nodes (cLNs) of the mice were ground on a cell strainer and rinsed with PBS to obtain single cells. Bone marrow cells were prepared by flushing the femur with a syringe.

Single-cell suspensions from the lungs, cLNs and bone marrow were centrifuged and stained with Live/Dead Fixable Viability Stain 510 (564406, BD Biosciences). The cell samples were incubated with CD16/32 antibodies (101320, Biolegend) to block Fc receptors before surface staining and then stained with antibodies. For cytokine staining, cells were stimulated with the cell activation cocktail (423303, BioLegend), washed with PBS and finally stained with antibodies. Antibodies against the following molecules were used in this study: CD45 (557659, Biosciences), CD11b (101205, Biolegend), Ly6G (127616, Biolegend), Ly6C (128018, Biolegend), CD64 (139321, Biolegend), CD11c (117333, Biolegend), I-A/I-E (562564, Biosciences), SiglecF (155506, Biolegend), CD3 (100203, Biolegend), CD4 (100540, Biolegend), CD8 (100722, Biolegend), B220 (103251, Biolegend), F4/80 (123116, Biolegend), CD115 (135515, Biolegend), IFN-γ (505810, Biolegend) and IL-17A (506903, Biolegend). Samples were run on a BD LSRFortessa X-20 flow cytometer (BD Biosciences).

Total genomic DNA from lungs and oral silk ligatures was extracted by an OMEGA Soil DNA Kit (M5635-02, Omega Bio-Tek). Qualified DNA samples were performed PCR amplification of the nearly full-length bacterial 16S rRNA genes. PCR amplicons were then sequenced by the PacBio Sequel platform (Shanghai Personal Biotechnology Co., Ltd., China). Microbiome bioinformatics was performed with QIIME software and R packages (v3.2.0). The alpha diversity was analysed by the ASV table in QIIME2, and beta diversity was performed with Bray–Curtis metrics. Differentially abundant taxa between groups were analysed by linear discriminant analysis effect size (LEfSe) with the default parameters. The raw sequences were deposited in the NCBI Sequence Read Archive database with the accession number PRJNA1138471.

Identifying the presence of Pg was by fluorescence in situ hybridization (FISH) according to a published protocol (Rudney et al., 2005; Niu et al., 2024). Lung tissue slides were incubated with a specific Alexa Fluor 594-labelled Pg probe (5’-CAATACTCGTATCGCCCGTTATTC-3’) at 46°C overnight. Samples were washed with solution (100 mM Tris–HCl, 0.9 M NaCl, pH 7.5) for 25 min at 48°C, then distilled water, air-dried and counterstained with DAPI (Invitrogen). Images were obtained with a Leica DMi8 microscope (Leica, Wetzlar, Germany).

The experimental data are presented as the mean ± SEM. Unpaired Student’s t test was used for two-group comparisons. One-way ANOVA was used for comparisons among more than two groups. All statistical analyses were performed with GraphPad Prism software. Values of P < 0.05 were considered to indicate statistical significance. Information about the statistical details and methods is also included in Figure legends.

To determine the impact of PD on pulmonary fibrosis, the PD mouse model combined with pulmonary fibrosis was established. The molars of the mice were ligatured with silk sutures, and subgingival plaques from PD patients were inoculated in mouse oral cavity to establish PD with humanized oral microbiota, followed by BLM administration to induce pulmonary fibrosis ( Figure 1A ). PD caused severe alveolar bone loss ( Figure 1B ) and significantly increased the distance between the CEJ and ABC in mice ( Figure 1C ).

Moreover, the results of H&E and Masson’s trichrome staining indicated that PD induced more severe destruction of lung architecture and obviously increased collagen accumulation in mouse lungs ( Figure 1D ). Ashcroft scores, which assess the severity of fibrosis, also indicated that PD aggravated pulmonary fibrosis ( Figure 1E ). The lungs in the BLM combined with PD group were obviously heavier than those in the BLM group ( Figure 1F ). The hydroxyproline content, a marker of lung fibrosis severity, was markedly greater in lungs of the BLM combined with PD group than the BLM group ( Figure 1G ). Consistent with these observations, PD also obviously promoted the expression of fibrotic genes ( Figure 1H ). The western blot results also demonstrated that PD significantly increased the fibrotic protein levels such as Fibronectin and TGF-β1 ( Figures 1I, J ). Collectively, these results illustrated that PD exacerbated pulmonary fibrosis in mice.

Chronic inflammation is generally considered as a risk factor of pulmonary fibrosis, and many studies have verified that various immune cells are important for lung fibrosis (Yang et al., 2019). To explore immune mechanisms underlying PD-induced aggravation of pulmonary fibrosis, we next analyzed changes of immune cells. Myeloid cells and lymphoid cells in the lungs were analysed in an unbiased manner via flow cytometry at 0, 7 and 21 days after BLM treatment. Both the percentage and number of neutrophils were markedly greater in the lungs of the BLM+PD group than the BLM group at 0, 7 and 21 days after BLM administration ( Figures 2A, B ). Compared with the BLM group, the percentage and number of Th17 cells were also significantly greater in the BLM+PD group at the 3 indicated time points ( Figures 2C, D ). However, the percentage of other myeloid cells and lymphoid cells in the lungs did not always show the same significant trend as the number at the 3 indicated time points ( Supplementary Figure S1 ). This indicated that neutrophils and Th17 cells were the most important immune cells that mediated the PD-induced aggravation of pulmonary fibrosis.

Additionally, myeloid cells and lymphoid cells in the cLNs and bone marrow of mice were also analysed at 0, 7 and 21 days after BLM administration. PD markedly increased neutrophils and Th17 cells in the cLNs ( Figures 2E–H ). PD also led to obvious changes in other myeloid cells and lymphoid cells in the cLNs ( Supplementary Figures S2A–L ). A significant increase in B cells was also detected among the changes caused by PD in the cLNs ( Supplementary Figures S2A–F ). PD also affected myeloid cells and lymphoid cells in the bone marrow of mice at different time points ( Supplementary Figures S2M–R ). Compared with the BLM group, the percentage and number of neutrophils in the bone marrow were markedly greater in the BLM+PD group ( Figures 2I, J ). Together, these results demonstrated that for mice with pulmonary fibrosis, PD increased the expansion of neutrophils and Th17 cells in lungs, and that cLNs and bone marrow were likely the sources of increased neutrophils and Th17 cells.

To further investigate the importance of neutrophils and Th17 cells in PD-induced aggravation of pulmonary fibrosis, we depleted neutrophils or IL-17A (mainly generated by Th17 cells) to assess the severity of lung fibrosis in mice. The lung damage and collagen accumulation were notably attenuated in the BLM+PD+Anti-Ly6G group and BLM+PD+Anti-IL17A group than those in the BLM+PD group ( Figures 3A, B ). The weights of the lungs and the hydroxyproline level were significantly reduced in PD mice after depletion of neutrophils or IL-17A ( Figures 3C, D ). Consistent with above results, the expression of fibrosis-associated genes and proteins also markedly reduced in PD mice after depletion of neutrophils or IL-17A ( Figures 3E–G ).

We next evaluated the relationship between neutrophils and Th17 cells in PD-induced aggravation of pulmonary fibrosis. Neutrophils in the lungs were comparable between BLM-administrated mice and BLM+Anti-IL17A treated mice, while the percentage and number of neutrophils were significantly lower in BLM+PD+anti-IL17A mice than in BLM+PD mice ( Figures 4A, B ). These results indicated that IL-17A was the upstream signal triggering neutrophils to mediate PD-induced aggravation of pulmonary fibrosis. However, Th17 cells in lungs were comparable between the BLM+PD group and BLM+PD+Anti-Ly6G group ( Figures 4C, D ). This result indicated that neutrophils did not influence the number of Th17 cells in mice with pulmonary fibrosis. Hence, these results revealed that neutrophils and Th17 cells are critical for the PD-induced aggravation of pulmonary fibrosis and that Th17 cells regulate neutrophils via IL-17A.

Oral microbes were detected in the BALF microbiome of IPF patients, which possibly promoted the occurrence or progression of IPF (Tong et al., 2019). Therefore, we next assessed whether PD-related pathogenic bacteria transferred from the oral cavity to the lung and consequently impacted pulmonary fibrosis. We analysed both the lung microbiome and oral microbiome in the BLM group and BLM+PD group by 16S rRNA gene sequencing. For the lung microbiome, the alpha diversity and beta diversity were not notably different between the BLM+PD group and BLM group ( Supplementary Figure S3A, B ). In contrast, the alpha diversity and beta diversity of the oral microbiome were significantly different between the two groups ( Supplementary Figure S3C, D ). The LEfSe results of the enriched bacterial species in lungs and oral ligatures demonstrated that Porphyromonas gingivalis was among the top-ranked PD-related species in the BLM+PD group ( Figures 5A, B ).

To determinate the effect of Pg on pulmonary fibrosis in mice, we established a Pg-induced PD model by ligation of bilateral maxillary second molars and oral inoculation of Pg in the PG group and subsequently administrated BLM ( Figure 5C ). Pg was detected in lung sections of mice from both the BLM+PD group and BLM+PG group, which confirmed Pg translocating from the oral to the lungs ( Figure 5D ). According to the results of H&E and Masson’s trichrome staining, Pg caused more severe injury and significantly increased collagen accumulation in lungs of the mice ( Figures 5E, F ). Pg also notably promoted the expression of fibrotic genes ( Figure 5G ) and fibrotic proteins ( Figures 5J, K ). Compared with the BLM group, mice in the BLM+PG group presented markedly heavier lungs and higher levels of hydroxyproline in the lungs ( Figures 5H, I ). Overall, Pg translocated from the oral to the lung and aggravated BLM-induced pulmonary fibrosis in mice.

To explore immune mechanisms underlying Pg-driven exacerbation of pulmonary fibrosis, we performed flow cytometry analysis of the lungs, cLNs and bone marrow at 7 days after BLM administration, which was an early inflammation stage that determines the following development of pulmonary fibrosis. Consistent with the PD-induced infiltration of neutrophils and Th17 cells ( Figure 2 ), Pg also significantly increased neutrophils and Th17 cells in lungs of mice ( Figures 6A–D ). The percentages and numbers of other myeloid cells and lymphoid cells did not exhibit the same significant changes between the two groups ( Supplementary Figure S4A–C ). We further analysed immune cells in cLNs and the bone marrow in an unbiased manner. In line with subgingival plaque-induced PD, Pg also caused notable increases of neutrophils and Th17 cells in cLNs ( Figures 6E–H ). Additionally, Pg notably increased the percentage and number of B cells in cLNs ( Supplementary Figures S4D–F ). Compared with the BLM group, the percentage and number of neutrophils in the bone marrow of the BLM+PG group were greater ( Figures 6I, J ). To some extent, Pg also affected other myeloid cells and lymphoid cells in the bone marrow ( Supplementary Figures S4G, H ). Taken together, these results illustrated that Pg promoted the expansion of neutrophils and Th17 cells in lungs, cLNs and bone marrow of mice with BLM-induced pulmonary fibrosis.