Brain heart infusion (BHI) broth and bacto agar were obtained from BD Bioscience (San Diego, CA, USA). Chocolate agar was purchased from KisanBio (Seoul, Korea). Dulbecco’s modified eagle’s medium (DMEM) and Ham’s F-12 were purchased from Welgene (Daegu, Korea). Fetal bovine serum (FBS) was purchased from Gibco (Grand Island, NY, USA). Penicillin/streptomycin and trypsin-ethylenediaminetetraacetic acid (EDTA) were obtained from Hyclone (Logan, UT, USA). G418 and Hygromycin were purchased from Invitrogen (Thermo Fisher Scientific; Waltham, MA, USA) and Amresco (Solon, OH, USA), respectively. Recombinant murine macrophage colony-stimulating factor (M-CSF) and receptor activator of nuclear factor kappa-B ligand (RANKL) were purchased from R&D Systems (Minneapolis, MN, USA) and JW CreaGene (Seongnam, Korea), respectively. Polymyxin B sulfate salt (PMB), LPS from Escherichia coli O111:B4, and Nω-Nitro-L-arginine methyl ester hydrochloride (L-NAME) were purchased from Sigma-Aldrich (St. Louis, MO, USA). Tartrate-resistant acid phosphatase (TRAP) staining kit and mouse and rabbit specific horseradish peroxidase (HRP)/3,3’-Diaminobenzidine (DAB) detection immunohistochemistry (IHC) kit were purchased from CosmoBio (Tokyo, Japan) and Abcam (Cambridge, UK), respectively. Antibody specific to iNOS (#SC-7271) and rabbit polyclonal antibody against iNOS (#06-573) were purchased from Santa Cruz Biotechnology (Dallas, TX, USA) and Upstate Biotechnology (Lake Placid, NY, USA), respectively. Goat anti-rabbit IgG-HRP and anti-human CD25 IgG1-APC (#302610) were purchased from Southern Biotech (Birmingham, AL, USA) and Biolegend (San Diego, CA, USA), respectively. Coomassie Brilliant Blue R-250 staining and destaining solutions were obtained from Biosesang (Yongin, Korea).

N. bacilliformis KCTC 23360 (Korean Collection for Type Culture; Daejeon, Korea) was cultivated in BHI at 37°C with shaking for 6 h. After harvesting, to prevent uncontrolled bacterial growth or spread in the macrophage cell culture systems, the bacterial pellets were washed with phosphate-buffered saline (PBS) and subsequently heat-inactivated at 70°C for 1 h. To confirm complete inactivation, the heat-inactivated N. bacilliformis was cultured on a BHI agar plate, and the absence of bacterial colony formation was verified (data not shown). Neisseria meningitidis serogroup B American Type Culture Collection (ATCC) 13090 was kindly provided by Professor Jeon-Soo Shin (Yonsei University, Seoul, Korea) and used to demonstrate that the LOS of N. bacilliformis shares the same structural forms. The strain was grown in chocolate agar at 37°C in a jar for 6 h. LOS from N. bacilliformis and N. meningitidis were extracted using the LPS extraction kit (iNtRON Biotechnology; Seongnam, Korea) according to the manufacturer’s instructions. The isolated LOS was quantified using the Pierce™ chromogenic endotoxin quant kit (Thermo Fisher Scientific; Waltham, MA, USA).

Animal experiments were conducted under approval of the Institutional Animal Care and Use Committee of Seoul National University (SNU-211018-5-3). C57BL/6 mice were obtained from Dooyeol Biotech (Seoul, Korea). C57BL/6 background TLR2- and TLR4-deficient mice were provided by Professor Shizuo Akira (Osaka University, Osaka, Japan). All mice were maintained under specific pathogen–free conditions.

LIP was induced in 9- to 10-week-old male C57BL/6 mice. All animals were randomly assigned to one of the following groups (n = 5/group): control, LIP, LIP + live N. bacilliformis, and LIP + N. bacilliformis LOS. A 5–0 soft silk ligature was tied around the maxillary second molar and remained in place throughout the experimental period in all mice. Except for the first day when the ligature was placed on maxillary second molar, isoflurane anesthesia was used for all procedures. For the N. bacilliformis treatment group, N. bacilliformis (1 × 10⁹ CFU) with 1.5% carboxymethyl cellulose (CMC) was applied to the silk ligature every other day for a total of three treatments. In a separate experiment, N. bacilliformis LOS (5 or 20 μg/4 μl) was injected into the disto-palatal papilla on both sides of the right second molar (2 μl per site) using an insulin syringe with a 30G needle (Seongshim Medical Co., Anyang, Korea). Mice were treated around the ligature placement on the second molar with N. bacilliformis (1 × 10⁹ CFU) or 1.5% CMC (n = 5). The sample size (n = 5) was determined based on previous ligature-induced periodontitis studies in mice, where n = 5 per group has been commonly used and shown to be sufficient to detect statistically significant differences in alveolar bone loss (14, 15).

Seven days post-ligature placement, the mice were anesthetized and euthanized via cardiac perfusion with PBS. The molars were harvested and fixed in 4% paraformaldehyde overnight and then with PBS. Micro-CT scanning (Skyscan 1272; Bruker, Kontich, Belgium) was performed, and three-dimensional images were generated using CT-Volume software. Bone loss was assessed by measuring the distance from the cemento-enamel junction (CEJ) to the alveolar bone crest (ABC) using ImageJ software (National Institutes of Health, Bethesda, MD, USA). Alveolar bone loss was measured as the distance from the CEJ to the ABC. Measurements were taken at four sites per molar (third mesial root, distal and mesial roots of the second and first molars), and the mean CEJ–ABC distance was calculated for each group.

The maxillae were decalcified in 10% EDTA (pH 7) at 4°C for two weeks. After decalcification, the teeth were embedded in paraffin, positioned with the buccal side facing downward, and sectioned at a thickness of 5 µm. The sections were deparaffinized, rehydrated, and stained with hematoxylin/eosin (H&E) and TRAP, as previously described (16). For H&E staining, inflammatory cell infiltration was semi−quantitatively scored (0-3) based on the extent and density of inflammatory cell accumulation within the subepithelial and gingival connective tissue (17). For TRAP staining, number of osteoclast (N.Oc) and bone surface (BS) were quantified as N.Oc/BS using Image J. For each sample, regions of interest were defined in the alveolar bone between the maxillary first and second molars (M1–M2) and between the second and third molars (M2–M3). Bone surface (BS) and osteoclast numbers were measured in these regions, and osteoclast counts were normalized to BS (18). iNOS expression in gingival tissue was detected by immunohistochemistry (IHC) using an anti-iNOS antibody and an HRP/DAB detection kit according to the manufacturer’s protocol. iNOS immunostaining was quantified using the IHC Profiler plugin in ImageJ. H−scores were calculated from IHC−Profiler output as H−score = 3 × (% high positive) + 2 × (% positive) + 1 × (% low positive) for each region of interest (19). Images were captured using a digital upright fluorescence microscope (Olympus Corporation, Tokyo, Japan).

RAW 264.7 cells were obtained from the ATCC (Manassas, VA, USA) and cultured in DMEM containing 10% heat-inactivated FBS, 100 U/ml penicillin, and 100 μg/ml streptomycin at 37°C in a humidified incubator with 5% CO₂. The cells were seeded into 96-well plates at 200 μl of 3 × 10⁵ cells/ml/well and stimulated with the indicated stimuli for 24 h. In a separate experiment, bone marrow-derived macrophages (BMMs) were prepared as described previously (20). These cells were seeded into 96-well plates at 200 μl of 3 × 10⁵ cells/ml/well and stimulated with the indicated stimuli for 24 h. To examine osteoclast differentiation, BMMs were stimulated with 30 ng/ml RANKL and 30 ng/ml M-CSF for 2 days to generate committed osteoclasts. The committed osteoclasts were subsequently treated with N. bacilliformis LOS (3 or 30 ng/ml) for 22 h, and osteoclast activity was evaluated by TRAP staining.

NO production was analyzed by measuring nitrate levels in the supernatant. Equal volumes of Griess reagent (1% sulfanilamide, 0.1% naphthylethylenediamine dihydrochloride, and 2% phosphoric acid) and the supernatant were mixed and incubated at room temperature for 5 minutes. Nitrate levels were measured at 540 nm using a microplate reader (Tecan Group Ltd., Männedorf, Switzerland), with NaNO₂ used as a standard.

N. bacilliformis LOS, N. meningitidis LOS, and E. coli LPS were heated at 100°C for 5 min, loaded into each well, and separated using 15% sodium dodecyl sulfate-polyacrylamide electrophoresis (SDS-PAGE) gel. The gel was stained with Coomassie Brilliant Blue R-250 staining solution for 1 h, followed by overnight de-staining (45% methanol/10% glacial acetic acid). Silver staining was then performed as described previously (21).

RAW 264.7 cells were seeded into 6-well plates at 2 ml of 3 × 10⁵ cells/ml/well and stimulated with the indicated stimuli for 6 h. The detection of iNOS mRNA was performed as described previously (22) with the primers specific to murine iNOS (forward primer: 5′-CGCTTGGGTCTTGTTCACTC-3′, reverse primer: 5′-GGTCATCTTGTATTGTTGGG-3′) or GAPDH (forward primer: 5′-ACCCAGAAGACTGTGGATGG-3′, reverse primer 5′-CACATTGGGGGTAGGAACAC-3′). Melt curve and amplicon plot analyses demonstrated a single sharp peak for each primer pair, with no evidence of nonspecific amplification, thereby confirming amplification of a single product of the expected size.

RAW 264.7 cells were seeded into 6-well plates at 2 ml of 3 × 10⁵ cells/ml/well and treated with the indicated stimuli for 24 h. The cells were washed twice with PBS and lysed using RIPA buffer (LPS Solution, Daejeon, Korea). Western blotting was performed as described previously (23). Immunoreactive bands were detected using SUPEX ECL reagents (Neuronex, Daegu, Korea) and visualized with the Fusion FX6.0 imaging system (Vilber Lourmat, Marne-la-Vallée, France).

TLR2 and TLR4 activation was assessed by detecting CD25 expression, as described previously (24). Briefly, CHO/CD14/TLR2 and CHO/CD14/TLR4 cells were seeded into 24-well plates at 500 μl of 8 × 10⁵ cells/ml/well and incubated for 9 h. The cells were then stimulated with the indicated stimuli overnight. CD25 expression was analyzed using flow cytometry (FACSVerse, Becton Dickinson, Franklin Lakes, NJ, USA).

TNF-α and IL-6 were measured using ELISA kits, according to the manufacturer’s instructions (BioLegend).

The results are presented as mean ± standard deviation (S.D.) from at least three independent experiments unless stated otherwise. Statistical significance was examined by one-way ANOVA, followed by Tukey’s post-hoc test or Student’s t test using GraphPad Prism 6 software (GraphPad Software Inc., La Jolla, CA, USA) with significance (*) defined as P < 0.05.

We investigated whether N. bacilliformis exacerbates periodontitis using a mouse LIP model. Micro-CT analysis showed that the average CEJ-ABC distance was increased in the live N. bacilliformis-treated ligature group compared to the non-treated ligature group (mean values of 0.218 and 0.176 mm, respectively) (Figures 1A, B). In tissue sections with H&E staining, the control group exhibited a triangular-shaped gingiva and thick alveolar bone. In contrast, seven days post-ligature placement, mice showed destruction of the junctional epithelium and periodontal ligament, along with alveolar bone resorption compared to the control group (Figure 1C). N. bacilliformis-treated ligature mice showed more severe gingival tissue destruction and deeper subgingival pockets (Figure 1C) along with an increased accumulation of inflammatory cells within the subepithelial and gingival connective tissue (Figure 1D). Furthermore, TRAP-positive cells around the alveolar bone were increased in the N. bacilliformis-treated ligature mice compared to the non-treated ligature group, indicating increased osteoclast activity (Figures 1E, F). Since elevated NO levels act as a positive regulator of tissue destruction and bone resorption (13), we examined the expression of iNOS in gingival tissue using immunohistochemistry. As shown in Figures 1G, H, iNOS expression was markedly increased in N. bacilliformis-treated ligature mice compared to both control and non-treated ligature groups. These findings indicate that N. bacilliformis exacerbates periodontitis through the upregulation of NO production.

Since NO is primarily produced by macrophages (25), we treated the murine macrophage cell line RAW 264.7 with N. bacilliformis to determine whether it induces NO production. As shown in Figure 2A, similar to E. coli LPS and Pam2CSK4 (mimicking bacterial lipoproteins), heat-killed N. bacilliformis increased NO production in a dose-dependent manner. Next, since LPS is a well-known potent virulence factor of Gram-negative bacteria, we examined whether the NO induction by N. bacilliformis was due to its LOS using a lipid A-targeting LPS blocker, PMB. Pre-treatment with PMB substantially reduced NO production induced by both N. bacilliformis and E. coli LPS but not that by Pam2CSK4 (Figure 2B). These results suggest that N. bacilliformis LOS might play a critical role in NO production in macrophages.

To determine whether N. bacilliformis LOS induces NO production, we purified LOS from N. bacilliformis. As shown in Figure 3A, impurities in the LOS preparation was hardly detected in the Coomassie blue staining. When N. bacilliformis LOS was treated with Proteinase K, DNase I, heat, or lipoprotein lipase, NO induction remained unchanged compared to that of untreated N. bacilliformis LOS. However, PMB-treated N. bacilliformis LOS showed reduced NO production (Figure 3B), indicating NO induction by N. bacilliformis LOS was not due to impurities in the LOS preparation, but rather was caused by the LOS itself. Since Neisseria species are known to have LOS that lack the O-antigen (10), we determined whether N. bacilliformis LOS also lacks the O-antigen. In silver staining, the band pattern of N. bacilliformis LOS was similar to that of N. meningitidis LOS but distinct from that of E. coli LPS, which contains O-antigen ladders. Most of the bands were located in the lower region (Figure 3C). These results suggest that the endotoxin of N. bacilliformis is LOS, an O-antigen-deficient form of LPS. N. bacilliformis LOS increased NO production in a dose-dependent manner in RAW 264.7 cells (Figure 3D). Collectively, these findings indicate that LOS is the major factor responsible for NO induction by N. bacilliformis in macrophages.

iNOS is a critical enzyme responsible for the production of large amounts of NO (25). Thus, we examined the expression of iNOS in macrophages treated with N. bacilliformis LOS. N. bacilliformis LOS increased iNOS mRNA expression in a dose-dependent manner in RAW 264.7 cells (Figure 4A, Supplementary Table 1). Consistent with this result, an increase in iNOS protein level was also observed (Figure 4B, Supplementary Table 1). In addition, pretreatment with L-NAME, an iNOS inhibitor, abolished N. bacilliformis LOS–induced NO production, indicating that NO induction by LOS is mediated through iNOS upregulation (Figure 4C). These results indicate that N. bacilliformis LOS enhances iNOS expression in macrophages, ultimately leading to elevated NO production.

Both LPS and LOS are known to be recognized by TLR4 (26, 27). We further examined whether N. bacilliformis LOS induces an inflammatory response through TLR4. As shown in Figures 5A, B, N. bacilliformis LOS induces CD25 expression in CHO/CD14/TLR4 cells in a dose-dependent manner. However, unexpectedly, CD25 expression was also observed in CHO/CD14/TLR2 cells (Figures 5C, D). In addition, when N. bacilliformis LOS was treated with Proteinase K, DNase I, heat, or lipoprotein lipase, N. bacilliformis LOS−induced CD25 expression was not significantly altered by these treatments, implying that the observed TLR2 activation is unlikely to be due to residual lipoprotein contamination. To clearly determine whether N. bacilliformis LOS activates both TLR2 and 4 signaling pathways, primary BMMs derived from wild-type, TLR2-deficient, or TLR4-deficient mice were treated with N. bacilliformis LOS. In TLR4-deficient BMMs, NO production was completely abolished, whereas in TLR2-deficient BMMs, it was reduced by approximately 50% (Figure 5G). These results suggest that N. bacilliformis LOS primarily induces an inflammatory response via TLR4 and partially through TLR2.

Since N. bacilliformis LOS efficiently induces NO, we investigated whether it enhances periodontitis development in vivo. As shown in Figures 6A, B, N. bacilliformis LOS increased the CEJ-ABC distance compared to the non-treated ligature group, indicating that N. bacilliformis LOS accelerates periodontitis progression. Furthermore, N. bacilliformis LOS exacerbated ligature-induced destruction of the junctional epithelium and inflammatory cell infiltration within the subepithelial and gingival connective tissue (Figures 6C, D). TRAP-positive osteoclast cells and iNOS expression were increased in the N. bacilliformis LOS-treated ligature group compared to the non-treated ligature group (Figures 6E-H). In addition, N. bacilliformis LOS induced TRAP-positive multinucleated osteoclasts in committed osteoclasts, indicating that N. bacilliformis LOS promotes osteoclast differentiation (Figures 6I, J). These results indicate that N. bacilliformis LOS plays a critical role in N. bacilliformis-accelerated periodontitis.