Strains, constructs, and growth conditions
The strains used in the study are listed in Supplementary Table 1, the plasmids are listed in Supplementary Table 2, and the primers are listed in Supplementary Table 3. S. oralis was grown at 37 °C in the aerobic environment with Trypticase Soy Broth (TSB, Hope Bio-Technology) medium and TSB agar plates (TSB medium with 15 g/L agar). The knockout strains were constructed by replacing the target gene with the kanamycin gene sequences by homologous recombination. Fragments with ~500 bp each of the upstream and downstream of the target gene and the kanamycin gene sequences were introduced into S. oralis by competence stimulating peptide (CSP). The single clones were picked up on the TSB agar plates with kanamycin resistance for mutant identification. The overexpressed strains were constructed by introducing the plasmids inserted with the target gene into S. oralis by CSP and identified on the TSB agar plates with spectinomycin resistance. Complementary strains were constructed by introducing the overexpression plasmids into the mutant strains, which were picked up on the TSB agar plates with dual resistance to spectinomycin and kanamycin. S. oralis with overexpressed plasmids was induced with 8 mM inducer isopropyl-β-d-thiogalactoside (IPTG) at 37 °C for 1 h to complete the gene overexpression. If required, the following final concentrations of antibiotics were added to S. oralis culture medium: spectinomycin (1 mg/mL) and kanamycin (50 μg/mL).
E. coli DH5α was used for cloning and plasmid replication, and E. coli BL21 (DE3) was used for protein purification. E. coli was grown in the LB (Luria-Bertani, Sangon Biotech) broth medium and the LB agar plates (LB medium with 15 g/L agar) at 37 °C. The homologous recombinant plasmids were introduced into E. coli and single clones were picked up on the LB agar plates with resistance. If required, the following final concentrations of antibiotics were added to the E. coli culture medium: ampicillin (100 μg/mL) and kanamycin (50 μg/mL).
Protein purification
EF-Tu protein was produced and purified from freshly transformed E. coli BL21 (DE3) cells, which contained pET28a expression plasmids carrying the tuf gene35,36,51,52. The tuf gene was subcloned into the vector pET28a as a NdeI-HindIII fragment by homologous recombination and transformed into E. coli BL21 (DE3). E. coli BL21 (DE3) colonies were picked and resuspended in the LB medium containing kanamycin and incubated in a shaker at 37 °C until the absorbance (OD600) reached 0.6–0.8. The expression of recombinant protein was induced with 0.5 mM IPTG at 18 °C. After 16 h, E. coli BL21 (DE3) cells were collected and stored at −80 °C. To purify the proteins, cell precipitates were resuspended in the buffer at pH 7.5 containing 20 mM Tris–HCl, 100 mM NaCl, 1 mM DNase I, 1 mM PMSF, and 2 mM MgCl2. After being mixed well on ice, the cells were lysed by ultrasonication. The cell lysate was centrifuged at 4 °C, 12,000×g for 30 min, and the supernatant was incubated with Ni-NTA resin. The resin was washed sequentially with buffer A at pH 7.5 containing 20 mM Tris–HCl, 100 mM NaCl, 20 mM imidazole, and buffer B at pH 7.5 containing 20 mM Tris–HCl, 100 mM NaCl, and 100 mM imidazole. The final protein was eluted from the column with the elution buffer at pH 7.5 containing 20 mM Tris–HCl, 100 mM NaCl, and 300 mM imidazole. Protein concentration was measured using the BCA Protein Quantification Kit (Vazyme).
Metagenomic analysis
The information on the samples is listed in Supplementary Table 4. Quality assessment of raw sequencing data and quality control (QC) of sequencing data were achieved by FastQC. After QC, host DNA sequences were removed by KneadData and bowtie253. Subsequently, the species in clean data were annotated and classified using MetaPhlAn, and count tables were generated by executing the “rel_ab_w_read_stats” command54. The HUMAnN was used to analyze the functional information of microorganisms to obtain information on genes, pathways, and members of associated microbial communities. Subsequently, the relative abundances of different species and count tables were imported into the R software for further diversity analysis, statistical tests, and visualization. R packages such as ggplot255, RcolorBrewer56, ggsignif57, ade458, and vegan59 were used for downstream analysis. The biodiversity corresponding to the different groups was assessed by the values of α and β diversity. Principal co-ordinates analysis (PCoA) was implemented by using the ade4 package and mapped using the ggplot2 package. P-values corresponding to the relative abundances of different species in three groups were calculated by the Wilcoxon test using the ggsignif package.
For co-occurrence network analysis, correlations among species in each group were calculated based on the relative abundance at the species level. The thresholds for species’ relative abundance were all set at relative abundance >1% and occurring in more than 20% of the samples and the correlations between species were calculated based on Spearman. The final correlation values of |r| ≥ 0.6 and p < 0.05 were used for downstream analysis.
MS and pathway analysis
Overnight strains were diluted to absorbance OD600 was 0.1 for preparation. For bacterial surface protein sample collection, the treated S. oralis was collected by centrifugation and washed with a solution containing 40% sucrose (SCR) and 20 mM sodium azide (SCR). Immobilized trypsin (Thermo Fisher Scientific, Waltham, MA, USA), which had been activated with 50 mM ammonium bicarbonate (SCR), was added to the bacterial precipitate and reacted at 37 °C for 45 min. The isolated bacterial surface protein peptides were then collected and performed electrophoresis and Coomassie Brilliant Blue staining.
Clinical samples were obtained from healthy and periodontitis volunteers. Saliva samples were obtained by collecting 5 mL of saliva from volunteers, and subgingival plaque was obtained by scraping subgingival plaque from multiple tooth sites with a probe and collecting it in PBS. Samples were snap-frozen in liquid nitrogen immediately after collection and stored at −80 °C.
The corresponding gels were cut and decolorized with 50% acetonitrile (ACN)-50% 50 mmol/L NH4HCO3. Dithiothreitol (DTT), iodoacetamide (IAA), and 50% ACN–50% 50 mM NH4HCO3 were added sequentially to the decolorized samples to reduce alkylation. The samples were digested with trypsin, and the peptides were obtained with an extraction solution. The enzymatically cleaved peptides were desalted using a self-filling desalting column, and the solvent was evaporated in a vacuum centrifuge concentrator at 45 °C. Peptides were captured using Easy-nLC 1200 (ThermoFisher Scientific, USA) and 150 μm × 15 cm in-house-made column packed with Acclaim PepMap RPLC C18 (1.9 μm, 100 Å, Dr. Maisch GmbH, Germany). The samples were eluted from the capture column by mobile phase A: 0.1% formic acid in water and mobile phase B: 20% 0.1% formic acid in water–80% acetonitrile.
The nanoLC system was coupled to a Q Exactive™ Hybrid Quadrupole-Orbitrap™ Mass Spectrometer (Thermo Fisher Scientific, USA), which was operated at a Spray voltage of 2.2 kV and a Capillary temperature of 270 °C. The MS resolution was 70,000 at 400m/z, and the MS precursor m/z range was 300.0–1800.0. The top 20 most intense peptide ions from the preview scan in the Orbitrap were collected for the next step of MS analysis60. The MS raw files were analyzed using MaxQuant. The peptides identified with high confidence were selected for downstream protein identification analysis.
The protein abundance avidity was analyzed using GraphPad Prism 8.0. The KOBAS 3.0 web tool was used to perform the KEGG pathway analysis61, and the STRING database was used to complete the GO enrichment analysis62. The results of the KEGG pathway and GO enrichment analyses were visualized by the ggplot2 package. The co-occurrence network was implemented based on the R package igraph and visualized in Cytoscape software63. The co-occurrence network node centrality was implemented based on the closeness of nodes and feature vectors. Each node represented a protein gene, while edges represented associations between nodes.
Bacterial growth curve quantification
The frozen S. oralis was cultured 1:100 in TSB medium and grown overnight at 37 °C. Then they were transferred to fresh TSB medium, diluted to an absorbance (OD600) of 0.1, and shaken at 220 rpm in a 37 °C shaker. 100 μL of the bacterial solution was added to 96-well plates, and the absorbance (OD600) was tested at 0, 0.5, 1, 2, 4, 8, 12, and 24 h. Three replicates were performed at each time point to quantify the growth curve of S. oralis and to fit the curve. Meanwhile, the bacterial solution was coated on TSB agar plates diluted with PBS to calculate the bacterial density.
Biofilm growth curves were measured after adding S. oralis (OD600 = 0.1) to a 24-well plate containing coverslips and incubating at 37 °C for 1, 2, 4, 8, 12, and 24 h. The biofilms were fixed with 2.5% glutaraldehyde. After fixation at room temperature for 2 h, 10% crystal violet was added to the wells for 15 min for staining. After removing the supernatant, the biofilms were decolorized for 15 min, and the absorbance (OD595) of the decolorized solution was measured to quantify the content of the biofilm.
SEM photography
The biofilms were fixed with 2.5% glutaraldehyde at 4 °C overnight. After dehydration with a gradient ethanol series, the samples were performed critical point drying and gold spraying. The biofilm samples were subsequently photographed and analyzed with a field emission scanning electron microscope (Zeiss, SIGMA).
Biofilm quantification by fluorescence microscopy
Overnight cultures of S. oralis were diluted to appropriate absorbance (OD600 = 0.1) and incubated in confocal dishes at 37 °C for 1 h. After fixation with 2.5% glutaraldehyde, 5 μM of DiI membrane fluorescent dye (Yeasen) was added to the culture and stained at 37 °C for 30 min protected from light. The supernatant was gently discarded, and the biofilm was repeatedly washed with PBS to remove the unbound dye. The biofilms were then photographed with CLSM (Leica Stellaris 5 WLL), and the fluorescence intensity of the biofilms was quantified using ImageJ.
Bacterial movement tracking
S. oralis was stained with 5 μM of DiI dye for 30 min at 37 °C protected from light, and then washed with PBS to remove the unbound dye. S. oralis was then resuspended with TSB medium and added to the confocal dishes. It was immediately photographed with a Nikon Ti2-E fluorescence microscope for 10 min without a time delay interval. The time interval between photographs was set to 0.22 s. The individual bacterial movement trajectories in the video were analyzed using a MATLAB program to calculate their relevant movement parameters such as MSD, G-Probability, and displacement frequencies. GraphPad Prism 8.0 and Origin 2020 were used for graphical analysis.
Protein movement tracking
The purified protein was diluted to 4 mg mL−1, and 5-Fluorescein isothiocyanate (FITC, Yeasen) at a final concentration of 1 mg/mL was added to 100 μL of protein solution and incubated at 4 °C for 8 h. At the end of the reaction, 50 mM NH4Cl was added to the solution and reacted for 2 h to terminate the reaction. To remove unbound FITC and obtain a fluorescence-linked protein solution, the reaction solution was centrifuged in a 10 kDa ultrafiltration tube (Millipore) at 5000 rpm for 30 min at 4 °C. The solution was lyophilized at −80 °C.
The protein-FITC was diluted to the appropriate concentration and added dropwise to the coverslips64. The motion of individual proteins was tracked by TIRFM, and a video was taken for 10 min. GraphPad Prism 8.0 and Origin 2020 were used for graphical analysis.
Computer theoretical calculations
The GROMACS 2019.2 package, using the charmm36 force field, was used for the molecular dynamics simulations. The 3D structures of EF-Tu (AlphaFoldDB: P33170), Enolase (AlphaFoldDB: A0A428I769), and fibronectin (AlphaFoldDB: P02751) were obtained from the UniProt website. The HA plane was constructed by CHARMM-GUI using the corresponding CHARMM force field. EF-Tu was placed inside the box with the target protein and HA plane to form the system so that the system was 1.5 nm away from the box axis. Water and ions were added to the box for energy minimization. Subsequently, NVT, NPT equilibration, and finally, molecular dynamics simulations were performed for 10 ns. The simulation results were visualized by VMD software the final system was analyzed for equilibrium by RMSD analysis, and the potential energy of amino acid residues was calculated for EF-Tu.
MicroScale thermophoresis (MST) analysis
Enolase and fibronectin proteins were fluorescently labeled using rhodamine dissolved in the buffer (20 mM Hepes, 100 mM NaCl, pH 7.5). For MST measurements, purified EF-Tu protein and BSA protein were subjected to a dilution series of 16 serial 1:1 dilutions with ligand buffer (20 mM Tris–HCl, 100 mM NaCl, pH 7.5). Enolase-Rhodamine and fibronectin-Rhodamine proteins were diluted to the appropriate concentration and incubated with the prepared ligand solution 1:1 for 10 min. Samples were loaded into glass capillaries (NanoTemper Technologies), and microthermophoresis was performed using a 60% LED power supply and high MST power. Dissociation constants (Kd) were calculated from duplicate measured readings via NanoTemper Technologies software.
Periodontitis mouse model
8-week-old C57BL/6 female mice were selected for periodontitis animal modeling. All experiments were approved by the Animal Research Committee of the School of Stomatology, Wuhan University, China. After anesthesia, 5-0 filaments were ligated to the cervical region of the bilateral maxillary second molars of the mice. Overnight cultures of S. oralis were diluted to the absorbance (OD600) of 0.1, and 10 μL of the fresh bacterial solution was dropped on the ligature once daily. To observe the role of the tuf gene of S.oralis in the pathogenesis of periodontitis, Δtuf, pDL278tufΔtuf, and WT strains were used to coat the ligature lines in the cervical region of the teeth, and the ligated-only group served as a control group. Mice were euthanized 5 days after modeling and subsequent testing was performed. No significant alveolar bone resorption was observed in the ligation-only group at 5 days, and the effect of the tuf gene of S.oralis on alveolar bone resorption could be more clearly observed. To observe the effect of simeprevir, 1 μM final concentration of simeprevir was added to the ligature with or without bacterial solutions. The process was repeated for 7 days once a day. To observe the preventive effect of simeprevir on periodontitis in a mouse model of flora transplantation, subgingival plaque from patients with periodontitis was collected and cultured. In the bacterial infection group, the cultured mixed flora was added dropwise to the ligature. In the Simeprevir group, a 1 μM final concentration of simeprevir with or without mixed flora was added dropwise to the ligature, and the ligated-only group served as a control. The mice were euthanized after 7 days of once-daily repeats. To observe the supplementary therapeutic effects of simeprevir on periodontitis in mouse models of colony transplantation, the mouse model of periodontitis that had been ligated was first established, and the ligature was removed on day 7. The procedure with simeprevir and mixed flora solutions was repeated once a day for 7 days. Mice were euthanized on day 14 of the models. Finally, the bilateral maxillary alveolar bone was taken for micro-CT scanning and histological analysis.
Micro-CT and histological analysis
Maxillary alveolar bone was fixed with 4% paraformaldehyde at 4 °C for 48 h and then flushed overnight. The alveolar bone tissue was scanned with a high-resolution micro-computed tomography scanner (SkyScan) with parameters set to 55 kV, 200 μA, and a resolution of 5 μm/pixel. The second molar inter-root alveolar bone was selected as the region of interest (ROI). The reconstructed images were analyzed with CTAn (Bruker), including trabecular bone volume per tissue volume (BV/TV) and trabecular number (Tb. N).
The bone tissues were treated with 10% EDTA decalcification solution for 4 weeks to complete decalcification, dehydrated in an ethanol gradient, and then embedded in paraffin. The embedded tissues were cut into 5 μm-thick sections. The sections were then dewaxed and processed for HE (Solarbio) staining and TRAP (Solarbio) staining. Neutral gum was used to seal the sections.
For immunofluorescence experiments, the sections were digested with gastric enzymes. Then, non-specific antigens were blocked with goat serum. The relevant tissues were labeled with rabbit anti-iNOS and rabbit anti-MPO antibody solutions and incubated at 4 °C overnight. Following incubation, the tissues were treated with F4/80-FITC, Ly6G-FITC fluorescent antibody solution, and goat anti-rabbit DyLight 594 fluorescent secondary antibody solution. The tissues were then incubated in a 37 °C incubator for 1 h. After washing away the unconjugated fluorescent antibody with PBS, DAPI fluorescent sealer was added dropwise. Finally, the coverslips were placed and stored at 4 °C.
The sections were photographed with a scanning microscope (Leica, Aperio VERSA 8). Alveolar bone resorption and osteoclast infiltration were analyzed using ImageScope and ImageJ.
Immunotransmission electron microscopy photography
The sample was centrifuged to collect the precipitate and then resuspended and fixed for 2 h at 4 °C with the electron microscope fixative (Servicebio, G1102). The sample was centrifuged again and washed for 3 min with 0.1 M phosphate buffer PB (pH 7.4). A 1% agarose solution, which had been heated and dissolved in advance, was added to the EP tube after cooling. The precipitate was picked up with forceps and wrapped in agarose before it solidified. The agar block was added with 1% osmium tetroxide (Ted Pella Inc.) in 0.1 M phosphate buffer PB (pH 7.4) and fixed for 2 h at room temperature away from light. The supernatant was discarded, and the samples were rinsed 3 times with 0.1 M phosphate buffer PB (pH 7.4) for 15 min each time. The tissues were sequentially dehydrated in 30%–50%–70%–80%–95%–100%–100% alcohol for 20 min each time and finally immersed in 100% acetone twice for 15 min each time. The samples were permeated with the mix at 37 °C for 2–4 h, which contained acetone (Sinaopharm Group Chemical Reagent Co., Ltd, 10000418): 812 embedding agents (SPI, 90529-77-4) = 1:1. Next, the samples were treated with the mix with acetone: 812 embedding agents = 1:2 at 37 °C overnight. Subsequently, pure 812 embedding agents were used for 5–8 h at 37 °C. The pure 812 embedding agents were poured into the embedding plate, and the samples were inserted into the plate and placed in the oven at 37 °C overnight. The plates were then polymerized in an oven at 60 °C for 48 h, and the resin blocks were removed and set aside.
The tissue blocks were cut into ultrathin sections of 60–80 nm using an ultrathin sectioning machine (Leica, Leica UC7), and the slices were retrieved with a 150-mesh copper mesh with Formvar films. The copper mesh was stained in a 2% uranyl acetate saturated alcohol solution for 8 min away from light. After staining, the slices were sequentially washed 3 times with 70% alcohol and 3 times with ultrapure water. The copper mesh was then stained with 2.6% lead citrate solution for 8 min away from carbon dioxide. The slices were washed 3 times with ultrapure water and blotted slightly on filter paper. Copper mesh sections were placed in a copper mesh box and dried overnight at room temperature. The sections were observed under a transmission electron microscope (Hitachi, HT7800/HT7700), and images were collected.
MV extraction
An appropriate amount of overnight S. oralis was diluted to 8 L with TSB medium, added to several Petri dishes, and placed in a 37 °C incubator for 1 h. The supernatant was discarded, and the biofilm was resuspended in PBS. The collected bacterial broth was centrifuged at 10,000 rpm for 5 min to obtain the supernatant. The supernatant was sequentially passed through 0.45 and 0.22 μm filters (Millipore) to remove the bacteria and retain the liquid-containing vesicles. The supernatant was added to a 100 kDa ultrafiltration tube (Millipore) and centrifuged at 4 °C, 5000 rpm for concentration. Subsequently, the MV precipitates were obtained by centrifugation at 150,000 rpm for 3 h in an ultra-high-speed frozen centrifuge (Beckman). Protein concentration was measured using a BCA Protein Quantification Kit (Vazyme).
Pull-down assay
S. oralis with pDL278-tuf-his-tag and pDL278-eno-his-tag were centrifuged to collect the precipitate. SMM Buffer (0.02 M maleic acid, 0.5 M sucrose, and 0.02 M MgCl2) containing 5 mg/mL lysozyme (Biosharp) was prepared and supplemented with phenylmethylsulfonylfluoride (PMSF). The bacterial precipitate was resuspended with SMM Buffer and shaken at 37 °C, 220 rpm for 30 min in a shaker. The bacterial suspension was centrifuged at 4 °C, 6000 rpm for 5 min, and the supernatant was removed. The precipitate was repeatedly washed with PBS to remove the lysozyme. The bacteria precipitates were resuspended with an appropriate amount of SMM Buffer containing PMSF and sonicated on ice for 3 h until the suspension was clarified. The supernatant was collected after centrifugation and filtered with the 0.22 μm filters. The MVs were stored at −80 °C.
For the purified protein pull-down assay, the purified protein was diluted to 4 mg mL−1 and mixed in equal amounts. To characterize the inhibitory effect of simeprevir, 1 μM of simeprevir was added to the mixed protein solution.
An appropriate amount of magnetic beads was prepared by taking 30 μL of each sample and washing several times with Binding Buffer according to the commercial instructions of Protein A (or A/G) Immunoprecipitation Kit (Beaver Biosciences). The magnetic beads were resuspended with 1 mL of Binding Buffer, added with 30 μL of His-Tag antibody (ABclonal), and mixed on a shaker for 30 min at room temperature. The supernatant was discarded, and the unbound antibody was removed by washing several times with Binding Buffer. The pretreated beads were added to the samples and shaken overnight at 4 °C. After discarding the supernatant, the beads were washed several times with Washing Buffer, and the proteins on the beads were extracted with 100 μL of Elution Buffer. The mixture was added to a final concentration of 1×SDS protein loading buffer (Biosharp) and boiled at 98 °C for 10 min. The protein supernatant was collected and stored at −80 °C.
Western blotting
A final concentration of 1×SDS protein loading buffer was added to the samples, and protein denaturation was completed at 98 °C. The denatured proteins were performed to electrophoresis and transferred to PVDF (Roche) membranes. After washing with Tris-buffered saline Tween (TBST), the membranes were blocked with QuickBlock Blocking Buffer (Beyotime) for 15 min at room temperature on a shaker. The membranes were incubated with 1:1000 Rabbit anti-Flag-Tag antibody (MBL) at 4 °C overnight. The membranes were washed with TBST and incubated with 1:8000 Goat Anti-Rabbit IgG (H + L) antibody (ABclonal) for 1 h at room temperature on a shaker. After washing with TBST, the signal of Enoalse-Flag was detected with the Super-sensitive ECL chemiluminescent substrate (Biosharp). Subsequently, the membranes were washed with TBST and treated with Stripping Buffer (CWBio) for 10 min in a room-temperature shaker. After washing off the supernatant with TBST, the membranes were blocked with QuickBlock Blocking Buffer at room temperature and incubated with 1:1000 Mouse anti-His-Tag antibody (ABclonal) at 4 °C overnight, and the protein signal of EF-Tu-His was detected. The Western Blot results were quantified using ImageJ.
Oral plaque clinical trials
Twelve healthy oral volunteers who met the criteria were screened and divided equally into 2 groups of 6 each, one group for dynamic testing of the effect of simeprevir mouthwash on oral plaque growth and one group for plaque testing of the effect of simeprevir mouthwash on assisted toothbrushing. For the dynamic test group of the effect of simeprevir mouthwash on oral plaque growth, the volunteers were first cleaned with oral hygiene to remove plaque. Volunteers were gargled with 1 μM of simeprevir mouthwash for 30 s and were prohibited from brushing their teeth during the test period. The plaque staining was performed with a plaque indicator at 1, 2, 4, 6, 8 h after rinsing, and the Quigley–Hein plaque index was counted and photographed. The 6 volunteers were divided into 2 groups of 3 each again, and the subgingival plaques of 9 dental sites were collected with probes at 24 and 48 h, respectively, for measuring the bacterial density with plate counting. Similar to the above procedure, volunteers rinsed their mouths with double-distilled water and then performed their own control experiment. For the plaque test group of the effect of simeprevir mouthwash on assisted brushing, volunteers were first given an oral hygiene cleaning to remove plaque. The volunteers brushed their teeth twice a day during the test period and rinsed 1 μM of simeprevir mouthwash for 30 sec after brushing. Plaque staining with plaque indicator was performed on day 1, day 3, day 5, and day 7, and the Quigley–Hein plaque index was counted, photographed, and recorded. Similar to the above procedure, volunteers rinsed their mouths with double-distilled water and then performed their own control experiment.
Statistical analysis
Statistical significance was assessed by appropriate tests (see figure legends). Analysis was performed using GraphPad Prism 8.0, and P < 0.05 was considered significant. The unpaired t-test, One-way ANOVA, and Two-way ANOVA were used, and the asterisks indicate the level of significance: *P < 0.05, **P < 0.01, ***P < 0.001. No statistical methods were used to predetermine the sample size, and the researchers were not blinded to the allocation during the experiment and outcome assessment. No data were excluded from the analysis.
