Article pubs.acs.org/jpr
Proteomic and Transcriptional Analysis of Interaction between Oral Microbiota Porphyromonas gingivalis and Streptococcus oralis Kazuhiko Maeda,* Hideki Nagata, Miki Ojima, and Atsuo Amano Department of Preventive Dentistry, Osaka University Graduate School of Dentistry, Suita, Osaka 565-0871, Japan S Supporting Information *
ABSTRACT: Porphyromonas gingivalis, a major periodontal pathogen, forms biofilm with other oral bacteria such as streptococci. Here, by using shotgun proteomics, we examined the molecular basis of mixed-biofilm formation by P. gingivalis with Streptococcus oralis. We identified a total of 593 bacterial proteins in the biofilm. Compared to the expression profile in the P. gingivalis monobiofilm, the expression of three proteins was induced and that of 31 proteins was suppressed in the mixed biofilm. Additionally, the expression of two S. oralis proteins was increased, while that of two proteins was decreased in the mixed biofilm, as compared to its monotypic profile. mRNA expression analysis of selected genes using a quantitative reverse transcription polymerase chain reaction confirmed the proteomics data, which included overexpression of P. gingivalis FimA and S. oralis glyceraldehyde-3-phosphate dehydrogenase in association with the biofilm. The results also indicated that S. oralis regulates the transcriptional activity of P. gingivalis luxS to influence autoinducer-2-dependent signaling. These findings suggest that several functional molecules are involved in biofilm formation between P. gingivalis and S. oralis. KEYWORDS: Porphyromonas gingivalis, Streptococcus oralis, biofilm, microbiology, interaction, shotgun proteomics, quantitative RT-PCR
■
interacted with the FimA fimbriae with significantly high affinity and specificity.19 GAPDH is a tetrameric enzyme in the glycolytic pathway and is responsible for the phosphorylation of glyceraldehyde-3phosphate, which leads to the generation of 1,3-bisphosphoglycerate.20 However, multiple functions have been recently reported for GAPDH, including those in membrane fusion, microtubule binding, phosphotransferase activity, nuclear RNA export, DNA replication and repair, apoptosis, and viral pathogenesis.21,22 S. oralis GAPDH (GapC) shows high affinity for five P. gingivalis client proteins (tonB-dependent receptor protein (RagA4), 4-hydroxybutyryl-coenzyme A dehydratase (AbfD), GAPDH, NAD-dependent glutamate dehydrogenase (GDH), and malate dehydrogenase (MDH)), which suggests that these proteins regulate P. gingivalis biofilm formation with oral streptococci.23 P. gingivalis produces numerous adhesins, proteinases, and hemin-uptake systems along with communication signals, which contribute to colonization, persistence, and pathogenic potential such as the autoinducer AI-2.24−26 The pathogen is one of the organisms that produces LuxS, and its AI-2 activity is regulated under luxS transcription.25
INTRODUCTION Periodontal disease is an infectious disorder caused by complex interactions within a small subset of periodontal pathogens harbored in oral biofilms (dental plaque).1 Among these, Porphyromonas gingivalis, a Gram-negative anaerobic bacterium, is considered the most pathogenic.2−4 Interactions between P. gingivalis and early plaque-forming bacteria facilitate its colonization in the periodontal regions.5−7 P. gingivalis interacts with a variety of oral Gram-positive bacteria, including Actinomyces naeslundii,8 Actinomyces viscosus,9−11 Streptococcus gordonii,12,13 Streptococcus oralis,13,14 Streptococcus mutans,15 and Streptococcus sanguinis.13,16 Recently, the influence of early colonizing species on the structure and composition of the developing bacterial community in an in vitro subgingival 10-species biofilm was investigated.17 The biofilm model included S. oralis, Streptococcus anginosus, Actinomyces oris, Fusobacterium nucleatum, Veillonella dispar, Campylobacter rectus, Prevotella intermedia, P. gingivalis, Tannerella forsythia, and Treponema denticola. In the absence of streptococci, the overall biofilm was found to contain few P. gingivalis cells, and its structure was loose and dispersed. Therefore, it is likely that Streptococci facilitate the establishment of P. gingivalis in subgingival biofilms. We previously showed that P. gingivalis FimA fimbriae were involved in coaggregation with S. oralis, which mediated biofilm formation.18 It was also demonstrated that oral streptococcal cell surface glyceraldehyde-3-phosphate dehydrogenase (GAPDH) © XXXX American Chemical Society
Special Issue: Environmental Impact on Health Received: August 15, 2014
A
dx.doi.org/10.1021/pr500848e | J. Proteome Res. XXXX, XXX, XXX−XXX
Journal of Proteome Research
Article
trifluoroacetic acid (Wako). After centrifugation, the aqueous phase was desalted with C18-StageTips.31
Two similar isobaric chemical labeling methods, isobaric tags for relative and absolute quantification (iTRAQ)27 and tandem mass tags (TMT),28 have been widely used for quantitative proteomics. The performance of these quantification methods is better than that of label-free approaches.29 However, label-free quantification is also popular, and we used the spectral counting approach because it enables the simultaneous identification and quantification of proteins without the laborious and costly process of introducing stable isotopes into samples. In this study, we used a label-free shotgun proteome analysisbased approach to uncover the molecular basis of P. gingivalis− S. oralis mixed-biofilm formation. The proteomic profile suggests that several functional proteins play key roles in the biofilm formation between P. gingivalis and S. oralis.
■
Shotgun LC−MS/MS Analysis
LC−MS/MS analysis was performed using an UltiMate 3000 Nano LC System (Thermo Fisher Scientific, Waltham, MA) coupled to a Q-ExactiveHybrid Quadrupole-Orbitrap Mass Spectrometer (Thermo Fisher Scientific) with a nanoelectrospray ionization source.32,33 The digested sample was injected and enriched on a C18 Reversed Phase Trap Column (100 μm I.D. × 5 mm length; Thermo Fisher Scientific) at a flow rate of 4 μL/min. The C18 Reversed Phase Column (75 μm I.D. × 150 mm length; Nikkyo Technos Co. Ltd., Tokyo, Japan) was equilibrated with 2% acetonitrile with 0.1% formic acid. The sample was subsequently separated by the column at a flow rate of 300 nL/min with a linear gradient from 2−35% with 95% acetonitrile with 0.1% formic acid. The peptides were ionized using nanoelectrospray ionization in positive ion mode. Survey scans and MS/MS scans were obtained in positive ionization mode using a data-dependent acquisition. Survey scans were acquired at a resolving power of 70 000 at m/z 400−2000. Datadependent acquisition was used to select the 10 most intense precursor ions with an isolation window of 2 Th and fragmented by higher-energy collisional dissociation (HCD) with normalized collision energies of 32%. The resolution for HCD spectra was set to 17 500. The automatic gain controls for survey scans and MS/MS scans were set to 1 × 106 and 5 × 104 and the maximum ion injection times were 30 and 50 ms, respectively. Dynamic exclusion was set to 20 s.
MATERIALS AND METHODS
Bacterial Strains and Growth Conditions
P. gingivalis ATCC 33277 and S. oralis ATCC 9811 were maintained as frozen stocks in our laboratory. P. gingivalis was cultured in prereduced trypticase soy broth (Becton, Dickinson, and Company (BD), Sparks, MD) containing 1 mg/mL yeast extract (BD), 5 μg/mL hemin (Sigma-Aldrich Japan K. K., Tokyo, Japan), and 1 μg/mL menadione (Sigma-Aldrich) for 24 h in the Anaerobic System 1025 (Forma Scientific Inc., Marietta, OH) with an 80% N2/10% CO2/10% H2 atmosphere at 35 °C. S. oralis was cultured anaerobically at 37 °C for 16 h in brain heart infusion (BHI) broth (BD). Proteomics of Model Biofilm Formation
Protein Identification
P. gingivalis and S. oralis were cultured to the mid log phase. They were harvested by centrifugation at 5000 × g for 30 min and washed with prereduced 20 mM phosphate buffer containing 0.15 M NaCl (PBS, pH 6.0) twice. After centrifugation at 5000 × g for 30 min, the bacteria were resuspended in prereduced PBS. S. oralis (1.5 × 109 CFU/mL) in prereduced PBS (100 mL) were mixed with an equal number of P. gingivalis in prereduced PBS (100 mL). S. oralis or P. gingivalis alone were used as controls and resuspended in prereduced PBS (200 mL). The mixtures of bacteria were held in prereduced PBS (200 mL) in an anaerobic chamber at 37 °C for 12 h. After centrifugation at 10 000 × g for 30 min, the bacteria were pelleted and lysed by sonication in 10 mM Tris-HCl (pH 9.0), Phosphatase Inhibitor Cocktail II (Calbiochem, Darmstadt, Germany), endonuclease (Life Technologies, Carlsbad, CA), and complete ethylenediaminetetraacetic acid-free Protease Inhibitor Cocktail Tablets (Roche Diagnostics, Indianapolis, IN). After centrifugation at 15 000 × g for 30 min, the supernatants were collected. Trichloroacetic acid (Wako Pure Chemical Industries, Ltd., Osaka, Japan)-induced precipitation was performed as described previously.30 After centrifugation at 15 000 × g for 30 min, the pellets were suspended in 200 μL of 50 mM Tris-HCl (pH 9.0), 6 M urea (Wako), 2 M thiourea (Wako), and 2% sodium deoxycholate (Wako).
All database searches were performed using Mascot Distiller version 2.3 (Matrix Science Ltd., London, UK) against the UniProt database with the P. gingivalis database (NCBI Taxonomy ID: 837) containing 8700 protein sequences (entries as of February 6, 2013) and the S. oralis database (NCBI Taxonomy ID: 1303) containing 18 400 protein sequences (entries as of February 6, 2013). Only tryptic peptides with up to two missed cleavages were accepted. Carbamidomethylation was a fixed modification. Oxidation of methionine; deamidation of asparagine and glutamine; and phosphorylation of serine, threonine, and tyrosine were variable modifications. MS/MS spectra were searched against the UniProt combined database with P. gingivalis, S. oralis, and common contaminants such as human keratin and trypsin. The precursor and product ion mass tolerances were 10 ppm and 0.01 Da, respectively (Tables 1−4). Scaffold (version 3; Proteome Software, Portland, OR) was used to validate peptide and protein identifications. Probability thresholds of 50% were selected for peptides and proteins, with one peptide assigned per protein. A protein quantification analysis using spectral counting was performed for the samples. The “number of assigned spectra” function of the Scaffold software was used to estimate the total number of spectra that matched to an identified protein in each sample. The false discovery rate (FDR) for proteins was less than 1.0%; total spectra were assigned with high confidence. Isoforms and members of a protein family were identified separately and were grouped under the same gene name by the Scaffold software. The quantitative Venn diagram was obtained using the freely available online tool BioVenn.34 The pI/Mw tool of the Swiss Institute of Bioinformatics (Lausanne, Switzerland) was used to determine the pI (http://web.expasy.org/compute_pi/), and the theoretical pIs and masses of the identified proteins are listed in Table S1 of the Supporting Information. The P. gingivalis proteins identified
In-Solution Digestion
The samples (100 μg) were reduced with 10 mM dithiothreitol (Wako) for 60 min at 37 °C and alkylated with 55 mM iodoacetamide (Wako) for 30 min in the dark at 25 °C. The reduced and alkylated samples were diluted 10-fold with 50 mM Tris-HCl (pH 9.0) and digested with trypsin (Roche) at 37 °C for 16 h (trypsin:protein ratio of 1:20 (w/w)). An equal volume of ethyl acetate was added to each sample solution, and the mixtures were acidified with the final concentration of 0.5% B
dx.doi.org/10.1021/pr500848e | J. Proteome Res. XXXX, XXX, XXX−XXX
Journal of Proteome Research
Article
Table 1. List of P. gingivalis Proteins Identified and Quantified in This Study to Be Induced in P. gingivalis−S. oralis Biofilm QSpec analysis accession number
gene number
descriptiona
Proteins involved in information storage and PTMs YP_001928529 PGN_0413 DNA gyrase subunit B, GyrB YP_001928754 PGN_0638 RNA polymerase sigma factor, RpoD Cell wall and membrane transport functions YP_001928296 PGN_0180 Type I fimbrillin, FimA Uncharacterized proteins YP_001930158 PGN_2043 probable transcriptional regulatory protein a
log2 ratio mixed/mono
p-value
LogFoldChange
z-statistic
FDR
2.58 2.81
0.066 0.029
0.505 0.895
1.7746 2.9452
0.027525 0.010499
4.79
0.009
2.279
17.4546
3.38
0.019
1.465
3.6803
0 0.048287
Bold fonts in description show transcriptional genes selected in Figure 3.
(rpoD), DNA gyrase subunit B (gyrB), adenosine triphosphate (ATP)-dependent DNA helicase (recQ), ATP-binding cassette (ABC) transporter ATP-binding protein (suf C), universal stress protein (uspA), immunoreactive 23 kDa antigen (ompA), carboxyl-terminal processing protease (prc), 2-oxoglutarate ferredoxin oxidoreductase subunit (oafo-sf), superoxide dismutase (sod), enolase (eno), malonyl CoA-acyl carrier protein transacylase (fabD), redox-sensing transcription factor (oxyR), rnf ABCDGE Type G (rnf G), and hemin binding lipoprotein ( fetB) were designed to produce SYBR Green-labeled probes using Beacon Designer version 7 software (PREMIER Biosoft International, Palo Alto, CA) (Table S3).39 The primer sequence information for gapA and arginine-specific cysteine proteinase (rgpB) was obtained from Maeda et al.,23 and that for fimA, luxS, and 16S rRNA was obtained from James et al.25 RNA was extracted using the PureLink RNA Mini Kit (Life Technologies) and treated with RNase-free DNase I (QIAGEN Inc., Valencia, CA). An iScript cDNA synthesis Kit (Bio-Rad Laboratories, Hercules, CA) was used to generate cDNA from RNA templates (1 μg). Real-time RT-PCR was performed on the Rotor Gene 6000 (QIAGEN) using the QuantiFast SYBR Green Kit (QIAGEN). Results were analyzed with Rotor-Gene 6000 Series Software version 1.7 (QIAGEN). Melting curve profiles were examined to verify a single peak for each sample, and the transcript copy number was calculated. RNA extracts were prepared in duplicate from independent experiments, while cDNA samples were loaded in triplicate. Transcript levels were normalized to 16S rRNA. P. gingivalis without S. oralis served as a control. Targeted mRNA levels are expressed as fold differences after contact with S. oralis compared to controls. (ii). qRT-PCR of S. oralis. Primers for 16S rRNA and GapC were designed to produce TaqMan probe using Beacon software (Table S3). RNA was extracted and used to generate cDNA as described above. Real-time RT-PCR was performed using the Rotor-Gene Probe PCR Kit (QIAGEN). RNA extracts were prepared in duplicate from independent experiments, while cDNA samples were loaded in triplicate. Transcript levels were normalized to 16S rRNA. S. oralis without P. gingivalis served as a control. Targeted mRNAs levels are expressed as fold differences after contact with P. gingivalis compared to controls.
were categorized into COGs using BLAST (http://www.ncbi. nlm.nih.gov/). Protein Quantification
Statistical methods for protein quantification were performed accoding to Vera et al.35 In brief, three independent samples (i.e., biologic replicates) were analyzed for each cell subpopulation. Spectral counts were normalized by dividing the number of spectra of each protein by the sum of all spectra in the sample and multiplying the result with the sum of spectra of the sample with the most obtained spectra. The log2 + 0.5 (to avoid undefined values) of individual spectra counts was calculated, as were the log2 ratios between the mixed biofilm and control samples. Student’s t-test (n = 3) was performed, and proteins had to fulfill the following criteria to be regarded as differentially expressed, if the log2 (normalized spectral count) ratio was >2, this protein was considered to be induced in biofilms. When the log2 ratio was
