Correspondence: Koji Nakayama, Division of Microbiology and Oral Infection, Department of Molecular Microbiology and Immunology, Nagasaki University Graduate School of Biomedical Sciences, Nagasaki 852-8588, Japan. Tel.: +81 95 819 7648; fax: +81 95 819 7650; e-mail: email@example.com
The Gram-negative bacterium Porphyromonas gingivalis possesses a number of potential virulence factors for periodontopathogenicity. In particular, cysteine proteinases named gingipains are of interest given their abilities to degrade host proteins and process other virulence factors such as fimbriae. Gingipains are translocated on the cell surface or into the extracellular milieu by the Por secretion system (PorSS), which consists of a number of membrane or periplasmic proteins including PorK, PorL, PorM, PorN, PorO, PorP, PorQ, PorT, PorU, PorV (PG27, LptO), PorW and Sov. To identify proteins other than gingipains secreted by the PorSS, we compared the proteomes of P. gingivalis strains kgp rgpA rgpB (PorSS-proficient strain) and kgp rgpA rgpB porK (PorSS-deficient strain) using two-dimensional gel electrophoresis and peptide-mass fingerprinting. Sixteen spots representing 10 different proteins were present in the particle-free culture supernatant of the PorSS-proficient strain but were absent or faint in that of the PorSS-deficient strain. These identified proteins possessed the C-terminal domains (CTDs), which had been suggested to form the CTD protein family. These results indicate that the PorSS is used for secretion of a number of proteins other than gingipains and that the CTDs of the proteins are associated with the PorSS-dependent secretion.
The Gram-negative bacterium Porphyromonas gingivalis, a major pathogen of periodontal disease, possesses a number of virulence factors, including fimbriae, hemagglutinins, lipopolysaccharides and proteinases. Extracellular and surface proteinases with high hydrolytic activities named gingipains are of particular importance as they have the ability to destroy periodontal tissue directly and/or indirectly (Potempa et al., 2000; Andrian et al., 2007). Gingipains are encoded by three separate genes, rgpA, rgpB and kgp, on the P. gingivalis chromosome (Curtis et al., 1999). The kgp and rgpA genes encode polyproteins comprising the signal peptide, propeptide, Lys- and Arg-specific proteinase domains, adhesin domains and C-terminal domain (CTD). The rgpB gene encodes a protein comprising the signal peptide, propeptide, Arg-specific proteinase domain and CTD. These proteins are synthesized as polyproteins in the cytoplasm, are translocated across two membranes, inner and outer membranes, and secreted onto the bacterial cell surface.
In our previous studies (Sato et al., 2010; Shoji et al., 2011) we found that gene products of rgpA, rgpB and kgp were translocated across the outer membrane by the Por secretion system (PorSS) in which porK, porL, porM, porN, porO, porP, porQ, porT, porU, porV (PG27, lptO), porW and sov genes were involved. Expression of some of these genes is regulated by a two-component system, the PorX response regulator and PorY histidine sensor kinase (Sato et al., 2010). Primary gene products of rgpA, rgpB and kgp have common motifs in their CTD regions. The P. gingivalis genome encodes a number of putative CTD-containing proteins (Seers et al., 2006). Nguyen et al. (2007) showed that CTD-containing proteins were also found in predicted proteins of other bacteria in the Bacteroidetes phylum, such as Prevotella intermedia and Tannerella forsythia. Among P. gingivalis CTD-containing proteins, TapA and HBP35 have been found to be translocated across the outer membrane by the PorSS (Kondo et al., 2010; Shoji et al., 2011). Studies of Gram-negative bacteria have identified at least eight different protein secretion systems, including types I–VI, the two-partner secretion system and the chaperone/usher system (Economou et al., 2006). PorSS is not related to the previously known bacterial protein secretion systems.
PorK, PorL, PorM, PorN, PorW, PorT and Sov involved in the P. gingivalis PorSS share similarity in amino acid sequence with Flavobacterium johnsoniae gliding motility proteins, GldK, GldL, GldM, GldN, SprE, SprT and SprA, respectively (Braun et al., 2005; Rhodes et al., 2010; Sato et al., 2010). In F. johnsoniae, disruption of the sprT gene resulted in defects in translocation of the gliding motility protein SprB and secretion of chitinase, suggesting that the PorSS is linked to gliding motility of bacteria in the Bacteroidetes phylum (Sato et al., 2010). Genes homologous to the PorSS-related genes are also found in genomes of other members of the Bacteroidetes phylum, including important periodontal pathogens such as P. intermedia and T. forsythia (Sato et al., 2010). In the present study, proteomic analyses of particle-free culture supernatants and vesicle fractions from porK+ and porK strains with the genetic background of rgpA rgpB kgp and outer membrane fractions from wild-type and porK strains were performed to identify P. gingivalis proteins that were secreted into the extracellular milieu by the PorSS.
Materials and methods
Bacterial strains and culture conditions
Bacterial strains and plasmids used in this study are listed in Table 1. Porphyromonas gingivalis cells were grown anaerobically (10% CO2, 10% H2, 80% N2) in enriched brain heart infusion medium and on enriched trypticase soy agar (Nakayama et al., 1995). For blood agar plates, defibrinated laked sheep blood was added to enriched trypticase soy agar at 5%. For selection and maintenance of antibiotic-resistant P. gingivalis strains, antibiotics were added to the medium at the following concentrations: erythromycin (Em), 10 μg mL−1; tetracycline (Tc), 0.7 μg mL−1.
Table 1. Bacterial strains and plasmids used in this study
Porphyromonas gingivalis deletion mutants were constructed as follows. DNA regions upstream and downstream of a gene were PCR-amplified from the chromosomal DNA of P. gingivalis ATCC 33277T using pairs of primers (PGN gene number-U-F plus PGN gene number-U-R and PGN gene number-D-F plus PGN gene number-D-R), respectively, where ‘U’ indicates upstream, ‘F’ indicates forward, ‘D’ indicates downstream and ‘R’ indicates reverse. Primers used in this study are listed in Supporting information, Table S1. Amplified DNAs upstream and downstream of each gene were double-digested with NotI plus BamHI and KpnI plus BamHI, respectively. Both digested products were ligated together with pBluescript II SK(−) which had been digested with NotI plus KpnI, resulting in pKD945 (for rgpA mutagenesis) and pKD947 (for rgpB mutagenesis). The 1.5-kb BamHI cepA (pKD1002) and 2.7-kb BamHI–BglII tetQ (pKD375) DNA fragment was inserted into the BamHI site of pKD945 and pKD947 to yield pKD946 (ΔrgpA::cepA) and pKD948 (ΔrgpB::tetQ), respectively. pKD946 was digested with NotI and KpnI and introduced into P. gingivalis KDP129 (kgp) by electroporation to yield strain KDP980 (kgp::cat ΔrgpA::cepA). pKD948 was digested with NotI and KpnI and introduced into P. gingivalis KDP980 by electroporation to yield strain KDP981 (kgp::cat ΔrgpA::cepA ΔrgpB::tetQ). Porphyromonas gingivalis KDP981 was then transformed to be Em-resistant with NotI–KpnI-digested pKD981 (ΔporK::ermF) to yield strain KDP982 (kgp::cat ΔrgpA::cepA ΔrgpB::tetQ ΔporK::ermF).
Particle-free culture supernatant and vesicle fractions were obtained as described previously (Potempa et al., 1995). Porphyromonas gingivalis cell cultures were centrifuged at 6000 g for 30 min at 4 °C and the culture supernatant was separated from pellet cells. The culture supernatant was subjected to ultracentrifugation at 100 000 g for 60 min at 4 °C and the particle-free culture supernatant was separated from vesicles. The proteins in the particle-free culture supernatant and vesicle fractions were precipitated with 10% trichloroacetic acid at 4 °C and the precipitated proteins were harvested by centrifugation at 4 °C for 20 min and the pellet was washed three times with cold diethyl ether, dried at room temperature for 30 min and the pellet resuspended in cell lysis solution (7 M urea, 2 M thiourea, 4% CHAPS, 1 mM EDTA and 5 mM tributylphosphine).
For isolation of the outer membrane fraction, P. gingivalis cells were harvested by centrifugation at 10 000 g for 30 min at 4 °C and resuspended with PBS containing 0.1 mM N-alpha-tosyl-L-lysine chloromethyl ketone (TLCK) and 0.1 mM leupeptin. Cells were disrupted in a French pressure cell at 100 Mpa by two passes. The remaining intact bacterial cells were removed by centrifugation at 2400 g for 10 min, and the supernatant was subjected to ultracentrifugation at 100 000 g for 60 min at 4 °C. The pellet was then treated with 1% (v/v) Triton X-100 in PBS containing 20 mM MgCl2 for 30 min at 20 °C. The outer membrane fraction was obtained as a precipitate by ultracentrifugation at 100 000 g for 60 min at 4 °C.
Two-dimensional gel electrophoresis (2D-PAGE)
Sample was applied to an IPG strip (13 cm; GE Healthcare) with a pH range from 4 to 7 (first dimension) swollen with a rehydration solution [7 M urea, 2 M thiourea, 4% CHAPS, 0.5% IPG buffer (pH 4–7; GE Healthcare), 1 mM EDTA, 12 μL mL−1 destreak reagent (GE Healthcare), and bromophenol blue]. The second dimension (SDS-PAGE) was performed in polyacrylamide gels and the proteins were stained with Coomassie Brilliant Blue R250.
MS analysis and database search for protein identification
Proteins were identified by peptide mass fingerprinting (PMF) after in-gel tryptic digestion as previously described (Sato et al., 2010). A gel plug containing proteins was subjected to the following procedures: washing with 50% (v/v) acetonitrile, washing with 100% acetonitrile, reduction with 10 mM DTT, alkylation with 55 mM iodoacetamide, washing/dehydration with 50% (v/v) acetonitrile, and digestion for 10 h with 10 μg mL−1 trypsin. The resulting peptides were extracted from the gel plug with 0.1% (v/v) trifluoroacetic acid/50% (v/v) acetonitrile. Digests were spotted on a MALDI target using α-cyano-4-hydroxycinnamic acid as a matrix. Spectra were acquired on a 4800 MALDI TOF/TOF analyser (Applied Biosystems, Foster City, CA). Data analysis and MS database searching were performed using GPS Explorer™ and mascot software.
Quantification of mRNA by real-time quantitative PCR (qPCR)
Total RNA was isolated from P. gingivalis cells grown to mid-exponential phase (OD600 nm of c. 1.0) by using an RNeasy Mini kit (Qiagen Science). DNA was removed with RNase-Free DNase. cDNA was generated in a reaction mixture containing a random primer (Promega), dNTP mixture, RNase inhibitor, DTT, Superscript III Reverse Transcriptase (Invitrogen) and DEPC-treated water. Real-time qPCR was performed using Brilliant SYBR Green II QPCR Master Mix (Stratagene) with an Mx3005P™ Real-Time PCR System (Stratagene) according to the manufacturer's instructions. Primers for the real-time qPCR are listed in Table S1 and were designed using the primer3 program. Real-time qPCR conditions were as follows: one cycle at 95 °C for 10 min, and 35 cycles of 95 °C for 30 s and 60 °C for 1 min. At each cycle, accumulation of PCR products was detected by the reporter dye from the dsDNA-binding SYBR Green. To confirm that a single PCR product was amplified, after the PCR, a dissociation curve (melting curve) was constructed in the range 55–95 °C. All data were analysed using Mx3005P software. The expression level of each targeted gene was normalized to that of the 16S rRNA gene, which was used as a reference. All PCR reactions were carried out in triplicate. The efficiency of primer binding was determined by linear regression by plotting the cycle threshold (CT) value versus the log of the cDNA dilution. Relative quantification of transcript was determined using the comparative CT method () calibrated to 16S rRNA gene. qPCR experiments were performed multiple times independently, yielding comparable results.
Results and discussion
Identification of PorSS-dependently secreted proteins of P. gingivalis
We constructed an rgpA rgpB kgp porK mutant from an rgpA rgpB kgp strain and compared secreted proteins between the rgpA rgpB kgp and rgpA rgpB kgp porK strains to avoid degradation of secreted and surface proteins by gingipains as the wild-type strain secreted gingipains that had the ability to process both secreted and surface proteins, while the porK mutant secreted no gingipains. 2D-PAGE of the particle-free (membrane-free) culture supernatants from the kgp rgpA rgpB and kgp rgpA rgpB porK mutants was performed. As a control, three protein spots in each 2D gel, which exhibited the same amounts of proteins with the same molecular masses and isoelectric points, were subjected to MALDI-TOF mass analysis, resulting in the same proteins (PGN_0916, PGN_1367 and PGN_1587; Fig. 1). Their molecular masses and isoelectric points calculated from their amino acid sequences were 69 044 and 4.88 for PGN_0916, 49 199 and 5.99 for PGN_1367, and 30 130 and 5.30 for PGN_1587, respectively, which were consistent with their positions on the 2D gels. At least 16 protein spots, which were present in the particle-free culture supernatant of the kgp rgpA rgpB strain, were absent or faint in that of the kgp rgpA rgpB porK mutant (Fig. 1). Relative amounts (kgp rgpA rgpB porK versus kgp rgpA rgpB) of the protein spots were calculated (Table 2). The protein spots were then subjected to MALDI-TOF mass analysis. PMF analysis of the spots, in comparison with the genome database of P. gingivalis ATCC 33277T (Naito et al., 2008), allowed the identification of 10 proteins (Table 2). An immunoreactive 46-kDa antigen (PGN_1767) was identified in two different protein spots [spot 10 (33 kDa) and spot 8 (42 kDa)]. Both 33- and 42-kDa PGN_1767 proteins contained the D42-R66 fragment at the most N-terminal position, whereas the 42-kDa protein possessed the G403-R418 fragment in the CTD, but the 33-kDa protein did not, suggesting that the 42-kDa PGN_1767 protein was processed at the C-terminal end to yield the 33-kDa PGN_1767 protein. PGN_0659 (35-kDa hemin binding protein, HBP35) was identified in four different spots [one (spot 9) with a molecular mass of 36 kDa and three (spots 12, 13 and 14) with a molecular mass of 28 kDa] in 2D-PAGE. The three 28-kDa protein spots had different isoelectric points. All of the 28- and 36-kDa HBP35 proteins contained the A61-K87 fragment at the most N-terminal position, whereas the 36-kDa protein possessed the D244-R329 fragment at the C-terminal end, but the 28-kDa proteins had the E234-K273 or D244-K273 fragment, suggesting that the 36-kDa HBP35 protein was processed at the C-terminal end to yield the 28-kDa HBP35 proteins. HBP35 exhibits thioredoxin and hemin-binding activities and has an important role in heme acquisition for growth (Shoji et al., 2010). PGN_0898 (spot 15) is a bacterial peptidylarginine deiminase (PAD). Wegner et al. (2010) showed that deletion of the PAD (PGN_0898)-encoding gene resulted in complete abrogation of protein citrullination. Inactivation of Arg-gingipains, but not Lys-gingipain, led to decreased citrullination, suggesting that host peptides generated by proteolytic cleavage at Arg-X peptide bonds by Arg-gingipains were citrullinated at the C terminus by PAD. Citrullinated bacterial and host peptides may cause the autoimmune response in rheumatoid arthritis (Lundberg et al., 2010). CPG70 (PGN_0335, spot 4) exhibits Lys- and Arg-specific metallocarboxypeptidase activity. A previous study (Chen et al., 2002) suggested that CPG70 may have an important role in C-terminal processing of cell surface proteins containing Arg-gingipains, Lys-gingipain and adhesins of P. gingivalis. TapA (PGN_0152) was identified in two different protein spots [spot 7 (44 kDa) and spot 6 (48 kDa)]. TapA is associated with the periplasmic tetratricopeptide repeat protein TprA that is upregulated in host tissues of the subcutaneous chamber model, and is involved in the virulence of P. gingivalis W83 as mice infected with the tapA and tprA mutants showed higher survival rates than those infected with the wild-type (Yoshimura et al., 2008; Kondo et al., 2010). PGN_1416 (spot 3) is considered to be a lysyl endopeptidase in the P. gingivalis genome database, whereas PGN_0291 (spots 1 and 2), PGN_0654 (spot 11), PGN_0795 (spot 5) and PGN_1476 (spot 16) are hypothetical proteins (Naito et al., 2008). A number of proteins appeared to be more abundant in the particle-free supernatant of the rgpA rgpB kgp porK strain than that of the rgpA rgpB kgp strain, particularly in the pH 6–7/35- to 55-kDa region of the gel. However, it was not reproducible and the proteins included no CTD proteins (data not shown).
Table 2. Identification of protein spots shown in Fig. 1
Ratio of the amount of each protein spot in the kgp rgpA rgpB porK mutant to that in the kgp rgpA rgpB mutant.
Probable lysyl endopeptidase precursor
Conserved hypothetical protein with zinc carboxypeptidase domain (CPG70)
Immunoreactive 61-kDa antigen (TapA)
Immunoreactive 61-kDa antigen (TapA)
Immunoreactive 46-kDa antigen
35-kDa hemin binding protein (HBP35)
Immunoreactive 46-kDa antigen
Hypothetical protein (putative lipoprotein)
35-kDa hemin binding protein (HBP35)
35-kDa hemin binding protein (HBP35)
35-kDa hemin binding protein (HBP35)
Probable peptidylarginine deiminase (PAD)
Next, we compared a 2D-gel profile of the vesicle fraction of the rgpA rgpB kgp strain with that of the rgpA rgpB kgp porK strain (Fig. 2). We found two (spots 26 and 39) and seven (spots 45, 46, 48, 49, 53, 56 and 59) spots of CTD proteins in the vesicle fractions of the rgpA rgpB kgp and rgpA rgpB kgp porK strains, respectively. Five spots (spots 45, 46, 48, 56 and 59) were only observed in the rgpA rgpB kgp porK strain. CPG70 (PGN_0335) was identified in spots 45 and 46 (Table S2). PGN_1476, PAD (PGN_0898) and TapA (PGN_0152) were identified in spots 48, 56 and 59, respectively. An immunoreactive 46-kDa antigen (PGN_1767) and HBP35 (PGN_0659) were identified in both strains, but spot 49 (PGN_1767) and spot 53 (PGN_0659) of the rgpA rgpB kgp porK strain were clearly larger than spot 26 (PGN_1767) and spot 39 (PGN_0659) of the rgpA rgpB kgp strain, respectively. These six CTD proteins were also found in the particle-free supernatant of the rgpA rgpB kgp porK strain (Table 2). Molecular mass and PMF analyses revealed that the six CTD proteins found in the particle-free culture supernatant of the rgpA rgpB kgp strain were processed at the C terminus compared with those found in the vesicle fraction of the rgpA rgpB kgp porK strain (Table 3, Figs S1 and S2).
Table 3. Comparison of molecular masses of protein spots
Molecular mass of protein spot (kDa)
Particle-free culture supernatant of rgpA rgpB kgp in Fig. 1
Numbers in parentheses indicate protein spot numbers in each figure.
HP, hypothetical protein; IR, immunoreactive.
48 (6), 44 (7)
100 (45), 98 (46)
36 (9), 28 (12, 13, 14)
IR 46 kDa antigen
42 (8), 33 (10)
We also determined patterns of 2D-gels of the outer membrane fractions from the wild-type and porK strains (Fig. 3, Table S3). Spot 4 (PGN_1689), spot 5 (PGN_1366), spot 6 (PGN_0287), spot 8 (PGN_0293), spot 10 (PGN_1432), spot 11 (PGN_1808), spot 13 (PGN_0293), spot 14 (PGN_0293), spot 18 (PGN_0290), spot 19 (PGN_0293), spot 20 (PGN_0293) and spot 23 (PGN_0729) were present only in the wild-type. Judging from the molecular masses of the protein spots, the proteins appeared to be proteolytically processed products in the wild-type. None of them possessed CTD at the C-terminal region.
To examine the effect of the porK mutation on the expression of the 10 secreted protein-encoding genes at a transcriptional level, RT-PCR analysis was performed and relative amounts of mRNA were determined (Fig. 4). The genes encoding PGN_0291, PGN_0335 and PGN_0898 in the PorSS-deficient strain (kgp rgpA rgpB porK) were expressed at the same level as those in the PorSS-proficient strain (kgp rgpA rgpB), whereas the genes encoding PGN_0152, PGN_0654, PGN_0659, PGN_0795, PGN_1416 and PGN_1767 were about 50% downregulated in the PorSS-deficient strain compared with the PorSS-proficient strain. The gene encoding PGN_1476 in the PorSS-deficient strain was expressed about three times more than that in the PorSS- proficient strain. As the relative amounts of the protein spots were < 20% (Table 2), the results suggest that decrease of the 10 secreted proteins in the PorSS-deficient mutant are mostly dependent on the defect in the PorSS.
The 10 PorSS-dependently secreted proteins as well as precursor forms of Arg-gingipains (RgpA and RgpB) and Lys-gingipain (Kgp) had CTDs in which the conserved DxxG and GxY motifs and the conserved Lys residue are located (Seers et al., 2006; Fig. 5). Seers et al. (2006) reported that 34 CTD family proteins with sequence similarity to the C-terminal region of the RgpB precursor were identified by a blast search with the P. gingivalis W83 genome, which include the 10 proteins identified in the present study. Slakeski et al. (2010) suggested that the CTD of RgpB is essential for covalent attachment to the cell surface by an A-LPS anchor containing anionic polysaccharide repeating units. In our previous studies (Kondo et al., 2010; Shoji et al., 2011), we demonstrated that HBP35 and TapA were modified by A-LPS and anchored on the bacterial cell surface. In addition, the green fluorescent protein–CTD fusion study revealed that the CTDs of CPG70, PAD and HBP35 as well as RgpB play roles in PorSS-dependent translocation and glycosylation (Shoji et al., 2011). We suggested in the study both that the CTD region functions as a recognition signal for the PorSS and that glycosylation of CTD proteins occurs after removal of the CTD region. Cleaved CTD fragments of HBP35, CPG70, PAD, RgpB and PGN_1767 have recently been found in the culture supernatants of P. gingivalis (Glew et al., 2012), which is consistent with the present study and supports our model (Shoji et al., 2011).
Our results strongly indicate that the P. gingivalis secreted proteins with CTDs, which are responsible for colony pigmentation, hemagglutination, adherence and modification/processing of the bacterial surface proteins and host proteins, are translocated to the cell surface by the PorSS.
In the present study, using 2D-PAGE and MS we identified 10 proteins secreted into the extracellular milieu by the PorSS. All of the proteins possessed CTDs. They included HBP35 in heme acquisition, TapA in virulence, PAD in citrullination of C-terminal Arg residues of the surface proteins and CPG70 in processing of C-terminal Arg and Lys residues. These results indicate that the PorSS is used for secretion of a number of proteins other than gingipains and that the CTDs of the proteins are associated with the PorSS-dependent secretion.
Note added post-publication
As the secretion system and components of PorSS are not similar to those of the type I to type VIII protein secretion systems, we suggest that the PorSS could be referred to as the type IX secretion system (T9SS).
We thank Dr Fuminobu Yoshimura and Ms Mikie Sato for help with PMF analysis. This work was supported by Grants-in-Aid for Scientific Research (to K.S. and K.N.) from the Ministry of Education, Culture, Sports, Science, and Technology of Japan and the Global COE Program at Nagasaki University (to K.N.).