Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Open Access

Peer-reviewed

Research Article

Proteomics of Protein Secretion by Aggregatibacter actinomycetemcomitans

Abstract

The extracellular proteome (secretome) of periodontitis-associated bacteria may constitute a major link between periodontitis and systemic diseases. To obtain an overview of the virulence potential of Aggregatibacter actinomycetemcomitans, an oral and systemic human pathogen implicated in aggressive periodontitis, we used a combined LC-MS/MS and bioinformatics approach to characterize the secretome and protein secretion pathways of the rough-colony serotype a strain D7S. LC-MS/MS revealed 179 proteins secreted during biofilm growth. Further to confirming the release of established virulence factors (e.g. cytolethal distending toxin [CDT], and leukotoxin [LtxA]), we identified additional putative virulence determinants in the secretome. These included DegQ, fHbp, LppC, Macrophage infectivity protein (MIP), NlpB, Pcp, PotD, TolB, and TolC. This finding indicates that the number of extracellular virulence-related proteins is much larger than previously demonstrated, which was also supported by in silico analysis of the strain D7S genome. Moreover, our LC-MS/MS and in silico data revealed that at least Type I, II, and V secretion are actively used to excrete proteins directly into the extracellular space, or via two-step pathways involving the Sec/Tat systems for transport across the inner membrane, and outer membrane factors, secretins and auto-transporters, respectively for delivery across the outer membrane. Taken together, our results provide a molecular basis for further elucidating the role of A. actinomycetemcomitans in periodontal and systemic diseases.

Citation: Zijnge V, Kieselbach T, Oscarsson J (2012) Proteomics of Protein Secretion by Aggregatibacter actinomycetemcomitans. PLoS ONE 7(7): e41662. https://doi.org/10.1371/journal.pone.0041662

Editor: Jamunarani Vadivelu, University of Malaya, Malaysia

Received: April 25, 2012; Accepted: June 25, 2012; Published: July 25, 2012

Copyright: © 2012 Zijnge et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Funding: This work was supported by TUA grants from the County Council of Västerbotten, Sweden (JO), and by a grant from the Swedish Research Council (822-2007-3473; TK). The instrumentation and bioinformatics infrastructure of the proteomics platform of the Chemistry Biology Center (KBC) at Umeå University and SLU/Umeå was supported by grants of the Wallenberg and the Kempe Foundations. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The authors have declared that no competing interests exist.

Introduction

The destruction of the periodontium in periodontitis is the most common cause of tooth-loss worldwide. Periodontitis is a bacterially induced chronic inflammatory disease of the periodontium, which is also epidemiologically associated with systemic conditions such as cardiovascular diseases and rheumatoid arthritis. Periodontitis is associated with a defined subgingival microbial composition of the dental plaque biofilm, however the contribution of these bacteria to disease progression is poorly understood [1]–[4]. Hence, research in this area can improve the understanding of periodontitis and additional chronic diseases, and possibly lead to the identification of novel background mechanisms for increased cardiovascular risk. A commonly used model organism in periodontitis is the Gram-negative oral and systemic human pathogen Aggregatibacter actinomycetemcomitans, which is associated with aggressive periodontitis and endocarditis [5]–[7]. The virulence mechanisms executed by A. actinomycetemcomitans and their possible implication in periodontal and systemic disease are not clearly understood. The extracellular proteome, or secretome, of this organism is presumed to represent a key element. A. actinomycetemcomitans releases several factors that may play a role in modulating the host response, including leukotoxin (LtxA) [8], cytolethal distending toxin (CDT) [9], GroEL [10], peptidoglycan-associated lipoprotein (PAL) [11], and lipopolysaccharide (LPS) [12]. However, as a whole the secretome of this species is largely unexplored. Moreover, there is lacking knowledge regarding the set of secretion machineries that is used to deliver proteins to the exterior by A. actinomycetemcomitans. We and others have demonstrated that A. actinomycetemcomitans can deliver some of its virulence factors into human cells via outer membrane vesicles (OMVs) [13], [14]. Moreover, the species can excrete free-soluble surface components with proinflammatory activity independently of OMVs [15]. As both OMVs and free-soluble surface material are abundantly produced locally in the plaque biofilm, their potential entry into the circulation may constitute a significant source of inflammatory stimulants along with the planktonic bacteria in the bloodstream [16], [17]. To obtain a comprehensive overview of the potentially virulence-related extracellular proteins of A. actinomycetemcomitans, we have used liquid chromatography-tandem mass spectrometry (LC-MS/MS) to characterize the secretome of strain D7S (serotype a) during biofilm growth. Together with an in silico approach to identify the protein secretion machineries and candidate substrates for extracellular release in the strain D7S genome, this study provides a comprehensive overview of the active secretion pathways and a molecular basis for the pathogenic potential of A. actinomycetemcomitans.

Results and Discussion

Identification of Virulence-related Proteins in the A. actinomycetemcomitans Strain D7S Secretome

The rough-colony serotype a A. actinomycetemcomitans strain D7S was selected for this study as virulence properties of this strain and its derivatives have been frequently assessed in functional studies [14], [15], [18], [19], [20]–[22], and as its genome was recently characterized [23]. To identify the proteins extracellularly secreted by strain D7S, it was cultivated in biofilm form (Fig. 1A), and harvested at approximately early stationary phase (data not shown). Secretome preparations (∼20 µg protein; protein concentration ∼0.75 µg/µl) from two independent experiments were then analyzed by LC-MS/MS (Fig. 1B, and Fig. S1). Database searches (NCBInr) resulted in the identification of 179 proteins, out of which 106 were present in both of the secretome preparations (Fig 1C, and Table S1). This result is similar to an earlier study revealing approximately 120 proteins in the extracellular proteome of A. actinomycetemcomitans strain NCTC9710 (serotype c), albeit only five of the secreted proteins were identified [24]. The programs SosuiGramN, Cello 2.5 and PsortB were used to predict the subcellular localizations of the 179 proteins. According to our findings, 92 proteins were predicted to be cytoplasmic (51.4%), 40 periplasmic (22.3%), 19 located in the outer membrane (10.6%), and three to be extracellular (1.7%), whereas the subcellular locations of 25 (14%) proteins could not be predicted (Fig. S2, and Table S1). The relatively high abundance of proteins found by LC-MS/MS predicted to be either periplasmic or located in the outer membrane is consistent with the release of OMVs and free-soluble surface material by A. actinomycetemcomitans [15], [25]. OMVs are a common source of periplasmic and outer membrane proteins [26], and several are associated with, and secreted via A. actinomycetemcomitans OMVs [25]. The OMV’s (diameter 50–200 nm according to electron microscopy [15], [18]) are not expected to be filtered out (pore size 0.22 µm) during the preparation of extracellular protein extracts from strain D7S. Of the proteins identified by LC-MS/MS, 102 (57%) did not contain a signal sequence (Table S1). This is in the same range as similar studies with Campylobacter concisus and Listeria spp., respectively, which revealed that 57% [27] or 49% [28] of the extracellular proteins lacked a signal sequence. Identification of cytoplasmic proteins in the extracellular fraction may be a result of cell lysis or cell autolysis occurring during growth [29]. However, among the proteins identified by LC-MS/MS (Table S1) we did not find cyclic AMP receptor protein (CRP; GI:293390977). This cytoplasmic protein was used as a lysis marker in our earlier studies assessing the release of proteins by strain D7S grown as biofilm [15], [18]. Notably, CRP was not detected in extracellular supernatants of D7S cultivated as biofilm for up to 3 days, but could be released upon deliberate lysis of the A. actinomycetemcomitans cells [15], [18]. Hence, the absence of CRP in our present secretome preparations would argue against release due to bacterial lysis. An alternative explanation why proteins without signal seqeuence appeared in the secretome preparations may be non-classical secretion, e.g. translocation via hitherto uncharacterized routes for protein transport [30]. There is substantial evidence of cytoplasmic proteins being incorporated into OMVs of different bacterial species [31], [32]. Moreover, it has been suggested that certain cytoplasmic proteins may in fact have dual functions and can be targeted by the cell to different subcellular sites [33].

thumbnail
Download:
Figure 1. Analysis of the secretome of A. actinomycetemcomitans strain D7S during biofilm growth.

(A) Atomic force micrograph of strain D7S grown as biofilm. Bundles of Flp pili are indicated by arrows. Bar = 700 nm. (B) Coomassie blue-stained SDS-PAGE of the two independent secretome preparations, denoted 1 and 2 (protein concentration ∼0.75 µg/µl; ∼20 µg protein applied on the gel), obtained from strain D7S biofilms, and which were analyzed in this study. The sizes (kDa) of the proteins in the prestained molecular mass marker (M) are indicated along the left side. (C) Venn diagram obtained from the comparison of the LC-MS/MS-identified proteins of the two strain D7S secretome preparations (1 and 2).

https://doi.org/10.1371/journal.pone.0041662.g001

To assess the virulence potential of the strain D7S secretome, all proteins identified by LC-MS/MS (n = 179) were manually searched for their earlier reported associations with virulence-related activity in A. actinomycetemcomitans or, when applicable, in other organisms. From this screening, 26 proteins were of particular interest (Table 1). In accordance with previous studies, the secretome included several (n = 17) proteins demonstrated earlier to contribute to the pathogenicity of A. actinomycetemcomitans: CdtA (GI:293392175), CdtB (GI:293392176), and CdtC (GI:293392177) constitute a tripartite (CdtABC) genotoxin, produced by several Gram-negative organisms, and which can induce G2 cell cycle arrest, progressive cellular distention, and/or apoptosis in many cell types [34], [35]. The CdtB protein acts as a type I DNase [36], [37], whereas CdtA and CdtC are involved in host cell recognition and internalization of CdtB in host cells [38], [39]. The chaperonin GroEL (GI:293391167) can activate a plethora of mammalian cells, including macrophages, keratinocytes and periodontal ligament epithelial cells [10], [15], [40]. LtxA (GI:293390491) is a well-studied virulence determinant in A. actinomycetemcomitans that contributes to pathogenesis by killing lymphoid and myeloid cells [41]–[44]. Macrophage infectivity protein (MIP; GI:293391100) is a surface-exposed lipoprotein that is involved in intracellular survival and persistence of several species, including A. actinomycetemcomitans, Legionella pneumophila and Neisseria spp. [45]–[48], and which meets several important criteria for a potential meningococcal vaccine antigen [49]. The outer membrane proteins (OMP) Omp18/16 (GI:293391272), Omp29 (GI:293390286; also referred to as OmpA and Omp34), Omp39 (GI:293391956), Omp100 (GI:293390333; also referred to as ApiA), and YaeT (GI:293390810) exhibit antigenicity in the host [50]–[53]. Notably, Omp39 and YaeT are antigenic in patients carrying the highly leukotoxic A. actinomycetemcomitans JP2 clone [52], and Omp29 is associated with the entry of A. actinomycetemcomitans into gingival epithelial cells [50]. Omp64 has been suggested to play a role in iron acquisition [51]. It cannot be excluded that the Omp29-(OmpA)-like protein (GI:293390327; ∼76% amino acid identity to Omp29) could also exhibit antigenicity in the host similar to A. actinomycetemcomitans Omp18/16, Omp29, Omp39, and Omp100. PAL (GI:293390507) is a lipoprotein that has proinflammatory activity in vitro [11], [15], [18]. RcpA (GI:293391332), TadD (GI:293391326), TadE (no assigned GI in the D7S genome database used; see also Type II secretion below), and TadG (GI:293391324) are part of the A. actinomycetemcomitans tad (tight adherence) gene locus (flp-1, flp-2, tadV, rcpCAB, tadZABCDEFG) [54] encoding a macromolecular machinery for Flp pili biogenesis [55], [56], which is required for virulence in a rat model for periodontal disease [57]. Whereas studies on Porphyromonas gingivalis fimbriae have indicated their modulation of signaling pathways mediating proinflammatory or proadhesive effects [58], such function of A. actinomycetemcomitans fimbriae or fimbrial components has not yet been experimentally demonstrated. To corroborate our LC-MS/MS results, we used immunoblotting with specific antibodies to confirm the presence of selected of the above proteins in the extracellular medium of strain D7S biofilm samples used for the LC-MS/MS analysis (Fig. S3). Extracellular release of GroEL and PAL by strain D7S grown as a biofilm is consistent with our published data [15].

thumbnail
Download:
Table 1. Virulence-related proteins in the A. actinomycetemcomitans strain D7S secretome identified by LC-MS/MS.

https://doi.org/10.1371/journal.pone.0041662.t001

Further to the above proteins, the D7S secretome included eight additional proteins that could potentially contribute to the pathogenicity of A. actinomycetemcomitans (Table 1): DegQ (GI: 293390325) is a periplasmic protease, which has been suggested to play a protective role when E. coli enters a host organism [59]. In Actinobacillus pleuropneumoniae, LppC (GI:293390346) is an outer membrane lipoprotein that can induce a specific antibody response in pigs [60]. NlpB (GI:293391894; also referred to as DapX) is an outer membrane lipoprotein that was demonstrated to be essential for virulence of Yersinia pestis in murine models of bubonic and septicemic plague [61]. Pcp (GI:293392122) is a lipoprotein highly abundant on the surface of Y. pestis, and has been hypothesized to be important for pathogenicity [62]. PotD (GI:293390241) is a surface-associated spermidine- and putrescine-binding protein in Streptococcus pneumoniae and a virulence factor in murine models of systemic and pulmonary infection [63]. TolB (GI: 293390508) was identified in a genome-wide screen of Salmonella typhimurium as a virulence factor in mice contributing to resistance to deoxycholate and serum and bacterial survival in J774 cells [64]. TolC (GI: 293389951) is involved in multidrug resistance and plays a key role in virulence in several Gram-negative organisms [65]–[68]. Finally, a putative 28-kDa lipoprotein (GI:293391790) shows strong similarity (>90%) with the factor H-binding protein (fHbp) of Neisseria spp. and contains the conserved lipoprotein C domain (pfam08794). Factor H-binding protein attaches to the human regulatory protein factor H that regulates homeostasis of the complement system. By binding factor H, bacteria mimic host tissue and avoid complement initiated cell lysis [69]. The novel observation that A. actinomycetemcomitans possess an fHbp homologue may explain the finding that complement-mediated phagocytosis of A. actinomycetemcomitans is generally inefficient [70].

Identification of Virulence-related Proteins among the Proteins Predicted to be Secreted by A. actinomycetemcomitans Strain D7S

In the present study we assessed the secretome composition during biofilm growth using LC-MS/MS. It cannot be excluded that some proteins targeted for secretion may have been missed due to the experimental setup and analysis methods used. For instance, previous analysis of 2D-PAGE protein patterns during various growth conditions revealed differential protein levels for up to 8% of the secreted subproteome of A. actinomycetemcomitans strain NCTC9710 [24]. To identify such proteins and to gain further insight into the pathogenic potential of this organism, we therefore determined the theoretical secretome by in silico analysis of the strain D7S genome using a rational approach formalized in a bioinformatics workflow [71]. In this workflow, proteins with a signal peptide and proteins predicted to be extracellularly secreted using any of the software packages SosuiGramN, Cello 2.5 and PsortB were identified. From the identified proteins, those with at least two predicted alpha helical transmembrane domains were regarded as putatively attached to the inner membrane and discarded from the list of proteins. Hence, the final list contained 250 proteins having the potential to be extracellularly secreted. Out of these, 73 proteins (29.2%) were detected by LC-MS/MS. The 250 proteins were grouped into lipoproteins and outer membrane proteins forming beta barrels, respectively (Table S2A, and Table S2B). The remaining proteins were categorized according to their predicted subcellular localization (Table S2C). Analogously to the proteins identified by LC-MS/MS, the 250 proteins found by in silico analysis were manually screened for their earlier reported associations with virulence-related activity in A. actinomycetemcomitans or, when applicable, in other organisms. As summarized below, this screening revealed seven proteins not identified by LC-MS/MS, which were of particular interest (Table 2).

thumbnail
Download:
Table 2. Additional virulence-related proteins identified among the proteins predicted to be secreted by A. actinomycetemcomitans strain D7S.

https://doi.org/10.1371/journal.pone.0041662.t002

Lipoproteins.

Proteins containing a lipobox are translocated across the inner membrane in a Sec dependent way and can remain attached to the inner membrane or guided to the outer membrane by the LolCDE complex [72] where they can become exposed to the exterior or secreted. In silico analysis of the D7S genome revealed 71 proteins containing a lipobox, out of which 23 were present in the secretome as determined by LC-MS/MS (Table S2A). Among the proteins not found by LC-MS/MS, four could have a potential to contribute to A. actinomycetemcomitans virulence. DspB (Dispersin B; also referred to as Chb [GI:293391439]) is a glycoside dehydrogenase that degrades poly-N-acetylglucosamine, the main component of the A. actinomycetemcomitans biofilm matrix. By degrading the matrix, DspB is involved in dispersion of A. actinomycetemcomitans cells from the attached biofilm [73]. OapB (GI:293391876) belongs to a growing family of lysozyme inhibitors contributing to lysozyme tolerance in Gram-negative bacteria [74]. This function may be beneficial for survival in the oral cavity where lysozyme is an important component of the antibacterial activity of saliva. The outer membrane antigenic lipoprotein B (GI:293391443) exhibits ∼80% amino acid sequence similarity to the lipoprotein NlpD of Yersinia pestis, which is essential for the development of bubonic and pneumonic plague in mice [75]. Finally, VacJ (GI:293391047) is exposed on the surface of Shigella flexneri cells, and is essential for the spreading of this species through the intercellular spaces of tissues and, moreover, induces protrusions of eukaryotic cells upon expression in intracellular bacteria. This facilitates migration of the bacteria to the cytoplasm of the next cell [76].

Beta barrel forming proteins.

Using the BCCS and TMB-Hunt packages with the BCCS >3 and/or TMB-Hunt ≥4 criteria, 35 beta barrel forming proteins were found to be encoded by the D7S genome, including 11 that were identified by LC-MS/MS (Table S2B). Among the proteins not identified by LC-MS/MS, EmaA (GI293390626, GI293390627, and GI293390628) is a candidate virulence-related protein that could promote adhesion of A. actinomycetemcomitans [77].

Predicted subcellular localization of the remaining proteins based on a most-votes analysis.

The subcellular localization of the remaining 144 proteins identified by in silico analysis, containing 39 identified by LC-MS/MS, could not be predicted solely based on the presence of a signal peptide or secondary structure. Therefore a majority of votes analysis was used to categorize them (Table S2C). We are aware of that the subcellular localization prediction tools should be used with caution. This is exemplified by the 53 proteins for which no subcellular localization could be predicted, and for the ten cytoplasmically predicted proteins that nevertheless do contain a signal peptide. Among the proteins not detected by LC-MS/MS, Aae (GI:293391897) modulates binding of A. actinomycetemcomitans to human buccal epithelial cells [78], and OapA (GI:293391875) is a surface-exposed protein in Haemophilus influenzae that mediates epithelial adhesion via an unknown receptor [79].

Delineation of Active Protein Secretion Pathways in A. actinomycetemcomitans Strain D7S

After having assessed the virulence potential of the A. actinomycetemcomitans strain D7S secretome by two complementary approaches, we next wanted to disclose the active pathways for protein secretion of this strain. Blast searches for secretion systems as defined in the Transporter Classification Database (TCDB) revealed the presence of homologues to Sec, Tat, Type I, Type II, and Type V secretion systems in the D7S genome. Moreover, we identified secretins, the Outer Membrane Factor, and Outer Membrane Protein Insertion Porins (TCDB classification numbers indicated in corresponding sections below). On the other hand, Blast searches against typical Type III (#3.A.6), Type IV (#3.A.7), Type VI (#9.A.34) or Type VII (#1.C.95, #9.A.44, #9.A.25) proteins revealed no homologues to these secretion systems in the D7S genome. The functionality of the identified protein secretion systems encoded by strain D7S was then primarily assessed by matching them with their substrates among the 179 secretome proteins identified by LC-MS/MS. Our findings regarding the respective secretion systems are summarized below, in Fig. 2, and in Table S3. As indicated in Table S3, in some cases a protein was found to have several gene identifiers in the D7S genome database used.

thumbnail
Download:
Figure 2. Schematic representation of functionally active protein secretion systems (Ss) present in A. actinomycetemcomitans strain D7S.

(A) Secretin and OMF-related secretion systems. (B) OmpIP and Type V-related secretion systems. Secretion systems are matched with a selection of their substrates among the virulence-related proteins identified by LC-MS/MS analysis and/or by in silico analysis of the strain D7S genome.

https://doi.org/10.1371/journal.pone.0041662.g002

General secretory pathway (# 3.A.5).

Genetic determinants encoding general secretory pathway (Sec) components, a major pathway for translocation of proteins across the inner membrane, was found in the strain D7S genome (Table S3A). Proteins designated for secretion by the Sec pathway contain a specific N-terminal signal sequence that directs them to the appropriate machinery [80]. Among the secretome proteins, 74 (41.3%) were found to carry such N-terminal signal sequence (Table S1), indicating that they are targeted for secretion via the general secretory pathway. This is consistent with a functionally active general secretory pathway in A. actinomycetemcomitans strain D7S.

Twin arginine pathway (# 2.A.64).

Homologues to TatABC were identified in the A. actinomycetemcomitans D7S genome (Table S3A). The Tat translocase is composed of two or three inner membrane located proteins, TatA, TatB and TatC. The Tat secretion machinery is dedicated to the translocation of folded proteins across the membrane, which is a clear distinction from the Sec translocation pathway that translocates unfolded peptides [80]. Substrates targeted for the Tat pathway have an N-terminal signal sequence that resembles the NHC signal sequence organization also found in Sec-directed substrates. The distinguishing feature of Tat-directing signal sequences is the consensus motif (ST)-R-R-X-F-L-K in the hydrophobic domain. Among the extracellularly secreted proteins identified by LC-MS/MS, three appeared to contain a Tat signal sequence (Table S1). This suggests that the Tat pathway was functionally active during the present experimental setup.

Type I (ATP binding cassette, # 3.A.1).

Our database searches revealed 53 proteins encoded by A. actinomycetemcomitans strain D7S having an ABC transporter domain (PS50893). Of these proteins, 16 are involved in secretion whereas the rest function as substrate uptake systems (Table S3B). In silico analysis revealed five systems for biosynthesis purposes, e.g. the transport of LPS (Wzm/Wzt and LptABCDEFG), the periplasmic cytochrome biosynthesis (CydCD and CcmABCDE) and lipoprotein (LolABCDE) transport to the outer membrane. Other proteins are part of a Type I secretion system that is dedicated to secrete a substrate to the exterior. These systems are LtxCABD, MacAB, a putative exporter involved in toluene tolerance, MdlB (GI:293390739), and exporters with unknown function (GI:293389953, 293390276, 293391310). LtxA is the secreted substrate of the LtxCABD system. Our LC-MS/MS data revealed the extracellular release of LtxA and several additional ABC transporters during the growth conditions applied (Table S1), which is consistent with a functionally active Type I secretion in strain D7S.

Type II (Main terminal branch; # 3.A.15).

Our in silico analysis of the A. actinomycetemcomitans strain D7S genome also revealed the presence of several Type II secretion-related genes (Table S3C). Type II secretion systems are protein complexes that mediate the translocation of Sec- or Tat-secreted proteins across the outer membrane, and share homology with Type IV pilus production (T4P) and competence systems [81], [82], [83]. In A. actinomycetemcomitans, a Type IV pilus (T4P)-like gene cluster named PilABCD has been identified that is involved in natural competence but not in pilus formation [21]. However, that study did not reveal the identity of the outer membrane protein that forms the pore for DNA uptake. An upstream gene cluster that is also involved in competence, ComABCD [84], encodes the HofQ secretin. Hence, it cannot be excluded that HofQ may function as a secretin needed for DNA uptake via PilABCD. The close resemblance of the proteins involved in T4P, T2SS and competence has led others to hypothesize a system in which DNA binding and uptake is mediated by a pilus (or pseudo-pilus) structure that combines T4P and competence-specific pili [85]. Whether PilABCD is part of a T4P or a T2SS and thereby involved in secretion is unknown but it might be a potential candidate secretion system to deliver Sec translocated proteins across the outer-membrane.

Type II (Tight adherence [Tad]; # 3.A.7.15.1).

The tad gene cluster in A. actinomycetemcomitans (flp-1, flp-2, tadV, rcpCAB, tadZABCDEFG) encodes long filamentous fimbrils composed of bundled Flp pili [54], [86], and has been classified as a novel lineage of Type II secretion [56]. The function of this macromolecular transport system in A. actinomycetemcomitans and other species is not entirely understood. It has been suggested that the pseudopilins TadE and TadF form an oligomeric structure in the periplasm, which could guide pilus assembly and provide contact with the outer membrane RcpA secretin complex [56]. Expression of the Tad-pilus is important for the characteristic rough-colony morphology of A. actinomycetemcomitans strains including D7S [20], [87]. Strain D7S encodes a complete tad gene locus, and mutational analysis has revealed that fimbriae expression requires the function of flp-1, rcpA, rcpB, tadB, tadD, tadE, and tadF [22]. The concerted finding of several Tad locus proteins (TadD, TadE, TadG and RcpA) in the strain D7S secretome (Table S1), and the production of bundle forming pili by D7S [20](Fig. 1A) is consistent with functionally active Type II secretion. However, notably our BLAST searches of the strain D7S genome database failed to detect either flp-2 or tadE (Table S3C) despite that these genes were earlier characterized in this strain [22]. The reason for this discrepancy is not known.

Type IV (# 3.A.7).

Eleven A. actinomycetemcomitans Type IV secretion system (T4SS) determinants were earlier found to be encoded on a plasmid (pVT745) in one strain (VT745), whereas they were present on the chromosome of a different strain (VT747) [88], [89]. Homologues to these T4SS determinants are encoded on the conjugative plasmid S57 (92% nucleotide identities to pVT745) [90]. One additional gene (GI: 1243302) is located within the gene cluster and may serve as an entry exclusion protein that is thought to inhibit DNA transfer after mating pairs have been formed [88], [91]. Also the novel plasmid S25 harbours a set of ten T4SS related proteins [90]. One gene (GI: 8537963) is probably a VirB3/B4 fused protein. However neither of the plasmids pVT745, S25 or S57 is carried by strain D7S [23]. Moreover, our in silico analysis revealed no T4SS related genes or putative T4SS protein substrates such as CagA, CagE, and VirB1 [92], [93] to be encoded on the D7S chromosome. This is consistent with our LC-MS/MS data.

Type V (# 1.B.12, # 1.B.40, # 1.B.54).

A Prosite search with PS51208 for the autotransporter beta domain revealed only Aae as a possible autotransporter in the A. actinomycetemcomitans D7S genome (Table S3C). Literature searches showed that EmaA [77] and Omp100 [78] are also considered to be autotransporters in A. actinomycetemcomitans. This assumption was based on the presence of a C-terminal YadA region and the predicted formation of beta barrels. EmaA and Omp100 have thereby been assigned as autotransporters of the AT-2 family, but lack the typical autotransporter beta domain. Due to the absence of an autoproteolytic domain, Aae, Omp100 and EmaA are most likely exposed on the cell surface. Indeed, Aae and EmaA have been shown to form surface structures on A. actinomycetemcomitans cells and are involved in adhesion to buccal epithelial cells and collagen [77], [94]. Identifying Omp100 among the extracellular proteins (Table S1) is consistent with functionally active Type V secretion in strain D7S.

Secretins (# 1.B.22).

In our in silico analysis of the A. actinomycetemcomitans strain D7S genome, we identified HofQ and RcpA as proteins that carry one and two of the secretin consensus sequences, respectively [95] (Table S3C). RcpA is essential for Tad-pilus formation and forms the pore through which the pilus extends through the outer membrane [56], [96]. Production of bundle forming pili by strain D7S [20](Fig. 1A) is consistent with the RcpA secretin being functionally active. HofQ shares homology with the competence protein ComE of Haemophilus influenzae and the Pseudomonas aeruginosa PilQ protein involved in type II secretion/type IV pilus biogenesis. HofQ may function in natural transformation, fibronectin binding and Type IV pilus biogenesis [97], although the exact mechanism is unclear.

Outer membrane factor (OMF) (# 1.B.17).

The TolC family of outer membrane proteins is ubiquitous among Gram-negative organisms and their role in protein secretion has been thoroughly characterized [98]. The archetypical TolC protein acts as an outer membrane factor by forming a pore, and is recruited by ABC transporters to export substrates to the exterior [99]. Hitherto, TdeA (GI: 293392240), also referred to as TolC [100], has been the only Outer Membrane Factor recognized in A. actinomycetemcomitans, and is a component of a drug efflux pump that plays a role in LtxA secretion [101]. Interestingly, our BLAST searches with OMF proteins against the strain D7S genome revealed the presence of one additional, tentative OMF, denoted TolC (GI: 293389951) (Table S3C), exhibiting ∼24% amino acid identity to TdeA. Albeit the putative role of this novel OMF in protein secretion remains to be experimentally confirmed, the recognition of TdeA, TolC, and several ABC transporters in the D7S secretome (Table S1) would be consistent with both OMF proteins playing an active role.

Outer membrane protein insertion porins (# 1.B.33).

The Outer Membrane Protein Insertional Porin (OmpIP) or Beta barrel Assembly Machinery (BAM) is responsible for the folding and insertion of outer membrane proteins in the outer membrane [102], and its functional activity in D7S is consistent with the recognition of several OMPs in the secretome (Table S1). In the strain D7S genome we found homologues to three BAM chaperones, and to BamA, BamC, BamD and BamE (Table S3C). However, similar to genomic analysis of Neisseria spp. [103], [104], we did not find a BamB homologue in the D7S genome, or in any other A. actinomycetemcomitans genomes available at the databases in NCBI. Studies in E. coli suggest that absence of BamB may hamper the correct insertion of outer membrane proteins and compromise the outer membrane permeability [105]. Hence, analogously to the hypothesis regarding Neisseria spp. [104], BamC may compensate for lack of BamB in A. actinomycetemcomitans. This remains to be tested.

Concluding Remarks

In the present work we have assessed the virulence-potential of the extracellular proteome of the A. actinomycetemcomitans serotype a strain D7S using two complementary approaches, LC-MS/MS analysis of the secretome during biofilm growth, and in silico analysis of the D7S genome. Although the virulence potential of A. actinomycetemcomitans may vary among strains, these approaches together underscored that this organism releases a much larger arsenal of virulence-related proteins than previously demonstrated. Moreover, combining our LC-MS/MS and in silico data revealed active use of at least Type I, II, and V secretion to translocate proteins directly or via two-step pathways into the extracellular space. This includes the Sec/Tat systems for transport across the inner membrane, and outer membrane factors, secretins and auto-transporters for delivery across the outer membrane. The possible presence of non-classical protein secretion in A. actinomycetemcomitans, the formation of outer membrane vesicles and the unknown sorting pathway of lipoproteins to and across the outer membrane emphasizes the importance of combining in silico data with experimental evidence. We conclude that our present results provide a molecular basis for further disclosing the role of A. actinomycetemcomitans in periodontal and systemic disease. For instance, studies investigating the expression, secretion and function of novel putative extracellular virulence factors such as DegQ, fHbp, LppC, MIP, NlpB, Pcp, and PotD may provide new evidence how bacterial modulation of host cytokine expression repertoires may lead to destructive inflammation.

Materials and Methods

Bacterial Strains and Growth Conditions

The A. actinomycetemcomitans serotype a rough-colony strain D7S, and its pal mutant derivative, D7S-p [18] were used in this study. Strain D7S was originally isolated from a patient with aggressive periodontal disease [20]. The strain D7S was kindly donated by Dr. Casey Chen, University of Southern California. This strain has since been used in several of our earlier studies that were cited in our present work [14], [15], [18]. The strains were cultured on blood agar plates (5% defibrinated horse blood, 5 mg hemin/l, 10 mg Vitamin K/l, Columbia agar base) incubated in air supplemented with 5% CO2, at 37°C for 3 days as previously described [18]. For biofilm growth, 2×108 bacterial cells were inoculated in 2 ml tryptic soy broth (Difco) in 24-well cell culture plates (Nunc), which were incubated in static culture in air supplemented with 5% CO2, at 37°C for 42 h. To assess the growth of A. actinomycetemcomitans biofilms, biofilm samples were stained with crystal violet and the amount of bound dye, which is proportional to the biofilm mass was quantitated by measuring its absorbance at 590 nm.

Preparation of the A. actinomycetemcomitans Strain D7S Secretome

Following biofilm cultivation, 2 ml of the growth medium of a single well was carefully collected and then centrifuged for 10 min at 10.000×g to pellet down remaining bacterial cells. Supernatants were then filtered through 0.45 µm and subsequently 0.22 µm membranes prior to being desalted and concentrated into 120 µl H2O with Pall 10 K membrane filters according to the manufacturer’s instructions (Pall Corporation). Protein concentrations were determined using the Bradford Reagent (Sigma-Aldrich).

SDS-PAGE and Western Immunoblotting

The quality of secretome preparations (protein concentration ∼0.75 µg/µl) was confirmed by running a fraction (∼5 µg protein) of the sample on an 8–16% linear gradient SDS-PAGE gel (Criterion, Bio-Rad). As control samples in immunoblots, A. actinomycetemcomitans whole cell preparations (∼1 µg/µl protein) equivalent to 10 µg protein were loaded on the gels. The procedures used for SDS-PAGE and immunoblot analysis have been described previously [11], [106]. Gels were stained using non-ammonical Silver-staining (BioRad). For immunoblot detection, we used polyclonal antisera raised in rabbits specific for E. coli GroEL (Sigma-Aldrich) and A. actinomycetemcomitans PAL [107]. The antisera were used at a final concentration of 1∶8000 and 1∶10.000, respectively. As secondary antibody, anti-rabbit horseradish peroxidase (HRP)-conjugate was used (1∶10.000). Immunoreactive bands were visualized using SuperSignal® (Pierce) and the ChemiDoc™ XRS + System (Bio-Rad).

Atomic Force Microscopy

For atomic force microscopy, bacterial cells were suspended in ultrapure water (Millipore) and 10 µl bacterial suspension was then placed on a freshly cleaved mica surface. The samples were incubated for 5 min at room temperature and blotted dry before being placed into a desiccator for at least 2 h. Imaging was performed using a Nanoscope V Atomic Force Microscope (Bruker AXS) using Tapping Mode with standard silicon cantilevers. Final images were plane fitted in both x and y axes and are presented in amplitude mode.

LC-MS/MS Analysis and Data Processing

For mass-spectrometry, protein samples equivalent to approximately 20 µg protein (protein concentration ∼0.75 µg/µl) were separated in 12 cm long 12% SDS-PAGE gels [106] containing 2 M urea. Subsequent to electrophoresis the gels were fixated using 10% acetic acid, 30% ethanol and stained using hot Coomassie blue [108]. In-gel digestion of peptides for analysis by mass spectrometry was carried out essentially as described earlier [109]. LC-MS/MS analysis of peptides was performed using an HCT-Ultra ETD II ion trap mass spectrometer (Bruker) linked to an Easy-nLC system (Proxeon). Spectra were acquired using the enhanced scanning mode covering a mass range from m/z 400 to m/z 1300. The LC separation of peptides was performed using a 5 µm C18 column (375 µm OD/75 µm ID×10 cm) (NanoSeparations) and a 60 min gradient ranging from 1 to 50 percent of acetonitrile. The flow rate was 300 nl min−1. The LC-MS/MS datasets were processed using Bruker DataAnalysis 4.0 SP4. Database searches using the peaklist files of the processed mass spectra were performed in the bacterial section of the NCBInr database using ProteinScape 2.1 (Bruker) and in-house licenses of Mascot 2.3.01 (www.matrixscience.com) and of Phenyx 2.6 (www.genbio.com). The search parameters allowed for one missed cleavage site and a mass error of 0.3 Da for both the MS and MS/MS mode. In addition, variable modifications including methionine oxidation, N-terminal acetylation, and derivation of cysteine by propionamide were considered. The Mascot.dat files of the database searches were submitted to the EBI-Pride repository (accession numbers: 22453–22456). Non-redundant protein lists were compiled from the database searches using the ProteinExtractor of ProteinScape 2.1 and settings for spectra acceptance as follows: Mascot score >100 and at least one peptide with a peptide ion score >55. Peptides with a Mascot ion score <30 were ignored. As for Phenyx scores, the minimum threshold for protein acceptance was 18 and at least one peptide with a score of 10 as required. Peptides with a Phenyx score <7 were not considered.

The non-redundant protein lists created by the ProteinExtractor were further inspected manually to ensure that each protein identification was based on at least two different peptide identifications. It is an intrinsic feature of the database searches than the identified proteins not only included hits from the sequenced genomes of A. actinomycetemcomitans strain D7S but also from strain D11S and from the partly sequenced genomes of additional A. actinomycetemcomitans strains present in the NCBI nr database. As for the protein identifications obtained from other strains than D7S, BLAST searches were performed to identify the corresponding D7S homologues.

In Silico Analysis

The whole genome shotgun sequence of A. actinomycetemcomitans strain D7S with accession number ADCF00000000 [23] was downloaded from the European Bioinformatics Institute (http://www.ebi.ac.uk/), and used in this work. During the analyses a number of annotations appeared inconsistent because of abnormal length of the amino acid sequence compared with homologues in other species, multiple assignments for the same protein for sequences directly next to each other, and annotations that rather than complete protein sequences only represented a signal sequence or the absence of such. Such annotations were therefore compared with their homologues in the A. actinomycetemcomitans serotype c strain D11S-1 genome [90]. When the present manuscript was under revision, the D7S whole genome shotgun project was superseded by the complete genome record (CP003496).

All annotated ORFs were analyzed for the presence of protein transport systems using the Transporter Classification Database [110] and combined with the results from TransportDB (http://www.membranetransport.org/) [111], the ABCdb database for ABC transporters in Archea and Bacteria (http://www-abcdb.biotoul.fr/) and Prosite searches with (PS50893, PS51012, PS50928, PS50990, PS50929, PS00211) for ABC transporter motifs.

The presence of signal sequences was determined using SignalP (http://www.cbs.dtu.dk/services/SignalP/) [112], Phobius (http://phobius.sbc.su.se/) [113], Predisi (http://www.predisi.de/) [114] and PSortb (http://www.psort.org/psortb/) [115] and results were interpreted using a most-votes approach. LipoP (http://www.cbs.dtu.dk/services/LipoP/) [116] was used to identify proteins with a lipobox. TatP (http://www.cbs.dtu.dk/services/TatP/) [112] and TatFind (http://signalfind.org/tatfind.html) [117] were used to identify proteins secreted via the Tat pathway. In addition the PattinProt software (http://npsa-pbil.ibcp.fr/cgi-bin/npsa_automat.pl?page  = /NPSA/npsa_pattinprot.html) was used to search for Tat and other motifs. The probable secondary structure of the proteins was analyzed by the search for beta-barrels with BOMP (http://services.cbu.uib.no/tools/bomp) [118] and with TMB-Hunt (http://bmbpcu36.leeds.ac.uk/~andy/betaBarrel/AACompPred/aaTMB_Hunt.cgi) [119]. The presence of transmembrane alfa-helixes was inspected with the TMHMM (http://www.cbs.dtu.dk/services/TMHMM/) [120], ThumbUP (http://sparks.informatics.iupui.edu/Softwares-Services_files/thumbup.htm) [121] and Phobius. Tools to predict subcellular localization were SosuiGramN (http://bp.nuap.nagoya-u.ac.jp/sosui/) [122], Cello 2.5 (http://cello.life.nctu.edu.tw/) [123], PsortB, and a most-votes analysis. If SosuiGramN, Cello 2.5, and PsortB all predicted different subcellular localizations for a particular protein, the protein was grouped into “three different predicted subcellular localizations”.

Supporting Information

Figure S1.

Silver-stained SDS-PAGE of the secretome preparations (protein concentration ∼0.75 µg/µl) of strain D7S grown as biofilm. Samples (∼20 µg protein) of secretome preparation 1 (A), and preparation 2 (B), were applied on the gel. The indicated gel bands were excised from the gel and processed for LC-MS/MS analysis. The approximate locations of the protein bands (10 to 150 kDa) of the prestained molecular weight marker are indicated.

https://doi.org/10.1371/journal.pone.0041662.s001

(TIF)

Figure S2.

Distribution of the identified A. actinomycetemcomitans strain D7S secretome proteins according to their predicted subcellular localization. One hundred and seventy-nine different proteins were identified by LC-MS/MS.

https://doi.org/10.1371/journal.pone.0041662.s002

(TIF)

Figure S3.

Immunoblot detection of GroEL and PAL released by A. actinomycetemcomitans strain D7S grown as biofilm. Lane 1: a representative, filtered and concentrated supernatant sample (protein concentration ∼0.75 µg/µl; ∼5 µg protein applied on the gel) was applied in lane 1. The following whole cell preparation samples (protein concentration ∼1 µg/µl; 10 µg loaded each) were used as controls: lane 2. D7S, lane 3. D7S-p (PAL-deficient derivative of D7S). Polyclonal antisera specific for E. coli GroEL, and A. actinomycetemcomitans PAL were used for immunoblot detection. Sizes (kDa) of proteins in the prestained molecular weight marker (M) are indicated.

https://doi.org/10.1371/journal.pone.0041662.s003

(TIF)

Table S1.

Proteins identified by LC-MS/MS in two independent secretome preparations of A. actinomycetemcomitans strain D7S grown as biofilm.

https://doi.org/10.1371/journal.pone.0041662.s004

(XLS)

Table S2.

Identification of virulence-related proteins among the proteins predicted to be secreted by A. actinomycetemcomitans strain D7S using in silico analysis. (A) Proteins containing a lipobox. (B) Proteins predicted to form a beta-barrel. (C) Proteins for which the subcellular localization was predicted based on a most votes analysis.

https://doi.org/10.1371/journal.pone.0041662.s005

(XLS)

Table S3.

Protein secretion machinery components encoded in the genome of A. actinomycetemcomitans strain D7S. (A) Sec and Tat secretion machinery components. (B) Secretory ABC transporter system components. (C) Secretion systems tentatively involved in the transport of proteins across the outer membrane.

https://doi.org/10.1371/journal.pone.0041662.s006

(XLS)

Acknowledgments

Elisabeth Granström is thanked for excellent technical assistance. The polyclonal GroEL antiserum was kindly provided by Dr. Sun Nyunt Wai.

Author Contributions

Conceived and designed the experiments: VZ TK JO. Performed the experiments: VZ TK. Analyzed the data: VZ TK JO. Contributed reagents/materials/analysis tools: TK JO. Wrote the paper: VZ JO.

References

  1. 1. Beck JD, Slade G, Offenbacher S (2000) Oral disease, cardiovascular disease and systemic inflammation. Periodontol 2000 23: 110–120.
  2. 2. Darveau RP (2010) Periodontitis: a polymicrobial disruption of host homeostasis. Nat Rev Microbiol 8: 481–490.
  3. 3. de Pablo P, Chapple IL, Buckley CD, Dietrich T (2009) Periodontitis in systemic rheumatic diseases. Nat Rev Rheumatol 5: 218–224.
  4. 4. Hugoson A, Norderyd O (2008) Has the prevalence of periodontitis changed during the last 30 years? J Clin Periodontol 35: 338–345.
  5. 5. Slots J, Genco RJ (1984) Black-pigmented Bacteroides species, Capnocytophaga species, and Actinobacillus actinomycetemcomitans in human periodontal disease: virulence factors in colonization, survival, and tissue destruction. J Dent Res 63: 412–421.
  6. 6. Socransky SS, Haffajee AD (2005) Periodontal microbial ecology. Periodontol 2000 38: 135–187.
  7. 7. van Winkelhoff AJ, Slots J (1999) Actinobacillus actinomycetemcomitans and Porphyromonas gingivalis in nonoral infections. Periodontol 2000 20: 122–135.
  8. 8. Tsai CC, McArthur WP, Baehni PC, Hammond BF, Taichman NS (1979) Extraction and partial characterization of a leukotoxin from a plaque-derived Gram-negative microorganism. Infect Immun 25: 427–439.
  9. 9. Sugai M, Kawamoto T, Peres SY, Ueno Y, Komatsuzawa H, et al. (1998) The cell cycle-specific growth-inhibitory factor produced by Actinobacillus actinomycetemcomitans is a cytolethal distending toxin. Infect Immun 66: 5008–5019.
  10. 10. Goulhen F, Hafezi A, Uitto VJ, Hinode D, Nakamura R, et al. (1998) Subcellular localization and cytotoxic activity of the GroEL-like protein isolated from Actinobacillus actinomycetemcomitans. Infect Immun 66: 5307–5313.
  11. 11. Paul-Satyaseela M, Karched M, Bian Z, Ihalin R, Boren T, et al. (2006) Immunoproteomics of Actinobacillus actinomycetemcomitans outer-membrane proteins reveal a highly immunoreactive peptidoglycan-associated lipoprotein. J Med Microbiol 55: 931–942.
  12. 12. Kiley P, Holt SC (1980) Characterization of the lipopolysaccharide from Actinobacillus actinomycetemcomitans Y4 and N27. Infect Immun 30: 862–873.
  13. 13. Demuth DR, James D, Kowashi Y, Kato S (2003) Interaction of Actinobacillus actinomycetemcomitans outer membrane vesicles with HL60 cells does not require leukotoxin. Cell Microbiol 5: 111–121.
  14. 14. Rompikuntal PK, Thay B, Khan MK, Alanko J, Penttinen AM, et al. (2012) Perinuclear localization of internalized Outer Membrane Vesicles carrying active Cytolethal Distending Toxin (CDT) from Aggregatibacter actinomycetemcomitans. Infect Immun 80: 31–42.
  15. 15. Oscarsson J, Karched M, Thay B, Chen C, Asikainen S (2008) Proinflammatory effect in whole blood by free soluble bacterial components released from planktonic and biofilm cells. BMC Microbiol 8: 206.
  16. 16. Asikainen SE (2009) Periodontal bacteria and cardiovascular problems. Future Microbiol 4: 495–498.
  17. 17. Kebschull M, Demmer RT, Papapanou PN (2010) “Gum bug, leave my heart alone!”–epidemiologic and mechanistic evidence linking periodontal infections and atherosclerosis. J Dent Res 89: 879–902.
  18. 18. Karched M, Ihalin R, Eneslätt K, Zhong D, Oscarsson J, et al. (2008) Vesicle-independent extracellular release of a proinflammatory outer membrane lipoprotein in free-soluble form. BMC Microbiol 8: 18.
  19. 19. Nalbant A, Chen C, Wang Y, Zadeh HH (2003) Induction of T-cell apoptosis by Actinobacillus actinomycetemcomitans mutants with deletion of ltxA and cdtABC genes: possible activity of GroEL-like molecule. Oral Microbiol Immunol 18: 339–349.
  20. 20. Wang Y, Goodman SD, Redfield RJ, Chen C (2002) Natural transformation and DNA uptake signal sequences in Actinobacillus actinomycetemcomitans. J Bacteriol 184: 3442–3449.
  21. 21. Wang Y, Shi W, Chen W, Chen C (2003) Type IV pilus gene homologs pilABCD are required for natural transformation in Actinobacillus actinomycetemcomitans. Gene 312: 249–255.
  22. 22. Wang Y, Chen C (2005) Mutation analysis of the flp operon in Actinobacillus actinomycetemcomitans. Gene 351: 61–71.
  23. 23. Chen C, Kittichotirat W, Chen W, Downey JS, Si Y, et al. (2010) Genome sequence of naturally competent Aggregatibacter actinomycetemcomitans serotype a strain D7S-1. J Bacteriol 192: 2643–2644.
  24. 24. Fletcher JM, Nair SP, Ward JM, Henderson B, Wilson M (2001) Analysis of the effect of changing environmental conditions on the expression patterns of exported surface-associated proteins of the oral pathogen Actinobacillus actinomycetemcomitans. Microb Pathog 30: 359–368.
  25. 25. Kato S, Kowashi Y, Demuth DR (2002) Outer membrane-like vesicles secreted by Actinobacillus actinomycetemcomitans are enriched in leukotoxin. Microb Pathog 32: 1–13.
  26. 26. Ellis TN, Kuehn MJ (2010) Virulence and immunomodulatory roles of bacterial outer membrane vesicles. Microbiol Mol Biol Rev 74: 81–94.
  27. 27. Kaakoush NO, Man SM, Lamb S, Raftery MJ, Wilkins MR, et al. (2010) The secretome of Campylobacter concisus. Febs J 277: 1606–1617.
  28. 28. Trost M, Wehmhoner D, Karst U, Dieterich G, Wehland J, et al. (2005) Comparative proteome analysis of secretory proteins from pathogenic and nonpathogenic Listeria species. Proteomics 5: 1544–1557.
  29. 29. Tullius MV, Harth G, Horwitz MA (2001) High extracellular levels of Mycobacterium tuberculosis glutamine synthetase and superoxide dismutase in actively growing cultures are due to high expression and extracellular stability rather than to a protein-specific export mechanism. Infect Immun 69: 6348–6363.
  30. 30. Bendtsen JD, Kiemer L, Fausboll A, Brunak S (2005) Non-classical protein secretion in bacteria. BMC Microbiol 5: 58.
  31. 31. Ferrari G, Garaguso I, Adu-Bobie J, Doro F, Taddei AR, et al. (2006) Outer membrane vesicles from group B Neisseria meningitidis delta gna33 mutant: proteomic and immunological comparison with detergent-derived outer membrane vesicles. Proteomics 6: 1856–1866.
  32. 32. Galka F, Wai SN, Kusch H, Engelmann S, Hecker M, et al. (2008) Proteomic characterization of the whole secretome of Legionella pneumophila and functional analysis of outer membrane vesicles. Infect Immun 76: 1825–1836.
  33. 33. Song C, Kumar A, Saleh M (2009) Bioinformatic comparison of bacterial secretomes. Genomics Proteomics Bioinformatics 7: 37–46.
  34. 34. Guerra L, Cortes-Bratti X, Guidi R, Frisan T (2011) The biology of the cytolethal distending toxins. Toxins 3: 172–190.
  35. 35. Smith JL, Bayles DO (2006) The contribution of cytolethal distending toxin to bacterial pathogenesis. Crit Rev Microbiol 32: 227–248.
  36. 36. Elwell CA, Dreyfus LA (2000) DNase I homologous residues in CdtB are critical for cytolethal distending toxin-mediated cell cycle arrest. Mol Microbiol 37: 952–963.
  37. 37. Lara-Tejero M, Galan JE (2000) A bacterial toxin that controls cell cycle progression as a deoxyribonuclease I-like protein. Science 290: 354–357.
  38. 38. Akifusa S, Heywood W, Nair SP, Stenbeck G, Henderson B (2005) Mechanism of internalization of the cytolethal distending toxin of Actinobacillus actinomycetemcomitans. Microbiology 151: 1395–1402.
  39. 39. Boesze-Battaglia K, Brown A, Walker L, Besack D, Zekavat A, et al. (2009) Cytolethal distending toxin-induced cell cycle arrest of lymphocytes is dependent upon recognition and binding to cholesterol. J Biol Chem 284: 10650–10658.
  40. 40. Paju S, Goulhen F, Asikainen S, Grenier D, Mayrand D, et al. (2000) Localization of heat shock proteins in clinical Actinobacillus actinomycetemcomitans strains and their effects on epithelial cell proliferation. FEMS Microbiol Lett 182: 231–235.
  41. 41. Henderson B, Ward JM, Ready D (2010) Aggregatibacter (Actinobacillus) actinomycetemcomitans: a triple A* periodontopathogen? Periodontol 2000 54: 78–105.
  42. 42. Kachlany SC (2010) Aggregatibacter actinomycetemcomitans leukotoxin: from threat to therapy. J Dent Res 89: 561–570.
  43. 43. Lally ET, Golub EE, Kieba IR, Taichman NS, Rosenbloom J, et al. (1989) Analysis of the Actinobacillus actinomycetemcomitans leukotoxin gene. Delineation of unique features and comparison to homologous toxins. J Biol Chem 264: 15451–15456.
  44. 44. Simpson DL, Berthold P, Taichman NS (1988) Killing of human myelomonocytic leukemia and lymphocytic cell lines by Actinobacillus actinomycetemcomitans leukotoxin. Infect Immun 56: 1162–1166.
  45. 45. Cianciotto NP, Fields BS (1992) Legionella pneumophila mip gene potentiates intracellular infection of protozoa and human macrophages. Proc Natl Acad Sci U S A 89: 5188–5191.
  46. 46. Echenique-Rivera H, Muzzi A, Del Tordello E, Seib KL, Francois P, et al. (2011) Transcriptome analysis of Neisseria meningitidis in human whole blood and mutagenesis studies identify virulence factors involved in blood survival. PLoS Pathog 7: e1002027.
  47. 47. Leuzzi R, Serino L, Scarselli M, Savino S, Fontana MR, et al. (2005) Ng-MIP, a surface-exposed lipoprotein of Neisseria gonorrhoeae, has a peptidyl-prolyl cis/trans isomerase (PPIase) activity and is involved in persistence in macrophages. Mol Microbiol 58: 669–681.
  48. 48. Maeda T, Maeda H, Yamabe K, Mineshiba J, Tanimoto I, et al. (2009) Highly expressed genes in a rough-colony-forming phenotype of Aggregatibacter actinomycetemcomitans: implication of a mip-like gene for the invasion of host tissue. FEMS Immunol Med Microbiol 58: 226–236.
  49. 49. Hung MC, Salim O, Williams JN, Heckels JE, Christodoulides M (2011) The Neisseria meningitidis macrophage infectivity potentiator protein induces cross-strain serum bactericidal activity and is a potential serogroup B vaccine candidate. Infect Immun 79: 3784–3791.
  50. 50. Kajiya M, Komatsuzawa H, Papantonakis A, Seki M, Makihira S, et al. (2011) Aggregatibacter actinomycetemcomitans Omp29 is associated with bacterial entry to gingival epithelial cells by F-actin rearrangement. PLoS One 6: e18287.
  51. 51. Komatsuzawa H, Asakawa R, Kawai T, Ochiai K, Fujiwara T, et al. (2002) Identification of six major outer membrane proteins from Actinobacillus actinomycetemcomitans. Gene 288: 195–201.
  52. 52. Rylev M, Abduljabar AB, Reinholdt J, Ennibi OK, Haubek D, et al. (2011) Proteomic and immunoproteomic analysis of Aggregatibacter actinomycetemcomitans JP2 clone strain HK1651. J Proteomics 74: 2972–2985.
  53. 53. Wilson ME, Hamilton RG (1995) Immunoglobulin G subclass response of juvenile periodontitis subjects to principal outer membrane proteins of Actinobacillus actinomycetemcomitans. Infect Immun 63: 1062–1069.
  54. 54. Kachlany SC, Planet PJ, DeSalle R, Fine DH, Figurski DH (2001) Genes for tight adherence of Actinobacillus actinomycetemcomitans: from plaque to plague to pond scum. Trends Microbiol 9: 429–437.
  55. 55. Perez BA, Planet PJ, Kachlany SC, Tomich M, Fine DH, et al. (2006) Genetic analysis of the requirement for flp-2, tadV, and rcpB in Actinobacillus actinomycetemcomitans biofilm formation. J Bacteriol 188: 6361–6375.
  56. 56. Tomich M, Planet PJ, Figurski DH (2007) The tad locus: postcards from the widespread colonization island. Nat Rev Microbiol 5: 363–375.
  57. 57. Schreiner HC, Sinatra K, Kaplan JB, Furgang D, Kachlany SC, et al. (2003) Tight-adherence genes of Actinobacillus actinomycetemcomitans are required for virulence in a rat model. Proc Natl Acad Sci U S A 100: 7295–7300.
  58. 58. Hajishengallis G, Wang M, Liang S (2009) Induction of distinct TLR2-mediated proinflammatory and proadhesive signaling pathways in response to Porphyromonas gingivalis fimbriae. J Immunol 182: 6690–6696.
  59. 59. Kucz N, Meltzer M, Ehrmann M (2007) Periplasmic proteases and protease inhibitors. In: Ehrmann M, editor. Washington, DC: ASM Press.
    • 60. Oldfield NJ, Donovan EA, Worrall KE, Wooldridge KG, Langford PR, et al. (2008) Identification and characterization of novel antigenic vaccine candidates of Actinobacillus pleuropneumoniae. Vaccine 26: 1942–1954.
    • 61. Bland DM, Eisele NA, Keleher LL, Anderson PE, Anderson DM (2011) Novel genetic tools for diaminopimelic acid selection in virulence studies of Yersinia pestis. PLoS One 6: e17352.
    • 62. Myers-Morales T, Cowan C, Gray ME, Wulff CR, Parker CE, et al. (2007) A surface-focused biotinylation procedure identifies the Yersinia pestis catalase KatY as a membrane-associated but non-surface-located protein. Appl Environ Microbiol 73: 5750–5759.
    • 63. Ware D, Jiang Y, Lin W, Swiatlo E (2006) Involvement of potD in Streptococcus pneumoniae polyamine transport and pathogenesis. Infect Immun 74: 352–361.
    • 64. Bowe F, Lipps CJ, Tsolis RM, Groisman E, Heffron F, et al. (1998) At least four percent of the Salmonella typhimurium genome is required for fatal infection of mice. Infect Immun 66: 3372–3377.
    • 65. Buckley AM, Webber MA, Cooles S, Randall LP, La Ragione RM, et al. (2006) The AcrAB-TolC efflux system of Salmonella enterica serovar Typhimurium plays a role in pathogenesis. Cell Microbiol 8: 847–856.
    • 66. Ferhat M, Atlan D, Vianney A, Lazzaroni JC, Doublet P, et al. (2009) The TolC protein of Legionella pneumophila plays a major role in multi-drug resistance and the early steps of host invasion. PLoS One 4: e7732.
    • 67. Gil H, Platz GJ, Forestal CA, Monfett M, Bakshi CS, et al. (2006) Deletion of TolC orthologs in Francisella tularensis identifies roles in multidrug resistance and virulence. Proc Natl Acad Sci U S A 103: 12897–12902.
    • 68. Posadas DM, Martin FA, Sabio y Garcia JV, Spera JM, Delpino MV, et al. (2007) The TolC homologue of Brucella suis is involved in resistance to antimicrobial compounds and virulence. Infect Immun 75: 379–389.
    • 69. Lambris JD, Ricklin D, Geisbrecht BV (2008) Complement evasion by human pathogens. Nat Rev Microbiol 6: 132–142.
    • 70. Permpanich P, Kowolik MJ, Galli DM (2006) Resistance of fluorescent-labelled Actinobacillus actinomycetemcomitans strains to phagocytosis and killing by human neutrophils. Cell Microbiol 8: 72–84.
    • 71. Desvaux M, Dumas E, Chafsey I, Chambon C, Hebraud M (2010) Comprehensive appraisal of the extracellular proteins from a monoderm bacterium: theoretical and empirical exoproteomes of Listeria monocytogenes EGD-e by secretomics. J Proteome Res 9: 5076–5092.
    • 72. Tokuda H (2009) Biogenesis of outer membranes in Gram-negative bacteria. Biosci Biotechnol Biochem 73: 465–473.
    • 73. Kaplan JB (2010) Biofilm dispersal: mechanisms, clinical implications, and potential therapeutic uses. J Dent Res 89: 205–218.
    • 74. Callewaert L, Aertsen A, Deckers D, Vanoirbeek KG, Vanderkelen L, et al. (2008) A new family of lysozyme inhibitors contributing to lysozyme tolerance in gram-negative bacteria. PLoS Pathog 4: e1000019.
    • 75. Tidhar A, Flashner Y, Cohen S, Levi Y, Zauberman A, et al. (2009) The NlpD lipoprotein is a novel Yersinia pestis virulence factor essential for the development of plague. PLoS One 4: e7023.
    • 76. Suzuki T, Murai T, Fukuda I, Tobe T, Yoshikawa M, et al. (1994) Identification and characterization of a chromosomal virulence gene, vacJ, required for intercellular spreading of Shigella flexneri. Mol Microbiol 11: 31–41.
    • 77. Tang G, Ruiz T, Barrantes-Reynolds R, Mintz KP (2007) Molecular heterogeneity of EmaA, an oligomeric autotransporter adhesin of Aggregatibacter (Actinobacillus) actinomycetemcomitans. Microbiology 153: 2447–2457.
    • 78. Yue G, Kaplan JB, Furgang D, Mansfield KG, Fine DH (2007) A second Aggregatibacter actinomycetemcomitans autotransporter adhesin exhibits specificity for buccal epithelial cells in humans and Old World primates. Infect Immun 75: 4440–4448.
    • 79. Prasadarao NV, Lysenko E, Wass CA, Kim KS, Weiser JN (1999) Opacity-associated protein A contributes to the binding of Haemophilus influenzae to chang epithelial cells. Infect Immun 67: 4153–4160.
    • 80. Natale P, Bruser T, Driessen AJ (2008) Sec- and Tat-mediated protein secretion across the bacterial cytoplasmic membrane–distinct translocases and mechanisms. Biochim Biophys Acta 1778: 1735–1756.
    • 81. Hazes B, Frost L (2008) Towards a systems biology approach to study type II/IV secretion systems. Biochim Biophys Acta 1778: 1839–1850.
    • 82. Kostakioti M, Newman CL, Thanassi DG, Stathopoulos C (2005) Mechanisms of protein export across the bacterial outer membrane. J Bacteriol 187: 4306–4314.
    • 83. Sandkvist M (2001) Biology of type II secretion. Mol Microbiol 40: 271–283.
    • 84. Chen I, Dubnau D (2004) DNA uptake during bacterial transformation. Nat Rev Microbiol 2: 241–249.
    • 85. Chen I, Dubnau D (2003) DNA transport during transformation. Front Biosci 8: s544–556.
    • 86. Inouye T, Ohta H, Kokeguchi S, Fukui K, Kato K (1990) Colonial variation and fimbriation of Actinobacillus actinomycetemcomitans. FEMS Microbiol Lett 57: 13–17.
    • 87. Kachlany SC, Planet PJ, Bhattacharjee MK, Kollia E, DeSalle R, et al. (2000) Nonspecific adherence by Actinobacillus actinomycetemcomitans requires genes widespread in bacteria and archaea. J Bacteriol 182: 6169–6176.
    • 88. Galli DM, Chen J (2006) Entry exclusion activity on conjugative plasmid pVT745. Plasmid 55: 158–163.
    • 89. Novak KF, Dougherty B, Pelaez M (2001) Actinobacillus actinomycetemcomitans harbours type IV secretion system genes on a plasmid and in the chromosome. Microbiology 147: 3027–3035.
    • 90. Chen C, Kittichotirat W, Si Y, Bumgarner R (2009) Genome sequence of Aggregatibacter actinomycetemcomitans serotype c strain D11S-1. J Bacteriol 191: 7378–7379.
    • 91. Marrero J, Waldor MK (2007) Determinants of entry exclusion within Eex and TraG are cytoplasmic. J Bacteriol 189: 6469–6473.
    • 92. Alvarez-Martinez CE, Christie PJ (2009) Biological diversity of prokaryotic type IV secretion systems. Microbiol Mol Biol Rev 73: 775–808.
    • 93. Teng YT, Hu W (2003) Expression cloning of a periodontitis-associated apoptotic effector, cagE homologue, in Actinobacillus actinomycetemcomitans. Biochem Biophys Res Commun 303: 1086–1094.
    • 94. Rose JE, Meyer DH, Fives-Taylor PM (2003) Aae, an autotransporter involved in adhesion of Actinobacillus actinomycetemcomitans to epithelial cells. Infect Immun 71: 2384–2393.
    • 95. Yen MR, Peabody CR, Partovi SM, Zhai Y, Tseng YH, et al. (2002) Protein-translocating outer membrane porins of Gram-negative bacteria. Biochim Biophys Acta 1562: 6–31.
    • 96. Planet PJ, Kachlany SC, Fine DH, DeSalle R, Figurski DH (2003) The Widespread Colonization Island of Actinobacillus actinomycetemcomitans. Nat Genet 34: 193–198.
    • 97. Mullen LM, Bosse JT, Nair SP, Ward JM, Rycroft AN, et al. (2008) Pasteurellaceae ComE1 proteins combine the properties of fibronectin adhesins and DNA binding competence proteins. PLoS One 3: e3991.
    • 98. Wandersman C, Delepelaire P (1990) TolC, an Escherichia coli outer membrane protein required for hemolysin secretion. Proc Natl Acad Sci U S A 87: 4776–4780.
    • 99. Davidson AL, Dassa E, Orelle C, Chen J (2008) Structure, function, and evolution of bacterial ATP-binding cassette systems. Microbiol Mol Biol Rev 72: 317–364, table of contents.
      • 100. Tang G, Kawai T, Komatsuzawa H, Mintz KP (2012) Lipopolysaccharides mediate leukotoxin secretion in Aggregatibacter actinomycetemcomitans. Mol Oral Microbiol 27: 70–82.
      • 101. Crosby JA, Kachlany SC (2007) TdeA, a TolC-like protein required for toxin and drug export in Aggregatibacter (Actinobacillus) actinomycetemcomitans. Gene 388: 83–92.
      • 102. Knowles TJ, Scott-Tucker A, Overduin M, Henderson IR (2009) Membrane protein architects: the role of the BAM complex in outer membrane protein assembly. Nat Rev Microbiol 7: 206–214.
      • 103. Bos MP, Robert V, Tommassen J (2007) Functioning of outer membrane protein assembly factor Omp85 requires a single POTRA domain. EMBO Rep 8: 1149–1154.
      • 104. Volokhina EB, Beckers F, Tommassen J, Bos MP (2009) The beta-barrel outer membrane protein assembly complex of Neisseria meningitidis. J Bacteriol 191: 7074–7085.
      • 105. Charlson ES, Werner JN, Misra R (2006) Differential effects of yfgL mutation on Escherichia coli outer membrane proteins and lipopolysaccharide. J Bacteriol 188: 7186–7194.
      • 106. Laemmli UK (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227: 680–685.
      • 107. Ihalin R, Karched M, Eneslätt K, Asikainen S (2006) Characterization of immunoaffinity purified peptidoglycan-associated lipoprotein of Actinobacillus actinomycetemcomitans. J Chromatogr B Analyt Technol Biomed Life Sci 831: 116–125.
      • 108. Westermeier R, Naven T (2002) Proteomics in practice – a laboratory manual of proteome analysis. Weinheim: Wiley-VCH.
        • 109. Yao D, Kieselbach T, Komenda J, Promnares K, Prieto MA, et al. (2007) Localization of the small CAB-like proteins in photosystem II. J Biol Chem 282: 267–276.
        • 110. Saier MH Jr, Yen MR, Noto K, Tamang DG, Elkan C (2009) The Transporter Classification Database: recent advances. Nucleic Acids Res 37: D274–278.
        • 111. Ren Q, Kang KH, Paulsen IT (2004) TransportDB: a relational database of cellular membrane transport systems. Nucleic Acids Res 32: D284–288.
        • 112. Bendtsen JD, Nielsen H, von Heijne G, Brunak S (2004) Improved prediction of signal peptides: SignalP 3.0. J Mol Biol 340: 783–795.
        • 113. Kall L, Krogh A, Sonnhammer EL (2007) Advantages of combined transmembrane topology and signal peptide prediction–the Phobius web server. Nucleic Acids Res 35: W429–432.
        • 114. Hiller K, Grote A, Scheer M, Munch R, Jahn D (2004) PrediSi: prediction of signal peptides and their cleavage positions. Nucleic Acids Res 32: W375–379.
        • 115. Yu NY, Wagner JR, Laird MR, Melli G, Rey S, et al. (2010) PSORTb 3.0: improved protein subcellular localization prediction with refined localization subcategories and predictive capabilities for all prokaryotes. Bioinformatics 26: 1608–1615.
        • 116. Juncker AS, Willenbrock H, Von Heijne G, Brunak S, Nielsen H, et al. (2003) Prediction of lipoprotein signal peptides in Gram-negative bacteria. Protein Sci 12: 1652–1662.
        • 117. Rose RW, Bruser T, Kissinger JC, Pohlschroder M (2002) Adaptation of protein secretion to extremely high-salt conditions by extensive use of the twin-arginine translocation pathway. Mol Microbiol 45: 943–950.
        • 118. Berven FS, Flikka K, Jensen HB, Eidhammer I (2004) BOMP: a program to predict integral beta-barrel outer membrane proteins encoded within genomes of Gram-negative bacteria. Nucleic Acids Res 32: W394–399.
        • 119. Garrow AG, Agnew A, Westhead DR (2005) TMB-Hunt: a web server to screen sequence sets for transmembrane beta-barrel proteins. Nucleic Acids Res 33: W188–192.
        • 120. Krogh A, Larsson B, von Heijne G, Sonnhammer EL (2001) Predicting transmembrane protein topology with a hidden Markov model: application to complete genomes. J Mol Biol 305: 567–580.
        • 121. Zhou H, Zhou Y (2003) Predicting the topology of transmembrane helical proteins using mean burial propensity and a hidden-Markov-model-based method. Protein Sci 12: 1547–1555.
        • 122. Imai K, Asakawa N, Tsuji T, Akazawa F, Ino A, et al. (2008) SOSUI-GramN: high performance prediction for sub-cellular localization of proteins in Gram-negative bacteria. Bioinformation 2: 417–421.
        • 123. Yu CS, Lin CJ, Hwang JK (2004) Predicting subcellular localization of proteins for Gram-negative bacteria by support vector machines based on n-peptide compositions. Protein Sci 13: 1402–1406.