Báo cáo khoa học: The secretome of Campylobacter concisus pot

The secretome of Campylobacter concisus Nadeem O. Kaakoush 1 , Si Ming Man 1 , Sarah Lamb 1 , Mark J. Raftery 2 , Marc R. Wilkins 1 , Zsuzsanna Kovach 1 and Hazel Mitchell 1 1 School of Biotechnology and Biomolecular Sciences, University of New South Wales, Sydney, Australia 2 Biological Mass Spectrometry Facility, University of New South Wales, Sydney, Australia Introduction Since the discovery of Campylobacter jejuni, which is recognized as the leading cause of bacterial gastroenteri- tis in both the developing and developed world [1], a considerable body of research has focused on this patho- genic bacterial species. The virulence factors integral to the pathogenesis of C. jejuni include motility, adhesion, expression of toxins, and invasion into host cells [2], making it well suited to the conditions of the gastroin- testinal tract. Following colonization of the intestinal mucosa, C. jejuni adheres to epithelial cells via surface- associated adhesins [2]. The bacterium then employs its flagellar apparatus as a secretion organelle, through which it secretes invasion antigens that promote cellular invasion [3]. Evidence also supports a paracellular invasion mechanism by which C. jejuni disrupts tight junctions of epithelial cells [4,5]. Additionally, C. jejuni can induce apoptotic cell death through the expression and secretion of a cytolethal distending toxin within the cells [6]. Although research into the pathogenesis of C. jejuni has expanded over the past decades, little is known about other members of the Campylobacter genus. Over recent years, evidence has emerged suggesting that a number of non-jejuni Campylobacter species may also be potential pathogens of the human intesti- nal tract. For example, Van Etterijck et al. [7], Vandamme et al. [8] and Johnson and Finegold [9] have reported the isolation of Campylobacter concisus from fecal samples of patients with gastrointestinal disorders. As a result of these and other studies, Keywords Campylobacter concisus; Crohn’s disease; secretome; virulence; zonula occludens Correspondence H. Mitchell, School of Biotechnology and Biomolecular Sciences, The University of New South Wales, Sydney, NSW 2052, Australia Fax: +61293851483 Tel: +61293852040 E-mail: H.Mitchell@unsw.edu.au (Received 3 November 2009, revised 18 January 2010, accepted 20 January 2010) doi:10.1111/j.1742-4658.2010.07587.x A higher prevalence of Campylobacter concisus and higher levels of IgG antibodies specific to C. concisus in Crohn’s disease patients than in con- trols were recently detected. In this study, 1D and 2D gel electrophoresis coupled with LTQ FT-MS and QStar tandem MS, respectively, were per- formed to characterize the secretome of a C. concisus strain isolated from a Crohn’s disease patient. Two hundred and one secreted proteins were iden- tified, of which 86 were bioinformatically predicted to be secreted. Searches were performed on the genome of C. concisus strain 13826, and 25 genes that have been associated with virulence or colonization in other organisms were identified. The zonula occludens toxin was found only in C. concisus among the Campylobacterales, although expanded searches revealed that this protein was present in two e-proteobacterial species from extreme mar- ine environments. Alignments and structural threading indicated that this toxin shared features with that of other virulent pathogens, including Neisseria meningitidis and Vibrio cholerae. Further comparative analyses identified several associations between the secretome of C. consisus and putative virulence factors of this bacterium. This study has identified several factors putatively associated with disease outcome, suggesting that C. concisus is a pathogen of the gastrointestinal tract. Abbreviations OMP, outer membrane protein; TCA, trichloroacetic acid; Zot, zonula occludens toxin. 1606 FEBS Journal 277 (2010) 1606–1617 ª 2010 The Authors Journal compilation ª 2010 FEBS C. concisus has recently been suggested to be a putative agent of diarrheal diseases [10–12]. A recent study in our laboratory resulted in the culture of several non-jejuni Campylobacter species from biopsy samples obtained from children newly diagnosed with Crohn’s disease [13]. Using a species-specific PCR, C. concisus was shown to be present in 51% of children with Crohn’s disease, a significantly higher frequency than in controls (2%) [13]. Investigation of the IgG antibody response in sera from children shown to be PCR-positive showed a significantly higher level of C. concisus antibodies to be present in patients with Crohn’s disease than in controls [13], suggesting that children infected with C. concisus mounted an IgG response to this bacterium. To date, there is limited information regarding the molecular basis of the pathogenesis of C. concisus. A study by Engberg et al. [14] has shown that C. concisus is able to produce a toxin similar to cytolethal distending toxin, and that cell lysates from C. concisus are able to induce cytopathic effects in a monkey kidney epithelial cell line. Two further studies have shown that C. concisus isolates possess cell-bound and secreted hemolytic activities [15,16]. In relation to animal models, a study conducted in 2008 showed that some strains of C. concisus have the ability to colonize the mouse intestinal tract, and that C. concisus can be cultured from the liver, ileum and jejunum of infected mice [17]. In this study, proteomics coupled with MS and comparative bioinformatic searches were performed to identify the secreted proteins and putative virulence factors of C. concisus. Results and Discussion The secretome of C. concisus The secreted proteins of a pathogen can be divided into three major groups on the basis of their function within the cell: those involved in cell survival; those involved in protection against stresses; and those asso- ciated with virulence and ⁄ or colonization of the host. The secretome is an essential component of a patho- gen’s arsenal, and can ultimately reflect its invasive- ness. As such, the characterization of the secretome of C. concisus can assist in unraveling the pathogenic potential of this bacterium (391 proteins were pre- dicted to be secreted by C. concisus strain 13826 using the signalp 3.0 server). In addition, given that the flagellar secretion system has a significant role in the virulence of C. jejuni and that similarities are present between the flagellar systems of C. jejuni and C. conci- sus, detection and identification of the secreted pro- teins of C. concisus was undertaken. The secreted proteins of C. concisus strain UNSWCD, which was isolated from a cecal biopsy sample from a child with Crohn’s disease [13], were purified from liquid cultures and separated using 1D-PAGE and 2D-PAGE (pI 4–7 and 7–10) (Fig. 1). Bands or spots were excised and digested, and proteins were identified using the appropriate MS protocol. The use of two independent methods, each with its own advantages, ensured the identification of the highest number of proteins within the purified sample. 1D-PAGE coupled with LTQ FT-MS allows for the 209 124 80 49.1 34.8 28.9 20.6 7.1 kDa AB C 4 7 7 10 Fig. 1. One-dimensional (A) and two-dimen- sional (B, C) PAGE on C. concisus UNSWCD secreted proteins. The fragment of the gel in (A) bordered by a dashed line was sectioned into 25 gel slices and processed for MS analyses. Proteins were also run on 2D-PAGE gels at pI 4–7 (B) and pI 7–10 (C), and all spots on both gels were extracted for MS analyses. N. O. Kaakoush et al. Campylobacter concisus secretome FEBS Journal 277 (2010) 1606–1617 ª 2010 The Authors Journal compilation ª 2010 FEBS 1607 identification of a high percentage of proteins from a complex fraction. In contrast, 2D-PAGE coupled with QStar tandem MS allows for the identification of lower-abundance proteins that could be overlooked in the process of analyzing complex fractions. The purifi- cation of secreted proteins from large volumes of med- ium may result in relatively higher quantities of contaminants (e.g. salts) within the protein fraction. Together with the presence of extracellular proteases that will degrade proteins, this may explain the streak- ing observed in the 2D-PAGE gels. The combined results from both methods consisted of 201 identified proteins within the purified fraction (PRIDE accession number: 11363) (Table S1; Figs S1 and S2). As cellular lysis and degradation during the growth of bacterial cultures and high-abundance proteins may result in nonsecretory contaminants within the identi- fied proteins, the 201 proteins were analyzed for the presence of a signal peptide, using the signalp 3.0 ser- ver. Signal peptides interact with signal peptidases to cleave proteins at specific sites, allowing proteins to be folded and exported via secretion. Of the 201 proteins, 69 were predicted to have a signal peptide, confirming that they are secreted proteins (Table S1; Table 1). Seventeen of the remaining 132 proteins were found, using the secretomep 2.0 server, to be nonclassically secreted (Table S1; Table 1). Functional classification of the 115 remaining proteins revealed that the major- ity of proteins, bioinformatically predicted to be non- secretory were involved in cellular survival (Table 2). Enriched functions included amino acid metabolism (n = 21), carbohydrate metabolism (n = 16), and elongation factors and chaperones (n = 15) (Table 2). It is possible that the high abundance of metabolic proteins, elongation factors and chaperones within cells may result in many of these proteins contaminat- ing the secretory fraction; however, owing to their high stringency, the bioinformatic predictio n processes employed may have also overlooked true-positives. The 86 confirmed secreted proteins of C. concisus UNSWCD were also grouped on the basis of their functions (Table 1). The proteins identified were either: (a) related to bacterial physiology, such as metabolic and solute-binding⁄ transport proteins; (b) involved in host-related functions, such as virulence factors; or (c) associated with protection against environmental stres- ses, such as oxidative stress response proteins. An additional 12 proteins previously annotated as putative or hypothetical were identified in the secretome of C. concisus. The first major group of proteins identified was involved in bacterial physiology and survival. For example, seven solute-binding proteins were identified. The presence of these proteins is to be expected, as many of them have functions linked to cellular metabolic processes. The proteins identified were two solute-binding family 1 proteins, two C4-dicarboxy- late-binding periplasmic proteins, extracellular tung- state-binding protein, glutamine-binding periplasmic protein, and d-methionine-binding lipoprotein MetQ. Each protein is specific for a different substrate, which it transports into or out of the cell across the periplas- mic membrane. For example, the C4-dicarboxylate- binding protein is a high-affinity transporter of C4-dicarboxylates such as fumarate. Bacteria such as C. concisus, which can respire anaerobically, are able to utilize fumarate as a terminal electron acceptor for this process. Similarly, MetQ binds d-methionine for bacterial utilization, and the tungstate-binding protein has been shown to protect cytochrome c oxidase from tungsten inhibition by binding free tungsten in the plasma membrane [18], thus allowing cytochrome c to function uninhibited. Cytochrome c is involved in an electron transfer system in which cytochrome c oxidase is the terminal electron acceptor, helping to establish a proton gradient, allowing the cell to synthesize ATP. It is interesting to note that both the cytochrome c assembly protein and cytochrome c oxidase were also identified as secreted proteins, implying that their function, as well as the function of tungstate-binding protein, is critical to cell survival. Another protein identified was the methyl-accepting chemotaxis protein, which is involved in signal trans- duction and chemotaxis. This protein is a member of a family of signal transducers in which sensory adapta- tion is mediated by the methylation of proteins. Stud- ies have shown that these proteins are involved in the general sensory control of both gliding and flagellar motility [19]. Given that C. concisus may rely on its flagella to both access and bind to the epithelial sur- face of the gastrointestinal tract, the secretion of proteins involved in the control of chemotaxis and flagellar motility highlights not only the importance of motility in cellular survival, but also the fact that secreted proteins may play a role in the pathogenesis of the bacterium. Examples of proteins that were found to be involved in the oxidative stress response are cop- per ⁄ zinc superoxide dismutase, superoxide dismutase (Fe), and protease Do. Superoxide dismutase is involved in the catalysis of superoxides, such as those produced by phagocytes during oxidative burst killing of pathogens [20]. Protease Do is a serine protease identical to the product of the high temper- ature requirement A gene (htrA), which has been described in Escherichia coli. HtrA protein has been Campylobacter concisus secretome N. O. Kaakoush et al. 1608 FEBS Journal 277 (2010) 1606–1617 ª 2010 The Authors Journal compilation ª 2010 FEBS Table 1. Functional classification of C. concisus UNSWCD secreted proteins bioinformatically predicted to be secreted (n = 86). Proteins from Table S1 that contained a signal peptide in their amino acid sequence or were found to be nonclassically secreted were chosen for fur- ther classification. Classification GI number ORF Protein name ABC transporter 157165101 CCC13826_1247 Nlpa lipoprotein 157165776 CCC13826_1793 Putative carbon storage regulator-like protein 157164930 CCC13826_2116 Hypothetical protein 157164768 CCC13826_2204 Periplasmic molybdate-binding protein Protein transport 157164610 CCC13826_0922 Translocation protein TolB 157165665 CCC13826_1073 Outer membrane lipoprotein carrier protein 157165672 CCC13826_1131 Curli production component CsgG protein Solute transport 158605017 CCC13826_0202 Extracellular tungstate-binding protein 157163977 CCC13826_0329 C4-dicarboxylate-binding periplasmic protein 157165691 CCC13826_0764 C4-dicarboxylate-binding periplasmic protein 158604975 CCC13826_1248 D-Methionine-binding lipoprotein MetQ 157165339 CCC13826_1924 Glutamine-binding periplasmic protein 157164859 CCC13826_1925 Solute-binding family 1 protein 157164888 CCC13826_1926 Solute-binding family 1 protein Response regulator 157165180 CCC13826_1398 Putative cytochrome c-type periplasmic protein 157165649 CCC13826_2064 DNA-binding response regulator 157165183 CCC13826_2088 Methyl-accepting chemotaxis sensory transducer Oxidoreduction 157164382 CCC13826_0059 Thioredoxin domain-containing protein 157165326 CCC13826_0161 Copper ⁄ zinc superoxide dismutase 157165242 CCC13826_0201 Molybdopterin oxidoreductase Fe 4 S 4 subunit 157165413 CCC13826_0328 Superoxide dismutase (Fe) 157163886 CCC13826_0758 Thiol peroxidase 157164029 CCC13826_1064 Protease Do 157164957 CCC13826_1400 Dyp-type peroxidase 157165527 CCC13826_1459 Hydrogenase-4 component I 157164172 CCC13826_1929 Flavodoxin FldA Cytochrome c maturation 157164310 CCC13826_1461 Cytochrome c assembly protein 157165298 CCC13826_1472 Cytochrome c oxidase Protein folding 157165622 CCC13826_0002 Disulfide isomerase DsbA 157164923 CCC13826_1633 Methionine sulfoxide reductase family protein 157164051 CCC13826_1670 TPR repeat-containing protein Protein synthesis 157165109 CCC13826_0029 L-Asparaginase 157165495 CCC13826_0170 50S ribosomal protein L11 157164773 CCC13826_0171 50S ribosomal protein L1 157164220 CCC13826_0172 50S ribosomal protein L10 157164672 CCC13826_0341 Ribosome recycling factor 157164041 CCC13826_0468 Glutamyl-tRNA synthetase 157164944 CCC13826_1737 Sigma 54 modulation protein 157165065 CCC13826_1871 tRNA synthetase, class II 157165615 CCC13826_1872 Adenylosuccinate synthetase 157165454 CCC13826_1997 Amino acid carrier protein AlsT 157165048 CCC13826_2114 Diaminopimelate epimerase Fatty acid synthesis 157164063 CCC13826_0560 3-Oxoacyl-(acyl carrier protein) synthase II 157164484 CCC13826_0562 3-Ketoacyl-(acyl carrier protein) reductase 157164819 CCC13826_1552 b-Ketoacyl-acyl carrier protein synthase II 157164251 CCC13826_2069 Holo-(acyl carrier protein) synthase Nitrogen metabolism 157164023 CCC13826_0863 Carbon–nitrogen family hydrolase 157165227 CCC13826_0865 Periplasmic nitrate reductase 157165732 CCC13826_0868 Nitrate reductase 157164517 CCC13826_1729 Methyl-accepting chemotaxis protein Sulfur metabolism 157164952 CCC13826_0231 Arylsulfotransferase 158604950 CCC13826_2236 Thiosulfate sulfurtransferase Sugar metabolism 157165684 CCC13826_0411 N-acetylglucosamine deacetylase 157164467 CCC13826_0872 Phosphomannomutase 157165286 CCC13826_1509 Phosphoenolpyruvate carboxykinase N. O. Kaakoush et al. Campylobacter concisus secretome FEBS Journal 277 (2010) 1606–1617 ª 2010 The Authors Journal compilation ª 2010 FEBS 1609 found to play a role in the intramacrophagic replica- tion of E. coli, with mutations in the htrA gene lead- ing to reduced bacterial virulence in mice [21]. The microaerophilic nature of C. concisus may explain the secretion of proteins involved in combating oxi- dative stress; however, the additional role of these proteins in neutralizing the oxidative bursts produced by polymorphonuclear cells in response to infection may reflect the survival strategies of C. concisus within its host. In the group pertaining to host-related functions, an outer membrane fibronectin-binding protein, known to be involved in adhesion to the host cell,was identified. Fibronectin is a large glycoprotein that is a component of the extracellular matrix of the human intestinal epithelium. Studies on C. jejuni have shown that the bacterium binds to fibronectin on the basolateral sur- face of human colonic cells [22]. The secretion of an extracellular binding protein that is specific to recep- tors in the intestinal epithelium is especially significant if C. concisus plays a pathogenic role in humans, as this protein assists in the adhesion to, and subsequent colonization of, the host cells [22]. Other virulence factors, CjaA and CjaC, were identified among the secreted proteins analyzed. CjaA and CjaC are poten- tially surface-exposed proteins that are homologs of ABC-transport proteins and known to be highly immunodominant in C. jejuni. Additionally, an S-layer-RTX protein was also found to be secreted by C. concisus UNSWCD. RTX proteins are pore-form- ing toxins synthesized by a diverse group of Gram-neg- ative pathogens. RTX-mediated cytotoxicity comprises two phases: a passive phase of adsorption onto the tar- get cell surface; and a membrane insertion phase [23]. The two forms of host cell death associated with this type of toxin include apoptosis and necrosis [23]. Finally, the flagellin-like protein FlaC, encoded by ccc13826_2187, was secreted by C. concisus UNSWCD. This protein is secreted by the flagellar system of C. jejuni, and mutants in the flaC gene showed a significantly reduced level of invasion into HEp-2 cells [24]. Table 1. (Continued.) Classification GI number ORF Protein name 157165707 CCC13826_2070 Fructose bisphosphate aldolase Tricarboxylic acid cycle 157164353 CCC13826_1283 Succinate dehydrogenase flavoprotein subunit Electron transfer 157164059 CCC13826_1119 Hypothetical protein 157165471 CCC13826_1395 Radical SAM domain-containing protein DNA ⁄ RNA 157164945 CCC13826_0021 DNA-binding protein HU 1 157165374 CCC13826_0649 DNA primase 157165135 CCC13826_0888 ComEC ⁄ Rec2 family protein Membrane and cell wall synthesis 157164040 CCC13826_0084 Glutamate racemase 2 157164709 CCC13826_0131 Peptidoglycan-associated lipoprotein 157164747 CCC13826_0576 ADP-glyceromanno-heptose-6-epimerase 157164370 CCC13826_0924 ADP-heptose-LPS heptosyltransferase II 157164830 CCC13826_1534 Lipopolysaccharide biosynthesis protein Motility 157164686 CCC13826_2187 Flagellin-like protein FlaC 157164988 CCC13826_2297 Flagellin B Virulence 157164816 CCC13826_0664 Surface antigen CjaA 157164740 CCC13826_0739 Outer membrane fibronectin-binding protein 157165273 CCC13826_0963 Antigen CjaC 157164622 CCC13826_1253 a-Macroglobulin family protein 157165243 CCC13826_1838 S-layer-RTX protein Unknown 157164911 CCC13826_0243 Putative periplasmic protein 157164523 CCC13826_0601 Putative outer membrane protein 157165071 CCC13826_0859 Hypothetical protein 158604901 CCC13826_1186 Conserved hypothetical protein 157163941 CCC13826_1330 Hypothetical protein 157164030 CCC13826_1407 Hypothetical protein 157165213 CCC13826_1428 Hypothetical protein 157164793 CCC13826_1524 Hypothetical protein 157165368 CCC13826_1643 FAD-binding domain-containing protein 157164111 CCC13826_1707 Hypothetical protein 157164095 CCC13826_1803 FAD-binding domain-containing protein 157163930 CCC13826_2060 Outer membrane protein Campylobacter concisus secretome N. O. Kaakoush et al. 1610 FEBS Journal 277 (2010) 1606–1617 ª 2010 The Authors Journal compilation ª 2010 FEBS Putative virulence factors of C. concisus Increasing reports suggesting a pathogenic role for C. concisus in the intestinal tract of humans highlights the importance of understanding the molecular mecha- nisms by which this bacterium may cause disease in its host. Therefore, blast searches of genes and proteins previously associated with virulence or colonization in other organisms were performed on the available gen- ome of C. concisus 13826, to determine whether any of these were present. This resulted in the identification of 25 potential candidates. These included known invasins, adhesins, hemolysins and iron-associated virulence fac- tors, such as invasin InvA, fibronectin-binding protein CadF, hemolysin TlyA, and siderophore esterase IroE (Table 3). Another virulence-associated factor identified was outer membrane protein (OMP) 18, encoded by ccc13826_0923 (Table 3). This protein is known to be an immunodominant antigen in both C. jejuni and Helicobacter pylori [25,26], and a study by Rathinavelu et al. [27] demonstrated that OMP18 was capable of inducing dendritic cell maturation and function, as well as initiating a Th-1-mediated immune response. Two proteins were found to be involved in twitching motility (Table 3), a form of surface translocation that enables the bacterium to crawl along surfaces [28]. This form of bacterial motility has been implicated in virulence and cytotoxicity in E. coli, Pseudomo- nas aeruginosa and Neisseria spp. [28]. Two genes encoding zonula occludens toxin (Zot) were identified in C. concisus 13826 (Table 3). More- over, two hypothetical proteins encoded by ccc13826_0191 and ccc13826_1210 were identified with Table 2. Functional classification of C. concisus UNSWCD identi- fied proteins bioinformatically predicted to be nonsecretory (n = 115). Functional classification Number of proteins Amino acid metabolism 21 Carbohydrate metabolism 16 Elongation factors and chaperones 15 Lipid metabolism 3 Metabolism of cofactors and vitamins 4 Nucleotide metabolism 6 Redox 8 Signal transduction and chemotaxis 8 Transcription 11 Translation 13 Transport and secretion 3 Unknown 7 Table 3. Putative virulence and colonization factors found in C. concisus 13826. GI ORF Protein name Present in other Campylobacterales 158604981 CCC13826_0115 Integral membrane protein MviN Yes 157164228 CCC13826_0191 Hypothetical protein Yes a 158605018 CCC13826_0222 Twitching motility protein Yes 112800340 CCC13826_0315 Symbiosis island integrase Yes 157101450 CCC13826_0608 Hemolysin activator-related protein HecB Yes 157164816 CCC13826_0664 Surface antigen, CjaA Yes 157164010 CCC13826_0706 Phage integrase family protein Yes 157164740 CCC13826_0739 Fibronectin-binding protein CadF Yes 157164577 CCC13826_0816 Acid membrane antigen A Yes 157164304 CCC13826_0923 Outer membrane protein 18 Yes 157165273 CCC13826_0963 Antigen CjaC Yes 157164212 CCC13826_1210 Hypothetical protein Yes a 158604972 CCC13826_1235 Iron-regulated colicin 1 receptor Yes 158604973 CCC13826_1236 Siderophore esterase IroE Yes 158604983 CCC13826_1443 Peptidase U32 (collagenase) Yes 157164343 CCC13826_1584 Twitching motility protein Yes 157165243 CCC13826_1838 S-layer-RTX protein Yes 157164705 CCC13826_2000 RNase R (VacB homolog) Yes 157164282 CCC13826_2017 Invasin InvA Yes 157164345 CCC13826_2024 Hemolysin TlyA Yes 157164617 CCC13826_2075 Zonula occludens toxin No 157165505 CCC13826_2148 Invasion protein CiaB Yes 157164974 CCC13826_2191 Phospholipase PldA Yes 157163885 CCC13826_2197 Hsp12 variant C Yes 157164938 CCC13826_2276 Zonula occludens toxin No a Found only in C. jejuni ssp. doylei 269.97. N. O. Kaakoush et al. Campylobacter concisus secretome FEBS Journal 277 (2010) 1606–1617 ª 2010 The Authors Journal compilation ª 2010 FEBS 1611 47% and 46% similarity, respectively, to C. concisus Zot. Zot is known to mimic a physiological modulator of intercellular tight junctions [29], and is used by viru- lent pathogens such as Vibrio cholerae and Neisse- ria meningitidis to increase tissue permeability [30]. In contrast to the activities of Clostridium difficile toxins A and B, the changes in tight junctions after exposure to Zot are reversible and are not associated with the destruction of the tight junction complex [31]. Further characterisation of Zot has indicated that its C-termi- nal domain causes delocalization of occludin and ZO-1 from Caco-2 cell–cell contacts [32]. Furthermore, expo- sure of Caco-2 cell monolayers to a peptide synthe- sized on the basis of the active domain of V. cholerae Zot caused the redistribution of ZO-1 away from cell junctions [33]. The peptide also caused a reversible reduction in transepithelial electrical resistance and an increase in lucifer yellow permeability [33]. Searches for Zot homologs within the host-related Campylobacterales order revealed this toxin to be only present in C. concisus, suggesting that it may have an important role in the pathogenesis of the bacterium. As a result of its specificity, the possibility that C. concisus acquired Zot from another pathogen through gene transfer was strong. However, further expanded searches against the genomes of all e-Proteobacteria found Zot homologs within the genomes of the sulfur- metabolizing bacteria Nautilia profundicola and Caminibacter mediatlanticus. These bacterial species have been isolated from extreme environments such as deep sea vents [34,35]. It is unknown why bacteria from these extreme environments would require a toxin that targets tight junctions, but one possibility could be that this toxin aids N. profundicola in penetrating the sheath lining covering the worms that it colonizes. Alignment of the Zot amino acid sequences of C. concisus, Ne. meningitidis and V. cholerae demon- strated that four highly conserved domains exist within these proteins that are likely to be important for toxin activity (Fig. 2). However, no domains within the sequences of C. concisus Zot and Ne. meningitidis Zot aligned with the previously identified active domain of V. cholerae Zot (FCIGRL), suggesting that these toxins may have different mechanisms of action. Ter- tiary structure prediction indicated that Zot structures from these three bacterial species were highly variable (Fig. 3); however, high-scoring templates for structure generation (mean sequence identity, 11.4%; mean sequence length, 86%) were not available, and this may have contributed to the differences observed in the 3D structures. Analysis of the secondary structures generated from the tertiary structure prediction showed an overall similarity between the Zot secondary struc- tures of C. concisus, Ne. meningitidis and V. cholerae, with the exception of a few minor changes (Fig. 3). These findings support the hypothesis that C. concisus is capable of attaching to and invading host cells through a paracellular mechanism in which it targets the host cell tight junctions by expressing Zot. Interactions between virulence factors and the C. concisus secretome Further analyses on the secretome of C. concisus UNSWCD included the identification of interactions between individual secreted proteins and the virulence factors outlined in Table 3. Physical and functional associations between proteins were searched for using the string database for known and predicted protein interactions, and six secreted proteins were recognized to be putatively interacting with virulence factors. The disulfide bond-forming protein DsbA, encoded by ccc13826_0002, was secreted by C. concisus UNSWCD. DsbA is reported to be essential for the pathogenic process of many bacteria [36,37], where it plays a critical role in the production of secreted virulence factors in pathogens [38]. One such example is the secretion of the pertussis toxin by Bordetella pertussis [39]. The inactiva- tion of DsbA results in the perturbation of redox homeo- stasis within the bacterial periplasm, and, as a result, sulfydryl-containing proteins are not properly folded. Translocation protein TolB, encoded by ccc1 3826_0922, and ADP-heptose-LPS heptosyltransferase II, encoded by ccc13826_0924, were found within the network of OMP18, an immunodominant antigen in both C. jejuni and H. pylori [25,26]. These associations would probably have resulted from the proximity of the ORFs of the two secreted proteins to ccc13826_0923, which encodes OMP18. Holo-(acyl carrier protein) synthase, encoded by ccc13826_2069, was found to interact with FlaC, a flagellin-like protein. As previously discussed, this pro- tein is secreted by C. jejuni and is capable of binding host cells and modulating the invasion process [24]. The carbon–nitrogen family hydrolase was putatively associ- ated with the invasion protein CiaB. Approximately 14 Cia proteins have been shown to be synthesized when C. jejuni is cocultured with epithelial cells, a subset of which are secreted in the presence of eukaryotic cells. Mutation of the ciaB gene has been shown to inhibit the secretion of all other Cia proteins and significantly reduce the number of internalized C. jejuni cells when compared with the wild-type parent strain [3,40]. One further association between the secretome and virulence was found between the peptidoglycan-associ- ated lipoprotein encoded by ccc13826_0131 and invasin Campylobacter concisus secretome N. O. Kaakoush et al. 1612 FEBS Journal 277 (2010) 1606–1617 ª 2010 The Authors Journal compilation ª 2010 FEBS InvA. This invasin is an essential component of the invasion-associated type III secretion system in Salmo- nella spp. [41]. InvA-expressing bacteria enter the host cell through the invasin-mediated pathway, and are subsequently delivered to lysosomes [42]. Interestingly, even though C. jejuni possesses InvA within its gen- ome, Watson and Gala ´ n [42] have shown that it avoids delivery into lysosomes after entering the cell via a unique caveolae-dependent entry pathway. Conclusions Analysis of the secretome of C. concisus identified a number of proteins that are involved in the general func- tion of the bacterial cell, as well as a number of potential virulence factors. The presence of virulence factors in the secreted proteins of C. concisus, which are likely to come into contact with the host cell more readily than mem- brane-bound proteins, increases the likelihood that C. concisus is a pathogen of the gastrointestinal tract. Experimental procedures Materials Blood Agar Base No. 2, Brain Heart Infusion medium, defibrinated horse blood and gas-generating CampyGen packs were from Oxoid (Heidelberg West, Victoria, Austra- lia). Bicinchoninic acid, BSA, Chaps, copper II sulfate, b-cyclodextrin, dithiothreitol, iodoacetamide and trichloro- acetic acid (TCA) were from Sigma (Castle Hill, NSW, Australia). Vancomycin was from Eli Lilly (North Ryde, Fig. 2. Sequence alignment of zonula occlu- dens toxins found in the C. concisus 13826, V. cholerae 86015 and Ne. meningitidis MC58 genomes. Four highly conserved domains are indicated by boxes. N. O. Kaakoush et al. Campylobacter concisus secretome FEBS Journal 277 (2010) 1606–1617 ª 2010 The Authors Journal compilation ª 2010 FEBS 1613 NSW, Australia). Tris base and SDS were from Amersham Biosciences (Melbourne, Australia). All other reagents were of analytical grade. Bioinformatics and protein modeling blastp searches were performed using complete protein sequences available at the NCBI database (http://www. ncbi.nlm.nih.gov/) against the genome of C. concisus 13826 (CP000792; GI:157101370). The Kyoto Encyclopedia of Genes and Genomes [43], available at (http://www.geno- me.jp/kegg), was employed to determine the biochemical pathways to which genes were assigned. The Search Tool for the Retrieval of Interacting Proteins (string) is a data- base of known and predicted protein–protein interactions available at http://string.embl.de/. string was employed to examine interactions between proteins. The presence and location of signal peptide cleavage sites in the amino acid sequences were predicted using the default settings for Gram-negative bacteria on the signalp 3.0 server (http://www.cbs.dtu.dk/services/SignalP/) [44]. Nonclassical- ly secreted proteins were predicted using the secretomep 2.0 server (http://www.cbs.dtu.dk/services/SecretomeP/). MS data were submitted to the Proteomics Identifications (PRIDE) database, available at http://www.ebi.ac.uk/pride/. Protein structure files were compiled from the protein data bank available at http://www.rcsb.org/pdb. Comparative modeling of proteins was performed using the loopp parallel driver version 3.2, available at http://cbsuapps.tc.cornell.edu/ loopp.aspx. Protein structures were viewed using deepview ⁄ swiss-pdbviewer [45]. Growth of C. concisus and preparation of secreted proteins C. concisus UNSWCD was grown on horse blood agar supplemented with 6% defibrinated horse blood and 5.0 lgÆ mL )1 vancomycin. Cultures were incubated at 37 °C under microaerobic conditions generated using Campylo- bacter gas-generating kits (Cat. no. BR0056A; Oxoid). The purity of bacterial cultures was confirmed by motility and morphology observed under phase contrast microscopy. Secreted proteins were prepared using a modified version of the method described by Bumann et al. [46]. Briefly, log phase C. concisus was harvested and inoculated into 50 mL of Brain Heart Infusion broth. Six 50 mL cultures were grown overnight at 37 °C under microaerobic conditions. Exponential cultures were centrifuged at 4 °C and 4000 g for 15 min, and the supernatant was filtered through a 0.45 lm membrane filter to remove residual bacteria. Secreted pro- teins were precipitated using a previously described modified TCA method [47]. Three hundred milliliters of filtrate was mixed with 95 mL of prechilled TCA and incubated on ice–water for 15 min. The mixture was then centrifuged for 10 min at 4000 g at 4 °C before the pellet was resuspended in 10 mL of acetone, after which it was centrifuged, washed C. concisus Zot Ne. meningitidis Zot V. cholerae Zot A C. concisus Zot Ne.meningitidis Zot V. cholerae Zot B Fig. 3. Predictions of the tertiary structures (A) and secondary structures (B) of zonula occludens toxins found in the C. concisus 13826, V. cholerae 86015 and Ne. meningitidis MC58 genomes. Arrows represent strands, rectangles represent helices and lines represent loops within the secondary structures (B). Campylobacter concisus secretome N. O. Kaakoush et al. 1614 FEBS Journal 277 (2010) 1606–1617 ª 2010 The Authors Journal compilation ª 2010 FEBS with acetone twice, and air-dried. Proteins were then resus- pended in the appropriate buffer and stored at )80 °C. Estimation of the protein content of the samples was performed using the bicinchoninic acid method, employing a microtiter protocol (Pierce, Rockford, IL, USA). Absor- bances were measured using a Beckman Du 7500 spectro- photometer. One-dimensional PAGE Secreted proteins (40 lg) were resuspended in 40 lLof SDS ⁄ PAGE sample buffer (0.375 m Tris, pH 6.8, 0.01% SDS, 20% glycerol, 40 mgÆmL )1 SDS, 31 mgÆmL )1 dith- iothreitol, 1 lgÆmL )1 bromophenol blue). For electropho- retic analyses, proteins were further denatured by heating at 95 °C for 5 min. Proteins were separated on 12% SDS ⁄ PAGE gels by electrophoresis for 2 h at 100 V. Gels were stained using Coomassie Brilliant Blue G-250 (Bio-Rad, Gladesville, Australia). Two-dimensional PAGE Strip rehydration, isoelectric focusing and SDS ⁄ PAGE were performed according to the protocol supplied with the Ready- Strip IPG strips (Bio-Rad). For each strip, protein aliquots (200 lg) were suspended in 245 lL of a rehydration buffer consisting of 8 m urea, 100 mm dithiothreitol, 65 mm Chaps, 40 mm Tris ⁄ HCl, pH 8.0, and 10 lL pH 4–7 IPG buffer. Nuclease buffer (5 lL) was added, and the mixture was incu- bated at 4 °C for 20 min. The sample was then centrifuged at 7230 g for 15 min at 4 °C, and the supernatant was loaded for the first dimension of chromatography onto an 11 cm Ready- Strip IPG (Bio-Rad) of the appropriate pI range, and left to incubate, sealed, for 24 h at room temperature. Isoelectric focusing was performed using an IsoeletrIQ Focusing System (Proteome Systems, Sydney, NSW, Australia). The machine was programmed to run at 300 V for 4 h, 10 000 V for 8 h, and 10 000 V for 22 h, or until 80 000 Volt-hours was reached. After focusing, strips were equilibrated sequentially in two buffers of 6 m urea, 20% (w ⁄ w) glycerol, 2% (w ⁄ v) SDS, and 375 mm Tris ⁄ HCl: the first one contained 130 mm dithiothreitol, and the second one contained 135 mm iodoace- tamide. Strips were rinsed briefly with 0.375 m (pH 8.0) Tris before SDS ⁄ PAGE was performed using Criterion 12.5% Tris ⁄ HCl Precast gels (Bio-Rad), run at 200 V for approxi- mately 45 min. Gels were fixed individually in 0.1 L of fixing solution [50% (v ⁄ v) methanol, 10% (v ⁄ v) acetic acid] for a minimum of 1 h, and were subsequently stained using a sensi- tive ammoniacal silver method based on silver nitrate [48]. MS Protein-containing spots or bands excised from the gels were digested according to a previously described method [49]. Digests originating from 2D-PAGE spots were analyzed on an API QStar Pulsar I tandem MS instrument, using a previ- ously described protocol [48]. Digests (2.5 lL) originating from 1D-PAGE bands were separated by nano-LC using an Ultimate 3000 HPLC and autosampler system (Dionex, Amsterdam, The Netherlands). Samples were concentrated and desalted onto a micro C18 precolumn (500 lm · 2 mm; Michrom Bioresources, Auburn, CA, USA) with H 2 O ⁄ CH 3 CN (98 : 2, 0.05% heptafluorobutyric acid) at 15 lLÆmin )1 . After a 4 min wash, the precolumn was switched (Valco 10 port valve; Dionex) in line with a fritless nano column (75 lm ·  10 cm) containing C18 medium (5 lm, 200 A ˚ ; Magic, Michrom), manufactured according to Gatlin et al. [50]. Peptides were eluted using a linear gradient of H 2 O ⁄ CH 3 CN (98 : 2, 0.1% formic acid to 64 : 36, 0.1% formic acid) at 250 nLÆmin )1 over 30 min. High voltage (1800 V) was applied to a low-volume tee (Upchurch Scien- tific; Oak Harbor, WA, USA), and the column tip was posi- tioned  0.5 cm from the heated capillary (T = 250 °C) of an LTQ FT Ultra mass spectrometer (Thermo Electron, Bremen, Germany). Positive ions were generated by electro- spray, and the LTQ FT Ultra was operated in data-depen- dent acquisition mode. A survey scan (m ⁄ z 350–1750) was acquired in the Fourier transform ion cyclotron resonance cell (resolution = 100 000 at m ⁄ z 400, with an accumulation target value of 1 000 000 ions in the linear ion trap). Up to six of the most abundant ions (> 3000 counts) with charge states of > +2 were sequentially isolated and fragmented within the linear ion trap, using collisionally induced dissoci- ation with an activation of q = 0.25 and activation time of 30 ms at a target value of 30 000 ions. m ⁄ z ratios selected for MS ⁄ MS were dynamically excluded for 30 s. Peak lists were generated using mascot daemon ⁄ extract_msn (Matrix Sci- ence, London, UK), using the default parameters, and sub- mitted to the database search program mascot (version 2.2; Matrix Science). Search parameters were as follows: precur- sor tolerance was 4 p.p.m., and product ion tolerances were ±0.4 Da; Met(O) was specified as a variable modification, enzyme specificity was trypsin, one missed cleavage was pos- sible, and the NCBI nr (July, 2009) or C. concisus (strain 13826) databases were searched (Figs S1 and S2). A false- positive rate of  2% was applied to searches from the LTQ FT-MS data. Acknowledgements This work was made possible by the support of the National Health and Medical Research Council, Aus- tralia. References 1 Moore JE, Corcoran D, Dooley JS, Fanning S, Lucey B, Matsuda M, McDowell DA, Megraud F, Millar BC, N. O. Kaakoush et al. Campylobacter concisus secretome FEBS Journal 277 (2010) 1606–1617 ª 2010 The Authors Journal compilation ª 2010 FEBS 1615 [...]... Characterization of a haemolytic phospholipase A(2) activity in clinical isolates of Campylobacter concisus J Med Microbiol 53, 483–493 16 Istivan TS, Smith SC, Fry BN & Coloe PJ (2008) Characterization of Campylobacter concisus hemolysins FEMS Immunol Med Microbiol 54, 224–235 17 Aabenhus R, Stenram U, Andersen LP, Permin H & Ljungh A (2008) First attempt to produce experimental Campylobacter concisus infection... 691–701 Galan JE, Ginocchio C & Costeas P (1992) Molecular and functional characterization of the Salmonella invasion gene invA: homology of InvA to members of a new protein family J Bacteriol 174, 4338–4349 Campylobacter concisus secretome 42 Watson RO & Galan JE (2008) Campylobacter jejuni survives within epithelial cells by avoiding delivery to lysosomes PLoS Pathog 4, e14, doi:10.1371/journal ppat.0040014... induction of apoptotic death by cytolethal distending toxin Infect Immun 73, 5194–5197 Van Etterijck R, Breynaert J, Revets H, Devreker T, Vandenplas Y, Vandamme P & Lauwers S (1996) Isolation of Campylobacter concisus from feces of children with and without diarrhea J Clin Microbiol 34, 2304– 2306 Vandamme P, Falsen E, Pot B, Hoste B, Kersters K & De Ley J (1989) Identification of EF group 22 campylobacters.. .Campylobacter concisus secretome 2 3 4 5 6 7 8 9 10 11 12 13 14 N O Kaakoush et al O’Mahony R et al (2005) Campylobacter Vet Res 36, 351–382 Poly F & Guerry P (2008) Pathogenesis of Campylobacter Curr Opin Gastroenterol 24, 27–31 Konkel ME, Klena JD, Rivera-Amill V, Monteville MR, Biswas D, Raphael B & Mickelson J (2004) Secretion of virulence proteins from Campylobacter jejuni... Transcellular translocation of Campylobacter jejuni across human polarised epithelial monolayers FEMS Microbiol Lett 179, 209–215 MacCallum A, Hardy SP & Everest PH (2005) Campylobacter jejuni inhibits the absorptive transport functions of Caco-2 cells and disrupts cellular tight junctions Microbiology 151, 2451–2458 Hickey TE, Majam G & Guerry P (2005) Intracellular survival of Campylobacter jejuni in... (2005) Identification of a fibronectin-binding domain within the Campylobacter jejuni CadF protein Mol Microbiol 57, 1022–1035 23 Lally ET, Hill RB, Kieba IR & Korostoff J (1999) The interaction between RTX toxins and target cells Trends Microbiol 7, 356–361 24 Song YC, Jin S, Louie H, Ng D, Lau R, Zhang Y, Weerasekera R, Al Rashid S, Ward LA, Der SD et al (2004) FlaC, a protein of Campylobacter jejuni... secreted through the flagellar apparatus, binds epithelial cells and influences cell invasion Mol Microbiol 53, 541–553 25 Burnens A, Stucki U, Nicolet J & Frey J (1995) Identification and characterization of an immunogenic outer membrane protein of Campylobacter jejuni J Clin Microbiol 33, 2826–2832 26 Kimmel B, Bosserhoff A, Frank R, Gross R, Goebel W & Beier D (2000) Identification of immunodominant... Leach ST, Lemberg DA, Dutt S, Stormon M, Otley A, O’Loughlin EV, Magoffin A et al (2009) Detection and isolation of Campylobacter species other than C jejuni from children with Crohn’s disease J Clin Microbiol 47, 453–455 Engberg J, Bang DD, Aabenhus R, Aarestrup FM, Fussing V & Gerner-Smidt P (2005) Campylobacter concisus: an evaluation of certain phenotypic and genotypic characteristics Clin Microbiol... available: Table S1 Secreted proteins identified from cultures of C concisus UNSWCD (n = 201) Fig S1 Peptide summary reports for LTQ FT-MS searches against the NCBInr database (July, 2009) Fig S2 Peptide summary reports for LTQ FT-MS searches against the C concisus 13826 proteome This supplementary material can be found in the online version of this article Please note: As a service to our authors and... group 22 campylobacters from gastroenteritis cases as Campylobacter concisus J Clin Microbiol 27, 1775–1781 Johnson CC & Finegold SM (1987) Uncommonly encountered, motile, anaerobic gram-negative bacilli associated with infection Rev Infect Dis 9, 1150–1162 Aabenhus R, Permin H, On SL & Andersen LP (2002) Prevalence of Campylobacter concisus in diarrhoea of immunocompromised patients Scand J Infect Dis . virulence factors and the C. concisus secretome Further analyses on the secretome of C. concisus UNSWCD included the identification of interactions between. gen- ome of C. concisus 13826, to determine whether any of these were present. This resulted in the identification of 25 potential candidates. These included