SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Helicobacter pylori is one of the most common bacterial pathogens, infecting about 50% of the world population. The presence of a pathogenicity island (PAI) in H. pylori has been associated with gastric disease. We present evidence that the H. pylori protein encoded by the cytotoxin-associated gene A (cagA) is translocated and phosphorylated in infected epithelial cells. Two-dimensional gel electrophoresis (2-DE) of proteins isolated from infected AGS cells revealed H. pylori strain-specific and time-dependent tyrosine phosphorylation and dephosphorylation of several 125–135 kDa and 75–80 kDa proteins. Immunoblotting studies, matrix-assisted laser desorption/ionization mass spectrometry (MALDI-MS), cell fractionation and confocal microscopy demonstrated that one of the 125–135 kDa proteins represents the H. pylori CagA protein, which is translocated into the host cell membrane and the cytoplasm. Translocation of CagA was dependent on functional cagA gene and virulence (vir) genes of a type IV secretion apparatus composed of virB4, virB7, virB10, virB11 and virD4 encoded in the cag PAI of H. pylori. Our findings support the view that H. pylori actively translocates virulence determinants, including CagA, which could be involved in the development of a variety of gastric disease.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Helicobacter pylori, a Gram-negative microaerophilic bacterial pathogen and class I carcinogen, is specialized to live for decades in the extreme environment of the human stomach. H. pylori colonizes and interacts with gastric epithelial cells throughout the host's life and induces gastric inflammation, which can progress to a variety of diseases, such as peptic ulcer, mucosa-associated lymphoma or gastric cancer ( Covacci et al., 1999 ; Cover and Blaser, 1999). The recent determination of two complete H. pylori genome sequences of strains 26695 and J99 provided hints to the presence of several putative virulence factors ( Tomb et al., 1997 ; Alm et al., 1999 ; Taylor, 1999), those studied in most detail being involved in the production of abundant amounts of urease, the expression of the vacuolating toxin (VacA) or CagA and adhesion to tissue-specific cell receptors ( Guruge et al., 1998; Covacci et al., 1999 ). The major disease-associated genetic difference in H. pylori isolates is the presence (cag+ or type I strains) or absence (cag or type II strains) of the cag pathogenicity island (PAI), a locus of about 40 kb containing up to 31 genes ( Censini et al., 1996 ; Akopyants et al., 1998 ).

Six of the H. pylori cag genes are homologous to the well-known virulence genes virB4, virB7, virB9, virB10, virB11 and virD4 from Agrobacterium tumefaciens, Bordetella pertussis or Legionella pneumophila ( Winans et al., 1996 ; Christie, 1997; Covacci et al., 1999 ). By analogy of the cag genes to the VirB system of the agrobacterial transfer (T)-DNA-transfer machinery to plant cells, it was suggested recently that the cag PAI might code for an ancient type IV secretion transporter that is capable of exporting a variety of proteinaceous material, and probably also nucleoprotein particles ( Censini et al., 1996 ; Christie, 1997; Backert et al., 1998 ). In A. tumefaciens, VirB4 exhibits ATPase activity and is tightly associated with the inner cytoplasmic membrane. H. pylori strain 26695 contains altogether four virB4 gene copies. One of them is located inside, but three are outside of the cag PAI. VirB7 is a lipoprotein that, after recruiting VirB9 to form a heterodimer, is postulated to play a role in stabilizing other agrobacterial VirB proteins by the formation of intermolecular disulphide bridges, leading to the assembly of a functional transporter. This heterodimer seems to stabilize VirB10 in higher order complexes. VirB11 is also essential in the assembly of the transporter and exhibits ATPase activity. A ring-shaped structure composed of six monomeric H. pylori VirB11 proteins has been observed in vitro ( Krause et al., 2000 ). A potential virB2 gene encoding a pilus-like structure has not been identified yet in H. pylori. Interestingly, members of the VirD4 protein family that have been suggested to link the T-DNA complex directly to the exporting membrane channel have only been detected in agrobacterial Ti-plasmid and conjugative DNA transfer systems of broad-host-range plasmids but not in protein transporters, such as that of B. pertussis ( Pansegrau and Lanka, 1996). However, despite the significance of potential virulence genes present, their functional importance as a protein and/or DNA transporter is not known, and the mechanisms by which H. pylori causes gastritis, ulcers or cancer remain poorly understood.

It has been demonstrated that H. pylori triggers, in a cag-dependent manner, cellular events such as the reorganization of the actin cytoskeleton and pedestal formation ( Segal et al., 1996 ), the release of cytokines and the activation of host signalling pathways, leading to the activation of the transcription factors AP-1 and NF-κB ( Tummuru et al., 1995 ; Aihara et al., 1997 ; Keates et al., 1997 ; Münzenmaier et al., 1997 ; Sharma et al., 1998 ; Li et al., 1999 ; Naumann et al., 1999 ). The development of severe gastric disease, which is typically associated with an infection of strains carrying the cag PAI, suggests that the activation of the above-mentioned components of the proinflammatory cellular response is crucial for the clinical outcome. Although recent data described in detail the cellular host response leading to the transcriptional activation of certain target genes ( Naumann et al., 1999 ), no bacterial components directing this signalling have been identified so far. H. pylori was recently shown to induce tyrosine phosphorylation of 105–145 kDa proteins ( Segal et al., 1996 ; 1997; Su et al., 1999 ), and deconvolution immune fluorescence microscopy suggested that a percentage of bacteria attached to the cell surface is associated with a high concentration of phosphorylated CagA, which could induce cellular changes, such as elongation and spreading of host cells, including the production of filopodia and lamellipodia ( Segal et al., 1999 ).

We are interested in clarifying the role of the type IV secretion apparatus of H. pylori in the induction of host signalling responses. Our experimental approach involves the identification of bacterial proteins targeted into the host cell. Data presented here demonstrate that CagA is translocated into the host membrane and into the cytoplasm by the H. pylori type IV secretion machinery encoded in the cag PAI. The CagA protein becomes phosphorylated during infection in a strain- and time-dependent manner. This process is embedded in a scenario of certain protein tyrosine phosphorylation and dephosphorylation events within the host cell that may be crucial for H. pylori virulence.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Type IV secretion-dependent phosphorylation of the H. pylori CagA protein during infection

In order to understand the molecular mechanisms of the H. pylori infection and the potential role of the type IV secretion apparatus encoded in the cag PAI, we have produced a set of vir gene knock-out mutants according to our current working model depicting the putative position of the H. pylori Vir proteins assembled as a transporter complex in the bacterial membrane by analogy with the agrobacterial T-DNA secretion machinery ( Fig. 1). To study the cellular cross-talk between the pathogen and the host cell, subconfluent monolayers of the human gastric adenocarcinoma cell line AGS were infected with different H. pylori strains, followed by analysis of the phosphotyrosine-modified proteins using an anti-phosphotyrosine antibody. Several H. pylori type I-wild-type strains, including P1, P12 and P210, induced tyrosine phosphorylation as well as dephosphorylation of 125–135 kDa proteins and dephosphorylation of 75–80 kDa proteins ( Fig. 2). Tyrosine phosphorylation and dephosphorylation events have not been detected in the type I strain J99, in the type II strain 1061 lacking the complete cag PAI and in mutants for the core components virB4, virB7, virB10 (data not shown), virB11 and virD4 ( Fig. 2, top) of the type IV transporter encoded in the cag PAI. These results were obtained using a multiplicity of infection (MOI) of 50 for a 2 h infection and did not change significantly at MOIs of 100 or 200. Molecular weight differences of phosphorylated 125–135 kDa proteins observed in infections with different H. pylori wild-type strains suggested that this protein could represent a phosphorylated bacterial protein rather than a host protein. The observation that the isogenic mutant of cagA abolished the presence of the respective 125 kDa protein and that CagA is not a proposed membrane component of the type IV secretion apparatus ( Covacci et al., 1999 ) led to the assumption that CagA is a bacterial protein translocated in the host cell. Reprobing of the blot with an anti-CagA antibody confirmed that one of the phosphorylated 125–135 kDa proteins represents the CagA protein ( Fig. 2, bottom). CagA proteins are also present in attached H. pylori of strain J99 and vir gene mutants, but exclusively in the non-phosphorylated form ( Fig. 2, compare top and bottom panels). As a control, H. pylori without cell contact does not show phosphorylated CagA, suggesting that host cell contact is a prerequisite for CagA phosphorylation ( Fig. 2, top). These observations are consistent with the idea that CagA is phosphorylated only during infections of host cells with H. pylori strains encoding functional genes of the proposed type IV secretion machinery. However, we cannot rule out the existence of a polar effect in some of our vir mutants. The motility of some of the mutants was slightly reduced, but none of these mutations significantly affected the ability of the bacterium to adhere to host cells, or the production of the CagA protein ( Fig. 2, bottom).

image

Figure 1. Hypothetical model of the H. pylori type IV secretion machinery encoded in the cag PAI. The putative position of the core proteins VirB4, VirB7, VirB9, VirB10, VirB11 and VirD4 in the bacterial inner membrane (IM) and outer membrane (OM) is presented by homology with models of the agrobacterial and B. pertussis type IV secretion apparatus ( Winans et al., 1996 ; Christie, 1997). The agrobacterial VirB7 and VirB9 proteins form intermolecular heterodimers linked via a disulphide bridge (B7-s-s-B9). Purified H. pylori VirB11 proteins form a hexameric ring in vitro ( Krause et al., 2000 ). Potential ATP-binding domains in the VirB4, VirB11 and VirD4 proteins are indicated as determined from the Prosite database using the husar program ( http://www.genius.embnet.dkfz-heidelberg.de). According to this working model, we have produced isogenic knock-out mutants of the respective genes in H. pylori strain P1 to investigate their relevance in the H. pylori infection process.

Download figure to PowerPoint

image

Figure 2. Phosphorylation of the H. pylori CagA protein during infection. Immunoblot analysis of AGS cells infected with H. pylori wild-type strains and mutants. Total protein isolated from H. pylori strain P1 served as a control. The blot was probed with an antiphosphotyrosine antibody (top) and reprobed with an anti-CagA serum (bottom). Representative results from eight independent experiments are shown. In each experiment, H. pylori infection was for 2 h using a MOI of 50. Molecular size markers (in kDa) are indicated.

Download figure to PowerPoint

In a time course, phosphorylation of CagA appeared 15 min after attachment and was strongly established after 1 h accompanied by a strong dephosphorylation of the 75–80 kDa proteins ( Fig. 3, top). Reprobing of this blot with an anti-CagA antibody showed gradually increasing amounts of the CagA protein present during infection, which is probably the result of an increasing number of attached bacteria and translocated CagA during infection ( Fig. 3, bottom). Incubation of AGS cells with the supernatant of the H. pylori culture did not induce changes in the tyrosine phosphorylation pattern of proteins, suggesting that CagA is not secreted into the medium.

image

Figure 3. Time-dependent induction of tyrosine phosphorylation of the CagA protein (see arrow) and dephosphorylation of 75–80 kDa proteins (see arrowhead) in AGS cells upon infection with H. pylori wild-type strain P1. As a control, the supernatant (sn) of H. pylori strain P1 was added to AGS cells for 180 min. The blot was probed with an antiphosphotyrosine antibody (top) and reprobed with an anti-CagA serum (bottom). Representative results from three independent experiments are shown. A MOI of 50 was used in each experiment. Molecular size markers (in kDa) are indicated.

Download figure to PowerPoint

Identification of CagA by two-dimensional gel electrophoresis and matrix-assisted laser desorption/ionization mass spectrometry

To investigate protein tyrosine phosphorylation and dephosphorylation events of infected AGS cells in more detail and to confirm the identification of CagA by matrix-assisted laser desorption/ionization mass spectrometry (MALDI-MS) peptide mass fingerprinting, we compared two-dimensional gel electrophoresis (2-DE) phosphotyrosine patterns of non-infected and infected AGS cells after removal of the attached bacteria. Figure 4A shows a silver-stained gel of AGS cell proteins to which quantities of protein samples isolated from AGS cells infected with wild-type H. pylori strain P1 or mutants, such as virD4 or cagA, have been standardized. Respective 2-DE blots probed with an anti-phosphotyrosine antibody are shown in Fig. 4B. These blots revealed several significant changes with respect to the phosphorylation status of proteins during host cell infection. In the complete scenario, we detected about six major differences. Most of them were composed of at least four completely or partially dephosphorylated proteins (black circle), and two major 100 and 130 kDa phosphorylated protein species appeared only after infection with wild-type H. pylori (white circle).

image

Figure 4. Phosphorylation patterns of proteins isolated from H. pylori-infected AGS cells analysed by two-dimensional gel electrophoresis (2-DE). Proteins were prepared from non-infected or AGS cells infected with H. pylori P1 wild-type, virD4 and cagA mutants.A. Silver-stained gel of proteins from non-infected AGS cells.B. Immunoblot analysis with an antiphosphotyrosine antibody. Several proteins were found to be phosphorylated and dephosphorylated during infection with the H. pylori wild-type strain. The white circle indicates a set of two major 100 and 130 kDa phosphorylated protein species. The arrow indicates the position of the CagA protein, as later confirmed with MALDI-MS (see Fig. 5). The black circle surrounds a group of several dephosphorylated protein species after infection with wild-type H. pylori. H. pylori infection was for 2 h using a MOI of 50. Results are representative of three independent experiments.

Download figure to PowerPoint

To identify possible H. pylori proteins on these patterns, we compared them with the silver-stained 2-DE protein expression patterns of H. pylori wild-type strains P1 ( Fig. 5A), 26659, J99 and cagA mutant (data not shown). Overlays of these images revealed one major difference between several wild-type strains and the cag mutant: only the wild-type strains gave rise to a 130 kDa protein (compare Figs 4B and 5A, see arrow). We excised this protein from the Coomassie-stained gel and identified more than 30 peptides (marked with asterisks) by in-gel tryptic digestion followed by MALDI-MS ( Fig. 5B). All peptides represented amino acid sequences of the H. pylori CagA protein, giving rise to a sequence coverage of 32%. The position of the peptides in the complete CagA protein sequence is shown in Fig. 5C. Two potential tyrosine phosphorylation sites in the CagA sequence are shadowed with a black circle. Our experiments finally demonstrate that one of the 125–135 kDa phosphorylated proteins is indeed CagA. The lack of signals on 2DE blots during infection with the H. pylori vir mutants, such as virD4 ( Fig. 4B, white circles), supports the hypothesis that translocation into the host cell is required for phosphorylation of the CagA protein.

image

Figure 5. Identification of CagA protein by two-dimensional gel electrophoresis (2-DE) and MALDI-MS.A. Silver-stained gel of proteins from H. pylori strain P1. The arrow indicates the position of the CagA protein, as confirmed with MALDI-MS. B. CagA MALDI-MS. Peaks marked with an asterisk were matched against the H. pylori database and correspond to CagA.C. Amino acid sequence of the 132.4 kDa CagA protein from H. pylori strain 26695 (ORF 547; Tomb et al., 1997 ; accession no. P55980). Peptide sequences identified by MALDI-MS after digestion with trypsin are highlighted in black. Overlapping peptide sequences are marked with bars. Two potential tyrosine phosphorylation sites at positions 116–123 and 892–900 have been determined with the Prosite database using the husar program ( http://www.genius.embnet.dkfz-heidelberg.de). Tyrosine residues in these motifs are shadowed with a black circle.

Download figure to PowerPoint

Translocation of CagA into the host cell membrane and cytoplasm

To confirm our hypothesis and to identify the CagA protein within cells, AGS cells were infected with H. pylori wild-type strain P1 and its virB11 mutant for 2 h followed by the separation of membrane as well as cytoplasmic fractions. The 130 kDa CagA protein was identified by immunoblot analysis with an anti-phosphotyrosine antibody and the anti-CagA antibody ( Fig. 6, top). Phosphorylated CagA protein is present in the membrane as well as in the cytoplasm during infection with wild-type H. pylori. In contrast, the isogenic virB11 mutant as an important key protein of the type IV secretion apparatus ( Fig. 1) did not reveal phosphorylated CagA protein at all. This membrane fraction contained CagA as a result of bacteria attached to the host membrane, but in lower amounts. As a control, CagA protein prepared from H. pylori without host cell contact was not phosphorylated. To verify that the presence of CagA in the host is not a preparation artifact, the blot was reprobed with an antibody against H. pylori urease ( Fig. 6, bottom). Urease is a protein that is secreted into the medium in high amounts, but is also present in the bacterial cytosol (data not shown). This protein has never been shown to be translocated into host cells. We observed cross-contamination with urease in the membrane because of attached bacteria but no cross-contamination in the host cytosol, suggesting that the presence of CagA protein in the host membrane and cytoplasmic fractions is caused by translocation of the protein from the bacterium into the host.

image

Figure 6. Translocation of CagA into the host cell. Fractionation of membrane (m) and cytoplasmic (c) proteins after infection of AGS cells with H. pylori wild-type strain P1 and the isogenic virB11 mutant was analysed by SDS–PAGE. Identification of the phosphorylated CagA protein with an antiphosphotyrosine antibody (top). Identification of CagA with an anti-CagA serum (middle). The blot shown in the middle was reprobed with an antibody against H. pylori urease (β-subunit) as a control (bottom). AGS cells were challenged with a MOI of 100 for 2 h. Total proteins isolated from H. pylori strain P1 and the virB11 mutant served as a control. Representative results of three independent experiments are shown.

Download figure to PowerPoint

To visualize CagA protein directly in the host cell and to confirm its translocation from the bacterium into the host membrane and cytoplasm, confocal microscopy was applied. AGS cells were exposed to H. pylori wild-type strains and virB4, virB7, virB10, virB11, virD4 and cagA mutant strains. CagA from type I wild-type strains was detected inside or immediately beneath attached bacteria ( Fig. 7B–D, arrows) as well as in the cytoplasm of host cells ( Fig. 7C and D, arrowheads). Although intracellular H. pylori have been observed sporadically, signals of accumulated CagA in the host cytoplasm (as indicated by the arrowheads) did not show co-localization with bacteria. In contrast, CagA from strains mutated in vir genes, as important components of the type IV secretion apparatus, was exclusively located in bacterial cells, suggesting that translocation was blocked or greatly reduced ( Fig. 7F and G). Non-infected AGS cells were devoid of the CagA signal, and type II strain 1061 lacking the complete cag PAI or cagA mutant showed no CagA signal, as expected (data not shown).

image

Figure 7. Identification of CagA in the host membrane and cytoplasm by laser-scanning confocal immune fluorescence microscopy. H. pylori wild type (top, A–D) and virB11 mutant (bottom, E–H) attached to AGS cells were used to localize CagA protein in the host cell.A and E. Phase-contrast microscopy.B and F. Fields stained for H. pylori.C and G. Fields stained for CagA.D and H. Merged image showing staining for H. pylori (green), CagA (red) and actin (white). Arrows indicate co-localization of H. pylori and CagA (yellow) in the membrane. Arrowheads indicate CagA protein accumulations in the cytoplasm. For better visualization, grey values of confocal images were depicted as pseudocolours. Scale bar = 25 µm. AGS cells were challenged with a MOI of 100 for 2 h. Results are representative of three independent experiments.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

The delivery of bacterial virulence factors into the host by a type IV secretion apparatus has been well described for the agrobacterial T-DNA and B. pertussis toxin export systems (for reviews, see Winans et al., 1996 ; Christie, 1997). By comparison, very little is known about the function of the newly discovered type IV secretion systems in Legionella pneumophila ( Vogel and Isberg, 1999), Brucella suis ( O'Callaghan et al., 1999), Rickettsia prowazekii ( Andersson et al., 1998 ) or Helicobacter pylori ( Covacci et al., 1999 ). The present data provide several lines of evidence for a function of the H. pylori type IV secretion machinery in actively translocating an important virulence determinant, CagA, into the host cell. Although no function for this immune-dominant antigen has been reported so far, we made the following observations that may deepen our understanding of the function of the H. pylori virulence system. (i) The CagA protein is tyrosine phosphorylated during infection with cultured gastric cells in a strain- and time-dependent manner. The CagA protein is not tyrosine phosphorylated in the H. pylori cells without host cell contact. Only several type I, but no type II, wild-type strains were able to induce CagA phosphorylation. (ii) CagA was not secreted into the medium, as supported by the observation that supernatants of H. pylori type I strains did not induce changes in the phosphotyrosine pattern of infected AGS cells. (iii) We have shown the insertion of CagA into the membrane (at positions at which H. pylori is attached) and the cytosol of the host by two independent experimental approaches including fractionation studies and confocal microscopy. (iv) Injection and tyrosine phosphorylation of CagA in the host cell appears to be dependent on important functional core components (vir genes) and the cagA gene of the H. pylori type IV secretion apparatus encoded by the cag PAI (compare Fig. 1). (v) Translocation and phosphorylation of the CagA protein is accompanied by additional tyrosine phosphorylation and dephosphorylation events of several proteins in the host.

To facilitate studies of the H. pylori-induced signalling processes that underlie cag-dependent mechanisms, we have produced a set of cag mutants. The cag mutants were from strain P1, a type I strain that actually encodes all the putative homologues of the agrobacterial virB4, virB7, virB9, virB10, virB11 and virD4 genes reported so far ( Covacci et al., 1999 ), and provokes a very strong host response. We have produced mutants from all these genes, except the virB9 gene; although present in P1, we were not yet able to produce a functional mutant. However, all the open reading frames (ORFs) tested were needed for successful translocation and phosphorylation of the H. pylori CagA protein. Although we cannot completely rule out the occurrence of polar effects on downstream genes as a result of the integrated CmR cassette in some of these mutants, the protein expression patterns and the patterns of tyrosine-phosphorylated proteins of infected AGS cells in 2-DE were consistently observed with different MOIs similar to that produced by type II strains lacking the complete cag PAI. Moreover, our vir mutants were not significantly reduced in their ability to attach to the host cell when compared with its wild type, and also did not significantly affect the expression level of the CagA protein. However, our results, i.e. (i) the inability of the H. pylori vir gene mutant strains to translocate and phosphorylate the CagA protein, and (ii) the homology of these vir genes to that of A. tumefaciens or B. pertussis, are collectively consistent with a model in which different proteins encoded in the cag PAI form a contact-dependent and type IV-like secretion apparatus responsible for direct injection of CagA into the host cell ( Fig. 1). Our results led us to suggest that intact virB4, virB7, virB10, virB11, virD4, cagA and possibly also virB9 gene products are essential in the injection process of CagA.

Our data considerably extend those of Segal et al. (1999) , who first presented evidence of CagA phosphorylation during the infection process. These authors showed that CagA is necessary to induce a growth factor-like phenotype (hummingbird) in host gastric cells and another CagA-independent cellular phenotype called SFA (stress fibre associated). We have also identified phosphorylated CagA protein by several approaches, including MALDI-MS. In addition, searches in the Prosite database revealed two potential tyrosine phosphorylation sites in the CagA sequence of strain 26695. In another strain, J99, CagA is also present during infection, but not in the phosphorylated form. Alignments of CagA protein sequences from strains J99 and 26695 revealed that CagA of J99 is mutated in respective tyrosine residues, which supports the view that both sites detected in strain 26695 could indeed have a function in vivo (unpublished data). This conclusion is supported by our observation that, only in infections with H. pylori type I strains showing phosphorylated CagA protein, were dephosphorylated 75–80 kDa proteins additionally observed in a time-dependent manner. The potential role of these tyrosine phosphorylation sites in the infection process and possibly in gastric disease has to be elucidated in future studies.

The potential function of the translocated CagA protein in the host is unknown. Database searches of CagA sequences did not reveal significant homologies to any known proteins. Recent mutational studies have implicated several genes in the cag PAI in the induction of interleukin (IL)-8 but some, most strikingly cagA, were not involved ( Tummuru et al., 1995 ; Censini et al., 1996 ; Li et al., 1999 ). By analogy with the enteropathogenic Escherichia coli Tir protein, which is secreted in the host membrane by the bacterial type III secretion machinery and subsequently tyrosine phosphorylated, thereafter acting as a receptor for intimin binding on the bacterial surface ( Kenny et al., 1997 ), it is tempting to speculate that CagA could act as a receptor for H. pylori in a similar manner to attract additional bacteria to adhere. Alternatively, other unknown function(s) should also be considered, as CagA has also been detected in the host cytoplasm. As outlined above, translocation and tyrosine phosphorylation of CagA are temporally correlated with dephosphorylation of 75–80 kDa proteins. The nature and function of these host proteins is unknown as yet. It will be very interesting in future to study the correlation between the two events, which may shed new light on H. pylori-induced signalling processes in gastric disease.

In conclusion, our study has revealed that CagA is the first H. pylori protein shown to be translocated into the host. The present in vitro studies emphasize the importance of several agrobacterial vir gene homologues in the translocation and tyrosine phosphorylation of the H. pylori CagA protein, possibly by an active process via a type IV secretion apparatus encoded in the cag PAI. This event is temporally embedded in a scenario of additional protein tyrosine phosphorylation and dephosphorylation events within the host cell, which may be crucial for H. pylori virulence. We feel that further molecular dissection of the cag PAI, as well as CagA translocation and tyrosine phosphorylation-dependent signalling processes, will significantly contribute to an understanding of H. pylori pathogenesis mechanisms.

Experimental procedures

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Bacterial strains and mutagenesis

Type I H. pylori strains P1, P12 and P92 (Sydney strain SS1) are clinical isolates that have been described previously ( Schmitt and Haas, 1994; Corthesy-Theulaz et al., 1996 ; Lee et al., 1997 ). H. pylori strain P210 is a clinical strain isolated from a patient with gastric cancer. The type II H. pylori strain 1061 was kindly provided by J. Bijlsma and H. Kusters (Department of Medical Microbiology, Vrije Universiteit Amsterdam, The Netherlands). Strains 26695 and J99 are H. pylori isolates obtained from TIGR and AstraZeneca respectively ( Tomb et al., 1997 ; Alm et al., 1999 ). Isogenic P1 knock-out mutants have been constructed by insertion of a chloramphenicol resistance gene cassette (CmR, 1 kb BamHI/BglII fragment of plasmid pTnMax1) in cloned virB4 (P205), virB7 (P206), virB10 (P207), virB11 (P208), virD4 (P209) and cagA (P211) genes according to a standard protocol ( Haas et al., 1993 ). Briefly, the genes of interest were amplified by polymerase chain reaction (PCR) using primers based on the sequences of the genes from strains 26695 and J99 in the AstraZeneca H. pylori genome database ( http://www.astra-boston.com/hpylori/). The PCR products were cloned in the pGEM-T vector (Promega). After mutagenesis by insertion of the CmR cassette in the vir genes, 3–5 µg of supercoiled plasmid DNA was added to 1 × 108 bacteria ml−1 brain–heart infusion (BHI) medium. After incubation for 6 h, bacteria were grown on agar plates containing 4–6 µg ml−1 chloramphenicol to select for chloramphenicol-resistant transformants obtained after 4–5 days according to a standard procedure ( Haas et al., 1993 ). Correct integration of the CmR cassette into the H. pylori chromosome by double cross-over recombination was confirmed by PCR. All H. pylori strains were grown on horse serum agar plates supplemented with vancomycin (10 µg ml−1), nystatin (1 µg ml−1) and trimethoprim (5 µg ml−1) and, if necessary, with chloramphenicol (4–6 µg ml−1). Incubation was at 37°C for 2 days in an anaerobic jar containing a gas mix of 5% O2, 10% CO2 and 85% N2 (Oxoid).

Infection assays and immunoblot analysis

AGS cells (ATCC CRL 1739; a human gastric adenocarcinoma epithelial cell line) were grown in 25 cm2 flasks with RPMI/10% FCS for 1 or 2 days respectively. The cells were washed once with PBS, and 4 ml of fresh medium was added to each flask. H. pylori (1 × 108) were resuspended in 0.5 ml of PBS and added to 2 × 106 AGS cells at a MOI of 50. H. pylori culture supernatant was obtained after completely pelleting bacteria at 3000 g for 10 min. After incubation in a 5% CO2/95% air incubator for 1–3 h, AGS cells were washed once with PBS (containing 1 mM sodium vanadate) to remove non-adherent bacteria. Whole-cell lysates with attached bacteria were made by pelleting the cells at 600 g and resuspending the pellet in an equal amount of 2 × SDS lysis buffer (250 mM Tris-HCl, pH 8.0, 4% SDS, 20% glycerol, 0.002% bromophenol blue, 6% 2-mercaptoethanol). SDS–PAGE and electroblotting to Immobilon-P membranes were performed as described by the manufacturers (Millipore, Bio-Rad). Antibody binding and detection were performed with the ECL system (Amersham). Mouse monoclonal antiphosphotyrosine antibody PY99 was purchased from Santa Cruz Biotechnology. Polyclonal antiurease antibodies have been described recently ( Gomez-Duarte et al., 1998 ) . Polyclonal anti-CagA antibodies were produced by immunization of mice with a fusion protein containing a CagA polypeptide that corresponds to residues 110–769 of the CagA sequence reported recently ( Tummuru et al., 1993 ).

Cellular fractionation

For the preparation of membrane/nuclear and cytoplasmic fractions of infected host cells, we followed a special protocol to remove adherent and non-adherent bacteria. For this purpose, 2 × 106 infected cells were washed with PBS (containing 1 mM sodium vanadate) to remove non-adherent bacteria. Adherent bacteria were detached from AGS cells by incubation with 2 ml of trypsin for 5 min. The following steps were carried out at 4°C with precooled solutions. PBS/10% FCS (4 ml) was added to the cells, followed by centrifugation at 600 × g for 5 min. The majority of the remaining bacteria was removed by several washing steps in PBS. The pellets with purified AGS cells were washed again in 1 ml of buffer [10 mM Tris-HCl, pH 7.9, 10 mM KCl, 1.5 mM MgCl2, 10% glycerol, 10 mM K2HPO4, 1 mM sodium vanadate, 10 mM NaF, 0.5 mM dithiothreitol (DTT) and 0.5 mM phenylmethylsulphonyl fluoride (PMSF)]. The cells were centrifuged again at 600 × g for 5 min, and the pellet was resuspended in 200 µl of buffer. Nonidet-P40 (2 µl) was added to the cells, gently mixed and incubated for 1 min on ice. Membrane/nuclear fractions were collected by pelleting at 1000 g for 10 min. Cytoplasmic fractions were purified from the remaining cell debris by centrifugation at 12 000 g for 30 min. The fractions were resuspended in an equal amount of 2 × SDS lysis buffer.

Two-dimensional electrophoresis (2-DE)

For the resolution of proteins isolated from infected AGS cells by 2-DE, we followed the protocol described above and removed adherent and non-adherent bacteria. The purified AGS cells were washed again with PBS and sonicated in the presence of proteinase and phosphatase inhibitors (0.5 mM PMSF, 2 µg ml−1 aprotinin, 2 µg ml−1 leupeptin and 1 mM sodium vanadate). Proteins from H. pylori strains P1, J99 and 26695 were isolated from bacterial pellets obtained after centrifugation at 2000 g. All protein fractions were treated with 9 M urea, 70 mM DTT and 2% Triton X-100 to obtain completely denatured and reduced proteins. For the resolution of proteins, we applied a 2-DE gel system ( Jungblut and Seifert, 1990) in a 7 cm × 8 cm version. Protein (50 µg) was loaded to the anodic side of the isoelectric focusing (IEF) gel. The proteins were detected by silver staining ( Jungblut and Seifert, 1990), by staining with Coomassie brilliant blue R250 and by immunoblot analysis as described above.

Peptide mass spectrometry

For the identification of CagA protein, 300 µg of total protein extract was applied to large-scale 2-DE gel (20 cm × 30 cm) with a resolution power of about 5000 protein species. Respective spots were excised from the gel and subjected to in-gel tryptic digestion using a peptide-collecting device ( Otto et al., 1996 ). The peptide solution was mixed with an equal volume of a saturated α-cyano-4-hydroxy cinnamic acid solution in 50% acetonitrile, 0.3% TFA, and 2 µl was applied to the sample template of a MALDI-MS (Voyager Elite; Perseptive), as described recently ( Jungblut et al., 1999 ). Peptide mass fingerprints were searched using the program msfit ( http://prospector.ucsf.edu/ucsfhtml/msfit.htm), reducing the proteins of the NCBI database to the H. pylori proteins and to a molecular mass range estimated from 2-DE ± 20%, allowing a mass accuracy of 0.1 Da for the peptide mass.

Laser scanning confocal immunofluorescence (IF) microscopy

To localize CagA proteins directly during infection, AGS cells were exposed to H. pylori wild types, vir and cagA mutant strains. Monolayers of 2 × 104 AGS cells were grown overnight in 16-well Lab-Tek chambers (Nunc) and infected with H. pylori cells of the indicated strains at an MOI of 50. Three hours after infection, the cells were washed with PBS to remove non-adherent bacteria and fixed with 4% formaldehyde in PBS. The cultures were permeabilized with 0.1% Triton X-100 in PBS for 30 min and stained with antibodies against H. pylori[rabbit serum 1:100 (Dako)/anti-rabbit IgG-Alexa 594 (Molecular Probes)], CagA (mouse serum 1:200, anti-mouse IgG-Cy5; Jackson), and filamentous actin was visualized using phalloidin-AlexaFluor 488 (Molecular Probes). Specimens were analysed with a confocal microscope (Leica TCS-SP).

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

We are grateful to P. R. Suryanarayanan and Corinna Kik for help in cloning experiments, and Terry Kwok, Stefan Moese and Efterpi Kardalinou for critical discussion of the data. We are also grateful to Stephanie Lamer for help with the MALDI-MS, and Jani O'Rourke (University of New South Wales, School of Microbiology and Immunology, Sydney, Australia), Jetta Bijlsma, Hans Kusters (Department of Medical Microbiology, Vrieje Universiteit Amsterdam, The Netherlands) and Galip Karaali for providing H. pylori strains SS1, 1061 and P210 respectively. This work was supported by grants from the Fonds der Chemischen Industrie to T.F.M. and M.N.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  • 1
    Aihara, M., Tsuchimoto, D., Takizawa, H., Azuma, A., Wakebe, H., Ohmoto, Y., et al. (1997) Mechanisms involved in Helicobacter pylori-induced interleukin-8 production by a gastric cancer cell line, MKN45. Infect Immun 65: 3218 3224.
  • 2
    Akopyants, N.S., Clifton, S.W., Kersulyte, D., Crabtree, J.E., Youree, B.E., Reece, C.A., et al. (1998) Analyses of the cag pathogenicity island of Helicobacter pylori. Mol Microbiol 28: 37 53.
  • 3
    Alm, R.A., Ling, L.S., Moir, D.T., King, B.L., Brown, E.D., Doig, P.C., et al. (1999) Genomic sequence comparison of two unrelated isolates of the human gastric pathogen Helicobacter pylori. Nature 397: 176 180 [published erratum in Nature 397: 719].
  • 4
    Andersson, S.G., Zomorodipour, A., Andersson, J.O., Sicheritz-Ponten, T., Alsmark, U.C., Podowski, R.M., et al. (1998) The genome sequence of Rickettsia prowazekii and the origin of mitochondria. Nature 396: 133 140.
  • 5
    Backert, S., Von Nickisch-Rosenegk, E., Meyer, T.F. (1998) Potential role of two Helicobacter pylori relaxases in DNA transfer? Mol Microbiol 30: 673 674.
  • 6
    Censini, S., Lange, C., Xiang, Z., Crabtree, J.E., Ghiara, P., Borodovsky, M., et al. (1996) cag, a pathogenicity island of Helicobacter pylori, encodes type I-specific and disease-associated virulence factors . Proc Natl Acad Sci USA 93: 14648 14653.
  • 7
    Christie, P.J. (1997) Agrobacterium tumefaciens T-complex transport apparatus: a paradigm for a new family of multifunctional transporters in eubacteria. J Bacteriol 179: 3085 3094.
  • 8
    Corthesy-Theulaz, I., Porta, N., Pringault, E., Racine, L., Bogdanova, A., Kraehenbuhl, J.P., et al. (1996) Adhesion of Helicobacter pylori to polarized T84 human intestinal cell monolayers is pH dependent. Infect Immun 64: 3827 3832.
  • 9
    Covacci, A., Telford, J.L., Del Giudice, G., Parsonnet, J., Rappuoli, R. (1999) Helicobacter pylori virulence and genetic geography. Science 21: 1328 1333.
  • 10
    Cover, T.L. & Blaser, M.J. (1999) Helicobacter pylori factors associated with disease. Gastroenterology 117: 257 261.
  • 11
    Gomez-Duarte, O.G., Lucas, B., Yan, Z.X., Panthel, K., Haas, R., Meyer, T.F. (1998) Protection of mice against gastric colonization by Helicobacter pylori by single oral dose immunization with attenuated Salmonella typhimurium producing urease subunits A and B. Vaccine 16: 460 471.
  • 12
    Guruge, J.L., Falk, P.G., Lorenz, R.G., Dans, M., Wirth, H.P., Blaser, M.J., et al. (1998) Epithelial attachment alters the outcome of Helicobacter pylori infection. Proc Natl Acad Sci USA 95: 3925 3930.
  • 13
    Haas, R., Meyer, T.F., Van Putten, J.P. (1993) Aflagellated mutants of Helicobacter pylori generated by genetic transformation of naturally competent strains using transposon shuttle mutagenesis. Mol Microbiol 8: 753 760.
  • 14
    Jungblut, P.R. & Seifert, R. (1990) Analysis by high-resolution two-dimensional electrophoresis of differentiation-dependent alterations in cytosolic protein pattern of HL-60 leukemic cells. J Biochem Biophys Methods 21: 47 58.
  • 15
    Jungblut, P.R., Schaible, U.E., Mollenkopf, H.J., Zimny-Arndt, U., Raupach, B., Mattow, J., et al. (1999) Comparative proteome analysis of Mycobacterium tuberculosis and Mycobacterium bovis BCG strains: towards functional genomics of microbial pathogens. Mol Microbiol 33: 1103 1117.
  • 16
    Keates, S., Hitti, Y.S., Upton, M., Kelly, C.P. (1997) Helicobacter pylori infection activates NF-kappa B in gastric epithelial cells. Gastroenterology 113: 1099 1109.
  • 17
    Kenny, B., DeVinney, R., Stein, M., Reinscheid, D.J., Frey, E.A., Finlay, B.B. (1997) Enteropathogenic E. coli (EPEC) transfers its receptor for intimate adherence into mammalian cells. Cell 91: 511 520.
  • 18
    Krause, S., Bárcena, M., Pansegrau, W., Lurz, R., Carazo, J.M., Lanka, E. (2000) Sequence related protein export NTPases encoded by the conjugative transfer region of RP4 and by the cag pathogenicity island of Helicobacter pylori share similar hexameric ring structures. Proc Natl Acad Sci USA 97: 3067 3072.
  • 19
    Lee, A., O'rourke, J., De Ungria, M.C., Robertson, B., Daskalopoulos, G., Dixon, M.F. (1997) A standardized mouse model of Helicobacter pylori infection: introducing the Sydney strain. Gastroenterology 112: 1386 1397.
  • 20
    Li, S.D., Kersulyte, D., Lindley, I.J., Neelam, B., Berg, D.E., Crabtree, J.E. (1999) Multiple genes in the left half of the cag pathogenicity island of Helicobacter pylori are required for tyrosine kinase-dependent transcription of interleukin-8 in gastric epithelial cells. Infect Immun 67: 3893 3899.
  • 21
    Münzenmaier, A., Lange, C., Glocker, E., Covacci, A., Moran, A., Bereswill, S., et al. (1997) A secreted/shed product of Helicobacter pylori activates transcription factor nuclear factor-kappa B. J Immunol 159: 6140 6147.
  • 22
    Naumann, M., Wessler, S., Bartsch, C., Wieland, B., Covacci, A., Haas, R., et al. (1999) Activation of activator protein 1 and stress response kinases in epithelial cells colonized by Helicobacter pylori encoding the cag pathogenicity island. J Biol Chem 274: 31655 31662.
  • 23
    O'callaghan, D., Cazevieille, C., Allardet-Servent, A., Boschiroli, M.L., Bourg, G., Foulongne, V., et al. (1999) A homologue of the Agrobacterium tumefaciens VirB and Bordetella pertussis Ptl type IV secretion systems is essential for intracellular survival of Brucella suis. Mol Microbiol 33: 1210 1220.
  • 24
    Otto, A., Thiede, B., Müller, E.-C., Scheler, C., Wittmann-Liebold, B., Jungblut, P. (1996) Identification of human myocardial proteins separated by two-dimensional electrophoresis using an effective sample preparation for mass spectrometry. Electrophoresis 17: 1643 1650.
  • 25
    Pansegrau, W. & Lanka, E. (1996) Enzymology of DNA transfer by conjugative mechanisms. Prog Nucleic Acid Res Mol Biol 54: 197 251.
  • 26
    Schmitt, W. & Haas, R. (1994) Genetic analysis of the Helicobacter pylori vacuolating cytotoxin: structural similarities with the IgA protease type of exported protein. Mol Microbiol 12: 307 319.
  • 27
    Segal, E.D., Falkow, S., Tompkins, L.S. (1996) Helicobacter pylori attachment to gastric cells induces cytoskeletal rearrangements and tyrosine phosphorylation of host cell proteins. Proc Natl Acad Sci USA 93: 1259 1264.
  • 28
    Segal, E.D., Lange, C., Covacci, A., Tompkins, L.S., Falkow, S. (1997) Induction of host signal transduction pathways by Helicobacter pylori. Proc Natl Acad Sci USA 94: 7595 7599.
  • 29
    Segal, E.D., Cha, J., Lo, J., Falkow, S., Tompkins, L.S. (1999) Altered states: involvement of phosphorylated CagA in the induction of host cellular growth changes by Helicobacter pylori. Proc Natl Acad Sci USA 96: 14559 14564.
  • 30
    Sharma, S.A., Tummuru, M.K., Blaser, M.J., Kerr, L.D. (1998) Activation of IL-8 gene expression by Helicobacter pylori is regulated by transcription factor nuclear factor-kappa B in gastric epithelial cells. J Immunol 160: 2401 2407.
  • 31
    Su, B., Johansson, S., Fallman, M., Patarroyo, M., Granstrom, M., Normark, S. (1999) Signal transduction-mediated adherence and entry of Helicobacter pylori into cultured cells. Gastroenterology 117: 595 604.
  • 32
    Taylor, D.E. (1999) Helicobacter pylori and its genome: lessons from the treasure map. Can J Gastroenterol 13: 218 223.
  • 33
    Tomb, J.F., White, O., Kerlavage, A.R., Clayton, R.A., Sutton, G.G., Fleischmann, R.D., et al. (1997) The complete genome sequence of the gastric pathogen Helicobacter pylori. Nature 388: 539 547 [published erratum in Nature 389: 412].
  • 34
    Tummuru, M.K., Cover, T.L., Blaser, M.J. (1993) Cloning and expression of a high-molecular-mass major antigen of Helicobacter pylori: evidence of linkage to cytotoxin production . Infect Immun 61: 1799 1809.
  • 35
    Tummuru, M.K., Sharma, S.A., Blaser, M.J. (1995) Helicobacter pylori picB, a homologue of the Bordetella pertussis toxin secretion protein, is required for induction of IL-8 in gastric epithelial cells. Mol Microbiol 18: 867 876.
  • 36
    Vogel, J.P. & Isberg, R.R. (1999) Cell biology of Legionella pneumophila. Curr Opin Microbiol 2: 30 34.
  • 37
    Winans, S.C., Burns, D.L., Christie, P.J. (1996) Adaptation of a conjugal transfer system for the export of pathogenic macromolecules. Trends Microbiol 4: 64 68.