Keywords

1 Introduction

Helicobacter pylori is one of the best adapted human pathogens that colonizes the surface region in the gastric mucosa of the stomach. Approximately half of the world’s population carries this microbe, often causing asymptomatic gastritis in infected people, but also more severe gastric diseases such as mucosa-associated lymphoid tissue (MALT) lymphoma and gastric cancer may arise. Colonization commonly occurs early in childhood and H. pylori can persist lifelong, if not treated by antimicrobial therapy. Although H. pylori infections are commonly associated with elevated inflammation parameters that are generated by the host innate and adaptive immune systems, the bacteria are not eliminated. Various mechanisms of immune evasion were documented and H. pylori became a prime example of chronic bacterial infections. Host-pathogen interactions are highly complex and determine the clinical outcome of infections. The development of gastric diseases is controlled by the bacterial genotype, genetic predisposition of the host, and environmental factors. For example, specific polymorphisms in host genes with important roles in pro-inflammatory and immune-regulatory signal transduction such as interleukin-1β (IL-1β), interleukin-8 (IL-8), tumor necrosis factor alpha (TNF-α), or Nod-1 ( nucleotide oligomerization domain-1) have been linked to a higher risk of developing H. pylori-triggered gastric diseases (for more details, see Chaps. 3, 13, and 16 of this book). H. pylori strains are highly heterogeneous both in their DNA sequences and virulence properties. Dozens of bacterial genes have been described to control the pathogenesis of H. pylori. There are two classical virulence factors expressed by H. pylori, the CagA protein encoded in the cag (cytotoxin-associated genes) pathogenicity island (cagPAI) and the vacuolating cytotoxin A (VacA). VacA interacts with multiple host molecules and can trigger various downstream signaling cascades as discussed in Chap. 5. Here we summarize our current knowledge on the multiple cagPAI and CagA functions as well as the multitude of affected host signaling cascades with focus on their importance in H. pylori pathogenesis.

2 The cagPAI Encodes a Type IV Secretion System

In the H. pylori research field, worldwide interest is focused on the effector protein CagA. CagA is encoded by highly virulent type I isolates, but is absent in less virulent type II strains. Thus, the protein has been recognized as a molecular marker for the cagPAI locus (Hatakeyama 2003; Backert et al. 2015). The cagPAI encodes functional components of a type IV secretion system ( T4SS). This T4SS represents a pilus-like structure (called the T4SS pilus), induced upon host cell contact and protruding from the bacterial membrane (Fig. 4.1a). T4SS machineries are evolutionary related to DNA conjugation systems (Backert and Meyer 2006). The class of T4SSs is highly diverse both with regard to the delivered factors (DNA, proteins, DNA-protein complexes, or peptidoglycan) and recipient organisms. The latter can be either a bacterium of the same or other species or even species from different kingdoms (e.g., mammalian, fungal, or plant cells). In addition to H. pylori, pathogenicity-linked T4SSs have also been found in Agrobacterium, Legionella, Bordetella, Bartonella, and other species. T4SSs commonly comprise 11 VirB proteins (encoded by the virB1–virB11 genes) and the NTPase coupling protein (VirD4). The prototypic and by far best characterized T4SS is the T-DNA transfer apparatus of Agrobacterium tumefaciens (Waksman and Orlova 2014). The agrobacterial VirB proteins can be classified in three groups: (1) the core or putative channel components (VirB6–VirB10), (2) the pilus-associated proteins (VirB2, and possibly VirB3 and VirB5), and (3) the energetic factors (the NTPases: VirB4 and VirB11). In addition, VirB1 is an enzyme having muraminidase activity, allowing localized lysis of murein layer in the membrane to achieve T4SS assembly at a specific site of the cell (Backert et al. 2015). A model for the individual steps in assembling of the agrobacterial T4SS has been described in numerous review articles (Backert and Meyer 2006; Waksman and Orlova 2014). When having a look at the H. pylori T4SS, all 11 VirB and VirD4 orthologs are encoded in the cagPAI as well as some accessory proteins (Fischer 2011), resulting in a T4SS model (Fig. 4.1b). Immunogold labeling and electron microscopy (EM) have shown that the tips of the T4SS pilus are covered with CagL (Kwok et al. 2007). In addition, EM revealed that CagL, CagI, and CagH proteins were involved in T4SS pilus formation and deletion of a conserved C-terminal hexapeptide motif, which is shared among these proteins, abolished CagA delivery (Shaffer et al. 2011). The T4SS pilus is also decorated locally or completely by CagY (VirB10) (Rohde et al. 2003). CagY is a very large protein (about 250-kDa) that contains two transmembrane domains with the mid region (also called the repeat domain) exposed to the extracellular environment (Rohde et al. 2003). It carries an unusual sequence structure owing to an extraordinary number of direct DNA repeats that are predicted to result in in-frame intersections or deletions (Barrozo et al. 2013). Interestingly, CagY rearrangements are driven by the host and are sufficient to result in gain or loss of function in the T4SS. This molecular switch allowed modification of the immune response to benefit persistent infection of H. pylori (Barrozo et al. 2013). Major other components of the T4SS structure, also visualized by EM, are the VirB7 and VirB9 proteins (Tanaka et al. 2003). Interestingly, T4SS pilus formation requires a host cell receptor (Kwok et al. 2007). This is achieved by a number of T4SS proteins, including CagI, CagL, CagY, and CagA, interacting with the host cell integrin member α5β1, followed by translocation of CagA into the host (Kwok et al. 2007; Jiménez-Soto et al. 2009). Binding of CagA to phosphatidylserine has also been shown to play a role in the translocation process (Murata-Kamiya et al. 2010).

Fig. 4.1
figure 1

Model for the assembled type IV secretion (T4SS) complex in Helicobacter pylori (Hp). (a) The T4SS encoded by the cag pathogenicity island is a multicomponent protein complex spanning the inner and outer membranes. Typical T4SS pili are shown by scanning electron microscopy, connecting the bacterium with the membrane of AGS gastric epithelial cells (arrows). (b) The Hp T4SS exhibits homology to the VirB/VirD4 T4SS machinery of Agrobacterium tumefaciens. T4SS assembly and subcellular localization of the proteins are shown in a simplified manner. Pilus components, core complex proteins, energetic components, and others factors are highlighted with different colors as indicated. The reported substrates for the Hp T4SS are CagA and peptidoglycan. (For more details, see text. Panel a was from Hauck (2007) with kind permission from Nature Publishing and panel b was adapted from Tegtmeyer and coworkers (2011) with kind permission from Wiley.)

Full size image

3 Crystal Structures of cagPAI Proteins

In the last few years, several pieces of information have been gathered about the architecture of bacterial T4SSs, both from EM and single-crystal X-ray diffraction studies. The most complete picture of the apparatus comes from the 18Å–23Å EM structure of the T4SS of Escherichia coli conjugative plasmid R388 (Low et al. 2014). This complex includes 8 proteins (from VirB3 to VirB10) that form a supramolecular assembly spanning the entire cell envelope. The complex can be divided in two parts, the core complex, characterized by C14 symmetry, and the inner membrane complex, with a lower C2 symmetry. The two complexes are connected by a small stalk. In the overall, the entire membrane-spanning machinery is about 340Å high and 255Å wide. A more detailed description at the atomic level comes from the crystal structure of the O-layer (Chandran et al. 2009; Terradot and Waksman 2011) and from EM studies of the outer membrane T4SS core complex from plasmid pKM101 (Fronzes et al. 2009). The O-layer includes 14 copies of VirB10, VirB7, and VirB9, with the VirB10 components forming a tetradecameric channel and the 14 VirB7/VirB9 complexes surrounding and stabilizing it. The T4SS pilus is composed of the major subunit VirB2 and VirB5 (CagL) (Backert et al. 2008). The structure of the latter, which decorates the pilus surface, has been determined (Barden et al. 2013). CagL consists of a four-helix bundle, containing the Arg-Gly-Asp (RGD) motif that represents a recognition site for integrin binding (Kwok et al. 2007). Finally, the structure of VirB11 ATPase revealed a hexameric ring and functions as a gating molecule at the inner membrane, which is proposed to cycle through closed and open forms by ATP binding/hydrolysis (Yeo et al. 2000; Hare et al. 2007).

The crystal structures of some other members of the cagPAI, which are important for CagA secretion or IL-8 induction, have also been determined: CagZ, a 23-kDa protein essential for CagA translocation, consists of a single compact L-shaped domain containing seven α-helices (Cendron et al. 2004); CagS is a 23-kDa single-domain protein characterized by an all-α structure and an unusually high methionine content (Cendron et al. 2007); CagD is a covalent dimer in which each monomer folds as a single domain that is composed of five β-strands and three α-helices (Cendron et al. 2009).

The translocated effector protein CagA in strain 26695 consists of 1,186 amino acids and is structurally organized in several domains. Various crystal structures of the large N-terminal portion, residues 1–876 and 261–829 (Hayashi et al. 2012) and residues 1–884 (Kaplan-Türköz et al. 2012), consist of three domains, one of which is made by two sub-domains (or four domains in total, according to a different interpretation). Domain I (24–221) is composed of 10 α-helices. It makes very few contacts with the other domains, such that its orientation is mobile with respect to the rest of the molecule. Domain II (residues 303–644) constitutes the molecular core and contains a large antiparallel β-sheet flanked by an all α-helical subdomain (residues 370–446) and two long α-helices. A long, partially flexible α-helix connects domain II to domain III (residues 645–824), whose structure is that of a classical four-helix bundle. The C-terminal region (residues 825–1186, about 30% of the entire protein) of CagA seems to be intrinsically disordered, a fact that could possibly facilitate the interaction with other proteins inside the host cell. In fact, the structure of microtubule affinity-regulating kinase 2 (MARK2) in complex with 120 amino acids of the C-terminal region (residues 885–1005) has been determined, but only a 14-residue peptide of CagA is ordered in the crystal (Nesić et al. 2010). Recently, the complex between domain I of CagA (residues 25–220) with the apoptosis stimulating protein of p53-2 (ASPP2) has been also solved (Nešić et al. 2014), confirming that different parts of CagA interact with different target proteins to hijack host cell-signaling cascades associated with disease outcome.

4 Pathological Function of the cagPAI Type IV Secretion System

Clinical, epidemiological, and functional studies have shown that the presence of cagPAI and CagA is associated with the development of gastric disease. Various animal infection models have been developed and provided comprehensive evidence for the significance of cagPAI and CagA in H. pylori pathogenesis (Ogura et al. 2000; Rieder et al. 2005; Franco et al. 2005; Noto et al. 2013). For example, Mongolian gerbils infected with highly virulent cagPAI-positive H. pylori have been found to develop similar pathology as compared to infected humans. Gerbils developed gastric dysplasia in almost all animals by 4-week infection, which was accompanied by adenocarcinomas in ~25% of gerbils (Franco et al. 2005). After 8 weeks, ~75% of infected animals exhibited gastric adenocarcinomas. Importantly, infection with isogenic mutants indicated that CagA and the T4SS were necessary for gastric cancer development in the gerbil model (Franco et al. 2005; Noto et al. 2013). For more details and other models, please refer to Chap. 10 of this book. In addition to the H. pylori infection model systems, a first direct causal link between CagA and oncogenesis in vivo was reported by the production of transgenic C57BL/6J mice expressing CagA (Ohnishi et al. 2008). After 72 weeks, these transgenic mice exhibited gastric epithelial hyperplasia and some mice developed polyps and adenocarcinomas in the stomach and small intestine. Systemic CagA expression in these mice further induced leukocytosis with IL-3/GM-CSF hypersensitivity, and some animals exhibited myeloid leukemias and B-cell lymphomas (Ohnishi et al. 2008). These findings were supported using two other model organisms, zebrafish and Drosophila (Neal et al. 2013; Reid et al. 2012; Muyskens and Guillemin 2011). Transgenic expression of CagA in these systems exhibited significantly increased rates of intestinal epithelial cell proliferation, upregulation of c-Jun N-terminal kinase (JNK) signaling, and Wnt target genes associated with small cell carcinoma and adenocarcinoma (Neal et al. 2013; Wandler and Guillemin 2012). These experiments demonstrate that H. pylori can trigger the development of gastric adenocarcinoma in gerbils and other model systems in a manner dependent on a functional T4SS and that sole expression of CagA is sufficient to produce severe malignant lesions in transgenic mice. Thus, CagA and the cagPAI T4SS play central roles during H. pylori pathogenesis in vivo.

5 Phosphorylation-Dependent Host Cell Signaling of Translocated CagA

CagA represents a prime example of tyrosine phosphorylatable effector proteins ( CagAPY) of bacteria. Site-directed mutagenesis and mass spectrometry revealed numerous phosphorylation sites in CagA known as the Glu-Pro-Ile-Tyr-Ala (EPIYA) motifs A, B, C, and/or D (Hatakeyama 2003; Yamaoka 2010; Backert et al. 2010). The host tyrosine kinases active on these EPIYA motifs were identified as members of the c-Src and c-Abl families (Wessler and Backert 2008). The resulting phosphotyrosines together with some flanking residues commonly act as recognition motifs for eukaryotic signaling factors. They recruit in particular cellular binding partners that contain SH2 (Src homology 2) domains, but not PTB (phosphotyrosine binding) domains and thereby target and subvert eukaryotic signal transduction pathways in ways that benefit the pathogen (Selbach et al. 2009). This indicates that CagA was specifically designed during evolution to target SH2 domain containing host cell factors (Fig. 4.2a). Altogether, 12 phospho-dependent interaction partners have been identified over the years (Table 4.1). The first reported interaction partner of CagAPY was the tyrosine phosphatase SHP-2 (Higashi et al. 2002). Since then, nine other host cell factors were also found to interact with CagA in a phosphorylation-dependent fashion: the tyrosine phosphatases SHP-1; phosphoinositide-3-kinase (PI3K); the signaling adaptor proteins Crk, Grb2, and Grb7; the tyrosine kinases Csk, c-Src, and c-Abl; as well as the Ras GTPase-activating protein Ras-GAP (Tsutsumi et al. 2003; Suzuki et al. 2005; Tammer et al. 2007; Selbach et al. 2009; Zhang et al. 2015). Thus, CagAPY seems to mimic a tyrosine-phosphorylated host cell protein and therefore appears to function as a master key or picklock to hijack specific cascades of the host. The various CagAPY-SH2 domain interactions appear to have complex roles in H. pylori-triggered cytoskeletal rearrangements, cell scattering, and elongation (summarized in Fig. 4.2a).

Fig. 4.2
figure 2

Model for the role of H. pylori CagA in host cell-signaling processes which may affect pathogenesis. CagA phosphorylation-dependent (a) and phosphorylation-independent (b) signal transduction events are shown. CagA is translocated across the host cell membrane of infected gastric epithelial cells which requires integrin β1 and phosphatidylserine. The tyrosine kinases c-Src and c-Abl phosphorylate delivered CagA. CagA can then modulate various signaling cascades associated with cell polarity, cell proliferation, actin-cytoskeletal rearrangements, cell elongation, disruption of tight and adherens junctions, pro-inflammatory responses, and suppression of apoptosis, as depicted. Black arrows indicate activated signaling pathways and red arrows correspond to inactivated cascades. (For more details, see text. Panels a and b were updated from Backert and coworkers (2010) with kind permission from Wiley.)

Full size image
Table 4.1 Host cell proteins described to interact with phosphorylated CagA and proposed roles in H. pylori infections
Full size table

Infection of AGS gastric epithelial cells with H. pylori in vitro results in migration of the cells and an elongated morphology, known as the “hummingbird phenotype” (Segal et al. 1999). This phenotype requires host cell motility by a yet unknown T4SS factor (Churin et al. 2003), while cell elongation is clearly induced by CagAPY (Backert et al. 2001). Transfection experiments have shown that CagAPY-SHP-2 interaction activates the phosphatase activity of SHP-2, which contributes to AGS cell elongation by stimulating the Rap1 → B-Raf → Erk signaling pathway (Higashi et al. 2004). It was also shown that the CagAPY-SHP-2 complex suppresses the activation of epidermal growth factor receptor (EGFR) and downstream signaling, which is associated with enhanced H. pylori survival (Bauer et al. 2012). The authors found that an antimicrobial peptide, human β-defensin 3 (hBD3), which is highly active against H. pylori, is downregulated by EGFR suppression. These findings revealed a novel mechanism how T4SS-positive strains make use of CagA to evade a key innate mucosal defense pathway to support persistent H. pylori infection (Bauer et al. 2012).

Additional reports demonstrated that the elongation phenotype also requires tyrosine dephosphorylation of three well-known actin-binding proteins—vinculin, ezrin, and cortactin (Selbach et al. 2003, 2004; Moese et al. 2007). The phosphatase involved in this scenario is not SHP-2 and still remains to be identified. Instead it was shown that CagAPY can hamper c-Src activity in two ways, by direct binding of both proteins to each other and by interaction of CagAPY with Csk, a negative regulator of c-Src (Tsutsumi et al. 2003; Selbach et al. 2003). Since c-Src is the first kinase phosphorylating CagA, inactivation of c-Src by CagAPY forms a typical negative feedback-loop mechanism, controlling the amount of intracellular CagAPY. Interestingly, vinculin, ezrin, and cortactin are also targets of c-Src, and c-Src inactivation by CagAPY causes the tyrosine dephosphorylation of these factors (Tegtmeyer and Backert 2011). Moreover, binding of CagAPY to PI3K and/or CrkII is involved in activating the small Rho family GTPase members Rac1 and Cdc42 (Suzuki et al. 2005; Selbach et al. 2009), while binding of CagAPY to SHP-2 or Grb2 can stimulate pro-inflammatory and proliferative responses through the mitogen-activated protein (MAP) kinase cascades (Fig. 4.2a). Finally, CagAPY can also interact with SHP-1, Grb7, and Ras-GAP with yet unknown consequences for the host cell (Fig. 4.2a). Taken together, CagAPY can bind to a remarkably high number of host cell factors to activate signaling mediating cell scattering, elongation, and probably other phenotypes.

Translocation and phosphorylation of CagA appear also important for the interplay of H. pylori with macrophages. Heme oxygenase (HO-1), an anti-inflammatory enzyme, is released by macrophages in response to CagA phosphorylation in H. pylori-infected human patients and mice, while blocking of phagocytosis prevented CagAPY and HO-1 induction (Gobert et al. 2014). Genetic ablation of the hmox-1 gene in mice resulted in increased gastritis, which was associated with enhanced M1/Th1/Th17 responses, reduced regulatory macrophage response, as well as lower H. pylori colonization. These findings provide a mechanism by which H. pylori manipulate immune responses, supporting its own survival by induction of macrophage HO-1 (Gobert et al. 2014).

6 Phosphorylation-Independent Signaling of CagA

Early microarray studies of H. pylori- infected T84 cells showed that expression of 670 host genes changed and 479 of these genes occurred independent of the phosphorylation state of the CagA protein (El-Etr et al. 2004). Thus, not all the interactions of intracellular CagA require its tyrosine phosphorylation. Subsequently, 12 cellular binding partners of non-phosphorylated CagA have been reported (Table 4.2). Non-phospho CagA interactions have been found to induce loss of cell polarity, mitogenic responses, and pro-inflammatory signaling (Fig. 4.2b). The first interaction partner of non-phosphorylated CagA described has been the adapter protein Grb2 (Mimuro et al. 2002). Interestingly, Grb2 is the only binding factor that has been described to interact with both non-phosphorylated and phosphorylated CagA (Mimuro et al. 2002; Selbach et al. 2009). In particular, non-phosphorylated CagA was shown to utilize Grb2 for recruiting Grb2-associated Sos (son of sevenless), a guanine-exchange factor (GEF) of the small GTPase Ras, to the plasma membrane (Fig. 4.2b). This CagA/Grb2/Sos complex triggers Ras-GTP production, which in turn activates the Raf → MEK → ERK signaling pathway contributing to cell scattering (Mimuro et al. 2002) and stimulation of nuclear responses mediating cell proliferation and transcription of the anti-apoptotic myeloid cell leukemia sequence-1 (MCL-1) protein (Mimuro et al. 2007). CagA was also described to function as a mimetic of the eukaryotic Grb2-associated binder (Gab) adaptor protein in transgenic Drosophila which feeds into the same MAP kinase signaling pathway (Hatakeyama 2003; Botham et al. 2008). CagA-triggered MAP kinase activation can stimulate the transcription factor NF-κB mediating the onset of multiple target genes such as IL-8 (Brandt et al. 2005). In addition, it was shown that CagA targets various tumor suppressors such as p53 and RUNX3. CagA-positive H. pylori strains more strongly suppressed p53 as compared with low-risk strains during infection in vivo and in vitro. Degradation of p53 protein was shown to be induced by CagA-mediated signaling via host-specific E3 ubiquitin ligases, thus contributing in tumorigenicity (Wei et al. 2015). In addition, non-phosphorylated CagA can bind to RUNX3, which is often inactivated in gastric cancer. Interaction with RUNX3 proceeds by a WW domain located at the amino-terminus of CagA (Tsang et al. 2010). Importantly, CagA induces the ubiquitination and subsequent degradation of RUNX3, in this way turning off the transcriptional activity of RUNX3 (Fig. 4.2b). These data together suggest the presence of distinct EPIYA-independent domains within CagA, which have crucial functions in protein targeting and modification of host cell transcription.

Table 4.2 Host cell proteins described to interact with non-phosphorylated CagA and proposed roles in H. pylori infections
Full size table

Another remarkable outcome of phosphorylation-independent CagA activities in polarized epithelial cells is the disturbance of the cell-to-cell integrity (Fig. 4.2b). In particular, the proper architecture of the gastric epithelium is controlled by highly organized tight and adherent junctions (Wessler and Backert 2008). Infection and transfection experiments showed that CagA interferes with these intercellular junctions in multiple ways. For example, CagA associates with ZO-1 (zona occludens-1), a tight-junction scaffolding protein and the transmembrane protein JAM (junctional adhesion molecule), causing an ectopic assembly of tight-junction components at sites of bacterial attachment (Amieva et al. 2003). The non-phosphorylated form of CagA has also been reported to bind to the cell-cell junctional transmembrane protein E-cadherin (Murata-Kamiya et al. 2007). Later on, immunoprecipitation studies showed that CagA forms a complex with c-Met recruiting E-cadherin and the armadillo-domain protein p120 catenin, suggesting that binding of CagA to E-cadherin is probably not direct (Oliveira et al. 2009). However, there is much debate going on whether or not the 135-kDa c-Met receptor is phosphorylated and activated upon infection with H. pylori (Snider and Cardelli 2009). Thus, the function of c-Met signaling during H. pylori infection is not fully clarified and should be investigated more thoroughly in future. Some controversy also exists whether CagA can disrupt the E-cadherin complex associated with the release of β-catenin, which has been proposed for transfected CagA or H. pylori-infected AGS cells (Franco et al. 2005; Murata-Kamiya et al. 2007). It has been noted by some authors that AGS cells do not express E-cadherin and exhibit abnormal β-catenin allocation, making them not suitable for the study of related signaling (Oliveira et al. 2009). Considering this fact, it was shown, using MDCK cells expressing wild-type E-cadherin and β-catenin without any mutations, that H. pylori-induced β-catenin signal transduction proceeds independently of CagA during infection (Sokolova et al. 2008). Similarly, some controversy also exists with regard to the proposed interaction of CagA with transforming growth factor beta-activated kinase 1 (TAK1) (Lamb et al. 2009; Sokolova et al. 2014) (Table 4.2).

However, the importance of CagA in inducing the loss of cell polarity is much clearer. The host kinase Par1b (partitioning-defective 1), also called MARK2 (microtubule affinity-regulating kinase), is a central regulator of cell polarity and was reported to have a crucial impact on H. pylori-induced signal transduction. Non-phosphorylated CagA can directly bind Par1b that results in the inhibition of its kinase activity, triggering the loss of cell polarity (Saadat et al. 2007; Nesić et al. 2010). Furthermore, more recent studies showed that CagA not only binds to Par1b but also to other members of this kinase family (Par1a, Par1c, and Par1d) and that these interactions contribute to the H. pylori-triggered AGS cell elongation phenotype (Lu et al. 2009). Recent data also suggest that CagA can also interact with glycogen synthase kinase 3 (GSK-3) and acts as a pathogenic scaffold protein that induces a Snail-mediated epithelial-mesenchymal transition via the depletion of GSK-3 activity (Lee et al. 2014). Taken together, these results suggest that transfected CagA can interfere with Par1 members, c-Met, GSK-3, and E-cadherin signaling and may also activate NF-κB, thereby contributing to H. pylori-induced pro-inflammatory responses. Finally, there are two more reported binding partners of CagA, α-Pix (Baek et al. 2007), and integrin β1 (Jiménez-Soto et al. 2009; Kaplan-Türköz et al. 2012). While the interaction with α-Pix has a proposed role in inflammation, CagA-integrin β1 interaction may be involved in delivery of CagA into the host cell (Fig. 4.2b). Importantly, the downstream pathways of CagA emerged to be highly diverse and possible cross talk among them and other bacterial factors need to be dissected in more detail.

7 T4SS-Dependent but CagA-Independent Cellular Signaling Induced by H. pylori

In this section, various T4SS-dependent but CagA-independent signaling events induced by H. pylori will be discussed (Fig. 4.3). Early studies indicated that H. pylori can actively inhibit its own uptake and killing by professional phagocytes (Ramarao et al. 2000). This anti-phagocytic effect was dependent on vital bacteria expressing the T4SS, because various isogenic virB mutants blocked this phenotype (Ramarao et al. 2000). Interestingly, the actual factor involved was not CagA, because isogenic ΔcagA mutants also abrogated phagocytosis. These experiments indicated that H. pylori express a yet unknown T4SS effector with anti-phagocytic capability that may play a crucial function in the immune escape of this persistent pathogen (Fig. 4.3). Infection of bone-marrow-derived macrophages by H. pylori-induced pathology via microRNAs, such as miR-155, as important regulators of inflammatory and innate immune responses. Increase of miR155 expression was T4SS-dependent but CagA-independent and resulted in reduced macrophage apoptosis (Koch et al. 2012). However, most of the studies were performed to investigate the interaction of H. pylori with gastric epithelial cells. Phosphoproteomics of epithelial cells infected with H. pylori revealed the induction of multiple tyrosine-phosphorylated proteins. The majority of enriched phospho-peptides were from kinases of the MAPK family and the use of isogenic mutants showed that both CagA and the T4SS are key regulators of tyrosine phosphorylation events (Glowinski et al. 2014). In addition, histone H3 phosphorylation was found to be altered by a T4SS-dependent but CagA-independent pathway (Fig. 4.3). Infection with cagPAI-positive H. pylori strains decreased H3 phosphorylation levels at two phosphorylation sites, serine residue 10 and threonine residue 3 (Fehri et al. 2009; Ding et al. 2010). It appeared that mitotic histone H3 kinases such as Aurora B and vaccinia-related kinase 1 (VRK1) were not fully activated in H. pylori-infected cells, leading to a transient pre-mitotic cell cycle arrest (Fehri et al. 2009). Together, these data demonstrate that H. pylori subvert cellular key processes, including cell cycle progression, by a not yet identified T4SS effector. Furthermore, the results of numerous reports revealed that other components of the T4SS, but not CagA itself, were necessary for the induction of pro-inflammatory signaling, including the activation of transcription factors NF-κB and AP-1 (Fig. 4.3). This implicated that the T4SS might deliver effectors in addition to CagA or that the T4SS itself triggers the effect. Despite systematic mutagenesis of all cagPAI genes and other efforts, the hypothetical additional effector remained unknown for many years. One proposed candidate was bacterial peptidoglycan, because it can be identified by Nod1, an intracellular pathogen-recognition molecule (Viala et al. 2004). These results implicated that T4SS-dependent delivery of peptidoglycan is responsible for activation of Nod1 → NF-κB-dependent pro-inflammatory responses such as secretion of IL-8 (Viala et al. 2004). However, the actual bacterial T4SS factor(s) and pathways that activate both transcription factors, NF-κB and AP-1, are highly controversial in the literature and still not fully defined (Backert and Naumann 2010). Remarkably, T4SS-positive H. pylori can activate the NF-κB-dependent induction of a DNA-editing enzyme (AID) in gastric epithelial cells, that leads to the accumulation of mutations in tumor suppressor p53 (Matsumoto et al. 2007). Thus, induction of AID by H. pylori infection might be a mechanism whereby gastric carcinogenesis-related gene mutations accumulate.

Fig. 4.3
figure 3

Model for the role of H. pylori T4SS-dependent but CagA-independent host cell-signaling processes which may affect pathogenesis. The multitude of known T4SS-dependent but CagA-independent pathways involve in the activation of receptor and non-receptor tyrosine kinases, pro-inflammatory signaling, Rho GTPase activation, scattering and motility of gastric epithelial cells, as well as suppression of histone phosphorylation and H. pylori phagocytosis by immune cells. Two particular T4SS factors have been reported to be involved in some but not all of these responses. The known signaling functions for injected peptidoglycan as well as pilus-exposed or recombinant CagL are shown. For numerous other pathways, the actual T4SS factor is yet unknown as also indicated. (For more details see text. This figure was updated from Backert and coworkers (2010) with kind permission from Wiley.)

Full size image

Infection of gastric epithelial cells with H. pylori was also reported to profoundly activate various receptor tyrosine kinases (RTKs) in a T4SS-dependent manner including EGFR (Keates et al. 2001; Churin et al. 2003), hepatocyte growth factor receptor c-Met, and Her2/Neu (Churin et al. 2003). Studies on the downstream signaling indicated that each of these RTKs can activate the MAP kinase members MEK and ERK1/2 (Fig. 4.3). However, while induction of EGFR has been shown to induce pro-inflammatory responses leading to the secretion of IL-8 (Keates et al. 2001), activation of c-Met (but not EGFR or Her2/Neu) was involved in cell scattering and motogenic responses of infected gastric epithelial cells (Churin et al. 2003). At later time points of infection, EGFR can be inactivated by CagA as mentioned above (Bauer et al. 2012). Interestingly, the small Rho GTPases Rac1 and Cdc42 and the non-receptor tyrosine kinase c-Abl are also activated by a T4SS-dependent but CagA-independent process and play a role in stimulating scattering and motility of infected gastric epithelial cells (Fig. 4.3). However, the actual T4SS factor involved for many of the above events is still unknown.

In vitro studies showed a profound role of recombinant CagL in activating the host tyrosine kinases EGFR, ErbB3/Her3, FAK, and c-Src (Tegtmeyer et al. 2010). Investigation on the molecular mechanism of EGFR activation by CagL has demonstrated the involvement of ADAM17, a metalloprotease implicated in catalyzing ectodomain shedding of receptor tyrosine kinase ligands. In non-stimulated cells, the inactive form of ADAM17 forms a complex with integrin α5β1 (Fig. 4.3). During acute H. pylori infection, however, it was demonstrated that CagL binding to integrin α5β1 activates ADAM17 by dissociating ADAM17 from the complex (Saha et al. 2010). In addition, CagL immobilized on petri dishes binds host cells and, thus, mimics human fibronectin (Tegtmeyer et al. 2010). Fibronectin is a 250-kDa protein containing an RGD motif that plays crucial roles in promoting cell adhesion, migration, and intracellular signaling. It was shown that purified CagL alone can directly trigger intracellular signaling pathways upon contact with mammalian cells and can even complement the spreading defect of fibronectin-/- knockout cells in vitro (Tegtmeyer et al. 2010). Treatment of AGS cells with purified CagL was also demonstrated to be sufficient to result in IL-8 induction, which required the RGD motif in CagL and activation of integrin α5β1 (Gorrell et al. 2012). Using different wild-type or ΔcagL mutant strains showed that IL-8 induction occurred independently of CagA translocation. Further studies revealed another surface-exposed Phe-Glu-Ala-Asn-Glu (FEANE) interaction motif located close to the RGD site. This site enhanced the interaction of CagL with integrin α5β1 supporting CagA translocation and was referred to as RGD helper sequence, RHS (Conradi et al. 2012). CagL was also shown to be related to increased gastrin expression resulting in hypergastrinemia, a major risk factor for gastric adenocarcinoma formation. Gastric epithelial cells stably transfected with a human gastrin promoter luciferase construct increased the promoter activation of gastrin via integrin-linked kinase (ILK) and integrin αVβ5 (Wiedemann et al. 2012). Another role of CagL was seen in bone-marrow derived DCs, where it resulted in induction of pro-IL-1β and formation of mature IL-1β and NLRP3 (NOD-like receptor pyrin domain-containing 3) induction in response to H. pylori infection. This activity was mediated via different host innate immune receptors including Toll-like receptor 2 (TLR2) and nucleotide-binding oligomerization domain 2 ( NOD2) (Kim et al. 2013). CagL and the activity of MAP kinases were also important for H. pylori-mediated regulation of eosinophil migration (Nagy et al. 2011). The results indicate that H. pylori increases production of the chemokines CCL2, CCL5, and granulocyte-macrophage colony-stimulating factor (GMC-SF) by gastric epithelial cells and that these molecules induce eosinophil migration. Interestingly, CagL sequence analyses revealed that isolates from the gastric cancer patients had a higher rate of amino acid sequence polymorphisms—Y58 and E59—than those of the non-gastric cancer patients (Yeh et al. 2011). The CagL Y58E59 polymorphism increased risk of gastric cancer up to 4.6-fold and infected patients had higher integrin α5β1 expression than noninfected patients. Furthermore, CagL-Y58E59 H. pylori infection predisposed an upward shift in integrin α5β1 in the corpus, leading to more severe corpus chronic inflammation (Yeh et al. 2011). However, expression of isogenic CagL Y58/E59 variants in H. pylori strain 26695 significantly blocked translocation and phosphorylation of CagA as compared to complemented wild-type CagL (Tegtmeyer et al. 2014). The involved signaling should be studied in detail in future studies.

8 Conclusions

H. pylori represents a highly successful human pathogen, which can trigger severe clinical symptoms in a small subset of patients. The investigation of bacteria-host interactions and virulence factors such as CagA and the T4SS has provided us with crucial insights in mechanisms leading to H. pylori pathogenesis. A list of more than 20 known cellular interaction partners of CagA is quite amazing for a bacterial effector protein. The current model suggests that CagA mimics a eukaryotic signaling factor either located in a large multiprotein complex or simultaneously in various subcellular areas of infected host cells. The large variety of binding partners also reflects the integrated network of complex signal transduction pathways in target cells, which may have important impact on the multi-step pathogenesis of H. pylori. In the future, it will be important to search for additional translocated effector molecules of the cagPAI T4SS and various other T4SSs present in the H. pylori chromosome (Backert et al. 2015). Finally, the importance of CagA for H. pylori itself is also not yet clear. Using a polarized epithelium model system, ΔcagA mutants were shown to be defective in cell surface colonization, but exogenous addition of iron to the apical medium partially rescues this defect, suggesting that one of CagA’s effects on host cells is to facilitate iron acquisition from the host via a mechanism involving the transferrin receptor (Tan et al. 2011). To test whether CagA is important in promoting iron acquisition in vivo, the colonization of H. pylori in iron-replete vs. iron-deficient Mongolian gerbils was carried out. While wild-type H. pylori and ΔcagA mutants colonized iron-replete gerbils at similar levels, ΔcagA mutants are markedly impaired in colonizing iron-deficient gerbils (Tan et al. 2011). Iron depletion accelerated the development of H. pylori-induced premalignant and malignant lesions in a CagA-dependent manner (Noto et al. 2013). H. pylori strains harvested from iron-depleted gerbils or grown under iron-limiting conditions exhibited enhanced virulence and induction of inflammatory factors. Future studies should investigate the molecular basis of this important disease-associated phenomenon.