Perturbation of the non-canonical Wnt signaling pathway by the Helicobacter pylori CagA oncoprotein

概要

Gastric cancer is the fifth most frequent cancer and the third leading cause of cancer deaths worldwide. Also notably, more than half of the world total gastric cancer-related deaths are from East Asia, mainly in China, Japan and Korea. Since its discovery in 1983 as a spiral or slightly curved gram-negative bacillus colonizing the stomach, there are increasing numbers of studies pointing out that H. pylori infection plays a key role in chronic atrophic gastritis and peptic ulcers. Subsequent epidemiological investigation indicated H. pylori infection might be a significant etiological cause of gastric cancer. H. pylori is subdivided into two strains, cagA gene-positive strain and cagA gene-negative strain. Chronic infection of both two strains can cause simple gastritis, but cagA-gene-positive strains causes more severe chronic gastritis, atrophic gastritis, intestinal metaplasia and gastric cancer.

The Helicobacter pylori virulence factor CagA (cytotoxin-associated gene A), encoded by cagA-gene, is an oncoprotein varying in its size from 130 to 145 kDa, because of structural polymorphism in the carboxy-terminal (C-terminal) region. About 40% of Helicobacter pylori strains isolated in Western countries do not contain the cagA gene, i.e., the cagA-negative strains. In contrast, practically all the H. pylori strains isolated in East Asian countries carry the cagA gene, i.e., the cagA-positive strains.

Upon stable attachment to gastric epithelial cells, H. pylori cagA-positive strains initiate delivery of CagA into the cytoplasm of the host cells via TFSS. Delivered CagA then is localized to the inner surface of the cell membrane by physically interacting with the membrane phosphatidylserine (PS) through its N-terminal region.

The CagA proteins are highly variable at the C-terminal regions and can be subdivided into two species, Western CagA and East Asian CagA. The Western CagA proteins, produced by H. pylori strains circulating all over the world other than East Asia, possess EPIYA-A and EPIYA- B segments, followed by variable numbers of EPIYA-C segments (ABC type CagA). The CagA proteins of H. pylori strains isolated in East Asian countries, known as East Asian CagA, possess the same EPIYA-A and EPIYA-B segments as Western CagA, followed by the unique EPIYA-D segment (ABD type CagA). CagA undergoes tyrosine phosphorylation on these EPIYA segments by Src family kinases. The phosphorylated CagA interacts with and thereby deregulates the protein tyrosine phosphatase SHP2, which in turn aberrantly activates Ras-Erk MAP kinase signaling, causing dysregulated cell proliferation. Moreover, CagA also interacts with the polarity-regulating serine/threonine kinase, partitioning defective-1 (PAR1, also known as microtubule affinity regulating kinase [MARK]), through the C-terminal CagA multimerization (CM) motif to inhibit its kinase activity. The CagA-PAR1 interaction leads to the disruption of tight junctions (TJs) and causes a loss of epithelial apical–basal (A-B) polarity. Moreover, it is reported that the PAR1 can perturb the Wnt/PCP pathway.

The planar cell polarity (PCP) is cellular polarity in the axis orthogonal to the apical–basal axis. The planar cell polarity is controlled by the non-canonical Wnt/PCP pathway, which is conserved through animal evolution.

The Wnt/PCP pathway regulates cell fate determination, cell migration, cell polarity, neural patterning and organogenesis during embryonic development. Consistently, perturbation of the Wnt/PCP signaling pathway is related to several disease, such as polycystic kidney disease and neural tube defects (NTDs). Although no study has shown the relationship between the Wnt/PCP pathway and gastric cancer, it is reported that PAR1 is functionally associated with the Wnt/PCP pathway and delivered CagA forms a physical complex with PAR1.

In a previously reported paper, the authors injected the xPAR1 mRNA into Xenopus laevis embryos; they found that the convergent extension movements (CEM) in gastrulation and neurulation was inhibited, and at the tadpole stage the embryos showed a dorsal bending phenotype with a shortened body axis. The convergent extension movements (CEM) is a key process by which tissues undergo narrowing along one axis and concomitant extension along another axis. It has already been indicated that the Wnt/PCP pathway could establish cell polarity to control the CEM of cells. These phenotypes showed that xPAR1 modifies the Wnt/PCP pathway.

CagA can interact with PAR1 via CagA multimerization (CM) motif to inhibit its kinase activity. Therefore, I hypothesized that maybe CagA also can play a role in perturbation of the Wnt/PCP signaling pathway. The Xenopus laevis embryo is used as a precious tool in my experiments to analyze the function of CagA in perturbation of Wnt/PCP pathway.

In this study, I planned to decipher the role of CagA was played in Wnt/PCP pathway, the ectopic expression of CagA-WT, CagA-ΔCM (CM deleted CagA), CagA-N’ (N-terminal region of CagA) and CagA-C’-ΔCM (the C-terminal region of CagA-ΔCM) in Xenopus laevis embryos. Several experiments related to the CEM were performed by the mRNA microinjection using Xenopus laevis embryos, such as the blastopore closure assay, the neural tube closure assay, the analysis of body axis elongation in Tailbud embryo and the animal cap assay.

In the blastopore closure assay, the expression of CagA and its deletion mutants (CagA-ΔCM and CagA-N’) induced the inhibition of the blastopore closure. These results indicated that the N-terminal region of CagA is sufficient to inhibit the blastopore closure during Xenopus laevis gastrulation.

In the neural tube closure assay, embryos expressing CagA and its deletion mutants (CagA-ΔCM and CagA-N’) showed poor neural tube closure. These results indicated that the N- terminal region of CagA is sufficient to inhibit the neural tube closure during Xenopus laevis neurulation.

In the analysis of body axis elongation, embryos expressing CagA and its deletion mutants (CagA-ΔCM and CagA-N’) showed the inhibition of CEM induced anterior-posterior axis formation which is observed as dorsal flexure with a shortened body axis. These results indicated that the N-terminal region of CagA is sufficient to inhibit anterior-posterior body axis elongation during the Xenopus laevis Tailbud.

In animal cap assay, the CagA and its deletion mutants (CagA-ΔCM and CagA-N’) expressed animal caps exhibited the inhibition of Activin A-induced elongation. These results indicated that CagA-N’ inhibits Activin A-induced animal cap elongation.

The results of blastopore closure assay, neural tube closure assay, analysis of body axis elongation assay and animal cap assay suggested that the N-terminal region of CagA inhibits mesodermalization and/or CEM.

To examine the effect of CagA expression in mesodermal specification, the expression of Xbra, a mesoderm-specific transcriptional co-activator, was analyzed by RT-PCR using the CagA expressed animal caps. In this experiment, the Activin A-induced Xbra expression was observed in animal caps injected with CagA and its deletion mutants (CagA-ΔCM, CagA-N’ and CagA-C’-ΔCM), suggesting that these animal caps differentiated into mesodermal tissues. It means that the N-terminal region of CagA did not affect the cell fate specification.

These data obtained from in vivo development and ex vivo mesodermalization of CagA- expressing Xenopus laevis embryos indicated that the N-terminal region of CagA has an ability to inhibit the CEM during Xenopus laevis embryogenesis, suggesting that the N-terminal region of CagA is sufficient to perturb the Wnt/PCP pathway without affecting the cell fate specification.

Furthermore, the results from this study by using HEK 293T cell and mammal epithelial cell AGS disclosed that CagA could interact with the four-transmembrane protein Vangl1/2 which is a core-protein of Wnt/PCP signaling pathway via its N-terminal region. In vitro result of the CagA responsible region is consistent with the in vivo data. The binding domain of Vangl2 was also confirmed. Moreover, this study also determined that CagA inhibits the expression of Vangl2 in gastric cell line.

In conclusion, this study demonstrated the importance of the N-terminal region of H. pylori CagA in the perturbation of the Wnt/PCP signaling pathway. Although, the relationship between the Wnt/PCP signaling pathway and gastric cancer is not well known, this work may dedicate a new insight into the mechanism of cagA-positive H. pylori-mediated gastric lesions that promote gastric carcinogenesis.