Establishment and analysis of a mouse model conditionally expressing the Helicobacter pylori CagA oncoprotein

概要

Gastric cancer is the third leading cause of cancer related deaths, accounting for over 800,000 deaths worldwide. Over 50% of the total number of gastric cancer cases are in East Asian countries such as Japan, Korea, and China. The strongest risk factor for the development of gastric cancer is infection with the gastric pathogen, Helicobacter pylori (H. pylori). H. pylori is subdivided into cagA-positive and cagAnegative strains, where almost all the isolated H. pylori strains are cagA-positive in East Asian countries. Infection with cagA-positive H. pylori increases the risk of gastric cancer by over ten-folds compared to infection with cagA-negative strains. In 2014, the IARC announced that 80% of all stomach cancers are caused by H. pylori infection, which may be grossly underestimated in East Asian countries. When infected by cagA-positive H. pylori, the cagA-encoded CagA protein is delivered into gastric epithelial cells via the bacterial type IV secretion system (T4SS), where it localizes to the inner leaflet of the plasma membrane. There, CagA undergoes tyrosine phosphorylation by host protein tyrosine kinases (e.g. c-Src, c-Abl) specifically on the C-terminal EPIYA-repeat region of CagA. Tyrosine-phosphorylated CagA then binds to SHP2 protein tyrosine phosphatase, a bona fide oncoprotein, which aberrantly activates SHP2, resulting in full activation of Ras-Erk signaling. In a phosphorylation-independent manner, the CM motif binds to and inhibits partitioning-defective 1 (PAR1) serine/threonine kinase, which disrupts tight junctions and causes the loss of apical-basal polarity in polarized epithelial cells. Loss of epithelial apical-basal polarity by CagA contributes to neoplastic transformation of cells, as normal epithelial polarity constrains abnormal cell proliferation. The purpose of this study is to establish a cagA mice model which can inducibly and sitespecifically express high levels of CagA in the gastric epithelium, as observed in human gastric cancer cases. By using this model, we aim to determine the acute and long-term impacts of CagA expression in the gastric epithelium, which will help shed light on the mechanism for which CagA protein initiates and perpetuates gastric carcinogenesis in vivo.

The newly established mouse model expressing H. pylori CagA oncoprotein, CAG-cagA mouse, was designed by placing the 3xFlag tagged cagA gene (cagA) downstream of the cytomegalovirus early enhancer element fused chicken-beta actin (CAG) promotor, for high ubiquitous expression of the CagA protein. A transcriptional stop sequence (polyA) was placed between two loxP sequences (LSL cassette) and was inserted prior to the cagA gene, thereby regulating CagA expression via the cre-loxP system. This transgene was targeted to the Rosa26 locus as the Rosa26 locus has been reported to be a safe integration site which allows for strong and predictable expression of a single transgenes carrying an exogenous promoter. In the absence of cre recombinase, there is no expression from the established target allele. After recombination by cre-recombinase, the LSL cassette is excised and CagA is expressed. By utilizing ubiquitous and tissue-specific promoters for cre and creER recombinase, we examined the impacts of CagA on systemic and gastric specific expression in mice.

To examine the impacts of systemic CagA expression, CAG-cagA mice were mated with CAGcreER mice. When CagA expression was induced during embryogenesis, CAG-cagA; CAG-creER embryos at E13.5 displayed stunted growth compared to control littermates, indicating CagA expression perturbed embryogenesis, which induced embryonic lethality and subsequent resorption of embryos. The results suggest that deregulation and aberrant activation of SHP2 by CagA in this mouse model is not tolerated during embryonic development in mice, ultimately leading to embryonic lethality. In adult mice, expression of CagA resulted in overall weakness in mice characterized by hunched posture, lack of movement, ruffled fur, and squinted eyes. At 1 week post-induction, CAG-cagA; CAG-creER mice displayed roughly 20% bodyweight loss compared to weight prior to injection. Relatively no change in weight was observed in CAG-cagA control mice. Examination of stomach tissue revealed large disruptions to the glandular structure where cells displaying abnormal mitosis was observed. Further evaluation of the stomach revealed drastic reduction of overall mucin in CAG-cagA; CAG-creER mice due to the loss of mucus neck cells. Furthermore, drastic reduction of parietal cells which was accompanied by Spasmolytic Polypeptideexpressing Metaplasia (SPEM) in CAG-cagA; CAG-creER mice. Regardless of the drastic loss of gastric cell populations, apoptosis was not observed in the gastric epithelium. Instead, severe DNA damage was observed in the presence of CagA in the gastric epithelium, which may have resulted in the inhibition of cell turnover, as p21 accumulation was observed in response to CagA-induced DNA damage. Never-theless, CagA expression in other major organs such as the liver, kidney, spleen, heart, and lung displayed no significant histological differences to that of CAG-cagA control mice. Additionally, no drastic changes which could explain sudden death in mice were observed in CD45+ myeloid and lymphoid cells populations, suggesting that the death of mice was unlikely to be caused by changes to the immune system. Taken together, these observations suggest that the phenotypes observed in mice were primarily caused by CagA expression in the stomach.

To determine the impacts of long term CagA expression in the gastric epithelium, CagA was expressed specifically in the stomach by crossing mice with gastric specific promoter driven-creER mice. Prolonged expression of CagA specifically in the gastric epithelium of mice resulted in the development of gastric hyperplasia (1/30) and gastric adenocarcinoma (1/30) in the absence of inflammation and immune cell response. In both cases, tumors were localized above the isthmus region, suggesting that CagA may impact the function of stem cell populations in the isthmus region. Furthermore, p53 expression was significantly attenuated in both gastric hyperplasia and adenocarcinoma. To determine the impacts of CagA expression in stem cell populations and the role of p53 loss in CagA-induced gastric cancer development, gastric organoids were established from CAG-cagA; CAG-creER mice and wild type (WT) control littermates.

Expression of CagA in gastric organoids perturbed growth and induced accumulation of debris in the inner lumen of the organoids. The CAG-cagA; CAG-creER organoids were significantly smaller in size (diameter) than control organoids. Analysis by MTS assay, which evaluates mitochondrial activity and growth rate, showed overall decreased growth of organoids expressing CagA signifying that CagA can perturb organoid development and maintenance. Knockdown of p53 in gastric organoids rescued CagAinduced growth inhibition. Taken together, these results indicate that in the presence of p53, CagA expression perturbs gastric stem cell growth and maintenance by inducing DNA-damage associated p21 accumulation and growth arrest. However, after loss of p53, cells are rescued from CagA-induced growth arrest. This suggests that CagA induces DNA damage while inhibiting DNA damage-associated apoptosis, thereby perpetuating the accumulation of mutations in the stem cell population. Mutation in tumor suppressor genes such as p53 may be a key event which potentiates CagA-dependent carcinogenesis.

In this study, we demonstrated in vivo that gastric tissue specific expression of CagA induces gastric cancer through induction of DNA damage in stem cell and progenitor cells in the isthmus region, which through accumulation of mutations, are transformed into cancer cells. The results of this study suggest that attenuation or loss of p53 plays an important role in CagA induced gastric carcinogenesis. Furthermore, while cagA-positive H. pylori is most notably associated with the development of different gastro-intestinal diseases, H. pylori infection has been shown to be associated with extra-gastric diseases including but not limited to ischemic heart disease, anemia, insulin resistance, mucosa-associated lymphoid tissue (MALT) lymphoma, and type 2 diabetes. cagA-positive H. pylori hijack host exosomes to secrete exosomes-enriched with CagA protein into circulation. The newly established CAG-cagA mice model will be an essential tool to elucidating the function of CagA in both gastric cancer and cagA-positive H. pylori associated extra-gastric diseases.