Genome evolution and virulence in H. pylori: Identifying the genes/alleles underlying phenotype variation

An estimated 50% of all people carry the stomach bacterium Helicobacter pylori (H. pylori). This organism is responsible for gastric problems like gastritis and gastric ulcers, and is one of the major causes of gastric cancer worldwide. Large numbers of people carry this organism asymptomatically and many questions remain about why serious symptoms develop in a subset of infected humans.These extremely recombinant bacteria may take different evolutionary trajectories in different people, and some genomic changes may be associated with gastric cancer. To test this, and learn more about the genetics of cancer-associated H. pylori, different approaches were used.First, evolution of H. pylori populations was investigated looking at both core and accessory genomes and revealed traces of the long and complex history of the Americas in the bacterial genomes, as well as a similar evolution in core and accessory genome. This was the first time accessory genome of H. pylori was studied that way. Secondly, evolution occurring in the bacterial genome during colonisation of a single host was studied in mice model. This analysis revealed small changes during the passage from a human host to a mice host, and during the long-term colonisation of mice stomach. Then a Genome Wide Association Study (GWAS) approach was applied to a large isolate collection sampled across Europe comprising strains isolated from cancer patients and strains from asymptomatic or gastritis-suffering patients. This approach identified 11 polymorphisms in 9 genes (3 cagPAI genes, babA, hpaA, 1 outer membrane protein coding gene HP1055 and 3 other core genes (HP0747, HP0709 and HP0468) associated with cancer and a preliminary risk score was built to identify high risk strains. Finally, variations observed among clinical isolates of H. pylori from European patients with different pathologies in terms of motility and ability to trigger cytokine production in two types of cells were quantified. Motility variations were not associated with the disease type, but a link was observed for cytokine production. This was compared to genomic variations, confirming the role of known genomic factors such as cagPAI genes and sheding light to possible functions of a number of new genes