Helicobacter pylori : cellular interactions and pathogenesis

Abstract

Helicobacter pylori colonizes the stomachs of about half of die world's population. It inevitably causes an inflammatory response in all its hosts, but in a subset of individuals the infection leads to severe gastric diseases such as peptic ulcers, gastric cancer or MALT lymphoma.

The relative contributions of the host, microbe and environment to disease progression are not known. Host factors such as HLA-type, secretor status and Il-1beta allele type have been identified as contributing to the risk of disease. Microbial factors that have been recognized as virulence factors are, e.g., the vacuolating cytotoxin, and the Cag Pathogenicity Island (PAI) that encodes a type IV secretion system. The route of transmission of H. pylori is not known. Once colonization is established, it persists for life unless eradicated with antibiotics. Resistance to antibiotics in H. pylori is increasing; especially resistance to clarithromycin results in treatment failure. The genetic diversity of H. pylori strains is remarkable and it reflects a panmictic population structure with free recombination. This study aimed to characterize defined host, microbial and environmental factors that might impact on the outcome of the infection. These factors were: (i) production of specific receptors (Leb and NetiAcalpha2,3Galbeta1,4 glycans) and presence or absence of gastric acid; (ii) bacterial genotype; and (iii) presence or absence of a microflora competing with H. pylori colonization. The host factors were studied in genetically modified, transgenic, FVB/N mice. Bacterial genotypes were determined by microarray in clinical isolates obtained from patients with gastric adenocarcinoma and controls. The effect of the microflora was assessed by comparing germ-free and conventionally raised animals.

After whole-genome genotyping of two subclones of one strain of H. pylori, we found that the isolates were more similar to each other than to any other strain that had been genotyped, but had undergone considerable divergence including excision of the Cag PAI. Both isolates colonized germfree Leb producing mice to the same density. The Cag PAI+ isolate colonized conventionally raised animals to the same extent as germ-free, while the Cag PAI- isolate was unable to colonize conventionally raised animals. Evolution of the isolates in ex-germ-free and conventional mice after three or ten month infections was limited.

Among the factors that affect the appearance and spread of acquired antibiotic resistance, the mutation frequency and biological cost of resistance are of special importance. We used clinical pairs of H. pylori isolated before and after treatment with clarithromycin. The mutation frequency of a panel of H. pylori isolates was found to be higher than in enteric bacteria, (median 10-6). 1/4 of the isolates had a mutation frequency higher than Enterobacteriaceae mismatch repair defective mutants. Clarithromycin resistance was associated with a biological cost, as measured by decreased competitive ability of the resistant mutants compared to wildtype in colonization of mice. In clinical isolates, this cost could be reduced by compensatory mutations at extra-genie site(s), indicating that compensation is a clinically relevant phenomenon that could act to stabilize resistant bacteria in a population.

We also studied the influence of specific bacterial binding to Le Leb and sialylated glycoconjugates, as well as the presence and absence of acid on the fecal-oral transmission of H. pylori between gnotobiotic transgenic mice. Transmission only occurred in transgenic mice without parietal cells, indicating that loss of acid production in the host facilitates transmission of H. pylori.

Genotypes were compared among 13 strains that exhibited the ability to bind to Le b and/or NeuAcalpha2,3GaIbeta1,4 glycans and two strains that did not bind. Strains were selected from a panel of 96 strains from a case-control study of gastric cancer. Colonization ability and host response to the gentyped isolates were assessed in transgenic mice. H. pylori-induced expression of three genes were analyzed by quantitative real-time RT- PCR: heat shock protein 70, polymeric immunoglobulin receptor and interleukin-10. These markers were differentially induced (compared to uninfected controls) by the different isolates. We found that genes encoding specificity subunits of type 1 restriction modification systems (hsdS) in H. pylori were correlated with induction of a strong host response. The HsdS proteins are DNA binding proteins that confer sequence specificity for the type I R- M enzyme. We suggest that these regulate virulence gene expression.

Our work describes part of the interplay between H. pylori and its host that lead to persistent infection. In addition, we have identified host genes that can potentially be used as molecular markers of a robust host response as well as genes in the bacterium that might regulate virulence of the isolate. This could help identify patients at risk for severe gastric disease due to H. pylori infection, to avoid massive treatment programs that could affect the total commensal and pathogenic microflora.