Helicobacter pylori bacterial diversity and human disease : microbiological and epidemiological studies with special reference to gastric cancer

Abstract

The aims of this thesis were to explore if the coccoid form of H. pylori is a degenerative, dead form of the bacteria. We also wanted to study if multiple strain infections of H. pylori existed in Swedish subjects, if the strain variations were connected to certain disease groups, and if specific protein markers could be found in H. pylori strains isolated from patients with different gastric diseases. Further, we wanted to detect specific risk modifiers in the host as well as in the bacteria that influence the association between H. pylori infection and gastric cancer, and we investigated if all strain types of H. pylori were associated with an increased gastric cancer risk.

Degenerative changes, observed as changes in intracellular composition and surface properties, were analyzed in the coccoid form of H. pylori by several techniques in vitro. The coccoid form was found to be a dead, non-viable form of the microorganism which pose no risk for transmission of infection from the environment. To test the inter- and intrapatient variation of H. pylori strains, genomic fingerprints and cagA status were determined by PCR for a total of 30 clinical isolates from three different disease groups. From each isolate, ten bacterial colonies were tested. The degree of homogeneity within single strains was high, since all ten colonies from each isolate gave identical fingerprints. No multiple strain infections, but an extensive inter patient variation, was observed.

Subclones of H. pylori with different cagA status, indicating differences in virulence, were observed in five subjects. In a patient with multiple strain infection, who underwent clarithromycin eradication treatment, the underrepresented strain was found to be mutated in the 23S rRNA gene, was resistant to the antibiotic, and underwent clonal expansion. This resulted in recurrence of the infection.

To investigate whether H. pylori strains isolated from patients with different gastric diseases express disease specific proteins, two-dimensional gel electrophoresis (2-D PAGE) on 16 strains of H. pylori was performed. No marker proteins were found, but some clustering of strains isolated from the same disease group was observed by dendrogram analysis. The extensive strain variation that H. pylori exhibits, not only on the genome, but also on the protein level, was highlighted.

Bacterial factors, the contribution of the diagnostic methods used for detection of H. pylori infection, and the associated gastric cancer risks were explored in a hospital-based case control study. In total, 72 gastric cancer cases and 324 controls were included at the time of endoscopy. H. pylori status was determined by culture, immunohistochemistry, ELISA and immunoblotting. Strains of H. pylori were tested for presence of the cagA gene and the vacA gene (s1/s2, m1/m2).

Positive antibody reactions to the proteins VacA, CagA, and four other proteins were detected. These results were used to classify H. pylori strains into highly virulent type I (CagA/VacA positive), intermediate, and type 11 with low virulence (CagA/VacA negative). In this study, H. pylori strain type I and the intermediate type, as well as antibodies towards CagA, were associated with increased risk of gastric adenocarcinoma. Subjects with increasing antibody levels and subjects diagnosed with past or present H. pylori infection are also at an increased risk.

In conclusion, H. pylori strains are highly diverse, and strain variations influence the pathogenicity of the infecting strain. Highly virulent strains of H. pylori are associated with an increased gastric cancer risk. Treatment of H. pylori infection could be reserved only for those infected by virulent strains in the future, as our results show that the type of strain determines the nature of disease complicating chronic H. pylori infection.