Comparative genomics of Helicobacter pylori isolates recovered from ulcer disease patients in England

Abstract

Background: Genomic diversity of H. pylori from many different human populations is largely unknown. We compared genomes of 65 H. pylori strains from Nottingham, England. Molecular analysis was carried out to identify rearrangements within and outside the cag-pathogenicity-island (cag PAI) and DNA sequence divergence in candidate genes. Phylogenetic analysis was carried out based on various high-resolution genotyping techniques.

Results: Analyses of virulence genes (cagT, cagE, cagA, vacA, iceA, oipA and babB) revealed that H. pylori strains from England are genetically distinct from strains obtained from other countries. The toxigenic vacA s1m1 genotype was found to be less common and the plasticity region cluster was found to be disrupted in all the isolates. English isolates showed a predominance of iceA1 alleles and a functional proinflammatory oipA gene. The English H. pylori gene pool revealed several Asian/oriental features. This included the predominance of cagA - glr (cagA right junction) motif types III and II (up to 42%), presence of vacA m1c alleles and phylogenetic affinity towards East Asian / Amerindian gene pools based on fluorescent amplified fragment length polymorphism (FAFLP) analysis and glmM sequence analysis.

Conclusion: Overall, our results demonstrated genetic affinities of H. pylori in England with both European and the Asian gene pools and some distinctive genetic features of virulence genes that may have evolved in this important European population.

Figure 1

Diagram showing the percent distribution of different genetic loci in the English isolates. The presence of the cag PAI with the motifs on its right end, the vacA and iceA genotypes and the presence of the glmM and babB house keeping genes and characteristics of the virulence gene oipA and the genes included in the plasticity region cluster are shown. The distribution was determined by PCR using primer sequences from the reference articles denoted by the number in superscript.

Figure 2

(A) shows a gel image of representative FAFLP patterns of English H. pylori strains with the bands in red indicating the internal lane standards (Genescan Rox 500). (B) Representative phylogenetic tree indicating genetic relationships between English H. pylori strains (n = 16) and those from other world populations (HupB, n = 3; Ire, n = 3; Italy (Sard), n = 2; R = 2; E. Asia (Hu), n = 2; India (L, MS, BJ), n = 5) based on the FAFLP profiles generated for the strains from different countries (described in the materials and methods section). The neighbour joining network tree created based on the level of similarity between the amplitypes shows that European strains form a distinct cluster although two other clusters showing associations with the Indian strains are also seen. (C) Cluster analysis of the 296 bp glmM was generated with 10 English strains, 2 strains each from India (L), E. Asia (Hu), Peru (Sjm) and S. Africa (R). 4 each from Ire and HupB and the sequenced HP26695. The rounded circle here shows the grouping of English islolate (N 17) with the Indian (L) and E. Asian (Hu) strains. (D) Tree constructed from the babB gene products from different strains (N, n = 8; Ire, n = 4; HupB and MS, L, n = 3 respectively, 2 each of Hu and Sjm with HP26695). (E) Genetic relationships of the English strains based on the 250 bp sequence of the 5'end of the cagA gene (England, n = 7; East Asia, n = 4; S. Africa, n = 1; W. Africa, n = 2; S. America, n = 2; central America, n = 1; India, n = 2; Bangladesh, n = 2 and Holland, n = 1). A few strains from Nottingham showed sequence similarity with the Asian strains as indicated by the circles in green. (F) Phylogenetic tree based on the REP fingerprinting patterns generated by the English strains (n= 5) in comparison with 5 strains from Ire and 6 from India (L).