New implications on genomic adaptation derived from the Helicobacter pylori genome comparison

Abstract

Background: Helicobacter pylori has a reduced genome and lives in a tough environment for long-term persistence. It evolved with its particular characteristics for biological adaptation. Because several H. pylori genome sequences are available, comparative analysis could help to better understand genomic adaptation of this particular bacterium.

Principal findings: We analyzed nine H. pylori genomes with emphasis on microevolution from a different perspective. Inversion was an important factor to shape the genome structure. Illegitimate recombination not only led to genomic inversion but also inverted fragment duplication, both of which contributed to the creation of new genes and gene family, and further, homological recombination contributed to events of inversion. Based on the information of genomic rearrangement, the first genome scaffold structure of H. pylori last common ancestor was produced. The core genome consists of 1186 genes, of which 22 genes could particularly adapt to human stomach niche. H. pylori contains high proportion of pseudogenes whose genesis was principally caused by homopolynucleotide (HPN) mutations. Such mutations are reversible and facilitate the control of gene expression through the change of DNA structure. The reversible mutations and a quasi-panmictic feature could allow such genes or gene fragments frequently transferred within or between populations. Hence, pseudogenes could be a reservoir of adaptation materials and the HPN mutations could be favorable to H. pylori adaptation, leading to HPN accumulation on the genomes, which corresponds to a special feature of Helicobacter species: extremely high HPN composition of genome.

Conclusion: Our research demonstrated that both genome content and structure of H. pylori have been highly adapted to its particular life style.

Figure 1. Phylogenic trees of strains based on the genomic rearrangements using different reference genomes (A and B) or on the sequences of 4 groups of each 10 concatenated genes (I–IV).

A. using reference genome from strains J99, HPAG1, Shi470 or P12. B. using reference genome from strains 26695, B38, 51, 52 or G27. A10-A16 represents the ancestor genome structure at different evolutionary stage and A16 is the last common ancient of H. pylori. I–IV represents the phylogeny inferred from each group of 10 concatenated core genes.

Figure 2. Distribution of inverted sequences on H. pylori genomes.

The REPuter output files were combined with the Mauve alignment “coordinates output file” to make a plot with java script codes. The grey background figure shows the conserved blocks between genomes produced by Mauve. All the color lines indicate a pair of inverted repeats (>25 bp) at the position of genomes. Red arrows show the positions of some repeats mentioned in the text.

Figure 3. The phylogenic tree of repeats HP0227 and HP1342 with their homologs in different genomes.

The arrows indicate the homologs in strain HPAG1.

Figure 4. Sketch map of genomic structure in an inversion.

Each arrow-formed block represents a gene open reading frame (ORF). The solid block indicates a pair of inverted repeats. The density from pale to dark shows the nucleotide identity degree of various regions between these two inverted repeats in which the dark region means identical. Small black lines indicate other genomic sequences. The letters within the blocks are the names of genes.

Figure 5. The phylogenic tree of repeats HP0722 and HP0725 with their homologs in different genomes.

Figure 6. Sketch map of core genome for protein-coding genes.

The central cycle represents core genome. The secondary cycle indicates the genes that are shared by at least two genomes. Other small partial cycles show the strain-specific genes. The number within the cycles means the number of genes in this category.

Figure 7. The portions of types of premature mutations in pseudogenes.

a. premature mutations in pseudogenes in strain B38. b. premature mutations in pseudogenes in other strains.

Figure 8. Creation of homopolynucleotide (HPN) possibly by recombination.

The left is the locus-tag of gene. The right is the partial sequences of gene.

Figure 9. The comparison of homopolynucleotide (HPN) content in H. pylori strains with other closely related species.

Figure 10. The comparison of homopolynucleotide (HPN) content in H. pylori 26695 with other bacteria, from Archaea to Eubacteria species.

The abbreviations are: AbT469, Aciduliprofundum boonei T469; Ab, Acinetobacter baumannii AB0057; Bq, Bartonella quintana; Cc, Chlamydophila caviae GPIC; Ct, Clostridium thermocellum ATCC 27405; CPCC 7424, Cyanothece sp. PCC 7424; Ch, Cytophaga hutchinsonii ATCC 33406; Lr, Lactobacillus reuteri DSM 20016; Lb, Leptospira biflexa serovar Patoc; Lc, Leuconostoc citreum KM20; Mb, Methanosarcina barkeri; Pm, Proteus mirabilis strain HI4320; Spn, Streptococcus pneumoniae ATCC; Spy, Streptococcus pyogenes M1 GAS; St, Streptococcus thermophilus CNRZ1066; Sd, Sulfurospirillum deleyianum DSM 6946; Tl, Thermotoga lettingae TMO; Vp, Veillonella parvula DSM 2008; As, Aliivibrio salmonicida LFI1238; Cj, Campylobacter jejuni RM1221; EcK12, Escherichia coli K12; Hp, Helicobacter pylori 26695.