Background

Helicobacter pylori is a Gram-negative bacterium that colonises human stomach, resulting in various gastric diseases including gastritis, peptic ulcer disease and gastric cancer. About half of the world population is infected with H. pylori with prevalence in developing countries reaching up to 90% [1, 2]. H. pylori is well-known for its genome’s ability to mirror the history of human migration history due to its very long association with humans and transmission being within families [2,3,4,5]. This coevolution has led to the emergence of seven different MLST population types of H. pylori, based on the geographical origins, namely hpAfrica1, hpAfrica2, hpNEAfrica, hpEastAsia, hpAsia2, hpEurope and hpSahul [2, 3, 5].

Helicobacter pylori harbours large number of type II restriction-modification (R-M) systems [6,7,8,9]. The type II systems have separate DNA methyltransferase (MTase) and restriction endonuclease proteins (REase) that act on the same DNA sequence motif. Apart from protecting host DNA from foreign DNA, DNA methylation has been implicated in the regulation of bacterial gene expression [10]. The Type II RM systems in H. pylori are substantially diverse among strains and therefore can be used as a biomarker to trace geographical association of H. pylori [11, 12].

Helicobacter pylori strains of hpNEAfrica population type are mainly found circulating in Ethiopia, Somalia, Sudan and Nilo-Saharan speakers in northern Nigeria [2, 13]. No complete genome of a H. pylori strain originated from this region is available in public databases. Our complete genome and methylome data of H. pylori HP14039 hence will provide further indicators on the evolution and genetic diversity of this human gastric pathogen.

Methods

Bacterial culture and genomic DNA extraction

Helicobacter pylori strain HP14039 was isolated from a patient gastric biopsy sample onto selective agar plates. The selective plates used were Columbia blood agar plates (CBA) containing 5% horse blood (PathWest Laboratory Medicine WA Media, Australia) with Dent supplement (Oxoid, UK). The plates were incubated for 3–4 days at 37 °C in a 10% CO2 environment. The genomic DNA extraction was performed on 48 h bacterial culture using phenol–chloroform method [14]. Cells were harvested from culture plates and washed with PBS (pH 8) followed by centrifugation at 14,000 rpm for 1 min. Following the removal of supernatant, the pellet was resuspended in 50 µl of 0.5 M EDTA and 200 µl of sodium dodecyl sulphate and incubated at 50 °C for 2 h. Resultant lysate was thoroughly mixed with one volume 25:24:1 phenol:chloroform:isoamyl alcohol solution in a phase separating gel tube and spun at 14,000 rpm for 5 min; repeated once, then again subsequently with 24:1 chloroform:isoamyl alcohol. The aqueous layer was transferred to two volumes ice cold ethanol and gently mixed immediately. Precipitated DNA was then washed with 70% ethanol and solubilised in TE buffer. DNA quality and quantity were assessed using both Nanodrop (Thermofisher, USA) and Qubit (Thermofisher, USA).

PacBio and Illumina MiSeq genome sequencing

The genomic DNA was sequenced using two sequencing platforms, Pacbio RSII and Illumina MiSeq. The PacBio sequencing was conducted by Novogene (HK) Co., Ltd, Hong Kong. For Illumina MiSeq sequencing, the genomic library was prepared using Nextera XT kit (Illumina, San Diego, USA) according to manufacturer’s protocol and sequenced using 2 × 300 paired-end protocol on an Illumina MiSeq instrument.

Genome assembly and annotation

The Pacbio raw reads were assembled into a single contig using Canu assembler v1.7 [15], after which the assembly was circularized using Circlator v1.5.5 [16]. The circularised contig was subjected to further correction by mapping of Illumina MiSeq-generated paired-end reads using CLC Genomics Workbench 11 with default parameters (QIAGEN). Genome annotation was performed using Prokka v1.12 [17]. The annotation features are available in Additional file 1. Genome completeness and contamination of HP14039 genome was assessed using the taxonomy_wf workflow at species level available in CheckM v1.0.13.

Processing of PacBio methylome data

All raw data in bax.h5 format were converted and merged into a bam file using bax2bam v0.0.8 prior to alignment to H. pylori HP14039 complete genome sequence using blasr v5.3.2 with default parameters. The aligned bam output file was then subjected to ipdSummary v2.3 to detect kinetic variations that were predictive of DNA modification events. Finally, the methylated DNA motifs were deduced using MotifMaker v0.3.1 [18, 19]. The density of methylated sites was plotted using Circos v0.69-6 with a window of 5000 bp [20].

Phylogenetic analysis

The complete genome of HP14039, and 47 publicly available H. pylori complete genomes from NCBI database and 12 draft genomes of H. pylori strains isolated from our patients who were born in Northeast Africa, were used for core genome phylogeny analysis. The accession numbers of all H. pylori genomes used in this study are provided in Additional file 2: Table S1. For consistency, all genomes were annotated by Prokka v1.12 prior to using Roary v3.12.0 [21] for core genome analysis. In the Roary pipeline, sequence alignment of multiple core genes was performed using MAFFT v7.271 [22] and we specified that a gene must be present in all H. pylori strains to be considered as a core gene with the percentage identity cut-off of 95. The core alignment was then used to construct a neighbour joining tree using Mega v7.0.2 [23] and the output phylogenetic tree was visualized using Figtree v1.4.3 (http://tree.bio.ed.ac.uk/software/figtree).

Quality assurance

Species confirmation was performed by using biochemical tests (urease, catalase and oxidase positives) and PCR with seven species-specific housekeeping genes (atpA, efp, mutY, ppa, trpC, ureI and yphC). Bacterial culture of pure single colony was used for genomic DNA extraction.

Results and discussion

Strain metadata and genomic characteristics

Helicobacter pylori strain HP14039 was isolated from the gastric biopsy of an Australian resident who was born in Somalia, located in the Northeast Africa region.

The genome of H. pylori HP14039 was sequenced using PacBio and Illumina technologies at 257× and 159× genome coverages, respectively. The final assembled genome is 1,678,260 bp in length with 1574 coding sequences, 36 tRNA genes and 38.7% G + C content. Genome assessment using CheckM revealed no contamination and 99.13% genome completeness. We found that it completely lacks the cag pathogenicity island, which is one of the major virulence factors and is thought to be associated with the development of gastric cancer [24]. Other major H. pylori virulence factors present in H. pylori HP14039 are listed in Table 1.

Table 1 Presence and absence of major virulence factors in HP14039 complete genome and other hpNEAfrica draft genomes (“+” means presence and “–” means absence)
Full size table

Methylome of HP14039

Pacbio SMRT sequencing technology has the advantage of being able to detect the epigenetic state of sequenced DNA, and allow identification of modified nucleotides and methylated motifs. In HP14039 genome, a total of 62,407 methylated genomic positions were detected (m6A and m4C). The distribution of methylated bases over the HP14039 chromosome is presented in Fig. 1. Fifteen functional MTases were identified of which thirteen were assigned to their MTase genes based on previous studies [8, 25, 26]. Two methylated motifs, WCANHNNNNTG and CTANNNNNNNTAYG detected in this study were not described in earlier studies. All recognition sequence motifs and their corresponding MTases are summarised in Table 2.

Fig. 1
figure 1

Circos plot displaying the density of methylated sites in HP14039 complete genome (5000 bp window). Open reading frames are highlighted in red in the outer ring. The inner histogram displays the abundance of methylated bases within every 5000 bp region over the chromosome, with an average of 367.9 ± 49.7. The hypermethylated and hypomethylated regions with methylated sites exceeding two standard deviations (> 466.7 and < 267.9, respectively) are highlighted in yellow and dark blue, respectively. The three outer green tracks within the histogram represent density values ranging from 600 to 481 (from outer to inner). The three inner orange tracks within the histogram represent density values ranging from 320 to 201 (from outer to inner)

Full size image
Table 2 Methylated motifs detected in HP14039
Full size table

Phylogeny

The neighbour joining tree was constructed using core genome alignment derived from 48 complete H. pylori genomes including HP14039, and additionally 12 draft genomes of H. pylori strains isolated from patients originated from similar African region as HP14039. Among the 12 clinical strains that were included, two were from Somalia, identical to that of HP14039; four each from Sudan and Ethiopia, respectively; and the remaining two were from Eritrea. As H. pylori infection is common in early childhood [27], it is therefore highly likely that the patients have acquired these individual strains locally when young prior to their migration to Australia. The phylogenetic tree showed clear separation of H. pylori population types (Fig. 2). As expected, HP14039, along with other 12 clinical strains with similar geographical origins, were found clustered together. Importantly, HUP-B14, ELS37 and SJM180, which were isolated from Spain, El Salvador and Peru, respectively, were found to be closely related to hpNEAfrica and hpAfrica1 populations despite previous reports that these strains belong to the hpEurope population [28]. This indicates that the birthplace of the patient plays a more important and accurate role in determining the population type of a H. pylori isolate, than the geographical origin where the clinical isolate was acquired, as countless individuals are constantly migrating and moving in today’s globalised world.

Fig. 2
figure 2

Core genome based phylogenetic tree of HP14039 with 47 complete genomes and 12 draft genomes. HP14039 is highlighted in red

Full size image