Ancestral European roots of Helicobacter pylori in India

Abstract

Background: The human gastric pathogen Helicobacter pylori is co-evolved with its host and therefore, origins and expansion of multiple populations and sub populations of H. pylori mirror ancient human migrations. Ancestral origins of H. pylori in the vast Indian subcontinent are debatable. It is not clear how different waves of human migrations in South Asia shaped the population structure of H. pylori. We tried to address these issues through mapping genetic origins of present day H. pylori in India and their genomic comparison with hundreds of isolates from different geographic regions.

Results: We attempted to dissect genetic identity of strains by multilocus sequence typing (MLST) of the 7 housekeeping genes (atpA, efp, ureI, ppa, mutY, trpC, yphC) and phylogeographic analysis of haplotypes using MEGA and NETWORK software while incorporating DNA sequences and genotyping data of whole cag pathogenicity-islands (cagPAI). The distribution of cagPAI genes within these strains was analyzed by using PCR and the geographic type of cagA phosphorylation motif EPIYA was determined by gene sequencing. All the isolates analyzed revealed European ancestry and belonged to H. pylori sub-population, hpEurope. The cagPAI harbored by Indian strains revealed European features upon PCR based analysis and whole PAI sequencing.

Conclusion: These observations suggest that H. pylori strains in India share ancestral origins with their European counterparts. Further, non-existence of other sub-populations such as hpAfrica and hpEastAsia, at least in our collection of isolates, suggest that the hpEurope strains enjoyed a special fitness advantage in Indian stomachs to out-compete any endogenous strains. These results also might support hypotheses related to gene flow in India through Indo-Aryans and arrival of Neolithic practices and languages from the Fertile Crescent.

Figure 1

Detailed characteristics of Indian H. pylori isolates used in the study. [Yellow, region amplified or present; Blue, region absent or rearranged; -, region failed to amplify].

Figure 2

Neighbor joining tree (Kimura 2-parameter) (right) showing the global population structure of H. pylori wherein Indian isolates are highlighted. The phylogenetic tree was based on a total of 23 sequence records of South and North Indian isolates while incorporating ~400 other sequence records from pubMLST database representing different H. pylori populations and sub populations in the world. The population genetic structure was investigated by determining the multilocus haplotypes based on concatenated sequences of seven unlinked housekeeping genes that are scattered around the H. pylori chromosome. Individual isolates were assigned to bacterial populations called hpEastAsia (sub-populations: hspEAsia, hspMaori, hspAmerind), hpEurope, hpAfrica1 (hspSAfrica, hspWAfrica), hpAsia2 and hpAfrica2 [11]. Representatives from each of these (sub)-populations were chosen for subsequent analysis of the cagPAI. Isolates from the population hpAfrica2 do not contain cagPAI. Phylogenetic relationships were also estimated through NETWORK analysis (left) based on 665 mutating positions that revealed the co-evolution of the H. pylori genome. The Ladakhi (yellow) and other Indian (light green) lineages were more clearly discerned within the European (dark green) cluster (centre box), when analyses based on the remaining 650 mutating positions were performed. For the Neighbor-joining tree (right), the bootstrap values of the interior branches as calculated in MEGA, were significantly high to indicate the correct topology of the branches within the clades.

Figure 3

Comparative genomic analysis of the cagPAIs from Indian isolates. A) PCR based analysis of the complete cagPAI of 5 representative hpEurope Isolates: 3K, 4K, 3C, MS40 and MS38 from India. Overlapping PCR primers amplified the whole cagPAI indicating the intactness of the PAI in these isolates. B) Global pair-wise alignments of whole cagPAI sequences of different H. pylori isolates were generated by VISTA using default parameters [47]. The OK129 genome was taken as the base sequence (not shown) and rest of the sequences were aligned against it. The X-axis denotes length of the sequence under consideration and the Y-axis conveys homology in % with the base genome sequence). The Indian hpEurope isolate, 3K was aligned with other whole cagPAI sequences from GenBank along with the cagPAIs of HP 26695, HPJ99 and HPAG1. The accession numbers for the public domain sequences of the cagPAIs from Europe [9] and Japan [49] that we used in our analyses, were as follows – Ca73 (AY330638 and AY330639), Du23:2 (AY330643 and AY330644), Du52:2 (AY330640, AY330641 and AY330642), F80 (AB120421), OK112 (AB120425), F16 (AB120416), F17 (AB120417), F28 (AB120418), F79 (AB120420), OK101 (AB120422), OK109 (AB120424). Sequence of the French isolate, Fr 908 was determined in this study (EF195721). While the cagPAI sequence of the Indian isolate 3K (hpEurope) was found to be genetically highly similar to and aligning closely with the 26695 sequence, it also revealed significant sequence similarities with other isolates of European origins (that harbor Western type of cag EPIYA sequences) such as HPAG1, OK112, Du52, Du23, Ca73, J99 and Fr908. It was however largely unrelated to the East Asian like isolates (mainly harboring Asian type cag EPIYA sequences) such as F16, F28, F79, OK109, F17, OK101 and F80.

Figure 4

Phylogenetic tree based on the 5' end sequence of the cagA (an informative 219 bp segment of cagA was used to align sequences from unrelated isolates) suggests possible common origins for isolates from ethnic Indians and the tribal. Representative Indian genotypes (3K, MS4, Ms7 and MS15) based on this 219 bp sequence clustered tightly with previously determined genotypes of strains obtained from ethnic Bengalis [India3B (AF202219), India7A (AF202220), India9A (AF202221), India10A (AF202222), India17A (AF202223), India18A (AF202224), India19A (AF202225), DH140 (AY169293), DH200 (AY169294), DH29 (AY169295), DH37 (AY169296), DH60 (AY169297), DH93 (AY169298)] and Santhal and Oraon tribals [Sant4 (AY162446), Sant53 (AY162447), Sant64 (AY162448), Sant67 (AY162449), Sant69 (AY162450), Oraon1 (AY162451), Oraon10 (AY162452), Oraon4 (AY162453)] [20]. All the East Asian strains [China27 (AJ252979), China29 (AJ252980), China40 (AJ252982), China48 (AJ252983), China47 (AJ252985), China59 (AJ252986), Hongkong77 (AF198485), Hongkong81 (AF198486), Hongkong97-42 (AF239733), Japan GC4 (AF198484), Japan32 (AJ239726)], however, clustered together and formed a separate cluster.