Local genes for local bacteria: Evidence of allopatry in the genomes of transatlantic Campylobacter populations

Authors: PASCOE, BEN - University Of Bath; MERIC, GUILLAUME - University Of Bath; YAHAR, KOJI - Kurume University; WIMARALATHNA, HELEN - Oxford University; MURRAY, SUSAN - Swansea University; HITCHINGS, MATTHEW - Swansea University; SPROSTON, EMMA - Health Canada; CARRILLO, CATHERINE - Health Canada; TABOADA, EDUARDO - Public Health Agency Of Canada; COOPER, KERRY - California State University; Huynh, Steven; CODY, ALISON - Oxford University; JOLLEY, KEITH - Oxford University; MAIDEN, MARTIN M. - Oxford University; MCCARTHY, NOEL - Oxford University; DIDELOT, XAVIER - Imperial College; Parker, Craig; SHEPPARD, SAMUEL - University Of Bath

Submitted to: Molecular Ecology Resources
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 5/7/2017
Publication Date: 5/10/2017
Citation: Pascoe, B., Meric, G., Yahar, K., Wimaralathna, H., Murray, S., Hitchings, M.D., Sproston, E.L., Carrillo, C.D., Taboada, E.N., Cooper, K.K., Huynh, S., Cody, A., Jolley, K.A., Maiden, M.J., McCarthy, N., Didelot, X., Parker, C., Sheppard, S.K. 2017. Local genes for local bacteria: evidence of allopatry in the genomes of transatlantic Campylobacter populations. Molecular Ecology Resources. doi: 10.1111/mec.14176.

Interpretive Summary: The genetic structure of bacterial populations can be related to the geographical isolation source. In some species, there is a strong correlation between geographical distances and genetic distances, which can be caused by different evolutionary mechanisms. The historical ancestry Helicobacter pylori can be reconstructed in concordance with human patterns of migration and Mycobacterium tuberculosis show enough non-overlapping geographical clustering to successfully manage clinical outbreaks. In Campylobacter, source attribution from genomic data has been successful in distinguishing the host source of infection, but not geographical origin. In this study, we investigate biogeographical signals in highly recombining genes to determine the extent of clustering in geographically distinct Campylobacter population genomes. Whole genome sequences from 294 Campylobacter isolates from North America and the UK were analyzed and the genetic inheritance of 15 closely matched pairs of isolates was quantified. We demonstrate that isolates from within the same country shared more DNA than isolates from different countries. Using a pairwise approach, we identify regions of high diversity and test their correlation with geographical signal. The seven genes that demonstrated the greatest clustering by geography were used to attribute putative geographical source using an attribution model to infer geographical origjns. A further 383 UK clinical isolates were used to detect signals of foreign travel. Patient records indicated that 46 cases had travelled abroad less than two weeks prior to sampling and 34 (74%) of those Campylobacter genomes were deemed to be from a non-UK origin. Detection of signals of biogeographical differences in Campylobacter genomes will contribute to improved source and geographical attribution of clinical Campylobacter infection and inform public policy and contribute to intervention strategies to reduce campylobacteriosis.

Technical Abstract: The genetic structure of bacterial populations can be related to the geographical isolation source. In some species, there is a strong correlation between geographical distances and genetic distances, which can be caused by different evolutionary mechanisms. The patterns of ancient admixture in Helicobacter pylori can be reconstructed in concordance with human patterns of migration, whereas in Mycobacterium tuberculosis it is the lack of genetic recombination that allows allopatric clusters. In Campylobacter, source attribution based on genomic data has been successful in distinguishing the host source of infection, but not geographical origin. We investigate biogeographical signals in highly recombining genes to determine the extent of clustering in geographically distinct Campylobacter population genomes. Whole genome sequences from 294 Campylobacter isolates from North America and the UK were analysed and the genetic inheritance of 15 closely matched pairs of isolates was quantified. Isolates from within the same country shared more DNA than isolates from different countries. Using a pairwise approach, we identify regions of high diversity and test their correlation with geographical signal. The seven genes that demonstrated the greatest clustering by geography were used to attribute putative source using Structure. A further 383 UK clinical isolates were used to detect signals of foreign travel. Patient records indicated that 46 cases had travelled abroad less than two weeks prior to sampling and 34 (74%) of those Campylobacter genomes were deemed to be from a non-UK origin. Detection of signals of biogeographical differences in Campylobacter genomes will contribute to improved source attribution of clinical Campylobacter infection and inform public policy and contribute to intervention strategies to reduce campylobacteriosis.