Suspected cases of intracontinental Burkholderia pseudomallei sequence type homoplasy resolved using whole-genome sequencing. Issue 11 (14th November 2017)

• Record Type: Journal Article
• Authors: Aziz, Ammar; Sarovich, Derek S.; Harris, Tegan M.; Kaestli, Mirjam; McRobb, Evan; Mayo, Mark; Currie, Bart J.; Price, Erin P.
• Is Part Of: Microbial genomics. Volume 3:Issue 11(2017)
• Publication Date: 2017-11-14
• Subjects: genomics -- melioidosis -- phylogenetics -- homoplasy -- source tracing; Microbial genomics -- Periodicals; 572.8629
• DOI: 10.1099/mgen.0.000139
• Languages: English
• ISSNs: 2057-5858

Abstract

Abstract : Burkholderia pseudomallei is a Gram-negative environmental bacterium that causes melioidosis, a disease of high mortality in humans and animals. Multilocus sequence typing (MLST) is a popular and portable genotyping method that has been used extensively to characterise the genetic diversity of B. pseudomallei populations. MLST has been central to our understanding of the underlying phylogeographical signal present in the B. pseudomallei genome, revealing distinct populations on both the intra- and the inter-continental level. However, due to its high recombination rate, it is possible for B. pseudomallei isolates to share the same multilocus sequence type (ST) despite being genetically and geographically distinct, with two cases of 'ST homoplasy' recently reported between Cambodian and Australian B. pseudomallei isolates. This phenomenon can dramatically confound conclusions about melioidosis transmission patterns and source attribution, a critical issue for bacteria such as B. pseudomallei that are of concern due to their potential for use as bioweapons. In this study, we used whole-genome sequencing to identify the first reported instances of intracontinental ST homoplasy, which involved ST-722 and ST-804 B. pseudomallei isolates separated by large geographical distances. In contrast, a third suspected homoplasy case was shown to be a true long-range (460 km) dispersal event between a remote Australian island and the Australian mainland. Our results show that, whilst a highly useful and portable method, MLST can occasionally lead to erroneous conclusions about isolate origin and disease attribution. In cases where a shared ST is identified between geographically distant locales, whole-genome sequencing should be used to resolve strain origin.