Culture‐independent tracking of Vibrio cholerae lineages reveals complex spatiotemporal dynamics in a natural population

Abstract

Populations of the bacterium Vibrio cholerae consist of dozens of distinct lineages, with primarily (but not exclusively) members of the pandemic generating lineage capable of causing the diarrhoeal disease cholera. Assessing the composition and temporal dynamics of such populations requires extensive isolation efforts and thus only rarely covers large geographic areas or timeframes exhaustively. We developed a culture-independent amplicon sequencing strategy based on the protein-coding gene viuB (vibriobactin utilization) to study the structure of a V. cholerae population over the course of a summer. We show that the 26 co-occurring V. cholerae lineages continuously compete for limited space on nutrient-rich particles where only a few of them can grow to large numbers. Differential abundance of lineages between locations and size-fractions associated with a particle-attached or free-swimming lifestyle could reflect adaptation to various environmental niches. In particular, a major V. cholerae lineage occasionally grows to large numbers on particles but remain undetectable using isolation-based methods, indicating selective culturability for some members of the species. We thus demonstrate that isolation-based studies may not accurately reflect the structure and complex dynamics of V. cholerae populations and provide a scalable highthroughput method for both epidemiological and ecological approaches to studying this species.

Population analysis ofVibrio choleraein aquatic reservoirs reveals a novel sister species(Vibrio paracholeraesp. nov.) with a history of association with human infections

“…High-throughput 1 sequencing of viuB marker gene amplicons was used to analyse the subspecies composition of V. 1 cholerae in these two populations. Amplicons of this gene were annotated following a previously 1 established scheme (11), in which diversity within the V. cholerae species is measured based on 1 relative abundance and distribution of viuB alleles. Each allele represents a V. cholerae lineage, 1 the diversity of which is roughly equivalent to that of a clonal complex as traditionally defined 1 by Multi-locus Sequence Typing (11).…”

“…Amplicons of this gene were annotated following a previously 1 established scheme (11), in which diversity within the V. cholerae species is measured based on 1 relative abundance and distribution of viuB alleles. Each allele represents a V. cholerae lineage, 1 the diversity of which is roughly equivalent to that of a clonal complex as traditionally defined 1 by Multi-locus Sequence Typing (11). A single viuB allele (viuB-73) can be found to be uniquely 1 associated with the pandemic generating (PG) lineage which is mostly composed of V. cholerae 1 O1 strains (11).…”

“…It has recently become possible to investigate the details of the population structure of V. cholerae and its close relatives, using a molecular marker based on a single copy housekeeping gene (viuB, vibriobactin utilization protein subunit B), which provides subspecies level resolution (11). This method was used to study a cholera-free region on the east coast of the USA, the Oyster Pond ecosystem (Falmouth, USA), where differences in abundance of individual alleles in particular locations/habitats indicated potential adaptation to ecological conditions at the subspecies level (11,12). A similar study was performed in V. cholerae populations in an inland location (Dhaka) in choleraendemic Bangladesh (11,13).…”

“…This method was used to study a cholera-free region on the east coast of the USA, the Oyster Pond ecosystem (Falmouth, USA), where differences in abundance of individual alleles in particular locations/habitats indicated potential adaptation to ecological conditions at the subspecies level (11,12). A similar study was performed in V. cholerae populations in an inland location (Dhaka) in choleraendemic Bangladesh (11,13).…”

Dynamic Subspecies Population Structure of Vibrio Cholerae in Dhaka, Bangladesh

“…Processing of amplicon sequence reads was performed following the procedure described by Kirchberger et al [21]. Brie y, de-multiplexed raw reads were processed in R [26] using the DADA2 pipeline [27].…”