Genetic Characterization of Vibrio cholerae O1 isolates from outbreaks between 2011 and 2015 in Tanzania

Abstract

Background: Cholera outbreaks have occurred in Tanzania since 1974. To date, the genetic epidemiology of these outbreaks has not been assessed.

Methods: 96 Vibrio cholerae O1 isolates from five regions were characterized, and their genetic relatedness assessed using multi-locus variable-number tandem-repeat analysis (MLVA) and whole genome sequencing (WGS).

Results: Of the 48 MLVA genotypes observed, 3 were genetically unrelated to any others, while the remaining 45 genotypes separated into three MLVA clonal complexes (CCs) - each comprised of genotypes differing by a single allelic change. In Kigoma, two separate outbreaks, 4 months apart (January and May, 2015), were each caused by genetically distinct strains by MLVA and WGS. Remarkably, one MLVA CC contained isolates from both the May outbreak and ones from the 2011/2012 outbreak in Dar-es-Salaam. However, WGS revealed the isolates from the two outbreaks to be distinct clades. The outbreak that started in August 2015 in Dar-es-Salaam and spread to Morogoro, Singida and Mara was comprised of a single MLVA CC and WGS clade. Isolates from within an outbreak were closely related differing at fewer than 5 nucleotides. All isolates were part of the 3^rd wave of the 7^th pandemic and were found in four clades related to isolates from Kenya and Asia.

Conclusions: We conclude that genetically related V. cholerae cluster in outbreaks, and distinct strains circulate simultaneously.

Fig. 1

Tanzania map indicating the five geographical regions where Vibrio cholerae was isolated between January and November, 2015. The MLVA CC of the isolates at each location is indicated.

Fig. 2

Distribution genotypes in MLVA clonal complexes depicting the genetic relatedness of the V. cholerae O1. Genotypes identified in a) CC1, two outbreaks: Dar es Salaam in 2011/2012 (boxes with solid lines) and Kigoma in May, 2015 (boxes with dotted lines). The symbol WGS! indicates the location of two genetic characteristics that differentiated between the two portions of the diagram. Those characteristics were serotype: Inaba versus Ogawa and a large (>70 SNVs) difference in the WGS. The relatedness of genotypes is diagrammed in b) for CC2, the August – November 2015 outbreak and c) for CC3, the January 2015 outbreak in Kigoma

Fig. 3

The 7^th pandemic V. cholerae phylogenetic tree depicting the genetic relatedness and position of Tanzanian isolates among selected other sequenced V. cholerae isolates (Additional file 1: Table S1). The phylogenetic tree was constructed on an alignment of the variable sites of 445 strains with predicted recombination sites removed consistent with previous publications [15]. The location and date of collection of selected isolates is noted

Fig. 4

Maximum likelihood tree of genotypes from 39 Tanzanian isolates. Each isolate is identified by a number, its MLVA genotype and the location where it was collected. The date of collection and MLVA CC are assigned to the four clades