Genetic diversity and microevolution of Burkholderia pseudomallei in the environment

Abstract

Background: The soil dwelling Gram-negative pathogen Burkholderia pseudomallei is the cause of melioidosis. The diversity and population structure of this organism in the environment is poorly defined.

Methods and findings: We undertook a study of B. pseudomallei in soil sampled from 100 equally spaced points within 237.5 m(2) of disused land in northeast Thailand. B. pseudomallei was present on direct culture of 77/100 sampling points. Genotyping of 200 primary plate colonies from three independent sampling points was performed using a combination of pulsed field gel electrophoresis (PFGE) and multilocus sequence typing (MLST). Twelve PFGE types and nine sequence types (STs) were identified, the majority of which were present at only a single sampling point. Two sampling points contained four STs and the third point contained three STs. Although the distance between the three sampling points was low (7.6, 7.9, and 13.3 meters, respectively), only two STs were present in more than one sampling point. Each of the three samples was characterized by the localized expansion of a single B. pseudomallei clone (corresponding to STs 185, 163, and 93). Comparison of PFGE and MLST results demonstrated that two STs contained strains with variable PFGE banding pattern types, indicating geographic structuring even within a single MLST-defined clone.

Conclusions: We discuss the implications of this extreme structuring of genotype and genotypic frequency in terms of micro-evolutionary dynamics and ecology, and how our results may inform future sampling strategies.

Figure 1. The presence of B. pseudomallei in 100 spaced sampling points within an area of disused land in northeast Thailand.

Figure 2. Graph of the proportion of all pairwise comparisons showing 0, 1, 2 … 7 allelic mismatches for each of 200 primary colonies (strains) examined at three independent sampling points.

The “trials” data represents mean values for 100 random samples of 200 strains drawn from the combined data set of 600 strains from all three sampling points (with replacement). Error bars are based on the 5^th and 95^th percentiles of the 100 random samples. No pw comparisons (real or trial data) differ at all seven loci since the locus ndh is monomorphic (invariant).