A genome-wide scan for genes under balancing selection in the plant pathogen Ralstonia solanacearum

Abstract

Background: Plant pathogens are under significant selective pressure by the plant host. Consequently, they are expected to have adapted to this condition or contribute to evading plant defenses. In order to acquire long-term fitness, plant bacterial pathogens are usually forced to maintain advantageous genetic diversity in populations. This strategy ensures that different alleles in the pathogen's gene pool are maintained in a population at frequencies larger than expected under neutral evolution. This selective process, known as balancing selection, is the subject of this work in the context of a common bacterial phytopathogen. We performed a genome-wide scan of Ralstonia solanacearum species complex, an aggressive plant bacterial pathogen that shows broad host range and causes a devastating disease called 'bacterial wilt'.

Results: Using a sliding window approach, we analyzed 57 genomes from three phylotypes of the R. solanacearum species complex to detect signatures of balancing selection. A total of 161 windows showed extreme values in three summary statistics of population genetics: Tajima's D, θ_w and Fu & Li's D*. We discarded any confounding effects due to demographic events by means of coalescent simulations of genetic data. The prospective windows correspond to 78 genes with known function that map in any of the two main replicons (1.7% of total number of genes). The candidate genes under balancing selection are related to primary metabolism and other basal activities (51.3%) or directly associated to virulence (48.7%), the latter being involved in key functions targeted to dismantle plant defenses or to participate in critical stages in the pathogenic process.

Conclusions: We identified various genes under balancing selection that play a significant role in basic metabolism as well as in virulence of the R. solanacearum species complex. These genes are useful to understand and monitor the evolution of bacterial pathogen populations and emerge as potential candidates for future treatments to induce specific plant immune responses.

Keywords: Balancing selection; Fu & Li’s D*; Pathogenesis; Ralstonia solanacearum; Tajima’s D; Virulence related genes; Watterson’s theta.

Fig. 1

Two-dimensional plot of Tajima’s D and Fu & Li’s D* values for all windows. Shaded dots represent the 5% top windows of the distribution of both measures of BS

Fig. 2

Analysis of genomic region corresponding to the phcBSR operon in strain GMI1000 showing sliding window analyses for two statistics: Tajima’s D (solid line) and Fu & Li’s D* (dotted line). All three genes of the operon comprise about 3.9 Kb of the genome; however, the gene (RSc2734) physically preceding this operon is also shown for comparison purposes. Windows are 200 bp and asterisks indicate the windows with extreme values of respective statistics. Arrows represent gene arrangement in the genome of strain GMI1000