Identification of virulence factors in the phytopathogen P. syringae by molecular evolution Public Deposited

Downloadable Content

Download PDF
Last Modified
  • March 21, 2019
  • Fisher, Emily Jane
    • Affiliation: College of Arts and Sciences, Department of Biology
  • The interaction between pathogen and host is often likened to an evolutionary arms race in which the pathogen evolves a novel mechanism to infect the host and the host, in turn, evolves better defenses against the pathogen. This process repeats throughout the history of the pathogen-host interaction as new virulence and defense mechanisms arise. My work used this evolutionary interplay between pathogen and host to identify novel virulence genes in the model bacterial pathogen Pseudomonas syringae. My hypothesis was that genes in the P. syringae genome that determine pathogen virulence may be rapidly evolving due to selective pressure from the plant host immune system. Therefore, rapidly-evolving genes in P. syringae may be required for virulence. I identified genes purported to be rapidly-evolving in the genome of P. sryingae by two methods: codon volatility and comparison of homologs. The codon volatility method identified 31 candidate virulence genes based on the proportion of volatile codons in each gene. Three of these 31 have previously-predicted roles in virulence, though only one has been demonstrated to function in pathogen growth on plants. For the second method, I compared the genomes of three isolates of P. syringae and identified 20 genes with high numbers of nonsynonymous (dN) mutations compared to silent, synonymous mutations (dS). High dN/dS ratios are consistent with positive selection on these genes and we therefore anticipated that these genes would function in bacterial virulence. Mutational analysis of 9 of these candidates revealed that two candidate genes had newly-discovered roles in virulence. Two additional genes were previously-identified to function in virulence, and two others are required for growth in minimal medium. Including only those genes with direct functions in bacterial virulence during infection, my candidate list based on dN/dS ratios gives a 4-fold enrichment in virulence genes compared to the genome as a whole. We therefore conclude that rapid evolution rate as evaluated by dN/dS is a predictor of virulence function and propose that this method could be useful in other systems to identify novel virulence determinants.
Date of publication
Resource type
Rights statement
  • In Copyright
  • Dangl, Jeffery L.
  • Open access

This work has no parents.