Phylogenetic characterization of virulence and resistance phenotypes of Pseudomonas syringae

Abstract

Individual strains of the plant pathogenic bacterium Pseudomonas syringae vary in their ability to produce toxins, nucleate ice, and resist antimicrobial compounds. These phenotypes enhance virulence, but it is not clear whether they play a dominant role in specific pathogen-host interactions. To investigate the evolution of these virulence-associated phenotypes, we used functional assays to survey for the distribution of these phenotypes among a collection of 95 P. syringae strains. All of these strains were phylogenetically characterized via multilocus sequence typing (MLST). We surveyed for the production of coronatine, phaseolotoxin, syringomycin, and tabtoxin; for resistance to ampicillin, chloramphenicol, rifampin, streptomycin, tetracycline, kanamycin, and copper; and for the ability to nucleate ice at high temperatures via the ice-nucleating protein INA. We found that fewer than 50% of the strains produced toxins and significantly fewer strains than expected produced multiple toxins, leading to the speculation that there is a cost associated with the production of multiple toxins. None of these toxins was associated with host of isolation, and their distribution, relative to core genome phylogeny, indicated extensive horizontal genetic exchange. Most strains were resistant to ampicillin and copper and had the ability to nucleate ice, and yet very few strains were resistant to the other antibiotics. The distribution of the rare resistance phenotypes was also inconsistent with the clonal history of the species and did not associate with host of isolation. The present study provides a robust phylogenetic foundation for the study of these important virulence-associated phenotypes in P. syringae host colonization and pathogenesis.

FIG.1.

Phylogenetic tree and distribution of toxin and resistance phenotypes. A linearized neighbor-joining MLST tree from combined rpoD, gyrB, gltA, and gapA data set, constructed using the K2P-γ (α = 0.2) substitution model, is shown. Numbers above the nodes are bootstrap scores based on 1,000 pseudoreplicates. The genetic distance scale is presented below the tree. The tree is identical in its gross topology to one produced by maximum likelihood. Strain designations are presented on the right, along with the host of isolation. The black and white grid represents the presence or absence, respectively, of the assayed phenotypes. Tab, tabtoxin; Phas, phaseolotoxin; Cor, coronatine; Syr, syringomycin; INA, ice nucleation; Kan, kanamycin; Tet, tetracycline; Str, streptomycin; Rif, rifampin; Chl, chloramphenicol; Amp, ampicillin; Cu, copper.