Evolutionary history of the phl gene cluster in the plant-associated bacterium Pseudomonas fluorescens

Abstract

Pseudomonas fluorescens is of agricultural and economic importance as a biological control agent largely because of its plant association and production of secondary metabolites, in particular 2,4-diacetylphloroglucinol (2,4-DAPG). This polyketide, which is encoded by the eight-gene phl cluster, has antimicrobial effects on phytopathogens, promotes amino acid exudation from plant roots, and induces systemic resistance in plants. Despite its importance, 2,4-DAPG production is limited to a subset of P. fluorescens strains. Determination of the evolution of the phl cluster and understanding the selective pressures promoting its retention or loss in lineages of P. fluorescens will help in the development of P. fluorescens as a viable and effective inoculant for application in agriculture. In this study, genomic and sequence-based approaches were integrated to reconstruct the phylogeny of P. fluorescens and the phl cluster. It was determined that 2,4-DAPG production is an ancestral trait in the species P. fluorescens but that most lineages have lost this capacity through evolution. Furthermore, intragenomic recombination has relocated the phl cluster within the P. fluorescens genome at least three times, but the integrity of the cluster has always been maintained. The possible evolutionary and functional implications for retention of the phl cluster and 2,4-DAPG production in some lineages of P. fluorescens are discussed.

FIG. 1.

The P. fluorescens Pf-5 phl cluster is located within the core Pseudomonas genome. The phl cluster genes are surrounded by a box. The arrows on either side represent the genes flanking the cluster in P. fluorescens Pf-5 and indicate ORF orientation. Also shown above the arrows are gene numbers. (a) Comparison of P. fluorescens Pf-5 with the sequenced Pseudomonas species. In this image P. syringae 1448a and B728a are not included, as both have the same result as for DC3000. Identical shaded patterns indicate homologous genes, which are found both 5′ and 3′ of the phl cluster. Homologous genes are contiguous in each of the species examined. (b) PCR results using primers designed to P. fluorescens Pf-5 flanking genes, PFL_5950 and PFL_5961. The positions of the primers are indicated by arrows in both the P. fluorescens Pf-5 and Pf0-1 diagrams in panel a. Shown here are the PCR results for two phl-positive strains (F113 and Pf-5) and two phl-negative strains (Pf0-1 and SBW25).

FIG. 2.

Alignment of the protein sequences of the putative gene identified in the Pf-5, F113, and Q2-1 lineages. This gene is located in the region following the phlE gene from the phl cluster in all lineages and may constitute an additional phl cluster gene.

FIG. 3.

The genomic locations of the phl gene cluster differ between P. fluorescens strains. Shown here are representatives of the main genomic location groups, F113, Q2-1/Q1-87, and Pf-5. The white arrows indicate genes flanking the phl cluster. The black arrows surrounded by a box represent the eight genes of the phl cluster. Arrows also indicate the orientation of the ORF. Homology between the three strains is indicated by red shading. It is seen that the phl cluster and the sequence immediately 3′ are homologous. Overall the 3′ flanking region differs among the three representative strains. In the 5′ region, strains F113 and Q2-1/Q1-87 are similar to each other but are distinct from strain Pf-5.

FIG. 4.

Evolutionary relationships using all sites in the concatenated HK tree consisting of 21 phl-positive strains (shown in color) and 15 phl-negative strains (in black) (a) and in the phlD gene tree among our collection of 21 phl-positive strains (shown in color) (b). The three colors for the phl-positive strains on both phylogenetic trees represent the three different genomic context groups identified. Phylogenetic trees were constructed using the neighbor-joining method based on the Jukes-Cantor distance method (29, 55). Bootstraps of 1,000 computer-generated replicates were used and are shown at the branch nodes on the two phylogenetic trees.

FIG. 5.

Genus tree demonstrating the phylogenetic relationships among Pseudomonas species. Shown here are representatives of Pseudomonas species, i.e., P. syringae, P. fluorescens, P. putida, P. entomophila, and P. aeruginosa. Highlighted by a box are the P. fluorescens species with two phl-positive strains (F113 and Pf-5) and two phl-negative strains (Pf0-1 and SBW25). This concatenated HK tree was constructed using the neighbor-joining method based on the Jukes-Cantor distance method (29, 55). Bootstraps of 1,000 computer generated replicates were used and are shown at the branch nodes on the tree.