Specificity and genetic polymorphism in the Vfm quorum sensing system of plant pathogenic bacteria of the genus Dickeya

Abstract

The Vfm quorum sensing (QS) system is preponderant for the virulence of different species of the bacterial genus Dickeya. The vfm gene cluster encodes 26 genes involved in the production, sensing or transduction of the QS signal. To date, the Vfm QS signal has escaped detection by analytical chemistry methods. However, we report here a strain-specific polymorphism in the biosynthesis genes vfmO and vfmP, which is predicted to be related to the production of different analogues of the QS signal. Consequently, the Vfm communication could be impossible between strains possessing different variants of the genes vfmO/P. We constructed three Vfm QS biosensor strains possessing different vfmO/P variants and compared these biosensors for their responses to samples prepared from 34 Dickeya strains possessing different vfmO/P variants. A pattern of specificity was demonstrated, providing evidence that the polymorphism in the genes vfmO/P determines the biosynthesis of different analogues of the QS signal. Unexpectedly, this vfmO/P-dependent pattern of specificity is linked to a polymorphism in the ABC transporter gene vfmG, suggesting an adaptation of the putative permease VfmG to specifically bind different analogues of the QS signal. Accordingly, we discuss the possible involvement of VfmG as co-sensor of the Vfm two-component regulatory system.

Fig. 1

Functional annotation of the vfm gene cluster of Dickeya dadantii 3937 (taken from the GenBank accession NC_014500). In the upper part, the 26 genes of the gene cluster are presented by arrows. Green arrows correspond to genes annotated as biosynthesis genes, red arrows are for genes annotated as regulatory or transporter genes and black arrows for genes encoding hypothetical proteins. The length of arrows is proportional to the gene length according to the scale indicated at the right. The predicted function of each gene is shown below, with the same colour code. Abbreviations: ABC: ATP binding cassette, ACP: acyl carrier protein, MATE: multidrug and toxic compound extrusion. The corresponding GenBank tags are indicated for each vfm genes.

Fig. 2

Distribution of strains of the five Vfm genetic groups within the 12 species of Dickeya. The length of each block is proportional to the number of strains indicated in the block. The colour of the blocks refers to the Vfm genetic group as indicated at the right of the figure. These data are based on the analysis of the 126 Dickeya genomic sequences available in GenBank until April 2021 plus the genomic sequences of the strains D. fangzhongdai NCPPB 2929, D. dadantii NCPPB 3065 and D. dadantii CFBP 3964 obtained in the present study.

Fig. 3

Characterization of variable regions in the vfm gene cluster between strains of Dickeya belonging to the same species but to different Vfm genetic groups. Top: Map of the vfm gene cluster. Genes are represented by arrows whose length is proportional to the gene length according to the scale indicated at the right. The genes vfmO, vfmP, vfmW and vfmG are indicated by the corresponding letter. White boxes below each of these four genes indicate the positions of the nucleotide sequences used to build the phylogenetic trees described in Fig. 4. A–K. Schematic representation of the alignment of nucleotide sequences of the vfm gene clusters for pairs of strains belonging to the same species but to different Vfm genetic groups (I, II, III or IV). Grey boxes correspond to the regions in which the aligned sequences of strain pairs share a nucleotide identity higher than 93%. White boxes correspond to the predicted recombined regions in which both aligned sequences share a nucleotide identity lower than 77%. Values of nucleotide identities shared by each strain pairs are indicated inside, below or above each box. A. D. dadantii 3937 (group I) versus D. dadantii NCPPB 3537 (group IV). B. D. fangzhongdai NCPPB 2929 (group I) versus D. fangzhongdai S1 (group IV). C. D. chrysanthemi NCPPB 516 (group II) versus D. chrysanthemi NCPPB 402T (group IV). D. D. fangzhongdai NCPPB 2929 (group I) versus D. fangzhongdai DSM 101947T (group III). E. D. chrysanthemi NCPPB 516 (group II) versus D. chrysanthemi NCPPB 3533 (group III). F. D. oryzae EC2 (group II) versus D. oryzae EC1 (group III). G. D. chrysanthemi NCPPB 3533 (group III) versus D. chrysanthemi NCPPB 402T (group IV). H. D. fangzhongdai DSM 101947T (group III) versus D. fangzhongdai S1 (group IV). I. D. solani IPO 2222T (group III) versus D. solani RNS 05.1.2A (group IV). J. D. undicola FVG1‐MFK‐O17 (group III) versus D. undicola 2B12T (group IV). K. D. zeae MS1 (group III) versus D. zeae NCPPB 2538T (group IV).

Fig. 4

Phylogenetic trees of the variable regions of the genes vfmO, vfmP, vfmG and vfmW of a selection of representative Dickeya strains. A. vfmO tree. B. vfmP tree. C. vfmG tree. D. vfmW tree. The 18 selected strains correspond to triplets or pairs of strains, each triplet or pair belonging to the same species but to different Vfm genetic groups, except for D. poaceiphila whose two strains belong to group V. The colour code referring to their Vfm genetic group (strains I to V) is given at the top left corner. To simplify the figure, the names of the species are indicated as follows: Dch, D. chrysanthemi (three strains); Dda, D. dadantii (two strains); Dfa, D. fangzhongdai (three strains); Dor, D. oryzae (two strains); Dpo, D. poaceiphila (two strains); Dso, D. solani (two strains); Dun, D. undicola (two strains); Dze, D. zeae (two strains). The nucleotide sequences corresponding to the variable regions of the genes vfmO (727 nt), vfmP (901 nt), vfmG (2149 nt) and vfmW (229 nt) have been obtained from genome sequences available in GenBank (accession numbers given in Table S1), except for the strain NCPPB 2929 sequenced in the present study. The positions of these nucleotide sequences in the vfm gene cluster are indicated in Fig. 3. Nucleotide sequences have been aligned with MUSCLE (Edgar, 2004), the phylogenetic tree was reconstructed using the maximum likelihood method implemented in the PhyML program (Guindon and Gascuel, 2003) and graphical representation of the phylogenetic tree was performed with TreeDyn (Chevenet et al., 2006). Numbers at nodes correspond to bootstrap values (500 replicates of the original alignment). Bootstrap values are indicated only for the nodes that separate the different variants of the genes vfmO or vfmP or the main phyla for the genes vfmG or vfmW. The scale bar represents the average number of substitutions per site.