Phosphodiesterase Genes Regulate Amylovoran Production, Biofilm Formation, and Virulence in Erwinia amylovora

Abstract

Cyclic di-GMP (c-di-GMP) is a ubiquitous bacterial second messenger molecule that is an important virulence regulator in the plant pathogen Erwinia amylovora Intracellular levels of c-di-GMP are modulated by diguanylate cyclase (DGC) enzymes that synthesize c-di-GMP and by phosphodiesterase (PDE) enzymes that degrade c-di-GMP. The regulatory role of the PDE enzymes in E. amylovora has not been determined. Using a combination of single, double, and triple deletion mutants, we determined the effects of each of the four putative PDE-encoding genes (pdeA, pdeB, pdeC, and edcA) in E. amylovora on cellular processes related to virulence. Our results indicate that pdeA and pdeC are the two phosphodiesterases most active in virulence regulation in E. amylovora Ea1189. The deletion of pdeC resulted in a measurably significant increase in the intracellular pool of c-di-GMP, and the highest intracellular concentrations of c-di-GMP were observed in the Ea1189 ΔpdeAC and Ea1189 ΔpdeABC mutants. The regulation of virulence traits due to the deletion of the pde genes showed two patterns. A stronger regulatory effect was observed on amylovoran production and biofilm formation, where both Ea1189 ΔpdeA and Ea1189 ΔpdeC mutants exhibited significant increases in these two phenotypes in vitro In contrast, the deletion of two or more pde genes was required to affect motility and virulence phenotypes. Our results indicate a functional redundancy among the pde genes in E. amylovora for certain traits and indicate that the intracellular degradation of c-di-GMP is mainly regulated by pdeA and pdeC, but they also suggest a role for pdeB in regulating motility and virulence.IMPORTANCE Precise control of the expression of virulence genes is essential for successful infection of apple hosts by the fire blight pathogen, Erwinia amylovora The presence and buildup of a signaling molecule called cyclic di-GMP enables the expression and function of some virulence determinants in E. amylovora, such as amylovoran production and biofilm formation. However, other determinants, such as those for motility and the type III secretion system, are expressed and functional when cyclic di-GMP is absent. Here, we report studies of enzymes called phosphodiesterases, which function in the degradation of cyclic di-GMP. We show the importance of these enzymes in virulence gene regulation and the ability of E. amylovora to cause plant disease.

Keywords: EAL domain; cyclic di-GMP; exopolysaccharide; fire blight; flagellar motility; levan.

FIG 1

Intracellular levels of c-di-GMP for WT E. amylovora Ea1189, pde and edcA mutant strains, and complemented mutants, measured in strains grown in LB using ultraperformance liquid chromatography coupled with tandem mass spectrometry. Data represent three biological replicates, and error bars represent standard error of the means. Different letters above the bars indicate statistically significant differences (P < 0.05 by Tukey’s honestly significant difference [HSD] test).

FIG 2

Swimming motility (area of colony expansion) for WT E. amylovora Ea1189, pde and edcA mutant strains, and complemented mutants measured 48 hpi on a motility agar plate. Data represent three biological replicates, and error bars represent standard error of the means. Different letters above the bars indicate statistically significant differences (P < 0.05 by Tukey’s honestly significant difference [HSD] test).

FIG 3

Amylovoran production in vitro for WT E. amylovora Ea1189, pde and edcA mutant strains, complemented mutants, and Ea1189 Δams (negative control) 48 hpi at 28°C in MBMA minimal media amended with sorbitol. Data represent three biological replicates, and error bars represent standard error of the means. Letters above the bars indicate statistically significant differences (P < 0.05 by Tukey’s honestly significant difference [HSD] test).

FIG 4

Quantification of in vitro biofilm production for WT E. amylovora Ea1189, pde and edcA mutant strains, and complemented mutants on glass coverslips 48 hpi at 28°C. Data represent three biological replicates, and error bars represent standard error of the means. Different letters above the bars indicate statistically significant differences (P < 0.05 by Tukey’s HSD test).

FIG 5

Diameters of necrotic lesions in immature pears (cultivar Bartlett) 5 days postinoculation with WT Ea1189, pde mutant strains, and complemented mutants and representative images of these infections. Data represent two biological replicates, and error bars represent standard error of the means. Different letters above the bars indicate statistically significant differences (P < 0.05 by Tukey’s HSD test).

FIG 6

(A) Volume of ooze droplet emerging from disease lesions on immature pears infected with WT E. amylovora Ea1189, pde mutants, and complemented mutant strains. Data for double and triple mutants, as well as for the negative control, Ea1189 Δams, are not shown here due to the absence of ooze emergence in immature pears infected with these strains. (B) Amylovoran content in the ooze droplets emerging from immature pears infected with WT E. amylovora Ea1189, pde mutant strains, and complemented mutants. Data for double and triple mutants, as well as for the negative control, Ea1189 Δams, are not shown here, due to the absence of ooze formation in immature pears infected with these strains. For both panels A and B, data represent three biological replicates, and error bars represent standard error of the means. Different letters above the bars indicate statistically significant differences (P < 0.05 by Tukey’s HSD test).

FIG 7

(A) hrpL expression levels 18 hpi in Hrp-MM in WT E. amylovora Ea1189, pde and edcA mutant strains, and complemented mutants. Gene expression levels for the mutants and complemented strains are normalized relative to expression levels recorded in Ea1189. (B) amsG expression levels 18 hpi in MBMA medium in WT E. amylovora Ea1189, pde and edcA mutant strains, and complemented mutants. Gene expression levels for the mutants and complemented strains are normalized relative to expression levels recorded in Ea1189. For both panels A and B, data represent three biological replicates, and error bars represent standard error of the means. Different letters above the bars indicate statistically significant differences (P < 0.05 by Tukey’s HSD test).

FIG 8

Translocation levels of type III effector protein DspE for WT E. amylovora Ea1189, pde mutant strains, and complemented mutants (letters represent respective pde genes) measured in a tobacco model using a DspE-CyaA colorimetric assay. Data represent three biological replicates, and error bars represent standard error of the means. Different letters above the bars indicate statistically significant differences (P < 0.05 by Tukey’s HSD test).