Lack of RsmA-mediated control results in constant hypervirulence, cell elongation, and hyperflagellation in Pectobacterium wasabiae

Abstract

The posttranscriptional regulator RsmA controls the production of plant cell wall degrading enzymes (PCWDE) and cell motility in the Pectobacterium genus of plant pathogens. In this study the physiological role of gene regulation by RsmA is under investigation. Disruption of rsmA gene of the Pectobacterium wasabiae strain, SCC3193 resulted in 3-fold decrease in growth rate and increased virulence. The comparison of mRNA levels of the rsmA(-) mutant and wild-type using a genome-wide microarray showed, that genes responsible for successful infection, i.e. virulence factors, motility, butanediol fermentation, various secretion systems etc. were up-regulated in the rsmA(-) strain. The rsmA(-) strain exhibited a higher propensity to swarm and produce PCWDE compared to the wild-type strain. Virulence experiments in potato tubers demonstrated that in spite of its more efficient tissue maceration, the rsmA(-) strain's ability to survive within the host is reduced and the infection site is taken over by resident bacteria. Taken together, in the absence of RsmA, cells revert to a constitutively infective phenotype characterized by expression of virulence factors and swarming. We hypothesize that lack of control over these costly energetic processes results in decreased growth rate and fitness. In addition, our findings suggest a relationship between swarming and virulence in plant pathogens.

Figure 1. RsmA has a central role in regulating virulence factor synthesis in Pectobacteria.

Arrowheads indicate positive regulatory effects and barred lines indicate negative effects.

Figure 2. In Pectobacteria rsmA expression level affects transcriptome.

A global mRNA profile of the rsmA ⁻ of Pw was compared with its parental strain. The results were compared with a previous study of the expI ⁻ strain of Pa . In the expI ⁻ strain the RsmA level is increased . The genes included in the figure are those which passed two criteria: firstly, the expression level changes were in the opposite direction in response to rsmA inactivation as compared to rsmA up-regulation (expI⁻ strain); secondly, the genes have been associated with virulence. The scale bar(top of figure) represents the fold change in mRNA levels between the wild-type and the rsmA⁻ Pw strain, with positive values representing up-regulation and negative values representing down-regulation in the rsmA⁻ strain.

Figure 3. The rsmA mutant produces high levels of pectinolytic enzymes which constitute the majority of the proteins that are secreted.

Wild-type (wt), rsmA⁻, and rsmB⁻ cells were grown in minimal medium containing either 0.2% glucose (glc) or 0.2% polygalacturonic acid (PGA). Cell-free supernatants were then collected at various timepoints to evaluate protein content. A. The growth curves obtained for each strain and growth condition. The arrows indicate the timepoints used for collecting supernatant samples. B–D. Activity of polygalacturonase (Peh) (B), pectate lyases (Pel) (C), and proteases (Prt) (D) in supernatant samples collected from the exponential phase (white) and the early stationary phase (black) for each strain indicated. The average activity values for three independent experiments are represented. Error bars indicate the standard deviation values. E. Proteins in the wild-type and rsmA⁻ strain culture supernatants were concentrated20 times by acetone precipitation and separated by SDS PAGE. M indicates molecular weight marker.

Figure 4. Plasmid expression of rsmA or rsmB can complement the corresponding knock-out mutants.

A. The production of protease (Prt) was assayed using a specific, semi-quantitative indicator plate, where the halo around the inoculation site was proportional to the amount of enzyme produced. Exogenous expression of rsmA and rsmB (prsmA and prsmB, respectively) was achieved, with pBSK being the vector control. B. To assess glycogen accumulation, cells were grown on Kornberg's medium then exposed to iodine vapors. Darker staining represents higher levels of glycogen. Assays were performed at least three times, and no significant variations were observed.

Figure 5. RsmA affects the swimming and swarming motility of Pw.

A. The swimming ability of wild-type, rsm mutants (rsmA ⁻ and rsmB ⁻), and the respective overexpression strains (wt+prsmA and wt+prsmB) were evaluated based on the migration speed of these cells in 0.25% soft agar plates containing polygalacturonic acid. Cells were imaged 48 h after inoculation. B. The swarming ability of the wild-type, rsm mutants (rsmA ⁻ and rsmB ⁻), and the respective overexpression strains (wt+prsmA and wt+prsmB) were evaluated. The ability of the cells to migrate on the surface of the medium, away from the site of inoculation was examined on 0.4% agar plates containing 10% potato extract. Plates were imaged 24 h after inoculation. Swarming of the wild-type strain was not observed until 72 h after inoculation (not shown). All images are representative of five different experiments performed, and no significant variability was observed between the experiments.

Figure 6. Pw cells change their shape and flagellation depending on their growth environment.

Wild-type (wt) and rsmA⁻ strains grown in different media, as indicated, had their flagella stained according to West et al. . Representative images of cells grown in the liquid medium (0.2% PGA liquid broth), swarming plate (0.4% agar containing 10% potato extract)and potato tuber are shown. The samples from potato tubers were collected 20 h after inoculation, all the other samples 24 h after inoculation. On the swarming plate the wild-type cells became long and hyperflagellated after 72 h of inoculation (not shown).

Figure 7. The role of RsmA in regulating the tricarboxylic acid (TCA) cycle and glyoxylate bypass.

Blue arrows indicate reactions catalysed by the enzymes corresponding to genes up-regulated in the rsmA⁻ strain according to microarray data. Red arrows indicate reactions catalysed by the enzymes corresponding to genes down-regulated in the rsmA⁻ strain. icd–isocitrate dehydrogenase; sucA–2-ketoglutarate dehydrogenase; sucCD–succinyl CoA synthetase; sdhACDB–succinate dehydrogenase; fumA–fumarase; mdh–malate dehydrogenase; gltA–citrate synthase; acn–aconitase; aceA–isocitratelyase; aceB–malate synthase.

Figure 8. RsmA regulates butanediol fermentation in potato tubers.

A. Potato tubers infected with wild-type (wt) or rsmA⁻ strains were assayed for activity of the butanediol fermentation pathway. Specifically, production of acetoin (an intermediate in the butanediol fermentation pathway) was detected spectrophotometrically (at 530 nm) using the Voges-Proskauer method 24 h after inoculation . The experiment was performed in triplicate and error bars indicate standard deviation values. The colorimetric reaction associated with each result is shown below the graph. B. The butanediol fermentation pathway in Pectobacteria. The reactions are catalysed by enzymes coded by budB (acetolactate synthase), budA (α-acetolactatedecarboxylase), and budC (acetoinreductase).

Figure 9. The RsmA mutant is able to grow aerobically on citric acid.

Wild-type (wt), rsmA⁻, and rsmB⁻ strains were tested for their ability to grow aerobically in minimal medium containing 0.2% citric acid as the sole carbon source. The absorbance of cultures at 580 nm was recorded up to 27 h post inoculation. Experiments were performed in triplicate.

Figure 10. The virulence and survival of wild-type and rsmA⁻ strains over the course of potato tuber infections.

Potato tubers (n = 30) inoculated with 5×10⁶ CFU of wild-type (wt) or rsmA ⁻ cells were placed in plastic boxes and incubated at 30°C. A. At 20 h and 40 h post inoculation, macerated tissue was removed from the inoculation site and weighed. Error bars indicate standard deviation values. B. The number of bacteria residing in the infection sites infected with wt, rsmA ⁻ cells or buffer was determined by plating dilution series (green, blue and yellow). The number of rsmA⁻ bacteria present in infection sites was determined using Cm-containing plates to avoid overgrowth by other bacteria inhabiting the infection sites (red). Circles and triangles represent data from the 20 h and 40 h time points assayed, respectively. Horizontal lines represent the median values.