Hydroxycinnamic Acid Degradation, a Broadly Conserved Trait, Protects Ralstonia solanacearum from Chemical Plant Defenses and Contributes to Root Colonization and Virulence

Abstract

Plants produce hydroxycinnamic acid (HCA) defense compounds to combat pathogens, such as the bacterium Ralstonia solanacearum. We showed that an HCA degradation pathway is genetically and functionally conserved across diverse R. solanacearum strains. Further, a feruloyl-CoA synthetase (Δfcs) mutant that cannot degrade HCA was less virulent on tomato plants. To understand the role of HCA degradation in bacterial wilt disease, we tested the following hypotheses: HCA degradation helps the pathogen i) grow, as a carbon source; ii) spread, by reducing HCA-derived physical barriers; and iii) survive plant antimicrobial compounds. Although HCA degradation enabled R. solanacearum growth on HCA in vitro, HCA degradation was dispensable for growth in xylem sap and root exudate, suggesting that HCA are not significant carbon sources in planta. Acetyl-bromide quantification of lignin demonstrated that R. solanacearum infections did not affect the gross quantity or distribution of stem lignin. However, the Δfcs mutant was significantly more susceptible to inhibition by two HCA, namely, caffeate and p-coumarate. Finally, plant colonization assays suggested that HCA degradation facilitates early stages of infection and root colonization. Together, these results indicated that ability to degrade HCA contributes to bacterial wilt virulence by facilitating root entry and by protecting the pathogen from HCA toxicity.

Fig. 1

Hydroxycinnamic acid (HCA) degradation pathway and genes in R. solanacearum GMI1000. A, The HCA degradation pathway is shown with enzyme names in boldface. Fcs, Fca, and Vdh convert the HCAs ferulate, p-coumarate, and caffeate to the phenolic acids vanillate, p-hydroxybenzoate, and protocatechuate, respectively. VanAB and PobA convert vanillate and p-hydroxybenzoate to protocatechuate, which is further metabolized by the β-ketoadipate enzymes; B, A locus containing genes encoding multiple enzymes in the HCA degradation pathway. White arrows indicate ORFs encoding HCA degradation, and grey arrows indicate neighboring ORFS. RSp0221, RSp0224, and RSp0228 encode transcriptional regulators of the LysR family, MarR family, and Fis family, respectively. The dashed line above the genes indicates the region that was precisely excised to create the feruloyl-CoA synthetase deletion mutant (Δfcs). The solid line below the genes indicates the region used to genetically complement the Δfcs mutation in the Δfcs+fcs strain.

Fig. 2

HCA degradation is widely conserved in the R. solanacearum species complex. A, Genetic and functional conservation of HCA degradation. A whole genome comparison phylogenetic tree is shown on the left. Presence of HCA degradation genes and growth of R. solanacearum strains on ferulate (Fer), p-coumarate (Cou), vanillin (Van), vanillate (VA), p-hydroxybenzoate (HBA), and protocatechuate (PCA) are indicated. ^aGrowth phenotype differs from genotype prediction; B, fcs encodes a functional feruloyl-CoA synthetase in strain GMI1000. Strains were grown in minimal media supplemented with 0.2 mM succinate, p-coumarate, caffeate, ferulate, or no carbon (−) for 72 hr. Bars represent the mean of 3 biological replicates and error bars indicate standard error of the mean.

Fig. 3

HCA degradation is required for full virulence of R. solanacearum. A, Disease progress of WT and Δfcs mutant strains on susceptible tomato plants grown at 28 °C. Twenty-one-day-old unwounded plants (cv. Bonny Best) grown at constant 28 °C were inoculated by pouring a bacterial suspension into the soil of each pot. Symptoms were rated using a 0 to 4 disease index scale. Each point represents the mean disease index of a total of 82 plants per strain, in 6 biological replicates. Bars indicate standard error of the mean. Disease progress of the Δfcs mutant was significantly slower than that of wild-type (P=0.0123, two-way repeated measures ANOVA); B, Survival analysis of the above dataset showing the rate of symptom onset after inoculations with WT and the Δfcs mutant; C, Disease progress of strains on tomato plants grown in a 24 °C day /19 °C night cycle (1 biological replicate with N=16 plants per strain).

Fig. 4

Hydroxycinnamic acid degradation does not enhance R. solanacearum growth in plant associated environments. A-C, ex vivo bacterial growth in: A, water extract of potting soil; B, tomato root exudate; and C, tomato xylem sap harvested from stems of un-inoculated, healthy plants. Graphs show the mean of 3 replicates.

Fig. 5

HCA degradation contributes to root entry and competitive fitness following soil soak inoculation of tomato. A-B, Plants grown at 28 °C were soil-soak inoculated with suspensions of WT or Δfcs bacteria. At 3- and 6-days post inoculation (dpi); A, 300 mg root tissue; or B, 100 mg of midstem tissue were harvested, ground, and dilution plated to determine cell density of R. solanacearum (N=30 plants for root colonization at 3 and 6 dpi; N=20 for stem colonization at 3 dpi and N=30 at 6 dpi); Solid lines represent the median population sizes and the dashed lines represent the limit of detection. WT-gm colonized roots better than Δfcs-gm at 3 days after inoculation (P<0.0094; t-test); C, Competitive fitness of WT vs. Δfcs bacteria following soil-soak inoculation. Tomato plants were co-inoculated with mixtures of reciprocally-marked WT and Δfcs strains. At the first stage of disease (less than 25% leaves wilted), midstem tissue was harvested, ground, and dilution plated. Population size of each strain was normalized by initial inoculum. Median competitive index (CI) of the Δfcs mutant was 0.46 (P<0.0001, Wilcoxon Signed Rank Test; N=13 plants per co-inoculation, 26 total); D, Competition of Δfcs and WT bacteria in tomato stem following direct stem inoculation. Tomato plants were co-inoculated via a cut leaf petiole with 4,000 CFU in a 1:1 suspension of reciprocally-marked WT and Δfcs strains. Midstem tissue was harvested at the first sign of symptoms, ground, and dilution plated. Population size of each strain was normalized by initial inoculum. Median CI of the Δfcs mutant was 0.71 (P=0.225, Wilcoxon Signed Rank Test; N=14 plants per co-inoculation, 28 total).

Fig. 6

HCA degradation by R. solanacearum did not affect total lignin quantity or distribution in tomato stems. A, Mean gross lignin content in tomato stems at 6 days post soil-soak inoculation. Whole stems of healthy (mock-inoculated) or infected tomato plants (n=6 per condition) were dried, ground, and analyzed by the acetyl bromide lignin quantification assay using wood pulp inulin as a standard. Error bars indicate standard error of the mean. Similar results were obtained at 3 and 9 days after inoculation; B-D, Phloroglucinol HCl-stained cross-sections of stems from representative healthy or symptomatic infected plants. Pink precipitate indicates lignin.

Fig. 7

HCA degradation protected R. solanacearum from HCA toxicity. Bacterial growth in succinate minimal medium supplemented with increasing concentrations of: A, p-coumarate (p-Cou); B, caffeate (Caf); or C, ferulate (Fer). Culture optical density was measured by a plate reader 48 hr after inoculation with 10⁵ CFU/ml of bacteria. Growth of each strain was calculated relative to that of wild-type bacteria growing without HCAs. Error bars indicate standard error of the mean. The WT strain was less inhibited than the Δfcs mutant by p-coumarate and caffeate (t-test; P< 0.005).