Ralstonia solanacearum pandemic lineage strain UW551 overcomes inhibitory xylem chemistry to break tomato bacterial wilt resistance

Abstract

Plant-pathogenic Ralstonia strains cause bacterial wilt disease by colonizing xylem vessels of many crops, including tomato. Host resistance is the best control for bacterial wilt, but resistance mechanisms of the widely used Hawaii 7996 tomato breeding line (H7996) are unknown. Using growth in ex vivo xylem sap as a proxy for host xylem, we found that Ralstonia strain GMI1000 grows in sap from both healthy plants and Ralstonia-infected susceptible plants. However, sap from Ralstonia-infected H7996 plants inhibited Ralstonia growth, suggesting that in response to Ralstonia infection, resistant plants increase inhibitors in their xylem sap. Consistent with this, reciprocal grafting and defence gene expression experiments indicated that H7996 wilt resistance acts in both above- and belowground plant parts. Concerningly, H7996 resistance is broken by Ralstonia strain UW551 of the pandemic lineage that threatens highland tropical agriculture. Unlike other Ralstonia, UW551 grew well in sap from Ralstonia-infected H7996 plants. Moreover, other Ralstonia strains could grow in sap from H7996 plants previously infected by UW551. Thus, UW551 overcomes H7996 resistance in part by detoxifying inhibitors in xylem sap. Testing a panel of xylem sap compounds identified by metabolomics revealed that no single chemical differentially inhibits Ralstonia strains that cannot infect H7996. However, sap from Ralstonia-infected H7996 contained more phenolic compounds, which are known to be involved in plant antimicrobial defence. Culturing UW551 in this sap reduced total phenolic levels, indicating that the resistance-breaking Ralstonia strain degrades these chemical defences. Together, these results suggest that H7996 tomato wilt resistance depends in part on inducible phenolic compounds in xylem sap.

Keywords: Hawaii 7996 tomato; Ralstonia Race 3 biovar 2; Ralstonia solanacearum species complex; antimicrobial phenolics; bacterial wilt; bacterial wilt resistance.

FIGURE 1

Ralstonia strains UW551 and GMI1000 differ in virulence and induction of defence responses on wilt‐susceptible and wilt‐resistant tomato plants. (a, b) Disease progression of Ralstonia GMI1000 (blue line) or Ralstonia UW551 (orange line) following soil‐soak inoculation of susceptible host MM (purple circles) or resistant host H7996 (green circles) was rated on a 0–4 disease index scale. (c) Bacterial population sizes determined 5 days after unwounded 17‐day‐old tomato plants were petiole‐inoculated with Ralstonia GMI1000 (open blue circles), Ralstonia UW551 (open orange squares), or a 1:1 mixture of both strains (closed symbols). Horizontal bars indicate geometric mean; asterisk indicates difference at p < 0.05, Mann–Whitney test. Numbers below each column indicate the number of plants without detectable Ralstonia cells. (d) Relative expression levels of tomato reporter genes PR‐1a, indicating activation of the salicylic acid (SA) pathway; PR‐1b, indicating activation of the SA and ethylene (ET) pathways; and Pin‐2, indicating activation of the jasmonic acid (JA) pathway. Gene expression was measured by reverse transcription‐quantitative PCR in roots (r) or stems (s) of whole H7996 tomato plants 96 h after inoculation with Ralstonia strain GMI1000 (blue bar), inoculation with UW551 (orange bars), or co‐inoculation with both strains (black bars). Gene expression levels are relative to those in water‐inoculated healthy control plants of the same genotype. Bars show normalized mean fold induction relative to mock‐inoculated control plants (±SE). Expression levels were similar in both tissue types infected by the same bacterial strain, but PR1‐a and PR1‐b expression levels were much higher in response to GMI1000 than to UW551; asterisks indicate p < 0.05, t test. For (a–d) all experiments contained at least three biological assays each with a total of 9–15 plants.

FIGURE 2

Grafting bacterial wilt‐resistant H7996 scions to susceptible rootstocks increases plant defence gene expression in roots. (a) Disease progress of Ralstonia GMI1000 (blue line) on soil‐soak inoculated reciprocally grafted Money Maker (MM) (green)‐H7996 (purple) chimeric plants. Green circles with open bottoms indicate mean disease index of susceptible MM rootstocks grafted to resistant H7996 scions. Green circles with open tops indicate mean disease index of resistant H7996 rootstock grafted to susceptible MM scions. (b–d) Relative expression of reporter genes of the tomato salicylic acid (SA) pathway (PR‐1a), the SA and ethylene (ET) pathways (PR‐1b), and the jasmonic acid (JA) pathway (Pin‐2). Gene expression was measured by reverse transcription‐quantitative PCR in H7996 roots (r; dotted bar) or stems (s; solid bar) of grafted tomato plants 96 h after soil‐soak inoculation with either Ralstonia GMI1000 (blue bars) or Ralstonia UW551 (orange bars). Self‐grafted plants were included as controls for the effect of grafting. Expression levels are relative to those in healthy water‐inoculated control plants. Data are shown as mean fold change expression of three independent experiments; bars show normalized mean fold induction relative to mock‐inoculated control plants of the same genotype (±SE). Letters indicate differences among treatments (p < 0.05, analysis of variance).

FIGURE 3

Resistance‐breaking UW551 overcomes the inhibitory xylem conditions formed during infection of Hawaii 7996 (H7996). (a) Growth of diverse Ralstonia strains (indicated on x‐axis) on ex vivo xylem sap harvested from H7996 plants that were healthy or soil‐soak inoculated by diverse Ralstonia strains (indicated on the y‐axis). Growth was quantified as area under a 24‐h growth curve indicated by shading from white (low area or little to no growth) to black (maximum growth). Asterisks indicate a significant difference compared to each strain's growth in sap from healthy plants, determined by t test, ***p < 0.001, **p < 0.01, *p < 0.05. (b) Growth of Ralstonia GMI1000 and UW551 (indicated on x‐axis) on ex vivo xylem sap harvested from Bonny Best (susceptible), Money Maker (susceptible), and CRA66 (resistant) tomatoes that were infected by GMI1000 or UW551 (indicated on the y‐axis). Growth was quantified as area under the growth curve as for (a) above. For all ex vivo sap experiments, bacterial growth was measured spectrophotometrically over 24 h in a plate reader as OD₆₀₀. (c) Growth of Ralstonia strain GMI1000 in ex vivo xylem sap harvested from H7996 tomato plants that were healthy (black line) or infected by GMI1000 (blue lines). Xylem sap was filter‐sterilized, dehydrated, and rehydrated with minimal medium to 1× original concentration (solid lines) or to 6× original concentration (dashed line). Data are shown as means of three independent experiments, each containing three technical replicates.

FIGURE 4

Resistance‐breaking Ralstonia UW551 alters the H7996 stem environment to favour GMI1000. (a) Growth of Ralstonia GMI1000 in ex vivo xylem sap that was first harvested from GMI1000‐infected H7996 plants, then used as a growth medium for GMI1000 (blue dots) or UW551 (orange dots), after which it was filter‐sterilized and supplemented with 20 mM sucrose to ensure adequate carbon source, and finally inoculated with GMI1000. (b) Colonization of H7996 tomato stems by Ralstonia GMI1000 and UW551; stems were sampled 4 days after plants were soil‐soak inoculated with GMI1000 first and then with UW551 24 h later (left box) or in the reverse order (right box). Each dot indicates the bacterial population size in one plant; n = 30 plants for each treatment; the number of plants with undetectable populations is given in parentheses below each column; horizontal bar indicates geometric mean population. No significant differences were found within strains between inoculation order at p < 0.05, t test. (c) Relative expression levels of salicylic acid (SA), SA/ethylene (ET), and jasmonic acid (JA) defence pathway marker genes, measured by reverse transcription‐quantitative PCR in tomato roots (r) or stems (s) at 96 h after soil‐soak inoculation with either Ralstonia strain GMI1000 first (blue) or UW551 first (orange) as described above. Data are shown as mean expression levels from three independent experiments, each containing RNA from 12 plant samples. Bars show normalized mean fold induction relative to water‐inoculated control plants (±SE). Relative expression levels did not differ between tissue types, but expression of PR1‐a and PR1‐b differed in response to the two bacterial strains as indicated by asterisks (p < 0.05, t test).

FIGURE 5

Neither β‐1,3‐glucanases nor the AcrA or DinF efflux pumps explain why Ralstonia strain UW551 breaks resistance or grows better in Hawaii 7996 (H7996) xylem sap. Expression of β‐1,3‐glucanase genes in stems of wilt‐resistant H7996 tomato plants at 48 h (a) or 120 h (b) after soil‐soak inoculation with Ralstonia UW551 (orange), Ralstonia GMI1000 (blue), or water (grey). Reverse transcription‐quantitative PCR was used to measure expression of class III acid, class III basic, and class II acid β‐1,3‐glucanase genes; expression levels are presented as fold change relative to soil‐soak inoculated wilt‐susceptible Money Maker (MM) plants at 48 or 120 h post‐inoculation (hpi). Bars show normalized mean fold induction relative to MM plants (±SE). Asterisks indicate a significant difference compared to GMI1000 (p < 0.05, t test). Wilt disease progression on wilt‐susceptible MM tomato of wild type and ∆acrA and ∆dinF deletion mutants of Ralstonia UW551 (c) and Ralstonia GMI1000 (d). Unwounded 17‐day‐old tomato seedlings were soil‐soak inoculated to 5 × 10⁷ cfu/g of soil, and plants were rated daily on a 0 to 4 disease index scale. Symbols indicate mean disease index; results reflect three independent experiments, each with 15 plants per treatment (total n = 45). (e) Doubling time of Ralstonia UW551 wild type and ∆acrA and ∆dinF deletion mutants in ex vivo xylem sap from healthy or infected tomato plants as indicated in the legend. The wild‐type strain grew similarly in all saps tested, but loss of either dinF or acrA slowed growth of UW551 in sap from resistant H7996 whether plants were infected by Ralstonia UW551 or Ralstonia GMI1000 (adjusted p < 0.05 by analysis of variance with Tukey's multiple comparisons). Symbols indicate the average doubling times for three independent experiments, each containing three technical replicates; bars indicate standard error.

FIGURE 6

Metabolomic profiling of tomato xylem sap from H7996 following infection by Ralstonia strain GMI1000 or UW551. Gas chromatography–mass spectrometry metabolomics analysis of ex vivo xylem sap from healthy wilt‐resistant H7996 plants and from plants previously infected by GMI1000 or resistance‐breaking strain UW551 was conducted to determine the relative concentrations of the corresponding metabolite in each sample. (a) 2D principal component analysis score plot illustrating the grouping of the variable conditions: H7996 ex vivo xylem sap harvested from plants infected with resistance‐breaking UW551 (orange, H7U) and GMI1000 (blue, H7G). Ellipses on the score plot represent a 95% confidence interval. (b) Volcano plot combining results from fold change (FC) analysis and t tests (p < 0.05) into a single graph to discern identified compounds based on biological interest, statistical significance, or both. Complete metabolomic results are shown in Table S2.

FIGURE 7

UW551 detoxifies ex vivo xylem sap in part by reducing phenolic compounds. (a) Total phenolic compounds in xylem sap harvested from wilt‐susceptible Money Maker (MM) or resistant H7996 tomato plants infected with Ralstonia strain GMI1000 (blue dots), strain UW551 (orange dots), or water (black dots) as measured by the Folin–Ciolcalteu assay. Infection by either Ralstonia strain increased total phenolics in sap from both resistant and susceptible tomato plants relative to in healthy plants (black dots). However, resistant H7996 plant sap contained more phenolics after GMI1000 infection (blue dots) than after UW551 infection (orange dots). (b) Phenolic levels in xylem sap following 24 h growth of Ralstonia GMI1000 on ex vivo sap harvested from H7996 plants infected with Ralstonia GMI1000 (blue dots), UW551 (orange dots), or water (black dots). (c) Phenolic levels in xylem sap following 24 h of growth of resistance‐breaking Ralstonia UW551 on sap harvested from H7996 plants infected with Ralstonia strain GMI1000 (blue dots), strain UW551 (orange dots), or water (black dots). Each dot indicates the average of three technical replicates; each experiment contained three biological replicates. Columns marked with differing lowercase letters are different (adjusted p < 0.05 by analysis of variance with Tukey's multiple comparisons).