Enhanced disease susceptibility 1 and salicylic acid act redundantly to regulate resistance gene-mediated signaling

Abstract

Resistance (R) protein-associated pathways are well known to participate in defense against a variety of microbial pathogens. Salicylic acid (SA) and its associated proteinaceous signaling components, including enhanced disease susceptibility 1 (EDS1), non-race-specific disease resistance 1 (NDR1), phytoalexin deficient 4 (PAD4), senescence associated gene 101 (SAG101), and EDS5, have been identified as components of resistance derived from many R proteins. Here, we show that EDS1 and SA fulfill redundant functions in defense signaling mediated by R proteins, which were thought to function independent of EDS1 and/or SA. Simultaneous mutations in EDS1 and the SA-synthesizing enzyme SID2 compromised hypersensitive response and/or resistance mediated by R proteins that contain coiled coil domains at their N-terminal ends. Furthermore, the expression of R genes and the associated defense signaling induced in response to a reduction in the level of oleic acid were also suppressed by compromising SA biosynthesis in the eds1 mutant background. The functional redundancy with SA was specific to EDS1. Results presented here redefine our understanding of the roles of EDS1 and SA in plant defense.

Figure 1. Morphological, molecular, and defense phenotypes of ssi2 ndr1-1 sid2-1 and ssi2 eds1-1 sid2-1 plants.

(A) Comparison of the morphological phenotypes displayed by 3-week-old soil-grown plants (scale, 0.5 cm). (B) Microscopy of trypan blue-stained leaves from wt (SSI2, Col-0 ecotype), ssi2, ssi2 eds1-1, ssi2 sid2-1 and ssi2 eds1-1 sid2-1 plants (scale bars, 270 microns). (C) SA and SAG levels in indicated genotypes. The error bars indicate SD. Asterisks indicate data statistically significant from wt Nö ecotype (SSI2) (P<0.05, n = 4). (D) Expression of PR-1 and PR-2 genes in indicated genotypes. Total RNA was extracted from 4-week-old plants and used for RNA gel-blot analysis. Ethidium bromide staining of rRNA was used as the loading control. The PR-1 transcript levels in EDS1 SID2 F2 plants were similar to those of wt plants (data not shown). (E) RT-PCR analysis of various R genes in indicated genotypes. The level of β-tubulin was used as an internal control to normalize the amount of cDNA template. (F) RT-PCR analysis of various R genes in indicated genotypes. The level of β-tubulin was used as an internal control to normalize the amount of cDNA template. The expression of R genes in EDS1 SID2 F2 plants was similar to that of wt plants (data not shown). (G) Levels of Myc-tagged RPM1 protein in indicated genotypes. Levels of Rubisco were used as the loading control.

Figure 2. Restoration of ssi2 phenotypes in ssi2 eds1-1 sid2-1 plants and glycerol responsiveness of eds1-1 sid2-1 plants.

(A) Visual phenotypes of water- or BTH–treated wt (SSI2; Col-0 ecotype) and ssi2 eds1-1 sid2-1 plants. The plants were photographed at 2 days post treatment (dpt). (B) Microscopy of trypan blue-stained leaves from BTH–treated wt (SSI2; Col-0 ecotype), sid2, eds1-1 and ssi2 eds1-1 sid2-1 plants. The plants were treated with BTH and stained at 2 dpt (scale bars, 270 microns). (C) RT–PCR analysis of R genes in water- or BTH-treated ssi2 eds1-1 sid2-1 plants. Untreated wt (SSI2; Col-0 ecotype) and ssi2 plants were used as controls. The expression of R genes in EDS1 SID2 F2 plants was similar to that of wt plants (data not shown). The level of β-tubulin was used as an internal control to normalize the amount of cDNA template. (D) RT–PCR analysis of various R genes in water- or glycerol-treated sid2-1 and eds1-1 sid2-1 plants. The glycerol-treated wt (SSI2; Col-0 ecotype) and eds1-1 were included as additional controls. The expression of R genes in water- or glycerol-treated EDS1 SID2 F2 plants was similar to that of water- or glycerol-treated wt plants, respectively (data not shown). The expression of R genes in wt and eds1-1 plants was similar to that seen in sid2-1 or eds1-1 sid2-1 plants. The plants were treated with water or glycerol for three days and analyzed for 18∶1 levels and R gene expression. The level of β-tubulin was used as an internal control to normalize the amount of cDNA template. The 18∶1 content of each genotype is shown as mol%±SD.

Figure 3. Interaction phenotypes of TCV with HRT ssi2 eds1-1 sid2-1 and HRT eds1-1 sid2-1 plants.

(A) Percentage TCV susceptible plants. HRT and hrt indicate resistant and susceptible ecotypes Di-17 and Col-0, respectively. Approximately 70–100 plants were scored for each genotype three-weeks post inoculation and all susceptible plants showed crinkling phenotype and drooping of the bolt . (B) Expression of PR-1 gene in indicated genotypes after mock- or TCV-inoculation. Total RNA was extracted from inoculated leaves at 3 dpi. Ethidium bromide staining of rRNA was used as the loading control. (C) HR formation in indicated genotypes at 3 dpi. The HR response in TCV-inoculated HRT EDS1 SID2 F2 plants was similar to that seen in TCV-inoculated Di-17, HRT sid2-1 or HRT eds1-1 plants. Plants lacking HRT (Col-0, Nö ecotypes or EDS1 SID2 F2's) did not show any HR. (D) Lesion size in indicated genotypes at 3 dpi. Lesion size was determined from ∼23 individual leaves from each genotype. Statistical significance was determined using Students t-test. Asterisks indicate data statistically significant from those of HRT, HRT EDS1 SID2, HRT sid2-1 or HRT eds1-1 plants (P<0.05, n = 23). The error bars indicate SD. (E) ELISA showing levels of TCV CP in the inoculated leaves of indicated genotypes at 3 dpi. Asterisks indicate data statistically significant from results for HRT (Di-17 ecotype) plants (P<0.05, n = 4). The error bars indicate SD. (F) Transcript levels of TCV CP in the inoculated leaves of indicated genotypes at 3 dpi. Ethidium bromide staining of rRNA was used as the loading control. (G) Expression of PR-1 gene in indicated genotypes. Total RNA was extracted from inoculated leaves at 3 dpi. Ethidium bromide staining of rRNA was used as the loading control. The PR-1 gene expression in TCV-inoculated HRT EDS1 SID2 F2 plants was similar to that observed in TCV-inoculated HRT, HRT eds1-1 or HRT sid2-1 plants (data not shown). (H) RT–PCR analysis showing HRT and PR-2 transcript levels in indicated genotypes. The plants were inoculated with TCV and leaf samples were harvested 24 h post inoculation. The level of β-tubulin was used as an internal control to normalize the amount of cDNA template.

Figure 4. Interaction phenotypes of virulent or AvrRPT2-expressing P. syringae with eds1 sid2 plants.

(A) Photograph showing phenotypes produced upon infiltration of 10⁵ CFU/ml bacteria (AvrRPT2). All genotypes were in the Col-0 background. The leaves were photographed at 3 days post inoculation (dpi). The pathogen-inoculated EDS1 SID2 F2 plants showed absence of any visible symptoms in response to bacterial inoculations, similar to Col-0 plants (data not shown). (B) Growth of virulent or avirulent (expressing AvrRPT2) P. syrinage on indicated genotypes. The error bars indicate SD. Asterisks and omega symbols indicate data statistically significant from wt (Col-0) or sid2 (P<0.05, n = 4), respectively. All genotypes are in the Col-0 background. (C) Levels of HA-tagged RPS2 protein at 0, 2, 4, 8, and 24 h post inoculation with P. syringae expressing AvrRPT2. Levels of Rubisco were used as the loading control. (D) Levels of HA-tagged RPS2 protein in indicated genotypes. Levels of Rubisco were used as the loading control.

Figure 5. Interaction phenotypes of H. arabidopsidis biotype Emco5 expressing Atr8 with RPP8 eds1-2 nahG or RPP8 eds1-2 sid2-1 plants.

(A) Whole leaf pictures showing growth of Emco5 on the cotyledons from indicated genotypes. All genotypes were in the Ler background. Cotyledons were photographed 10 days after inoculation. (B) Trypan blue stained leaf showing microscopic HR on Ler and Ler nahG leaves, and trailing necrosis on eds1-2 and eds1-2 nahG leaves (scale bars, 270 microns). Both high (100×) and low magnification (100×) images of eds1-2 nahG leaf are shown. Pathogen inoculations were carried out in F2, F3, and F4 generations with consistent results. The F2 plants showing wt genotype at the mutant locus were resistant to pathogen infection (data not shown). (C) Quantification of pathogen growth on RPP8 EDS1, RPP8 eds1-2 and RPP8 eds1-2 nahG plants. Approximately, 40–60 cotyledons were assayed for each genotype. Asterisks indicate absence of spores. All genotypes were in the Ler background. (D) Quantification of pathogen growth on RPP8 sid2, RPP8 eds1-2, and RPP8 eds1-2 sid2-1 plants. All genotypes were in the ssi2 background. Approximately, 40–60 cotyledons were assayed for each genotype. Asterisks indicate absence of spores.

Figure 6. Effect of SA pretreatment and EDS1 overexpression on pathogen resistance.

(A) Expression of EDS1 and PR-1 in EDS1 (Col-0) and 35S-EDS1 (Col-0) plants. Total RNA was extracted from 4-week-old plants and ethidium bromide staining of rRNA was used as the loading control. (B) Growth of P. syrinage AvrRPS4 on indicated genotypes (all in Col-0 background). Single asterisks indicate data statistically significant from results for water-treated wt (Col-0) (P<0.05, n = 4). Two asterisks indicate data statistically significant from results for SA–treated wt (Col-0) (P<0.05, n = 4). The error bars indicate SD.

Figure 7. Basal resistance in eds1 sid2 plants.

(A) Growth of virulent P. syrinage on indicated genotypes. The error bars indicate SD. Asterisks indicate data statistically significant from wt (Col-0 or Ws) (P<0.05, n = 4). The eds1-1 and eds1-22 are in Ws and Col-0 ecotypic backgrounds, respectively. (B) ELISA showing levels of TCV CP in the inoculated leaves of indicated genotypes at 3 dpi. The error bars indicate SD (n = 4).

Figure 8. Morphology, cell death, SA/SAG levels.

PR-1 and R gene expression ssi2 eds1-2 pad4-1 and ssi2 eds1-2 eds5-1 plants. (A) Comparison of the morphological phenotypes displayed by 4-week-old soil-grown wt (SSI2), ssi2, ssi2 eds1, ssi2 pad4, ssi2 eds5, ssi2 eds1 pad4, and ssi2 eds1 eds5 plants. (B) Microscopy of trypan blue-stained leaves from indicated genotypes. (C) Expression of PR-1 gene in indicated genotypes. Total RNA was extracted from 4-week-old plants and used for RNA gel-blot analysis. Ethidium bromide staining of rRNA was used as the loading control. (D) Endogenous SA levels in the leaves of 4-week-old soil-grown plants. Values are presented as mean of three replicates and the error bars represent SD. Statistical significance was determined using Students t-test. Asterisks indicate data statistically significant compared to SSI2 (Col-0) plants (P<0.05, n = 5). (E) Endogenous SAG levels in the leaves of 4-week-old soil-grown plants. Values are presented as mean of three replicates and the error bars represent SD. Asterisks indicate data statistically significant compared to SSI2 (Col-0) plants (P<0.05, n = 5). (F) RT–PCR analysis of R genes in indicated genotypes. The level of β-tubulin was used as an internal control to normalize the amount of cDNA template. The SSI2 EDS1, SSI2 PAD4, SSI2 EDS1 PAD4, and SSI2 EDS1 EDS5 F2 plants showed wt–like morphology, accumulated basal levels of SA and showed basal level expression of PR-1 and R genes (data not shown).

Figure 9. Models for signaling induced by low 18∶1 fatty acid levels and R genes.

(A) EDS1 and SA function upstream of R genes and regulate expression of R genes induced by low 18∶1 fatty acid levels. Mutations in EDS1 and SA-synthesizing enzyme, encoded by SID2, abolish constitutive upregulation of R genes and associated enhanced resistance in genetic backgrounds containing low 18∶1 levels. Similar to EDS1/SA, restored in defective crosstalk (RDC) 2 acts downstream of signaling induced by low levels of 18∶1 but upstream of R gene expression. Signaling induced by low 18∶1 fatty acid levels can also be suppressed by mutations in ACT1-encoded G3P acyltransferase , GLY1-encoded G3P dehydrogenase , or ACP4-encoded acyl carrier protein 4 , which normalize 18∶1 levels, or by blocking steps downstream of R gene expression (rdc3 and rdc4). Upregulation of R genes induced by low 18∶1 fatty acid levels does not require PAD4, SAG101, or EDS5, which are components of the resistance signaling pathway(s) initiated upon R-Avr interaction. (B) Direct or indirect interaction between host-encoded R and pathogen-encoded Avr products initiate resistance signaling, which requires EDS1 and SA. Exogenous application of SA induces expression of PR-1 and EDS1 genes but overexpression of EDS1 does not induce PR-1 expression or increase SA levels. The EDS1– and SA–dependent pathways have additive effects.