3.3 Role of jasmonic acid in Arabidopsis/Verticillium longisporum interaction
3.3.3 Expression analysis of known defense genes of the JA pathway in dde2-2 and coi1-t
Since disease development depends on COI1 but not on plant-derived JAs, jasmonate levels were determined (measurement done by the department of Prof. Feussner) in all three genotypes at 15 dpi in mock and infected petioles (Figure 15A). JA levels were increased significantly in infected wild type plants and were absent in dde2-2. In coi1-t, JA levels were slightly elevated in mock-infected plants and did not show a significant increase after infection.
Like JA levels, levels of the active hormone JA-Ile were increased in wild-type petioles and was absent in the dde2-2 mutant. The coi1-t mutant had increased JA-Ile levels already after mock infection and reacted to the fungus with a further increase. The lack of any biochemically detectable JA or JA-Ile in the infected dde2-2 mutant suggested that V. longisporum cannot synthesize JA. In order to investigate whether V. longisporum might produce a yet unknown JA mimic to activate COI1, transcript levels of two marker genes of the JA-Ile-dependent COI1
response, namely VSP2 and PDF1.2, were determined in petioles at 15 dpi,. Both genes were only induced in wild-type plants (Figure 15B) indicating that no fungal-derived JAs or JA mimics that would activate the established COI1-dependent defense genes are effective in V.
longisporum-infected dde2-2 plants. Consistent with the result those similar amounts of JA-Ile were found in wild-type plants after V. longisporum infection and wounding (Figure 11A and B), VSP2 transcript levels were induced to comparable levels under both conditions. In contrast, the JA/ET marker gene PDF1.2, which is highly expressed after infection with the foliar pathogen Botrytis cinerea, is not efficiently induced in V. longisporum-colonized plant tissue. The observed increase in ABA (Figure 15A), which is known to inhibit the JA/ET pathway, might explain the low PDF1.2 transcript levels (Anderson et al., 2004).
Figure 15 Activation of JA biosynthesis and signaling pathways in V. longisporum-infected wild-type, dde2-2 and coi1-t plants
(A) HPLC-MS/MS analysis for detection of JA-, JA-Ile and ABA levels in petioles from wild-type dde2-2 and coi1-t plants at 15 days after mock and V. longisporum infection. Data are the means (+/- SEM) of eight replicates from two independent experiments. Each replicate is a pool of four plants.
0.00
(B) Quantitative RT-PCR analysis of relative VSP2 and PDF1.2 transcript levels in petioles from wild-type dde2-2 and coi1-t plants at 15 days after mock and V. longisporum infection. Data indicate means (+/- SEM) of three independent experiments with 16 individual plants/experiment. Wounded petioles were harvested for RNA extraction after two hours (three biological replicates), Botrytis cinerea-infected samples were harvested after three days (four biological replicates). Relative transcript levels of the infected wild-type were set to 100%. Different letters denote significant differences between samples (one-way ANOVA followed by Tukey-Kramer multiple comparison test; P <
0.05 for (a) and (b) PDF1.2; P < 0,001 for (b) VSP2 and (c).
3.3.4 The role of salicylic acid defense pathway in the coi1-mediated tolerance
Increased resistance of the coi1 mutant has been detected before in a screen for resistance against the hemibiotrophic pathogen Pseudomonas syringae (Kloek et al., 2001). In this interaction, the bacterial JA-Ile mimic COR activates COI1 to suppress the SA pathway (Kloek et al., 2001, Laurie-Berry et al., 2006). To analyze, whether a similar scenario would explain the coi1-mediated tolerance towards V. longisporum, SA synthesis and SA signaling were analyzed in infected wild-type, dde2-2 and coi1-t plants (Figure 16A). Free SA levels did not increase after infection in wild-type plants and reached similar levels in all three infected genotypes.Lower basal SA levels were detected in dde2-2. In contrast, the SA glucoside (SAG) and the SA-derived metabolite dihydroxybenzoeic acid (DHBA) were elevated in all three genotype after infection. Like the relative levels of SAG and DHBA, which showed the highest values in the wild-type followed by intermediate levels in the dde2-2 mutant and even lower levels in coi1-t, PR-1 expression followed the same pattern (Figure 16B). However a slight increase in the SAG and DHBA levels was observed in mock-treated coi1-t plants.
Figure 16 Activation of SA biosynthesis and signaling pathways in V. longisporum-infected wild-type, dde2-2 and coi1-t plants
(A) HPLC-MS/MS analysis for detection of SA-, SAG and DHBA levels in petioles from wild-type, dde2-2 and coi1-t plants at 15 days after mock and V. longisporum infection. Data are the means (+/- SEM) of eight replicates from two independent experiments. Each replicate is a pool of four plants (same material as in Fig. 4).
(B) Quantitative RT-PCR analysis of relative PR-1 transcript levels in petioles from wild-type dde2-2 and coi1-t plants at 15 days after mock and V. longisporum infection. Data indicate means (+/- SEM) of three independent experiments with 16 individual plants/experiment (same material as in Fig. 4). Pseudomonas syringae pv. maculicola ES4326/avrRps4-infected leaf samples were harvested after three days (three biological replicates). Relative transcript levels of the V.l.-infected wild-type were set to 100%.
Different letters denote significant differences between samples (one-way ANOVA followed by Tukey-Kramer multiple comparison test; P < 0.01 for (a), SA and SAG; P < 0,05 for (a), DHBA and (b).
To investigate whether this slight increase in the SAG and DHBA levels confers resistance in coi1 plants, coi1-1/nahG double mutant plants were checked for susceptibility. coi1-1/nahG plants have also been previously known to lack induced expression of PR1 (Kloek et al., 2001).
Like coi1-t, coi1-1/nahG, nahG, coi1-1 and Col-gl were inoculated with a V. longisporum spore suspension. In two independent experiments, the leaf areas of mock and infected plants were measured at 15 dpi (Figure 17). coi1-1 showed similar resistance to the fungus like coi1-t but
type in one experiment (Figure 17A). In the second experiment, coi1/nahG showed no reduction in the leaf area after infection as compared to the wild type (Figure 17B). Since catechol, produced as a result of SA hydrolyses by NahG enzyme, might alter the susceptibility an independent crossing between coi1-1 and sid2-2 was initiated and analyzed. coi1-1/sid2-2 plants also showed reduced disease symptoms like coi1-1 plants (Figure 17C). These results indicate that the resistant/tolerant coi1 phenotype is not due to pre-induction of the SA pathway.
Figure 17 Disease phenotype of V. longisporum-infected wild type, nahG, coi1-1 and coi1-1/nahG plants
(A) Projected leaf area of mock- and V. longisporum-infected wild-type, nahG, coi1-1 and coi1-1/nahG plants first experiments.
(B) Projected leaf area of mock- and V. longisporum-infected wild-type, nahG, coi1-1 and coi1-1/nahG plants from second experiments.
(C) Projected leaf area of mock- and V. longisporum-infected wild-type, coi1-1, sid2-2 and coi1-1/sid2-2 plants from a single experiment at 19 dpi.
Data indicate means (+/- SEM) of 16-18 replicates. Stars indicate significant differences at P < 0.0001 (two-way ANOVA followed by Bonferroni multiple comparison test; ns, not significant) between V. longisporum - and mock-infected samples.