The phytopathogenic fungus Sclerotinia sclerotiorum detoxifies plant glucosinolate hydrolysis products via an isothiocyanate hydrolase

Abstract

Brassicales plants produce glucosinolates and myrosinases that generate toxic isothiocyanates conferring broad resistance against pathogens and herbivorous insects. Nevertheless, some cosmopolitan fungal pathogens, such as the necrotrophic white mold Sclerotinia sclerotiorum, are able to infect many plant hosts including glucosinolate producers. Here, we show that S. sclerotiorum infection activates the glucosinolate-myrosinase system, and isothiocyanates contribute to resistance against this fungus. S. sclerotiorum metabolizes isothiocyanates via two independent pathways: conjugation to glutathione and, more effectively, hydrolysis to amines. The latter pathway features an isothiocyanate hydrolase that is homologous to a previously characterized bacterial enzyme, and converts isothiocyanate into products that are not toxic to the fungus. The isothiocyanate hydrolase promotes fungal growth in the presence of the toxins, and contributes to the virulence of S. sclerotiorum on glucosinolate-producing plants.

Fig. 1. A. thaliana aliphatic ITC- and GL-deficient mutants are hyper-susceptible to S. sclerotiorum.

a Comparison of lesion areas caused by S. sclerotiorum 24 h post inoculation (hpi) on leaves of A. thaliana Col-0 wild-type, tgg1/tgg2 and myb28/myb29 mutants. The tgg1/tgg2 line is a myrosinase-defective mutant, and the myb28/myb29 line is deficient in aliphatic GL biosynthesis. b Relative quantification of S. sclerotiorum growth on A. thaliana lines for 24 h as quantified by qRT-PCR. The S. sclerotiorum Histone mRNA was normalized by the A. thaliana ACTIN2 mRNA to quantify relative fungal colonization. Data represent mean ± SEM (n = 6 inoculated leaves and plants, respectively; inoculation with detached leaves was repeated once) and were analyzed by one-way ANOVA (p < 0.001) followed by Tukey’s post-hoc test. Different letters above the bars indicate significant differences at p < 0.05. Source data are provided as a Source data file.

Fig. 2. 4MSOB-ITC is degraded by the fungus S. sclerotiorum.

a Quantification of 4MSOB-ITC in A. thaliana plants with and without S. sclerotiorum. Data represent mean ± SEM (n = 4 inoculated plants) and were analyzed by a Kruskal–Wallis rank sum test (p < 0.01) followed by a Games–Howell post-hoc test. Different letters above the bars indicate significant differences at p < 0.05. b Quantification of 4MSOB-ITC in fungal cultures during a time course. Twenty-five micromolar 4MSOB-ITC was used for each fungal culture. Data were analyzed by a linear regression (R² = 0.94, p < 0.001). Quantification of c 4MSOB-ITC mercapturic acid pathway conjugates, and d 4MSOB-ITC hydrolytic degradation products 4MSOB-amine and 4MSOB-acetamide, in the fungus-inoculated liquid medium supplemented with 4MSOB-ITC. Data represent mean ± SEM (n = 3 independent fungal cultures) and were analyzed by one-way ANOVA (p < 0.001) followed by Tukey’s post-hoc test. Different letters above the bars indicate significant differences at p < 0.05. e Proposed pathways for the degradation of 4MSOB-ITC by S. sclerotiorum. 4MSOB-ITC, 4-methylsulfinylbutyl isothiocyanate; 4MSOB-GSH, 4MSOB-ITC glutathione conjugate; 4MSOB-CG, 4MSOB-ITC cysteinylglycine conjugate; 4MSOB-CYS, 4MSOB-ITC cysteine conjugate; 4MSOB-NAC, 4MSOB-ITC N-acetylcysteine conjugate; 4MSOB-amine, 4-methylsulfinylbutylamine; 4MSOB-acetamide, 4-methylsulfinylbutylacetamide. Source data are provided as a Source data file.

Fig. 3. Ss12040 is a SaxA candidate gene in S. sclerotiorum.

a Maximum likelihood tree of fungal metallo-β-lactamase-like enzymes with similarity to known bacterial ITCases. Amino acid sequences were aligned with MEGA, and phylogenetic trees were constructed with PhyML using 1000 bootstraps. Numbers on each branch in the ML tree signify bootstrap values. b Induction of SaxA candidate gene Ss12040 expression by 4MSOB-ITC in vitro. The fungal culture medium was supplemented with 4MSOB-ITC (25 µM), with only ethanol being added to the control medium. Gene expression was determined 30 min post inoculation. Data represent mean ± SEM (n = 3 independent fungal cultures; in vitro induction of Ss12040 by 4MSOB-ITC was repeated once) and were analyzed by a two-tailed Student’s t test (**p < 0.01). c Expression of SaxA candidate gene Ss12040 24 hpi of different A. thaliana lines. Data represent mean ± SEM (n = 6 independent inoculated plants) and were analyzed by a Kruskal–Wallis rank sum test (p < 0.001) followed by a Games–Howell post-hoc test. Different letters above the bars indicate significant differences at p < 0.05. 4MSOB-ITC, 4-methylsulfinylbutyl isothiocyanate. Source data are provided as a Source data file.

Fig. 4. Enzyme kinetic analyses of heterologously expressed SsSaxA with 4MSOB-ITC and 2PE-ITC.

4MSOB-amine was analyzed as its FMOC (fluorenylmethoxycarbonyl chloride) derivative to avoid signal quenching by the buffer. Products in each reaction were measured 5 min after the enzyme was added. Data represent mean ± SEM (n = 3 independent reactions). 4MSOB-ITC, 4-methylsulfinylbutyl isothiocyanate; 4MSOB-amine, 4-methylsulfinylbutylamine; 2PE-amine, 2-phenylethylamine; 2PE-ITC, 2-phenylethyl isothiocyanate. Source data are provided as a Source data file.

Fig. 5. 2PE-ITC but not its metabolites 2PE-amine and 2PE-acetamide reduces S. sclerotiorum growth.

S. sclerotiorum was grown on PDA plates amended with different concentrations of these three metabolites and the diameter of fungal radial growth was measured 24 h after co-incubation with these compounds. Data represent mean ± SEM (n = 3 independent fungal cultures; growth assay was repeated once) and were analyzed by fitting a linear model using ANCOVA (2PE-ITC/DMSO, 2PE-ITC/2PE-Amine and 2PE-ITC/2PE-Acetamide, ***p < 0.001). 2PE-ITC, 2-phenylethyl isothiocyanate; 2PE-Amine, 2-phenylethylamine; 2PE-Acetamide, 2-phenylethylacetamide. Source data are provided as a Source data file.

Fig. 6. SsSaxA deletion reduces ITC degradation efficiency and growth in ITC-containing medium.

a Quantification of 4MSOB-ITC in both wild-type (WT) and ΔSsSaxA-1 cultures over a time course. Twenty-five micromolar 4MSOB-ITC was used for each culture. Data represent mean ± SEM (n = 3 independent fungal cultures) and were analyzed by one-way ANCOVA (***p < 0.001 for WT/ΔSsSaxA-1). Quantification of the degradation products b 4MSOB-amine and 4MSOB-acetamide and c 4MSOB-ITC conjugates in WT and ΔSsSaxA-1 cultures supplemented with 4MSOB-ITC. For degradation products, data were analyzed by a Kruskal–Wallis rank sum test (p < 0.01) followed by a Games–Howell post-hoc test; for conjugates, data were analyzed by one-way ANOVA (p < 0.001) followed by Tukey’s post-hoc test. Different letters above the bars indicate significant differences at p < 0.05. Deletion of SsSaxA in S. sclerotiorum resulted in reduced growth in culture medium in the presence of both d 4MSOB-ITC and e 2PE-ITC. Diameter of fungal radial growth was measured 24 h after co-incubation with these compounds. Data represent mean ± SEM (n = 3 independent fungal culture; growth assays of WT and ΔSsSaxA mutants with ITCs were repeated once) and were analyzed by one-way ANCOVA (4MSOB-ITC or 2PE-ITC treatment, ***p < 0.001 for WT/ΔSsSaxA-1). 4MSOB-ITC, 4-methylsulfinylbutyl isothiocyanate; 4MSOB-GSH, 4MSOB-ITC glutathione conjugate; 4MSOB-CG, 4MSOB-ITC cysteinylglycine conjugate; 4MSOB-CYS, 4MSOB-ITC cysteine conjugate; 4MSOB-NAC, 4MSOB-ITC N-acetylcysteine conjugate; 4MSOB-amine, 4-methylsulfinylbutylamine; 4MSOB-acetamide, 4-methylsulfinylbutylacetamide; 2PE-ITC, 2-phenylethyl isothiocyanate. Source data are provided as a Source data file.

Fig. 7. ΔSsSaxA mutants are significantly less virulent on A. thaliana Col-0 than the wild-type fungus.

a Representative images of A. thaliana Col-0 leaves infected by WT and ΔSsSaxA mutants. b Comparison of lesion area caused by WT and ΔSsSaxA mutants 24 hpi on leaves of A. thaliana. Data represent mean ± SEM (n = 9 inoculated leaves from separate plants). c Relative growth of WT and ΔSsSaxA mutants on A. thaliana Col-0 as determined by qRT-PCR. The fungal Histone mRNA was normalized to the A. thaliana ACTIN2 mRNA to quantify the relative fungal colonization 24 hpi. Data represent mean ± SEM (n = 6 independent inoculated plants). Data were analyzed by one-way ANOVA (p < 0.001) followed by a Tukey’s post-hoc test. Different letters above the bars indicate statistically significant differences at p < 0.05. Source data are provided as a Source data file.