A conserved enzyme of smut fungi facilitates cell-to-cell extension in the plant bundle sheath

Abstract

Smut fungi comprise one of the largest groups of fungal plant pathogens causing disease in all cereal crops. They directly penetrate host tissues and establish a biotrophic interaction. To do so, smut fungi secrete a wide range of effector proteins, which suppress plant immunity and modulate cellular functions as well as development of the host, thereby determining the pathogen's lifestyle and virulence potential. The conserved effector Erc1 (enzyme required for cell-to-cell extension) contributes to virulence of the corn smut Ustilago maydis in maize leaves but not on the tassel. Erc1 binds to host cell wall components and displays 1,3-β-glucanase activity, which is required to attenuate β-glucan-induced defense responses. Here we show that Erc1 has a cell type-specific virulence function, being necessary for fungal cell-to-cell extension in the plant bundle sheath and this function is fully conserved in the Erc1 orthologue of the barley pathogen Ustilago hordei.

Fig. 1. Erc1 is a virulence factor that is involved in cell-to-cell extension in maize.

a Disease symptoms caused by Ustilago maydis SG200, SG200Δerc1 mutant, SG200Δerc1/C strain, and SG200Δerc1/Erc1^M2x on Early Golden Bantam (EGB) maize leaves at 12 days post inoculation (dpi). Disease rates are given as a percentage of the total number of infected plants. n: indicates total number of infected maize seedlings in three independent biological experiments and letters above bars indicate significant differences (One-way ANOVA followed by Tukey multiple comparison test was performed, p < 0.05). b Quantification of cell-to-cell penetration efficiency of SG200, SG200Δerc1 and SG200Δerc1/C complementation strains. The graph depicts the percentage of trapped U. maydis hyphae in maize bundle sheath cells at 4 dpi. n: indicates total number of counted infected maize cells in three independent biological experiments. Asterisks above bars indicate significant differences (p < 0.05, Chi-square test). Data are presented as mean value ± SD. c Microscopic observation of trapped U. maydis SG200 and SG200Δerc1 hyphae in maize bundle sheath cells at 4 dpi via WGA-AF488/Propidium iodide staining. WGA-AF488 (green color -fungal cell wall): excitation at 488 nm and detection at 500–540 nm. PI (red color - plant cell wall): excitation at 561 nm and detection at 580–630 nm. Similar results were observed at least in three independent biological experiments. Calculated p values are shown in the Source Data.

Fig. 2. Erc1 is functionally conserved in smut fungi.

a The carbohydrate-binding module (CBM) is required for full function of Ustilago maydis Erc1. Disease assay was performed for U. maydis SG200, SG200Δerc1, SG200Δerc1/C strain, and SG200Δerc1/Erc1ΔCBM on EGB maize at 12 days post inoculation (dpi). Disease rates are given as a percentage of the total number of infected plants. n: indicates total number of infected maize seedlings in three independent biological experiments. Asterisks above bars indicate significant differences (p < 0.05, two-tailed student’s t-test). b Complementation of SG200Δerc1 mutant with Ustilago hordei UhErc1. Disease assay was performed for U. maydis SG200, SG200Δerc1, SG200Δerc1/C, and SG200Δerc1/UhErc1 strains on EGB maize plants at 12 days post inoculation (dpi). n: indicates total number of infected maize seedlings in three independent biological experiments. Asterisks above bars indicate significant differences (p < 0.05, two-tailed student’s t-test). c UhErc1 is a virulence factor during barley infection. Disease assay was performed with U. hordei DS199 and DS199Δerc1 mutant strains on 13-day-old barley seedlings. Pictures were taken at 8 dpi. Similar results were observed in three independent biological experiments. d Quantification of fungal biomass of DS199, DS199Δerc1 mutant, and DS199Δerc1/C complementation strains on barley at 8 dpi. qPCR was performed to determine fungal biomass by using gDNA that was isolated from U. hordei DS199, DS199Δerc1 mutant, and DS199Δerc1/C infected barley leaves. Three independent biological replicates were performed with total number of 15 infected barley seedlings for each experiment. Data were presented as mean value ± SD. Asterisks above bars indicate significant differences (p < 0.05, two-tailed student’s t-test). e Microscopic observation of trapped U. hordei DS199 and DS199Δerc1 mutant hyphae in barley bundle sheath cells at 8 dpi via WGA-AF488/Propidium iodide staining. WGA-AF488 (green color—fungal cell wall): excitation at 488 nm and detection at 500–540 nm. PI (red color—plant cell wall): excitation at 561 nm and detection at 580–630 nm. Similar results were observed in three independent biological experiments. f Quantification of cell-to-cell penetration efficiency of U. hordei DS199, DS1990Δuherc1 mutant and DS199Δerc1/C strains. The graph depicts the percentage of trapped U. hordei hyphae in barley bundle sheaths cells at 8 dpi. n: indicates total number of counted infected barley cells in three independent biological experiments. Data were presented as mean value ± SD. Asterisks above bars indicate significant differences (p < 0.05, Chi-square test). Calculated p values were presented in Source Data.

Fig. 3. Localization of Erc1 in Ustilago maydis SG200 during maize colonization.

a Erc1-mCherry was heterologously expressed in U. maydis SG200 strain under control of the native promotor with predicted native signal peptide for extracellular secretion. The SG200 strains expressing the Erc1-mCherry, UmPit2-mCherry (as a positive control for secretion) and cytosolic mCherry (int. mCherry; as a negative control for secretion) were inoculated on maize seedlings and at 4 dpi confocal microscopy was performed to monitor the localization of each recombinant protein. While both Erc1-mCherry and UmPit2-mCherry are secreted around the tip of the invasive hyphae, internal mCherry localizes to the fungal cytoplasm. The white graphs indicate the mCherry signal intensity along the diameter of the hyphae (illustrated by white lines in the image). White arrowheads indicate fungal hyphal tips and yellow arrowhead indicates apoplastic fluid after plasmolysis. Plasmolysis was performed with 1 M NaCl solution. Similar results were observed in three independent biological experiments. b Transmission electron micrographs after immunogold labeling of secreted Erc1-HA and GFP-HA with a monoclonal antibody recognizing HA epitopes. White arrowheads pointing to black dots indicate secretion of Erc1-HA to the biotrophic interface. FCW: Fungal cell wall, H: Hyphae, P: Plant cell cytoplasm, PCW: Plant cell wall. Similar results were observed in multiple transmission electron micrographs.

Fig. 4. Functional characterization of the Erc1 protein.

a Purification of His-Erc1-Myc-His recombinant protein. Sypro Ruby staining was performed to visualize the Erc1 recombinant protein. Western blot (WB) analysis was performed with α-His- and α-Myc- specific antibody to detect His-Erc1-Myc-His protein. While two bands were detectable in Sypro Ruby staining and WB performed with α-His, only one band was detectable with α-Myc-specific antibody. Similar results were observed at least in two independent biological replicates. b α-L-arabinofuranosidase activity assay with 4-nitrophenyl α-L-arabinofuranoside (4NPA) substrate. A commercial α-L-arabinofuranosidase from Aspergillus niger (AFASE) was used as a positive control. Data are presented as mean value of three independent biological experiments. c Activity-based protein profiling (ABPP) assay for Erc1. The Erc1, Erc1^M2x and AFASE recombinant proteins were incubated with the specific α-L-arabinofuranosidase inhibitor DL69. Plus (+) and minus (−) indicate the addition and the absence of the inhibitor, respectively. α-L-arabinofuranosidase specific probe ME868 was added as indicated (+/-). The probe was detected by scanning the in-gel fluorescence with Cy5 filter (Ex. 650 nm, Em. 670 nm). Protein of loaded samples were visualized via Sypro Ruby (Ex. 450 nm, Em. 610 nm). NP: no probe control. Similar results were observed in two independent biological experiments. d Complementation of SG200Δerc1 mutant with Afg2 and Afg3 under the native Erc1 promoter. Disease assay was performed for Ustilago maydis SG200, SG200Δerc1 mutant, SG200Δerc1/C, SG200Δerc1/Afg2 and SG200Δerc1/Afg3 strains on EGB maize plants at 12 days post inoculation (dpi). Disease rates are given as a percentage of the total number of infected plants. n: indicates total number of infected maize seedlings in three independent biological replicates. Letters above bars indicate significant differences (One-way ANOVA followed by Tukey multiple comparison test was performed, p < 0.05). e Thin layer chromatography (TLC) assay was performed to demonstrate β−1,3-glucanase activity of Erc1 on laminarin and laminarihexaose substrate. Erc1, Erc1^M1x, and Erc1^M2x recombinant proteins were incubated with laminarin and laminarihexaose. Samples were loaded on TLC Silica gel 60 F₂₅₄ plate. Glucose + Sucrose mix was used as reference. Laminarin, laminarihexaose and their hydrolysis products were visualized by spraying the TLC plate with detection solution. Arrowheads indicate released of hydrolyzed products. Similar results were observed in three independent biological experiments. f The signal intensity of bands representing glucose was quantified by using ChemiDoc Bio-Rad imaging machine. Data are presented as mean value ± SD of three independent biological experiments and asterisks above bars indicate significant differences (p < 0.05, two-tailed student’s t-test). Calculated p values were presented in Source Data.

Fig. 5. Smut Erc1 prevents induction of host defenses.

a A ROS-burst assay was performed with barley leaf disks following incubation with laminarihexaose, Erc1, Erc1^M2x, and a mix of laminarihexaose and Erc1 recombinant proteins. BSA was used as a negative control for background signal. Relative luminescence units (RLU) indicate ROS burst activity of treated barley leaf discs. The RLU are normalized with the buffer control. Each curve shows the mean of at least nine technical leaf discs with three independent biological experiments. Data are presented as mean value ± SD. b Statistical analysis was performed with sum of RLU values of each sample. Data are presented as mean value ± SD. Asterisks above bars indicate significant differences (p < 0.05, two-tailed student’s t test). c RT-qPCR for pathogenesis related (PR) gene expression on SG200 and SG200∆erc1 mutant strains infected maize leaves at 4 days post infection (dpi). The expression levels of maize PR genes, including PR1, PR3, PR4, PR5, and PRm6b were calculated relative to the GAPDH gene of maize. Data are presented as mean value ± SD. Asterisks above bars indicate significant differences (p < 0.05, pairwise Kruskal-Wallis H-test). Calculated p values are shown in the Source Data.

Fig. 6. Cell type-specific function of Erc1.

a Microscopic observation of trapped Ustilago maydis infected maize leaf and tassel tissues via WGA-AF488/Propidium iodide staining. WGA-AF488 (green color indicates fungal cell wall): excitation at 488 nm and detection at 500-540 nm. PI (red color indicates bundle sheath cells): excitation at 561 nm and detection at 580–630 nm. Unlike leaf tissue, no bundle sheaths cells are detectable in tassel tissue. b Protoplastation of maize leaf cells. Pictures were taken after treatment with plant cell wall degrading enzymes. While the used enzyme mix has the ability to convert mesophyll cells into protoplasts, bundle sheath cells remain intact. c Schematic depiction of the biotrophic interface of host-smut interaction. Fungus-derived proteins are depicted in gray. During host colonization, 1,3-β-glucan molecules are accumulated at the biotrophic interface and Erc1 may hydrolyze these 1,3-β-glucans to prevent the accumulation and subsequent recognition of DAMP molecules by the host plant. d While wild-type smut fungi have the ability to move from cell-to-cell in bundle sheath cells, Δerc1 mutants do not have the ability to fully suppress host immunity leading to the described cell-arrest phenotype (Modified from Fig. 1c). e, f Aniline blue staining with maize leaf cross-section for detection of callose. Intensity of aniline blue signal is depicted via graph (e). Orange bars indicate bundle sheath cell wall, green bars indicate mesophyll cell wall. Similar results were observed in two independent biological experiments.