2.1 Systematic identification of the nucleoporin family in F. graminearum and expression analysis during wheat infection

A systematic identification of nucleoporins (FgNups) in the F. graminearum genome was performed. Using Saccharomyces cerevisiae and Neurospora crassa nucleoporins as a reference, NCBI BlastP analysis identified 24 non-redundant candidate FgNups (Table S1 and S2 in S1 Appendix). Conserved domain analysis revealed that these FgNups proteins lack a consistent common domain but share certain specific domains among some members (S1 Fig). To further elucidate their evolutionary relationships, a phylogenetic tree containing ScNups and FgNups was constructed. The results showed that this protein family can be divided into six distinct clades (Fig 1A). Total of 20 nucleoporins, cluster on the same evolutionary branch as their homologues in S. cerevisiae, indicating that their functions may be relatively conserved in fungi. FgNup188, FgNup2, FgNup159 and FgNup82 are distributed on different branches from their homologues in S. cerevisiae, suggesting that they may have undergone functional differentiation or possess species-specific traits during evolution. To explore the potential functions of FgNups during infection, their expression patterns were analyzed from 0 to 7 d post-inoculation of wheat. Transcriptome data revealed that 10 FgNups genes were specifically upregulated during the infection stages (Fig 1B and S2 Fig). Subsequently, knockout mutants for these 10 genes (S3 Fig) were constructed and preliminarily assessed for vegetative growth phenotype (Fig 1C, D) and pathogenicity on wheat spikes (Fig 1E-G). Phenotypic analysis showed that among the 10 mutants, only the ΔFgNup2 mutant exhibited significant defects in hyphal growth and a marked reduction in pathogenicity.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 1. Systematic identification of the nucleoporin family in F.graminearum and analysis of their expression during wheat infection.

(A) Phylogenetic analysis of F. graminearum and S. cerevisiae Nups. The phylogenetic tree was obtained using MEGA-X. F. graminearum: Fg, purple square, S. cerevisiae: Sc, orange pentagon. (B) Analysis of FgNups expression levels during wheat infection from 0-7 d. (C) Morphology of PH-1, ΔFgNup2, ΔFgNup53, ΔFgNup133, ΔFgSeh1, ΔFgSec13, ΔFgNup192, ΔFgNup82, ΔFgPom34, ΔFgPom152, and ΔFgNic96 after 72 h of growth on PDA, CM, and MM media. (D) Colony diameters of the above strain on three different media after 3 d of growth, the data come from three independent repeated experiments, and significant differences are indicated by * (****P ≤ 0.0001). (E) Pathogenicity of the above strain on field wheat spikes. (F) Pathogenicity of the above strain on wheat coleoptiles. (G) Statistical analysis of lesion length on wheat coleoptiles infected by the above mutant strains, the data come from three independent repeated experiments, and significant differences are indicated by * (****P ≤ 0.0001).

https://doi.org/10.1371/journal.ppat.1014577.g001

2.2 FgNup2 regulates development, stress response, pathogenicity, and deoxynivalenol (DON) biosynthesis in F. graminearum

The deletion of FgNup2 resulted in significant hyphal growth defects, which were fully restored in the complemented strain (Fig 2A, B). Compared to the wild-type PH-1 and the complementation strain, the ΔFgNup2 mutant produced conidia with significantly altered morphology, characterized by increased length and decreased diameter (Fig 2C). Furthermore, the number of conidia produced by the mutant was significantly reduced (Fig 2D). In the wild-type strain PH-1, each compartment of the conidia usually contains only one nucleus (occasionally two). However, in the ΔFgNup2 mutant, we observed the abnormal presence of multiple nuclei in each compartment. Furthermore, nuclei in PH-1 hyphae are evenly distributed, while in the ΔFgNup2 mutant, nuclei often cluster together, suggesting abnormal nuclear division in this mutant. These results indicate that FgNup2 is crucial for maintaining normal nuclear distribution and division(Fig 2E). The ΔFgNup2 mutant exhibited enhanced tolerance to cell wall and cell membrane stressors (Congo Red and SDS) but increased sensitivity to osmotic and metal ion stress (Fig 2F, G). Pathogenicity assays were conducted on wheat spikes and coleoptiles using equal concentrations of conidial suspensions. Inoculation with the ΔFgNup2 mutant failed to produce any disease symptoms compared to PH-1 and ΔFgNup2-C (Fig 2H, I). DON production was quantified. The amount of DON produced by the ΔFgNup2 mutant was significantly lower than that of the wild-type strain (Fig 2J).

Download:

• PNG
  larger image
• TIFF
  original image

Fig 2. The nucleoporin FgNup2 regulates development, stress response, pathogenicity, and DON biosynthesis in F. graminearum.

(A) Colony morphology of PH-1, ΔFgNup2 and ΔFgNup2-C grown on PDA, CM and MM media for 72 h. (B) Colony diameter statistics of PH-1, ΔFgNup2 and ΔFgNup2-C grown on PDA, CM and MM media for 72 h. The data come from three independent repeated experiments, and significant differences are indicated by * (****P ≤ 0.0001) (C) ΔFgNup2 affects conidia number. CFW staining, DIC morphology observed under a fluorescence microscope, Bar = 20 μm. (D) Conidia production of PH-1, ΔFgNup2, and ΔFgNup2-C after 4 d of cultivation on CMC medium. (E) The nucleoporin FgNup2 is involved in regulating mitosis in F. graminearum. (F) The morphology of PH-1, ΔFgNup2, and ΔFgNup2-C colonies grown on PDA medium containing different stress agents. (G) Hyphal growth inhibition rates of PH-1, ΔFgNup2, and ΔFgNup2-C strains grown under stress conditions after 3 d, the data come from three independent repeated experiments, and significant differences are indicated by * (****P ≤ 0.0001). (H) Pathogenicity of PH-1, ΔFgNup2, and ΔFgNup2-C on wheat ears and wheat coleoptiles. (I) Statistics of wheat coleoptiles lesion length, the data come from three independent repeated experiments, and significant differences are indicated by * (****P ≤ 0.0001). (J) The amount of DON (per g dry weight) produced by the wild-type PH-1, ΔFgNup2 and ΔFgNup2-C in trichothecene biosynthesis-inducing (TBI) medium after 48 h inoculation, the data come from three independent repeated experiments, and significant differences are indicated by * (****P ≤ 0.0001).

https://doi.org/10.1371/journal.ppat.1014577.g002

2.3 FgNup2 interacts with the ubiquitin ligase FgBre1

RNA sequencing was performed on the wild-type strain PH-1 and the ΔFgNup2 mutant cultured for 48 h in trichothecene biosynthesis-inducing (TBI) medium, which contains arginine as the sole nitrogen source. This medium was previously used by Gardiner et al. to investigate the effect of humic acid on DON production [31]. Using a threshold of p-value<0.05 and |log2(fold change)| ≥ 2, we identified a total of 3,807 differentially expressed genes (DEGs) in the ΔFgNup2 mutant compared to PH-1. Among these, 2,075 genes were significantly upregulated and 1,732 were downregulated (Fig 3A). A heatmap of RPKM-normalized transcripts depicting the 3,807 DEGs is shown in S4 Fig. Functional annotation of these DEGs indicated their involvement in various biological processes, including carbohydrate, lipid, nucleotide, and amino acid metabolism. Cell cycle regulation; DNA replication and transcription, post-translational modification. Intracellular transport and the biosynthesis of secondary metabolites, cell wall and cell membrane (Fig 3B and C). An affinity capture approach was employed to identify its interaction partners. The N-terminal region of FgNup2 was fused to GFP and introduced into the ΔFgNup2 mutant. Proteins purified using GFP-trap agarose beads were analyzed by mass spectrometry (MS). By comparing the MS data with the F. graminearum PH-1 database, candidate FgNup2-interacting proteins were identified. Using a confidence threshold of false discovery rate (FDR)≤1% and more than 5 peptide-spectrum matches (PSMs), a list of 300 candidate interacting proteins was obtained. RNA-seq analysis revealed widespread dysregulation of transcription-related genes in ΔFgNup2. Among the 300 FgNup2-interacting candidates identified by affinity capture–mass spectrometry, we prioritized proteins with chromatin-regulatory functions, including FGSG_06769, which encodes the histone H2B ubiquitin ligase FgBre1 responsible for H2B monoubiquitination (H2Bub1) in F. graminearum, as the key candidate for further validation. The physical interaction between FgNup2 and FgBre1 was subsequently confirmed by multiple independent experiments, including yeast two-hybrid (Y2H) analysis, co-immunoprecipitation (Co-IP), GST-pull-down, and bimolecular fluorescence complementation (BiFC) assays (Fig 3D-G).

Download:

• PNG
  larger image
• TIFF
  original image

Fig 3. FgNup2 interacts with the ubiquitin ligase FgBre1.

(A) Volcano plot showing the number of genes with more than twofold expression change in ΔFgNup2. (B) GO enrichment analysis of differential genes. (C) EggNOG functional classification of differentially expressed genes in ΔFgNup2. (D) Yeast two-hybrid (Y2H) experiment verifying the interaction between FgNup2 and FgBre1. (E) Co-immunoprecipitation (Co-IP) experiment verifying the interaction between. (F) GST pull-down experiment verifying the interaction between FgNup2 and FgBre1. (G) Bimolecular fluorescence complementation (BiFC) assays verifying the interaction between FgNup2 and FgBre1.

https://doi.org/10.1371/journal.ppat.1014577.g003

2.4 FgNup2 regulates FgBre1-mediated histone modification

Based on the direct interaction between FgNup2 and FgBre1, we hypothesized that FgNup2 may participate in regulating H2Bub1 modification levels. To test this hypothesis, we examined H2Bub1 abundance in the wild-type strain PH-1 and the ΔFgNup2 mutant by Western blot. Under TBI conditions, H2Bub1 levels were markedly elevated in the wild type but significantly reduced in the ΔFgNup2 mutant compared to PH-1(Fig 4A, B). Notably, FgNup2 deletion did not affect FgBre1 protein expression, as confirmed by Western blot and qRT-PCR analyses (S5 Fig), excluding the possibility that the observed decrease in H2Bub1 resulted from reduced FgBre1 abundance.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 4. FgNup2 regulates FgBre1-mediated histone modification” would be more appropriate.

(A) Western blot experiment to verify the levels of H2Bub1 modification in PH-1 and ΔFgNup2 strains cultured under YEPD or TBI medium conditions. (B) Western blot experiment to verify the levels of H2Bub1 modification in PH-1, ΔFgNup2 and ΔFgNup2-C strains 48 h after under toxin-inducing conditions. (C) FgNup2 promotes the occurrence of H2Bub1 in F. graminearum through FgBre1. H3 was used as the internal control, and the intensity of the Western blot bands was quantified using ImageJ software. The values below the bands indicate the grey value of the detected protein bands relative to the H3 bands. (D) Colony morphology of PH-1, ΔFgBre1, PH-1::Bre1, and PH-1::Bre1^ΔFgNup2 after 72 h of growth on PDA, CM, and MM media. (E) Pathogenicity of PH-1, ΔFgNup2, ΔFgBre1, PH-1::Bre1 and PH-1::Bre1^ΔFgNup2 on wheat heads and wheat coleoptiles. (F) Statistical analysis of lesion lengths on wheat coleoptiles infected by PH-1, ΔFgNup2, ΔFgBre1, PH-1::Bre1 and PH-1::Bre1^ΔFgNup2, the data come from three independent repeated experiments, and significant differences are indicated by * (****P ≤ 0.0001). (G) Statistical analysis of DON toxin production induced in TBI medium for 48 h by PH-1, ΔFgNup2, ΔFgBre1, PH-1::Bre1 and PH-1::Bre1^ΔFgNup2, the data come from three independent repeated experiments, and significant differences are indicated by * (****P ≤ 0.0001, **P ≤ 0.01, ns P ≥ 0.05).

https://doi.org/10.1371/journal.ppat.1014577.g004

To determine whether FgNup2 regulates H2Bub1 through FgBre1, we constructed and compared four strains: the FgBre1 deletion mutant ΔFgBre1, the FgBre1 overexpression strain PH-1::FgBre1, and FgBre1 overexpression in the ΔFgNup2 background(PH-1::FgBre1^ΔFgNup2). Western blot analysis of H2Bub1 levels in these strains (Fig 4C) revealed the following: ΔFgBre1 showed virtually no detectable H2Bub1 signal, consistent with previous reports and confirming that FgBre1 is the essential for this modification. PH-1::FgBre1 exhibited significantly elevated H2Bub1 levels relative to the wild type, demonstrating that FgBre1 overexpression alone is sufficient to enhance H2Bub1, however, in PH-1::FgBre1^ΔFgNup2, despite FgBre1 overexpression, H2Bub1 levels were drastically reduced and comparable to those in the ΔFgNup2 single mutant. Furthermore, the PH-1::FgBre1^ΔFgNup2 strain displayed significantly reduced growth rate, pathogenicity, and DON toxin production compared to both the wild type and the FgBre1 overexpression strain (Fig 4D-G). Collectively, these findings demonstrate that FgNup2 is required for FgBre1-mediated H2Bub1 modification.

2.5 FgNup2 spatiotemporally regulates the nuclear import of FgBre1

Sequence analysis revealed that the N-terminus of FgBre1 contains a conserved putative nuclear localization signal (S6A Fig), suggesting that its nuclear import depends on nuclear pore trafficking. To investigate whether FgNup2 regulates the nuclear transport of FgBre1, we observed the subcellular localization of FgBre1-GFP in the wild-type PH-1 and the ΔFgNup2 mutant under TBI medium induction. In the wild-type strain, FgBre1 rapidly accumulated in the nucleus, with nearly complete nuclear import observed as early as 1 h toxin biosynthesis induction (Fig 5A). In stark contrast, the nuclear import of FgBre1 was significantly delayed in the ΔFgNup2 mutant. At 6 h, FgBre1 was primarily retained in the perinuclear region and failed to efficiently enter the nucleoplasm. Partial nuclear import was detected only at 12 h, and significant nuclear accumulation appeared at 24 h (Fig 5A). This delayed import phenotype was consistently observed in protoplasts prepared from the PH-1::Bre1-GFP and PH-1::Bre1-GFP^ΔFgNup2 strains, indicating it was not an artifact of hyphal morphology (Fig 5B). Subcellular fractionation followed by western blot analysis was performed to assess FgBre1 levels in nuclear and cytoplasmic extracts. The results were consistent with the fluorescence microscopy observations. At early induction stages, the mutant showed significantly reduced nuclear accumulation of FgBre1 compared to the wild-type, while its cytoplasmic levels remained relatively high (Fig 5C), We collected the hyphae of F. graminearum infecting the wheat leaves during the infection process, and observed the subcellular localization of FgBre1 through DAPI staining. The results showed that under natural infection conditions, FgBre1 was mainly located in the nucleus. While in the ΔFgNup2 mutant background, FgBre1 significantly accumulated in the perinuclear region (S6B, S6C Fig). Based on these findings, a schematic diagram illustrating the regulation of FgBre1 nuclear import by FgNup2 was drawn (Fig 5D).

Download:

• PNG
  larger image
• TIFF
  original image

Fig 5. Spatiotemporal regulation of FgBre1 nuclear entry by FgNup2 under toxin-inducing conditions.

(A) The localisation of FgBre1-GFP in the hyphae of PH-1 and ΔFgNup2 strains induced in TBI medium for 1 h, 3 h, 6 h, 12 h, 24 h, Bar = 10 μm, fluorescence intensity analysed with ImageJ software. (B) The localisation of FgBre1-GFP in the protoplasts of PH-1 and ΔFgNup2 strains induced in TBI medium for 1 h, 3 h, 6 h, 12 h, 24 h and 36 h, Bar = 10 μm. (C) Western blot experiment quantifying the content of FgBre1-GFP protein in nuclei and cytoplasm at 1 h, 3 h, 6 h, 12 h, 24 h and 36 h of induction in TBI medium. (D) Schematic diagram of the pattern of FgBre1 nuclear import regulated by FgNup2.

https://doi.org/10.1371/journal.ppat.1014577.g005

2.6 FgNup2 cooperates with FgImpα to regulate the nuclear localization of FgBre1

Fglmpa is a member of an importin protein family widely found in filamentous fungi, usually able to recognise nuclear localisation signals (NLS) and mediate the nuclear import of cargo proteins. Given that FgBre1 contains a predicted NLS, we speculate that Fglmpa might act as its nuclear import adaptor. we first performed a Co-IP assay. The results showed an interaction between FgBre1 and FgImpα under TBI medium conditions (Fig 6A, B). As a ΔFgImpα knockout mutant was unavailable, a competitive inhibition strategy was developed by overexpressing a 3 × Flag-tagged peptide containing a nuclear localization signal (NLS) and a nuclear export signal (NES). This NLS-NES construct acts as a high-affinity competitive cargo for FgImpα. Western blot analysis confirmed the successful expression of the Flag-tagged inhibitor (Fig 6C). In the PH-1/Impα-GFP strain, FgImpα-GFP was primarily localized to the nucleus after 36 h of toxin biosynthesis induction. However, in the presence of the NLS-NES competitor (PH-1/Impα-GFP + NLS), FgImpα-GFP became diffusely distributed throughout the cytoplasm (Fig 6D), indicating effective competition for Impα binding. Next, we examined the effect of this competition on FgBre1 localization. In the PH-1/Bre1-RFP Impα-GFP strain, FgBre1-RFP and FgImpα-GFP showed complete nuclear co-localization. In contrast, in the strain co-expressing the NLS-NES competitor (PH-1/Bre1-RFP+Impα-GFP + NLS), the co-localization of FgBre1 with FgImpα was significantly reduced (Fig 6B). Consistent with this, CO-IP assays showed that the interaction between FgBre1 and FgImpα was significantly weakened upon introduction of the NLS-NES competitor (Fig 6E). Together, these data indicate that the nuclear import of FgBre1 relies on functional FgImpα-mediated transport. Furthermore, fluorescence co-localization analysis revealed that the degree of co-localization between FgBre1 and FgImpα was significantly reduced in the ΔFgNup2 background (Fig 6F). Co-IP assays showed that the binding strength between FgBre1 and FgImpα was significantly lower in the ΔFgNup2 mutant compared to the wild-type strain (Fig 6G).

Download:

• PNG
  larger image
• TIFF
  original image

Fig 6. FgNup2 and FgImpα synergistically regulate the nuclear localisation of FgBre1.

(A) Co-immunoprecipitation (Co-IP) experiment confirmed the interaction between FgBre1 and FgImpα. (B) Effect of competitive inhibitors on the co-localization of FgBre1 with FgImpα. In the PH-1/Bre1-RFP Impα-GFP strain, significant fluorescence co-localization signals of FgBre1 and FgImpα were observed in the nuclear region, whereas in the PH-1/Bre1-RFP Impα-GFP NLS strain, the co-localization of FgBre1 and FgImpα was significantly weakened. Bar = 10 μm. Co-localization and fluorescence intensity were analyzed using ImageJ software. (C) Using Western Blot with Flag antibody to detect the expression of the 3 × Flag-NLS-NES competitive inhibitor vector in the PH-1 background strain. (D) Confocal laser microscopy observation of the effect of the inhibitor on the subcellular localisation of FgImpα-GFP. (E) Co-immunoprecipitation (Co-IP) detection of the effect of the inhibitor on the interaction between FgBre1 and FgImpα. (F) Under under toxin-inducing conditions, laser confocal microscopy was used to observe the fluorescence localisation and colocalisation of FgBre1-RFP and FgImpα-GFP in PH-1 and ΔFgNup2 strains, Bar = 10 μm. (G) Co-immunoprecipitation (Co-IP) experiments were used to detect the interaction between FgBre1 and FgImpα in PH-1 and ΔFgNup2 strains.

https://doi.org/10.1371/journal.ppat.1014577.g006

2.7 FgNup2 regulates development, stress response, and virulence in F. graminearum by promoting H2Bub1

To further investigate the impact of FgNup2 on H2Bub modification, ChIP-seq experiments were performed on PH-1 and ΔFgNup2 strains under TBI medium conditions. To ensure the reliability of H2Bub1 profiling, we performed comprehensive quality control for all ChIP-seq samples (S7 Fig). The sequencing depth was adequate across all ChIP samples (15.4–27.5 million reads). PCA analysis demonstrated tight clustering of biological replicates within each genotype and clear separation between PH-1 and ΔFgNup2, confirming high data reproducibility. RiP% values ranged from 12.3% to 30.5%, consistent with the expected enrichment efficiency for histone modifications. Peak distribution analysis revealed that H2Bub1 peaks were predominantly enriched in gene body regions, with characteristic depletion at transcription start sites (TSS) and transcription termination sites (TTS), matching the established distribution pattern of H2B monoubiquitination in actively transcribed genes [32]. Notably, this gene body enrichment pattern was markedly attenuated in ΔFgNup2, consistent with the global reduction of H2Bub1. Using Input as background and MACS2 software with broad parameters for peak calling on IP samples. ChIP-seq analysis revealed that, compared to the wild-type PH-1, a total of 677 genes with significantly downregulated H2Bub1 enrichment were identified in ΔFgNup2. Intersecting these genes with the 1,732 genes significantly downregulated in the ΔFgNup2 mutant based on our RNA-seq data yielded 91 overlapping genes (Fig 7A, B). GO functional enrichment analysis of these genes revealed their involvement in multiple important pathways, such as the MAPK signaling pathway, cell cycle, lipid metabolism, cell wall synthesis, metal ion stress response, sporulation, DON synthesis, and nucleocytoplasmic transport (Fig 7C). To validate the reliability of the ChIP-seq results, ChIP-qPCR and RT-qPCR were performed on 14 genes encompassing the above functions. The results were consistent with the sequencing data, showing significantly downregulated H2Bub1 levels and transcriptional levels for these genes in the ΔFgNup2 strain (Fig 7D, E and S8 Fig).

Download:

• PNG
  larger image
• TIFF
  original image

Fig 7. FgNup2 regulates the growth, development, stress response and virulence of F. graminearum by promoting the occurrence of H2bub1.

(A) Intersection analysis of genes with significantly downregulated H2Bub1 modifications in ΔFgNup2 and genes significantly downregulated in the transcriptome (RNA-seq). (B) Heatmap of RPKM-normalised expression levels of intersection genes in PH-1 and ΔFgNup2. (C) GO functional enrichment analysis of genes intersecting ChIP-seq and RNA-seq. (D) Using the IGV genome browser to display H2Bub1 ChIP-seq signal profiles of representative key gene chromatin regions in wild-type PH-1 and ΔFgNup2 strains under toxin-inducing conditions. the shown genes belong to different functional categories. (E) ChIP-qPCR validation of H2Bub1 modification levels at the gene loci shown in panel A in ΔFgNup2 and PH-1, the data come from three independent repeated experiments, and significant differences are indicated by * (****P < 0.0001, *P ≤ 0.05, ns P ≥ 0.05).

https://doi.org/10.1371/journal.ppat.1014577.g007

2.8 The key Phenylalanine residue F981 in the FG-repeat region of FgNup2 regulates FgBre1 nuclear transport and virulence

FgNup2 has three functional domains: an N-terminal nucleoporin domain containing an Impα-binding domain, a middle region containing a phenylalanine-glycine (FG) repeat sequence, and a C-terminal Ran-binding domain (Fig 8A). To identify the functional sites within FgNup2 that regulate virulence in F. graminearum, we first fused the three domains to a GFP-tagged vector and introduced them individually into the wild-type strain PH-1/Bre1-RFP. Co-IP assays using GFP affinity magnetic beads indicated that only the FgNup2–2 (FG-repeat) domain interacted with FgBre1, suggesting the FG-repeat plays a crucial role (Fig 8B, C). Subsequently, we mutated every 8 consecutive amino acids within the 143-amino-acid FgNup2–2 segment, generating 18 mutant strains for growth and pathogenicity screening (S9 Fig). The results showed that mutation of residues 980–987 significantly reduced growth rate and pathogenicity. Further alanine-scanning mutagenesis of these 8 residues (S10 Fig) revealed that only the mutation of phenylalanine at position 981 to alanine (F981A) caused significant reductions in growth rate, pathogenicity, and DON production (Fig 8D and S11A–S11B Fig). Furthermore, we examined the nuclear import of FgBre1 in the FgNup2^F981A strain at 6 h post induction and found that FgBre1 exhibited a nuclear import delay phenotype consistent with that of the ΔFgNup2 strain (Fig 8E and S11D Fig). The H2Bub1 modification level in the FgNup2^F981A strain was significantly lower than in the wild-type (S11E Fig). Concurrently, molecular docking simulations suggested an interaction between phenylalanine at position 981 of FgNup2 and FgBre1 (Fig 8F). Furthermore, we analyzed the sequence conservation of the F981 residue among different fungal species. The comparison results showed that F981 did not undergo substitution in a wide range of fungal groups from yeast to filamentous fungi(Fig 8G). To explore whether the F981 site is involved in regulating the interaction between FgNup2 and FgBre1, we performed Co-IP experiments in the wild-type PH-1 and the F981 point mutant strains. The results showed that in the F981 mutant strain, FgBre1 could still co-precipitate with FgNup2. However, compared with the wild type, the co-precipitation signal of FgBre1 in the mutant was significantly weakened (S11F Fig), indicating that the interaction between the two was not completely lost. The F981 site plays an important role in maintaining stable protein interactions between FgNup2 and FgBre1, but it is not an essential residue for this interaction.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 8. The key phenylalanine residue F981 in the FG repeat sequence of FgNup2 regulates FgBre1 nuclear transport and virulence.

(A) Schematic diagram of the FgNup2 protein domains, showing the N-terminal Importinα binding domain, central phenylalanine-glycine repeat sequences (FG repeats), and C-terminal Ran binding domain. (B) Interaction detection between FgNup2-1, FgNup2-2, and FgNup2-3 domains and FgBre1 using co-immunoprecipitation (Co-IP) experiments. (C) Yeast two-hybrid (Y2H) experiments verification of interactions between the different domains of FgNup2 and FgBre1. (D) Colony morphology of wild-type PH-1 and FgNup2^F981A point mutant strains after 72 h of cultivation on PDA, CM, and MM media. (E) Comparison of disease symptoms between wild-type PH-1 and FgNup2^F981A strains in wheat coleoptile and spike inoculation tests. (F) After 6 h under toxin-inducing conditions, we used a laser confocal microscope to check the subcellular localisation of FgBre1 in the FgNup2^F981A point mutant. (G) The overall level of H2Bub1 modification in the FgNup2^F981A strain was detected by Western Blot. (H) Molecular docking simulation of the interaction between FgNup2 and FgBre1. Visualization of the molecular docking model of the FgNup2 and FgBre1 protein complex. The enlarged section of the figure shows the interaction interface between FgNup2 and FgBre1 and highlights the key residues of FgNup2.

https://doi.org/10.1371/journal.ppat.1014577.g008

4.1 Fungal strains and construction of fungal deletion

The wild-type (WT) strain PH-1 (NRRL 31084) of F. graminearum was used as the parental strain as described previously [45]. Potato dextrose agar (PDA), minimal media (MM), complete medium (CM) were used to measure Determination of growth rates, conidiation in liquid carboxymethylcellulose (CMC) medium and trichothecene biosynthesis-inducing(TBI) medium A is used to induce F. graminearum to produce deoxynivalenol(DON). For the detailed recipe of the culture medium, see Table S3 in S1 Appendix.

Constructs for gene deletion of F. graminearum using a PEG-mediated protoplast transformation method were carried out as described previously. Briefly, fresh mycelia were treated with driselase (D9515, Sigma, MO, USA), lysozyme (S10038, Yuanye, Shanghai, China) and cellulose (S10042, Yuanye, Shanghai, China). The open reading frame (ORF) of each gene was replaced with hygromycin resistance cassette (HPH) [46]. Primers used to amplify the flanking sequences for each gene are listed in Table S4 in S1 Appendix. Deletion mutants were identified by PCR and Southern blot assays.

4.2 Phylogenetic analysis

Nucleoporin protein sequences of F. graminearum strain PH-1 and Saccharomyces cerevisiae strain S288C were retrieved from the NCBI database (https://www.ncbi.nlm.nih.gov/). Phylogenetic trees were constructed using MEGA-X with the maximum likelihood method based on the Jones-Taylor-Thornton substitution model. Bootstrap values were calculated from 1,000 replicates to assess branch support. The resulting tree was visualized and annotated using the iTOL web server (version 6; https://itol.embl.de/).

4.3 GFP and RFP fusion cassettes construction

To construct the FgNup2-GFP cassette, FgNup2 containing native promoter region and ORF (without the stop codon) was amplified with the relevant primers in Table S4 in S1 Appendix. The resulting PCR products were co-transformed with XhoI-digested pYF11 containing a neo cassette into the yeast strain XK1–25 using the Alkali-Cation Yeast Transformation Kit (MP Biomedicals, Solon, USA) to generate the recombined FgNup2-GFP fusion vector. Subsequently, the FgNup2-GFP fusion vector was recovered from the yeast transformant using the Yeast Plasmid Kit (Solarbio, Beijing, China) and then transferred into Escherichia coli strain DH5α for amplification. Using the similar strategy, FgImpα-GFP, FgBre1-GFP and FgBre1-RFP, FgNup2–1-GFP, FgNup2–2-GFP and FgNup2–3-GFP fusion cassettes were constructed. The FgNup2-GFP cassettes was respectively transformed into ΔFgNup2 to acquire complemented strains. Cassette pairs FgNup2-GFP + FgBre1-RFP, FgImpα-GFP + FgBre1-RFP, FgNup2–1-GFP + FgBre1-RFP, FgNup2–2-GFP + FgBre1-RFP, and FgNup2–3-GFP + FgBre1-RFP were transformed into PH-1 for Co-IP assay and co-localization assay.

4.4 Determination of stress sensitivity, virulence and DON production

To determine sensitivity to various stresses, 5 mm mycelial plugs of each strain taken from a 3-day-old colony edge were inoculated on PDA supplemented without/with each stress agent (0.2 g·L⁻¹ Congo red, 0.01% SDS, 1 mol·L⁻¹ KCl, 0.3 mol·L⁻¹ CaCl₂, 0.2 mol·L⁻¹ MgCl₂ and 0.1 mol·L⁻¹ LiCl), Add the stress agent to the sterilised PDA medium cooled to about 55°C, mix well and pour into plates. Use a puncher to take a 5 mm diameter fungal disc from the colony edge and inoculate it in the centre of a PDA plate containing the stress agent, then incubated at 25°C for 3 d in the dark. Each experiment was repeated three times independently.

For evaluating virulence on wheat spikelets, a 10 μL aliquot of conidial suspension (10⁵ conidia/ml) was injected into a floret in the central section spikelet of single flowering wheat heads of susceptible wheat (Triticum aestivum) cultivar Jimai 22. There were 20 replicates for each strain. At 15 d after inoculation, infected spikelets in each inoculated wheat head were recorded. When cultivating coleoptiles, the seeds are first placed in a petri dish lined with moist filter paper in the dark at 25°C for 24 h. Then, the coleoptiles are transferred to sterile transparent plastic containers and incubated at 16°C under a 12 h light/12 h dark cycle for 48 h to induce wheat seed germination. A sterile scalpel is used to make a 2 mm incision at the tip of the coleoptile, and 5 μL of conidial suspension (10⁵ spores/mL) is directly inoculated on the wound. After inoculation, the seedlings are grown in a growth chamber at 25°C and 95% humidity for 10 d, after which the lesion length is measured. Twenty replicates were used for each strain. Each virulence experiment was repeated three times.

Take 5 strains of F. graminearum and cultivate them in YEPD medium at 25°C and 180 rpm in a shaking incubator for 16 h to prepare mycelia. After filtering the mycelia, rinse them with sterile water and transfer them to a conical flask containing 30 mL of TBI medium. Cultivate them in the dark at 25°C and 180 rpm for 48 h. After the cultivation is completed, filter, wash the mycelia with sterile water, and transfer them to a pre-weighted aluminum foil dish. Dry them in a 55°C oven for 48 hours. The dried mycelia are used for dry weight determination. Collect the filtrate and dilute it 100 times before directly using the DON quantitative kit Wis008 (Wise Science, Zhenjiang, China) to determine the DON toxin content.

4.5 Western blotting and co-immunoprecipitation (Co-IP) assays

Fresh mycelia (500 mg) from each fungus were ground into a fine powder and suspended in 1 mL of extraction buffer containing 1% protease inhibitor. After homogenising with a vortex mixer, the lysate was centrifuged at 10,000 × g for 20 minutes at 4°C to extract total protein from F. graminearum. The resulting proteins were separated by 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to Immobilon-P transfer membrane (Millipore, Billerica, MA, USA). GFP, RFP and Flag-tagged proteins were detected with monoclonal anti-GFP (ab32146 and ab124754 Cambridge, UK) and anti-Flag (A9044, Sigma) antibodies, respectively. H2Bub1, level was detected with the antibody Cell Signalling 5546. H3 and GAPDH levels was detected by anti-H3 antibodies (Huabio EM1306 and EM1101) for the protein loading reference. For Co-IP assays, total proteins were extracted and incubated with anti-GFP agarose (KTSM1301, Shenzhen, China). Proteins eluted from agarose were detected by protein Western blotting as described above.

4.6 Yeast two-hybrid (Y2H) assays

To construct plasmids for Y2H analyses, cDNA was used to clone the coding sequence of each tested gene with primers in Table S4 in S1 Appendix. The cDNA fragment was inserted into the yeast GAL4-binding domain vector pGBKT7 and GAL4-activation domain vector pGADT7 (Clontech, Mountain View, CA, USA), respectively. The pairs of Y2H plasmids were transformed into the S. cerevisiae strain Y2H Gold with the lithium acetate/single-stranded DNA/polyethylene glycol transformation protocol. The pair of pGBKT7–53 and pGADT7 plasmids was used as positive control, and the pair of pGBKT7-Lam and pGADT7 plasmids was used as negative control. After cultured on minimal synthetic defined (SD) medium lacking Leu and Trp at 30°C for 3 d, the transformants were transferred to SD lacking His, Leu, Trp and Ade to assess protein–protein interaction.

4.7 GST pull-down assay

To express protein, the coding sequences of FgNup2 and FgBre1 were amplified from cDNA, and then, the fragments were inserted into pGEX-6p-1 and pET28a, respectively, to generate constructs for GST-tagged and His-tagged proteins. These plasmids were then transformed into the E. coli strain BL21. The resulting GST-tagged proteins were purified by the GST agarose beads (C600031; BBI, Shanghai, China) and eluted by reduced glutathione solution, and the His-tagged proteins were purified by Ni sepharose beads (20503ES10; Yeasen Biotechnology Co. Ltd, Shanghai, China) and eluted by imidazole. To conduct the GST pull-down assay, purified FgBre1-GST and FgNup2-His were mixed and incubated at 4°C for 2 h, and then, GST agarose beads were added and incubated at 4°C for 2 h. The resulting protein complexes were washed ten times with 1 × TBS to avoid nonspecific binding proteins and then eluted by 5 × SDS buffer and boiled for 10 min. Finally, protein samples were analyzed by Western blot using the mouse monoclonal anti-GST (LF305, EpizymeBio Co. Ltd) and the mouse monoclonal anti-His (LF307, EpizymeBio Co. Ltd) antibodies.

4.8 Microscopic examination

Conidia were stained with Calcofluor white to observe conidial morphology. The nuclear dye DAPI was used to determine FgNup2, FgBre1and FgImpα localization. To examine formation of DON toxisome, the Tri1-GFP fusion cassette was constructed and transformed into PH-1, ∆FgNup2, respectively. The resulting strains were cultured in TBI medium at 25°C for 1 h-48 h in a shaker at 180 rpm before microscopic examination. The following parameter sets of the confocal microscopy (Nikon, Tokyo, Japan). were used: The laser excitation wavelength was set at 488 nm for green fluorescence, at 405 nm for DAPI (blue fluorescence) and at 561 nm for RFP (red fluorescence). The intensity of fluorescence was acquired using the ImageJ software.

4.9 Bimolecular fluorescence complementation (BiFC) assay

The FgBre1-cYFP and FgNup2-nYFP fusion constructs were generated by cloning the related fragments into pHZ68 and pHZ65 vectors, respectively. FgBre1-cYFP and FgNup2-nYFP constructs were co-transformed into PH-1. In addition, construct pairs FgBre1-cYFP and FgNup2-nYFP and cYFP were used as negative controls. Transformants resistant to both hygromycin and zeocin were isolated and confirmed by PCR. Then the YFP signals were examined using confocal microscope (Nikon, Tokyo, Japan).

4.10 Affinity capture-mass spectrometry analysis

FgNup2 was tagged with GFP and transferred into the FgNup2 deletion mutant. The resulting transformant was used for protein extraction. After protein extraction, the supernatant was transferred to a sterilized tube. An aliquot of 25 μL of GFP-trap agarose beads was added to capture FgNup2-GFP interacting proteins, following the manufacturer’s instructions. After incubation at 4°C overnight, the agarose beads were washed three times with 1 mL of TBS (20 mM Tris–HCl, 500 mM NaCl, pH 7.5).

Proteins bound to the beads were then boiled with 50 μL TBS supplemented with 10 μL 10% SDS. After centrifugation at 5000g for 5 min at 4°C, the supernatant was digested with trypsin, and tryptic peptides were analyzed by Shanghai MeijiBio Technology Co., Ltd. using mass spectrometry. The false discovery rate (FDR) was used to determine relative protein confidence. The peptide-to-spectrum matches (PSMs) were quantified to determine relative protein abundance. Candidate interacting proteins were filtered based on high confidence (≤1% FDR) and reproducible presence across samples (PSM>5). Two biological replicates were conducted for the strain.

4.11 RNA extraction and qRT-PCR

Total RNA was isolated from hyphae collected in TBI medium using the RNAiso reagent (TaKaRa, Dalian, China). A reverse transcription kit was used for reverse transcription to generate complementary cDNA (Vazyme, Nanjing, China). The primer pairs used are shown in Table S4 in S1 Appendix. The actin gene of F. graminearum was used as the reference. Finally, the 2^−ΔΔCT method was used to analyze the data.

4.12 Chromatin immunoprecipitation (ChIP)-seq and ChIP-qPCR analyses

PH-1 and the FgNup2 deletion mutant fresh mycelia were cross-linked with 1% formaldehyde for 15 min and then stopped with 125 mM glycine. The cultures were ground with liquid nitrogen and resuspended in lysis buffer (250 mM HEPES [pH 7.5], 150 mM NaCl, 1 mM EDTA, 1% Triton, 0.1% deoxycholate, 10 mM DTT) and 1% protease inhibitor. The DNA was sheared into~300 bp fragments with 20 pulses of 10 s with 20 s of resting at 35% amplitude (Qsonica*sonicator, Q125, Branson, USA). After centrifugation, the supernatant was diluted with 10 × ChIP dilution buffer (1.1% Triton X-100, 1.2 mM EDTA, 16.7 mM Tris–HCl [pH 8.0], 167 mM NaCl). Samples were pre-washed with 20 mL of protein A agarose (sc-2001, Santa Cruz, CA, USA) for 1 h at 4°C. Then immunoprecipitation was performed using a monoclonal anti-H2Bub1 (Cell Signalling 5546). A mock sample was incubated with anti-IgG antibody (MA1–10406, Invitrogen-Thermo Fisher Scientific, Waltham, MA, USA).

DNA was immunoprecipitated with ethanol after washing, eluting, reversing the cross-linking, and digesting with proteinase K. Further, ChIP-enriched DNA with a monoclonal anti-H2Bub1 antibody was sequenced on an Illumina NovaSeq 6000.

Different peak analysis was based on the fold enrichment of peaks of different experiments. A peak was determined as different peak when the odds ratio between two groups was more than 2. Using the same method, genes associated with different peaks were identified and also do GO and KEGG enrichment analysis. ChIP-enriched DNA with a monoclonal anti-GFP antibody were used for quantitative PCR analysis using SYBR green I fluorescent dye detection with the relative primers. The relative enrichment of each gene was determined by quantitative PCR and calculated by normalizing the value of the immunoprecipitated sample with that of the input.

4.13 Construction of a Fusion Expression Vector Containing Nuclear Localization Signal/Nuclear Export Signal Elements

In this study, the nuclear localization signal (NLS) sequence used was PKKKRKV. To enhance nuclear import efficiency, it was constructed as a triple tandem repeat (PKKKRKV-PKKKRKV-PKKKRKV). The nuclear export signal (NES) element was LQLPPLERLTL, which conforms to the conserved hydrophobic residue arrangement pattern (LxxLxL or LxxxLxL) of classical nuclear export signals [47]. The above sequences were synthesized by Sangon Biotech (Shanghai) Co., Ltd. and subsequently ligated into a eukaryotic expression vector containing a 3 × Flag tag via PCR or restriction enzyme digestion sites, allowing fusion expression of the target sequence with the 3 × Flag tag. The recombinant plasmid was verified by restriction enzyme digestion and DNA sequencing before being used in subsequent experiments.

4.14 Quantification and statistical analysis

Statistical analysis was performed using GraphPad Prism (v8.0) and ImageJ 1.53e software. Two-tailed Student’s t test was used for two-group comparisons. One-way ANOVA followed by Tukey post hoc test and two-way ANOVA followed by Tukey post hoc test were used for multiple comparisons.