Characterization of RNA silencing components in the plant pathogenic fungus Fusarium graminearum

Abstract

The RNA interference (RNAi) plays a critical role in gene regulation in a variety of eukaryotic organisms. However, the role of RNAi remains largely unclear in plant pathogenic fungi. In this study, we explored the roles of core components of the RNAi pathway in Fusarium graminearum, the major causal agent of wheat head blight. Our results demonstrated that the hairpin RNA (hpRNA) can efficiently silence the expression level of target gene, and the argonaute protein FgAgo1 and dicer protein FgDicer2 are important in this silencing process. RNAi machinery was not involved in growth, abiotic stress and pathogenesis in F. graminearum under tested conditions. We firstly applied high-throughput sequencing technology to elucidate small RNA (17-40 nucleotides) (sRNA) transcriptome in F. graminearum, and found that a total of forty-nine micro-like-RNA (milRNA) candidates were identified in the wild-type and ∆FgDICER2, and twenty-four of them were FgDicer2-dependent. Fg-milRNA-4 negatively regulated expression of its target gene. Taken together, our results indicated that the hpRNA-induced gene silencing was a valuable genetic tool for exploring gene function in F. graminearum. FgAgo1 and FgDicer2 proteins played a critical role in the hpRNA mediated gene silencing process. In addition, FgDicer2 was involved in sRNA transcription and milRNA generation in this fungus.

Figure 1. Phenotypic analyses of hpRNA-induced single gene silencing in F. graminearum.

(a) Schematic representation of hpRNA-expressing constructs. The hairpin inverted-repeat targeted region of FgPKS12, FgCYP51A, or FgCNB1 was obtained by PCR and inserted into the multiple cloning site of hpRNA-induced gene silencing vector pSilent-1(b) Colony morphology of the RNAi-silenced FgPKS12 transformant. The wild-type HN9-1 and FgPKS12 gene silencing transformant were grown on PDA for 7 days at 25 °C. The silencing mutant FgPKS12-S1 displayed white colony, similar to the FgPKS12 deletion mutant (∆FgPKS12). (c) Susceptibility of FgCYP51A silencing transformant to the DMI fungicide triadimefon. FgCYP51A silencing transformant FgCYP51A-S7 were grown on PDA supplemented with triadimefon (5 μg mL⁻¹) for 3 days at 25 °C. Similar to the FgCYP51A deletion mutant (∆FgCYP51A), the silencing transformant showed increased sensitivity to triadimefon. (d) Growth pattern of the essential gene FgCNB1 silencing transformant. The FgCNB1 silencing transformant FgCNB1-S3 and the wild-type HN9-1 were cultured on PDA at 25 °C. The image was taken after 3 days of incubation. (e) The relative mRNA expression level of FgCNB1 in silencing transformants in comparison to that in the wild type. Line bars denote standard errors of three repeated experiments.

Figure 2. Simultaneous silencing of multiple genes in F. graminearum via expression of hpRNA with chimeric inverted repeats.

(a) Schematic representation of the chimeric inverted-repeat fragment construction. The target fragment of FgCYP51A, FgPKS12, and the DON biosynthesis gene FgTRI6 was constructed using the joint PCR strategy and then inserted into the multiple cloning site of pSilent-1. (b) Phenotypic analyses of the silencing transformant. The silencing transformant CPT-S2 showed increased susceptibility to triadimefon (5 μg mL⁻¹; left top), displayed white colony (left bottom), and produced drmatically less DON than the wild type HN9-1 (right panel).

Figure 3. The role of argonautes and dicers in hpRNA-induced gene silencing in F. graminearum.

(a) Growth pattern of HN9-1, FgCYP51A gene silencing transformant FgCYP51A-S7, argonaute gene mutants in the FgCYP51A-S7 background (S7-∆FgAGO1 and S7-∆ FgAGO 2), and dicer gene mutants in the FgCYP51A-S7 background (S7-∆FgDICER1 and S7-∆FgDICER2) on PDA with or without DMI fungicide triadimefon (5 μg mL⁻¹). The plate photographs were taken after 3 days of incubation at 25 °C. (b) The relative mRNA expression levels of FgCYP51A in the wild type, silencing transformant FgCYP51A-S7, S7-∆FgDICER1, S7-∆FgDICER2, S7-∆FgAGO1, and S7-∆FgAGO2 treated with triadimefon. Mycelia were collected after each strain was treated with 5 μg mL⁻¹ triadimefon for 6 h, then the RNA extracted from each strain was used for quantitative RT-PCR assays. Expression of the actin gene was used as the reference. Line bars denote standard errors of three repeated experiments.

Figure 4. Relative mRNA expression patterns of argonaute and dicer genes.

(a) The relative mRNA expression level of argonaute or dicer genes in conidia represents the relative amount of each gene mRNA in the mycelia determined by quantitative RT-PCR. Line bars indicate standard errors from three repeated experiments. (b) Localization of the FgDicer2-GFP fusion protein in mycelia and conidia. Fresh mycelia and conidia were collected and visualized with a Zeiss LSM780 confocal microscope for GFP signals. To observe nuclei, fresh mycelia or conidia were washed with sterilized water and stained with 10 μg mL⁻¹ 4′6-diamidino-2-phenylindole (DAPI, Sigma).

Figure 5. Characterization of small RNAs in the wild-type HN9-1 and ∆FgDICER2 of F. graminearum.

(a) The percentage of different sizes of small RNAs. Line bars indicate standard deviations of three repeated experiments. (b) Nucleotide frequency of the 5′ end of small RNAs in mycelia of the wild type and ∆FgDICER2. The percentages of small RNA distributions on the sense (c) and antisense (d) strands of supercontigs.

Figure 6. The milRNA expression patterns in F. graminearum.

(a) Comparisons of expression levels of 49 milRNAs candidates in the wild type and ∆FgDICER2. (b) Relative mRNA expression level of Fg-milRNA-4 target gene in the wild type and ∆FgDICER2. The relative mRNA expression levels of Fg-milRNA-3 target gene (FGSG_05319), Fg-milRNA-4 target gene (FGSG_04063) in ∆FgDICER2 represent the relative amount of each mRNA in the wild type determined via quantitative RT-PCR. Fg-milRNA-3 target-gene (FGSG_05319) was measured as a control. Line bars indicate standard errors from three repeated experiments. The star symptom meant there was significantly different.