Deoxynivalenol Biosynthesis in Fusarium pseudograminearum Significantly Repressed by a Megabirnavirus

Deoxynivalenol (DON) is a mycotoxin widely detected in cereal products contaminated by Fusarium. Fusarium pseudograminearum megabirnavirus 1 (FpgMBV1) is a double-stranded RNA virus infecting Fusarium pseudograminearum. In this study, it was revealed that the amount of DON in F. pseudograminearum was significantly suppressed by FpgMBV1 through a high-performance liquid chromatography–tandem mass spectrometry (HPLC-MS/MS) assay. A total of 2564 differentially expressed genes were identified by comparative transcriptomic analysis between the FpgMBV1-containing F. pseudograminearum strain FC136-2A and the virus-free strain FC136-2A-V-. Among them, 1585 genes were up-regulated and 979 genes were down-regulated. Particularly, the expression of 12 genes (FpTRI1, FpTRI3, FpTRI4, FpTRI5, FpTRI6, FpTRI8, FpTRI10, FpTRI11, FpTRI12, FpTRI14, FpTRI15, and FpTRI101) in the trichothecene biosynthetic (TRI) gene cluster was significantly down-regulated. Specific metabolic and transport processes and pathways including amino acid and lipid metabolism, ergosterol metabolic and biosynthetic processes, carbohydrate metabolism, and biosynthesis were regulated. These results suggest an unrevealing mechanism underlying the repression of DON and TRI gene expression by the mycovirus FpgMBV1, which would provide new methods in the detoxification of DON and reducing the yield loss in wheat.


Introduction
Fusarium is a genus of filamentous fungi ubiquitously existing in agricultural and natural ecosystems [1,2]. Some species in Fusarium cause diseases including wilts, blights, rots, and cankers on crops and some horticultural, ornamental, and forest plants [3,4]. Wheat crown rot (FCR), one of the most destructive wheat diseases worldwide, is caused by species in Fusarium [5]. The dominant causal agent of FCR is F. pseudograminearum [6][7][8]. The symptoms of FCR include dry seedlings at the seedling stage, browning and rot at the base of the stem at the adult stage, withered white ears at the filling stage, and shriveled grains at the harvest stage [9]. Besides the significant decrease in wheat yield, F. pseudograminearum produces a diverse array of toxic secondary metabolites including mycotoxins [10,11]. Cereals contaminated with mycotoxins are unsuitable for food or feed [12,13]. The mycotoxins produced by Fusarium include trichothecenes, fumonisins, etc. The most common Fusarium mycotoxin is deoxynivalenol (DON), which causes the disease of emesis, oral lesions, dermatitis, and hemorrhaging in human and livestock [14,15]. The biosynthesis of DON starts from the cyclization of farnesyl pyrophosphate mediated by trichodiene synthases (encoded by TRI5) [16]. Then, multiple steps are followed by at least six additional enzymes encoded by TRI4, TRI101, TRI11, TRI3, TRI1, and TRI8. Moreover, two transcription factors, TRI6 and TRI10; a transmembrane transporter, TRI12; and genes with unknown functions, TRI9 and TRI14, may also be involved in DON biosynthesis and the virulence of Fusarium [17][18][19][20][21].
Mycoviruses are viruses hosted in fungi. They have been described in many fungal species. Some of the viruses are hypovirulence-related with an effect on reducing the virulence of the pathogenic fungi. Besides the hypovirulence effect, some mycoviruses have an influence on the cultivating features of host fungi including the morphology and development of the colonies and spores. For example, Cryphonectria hypovirus 1 (CHV1) decreased the pigmentation and spore production of Cryphonectria parasitica [22]. Sclerotinia sclerotiorum hypovirulence-associated DNA virus 1(SsHDV1) caused abnormal colony morphology and the reduction in colony growth and small sclerotia production to Sclerotinia sclerotiorum [23]. Rosellinia necatrix megabirnavirus 1 (RnMBV1) suppressed the growth and melanin biosynthesis in Rosellinia necatrix [24].
To reveal the mechanism underlying the regulation of mycoviruses on fungi, highthroughput mRNA sequencing (RNAseq) was applied. For instance, the differential expression of genes related to fungal metabolism, transcription, translation, and ribosomal RNA processing was shown to be related to FgV1 infection [32]. Host cell transportrelated genes were down-regulated by Fusarium graminearum virus 3 (FgV3) [33]. Genes involved in RNA processing and ribosome transport assembly were down regulated during the infection of Fusarium graminearum virus 4 (FgV4) [33]. Genes related to glutamate metabolism, homoserine metabolism, cellular aldehyde metabolism, and lactate metabolism in F. graminearum were down-regulated by FgHV1 [34]. RNA silencing and virulence-related genes were considerably down-regulated in S. sclerotiorum strains infected by SsHADV-1 [35].
In this study, the DON production of the FpgMBV1-containing F. pseudograminearum strain was determined and compared with that of the virus-free strain by HPLC-MS/MS. The genome-wide transcriptional reprogramming in F. pseudograminearum under the infection of FpgMBV1 was outlined by RNA sequencing. Genes involved in the biosynthesis of DON and ergosterol and in RNA-silencing pathways were analyzed to unravel the interaction between FpgMBV1 and F. pseudograminearum. Data obtained in this study would provide a clue to the specific regulation mechanisms of mycovirus to DON biosynthesis and virulence of fungal pathogens including Fusarium.

DON Synthesis Was Inhibited by Mycovirus FpgMBV1
To investigate the effect of FpgMBV1 on the content of DON in F. pseudograminearum, the isogenic strains FC136-2A and FC136-2A-Vwere cultivated in wheat grains for 30 days at 25 • C. Then, 20mL acetonitrile/water/acetic acid (70:29:1, v/v/v) was used to extract DON from 5 g of these mixtures for each sample for HPLC-MS/MS analysis. Results showed that the content of DON in the FpgMBV1-containing F. pseudograminearum strain FC136-2A was 1.3 ± 0.1 µg/kg. The content of DON in the virus-free strain FC136-2A-Vwas 10.6 ± 1.4 µg/kg. Compared to the virus-free strain FC136-2A-V -, the content of DON in the FpgMBV1-containing strain FC136-2A was reduced by about 87% (p < 0.05) ( Figure 1). No significant difference was found for the content of 15-ADON in the FpgMBV1containing strain and the virus-free strain ( Figure S2). To reveal how the biosynthesis genes of DON were regulated by FpgMBV1, the FpgMBV1-containing F. pseudograminearum strain FC136-2A and the isogenic virus-free strain FC136-2A-Vwere used in the transcriptome analysis with three biological replicates for each strain using the DNBSEQ platform (BGI, Shenzhen, China). Gene expression profiles of the strains FC136-2A and FC136-2A-Vwere compared. An average of 7.24 Gb of data was yielded for each sample. The genome of F. pseudograminearum CS3096 was used as the reference genome. An average of 94.57% and 94.27% of total reads (56-57 million) were aligned to the genome, respectively. A total of 11,500 genes were detected as expressed, of which 11417 were known and 83 were predicted as unknown genes (Table S1). The absolute fold change ≥ 2 and adjusted p-value ≤ 0.001 were used to define differently expressed genes (DEGs) (Figure 2A). Compared to the gene expression of the strain FC136-2A-V -, a total of 2564 statistically significant DEGs were found in the strain FC136-2A. Among them, 1585 genes were up-regulated and 979 genes were down-regulated ( Figure 2B). The PHI database annotation was screened using the DIAMOND software. A total of 131 genes were associated with loss of pathogenicity, 479 genes with reduced virulence, and 32 genes with lethal consequences ( Figure 2C). In this study, the DEGs with PHI identity > 80% were used for further screening and the heat map construction. There were 18 genes categorized as reduced virulence genes including 4 up-regulated genes and 14 down-regulated ones ( Figure S1). There were four genes categorized as essential genes ( Figure S1). All these essential genes were up-regulated. No genes associated with loss of pathogenicity were identified in this study. Moreover, 12 of these genes were associated with DON and ergosterol biosynthesis (Table 1). Gene ontology (GO) terms significantly enriched in the three major functional ontologies: 57.5% for biological process (BP), 10.1% for cellular component (CC) and 32.4% for molecular function (MF) ( Figure 2D). For the 1585 up-regulated genes, the most enriched three for the BP were carbohydrate metabolic process, transmembrane transport, catechol-containing compound metabolic process, and oxidoreductase activity, catalytic activity, and cofactor binding for the MF ( Figure 3A, Table S2). For the 979 down-regulated genes, the most enriched three of the BP were sterol metabolic and biosynthetic process, steroid metabolic and biosynthetic process, ergosterol metabolic and biosynthetic process; and oxidoreductase activity, iron ion binding, and cofactor binding for the MF ( Figure 3B, Table S3). The most enrichedthree of the CC were integral component of membrane, an intrinsic component of membrane, membrane part, and membrane in all DEGs. There were 20 significantly enriched KEGG pathways for the up-regulated genes ( Figure 3C, Table S4) and down-regulated genes, respectively ( Figure 3D, Table S5). Sixteen pathways were related to metabolism.
The KEGG enrichment heat map shows that the fungal metabolic balance of amino acids and lipids was disturbed by FpgMBV1 ( Figure 4). The transcription and translation of cell membrane-associated genes was also regulated by FpgMBV1.
The KEGG enrichment heat map shows that the fungal metabolic balance of amino acids and lipids was disturbed by FpgMBV1 (Figure 4). The transcription and translation of cell membrane-associated genes was also regulated by FpgMBV1.

TRI Genes Regulating DON Biosynthesis Were Down-Regulated by FpgMBV1
The genes in the TRI gene cluster encoding key enzymes for DON biosynthesis were FgTRI1-FgTRI16 and FgTRI101 in F. graminearum [36]. Comparative transcriptomic results showed that 12 genes in the TRI gene cluster were down-regulated by FpgMBV1. They were FpTRI1, FpTRI3, FpTRI4, FpTRI5, FpTRI6, FpTRI8, FpTRI10, FpTRI11, FpTRI12, FpTRI14, FpTRI15, and FpTRI101 ( Figure 5). These TRI genes regulated by FpgMBV1 accounted for about two-thirds of the total gene number in the TRI gene cluster. Additionally, they were all significantly repressed.

Ergosterol Biosynthesis and Metabolism Were Inhibited by FpgMBV1
Fungal ergosterol content is significantly and positively correlated with DON content. In this study, ergosterol synthesis and metabolism were significantly enriched according to the results of GO enrichment of down-regulated genes ( Figure 6A). In the PHI database annotation, four DEGs related to ergosterol synthesis were significantly down-regulated by FpgMBV1. They were FgERG4 (PHI:2728), FgERG5B (PHI:3038), FgERG5A (PHI:3037), and FgERG3B (PHI:3036) ( Figure 6B). These results showed inhibition in ergosterol biosynthesis and metabolism in the FpgMBV1-containing strain FC136-2A.

Gene Expression Level by Quantitive Real-Time RT-PCR
Through quantitative real-time RT-PCR, the expression levels of 15 DEGs involved in metabolism, RNA silencing, virulence, and DON biosynthesis were confirmed. Primers are listed in Table S6. The expression levels of these representative genes were consistent with those in the transcriptomic data ( Figure 8).

Gene Expression Level by Quantitive Real-Time RT-PCR
Through quantitative real-time RT-PCR, the expression levels of 15 DEGs involved in metabolism, RNA silencing, virulence, and DON biosynthesis were confirmed. Primers are listed in Table S6. The expression levels of these representative genes were consistent with those in the transcriptomic data ( Figure 8).

Discussion
In this study, the production of key mycotoxins DON and 15-ADON was found to be inhibited by FpgMBV1 significantly. The expression of 12 out of 15 TRI genes was repressed in accordance with the HPLC-MS/MS data. Previously, mycovirus FgV1 was reported to cause decreased DON production (60-fold) in F. graminearum [26]. However, only the TRI12 gene in the TRI gene cluster was differently expressed in the FgV1-containing strains [32].
For FgMTV1/SX64, another mycovirus reducing the DON concentration significantly in F. graminearum, whether TRI gene expression was regulated is unknown [31]. The global regulation of most TRI genes by a mycovirus was first reported for FpgMBV1. Two transcription regulators, TRI6 and TRI10, were key regulators in the TRI gene cluster [17]. TRI6 has been identified as a global transcription regulator, not only enhancing the expression of the genes in the DON biosynthetic pathway but also involved in the upstream isoprenoid pathway for trichothecene accumulation [38,39]. TRI10 regulated the expression of TRI1, TRI3, TRI8, TRI11, TRI12, TRI14, and TRI15 in Fusarium spp. [40]. In this study, the significant down-regulation of TRI6 and TRI10 might be the reason for the down-regulation of other TRI genes. The significant disruption of TRI gene expression resulted in the low contents of DON mycotoxin. This suggests an unrevealing mechanism underlying the repression of DON and TRI gene expression by mycovirus FpgMBV1, which would provide new methods in the detoxification of DON and reducing the yield loss in wheat.
Other genes of biosynthetic pathways were differentially expressed under the infection of FpgMBV1. For example, four genes encoding key enzymes in ergosterol synthesis were down-regulated. They were FgERG3B, FgERG4, FgERG5A, and FgERG5B [41,42]. It has been reported that ergosterol synthesis has a strong, positive correlation with the content of mycotoxin DON in the infected grains [43,44]. Moreover, ergosterol is an important constituent of fungal membranes [45]. Membranes are essential for many cellular processes, including the defensing response against viruses in fungi [46,47]. Another significantly down-regulated gene, HMG-CoA-reductase (HMR1), is the key enzyme in the mevalonate pathway. HMR1 is involved in the biosynthesis of many primary and secondary metabolites [48]. Some mevalonate derivatives function in fungi-plant interaction, such as isoprenoids and gibberellins [49]. Another down-regulated gene, FgVELB, plays a critical role in regulating various cellular processes and acts as a negative regulator for lipid biosynthesis. The deletion mutant of the FgVELB gene in F. graminearum produces a very low level of DON [50,51]. These results demonstrated the specific genes involved in the regulation of DON production Further studies on these genes would clarify the virulence-related pathways in fungi, especially Fusarium.
In general, the gene expression profile of F. pseudograminearum was reprogrammed by FpgMBV1 with 2564 genes differentially expressed. The disrupted metabolism caused by FpgMBV1 included amino acid and lipid metabolism, ergosterol metabolic and biosynthetic processes, carbohydrate metabolism and biosynthesis. Some of these pathways were critical in the transcriptome of other fungi under the infection of mycoviruses. In C. parasitica, "biosynthesis of other secondary metabolites", "amino acid metabolism", "carbohydrate metabolism", and "translation" were enriched among the DEGs after CHV1 infection, demonstrating that virus infection resulted in massive but specific changes in primary and secondary metabolism. Some of the highly induced metabolites played key roles in the growth, development and pathogenicity of fungi [52]. The complex interaction between leucine metabolism and the global regulator of mycotoxin biosynthesis, TRI6, and virulence in F. graminearum has been explored [53]. In this study, the KEGG enrichment results showed that tryptophan, phenylalanine, tyrosine, glycine, serine, and threoninerelated genes were differentially expressed under the infection of FpgMBV1. Further studies on the regulation mechanism of FpgMBV1 would help in revealing the crosstalk between some primary metabolic pathways and mycotoxin biosynthesis and virulence in F. pseudograminearum.
Mycoviruses were triggers and targets of RNA silencing. Three genes (FpDCL2, FpAGO2, and FpRdR1) were up-regulated and two genes (FpDCL1 and FpAGO1) were down-regulated in F. pseudograminearum under the infection of FpgMBV1. They are genes for key components in the RNA-silencing machinery [54]. Gene expression of DCL2 and AGO2 were induced by CHV1 in C. parasitica [55]. Genes of SsAgl2, SsDcl1, and SsDCl2 were essential in defending against viruses in S. sclerotiorum [56]. The expression of FgDICER2 and FgAGO2 were suppressed by FgV1 in F. graminearum [57]. Most RNA-silencing genes in S. sclerotiorum were repressed by SsHADV-1 [35]. Considering the high diversity of these virus-fungi systems, the different regulation on the RNA-silencing pathway seems reasonable. The regulation network of FpgMBV1 is partially revealed in this study, while the underlying mechanism is to be explored.

Conclusions
Megabirnavirus FpgMBV1 significantly repressed the production of mycotoxin DON and the expression of 12 TRI genes in F. pseudograminearum. The expression of ergosterol biosynthesis and RNA-silencing-related genes and genes involved in metabolism were regulated by FpgMBV1 (Figure 9). Prospectively, FpgMBV1 is valuable in the detoxification of DON and the management of diseases caused by F. pseudograminearum.

Fungal Material and Growth Conditions
FpgMBV1-containing F. pseudograminearum strain FC136-2A and its isogenic FpgMBV1free strain FC136-2A-Vwere maintained in the laboratory, department of plant protection, Henan Agricultural University, Henan province, China [27]. For RNA sequencing, these two strains were grown at 25 • C in the dark on potato dextrose agar (PDA) medium (Becton, Dickinson, and Company, Sparks, MD, USA). For HPLC-MS/MS analysis, these strains were cultured in wheat grain media. Typically, 180 g of wheat seeds was soaked in distilled water for 12 h and boiled for 30 min then air-dried and autoclaved at 120 • C for 30 min in triangle flasks to complete the preparation of the wheat grain medium. Three-day-old PDA plugs of FC136-2A or FC136-2A-Vmycelia were cultured in wheat grain media for 30 days in the dark at 25 • C.

HPLC-MS/MS Analysis of Type B Trichothecene
Finely ground wheat grain (5.00 g) was weighted and extracted with 20 mL acetonitrile/water/acetic acid (70:29:1, v/v/v). The suspension was blended for 5 min at 2500 rpm using multi-tube vortexer UMV-2. The homogenized solution was centrifuged for 10 min after cooling (T = 4°C, 8000 rpm). Then, 750 µL of the upper layer and 2 mL water were mixed in a 15 mL tube at 11,000 rpm for 10 min. The upper layer was cleaned via nylon filters and used for HPLC-MS/MS analysis. HPLC analysis was performed using LC-MS-8060NX (Shimadzu, Japan) in gradient conditions. The separation of the toxin was performed using 100 nm × 2.1 mm, 1.9 µm (Thermo Fisher Scientific, Shanghai, China). The column temperature was set at 40°C and the injection volume was 1 µL. Solvent A was 1 mmoL·L-1 ammonium formate in 0.1% formic acid and solvent B was methanol and 1 mmoL·L-1 ammonium formate in 0.1% formic acid. A gradient at a flow rate of 0.2 mL·min-1 was performed within 26 min (Table S7)

Total RNA Extraction
The 0.5 mg mycelia of strains FC136-2A and FC136-2A-Vwere harvested from PDA plates and ground in liquid nitrogen. Total RNA isolation was conducted using TRIzol (Invitrogen, Carlsbad, CA, USA). The extracted RNA was treated with RNase-free DNase I (Promega, Madison, WI, United Kingdom). The concentration of the extracted RNAs was determined with Nanodrop system (NanoDrop, Madison, WI, USA). The RNA integrity was examined by the RNA integrity number (RIN) using an Agilent 2100 bioanalyzer (Agilent, Santa Clara, CA, USA).

cDNA Library Preparation and Sequencing
MGI Easy RNA Library Preparation Kit (BGI-Shenzhen, China) and DNA Clean Beads MGI Easy DNA Cleanbeads Kit (BGI-Shenzhen, China) were used to purify mRNA and synthesize cDNA and PCR amplification; then, purification of the PCR product was carried out following the kit instructions. The product was validated on the Agilent Technologies 2100 bioanalyzer for quality control. The double-stranded PCR products from the previous step were heated, denatured, and circularized by splint oligo sequencing to obtain the final library. The single-stranded circular DNA was formatted as the last library and amplified with phi29 polymerase to make a DNA nanoball, with more than 300 copies per molecule. DNBs were loaded into the patterned nanoarray, and 100 paired base reads were generated on the BGI seq500 platform (BGI-Shenzhen, China).

RNA-Seq Data Analysis
The raw data were filtered with BGI's filtering software SOAPnuke (version: v1.4.0) to remove reads containing splices (splice contamination), reads with unknown base N content greater than 10%, and reads in which bases with a quality value of lower than 15 accounted for more than 20% of the total number of bases. Using HISAT2 software (version: v2.1.0; http://www.ccb.jhu.edu/software/hisat), the clean reads were mapped to the F. pseudograminearum CS3096 genome (Taxonomy ID: 1028729). Bowtie2 (version: v2.2.5; http://bowtie-bio.sourceforge.net/Bowtie2/index.shtml) was used to compare the reference gene sequences and RSEM (version: v1.2.8) was applied to calculate the expression levels of genes and transcripts. Pearson correlation coefficients between every two samples were obtained using the core function in the R software, and principal component analysis (PCA) was performed using the princomp method. To identify significant virulence genes in fungus, DIAMOND software was used to annotate genes with query coverage of 50% and identity of 40% in the PHI database (version: V0.8.31). To improve the accuracy of DEGs, we defined genes with more than two-fold difference and Q-values ≤ 0.001 to be screened as significantly differentially expressed genes. Differential genes were functionally classified according to GO and KEGG annotation results and official classification. At the same time, enrichment analysis was performed using the phyper function in R software, with FDR correction for p-value, and functions with Q value ≤ 0.05 were considered significantly enriched.
The qRT-PCR was carried out in a QuantStudio 3 Real-Time PCR System (Thermo Fisher Scientific, Waltham, MA, USA) with 2×T5 Fast qPCR Mix (Tsingke Biotechnology Co., Ltd., Beijing, China). PCR amplification was performed under the following conditions: 95 • C for 1 min, followed by 40 cycles of 95 • C for 10 s, and 60 • C for 30 s. Meltcurve plots were analyzed for each gene tested after each PCR reaction. The ubiquitin gene of TEF1α (FPSE_11980) served as an internal reference gene. Primers for 15 of the DEGs were designed using Primer Premier 5 and are listed in Table S6.  Table S1: Statistics of the transcriptome-sequencing results, Table S2: The GO enrichment analysis of the up-regulated genes, Table S3: The GO enrichment analysis of the down-regulated genes, Table S4: The KEGG enrichment analysis of the up-regulated genes, Table S5: The KEGG enrichment analysis of the down-regulated genes, Table S6: qRT-PCR primers used in this study, Table S7: Gradient elution conditions of mobile phase.

Data Availability Statement:
The data presented in this study are available in Supplementary Materials Tables S1-S7 and Figures S1 and S2.

Acknowledgments:
We thank the editor for editing assistance, and the anonymous reviewers for constructive and helpful comments. We also acknowledge the BGI database development team for technical support.

Conflicts of Interest:
The authors declare no conflict of interest.