Unanticipated Large-Scale Deletion in Fusarium graminearum Genome Using CRISPR/Cas9 and Its Impact on Growth and Virulence

Abstract

Fusarium graminearum, a filamentous fungus, and causal agent of Fusarium head blight (FHB) in wheat and other cereals, leads to significant economic losses globally. This study aimed to investigate the roles of specific genes in F. graminearum virulence using CRISPR/Cas9-mediated gene deletions. Illumina sequencing was used to characterize the genomic changes due to editing. Unexpectedly, a large-scale deletion of 525,223 base pairs on chromosome 2, comprising over 222 genes, occurred in two isolates. Many of the deleted genes were predicted to be involved in essential molecular functions, such as oxidoreductase activity, transmembrane transporter activity, hydrolase activity, as well as biological processes, such as carbohydrate metabolism and transmembrane transport. Despite the substantial loss of genetic material, the mutant isolate exhibited normal growth rates and virulence on wheat under most conditions. However, growth rates were significantly reduced under high temperatures and on some media. Additionally, wheat inoculation assays using clip dipping, seed inoculation, and head point inoculation methods were performed. No significant differences in virulence were observed, suggesting that these genes were not involved in infection or alternative compensatory pathways, and allow the fungi to maintain pathogenicity despite the extensive genomic deletion.

Keywords: CRISPR/Cas9 editing; Fusarium graminearum; dual ribozyme sgRNA processing; large deletion.

Figure 1

Vector for CRISPR/Cas9 gene disruption of F. graminearum. (A) pEGC-sGR1 plasmid derived from pFGL821, featuring hygromycin B resistance, Cas9 gene with codon optimization driven by GPDA promoter, and a cassette encoding 5 sgRNAs, flanked by dual ribozymes, and codon-optimized GFP gene driven by the pTef1 promoter. (B) Example sgRNA cassette flanked by 5′ and 3′ sequences encoding a hammerhead ribozyme and HDV ribozyme, respectively, to process each sgRNA cassette.

Figure 2

Agarose gel (1%) showing PCR-amplified region of disrupted genes in transformed isolates. Ladder = ThermoScientific GeneRuler 100 bp Plus DNA ladder (Cat. No. SM0321). PCR products for (A) FGSG 03445 amplified by 4A-17-F and 4A-756-R, (B) FGSG 08238, amplified by 4B-9-F and 4B-673-R, (C) FGSG 04583, amplified by 4C-23-F and 4C-962-R, and (D) FGSG 04583, amplified by 4C-Fl-F and 4C-Fl-R. NTC = not template control reaction.

Figure 3

Location of target genes (A) and mapping depth of reads to genes (B) FGSG 08238, (C) FGSG 03445, and (D) FGSG 04583 in the F. graminearum PH-1 genome. Yellow indicates left single-mapped reads, grey indicated paired-mapped reads, and blue indicates right single-mapped reads. (E) Extent of deletion to chromosome 2, where dark grey indicates forward paired-mapped reads, medium grey indicates reverse paired-mapped reads, and light grey indicates proportion of single-mapped reads. Insertion of the transformation vector was detected in FGSG 08238 for Tk-1 and Tk-3, FGSG 03445 for Tk-1, and FGSG 04583 for Tk-1 and Tk-3 at the sgRNA cut sites as pairs of the single-mapped reads matched regions of pEGC-sGR1. Maximum read depth of the selected region is indicated on the right axis.

Figure 4

Summary of observed deletion sites on target genes in isolates Tk-1, Tk-3, and Tk-19. sgRNA target sites (grey) and detected cut sites (green) in CRISPR edited isolates Tk-1, Tk-3, and Tk-19. All isolates showed the same cut sites for gene FGSG 3455. Tk-1 and Tk-19 had the same cut sites in FGSG 8238, but Tk-3 was cut at the 297F target sites instead of 279F and had four less bp deleted at the B97 site. Tk-1 and Tk-19 were both cut at the expected C99 cut site, but Tk-3 had three more bp deleted in FGSG 4583. Tk-19 was also cut 175 bp from the C99. Location on chromosome 2 of the first base in the selection is indicated below the gene name. Sequence is split by dots for readability to indicate the nontargeted space between 5′ and 3′ sgRNA sites.

Figure 5

Analysis of disrupted genes by BLAST2GO. Multilevel summary pie charts summarize the genes’ (A) biological processes, (B) cellular components, and (C) molecular function. Numbers correspond to the quantity of genes predicted to each annotation. Some genes have multiple annotations.

Figure 6

Comparative wheat virulence assessment of wt and mutated F. graminearum isolates. (A) Visual comparison of disease symptoms after seedling inoculation with isolate Tk-3 vs. untreated control, 14 days post-inoculation (DAI); (B) subjective root disease ratings; (C) proportion of germinated seeds following inoculation; (D) clip dipping leaf lesions on inoculated leaves; (E) lesion measurements, 7 DAI; (F) visibly infected spikes post-point inoculation; (G) total count of visibly infected spikelets, 14 DAI. Error bars represent standard errors. ANOVA and Tukey’s HSD test revealed no statistically significant differences between wt and mutated isolates.

Figure 7

Growth rate characteristics of F. graminearum wt and mutated isolates. Growth rates of isolates grown at temperatures of 10, 20, and 30 °C grown on PDA. Vertical bars represent standard errors; bars sharing letters within temperatures are not significantly different at α ≤ 0.05 using Tukey’s HSD test (A). Growth rates of wt and mutant Tk-1 on YM agar, oatmeal agar, and CZ agar media. Vertical bars represent standard errors. Bars sharing letters within media are not significantly different at α ≤ 0.05 using Student’s t-test (B).