The FFSC isolates used in this study were obtained from rice seeds provided by the National Institute of Crop Science (Wanju, Republic of Korea). A total of 46 rice cultivars were examined, and 35 seeds from each cultivar were randomly chosen. All seeds were surface-sterilized with 1% sodium hypochlorite for 3 min and rinsed in sterile distilled water for 3 min; the water on the surface was removed using an autoclaved paper towel. Surface-sterilized seeds were placed on potato dextrose agar (PDA) containing 0.1% lactic acid to prevent bacterial contamination (Alakomi et al., 2000). PDA plates were incubated at 25°C for five days. To isolate F. fujikuroi strains, 293 candidate strains were initially selected based on colony morphology. We selected the final 28 strains based on morphological characteristics, such as conidial shapes. All strains grown on PDA were stored in 20% glycerol solution at -80°C.

The representative fungal strains used in this study were provided by the Rural Development Administration (Jeonju, Republic of Korea), Korean Collection for Type Cultures (Jeongeup, Republic of Korea), Fungal Plant Pathology Lab. (Seoul National University, Seoul, Republic of Korea), Chung-Ang University (Seoul, Republic of Korea), Soonchunhyang University (Asan, Republic of Korea), and Chonnam National University (Gwangju, Republic of Korea). Detailed information about all strains used in this study is listed in Table S1 .

The oligonucleotide primer pairs ITS4 (5′-TCCTCCGCTTATTGATATGC-3′)/ITS5 (5′- GGAAGTAAAAGTCGTAACAAGG-3′) and EF1T (5′-ATGGGTAAGGAGGACAAGAC-3′)/EF2T (5′-GGAAGTACCAGTGATCATGTT-3′) were used for the amplification of the ITS and TEF-1α regions, respectively (O'Donnell et al., 1998). All oligonucleotide primer sets were synthesized at an oligonucleotide synthesis facility (Bioneer, Daejeon, Republic of Korea). Amplification reactions were performed in a total volume of 20 μl AccuPower PCR Premix kit (Bioneer), containing 19 μl distilled water and 0.5 μl of each primer (20 μM). A small amount of mycelium grown on PDA was added to the reaction mixture using sterilized pipette tips. We conducted additional steps to facilitate the amplification of the target regions. The reaction mixture was heat shocked for 3 min at 95°C and cooled to 25°C. The PCR tubes were vortexed vigorously for 1 min to break down the cells. The thermal cycle was as follows: an initial 1-min denaturation at 95°C, followed by 40 cycles of 30-sec denaturation at 95°C, 1-min annealing at 53°C and 56°C for ITS and TEF-1α, respectively, 1 min at 72°C, followed by a final extension of 3 min at 72°C. The amplified products (5 μl) and 100 bp Plus DNA Ladder (Bioneer) were loaded on a 1.5% (w/v) agarose gel with Tris-acetate-EDTA (TAE) buffer. The products were purified using the α+ SolutionTM GEL/PCR Purification Kit (Alphagen, Changzhi, Taiwan) and submitted to the Bioneer sequencing service. The primers used for sequencing were the same as those used for amplification. DNA sequences were used to conduct a BLAST search against the NCBI GenBank database for species identification. Conventional PCR was performed on strains for which direct PCR did not work. For conventional PCR, fungal genomic DNA was extracted using the CTAB method (Leslie and Summerell, 2006). We added 0.5 μl of the genomic DNA to the total volume of 20 μl AccuPower PCR Premix (Bioneer). The thermal cycle for conventional PCR was the same as that for direct PCR; however, there was no additional heat shock step. The purified PCR products were submitted to the Bioneer sequencing service.

To determine their pathotypes, the F. fujikuroi isolates were subjected to PCR using the primer pairs B14J06375F2/B14J05375R2 and B20J12141F2/B20J12141R2 for B14- and B20-type, respectively (Niehaus et al., 2017). The F. fujikuroi B14 strain, but not the B20 strain, reportedly produces FBs; hence, we used B14 and B20 strains as controls (Niehaus et al., 2017). We used direct PCR protocols, as described above, with a modification of the annealing temperature to 60°C. The amplified products were loaded onto 1.5% (w/v) agarose gel.

To induce mycotoxins, such as FBs, from F. fujikuroi, we cultured 17 F. fujikuroi isolates on a rice medium. In brief, 1.5 g of rice and 1 ml of distilled water were autoclaved in 8 ml vials. The rice medium was inoculated with the PDA agar blocks on which the isolates were grown. The vials were incubated for three weeks at 25°C in the dark. To extract mycotoxins, the 1.5 g of rice cultures were freeze-dried overnight and ground into a powder. Mycotoxins were extracted with 3 ml of 85% methanol with vigorous shaking for 5 min. The samples were centrifuged at 470 ×g at 4°C for 5 min, and 500 μl of the supernatant was transferred into a new tube. The extract was filtered through a Millex syringe filter (hydrophilic PTFE membrane, 0.45 μm pore size), and filtrates were stored at 4°C until analysis. The level of FBs was measured using a commercial ELISA kit (AgraQuant Fumonisin Kit; Romer Labs, Austria), according to the manufacturer’s instructions. The sample was diluted 100,000 times in distilled water to ensure that the FB concentration was within the detection range. The toxigenic analysis was repeated three times.

The TEF-1α sequence of FFSC was used for the phylogenetic analysis of FFSC. Additional sequences of FFSC strains (MN696157, MN696156, MN695931, and MN386748.1) were obtained from GenBank (Qiu et al., 2020). The TEF-1α sequence of F. oxysporum strain (KX253985.1) was also obtained from GenBank. The DNA sequences were first trimmed using SeqMan software (DNASTAR Inc., Madison, USA) and then aligned and used to construct phylogenetic trees using MEGA11 software. We used the neighbor-joining (NJ) method for constructing phylogenetic trees and evaluated the topology of the trees by the bootstrap method based on 1000 replications. Analyses of mycotoxin production using ELISA kits were performed using a Synergy HTX microplate reader (BioTek, Winooski, USA) to measure absorbance. The absorbance value was analyzed to calculate the FB concentration in the rice medium using Microsoft Excel (Microsoft, Redmond, USA) and GraphPad Prism 8 (GraphPad Software, San Diego, USA), following the manufacturer’s instructions. Scientific illustrations were created using BioRender (Biorender.com), and other figures were generated using Microsoft PowerPoint (Microsoft).

Rapid detection and identification of fungal pathogens are important. Thus, we used a direct PCR approach for the detection and identification of fungal species. The representatives of major fungal lineages were used to demonstrate the application of PCR to various fungal species. The experimental scheme is shown in Figure 1 . Using the direct PCR approach, we amplified the ITS region of representatives of major fungal lineages ( Figure 2 ). Among the 37 fungal strains, the ITS region of 34 fungal strains (92%) was successfully amplified using direct PCR. However, direct PCR could not amplify the ITS region of three fungal strains: Aspergillus brasiliensis, Aspergillus niger, and Mucor mucedo. Genomic DNA was extracted, and the ITS region was amplified as previously described (Qiu et al., 2020). The ITS sequences of all the fungal strains analyzed matched the species identification ( Table S1 ). The ITS sequences of the 37 strains were used to construct a phylogenetic tree ( Figure 2 ).

We demonstrated the usefulness of the direct PCR approach by applying it to the identification of field isolates. FFSC, an important component of the genus Fusarium, consists of distinct species, such as F. fujikuroi, F. proliferatum, and F. verticillioides (Qiu et al., 2020). These strains can cause devastating cereal diseases. Therefore, we isolated FFSC strains from the seeds of 46 rice cultivars harvested in Jeonju and Miryang in the Republic of Korea ( Figure 1 ). The 28 FFSC candidate strains formed white mycelia that turned violet or orange with age on PDA. This pigmentation was the same as that in previous reports (Leslie and Summerell, 2006). For molecular identification, we used both the ITS and TEF-1α regions because the FFSC strains were indistinguishable in their ITS sequences. The target regions were amplified using a direct PCR approach, and the amplified products were directly sequenced ( Supplementary Figure S1 ). Based on sequence similarity, 28 candidate strains were identified as F. fujikuroi (17 strains), F. concentricum (7 strains), F. incarnatum (1 strain), F. oxysporum (1 strain), F. commune (1 strain), and Colletotrichum sp. (1 strain). Interestingly, 7 out of 17 F. fujikuroi strains were isolated from a rice cultivar, Chamdongjin ( Supplementary Figure S2 ). The other 10 strains were isolated from a variety of rice cultivars: Hyunpoom, Haepoom, Boramchal, Shindongjin, Mokyang, Alchanmi, Saebonghwang, and Mipoom. A phylogenetic tree was constructed based on the sequences of the TEF-1α gene ( Figure 3 ). In the phylogenetic tree, F. fujikuroi strains were divided into two groups, B14 and B20. The B14-type group includes the B14 strain, known to produce FB (Niehaus et al., 2017). In contrast, the B20 strain in the B20-type group did not produce FB. In F. fujikuroi, multiple alignments of the TEF-1α sequences revealed that B14-type F. fujikuroi strains have thymine at nucleotide position 618 of the TEF-1α gene sequences. In contrast, B20-type F. fujikuroi strains have guanine at this site ( Figure 4 ).

Seventeen F. fujikuroi strains were analyzed to examine their FB-producing ability. The strains were grown in a rice medium to induce FB production. Using ELISA, FBs were detected in the seven F. fujikuroi strains that clustered with the B14 strain ( Table 1 ). The FB concentration was very high, ranging from 46 to 205 mg/g, in the rice medium. However, FBs were not detected in the 10 strains that clustered with the B20-type group. This indicates that FB production correlates with phylogenetic group.

A previous study reported a relationship between mycotoxin production in F. fujikuroi isolates and the secondary metabolite genes PKS51 and NRPS31 (Niehaus et al., 2017). Therefore, we examined 17 F. fujikuroi strains using pathotype-specific primer pairs amplifying PKS51 and NRPS31 fragments ( Supplementary Figure S3 ). The PCR genotyping showed that the F. fujikuroi strains were divided into two groups: B14-type and B20-type. PKS51 primer pair could successfully amplify the gene in F. fujikuroi strains producing FB, while the PCR product of NRPS31 was obtained from FB-non-producing F. fujikuroi strains. However, two strains, HS96 and HS291, had no PCR amplification although we used both direct PCR and genomic DNA extraction. They produced FB in rice medium and were grouped with the B14-type strain in the phylogenetic analysis ( Table 1 and Figure 3 ).