Background

Rice bakanae disease (RBD), caused mainly by Fusarium fujikuroi, is a devastating disease of rice worldwide (Asmaul et al. 2020). This pathogenic fungus produces secondary metabolites, including gibberellin A3, fusaric acid, and fumonisins, which pose a threat to human and animal health (Amatulli et al. 2010; Fiyaz et al. 2016). Over the past few decades, changes in climate and cropping systems have exacerbated RBD epidemics in many rice-growing regions (Bashyal et al. 2016). Unfortunately, no RBD-resistant cultivars are available, and controlling the disease relies heavily on seed treatment with fungicides. However, the extensive use of systemic fungicides has led to a significant increase in fungicide resistance in F. fujikuroi. For instance, the benzimidazole fungicide carbendazim, which was once effective in controlling RBD, has lost its efficacy in China (Chen et al. 2014). Since then, the imidazole fungicide prochloraz has been used extensively for disinfection of F. fujikuroi in rice seeds (Kim et al. 2010), but prochloraz-resistant isolates have also frequently been detected in many regions (Leroux and Gredt 1997; Yin et al. 2009). Recently, a new myosin inhibitor fungicide called phenamacril has been registered for the control of RBD and has shown excellent efficacy (Zhang et al. 2015). Unexpectedly, resistant F. fujikuroi isolates have also been detected in the field of Zhejiang Province of China since 2016 (Hou et al. 2018). Therefore, there is an urgent need to introduce new fungicides to control RBD in China and other countries around the world.

The new generation of SDHIs can efficiently inhibit the energy metabolism of many pathogenic ascomycetes and basidiomycetes. Previous studies have demonstrated that several SDHIs, including pydiflumetofen and cyclobutrifluram, are highly active against Fusarium spp. (Shao et al. 2022; Liu et al. 2023; Sun et al. 2024). SDHIs act on the mitochondrial respiratory electron transport chain complex II, succinate dehydrogenase, which is composed of four subunits SdhA, -B, -C, and -D (Sdh1–4 in yeast) in eukaryotes. This process contributes to the proton gradient by supplying reducing equivalents resulting from succinate metabolism, and the reducing equivalents are then transported through the ubiquinone pool (Moosavi et al. 2019; Wu et al. 2023). To date, several studies have shown that mutations in subunits of SdhA, SdhB, SdhC, and SdhD may confer SDHI resistance in Fusarium and other various phytopathogenic fungi. In Fusarium pseudograminearum, FpSdhC1-A83V or FpSdhC1-R86K point mutation confers pydiflumetofen resistance (Li et al. 2023). Mutations of FaSdhB-H248Y, FaSdhC1-A64V, and FaSdhC1-R67K lead to resistance of Fusarium asiaticum to pydiflumetofen (Chen et al. 2021). Sun and colleagues found that some F. fujikuroi strains showed natural resistance to penflufen, which is mainly caused by amino-acid substitutions (Q46K, R283H, G430A, and Q586R) in SdhA (Sun et al. 2021). In addition, point mutations, including SdhB-H248L/D/Y, SdhC2-A83V/H144Y, SdhC2-G80R/A83V, and SdhD-S106F/E166K, are responsible for SDHI resistance in F. fujikuroi (Liu et al. 2023; Xue et al. 2023). However, the biological functions of the Sdh subunits have not yet been systematically documented in a phytopathogenic fungus.

To understand the biological function of all Sdh subunits in Fusarium spp., we carried out a gene deletion analysis for each of the five Sdh subunits in F. fujikuroi. Our results showed that FfSdhA, B, C1, and D deletion mutants exhibited resistance to all SDHIs tested, including pydiflumetofen, penflufen, and fluopyram. Moreover, FfSdhA, B, and D, but not FfSdhC1 and -C2, played critical roles in regulating hyphal growth, conidiation, virulence, and sensitivity to CaCl2 and oxidative stresses. In addition, FfSdhC1 also showed some defects in conidiation and sensitivity to SDHIs. Although FfSdhC2 deletion did not show detectable phenotypic changes compared to the wild-type strain, we failed to obtain a dual mutant of FfSdhC1+C2. In addition, the expression of FfSdhC2 increased dramatically in the SdhC1 deletion mutant compared to the wild-type. These results indicate that two paralogous FfSdhC genes are partially functional complementary in F. fujikuroi.

Results

The FfSdh subunits in F. fujikuroi

To determine whether the putative Sdh genes are expressed properly in F. fujikuroi, we first downloaded the SdhA (accession No. FFUJ_02582), SdhB (FFUJ_08745), SdhC1 (FFUJ_08951), SdhC2 (FFUJ_12139), and SdhD (FFUJ_00494) nucleotide sequences from Fungi DB (https://fungidb.org/fungidb/app), and identified the positions of introns and protein domains for these Sdh subunits. The nucleotide sequences of FfSdhA, -B, C1, -C2, and -D genes are 2272, 960, 660, 687, and 679 bp in length, respectively. The FfSdhA gene contains five introns and encodes 649 amino acids. For FfSdhB, FfSdhC1, FfSdhC2, and FfSdhD, each bears two introns (Additional file 1: Figure S1a), and encodes 278, 187, 192, and 182 amino acids, respectively (Additional file 1: Figure S1b). The locations of the introns in the FfSdh genes were verified by sequencing the 1802, 247, 180, 215, and 365-bp cDNA. Reverse transcription PCR results showed that FfSdhA, B, C1, C2 and D genes were able to be transcribed in F. fujikuroi (Fig. 1).

Fig. 1
figure 1

Identification of introns and expression of SDHs in F. fujikuroi. Amplification of FfSdhA, B, C1, C2, and D gene from genomic DNA and cDNA of the wild-type isolate JS16 using primers A-F + A-R, B-F + B-R, C1-F + C1-R, C2-F + C2-R, and D-F + D-R. The primer pairs generated the expected 2124, 370, 276, 323 and 495 bp from genomic DNA and 1,802, 247, 180, 215, and 365 bp from cDNA, respectively

Full size image

F. fujikuroi and other Fusarium spp. contains paralogous SdhC1 and -C2

Alignment assay showed that the deduced amino acids of FfSdhC1 exhibit a low degree of sequence identity (36.98%) with those of FfSdhC2 (Additional file 1: Figure S2a). Both FfSdhC1 and FfSdhC2 have three predicted transmembrane segments, specifically at the residues 83–106, 131–53, and 164–84 for FfSdhC1 and at the residues 82–109, 135–159, and 173–188 for FfSdhC2. They were predicted to localize in mitochondria by using the WoLF PSORT. Moreover, several other Fusarium species also contain two paralogous SdhCs with high identities for FfSdhC1 and -C2 (Additional file 1: Figure S2b, c), but the number of SdhC orthologues showed variation (1 to 3) among non Fusarium fungi (Additional file 1: Figure S2d, e). These results indicate that the orthologs of SdhC1 and SdhC2 are conserved in Fusarium spp.

Characterization of the FfSdh subunit mutants and expression of the SdhC genes

To investigate the functions of each FfSdh subunit in F. fujikuroi, we generated the gene deletion mutants using a homology recombination strategy (Additional file 1: Figure S3a, b). For each Sdh gene, at least two deletion mutants with similar phenotypes were identified by PCR with the corresponding primer pairs (Additional file 2: Table S1). As shown in Fig. 2a, the FfSdhA, B, and D deletion mutants grew dramatically slower than the wild-type JS16. The growth defect of ΔFfSdhA, -B, -D was restored to a wild-type level in the complementation strain, indicating that the hyphal growth defect was due to the deletion of SdhA, B, D gene. The FfSdhC1 deletion mutant had only a 17.93% reduction in hyphal growth compared to JS16 on PDA plates (Fig. 2b), while no growth defect was observed for the FfSdhC2 deletion mutant. These results indicate that FfSdhA, B, and D, and to a lesser extent, FfSdhC1, play an important role in F. fujikuroi growth.

Fig. 2
figure 2

Impact of the FfSdh deletion mutants on colony morphology and relative expression of FfSdhC1 and FfSdhC2 in the FfSdhC deletion mutants. a The wild-type JS16, FfSdhA, B, C1, C2, and D deletion mutants and complementation strains ΔFfSdhA-C, ΔFfSdhB-C, and ΔFfSdhD-C were grown on CM, MM, PDA at 25℃ for 7 days. b Mycelial growth diameter of each strain on CM, MM, and PDA plate was determined after 7 days of incubation. Line bars in each column denote the standard errors of three repeated experiments. Different letters represent statistically significant differences for each medium at P = 0.05 according to the one-way ANOVA test. c Relative expression of SdhC genes in each mutant is the relative amount of mRNA of each gene in the wild-type JS16. Line bars in each column denote the standard errors of three repeated experiments. Different letters represent statistically significant differences at P = 0.05 according to the one-way ANOVA test

Full size image

Although FfSdhC2 is not directly involved in regulating F. fujikuroi growth, we tried to generate a FfSdhC1 and FfSdhC2 double mutant in ΔFfSdhC2 background. While 210 G418-resistant transformants were recovered, all were ectopic mutants, suggesting that FfSdhC1 and FfSdhC2 might be essential for F. fujikuroi growth and partially functional complementary because the highly efficient homologous integration events in Fusarium. To further verify this, we determined the expression of FfSdhC1 and -C2 in ΔFfSdhC2 and ΔFfSdhC1 by real-time PCR, and the results revealed that expression of FfSdhC2 in ∆FfSdhC1 was increased more than three hundred-fold; while expression of FfSdhC1 in the ∆FfSdhC2 was similar to that in the wild-type JS16 (Fig. 2c).

Sensitivity of the FfSdh deletion mutants to SDHIs and other fungicides

Five different chemical groups of SDHIs, including thifluzamide, boscalid, pydiflumetofen, penflufen, and fluopyram exhibited variable activity against F. fujikuroi. While pydiflumetofen exhibited very high inhibitory activity against JS16, thifluzamide and boscalid at 10 µg/mL did not show obvious inhibitory activity against JS16 (Additional file 1: Figure S4a, b), and penflufen and fluopyram inhibited JS16 growth to a lesser extent. Compared to the wild-type, ΔFfSdhC2 did not show detectable changes in sensitivity to pydiflumetofen, penflufen, and fluopyram. However, ΔFfSdhA, -B, -C1, -D significantly reduced sensitivity to these three SDHIs (Fig. 3a, b). In addition, the FfSdh mutants did not change their sensitivity to other groups of fungicides dramatically, including the benzimidazole fungicide carbendazim, triazole fungicide tebuconazole, phenylpyrrole fungicide fludioxonil and cyanoacrylate fungicide phenamacril (Fig. 3c, d). To further verify this finding, we determined expression patterns of FfSdh genes under a SDHI treatment by real-time PCR assay. As shown in Fig. 3e, expression levels of FfSdhA, B, C1, and D were up-regulated 2–4-fold after the wild-type JS16 was treated with 1.5 µg/mL (EC90) pydiflumetofen for 4 h as compared to those of non-treatment, but the expression level of FfSdhC2 was up-regulated by 7 to 10-fold. These results indicate that FfSdh subunits may be specifically associated with the sensitivity to SDHIs but not other groups of fungicides.

Fig. 3
figure 3

Comparisons in sensitivity to different fungicides at EC50 concentration among the wild-type JS16 and the FfSdh deletion mutants. a Mycelial growth phenotypes of the wild-type JS16, the FfSdh gene deletion mutants, and the complementation strains treated with SDHIs, including pydiflumetofen, penflufen, and fluopyram. The strains were all incubated on YBA at 25℃ for 7 days. b Inhibition rate of pydiflumetofen, penflufen, and fluopyram against each strain after 7 days of incubation. Line bars in each column denote the standard errors of three repeated experiments. Different letters represent statistically significant differences for each fungicide at P = 0.05 according to the one-way ANOVA test. c Mycelial growth phenotypes of the wild-type JS16, the FfSdh gene deletion mutants, and the complementation strains treated with non-SDHI fungicides. The strains were incubated on CM at 25℃ for 7 days. d Inhibition rate of non-SDHIs against each strain. Line bars in each column denote the standard errors of three repeated experiments. Different letters represent statistically significant differences for each fungicide at P = 0.05 according to the one-way ANOVA test. e Relative transcription levels of the FfSdh gene in JS16 after pydiflumetofen treatment. Each strain was cultured in YEPD lipid medium for 12 h and then treated with pydiflumetofen (1.5 µg/mL) for 4 h. FfACTIN was used as an internal control. Different letters represent statistically significant differences for each fungicide at P = 0.05 according to the one-way ANOVA test

Full size image

Sensitivity of the FfSdh deletion mutants to various stresses

As shown in Fig. 4a, the FfSdhC1 and -C2 deletion mutants, the complementation strain ΔFfSdhA-C, ΔFfSdhB-C, and ΔFfSdhD-C were morphologically indistinguishable from the wild-type isolate JS16 on CM plates supplemented with NaCl (osmotic stress), Congo red (cell wall stress), CaCl2, and oxidative stresses. However, the ΔFfSdhA, -B, and -D mutants showed enhanced sensitivity to CaCl2 and were also hypersensitive to H2O2 (Fig. 4b). These results indicate that FfSdhA, B, and D are involved in regulating calcium and oxidative stresses in F. fujikuroi.

Fig. 4
figure 4

Sensitivity of the FfSdh gene deletion mutants to various stresses. a Mycelial growth phenotypes of the wild-type JS16, the FfSdhA, B, C1, C2, and D gene deletion mutants and the complementation strains under various stress conditions. b Inhibition rate of mycelial growth by different stresses. Line bars in each column denote the standard errors of three repeated experiments. Different letters represent statistically significant differences for each stress at P = 0.05 according to the one-way ANOVA test

Full size image

Involvement of the FfSdh in sporulation

Conidia of F. fujikuroi have an important role in the disease cycle of rice bakanae disease. As shown in Fig. 5, there was no significant difference in conidiation between ΔFfSdhC2 and the wild-type isolate JS16. However, the ΔFfSdhA, -B, and -D mutants did not produce conidia at all in the tested condition. Compared to JS16, ΔFfSdhC1 showed a decrease in conidial production. These results indicate that FfSdhA, B, and D play a critical role in conidiation and that FfSdhC1 is also involved in the regulation of sporulation in F. fujikuroi.

Fig. 5
figure 5

Sporulation of the wild-type JS16, the FfSdh gene deletion mutants, and the complementation strains ΔFfSdhA-C, FfSdhB-C, and FfSdhD-C. Line bars in each column denote standard errors of three repeated experiments. “n.d.” represents no conidia detected. Different letters represent statistically significant differences at P = 0.05 according to the one-way ANOVA test

Full size image

Virulence of the FfSdh deletion mutants

The wild-type JS16, ∆FfSdhC1, -C2, and the complementation strains caused typical symptoms of RBD, and the rice seedlings inoculated with these strains were significantly longer than those of the non-inoculated control, while the ∆FfSdhA, -B, and -D mutants significantly reduced in virulence (Fig. 6a, b). This result showed that FfSdhA, B, and D are indispensable for the virulence of F. fujikuroi.

Fig. 6
figure 6

Pathogenicity assay for the FfSdh gene deletion mutants. a Symptoms caused by the FfSdh gene deletion mutants on rice seedlings. b The length of rice seedlings inoculated with the JS16 or each of the FfSdh deletion mutants. CK indicates the seedling length of rice without inoculation. Line bars in each column denote the standard errors of three repeated experiments. Different letters represent statistically significant differences at P = 0.05 according to the one-way ANOVA test

Full size image

Discussion

In the current study, five SDH genes (FfSdhA, B, C1, C2, and D) were identified in F. fujikuroi. Phenotypic assays for the FfSdh gene deletion mutants showed that FfSdhA, B, and D play critical roles in hyphal growth, conidiation, and virulence in F. fujikuroi. Meanwhile, the FfSdhA, B, and D gene deletion mutants were hypersensitive to H2O2, which can be produced by host plants in response to fungal infection (Shao et al. 2016). Furthermore, the FfSdhA, B, and D deletion mutants exhibited resistance to all SDHIs tested, including pydiflumetofen, penflufen, and fluopyram. The occurrence of fungicide resistance is usually accompanied by fitness costs, which may influence resistance build-up in the absence of fungicide selection pressure (Liu et al. 2022). Thus, phenotypic characterization of ∆FfSdhA, -B, and -D are useful in evaluating the role of FfSdhA, B, and D in the development of SDHIs resistance. In addition, we also found that deletion of FfSdhC1 led to slightly decreased mycelial growth and sporulation, while the FfSdhC2 deletion mutant showed similar mycelial growth, conidation, and virulence. However, we could not obtain a double mutant of FfSdhC1+C2, similar to those reported in F. graminearum (Sun et al. 2023). These results suggest that the FfSdhC1 and C2 genes are partial functional complementarity in Fusarium spp..

It has been proposed that the paralogous genes might have evolved through duplication from an ancestral gene followed by functional specialization under certain stresses (Magadum et al. 2013). In general, paralogous genes resulting from gene duplication are a powerful source of genetic novelty and provide new substrates for evolution (Iohannes and Jackson 2023). However, the functions of gene duplication in fungicide resistance are largely unknown. In this study, we observed that the expression of FfSdhC2 in ∆FfSdhC1 was increased dramatically, which may lead to resistance of ∆FfSdhC1 to SDHIs. Recently, Steinhauer et al. found that some Zymoseptoria tritici isolates are naturally resistant to the stretched heterocycle amide group of SDHIs, which is caused by a gene paralog of SdhC (termed ZtSdhC3) encoding for an alternative C subunit of succinate dehydrogenase (alt-SdhC). The alt-SdhC associates with the three other Sdh subunits, leading to a fully functional enzyme. Importantly, a unique Qp-site residue within the alt-SdhC protein confers SHA-SDHIs resistance. Moreover, transposon insertions in the promoter of ZtSdhC3 are also associated with the most robust resistance phenotypes. These results establish that a dispensable paralogous gene determines SDHI fungicide resistance in natural pathogen populations (Steinhauer et al. 2019). Thus, it would be interesting to understand the roles of target paralogs in the sensitivity of many other fungal pathogens to fungicides.

In the current study, we found that the FfSdhC2 mutant did not alter its sensitivity to SDHIs. However, the FfSdhC1 mutant showed increased tolerance to SDHIs but not to other groups of fungicides. These results suggest that FfSdhC1, but not C2, might play a critical role in regulating SDHI sensitivity in F. fujikuroi. In line with this finding, Sun and colleagues found that as an FgSdhC paralog, FgSdhC1 is responsible for the natural resistance or less sensitivity of F. graminearum to a majority of SDHI fungicides (Sun et al. 2023), suggesting that the dispensable paralogous gene SdhCs may play an important role in mediating SDHI resistance in Fusarium populations. To date, 24 SDHIs have been launched in the market for controlling various fungal diseases. Since different SDHIs may target different SDH complexes, it is reasonable to believe that cross-reactivity patterns are complex among SDHIs. We expect that this finding will trigger more research on the development of SDHI resistance in pathogenic Fusarium and other fungal species.

Conclusions

In this study, we identified five Sdh genes in F. fujikuroi and carried out a gene deletion for FfSdhA, B, C1, C2, and D. Through phenotypic assays, we found that FfSdhA, B, C1, and D plays critical roles in regulating fungal growth and sensitivity to SDHIs. Although the FfSdhC2 deletion mutant did not show detectable phenotypic changes as compared to the wild-type strain, we failed to obtain a double mutant of FfSdhC1+C2, indicating that the two paralogous SdhC genes might be partially functional complementary, and are required for fungal growth in F. fujikuroi. The findings of this study provide useful information for enhancing our understanding of the functions of Sdh in pathogenic fungi.

Methods

Strain, media, and fungicides

The F. fujikuroi isolate JS16 collected from Zhejiang, China, was used as a wild-type strain for gene deletion experiments. This strain and target gene deletion mutants were grown on potato dextrose agar (PDA) (200 g potato, 20 g dextrose, 18 g agar, and 1 L water) or minimal medium (MM) (10 mM K2HPO4, 10 mM KH2PO4, 4 mM (NH4)2SO4, 2.5 mM NaCl, 2 mM MgSO4, 0.45 mM CaCl2, 9 mM FeSO4, 10 mM glucose, 10 g agar, and 1 L water, pH 6.9) for mycelial growth assay. The completed medium (CM) (10 g glucose, 2 g peptone, 1 g yeast extract, 1 g casamino acids, 6 g NaNO3, 0.52 g KCl, 0.52 g MgSO4⋅7 H2O, 1.52 g KH2PO4, 1 mL trace elements, 0.01% of vitamins, 10 g agar, and 1 L water, pH 6.5) (Correll et al. 1987) was used for non-SDHI fungicide sensitivity assay. Yeast bacteriological peptone agar medium (YBA) (10 g bacteriology peptone, 10 g yeast extract, 20 g sodium acetate, and 15 g agar in 1 L water) was used to determine the sensitivity of F. fujikuroi to SDHIs (Ishii et al. 2016). Yeast extract peptone dextrose medium (YEPD) (10 g peptone, 3 g yeast extract, 20 g D-glucose, made up to 1 L, pH 7.0) was used for spore production measurement.

Penflufen (22.4% active ingredient) and fluopyram (41.7%) were provided by Bayer Crop Sciences Co. Pydiflumetofen (200 g/L), boscalid (50%), and thifluzamide (98%) were obtained from Syngenta, BASF and Zhejiang Hangzhou Yulong Chemical Co., respectively. Thifluzamide was dissolved in methanol to obtain a stock solution of 5 mg of active ingredient/mL.

Verification of Sdh introns and analysis of Sdh gene expression in F. fujikuroi

F. graminearum contains five putative succinate dehydrogenase subunits (FgSdhA, B, C, and D with accession No. FGRAMPH1_01G24629, FGRAMPH1_01G18325, FGRAMPH1_01G04785, FGRAMPH1_01G28127, and FGRAMPH1_01G01879 in the FungiDB, respectively). Based on these five putativeFgSdhs, we obtained the FfSdhA, FfSdhB, FfSdhC1, FfSdhC2, and FfSdhD from F. fujikuroi genome by homologous comparison. FfSdhA, B, C1, C2, and D were predicted to possess five, two, two, two, and two introns, respectively. To verify the existence and size of these predicted introns, RNA was extracted from the mycelia of the wild-type isolate JS16 of F. fujikuroi using TRIzol (TaKaRa Biotechnology, Dalian, China) according to the manufacturer’s instructions. The first strand of cDNA was synthesized using 1 µg of total RNA per sample by HiScript II Q RT SuperMix kit (R223–01, Vazyme, Nanjing, China) according to the manufacturer’s instructions. PCR amplifications were performed with the primer pair A-F + A-R, B-F + B-R, C1-F + C1-R, C2-F + C2-R, and D-F + D-R (Additional file 2: Table S1) for FfSdhA, B, C1, C2, and D, respectively, using the JS16 cDNA and DNA as templates.

Relative expression levels of FfSdh genes were measured by real-time PCR assay. Briefly, each strain was cultured in 100 mL YEPD liquid medium for 12h at 25°C. Then, fresh mycelia were harvested for total RNA extraction and cDNA synthesis. Expression levels of Sdh genes were determined with qRT-PCR using HiScript II QRT SuperMix (Vazyme Biotech, Nanjing, China). For each sample, PCR amplifications with primer pair Ff-actin-F + Ff-actin-R were used as the endogenous control. The primers used for qRT-PCR assays are listed in Additional file 2: Table S1. The relative expression levels of target genes were calculated using the 2−ΔΔCt method (Kl 2001). The experiments were independently performed three times.

Phylogenic analysis of SdhC proteins in F. fujikuroi and Fusarium spp.

The sequence identities of FfSdhC1 and FfSdhC2 were analyzed using the DNAMAN software, and the locations of the transmembrane regions of FfSdhC1 and FfSdhC2 were analyzed using the DeepTMHMM (https://dtu.biolib.com/DeepTMHMM). Localization of FfSdhC1 and FfSdhC2 was predicted by using the WoLF PSORT (https://wolfpsort.hgc.jp/). Fusarium spp. orthologs of FfSdhC1 and FfSdhC2 were retrieved in ENSEMBL BLAST (https://fungi.ensembl.org/Multi/Tools/Blast), respectively. The identity of SdhC orthologues from Fusarium spp. or non Fusarium spp. was analyzed by the DNAMAN software.

Generation of deletion and complementation mutants

The target gene deletion mutants ΔFfSdhA, -B, -C1, -C2, and -D were constructed using a protocol described previously (Yun et al. 2014). Briefly, the open reading frame (ORF) of FfSdhs was replaced with hygromycin (HPH). Transformation for F. fujikuroi was performed via the polyethyleneglycol (PEG) mediated protoplast transformation method (Proctor et al. 1995). For ΔFfSdhA, -B, and -D complementation, ORFs of FfSdhA, B, and D fused with a geneticin (G418) resistance cassette were introduced into the corresponding mutant. To construct the ΔFfSdhC1+C2 dual-deletion mutant, the FfSdhC1 gene was tried to be knocked out in the ΔFfSdhC2 using geneticin (G418) resistance as a selection marker. All of the deletion mutants were verified by PCR with relevant primers listed in Additional file 2: Table S1.

Sensitivity of the FfSdh deletion mutants to SDHIs and non-SDHIs fungicides

To determine sensitivity of the FfSdh deletion mutants to SDHIs, a mycelial plug (5 mm in diameter) taken from the periphery of a 3-day-old colony was placed on the center of a YBA plate amended with pydiflumetofen at 0.03 µg/mL, penflufen at 0.20 µg/mL, and fluopyram at 0.35 µg/mL, respectively. The concentrations of other fungicides are indicated in the corresponding figures. In addition, to evaluate the sensitivity of the FfSdh deletion mutants to non-SDHIs, a mycelial plug (5 mm in diameter) taken from the periphery of a 3-day-old colony was placed in the center of a CM plate amended with carbendazim at 0.5 µg/mL, phenamacril at 0.17 µg/mL, tebuconazole at 0.07 µg/mL or fludioxonil at 0.045 µg/mL, respectively. Two replicates for each concentration were used for each mutant. After incubation at 25°C for 7 days, the colony diameter in each plate was measured in two perpendicular directions with the original mycelial plug diameter (5 mm) subtracted from each measurement. The experiment was repeated three times.

Stress assay

Phenotypic assay was performed on CM supplemented with NaCl, CaCl2, H2O2, or Congo red at various concentrations. Each plate was inoculated with a mycelial plug (5 mm in diameter) taken from the periphery of a 3-day-old colony of each isolate/mutant. After incubation at 25°C for 7 days in the dark, colony diameter was examined for each plate. The experiment was repeated three times.

Sporulation assay

For the sporulation assay, three mycelial plugs (5 mm in diameter) of each isolate/mutant taken from the periphery of a 3-days-old colony were inoculated in a 50 mL flask containing 30 mL of YEPD medium. The flasks were incubated at 25°C for 36 h in a shaker (180 rpm). For each isolate/mutant, the number of conidia in the broth was determined using a hemacytometer. The experiment was repeated three times.

Virulence assay

The RBD susceptible variety Zhongzao 39 was used for the virulence assay. The artificial inoculation of F. fujikuroi referred to a previously published protocol (Zhao et al. 2022). Briefly, after sterilization with 3% H2O2 for 2 h, rice seeds were further soaked for 2 days and then germinated for 1 day. Each sprouted seed together with a mycelial plug of each F. fujikuroi mutant (a thin layer of vermiculite was used to separate seed with mycelial plug) was incubated into a sterilized vermiculite test tube. Five replicates were used for each strain. Samples were then cultured in an incubator with 12 h of light and 12 h of darkness at 30°C. After 15 days of inoculation, the lengths of rice seedlings were examined and photographed. The experiment was repeated three times.