Genetic diversity of field Fusarium asiaticum and Fusarium graminearum isolates increases the risk of fungicide resistance

Abstract

Fusarium head blight (FHB) caused by Fusarium species, seriously threatens the safety of wheat (Triticum aestivum) production. Resistant cultivars and fungicides are frequently used to control these FHB pathogens. However, Fusarium species have been adapting the current FHB control approaches in a manner that raises concern for future FHB control strategies, which could lead to a greater risk of FHB outbreaks. In this study, a total of 521 strains of Fusarium were isolated from Sichuan province of China, to investigate the diversity of Fusarium species and the genes associated with their adaptation. Seven species were identified based on molecular markers and morphological analysis. The virulence assays showed that Fusarium asiaticum (Fa) and Fusarium graminearum (Fg) were the two major causal agents of FHB, with high virulence and more frequent isolates. Fungicide resistance analysis showed that four isolates had developed the resistance to carbendazim, and four isolates had developed the resistance to tebuconazole. Of note, two point-mutation variants (F200Y and E198Q) occurred in the β2-tubulin gene, leading to the carbendazim resistance. The landscape of genomic diversity was analyzed through whole-genome sequencing, revealing a total of 182,811 and 430,733 variants (including: single nucleotide polymorphisms, SNP, insertion and deletion, Indel, and structure variation, SV) among the Fa and Fg isolates, respectively. In addition, potential alterations in gene function (15.22%) were predicted among Fg variants. These alterations offer potential helps for the Fusarium species to adapt to various managements of FHB, which may increase risks in developing fungicide-resistant isolates. However, these annotated genetic variants are valuable resources for further genetic and genomic studies, as well as potential markers to assist disease risk assessment.

Background

Wheat (Triticum aestivum) is one of the most important staple crops in the world. However, Fusarium head blight (FHB) seriously threatens the safety of wheat production worldwide. It mainly infects the wheat head and absorbs nutrients, resulting in considerable yield reduction and the deterioration of flour quality (Bottalico and Perrone 2002; Salgado et al. 2015). After infection of FHB, mycotoxins, including NIV, DON, 3-ADON, and 15-ADON, are produced by Fusarium in wheat seeds (Cheat et al. 2015). These mycotoxins can cause anorexia, diarrhea, vomiting, and gastrointestinal bleeding in animals, and cannot be easily removed by temperature or chemical and physical treatments (Marin et al. 2013; Thapa et al. 2021). A great deal of efforts has been devoted to avoid the exposure of mycotoxin. However, human or animal consuming of food contaminated by mycotoxins still occasionally occurs (Al-Jaal et al. 2019). Due to global warming and cropping system changes, frequent FHB outbreaks have caused significant economic losses in wheat globally, and the situation appears getting worse (Vaughan et al. 2016; Xu et al. 2021).

FHB pathogens are complex and diverse. More than 19 phylogenetically distinct species have been reported to cause FHB on wheat (Leslie and Summerell 2006; Van der Lee et al. 2015). Different climatic conditions and crop rotations can lead to differences in the distribution and predominance of the pathogens. For instance, Fusarium graminearum (Fg) and Fusarium asiaticum (Fa) are the major pathogens under warmer climate conditions (Xu et al. 2021), while Fusarium culmorum, Fusarium nivale, and Fusarium poae are the major pathogens under cooler climate conditions (Dweba et al. 2017). Fa is the predominant species in the southern wheat-rice rotation regions of China, but Fg is the predominant species in the Northern China where wheat and maize rotation is very common (Zhang et al. 2012). Furthermore, Fusarium species isolated from different regions show differences in morphology, pathogenicity, and fungicide resistance (Xu et al. 2021; Wang et al. 2022). Therefore, an investigation of the diversity of Fusarium is valuable for evaluating the severity of FHB.

Various control strategies have been used to prevent Fa and Fg from infecting wheat, including timely fungicide application, resistance germplasm deployment, and special cultural practices. However, the genetic diversity of Fa and Fg can counteract the efficiency of these control strategies (Zeller et al. 2004; Talas et al. 2015a; Yang et al. 2020). There is a line of evidence that some Fa and Fg have adapted to FHB control strategies, especially fungicides (e.g., Tebuconazole, TEC, Carbendazim, MBC) (de Chaves et al. 2022). For example, point mutations at β1-tubulin (FGSG_09530) and β2-tubulin (FGSG_06611) have been reported to be associated with resistance to MBC in Fusarium (Chen et al. 2015). Spolti et al. (2014) isolated a Fg strain (Gz448NY11) from Steuben county, New York, USA, showing a resistance to TEC. In Anhui province of China, 8.23% of single-spore isolates of Fa were found to be resistant to MBC, and five types of point mutations (F167Y, E198L, E198K, F200Y, and E198Q) in the β2-tubulin gene conferred resistance to MBC (Chen et al. 2015). Previous studies have reported that CYP51-A (FGSG_04092) and CYP51-B (FGSG_01000) genes are related to the resistance to TEC in Fusarium by point mutations that cause overexpression of CYP51-A and CYP51-B genes (Ma et al. 2006; Qian et al. 2018). Qian et al. (2018) confirmed that the point mutation (Y137H) of CYP51-B led to the resistance to TEC by site-directed mutagenesis in Fg strain PH-1. A total of 150 TEC-resistant Fg strains were obtained from different areas of Henan province in China in 2018–2020, and six resistant strains possessed an amino acid mutation (S169T) in CYP51-B (Chen et al. 2021). Notably, there are no reports showing that overexpression of CYP51-A or CYP51-B could cause the resistance to TEC in Fusarium.

Whole-genome sequencing can provide information on genetic variation, such as single nucleotide polymorphisms (SNP), insertion and deletion (Indel), and structure variation (SV). Since the genome information of Fg is available, more than 10,000 SNPs have been identified, which are preferentially located at the ends of chromosomes or in inner chromosomal locations (Cuomo et al. 2007). Walkowiak et al. (2016) found 704,566 SNPs and Indels among 10 closely related members of the Fg species complexed with different mycotoxin genotypes. Similarly, Laurent et al. (2017) found 242,756 high-confidence genetic variants in six French isolates of Fg via whole-genome sequencing. Nevertheless, a large number of genetic variants suggests that a small number of isolates is not sufficient to explain all phenotypic, pathogenicity, and fungicide resistance variants. Hence, more isolates of Fa and Fg are required for discovering the genetic diversity, and the variants are potential markers for tracking the spreading of Fusarium populations, and aiding the assessing of the risk of FHB outbreaks (Oghenekaro et al. 2021).

The complex climate conditions and cultural practices in Sichuan province are conducive to the emergence of various Fusarium species (Huang and Ye 2005), which makes fungicide control strategies being less effective. In this study, we aimed to survey Fusarium species in Sichuan province, and to identify major pathogens of FHB using virulence assays. All Fusarium isolates can be used to investigate genetic variations by whole-genome sequencing. The fungicides MBC and TEC were used to verify the link between genetic variation and the diversity of biological functions, which could help us understand the diversity of Fusarium species and the genes associated with their adaptations to counteract current and future FHB control strategies, and for building the prediction model for potential FHB outbreaks.

Results

Identification of Fusarium species from wheat spikelet with visible FHB signs

A total of 521 strains of Fusarium species were obtained from wheat spikelet with visible FHB symptoms (Additional file 1: Table S1). The distribution of all isolates was shown in Fig. 1a. One or more molecular markers from each isolate were successfully obtained by PCR amplification and sequencing. These isolates had more than 98% sequence similarity with Fa, Fg, F. meridionale, F. avenaceum, F. tricinctum, F. flocciferum, or F. proliferatum (Additional file 1: Table S1). The identities of these seven Fusarium species were further confirmed by morphological analyses based on the descriptions of Leslie and Summerell (2006) (Fig. 1b). The frequency assays for each Fusarium species showed that 68.58% of them were Fa species and 27.59% were Fg species. Other species (including F. meridionale, F. avenaceum, F. tricinctum, F. flocciferum, and F. proliferatum) account for only 3.83% of the total isolates (Fig. 1c). The Fa isolates were the major pathogenic strains in the year of 2021. The Fg isolates were the major pathogenic strains in the year of 2022 (Fig. 1c). Therefore, Fa and Fg are the pathogens with a high frequency of FHB in Sichuan province.

Fig. 1

Identification of FHB in Sichuan province of China. a Distribution of the obtained Fusarium isolates in Sichuan province. b Morphological analyses of Fusarium species on PDA plates at the 7th day after inoculation (Scale bars = 1 cm). c Frequency of each Fusarium species. The detail information of each isolate can be found in Additional file 1: Table S1

Pathogenicity of Fusarium species

Our results showed that all isolates tested were able to infect the wheat spikelet (Fig. 2 and Additional file 1: Table S1). However, significant differences in infectivity were observed between different Fa or Fg isolates, where F. meridionale, F. avenaceum, F. tricinctum, F. lateritium, F. flocciferum, and F. proliferatum showed lower infectivity when compared with Fa and Fg isolates (Fig. 2 and Additional file 1: Table S1). These results indicate that Fa and Fg are highly pathogenic to wheat, despite there is a virulence diversity among different Fa or Fg isolates.

Fig. 2

Virulence assays on wheat. a Wheat heads were inoculated with conidial suspensions of Fusarium isolates. Infected wheat heads were photographed at 7th day after inoculation. Red arrows indicated the inoculation sites. Red dashed line marked the area of disease. b The numbers of infected and bleached spikelets at 7th day after inoculation. Values are means ± standard deviation of ten biological replicates per isolate. The isolate information was listed in Additional file 1: Table S1

Fungicide resistance in Fa and Fg isolates

Due to the reduced effectiveness of MBC and TEC on controlling the FHB in Sichuan province, all Fa and Fg isolates were analyzed using a higher concentration of MBC (50 μg/mL) and TEC (36 μg/mL) than previously reported (Yin et al. 2009; Qian et al. 2018; Chen et al. 2021). The results showed that 50 μg/mL of MBC completely lost the ability to inhibit the growth of Fa Lz189, Fa Lz263, Fa Lz503, and Fg Lz114 isolates, although the reference Fg isolate PH-1 could not grow. However, the growth of Fa Lz136, Fa Lz167, Fa Lz201, and Fg Lz179 isolates was partially inhibited in 36 μg/mL TEC compared with Fg PH-1 which could not grow either (Fig. 3a and Additional file 1: Table S1). Thus, Fa Lz189, Fa Lz263, Fa Lz503, and Fg Lz114 have evolved resistance to MBC, and Fa Lz136, Fa Lz167, Fa Lz201, and Fg Lz179 have evolved resistance to TEC.

Fig. 3

Fa and Fg isolates showed resistance to MBC and TEC. a Mycelial growth of Fg Lz114, Fa Lz503, Fa Lz167, and Fg Lz179 on PDA plates with the treatments of MBC and TEC at the 7th day after inoculation (Scale bars = 1 cm). Fg PH-1 was used as a control. CK means no addition of fungicide. Three biological replicates for each experiment. b Multiple protein sequence alignment of β2-tubulin protein sequence. Black arrows indicate the mutation sites. c Multiple protein sequence alignment of CYP51-B protein sequence. Fa Lz45, a control with same protein sequence, could not grow in 36 μg/mL TEC. d Relative expression levels of CYP51-A and CYP51-B in Fg Lz179. Mycelia were collected at 7th day after incubation on PDA plates supplied with 36 μg/mL TEC. There were three biological replicates. Asterisk indicates significance at P < 0.05

Mutation of β2-tubulin is associated with MBC resistance

Point mutations in the β1-tubulin and β2-tubulin genes have been reported to be associated with resistance to MBC in Fusarium species (Chen et al. 2015). The open reading frames of the β1-tubulin and β2-tubulin genes were identified by PCR and sequencing. The results showed that β1-tubulin gene had no changes when compared with the reference gene sequence. For the open reading frame of the β2-tubulin gene, point mutations at codons 198 (GAG → GCG, E198A) and 200 (TTC → TAC, F200Y) in Fa Lz189, Fa Lz263, and Fa Lz503, and a point mutation at codon 198 (GAG → GCG, E198A) were found in Fg Lz114 (Fig. 3b). Other sensitive strains were further checked, but they did not carry these mutations (data not shown). Thus, the mutations of β2-tubulin at codons 198 (GAG → GCG, E198A) and 200 (TTC → TAC, F200Y) are related to resistance to MBC in Fa and Fg.

Mutations of CYP51-B is associated with TEC resistance

The point mutations or overexpression of CYP51-A and CYP51-B genes are associated with resistance to TEC in Fusarium species (Ma et al. 2006; Qian et al. 2018). Therefore, the open reading frame of CYP51-A and CYP51-B genes were identified by PCR and sequencing. The results showed that CYP51-A gene had no changes compared with the reference gene sequence. For the CYP51-B gene, Fa Lz136, Fa Lz167, Fa Lz201, and Fg strain Lz179 showed no changes in CYP51-B gene compared with the Fa reference sequence and those Fa isolates that could not grow at 36 μg/mL TEC (Fig. 3c). We then investigated the expression of CYP51-A and CYP51-B genes by real-time PCR in Fa strains Lz136, Fa Lz167, Fa Lz201, and Fg strain Lz179. The results showed that CYP51-A exhibited a significantly higher expression level than Fg PH-1, but CYP51-B showed a significantly lower expression level than Fg PH-1 in Fg strain Lz179 (Fig. 3d).

Genetic diversity of the Fa and Fg isolates

To determine the sequence diversity in the two species, we sequenced the Fa and Fg isolates collected from Sichuan province. The whole-genome sequencing results showed that Fg had 99.31% genome coverage (≥ 4 ×) to the Fg reference genome, and Fa had 99.18% genome coverage (≥ 4 ×) by alignment with the Fa reference genome. These results show that the Fa and Fg isolates are secured, and can be used for further genetic variation analysis. By genetic variation analysis, a total of 165,888 SNPs, 10,714 Indels, and 6209 SVs variants were discovered in Fa isolates, respectively. The details information is in Additional file 1: Table S2. SNPs were evenly distributed in all four chromosomes, and mainly distributed at the ends of the chromosomes (Fig. 4a).

Fig. 4

Genetic variation analysis in Fa and Fg isolates. a Distribution of SNPs within 0.05 Mb window size in chromosome among Fa isolates. b Distribution of SNPs within 0.05 Mb window size in chromosome among Fg isolates. c Distribution of InDels among Fa isolates. d Distribution of InDels among Fg isolates. e Distribution of SVs. Different color columns represented different type of SVs among Fa isolates. f Distribution of SVs length. Different color columns represented different type of SVs among Fg isolates. All detail information can be found in Additional file 1: Tables S2 and S3

Among Fg isolates, a total of 379,318 SNPs, 25,356 Indels, and 8059 SVs were found. The detail information is in Additional file 1: Table S3. Compared with the density distribution of all SNPs in the genome, SNPs were evenly distributed in all four chromosomes, and mainly distributed in chromosome 1 and 2 (Fig. 4b). Besides, there are 15.22% potential alterations in gene function in Fg isolates due to SNPs (58,674), InDels (2432), and SVs (4450) by nonsynonymous SNPs, early stop codon, loss of stop codon, frameshift mutation, and SV in exons (Additional file 1: Table S3).

Among all variants, Fa isolates had a lower number of variants compared with Fg isolates. There was no significant difference in InDels length between Fa and Fg isolates (Fig. 4c, d). The chromosomal inversions (27.94%) were frequent discovered in the Fg isolates, and chromosomal insertions of large fragments frequently occurred in Fa (11.97%) (Fig. 4e, f). Notably, Fg isolates may have more variants that potentially affected genetic differentiation compared with Fa isolates.

Discussion

FHB is caused by several Fusarium species. The determination of the distribution of Fusarium species could allow effective monitoring of the occurrence of FHB (Dweba et al. 2017; Xu et al. 2021). In this study, seven species of Fusarium involved in FHB were isolated from diseased wheat spikelet (Additional file 1: Table S1), including Fa, Fg, F. meridionale, F. avenaceum, F. lateritium, F. flocciferum, and F. proliferatum. Fa and Fg were the dominant species, which is consistent with the previous studies in Asia (Huang and Ye 2005; Van der Lee et al. 2015; Xu et al. 2021). F. meridionale, F. flocciferum, and F. proliferatum were new isolates. Similarly, F. meridionale and F. proliferatum have been reported to be the major pathogens of maize and soybean in Sichuan province (Chang et al. 2018; Liu et al. 2020; Wang et al. 2021). All Fusarium isolates showed less pathogenicity towards wheat, except for Fa and Fg (Fig. 2a, b). These strains were isolated from spikelet of wheat which were intercropped with maize or rotated in rice fields in the year of 2021, where the soil and diseased crop residues could serve as the initial infection source of Fusarium (Parry et al. 1995). Because the straw returning is encouraged in China when applying intercropping or rotation (Yan et al. 2020), these isolates might come from the straw residues of maize, soybean, or rice that had been infected by Fusarium.

Because of the climatic changes, the distribution of Fusarium species has been altered (Zhang et al. 2012). In this study, the major pathogenic strains are the Fa isolates in 2021, and there is a change for Fg in 2022 (Fig. 1c). The drier conditions during the winter and warmer conditions during the infection and grain-colonization period occurred in 2022 compared to the year of 2021. A previous study showed that Fg isolates are associated with drier and cooler conditions during the winter, as well as warmer conditions during the infection period (Xu et al. 2017). Thus, it suggests that climatic conditions are related to the distribution of Fa and Fg.

The demethylation inhibitor type of fungicides, such as TEC and prochloraz, are highly effective against Fusarium (Pasquali et al. 2020). Yin et al. (2009) showed that the 50% mycelial growth inhibition (EC₅₀) values of TEC-resistant isolates ranged from 0.034 to 6.235 µg/mL in 159 isolates of Fa and Fg. Talas et al. (2015b) also showed that the EC₅₀ values of TEC-resistant isolates ranged from 5.4 to 62.2 µg/mL in 231 isolates. These values exhibited a normal distribution with a mean value of 22.2 µg/mL. So, different isolates have different EC₅₀ values for TEC (Yin et al. 2009; Talas et al. 2015b). In this study, we observed that TEC at a concentration of 36 µg/mL was unable to completely inhibit the mycelial growth of Fa and Fg isolates (Fig. 3a). This concentration was higher than the previously reported concentration (Yin et al. 2009; Qian et al. 2018; Chen et al. 2021). There is no doubt that TEC-resistant isolates are emerging in Sichuan province, which increases the risk of fungicide application. Fa and Fg share similar morphological and molecular characteristics, toxicology, and genome sequences (Lee et al. 2014; Walkowiak et al. 2016; Yang et al. 2020). Lee et al. (2014) have revealed more than 80% nucleotide similarity between Fa and Fg in trichothecene biosynthetic genes. Walkowiak et al. (2016) showed that Fa is more closely related to Fg, with 93.1% sequence similarity of the genome. In this study, our results also showed that Fa and Fg isolates have similar morphology and pathogenesis (Figs. 1, 2). However, Fg isolates had more variants compared with Fa isolates (Additional file 1: Table S2 and Table S3), and the distribution of SNPs and types of SVs were significantly different (Fig. 3). Thus, Fa may have a different direction of evolution compared with Fg, and can adapt to different conditions.

The isolates that used to determine the sequence diversity among Fa and Fg were isolated from different places and environments. A large number of variants may be linked to the polymorphism of biological functions, and may aid Fusarium in adapting to different stresses, including fungicides. The following two examples confirmed this assumption. Previous reports have shown that genetic variants in codon 198 aa and 200 aa in the open reading frame of β2-tubulin gene were related to resistance to MBC (Chen et al. 2015; Duan et al. 2015). The same variants were found in Fa Lz189, Fa Lz263, Fa Lz503, and Fg Lz114, and were confirmed by fungicide treatments on PDA plates (Fig. 3a). Over-expression of CYP51-A and specific point mutations in CYP51-B contributed to the TEC resistance (Ma et al. 2006; Yin et al. 2009; Qian et al. 2018). However, the CYP51-A and CYP51-B did not have sequence changes compared with the reference gene (Fig. 3c). This finding suggests that genes other than the CYP51 family may be the most important contributors to TEC resistance (Talas et al. 2015b). In the current study, CYP51-A exhibited a significantly higher expression level than Fg PH-1 (Fig. 3d). Compared with genome sequencing results for Fg PH-1, exons of the CYP51-A gene showed six non-synonymous SNPs for Fa, and three non-synonymous SNPs for Fg. Twenty SNPs of Fa and one SNP variant occurred upstream of the CYP51-A gene. The relationship between these SNP variants and over-expression of CYP51-A needs to be further verified in strain Fg Lz179.

Conclusions

In this study, Fa and Fg isolates have developed resistance to MBC and TEC in Sichuan province (Fig. 3a and Additional file 1: Table S1). Because of the extensive use of MBC and TEC, the frequency of emerging fungicide-resistant isolates will gradually increase (Chen et al. 2015, 2021). The increased resistant strains will undoubtedly erode the control effect of FHB. Among the genetic variants we discovered, 15.22% of them could potentially affect gene function among Fg isolates, which may change the biological functions (Additional file 1: Table S3). Thus, genetic diversity can aid Fusarium in adapting to different environmental stresses and counteracting FHB management approaches, and increases the risk of FHB outbreaks. The fungicide resistance assays confirmed that the landscape of genomic assortment linked to the biological functions (Laurent et al. 2017). Importantly, these 521 isolates can provide an isolate library for quick search of the resistance genes in Fusarium species. The genetic diversity data can also be used to develop special molecular markers to assist in FHB risk assessment.

Methods

Fungal isolates

Wheat spikelet with visible FHB signs and symptoms was selected from Sichuan province in the year of 2021 (FHB outbreak) and 2022. All wheat spikelet with visible FHB signs were collected from forty-three fields that were evenly distributed in wheat-producing areas of Sichuan province (Fig. 1a). Five or six wheat spikelet with visible FHB signs were collected from each field. The collected wheat spikelet with visible FHB signs were washed with tap water, and were cut into small pieces (approximately 1 mm³), and surface-sterilized with 75% ethanol (v/v) for 30 s, 1% NaClO (w/v) for 15 s, then rinsed three times with sterile distilled water. These pieces were placed on Petri dishes containing potato dextrose agar medium (PDA, Aoboxing Biotechnology, Beijing, China), then incubated at 25 ℃ for 7 days in the dark. Pure cultures were obtained by a single macroconidia isolation method described by Chang et al. (2018). Two or three single isolates were randomly selected from each wheat spikelet with visible FHB signs for further research. All isolates were stored at –80℃ with 20% glycerol at Sichuan Agriculture University.

Molecular markers analysis

For molecular identification, total genomic DNA was extracted from fungal mycelia that were collected at the 7th day from the PDA plates by the cetyl trimethyl ammonium bromide (CTAB) method (Lodhi et al. 1994). Partial gene sequences of internal transcribed spacer (ITS), β-tublin, translation elongation factor 1-alpha (EF-1α), and RNA polymerase beta large subunit II (RPB2) were amplified as molecular markers by PCR. Primers, PCR conditions, and product size are listed in Additional file 1: Table S4. PCR products were sequenced in BGI (https://en.genomics.cn/), and classified by blaste in the FUSARIUM-ID (http://www.fusariumdb.org) and Fusarium MLST (http://www.cbs.knaw.nl/Fusarium/) database (Geiser et al. 2004).

Morphological analysis

Morphological characteristics of fungal species were identified from isolates and macroconidia based on previous studies (Leslie and Summerell 2006). The color and morphology of isolates were observed at the 7th day from the PDA plates. Image Java (National Institutes of Health, Bethesda, MD, USA) was used to calculate the growth rate of the hyphal area on PDA plates. Three biological replicates of each isolate were used, and F. graminearum PH-1 (Fg PH-1) was as used as the control (CK). Macroconidia were produced in carboxymethyl cellulose liquid medium at 28 °C, with shaking (180 rpm) for 7 days, and collected at the 7th day (Capellini and Peterson 1965). The shape and septum of the macroconidia were recorded using at least 1000 conidia per isolate under a compound microscope (Nikon-80i, Japan).

Virulence assay

To determine if these Fusarium strains caused disease, 41 Fa isolates, 23 Fg isolates, 3 F. meridionale isolates, 4 F. avenaceum isolates, 3 F. tricinctum isolates, 2 F. lateritium isolates, 3 F. flocciferum isolates, and 2 F. proliferatum isolates were used for inoculation on wheat. The common wheat cultivar ‘shumai482’, which is susceptible to Fusarium infection, were grown in a greenhouse under a 16 h/8 h (day/night) cycle at 23 °C/18 °C. Plants were watered as necessary and fertilized before planting with 15-15-15 (N-P-K) compound fertilizer. After inoculation, the plants were maintained in a growth chamber at 25 ℃ under 16 h/8 h (day/night) cycle, with 90% moisture for 7 days. Two florets of a single central wheat (Triticum aestivum cv. ‘shumai482’) spikelet were inoculated at each point using a micropipette at the mid-anthesis stage with 1 × 10³ macroconidia. The average value of infected spikelet was used to represent the infectivity of the Fusarium isolates. Ten wheat spikelets were used per isolate, and Fg PH-1 was used as the control.

Fungicide-resistance assay for Fa and Fg isolates

To determine Fusarium resistance to fungicides, all Fg and Fa isolates were cultured by transferring 5 mm diameter plugs from the edge of a 3-day-old active colony to PDA medium. The concentrations of MBC (1.25 μg/mL) and TEC (36 μg/mL) that were determined to completely inhibit Fg PH-1 growth in PDA medium were used to select as resistant isolates (Additional file 2: Figure S1). Chen et al. (2015) reported that a concentration of 50 μg/mL of MBC can be used to select MBC-resistant isolates. We then used 50 μg/mL of MBC to further confirm the isolates with MBC resistance. All isolates were incubated at 25 °C for 7 days in the dark. Three biological replicates were used for each strain. Resistant isolates were further studied.

Cloning and sequencing of β1-tubulin and β2-tubulin gene in Fa and Fg isolates

Total RNA was extracted from the mycelia of resistant isolates that were grown on PDA plates for 7 days at 25 °C, using the E.Z.N.A.^® Total RNA Kit I (Omega Bio-Tek, Norcross, GA, USA) according to the manufacturer’s instructions. The RNA was reverse transcribed using the PrimeScript™ RT Reagent Kit with genomic DNA Eraser (Takara, Dalian, China) following the manufacturer’s protocol. The cDNA sequences of β1-tubulin (Fa: CP088260.1_6189185 to 6,190,813; Fg: FGSG_09530), β2-tubulin (Fa: CP088260.1_643341 to 645053; Fg: FGSG_06611), CYP51-A (Fa: CP088258.1_6962848 to 69644; Fg: FGSG_04092), and CYP51-B (Fa: CP088257.1_3457782 to 3459530; Fg: FGSG_01000) genes were amplified by PCR. Primers, PCR conditions, and product size were shown in Additional file 1: Table S4. PCR products were sequenced in BGI (https://en.genomics.cn/). The Fg isolates sequencing results were aligned with the Fg PH-1 reference gene using DNAman 7.0 software (Lynnon Biosoft, USA). The Fa isolates sequencing results were aligned with the Fa KCTC 16664 reference gene using DNAman 7.0 software (Lynnon Biosoft, USA).

Measure the expression levels of CYP51-A and CYP51-B in Fg isolates

The primers CYP51AF/CYP51AR and CYP51BF/CYP51BR were used to amplify CYP51-A and CYP51-B, respectively. The relative expression levels of CYP51-A and CYP51-B were analyzed using the 2^−ΔΔCt method. Actin (FGSG_07335) and β-tubulin (FGSG_09530) were used as the references to normalize the expression data. The Fg strain PH-1 was used as a calibrator. The qPCRs were performed using a MyiQ Real-Time PCR Detection System (Bio-Rad, Hercules, CA, USA). All the primers mentioned above are listed in Additional file 1: Table S4.

Sequencing and data processing

Genomic DNAs were extracted from the fungal mycelia of 140 Fa isolates (Additional file 1: Table S5) that were cultured on PDA plates for 7 days at 25 °C, and the extracted DNA were equally premixed to build a DNA library at the Annoroad Genome company (http://genome.annoroad.com/). The same process was used to build a DNA library from 140 Fg isolates (Additional file 1: Table S5). Whole-genome sequencing was performed in BGI company (https://en.genomics.cn/). High quality sequences (clean reads) were filtered as follows: (1) remove the adaptor-polluted reads (reads containing > 5 adapter-polluted bases); (2) remove the low-quality reads (Phred quality value < 19); (3) remove reads with the number of N bases accounting for more than 5%. The obtained clean reads after filtering were used for further statistical analyses. The Fg sequences were aligned to the reference genome Fg PH-1 (Submitted NCBI sequence: GCA_000240135.3), and Fa sequences were aligned to the reference genome Fa KCTC 16664 (GCA_025258505.1) using the BWA software (Li and Durbin 2009) and the BWA-MEM mode (Li 2013). Based on the alignment to the reference genome sequence, the software GATK (McKenna et al. 2010) was used to call SNPs and Indels present in the whole genome. The filtering settings were as follows: SNP: QD < 2.0, Read Pos Rank Sum < –8.0, FS > 60.0, QUAL < 30.0, DP < 4.0, MQ < 40.0, Mapping quality rank sum < –12.5, and INDEL: QD < 2.0, Read pos rank sum < –20.0, FS > 200.0, QUAL < 30.0, DP < 4.0. Finally, SNP and InDel data sets of high reliability were obtained. All potential chromosome SVs were detected by chromosomal structural variation analysis using the DELLY software (Rausch et al. 2012).

Statistical analyses

Student’s t-test (implemented in the DPS (Data Procession System) version 12.01 software (Zhejiang University, Hangzhou, China) was used to examine the significance of differences among average values of isolates grown rate on PDA plates, percent of infected spikelets and the relative expression levels of CYP51-A and CYP51-B genes. Statistical differences were analyzed using the Least Significant Difference test at P ≤ 0.05. In order to minimize errors, two independent tests were performed, and the average value of the two tests was taken as the final result.