Introduction

Fusarium species are important pathogens that cause Fusarium head blight (FHB) in wheat and other cereals and can reduce yield by 30–70% (Bai and Shaner 1994). More than ten Fusarium species, including Microdochium nivale (Fr.) Samuels and Hallett (=F. nivale Cesati ex Sacc.), can cause head blight throughout the world (Liddell 2003). Several of these species can produce two major groups of mycotoxins, trichothecenes (Mirocha et al. 2003), and zearalenone (Hesseltine et al. 1978). These Fusarium species can cause serious problems for the agricultural industry and are potential threat to the food supply (McMullen et al. 1997; Clear et al. 2000).

In Japan, 12 Fusarium species cause head blight on wheat, barley and rye (Aoki 2003). Of these 12, F. graminearum Schw. (=Gibberella zeae Schw. and Petch), F. culmorum (W. G. Smith) Sacc., F. avenaceum (Corda: Fr.) Sacc., and M. nivale predominate and are most important (Aoki 2003; Aoki et al. 2005). Although F. graminearum is distributed all over Japan, F. culmorum, F. avenaceum, and M. nivale are found mainly in the northern part of Honshu and Hokkaido. Fusarium graminearum, F. culmorum, and F. avenaceum produce trichothecenes such as nivalenol (NIV) and deoxynivalenol (DON), which contaminate the grain (Aoki 2003; Souma 2003).

Recently, molecular technique has been developed to monitor and quantify the Fusarium species causing FHB and producing mycotoxins (Bluhm et al. 2002; Edel et al. 1997; Jurado et al. 2005; Konstantinova and Yli-Mattila 2004; Kulik et al. 2004; Láday et al. 2004; Mishra et al. 2003; Nicholson et al. 1998, 2003, 2004; Parry and Nicholson 1996; Suga et al. 2004; Waalwijk et al. 2004; Wilson et al. 2004; Yli-Mattila et al. 2002, 2004a, b; Yoder and Christianson 1998). These specific molecular markers were mostly designed from internal transcribed spacer (ITS) (Bluhm et al. 2002; Edel et al. 1997; Mishra et al. 2003; Wilson et al. 2004; Yli-Mattila et al. 2002, 2004a, b) or intergenic spacer (IGS) regions of rDNA (Jurado et al. 2005; Konstantinova and Yli-Mattila 2004; Yli-Mattila et al. 2002, 2004a, b), β-tubulin gene (Yli-Mattila et al. 2002, 2004b) or mitochondrial DNA (mtDNA) (Láday et al. 2004), and are highly specific for identifying Fusarium species. However, some of these species-specific markers have a differential sensitivity for isolates of same species of Fusarium from different areas (Yli-Mattila et al. 2004b). Thus, developing species-specific molecular makers to detect local isolates is important for reliable species identification.

Among the Fusarium species producing mycotoxins, the F. graminearum complex is the most widely distributed in the world and has been separated into seven lineage groups (O’Donnell et al. 2000a). Recently, two more species, F. vorosii Tóth et al. and F. gerlachii Aoki et al. were described within the F. graminearum complex (Starkey et al. 2007). However, only three species of the F. graminearum complex, F. graminearum Schwabe s. str. (lineage 7 of F. graminearum), F. asiaticum O’Donnell et al. (lineage 6 of F. graminearum), and F. vorosii have been found in Japan (Aoki 2003; O’Donnell et al. 2004; Starkey et al. 2007). The F. graminearum s. str. and F. asiaticum are regarded as genetically distinct according to PCR-based assay (Aoki et al. 2005; Suga et al. 2004). Thus, genetic analysis is very useful for designing new PCR (polymerase chain reaction) primers to rapidly and successfully identify different Fusarium species causing FHB.

The ITS and 5.8S regions of rDNA and β-tubulin gene are useful for identifying Fusarium species and studying their phylogeny (O’Donnell and Cigelnik 1997; O’Donnell et al. 2000b; Yli-Mattila et al. 2002). Recently, the cytochrome b (cytb) gene was also used to study the phylogenetic relationship in basidiomycetous yeasts (Biswas et al. 2001) and fungi (Sierotzki et al. 2003). The aim of the present study was to compare the diversity of nucleotide sequences in ITS and 5.8S rDNA regions, β-tubulin gene and cytb gene of F. graminearum-complex, F. culmorum, F. avenaceum, and M. nivale collected in Japan. The divergence of nucleotide sequences was used to design species-specific primers to identify these fungi. Thus, developing a one-step, allele-specific PCR is important to rapidly identify F. graminearum-complex and F. culmorum, in particular, which are most widely distributing pathogens causing FHB in Japan.

Thiophanate–methyl, a benzimidazole fungicide, has been used to control wheat FHB for a long period in Japan. Very recently, however, Fusarium isolates resistant to this fungicide have been detected occasionally in Aomori and Oita prefectures (Iwama et al. 2007; Yoshimatsu et al. 2006). Benzimidazole resistance was reported much earlier for M. nivale (Tsuboki 1984). Therefore, we included molecular characterization of benzimidazole resistance in these fungi in this study. In addition, partial cytb gene fragments, including codons 129 and 143 known to be responsible for resistance to QoI fungicides in fungi (Gisi et al. 2002), from the F. graminearum-complex and F. culmorum were amplified.

Materials and methods

Fungal isolates and DNA extraction

Fungal isolates used in this study are listed in Table 1. All Fusarium and Microdochium isolates were maintained on potato dextrose agar (PDA). The DNA extraction protocol followed that of Chung and Tsukiboshi (2005).

Table 1 Fungal isolates used in this study, origin, and DDBJ/EMBL/GenBank accession numbers
Full size table

Polymerase chain reaction (PCR) amplification and sequencing of ITS and 5.8S rDNA regions, β-tubulin gene and cytb gene of Fusarium isolates

For ITS and 5.8S rDNA region amplification, primers of ITS1 and ITS4 (White et al. 1990) were used and PCR conditions were those of Chung and Tsukiboshi (2005). Partial β-tubulin gene fragment was amplified by primers FU-tubulin3 (5′-CGAGCCCGGTACCATGGACG-3′ and FU-tubulin2 (5′-GGTCGCCGTAAGAGGGGTTGG-3′), designed from sequences registered in GenBank, and PCR conditions were as follows: initial denaturation at 94°C for 3 min, 35 cycles of 94°C for 1 min, 60°C for 30 s, 72°C for 2 min; and final extension at 72°C for 5 min. Similarly, partial cytb gene fragment was amplified by primers RSCBF1 and RSCBR2 (Ishii et al. 2001), and PCR conditions were as follows: initial denaturation at 94°C for 2.5 min, 40 cycles of 94°C for 1 min, 50°C for 30 s, 72°C for 1.5 min; and final extension at 72°C for 8.5 min. Amplifications were done using 20 μl PCR reaction mixtures, containing 0.2 μM of each primer, 1 unit of Ex Taq DNA polymerase (Takara Bio, Otsu, Japan), and a dNTP mixture (containing 250 μM of each of dNTP) and Ex Taq reaction buffer (containing 2 mM MgCl2). PCR products were first purified with a Wizard PCR Prep DNA Purification System (Promega, Madison, WI, USA), and then the products were reacted with BigDye Terminator v3.1 Cycle Sequencing Kit (Applied Biosystems, Foster City, CA, USA) using the same primers for PCR under the following conditions: 25 cycles of 96°C for 10 s, 50°C for 5 s, 60°C for 4 min. Cycle sequencing reaction products were purified by ethanol precipitation, then analyzed with an ABI PRISM 3100 automated sequencer (Applied Biosystems). The sequences obtained in this study were deposited in the DDBJ/EMBL/GenBank databases.

Sequence alignment, molecular phylogenetic analyses and species-specific primer design

DNA sequences were aligned using Clustal X v1.8 (Thompson et al. 1997). Further visual alignments were done with the program Sequence Alignment (Se-Al) Editor v.2.0 (Rambaut 2000). Phylogenetic analysis of the aligned sequences was done by distance parsimony methods. The distance matrix for the aligned sequences was calculated using Kimura’s two-parameter method (Kimura 1980) and analyzed with the neighbor-joining (NJ) method (Saitou and Nei 1987), including positions with gaps as a fifth character because they may be informative for inferring phylogenies. Reliability of the inferred trees was estimated by 1,000 bootstrap resampling using the same program. Parsimony analysis was done with the program PAUP v4.0b (Swofford 2000) using the heuristic search option with 100 random stepwise-addition sequences to search for the most parsimonious tree. Bootstrap (Felsenstein 1985) values were generated with 1,000 replicate heuristic searches to estimate support for clade stability of the consensus tree using the same program. Fusarium oxysporum (AY928420) and F. solani (AY677295) from GenBank were used as outgroups in phylogenetic analyses. For increased reliability of the analyses, 17 ITS sequences from GenBank, including F. graminearum, F. culmorum, F. avenaceum and M. nivale (Table 2), were used to align our Fusarium species obtained from Japan. In addition, the consensus variation of nucleotides in ITS and 5.8S rDNA, β-tubulin gene and the cytb gene among Fusarium species was aligned by Clustal X v1.8 (Thompson et al. 1997) and used to design species-specific primers. The PCR condition were tested by gradient PCR using the TaKaRa PCR Thermal Cycler Dice (Otsu, Japan), and PCR products were run in 1.5% agarose gels containing 0.5 μg/ml ethidium bromide in Tris–acetate–EDTA (TAE) buffers.

Table 2 Additional Fusarium and Microdochium species selected for ITS sequence alignment and analysis, origin, and DDBJ/EMBL/GenBank accession numbers
Full size table

Evaluation of thiophanate–methyl resistance and determination of β-tubulin gene nucleotide sequence

Sensitivities of Fusarium and Microdochium isolates to thiophanate–methyl (Nippon/Soda, Japan) were tested using mycelial growth assay. Each isolate was cultured on PDA plates at 25°C for 5 days. Mycelial disks, 4 mm in diameter, were cut from the margins of colonies and transferred onto PDA plates amended with 0, 1, 10, 100, and 1,000 mg/l active ingredient (a.i.) of thiophanate–methyl. Thiophanate–methyl was added to PDA before autoclaving. After incubation at 25°C in the dark for 3 days, the diameter of each colony was measured and the minimum inhibitory concentration (MIC) of the fungicide was determined. The MIC value for the sensitive phenotype was <10 mg/l, between 10 mg/l and 100 mg/l for the less sensitive phenotype, and between 100 mg/l and 1,000 mg/l for the high resistance phenotype, and >1,000 mg/l for the very high resistance phenotype. The β-tubulin gene fragment, including amino acid codons 198 and 200, was amplified from each isolate by FU-tubulin 3 and FU-tubulin 2 primers, and its nucleotide sequence was determined. The PCR conditions and nucleotide sequencing followed as those described.

Results

Molecular phylogenetic analyses of Fusarium isolates based on ITS and 5.8S rDNA

The amplified sequences for ITS and 5.8S rDNA regions of F. graminearum complex and F. culmorum were 545 bp long. Those of F. avenaceum and M. nivale were 561 and 556 bp, respectively. The three Fusarium species are genetically closely related, in particular, only two nucleotide subtitutions were observed between the F. graminearum complex and F. culmorum. The three Fusarium species were distinct from M. nivale based on the differences in the ITS and 5.8S rDNA sequences (Table 3). Consequently, the aligned ITS sequences consisted of 624 characters.

Table 3 Comparison of nucleotide and deduced amino acid divergence of three regions in three species with Fusarium graminearum complex
Full size table

Two methods were used to study the phylogenetic relationship of these four Fusarium species. In the MP analysis, 418 characters were constant, 66 parsimony uninformative, and 140 parsimony informative. The ITS and 5.8S rDNA phylogram has a consistency index (CI) of 0.804, a retention index (RI) of 0.911, retention consistency (RC) of 0.733 and a tree length of 327. The result indicated that the F. graminearum-complex and F. culmorum were distinct from F. avenaceum and M. nivale, and this result was supported by a high bootstrap value (Fig. 1). Similar to the MP analysis, the NJ analysis also showed that the F. graminearum complex was closely related to F. culmorum based on the sequence of ITS and 5.8S rDNA regions and supported by a high bootstrap value (100%) (Fig. 2).

Fig. 1
figure 1

The ITS and 5.8S rDNA sequence-based tree generated using most parsimony analysis. The numbers at branch node indicate the confidence values from bootstrap analysis using 1,000 replications. Outgroup: Colletotrichum gloeosporioides

Full size image
Fig. 2
figure 2

The ITS- and 5.8S rDNA- sequence-based tree generated using neighbor-joining analysis. The numbers at branch node indicate the confidence values from bootstrap analysis using 1,000 replications. Bootstrap values above 52% are presented. Outgroup: Colletotrichum gloeosporioides

Full size image

Divergence of β-tubulin and cytb genes

The partial β-tubulin gene fragments, 470 bp long, were amplified from these Fusarium species and M. nivale. Japanese isolates of the F. graminearum complex were highly homologous with F. culmorum isolates and only 0.6–1.1% nucleotide divergence was observed between the two species, and the homology of deduced amino acid sequences was 100%. On the other hand, the partial β-tubulin gene nucleotide sequences of F. avenaceum and M. nivale had higher divergence of 11.9–12.1 and 12.3–13.2%, respectively, than did the F. graminearum complex and F. culmorum (Table 3).

The partially amplified cytb gene fragments from the four species were 286 bp long, corresponding to 95 amino acids. F. graminearum complex isolates differed by 0.0–0.7% in their nucleotides from F. culmorum isolates, but the deduced amino acid sequences were the same as F. culmorum isolates. F. avenaceum isolates had 3.8–10.5% nucleotide and 1.1–4.2% amino acid divergence from F. graminearum-complex isolates, while isolates of M. nivale had 14.7–15.0% nucleotide and 7.4% amino acid divergence from those of F. graminearum complex (Table 3).

Relationship of genetic variations in β-tubulin gene with thiophanate–methyl sensitivity

In the present study, the isolates of the F. graminearum complex and F. culmorum sensitive to thiophanate–methyl had the sequence GAA (Glu) at codon 198, while F. avenaceum and M. nivale isolates had GAG (Glu) (Table 4). Similarly, less thiophanate–methyl sensitive Chinese isolates of F. graminearum (kindly supplied by Prof. Zhou Minggou, Nanjing Agric. Univ., China) also revealed GAA (Glu) at codon 198. Sequences at codon 200 were TTT (Phe) in F. graminearum complex isolates, whereas those of F. culmorum, F. avenaceum, and M. nivale isolates were TTC (Phe). Thus, the mutation leading to the amino acid substitution at codon 198 or 200 of β-tubulin was not observed in the F. graminearum complex, F. culmorum and F. avenaceum isolates. In contrast, isolates very highly resistant to thiophanate–methyl, MAFF101046 and Mn4-16, and one highly resistant isolate of M. nivale, Mn5-11, had mutation at codon 198 (Table 4). The sequences in very highly resistant and highly resistant isolates of M. nivale were GCG (Ala) and AAG (Lys), respectively, both had substitution at GAG (Glu) at codon 198.

Table 4 Mutation and deduced amino acid substitutions in the partial sequence of the β-tubulin gene in isolates of the Fusarium graminearum-complex, F. culmorum, F. avenaceum, and Microdochium nivale, and different phenotypic responses to thiophanate–methyl
Full size table

Design of specific PCR primers for distinguishing F. graminearum complex from F. culmorum

The ITS and 5.8S rDNA regions of the isolates of F. graminearum-complex and F. culmorum analyzed in this study had high sequence homology. However, two consensus nucleotide substitutions were found in the ITS1 and ITS2 regions between F. graminearum complex and F. culmorum isolates (Fig. 3). Consequently, we designed the following species-specific primers to differentiate them: forward primer: ITS1-Fg1 (5′-GGACGGCCCGCCGCAGGAACCCT-3′) and reverse primer: ITS2-Fg (5′-CGCGACGATTACCAGTAACGATG-3′) for F. graminearum-complex, forward primer: ITS1-Fc1 (5′-GGACGGCCCGCCGCAGGAACCTT-3′) and reverse primer: ITS2-Fc1 (5′-CGCGACGATTACCAGTAACGATA-3′) for F. culmorum. PCR conditions were as follows: initial denaturation at 95°C for 2 min, 30 cycles of 95°C for 30 s, 68°C for 30 s, 72°C for 2 min; and final extension at 72°C for 5 min. Amplifications were done using 20 μl PCR reaction mixtures, containing 0.2 μM of each primer, 1 unit of Ex Taq DNA polymerase, and dNTP mixture (containing 250 μM of each of dNTP) and Ex Taq reaction buffer (containing 2 mM MgCl2). About 350 bp were amplified, and the results from PCR indicated that the two primer pairs were successfully used for distinguishing the F. graminearum complex from F. culmorum (Fig. 4). Then the Fusarium isolates obtained from Aomori, Oita, and Shimane Prefectures in Japan were examined using these primers. The results indicated that two thiophanate–methyl-resistant isolates, 19-1 and 20-2, from Aomori were F. culmorum and that the other two isolates, 20-1 and 20-3, were part of the F. graminearum complex. Moreover, two isolates, 1 and 329, from Oita and isolates OB02, OB03, 0B29, OB72, and OB73 from Shimane were F. graminearum complex (Fig. 5). The PCR results were further confirmed by sequencing the ITS-5.8S rDNA regions of these isolates and subjected to a Blast search of GenBank. In addition, we also used these species-specific primer pairs to test other Fusarium species, including F. tricinctum, F. equiseti, F. sporotrichioides, and M. nivale, but the result indicated that these primer pairs did not amplify the ITS and 5.8S rDNA regions in these species (data not shown).

Fig. 3
figure 3

The nucleotide sequence variation at ITS1 and ITS2 regions in Fusarium graminearum-complex and F. culmorum. The arrows indicate the change of bases. Nucleotides with underline are used to design the species-specific primers for identifying the F. graminearum-complex and F. culmorum

Full size image
Fig. 4
figure 4

The polymerase chain reaction (PCR) primer pairs ITS1-Fg1 + ITS2-Fg (left) and ITS1-Fc1 + ITS2-Fc1 (right) for identifying Fusarium graminearum (line 1–3) and F. culmorum (line 4–7). M DNA ladder marker, Lane 1 NIV7, Lane 2 H3, Lane 3 MAFF101032, Lane 4 MAFF236454, Lane 5 MAFF236455, Lane 6 MAFF236456, and Lane 7 MAFF101144

Full size image
Fig. 5
figure 5

The polymerase chain reaction (PCR) primer pairs ITS1-Fg1 + ITS2-Fg (up) and ITS1-Fc1 + ITS2-Fc1 (down) for identifying Fusarium isolates from Aomori Prefecture (Lane 1–4), Oita Prefecture (Lane 5–9) and Shimane Prefecture (Lane 10–14). Lane 15 (ZF52) and 16 (ZF4) are F. graminearum isolates from China. M DNA ladder marker

Full size image

Discussion

The molecular sequence of ITS-5.8S rDNA and the partial sequence of β-tubulin and of cytochrome b (cytb) genes indicated that F. graminearum complex were closely related with F. culmorum genetically, but distinct from F. avenaceum and M. nivale isolates. Molecular phylogenetic analysis of ITS and 5.8S rDNA regions showed that F. graminearum complex and F. culmorum were placed in the same clade, and F. avenaceum was phylogenetically closer to the F. graminearum complex and F. culmorum than to M. nivale.

In the present study, the sequence variations of ITS and 5.8S rDNA regions between the F. graminearum complex and F. culmorum were low. Thus, F. graminearum complex and F. culmorum could not be separated in the phylogenetic tree based on the sequence of this region. Similarly, several researchers indicated that the ITS and 5.8S rDNA regions were not suitable to identify the F. graminearum complex and F. culmorum because of their low polymorphic nucleotide sequences (Bateman et al. 1996; Edel et al. 1997; Schilling et al. 1996; Yli-Mattila et al. 2004a). Although Schilling et al. (1996) mentioned that insertion/deletions of a few bases existed the between F. graminearum complex and F. culmorum, the sequence variations were not polymorphic enough to identify the two species. However, two fixed nucleotide substitutions were found in the ITS1 and ITS2 regions between F. graminearum complex and F. culmorum. We aligned the ITS sequences in Japanese isolates of F. graminearum complex and F. culmorum and compared with isolates of the F. graminearum complex and F. culmorum from different countries. The alignment indicated that the two fixed nucleotide substitutions also existed between the F. graminearum-complex and F. culmorum isolates obtained in different countries.

In the present study, the primers designed for the two fixed nucleotide differences were effective in specifically identifying Japanese isolates of the F. graminearum complex and F. culmorum. The results revealed that a single nucleotide polymorphism (SNP) is useful for the differential detection of genetically close isolates. However, the SNP assay has been generally used to diagnose fungal species not to quantitate them. Recently, new methods were developed to quantitatively analyze a population of Fusarium species (Bluhm et al. 2002; Nicholson et al. 1998, 2003; Waalwijk et al. 2004). Thus, our species-specific primers may be useful for quantifying species populations. Bushnell et al. (2003) indicated that Fusarium species causing FHB could invade different plant tissues; however, we did not try to isolate Fusarium species from various plant tissues. We will do such isolations in the future. As we did with the ITS and 5.8S rDNA regions, we aligned the nucleotide sequences of partial β-tubulin gene and only a few nucleotide difference existed between the F. graminearum complex and F. culmorum. O’Donnell et al. (2004) indicated that the fixed nucleotide characters of β-tubulin gene existed in the F. graminearum complex; however, the partial β-tubulin gene sequences that we amplified (400 bp) had the same nucleotides as F. graminearum s. str., and we had difficulty in further identifying our F. graminearum complex isolates in this study. While F. avenaceum and M. nivale were distinct from the F. graminearum complex and F. culmorum, molecular phylogenetic analysis of the β-tubulin gene showed that the F. graminearum complex and F. culmorum located in the same clade as found in the analysis of ITS regions (data not shown). The results revealed high consensus between the F. graminearum complex and F. culmorum in the ITS region and the β-tubulin gene.

According to O’Donnell et al. (2004) and Starkey et al. (2007), the Japanese F. graminearum complex isolates belong to F. graminearum s. str., F. asiaticum or F. vorosii; however, we did not check whether the F. graminearum-complex isolates belonged to F. graminearum s. str. or to F. asiaticum. It is important to establish a way to distinguish F. graminearum s. str. and F. asiaticum from F. vorosii and to monitor the population in the field in future. Similarly, Nicholson et al. (1996) indicated that M. nivale causing wheat head blight included two varieties, M. nivale var. nivale and M. nivale var. majus. These two varieties were recently elevated to species status based on variations in the EF-1 alpha gene sequences (Glynn et al. 2005). M. nivale var. nivale is mainly found in Hokkaido, Japan, but we did not check these sequences for our Japanese M. nivale isolates. Doing so will be the focus of future research.

In the present study, we also examined the relationship of thiophanate–methyl resistance with the sequence variations in the β-tubulin gene. However, no mutations were found at amino acid codons 198 and 200 in the thiophanate–methyl-resistant isolates of the F. graminearum complex we have tested so far. This phenomenon was also reported for benzimidazole-resistant isolates of F. sambucinum Fuckel (=Gibberella pulicaris (Fr.) Sacc.) (Kawchuk et al. 2002) and F. graminearum (Li et al. 2003). In contrast to Fusarium species, thiophanate–methyl-resistant isolates of M. nivale had a mutation at codon 198 in the β-tubulin gene. The result revealed that in highly resistant and very highly resistant isolates of M. nivale, Glu (GAG) was substituted with Lys (AAG) and Ala (GCG), respectively. However, there was no nucleotide change at codon 200. This result indicated that the mutation at codon 198 was responsible for high resistance and very high resistance to thiophanate–methyl in M. nivale isolates.

The cytb gene sequence has been used to study the phylogeny of fungi (Biswas et al. 2001) due to their high sequence diversity. However, the partial cytb gene sequence that we analyzed indicated that the F. graminearum complex was closely related to F. culmorum as did the ITS-5.8 S rDNA regions and the β-tubulin gene. In the present study, only a few isolates of the F. graminearum complex and F. culmorum were analyzed. It is, therefore, necessary to analyze more isolates in the future to check the stability of the species identification system developed in this study.