The soilborne fungal pathogen Verticillium dahliae causes wilt disease on agricultural crops of many kinds. Our previous work revealed that 27 Japanese and nine foreign isolates of V. dahliae uniformly have the MAT1-2-1 gene encoding the mating type protein, although all the isolates are asexual (Usami et al. 2009). However, we were unable to find the opposite mating type gene, MAT1-1-1. Generally, sexual reproduction of ascomycetes is controlled by mating type genes (Turgeon 1998; Turgeon and Yoder 2000). In a heterothallic system, sexual reproduction occurs between a strain that has the MAT1-1 gene and one that has the MAT1-2 gene. In contrast, a homothallic fungus has both MAT1-1 and MAT1-2 genes or a gene combining both, and produces the sexual stage alone.

We inferred two possibilities from our results (Usami et al. 2009). First, V. dahliae might be heterothallic, and its mating type might be biased toward MAT1-2. Second, the MAT1-1-1 sequence was not amplified by polymerase chain reaction (PCR) primers we used, although V. dahliae is homothallic and each isolate has both MAT1-1-1 and MAT1-2-1. To assess these hypotheses, we needed to identify the sequence of MAT1-1-1.

In the present study, we assayed 41 isolates of V. dahliae (nos. 1–41 in Table 1), including 36 isolates described in Usami et al. (2009), using PCR and found an isolate that apparently lacks MAT1-2-1 (Fig. 1). For the experiment, total DNA of each isolate was prepared as previously described (Usami et al. 2002, 2005). We prepared a PCR reaction mixture (20 μl) containing 20 ng fungal total DNA, 100 pmol each of primers, VdMAT1-2a (5′-CGACCGCTACTATATTGGCCC-3′; nt 3327–3347 of accession AB505214) and VdMAT1-2b (5′-CTGCGACAGCAGATTCTGGGTTGCAAAGGC-3′; nt 3932–3903 of accession AB505214) (Usami et al. 2009), and 10 μl of Premix Ex Taq Hot Start Version (Takara Bio, Otsu, Shiga, Japan). Annealing sites and directions of the primers are shown in Fig. 2. PCR was conducted using the following program: 94°C for 3 min; 30 cycles of 94°C for 30 s, 65°C for 30 s, and 72°C for 60 s; and then 72°C for 3 min. Results of electrophoresis on a 2% (w/v) agarose gel containing 1 × Tris–acetate-EDTA (TAE) are shown in Fig. 1a. Amplification was absent only in isolate NBRC 6119. Total DNA of each isolate was digested with HindIII and electrophoresed on a 0.7% (w/v) agarose gel. The blot was probed with MAT1-2-1 fragment (nt 5223–5822 of DDBJ/EMBL/GenBank accession AB505214). The probe did not hybridize to any bands in NBRC 6119, confirming that NBRC 6119 does not contain MAT1-2-1 (Fig. 1b). Experimental procedures used for the Southern hybridization are described in Usami et al. (2009). The hybridization temperature was 60°C.

Table 1 Isolates of Verticillium dahliae used for this study
Full size table
Fig. 1
figure 1

PCR amplification using primers VdMAT1-2a and VdMAT1-2b designed from MAT1-2-1 sequence of Verticillium dahliae (a). Genomic Southern hybridization probed with a DNA sequence containing MAT1-2-1 of V. dahliae (b). Positions of respective primers and probes are shown in Fig. 2. Lanes 1–5, V. dahliae TV103, U48, ATCC 201177, NBRC 6126, and NBRC 6119

Full size image
Fig. 2
figure 2

Restriction maps of MAT1-1 idiomorph of Verticillium dahliae NBRC 6119 (top, DDBJ/EMBL/GenBank accession AB505215) and MAT1-2 idiomorph of V. dahliae TV103 (bottom, accession AB505214). Each idiomorph is marked by a gray box. White arrows indicate putative protein-coding regions with introns (black boxes). Homologous sequences of MAT1-1-3 and the DNA lyase gene are shown as white boxes. Annealing sites and directions of PCR primers are represented as arrowheads. Sequences used as probes for genomic Southern hybridization (shown in Figs. 1, 4) and screening of a genomic library of NBRC 6119 are also shown. Restriction sites: P, PstI; X, XbaI; B, BglII; E, EcoRI; V, EcoRV; H, HindIII; S, SacI

Full size image

We expected that an isolate lacking MAT1-2-1 would contain the MAT1-1-1 gene. Therefore, we constructed a genomic DNA library of NBRC 6119 using cloning vector Lambda BlueSTAR (Novagen, San Diego, CA, USA) and screened by plaque hybridization. We used a DNA fragment including part of a DNA lyase gene homologue (nt 6780–7881 of accession AB505214) (Fig. 2) as a screening probe because this gene was often found outside the MAT idiomorph (Kanematsu et al. 2007; Yokoyama et al. 2005, 2006; Yun et al. 2000) and probably exists adjacent to the MAT1-1 idiomorph of V. dahliae. Detailed procedures for library construction and screening are described by Usami et al. (2007).

Sequence analyses revealed that a DNA fragment obtained from the genomic library of NBRC 6119 includes a region (nt 1304–1759 of accession AB505215) that is homologous to the MAT1-1-1 gene of Gibberella avenacea (accession AJ535625; identity of nucleotide sequences is 56.5%). We obtained a cDNA sequence corresponding to this region using reverse transcription (RT)-PCR with primer pair VdMAT1-1-1F2 (5′-CCACTCGAAACCCCACCGTC-3′; nt 1902–1883 of accession AB505215) and VdMAT1-1-1R3 (5′-GGCCTCCATGTTGTAAGCGT-3′; nt 906–925 of accession AB505215). Annealing sites and directions of these primers are shown in Fig. 2. For the experiment, we extracted total RNA from mycelia that were statically cultured in potato broth containing 2% (w/v) sucrose for a week at 25°C in dark using RNAiso Plus (Takara Bio). After treating with deoxyribonuclease (Nippon Gene, Tokyo, Japan), we conducted reverse transcription using a First-strand cDNA Synthesis Kit (Takara Bio). We prepared the PCR mixture containing 10 ng of cDNA, 100 pmol of each primer, and 10 μl of Premix Ex Taq Hot Start Version (Takara Bio), followed by PCR using the program of 94°C for 3 min; 30 cycles of 94°C for 30 s, 60°C for 30 s, and 72°C for 60 s; and then 72°C for 5 min. The 1263-bp putative open reading frame (ORF) divided by a 56-bp intron was identified based on the nucleotide sequences of cDNA and genomic DNA. The sequence of this ORF is available in the DDBJ/EMBL/GenBank databases (accession AB469828). A putative 421-amino acid protein encoded by this ORF has a significant similarity to the MAT1-1-1 protein of fungus, and the most significant match is the MAT1-1-1 protein of Colletotrichum musae (UniProt ID: Q874U2; 32.9% identity and 62.6% similarity in 286-aa overlap). The DNA-binding motif in the MAT1-1-1 protein called the alpha box is well preserved among fungal species (Turgeon and Yoder 2000). The amino-acid sequence of alpha box in the putative MAT1-1-1 protein of V. dahliae has significant similarities to those of other fungi (Fig. 3).

Fig. 3
figure 3

The putative amino-acid sequence of alpha box (DNA-binding motif in MAT1-1-1 protein) of Verticillium dahliae NBRC 6119 is compared with that from Colletotrichum musae (UniProt ID: Q874U2), Gibberella avenacea (Q874U8), Paecilomyces tenuipes (Q874E6), Fusarium oxysporum f. sp. cubense (Q96UN5), and Gibberella fujikuroi (Q9C461). Amino-acid symbols that are identical to the sequence of V. dahliae NBRC 6119 are shaded in gray

Full size image

Mating-type idiomorphs of V. dahliae were identified by comparing nucleotide sequences of both MAT1-1 and MAT1-2 (Fig. 2). Nucleotide sequences of MAT1-1 (4124 bp) and MAT1-2 (3965 bp) idiomorphs are available in the DDBJ/EMBL/GenBank databases (accession AB505215 and AB505214, respectively). The sequence identities of 5′ (MAT1-1, nt 1–258 of accession AB505215; MAT1-2, nt 1707–1964 of accession AB505214) and 3′ (MAT1-1, nt 4383–5849 of accession AB505215; MAT1-2, nt 5930–7396 of accession AB505214) flanking regions between idiomorphs are 97.3 and 98.6%, respectively. In contrast, the sequences of MAT1-1 and MAT1-2 idiomorphs differ greatly from each other. The four genes, MAT1-1-1 (encoding a DNA-binding protein with an alpha box motif), MAT1-1-2 (encoding a putative transcriptional regulator), MAT1-1-3 (encoding a protein with a high mobility group [HMG] box motif), and MAT1-1-4 (encoding a metallothionein-like protein) were identified in the MAT1-1 idiomorph of ascomycetes (Turgeon and Yoder 2000). Alternaria alternata (Arie et al. 2000) and Cochliobolus heterostrophus (Turgeon and Yoder 2000) contain only MAT1-1-1 in their MAT1-1 idiomorphs. On the other hand, many fungal species contain three genes, MAT1-1-1, MAT1-1-2, and MAT1-1-3 (Arie et al. 2000; Kanamori et al. 2007; Kanematsu et al. 2007; Pöggeler et al. 2008; Turgeon and Yoder 2000; Yokoyama et al. 2005, 2006; Yun et al. 2000). Pyrenopeziza brassicae contains MAT1-1-4 instead of MAT1-1-2 (Singh and Ashby 1998). Although MAT1-1 idiomorph of V. dahliae included a region (nt 3758–4045 of accession AB505215) that is similar to MAT1-1-3 of Metarhizium anisopliae (accession AB258381) (56.2% identity of nucleotide sequence), we were unable to identify the introns and an ORF because the cDNA was not amplified by RT–PCR. No ORF or homologue corresponding to MAT1-1-2 or MAT1-1-4 was detected in the MAT1-1 idiomorph of V. dahliae NBRC 6119.

We designed PCR primers VdMAT1-1a (5′-GTCCCTGGAGGTAGGGAGTG-3′; nt 2457–2476 of accession AB505215) and VdMAT1-1b (5′-TGCTTCCTCCGTCAAGACGC-3′; nt 2866–2847 of accession AB505215) in the MAT1-1 idiomorph of V. dahliae NBRC 6119. Annealing sites and directions of these primers are shown in Fig. 2. The mating type idiomorph of each isolate can be examined using multiplex PCR with two primer pairs, VdMAT1-1a/VdMAT1-1b (for an ~400-bp fragment from MAT1-1 idiomorph) and VdMAT1-2a/VdMAT1-2b (for an ~600-bp fragment from MAT1-2 idiomorph), in the same reaction. We prepared the reaction mixture (20 μl) containing 20 ng fungal total DNA, 100 pmol of each primer, and 10 μl of Premix Ex Taq Hot Start Version (Takara Bio) and used the amplification program, which is described for PCR assays shown in Fig. 1. The PCR products were electrophoresed on a 2% (w/v) agarose gel containing 1 × TAE. We tested 61 isolates of V. dahliae, including 20 additional isolates (nos. 42–61 in Table 1). The PCR amplified ~600-bp fragments (MAT1-2 idiomorph) from 56 isolates and ~400-bp fragments (MAT1-1 idiomorph) from NBRC 6119 and four other isolates (Ud1-4-1, CA26, TO-0, and ATCC 26509) (Fig. 4a). None of the isolates had both fragments, suggesting that, of 61 isolates tested, five are MAT1-1 and 56 are MAT1-2. To confirm these results, we probed the gel blot of HindIII-digested DNA of each isolate with the MAT1-1-1 (nt 1304–1757 of accession AB505215) and MAT1-2-1 (nt 5223–5822 of accession AB505214) sequences, indicated in Fig. 2. The MAT1-1-1 probe hybridized to ~10-kbp bands in five isolates (NBRC 6119, Ud1-4-1, CA26, TO-0, and ATCC 26509) (lanes 41, 46, 54, 57, and 60 in Fig. 4b), and the MAT1-2-1 probe hybridized to ~3-kbp bands in the remaining 56 isolates (Fig. 4c). No isolate had both bands (Fig. 4b, c), and thus all isolates we tested appeared to have either MAT1-1 of MAT1-2 idiomorph. These results suggest that V. dahliae is potentially heterothallic.

Fig. 4
figure 4

Multiplex PCR assays using primer pairs VdMAT1-1a/VdMAT1-1b (for MAT1-1 idiomorph) and VdMAT1-2a/VdMAT1-2b (for MAT1-2-1 idiomorph) (a) Genomic Southern hybridization probed with sequences of MAT1-1-1 (b) and MAT1-2-1 (c). Fungal genomic DNA was digested with HindIII, and hybridization was performed at 60°C. The restriction site and the position of each primer and probe are shown in Fig. 2. Each lane no. corresponds to the isolate no. in Table 1

Full size image

Besides NBRC 6119, we found three Japanese (Ud1-4-1, CA26 and TO-0) and one foreign (ATCC 26509) isolate containing a MAT1-1 idiomorph using multiplex PCR and genomic Southern hybridization analyses (Fig. 4). Inclusively, we identified five isolates containing MAT1-1 idiomorphs among 61 isolates of V. dahliae in this study. In general, the mating type-ratio of progeny produced by sexual reproduction ought to remain at 1:1. However, the mating type-ratio of V. dahliae appears to be considerably biased. This finding suggests that, similar to the cases of asexual Fusarium oxysporum f. sp. lycopersici (Arie et al. 2000) and Cercospora spp. (Groenewald et al. 2006), sexual reproduction in V. dahliae is improbable or rare. Mating types of F. oxysporum f. sp. lycopersici (Arie et al. 2000) and C. apii (Groenewald et al. 2006) are biased toward MAT1-1. In contrast, those of C. apiicola (Groenewald et al. 2006) and V. dahliae (this study) are biased toward MAT1-2. It remains unclear why the isolates with MAT1-2 spread predominantly throughout the population of V. dahliae. Although some MAT1-1 isolates of this fungus possess sufficient pathogenicity on host plants, they might be disadvantaged in survival.

We did not observe sexual reproduction in any co-cultures between NBRC 6119 and a MAT1-2 isolate (isolates TV103, U22, TO20, TK15, TK23, U48, Cns, Chr208, or Ibh) (data not shown). Imperfect functioning of mating type genes or other sexual factors, e.g., pheromone receptors, and/or difference in karyotype might inhibit the occurrence of sexual reproduction. In the future, the probability of sexual reproduction in V. dahliae should be investigated by co-culturing Japanese isolates with MAT1-1 and MAT1-2 in addition to comparing the karyotypes of these two isolates.