Compatible/incompatible interactions between the tomato wilt fungus Fusarium oxysporum f. sp. lycopersici (FOL) and tomato Solanum lycopersicum are controlled by three avirulence genes (AVR1–3) in FOL and the corresponding resistance genes (I–I3) in tomato. The three known races (1, 2 and 3) of FOL carry AVR genes in different combinations. The current model to explain the proposed order of mutations in AVR genes is: i) FOL race 2 emerged from race 1 by losing the AVR1 and thus avoiding host resistance mediated by I (the resistance gene corresponding to AVR1), and ii) race 3 emerged when race 2 sustained a point mutation in AVR2, allowing it to evade I2-mediated resistance of the host. Here, an alternative mechanism of mutation of AVR genes was determined by analyses of a race 3 isolate, KoChi-1, that we recovered from a Japanese tomato field in 2008. Although KoChi-1 is race 3, it has an AVR1 gene that is truncated by the transposon Hormin, which belongs to the hAT family. This provides evidence that mobile genetic elements may be one of the driving forces underlying race evolution. KoChi-1 transformants carrying a wild type AVR1 gene from race 1 lost pathogenicity to cultivars carrying I, showing that the truncated KoChi-1 avr1 is not functional. These results imply that KoChi-1 is a new race 3 biotype and propose an additional path for emergence of FOL races: Race 2 emerged from race 1 by transposon-insertion into AVR1, not by deletion of the AVR1 locus; then a point mutation in race 2 AVR2 resulted in emergence of race 3.
Citation: Inami K, Yoshioka-Akiyama C, Morita Y, Yamasaki M, Teraoka T, et al. (2012) A Genetic Mechanism for Emergence of Races in Fusarium oxysporum f. sp. lycopersici: Inactivation of Avirulence Gene AVR1 by Transposon Insertion. PLoS ONE 7(8): e44101. doi:10.1371/journal.pone.0044101
Editor: Zhengguang Zhang, Nanjing Agricultural University, China
Received: April 3, 2012; Accepted: July 30, 2012; Published: August 27, 2012
Copyright: © Inami et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This research was financially supported by a Grant-in-Aid for Scientific Research from the Japan Society for the Promotion of Science to TA (18380030 and 22405018) and to KI (22–8126)(http://www.jsps.go.jp/j-grantsinaid/index.html). This work was supported in part by a grant from the Ministry of Agricuture, Forestry and Fisheries Research Council of Japan (H23–25 B5 102) (http://www.s.affrc.go.jp/docs/project/2011/pdf/2011_kekka_07.pdf). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
In the arms race between plants and pathogens, the pathogens can win by circumventing the immune system of host plants, e.g., by avoiding or suppressing defense mechanisms. In general, plants have two types of resistance: polygenic (horizontal), controlled by multiple genes, each with a small phenotypic effect, and monogenic (vertical), controlled by a single resistance (R) gene, which often confers a high level of resistance . Monogenic resistance generates immune responses (e.g. hypersensitive reaction, HR) to particular pathogen(s) , and has been effective and practical to use in modern plant breeding. This resistance is described by the ‘gene-for-gene theory’ , which explains the relationship between pathogen races and host plant cultivars by the interaction between an avirulence (AVR) gene in the race and an R gene in the cultivar. When a race possessing an AVR gene attacks a cultivar carrying the corresponding R gene, resistance is induced in the plant and the disease does not occur. A loss of function in an AVR gene allows the pathogen to avoid induction of resistance in the cultivar, the pathogen gains pathogenicity to that cultivar, and a new pathogenic race has emerged.
The ascomycete Fusarium oxysporum Schlecht. emend. Snyd. et Hans. causes vascular diseases of many plant species, yet each strain of this fungus has strictly defined host specificity . Strains that cause wilt disease only on tomato (Solanum lycopersicum L.) are classified as f. sp. lycopersici Snyd. et Hans. (FOL). Three races of FOL have been reported; their relationship with tomato cultivars is explained by the ‘gene-for-gene theory’ . Original descriptions of FOL races 1, 2 and 3 appeared before 1895 in England, in 1939 in the USA and in 1978 in Australia, respectively . In Japan, races 1, 2 and 3 were reported in Fukuoka in 1905, in 1966 and in 1997, respectively .
To date, the R genes I, I2 and I3 are known in tomato cultivars ; these R genes correspond to the avirulence genes AVR1, AVR2 and AVR3 in FOL, respectively (Table 1). Historically, race 1-resistant cultivars (I i2 i3), races 1 and 2-resistant cultivars (I I2 i3), and races 1, 2 and 3-resistant cultivars (I I2 I3) have been bred sequentially, each genotype corresponds to the emergence of a new race.
Table 1. Relationship between FOL races and tomato cultivars.doi:10.1371/journal.pone.0044101.t001
The FOL AVR genes (AVR1, AVR2 and AVR3) are unique to FOL , ,  and are carried in different combinations in different FOL races (Table 1). AVR1 ( = SIX4) is unique to race 1 , whereas AVR2 ( = SIX3) is found in races 1 and 2. Three nucleotide substitutions (G121A, G134A and G137C) in AVR2, which cause loss of avirulence function (avr2) have been found in race 3 . AVR3 ( = SIX1), which exists in all races , is known to have two silent mutations (lysine or glutamine at amino acid 164) that do not influence avirulence to I3 cultivars . FOL races can be determined by AVR gene combinations , .
Based on the knowledge of AVR genes, it was suggested that FOL races emerged as follows : race 1 (AVR1 AVR2 AVR3) lost the AVR1 locus and became race 2 (– AVR2 AVR3), which escapes recognition by the I gene; a nucleotide substitution in race 2 AVR2 resulted in race 3 (– avr2 AVR3), which evades recognition by both I and I2. Those mutations of AVR genes are consistent in many FOL isolates .
Mating type (MAT), vegetative compatibility group (VCG), and phylogeny have been used to characterize genetic relationships among FOL isolates , , . MAT and VCG correlate with the phylogenetic relationship . All FOL isolates belong to one of three clades (A1–A3) in the F. oxysporum phylogeny based on the intergenic region of ribosomal DNA (rDNA-IGS), suggesting a polyphyletic relationship with at least three FOL origins , . In Japanese isolates, race correlates with the phylogenetic relationship; races 1, 2 and 3 belong to clades A2, A1 and A3, respectively .
Masunaga el al. first reported emergence of race 3 in Japan in 1997 . It is now the number one wilt disease problem in Japan, since most commercial tomato cultivars are resistant to races 1 and 2 but susceptible to race 3. Japanese race 3 isolates all group in clade A3 and are MAT1-2 and VCG 0033 .
In 2008, a new outbreak of Fusarium wilt caused devastating damage to tomato production in greenhouses in Hidaka, Kochi Prefecture, Japan (Fig. S1A, B). The genotype of the affected cultivar was I I2 i3, which suggested the presence of race 3. However, certain characteristics of the pathogenic isolate did not match those reported for previously described Japanese race 3 isolates, suggesting a different biotype, and tomato wilt caused by the novel biotype of FOL race 3 has been occurring in Kochi to date. In this study, the novel biotype was analyzed by phenotypic, genetic and phylogenetic criteria; results suggest a new path for emergence of races.
Results and Discussion
A race 3 isolate, KoChi-1, belongs to a different lineage from the known race 3 isolates in Japan
A fungal isolate from the vascular tissues of diseased tomato in a greenhouse in Kochi Prefecture, Japan was identified as F. oxysporum based on morphology  and nucleotide sequence of the rDNA-internal transcribed spacer (ITS) region (DDBJ/EMBL/GenBank accession No. AB675383). Characteristics of the isolate, designated KoChi-1, are summarized in Table 2. In planta assays showed that KoChi-1 caused wilt disease on cvs. Ponderosa (i i2 i3), Momotaro (I i2 i3) and Walter (I I2 i3), but not on cv. Block (I I2 I3), indicating that KoChi-1 was race 3 (Table 2; Fig. 1A, B). This result was consistent with the fact that the commercial cultivar grown in the greenhouse was Momotaro-Fight (I I2 i3, Takii Seed, Kyoto, Japan).
Figure 1. Virulence of KoChi-1 and its transformants.
(A) KoChi-1 and its AVR1-complements were subjected to pathogenicity evaluation using four tomato cultivars, Ponderosa (i i2 i3), Momotaro (I i2 i3), Walter (I I2 i3) and Block (I I2 I3). The cv. Ponderosa does not have resistance to all FOL races, cv. Momotaro is resistant to FOL race 1 and susceptible to races 1 and 2, cv. Walter is susceptible to race 3 and resistant to races 1 and 2, and cv. Block is resistant to all FOL races. Inocula are as follows: KoChi-1 and its three transformants, K-B-b, K-2-11 and K-2-12; controls, race 1 MAFF 305121 (AVR1 AVR2 AVR3), race 2 JCM 12575 (– AVR2 AVR3) and race 3 Chz1-A (– avr2 AVR3). As a negative control, sterilized water was used (Mock). After three weeks of inoculation. (B) The disease severity of each individual was evaluated on external symptoms with 0~4 scale, respectively. The external symptoms were scored as follows: 0, no wilt or yellowing; 1, lower leaves are yellowing; 2, lower and upper leaves are yellowing; 3, lower leaves are yellowing and wilt and upper leaves are yellowing; 4, all leaves are wilt and yellowing or dead. The symptoms were evaluated after three weeks of inoculation. Four plants were used in each isolate, with three replicates.doi:10.1371/journal.pone.0044101.g001
Table 2. Summary of characteristcis of KoChi-1 and other FOL isolates.doi:10.1371/journal.pone.0044101.t002
Previous studies found that all race 3 isolates obtained in Japan (representative isolate Chz1-A is presented in Table 2) grouped in the A3 clade  (Table 2; Fig. S2), and were MAT1-2 and VCG 0033. However, we found that KoChi-1 belongs to the A2 clade (Table 2; Fig. S2), and is MAT1-1 and VCG 0030+0032. The A2 clade has been reported to include only race 1 isolates in Japan . Taken together, these characteristics suggest KoChi-1 is a novel biotype of race 3, distinct from the race 3 isolates previously reported in Japan.
KoChi-1 is the first reported race 3 isolate carrying the AVR1 locus, which itself is truncated by a transposon
Although previously reported race 3 isolates (e.g., Chz1-A) have no AVR1 locus , , Southern blot analysis using an AVR1 fragment from race 1 isolate MAFF 305121 (733 bp, nt 673–1406 bp, AB674509) as a probe presented that KoChi-1 possessed a single copy of AVR1 in its genome (Fig. 2A).
Figure 2. AVR1 in KoChi-1 genome was truncated by a transposon Hormin.
(A) Southern blot analysis to investigate the copy number of AVR1 gene. AVR1 probe was prepared using a primer set SIX4F/SIX4R (Table 3), and each gDNA was digested with restriction enzyme, EcoRV or NdeI (Fig. 2C). (B) Detection of AVR1 locus from KoChi-1 using a primer set SIX4f-F2/SIX4f-R2 (Table 3, Fig. 2C). (C) Schematic representation of AVR1 locus and AVR1 gene truncated by a transposon Hormin (avr1). The nonautonomous transposon Hormin (759 bp, shown in orange square) is inserted in the second exon of AVR1 in KoChi-1, Hormin harbors 15-bp tandem inverted repeats (TIRs, shown in blue triangle in white square) “CAGGGTTCAAATCCA” and 8-bp target site duplications (TSDs, shown with black line) “CACACCGG”. Arrows show primers. E, EcoRV site; N, NdeI site.doi:10.1371/journal.pone.0044101.g002
Then, we tried to amplify AVR1 from KoChi-1 using a primer set SIX4f-F2/SIX4f-R2 designed by Rep & Houterman to amplify AVR1 from race 1 (Table 3). The amplicon from KoChi-1 (2685 bp) was longer than that of MAFF 305121 (1924 bp) (Fig. 2B). The sequence of KoChi-1 AVR1 was deposited in DDBJ/EMBL/GenBank databases with the accession No. AB674508. In this paper, nucleotide positions are assigned according to AB674508 unless otherwise stated.
Table 3. Primers used in this study.doi:10.1371/journal.pone.0044101.t003
The structure of KoChi-1 AVR1 was compared with that of the race 1 isolate Fol004 (nt 326–2248 in AM234064; Fig. 2C). KoChi-1 contained a different number (13 bp, nt 30–42) of contiguous guanines and one cytosine deletion (nt 2136, AM234064) in addition to a 759 bp-insertion. This small number of polymorphisms suggests that the AVR1 locus is highly conserved. AVR1 in race 1 is composed of two exons (154 and 575 bp) and one intron (64 bp), and encodes a protein of 242 amino acids  (Fig. S4), but the KoChi-1 AVR1 sequence had a 759-bp insertion (nt 1043–1801) in exon 2.
BLASTN searches in the NCBI database suggested that the 759-bp insertion was a transposon with 15-bp terminal inverted repeats (TIRs; 5′-CAGGGTTCAAATCCA-3′; nt. 1043–1057, 1787–1801; Fig. 2C), and that both TIRs were flanked by 8-bp target site duplication (TSD; 5′-CACACCGG-3′; nt 1035–1042, 1802–1809; Fig. 2C). The sequence of the TIRs and the 5′ region of the transposon were highly homologous to the autonomous transposon Hornet1 from F. oxysporum (AF076626) . These characteristics are consistent with those of the hAT family of class II DNA transposons . Hence, we have designated this transposon Hormin (Hornet1 in miniature). Hormin does not encode transposases (and is therefore not autonomous) and may have emerged from Hornet1 through a series of mutations. A transposon identical to Hormin was previously reported in the alcohol dehydrogenase gene Adh1 in FOL NRRL 34936 . This is the first report of an F. oxysporum AVR gene truncated by a transposon.
According to the Broad Institute Fusarium genome database website (http://www.broadinstitute.org/annotation/genome/fusarium_group/MultiHome.html), only 2 isolates, FOL race 2 NRRL 34936 (Spain, MAT1-1, VCG 0030) and FOL race 3 NRRL 54003 (USA, MAT1-2, VCG 0033), carried Hormin-identical sequences (72 and 2 copies, respectively) among 13 isolates of Fusarium spp. In NRRL 34936, Hormin was distributed on almost every chromosome (Fig. S3). On the other hand, F. oxysporum f. spp. raphani (NRRL 54004, pathogenic to radish and Arabidopsis), pisi (NRRL 37622, pathogenic to pea), vasinfectum (NRRL 25433, pathogenic to cotton), melonis (NRRL 26406, pathogenic to melon), conglutinans (PHW808, pathogenic to cabbage), and two F. oxysporum isolates (Fo47, a nonpathogenic isolate; FOSC 3-a, pathogenic to immunocompromised humans) had several Hormin-like (85.8~99.8% homology) sequences.
KoChi-1 avr1 encodes a defective protein
The deduced amino acid sequence of KoChi-1 AVR1 with Hormin revealed a chimeric protein of 175 amino acids (avr1; Fig. S4) that may be nonfunctional. Here, we designate the AVR1 gene truncated with Hormin as avr1. To investigate the transcription of avr1, total RNA was extracted from tomato roots inoculated with KoChi-1 or MAFF 305121 (race 1, as a control). RT-PCR using primer set SIX4F/SIX4R (designed to amplify AVR1 including its intron) amplified a 734-bp fragment from MAFF 305121 RNA but not from KoChi-1 (Fig. 3). On the other hand, RT-PCR using primer SIX4F with primer hornet-like2 (designed on Hormin, see Table 3, Fig. 2C) generated a 440-bp fragment from KoChi-1 inoculated tomato only (Fig. 3), indicating that KoChi-1 avr1 is expressed in planta. Neither avr1 in KoChi-1 nor AVR1 in MAFF 305121 was expressed in mycelia grown on PDB or MM medium (data not shown). This expression pattern was consistent with that of AVR3 in FOL race 2 Fol007 .
Figure 3. Gene expression of AVR1, avr1, AVR2 (avr2) and AVR3.
Eight days after inoculation with race 1 MAFF 305121 (AVR1 AVR2 AVR3), race 3 KoChi-1 (avr1 avr2 AVR3) and the three transformants (avr1 AVR1 avr2 AVR3); K-B-b, K-2-11 and K-2-12, total RNA was extracted from the roots of tomato (cv. Ponderosa) and investigated the transcription of genes AVR1, avr1, AVR2 (avr2), AVR3, FEM1 and Actin with the primer sets SIX4F/SIX4R, SIX4F/hornet-like2, FP962/FP963, P12-F1/P12-R1, FP157/FP158 and Actin-f/Actin-r, respectively (Table 3). FEM1 and actin are used as controls for constitutively-expressed genes in fungal and plant tissues, respectively. Sterilized water is used as a negative control.doi:10.1371/journal.pone.0044101.g003
Other KoChi-1 AVR genes
KoChi-1 avr2 contains the previously known point mutation G121A; it is one of three mutations known to cause loss of AVR2 function in race 3 isolates . KoChi-1 AVR3 has a glutamine (E) type mutation (Table 2). To date, there have been no reports of E type AVR3 mutations in race 3 . Both avr2 and AVR3 of KoChi-1 were expressed during infection of tomato roots (Fig. 3).
Complementation of KoChi-1 avr1 with AVR1 results in loss of pathogenicity to cultivars carrying the I gene
KoChi-1 (avr1 avr2 AVR3) was transformed with the Fol004 (race 1) AVR1 gene. Each of three transformants (K-B-b, K-2-11 and K-2-12) had one copy of AVR1 integrated ectopically into chromosomal DNA to yield strains with the genotype (avr1 AVR1 avr2 AVR3) (Fig. S5); the AVR1 transgene was expressed (Fig. 3). Each of the three transformants lost pathogenicity to tomato cultivars carrying the I gene, e.g., Momotaro (I i2 i3) and Walter (I I2 i3) (Fig. 1A, B). This confirms that avr1 is not functional, and indicates that the mutation can be complemented by AVR1. It also indicates that the integrated AVR1 functioned in spite of coexisting with avr1.
How and where did KoChi-1 emerge?
According to the Broad Institute Fusarium genome database, FOL race 2 isolate NRRL 34936 bears AVR2, AVR3 and genes encoding small proteins secreted into tomato xylem on a small (ca. 2.2 Mb) chromosome. Since the chromosomal location of AVR1 is unknown, we investigated the location of KoChi-1 avr1 by CHEF Southern hybridization (Fig. 4A, B). avr1 was found on a small (ca. 2.5 Mb) chromosome together with avr2 and AVR3 (Fig. 4A, B; lane 8), which was also the case for AVR1 in race 1 isolates MAFF 305121 (1.6 Mb; Fig. 4A, B; lane 1). The small chromosome of each isolate had different size. However, although MAFF 103036 (a Japanese race 1 isolate) was found to carry AVR1 on a ca. 2.5 Mb chromosome, its AVR2 and AVR3 genes were found on a ca. 1.0 Mb chromosome (Fig. 4A, B; lane 2). Perhaps in MAFF 103036, chromosomal fragmentation resulted in relocation of AVR2 and AVR3 to an independent small chromosome. All race 2 and race 3 isolates carried AVR2 or avr2 and AVR3 on chromosomal DNA, but none of them had the AVR1 or avr1.
Figure 4. Localization of avr1/AVR1, AVR2 and AVR3 on the chromosomes of KoChi-1 and other FOL isolates.
(A) Karyotype of FOL isolates by CHEF-gel electrophoresis. Electrophoresis was performed in 1.0% Sea Kem gold agarose gel with CHEF Mapper XA Pulsed Field Electrophoresis System, as following condition; 260 hours run at 8°C, 1200-4800 s switch time at 1.5 V/cm. MAFF 305121 (AVR1 AVR2 AVR3, Japan); MAFF 103036 (AVR1 AVR2 AVR3, Japan); 73 (AVR1 AVR2 AVR3, Italy); Ita3 (AVR1 AVR2 AVR3, Italy); JCM 12575 (– AVR2 AVR3, Japan); NRRL 34936 (– AVR2 AVR3, Spain); Chz1-A (– avr2 AVR3, Japan); KoChi-1 (avr1 avr2 AVR3, Japan). The chromosomes of Saccharomyces cerevisiae and Schizosaccharomyces pombe were used as CHEF DNA size markers. (B) Southern blot analysis probed with AVR1 (upper), AVR2 (middle) and AVR3 (bottom). Probes to detect AVR1 (avr1), AVR2 (avr2) and AVR3 were prepared using primer sets SIX4F/SIX4R, SIX3-F1/SIX3R2 and P12-F2/P12-R1, respectively (Table 3).doi:10.1371/journal.pone.0044101.g004
Mobile elements, together with point mutation in the gene , , , are involved in the loss-of-function of AVR in fungal plant pathogens such as Magnaporthe oryzae and Cladosporium fulvum , , , , . Generally, mobile elements play a role in duplication and translocation of the genes/genomic regions in the genome , , sometimes they cause genetic mutations. AVR genes often locate on mobile element-rich regions in fungal plant pathogens, such as M. oryzae , Leptosphaeria maculans , Blumeria graminis , and F. oxysporum . In Phytophthora infestans, more than five hundreds of potential avirulence genes carrying RxLR motif located in mobile element-rich genomic regions . Moreover, in FOL NRRL 24936 (race 2), a large amount of mobile elements are located on the lineage specific (LS) chromosomes such as Chr03, Chr06, Chr14 (2.2 Mb; the small chromosome carrying AVR2 and AVR3) and Chr15. Of the 72 Hormin elements, 37 are located on LS chromosomes of NRRL 34936 (Fig. S3).
Unlike other fungal isolates, it is easy to speculate how races emerged sequentially in FOL due to its simple combinations of AVR genes and the small number of races. Based on the arms race model , FOL and its races are considered to have emerged as follows  (Fig. S6): First, a nonpathogenic F. oxysporum isolate acquired a small chromosome carrying AVR1, AVR2 and AVR3, and became FOL race 1. The deletion of the AVR1 locus in race 1 resulted in the emergence of race 2 (– AVR2 AVR3), and the point mutation in AVR2 (shown as avr2) in race 2 resulted in the emergence of race 3 (– avr2 AVR3). Refer to Table 2 for relationships among AVR genes, where phylogenetic groups, MAT and VCG of each isolate are also indicated. This study presented an alternative model: AVR1 in a race 1 isolate (AVR1 AVR2 AVR3) lost its function by a transposon insertion, resulting in the emergence of race 2 (avr1 AVR2 AVR3), and race 3 (avr1 avr2 AVR3) emerged from the race 2 as a result of the point mutation (G121A) in AVR2 (Fig. S6). If this scenario describes how KoChi-1 emerged, then where might it have happened? Soilborne pathogens are often carried with seed . KoChi-1 may have been imported on tomato seeds from a production field because we have not found race 2 isolates carrying AVR1 truncated by Hormin, so far, in Japan. There still is the possibility that KoChi-1 evolved via race 2 from a race 1 isolate belonging to the A2 clade in a particular field in Kochi Prefecture. Analysis of more isolates from Kochi, and seed production fields, will be necessary to test these hypotheses.
Materials and Methods
Fungal and plant materials
We sampled diseased tomato (cv. Momotaro-Fight) at a greenhouse in Hidaka, Kochi Prefecture, Japan (latitude, N33°31′53.0"; longitude, E133°21′57.3"; altitude, 32 m) on 4 Feb. 2009. Sampling was permitted by the owner of the private land and greenhouse. No other specific permits were required for the described field study. Moreover, the field study did not involve endangered or protected species. All of the isolates obtained from diseased individuals at the field were identified as F. oxysporum based on morphological characteristics . In addition, all isolates showed identical phenotypes including virulence, mating type (MAT), vegetative compatibility (VC), combination of avirulence genes (AVR) and sequence of rDNA-IGS and rDNA-ITS regions. One representative isolate (KoChi-1) was chosen for this study. FOL race 1 (MAFF 305121, Japan), race 2 (JCM 12575, Imaichi, Tochigi, Japan, 1988) and race 3 (Chz1-A, Yatsushiro, Kumamoto, Japan, 2006) isolates were used as controls. OSU-451B (race 1, VCG 0031; a gift from H. C. Kistler, USDA and University of Minnesota, USA), MN-66 (race 2, VCG 0030+0032; a gift from H. C. Kistler) and H-1-4 (race 3, VCG 0033; a gifte from Y. Hosobuchi, Sakata Seed, Japan) were used for vegetative compatibility group (VCG) determination. OSU-451B and MN-66 were imported to Japan under special permission of Ministry of Agriculture, and Forestry and Fisheries of Japan. All isolates were stored in 25% glycerol at −150°C.
Four race differential cultivars of tomato; Ponderosa (Noguchi Seed, Saitama, Japan), Momotaro (Takii seeds, Kyoto, Japan), Walter (gifted from National Institute of Vegetable and Tea Science, Mie, Japan) and Block (Sakata Seed, Yokohama, Japan) were used. Ponderosa (i i2 i3) is susceptible to all FOL races, Momotaro (I i2 i3) is resistant to FOL race 1 but susceptible to races 2 and 3, Walter (I I2 i3) is resistant to races 1 and 2 but susceptible to race 3, and Block (I I2 I3) is resistant to all races.
Race differential tomato cultivars were used to evaluate FOL pathogenicity. Each isolate was cultured on potato sucrose broth (PSB) for 5 days at 25°C and 120 rpm, and conidial suspensions (1.0×107 conidia/ml) were prepared. Two seeds of each cultivar were sown to soil (Kureha Soil, Kureha, Iwaki, Japan) in a plastic pot (7 cm-diam.) and were maintained in a growth chamber (16 hours light at 28°C/8 hours dark at 25°C). Roots of 15-day-old tomato were injured, dipped in a conidial suspension for 5 min, and replanted to well-moistened soil. Two weeks later, external symptoms of each plant were evaluated as follows: 0, no wilt or yellowing; 1, lower leaves yellowing; 2, lower and upper leaves yellowing; 3, lower leaves yellowing and wilting and upper leaves yellowing; 4, all leaves wilted and yellowing or dead.
DNA extraction and standard PCR
Fragments of rDNA-ITS (521 bp) and IGS (598 bp) regions were amplified using primer sets ITS1/ITS4  and FIGS11/FIGS12 , respectively (Table 3). We also amplified fragments of ca. 800 bp of AVR1, ca. 300 bp of AVR2 and ca. 900 bp of AVR3 using primer sets SIX4F/SIX4R, SIX3-F1/SIX3-R2 and P12-F2/P12-R1, respectively (Table 3). Each reaction mixture of 20 µl contained 20 ng of gDNA, 2.0 µl of 10×buffer (Takara Bio, Ohtsu, Japan), 1.6 µl of 2.5 mM (each) dNTPs (Takara-Bio), 8 pM of each primer, and 0.5 U of Ex-Taq polymerase (Takara Bio). Thermal conditions were as follows: One incubation at 94°C for 2 min; 30 cycles of: denaturation at 94°C for 30 s, annealing at 60°C for 30 s, and elongation at 72°C for 30 s; and a final extension at 72°C for 7 min. To amplify the fragment (ca. 2.0 kb) of the AVR1 locus by SIX4f-F2/SIX4f-R2 (Table 3), we modified the annealing temperature and extension time to 45°C and 2 min, respectively.
PCR amplicons purified with EXOSAP-IT (USB, Cleveland, USA) or 100 ng of plasmids were subjected to sequencing reaction using BigDye® Terminator v3.1 Cycle Sequencing Kit (Applied Biosystems) and analyzed with a 3130×l Genetic Analyzer (Applied Biosystems). Sequence was arranged with GENETYX ver. 13 (Genetyx, Tokyo, Japan).
Nucleotide sequences of the rDNA-IGS fragment from KoChi-1 were aligned with those from other FOL isolates using CLUSTAL X 2.0 . We constructed the phylogeny by the neighbor joining (NJ) method  based on Kimura's two-parameter model , using MEGA v. 4 . The statistical reliability of each node was assessed using 1000 bootstrap iterations. F. sacchari (synonym, Gibberella sacchari; mating population B of the G. fujikuroi-species complex) FGSC 7610 was used as an outgroup. All sequence data except for KoChi-1 were cited from the NCBI database.
Mating type (MAT) and vegetative compatibility group (VCG) determination
Mating type, MAT1-1 or MAT1-2, was determined by PCR using Gfmat1a/Gfmat1b or GfHMG11/GfHMG12, respectively (Table 3). The reaction mixture was prepared as described in the section of Standard PCR, reaction conditions were set as follows: One incubation at 94°C for 2 min; 30 cycles of: denaturation at 94°C for 30 s, annealing at 58°C for 30 s, and elongation at 72°C for 45 s; and a final extension at 72°C for 6 min.
We also identified the VCG type of each isolate. To date, four vegetative compatibility groups (VCGs), 0030+0032, 0031, 0033 and 0035, have been reported in FOL . The complementation test was performed using the tester isolates, OSU-451B (VCG 0031), MN-66 (VCG 0030+0032) and H-1-4 (VCG 0033). Each nitrate nonutilizing (nit) mutant of each isolate (nit1 and NitM) was prepared, and a compatibility test was performed following the procedures described previously .
Gene expression analysis
Tomato cv. Ponderosa was inoculated with F. oxysporum as described in the section entitled “Pathogenicity assay”. Eight days after inoculation, we vigorously washed the tomato roots with sterilize water. After drying with paper towels, roots were crushed in liquid nitrogen and total RNA was extracted with the SV Total RNA Isolation System (Promega) following the manufacturer's manual. From the extracted total RNA, cDNA was synthesized using TaKaRa RNA PCR Kit (AMV) Ver. 3.0 (TaKaRa Bio). Expression of target genes was examined with 5 ng of cDNA. To investigate expression of AVR1, avr1, AVR2, AVR3, FEM1 and the tomato actin gene, primer sets SIX4F/SIX4R, SIX4F/hornet-like2, FP962/FP963, and FP157/FP158, and Actin-f/Actin-r (Table 3) were used for PCR, respectively. FEM1 ,  and Actin  were used as controls for fungal and plant genes, respectively. Negative controls substituted sterile water for conidial suspension. Reaction mixtures were prepared as described above. Thermal conditions were: One incubation at 94°C for 2 min; 35 cycles of: denaturation at 94°C for 30 s, annealing at 57°C for 30 s, and elongation at 72°C for 30 s; and a final extension at 72°C for 7 min.
Complementation with AVR1 using Agrobacterium tumefaciens-mediated transformation (ATMT)
The AVR1 gene of FOL race 1 Fol004 was integrated into the KoChi-1 genome ectopically by the ATMT method. Transformation using the binary vector pPHSIX4c (carrying about 2.0 kb of AVR1 locus and phleomycin resistance gene)  was carried out following the procedure described earlier  with minor modifications. To suppress the growth of Agrobacterium after transformation, we used 25 µg/ml Melopen (Dainippon Sumitomo Phama, Osaka, Japan) and 50 µg/ml Zeocin (Invitrogen, San Diego, USA), respectively.
Contour-clamped homogeneous electric field (CHEF)-gel analysis
In addition to KoChi-1, we used several race 1 isolates; MAFF 305121 (Japan), MAFF 103036 (Japan), 73 (Italy; gift from Corby H. Kistler) and Ita3 (Italy; gift from Giorgia Ferro, The Regional Center For Agricultural Experimentation and Assistance, Italy): race 2 isolates; NRRL 24936 (Spain; gift from A. Di. Pietro, University of Cordoba, Spain) and JCM 12575 (Japan) and race 3 isolate, Chz1-A (Japan). Protoplasts were prepared following  with slight modification; we used enzyme solution containing 1.0% Lysing enzymes (Sigma, St. Louis, USA) and 1.0% Driselase (ASKA Pharmaceutical, Tokyo, Japan) for digestion of fungal cell wall, and Proteinase K (Nakarai Tesk, Kyoto, Japan) was used for plug purification.
CHEF gel electrophoresis was performed in 1.0% Sea Kem gold agarose gel (FMC BioProducts, Rockland, USA) with CHEF Mapper® XA Pulsed Field Electrophoresis System (BioRad, Hercules, USA). The condition to separate chromosomes was as described earlier  with slight modification; 260 hours run at 8°C, 1200–4800 s switch time at 1.5 V/cm. The running buffer 0.5xTBE was refreshed every 2 days. Chromosomes of Schizosaccharomyces pombe (BioRad) and Saccharomyces cereviciae (BioRad) were used as DNA size markers. The gel was stained with ethidium bromide to visualize chromosomes after running electrophoresis.
Southern blot analysis
Probes for AVR1/avr1, AVR2/avr2 and AVR3 were prepared using SIX4F/SIX4R, SIX3-F1/SIX3-R2 and P12-F2/P12-R1, respectively. For genomic Southern hybridization, 8.0 µg gDNA were digested with NdeI and BssHII, and incubated overnight at 37°C. The following procedure was performed as described earlier , note that Whatman Nytran SuPerCharge (SPC) nylon blotting membranes (Sigma) was used in this study.
The CHEF-gel was treated with 0.25 N HCl for 30 min, followed by denaturation buffer (0.5 M NaOH. 1.5 M NaCl), and the digested chromosomes were transferred to a nylon membrane (Hybond N+; Amersham, Amersham, UK) washed in 0.4 M NaOH for about 72 hours. The following procedure after transfer was performed as early study . For stripping the hybridized probe, the used membrane was washed twice, for 15 min each, with 0.2 M NaOH, 0.1% SDS at 37°C, then the membranes were soaked in 2×SSC for 5 min and dried.
Fusarium wilt of tomato caused by F. oxysporum f. sp. lycopersici in Kochi, Japan. (A) Location of the wilt disease emerged. Asterisk at the tip of bar presents Hidaka, Kochi Prefecture, Japan (latitude, N33°31′53.0"; longitude, E133°21′57.3"; altitude, 32 m). (B) Diseased tomato cultivar Momotaro-Fight (I I2 i3) in a greenhouse in Hidaka, Kochi prefecture, Japan. The diseased tomato plants wilted and the color of the leaves turned yellow. Severely diseased plants did not survive and white hyphae were observed on the lower part of their stems.
Phylogenetic relationship of tomato wilt fungus (FOL) isolates in Japan. KoChi-1 and other FOL races 1~3 isolates obtained in Japan were used. Race, the source, mating type (MAT) and vegetative compatibility group (VCG) were described in parentheses at the end of the isolates name. A hyphen indicates incompatible isolates with VCG testers. Gibberella fujikuroi strain FGSC 7610 was used as the outgroup. The phylogeny was constructed based on Kimura's two-parameter  as nucleotide substitution model using MEGA v. 4 . Bootstrap iterations are 1000 replications, the values are indicated at tree nodes. Bootstrap values greater than 70% are shown beside nodes. The FOL clades A1, A2 and A3 are consistent with the previous study . All sequence data are in the DDBJ/EMBL/GenBank databases; KoChi-1 (AB674508), MAFF 103043 (AB106032), JCM 12575 (AB106027), SUF 1330 (AB106035), MAFF 103038 (AB106031), MAFF 305121 (AB106021), MAFF 103036 (AB106020), MAFF 727501 (AB106022), Chz1-A (AB373819), F-1-1 (AB106037) and FGSC 7610 (AB106061).
Southern blot analysis to detect AVR1 and avr1 genes of KoChi-1 transformants. The probe was prepared using a primer set SIX4F/SIX4R (Table 3, Fig. 2C), each 8.0 µg gDNA was digested with NdeI. Race 1, MAFF 305121 (AVR1 AVR2 AVR3); race 3, KoChi-1 (avr1 avr2 AVR3); transformants, K-B-b, K-2-11 and K-2-12 (avr1 AVR1 avr2 AVR3).
The deduced amino acid sequences of AVR1 in race 1 and avr1 in KoChi-1. The AVR1 is composed of 242 amino acids. The deduced amino acid sequence of AVR1 with Hormin in KoChi-1 revealed a chimeric AVR1 composed of 175 amino acids (avr1) that may not function as AVR1. Black and orange characters show the amino acids encoded by AVR1 and Hormin, respectively. Asterisks show the homologous amino acid.
Hormin distributes on every chromosome of FOL race 2 NRRL 34936. Red arrowheads show the location of Hormin. The figures of the FOL chromosome was cited from the website of Broad Institute (http://www.broadinstitute.org/annotation/genome/fusarium_group/MultiHome.html).
A novel path of emergence of FOL races proposed in this study.
The authors would like to show our gratitude to Olen C. Yoder (Celgene, USA) for discussion of this research and reviewing the manuscript. For Yuji Hosobuchi (Sakata Seed, Kanagawa, Japan), Corby H. Kistler (U.S. Department of Agriculture, USA; University of Minnesota, USA), Antonio Di Pietro (University of Cordoba, Spain), Giorgia Ferro (The Regional Center For Agricultural Experimentation and Assistance, Italy), we would like to thank for providing fungal isolates. The authors also would like to acknowledge Martijn Rep and Petra M. Houterman (Swammerdam Institute for Life Science, University of Amsterdam, Amsterdam, The Netherlands) for their technical support including providing T-DNA for Agrobacterium-mediated transformation, and acknowledge to Kyowa Hakko Kogyo Co. (Tokyo, Japan) for the gift of Driselase.
Conceived and designed the experiments: KI TA. Performed the experiments: KI CYA TA. Analyzed the data: KI TA. Contributed reagents/materials/analysis tools: YM MY. Wrote the paper: KI TT TA.
- 1. Agrios GN (2005) Plant Pathology 5th Edition. California: Academic Press.
- 2. Flor HH (1956) The complementary genetic systems in flax and flax rust. Adv Genet 8: 29–54. doi: 10.1016/s0065-2660(08)60498-8
- 3. Armstrong GM, Armstrong JK (1981) Formae speciales and races of Fusarium oxysporum causing wilt diseases. In: Nelson PE, Toussoun TA, Cook RJ, editors. Fusarium: disease, biology, and taxonomy. : State University Press, pp. 391–399.
- 4. Arie T (2010) Phylogeny and phytopathogenicity mechanisms of soilborne Fusarium oxysporum. J Gen Plant Pathol 76: 403–405. doi: 10.1007/s10327-010-0264-z
- 5. Cai G, Gale LR, Schneider RW, Kistler HC, Davis RM, et al. (2003) Origin of race 3 of Fusarium oxysporum f. sp. lycopersici at a single site in California. Phytopathology 93: 1014–1022. doi: 10.1094/phyto.2003.93.8.1014
- 6. Komada H, Ogawa K, Aoki T (2011) Fusarium. Tokyo: Zenkoku Nouson Kyoiku Kyokai. (in Japanese).
- 7. Huang CC, Lindhout P (1997) Screening for resistance in wild Lycopersicon species to Fusarium oxysporum f. sp. lycopersici race 1 and race 2. Euphytica 93: 145–153.
- 8. Houterman PM, Cornelissen BJC, Rep M (2008) Suppression of plant resistance gene-based immunity by a fungal effector. PLoS Pathog 4: e1000061. doi: 10.1371/journal.ppat.1000061
- 9. Houterman PM, Ma L, van Ooijen G, de Vroomen MJ, Cornelissen BJC, et al. (2009) The effector protein Avr2 of the xylem-colonizing fungus Fusarium oxysporum activates the tomato resistance protein I-2 intracellularly. Plant J 58: 970–978. doi: 10.1111/j.1365-313x.2009.03838.x
- 10. Rep M, van der Does HC, Meijer M, van Wijk R, Houterman PM, et al. (2004) A small, cysteine-rich protein secreted by Fusarium oxysporum during colonization of xylem vessels is required for I-3-mediated resistance in tomato. Mol Microbiol 53: 1373–1383. doi: 10.1111/j.1365-2958.2004.04177.x
- 11. Lievens B, Houterman PM, Rep M (2009) Effector gene screening allows unambiguous identification of Fusarium oxysporum f. sp. lycopersici races and discrimination from other formae speciales. FEMS Microbiol Lett 300: 201–215. doi: 10.1111/j.1574-6968.2009.01783.x
- 12. van der Does HC, Lievens B, Claes L, Houterman PM, Cornelissen BJC, et al. (2008) The presence of a virulence locus discriminates Fusarium oxysporum isolates causing tomato wilt from other isolates. Environ Microbiol 10: 1475–1485. doi: 10.1111/j.1462-2920.2007.01561.x
- 13. Rep M, Meijer M, Houterman PM, van der Does HC, Cornelissen BJC (2005) Fusarium oxysporum evades I-3-mediated resistance without altering the matching avirulence gene. Mol Plant-Microbe Interact 18: 15–23. doi: 10.1094/mpmi-18-0015
- 14. Inami K, Yoshioka C, Hirano Y, Kawabe M, Tsushima S, et al. (2010) Real-time PCR for differential determination of the tomato wilt fungus, Fusarium oxysporum f. sp. lycopersici, and its races. J Gen Plant Pathol 76: 116–121. doi: 10.1007/s10327-010-0224-7
- 15. Kistler HC (1997) Genetic diversity in the plant-pathogenic fungus Fusarium oxysporum. Phytopathology 87, 474–479.
- 16. Kawabe M, Kobayashi Y, Okada G, Yamaguchi I, Teraoka T, et al. (2005) Three evolutionary lineages of tomato wilt pathogen, Fusarium oxysporum f. sp. lycopersici, based on sequences of IGS, MAT1, and pg1, are each composed of isolates of a single mating type and a single or closely related vegetative compatibility group. J Gen Plant Pathol 71: 263–272. doi: 10.1007/s10327-005-0203-6
- 17. O'Donnell K, Gueidan C, Sink S, Johnston PR, Crous PW, et al. (2009) A two-locus DNA sequence database for typing plant and human pathogens within the Fusarium oxysporum species complex. Fungal Genet Biol 46: 936–948. doi: 10.1016/j.fgb.2009.08.006
- 18. Masunaga T, Shiomi H, Komada H (1998) Identification of race 3 of Fusarium oxysporum f. sp. lycopersici isolated from tomato in Fukuoka prefecture. Ann Phytopathol Soc Jpn 64: 435 (in Japanese)..
- 19. Leslie JF, Summerell BA (2006) The Fusarium laboratory manual. Oxford: Blackwell Science. 212.
- 20. Hua-Van A, Daviere JM, Kaper F, Langin T, Daboussi MJ (2000) Genome organization in Fusarium oxysporum: clusters of class II transposons. Curr Genet 37: 339–347. doi: 10.1007/s002940050537
- 21. Kempken F, Windhofer F (2001) The hAT family: a versatile transposon group common to plants, fungi, animals, and man. Chromosoma 110: 1–9. doi: 10.1007/s004120000118
- 22. Corrales EAR, Rangel PRA, Meza CV, Gonzalez HGA, Torres GJC, et al. (2011) Fusarium oxysporum Adh1 has dual fermentative and oxidative functions and is involved in fungal virulence in tomato plants. Fungal Genet Biol 48: 886–895. doi: 10.1016/j.fgb.2011.06.004
- 23. van der Does HC, Duyvesteijn RGE, Goltstein PM, van Schie CCN, MandersEMM, et al (2008) Expression of effector gene SIX1 of Fusarium oxysporum requires living plant cells. Fungal Genet Biol 45: 1257–1264. doi: 10.1016/j.fgb.2008.06.002
- 24. Parlange F, Daverdin G, Fudal I, Kuhn ML, Balesdent MH, et al. (2009) Leptosphaeria maculans avirulence gene AvrLm4-7 confers a dual recognition specificity by the Rlm4 and Rlm7 resistance genes of oilseed rape, and circumvents Rlm4-mediated recognition through a single amino acid change. Mol Microbiol 71: 851–863. doi: 10.1111/j.1365-2958.2008.06547.x
- 25. Hogenhout SA, Van der Hoorn RA, Terauchi R, Kamoun S (2009) Emerging concepts in effector biology of plant-associated organisms. Mol Plant Microbe Interact 22: 115–122. doi: 10.1094/mpmi-22-2-0115
- 26. Kang S, Lebrun MH, Farrall L, Valent B (2001) Gain of virulence caused by insertion of a Pot3 transposon in a Magnaporthe grisea avirulence gene. Mol Plant Microbe In 14: 671–674. doi: 10.1094/mpmi.2001.14.5.671
- 27. Fudal I, Böhnert HU, Tharreau D, Lebrun MH (2005) Transposition of MINE, a composite retrotransposon, in the avirulence gene ACE1 of the rice blast fungus Magnaporthe grisea. Fungal Genet Biol 42: 761–772. doi: 10.1016/j.fgb.2005.05.001
- 28. Zhou E, Jia Y, Singh P, Correll JC, Lee FN (2007) Instability of the Magnaporthe oryzae avirulence gene AVR-Pita alters virulence. Fungal Genet Biol 44: 1024–1034. doi: 10.1016/j.fgb.2007.02.003
- 29. Li W, Wang B, Wu J, Lu G, Hu Y, et al. (2009) The Magnaporthe oryzae avirulence gene AvrPiz-t encodes a predicted secreted protein that triggers the immunity in rice mediated by the blast resistance gene Piz-t. Mol Plant Microbe In 22: 411–420. doi: 10.1094/mpmi-22-4-0411
- 30. Luderer R, Takken FLW, de Wit P, Joosten M (2002) Cladosporium fulvum overcomes Cf-2-mediated resistance by producing truncated AVR2 elicitor proteins. Mol Microbiol 45: 875–884. doi: 10.1046/j.1365-2958.2002.03060.x
- 31. Chuma I, Isobe C, Hotta Y, Ibaragi K, Futamata N, et al. (2011) Multiple Translocation of the AVR-Pita effector gene among chromosomes of the rice blast fungus Magnaporthe oryzae and related species. PLoS Pathog 7: e1002147. doi: 10.1371/journal.ppat.1002147
- 32. Dean RA, Talbot NJ, Ebbole DJ, Farman ML, Mitchell TK, et al. (2005) The genome sequence of the rice blast fungus Magnaporthe grisea. Nature 434: 980–986. doi: 10.1038/nature03449
- 33. Sacristan S, Vigouroux M, Pedersen C, Skamnioti P, Thordal-Christensen H, et al. (2009) Coevolution between a Family of Parasite Virulence Effectors and a Class of LINE-1 Retrotransposons. PLoS One 4: e7463. doi: 10.1371/journal.pone.0007463
- 34. Ma LJ, van der Does HC, Borkovich KA, Coleman JJ, Daboussi MJ, et al. (2010) Comparative genomics reveals mobile pathogenicity chromosomes in Fusarium. Nature 464: 367–373. doi: 10.1038/nature08850
- 35. Haas BJ, Kamoun S, Zody MC, Jiang RH, Handsaker RE, et al. (2009) Genome sequence and analysis of the Irish potato famine pathogen Phytophthora infestans. Nature 461: 393–398. doi: 10.1038/nature08358
- 36. Maor R, Shirasu K (2005) The arms race continues: battle strategies between plants and fungal pathogens. Curr Opin Microbiol 8: 399–404. doi: 10.1016/j.mib.2005.06.008
- 37. Arie T, Kaneko I, Yoshida T, Noguchi M, Nomura Y, et al. (2000) Mating-type genes from asexual phytopathogenic ascomycetes Fusarium oxysporum and Alternaria alternata. Mol Plant Microbe In 13: 1330–1339. doi: 10.1094/mpmi.2000.13.12.1330
- 38. Saitoh K, Togashi K, Arie T, Teraoka T (2006) A simple method for a mini-preparation of fungal DNA. J Gen Plant Pathol 72: 348–350. doi: 10.1007/s10327-006-0300-1
- 39. White TJ, Bruns T, Lee S, Taylor JW (1990) Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenetics. In: Innis MA, Gelfand DH, Sninsky JJ, White TJ, editors. PCR Protocols: A Guide to Methods and Applications. New York: Academic Press, Inc. pp. 315–322.
- 40. Larkin MA, Blackshields G, Brown NP, Chenna R, McGettigan PA, et al. (2007) Clustal W and Clustal X version 2.0. Bioinformatics 23: 2947–2948. doi: 10.1093/bioinformatics/btm404
- 41. Saitou N, Nei M (1987) The neighbor-joining method: a new method for reconstructing phylogenetic trees. Mol Biol Evol 4: 406–425.
- 42. Kimura M (1980) A simple method for estimating evolutionary rates of base substitutions through comparative studies of nucleotide sequences. J Mol Evol 16: 111–120. doi: 10.1007/bf01731581
- 43. Tamura K, Dudley J, Nei M, Kumar S (2007) MEGA4: molecular evolutionary genetics analysis (MEGA) software version 4.0. Mol Biol Evol 24: 1596–1599. doi: 10.1093/molbev/msm092
- 44. Correll JC, Klittich CJR, Leslie JF (1987) Nitrate nonutilizing mutants of Fusarium oxysporum and their use in vegetative compatibility tests. Phytopathology 77: 1640–1646. doi: 10.1094/phyto-77-1640
- 45. Schoffelmeer EAM, Vossen JH, van Doorn AA, Cornelissen BJC, Haring MA (2001) FEM1, a Fusarium oxysporum glycoprotein is covalently linked to the cell wall matrix and is conserved in filamentous fungi. Mol Genet Genomics 265: 143–152. doi: 10.1007/s004380000402
- 46. Michielse CB, van Wijk R, Reijnen L, Manders EMM, Boas S (2009) The nuclear protein Sge1 of Fusarium oxysporum is required for parasitic growth. PLoS Pathog 5: e1000637. doi: 10.1371/journal.ppat.1000637
- 47. Aimé S, Cordier C, Alabouvette C, Olivain C (2008) Comparative analysis of PR gene expression in tomato inoculated with virulent Fusarium oxysporum f. sp. lycopersici and the biocontrol strain F. oxysporum Fo47. Physiol Mol Plant Pathol 73: 9–15.
- 48. Mes JJ, Wit R, Testerink CS, de Groot F, Haring MA, Cornelissen BJC (1999) Loss of avirulence and reduced pathogenicity of a gamma-irradiated mutant of Fusarium oxysporum f. sp. lycopersici. Phytopathology 89: 1131–1137. doi: 10.1094/phyto.1922.214.171.1241
- 49. Barve MP, Arie T, Salimath SS, Muhelbauer FJ, Peever TL (2003) Cloning and characterization of the mating type (MAT) locus from Ascochyta rabiei (teleomorph: Dydimella rabiei) and a MAT phylogeny of legume associated Ascochyta spp. Fungal Genet Biol 39: 151–167. doi: 10.1016/s1087-1845(03)00015-x