Genetic variation of Fusarium oxysporum isolates forming fumonisin B1 and moniliformin

Thirty single-spore isolates of a toxigenic fungus, Fusarium oxysporum, were isolated from asparagus spears and identified by species-specific polymerase chain reaction (PCR) and translation elongation factor 1-α (TEF) sequence analysis. In the examined sets of F. oxysporum isolates, the DNA sequences of mating type genes (MAT) were identified. The distribution of MAT idiomorph may suggest that MAT1-2 is a predominant mating type in the F. oxysporum population. F. oxysporum is mainly recognised as a producer of moniliformin—the highly toxic secondary metabolite. Moniliformin content was determined by high-performance liquid chromatography (HPLC) analysis in the range 0.05–1,007.47 μg g−1 (mean 115.93 μg g−1) but, also, fumonisin B1 was detected, in the concentration range 0.01–0.91 μg g−1 (mean 0.19 μg g−1). There was no association between mating types and the mycotoxins biosynthesis level. Additionally, a significant intra-species genetic diversity was revealed and molecular markers associated with toxins biosynthesis were identified.


Introduction
Asparagus (Asparagus officinalis L.) is a worldwide wellknown vegetable not only due to its high nutritional value, but also because of the biologically active compounds present, including protodioscin and rutin, with a strong cytotoxic activity against many cancer cell lines (Kinghorn et al. 2003;Chin and Garrison 2008). However, the asparagus is often infected by fungi of the Fusarium genus, a causal agent of crown and root rot which decreases the quantity and quality of the yield. A visible chlorosis and wilting of the individual stalks, progressive discolouration of plants and destruction of roots and crown are the main symptoms of this serious disease (Logrieco et al. 2003). F. oxysporum Schlechtend. emend. Snyder & Hansen and F. proliferatum (Matsus.) Nirenberg are the most severe asparagus pathogens (Blok and Bollen 1995;von Bargen et al. 2009). Both species are well-known mycotoxins producers, so their precise identification is of prime concern. New molecular methods have been applied to complement, support and confirm identification based on morphological characters (Geiser et al. 2004).
F. oxysporum is identified worldwide and is known to be phylogenetically diverse and recognised as a moniliformin (MON) producer (Logrieco et al. 2003). Recent studies suggest that some F. oxysporum strains can also biosynthesise fumonisin B 1 -FB 1 (Waśkiewicz et al. 2010a;Stępień et al. 2011a, b). MON exhibits cytotoxic and cardiotoxic activity, causes developmental disorders and may also induce the development of Keshan disease, human myocardial impairment reported in rural areas of China and South Africa (Pineda-Valdes and Bullerman 2000). Fumonisins are polyketide-derived mycotoxins inhibiting a ceramide synthase, the key enzyme in the sphingolipid biosynthetic pathway, resulting in serious mycotoxicoses (Soriano and Dragacci 2004;Voss et al. 2007). Contamination of food with FB 1 has been associated with an oesophageal cancer of humans in Africa, China and the United States (Nicholson 2004). Considering an available toxicological evidence, the International Agency for Research on Cancer (IARC) describe FB 1 as a probable carcinogenic to humans (IARC 2002). Food processing does not solve the problem because the hydro-thermal stability of fumonisin is high and decomposition products are more toxic than fumonisin itself (Bullerman et al. 2002).
F. oxysporum is a large taxonomic unit, described as a species complex. So far, no known sexual stage of the F. oxysporum is available and the concept of a biological species and the sources of genetic variation are still discussed (Kistler 1997;O'Donnell et al. 2009;Kück and Pöggeler 2009). In most fungi, the mating type locus conferring mating behaviour consists of dissimilar DNA sequences (idiomorphs) in the mating partners (Pöggeler 2001). In heterothallic species, the mating type is controlled by a single locus with two idiomorphic alleles, designated MAT1-1 and MAT1-2 (Kerényi et al. 1999;Turgeon and Yoder 2000). Each MAT idiomorph carries one gene encoding a single MAT-specific DNA binding protein. These proteins probably play a crucial role in the pathways of cell speciation and sexual morphogenesis as regulatory transcription factors. The two MAT alleles contain a conserved alpha (ALPHA) box domain or a high mobility group (HMG) box domain, respectively (Yun et al. 2000). Evolutionary theory indicates that sexual reproduction plays an important role in a pathogen's evolution, i.e. the development of a new pathogenic race compatible with a diseaseresistant cultivar and of a new strain resistant to a fungicide. Recently, it was shown that the presence of the MAT1-2 idiomorph in the F. culmorum genome significantly affects mycelium growth (unpublished data). Assessing the possibility of mating by toxigenic strains is important for the design of successful control strategies, since these strategies are different for clonally and sexually reproducing organisms (McDonald and McDermott 1993). Moreover, there is poor information on the toxicity of pathogenic F. oxysporum isolates with different mating type alleles.
The present study focused on the genetic and toxigenic differentiation of F. oxysporum isolates. The principal aims of the study were to estimate mycotoxins biosynthesis yield and to detect mating type idiomorphs in the F. oxysporum genome. Additionally, the relationship between these features was assessed. Moreover, the level of intra-species genetic diversity were estimated and associations between molecular markers and mycotoxins (FB 1 and MON) formation were examined.

Fungal isolates
Fungi were collected from green (O-02, O-11, O-30) and white (remaining isolates) spears of asparagus 'Gijnlim' and 'Eposs' cultivated in Western Poland. Fungi were isolated from the basal part of spears (one spear per plant) on the borderline between discoloured and healthy tissues. After disinfection with 1% sodium hypochlorite, five sections (about 1 mm in diameter) of tissue were cut from spears and such a set sections was transferred onto a Petri dish with potato dextrose agar (PDA) medium. One of these fungal cultures from spears was randomly chosen for further work. Thirty single-spore cultures of F. oxysporum were used in the chemical and molecular analyses. All fungal isolates (O-01-O-30) were identified on the basis of their morphology (Booth 1971;Gerlach and Nirenberg 1982;Kwasna et al. 1991). The morphological features of the identified isolates were compared with standard isolates DSM-62287 (Deutsche Sammlung von Mikroorganismen und Zellkulturen, Germany) of F. oxysporum f. sp. asparagi. Isolates of DSM-62287 in this study were designated as O-31.
DNA preparations and molecular identification of F. oxysporum isolates by PCR and sequencing procedures Mycelia from 9-day-old single-spore cultures of F. oxysporum grown on liquid medium (5 g L −1 of glucose, 1 g L −1 of yeast extract) were separated from solution by vacuum filtration with a Büchner funnel. DNA was extracted and purified using a DNeasy Mini Kit (QIAGEN Inc., Hilden, Germany), according to the manufacturer's recommendations. Polymerase chain reaction (PCR) with species-specific primers designed based on a partial sequence of calmodulin gene (Mulè et al. 2004a, b) was used to confirm the morphological identification of each fungal isolate. For the PCR analysis, a forward primer CLOX1: 5′-CAGCAAAGCATCAGACCAC TATAACTC-3′ and a reverse primer CLOX2: 5′-CTTGTC AGTAACTGGACGTTGGTACT-3′ (Sigma, Pampisford, UK) were used. The amplification reactions were performed with a Taq PCR Core Kit (QIAGEN, Inc., Hilden, Germany). The reaction mixture and amplification conditions were as described previously (Waśkiewicz et al. 2010b). Amplification was carried out in a Biometra Tpersonal 48 thermocycler (Whatman Biometra, Goettingen, Germany).
Additionally, for 12 representative isolates (forming the highest and the lowest amount of FB 1 and MON), a part of the translation elongation factor 1-α (TEF) was amplified with the following primer pair: EF1-forward: 3′-ATGGGTAAG GARGACAAGAC-5′ (O'Donnell et al. 1998) and TEF_WIreverse: 3′-GCTCACGRGTCTGGCCATCCTTG′ (designed in this study). Primer TEF_WI is located about 90 nt downstream of the sequence of the reverse primer designed by O'Donnell et al. (1998). The reaction mixture contained: 20 ng of fungal DNA, 0.2 mM of each dNTP, 1 μM of each primer and 0.5 U of DreamTaq™ polymerase (Fermentas GMBH, St. Leon-Rot, Germany) in 1× reaction buffer with 2 mM MgCl 2 . Amplification was carried out using the following programme: an initial denaturation for 2 min at 94°C, followed by 45 cycles of denaturation at 94°C for 30 s, primer annealing at 58°C for 30 s and extension at 72°C for 1 min. The amplification was ended with an additional extension at 72°C for 5 min. Amplification products were cleaned and then sequenced on ABI PRISM® 310 Analyzer (Applied Biosystems), according to the procedure described previously (Irzykowska et al. 2005). The resulting sequences were checked manually and were aligned using the computer software package CLUSTAL_X (Thompson et al. 1997). The obtained TEF sequences were compared with all the sequences deposited at the Fusarium ID Database (http:// www.fusariumdb.org/index.php).

Genetic fingerprinting of F. oxysporum
Random amplified polymorphic DNA PCRs (RAPD-PCRs) were carried out using a Taq PCR Core Kit (QIAGEN, Inc., Hilden, Germany). The reaction mixture was as described previously (Irzykowska and Bocianowski 2008). Eight random 10-mer and one 11-mer primers, OPB-07, OPC-02, OPC-04, OPC-08, OPC-15, OPJ-10+A, OPL-11, OPL-12 and OPL-19 (Qiagen Operon, Cologne, Germany), were used to screen the isolates for DNA polymorphism. Amplification was carried out using the following programme: an initial denaturation for 2 min at 94°C, followed by 40 cycles of denaturation at 94°C for 30 s, primer annealing at 37°C for 1 min and extension at 72°C for 2 min. The amplification was ended with an additional extension at 72°C for 5 min. PCR was repeated twice in order to check reproducibility. The comparison of each band profile for each primer was performed on the basis of the presence (1) versus the absence (0) of RAPD products of the same length.

Electrophoresis conditions
The PCR products were separated by electrophoresis (4 V cm −1 ) in 1.5% agarose gels with 1× TBE buffer (89 mM Tris-borate and 2 mM EDTA, pH 8.0) and visualised under UV light following ethidium bromide staining. A Gene Ruler™ 100 bp DNA Ladder Plus (Fermentas GMBH, St. Leon-Rot, Germany) was used as a molecular size standard for PCR products.

Fungal cultures for mycotoxin estimation
For toxin quantification, rice cultures were prepared for individual F. oxysporum isolates. Long-grain white rice (RANI) tested as being mycotoxin-free was used as a medium for the formation of mycotoxins by F. oxysporum isolates. Rice samples of 65 g were soaked for 16 h with 81 ml of distilled water in 500-ml Erlenmeyer flasks and then sterilised for 15 min at 120°C. Such samples were inoculated with 5-mm disks of PDA medium overgrown by F. oxysporum mycelium. On rice substrate, fungal cultures were incubated for 3 weeks at 20°C. All samples for analyses were prepared and cultured in triplicate (three Erlenmeyer flasks with the prepared medium) and the mean values of the results were calculated.

Chemicals and reagents
Standards of pure FB 1 and MON were purchased from Sigma (St. Louis, MO, USA). Acetonitrile, methanol (highperformance liquid chromatography [HPLC] grade), disodium tetraborate, 2-mercaptoethanol and t-butyl-ammonium hydroxide were purchased from Sigma-Aldrich. Sodium dihydrogen phosphate, acetic acid, n-hexane, o-phosphoric acid and dichloromethane were purchased from POCh (Gliwice, Poland). Water for the HPLC mobile phase was purified using a Milli-Q system (Milipore, Bedford, MA, USA).

Fumonisin B 1 analysis
Samples (10 g) of dry 3-week-old cultures were homogenised for 3 min in 20 ml of methanol-water (3:1, v/v) and filtered through Whatman no. 4 filter paper, in accordance with the method described by Sydenham et al. (1990). The detailed procedure of the extraction and purification of FB 1 has been reported previously (Waśkiewicz et al. 2010a). Purified FB 1 was quantitatively determined by the HPLC method. The o-phosphoric acid (OPA) reagent (20 mg per 0.5 ml methanol) was prepared and diluted with 2.5 ml of 0.1 M disodium tetraborate (Na 2 B 4 O 7 × 10H 2 O) and then combined with 25 ml of 2-mercaptoethanol added to the solution. The FB 1 standard (5 μl) or extracts (20 μl) were derivatised with 20 or 80 μl of the OPA reagent. After 3 min, the reaction mixture (10 ml) was injected onto an HPLC column. Methanol-sodium dihydrogen phosphate (0.1 M in water) solution (77:23, v/v), adjusted to pH 3.35 with o-phosphoric acid, after filtration through a 0.45-mm Waters HV membrane, was used as the mobile phase with a flow rate of 0.6 ml min −1 . A Waters 2695 apparatus (Waters Division of Millipore, Milford, MA, USA), with a C-18 Nova Pak column (3.9 × 150 mm) and a Waters 2475 fluorescence detector (λ ex 0335 nm, λ em 0440 nm), was used in the metabolite quantitative determination. The identification and quantification of FB 1 was observed by comparison of the retention times and peak areas in the samples with those observed for FB 1 standard and using the relevant calibration curve (correlation coefficients for FB 1 was 0.9967). The limit of detection (LOD) was 10 ng g −1 for FB 1 and was defined as the concentration that was three times higher than the standard deviation of the blank signal. Recovery for FB 1 measured by the mycotoxins extraction from blank samples spiked with 10-100 ng g −1 of the compound was equal to 93%. The relative standard deviations (RSDs) were lower than 8%.

Moniliformin analysis
Culture samples (15 g) of each strain were homogenised with 75 ml of acetonitrile-methanol-water (16:3:1, v/v/v) and filtered (Whatman no. 4 filter paper). Moniliformin was extracted and purified according to the methods described by Waśkiewicz et al. (2010b). MON was quantified using a Waters 501 apparatus (Waters Division of Millipore) with a C-18 Nova Pak column (3.9 × 300 mm) and a Waters 486 UV detector (λ max 0229 nm). Acetonitrile-water (15:85, v/v) buffered with 10 ml of 0.1 M K 2 HPO 4 in 40% t-butylammonium hydroxide in 1 L of solvent (Sharman et al. 1991) was used as the mobile phase (flow rate00.6 ml min −1 ). The MON detection limit was 25 ng g −1 . Positive results (on the basis of retention time) were confirmed by the HPLC analysis and by comparison with the relevant calibration curve (the correlation coefficient for MON was 0.9990). The recovery for MON was 90%. The RSD was below 7%.

Statistical analysis
The one-way analysis of variance (ANOVA) of F. oxysporum occurrence was carried out in order to determine the differentiation of isolates regarding the concentrations of FB 1 and MON. Tukey's least significant differences (LSDs) for each trait were calculated. Homogeneous groups (not significantly different from each other) for the analysed traits were determined on the basis of least significant differences. The relationship between FB 1 and MON concentrations was estimated using correlation coefficients (Kozak et al. 2010).
The coefficients of genetic similarity (S) of the investigated isolates were calculated using the following formula (Nei and Li 1979): where N ij is the number of alleles present at the i-th and j-th isolates, N i is the number of alleles present at the i-th isolate, N j is the number of alleles present at the j-th isolate, and i, j01, 2, …, 31. On the basis of the calculated coefficients, isolates were grouped hierarchically using the unweighted pair group method of arithmetic means (UPGMA). The relationship among isolates was presented in the form of a dendrogram (Kumar et al. 2004).
The association between RAPD markers and the FB 1 and MON concentrations of F. oxysporum isolates was estimated by regression analysis (Bocianowski and Seidler-Łożykowska 2012). The molecular marker observations were treated as independent variables and considered in individual models. The relationship between MAT idiomorphs and the biosynthesis of FB 1 and MON were performed by the t-test (Snedecor and Cochran 1989). All statistical analyses were performed using the package GenStat v.7.1. (Payne et al. 2003).

Identification of F. oxysporum isolates
All of the tested isolates displayed morphology typical for F. oxysporum, quite similar to standard F. oxysporum f. sp. asparagi isolates DSM-62287 (here, O-31). Species-specific PCR assay targeting the calmodulin gene confirmed morphological species identification. Specific PCR products obtained for each isolate, including the reference one, was 534 bp in length.
Additionally, for representative isolates producing and non-producing mycotoxins, a 810-bp fragment of the TEF gene was sequenced. The sequences were compared with sequences deposited in the Fusarium ID database. The comparison with the TEF sequence in the Fusarium ID database showed 100% matching of all isolates examined with F. oxysporum, confirming the species identity (Table 1).

Evaluation of mycotoxins level
HPLC analysis revealed that the majority of F. oxysporum isolates produce both MON and FB 1 . The results of one-way ANOVA indicate that isolates differ significantly in their toxigenic activity and mycotoxin profile (Table 2). FB 1 was determined in the concentration range 0.01-0.91 μg g −1 (mean 0.19 μg g −1 ), while MON was detected in the range 0.5-1,007.47 μg g −1 (mean 115.93 μg g −1 ). Fumonisin B 1 and moniliformin were formed in the highest concentrations by isolates O-10 and O-15, respectively (Table 3). Reference isolate (O-310DSM-62287) produced only FB 1 at a concentration of 0.2 μg g −1 . The FB 1 biosynthesis level allowed to distinguish between four homogeneous groups among 31 isolates (Table 3). Two homogeneous groups (first group: O-15 isolate; second group: other isolates) were distinguished based on differences in the MON level (Table 3)

Genetic variability determination
To assess a genetic variation among F. oxysporum isolates, genetic fingerprinting by RAPD analyses with nine arbitrary primers was carried out. Three of them were excluded from further analysis because of monomorphism or low repeatability. Finally, 33 polymorphic products were obtained, ranging from 4 to 7 per primer ( Table 4). The size of the RAPD fragments ranged from 0.15 to 3.2 kbp. The relationship among isolates of F. oxysporum is presented in the form of a dendrogram (Fig. 2). Clustering analysis resolved two groups at the 30% similarity level. Each of these groups was divided into smaller subgroups at 60 and 52% similarity levels, respectively. The highest genetic similarity (equal to 1) was observed between the isolates O-03      (Table 5) as well as one marker (OPC-15b) with MON production were found (Fig. 3). The percentage variation of FB 1 concentration accounted by markers ranged from 11.2% (for OPB-07b) to 26.5% (for OPL-11c). In the case of MON concentration, the percentage variation accounted by OPC-15b marker was equal to 56.8%, which makes this marker a good candidate for further studies.

Discussion
When considering Fusarium fungi as mycotoxin producers, a proper identification of species infecting food and/or feed plants is very important. Traditional diagnostics based on  Fig. 2 Dendrogram of F. oxysporum isolates based on random amplified polymorphic DNA (RAPD) data analysis; the isolates were grouped hierarchically using the unweighted pair group method of arithmetic means (UPGMA). *Isolates with MAT1-1 idiomorph morphology as well as simple PCR in some cases is insufficient. The translation factor 1-α gene has high phylogenetic utility because it is very informative at the species level and non-orthologous copies of them have not been detected in the Fusarium genus (Geiser et al. 2004). Based on species-specific PCR and partial sequencing of the TEF gene, we proved undoubtedly that isolates forming FB 1 belong to F. oxysporum species.
The present study showed explicitly that some of the F. oxysporum isolates infecting asparagus spears have the potential to form and accumulate FB 1 , although, in general, they produced low amounts of fumonisin and significantly higher amounts of MON. The ability of F. oxysporum to form fumonisin has been questioned previously. Wang et al. (2010) reported that F. oxysporum strains isolated from asparagus spears in Zhejiang Province of China did not produce FB 1 , but only four F. oxysporum isolates were examined. Similarly, it was shown that F. oxysporum isolated from greenhouse-grown jimsonweed produced only MON at a concentration level of 3.5 μg g −1 (Abbas et al. 1992). On the other hand, Abbas et al. (1995) claimed that three of six isolates of F. oxysporum and seven isolates of F. oxysporum var. redolens from root lesions of eastern white pine seedlings produced FB 1 , ranging from traces (≥100) to 300.60 ng g −1 .
This statement was also supported by our earlier studies (Waśkiewicz et al. 2009).
Fumonisin formation is positively related to the FUM genes expression Stępień et al. 2011a, b). The FUM genes cluster is highly collinear among F. oxysporum, F. proliferatum and F. verticillioides, regarding gene number, orientation and order (Proctor et al. 2008). The amount of FB 1 produced in vitro depends on the substrate, genotype and conditions (Desjardins et al. 1992). It was shown that there is a significant impact of interacting environmental factors on the FUM transcript levels (Jurado et al. 2008). The correlation between FB 1 and MON levels was not revealed, probably because of the different biosynthetic pathways of these secondary metabolites (Voss et al. 2007). Naturally occurring genetic variance in the fungal population may explain why non-producing isolates were also recorded. The loss of toxigenic ability can occur due to even a single-point mutation that shifts the open reading frame or creates a premature stop codon (Proctor et al. 2006).
The detection of MAT genes in the genomes of mitosporic species is a first step into learning what causes asexuality (Kück and Pöggeler 2009). The obtained results confirm that F. oxysporum has a heterothallic origin and are in agreement with Turgeon's hypothesis that some Fusarium species may  (Turgeon 1998). Similar results were reported by Arie et al. (2000) for F. oxysporum isolates and by Kerényi et al. (2004) for several Fusarium species. The amplification of the MAT idiomorph sequence revealed that 30% of the examined F. oxysporum isolates possess the MAT1-1 idiomorph and as many as 70% possess the MAT1-2 allele. Similar results for mating types distribution in the fungal population were described for F. oxysporum originated from common bean (Karimian et al. 2010). The maximum effective reproductive strategy occurs when the mating type idiomorphs are present in a 1:1 ratio (Britz et al. 1998). In many Fusarium species, the ratio of MAT idiomorphs was significantly different from the theoretical 1:1 ratio expected in an idealised population, i.e. in a population of F. verticillioides, the mating ratio was 2:1 for isolates originated from maize in Italy, Brazil and the Philippines (Cumagun 2007;de Oliveira Rocha et al. 2011;Venturini et al. 2011). The same mating ratio was observed for an F. subglutinans population in South Africa (Britz et al. 1998). Among F. culmorum isolates originated from wheat and rye, the mating ratio was estimated as 2:3 (Irzykowska and Kosiada 2011). The achieved results may suggest that MAT1-2 is a predominant mating type in the F. oxysporum population in Poland, although a larger population of fungus originated from different geographic locations should be analysed before reaching a final conclusion. The predominance of one mating type can cause a limitation or lack of the possibility of sexual reproduction inside the population (Venturini et al. 2011). RT-PCR analysis proved that the F. oxysporum MAT genes are expressed (Yun et al. 2000). Lack of sexual reproduction may also be a result of the functional disorder of the yet-unidentified genes that are involved in successful sexual reproduction . Alternatively, asexual Fusarium spp. may require environmental conditions for mating that are uncommon when disease epidemics occur (Kerényi et al. 2004). To summarise, the molecular discrimination of MAT1-1 and MAT1-2 in F. oxysporum populations make possible the recognition of potentially compatible isolates that could be used in future experiments revealing whether the species is truly asexual or not. The genetic variability of F. oxysporum isolates was estimated previously by different molecular methods. Genetic fingerprinting by RAPD-PCR is an effective method for determining an inter-and intra-species genetic variation without prior knowledge of the genome sequence. RAPD has been used successfully to analyse the genetic variation of several Fusarium species, including F. oxysporum (Assigbetse et al. 1994;Clark et al. 1998;Werner and Irzykowska 2007;Karimian et al. 2010), F. proliferatum (von Bargen et al. 2009), F. culmorum (Miedaner et al. 2004;Irzykowska and Baturo 2008), F. graminearum (Ouellet and Seifert 1993;Carter et al. 2002), F. moniliforme, F. solani and F. avenaceum (Khalil et al. 2003). In the present study, a significant genetic variation among F. oxysporum isolates was revealed and shown in the form of a dendrogram. Clustering analysis resolved several fingerprint groups, six of which contained one isolate. This is consistent with the results of another study on the genetic diversity of F. oxysporum, where from 6 to 10 fingerprint groups were found (Cramer et al. 2003;Zanotti et al. 2006;Karimian et al. 2010). The DNA sequences of the most unlike isolates (O-02 and O-12) differ in analysed genome parts by almost 70%. Isolate O-02 originated from green spears of asparagus 'Gijnlim', whereas isolate O-12 originated from white spears of asparagus 'Eposs', thus, it is possible that the revealed variation between them resulted from race existence. However, we have found five pairs of isolates which were identical in the examined genome parts. All of these isolates originated from white spears of asparagus 'Eposs'. Such high genetic similarity can also be the result of the relatively low number of markers used in the study.
It is well known that a high level of intra-species genetic diversity is typical for species undergoing sexual cycle with the meiotic recombination (McDonald 1997; Kerényi et al. 2004). The ways available for genetic change in F. oxysporum are still largely unknown, but some possibilities exist beyond simple sexual or clonal reproduction (Kistler 1997). Daboussi and Langin (1994) reported that active fungal transposable elements may comprise up to 5% of the F. oxysporum genome and may have an impact on gene structure and function. Other molecular studies suggest a genetic duplication in the rDNA regions (O'Donnell and Cigelnik 1997).
Moreover, in this study, nine molecular markers associated with MON or FB 1 formation levels were found (Table 4). The obtained PCR products will be used in future for the preliminary discrimination of toxigenic isolates. RAPD marker OPC-15b (∼1,150 bp), connected with MON biosynthesis, will be cloned and sequenced to converse them to the specific and more universal SCAR marker. Amplified regions of the F. oxysporum genome may be involved in the study of toxin biosynthesis pathways. A genetic background of toxin biosynthesis is complex and, probably, some genes still remain unknown. Toxicity as the majority of significant physiological traits (i.e. toxin or protein biosynthesis) is inherited quantitatively, so continuous variation arises from the segregation of alleles at many interacting loci, whose effects depend on the environment (Edwards et al. 1987;Irzykowska et al. 2001;Bocianowski and Krajewski 2009). Proctor et al. (2008) reported that the FUM cluster is located in different genome parts in F. oxysporum, F. proliferatum and F. verticillioides. Perhaps the genetic bases of FB 1 formation by F. oxysporum strains have not yet been fully elucidated.
In conclusion, we postulate that F. oxysporum present in asparagus spears produces not only MON but also highly toxic FB 1 , which can be a significant concern for consumers. The mechanism of genetic changes occurring in F. oxysporum populations and the connection between the toxin profile and the genetic variability of fungus demand further attention. More extensive studies are necessary in order to elucidate the distribution of MATs in the fungus population.