Transmission of Fusarium boothii Mycovirus via Protoplast Fusion Causes Hypovirulence in Other Phytopathogenic Fungi

There is increasing concern regarding the use of fungicides to control plant diseases, whereby interest has increased in the biological control of phytopathogenic fungi by the application of hypovirulent mycoviruses as a possible alternative to fungicides. Transmission of hypovirulence-associated double-stranded RNA (dsRNA) viruses between mycelia, however, is prevented by the vegetative incompatibility barrier that often exists between different species or strains of filamentous fungi. We determined whether protoplast fusion could be used to transmit FgV1-DK21 virus, which is associated with hypovirulence on F. boothii (formerly F. graminearum strain DK21), to F. graminearum, F. asiaticum, F. oxysporum f. sp. lycopersici, and Cryphonectria parasitica. Relative to virus-free strains, the FgV1-DK21 recipient strains had reduced growth rates, altered pigmentation, and reduced virulence. These results indicate that protoplast fusion can be used to introduce FgV1-DK21 dsRNA into other Fusarium species and into C. parasitica and that FgV1-DK21 can be used as a hypovirulence factor and thus as a biological control agent.


Introduction
Although fungicides successfully control many diseases caused by plant-pathogenic fungi, fungal pathogens remain a major source of plant disease. Because of the development of fungicideresistant strains, and increasing public concern regarding environmental and food safety, there is renewed interest in biological control based on application of hypovirulent mycoviruses.
The potential of mycoviruses for managing plant-pathogenic fungi was first demonstrated for Cryphonectria parasitica [1]. The success of biological control with hypoviruses depends on their ability to reduce the virulence (to induce hypovirulence) of the target fungus. Hypoviruses can be transmitted from a hypovirulent strain to a virulent fungal strain by hyphal fusion (anastomosis) when the two strains are vegetatively compatible, but hypoviruses cannot be transmitted when applied by extracellular routes [2,3]. Because only closely related fungal strains are vegetatively compatible, vegetative incompatibility among many fungal species in agricultural ecosystems is a major barrier to the use of hypoviruses as biological control agents [4,5].
Double-stranded RNA mycoviruses have been described in yeasts, mushrooms, and filamentous fungi [6,7,8]. They are classified into five families based on virus structure and genome composition [9], but some are still unassigned to a genus or in some cases to a family. There is increasing evidence that mycoviruses reduce the growth and pathogenicity of fungal plant pathogens. As noted above, a virulence-attenuating dsRNA molecule has been described in C. parasitica, and five related mycoviruses have been completely sequenced [9]. Among them, Cryphonectria hypovirus 1 (CHV1) was successfully used as a biological control agent of C. parasitica in Europe, i.e., CHV1infected strains exhibited reduced virulence, reduced asexual and sexual sporulation, and reduced pigment production. CHV1 was unsuccessful as a biological control agent in North America, however, because the host fungus in North America has multiple vegetative compatibility groups (VCGs) that limit the spread of the virus [10].
The failure of mycovirus transmission caused by vegetative incompatibility can be overcome in the laboratory by using protoplast fusion [11]. Transmission of dsRNA mycoviruses via protoplast fusion has been reported in plant-pathogenic fungi including Aspergillus [12], F. poae [13], and Rosellinia necatrix [14].
We previously isolated the FgV1-DK21 virus from strain DK21 [15]. According to genealogical concordance phylogenetic species recognition (GCPSR), the F. graminearum species complex (Fg complex) comprises 13 phylogenetically distinct species based on DNA sequences from 13 independent genetic loci [16,17]. Strain DK21 was evaluated by GCPSR using DNA sequences from selected genes and it was identified as F. boothii ( Figure S1). FgV1-DK21 reduces the mycelial growth of F. boothii, increases its pigmentation, and reduces its virulence on wheat [15]. The 6,621 nucleotide-coding strand is polyadenylated and contains four open reading frames (ORFs 1 to 4) [18]. Pairwise sequence comparisons of the nucleotide and deduced amino acid sequences of ORFs 2 through 4 revealed no close relationships to other protein sequences currently available in GenBank while a phylogenetic analysis of the deduced amino acid sequence of ORF1, which encodes a putative RNA-dependent RNA polymerase (RdRp), and those of other mycoviruses revealed that this organism forms a distinct virus clade with some hypoviruses and is more distantly related to other mycoviruses [18]. While FgV1-DK21 does not encode a coat protein, the genome organization and accumulation of at least two subgenomic RNAs (sgRNAs) indicate that FgV1-DK21 belongs to a new, as yet unassigned genus of mycoviruses [18].
In this report, we present evidence that protoplast fusion can be used to expand hypovirus host range and to study hypovirusmediated alterations in new fungal hosts. The results of this study indicate that protoplast fusion can overcome the barriers to transmission caused by genetic diversity and multiple VCGs and thus will extend persistent and transmissible system with application of FgV1-DK21 for fungal disease control.

Results
Effect of FgV1-DK21 dsRNA on colony morphology and mycelial growth We first determined whether FgV1-DK21 can overcome the VCG barrier in other Fusarium species. One strain each of two species within the Fg complex [F. asiaticum [19] and F. graminearum [20]], and one strain of F. oxysporum f. sp. lycopersici (outgroup) were chosen (Table 1). To improve screening efficiency of fused protoplasts, virus-free recipients and the virus-infected donors were transformed with hygromycin B-and geneticin-resistance genes, respectively (Figures 1 and 2). After equal volumes of the two protoplast suspensions (1610 6 protoplasts/ml) were fused by the 60% polyethylene glycol (PEG 3350)-mediated method, virusinfected strains were finally selected on hygromycin B-containing PDA (see Materials and Methods). Several virus-infected strains (2, 8, and 9) were selected for F. asiaticum, F. graminearum, and F. oxysporum f. sp. lycopersici, respectively (data not shown).
The phenotypic changes of virus-infected strains of F. asiaticum and F. graminearum were similar to those of strain DK21. Like strain DK21, FgV1-DK21 recipient strains of F. asiaticum and F. graminearum had reduced growth rates and increased pigmentation relative to virus-free strains ( Figure 3A). In contrast, only slight morphological alterations were evident in the virus-infected F. oxysporum f. sp. lycopersici strain when growing on PDA ( Figure 3A). However, the FgV1-DK21-infected strains of Fusarium species produced less aerial hyphae than the virus-free strains. FgV1-DK21 dsRNAs were detected in F. graminearum and F. asaticum strains, but not in F. oxysporum f. sp. lycopersici when extracted total RNAs were separated on agarose gel ( Figure S2A). PCR amplified much more viral dsRNA in the virus-infected strains, F. asiaticum and F. graminearum, than in the virus-infected strain of F. oxysporum f. sp. lycopersici ( Figures 4B and S2C). We also sequenced DNA from portions of translation elongation factor 1a (TEF) gene and/ or histone H3 gene to determine whether dsRNA of strain DK21 was transferred into the desired recipient strain. The TEF and histone H3 genes have been used as phylogenetic markers to  investigate species limits in Fusarium [16,21]. As a consequence, the fixed nucleotide characters found in the virus-free strains of F. asiaticum (Histone H3 position 278; G) and F. graminearum (Histone H3 position 279; T) were also present in each virus-infected strain ( Figure 4). Although the results indicate that dsRNA of strain DK21 was transferred into the recipient strains, it is unclear whether the altered phenotypes of recipient strains were the result of virus transmission or protoplast fusion because the recipient strains were different in terms of their morphology and in pathogenicity [22]. To address this concern, we analyzed DNA polymorphism between uninfected and virus-infected strains using amplified fragment length polymorphism (AFLP) profiling. Because the AFLP technique is based on the selective PCR amplification of restriction fragments from a total digest of genomic DNA, it will generate fingerprints of any DNA regardless of the origin or complexity and thus reflect true DNA polymorphisms [23]. The genomic DNAs from virus-free and virus-infected strains were digested with EcoR I and Mse I for AFLP analysis (Text S1). The digested DNAs were ligated with two adapters and amplified by PCR using specific oligonucleotide primers. Identical AFLP profiles were observed when we compared the DNA fingerprints among both uninfected and virus-infected samples ( Figure S3A), indicating that the recipient strains screened from fused protoplasts were not substantially affected by protoplast fusion.

Hypovirulence of FgV1-DK21 in other Fusarium species
Based on the previous observation that the virulence of strain DK21 was significantly lower than that of the virus-free strain [15], we hypothesized that FgV1-DK21 dsRNA might also contribute to the hypovirulence in other Fusarium species. To explore this possibility, wheat head florets were inoculated with conidial suspensions of virus-free or virus-infected strains of F. asiaticum and F. graminearum at early-mid anthesis. Head blight was more severe on wheat plants inoculated with virus-free strains than with virus-infected strains of F. asiaticum and F. graminearum ( Figure 5).
For virulence assays with F. oxysporum f. sp. lycopersici, tomato seedlings growing in pots and at the four-leaf stage were inoculated with virus-free and virus-infected F. oxysporum f. sp. lycopersici strains by the root-dip method. At 3 weeks post-inoculation, seedlings were removed from the pots and their roots were observed for symptoms. Fusarium wilt had developed to the stem base in symptomatic seedlings, and the virus-infected strains were less virulent than the virus-free strains ( Figure 6A). At 4 weeks post inoculation, 46 of 60 plants (76.7%) inoculated with the virus-free strains were dead and 43 of 60 plants (71.7%) inoculated with virus-infected strains remained alive ( Figure 6B).
Transmission of FgV1-DK21 dsRNA from strain DK21 to C. parasitica We also tested whether protoplast fusion can be used to introduce FgV1-DK21 dsRNA into a filamentous fungus of a different genus. Cryphonectria parasitica and associated mycoviruses provide a good model for studying virus/virus and virus/host interactions. For this reason, C. parasitica was subjected to protoplast fusion and evaluated as a potential host of FgV1-DK21. Cryphonectria parasitica strain EP155 was transformed with the hygromycin B resistance gene and fused as a recipient strain with strain DK21 (virus donor) by protoplast fusion. Four strains produced by the fusion procedure were selected and compared with virus-free EP155 and CHV1-infected EP155. CHV1-infected colonies (UEP) were smaller than virus-free colonies and lacked the orange pigment of virus-free colonies ( Figure 7A). Colonies infected by FgV1-DK21 retained the orange color but were much smaller than virus-free EP155 or CHV1-infected EP155 colonies ( Figure 7A). FgV1-DK21 was detected in virus-infected colonies by RT-PCR ( Figures 7B and S2C). We also identified parallel bands among uninfected or virus-infected strains from AFLP profiling ( Figure S3B) indicating that the uninfected and virusinfected strains of EP155 had not been altered significantly by the fusion process. In a virulence test with apples, the areas of lesions caused by virus-free EP155, the CHV1-infected strain, and FgV1-DK21-infected strains were approximately 10, 4, and 0.5 cm 2 , respectively ( Figure 8A and B).

Discussion
Here, we demonstrate that the FgV1-DK21 functions as a hypovirulence factor, thereby leading to morphological changes and hypovirulence. Our studies also present evidence that the protoplast fusion system can be used as a means for studying hypovirus-mediated alterations and strain development potential. Protoplast fusion technology has frequently been used for the genetic manipulation of fungi (for establishing heterokaryons) when few molecular genetic tools are available [24,25]. By facilitating the discovery of new intermediates and hybrid antibiotics with beneficial properties [26], the large-scale enhancement of metabolite yields [27], and the construction of starchutilizing strains [28], protoplast fusion has shown great potential for industrial applications. The present study, however, required that the genetic background of the recipient isolates be maintained when protoplast fusion was used to transfer hypovirulent dsRNA mycovirus to the recipients. Protoplast fusion with specific selection marker(s) has proven to be a novel approach by which potential strains with desirable properties could be obtained with minimal disturbance to their genetic background and physiology [29]. The genetic variability of recipient isolates was studied by sequencing the TEF and histone H3 gene region and by AFLP analysis. Banding patterns from the PCR-based AFLP method using total DNA from the dsRNA-free and dsRNA-containing recipient isolates were almost identical in most, if not all, tested samples ( Figure S3). These observations suggest that the genetic background of the fungi suffered minimal disturbance from the fusion procedure and further confirmed the successful transmission of a dsRNA between the donor and the recipient isolates. Altogether, the present results demonstrate that protoplast fusion can be used to transmit FgV1-DK21 dsRNA both interspecifically and intergenerically.
Virulence-attenuating mycoviruses have been described in many plant-pathogenic fungi [9], however, few successful applications have been reported. Given that these mycoviruses are primarily transmitted horizontally via hyphal anastomosis or vertically from mycelium to spores, vegetative incompatibility within fungal species and the varying degree of vertical transmission efficiencies [10] are major barriers to their use as biological control agents. Consequently, the use of hypoviruses to control fungal diseases is restricted more by their limited natural transmission and/or lack of interspecies transmission than the availability of hypovirulence-associated mycoviruses. The development of efficient transmission and delivery methods might accelerate the use of hypovirulence-associated mycoviruses as biological control agents. In this study, transmitted virus (FgV1-DK21) replicated in the new hosts, though protoplast regeneration capacity was different between the strains tested. It is worth noting that hypovirulence induced in C. parasitica and measured in apple inoculations was greater with FgV1-DK21 dsRNA than with CHV1 ( Figure 7A). The mechanisms by which the FgV1-DK21 causes hypovirulence to their hosts are unknown. How the FgV1-DK21 affects fungal physiology and virulence will have implications for other phytopathogenic fungi. For example, putative transcription factor PRO1 is down-regulated by C. parasitica strains infected with different hypoviruses and required for female fertility, asexual spore development, and stable maintenance of viral infection [30]. In light of these, comparative analysis of background-or mycovirus-related transcriptome changes will aid the better understanding of mycovirus-fungal interactions.
Although the present study clearly demonstrated that the FgV1-DK21 dsRNA in new hosts can be transmitted to new species via protoplast fusion, it is necessary to test whether virus infected stains can transfer the FgV1-DK21 dsRNA to the same virulent species under field conditions. Therefore, an integrative knowledge including the effect of the timing and rate of application for practical application of the FgV1-DK21 dsRNA is required to establish efficacy and consistency of biological control. In addition, the success of biological control using hypovirulent mycovirus requires a sufficient understanding of the replication mechanism of the mycovirus and also of the interactions between the virus and host fungus, between the virus-fungus and plant, and between the virus-fungus-plant and the rest of the agro-ecosystem. Therefore, efficient and/or practical application of hypovirulent mycoviruses as biological control agents will require more research to elucidate their modes of action at the molecular level and to characterize their ecological fitness.

Fungal strains and culture conditions
All strains used in this study (Table 1) were stored in 25% (v/v) glycerol at 280uC and were reactivated on potato dextrose agar (PDA; Difco). For total RNA extraction, strains of the Fg complex were grown in 50 ml of liquid complete medium (CM) at 25uC at 150 r.p.m. for 5 days while strains of C. parasitica were grown in 50 ml of EP complete medium [31] at 26uC and 120 r.p.m. for 5 days. Mycelia were harvested by filtration through Miracloth (Calbiochem) and ground to a fine powder with a mortar and pestle in liquid nitrogen.

Construction of antibiotic resistant mutants
Protoplasts of fungal strains were prepared by treatment of fresh mycelia grown on YPG liquid medium (0.3% yeast extract, 1% peptone, 2% glucose) for 3 h at 30uC with 1 M NH 4 Cl containing 10 mg/ml of driselase (InterSpex Products), as described previously [32]. Plasmid DNA (20 mg) was directly added along with 1 ml of PEG solution (60% polyethylene glycol 3350, 10 mM Tris-HCl pH 7.5, 10 mM CaCl 2 ) to protoplast suspensions. Transformants with resistance to hygromycin B were obtained by transforming the fungal protoplasts with the plasmid pUCH1 [33] and selected for on regeneration medium containing 80 mg/ml of hygromycin B (Calbiochem). For construction of the geneticin-resistant mutant, the plasmid pII99 [34] was transformed into protoplasts of virus-free F. boothii. Following the transformation, FgV1-DK21 was transmitted by anastomosis and screened on PDA containing 50 mg/ml of geneticin (Duchefa). For genomic DNA extraction, fungal strains of the Fg complex were grown in 50 ml of CM at 25uC, 150 r.p.m. for 5 days. The mycelia were harvested by filtration through sterile Whatman no. 2 filter paper, ground in liquid nitrogen using a mortar and pestle, and then suspended in CTAB buffer [2% CTAB (cetyltrimethyl ammonium bromide), 20 mM EDTA, 0.1 M Tris-HCl, and 1.4 M NaCl]:2-mercaptoethanol (100:1). Genomic DNA was extracted sequentially with chloroform:isoamyl alcohol (24:1), precipitated with isopropanol. The extracted genomic DNA was extracted twice with phenol:chloroform:isoamyl alcohol (25:24:1), treated with RNase A (20 mg/ml) for 1 h at 37uC, precipitated with isopropanol, and then finally suspended in distilled water. For the Southern hybridization of hygromycin B-resistant mutants, the extracted genomic DNA was digested with Kpn I for 12 h, and the Southern hybridization of G418resistant mutants, it was digested with Spe I and Kpn I for 12 h at 37uC. A 10-ml of the digested DNA was separated on 0.8% agarose gel for 8 h. The gel was submerged twice in denaturation solution (1.5 M NaCl and 0.5 N NaOH) for 20 min at room temperature and capillary blotted onto a positively charged nylon transfer membrane (GE Healthcare) in 0.4 N NaOH. Probe labeling reactions were performed in 20 ml of 10 mM Tris-HCl pH 7.5, 7 mM MgCl 2 , 0.1 mM DTT,  30 mCi [a-32 P] dCTP, 3 mM dNTP mix, 10 pmoles of random primers and 2 U klenow fragment (TaKaRa). After hybridization, unhybridized probe is removed by washing in low stringency wash buffer (26SSC and 0.1% SDS) and high stringency wash buffer (0.16SSC and 0.1% SDS). Hybridization signal intensities were measured using a Bio-imaging Analyzer system (BAS-2500; Fuji Film).

Polymerase chain reaction (PCR) and nucleotide sequencing
PCR of TEF and histone H3 gene region was performed as described with modification [16,21,35] using the following conditions: one step at 94uC for 3 min; 35 cycles at 93uC for 45 sec, 55uC for 40 sec, and 72uC for 1 min; and finally one step at 72uC for 10 min. PCR products amplified from fungal strains of the Fg complex were extracted from an agarose gel with QIAquickH gel extraction kit (Qiagen) by following the manufacturer's instructions. DNA sequencing was performed at the National Instrumentation Center for Environmental Management of the Seoul National University with an ABI Prism 3730 XL DNA Analyzer (Applied Biosystems) according to manufacturer's instructions. The sequence data were analyzed using a BLAST search tool and were aligned using Clustal W [36].

Protoplast fusion
Protoplast fusion was performed according to a previously described method with modifications [37]. Young mycelia were prepared as described previously [32,38] and incubated for 3 h at 30uC with 1 M NH 4 Cl containing 5 mg/ml of driselase and 8 mg/ml of lysing enzyme (L1412; Sigma). Protoplasts were harvested by centrifugation at 2,5446g at 4uC for 10 min, washed twice with STC (1.2 M Sorbitol, 10 mM Tris-HCl pH 7.5, 50 mM CaCl 2 ), and suspended in 300 ml of MMC buffer (0.6 M Mannitol, 10 mM MOPS pH 7.0, and 10 mM CaCl 2 ). Equal volumes of the two protoplast suspensions (100 ml of 1610 7 protoplasts/ml) were mixed and placed on ice for 30 min. After 500 ml of PEG solution (60% PEG 3350, 10 mM MOPS pH 7.0, and 10 mM CaCl 2 ) was added to the protoplast suspension, the mixture was incubated at 20uC for 20 min. Protoplast fusants were regenerated in 700 ml of potato dextrose broth (PDB; Difco) for 7 days in the dark, plated on 15 ml of YCDA (0.1% yeast extract, 0.1% casein hydrolysate, 0.5% glucose, and 1.5% agar), and then selected on PDA containing 50 mg/ml of hygromycin B and 50 mg/ml of geneticin. Antibiotic-resistant colonies were screened again on hygromycin B-containing PDA.

RNA extraction and RT-PCR
Total RNA was isolated with extraction buffer according to a previously described method [39] and further treated with DNase I (TaKaRa) to remove genomic DNA. The samples were precipitated with ethanol and finally suspended in DEPC-treated water. To detect viral dsRNA in virus-infected colonies, cDNAs were synthesized with M-MLV reverse transcriptase (Promega) and oligo d(T) primer. The resulting cDNAs (20 ng of input RNA) were used to detect FgV1-DK21 (using primer pairs 59-TGTGGGAGAAGAAGTAT-GGCCT-39 and 59-ATCAGGAACCATTGAAAGAGTCC-39 (RdRp region) or 59-ATGGACACCAAGGATATTTA-39 and 59-TTAGGGGTGCAAGGCCCTTTTC-39 (ORF2 region)). PCR reactions were performed using the following conditions: one step at 94uC for 3 min; 35 cycles at 93uC for 45 sec, 60uC for 40 sec, and 72uC for 1 min 30 sec; and finally one step at 72uC for 10 min. PCR products were analyzed by 1% agarose gel electrophoresis.

Virulence assays
Virulence assays with F. boothii, F. asiaticum, and F. graminearum were performed as described [40] on wheat cv. Jokyoung. The plants were approximately 6 weeks old and had flowering heads. For production of conidial inoculum, five mycelial plugs were incubated in CMC liquid medium (1.5% carboxymethyl cellulose, 0.1% yeast extract, 0.05% MgSO 4 N7H 2 O, 0.1% NH 4 NO 3 , and 0.1% KH 2 PO 4 ) at 25uC and 150 r.p.m. for 5 to 7 days. Conidia were collected by filtering through six layers of sterile cheese cloth. A 10-ml volume of the spore suspension (10 5 conidia/ml) in 0.01% (v/v) Tween-20 was injected into one floret of each flowering wheat head. Wheat plants inoculated with 0.01% (v/v) Tween-20 alone served as a control. For each treatment, 10 replicate wheat heads were inoculated. Inoculated plants were placed in a growth chamber (25uC, 80% relative humidity, 14/10 h light/dark cycle). Wheat heads were examined for symptoms 14 days postinoculation.
Virulence of F. oxysporum f. sp. lycopersici strains was measured with a Fusarium wilt assay as described previously [41]. Ten-day-old tomato seedlings in the four-leaf stage were inoculated by dipping the roots for 3 min in a suspension containing 10 5 microconidia/ml of the F. oxysporum f. sp. lycopersici strains in distilled water. Twenty seedlings per treatment were planted in pots containing sterile soil and maintained in a growth chamber at 28uC with 14/10 h light/ dark cycle. Severity of disease symptoms was calculated using an index from 0 (healthy plant) to 4 (dead plant).
Virulence of C. parasitica strains was assayed as described previously with minor modifications [42]. Mycelial plugs were prepared from the edge of 7-day-old colonies on PDA. Apple tissues (5 mm diameter65 mm deep) were removed and the insides were filled with mycelial plugs. Following inoculation, they were sealed with plastic wrap to maintain humidity and incubated at 25uC with a 12/12 h light/dark cycle. The discolored area was measured at 14 days post-inoculation. All virulence assays were repeated three times. Statistical analysis was performed with the PASW statistics software (SPSS Inc.). Text S1 AFLP fingerprints of genomic DNAs of virusfree and virus-infected strains.

Supporting Information
(DOC)