A Host Factor Involved in Hypovirus Symptom Expression in the

JOURNAL OF VIROLOGY, Jan. 2008, p. 740–754 Vol. 82, No. 20022-538X/08/$08.00�0 doi:10.1128/JVI.02015-07Copyright © 2008, American Society for Microbiology. All Rights Reserved.

A Host Factor Involved in Hypovirus Symptom Expression in theChestnut Blight Fungus, Cryphonectria parasitica�†

M. Iqbal Faruk, Ana Eusebio-Cope, and Nobuhiro Suzuki*Agrivirology Laboratory, Research Institute for Bioresources, Okayama University, Kurashiki, Okayama 710-0046, Japan

Received 12 September 2007/Accepted 22 October 2007

The prototype hypovirus CHV1-EP713 causes virulence attenuation and severe suppression of asexualsporulation and pigmentation in its host, the chestnut blight fungus, Cryphonectria parasitica. We identified afactor associated with symptom induction in C. parasitica using a transformation of C. parasitica strain EP155with a full-length cDNA clone from a mild mutant virus strain, Cys(72). This was accomplished by usingmutagenesis of the transformant fungal strain TCys(72)-1 by random integration of plasmid pHygR, confer-ring hygromycin resistance. The mutant, namA (after nami-gata, meaning wave shaped), showed an irregularfungal morphology with reduced conidiation and pigmentation while retaining similar levels of virulence andvirus accumulation relative to TCys(72)-1- or Cys(72)-infected strain EP155. However, the colony morphologyof virus-cured namA (VC-namA) was indistinguishable from those of EP155 and virus-cured TCys(72)-1[VC-TCys(72)-1]. The phenotypic difference between VC-namA and VC-TCys(72)-1 was found only when thesestrains infected with the wild type or certain mutant CHV1-EP713 strains but not when infected with My-coreovirus 1. Sequence analysis of inverse-PCR-amplified genomic DNA fragments and cDNA identified theinsertion site of the mutagenic plasmid in exon 8 of the nam-1 gene. NAM-1, comprising 1,257 amino acids,shows sequence similarities to counterparts from other filamentous fungi and possesses the CorA domain thatis conserved in a class of Mg2� transporters from prokaryotes and eukaryotes. Complementation assays usingthe wild-type and mutant alleles and targeted disruption of nam-1 showed that nam-1 with an extension of thepHygR-derived sequence contributed to the altered phenotype in the namA mutant. The molecular mechanismunderlying virus-specific fungal symptom modulation in VC-namA is discussed.

Induction of symptoms by a virus in its host is a consequenceof alterations in complex physiological processes that involveinteractions between host and viral factors at the molecularlevel. Thus, it is difficult to investigate the molecular mecha-nism underlying macroscopic phenotypic alterations. This isparticularly true for mycoviruses/filamentous fungus systems,because genetic manipulation is often unavailable for hostfungi and/or viruses, and even virus introduction into host cellsis not always easy to perform. In this regard, the mycovirusesand Cryphonectria parasitica are placed in a unique position.

C. parasitica is an ascomycetous fungus that is the causalpathogen of chestnut blight disease. This fungus is the host ofa number of viruses, and some confer hypovirulence to the hostfungus, thus being of potential or practical use as biologicalcontrol agents (2, 19). Efficient DNA-mediated transformationis available for C. parasitica (9), which allows multiple trans-genic expressions of endogenous and exogenous genes andtargeted gene disruption by homologous recombination-medi-ated gene replacement (11, 32). Infectious cDNA clones areavailable for a few members of the family Hypoviridae (5, 8,23), while a transfection protocol with purified virus particleshas been established for members of the genus Mycoreovirus

(18). These established techniques provide a foundation tostudy fungal host-virus interactions.

The prototype hypovirus CHV1-EP713 moderately attenu-ates virulence (hypovirulence) of the chestnut blight fungus.Hypovirulence-associated phenotypic traits include severe re-duction in pigmentation, asexual sporulation, and loss of fe-male fertility. In this virus-host combination, viral symptomfactors have been studied using transformation and transfec-tion analyses. The viral double-stranded RNA genome ofCHV1-EP713 is 12.7 kb in size with two continuous openreading frames (ORFs), ORF A and ORF B (11). The ORFA-encoded polyprotein p69 is cleaved into a papain-like pro-tease, p29, and a basic protein, p40. The p29 protein, associ-ated with membranous vesicles derived from the trans-Golginetwork (20), plays multifunctional roles; that is, p29 containsan essential domain for virus viability (38), contributes to areduction in pigmentation and sporulation (7, 10, 34, 35, 37),suppresses RNA silencing (31), enhances the replication of thehomologous and heterologous viruses (34, 37), and is respon-sible for self-cleavage (6). The symptom determinant and rep-lication-enhancing activity domain resides in the N-terminalregion containing critical cysteine residues at positions 70 and72. The substitution of a glycine for Cys70 and Cys72 in thecontext of full-length virus cDNA results in mutant virusesCys(70) and Cys(72) that induce profoundly altered pheno-types and symptoms similar to those of a p29 null mutant �p29virus (35). Transgenic expression of wild-type p29 in the ab-sence of virus replication results in reductions in pigmentationand sporulation, while mutation at Cys70 or Cys72 abolishes itssuppressive activities (37). The other ORF A-encoded protein,p40, is also involved in virus RNA accumulation and symptom

* Corresponding author. Mailing address: Agrivirology Laboratory,Research Institute for Bioresources, Okayama University, Kurashiki,Okayama 710-0046, Japan. Phone: 81(86) 434-1230. Fax: 81(86) 434-1232. E-mail: [email protected].

† Supplemental material for this article may be found at http://jvi.asm.org/.

� Published ahead of print on 31 October 2007.

740

on April 8, 2018 by guest

http://jvi.asm.org/

Dow

nloaded from

http://jvi.asm.org/

induction only when provided in cis (from the CHV1-EP713genome) (36).

Perturbation of signaling pathways, e.g., the heterotrimericG protein pathway, is implicated in symptom expression byCHV1-EP713. Support for this comes from the pleiotropiceffects of CHV1 infection on the fundamental physiology inthe host fungus and similarities in transcription profiles be-tween fungal strains infected with the virus and defective in asignaling pathway. Using a catalogued expressed-sequence-tag(EST) library (13), Allen et al. (1) previously performed mi-croarray analysis of CHV1-EP713-infected and virus-free fun-gal strains to identify host genes that are responsive to infec-tion by the virus. Those studies showed that the macroscopicchanges induced by infection with CHV1-EP713 are accompa-nied by either the up- or down-regulation of at least 295 hostgenes (13.4% of 2,200 genes tested). Many of the responsivegenes are affected by the disruption of the G protein subunitsin the same direction (12). Furthermore, the disruption of a C.parasitica homologue of the Ste12 transcription factor that isdown-regulated by CHV1-EP713 infection results in virulenceattenuation and female sterility (14). Ste12 is a part of themitogen-activated protein kinase signaling cascade regulatingmating and pseudohyphal growth in Saccharomyces cerevisiae.

Here, we describe the development of a genetic screeningprotocol entailing the transformation of a wild-type funguswith a cDNA clone of the CHV1-EP713 variant Cys(72), mu-tagenesis by random plasmid integration, isolation of mutantstrains, and identification of mutated host genes by inversePCR (I-PCR). We report the characterization of a mutant C.parasitica strain, termed namA (nami-gata, meaning waveshaped), showing unusually altered symptoms upon infectionwith the hypovirus, but not with a distinct mycoreovirus, com-pared to its parental fungal strain. Furthermore, we character-ized a mutated host gene encoding a CorA Mg2� transporterdomain, which contributed to the mutant phenotype.

MATERIALS AND METHODS

Virus strains. Table 1 shows the virus strains used: the prototype hypovirusCHV1-EP713 (28); its deletion mutants �p29, �p40b, and �p69b (36); and thesite-directed mutant Cys(72) (35). The type member of the genus Mycoreovirus,MyRV1-Cp9B21, was described previously (19, 39).

Fungal strains and culturing. The field fungal isolate EP155 (ATCC 38755)and transformants with the EP155 background were infected or uninfected withviruses (Table 1). Fungal colonies were cultured at 25°C to 27°C for 5 to 7 daysin potato dextrose broth (PDB) (Difco) for RNA/DNA extraction and sphero-plast preparations or on potato dextrose agar (PDA) (Difco) for phenotypicobservation. Fungal strains were cultured on regeneration plates (9) for main-tenance and stored at 4°C in a refrigerator until use.

Transfection and transformation of C. parasitica spheroplasts. Spheroplastswere prepared from fungal strains with the EP155 backgrounds that were cul-tured in PDB as described previously by Churchill et al. (9) and transformed withfull-length cDNA clones of Cys(72) or genomic DNA fragments as describedpreviously by Choi and Nuss (8). Transfection of EP155 with in vitro-synthesizedRNA from cDNA of CHV1-EP713 and its mutant strains or purified virusparticles of MyRV1 was carried out according to methods described previouslyby Suzuki et al. (35) and Hillman et al. (18), respectively.

Phenotypic measurements of fungal colonies. Virulence was measured inapples (cv. Jonathan Gold) as areas of lesions induced by fungal strains asdescribed previously by Fulbright (15) and Hillman et al. (17). Sporulation andpigmentation were evaluated using methods described previously by Hillman etal. (17, 18). Fungal strains were cultured for 14 days on PDA plates that were 60mm in diameter: the first 4 days on the bench top and the last 10 days undermoderate light of approximately 3,000 lx using a 16-h photoperiod. Producedspores were liberated in 6 ml of 0.15% Tween 20 and filtered through a doublelayer of Miracloth (Calbiochem). Conidia were counted with the aid of a hemo-cytometer. To measure pigmentation, strains were cultured in 20 ml PDB for 12days and harvested onto Miracloth. The mycelia (300 mg) were homogenizedusing a mortar and pestle under liquid nitrogen. After the addition of 3 mlethanol, the absorbance at 454 nm was determined using an Ultrospec 2000spectrophotometer (Pharmacia Biotech) (18). Alternatively, pigment productionon PDA cultures was visually estimated.

Mutagenesis by random plasmid integration of a strain transformed with thevirus cDNA clone. A full-length cDNA clone of CHV1-EP713 mutant strainCys(72) that had a Cys-to-Gly substitution in p29 at Cys72 and that causedphenotypic alterations similar to those induced by �p29 in EP155 when trans-fected was described previously by Suzuki et al. (35). After being liberated fromthe pRFL4-based plasmid by NotI, Cys(72) cDNA was transferred into a trans-formation vector, pCPXBn1 (14), with the benomyl-resistant gene as the select-

TABLE 1. Viral and fungal strains

Strain Description Reference or source

ViralCHV1-EP713 Prototype of the family Hypoviridae 7Cys(72) CHV1-EP713 with a Cys-Gly substitution at p29 residue 72 35�p29 CHV1-EP713 lacking 87.5% of the p29 coding domain 35�p40b CHV1-EP713 lacking 99.5% of the p40 coding domain 36�p69b CHV1-EP713 lacking 96.1% of the p69 coding domain 36MyRV1-Cp9B21 Prototype of the genus Mycoreovirus 18

FungalEP155 Virus-free field isolate ATCC 38755TCys(72)-1 Transformant of EP155 with full-length cDNA of Cys(72) This studyVC-TCys(72)-1 Cys(72) virus-cured strain of TCys(72)-1 This studynamA Mutant of TCys(72)-1 obtained by random plasmid integration This studyVC-namA Cys(72) virus-cured strain of namA This studyVC-namA/p29 Transformant of VC-namA with the p29 coding sequence This studyVC-namA/p40 Transformant of VC-namA with the p40 coding sequence This studynamA/namC1 Transformant of namA with complementing plasmid pBSDnamC1 This studynamA/namC2 Transformant of namA with complementing plasmid pBSDnamC2 This studynamA/namC3 Transformant of namA with complementing plasmid pBSDnamC3 This studyTCys(72)-1/�nam1-1�TCys(72)-

1/�nam1-2nam-1 disruptant of TCys(72)-1 This study

VC-TCys(72)-1/�nam1-1�VC-TCys(72)-1/�nam1-2

Cys(72)-cured strains of TCys(72)-1/�nam1-1 and TCys(72)-1/�nam1-2 This study

TCys(72)-1/�nam1/nam1-HygR1 Transformant of TCys(72)-1/�nam1-1 with a nam-1 mutant allele of namA This study

VOL. 82, 2008 HOST FACTOR ASSOCIATED WITH HYPOVIRUS SYMPTOMS 741


http://jvi.asm.org/

Dow

nloaded from

http://jvi.asm.org/

able marker to form pBn1-Cys(72). Recombinant DNA procedures were per-formed according to methods described previously by Sambrook and Russell(30).

One of the transformant strains with pBn1-Cys(72), TCys(72)-1, was mu-tagenized by random insertion with a plasmid (pHygR) that contained the hy-gromycin B phosphotransferase gene (hph) controlled by the Aspergillus nidulanstrpC promoter (Ptrp) and terminator (Ttrp) (10) at the EcoRV site of pBluescriptSK(�) (Stratagene) (sequence is available upon request). Spheroplasts ofTCys(72)-1 were prepared as described above, retransformed with pHygR, andsubjected to single conidial isolation to obtain homokaryons with insertion mu-tations. The resulting transformants were screened for mutants with phenotypesthat were different from that of the parental strain TCys(72)-1. This screeningprocedure was based on the visual estimation of fungal phenotypes and compar-ison with TCys(72)-1 cultured on PDA plates in parallel.

The genetic stabilities of the phenotypes of TCys(72)-1 and mutated fungalstrains were examined by comparing the colony morphologies and drug resis-tances of approximately 50 single conidial isolates and isolates passaged fourtimes to those of the original isolates.

DNA isolation and Southern blot analysis. Total nucleic acids were preparedfrom mycelia of fungal strains cultured in 20 ml PDB as described previously bySuzuki et al. (37). Chromosomal DNA was enriched by digestion with RNase Aovernight at 37°C, followed by phenol-chloroform extraction and ethanol pre-cipitation. Ten micrograms of DNA was digested with the appropriate restrictionenzymes overnight at 37°C. The digested DNA was treated with phenol-chloro-form, ethyl alcohol precipitated, and suspended in water. The DNA was appliedto a well in 0.7% agarose gel, electrophoresed, and blotted onto a Hybond-N�nylon membrane (Amersham Biosciences, Buckingham, England). The mem-brane was washed twice in 5� SSC (1� SSC is 0.15 M NaCl plus 0.015 M sodiumcitrate [pH 7.0]), baked at 80°C for 10 min, and then probed with digoxigenin(DIG)-11-dUTP-labeled DNA fragments amplified from genomic DNA orcDNA according to methods recommended by the manufacturer (Roche Diag-nostics GmbH, Mannheim, Germany). Prehybridizations and hybridizationswere carried out at 42°C using the DIG Easy-Hyb Granules kit according to themanufacturer’s instructions (Roche). Hybridized bands were detected with aDIG detection kit and a CDP Star kit (Roche). Chemiluminescent signals werevisualized on film.

RNA preparation and Northern blot analysis. Total nucleic acids were pre-pared from mycelia of fungal strains cultured in PDB as described above. TotalRNA was enriched by eliminating fungal chromosomal DNA with two-rounddigestion of the extracted nucleic acids with RQ1 DNase I (Promega, Madison,Wis.) in the presence of an RNase inhibitor (40 units) (Toyobo, Tokyo, Japan)for 1 h at 37°C. After phenol, phenol-chloroform, and chloroform extractionsand ethanol precipitation, total RNA was suspended in sterile distilled water ata final optical density at 260 nm of 25.

For Northern blot analysis, single-stranded nucleic acids were precipitatedfrom total nucleic acids with LiCl (2 M) and treated twice with RQ1 DNase I.Total single-stranded RNA (ssRNA) was extracted twice by phenol-chloroformand suspended in sterile distilled water at an optical density at 260 nm of 25.When needed, poly(A) fractions were prepared from the total ssRNA isolatedfrom fungal mycelia using an mRNA isolation kit (Roche Diagnostics). TotalssRNA (25 �g) or poly(A) RNA (10 �g) was denatured, electrophoresed, andcapillary transferred onto a Hybond-N� nylon membrane (Amersham Bio-sciences, Buckingham, England) as described previously by Suzuki et al. (37).Procedures for prehybridization, hybridization, and detection of hybridizationsignals were described previously (37).

Plasmid rescue and I-PCR. Genomic DNA (10 �g) obtained from fungalstrains was cut with an appropriate restriction enzyme, which digested the in-serted plasmid at one site. After extraction by phenol-chloroform, the digestedproducts were self-linked by T4 DNA ligase (3 units) at 4°C overnight. ForI-PCR, the ligated DNA (200 ng) was mixed with the primer set HG15 and HG18(10 pmol) (see Table S1 in the supplemental material for sequences) and thelong and accurate (LA) Taq polymerase (2.5 units) (Takara) in a 50-�l reactionsolution (10 mM Tris-HCl [pH 8.0], 50 mM KCl, 2.5 mM deoxynucleosidetriphosphate). Reactions consisted of 4 min of initial denaturation at 94°Cfollowed by 35 thermal cycles: a denaturation step at 94°C for 30 s, an annealingstep at 52°C for 30 s, an extension step at 72°C for 10 min, and a final extensionstep at 72°C for 10 min. When needed, nested PCR was performed on I-PCRproducts in which the primer set consisting of HG21 and HG12 was used underthe same reaction conditions as those used for I-PCR.

Nucleotide sequencing of genomic DNA and cDNA of nam-1. The sequence ofgenomic DNA containing a 8.5-kb region of the nam-1 gene was determinedusing a primer-walking method. First, I-PCR products were sequenced withhph-specific primers to obtain the sequences of the junction regions of pHygR

integrated into chromosomes and the restriction enzyme sites used in I-PCR.The 6.5-kb DNA fragment that encoded the nam-1 gene was amplified bygenomic PCR with the primer set Nam5 and Nam25, whose sequences wereobtained by sequencing of the I-PCR products, and purified by a Wizard SV geland PCR clean-up system kit (Promega) according to the manufacturer’s pro-tocol. The PCR fragments or fragments cloned into the pGEM-T Easy vector(Promega) were sent to Macrogen, Inc. (Seoul, South Korea) to be sequencedwith a series of Nam primers listed in Table S1 in the supplemental material.

An almost-full-length cDNA of nam-1 was synthesized by using SuperScript IIreverse transcriptase (Invitrogen) and an oligo(dT) primer and amplified by tworounds of PCR with the primer set NamC1 and NamC15 and subsequently withthe primer set NamC2 and NamC16. This fragment was sequenced using a seriesof Nam and NamC primers (see Table S1 in the supplemental material). Bothterminal sequences of nam-1 mRNA were determined by using an RLM-RACEkit (Ambion). For 3� rapid amplification of cDNA ends (RACE), cDNA wassynthesized from mRNA isolated as described above using the 3� RACE adaptercontaining a dT tail as a primer; amplified by PCR with a primer set, the 3�RACE outer primer and a gene-specific primer, Nam59; and subsequently am-plified with the 3� inner primer and Nam60. The protocol for 5� RACE wasdescribed previously (39). The deoxyoligonucleotide supplied in the kit wasligated into mRNA treated with calf intestinal phosphatase and tobacco acidpyrophosphatase (decapping enzyme). The resulting RNA was subjected tocDNA synthesis using random decamers and SuperScript reverse transcriptaseand PCR amplification with primer sets consisting of the 5� RACE outer primerand Nam81 as well as the 5� inner primer and Nam82.

Both genomic and complementary DNAs were sequenced at least twice fromtwo directions. The determined sequences were examined for PCR misincorpo-ration.

Complementation assay. A few plasmid clones were prepared for functionalcomplementation of the namA mutant. After determining the sequences of bothflanking regions of the pHygR insertion sites, genomic DNA fragments of thewild-type allele were amplified on EP155 chromosomal DNA using LA PCR witholigonucleotide primer sets Nam5 and Nam25, Nam5 and Nam30-C, and Nam5and Nam42 (see Table S1 in the supplemental material for sequences). Theresulting fragments of approximately 6.5 kb, 7.5 kb, and 8.5 kb, respectively, werecloned into the pGEM-T Easy vector (Promega). After digestion with NotI, theDNA fragments were moved to vector pBSD1 to obtain the complementationconstructs pBSDnamC1, pBSDnamC2, and pBSDnam3. pBSD1 contained acoding domain for blasticidin S-deaminase that confers blasticidin S resistance atthe EcoRV site of pBluescript S(�) (Stratagene). Spheroplasts of the namAmutant were transformed with the complementing constructs as well as theempty vector pBSD1. The resulting transformants were selected on PDA con-taining 150 �g/ml blasticidin S (Nacalai Tesque, Kyoto, Japan) or a 1/10,000(vol/vol) dilution of the blasticidin S-based fungicide Bla-S (Nihon TokushuNouyaku, Nihonmatsu, Japan).

Construction of a replacement vector and targeted gene disruption. Targetedgene disruption was achieved by homologous recombination-based gene replace-ment previously established for this fungus (16, 21). A disruption plasmid clonewas constructed using pHygR with the pBluescript S(�) background. A genomicDNA fragment of approximately 1.7 kb (map positions �698 to 977) was am-plified by PCR using chromosomal DNA of EP155 as the template and theprimer set Nam5 and Nam10 (see Table S1 in the supplemental material) andcloned between the two NotI sites of the pGEM T-Easy vector (Promega). Afterliberation with NotI, the fragment was subcloned into the NotI site of pHygR.The 3� 1.3-kb region of nam-1 (map positions 3851 to 5116) was amplified byPCR using primers Nam2-C and Nam25. This fragment was inserted into theHindIII and KpnI sites of the intermediate plasmid. The resulting replacementvector was then used to transform TCys(72)-1, the parental strain for mutagen-esis.

Nucleotide sequence accession number. The sequence reported in this studywas deposited in the GenBank/EMBL/DDBJ database under accession numberAB300388.

RESULTS

Transformation of EP155 with a full-length cDNA clone ofCys(72). Several mutants of prototype hypovirus strain CHV1-EP713 were characterized previously (35, 36). Surmising thatdifferent host factors could be isolated using different viralmutants, we chose a site-directed mutant, Cys(72), that in-duced mild symptoms, allowing the host to sporulate relatively

742 FARUK ET AL. J. VIROL.


http://jvi.asm.org/

Dow

nloaded from

http://jvi.asm.org/

well (35). This moderate level of conidiation was important forthe screening procedure that required single conidial isolationto obtain homokaryons. Full-length cDNA of Cys(72) wasmoved into pCPXBn1 to obtain pBn1Cys(72) and was usedto transform EP155. More than 10 transformants withpBn1Cys(72) were selected on benomyl-containing PDA andexamined for copy numbers of the viral cDNA integrated intothe host chromosomes, phenotypic characteristics, and virusreplication. Most benomyl-resistant strains showed symptomstypical of strains transfected with the in vitro-synthesized RNAof Cys(72) (35). However, they have different copy numbers ofpBn1Cys(72), ranging from 1 to 3, and relative genetic stabilitywhen examined for their colony morphologies and abilities togrow on benomyl-PDA. Considering copy number and geneticstability, one of the transformants, TCys(72)-1, was selected formutagenesis by plasmid insertion. In Southern blot analyses(Fig. 1), a probe hybridizing the p29 coding sequence detectedonly a single major band when the chromosomal DNA ofTCys(72)-1 was digested by XbaI, KpnI, MfeI, and AfeI (Fig.1). These enzymes digest the cDNA sequence of Cys(72) atone site (KpnI or XbaI) or more than one site (the rest ofenzymes) downstream of the p29 coding sequence, while the

foundation vector pCPXBn1 has no (AflI or MfeI), one(XbaI), or more than one (KpnI) cut site. Southern analysisindicated that TCys(72)-1 contained one copy of pBn1Cys(72)per genome. Integrated cDNA was maintained through thenext generations in which virus infection could be launched bythe transmitted Cys(72) or initially from transcripts of thechromosomally integrated Cys(72) cDNA and, subsequently,by autonomous replication by the viral RNA polymerase com-plex (27).

Screening of the mutant collection for phenotypes differentfrom that of TCys(72)-1. Spheroplasts of TCys(72)-1 weretransformed with pHygR for mutagenesis by random plasmidinsertion and selected on hygromycin-containing PDA. A mu-tant population of homokaryonic single conidial isolates wascreated and observed for phenotypes differing from those ofparental strain TCys(72)-1 showing symptoms caused byCys(72). Of the 1,052 transformants containing pHygR, 109showed altered phenotypes, while many of the remainingtransformants manifested a phenotype that was indistinguish-able from that of TCys(72)-1. One of the altered strains, des-ignated namA, was subjected to further characterization in thisstudy. EP155 infected by CHV1-EP713 was severely reduced in

FIG. 1. Southern blot analysis of the parental strain TCys(72)-1. (A) Ten micrograms of genomic DNA prepared from virus-free EP155 andTCys(72)-1 was digested with four restriction enzymes (shown above the lanes). The restriction enzymes used were XbaI, KpnI, MfeI, and AfeI.Digested DNA was separated on a 0.7% agarose gel, transferred onto a Hybond N� nylon membrane, and probed with a 0.7-kb, DIG-11-dUTP-labeled p29 coding sequence. M refers to the 1-kb DNA ladder size marker (Fermentas, Life Science). (B) Physical map of the transformingplasmid pBn1-Cys(72). A circular form of pBn1-Cys(72) was used for the transformation of EP155 to prepare TCys(72)-1. Nevertheless, its linearform is shown to scale schematically. The digestion sites of restriction endonucleases and the probe sequence used in Southern blot analysis (A) areshown schematically. pBn1-Cys(72) is based on pCPXBn1, in which the cDNA to the genome of the Cys(72) mutant virus is flanked by the C.parasitica glyceraldehyde-phosphate dehydrogenase gene promoter (Pgpd) and terminator (Tgpd). The shaded bar represents the sequencedomains, derived from the foundation vector pCPXBn1 (14), which include the benomyl resistance gene, ampicillin resistance gene, and ColE1origin.



http://jvi.asm.org/

Dow

nloaded from

http://jvi.asm.org/

sporulation and pigmentation relative to virus-free EP155,while the parental strain TCys(72)-1 was restored in pigmen-tation like EP155 infected by a p29 null mutant virus, �p29, aspreviously reported (35). However, mutant namA had irregular

margins with a number of “protrusions” and was reduced inpigment production and asexual sporulation. The namA phe-notype resembled that of the EP67 strain containing a virus ofMichigan origin reported previously by Anagnostakis (see Fig.

FIG. 2. Characterization of the namA mutant. (A) Colony morphologies of TCys(72)-1, namA, and their virus-cured strains. The namA strain andits parental strain, TCys(72)-1, were cured of the Cys(72) mutant virus to obtain VC-namA and VC-TCys(72)-1, respectively, by a series of single conidialisolations. The strains were cultured on PDA medium for 10 days. VC-TCys(72)-1 and VC-namA showed indistinguishable phenotypes compared tovirus-free EP155. (B) Southern blot analysis for namA and VC-namA. Ten micrograms of chromosomal DNA of Cys(72)-containing namA and virus-freeVC-namA mutant strains was digested with restriction enzymes (HindIII, SacI, SpeI, EcoRI, and PstI) (shown above the lanes). The enzymes HindIII,SacI, and SpeI cut at one site in the pHygR plasmid sequence outside the hph gene, while EcoRI and PstI cleaved at two sites within and outside thehph gene. The number of hybridizing bands confirmed the single insertion of pHygR into the namA mutant. Chromosomal DNA of nontransformantstrain EP155 was also digested by HindIII. After transfer to a Hybond N� nylon membrane, digested DNA was hybridized to a DIG-11-dUTP-labeledhph probe. The estimated sizes of the hybridizing bands are indicated at the right. M refers to the 1-kb DNA ladder size marker (Fermentas). (C) Colonymorphology of VC-TCys(72)-1 and VC-namA strains infected with MyRV1. The photographs were taken at day 10 after inoculation of PDA.(D) Phenotypic properties of the VC-TCys(72)-1 and VC-namA strains infected with hypovirus strains. The hypovirus variants �p29, �p40b, and �p69band wild-type CHV1-EP713 were introduced into the fungal strains VC-TCys(72)-1 and VC-namA (see Materials and Methods for virus introduction).Phenotype alterations similar to those of the namA mutant are shown in VC-namA infected by �p29- and �p40b- but not in �p69b-infected VC-namA.Strain VC-TCys(72)-1 induced phenotypic changes indistinguishable from those of EP155 upon virus infection irrespective of virus strains. Back viewsof VC-namA strains infected with CHV1-EP713 and �p29, showing the characteristic white corrugated edge of namA, are shown. VC-TCys(72)-1 eitheruninfected or infected with the individual virus was cultured in parallel for comparison.



http://jvi.asm.org/

Dow

nloaded from

http://jvi.asm.org/

6a, bottom row, in reference 2). The numbers and sizes of theprotrusions of namA on PDA varied between cultures (e.g.,compare namA cultures in Fig. 2A and 7C), possibly due toslight environmental differences or unknown factors. EP155anastomosed with namA showed a phenotype similar to that ofTCys(72)-1, indicating that Cys(72) cDNA integrated into mu-tant chromosomes was unaltered during mutagenesis andlaunched infection by Cys(72).

The phenotype of the mutant namA is altered in a virusinfection-specific way. To determine whether the phenotype ofthe mutant namA was altered in the absence of virus infectionor in the presence of other virus strains, we attempted to curethe mutant namA strains of the Cys(72) virus. During thecourse of the genetic stability assay, pBn1-Cys(72) in namAwas found to be relatively stably maintained through asexualspores. However, of 36 single conidial isolates derived fromnamA, 2 lost benomyl resistance and/or signs of virus infection,possibly due to the deletion of the integrated pBn1-Cys(72).This phenomenon for transformants with cDNA of wild-typeCHV1-EP713 was previously reported by Chen et al. (4). Tak-ing advantage of this property, the namA strain was cured ofCys(72) by single conidial isolation. As expected, a few germ-lings (3 of 60) manifested the typical virus-free phenotypecharacterized by orange pigment production and high sporu-lation levels (Fig. 2A). These virus-free (virus-cured) strainswere termed VC-namA. Likewise, Cys(72) was cleared fromTCys(72)-1 to obtain a virus-free strain, VC-TCys(72)-1. Thesevirus-cured strains, VC-namA and VC-TCys(72)-1, did notharbor viral genomic double-stranded RNA (data not shown)and showed colony morphology attributes indistinguishablefrom those of wild-type EP155 (Fig. 2A and Table 2). VC-namA receiving Cys(72) back after being fused with Cys(72)-infected EP155 restored the namA phenotype (data notshown). This showed that the mutant phenotype found innamA was virus infection specific.

Southern blot analysis of the namA chromosomal DNA withSalI and XhoI suggested the integration of one copy of pHygRin the chromosomes of namA (data not shown). Utilizing dif-ferent restriction enzymes, we confirmed it and determinedwhether the integrated pHygR was retained in VC-namA.When chromosomal DNA was digested by HindIII, SacI, orSpeI, Southern blot analysis revealed that a probe containingthe hph sequence was detected in single bands at the samemigration positions corresponding to approximately 8 kb inboth namA and VC-namA (Fig. 2B). These three restrictionenzymes cut the pHygR plasmid at one single digestion site

outside the hph sequence. Somehow, digestion with HindIIIand SacI provided the hybridization signals with different in-tensities for namA and VC-namA. Digestion with EcoRI orPstI by cleaving the pHygR sequence at two sites, i.e., withinand outside the hph sequence, resulted in the detection of twosignals of over 10 kb and 0.6 kb (for EcoRI) and of 4.5 kb and0.7 kb (for PstI) in namA and VC-namA (Fig. 2B).

These combined results indicated that mutant namA con-tained a single copy of the mutagenic plasmid pHygR, whichwas maintained stably in VC-namA.

Phenotypic changes in VC-namA is virus species and virusstrain specific. As shown in Fig. 2, the namA mutant revertedto the wild-type (parental) phenotype upon virus elimination,and infection of VC-namA with certain strains of CHV1-EP713 was required to regain the phenotype. We next deter-mined whether the phenotype alteration manifested by VC-namA was virus species specific. A relatively well-characterizedmycoreovirus of C. parasitica, MyRV1, was chosen because atransfection protocol was available for this virus (18). Interest-ingly, unlike the VC-namA strain infected with Cys(72), nosignificant difference was observed in colony morphologies ofVC-namA and EP155 or VC-TCys(72)-1 that were infectedwith MyRV1 (Fig. 2C). Thus, the mutated gene in VC-namAwas considered to be involved in symptom induction by wild-type CHV1 but not by MyRV1.

Furthermore, several CHV1-EP713 mutants (�p29, �p40b,and �p69b) were introduced into VC-namA and VC-TCys(72)-1 by fusing them with the donor strain EP155 in-fected with the respective virus strains. Upon infection with thewild type, CHV1-EP713, �p29, or �p40b, VC-namA showed areduced growth rate and repressed pigmentation with irregularmycelial extensions (Fig. 2D). As observed with namA, levelsof growth suppression and numbers and lengths of hyphalextensions were different between cultures of single strainsinfected with each virus. These phenotypic characteristics werevery similar to those exhibited by namA, being distinguishablefrom VC-TCys(72)-1 or EP155 infected with the respectiveviruses that showed regular symptoms as previously reported(36). Irregular colony morphology of virus-infected VC-namAwas concurrent with the firm colony edge (Fig. 2D). Interest-ingly, no significant differences in colony morphology were,however, detected between VC-namA and EP155 or VC-TCys(72)-1 strains when infected with �p69b (Fig. 2D).

Comparative measures of other biologic properties of VC-namA and EP155 infected with different viral strains are sum-marized in Table 2. Conidiation levels are usually decreased

TABLE 2. Levels of asexual sporulation and virulence of fungal strains EP155 and VC-namA infected with several hypovirus variantsand a mycoreovirus, MyRV1a

StrainAvg no. of conidia/ml � SD Avg lesion area (cm2) � SD

EP155 VC-namA EP155 VC-namA

Virus-free 1.9 � 108 � 5.9 � 107 1.6 � 108 � 3.9 � 107 17.15 � 3.09 17.43 � 4.35CHV1-EP713 2.8 � 106 � 1.3 � 106 2.2 � 104 � 7.5 � 103 9.36 � 0.97 9.47 � 0.76�p29 6.3 � 107 � 3.4 � 107 8.8 � 105 � 5.7 � 105 7.86 � 1.09 8.26 � 1.05�p40b 8.3 � 107 � 3.5 � 107 1.6 � 106 � 7.1 � 105 7.46 � 2.31 7.67 � 1.42�p69b 1.1 � 108 � 4.6 � 107 9.8 � 107 � 3.2 � 107 2.57 � 0.11 2.85 � 1.29MyRV1-Cp9B21 4.0 � 107 � 7.0 � 106 7.5 � 107 � 1.4 � 107 2.55 � 0.18 2.49 � 0.17

a Averages and standard deviations were calculated from six measurements for conidiation and virulence for each strain. In the virulence assay, the areas of the lesionsinduced on commercially purchased apples were measured at day 14 postinoculation.



http://jvi.asm.org/

Dow

nloaded from

http://jvi.asm.org/

upon infection by CHV1-EP713 derivatives, although differentviral strains have different magnitudes of impact (36). Withabnormality of colony morphology, a further decrease inconidiation was found in virus-infected VC-namA. When in-fected with wild-type CHV1-EP713, �p29, �p40b, or Cys(72),VC-namA manifested a 1- to 2-log reduction in conidiationlevels relative to those of EP155 or VC-TCys(72)-1 infectedwith the respective viruses. For example, VC-namA infectedwith wild-type CHV1-EP713 produced conidia at 2.2 � 104

conidia/ml, a reduction of over 2 magnitudes relative to EP155infected by CHV1-EP713 (2.8 � 106 conidia/ml). However, nosignificant difference in conidiation was found in strains VC-namA and EP155 infected with either �p69b or MyRV1. Un-infected VC-namA and EP155 also sporulated at similar levels(1.9 � 108 versus 1.6 � 108 conidia).

Pigment production was also reduced in VC-namA infectedwith �p29 or �p40b compared to EP155 infected with therespective viruses, while VC-namA showed a level of pigmentproduction similar to those of EP155 and VC-TCys(72)-1 (Fig.2A and D). Therefore, repression in pigment production andasexual sporulation were observed only in strains that showedaltered colony morphologies upon virus infection (Fig. 2A andD and Table 2).

In contrast, no significant difference in virulence was ob-served between the VC-namA and EP155 backgrounds, re-gardless of virus infection, when assayed with apples (Table 2).Lesions of similar sizes (ranging from 7.46 to 9.47 cm2) wereformed on apples inoculated by the fungal strains EP155 orVC-namA that carried wild-type, �p29, and �p40b virusstrains. Uninfected EP155 or VC-namA induced much largerlesions (17.15 and 17.43 cm2), while �p69b and MyRV1 causeda severe reduction in the size of the lesions formed by thosefungal strains (2.49 to 2.85 cm2).

Transgenic supply of either p29 or p40 confers the namAphenotype to VC-namA infected by �p69. The observation thatVC-namA shows an irregular phenotype similar to that ofnamA upon infection by �p29 or �p40, but not by �p69, raisesthe possibility that the manifestation of the namA phenotype inVC-namA requires either p29 or p40. To test this, effects oftransgenic expression of the proteins on the colony morphol-ogy of �p69-infected VC-namA were investigated. Interest-ingly, transformants of VC-namA with the p29 coding domain(VC-namA/p29) were similar in phenotype to namA when in-fected with �p69 (Fig. 3). Likewise, a transgenic supply of p40(VC-namA/p40) (Fig. 3) or p69 (VC-namA/p69) (data notshown) altered the phenotype of �p69-infected VC-namA intothe namA type. �p69-infected VC-namA/p29 and VC-namA/p40 were similar to each other while slightly different in theshapes of mycelial extensions. In contrast, �p69-infected VC-namA that had been transformed with the empty vector pCPX-BSD1 showed symptoms that were identical to those of �p69-infected EP155. These results indicate that either p29 or p40 isrequired for VC-namA to show the namA phenotype wheninfected with �p69.

Determining the insertion site of the mutagenic plasmidHygR and the sequence of its flanking region. The mutantnamA was expected to have two different plasmid clones,pBn1Cys(72) and the mutagenic plasmid pHygR. Based on theassumption that a copy of the hph sequence detected by South-ern blotting (Fig. 2B) contributed to the phenotype of namA,

we used I-PCR to amplify DNA fragments adjacent to theinsertion site. As shown in Fig. 4A, I-PCR on HindIII-, XhoI-,and SacI-digested chromosomal DNA of namA and VC-namAwith the primer set HG15 and HG18 allowed the amplificationof a large DNA fragment. A series of hygromycin resistancegene-specific (HG) primers are specific to the hph sequence(see Table S1 in the supplemental material). The sizes ofI-PCR-amplified fragments were approximately 9 kb and 8 kbfor HindIII and SacI digests, respectively, of chromosomalDNA from namA or VC-namA (Fig. 4A). I-PCR on XhoI-digested DNA of namA and VC-namA amplified two majorfragments of approximately 6 kb and 3.5 kb and one minor4.5-kb fragment. DNA of the expected size of 6 kb was pro-duced from mutagenic plasmid DNA by I-PCR (pHygR) (Fig.4A), while no amplification was observed from genomic DNAfrom EP155. Nested PCR was performed on I-PCR productsas shown in Fig. 4B. The nested PCR led to the amplificationof large amounts of specific single DNA fragments (9 kb forHindIII, 4.5 kb for XhoI, and 8 kb for SacI) (Fig. 4B). The sizesof the nested-PCR-amplified fragments were in accordancewith the sizes of the DNA fragment observed in the Southernblot analysis (Fig. 2B and 4B). The lack of differences in I-PCRand nested PCR profiles between namA and VC-namA pro-vided additional evidence that pHygR was retained stably inVC-namA.

These inverse and nested PCR fragments were sequenceddirectly or after being cloned into the pGEM-T Easy plasmid.A chromosomal DNA fragment of approximately 8.5 kb span-ning both sides of the integrated plasmid was sequenced. I-PCR and nested PCR on SacI-cleaved/self-ligated DNA pro-vided the flanking region on the opposite side of that obtainedby HindIII and XhoI DNA. The sequence of the 8.5-kb region,excluding the pHygR sequence, and the plasmid insertion siteare shown in Fig. S1 in the supplemental material. The corre-sponding region in the wild-type allele was also sequenced. Thesequence of the inserted mutagenic plasmid is available uponrequest. Linearization of plasmid pHygR at the AmpR codingdomain and deletion of a small region of the plasmid occurredduring the insertion event.

FIG. 3. Effects of transgenic expression of the ORF A-encodedproteins on the colony morphology of VC-namA. Spheroplasts of VC-namA were transformed with pCPXBSD1 containing the coding do-mains for p29 (VC-namA/p29) or p40 (VC-namA/p40). To introduce�p69b, independent transformants were fused with EP155 infectedwith the mutant viral strain. Virus-free transformed fungal coloniesand corresponding �p69b-infected fungal colonies were cultured onPDA for 10 days. A fungal strain transformed with the empty vectorwas grown in parallel (pCPXBSD).



http://jvi.asm.org/

Dow

nloaded from

http://jvi.asm.org/

To ensure insertion site and sequence integrity, genomic LAPCR was performed (Fig. 4C). When PCRs were conductedwith the primer set Nam2 and Nam5 (see Fig. 5A for primermap positions), each specific to the endogenous sequence, allthree fungal strains tested, EP155, namA, and VC-namA, gaveproducts of the expected size of 5 kb (Fig. 4C). The primer setconsisting of Nam5 and hph-specific primer HG21 amplified a6-kb fragment in namA and VC-namA but not in EP155 thatdid not harbor pHygR.

Complementation assay. To determine whether the inser-tion mutation identified above was associated with the virusinfection-specific phenotypic changes in namA (Fig. 2A), threeDNA fragments of approximately 6.5 kb, 7.5 kb, and 8.5 kb,containing the corresponding unaltered region in the insertionsite, were LA PCR amplified from chromosomal DNA fromwild-type EP155 (wild-type allele). These fragments werecloned into plasmid pBSD1 containing the blasticidin resis-tance gene as the selectable marker (pBSDnamC1, pBSD-namC2, and pBSDnamC3) (Fig. 5A) and transformed intomutant strain namA to determine their abilities to complementthe phenotypic abnormalities. For each transforming DNAclone, several transformants were obtained. Colony morphol-ogies of representative isolates are shown in Fig. 5B. Conse-quently, all the plasmid clones complemented the altered phe-notype of the mutant and conferred a TCys(72)-1 phenotype tonamA; that is, namA transformed with the DNA fragmentsmanifested regular mycelial growth without hyphal extensionsand restoration of pigment production to the level shown byTCys(72)-1 (namA/namC1, namA/namC2, and namA/namC3)(Fig. 5B). Transformants with pBSDnamC2 showed a slightincrease in pigmentation compared to those with the other twocomplementing DNA clones. Transformation with the emptyvector pBSD1 failed to restore the phenotype of TCys(72)-1(namA/pBSD1) (Fig. 5B). Two independent transformantswith pBSD1 were similar in pigmentations and growth ratesbut different in the irregular margins from each other (Fig. 5).These results indicated that the 6.5-kb DNA fragment wassufficient for the complementation of defects in a gene(s) in-volved in the namA phenotype.

Characterization of the nam-1 gene. Based on data from thecomplementation assay (Fig. 5), the 6.5-kb fragment may con-tain sufficient sequence for a functional gene mutated in namA.cDNA from mRNA derived from the wild-type allele was syn-thesized, and its nucleotide sequence was determined to revealthe organization of the gene, termed nam-1. The exon-intronstructure of nam-1, which is composed of eight exons andseven introns, is shown in Fig. 5A. An ORF starts at the AUGtriplet at map positions 19 to 21 and ends with the TAG codonat map positions 4381 to 4383 (the transcription start site onthe genomic DNA is numbered 1). Most splicing sites conformto the GT/AG rule, except for the first intron, where it is5�-GT—————CG-3� rather than 5�-GT—————AG-3�.A putative CAAT box and a TATA box (see Fig. S1 in thesupplemental material) are found at positions �95 to �98 and�155 to �158, respectively, and are more distantly locatedfrom the transcription start site than usual. Additional TATAand CAAT sequences are found in further upstream regions(see Fig. S1 in the supplemental material). The nucleotidesequences of the genomic DNA and cDNA of nam-1 mRNAand the deduced amino acid sequence of NAM-1 comprising

FIG. 4. Amplification of pHygR plasmid insertion regions on the namAchromosomal DNA. (A) I-PCR for amplification of upstream and down-stream DNA regions flanking the pHygR insertion site. I-PCR with theprimer set HG15 and HG18, specific for the hph sequence, was performed onthe genomic DNA of namA and VC-namA (approximately 200 ng) digestedwith HindIII, XhoI, or SacI (indicated above the lanes) and self-ligated.HindIII-digested, self-ligated DNA of virus-free EP155 was used as a nega-tive control, while pHygR plasmid DNA (10 ng) was used as a positivecontrol. Purified pHygR DNA was also electrophoresed (lane plasmid). Analiquot of each I-PCR product was analyzed by 0.7% agarose gel electro-phoresis using the 0.5� TAE buffer system (20 mM Tris-acetate, 0.5 mMEDTA [pH 7.8]), stained in ethidium bromide for 30 min, and photographedunder a UV transilluminator. M refers to 1-kb DNA ladder size markers inthis and subsequent figures (B and C) (Fermentas, Life Science). (B) NestedPCR amplification on the I-PCR products. One microliter of I-PCR productsshown in Fig. 4A was subjected to nested PCR with the hph-specific primerset HG21 and HG12. I-PCR products from plasmid DNA (pHygR) andEP155 were also used as positive and negative controls, respectively. Ampli-fied DNA fragments were electrophoresed as described in the legend of Fig.4A. (C) LA PCR analysis of namA and VC-namA. Genomic PCR wasperformed using primers designed from the endogenous DNA sequence(nam-1 gene) (Nam2 and Nam5) and the mutagenic plasmid (HG21). Theprimer set Nam2 and Nam5 allowed the amplification of approximately 5-kbfragments in all the fungal strains, EP155, namA, and VC-namA, while theprimer set of hph-specific HG21 and nam-1-specific Nam5 gave products of6-kb DNA fragments in namA and VC-namA but not in EP155.



http://jvi.asm.org/

Dow

nloaded from

http://jvi.asm.org/

1,257 amino acids are shown in Fig. S1 in the supplementalmaterial. To confirm the expression of the nam-1 gene inwild-type strain EP155 and the parental strain TCys(72)-1,Northern analysis was pursued. It was, however, difficult todetect nam-1 transcripts when total RNA fractions were used.Thus, poly(A)-enriched RNA fractions were prepared forNorthern analysis. As a result, nam-1 was shown to be ex-pressed in the two C. parasitica strains (Fig. 6A). No significantdifference in nam-1 transcription levels between fungal colo-nies infected and uninfected with the wild-type CHV1 orCys(72) was found, suggesting that nam-1 was not responsiveto hypovirus infection (Fig. 6A).

Dawe et al. (13) constructed a publicly available EST librarycontaining 2,200 genes from C. parasitica. nam-1 was not foundin the EST collection. However, a BLAST search of the GenBank

library with the deduced amino acid sequence of NAM-1revealed similar hypothetical proteins from other ascomycet-ous, filamentous fungi including Neurospora crassa, Gibberellazeae, and Chaetomium globosum. The sequence similaritieswere found over the entire sequences (see Fig. S2 in the sup-plemental material). The levels of identities found in NAM-1and its counterpart were 38%, with an E value of 2e-172, for N.crassa; 34%, with an E value of 3e-121, for G. zeae; and 34%,with an E value of 3e-121, for C. globosum. Highly conservedsequence stretches were detected in the middle and C-terminalregions. The equivalent, hypothetical proteins of N. crassa andC. globosum have a C-terminal extension of approximately 250amino acids (aa) and a deletion of approximately 200 aa in thecentral region, respectively, thus being different in size fromNAM-1 (see Fig. S2 in the supplemental material).

FIG. 5. Complementation test of the namA mutant. (A) Illustration of plasmid vector constructs for complementation assay. Three nam-1gene-containing fragments were amplified by LA PCR from genomic DNA of EP155 with the wild-type nam-1 allele using primer sets Nam5 andNam25, Nam5 and Nam30-C, and Nam5 and Nam42. The amplified DNA fragments of 6.5 kb, 7.5 kb, and 8.5 kb were cloned into the pBSD1vector to prepare the complementation constructs pBSDnamC1, pBSDnamC2, and pBSDnamC3. pBSD1 contains a coding domain for theblasticidin S resistance gene (blasticidin S-deaminase). Eight exons of the nam-1 gene are indicated by green boxes, while the introns are depictedas gray bars. The pHygR insertion site in the eighth exon of nam-1 is indicated by a triangle. Map positions of primers and recognition sites ofrestriction enzymes used in I-PCR and Southern blot analyses (Fig. 2 and 4) are shown. An exon-intron structure of the gene immediatelydownstream of nam-1, which encodes a carboxypeptidase (M. I. Faruk and N. Suzuki, unpublished data), is also shown in the same manner.(B) Colony morphology of the complemented strains of namA. Three representative complementation strains, designated namA/namC1, namA/namC2, and namA/namC3, are shown. Those strains were obtained by transforming namA with pBSDnamC1, pBSDnamC2, and pBSDnamC3,respectively. All complemented strains were grown on PDA in parallel with EP155, TCys(72)-1, namA, and namA transformed with vector pBSD1for 7 days at 25 to 27°C on the bench top.



http://jvi.asm.org/

Dow

nloaded from

http://jvi.asm.org/

A search of motif databases with the NAM-1 sequence iden-tified the CorA domain at its C-terminal region (aa 792 to1170), as defined by PFAM family PF01544 (Fig. 6B), in ad-dition to a number of consensus sequences including glycosy-lation sites, phosphorylation sites, myristoylation sites, andleucine zipper motifs that are frequently found in protein se-

quences. Interestingly, the CorA motif was conserved in thecounterparts of filamentous fungi. The N-terminal portion ofNAM-1 contains the PFAM DUF1777 domain at aa 133 to 253shared by the counterpart of G. zeae but not by that of N. crassaor C. globosum. The CorA consensus sequence is found inprokaryotic and eukaryotic members of the CorA Mg2� trans-

FIG. 6. Expression of nam-1 in fungal colonies and functional domains of NAM-1. (A) Northern blot analysis of nam-1. Transcription levelsof nam-1 between hypovirus-infected and uninfected colonies were compared. Ten micrograms of poly(A)-enriched RNA isolated from themycelia of wild-type strain EP155, CHV1-EP713-infected EP155 (CHV1-EP713), and TCys(72)-1 was electrophoresed in a 1.4% gel underdenaturing conditions and transferred onto a nylon membrane for hybridization with a nam-1-specific, DIG-11-dUTP-labeled probe (top). Todetect -tubulin mRNA as a loading marker, 3 �g of RNA from the same poly(A)-enriched fraction as that in the top panel was probed by aDIG-11-dUTP-labeled probe amplified on -tubulin cDNA (bottom). (B) Schematic diagram showing two domains, DUF1777 and CorA, ofNAM-1 from C. parasitica. A PFAM search (http://www.sanger.ac.uk/Software/Pfam/search.shtml) with NAM-1 revealed a few domains: CorA asa trusted match and other domains as potential matches represented by DUF1777 with the lowest E value. Positions of the CorA Mg2� transporterdomain and a less characterized domain, DUF1777, are shown in the NAM-1 sequence. (C) Sequence alignment of NAM-1 and its counterpartsfrom other filamentous fungi and prokaryotes. The sequences of the C. parasitica NAM-1 were aligned with those of corresponding hypotheticalproteins from the ascomycetous fungi Neurospora crassa, Gibberella zeae, and Chaetomium globosum and two bacteria, Burkholderia vietnamiensisand Azotobacter vinelandii. For the overall alignment, see Fig. S2 in the supplemental material. All proteins contain the CorA domain, while onlythe counterpart from G. zeae possesses the DUF1777 domain. Accession numbers for the GenBank database are SM_955371 for N. crassa,SM_381249 for G. zeae, SM_001226467 for C. globosum, Q4BIY7 for B. vietnamiensis, and Q4IVJ8 for A. vinelandii. The two transmembranedomains, TM1 and TM2, within the CorA consensus sequence are boxed. At the C terminus of TM1, the signature sequence GMN is strictlyconserved. Asterisks refer to amino acids that are strictly conserved, while colons and dots indicate changes to chemically similar amino acids.



http://jvi.asm.org/

Dow

nloaded from

http://jvi.asm.org/

porter family (25), while the DUF1777 domain is less charac-terized and defined. The three-dimensional structures of oneof the CorA members were recently revealed at 2.9-Å resolu-tion and clearly showed two transmembrane domains (24). ThePSORT prediction (http://psort.nibb.ac.jp/form.html) also re-vealed two transmembrane sequences of NAM-1, at aa 1111 to1138 (transmembrane domain 1 [TM1]) and 1148 to 1174(TM2). The NAM-1 transmembrane region within the CorAdomain can be aligned with the corresponding regions of otherCorA homologues even with prokaryotic members. TheNAM-1 counterparts from the three filamentous ascomycetousfungi and those of two bacteria, Burkholderia vietnamiensis andAzotobacter vinelandii, with the highest sequence identitiesamong prokaryotic CorA members are included in the align-ment shown in Fig. 6C. It is also noteworthy that TM1 containsthe GMN motif at its C terminus (Fig. 6C), which was previ-ously shown to be highly preserved in prokaryotic and eukary-otic CorA homologues (25).

Disruption of nam-1 in TCys(72)-1 results in virus infection-specific alteration of the colony phenotype. To further confirmthe involvement of mutations of the nam-1 gene in the phe-notypic change found in namA, a nam-1 disruption constructthat contained an hph gene cassette inserted between the 5�and 3� sequences of nam-1 was prepared (Fig. 7A). The con-struct was designed to delete six exons in the central region ofthe nam-1 gene (Fig. 7A) based upon homologous recombina-tion-based gene replacement. A population of approximately120 transformants of TCys(72)-1 with the construct wasscreened for disruptants using selection on hygromycin-con-taining PDA and subsequently by genomic PCR. Several inde-pendent disruptant candidates, such as TCys(72)-1/�nam1-1and TCys(72)-1/�nam1-2 were obtained. Targeted disruptionwas further confirmed by Southern blot analysis (Fig. 7B); thatis, probe 1, spanning map positions 191 to 1381, hybridized1.8-kb DNA fragments in TCys(72)-1/�nam1-1 and TCys(72)-1/�nam1-2, while EP155 and TCys(72)-1 gave a signal corre-sponding to 5.3 kb when genomic DNA was digested by XbaI.No signal was found in the disruptant candidates with probe 2harboring the nam-1 sequence that was expected to be deleted,while a 5.3-kb band was detected in EP155 and TCys(72)-1. Anhph probe (probe 3) detected the expected signals (2.2 kb) onlyin disruptant candidates [TCys(72)-1/�nam1-1 and TCys(72)-1/�nam1-2] but not in EP155 or TCys(72)-1. These data clearlyindicated that TCys(72)-1/�nam1-1 and TCys(72)-1/�nam1-2were nam-1 disruptants.

TCys(72)-1/�nam1-1 and TCys(72)-1/�nam1-2 showed sim-ilar colony morphologies (Fig. 7C). The disruptants had un-even, dense aerial hyphae and lacked the protrusions pos-sessed by namA (Fig. 7C). The disruptants were also reducedin growth rate and pigmentation compared with TCys(72)-1(Fig. 7C). The suppression in pigmentation found in the dis-ruptants was pronounced when cultured in PDB (Fig. 8). Lev-els of pigmentation suppression were exhibited by the dis-ruptants and were similar to those found in namA, which werelower than those in TCys(72)-1 and much lower than those inEP155 or VC-namA (Fig. 8). These showed the involvementof nam-1 in host pigmentation. To determine the phenotypeof virus-free nam-1 disruptants, TCys(72)-1/�nam1-1 andTCys(72)-1/�nam1-2 were cured of the Cys(72) virus as in thecase of namA and TCys(72)-1 [VC-TCys(72)-1/�nam1-1 and

VC-TCys(72)-1/�nam1-2]. In the absence of virus infection,the disruptants VC-TCys(72)-1/�nam1-1 and VC-TCys(72)-1/�nam1-2 were identical to VC-namA and EP155 (data notshown).

FIG. 7. Disruption of the nam-1 gene of TCys(72)-1. (A) Exon/intron organization of the C. parasitica nam-1 gene (top) and organi-zation of the nam-1 gene disruption construct. The 6-kb genomic DNAcontaining nam-1 is composed of eight exons (green boxes) and sevenintrons (gray bars). A nam-1 gene disruption construct was generatedby replacing a fragment covering the central six exons between primersNam10 and Nam2 with a 2.3-kb cassette composed of the Escherichiacoli hygromycin B phosphotransferase gene (hph) flanked by the A.nidulans trpC promoter (Ptrp) and terminator (Ttrp). The DNA oper-ation was carried out with the aid of PCR using primers Nam5, Nam10,Nam2, and Nam25. The primers’ positions are indicated above thegene organization schematic. DNA regions used as probes (probe 1,probe 2, and probe 3) for Southern blot analysis are indicated by a barbelow the diagrams. The recognition site of XbaI used in Southernblotting (Fig. 7B) is indicated on the nam-1 gene organization. A 1-kbscale bar is shown. (B) Southern hybridization analysis of the nam-1disruption mutants. Endonuclease XbaI-digested DNA (10 �g) ofEP155, TCys(72)-1, and the disruptant candidates TCys(72)-1/�nam1-1 and TCys(72)-1/�nam1-2 was separated in a 1% agarose gel,blotted onto a nylon membrane, and probed with DIG-11-dUTP-labeled probe 1. The same blot was reprobed sequentially by probe 2and probe 3. (C) Effect of nam-1 disruption on phenotypic propertiesof TCys(72)-1. Two independent nam-1 disruption candidates,TCys(72)-1/�nam1-1 and TCys(72)-1/�nam1-2, were grown on PDAfor 10 days on the laboratory bench at 25 to 27°C. TCys(72)-1, namA,EP155, and VC-namA were cultured in parallel.



http://jvi.asm.org/

Dow

nloaded from

http://jvi.asm.org/

These results showed that the disruption of nam-1 in theTCys(72)-1 backgrounds had a virus infection-specific effect oncolony morphology. The data also suggested that nam-1 wasinvolved in the altered phenotypic changes in namA, but thedisruption of the gene was not sufficient to reproduce its phe-notype.

Altered, longer versions of nam-1 transcripts contribute tothe namA phenotype. The observation that the morphologiesof TCys(72)-1/�nam1-1 and namA were not identical (Fig. 7C)suggested that the phenotype of namA may be due to expres-sion of the modified nam-1 transcripts with altered functionrather than because of a simple disruption of its function. Toexamine this possibility, EP155, namA, and TCys(72)-1/�nam1-1 were subjected to Northern analysis. Consequently,while no signal was observed in TCys(72)-1/�nam1-1 at theexpected size (approximately 4 kb), transcripts of nam-1 weredetected in EP155 and namA (Fig. 9A). Nevertheless, tran-scripts of a slightly large size than that in TCys(72)-1 or EP155were detected in namA (Fig. 6A and 9A, compare lane 1 to

FIG. 9. Contribution of an altered longer version of nam-1 tran-scripts to the namA phenotype. (A) Northern hybridization analysis todetect nam-1 transcripts. Ten micrograms of poly(A)-enriched RNAisolated from the mycelia of wild-type strain EP155, the namA mutant,and the nam-1 disruptant with the TCys(72)-1 backgrounds, TCys(72)-1/�nam1, was electrophoresed in a 1.4% gel under denaturing condi-tions and transferred onto a nylon membrane for hybridization with anam-1-specific, DIG-11-dUTP-labeled probe (top). -Tubulin mRNAwas used as a loading marker. To detect -tubulin mRNA as a loadingmarker, 3 �g of RNA from the same poly(A)-enriched fraction as thatin the top panel was probed by a DIG-11-dUTP-labeled probe ampli-fied on -tubulin cDNA (bottom). (B) Effects of expression of modi-fied nam-1 on the colony morphology of TCys(72)-1/�nam1-1. A 6-kbgenomic DNA fragment carrying the HygR sequence on exon 8 ofnam-1 was amplified on the genomic DNA of namA by LA PCR usingprimer sets Nam5 and HG17 and cloned into the NotI site of pBSD1(pBSD-nam1-HygR). Exons/introns of nam-1, the pHygR integrationsite, and primers’ map positions are shown in Fig. 5A. Spheroplasts ofthe TCys(72)-1/�nam1-1 strain were transformed with pBSD-nam1-HygR to generate TCys(72)-1/�nam1-1/nam1-HygR1. The resultingtransformant strain was cultured on PDA in parallel with theTCys(72)-1, TCys(72)-1/�nam1-1, and the namA mutant strain for 8days on the laboratory bench at 25 to 27°C.

FIG. 8. Comparison of pigment production of nam-1 disruptionmutants with those of other strains. (A) Pigment production of anam-1 disruptant [TCys(72)-1/�nam1-1] in PDB medium. The disrup-tion mutant was cultured in 20 ml PDB (Difco) in parallel with itsparental strain, TCys(72)-1, Cys(72)-containing and VC-namA strains,and EP155 for 12 days on a laboratory bench at 25 to 27°C andphotographed. (B) Relative levels of pigmentation of different fungalstrains. Three hundred milligrams of harvested mycelia from eachstrain shown in Fig. 8A was pulverized under liquid nitrogen andhomogenized in 3 ml absolute ethanol (18). An aliquot of the mixturewas transferred into a 1.5-ml microcentrifuge tube, incubated at roomtemperature for 3 h followed by centrifugation at 12.5 krpm for 1 min,and photographed. (C) Graphical representation of relative levels ofpigmentation of different fungal strains. Pigmentation was measured asthe absorbance at 454 nm by a densitometer. Ten replicates for eachisolate of namA, TCys(72)-1/�nam1-1, TCys(72)-1, EP155, and VC-namA were used for the measurements to calculate averages andstandard deviations.



http://jvi.asm.org/

Dow

nloaded from

http://jvi.asm.org/

lane 2). This may indicate that the plasmid insertion into thelast exon of the nam-1 gene in namA (Fig. 5) leads to theproduction of extended transcripts containing the partialpHygR sequence. Data from 3� RACE supported this possi-bility and showed that the altered nam-1 transcripts in namAcontained a deletion of the 3�-terminal region of approximately150 nucleotides (nt) and an extension of approximately 450 ntderived from the pHygR plasmid sequence (see Fig. S3 in thesupplemental material). The integration of pHygR induced anamino acid sequence change at the C terminus from ——RSTEGKPPGGRIKNGNIL [in EP155 or TCys(72)-1] to ——TNTVLLV (in namA). Thus, it seemed likely that the genera-tion of modified transcripts was involved in the namAphenotype.

To test this possibility, the nam-1–pHygR sequence was am-plified from genomic DNA of namA by LA PCR with theprimer pair Nam5 and HG17 (see Fig. 5A for primers’ mappositions) and cloned into the NotI site of pBSD1. The result-ing plasmid, pBSD1-nam1-HygR, which covered the extendedregion of the altered nam-1 transcripts, was used to transformTCys(72)-1/�nam1-1. Transformation of TCys(72)-1/�nam1-1with pBSD1-nam1-Hyg resulted in colony morphology thatresembled that of namA (Fig. 9B) and was characterized by anirregular margin with mycelial protrusions and reduced pig-mentation. However, unlike in namA, no firm colony edge wasobserved in TCys(72)-1/�nam1-1 transformed with pBSD1-nam1-HygR. Fungal colonies of TCys(72)-1/�nam1-1 culturedin parallel were reduced in pigmentation but increased in aer-ial hyphae relative to TCys(72)-1 (Fig. 9B). These resultsshowed that the extension of transcripts of the nam-1 genecontributed to the phenotype of namA.

DISCUSSION

To understand virus-host interactions, we need to identifyhost factors involved in virus replication and macroscopicsymptom induction. Allen et al. (1) and Deng et al. (14) pre-viously used microarray analyses to identify hypovirus infec-tion-responsive genes in C. parasitica and subsequently vali-dated the approach. Here, we describe the development of agenetic screening protocol where a collection of fungal strainscarrying artificially induced mutations is mined for host factorsinvolved in hypovirus symptom expression. The method in-volved (i) the integration of the full-length cDNA to thegenomic RNA of a hypovirus single-amino-acid substitutionmutant, Cys(72), into the host genome, allowing the launch ofvirus infection in theoretically every fungal cell (7); (ii) mu-tagenesis by plasmid insertion; (iii) isolation of mutants ofinterests; and (iv) identification of causative mutated genes byI-PCR. This method enables the elimination of an inoculationstep that usually requires tremendous labor and time in screen-ing. Plasmid insertion for mutagenesis allows the direct iden-tification of mutated genes by I-PCR or plasmid rescue. Fur-thermore, this mutagenesis approach is important becauselittle is known about natural fungal host variations in symptomexpression that are governed genetically. This is in contrast toplant viral hosts for which a number of natural mutants withdifferent symptom responses to certain viruses and virus sus-ceptibility are reported at the variety or ecotype level (26).

The namA mutant isolated by this method manifested a

phenotype that was distinguishable from that of its parent,TCys(72)-1, while the virus-cured corresponding strains wereidentical in colony morphology [VC-namA compared to VC-TCys(72)-1] (Fig. 2A); that is, the irregular phenotype in namAwas produced only upon infection by Cys(72). More intrigu-ingly, the nature of the mutant phenotype is virus specific andvirus strain specific. When infected with CHV1-EP713 or itsmutant virus strain �p29 or �p40b, VC-namA shows symptomssimilar to those of namA but different from those of EP155 orVC-TCys(72)-1 infected with the respective virus strains. Col-onies of VC-namA infected with the hypoviral strains werecharacterized by irregular colony morphology with mycelialextensions and reduced pigmentation and sporulation (Table 2and Fig. 2A and D and 8) while retaining the ability to confersimilar levels of hypovirulence (Table 2) and support similarlevels of viral accumulation (see Fig. S4 in the supplementalmaterial). However, infection with �p69b results in the induc-tion of symptoms that are indistinguishable between VC-namAand VC-TCys(72)-1 (Table 2 and Fig. 2D), which suggests arequirement of p29 or p40 for phenotypic differences betweenthe two host backgrounds. This idea was proven to be correctby transformation analysis in which a transgenic supply ofeither p29 or p40 conferred the namA phenotype to VC-namAinfected by �p69 (Fig. 3). To our knowledge, this is the firstdescription of such a virus-specific phenotype in filamentousfungal hosts. �p69b and MyRV1 induce a similar set of phe-notypic alterations: fungal colonies infected by the two virusesmanifest severe reductions in the growth of aerial hyphae andattenuation of virulence with apple assay while retaining levelsof pigmentation and conidiation comparable to those exhibitedby virus-free colonies (18, 36). Previously, p29 and/or p40 wasshown to contribute to the restoration of the growth of aerialhyphae of fungal colonies infected by �p69b (36) and MyRV1(34), while their modes of action were considered to be dif-ferent from each other. These activities of p29 and p40 maybe related to the infection-specific phenotypic alterations inVC-namA that require the growth of aerial mycelia.

We demonstrated that plasmid insertion into nam-1 contrib-uted to the phenotype of namA by complementation and dis-ruption assays. The wild-type allele of nam-1 of EP155 com-plemented the colony morphology of namA, and the disruptionof the nam-1 gene in the TCys(72)-1 background resulted ininfection-specific phenotypic alterations similar to but distin-guishable from the phenotype exhibited by namA (Fig. 7C).The observation that disruptants of nam-1 in the TCys(72)-1background [TCys(72)-1/�nam1-1] did not show a phenotypeidentical to that of namA may be accounted for in several waysthat are not mutually exclusive. As expected from the insertionsite of plasmid pHygR (Fig. 5A), Northern blot analysis re-vealed that sizes of transcripts different from those in EP155were produced in namA (Fig. 9A). Sequence analysis of the 3�RACE clones showed that the nam-1 transcripts in namAcontained a deletion of 150 nt and an extension of approxi-mately 450 nt derived from pHygR at the 3� terminus thatcould direct the synthesis of a modified version of NAM-1 (seeFig. S3 in the supplemental material). This altered NAM-1 isconsidered to contribute to the irregular growth of mycelia andrepressed pigmentation and conidiation in namA by playing adifferent role from that of wild-type NAM-1. This notion issupported by the data shown in Fig. 9B in which transformants



http://jvi.asm.org/

Dow

nloaded from

http://jvi.asm.org/

of TCys(72)-1/�nam1-1 with the DNA fragment including thenam-1 and pHygR sequences of namA [TCys(72)-1/�nam1/nam1-HygR1] show a greater resemblance to namA in colonyphenotype. TCys(72)-1/�nam1/nam1-HygR1 exhibited still aslightly different phenotype from that of namA.

Given that the mutation of this gene results in a severereduction in pigmentation and sporulation in addition to ab-normal growth characteristics (Fig. 2 and Table 2), it is likelythat the expression of NAM-1 is involved in the alleviation ofsymptoms caused by hypoviruses. The molecular mechanismunderlying the alleviation of hypovirus-mediated symptom ex-pression has yet to be unraveled. However, some clues to itsfunctional role were supplied by computer-assisted analysis.A BLAST search identified sequence similarities betweenNAM-1 and counterparts in other filamentous fungi that com-prised 952 to 1,506 aa (see Fig. S2 in the supplemental mate-rial). A motif search showed that NAM-1 has a well-definedfunctional domain, CorA, as well as frequently observed con-sensus sequences. CorA is found at the C-terminal region of allthe NAM-1 counterparts in filamentous fungi (Fig. 6) and ineukaryotic and prokaryotic CorA family members. CorA mem-bers are one of the classes of Mg2� transport systems that serveas membrane-associated Mg2� transporters to maintain ho-meostatis of this most abundant divalent cation in cells (25).Typical CorA homologues are 300 to 500 aa in length with twodomains: a large N-terminal cytoplasmic domain and a shorterC-terminal transmembrane domain. Unlike other well-charac-terized CorA members, NAM-1 contains an extremely longN-terminal portion. S. cerevisiae also possesses two orthologsof CorA, ALR1 (encoding 859 aa) and ALR2 (encoding 858aa), that are responsible for aluminum resistance and that alsohave larger N-terminal extensions. The Alr proteins possiblymediate Mg2� uptake in Al3�-induced Mg2� deficiency, andtheir N-terminal regions are presumably involved in Al3� bind-ing (25). However, no discernible sequence similarities be-tween the N-terminal portions of NAM-1 and Alr1p, Alr2p, orother known sequences in the GenBank database were found,excluding the filamentous counterparts shown in Fig. 6 andFig. S2 in the supplemental material. A motif search identifiedan additional domain, DUF1777 (PFAM), in the N-terminalregion of NAM-1 whose functional role has yet to be assignedclearly (Fig. 6). The long N-terminal portion of NAM-1 islikely to be involved in unrecognized functions as in the yeastAlr proteins.

Mg2� is involved in numerous enzymatic reactions as a co-factor and phosphotransfer influencing structures of nucleicacids, proteins, and membranes and controlling the activities ofCa2� and K� channels in the plasma membrane (25, 33).Therefore, defects in CorA member proteins would results indetrimental effects on fundamental cellular functions, e.g., le-thality in pathogenic bacteria (29) and growth arrest in yeast(22). Although whether NAM-1 is a functional Mg2� trans-porter like other CorA members remains to be elucidatedbiochemically, the conservation of the CorA domain contain-ing the two transmembrane sequences and the GMN motif inNAM-1 (Fig. 6) strongly suggests that nam-1 encodes an Mg2�

uptake system most probably in the plasma membrane basedon the PSORT prediction. The pHygR integration into nam-1of namA causing a replacement of 18 aa with seven unrelatedamino acids (see Fig. S3 in the supplemental material) may

lead to the abnormal functionality of NAM-1. Therefore, thereduced pigmentation and sporulation found in namA andnam-1 disruptants (Fig. 8) may be a consequence of a pertur-bation in Mg2� homeostasis. Coordinated hyphal growth offungi is also regulated in a Ca2�-dependent manner (3). Ad-dressing how phenotypic alterations specific to infection byCHV1 viral strains occur in namA and nam-1 disruptants andwhether those alterations happen upon infection by other hy-povirus species like CHV2, CHV3, and CHV4 (28) is an in-teresting challenge that warrants further investigation.

ACKNOWLEDGMENTS

This work was supported in part by the Okayama University COEprogram Establishment of Plant Health Science to N.S.

We are grateful to Donald L. Nuss for his generous gift of a numberof plasmid clones and fungal strains used in this study. We thankBradley I. Hillman and Shin Kasahara for the generous gift of plasmidspBSD1 and pCPXBn1 and the 9B21 fungal strains. We thank all ofthese colleagues for fruitful discussions.

REFERENCES

1. Allen, T. D., A. L. Dawe, and D. L. Nuss. 2003. Use of cDNA microarrays tomonitor transcriptional responses of the chestnut blight fungus Cryphonectriaparasitica to infection by virulence-attenuating hypoviruses. Eukaryot. Cell2:1253–1265.

2. Anagnostakis, S. L. 1982. Biological control of chestnut blight. Science215:466–471.

3. Brand, A., S. Shanks, V. M. S. Duncan, M. Yang, K. Mackenzie, and N. A. R.Gow. 2007. Hyphal orientation of Candida albicans is regulated by a calcium-dependent mechanism. Curr. Biol. 17:347–352.

4. Chen, B., G. H. Choi, and D. L. Nuss. 1993. Mitotic stability and nuclearinheritance of integrated viral cDNA in engineered hypovirulent strains ofthe chestnut blight fungus. EMBO J. 12:2991–2998.

5. Chen, B., and D. L. Nuss. 1999. Infectious cDNA clone of hypovirus CHV1-Euro7: a comparative virology approach to investigate virus-mediated hypo-virulence of the chestnut blight fungus Cryphonectria parasitica. J. Virol.73:985–992.

6. Choi, G. H., D. M. Pawlyk, and D. L. Nuss. 1991. The autocatalytic proteasep29 encoded by a hypovirulence-associated virus of the chestnut blight fun-gus resembles the potyvirus-encoded protease HC-Pro. Virology 183:747–752.

7. Choi, G. H., and D. L. Nuss. 1992. A viral gene confers hypovirulence-associated traits to the chestnut blight fungus. EMBO J. 11:473–477.

8. Choi, G. H., and D. L. Nuss. 1992. Hypovirulence of chestnut blight fungusconferred by an infectious viral cDNA. Science 257:800–803.

9. Churchill, A. C. L., L. M. Ciufetti, D. R. Hansen, H. D. Van Etten, and N. K.Van Alfen. 1990. Transformation of the fungal pathogen Cryphonectria para-sitica with a variety of heterologous plasmids. Curr. Genet. 17:25–31.

10. Craven, M. G., D. M. Pawlyk, G. H. Choi, and D. L. Nuss. 1993. Papain-likeprotease p29 as a symptom determinant encoded by a hypovirulence-asso-ciated virus of the chestnut blight fungus. J. Virol. 67:9513–9521.

11. Dawe, A. L., and D. L. Nuss. 2001. Hypoviruses and chestnut blight: exploit-ing viruses to understand and modulate fungal pathogenesis. Annu. Rev.Genet. 35:1–29.

12. Dawe, A. L., G. C. Segers, T. D. Allen, V. C. McMains, and D. L. Nuss. 2004.Microarray analysis of Cryphonectria parasitica G and G-signalling path-ways reveals extensive modulation by hypovirus infection. Microbiology 150:4033–4043.

13. Dawe, A. L., V. C. McMains, M. Panglao, S. Kasahara, B. Chen, and D. L.Nuss. 2003. An ordered collection of expressed sequences from Cryphonec-tria parasitica and evidence of genomic microsynteny with Neurospora crassaand Magnaporthe grisea. Microbiology 149:2373–2384.

14. Deng, F., T. D. Allen, and D. L. Nuss. 2007. Ste12 transcription factorhomologue CpST12 is down-regulated by hypovirus infection and requiredfor virulence and female fertility of the chestnut blight fungus Cryphonectriaparasitica. Eukaryot. Cell 6:235–244.

15. Fulbright, D. W. 1984. Effect of eliminating dsRNA in hypovirulent Endothiaparasitica. Phytopathology 74:722–724.

16. Gao, S., G. H. Choi, L. Shain, and D. L. Nuss. 1996. Cloning and targeteddisruption of enpg-1, encoding the major in vitro extracellular endopoly-galacturonase of the chestnut blight fungus, Cryphonectria parasitica. Appl.Environ. Microbiol. 62:1984–1990.

17. Hillman, B. I., R. Shapira, and D. L. Nuss. 1990. Hypovirulence-associatedsuppression of host functions in Cryphonectria parasitica can be partiallyrelieved by high light intensity. Phytopathology 80:950–956.

18. Hillman, B. I., S. Supyani, H. Kondo, and N. Suzuki. 2004. A reovirus of the



http://jvi.asm.org/

Dow

nloaded from

http://jvi.asm.org/

fungus Cryphonectria parasitica that is infectious as particles and related tothe Coltivirus genus of animal pathogens. J. Virol. 78:892–898.

19. Hillman, B. I., and N. Suzuki. 2004. Viruses in the chestnut blight fungus.Adv. Virus Res. 63:423–472.

20. Jacob-Wilk, D., M. Turina, and N. K. Van Alfen. 2006. Mycovirus Cryphonec-tria hypovirus 1 elements cofractionate with trans-Golgi network membranesof the fungal host Cryphonectria parasitica. J. Virol. 80:6588–6596.

21. Kasahara, S., P. Wang, and D. L. Nuss. 2000. Identification of bdm-1, a geneinvolved in G protein -subunit function and -subunit accumulation. Proc.Natl. Acad. Sci. USA 97:412–417.

22. Lee, J., and R. C. Gardner. 2006. Residues of the yeast ALRI protein that iscritical for magnesium uptake. Curr. Genet. 49:7–20.

23. Lin, H., X. Lan, H. Liao, T. B. Parsley, D. L. Nuss, and B. Chen. 2007.Genome sequence, full-length infectious cDNA clone, and mapping of viraldouble-stranded RNA accumulation determinant of hypovirus CHV1-EP721. J. Virol. 81:1813–1820.

24. Lunin, V. V., E. Dobrovetsky, G. Khutoreskaya, R. Zhang, A. Joachimiak,D. A. Doyle, A. Bochkarev, M. E. Maguire, A. M. Edwards, and C. M. Koth.2006. Crystal structure of the CorA Mg2� transporter. Nature 440:833–837.

25. Maguire, M. E. 2006. Magnesium transporter: properties, regulation andstructure. Front. Biosci. 11:3149–3163.

26. Maule, A. J., C. Caranta, and M. I. Boultion. 2007. Source of naturalresistance to plant viruses: status and prospects. Mol. Plant Pathol. 8:223–231.

27. Nuss, D. L. 2005. Hypovirulence: mycoviruses at the fungal-plant interface.Nat. Rev. Microbiol. 3:632–642.

28. Nuss, D. L., B. I. Hillman, D. Rigling, and N. Suzuki. 2005. Family Hypo-viridae, p. 597–601. In C. M. Fauquet et al., Virus taxonomy. Eighth Reportof the International Committee for the Taxonomy of Viruses. AcademicPress, San Diego, CA.

29. Pfeiffer, J., J. Guhl, B. Waidner, M. Kist, and S. Bereswill. 2002. Magnesiumuptake by CorA is essential for viability of the gastric pathogen Helicobacterpylori. Infect. Immun. 70:3930–3934.

30. Sambrook, J., and D. W. Russell. 2001. Molecular cloning: a laboratorymanual, 3rd ed. Cold Spring Harbor Laboratory Press, Cold Spring Har-bor, NY.

31. Segers, G. C., R. van Wezel, X. Zhang, Y. Hong, and D. L. Nuss. 2006.Hypovirus papain-like protease p29 suppresses RNA silencing in the naturalfungal host and in a heterologous plant system. Eukaryot. Cell 5:896–904.

32. Segers, G. C., X. Zhang, F. Deng, Q. Sun, and D. L. Nuss. 2007. Evidencethat RNA silencing functions as an antiviral defense mechanism in fungi.Proc. Natl. Acad. USA 104:12902–12906.

33. Shabala, S., and Y. Hariadi. 2005. Effects of magnesium availability on theactivity of plasma membrane ion transporters and light-induced responsesfrom broad bean leaf mesophyll. Planta 221:56–65.

34. Sun, L.-Y., D. L. Nuss, and N. Suzuki. 2006. Synergism between a mycoreo-virus and a hypovirus mediated by the papain-like protease p29 of theprototypic hypovirus CHV1-EP713. J. Gen. Virol. 87:3703–3714.

35. Suzuki, N., B. Chen, and D. L. Nuss. 1999. Mapping of a hypovirus p29protease symptom determinant domain with sequence similarity to potyvirusHC-Pro protease. J. Virol. 73:9478–9484.

36. Suzuki, N., and D. L. Nuss. 2002. The contribution of p40 to hypovirus-mediated modulation of fungal host phenotype and viral RNA accumulation.J. Virol. 76:7747–7759.

37. Suzuki, N., K. Maruyama, M. Moriyama, and D. L. Nuss. 2003. Hypoviruspapain-like protease p29 functions in trans to enhance viral double-strandedRNA accumulation and vertical transmission. J. Virol. 77:11697–11707.

38. Suzuki, N., L. M. Geletka, and D. L. Nuss. 2000. Essential and dispensablevirus-encoded elements revealed by efforts to develop hypoviruses as geneexpression vector. J. Virol. 74:7568–7577.

39. Suzuki, N., S. Supyani, K. Maruyama, and B. I. Hillman. 2004. Completegenome sequence of Mycoreovirus 1/Cp9B21, a member of a new genuswithin the family Reoviridae, from the chestnut blight fungus Cryphonectriaparasitica. J. Gen. Virol. 85:3437–3448.



http://jvi.asm.org/

Dow

nloaded from

http://jvi.asm.org/

Documents

A Host Factor Involved in Hypovirus Symptom Expression in the