Characterization of an RNA-Dependent RNA ... - Journal of Virology

JOURNAL OF VIROLOGY, Mar. 1997, p. 2264–2269 0022-538X/97/$04.0010 Copyright q 1997, American Society for Microbiology

Vol. 71, No. 3

Characterization of an RNA-Dependent RNA Polymerase Activity Associated with La France Isometric Virus† MICHAEL M. GOODIN,‡ BETH SCHLAGNHAUFER, TIFFANY WEIR,

AND

C. PETER ROMAINE*

Department of Plant Pathology, Pennsylvania State University, University Park, Pennsylvania 16802 Received 12 August 1996/Accepted 19 November 1996

Purified preparations of La France isometric virus (LIV), an unclassified, double-stranded RNA (dsRNA) virus of Agaricus bisporus, were associated with an RNA-dependent RNA polymerase (RDRP) activity. RDRP activity cosedimented with the 36-nm isometric particles and genomic dsRNAs of LIV during rate-zonal centrifugation in sucrose density gradients, suggesting that the enzyme is a constituent of the virion. Enzyme activity was maximal in the presence of all four nucleotides, a reducing agent (dithiothreitol or b-mercaptoethanol), and Mg21 and was resistant to inhibitors of DNA-dependent RNA polymerases (actinomycin D, a-amanitin, and rifampin). The radiolabeled enzyme reaction products were predominantly (95%) singlestranded RNA (ssRNA) as determined by cellulose column chromatography and ionic-strength-dependent sensitivity to hydrolysis by RNase A. Three major size classes of ssRNA transcripts of 0.95, 1.3, and 1.8 kb were detected by agarose gel electrophoresis, although the transcripts hybridized to all nine of the virion-associated dsRNAs. The RNA products synthesized in vitro appeared to be of a single polarity, as they hybridized to an ssDNA corresponding to one strand of a genomic dsRNA and not to the complementary strand. Similarly, reverse transcription-PCR with total cellular ssRNA as a template and strand-specific primers targeting a genomic dsRNA during synthesis of cDNA suggested that only the coding strand was transcribed in vivo. Our data indicate that the RDRP activity associated with virions of LIV is probably a transcriptase engaged in the synthesis of ssRNA transcripts corresponding to each of the virion-associated dsRNAs. dsRNA strands remains intact following transcription. In contrast, a semiconservative mechanism is utilized by fungal viruses in the Partiviridae, in which one of the parental strands of dsRNA is displaced by the nascent strand (3, 4, 22). In this study, we report on the properties of an RNA-dependent RNA polymerase (RDRP) activity with respect to requirements for catalysis, nature of the RNA products synthesized in vitro, and association with virions of LIV.

Sinden and Hauser (31) were the first to describe a serious infectious disorder of the cultivated mushroom Agaricus bisporus, known as La France disease. Today the disease is considered an important limiting factor worldwide in the commercial cultivation of mushrooms (10). The most diagnostic symptoms of the disease include the degeneration and death of the mycelium, a delayed appearance of mushrooms, the development of deformed mushrooms, premature opening of the veil, and a reduced yield. The viral nature of La France disease was proposed in 1962 by Hollings (16) following the observation of virus-like particles (VLPs) in extracts from diseased basidiocarps. Since then, a wealth of correlative evidence implicating La France isometric virus (LIV) in the etiology of the disease has accumulated (6, 10, 12, 23, 25–27, 33, 34). LIV has 36-nm isometric particles that encapsidate nine double-stranded RNAs (dsRNAs) of 0.8 to 3.8 kb (12, 33, 34). Three of the dsRNAs have been completely sequenced, although functions have not been ascribed to the putatively encoded polypeptides (15). Virions of LIV are composed of three structural polypeptides which, depending on the method of purification, are estimated to be 63, 66, and 129 kDa (12) or 65, 115, and 120 kDa (33). The complexity of the genome renders LIV distinct from other dsRNA viruses belonging to the recognized taxonomic groups (22). Information on replication strategy can be used to support the taxonomic position of viruses. For example, fungal viruses classified in the Totiviridae use a conservative mode of replication (11, 22, 30) such that the integrity of the parental

MATERIALS AND METHODS Source of tissue. Healthy and diseased mushrooms were collected at commercial sites and stored at 2808C. Purification of virus. (i) Method I. A 100-g sample of mushroom tissue was homogenized in a Waring blender with 300 ml of a solution containing 50 mM Tris-HCl, 1 mM EDTA, 0.1% b-mercaptoethanol (BME), and 0.1% Nonidet P-40, pH 8.0. The homogenate was clarified by centrifugation at 6,000 3 g for 20 min. The resultant supernatant was collected and incubated for 3 to 4 weeks at 48C, after which time a precipitate enriched in virus was collected by centrifugation at 3,000 3 g for 10 min. The pellet was resuspended in a minimal volume of the supernatant and stored at 2808C. (ii) Method II. More highly purified preparations of LIV were obtained by a modification of the subcellular fractionation procedures described by Rogers et al. (24) and Morten and Hicks (21). Mushrooms (100 g) were homogenized in a Waring blender for 4 min with 200 ml of 100 mM Tris-HCl–0.2 mM EDTA, pH 7.5 (Tris buffer), containing 15% sucrose. The homogenate was mixed with an equal volume of Tris buffer containing 1% (wt/vol) sucrose and centrifuged at 1,300 3 g for 10 min at 48C. The supernatant was centrifuged at 14,000 3 g for 30 min, and the pellet was resuspended in 4 ml of Tris buffer with 20% sucrose, extracted with 0.1 volume of chloroform, and centrifuged at 10,000 3 g for 10 min. The supernatant was centrifuged at 50,000 3 g for 30 min, and the virusenriched pellets were resuspended in 4 ml of Tris buffer with 20% sucrose. Two milliliters of the virus preparation was layered onto a sucrose step gradient consisting of 18 ml each of 60 and 35% sucrose prepared in Tris buffer. Gradients were centrifuged at 64,000 3 g for 90 min at 48C and partitioned into 3-ml fractions (model 640 density gradient fractionator; ISCO, Lincoln, Nebr.). Each fraction was analyzed for RDRP activity, dsRNA, protein, and VLPs. Polymerase assays. The standard reaction cocktail contained, in a final volume of 50 ml, 50 mM Tris-HCl (pH 8.0), 10 mM MgCl2, 5 mM dithiothreitol (DTT), 40 mg of actinomycin D per ml, 60 U of RNasin (Pharmacia, Piscataway, N.J.), 1 mM (each) ATP, CTP, and GTP, 50 mCi of [3H]UTP (35 to 50 Ci/mmol; ICN Biomedical, Irvine, Calif.) or [a-32P]UTP (.600 Ci/mmol; ICN), and, unless stated otherwise, 2 ml of virus purified by method I. In some experiments, the

* Corresponding author. Mailing address: Department of Plant Pathology, 209 Buckhout Laboratory, Pennsylvania State University, University Park, PA 16802. Phone: (814) 865-7132. Fax: (814) 8632065. E-mail: cpr2@psu.edu. † Department of Plant Pathology contribution 2043. ‡ Present address: Department of Plant Biology, University of California, Berkeley, Berkeley, CA 94720. 2264

VOL. 71, 1997

LIV-ASSOCIATED RNA POLYMERASE

2265

FIG. 1. Analysis of extracts of healthy and diseased mushrooms for LIV by agarose gel electrophoresis of dsRNA (A) and SDS-PAGE of proteins (B). Clarified homogenates (supernatant obtained at 6,000 3 g) of healthy and diseased mushrooms were incubated for 4 weeks at 48C and then centrifuged at 3,000 3 g. The resultant supernatants and precipitates were analyzed for the LIV dsRNAs and virion-specific polypeptides. C, clarified homogenate; S, supernatant; P, precipitate. The sizes of the dsRNAs and polypeptides are indicated in the left and right margins, respectively.

standard reaction cocktail was amended with 50 mg of a-amanitin per ml, 50 mg of rifampin per ml, 2 to 10 mM BME, or 10 mM MnCl2. Typically, the enzyme reaction was allowed to proceed for 1.5 h at 378C and was terminated by spotting 10 ml of the reaction cocktail onto 2.3-cm-diameter filter paper discs (3MM; Whatman Ltd., Maidstone, England). Discs were washed batchwise five times for 10 min each time in cold 5% trichloroacetic acid (TCA) containing 0.02% uracil and then washed twice for 5 min each in 95% ethanol. Following washing, discs were air dried and heated for 20 min at 508C in capped scintillation vials containing 300 ml of a 90% aqueous solution of NCS tissue solubilizer (Amersham, Arlington Heights, Ill.). Ten milliliters of Ecoscint (National Diagnostics, Manville, N.J.) was added to each vial, and radioactivity was determined by liquid scintillation counting on an LS 7000 instrument (Beckman, Fullerton, Calif.). TCA-precipitable radioactivity was determined by Cerenkov counting when [a-32P]UTP was used in polymerase assays. For sizing the polymerase reaction products, virus purified by method II was used as a source of enzyme, the standard reaction cocktail was amended with 0.1 mM UTP and 50 mCi of [a-32P]UTP, and reactions were carried out for 3 to 12 h at 308C. Extraction of RDRP reaction products. Polymerase reaction products were extracted with phenol (18) and resuspended in water or 10 mM Tris-HCl–1 mM EDTA, pH 7.5 (TE). Electrophoresis of nucleic acid. RDRP reaction products were analyzed on formaldehyde-denaturing 1.2% agarose gels (18). LIV dsRNAs were resolved on either nondenaturing 1% agarose gels prepared in Tris-acetate-EDTA buffer (18) or formaldehyde-denaturing agarose gels. Electrophoresis of protein. Polypeptides were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) with a 7.5% stacking gel as described previously (12). Cellulose column chromatography. RDRP reactions were terminated by the addition of 1 volume of 23 STE (13 STE is 50 mM Tris, 0.1 M NaCl, and 1 mM EDTA, pH 7.0) containing 1% SDS. The reaction mixture was incubated for 10 min at 658C, and nucleic acids were extracted with phenol (18) and resuspended in water or TE. The nucleic acid preparation was modified to include STE containing 35% ethanol (STE-35) by the addition of a 53 STE stock and absolute ethanol and loaded onto a column containing 1 g of CF-11 cellulose (Whatman) equilibrated with STE-35. The column was washed with 100 ml of STE-35 to remove oligonucleotides and unincorporated radionucleotides, then with 36 ml of STE containing 15% ethanol to release single-stranded RNA (ssRNA), and finally with 36 ml of STE to elute dsRNA. Nuclease sensitivity assays. The sensitivity of the polymerase reaction products to hydrolysis by DNase was determined in a solution containing 10 mM Tris-HCl, 10 mM MgCl2, 2 mM CaCl2 [pH 7.5], 0.6 U of RNasin per ml, and 1 U of RNase-free DNase I (Promega, Madison, Wis.). RNase sensitivity trials were carried out in the presence of 1 mg of RNase A (Sigma Chemical Co., St. Louis, Mo.) per ml under high (0.3 M NaCl)-salt and low (0.03 M NaCl)-salt conditions (34). Nuclease reactions were conducted in a final volume of 100 ml for 30 min at 378C. Synthesis of cDNA. The LIV-specific 3.8-kb dsRNA was isolated from diseased basidiocarps by phenol extraction and cellulose column chromatography (20, 34) and purified by agarose gel electrophoresis (18). The gel-purified dsRNA was denatured with methylmercury hydroxide (1) and converted into bluntended cDNA according to the method of Gubler and Hoffman (14) as described by Klickstein et al. (17), with the only modification being the omission of RNase A in the blunt-ending reaction. Blunt-ended cDNAs were ligated to SmaIdigested pBlueScript KS1 phagemid and transformed into Escherichia coli SURE according to the manufacturer’s instructions (Stratagene, La Jolla, Calif.). Putative transformants were screened for DNA inserts by the alkaline lysis procedure (18). A 1.7-kb cDNA corresponding to the 3.8-kb dsRNA was con-

firmed by Northern hybridization to LIV-associated dsRNAs (7). Subsequently, strand-specific ss-cDNAs were rescued by phage R408 from E. coli XL1-Blue (Stratagene) that had been transformed with the 1.7-kb cDNA subcloned in the same orientation at the SmaI site of pBluescript KS1 or KS2 (29). Northern hybridization. RNA was capillary transferred to Nytran membranes (Schleicher & Schuell, Keene, N.H.) in 103 SSC (103 SSC is 1.5 M NaCl plus 0.15 sodium citrate) from either formaldehyde-denaturing (18) or nondenaturing (19) agarose gels. Prehybridization and hybridization were carried out at 428C in a solution containing 50% formamide, 63 SSC, 53 Denhardt’s solution (0.1% Ficoll, 0.1% polyvinylpyrrolidone, 0.1% bovine serum albumin), 0.5% SDS, and 100 mg of sheared salmon sperm DNA per ml. Blots were prehybridized for at least 2 h and hybridized for a minimum of 16 h with fresh hybridization solution containing the in vitro-synthesized [a-32P]UTP-labeled polymerase reaction products. Following hybridization, blots were washed twice at room temperature, twice at 428C, and twice again at room temperature. Each wash was done for 15 min in 0.13 SSC containing 0.1% SDS, except for the final wash, in which SDS was excluded. Southern hybridization. Rescued ss-cDNAs corresponding to the complementary strands of the 1.7-kb cDNA to the LIV 3.8-kb dsRNA were separated by electrophoresis on 1% agarose gels, denatured in 1.5 M NaCl–0.5 M NaOH, neutralized in 1 M Tris-HCl–1.5 M NaCl, pH 7.4, and capillary transferred in 103 SSC to Nytran membranes. The conditions for hybridization and washing were as described for Northern blots. RT-PCR. Reverse transcription (RT)-PCR with primers targeting the putative coding strand (reverse primer) or the noncoding strand (forward primer) of the LIV 1.3-kb dsRNA was carried out essentially as described by Romaine and Schlagnhaufer (27). The only modification of the procedure was that the forward and reverse primers were used separately to initiate synthesis of cDNA to a total cellular ssRNA fraction prepared from diseased mushrooms by the acid guani-

FIG. 2. Kinetics of RDRP activity in the precipitates from diseased (LIVenriched) and healthy tissues. Activity is defined as the incorporation of 3 [ H]UTP into TCA-precipitable product. Time course experiments were carried out with the standard reaction cocktail described in Materials and Methods. Data points represent the means of two identical experiments.

2266

GOODIN ET AL.

J. VIROL.

TABLE 1. Characteristics of the RNA polymerase activitya Relative incorporation (%)

Modification of complete medium

None................................................................................................ 100b 2Actinomycin D ........................................................................... 109 1a-Amanitin (50 mg/ml) .............................................................. 88 1Rifampin (50 mg/ml).................................................................. 87 2Mg, 1Mn (10 mM).................................................................... ,1 2Mg ................................................................................................ ,1 2ATP, 2CTP, 2GTP .................................................................. ,1 2DTT ............................................................................................. 14 2DTT, 1BME (2 mM) ............................................................... 42 2DTT, 1BME (5 mM) ............................................................... 62 2DTT, 1BME (10 mM) ............................................................. 84 a All reactions were carried out at 378C for 90 min in 50 ml of a standard reaction cocktail with the addition (1) or omission (2) of components as indicated. Values are the means of two experiments. b Incorporation in complete medium was 43,000 cpm.

dinium-phenol method (5) and cellulose column chromatography. For subsequent PCR amplification, both primers were included in the reaction mixture.

FIG. 3. Characterization of the LIV polymerase products synthesized in vitro by cellulose column chromatography. Phenol-extracted 3H-labeled polymerase product (53,000 cpm) was loaded onto a column containing 1 g of cellulose resin. The column was washed exhaustively with STE-35 to remove unincorporated nucleotides and oligonucleotides. ssRNA was eluted from the column by washing with STE-15, and dsRNA was eluted by washing with STE. Six 3-ml fractions were collected for each eluant, and radioactivity was determined for 1 ml of each fraction by liquid scintillation counting.

RESULTS Analysis of LIV-enriched precipitates. Precipitates (3,000 3 g) derived from clarified homogenates of diseased mushrooms that had been incubated for 4 weeks at 48C (method I) were highly enriched in LIV. The LIV-associated dsRNAs in the range of 0.8 to 3.8 kb were detected by agarose gel electrophoretic analysis of clarified homogenates as well as the derived precipitates (Fig. 1). The lack of detectable dsRNA in the postprecipitation supernatant suggested quantitative recovery of virus from solution. No dsRNA was recovered from comparable preparations of healthy mushrooms. Similarly, SDSPAGE analysis revealed that precipitates from La France disease-affected mushrooms contained primarily the three LIV-specific polypeptides of 63, 66, and 129 kDa (Fig. 1). Precipitates prepared from clarified homogenates of healthy mushrooms lacked the three polypeptides and, in fact, showed a paucity of protein. Electron microscopic examination revealed that the precipitates from diseased mushrooms contained aggregates of 36-nm isometric particles characteristic of LIV, whereas those from healthy mushrooms were devoid of VLPs but contained amorphous material (13). In vitro polymerase activity. RDRP activity was detected in LIV-enriched precipitates prepared from diseased mushrooms, whereas no appreciable enzyme activity was detected in comparable preparations from healthy mushrooms (Fig. 2). Incorporation of radiolabeled UMP increased steadily for 3 h, after which time a decrease in the accumulation of TCA-

precipitable product occurred, presumably due to the activity of endogenous ribonucleases. Requirements for catalysis. The RDRP activity was largely insensitive to polymerase inhibitors actinomycin D, a-amanitin, and rifampin, suggesting that the observed activity was not catalyzed by cellular transcriptases (Table 1). Maximal polymerase activity required the presence of Mg21, which could not be replaced by Mn21, the four nucleoside triphosphates, and a reducing agent, either DTT or BME. Characterization of polymerase products. The polymerase products synthesized in vitro by the RDRP activity were insen-

TABLE 2. Nuclease sensitivity of the in vitro-synthesized polymerase producta Treatment

Relative radioactivity (%)

Water.............................................................................................. 100 DNase I.......................................................................................... 72 RNase A (0.03 M NaCl) ............................................................. 5 RNase A (0.3 M NaCl) ............................................................... 16 a Aliquots of phenol-extracted radiolabeled polymerase product (5,600 cpm) were treated for 30 min at 378C with water, DNase I, RNase A under low-salt conditions (0.03 M NaCl), and RNase A under high-salt conditions (0.3 M NaCl). Values are the means of two experiments.

FIG. 4. Northern hybridization of [a-32P]UTP-labeled polymerase reaction products to the disease-specific (D) LIV dsRNAs. dsRNA isolated from healthy (H) mushrooms was included as a control. The dsRNAs were separated under nondenaturing conditions and transferred to nylon membranes by capillary blotting in 103 SSC following incubation in a solution containing 50% formamide, 9% formaldehyde, and 40 mM Tris acetate (pH 8.6) containing 2 mM EDTA for 30 min at 558C. Hybridization conditions were as described in Materials and Methods. Shown are the original ethidium bromide (EtBr)-stained gel (A) and autoradiograms of the radioactive membranes exposed for 2 (B) and 24 (C) h. The sizes of the dsRNAs are indicated in the left margin.

VOL. 71, 1997


2267

FIG. 5. Southern hybridization of [a-32P]UTP-labeled polymerase product to complementary strands of a 1.7-kb cDNA clone corresponding to the 3.8-kb dsRNA. The cDNA was subcloned into the pBluescript vectors KS1 and KS2 in the same orientation with respect to the multiple-cloning site, allowing recovery of the complementary strands by phage R408-mediated rescue. Equivalent amounts of each strand of cDNA were electrophoresed in nondenaturing agarose gels and visualized by ethidium bromide staining (A) and autoradiography after hybridization with the ssRNA probes (B). The size of the helper phage DNA is indicated in the left margin. The relative positions of the phage and LIV-specific DNAs are indicated in the right margin.

sitive to hydrolysis by DNase I but were hydrolyzed by RNase A under high- and low-salt conditions, indicating that the products were ssRNA (Table 2). Moreover, approximately 95% of the radiolabeled products were eluted from a cellulose column in the ssRNA fraction, with the remainder being eluted in the dsRNA fraction (Fig. 3). Northern hybridization analysis disclosed that transcripts corresponding to all nine of the LIV-associated dsRNAs were transcribed in vitro, with the hybridization signal being most intense for the 1.3-kb and 2.8- to 3.1-kb dsRNAs (Fig. 4). In contrast, the labeled transcription products failed to hybridize to a 2.4-kb dsRNA found in healthy mushrooms. Southern

FIG. 6. Verification by RT-PCR of the single-stranded nature of the LIV transcripts synthesized in vivo. Total cellular RNA isolated from diseased mushrooms was fractionated into ssRNA and dsRNA by cellulose column chromatography. PCR primers were specific for the 1.3-kb LIV dsRNA. Either the forward or the reverse primer, which hybridized to the noncoding or the coding strand, respectively, was used to prime cDNA synthesis. The second primer was then added to the reaction mixture for PCR amplification. The expected 128-bp amplicon was generated in reactions where the reverse primer was used to prime cDNA synthesis. A 123-bp marker is shown.

FIG. 7. Separation of the in vitro-synthesized transcription products by electrophoresis on denaturing formaldehyde-agarose gels. Sucrose gradient-purified virus (method II) was incubated in a standard polymerase reaction cocktail containing 0.1 mM UTP and 50 mCi of [a-32P]UTP at 218C for 12 h. Following phenol extraction and ethanol precipitation, transcripts were separated by electrophoresis and then transferred to Nytran membranes by capillary blotting. Membranes were dried and exposed to X-ray film at 2808C. The sizes (in kilobases) of the transcripts are indicated in the left margin.

hybridization of the RDRP products to one strand of an ssDNA corresponding to the 3.8-kb LIV dsRNA, but not to the complementary strand, demonstrated that only transcripts of a single polarity were transcribed (Fig. 5). An experiment was carried out to determine if the single polarity transcripts synthesized in vitro by the RDRP activity reflected the nature of the products synthesized in vivo. Total cellular ssRNA from diseased mushrooms served as a template for RT-PCR in which primers directed against the putative coding or noncoding strand of the 1.3-kb dsRNA were used individually to prime RT prior to PCR amplification with both primers. The predicted PCR product of 128 bp was generated only when the reverse primer, which annealed to the putative coding strand, was used to initiate synthesis of cDNA (Fig. 6). Sizing of RNA products. Several high-molecular-weight RNA products were synthesized in vitro with sucrose gradient-purified viral preparations (method II) as a source of enzyme in the RDRP reaction mixture. Three major size classes of ssRNA with estimated molecular sizes of 0.95, 1.3, and 1.8 kb, as well as higher-molecular-weight ssRNA, were observed on autoradiograms of electrophoretically separated transcripts (Fig. 7). Association of RDRP activity with LIV. Following rate-zonal centrifugation in sucrose density gradients, RDRP activity was associated with the gradient fraction containing the LIV-associated dsRNAs (fraction 4 [Fig. 8]) and the virion-specific polypeptides and 36-nm isometric VLPs (13). The corresponding fractions of gradients containing extracts from healthy mushrooms were devoid of RDRP activity, dsRNA, and VLPs. DISCUSSION We have characterized an RDRP activity associated with virions of LIV, the suspected etiologic agent of La France disease of A. bisporus. The requirements for catalysis of the LIV-associated enzyme were generally similar to those described for other viral RDRPs (2, 11, 35). It should be noted that Sriskantha et al. (32) associated an RDRP activity with

2268

GOODIN ET AL.

J. VIROL.

FIG. 8. Association of the polymerase with virions of LIV following rate-zonal centrifugation in sucrose density gradients. The UV absorbance profiles of discontinuous sucrose gradients containing preparations from healthy (dashed line) or diseased (solid line) mushrooms are shown. Gradient fractions were assayed for polymerase activity (histogram in panel A) and LIV-specific dsRNAs (B). Polymerase activity cosedimented with the LIV dsRNAs (fraction 4). The sizes (in kilobases) of the dsRNAs are shown in the left margin of panel B.

35-nm isometric VLPs from La France disease-affected mushrooms. However, this polymerase synthesized in vitro two dsRNAs that were homologous to and the same size as the genomic dsRNAs of 2.1 and 6.5 kb. Therefore, the viral polymerases, and hence the viruses themselves (described in the present study and by Sriskantha and coworkers), are distinct. All lines of evidence support the role of LIV in the etiology of La France disease (6, 10, 12, 23, 25–27, 33, 34), whereas the biological relevance of the 35-nm virus characterized by Sriskantha et al. (32) has not been determined. We found that RDRP activity cosedimented with 36-nm isometric particles of LIV during rate-zonal centrifugation in sucrose density gradients, demonstrating that the enzyme is a constituent of the virion. We also established that under conditions of centrifugation that were identical to those used by Morten and Hicks (21), mitochondria sedimented to the same position as LIV particles that were obtained by chloroform extraction of preparations from diseased mushrooms (13). Therefore, the association of LIV dsRNA with mitochondria, as proposed by Morten and Hicks (21), remains equivocal. We determined that three major size classes of ssRNA transcripts with estimated molecular sizes of 0.95, 1.3, and 1.8 kb, as well as higher-molecular-weight ssRNA, were synthesized in vitro by the virion-associated RDRP activity. Assuming that the observed hybridization signal is proportional to the level of transcription, the predominant transcripts corresponded to the 1.3-kb and 2.8- to 3.1-kb dsRNAs, although transcripts of all of the dsRNAs were detected. The RDRP specifically catalyzed LIV RNA, because the enzyme products failed to hybridize to an endogenous 2.4-kb dsRNA associated with fungal vesicles from healthy mushrooms (28). We are uncertain at this time if our inability to synthesize discrete full-length transcripts of each of the LIV dsRNAs can be attributed to contaminating ribonucleases or to suboptimal reaction conditions. The inability to synthesize full-length transcripts with isolated viral polymerases has been noted elsewhere. White and Wang (35) observed heterogeneity in the sizes of the transcripts synthesized by the RDRP associated with the Giardia lamblia virus. In the case of the RDRP activities associated with yeast L-A virus, full-length transcription requires a host factor or specific chemical agents (8, 9). Conservative transcription occurs in members of the Totiviridae (8, 11, 22, 30) and is characterized by the accumulation of radiolabel in ssRNA molecules while the integrity of the parental dsRNA template is preserved. Members belonging to

a second family of dsRNA viruses known as the Partiviridae replicate by way of a semiconservative mechanism whereby nascent transcripts remain hybridized to the parental template while the complementary parental strand is displaced (3, 4, 22). Telltale of this strategy for replication is the generation of labeled dsRNA molecules during transcription. We found that the preponderance of the labeled products synthesized in vitro were ssRNA transcripts and that, in the case of at least one of the genomic dsRNAs, they were of a single polarity, presumably positive strand. Moreover, RT-PCR analysis revealed that for one of the genomic dsRNAs, transcripts corresponding only to the positive strand were synthesized in vivo. Taken together, these findings seem to indicate that free negative strands are not generated during replication but rather that positive strands are synthesized and, ultimately, encapsidated and converted to dsRNAs within the virion. However, our data do not directly address the question of whether transcription occurs by a conservative or semiconservative mechanism, because although traces of the in vitro-synthesized product were eluted in the dsRNA fraction, we could not detect label in dsRNA molecules by nondenaturing gel electrophoresis in enzyme time course studies (13). We do not know whether the labeled ssRNAs obscured small amounts of dsRNA that may have been synthesized or whether labeled dsRNA was absent altogether. Further, polymerase assays carried out in the presence of RNase A did not yield sufficient precipitable labeled product to be detected by gel electrophoresis. Continued investigation into the replication and genome organization of LIV should aid in the understanding of its taxonomy, strategy for replication, and pathogenicity. ACKNOWLEDGMENTS We thank Douglas Furtek, Ross Hardison, Richard Arteca, and David Schultz for technical counsel. This work was supported in part by a grant from the Pennsylvania Department of Agriculture. REFERENCES 1. Bailey, J. M., and N. Davidson. 1976. Methylmercury as a reversible denaturing agent for agarose gel electrophoresis. Anal. Biochem. 70:75–85. 2. Ben-Tzvi, B.-S., Y. Koltin, M. Mevarech, and A. Tamarkin. 1984. RNA polymerase activity in virions from Ustilago maydis. Mol. Cell. Biol. 4:188– 194. 3. Buck, K. W. 1978. Semiconservative replication of double-stranded RNA by a virion-associated RNA polymerase. Biochem. Biophys. Res. Commun. 84: 639–645.

VOL. 71, 1997 4. Buck, K. W., M. A. Romanos, J. S. P. MacFadden, and C. J. Rawlinson. 1981. In vitro transcription of double-stranded RNA by virion-associated RNA polymerases of viruses from Gaeumannomyces graminis. J. Gen. Virol. 37: 215–219. 5. Chomczynski, P., and N. Sacchi. 1987. Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal. Biochem. 162:156–159. 6. Dieleman-van Zaayen, A., and J. H. M. Temmink. 1968. A virus disease of cultivated mushrooms in the Netherlands. Neth. J. Plant Pathol. 74:48–51. 7. Feinberg, A. P., and B. Vogelstein. 1983. A technique for radiolabeling DNA restriction endonuclease fragments. Anal. Biochem. 132:6–13. 8. Fujimura, T., and R. B. Wickner. 1989. Reconstitution of template dependent in vitro transcriptase activity of a yeast double-stranded RNA virus. J. Biol. Chem. 264:10872–10877. 9. Fujimura, T., and R. B. Wickner. 1988. Replicase of L-A virus-like particles of Saccharomyces cerevisiae. J. Biol. Chem. 263:454–460. 10. Ghabrial, S. A. 1994. New developments in fungal virology. Adv. Virus Res. 43:303–387. 11. Ghabrial, S. A., and W. M. Havens. 1989. Conservative transcription of Helminthosporium victoriae 190S double-stranded RNA in vitro. J. Gen. Virol. 70:1025–1035. 12. Goodin, M. M., B. Schlagnhaufer, and C. P. Romaine. 1992. Encapsidation of the La France disease-specific double-stranded RNAs in 36-nm isometric virus-like particles. Phytopathology 82:285–290. 13. Goodin, M. M., and C. P. Romaine. Unpublished data. 14. Gubler, U., and B. J. Hoffman. 1983. A simple and very effective method for generating cDNA libraries. Gene 25:263–269. 15. Harmsen, M. C., B. Tolner, A. Kram, S. J. Go, A. deHaan, and J. G. H. Wessels. 1991. Sequences of three dsRNAs associated with La France disease of the cultivated mushroom (Agaricus bisporus). Curr. Genet. 20:137– 144. 16. Hollings, M. 1962. Viruses associated with dieback disease of cultivated mushrooms. Nature (London) 196:962–965. 17. Klickstein, L. B., R. L. Neve, E. A. Golemis, and G. Gyuris. 1987. Conversion of mRNA into double-stranded cDNA, p. 5.5.2–5.5.14. In F. M. Ausubel, R. Brent, R. E. Kingston, D. D. Moore, J. G. Seidman, J. A. Smith, and K. Struhl (ed.), Current protocols in molecular biology. Greene Publishing Associates-Wiley-Interscience, New York, N.Y. 18. Maniatis, T., E. F. Fritsch, and J. Sambrook. 1982. Molecular cloning: a laboratory manual, p. 363–402. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. 19. Matsumoto, Y., R. Fishel, and R. B. Wickner. 1990. Circular single-stranded RNA replicon in Saccharomyces cerevisiae. Proc. Natl. Acad. Sci. USA 87: 7628–7632. 20. Morris, T. J., and J. A. Dodds. 1979. Isolation and analysis of double-


21. 22.

23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35.

2269

stranded RNA from virus-infected plant and fungal tissue. Phytopathology 69:854–858. Morten, K. J., and R. G. T. Hicks. 1992. Changes in double-stranded RNA profiles in Agaricus bisporus. FEMS Microbiol. Lett. 91:159–164. Murphy, F. A., C. M. Fauquet, D. H. L. Bishop, S. A. Ghabrial, A. W. Jarvis, G. P. Martelli, M. A. Mayo, and M. D. Summers (ed.). 1995. Virus taxonomy. Classification and nomenclature of viruses. Sixth report of the International Committee on Taxonomy of Viruses. Springer-Verlag, New York, N.Y. Passmore, E. L., and R. R. Frost. 1979. The detection of virus-like particles in mushrooms and mushroom spawns. Phytopathol. Z. 80:85–87. Rogers, H. J., K. W. Buck, and C. M. Brasier. 1987. A mitochondrial target for double-stranded RNA in diseased isolates of the fungus that causes Dutch elm disease. Nature (London) 329:558–560. Romaine, C. P., and B. Schlagnhaufer. 1989. Prevalence of double-stranded RNAs in healthy and La France disease-affected basidiocarps of Agaricus bisporus. Mycologia 81:822–825. Romaine, C. P., and B. Schlagnhaufer. 1991. Hybridization analysis of the single-stranded RNA bacilliform virus associated with La France disease of Agaricus bisporus. Phytopathology 81:1336–1340. Romaine, C. P., and B. Schlagnhaufer. 1995. PCR analysis of the viral complex associated with La France disease of Agaricus bisporus. Appl. Environ. Microbiol. 61:2322–2325. Romaine, C. P., B. Schlagnhaufer, and M. M. Goodin. 1994. Vesicle-associated double-stranded ribonucleic acid genetic elements in Agaricus bisporus. Curr. Genet. 25:128–134. Russell, M., S. Kidd, and M. R. Kelly. 1986. An improved filamentous helper phage for generating single-stranded plasmid DNA. Gene 45:333–338. Sclafani, R. A., and W. L. Fangman. 1984. Conservative replication of double-stranded RNA in Saccharomyces cerevisiae by displacement of progeny single strands. Mol. Cell. Biol. 4:1618–1626. Sinden, J. W., and E. Hauser. 1950. Report of two new mushroom diseases. Mushroom Sci. 1:96–100. Sriskantha, A., M. P. Wach, B. Schlagnhaufer, and C. P. Romaine. 1986. Synthesis of double-stranded RNA in a virus-enriched fraction from Agaricus bisporus. J. Virol. 57:1004–1009. Van der Lende, T. R. 1994. Double-stranded RNAs and proteins associated with the 34 nm virus particles of the cultivated mushroom Agaricus bisporus. J. Gen. Virol. 75:2533–2536. Wach, M. P., A. Srikantha, and C. P. Romaine. 1987. Double-stranded RNAs associated with La France disease of the cultivated mushroom. Phytopathology 77:1321–1325. White, T. C., and C. C. Wang. 1990. RNA-dependent RNA polymerase activity associated with the double-stranded RNA virus of Giardia lamblia. Nucleic Acids Res. 18:553–559.