Soluble, Template-Dependent Extracts from Nicotiana Benthamiana Plants Infected with Potato Virus X Transcribe Both Plus- and Minus-Strand RNA Templates

Home , Black beetle virus, Turnip crinkle virus

Virology 275, 444–451 (2000) doi:10.1006/viro.2000.0512, available online at http://www.idealibrary.com on View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Elsevier - Publisher Connector

Soluble, Template-Dependent Extracts from Nicotiana benthamiana Plants Infected with Potato Virus X Transcribe both Plus- and Minus-Strand RNA Templates

Carol A. Plante,* Kook-Hyung Kim,*,1 Neeta Pillai-Nair,* Toba A. M. Osman,† Kenneth W. Buck,† and Cynthia L. Hemenway*,2

*Department of Biochemistry, North Carolina State University, Raleigh, North Carolina 27695; and †Department of Biology, Imperial College of Science, Technology, and Medicine, London, SW7 2AZ, United Kingdom Received May 4, 2000; returned to author for revision June 22, 2000; accepted July 7, 2000

We have developed a method to convert membrane-bound replication complexes isolated from Nicotiana benthamiana plants infected with potato virus X (PVX) to a soluble, template-dependent system for analysis of RNA synthesis. Analysis of RNA-dependent RNA polymerase activity in the membrane-bound, endogenous template extracts indicated three major products, which corresponded to double-stranded versions of PVX genomic RNA and the two predominant subgenomic RNAs. The endogenous templates were removed from the membrane-bound complex by treatment with BAL 31 nuclease in the presence of Nonidet P-40 (NP-40). Upon the addition of full-length plus- or minus- strand PVX transcripts, the corresponding-size products were detected. Synthesis was not observed when red clover necrotic mosaic dianthovirus (RCNMV) RNA 2 templates were added, indicating template specificity for PVX transcripts. Plus-strand PVX templates lacking the 3Ј terminal region were not copied, suggesting that elements in the 3Ј region were required for initiation of RNA synthesis. Extracts that supported RNA synthesis from endogenous templates could also be solublized using sodium taurodeoxycholate and then rendered template-dependent by BAL 31 nuclease/NP-40 treatment. The solubilized preparations copied both plus- and minus-strand PVX transcripts, but did not support synthesis from RCNMV RNA 2. These membrane-bound and soluble template-dependent systems will facilitate analyses of viral and host components required for PVX RNA synthesis. © 2000 Academic Press

INTRODUCTION (Bates et al., 1995), and turnip crinkle virus (Song and Simon, 1994). Some of these template-dependent sys- The development of template-dependent, RNA-depen- tems have been utilized for analyses of host and viral dent RNA polymerase (RdRp) preparations from tissues or cells infected with plus-strand RNA viruses has facil- protein components contained in the RdRp complexes itated analyses of the components and reactions neces- (for reviews, see Buck, 1996; and Lai, 1998). Several of sary for RNA synthesis. Such systems have been devel- the plant viral RdRp preparations have been useful for oped for several viruses that infect a variety of hosts, the determination of template requirements for initiation including bacteriophage Q␤ (Blumenthal and Car- of minus-strand RNA synthesis (for reviews, see Buck, michael, 1979), black beetle virus (Saunders and Kaes- 1996; and Dreher, 1999). The BMV and TCV RdRp prep- berg, 1985), flockhouse virus (Wu et al., 1992), polio virus arations have additionally been utilized for analysis of (van Dyke and Flanegan, 1980; Baron and Baltimore, plus-strand RNA synthesis in vitro (Miller et al., 1985; 1982), Sindbis virus (Lemm et al., 1998), and the plant Adkins et al., 1997; Guan et al., 1997, 2000a, b; Siegel et viruses alfalfa mosaic virus (Houwing and Jaspars, 1986; al., 1997; Wang and Simon, 1997; Sivakumaran and Kao, Quadt et al., 1991), brome mosaic virus (Miller and Hall, 1999; Sivakumaran et al., 1999) and for studying recom- 1983; Quadt and Jaspars, 1990), cucumber mosaic virus bination mechanisms (Nagy et al., 1998; Nagy and Si- (Hayes and Buck, 1990; Quadt and Jaspars, 1991), turnip mon, 1998a, b). For the most part, such requirements yellow mosaic virus (Mouches et al., 1974; Garbouri- have been investigated for plant viruses containing ei- Bouzid et al., 1991; Deiman et al., 1997; Singh and Dreher, ther a tRNA-like 3Ј terminus or other predicted struc- 1997), tobacco mosaic virus (Osman and Buck, 1996; tures, but not for viruses containing polyadenylated ter- Watanabe et al., 1999), red clover necrotic mosaic virus mini such as those in the Potexvirus genus. Potato virus X (PVX), the type member of the Potexvirus genus, is a flexuous rod-shape particle containing a 1 Current address: School of Applied Biology and Chemistry, College capped and polyadenylated genomic RNA of 6.4 kb of Agriculture and Life Science, Seoul National University, Suwon, (Skryabin et al., 1988a, b). This RNA contains an 84- Korea 441-744. 2 Ј To whom correspondence and reprint requests should be ad- nucleotide (nt) 5 nontranslated region (NTR), five open dressed. reading frames (ORFs), and a 72-nt 3Ј NTR (Fig. 1A). The

tabacum (NT-1) protoplasts and plants (Kim and Hemen- way, 1996; Miller et al., 1998, 1999). Conserved oc- tanucleotide sequence elements located upstream of the two major PVX sgRNAs are important for sgRNA accumulation (Kim and Hemenway, 1997, 1999), and interactions between these elements and complementary terminal sequences are important for optimal plus-strand RNA accumulation. Elements in the 3Ј NTR also effect PVX RNA accumulation in NT-1 cells (Pillai-Nair, Kim, and Hemenway, unpublished data). The development of a soluble, template-dependent system would allow for mechanistic studies of PVX RNA synthesis in vitro and identification of protein components and activities necessary for PVX replication. An RdRp preparation from barley plants infected with an- other potexvirus, foxtail mosaic virus (FMV), was developed, but treatment of these preparations with micrococcal nuclease for removal of endogenous template rendered them inactive (Rouleau et al., 1993). Previously, we isolated extracts containing membrane-bound PVX RdRp from Nicotiana tabacum, but also were not able to elim- inate the endogenous viral RNA (Doronin and Hemen- way, 1996). Li et al. (1998) recently demonstrated template-dependent RdRp activity in Escherichia coli trans- FIG. 1. PVX genome organization and transcripts derived from PVX formed with the PVX replicase protein using short PVX cDNA clones. A illustrates the PVX genome, depicting five open read- templates; however, this system cannot be utilized to ing frames as boxes and the two major subgenomic RNAs as arrows study the in vivo composition of the replication complex. below the genome. Transcripts utilized for template-dependent RdRp assays were derived from various PVX cDNA clones shown in B. Here we report the isolation and solubilization of a tem- Plus-strand RNA templates were derived from wild-type pMON8453, plate-dependent RdRp activity isolated from Nicotiana p32 (containing an internal deletion of most of ORF 1, as noted by benthamiana plants infected with PVX that supports RNA dotted line), and ⌬84 (lacking the first 84 nucleotides of the 5Ј end) by synthesis from both plus- and minus-strand PVX tem- transcription with bacteriophage T7 RNA polymerase (shown as arrow plates. to left) of SpeI-digested plasmids. These transcripts contain 17 A residues and a G residue (from the SpeI site) at their 3Ј termini. Truncated plus-strand transcripts lacking the 3Ј region of the PVX RESULTS AND DISSUSSION genome were generated by transcription of ⌬84 digested with BstEII. Full-length minus-strand transcripts were obtained by transcription of Products produced from endogenous templates are BstBI-digested p75 with with bacteriophage SP6 RNA polymerase. primarily double-stranded With the ultimate goal of developing an extract con- first ORF encodes a 165-kDa protein (P1) that has ho- taining a template-dependent RdRp, we first developed a mology to other known RNA-dependent RNA polymerase membrane-containing RdRp preparation from N. proteins (Rozanov et al., 1992) and is the only viral protein benthamiana plants inoculated with PVX. Similar to the required for RNA synthesis in tobacco (Longstaff et al., procedure published for TMV (Osman and Buck, 1996), a 1993). Three internal reading frames (ORFs 2–4), referred P30 pellet was resuspended and purified through a su- to as the triple gene block, encode proteins involved in crose density gradient. Similar to the TMV system, anal- cell-to-cell movement (Beck et al., 1991; Angell et al., ysis of RdRp activity on endogenous PVX templates in- 1996). ORF 5 encodes the coat protein, which functions dicated that the near-colorless, sucrose gradient fraction in cell-to-cell movement (Chapman et al., 1992; Baul- 3 (F3) and the dark green fraction 4 (F4) contained the combe et al., 1995) and encapsidation. Replication of peak of activity (data not shown). As indicated in Fig. 2 PVX in plant tissue results in the production of genomic- (lane 3), analysis of the RNA products using F3 from length plus- and minus-sense RNAs, two major plus- infected plants (Inf) on a denaturing gel indicated RNAs sense subgenomic RNAs, and corresponding double- migrating at approximately 6.4, 2.5, and 0.9 kb. Products stranded RNAs (Dolja et al., 1987). We have demon- treated with S1 nuclease prior to electrophoresis exhib- strated that sequence elements, as well as a stem–loop ited a similar profile (Fig. 2, lane 4), indicating that these structure located in the 5Ј nontranslated region of the products are primarily double-stranded. In contrast, ex- PVX genome, are required for both genomic and sub- tracts from mock inoculated plants (M), did not support genomic plus-strand RNA accumulation in Nicotiana synthesis of these reaction products (lane 2). This profile 446 PLANTE ET AL.

nuclease or treatment with KCl in the presence of detergents did not significantly remove the endogenous templates (data not shown). As the endonuclease BAL 31 was reported to digest tRNA (Gray et al., 1975) and hydrolyze duplex RNA from both the 5Ј and the 3Ј ends (Bencen et al., 1984), we tested this nuclease for removal of the endogenous PVX templates in the infected plant extracts. Titration of this enzyme in the presence or absence of various detergents indicated that 10 units of BAL 31 nuclease in the presence of 5 mM calcium

acetate (Ca(OAc)2) and 0.1% Nonidet P-40 (NP-40) was effective at removal of the PVX templates from the sucrose gradient fraction 3. We found that optimal template removal required both BAL 31 and addition of NP-40 concentrations equal to or greater than 0.1%. Presum- ably, the detergent helped to open up the complex and render the template accessible to the nuclease. As shown in Fig. 3 (lanes 3 and 4), the activity of F3 was

somewhat stimulated by addition of NP-40/Ca(OAc)2.In contrast, extract treated with BAL 31 nuclease and NP-

FIG. 2. Products derived from extracts containing endogenous tem- 40/Ca(OAc)2 (lane 5) did not support RNA synthesis in plates. Extracts prepared from N. benthamiana plants 5 days postin- the absence of added template. oculation were tested for RdRp activity and reaction products were Addition of EGTA to the BAL 31 nuclease treated ex- analyzed by denaturing 1% agarose/formaldehyde gel electrophoresis. tract was utilized to chelate the Caϩ2 and inactivate the RNA markers were loaded in lane 1; the sizes of these markers (6.4, 4.1, 2.7, and 0.8 kb) are noted at the left. Lanes 2 and 3 contain products nuclease. Subsequently, extracts were tested for RNA from RdRp reactions with F3 samples from mock-inoculated (M) or synthesis by the addition of various templates (Fig. 1B) in infected (Inf) plants, respectively. RNA products isolated from RdRp the presence of the RNase inhibitor, RNasin. The wild- reactions with infected plant extracts that were additionally treated with type PVX plasmid, pMON8453 (Hemenway et al., 1990) ϩ 200 units of S1 nuclease were loaded in lane 4 (Inf S1). The arrows was linearized with SpeI and used to generate full- shown at the right mark positions of the major products from the RdRp reactions. length, plus-strand templates. A truncated plus-strand template was generated by transcription of p32, which was derived from pMON8453 by deletion of most of the of three major double-stranded products that correspond replicase gene (nts 203–4138) (Kim and Hemenway, to sizes of genomic and the two major subgenomic RNAs 1996). Two additional truncated plus-strand transcripts ⌬ was previously reported for an extract derived from PVX- were obtained from the 84 plasmid, which contains a Ј infected N. tabacum plants by a somewhat different pro- deletion of the 5 NTR (nts 1–84) of pMON8453 (Kim and cedure (Doronin and Hemenway, 1996). In addition, the Hemenway, 1996). The first truncated template, referred ⌬ polymerase activity was completely abolished from PVX- to as 84, was approximately 6.3 kb in size (nts 84–6436) ⌬ inoculated extract when any of the four nucleotides were and was directly transcribed from 84 DNA digested ⌬ eliminated from the replication mixture, demonstrating with SpeI. A second truncated transcript, 84/BstEII (nts Ј Ј that the activity was due to RdRp complexes and not 84–2736), lacked both the 5 and the 3 termini and was ⌬ terminal transferase activity (data not shown). made from 84 linearized with BstEII. RNA2 from the unrelated virus, red clover necrotic mosaic virus (RC- NMV), was utilized as a negative control. Extracts rendered template-dependent are specific As shown in Fig. 3, no products (1.47 kb expected size) for positive- and negative-strand PVX templates were detected when templates from RCNMV were in- For other plant viral systems, either micrococcal nu- cluded (lane 11, respectively). The major product of the clease digestion in the presence or absence of detergent polymerase reaction upon addition of a full-length PVX or treatment with high concentrations of KCl in the pres- plus-sense RNA (pMON8453) corresponded in size to a ence of detergent have been effective at removal of 6.4-kb double-stranded species (lane 6), indicating that endogenous templates from RdRp complexes (for review, the product is likely a double-stranded form of the input see Hayes and Buck, 1993). Although micrococcal nucle- template. Comparison of the band intensity for this prod- ase treatment of the potexvirus, FMV, resulted in removal uct (lane 6) and the products derived from the endoge- of endogenous template, the resulting extract lost activity nous template sample (lane 4) indicated that activity (Rouleau et al., 1993). In the case of the PVX system, levels are lower for the template-dependent extract. Typ- treatment with very high concentrations of micrococcal ically, we observed a 5- to 20-fold decrease in product TEMPLATE-DEPENDENT EXTRACTS FROM N. benthamiana 447

FIG. 3. Analysis of template-dependent extracts. Sucrose gradient fraction 3 samples were incubated in the absence (Ϫ) or in the presence (ϩ) of 0.1% NP-40/5 mM Ca(OAc)2, 10 unit BAL 31, and 5 ␮g exogenous template. Products were analyzed on a vertical 1.5% agarose, 0.1% sodium dodecyl sulfate gel. Lane 1 contains 32P-labeled bacteriophage lambda DNA digested with BstEII as a double-stranded size marker; the sizes of these markers are noted at the left. RNA transcripts were loaded into lane 2 and their sizes are noted at the right of the gel. Lanes 3 and 4 contain the RdRp products of extracts containing endogenous viral template in the absence or presence of NP-40/calcium acetate, respectively. After treatment of F3 samples with NP-40/calcium acetate and BAL 31, RdRp reactions were conducted without added template (lane 5) or after addition of transcripts derived from pMON8453 (lane 6), ⌬84 (lane 7), ⌬84/BstEII (lane 8), p32 (lane 9), p75 (lane 10), and RCNMV RNA 2 (lane 11). The positions of expected template-sized products are noted by arrowheads to the right of each lane. levels upon conversion of the endogenous template a polyadenlyated 3Ј terminus and is predicted to form preparations to template-dependent extracts, and the several stable stem–loop structures within the 3Ј non- extent of this decrease varied with different extracts and translated region. In addition, several elements in this Bal31 preparations. Products similar in size to those region are important for minus-strand RNA accumulation obtained with full-length PVX transcripts were produced in vivo (Pillai-Nair, Kim, and Hemenway, unpublished when ⌬84 was incubated in the polymerase assay (lane results). Future experiments will determine whether 3Ј 7), suggesting that the 5Ј NTR is not necessary for elements necessary for PVX RNA synthesis in vivo also initiation of negative-strand RNA synthesis in these prep- are necessary for synthesis in vitro. arations. Reactions containing RNA transcripts from p32 For analysis of a minus-strand PVX template, plasmid resulted in products migrating at approximately 2.5 kb, p75 was generated from pMON8453 in order to synthe- which is the expected size of a duplex product of the p32 size full-length negative-strand PVX templates. The template (lane 9). Given that the p32 transcripts lack pMON8453 clone was modified to contain a BstBI site most of the replicase coding sequence, these data indi- upstream of the 5Ј terminus of the PVX cDNA. As a result cate that initiation of minus-strand RNA synthesis in this of the BstB1 site insertion, a nontemplated guanylate system was not dependent on nts 203–4138 of the PVX residue is generated at the 3Ј terminus of p75 minus- genome. strand RNA template during transcription with bacterio- In contrast, ⌬84/BstEII that contained only the coding phage SP6 polymerase. Nontemplated residues have region of PVX ORF 1 did not support transcription of a been found in viral RNAs isolated from tissues infected template-size product of 2.7 kb (lane 8), indicating that with PVX (Dolja et al., 1987), CMV (Collmer and Kaper, signals in the 3Ј region were necessary for initiation of 1985), and Semliki Forest virus RNA (Wengler and Gross, minus-strand RNA synthesis in this system. It has been 1979). In addition, for both CMV (Wu and Kaper, 1994) and shown for many viruses that both sequences and struc- BMV (Sivakumaran and Kao, 1999) RdRp systems, a tures in the 3Ј end of the viral plus-strand RNAs are nontemplated guanylate is required for the initiation of important for the initiation of minus-strand RNA synthesis plus-strand RNA synthesis. As shown in Fig. 3 (lane 10), in vivo and in vitro (for reviews, see Buck, 1996; Dreher, the BAL 31 treated PVX RdRp preparation was able to 1999; and Hemenway and Lommel, 2000). PVX RNA has copy the minus-strand p75 template, resulting in prod- 448 PLANTE ET AL.

TDC resulting in the most solubilized activity. After treatment of the S100 with nuclease BAL 31, no polymerase activity could be detected without the addition of template (lane 5). However, the addition of p32 truncated plus-strand template to the BAL 31 treated S100 sample resulted in synthesis of the corresponding 2.5-kb double- stranded product (lane 6). Full-length minus-strand p75 template was also successfully copied in this system (lane 7). In contrast, RCNMV RNA 2 template was not copied (lane 8), demonstrating that the soluble system is not only template-dependent but also specific for PVX RNA. Development of this template-dependent RdRp system that can utilize both plus- and minus-strand PVX RNA templates will enable us to analyze the mechanisms of RNA synthesis and determine the extent to which regu- latory elements required in vitro are similar to sequences/structures that we have shown to be important for PVX RNA accumulation in vivo (Kim and Hemenway, 1996, 1997, 1999; Miller et al., 1998, 1999). An advantage of the in vitro system is that it will enable us obtain FIG. 4. RNA synthesis in solubilized, template-dependent extracts. information specific to RNA synthesis rather than other Extracts incubated with 0.5, 1, and 2% TDC were centrifuged at 100,000g and RdRp activity in the S100 samples was assayed (lanes aspects of the replication cycle, such as translation and 2–4, respectively). After treatment of S100 samples with NP-40/calcium encapsidation, which are frequently difficult to distin- acetate/BAL 31, RdRp activity was assayed in the absence (lane 5) or guish in vivo. presence of transcripts derived from p32 (lane 6), p75 (lane 7), or RCNMV RNA 2 (lane 8). The sizes of DNA markers shown in lane 1, as well as conditions for electrophoresis and visualization of products, MATERIALS AND METHODS were as described in the legend for Fig. 3. Materials All restriction enzymes, nucleases, and polymerases ucts similar in size to double-stranded genomic PVX were purchased from New England Biolabs. RNasin and RNA. The extent to which the extra G nucleotide in the RQ1 RNase-free DNase were obtained from Promega. p75 template is necessary for RNA synthesis remains to Ribonucleotides were purchased from Roche. Sigma be determined. Also shown in Fig. 3, the RNA synthe- supplied actinomycin D, creatine kinase, creatine phos- sized from the p32 template gave a more discrete band phate, leupeptin, pepstatin, and sodium taurodeoxy- on this native gel than those observed with the full-length cholate. Radiolabeled [␣-32P]CTP (800Ci/mmol) was a templates. It is likely that the diffuse nature of these product of Amersham. The RCNMV RNA2 cDNA clone full-length product bands reflects heterogeneity in the used to generate corresponding transcripts was a gen- templates generated by in vitro transcription with bacte- erous gift of Dr. Steven Lommel (NCSU). riophage T7 or SP6 RNA polymerases and/or to less efficient copying of the longer templates by the PVX Synthesis of RNA templates RdRp complex. Runoff transcripts derived from various PVX cDNA clones were utilized for template-dependent assays (Fig. Solubilized extracts retain template-dependent activity 1B). The wild-type PVX plasmid, pMON8453, ⌬84, and In order to develop a solubilized template-dependent p32 were transcribed with bacteriophage T7 RNA poly- extract, we first determined the conditions for solubiliza- merase as described previously (Kim and Hemenway, tion of the RdRp activity using a procedure similar to that 1996). A truncated plus-strand transcript (2.6 kb, contain- described by Osman and Buck (1997). The sucrose gra- ing nts 84–2736) lacking both 5Ј and 3Ј termini was made dient F3 from PVX infected plants was incubated with from a BstEII-digested ⌬84 template. The plasmid p75 various concentrations of sodium taurodeoxycholate was used to generate minus-strand transcripts by tran- (TDC). After centrifugation for1hat100,000g, both pellet scription with bacteriophage SP6 RNA polymerase. Con- (data not shown) and supernatant fractions were tested trol transcripts corresponding to an unrelated virus were for RdRp activity. As shown in Fig. 4 (lanes 2–4), treat- derived from a red clover necrotic mosaic RNA2 cDNA ment with increasing amounts of TDC resulted in in- clone that was digested with SmaI and transcribed with creasing levels of activity in the S100 samples, with 2% bacteriophage T7 RNA polymerase (Xiong and Lommel, TEMPLATE-DEPENDENT EXTRACTS FROM N. benthamiana 449

1991). RNA transcripts were purified by phenol/chloro- tions were incubated with 10 units of BAL 31 nuclease in form extraction, passage through Qiagen RNeasy spin the presence of 0.1% NP-40 and 5 mM calcium acetate at columns, and ethanol precipitation. The quality and rel- 30°C for 30 min. The reaction was terminated by the ative concentrations of transcripts were checked by elec- addition of EGTA to a concentration of 25 mM and placed trophoresis on 1.0% agarose gels at 4°C and visualized on ice for 30 min. Treated fractions were tested for RdRp by ethidium bromide staining as described previously activity with or without template RNA (5 ␮g) addition in 25 (Kim and Hemenway, 1996). ␮l of buffer containing 50 mM Tris–HCl, pH 8.2; 4%

glycerol, 1 mM MnCl2,1mMATP,1mMUTP,1mMGTP, Preparation of extracts containing RdRp activity 25 ␮M CTP, 0.2 ␮Ci/␮l[␣-32P]CTP (800 Ci/mmol), 10 mM DTT, 0.2 ␮g/␮l actinomycin D, 800 U/ml RNasin, 5 mM Each leaf of 10 N. benthamiana plants was dusted with creatine phosphate, and 0.2 ␮g/␮l creatine phosphoki- carborundum (320 grit) and inoculated with 50 ␮lof10 nase. Samples were incubated for 30 min at 30°C and ␮g/ml of PVX in 0.1 M sodium phosphate buffer, pH 7.0. terminated by addition of phenol/chloroform. After con- Plants were maintained at 25°C in light for 12 h and at centration by ethanol precipitation, RNA products were 22°C in darkness for 12 h. Extracts were prepared using analyzed by electrophoresis as described below. a protocol similar to that utilized for tobacco mosaic virus Although the sucrose gradient F4 exhibited the maxi- (Osman and Buck, 1996). At 5 days postinoculation, mum amount of RdRp activity, the BAL 31 nuclease treat- leaves were collected and homogenized in 3 vol of cold ment was not effective in eliminating the endogenous Buffer A (50 mM Tris–HCl, pH 8.2; 15 mM MgCl , 120 mM 2 template from this fraction (data not shown). For this KCl, 20% glycerol, 1 mM DTT, 1 ␮M leupeptin, 1 ␮M reason, F3 was used for all template-dependent studies. pepstatin). The crude homogenate was filtered through In contrast to the endogenous template preparation, both muslin and centrifuged at 500g for 10 min at 4°C. The the template-dependent and the soluble template-de- supernatant was removed and centrifuged at 30,000g for pendent preparations were highly unstable, with approx- 30 min. The resulting pellet (P30) was resuspended in imately 90% of activity being lost upon freezing and Buffer B (50 mM Tris–HCl, pH 8.2; 15 mM MgCl ,5% 2 thawing. glycerol, 1 mM DTT, 1 ␮M leupeptin, 1 ␮M pepstatin) and loaded onto 20–60% continuous sucrose gradients. After Analysis of RdRp products centrifugation at 189,000g for1hat4°C, fractions (approximately 5 ml each) were collected from the top of the Products were isolated from RdRp reactions by phe- gradient. Sucrose gradient fractions stored at Ϫ80°C nol/chloroform extraction and ethanol precipitation. Al- were stable for up to 1 year. ternatively, to determine whether the RNA was double- or Sucrose gradient fractions were analyzed for RdRp single-stranded in nature, products from RdRp assays activity by combining 100 ␮l of each fraction with 25 ␮lof were resuspended in 200 ␮l of buffer containing 0.2 M 5ϫ RdRp reaction buffer, giving final concentrations of 50 NaCl, 0.05 M sodium acetate, pH 4.5, 1 mM ZnSO4, 0.5% mM Tris–HCl, pH 8.2; 4% glycerol, 10 mM MgCl2,1mM glycerol), and 200 U of S1. These samples were incu- ATP,1mMUTP,1mMGTP,25␮M CTP, 0.2 ␮Ci/␮l bated at 37°C for 20 min, extracted with phenol/chloro- [␣-32P]CTP (800 Ci/mmol), 10 mM DTT, 0.2 ␮g/␮l actino- form, and ethanol precipitated. All RNA products were mycin D, and 800U/ml RNasin. Samples were incubated loaded onto either nondenaturing vertical, 1.5% agarose for1hat30°C and RNA products were extracted with gels (Tris–acetate, 0.1% SDS, and 40 mM iodoacetate) or equal volumes of phenol/chloroform and ethanol precip- denaturing 1% agarose gels containing formaldehyde. itated. Gradient fractions 3 and 4 typically exhibited the Gels were dried and exposed to autoradiographic film or peak activities. analyzed using a phosphorimager and the Image Quant program. Conversion of extract to soluble, template-dependent system ACKNOWLEDGMENTS Prior to removal of endogenous template, the active This research was supported by National Institutes of Health Grant gradient fractions were diluted to reduce the sucrose GM49841 awarded to C.H and a grant from the UK Biotechnology and concentration and subsequently centrifuged at 50,000g. Biological Sciences Research Council to K.W.B. C.P. was a recipient of a North Carolina State University Biotechnology Student Award and After resuspension in buffer B, the P50 was stable for up College of Agriculture and Life Science Study Abroad Program Award. to 1 month at Ϫ80°C. This P50 fraction was either directly used to remove the endogenous template or first REFERENCES treated with TDC. For solubilization, 100 ␮l of P50 from F3 or F4 was adjusted to 2% TDC, incubated on ice for Adkins, S., Siegel, R. W., Sun, J. H., and Kao, C. C. (1997). Minimal templates directing accurate initiation of subgenomic RNA synthesis 1 h, and centrifuged at 100,000g for 1 h. using TDC as in vitro by the brome mosaic virus RNA-dependent RNA polymerase. described by Osman and Buck (1997). For removal of the RNA 3, 634–647. endogenous template, either the P50 or the S100 frac- Angell, S. M., Davies, C., and Baulcombe, D. C. (1996). Cell-to-cell 450 PLANTE ET AL.

movement of potato virus X is associated with a change in the Hemenway, C., Weiss, J., O’Connel, K., and Tumer, N. E. (1990). Char- size-exclusion limit of plasmodesmata in trichome cells of Nicotiana acterization of infectious transcripts from a potato virus X cDNA clevelandii. Virology 216, 197–201. clone. Virology 175, 365–371. Baron, M. K., and Baltimore, D. (1982). In vitro copying of viral positive Hemenway, C. L., and Lommel, S. A. (2000). Manipulating plant viral strand RNA by poliovirus replicase. Characterization of the reaction RNA transcription signals. In “Genetic Engineering Principles and and its products. J. Biol. Chem. 257, 12359–12366. Methods” (J. K. Setlow, Ed.). Vol. 22, in press, Plenum Press, NY. Bates, H. J., Farjah, M., Osman, T. A. M., and Buck, K. W. (1995). Isolation Houwing, C. J., and Jaspars, E. M. J. (1986). Coat protein blocks the in and characterization of an RNA-dependent RNA polymerase from vitro transcription of virion RNAs of alfalfa mosaic virus. FEBS Lett. Nicotiana clevelandii plants infected with red clover necrotic mosaic 209, 264–268. dianthovirus. J. Gen. Virol. 76, 1483–1491. Kim, K.-H., and Hemenway, C. (1996). The 5Ј nontranslated region of Baulcombe, D. C., Chapman, S., and Santa Cruz, S. (1995). Jellyfish potato virus X RNA affects both genomic and subgenomic RNA green fluorescent protein as a reporter for virus infections. Plant J. 7, synthesis. J. Virol. 70, 5533–5540. 1045–1053. Kim, K.-H., and Hemenway, C. (1997). Mutations that alter a conserved Beck, D. L., Guilford, P. J., Voot, D. M., Andersen, M. T., and Forster, R. L. element upstream of the potato virus X triple block and coat protein (1991). Triple gene block proteins of white clover mosaic potexvirus genes affect subgenomic RNA accumulation. Virology 232, 187–197. are required for transport. Virology 183, 695–702. Kim, K.-H., and Hemenway, C. L. (1999). Long distance RNA–RNA Bencen, G. H., Wei, C. F., Robberson, D. L., and Gray, H. B. J. (1984). interactions and conserved sequence elements affect potato virus X Terminally directed hydrolysis of ribonucleic acid catalyzed by a plus-strand RNA accumulation. RNA 5, 636–645. species of the BAL 31 nuclease from Altermonas espejiana. J. Biol. Lai, M. M. C. (1998). Cellular factors in the transcription and replication Chem. 259, 13584–13589. of viral RNA genomes: A parallel to DNA-dependent RNA transcrip- Blumenthal, T., and Carmichael, G. G. (1979). RNA replication: Function tion. Virology 244, 1–12. and structure of Qbeta-replicase. Annu. Rev. Biochem. 48, 525–548. Lemm, J. A., Bergqvist, A., Read, C. M., and Rice, C. M. (1998). Template- Buck, K. W. (1996). Comparison of the replication of postive-stranded dependent initiation of Sindbis virus RNA replication in vitro. J. Virol. RNA viruses of plants and animals. Adv. Virus Res. 47, 159–251. 72, 6546–6553. Chapman, S., Hills, G., Watts, J., and Baulcombe, D. (1992). Mutational Li, Y. I., Cheng, Y. M., Huang, Y. L., Tsai, C. H., Hsu, Y. H., and Meng, M. analysis of the coat protein gene of potato virus X: Effects on virion (1998). Identification and characterization of the Escherichia coli- morphology and viral pathogenicity. Virology 191, 223–230. expressed RNA-dependent RNA polymerase of bamboo mosaic vi- Collmer, C. W., and Kaper, J. M. (1985). Double-stranded RNAs of rus. J. Virol. 72, 10093–10099. cucumber mosaic virus and its satellite contain an unpaired terminal Longstaff, M., Brigneti, G., Boccard, F., Chapman, S., and Baulcombe, D. guanosine: Implications for replication. Virology 145, 249–259. (1993). Extreme resistance to potato virus X infection in plants ex- Deiman, B. A. L. M., Seron, K., Jaspars, E. M. J., and Pleij, C. W. A. (1997). pressing modified components of the putative viral replicase. EMBO Efficient transcription obtained by a new procedure of the tRNA-like J. 12, 379–386. structure of turnip yellow mosaic virus by a template-dependent and Miller, E. D., Plante, C. A., Kim, K.-H., and Hemenway, C. (1998). Stem– specific viral RNA polymerase. J. Virol. Methods 64, 181–195. loop structure in the 5Ј region of potato virus X genome required for Dolja, V. V., Grama, D. P., Morozov, S. Y., and Atabekov, J. G. (1987). plus-strand RNA accumulation. J. Mol. Biol. 284, 591–608. Potato virus X related single- and double-stranded RNAs. FEBS Lett. Miller, E. D., Kim, K.-H., and Hemenway, C. (1999). Restoration of a 214, 308–312. stem–loop structure required for potato virus X accumulation indi- Doronin, S. V., and Hemenway, C. (1996). Synthesis of potato virus X cates selection for a mismatch and a GNRA tetraloop. Virology 260, RNAs by membrane-containing extracts. J. Virol. 70, 4795–4799. 342–353. Dreher, T. W. (1999). Functions of the 3Ј-untranslated regions of positive Miller, W. A., and Hall, T. C. (1983). Use of micrococcal nuclease in the strand RNA viral genomes. Annu. Rev. Phytopathol. 37, 151–174. purification of highly template dependent RNA-dependent RNA poly- Garbouri-Bouzid, R., David, C., and Haenni, A. L. (1991). The 3Ј promoter merase from brome mosaic virus-infected barley. Virology 25, 236– region involved in RNA synthesis directed by turnip yellow mosaic 241. virus genome in vitro. FEBS Lett. 294, 56–58. Miller, W. A., Dreher, T., and Hall, T. C. (1985). Synthesis of brome Gray, H. B., Ostrander, D. A., Hodnett, J. L., Legerski, R. J., and Robber- mosaic virus subgenomic RNA in vitro by initiation on minus-sense son, D. L. (1975). Extracellular nucleases of Psudomonas BAL31. I. genomic RNA. Nature 313, 68–70. Characterization of single strand-specific deoxyriboendonuclease Mouches, C., Bove, C., and Bove, J. M. (1974). Turnip yellow mosaic and double-strand deoxyriboexonuclease activities. Nucleic Acids virus RNA replicase: Partial purification of the enzyme from the Res. 2, 1459–1492. solubilized enzyme–template complex. Virology 58, 409–423. Guan, H., Song, C., and Simon, A. E. (1997). RNA promoters located on Nagy, P. D., Zhang, C., and Simon, A. E. (1998). Dissecting RNA recom- (Ϫ)-strands of a subviral RNA associated with turnip crinkle virus. bination in vitro: Role of RNA sequences and the viral replicase. RNA 3, 1401–1412. EMBO J. 7, 2392–2403. Guan, H., Carpenter, C. D., and Simon. A. E. (2000a). Analysis of Nagy, P. D., and Simon, A. E. (1998a). In vitro characterization of late cis-acting sequences involved in plus-strand synthesis of a turnip steps of RNA recombination in turnip crinkle virus. I. Role of the crinkle virus-associated satellite RNA identifies a new carmovirus motif1–hairpin structure. Virology 249, 379–392. replication element. Virology 268, 345–354. Nagy, P. D., and Simon, A. E. (1998b). In vitro characterization of late Guan, H., Carpenter, C. D., and Simon. A. E. (2000b). Requirement of a steps of RNA recombination in turnip crinkle virus. II. The role of the 5Ј-proximal linear sequence on minus-strands for plus-strand syn- priming stem and flanking sequences. Virology 249, 393–405. thesis of a satellite RNA associated with turnip crinkle virus. Virology Osman, T. A., and Buck, K. W. (1996). Complete replication in vitro of 268, 355–363. tobacco mosaic virus RNA by a template-dependent, membrane- Hayes, R. J., and Buck, K. W. (1990). Complete replication of a eukaryotic bound RNA polymerase. J. Virol. 70, 6227–6234. virus RNA in vitro by a purified RNA-dependent RNA polymerase. Cell Osman, T. A. M., and Buck, K. W. (1997). The tobacco mosaic virus RNA 63, 363–368. polymerase complex contains a plant protein related to the RNA- Hayes, R. J., and Buck, K. W. (1993). Analysis of replication complexes binding subunit of yeast eIF-3. J. Virol. 71, 6075–6082. of positive strand RNA plant viruses. In “Molecular Virology: A Prac- Quadt, R., and Jaspars, E. M. (1990). Purification and characterization of tical Approach” (A. J. Davison and R. M. Elliott, Eds.), pp. 1–34. IRL brome mosaic virus RNA-dependent RNA polymerase. Virology 178, Press, Oxford. 189–194. TEMPLATE-DEPENDENT EXTRACTS FROM N. benthamiana 451

Quadt, R., and Jaspars, E. M. (1991). Characterization of cucumber mosaic Skryabin, K. G., Morozov, S. Y., Kraev, A. S., Rozanov, M. N., Chernov, virus RNA-dependent RNA polymerase. FEBS Lett. 279, 273–276. B. K., Lukasheva, L. I., and Atabekov, J. G. (1988b). Conserved and Quadt, R., Rosdorff, H. J., Hunt, T. W., and Jaspars, E. M. (1991). Analysis variable elements in RNA genomes of potexviruses. FEBS Lett. 240, of the protein composition of alfalfa mosaic virus RNA- dependent 33–40. RNA polymerase. Virology 182, 309–315. Song, C., and Simon, A. E. (1994). RNA-dependent RNA polymerase Rouleau, M., Bancroft, J. B., and Mackie, G. A. (1993). Partial purification from plants infected with turnip crinkle virus can transcribe (ϩ)-and and characterization of foxtail mosaic potexvirus RNA-dependent (Ϫ)-strands of virus-associated RNAs. Proc. Natl. Acad. Sci. USA 91, RNA polymerase. Virology 197, 695–703. 8792–8796. Rozanov, M. N., Koonin, E. V., and Gorbalenya, A. E. (1992). Conserva- van Dyke, T. A., and Flanegan, J. B. (1980). Identification of poliovirus tion of the putative methyltransferase domain: A hallmark of the polypeptide p63 as a soluble RNA-dependent RNA polymerase. J. Vi- “Sindbis-like” supergroup of positive-strand RNA viruses. J. Gen. rol. 35, 732–740. Virol. 73, 2129–2134. Wang, J., and Simon, A. E. (1997). Analysis of the two subgenomic RNA Saunders, K., and Kaesberg, P. (1985). Template-dependent RNA poly- promoters for turnip crinkle virus in vivo and in vitro. Virology 232, merase from black beetle virus-infected Drosophila melanogaster 174–186. cells. Virology 147, 373–381. Watanabe, T., Honda, A., Iwata, A., Ueda, S., Hibi, T., and Ishihama, Siegel, R. W., Adkins, S., and Kao, C. C. (1997). Sequence-specific A. (1999). Isolation from tobacco mosaic virus-infected tobacco of recognition of a subgenomic RNA promoter by a viral RNA polymer- a solubilized template-specific RNA-dependent RNA polymerase ase. Proc. Natl. Acad. Sci. USA 94, 11238–11243. containing a 126K/183K protein heterodimer. J. Virol. 73, 2633– Singh, R. N., and Dreher, T. W. (1997). Turnip yellow mosaic virus RNA-dependent RNA polymerase: Initiation of minus strand synthe- 2640. sis in vitro. Virology 233, 430–439. Wengler, G., and Gross, H. S. (1979). Replicative form of Semliki Forest Sivakumaran, K., and Kao, C. C. (1999). Initation of genomic plus-strand virus RNA contains an unpaired guanosine. Nature 82, 754–756. Ј RNA synthesis from DNA and RNA templates by a viral RNA-depen- Wu, G., and Kaper, J. M. (1994). Requirement of 3 -terminal guanosine in Ϫ dent RNA polymerase. J. Virol. 73, 6415–6423. ( )-stranded RNA for in vitro replication of cucumber mosaic virus Sivakumaran, K., Kim, C.-H., R. T., Jr., and Kao, C. C. (1999). RNA satellite RNA by viral RNA-dependent RNA polymerase. J. Mol. Biol. sequences and secondary structural determinants in a minimal viral 238, 655–657. promoter the directs replicase recognition and initation of genomic Wu, S. X., Ahlquist, P., and Kaesberg, P. (1992). Active complete in vitro plus-strand RNA synthesis. J. Mol. Biol. 294, 667–682. replication of nodavirus RNA requires glycerophospholipid. Proc. Skryabin, K. G., Kraev, A. S., Morozov, S. Y., Rozanov, M. N., Chernov, Natl. Acad. Sci. USA 89, 11136–11140. B. K., Lukasheva, L. I., and Atabekov, J. G. (1988a). The nucleotide Xiong, Z. G., and Lommel, S. A. (1991). Red clover necrotic mosaic virus sequence of potato virus X RNA. Nucleic Acids Res. 16, 10929–10930. infectious transcripts synthesized in vitro. Virology 182, 388–392.