RNA Binding Activity of NIa Proteinase of Tobacco Etch Potyvirus

VIROLOGY

237, 327–336 (1997) VY978802

ARTICLE NO.

RNA Binding Activity of NIa Proteinase of Tobacco Etch Potyvirus Jose´-Antonio Daro`s and James C. Carrington1 Institute of Biological Chemistry, Washington State University, Pullman, Washington 99164; and Department of Biology, Texas A&M University, College Station, Texas 77843 Received April 25, 1997; returned to author for revision June 24, 1997; accepted August 19, 1997 The C-terminal domain of NIa protein (NIaPro) from tobacco etch potyvirus (TEV) is a sequence-specific proteinase required for processing of the viral polyprotein. This proteinase also interacts with NIb, the TEV RNA-dependent RNA polymerase. NIaPro and two NIaPro-containing polyproteins (NIa and 6/NIa) were analyzed from extracts of recombinant Escherichia coli. Using RNA–protein blot and UV-crosslinking assays, NIaPro and the NIaPro-containing polyproteins were shown to possess RNA-binding activity. NIaPro bound nonspecifically to several RNAs, including plus- and minus-strands of the TEV 5* and 3* noncoding regions. Saturation binding data obtained using the UV-crosslinking assay were consistent with a possible cooperative RNA-binding activity of NIaPro. In addition, the RNA-binding activities of NIaPro and full-length NIa protein were similar. Based on its RNA-binding activity and other known functions, NIaPro or a NIaPro-containing polyprotein is proposed to serve one or more direct roles during TEV RNA synthesis. q 1997 Academic Press

INTRODUCTION

Carrington, 1995b). The cysteine-type HC-Pro participates in genome amplification, long-distance movement of virus within plants, and transmission by aphid vectors (Atreya et al., 1992; Atreya and Pirone, 1993; Cronin et al., 1995; Klein et al., 1994; Thornbury et al., 1985). Both P1 and HC-Pro encoded by related potyviruses have been shown to possess nonspecific RNA-binding activity (Brantley and Hunt, 1993; Maia and Bernardi, 1996; Soumounou and Laliberte, 1994). The NIa proteinase is actually a polyprotein consisting of two domains separated by an inefficiently utilized NIa cleavage site (Dougherty and Parks, 1991; Schaad et al., 1996). The NIa proteinase, which is most closely related to the picornaviral 3C proteinase, contains a fold similar to serine-type proteinases but with a Cys rather than Ser at the active site (Allaire et al., 1994; Bazan and Fletterick, 1989; Dougherty et al., 1989b; Gorbalenya et al., 1989; Matthews et al., 1994). Specificity of the proteinase involves an extended heptapeptide motif surrounding each cleavage site (Carrington and Dougherty, 1988; Dougherty et al., 1989a). Within the proteolytic domain is a suboptimal NIa cleavage site located 24 amino acid residues from the carboxyl terminus, processing at which may regulate enzymatic activity (Kim et al., 1995; Parks et al., 1995). The amino-terminal region of NIa contains the VPg domain which, either independently or as part of full-length NIa, attaches covalently to the 5* terminus of viral RNA via a phosphodiester linkage with Tyr62 (Murphy et al., 1990a, 1990b; Riechmann et al., 1989; Shahabuddin et al., 1988). The VPg provides an essential activity (Murphy et al., 1996; Schaad et al., 1996), most likely serving during the initiation steps of viral RNA synthesis. In addition to its proteolytic and VPg functions,

Tobacco etch virus (TEV) is a member of the family Potyviridae, which belongs to the ‘‘picornavirus supergroup’’ of positive-strand RNA viruses. This virus contains an RNA genome of approximately 10 kb that is polyadenylated at the 3* end and covalently linked to a virusencoded protein (VPg) at the 5* end (Riechmann et al., 1992). The TEV genome is translated into a single polyprotein that undergoes proteolytic processing catalyzed by three virus-encoded proteinases—P1, HC-Pro, and NIa (Dougherty and Semler, 1993). The P1 and HC-Pro proteinases each catalyze only one autoproteolytic reaction at their respective carboxyl termini (Carrington et al., 1989; Verchot et al., 1991), while the NIa proteinase catalyzes the remaining seven or eight cleavages through a series of autoproteolytic and trans-proteolytic reactions (Carrington and Dougherty, 1987b; GarcõB a et al., 1989; Hellmann et al., 1988). Each proteinase has a general organization in which the proteolytically active region is located proximal to the carboxyl terminus in an independently folded domain (Carrington et al., 1989; Carrington and Dougherty, 1987b; Dougherty and Parks, 1991; Verchot et al., 1991). Besides their key role in polyprotein processing, each of the TEV proteinases serves functions during other phases of the infection process. The serine-type P1 proteinase serves as a trans-active accessory factor during genome amplification, although the precise function of P1 in the replication process is not clear (Verchot and

1 To whom correspondence and reprint requests should be addressed. Fax: (509) 335-2482. E-mail: [email protected].

0042-6822/97 $25.00

327

AID

VY 8802

/

6a4f$$$561

09-30-97 21:53:18

Copyright q 1997 by Academic Press All rights of reproduction in any form reserved.

vira

AP: VY

DARO`S AND CARRINGTON

328

NIa interacts with NIb, the RNA-dependent RNA polymerase, in the yeast two-hybrid system (Hong et al., 1995; Li et al., 1997). This interaction involves primarily the NIaPro domain of NIa (Li et al., 1997). These data led to the suggestion that, in addition to its proteolytic function, NIa serves multiple roles during initiation of RNA synthesis by recruiting NIb polymerase through protein–protein interactions and by priming RNA synthesis through the VPg activity (Li et al., 1997; Schaad et al., 1996). Picornaviral 3C proteinase also performs functions during RNA replication. The 3C proteinase appears to interact with a 5*-proximal cloverleaf structure to promote positive-strand RNA synthesis, and this interaction is enhanced when 3C is in the context of the 3CD polyprotein (Andino et al., 1993, 1990a, 1990b; Harris et al., 1994; Xiang et al., 1995). The 3CD polyprotein also interacts with 3AB, the VPg precursor (Lama et al., 1995; Molla et al., 1994; Paul et al., 1994; Xiang et al., 1995). These interactions may provide a mechanism to deliver 3D polymerase to initiation sites after 3C-mediated cleavage. Although NIa performs pivotal functions during initiation of potyviral RNA synthesis, nothing is known about its potential to interact with RNA. Here, we identify and analyze a RNA-binding activity associated with the proteolytic domain of NIa (NIaPro).

His, and Cys339 to Ala substitution. pBS5NCR and pBS3NCR contained cDNA representing the 5* and 3* noncoding region (NCR) of TEV RNA, respectively, inserted between the KpnI and SacI sites of the plasmid pBluescript KS (0) (Stratagene). All plasmids were constructed by PCR amplification of the corresponding TEV cDNA from plasmid pTEV-7DA or pTEV-7DA(C-A) using Pfu DNA polymerase (Stratagene). Expression and purification of TEV proteins

Plasmid pTEV-7DA (Verchot and Carrington, 1995a) contained the complete complementary DNA representing the TEV genome (Allison et al., 1986). pTEV-7DA(CA) was identical to pTEV-7DA except for substitution of an Ala codon for the Cys339 codon within the NIa coding sequence (Carrington and Dougherty, 1987a). Plasmid pTrc7H-Pro (Parks et al., 1995) contained the coding sequence for TEV NIaPro (codons 189–430 within the NIa sequence), with an additional 11 amino acids (MMHHHHHHHAM) at the amino terminus, cloned in the Escherichia coli expression vector pTrc99A (Pharmacia). Several plasmids were constructed using the E. coli expression vector pET-23d(/) (Novagen). pETNIaPro contained the NIaPro coding sequence with a Met initiator codon. pETNIaPro6H contained the NIaPro coding sequence with a Met initiator codon and a six-His coding sequence at the 3* end. pETNIaPro(C-A)6H was identical to pETNIaPro6H except for the Cys339 to Ala codon substitution within the NIaPro coding sequence. pETVPg6H contained the coding region of the VPg domain of NIa (codons 1–188) with initiator Met and carboxyl-terminal sixHis coding sequences. pETNIa(C-A)6H contained the full-length NIa coding sequence with an initiator Met and carboxyl-terminal six-His sequences, as well as with the Cys339 to Ala codon substitution. pET6/NIa(C-A)6H contained the coding sequence of the 6-kDa/NIa polyprotein with an amino-terminal Met-Gly, carboxyl-terminal six-

E. coli [strain BL21(DE3)pLysS, Novagen] was grown to 0.6 OD600 in Luria broth containing 50 mg/L ampicillin and 34 mg/L chloramphenicol. Expression was induced by addition of isopropyl thio-b-D-galactoside (IPTG) to a final concentration of 0.4 mM. Cells were harvested 2 hr later by centrifugation (5500 g for 10 min) and washed once in 100 mM Tris–HCl, pH 7.5. When extracted under denaturing conditions, cells were resuspended in 0.01 original volume of 6 M guanidine hydrochloride, 100 mM Tris–HCl, pH 7.5, and the lysates were clarified by centrifugation (12,000 g for 15 min). When extracted under native conditions, cells were resuspended in 0.01 original volume of 100 mM Tris–HCl, pH 7.5, and frozen at 0807. After thawing, phenylmethylsulfonyl fluoride (PMSF), MgCl2 , and DNase I were added to final concentrations of 1 mM, 1 mM, and 20 U/ml, respectively, and the lysate was shaken gently for 30 min at room temperature. Finally, the lysates were clarified by centrifugation (12,000 g for 15 min) and NaCl added to the supernatant to a final concentration of 300 mM. Proteins were isolated by affinity (Ni2/-NTA Superflow, Qiagen) and cation exchange (SP Sepharose, Pharmacia) chromatography using a Biocad Sprint system (PerSeptive Biosystems) equipped with 4.6 1 50-mm columns (0.83 ml column volume, cv). The columns were packed at 3 ml/min and run at 2 ml/min. Ni2/-NTA columns were packed in 100 mM Tris–HCl, pH 7.5, 300 mM NaCl and SP Sepharose columns in 25 mM sodium phosphate, pH 7.0. For purification under denaturing conditions, the Ni2/-NTA column was equilibrated with 10 cv of 6 M guanidine hydrochloride, 100 mM Tris–HCl, pH 7.5. Extract (2 ml) was loaded and the column was washed with 10 cv of equilibration buffer, 10 cv of 8 M urea, 300 mM NaCl, 100 mM Tris–HCl, pH 7.5, and then 10 cv of the same buffer including 50 mM imidazol. Finally the proteins were eluted with 10 cv of 8 M urea, 300 mM NaCl, 100 mM Tris–HCl, pH 7.5, 250 mM imidazol. The elution of proteins was monitored by absorbance at 280 nm through a continuous flow cell. Fractions were collected, concentrated approximately 10-fold with a Centricon centrifugal concentrator (Amicon) spun at 5000 g, and analyzed by 12.5% SDS–PAGE and Coomassie blue staining. NIaPro and NIa(C-A) were also purified under native conditions using Ni2/-NTA columns that were equili-

AID

vira

MATERIALS AND METHODS Plasmids

VY 8802

/

6a4f$$$562

09-30-97 21:53:18

AP: VY

TEV NIa PROTEINASE BINDS RNA

brated with 10 cv of 100 mM Tris–HCl, pH 7.5, 300 mM NaCl. Columns were loaded with 25 ml of extract and washed with 10 cv of equilibration buffer, 10 cv of 50 mM sodium phosphate, pH 7.0, 300 mM NaCl, and 10 cv of 50 mM sodium phosphate, pH 7.0, 10% glycerol. A final wash was done with 10 cv of 50 mM sodium phosphate, pH 7.0, 10% glycerol, 100 mM imidazol, and proteins were eluted with 10 cv of 50 mM sodium phosphate, pH 7.0, 10% glycerol, 250 mM imidazol. NIaPro was further purified by cation exchange chromatography. Protein-containing fractions eluting from the Ni2/-NTA column were identified by monitoring absorbance (280 nm), pooled, and loaded onto a SP Sepharose column that had been equilibrated with 1 cv of 2 M NaCl, 25 mM sodium phosphate, pH 7.0, 10% glycerol, followed by 10 cv of 25 mM sodium phosphate, pH 7.0, 10% glycerol. The protein was eluted with a linear gradient (10 cv) of NH4Cl (0 to 0.5 M) in 25 mM sodium phosphate, pH 7.0, 10% glycerol. The presence and purity of NIaPro in the eluted fractions were analyzed in duplicate 12.5% SDS–PAGE gels, one stained with Coomassie blue and the other processed for immunoblot analysis using polyclonal anti-NIa serum (Restrepo et al., 1990) and a chemiluminescence detection system. Selected fractions were pooled and the protein was concentrated approximately 10-fold with a Centricon centrifugal concentrator spun at 5000 g. The protein preparation was desalted by size exclusion chromatography using a Sephadex G-50 spin column (Boehringer) equilibrated with 25 mM sodium phosphate, pH 7.0, 10% glycerol, divided into aliquots and stored at 0807. The concentration of NIaPro and NIa(C-A) in each preparation was determined by measuring the absorbance at 280 nm and using the molar extinction coefficient of 33,810 and 42,710 M01 cm01, respectively, which were calculated for the recombinant proteins using the program PEPDATA from the GCG nucleic acid sequence analysis package (Devereux et al., 1984). In vitro synthesis of RNA

329

radiation. Plasmid pBS5NCR was used to synthesize positive or negative polarity transcripts corresponding to the TEV 5* NCR by linearization/transcription reactions using BamHI/T3 RNA polymerase or EcoRI/T7 RNA polymerase, respectively. Plasmid pBS3NCR was used to synthesize positive or negative polarity transcripts corresponding to the TEV 3* NCR by linearization/transcription reactions using BglII/T3 RNA polymerase or EcoRI/T7 RNA polymerase, respectively. Nonspecific TEV-related RNA transcripts were synthesized with pGEM-4 as template using T7 RNA polymerase after linearization with NheI. RNA–protein blot assay In experiments with extracts of total E. coli proteins, bacterial cultures (5 ml) were grown to 0.6 OD600 and induced with IPTG (0.4 mM). Cells were harvested 2 hr later by centrifugation and total SDS-soluble proteins were extracted by boiling in dissociation buffer (0.5 ml). In experiments with proteins isolated under denaturing conditions, the amounts of NIaPro, NIaPro(C-A), VPg domain, NIa(C-A), and 6/NIa(C-A) preparations were normalized by comparing the intensity of the blue Coomassie-stained bands. Normalized amounts of the purified proteins were subjected to 12.5% SDS–PAGE and transferred electrophoretically to nitrocellulose membranes. The membranes were blocked for 1 hr in RNA-binding buffer (20 mM HEPES, 6 mM Tris, 25 mM NaCl, 5 mM MgCl2 , 1 mM EDTA, 1 mM DTT, pH 7.0) plus 5% nonfat milk. After three washes with RNA-binding buffer, the proteins were denatured with two successive incubations for 5 min in 6 M guanidine hydrochloride in RNAbinding buffer. The proteins were renatured by successive incubations of 5 min each in RNA-binding buffer with reduced concentrations of guanidine hydrochloride (3, 1.5, 0.75, 0.37, 0.19, and 0 M) (Masson et al., 1993). The membranes were washed in RNA-binding buffer plus 0.1% Nonidet P-40 and then incubated for 1 hr at room temperature with 106 cpm/ml of 32P-labeled TEV 5*NCR (0) RNA in this last buffer. The membranes were washed three times for 5 min in RNA-binding buffer plus 0.1% Nonidet P-40, dried, and subjected to autoradiography.

In vitro transcription reactions were carried out in a volume of 50 ml in 40 mM Tris–HCl, pH 7.9, 6 mM MgCl2 , 2 mM spermidine, 10 mM dithiothreitol (DTT), 2 mM each NTP, 10 mg of linearized plasmid, 30 U RNasin, and 100 U of the appropriate RNA polymerase for 2 hr at 407. Fifty units of RNase-free DNase I were then added followed by incubation for 30 min at 377. To synthesize 32P-labeled RNA at different specific activities, nonlabeled UTP was replaced by 150 mCi of [a-32P]UTP, either at 800 Ci/mmol or 8 Ci/mmol, and the transcription and DNase I reactions were shortened to 30 and 15 min, respectively. The transcripts were extracted using phenol-chloroform and filtered through a Sephadex G-50 spin column preequilibrated with DEPC-treated deionized H2O. Transcripts were quantified by measuring the absorbance at 260 nm and, in the case of the labeled transcripts, Cerenkov

In a typical experiment, labeled RNA probes were incubated with NIaPro or NIa(C-A) in binding buffer (unless stated otherwise, 20 mM MES, 9 mM Tris–HCl, 25 mM NaCl, 1 mM EDTA, 1 mM DTT, pH 6.0) for 1 hr at room temperature in a total volume of 10 ml, deposited as a drop on a parafilm strip, and irradiated on ice with UV light (1.8 J) using a Stratalinker (Stratagene). The concentrations of protein and probe used in each experiment are stated throughout the text. Nonprotected RNA was digested for 1 hr at 377 by the addition of 1 mg of RNase A. The proteins were subjected to 12.5% SDS–PAGE.

AID

vira

VY 8802

/

6a4f$$$562

09-30-97 21:53:18

UV crosslinking/label transfer assay

AP: VY

DARO`S AND CARRINGTON

330

The gel was dried and subjected to autoradiography or phosphorimage analysis using a Phosphorimager Fujix Bas 2000 (Fuji). Radioactivity levels were measured as photostimulated luminescence (PSL) units. A variety of buffers were used to test the effect of pH and ionic conditions on the RNA-binding properties of NIaPro. Buffers used to test the effects of pH contained 1 mM EDTA, 1 mM DTT, and the following: pH 6.0 buffer, 20 mM MES, 9 mM Tris; pH 6.5 buffer, 20 mM MES, 14 mM Tris; pH 7.0 buffer, 20 mM HEPES, 6 mM Tris; pH 7.5 buffer, 20 mM HEPES, 12.5 mM Tris; pH 8.0 buffer, 20 mM HEPES, 24 mM Tris; or pH 8.5 buffer, 20 mM Tris, 7 mM HEPES. Buffers used to test the effects of various salts contained NaCl, KCl, or NH4Cl at concentrations indicated in the Results. Immunoprecipitation NIaPro was crosslinked with 32P-labeled TEV 5*NCR (0) RNA and the unprotected RNA was digested with RNase A as described above. This preparation was denatured by addition of SDS (2%), subjected to centrifugation to remove insoluble material, and diluted 10-fold with 50 mM Tris–HCl, pH 8.0, 100 mM NaCl, 1% Nonidet P40, 0.1% SDS, 0.2% sodium deoxycholate. The mixture was submitted to immunoprecipitation analysis using polyclonal anti-NIa and anti-HC-Pro antibodies essentially as described previously (Verchot et al., 1991). Immunoprecipitated proteins were analyzed by SDS–12.5% PAGE and autoradiography.

RNA bound to the NIa and 6/NIa bands was relatively high compared to that bound to the NIaPro band. However, in multiple experiments, this differential level of RNA binding was not consistent. Relatively little RNA-binding activity was detected with the VPg domain (Fig. 1B, lane 3) or with any proteins present in the set of molecular mass standards (lane M*). As each recombinant protein tested in the experiment shown in Figs. 1A and 1B contained a His tag, it was possible that the RNA-binding activity of NIa, 6/NIa, and NIaPro was due to a nonspecific effect of the additional His residues. To test this possibility, NIaPro was expressed in E. coli with and without the His-tag, and total protein extracts containing these proteins were subjected to SDS–PAGE and either Coomassie blue staining or RNA–protein blot assay. Extracts from control vector pET-23d(/)-transformed E. coli, and purified NIaPro containing the His-tag, were also tested in parallel. The NIaPro variants without (Fig. 1C, lane 2; 27 kDa) and with (lane 3; 28 kDa) the His tag were detected by Coomassie blue staining among a complex background of E. coli proteins. In the RNA–protein blot assay, the His-tagged and non-His-tagged NIaPro variants in the total protein extracts, and purified NIaPro, bound to the radiolabeled probe (Fig. 1D, lanes 2–4). Only two E. coli-derived bands were labeled with the probe—one with an electrophoretic position corresponding to approximately 35 kDa and one at or near the SDS–PAGE running front (Fig. 1D, lanes 1–3). These results suggest that RNA-binding activity of NIaPro in this assay is independent of the presence or absence of a His-tag.

RESULTS NIaPro binds RNA in a RNA–protein blot assay

Purification of NIaPro and RNA-binding activity in a UV crosslinking/label transfer assay

Initial experiments were designed to determine whether or not NIa possesses an RNA-binding activity using an RNA–protein blot assay. The potential RNAbinding activity of the N-terminal VPg domain, C-terminal proteinase (NIaPro) domain, and 6/NIa polyprotein were also tested. To prevent NIa-mediated processing between the VPg and proteinase domains of NIa, and between the 6-kDa protein and NIa within 6/NIa polyprotein, substitutions resulting in a change of Cys339 to Ala at the NIa active site were introduced. In addition, the effect of the Cys339 to Ala substitution was tested in NIaPro. Each protein contained a carboxyl-terminal His-tag and was enriched by affinity chromatography under denaturing conditions. Major proteins with electrophoretic mobilities consistent with those of the expected sizes were detected after SDS–PAGE and staining with Coomassie blue (Fig. 1A). In RNA–protein blots using 32P-labeled 5* NCR(0) TEV RNA as a probe, both NIa and 6/NIa bound RNA (Fig. 1B, lanes 4 and 5). The probe also bound to NIaPro, regardless of whether or not the protein contained the Cys339 to Ala substitution (Fig. 1B, lanes 1 and 2). In this experiment, the intensity of radiolabeled

To study the RNA-binding activity of NIaPro in greater detail, the His-tagged NIaPro from pTrc7H-Pro-containing cells was purified under nondenaturing conditions. For simplicity, the His-tagged derivative protein subsequently will be referred to as NIaPro. Soluble NIaPro was purified by Ni2/-affinity, cation exchange, and size-exclusion chromatography steps. Aliquots at each stage were resolved by SDS–PAGE and analyzed by staining with Coomassie blue or by anti-NIa immunoblot. An anti-NIa-reactive protein corresponding in size to the anticipated product (28 kDa) was detected in total nonfractionated extracts (Figs. 2A and 2B, lane 1) and was enriched after the Ni2/-NTA chromatography step (Figs. 2A and 2B, lane 2). The minor amounts of contaminating protein that eluted from the Ni2/-NTA column were reduced substantially by SP Sepharose chromatography, during which NIaPro eluted at 200–300 mM NH4Cl (Figs. 2A and 2B, lane 3). Protein used for all subsequent experiments was desalted using size-exclusion chromatography (Figs. 2A and 2B, lane 4). At all stages of purification, a second anti-NIa-reactive species that migrated slightly

AID

vira

VY 8802

/

6a4f$$$562

09-30-97 21:53:18

AP: VY

TEV NIa PROTEINASE BINDS RNA

331

FIG. 1. Analysis of RNA-binding activity of NIaPro, VPg domain, NIa, and 6/NIa polyprotein by RNA–protein blot assay. (A) Coomassie brilliant blue staining of a SDS–PAGE gel containing different His-tagged proteins: NIaPro (lane 1), NIaPro (C-A) (lane 2), VPg domain (lane 3), NIa (C-A) (lane 4), 6/NIa (C-A) (lane 5) or molecular mass standards (lanes M and M*). (B) RNA–protein blot. Proteins from a SDS–PAGE gel identical to that shown in (A) were transferred to a nitrocellulose membrane, probed with 32P-labeled TEV 5*NCR (0) RNA, and subjected to autoradiography. (C) Coomassie brilliant blue staining of a SDS–PAGE gel containing total SDS-soluble proteins from E. coli transformed with the plasmids indicated (lanes 1–3), purified His-tagged NIaPro from pTrc7H-Pro-transformed cells (lane 4) or molecular mass standards (lane M). (D) RNA–protein blot. Proteins from an SDS–PAGE gel identical to that shown in (C) were transferred to a nitrocellulose membrane, probed with 32P-labeled TEV 5*NCR (0) RNA, and subjected to autoradiography. The positions of molecular mass markers (lanes M) (shown in kilodaltons) are indicated at the left of each panel.

faster than NIaPro was detected. This protein most likely corresponds to the carboxyl-terminal truncated, self-processed form of NIaPro that lacks 24 residues (Kim et al., 1995; Parks et al., 1995). Both the full-length and truncated forms of NIaPro bound radiolabeled RNA in the

FIG. 2. Analysis of NIaPro by SDS–PAGE after sequential chromatographic steps. (A) Coomassie brilliant blue staining. (B) Anti-NIa immunoblot. Samples analyzed were from the soluble E. coli extract (lane 1), eluate from Ni2/-NTA affinity column (lane 2), NIaPro-containing peak fraction from SP Sepharose column (lane 3), flow-through of Sephadex G-50 column (lane 4), and protein molecular mass markers (lane 5). The positions of molecular mass markers (shown in kilodaltons) are indicated at the left of (A).

nitrocellulose-based RNA–protein blot assay (Fig. 1D, lane 4). The RNA-binding activity of soluble NIaPro was tested using a UV crosslinking/label transfer assay with a 32Plabeled RNA probe corresponding to the (0)-strand of the TEV 5* noncoding region (NCR). Probe (0.5 nM) was incubated with the protein preparation purified with the Ni2/-NTA column and the fractions were recovered from the SP Sepharose column. Prior to crosslinking, the amount of NH4Cl was normalized in each fraction (50 mM in binding assay). The mixtures were incubated for 1 hr, subjected to UV radiation, digested with RNase A, and analyzed by SDS–PAGE and autoradiography. Also, the presence of NIaPro in each fraction was tested by immunoblot assay. A 32P-labeled species corresponding in size to NIaPro was detected only in the fractions where NIaPro was present (Figs. 3A and 3B). To confirm that the protein labeled in the crosslinking assays was NIaPro, the 32P-labeled crosslinked product was subjected to immunoprecipitation analysis with anti-NIa and antiHC-Pro sera. The anti-NIa serum reacted with the crosslinked protein, while the anti-HC-Pro serum did not (Fig. 3C). These data demonstrate that soluble NIaPro possesses RNA-binding activity in vitro. The UV crosslinking/label transfer assay was used to test the binding of NIaPro to different 32P-labeled RNA probes. Probes (0.5 nM) corresponding to the (/)- and (0)-strands of the TEV 5* and 3* NCRs and a sequence derived from the vector pGEM-4 were incubated with or without NIaPro (1 mM) and subjected to UV-crosslinking. After RNase A digestion, SDS–PAGE, and autoradiogra-

AID

vira

VY 8802

/

6a4f$$$563

09-30-97 21:53:18

AP: VY

DARO`S AND CARRINGTON

332

was also detected in some reactions (Fig. 4) and might correspond to a crosslinked dimer of NIaPro. The probe representing the 5* NCR(0) consistently provided the best label transfer in the crosslinking experiments and was used in all subsequent experiments. Nucleic acid binding activity of NIaPro is highly dependent on pH, ionic conditions, and type of nucleic acid To identify optimal conditions under which NIaPro binds RNA, the effects of pH, ionic strength, and ionic composition were analyzed using the UV crosslinking assay with 32P-labeled 5* NCR(0) RNA. Protein (1 mM) was incubated with probe (0.5 nM) for 5 min at room temperature in buffers ranging between pH 6.0 and 8.5 prior to crosslinking. The extent of labeling of NIaPro was determined quantitatively by phosphoimage analysis. Maximal binding occurred at pH 6.0, with activity decreasing as the pH was increased (Fig. 5A). The effect of NaCl concentration at pH 6.5 was tested (Fig. 5B). The NaCl concentration at which binding was reduced to 50% of maximal levels (I50) was 112 mM. The inhibitory effect of salt was independent of the cation composition, as NaCl, KCl, and NH4Cl affected the RNA-binding reaction to similar extents (data not shown). The specificity of the NIaPro RNA-binding reaction was tested in crosslinking experiments using various ssRNA, ssDNA, and dsDNA molecules as competitors. NIaPro (1 mM) was incubated with 32P-labeled 5*NCR(0) probe (0.45 nM) plus nonlabeled pGEM-4-derived RNA transcripts, ssDNA phagemid corresponding to pTL7SN-0027 (Carrington et al., 1990) and linearized pBS5NCR dsDNA, each at 1-, 10-, 100-, and 1000-fold weight excess over the radiolabeled probe. The ssRNA and ssDNA competitors reduced bindFIG. 3. Copurification of NIaPro and RNA-binding activity. (A) UVcrosslinking assay using 32P-labeled TEV 5*NCR (0) RNA probe and no protein (lane 1), eluate from the Ni2/-NTA column (lane 2), flowthrough from the SP Sepharose column (lane 3) or fractions eluted from the SP Sepharose column with a NH4Cl gradient (lanes 4–10). Crosslinked products were detected by SDS–PAGE and autoradiography. (B) Anti-NIa immunoblot analysis of the same fractions used in lanes 2–10 in the UV-crosslinking assay shown in (A). (C) Immunoprecipitation of the radiolabeled NIaPro-crosslinked product. 32P-labeled TEV 5*NCR (0) RNA was crosslinked in the presence of no protein (lane 1) or purified NIaPro (lanes 2–4) and treated with RNase A. Products were subjected to no immunoprecipitation (lanes 1 and 2) or immunoprecipitation with anti-NIa (lane 3) or anti-HC-Pro (lane 4) sera. Crosslinked products were detected by autoradiography. The positions of molecular mass markers (shown in kilodaltons) are indicated at the left of each panel.

phy, a 32P-labeled species corresponding in size to NIaPro was detected using each probe and only in reactions containing protein (Fig. 4). A weak band or set of bands that migrated to the electrophoretic position corresponding to protein(s) of approximately 60-kDa molecular mass

FIG. 4. Analysis of NIaPro RNA-binding activity using various radiolabeled RNA probes. 32P-labeled RNA probes were TEV 5*NCR (/) (lanes 1 and 2), TEV 5*NCR (0) (lanes 3 and 4), TEV 3*NCR (/) (lanes 5 and 6), TEV 3*NCR (0) (lanes 7 and 8), and an RNA derived from the polylinker region of pGEM-4 (lanes 9 and 10). Each probe was subjected to UV-crosslinking in the presence or absence of purified NIaPro. Reaction products were digested with RNase A, resolved by SDS– PAGE, and detected by autoradiography. The positions of molecular mass markers (shown in kilodaltons) are indicated at the left.

AID

vira

VY 8802

/

6a4f$$$563

09-30-97 21:53:18

AP: VY

TEV NIa PROTEINASE BINDS RNA

333

1.3 1 1006 M and a Hill coefficient of 4.5 (Fig. 6B). The sigmoidal curve shown in Fig. 6A and a Hill coefficient of 4.5 is consistent with NIaPro binding to RNA in a cooperative manner. As full-length NIa could conceivably possess RNAbinding activities that differ from NIaPro, RNA-binding activity with purified, His-tagged NIa was analyzed using the crosslinking assay with 32P-labeled 5*NCR(0) probe. The proteolytically inactive variant of NIa (Cys339 to Ala substitution) was purified under native conditions (data not shown), and 0.01 to 5 mM protein was used in the crosslinking assay with 50 nM probe. The saturation curve obtained for full-length NIa was comparable to NIaPro (Fig. 6C). The Hill plot revealed a midpoint of maximal binding at 1.5 1 1006 M and Hill coefficient of 4.0 (Fig. 6D). These data indicate that full-length NIa and NIaPro possess similar RNA-binding activities. DISCUSSION

FIG. 5. Dependence of NIaPro RNA-binding activity on pH and ionic conditions. NIaPro and 32P-labeled TEV 5*NCR (0) RNA were reacted at different pH and ionic conditions, and subjected to UV-crosslinking and digestion with RNase A. Radiolabeled products were resolved by SDS–PAGE and quantified by phosphoimage analysis. (A) Effect of pH. (B) Effect of NaCl at pH 6.5.

In this report, RNA–protein blots and UV-crosslinking assays were used to demonstrate that the proteinase domain of NIa possesses RNA-binding activity. The NIaPro RNA-binding activity was sequence-nonspecific and had a low pH optimum. NIaPro displayed characteristics

ing by 50% (I50) at calculated weight excesses of 65- and 74-fold, respectively, whereas the dsDNA competitor had a calculated I50 of 695-fold. In addition, experiments to test the competitor effects of the TEV full-length (/) and (0), TEV 5*NCR(/), TEV 5*NCR(0), TEV3*NCR(/), TEV3*NCR(0), and pGEM-4 transcripts were done. None of these RNAs exhibited competitor activity that was reproducibly higher than the activity of the others (data not shown), suggesting that the RNA-binding activity of NIaPro in these experiments was sequence-nonspecific. Possible cooperativity of NIaPro and NIa RNA-binding activity The RNA-binding properties of NIaPro were analyzed using the crosslinking assay with 32P-labeled 5*NCR(0) probe and varying amounts of NIaPro (0.01 to 5 mM). In this experiment, the maximum amount of probe (50 nM) that was practical to use was added to each reaction. Binding activity was not detected in reactions containing less than 0.8 mM NIaPro. However, binding activity was detected at 0.8 mM NIaPro and increased until reaching a maximum at approximately 2.0 mM (Fig. 6A). Saturation of binding was achieved at a concentration corresponding to approximately 50 NIaPro molecules per RNA molecule. The plot of NIaPro bound as a function of NIaPro concentration revealed a sigmoidal curve. A Hill plot was generated, revealing a midpoint of maximal binding at

AID

VY 8802

/

6a4f$$$563

09-30-97 21:53:18

FIG. 6. Saturation RNA-binding curves of NIaPro and NIa. (A and C) P-labeled TEV 5*NCR (0) RNA (50 nM) was bound to increasing amounts (0.01 to 5 mM) of NIaPro (A) or NIa (C) and subjected to UV crosslinking. After digestion with RNase A and SDS–PAGE, radiolabeled products were quantified by phosphoimage analysis. (B and D) Hill plot of RNA-binding activity of NIaPro (B) or NIa (D) using data obtained at protein concentrations between 1.2 and 2.0 mM, where Y is the ratio of protein bound to total protein. Maximal protein binding was determined as the mean protein binding measured at saturation (between 2.6 and 5 mM). The lines represent the best fit determined by least-squares analysis [(B) y Å 4.48x / 26.37, r Å 0.95, (D) y Å 3.98x / 23.19, r Å 0.91]. 32

vira

AP: VY

DARO`S AND CARRINGTON

334

consistent with that of a cooperative RNA-binding protein. In addition, similar RNA-binding activity was detected for full-length NIa. Three functions have now been associated with NIaPro—a sequence-specific proteinase activity (Carrington and Dougherty, 1987b), TEV RNA polymerase (NIb) binding activity (Hong et al., 1995; Li et al., 1997), and RNA-binding activity. This combination of activities leads to speculation that the proteinase domain, alone or as part of a polyprotein, serves direct functions during TEV RNA synthesis in addition to its role in polyprotein processing. In the titration experiments, the midpoints of maximal binding of NIaPro and NIa to RNA were determined to be 1.3 and 1.5 mM, respectively. It is conceivable that these values correspond to the macroscopic binding constant (Kd). If so, these values are considerably higher than those of many sequence-specific RNA-binding proteins, such as HIV tat and rev proteins (Kd Å 5 nM and 1 nM, respectively; Burd and Dreyfuss, 1994). On the other hand, they are comparable to other sequence nonspecific RNA-binding proteins, like the major hnRNP proteins (Kd Å 0.1 to 1 mM) (Burd and Dreyfuss, 1994) and the poliovirus 3D polymerase (Kd Å Ç3 mM) (Pata et al., 1995). However, interpretation of the binding saturation curves in terms of apparent Kd may not be appropriate. Although the RNA probe and protein in all UV crosslinking assays were allowed to reach equilibrium for 1 hr, it is possible that the proportion of protein bound was affected by the crosslinking reaction. In addition, it is possible that, due to the chemistry of the UV-crosslinking reaction, the reaction products measured did not accurately reflect the true fraction of protein bound to RNA. Because of these reservations, it is also difficult to conclude confidently that NIaPro and NIa are cooperative RNA-binding proteins, although the saturation curve data obtained are consistent with results expected from a cooperative RNAbinding activity. Both potyviral NIaPro and picornaviral 3C proteinase lack clearly recognizable conserved RNA-binding motifs (Burd and Dreyfuss, 1994). Crystallographic analysis of human rhinovirus-14 (HRV-14) 3C proteinase suggested that the presence and correct positioning of Arg87 are important for RNA-binding activity (Matthews et al., 1994). In fact, a substitution affecting Arg87 of poliovirus 3C eliminates virus viability (Hammerle et al., 1992). It was also proposed that a conserved domain, which in HRV14 3C proteinase consists of residues 74–90, is involved in the RNA-binding activity (Walker et al., 1995). In comparisons of sequences of TEV NIaPro and picornavirus 3C proteinases, picornaviral Arg87 aligns with Arg80 in NIaPro (unpublished observations). Residues critical for NIaPro RNA-binding activity, however, have yet to be identified experimentally. NIaPro bound in vitro to all RNAs assayed, including both (/) and (0) TEV 5* and 3* noncoding regions, and to RNA unrelated to TEV, with approximately the same

affinity. This is also the case for the two other potyviral proteinases, P1 and HC-Pro (Brantley and Hunt, 1993; Maia and Bernardi, 1996; Soumounou and Laliberte, 1994). It is possible that NIaPro normally binds RNA nonspecifically, that it binds specifically but to a sequence not analyzed in this study, or that the conditions under which putative sequence specificity occurs were not faithfully duplicated. It is also possible that host or other viral proteins participate in the physiological RNA-binding complexes. In this regard it is important to mention that poliovirus 3C proteinase binds more efficiently to its target RNA in the context of a 3CD polyprotein, and that another polyprotein, 3AB, also participates in the binding complex (see Introduction). Although a systematic analysis of NIaPro-containing polyproteins was not done, fulllength NIa lacked sequence specificity with any of the probes analyzed in this study (unpublished observations). NIaPro RNA-binding activity was inhibited by salt, showing an I50 of 112 mM NaCl at pH 6.5. This level of salt sensitivity was similar to that detected with cauliflower mosaic virus gene I protein (Citovsky et al., 1991), alfalfa mosaic virus movement protein (Schoumacher et al., 1992), and barley yellow mosaic virus capsid protein (Reichel et al., 1996), but was somewhat lower than that detected with potyviral P1 (Brantley and Hunt, 1993; Soumounou and Laliberte, 1994) and HC-Pro (Maia and Bernardi, 1996). Although the in vitro RNA-binding activity of NIaPro is relatively weak in the presence of salt, it is difficult to evaluate the significance of this observation. Conceivably, NIaPro could possess a weak binding activity to permit transient interaction with RNA. Alternatively, NIaPro RNA-binding activity in vivo could be strengthened by interaction with other viral or cellular proteins or by specific subcellular microenvironments. What might be the function of the RNA-binding activity of NIaPro? It is possible that NIb polymerase is recruited to initiation complexes through interaction with NIaPro that is bound to the RNA template. During initiation of (/)-strand RNA synthesis, NIaPro might interact with either the 3* end of a (0)-strand template or with the 5* end of the (/)-strand that is displaced from its complementary sequence in a partially denatured dsRNA structure. In either case, it is important to consider that the RNA-binding function of NIaPro may be relevant only within the context of a polyprotein. Two such polyproteins, NIa and 6/NIa, have been identified within TEVinfected cells (Restrepo-Hartwig and Carrington, 1994). In NIa, NIaPro is linked to the VPg domain. In 6/NIa, NIa is linked to a small endoplasmic reticulum-targeted polypeptide (Restrepo-Hartwig and Carrington, 1994; Schaad et al., 1997). The 6/NIa polyprotein has been proposed to serve as the critical membrane-bound VPg precursor during initiation of RNA synthesis. In this model, the NIaPro domain within a membrane-bound 6/ NIa polyprotein would bind near the terminus of a tem-

AID

vira

VY 8802

/

6a4f$$$563

09-30-97 21:53:18

AP: VY

TEV NIa PROTEINASE BINDS RNA

335

Allaire, M., Chernaia, M. M., Malcolm, B. A., and James, M. N. (1994). Picornaviral 3C cysteine proteinases have a fold similar to chymotrypsin-like serine proteinases. Nature 369, 72–76. Allison, R., Johnston, R. E., and Dougherty, W. G. (1986). The nucleotide sequence of the coding region of tobacco etch virus genomic RNA: Evidence for the synthesis of a single polyprotein. Virology 154, 9– 20. Andino, R., Rieckhof, G. E., Achacoso, P. L., and Baltimore, D. (1993). Poliovirus RNA synthesis utilizes an RNP complex formed around the 5*-end of viral RNA. EMBO J. 12, 3587–3598. Andino, R., Rieckhof, G. E., and Baltimore, D. (1990a). A functional ribonucleoprotein complex forms around the 5* end of poliovirus RNA. Cell 63, 369–380. Andino, R., Rieckhof, G. E., Trono, D., and Baltimore, D. (1990b). Substitutions in the protease (3Cpro) gene of poliovirus can suppress a mutation in the 5* noncoding region. J. Virol. 64, 607–612. Atreya, C. D., Atreya, P. L., Thornbury, D. W., and Pirone, T. P. (1992). Site-directed mutations in the potyvirus HC-PRO gene affect helper component activity, virus accumulation, and symptom expression in infected tobacco plants. Virology 191, 106–111. Atreya, C. D., and Pirone, T. P. (1993). Mutational analysis of the helper component-proteinase gene of a potyvirus: Effects of amino acid substitutions, deletions, and gene replacement on virulence and aphid transmissibility. Proc. Natl. Acad. Sci. USA 90, 11919–11923. Bazan, J. F., and Fletterick, R. J. (1989). Detection of a trypsin-like serine protease domain in flaviviruses and pestiviruses. Virology 171, 637– 639. Brantley, J. D., and Hunt, A. G. (1993). The N-terminal protein of the polyprotein encoded by the potyvirus tobacco vein mottling virus is an RNA binding protein. J. Gen. Virol. 74, 1157–1162. Burd, C. G., and Dreyfuss, G. (1994). Conserved structures and diversity of functions of RNA-binding proteins. Science 256, 615–621. Carrington, J. C., Cary, S. M., Parks, T. D., and Dougherty, W. G. (1989). A second proteinase encoded by a plant potyvirus genome. EMBO J. 8, 365–370. Carrington, J. C., and Dougherty, W. G. (1987a). Processing of the tobacco etch virus 49K protease requires autoproteolysis. Virology 160, 355–362.

Carrington, J. C., and Dougherty, W. G. (1987b). Small nuclear inclusion protein encoded by a plant potyvirus genome is a protease. J. Virol. 61, 2540–2548. Carrington, J. C., and Dougherty, W. G. (1988). A viral cleavage site cassette: Identification of amino acid sequences required for tobacco etch virus polyprotein processing. Proc. Natl. Acad. Sci. USA 85, 3391–3395. Carrington, J. C., Freed, D. D., and Oh, C.-S. (1990). Expression of potyviral polyproteins in transgenic plants reveals three proteolytic activities required for complete processing. EMBO J. 9, 1347–1353. Citovsky, V., Knorr, D., and Zambryski, P. (1991). Gene I, a potential cell-to-cell movement locus of cauliflower mosaic virus, encodes an RNA-binding protein. Proc. Natl. Acad. Sci. USA 88, 2476–2480. Cronin, S., Verchot, J., Haldeman-Cahill, R., Schaad, M. C., and Carrington, J. C. (1995). Long-distance movement factor: A transport function of the potyvirus helper component-proteinase. Plant Cell 7, 549–559. Devereux, J., Haeberli, P., and Smithies, O. (1984). A comprehensive set of sequence analysis programs for the VAX. Nucleic Acids Res. 12, 387–395. Dougherty, W. G., Cary, S. M., and Parks, T. D. (1989a). Molecular genetic analysis of a plant virus polyprotein cleavage site: A model. Virology 171, 356–364. Dougherty, W. G., and Parks, T. D. (1991). Post-translational processing of the tobacco etch virus 49-kDa small nuclear inclusion polyprotein: Identification of an internal cleavage site and delimitation of VPg and proteinase domains. Virology 183, 449–456. Dougherty, W. G., Parks, T. D., Cary, S. M., Bazan, J. F., and Fletterick, R. J. (1989b). Characterization of the catalytic residues of the tobacco etch virus 49-kDa proteinase. Virology 172, 302–310. Dougherty, W. G., and Semler, B. L. (1993). Expression of virus-encoded proteinases: Functional and structural similarities with cellular enzymes. Microbiol. Rev. 57, 781–822. GarcõB a, J. A., Riechmann, J. L., and LaõB n, S. (1989). Proteolytic activity of the plum pox potyvirus NIa-like protein in Escherichia coli. Virology 170, 362–369. Gorbalenya, A. E., Donchenko, A. P., Binov, V. M., and Koonin, E. V. (1989). Cysteine proteases of positive strand RNA viruses and chymotrypsin-like serine proteases. FEBS Lett. 243, 103–114. Hammerle, T., Molla, A., and Wimmer, E. (1992). Mutational analysis of the proposed FG loop of poliovirus proteinase 3C identifies amino acids that are necessary for 3CD cleavage and might be determinants of a function distinct from proteolytic activity. J. Virol. 66, 6028– 6034. Harris, K. S., Xiang, W., Alexander, L., Lane, W. S., Paul, A. V., and Wimmer, E. (1994). Interaction of poliovirus polypeptide 3CDpro with the 5* and 3* termini of the poliovirus genome: Identification of viral and cellular cofactors needed for efficient binding. J. Biol. Chem. 269, 27004–27014. Hellmann, G. M., Shaw, J. G., and Rhoads, R. E. (1988). In vitro analysis of tobacco vein mottling virus NIa cistron: Evidence for a virus-encoded protease. Virology 163, 554–562. Hong, Y., Levay, K., Murphy, J. F., Klein, P. G., Shaw, J. G., and Hunt, A. G. (1995). The potyvirus polymerase interacts with the viral coat protein and VPg in yeast cells. Virology 214, 159–166. Kim, D. H., Park, Y. S., Kim, S. S., Lew, J., Nam, H. G., and Choi, K. Y. (1995). Expression, purification, and identification of a novel selfcleavage site of the NIa C-terminal 27-kDa protease of turnip mosaic potyvirus C5. Virology 213, 517–525. Klein, P. G., Klein, R. R., RodrõB guez-Cerezo, E., Hunt, A. G., and Shaw, J. G. (1994). Mutational analysis of the tobacco vein mottling virus genome. Virology 204, 759–769. Lama, J., Sanz, M. A., and Rodriguez, P. L. (1995). A role for 3AB protein in poliovirus genome replication. J. Biol. Chem. 270, 14430–14438. Li, X. H., Valdez, P., Olvera, R. E., and Carrington, J. C. (1997). Functions of the tobacco etch virus RNA polymerase (NIb): Subcellular transport and protein–protein interaction with VPg/proteinase (NIa). J. Virol. 71, 1598–1607.

AID

vira

plate RNA molecule (or to the nearby, displaced complementary sequence) and recruit NIb through a protein– protein interaction (Hong et al., 1995; Li et al., 1997). Interaction of the membrane-bound precursor polyprotein with RNA may be facilitated by cooperative interactions between the NIaPro domain of the precursor and the NIaPro domain linked to the VPg at the 5* terminus of the displaced strand. Synthesis of a new RNA strand by NIb polymerase would be initiated through a priming mechanism involving the VPg domain within 6/NIa precursor. The NIaPro proteolytic activity within 6/NIa would then catalyze autoproteolysis between the 6-kDa protein and NIa (Carrington and Dougherty, 1987a; Dougherty and Parks, 1991; Restrepo-Hartwig and Carrington, 1992). ACKNOWLEDGMENTS We thank Dr. W. G. Dougherty for supplying pTrc7H-Pro and Dr. David P. Giedroc for helpful discussions and advice. This work was supported in part by the National Institutes of Health (AI27832) and by a postdoctoral fellowship to J.A.D. from the Ministry of Education and Science of Spain.

REFERENCES

VY 8802

/

6a4f$$$564

09-30-97 21:53:18

AP: VY

DARO`S AND CARRINGTON

336

Maia, I. G., and Bernardi, F. (1996). Nucleic acid binding properties of a bacterially expressed potato virus Y helper component-proteinase. J. Gen. Virol. 77, 869–877. Masson, N., Hurst, H. C., and Lee, K. A. W. (1993). Identification of proteins that interact with CREB during differentiation of F9 embryonal carcinoma cells. Nucleic Acids Res. 21, 1163–1169. Matthews, D. A., Smith, W. W., Ferre, R. A., Condon, B., Budahazi, G., Sisson, W., Villafranca, J. E., Janson, C. A., McElroy, H. E., Gribskov, C. L., and Worland, S. (1994). Structure of human rhinovirus 3C protease reveals a trypsin-like polypeptide fold, RNA-binding site, and a means for cleaving precursor polyprotein. Cell 77, 761–771. Molla, A., Harris, K. S., Paul, A. V., Shin, S. H., Mugavero, J., and Wimmer, E. (1994). Stimulation of poliovirus proteinase 3Cpro-related proteolysis by the genome-linked protein VPg and its precursor 3AB. J. Biol. Chem. 269, 27015–27020. Murphy, J. F., Klein, P. G., Hunt, A. G., and Shaw, J. G. (1996). Replacement of the tyrosine residue that links a potyviral VPg to the viral RNA is lethal. Virology 220, 535–538. Murphy, J. F., Rhoads, R. E., Hunt, A. G., and Shaw, J. G. (1990a). The VPg of tobacco etch virus RNA is the 49 kDa proteinase or the Nterminal 24 kDa part of the proteinase. Virology 178, 285–288. Murphy, J. F., Rychlik, W., Rhoads, R. E., Hunt, A. G., and Shaw, J. G. (1990b). A tyrosine residue in the small nuclear inclusion protein of tobacco vein mottling virus links the VPg to the viral RNA. J. Virol. 65, 511–513. Parks, T. D., Howard, E. D., Wolpert, T. J., Arp, D. J., and Dougherty, W. G. (1995). Expression and purification of a recombinant tobacco etch virus NIa proteinase: Biochemical analysis of the full-length and a naturally occurring truncated proteinase form. Virology 210, 194– 201. Pata, J. D., Schultz, S. C., and Kirkegaard, K. (1995). Functional oligomerization of poliovirus RNA-dependent RNA polymerase. RNA 1, 466– 477. Paul, A. V., Cao, X., Harris, K. S., Lama, J., and Wimmer, E. (1994). Studies with poliovirus polymerase 3Dpol: Stimulation of poly(U) synthesis in vitro by purified poliovirus protein 3AB. J. Biol. Chem. 269, 29173– 29181. Reichel, C., Maas, C., Schulze, S., Schell, J., and Steinbiss, H. H. (1996). Cooperative binding to nucleic acids by barley yellow mosaic bymovirus coat protein and characterization of a nucleic acid-binding domain. J. Gen. Virol. 77, 587–592. Restrepo, M. A., Freed, D. D., and Carrington, J. C. (1990). Nuclear transport of plant potyviral proteins. Plant Cell 2, 987–998. Restrepo-Hartwig, M. A., and Carrington, J. C. (1992). Regulation of nu-

clear transport of a plant potyvirus protein by autoproteolysis. J. Virol. 66, 5662–5666. Restrepo-Hartwig, M. A., and Carrington, J. C. (1994). The tobacco etch potyvirus 6-kilodalton protein is membrane-associated and involved in viral replication. J. Virol. 68, 2388–2397. Riechmann, J. L., LaõB n, S., and GarcõB a, J. A. (1989). The genome-linked protein and 5* end RNA sequence of plum pox potyvirus. J. Gen. Virol. 70, 2785–2789. Riechmann, J. L., LaõB n, S., and GarcõB a, J. A. (1992). Highlights and prospects of potyvirus molecular biology. J. Gen. Virol. 73, 1–16. Schaad, M. C., Haldeman-Cahill, R., Cronin, S., and Carrington, J. C. (1996). Analysis of the VPg-Proteinase (NIa) encoded by tobacco etch potyvirus: Effects of mutations on subcellular transport, proteolytic processing and genome amplification. J. Virol. 70, 7039–7048. Schaad, M. C., Jensen, P., and Carrington, J. C. (1997). Formation of plant RNA virus replication complexes on membranes: Role of an endoplasmic reticulum-targeted viral protein. EMBO J. 13, 4049– 4059. Schoumacher, F., Erny, C., Berna, A., Godefroy-Colburn, T., and StussiGaraud, C. (1992). Nucleic acid binding properties of the alfalfa mosaic virus movement protein produced in yeast. Virology 188, 896– 899. Shahabuddin, M., Shaw, J. G., and Rhoads, R. E. (1988). Mapping of the tobacco vein mottling virus VPg cistron. Virology 163, 635–637. Soumounou, Y., and Laliberte, J. F. (1994). Nucleic acid-binding properties of the P1 protein of turnip mosaic potyvirus produced in Escherichia coli. J. Gen. Virol. 75, 2567–2573. Thornbury, D. W., Hellmann, G. M., Rhoads, R. E., and Pirone, T. P. (1985). Purification and characterization of potyvirus helper component. Virology 144, 260–267. Verchot, J., and Carrington, J. C. (1995a). Debilitation of plant potyvirus infectivity by P1 proteinase-inactivating mutations and restoration by second-site modifications. J. Virol. 69, 1582–1590. Verchot, J., and Carrington, J. C. (1995b). Evidence that the potyvirus P1 protein functions as an accessory factor for genome amplification. J. Virol. 69, 3668–3674. Verchot, J., Koonin, E. V., and Carrington, J. C. (1991). The 35-kDa protein from the N-terminus of a potyviral polyprotein functions as a third virus-encoded proteinase. Virology 185, 527–535. Walker, P. A., Leong, L. E.-C., and Porter, A. G. (1995). Sequence and structural determinants of the interaction between the 5*-noncoding region of picornavirus RNA and rhinovirus protease 3C. J. Biol. Chem. 270, 14510–14516. Xiang, W., Harris, K. S., Alexander, L., and Wimmer, E. (1995). Interaction between the 5*-terminal cloverleaf and 3AB/3CDpro of poliovirus is essential for RNA replication. J. Virol. 69, 3658–3667.

AID

vira

VY 8802

/

6a4f$$$564

09-30-97 21:53:18

AP: VY