This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Han, S.
Right arrow Articles by Sanfaçon, H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Han, S.
Right arrow Articles by Sanfaçon, H.
Right arrowPubmed/NCBI databases
*Protein
*Substance via MeSH

 Previous Article  |  Next Article 

Journal of Virology, January 2003, p. 523-534, Vol. 77, No. 1
0022-538X/03/$08.00+0     DOI: 10.1128/JVI.77.1.523-534.2003
Copyright © 2003, American Society for Microbiology. All Rights Reserved.

Tomato Ringspot Virus Proteins Containing the Nucleoside Triphosphate Binding Domain Are Transmembrane Proteins That Associate with the Endoplasmic Reticulum and Cofractionate with Replication Complexes{dagger}

Sumin Han1 and Hélène Sanfaçon2*

Department of Botany, University of British Columbia, Vancouver, British Columbia V6T 1Z4,1 Pacific Agri-Food Research Centre, Summerland, British Columbia V0H 1Z0, Canada2

Received 8 August 2002/ Accepted 1 October 2002


arrow
ABSTRACT
 
Replication of all known positive-strand RNA viruses occurs in replication complexes associated with intracellular membranes. The putative nucleoside triphosphate binding (NTB) protein of Tomato ringspot virus (ToRSV) contains a stretch of hydrophobic residues at its C terminus, suggesting that it may act as a membrane anchor for the replication complex. Anti-NTB antibodies detected two predominant proteins in membrane-enriched fractions (the 66-kDa NTB and 69-kDa NTB-VPg proteins) along with other, larger proteins. The proteins containing the NTB domain cofractionated with markers of the endoplasmic reticulum (ER) and with ToRSV-specific RNA-dependent RNA polymerase activity in sucrose gradients. ToRSV infection induced severe changes in the morphology of the ER in plants expressing an ER-targeted green fluorescent protein (ER-GFP), and proteins containing the NTB domain colocalized with ER-GFP in indirect immunofluorescence assays. The proteins containing the NTB domain have properties of integral membrane proteins. Proteinase K protection assays using purified membranes from infected plants revealed that although the central portion of the NTB domain is exposed to the cytoplasmic face of the membranes, an 8-kDa fragment, recognized by anti-VPg antibodies, is protected by the membranes. This fragment probably consists of the 3-kDa VPg and the 5-kDa stretch of hydrophobic residues at the C terminus of the NTB protein, suggesting a luminal location for the VPg in at least a portion of the molecules. These results provide evidence that proteins containing the NTB domain are transmembrane proteins associated with ER-derived membranes and support the hypothesis that one or several of the proteins containing the NTB domain anchor the replication complex to the ER.


arrow
INTRODUCTION
 
Replication of the genomes of all characterized positive-strand RNA viruses occurs in large complexes which are associated with intracellular membranes (5). Many positive-strand RNA viruses induce extensive proliferation and modification of intracellular membranes in their hosts, which often results in the accumulation of numerous membranous vesicles. Double-stranded RNA (dsRNA) replication intermediates and viral replication factors are associated with the membranous vesicles, which are thought to be the site of the replication (6, 7, 18, 25, 48, 49, 52). The requirement for intact membranes for successful virus replication has also been demonstrated using cell-free replication systems (2, 35, 59). Different viruses associate with and modify different types of intracellular membranes (5). Although the importance of the association of viral replication factors with intracellular membranes is well documented, the mechanisms by which replication complexes are fixed on specific membrane sites remain poorly understood.

Picornaviruses and several plant viruses with similar genomic organization have been shown to replicate in association with membranes derived from the endoplasmic reticulum (ER) (6, 42, 47, 51). One or several viral proteins are thought to act as membrane anchors for the complex, while other viral replication factors are brought in the complex either as part of larger polyproteins or through protein-protein interactions with the membrane anchor. The 3AB, 2BC, and 2C proteins of picornaviruses (references 39 and 60 and references therein) and the 6-kDa protein of potyviruses (47) are integral membrane proteins and have been suggested to play a role as membrane anchors for the replication complex. In some of these proteins (e.g., the picornavirus 3AB and the potyvirus 6-kDa proteins), the association with membranes is mediated by transmembrane domains consisting of stretches of hydrophobic residues, while in others (e.g., the picornavirus 2C protein) it is mediated by amphipathic helices. The membrane anchors of picornavirus-like viruses are produced by proteolytic processing of large polyprotein precursors. The mature proteins as well as larger intermediate precursors have been detected in infected cells in association with intracellular membranes. In many cases, the intermediate precursors have activities which are distinct from those of the corresponding mature proteins. For example, the picornavirus 3AB protein was shown to play critical roles in virus replication (binding the viral RNA and binding to other viral proteins to stimulate their activity) (1, 39, 60). Therefore, the coordination of the processing of the polyproteins and of the association of the precursors and/or mature proteins with membranes is crucial for the regulation of the genome replication. Yet this process is not well understood, and in many cases it is still not clear whether the membrane anchors associate with the membranes as mature proteins, intermediate precursors, or large polyproteins (28, 53).

Tomato ringspot virus (ToRSV) (genus Nepovirus, subgroup III, family Comoviridae) is a bipartite single-stranded, positive-sense RNA virus. Each RNA is covalently linked to a small virus-encoded protein (VPg) at its 5' end and encodes one large polyprotein. Nepoviruses have a genomic organization similar to that of the animal picornaviruses (34, 44). The RNA1-encoded polyprotein (P1) contains the domains for proteins likely to be involved in replication, including a putative nucleoside triphosphate binding (NTB) protein, the VPg, the 3C-like proteinase (Pro), and the RNA-dependent RNA polymerase (Pol) (Fig. 1A) (43). The proteinase processes P1 at five cleavage sites in vitro, and the precise locations of the NTB-VPg, VPg-Pro, and Pro-Pol cleavage sites have been determined experimentally (54, 55). The NTB protein has sequence elements similar to those found in known RNA helicases (24) and has homology with the picornavirus 2C and 3A proteins. A stretch of hydrophobic residues has been identified at the C terminus of the NTB protein (43), suggesting that it is a transmembrane domain and that the NTB protein, or a larger polyprotein containing this protein, may act as a membrane anchor for the replication complex. Previous studies on the cytopathology of nepovirus-infected cells have revealed the presence of diffuse inclusions, often near the nuclei, which consist of complex membranous structures, some of which form vesicles (19). Indeed, perinuclear membranous structures produced by massive proliferation of ER have been shown to be the site of viral replication for a comovirus (6, 7) and a nepovirus (Grapevine fanleaf virus [GFLV], subgroup I) (42).



View larger version (40K):
[in this window]
[in a new window]
  		FIG. 1. Immunodetection of viral protein precursors containing the NTB domain. (A) Schematic diagram of the ToRSV RNA1-encoded polyprotein. Identified cleavage sites are indicated by vertical lines, and the putative functions of individual domains are indicated above the diagram. The hydrophobic domain at the C terminus of the NTB domain is represented as an asterisk. The regions of the NTB and VPg domains present in the fusion protein or peptide used as antigens to raise the corresponding antibodies (Abs) are shown below the diagram. (B) Immunoblot analysis of crude membrane fraction (P30) from healthy (lanes H) or ToRSV-infected (lanes I) plants. The proteins were separated by SDS-PAGE (8% polyacrylamide), detected by antibodies raised against the NTB (NTB Abs) or VPg (VPg Abs) domain, and developed by using the chemiluminescence detection system as described in Materials and Methods. Arrows point to proteins detected by the NTB domain antibodies in the infected but not in the healthy extracts, and the numbers associated with the arrows indicate the estimated molecular masses of these proteins (in kilodaltons). Migrations of molecular mass standards are shown on the right. (C) Detection of the NTB and NTB-VPg proteins by using the colorimetric detection system. Only the relevant portion of the gel is shown.

In this study, the nature of the interaction between ToRSV proteins containing the NTB domain and intracellular membranes was investigated. The proteins containing the NTB domain were found to associate with membranes having properties of the rough ER which were active in synthesis of ToRSV RNA. This suggests that one (or several) of the proteins containing the NTB domain may act as an anchor for the replication complexes. Proteins containing the NTB domain were shown to be integral membrane proteins. Surprisingly, proteinase K protection experiments suggest a luminal location for the VPg domain in at least a portion of the NTB-VPg molecules.


arrow
MATERIALS AND METHODS
 
Infection of plants and protoplasts. ToRSV was propagated routinely in Cucumis sativus var. Straight Eight. Nicotiana benthamiana transgenic plants expressing the green fluorescent protein (GFP) targeted to the lumen of the ER were generously provided by D. Baulcombe (John Innes Centre, Norwich, United Kingdom). Plants were inoculated at leaf stage 5 to 7 by using 40 µl of fresh leaf extracts of ToRSV-infected cucumber plants diluted 1:10 in phosphate-buffered saline (PBS) as described previously (57).

Mesophyll protoplasts from the ER-GFP N. benthamiana plants were prepared and transfected with 10 µg of purified ToRSV viral RNA by polyethylene glycol-mediated transfection as described previously (57).

Membrane fractionation. Leaf tissue from ToRSV-infected C. sativus was used for the isolation of a crude membrane fraction as described previously (47). Briefly, 1 g of tissue was ground in 4 ml of homogenization buffer (47). Nuclei, chloroplasts, cell wall, and debris were removed by centrifugation at 3,700 x g at 4°C for 10 min. The supernatant (S3) was centrifuged at 30,000 x g at 4°C for 30 min, resulting in soluble (S30) and crude membrane (P30) fractions. The P30 fraction was reconstituted in homogenization buffer with the aid of a Dounce homogenizer and centrifuged again at 30,000 x g for 20 min at 4°C, resulting in washed crude membrane fraction P30-2.

Membranes present in S3 or P30-2 fractions were fractionated on 20 to 45% sucrose gradients as described previously (47). The gradients were subjected to centrifugation at 143,000 x g for 4 h at 4°C. The sucrose gradient was subdivided into 13 fractions that were analyzed by immunoblotting or RNA-dependent RNA polymerase (RdRp) assays (see below). rRNAs were extracted from 50 µl of gradient fractions by using phenol-chloroform and precipitated with ethanol.

To concentrate and purify the intracellular membranes from sucrose gradient fractions, fractions 4 to 6 from a sucrose gradient obtained in the presence of 3 mM MgCl2 were pooled. The membranes were collected by centrifugation at 140,000 x g for 45 min and resuspended in buffer (50 mM Tris-HCl [pH 8.0], 10 mM KCl, 1 mM EDTA), and the presence of viral proteins in the purified membranes was examined by immunoblotting (see below). These purified membranes were used for the proteinase K experiments (see below).

Immunoblot analysis. Samples were separated by SDS-polyacrylamide gel electrophoresis (SDS-PAGE) (29), and the proteins were transferred to Sequi-blot polyvinylidene difluoride membranes (Bio-Rad) as described previously(56). The membranes were blocked with PBS containing 5% skim milk powder and incubated with the primary antibodies (diluted in PBS buffer) for 1 h at room temperature. The anti-NTB, anti-VPg, anti-movement protein (anti-MP), and anti-coat protein (anti-CP) antibodies were described previously (45, 54, 56). The anti-Bip and anti-ß-xylosyl antibodies were gifts from M. Chrispeels (University of California, San Diego) and A. Sturm (Friedrich Miescher Institute), respectively. Membranes were incubated with the secondary antibodies (goat anti-rabbit immunoglobulin G), which were conjugated to either horseradish peroxidase (Amersham, Inc.) or alkaline phosphatase (Sigma), for 1 h at room temperature and developed with a substrate for chemiluminescence (ECL Plus; Amersham, Inc.) for the peroxidase-conjugated antibodies or with a substrate for colorimetric detection (nitroblue tetrazolium-5-bromo-4-chloro-3-indolylphosphate; Invitrogen) for the alkaline phosphatase-conjugated antibodies.

RdRp activity assays. RdRp assays using endogenous templates present in membrane-enriched fractions from infected plants were conducted as described previously (47). Briefly, gradient fractions or resuspended P30-2 fractions were mixed with an equal volume of 2x RdRp buffer (47) containing [{alpha}-32P]UTP and incubated for 1 h at 24°C. The products were resuspended in 10 µl of deionized H2O, and 5 µl of each sample was digested with RNase A (1.25 µg/ml; Sigma) in the presence of 233 mM NaCl, 3.3 mM Tris-HCl (pH 7.4), and 10 mM EDTA at 30°C for 15 min. The RNase A-treated products were then incubated with proteinase K (5 mg/ml) in the presence of 2% SDS at 37°C for 30 min. For assay of inhibition of host transcription, actinomycin D (50 µg/ml) and DNase I (12.5 U/ml) were added to the reaction mixture. The products were analyzed by electrophoresis on a 1% agarose gel, followed by autoradiography. As controls, ToRSV single-stranded RNA (ssRNA) was extracted from purified virus particles (57) and ToRSV dsRNA was purified from infected leaves (16). The purified ssRNAs and dsRNAs were run on the same agarose gel along with the RdRp reactions. Dot blot hybridizations were performed using the labeled RdRp products from reactions containing P30-2 fractions from infected and noninfected plants as probes. Plus- and minus-strand transcripts specific to a 1.7-kb region of RNA2 were synthesized by digestion of pT7-X-MP (9) with the appropriate restriction enzyme followed by in vitro transcription with T3 RNA polymerase for the plus-strand transcript and T7 RNA polymerase (Invitrogen) for the minus-strand transcript. One microgram of each transcript was fixed to Zeta-Probe blotting membrane (Bio-Rad) by baking at 80°C for 1 h and hybridized with heat-denatured RdRp products as recommended by the supplier (Bio-Rad).

Immunofluorescence analysis of transfected protoplasts. Immunofluorescent staining of transfected protoplasts was performed essentially as described previously (6). Briefly, protoplasts harvested at 30 h posttransfection were allowed to settle on poly-L-lysine-coated coverslips for 2 min. One volume of fixing solution (6) was then added to the protoplast suspension. After incubation for 15 min, the coverslips were immersed in fixing solution and allowed to incubate for another 30 min. The cells were washed three times with PBS and permeabilized with a 0.5% Triton X-100 solution in PBS for 10 min. Nonspecific antibody binding was reduced by incubation for 10 min in blocking solution (6). Subsequently, the protoplasts were incubated for 1 h with the anti-NTB antibodies (diluted 1:5,000 in blocking solution). After five washes with PBS, the protoplasts were incubated with goat anti-rabbit antibodies conjugated to Cy3 (Jackson Immuno Research Laboratories), diluted 1:250 in blocking buffer, for 1 h. After five washes with PBS, the coverslips were mounted on microscope slides by using the ProLong Antifade kit (Molecular Probes, Inc.).

Fluorescence microscopy. Protoplasts were viewed by fluorescence microscopy with an Axiophot microscope (Zeiss). For detecting GFP fluorescence, a BP 450/490 exciter filter/LP 520 barrier filter set was used. For detecting Cy3 fluorescence, a BP 546/12 exciter filter/LP 590 barrier filter set was used. Photomicrographs were taken with Kodak Ektachrome 160 film. For analysis of infected plant tissues, leaf disks containing infection sites were excised, mounted as live tissue on microscope slides, and investigated immediately. A Bio-Rad Micro Radiance confocal microscope was used to obtain the images.

Membrane protein extraction analysis. Crude membrane fractions (P30) were resuspended in 0.1 M Na2CO3 (pH 10.5) or 1 M KCl. For each extraction, the samples were incubated for 30 min on ice and then subjected to centrifugation at 30,000 x g at 4°C for 30 min. The pellets (P30-2) were resuspended in protein loading buffer (29) in a volume equal to that of the corresponding supernatant. Triton X-114 phase partitioning was performed as previously described (4). The volumes of the final aqueous and detergent-soluble phases were equalized. The samples were then analyzed by immunoblotting as described above.

Structure predictions. The secondary structure of NTB-VPg was predicted by using the consensus secondary structure prediction method available at the NPS@ website (11). Prediction of the topological orientation of NTB-VPg was conducted by using the transmembrane hidden Markov model (TMHMM) algorithm (27).

Proteinase K protection assays. Samples enriched in membrane-bound NTB-VPg were obtained as described above by purifying membranes by using an ultracentrifugation of pooled fractions from a sucrose gradient obtained in the presence of 3 mM MgCl2. Aliquots of the resuspended membranes were incubated with proteinase K (0.5 mg/ml) for 6 min at room temperature in the absence or presence of 0.4% Triton X-100. The samples were then incubated with phenylmethylsulfonyl fluoride (0.2 mg/ml) for 8 min on ice and analyzed by immunoblotting as described above.


arrow
RESULTS
 
Various viral proteins containing the NTB domain are detected in crude membrane fractions. To determine if the NTB protein or larger polyproteins containing the NTB domain are present in infected plants, membrane-enriched fractions (P30 fractions) of ToRSV-infected and healthy plant cell extracts were analyzed by immunoblotting with two previously described antisera: polyclonal antiserum raised against a fusion protein containing the middle region of the NTB domain and polyclonal antiserum raised against a synthetic peptide corresponding to the central region of VPg (54) (Fig. 1A). The bands were visualized by using a chemiluminescence-based detection assay. The anti-NTB antibodies did not recognize any plant proteins in the healthy control, while the anti-VPg antibodies cross-reacted with several plant proteins in the healthy control (a predominant protein with a molecular mass of approximately 110 kDa and a protein of approximately 75 kDa). Several protein bands were detected with the anti-NTB and anti-VPg antibodies in infected but not in healthy P30 fractions (Fig. 1B). These proteins were not detected in S30 fractions corresponding to the cytosolic fraction (data not shown). A predominant band with an apparent molecular mass of approximately 69 kDa was detected with both antibodies and corresponds to the NTB-VPg polyprotein. The mature NTB protein has a calculated molecular mass of 66 kDa and would migrate very closely to the 69-kDa NTB-VPg on SDS-polyacrylamide gels. It was therefore possible that the NTB protein was also present in infected plants but that the chemiluminescent detection method did not allow the resolution of the two closely spaced protein bands. To test this, we conducted new immunoblotting experiments with the same extracts by using a colorimetric detection method. The colorimetric detection was less sensitive than the chemiluminescent detection. However the resolution of closely spaced protein bands was improved (Fig. 1C). Using this method, a protein of approximately 66 kDa was detected in addition to the 69-kDa NTB-VPg. The 66-kDa protein was recognized by the anti-NTB antibodies but not by the anti-VPg antibodies and therefore probably corresponded to the mature NTB protein (Fig. 1C).

Other viral proteins detected by the anti-NTB antibodies were present in much smaller amounts, and their relative concentrations varied from extract to extract. The apparent molecular masses of these proteins were 90, 110, and 130 kDa (Fig. 1B). The 90-kDa protein was also detected by the anti-VPg antibodies. Detection of the 110-kDa protein by the anti-VPg antibodies was obscured by cross-reaction of the antibodies with a 110-kDa protein also present in the healthy plant extract. The 90-and 110-kDa proteins may correspond to the X2-NTB-VPg and NTB-VPg-Pro polyproteins, respectively. The 130-kDa protein was only partially recognized by the anti-VPg antibodies, raising the possibility that it may consist of a mixture of two proteins, only one of which contains the VPg domain. Based on its apparent molecular mass, the 130-kDa protein may correspond to a polyprotein containing X1-X2-NTB-VPg (possibly mixed with X1-X2-NTB). Alternatively, it may be a dimer form of NTB-VPg (possibly mixed with a dimer form of NTB). The precise nature of these proteins could not be determined, as antisera against other domains of the P1 polyprotein were not available. These results show that several viral proteins containing the NTB domain are present in membrane-enriched fractions and are consistent with the suggestion that the NTB protein is a membrane-associated protein.

Viral proteins containing NTB cofractionate with ER-derived membranes in sucrose gradients. To characterize further the type of membranes to which the NTB protein attaches, cellular membranes from ToRSV-infected leaves were fractionated on sucrose gradients. Extracts from infected leaves were obtained as described in Materials and Methods and clarified by a single centrifugation at 3,000 x g to eliminate the majority of the cell wall and debris. The resulting S3 fraction was then analyzed on 20 to 45% sucrose gradients in the presence or absence of 3 mM MgCl2. The presence of 3 mM MgCl2 in the extraction buffer preserves the integrity of the association of ribosomes with the rough ER (rER), while in the absence of MgCl2, the association of ribosomes with the rER is disrupted, resulting in a shift of rER-containing fractions towards the top of the gradient (58). In this experiment, we chose to analyze S3 fractions to ensure that all intracellular membranes (including Golgi membranes) were present in the extracts. Sucrose gradient fractions were analyzed by immunoblot assay using antibodies raised against the Bip protein (a marker of the ER [50]) and against proteins containing xylose-ß,1->2-mannose modifications (a marker of the medial- and trans-Golgi [61]). Two peaks containing Bip were detected in gradients derived from infected leaves extracted in the presence of 3 mM MgCl2 (Fig. 2A, Bip), a result previously described by others (47). These two peaks were also detected in gradients derived from healthy leaves, although the concentration of Bip was lower in these gradients (data not shown). The peak containing Bip at the bottom of the gradient (fractions 1 to 6) was shifted towards the top of the gradient in the absence of MgCl2 (fractions 8 to 12), suggesting that these fractions contain rER. The proteins containing ß-xylosyl were detected near the top of the gradient (fractions 9 to 13) in the presence or absence of MgCl2, indicating that these fractions contained the Golgi apparatus (Fig. 2A, ß-xylosyl). Immunoblotting of the sucrose gradient fractions was also conducted with anti-NTB antibodies. The 66/69-kDa band was found in fractions 4 to 6 in gradients obtained from extracts prepared in the presence of 3 mM MgCl2 and was shifted up to fractions 9 to 12 in the gradient obtained from extracts prepared in the absence of MgCl2 (Fig. 2A, NTB). This shift was similar to that observed with the bottom Bip-containing peak and suggested that proteins contained in the 66/69-kDa band were associated with membranes having properties in common with the rER. The 66/69-kDa peak in the gradient obtained from extracts prepared in the presence of 3 mM MgCl2 (fractions 4 to 6, with a maximum in fraction 5) was slightly displaced compared to the bottom Bip-containing peak in the same gradient (fractions 1 to 6, with a maximum in fractions 4 and 5). One possible interpretation of this observation is that the viral proteins contained in the 66/69-kDa peak were associated with only a subpopulation of the rER-derived membranes present in infected plants. The distribution of the 66/69-kDa band was compared with those of MP (present in tubular structures in the cell wall [56]) and CP. Antibodies raised against MP detected a protein of the expected size for the mature MP in a few fractions at the bottom of either gradient (fractions 1 to 3), suggesting that these fractions were enriched in residual cell wall material that was not eliminated at the clarification step (Fig. 2A, MP). CP was detected in a broader area of the sucrose gradients (Fig. 2A, CP, fractions 1 to 7). The MP- and CP-containing peaks did not shift towards the top of the gradient in the absence of MgCl2, indicating that these proteins did not associate with rER-derived membranes and did not cofractionate with the 66/69-kDa peak.



View larger version (36K):
[in this window]
[in a new window]
  		FIG. 2. Western blot analyses of fractions from sucrose density gradients. (A) Tissue extracts were prepared and fractionated on a 20 to 45% sucrose density gradient in the presence or absence of 3 mM MgCl2. The direction of sedimentation is shown, with fraction 13 representing the top of each gradient. The proteins from each fraction were separated by SDS-PAGE (12% polyacrylamide); detected by antibodies raised against Bip, NTB, ß-xylosyl-containing proteins, MP, and CP; and developed by using the chemiluminescence-based secondary antibody system. The relevant portion of each immunoblot is shown. In the NTB immunoblot, the predominant 66/69-kDa band (corresponding to the NTB and NTB-VPg proteins) is shown. In the ß-xylosyl immunoblot, only ß-xylosyl-containing proteins with apparent molecular masses of approximately 50 kDa are shown. The concentration of MgCl2 used in each sucrose gradient is shown on the left side of the gels. (B) Fractions 4 to 6 from the 3 mM MgCl2 gradient prepared with ToRSV-infected (lanes I) plant extracts were pooled, concentrated by ultracentrifugation, and resuspended in buffer as described in Materials and Methods. Equivalent sucrose gradient fractions were also used to prepare purified membranes from healthy plants (lanes H). Proteins were separated by SDS-PAGE (15% polyacrylamide) and detected by immunoblotting with anti-NTB and anti-VPg antibodies. Migrations of molecular mass standards (in kilodaltons) are shown on the right, and arrows on the left point to the proteins detected by the anti-NTB and anti-VPg antibodies. The estimated molecular masses (in kilodaltons) are indicated by the numbers associated with the arrows.

Using immunoblot analysis of the diluted sucrose gradient fractions with the anti-NTB antibodies, we could detect the predominant 66/69-kDa band but not other, larger proteins containing the NTB domain. To further analyze the nature of the NTB domain-containing proteins associated with the intracellular membranes contained in fractions 4 to 6 (gradients prepared in the presence of 3 mM MgCl2), the membranes were purified and concentrated by an ultracentrifugation step (see Materials and Methods). Immunoblotting experiments using these purified membranes revealed the presence of the NTB-VPg polyprotein (69-kDa protein detected by the anti-NTB and anti-VPg antibodies) and of the 130- and 90-kDa proteins (also detected by both antibodies) (Fig. 2B). These proteins were not present in controls prepared in a similar manner by using sucrose gradients containing extracts from noninfected (healthy) plants. The 110-kDa plant protein that cross-reacted with the anti-VPg antibodies in the healthy and infected P30 extracts (Fig. 1) was not detected by these antibodies in the healthy or infected purified membranes (Fig. 2B), because this protein separated to different fractions in the sucrose gradient (data not shown). These results suggested that NTB-VPg and possibly other proteins containing the NTB domain were associated with membranes having properties in common with the rER.

Viral RNA synthesis activity cofractionates with proteins containing the NTB domain. We next investigated whether RdRp activity cofractionated with the NTB domain-containing proteins and with the rER. As a first step, we established a ToRSV RdRp assay using endogenous templates present in washed membrane-enriched fractions (P30-2 fractions) from healthy or infected plants. The [{alpha}-32P]UTP-labeled RNA products were separated by electrophoresis in 1% agarose under nondenaturing conditions. In infected membrane fractions, an RNA product that migrated to a position corresponding to that of ToRSV dsRNAs purified from infected plants was synthesized (Fig. 3A, lane 1). The band was rather diffuse, indicating that it may represent a mixture of RNA1 and RNA2. Due to their large and similar sizes, the dsRNA products of RNA1 (7.2 kb) and RNA2 (8 kb) purified from infected plants are not resolved into two separate bands on agarose gels (data not shown). The labeled band was not detected in the RNA products synthesized in membrane fractions derived from healthy plants (lane 2). The labeled RNA was also synthesized in the presence of DNase I (which degrades endogenous DNA) and actinomycin D (which inhibits transcription), indicating that it was not derived from DNA-dependent host transcription (lane 3). The RNA product was resistant to RNase A digestion, confirming that the labeled RNA was dsRNA produced by an RdRp and not a single-stranded poly(U) extension synthesized through the action of a terminal uridylyl transferase (lane 5). To confirm that the labeled dsRNAs were specific to ToRSV, hybridization experiments were conducted with the labeled RdRp products from infected membrane fractions as a probe and ToRSV minus- and plus-strand synthetic RNA2 transcripts bound to a filter (Fig. 3B). A strong hybridization was detected with minus-strand synthetic RNA2, indicating that the ToRSV plus strand was synthesized from the endogenous templates. In contrast, the labeled RdRp products hybridized very weakly to ToRSV plus-strand RNA2, suggesting that the ToRSV minus strand was synthesized at very low levels (if at all). As expected, the products synthesized with noninfected tissue extracts did not hybridize with either of the RNA transcripts. Similar results were obtained with synthetic RNAs derived from ToRSV RNA1 (data not shown). Other studies have also shown that only dsRNA replication products are observed in replication assays using extracts from plants infected with other picornavirus-like viruses, including another nepovirus (Grapevine chrome mosaic virus, subgroup II) (31), a comovirus (cowpea infected with Cowpea mosaic virus [CPMV]) (17), and a potyvirus (Plum pox virus) (33). Analysis of the RdRp products of other plant picornavirus-like viruses has revealed that plus-strand RNA is synthesized predominantly; i.e., the minus strand is synthesized at very low levels or not at all (17, 31, 33, 47).



View larger version (41K):
[in this window]
[in a new window]
  		FIG. 3. Analysis of membrane-bound ToRSV RdRp activity. (A) Detection of [{alpha}-32P]UTP-labeled RdRp products synthesized with P30-2 fractions from ToRSV-infected (lanes I) and healthy (lanes H) plants. The RdRp products were separated on 1% agarose gels. Total products (lanes 1 and 2), DNase I- and actinomycin D (Act.D)-resistant products (lanes 3 and 4), and RNase A-resistant products (lanes 5 and 6) are shown. The migrations of ToRSV ssRNA1 and ssRNA2 (ss1 and ss2, respectively) and dsRNAs purified from infected plants (the double-stranded forms of RNA1 and RNA2 do not separate on 1% agarose gels and migrate as one diffuse band) are shown on the left. In lane 5, the presence of 2% SDS in the proteinase K digestion buffer used after the RNase A treatment interfered slightly with the migration of the sample, resulting in a slower migration of the dsRNA products. (B) Dot blot hybridization analysis of labeled RdRp products with P30-2 fractions from ToRSV-infected (lanes I) and healthy (lanes H) plants. Transcripts corresponding to the positive (+) or negative (-) strand of a region of ToRSV RNA2 were synthesized in vitro as described in Materials and Methods and blotted onto Zeta-Probe membranes (Bio-Rad). The polarity of the ToRSV transcripts is indicated at the top. (C) Fractionation of RdRp activity, proteins containing the NTB or VPg domain, Bip, and rRNAs in a 20 to 45% sucrose gradient with P30-2 extracts from ToRSV-infected tissue. Each fraction was analyzed for RNase A-resistant RdRp activity (top); subjected to immunoblot analysis using anti-NTB, anti-VPg, and anti-Bip antibodies; and tested for the presence of rRNA. For the anti-NTB and anti-VPg antibodies, only the portion of the gel containing the predominant 66/69-kDa band is shown.

Membrane-enriched fractions (P30-2) from ToRSV-infected tissues were separated by sucrose gradient analysis in the presence of 3 mM MgCl2 as described above. In this experiment we chose to analyze P30-2 fractions that were enriched in ER membranes to allow a better separation of the membranes on the sucrose gradients and to increase the concentration of membranes present in each sucrose gradient fraction. Analysis of the RdRp activity in the different fractions revealed a peak of activity in fractions near the bottom of the gradient (Fig. 3C, fractions 1 to 4). Proteins containing the NTB domain cofractionated with the peak of ToRSV RdRp activity (only the 66/69-kDa protein band is shown in Fig. 3C). The 66/69-kDa protein band was also detected by the anti-VPg antibodies, confirming the presence of NTB-VPg in these fractions. As shown above, proteins containing the NTB domain also cofractionated with a peak containing Bip at the bottom of the gradient. Noticeably, the 66/69-kDa protein (and Bip marker) sedimented faster in this gradient than in the gradient prepared with clarified extracts (Fig. 2A, S3 fractions). Similar differential sedimentation of the Bip marker were also obtained by others when comparing gradients prepared using concentrated purified P30-2 fractions or more diluted S3 fractions (47). Extraction of rRNAs from the different fractions confirmed that the bottom peak containing Bip also contained rRNA, suggesting the presence of rER in these fractions. Taken together, the results presented in Fig. 2 and 3 indicated that the proteins containing the NTB domain cofractionated with replication complexes associated with membranes having properties of the rER.

ToRSV infection induces morphological changes of the ER. To investigate whether infection by ToRSV induces morphological changes of the ER, we used transgenic N. benthamiana plants which express GFP targeted to the lumen of the ER. GFP fluorescence in epidermal cells of leaves from ToRSV-infected plants and mock-inoculated plants was examined by confocal fluorescence microscopy. As described by others (6, 47), green fluorescence was associated with the cortical ER network and with the nuclear envelope in healthy plant cells (Fig. 4A [the green fluorescence has been converted to white in the figure]). In contrast, the morphology of the ER was drastically altered in ToRSV-infected epidermal cells (Fig. 4B and C). Large bodies of fluorescence were observed, which were often (but not always) located in close proximity to the nucleus (Fig. 4C). These morphological changes were found in clusters of cells (probably corresponding to foci of infection) in inoculated leaves at 4 days postinoculation and in systemically infected leaves at 7 days postinoculation, i.e., before obvious symptoms developed in the leaves. Once lesions became necrotic, the cells were in general too damaged to allow a clear visualization of the ER structure. However, ER aggregates were visible in cells surrounding the necrotic lesions. The ER aggregates were never observed in mock-inoculated plants. Taken together, these results suggest that the formation of ER aggregates is a consequence of ToRSV infection.



View larger version (80K):
[in this window]
[in a new window]
  		FIG. 4. Confocal fluorescence micrograph of mock-infected (A) or ToRSV-infected (B and C) epidermal cells of ER-GFP transgenic N. benthamiana. (A) Epidermal cells of a mock-inoculated leaf. (B and C) Epidermal cells of a ToRSV-inoculated leaf (4 days postinoculation), showing aggregates of ER-GFP fluorescence. (C) Close-up view of two ToRSV-infected epidermal cells, showing a perinuclear location of the ER-GFP aggregates. Bars, 20 µm. Arrows point to nuclei. The green GFP fluorescence is shown in white over the dark background.

Changes in the morphology of the ER were also observed by epifluorescence microscopy in live mesophyll protoplasts prepared from ER-GFP transgenic N. benthamiana transfected with ToRSV viral RNA. In mock-transfected protoplasts, the green fluorescence was in general very weak. However, the cortical ER network was visible (Fig. 5A, panels 1), and in some cases, a perinuclear weak green fluorescence was also observed (data not shown). In protoplasts transfected with ToRSV RNA (30 h postinoculation), large bodies of fluorescence were observed in 10 to 15% of the protoplasts (which corresponds to the percentage of protoplasts successfully infected by ToRSV as measured in a separate experiment [see below]). In some cases, the fluorescence appeared as several large aggregates in the protoplast (panels 2), and in other cases it appeared as a halo around the nucleus (panels 3). Therefore, morphological changes in the ER were also induced in protoplasts after infection with ToRSV viral RNA.



View larger version (27K):
[in this window]
[in a new window]
  		FIG. 5. Immunofluorescence localization of viral proteins containing the NTB domain in infected protoplasts. (A) Visualization of ER-GFP in mock-transfected (panels 1) and ToRSV-transfected (panels 2 and 3) live protoplasts. The left and right micrographs in each horizontal row show bright-field and ER-GFP fluorescence, respectively. (B) Immunofluorescence localization of proteins containing the NTB domain in ToRSV-transfected protoplasts. The protoplasts were fixed and immunostained with anti-NTB and Cy3-conjugated secondary antibodies. The micrographs in each horizontal row show the Cy3 fluorescence corresponding to the location of the NTB protein (NTB), the ER-GFP fluorescence (ER-GFP), and the digitally superimposed images, where green and red signals that coincide produce a yellow signal (Merge). In panels 1, two protoplasts are present. In the protoplast on the left, a successful infection by ToRSV was established, and NTB specific fluorescence is detected. The protoplast in the right is apparently not infected, and NTB fluorescence is not detected. Bars, 10 µm.

Proteins containing the NTB domain colocalize with ER membranes in infected protoplasts. To determine whether the proteins containing the NTB domain colocalize with ER membrane aggregates in infected plant cells, protoplasts prepared from ER-GFP N. benthamiana were transfected with viral RNA, fixed at 30 h postinoculation, and immunostained with anti-NTB antibodies. The use of goat anti-rabbit-Cy3 as secondary antibodies allowed us to simultaneously visualize the red fluorescence (Cy3) associated with the anti-NTB antibodies and the green fluorescence from the ER-GFP. Although some red autofluorescence from the chloroplasts remained after the fixation procedure, the Cy3 red fluorescence was much more intense and easily distinguishable from the chloroplast autofluorescence (compare the noninfected protoplast on the right of Fig. 5B, panels 1, NTB, to the ToRSV-infected protoplast on the left). By using epifluorescence microscopy, the percentage of protoplasts successfully infected by ToRSV was estimated by analyzing the number of protoplasts displaying the Cy3 red fluorescence and therefore expressing proteins containing the NTB domain. In approximately 15% of the protoplasts, a successful ToRSV infection was established. In ToRSV-infected protoplasts, the NTB-associated red fluorescence was localized in one or several large fluorescent bodies per cell (Fig. 5B, panels 1 [protoplast on the left] and 2). In approximately 90% of the ToRSV-infected protoplasts, the aggregates of red fluorescence associated with the anti-NTB antibodies colocalized with aggregates of green fluorescence associated with the ER-GFP as shown in the digitally superimposed image (Fig. 5B, Merge). These results are in agreement with the sucrose gradient fractionation studies presented above and confirm the suggestion that proteins containing the NTB domain associate with membranes derived from the ER.

Proteins containing the NTB domain are integral membrane proteins. To further characterize the nature of the association of NTB domain-containing proteins with cellular membranes, membrane-enriched fractions (P30) were extracted with different agents known to dislodge peripheral, luminal, or integral membrane proteins. Treatment with 1 M KCl solubilizes peripheral proteins, while integral and luminal proteins remain associated with the membranes (46). Under high-pH conditions (0.1 M Na2CO3, pH 10.5), membrane vesicles are converted to open membrane sheets, allowing the release of peripheral and luminal proteins but not of integral membrane proteins (20, 26). Finally, membrane-enriched fractions were subjected to a Triton X-114 phase partition analysis. Integral membrane proteins partition to the detergent phase, while other proteins are found in the aqueous phase (4).

After treatment with the various agents, the membranes were collected by centrifugation and analyzed by immunoblotting. The immunoblots were probed sequentially with antibodies against the NTB protein, VPg, and Bip (a soluble ER luminal protein) (Fig. 6). In these experiments, the proteins containing the NTB domain, including NTB-VPg (revealed by both the anti-NTB and anti-VPg antibodies), remained associated with the membrane pellet following extraction with 0.1 M Na2CO3 (pH 10.5) and 1 M KCl. Furthermore, they partitioned into the detergent fraction after extraction with Triton X-114 (Fig. 6A and B). In contrast and as expected, the luminal Bip protein was released into the supernatant fraction after treatment with 0.1 M Na2CO3 (pH 10.5), remained in the pellet after treatment with 1 M KCl, and partitioned into the aqueous phase after extraction with Triton X-114. Taken together, these results indicate that the proteins containing the NTB domain are associated with cellular membranes as true integral membrane proteins.



View larger version (40K):
[in this window]
[in a new window]
  		FIG. 6. Extraction and immunoblot analysis of proteins containing the NTB domain from infected plants. The P30 fraction from infected plants was extracted with 1 M KCl or 0.1 M Na2CO3 (pH 10.5) and subjected to centrifugation at 30,000 x g, yielding S30 (lanes S) and P30 (lanes P) fractions. The total P30 fraction was also separated into an aqueous phase (lane AP) and a detergent-soluble phase (lane DP) after treatment with Triton X-114. The supernatant and pellet fraction or aqueous phase and detergent phase were loaded on SDS-12% polyacrylamide gels in equivalent amounts and analyzed by immunoblotting with anti-NTB (A), anti-VPg (B), and anti-Bip (C) antibodies and the chemiluminescence-based secondary antibody system. Only the portion of the gel containing the predominant 66/69-kDa band is shown.

The VPg domain of proteins containing the NTB and VPg domains is protected from proteinase K digestion by its association with cellular membranes. We have shown above that several NTB domain-containing proteins are strongly associated with ER-derived membranes. The predominant proteins in membrane-enriched fractions were the mature NTB protein and the NTB-VPg polyprotein (Fig. 1). The presence of the NTB-VPg polyprotein in association with replication complexes was confirmed by its detection with anti-NTB and anti-VPg antibodies (Fig. 3). Because of the suggested role of the VPg as a primer in viral RNA replication in the related picornaviruses (37), it was of interest to study the topology of NTB-VPg in the membranes. We initiated this topology study by analyzing the deduced amino acid sequence of the NTB-VPg polyprotein. The hydrophobicity profile of NTB-VPg revealed that the NTB domain contains two potential hydrophobic domains, one at each of the N and C termini of the protein (Fig. 7A and B). We have used a consensus of several prediction methods to analyze the possible secondary structure of NTB-VPg in the region of the short N-terminal hydrophobic domain (see Materials and Methods). An {alpha}-helix was predicted for the segment of amino acids (aa) 37 to 67 (numbered from the first amino acid of the NTB domain). An {alpha}-helix projection of this segment revealed the presence of a putative amphipathic helix (data not shown). The C-terminal hydrophobic domain was previously suggested to be a transmembrane domain (43). This domain consists of two hydrophobic stretches of 19 and 20 aa separated by a single charged Lys residue (Fig. 7C). Lys residues were also present at the C and N termini of the hydrophobic domain. As stretches of 20 to 25 nonpolar amino acids have been shown to be of sufficient length to span the hydrophobic lipid bilayer (32), the sequence of the ToRSV putative transmembrane domain suggested that it either could adopt a hairpin structure and span the membrane twice or could span the membrane only once. In an attempt to predict the possible topology of NTB-VPg, the deduced amino acid sequence was analyzed with the TMHMM algorithm (27). As expected, a transmembrane domain was predicted at the C terminus of the protein. The region of the NTB protein upstream of the putative transmembrane domain was predicted to be localized on the cytoplasmic face of the membranes (100% prediction) (data not shown). Interestingly, the C-terminal segment of NTB-VPg containing the VPg domain was predicted to be on the luminal face of the membrane (60% prediction).



View larger version (28K):
[in this window]
[in a new window]
  		FIG. 7. Predicted structure of the NTB-VPg polyprotein and proteinase K treatment of membrane-enriched fractions from sucrose gradients. (A) Kyte-Doolittle hydropathy blot for NTB-VPg. The position of the start of the VPg domain is indicated by an arrow above the graph. (B) Graphical summary of the prediction for the 620 aa of NTB-VPg. The light grey square, the asterisk, and the black square represent the putative amphipathic {alpha}-helix, the putative transmembrane domain, and the VPg, respectively. Abs, antibodies. (C) Amino acid sequence of the putative transmembrane domain. Positively charged amino acids are shown in boldface, and two hydrophobic stretches are underlined. The NTB-VPg cleavage site downstream of the transmembrane domain is shown by an arrow. (D) Proteinase K protection assays using membrane-enriched fractions from ToRSV-infected plants. Purified membranes from Fig. 2B were subjected to proteinase K (Prot. K) digestion in the absence (-) or presence (+) of Triton X-100. The proteins were separated by SDS-PAGE (18% polyacrylamide); analyzed by immunoblotting with anti-Bip (lanes 1 to 3), anti-NTB (lanes 4 to 6), and anti-VPg (lanes 7 to 12) antibodies; and developed by using the chemiluminescence-based secondary antibody system. The arrowhead points to the 8-kDa protected fragment detected by the anti-VPg antibodies. Migrations of molecular mass standards are shown on the left.

To study the topology of proteins containing the NTB and VPg domains, samples enriched in membrane-bound NTB-VPg were treated with proteinase K. Using this technique, only portions of the proteins which are exposed to the cytoplasmic face should be susceptible to the proteinase. In contrast, parts of the proteins embedded in the membrane are protected from digestion unless the membranes are solubilized by a detergent. We have shown above that intracellular membranes purified from sucrose gradient fractions contain the NTB-VPg polyprotein along with smaller amounts of other, larger proteins recognized by the anti-NTB and anti-VPg antibodies (Fig. 2B). These samples, which did not contain substantial amounts of plant proteins cross-reacting with the anti-NTB or anti-VPg antibodies, were used to conduct the proteinase K protection assays. To test the efficiency of the proteinase K protection assay, we used the Bip ER luminal protein as a control (Fig. 7D, lanes 1 to 3). Immunoblotting of samples collected prior to the proteinase K digestion allowed the recognition of the full-length 70-kDa protein by the anti-Bip antibodies. Small amounts of a 31-kDa fragment that may represent a degradation product of the protein were also detected. As expected, the 70-kDa protein (and the 31-kDa protein) were protected from degradation in the absence of the detergent but were degraded in the presence of Triton X-100. Small amounts of two 30-kDa fragments were resistant to proteinase K digestion in the presence or absence of Triton X-100. Therefore, this resistance to proteinase digestion was not due to membrane protection. Very similar results were described for studies using Kar2p, which is a yeast ER luminal protein homologous to Bip (14). The immunoblots were stripped of the anti-Bip antibodies and reprobed sequentially with the anti-NTB antibodies raised against the central region of the NTB protein and with the anti-VPg antibodies. In samples collected prior to the proteinase K digestion, only the predominant 66/69-kDa protein is shown (Fig. 7D, lanes 4 and 7). The anti-NTB antibodies failed to detect any protected fragments in the presence or absence of detergent, suggesting that the region of the NTB protein which is recognized by the antibodies is exposed to the cytoplasmic face of the membrane (lanes 4 to 6). Interestingly, an 8-kDa segment was detected by the anti-VPg antibodies when the proteinase K digestion was performed in the absence of detergent (lane 8). This fragment was completely digested when Triton X-100 was added to the reaction mixture (lane 9). A proteinase K protection experiment was also performed on the purified membrane samples from noninfected plant extracts. The anti-VPg antibodies did not detect any protected fragment in the presence or absence of Triton X-100 (lanes 11 and 12). The size of the 8-kDa protected fragment corresponds to the predicted size for a fragment containing the putative transmembrane domain at the C terminus of the NTB protein (with a calculated molecular mass close to 5 kDa) followed by the entire VPg domain (3 kDa) (Fig. 7B). This result confirms the prediction that the central portion of the NTB domain is exposed to the cytoplasmic face of the membranes while the VPg domain adopts a luminal orientation in at least a portion of the molecules.


arrow
DISCUSSION
 
In this study, we have shown that several proteins containing the NTB domain are present in membrane-enriched fractions isolated from infected plant leaves. Among these proteins, the mature NTB protein and the NTB-VPg polyprotein were abundant in all of the extracts tested. The accumulation of NTB-VPg suggests that it is a stable intermediate in the processing pathway. We have previously shown that the cleavage sites on the RNA1-encoded polyprotein (including the NTB-VPg cleavage site) are not processed in trans by the proteinase in vitro (9, 55). Therefore, the NTB and NTB-VPg proteins detected in infected plant extracts are likely to be produced by alternative cis-processing pathways of the P1 polyprotein. Such alternative processing strategies have also been suggested for the closely related CPMV (23). In vitro, the ToRSV NTB-VPg cleavage site is a predominant cis-cleavage site, and accumulation of NTB-VPg is not detected. Rather, the mature NTB and the VPg-Pro intermediate were shown to accumulate (54, 55). The accumulation of NTB-VPg in infected plants suggests that cellular factors (such as the membrane environment) modulate the recognition of the NTB-VPg cleavage site by the ToRSV proteinase.

The accumulation of ToRSV NTB-VPg in infected plants also raises the possibility that it may play an important role in the biology of the virus which may be distinct from that of the mature NTB protein. Because NTB-VPg cofractionates with the viral replication complex, an active role of this protein in the replication of the virus seems probable. The accumulation of processing intermediates containing a putative membrane anchor (such as the hydrophobic domain at the C terminus of the NTB protein) and the VPg has been observed for several picornavirus-like viruses. The picornavirus 3AB, the Tobacco etch virus (TEV) (potyvirus) 6K-NIa, and the CPMV NTB-VPg (60K) proteins and now the ToRSV NTB-VPg polyproteins are examples of such processing intermediates which accumulate in infected tissues (1, 23, 41, 60; this study). The poliovirus (PV) 3AB protein has been shown to play critical roles in the virus replication cycle, and the presence of the VPg moiety (3B) on 3AB is essential for its activity (1, 39, 60). Improving proteolytic cleavage at the 3A/3B site of the Hepatitis A virus polyprotein results in impaired processing and particle formation (28), suggesting that accumulation of the 3AB protein is essential for the virus replication cycle. Future work will be necessary to dissect the possible biological function(s) of the ToRSV NTB-VPg polyprotein.

Using transgenic plants expressing GFP targeted to the ER, we showed that ToRSV infection induces drastic modifications of the ER. Large GFP-containing structures that often had a perinuclear location were found in infected tissues, and proteins containing the NTB domain colocalize with these structures. Sucrose gradient fractionation analysis confirmed that proteins containing the NTB domain cofractionate with membranes derived from the rough ER that are active in viral replication. Nepovirus-infected cells are characterized by the presence of diffuse inclusions (containing membrane vesicles), often near the nuclei (19). Recently, the replication activity of another nepovirus (GFLV) has been shown to be associated with ER-derived membrane structures very similar to the ones described in this study, and they were also located at a perinuclear location (42). Replication-associated proteins were also shown to colocalize at this site (21, 42). Infection of ER-GFP plant cells by other picornavirus-like viruses (including CPMV and TEV) is also characterized by the induction of GFP aggregates similar to the ones described here, and the replication proteins of these viruses colocalize with the modified ER membranes (6, 7, 47). Several viral proteins expressed outside the context of the virus genome have been shown to localize to intracellular membranes and to induce the alteration of the morphology of these membranes. For example, the morphology of ER and/or Golgi membrane is altered after expression of the PV 2C and 2BC (10) and 3A and 3AB (12, 15) proteins, the TEV 6-kDa protein (41), and the CPMV 60K and 32K proteins (8). On the basis of sequence similarities with the PV 3A and 2C and the CPMV 60-kDa proteins, it is likely that NTB and/or proteins containing the NTB domain are also responsible (at least in part) for the alterations of the ER morphology in ToRSV-infected cells. The association of the replication complex with membranes derived from the rough ER would provide opportunities for the coupling between the translation and the replication of the virus genome. Coupling between the viral translation, the formation of membranous vesicles, and viral RNA synthesis is required for the formation of the PV replication complex (18, 36), although the mechanisms by which the virus regulates the switch from translation to replication are not yet clearly understood (1, 3, 22).

We have shown that proteins containing the NTB domain associate with membranes as integral membrane proteins, suggesting that they may act as anchors for the ToRSV replication complex. Active replication complexes are presumed to assemble on the cytoplasmic face of the membranes (reference 49 and references therein). For Brome mosaic virus and possibly other viruses, the replication complexes are subsequently sequestered in cytoplasmic invaginations or spherules connected to the cytoplasm by a neck (49). In agreement with this model, our analysis of the topology of the proteins containing the NTB domain revealed that at least the central part of that domain is exposed to the cytoplasmic face of the membranes. The results of the proteinase K protection assay also revealed that an 8-kDa fragment containing the VPg domain is embedded in the membranes. Because NTB-VPg is one of the predominant proteins in the purified membranes used in the protection experiments (Fig. 2B), it is likely that the protected fragment is derived mainly from NTB-VPg, although we cannot exclude the possibility that digestion of other, larger NTB- and VPg-containing polyproteins by the proteinase K may also result in the protection of this fragment. As mentioned above, the size of the protected fragment corresponds to the predicted size for a fragment containing the entire hydrophobic domain at the C terminus of the NTB protein followed by the VPg. This would suggest that the transmembrane domain spans the membrane once, resulting in the luminal location for the VPg. As the VPg is likely to play an active role in the replication of the viral RNA (possibly as a primer in RNA synthesis [37]), a luminal location for the VPg may seem surprising. Recently, Ritzenthaler and colleagues demonstrated that small membranous vesicles contained in ER-enriched sucrose gradient fractions purified from GFLV-infected tissues could be immunotrapped by anti-VPg antibodies, suggesting that at least some of the VPg population is present on the cytoplasmic face of the membranes in GFLV-infected cells (42). Several points should be made that may help reconcile this apparently conflicting observation. First, our results do not exclude the possibility that NTB-VPg (or other proteins containing the NTB and VPg domains) displays a dual topology in the membranes. In recent years, it has become evident that certain cellular and viral transmembrane proteins exhibit two or more distinct topological orientations which influence their biological functions (reference 30 and references therein). The TMHMM analysis of NTB-VPg gave a 100% prediction for a cytosolic orientation of the region of the NTB protein upstream of the transmembrane domain but only a 60% prediction for a luminal orientation of the VPg, suggesting that alternative topological orientations are possible. According to this prediction, the alternative topology would be possible only if the transmembrane domain spans the membrane twice, resulting in the VPg being exposed to the cytosol. As mentioned above, the presence of two adjacent hydrophobic stretches of 19 to 20 aa suggests that such a conformation is theoretically possible. Unfortunately, the proteinase K protection assay used in this study would not allow us to detect proteins with this alternative topology. Indeed, if the VPg was exposed to the cytoplasm, a 5-kDa fragment corresponding to the NTB transmembrane domain would be protected from proteinase K digestion but would not be detected in our assay, as the anti-NTB antibodies were not raised against this part of the protein. Additional experiments will be needed to investigate the possibility that other topological orientations of NTB-VPg (or other larger polyproteins) exist in infected plants. The second point is that NTB-VPg is probably not the donor for the VPg in infected plants. As discussed above, the NTB and NTB-VPg proteins are probably produced by alternative cis-processing events and therefore probably associate with the membranes after their release from the P1 polyprotein. As a consequence, the Pro and VPg-Pro (and/or Pro-Pol and VPg-Pro-Pol) are likely also produced in infected plants. The VPg-Pro or VPg-Pro-Pol are therefore possible donors for the active VPg, as is suggested for CPMV (38). We are currently raising polyclonal antibodies against the Pro and Pol domains in an attempt to detect such intermediates in infected plant cells. It should be noted that the VPg-Pro-Pol intermediate was detected in plants infected by Tomato black ring virus (genus Nepovirus, subgroup II) (13) and that several large polyproteins containing VPg were also detected in ER-enriched fractions from GFLV-infected plant extracts (42). Finally, while it is possible that the luminal orientation of the VPg in at least some of the membrane-associated proteins may play an important (as yet unknown) biological role, it may also represent a means to dispose of excess VPg and regulate the replication of the virus. Similarly, in TEV, the export of NIa (containing the VPg domain) in the nucleus has been suggested to help dispose of excess NIa in the cytoplasm, and this process is tightly regulated by autoproteolysis (40).

Our results provide support for a role of the transmembrane domain at the C terminus of the NTB protein in the association of proteins containing the NTB domain with the membranes. In agreement with this, analysis of the hydrophobicity profile of NTB-VPg from a comovirus (CPMV) and three other nepoviruses (GFLV, Grapevine chrome mosaic virus, and Tomato black ring virus) also revealed the presence of large hydrophobic domains (40 to 50 residues) immediately upstream of the hydrophilic VPg domain (data not shown). The possible role of the putative amphipathic helix at the N terminus of ToRSV NTB in the membrane association remains to be determined. A similar amphipathic helix was found at the N terminus of the CPMV NTB domain (data not shown) and in the N-terminal region of the NTB domains of other nepoviruses (in these cases, however, the analysis was complicated by the fact that the proteinase cleavage sites at the N terminus of the NTB domain have not been experimentally determined). The ToRSV-encoded NTB-VPg shares similarities with picornavirus 2C and 3AB proteins (Fig. 8A). An amphipathic helix at the N terminus of the 2C protein was shown to be essential for association of the protein with membranes (1, 60) and may be the functional equivalent of the putative N-terminal amphipathic helix in the ToRSV NTB domain. It was proposed that membrane association of the picornavirus replication complex is mediated by the 2C and 3AB proteins (1, 60), possibly in the form of larger 2C- and 3AB-containing polyproteins. Our results suggest that the NTB protein (as a mature protein or as a larger polyprotein) acts as an anchor for the replication complex. By analogy with the picornavirus proteins, we suggest a model in which NTB-VPg (and NTB) is attached to the membranes by both termini (Fig. 8B), with its central portion exposed to the cytoplasm and accessible for protein-protein interactions with other viral (or plant) proteins. Further investigation of protein-protein interactions involved in ToRSV RNA synthesis should be helpful in defining the mechanisms of replication complex assembly.



View larger version (26K):
[in this window]
[in a new window]
  		FIG. 8. Model for the topology of NTB-VPg. (A) Schematic diagrams of picornavirus 2C-3A-3B and ToRSV NTB-VPg. The light grey square, the asterisk, and the black square represent the putative amphipathic {alpha}-helix, the transmembrane domain, and VPg, respectively. (B) Model for the topology of ToRSV NTB-VPg. The N-terminal putative amphipathic {alpha}-helix is partly buried within membranes. The C-terminal transmembrane domain spans the membrane once, and VPg is buried within the ER lumen. The central part of NTB is exposed to the cytoplasmic face of the membranes. NTP, nucleoside triphosphate.


arrow
ACKNOWLEDGMENTS
 
We thank M. Weis and E. Humphrey (Department of Botany, University of British Columbia) for help with confocal microscopy, M. Chrispeels (University of California, San Diego) for the anti-Bip antibodies, A. Sturm (Friedrich Miescher Institute) for the anti-ß-xylosyl antibodies, D. Baulcombe (John Innes Centre) for the ER-GFP transgenic plants, A. Wang for performing preliminary immunoblotting experiments with the anti-NTB and anti-VPg antibodies and for valuable discussions in the early stages of this project, Y. Xiang for preliminary extraction of membrane-associated proteins with various biochemical treatments, and J. Chisholm for expert technical assistance and valuable discussions. D. A. Theilmann, J. Wellink, and J. Chisholm are gratefully acknowledged for critical reading of the manuscript.

This study was supported in part by a grant from Natural Sciences and Engineering Research Council of Canada to H.S.


arrow
FOOTNOTES
 
* Corresponding author. Mailing address: Pacific Agri-Food Research Centre, 4200 Highway 97, Summerland, British Columbia V0H 1Z0, Canada. Phone: (250) 494-6393. Fax: (250) 494-0755. E-mail: SanfaconH{at}agr.gc.ca. Back

{dagger} Contribution no. 2175 of the Pacific Agri-Food Research Centre. Back


arrow
REFERENCES
 
    1
  1. Agol, V. I., A. V. Paul, and E. Wimmer. 1999. Paradoxes of the replication of picornaviral genomes. Virus Res. 62:129-247.[CrossRef][Medline]
    2. 2
  2. Barton, D. J., E. P. Black, and J. B. Flanegan. 1995. Complete replication of poliovirus in vitro: preinitiation RNA replication complexes require soluble cellular factors for the synthesis of VPg-linked RNA. J. Virol. 69:5516-5527.[Abstract]
    3. 3
  3. Barton, D. J., B. J. Morasco, and J. B. Flanegan. 1999. Translating ribosomes inhibit poliovirus negative-strand RNA synthesis. J. Virol. 73:10104-10112.[Abstract/Free Full Text]
    4. 4
  4. Bordier, C. 1981. Phase separation of integral membrane proteins in Triton X-114 solution. J. Biol. Chem. 256:1604-1607.[Abstract/Free Full Text]
    5. 5
  5. Buck, K. W. 1996. Comparison of the replication of positive-stranded RNA viruses of plants and animals. Adv. Virus Res. 47:159-251.[Medline]
    6. 6
  6. Carette, J. E., M. Stuiver, J. van Lent, J. Wellink, and A. van Kammen. 2000. Cowpea mosaic virus infection induces a massive proliferation of endoplasmic reticulum but not Golgi membranes and is dependent on de novo membrane synthesis. J. Virol. 74:6556-6563.[Abstract/Free Full Text]
    7. 7
  7. Carette, J. E., K. Guhl, J. Wellink, and A. van Kammen. 2002. Coalescence of the sites of cowpea mosaic virus RNA replication into a cytopathic structure. J. Virol. 76:6235-6243.[Abstract/Free Full Text]
    8. 8
  8. Carette, J. E., J. van Lent, S. A. MacFarlane, J. Wellink, and A. van Kammen. 2002. Cowpea mosaic virus 32- and 60-kilodalton replication proteins target and change the morphology of endoplasmic reticulum membranes. J. Virol. 76:6293-6301.[Abstract/Free Full Text]
    9. 9
  9. Carrier, K., F. Hans, and H. Sanfaçon. 1999. Mutagenesis of amino acids at two tomato ringspot nepovirus cleavage sites: effect on proteolytic processing in cis and in trans by the 3C-like protease. Virology 258:161-175.[CrossRef][Medline]
    10. 10
  10. Cho, M. W., N. Teterina, D. Egger, K. Bienz, and E. Ehrenfeld. 1994. Membrane rearrangement and vesicle induction by recombinant poliovirus 2C and 2BC in human cells. Virology 202:129-145.[CrossRef][Medline]
    11. 11
  11. Combet, C., C. Blanchet, C. Geourjon, and G. Deleage. 2000. NPS@: network protein sequence analysis. Trends Biochem. Sci. 25:147-150.[CrossRef][Medline]
    12. 12
  12. Datta, U., and A. Dasgupta. 1994. Expression and subcellular localization of poliovirus VPg-precursor protein 3AB in eukaryotic cells: evidence for glycosylation in vitro. J. Virol. 68:4468-4477.[Abstract/Free Full Text]
    13. 13
  13. Demangeat, G., O. Hemmer, J. Reinbolt, M. A. Mayo, and C. Fritsch. 1992. Virus-specific proteins in cells infected with tomato black ring nepovirus: evidence for proteolytic processing in vivo. J. Gen. Virol. 73:1609-1614.[Abstract/Free Full Text]
    14. 14
  14. Den Boon, J. A., J. Chen, and P. Ahlquist. 2001. Identification of sequences in Brome mosaic virus replicase protein 1a that mediate association with endoplasmic reticulum membranes. J. Virol. 75:12370-12381.[Abstract/Free Full Text]
    15. 15
  15. Doedens, J. R., and K. Kirkegaard. 1995. Inhibition of cellular protein secretion by poliovirus proteins 2B and 3A. EMBO J. 14:894-907.[Medline]
    16. 16
  16. Eastwell, K. C., and M. G. Bernardy. 1996. Association of high molecular weight double-strand RNA with little cherry disease. Can. J. Plant Pathol. 18:203-208.
    17. 17
  17. Eggen, R., A. Kaan, R. Goldbach, and A. van Kammen. 1988. Cowpea mosaic virus RNA replication in crude membrane fractions from infected cowpea and Chenopodium amaranticolor. J. Gen. Virol. 69:2711-2720.[CrossRef]
    18. 18
  18. Egger, D., N. Teterina, E. Ehrenfeld, and K. Bienz. 2000. Formation of the poliovirus replication complex requires coupled viral translation, vesicle production and viral RNA synthesis. J. Virol. 74:6570-6580.[Abstract/Free Full Text]
    19. 19
  19. Francki, R. I. B., R. G. Milne, and T. Hatta. 1985. Nepovirus group, p. 23-38. In Atlas of plant viruses, vol. II. CRC Press, Boca Raton, Fla.
    20. 20
  20. Fujiki, Y. A., S. Hubbard, S. Fowler, and P. B. Lazarow. 1982. Isolation of intracellular membranes by means of sodium carbonate treatment: application to endoplasmic reticulum. J. Cell Biol. 93:97-102.[Abstract/Free Full Text]
    21. 21
  21. Gaire, F., C. Schmitt, C. Stussi-Garaud, L. Pinck, and C. Ritzenthaler. 1999. Protein 2A of grapevine fanleaf nepovirus is implicated in RNA2 replication and colocalizes to the replication site. Virology 264:25-36.[CrossRef][Medline]
    22. 22
  22. Gamarnik, A. V., and R. Andino. 1998. Switch from translation to RNA replication in a positive-stranded RNA virus. Genes Dev. 12:2293-2304.[Abstract/Free Full Text]
    23. 23
  23. Goldbach, R., and J. Wellink. 1996. Comoviruses: molecular biology and replication, p. 35-76. In B. D. Harrison and A. F. Murant (ed.), The plant viruses, vol. 5. Polyhedral virions and bipartite RNA genomes. Plenum Press, New York, N.Y.
    24. 24
  24. Gorbalenya, A. E., and E. V. Koonin. 1989. Viral proteins containing the purine NTP-binding sequence pattern. Nucleic Acids Res. 17:8413-8440.[Abstract/Free Full Text]
    25. 25
  25. Gosert, R., A. Kanjanahaluethai, D. Egger, K. Bienz, and S. C. Baker. 2002. RNA replication of mouse hepatitis virus takes place at double-membrane vesicles. J. Virol. 76:3697-3708.[Abstract/Free Full Text]
    26. 26
  26. Howell, K. E., and G. E. Palade. 1982. Hepatic Golgi fractions resolved into membrane and content subfractions. J. Cell Biol. 92:822-832.[Abstract/Free Full Text]
    27. 27
  27. Krogh, A., B. Larsson, G. von Heijne, and E. L. L. Sonnhammer. 2001. Predicting transmembrane protein topology with a hidden Markov model: application to complete genomes. J. Mol. Biol. 305:567-580.[CrossRef][Medline]
    28. 28
  28. Kusov, Y., and V. Gaus-Muller. 1999. Improving proteolytic cleavage at the 3A/3B site of the hepatitis A virus polyprotein impairs processing and particle formation, and the impairment can be complemented in trans by 3AB and 3ABC. J. Virol. 73:9867-9878.[Abstract/Free Full Text]
    29. 29
  29. Laemmli, U. K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680-685.[CrossRef][Medline]
    30. 30
  30. Lambert, C., and R. Prange. 2001. Dual topology of the hepatitis B virus large envelope protein: determinants influencing post-translational pre-S translocation. J. Biol. Chem. 276:22265-22672.[Abstract/Free Full Text]
    31. 31
  31. Le Gall, O., T. Candresse, and J. Dunez. 1997. An RNA-dependent-RNA-polymerase activity associated with grapevine chrome mosaic nepovirus infection. Arch. Virol. 142:151-156.[CrossRef][Medline]
    32. 32
  32. Lemmon, M. A., K. R. MacKenzie, I. T. Arkin, and D. M. Engelman. 1997. Transmembrane {alpha}-helix interactions in folding and oligomerization of integral membrane proteins, p. 3-23. In G. von Heijne (ed.), Membrane protein assembly. Springer, Heidelberg, Germany.
    33. 33
  33. Martin, M. T., and J. A. Garcia. 1991. Plum pox potyvirus RNA replication in a crude membrane fraction from infected Nicotiana clevelandii leaves. J. Gen. Virol. 72:785-790.[Abstract/Free Full Text]
    34. 34
  34. Mayo, M. A., and D. J. Robinson. 1996. Nepoviruses: molecular biology and replication, p. 139-185. In B. D. Harrison and A. F. Murant (ed.), The plant viruses, vol. 5. Polyhedral virions and bipartite RNA genomes. Plenum Press, New York, N.Y.
    35. 35
  35. Molla, A., A. V. Paul, and E. Wimmer. 1993. Effects of temperature and lipophilic agents on poliovirus formation and RNA synthesis in a cell-free system. J. Virol. 67:5932-5938.[Abstract/Free Full Text]
    36. 36
  36. Novak, J. E., and K. Kirkegaard. 1994. Coupling between genome translation and replication in an RNA virus. Genes Dev. 8:1726-1737.[Abstract/Free Full Text]
    37. 37
  37. Paul, A. V., J. H. van Boom, D. Filippov, and E. Wimmer. 1998. Protein-primed RNA synthesis by purified poliovirus RNA polymerase. Nature 393:280-284.[CrossRef][Medline]
    38. 38
  38. Peters, S. A., J. M. Mesnard, I. M. Kooter, J. Verver, J. Wellink, and A. van Kammen. 1995. The cowpea mosaic virus RNA 1-encoded 112 kDa protein may function as a VPg precursor in vivo. J. Gen. Virol. 76:1807-1813.[Abstract/Free Full Text]
    39. 39
  39. Porter, A. G. 1993. Picornavirus nonstructural proteins: emerging roles in virus replication and inhibition of host cell functions. J. Virol. 67:6917-6921.[Free Full Text]
    40. 40
  40. Restrepo-Hartwig, M. A., and J. C. Carrington. 1992. Regulation of nuclear transport of a plant potyvirus protein by autoproteolysis. J. Virol. 66:5662-5666.[Abstract/Free Full Text]
    41. 41
  41. Restrepo-Hartwig, M. A., and J. C. Carrington. 1994. The tobacco etch potyvirus 6-kilodalton protein is membrane-associated and involved in viral replication. J. Virol. 68:2388-2397.[Abstract/Free Full Text]
    42. 42
  42. Ritzenthaler, C., C. Laporte, F. Gaire, P. Dunoyer, C. Schmitt, S. Duval, A. Piéquet, A. M. Loudes, O. Rohfritsch, C. Stussi-Garaud, and P. Pfeiffer. 2002. Grapevine fanleaf virus replication occurs on endoplasmic-derived membranes. J. Virol. 76:8808-8819.[Abstract/Free Full Text]
    43. 43
  43. Rott, M. E., A. Gilchrist, L. Lee, and D. M. Rochon. 1995. Nucleotide sequence of the tomato ringspot virus RNA1. J. Gen. Virol. 76:465-473.[Abstract/Free Full Text]
    44. 44
  44. Sanfaçon, H. 1995. Nepoviruses, p. 129-141. In R. P. Singh, U. S. Singh, and K. Kohmoto (ed.), Pathogenesis and host specificity in plant diseases, vol. III. Viruses and viroids. Pergamon Press, Oxford, United Kingdom.
    45. 45
  45. Sanfaçon, H., A. Wieczorek., and F. Hans. 1995. Expression of the tomato ringspot nepovirus movement and coat proteins in protoplasts. J. Gen. Virol. 76:2299-2303.[Abstract/Free Full Text]
    46. 46
  46. Sankaram, M. B., and D. Marsh. 1993. Protein-lipid interactions with peripheral membrane proteins, p. 127-162. In A. Watts (ed.), Protein-lipid interactions. Elsevier Science Publishers B.V., Amsterdam, The Netherlands.
    47. 47
  47. Schaad, M. C., P. E. Jensen, and J. C. Carrington. 1997. Formation of plant RNA virus replication complexes on membranes: role of an endoplasmic reticulum-targeted viral protein. EMBO J. 16:4049-4059.[CrossRef][Medline]
    48. 48
  48. Schlegel, A., T. H. Giddings, Jr., M. S. Ladinsky, and K. Kirkegaard. 1996. Cellular origin and ultrastructure of membranes induced during poliovirus infection. J. Virol. 70:6576-6588.[Abstract/Free Full Text]
    49. 49
  49. Schwartz, M., J. Chen, M. Janda, M. Sullivan, J. den Boon, and P. Ahlquist. 2002. A positive-strand RNA virus replication complex parallels form and function of retrovirus capsids. Mol. Cell 9:505-514.[CrossRef][Medline]
    50. 50
  50. Staehelin, L. A. 1997. The plant ER: a dynamic organelle composed of a large number of discrete functional domains. Plant J. 11:1151-1165.[CrossRef][Medline]
    51. 51
  51. Suhy, D. A., T. H. Giddings, Jr., and K. Kirkegaard. 2000. Remodeling the endoplasmic reticulum by poliovirus infection and by individual viral proteins: an autophagy-like origin for virus-induced vesicles. J. Virol. 74:8953-8965.[Abstract/Free Full Text]
    52. 52
  52. Teterina, N. L., D. Egger, K. Bienz, D. M. Brown, B. L. Semler, and E. Ehrenfeld. 2001. Requirements for assembly of poliovirus replication complexes and negative-strand RNA synthesis. J. Virol. 75:3841-3850.[Abstract/Free Full Text]
    53. 53
  53. Towner, J. S., M. M. Mazanet, and B. L. Semler. 1998. Rescue of defective poliovirus RNA replication by 3AB-containing precursor polyproteins. J. Virol. 72:7191-7200.[Abstract/Free Full Text]
    54. 54
  54. Wang, A., K. Carrier, J. Chisholm, A. Wieczorek, C. Huguenot, and H. Sanfaçon. 1999. Proteolytic processing of tomato ringspot nepovirus 3C-like protease precursors: definition of the domains for the VPg, protease and putative RNA-dependent RNA-polymerase. J. Gen. Virol. 80:799-809.[Abstract]
    55. 55
  55. Wang, A., and H. Sanfaçon. 2000. Proteolytic processing at a novel cleavage site in the N-terminal region of the tomato ringspot nepovirus RNA-1 encoded polyprotein in vitro. J. Gen. Virol. 82:1785-1790.[Abstract/Free Full Text]
    56. 56
  56. Wieczorek, A., and H. Sanfaçon. 1993. Characterization and subcellular localization of tomato ringspot nepovirus putative movement protein. Virology 194:734-742.[CrossRef][Medline]
    57. 57
  57. Wieczorek, A., and H. Sanfaçon. 1995. An improved method for the generation and transfection of protoplasts from Cucumis sativus cotyledons. Plant Cell Rep. 114:603-610.
    58. 58
  58. Wienecke, K., R. Glas, and D. G. Robinson. 1982. Organelles involved in the synthesis and transport of hydroxyproline-containing glycoproteins in carrot root disks. Planta 155:58-63.[CrossRef]
    59. 59
  59. Wu, S. X., P. Ahlquist, and P. Kaesberg. 1992. Active complete in vitro replication of nodavirus RNA requires glycerophospholipid. Proc. Natl. Acad. Sci. USA 89:11136-11140.[Abstract/Free Full Text]
    60. 60
  60. Xiang, W., A. V. Paul, and E. Wimmer. 1997. RNA signals in entero- and rhinovirus genome replication. Semin. Virol. 8:256-273.[CrossRef]
    61. 61
  61. Zhang, G. F., and L. A. Staehelin. 1992. Functional compartmentation of the Golgi apparatus of plant cells. Plant Physiol. 99:1070-1083.[Abstract/Free Full Text]

Journal of Virology, January 2003, p. 523-534, Vol. 77, No. 1
0022-538X/03/$08.00+0     DOI: 10.1128/JVI.77.1.523-534.2003
Copyright © 2003, American Society for Microbiology. All Rights Reserved.




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Han, S.
Right arrow Articles by Sanfaçon, H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Han, S.
Right arrow Articles by Sanfaçon, H.
Right arrowPubmed/NCBI databases
*Protein
*Substance via MeSH

Copyright © 2009 by the American Society for Microbiology.   For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»