| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
NS
1 Departamento de Bioquímica y Biología Molecular, Facultad de Farmacia, Universidad de Santiago de Compostela, 15782-Santiago de Compostela, Spain
2 Center for Biosystems Research, University of Maryland Biotechnology Institute and VA-MD Regional College of Veterinary Medicine, University of Maryland, College Park, MD 20742, USA
Correspondence
Javier Benavente
bnjbena{at}usc.es
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
NS has previously been shown to bind single-stranded (ss) RNA in vitro in a sequence-independent manner. The results of the present study further reveal that NS binds poly(A), poly(U) and ssDNA, but not poly(C), poly(G) or duplex nucleic acids, suggesting that NS has some nucleotide-sequence specificity for ssRNA binding. The current findings also show that NS is present in large ribonucleoprotein complexes in the cytoplasm of avian reovirus-infected cells, indicating that it exists in intimate association with ssRNAs in vivo. Removal of RNA from the complexes generates a NS protein form that sediments between 4·5 and 7 S, suggesting that RNA-free NS associates into small oligomers. Expression and purification of recombinant NS in insect cells allowed us to generate specific antibodies and to perform a variety of assays. The results of these assays revealed that: (i) RNA-free NS exists as homodimers and homotrimers; (ii) the minimum RNA size for NS binding is between 10 and 20 nt; (iii) NS does not have a preference for viral mRNA sequences; and (iv) its RNA-binding activity is conformation-dependent. Baculovirus expression of point and deletion NS mutants in insect cells showed that the five conserved basic amino acids that are important for RNA binding and ribonucleoprotein-complex formation are dispersed throughout the entire NS sequence, suggesting that this protein binds ssRNA through conformational domains. Finally, the properties of the avian reovirus protein NS are compared with those of its mammalian reovirus counterpart. |
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
; Robertson & Wilcox, 1986; van der Heide, 2000). These cytoplasmic-replicating viruses have a genome formed by 10 segments of double-stranded RNA (dsRNA) encased by a non-enveloped, double-protein capsid shell (Spandidos & Graham, 1976). Their genome encodes at least eight structural proteins and four non-structural proteins (Bodelón et al., 2001; Schnitzer, 1985; Varela & Benavente, 1994).
Avian reovirus genes are transcribed by a core-associated RNA polymerase to produce mRNAs that are identical to the positive strands of the dsRNA segments, possessing a type 1 cap at their 5' ends and lacking a 3' poly(A) tail (Martínez-Costas et al., 1995). Reoviral mRNAs are used as templates for the synthesis of viral proteins and minus-strand RNAs. Avian reovirus replication and morphogenesis occur within globular viral factories that are not microtubule-associated (Tourís-Otero et al., 2004a).
The 367 aa avian reovirus NS protein, which is encoded by the S4 gene (Chiu & Lee, 1997; Schnitzer, 1985; Varela & Benavente, 1994), binds single-stranded RNA (ssRNA) in a sequence-independent fashion (Yin & Lee, 1998, 2000) and is present in cytoplasmic globular inclusions in infected cells through an association with the non-structural protein µNS (Tourís-Otero et al., 2004b). In this study, we have further characterized the nucleic acid-binding activity of avian reovirus NS.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
). Propagation of baculoviruses in Sf9 cells has also been described previously (Hsiao et al., 2002).
Cloning and generation of recombinant baculoviruses.
Cloning and sequencing of the recombinant plasmid pCR2.1-S4, containing the 1733 NS-encoding S4 gene, has been reported previously (Tourís-Otero et al., 2004b). To express a recombinant NS (rNS) protein in insect cells, the S4-coding sequence of the pCR2.1-S4 plasmid was PCR-amplified by using the forward primer 5'-GGAATTCGCCATGGACAACACCGTGC-3' (EcoRI site underlined) and the reverse primer 5'-GCGTCTAGACTACGCCATCCTAGCTGG-3' (XbaI site underlined). The PCR product was digested and cloned into the EcoRI and XbaI sites of pFastBac1 (Bac-to-Bac system; Invitrogen) to generate pFastBac1-S4, which was then used to produce the baculovirus Bac-NS according to the supplier's protocol. This baculovirus expresses the S4 gene under the control of the polyhedrin promoter.
To generate recombinant baculoviruses expressing deleted NS mutants, PCR amplification was performed with the following primers: to express the N-terminal mutant rNS 
N11, the forward primer was 5'-GGAATTCATGAACACATCCGGCGCACGTG-3' (EcoRI site underlined) and the reverse primer was 5'-GCGTCTAGACTACGCCATCCTAGCTGG-3' (XbaI site underlined). To express C-terminal mutants, the forward primer was 5'-GGAATTCGCCATGGACAACACCGTGC-3' and the reverse primers were 5'-GCGTCTAGATCAGATCATCCAATTACC-3' for C16, 5'-GCGTCTAGACTAATTCATGGCGAAGCCCTG-3' for C50 and 5'-GCGTCTAGACTACGCGTCCAACTCAACC-3' for C100 (XbaI sites underlined). To express point NS mutants, we performed site-directed mutagenesis of pFastBac1-S4 by using a QuikChange site-directed mutagenesis kit (Stratagene) according to the manufacturer's instructions. Mutations were performed to change to leucine those lysine and arginine residues indicated in Table 1. The mutant pFastBac1-S4 sequences were verified by restriction analysis and nucleotide sequencing. These mutants were used to generate recombinant baculoviruses as described above.
|
For velocity-sedimentation analysis, S10 fractions were left untreated or treated with either 50 µg RNase A ml–1 (Sigma-Aldrich) or 50 U RNase V1 ml–1 (Ambion) for 15 min at 37 °C. An aliquot of the RNase-treated sample was adjusted to 1 M NaCl and incubated for 15 min at 37 °C. These samples, as well as a reticulocyte lysate that had been programmed with total RNA isolated from avian reovirus-infected cells, were loaded onto 10–40 % sucrose gradients in STE buffer. After centrifugation at 150 000 g for 16 h at 4 °C in a Beckman SW50.1 rotor, 15 fractions of 300 µl were collected from the top of the tubes. These fractions and the pellets, as well as aliquots of both the original extract and purified avian reovirions, were boiled in Laemmli sample buffer and analysed by 12 % SDS-PAGE and autoradiography. Protein standards of known molecular masses and sedimentation coefficients were all from Sigma: BSA (66 kDa, 4·3S), gamma-globulin (156 kDa, 7S), catalase (250 kDa, 11·3S) and thyroglobulin (670 kDa, 19S). The standards were run on identical sucrose gradients, collected from the top and detected by measurement of A280.
For Sepharose bead-binding assays, the radioactive S10 fraction was subjected to ultracentrifugation (150 000 g for 2 h). Aliquots of the supernatant (100 µl) were supplemented with different NaCl concentrations and mixed with 50 µl of each of the various Sepharose beads indicated in Fig. 3 (Sigma). After incubation for 30 min at 4 °C in STE buffer, the mixtures were centrifuged for 30 s and subjected to five rounds of washing (with 1 ml STE buffer) and centrifugation. The pelleted beads were washed with STE buffer containing increasing salt concentrations (0·4–2·0 M NaCl) and centrifuged. The original extracts, the different washes and the final pellets were boiled in Laemmli sample buffer and analysed by 12 % SDS-PAGE and autoradiography. For competition assays, the supernatants of ultracentrifuged S10 extracts were incubated for 15 min at 4 °C in STE buffer with the indicated amounts of GTP or competitor soluble nucleic acids (0·5 mg ml–1) prior to the binding assays.
|
NS and generation of polyclonal antibodies.NS per cell and incubated at 27 °C for 72 h. The cells were then pelleted by low-speed centrifugation and resuspended in cold lysis buffer [10 mM Tris/HCl (pH 7·5), 1 mM EDTA, 10 mM NaCl and 0·5 % Triton X-100] at 2x106 cells ml–1. The extract was incubated for 10 min at 4 °C and then centrifuged for 30 min at 2500 g, the pellet was discarded and the supernatant was centrifuged at 4 °C for 1 h at 28 000 r.p.m. in a SW40 rotor (Beckman). The new supernatant was discarded and the rNS-enriched pellet was resuspended in lysis buffer containing 1 M NaCl, incubated for 15 min at 4 °C and passed through a 22-gauge needle. The mixture was centrifuged again at 4 °C for 1 h at 30 000 r.p.m. in a SW40 rotor, the pellet was discarded and the resulting supernatant was diluted five times with lysis buffer and subjected to poly(A)–Sepharose affinity chromatography. The flow-through fraction was discarded and the column was washed with buffer containing increasing salt concentrations. Protein rNS was eluted with buffer containing 0·8 M NaCl and concentrated by centrifugation using a Centricon YM-10 (Millipore). Purified rNS was used as immunogen to raise a rabbit polyclonal antiserum, as described elsewhere (Bodelón et al., 2001).
Immunoprecipitation, immunoblotting, chemical cross-linking and in vitro translation.
Immunoprecipitation and immunoblotting were performed as described previously (Tourís-Otero et al., 2004a). For chemical cross-linking, glutaraldehyde (Sigma) was diluted to the working concentration in PBS and incubated with purified rNS for 30 min at room temperature. The reactions were stopped by adding Tris/HCl (pH 7·5) to a final concentration of 50 mM, followed by a 15 min incubation. Isolation of total RNA from avian reovirus-infected cells and its translation in rabbit reticulocyte lysates have been described previously (Varela & Benavente, 1994).
Gel mobility-shift assays.
Uncapped and polyadenylated luciferase mRNA was purchased from Promega and poly(A) from Sigma. All other ssRNA probes were generated by in vitro transcription using the T7 RiboMAX Express RNA-production system (Promega). The DNA template for avian reovirus s1 mRNA transcription was generated by PCR amplification of S1 sequences contained within the pBsct-S1 plasmid (Bodelón et al., 2001) with the forward primer 5'-GCGTAATACGACTCACTATAGGCTTTTTCAATCCCTT-3' (T7 promoter sequence underlined) and the reverse primer 5'-GATGAATAACCAATCCCAGTAC-3'. The template for the synthesis of the 10 nt RNA was generated by annealing the oligonucleotides 5'-GCGTAATACGACTCACTATAGG-3' and 5'-GCGTGGTACCTATAGTGAGTCGTATTACGC-3'. The template for the synthesis of the 20 nt RNA was generated by annealing the oligonucleotides 5'-GCGTAATACGACTCACTATAGG-3' and 5'-GAGAATTCACGCGTGGTACCTATAGTGAGTCGTATTACGC-3'. The latter primer was also used as a 40 nt ssDNA probe. Annealing of this primer to its complement produced a 40 bp dsDNA probe. A 20 bp dsRNA was generated by annealing the 20 nt RNA with the RNA synthesized by T7 run-off transcription of NotI-predigested plasmid pCI-Neo.
The 5' end of the probes (20 pmol) was dephosphorylated by incubation with 20 U alkaline phosphatase (Roche Diagnostics) and then radiolabelled by incubation for 1 h at 37 °C with 50 µCi (1·85 MBq) [
-32P]ATP (ICN) and 5 U T4 polynucleotide kinase (Promega) in 70 mM Tris/HCl (pH 7·6), 10 mM MgCl2 and 5 mM dithiothreitol. Unincorporated nucleotides were removed by Sepharose G50 chromatography.
Gel mobility-shift assays were performed by incubating 75 ng to 2 µg purified rNS in PBS with 10 000 c.p.m. of each radiolabelled probe in 20 µl STE buffer containing 10 U RNasin for 15 min at 4 °C. The samples were mixed with gel-loading buffer (0·25 % bromophenol blue, 0·25 % xylene cyanol, 30 % glycerol in H2O) and subjected to electrophoresis in 10 % polyacrylamide native gels in TBE buffer. At the end of the runs, the gels were dried and exposed to X-ray films. For competition assays, 1 µg purified rNS was preincubated with 0–2 µg non-radioactive RNA prior to the addition of 50 ng radiolabelled s1 mRNA probe.
Membrane-filter assays.
North-Western blot assays were performed with extracts from insect cells that had been infected for 72 h with wild-type baculovirus or with recombinant baculovirus Bac-NS, as described by González & Ortín (1999). Proteins were separated in 12 % SDS-PAGE gels and transferred onto nitrocellulose membranes in 25 mM Tris, 192 mM glycine (pH 8·3). Membranes were incubated for 16 h at 4 °C in renaturing buffer [50 mM NaCl, 1 mM EDTA, 0·02 % Ficoll, 0·02 % BSA, 0·02 % polyvinylpyrrolidone, 0·1 % Triton X-100, 10 mM Tris/HCl (pH 7·5)], then incubated for 2 h at room temperature in renaturing buffer containing 106 c.p.m. 32P-radiolabelled s1 mRNA, washed four times (1 h each) with renaturing buffer and finally dried and exposed to X-ray film.
For membrane-filter assays, serial threefold dilutions of purified rNS, ranging from 0·1 to 9 µg, were prepared in STE buffer. BSA was then added to equalize the protein content of each sample to 9 µg and the final volume to 150 µl. Half of each sample was incubated with 3·5 M urea at 50 °C for 15 min and the other half was mock-incubated. Aliquots containing one-third of each sample were deposited onto three different 0·2 µm nitrocellulose membranes (Bio-Rad) by using a dot-blot apparatus (Millipore). The first membrane was incubated with Ponceau S solution (0·1 % Ponceau S in 5 % acetic acid; Sigma), the second membrane with anti-rNS antibodies and the third membrane first for 1 h in STE buffer containing 0·5 mg BSA ml–1 and then an additional hour in the same buffer containing 100 000 c.p.m. 32P-radiolabelled s1 mRNA. The membrane was finally washed five times in STE buffer containing 0·5 mg BSA ml–1, then dried and exposed to X-ray film.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
NS is found in large ribonucleoprotein complexes in avian reovirus-infected cellsNS to associate with nucleic acids in vivo, both a reticulocyte lysate programmed with viral RNA and a virus-free cytoplasmic extract from avian reovirus-infected cells were subjected to velocity-sedimentation analysis through 10–40 % sucrose gradients. The 35S-radiolabelled proteins present in the original extracts, in the gradient fractions and in the pellets, as well as in purified avian reovirions, were analysed by SDS-PAGE and autoradiography (Fig. 1). Most of the in vitro-translated NS sedimented in fractions 4 and 5, a narrow zone between the 4·3S and 7S protein markers (Fig. 1a), although faster than expected for a globular monomer of 40 kDa (3·5S), suggesting that it is an oligomer. In contrast, most of the NS present in extracts of infected cells was detected in the pellet fraction (lane P in Fig. 1b), indicating that NS is present in large complexes. To assess whether NS is associated with RNA within the complexes, the extract from infected cells was treated with either RNase V1 or RNase A before centrifugation. The RNase V1 treatment did not alter the sedimentation profile of NS, suggesting that NS is not associated with dsRNA (Fig. 1c). In contrast, the RNase A treatment caused a shift in the NS distribution to slower-sedimenting fractions (Fig. 1d), suggesting both that ssRNA is present in the NS-containing complexes and that NS is associated with ssRNA within the complexes. The RNase A treatment did not, however, generate a homogeneous NS population, as the protein sedimented in a broad zone between fractions 4 and 15, with a large NS fraction still present in the pellet (Fig. 1d). Subsequent incubation of the RNase A-treated extract with 1 M NaCl caused NS to sediment in the same fractions as the in vitro-translated protein, suggesting that the combined treatment caused total RNA removal from the complexes (Fig. 1e). Collectively, our results demonstrate that NS is associated with ssRNA within large ribonucleoprotein complexes in infected cells.
|
NS in insect cells, purification and antibody generationNS for antibody production and biochemical characterization, we generated the recombinant baculovirus Bac-NS, which contains the avian reovirus S4 gene under the control of the polyhedrin promoter. Expression of high levels of a 40 kDa protein, designated rNS, was observed upon infection of Sf9 cells with Bac-NS, but not with wild-type baculovirus (Fig. 2a, compare lanes 1 and 2). The authenticity of the rNS protein synthesized in insect cells was confirmed in different ways. Firstly, rNS and NS from avian reovirus-infected cells have identical electrophoretic mobilities (Fig. 2a, compare lanes 10 and 11) and both are able to bind to poly(A)–Sepharose and to form large ribonucleoprotein complexes (see below, Fig. 6a). Secondly, anti-rNS antibodies recognized the NS protein from avian reovirus-infected cells in immunoprecipitation and immunoblotting (Fig. 2b, c).
|
|
NS occurred 72 h after infection of Sf9 cells with 5 p.f.u. Bac-NS per cell (data not shown) and this time was therefore chosen for large-scale rNS production. Based on its ability to form large ribonucleoprotein complexes and to bind ssRNA, we devised a simple and highly efficient scheme for rNS purification, as described in Methods (Fig. 2a
). Virtually pure rNS was eluted from a poly(A)–Sepharose column with buffer containing 0·8 M NaCl, as judged by Coomassie blue staining of an SDS-PAGE gel (lane 10). A sample of NS similarly purified from avian reovirus-infected cells is shown in lane 11. We generally obtained 15 mg pure rNS from 1 l Bac-NS-infected Sf9 cells. The purified protein was used as immunogen to generate polyclonal anti-rNS antibodies in rabbits.
Cross-linking with glutaraldehyde was performed next, to evaluate the oligomeric state of purified rNS. The immunoblot shown in Fig. 2(d) revealed that, whereas the anti-NS antibodies only recognized a 40 kDa protein band before cross-linking (lane 1), two additional bands of 80 and 110 kDa showed up after incubation with glutaraldehyde (lanes 2 and 3), suggesting that they correspond to NS homodimers and homotrimers.
Binding to immobilized nucleic acids
To further characterize the in vitro nucleic acid-binding activity of NS, a cytoplasmic extract from 35S-amino acid-labelled avian reovirus-infected cells was ultracentrifuged and incubated with different resin-immobilized nucleic acids; the beads were subsequently washed with buffer containing increasing salt concentrations (Fig. 3a). The radioactive proteins in the original extracts (lane 8), in the flow-through fractions (lane 1) and in the different washes (lanes 2–6), as well as those that remained attached to the matrix after the final wash (lane 7), were resolved by SDS-PAGE and visualized by autoradiography. Only the viral proteins A and NS displayed binding activity for the nucleic acids used in these assays, but not for control Sepharose beads (upper panel). Protein A bound very tightly to poly(I : C), as it remained resin-bound even in the presence of 2 M NaCl and was able to bind poly(I : C)–Sepharose in the presence of 2·5 M NaCl (Fig. 3b). On the other hand, NS bound with moderate affinity to poly(A), poly(U) and ssDNA, but not to poly(G), poly(C), poly(I : C) or dsDNA. Based on the salt concentration required for elution, we conclude that NS exhibits a higher affinity for poly(A) than for poly(U), and for poly(U) than for ssDNA (Fig. 3a). A similar conclusion was reached when the poly(A)-binding assays were performed with in vitro-translated NS (data not shown), or when monitoring NS attachment to, rather than elution from, the affinity beads at different salt concentrations (Fig. 3b).
To further characterize the nucleic acid-binding activities of A and NS, competition-binding assays with soluble, non-radioactive nucleic acids were performed. As shown in Fig. 3(c), the binding of A to poly(I : C) could only be inhibited by soluble viral dsRNA and not by ssRNA, ssDNA or dsDNA, confirming previous observations that A binds exclusively to dsRNA in a sequence-independent manner (Martínez-Costas et al., 2000; Yin et al., 2000). The binding of NS to poly(A) and poly(U) could only be competed by soluble poly(A) or poly(U), not by the other nucleic acids tested.
It has been reported that GTP concentrations over 0·5 mM outcompete the binding of mammalian reovirus non-structural protein NS (mrNS) to poly(A) and poly(U) (Richardson & Furuichi, 1985). To assess whether a similar situation holds true for avian reovirus NS (arNS), radiolabelled cytoplasmic extracts of both avian and mammalian reovirus-infected cells were supplemented with different GTP concentrations before incubation with poly(A)–Sepharose beads. After washing the beads with binding buffer, the attached proteins were eluted by boiling in Laemmli sample buffer and analysed by SDS-PAGE and autoradiography (Fig. 3d). Whilst GTP concentrations over 1 mM reduced the binding affinity of mrNS to poly(A) (compare lanes 11 and 12), they apparently enhanced the affinity of arNS for poly(A) (compare lanes 5 and 6), suggesting that GTP has opposing effects on the RNA-binding activity of the two NS proteins.
Gel mobility-shift assays
Consistent with the results of the Sepharose-immobilized nucleic acids, our gel-shift assays revealed that rNS forms complexes with radiolabelled ssRNA and ssDNA, but not with dsRNA or dsDNA (Fig. 4a), and that the intensity of the ssRNA–rNS band was directly dependent upon the amount of rNS used in the assays (Fig. 4b). We further found that purified rNS also forms complexes with poly(A) and luciferase mRNA (data not shown). When fixed amounts of rNS were mixed with increasing amounts of unlabelled s1 mRNA or luciferase mRNA before the addition of the radiolabelled avian reovirus s1 mRNA probe, the two unlabelled mRNAs competed with similar efficiencies with radiolabelled s1 mRNA for rNS binding (Fig. 4c). A similar competition efficiency was also exhibited by poly(A) (data not shown). These results suggest that NS does not have a preference for avian reovirus sequences.
|
NS binding, fixed amounts of 32P-labelled ssRNAs of different sizes were incubated with increasing concentrations of rNS and subjected to gel-shift analysis. The results revealed that, whilst rNS formed complexes readily with a 20 nt RNA, it did not do so with a 10 nt RNA (Fig. 4d), indicating that the minimum RNA size required for NS binding is between 10 and 20 nt.
Filter-binding assays
The RNA-binding activity of rNS was also analysed by filter-binding assays. For this, the proteins present in extracts from both Bac-NS- and wild-type baculovirus-infected Sf9 cells were resolved by SDS-PAGE and transferred onto nitrocellulose filters, and the filters were then probed with 32P-labelled s1 mRNA (Fig. 5a). Whilst RNA binding by a cellular or baculovirus 30 kDa protein was evident in extracts from baculovirus-infected cells (lanes 3 and 4), no RNA binding by a protein migrating in the position of rNS (40 kDa) could be observed (lane 4). These results suggest that filter-immobilized NS monomers do not bind ssRNA. Further RNA-binding assays were performed with purified rNS spotted directly onto membranes, either before or after denaturation with urea (Fig. 5b). Ponceau S staining of one membrane revealed that all wells contained amounts of similar protein (top panel). Incubation of the second membrane with anti-NS antibodies demonstrated that both native and denatured rNS were retained by the membrane with similar efficiencies (medium panel). Probing the third membrane with 32P-labelled avian reovirus s1 mRNA revealed that only native rNS displayed RNA-binding activity (bottom panel).
|
NS mutagenesis on RNA binding and complex formationNS that might be important for RNA binding and complex formation, we initially assayed the complex-forming and RNA-binding capacities of rNS versions lacking end sequences. We first generated the recombinant baculovirus Bac-NS N11, which encodes a NS version lacking the 11 most N-terminal residues, including two arginines at positions 6 and 11. Both rNS N11 and full-length rNS were detected in soluble extracts from baculovirus-infected cells. These extracts were subsequently subjected to poly(A)–Sepharose-binding and velocity-sedimentation analysis, and the resulting fractions were subsequently analysed by Western blot using anti-rNS antibodies (Fig. 6a). As with NS from avian reovirus-infected cells, full-length rNS bound poly(A) (left top panel) and formed large complexes (centre top panel), which could be partially disassembled by preincubation with RNase A (right top panel), suggesting that no other avian reovirus factors are required for the NS activities. In contrast, rNS N11 did not bind poly(A) or form large complexes (Fig. 6a, bottom panels), indicating that the region comprising the first 11 NS residues is important for RNA binding and complex formation.
We next generated recombinant baculoviruses that expressed proteins lacking the C-terminal 16, 50 and 100 residues from the C terminus of NS. Unfortunately, none of the C-terminally truncated proteins expressed in insect cells were found in the soluble fractions, which precluded their use for further analysis (Fig. 6b). These results, however, suggest that the NS C-terminal sequences are important for stability and/or solubility of the viral protein.
Site-directed mutagenesis was performed next to identify specific residues that are important for the activities of NS. An alignment of the avian reovirus 1733 and S1133 NS sequences with those of their counterparts from muscovy duck reovirus, Nelson Bay virus and mammalian reovirus type 3 revealed that the only conserved basic residues in all four proteins were those at avian reovirus NS positions 6, 11, 29, 67, 80, 223, 234, 251, 262, 287, 310 and 365 (indicated by triangles in Fig. 7). Recombinant baculoviruses expressing mutated NS in which each of these basic residues was changed to leucines were generated and used to infect insect cells. All mutants were detected in the soluble fractions (Fig. 6c), which allowed us to test their capacity to bind poly(A) and to form complexes. The results, summarized in Table 1, revealed that mutation of the arginines at the NS positions 6, 11, 234, 287 and 365 (indicated by filled triangles in Fig. 7) prevented RNA binding and complex formation, whereas mutation of the other conserved basic residues did not (indicated by empty triangles in Fig. 7). Furthermore, all mutants that did not bind RNA exhibited the same electrophoretic mobility as non-mutated rNS, indicating that they did not undergo degradation (Fig. 6c).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
NS, as a major non-structural, ssRNA-binding protein that accumulates in viral factories of avian reovirus-infected cells (Yin & Lee, 1998; Tourís-Otero et al., 2004b) identifies NS as a good candidate for playing key roles in RNA packaging and replication. It was therefore of interest to characterize the nucleic acid-binding properties of this protein.
Our results reveal that NS functions as an RNA-binding protein in infected cells and that it possesses ssRNA- and ssDNA-binding activity in vitro. Our observation that avian reovirus NS binds poly(A) and poly(U), but not poly(C) or poly(G), suggests some specificity in the NS–RNA interaction. Although the significance of this specificity is not clear, it could be related to viral mRNA recognition, as all avian reovirus mRNAs sequenced so far contain 5'-GCUUUUU-3' and 5'-UUAUUCAUC-3' at their 5' and 3' ends, respectively, and these U-rich sequences are predicted to adopt a single-stranded conformation (data not shown). However, this hypothesis is not supported by our findings that rNS does not have a preference for viral sequences in gel-shift assays and that rNS associates with RNA in insect cells. Furthermore, the fact that poly(A), luciferase mRNA and s1 mRNA compete for the binding of NS to the s1 mRNA with similar efficiencies suggests that NS lacks sequence specificity and has no preference for capped and/or polyadenylated RNAs.
The possibility exists that viral cofactors enhance the RNA-binding specificity of NS within infected cells; a good candidate would be µNS, which associates with NS and mediates its recruitment into viral factories (Tourís-Otero et al., 2004b). Another possibility is that cellular mRNAs are excluded from the factories, obviating the need for inclusion-associated NS to have a preference for viral sequences. In this scenario, sequence-independent RNA-binding activity would still allow NS to concentrate and retain viral mRNAs within the factories, to provide them with RNase protection, to destabilize duplex regions and/or to improve their template efficiency for minus-strand synthesis. Sequestration of viral mRNAs within factories would direct them towards replication and would preclude their release for cytosolic translation.
The results of our RNA filter-binding assays suggest that native NS conformation is essential for RNA binding. One likely explanation is that NS RNA-binding activity is dependent on multimerization, as has been reported to occur with rotavirus NSP2 (Boyle & Holmes, 1986; Taraporewala et al., 1999). Consistent with this hypothesis, native rNS spotted onto a membrane filter binds RNA (Fig. 5b, bottom panel), whereas the protein subjected to SDS-PAGE and renatured does not (Fig. 5a, lane 4). These findings appear to suggest that oligomeric NS, but not monomeric NS, possesses RNA-binding activity. However, we cannot rule out the possibility that a properly folded NS monomer binds RNA as well, and that the lack of RNA-binding activity shown by rNS in the North-Western blot assay is due to inability of the renaturing conditions to induce correct NS folding.
The fact that five conserved arginine residues, dispersed widely along the primary NS sequence, are important for poly(A) binding and complex formation suggests that NS interacts with RNA through conformational domains and not through a linear RNA recognition motif, like the linear motifs found in various nucleic-acid binding proteins (Burd & Dreyfuss, 1994; Kenan et al., 1991; Kochan et al., 2003; Lazinski et al., 1989; Merrill et al., 1988; Query et al., 1989). One possibility is that NS interacts with RNA through various domains located at different positions in the primary NS sequence, as occurs with several other RNA-binding proteins (Cordingley et al., 1990; Fillmore et al., 2002; González & Ortín, 1999; Rould et al., 1989). Another possibility is that the five basic residues that are important for RNA binding may cluster together on native NS to create a conformational RNA-binding motif. These basic residues are likely to be surface-exposed and to make direct contacts with RNA. Alternatively, they could be important for maintaining the native NS conformation, although this possibility is unlikely, as the mutated rNS versions that do not bind RNA still interact with µNS and are recruited to µNS inclusions (unpublished data). Furthermore, all point mutants and the N-terminal NS truncation used in this study are present in the soluble fraction, whereas the NS C-terminal truncations are insoluble. Collectively, these results suggest both that the C-terminal sequences of NS are important for its conformational stability and that the point mutants and the N-terminal truncation are not misfolded.
The avian and mammalian reovirus NS proteins have similar characteristics. Both are proteins of 
40 kDa that bind DNA and RNA non-specifically in vitro, that assemble into ribonucleoprotein complexes and that accumulate into viral factories through an association with µNS (Gillian & Nibert, 1998; Gillian et al., 2000; Gomatos et al., 1981; Huismans & Joklik, 1976; Miller et al., 2003; Richardson & Furuichi, 1985; Tourís-Otero et al., 2004b; Yin & Lee, 1998). However, the results of this study also revealed specific differences between the two proteins. Firstly, whereas the 4·3–7S RNA-free form of arNS exits as dimers and trimers, the RNA-free form of mrNS appears to contain more monomeric subunits, as it sediments between 7 and 9 S (Gillian & Nibert, 1998; Gillian et al., 2000). Secondly, whilst mrNS binds equally well to all four ribopolynucleotides at low salt concentrations and has a slight preference for poly(U) at 0·4 M NaCl (Huismans & Joklik, 1976), arNS does not bind poly(C) or poly(G) and exhibits a slight preference for poly(A) over poly(U). Thirdly, GTP exerts opposite effects on the binding affinity of mrNS and arNS to poly(A)–Sepharose (Richardson & Furuichi, 1985). Fourthly, mrNS has been detected in the nucleus and cytoplasm of infected and transfected cells and has been reported to bind native and denatured DNA (Miller et al., 2003; Shelton et al., 1981), whereas arNS is not present within the nucleus and has no affinity for dsDNA (Tourís-Otero et al., 2004b). Finally, the N-terminal 118 aa of mrNS have been reported to be sufficient for RNA binding and complex formation (Gillian & Nibert, 1998), whereas basic residues located at both the N and C termini of arNS are required for these activities.
|
ACKNOWLEDGEMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Boyle, J. F. & Holmes, K. V. (1986). RNA-binding proteins of bovine rotavirus. J Virol 58, 561–568.
Burd, C. G. & Dreyfuss, G. (1994). Conserved structures and diversity of functions of RNA-binding proteins. Science 265, 615–621.
Chiu, C. J. & Lee, L. H. (1997). Cloning and nucleotide sequencing of the S4 genome segment of avian reovirus S1133. Arch Virol 142, 2515–2520.[CrossRef][Medline]
Cordingley, M. G., LaFemina, R. L., Callahan, P. L. & 8 other authors (1990). Sequence-specific interaction of Tat protein and Tat peptides with the transactivation-responsive sequence element of human immunodeficiency virus type 1 in vitro. Proc Natl Acad Sci U S A 87, 8985–8989.
Fillmore, G. C., Lin, H. & Li, J. K.-K. (2002). Localization of the single-stranded RNA-binding domains of bluetongue virus nonstructural protein NS2. J Virol 76, 499–506.
Gillian, A. L. & Nibert, M. L. (1998). Amino terminus of reovirus nonstructural protein NS is important for ssRNA binding and nucleoprotein complex formation. Virology 240, 1–11.[CrossRef][Medline]
Gillian, A. L., Schmechel, S. C., Livny, J., Schiff, L. A. & Nibert, M. L. (2000). Reovirus protein NS binds in multiple copies to single-stranded RNA and shares properties with single-stranded DNA binding proteins. J Virol 74, 5939–5948.
Gomatos, P. J., Prakash, O. & Stamatos, N. M. (1981). Small reovirus particles composed solely of sigma NS with specificity for binding different nucleic acids. J Virol 39, 115–124.
González, S. & Ortín, J. (1999). Characterization of influenza virus PB1 protein binding to viral RNA: two separate regions of the protein contribute to the interaction domain. J Virol 73, 631–637.
Grande, A. & Benavente, J. (2000). Optimal conditions for the growth, purification and storage of the avian reovirus S1133. J Virol Methods 85, 43–54.[CrossRef][Medline]
Hsiao, J., Martínez-Costas, J., Benavente, J. & Vakharia, V. N. (2002). Cloning, expression, and characterization of avian reovirus guanylyltransferase. Virology 296, 288–299.[CrossRef][Medline]
Huismans, H. & Joklik, W. K. (1976). Reovirus-coded polypeptides in infected cells: isolation of two native monomeric polypeptides with affinity for single-stranded and double-stranded RNA, respectively. Virology 70, 411–424.[CrossRef][Medline]
Jones, R. C. (2000). Avian reovirus infections. Rev Sci Tech 19, 614–625.[Medline]
Kenan, D. J., Query, C. C. & Keene, J. D. (1991). RNA recognition: towards identifying determinants of specificity. Trends Biochem Sci 16, 214–220.[CrossRef][Medline]
Kochan, G., Gonzalez, D. & Rodriguez, J. F. (2003). Characterization of the RNA-binding activity of VP3, a major structural protein of Infectious bursal disease virus. Arch Virol 148, 723–744.[CrossRef][Medline]
Lazinski, D., Grzadzielska, E. & Das, A. (1989). Sequence-specific recognition of RNA hairpins by bacteriophage antiterminators requires a conserved arginine-rich motif. Cell 59, 207–218.[CrossRef][Medline]
Martínez-Costas, J., Varela, R. & Benavente, J. (1995). Endogenous enzymatic activities of the avian reovirus S1133: identification of the viral capping enzyme. Virology 206, 1017–1026.[CrossRef][Medline]
Martínez-Costas, J., Gónzalez-López, C., Vakharia, V. N. & Benavente, J. (2000). Possible involvement of the double-stranded RNA-binding core protein A in the resistance of avian reovirus to interferon. J Virol 74, 1124–1131.
Merrill, B. M., Stone, K. L., Cobianchi, F., Wilson, S. H. & Williams, K. R. (1988). Phenylalanines that are conserved among several RNA-binding proteins form part of a nucleic acid-binding pocket in the A1 heterogeneous nuclear ribonucleoprotein. J Biol Chem 263, 3307–3313.
Miller, C. L., Broering, T. J., Parker, J. S. L., Arnold, M. M. & Nibert, M. L. (2003). Reovirus NS protein localizes to inclusions through an association requiring the µNS amino terminus. J Virol 77, 4566–4576.
Query, C. C., Bentley, R. C. & Keene, J. D. (1989). A common RNA recognition motif identified within a defined U1 RNA binding domain of the 70K U1 snRNP protein. Cell 57, 89–101.[CrossRef][Medline]
Richardson, M. A. & Furuichi, Y. (1985). Synthesis in Escherichia coli of the reovirus nonstructural protein NS. J Virol 56, 527–533.
Robertson, M. D. & Wilcox, G. E. (1986). Avian reovirus. Vet Bull 56, 155–174.
Rould, M. A., Perona, J. J., Söll, D. & Steitz, T. A. (1989). Structure of E. coli glutaminyl-tRNA synthetase complexed with tRNA(Gln) and ATP at 2·8 Å resolution. Science 246, 1135–1142.
Schnitzer, T. J. (1985). Protein coding assignment of the S genes of the avian reovirus S1133. Virology 141, 167–170.[CrossRef][Medline]
Shelton, I. H., Kasupski, G. J., Jr, Oblin, C. & Hand, R. (1981). DNA binding of a nonstructural reovirus protein. Can J Biochem 59, 122–130.[Medline]
Spandidos, D. A. & Graham, A. F. (1976). Physical and chemical characterization of an avian reovirus. J Virol 19, 968–976.
Taraporewala, Z. F., Chen, D. & Patton, J. T. (1999). Multimers formed by the rotavirus nonstructural protein NSP2 bind to RNA and have nucleoside triphosphatase activity. J Virol 73, 9934–9943.
Tourís-Otero, F., Cortez-San Martín, M., Martínez-Costas, J. & Benavente, J. (2004a). Avian reovirus morphogenesis occurs within viral factories and begins with the selective recruitment of NS and 
A to µNS inclusions. J Mol Biol 341, 361–374.[CrossRef][Medline]
Tourís-Otero, F., Martínez-Costas, J., Vakharia, V. N. & Benavente, J. (2004b). Avian reovirus nonstructural protein µNS forms viroplasm-like inclusions and recruits protein NS to these structures. Virology 319, 94–106.[CrossRef][Medline]
van der Heide, L. (2000). The history of avian reovirus. Avian Dis 44, 638–641.[CrossRef][Medline]
Varela, R. & Benavente, J. (1994). Protein coding assignment of avian reovirus strain S1133. J Virol 68, 6775–6777.
Yin, H. S. & Lee, L. H. (1998). Identification and characterization of RNA-binding activities of avian reovirus non-structural protein NS. J Gen Virol 79, 1411–1413.[Abstract]
Yin, H. S. & Lee, L. H. (2000). Characterization of avian reovirus non structural protein NS synthesized in Escherichia coli. Virus Res 67, 1–9.[CrossRef][Medline]
Yin, H. S., Shien, J.-H. & Lee, L. H. (2000). Synthesis in Escherichia coli of avian reovirus core protein A and its dsRNA-binding activity. Virology 266, 33–41.[CrossRef][Medline]
Received 30 July 2004;
accepted 11 January 2005.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| INT J SYST EVOL MICROBIOL | MICROBIOLOGY | J GEN VIROL |
| J MED MICROBIOL | ALL SGM JOURNALS | |
and those whose mutation did not interfere with the binding activity of 
. The sequences shown in the figure correspond to avian reovirus 1733 (1733; GenBank accession no. AY303992), avian reovirus S1133 (Avian; accession no. U95952), muscovy duck reovirus (Muscovy; accession no. AJ133122), Nelson Bay reovirus (Nelson; accession no. AF059726) and mammalian reovirus type 3 Dearing (Mammalian; accession no. NC_004283).