Rotavirus protein NSP3 (NS34) is bound to the 3' end consensus sequence of viral mRNAs in infected cells

Recognition of eIF4G by rotavirus NSP3 reveals a basis for mRNA circularization

Rotaviruses, segmented double-stranded RNA viruses, co-opt the eukaryotic translation machinery with the aid of nonstructural protein 3 (NSP3), a rotaviral functional homolog of the cellular poly(A) binding protein (PABP). NSP3 binds to viral mRNA 3' consensus sequences and circularizes mRNA via interactions with eIF4G. Here, we present the X-ray structure of the C-terminal domain of NSP3 (NSP3-C) recognizing a fragment of eIF4GI. Homodimerization of NSP3-C yields a symmetric, elongated, largely alpha-helical structure with two hydrophobic eIF4G binding pockets at the dimer interface. Site-directed mutagenesis and isothermal titration calorimetry documented that NSP3 and PABP use analogous eIF4G recognition strategies, despite marked differences in tertiary structure.

Recognition of the rotavirus mRNA 3' consensus by an asymmetric NSP3 homodimer

Rotaviruses, the cause of life-threatening diarrhea in humans and cattle, utilize a functional homolog of poly(A) binding protein (PABP) known as nonstructural protein 3 (NSP3) for translation of viral mRNAs. NSP3 binds to viral mRNA 3' consensus sequences and circularizes the mRNA via interactions with eIF4G. The X-ray structure of the NSP3 RNA binding domain bound to a rotaviral mRNA 3' end has been determined. NSP3 is a novel, heart-shaped homodimer with a medial RNA binding cleft. The homodimer is asymmetric, and contains two similar N-terminal segments plus two structurally different C-terminal segments that intertwine to create a tunnel enveloping the mRNA 3' end. Biophysical studies demonstrate high affinity binding leading to increased thermal stability and slow dissociation kinetics, consistent with NSP3 function.

Cap-Poly(A) synergy in mammalian cell-free extracts. Investigation of the requirements for poly(A)-mediated stimulation of translation initiation

The 5' cap and 3' poly(A) tail of eukaryotic mRNAs cooperate to stimulate synergistically translation initiation in vivo, a phenomenon observed to date in vitro only in translation systems containing endogenous competitor mRNAs. Here we describe nuclease-treated rabbit reticulocyte lysates and HeLa cell cytoplasmic extracts that reproduce cap-poly(A) synergy in the absence of such competitor RNAs. Extracts were rendered poly(A)-dependent by ultracentrifugation to partially deplete them of ribosomes and associated initiation factors. Under optimal conditions, values for synergy in reticulocyte lysates approached 10-fold. By using this system, we investigated the molecular mechanism of poly(A) stimulation of translation. Maximal cap-poly(A) cooperativity required the integrity of the eukaryotic initiation factor 4G-poly(A)-binding protein (eIF4G-PABP) interaction, suggesting that synergy results from mRNA circularization. In addition, polyadenylation stimulated uncapped cellular mRNA translation and that driven by the encephalomyocarditis virus internal ribosome entry segment (IRES). These effects of poly(A) were also sensitive to disruption of the eIF4G-PABP interaction, suggesting that 5'-3' end cross-talk is functionally conserved between classical mRNAs and an IRES-containing mRNA. Finally, we demonstrate that a rotaviral non-structural protein that evicts PABP from eIF4G is capable of provoking the shut-off of host cell translation seen during rotavirus infection.

Efficient translation of rotavirus mRNA requires simultaneous interaction of NSP3 with the eukaryotic translation initiation factor eIF4G and the mRNA 3' end

In contrast to the vast majority of cellular proteins, rotavirus proteins are translated from capped but nonpolyadenylated mRNAs. The viral nonstructural protein NSP3 specifically binds the 3'-end consensus sequence of viral mRNAs and interacts with the eukaryotic translation initiation factor eIF4G. Here we show that expression of NSP3 in mammalian cells allows the efficient translation of virus-like mRNA. A synergistic effect between the cap structure and the 3' end of rotavirus mRNA was observed in NSP3-expressing cells. The enhancement of viral mRNA translation by NSP3 was also observed in a rabbit reticulocyte lysate translation system supplemented with recombinant NSP3. The use of NSP3 mutants indicates that its RNA- and eIF4G-binding domains are both required to enhance the translation of viral mRNA. The results reported here show that NSP3 forms a link between viral mRNA and the cellular translation machinery and hence is a functional analogue of cellular poly(A)-binding protein.

eIF4 initiation factors: effectors of mRNA recruitment to ribosomes and regulators of translation

Eukaryotic translation initiation factor 4F (eIF4F) is a protein complex that mediates recruitment of ribosomes to mRNA. This event is the rate-limiting step for translation under most circumstances and a primary target for translational control. Functions of the constituent proteins of eIF4F include recognition of the mRNA 5' cap structure (eIF4E), delivery of an RNA helicase to the 5' region (eIF4A), bridging of the mRNA and the ribosome (eIF4G), and circularization of the mRNA via interaction with poly(A)-binding protein (eIF4G). eIF4 activity is regulated by transcription, phosphorylation, inhibitory proteins, and proteolytic cleavage. Extracellular stimuli evoke changes in phosphorylation that influence eIF4F activity, especially through the phosphoinositide 3-kinase (PI3K) and Ras signaling pathways. Viral infection and cellular stresses also affect eIF4F function. The recent determination of the structure of eIF4E at atomic resolution has provided insight about how translation is initiated and regulated. Evidence suggests that eIF4F is also implicated in malignancy and apoptosis.

A four-nucleotide translation enhancer in the 3'-terminal consensus sequence of the nonpolyadenylated mRNAs of rotavirus

The 5' cap and poly(A) tail of eukaryotic mRNAs work synergistically to enhance translation through a process that requires interaction of the cap-associated eukaryotic initiation factor, eIF-4G, and the poly(A)-binding protein, PABP. Because the mRNAs of rotavirus, and other members of the Reoviridae, contain caps but lack poly(A) tails, their translation may be enhanced through a unique mechanism. To identify translation-enhancement elements in the viral mRNAs that stimulate translation in vivo, chimeric RNAs were prepared that contained an open reading frame for luciferase and the 5' and 3' untranslated regions (UTRs) of a rotavirus mRNA or of a nonviral mRNA. Transfection of the chimeric RNAs into rotavirus-infected cells showed that the viral 3' UTR contained a translation-enhancement element that promoted gene expression. The element did not enhance gene expression in uninfected cells and did not affect the stability of the RNAs. Mutagenesis showed that the conserved sequence GACC located at the 3' end of rotavirus mRNAs operated as an enhancement element. The 3'-GACC element stimulated protein expression independently of the sequence of the 5' UTR, although efficient expression required the RNA to contain a cap. The results indicate that the expression of viral proteins in rotavirus-infected cells is specifically up-regulated by the activity of a novel 4-nt 3' translation enhancer (TE) common to the 11 nonpolyadenylated mRNAs of the virus. The 4-nt sequence of the rotavirus 3' TE represents by far the shortest of any of the sequence enhancers known to stimulate translation.

Efficient rescue of infectious bursal disease virus from cloned cDNA: evidence for involvement of the 3'-terminal sequence in genome replication

To study the mechanism of replication of infectious bursal disease virus (IBDV), and to determine factors on the IBDV RNA which are involved in viral replication, we used cloned full-length cDNA of both the A- and B-segments to generate infectious IBDV. Infectious IBDV was rescued from plasmids that contained full-length IBDV cDNA behind a T7 promoter, by transfecting these plasmids into cells which were infected with a recombinant Fowlpox virus that expressed T7 RNA polymerase. By using the cDNA transfection system we evaluated the effect of the length of the 3' terminus of the A-segment plus strand of IBDV. Although wild-type IBDV predominantly contains four cytosines at the 3' terminus, no difference in virus yield was found when virus was rescued from cDNAs containing three to six adjacent cytosines. When the 3' terminus was shorter than three cytosines the efficiency to generate infectious IBDV from cDNA was reduced, but IBDV could still be recovered reproducibly. The rescued viruses from cDNAs containing 3'-terminal deletions appeared to have a restored 3'-terminal sequence. The missing nucleotides are probably restored by using complementary bases of a stem-loop structure as template.

Identification of the RNA-binding, dimerization, and eIF4GI-binding domains of rotavirus nonstructural protein NSP3

The rotavirus nonstructural protein NSP3 is a sequence-specific RNA binding protein that binds the nonpolyadenylated 3' end of the rotavirus mRNAs. NSP3 also interacts with the translation initiation factor eIF4GI and competes with the poly(A) binding protein. Deletion mutations and point mutations of NSP3 from group A rotavirus (NSP3A), expressed in Escherichia coli, indicate that the RNA binding domain lies between amino acids 4 and 149. Similar results were obtained with NSP3 from group C rotaviruses. Data also indicate that a dimer of NSP3A binds one molecule of RNA and that dimerization is necessary for strong RNA binding. The dimerization domain of NSP3 was mapped between amino acids 150 and 206 by using the yeast two-hybrid system. The eukaryotic initiation factor 4 GI subunit (eIF-4GI) binding domain of NSP3A has been mapped in the last 107 amino acids of its C terminus by using a pulldown assay and the yeast two-hybrid system. NSP3 is composed of two functional domains separated by a dimerization domain.

Rotavirus RNA-binding protein NSP3 interacts with eIF4GI and evicts the poly(A) binding protein from eIF4F

Most eukaryotic mRNAs contain a 5'cap structure and a 3'poly(A) sequence that synergistically increase the efficiency of translation. Rotavirus mRNAs are capped, but lack poly(A) sequences. During rotavirus infection, the viral protein NSP3A is bound to the viral mRNAs 3' end. We looked for cellular proteins that could interact with NSP3A, using the two-hybrid system in yeast. Screening a CV1 cell cDNA library allowed us to isolate a partial cDNA of the human eukaryotic initiation factor 4GI (eIF4GI). The interaction of NSP3A with eIF4GI was confirmed in rotavirus infected cells by co-immunoprecipitation and in vitro with NSP3A produced in Escherichia coli. In addition, we show that the amount of poly(A) binding protein (PABP) present in eIF4F complexes decreases during rotavirus infection, even though eIF4A and eIF4E remain unaffected. PABP is removed from the eIF4F complex after incubation in vitro with the C-terminal part of NSP3A, but not with its N-terminal part produced in E.coli. These results show that a physical link between the 5' and the 3' ends of mRNA is necessary for the efficient translation of viral mRNAs and strongly support the closed loop model for the initiation of translation. These results also suggest that NSP3A, by taking the place of PABP on eIF4GI, is responsible for the shut-off of cellular protein synthesis.

Amino terminus of reovirus nonstructural protein sigma NS is important for ssRNA binding and nucleoprotein complex formation

Reovirus nonstructural protein sigma NS exhibits a ssRNA-binding activity thought to be involved in assembling the reovirus mRNAs for genome replication and virion morphogenesis. To extend analysis of this activity, recombinant sigma NS (r sigma NS) was expressed in insect cells using a recombinant baculovirus. In infected-cell extracts, r sigma NS was found in large complexes (> or = 30 S) that were disassembled into smaller, 13-19 S complexes upon treatment with RNase A. R sigma NS also bound to poly(A)-Sepharose beads both before and after purification. Treatment with high salt during purification caused r sigma NS to sediment in even smaller, 7-9 S complexes, consistent with more complete loss of RNA. To localize the RNA-binding site, limited proteolysis was used to fragment the r sigma NS protein. Upon mild treatment with thermolysin, 11 amino acids were removed from the amino terminus of r sigma NS, and the resulting protein no longer bound to poly(A). In addition, when r sigma NS in cell extracts was treated with thermolysin to generate the amino-terminally truncated from, it sedimented at 7-9 S, also consistent with the loss of RNA-binding capacity. To confirm these findings, a deletion mutant lacking amino acids 2-11 was constructed and expressed in insect cells from a recombinant baculovirus. The mutant protein in cell extracts showed greatly reduced poly(A)-binding activity and sedimented as 7-9 S complexes. These data suggest that the first 11 amino acids of sigma NS, which are predicted to form an amphipathic alpha-helix, are important for both ssRNA binding and formation of complexes larger than 7-9 S.

Genetics of the rotaviruses

Genetic analyses have contributed significantly to our understanding of the biology of the rotaviruses. The distinguishing feature of the virus is a genome consisting of 11 segments of double-stranded RNA. The segmented nature of the genome allows reassortment of genome segments during mixed infections, which is the major distinguishing feature of rotavirus genetics. Reassortment has been a powerful tool for mapping viral mutations and other determinants of biological phenotypes to specific genome segments. However, more detailed genetic analysis of rotaviruses is currently limited by the inability to perform reverse genetics. Development of a reverse genetic system will facilitate analysis of the molecular mechanisms involved in various genetic, biochemical, and biological phenomena of the virus.

Stable protein-RNA interaction involves the terminal domains of bluetongue virus mRNA, but not the terminally conserved sequences

The interaction of bluetongue virus (BTV) proteins with viral RNA was investigated in vitro by means of a biochemical approach. By subjecting cytoplasmic extracts from virus-infected baby hamster kidney cells and in vitro synthesized radiolabeled RNA to ultraviolet cross-linking assays, we demonstrated that, of all the BTV proteins, NS2 becomes most intimately associated with the labeled viral RNA. Competition binding studies indicated that NS2 has the greatest affinity for the 3' region of the viral transcripts. By analyzing the binding efficiency of NS2 to mutant RNA transcripts which lacked the fully conserved 5'- and/or 3'-terminal hexanucleotides, we have established that these sequences are not necessary for optimal binding. The specificity of the NS2-RNA interaction was investigated by competition experiments with unlabeled BTV-specific homologous and heterologous competitor RNAs as well as with viral double-stranded RNA (dsRNA). Although apparent differences in the ability of NS2 to bind to the different RNA transcripts were observed, it did not bind to the dsRNA.

In vivo and in vitro phosphorylation of rotavirus NSP5 correlates with its localization in viroplasms

NSP5 (NS26), the product of rotavirus gene 11, is a phosphoprotein whose role in the virus replication cycle is unknown. To gain further insight into its function, we obtained monoclonal antibodies against the baculovirus-expressed protein. By immunoprecipitation and immunoblotting experiments, we showed that (i) NSP5 appears in many different phosphorylated forms in rotavirus-infected cells; (ii) immunoprecipitated NSP5 from rotavirus-infected cells can be phosphorylated in vitro by incubation with ATP; (iii) NSP5, produced either by transient transfection of rotavirus gene 11 or by infection by gene 11 recombinant vaccinia virus or baculovirus, can be phosphorylated in vivo and in vitro; (iv) NSP5 expressed in Escherichia coli is phosphorylated in vitro, and thus NSP5 is a potential protein kinase; and (v) NSP5 forms dimers and interacts with NSP2. The intracellular localization of NSP5 in the course of rotavirus infection and after transient expression in COS7 cells has also been investigated. In rotavirus-infected cells, NSP5 is localized in viroplasms, but it is widespread throughout the cytoplasm of transfected COS7 cells. NSP5 produced by transfected COS7 cells did not acquire the multiphosphorylated forms observed in rotavirus-infected COS7 cells. Thus, there is a tight correlation between the localization of NSP5 in the viroplasms and its protein kinase activity in vivo or in vitro. Our results suggest that cellular or viral cofactors are indispensable to fully phosphorylate NSP5 and to reach its intracellular localization.

Serine protein kinase activity associated with rotavirus phosphoprotein NSP5

The rotavirus nonstructural protein NSP5, a product of the smallest genomic RNA segment, is a phosphoprotein containing O-linked N-acetylglucosamine. We investigated the phosphorylation of NSP5 in monkey MA104 cells infected with simian rotavirus SA11. Immunoprecipitated NSP5 was analyzed with respect to phosphorylation and protein kinase activity. After metabolic labeling of NSP5 with 32Pi, only serine residues were phosphorylated. Separation of tryptic peptides revealed four to six strongly labeled products and several weakly labeled products. Phosphorylation at multiple sites was also shown by two-dimensional polyacrylamide gel electrophoresis (PAGE), where several isoforms of NSP5 with different pIs were identified. Analysis by PAGE of protein reacting with an NSP5-specific antiserum showed major forms at 26 to 28 and 35 kDa. Moreover, there were polypeptides migrating between 28 and 35 kDa. Treatment of the immunoprecipitated material with protein phosphatase 2A shifted the mobilities of the 28- to 35-kDa polypeptides to the 26-kDa position, suggesting that the slower electrophoretic mobility was caused by phosphorylation. Radioactive labeling showed that the 26-kDa form contained additional phosphate groups that were not removed by protein phosphatase 2A. The immunoprecipitated NSP5 possessed protein kinase activity. Incubation with [gamma-32P]ATP resulted in 32P labeling of 28- to 35-kDa NSP5. The distribution of 32P radioactivity between the components of the complex was similar to the phosphorylation in vivo. Assays of the protein kinase activity of a glutathione S-transferase-NSP5 fusion polypeptide expressed in Escherichia coli demonstrated autophosphorylation, suggesting that NSP5 was the active component in the material isolated from infected cells.

The 3'-terminal consensus sequence of rotavirus mRNA is the minimal promoter of negative-strand RNA synthesis

We used an in vitro template-dependent replicase assay (D. Chen, C. Zeng, M. Wentz, M. Gorziglia, M. Estes, and R. Ramig. J. Virol. 68:7030-7039, 1994) to identify the cis-acting signals required for replication of a genome segment 9 template from the group A rotavirus strain OSU. The replicase phenotypes for a panel of templates with internal deletions or 3'-terminal truncations indicated that no essential replication signals were present within the open reading frame and that key elements were present in the 5' and 3' noncoding regions. Chimeric constructs containing portions of viral sequence ligated to a nonviral backbone were generated to further map the regions required for in vitro replication of segment 9. The data from these constructs showed that the 3'-terminal seven nucleotides of the segment 9 mRNA provided the minimum requirement for replication (minimal promoter). Analysis of additional chimeric templates demonstrated that sequences capable of enhancing replication from the minimal promoter were located immediately upstream of the minimal promoter and at the extreme 5' terminus of the template. Mutational analysis of the minimal promoter revealed that the 3'-terminal -CC residues are required for efficient replication. Comparison of the replication levels for templates with guanosines and uridines at nucleotides -4 to -6 from the 3' terminus compared with levels for templates containing neither of these residues at these positions indicated that either or both residues must be present in this region for efficient replication in vitro.

Rotavirus VP1 alone specifically binds to the 3' end of viral mRNA, but the interaction is not sufficient to initiate minus-strand synthesis

Recent studies have shown that disrupted (open) rotavirus cores have an associated replicase activity which supports the synthesis of dsRNA from viral mRNA in a cell-free system (D. Chen, C. Q.-Y. Zeng, M. J. Wentz, M. Gorziglia, M. K. Estes, and R. F. Ramig, J. Virol. 68:7030-7039, 1994). To determine which of the core proteins, VP1, VP2, or VP3, recognizes the template mRNA during RNA replication, SA11 open cores were incubated with 32P-labeled RNA probes of viral and nonviral origin and the reaction mixtures were analyzed for the formation of RNA-protein complexes by gel mobility shift assay. In mixtures containing a probe representing the 3' end of SA11 gene 8 mRNA, two closely migrating RNA-protein complexes, designated s and f, were detected. The interaction between the RNA and protein of the s and f complexes was shown to be specific by competitive binding assay with tRNA and brome mosaic virus RNA. By electrophoretic analysis of RNA-protein complexes recovered from gels, VP1 was shown to be the only viral protein component of the complexes, thereby indicating that VP1 specifically recognizes the 3' end of gene 8 mRNA. Analysis of VP1 purified from open cores by glycerol gradient centrifugation verified that VP1 recognizes the 3' end of viral mRNA but also showed that in the absence of other viral proteins, VP1 lacks replicase activity. When reconstituted with VP2-rich portions of the gradient, VP1 stimulated levels of replicase activity severalfold. These data indicate that VP1 can bind to viral mRNA in the absence of any other viral proteins and suggest that VP2 must interact with the RNA-protein complex before VP1 gains replicase activity.

cis-Acting signals that promote genome replication in rotavirus mRNA

A previous study has shown that rotavirus cores have an associated replicase activity which can direct the synthesis of double-stranded RNA from viral mRNA in a cell-free system (D. Y. Chen, C. Q.-Y. Zeng, M. J. Wentz, M. Gorziglia, M. K. Estes, and R. F. Ramig, J. Virol. 68:7030-7039, 1994). To define the cis-acting signals in rotavirus mRNA that are important for RNA replication, gene 8 transcripts which contained internal and terminal deletions and chimeric transcripts which linked gene 8-specific 3'-terminal sequences to the ends of nonviral sequences were generated. Analysis of these RNAs in the cell-free system led to the identification of a cis-acting signal in the gene 8 mRNA which is essential for RNA replication and two cis-acting signals which, while not essential for replication, serve to enhance the process. The sequence of the essential replication signal is located at the extreme 3' end of the gene 8 mRNA and, because of its highly conserved nature, is probably a common feature of all 11 viral mRNAs. By site-specific mutagenesis of the gene 8 mRNA, residues at positions -1, -2, -5, -6, and -7 of the 3' essential signal were found to be particularly important for promoting RNA replication. One of the cis-acting signals shown to enhance the replication in the cell-free system was located near the 5' end of the 3' untranslated region (UTR) of the gene 8 mRNA, while remarkably the other was located in the 5' UTR of the message. The existence of an enhancement signal in the 5' UTR raises the possibility that the 5' and 3' ends of the rotavirus mRNA may interact with each other and/or with the viral replicase during genome replication.

Nondefective rotavirus mutants with an NSP1 gene which has a deletion of 500 nucleotides, including a cysteine-rich zinc finger motif-encoding region (nucleotides 156 to 248), or which has a nonsense codon at nucleotides 153-155

We isolated two nondefective bovine rotavirus mutants (A5-10 and A5-16 clones) which have nonsense mutations in the early portion of the open reading frame of the NSP1 gene. In the NSP1 gene (1,587 bases long) of A5-10, a nonsense codon is present at nucleotides 153 to 155 just upstream of the coding region (nucleotides 156 to 230) of a cysteine-rich Zn finger motif. A5-16 gene 5 (1,087 bases long) was found to have a large deletion of 500 bases corresponding to nucleotides 142 to 641 of a parent A5-10 NSP1 gene and to have a nonsense codon at nucleotides 183 to 185, which resulted from the deletion. Expression of gene 5-specific NSP1 could not be detected in MA-104 cells infected with the A5-10 or A5-16 clone or in an in vitro translation system using the plasmids with gene 5 cDNA from A5-10 or A5-16. Nevertheless, both A5-10 and A5-16 replicated well in cultured cells, although the plaque size of A5-16 was extremely small.

Recovery and characterization of a replicase complex in rotavirus-infected cells by using a monoclonal antibody against NSP2

Replication of the rotavirus genome involves two steps: (i) transcription and extrusion of transcripts and (ii) minus-strand RNA synthesis in viral complexes containing plus-strand RNA. In this study, we showed evidence for the importance of the viral nonstructural protein of rotavirus, NSP2, in the replication of viral RNAs. RNA-binding properties of NSP2 were tested by UV cross-linking in vivo (in rotavirus-infected MA104 cells and recombinant baculovirus-expressing NSP2-infected Sf9 cells). In rotavirus-infected cells, NSP2 is bound to the 11 double-stranded RNA genomic segments of rotavirus. Quantitative analysis (using hydrolysis by RNase A) is consistent with NSP2 being directly bound to partially replicated viral RNA. Using various monoclonal antibodies and specific antisera against the structural (VP1, VP2, and VP6) and nonstructural (NSP1, NSP2, NSP3, and NSP5) proteins, we developed a solid-phase assay for the viral replicase. In this test, we recovered a viral RNA-protein complex with replicase activity only with a monoclonal antibody directed against NSP2. Our results indicated that these viral complexes contain the structural proteins VP1, VP2, and VP6 and the nonstructural protein NSP2. Our results show that NSP2 is closely associated in vivo with the viral replicase.

Structure and function of rotavirus nonstructural protein NSP3

The genomes of viruses in the family Reoviridae consist of segmented double-stranded RNA. There are 10 to 12 segments depending on the genus. The 5' ends and the 3' ends of the RNAs present conserved motifs for each virus genus. These conserved motifs have been hypothesized to play a role in genomic segment assortment during virus morphogenesis. Using a set of monoclonal antibodies we have tried to identify rotaviral proteins that bind to RNA during infection in cell culture. This methodology takes advantage of being able to label RNA in vitro to high specific activity and also of solid phase processing of RNA-protein complexes. After cross-linking the RNA to protein in infected cells, protein-RNA complexes are precipitated with a specific MAb; then, the RNA in the complex is labeled in vitro and the protein or nucleic acid moieties are analyzed by usual protocols. This paper describes results using an anti NSP3 MAb. In infected cells, we have shown that NSP3 binds to the eleven messenger RNAs, and that a sequence from nucleotides 8 to 15 is protected from digestion with RNAse T1 by NSP3 in the RNA-protein complex. The availability of recombinant protein NSP3 expressed in the baculovirus-insect cell system has allowed the sequence specificity of NSP3 to be studied in vitro. The minimal sequence recognized by NSP3 is GACC. The role of NSP3 in rotavirus replication is discussed based on these results and by comparison with other RNA-binding proteins of members of the Reoviridae family.

Template-dependent, in vitro replication of rotavirus RNA

A template-dependent, in vitro rotavirus RNA replication system was established. The system initiated and synthesized full-length double-stranded RNAs on rotavirus positive-sense template RNAs. Native rotavirus mRNAs or in vitro transcripts, with bona fide 3' and 5' termini, derived from rotavirus cDNAs functioned as templates. Replicase activity was associated with a subviral particle containing VP1, VP2, and VP3 and was derived from native virions or baculovirus coexpression of rotavirus genes. A cis-acting signal involved in replication was localized within the 26 3'-terminal nucleotides of a reporter template RNA. Various biochemical and biophysical parameters affecting the efficiency of replication were examined to optimize the replication system. A replication system capable of in vitro initiation has not been previously described for Reoviridae.

Products of the porcine group C rotavirus NSP3 gene bind specifically to double-stranded RNA and inhibit activation of the interferon-induced protein kinase PKR

The porcine group C rotavirus (Cowden strain) NSP3 protein (the group C equivalent of the group A gene 7 product, formerly called NS34) shares homology with known double-stranded RNA-binding proteins, such as the interferon-induced, double-stranded RNA-dependent protein kinase PKR. A clone of NSP3, expressed both in vitro and in COS-1 cells, led to the synthesis of minor amounts of a product with an M(r) of 45,000 (the expected full-length M(r) of NSP3) and major amounts of products with M(r)s of 38,000 and 8,000. Restriction enzyme digestion analysis prior to expression in vitro and amino-terminal sequence analysis suggest that the products with M(r)s of 38,000 and 8,000 are cleavage products of the protein with an M(r) of 45,000. The full-length protein and the product with an M(r) of 8,000, both of which contain the motif present in double-stranded RNA-binding proteins, bound specifically to double-stranded RNA. The products with M(r)s of 45,000 and 8,000 were also detected in Cowden strain-infected MA104 cells. NSP3 products expressed in COS-1 cells were capable of inhibiting activation of the double-stranded RNA-dependent protein kinase similar to other double-stranded RNA-binding proteins, and NSP3 products expressed in HeLa cells were capable of rescuing the replication of an interferon-sensitive deletion mutant of vaccinia virus.

Deletion mapping of the rotavirus metalloprotein NS53 (NSP1): the conserved cysteine-rich region is essential for virus-specific RNA binding

NS53 (NSP1), the gene 5 product of the group A rotaviruses, is a minor nonstructural protein of 486 to 495 amino acids which binds zinc and contains an amino-terminal highly conserved cysteine-rich region that may form one or two zinc fingers. To study the structure-function of the gene 5 product, wild-type and mutant forms of NS53 were produced by using a recombinant baculovirus expression system and a recombinant vaccinia virus/T7 (vTF7-3) expression system. Analysis of the RNA-binding activity of the wild-type NS53 immobilized onto protein A-Sepharose beads with NS53-specific antiserum showed that the protein exhibited specific affinity for all 11 rotavirus mRNAs. The use of short virus-specific RNA probes indicated that NS53 specifically recognizes an element located near the 5' ends of viral mRNAs. Analysis of the RNA-binding activity of deletion mutants of NS53 showed that the RNA-binding domain resides within the first 81 amino acids of the protein and that the highly conserved cysteine-rich region within this region of the protein is essential for the activity. Gel electrophoresis and Western immunoblot analyses of intracellular fractions derived from infected cells revealed that large amounts of NS53 were present in the cytosol and in association with the cytoskeletal matrix. Indirect immunofluorescence analysis of cells programmed to transiently express mutant forms of NS53 using vTF7-3 indicated that the intracellular localization domain resides between amino acids 84 and 176 of NS53. Together, these data show that the RNA-binding domain and the intracellular localization domain lie upstream from the region of NS53 previously determined not to be essential for replication of rotaviruses in cell culture (J. Hua and J. T. Patton, Virology 198:567-576, 1994).

Expression of two bovine rotavirus non-structural proteins (NSP2, NSP3) in the baculovirus system and production of monoclonal antibodies directed against the expressed proteins

Studies on rotavirus non-structural proteins have been hampered in the past by difficulties in obtaining monospecific reagents. To make such reagents available, we have expressed in the baculovirus system NSP2 and NSP3 (formerly called NS35 and NS34, respectively) of the bovine rotavirus RF and produced hybridomas against these proteins. Full-length DNA copies of RNA segments 7 (coding for NSP3) and 8 (coding for NSP2) of the virus strain RF were cloned and sequenced. Each cDNA was inserted in the transfer vector pVL941 and used to transfect Spodoptera frugiperda cells (Sf9). Recombinant baculoviruses encoding these proteins were obtained. Infection of Sf9 cells with these recombinant viruses resulted in a high level of expression of NSP2 and NSP3 (range of 1 microgram per 10(6) cells). Monoclonal antibodies (MAbs) were elicited by immunization of BALB/c mice with adjuvented, unpurified recombinant proteins in the rear foot pads. Fusion was performed using lymphocytes from popliteal lymph nodes with SP2/O-Ag14 myeloma line. Screening was by differential indirect immunofluorescent staining on monolayers of Sf9 cells infected with each recombinant virus. Two MAbs proved to be reactive against NSP3 and a single one against NSP2. They showed high specificity by immunofluorescence, immunoprecipitation and Western blot. The isotype of these MAbs was IgG1. Oligomeric forms of NSP3 and NSP2 proteins were detected and the existence of intra-chain disulfide bridge in NSP2 protein was suggested. The levels of synthesis and cellular localization of NSP3 and NSP2 proteins were different as shown by immunoprecipitation and immunofluorescence.