Identification of an RNA Binding Region within ... - Journal of Virology

JOURNAL OF VIROLOGY, June 1995, p. 3799–3806 0022-538X/95/$04.0010 Copyright q 1995, American Society for Microbiology

Vol. 69, No. 6

Identification of an RNA Binding Region within the N-Terminal Third of the Influenza A Virus Nucleoprotein CARMEN ALBO,1 ALFONSO VALENCIA,2

AND

AGUSTIŃ PORTELA1*

Centro Nacional de Microbiologıá, Virologıá e Inmunologıá Sanitarias, Instituto de Salud Carlos III, Majadahonda 28220,1 and Centro Nacional de Biotecnologıá, Universidad Auto ńoma, Cantoblanco 28049,2 Madrid, Spain Received 28 November 1994/Accepted 27 February 1995

The influenza A virus nucleoprotein (NP) has been examined with regard to its RNA-binding characteristics. NP, purified from virions and devoid of RNA, bound synthetic RNAs in vitro and interacted with the ribonucleotide homopolymers poly(A), poly(G), poly(U), and poly(C) in a salt-dependent manner, showing higher binding affinity for polypyrimidine homopolymers. To map the NP regions involved in RNA binding, a series of deleted forms of the NP were prepared, and these truncated polypeptides were tested for their ability to bind poly(U) and poly(C) homopolymers linked to agarose beads. Proteins containing deletions at the N terminus of the NP molecule showed reduced RNA-binding activity, indicating that this part of the protein was required to bind RNA. To identify the NP region or regions which directly interact with RNA, proteins having the maltose-binding protein fused with various NP fragments were obtained and tested for binding to radioactively labeled RNAs in three different assays: (i) nitrocellulose filter binding assays, (ii) gel shift assays, and (iii) UV light-induced cross-linking experiments. A maltose-binding protein fusion containing the N-terminal 180 amino acids of NP behaved as an RNA-binding protein in the three assays, demonstrating that the N terminus of NP can directly interact with RNA. This NP region could be further subdivided into two smaller regions (amino acids 1 to 77 and 79 to 180) that also retained RNA-binding activity. from a recombinant cDNA, displays DNA- and RNA-binding activity in vitro (3, 26, 59). It has been shown that NP can bind any RNA longer than 15 nucleotides (59) and that the protein does not show higher binding affinity for RNA molecules containing viral sequences (3, 26, 59). This situation contrasts with that found in other viral proteins, such as, for example, the nucleocapsid protein of vesicular stomatitis virus (6, 38), the capsid protein of Sindbis virus (57), and the coat protein of tobacco mosaic virus (5), which selectively encapsidates viral RNAs over other nonviral transcripts in vitro. Despite the fact that NP binds with similar affinities to viral and nonviral RNA molecules, purified NP freed of RNA, obtained either from virions (20, 33, 41, 49, 60) or from influenza virus-infected cells (36), can be complexed with synthetic viruslike RNAs and can reconstitute RNP complexes, making them fully competent to transcribe the encapsidated RNA in vitro and to express it in vivo. In addition to its structural role, genetic and biochemical evidence indicates that the NP is involved in the process of virus-specific RNA synthesis. Thus, temperature-sensitive mutants in the NP gene show defects in RNA synthesis (reviewed in reference 34), antibodies to the NP inhibit the transcriptase activity associated with disrupted virions (2, 56), and nucleocapsids in which most of the NP has been removed are unable to synthesize template-sized RNA transcripts (20). Furthermore, free NP molecules not associated with nucleocapsids are required for antitermination during viral mRNA synthesis and for elongation of virion RNA molecules (4, 50). The 498-amino-acid influenza A virus NP is highly conserved among influenza A viruses (19, 48, 51). The protein is rich in arginine (9.6%), glycine (8.2%), and serine (8.2%) residues and has a net positive charge at neutral pH (58). However, there are no major clusters of predominantly basic residues which could presumably bind the RNA genome (58). In addition, the protein does not contain any of the canonical consensus RNA binding motifs (RGG [24], KH [1, 53], arginine

Influenza A viruses contain a segmented genome made up of eight negative-sense single-stranded RNA molecules. In the virus particle, the genome is found in the form of ribonucleoprotein (RNP) complexes, in which the virion RNA is associated with four virus-encoded polypeptides, the P proteins (PB1, PB2, and PA), which constitute the viral polymerase, and the nucleoprotein (NP). It has been shown that these four proteins are absolutely required for expression of influenza virus-like RNAs in vivo (11, 21, 25, 37) and that the RNP complexes constitute the functional templates for replication and transcription of the viral genome (reviewed in reference 30). In the virions, the P proteins are present in catalytic amounts and appear to be located at or very close to the end of each RNP complex, as shown by immunoelectron microscopy studies (39). In agreement with these results, it has recently been reported that the P proteins bind specifically to the 59 and 39 conserved sequences found in all eight viral RNA segments (16, 55). The NP is a major structural component of the viral particle and encapsidates the viral RNA (14, 27, 29, 43). It has been calculated, on the basis of the stoichiometry of NP in RNPs, that one NP interacts with 20 nucleotides of virion RNA (10). It has been suggested that the RNA is coiled around the outside of the RNP complex, because this structure is maintained when the RNA is removed by treatment with RNases or is displaced from the nucleocapsids by polyvinyl sulfate (14, 26, 27, 43). In addition, a peculiar feature that distinguishes the influenza virus RNP complex from the nucleocapsids of other negative-sense single-stranded RNA viruses is the inability to band the influenza virus RNPs in CsCl gradients unless the NP had been previously cross-linked to the RNA (26, 29). The influenza virus NP, isolated from virions or obtained

* Corresponding author. Phone: (1) 6380011, ext. 252. Fax: (1) 6391859. 3799

3800

ALBO ET AL.

rich [31], and RRM [23]) found in other RNA-binding proteins (reviewed in reference 8). To better understand the functions of the NP, both as a structural component and as a functional component of the RNP complexes, we decided to analyze its RNA-binding properties and to map the protein regions involved in RNA binding. MATERIALS AND METHODS Plasmids. Plasmid pGEM-NP (37) contains a full-length cDNA encoding the NP gene of influenza A/Victoria/3/75 virus (12), cloned in the polylinker of pGEM-3 vector downstream from the T7 RNA polymerase promoter. The nucleotide sequence of the cloned gene was determined by the dideoxy method with T7 DNA polymerase (Sequenase) and NP-specific primers. The deduced amino acid sequence was identical to that reported for the A/Beijing/39/75 (H3N2) strain (51), except for one amino acid substitution (D to N) at residue 34. Plasmids derived from pGEM-NP, encoding deletion mutants of NP, were obtained by standard techniques (46) by taking advantage of restriction sites present in the cDNA. Plasmids containing the maltose-binding protein (MBP) fused to various fragments of the NP were obtained by subcloning various NP cDNA fragments in the polylinker of plasmid pMAL-c (which encodes a malE-lacZa gene fusion) (New England Biolabs) downstream from the malE gene. All plasmids derived from pGEM-NP and pMAL-c were characterized by restriction analysis and sequenced across the fusion junctions. Plasmid pPB2CAT9 (a gift from M. Krystal), which mirrors plasmid pIVACAT1 (33), contains the coding sequence of the chloramphenicol acetyltransferase (CAT) gene in negative polarity flanked by the 59 and 39 nontranslated sequences of the influenza virus segment encoding PB2. Plasmid pBSKpolyA (a gift from F. Gebauer) contains a fragment of 73 A residues cloned in the polylinker of pBluescript II SK2 (Stratagene). These two plasmids were used to synthesize the 32P-labeled RNAs used in the RNA binding assays. Preparation of RNA probes. Plasmids pPB2CAT9 and pBSKpolyA were digested with the appropriate restriction enzyme and transcribed in vitro with T7 RNA polymerase in the presence of [a-32P]CTP. After the transcription reaction, plasmid DNA was digested with DNase I and the RNA transcripts were purified by chromatography on Sephadex G50-80 columns. Plasmid pPB2CAT9 was digested with either HgaI or RsaI to yield a full-length influenza virus-like CAT RNA or a short transcript (71 nucleotides in length), respectively. Plasmid pBSKpolyA was digested with XbaI and transcribed with T7 RNA polymerase to yield a 120-nucleotide-long RNA containing 73 U residues. Purification of NP from virions. Influenza A/Victoria/3/75 virus was grown in embryonated eggs, and viral cores were prepared as described previously (2). RNP complexes were then centrifuged through a CsCl-glycerol step gradient to dissociate the complexes into their RNA and protein components (20, 36). The fractions of the gradient enriched in NP were harvested, dialyzed (as described in reference 41), and stored at 2808C. In vitro transcription and translation. The pGEM-NP-derived plasmids were linearized with XbaI restriction enzyme and transcribed in vitro with T7 RNA polymerase (Promega), and the resultant RNAs were translated in a rabbit reticulocyte lysate (Boehringer Mannheim) in the presence of [35S]methionine (Amersham) according to the instructions recommended by the manufacturer. Proteins dS and dH were obtained from RNA transcripts derived from plasmid pGEM-NP digested with the HindIII and ScaI restriction enzymes, respectively. Ribonucleotide homopolymer binding assays. Experiments were essentially done as described previously (9, 54). Briefly, after in vitro translation, an aliquot of the rabbit reticulocyte lysate (3 ml) was diluted to 16 ml in binding buffer (10 mM Tris-HCl [pH 7.5], 100 mM NaCl, 2.5 mM MgCl2) and supplemented with 1 mM CaCl2 and 2 U of micrococcal nuclease (Boehringer Mannheim). After incubation for 10 min at 308C, the mixture was then supplemented with EGTA (up to 5 mM) to inactivate the nuclease and diluted with 150 ml of buffer A (binding buffer containing 0.6% Triton X-100 and the indicated concentration of NaCl [ranging from 0.1 to 2 M]). The samples were then incubated with 40 ml of the ribonucleotide homopolymers bound to agarose beads (Sigma [0.6 mg of polyribonucleotide per ml of agarose]). After incubation for 10 min on a rocking platform at 48C, the agarose beads were pelleted with a brief spin (15 s) in a microcentrifuge, resuspended in buffer A, and pelleted through a 100-ml cushion containing buffer A and 50% glycerol. The agarose beads were then washed twice more in buffer A, dried, and boiled in sodium dodecyl sulfate (SDS)-sample buffer. The eluted proteins were resolved by SDS-polyacrylamide gel electrophoresis (SDS-PAGE) and visualized by fluorography. The percentage of protein bound to the agarose beads was determined by densitometric analyses of the fluorograms. The experiments to determine the binding affinities of purified NP to ribonucleotide homopolymers were carried out as described above, except that the nuclease and EGTA steps were omitted. Upon elution of the bound proteins in SDS-sample buffer, they were resolved by SDS-PAGE, transferred to nitrocellulose paper, and developed with an anti-NP-specific antiserum. Expression of the MBP-NP fusions. All pMAL-c-derived plasmids were transferred to Escherichia coli XL1-Blue cells (7) for expression of the corresponding

J. VIROL. MBP fusion. The conditions for induction with IPTG (isopropyl-b-D-thiogalactopyranoside), preparation of the cell extracts, and binding of the fusion proteins to amylose resin columns were those recommended by the manufacturer (New England Biolabs). Proteins bound to the amylose resin column were successively washed with 10 column volumes of buffer B (10 mM Tris-HCl [pH 7.5], 1 mM dithiothreitol, 2 M NaCl) and buffer C (10 mM Tris-HCl [pH 7.5]), 1 mM dithiothreitol, 100 mM NaCl). The fusion proteins were then eluted from the column in buffer C containing 10 mM maltose, adjusted to 20% glycerol, and stored at 2208C. Nitrocellulose filter binding assays. Aliquots of the MBP fusions (100 to 500 ng) were mixed with 1 to 3 ng of 32P-labeled RNAs (20,000 cpm) in binding buffer (final reaction volume of 50 ml). After 10 min of incubation on ice, samples were filtered through nitrocellulose paper (Millipore) (0.22-mm-pore size), which had been previously soaked in binding buffer, and washed with 10 ml of the same buffer. Radioactivity bound to the filter was measured by Cerenkov counting, and the efficiency of binding was determined as a percentage of the radioactivity included in the experimental samples. Gel shift assays. Mixtures of RNA and MPB fusions were prepared and incubated as described in the previous section. Upon incubation, the samples were supplemented with glycerol (up to a final concentration of 10%) and electrophoresed in 0.3% (wt/vol) agarose gels containing 0.53 TAE buffer (40 mM Tris base, 19 mM glacial acetic acid, 1 mM EDTA). The agarose gel was soaked in a 5% trichloroacetic acid solution for 30 min, washed several times with water, dried, and exposed to X-ray films. UV light-induced cross-linking assays. Mixtures of RNA and MPB fusions were prepared as described above and incubated at 228C. After incubation, the samples were irradiated for 10 min with a UV lamp (250 nm) placed 4 to 5 cm above the samples and then were treated with 3 mg of RNase A (to remove all but the few nucleotides of the RNA probe which were directly linked to the protein) for 20 min at 308C. Samples were then boiled in SDS-sample buffer, subjected to SDS-PAGE, and analyzed by autoradiography. Sequence analyses. Sequence searches were carried out with GENEQUIZ (47). This tool facilitates searches in the most updated sequence databases and includes algorithms for sequence searches and multiple-sequence alignment. Secondary structure was predicted with the PHD program implemented at the EMBL file server (45). Database searches with specific sequence patterns were carried out with SCRUTINEER (52).

RESULTS AND DISCUSSION RNA-binding characteristics of NP. In order to determine the NP region(s) required for RNA binding, it was important to know some specific features of the NP-RNA interaction. To this end, NP devoid of RNA was purified by centrifugation of RNP cores through CsCl-glycerol gradients (Fig. 1A) and its RNA-binding characteristics were analyzed in different assays. The purified protein retained RNA-binding activity in nitrocellulose filter binding assays (35), but as previously reported by others (3, 26, 59), we failed to detect specific binding of the protein to RNA molecules containing viral sequences. It was then decided to investigate the binding of the NP to the ribonucleotide homopolymers poly(G), poly(A), poly(C), and poly(U), immobilized on agarose beads, at various salt concentrations. This kind of approach has been used to classify (54) and characterize the RNA-binding properties of heterogeneous nuclear RNPs (9, 24) and other RNA-binding proteins (15, 53, 61). As shown in Fig. 1B, at low salt concentrations (0.1 M NaCl), NP bound all four RNA homopolymers, and this interaction was still maintained at 0.5 M NaCl, indicating the high affinity of the protein for RNA. It was necessary to increase the salt concentration up to 1 M to detect significant differences in binding specificity; at these high salt concentrations, the NP bound only to poly(C) and poly(U) homopolymers. It should be noted that the results shown in Fig. 1B actually reflect the binding of NP to the agarose-bound homopolymers, because NP was not detected when the agarose beads were not included in the reaction mixture (Fig. 1B, lane b). It has been suggested that the NP-RNA interaction involves electrostatic interactions of basic residues of NP with the negatively charged phosphate backbone of the RNA. However, the fact that NP displays a binding preference towards pyrimidine ribonucleotide homopolymers suggests that there are

VOL. 69, 1995

RNA-BINDING CHARACTERISTICS OF INFLUENZA A VIRUS NP

FIG. 1. Properties of the NP-RNA interaction. (A) Purification of the NP polypeptide. RNP complexes purified from influenza A/Victoria/3/75 virus virions were centrifuged through a CsCl-glycerol step gradient. The fractions of the gradient enriched in NP were pooled, dialyzed, and analyzed by SDS-PAGE and Coomassie blue staining. Lane 1, molecular weight markers (positions indicated in thousands); lane 2, purified virions; lane 3, purified NP. The positions of influenza virus NP and M1 polypeptides are indicated on the right. (B) Binding of purified NP to ribonucleotide homopolymers. Purified NP was incubated with agarose-bound polyribonucleotides poly(G), poly(A), poly(C), and poly(U), in a buffer containing the indicated NaCl concentrations (ranging from 0.1 to 2 M). After washing (see Materials and Methods for details), bound proteins were resolved by SDS-PAGE, blotted to nitrocellulose sheets, and probed with an anti-NP serum. For lane a, an amount of NP equivalent to the material used for each binding reaction mixture was loaded as a marker in the polyacrylamide gel. For lane b, the binding reaction was carried out in the absence of the agarose beads.

also contacts between amino acid residues and nucleotide bases. Although the relationship of this NP binding preference with binding of NP to a virus-specific RNA sequence is not straightforward, it is worth mentioning that the influenza virus RNA segments, unlike those of other RNA viruses, are rich in uridine residues (44) and that the 39 conserved ends of virion (viral RNA) [59 . . . CCUGCUUU(U or C)GCUOH39] and template (cRNA) (59 . . . CCUUGUUUCUACUOH39) RNA molecules are rich in pyrimidine residues (13). Binding of translated NP and truncated NPs to poly(U) and poly(C). To identify the domain or domains of the NP that confer RNA binding, it was decided to evaluate the RNAbinding activity of in vitro-translated NP fragments to poly(U) and poly(C) homopolymers immobilized on agarose beads. It should be mentioned that a similar experimental approach has been utilized to map the protein regions relevant for RNA binding in many RNA-binding proteins, which bind RNA homopolymers in vitro but whose specific RNA targets are presently unknown (9, 15, 53, 61). A set of plasmids encoding internal in-frame and carboxyterminal deletions in the NP were prepared (Fig. 2A). Each plasmid was then transcribed and translated in vitro (Fig. 2B), and the individual protein products were assayed for binding to poly(U) and poly(C) at various salt concentrations (Fig. 2). Under the assay conditions used, ;50 and ;25% of the intact NP bound to poly(U) at NaCl concentrations of 0.1 and 0.2 M, respectively. This binding efficiency (percentage of protein bound to total input protein) is similar to that described for other RNA-binding proteins assayed in similar tests (9, 53, 61). In addition, b-globin protein (translated in vitro) did not bind

3801

poly(U) at any of the salt concentrations tested (data not shown), and none of the NP-derived translated polypeptides were detected when the agarose beads were not included in the incubation mixture (data not shown). As shown in Fig. 2C, D, and E, mutants lacking carboxy-terminal amino acids (d1, dS, dH, and d10) bound poly(U), suggesting that the C terminus of NP is not required for RNA binding. This suggestion was further supported by the fact that mutant d12, which includes amino acids 355 to 498 of NP, did not bind the RNA homopolymer. In addition, residues 256 to 339 do not seem to be the major determinants of RNA binding, because mutants d3 and d8, which lack these sequences, efficiently bound the homopolymer. However, removal of the N-terminal segment of NP strongly reduced the RNA-binding activity of the mutated proteins. Thus, the binding activity of mutants d6 (lacking residues 78 to 161) and d7 (lacking residues 161 to 254) to poly(U) was reduced at 0.1 M NaCl (Fig. 2C), and it was practically abolished when these proteins were tested under more stringent conditions (0.2 M NaCl) (Fig. 2D). In fact, these two mutants, but not proteins d1, d3, and d8, showed reduced binding to poly(C) at 0.1 M NaCl (Fig. 2E). Another N-terminally truncated mutant, protein d11 (lacking residues 5 to 78), showed reduced RNA-binding activity when tested at 0.2 M NaCl (Fig. 2D), although it bound poly(U) at 0.1 M NaCl (Fig. 2C). Altogether, these results indicate that amino acids 1 to 254 of NP are necessary for RNA-binding activity, whereas the carboxy-terminal part of the protein seems to be dispensable for this function. The apparently contradictory results obtained with proteins d7 and d8 deserve further comment. Protein d8 lacks amino acids 161 to 339 and binds RNA. It can therefore be interpreted that the deleted region is not required for RNA binding. If this were the case, mutant d7, lacking residues 161 to 254, should bind RNA, but it did not. These results indicated that deletions in the region extending from amino acids 161 to 254 affected the structure and/or the RNA binding domain of NP, although that effect could be compensated for by more extensive deletions. (Further comments are provided below.) In this context, it should be noted that retention of RNAbinding activity by a mutated protein implies that a functional RNA binding domain has been exposed. However, a negative result could be due to the lack of such a RNA binding domain in the mutated protein or to being masked because of disruption of the integrity of the protein. Expression and purification of MBP fusions containing NP fragments. Although the results presented above stressed the importance of the N terminus of NP for efficient binding to RNA, it could not be concluded that this protein region contains an RNA binding determinant (i.e., is sufficient for RNAbinding activity). To determine the NP regions that directly interact with RNA, it was necessary to show that NP fragments fused to a non-nucleic acid-binding protein converted the resulting protein into an RNA-binding protein. For this purpose, plasmids containing the coding sequence of the MBP fused at the C terminus to various NP coding sequences were prepared, and the corresponding fusion polypeptides were purified by affinity chromatography on amylose resin columns. In preliminary experiments, it was observed that the protein containing the intact NP fused to the MBP (MBP-NP), purified from the amylose resin, was associated with bacterial RNA (data not shown). This fact could hamper interpretation of the RNA binding experiments; however, it was found that washing the fusion protein, while still bound to the amylose resin, in a buffer containing 2 M NaCl rendered an MBP-NP fusion practically devoid of RNA (not shown). Thus, all MBP fusions were

3802

ALBO ET AL.

J. VIROL.

FIG. 2. Binding of in vitro-translated NP and various truncated forms of NP to poly(U) and poly(C) immobilized on agarose beads. (A) Diagrams of the proteins used in the binding assays. For proteins containing internal in-frame deletions, the numbers in parentheses indicate deleted amino acids. Protein d1 contains an additional 16 amino acids at the C terminus, which were derived from the cloning procedure. The RNA-binding activities of the various proteins are summarized as 1 for those proteins that bound like the intact NP in the assays shown in panels C through E, 1/2, for those that showed reduced binding in any of the assays, or 2 for those that did not bind the homopolymers at all in any of the assays. (B) Plasmids that encode for the proteins shown in panel A were transcribed and translated in vitro with a rabbit reticulocyte lysate. A sample of the translation reaction mixture was analyzed by SDS-PAGE and fluorography. (C, D, and E) in vitro-translated proteins were incubated with either poly(U) (C and D) or poly(C) (E) bound to agarose beads in a buffer containing either 0.1 M (C and E) or 0.2 M (D) NaCl as described in Materials and Methods. Bound proteins were then eluted from the agarose beads, resolved by SDS-PAGE, and analyzed as described for panel B. Panels B, C, and D were exposed for the same period of time, and the amount of protein shown in panel B was 50% of the amount used in the binding reaction mixtures.

purified through the same purification scheme, including the 2 M NaCl wash. The purity and size of the purified proteins were analyzed by SDS-PAGE. As shown in Fig. 3, the relative mobilities of the proteins were consistent with their predicted sizes and the protein samples contained no detectable amounts of contaminant bacterial proteins. It should be noted, however, that in some of the samples (MBP-NP, F1, F4, and F5), some other protein bands with electrophoretic mobilities higher than those of the expected fusion polypeptide were observed. Some of these bands had the same mobility in the four lanes, and they were recognized by an anti-MBP serum in Western immunoblotting experiments (data not shown). This result indicated, therefore, that these faster-moving bands were MBP fusions containing short NP fragments that, most likely, appeared as a consequence of the action of bacterial proteases on specific NP sequences. RNA-binding properties of MBP fusions. The fusion proteins were then tested for their ability to bind RNA in three complementary assays: nitrocellulose filter binding, gel retardation, and UV light-induced cross-linking. The first assay is the less-stringent assay because it only requires the RNA-NP interaction to be stable in binding buffer; the second assay requires, in addition, that the RNA-NP complexes be main-

tained under electrophoresis conditions. The third assay is very stringent; UV light is a zero-length cross-linking agent, thought to cross-link proteins to nucleic acids at the contact points, and therefore only proteins closely associated with RNA can be covalently linked by UV irradiation (42). For these experiments, we previously titrated the amount of purified NP, obtained from virions, that retained 100% of a 32 P-labeled RNA on nitrocellulose filter retention assays (100 ng of NP when 2 ng of the full-length CAT RNA probe is used). All of the MBP fusions were then tested in the three RNA binding assays at this protein/RNA molar ratio. Under these assay conditions, MBP-NP retained practically all input RNA in the filter binding assays (Fig. 4A). The other fusion proteins, except for F3, retained amounts of RNA ranging from 25 to 61% of the input probe (as opposed to 0.2% retained by MBP and F3), indicating that they have RNA-binding activity. It should be mentioned that the relative amount of bound RNA did not change when the input RNA was a fulllength CAT RNA, a shorter derivative of this RNA (71 nucleotides long), or a short poly(U)-rich RNA derived from plasmid pBSKpolyA (see Materials and Methods [data not shown]). Moreover, when the experiments were done at lower protein/RNA ratios the percentage of bound RNA was proportionally reduced compared with those shown in Fig. 4A

VOL. 69, 1995


3803

FIG. 3. Analyses of the MBP fusions. (A) Schematic diagrams of the MBP fusions containing either the intact NP or various NP fragments. The MBP is not shown to scale. Proteins F1 and F9 contain, at the C terminus, 11 additional amino acids (KLGTGRRFTTS) derived from the cloning procedure. Protein F4 also contains an additional 19 amino acids (PRVDLQACKLGTGRRFTTS) at the C terminus. A summary of the RNA-binding properties of each fusion protein, as determined from the data shown in Fig. 4, is also indicated. ND, not determined. (B) Analysis of the purified proteins indicated in panel A by SDSPAGE and Coomassie blue staining. Lanes BSA, 0.25, 0.5, and 1 mg of bovine serum albumin (BSA) loaded in the gel as standards; lane M, molecular weight markers (positions indicated in thousands).

(data not shown), indicating that the differences observed in the assay reflected differences in RNA binding affinity. The protein-RNA complexes were also analyzed by nondenaturing agarose gel electrophoresis. As shown in Fig. 4B, complexes between the MBP-NP and the RNA were stable and remained in the loading wells, as has been previously reported for a recombinant NP purified from bacteria (26). Proteins F1, F2, F4, F5, F8, and F9 retarded the mobility of variable amounts of the labeled probe, whereas proteins MBP, F6, and F3 did not alter the mobility of the RNA. Upon comparison of the results obtained in the filter binding and gel retardation experiments, it is clear that only proteins MBP-NP, F1, F4, F5, and F9 behave similarly in the two assays. In contrast, proteins F2 and F8, which retained 28 and 47% of the input RNA in the filter retention assays, respectively, retarded the mobility of a very small amount of RNA in the gel shift assays, and the RNA-binding activity of protein F6 was not detected in the latter assay. These results indicate that F2, F6, and F8 had reduced RNA-binding activity compared with the MBP-NP fusion and suggested that they had lost part of the RNA binding region. Because F2, F6, and F8 proteins contained deletions at the N-terminal part of the NP molecule, it could be implied that this protein region bears an RNA binding domain. In accordance with this interpretation, protein F1, containing amino acids 1 to 180, behaves as the MBP-NP fusion in the two RNA binding assays, demonstrating the presence of an RNA binding domain within this NP region. This domain could be further subdivided into two smaller regions, because proteins F8 (amino acids 1 to 77) and F9 (amino acids

FIG. 4. RNA-binding activities of the MBP fusions. Mixtures of the indicated protein and a 32P-labeled RNA were incubated as described in Materials and Methods. (A) Nitrocellulose filter binding assays. The incubation mixtures were filtered through nitrocellulose paper, and the amount of RNA retained in the filter was expressed as a percentage of the input RNA. The values indicated are the averages of several determinations, and bars at the top of each rectangle indicate the range of values obtained in two to five independent experiments. (B) Gel retardation assays. Incubation mixtures containing the short CAT RNA probe (see Materials and Methods) were electrophoresed in a 0.3% agarose gel (0.53 TAE) and analyzed by autoradiography. Only the two regions of the gel that contained radioactivity are shown. The region indicated by COMPLEX is the origin of the gel which corresponds to the RNA-protein complexes, and the region indicated by FREE, which corresponds to naked RNA, was separated by 6 cm from the gel origin. (C) UV light-induced cross-linking experiments. The protein-RNA mixtures were irradiated with UV light, treated with RNase A, and resolved by SDS-PAGE, and proteins containing residual cross-linked radioactive nucleotides were visualized by autoradiography. The RNA used was the full-length CAT RNA.

3804

ALBO ET AL.

79 to 180) retained RNA-binding activity in the two assays. It appears, however, that the region encompassing amino acids 79 to 180 was the most important region for RNA binding because protein F9 behaves similarly to MBP-NP in the gel shift assays. Finally, whether the fusion proteins could be UV crosslinked to RNA was analyzed. As shown in Fig. 4C, NP, obtained from virions, and the MBP-NP fusion were cross-linked to RNA with similar efficiencies, whereas proteins F1 (which includes amino acids 1 to 180) and F9 (which includes amino acids 79 to 180) showed reduced cross-linking efficiency compared with MBP-NP. Longer exposure of the gel showed a faint radioactive signal associated with proteins F5 and F8, but no signal was found associated with MBP and the F2 and F6 proteins. These results provided further support to the conclusions drawn from the filter binding and gel retardation studies and demonstrate that the amino-terminal 180 amino acids or an NP fragment containing residues 79 to 180 can directly interact with RNA. Furthermore, it should be emphasized that these results were in full agreement with those obtained with the NP-deleted derivatives (Fig. 2). It should be noted that the purified NP and the MBP-NP fusion exhibited similar RNA-binding activities in filter retention and gel retardation assays and cross-linked similar amounts of RNA; therefore, the presence of extra amino acids at the N terminus of NP did not significantly affect its RNAbinding activity. In this context, it should be mentioned that a recombinant NP containing 32 heterologous amino acids at the N terminus also retained RNA-binding activity (26). Some of the fusion proteins (F1, F4, F5, and F9) behaved similarly to the MBP-NP fusion in the filter retention and gel retardation assays. However, none of these fusion proteins cross-linked to RNA as efficiently as the MBP-NP fusion, suggesting that other NP regions in addition to the N-terminal domain are required for total RNA-binding activity or proper folding of the RNA binding domain. During the manuscript submission process, a study by Kobayashi and coworkers (28) describing the RNA-binding properties of truncated NPs by Northwestern blot analysis was published. Kobayashi et al. mapped an RNA binding region between residues 91 and 188 of the influenza A virus NP. However, they could not rule out that another RNA binding site could exist in the first 90 amino acids of the NP, as indicated in the manuscript. The results presented here show that two regions of the NP (amino acids 1 to 77 and 79 to 180) display RNA-binding activity, thus extending and complementing the results reported. Homologies to the NP region extending from amino acids 1 to 180. The N-terminal region (amino acids 1 to 180) of NP is not particularly rich in basic amino acids (17% versus 15% of the full-length NP) and does not contain any significant amino acid sequence homology to any previously identified RNA binding motif. In addition, searches in all available databases revealed no obvious homologous sequences to the NP regions encompassing amino acids 1 to 180, 1 to 77, and 79 to 180, except for the Orthomyxoviridae family NPs. The Orthomyxoviridae family includes the three types (A, B, and C) of influenza virus and two tick-borne viruses (Dhori and Thogoto) (17). Alignment of the amino acid sequence extending from residues 1 to 180 of the influenza A virus NP with that of the predicted NP sequences of influenza B and C and Dhori viruses reveals the conservation of a region that includes amino acids 142 to 180 (Fig. 5A) of influenza A virus NP. No such long conserved region was found between residues 1 to 141 of NP, because multiple gaps had to be introduced for proper alignment of this part of the protein (not shown). Therefore,

J. VIROL.

FIG. 5. Primary structure of a segment of the influenza A/Victoria/3/75 virus NP and its sequence similarity to other proteins. (A) Amino acid alignment of the Orthomyxoviridae family NPs. The sequences shown correspond to the NPs of the influenza A/Victoria/3/75 virus (INF A), B/Singapore/222/79 (INF B) (32), C/California/78 (INF C) (40), and Dhori virus isolate Indian/1313/61 (DHO) (18). Numbers on the left indicate positions of the initial amino acid in the corresponding protein. Highly conserved positions in at least three of the proteins are highlighted, and those conserved across the four proteins (up to residue 180) are indicated by asterisks. Some of the residues of the A/Victoria/3/75 NP are numbered (top). The predicted secondary structure of this region is shown at the top. (B) Alignment of the influenza A/Victoria/3/75 virus NP and truncated proteins d7 (D7) and d8 (D8). Positions maintained in at least two of the aligned sequences are highlighted.

although our data indicate that residues within amino acids 1 to 141 are important for the RNA-binding activity of NP, the lack of conserved features in this region precluded any speculation on the relevant residues required for this activity. We thus focused on the region shown in Fig. 5A that includes several residues conserved across all four NPs. This region is predicted to form a similar structure (alpha helix-loop-beta sheet-loop-alpha helix) in all four proteins. Remarkably, this region overlaps a 24-residue region (amino acids 158 to 182) that is completely conserved (no amino acid substitution) in more than 100 NP sequences determined for influenza A viruses (19, 48, 51) and includes the largest stretch of consecutive amino acids (GSTLPRRSG) conserved among influenza A and B viruses. Further indirect evidence to support the hypothesis that this conserved region may be relevant for RNA binding comes from the fact that protein d8, which bound poly(U) (Fig. 2), shares with the intact NP a G residue at position 169 and an R residue at position 175 and is predicted to fold as the NP in this region (Fig. 5B), whereas protein d7, which showed reduced RNA binding to the RNA homopolymers (Fig. 2), does not contain those residues at the indicated positions and is not predicted to fold as the intact NP. Searches for proteins sharing sequence similarities to the NP region encompassing amino acids 142 to 180 yielded no obvious homologous sequence, except for a region between amino acids 148 to 163 of NP homologous with amino acids 185 to 199 of the influenza A virus PB1 protein (22), one of the subunits of the influenza virus polymerase (reviewed in reference 30). However, RNA binding studies with the PB1 polypeptide are required to demonstrate that this region is involved in RNA binding. We are currently preparing NP derivatives containing point mutations within the N-terminal 180 amino acids of NP to identify specific residues relevant for the RNA-binding activity of the protein. In particular, we are mutating the conserved residues, indicated by an asterisk in Fig. 5A, that may be part of a new RNA binding motif. ACKNOWLEDGMENTS This work was supported by Comisıón Interministerial de Ciencia y Tecnologıá (grant BIO92-1044), Fondo de Investigaciones Sanitarias

VOL. 69, 1995


(grant 92/0893E), and the EC (SCIENCE Program, grant SC1-CT910688). We thank Jose A. Melero and Juan Ortıń for critical reading of the manuscript and Juan Ortıń for many helpful suggestions and for providing the pGEM-NP plasmid. We also thank Angel del Pozo for the artwork.

25.

26.

REFERENCES 1. Ashley, C. T., Jr., K. D. Wilkinson, D. Reines, and S. T. Warren. 1993. FMR1 protein: conserved RNP family domains and selective RNA binding. Science 262:563–566. 2. Ba ´rcena, J., M. Ochoa, S. de la Luna, J. A. Melero, A. Nieto, J. Ortıń, and A. Portela. 1994. Monoclonal antibodies against influenza virus PB2 and NP polypeptides interfere with the initiation step of viral mRNA synthesis in vitro. J. Virol. 68:6900–6909. 3. Baudin, F., C. Bach, S. Cusak, and R. W. H. Ruigrok. 1994. Structure of influenza virus RNP. I. Influenza virus nucleoprotein melts secondary structure in panhandle RNA and exposes the bases to the solvent. EMBO J. 13:3158–3165. 4. Beaton, A. R., and R. M. Krug. 1986. Transcription antitermination during influenza viral template RNA synthesis requires the nucleocapsid protein and the absence of a 59 capped end. Proc. Natl. Acad. Sci. USA 83:6282– 6286. 5. Bloomer, A. C., and P. J. G. Butler. 1986. Tobacco mosaic virus. Structure and self-assembly, p. 19–57. In M. H. V. Van Regenmortel and H. FraenkelConrat (ed.), The plant viruses, vol. 2. Plenum Publishing Corp., New York. 6. Blumberg, B. M., C. Giorgi, and D. Kolakofsky. 1983. N protein of vesicular stomatitis virus selectively encapsidates leader RNA in vitro. Cell 32:559– 567. 7. Bullock, W. O., J. M. Ferna ńdez, and J. M. Short. 1987. XL1-Blue: a high efficiency plasmid transforming recA Escherichia coli strain with beta-galactosidase selection. BioTechniques 5:376–378. 8. Burd, C. G., and G. Dreyfuss. 1994. Conserved structures and diversity of functions of RNA-binding proteins. Science 265:615–621. 9. Burd, C. G., E. L. Matunis, and G. Dreyfuss. 1991. The multiple RNAbinding domains of the mRNA poly(A)-binding protein have different RNAbinding activities. Mol. Cell. Biol. 11:3419–3424. 10. Choppin, P. W., and R. W. Compans. 1975. The influenza viruses and influenza, p. 15–51. In E. D. Kilbourne (ed.), Academic Press, New York. 11. de la Luna, S., J. Martıń, A. Portela, and J. Ortıń. 1993. Influenza virus naked RNA can be expressed upon transfection into cells co-expressing the three subunits of the polymerase and the nucleoprotein from simian virus 40 recombinant viruses. J. Gen. Virol. 74:535–539. 12. de la Luna, S., C. Martıńez, and J. Ortıń. 1989. Molecular cloning and sequencing of influenza virus A/Victoria/3/75 polymerase genes: sequence evolution and prediction of possible functional domains. Virus Res. 13:143– 156. 13. Desselberger, U., V. R. Racaniello, J. J. Zazra, and P. Palese. 1980. The 39 and 59-terminal sequences of influenza A, B and C virus RNA segments are highly conserved and show partial inverted complementarity. Gene 8:315– 328. 14. Duesberg, P. H. 1969. Distinct subunits of the ribonucleoprotein of influenza virus. J. Mol. Biol. 42:485–499. 15. Fabre, E., W. C. Boelens, C. Wimmer, I. W. Mattaj, and E. C. Hurt. 1994. Nup145p is required for nuclear export of mRNA and binds homopolymeric RNA in vitro via a novel conserved motif. Cell 78:275–289. 16. Fodor, E., D. C. Pritlove, and G. G. Brownlee. 1994. The influenza virus panhandle is involved in the initiation of transcription. J. Virol. 68:4092– 4096. 17. Francki, R. I. B., C. M. Fauquet, D. L. Knudson, and F. Brown. 1991. Classification and nomenclature of viruses. Arch. Virol. Suppl. 2:263–272. 18. Fuller, F. J., W. C. Clay, E. Z. Freedman, M. McEntee, and J. A. Barnes. 1989. Nucleotide sequence of the major structural protein genes of the tick-borne, orthomyxo-like Dhori/Indian/1313/61 virus, p. 279–286. In D. Kolakofsky and B. W. J. Mahy (ed.), Genetics and pathogenicity of negative strand virus. Elsevier Science Publishers, New York. 19. Gorman, O. T., W. J. Bean, Y. Kawaoka, I. Donatelli, Y. Guo, and R. G. Webster. 1991. Evolution of influenza A virus nucleoprotein genes: implications for the origins of H1N1 human and classical swine viruses. J. Virol. 65:3704–3714. 20. Honda, A., K. Ue´da, K. Nagata, and A. Ishihama. 1988. RNA polymerase of influenza virus: role of NP in RNA chain elongation. J. Biochem. 104:1021– 1026. 21. Huang, T.-S., P. Palese, and M. Krystal. 1990. Determination of influenza virus proteins required for genome replication. J. Virol. 64:5669–5673. 22. Kawaoka, Y., S. Krauss, and R. G. Webster. 1989. Avian-to-human transmission of the PB1 gene of influenza A viruses in the 1957 and 1968 pandemics. J. Virol. 63:4603–4608. 23. Kenan, D. J., C. C. Query, and J. D. Keene. 1991. RNA recognition: towards identifying determinants of specificity. Trends Biochem. Sci. 16:214–220. 24. Kiledjian, M., and G. Dreyfuss. 1992. Primary structure and binding activity

27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40.

41.

42.

43.

44. 45. 46.

47.

48.

49.

50.

51.

52.

3805

of the hnRNP U protein: binding RNA through RGG box. EMBO J. 11: 2655–2664. Kimura, N., M. Nishida, K. Nagata, A. Ishihama, K. Oda, and S. Nakada. 1992. Transcription of a recombinant influenza virus RNA in cells that can express the influenza virus RNA polymerase and nucleoprotein genes. J. Gen. Virol. 73:1321–1328. Kingsbury, D. W., I. M. Jones, and K. G. Murti. 1987. Assembly of influenza ribonucleoprotein in vitro using recombinant nucleoprotein. Virology 156: 396–403. Kingsbury, D. W., and R. G. Webster. 1969. Some properties of influenza virus nucleocapsids. J. Virol. 4:219–225. Kobayashi, M., T. Toyoda, D. M. Adyshev, Y. Azuma, and A. Ishihama. 1994. Molecular dissection of influenza virus nucleoprotein: deletion mapping of the RNA binding domain. J. Virol. 68:8433–8436. Krug, R. M. 1971. Influenza viral RNPs newly synthesized during the latent period of viral growth in MDCK cells. Virology 44:125–136. Krug, R. M., F. V. Alonso-Caplen, I. Julkunen, and M. G. Katze. 1989. Expression and replication of the influenza virus genome, p. 89–152. In R. M. Krug (ed.), The influenza viruses. Plenum Press, New York. Lazinski, D., E. Grzadzielska, and A. Das. 1989. Sequence-specific recognition of RNA hairpins by bacteriophage antiterminators requires a conserved arginine-rich motif. Cell 59:207–218. Londo, D. R., A. R. Davis, and D. P. Nayak. 1983. Complete nucleotide sequence of the nucleoprotein gene of influenza B virus. J. Virol. 47:642– 648. Luytjes, W., M. Krystal, M. Enami, J. D. Parvin, and P. Palese. 1989. Amplification, expression, and packaging of a foreign gene by influenza virus. Cell 59:1107–1113. Mahy, B. W. J. 1983. Mutants of influenza virus, p. 192–254. In P. Palese and D. W. Kingsbury (ed.), Genetics of influenza viruses. Springer, Vienna. Martin, J. 1991. Ph.D. thesis. Universidad Auto ńoma, Madrid. Martıń, J., C. Albo, J. Ortıń, J. A. Melero, and A. Portela. 1992. In vitro reconstitution of active influenza virus ribonucleoprotein complexes using viral proteins purified from infected cells. J. Gen. Virol. 73:1855–1859. Mena, I., S. de la Luna, C. Albo, J. Martıń, J. Ortıń, and A. Portela. 1994. Synthesis of biologically active influenza virus core proteins using a vaccinia virus-T7 RNA polymerase expression system. J. Gen. Virol. 75:2109–2114. Moyer, S. A., S. Smallwood-Kentro, A. Haddad, and L. Prevec. 1991. Assembly and transcription of synthetic vesicular stomatitis virus nucleocapsids. J. Virol. 65:2170–2178. Murti, K. G., R. G. Webster, and I. M. Jones. 1988. Localization of RNA polymerases on influenza viral ribonucleoproteins by immunogold labelling. Virology 164:562–566. Nakada, S., R. S. Creager, M. Krystal, and P. Palese. 1984. Complete nucleotide sequence of the influenza C/California/78 virus nucleoprotein gene. Virus Res. 1:433–441. Parvin, J. D., P. Palese, A. Honda, A. Ishihama, and M. Krystal. 1989. Promoter analysis of influenza virus RNA polymerase. J. Virol. 63:5142– 5152. Pashev, I. G., S. I. Dimitrov, and D. Angelov. 1991. Crosslinking proteins to nucleic acids by ultraviolet laser irradiation. Trends Biochem. Sci. 16:323– 326. Pons, M. W., I. T. Schulze, G. K. Hirst, and R. Hauser. 1969. Isolation and characterization of the ribonucleoprotein of influenza virus. Virology 39: 250–259. Ritchey, M. B., P. Palese, and E. D. Kilbourne. 1976. RNAs of influenza A, B, and C viruses. J. Virol. 18:738–744. Rost, B., and C. Sander. 1993. Prediction of protein secondary structure at better than 70% accuracy. J. Mol. Biol. 232:584–599. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y. Scharf, M., R. Schneider, G. Casari, P. Bork, A. Valencia, C. Ouzounis, and C. Sander. 1994. GENEQUIZ, a workbench for sequence analysis, p. 348– 353. In R. Altman, D. Brutlag, P. Karp, R. Lathrop, and D. Seearls (ed.), Second International Conference in Intelligent Systems for Molecular Biology. AAAI Press, Menlo Park, Calif. Scholtissek, C., S. Ludwig, and W. M. Fitch. 1993. Analysis of influenza A virus nucleoproteins for the assessment of molecular genetic mechanisms leading to new phylogenetic virus lineages. Arch. Virol. 131:237–250. Seong, B. L., and G. G. Brownlee. 1992. A new method for reconstituting influenza polymerase and RNA in vitro: a study of the promoter elements for cRNA and vRNA synthesis in vitro and viral rescue in vivo. Virology 186: 247–260. Shapiro, G. I., and R. M. Krug. 1988. Influenza virus RNA replication in vitro: synthesis of viral template RNAs and virion RNAs in the absence of an added primer. J. Virol. 62:2285–2290. Shu, L. L., W. J. Bean, and R. G. Webster. 1993. Analysis of the evolution and variation of the human influenza A virus nucleoprotein gene from 1933 to 1990. J. Virol. 67:2723–2729. Sibbad, P. R., and P. Argos. 1990. SCRUTINEER, a computer program that flexibly seeks and describes motifs and profiles in protein sequence data-

3806

ALBO ET AL.

bases. Comput. Appl. Biosci. 3:279–288. 53. Siomi, H., M. Choi, M. C. Siomi, R. L. Nussbaum, and G. Dreyfuss. 1994. Essential role for KH domains in RNA binding: impaired RNA binding by a mutation in the KH domain of FMR1 that causes fragile X syndrome. Cell 77:33–39. 54. Swanson, M. S., and G. Dreyfuss. 1988. Classification and purification of proteins of heterogeneous nuclear ribonucleoprotein particles by RNA-binding specificities. Mol. Cell. Biol. 8:2237–2241. 55. Tiley, L. S., M. Hagen, J. T. Matthews, and M. Krystal. 1994. Sequencespecific binding of the influenza virus RNA polymerase to sequences located at the 59 ends of the viral RNAs. J. Virol. 68:5108–5116. 56. van Wyke, K. L., W. J. Bean, Jr., and R. G. Webster. 1981. Monoclonal antibodies to the influenza A virus nucleoprotein affecting RNA transcription. J. Virol. 39:313–317. 57. Weiss, B., H. Nitschko, I. Ghattas, R. Wright, and S. Schlesinger. 1989.

J. VIROL.

58. 59.

60.

61.

Evidence for specificity in the encapsidation of Sindbis virus RNAs. J. Virol. 63:5310–5318. Winter, G., and S. Fields. 1981. The structure of the gene encoding the nucleoprotein of human influenza virus A/PR/8/34. Virology 114:423–428. Yamanaka, K., A. Ishihama, and K. Nagata. 1990. Reconstitution of influenza virus RNA-nucleoprotein complexes structurally resembling native viral ribonucleoprotein cores. J. Biol. Chem. 265:11151–11155. Yamanaka, K., N. Ogasawara, H. Yoshikawa, A. Ishihama, and K. Nagata. 1991. In vivo analysis of the promoter structure of the influenza virus RNA genome using a transfection system with an engineered RNA. Proc. Natl. Acad. Sci. USA 88:5369–5373. Ye, L., and M. Sugiura. 1992. Domains required for nucleic acid binding activities in chloroplast ribonucleoproteins. Nucleic Acids Res. 20:6275– 6279.