Studies on the intranuclear localization of influenza virus-specific proteins

Nuclear and nucleolar targeting of influenza A virus NS1 protein: striking differences between different virus subtypes

Influenza A virus nonstructural protein 1 (NS1A protein) is a virulence factor which is targeted into the nucleus. It is a multifunctional protein that inhibits host cell pre-mRNA processing and counteracts host cell antiviral responses. We show that the NS1A protein can interact with all six human importin alpha isoforms, indicating that the nuclear translocation of NS1A protein is mediated by the classical importin alpha/beta pathway. The NS1A protein of the H1N1 (WSN/33) virus has only one N-terminal arginine- or lysine-rich nuclear localization signal (NLS1), whereas the NS1A protein of the H3N2 subtype (Udorn/72) virus also has a second C-terminal NLS (NLS2). NLS1 is mapped to residues 35 to 41, which also function in the double-stranded RNA-binding activity of the NS1A protein. NLS2 was created by a 7-amino-acid C-terminal extension (residues 231 to 237) that became prevalent among human influenza A virus types isolated between the years 1950 to 1987. NLS2 includes basic amino acids at positions 219, 220, 224, 229, 231, and 232. Surprisingly, NLS2 also forms a functional nucleolar localization signal NoLS, a function that was retained in H3N2 type virus NS1A proteins even without the C-terminal extension. It is likely that the evolutionarily well-conserved nucleolar targeting function of NS1A protein plays a role in the pathogenesis of influenza A virus.

Rescue of influenza virus expressing GFP from the NS1 reading frame

In this study, several influenza NS1 mutants were examined for their growth ability in interferon (IFN)-deficient Vero cells treated with human interferon alpha (IFN-alpha). Mutants with an intact RNA binding domain showed similar growth properties as the wild-type virus, whereas viruses carrying an impaired RNA binding domain were dramatically attenuated. Relying on the ability of the first half of the NS1 protein to antagonize the IFN action, we established a rescue system for the NS gene based on the transfection of one plasmid expressing recombinant NS vRNA and subsequent coinfection with an IFN sensitive helper virus followed by adding of human IFN-alpha as a selection drug. Using this method, a recombinant influenza A virus expressing green fluorescence protein (GFP) from the NS1 reading frame was rescued. To ensure the posttranslational cleavage of GFP from the N-terminal 125 amino acids (aa) of NS1 protein, a peptide sequence comprising a caspase recognition site (CRS) was inserted upstream the GFP protein. Although a rather long sequence of 275 aa was inserted into the NS1 reading frame, the rescued recombinant vector appeared to be genetically stable while passaging in Vero cells and was able to replicate in PKR knockout mice.

Transfectant influenza A viruses with long deletions in the NS1 protein grow efficiently in Vero cells

We established a reverse genetics system for the nonstructural (NS) gene segment of influenza A virus. This system is based on the use of the temperature-sensitive (ts) reassortant virus 25A-1. The 25A-1 virus contains the NS gene from influenza A/Leningrad/134/57 virus and the remaining gene segments from A/Puerto Rico (PR)/8/34 virus. This particular gene constellation was found to be responsible for the ts phenotype. For reverse genetics of the NS gene, a plasmid-derived NS gene from influenza A/PR/8/34 virus was ribonucleoprotein transfected into cells that were previously infected with the 25A-1 virus. Two subsequent passages of the transfection supernatant at 40 degreesC selected viruses containing the transfected NS gene derived from A/PR/8/34 virus. The high efficiency of the selection process permitted the rescue of transfectant viruses with large deletions of the C-terminal part of the NS1 protein. Viable transfectant viruses containing the N-terminal 124, 80, or 38 amino acids of the NS1 protein were obtained. Whereas all deletion mutants grew to high titers in Vero cells, growth on Madin-Darby canine kidney (MDCK) cells and replication in mice decreased with increasing length of the deletions. In Vero cells expression levels of viral proteins of the deletion mutants were similar to those of the wild type. In contrast, in MDCK cells the level of the M1 protein was significantly reduced for the deletion mutants.

Expression and characterisation of the influenza A virus non-structural protein NS1 in yeast

The influenza A virus non-structural protein NS1 was produced using a copper-inducible expression system in the yeast Saccharomyces cerevisiae. The protein produced had a molecular weight of 26 kDa by SDS-PAGE and was reactive with anti-NS1 antisera. The recombinant NS1 protein was targetted to the nucleolus/nuclear envelope fraction of the yeast cell nucleus, showing that its localisation signals remain functional in yeast. In addition, immune-electron microscopy detected cytoplasmic inclusions reminiscent of those seen in cells infected with some influenza strains. The NS1 protein was shown to be capable of in vivo self-interaction which probably forms the basis of its propensity to form inclusions. Expression of the protein was found to be toxic to yeast cells expressing it, supporting a role for the protein in the shutdown of influenza virus-infected cells. Deletion mapping of NS1 pointed to 2 regions of the molecule being important for this toxicity: a basic C-terminal stretch which has been shown to act as a nuclear localisation signal, and an N-terminal region implicated in RNA binding.

Influenza virus NS1 protein stimulates translation of the M1 protein

The influenza virus NS1 protein was shown to stimulate translation of the M1 protein. M-CAT RNA, which contains the chloramphenicol acetyltransferase (CAT) reporter gene and the terminal noncoding sequence of segment 7 (coding for the M1 and M2 proteins), was ribonucleoprotein transfected into clone 76 cells expressing the influenza virus RNA polymerase and NP proteins required for the transcription and replication of influenza virus ribonucleoproteins. When the cells were superinfected with a recombinant vaccinia virus which expresses the NS1 protein, CAT expression from the M-CAT RNA was significantly stimulated but transcription was not altered. The expression of NS-CAT RNA, which contains noncoding sequences of segment 8 (coding for the NS1 and NS2 proteins), was not altered by the NS1 protein. Site-directed mutagenesis showed that the sequence GGUAGAUA upstream of the initiation codon on segment 7 was required for stimulation.

Naturally occurring NS gene variants in an avian influenza virus isolate

The A/Turkey/Wisconsin/68 (H5N9) isolate of avian influenza (AI) consists of two virus populations which have different NS genes and differ in their biological responses in chicken embryos. They were classified as being either rapidly embryo-lethal (REL) or slowly embryo-lethal (SEL), (Avian Dis., 33 (1989) 695-706). In this study, sequence analysis identified only two nucleotide differences between the two NS genes, creating single amino acid differences in both the NS1 and the NS2 protein. The difference in the NS1 protein appears to be neutral, while the differences in the NS2 places a phenylalanine at position 48. This amino acid has not been previously demonstrated at this position in an NS2 sequence and its presence results in a distinct hydrophobic shift in the region. The sequence specifying the phenylalanine also creates an EcoRI site in the cDNA of the REL NS gene. Analysis of several clones showed that this site appears to co-segregate with the REL characteristic. Molecular differences between the two NS gene variants were reflected by differences in the kinetics of early protein synthesis in infected cells. In particular, the NS2 protein is in higher concentration (relative to the NS1) in SEL-infected cells than in REL-infected cells. No differences were detectable, however, in the rates of viral replication, either in cell culture or in embryos. Also, the REL or SEL rate was established early during infection of the embryo and could not be competed out by the other variant population 3 h after inoculation. Thus, these two natural NS gene variants appear to specify early differences which influence the time of death of an infected embryo but the differences do not appear to influence virus replication.

Viral RNA polymerases

Recent progress in molecular biological techniques revealed that genomes of animal viruses are complex in structure, for example, with respect to the chemical nature (DNA or RNA), strandedness (double or single), genetic sense (positive or negative), circularity (circle or linear), and so on. In agreement with this complexity in the genome structure, the modes of transcription and replication are various among virus families. The purpose of this article is to review and bring up to date the literature on viral RNA polymerases involved in transcription of animal DNA viruses and in both transcription and replication of RNA viruses. This review shows that the viral RNA polymerases are complex in both structure and function, being composed of multiple subunits and carrying multiple functions. The functions exposed seem to be controlled through structural interconversion.

Isolation and preliminary characterization of a highly cytolytic influenza B virus variant with an aberrant NS gene

By repeated backcrosses of influenza virus A/Aichi/2/68 (H3N2) with B/Yamagata/1/73 in MDCK cells, a virus clone with HA of B serotype (clone B/610B5B/201, or clone 201) was obtained, which formed sharp plaques in MDCK cells and induced a severe cell lysis early after infection. Its structural proteins were indistinguishable from those of B/Yamagata. Electrophoresis of the RNA of the clone also showed an identical pattern to that of B/Yamagata except RNA segment 8 (NS gene), which migrated faster than the corresponding segment of B/Yamagata in a 2.8% polyacrylamide gel. Within the clone 201-infected MDCK cells, only one species of nonstructural (NS) polypeptide was demonstrable, which had the same electrophoretic mobility as NS2 of B/Yamagata, and any band which might be taken as the counterpart of NS1 of B/Yamagata was not detectable on the gel. Peptide mapping revealed that NS of clone 201 was structurally different from either NS1 or NS2 of wild-type B/Yamagata. NS gene and its function of clone 201 was successfully transferred to B/Lee/40 by genetic reassortment.

Analysis of an influenza A virus mutant with a deletion in the NS segment

The influenza virus host range mutant CR43-3, derived by recombination from the A/Alaska/6/77 and the cold-adapted and temperature-sensitive A/Ann Arbor/6/60 viruses, has previously been shown to possess a defect in the NS gene. To characterize this defect, nucleotide sequence data were obtained from cloned cDNAs. The CR43-3 NS gene was found to be 854 nucleotides long and to derive from the NS gene of the A/Alaska/6/77 parent virus by an internal deletion of 36 nucleotides. Direct sequencing of RNA 8 of CR43-3 virus confirmed that the deletion in the NS1-coding region was not an artifact that was generated during the cloning procedure. Protein analysis indicated that the NS1 protein of CR43-3 virus was synthesized in equal amounts in the restrictive (MDCK) cells as well as in the permissive (PCK) host cells. Also, indirect immunofluorescence studies of virus-infected cells showed that the NS1 protein of CR43-3 virus, like that of the parent viruses, accumulates in the nuclei of both cell systems. Although no differences in synthesis or localization of the NS1 protein could be detected, a consistent reduction in M1 protein was noted in CR43-3 virus-infected, nonpermissive cells as compared with that of the permissive host. Since analysis of the CR43-3 virus required us to obtain the NS nucleotide sequence of the 1977 isolate A/Alaska/6/77, we were able to compare this sequence with those of corresponding genes of earlier strains. The result of this analysis supports the idea of a common lineage of human influenza A viruses isolated over a 43-year period.

Efficient expression of influenza virus NS1 nonstructural proteins in Escherichia coli

RNA segment 8 of the influenza A virus genome codes for two nonstructural proteins, NS1 and NS2, for which the functions are unknown. Cloned cDNA copies of this gene from three different influenza A virus strains were inserted into an Escherichia coli plasmid expression vector, pAS1, carrying the strong regulatable lambda phage promoter, PL. After induction, the NS1 proteins were overproduced to levels of 20-25% of total cellular protein. This was surprising in that the codon composition for these eukaryotic genes is similar to that for weakly expressed proteins in E. coli. Thus, under the appropriate conditions, it appears that high level expression of genes containing a relatively large proportion of minor codons can be obtained. The NS1 protein produced in bacteria from a cloned cDNA copy of the A/PR/8/34 virus NS gene was purified to apparent homogeneity and used to generate a high-titer monospecific rabbit antiserum. Immunoprecipitation studies showed this antibody to be crossreactive against the NS1 proteins produced by several different influenza A virus strains. Immunofluorescence experiments in Madin-Darby canine kidney cells showed the NS1 proteins to be located in the nucleoplasm early in infection for all strains examined. With some of the strains, NS1-specific immunofluorescence was observed predominantly in the nucleoli later in infection. This technology can be used to obtain other viral proteins in pure form for structural, functional, and immunological studies.

Immunologic studies on the influenza A virus nonstructural protein NS1

We purified the major influenza virus nonstructural protein, designated NS1, from cytoplasmic inclusions that were solubilized and used to raise antisera in rabbits. One of the antisera was found to be specific for NS1 by complement fixation tests and analyses of immune precipitates. Antiserum to NS1 isolated from cells infected with A/WSN/33 virus specifically precipitated NS1 from extracts of cells infected with seven distinct isolates of influenza A virus representing five different antigenic subtypes. These included A/WSN/33, A/PR/8/34, A/FW/5/50, A/USSR/90/77, A/RI/5+/57, A/Victoria/3/75, and A/Swine /1977/31; however, NS1 from cells infected with B/Lee/40 virus was not precipitated. Radioimmunoassays using radioiodinated NS1 protein from A/WSN virus-infected cells and unlabeled cytoplasmic extracts of cells infected with various strains of influenza virus as competitors indicated significant antigenic cross-reactivities for the NS1 proteins of all influenza A viruses tested. The results suggest a gradual antigenic drift over the 45 yr separating the earliest and most recent virus isolates examined. Thus, compared with the virion neuraminidase and hemagglutinin antigens, NS1 appears to be highly conserved in different influenza A virus isolates.

Influenza virus-specific RNA and protein syntheses in cells infected with temperature-sensitive mutants defective in the genome segment encoding nonstructural proteins

Virus-specific protein and RNA syntheses have been analyzed in chicken embryo fibroblast cells infected with two group IV temperature-sensitive (ts) mutants of influenza A (fowl plague) virus in which the ts lesion maps in RNA segment 8 (J. W. Almond, D. McGeoch, and R. D. Barry, Virology 92:416-427, 1979), known to code to code for two nonstructural proteins, NS1 and NS2. Both mutants induced the synthesis of similar amounts of all the early virus-specific proteins (P1, P2, P3, NP, and NS1) at temperatures that were either permissive (34 degrees C) or nonpermissive (40.5 degrees C) for replication. However, the synthesis of M protein, which normally accumulates late in infection, was greatly reduced in ts mutant-infected cells at 40.5 degrees C compared to 34 degrees C. The NS2 protein was not detected at either temperature in cells infected with one mutant (mN3), and was detected only at the permissive temperature in cells infected with mutant ts47. There was no overall reduction in polyadenylated (A+) complementary RNA, which functions as mRNA, in cells infected with these mutants at 40.5 degrees C compared to 34 degrees C, nor was there any evidence of selective accumulation of this type of RNA within the nucleus at the nonpermissive temperature. No significant differences in ts mutant virion RNA transcriptase activity were detected by assays in vitro at 31 and 40.5 degrees C compared to wild-type virus. Virus-specific non-polyadenylated (A-) complementary RNA, which is believed to act as the template for new virion RNA production, accumulated normally in cells at both 34 and 40.5 degrees C, but at 40.5 degrees C accumulation of new virion RNA was reduced by greater than 90% when compared to accumulation at 34 degrees C.

Intracellular development of membrane protein of influenza virus

The intracellular development of membrane protein (MP) of influenza A virus was investigated by immunofluorescent staining. Monospecific antiserum was prepared by immunizing rabbits with MP eluted from SDS-polyacrylamide gels of SDS-disrupted NWS virions. In the productive infection in clone 1-5C-4 cells, MP antigen was first detected over the whole cell at 4 hr after infection, concomitantly with the appearance of hemagglutinin (HA) antigen in the cytoplasm, and bright nuclear fluorescence was then observed. Nucleoprotein (NP) antigen was detected in the nucleus prior to the appearance of fluorescence of MP antigen and thereafter the cytoplasmic fluorescence developed. Late in infection, all of these three antigens were observed predominantly in the cytoplasm with stronger fluorescence at the cell surface. Essentially similar findings were obtained in the abortive infections in L cells and BHK cells. The above results suggest that the membrane protein of influenza A virus is present in the nucleus as well as in the cytoplasm of infected cells.

Use of specific radioactive probes to study transcription and replication of the influenza virus genome

Specific radioactive probes have been obtained for both influenza virion RNA (vRNA) and for its complement (complementary RNA or cRNA): 32P-labeled complementary DNA (cDNA) synthesized with the avian sarcoma virus reverse transcriptase, and [125I]vRNA, respectively. From the kinetics of annealing of these two probes to RNA from canine kidney cells infected with the WSN strain of influenza virus, we have determined the average number of cRNA and vRNA sequences in the nucleus and cytoplasm as a function of time after infection. Immediately after infection, a small amount of vRNA is detected, presumably from the inoculum virus. As expected, the amount of cRNA is insignificant. During the first 1.75 h of infection, the most significant increase observed is in cRNA sequences. Most of these cRNA sequences are found in the cytoplasm, but a significant amount (30%) is found in the nucleus. During this time, a small but significant increase in vRNA is also detected in the nucleus and cytoplasm. From 1.75 to 2.75 h, the absolute amounts of both cRNA and vRNA increase, predominantly in the cytoplasm, with cRNA remaining as the majority species. Subsequently, the amount of vRNA increases with respect to cRNA and becomes the majority species. At 3.75 h, 95% of both cRNA and vRNA are found in the cytoplasm. Addition of actinomycin D at 1.75 h completely suppresses the subsequent ninefold increase in cRNA and does not have a significant effect on the subsequent 14-fold increase in cytoplasmic vRNA. This assay is also able to detect the cRNA produced as a result of primary transcription, operationally defined as the cRNA produced in the presence of 100 mug of cycloheximide per ml added at zero time of infection. Increases in cRNA in the presence of cycloheximide are detectable in both the nucleus and the cytoplasm. Addition of actinomycin D as well as cycloheximide at zero time completely suppresses the appearance of cRNA in the cytoplasm, whereas a large fraction (50%) of the increase in nuclear cRNA still occurs.

Influenza viral mRNA contains internal N6-methyladenosine and 5'-terminal 7-methylguanosine in cap structures

Influenza viral complementary RNA (cRNA), i.e., viral mRNA was radioactive when purified from the cytoplasmic fraction of cordycepin-treated canine kidney cells that were incubated with [methyl-3H]methionine during infection. Approximately 55 to 60% of the methyl-3H radioactivity was in internal N6-methyladenosine, a feature distinguishing this mRNA from those viral mRNA's that are known to be synthesized in the cytoplasm. The remaining methyl-3H radioactivity was in 5'-terminal cap structures that consisted of 7-methylguanosine in pyrophosphate linkage to 2'-o-methyladenosine, N6, 2'-O-dimethyladenosine, or 2'-O-methylguanosine. Methylated adenosine was the predominant penultimate nucleoside in caps, suggesting that cRNA synthesis in infected cells initiates preferentially with adenosine at the 5' end. In contrast to cRNA, influenza virion RNA segments extracted from purified virus contained mainly 5'-terminal ppA and no detectable cap structures.

Cell-free translation of influenza virus mRNA

Cytoplasmic poly (A)-rich RNA extracted from fowl plague virus-infected cells was found to program efficiently the translation of two major peptides in the wheat germ cell-free system. These peptides have the same electrophoretic mobility, on polyacrylamide gels, as the two major virion proteins M and NP. [35S] methionine tryptic peptide analysis by one-dimensionalthin-layer ionophoresis and finger printing by two-dimensional thin-layer ionophoresis and chromatography show a high degree of similarity between the two in vitro products and the authentic viral proteins M and NP. Although virion RNA is devoid of any poly (A) sequence, it is confirmed here that the viral complementary cytoplasmic RNA contains poly (A) stretches of varying lengths. Intact purified virion was found to promote the synthesis of very low amounts of the same NP and M proteins in this cell-free system. Quantitative aspects of data would indicate that this is due to minute amounts of complementary viral RNA associated with the virion or with the virion RNA itself. In conclusion, it is shown diectly by cell-free translation of authentic viral products that the influenza virion is "negative stranded" (Baltimore, 1971), at least for its two major structural proteins.

Purification of influenza viral complementary RNA: its genetic content and activity in wheat germ cell-free extracts

Influenza viral complementary RNA (cRNA) was purified free from any detectable virion-type RNA (vRNA), and its genetic content and activity in wheat germ cell-free extracts were examined. After phenol-chloroform extraction of cytoplasmic fractions from infected cells, poly(A)-containing viral cRNA is found in two forms: in single-stranded RNA and associated with vRNA in partially and fully double-stranded RNA. To purify single-stranded cRNA free of these double-stranded forms, it was necessary to employ, as starting material, RNA fractions in which cRNA was predominantly single stranded. Two RNA fractions were successfully employed as starting material: polyribosomal RNA and the total cytoplasmic RNA from infected cells treated with 100 mug of cycloheximide (CM) per ml at 3 h after infection. In WSN virus-infected canine kidney (MDCK) cells, the addition of CM at 3 h after infection stimulates the production of cRNA threefold and causes a very large increase in the proportion of the cytoplasmic cRNA which is single stranded; double-stranded RNA forms are greatly reduced in amount. Total cRNA was obtained by oligo(dT)-cellulose chromatography, and single-stranded cRNA was separated from double-stranded forms by Sepharose 4B chromatography. The cRNA preparation purified from polyribosomes consists of 95% single-stranded cRNA, with the remaining 5% apparently being double-stranded RNA forms. The cRNA preparation purified from CM-treated cells (CM cRNA) is even more pure: 100% of the radiolabeled RNA is single-stranded cRNA. Annealing experiments, in which a limited amount of 32P-labeled genome RNA was annealed to the cRNA, indicate that the purified cRNA contains at least 84 to 90% of the genetic information in the vRNA genome. Purified viral cRNA (CM cRNA) is very active in directing the synthesis of virus-specific proteins in wheat germ cell-free extracts.