The NB Protein of Influenza B Virus Is Not Necessary for Virus Replication In Vitro

Abstract

The NB protein of influenza B virus is thought to function as an ion channel and therefore would be expected to have an essential function in viral replication. Because direct evidence for its absolute requirement in the viral life cycle is lacking, we generated NB knockout viruses by reverse genetics and tested their growth properties both in vitro and in vivo. Mutants not expressing NB replicated as efficiently as the wild-type virus in cell culture, whereas in mice they showed restricted growth compared with findings for the wild-type virus. Thus, the NB protein is not essential for influenza B virus replication in cell culture but promotes efficient growth in mice.

The genome of Influenza B virus, a member of the family Orthomyxoviridae, consists of eight negative-strand RNA segments, which encode 11 proteins. Of these, nine are also found in influenza A virus: three RNA-dependent RNA polymerase subunits (PB1, PB2, and PA), hemagglutinin (HA), nucleoprotein (NP), neuraminidase (NA), matrix protein (M1), and two nonstructural proteins (NS1 and NS2). Two proteins, NB and BM2, are unique to influenza B virus. NB is encoded by RNA segment 6, which also encodes NA, while BM2 is encoded by segment 7.

The NB protein of influenza B virus is a type III integral membrane protein, expressed abundantly on the surface of virus-infected cells (5, 30, 31), and is incorporated into virions (5, 7). This small protein (100 amino acids) possesses an 18-residue N-terminal ectodomain, a 22-residue transmembrane domain, and a 60-residue cytoplasmic tail (5, 42). From previous studies measuring membrane currents and by analogy with the M2 protein of influenza A virus (12, 13, 35), NB is thought to function as an ion channel protein. However, the electrophysiological measurements of the NB protein based on the lipid bilayer system are difficult to interrupt. That is, proteins and peptides containing hydrophobic domains, which are believed to lack ion channel activity in cells, can yield channel recordings in lipid bilayers (22, 39, 40). Moreover, in the studies of Fischer et al. (13) and Sunstrom et al. (35), amantadine was used to demonstrate the loss of channel activity by the NB protein, despite the inability of this drug to inhibit influenza B virus replication. Thus, the available evidence challenges the notion that the NB protein has ion channel activity.

Recently, members of our group and others (14, 16, 23) established a reverse genetics system for generating influenza A virus entirely from cloned cDNAs, based on the in vivo synthesis of viral RNA. This innovation led us to devise a similar system to determine the importance of the NB protein in the life cycle of influenza B virus (Hoffmann et al. [17] and Jackson et al. [20] reported a reverse-genetics strategy for influenza B virus while this report was being prepared). Here, we generated influenza B virus NB knockout viruses by reverse genetics and tested their growth characteristics and other properties both in cell culture and in mice.

MATERIALS AND METHODS

Cells, viruses, and antibodies.

293T human embryonic kidney cells and Madin-Darby canine kidney (MDCK) cells were maintained in Dulbecco's modified Eagle medium supplemented with 10% fetal calf serum and in minimal essential medium containing 5% newborn calf serum, respectively. The 293T cell line is a derivative of the 293 line, into which the gene for the simian virus 40 T antigen was inserted (9). All cells were maintained at 37°C in 5% CO₂. An anti-NB rabbit serum was generated against the synthesized peptide NKRDDISTPRAGVD (amino acid residues 70 to 83 of the NB protein) coupled to keyhole limpet hemocyanin.

Construction of plasmids.

The cDNAs of B/Lee/40 virus were synthesized by reverse transcription of viral RNA with an oligonucleotide complementary to the conserved 3′ end of the viral RNA. The cDNA was amplified by PCR with gene-specific oligonucleotide primers containing BsmBI sites, and PCR products were cloned into the pT7Blueblunt vector (Novagen, Madison, Wis.). After digestion with BsmBI, the fragment was cloned into the BsmBI sites of a plasmid vector, which contains the human RNA polymerase I promoter and the mouse RNA polymerase I terminator, separated by BsmBI sites. These plasmids for the expression of virion RNA are referred to as PolI constructs in this report. The cDNAs encoding the PB2, PB1, PA, and NP genes of B/Lee/40 virus were cloned into the eukaryotic expression vector pCAGGS/MCS (controlled by the chicken β-actin promoter) (21, 24), resulting in pCABLeePB2, pCABLeePB1, pCABLeePA, and pCABLeeNP, which express the PB2, PB1, PA, and NP proteins, respectively.

The NB knockout mutants were constructed as follows. Mutated NA genes (see Fig. 1) were amplified by PCR from the PolI construct containing the B/Lee/40 NA gene and then digested with BsmBI. The sequences of the primers will be provided upon request. The BsmBI-digested fragment was cloned into the BsmBI sites of the PolI plasmid. The resulting constructs were designated pPolBLeeNBstop#1, pPolBLeeNBstop#2, and pPolBLeeNBstop#3. All of the constructs were sequenced to ensure that unwanted mutations were not present.

FIG. 1.

Schematic diagram of mutations introduced into the NA segment. Mutations are shown in bold (-, deletion; ∗, insertion). The numbers shown are nucleotide positions.

Plasmid-based reverse genetics.

Transfectant viruses were generated as reported earlier (23). Briefly, 12 plasmids (eight PolI constructs for eight RNA segments and four protein-expression constructs for polymerase proteins and NP) were mixed with transfection reagent (Trans IT LT-1 [Panvera, Madison, Wis.]), incubated at room temperature for 10 min, and added to 10⁶ 293T cells cultured in Opti-MEM (Invitrogen) containing 0.3% bovine serum albumin. Forty-eight hours later, viruses in the supernatants were collected and amplified in MDCK cells for the production of stock viruses.

Indirect immunofluorescence assay.

MDCK cells were infected with viruses at a multiplicity of infection (MOI) of 1 to 2 PFU per cell. After 8 h of infection, cells were fixed with 3% formaldehyde solution and permeated with 0.1% Triton X-100. Antigens were detected with rabbit anti-NB peptide rabbit serum as a primary antibody and fluorescein isothiocyanate-conjugated anti-rabbit immunoglobulin G as a secondary antibody.

Immunoprecipitation.

Influenza B virus-infected MDCK cells (MOI, 5 PFU/cell) were labeled with a mixture of [³⁵S]Met and [³⁵S]Cys (50 μCi/ml each) (Tran³⁵S-label; ICN Biochemicals) at 7 h postinfection for 2 h. The radiolabeled cells were lysed in RIPA buffer containing 10 mM Tris-HCl (pH 7.5)-100 mM NaCl-1 mM EDTA-0.5% Triton X-100 and then centrifuged. The anti-NB rabbit serum was added to the supernatant and incubated overnight at 4°C. Protein A-Sepharose beads were then added and incubated for 1 h at room temperature. The immune complexes were washed and separated on 4 to 20% gradient polyacrylamide gels (ISC BioExpress, Kaysville, Utah). The gels were dried and examined by autoradiography.

Replicative properties of transfectant viruses.

MDCK cells were infected with viruses at an MOI of 0.001 PFU per cells, overlaid with MEM medium containing 0.5 μg of trypsin per ml, and incubated at 37°C. Supernatants were assayed at different times for infectious virus in plaque assays on MDCK cells.

Experimental infection.

Five-week-old female BALB/c mice, anesthetized with methoxyflurane, were infected intranasally with 50 μl of virus. The dose lethal for 50% of mice (MLD₅₀) was determined as previously described (15). The replicative capacity of virus was determined by intranasally infecting mice (1.0 × 10⁴ PFU) and determining titers of virus in organs at 3 days postinfection, as described by Bilsel et al. (6).

RESULTS

Generation of B/Lee/40 virus by reverse genetics.

As a first step in determining the role(s) of the NB protein in virus replication, we generated B/Lee/40 (B/Lee) virus entirely from cloned cDNA, using plasmid-based reverse genetics (23). The plasmids contained cDNAs encoding all eight segments of B/Lee virus, flanked by the human RNA polymerase I promoter and the mouse RNA polymerase I terminator. We then transfected 293T cells with four plasmids expressing the PA, PB1, PB2, and NP proteins of B/Lee virus and eight plasmids that directed the production of eight viral RNA segments of B/Lee virus. Forty-eight hours after transfection, the virus, designated B/LeeRG, was recovered from the supernatant of 293T cells (10^3.5 50% tissue culture infectious dose).

NB protein knockout viruses are viable.

Using this reverse genetics system, we next attempted to generate mutant viruses that did not express the NB protein. We made three mutant PolI constructs, designated pPolBLeeNBstop#1, pPolBLeeNBstop#2, and pPolBLeeNBstop#3 (Fig. 1). In all mutant constructs, the initiation codon of the NB protein was converted from ATG to GCG (Met to Ala), and the codon at amino acid position 41 of the NB protein was changed from AAA to TAA (stop codon). pPolBLeeNBstop#2 has a single nucleotide deletion downstream of the mutated initiation codon, which was expected to alter the reading frame of the NB protein. pPolBLeeNBstop#3 has a nucleotide insertion downstream of the mutated initiation codon, which also was expected to alter the reading frame of the NB protein. At 48 h after transfection of 293T cells with each mutant NA PolI plasmid, together with seven other PolI plasmids and four protein expression plasmids, BLeeNBstop#1, BLeeNBstop#2, and BLeeNBstop#3 viruses were recovered from the supernatant (10^3.5 50% tissue culture infectious dose), indicating that all viruses lacking the NB protein were generated with an efficiency equivalent to that of the wild-type B/Lee virus. The transfectant viruses present in the supernatant were grown in MDCK cells and used as stock viruses. Sequencing of the NA gene of each stock virus confirmed the stability of the desired mutations and ruled out the introduction of additional mutations.

To confirm that the three mutant viruses did not express the NB protein, as intended, we performed indirect immunofluorescence assays and immunoprecipitation assays using virus-infected MDCK cells (Fig. 2A and B). None of the mutants was positive, in contrast to the B/LeeRG virus, which expressed NB. In immunoprecipitation studies, the NB protein was identified as an 18-kDa protein (high-mannose form) and as ∼30- to 50-kDa proteins (heterogeneous form) in agreement with the previously reported results (42, 43). Several cells infected with BLeeNBstop#1 virus showed faint, diffuse cytoplasmic staining in the immunofluorescence assays, which might indicate the production of a short NB peptide produced by alternative initiation and readthrough of the stop codon introduced. Thus, all three mutant viruses were viable and did not express the full-length NB protein.

FIG. 2.

Analysis of the expression of NB protein. (A) Detection of NB protein in infected MDCK cells by immunofluorescence assay. B/LeeRG, B/LeeRG-infected; WSN, A/WSN/33-infected; Control, uninfected; #1, #2, and #3, BLeeNBstop#1-, BLeeNBstop#2-, and BLeeNBstop#3-infected cells, respectively. (B) Detection of NB protein in virus-infected MDCK cells by immunoprecipitation assays. Radiolabeled NB proteins were immunoprecipitated with a rabbit anti-NB peptide serum and analyzed on 4 to 20%-gradient polyacrylamide gels as described in Materials and Methods. Lane #1, BLeeNBstop#1-infected; lane #2, BLeeNBstop#2-infected; lane #3, BLeeNBstop#3-infected; lane C, uninfected cell lysate. Molecular mass markers (kDa) are indicated.

Growth properties of NB knockout viruses in cell culture.

MDCK cells were infected with B/LeeRG, BLeeNBstop#1, BLeeNBstop#2, or BLeeNBstop#3 viruses at an MOI of 0.001 PFU per cell and incubated at 37°C. The supernatants were collected at different times postinfection, and titers of virus were determined by plaque assays with MDCK cells. BLeeNBstop#1, BLeeNBstop#2, and BLeeNBstop#3 viruses showed growth kinetics similar to those of B/LeeRG, with virus titers reaching 10⁷ PFU/ml at 36 h postinfection (Fig. 3). These results indicate that in cell culture, influenza B virus can undergo multiple cycles of replication and grow well without the NB protein.

FIG. 3.

Growth curves for B/LeeRG and mutant viruses. MDCK cells were infected with virus (0.001 PFU) and incubated at 37°C. At the indicated times after infection, titers of virus were determined in the supernatant. The values are means (± standard deviation) of three determinations.

Replication of NB knockout viruses in mice.

To determine the role of NB in influenza B virus replication in vivo, we first compared the MLD₅₀ of the wild-type and mutant viruses (Table 1). The MLD₅₀ values for NB knockout viruses were at least 1 log higher than the value for B/LeeRG. In tests of virus replication in the lungs and nasal turbinates of mice infected with 10⁴ PFU of virus (Table 1), B/LeeRG grew well in both sites, while the growth of mutant viruses was restricted, as shown by titers of virus that were generally more than 1 log lower than the titer for mutant viruses. Thus, although not required for growth in cell culture, the NB protein appears important for efficient replication of influenza B virus in mice.

TABLE 1.

Role of NB in virus replication in mice^a

Virus	Titer of virus (mean log PFU ± SD/g) in:		MLD50 (PFU)
	Lungs	Nasal turbinates
B/LeeRG	7.9 ± 0.2	6.5 ± 0.2	2.1 × 103
BLeeNBstop#1	5.2 ± 0.6	4.9 ± 0.3	4.3 × 104
BLeeNBstop#2	5.7 ± 0.1	3.9 ± 0.2	>1.5 × 105
BLeeNBstop#3	6.6 ± 0.04	3.4 ± 0.4	1.5 × 104

^aBALB/c mice, anesthetized with methoxyflurane, were infected intranasally with 50 μl of virus (10⁴ PFU). Three mice from each virus-infected group were sacrificed on day 3 postinfection for virus titration. The MLD₅₀ was determined as described previously (15).

DISCUSSION

Here we demonstrate that the NB protein is not essential for replication of influenza B virus in cell culture but promotes efficient replication in vivo. In this regard, NB is similar to the M2 protein of A/WSN/33 influenza virus, although the requirement for NB during in vivo replication appears less stringent than that for the M2 protein. An A/WSN/33 mutant lacking the transmembrane and cytoplasmic domains of M2 was severely attenuated in mice (41), and a mutant of A/Udorn/72 (H3N2) lacking nucleotides encoding amino acid residues 29 to 31 of the M2 protein was attenuated even in cell culture (36). Although the ion channel activity of M2 is experimentally well established (10, 18, 27, 33, 34), such activity has not been unequivocally demonstrated for the NB protein. Thus, the limited dependency of influenza B virus on NB function may suggest either that the virus does not depend as much on ion channel activity as influenza A virus does or that NB has functions other than ion channel activity. Since NB is highly conserved among influenza B virus strains, such functions must be important for viral replication in a natural setting.

Current human vaccines are inactivated vaccines that reduce the severity of, but are limited in their ability to prevent, viral infection. Clinical trials of cold-adapted live attenuated vaccines have generated promising results with respect to both efficacy and safety (1, 2, 3, 4, 8, 19, 25, 32, 38, 44). However, a molecular basis for the attenuation of the master vaccine strain of influenza B viruses remains unknown. Thus, it is important to produce an influenza B virus with known attenuating mutations. It would be ideal to produce a master vaccine strain which contains attenuating mutations exclusively in genes other than the HA and NA genes, so that only the latter genes need replacement with those of a field strain for vaccine production. However, with the invention of reverse genetics, it is no longer difficult to modify even the HA and NA genes for vaccine production. Thus, the mutations to knockout NB expression may be included, in addition to other attenuating mutations, in vaccine strains, considering that no growth defect was detected with NB knockout viruses in cell culture.

Although the replicative abilities of NB knockout viruses were similar to each other in MDCK cells, they differed in mice. This difference in replicative ability among the mutants in mice may originate from different levels of NA expression. To knock out NB expression, the upstream sequence of the NA protein was modified. This might have altered NA protein expression levels, resulting in various extents of attenuation in vivo. Further experiments are in progress to clarify this point.

Thus far, five viral proteins have been reported to act as ion channels: the M2 protein of influenza A virus, the NB protein of influenza B virus, Vpu and Vpr of human immunodeficiency virus type 1 (HIV-1), and Kcv of chlorella virus (11, 26, 28, 29, 33, 34, 35). The Vpr and Kcv proteins have been demonstrated to play an important role in the viral life cycle. The Vpu gene of HIV-1 can be deleted without completely abrogating HIV-1 replication in vitro. In the present study, we show that the NB protein is not necessary for viral growth in cell culture but appears to be required for efficient influenza B virus replication in mice. The lack of an apparent growth defect of NB knockout viruses in cell culture suggests the difficulty of determining the true function of this protein. Since the effects of mutations in the M2 protein differ among viral strains (36, 37), it is possible that deletion of the NB protein may produce growth defects in some influenza B viruses but not in others. Hence, it will be important to determine the effects of NB deletion in influenza B virus strains other than B/Lee/40.