Influenza A and B Virus Intertypic Reassortment through Compatible Viral Packaging Signals

Human embryonic kidney 293T (293T) cells (ATCC CRL-11268) and Madin-Darby canine kidney (MDCK) cells (ATCC CCL-34) were maintained in Dulbecco's modified Eagle's medium (DMEM; Mediatech, Inc.) supplemented with 10% fetal bovine serum (FBS; Atlanta Biologicals) and 1% PSG (penicillin, 100 units/ml; streptomycin, 100 μg/ml; l-glutamine, 2 mM; Mediatech, Inc.). Cells were grown at 37°C in a 5% CO₂ atmosphere. MDCK cells constitutively expressing HA (MDCK-HA) from the pandemic H1N1 virus A/California/4_NYICE_E3/2009 (pH1N1/E3) were previously described (4). MDCK-HA cells stably expressing HA from influenza virus A/Puerto Rico/8/1934 (H1N1; PR8), B/Lee, influenza virus B/Victoria/1/1987 (B/Vic), and influenza virus B/Yamagata/16/1988 (B/Yam) were generated by cotransfecting each pCAGGS HA plasmid and pCB7 (3:1 ratio) for eukaryotic expression of HA and hygromycin B resistance, respectively (4, 26, 29). HA-expressing cells were grown in DMEM–10% FBS–1% PSG supplemented with 200 μg/ml hygromycin B (Corning). After viral infections, cells were maintained at 33°C in a 5% CO₂ atmosphere in DMEM containing 0.3% bovine serum albumin (BSA), 1% PSG, and 1 μg/ml tosylsulfonyl phenylalanyl chloromethyl ketone (TPCK)-treated trypsin (Sigma) (30).

Recombinant wild-type (rWT) PR8 (31), B/Yam (32), and pH1N1 (33) and chimeric influenza A/B virus in the genetic background of PR8 were rescued using ambisense reverse genetics plasmids (pDZ) as previously described (30, 34, 35). Influenza A virus X31 is a recombinant virus with a PR8 backbone that contains HA and NA from influenza virus A/Hong Kong/1/1968 (H3N2) (36). The sciIAV on the pH1N1 backbone expressing the green fluorescent protein (GFP) in place of HA has been previously described (4). Using the same approach, a sciIAV on the PR8 backbone was generated (31). To rescue sciIBV, influenza virus B/Brisbane/60/2008 (B/Bris) ambisense plasmids (pDP2002) were used as previously described (37). Virus stocks were propagated in MDCK (rWT) or MDCK-HA (sciIAV or sciIBV) cells for 3 days at 33°C in a 5% CO₂ atmosphere (30). The virus titers of the stocks were determined by standard plaque assay (PFU/ml) in MDCK or MDCK-HA cells (4, 38). The virus titers in the tissue culture supernatants (TCSs) of multicycle growth analyses were determined by a focus formation assay (focus-forming units [FFU]/ml) by direct GFP quantification with fluorescence (sciIAV or sciIBV) or immunofluorescence (rWT) assays (4). Monoclonal antibodies (MAbs) against NP from influenza A virus (MAb HT103) (39) or influenza B virus (MAb B017; Abcam) and fluorescein isothiocyanate (FITC)-conjugated secondary antibodies were used, as described below and previously (4, 40).

The reverse genetics plasmids used to generate PR8-sciIAV (31) and the influenza A virus ΔHA/GFP plasmid pPolI HA(45)GFP(80) have been previously described (28). The reverse genetics plasmids used to rescue B/Brisbane/60/2008 were generated by extracting RNA (kindly provided by the CDC) using an RNeasy minikit (Qiagen). Segments 1 to 3 and 7 were then cloned using a two-step reverse transcription-PCR (RT-PCR) strategy modeled after that used for influenza A virus (41, 42). Avian myeloblastosis virus reverse transcriptase and RNasin were used for the extension of vRNA as described by the manufacturer (Promega). For these segments, AarI and BsmB1 restriction sites were used to clone each segment into pDP2002 (43). Segments 4 to 6 and 8 were cloned according to the one-step RT-PCR strategy previously described (44). For all 8 segments, QuikLigase (New England BioLabs [NEB]) was used for ligation of the digested PCR product into pDP2002. Plasmid clones were sequenced by the Sanger sequencing method according to the BigDye Terminator kit protocol (BigDye Terminator [v3.1] cycle sequencing kit; Life Technologies) and run on a 3500xL genetic analyzer (ABI).

pCAGGS plasmids carrying B/Vic, B/Yam, and B/Lee HAs were generated by RT-PCR from RNA isolated from infected MDCK cells, using the EcoRI and NheI restriction enzymes (NEB). Minigenome (MG) plasmids containing GFP or firefly luciferase (FFluc) flanked by the 3′ and 5′ NCRs of the influenza B virus HA segment (IBV MG) were cloned into the human pPolI plasmid using SapI (NEB), as previously described for the IAV MG plasmid (31, 45). Ambisense pDZ plasmids carrying PR8 or B/Yam PB2, PB1, PA, and NP or pCAGGS carrying lymphocytic choriomeningitis virus (LCMV) NP were previously described (31, 32, 46). Chimeric influenza A/B virus constructs were generated by subcloning the entire ORF of B/Yam HA or NA into vRNA expression plasmids that contained the 3′ and 5′ NCRs and packaging signals from influenza A virus HA (28) or NA (47). B/Yam HA was cloned into human pPolI HA(45)GFP(80) (28) using NheI and XhoI (NEB), and B/Yam NA was cloned into pHW2000 Flu-GFP (47) using MluI and SpeI (NEB). Influenza B virus ΔHA/GFP plasmid pPolI HA(228)GFP(108) was generated as previously described (28). The B/Yam HA segment was amplified by RT-PCR from RNA isolated from infected MDCK cells and cloned into the human pPolI plasmid using SapI (NEB) and then digested with BglII and BsaI (NEB) to cut 228 nucleotides (nt) from the 3′ end and 108 nt from the 5′ end. The product from PCR amplification of GFP using primers flanked by BglII and BsaI was subcloned between these two restriction sites. Following rescue of sciIBV (228)GFP(108), sequencing revealed that pPol HA(228)GFP(108) contained a primer duplication mutation. This mutation introduced the primer twice (on the 3′ end) immediately upstream of the GFP ORF. All plasmids were generated using standard cloning techniques and purified using a Wizard SV kit (Promega). The primers (Integrated DNA Technologies, Inc.) used for the generation of the described plasmid constructs are available upon request. All plasmid constructs were verified by DNA sequencing (ACGT, Inc.).

For the characterization of MDCK-HA cells, confluent monolayers of parental MDCK or MDCK-HA cells (10⁵ cells, 48-well plate format) were fixed with 4% paraformaldehyde for 15 min at room temperature and blocked overnight with 2.5% BSA at 4°C. Cells were then incubated with mouse MAb 31C2 or 29E3 (pH1N1 HA, 1 μg/ml) (48), MAb PY102 (PR8 HA, 1.5 μg/ml) (49), or a 1:1,000 dilution of goat polyclonal antibody (PAb) against HA of influenza virus B/Hong Kong/8/1973 (PAb NR-3165; NIAID Biodefense and Emerging Infectious Research Resources Repository [BEI Resources]) for 1 h at 37°C. After washing with 1× phosphate-buffered saline (PBS), the cells were incubated with FITC-conjugated secondary antimouse (1:140; Dako) or antigoat (1:200; Jackson ImmunoResearch) antibodies and 4′,6′-diamidino-2-phenylindole (DAPI; Research Organics) for 30 min at 37°C. After three washes with 1× PBS, the cells were visualized using a fluorescence microscope, photographed (Cooke Sensicam QE) with a ×20 objective (actual magnification, ×200), and colored and merged using Adobe Photoshop CS4 (v11.0) software. Representative images from at least three independent fields are shown.

For virus characterization, confluent monolayers of MDCK cells (10⁵ cells, 48-well plate format) infected at a multiplicity of infection (MOI) of 10 were fixed at 8 h postinfection as described above and permeabilized with 0.2% Triton X-100. Viral HA or NP was detected using antibodies PY102 and HT103, respectively, for influenza A virus or antibodies NR-3165 and B017, respectively, for influenza B virus. NA was stained with mouse anti-PR8 MAb NR-4540 (1:500 dilution; BEI Resources) for influenza A virus and goat anti-B/Lee PAb NR-3114 (1:1,000 dilution; BEI Resources) for influenza B virus and stained with FITC-conjugated secondary antibodies and DAPI, as described above. For phenotypic analysis of influenza A/B reassortant viruses, 1 μg/ml of mouse MAbs against influenza A virus HA (29E3), NA (10C9), nonstructural region 1 (NS1; 1A7), or NP (NR-4545) or goat PAbs against influenza B virus HA (NR-3165) or NA (NR-3114) were used as described above.

Parental MDCK and MDCK-HA cells (10⁶ cells, 6-well plate format) were trypsinized and stained in suspension as previously described (4). Briefly, cells were washed and incubated with 31C2, PY102, NR-3165, rabbit anti-B/Lee PAb (V-63, 1:1,000) (18), and mouse anti-B/Yam MAb (MAb 15B6, 1:1 of hybridoma TCS) (18) for 30 min at 4°C. After washing 3 times in staining buffer (1× PBS with 5% FBS). cells were stained with FITC-conjugated secondary antimouse or antirabbit antibodies (1:140; Dako) for 30 min at 4°C. Cells were washed 3 times with staining buffer, and the percentage of fluorescent cells was determined using a C6 4-color flow cytometer (Accuri). The frequencies of FL1-positive cells that fell into an appropriate side scatter gate were analyzed using FlowJo software (Tree Star). Parental cells stained with antibodies were used as a negative control. To evaluate MG activity, 293T cells were transfected as described below and 48 h later were trypsinized, washed 3 times in staining buffer, and analyzed using a C6 flow cytometer. The percentages and mean fluorescence intensities (MFIs) of GFP-positive (GFP⁺) cells were determined by gating on FL1-positive cells.

Parental MDCK or MDCK-HA cells were evaluated for protein expression by Western blotting as previously described (50). Briefly, samples were lysed in passive lysis buffer (Promega), separated on a 10% SDS-polyacrylamide gel, and transferred onto nitrocellulose membranes (Bio-Rad) for 2 h at 175 mA. After blocking for 1 h at room temperature with 10% dry milk in 1× PBS–0.1% Tween 20, membranes were incubated with antibodies against HA (31C2, 1 μg/ml; PY102, 1 μg/ml; NR-3165, 1:1,000) or mouse monoclonal antiactin antibody (1:1,000; Sigma) as a loading control. The membranes were then washed 3 times with 1× PBS–0.1% Tween 20 and probed with secondary horseradish peroxidase (HRP)-conjugated anti-mouse or anti-goat Ig (1:2,000 dilution; GE Healthcare) for 1 h at room temperature. After washing 3 times with 1× PBS–0.1% Tween 20, protein bands were detected using a chemiluminescence kit (Denville Scientific Inc.) and photographed using a Kodak ImageStation.

Multicycle growth analyses were performed by infecting confluent monolayers of parental MDCK or MDCK-HA cells (5 × 10⁵ cells, 12-well plate format) at an MOI of 0.001 (in triplicate), and the cells were incubated at 33°C for 3 days (26). At the times postinfection indicated below, GFP expression was assessed by fluorescence microscopy, and viral titers in the TCS were measured by evaluating the number of FFU/ml in a focus formation assay. Briefly, confluent wells of MDCK-HA cells (5 × 10⁴ cells, 96-well plate format, in triplicate) were infected with 10-fold serial dilutions of the TCS. At 18 h postinfection, the cells were washed with 1× PBS and foci were visualized using a fluorescence microscope and enumerated. Titrations of chimeric influenza A/B viruses were performed in a manner similar to that described above, but individual infected cells were detected at 10 h postinfection by immunofluorescence using influenza virus NP MAb HT103 (PR8) or B017 (B/Yam). Mean values and standard deviations (SDs) were calculated using Microsoft Excel software. The limit of detection was determined by the initial dilution of TCS for titration, where 50 μl of a 1:10 dilution was used.

The plaque phenotype of PR8 or pH1N1 sciIAVs in parental MDCK or MDCK-HA cells was determined by infecting confluent monolayers of cells (10⁶ cells, 6-well plate format) with specified viruses for 3 days at 33°C in postinfection medium containing agar. Monolayers were then fixed with 4% paraformaldehyde for 15 min at room temperature and stained with 0.1% crystal violet (Fisher Scientific). Plaque assays for chimeric influenza A/B viruses were performed similarly, but after infection and fixation, the monolayers were permeabilized and immunostained as previously described (29) using HT103 or B017 to detect influenza A or B virus NP, respectively.

Receptor-destroying enzyme-treated (Denka Seiken) sera or MAbs were serially diluted in 96-well V-bottom plates and mixed 1:1 with 8 hemagglutinating units (HAU) of influenza A or B virus HA-pseudotyped PR8-sciIAV or rWT PR8, X31, or B/Yam for 30 min at room temperature. Hemagglutination inhibition (HAI) assay titers were determined by adding 0.5% turkey red blood cells (RBCs) to the virus-antibody mixtures for 30 min on ice, as previously described (51). For sciIAV HAI assays, MAbs 31C2, PY102, NR-3165, and V-63 or PAbs anti-B/Vic and anti-B/Yam sheep (which were obtained from the CDC and which were for the years 2012 and 2013) were used.

Human 293T cells were transfected in suspension (5 × 10⁵ cells, 12-well plate format, in triplicate) with 500 ng each of influenza A or B virus ambisense pDZ PB2, PB1, and PA (3P) plasmids in the absence (negative control) or presence of the respective pDZ NP plasmid, together with influenza A or B virus MG plasmids expressing GFP or FFluc. A plasmid expressing Renilla luciferase under the control of the simian virus 40 (SV40) promoter (Promega) was cotransfected to normalize transfection efficiencies (52). In all cases, cells were cotransfected with empty pDZ plasmid to achieve equimolar DNA concentrations. Lipofectamine 2000 transfection reagent (Invitrogen) was used at a 1:1 ratio with plasmid DNA. At 48 h posttransfection, cells were prepared for flow cytometry or cell lysates were prepared using passive lysis buffer (Promega). Renilla luciferase and FFluc expression was determined using Stop & Glow reagents (Promega) and quantified with a Lumicount luminometer (Packard). The fold induction over the level of induction for the negative controls was calculated, and statistics (Student's two-tailed t test) were calculated with GraphPad Prism software. For competition experiments, increasing concentrations (250, 500, or 1,000 ng) of competing PR8, B/Yam, or LCMV NP expression plasmids were added, in addition to 3P plus NP and MG vRNA expression plasmids, and values were normalized to the fold induction in the absence of competitors (percent reporter expression).

Rescue transfections where sciIAV or sciIBV was each rescued with the ΔHA/GFP construct pPolI influenza A virus HA(45)GFP(80) or pPolI influenza B virus HA(228)GFP(108) were performed as previously described (26). Briefly, 293T cells (10⁶ cells, 6-well plate format) were cotransfected with seven ambisense rescue plasmids (for PB2, PB1, PA, NP, NA, M, and NS) for PR8 (31) or B/Bris (37), together with pCAGGS HA protein expression plasmids to facilitate initial rescue (26), and ΔHA/GFP vRNA expression plasmids, using Lipofectamine 2000 (1:1 ratio). The GFP fluorescence from vRNA expression plasmids in rescue transfections was monitored at 2 days posttransfection. Directly afterward, TCSs were collected and passaged onto fresh HA-expressing MDCK cells. The rescue of sciIAV or sciIBV with each ΔHA/GFP construct was confirmed by fluorescence and standard hemagglutination assay at 2 to 4 days postinfection (30). Triplicate transfections were performed in five independent experiments.

For immunogenicity and pathogenicity experiments, female 6- to 8-week-old C57BL/6 mice were purchased from the National Cancer Institute (NCI) and maintained at the University of Rochester under specific-pathogen-free conditions. All mouse experiments were approved by the University Committee of Animal Resources and complied with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Research Council (53). Infections were performed as previously described (4). Briefly, mice were anesthetized intraperitoneally (i.p.) with 2,2,2-tribromoethanol (Avertin; 240 mg/kg of body weight) and inoculated intranasally (i.n.) with 30 μl of virus preparations (4). Morbidity was monitored by determination of the percentage of body weight loss relative to the starting weight. Mice losing greater than 25% of their body weight were euthanized humanely. At 2 weeks postinfection, mouse sera were collected by submandibular bleeding and stored at −20°C. Naive serum was collected from animals prior to infection and used as a negative control. The 50% mouse lethal dose (MLD₅₀) was determined by infecting five groups of mice (n = 4) with 10-fold serial dilutions of 10⁵ to 10¹ PFU and calculated using the method of Reed and Muench (54).

Enzyme-linked immunosorbent assays (ELISAs) were performed as previously described by coating plates for 18 h at 4°C with lysates from PR8- or B/Yam-infected MDCK cells (4). After blocking with 1% BSA, the plates were incubated with 2-fold dilutions of serum for 1 h at 37°C (1:250 starting dilution), washed with H₂O, and incubated with HRP-conjugated goat anti-mouse IgG (1:2,000; Southern Biotech) for 30 min at 37°C. Reactions were then developed with tetramethylbenzidine substrate (BioLegend) for 5 min at room temperature, quenched with 2 N H₂SO₄, and read at 450 nm (Vmax kinetic microplate reader; Molecular Devices).

The minimal requirement for intertypic virus reassortment is conservation of protein function. To determine the functional compatibility of influenza B virus HA with an influenza A virus backbone, we chose to utilize our HA-deficient sciIAV approach (26). This virus requires HA protein trans-complementation, because the fourth segment is modified to encode GFP (28). MDCK cells that stably express full-length influenza virus type A (pH1N1/E3 and PR8) or type B (B/Lee, B/Vic, and B/Yam) HA proteins were first generated (Fig. 1). All MDCK-HA cells expressed the viral glycoproteins on the cell surface (Fig. 1A) at comparable levels (Fig. 1B). Flow cytometric detection of HA then demonstrated the distinct antigenic signature of the various MDCK-HA stable cell lines (Fig. 1C).

To evaluate intertypic HA compatibility, sciIAV was used to infect MDCK-HA cells, where productive infection can result only from functional HA complementation. An sciIAV that contained seven segments of PR8 plus the influenza A virus HA(45)GFP(80) transgene was used to infect MDCK-HA cells at a low MOI, and any virus progeny would be pseudotyped with HA from the cells (Fig. 2) (28). Regardless of homo- or heterotypic HA expression, the MDCK-HA cells were able to complement the HA-deficient, GFP-expressing sciIAV over multiple cycles of infection, as characterized by fluorescence microscopy that showed the spread of GFP expression (Fig. 2A) and the production of infectious progeny in the TCS (Fig. 2B). As expected, no propagation of virus was seen in parental MDCK cells (26). These results were corroborated by plaque assays (Fig. 2C). Similar trans-complementation was achieved when cells were infected with an evolutionarily distinct sciIAV containing the backbone of the pH1N1 reassortant virus that emerged in 2009 (4), which possesses internal segments of both avian and swine influenza virus origin (Fig. 3) (55). Influenza virus type A or B MDCK-HA cells also complemented the growth of pH1N1-sciIAV, as shown by the spread of GFP expression (Fig. 3A), virus production (Fig. 3B), and the plaque size phenotype (Fig. 3C). The identity of the MDCK-HA cells was further confirmed by performing HAI assays of the HA-pseudotyped sciIAV using MAbs and PAbs specific for each virus isolate (Table 1). Together, these results indicate that full-length influenza B virus HA protein complements HA-deficient sciIAV to generate an infectious influenza virus.

In a previous study, Muster and colleagues demonstrated that a replicating influenza A virus could be rescued when the NCR of influenza A virus NA was replaced with the NCR of the influenza B virus nonstructural (NS) segment (20). This demonstrated that influenza A virus RdRp can utilize the promoter activity of influenza B virus NCR, but importantly, the regions later to be defined as packaging signals (within the ORF) were left unchanged. In order to more fully define the NCR and RdRp compatibilities between influenza A and B viruses (Fig. 4), we used type A virus MG plasmids (52) to generate similar virus-like RNA expression plasmids that transcribe GFP or FFluc, which is flanked by influenza B virus NCRs (Fig. 4A). To normalize for cellular translation levels, a protein expression plasmid encoding Renilla luciferase under the control of the SV40 promoter was cotransfected with the MG plasmids. By flow cytometry, a significantly higher percentage of cells was found to express GFP from the MG when homotypic RdRp rather than heterotypic RdRp was cotransfected (Fig. 4B). Additionally, homotypic MG activity was greater in GFP⁺ cells, as the MFI was higher (Fig. 4B). Luminescence quantification of MG activity provided similar results, where influenza A or B virus RdRp could transcribe the heterotypic MG, albeit it transcribed the heterotypic MG to lower levels than it transcribed the homotypic MG (Fig. 4C). Although heterotypic RdRp is not as efficient as homotypic RdRp for MG replication and transcription, these data suggest that the lack of type A and B virus heterotypic reassortment is not due to an incompatibility of promoters or polymerase activities.

Other discrepancies that might impede heterotypic reassortment could include the inhibitory effect of influenza B virus NP on influenza A virus replication (56). Indeed, we observed that coexpression of influenza B virus NP inhibited IAV MG expression, whereas that of the NP of another negative-stranded RNA virus (LCMV) did not (Fig. 4D). Whether this inhibition results from affinity differences and substrate (vRNA/cRNA) binding competition, a specific virus, host protein interaction with influenza B virus NP, or some other mechanism is unknown. However, even in systems devoid of influenza B virus NP, virus rescue still does not occur when the HA segment of influenza B virus is used to complement reverse genetics plasmids encoding seven segments of influenza A virus (18, 19), suggesting that a separate mechanism is responsible for the lack of intertypic reassortants in the absence of this protein.

The packaging signals of influenza A virus have been studied extensively (27, 28, 57–67), yet such nucleotide signals have not been described for influenza B virus. Using an approach that added various lengths of the 3′- and 5′-terminal coding region of influenza B virus HA to a GFP gene, we generated an IBV HA(228)GFP(108) vRNA construct that was routinely incorporated into influenza B virus virions in the presence of exogenously expressed HA protein (Fig. 5A). Using reverse genetics plasmids and influenza B virus-infected MDCK-HA cell lines, we rescued a sciIBV. To assess the ability of influenza A or B virus to incorporate a heterotypic transgene, we tested if sciIAV could incorporate the influenza B virus HA(228)GFP(108) vRNA or if sciIBV could incorporate the influenza A virus HA(45)GFP(80) vRNA (Fig. 5B). While the RdRp of the influenza A or B viruses could replicate and transcribe GFP from either type A or type B influenza virus vRNAs (Fig. 5C), sciIAV could incorporate only the influenza A virus HA(45)GFP(80) construct and sciIBV could incorporate only the influenza B HA(228)GFP(108) construct when rescue transfection TCSs were passaged onto fresh MDCK-HA cells (Fig. 5D). This observation was verified by five independent experiments, each including triplicate transfections (Fig. 5E). These results suggest that packaging signals play an important role in the specific incorporation of type A or B influenza virus vRNAs into nascent virions.

To query if cis-acting packaging signals from influenza A virus were sufficient for incorporation of influenza B vRNAs into influenza A viruses, we performed gain-of-function experiments (Fig. 6). Using a strategy similar to that used to generate sciIAV (26, 47), we constructed chimeric influenza A/B virus genomic segments, where the entire ORF of B/Yam HA or NA replaced that of influenza A virus, except that the type A packaging signals and NCRs were left unchanged (Fig. 6A). Plasmids encoding the chimeric influenza A/B virus segments with the remaining PR8 reverse genetics plasmids were used to rescue recombinant virus containing HA, NA, or HA plus NA of influenza B virus (here referred to as PR8 BHA, BNA, and BHANA, respectively). The identities of the replication-competent, plaque-purified recombinant viruses were confirmed by RT-PCR (data not shown) and immunofluorescence of infected cells (Fig. 6B).

With the rescued type A viruses containing the glycoproteins of influenza B virus, we sought to characterize any effects on virus fitness. PR8 BHA and BNA have replication kinetics similar to those of rWT PR8, but PR8 BNA peak titers were moderately lower. In contrast, PR8 BHANA displayed slightly delayed kinetics comparable to those of rWT B/Yam, which grows slower than PR8 (Fig. 6C). Analysis of the plaque size phenotype provided results that agree with the observations presented above (Fig. 6D). Both the production of recombinant chimeric influenza virus and the efficient in vitro replication of PR8 BHA, BNA, and BHANA indicate that manipulation of influenza A virus packaging signals is sufficient to incorporate an influenza B viral segment with an homologous function and can allow the generation of intertypic reassortant viruses.

To evaluate the ability of chimeric influenza A/B viruses with recombinant glycoproteins to thrive in vivo, we sought to first test the morbidity induced by intranasal infection of mice (Fig. 7). A dose of 10⁵ PFU was selected for all viruses except PR8 (10 PFU), because of the mild pathogenicity of influenza B virus in C57BL/6 mice (68). The morbidity phenotype of PR8 BHANA mirrored that of B/Yam, as determined by body weight loss (<5%), and infection with PR8 BHANA did not lead to any mortality, whereas infection with PR8 BHA led to moderate weight loss (>15%). In contrast, PR8 BNA at the same dose was fatal to mice, with a calculated MLD₅₀ of 50 PFU (Fig. 7B), which is only twice that of rWT PR8 (MLD₅₀, ∼25 PFU) (68) and suggests that PR8 HA is a major pathogenicity factor in mice, similar to the results found for the pandemic 1918 Spanish influenza virus (69).

To demonstrate the immunogenicity of influenza A/B viruses with recombinant glycoproteins, systemic IgG antibodies were evaluated in mice infected with sublethal doses of virus. As expected, chimeric and rWT viruses induced comparable amounts of total anti-influenza A antibodies, as determined by ELISA (Fig. 7C). This outcome is perhaps due to saturation of the assay by non-HA and non-NA antibodies, such as anti-NP antibodies (70). When evaluating anti-influenza B virus antibodies, we found that only chimeric viruses carrying influenza B virus HA (PR8 BHA and PR8 BHANA) induced detectable levels of total anti-influenza B virus antibodies (Fig. 7D), suggesting that type B virus NA was not strongly immunogenic in this model. To validate our approach as a potential vaccine platform, we showed that antibodies from mice infected with influenza A/B viruses with recombinant glycoproteins had potent HAI activity (71) against homologous virus (Table 2). These strategies, coupled with intrinsic and further attenuating mutations, such as those found in the live attenuated influenza vaccine (72), could be synergized to generate a safe, immunogenic, multivalent vaccine against type A and B viruses using a single type A vaccine virus backbone.

Reassortment occurs readily within the different influenza virus types; thus, the ability of chimeric influenza A/B virus segments to move between homotypic genetic backbones during coinfection was analyzed (Fig. 8). To both assess reassortment ability and verify that chimeric influenza A/B glycoproteins can also function in pH1N1 virus, we coinfected MDCK cells with equivalent amounts of rWT pH1N1 and PR8 BHANA (MOI, 3). At 16 h postinfection, TCS was collected to isolate individual virus clones by plaque assay and to identify reassortment events using IFA against influenza B virus HA or NA or influenza A virus HA, NA, NP, or NS1 (Fig. 8A). Viruses that possessed either the influenza B virus HA or the influenza B virus NA in the context of novel constellations of PR8 or pH1N1 segments were identified (Fig. 8B). In agreement with the generation of PR8 BHA and PR8 BNA, we observed the compatibility of influenza B virus HA or NA with pH1N1, suggesting that the enzymatic and functional activities of these proteins were functional. For instance, one of the isolated viruses contained the viral HA from pH1N1, while NP, NA, and NS were derived from PR8 BHANA (phenotype 1). Another viral isolate contained similar segment constellations, but in this case the NS segment was derived from pH1N1, as determined by the lack of recognition with the PR8 NS1 MAb 1A7 (phenotype 2) (73). Finally, a virus containing HA, NP, and NS of PR8 HANA and NA from pH1N1 was also identified (phenotype 3). Altogether, these results suggest that the packaging signals of influenza A virus are sufficient in granting intertypic reassortment when the genes, including those from influenza B viruses, possess homologous functions and are at least in part responsible for the lack of reassortment between influenza A and B viruses.