Avian Influenza A Virus Polymerase Association with Nucleoprotein, but Not Polymerase Assembly, Is Impaired in Human Cells during the Course of Infection ^▿

Cell and viruses.

293T and DF1 cells were grown in complete Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal calf serum (FCS). QT6 cells were grown in Ham-F10 medium supplemented with 10% FCS, 1% chicken serum, and 2% tryptose phosphate broth. MDCK cells were grown in modified Eagle's medium supplemented with 5% FCS.

Influenza viruses A/Paris/908/97 ([P908] H3N2) and A/Mallard/Marquenterre/Z237/83 ([MZ] H1N1) were isolated by the National Influenza Center (northern France) at the Pasteur Institute in Paris (France). Viruses P908 and MZ were propagated in MDCK cells and embryonated hen eggs, respectively.

Plaque assays.

Unless indicated otherwise, P908- and MZ-derived viruses were titrated on MDCK and QT6 cells, respectively, using a plaque assay procedure adapted from Matrosovich et al. (21). When MDCK cells were used, the overlay was prepared with 1.2% Avicel (RC581; FMC BioPolymer) in complete minimal essential medium (MEM) and l-(tosylamido-2-phenyl) ethyl chloromethyl ketone (TPCK)-trypsin (Worthington) at a final concentration of 1 μg/ml. When QT6 cells were used, the overlay was prepared with 0.3% Avicel, 25% MEM, 75% Ham-10 medium, and TPCK-trypsin at a final concentration of 0.25 to 0.4 μg/ml.

Plasmids.

The four plasmids containing the promoter of Pol I (pPol I) encoding the sequences corresponding to the hemagglutinin (HA), neuraminidase, M, and NS genomic segments of A/WSN/33 (WSN) virus (8) were kindly provided by G. Brownlee (Sir William Dunn School of Pathology, Oxford, United Kingdom). Plasmids pHMG-PB1, -PB2, -PA, and -NP, encoding the PB1, PB2, PA, and NP proteins of influenza viruses P908 and MZ have been described earlier (16). The PB1, PB2, PA, and NP cDNAs were amplified using the recombinant pHMG plasmids as templates and primers specific for the coding sequences that contained additional nucleotides corresponding to the 3′ noncoding region (NCR) and 5′ NCR of the PB1, PB2, PA, and NP genomic segments, respectively, of the WSN virus. The amplicons were cloned between the two BbsI sites of the pPR7 plasmid (1), resulting in pPR7-P908-PB1, pPR7-P908-PB2, pPR7-P908-PA, and pPR7-P908-NP plasmids on the one hand, and pPR7-MZ-PB1, pPR7-MZ-PB2, pPR7-MZ-PA, and pPR7-MZ-NP plasmids on the other hand. The following sequences were fused to the last amino acid of the PB2 open reading frame (ORF) into the pPR7-P908-PB2 and pPR7-MZ-PB2 plasmids using standard PCR and cloning procedures: (i) a linker sequence encoding three Ala residues, (ii) the HA tag (YPYDVPDYA) or the One-Strep tag (WSHPQFEKGGGSGGGSGGGSWSHPQFEK) coding sequence, (iii) a stop codon combined with an NheI restriction site (TAGCATGC), and (iv) the 5′ terminal 143 nucleotides of the PB2 segment. Alternatively, the One-Strep tag sequence followed by a linker sequence encoding three Ala residues was inserted upstream of the PB2 ORF into the pPR7-P908-PB2 and pPR7-MZ-PB2 plasmids. The E627K mutation was introduced in the pPR7-MZ-PB2 plasmids by site-directed mutagenesis using a QuickChange Site-Directed Mutagenesis kit (Stratagene). All constructs were verified by sequencing positive clones using a Big Dye terminator sequencing kit and an automated sequencer (Perkin Elmer). The sequences of the oligonucleotides used for amplification and sequencing can be provided upon request.

Production of recombinant viruses by reverse genetics.

The method used for the production of recombinant influenza viruses by reverse genetics was adapted from previously described procedures (8). Briefly, the four pPol I WSN plasmids, together with the four pPR7-P908 or pPR7-MZ plasmids (including pPR7-PB2, pPR7-PB2-HA, pPR7-PB2-Nstrep [where Nstrep indicates the presence of the One-Strep tag sequence at the N terminus], or pPR7-PB2-Cstrep [carrying the One-Strep tag at the C terminus]) and the corresponding four pHMG-P908 or pHMG-MZ plasmids (0.5 μg of each), were cotransfected into a subconfluent monolayer of cocultivated 293T and MDCK cells (4 × 10⁵ and 3 × 10⁵ cells, respectively, seeded in a 35-mm dish), using 10 μl of the Fugene 6 transfection reagent (Roche). After 24 h of incubation at 37°C, the supernatant was removed, and cells were washed twice with DMEM and incubated at 37°C in DMEM containing TPCK-trypsin at a final concentration of 0.5 μg/ml for 48 h. The efficiency of reverse genetics was evaluated by titrating the supernatant on MDCK and QT6 cells in plaque assays.

Subsequent amplifications of the P908-WSN recombinant virus and derivatives were performed on MDCK cells at a multiplicity of infection (MOI) of 0.001 or 0.0001 for 2 to 3 days at 35°C in MEM containing TPCK-trypsin at a concentration of 1 μg/ml. The MZ-WSN recombinant virus and derivatives were amplified on DF1 cells at an MOI of 0.001 or 0.0001 for 2 to 3 days at 37°C in DMEM supplemented with 2.5% FCS. The exact conditions of amplification used for a given virus can be provided upon request.

vRNA extraction, RT-PCR, and sequence analysis.

Viral RNA was prepared from all viral stocks using a QIAamp Viral RNA Mini kit (Qiagen) according to the manufacturer's recommendations. Five microliters of vRNA was subjected to reverse transcription PCR (RT-PCR) and amplification using a Titan One-Step RT-PCR kit (Roche). The PB1, PB2, PA, and NP segments of the P908-WSN and MZ-WSN viruses and the PB2 segments of the tagged viruses and of a PB2 mutant virus carrying a Glu-to-Lys substitution at residue 627 (PB2-627) were amplified as one single amplicon, using specific oligonucleotides anchored on the 3′ and 5′ terminal 22 to 38 nucleotides of the segment, or as two overlapping amplicons, in which case an additional pair of internal oligonucleotides was used. The products of amplification were purified using a QIAquick gel extraction kit (Qiagen) and sequenced using a Big Dye terminator sequencing kit and an automated sequencer (Perkin Elmer). The sequences of the oligonucleotides used for amplification and sequencing can be provided upon request.

Purification of the PB2 proteins fused to the One-Strep tag (PB2-Strep) on Strep-tactin beads.

At 6 h following infection of 293T cells (2 ×10⁷ cells in 100-mm dishes) or DF1 cells (5 × 10⁶ cells in 100-mm dishes) with the PB2-wild type (wt) and PB2-Cstrep or PB2-wt and PB2-Nstrep viruses at an MOI of 2, cells were washed twice with phosphate-buffered saline (PBS), collected with a cell scraper, and centrifuged at 450 × g for 5 min. The packed cell volume was estimated. Cells were resuspended in five packed cell volumes of a lysis buffer containing 50 mM Tris-HCl, pH 8, 300 mM NaCl, 2 mM MgCl₂, 0.3% NP-40, 1 mM dithiothreitol, 10% glycerol, 1× protease inhibitor cocktail (Sigma), and 0.25 μ/ml of benzonase nuclease (Novagen). Cell lysates were incubated on a wheel for 1 h at 4°C and then centrifuged at 16,000 × g for 20 min. The supernatant was transferred to a fresh tube, 1/20 of the volume was frozen at −80°C, and the remaining volume was incubated on a wheel overnight with 1 mg of Strep-tactin beads (IBA GmbH). The beads were washed as recommended by the supplier, and bound proteins were eluted in 50 μl of Laemmli buffer, preheated at 95°C for most experiments, or in 50 μl of elution buffer containing 2.5 mM desthiobiotin (IBA GmbH) when the Nstrep viruses were used.

Western blot assays.

Cell lysates and Strep-tactin elution samples were analyzed by electrophoresis on a denaturing NuPAGE 4 to 12% Bis-Tris gel (Invitrogen), followed by Western blotting using polyvinylidene difluoride membranes. The membranes were incubated overnight at 4°C with a mouse monoclonal antibody directed against the C-terminal domain of RNA Pol II (diluted 1:1,000; 8WG16; Covance) or with rabbit polyclonal antibodies directed against the PB1, PB2, or PA proteins (diluted 1:5,000; kindly provided by J. Ortin, Centro Nacional de Biotecnologia, Madrid, Spain) or against A/PR/8/34 virions (37) (diluted 1:10,000 for the detection of NP and M1 proteins) in PBS with 1% bovine serum albumin, and 0.25% Tween 20. Membranes were then incubated for 1 h at room temperature with peroxidase-conjugated secondary antibodies (GE Healthcare), and with the ECL+ substrate (GE Healthcare). After the membranes were scanned using a G-Box (SynGene), the signals corresponding to the PB2, PB1, PA, NP, or RNA Pol II proteins were quantified using the GeneTools software (SynGene).

Indirect immunofluorescence assay.

Semiconfluent 293T cells on coverslips were infected with the PB2-wt or PB2-HA recombinant viruses at an MOI of 2. At 2 and 6 h postinfection (p.i.), cells were fixed with PBS-4% paraformaldehyde for 20 min, permeabilized with PBS-0.1% Triton X-100 for 10 min, and then incubated with either the 16B12 anti-HA (Covance) or IA-52 anti-NP (Oxford Biotechnology) monoclonal antibody, diluted 1:200. After subsequent incubation with AlexaFluor 555-coupled anti-mouse immunoglobulin G and AlexaFluor 488-coupled anti-rabbit immunoglobulin G secondary antibodies (diluted 1:500; Molecular Probes), the samples were analyzed under a fluorescence microscope (Leica).

Production of reassortant influenza viruses with core proteins derived from an avian or a human influenza virus isolate.

Recombinant viruses expressing the core proteins (PB1, PB2, PA, and NP) derived from the human isolate P908 or from the avian isolate MZ in a WSN virus background were produced by reverse genetics, as described in the Materials and Methods section. The resulting P908-WSN and MZ-WSN viruses were amplified on MDCK (mammalian: canine) and DF1 (avian: chicken) cells, respectively. vRNA was extracted and subjected to RT-PCR using primers designed to amplify the whole PB1, PB2, PA, and NP segments. Direct sequencing of the PCR products demonstrated that the PB1, PB2, PA, and NP coding sequences were unchanged in the P908-WSN and MZ-WSN reassortant viruses compared to the P908 and MZ parental viruses, respectively (data not shown).

MDCK cells were infected at an MOI of 0.001 and incubated at 37, 35, or 33°C for 72 h. Virus titers in the culture supernatant were determined at different times p.i. by plaque assays. Viral multiplication was only moderately affected by temperature variations for P908 and P908-WSN viruses (Fig. 1A), whereas it was strongly delayed at low temperatures in the case of MZ and MZ-WSN (Fig. 1B), in agreement with our previous observations (16, 20). The growth curves of P908 and P908-WSN viruses were very similar: maximum titers were observed at 48 h p.i. with a 2-fold (at 37°C) to 13-fold (at 33°C) difference in favor of P908 (Fig. 1A). In contrast, at 48 h p.i., the MZ virus grew to titers 5-fold (at 37°C) to 10-fold (at 33°C) lower than those of the reassortant MZ-WSN virus (Fig. 1B). As a consequence, the growth advantage of P908 over MZ was attenuated when P908-WSN and MZ-WSN were compared (e.g., at 48 h p.i. the difference in titers between the reassortant viruses compared to the difference between the parental viruses was about 1 log versus 2 log at 37°C, 2 log versus 4 log at 35°C, and 4 log versus 6 log at 33°C). Still, the P908-WSN virus conserved a clear growth advantage over MZ-WSN.

Host range and temperature sensitivity of the parental and reassortant viruses were further examined by performing plaque assays at 37 or 33°C on MDCK and QT6 (avian: quail) cells in parallel. Plaques formed by the avian-derived MZ and MZ-WSN viruses were reduced in number and in size on MDCK (Fig. 1D) compared to QT6 cells (Fig. 1F). Sensitivity to low temperatures of these viruses was very apparent on MDCK cells, as almost no visible plaques were formed at 33°C even when lower dilutions of the viral stock were tested (Fig. 1D). This was not the case on QT6 cells, where plaque size was only slightly reduced at 33 compared to 37°C (Fig. 1F). Unlike what was observed with MZ and MZ-WSN viruses, plaque formation by the human-derived P908 and P908-WSN viruses was more efficient on MDCK (Fig. 1C) than on QT6 cells (Fig. 1E). On MDCK cells, sensitivity to low temperatures of the P908 and P908-WSN viruses was very moderate compared to their MZ counterparts (compare Fig. 1C and D) although slightly more pronounced for the P908-WSN than for the P908 virus (Fig. 1C). On QT6 cells, sensitivity to low temperatures of the P908 and P908-WSN viruses was also moderate (Fig. 1E), as observed for the MZ and MZ-WSN viruses (Fig. 1F).

Overall, the P908-WSN and MZ-WSN reassortant viruses showed growth properties close to the parental P908 and MZ viruses, respectively. This similarity pointed to the fact that the differences between P908 and MZ with respect to growth potential and temperature sensitivity on mammalian cells were determined mainly by the RNP genes and validated the use of the reassortant viruses as a tool to further investigate the molecular mechanisms involved.

Production of influenza viruses expressing a PB2 protein fused to the One-Strep purification tag.

The PB2 segment was further engineered in order to generate P908-WSN and MZ-WSN viruses expressing a PB2 protein fused to the One-Strep affinity purification tag, a polypeptide which binds specifically to a derivative of streptavidin called Strep-tactin (32). The sequence encoding the One-Strep tag was fused to the sequence encoding either the C terminus (Fig. 2A) or the N terminus of PB2 (Fig. 2B). In the former case, a duplication of the last 109 nucleotides encoding PB2 was inserted between the stop codon of the PB2-Strep sequence and the 5′ NCR in order to ensure the conservation of a packaging signal overlapping the coding and noncoding regions at the 5′ end of the segment, as described earlier (4) (Fig. 2A). The resulting P908-Cstrep-WSN and P908-Nstrep-WSN viruses were amplified on MDCK cells. The MZ-Cstrep-WSN and MZ-Nstrep-WSN viruses were amplified on DF1 cells. vRNA was extracted and subjected to RT-PCR using primers designed to amplify the whole PB2 segment. Direct sequencing of the PCR products confirmed that the One-Strep coding sequence was present, as well as the PB2-ORF duplication in the case of Cstrep viruses (data not shown).

Growth properties of the viruses expressing a tagged PB2 or a wt PB2 protein were compared at 37°C. The P908-Cstrep-WSN and P908-WSN viruses showed the same kinetics of multiplication following infection of MDCK cells at a low MOI (0.001) (Fig. 2C). The MZ-Cstrep-WSN virus grew at a slower rate than MZ-WSN and achieved much lower titers on MDCK cells (Fig. 2C), whereas both viruses showed similar growth curves on QT6 cells (Fig. 2E). Thus, the presence of the One-Strep tag at the C terminus of PB2 did not alter the fitness of the virus per se but caused a growth defect specifically in mammalian cells. The growth defect of the MZ-Cstrep-WSN virus on MDCK cells was also observed at 40°C (data not shown). Unlike the MZ-Cstrep-WSN virus, the MZ-Nstrep-WSN virus showed a growth curve similar to that of the MZ-WSN virus on MDCK cells (Fig. 2D).

Expression levels of the PB2, PB1, NP, and M1 proteins were examined in MDCK and QT6 cells infected at a high MOI with the PB2-Cstrep or PB2-Nstrep viruses or with the PB2-wt viruses as a control. Total cell extracts were prepared at 6 h p.i and were analyzed by Western blotting. Both the PB2-Strep and PB2-wt proteins were revealed when an anti-PB2 polyclonal antibody was used, whereas a monoclonal antibody specific for the One-Strep tag reacted exclusively with the PB2-Strep proteins (Fig. 3). The overall levels of PB2, PB1, NP, and M1 proteins were reduced in cells infected with the MZ-WSN viruses compared to their P908-WSN counterparts. This reduction was more pronounced in MDCK (Fig. 3) than in QT6 infected cells (Fig. 3). The presence of a One-Strep tag at the C terminus of PB2 had little effect on the accumulation of viral proteins, except for a fourfold increase of NP accumulation in cells infected with the P908-Cstrep-WSN compared to the P908-WSN virus (Fig. 3; also data not shown for the estimated fourfold difference based on Western blot analysis of serial dilutions of both types of cell extracts). Remarkably, the PB2, PB1, NP, and M1 proteins were detected at similar levels in MDCK cells infected with the MZ-Cstrep-WSN and MZ-WSN viruses (Fig. 3) (M1 signals were detectable only after a longer exposure time [data not shown]). This observation suggests that the reduced growth of the MZ-Cstrep-WSN virus observed upon multicycle multiplication on MDCK cells (Fig. 2C) does not correspond to a major interference of the Cstrep tag with the Pol activity at an early stage of the viral cycle, and it contributes to the validation of the MZ-Cstrep-WSN virus as an appropriate tool in the present study. The presence of a One-Strep tag at the N terminus of PB2 had no significant effect on the accumulation of the NP and M1 proteins but led to slightly decreased levels of the PB2 and PB1 proteins in cells infected with P908-Nstrep-WSN as well as MZ-Nstrep-WSN (Fig. 3) (M1 signals were detectable only after a longer exposure time [data not shown]). When the same comparative Western blot analysis was performed for 293T and DF1 cells infected at a high MOI, the results were similar to those obtained for MDCK and QT6 cells, respectively (data not shown).

An indirect immunofluorescence assay was used to analyze 293T cells at 2 and 6 h following infection at high MOIs with MZ-WSN or P908-WSN viruses expressing an HA-tagged PB2 protein. Striking differences in the intensity and cellular localization of NP staining were observed. In P908-HA-WSN-infected cells, the NP protein was nuclear at 2 h p.i. (Fig. 4, columns C and D, row a) and mainly cytoplasmic at 6 h p.i. (Fig. 4, columns C and D, row c). In MZ-HA-WSN-infected cells, the NP-specific fluorescence was almost undetectable at 2 h p.i. (Fig. 4, columns C and D, row b) and limited to the nucleus at 6 h p.i. (Fig. 4, columns C and D, row d). At both time points, the PB2-HA protein was detected mainly in the nucleus of infected cells (Fig. 4, columns A and B, rows a to d), and the signal was much stronger in P908-HA-WSN-infected cells (as a consequence, different settings had to be used for P908-HA-WSN and MZ-HA-WSN samples shown in Fig. 4, columns A and B). These observations were in agreement both with the delayed viral multiplication (Fig. 1) and reduced accumulation of PB2 and NP proteins (Fig. 3) observed upon infection of MDCK cells with the MZ-WSN compared to the P908-WSN virus.

We next asked whether the replicative potential of MZ-WSN and P908-WSN viruses could be correlated to the efficiency with which the RNP components interacted with each other and with cellular partners, by performing copurification experiments on cells infected with the corresponding PB2-Strep viruses. Because reagents are more readily available for human and chicken cellular proteins, these experiments were set up on 293T and DF1 cells.

Copurification of viral core proteins from human or avian cells infected with PB2-Strep viruses.

We first compared how efficiently the P908 and MZ viral Pols assembled and interacted with the NP in 293T and DF1 cells. Confluent cell monolayers were infected at a high MOI with the P908- or MZ-Cstrep-WSN viruses in parallel with the control P908- or MZ-WSN viruses and incubated at 37°C or 33°C. Total cell extracts were prepared at 6 h p.i. as described in the Materials and Methods section and were incubated with Strep-tactin beads. After the beads were washed, PB2-Strep complexes were eluted using Laemmli buffer, separated by electrophoresis on a denaturing gel, and analyzed by Western blotting using antibodies specific for the viral PB2, PB1, PA, and NP proteins.

For each of the Pol subunits, protein levels upon infection with the MZ-Cstrep-WSN or P908-Cstrep-WSN virus were compared in total lysates prepared from infected 293T or DF1 cells and in the corresponding Strep-tactin eluates. For the purpose of quantification, serial dilutions of the pre- and postpurification samples were analyzed using the G-Box and GeneTools software (Syngene). Very reproducible results were obtained in three independent experiments. Similar observations were made at 37°C and 33°C. In total cell lysates, for each of the Pol subunits, the signal ratios of MZ-Cstrep-WSN to P908-Cstrep-WSN were about 1:4 to 1:8 in 293T cells (Fig. 5E and G) and 1:1 in DF1 cells (Fig. 5F and H). In Strep-tactin eluates derived from 293T infected cell extracts, the protein ratios of MZ-Cstrep-WSN to P908-Cstrep-WSN were still about 1:4 to 1:8 for PB2-Strep as well as for the PB1 and PA subunits (Fig. 5A, E, and G). That the ratios before and after purification for all three subunits were the same suggests that assembly of the MZ Pol complex is not impaired in 293T cells although each subunit is underexpressed. Similarly, the 1:1 protein ratios of MZ to P908 in DF1 samples was conserved throughout the purification process (Fig. 5B, F, and H), suggesting that both viral Pols are being assembled with the same efficiency in avian cells.

Analysis of the levels of NP protein upon infection with MZ- and P908-Cstrep-WSN viruses also showed marked differences in 293T but not in DF1 cells. In DF1 cell lysates analyzed prior to purification, similar NP signals were detected whether the cells had been infected with the MZ- or the P908-Cstrep-WSN virus (Fig. 5D, cell lysates, F, and H). Subsequently, similar levels of MZ- and P908-NP appeared to copurify with MZ- and P908-PB2-Cstrep, respectively (Fig. 5D, eluates, F, and H). In contrast, 293T cells infected with the MZ-Cstrep-WSN virus showed significantly reduced NP levels compared to P908-Cstrep-WSN-infected cells. The difference was about 16- to 32-fold when cells had been incubated at 37°C and 64- to 128-fold when cells were incubated at 33°C (Fig. 5C, lysates, E, and G). The relatively high levels of NP in P908-Cstrep-WSN-infected cells (Fig. 3 and related comments) accounted in part for the broad difference. When this corrective factor was taken into account, the reduction of MZ-NP levels was still repeatedly more pronounced than the reduction of MZ-Pol levels in 293T cells incubated at 33°C. After Strep-tactin purification, the estimated ratios of MZ-NP to P908-NP were about 1:256 and <1:512 when 293T cells were incubated at 37 and 33°C, respectively (Fig. 5C, eluates, E, and G). These lower ratios compared to those observed in total cell extracts (about 1:32 and 1:128 at 37 and 33°C, respectively, as mentioned above) raised the possibility of a specific defect of the association between the MZ-NP and MZ-Pol in 293T cells. Whether this defect was more pronounced at 33°C than at 37°C could not be established because of weak MZ-NP signals in 293T cells infected at 33°C (Fig. 5G).

Copurification experiments were repeated using the MZ- and P908-WSN viruses tagged at the N terminus instead of the C terminus of PB2. When protein ratios were compared in Strep-tactin eluates (Fig. 6) and in total cell extracts (Fig. 6), the ratios of MZ-PB2 to P908-PB2 and of MZ-PB1 to P908-PB1 appeared essentially unchanged. In contrast, the ratio of MZ-NP to P908-NP was decreased at least twofold in Strep-tactin eluates compared to total cell extracts. The extent to which the NP ratio was decreased could not be evaluated precisely because the signals corresponding to MZ-NP were at background levels (Fig. 6). The overall signal intensity in experiments with Nstrep viruses was low compared to Cstrep viruses due to low levels of expression (Fig. 3) and purification yield (data not shown) of PB2-Nstrep compared to PB2-Cstrep proteins. However, the data obtained with Nstrep viruses indicated a defect of the association between MZ-NP and MZ-Pol in 293T cells and were in agreement with those obtained with Cstrep viruses.

The impact of a Glu-to-Lys substitution at residue 627 of MZ-PB2 on the efficiency of copurification of the viral core proteins was finally examined. Avian-like MZ-WSN and MZ-Cstrep-WSN viruses with a PB2-627 Glu-to-Lys substitution (627K) were produced. A copurification experiment was performed on cell extracts prepared from 293T cells infected at a high MOI with the MZ-627K-Cstrep-WSN, MZ-Cstrep-WSN, or P908-Cstrep-WSN virus in parallel with the control MZ-627K-WSN, MZ-WSN, or P908-WSN virus. No reduction of the ratio of MZ-NP to P908-NP in Strep-tactin eluates compared to total cell extracts was observed upon infection with the MZ-627K-Cstrep-WSN virus, unlike with the MZ-Cstrep-WSN virus (a fourfold reduction) (Fig. 7A). This observation indicated that the presence of a Lys instead of a Glu at PB2 residue 627 fully corrected the defect of NP association with the Pol in MZ-Cstrep-WSN-infected cells. Following infection of MDCK cells at a low MOI (0.001), the mutant MZ-627K-WSN and MZ-627K-Cstrep-WSN viruses grew more efficiently than their wt counterparts (Fig. 7B). Notably, the mutant MZ-627K-Cstrep-WSN still showed delayed growth kinetics compared to the untagged mutant MZ-627K-WSN virus, which strongly suggests that the growth defect due to the presence of the Cstrep tag was not corrected by the Glu627Lys substitution and, thus, was unrelated to the defect of NP-Pol association.

Copurification of viral Pol and cellular RNA Pol II from human or avian cells infected with PB2-Cstrep viruses.

Engelhardt et al. showed the existence of a physical association between the influenza WSN virus Pol and the large subunit of the cellular RNA Pol II in a transcriptionally active, hyperphosphorylated form (7). Here, we compared how efficiently the P908 and MZ viral Pols interacted with the Pol II enzyme in 293T and DF1 cells. Total cell lysates were analyzed by Western blotting using the monoclonal antibody 8WG16, which allows detection of two forms of the Pol II large subunit C-terminal domain corresponding to different phosphorylation states of the protein (7). In 293T as well as in DF1 cells, both forms of Pol II were detected at similar levels whether the cells had been mock infected or infected with the MZ-Cstrep-WSN or the P908-Cstrep-WSN virus (data not shown). Following purification on Strep-tactin beads, some Pol II was detected specifically in the samples corresponding to cells infected with PB2-Cstrep viruses (Fig. 5A and B), and this protein was found to correspond to the slow mobility form (data not shown). Comparison of the signal ratios of PB2/PB1/PA to Pol II indicated that the human Pol II enzyme (Fig. 5A, E, and G) as well as the chicken Pol II enzyme (Fig. 5B, F, and H) copurified more efficiently with the MZ Pol than with the P908 Pol. Depending on the experiments, the ratio of Pol II to PB1 was two- to eightfold higher in cells infected with MZ-Cstrep-WSN than with P908-Cstrep-WSN. Overall, these observations clearly indicated that the replication defect of the MZ virus in 293T cells was not related to an impaired binding of the Pol complex to the cellular Pol II.