Free access
Research Article
24 May 2021

Development of a Genetically Stable Live Attenuated Influenza Vaccine Strain Using an Engineered High-Fidelity Viral Polymerase


RNA viruses demonstrate a vast range of variants, called quasispecies, due to error-prone replication by viral RNA-dependent RNA polymerase. Although live attenuated vaccines are effective in preventing RNA virus infection, there is a risk of reversal to virulence after their administration. To test the hypothesis that high-fidelity viral polymerase reduces the diversity of influenza virus quasispecies, resulting in inhibition of reversal of the attenuated phenotype, we first screened for a high-fidelity viral polymerase using serial virus passages under selection with a guanosine analog ribavirin. Consequently, we identified a Leu66-to-Val single amino acid mutation in polymerase basic protein 1 (PB1). The high-fidelity phenotype of PB1-L66V was confirmed using next-generation sequencing analysis and biochemical assays with the purified influenza viral polymerase. As expected, PB1-L66V showed at least two-times-lower mutation rates and decreased misincorporation rates, compared to the wild type (WT). Therefore, we next generated an attenuated PB1-L66V virus with a temperature-sensitive (ts) phenotype based on FluMist, a live attenuated influenza vaccine (LAIV) that can restrict virus propagation by ts mutations, and examined the genetic stability of the attenuated PB1-L66V virus using serial virus passages. The PB1-L66V mutation prevented reversion of the ts phenotype to the WT phenotype, suggesting that the high-fidelity viral polymerase could contribute to generating an LAIV with high genetic stability, which would not revert to the pathogenic virus.
IMPORTANCE The LAIV currently in use is prescribed for actively immunizing individuals aged 2 to 49 years. However, it is not approved for infants and elderly individuals, who actually need it the most, because it might prolong virus propagation and cause an apparent infection in these individuals, due to their weak immune systems. Recently, reversion of the ts phenotype of the LAIV strain currently in use to a pathogenic virus was demonstrated in cultured cells. Thus, the generation of mutations associated with enhanced virulence in LAIV should be considered. In this study, we isolated a novel influenza virus strain with a Leu66-to-Val single amino acid mutation in PB1 that displayed a significantly higher fidelity than the WT. We generated a novel LAIV candidate strain harboring this mutation. This strain showed higher genetic stability and no ts phenotype reversion. Thus, our high-fidelity strain might be useful for the development of a safer LAIV.


RNA viruses demonstrate heritably heterogeneous populations, termed quasispecies, because of error-prone virus genome replication by viral RNA-dependent RNA polymerases (RdRp) (13). Although error-prone replication has a risk of producing lethal viruses, the diversified virus population is useful in evolutionary processes such as drug resistance (48). The crucial mechanism of error-prone replication by viral RdRp is the lack of proofreading ability during RNA genome replication (911). Studies on poliovirus and other RNA viruses have revealed that certain amino acid residues in RdRp are associated with its fidelity, which can be modulated by introducing changes in these amino acid residues (1216).
The genome of influenza A virus (IAV) comprises eight different viral RNA (vRNA) segments. The influenza viral polymerase is composed of polymerase basic protein 1 (PB1), polymerase basic protein 2 (PB2), and polymerase acidic protein (PA); of these, PB1 is involved in the polymerization of the RNA chain (1719). Amino acid residues that affect the fidelity of influenza viral polymerase have recently been identified. A Val43-to-Ile (V43I) mutation in PB1 has been reported to show increased selectivity to nucleotides in influenza viral polymerase (20, 21). We recently reported that a Tyr82-to-Cys (Y82C) mutation in PB1 induces an influenza virus mutator phenotype (13). Lin et al. also reported that a Ser216-to-Gly (S216G) mutation reduces the fidelity of influenza viral polymerase (22).
IAV is one of the most serious zoonotic pathogens that cause seasonal epidemics and periodic pandemics worldwide. Vaccination is the most useful approach to prevent infection and exacerbation postinfection. Influenza virus vaccines include inactivated and live vaccines. Although inactivated vaccines have been utilized for influenza prophylaxis for over 80 years (23), a live attenuated influenza vaccine (LAIV), commercially known as FluMist, has been approved for use in the United States since 2003 and in the European Union since 2011. The LAIV currently in use is administered intranasally, to mimic the natural mode of infection of the influenza virus. Therefore, the LAIV could establish local protection at the site of infection in the upper respiratory tract, resulting in the induction of more robust and long-lasting immunity than that induced by inactivated vaccines (24, 25). Hemagglutinin (HA) and neuraminidase (NA) are the major antigenic determinants of influenza viruses. LAIV includes the HA and NA segments of the currently prevalent virus strains. The other six gene segments, those encoding PB2, PB1, PA, nucleoprotein (NP), matrix proteins, and nonstructural proteins, are derived from the master donor virus (MDV) (26). The MDV of the current LAIV was generated by carrying out serial passages of the A/Ann Arbor/6/60 (A/AA/6/60) virus at low temperatures (26). Although the current LAIV grows at temperatures around the human body temperature (34°C to 37°C), the viral growth of LAIV is limited to 38°C and 39°C (i.e., is temperature sensitive [ts]). The amino acid mutations responsible for the ts phenotype have been identified as PB2-N265S, PB1-K391E/E581G/A661T, and NP-D34G, which are sufficient to impart the ts phenotype to the A/AA/6/60 virus (27). The ts phenotype of the current LAIV is stable following propagation in animals and humans (28). However, previous studies have demonstrated that mutations can be generated during the production of LAIV and in the course of viral propagation in vaccinated children (29, 30). Thus, the generation of mutations that facilitate pathogenesis in the current LAIV should be considered. In fact, a recent report demonstrated reversion of the ts phenotype of the LAIV strain currently in use to a pathogenic virus in cultured cells (31).
Although live attenuated vaccines for RNA viruses are effective, these can revert to virulence. Therefore, a high-fidelity live attenuated vaccine is important because suppressing error-prone replication could increase the genetic stability of the live vaccine. In fact, high-fidelity viral polymerase has been shown to reduce the virulence of poliovirus; thus, restricting the diversity of virus quasispecies could be an effective strategy for the establishment of safer, live attenuated vaccines for many RNA viruses (16, 32). Thus, we hypothesized that high-fidelity influenza viral polymerase reduces the diversity of influenza virus quasispecies and inhibits the reversion of the ts phenotype introduced in the LAIV.
Ribavirin is a guanosine nucleotide analog that inhibits vRNA polymerase or acts as a mutagen, leading to lethal mutagenesis, thereby inhibiting RNA virus replication (3335). Ribavirin has been used to isolate high-fidelity mutants of poliovirus and other RNA viruses (15, 3639). In this study, we aimed to identify the amino acid residues in influenza viral polymerase that are involved in the fidelity of influenza virus replication by screening for a high-fidelity polymerase through serial virus passages under selection with ribavirin; we then aimed to apply this engineered polymerase to the LAIV.


Isolation of a mutant influenza virus with low sensitivity to ribavirin.

We generated a recombinant WSN/33 strain using a plasmid-driven reverse-genetics method (40) and checked its sensitivity to ribavirin. MDCK cells were infected with WSN/33 virus and cultured in a medium containing 0, 12.5, 25, 50, and 100 μM ribavirin. The supernatant was collected at 36 h postinfection, and the virus titer was determined using a plaque assay. Virus propagation in 50 μM ribavirin-treated cells was approximately 90% lower than that in untreated cells (Fig. 1A).
FIG 1 Isolation of a mutant less sensitive to ribavirin. (A) Sensitivity of the WSN/33 virus to ribavirin. MDCK cells were infected with WSN/33 virus at an MOI of 0.1 and then cultured in the presence or absence of ribavirin. At 36 h postinfection, the culture supernatant was recovered, and the virus titer was determined using a plaque assay. Each result is represented by a value relative to that in the absence of ribavirin. The graph indicates average values with standard deviations from three independent experiments. (B and C) Sensitivity of P1, P4, P8, and P12 viruses (B) and PB1-L66V mutant virus (C) to ribavirin. Experiments were carried out as described for panel A. Each result is represented by a value relative to that in the absence of ribavirin. The graph indicates average values with standard deviations from three independent experiments. P values are based on Student's t test (panel B, *, P = 0.0029; **, P = 0.0015) (panel C, *, P = 0.0045; **, P = 0.0221).
To isolate the ribavirin escape mutant, we serially passaged the WSN/33 virus in MDCK cells at a multiplicity of infection (MOI) of 0.01 to 0.05 PFU per cell from passages 1 to 8 (P1 to P8) with 50 μM ribavirin. After eight serial passages, the concentration of ribavirin was increased to 100 μM, and the virus was passaged four times to isolate a mutant that could replicate at a higher concentration of ribavirin (P9 to P12). We then examined ribavirin sensitivity in P1, P4, P6 and P12 viruses. The results showed that the P12 virus was less sensitive to 50 μM and 100 μM ribavirin (Fig. 1B, *, P = 0.0029 using Student's t test; **, P = 0.0015 using Student's t test).
Next, we performed a plaque assay to isolate ribavirin escape variants. MDCK cells were infected with P12 virus and then overlaid with a culture medium and agarose mixture containing 100 μM ribavirin. In total, six plaques were picked and replicated once without ribavirin using MDCK cells. All six clones were then subjected to Sanger sequencing, to determine the existence of a mutation in the PB1 gene that might be involved in viral polymerase fidelity. We found that the PB1-L66V mutation was successfully introduced into five of the six PB1 clones, whereas silent mutations were introduced into one PB1 clone. To examine whether the PB1-L66V mutation reduces sensitivity to ribavirin, we generated a recombinant WSN/33–PB1-L66V mutant using a plasmid-driven reverse-genetics method (40). MDCK cells were infected with the WSN/33–PB1-WT virus (PB1-WT) or WSN/33–PB1-L66V mutant virus (PB1-L66V) and cultured in medium containing 0, 12.5, 25, 50, and 100 μM ribavirin. PB1-L66V was confirmed to be less sensitive to ribavirin than PB1-WT (Fig. 1C; *, P = 0.0045 using Student's t test; **, P = 0.0221 using Student's t test).

Involvement of PB1-L66V in the high-fidelity phenotype.

The PB1 subunit of the influenza virus RdRp plays a crucial role in viral genome replication (17, 41). As L66 in the PB1 subunit is located in the putative vRNA-binding domain, which is highly conserved among IAVs (42), the L66V mutation in the PB1 subunit can be inferred to influence the viral polymerase activity. To test whether the PB1-L66V mutation of the WSN/33 virus affects the viral polymerase activity, we generated a PB1-L66V expression vector and transfected it into 293T cells along with other expression vectors for viral ribonucleoprotein (vRNP)- and vRNA-encoding firefly luciferase expression. Luciferase activity was measured at 24 h posttransfection. Both the level of polymerase activity (Fig. 2A) and the expression levels of PB1 protein (Fig. 2B) were similar between PB1-WT and PB1-L66V.
FIG 2 Characteristics of the PB1-L66V mutant. (A and B) Comparison of viral polymerase activities. 293T cells were cotransfected with viral polymerase subunits, NP, and model vRNA coding firefly luciferase expression vectors. Firefly luciferase activities were measured at 24 h posttransfection. (A) The luciferase activity of PB1-L66V was calculated relative to that of PB1-WT. The graph indicates average values with standard deviations from three independent experiments. (B) Transfected PB1 proteins were confirmed using anti-PB1 polyclonal antibody (top). β-Actin was detected as a loading control (bottom). pol (−), PB1 expression vector was not transfected. (C) Comparison of mutation frequencies. MDCK cells were infected with PB1-WT, PB1-L66V, or PB1-V43I viruses at an MOI of 10, and the total RNA from the infected cells was collected at 15 h postinfection. The cDNA of segment 5 or segment 8 was amplified using reverse transcription PCR. Then, DNA libraries were amplified using a second PCR with the Nextera XT index kit, which includes index tags for discrimination of the sequencing samples. DNA libraries were subjected to NGS, after which the ratio of mutation frequency was calculated. The graph indicates average values with standard deviations from three independent experiments. P values are based on Student's t test (*, P = 0.0064; **, P = 0.0149).
According to the findings with other RNA viruses, ribavirin-resistant mutants demonstrate increased viral polymerase fidelity during RNA replication (15, 3639). To confirm whether the PB1-L66V mutation also affects the fidelity of influenza viral polymerase, we compared the mutation rates between the PB1-WT and PB1-L66V viruses. In addition, we analyzed PB1-V43I as a control IAV of high fidelity, because it has previously been reported that the PB1-V43I virus exhibits a lower mutation rate (20, 21). MDCK cells were infected with PB1-WT, PB1-L66V, or PB1-V43I viruses, and total RNA was collected at 15 h postinfection. After reverse transcription of the vRNA, the cDNA of segment 5 or segment 8 was amplified using PCR. The number of mutations among these amplicons was determined using next-generation sequencing (NGS), followed by calculation of the mutation frequencies (Fig. 2C). The segment 5 gene mutation efficiencies of PB1-L66V and PB1-V43I decreased by 2.07-fold (Fig. 2C; *, P = 0.0064 using Student's t test) and 1.24-fold, respectively, relative to that of PB1-WT. Similarly, the segment 8 gene mutation efficiencies of both the PB1 variants decreased by 1.86-fold (PB1-L66V) (Fig. 2C; **, P = 0.0149 using Student's t test) and 1.21-fold (PB1-V43I) relative to that of PB1-WT. These results suggest that the PB1-L66V polymerase leads to higher genetic stability during viral genome replication than the PB1-WT polymerase and PB1-V43I polymerase, which is known as a high-fidelity polymerase.

Leu-to-Val change at position 66 in PB1 reduces nucleotide triphosphate misincorporation during viral RNA synthesis.

Next, we verified the involvement of the PB1-L66V mutation in the high-fidelity phenotype using biochemical assays. The fidelities of the viral polymerase were compared using a cell-free RNA synthesis system performed under a UTP-limited condition (referred to hereafter as a limited elongation assay), as described previously (13, 21). A 120-nucleotide (120-nt) genome (Fig. 3A) was used as the template in the limited elongation assay. In the absence of UTP, viral genome replication by the viral RdRp paused at the first adenine residue of the template and a 60-nt RNA product was synthesized (Fig. 3B, lanes 1 and 2, 60-nt band). However, misincorporation of the first adenine residue causes the viral RdRp to proceed to the second adenine residue, resulting in generation of a 109-nt band (Fig. 3B, lanes 1 and 2, 109-nt band). The fidelity of the viral polymerase can be determined by comparing the amounts of the 109-nt RNA product. In the presence of UTP, the amount of 120-nt RNA product derived from PB1-L66V was comparable to that from PB1-WT, indicating that the PB1-L66V mutation did not significantly affect the viral polymerase activity (Fig. 3B, lanes 3 and 4). In contrast, the amount of 109-nt RNA derived from PB1-L66V was lower than that from PB1-WT (Fig. 3B, lanes 1 and 2, 109-nt band). Quantitative results also showed that the synthesis of 109-nt RNA products derived from PB1-L66V polymerase was suppressed, compared to that from PB1-WT polymerase (Fig. 3C, middle panel; *, P = 0.0093 using Student's t test). Taken together, these results suggest that the PB1-L66V mutation conferred a high-fidelity phenotype to the influenza viral polymerase.
FIG 3 Biochemical assay to assess high-fidelity RNA synthesis by PB1-L66V. (A) Sequence of 120-nt model viral genome template for limited elongation assay. The adenine residue marked with an asterisk appears as the first template adenine base during RNA synthesis. (B) Illustration of the in vitro limited elongation assay. In the presence of ATP, GTP, CTP, and UTP, viral RdRp produces the full-length RNA product (120 nt). In the absence of UTP, the viral genome replication by viral RdRp paused at the first adenine residue of the template and a 60-nt RNA product was synthesized. Misincorporation of nucleotides at the first adenine residue causes the viral RdRp to proceed to the second adenine residue, resulting in the generation of a 109-nt band. The first and second adenines from the 3′ terminus are indicated by an asterisk and an underline, respectively. The result of the limited elongation assay is shown in the right panel. Assays were performed with the 120-nt model vRNA as a template and 40 ng of vRNP at 37°C for 1 h, in the absence (lanes 1 and 2) or presence (lanes 3 and 4) of UTP. (C) Quantitative results of high-fidelity RNA synthesis by PB1-L66V. Limited elongation assays were performed under the same conditions as described for panel B using 20 ng, 40 ng, and 60 ng of vRNP. The amounts of the RNA products (60 nt, 109 nt, and 120 nt long) derived from the PB1-L66V mutant were compared with those from PB1-WT. Band intensities (shown as arbitrary units) were determined after subtracting the background using ImageJ version, an image analysis software. Quantitative results are presented as averages with standard deviations from three independent experiments (circles, vRNP derived from PB1-WT; triangles, vRNP derived from PB1-L66V). Significance was determined using Student's t test (*, P = 0.0093).

Regulation of the fidelity of influenza virus replication by amino acid residue 66 in the PB1 subunit.

We hypothesized that L66 in the PB1 subunit plays a crucial role in the regulation of viral polymerase fidelity. Therefore, we generated mutant viruses in which leucine 66 of PB1 was replaced with other amino acids and then examined the fidelity of the PB1-L66 variants in comparison to that of the WT virus. Five recombinant viruses, PB1-L66D, PB1-L66H, PB1-L66P, PB1-L66W, and PB1-L66Y, were not found to propagate in MDCK cells; however, the propagation of other variants was observed (Table 1, “Virus titer” column, indicated as not detected [ND]), implying that these five mutations impaired the viral polymerase activity. We then determined the number of nucleotide mutations using NGS, as shown in Fig. 2C, and calculated the mutation frequencies in the PB1-L66 variants that were successively isolated. The mutation frequency of PB1-L66V (5.0 mutations per 105 nucleotides) was 2-fold lower than that of PB1-WT (10.1 mutations per 105 nucleotides). Furthermore, PB1-L66F (5.3 mutations per 105 nucleotides) and PB1-L66G (6.3 mutations per 105 nucleotides) also showed a tendency for a lower mutation frequency than PB1-WT (Table 1, “Mutational frequency” column). Interestingly, PB1-L66C (17.5 mutations per 105 nucleotides) and PB1-L66T (15.7 mutations per 105 nucleotides) exhibited higher mutation frequencies than PB1-WT. These results indicate that amino acid residue 66 in PB1 regulates the fidelity of influenza virus replication.
TABLE 1 Virus growth and mutation frequencies of PB1-L66 mutant viruses
VirusVirus titer (PFU/ml) (log10)Total no. of mutations identifiedTotal no. of nt sequencedMutational frequency
PB1-WT6.94524,492,33310.1 × 10−5
PB1-L66V7.42605,234,2565.0 × 10−5
PB1-L66A7.35125,207,3999.8 × 10−5
PB1-L66C7.49095,191,64717.5 × 10−5
PB1-L66E7.35174,955,98510.4 × 10−5
PB1-L66F5.73005,640,7605.3 × 10−5
PB1-L66G5.82964,698,5136.3 × 10−5
PB1-L66I7.36044,861,67912.4 × 10−5
PB1-L66K6.34265,232,8438.1 × 10−5
PB1-L66M7.34154,038,98010.3 × 10−5
PB1-L66N6.95295,287,16410.0 × 10−5
PB1-L66Q7.24484,941,4459.1 × 10−5
PB1-L66R6.34944,687,81010.5 × 10−5
PB1-L66S7.35785,536,35710.4 × 10−5
PB1-L66T6.08305,281,30815.7 × 10−5
ND, not detected.
NT, not tested.

Engineering the WSN/33 strain possessing a ts phenotype based on the current LAIV.

PB2-N265S, PB1-K391E/E581G/A661T, and NP-D34G are reported to be involved in the ts phenotype of the current LAIV strain and sufficient to confer the ts phenotype to the A/AA/6/60 strain (27). Therefore, we tested whether these amino acid mutations confer the ts phenotype to the WSN/33 strain as well. Recombinant viruses containing PB2-N265S, PB1-K391E/E581G/A661T, and NP-D34G, with or without PB1-L66V mutations in the WSN/33 virus, were generated using a reverse-genetics approach and designated ts-PB1-L66V and ts-PB1-L66, respectively. MDCK cells were infected with WSN/33, ts-PB1-L66, or ts-PB1-L66V viruses and incubated for 1 h at 34°C. Plaque assays were then carried out, and the cells were cultured for 3 days at 34°C or 38°C. The WSN/33 virus formed plaques at 34°C and 38°C. However, ts-PB1-L66 and ts-PB1-L66V did not form plaques at 38°C, whereas plaques were clearly visible at 34°C (Fig. 4A). These results indicated that the PB2-N265S, PB1-K391E/E581G/A661T, and NP-D34G mutations are effective at inducing the ts phenotype in the WSN/33 virus.
FIG 4 Generation of PB1-L66V mutant possessing temperature-sensitive phenotype. (A) Temperature sensitivity of ts-PB1-L66 and ts-PB1-L66V. MDCK cells were infected with WSN/33, ts-PB1-L66, or ts-PB1-L66V, and plaque assays were then carried out. Cells were cultured at 34°C or 38°C. Three days postinfection, the cells were stained with amido black 10B. (B) Comparison of viral growth kinetics between ts-PB1-L66 and ts-PB1-L66V viruses. MDCK cells were infected with ts-PB1-L66 and ts-PB1-L66V at an MOI of 0.001, and the culture supernatants were collected at 12, 24, 36, 48, and 60 h postinfection. The amount of virus in the culture supernatants at each collection time was measured using plaque assays. The results are presented as averages with standard deviations from three independent experiments (circles, ts-PB1-L66 virus; triangles, ts-PB1-L66V virus). (C) Comparison of mutation frequencies. MDCK cells were infected with the indicated viruses at an MOI of 10, and the total RNA in infected cells was collected 15 h postinfection. The amplicon of segment 8 gene derived from each virus sample was prepared as described in the legend to Fig. 2C and subjected to NGS. The graph indicates average values with standard deviations from three independent experiments. P values are based on Student's t test (*, P = 0.0090).
Next, we compared the viral growth kinetics between ts-PB1-L66 and ts-PB1-L66V. MDCK cells were infected with ts-PB1-L66 and ts-PB1-L66V at an MOI of 0.001, followed by recovery of the supernatants every 12 h. The amounts of virus in the culture supernatants at each collection time were then measured using plaque assays. There was no difference between the two groups in terms of viral growth (Fig. 4B).
Because the amino acid mutations that confer the ts phenotype are introduced in the PB1 subunit of the ts-PB1-L66V mutant virus, it is possible that these mutations affected the fidelity of the viral polymerase derived from PB1-L66V. To examine the effect of mutations introduced in the ts phenotype on fidelity, we analyzed the mutation frequencies of the ts-PB1-L66 and ts-PB1-L66V viruses (Fig. 4C). MDCK cells were infected with ts-PB1-L66 and ts-PB1-L66V viruses, and total RNA was collected at 15 h postinfection. After reverse transcription of the vRNA, the cDNA of segment 8 was amplified, and the amplicons were subjected to NGS. The difference in mutation frequency between ts-PB1-L66 and ts-PB1-L66V showed the same tendency as that between WSN/33 PB1-WT and PB1-L66V shown in Fig. 2C, suggesting that the mutations introduced to generate the ts phenotype had no effect on viral polymerase fidelity.

High-fidelity polymerase mutant PB1-L66V suppressed the reversion of the ts phenotype to the WT phenotype.

The strain of LAIV currently in use gradually adapted to nonpermissive temperature during serial passages in cultured cells, finally leading to the isolation of an LAIV strain in which the ts phenotype was abolished (31). Considering the improved fidelity of the ts-PB1-L66V virus introduced with the ts phenotype (Fig. 4C), we hypothesized that generation of the reversion mutation related to the ts phenotype is suppressed in ts-PB1-L66V. To prove this hypothesis, serial virus passages of ts-PB1-L66 and ts-PB1-L66V were carried out at a permissive temperature. MDCK cells were infected with ts-PB1-L66 and ts-PB1-L66V and cultured at the permissive temperature. When a cytopathic effect was observed, the supernatant was recovered and named passage 1 (P1). The virus passages were then continued until P25, and temperature sensitivity of the P1, P5, P10, P15, P20, and P25 viruses was checked at the nonpermissive temperature. According to the three independent experiments of virus passages, the ts-PB1-L66 virus overcame the growth restriction imposed by the nonpermissive temperature at P15, P20, and P25 (Fig. 5A). On the other hand, adaptation to nonpermissive temperature in the ts-PB1-L66V virus was not observed until at least 25 passages (Fig. 5A). These results suggested that the PB1-L66V mutation that resulted in the high-fidelity phenotype prevented reversal of the ts phenotype.
FIG 5 Maintenance of the temperature-sensitive phenotype during serial virus passages. (A) Kaplan-Meier curves according to the loss of ts phenotype. MDCK cells were infected with ts-PB1-L66 and ts-PB1-L66V at an MOI of 0.001 and cultured at 34°C. When a cytopathic effect was observed, the supernatant was recovered and named P1. The supernatant P1 was diluted 102 times and used for the second passage of infection, after which the cells were cultured at 34°C. This procedure was repeated 25 times. The ts phenotype was then checked at the indicated passage numbers. The experiments were repeated three times independently. The period when the virus lost the ts phenotype was derived using the Kaplan-Meier method. P values are based on log rank test (*, P = 0.0246). (B) Nucleotide and amino acid substitutions detected in the ts-PB1-L66 virus showing loss of ts phenotype. a, Data represent results from three independent virus serial passages. b, Amino acid substitutions responsible for the ts phenotype. c, n.d., not detected.
We then checked the sequence of the viral genomes derived from the viruses that adapted to the nonpermissive temperatures. PB2, PB1, PA, and NP, which compose the vRNP complex, were analyzed using Sanger sequencing. The amino acid changes detected among the viruses that adapted to the nonpermissive temperatures are summarized in Fig. 5B. Four amino acid substitutions, N296D, I538V, M202V, and S265N, were identified in PB2. Two substitutions, H431Y and G581E, were detected in PB1. Only the G34D amino acid substitution was identified in NP, while there was no substitution in PA. It should be noted that N265S in PB2, E581G in PB1, and D34G in NP were involved in the introduction of the ts phenotype.


Recently, several high-fidelity RNA virus polymerases have been isolated using nucleotide analogs (15, 3639). In this study, after serial virus passages with ribavirin selection, we identified the PB1-L66V mutation, which conferred a high-fidelity phenotype to influenza viral polymerase. Indeed, NGS analysis showed that the mutation frequency of PB1-L66V was at least 2-fold lower than that of PB1-WT. PB1-V43I is an amino acid substitution involved in improving the replication fidelity of influenza virus RNA polymerase (20), and there are no other reports describing amino acid substitutions resulting in a high-fidelity phenotype in influenza virus RNA polymerase. In this study, we showed that the fidelity of PB1-L66V was higher than that of PB1-V43I. Influenza virus genome replication requires viral factors, including viral polymerase, as well as cellular factors from host cells (i.e., host factors) (4346). Therefore, we performed in vitro vRNA synthesis to examine whether the high-fidelity phenotype of the PB1-L66V polymerase is derived from the polymerase itself or depends on interaction with host factors. Limited elongation assays showed that the rate of misincorporation in the newly synthesized vRNA was reduced by the PB1-L66V mutation, indicating that the high-fidelity phenotype expressed by the PB1-L66V mutation was responsible for the viral polymerase itself.
Structural insights into the PB1 subunit of influenza virus polymerase are available (17, 47). L66 of the PB1 subunit is located in the domain that constitutes the assumed entrance tunnel domain for the incoming nucleotide triphosphate (NTP) (42) (Fig. 6). It is assumed that the PB1-L66 mutation induces a structural change in the NTP entrance tunnel site, leading to successful polymerization for nucleotide selection. Interestingly, V43 (20, 21) and Y82 (13) of the PB1 subunit, which were previously reported as amino acid residues involved in the fidelity of influenza virus genome replication, are also located in this domain (Fig. 6). Thus, this domain could be crucial for controlling the fidelity of influenza virus replication. However, the possibility that other mutations, except for those in the PB1 subunit, may change the fidelity of influenza virus replication cannot be excluded. Indeed, Walmacq et al. demonstrated that the Rpb9 subunit of yeast RNA polymerase II, which is not essential for RNA polymerization, affects the fidelity of transcription (48).
FIG 6 Location of PB1 amino acid mutations involved in fidelity. Cartoon representations of the PB1 subunit (left) and PB1-PB2-PA heterotrimeric polymerase complex (right). PB1-Val43, -Leu66, and -Tyr82 are represented in a space-filling Corey-Pauling-Koltun model. PB2 (red)/PA (green) and PB1 (blue) subunits are indicated in a surface representation and ribbon diagram structure, respectively. The vRNA duplex chain is shown in yellow.
Five amino acid substitutions, PB2-N265S, PB1-K391E/E581G/A661T, and NP-D34G, have been found to be involved in the ts phenotype (27). Both ts-PB1-L66 and ts-PB1-L66V, which included PB2-N265S, PB1-K391E/E581G/A661T, and NP-D34G, showed the ts phenotype. The ts phenotype incorporated into ts-PB1-L66V was genetically stable, whereas the ts phenotype in ts-PB1-L66 was abolished until passage 25. Sequencing analyses revealed that N296D, I538V, M202V, and S265N amino acid substitutions were identified in PB2, H431Y and G581E were detected in PB1, and G34D was identified in NP. Zhou et al. reported that a single amino acid substitution, PB1-E51K, recovered the polymerase activity at nonpermissive temperatures, although the virus possessed all the mutations, PB2-N265S, PB1-K391E/E581G/A661T, and NP-D34G. PB1-I171V or PA-N350K also rescued the polymerase activity at nonpermissive temperatures, without any mutations in the amino acid residue involved in the ts phenotype (31). In this study, the mutations involved in the loss of the ts phenotype were not found in ts-PB1-L66 viruses that adapted to nonpermissive temperatures. However, the PB2-S265, PB1-G581, and NP-G34 mutations responsible for the ts phenotype were found to revert to the WT amino acid sequence.
Usually, an influenza infection in individuals lasts for less than a week. However, especially in immunocompromised individuals, prolonged viral infection is often observed, which results in the induction of mutation accumulation over extended periods. Therefore, in immunocompromised individuals, there is a risk of drug resistance due to the diversification of viral quasispecies. It has been reported that a de novo drug-resistant variant emerged in immunocompromised individuals infected with the 2009 pandemic influenza virus, due to prolonged infection after initiation of oseltamivir treatment (49). The ts phenotype of the LAIV currently in use is known to be relatively stable following replication in individuals (28). However, prolonged propagation and reversal of the pathogenic phenotype may occur in the LAIV currently in use, especially in infants and elderly people with weakened immune systems. Considering this, it is not approved for use in infants and elderly people, who need protection from an influenza virus infection the most. In this study, we isolated influenza viral polymerase PB1-L66V, which induces high genetic stability in virus replication, and demonstrated that the PB1-L66V mutation delayed the reversal of the ts phenotype to a greater degree than the WT. Restricting the diversity of viral quasispecies using a high-fidelity influenza viral polymerase could be an interesting strategy for the development of a stable LAIV.



293T cells (kindly provided by Yoshihiko Kawaoka) and MDCK cells (American Type Culture Collection, Manassas, VA, USA) were maintained in Dulbecco’s modified Eagle’s medium (DMEM) with high glucose (Sigma-Aldrich, St. Louis, MO, USA) containing 10% fetal bovine serum (Nichirei Biosciences, Tokyo, Japan) and penicillin/streptomycin (Sigma-Aldrich).

Plasmid construction for generating recombinant viruses.

To generate plasmids encoding an amino acid point mutation at the Leu66 residue of the PB1 subunit, the pPolI-WSN-PB1 vector (40), which contains a sequence encoding the WT PB1, was used as the backbone vector. Each mutation was introduced into the vector by using inverted PCR with the PrimeSTAR mutagenesis basal kit (TaKaRa Bio, Otsu, Japan) and the mutagenic primers listed in Table 2. The inverted PCR products were diluted 10 times with H2O without purification and rapidly transformed into Competent Quick DH5α (Toyobo, Osaka, Japan), according to the manufacturer’s instructions. To construct pPolI-WSN-PB1-K391E/E581G/A661T, the first pPolI-WSN-PB1-K391E was generated by introducing the PB1-K391E mutation into the pPolI-WSN-PB1 vector, as described above. Then, the PB1-E581G mutation was inserted into pPolI-WSN-PB1-K391E and designated pPolI-WSN-PB1-K391E/E581G. Finally, pPolI-WSN-PB1-K391E/E581G/A661T was generated by introducing the PB1-A661T mutation into pPolI-WSN-PB1-K391E/E581G. The plasmids pPolI-WSN-PB2-N265S, pPolI-WSN-PB1-V43I, and pPolI-WSN-NP-D34G were constructed by introducing these point mutations into pPolI-WSN-PB2, pPolI-WSN-PB1, and pPolI-WSN-NP vectors (40), respectively, using inverted PCR, as described above. The PB1-L66V expression vector was generated by introducing the PB1-L66V mutation into pcDNA-PB1 vector (40) by using inverted PCR, as described above.
TABLE 2 Primers used in this study
Primer nameSequence (5′ to 3′)

Generation of recombinant viruses.

Recombinant influenza viruses were generated using the reverse-genetics method (40, 50). 293T cells were transfected with viral protein and vRNA expression vectors using polyethylenimine (Polysciences, Warrington, PA, USA). Twenty-four hours posttransfection, the culture medium was changed to Opti-MEM I (Gibco, Thermo Fisher Scientific, Waltham, MA, USA) containing 3.5 μg/ml N-p-tosyl-l-phenylalanine chloromethyl ketone-treated trypsin (Sigma-Aldrich). After incubation for 24 h, the cell culture supernatant was collected. The supernatants were inoculated into MDCK cells to amplify the recovered viruses.

Plaque assay.

A confluent monolayer culture of MDCK cells (1.5 × 106 cells) in 6-well tissue culture plates was washed with serum-free DMEM and then infected with influenza viruses. After virus adsorption at 34°C for 1 h, the cells were washed with serum-free DMEM and then overlaid with a DMEM and 0.8% agarose (Sigma-Aldrich) mixture containing 0.2% bovine serum albumin (Sigma-Aldrich), 1× vitamin solution (Gibco), 1× MEM amino acid mixture (Gibco), and 3.5 μg/ml N-p-tosyl-l-phenylalanine chloromethyl ketone-treated trypsin (Sigma-Aldrich). After incubation at 34°C or 38°C for 2 to 3 days, the cells were fixed and stained using 0.5% amido black 10B (Fujifilm Wako Pure Chemical Corporation, Osaka, Japan), as described previously (51).

Minireplicon reporter assay system.

293T cells were seeded into 6-well plates and transfected with pcDNA-PB2, pcDNA-PB1, pcDNA-PA, pCAGGS-NP (40), pHH-vNSLuc (52), and pRL-CMV (Promega, Fitchburg, WI, USA) using polyethylenimine. Twenty-four hours posttransfection, the cells were lysed with passive lysis buffer (Promega), followed by measurement of the firefly luciferase (Fluc) and Renilla luciferase (Rluc) activities. Fluc activity was normalized to that of the Rluc transfection control, and the influenza viral polymerase activity derived from the PB1-L66V mutant was calculated relative to the normalized Fluc activity of PB1-WT, which was taken as 100.

Western blotting.

Western blotting of the viral proteins was performed as previously described (53). Twenty-four hours posttransfection, the number of cells was counted, and 5.0 × 105 cells of each sample were lysed in 20 mM Tris-HCl (pH 7.9), 150 mM NaCl, 1 mM EDTA, and 0.1% NP-40. Viral proteins in the cell lysate were separated using SDS-PAGE, detected using Western blotting, and visualized on an ImageQuant LAS 4000 system (GE Healthcare, Milwaukee, WI, USA). PB1 and β-actin (Sigma-Aldrich) were detected using a rabbit polyclonal antibody (54) and mouse monoclonal antibody, respectively. These antibodies were diluted 103 times with 5% nonfat dry milk before use. The horseradish peroxidase (HRP)-conjugated secondary antibodies were diluted 3 × 103 times with Tris-buffered saline containing 0.1% Tween 20 before use. HRP was detected using ImmunoStar LD (Fujifilm Wako Pure Chemical Corporation).

Cell-free model virus genome replication.

The vRNPs were prepared from purified PB1-WT and PB1-L66V viruses, as described previously (52). Micrococcal nuclease-treated vRNPs were prepared by incubation of vRNPs at 25°C for 3 h with 0.83 U of micrococcal nuclease (Worthington Biochemical Corp., Lakewood, NJ, USA) per microliter in the presence of 3 mM CaCl2. The nuclease reaction was terminated by adding EGTA at a final concentration of 3.5 mM; the micrococcal nuclease-treated vRNPs were used as the enzyme source for the limited elongation assay.
A limited elongation assay was performed as described previously (21, 55). Each vRNP and 120-nt model vRNA template was incubated at 37°C for 1 h in a final volume of 25 μl containing 50 mM HEPES-NaOH (pH 7.9), 3 mM MgCl2, 50 mM KCl, 1.5 mM dithiothreitol, 25 μM GTP, 50 μM CTP, 500 μM ATP, 5 μCi of [α-32P]GTP (3,000 Ci/mmol), 8 U of RNase inhibitor (Toyobo), 250 μM ApG dinucleotide, 0.1 pmol of model vRNA, and vRNP (20, 40, and 60 ng of NP equivalents), in the absence or presence of 500 μM UTP. RNA products were purified using phenol-chloroform extraction, subjected to 6% PAGE with 8 M urea, and visualized using autoradiography (Typhoon 9400; GE Healthcare).

Analyses of nucleic acid substitutions.

MDCK cells were infected with viruses and cultured at a nonpermissive temperature (38°C). vRNA was collected from the infected cells, and cDNA derived from PB2, PB1, PA, and NP, which constitute the vRNP complex, was generated using specific primers. After amplicon generation, Sanger sequencing was performed by Eurofins Genomics Co., Ltd. (Tokyo, Japan).

Mutation frequency determination using NGS.

Amplicon generation and NGS were performed as previously reported (56). Briefly, vRNA was extracted using the QuickGene RNA cultured cell HS kit (Kurabo, Osaka, Japan). Total RNA was reverse-transcribed with oligonucleotide primers Seg5 1 For and Seg8 1 For (Table 2), which are complementary to the nucleotide sequences between positions 1 and 20 on the negative-sense RNA of segments 5 and 8, respectively. The cDNA derived from segment 5 was amplified using fusion primers, including the adaptors for the MiSeq sequencing system (Illumina, San Diego, CA, USA): seg5 217 For overhang and seg5 465 Rev overhang, corresponding to segment 5 between the nucleotide sequence positions 217 to 236 and 446 to 465, respectively (Table 2). To amplify segment 8 cDNA, primers seg8 63 For overhang and seg8 311 Rev overhang, corresponding to segment 8 between the nucleotide sequence positions 63 to 85 and 288 to 311, respectively, were used. DNA libraries were then amplified using a second PCR with the Nextera XT index kit (Illumina), which includes index tags for discriminating the sequencing samples. Quality checks of the DNA libraries and NGS operations were carried out by Bioengineering Lab Co., Ltd., Kanagawa, Japan. It is known that mutation frequencies higher than one mutation per 105 nucleotides are quantitatively reliable for the Illumina system (22). The mutation frequency was determined as previously reported (21, 56). Briefly, we eliminated the low-quality reads (average Phred score of <31) at one sequence position. The reads that passed through the quality filter were aligned using ClustalW, as described previously (57), and the number of mutations in each sequence position was counted. The intrinsic error rate associated with NGS was subtracted from the mutation frequency obtained for each sample, as described previously (56).

Molecular modeling.

Ribbon diagrams and space-filling representations of the influenza virus polymerase were constructed using structural data (PDB ID no. 6T0V) and MolFeat software version (FiatLux, Tokyo, Japan).

Data availability.

Raw sequence data from this experiment are available in the DDBJ Sequence Read Archive under accession number DRA010053.


We thank Yoshihiro Kawaoka (The University of Tokyo) for kindly providing the plasmids for the reverse-genetics system and Kyosuke Nagata (University of Tsukuba) for kindly providing the anti-PB1 antibody. We thank Editage for English language editing.
This study was financially supported by JSPS KAKENHI grant numbers JP16J10490 and JP20K18915 (K.M.), the Kawasaki Foundation for Medical Science and Medical Welfare (K.M.), the Mochida Memorial Foundation for Medical and Pharmaceutical Research (T.N.), a grant for promotion of science and technology in Okayama Prefecture by the Ministry of Education, Culture, Sports, Science, and Technology (T.N.), and the Teraoka Scholarship Foundation (T.N.).
The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.


Domingo E, Holland JJ. 1997. RNA virus mutations and fitness for survival. Annu Rev Microbiol 51:151–178.
Drake JW, Holland JJ. 1999. Mutation rates among RNA viruses. Proc Natl Acad Sci U S A 96:13910–13913.
Peck KM, Lauring AS. 2018. Complexities of viral mutation rates. J Virol 92:e01031-17.
Coffin JM. 1995. HIV population dynamics in vivo: implications for genetic variation, pathogenesis, and therapy. Science 267:483–489.
Havlir DV, Eastman S, Gamst A, Richman DD. 1996. Nevirapine-resistant human immunodeficiency virus: kinetics of replication and estimated prevalence in untreated patients. J Virol 70:7894–7899.
Hurt AC, Barr IG. 2008. Influenza viruses with reduced sensitivity to the neuraminidase inhibitor drugs in untreated young children. Commun Dis Intell Q Rep 32:57–62.
Lech WJ, Wang G, Yang YL, Chee Y, Dorman K, McCrae D, Lazzeroni LC, Erickson JW, Sinsheimer JS, Kaplan AH. 1996. In vivo sequence diversity of the protease of human immunodeficiency virus type 1: presence of protease inhibitor-resistant variants in untreated subjects. J Virol 70:2038–2043.
Najera I, Holguin A, Quinones-Mateu ME, Munoz-Fernandez MA, Najera R, Lopez-Galindez C, Domingo E. 1995. Pol gene quasispecies of human immunodeficiency virus: mutations associated with drug resistance in virus from patients undergoing no drug therapy. J Virol 69:23–31.
Sanjuan R, Domingo-Calap P. 2016. Mechanisms of viral mutation. Cell Mol Life Sci 73:4433–4448.
Smith EC, Blanc H, Surdel MC, Vignuzzi M, Denison MR. 2013. Coronaviruses lacking exoribonuclease activity are susceptible to lethal mutagenesis: evidence for proofreading and potential therapeutics. PLoS Pathog 9:e1003565.
Steinhauer DA, Domingo E, Holland JJ. 1992. Lack of evidence for proofreading mechanisms associated with an RNA virus polymerase. Gene 122:281–288.
Gnadig NF, Beaucourt S, Campagnola G, Borderia AV, Sanz-Ramos M, Gong P, Blanc H, Peersen OB, Vignuzzi M. 2012. Coxsackievirus B3 mutator strains are attenuated in vivo. Proc Natl Acad Sci U S A 109:E2294–E2303.
Naito T, Shirai K, Mori K, Muratsu H, Ushirogawa H, Ohniwa RL, Hanada K, Saito M. 2019. Tyr82 amino acid mutation in PB1 polymerase induces an influenza virus mutator phenotype. J Virol 93:e00834-19.
Rozen-Gagnon K, Stapleford KA, Mongelli V, Blanc H, Failloux AB, Saleh MC, Vignuzzi M. 2014. Alphavirus mutator variants present host-specific defects and attenuation in mammalian and insect models. PLoS Pathog 10:e1003877.
Sadeghipour S, Bek EJ, McMinn PC. 2013. Ribavirin-resistant mutants of human enterovirus 71 express a high replication fidelity phenotype during growth in cell culture. J Virol 87:1759–1769.
Vignuzzi M, Wendt E, Andino R. 2008. Engineering attenuated virus vaccines by controlling replication fidelity. Nat Med 14:154–161.
Fan H, Walker AP, Carrique L, Keown JR, Serna Martin I, Karia D, Sharps J, Hengrung N, Pardon E, Steyaert J, Grimes JM, Fodor E. 2019. Structures of influenza A virus RNA polymerase offer insight into viral genome replication. Nature 573:287–290.
Nagata K, Kawaguchi A, Naito T. 2008. Host factors for replication and transcription of the influenza virus genome. Rev Med Virol 18:247–260.
Te Velthuis AJ, Fodor E. 2016. Influenza virus RNA polymerase: insights into the mechanisms of viral RNA synthesis. Nat Rev Microbiol 14:479–493.
Cheung PP, Watson SJ, Choy KT, Fun Sia S, Wong DD, Poon LL, Kellam P, Guan Y, Malik Peiris JS, Yen HL. 2014. Generation and characterization of influenza A viruses with altered polymerase fidelity. Nat Commun 5:4794.
Naito T, Mori K, Ushirogawa H, Takizawa N, Nobusawa E, Odagiri T, Tashiro M, Ohniwa RL, Nagata K, Saito M. 2017. Generation of a genetically stable high-fidelity influenza vaccine strain. J Virol 91:e01073-16.
Lin RW, Chen GW, Sung HH, Lin RJ, Yen LC, Tseng YL, Chang YK, Lien SP, Shih SR, Liao CL. 2019. Naturally occurring mutations in PB1 affect influenza A virus replication fidelity, virulence, and adaptability. J Biomed Sci 26:55.
Hannoun C. 2013. The evolving history of influenza viruses and influenza vaccines. Expert Rev Vaccines 12:1085–1094.
Clements ML, Murphy BR. 1986. Development and persistence of local and systemic antibody responses in adults given live attenuated or inactivated influenza A virus vaccine. J Clin Microbiol 23:66–72.
Shannon I, White CL, Nayak JL. 2019. Understanding immunity in children vaccinated with live attenuated influenza vaccine. J Pediatric Infect Dis Soc 9(Suppl 1):S10–S14.
Maassab HF. 1967. Adaptation and growth characteristics of influenza virus at 25 degrees C. Nature 213:612–614.
Jin H, Lu B, Zhou H, Ma C, Zhao J, Yang CF, Kemble G, Greenberg H. 2003. Multiple amino acid residues confer temperature sensitivity to human influenza virus vaccine strains (FluMist) derived from cold-adapted A/Ann Arbor/6/60. Virology 306:18–24.
Murphy BR, Coelingh K. 2002. Principles underlying the development and use of live attenuated cold-adapted influenza A and B virus vaccines. Viral Immunol 15:295–323.
Buonagurio DA, Bechert TM, Yang CF, Shutyak L, D'Arco GA, Kazachkov Y, Wang HP, Rojas EA, O'Neill RE, Spaete RR, Coelingh KL, Zamb TJ, Sidhu MS, Udem SA. 2006. Genetic stability of live, cold-adapted influenza virus components of the FluMist/CAIV-T vaccine throughout the manufacturing process. Vaccine 24:2151–2160.
Buonagurio DA, O'Neill RE, Shutyak L, D'Arco GA, Bechert TM, Kazachkov Y, Wang HP, DeStefano J, Coelingh KL, August M, Parks CL, Zamb TJ, Sidhu MS, Udem SA. 2006. Genetic and phenotypic stability of cold-adapted influenza viruses in a trivalent vaccine administered to children in a day care setting. Virology 347:296–306.
Zhou B, Meliopoulos VA, Wang W, Lin X, Stucker KM, Halpin RA, Stockwell TB, Schultz-Cherry S, Wentworth DE. 2016. Reversion of cold-adapted live attenuated influenza vaccine into a pathogenic virus. J Virol 90:8454–8463.
Lauring AS, Jones JO, Andino R. 2010. Rationalizing the development of live attenuated virus vaccines. Nat Biotechnol 28:573–579.
Graci JD, Cameron CE. 2006. Mechanisms of action of ribavirin against distinct viruses. Rev Med Virol 16:37–48.
Oxford JS. 1975. Inhibition of the replication of influenza A and B viruses by a nucleoside analogue (ribavirin). J Gen Virol 28:409–414.
Wray SK, Gilbert BE, Noall MW, Knight V. 1985. Mode of action of ribavirin: effect of nucleotide pool alterations on influenza virus ribonucleoprotein synthesis. Antiviral Res 5:29–37.
Pfeiffer JK, Kirkegaard K. 2003. A single mutation in poliovirus RNA-dependent RNA polymerase confers resistance to mutagenic nucleotide analogs via increased fidelity. Proc Natl Acad Sci U S A 100:7289–7294.
Stapleford KA, Rozen-Gagnon K, Das PK, Saul S, Poirier EZ, Blanc H, Vidalain PO, Merits A, Vignuzzi M. 2015. Viral polymerase-helicase complexes regulate replication fidelity to overcome intracellular nucleotide depletion. J Virol 89:11233–11244.
Tian D, Meng XJ. 2016. Amino acid residues Ala283 and His421 in the RNA-dependent RNA polymerase of porcine reproductive and respiratory syndrome virus play important roles in viral ribavirin sensitivity and quasispecies diversity. J Gen Virol 97:53–59.
Van Slyke GA, Arnold JJ, Lugo AJ, Griesemer SB, Moustafa IM, Kramer LD, Cameron CE, Ciota AT. 2015. Sequence-specific fidelity alterations associated with West Nile virus attenuation in mosquitoes. PLoS Pathog 11:e1005009.
Neumann G, Watanabe T, Ito H, Watanabe S, Goto H, Gao P, Hughes M, Perez DR, Donis R, Hoffmann E, Hobom G, Kawaoka Y. 1999. Generation of influenza A viruses entirely from cloned cDNAs. Proc Natl Acad Sci U S A 96:9345–9350.
Kobayashi M, Toyoda T, Ishihama A. 1996. Influenza virus PB1 protein is the minimal and essential subunit of RNA polymerase. Arch Virol 141:525–539.
Gonzalez S, Ortin J. 1999. Characterization of influenza virus PB1 protein binding to viral RNA: two separate regions of the protein contribute to the interaction domain. J Virol 73:631–637.
Kawaguchi A, Nagata K. 2007. De novo replication of the influenza virus RNA genome is regulated by DNA replicative helicase, MCM. EMBO J 26:4566–4575.
Momose F, Handa H, Nagata K. 1996. Identification of host factors that regulate the influenza virus RNA polymerase activity. Biochimie 78:1103–1108.
Naito T, Kiyasu Y, Sugiyama K, Kimura A, Nakano R, Matsukage A, Nagata K. 2007. An influenza virus replicon system in yeast identified Tat-SF1 as a stimulatory host factor for viral RNA synthesis. Proc Natl Acad Sci U S A 104:18235–18240.
Sugiyama K, Kawaguchi A, Okuwaki M, Nagata K. 2015. pp32 and APRIL are host cell-derived regulators of influenza virus RNA synthesis from cRNA. Elife 4:e08939.
Pflug A, Guilligay D, Reich S, Cusack S. 2014. Structure of influenza A polymerase bound to the viral RNA promoter. Nature 516:355–360.
Walmacq C, Kireeva ML, Irvin J, Nedialkov Y, Lubkowska L, Malagon F, Strathern JN, Kashlev M. 2009. Rpb9 subunit controls transcription fidelity by delaying NTP sequestration in RNA polymerase II. J Biol Chem 284:19601–19612.
Memoli MJ, Hrabal RJ, Hassantoufighi A, Jagger BW, Sheng ZM, Eichelberger MC, Taubenberger JK. 2010. Rapid selection of a transmissible multidrug-resistant influenza A/H3N2 virus in an immunocompromised host. J Infect Dis 201:1397–1403.
Takizawa N, Ogura Y, Fujita Y, Noda T, Shigematsu H, Hayashi T, Kurokawa K. 2019. Local structural changes of the influenza A virus ribonucleoprotein complex by single mutations in the specific residues involved in efficient genome packaging. Virology 531:126–140.
Kuroki T, Lee S, Hirohama M, Taku T, Kumakura M, Haruyama T, Nagata K, Kawaguchi A. 2018. Inhibition of influenza virus infection by Lentinus edodes mycelia extract through its direct action and immunopotentiating activity. Front Microbiol 9:1164.
Shimizu K, Handa H, Nakada S, Nagata K. 1994. Regulation of influenza virus RNA polymerase activity by cellular and viral factors. Nucleic Acids Res 22:5047–5053.
Takizawa N, Kimura T, Watanabe T, Shibasaki M. 2018. Anti-influenza virus activity of a salcomine derivative mediated by inhibition of viral RNA synthesis. Arch Virol 163:1607–1614.
Naito T, Momose F, Kawaguchi A, Nagata K. 2007. Involvement of Hsp90in assembly and nuclear import of influenza virus RNA polymerase subunits. J Virol 81:1339–1349.
Momose F, Naito T, Yano K, Sugimoto S, Morikawa Y, Nagata K. 2002. Identification of Hsp90 as a stimulatory host factor involved in influenza virus RNA synthesis. J Biol Chem 277:45306–45314.
Mori K, Murano K, Ohniwa RL, Kawaguchi A, Nagata K. 2015. Oseltamivir expands quasispecies of influenza virus through cell-to-cell transmission. Sci Rep 5:9163.
Thompson JD, Higgins DG, Gibson TJ. 1994. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res 22:4673–4680.

Information & Contributors


Published In

cover image Journal of Virology
Journal of Virology
Volume 95Number 1224 May 2021
eLocator: 10.1128/jvi.00493-21
Editor: Stacey Schultz-Cherry, St. Jude Children's Research Hospital


Received: 26 March 2021
Accepted: 28 March 2021
Accepted manuscript posted online: 9 April 2021
Published online: 24 May 2021


Request permissions for this article.


  1. high fidelity
  2. influenza
  3. live attenuated vaccine
  4. viral polymerase



Kotaro Mori [email protected]
Department of Microbiology, Kawasaki Medical School, Okayama, Japan
Present address: Kotaro Mori, Medical Genomics Center, National Center for Global Health and Medicine, Tokyo, Japan.
Ryosuke L. Ohniwa
Division of Biomedical Science, Faculty of Medicine, University of Tsukuba, Ibaraki, Japan
Center for Biotechnology, National Taiwan University, Taipei, Taiwan
Naoki Takizawa
Laboratory of Virology, Institute of Microbial Chemistry (BIKAKEN), Tokyo, Japan
Tadasuke Naito
Department of Microbiology, Kawasaki Medical School, Okayama, Japan
Department of Microbiology, Kawasaki Medical School, Okayama, Japan


Stacey Schultz-Cherry
St. Jude Children's Research Hospital

Metrics & Citations


Note: There is a 3- to 4-day delay in article usage, so article usage will not appear immediately after publication.

Citation counts come from the Crossref Cited by service.


If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. For an editable text file, please select Medlars format which will download as a .txt file. Simply select your manager software from the list below and click Download.

View Options

Figures and Media






Share the article link

Share with email

Email a colleague

Share on social media

American Society for Microbiology ("ASM") is committed to maintaining your confidence and trust with respect to the information we collect from you on websites owned and operated by ASM ("ASM Web Sites") and other sources. This Privacy Policy sets forth the information we collect about you, how we use this information and the choices you have about how we use such information.
FIND OUT MORE about the privacy policy