The substitution of T271A in PB2 protein could enhance the infectivity and pathogenicity of Eurasian avian-like H1N1 swine influenza viruses in mice

Background Currently, Eurasian avian-like H1N1 (EA H1N1) swine influenza viruses (SIVs) are widely prevalent in pigs in China, with sporadic human cases reported as well. As one of the key molecular makers detected in avian H5N1 and H 7N9 viruses and pandemic H1N1 2009 virus, contributions of T271A in PB2 protein to the EA H1N1 viruses are still unknown. In this study, we investigated the effects of residue 271 in PB2 protein on the viral properties of EA H1N1 viruses. Methods 271 PB2 protein were studied in cells and mice. The results showed that the substitution PB2-T271A increased the viral replication in mammalian and avian cell lines. In addition, the mutation enhanced the viral infectivity, virulence and pathogenicity in mice. The viral titers of lung tissue in the rgHuN 271A virus were higher than that of the rgHuN 271T at 1, 4, and 7 dpi. The MID 50 of the rgHuN 271A and rgHuN 271T virus were 10 1.1 TCID 50 and 10 1.9 TCID 50 , respectively. Besides, the substitution of PB2-T271A enhanced the viral polymerase activity in mammalian cells. 271A in PB2 were higher than EA H1N1 virus 271T protein vivo vitro . Therefore, the PB2-T271A mutation should be continually monitored influenza circulating and IL-6: interleukin-6; IL-8: interleukin-8; IL-10: interleukin-10; CCL5: chemokine 5; IFN-α1: interferon α1


Background
The contributions of molecular markers to the adaptation, virulence and transmission of influenza A virus have been broadly studied. These molecular markers include a series of substitutions in hemagglutinin (HA), neuraminidase (NA), and polymerase basic protein 2 (PB2). Amongst them, mutations in the polymerase complex (PB2, PB1 and PA) play an important role in the avian influenza viruses breaking the species barrier and increase their fitness in the mammalian hosts. The host-adaptive mutations of E627K and D701N in PB2 protein have been well recognized as the molecular biomarkers that enhance the viral replication, pathogenesis, and virulence of H5N1, H7N7, and H7N9 influenza viruses in mammals [1][2][3][4].
In 2007, to discover host-associated genomic signatures, a total of 23,581 full genome sequences of avian and human influenza viruses were analyzed [5]. Many conservative residues between human and avian influenza viruses were identified as potential host-associated markers, including PB2-T271A [5][6][7]. In avian influenza viruses, the 271 site in the PB2 protein was threonine (T); nevertheless, the 271 site resulted in alanine (A) in human influenza viruses. Subsequent studies showed that PB2 residue 271A could enhance viral polymerase activity and the viral growth of 2009 H1N1 pandemic influenza viruses in mammalian cells [8][9][10]. Besides, the PB2-T271A mutation could contribute to the efficient transmission of 2009 H1N1 pandemic influenza virus in guinea pigs [9]. Similar results were found in H7N9 and H5N1 influenza viruses [11][12][13]. Besides, the combination of the substitutions PB2-T271A and 590/591 SR increased the viral replication and adaptation of the 2009 pandemic H1N1 and triple reassortant swine influenza viruses (SIVs) in mammals [14].
Eurasian avian-like H1N1 (EA H1N1) SIVs have been predominant in pigs in China since 2005 [15]. After a long term evolution, the EA H1N1 viruses were reported to have the potential to cause an influenza pandemic [16].
As a matter of fact, sporadic human cases of infection with EA H1N1 viruses have been continuously reported in China [17][18][19][20]. In a previous study, the substitution of D701N in the PB2 protein was shown to enhance the viral replication and pathogenicity of EA H1N1 viruses [21]. Besides, the substitution Q357K in the NP protein plays an important role in the virulence phenotype of EA H1N1 viruses [22]. Our previous study reported that the pathogenicity of two human-isolated EA H1N1 viruses (A/Hunan/42443/2015 (H1N1, HuN) and A/Jiangsu/1/2011 (H1N1, JS1) viruses) were divergent [19]. Sequence analysis showed that the two viruses had different amino acids at the 271 site, with A and T in HuN and JS1 viruses, respectively. Whether the 271 residue contributed to the viral virulence is still to be shown. Thus, we analyzed the effects of the residue 271 on the PB2 protein of HuN EA H1N1 SIVs in vivo and in vitro.

Site-directed mutagenesis
The eight gene plasmids of HuN virus were constructed and stored by the Chinese National Influenza Center. A plasmid with a signal mutation A271T in the PB2 protein was constructed using the forward primer 5'-GTAAGAAGAGCAACAGTGTCAGCAG-3' and the reverse primer 5'-CTGCTGACACTGTTGCTCTTCTTAC-3'. To ensure the presence of the point mutation and the absence of unwanted mutations, the viral cDNA was sequenced.

Virus rescue and virus titration
As previously described [23], the eight plasmids containing the cDNA of the HuN 271A (with residue 271A in PB2 protein) or HuN 271T (with residue 271T in PB2 protein) virus were co-transfected into co-cultured 293T/MDCK cells on six-well plates using SuperFect transfection reagent (Qiagen, Valencia, CA, USA). After 48 hours, cultured supernatants were inoculated into 9-day-old specific pathogen free (SPF) embryonated eggs. To avoid unwanted mutations, we sequenced both the recombinant viruses. Virus titration was performed in MDCK cells. The 50% tissue culture infective dose (TCID 50 ) was calculated by the Reed-Muench formula [24].

Virus growth kinetics study
To detect the growth kinetics of both recombinant viruses, A549, DF-1 and PK15 cells were inoculated with the viruses at the multiplicity of infection (MOI) of 0.01. After incubating at 37°C for 1 hour, the supernatants were removed and cells were washed twice with PBS. DMEM containing TPCK-trypsin, penicillin and streptomycin was added . Supernatants were collected at specified time points of 0, 12, 24, 36, 48, 60, 72, 84, and 96 hours post inoculation (hpi) and stored at -80 °C until use. Virus titers were determined by a TCID 50 assay in MDCK cells.

Animal experiments
To evaluate the contribution of the substitution PB2-T271A to pathogenicity of the EA H1N1 SIVs, we compared the infectivity, replication and virulence of recombinant viruses in mice. Eight-to ten-week-old SPF C57BL/6J female mice were used in the experiment. Mice were randomly assigned. Each group had five mice. The mice were anesthetized with pentobarbital sodium and inoculated intranasally with 50 μl of 10-fold serial dilutions of each recombinant virus in phosphate-buffered saline (PBS), with doses ranging from 10 1 -10 6 TCID 50 . Mock mice were inoculated with 50 μl PBS. Body weight were recorded daily for 14 days post inoculation (dpi). When the mice lost 30% of their body weight, they were killed humanely. The 50% mouse lethal dose (MLD 50 ) was calculated based on the survival rate of the mice. Serum from the surviving mice was collected on the 14th day after post-infection, and we detected hemagglutinin inhibiting (HI) antibody titers to calculate the 50% mouse infective dose (MID 50 ).
To evaluate the tropism and replication capacity of both recombinant viruses, we inoculated mice with 10 5 TCID 50 viruses. Three mice were euthanized at 1, 4, and 7 dpi, respectively. Respiratory tract samples containing lung, trachea, and nasal turbinate were collected. The right lobes of the lungs were pumped with 10% formaldehyde solution, which were used for histological analysis. The viral titers were determined by TCID 50 assay in MDCK cells.

Histopathological analysis
The right lungs were fixed and dehydrated. The tissue blocks were placed in a mold containing melted paraffin wax and solidified into wax blocks. Lung tissues were sectioned at 4-5 μm thick. Then hematoxylin and eosin staining (H&E) was performed to observe viral infection.

Polymerase activity assay
Monolayers of 293T cells were transfected with a firefly luciferase reporter plasmid, a Renilla luciferase expressing plasmid and four plasmids expressing the polymerase genes PB2, PB1, PA and NP. The Renilla luciferase was used as an internal transfection control for the dual luciferase assay. After 24 hours of transfection, luciferase activity was measured using EnVision 2104 Multilabel Reader (PerkinElmer, USA) and the assay was performed in triplicate.

Western blot assay
293T cells transfected with plasmids in the polymerase activity assay were lysed with passive lysis buffer (Promega, Madison, USA). Cells lysates were mixed with SDS-PAGE sample loading buffer (Beyotime Biotech, China) and heated at 95 ℃ for 10 min. The samples were separated by SDS-PAGE and transferred to a nitrocellulose membrane (Millipore, Billerica, USA). The membrane was blocked with 5% nonfat milk for 2 hours and then incubated with anti-PB2 (GTX125926, GeneTex, Irvine, USA). The primary antibody was incubated overnight at 4 ℃. The membrane was incubated with goat anti-rabbit secondary antibody for 1 hour. The images were obtained by using the MiniChemi Imaging and Analysis System (Sage Creation, China).

Statistical analysis
The student's t test and two-way ANOVA were calculated using GraphPad Prism software 6.0. A P value < 0.05 was considered to be statistically significant.

The substitution of PB2-T271A enhanced the viral replication in cells
To determine the effect of PB2-T271A substitution on virus replications in vitro, we performed growth kinetics analysis of both recombinant viruses in A549, PK15, and DF-1 cells (Fig. 1). It was obvious that the rgHuN 271T virus rarely replicated in DF-1 cells. Nevertheless, the rgHuN 271A virus had significantly better growth advantages than the rgHuN 271T virus in DF-1 cells during 36 to 96 hpi (P < 0.01) (Fig. 1a). Furthermore, the replication titers of rgHuN 271A virus were significantly higher than those of rgHuN 271T virus from 36 to 96 hpi in A549 cells (P < 0.01) (Fig. 1b). In PK15 cells, both recombinant viruses could replicate efficiently (Fig. 1c). Generally, the replication titers of rgHuN 271A virus were higher than the rgHuN 271T virus (Fig. 1c). There were statistical differences between the two recombinant viruses at 36, 48, 72, 84, and 96 hpi (P < 0.01 or P < 0.05). These results indicated that the mutation PB2-T271A could enhance the viral replication in mammalian and avian cell lines.

The substitution of PB2-T271A enhanced viral morbidity and mortality in mice
To evaluate the effects of the PB2-T271A mutation on pathogenicity in mice, the recombinant viruses at the indicated dose (10 1 -10 6 TCID 50 ) were inoculated intranasally with five mice in each group. At the inoculation dose of 10 1 -10 3 TCID 50 with the recombinant rgHuN 271A or rgHuN 271T viruses, we observed that no mice died and the body weight loss was <10% (Fig. 2a-b). When inoculated with 10 4 TCID 50 of the recombinant viruses, the body weight loss of the rgHuN 217A group mice was approximately 20%. However, the body weight loss of the rgHuN 271T group mice was <10% (Fig. 2a-b). At the inoculation dose of 10 5 TCID 50 , four-fifths of the mice of the rgHuN 271A virus died (Fig. 2c). Two mice died at 7 dpi; another two mice died at 8 dpi. Meanwhile, no mice died at the inoculation dose of 10 5 TCID 50 with rgHuN 271T virus (Fig. 2d). When infection with the highest dose of 10 6 TCID 50 of the recombinant viruses, all mice of rgHuN 271A virus died at 6 dpi and four mice of rgHuN 271T virus died at 8 dpi; one mouse survived until the end of the experiment (Fig. 2c-d). To evaluate the virulence of the recombinant viruses, we calculated the MLD 50 . The MLD 50 of the rgHuN 271A virus was 10 4.5 TCID 50 (Fig. 2c).
Nevertheless, the MLD 50 of the rgHuN 271T virus was 10 5.7 TCID 50 (Fig. 2d). The MID 50 values of the rgHuN 271A and rgHuN 271T virus were 10 1.1 TCID 50 and 10 1.9 TCID 50 , respectively (Table 1). These results indicated that the substitution T271A in the PB2 protein could increase the virulence of the EA H1N1 SIVs in mice.

The substitution of PB2-T271A enhanced the viral replication in mice
To evaluate the replication of the recombinant viruses in the respiratory tracts of mice, we collected the tracheal, nasal turbinate, and lung tissue samples from mice at 1, 4, and 7 dpi with the inoculated dose of 10 5 TCID 50 .
Both recombinant viruses could replicate well in the respiratory tract tissues. The rgHuN 271A viral titers of nasal turbinate were higher than that of rgHuN 271T , with no statistical difference (Fig. 3). There was no significant change in the viral titer of trachea tissues between the two recombinant viruses. However, the viral titers of lung tissues in the rgHuN 271A virus were significantly higher than those of the rgHuN 271T at 1, 4, and 7 dpi (P < 0.01) (Fig. 3). These data showed that the substitution PB2-T271A could increase the replication of EA H1N1 viruses in mice.

The substitution of PB2-T271A enhanced viral pathogenicity in mice
To compare the pulmonary histopathological damage induced by the recombinant viruses, the H&E staining was performed with lung tissue sections (Fig. 4). At 1 dpi, the lung damage of both recombinant viruses was rarely observed (Fig. 4a-b). The cell exfoliation and local bronchitis occurred in the mice lung of the rgHuN 271A virus group compared to rgHuN 271T viruses at 4 dpi (Fig. 4c-d). Interstitial bronchial pneumonia and focal alveolar damage of the lung were manifested at 7 dpi in the group of rgHuN 271A virus; however, scattered pathological lesions of the lungs were observed in rgHuN 271T virus (Fig. 4e-f). These results indicated that the substitution PB2-T271A could increase the viral pathogenicity in mice.

The substitution of PB2-T271A enhanced the proinflammatory response in mice
To compare the proinflammatory response of the recombinant rgHuN 271A and rgHuN 271T viruses, we detected the cytokines and chemokine expression levels at 1, 4, and 7 dpi including IL-6, IL-8, IL-10, CCL5, and IFN-α1 (Fig. 5). The recombinant rgHuN 271A virus induced higher levels of IL-6, IL-8, IL-10, and IFN-α1 than those of rgHuN 271T virus at 1, 7 dpi (Fig. 5), even if there was no statistic difference. These results indicated that the mutation of T271A in PB2 protein could increase proinflammation response in mice.

The substitution of PB2-T271A enhanced the viral polymerase activity in mammalian cells
To measure the impact of mutation PB2-T271A in the recombinant viruses on the transcription and replication of viral RNA in 293T cells, the luciferase reporter assay was conducted. From the results, we can see that the PB2-T271A substitution significantly increased the polymerase activity (P < 0.01). This was increased approximately four-fold compared to the polymerase activity of rgHuN 271T virus (Fig. 6). The PB2 proteins of the rgHuN 271A and rgHuN 271T viruses were expressed at similar levels in 293T cells by Western blot analysis (Fig. 6).

Discussion
As a molecular maker, PB2-T271A was able to enhance the polymerase activity and viral replication in mammal cells for the 2009 pandemic H1N1 and H7N9 influenza virus [9,11]. On the other hand, as a the zoonotic influenza virus, EA H1N1 viruses circulating in pigs in China were considered most likely to cause the next pandemic [16]. However, studies of the molecular markers in the EA H1N1 viruses were limited in number. In this study, we systematically analyzed the contributions of PB2-271 substitution of EA H1N1 SIVs both in vivo and in vitro. We found that PB2-T271A substitution enhanced the viral replication ability and polymerase activity of EA H1N1 viruses in cell lines, and more significantly, it increased the infectivity, replication, and virulence of EA H1N1 SIVs in mice.
Since 2005, EA H1N1 viruses have been predominant in China [15]. Several molecular makers played an important role in EA H1N1 viruses. The PB2-R251K, PB2-D701N, and NP-Q357K of EA H1N1 viruses contributed to the viral replication and pathogenicity in mice [21,22,30]. Furthermore, these mutations enhanced the polymerase activity. The PB2-T271A substitution showed similar results. It enhanced the viral replication in vivo and in vitro ( Fig. 1-2). In addition, the residue 271 in PB2 was located near the cap-binding domain, thus affecting the interaction between PB2 and the cap structure [8,31]. This might increase the viral polymerase activity and vRNA synthesis [32].
PB2-271A and PB2-271T are conserved in human influenza viruses and avian influenza viruses, respectively ( Table 2) [10]. The sequence analysis showed that 99.6% of human H1N1 influenza viruses possessed A at position 271 in PB2 protein. However, 98.2% of avian H1N1 influenza viruses contained T at position 271 in PB2. In swine populations, the prevalence of A (71.0%) was higher than that of T (22.4%) for H1N1 SIVs (Table 2). Considering the higher virulence exhibited by 271A viruses than 271T viruses in mice, surveillance of the EA H1N1 viruses should be strengthened, especially for those with 271A residues.

Conclusions
In this study, we evaluated the effect of the PB2-T271A mutation on EA H1N1 SIVs in vivo and in vitro. The substitution PB2-T271A increased the viral polymerase activity and growth properties in mammalian cells, and the infectivity and virulence in mice. Thus, substantial attention should be paid on the PB2-T271A mutation.  were treated as positive. The HI titer "-" was considered as negative.  Figure 1 Replication kinetics of the two recombinant viruses in DF-1, A549 and PK15 cells. Cells were inoculated at a multiplicity of infection 0.01 with the rgHuN271A and rgHuN271T viruses in triplicate. The supernatants were collected at the indicated time points for titration in MDCK cells. a DF-1 cells. b A549 cells. c PK15 cells. * indicated P < 0.05 when comparing the growth kinetics of the rgHuN271A and rgHuN271T viruses. ** P < 0.01.

Figure 2
Body weight changes and survival rates of the rgHuN271A and rgHuN271T viruses in mice. Mice were inoculated intranasally with 50 μl at the dose of 101-106 TCID50 of the recombinant viruses rgHuN271A or rgHuN271T. The mice of the control group were inoculated intranasally with 50 μl PBS. a and b Body weight changes and survival rate of rgHuN271A virus in mice. c and d Body weight changes and survival rate of rgHuN271T virus in mice. Body weight and survival were recorded daily for 14 dpi.  The lungs histopathology changes with the recombinant viruses by the H&E staining. The mice lungs were collected with 10% neutral buffered formalin at 1, 4, and 7 dpi. a-g show the results of H&E staining.

Figure 5
Detection of cytokine proteins in mice lungs infected with the recombinant viruses. Lungs were collected with inoculated 105 TCID50 of the two recombinant viruses at 1, 4, and 7 dpi. IL-6, IL-8, IL-10, CCL5 and IFN-α1 of lungs in mice were detected.

Figure 6
Viral polymerase activity with the two recombinant viruses in 293T cells. A firefly luciferase reporter plasmid, a Renilla luciferase expressing plasmid and the polymerase complex (PB1, PA, NP, and PB2271A or PB2271T) were transfected in 293T cells. After 24 hpi, the polymerase activity was detected. The relative activity (%) means the value of firefly luciferase reporter plasmid was normalized against by the value of Renilla luciferase expressing plasmid. Expression of the rgHuN271A and rgHuN271T PB2 proteins in transfected 293T cells was determined by Western blot assay. **P < 0.01.