Generation of a Reassortant Influenza A Subtype H3N2 Virus Expressing Gaussia Luciferase

Reporter influenza A viruses (IAVs) carrying fluorescent or luminescent genes provide a powerful tool for both basic and translational research. Most reporter IAVs are based on the backbone of either subtype H1N1 viruses, A/Puerto Rico/8/1934 (PR8) or A/WSN/1933, but no reporter subtype H3N2 virus is currently available to our knowledge. Since the IAV subtype H3N2 co-circulates with H1N1 among humans causing annual epidemics, a reporter influenza A subtype H3N2 virus would be highly valuable. In this study, the segments of A/Wyoming/3/03 (NY, H3N2) virus encoding hemagglutinin and neuraminidase, respectively, were reassorted with the six internal genes of PR8 where the NS gene was fused with a Gaussia luciferase (Gluc) gene. Using reverse genetics, NY-r19-Gluc, a replication competent reassortant influenza A subtype H3N2 virus expressing reporter Gluc was successfully generated. This reporter virus is stable during replication in Madin-Darby canine kidney (MDCK) cells, and preliminary studies demonstrated it as a useful tool to evaluate antivirals. In addition, NY-r19-Gluc virus will be a powerful tool in other studies including the application of diagnostic and therapeutic antibodies as well as the evaluation of novel vaccines.


Introduction
The influenza A virus (IAV) is a major cause of respiratory illness in humans, accounting for up to 650,000 annual deaths worldwide [1]. The IAV belongs to the family Orthomyxoviridae of enveloped viruses, and the genome consists of 8 negative-sense, single-stranded viral RNA (vRNA) segments, coding for at least 11 proteins [2]. Hemagglutinin (HA) and neuraminidase (NA) are the major surface glycoproteins, based on which IAVs can be classified into different subtypes. There are 16 subtypes of HA (H1-H16) and 9 subtypes of NA (N1-N9) identified in wild birds [3]. In addition, two new subtypes (H17N10 and H18N11) have recently been isolated in bats [4,5]. Currently, IAV subtypes H1N1 and H3N2 as well as two influenza B virus strains of Yamagata and Victoria lineages co-circulate in humans [6], while other subtypes such as H5N1 and H7N9 can gain the ability to infect humans occasionally [7,8].

Generation of Reporter IAV of H3N2
As described previously, wild-type A/NY (H3N2) was generated by using the reverse genetics system with indicated rescue plasmids [32]. Briefly, eight pDZ plasmids (0.5 µg each) representing the eight segments of the IAV genome were transfected into 293T/MDCK cocultures using Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA). After 48 h the virus was harvested from the supernatant.
In order to generate reporter IAV H3N2, three strategies were employed respectively. First, pDZ-NY/NS-Gluc instead of pDZ-NY/NS was used to generate a recombinant virus in which the NS gene of the NY virus was fused with the reporter gene Gaussia luciferase, which is the same as the strategy used to construct PR8/NS-Gluc previously described [28]. The recombinant virus was designated as NY-Gluc accordingly.
Second, the NS segment of the NY virus was replaced with PR8/NS-Gluc, and the virus was designated as NY-r18-Gluc.
Third, the six segments other than HA and NA of the NY virus were replaced with those of PR8-Gluc, and the virus was designated as NY-r19-Gluc. In parallel, a reassortant NY-r19 virus was generated by replacing the six internal segments of NY with those of wild-type PR8 virus.
The TCID 50 values were determined using MDCK cells and the titer was calculated by the Reed-Muench method [33].

Virus Propagation
The viruses of NY, NY-Gluc, NY-r18-Gluc, NY-r19, and NY-r19-Gluc were propagated in MDCK cells. MDCK cells of 80% confluence grown in 10 cm dishes were infected with the indicated viruses at a multiplicity of infection (MOI) of 0.01 TCID 50 per cell. After incubation at 37 • C for 1 h, the cells were washed and fresh medium was added. The viruses were harvested respectively when most cells showed the cytopathic effect (CPE) and titrated for TCID 50 values.
In order to test the genetic stability of reporter IAV NY-r19-Gluc, the virus was serially passaged in MDCK cells eight times, labeled as P1 to P8 accordingly.

Multicycle Replication Assay
A multicycle replication assay was performed to compare the replication properties of the indicated viruses in vitro. Briefly, MDCK cells were seeded into 24-well plates 1 day before inoculation with the indicated viruses at a MOI of 0.01. After incubation at 37 • C for 1 h, the cells were washed, followed by addition of fresh medium. At various time points, aliquots were removed for determination of Gluc activity and viral titers.

Antiviral Determination
MDCK cells grown in 24-well plates were infected with NY-r19-Gluc virus at a MOI of 0.1. After 1 h of incubation at 37 • C, the cells were washed, and fresh Opti-MEM containing 2 µg/mL TPCK-trypsin was added. For antiviral determination, indicated concentrations of TBHQ or arbidol were present all through the process. At around 72 h post-infection (p.i.), aliquots were removed for the luciferase assay.

Reporter Influenza A Subtype H3N2 Virus with a Gluc Gene Fused to NS1 Loses the Reporter Gene Rapidly
Previously, a reporter influenza A subtype H1N1 virus PR8-Gluc was successfully generated by fusing a Gluc gene to the NS1 open reading frame [28]. In this study, a similar strategy was initially employed to generate the reporter influenza A subtype H3N2 virus, NY-Gluc ( Figure 1a). However, as the recombinant virus was passaged in MDCK cells, the expression level of the reporter Gluc decreased rapidly, and only negligible Gluc was detectable after two passages (Figure 1b). This result suggested that the NY-Gluc is genetically unstable in vitro.
Previously, a reporter influenza A subtype H1N1 virus PR8-Gluc was successfully generated by fusing a Gluc gene to the NS1 open reading frame [28]. In this study, a similar strategy was initially employed to generate the reporter influenza A subtype H3N2 virus, NY-Gluc ( Figure 1a). However, as the recombinant virus was passaged in MDCK cells, the expression level of the reporter Gluc decreased rapidly, and only negligible Gluc was detectable after two passages (Figure 1b). This result suggested that the NY-Gluc is genetically unstable in vitro.

Reporter Influenza A Subtype H3N2 Virus Carrying PR8/NS-Gluc Segment Induces CPE Alteration
To generate the reporter influenza A subtype H3N2 virus, another strategy was explored based on previous work by Fukuyama et al. [30]. They successfully generated a reporter IAV expressing fluorescent Venus based on A/Vietnam/1203/2004 (VN1203; H5N1) [30]. In this strategy, the PR8 NS gene is used to express NS1-Venus chimeric protein since the NS gene contributes little to the pathogenicity of VN1203 in mice [34].
As shown in Figure 2a, a reassortant virus containing PR8 NS-Gluc segment derived from a recombinant virus PR8-Gluc and the other seven segments of wildtype NY virus was successfully rescued using reverse genetics. The virus was designated as NY-r18-Gluc. Unlike the NY-Gluc virus, the NY-r18-Gluc virus could express Gluc protein at a high level, and no obvious decrease of Gluc expression was observed after serial passage of the viruses, suggesting that the reporter Gluc gene is functionally maintained (Figure 2b). in Figure 2c, the infection of NY-r18-Gluc of passages 1 to 3 respectively caused cells to become round and detached from the culture plates, which is similar to the CPE induced by wildtype NY virus infection. However, after four passages, the NY-r18-Gluc infected cells became fibrous, and formed numbers of multinucleated syncytia (Figure 2c). This phenomenon implies that NY-r18-Gluc has evolved during passaging in MDCK cells, and the precise mutations and underlying mechanisms remain to be investigated, leaving the reporter virus with limited importance at present.  However, an altered CPE upon NY-r18-Gluc infection into MDCK cells was observed. As shown in Figure 2c, the infection of NY-r18-Gluc of passages 1 to 3 respectively caused cells to become round and detached from the culture plates, which is similar to the CPE induced by wildtype NY virus infection. However, after four passages, the NY-r18-Gluc infected cells became fibrous, and formed numbers of multinucleated syncytia (Figure 2c). This phenomenon implies that NY-r18-Gluc has evolved during passaging in MDCK cells, and the precise mutations and underlying mechanisms remain to be investigated, leaving the reporter virus with limited importance at present.

A 6:2 Genetic Reassortant Reporter Influenza A Subtype H3N2 Virus is Genetically Stable
X-31, which is a 6:2 genetic reassortant encoding the HA and NA genes of A/Hong Kong/1/1968 (H3N2) in a backbone comprising the six internal genes of PR8, is often used in studies requiring an IAV H3N2 subtype [35][36][37][38][39][40][41][42][43]. A similar strategy was explored here, and a reassortant virus NY-r19 encoding HA and NA of the NY virus in the backbone of PR8 was successfully rescued by reverse genetics. In parallel, we further generated a reporter virus NY-r19-Gluc containing HA and NA of the NY virus in the backbone of PR8 where a reporter Gluc gene was fused to NS1 (Figure 3a).  Figure 3b, no loss of Gluc expression was observed upon virus passaging (for at least 8 passages), suggesting that the reassortant virus is stable in vitro. The CPEs in MDCK cells respectively infected with NY-r19-Gluc of early and late passages were also compared, and no difference was detected (Figure 3c) Together, these results show that reassortant NY-r19-Gluc virus is a good candidate as a reporter influenza A subtype H3N2 virus for further characterization and application.  The stability of Gluc in the reassortant virus NY-r19-Gluc was determined by serial passages in MDCK cells. As shown in Figure 3b, no loss of Gluc expression was observed upon virus passaging (for at least 8 passages), suggesting that the reassortant virus is stable in vitro. The CPEs in MDCK cells respectively infected with NY-r19-Gluc of early and late passages were also compared, and no difference was detected (Figure 3c)

Characterization of the Reassortant Reporter NY-r19-Gluc Virus in Vitro
Together, these results show that reassortant NY-r19-Gluc virus is a good candidate as a reporter influenza A subtype H3N2 virus for further characterization and application.

Characterization of the Reassortant Reporter NY-r19-Gluc Virus in Vitro
To determine the replication properties of the recombinant virus, the growth kinetics of NY-r19-Gluc virus was evaluated and compared to that of wild-type NY and NY-r19. As shown in Figure 4a, NY-r19 showed a growth pattern similar to that of wild-type NY, while NY-r19-Gluc replicated less efficiently, with a titer roughly 15 times lower than the wild-type virus. The expression of Gluc over the virus growth period of NY-r19-Gluc was also detected. Unlike virus titers, the Gluc activity was not detectable until 36 h p.i., suggesting a slight delay in luminescence kinetics (Figure 4b). However, the Gluc level grew rapidly after that and showed a significant correlation with the accumulation of the infectious viruses (Figure 4c). In addition, the luciferase signal correlated well with the MOI of NY-r19-Gluc infecting MDCK cells (Figure 4d). the Gluc level grew rapidly after that and showed a significant correlation with the accumulation of the infectious viruses (Figure 4c). In addition, the luciferase signal correlated well with the MOI of NY-r19-Gluc infecting MDCK cells (Figure 4d).
Collectively, these data demonstrate that the reporter influenza A subtype H3N2 virus NY-r19-Gluc is replication-competent, and viral proliferation can be monitored by the Gluc assay.

Use of NY-r19-Gluc to Test Group-2-Specific HA Inhibitors
Considering that HA is a promising target for antiviral development and given the subtypespecificity of HA inhibitory small molecules, we tested NY-r19-Gluc against some group-2-specific HA inhibitors. To achieve this, MDCK cells were infected with NY-r19-Gluc at a MOI of 0.1, in presence of TBHQ, an inhibitor specifically targeting group-2 HAs, and arbidol, a broad spectrum HA inhibitor, respectively. At 72 h p.i., the expression level of Gluc was determined to monitor virus replication.
As shown in Figure 5, NY-r19-Gluc virus was sensitive to both TBHQ and arbidol, with an IC50 of 3.4 μM and 0.9 μM, respectively. In contrast, PR8-Gluc virus, which possesses a group-1 HA, was inhibited by arbidol but not by TBHQ. These results indicate that the NY-r19-Gluc virus can be used to evaluate IAV antivirals. Collectively, these data demonstrate that the reporter influenza A subtype H3N2 virus NY-r19-Gluc is replication-competent, and viral proliferation can be monitored by the Gluc assay.

Use of NY-r19-Gluc to Test Group-2-Specific HA Inhibitors
Considering that HA is a promising target for antiviral development and given the subtypespecificity of HA inhibitory small molecules, we tested NY-r19-Gluc against some group-2-specific HA inhibitors. To achieve this, MDCK cells were infected with NY-r19-Gluc at a MOI of 0.1, in presence of TBHQ, an inhibitor specifically targeting group-2 HAs, and arbidol, a broad spectrum HA inhibitor, respectively. At 72 h p.i., the expression level of Gluc was determined to monitor virus replication.
As shown in Figure 5, NY-r19-Gluc virus was sensitive to both TBHQ and arbidol, with an IC 50 of 3.4 µM and 0.9 µM, respectively. In contrast, PR8-Gluc virus, which possesses a group-1 HA, was inhibited by arbidol but not by TBHQ. These results indicate that the NY-r19-Gluc virus can be used to evaluate IAV antivirals. Infections by PR8-Gluc were performed in parallel as a control. Virus infections were monitored by Gluc examination, and the inhibitory effects were analyzed using GraphPad Prism 5.

Discussion
Reporter viruses provide a powerful tool for basic virology studies as well as antiviral development. However, the introduction of foreign genes or modifications of existing viral genes can lead to altered the virological properties, such as delayed replication kinetics or reduced virulence; moreover, such alterations can create an evolutionary pressure, leading to loss of the reporter genes [44]. In the present study, the strategy of fusing reporter genes to the NS1 open reading frame that was successfully used to construct reporter influenza A subtype H1N1 viruses [28,45] failed to generate a reporter influenza A subtype H3N2 virus (Figure 1). This is not surprising, since our previous study demonstrated that the vRNA replication of an IAV segment can be impaired by the insertion of exogenous genes, resulting in an imbalance of the eight segments during genome replication [33]. We speculate that the NS segment of IAV H3N2 is more vulnerable than that of H1N1 to the modification.
A second strategy using the PR8 NS gene instead of that of the NY virus to express NS1-Gluc chimeric protein was also explored in this study, and a reporter influenza A subtype H3N2 virus NY-r18-Gluc was successfully rescued. This reporter virus can express a high level of Gluc protein, and the reporter gene is adequately maintained upon passaging in MDCK cells (Figure 2b). However, we observed a change in CPE in the MDCK cells infected with NY-r18-Gluc virus after passage 4 ( Figure  2c). It is likely that the mutations in the genome of the recombinant virus occurred during the passages, and the mutations are likely residues within the HA gene, since the CPE alteration includes syncytia formation, which is mediated by HA. Thus, this recombinant virus may have limited utility as a research tool.
Using a third strategy, we successfully generated a reporter influenza subtype H3N2 virus NY-r19-Gluc, which is a 6:2 genetic reassortant encoding the HA and NA genes of the NY virus in a backbone comprising the six internal genes of PR8, where a Gluc gene is fused to the NS segment as previously described [28]. Although slightly attenuated compared to wildtype virus, the reassortant virus is replication-competent and stable in vitro; furthermore, the expression level of Gluc protein can be used to monitor virus replication accurately (Figures 3,4).
The usefulness of the NY-r19-Gluc virus as a tool to evaluate subtype-specific HA inhibitors was validated with TBHQ, an inhibitor specifically targeting group-2 HAs, and arbidol, a broad-spectrum HA inhibitor [21,22,24]. The luciferase assay clearly demonstrated that the NY-r19-Gluc virus is sensitive to both TBHQ and arbidol, with IC50 values comparable to the results using traditional methods [21,46,47]. Similarly we think that the NY-r19-Gluc virus can be used to evaluate antibodies or vaccines against IAVs.

Discussion
Reporter viruses provide a powerful tool for basic virology studies as well as antiviral development. However, the introduction of foreign genes or modifications of existing viral genes can lead to altered the virological properties, such as delayed replication kinetics or reduced virulence; moreover, such alterations can create an evolutionary pressure, leading to loss of the reporter genes [44]. In the present study, the strategy of fusing reporter genes to the NS1 open reading frame that was successfully used to construct reporter influenza A subtype H1N1 viruses [28,45] failed to generate a reporter influenza A subtype H3N2 virus (Figure 1). This is not surprising, since our previous study demonstrated that the vRNA replication of an IAV segment can be impaired by the insertion of exogenous genes, resulting in an imbalance of the eight segments during genome replication [33]. We speculate that the NS segment of IAV H3N2 is more vulnerable than that of H1N1 to the modification.
A second strategy using the PR8 NS gene instead of that of the NY virus to express NS1-Gluc chimeric protein was also explored in this study, and a reporter influenza A subtype H3N2 virus NY-r18-Gluc was successfully rescued. This reporter virus can express a high level of Gluc protein, and the reporter gene is adequately maintained upon passaging in MDCK cells (Figure 2b). However, we observed a change in CPE in the MDCK cells infected with NY-r18-Gluc virus after passage 4 ( Figure 2c). It is likely that the mutations in the genome of the recombinant virus occurred during the passages, and the mutations are likely residues within the HA gene, since the CPE alteration includes syncytia formation, which is mediated by HA. Thus, this recombinant virus may have limited utility as a research tool.
Using a third strategy, we successfully generated a reporter influenza subtype H3N2 virus NY-r19-Gluc, which is a 6:2 genetic reassortant encoding the HA and NA genes of the NY virus in a backbone comprising the six internal genes of PR8, where a Gluc gene is fused to the NS segment as previously described [28]. Although slightly attenuated compared to wildtype virus, the reassortant virus is replication-competent and stable in vitro; furthermore, the expression level of Gluc protein can be used to monitor virus replication accurately (Figures 3 and 4).
The usefulness of the NY-r19-Gluc virus as a tool to evaluate subtype-specific HA inhibitors was validated with TBHQ, an inhibitor specifically targeting group-2 HAs, and arbidol, a broad-spectrum HA inhibitor [21,22,24]. The luciferase assay clearly demonstrated that the NY-r19-Gluc virus is sensitive to both TBHQ and arbidol, with IC 50 values comparable to the results using traditional methods [21,46,47]. Similarly we think that the NY-r19-Gluc virus can be used to evaluate antibodies or vaccines against IAVs.
In summary, a stable replication-competent reassortant influenza A subtype H3N2 virus carrying a Gluc gene was generated here. This reporter virus can be adapted as a powerful tool, providing simple and robust assays especially for IAV studies of subtype-specificity.