Identification and Characterization of Novel Antibody Epitopes on the N2 Neuraminidase

ABSTRACT The influenza virus neuraminidase (NA) is becoming a focus for novel vaccine designs. However, the epitopes of human anti-NA antibodies have been poorly defined. Using a panel of 10 anti-N2 monoclonal antibodies (MAbs) that bind the H3N2 virus A/Switzerland/9715293/2013, we generated five escape mutant viruses. These viruses contained mutations K199E/T, E258K, A272D, and S331N. We found that mutations at K199 and E258 had the largest impact on MAb binding, NA inhibition and neutralization activity. In addition, a natural isolate from the 2017-2018 season was found to contain the E258K mutation and was resistant to numerous antibodies tested. The mutation S331N, was identified in virus passaged in the presence of antibody; however, it had little impact on MAb activity and greatly decreased viral fitness. This information aids in identifying novel human MAb epitopes on the N2 and helps with the detection of antigenically drifted NAs. IMPORTANCE The influenza virus neuraminidase is an emerging target for universal influenza virus vaccines. However, in contrast to influenza virus hemagglutinin, we know little about antibody epitopes and antigenic sites on the neuraminidase. Characterizing and defining these sites is aiding vaccine development and helping to understand antigenic drift of NA.

I nfluenza viruses are respiratory pathogens that cause seasonal outbreaks and, occasionally, global pandemics. Annually, influenza viruses cause significant morbidity and mortality (1). Vaccinations against influenza virus are administered seasonally; however, vaccine effectiveness is generally low (20 to 60%) (2). Representative strains to be included in influenza virus vaccines are selected based on the antigenicity of their hemagglutinin (HA), the most abundant glycoprotein on the surface of the virion (3,4). Antibody responses induced by vaccination generally target the hypervariable, immunodominant head domain of the HA (5-7). Anti-HA antibodies have long been considered the "gold standard" of anti-influenza virus immunity since they can readily neutralize the virus and induce-at least in some animal models-sterilizing immunity. Subsequently, influenza virus vaccines are standardized by HA content, whereas the amount of the second viral glycoprotein, the neuraminidase (NA), in any given formulation is variable (8). However, there is increasing evidence that anti-NA immunity can substantially contribute to protection and should also be standardized in influenza virus vaccines (9)(10)(11).
The NA is a sialidase which cleaves terminal sialic acids from N-linked glycans on glycoproteins. The protein is enzymatically active as a homotetramer and has two domains, the head (which contains the active site) and the stalk (12). The NA is primarily involved in viral transmission through the cleavage of decoy receptors in the mucosa, preventing viral aggregation and releasing newly formed virions from infected cells; achieving these functions through its enzymatic activity (13,14). There are nine subtypes of NA which are organized into group 1 (N1, N4, N5, and N8) and group 2 (N2, N3, N6, N7, and N9) (4). Currently, NA inhibitors are prescribed to aid in reducing influenza disease progression and virus transmission. There are several neuraminidase inhibitors on the market that block the enzymatic activity of the NA. These include Relenza (zanamivir), Tamiflu (oseltamivir), Rapivab (peramivir), and Inavir (laninamivir). Unfortunately, influenza viruses can become resistant to these inhibitors, greatly reducing their efficacy. The mutations E119V, R292K, and N294S have been shown to confer oseltamivir resistance in N2 containing viruses (15).
There have been several studies using MAbs to map antigenic regions of the N2 NA (8,22,(26)(27)(28)(29)(30). Early reports described seven "families" of antigenic regions (29). Later studies identified residues that were critical for MAb binding, NAI, and neutralization activity (8,22,(26)(27)(28)30). However, with the exception of Chen et al. (8) and Stadlbauer et al. (22), as well as Powell and Pekosz (31), these studies were performed using murine antibodies and were therefore not a true reflection of the epitopes targeted by human MAbs. Here, we used a panel of MAbs from Chen et al. (8) to define novel epitopes on the human N2 from the isolate A/Switzerland/9715293/2013 using escape mutagenesis. We then characterized escape mutant viruses (EMVs) to examine how each escape mutation impacted the panel's binding, NAI, and neutralization activities. Knowing which residues are primarily targeted by the human antibody response can aid in determining whether novel H3N2 isolates will be antigenically distinct from one another. In addition, an understanding of the anti-N2 antibody response can provide insights for rational vaccine design, which may be critical for future NA containing vaccines.

RESULTS
Establishing a panel of anti-N2 monoclonal antibodies. As stated above, many of the current MAb epitopes identified in literature have been elucidated using murine antibodies (8,22,(26)(27)(28)(29)(30). To complement previous studies, we used several MAbs identified and isolated in Chen et al. (8). For epitope analysis we chose a panel of 10 MAbs: 229-1F06, 229-2E02, 229-2G05, 235-1C02, 235-1E06, 229-2B04, 228-1B03, 229-1G03, 229-1D05, and 229-2C06. Some MAbs, such as 229-2E02 and 229-2B04, had low crossreactivity; however, several MAbs were able to cross-react with avian N2, as well as with N3 and N9 (8). To test NAI activity, we performed an enzyme-linked lectin assay (ELLA) in the presence of MAb and found that all MAbs, aside from 228-1B03, exhibited NAI activity against the H3N2 virus A/Switzerland/9715293/2013, a previous vaccine strain (Table 1). In addition, all MAbs aside from 228-1B03 efficiently neutralized the virus in plaque reduction neutralization assays (PRNAs) ( Table 1). Furthermore, all MAbs aside from 229-1F06 were previously tested for in vivo protection. The MAbs with low cross-reactivity mentioned above were not protective, while the more cross-reactive MAbs were 60 to 100% protective (8). We included an array of MAbs with various properties to aid in understanding if particular types of MAbs target specific regions of the NA.
Escape mutations are located in various regions of the protein. The K199E/T mutations are the closest to the NA active site ( Fig. 1A and B). E258K is located on the bottom of the NA (Fig. 1C). Both A272D and S331N are located on the side of the NA ( Fig. 1A and B). Notably, mutations at E258 and A272 have not been previously reported. The S331N mutation causes the gain of a putative N-linked glycosylation site in the NA. In addition, each EMV contained mutations in several other segments ( Table 3). The irrelevant IgG control virus did not contain any NA mutations; however, it did acquire mutations in its PA, HA, M, and NS1 proteins. Once identified, we decided to further characterize the impact of each NA escape mutation on NA antigenicity and viral fitness.
Escape mutations cause changes in MAb binding, neuraminidase inhibition, and neutralization activities. Using our panel of antibodies, we investigated which NA mutations had effects on MAb binding, neutralization and NA inhibition activities. The NA inhibition and neutralization activities of these antibodies against wild-type virus vary (Table 1). We first tested whether the NA mutations had an impact on binding using an immunofluorescent staining assay. We found that the residues K199 and E258 were the most important for MAb binding ( Fig. 2A). Notably, MAb 228-1B03, which does not have NAI or neutralizing activity, lost binding to the E258K EMV.
We next investigated how each escape mutation affected both the NAI and neutralization activity of each MAb. We found that the mutations in our EMVs conferred     2C and Table 5). We measured neutralization through PRNAs and  Table 5. These results indicate that both the K199 and the E258 residues are critical for a number of MAbs.
Escape mutations have effects on virus growth in vitro. A majority of the identified escape mutations have not been identified in natural isolates. Therefore, we assessed how these mutations alter virus fitness in vitro to determine the impact of each NA escape mutation on viral replication. Each EMV was generated using MDCK cells; however, we performed growth curve assays using both MDCK cells and the human lung epithelial cell line, A549. All EMVs aside from the S331N EMV grew to similar, high levels in MDCK cells with titers peaking around 1 Â 10 7 PFU/ml (Fig. 3A). However, the growth kinetics of each EMV in A549 cells varied (Fig. 3B). Both the K199T and the E258K EMVs grew to higher titers than the irrelevant IgG control virus (with peak titers of 3.12 Â    Table 2). The K199T and E258K EMVs did not contain this HA mutation and also grew to the highest titers in A549 cells, suggesting that N262T is altering viral fitness. Due to poor replication kinetics and low infectivity of recent H3N2 viruses in mice, we were unable to complete in vivo fitness studies (32).

DISCUSSION
The epitopes for human anti-HA MAbs have been studied extensively. However, the N2 antigenic sites for human anti-NA MAbs have not been fully defined. To accurately designate MAb epitopes on the N2, we began by determining whether the mutations found using our panel of antibodies overlapped with those previously identified. We found that position 199 was identified as an important MAb interacting residue by Colman et al. (29), Gulati et al. (30), and Stadlbauer et al. (22).
In our study, we found that 199 was critical for MAbs 229-1D05 and 235-1C02. This residue is located in a loop that "hangs" over the opening to the enzymatic site. Additional studies have confirmed our results and also identified 235-1E06 as another MAb that targets 199 (28)(29)(30). Mutations A272D and S331N have not been identified in other studies. Our A272D EMV showed resistance toward the MAb 229-1F06 in neutralization assays; however, it retained NAI activity. In addition, the S331N EMV did not become resistant to any MAbs in the panel. Despite 331 being identified by Colman et al. (27) as a variable residue, our S331N EMV did not behave as a proper escape mutant. These data suggest that mutations such as A272D and S331N may not be sufficient for complete escape from MAb NAI activity. We also identified the escape mutation at residue 258, which is located on the bottom of the NA. This mutation was important for resistance to 229-1F06 and 229-2C06. In addition, E258K was present in A/New York/PV190/2017 and allowed the virus to completely escape the MAb 229-2C06 (which selected the E258K EMV). This makes E258 a unique target for human antibodies, suggesting that changes at 258 and elsewhere are occurring to avoid antibody responses. The identification of these important sites for MAb binding can aid in the detection of antigenically distinct N2 proteins. In particular, residue 199 seems to be the most critical for binding, NAI and neutralization activity of several MAbs (28)(29)(30). The emergence of mutations in these epitopes could indicate antigenic drift is occurring in viral populations.
In addition, we noticed that escape mutations had a stronger impact on MAb neutralization activity compared to NAI activity. This may be attributed to the mechanisms of neutralization for anti-NA MAbs, which rely on inhibiting NA activity and preventing viral spread rather than preventing initial infection (12,13). NAI activity can be achieved by either directly or indirectly blocking the NA enzymatic site from cleaving sialic acid. However, neutralization activity (if determined through PRNAs) can only be achieved by MAbs with NAI activity that is so potent it prevents viral spread before detectable plaques can form. Our data show that subtle changes in NAI activity for a single MAb may cause much larger changes in neutralization activity. For example, the A272D EMV escaped neutralization by the MAb 229-1F06; however, there was only a 4fold increase in the NAI IC 50 . We did not observe any MAbs that lost NAI activity but retained neutralization activity. Of note, the MAbs with the weakest NAI activity-229-2E02, 229-2G05, 229-2B04, and 228-1B03-did not generate EMVs. For future studies, it would be important to determine how well human sera can inhibit our identified escape mutant viruses, along with new H3N2 isolates, to fully determine the extent of effects of NA mutations on overall anti-NA responses in the human host. By monitoring drift in both HA and NA, we could potentially greatly increase vaccine efficacy and reduce the overall morbidity and mortality caused by influenza virus.
Antibodies. Plasmids encoding each antibody were provided by Patrick Wilson from the University of Chicago. The isolation and reactivity of each antibody has been previously described (8). Antibodies were produced by cotransfection of heavy and light chain plasmids using an ExpiFectamine 293 transfection kit according to manufacturer's instructions (Thermo Fisher). Briefly, heavy-and light-chain plasmids were mixed with ExpiFectamine and incubated for 15 min in Opti-MEM (Gibco). HEK293F cells (Thermo Fisher) were diluted to 7 Â 10 7 cells in 30 ml of Expi293 expression medium (Thermo Fisher). The transfection mix was then added to cells, followed by incubation for 7 days at 37°C with 8% CO 2 with shaking. All antibodies were purified using gravity flow columns packed with protein G-sepharose. They were eluted into a 50-ml Falcon tube with 5 ml of 2 M Tris (pH 10) and concentrated using an Amicon Ultra 30-kDa filter unit (Millipore) (34).
Plaque assays. Virus titers were determined using a standard influenza virus plaque assay. MDCK cells were plated at 8 Â 10 5 cells per ml in a 12-well cell culture plate and incubated overnight at 37°C with 5% CO 2 . The following day, allantoic fluid or cell culture supernatant was diluted 1:10 six times using 1Â minimal essential medium (MEM; 10% 10Â MEM [Gibco], 2 mM L-glutamine [Gibco], 0.1% sodium bicarbonate [Gibco], 10 mM HEPES, 1% 100 U/ml penicillin-100 mg/ml streptomycin solution [Gibco], and 0.2% bovine serum albumin [BSA]). The MDCK cells were washed one time using 1Â phosphate-buffered saline (PBS) and then infected with 200-ml portions of each virus dilution. The cells were then incubated at 33°C with 5% CO 2 for 40 min (the plates were rocked every 10 min). After 40 min, the virus dilutions were aspirated and immediately replaced with an agarose overlay containing 2Â MEM, 0.1% diethylaminoethyl (DEAE)-dextran, 1 mg/ml tolylsulfonyl phenylalanyl chloromethyl ketone (TPCK)treated trypsin, and 0.64% Oxoid agarose. The plates were incubated for 3 days at 33°C with 5% CO 2 . To quantify virus titers, plates were fixed using 3.7% paraformaldehyde (PFA) overnight at 4°C. The overlay was then removed, and the cells were stained with a solution of 20% methanol containing 0.5% crystal violet powder.
Immunofluorescence. MDCK cells were plated in a 96-well cell culture plate at 3 Â 10 4 cells/well and incubated overnight at 37°C with 5% CO 2 . The following day, the viruses were diluted to a multiplicity of infection (MOI) of 5 in 1Â MEM. Cells were washed with 1Â PBS and then infected with 100 ml of diluted virus. The plates were then incubated for 18 h at 33°C with 5% CO 2 . The following day, the cells were fixed using 3.7% PFA at 200 ml per well. For immunofluorescence staining, the PFA was first aspirated from cells and then replaced with 200 ml of blocking solution containing 3% milk (American Bio) diluted in 1Â PBS, followed by incubation for 1 h at room temperature. The blocking solution was removed and replaced with 1% milk. The primary antibodies were diluted to 300 mg in 1Â PBS and then added at a 1:10 dilution to the 1% milk for a final concentration of 30 mg per well. Each plate was then incubated for 1 h at room temperature with shaking. The primary antibodies were then aspirated, and the plate was washed three times using 1Â PBS. The secondary antibody Alexa Fluor 488-goat antihuman IgG(H1L) (Invitrogen) was diluted 1:500 in 1% milk and added to the plate at 100 ml/well. The plate was then incubated for 1 h at room temperature in the dark with shaking. Finally, the secondary antibody was aspirated, and the plate was washed again three times using 1Â PBS. A final 50 ml of 1Â PBS was added to each well to prevent the cells from drying out. We used a Celigo S adherent cell cytometer (Nexcelom Bioscience) with the two-channel "Target 1 1 2" (merge) setting to visualize the immunofluorescence. Exposure time, gain and focus (set using image-based auto focus with the 488-nm signal as the target) were automatically determined by the machine. Fluorescence was calculated using the default analysis settings, and the percent fluorescence was determined based on the wild-type signal. The images are representative of two independent immunofluorescence assays.
Escape mutagenesis. EMVs were generated using wild-type A/Switzerland/9715293/2013. First, virus was diluted to an MOI of 0.01 in 1Â MEM supplemented with 1 mg/ml of TPCK-treated trypsin. Antibodies were diluted to 0.25Â IC 50 and added to the diluted virus (for 1.5 ml/well). The virus-MAb mixture was incubated for 1 h at room temperature before being added to a confluent monolayer of MDCK cells. The cells were incubated for 3 days at 33°C with 5% CO 2 . The cell culture supernatant was then collected spun at 13,000 Â g for 5 min to pellet the cells. Aliquots were stored at 280°C. For subsequent passages, cell culture supernatant was diluted 1:10 in 500 ml of 1Â MEM supplemented with 1 mg/ml of TPCK-treated trypsin and added to a confluent monolayer of MDCK cells. This was incubated at 33°C with 5% CO 2 for 40 min. Then an additional 1 ml of 1Â MEM with 1 mg/ml of TPCK-treated trypsin and 0.5Â IC 50 of MAb was added, and the plates were incubated at 33°C for 3 days. At each passage, the amount of MAb was doubled. Virus was passaged with a MAb up to 128Â IC 50 . Each passage was screened for the presence of EMVs using a plaque assay that contained .128Â IC 50 of antibody in the overlay. Individual plaques were injected into 10-day-old SPF eggs and propagated as described above.
RNA extractions and deep sequencing. The E.Z.N.A viral RNA extraction kit (Omega Bio-Tek) was used for RNA extractions according to the manufacturer's instructions. Isolated RNA was stored at 280°C until sequencing. Using the MiSeq v2, 300-cycle reagent kit (Illumina), next-generation sequencing was performed, and the genome was assembled using a pipeline developed at the Icahn School of Medicine at Mount Sinai (35). Full-length sequences were aligned to the deep sequenced segments of wild-type A/Switzerland/9715293/2013 to identify point mutations using MUSCLE in MEGA 7.0 (36).
Enzyme-linked lectin assay (ELLA). To determine NA activity for each virus, flat-bottom Immulon 4HBX microtiter plates (Thermo Scientific) were coated with 100 ml/well with 25 mg/ml of fetuin (Sigma) diluted in 1Â PBS and incubated overnight at 4°C. The following day, virus was serially diluted (3-fold) in sample dilutant buffer (1Â PBS with 0.9 mM CaCl 2 , 0.5 mM MgCl 2 1% bovine serum albumin (BSA), and 0.5% Tween 20) in a sterile 96-well plate. An additional volume of sample dilutant was then added at a 1:1 ratio. The diluted virus was then incubated for 1 h at room temperature, shaking. The fetuin-coated plates were then washed three times 220 ml of PBS containing 0.1% Tween 20 (PBS-T) per well using an AquaMax 3000 automated plate washer before virus dilutions were transferred to the plate. The samples were then incubated at 33°C for 18 h (overnight). The plates were washed six times with PBS-T, and then 100 ml of peroxidase-conjugated peanut agglutinin (PNA; Sigma) was added at 5 mg/ml, and the plates were incubated for 2 h at room temperature in the dark. PNA was diluted in conjugate dilutant buffer (1Â PBS with 0.9 mM CaCl 2 , 0.5 mM MgCl 2 , and 1% BSA). The PNA was removed, and the plates were washed three times with PBS-T. SigmaFast o-phenylenediamine dihydrochloride (OPD; Sigma) was diluted in water, added at 100 ml per well, and incubated for 7 min at room temperature. Development was stopped by the addition of 50 ml of 3 M hydrochloric acid, and the absorbance at 490 nm was read using a Synergy H1 hybrid multimode microplate reader (Bio-Tek). Prism 7.0 was used to determine the effective concentration of each virus that would yield detectable NA activity. Each ELLA was performed in triplicate.
Neuraminidase inhibition assay. To determine the IC 50 of each MAb, flat-bottom Immulon 4HBX microtiter plates (Thermo Scientific) were again coated with 100 ml/well of 25 mg/ml fetuin (Sigma) diluted in 1Â PBS, and incubated overnight at 4°C. Antibodies were diluted to 120 mg/ml in sample dilutant buffer and then serially diluted 1:3 in a sterile 96-well plate. Virus was diluted to its calculated effective concentration, added to the MAb dilutions at a 1:1 ratio, and incubated for 1 h at room temperature, shaking. Fetuin-coated plates were washed three times with PBS-T, and the virus-MAb dilutions were added. The assay was then performed according to the ELLA procedure above. IC 50 values were determined using Prism 7.0. NAI assays were completed in duplicates.
Plaque reduction neutralization assays. Neutralization IC 50 values were determined using PRNAs. First, MDCK cells were seeded at 8 Â 10 5 cells/ml onto 12-well plates. The following day, MAbs were diluted to 100 mg/ml in 300 ml of 1Â MEM and then serially diluted 1:5 in a 24-well plate to a final concentration of 0.032 mg/ml in 1Â MEM. Virus was diluted to 1 Â 10 3 PFU and added to each of the antibody dilutions (50 ml/well). The virus-MAb mixture was incubated at room temperature for 1 h, shaking. The MDCK cells were then washed one time with 1Â PBS and immediately infected with 200 ml of the virus-MAb mixture and incubated at 33°C with 5% CO 2 , with the plates rocked every 10 min. In the meantime, the overlay was prepared by diluting MAbs to 100 mg/ml in 625 ml of 2Â MEM and then serially diluted 1:5. Then, a mixture of 1Â DEAE-dextrane and 1 mg/ml TPCK-treated trypsin in sterile water for injection (Gibco) was added at 180 ml per well. After the 40 min, the inoculum was aspirated (three wells at a time) and immediately replaced by the overlay mixture containing 360 ml of 2% Oxoid agarose so that the MAb concentration within the agarose matched the same MAb concentration of the inoculum. The plates were incubated at 33°C with 5% CO 2 for 3 days and then fixed with 3.7% PFA overnight at 4°C. The overlay was removed, and the cells were stained as described above. PRNAs were conducted in duplicates.
Growth kinetics. Sterile 24-well plates were seeded with MDCK or A549 cells at 4 Â 10 5 cells/well and incubated overnight at 37°C with 5% CO 2 . The following day, the viruses were diluted to an MOI of 0.01 (5 Â 10 3 PFU/well) in 1Â MEM supplemented with 0.2 mg/ml TPCK-treated trypsin. The cells were washed with 1Â PBS before being infected with virus. An aliquot of this initial virus dilution was kept and used to determine the initial titer of virus at infection. Viruses were incubated for 72 h at 33°C with N2 Neuraminidase Antibody Epitopes January/February 2021 Volume 6 Issue 1 e00958-20 msphere.asm.org 9 5% CO 2 . Cell culture supernatant was sampled every 12 h and frozen at 220°C until virus titers were determined using plaque assays. Each experiment was performed in duplicate.