Impact of Prior Seasonal H3N2 Influenza Vaccination or Infection on Protection and Transmission of Emerging Variants of Influenza A(H3N2)v Virus in Ferrets

Viruses.

The seasonal H3N2 viruses A/Beijing/32/1992 (Beijing/92) and A/Perth/16/2009 (Perth/09) were propagated in the allantoic cavity of 10-day-old embryonated eggs at 34°C for 48 h. The A(H3N2)v virus, A/Indiana/08/2011 (IN/11), was grown in Madin-Darby kidney (MDCK) cells at 37°C for 48 h. IN/11 is a representative H3N2v virus containing the M gene from A(H1N1)pdm09 virus (12). All virus stocks were clarified by centrifugation after collection and stored at −80°C. Stocks were titrated in a standard plaque assay on MDCK cells and expressed as PFU (19).

Whole-virus vaccine preparation.

Virus was concentrated from large volumes of allantoic fluid by ultracentrifugation at 17,000 rpm for 3 h at 4°C. Concentrated viruses were purified over a 30- to 60%-sucrose cushion as previously described (20). The HA gene of each vaccine stock was confirmed by sequencing to ensure no inadvertent mutations were introduced during propagation of new vaccine stocks. Concentrated viruses were inactivated with 0.025% formalin at 4°C for 3 days. The treatment resulted in the complete loss of infectivity of the virus, as determined by titration of vaccine preparations in eggs. The vaccine doses used are expressed as amounts of total protein measured by Bradford assay (Bio-Rad Laboratories, Hercules, CA). Evaluation of the HA protein content of purified vaccines was done using a Coomassie gel system as previously described (21). The HA protein was estimated to make up 40 to 55% of the total protein of purified vaccines.

Ferret immunizations.

All animal experiments were performed in biosafety level 3 laboratories with enhancements (BSL3+) as outlined in Biosafety in Microbiological and Biomedical Laboratories (22).

Animal research was conducted under the guidance of the Centers for Disease Control and Prevention's Institutional Animal Care and Use Committee in an Association for Assessment and Accreditation of Laboratory Animal Care International-accredited animal facility. Adult male Fitch ferrets, 5 to 8 months of age (Triple F Farms, Sayre, PA) and serologically negative for currently circulating influenza viruses as determined by hemagglutination inhibition (HI) assay (described below), were used in this study. Ferrets were vaccinated twice (with 3 to 5 weeks between vaccinations) intramuscularly (i.m.) with an adult human dose (in a 0.5-ml prefilled syringe) of the seasonal (2011-2012) commercially available split-product TIV. In some experiments, ferrets were injected i.m. with 15 μg of inactivated whole-virus vaccine prepared from Beijing/92, Perth/09, or IN/11 virus. Prior to initial vaccination, vaccine boost, and viral challenge, all ferrets were bled for collection of serum and analyzed by HI assay. For determination of prechallenge antibody titers, sera were collected 5 weeks following the final vaccination. For primary Perth/09 or Beijing/92 virus infections, ferrets were first anesthetized with a ketamine-xylazine-atropine cocktail given i.m. followed by a 1-ml intranasal (i.n.) inoculation (500 μl per nostril) of 10⁶ PFU of infectious virus diluted in phosphate-buffered saline (PBS). Nasal wash titers were collected (described below) on day 3 postinoculation (p.i.), and sera were collected at 6 weeks p.i. to confirm viral replication and seroconversion.

Virus challenge and transmission experiments.

Baseline serum, temperature, and weight measurements were obtained before virus inoculations. Temperatures were measured with a subcutaneous (s.c.) implantable temperature transponder (BioMedic Data Systems, Seaford, DE). For A(H3N2)v virus challenges, ferrets were inoculated with 10⁶ PFU of virus as described above. Virus challenges occurred 5 weeks after the final vaccine boost and 6 weeks after prior infection. For respiratory droplet transmission experiments, naive-contact animals were placed in adjacent cages 24 h after challenge using a previously established model (11, 23). Following inoculation, animals were monitored daily for changes in body weight and temperature as well as clinical signs of illness. Nasal washes were collected on alternate days postchallenge with 1 ml of PBS containing bovine serum albumin (BSA) and antibiotics (11, 23). Viral titers of the nasal washes were determined by plaque assay, and sera were collected from the contact ferrets approximately 21 days postcontact (p.c.) to check for seroconversion via HI.

Hemagglutination inhibition.

HI assays were performed using infectious virus stocks or a World Health Organization (WHO) influenza reagent kit (according to instructions) obtained through the Influenza Reagent Resource, Influenza Division, WHO Collaborating Center for Surveillance, Epidemiology and Control of Influenza, Centers for Disease Control and Prevention (Atlanta, GA). Sera were initially treated with receptor-destroying enzyme (RDE) from Vibrio cholerae (Denka Seiken, Tokyo, Japan) overnight at 37°C. The enzyme was then inactivated at 56°C for 30 min and PBS was added to the sera for a final dilution of 1:10. The sera were then absorbed to turkey RBCs (tRBCs) for 30 min at 4°C to remove any nonspecific agglutinins. The HI assay was performed in V-bottom 96-well plates using 4 hemagglutinating units (HAU) of virus and 0.5% tRBCs. Ferret sera were tested for HI antibody titers against a selection of the following antigens: Beijing/92, Perth/09, Ind/11, A/California/07/2009 (pH1N1), and B/Brisbane/60/2008 (influenza B).

Ferret mucosal wash collection.

Both upper respiratory tract (URT; sampled by tracheal/nasal wash) and lower respiratory tract (LRT; bronchoaveolar [lung] wash samples) were collected 5 weeks after prior infection or final vaccination using a novel protocol adapted from protocols established for mice (24). Ferrets were exsanguinated by cardiac puncture under ketamine anesthesia to best avoid blood contamination of the mucosal samples. The chest cavity was opened to expose the lungs and an incision was made to expose the trachea. The trachea was cut just below the larynx and clamped before the entire respiratory tract was carefully removed with the heart. For LRT washes, a volume of 10 ml of PBS containing antibiotics was infused slowly down the trachea into the lung using a pipette. After filling, the lungs were gently massaged and inverted, and the fluid was allowed to drip back through the trachea and collected in 35-mm petri dishes. Approximately 6 to 8 ml of fluid was retrieved and immediately kept on ice. The URT washes from the same animals were collected by flushing 5 ml of PBS with antibiotics through the open tracheal incision and forward into the nasal passages 3 times. The supernatant from lavage fluid was collected following centrifugation at 200 × g at 4°C for 10 min and frozen at −20°C until analyzed by enzyme-linked immunosorbent assay (ELISA). Any sample with blood contamination was excluded from analysis.

ELISA.

Ferret IgG and IgA antibodies were detected using a modified ELISA procedure as previously described (20, 25). Flat-bottom 96-well microtiter plates (Immulon II; Dynatech Laboratories, Chantilly, VA) were coated with 100 HAU per well of purified Perth/09 inactivated virus overnight at 4°C. Serum samples were added to the plates in 4-fold serial dilutions starting at 1:100 for IgG and 1:20 for IgA. LRT/URT samples were added at starting dilutions of 1:10 for IgG and 1:2 for IgA. Bound antibody was detected by IgG horseradish peroxidase (HRP)-conjugated antibody (Bethyl Laboratories, Montgomery, TX) or by a goat anti-ferret IgA followed by a rabbit anti-goat HRP-conjugated antibody (Bethyl). The absorbance was measured at 490 nm following the addition of o-phenylenediamine(dihydrochloride) (OPD) in citrate buffer (Sigma). Titers are expressed as the highest dilution that yielded an optical density greater than the mean plus two standard deviations of similarly diluted PBS control sera.

Statistics.

Statistical significance was determined by one of the following methods: Student's t test, 2-way analysis of variance (ANOVA) with a Bonferroni's posttest, or area under the curve (AUC) with a Mann-Whitney posttest.

Transmission of seasonal H3N2 and A(H3N2)v viruses.

To examine whether seasonal H3N2 vaccination or prior infection affects the transmission efficiency of A(H3N2)v or homologous seasonal H3N2 viruses, we first determined the transmission efficiencies of these viruses in naive ferrets. For respiratory droplet transmission, ferrets were housed in adjacent cages with a perforated side wall, allowing air exchange between ferrets in the absence of direct or indirect contact (11, 23). Ferrets were inoculated with 10⁶ PFU of Perth/09, representative of H3N2 viruses circulating in the 2011-2012 Northern Hemisphere season. Nasal washes were collected every other day, and clinical signs were monitored daily for both inoculated and contact ferrets. All Perth/09 virus-inoculated ferrets exhibited a mean temperature increase of 1.0°C (Table 1) and shed high titers of infectious virus as early as day 1 that were sustained at titers of ≥10^4.0 PFU/ml for 5 days p.i. (Fig. 1A). Although viral titers were detected in the nasal washes of only 2 of the 3 contact ferrets, all 3 animals displayed body temperature elevation and seroconverted to Perth/09 virus (Table 1). Thus, consistent with the experimental transmission data obtained with other seasonal human influenza viruses (11, 23), Perth/09 H3N2 virus exhibited efficient respiratory droplet transmission.

Ferrets inoculated with A(H3N2)v (IN/11) virus exhibited modest weight loss and a maximum temperature increase of 2.2°C on day 2 p.i. (Table 1). Similar to the results described for the seasonal Perth/09 virus, IN/11 virus replicated efficiently in the upper respiratory tract of inoculated ferrets and spread efficiently to contact ferrets (Fig. 1B). The results are consistent with earlier ferret studies which found that A(H3N2)v viruses isolated since 2010 exhibited efficient respiratory droplet transmission (12).

Transmission of A(H3N2)v virus following seasonal TIV immunization.

To test whether seasonal influenza virus vaccination reduces the transmission of A(H3N2)v virus to naive ferrets, we set up experiments in which TIV-immune ferrets were challenged with IN/11 virus and subsequent transmission to naive animals was analyzed. Prior to virus challenge, all nine vaccinated ferrets displayed HI titers of ≥40 to each of the three influenza virus HA antigens present in the 2011-2012 TIV but lacked detectable cross-reactive HI antibodies to IN/11 virus antigen (Table 2). Seasonal TIV immunization did not protect ferrets from weight loss, fever (Table 1), or virus replication (Fig. 2A) following IN/11 virus challenge. As late as 5 days p.i., virus titers of ≥10^2.6 PFU/ml were still present in nasal washes of all vaccinated ferrets, and A(H3N2)v virus spread to six of nine naive-contact ferrets (Fig. 2A and Table 1). Virus transmission occurred with kinetics similar to that observed in naive ferrets, with the majority of contacts displaying viral titers in their nasal washes by day 3 p.c.

A separate group of TIV-immune ferrets were challenged with homologous Perth/09 virus for comparison. Despite prechallenge HI titers of ≥40 to homologous virus (Table 2), TIV did not provide sterilizing immunity against Perth/09 virus challenge. All ferrets displayed viral titers (≥10^2.1 PFU/ml) in their nasal washes for 5 days p.i., and respiratory droplet transmission of Perth/09 virus still occurred; virus spread to four of nine naive-contact ferrets (Fig. 2B), resulting in seroconversion to Perth/09 virus (Table 1).

Impact of vaccination with a historical seasonal H3N2 virus on A(H3N2)v virus transmission.

Serologic studies have previously demonstrated that the young adult population has the highest percentage of individuals with preexisting cross-reactive antibodies to A(H3N2)v virus (8–10). The individuals within this age range would have encountered antigenically related H3N2 viruses through virus infection or vaccination in the 1990s. We first performed antigenic analyses of 6 representative seasonal H3N2 viruses (spanning the years 1990 to 2013) by HI assay and the use of reference ferret antisera. The only seasonal H3N2 virus that displayed limited cross-reactivity with antisera to A(H3N2)v virus was A/Beijing/32/1992 (Beijing/92) virus, a component in the 1993-1994 seasonal TIV. Ferret antisera to Beijing/92 virus infection reacted to low titers (HI titer = 20) of IN/11 virus. These results are consistent with previous antigenic characterization which demonstrated that A(H3N2)v virus retains antigenic properties more similar to human seasonal viruses from the early 1990s than to currently circulating human A(H3N2) viruses (4).

We next assessed the ability of vaccination with inactivated whole-virus monovalent vaccines prepared from Beijing/92, Perth/09, and IN/11 viruses to limit infection and transmission of A(H3N2)v virus. Groups of ferrets were vaccinated i.m. with 15 μg of formalin-inactivated vaccine and two boosts in order to obtain HI titers of ≥40 against the homologous viral strain. HI antibody titers of ≥40 have been correlated with reduction in influenza-like illness in adult populations (26–28). The vaccines failed to elicit cross-reactive HI antibodies against heterologous H3N2 or A(H3N2)v viruses using the standard starting dilution of 1:10 serum (Table 2). However, a repeat of the HI test using a starting dilution of 1:4 revealed that Beijing/92-immune sera gave HI titers of 4 to 8 to IN/11 virus, whereas Perth/09-immune sera had undetectable (<4) HI titers to IN/11 virus (data not shown). Five weeks after the final vaccine boost, ferrets were set up in transmission cages and vaccinated ferrets were challenged with 10⁶ PFU of IN/11 virus. As indicated in Fig. 3A, relatively high viral titers were detected in the nasal washes of Perth/09-vaccinated ferrets and transmission of IN/11 virus occurred in 2 of the 3 contact animals. Furthermore, Perth/09-vaccinated ferrets displayed weight loss and fever consistent with that observed in unimmunized PBS controls (Table 1). Despite the low HI cross-reactivity to IN/11 virus, Beijing/92-vaccinated ferrets also shed high titers of infectious virus and full transmission of IN/11 virus to naive contacts was observed (Fig. 3B). All three of the contacts began to shed virus by day 3 p.c. and seroconverted to A(H3N2)v virus (Table 1). Consistent with unimmunized controls, Beijing/92-vaccinated ferrets displayed a mean maximum weight loss of 7.5% and a fever spike of 1.3°C (Table 1). For comparison, a transmission efficiency experiment with IN/11 virus was performed among homologous IN/11-vaccinated ferrets. We found that in contrast to the lack of protection against IN/11 virus transmission conferred by heterologous vaccination, homologous-virus challenge of IN/11-vaccinated ferrets blunted transmission to the naive-contact ferrets (Fig. 3C). Viral titers were detected only in the nasal washes of the inoculated ferrets on day 1 p.i., and vaccinated animals displayed marginal weight loss (1.6%) and reduced fever (0.8°C) (Table 1). Taken together, our results show that monovalent vaccines prepared from Perth/09 virus or an antigenically related seasonal influenza H3N2 (Beijing/92) virus failed to substantially reduce A(H3N2)v virus shedding and subsequent transmission to naive hosts. Furthermore, the vaccine prepared from the 1992 seasonal H3N2 virus (Beijing/92) offered no added benefit over a contemporary seasonal H3N2 vaccine.

Effect of prior seasonal H3N2 virus infection on cross-protection and transmission of A(H3N2)v virus.

Next, we evaluated the level of cross-protection against A(H3N2)v virus that develops from seasonal H3N2 virus infection. Although seasonal H3N2 viruses are antigenically distinct from the A(H3N2)v viruses, prior influenza virus infection generally induces an effective mucosal immune response and may provide greater cross-protection than that induced by inactivated influenza virus vaccines given parenterally (16, 29). Ferrets were inoculated i.n. with 10⁶ PFU of Perth/09 virus or PBS for the controls and challenged 6 weeks later with either Perth/09 or IN/11 virus. Following homologous-virus challenge with Perth/09, prior infection resulted in significantly lower viral titers in nasal washes than in those of the PBS controls on all days postchallenge (P = 0.004, P = 0.004, and P = 0.003 for days 2, 4, and 6, respectively) (Fig. 4A). Of the 6 ferrets that received primary virus infections, 5 displayed a complete lack of detectable virus in nasal washes. Primary infection with Perth/09 virus also significantly lowered viral titers in nasal washes following heterologous IN/11 virus infection (P = 0.002, P = 0.004, and P = 0.003 for days 2, 4, and 6, respectively) (Fig. 4B). Moreover, prior infection lowered the mean maximum weight loss by more than half (5.9% in PBS controls compared to 2.3% for Perth/09-immune ferrets) (Table 3).

Next we assessed the effect of primary seasonal H3N2 virus infection on the transmission of A(H3N2)v (IN/11) virus to naive animals. For comparison, a group of H3N2-immune ferrets received a homologous Perth/09 virus challenge. No virus shedding was detected from any of the six Perth/09-challenged animals on any day p.i. (data not shown), and immune ferrets displayed minimal weight loss (2.1%) and fever (maximum temperature increase = 0.7°C) (Table 1). Moreover, none of the contact ferrets shed virus (data not shown) or showed a serological response to Perth/09 virus (Table 1). Following heterologous virus challenge, IN/11 virus titers were substantially reduced in Perth/09-immune ferrets and were detected only through day 3 p.i. (Fig. 4C). Only one of six contact ferrets shed infectious virus and seroconverted to IN/11 virus. We also assessed whether primary infection with the antigenically related Beijing/92 virus would provide equivalent or greater protection against A(H3N2)v virus infection. Nasal wash viral titers were markedly reduced in the Beijing/92-immune ferrets and no transmission of IN/11 virus was observed (Fig. 4D). Taken together, the results show that prior seasonal virus infection effectively limited virus growth and blunted the transmission of A(H3N2)v virus.

IgG and IgA in upper and lower respiratory tracts of ferrets following primary virus infection or TIV immunization.

To elucidate the differences observed in the transmission efficiency of A(H3N2)v virus following Perth/09 virus infection versus vaccination, prechallenge sera and mucosal washes were collected and tested for antiviral IgG and IgA antibodies. Immune and control ferrets were necropsied for collection of mucosal washes from the upper respiratory tract (URT) and lower respiratory tract (LRT; bronchoaveolar-lung). In general, IgG and IgA responses among vaccinated and previously infected ferrets were increased compared to those of naive animals. As shown in Table 4, the highest antibody titers were IgG titers observed in the sera and LRTs of ferrets that received prior infections. Lung IgG titers were approximately 10,000-fold higher than those induced by TIV immunization. Serum IgA titers induced by either mode of immunization were similar; however, prior infection induced IgA responses in the URT that were 16-fold higher than those elicited by vaccination. Taken together, the results show that Perth/09 virus infection resulted in consistently higher IgG and URT IgA titers than TIV immunization.