VACCINES AND ANTIVIRAL AGENTS

crossm Vaccine Efficacy of Inactivated, Chimeric Hemagglutinin H9/H5N2 Avian Influenza Virus and Its Suitability for the Marker Vaccine Strategy Se Mi Kim,a,b Young-Il Kim,a,b Su-Jin Park,a,b Eun-Ha Kim,a,b Hyeok-il Kwon,a,b Young-Jae Si,a,b In-Won Lee,a Min-Suk Song,a,b Young Ki Choia,b Department of Microbiology, College of Medicine and Medical Research Institute, Chungbuk National University, Cheongju, South Koreaa; Zoonotic Infectious Diseases Research Center, Chungbuk National University, Cheongju, South Koreab

ABSTRACT In order to produce a dually effective vaccine against H9 and H5 avian

influenza viruses that aligns with the DIVA (differentiating infected from vaccinated animals) strategy, we generated a chimeric H9/H5N2 recombinant vaccine that expressed the whole HA1 region of A/CK/Korea/04163/04 (H9N2) and the HA2 region of recent highly pathogenic avian influenza (HPAI) A/MD/Korea/W452/14 (H5N8) viruses. The chimeric H9/H5N2 virus showed in vitro and in vivo growth properties and virulence that were similar to those of the low-pathogenic avian influenza (LPAI) H9 virus. An inactivated vaccine based on this chimeric virus induced serum neutralizing (SN) antibodies against both H9 and H5 viruses but induced cross-reactive hemagglutination inhibition (HI) antibody only against H9 viruses. Thus, this suggests its compatibility for use in the DIVA strategy against H5 strains. Furthermore, the chimeric H9/H5N2 recombinant vaccine protected immunized chickens against lethal challenge by HPAI H5N8 viruses and significantly attenuated virus shedding after infection by both H9N2 and HPAI H5N8 viruses. In mice, serological analyses confirmed that HA1- and HA2 stalk-specific antibody responses were induced by vaccination and that the DIVA principle could be employed through the use of an HI assay against H5 viruses. Furthermore, each HA1- and HA2 stalk-specific antibody response was sufficient to inhibit viral replication and protect the chimeric virusimmunized mice from lethal challenge with both mouse-adapted H9N2 and wild-type HPAI H5N1 viruses, although differences in vaccine efficacy against a homologous H9 virus (HA1 head domain immune-mediated protection) and a heterosubtypic H5 virus (HA2 stalk domain immune-mediated protection) were observed. Taken together, these results demonstrate that the novel chimeric H9/H5N2 recombinant virus is a lowpathogenic virus, and this chimeric vaccine is suitable for a DIVA vaccine with broadspectrum neutralizing antibody against H5 avian influenza viruses.

Received 31 August 2016 Accepted 15 December 2016 Accepted manuscript posted online 11 January 2017 Citation Kim SM, Kim Y-I, Park S-J, Kim E-H, Kwon H-i, Si Y-J, Lee I-W, Song M-S, Choi YK. 2017. Vaccine efficacy of inactivated, chimeric hemagglutinin H9/H5N2 avian influenza virus and its suitability for the marker vaccine strategy. J Virol 91:e01693-16. https://doi.org/ 10.1128/JVI.01693-16. Editor Jae U. Jung, University of Southern California Copyright © 2017 American Society for Microbiology. All Rights Reserved. Address correspondence to Young Ki Choi, [email protected].

IMPORTANCE Current influenza virus killed vaccines predominantly induce antihem-

agglutinin (anti-HA) antibodies that are commonly strain specific in that the antibodies have potent neutralizing activity against homologous strains but do not crossreact with HAs of other influenza virus subtypes. In contrast, the HA2 stalk domain is relatively well conserved among subtypes, and recently, broadly neutralizing antibodies against this domain have been isolated. Therefore, in light of the need for a vaccine strain that applies the DIVA strategy utilizing an HI assay and induces broad cross-protection against H5N1 and H9N2 viruses, we generated a novel chimeric H9/ H5N1 virus that expresses the entire HA1 portion from the H9N2 virus and the HA2 region of the heterosubtypic H5N8 virus. The chimeric H9/H5N2 recombinant vaccine protected immunized hosts against lethal challenge with H9N2 and HPAI H5N1 March 2017 Volume 91 Issue 6 e01693-16

Journal of Virology

jvi.asm.org 1

Kim et al.

Journal of Virology

viruses with significantly attenuated virus shedding in immunized hosts. Therefore, this chimeric vaccine is suitable as a DIVA vaccine against H5 avian influenza viruses. KEYWORDS H9N2, H5N8, H5N1, avian influenza virus, chimeric vaccine, poultry

A

vian influenza virus (AIV) infection in poultry can cause a range of disease symptoms from asymptomatic infection to respiratory disease and decreased egg production to severe, systemic disease with nearly 100% mortality rates. Genetic features and/or the severity of disease in poultry determines whether the virus is classified as a low-pathogenic avian influenza (LPAI) or a highly pathogenic avian influenza (HPAI) virus. LPAI viruses include viruses of all the subtypes H1 to H16. On the other hand, HPAI viruses have traditionally been of either the H5 or H7 subtype (1). The hemagglutinin (HA) genes of HPAI viruses possess a key virulence determinant in the cleavage site, a site comprised of multiple basic amino acids that is cleavable by furin-like cellular enzymes, which leads to systemic infection and mortality (2). Among the many subtypes, the H9N2 subtype is believed to spread rapidly and has become one of the most prevalent LPAI viruses in the domestic poultry industry. In fact, H9N2 viruses have been become panzootic in Eurasia and worldwide, resulting in great economic losses to the poultry industry due to decreased egg production and moderate to high mortality rates (3–8). Since the mid-1990s, HPAI H5N1 viruses have ravaged domestic poultry in Asia, with sporadic human infections (9). Since 2003, there has been an unprecedented spread of these viruses toward Africa and Europe, giving rise to at least 10 distinct phylogenetic clades based on the positions of their HA genes (10). The number of countries and regions affected by Asian H5N1 HPAI viruses reached a maximum in 2006, with 55 countries and regions being affected (11). Furthermore, recently, Asian H5N8 HPAI viruses spread to some European and, for the first time, North American countries (Canada and the United States), resulting in the emergence of the novel combination of the H5 HA and neuraminidase (NA) subtypes associated with the clade 2.3.4.4 HA and, thus, H5N2, H5N8, and H5N9 HPAI viruses that appeared in North America and Europe (11–13). AIVs of both the H5 and H9 subtypes not only cause seriously economic losses to the poultry industry but also endanger human public health (14, 15). For example, along with the poultry outbreaks caused by Asian H5 HPAI viruses, human infections have been continuously reported since 2003 (16). In fact, the spread of H5N1 was confirmed as the cause of more than 449 deaths worldwide with a fatality rate of ⬃59% in 16 countries (17). In addition, although fatalities have not yet been reported, H9N2 has occasionally been isolated from pigs and humans (18–20). These unprecedented outbreaks of avian influenza viruses of the H5 and H9 subtypes and zoonotic infections threaten not only the poultry industry but also public health. Therefore, avian H9N2 and HPAI H5N1 viruses are considered some of the top threats on the WHO list of pandemic candidates. Most human cases of H9N2 and HPAI H5N1 virus infections are the result of direct contact with infected poultry (21). Therefore, to protect the poultry industry and reduce the risk of human infection through direct contact with infected poultry, some countries have used various commercial HPAI H5N1 and LPAI H9N2 virus poultry vaccines (22–24). However, the H5N1 and H9N2 vaccines have not been effective despite the large amount of inactivated vaccines used in areas of endemicity (25–29). There are several major challenges for vaccinations of poultry against H5 and H9 avian influenza viruses. The first challenge is that vaccine production cannot keep up with the rapid changes in influenza virus, resulting in mismatches between strains used in vaccines and locally circulating viruses (22, 23). Therefore, a successful avian vaccine strain should have high cross-reactivity with cocirculating AIV strains. The second hurdle for avian influenza vaccines is the ability to screen out vaccinated animals from infected animals. This is commonly recommended for poultry influenza vaccines when there is a risk of major spread and depopulation is not feasible or desirable (30). Thus, to align with the DIVA (differentiating infected from vaccinated March 2017 Volume 91 Issue 6 e01693-16

jvi.asm.org 2

Efficacy of Chimeric HA H9/H5N2 DIVA Vaccine

Journal of Virology

animals) strategy, a vaccine needs to contain additional critical factors, such as adaptations of heterologous NA, nucleoprotein (NP), or nonstructural protein (NS) genes, to differentiate between infected and vaccinated animals. However, the assay for neuraminidase inhibition and enzyme-linked immunosorbent assays (ELISAs) to screen for the differentiation of NP or NS proteins are relatively complicated and require specific equipment. In comparison, an hemagglutination inhibition (HI) assay is relatively simple and easy to use with immunized sera. Current killed influenza virus vaccines predominantly induce anti-HA antibodies that specifically target antigenic sites in the globular head domain of the HA1 region and block receptor binding (31, 32). These responses are commonly strain specific in that the antibodies have potent neutralizing activity against homologous strains but do not cross-react with HAs of other influenza virus subtypes (33). In contrast, the HA2 stalk domain, although less immunogenic (6, 34), is group specific (group 1 HA and group 2 HA) and relatively well conserved among different subtypes (35). Furthermore, the HA2 stalk domain is phylogenic HA group specific, and recently, broadly neutralizing antibodies against this domain have been isolated (33, 34). Hence, in light of the need for a vaccine that applies the DIVA strategy utilizing an HI assay and that induces broad cross-protection against H5N1 and H9N2 viruses, we generated a novel chimeric H9/H5N2 virus that expresses the entire HA1 portion from H9N2 and the HA2 region of heterosubtypic H5N8 virus. These chimeric recombinant viruses possess growth properties similar to those of wild-type influenza virus. Moreover, an inactivated vaccine based on this chimeric virus protected chickens against clinical disease and significantly attenuated virus shedding after H9N2 and HPAI H5N1 virus challenge infections. Serological analyses confirmed that HA1- and HA2 stalk-specific antibody responses were induced by vaccination and complied with the DIVA principle by an HI assay. Furthermore, each HA1- and HA2 stalk-specific antibody response was sufficient to inhibit viral replication and protect chimeric virus-immunized mice from lethal challenge with either mouse-adapted (ma) H9N2 (maH9N2) or wild-type HPAI H5N1 viruses. These results suggest that the HA2 stalk region of the H5 HA protein contains functional epitopes that are attractive targets for a broad-spectrum H5 influenza virus vaccine and that this chimeric vaccine is suitable as a DIVA vaccine against H5 avian influenza viruses. RESULTS Generation of chimeric HA constructs and chimeric H9/H5N2 virus. Although the HA globular head domains of each group exhibit limited homology across HA subtypes, the HA2 ectodomains are highly conserved among various influenza virus subtypes. The level of amino acid homology between HA2 regions of H5 or H9 viruses was as high as ⬎95% within the same HA subtype, while only 64 to 65% identity between the HA2 regions of H9 and H5 viruses was observed (Fig. 1C). Thus, to develop a multisubtypic vaccine through reverse genetics, we generated a chimeric HA (cHA) H9/H5 construct comprised of the entire HA1 domain from A/CK/Korea/04163/04 (H9N2) (04163/04) and the HA2 region from a A/MD/W452/14 (H5N8) (W452/14) (Fig. 1B and C). Briefly, each HA1 and HA2 region was amplified with subtype-specific primer sets, and the two segments were combined in the cHA H9/H5 gene by overlapping PCR and then cloned into vpHW2000 (36). The chimeric H9/H5N2 virus was generated in the 04163/04 (H9N2) backbone by using a reverse-genetics method as described previously (36). The cHA H9/H5N2 virus has low-pathogenic characteristics similar to those of the parental H9N2 virus. We compared biological features of the cHA H9/H5N2 virus with the parental viruses 04163/04 (H9N2) and W452/14 (H5N8). No significant difference was observed in replication titers of the cHA H9/H5N2 virus compared with the parental 04163/04 (H9N2) and W452/14 (H5N8) viruses (Table 1), as the cHA H9/H5N2 virus reached 7.2 log10 50% tissue culture infective doses (TCID50)/ml in Madin-Darby canine kidney (MDCK) cells and 8.5 log10 50% egg infectious doses (EID50)/ml in chicken embryo eggs. Next, to investigate the pathogenic potential of the cHA H9/H5N2 virus, March 2017 Volume 91 Issue 6 e01693-16

jvi.asm.org 3

Kim et al.

Journal of Virology

FIG 1 Construction strategy for the chimeric cHA H9/H5N2 plasmid and alignment of HA2 regions. (A) The HA0 protein is composed of HA1 and HA2 domains from different viruses separated by the cleavage site and fusion peptide (A). (B) Schematic representation of the chimeric H9/H5HA protein used in this study. Fusion of HA1 of A/CK/Korea/04163/04 (H9N2), containing the cleavage sites, and the HA2 region of A/MD/Korea/W452/14 (H5N8), including the fusion peptide, was achieved by overlapping PCR with specific primer sets, and the construct was then cloned into vpHW2000. (C) The amino acid alignment and percent identities of HA2 regions of the 04163/04 (H9N2), HC09/09 (H9N2), W452/14 (H5N8), and ma81/07 (H5N2) viruses were analyzed by using the DNAStar Lasergene sequence analysis software package, version 5.0 (DNAStar, Madison, WI).

we examined the 50% lethal dose in chickens and mice. As expected, the parental 04163/04 (H9N2) virus did not induce mortality in mice or chickens. Moreover, the cHA H9/H5N2 virus did not induce mortality even at the maximum infection doses in mice (⬎7.2 log10 EID50/ml [50% mouse lethal dose {MLD50}]) and chickens (⬎8.5 log10 EID50/ml [50% chicken lethal dose {CLD50}]). In contrast, the HPAI W452/14 (H5N8) virus showed 5.8 log10 EID50/ml for the MLD50 and 2.5 log10 EID50/ml for the CLD50, which was in agreement with data from a previous study (37).

TABLE 1 Comparison of growth properties and virulence of each tested virus in vitro and in vivoa Viral titer (log10 TCID50/ml) Virus 01310/01 04163/04 W452/14 cHA H9/H5N2 ma163/04 ma81/07 aNA,

Strain H9N2 H9N2 H5N8 H9/H5N2 H9N2 H5N2

With trypsin 7.2 7.2 6.8 7.2 7.5 6.5

Without trypsin ⬍2.5 ⬍2.5 6.0 ⬍2.5 ⬍2.5 ⬍2.5

Viral titer (log10 EID50/ml) 9.5 9.5 8.9 8.5 9.5 7.5

MLD50 (log10 EID50/ml) NA ⬎7.2 5.8 ⬎7.2 6.5 2.6

CLD50 (log10 EID50/ml) NA ⬎9.5 2.5 ⬎8.5 ⬎9.5 ⬎7.5

not applicable.

March 2017 Volume 91 Issue 6 e01693-16

jvi.asm.org 4

Efficacy of Chimeric HA H9/H5N2 DIVA Vaccine

Journal of Virology

FIG 2 Serum antibody responses in chickens administered the 04163/04 (H9N2) and cHA H9/H5N2 vaccines. Groups of chickens were vaccinated with cHA H9/H5N2, 04163/04 (H9N2), or PBS-alum (mock) twice as described in Materials and Methods, and sera were collected 2 and 4 weeks after the first vaccination. Serum antibody responses were evaluated by HI (A) and serum neutralization (B) assays with cHA H9/H5N2, 04163/04 (H9N2), HC09/09 (H9N2), W452/14 (H5N8), and ma81/07 (H5N2) viruses. The detection limit of each test was ⬍10 HI.

The cHA H9/H5N2 vaccine induces H9-specific HI antibodies with broad SN titers against H5N8 virus in chickens. To test the immunogenicity and protective ability of cHA H9/H5N2 vaccines, groups of specific-pathogen-free (SPF) chickens (influenza A virus negative by an HI assay) were immunized twice with the cHA H9/H5N2 vaccine, the H9N2 (04163/04) vaccine, or the phosphate-buffered saline (PBS)-alum control (mock). Two weeks after each vaccination, chicken sera were collected to determine the mean HI and serum neutralizing (SN) titers against homologous H9N2 and heterologous H5 viruses (Fig. 2). All vaccinated groups showed increases in mean HI titers following boosting vaccinations against the homologous strain, except for the mock-infected group (Fig. 2A). Briefly, H9N2 immunization induced a relatively high HI geometric mean titer (GMT) (HI titer of 806) against homologous 04163/04 (H9N2) strains and a moderate GMT (HI titer of 201) against recent heterologous HC09/09 (H9N2) strains. However, the H9N2 vaccine does not induce a cross-reactive HI titer against heterosubtypic H5 viruses (W452/14 [H5N8] and ma81/07 [H5N2] viruses) (Fig. 2A). The cHA H9/H5N2-immunized groups showed an HI GMT of ⬎40 against homologous and heterologous H9N2 viruses even after one dose of the vaccine. In addition, after the second immunization with the cHA H9/H5N2 vaccine, HI antibodies increased and were at least 2-fold higher against both homologous cHA H9/H5N2 and 04163/04 (H9N2) (HI GMT, ⬎650) and heterologous H9N2 (HI GMT, 172) (HC09/09) strains. However, the cHA H9/H5N2 vaccine could not induce detectable HI titers against H5 strains. Moreover, mock-vaccinated chickens had HI titers against all five viruses that were below the limit of detection, and no groups of chickens showed protective antibody titers against H5 viruses. To evaluate whether the above-noted HI antibodies could neutralize influenza virus, serum samples were tested by a microneutralization (MN) assay (Fig. 2B). The H9N2 vaccine (04163/04) groups showed SN titers ranging from 320 to 1,280 against homologous and heterologous H9N2 strains, including the cHA H9/H5N2 virus. However, this vaccine did not elicit any detectable SN titer against H5 strains (Fig. 2B). In contrast, the cHA H9/H5N2-immunized groups elicited relatively high SN titers (320 to 1,280) against a homologous cHA H9/H5N2 virus and various H9N2 strains. Furthermore, it is noteworthy that the cHA H9/H5N2-immunized groups showed SN titers of 20 to 80 against heterosubtypic W452/14 (H5N8) and ma81/07 (H5N2) strains (Fig. 2B). None of the H9N2- or mock-vaccinated groups elicited detectable titers against H5 viruses. These results suggest that the cHA H9/H5N2 vaccine could not induce an HI titer against the HA2 region (H5 stem region) of cHA H9/H5 strains, although it induced strong neutralization activity against the A/MD/Korea/W452/14 (H5N8) virus. The cHA H9/H5N2 vaccine induces an IgY response against H5N8 virus in chickens. Although HI and SN titers provide a representative picture of the elicited March 2017 Volume 91 Issue 6 e01693-16

jvi.asm.org 5

Kim et al.

Journal of Virology

FIG 3 Qualitative IgY responses elicited by H9N2 and cHA H9/H5N2 vaccines in chickens. An ELISA was adapted to detect the qualitative IgY responses of each vaccine against the 04163/04 (H9N2) (A) and W452/14 (H5N8) (B) viruses and the purified sM2-HA2 protein of H5N8 (C) with sera from each group of vaccinated chickens. Sera from each group of immunized chickens were incubated with proteins from 04163/04 (H9N2) or W452/14 (H5N8) or with the sM2-HA2 protein of H5N8 as described in Materials and Methods. Sera from PBS-alum-immunized chickens were used as a negative control (naive sera). Asterisks indicate samples significantly different from those of the mock group (*, P ⬍ 0.05; **, P ⬍ 0.001; ***, P ⬍ 0.0001) as determined by a t test.

immune response, qualitative aspects of the antibodies elicited by vaccination (i.e., IgY avidity) may also influence protective efficacy (33). Therefore, we determined endpoint titers of anti-HA IgY responses against the 04163/04 (H9N2) and W452/14 (H5N8) viruses (Fig. 3). Similarly to the HI and MN assays, all vaccines induced reasonable IgY responses against the homologous 04163/04 (H9N2) strain (Fig. 3A), but the H9N2 vaccine did not elicit an IgY response against the heterosubtypic W452/14 H5N8 virus above the baseline (Fig. 3B). However, the cHA H9/H5N2-vaccinated groups showed high IgY responses against both H9N2 and H5N8 viruses (Fig. 3). For example, the cHA H9/H5N2 vaccine induced an IgY response (above an optical density [OD] value of 1.5) almost comparable to that of the H9N2 vaccine against the 04163/04 (H9N2) strain (Fig. 3A). Furthermore, only the cHA H9/H5N2 vaccine induced a positive IgY response against heterosubtypic strain W452/14 (H5N8) (as high as an OD of 1.5) (Fig. 3B). To confirm whether the antibodies produced were specific for the HA2 region of H5N8, an additional ELISA using the sM2-HA2 region of H5N8 was conducted with each of the chicken sera. Compared with the H9N2-vaccinated group, sera from cHA H9/H5N2vaccinated birds showed significantly high IgY responses against H5N8 HA2-specific antigen (Fig. 3C). These results suggest that the cHA H9/H5N2 vaccine induces strong serum IgY responses against both the HA1 and HA2 regions in immunized chickens. The cHA H9/H5N2 vaccine cross-protects vaccinated chickens against H9N2 and H5N8 viruses. To demonstrate whether the chimeric H9/H5N2 viruses could also elicit cross-protection against challenge with a heterologous virus, each vaccinated group of chickens was challenged with 109.5 EID50 of A/CK/Korea/04163/04 (H9N2) (the maximum titer) or 104.5 EID50 of W452/14 (H5N8) (100 CLD50) in a volume of 0.5 ml through the intranasal route. During the chicken studies, no marked clinical disease March 2017 Volume 91 Issue 6 e01693-16

jvi.asm.org 6

Efficacy of Chimeric HA H9/H5N2 DIVA Vaccine

Journal of Virology

FIG 4 Survival rates of 04163/04 (H9N2)- and cHA H9/H5N2-vaccinated chickens. Both vaccination doses contained 3.5 ␮g HA/dose with an alum adjuvant. The efficacy of vaccination was verified by the survival rate following infections. Groups of immunized chickens were intranasally challenged with the 04163/04 (H9N2) (109.5 EID50) (A) or W452/14 (H5N8) (104.5 EID50) (B) virus 2 weeks after the last vaccination. Survival was recorded for 14 dpi. Asterisk indicate results that are significantly different from those of the control (*, P ⫽ 0.028) as determined by a t test.

symptoms were observed in mock- or H9N2-infected birds (Fig. 4A). The chickens in the mock-infected group showed significantly high virus titers in lung and nasal and cloacal swabs from 3 days postinfection (dpi), which persisted until 5 dpi (2.7 to 4.3 log10 EID50/ml), compared with those of other vaccine groups (Fig. 5A to C). However, the H9N2- and cHA H9/H5N2-vaccinated groups showed significantly attenuated viral titers in lung and nasal and cloacal swabs compared with the other groups, and virus was not detected in any of the three specimens at 5 dpi (Fig. 5B and C).

FIG 5 Viral titrations in H9N2- and cHA H9/H5N2-vaccinated chickens following challenge with the wild-type 04163/04 (H9N2) virus. After intranasal virus challenge with 04163/04 (H9N2), lungs (A), nasal swabs (B), and cloacal swabs (C) were harvested at 3 and 5 days postinfection. Virus titers are expressed as log10 EID50 per milliliter. The limit of virus detection was set to 0.7 log10 EID50/g or 0.7 log10 EID50/ml. Statistical significance compared to mock vaccination was determined by a t test (*, P ⬍ 0.05; **, P ⬍ 0.001; ***, P ⬍ 0.0001). March 2017 Volume 91 Issue 6 e01693-16

jvi.asm.org 7

Kim et al.

Journal of Virology

FIG 6 Viral titrations in H9N2- and cHA H9/H5N2-vaccinated chickens following challenge with the wild-type HPAI 452/14 (H5N8) virus. After intranasal challenge with HPAI W452/14 (H5N8) virus, tracheal and cloacal swabs were collected at 3 and 5 dpi. In addition, lung, brain, kidney, spleen, heart, liver, and colon were also harvested from the PBS-alum (A), H9N2 (B), and cHA H9/H5N2 (C) vaccine groups. Virus titers are expressed as log10 EID50 per milliliter. The limit of virus detection was set to 0.7 log10 EID50/g. † indicates that there were no samples collected because the chickens in this group died.

In the W452/14 HPAI (H5N8) virus infection study, the mock and H9N2 vaccine groups of chickens all died within 5 dpi (Fig. 4B). However, the cHA H9/H5N2 vaccine group showed only 20% (1/5) mortality even when the birds were inoculated with a dose as high as 100 CLD50. However, mild to moderate clinical symptoms, such as decreased morbidity and food consumption, were observed in W452/14-infected birds at 3 dpi. Moreover, the remaining birds started to recover by 5 dpi and fully recovered over the 14-day period (Fig. 4B). There were no significant differences in virus titers in each specimen between the mock and H9N2 vaccine groups (Fig. 6A and B). The highest virus titer was observed in infected lungs (as high as 6.5 log10 EID50/g) of the mock and H9N2 vaccine groups at 3 dpi (Fig. 6A and B). Since HPAI H5N8 virus causes systemic infection (37), three birds in each group were euthanized on days 1, 3, and 5 postchallenge, and organ tissue samples (brain, kidney, spleen, heart, liver, and intestine) were collected for virus titrations. Besides the respiratory tracts, the W452/14 virus was detected in brain (3.0 to 4.5 log10 EID50/g), kidney (5.0 to 5.4 log10 EID50/g), spleen (2.4 to 2.5 log10 EID50/g), heart (6.1 to 6.8 log10 EID50/g), liver (1.5 to 3.6 log10 EID50/g), and colon (1.5 to 2.5 log10 EID50/g) in both the H9N2 and mock vaccine groups (Fig. 6A and B). In contrast, the cHA H9/H5N2 vaccine group showed significantly attenuated virus titers in all organs tested, and virus was not detected in any tissue or swab specimens at 5 dpi. Moreover, virus was not detected in brain tissues even at 3 dpi in the cHA H9/H5N2 vaccine group (Fig. 6C). The cHA H9/H5N2 virus is not pathogenic in mice, and its use as a vaccine induces a strong antibody response against H9 and H5 viruses. To confirm the broad spectrum of cross-protection of the cHA H9/H5N2 vaccine in a mammalian model, BALB/c mice were used to determine the pathogenic potential and crossprotective efficacy of cHA H9/H5N2 against H9N2 and HPAI H5 viruses. The MLD50 was March 2017 Volume 91 Issue 6 e01693-16

jvi.asm.org 8

Efficacy of Chimeric HA H9/H5N2 DIVA Vaccine

Journal of Virology

FIG 7 Serum antibody responses in mice administered the 04163/04 (H9N2) and cHA H9/H5N2 vaccines. Groups of mice were treated with cHA H9/H5N2, A/CK/Korea/04163/2004, or mock treatment twice as described in Materials and Methods, and sera were collected 2 and 4 weeks after the first vaccination. Serum antibody responses were evaluated by HI (A) and serum neutralization (B) assays with the cHA H9/H5N2, 04163/04 (H9N2), HC09/09 (H9N2), W452/14 (H5N8), and ma81/07 (H5N2) viruses. The detection limit of each test was ⬍10 HI.

determined in order to quantify the virulence of the cHA H9/H5N2 virus and compare it with those of the wild-type and mouse-adapted strains. The MLD50 values for ma163/04 (H9N2), W452/14 (H5N8), and ma81/07 (H5N2) were 106.5, 105.8, and 102.6 EID50 of the stock virus titers, respectively. However, the maximum titer of cHA H9/H5N2 (107.2 EID50) did not induce mortality in mice (Table 1). To assess the immunogenicity of the cHA H9/H5N2 vaccine in mice, mice were immunized twice with the inactivated H9N2 or cHA H9/H5N2 vaccine intramuscularly (i.m.) 14 days apart. Similar to the chicken data, both vaccines induced strong HI titers against homologous strains (HI GMT of 320 to 640) and showed broad spectra of HI titers against the heterologous HC09/09 (H9N2) strain (HI GMT of 160). However, neither the H9N2 nor the cHA H9/H5N2 vaccine induced detectable HI titers against heterosubtypic H5 viruses in sera from immunized mice (Fig. 7A). In an SN assay, sera from the H9N2 vaccine-immunized mice induced mean SN titers of 320 to 640 against homologous 04163/04 strains and also induced mean MN titers of ⬎160 against the heterologous HC09/09 strain. Interestingly, sera from cHA H9/H5N2-immunized mice showed strong SN titers against both H9N2 strains and heterosubtypic H5 viruses, even with only one vaccination. Sera from boosted mice showed a mean MN titer of ⬎160 against all H9N2 viruses and induced MN titers of ⬎80 against two different heterosubtypic H5 strains (Fig. 7B). In an ELISA, all vaccines induced reasonable IgG responses against the homologous 04163/04 H9N2 strain (Fig. 8A), while the H9N2 vaccine did not elicit any IgG response above the baseline against the heterosubtypic W452/06 (H5N8) virus (Fig. 8B). In contrast, the cHA H9/H5N2-vaccinated groups showed the highest IgG response against both H9N2 and H5N8 viruses (Fig. 8). To confirm whether the antibodies produced were specific for the HA2 region of H5N8, an ELISA specific for the sM2-HA2 region of H5N8 was conducted with sera from each mouse. Compared with the H9N2-vaccinated group, the sera from cHA H9/H5N2-vaccinated mice showed significantly high IgG responses against an H5N8 HA2-specific antigen (Fig. 8C). These results suggest that the cHA H9/H5N2 vaccine induces strong serum IgG responses against both the HA1 and HA2 regions in immunized mice. The cHA H9/H5N2 vaccine cross-protects vaccinated mice against H9N2 and H5Nx viruses. Next, to assess the protective efficacy of the inactivated cHA H9/H5N2 and H9N2 vaccines, each group of mice was inoculated with 10 MLD50 of the ma163/04 (H9N2), W452/14 (H5N8), or maW81/07 (H5N2) virus through the intranasal route. Daily body weight was monitored over the course of infection (14 days). The ma163/04 (H9N2)-infected mice presented clinical manifestations, including conspicuous body weight loss at 3 dpi, and all mice in the mock group died by 12 dpi (Fig. 9A). However, the H9N2- and cHA H9/H5N2-vaccinated groups survived, with ⬍10% body weight loss, following ma163/04 (H9N2) infection. In the W452/14 (H5N8) challenge study, the H9N2 and mock groups of mice exhibited ⬎20% decreases in body March 2017 Volume 91 Issue 6 e01693-16

jvi.asm.org 9

Kim et al.

Journal of Virology

FIG 8 Qualitative IgG responses following H9N2 and cHA H9/H5N2 vaccination of mice. An ELISA was adapted to detect the qualitative IgG responses of vaccines against the 04163/04 (H9N2) (A) and W452/14 (H5N8) (B) viruses and the purified sM2-HA2 protein of H5N8 (C). Sera from each group of immunized mice were incubated with purified proteins from 04163/04 (H9N2) or W452/14 (H5N8) or the sM2-HA2 protein of H5N8 as described in Materials and Methods. Sera from mock-immunized mice was used as a negative control (naive sera). Asterisks indicate samples that were significantly different from the samples of the mock group (*, P ⬍ 0.05; **, P ⬍ 0.001; ***, P ⬍ 0.0001) as determined by a t test.

weight, and all mice died by 13 dpi (Fig. 9B). In contrast, cHA H9/H5N2-immunized mice showed only 10% mortality (1 of 10), with 10 to 12% body weight loss, but the rest (9 of 10) recovered their body weight starting at 9 dpi. To confirm the broad cross-protection of the cHA H9/H5N2 vaccine against heterologous H5 viruses, we additionally challenged mice with highly lethal strain ma81/07 (H5N2) (Fig. 9C) (38). Surprisingly, 80% (8/10) of the cHA H9/H5N2-immunized mice survived infection with the mouse-adapted ma81/07 (H5N2) virus for 14 dpi and recovered their body weight beginning at 8 dpi. This mortality rate is almost comparable to that of the homologous strain (W452/14 [H5N8]), suggesting the broad cross-protective effect of the HA2 region even at 10 times the MLD50. To investigate the association between mouse survival rates and virus titers, five mice in each group were euthanized at 1, 3, 5, 7, and 9 dpi, and tissues were collected for virus titration. Infected mice in the mock group showed the highest lung viral titers (3.3 to 4.4 log10 EID50/g) and spread of the W452/14 virus into brain (1.2 to 2.0 log10 EID50/g) (Table 2). In the H9N2 vaccine groups, the lung viral titer of the ma163/04 virus was significantly attenuated at 3 dpi compared to that in the mock group (P ⬍ 0.005), and the virus was cleared by 5 dpi. However, there were no significant differences in virus titers in each specimen following infection with the ma81/07 and W452/14 viruses between the mock and the H9N2 vaccine groups. Moreover, the W452/14 virus was detected even in brain (0.7 to 1.6 log10 EID50/g) at titers almost comparable to those in the mock group (1.2 to 2.0 log10 EID50/g). In contrast, the cHA H9/H5N2 vaccine group showed significantly attenuated viral titers in all tissues tested, and the virus was not detected in any tissues by 7 dpi. Following maH9N2 infection, the virus was cleared by as early as 5 dpi. It is noteworthy that the virus was not detected in brain tissues at any time point, even in the case of HPAI H5N8 virus infection. March 2017 Volume 91 Issue 6 e01693-16

jvi.asm.org 10

Efficacy of Chimeric HA H9/H5N2 DIVA Vaccine

Journal of Virology

FIG 9 Survival rates and weight loss of 04163/04 (H9N2)- and cHA H9/H5N2-vaccinated mice. Groups of mice immunized with each vaccine were challenged intranasally with 10 MLD50 of mouse-adapted strain 04163/04 (H9N2) (107.5 EID50) (A), W452/14 virus (H5N8) (106.8 EID50) (B), or maW81 (H5N2) (103.6 EID50) (C) after a booster vaccination. Body weight is presented as a percentage of the animal’s weight on the day of inoculation. The data are presented as the means ⫾ the standard errors of the means. Survival was recorded for 14 dpi. Asterisks indicate samples that were significantly different from those of the control (*, P ⬍ 0.05; **, P ⬍ 0.005; ***, P ⬍ 0.0001) as determined by a t test.

DISCUSSION Of the numerous influenza virus subtypes, LPAI H9N2 viruses have been circulating in multiple avian species in Eurasia, resulting in great economic losses to the poultry industry due to decreased egg production and moderate to high mortality rates (5, 7). In addition, H9N2 influenza viruses have become panzootic in Eurasia during the past decade and have been isolated from terrestrial poultry worldwide. Besides the H9N2 viruses, HPAI H5 viruses have become widespread, causing infection of wild waterfowl and domestic poultry in Central and South Asia, the Middle East, Europe, and Africa (4, 10, 12–15, 21, 39). In consideration of the veterinary and public health burdens caused by H9N2 and HPAI H5 viruses, the main objective of this study was to develop a broad, cross-protective, dual H5 and H9 vaccine with an easy-to-implement DIVA strategy. To this end, we adapted a novel chimeric H9/H5 HA segment that expresses the entire HA1 portion from H9N2 virus and the HA2 region of heterosubtypic H5N1 virus. Previously, one group reported an approach for the generation of chimeric HA March 2017 Volume 91 Issue 6 e01693-16

jvi.asm.org 11

March 2017 Volume 91 Issue 6 e01693-16

Lung Brain

Lung

W452/14 (H5N8)

ma81/07 (H5N2)

6.8

—b

3.3 ⫾ 0.3

3.0 ⫾ 0.2 —

5.6 ⫾ 0.2

2.1 ⫾ 0.3 —b

5.1 ⫾ 0.5

5.5

4.1 ⫾ 0.5 0.7 3.2 ⫾ 0.2

2.0 ⫾ 0.2 — 2.4 ⫾ 0.5

1.8 ⫾ 0.3 —

5 —

3.8 1.6

3 2.4

3.0 ⫾ 0.2 —

9 —

3.6 ⫾ 0.2 —

7 —

1 3.3 ⫾ 0.3

5 —

3 1.6

1 3.3 ⫾ 0.3

—

— —

7 —

—

— —

9 —

5.1 ⫾ 0.5

3.6 ⫾ 0.2 —

1 6.1

5.1 ⫾ 0.5

3.3 ⫾ 0.3 —

3 5.6

Mock group

5.0 ⫾ 0.7

3.4 ⫾ 0.5 2.0

5 5.0

5.6

4.4 ⫾ 0.4 1.2

7 5.6

indicate that the tissue was negative for virus detection (lower limit of 0.7 log10 EID50/g). Viruses were titrated in egg. Virus titrations are shown only for virus detected in multiple organs; except for W149 and W452, other viruses were not detected in multiple organs. bMice that died.

aDashes

Organ Lung

Challenge virus ma163 (H9N2)

cHA H9/H5N2-vaccinated group

H9N2-vaccinated group

Mean virus titer (EID50/g) ⴞ SD at day postinfection

TABLE 2 Comparison of tissue virus titers in mice immunized with each vaccine following lethal challenge with H9N2, H5N2, and H5N8 virusesa

—b

—b —b

9 —b

Kim et al. Journal of Virology

jvi.asm.org 12

Efficacy of Chimeric HA H9/H5N2 DIVA Vaccine

Journal of Virology

influenza viruses through the substitution of the globular head domain of a different subtype. This approach results in a chimeric virus expressing the foreign globular head domain and part of the original HA1 and HA2 regions (40, 41). However, these viruses might not be fit for the DIVA strategy in vaccinated hosts since they still express large portions of the parental HA1 region, which could hamper clear differentiation of the HA1 and HA2 antibody responses by serology assays (42). Furthermore, although the globular head domain is considered a major antigenic component of the HA protein, immunogenicity outside the globular head domain of the HA1 protein could not be ruled out as being crucial for protection against challenge virus. In addition, Wang et al. demonstrated the generation of recombinant viruses that express HA possessing the HA1 portion of A/California/07/2009 (Cal/09) and the HA2 region from A/South Dakota/ 6/07 (SD/07). It should be noted that both Cal/09 and SD/07 are H1N1 viruses and thus have a high degree of similarity in their HA2 regions (43). However, to our knowledge, this is the first study to show the generation of a chimeric HA influenza A virus between two different subtypes (H9N2 and H5N8) that contains a very low degree of similarity in their HA1 and HA2 regions in order to induce the maximum protective antibody response. Our in vitro study revealed that the cHA H9/H5N2 recombinant virus possesses growth properties similar to those of the wild-type H9N2 influenza virus, which can grow with the addition of exogenous trypsin supplementation. In contrast, the HPAI H5N8 virus grew even without the addition of exogenous trypsin (Table 1). In animal studies, the cHA H9/H5N2 virus infected chickens and mice, but the virus was not detected outside the respiratory tract (mice) or intestinal organs (chickens), which is characteristic of LPAI viruses. Therefore, these results suggest that the cHA H9/H5N2 virus has characteristics of LPAI viruses and will be safe for vaccine production. It is noteworthy that the inactivated cHA H9/H5N2 vaccine induced an H9-specific HI titer (HI titer of 160 to 1,280) comparable to those induced by the H9N2 vaccine (Fig. 2A and 7A), while there was no detectable cross-reactive HI titer against H5 viruses, which express the HA2 region of the cHA H9/H5N2 virus. However, in the serum neutralization assay, only the cHA H9/H5N2 vaccine induced broad serum neutralizing antibodies against both homologous and heterologous H9N2 (HI titer of 160 to 640) and HA2 stalk H5 (SN titer of 80) viruses in both chicken and mouse models (Fig. 2B and 7B). In addition, an ELISA confirmed that only the cHA H9/H5N2 vaccine induced IgG responses against both H9N2 and H5N8 viruses (Fig. 3B and 8B). To our knowledge, this study is the first to report that a chimeric virus containing the HA protein with the whole HA1 and HA2 regions from two different subtypes can induce both HA1- and HA2-specific antibodies. Thus, the cHA H9/H5N2 vaccine can be implemented in a DIVA strategy for poultry, and a simple HI assay can be used to differentiate vaccinated animals from those infected with an H5 virus. Although a dose of 3.5 ␮g HA is relatively high for mouse vaccination, we wanted to use the same dose of the HA antigen in both chickens and mice to evaluate cross-protective efficacy. In animal challenge studies, both H9N2- and cHA H9/H5N2vaccinated animals showed significantly attenuated virus shedding after H9N2 and HPAI H5N8 virus challenges. Although each group of mice was vaccinated with 3.5 ␮g HA/dose of the H9N2 or cHA H9/H5N2 vaccine, they showed moderate body weight loss until 5 dpi (⬎10%) following homologous mouse-adapted virus infections (Fig. 9A and B). However, these values were lower than those seen with infection by the heterologous ma81/07 virus, which caused ⬎15% body weight loss (Fig. 9C). Many previous studies reported that even though a vaccine is able to provide protection from mortality, immunized animals generally shed virus for up to 3 days after virus challenge (44–46). In agreement with data from those studies, the heterologous H5N2 and H5N8 viruses were detected until 5 dpi, but the virus detection periods and viral titers were significantly reduced (at least 10 times) compared to those of the H9N2 vaccine group (P ⬍ 0.05) (Table 2). It is noteworthy that in the cHA H9/H5N2 vaccine group, the virus was not detected in brain tissue during the testing period. These results clearly March 2017 Volume 91 Issue 6 e01693-16

jvi.asm.org 13

Kim et al.

Journal of Virology

demonstrate that the cHA H9/H5N2 vaccine can inhibit the replication and spread of the HPAI H5N8 virus in internal organs of infected hosts. The inability of LPAI vaccines to effectively protect against infection with antigenically drifted viruses or newly emerging viruses underlines the need for the development of cross-reactive influenza vaccines that induce immunity against a variety of virus subtypes. In this study, we demonstrate that immunization of mice with the cHA H9/H5N2 vaccine induced broad cross-reactive antibody responses that protected immunized hosts against homologous and heterologous lethal challenges with H5N8 and maH5N2 viruses (Fig. 7 to 9). In contrast, all H9N2-vaccinated mice succumbed to death by 13 dpi (Fig. 9B and C). It should be noted that the chimeric H9/H5N2 vaccine groups also showed about 15% weight loss for up to 7 days following maH5N2 virus challenge (Fig. 9C), although most mice recovered after 7 dpi. These results might be explained by the small amino acid variations between the HA2 regions of viruses of the same H5 subtype. Although the HA2 regions have been considered to be conserved within the same groups of viruses (47), our amino acid comparison results show that there is 96% (8/220) amino acid homology between the HA2 regions of H5N8 and maH5N2 viruses, even though they were clustered in the same HA group, group I (Fig. 1C). For comparison, ⬍65% identity between H9 and H5 viruses was observed. Therefore, the different protection efficacies between strains might be explained by sequence variations in HA2 regions, which were previously considered to be conserved within the same subtype. Taken together, our results suggest that the cHA H9/H5N2 vaccination strategy provides robust protection against homologous, heterologous, and heterosubtypic viruses of both subtypes. Furthermore, each HA1- and HA2 stalk-specific antibody response was sufficient to inhibit viral replication and protect chimeric virus-immunized mice from lethal challenge with mouse-adapted H9N2, H5N2, or wild-type HPAI H5N1 virus. This finding suggests that the HA2 stalk region of the H5 HA protein contains functional epitopes that are attractive targets for broad-spectrum H5 influenza viruses in both chicken and mouse models. MATERIALS AND METHODS Chimeric virus generation using reverse genetics. To construct the chimeric H9/5 HA (hemagglutinin) gene, the HA1 sequences of the H9N2 LPAI virus strain A/CK/Korea/04163/04 and the HA2 sequences of the H5N8 HPAI virus strain A/MD/Korea/W452/14 (W452/14) were selected. The chimeric HA gene (cHA H9/H5) was generated by overlapping PCR with specific primer sets (H9/H5 HA2-Forword [CTATTTGGAGCTATAGCA] and H9/H5 HA1-Reverse [AGCTATGGCACCAAATAG]) and then cloned into vpHW2000 as previously described (36). The chimeric influenza H9/H5N2 (cHA H9/H5N2) viruses were produced by reverse genetics as previously described (36). The rescued cHA H9/H5N2 virus was the 1:7 recombinant virus that contained the polybasic cleavage site-deleted cHA H9/H5 HA gene and seven internal genes of A/CK/04163/2004 (H9N2), and its genetic composition was confirmed by full-length sequencing. Vaccine preparation. The cHA H9/H5N2 and A/CK/Korea/04163/2004 (04163/04) (H9N2) viruses were propagated in 10-day-old SPF embryonated chicken eggs. Clarified allantoic fluids were loaded into ultracentrifuge tubes and underlaid with a 20% sucrose cushion. Ultracentrifugation was performed at 13,000 ⫻ g for 3 h at 4°C. Supernatants were discarded, and virus pellets were resuspended overnight at 4°C in 1⫻ phosphate-buffered saline. Purified viruses were inactivated by treatment with 0.025% formalin at 4°C for at least 1 week, which resulted in the complete loss of infectivity of the virus, as confirmed by the absence of detectable infectious virus following inoculation of the vaccines into eggs. The samples were then aliquoted and stored at ⫺80°C until further analysis. The presence of viruses in the pellet was confirmed by a hemagglutination assay. Virus titrations. Virus titers in virus stocks were determined by performing endpoint titrations in 10-day-old embryonated chicken eggs, monolayers of MDCK cells, or both. Eggs or cells were inoculated with 10-fold serial dilutions of each sample in a 1⫻ phosphate-buffered saline solution or fetal bovine serum (FBS)-free medium containing antibiotics. After 48 h of incubation at 35°C, the presence of viruses was detected by a standard HA assay using 0.5% chicken erythrocytes. Mean virus titers were expressed as log10 EID50 or as log10 TCID50 per unit sample (grams or milliliters) tested. The limit of virus detection was set at 0.7 log10 EID50 or 2.5 log10 TCID50 per unit sample tested. Immunization and experimental infection of chickens. Six-week-old female SPF White Leghorn chickens (CAVac Lab Co., Ltd., Daejeon, South Korea) were used in this study. One vaccine dose (0.5 ml) contained 3.5 ␮g HA/dose of influenza A 04163/04 (H9N2) virus and cHA H9/H5N2 with an alum adjuvant. Each vaccine was administered to all subjects i.m. 14 days apart. Each group of birds (n ⫽ 21) was then experimentally challenged with 04163/04 (H9N2) (109.5 EID50 [maximum titer]) or W452/14 (H5N8) (104.5 EID50 [100 times the CLD50]) virus in a volume of 0.5 ml by the intranasal route. Three birds March 2017 Volume 91 Issue 6 e01693-16

jvi.asm.org 14

Efficacy of Chimeric HA H9/H5N2 DIVA Vaccine

Journal of Virology

in each group were euthanized on days 1, 3, and 5 postchallenge, and oropharyngeal, cloacal swabs, and organ tissue samples (trachea, lungs, brain, kidney, spleen, heart, liver, and intestine) were collected for virus titrations. Immunization and experimental infection of mice. Female inbred BALB/c mice(6 weeks old) were purchased from Samtako (Seoul, South Korea) and maintained under conventional conditions. Mice were immunized with 3.5 ␮g HA/dose (0.1 ml) of the inactivated H9N2 or cHA H9/H5N2 vaccine through two i.m. injections 14 days apart. A PBS-alum-immunized group (mock) was included as a negative control. The same antigenic dose was used in mice at a 0.1-ml volume (50 ␮l in each thigh) with an alum adjuvant. The mice were grouped into three challenge groups: ma163/04 (A/CK/Korea/04163/04) (mouse-adapted H9N2 strain [44]), W452/14 (wild-type H5N8 [37]), and ma81/07 (A/AB/Korea/MA81/07) (mouse-adapted H5N2 strain [38]). Groups of mice (n ⫽ 35) were intranasally inoculated with 10 times the MLD50 of each representative virus. Five mice from each group were euthanized at 1, 3, 5, 7, and 9 dpi to collect the lungs and various organs (turbinate, heart, brain, kidney, and liver) for virus titration. Body weights and survival (n ⫽ 10) were monitored daily for 14 days. Hemagglutination inhibition assay. HI assays were adapted as described previously (48). Immunization sera were treated with receptor-destroying enzyme (RDE) to inactivate nonspecific inhibitors with a final serum dilution of 1:10. RDE-treated sera were serially diluted in PBS at 2-fold increments in a 25-␮l volume in a 96-well plate, and equal volumes of virus (8 hemagglutination units in 50 ␮l) were added to each well. The plate was incubated at room temperature (RT) for 30 to 60 min. Negative hemagglutination results appeared as dots in the centers of the wells. The titer was calculated as the highest dilution factor that produced a positive reading. The limit of detection for the HI assays was set to ⬍10 HI units. Microneutralization assay. Serum neutralizing antibody titers were measured in MDCK cells as described previously by Rowe et al. (49), with slight modifications. Briefly, a monolayer of MDCK cells was cultured in minimal essential medium (MEM) supplemented with 10% FBS. Heat-inactivated sera were serially diluted 2-fold and mixed with 100 times the TCID50 of the cHA H9/H5N2, 04163/04 (H9N2), A/CK/Korea/01310/01 (H9N2) (01310/01), A/CK/Korea/HC09/09 (H9N2) (HC09/09), W452/14 (H5N8), and A/AB/Korea/ma81/07 (H5N2) (ma81/07) viruses. Neutralizing antibody titers were expressed as the reciprocal of the highest dilution of serum that was able to neutralize 100 times the TCID50 of the virus in cells. ELISA. ELISA plates (Thermo Scientific) were coated overnight at 4°C with each purified virus (1 mg/ml) or the sM2-HA fusion protein (kindly provided by Chul-Joong Kim, Chungnam National University, Daejeon, South Korea) diluted in carbonate-bicarbonate coating buffer (pH 9.4; Sigma). Plates were blocked for 1 h at RT with PBS containing 0.1% Tween 20 (PBST) and 5% nonfat dry milk powder. Mouse and chicken sera were prediluted 1:100, serially diluted in 1:2 steps in PBST containing 2% nonfat dry milk powder, and incubated on the plates for 1 h at RT. After extensive washing with PBST (3 times with 100 ␮l/well), the plates were incubated for 1 h at RT with anti-mouse horseradish peroxidase (HRP)conjugated IgG (Abcam) and rabbit anti-chicken HRP-conjugated IgY (IgG) (Thermo Fisher) diluted in PBST containing 2% nonfat dry milk powder. After three more washes with PBST, the plates were overlaid with the o-phenylenediamine dihydrochloride (SigmaFast OPD; Sigma) substrate. Reactions were stopped by using 3 M HCl. Optical density (OD) was read at 490 nm using an iMark microplate absorbance reader (Bio-Rad). Ethics statement. Virus preparation, titration, all animal studies, and serological testing for chimeric H9/H5N2 and H5N1 viruses were performed in an enhanced biosafety level 3 (BSL-3⫹) containment facility at Chungbuk National University, approved by the Korean Center for Disease Control and Prevention (K-CDC) (permit no. KCDC-14-3-07). All animal experiments were conducted in strict accordance with and adherence to relevant policies regarding animal handling as mandated under the guidelines for animal use and care of the K-CDC and were approved by the Medical Research Institute (approval no. CBNUA-767-14-01), a member of LARC (Laboratory Animal Research Center of Chungbuk National University).

ACKNOWLEDGMENTS This research was supported by the Korea Healthcare Technology R&D Project funded by the Ministry of Health (grant no. A103001). We thank Chul-Joong Kim and Jong Soo Lee for kindly providing the sM2-HA fusion proteins used for the ELISAs.

REFERENCES 1. Alexander DJ. 2007. An overview of the epidemiology of avian influenza. Vaccine 25:5637–5644. https://doi.org/10.1016/j.vaccine.2006.10.051. 2. Rott R. 1992. The pathogenic determinant of influenza virus. Vet Microbiol 33:303–310. https://doi.org/10.1016/0378-1135(92)90058-2. 3. Wu R, Sui Z, Zhang H, Chen Q, Liang W, Yang K, Xiong Z, Liu Z, Chen Z, Xu D. 2008. Characterization of a pathogenic H9N2 influenza A virus isolated from central China in 2007. Arch Virol 153:1549 –1555. https:// doi.org/10.1007/s00705-008-0139-1. 4. Monne I, Joannis TM, Fusaro A, De Benedictis P, Lombin LH, Ularamu H, Egbuji A, Solomon P, Obi TU, Cattoli G. 2008. Reassortant avian influenza March 2017 Volume 91 Issue 6 e01693-16

virus (H5N1) in poultry, Nigeria, 2007. Emerg Infect Dis 14:637– 640. https://doi.org/10.3201/eid1404.071178. 5. Lee Y-J, Shin J-Y, Song M-S, Lee Y-M, Choi J-G, Lee E-K, Jeong O-M, Sung H-W, Jae-Hong K, Kwon Y-K. 2007. Continuing evolution of H9 influenza viruses in Korean poultry. Virology 359:313–323. https://doi.org/10.1016/ j.virol.2006.09.025. 6. Toroghi R, Momayez R. 2006. Biological and molecular characterization of avian influenza virus (H9N2) isolates from Iran. Acta Virol 50:163–168. 7. Guo Y, Krauss S, Senne D, Mo I, Lo K, Xiong X, Norwood M, Shortridge K, Webster R, Guan Y. 2000. Characterization of the pathogenicity of memjvi.asm.org 15

Kim et al.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22. 23.

24.

25.

26.

27.

bers of the newly established H9N2 influenza virus lineages in Asia. Virology 267:279 –288. https://doi.org/10.1006/viro.1999.0115. Choi Y, Ozaki H, Webby R, Webster R, Peiris J, Poon L, Butt C, Leung Y, Guan Y. 2004. Continuing evolution of H9N2 influenza viruses in Southeastern China. J Virol 78:8609–8614. https://doi.org/10.1128/JVI.78.16.8609-8614.2004. Hatta M, Kawaoka Y. 2002. The continued pandemic threat posed by avian influenza viruses in Hong Kong. Trends Microbiol 10:340 –344. https://doi.org/10.1016/S0966-842X(02)02388-0. Eagles D, Siregar E, Dung D, Weaver J, Wong F, Daniels P. 2009. H5N1 highly pathogenic avian influenza in Southeast Asia. Rev Sci Tech 28: 341–348. https://doi.org/10.20506/rst.28.1.1864. Saito T, Tanikawa T, Uchida Y, Takemae N, Kanehira K, Tsunekuni R. 2015. Intracontinental and intercontinental dissemination of Asian H5 highly pathogenic avian influenza virus (clade 2.3. 4.4) in the winter of 20142015. Rev Med Virol 25:388 – 405. https://doi.org/10.1002/rmv.1857. Pasick J, Berhane Y, Joseph T, Bowes V, Hisanaga T, Handel K, Alexandersen S. 2015. Reassortant highly pathogenic influenza A H5N2 virus containing gene segments related to Eurasian H5N8 in British Columbia, Canada, 2014. Sci Rep 5:9484. https://doi.org/10.1038/srep09484. Ip HS, Torchetti MK, Crespo R, Kohrs P, DeBruyn P, Mansfield KG, Baszler T, Badcoe L, Bodenstein B, Shearn-Bochsler V. 2015. Novel Eurasian highly pathogenic avian influenza A H5 viruses in wild birds, Washington, USA, 2014. Emerg Infect Dis 21:886 – 890. https://doi.org/10.3201/ eid2105.142020. Zhao K, Gu M, Zhong L, Duan Z, Zhang Y, Zhu Y, Zhao G, Zhao M, Chen Z, Hu S. 2013. Characterization of three H5N5 and one H5N8 highly pathogenic avian influenza viruses in China. Vet Microbiol 163:351–357. https://doi.org/10.1016/j.vetmic.2012.12.025. Tang Y, Wu P, Peng D, Wang X, Wan H, Zhang P, Long J, Zhang W, Li Y, Wang W. 2009. Characterization of duck H5N1 influenza viruses with differing pathogenicity in mallard (Anas platyrhynchos) ducks. Avian Pathol 38:457– 467. https://doi.org/10.1080/03079450903349147. Li K, Guan Y, Wang J, Smith G, Xu K, Duan L, Rahardjo A, Puthavathana P, Buranathai C, Nguyen T. 2004. Genesis of a highly pathogenic and potentially pandemic H5N1 influenza virus in eastern Asia. Nature 430: 209 –213. https://doi.org/10.1038/nature02746. WHO. 2015. Cumulative number of confirmed human cases for avian influenza A (H5N1) reported to WHO, 2003-2015. WHO, Geneva, Switzerland. http://www.who.int/influenza/human_animal_interface/ EN_GIP_20151214cumulativeNumberH5N1cases.pdf?ua⫽1. Butt K, Smith GJ, Chen H, Zhang L, Leung YC, Xu K, Lim W, Webster RG, Yuen K, Peiris JM. 2005. Human infection with an avian H9N2 influenza A virus in Hong Kong in 2003. J Clin Microbiol 43:5760 –5767. https:// doi.org/10.1128/JCM.43.11.5760-5767.2005. Yu H, Zhou Y-J, Li G-X, Ma J-H, Yan L-P, Wang B, Yang F-R, Huang M, Tong G-Z. 2011. Genetic diversity of H9N2 influenza viruses from pigs in China: a potential threat to human health? Vet Microbiol 149:254 –261. https://doi.org/10.1016/j.vetmic.2010.11.008. Chen H, Yuan H, Gao R, Zhang J, Wang D, Xiong Y, Fan G, Yang F, Li X, Zhou J. 2014. Clinical and epidemiological characteristics of a fatal case of avian influenza A H10N8 virus infection: a descriptive study. Lancet 383:714 –721. https://doi.org/10.1016/S0140-6736(14)60111-2. Peiris JM, De Jong MD, Guan Y. 2007. Avian influenza virus (H5N1): a threat to human health. Clin Microbiol Rev 20:243–267. https://doi.org/ 10.1128/CMR.00037-06. James-Berry CM. 2013. Vaccine control of avian influenza H5N1 in poultry: need for a positive marker. J Vaccines Vaccin 4:168. Park KJ, Kwon H, Song M-S, Pascua PNQ, Baek YH, Lee JH, Jang H-L, Lim J-Y, Mo I-P, Moon H-J. 2011. Rapid evolution of low-pathogenic H9N2 avian influenza viruses following poultry vaccination programmes. J Gen Virol 92:36 –50. https://doi.org/10.1099/vir.0.024992-0. Vergne T, Grosbois V, Jobre Y, Saad A, El Nabi AA, Galal S, Kalifa M, El Kader SA, Dauphin G, Roger F. 2012. Avian influenza vaccination of poultry and passive case reporting, Egypt. Emerg Infect Dis 18: 2076 –2078. https://doi.org/10.3201/eid1812.120616. Pu J, Wang S, Yin Y, Zhang G, Carter RA, Wang J, Xu G, Sun H, Wang M, Wen C. 2015. Evolution of the H9N2 influenza genotype that facilitated the genesis of the novel H7N9 virus. Proc Natl Acad Sci U S A 112: 548 –553. https://doi.org/10.1073/pnas.1422456112. Sun Y, Pu J, Jiang Z, Guan T, Xia Y, Xu Q, Liu L, Ma B, Tian F, Brown E. 2010. Genotypic evolution and antigenic drift of H9N2 influenza viruses in China from 1994 to 2008. Vet Microbiol 146:215–225. https://doi.org/ 10.1016/j.vetmic.2010.05.010. Zhang P, Tang Y, Liu X, Peng D, Liu W, Liu H, Lu S, Liu X. 2008.

March 2017 Volume 91 Issue 6 e01693-16

Journal of Virology

28.

29.

30.

31. 32.

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

Characterization of H9N2 influenza viruses isolated from vaccinated flocks in an integrated broiler chicken operation in eastern China during a 5 year period (1998-2002). J Gen Virol 89:3102–3112. https://doi.org/ 10.1099/vir.0.2008/005652-0. Li C, Yu K, Tian G, Yu D, Liu L, Jing B, Ping J, Chen H. 2005. Evolution of H9N2 influenza viruses from domestic poultry in Mainland China. Virology 340:70 – 83. https://doi.org/10.1016/j.virol.2005.06.025. Guan Y, Smith GJ. 2013. The emergence and diversification of panzootic H5N1 influenza viruses. Virus Res 178:35– 43. https://doi.org/10.1016/ j.virusres.2013.05.012. FAO/OIE/WHO. 2004. FAO/OIE/WHO technical consultation on the control of avian influenza. WHO, Geneva, Switzerland. http://www.fao.org/ avianflu/documents/expertconsult.pdf. Hashem AM. 2015. Prospects of HA-based universal influenza vaccine. Biomed Res Int 2015:414637. https://doi.org/10.1155/2015/414637. Chiu F-F, Venkatesan N, Wu C-R, Chou A-H, Chen H-W, Lian S-P, Liu S-J, Huang C-C, Lian W-C, Chong P. 2009. Immunological study of HA1 domain of hemagglutinin of influenza H5N1 virus. Biochem Biophys Res Commun 383:27–31. https://doi.org/10.1016/j.bbrc.2009.03.106. Sui J, Hwang WC, Perez S, Wei G, Aird D, Chen L, Santelli E, Stec B, Cadwell G, Ali M. 2009. Structural and functional bases for broadspectrum neutralization of avian and human influenza A viruses. Nat Struct Mol Biol 16:265–273. https://doi.org/10.1038/nsmb.1566. Tan GS, Krammer F, Eggink D, Kongchanagul A, Moran TM, Palese P. 2012. A pan-H1 anti-hemagglutinin monoclonal antibody with potent broad-spectrum efficacy in vivo. J Virol 86:6179 – 6188. https://doi.org/ 10.1128/JVI.00469-12. Pica N, Palese P. 2013. Toward a universal influenza virus vaccine: prospects and challenges. Annu Rev Med 64:189 –202. https://doi.org/ 10.1146/annurev-med-120611-145115. Song M-S, Baek YH, Pascua PNQ, Kwon H, Park S-J, Kim E-H, Lim G-J, Choi Y-K. 2013. Establishment of Vero cell RNA polymerase I-driven reverse genetics for influenza A virus and its application for pandemic (H1N1) 2009 influenza virus vaccine production. J Gen Virol 94:1230 –1235. https://doi.org/10.1099/vir.0.051284-0. Kim Y-I, Pascua PNQ, Kwon H-I, Lim G-J, Kim E-H, Yoon S-W, Park S-J, Kim SM, Choi E-J, Si Y-J. 2014. Pathobiological features of a novel, highly pathogenic avian influenza A (H5N8) virus. Emerg Microbes Infect 3:e75. https://doi.org/10.1038/emi.2014.75. Song M-S, Pascua PNQ, Lee JH, Baek YH, Lee O-J, Kim C-J, Kim H, Webby RJ, Webster RG, Choi YK. 2009. The polymerase acidic protein gene of influenza A virus contributes to pathogenicity in a mouse model. J Virol 83:12325–12335. https://doi.org/10.1128/JVI.01373-09. Claas EC, Osterhaus AD, Van Beek R, De Jong JC, Rimmelzwaan GF, Senne DA, Krauss S, Shortridge KF, Webster RG. 1998. Human influenza A H5N1 virus related to a highly pathogenic avian influenza virus. Lancet 351:472– 477. https://doi.org/10.1016/S0140-6736(97)11212-0. Krammer F, Pica N, Hai R, Margine I, Palese P. 2013. Chimeric hemagglutinin influenza virus vaccine constructs elicit broadly protective stalk-specific antibodies. J Virol 87:6542–6550. https://doi.org/10.1128/JVI.00641-13. Hai R, Krammer F, Tan GS, Pica N, Eggink D, Maamary J, Margine I, Albrecht RA, Palese P. 2012. Influenza viruses expressing chimeric hemagglutinins: globular head and stalk domains derived from different subtypes. J Virol 86:5774 –5781. https://doi.org/10.1128/JVI.00137-12. Pica N, Hai R, Krammer F, Wang TT, Maamary J, Eggink D, Tan GS, Krause JC, Moran T, Stein CR. 2012. Hemagglutinin stalk antibodies elicited by the 2009 pandemic influenza virus as a mechanism for the extinction of seasonal H1N1 viruses. Proc Natl Acad Sci U S A 109:2573–2578. https:// doi.org/10.1073/pnas.1200039109. Wang W, Suguitan AL, Zengel J, Chen Z, Jin H. 2012. Generation of recombinant pandemic H1N1 influenza virus with the HA cleavable by bromelain and identification of the residues influencing HA bromelain cleavage. Vaccine 30:872–878. https://doi.org/10.1016/j.vaccine.2011.11.101. Park KJ, Song M-S, Kim E-H, Kwon H, Baek YH, Choi E, Park S-J, Kim SM, Kim Y, Choi W-S. 2015. Molecular characterization of mammalian-adapted Korean-type avian H9N2 virus and evaluation of its virulence in mice. J Microbiol 53:570–577. https://doi.org/10.1007/s12275-015-5329-4. Noh J-Y, Park J-K, Lee D-H, Yuk S-S, Kwon J-H, Lee S-W, Lee J-B, Park S-Y, Choi I-S, Song C-S. 2016. Chimeric bivalent virus-like particle vaccine for H5N1 HPAI and ND confers protection against a lethal challenge in chickens and allows a strategy of differentiating infected from vaccinated animals (DIVA). PLoS One 11:e0162946. https://doi.org/10.1371/ journal.pone.0162946. Lu J, Wu P, Zhang X, Feng L, Dong B, Chu X, Liu X, Peng D, Liu Y, Ma H. jvi.asm.org 16

Efficacy of Chimeric HA H9/H5N2 DIVA Vaccine

2016. Immunopotentiators improve the efficacy of oil-emulsioninactivated avian influenza vaccine in chickens, ducks and geese. PLoS One 11:e0156573. https://doi.org/10.1371/journal.pone.0156573. 47. Steel J, Lowen AC, Wang TT, Yondola M, Gao Q, Haye K, García-Sastre A, Palese P. 2010. Influenza virus vaccine based on the conserved hemagglutinin stalk domain. mBio 1:e00018-10. https://doi.org/10.1128/ mBio.00018-10.

March 2017 Volume 91 Issue 6 e01693-16

Journal of Virology

48. Hirst GK. 1942. The quantitative determination of influenza virus and antibodies by means of red cell agglutination. J Exp Med 75:49 – 64. https://doi.org/10.1084/jem.75.1.49. 49. Rowe T, Abernathy RA, Hu-Primmer J, Thompson WW, Lu X, Lim W, Fukuda K, Cox NJ, Katz JM. 1999. Detection of antibody to avian influenza A (H5N1) virus in human serum by using a combination of serologic assays. J Clin Microbiol 37:937–943.

jvi.asm.org 17