Virus Research 200 (2015) 9–18

Contents lists available at ScienceDirect

Virus Research journal homepage: www.elsevier.com/locate/virusres

H5N1 influenza virus-like particle vaccine protects mice from heterologous virus challenge better than whole inactivated virus Zhiguang Ren a,b , Xianliang Ji b,c , Lingnan Meng b,d , Yurong Wei b,e , Tiecheng Wang b , Na Feng b , Xuexing Zheng b , Hualei Wang b , Nan Li b , Xiaolong Gao b , Hongli Jin f , Yongkun Zhao b , Songtao Yang b , Chuan Qin a , Yuwei Gao b,g,h,∗ , Xianzhu Xia a,b,g,h,∗ a

Institute of Laboratory Animal Sciences, Chinese Academy of Medical Sciences & Peking Union Medical College, Beijing, China Institute of Military Veterinary, Academy of Military Medical Sciences, Changchun, Jilin Province, China c College of veterinary Medicine, Inner Mongolia Agricultural University, Inner Mongolia Autonomous Region, Huhhot, China d College of Animal Science and Technology, Jilin Agricultural University, Changchun, Jilin Province, China e College of Animal Science and Technology, Shihezi University, Shihezi, Xinjiang Province, China f Changchun SR Biological Technology Co., Ltd, Changchun, Jilin Province, China g Jiangsu Co-innovation Center for Prevention and Control of Important Animal Infectious Diseases and Zoonoses, Yangzhou, Jiangsu Province, China h Changchun Veterinary Research Institute, Chinese Academy of Agricultural Sciences, Changchun, Jilin Province, China b

a r t i c l e

i n f o

Article history: Received 2 November 2014 Received in revised form 6 January 2015 Accepted 10 January 2015 Available online 17 January 2015 Keywords: H5N1 influenza Protection Virus-like particles Vaccine Cross-protective

a b s t r a c t The highly pathogenic avian influenza (HPAI) H5N1 virus has become highly enzootic since 2003 and has dynamically evolved to undergo substantial evolution. Clades 2.3.2.1 and 2.3.4 have become the most dominant lineage in recent years, and H5N8 avian influenza outbreaks have been reported Asia. The current approach to generate influenza virus vaccines uses embryonated chicken eggs for large-scale production, although such vaccines have been poorly immunogenic to heterologous virus challenge. In the current study, virus-like particles (VLP) based on A/meerkat/Shanghai/SH-1/2012 (clade 2.3.2.1) and comprising hemagglutinin (HA), neuraminidase (NA), and matrix (M1) were produced using a baculovirus expression system to develop effective protection for different H5 HPAI clade challenges. Mice immunized with VLP demonstrated stronger humoral and cellular immune responses than mice immunized with whole influenza virus (WIV), with 20-fold higher IgG antibody titers against A/meerkat/Shanghai/SH-1/2012 after boost. Notably, the WIV vaccine group showed partial protection (80% survival) to homologous challenge, little protection (40% survival) to heterologous challenge, and 20% survival to H5N8 challenge, whereas all mice in the VLP + CFA group survived. These results provide insight for the development of effective prophylactic vaccines based on VLPs with cross-clade protection for the control of current H5 HPAI outbreaks in humans. © 2015 Elsevier B.V. All rights reserved.

1. Introduction Influenza A virus causes acute viral respiratory disease and recurrent outbreaks, which significantly affect human health and the global economy (Quan et al., 2010). The first human outbreak of highly pathogenic avian influenza (HPAI) H5N1 was identified in 1997, and more than 649 confirmed human infections and 385 fatalities have occurred as of January 9, 2014 (http://www.who.int/csr/don/2014 01 09 h5n1/en/). HPAI H5N1 has become highly enzootic since 2003 and has dynamically

∗ Corresponding authors at: The Military Veterinary Institute, Academy of Military Medical Science of PLA 666 Liuyingxi st. Changchun 130122, PR China, Tel./fax: +86431 86985516. E-mail addresses: [email protected] (Y. Gao), [email protected] (X. Xia). http://dx.doi.org/10.1016/j.virusres.2015.01.007 0168-1702/© 2015 Elsevier B.V. All rights reserved.

evolved to undergo substantial evolution. HPAI H5N1 was found to co-circulate, particularly in China, Vietnam, Indonesia, Egypt, Cambodia, and Bangladesh, and clades 2.3.2.1 and 2.3.4 have become the predominant lineage (Le and Nguyen, 2014; Watanabe et al., 2011) (http://www.who.int/entity/influenza/vaccines/virus/201402 h5h7h9h10 vaccinevirusupdate.pdf). However, it is difficult to predict the specific isolate from different H5N1 clades with the potential to create a pandemic (Prabakaran et al., 2013). H5N8 avian influenza outbreaks have been reported in the Republic of Korea (Lee et al., 2014) and have been isolated from wild duck in China (Fan et al., 2014). Therefore, the development of a vaccine that induces cross-protection against different antigenic H5 subtypes is crucial. The current approach to generate influenza virus vaccines uses embryonated chicken eggs for large-scale production (Robertson and Engelhardt, 2010). However, manufacturing problems,

10

Z. Ren et al. / Virus Research 200 (2015) 9–18

including limited production capacity due to insufficient egg supply and pandemic allergic reactions to egg proteins, in recent years have shown that the current methods of production are too fragile and time-consuming to ensure an adequate supply of influenza virus vaccine (Erlewyn-Lajeunesse et al., 2009). In addition, shortfalls in the response of the vaccine supply for the influenza season have been noted, and it takes months to identify new potential strains (Quan et al., 2007; Rezaei et al., 2013). More importantly, the HPAI H5N1 viruses that were responsible for recent epizootic outbreaks in Asia are lethal to chicken eggs (Choi et al., 2013a; Qiao et al., 2003). However, recombinant noninfectious virus-like particles (VLPs) produced in a baculovirus system can avoid the handling of live influenza viruses during the vaccine manufacturing process; thus, this approach represents a promising and novel technology for the creation of low-cost, safe and high-yielding commercial vaccines for influenza virus. VLPs are structurally native and immunologically relevant viral antigens because the hemagglutinin (HA) antigen is presented to the host in a native particulate form without chemical inactivation compared to the WIV vaccine (Bright et al., 2008; Bright et al., 2007; Jin et al., 2008). Several different constructs of VLPs that contain influenza HA or HA- neuraminidase (NA) and the influenza matrix protein M1 have been shown to induce high titers of virus-specific antibodies in vaccinated mice. These VLPs provide protection against heterologous virus challenge (Choi et al., 2013a; Prabakaran et al., 2013). However, the protective efficacy of VLPs has not been evaluated for VLPs against different H5N1 virus clades and H5N8 virus or compared with that of whole inactivated virus (WIV). This study generated H5N1 influenza VLPs with NA-HA-M1 using a baculovirus expression system, and then we investigated the immunogenicity and protective efficacy of H5N1 WIV, VLPs and VLPs plus complete Freund’s adjuvant (CFA) against different H5N1 clades in mice. 2. Materials and methods 2.1. Viruses and cells HPAI H5N1 virus A/meerkat/Shanghai/SH-1/2012 (SH-1; clade 2.3.2.1), A/duck/Jilin/JL-SIV/2013 (JL-SIV; clade 2.3.4), and H5N8 virus A/mallard duck/Shanghai/SH-9/2013 (SH-9) were originally isolated and stored in the Changchun Veterinary Research Institute. All the viruses were grown in 9-day-old embryonated hen’s eggs for 24–48 h at 37 ◦ C. Allantoic fluids were harvested from infected eggs, stored overnight at 4 ◦ C and centrifuged to remove cell debris. The virus was purified through a discontinuous sucrose gradient (20–30–60% layers) and ultracentrifuged at 28,000 rpm for 1 h. Purified virus was mixed with formalin at a final concentration of 1:4000 (vol/vol) as described previously to inactivate the virus. Hemagglutination assays were performed to determine the virus titer. Spodoptera frugiperda Sf9 insect cells (Invitrogen, USA) were maintained in TMN insect medium (Appilchem, Germany) at 27 ◦ C. Madin-Darby canine kidney (MDCK) cells were maintained and grown in Dulbecco’s modified Eagle’s medium (DMEM) plus 10% fetal bovine serum. All experiments using highly pathogenic virus were conducted in a biosafety level 3 (BSL3) facility in compliance with WHO recommendations and approved by the Changchun Veterinary Research Institute.

Fig. 1. Construction of pFastbac 1 plasmid. The NA-HA-M1 gene was synthesized with the Not I and Sph I restriction sites and ligated with pFastBac 1 transfer vector as indicated. The NA-HA-M1 gene was paired and cloned in bacmid recombination.

synthesized by Shanghai Generay Biotech Co., Ltd. (Fig. 1). Genes were digested with Not I and Sph I and cloned into a pFastBac 1 transfer vector (Invitrogen, USA). The final plasmid pFastBac-NAHA-M1 was transformed into E. coli DH10Bac competent cells that contained the AcMNPV baculovirus genome (Invitrogen, USA), and recombinant bacmids were produced using site-specific homologous recombination, as previously described (Pushko et al., 2005). Recombinant bacmid DNA was transfected into 1 × 106 Sf9 cells seeded in 6-well plates using Cellfectin reagent (Invitrogen, USA). The pFastBac 1 transfer vector was transformed into E.coli DH10 cells and transfected into Sf9 cells to obtain the baculovirus (BV). The resulting recombinant baculovirus (rBV) or BV was collected from the culture medium 72 h post-infection according to the manufacturer’s instructions and stored at 4 ◦ C. Sf9 cells were infected with the rBV or BV at a multiplicity of infection (MOI) of 3 to produce VLPs. Titers of recombinant baculovirus stocks were determined by the BacPAK Baculovirus Rapid Titer Kit (Clontech Laboratories, Inc., USA) according to the manufacturer’s instructions. Culture supernatants that contained VLPs or BV were harvested 3 days post-infection and clarified by lowspeed centrifugation at 2000 rpm for 20 min at 4 ◦ C followed by ultracentrifugation at 30,000 rpm for 60 min to pellet. Sedimented particles were suspended in phosphate-buffered saline (PBS) at 4 ◦ C overnight and further purified through a 20%–30%–60% discontinuous sucrose gradient at 30,000 rpm for 1 h at 4 ◦ C. Indirect fluorescence assays (IFAs) were performed to assess the expression of HA, NA and M1 proteins, as previously described (Ma et al., 2014). Sf9 cells were infected with rBV, and the cells were cultured at 27 ◦ C for 48 h. Supernatants were discarded, and the cells were fixed with 80% pre-cooled acetone at −20 ◦ C for 2 h. Cells were washed in PBS (0.01 mol/L) and incubated with antiserum (1:200) from chickens immunized with H5N1 WIV at 37 ◦ C for 2 h. The cells were then washed with PBS and incubated with FITC-conjugated rabbit anti-chicken secondary antibody (Bioss, China) diluted in 0.1% Evans blue at 37 ◦ C for 1 h. Finally, the cells were washed with PBS and observed using fluorescence microscopy (Olympus IX51).

2.2. VLP production and protein expression 2.3. Western blot analysis Genes encoding NA, HA, M1 (4574 bp) with the polyhedrin promoter, polyadenylation signal of A/meerkat/Shanghai/SH-1/2012 (SH-1; clade 2.3.2.1), and Not I and Sph I restriction sites were

Characterization of influenza VLPs, BV and WIV was performed using a sodium dodecyl sulfate (SDS) 12% polyacrylamide gel,

Z. Ren et al. / Virus Research 200 (2015) 9–18 Table 1 Preparation of VLPs antigens. Antigen

Protein concentration (␮g/ml)

Hemagglutination unit (log2 )

Culture supernatants Purified VLPs

– 1789

7 12

The protein concentration of culture supernatants was not determined.

silver stain, Coomassie stain and Western blot analysis as described by Pushko et al. (Pushko et al., 2005). Total VLP protein amounts of 10 ␮g, 2.5 ␮g and 0.6 ␮g were loaded into the gel and transferred onto a PVDF membrane using a Mini Trans-Blot (Bio-Rad, CA). Membranes were blocked using a Blocking Buffer solution (Thermo, USA) at room temperature for 1 h and then incubated with chicken antiserum (1:500 v/v) overnight at 4 ◦ C. Membranes were then incubated with horseradish peroxidase (HRP)-conjugated goat anti-chicken IgG (1:50,000 v/v, Sigma, USA), and proteins were visualized using an ECL system (Thermo, USA). 2.4. Electron microscopy Sf9 cells infected with rBVs expressing HA, NA and M1 were fixed with 1% phosphotungstic acid for 1–2 min to examine the budding of VLPs. A piece of paper was used to wick away excess stain, and the samples were air dried for 1–3 min. Influenza VLPs were observed using an H8100 transmission electron microscope (Hitachi Ltd., Hitachi, Japan) at a magnification of 40,000×. 2.5. Immunization and challenge BALB/c mice (Mus musculus, females, 6–8 weeks of age) were purchased from the Changchun Animal Breeding Center for Medical Research, Changchun, China. Protein concentrations of purified influenza VLPs were analyzed using a BCA assay kit (Thermo, USA). Ten mice received two intramuscular injections (weeks 0 and 3) of 0.2 ml of vaccine containing 10 ␮g (contained 27 HA units) of VLP or VLP + CFA (Sigma, USA) (Table 1). The WIV group was immunized with 27 HA units of whole inactive A/meerkat/Shanghai/SH-1/2012 (SH-1; clade 2.3.2.1) virus. Ten mice from the same litter were immunized with a vaccine prepared using PBS as a mock (control) group. Mock or vaccinated mice were anesthetized with isoflurane and intranasally infected with 10MLD50 of A/meerkat/Shanghai/SH-1/2012, A/duck/Jilin/JLSIV/2013 or A/mallard duck/Shanghai/SH-9/2013 in 50 ␮l of PBS per mouse at week 2 after the second immunization for challenge studies. Mice were observed daily to monitor body weight changes and mortality. All housing conditions and experimental procedures (for mice) complied with the ethical guidelines of the International Association for the Study of Pain, and the Animal Care and Use Committee of the Chinese People’s Liberation Army approved the experimental procedures (No:SYXK2009-045). 2.6. Antibody responses and hemagglutination inhibition (HAI) titers Blood samples were collected by retro-orbital plexus puncture 0, 3 and 5 weeks after immunization. After, the samples were clotted (approximately 2 h at room temperature), and sera were collected after a brief spin at 3000 rpm for 10 min. Lung homogenates were obtained 4 days after the challenge infections. Whole lung extracts were prepared using frosted glass slides and centrifuged at 3000 rpm for 10 min to collect supernatants. Supernatants were stored at −80 ◦ C for further study.

11

Influenza virus-specific IgG, IgG1 and IgG2a antibodies were identified in sera using enzyme-linked immunosorbent assays (ELISA), as described previously (Quan et al., 2007). Briefly, each well of a 96-well plate was coated with inactivated A/meerkat/ Shanghai/SH-1/2012, A/duck/Jilin/JL-SIV/2013 or A/mallard duck/ Shanghai/SH-9/2013 at a concentration of 5 ␮g/ml in coating buffer (0.1 M sodium carbonate, pH 9.5) at 4 ◦ C overnight. Non-specific binding was blocked with PBS containing Tween 20 (0.05%) and BSA (1%) at 25 ◦ C for 2 h, and plates were incubated with serial dilutions of each serum sample at 25 ◦ C for 2 h. Plates were washed in PBS-Tween 20 (0.05%) and incubated with HRP-labeled goat anti-mouse IgG, IgG1 or IgG2a (1:5,000 v/v, Southern Biotechnology, USA) at 25 ◦ C for 1 h. Plates were thoroughly washed in PBS-Tween 20 (0.05%) and incubated with a TMB (Sigma, USA) substrate at 25 ◦ C for 30 min. The reaction was stopped with 50 ␮l of 2 M H2 SO4 , and colorimetric changes were measured as the optical density (O.D.450 nm) using a spectrophotometer (Bio-Rad). HAI titers were determined using 0.85% chicken red blood cells and 4 HA units per well of inactivated A/meerkat/Shanghai/SH-1/2012, A/duck/Jilin/JL-SIV/2013 or A/mallard duck/Shanghai/SH-9/2013 reassortants.

2.7. Lung viral titers and microneutralization assay Lung homogenates were titrated for virus infectivity in eggs from initial dilutions of 1:10. The limit of virus detection was 101.2 EID50 /ml (Pushko et al., 2007). Virus neutralization assays were performed using MDCK cells, as previously described (Suguitan et al., 2006). Pooled sera (2 mice/pool) were heat inactivated at 56 ◦ C for 30 min, and serial two-fold dilutions of pooled sera were mixed with 100 TCID50 of the A/meerkat/Shanghai/SH-1/2012, A/duck/Jilin/JL-SIV/2013 or A/mallard duck/Shanghai/SH-9/2013 strain separately. MDCK monolayer cells were added to the mixtures and incubated at 37 ◦ C for 1 h. The highest serum dilution in which no cytopathic effect was observed was recorded as the neutralizing antibody titer.

2.8. Cytokine assaysusing ELISPOT An enzyme-linked immunospot assay (ELISpot) was performed as previously described to analyze antigen-specific T cell activation induced by VLPs and WIV (Bower et al., 2004; Quan et al., 2007). Mouse IFN-␥ and IL-4 ELISPOT kits were purchased from Mabtech AB, Sweden. Anti-IFN-␥and anti-IL-4 antibodies were used to coat Multiscreen 96-well filtration plates (Millipore). Freshly prepared mouse splenocytes (1 × 106 cells) were isolated 4 days post-challenge and added to each well. Plates were blocked with 1640 medium containing 10% FBS for 2 h at 37 ◦ C and stimulated with inactivated A/meerkat/Shanghai/SH-1/2012 (0.2 ␮g per well). Plates were incubated for 24 hat 37 ◦ C in 5% CO2 followed by incubation with biotinylated detection antibody and streptavidinHRP. Finally, the spots were generated using a BCIP/NBT substrate solution after a 20-min incubation at 37 ◦ C in the dark. Spot forming units (SFU) were counted in an ELISpot reader system (Multispotreader Spectrum, AID, Strasberg, Germany).

2.9. Statistical analysis All data are expressed as means ± SD, and Student’s t-test was used for comparisons of treated groups with the mock group. Samples from VLP or VLP + CFA vaccinated animals were compared with WIV-vaccinated animals, and values of P < 0.05 were considered significant.

12

Z. Ren et al. / Virus Research 200 (2015) 9–18

3. Results 3.1. Expression and characterization of a VLP vaccine for A/meerkat/Shanghai/SH-1/2012 We produced HPAI H5N1 virus A/meerkat/Shanghai/SH-1/2012 VLPs in Sf9 cells infected with rBVs-NA-HA-M1 following a previously described procedure (Pushko et al., 2005). rBV vectors were constructed to express the H5N1 NA, HA and M1 genes in tandem, and each gene was flanked with a 5 polyhedrin promoter (PolH) and 3 transcription termination signal (p(A))(Fig. 1). IFA determined the expression levels of NA-HA-M1 using chicken antisera raised against WIV. Specific fluorescence was observed in rBV-NAHA-M1-infectedinsect cells but not in uninfected mock cells (Fig. 2A and B). Culture supernatants from rBVs-NA-HA-M1-infected cells showed hemagglutination activity (27 hemagglutination units; HAU) for 0.85% chicken red blood cells, but no activity was observed in uninfected mock cells (Fig. 2D). Purified H5N1 VLPs exhibited a consistent size of approximately 100 nm under transmission electron microscopy (Fig. 2 C); in addition, H5N1 VLPs were spherical and pleomorphic with surface spikes characteristic of influenza virus NA and HA proteins on virions. Purified H5N1 VLPs or baculovirus (BV) proteins were confirmed using SDS-PAGE, Coomassie blue-stained gels (Fig. 2E) and Western blotting using chicken polysera followed by incubation with HRP-conjugated secondary antibodies (Fig. 2F). These results suggest that H5N1 VLPs contained NA, HA and M1 with functional activity, resembled influenza virions in morphology and size, and were structurally intact. The expected molecular weights of HA, NA and M1 (70, 52 and 25 kDa, respectively) were observed in purified H5N1 VLPs. In the 10 ␮g group, only small amounts of the baculoviral proteins were observed; baculoviral proteins gp64 and vp39 (64 and 39 kDa, respectively) are indicated in the 10 ␮g BV protein in the Coomassie blue-stained gels but were not shown by Western blotting. We determined the baculovirus titers of the 10 ␮g H5N1 VLPs. The baculovirus titer was 2.8 × 105 pfu/ml, whereas the majority of baculovirus was detected in rBV with titers of 3.2 × 108 pfu/ml, which may result from the baculovirus virions having very similar densities that cannot be separated efficiently by density gradient ultracentrifugation. To determine the amounts of H5N1 VLPs and WIV protein, 10 ␮g (contained 27 HA units) of VLP or WIV was evaluated by silver stained gels (Fig. 2G). An equal amount of HA (70 kDa) was observed, but the NA and M1 proteins of the WIV were more than the VLP, and 10 ␮g BV protein was also determined. These results suggest that 10 ␮g (contained 27 HA units) of H5N1 VLPs and WIV contained equal amounts of HA, and some baculovirus is found in the purified VLP vaccine. 3.2. H5N1 VLPs and WIV elicit immune responses in mice To evaluate VLP and WIV immunogenicity, groups of mice (n = 10) were immunized twice intramuscularly (weeks 0 and 3) with 10 ␮g (contained 27 HA units) of VLP or whole inactive A/meerkat/Shanghai/SH-1/2012 virus. We determined the HAI titers against A/meerkat/Shanghai/SH-1/2012, A/duck/Jilin/JLSIV/2013 and A/mallard duck/Shanghai/SH-9/2013 viruses in immune sera collected 0, 3, and 5 weeks after immunization (Fig. 3A–C). VLP and VLP + CFA group sera showed progressive increases in HAI titers up to 128–256 against the homologous A/meerkat/Shanghai/SH-1/2012 strain, and these results were significantly (P < 0.05 or P < 0.01) higher than those observed in the WIV-immunized group at week 5. We then determined crossreactive HAI activities against the A/duck/Jilin/JL-SIV/2013 virus. As expected, the HAI titers of the VLP and VLP + CFA group were 32–64 at week 5 after vaccination, whereas the WIV-immunized group showed a titer lower than 32. Importantly, the HAI titer

induced by the VLP + CFA group was 64. HAI cross-reactive activities against the A/mallard duck/Shanghai/SH-9/2013 virus were determined. The HAI titers of the VLP + CFA group were 32–64 at week 5 after vaccination, and the VLP and WIV-immunized group showed a titer lower than 32. We next determined the levels of total IgG antibody responses specific to A/meerkat/Shanghai/SH-1/2012, A/duck/Jilin/JL-SIV/2013, and A/mallard duck/Shanghai/SH-9/2013 viruses at 0, 3, and 5 weeks after immunization with VLP or WIV (Fig. 3D, E, F). The IgG responses specific to the A/meerkat/Shanghai/SH-1/2012 virus and cross-reactive to the A/duck/Jilin/JL-SIV/2013 and A/mallard duck/Shanghai/SH-9/2013 virus increased with time post-immunization, which indicates the progressive maturation of virus-specific antibodies. The VLP group showed a 20-fold higher IgG titer against A/meerkat/Shanghai/SH1/2012 and the VLP + CFA group showed 3–4-fold higher IgG titer against A/duck/Jilin/JL-SIV/2013 or A/mallard duck/Shanghai/SH9/2013 than the WIV group at week 5. IgG2a and IgG1 antibody responses specific to the A/meerkat/Shanghai/SH-1/2012 virus were observed in immune sera at week 5 (Fig. 3G). All groups demonstrated higher IgG1 and IgG2a titers than the mock groups (P < 0.01), and the IgG2a titers were higher than the IgG1 titers in all groups that showed IgG2a-dominant antibody responses specific to the A/meerkat/Shanghai/SH-1/2012 virus. We then determined the microneutralization activity of immunized sera against the A/meerkat/Shanghai/SH-1/2012, A/duck/Jilin/JL-SIV/2013, and A/mallard duck/Shanghai/SH-9/2013 viruses in MDCK cells. The VLP and VLP + CFA group sera showed 2- to 3-fold higher neutralizing antibody titers against A/meerkat/Shanghai/SH-1/2012 and A/duck/Jilin/JL-SIV/2013 virus compared with the WIV group (Fig. 3F). As expected, all groups showed higher microneutralization titers against the A/meerkat/Shanghai/SH-1/2012 virus than the A/duck/Jilin/JL-SIV/2013 and A/mallard duck/Shanghai/SH9/2013 virus. These results indicate that the VLP and VLP + CFA groups showed higher immunogenicity than the WIV group, and these VLPs induced virus-specific antibody responses with crossreactivities to the A/meerkat/Shanghai/SH-1/2012, A/duck/Jilin/JLSIV/2013, and A/mallard duck/Shanghai/SH-9/2013 viruses. 3.3. Immunization with H5N1 VLPs confers protection from challenges with homologous and heterologous strains VLP- or WIV-vaccinated mice were challenged with homologous (A/meerkat/Shanghai/SH-1/2012, clade 2.3.2.1) and heterologous (A/duck/Jilin/JL-SIV/2013, clade 2.3.4, A/mallard duck/Shanghai/SH-9/2013, H5N8) strains to investigate whether vaccinated mice were cross-protected against a lethal challenge (Fig. 4). All mice in the VLP and VLP + CFA groups survived a lethal virus challenge with A/meerkat/Shanghai/SH-1/2012 and A/duck/Jilin/JL-SIV/2013, and VLP + CFA groups 100% survived a lethal virus challenge with A/mallard duck/Shanghai/SH-9/2013 (Fig. 4B–F). In contrast, the WIV and mock groups showed a significant and progressive loss in bodyweight and shivering with challenge infection, which indicated that these mice suffered severe illness due to A/meerkat/Shanghai/SH-1/2012 or A/duck/Jilin/JLSIV/2013 viral infection. The VLP + CFA group lost less than 5% bodyweight after A/meerkat/Shanghai/SH-1/2012, A/duck/Jilin/JLSIV/2013, and A/mallard duck/Shanghai/SH-9/2013 challenges, and 100% of the animals survived the infection. Similarly, the VLP group showed less than a 5% and 7% loss in bodyweight after A/meerkat/Shanghai/SH-1/2012 and A/duck/Jilin/JL-SIV/2013 challenges, respectively, and all animals survived. After A/mallard duck/Shanghai/SH-9/2013 challenge, the animals showed an approximately 15% loss in bodyweight, and 60% of the animals survived. Moreover, the VLP and VLP + CFA groups regained their body weight more rapidly. In contrast, the WIV group showed a 15%, 18%, and 14% loss in bodyweight. For the A/duck/Jilin/JL-SIV/2013

Z. Ren et al. / Virus Research 200 (2015) 9–18

13

Fig. 2. Recombinant baculovirus (rBV) expressed in Sf9 cells, electron microscopy, hemagglutination assays of rBV culture supernatants, silver stained SDS-PAGE, Coomassie blue-stained gel, and Western blot analysis of influenza H5N1 VLPs. IFA assay of NA, HA, and M1 proteins expressed in Sf9 cells infected with rBV. (A) rBV-NA-HA-M1-infected Sf9 cells. (B) Mock-infected Sf9 cells. The IFA assay was performed using chicken polysera. (C) Electron microscopy of influenza H5N1 VLPs. Bars represents 100 nm. Culture supernatants from rBV-NA-HA-M1-infected cells showed hemagglutination activity (27 hemagglutination units; HAU) in 0.85% of chicken red blood cells but not in uninfected mock cells (D). (E) Analysis of purified BV protein (10 ␮g of purified baculovirus protein) and H5N1 VLPs (10, 2.5 and 0.6 ␮g total protein) using SDS-PAGE and Coomassie blue stained gel and (F) Western blotting using chicken polysera followed by incubation with HRP-conjugated secondary antibodies. The expected molecular weights of HA, NA and M1 are 70, 52, and 25 kDa, respectively. (G) Silver stained gel showing the following: Lane 1:27 HAU purified WIV protein; Lane 2: 10 ␮g of purified H5N1 VLP protein; Lane 3: 10 ␮g of purified baculovirus protein; and M: standard molecular size marker.

challenge, the weight of the WIV group was 1–2 g less than the other group because the mice that succumbed more easily to influenza infection lost more bodyweight than the other group. These groups showed only 80%, 40%, and 20% protection to challenge with the A/meerkat/Shanghai/SH-1/2012, A/duck/Jilin/JL-SIV/2013, and A/mallard duck/Shanghai/SH-9/2013 strains, respectively. Mice in the mock group died from influenza infection, which indicated incomplete protection against both H5N1 strains. These results suggest that the VLP + CFA group generated protective immune responses against A/meerkat/Shanghai/SH-1/2012 (clade 2.3.2.1), A/duck/Jilin/JL-SIV/2013 (clade 2.3.4), and A/mallard duck/Shanghai/SH-9/2013 (H5N8) lethal challenges.

3.4. Vaccination with H5N1 VLPs elicits cell-mediated immunity We next measured the IFN-␥ and IL-4 cytokine levels in splenocytes from mice immunized with VLP or WIV to investigate the cellular immune responses of the VLP and WIV groups. Splenocytes were harvested from mice at day 4 post-challenge and stimulated with inactivated A/meerkat/Shanghai/SH-1/2012 virus to examine IFN-␥ (T-helper 1) and IL-4 (T-helper 2) cytokine production (Fig. 5A and B). Mice immunized with VLP or VLP + CFA showed significantly (P < 0.01) higher numbers of IFN-␥- and IL-4-secreting cells

compared with mice immunized with WIV. In addition, the WIV group showed mixed Th1- and Th2-like responses. The VLP and VLP + CFA groups showed higher levels of IFN-␥, which appeared to correlate with Th1 (IgG2a) antibody isotypes. These results indicated that H5N1 VLPs induced both Th1- and Th2-type cellular immune responses, which expanded rapidly in response to influenza virus infection.

3.5. H5N1 VLPs provide effective control of challenge virus replication We further evaluated the effect of VLPs and WIV in the suppression of viral replication in the lung. Fig. 6 shows that all surviving groups had lung virus loads lower than 103 for the A/meerkat/Shanghai/SH-1/2012 challenge virus, 105 for the A/duck/Jilin/JL-SIV/2013 challenge virus, or 104 for the A/mallard duck/Shanghai/SH-9/2013 challenge virus. The VLP group showed a 5-fold lower lung viral load than the WIV group with the A/meerkat/Shanghai/SH-1/2012 virus challenge and a 20-fold lower lung viral load with the A/duck/Jilin/JL-SIV/2013 virus challenge. The VLP + CFA group showed much lower viral loads with the A/meerkat/Shanghai/SH-1/2012, A/duck/Jilin/JLSIV/2013, or A/mallard duck/Shanghai/SH-9/2013 virus

14

Z. Ren et al. / Virus Research 200 (2015) 9–18

Fig. 3. Humoral responses and microneutralization activity. (A)-(C) HAI titers. HAI titers against A/meerkat/Shanghai/SH-1/2012 (A), A/duck/Jilin/JL-SIV/2013 (B), or A/mallard duck/Shanghai/SH-9/2013 (C) viruses at weeks 0, 3, and 5 after the first immunization were determined in groups of mice intramuscularly immunized with VLP, VLP + CFA or WIV. At week 5 post-vaccination, the serum HAI levels of the VLP + CFA group increased substantially and were significantly different from the WIV-immunized group. (D)–(F) IgG serum antibodies specific to A/meerkat/Shanghai/SH-1/2012 (D), A/duck/Jilin/JL-SIV/2013 (E), or A/mallard duck/Shanghai/SH-9/2013 (F) virus at weeks 0, 3 and 5 were determined. The highest dilution of serum showing a mean optical density at 450 nm greater than the mean plus 2 standard deviations above mock serum samples was expressed as the titer. At weeks 3 and 5, the VLP and VLP + CFA groups exhibited significantly higher IgG titers against the A/meerkat/Shanghai/SH-1/2012 virus, and the VLP + CFA group exhibited significantly higher IgG titers against the A/duck/Jilin/JL-SIV/2013 and A/mallard duck/Shanghai/SH-9/2013 virus. (G) IgG2a and IgG1 responses. Serum was serially diluted, and ELISA was performed for serum antibodies specific to the A/meerkat/Shanghai/SH-1/2012 virus. (H) Serum microneutralization titers. At week 5 post-vaccination, serum microneutralization was determined against A/meerkat/Shanghai/SH-1/2012, A/duck/Jilin/JL-SIV/2013, and A/mallard duck/Shanghai/SH-9/2013. Titers in the VLP + CFA groups increased substantially and were significantly higher than those in the WIV-immunized group. The data are shown as the means ± SD (n = 5, 5 mice per group), *P < 0.05, **P < 0.01 compared with the WIV group.

challenges than the WIV group. These results further demonstrate the VLP + CFA induced protective immune responses against A/meerkat/Shanghai/SH-1/2012, A/duck/Jilin/JL-SIV/2013, and A/mallard duck/Shanghai/SH-9/2013 lethal challenges. 4. Discussion The emergence of new subclades of H5N1 influenza virus in poultry represents the most likely culprit for future pandemics, and H5N8 has been reported in the Republic of Korea and China (Fan et al., 2014; Lee et al., 2014). Thus, the development of a safe and broadly protective vaccine for different H5 clades is a high priority for preparedness (Choi et al., 2013b; Watanabe et al., 2011). However, serological studies of confirmed cases of influenza

A (H5N1) infection suggest that cross-protection between two influenza A (H5N1) clades may be limited. Previous studies of an influenza (H5N3) WIV vaccine administered with an MF59 adjuvant in two doses showed cross-neutralization against influenza A (H5N1), but vaccines without adjuvant were poorly immunogenic (Nicholson et al., 2001; Poland, 2006; Stephenson et al., 2003). The development of influenza VLPs in insect cells has been suggested as an influenza vaccine strategy to overcome several drawbacks attributed to the egg-based system, such as the possible disruption of vaccine supplies due to a shortage of fertilized chicken embryos or potential low yields in the production of highly pathogenic influenza viruses (Quan et al., 2007; Shortridge et al., 1998). Indeed, several studies have reported that influenza VLPs could protect against heterologous strains of influenza virus challenge, which

Z. Ren et al. / Virus Research 200 (2015) 9–18

15

Fig. 4. Protection of mice from lethal A/meerkat/Shanghai/SH-1/2012 (clade 2.3.2.1), A/duck/Jilin/JL-SIV/2013 (clade 2.3.4), and A/mallard duck/Shanghai/SH-9/2013 (H5N8) challenge. Each group of mice (n = 5) was intramuscularly injected twice (weeks 0 and 3) with 10 ␮g of VLP (contained 27 HA units). The WIV group was immunized with 27 HA units of whole inactive A/meerkat/Shanghai/SH-1/2012 virus. Immunized and mock mice were intranasally infected with a lethal dose of A/meerkat/Shanghai/SH-1/2012 (10 MLD50 ), A/duck/Jilin/JL-SIV/2013 (10 MLD50 ), and A/mallard duck/Shanghai/SH-9/2013 (10 MLD50 ) viruses at week 5 after the first immunization. Mice were monitored daily for 14 days. Body weight changes after A/meerkat/Shanghai/SH-1/2012 (A), A/duck/Jilin/JL-SIV/2013 (C), or A/mallard duck/Shanghai/SH-9/2013 challenge (E). Percent survival after A/meerkat/Shanghai/SH-1/2012 (B), or A/duck/Jilin/JL-SIV/2013(D), or A/mallard duck/Shanghai/SH-9/2013 (F) challenge.

supports influenza VLPs as promising candidates against different H5N1 clades for influenza vaccines (Inn et al., 2014; Quan et al., 2007). In this study, we generated H5N1 influenza VLPs with NA-HAM1 using a baculovirus expression system and purified H5N1 VLPs through a discontinuous sucrose gradient. The baculovirus titers of the VLPs were 2.8 × 105 pfu/ml because baculovirus virions have very similar densities and cannot be separated efficiently by density gradient ultracentrifugation (Krammer and Grabherr, 2010; Krammer et al., 2010). The presence of baculovirus in VLPs demonstrates higher immunogenicity compared to mammalian cell-based systems in terms of eliciting differentiated IgG and mucosal IgA responses (Margine et al., 2012). Interestingly, a recent report showed that MyD88−/− mice have a diminished response to vaccination with baculovirus-derived VLPs compared to wild-type mice because baculovirus-induced signaling occurs mainly through TLR9 activation and downstream MyD88 (Abe et al., 2005; Abe et al., 2009; Kang et al., 2011). Baculoviruses have strong adjuvant properties to promote DC maturation, humoral and CTL responses against co-administered Ag, and the production of inflammatory

mediators through mechanisms primarily mediated by type I IFN (Hervas-Stubbs et al., 2007). The baculoviruses present in VLPs act as an adjuvant to induce strong humoral and cellular immune responses against homologous and heterologous influenza viruses in our experiment. Due to limitations of purifying VLPs through a discontinuous sucrose gradient, some baculovirus was also in the purified VLPs safety concerns in preclinical results with noninactivated preparations may be hard to translate into clinical trials where these preparations need to be inactivated (Margine et al., 2012). A report showed that after separation from host contaminants, baculoviruses, and nucleic acids using ion exchange chromatography and inactivation of residual baculovirus with 0.2% beta-propiolactone, an insect cell-derived H1N1(A/California/04/2009) pandemic influenza vaccine candidate proved to be both immunogenic and well-tolerated in healthy adults in the midst of a pandemic in Mexico (LopezMacias et al., 2011). Considering safety in human clinical trials, our H5N1 VLP needs improvement to limit the baculovirus in the VLP.

16

Z. Ren et al. / Virus Research 200 (2015) 9–18

Fig. 5. Cytokine-secreting splenocytes on day 4 post-challenge. Cytokine phenotypes for IFN-␥ (A) and IL-4 (B). Splenocytes were isolated from immunized mice 5 weeks after the first immunization and on day 4 post-challenge. ELISPOT assays were used to determine cytokine-secreting cells after stimulation with inactive A/meerkat/Shanghai/SH-1/2012 virus. Splenocytes from mock or immunized mice in the absence of inactivated A/meerkat/Shanghai/SH-1/2012 virus stimulator were used as background controls (cells alone). Spots representing cytokine-producing cells from the spleen were counted and expressed based on 106 cells per well. The data are shown as the means ± SD (n = 3, 3 mice per group), **P < 0.01 compared with the WIV group.

Fig. 6. Lung viral load on day 4 post-challenge. Mouse lungs were collected on day 4 post-challenge. Each lung was ground and cleared in 1 ml of DMEM. Lung homogenates were titrated for virus infectivity in eggs. The lung viral titer was expressed as the number of EID50 /ml. The data are shown as the means ± SD (n = 3, 3 mice per group), **P < 0.01, ***P < 0.001 compared with the WIV group.

This study demonstrated that mice immunized with VLP or VLP plus CFA adjuvant vaccines were effectively protected from disease and death when challenged with homologous (A/meerkat/Shanghai/SH-1/2012, clade 2.3.2.1) and heterologous (A/duck/Jilin/JL-SIV/2013, clade 2.3.4; H5N8 A/mallard duck/Shanghai/SH-9/2013) strains of influenza virus. In particular, our VLPs showed the advantage of inducing strong humoral and cellular immune responses against homologous and heterologous influenza viruses with or without the CFA adjuvant. These results highlight the potential of VLP vaccines to serve as effective immunogens against heterologous influenza viruses as opposed to WIV vaccines. HAI antibody titers of formalin-inactivated split virus in the range of 1:40 are required to confer 50% protection against infection for human influenza vaccinations (Bright et al., 2007; Hobson et al., 1972). In our study, IgG2a isotype antibody responses assisted in viral clearance and increased protection against lethal influenza challenge, and these antibodies likely interact with complement components and Fc receptors with high affinity (Gessner et al., 1998; Heusser et al., 1977; Huber et al., 2006). In addition, our results suggest that interactions between T-helper cells and B cells helped B cells to differentiate into plasma cells (Perrone et al., 2009). The WIV vaccine exhibited poor and partial immunogenic protection (80% survival) against homologous A/meerkat/

Shanghai/SH-1/2012 virus challenge and showed 40% protection against heterologous A/duck/Jilin/JL-SIV/2013 virus challenge and only 20% protection against A/mallard duck/Shanghai/SH-9/2013 virus challenge. WIV has also been used as an influenza vaccine in the early years of influenza vaccination, and split and subunit vaccines are most frequently used for immunization against the yearly influenza epidemics among the available vaccine formulations in recent years. In contrast, split or subunit vaccines were shown to require adjuvant systems or significantly higher antigen doses to induce immune responses (van der Velden et al., 2012). A previous study showed that an alum-adjuvanted splitvirion H5N1 vaccine of two 30 ␮g hemagglutinin (HA) doses was needed to induce an immune response that met two of three criteria for European Union licensure (Bresson et al., 2006). One study found that two doses of a non-adjuvanted vaccine with 90 ␮g of hemagglutinin (HA) induced an antibody response at protective levels in only half of an immunologically naïve population (Ferguson et al., 2006). Another study found that two doses of a 15 ␮g subunit MF59-adjuvanted H5N1 (clade 1) vaccine induced cross-reactive antibodies to a clade 2 heterologous strain response in non-elderly and elderly adults (Banzhoff et al., 2009). As shown by Felix Geeraedts in Balb/c mice (Th2 prone) and in C57Bl/6 mice (Th1 prone), WIV induced consistently higher hemagglutination inhibition titers, virus-neutralizing antibody titers, and secretion of proinflammatory cytokines by DCs than split or subunit vaccines (Geeraedts et al., 2008). A previous study showed that the immunization of healthy adults (18–59 years) and elderly subjects who received 7.5 ␮g of HA antigen A/Vietnam/1203/2004 Vero platform whole-virus vaccine and a booster vaccination of either 7.5 ␮g or 3.75 ␮g of A/Indonesia/05/2005 or A/Vietnam/1203/2004 HA, was well tolerated and induced long-lasting cross-clade immunological memory that was effectively boosted for 1-2 years (van der Velden et al., 2012). In one study, mice received one or two priming immunizations with a Vero platform whole-virus clade 1 H5N1 vaccine, and six months after the first priming immunization, the mice received either a booster immunization with the same clade 1 vaccine or a heterologous clade 2.1 vaccine. The broadest protective immunity was provided by an immunization regimen consisting of one or two priming immunizations with a clade 1 vaccine and a boosting immunization with a clade 2.1 vaccine (Sabarth et al., 2012). The Vero platform whole-virus vaccine has cross-clade antibody responses after a boosting immunization with a heterologous virus vaccine, in contrast to traditional egg-based manufacturing processes that require the use of reassortant viruses, which only contain HA and NA proteins derived from H5N1. The use of the Vero platform allows H5N1 vaccines to be manufactured using wild-type virus, providing authentic antigenicity with respect to all internal proteins (Barrett et al., 2010). In the lethal homologous A/meerkat/Shanghai/SH-1/2012 virus challenge, the groups immunized with VLP and VLP + CFA showed 100% protection, whereas the WIV group showed 80% protection. This result may have occurred because the VLPs mimic the antigenic epitopes, which retain stronger humoral and cellular immune responses than WIV (Rodriguez-Limas et al., 2013), whereas the WIV group exhibited poor and partial immunogenic protection (80% survival). More importantly, the groups immunized with VLP + CFA showed 100% protection in the lethal heterologous A/duck/Jilin/JL-SIV/2013 and A/mallard duck/Shanghai/SH-9/2013 virus challenge. The amino acid sequence homologies of the major site that determines antigenicity in the HA1 subunit between A/meerkat/Shanghai/SH-1/2012 and A/duck/Jilin/JL-SIV/2013 or A/mallard duck/Shanghai/SH-9/2013 are 86% and 89%, respectively, which may explain the reason that the VLP and VLP + CFA vaccine groups generated complete protection against the heterologous virus.

Z. Ren et al. / Virus Research 200 (2015) 9–18

In conclusion, our results demonstrate that 10 ␮g of VLP with or without CFA was sufficient to protect mice from lethal homologous or heterologous challenge and was superior to vaccination with WIV. In addition, neutralizing antibodies, humoral antibodies and cellular immune responses induced by influenza VLPs expanded rapidly upon virus challenge and inhibited viral replication. These results provide insight for the development of effective prophylactic vaccines based on VLPs to control different H5N1 HPAI clade virus outbreaks. Acknowledgements This work was supported by the National High Technology Research and Development Program (No.2012AA02A403 and No.2012AA022006) and the Special Fund for Agro-scientific Research in the Public Interest (No.201303042). References Abe, T., Hemmi, H., Miyamoto, H., Moriishi, K., Tamura, S., Takaku, H., Akira, S., Matsuura, Y., 2005. Involvement of the toll-like receptor 9 signaling pathway in the induction of innate immunity by baculovirus. J. Virol. 79, 2847–2858. Abe, T., Kaname, Y., Wen, X.Y., Tani, H., Moriishi, K., Uematsu, S., Takeuchi, O., Ishii, K.J., Kawai, T., Akira, S., Matsuura, Y., 2009. Baculovirus induces Type I interferon production through Toll-like receptor-dependent and -independent pathways in a cell-type-specific manner. J. Virol. 83, 7629–7640. Banzhoff, A., Gasparini, R., Laghi-Pasini, F., Staniscia, T., Durando, P., Montomoli, E., Capecchi, P.L., di Giovanni, P., Sticchi, L., Gentile, C., Hilbert, A., Brauer, V., Tilman, S., Podda, A., 2009. MF59-adjuvanted H5N1 vaccine induces immunologic memory and heterotypic antibody responses in non-elderly and elderly adults. PloS One 4, e4384. Barrett, P.N., Portsmouth, D., Ehrlich, H.J., 2010. Developing cell culture-derived pandemic vaccines. Curr. Opin. Mol. Therap. 12, 21–30. Bower, J.F., Green, T.D., Ross, T.M., 2004. DNA vaccines expressing soluble CD4envelope proteins fused to C3d elicit cross-reactive neutralizing antibodies to HIV-1. Virology 328, 292–300. Bresson, J.L., Perronne, C., Launay, O., Gerdil, C., Saville, M., Wood, J., Hoschler, K., Zambon, M.C., 2006. Safety and immunogenicity of an inactivated splitvirion influenza A/Vietnam/1194/2004 (H5N1) vaccine: phase I randomised trial. Lancet 367, 1657–1664. Bright, R.A., Carter, D.M., Crevar, C.J., Toapanta, F.R., Steckbeck, J.D., Cole, K.S., Kumar, N.M., Pushko, P., Smith, G., Tumpey, T.M., Ross, T.M., 2008. Cross-clade protective immune responses to influenza viruses with H5N1 HA and NA elicited by an influenza virus-like particle. PloS One 3, e1501. Bright, R.A., Carter, D.M., Daniluk, S., Toapanta, F.R., Ahmad, A., Gavrilov, V., Massare, M., Pushko, P., Mytle, N., Rowe, T., Smith, G., Ross, T.M., 2007. Influenza viruslike particles elicit broader immune responses than whole virion inactivated influenza virus or recombinant hemagglutinin. Vaccine 25, 3871–3878. Choi, J.G., Kang, H.M., Jeon, W.J., Choi, K.S., Kim, K.I., Song, B.M., Lee, H.S., Kim, J.H., Lee, Y.J., 2013a. Characterization of clade 2.3.2.1 H5N1 highly pathogenic avian influenza viruses isolated from wild birds (mandarin duck and Eurasian eagle owl) in 2010 in Korea. Viruses 5, 1153–1174. Choi, J.G., Kim, M.C., Kang, H.M., Kim, K.I., Lee, K.J., Park, C.K., Kwon, J.H., Kim, J.H., Lee, Y.J., 2013b. Protective efficacy of baculovirus-derived influenza virus-like particles bearing H5 HA alone or in combination with M1 in chickens. Vet. Microbial. 162, 623–630. Erlewyn-Lajeunesse, M., Brathwaite, N., Lucas, J.S.A., Warner, J.O., 2009. Recommendations for the administration of influenza vaccine in children allergic to egg. BMJ 339, http://dx.doi.org/10.1136/bmj.b3680. Fan, S., Zhou, L., Wu, D., Gao, X., Pei, E., Wang, T., Gao, Y., Xia, X., 2014. A novel highly pathogenic H5N8 avian influenza virus isolated from a wild duck in China. Influenza Other Respir. Viruses 8, 646–653. Ferguson, N.M., Cummings, D.A., Fraser, C., Cajka, J.C., Cooley, P.C., Burke, D.S., 2006. Strategies for mitigating an influenza pandemic. Nature 442, 448–452. Geeraedts, F., Bungener, L., Pool, J., ter Veer, W., Wilschut, J., Huckriede, A., 2008. Whole inactivated virus influenza vaccine is superior to subunit vaccine in inducing immune responses and secretion of proinflammatory cytokines by DCs. Influenza Other Respir. Viruses 2, 41–51. Gessner, J.E., Heiken, H., Tamm, A., Schmidt, R.E., 1998. The IgG Fc receptor family. Ann. Hematol. 76, 231–248. Hervas-Stubbs, S., Rueda, P., Lopez, L., Leclerc, C., 2007. Insect baculoviruses strongly potentiate adaptive immune responses by inducing type I IFN. J. Immunol. 178, 6653. Heusser, C.H., Anderson, C.L., Grey, H.M., 1977. Receptors for IgG: subclass specificity of receptors on different mouse cell types and the definition of two distinct receptors on a macrophage cell line. J. Exp. Med. 145, 1316–1327. Hobson, D., Curry, R.L., Beare, A.S., Ward-Gardner, A., 1972. The role of serum haemagglutination-inhibiting antibody in protection against challenge infection with influenza A2 and B viruses. J. Hyg. 70, 767–777.

17

Huber, V.C., McKeon, R.M., Brackin, M.N., Miller, L.A., Keating, R., Brown, S.A., Makarova, N., Perez, D.R., Macdonald, G.H., McCullers, J.A., 2006. Distinct contributions of vaccine-induced immunoglobulin G1 (IgG1) and IgG2a antibodies to protective immunity against influenza. Clin. Vaccine Immunol. 13, 981–990. Inn, K.S., Lee, G.J., Quan, F.S., 2014. A pandemic H1N1 influenza virus-like particle vaccine induces cross-protection in mice. Immunol. Invest. 43, 236–254. Jin, R., Lv, Z., Chen, Q., Quan, Y., Zhang, H., Li, S., Chen, G., Zheng, Q., Jin, L., Wu, X., Chen, J., Zhang, Y., 2008. Safety and immunogenicity of H5N1 influenza vaccine based on baculovirus surface display system of Bombyx mori. PloS One 3, e3933. Kang, S.M., Yoo, D.G., Kim, M.C., Song, J.M., Park, M.K., Eunju, O., Quan, F.S., Akira, S., Compans, R.W., 2011. MyD88 plays an essential role in inducing B cells capable of differentiating into antibody-secreting cells after vaccination. J. Virol. 85, 11391–11400. Krammer, F., Grabherr, R., 2010. Alternative influenza vaccines made by insect cells. Trends Mol. Med. 16, 313–320. Krammer, F., Schinko, T., Palmberger, D., Tauer, C., Messner, P., Grabherr, R., 2010. Trichoplusia ni cells (High Five(TM)) are highly efficient for the production of influenza A virus-like particles: a comparison of two insect cell lines as production platforms for influenza vaccines. Mol. Biotechnol. 45, 226–234. Le, T.H., Nguyen, N.T., 2014. Evolutionary dynamics of highly pathogenic avian influenza A/H5N1 HA clades and vaccine implementation in Vietnam. Clin. Exp. Vaccine Res. 3, 117–127. Lee, Y.J., Kang, H.M., Lee, E.K., Song, B.M., Jeong, J., Kwon, Y.K., Kim, H.R., Lee, K.J., Hong, M.S., Jang, I., Choi, K.S., Kim, J.Y., Lee, H.J., Kang, M.S., Jeong, O.M., Baek, J.H., Joo, Y.S., Park, Y.H., Lee, H.S., 2014. Novel Reassortant Influenza A(H5N8) Viruses, South Korea. Emerg. Infect. Dis. 20, 1087–1089. Lopez-Macias, C., Ferat-Osorio, E., Tenorio-Calvo, A., Isibasi, A., Talavera, J., ArteagaRuiz, O., Arriaga-Pizano, L., Hickman, S.P., Allende, M., Lenhard, K., Pincus, S., Connolly, K., Raghunandan, R., Smith, G., Glenn, G., 2011. Safety and immunogenicity of a virus-like particle pandemic influenza A (H1N1) 2009 vaccine in a blinded, randomized, placebo-controlled trial of adults in Mexico. Vaccine 29, 7826–7834. Ma, J., Wang, H., Zheng, X., Xue, X., Wang, B., Wu, H., Zhang, K., Fan, S., Wang, T., Li, N., Zhao, Y., Gao, Y., Yang, S., Xia, X., 2014. CpG/Poly (I:C) mixed adjuvant priming enhances the immunogenicity of a DNA vaccine against eastern equine encephalitis virus in mice. Int. Immunopharmacol. 19, 74–80. Margine, I., Martinez-Gil, L., Chou, Y.Y., Krammer, F., 2012. Residual Baculovirus in insect cell-derived influenza virus-like particle preparations enhances immunogenicity. PloS One 7, e51559. Nicholson, K.G., Colegate, A.E., Podda, A., Stephenson, I., Wood, J., Ypma, E., Zambon, M.C., 2001. Safety and antigenicity of non-adjuvanted and MF59-adjuvanted influenza A/Duck/Singapore/97 (H5N3) vaccine: a randomised trial of two potential vaccines against H5N1 influenza. Lancet 357, 1937–1943. Perrone, L.A., Ahmad, A., Veguilla, V., Lu, X., Smith, G., Katz, J.M., Pushko, P., Tumpey, T.M., 2009. Intranasal vaccination with 1918 influenza virus-like particles protects mice and ferrets from lethal 1918 and H5N1 influenza virus challenge. J. Virol. 83, 5726–5734. Poland, G.A., 2006. Vaccines against avian influenza–a race against time. N. Engl. J. Med. 354, 1411–1413. Prabakaran, M., Kolpe, A.B., He, F., Kwang, J., 2013. Cross-protective efficacy of bivalent recombinant baculoviral vaccine against heterologous influenza H5N1 challenge. Vaccine 31, 1385–1392. Pushko, P., Tumpey, T.M., Bu, F., Knell, J., Robinson, R., Smith, G., 2005. Influenza viruslike particles comprised of the HA, NA, and M1 proteins of H9N2 influenza virus induce protective immune responses in BALB/c mice. Vaccine 23, 5751–5759. Pushko, P., Tumpey, T.M., Van Hoeven, N., Belser, J.A., Robinson, R., Nathan, M., Smith, G., Wright, D.C., Bright, R.A., 2007. Evaluation of influenza virus-like particles and Novasome adjuvant as candidate vaccine for avian influenza. Vaccine 25, 4283–4290. Qiao, C.L., Yu, K.Z., Jiang, Y.P., Jia, Y.Q., Tian, G.B., Liu, M., Deng, G.H., Wang, X.R., Meng, Q.W., Tang, X.Y., 2003. Protection of chickens against highly lethal H5N1 and H7N1 avian influenza viruses with a recombinant fowlpox virus co-expressing H5 haemagglutinin and N1 neuraminidase genes. Avian Pathol. 32, 25–32. Quan, F.S., Huang, C., Compans, R.W., Kang, S.M., 2007. Virus-like particle vaccine induces protective immunity against homologous and heterologous strains of influenza virus. J. Virol. 81, 3514–3524. Quan, F.S., Vunnava, A., Compans, R.W., Kang, S.M., 2010. Virus-like particle vaccine protects against 2009 H1N1 pandemic influenza virus in mice. PloS One 5, e9161. Rezaei, F., Mirshafiey, A., Shahmahmoodi, S., Shoja, Z., Ghavami, N., MokhtariAzad, T., 2013. Influenza virus-like particle containing two different subtypes of hemagglutinin confers protection in mice against lethal challenge with A/PR8 (H1N1) and A/HK (H3N2) viruses. Iran. Red Crescent Med. J. 15, 75–82. Robertson, J.S., Engelhardt, O.G., 2010. Developing vaccines to combat pandemic influenza. Viruses 2, 532–546. Rodriguez-Limas, W.A., Sekar, K., Tyo, K.E., 2013. Virus-like particles: the future of microbial factories and cell-free systems as platforms for vaccine development. Curr. Opin. Biotechnol. 24, 1089–1093. Sabarth, N., Savidis-Dacho, H., Schwendinger, M.G., Bruhl, P., Portsmouth, D., Crowe, B.A., Kistner, O., Barrett, P.N., Kreil, T.R., Howard, M.K., 2012. A cell culturederived whole-virus H5N1 vaccine induces long-lasting cross-clade protective immunity in mice which is augmented by a homologous or heterologous booster vaccination. Vaccine 30, 5533–5540. Shortridge, K.F., Zhou, N.N., Guan, Y., Gao, P., Ito, T., Kawaoka, Y., Kodihalli, S., Krauss, S., Markwell, D., Murti, K.G., Norwood, M., Senne, D., Sims, L., Takada, A., Webster, R.G., 1998. Characterization of avian H5N1 influenza viruses from poultry in Hong Kong. Virology 252, 331–342.

18

Z. Ren et al. / Virus Research 200 (2015) 9–18

Stephenson, I., Nicholson, K.G., Colegate, A., Podda, A., Wood, J., Ypma, E., Zambon, M., 2003. Boosting immunity to influenza H5N1 with MF59-adjuvanted H5N3 A/Duck/Singapore/97 vaccine in a primed human population. Vaccine 21, 1687–1693. Suguitan Jr., A.L., McAuliffe, J., Mills, K.L., Jin, H., Duke, G., Lu, B., Luke, C.J., Murphy, B., Swayne, D.E., Kemble, G., Subbarao, K., 2006. Live, attenuated influenza A H5N1 candidate vaccines provide broad cross-protection in mice and ferrets. PLoS Med. 3, e360. van der Velden, M.V., Aichinger, G., Pollabauer, E.M., Low-Baselli, A., Fritsch, S., Benamara, K., Kistner, O., Muller, M., Zeitlinger, M., Kollaritsch, H.,

Vesikari, T., Ehrlich, H.J., Barrett, P.N., 2012. Cell culture (Vero cell) derived whole-virus non-adjuvanted H5N1 influenza vaccine induces long-lasting cross-reactive memory immune response: homologous or heterologous booster response following two dose or single dose priming. Vaccine 30, 6127–6135. Watanabe, Y., Ibrahim, M.S., Ellakany, H.F., Kawashita, N., Mizuike, R., Hiramatsu, H., Sriwilaijaroen, N., Takagi, T., Suzuki, Y., Ikuta, K., 2011. Acquisition of human-type receptor binding specificity by new H5N1 influenza virus sublineages during their emergence in birds in Egypt. PLoS Pathog. 7, e1002068.