Healthy fertilized White Leghorn eggs were purchased from a local farm in Korea and hatched in our laboratory. Chickens aged 4 weeks were used in the immunization experiments. Immunized chickens (384HA, 192HA and 96HA) were housed separately in cages arranged in the animal center of Chungnam National University. The chickens were moved to a Biosafety Level-3 (BSL-3) poultry isolator to perform the challenge study with HP H5N6 and HP H5N1 viruses, and each group of infected chickens was housed separately in a poultry isolator during the study period.

HP H5N6; A/Waterfowl/Korea/S57/2016 (clade 2.3.4.4) was isolated in our laboratory from the feces of wild birds and was used as a seed strain for vaccine development and homologous virus challenge studies. Dr. Malik Peiris of the University of Hong Kong kindly provided the HP H5N1 A/Hong Kong/213/2003 (clade 1) virus. Virus strains were propagated in the allantoic fluid of 10-day-old embryonated chicken eggs at 37 °C for 72 h.

All experiments related to highly pathogenic influenza viruses were conducted in BSL-3 facility certified by the Korean government.

A reassorted virus was generated using a plasmid-based reverse genetic system, as described by Hwang et al. [16]. Briefly, RNA was extracted from A/Waterfowl/Korea/s57/2016 (H5N6) cells using an RNeasy mini kit (Qiagen, Germany). The gene sequence of HA, which encodes a polybasic amino acid cleavage site, was deleted using overlapping primer PCR. The polybasic amino acid-depleted HA of H5N6 and NA of H5N6 were cloned into the pHW 2000 vector. PB2, PB1, PA, NP, M, and NS plasmids of A/Puerto Rico/8/34 (PR8) (H1N1) were kindly provided by Dr. Robert Webster of St. Jude Children’s Research Hospital, Memphis, TN, USA. One microgram of each plasmid was introduced into Vero cells at 80% confluency in 6-well cell culture plates, and then the Vero cells were transfected for 48 h. The cell supernatant (300 $\mu$L) was treated with L-1-tosylamido- 2-phenylethyl chloromethyl ketone (TPCK)-trypsin (1 $\mu$g/mL) and inoculated into the allantoic cavity of 10-day-old embryonated chicken eggs. The reassorted viruses were validated through genetic analysis, and the viral titer was determined using a hemagglutination assay (HA).

The H5N6 vaccine virus was propagated in 10-day-old embryonated eggs (37 °C). The allantoic fluid was harvested 72 h post-inoculation and centrifuged at 14,981 $\times$ g for 10 min at 4 °C to remove the cell debris. The harvested fluid was filtered using 0.45-$\mu$m and 0.22-$\mu$m bottle top vacuum membrane filters (Corning Inc., Corning, NY, USA). The clarified vaccine virus was concentrated to 1/10th (v/v) of the original volume using a tangential flow filtration (TFF) system (Millipore Corporation, Bedford, MA, USA). The virus was inactivated by incubating with 0.02% formaldehyde (Sigma-Aldrich Inc., St. Louis, MO, USA) for 24 h at 4 °C. Diafiltration of the vaccine virus with phosphate-buffered saline (PBS; pH 7.40) was performed using the TFF system.

Complete inactivation of the vaccine virus was confirmed by inoculating the vaccine virus into the allantoic fluid of 10-day-old embryonated eggs, followed by the hemagglutination (HA) assay using 0.5% Turkish erythrocytes.

The inactivated vaccine virus (antigen) titer was determined using an HA assay with 0.5% turkey blood. Antigen doses were selected based on the HA units. The 384HA, 192HA, and 96HA units of antigen-containing solutions were prepared using PBS (pH = 7.40). The oil-adjuvanted inactivated H5N6 vaccine was formulated by mixing 30% antigens diluted in PBS and 70% oil adjuvant (Montanide ISA 70 VG, SEPPIC Co., Puteaux, France). Three groups of 4-week-old chickens (n = 10) were immunized intramuscularly (i.m.) using 500 $\mu$L of the vaccine formulation containing three inactivated H5N6 antigen doses of 384 HA, 192HA, and 96HA units. Further, 10, 5, and 5 chickens were immunized i.m. using 384HA, 192HA, and 96HA vaccines. Unimmunized chickens (n = 1 per group) were used as a negative control.

To determine the long-term efficacy of an immune response elicited by the vaccine against homologous virus infections, each group of immunized chickens (384HA (n = 10), 192HA (n = 5), and 96HA (n = 5)) was intranasally (i.n.) challenged using 1 mL of 100 TCID${}_{50}$/mL of A/Waterfowl/Korea/s57/2016 (H5N6) (clade 2.3.4.4) after 12 weeks of vaccination. To evaluate the cross-clade protective immunity of the inactivated H5N6 vaccine, 384HA inactivated H5N6 antigens containing vaccine immunized chickens (n = 4) were i.n. challenged with 1 mL of 100 TCID${}_{50}$/mL of A/Hong Kong/213/2003 (H5N1) (clade 1) 4 weeks after vaccination. Five unimmunized chickens infected with 1 mL of 100 TCID${}_{50}$/mL of A/Waterfowl/Korea/s57/2016 (H5N6) (clade 2.3.4.4) and 4 unimmunized chickens i.n. infected with 1 mL of 100 TCID${}_{50}$/mL of A/Hong Kong/213/2003 (H5N1) (clade 1) were maintained as positive controls with each virus challenge group simultaneously.

To further examine the effectiveness of the 96HA vaccine, we infected five chickens that had received two doses of the 96HA H5N6 vaccine with 1 mL of 100 TCID50/mL of HP H5N6 virus after 12 weeks of vaccination. A separate group of unimmunized chickens (n = 3) i.n. infected with1 mL of TCID${}_{50}$/mL of HP H5N6 homologous virus was used as a control.

Blood was collected from chickens in each vaccinated group at 2-week intervals two weeks after immunization. Sera were extracted from each blood sample and heat-inactivated at 56 °C for 30 min. An HI assay was performed to determine the antibody titer in chickens against homologous HP H5N6 and heterologous HP H5N1 virus infections. Briefly, sera were diluted 10-fold before performing 2-fold serial dilutions in PBS (pH 7.40) in V-bottom 96-well plates. Eight hemagglutination (HA) units (25 $\mu$L) of H5N6 virus were added and the plates were incubated at room temperature for 30 min. Then, 50 $\mu$L of 0.5% Turkey’s red blood cells were added, followed by incubation for 30 min at room temperature. HI titers were calculated as the reciprocal of the dilution factor that showed complete inhibition of hemagglutination.

The trachea and cloacae of surviving chickens were swabbed at 1, 3, 5, and 7 days post infection (p.i.). Tracheal and cloacal swab samples were collected in 250 $\mu$L and 500 $\mu$L PBS (pH 7.40) (per sample), respectively. The swab samples were homogenized through vortexing and filtered using a 0.22-$\mu$m Ministart syringe filter (Sartorius Stedium Biotech GmbH, Goettingen, Germany).

Viral RNA was extracted from tracheal and cloacal swab samples of immunized and unimmunized chickens challenged with HP H5N6 or HP H5N1, which were collected on day 3 p.i. using a HiGene${}^{\text{TM}}$ Viral RNA/DNA prep kit (column type) (BIOFACT Co., Ltd., South Korea). Briefly, 250 $\mu$L of VL buffer and 20 $\mu$L of proteinase K were added to 150 $\mu$L of each tracheal or cloacal sample collected on day three p.i. The samples were then vortexed for 1 min and incubated at room temperature for 10 min before 350 $\mu$L of 100% ethanol was added to each sample. The lysate/ethanol mixture was transferred to a capsule column and centrifuged at 13,000 rpm for 1 min. The column was washed with 500 $\mu$L of VW1 buffer which was followed by centrifugation at 13,000 rpm for 1 min. Five hundred microliters of 100% ethyl alcohol were washed twice followed by centrifugation at 13,000 rpm for 1 min. RNA was eluted using 30 $\mu$L of nuclease-free water.

To detect the virus, real-time quantitative fluorescent PCR was performed using the TOPreal${}^{\text{TM}}$ one-step RT qPCR kit (TaqMan probe) (Enzynomics, Daejeon, South Korea) and H5 influenza primers and probes. A total of 20 $\mu$L of PCR reaction mixture was prepared by mixing 5 $\mu$L of TOPreal${}^{\text{TM}}$ one-step RT qPCR kit (TaqMan probe), 1 $\mu$L of 10 pmol of each primer (H5 1504 Forward, 5${}^{\prime }$-ACA TAT GAC TAC CCA CAR TAT TCA G – 3${}^{\prime }$, H5 1655 Reverse; 5${}^{\prime }$- AGA CCA GCT AYC ATG ATT GC-3${}^{\prime }$), 1 $\mu$L of H5 probe (5${}^{\prime }$-FAM-TCW ACA GTG GCG AGT TCC CTA GCA-TAMRA-3${}^{\prime }$), 10 $\mu$L of template RNA, and 2 $\mu$L of nuclease-free water. RNA was amplified in a Rotor-Gene 6000 system (QIAGEN, Hilden, Germany) using the RT qPCR cycle with reverse transcription at 50 °C for 30 min, initial denaturation at 95 °C for 10 min, followed by 45 cycles of denaturation at 95 °C for 5 s, annealing, and elongation at 60 °C for 30 s. A standard curve for H5 influenza virus was generated using the plaque assay data of stock viruses.

To determine tissue histopathology, on day 6 p.i., lung and tracheal tissues were collected from immunized chickens challenged with HP H5N6 or HP H5N1 viruses. Tissues from HP H5N6 infected unimmunized chickens and HP H5N1 infected unimmunized chickens were obtained on days 1 and 3 p.i., respectively. The tissues were immersed in phosphate-buffered formalin (10%) (Triangle Biomedical Sciences, Durham, USA) and fixed for 6 h while shaking. The tissues were then washed under running tap water and embedded in paraffin. Tissues were prepared into 5-$\mu$m sections. The sections were stained with Harris hematoxylin (H) for 1 min and counterstained with 1% eosin Y solution (E) for 1 min. The stained tissue sections were preserved using Canada balsam, observed under an Olympus DP70 microscope (Olympus Corporation, Tokyo, Japan), and photographed.

Experimental data were analyzed using the Wilcoxon signed-rank test and multiple t-tests in GraphPad Prism software (Version 8.0.1) (GraphPad software. Inc., San Diego, CA, USA). Statistical p-values less than 0.05 (p 0.05) were considered significant.

Chickens immunized with a single dose of the 384HA, 192HA, and 96HA vaccines and two doses of the 96HA vaccine were challenged with homologous HP H5N6 at 12 weeks p.v. Survival of chickens was observed up to 14 days p.i. The 384HA- or 192HA-immunized chickens exhibited 100% survival rate, while survival of the 96HA vaccinated chickens dropped to 20% on day 6 p.i. and remained constant until day 14 p.i. Chickens that received two doses of the 96HA vaccine also showed 100% survival up to 14 days p.i. When unimmunized chickens were infected with the homologous HP H5N6 virus, all chickens (n = 5) died within two days of infection (Fig. 1A).

Fig. 1.

Survival Rate of immunized chickens after the challenge. Immunized chickens were intranasally infected with 100 TCID${}_{50}$/mL of homologous A/waterfowl/Korea/s57/2016 (H5N6) (clade 2.3.4.4) or heterologous A/Hong Kong/213/2003 (H5N1) (clade 1) viruses. Survival was observed until 14 days after infection. (A) Survival rates of chickens challenged with the homologous H5N6 virus. (B) Survival rate of chickens challenged with the heterologous H5N1 virus.

Interestingly, all chickens immunized with the 384HA oil-adjuvanted inactivated vaccine and infected with heterologous HP H5N1 virus (n = 4) four weeks p.v. survived throughout the 14 days of observation. All unimmunized chickens infected with heterologous HP H5N1 (n = 4) survived until day 6 p.i. (Fig. 1B).

Sera were extracted from blood collected from the immunized chickens starting at 2 weeks p.v. up to 12 weeks p.v. The HI assay was performed to determine antibody titers in immunized chickens. The mean hemagglutination inhibition (HI) titer was 240 against homologous HP H5N6 and 80 against heterologous HP H5N1 influenza virus (clade 1) 4 weeks p.v. in 384HA-vaccinated chickens (Fig. 2A,C).

Fig. 2.

Antibody titers in immunized chickens. 4-week-old chickens were injected with 500 $\mu$L of 384HA units of oil-adjuvanted inactivated H5N6 vaccine into the pectoral muscle. 2 weeks after vaccination, sera were collected from chickens for 12 weeks. Antibody titers in the sera were determined using the HI assay. (A) Mean HI titers in sera of vaccinated chickens 4 weeks after immunization against HP H5N6 (clade 2.3.4.4) and HP H5N1 (clade 1) viruses. (B) Mean HI titer of 384HA vaccinated chickens up to 12 weeks p.v. against HP H5N6 (clade 2.3.4.4). (C) Individual antibody titers of 192HA and 96HA vaccine-immunized chickens at week 12 p.v. against HP H5N6 (clade 2.3.4.4) infection. Data are presented as mean titers of vaccinated chickens $\pm$ standard deviation. * p 0.05. The limit of detection was 10HI.

The mean HI titer peaked 6 weeks after vaccination and was reported to be 156 against homologous HP H5N6 infection. The vaccinated chickens showed a mean HI titer of 60 against A/waterfowl/Korea/s57/2016 (H5N6) at 12 weeks p.v., which was significantly higher than the HI titers in unimmunized chickens (Fig. 2B).

To determine the viral titer, the trachea and cloacae of surviving chickens were swabbed on day 3 p.i. No virus was detected in the tracheal and cloacal samples of 384HA-,192HA-, and 96HA-vaccinated chickens that were challenged with the homologous H5N6 virus (Fig. 3A). Cloacal and tracheal swabs from chickens vaccinated with two doses of the 96HA vaccine and challenged with HP H5N6 virus did not show any viral titers. However, unimmunized chickens exposed to HP H5N6 showed a significantly high viral titer of (411.9 PFU/mL) in the trachea on day 1 p.i. and no viruses were shed through the cloacae 1 day p.i. (Supplementary Fig. 1). We could not measure the viral titer of unimmunized chickens in subsequent days since they died 2 days p.i. Interestingly, no virus was detected in tracheal and cloacal swabs when the 384HA vaccine was administered to chickens infected with HP H5N1, while HP H5N1 infected unimmunized chickens showed significantly higher viral titers in the cloacal (Log${}_{10}$ 4.09 PFU/mL, p 0.0001) and tracheal (Log${}_{10}$ 5.06 PFU/mL, p 0.0001) swabs (Fig. 3B).

Fig. 3.

Viral titers in tracheal and cloacal swabs of vaccinated chickens after challenge with HP H5N6 or HP H5N1 viruses. Tracheal and cloacal swabs were collected in PBS on day 3 p.i. from chickens that received the vaccine and were challenged with HP H5N6 A/Waterfowl/Korea/S57/2016 (clade 2.3.4.4) and HP H5N1 A/Hong Kong/213/2003 (clade 1) viruses. RNA was extracted from the swab samples, and RT qPCR was performed using H5 influenza primers and probes. (A) Viral titers of immunized chickens challenged with HP H5N6 (clade 2.3.4.4) (384HA; n = 10, 192HA; n = 5, and 96HA; n = 4). (B) Viral titers in immunized chickens challenged with HP H5N1 (clade 1) (n = 4). Unimmunized chickens infected with HP H5N6 homologous virus (n = 5) and unimmunized chickens infected with HP H5N1 heterologous virus (n = 4) were used as controls. Data are represented as Log10 plaque forming units (PFU)/mL of viral titers in vaccinated chickens $\pm$ standard deviation. **** p 0.0001. The limit of detection was 2.2 Log10 PFU/mL.

None of the 384HA-immunized chickens that were infected with homologous HP H5N6 virus exhibited signs of pneumonia in the lung tissue (Fig. 4A). However, lung tissues of 192HA- and 96HA-vaccinated chickens that were challenged with homologous HP H5N6 virus showed symptoms of mild pneumonia (Fig. 4B,C) when compared with the lung tissues of only HP H5N6 infected chickens, which showed signs of severe pneumonia (Fig. 4D). When 384HA-immunized chickens were infected with the HP H5N1virus, their lung tissues showed mild pneumonia (Fig. 5A) compared to the lung tissues of HP H5N1-infected, unimmunized chickens, which showed signs of severe pneumonia (Fig. 5B). The lung tissues of unimmunized and non-infected chickens did not show any signs of pneumonia (Figs. 4E,5C).

Fig. 4.

Pathology of lung tissues of immunized chickens challenged with HP H5N6: A/Waterfowl/Korea/S57/2016 (clade2.3.4.4). (A) Lung tissue from 384HA-immunized chicken. (B) Lung tissue from 192HA-immunized chicken. (C) Lung tissue from 96HA-immunized chicken. (D) Lung tissue from HP H5N6-challenged, unimmunized chicken. (E) Uninfected chicken lung tissue.

Fig. 5.

Pathology of lung tissues of immunized chickens challenged with HP H5N1: A/ Hong Kong/213/2003 (clade 1). (A) Lung tissue from 384HA-immunized chicken. (B) Lung tissue from H5N1 infected unimmunized chicken. (C) Uninfected chicken lung tissue.