Protection of layers and breeders against homologous or heterologous HPAIv by vaccines from Korean national antigen bank

Korean government has selected and stocked five type antigens of two clades as Korean national antigen bank having high possibility of introduction to Korea. We aimed to evaluate the efficacy of the clade 2.3.2.1c and 2.3.4.4c H5Nx vaccines from the Korean avian influenza (AI) national antigen bank for emergency preparedness for their potency and protective efficacy against lethal homologous and heterologous viruses in layer and breeder chickens practically. The PD50 (dose of vaccine that protects 50% of chickens from viral challenge) of all vaccinated groups was >50, which was satisfied with minimum antigen requirement of OIE, and the PD50 levels of the two vaccines differed depending on strain and chicken breed. In homologous challenge, all vaccinated groups exhibited 100% survival with no clinical symptoms and high levels of pre-challenge protective immunity (7.2–8.5 log2), although they did not completely prevent virus shedding. On the other hand, against heterologous virus challenge, vaccinated animals exhibited 62.5–80% survival with lower antibody titers (2.3–3.4 log2) and a longer period of virus shedding (14 days post infection [dpi]). Our results suggest that the clade 2.3.2.1c and 2.3.4.4c H5Nx vaccines are good candidates for emergency vaccination of commercial chickens and support the idea that close genetic matching between vaccine and challenge virus provides the best protection.


Study 1: Potency of vaccines against homologous viruses in commercial chickens. Clinical pro-
tection. In layer and breeder chickens, vaccination with a 1 dose of rgKA435/2.3.2.1c conferred 100% clinical protection from challenge with homologous virus, with no clinical symptoms, whereas vaccination with 0.1 dose resulted in 20% mortality by 8 dpi only in layers (Fig. 1). Vaccination with 0.01 dose resulted in higher mortality and clinical signs of infection than the 1 dose and 0.1 dose groups. Vaccination of layer chickens with 0.01 doses led to 30% mortality by 8 dpi, with two chickens dying between 7 and 8 dpi with neurological signs and diarrhea (Fig. 1A). Vaccination of breeder chickens with 0.01 dose led to 60% mortality by 5 dpi (Fig. 1C), with four chickens dying between days 4 and 5 with neurological signs. For rgES2/2.3.4.4 C, vaccination with 0.01 dose resulted in no mortality in layer chickens (Fig. 1B), but 25% mortality in breeder chickens (Fig. 1D). The mean time to death (MDT) in both 0.01 dose vaccination groups was 4.6-6.0 days [ Table 1]. For sham-treated chickens, mean time to death was 2.0-3.7 days.

Discussion
It is crucial that vaccine potency and efficacy are evaluated in commercial chickens prior to their emergency use in the field, because the efficacy of AI vaccines differs among poultry species (chicken, duck, and quail) and breeds (layer and breeder) 19 . Hence, we evaluated vaccine efficacy against homologous and heterologous HPAIV in commercial chickens by measuring antibody titers after vaccination and monitoring clinical signs and virus shedding post-infection.
The minimum antigen requirement for licensing vaccines is 50 PD 50 per dose, which ensures that there is sufficient antigen mass or virus titer to be efficacious in the field 20,21 . The two representative vaccines that we obtained from the Korean national AI antigen bank satisfied this criterion. The PD 50 values of rgES2/2.3.4.4c in layers and breeders were higher than those of rgKA435/2.3.2.1c. Moreover, rgES2/2.3.4.4c yielded a 100% survival rate, depending on dose (1 to 0.01), and no virus shedding was observed in OP and CL at 3 dpi except in one layer chicken that received 0.1 dose [ Table 1]. The differences in PD 50 among vaccine strains may be attributed to differences in the virulence of homologous challenge viruses, despite their similar immunogenicity. This result corresponds with a report that one virus within clade 2.3.2.1c 22 had a higher Lethal Dose 50 (LD 50 ) and shorter Mean Death Time (MDT) in chickens than ES2/2.3.4.4c 23 . Finally, differences in pathogenicity between viruses result in differences in mortality among vaccinated chickens, again depending on the dose, and were reflected in the PD 50 value 18 .
Antibody titers in commercial chickens are lower than those in SPF chickens due to average 35 weeks age and puberty. However, in previous study, two representative vaccines from the national AI antigen bank had higher potency in both kinds of commercial chickens than in SPF chickens 9 . The layer chickens used in this study were 35-week-old and had brown feathers, which are recognized as exhibiting reduced virulence against HPAI 24 . Additionally, the pathogenesis of avian influenza might be showed in the difference of susceptibility depending on age in various reports [25][26][27][28][29] although it was recently reported that age is not a determinant factor in susceptibility to H5N2 HPAIv 30 . Our results are not consistent with a previous study showing that commercial chickens experienced less immunization than laboratory chickens due to maternal antibodies, immunosuppressive viruses, and the use of a lower vaccine dose 31 .
For emergency vaccination, it was important to evaluate cross-protection against heterologous viruses in commercial chicken. Although they did not completely prevent virus shedding in commercial chickens, the two vaccine groups exhibited a 100% survival rate and higher antibody titer against homologous challenge. By contrast, heterologous vaccine groups yielded a 62.5-80% survival rate with lower HI titer. This result corresponds with a previous report that protection in low antibody titer might be satisfied minimum antibody titer [> 3 log 2 ] for survival in case the vaccine and field viruses 32 . Meanwhile, some layer chickens under 3 log 2 HI titer could survive challenge with heterologous virus, possibly due to their H9N2 antibody titer of 4.5-5.5 log 2 . Seo et. al 33 . reported that most young chickens infected with an H9N2 influenza virus survived lethal challenge with an H5N1 influenza virus, but infected birds shed H5N1 influenza virus in their feces due to adoptive transfer of T lymphocytes www.nature.com/scientificreports www.nature.com/scientificreports/ or CD8 + T cells. However, further study is needed on how H9N2 vaccination affects H5 HPAI vaccination to host.
Notably, virus shedding in OP and CL was observed until 14 dpi in one breeder chicken (Fig. 6B). This could be due to the distinguishable amino acidic differences (10.5%, data not shown) in whole hemagglutinin (HA) similarity between KA435/2.3.2.1c and ES2/2.3.4.4c. Antigenic matching of HA between vaccine and field viruses provides the best protection against mortality and virus shedding, assuming a comparable host immune response 34 . Therefore, emergency vaccination should only be considered if the vaccine is a 95% or better match to the strains circulating in Korea.
In conclusion, our study demonstrated the potency and efficacy of two representative vaccines, clade 2.3.2.1c and 2.3.4.4c H5Nx, and confirmed that they offered good protection against homologous challenge in commercial chickens. In addition, we found that vaccine potency may be influenced by the virulence of the challenge virus, as well as chicken breed. Cross-protection testing revealed that survival rate was lower, and virus shedding period was longer, when the vaccine and field strain were mismatched. Our findings suggest that these two representative vaccines effectively protect commercial chickens against homologous viruses, but are significantly less protective when the vaccine and field strain are mismatched.

Materials and Methods
Viruses and vaccine development. Two different H5 HPAIVs were used as inactivated-vaccine seed strains and challenge strains. These strains were selected from the Korean national AI antigen bank.

Animals (layer and breeder chickens).
Vaccine efficacy and potency experiments used layer(Hy-line brown) and breeder(Ross) chickens obtained from Korean commercial chicken farms. Specifically, the animals were 35-week-old layer and breeder chickens serologically positive for H9 due to H9N2 LPAI vaccination and negative for H5, as determined by the hemagglutination inhibition (HI) assay. All experiments with live H5 virus were performed in biosafety level 3 facilities, following guidelines approved by the Animal Ethics Committee of the Animal and Plant Quarantine Agency, Korea (Approval number: 2019-176).

Study 1: Potency of vaccines in commercial chickens against homologous viruses.
To evaluate the potency (in terms of PD 50 , the dose of vaccine that protects 50% of chickens from viral challenge) and efficacy of the inactivated vaccines, 40 35-week-old layer chickens and 32 35-week-old breeder chickens for each vaccine were divided into four groups (10 chickens per group in layers; 8 chickens per group in breeders): three immunization groups and one non-immunization (sham) group. Immunizations, which were intramuscular, delivered 1, 0.1, or 0.01 doses, obtained by serially diluting the vaccines in PBS and mixing the dilutions (30:70, w/w) with the adjuvant Montanide ISA VG70. The sham group was inoculated with PBS and the adjuvant Montanide ISA VG70. At 3 weeks post-vaccination (wpv), chickens were challenged intranasally with 0.1 ml of PBS containing 10 6 EID 50 (the amount of virus that will infect 50 percent of inoculated eggs) of virus homologous to the vaccine strain. Post-challenge, chickens were monitored daily for clinical signs and survival. PD 50 was calculated using mortality as the endpoint, as described previously 36 .
Serology and antibody assays. Serum samples were collected from all chickens prior to vaccination and weekly for 3 weeks following vaccination. Blood samples were also obtained from all living chickens at 14 days post-challenge. Hemagglutination inhibition (HI) assays were performed using standard methods and homologous antigens (8 HA units, as determined using chicken RBCs) 37 .
Post-challenge virus shedding. Oropharyngeal and cloacal swabs were collected from animals in all groups at 3, 5, 7, 10, and 14 days post-challenge (dpc). Each oropharyngeal or cloacal sample was suspended in 1 ml of maintenance medium containing antibiotic-antimycotic mixture (Invitrogen, Carlsbad, CA, USA). Samples were used for inoculation of Dermal Fibroblast 1 (DF1) cells, and virus growth was determined based on cytopathic effects (CPE) and HA activity. Virus titers were calculated as described elsewhere 36 , and the limit of virus detection was <1. Statistical significance of differences between measurements was determined using Student's t-test, with a P-value <0.05 indicating a significant difference.

Study 2: Efficacy of vaccines in commercial chickens against challenge with heterologous vaccines.
To evaluate the efficacy of two representative vaccines (rgKA435/2.3.2.1c and rgES2/2.3.4.4c) against challenge with heterologous viruses, 60 35-week-old layer chickens and 48 35-week-old breeder chickens for each vaccine were divided into six groups (10 chickens per group in layers and 8 chickens per group in breeders): four immunized and two non-immunized groups. The immunized groups were intramuscularly vaccinated with a single dose of rgKA435/2.3.2.1c or rgES2/2.3.4.4c vaccine with adjuvant Montanide ISA VG70. The two sham groups were inoculated with a mixture of PBS and adjuvant Montanide ISA VG70. To assess immunogenicity post-vaccination, all chickens in each group were bled weekly, and the HI assay was used to measure serum antibody levels in each group using homologous and heterologous antigens. Three weeks after vaccination, chickens were intranasally challenged with 10 6.0 EID 50 /0.1 ml of homologous or heterologous strain (KA435/2.3.2.1c and ES2/2.3.4.4c). The protective efficacy of the vaccine was determined by evaluating clinical signs, mortality, and