Influenza A viruses continuously circulate in wild aquatic birds worldwide. Seventeen hemagglutinin (HA) subtypes and nine neuraminidase (NA) subtypes have been identified – two key surface proteins of the virus. HA enables viral attachment and entry into host cells, and its external domain contains the major antibody-recognition sites.
Avian and human influenza viruses differ in receptor specificity: avian viruses bind α2,3-linked sialic acids, whereas human viruses bind α2,6-linked sialic acids. This distinction determines which species can be infected.
Subtypes A(H5) and A(H7) are of particular concern because they may evolve into highly pathogenic avian influenza viruses capable of causing severe disease with high mortality. A(H5) viruses of the A/goose/Guangdong/1/1996 lineage were first detected in Hong Kong in 1997. Since then, they have spread across regions and become established in domestic and wild bird populations. These viruses have infected more than 60 mammalian species, caused mass mortality among marine mammals, and triggered large outbreaks on U.S. dairy farms. More than a thousand human infections have been reported, nearly half of them fatal.
Continuous circulation of A(H5) has driven substantial genetic and antigenic diversification of hemagglutinin. Moreover, it complicates vaccine development: the World Health Organization must update the list of candidate vaccine viruses twice each year. Since 2006, 48 variants have already been recommended, highlighting the scale of the challenge. Traditional inactivated vaccines require long production times – up to six months – leaving populations vulnerable to infection.
Understanding how A(H5) viruses have evolved and which antigenic features they have acquired is essential for developing more universal vaccines. Scientists produced a high-resolution antigenic map of historical, recent, and contemporary A(H5) influenza viruses. This map helps quantify how distinct viruses are from one another and identify centrally positioned antigens – those that can elicit immune responses across multiple variants.
Using this map, researchers designed vaccine antigens with high immunogenicity and broad cross-reactivity. Experiments in ferrets confirmed that such antigens elicit strong immune responses and confer protection against influenza.
Development of a Central Vaccine Antigen
An alternative to continuously producing new candidate vaccines is to select or engineer antigens that trigger cross-reactive immune responses. Among the antigens closest to the map’s centre, researchers identified 12 highly reactive variants. Eight of these twelve variants were capable of binding α2,6-linked sialic acids.
Based on these observations, the team constructed three candidate vaccine antigens (CVAs): Vietnam, Indonesia, and Anhui. These antigens acquired dual receptor-binding profiles – recognizing both α2,3- and α2,6-linked sialic acids – and were positioned closer to the center of the antigenic map than their original counterparts. After ferret immunization, all three CVAs elicited antibodies that effectively inhibited hemagglutination across a broad range of viruses, with CVA–Anhui producing the strongest antibody response.
Additional experiments identified an amino acid substitution that increased the stability of CVA–Anhui during passages in chicken eggs. This substitution was incorporated into the final antigen design, yielding AC–Anhui, which mapped even closer to the antigenic centre and was selected for further preclinical evaluation.
Preclinical Ferret Study
In the ferret experiment, researchers evaluated two aspects: the protective efficacy of the AC–Anhui vaccine and the impact of the introduced amino acid substitution on the strength and breadth of hemagglutination-inhibiting (HI) antibody responses. Two genetically and antigenically distinct viruses were used for testing – H5N1Giza and H5N6Sichuan. Ferrets received two doses of inactivated vaccines 28 days apart. These vaccines contained either the central antigen AC–Anhui, its wild-type counterpart, or HA proteins from H5N1 Giza and H5N6Sichuan. PBS served as the control.
After booster vaccination, AC–Anhui induced stronger and broader HI antibody responses than the wild-type antigen. It also generated antibodies against H5N1Giza and H5N6Sichuan, whereas the wild-type variant did not.
Four weeks after the booster dose, ferrets were challenged with the H5N1Giza or H5N6 Sichuan viruses. In the mock-vaccinated group, H5N1 Giza caused the most tremendous average body-weight loss (11.4%) and reduced activity, accompanied by increased respiratory rate. In contrast, vaccinated ferrets showed no significant changes in body weight, activity, or clinical condition. Vaccines containing AC–Anhui or HA from H5N1Giza prevented viral spread to the brain, whereas the wild-type Anhui HA vaccine did not.
Upon infection with the more virulent H5N6Sichuan virus, the AC–Anhui vaccine again reduced disease severity, lowered respiratory tract viral loads, and prevented extra-respiratory spread. Across most metrics, it outperformed the wild-type Anhui HA vaccine and was comparable to the HA H5N6 Sichuan vaccine.
Reference
A vaccine central in A(H5) influenza antigenic space confers broad immunity
