Reverse engineering the antigenic architecture of the haemagglutinin from influenza H5N1 clade 1 and 2.2 viruses with fine epitope mapping using monoclonal antibodies
Graphical abstract
Highlights
► We reverse engineered the antigenic structure of the haemagglutinin of two highly pathogenic clade 1 and 2.2 H5N1 viruses. ► Fine mapping reveal the localisation and fine structure of cross-clade antigenic sites in the globular head of the HA molecule. ► The overlapping cross-reactivity of these antigenic sites suggests that the reactive amino acid positions relate to the antigenic evolution of the H5 clade 1 and 2.2 viruses. ► Significantly, there is overlap with these antigenic positions and those required for direct mammalian transmission.
Introduction
The first documented cases of H5N1 subtype avian influenza A virus crossing the species barrier from poultry to humans occurred in Hong Kong in 1997 resulting in several deaths (de Jong et al., 1997, Subbarao et al., 1998). Since 1997, highly pathogenic avian H5N1 influenza A virus (HPAIV) out breaks have occurred frequently (Webster and Govorkova, 2006, Webster et al., 2005, Webster et al., 2007), establishing an endemic presence in poultry and wild bird populations in south-east Asia (Komar and Olsen, 2008). From this origin, HPAIV H5N1 has spread to avian populations across Europe and emerged in Turkey, Russia, and Africa (Webster and Govorkova, 2006). Human infection by HPAIV strains of H5N1 have resulted in high levels of mortality and morbidity and raised serious concerns about its pandemic potential. Infection of humans has thus far been limited, with no convincing evidence for sustained human to human transmission (Imai and Kawaoka, 2012, Imai et al., 2012, Yang et al., 2007a). However, concern over the sustained human to human transmission of HPAIV has heightened by the recent demonstration that a small number of changes to the receptor binding site (RBS) of haemagglutinin (HA) can confer sustained mammalian transmission (Imai et al., 2012). Therefore currently circulating H5N1 viruses in poultry and wild-aquatic birds have the potential to cause serious illness in humans without the requirement for re-assortment with human influenza viruses, or the need for an intermediate mammalian host (Nguyen et al., 2005, Smith et al., 2009, Wallace et al., 2007, Wan et al., 2008).
HA is the principal surface antigen on the influenza virus, and as such is the target for neutralising antibodies (Ducatez et al., 2011, Ekiert et al., 2009). Antigenic drift mutations that produce amino acid (aa) substitutions in HA allow the virus to escape immune surveillance and cause recurrent epidemics (Hoper et al., 2012, Kaverin et al., 2007). These mutations often correlate with changes to the phenotypic features of the virus such as virulence in mice and affinity for sialic acid receptors (Krylov et al., 2010, Maines et al., 2011, Stevens et al., 2006, Yamada et al., 2006, Yang et al., 2007b, Yen et al., 2009). Although poultry continues to be the major source of human infection of H5N1, there is the constant risk of emergence of human-adapted H5 viruses. Human adaptation of avian H5N1 viruses poses a significant pandemic threat. A greater understanding of the antigenic structure and evolution of the H5 HA proteins can help researchers to prepare for, and mitigate the risks associated with a H5 influenza pandemic.
In order to ensure vaccines and immunologic diagnostic tests contain the most relevant variants of circulating viruses, the antigenicity of H5N1 influenza isolates from outbreaks of disease and epidemics must be regularly monitored. A number of neutralising epitopes of the H5 HA have been operationally mapped, however such early reports used the H3 crystal structure for antigenic mapping (Philpott et al., 1990). The resolution of the crystallographic structures of the HA from the HPAIV H5N1 A/Vietnam/1203/04 and A/Vietnam/1194/04 strains have presented the opportunity to establish more precise structural epitope maps (Stevens et al., 2006, Yamada et al., 2006).
In the present study, we compared the differences in antigenicity across three currently circulating HPAIV H5N1 strains from clades 1 (A/Vietnam/1194/04), 2.1 (A/Indonesia/5/05) and 2.2 (A/Bar-headed Goose/Qinghai/65/05). The mutants were characterised by their cross-reaction with three monoclonal antibodies (mAbs) raised against the HA of each parent virus. This allowed us to antigenically map the localisation and fine structure of clade specific epitopes on the three-dimensional crystallographic structure of the A/Vietnam/1194/04 H5 HA molecule and a homology model of the A/Bar-headed Goose/Qinghai/65/05HA. The particular features of each antigenic mutation are discussed on a structural level. Our mapping revealed clade specific differences in fine antigenic structure of HA from recent H5N1 isolates, which has broad reaching implications for future monitoring efforts of emerging H5N1 strains and for designing novel vaccines.
Section snippets
Viruses
H5N1 isolates were obtained through the World Health Organisation Global Influenza Surveillance Network. The HA and neuraminidase (NA) genes of the H5N1 viruses A/Vietnam/1194/04 (NIBRG-14; NIBSC, London, UK), A/Bar-headed Goose/Qinghai/1A/05 (Seed virus 163222; St. Judes, TN, USA) and A/Indonesia/5/05 (IBDC-RG2; CDC, GA, USA) were produced using reverse genetics with 6 genes from A/Puerto Rico/8/34. These viruses correspond to A/Vietnam/1194/04, A/Bar-headed Goose/Qinghai/65/05 and
Results
MAbs against A/Vietnam/1194/04 (Clade 1), A/Indonesia/5/05 (Clade 2.1) and A/Bar-headed Goose/Qinghai/1A/05 (Clade 2.2) were raised by immunising mice. The mAbs 153b (anti-A/Vietnam/1194/04), 4c11 (anti-A/Bar-headed Goose/Qinghai/1A/05) and 5C9 (anti-A/Indonesia/5/05) were cross-reacted with the HA of each virus in HI and ELISA to confirm specificity as previously described (Bodle et al., 2012).
Discussion
Vaccination to elicit a neutralising Ab response remains the most effective prophylaxis against influenza infection. However high mutation rates within the major neutralising epitope regions of H5 HA, and the circulation of 9 phylogenetic variant H5N1 clades, hampers the development of universal vaccines (Prabakaran et al., 2010, Wang and Palese, 2009). Understanding the evolution of major neutralising epitopes on the HA of circulating H5N1 clades may aid in the development of a broad-spectrum
Funding
J.L. is an Australian NHMRC Senior Research Fellow. T.V. and M.B are Australian NHMRC Career Development Research Fellows. This work was supported by the strategic award grant 594875 from the National Health and Medical Research Council of Australia.
References (63)
- et al.
Eight-plasmid system for rapid generation of influenza virus vaccines
Vaccine
(2002) - et al.
The role of receptor binding specificity in interspecies transmission of influenza viruses
Current Opinion in Virology
(2012) - et al.
Protein disorder prediction: implications for structural proteomics
Structure
(2003) - et al.
Effect of receptor binding domain mutations on receptor binding and transmissibility of avian influenza H5N1 viruses
Virology
(2011) - et al.
Structural features of the avian influenza virus hemagglutinin that influence virulence
Veterinary Microbiology
(2000) - et al.
Evolution and adaptation of H5N1 influenza virus in avian and human hosts in Indonesia and Vietnam
Virology
(2006) - et al.
Development of an enzyme-linked immunoassay for the quantitation of influenza haemagglutinin: an alternative method to single radial immunodiffusion
Influenza and Other Respiratory Viruses
(2012) - et al.
Deciphering protein sequence information through hydrophobic cluster analysis (HCA): current status and perspectives
Cellular and Molecular Life Sciences
(1997) - et al.
Prediction of unfolded segments in a protein sequence based on amino acid composition
Bioinformatics
(2005) - et al.
A pandemic warning?
Nature
(1997)