Variation and infectivity neutralization in influenza

Abstract

Worldwide epidemics of influenza are caused by viruses that normally infect other species, particularly waterfowl, and that contain haemagglutinin membrane glycoproteins (HAs) to which the human population has no immunity. Anti-HA immunoglobulins neutralize influenza virus infectivity. In this review we outline structural differences that distinguish the HAs of the 16 antigenic subtypes (H1-16) found in viruses from avian species. We also describe structural changes in HA required for the effective transfer to humans of viruses containing three of them, H1, H2 and H3, in the 1918 (Spanish), the 1957 (Asian) and the 1968 (Hong Kong) pandemics, respectively. In addition, we consider changes that may be required before the current avian H5 viruses could pass from human to human.

Figure 1

The structure of influenza haemagglutinin membrane glycoproteins (HAs). Left: top- and side-view of a ribbon diagram of the HA structure. Two monomers of the HA trimer are drawn in grey, one is coloured. Each monomer is synthesized as a single polypeptide chain that is post-translationally cleaved into two chains – HA1 and HA2 – which are shown in blue and red, respectively. Right: superposed ribbon diagrams of the receptor-binding domains of a clade 3 HA (represented by H1) and a clade 5 HA (represented by H7). Clades 1, 2 and 4 are represented by subtypes H9, H13 and H3, respectively.

Figure 2

The haemagglutinin membrane glycoprotein (HA) receptor-binding site. Left: side-view of HA (same colour code as in Figure 1), with the receptor-binding site of one monomer highlighted. Top right: sialic acid (yellow) and residue side-chains and main-chain atoms that interact with it are presented as ball-and-stick models. Hydrogen bonds are presented as dashed lines. Bottom right: overview of sialic acid in α2,3- and α2,6-linkages to galactose bound to avian H3 and human H3 (left and right panels, respectively). In these views, the only side-chain presented is that of residue 226 of HA1.

Figure 3

Distribution of sequence changes in haemagglutinin membrane glycoproteins (HAs) of the Hong Kong pandemic era during the 1968–2005 period. The space-filling models represent, in yellow, the virus receptor-binding site and, in green, substituted amino acids. (a) All substitutions in HAs of viruses isolated between 1968 and 2005; (b) amino acid substitutions that were retained in subsequent years; (c) amino acid substitutions detected in monoclonal antibody-selected variants of A/Hong Kong/68 HA. The α-carbon tracings of the HA trimers are coloured blue and red to denote the HA1 and HA2 polypeptide chains that form each subunit of the trimer.

Figure 4

Neutralizing antibody Fab–haemagglutinin membrane glycoprotein (HA) complexes. Ribbon diagrams of the complexes showing one A/Hong Kong/68 HA monomer and, from left to right, Fabs (in green) of antibody 1, antibody 2 and antibody 3 which select mutations at HA1 residues 157, 226 and 63, respectively. These residues are coloured red in the complexes. Amino acids in the receptor-binding site are shown as yellow space-filling models.

Figure 5

The relationship between inhibition of virus binding to cells by antibodies and neutralization. The ratios of the number of virus plaques to the number of plaques without antibody (plain lines, filled symbols), and the ratio of cell-bound virus to cell-bound virus without antibody (dashed lines, open symbols), are plotted on a semilogarithmic scale as a function of antibody concentration. The blue, green and red curves correspond to antibodies 1, 2 and 3, respectively.