As we discussed previously, attachment of all influenza A virus strains to cells requires sialic acids. However, there are a number of chemically different forms of sialic acids, and influenza virus strains vary in their affinity for them. These differences may determine which animal species can be infected.
In the example shown below, sialic acid is linked to the sugar galactose by what is called an alpha(2,3) linkage. This means that the carbon atom at position number 2 of the sialic acid hexose is joined, via an oxygen atom, to the carbon at position 3 of the hexose of galactose.
Avian influenza virus strains preferentially bind to sialic acids attached to galactose via an alpha(2,3) linkage. This is the major sialic acid on epithelial cells of the duck gut. In contrast, human influenza virus strains preferentially attach to sialic acids attached to galactose by an alpha(2,6) linkage. This is the major type of sialic acid present on human respiratory epithelial cells. Alpha(2,3) linked sialic acids are found on ciliated epithelial cells, which are a minor population within the human respiratory tract, and also on some epithelial cells in the lower tract.
This receptor specificity has implications for human infection with avian influenza virus strains. For example, highly pathogenic avian H5N1 influenza viruses undergo limited replication in the human respiratory tract due to the presence of some cells with alpha(2,3) linked sialic acids. However, efficient human to human transmission requires that the avian viruses recognize sialic acids with alpha(2,6) linkages. Consistent with this hypothesis, the results of studies of early influenza virus isolates from the 1918, 1957, and 1968 pandemics suggest that these viruses preferentially recognized alpha(2,6) linked sialic acids.
Epithelial cells of the pig trachea produce both alpha(2,3) and alpha(2,6) linked sialic acids. This is believed to be the reason why pigs can be infected with both avian and human influenza virus strains and serve as a ‘mixing vessel’ for the emergence of new viruses. However, Peter Palese has his doubts, as recorded in The Great Influenza by John Barry:
Dr. Peter Palese….considers the mixing-bowl theory unnecessary to explain antigen shift: “It’s equally likely that co-infection of avian and human virus in a human in one cell in the lung [gives] rise to the virus…There’s no reason why mixing couldn’t occur in the lung, whether in pig or man. It’s not absolute that there are no sialic acid receptors of those types in other species. It’s not absolute that that avian receptor is really that different from the human, and, with one single amino acid change, the virus can go much better in another host”.
Ito T, Couceiro JN, Kelm S, Baum LG, Krauss S, Castrucci MR, Donatelli I, Kida H, Paulson JC, Webster RG, & Kawaoka Y (1998). Molecular basis for the generation in pigs of influenza A viruses with pandemic potential. Journal of Virology, 72, 7367-7373
Matrosovich M, Tuzikov A, Bovin N, Gambaryan A, Klimov A, Castrucci MR, Donatelli I, & Kawaoka Y. (2000). Early alterations of the receptor-binding properties of H1, H2, and H3 avian influenza virus hemagglutinins after their introduction into mammals. Journal of Virology, 74, 8502-8512