Newly Developed Glycan Array Can Distinguish Human, Avian Flu Forms

Using a newly developed glycan array, scientists at the Scripps Research Institute have analyzed the binding specificities of viral coat proteins from both human and avian forms of the deadly 1918 influenza virus.

Though the human and avian forms of the 1918 influenza viral coat protein differ by only two amino acids, the glycan array was able to distinguish between the two.

"The most immediate application for [this glycan array] would be for surveillance of the current circulating [H5N1] avian flu viruses. The array could detect any mutation that might change their specificity to attach to or recognize human structures," said Scripps research professor and study leader James Paulson, who with colleague Ian Wilson published the research in this month's issue of the Journal of Molecular Biology. "It is remarkable that only one or two mutations are required to change the specificity to cross species."

About 20 years ago, Paulson and Wilson collaborated to analyze the 1968 pandemic H3N2 influenza virus. They found that the virus contained two avian genes, and the rest of the genes were human influenza virus genes. It was hypothesized that the avian and human forms of the influenza virus had mixed together in a common host — probably a pig — and then recombined.

To analyze the binding specificities of viral coat proteins from the 1968 influenza virus, the researchers used human erythrocytes that were enzymatically modified to contain cell surface sialyloligosaccharides. Differential agglutination of cells containing different viral sequences revealed different receptor binding types.

They found that avian viruses preferentially bound to a type of sugar chain called a2-3 linked sialic acids, while human viruses preferentially bound to a2-6 linked sialic acids.

For the more recent 1918 influenza virus study, the researchers had a new tool that allowed them to skip the step of preparing enyzmatically modified human erythrocytes.

The new tool — a glycan array that contains 200 carbohydrates and glycoproteins — was made as part of a research initiative funded by the National Institute of General Medical Sciences called the Corsortium for Functional Glycomics. A paper showing that the array could be used to study a wide variety of glycan binding proteins was published in December 2004 in the Proceedings of the National Academy of Sciences.

Paulson is the director of the CFG at Scripps, while Wilson is a member of the consortium.

"Ian was separately working with another group of investigators of the 1918 flu, and he suggested … we try these [flu] mutants on the new array," said Paulson. "So 20 years later, we began collaborating on a different flu. … It turns out that the array works spectacularly well with being able to detect the bird and human structures."

Researchers obtained samples of the 1918 flu virus from the bodies of soldiers who had died from influenza in 1918. Enough viral sample was obtained that researchers were able to reconstruct the sequence of the gene that coded for the viral protein hemagglutinin.

Hemagglutinin is the receptor that makes it possible for the influenza virus to infect cells of the host organism. To see how the virus adapts from one species to another through mutations to the hemagglutinin protein, Wilson and his colleagues changed particular amino acids in the binding site of the 1918 virus' hemagglutinin. They found that influenza strains that differed by only a single sialic acid linkage, or one sugar molecule, bound to different portions of the glycan array.

Viruses defined as avian by their ability to bind to a2-3 linked sialic acids bound to receptors on intestinal epithelial cells, while viruses defined as human by their ability to bind to a2-6 sialic acids bound to receptors on epithelial cells of the lungs and upper respiratory tract.

"We showed which mutations in the 1918 virus were responsible for the shift in [species] specificity," said Paulson. "They were different from the ones that were responsible for the shift in the 1968 H3 hemagglutinins. That shows that different sets of mutations can cause the same phenotypic change for recognition of bird and human receptors."

Paulson added that the 1918 flu virus study shows that looking at the phenotypic outcome is more important than looking at genetic mutation.

"In the H3 virus, there were two amino acids that were important for that [species] switch," he said. "In the 1918 virus, there were also two amino acids, but they were different amino acids. Just by looking at the gene sequence, you wouldn't be able to tell what the specificity was. What this array does is it looks at the relevant outcome due to the change in genetic sequence."

Going forward, the researchers would like to mutate the H5 virus' hemagglutinin to see what amino acids would have to be modified in order for the viral receptor to mutate from an avian form to a human form.

"You could start with the avian H5 hemagglutinin and do the same thing [as with the 1918 virus]," said Paulson.

Though a few people in the CFG are talking about commercializing the glycan array, the array is not commercially available at the moment, Paulson said. It is, however, available free of charge to the academic community through the CFG.

— Tien Shun Lee ([email protected])