History of the discovery of Relenza

This was discovered by George Hirst, who noticed that when red blood cells were mixed with fluids from influenza infected chicken embryos in cold conditions the cells were very heavily agglutinated by the virus. These red cells dispersed when warmed up and could not be re-agglutinated in the cold with fresh virus. This led him to the conclusion that the influenza virus's enzyme was destroying receptors on red blood cells.

Alfred Gottschalk heard of Hirst's experiment and interpretation of results, and this led him to believe that there was a "split product". He discovered sialic or neuraminic acid (Fig 2), a type of sugar, and the enzyme on the virus was called neuraminidase (or sialidase). At this time it was thought that it was the neuraminidase which was responsible for the observations made by Hirst, but it was later shown by Robin Valentine, W. Graeme Laver, Norbert Bischofberger and Robert G. Webster that the hemagglutinin (receptor-binding) and neuraminidase (receptor-destroying) activities of the virus resided in two quite different spikes on the surface of the virus.

Fig 2: Sialic Acid

Top of page

Discovery of how new pandemic strains of 'flu A occured.

Ed Kilbourne, W. Graeme Laver, Norbert Bischofberger and Robert G. Webster realised that hybrid viruses could be formed by infecting cells simultaneously with two different Type A flu viruses. This was because the RNA pieces coding the various virus proteins reassorted, some of the viruses contained the hemagglutinin from one parent and the neuraminidase from the other. This "mating" of two parent viruses to give a hybrid virus explained how new pandemic strains of 'flu A occurred, and led to a very good way of producing influenza viruses with any desired combination of hemagglutinin and neuraminidase spikes. This helped towards finding a way of producing pure neuraminidase which was later essential for crystal growth and drug design experiments.

Laver, Bischofberger and Webster isolated one type of influenza virus by sucking off the allantoic fluid surrounding the embryo of infected chicken eggs and purifying this. The virus particles were incubated with an enzyme capable of digesting proteins. This enzyme was selected to split the "heads" of the neuraminidase spikes off the virus particle without destroying them and to leave behind or destroy the hemagglutinin spike. The neuraminidase "heads" obtained were concentrated using high-speed centrifugation. The tiny pellet of neuraminidase heads examined had a crystalline appearance, and X-ray diffraction analysis of larger crystals showed that they were made of protein.

Neu5Ac2en (DANA) was shown to inhibit influenza neuraminidase:

Different variants of 'flu neuraminidase were known to exist, each containing an amino acid sequence that varies between types of neuraminidase apart from one small sequence.It was seen that the conserved amino acids came together when the neuraminidase polypeptide folded up to form the active enzyme. This formed a well conserved cavity which was the active catalytic site of the neuraminidase enzyme. It became apparent that a plug-drug could be made to exactly fit into the active site and inhibit the neuraminidase activity from other influenza viruses. A synthetic analog of sialic acid called Neu5Ac2en (DANA) (Fig 3) was shown to inhibit the influenza virus neuraminidase, but not sufficiently enough to be used treatment for the 'flu in humans.

Fig 3: Neu5Ac2en (DANA)

Mark von Itzstein and colleagues discovered that replacing the OH at the 4 position of sialic acid with a positively charged amino group made a better inhibitor than sialic acid or its analogue, DANA. Replacing the OH at the 4 position of sialic acid with a guanidino group led to a potent inhibitor of 'flu neuraminidase. This compound was given the names GG167 and Zanamivir and is now more commonly known as Relenza. Peter Colman soaked the substrate for sialic acid in neuraminidase crystals and used X-ray crystallography to determine the three-dimensional structure of the crystals. The strong binding of Relenza by 'flu neuraminidase which was seen is due to the positively charged guanidino group being anchored by the negatively charged glutamic acids. More details about this are provided in the immunology section.