Skip Navigation

Chink found in flu virus enzymes

August 16, 2006 By Alison Abbott This article courtesy of Nature News.

New drug targets could combat resistance.

Biologists have uncovered subtle variations in the structure of a key flu virus enzyme, a finding that might lead to better anti-flu drugs.

The discovery, published this week in Nature1, is particularly important because some flu viruses are becoming resistant to all currently available drugs.

As well as offering a new target for drugs, the work "helps our understanding of how drug resistance develops", says Frederick Hayden, an expert on flu drugs at the World Health Organization in Geneva, Switzerland.

All flu viruses have the proteins haemagglutinin and neuraminidase on their coats. Haemagglutinin (H) helps the virus attach to cells and has 16 known forms. Neuraminidase (N) is an enzyme that comes in nine forms. It allows the virus to be released from infected cells and infect new ones. Different flu virus strains are named after the different forms of two proteins that they carry.

The strain that is causing most worry at the moment 'bird flu' is the often lethal virus named H5N1, currently circulating in birds and humans. Experts fear it may mutate into a pandemic strain that could kill millions of people.

Current best

This helps our understanding of how drug resistance develops.
Frederick Hayden
WHO flu expert
At the moment, the best defence against it is the drug oseltamivir, better known as Tamiflu, which is taken by mouth. The closely related zanamivir (Relenza) is also effective, but it has to be inhaled a problem when the lungs are damaged by the virus.

Both oseltamivir and zanamivir are designed to block neuraminidase. The chemists who developed the drugs based their designs on the structures of the N2 and N9 forms of the enzyme, because these were the only ones available at the time. But both drugs are also effective against flu viruses with other types of neuraminidase. Scientists assumed this was because the 'active site' the part of the enzyme that cuts viruses free from cells and also where the drugs bind must be identical in all of the types. This made sense because the sequences of amino acids in all neuraminidase active sites are virtually identical.

But John Skehel from the National Institute for Medical Research in London, UK, had a hunch this might not be the case. With his colleagues he used X-ray crystallography to determine the structures of three other types of neuraminidase: N1, N4 and N8. They found that the three-dimensional structures of the active site were indeed different to those in N2 and N9. When N1, N4 and N8 bind their normal targets, a small cavity close to the active site opens up. It seems that amino acids well outside the active sites influence whether or not such a cavity opens.

So drug developers were lucky that oseltamivir and zanamivir were able to block all neuraminidase subtypes. Skehel suggests that they could now exploit information from his studies to find new drugs ones that may even be subtype-specific, or perhaps less likely to allow resistance to develop.

"We have to remember, though, that there is more to making a drug than just getting a small molecule to fit into an active site," Hayden cautions. It has to be well tolerated by patients and have the right chemistry to enable it to get to the virus in the body.

Other drugs

At the moment, researchers may have a bit of time to develop new leads, since resistance to oseltamivir seems to be developing relatively slowly. And pharmacologists still have a few tricks up their sleeves. Other neuraminidase inhibitors are starting to enter clinical trials, including an intravenous versions of zanamivir and another promising drug called peramivir.

Zanamivir is already known to be effective it's just that the inhaled version is impractical. "I'm very excited about an intravenous neuraminidase inhibitor," says Hayden. "It answers a real medical need because there is currently no injectable drug that can be given to very sick people."

Visit our newsblog to read and post comments about this story.


  1. Russell R. J., et al. Nature, doi: 10.1038/nature05114 (2006).


Need Assistance?

If you need help or have a question please use the links below to help resolve your problem.