Research names five mutations making potent H5N1 strain airborne

By Tina Hesman Saey

Web edition: June 21, 2012
Print edition: July 14, 2012; Vol.182 #1 (p. 8)

A controversial research paper banned from publication in 2011 because it contained potentially dangerous information is now available for the world to see.

The study, appearing in the June 22 Science, details experiments in which researchers in the Netherlands created a version of the H5N1 bird flu virus that can be passed through the air from one ferret to another. The H5N1 avian influenza virus currently does not spread between people through coughs or sneezes, but the new work suggests that only a few mutations would be needed to turn H5N1 from a virus that requires close contact into one that could spread through the air.

A U.S. government advisory board originally ruled that the Science paper and a similar one published in the June 21 Nature by researchers from the University of Wisconsin–Madison and Japanese colleagues should not be published in full because terrorists might use information about the flu virus mutations to create a biological weapon (SN: 06/02/12, p. 20). The panel reversed the decision in March after seeing revised versions of both papers and speaking with the scientists.

All the hoopla may have come down to one line in the original version of the Science paper. The potentially offending statement implied that the airborne version of the virus might be as deadly as the original H5N1 strain, which has killed more than half of the people it has infected, says Vincent Racaniello, a virologist at Columbia University who has read both versions of the paper.

The revised version of the paper clearly states that ferrets that caught the airborne version of the virus from other ferrets did not die. “The data are the same, but the way it is explained is very different,” Racaniello says. “Everything is explained and put in context.”

In the study, researchers led by Ron Fouchier of Erasmus Medical Center in the Netherlands performed experiments to determine if the H5N1 virus could mutate into an airborne form capable of spreading among mammals, including humans. The researchers genetically engineered into H5N1 three mutations previously found in other flu viruses that had caused pandemics.

Two of the mutations alter a protein called hemagglutinin on the outside of the virus. Flu viruses invade human and animal cells by using hemagglutinin (the H in H5N1) as a grappling hook to snag proteins that stud the surfaces of the cells. H5N1’s standard version of hemagglutinin grasps proteins found in the bird intestinal tract and in human lungs. The changes the researchers engineered in hemagglutinin have been shown to help the virus grab proteins found in the nose and throat, where the virus could more easily be coughed or sneezed out.

A third mutation in a protein called PB2 helps the virus copy its genetic material more easily at the relatively cool temperatures found in human respiratory tracts.

But those engineered mutations did not allow the virus to spread between ferrets through air, the team found. So the researchers put evolution to work by infecting one ferret, then harvesting viruses from the infected ferret’s sneezes and inoculating a second ferret. Each time the virus reproduces in an infected animal, it may pick up new mutations. After three to four cycles, the virus was already reproducing better than it had before in the ferrets’ upper respiratory tracts, Fouchier says.

After 10 passages the researchers tested the virus’s ability to spread through the air and found that sick ferrets could infect a ferret in a neighboring cage. Fouchier and his colleagues examined the virus’s genes and discovered many mutations. “We show that as little as five mutations and less than 10 are sufficient to make H5N1 virus airborne,” Fouchier says.

Only two mutations (beyond the ones the researchers engineered) were found in all of the airborne viruses, both in the gene for hemagglutinin. One of the mutations, known as H103Y, may make the hemagglutinin protein more stable, Fouchier says. The other mutation, T156A, helps the virus cling more securely to the upper respiratory tract proteins. That mutation is found frequently in nature, including in all human cases of H5N1 infection in Egypt, where the virus is currently active. Egypt has had 168 cases of H5N1 and 60 deaths since 2003, including 10 cases and five deaths so far this year.

Although these changes in hemagglutinin differ from ones found by the Wisconsin group, the mutations may change the protein’s function in similar ways. The virus requires few if any other changes in its other genes to become airborne, Fouchier says.

And since several of the mutations needed to make H5N1 airborne are already found in nature, it is possible that the virus could mutate in a single human host to become a pandemic strain, Derek Smith and Colin Russell of the University of Cambridge and colleagues report in the June 22 Science.