The work cannot be used to predict whether that combination of changes — or other combinations which would have the same effect — will ever come into being. But it may give pause to those who argue that a bird flu virus cannot evolve to become transmissible from human to human.
The research is the latest in a series of so-called gain of function studies, controversial experiments that deliberately alter bird flu viruses to the point where they can spread between mammals in the way seasonal flu viruses do.
"I think this whole set of experiments really shows that bird flu viruses can become airborne. Whether they will is a completely different question," senior author Ron Fouchier said of the paper, published in the journal Cell.
Many flu scientists insist it is critical to understand what changes enable that evolution. But the many vocal critics of the work denounce these types of studies as unnecessarily dangerous and without merit.
In fact, earlier studies by Fouchier, a virologist at the Erasmus Medical Centre in Rotterdam, the Netherlands, and Yoshihiro Kawaoka of the University of Wisconsin-Madison, were blocked from publication for months in late 2011 and early 2012 after a panel of biosecurity experts that advises the U.S. government ruled that the list of mutations needed to give H5N1 viruses the capacity to spread by the airborne route should not be disclosed.
The National Science Advisory Board for Biosecurity recommended those first studies detailing how to make airborne-transmissible H5N1 viruses should be published in redacted form, with the key details removed.
The scientists and the journals trying to publish the papers objected and after months of debate and dispute the papers were eventually cleared for publication in May and June of 2012. Since that time, though, opposition to this type of research within the broader scientific community has grown.
Michael Osterholm was a member of the NSABB, and was one of the committee members who objected to putting this type of information into the public domain. His views on the issue have not changed in the intervening two years.
While some critics argue these studies should not be done at all, Osterholm doesn't take that view. But he insists that making this kind of information readily available makes it possible for scientists anywhere — even those working in laboratories which don't meet the highest standards of biosafety and biosecurity — to make dangerous viruses.
"We now have intentionally lowered the bar to the point ... where we have made it possible for laboratories around the world to do this work," said Osterholm, who is the director of the Center for Infectious Diseases Research and Policy at the University of Minnesota.
"I think what remains very problematic is just the enabling information it provides. There are a lot of laboratories around the world that are not funded by the U.S. government, do not have specific restrictions or limitations on what they can do relative to biosafety."
At the time of the 2011-12 dispute over the earlier gain-of-function papers, an article like the one Fouchier published in Cell would have had to be reviewed by the NSABB if the work had been conducted with U.S. government funding. (This work was.) But Osterholm said the paper was not referred to the NSABB. He said, in fact, that the group has not met in the past 18 months.
This new study built on the earlier work, with Fouchier and his team trying to see how few mutations were needed to allow for mammal-to-mammal transmission of this bird flu virus.
They identified a number of mutations they felt were critical, and then began the laborious trial-and-error work of making viruses with various combinations of the changes. They then infected ferrets — animals which are often used as proxies for people in flu studies — with the engineered viruses and put healthy ferrets in cages nearby to see if infection would spread from the sick animals to the healthy ones.
The process led to two viruses with five mutations each that were able to spread from ferret to ferret. In one case, one of four healthy ferrets contracted the infection, and in the other, two of four became infected.
Given the small number of animals in each experiment — a deliberate attempt to minimize the number used and euthanized — one cannot tell whether these engineered viruses are highly or moderately transmissible, acknowledged Adolfo Garcia-Sastre, an influenza expert and director of the Global Health and Emerging Pathogens Institute at Mount Sinai Hospital in New York.
"It is challenging and very expensive to conduct experiments using more animals," Garcia-Sastre said via email. "(But) the data is real."
The mutations engineered into these viruses are known to affect which receptors the viruses attach to — the type more common in birds or those found more predominantly in human upper airways — as well as the virus's ability to replicate and to remain stable at higher temperatures.
Fouchier said the combination of mutations his group used likely would not work with all H5N1 viruses or all bird flu viruses. For instance, it is known that some of the changes that appeared to be needed to make the viruses transmissible can be achieved by more than one mutation.
One important thing to note: The changes identified by Fouchier's team appear to weaken the bird flu viruses.
To date about 60 per cent of people known to have been infected with H5N1 have died, leading to serious concerns about what the virus would do if it acquires the ability to easily infect people. But in this study the two viruses the team developed did not kill any of the infected ferrets.
Follow (at)HelenBranswell on Twitter.