Researchers have identified five changes to a strain of the H5N1 avian influenza virus which can make the virus transmissible between ferrets via respiratory droplets. The findings underscore the risk that a similarly transmissible virus might evolve naturally and cause a human pandemic. The work should also assist efforts to develop global influenza biosurveillance, as well as drugs and vaccines to protect against this threat.
This research, published in full after considerable international debate, along with another study that assesses the virus’ potential to evolve this ability naturally, appears in the 22 June 2012 edition of Science. The issue also includes six commentary articles that discuss the work’s biosecurity implications.
Since its first detection in 1997, the H5N1 virus has devastated the poultry industries of numerous countries and caused over 300 human deaths—about half of all those infected. The virus does not spread easily between humans, but if it did, experts say it could cause a deadly pandemic.
The study and another that was recently published in Nature have been controversial. Critics have been concerned that the research could be used for harmful purposes, or that the mutated virus might accidentally be released from the laboratory where it is contained.
The decision to publish this research in full is consistent with the recommendations of the National Science Advisory Board for Biosecurity and the World Health Organization which have both determined that the study’s benefits for public health and safety outweigh its risks. All the experimentation was conducted under strict biosecurity conditions called BSL3+ (Biosafety Level 3+, which requires specific training and equipment to protect against agents that can have serious or lethal effects when inhaled) that were monitored by the Netherlands Erasmus Medical Centre, the U.S. National Institutes of Health, and U.S. Centers for Disease Control and Prevention.
“It’s our hope that…publication will help to make the world safer, particularly by stimulating many more scientists and policymakers to focus on preparing defenses,” said Bruce Alberts, Science’s editor in chief, in a teleconference for journalists on 20 June.
Anthony Fauci, director of the National Institutes of Allergy and Infectious Diseases, which helped fund the research, also described the rationale for full publication.
“When it’s out there in the general literature, you can get anyone from X-ray crystallographers to structural biologists, to physiologists, to viral epidemiologists getting involved,” he said. “Being in the free and open literature would make it much easier to get a lot of the good guys involved [compared to] the risk of getting the rare bad guy involved.”
Could H5N1 Spread Easily Among Mammals?
To determine what it would take for the H5N1 virus to become capable of airborne transmission via tiny droplets or “aerosols,” Sander Herfst and Ron Fouchier of the Erasmus Medical Center in Rotterdam, the Netherlands and colleagues began by genetically modifying the virus.
The researchers changed three amino acids that would be expected to increase the virus’ affinity for mammalian hosts. They infected ferrets—which show symptoms of flu similar to humans—by applying the modified virus directly inside the animals’ noses. Then the researchers swabbed the noses of those ferrets and inoculated more ferrets, “passaging” the virus several times to see how it evolved. As the experiment proceeded, the researchers took tissue samples from the ferrets and sequenced the viruses they found.
Fouchier’s team consistently identified the mutations that had been added at the beginning. In addition, several other novel mutations appeared during the transmission experiments, which seemed to give the viruses an advantage for replicating in the nose and throat and potentially for being transmitted by air.
Next the researchers tested whether the “selected” mutant viruses could be spread to uninfected ferrets through the air, by respiratory droplets. They placed pairs of cages containing healthy and infected ferrets next to each other but not in direct contact. Most of the healthy ferrets became infected with influenza virus. The animals later recovered, and in fact the mutant viruses were only found to be fatal to ferrets during pathology experiments when the virus was inoculated directly into the throat at extremely high doses.
The researchers sequenced the airborne, infectious viruses and found five mutations that together conferred the ability for airborne transmission: the three that were introduced initially and two that were selected during passaging. Four of the changes were in hemagglutinin, the protein on the virus’ surface that helps it enter host cells. The fifth was in the polymerase 2 protein, which helps the virus replicate its genome.
These mutations have all been detected individually, or in partial combination, in viruses found in nature. It has long been thought that, to trigger a pandemic, influenza viruses must first mix their genomes with another virus in an animal host. In these experiments, however, such “reassortment” wasn’t necessary for the mutant virus to change its transmissibility characteristics.
The study does indicate that currently available drugs and vaccines are effective against/H5N1 in ferrets and this may be the case for humans too, though further research will be necessary. In test tube experiments, strains of the modified virus responded to the antiviral drug oseltamivir and to antibodies from ferrets that had received candidate H5N1 vaccines.
The Potential for a Natural Outbreak
Several circulating H5N1 avian influenza strains already have two of the mutations known to make experimental strains of the virus transmissible between mammals via respiratory droplets. That’s the conclusion of an analysis of surveillance data by Colin Russell and Derek Smith of the University of Cambridge in the United Kingdom and colleagues, published in the same issue of Science.
These viruses might thus need only three more mutations to resemble the virus revealed in the Fouchier study, or just two mutations to resemble the virus in the recently published Nature study by Masaki Imai, Yoshihiro Kawaoka, and colleagues.
It’s currently impossible to estimate precisely the probability that these airborne-transmissible viruses will evolve naturally, but the results suggest that the remaining mutations could evolve within a single mammalian host. The possibility of an airborne-transmissible H5N1 virus evolving in nature is “a potentially serious threat,” the researchers write.
“We now know that we’re living on a fault line,” Smith said. “What we have discovered in this working collaboration with Drs. Fouchier and Kawaoka is that it’s an active fault line.”
In the second part of the study, Russell and colleagues analyzed factors that might increase or decrease the likelihood that the virus will develop the mutations after infecting humans or other mammals. Using a mathematical model of a virus’ evolution within its host, they identified six factors that could make the virus more likely to evolve the full set of mutations, and two factors that would make it less likely.
“The path is now clear for what needs to happen next…in order to more accurately assess the risks of these viruses emerging in nature,” Smith said.
Rino Rappuoli of Novartis Vaccines and Diagnostics, who authored one of the accompanying commentary articles in Science, said during the teleconference that it should now in theory be possible to effectively prepare for an H5N1 pandemic by producing a vaccine faster and in larger quantities than during past pandemics.
Read all the articles in Science‘s special section on the H5N1 virus (free to the public).
Read more of Science‘s special commentary articles on H5N1 influenza research (free to the public).
Listen to Science Editor-in-Chief discuss the research on the Science Podcast.