H5N1 transmission experiment stirs concern

Nov 17, 2011 (CIDRAP News) – A national biosecurity board that monitors "dual use" research is apparently worried about an as-yet-published study in which a mutant form of H5N1 avian influenza virus was found to be easily transmissible in ferrets, which are considered good models for flu in humans.

A National Public Radio (NPR) report today said the National Science Advisory Board for Biosecurity (NSABB) is scrutinizing the research by Dr. Ron Fouchier of Erasmus University Medical Center in the Netherlands. The board provides guidance on biological research that has a legitimate purpose but could be misused to endanger public health.

Dr. Paul S. Keim, acting chair of the NSABB, said today that the board is conducting a review of H5N1 virus transmission in mammals, but because of the board's confidentiality rules, he could not give any details. Keim is director of pathogen genomics in the Translational Genomics Research Institute at Northern Arizona University.

Referring to the Office of Biotechnology Activities in the National Institutes of Health's Office of Science Policy, Keim told CIDRAP News, "We have now been directed by OBA staff that we can acknowledge that a review process on the H5N1 transmissibility in mammals is under way. We are not allowed to provide additional details."

The H5N1 virus causes human illness relatively rarely, but it is often deadly when it does, with a case-fatality rate of about 60% among cases confirmed by the World Health Organization. Though the virus has circulated in poultry in many countries since 2003, it has not gained the ability to spread easily in humans. Scientists worry that if it did gain that ability, it could spark a fearsome pandemic.

With the goal of identifying genetic changes that could lead to greater person-to-person transmissibility, a number of researchers have introduced mutations in the virus and studied how the mutant strains behaved in animals.

Fouchier gave a general description of his experiments at a European meeting in September, according to a news story published in Scientific American after the meeting. He and his team introduced various mutations into the virus and watched their effects on its ability to attach to human respiratory tract cells. They found that with as few as five single mutations, the virus could bind to nasal and tracheal cells, according to the story.

But when tested in ferrets, this mutant virus still didn't spread very easily through close contact. Fouchier and his team then undertook to let the virus evolve naturally—a project that he described as "really, really stupid," according to the story. They inoculated one ferret with the mutant virus, and after it got sick, they exposed a second ferret to infectious material from the first one.

After they repeated this process 10 times, "H5N1 became as easily transmissible as seasonal flu," the story said. Fouchier said he concluded from this that H5N1 viruses "can become airborne" and do not need to reassort with other mammalian flu viruses to do so.

The report quoted another expert, Albert Osterhaus, also of Erasmus University, as saying the individual mutations that Fouchier introduced have already occurred naturally in animals, but not together.

Further details about Fouchier's study have been hard to come by. In response to a CIDRAP News query last week, Fouchier said he couldn't comment on the study until it is published. Other flu experts, including those at the Centers for Disease Control and Prevention (CDC), also declined to discuss the study or didn't reply to queries last week.

The NPR report said the Fouchier study is the subject of a "fierce debate" among disease experts and is being reviewed by the NSABB. The experiment was criticized by some experts quoted in the NPR report.

"It's just a bad idea for scientists to turn a lethal virus into a lethal and highly contagious virus, and it's a second bad idea to publish how they did it so others can copy it," said Thomas V. Inglesby, MD, director of the Center for Biosecurity at the University of Pittsburgh Medical Center, who is not a member of the NSABB.

While biology has a culture of openness and relies on the full sharing of findings, occasional exceptions to this policy are warranted, and Fouchier's study calls for an exception, he told NPR.

It was not immediately clear just how the NSABB might influence whether and in what form Fouchier's findings would be published. The board makes recommendations on policies governing publication, public communication, and dissemination of dual use research methods and results, according to information on the OBA Web site. It is up to the federal government to convene relevant agencies to determine how to respond to the recommendations.

One previous case in which the NSABB influenced a research publication pertained to reports on the reconstruction of the 1918 pandemic flu virus using material from preserved tissue samples, according to Keim.

"In this case NSABB recommended that the papers be modified to better represent the biosafety aspects of the research and also to explain the positive benefits of the work to public health," he told CIDRAP News. "There were no restrictions recommended by NSABB on the actual data or limitations to the results distribution."

Some other recent studies have tested the ability of mutant or reassortant strains of H5N1 to replicate and spread in animals, but the findings were less dramatic than Fouchier's.

For example, the December issue of the Journal of Virology includes a study in which researchers at St. Jude Children's Research Hospital in Memphis inserted the hemagglutinin gene from a 1997 Hong Kong strain of H5N1 virus into a pandemic 2009 H1N1 virus. When this hybrid was grown serially in human lung epithelial cells, it became more pathogenic in mice, "suggesting that these viruses may easily adapt to humans and become more virulent," the report says. The authors concluded that natural reassortment between circulating 2009 H1N1 and H5N1 viruses could lead to viruses with increased pathogenicity in mammals.

Dr. Lynn Enquist, the Journal of Virology's editor-in-chief and a member of the NSABB, told NPR that the journal's staff carefully considered whether to publish the paper. He observed that the type of reassortant the scientists created could occur in nature.

Another recent study, by researchers from the CDC and Scripps Research Institute, suggested that it would take a series of complex changes for H5N1 viruses to achieve airborne transmissibility in ferrets.

They determined that a particular strain of H5N1, when modified with certain mutations from previous pandemic viruses, could spread in ferrets by direct contact but not by airborne respiratory droplets, according to their report in Virology. Only when they inserted a neuraminidase gene from a human flu virus did this virus achieve a modest ability to spread via respiratory droplets.

"The complex genetic changes required by a clade 2.2 H5N1 virus to reach a low level of transmissibility in ferrets would indicate that considerable functional evolution is still required for acquisition of transmissibility in humans," the report concludes.

See also:

Information about the NSABB

Nov 17 NPR report

Scientific American story describing Fouchier's study

Journal of Virology study abstract

Virology study abstract

October 2005 CIDRAP News report on re-creation of 1918 flu virus

This week's top reads