An approach that adds a safety feature to laboratory-based studies on avian influenza viruses is presented in a paper published online this week in Nature Biotechnology. This method could further strengthen the biosafety of 'gain-of-function' influenza studies, and may reduce the risk of the virus causing illness in humans that come into contact with these laboratory strains.

While some strains of avian influenza can be contracted after direct contact with infected birds, these strains cannot currently spread efficiently from human to human. Recent studies in ferrets - the animal model used for studies investigating human influenza - demonstrated that several mutations could lead to certain avian influenza viruses 'gaining' the capacity for airborne transmission, to some extent, between these ferrets. These 'gain-of-function' influenza studies are carried out in order to try to aid global efforts to prevent pandemics, however concerns have been raised about the potential for accidental or deliberate release of these laboratory strains of influenza virus.

Though such studies are already conducted under specific safety guidelines, Benjamin tenOever and colleagues devise a strategy that they suggest could further ensure safety inside, and potentially outside, the laboratory. Different species express different microRNAs, strings of nucleic acid which are able to suppress expression of genes containing particular target sequences. The authors reasoned that if an influenza strain expressing a microRNA target site infects an organism whose cells express that microRNA, influenza expression and replication will be shut down. After identifying a microRNA expressed in human and mouse but not ferret lungs, they inserted its target sequence into the influenza genome. They show that when ferrets were exposed to the resulting influenza virus it could be transmitted normally between the animals, whilst mice infected with this strain showed no evidence of illness.

The team caution that it is not yet known whether this virus will be nonpathogenic in humans; however, this approach holds the potential to add a molecular layer to the physical barriers already used for biocontainment of pathogens in the laboratory.