Are we Engineering a Flu Pandemic?

Avian Flu Virus

At a mid-September influenza conference in Malta, Ron Fouchier of the Erasmus Medical Centre in Rotterdam, the Netherlands, reported that his team had managed to take the avian H5N1 virus (which has high mortality in humans but is difficult to transmit) and engineer it so that it became highly contagious, at least among lab animals. They had in fact engineered a super-bug which, although currently contained in their facility, has the capacity to cause a devastating flu pandemic. In a sort of scientific synchronicity, similar experiments have been reported by virologist Yoshihiro Kawaoka and his group at the University of Wisconsin, Madison.

Not surprisingly this has raised some concern. The scientists, who are publishing their results in Science and Nature, have been asked by the U.S. National Science Advisory Board for Biosecurity to remove details that could allow people to replicate their results for destructive purposes. Many people have stated that this is the kind of work that should not have been done in the first place. The BBC recently carried a story on it here.

In their experiments, the Erasmus team were able to show that a limited set of five mutations were all it took to go from the current H5N1 virus to a highly infectious form, falsifying claims that H5N1 could not evolve to become a super-bug. This is a valuable scientific contribution which could be critical to future public health, but many people understandably have concerns about carrying out such dangerous experiments and providing information on methods to the world at large. Interestingly the same media storm and desire for censorship was not produced by the 2005 publication in Science of the reconstruction of the 1918 spanish flu virus.

While I consider this research to be seriously risky, I can see the value of the information that it produces. Understanding what genes need to change to activate a virus so that it spreads in the human population allows us to watch carefully for this starting to happen in wild virus samples. It also may allow us to get a head start in developing vaccines. It certainly should convince people that we need to take this seriously.

There are two main concerns. The first is containment: We need to be sure that viruses created in the lab as part of an experiment cannot get released into the wild. In the US, these labs maintain “Andromeda Strain“-like biological security and are run with official regulatory oversight. The NIH Office of Biotechnology Activities publishes a detailed FAQ document related to the safety practices for working with highly pathogenic influenza viruses. Researchers are highly motivated to avoid being exposed, and they do take antivirals to reduce their personal risk. The concern is greater for research going on in non-western countries where oversight may be low.

The second fear is that information is going to fall into the hands of people who will use it to purposely engineer a virus to use as a weapon against a country or against society in general. Despite this concern, experimental details will get out. Two legitimate research groups are known to already have done this work. Scientific advances tend to have a time of arrival when the conditions, ideas and techniques fall into place and enable people to spontaneously come up with similar results. Therefore even if the details of the reports from these two labs are suppressed, other people will try to do it anyway. Additionally, there is little security in trying to restrict the information to only the community of viral researchers, their students, postgrads and whoever is related to their teams. These groups are unhappy already that they are being told to keep the details secret. Someone somewhere will post it on the internet eventually.

I would ask what details a trained experimental virologist really needs to proceed with these experiments beyond that which has already been reported? Here is the method that the Erasmus group used to prepare their virus:

  1. Experimenters obtained samples of H5N1 virus.
  2. Next they generated mutated versions of the virus and determined how well they bound with cells of the respiratory tract. I’m sure that this step is a well known procedure for virologists. It might involve incubating the virus, exposing it to UV or mutagenic chemicals, and testing for growth in vitro. They found three mutations which were effective in adapting bird flu to ferrets.
  3. The experimenters transfered the virus to a ferret via nasal inoculation.
  4. When the ferret got sick they placed its new viral particles into the nose of another ferret.
  5. After ten rounds of inoculation they found that the virus could transmit through the air to ferrets in adjacent cages without help.

Note that I am just summarizing the information that is already available on-line in the Scientific American and New Scientist articles. In the long term, we cannot really do anything about details getting into the wrong hands.

We should be realistic about what kind of malevolent people could actually create a killer virus. It is not something that someone could easily do with rudimentary science knowledge and equipment bought on the internet. You would have to be well schooled in virology and experimental technique. You would need a facility which can maintain all the proper containment, and you would need high-end equipment such as access to electron microscopes and necessary support people.

That leaves government sponsored terrorism. But creating an indiscriminate biological weapon seems like an illogical move for such governments. While they could cause a primary epidemic in the West, with international movement of people being as it is, it would quickly become a pandemic. Releasing a virus like this is more of a suicide move than a way to target an enemy.

History is strewn with examples of disasters which the population had been assured were impossible, but we do know that nature will produce these pandemics eventually. We should balance these risk scenarios carefully.